The Benefits and Costs
of the Clean Air Act,_
197O to 199O ^ ^
~-"
Prepared for
US Congress
by
US Environmental Protection Agency
May 3, 1996-DRAFT
-------
-------
Table of Contents
2
3 Acknowledgments v
4
s Executive Summary , vii
6 Purpose of the Study ...,.,... ;vii
7 Study Design '. .. v *.,. ,... vii
* Summary of Results , , .,,,,....... viii
9 -• Conclusions and Future Directions i 4...... xviii
10 \
;/ Tables ..., xxi
* ?'.
12 • - "
13 Figures . . * xzvii
14 -" . ' -
is . Acronyms and Abbreviations ...'..._..* xxix
Introduction .. 1
Background and Purpose .y. , ,±, . r , .1
Clean Air Act Requirements, 1970to 1990u^.Jv."-^ * 1
19 Section 812 of me Clean Air Act Amendmentrjof 1990 J, 2
20 Analytical Design and Revie%ii* * 3
21 Review Process ^ ^iiiSi ".*... 8
22 Report Organization -, -«• •««'• •' • • $
23 Cost and Macroeconomic Effects 11
24 Direct Compliance Cosfe»;igi>^ , 11
25 Indirect Effects of the
26 Conclusions ,...'. 14
27 Emissions • 15
a Sector-Specific.Approach 16
29 Summary of Results 18
jo Air Quality 21
31 General Methodology 23
32 Sample Results 24
33 Physical Effects 31
34 Human Health and Welfare Effects Modeling Approach 31
15 Key Analytical Assumptions 35
Health Effects Modeling Results 38
Other Physical Effects • •- 39
-------
Table of Contents
i Economic Valuation .....'.. 43
2 Unit Valuation 43
3 Aggregation of Benefits 47
4 Uncertainty in the Benefits Analysis 51
s Appendix A: Cost and Macroeconomic Modeling 57
6 Macroeconomic Modeling „.....*.".'., 58
i Direct Compliance Expenditures Data 67
i Assessment Results ...„,.,,... 77
9 Cost and Macroeconomic Modeling References 84
10 Appendix B: Emissions Modeling ...., ............. 87
n Industrial Boilers and Processes 89
12 Off-Highway Vehicles 98
a On-Highway ...«„- 101
u Utilities ...:.; .".V-"•"•*•. .-• -H4
a Commercial/Residential & , -..;..., 122
16 Emissions Modeling References ,.;-„,-, 133
i? Appendix C: Air Quality Modeling ,U*> 137
y« Introduction .;*: ^ 137
19 Carbon Monoxide -••*:<•> *% • * * • 138
M Sulfur Dioxide ***A ^'S^r'- '• 142
21 Nitrogen Oxides ...»t :*s: .Jf, '*'!**?• •"• 144
22 Acid Deposition, . /^T.^JSi.. ' 147
23 Particular Matter , '^s^f^:. 153
24 Ozone .... ^ .* ». V, iiiXCi. 160
25 %sibiiity T.^,.^%v....:^5r: ies
26 AirQuaUtyModelingBe|&ences .?..... 175
27 Appendix D: Human Health and Visibility Effects of Criteria Pollutants 177
a ' Introduction and Overview^ 177
29 Concentration-Response Functions ~ 186
x Health Effects Model 214
31 5lCi»erResuhs ^IPT .....: 225
32 ^y«iiMod«|^keferences 225
~ "."__", r" ' - .''-•*"
33 Appendix E: Ecological Effects of Criteria Pollutants 231
34 Introduction • 231
33 Benefits From Avoidance of Damages to Aquatic Ecosystems 231
it Benefits from Avoided Damages to Wetland Ecosystems 241
37 Benefits from Avoided Damages to Forests 244
3s Ecosystem Effects References 251
u
-------
Table of Contents
i Appendix F: Effects of Criteria Pollutants on Agriculture 255
2 Introduction 255
/ Ozone Concentration Data 256
4. Yield Change Estimates 259
i Economic Impact Estimates 263
6 Conclusions 265
7 Agricultural Effects References .-;« 303
s Appendix G: Lead Benefits Analysis » .305
9 Methods Used to Measure and Value Health Effects , ; 305
10 Industrial Processes and Boilers and Electric Utilities ...,,,,,. , 323
n Reduction in Health Effects Attributable to Gasoline Lead Reductions 343
12 Lead Benefits Analysis References , , 349
1.3 Appendix H: Air Toxics 355
14 Introduction 355
u Limited Scope of this Assessment , 355
u History of Air Toxics Standards under the Clean Air Act of 1970 357
u Quantifiable Stationary Source Air Toxics Benefits „>,,,,, 357
it Non-utility Stationary Source Cancer Incidence Reductions .. 361
19 Mobile Source HAP Exposure Reductions 367
Non-Cancer Health Effects ...,,,. „ 369
Ecological Effects L- *..*..... 369
22 Conclusions — ResearchNeeds ,;„-».,.. 371
23 Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants 373
24 Methods Used1»Value HeaWiEffecfe 373
25 Results of Valuation of Health and Welfare Effects -384
26 Economic Valuation References .,»-»; 388
27 Appendix J: Future Directions ..* 389
21 Research Implications w., „. ..., 389
29 Future Section 812 Analyses 389
in
-------
[This page intentionally blank].
IV
-------
Acknowledgments
2 This project is managed under the direction of Robert D. Brenner, Director of the U.S. EPA Office
3 of Air and Radiation/Office of Policy Analysis and Review and Richard D. Morgenstern, Associate
4 , Assistant Administrator for Policy Planning and Evaluation, U.S. EPA (currently on leave as Visiting
5 Scholar, Resources for the Future). The principal project managers are Jjm DeMocker,
« EPA/OAR/OPAR; AlMcGartland, Director, EPA/OPPE/OEE; andTomGillis,If'ATOPPE/OEE.
7 Many EPA staff contributed or reviewed portions of this draft document, including Joel Schwartz,
* Michael Shapiro, Peter Preuss, Tracey Woodruff, Diane DeWitt, Dan Axelrad, Joel Scheraga, Anne
9 Grambsch, Jenny Weinberger, Allyson Siwik, Richard Scheffe, Vasu Kilaru, Amy Vasu, Kathy
10 Kaufmann, Mary Ann Stewart, Eric Smith, Dennis J. Kotchmar, Warren Freas, Tom Braverman,
11 Bruce Polkowsky, David Mobley, Sharon Nizich, David Meisenheimer, Fred Dimmick, Harvey
12 Richmond, John Haines, John Bachmann, Ron Evans, Tom McMullen, Das Mussatti, Bill Vatavuk,
u Larry Sorrels, Dave McKee, Susan Stone, Melissa McCullough, Rosalina Rodriguez, Vickie Boothe,
H Tom Walton, Michele McKeever, Vicki Atwell, Kefly Rimer, Bob Fegley, Aparna Koppikar, Les
a Grant, Judy Graham, Robin Dennis, Dennis Leaf, Ann Watkins, Penny Carey, Joe Somers, Pam
16 Brodowicz, Byron Bunger, Allen Basala, David Lee, Bill O'Neill, Susan Herrod, and Susan
n Stendebach. Allyson Siwik of EPA/OAR/OAQPS and Bob Fegley of EPA/ORD/OSPRE played
is particularly important roles hi coordinating substantive and review contributions from then- respective
19 offices. " TI" •" "-'
-
A number of contractors developed key elements of the analysis and supporting documents. These
21 contractors include Bob Unsworth and Jim Neumann of Industrial Economics, Incorporated (lEc);
22 Leland Deck, Lisa Akeson, Brad Firlie, Susan Eeane, Kathleen Cunningham, and John Voyzey of Abt
23- Associates; Bruce Braine, Patricia Kim, Sandeep Kohli, Anne Button, Barry Galef, Cynde Sears, and
24 Tony Bansal of ICFResources; John Langstaff, Michelle Woolfolk, Shelly Eberly, Chris Emery, Till
25 Stoekenius', and Andy Gray of ICF/Systems Applications International (ICF/SAI); Dale Jorgenson, '
26 Peter Wilcoxen, and Richard Goettle oCJorgenson Associates; Jim Lockhart of the Environmental Law
27 Institute (ELI); Beverly GopMch, Rehan Aziz, Noel Roberts, and Lucille Bender of Computer
28 Sciences Corporation; Margaret Sexsmith of Analytical Sciences, Incorporated; Ken Meardon of
29 Pacific Environmental Services QPES); David South, Gale Boyd, Melanie Tomkins, and K. Guziel of
30 Argorme National Laboratory (ANL); Don Garner; Rex Brown and Jacob Ulvila of Decision Science
31 Consortium; and Jim Wilson and Dianne P. Crocker of Pechan Associates. The SARMAP AQM runs
32 were provided by Carol Bohnenkamp of EPA Region 9 and Saffet Tanrikulu of the California Air
33 Resources Board. ;>
34 Science Advisory Board review of this report is supervised by Donald G. Barnes, Director of the
15 SAB Staff. SAB staff coordinating the reviews have included Jack Kooyoomjian, Sam Rondberg, Fred
36 Talcott, and Randall Bond. Diana Pozun provided administrative support.
37 The SAB CAACAC is chaired by Richard Schmalensee of MTT. Members include Morton
3s Lippmann of New York University Medical Center, William Nordhaus of Yale University, Paul
39 Portney of Resources for the Future, Kip Viscusi of Duke University, A. Myrick Freeman of Bowdoin
College, Maureen Cropper of the World Bank, Ronald Cummings of Georgia State University, Daniel
Dudek of the Environmental Defense Fund, Robert Mendelsohn of Yale University, Wayne Kachel of
42 Martin-Lockheed Corporation, William Cooper of Michigan State University, Thomas Tietenberg of
-------
Acknowledgments
; Colby College, Paul Lioy of the Robert Wood Johnson School of Medicine, Roger McClellan of the
2 Chemical Industry Institute of Toxicology, Richard Conway of Union Carbide Corporation, and
3 Wallace Gates of the University of Maryland.
4 The SAB CAACAC Physical Effects Subcommittee is chaired by Morton, Lippmann. Members
5 included David V. Bates of the University of British Columbia, A. Myrick Freeman of Bowdoin
6 College, Gardner Brown, Jr. of the University of Washington, Timothy Larson of the University of
7 Washington, Lester Lave of Carnegie Mellon University, Joseph Meyer of the University of Wyoming,
a Robert Rowe of Hagler Bailly, Incorporated, George Taylor of the University of Nevada, Bernard
9 Weiss of the University of Rochester Medical Center, and George Wolff of the General Motors
10 Research Laboratory. • '"--"".--
// The SAB CAACAC Air Quality Subcommittee is chaired by George Wolff. Members have
12 included Benjamin Liu of the University of Minnesota, Peter Mueller of the Electric Power Research
13 Institute, Warren White of Washington University, Joe Mauderly of the Lovelace Biomedical &
14 Environmental Research Institute, Philip Hopke of Clarkson University, Paulette Middleton of Science
is Policy Associates, James H. Pierce, Jr. of the Texas Natural Resource Conservation Commission, and
n Harvey Jeffries of the University of North Carolina. Chapel Hill.
r- - f _ : . r _ - f
17 A number of interagency review meetings have been held during the course of the development of
is this analysis. Agencies and Departments,which have been represented at these meetings include the
19 Council oh Environmental Quality, the Council of Economic Advisors, the Department of Energy, the
20 National Acid Precipitation Assessment Program, the Department of Commerce, the Department of
21 Labor, and the Office of Management and Budget.
22 This report could not have beeiproduced wflfcout the support of key administrative support staff.
23 The project managers are grateful tpMona Smoke, Carolyn Hicks, Eunice Javis, Gloria Booker,
24 Thelma Butler; Wanda Farrar, Ladonya Langston, and Eileen Pritchard for their timely and tireless
25 support on this project, _-C : :^5.>
VI
-------
24
Executive Summary
In 1990, Americans received roughly 20 dollars of value in reduced
risks of death, illness, and other adverse effects for every one dollar
spent to control air pollution.
Purpose\ ofthe Study
* •
6 Throughout the history of the federal Clean Air Act, questions have been raised as to whether the
7 health and environmental benefits of air pollution control justify the costs incurred by industry,
* taxpayers, and consumers. For the most part, questions about the costs and benefits of individual
9 regulatory standards continue to be addressed during the regulatory development process through
10 Regulatory Impact Analyses (RIAs) and other analyses which evaluate regulatory costs, benefits, and
/; design variables such as scope, stringency, and timing. There have never been, however, any
n comprehensive, long-term, scientifically valid and reliable studies which answered the broader question:
•» "How do the overall health, welfare, ecological, and economic benefits of Clean Air Act
programs compare to the costs of these programs?" '
is To address this void, Congress added to the 1990 Clean Air Act Amendments a requirement under
if section 812 that EPA conduct periodic, scientifically-reviewed studies to assess the benefits and the costs
17 of the Clean Air Act. Congress further required EPA to conduct the assessments to reflect central
is tendency, or "best estimate/' assumptions rather than the conservative assumptions sometimes deemed
19 appropriate for setting protective standards.
20 This report is the first in mis ongoing series of Reports to Congress. By examining the benefits and
21 costs of the 1970 and 1977Amendments, this report finally addresses the question of the overall value of
22 America's historical investment in cleaner air. The first Prospective Study, now in progress, will
23 evaluate the benefits and costs of the 1990 Amendments.
Study
25 Designing a study to evaluate effectively the benefits and costs of the entire Clean Air Act over a 20-
26 year period throughout the U.S. was a formidable challenge, particularly given limitations in historical
27 data, scientific tools, and available resources. .As a result of these constraints, the levels of geographical,
21 industry-specific and pollutant-specific detail which could be incorporated in the assessment were
29 limited. For example, the emissions data and models available for this study supported development of
x emissions projections at the state, rather man county, level. While relying on state-level data allows for
• development of reasonable national-level estimates of the benefits and costs of the Clean Air Act, it is
* not possible use the results of this study to identify costs and benefits of the Clean Air Act for particular
Vll
-------
Executive Summary
i locations. The need to rely on national aggregate compliance expenditure data similarly precludes
2 estimation of costs and benefits by individual company or industry. Finally, atmospheric transport and
3 transformation of pollutants from one species to another (e.g., transformation of gaseous sulfur dioxide
4 to particulate sulfates) make it difficult to estimate benefits and costs by individual pollutant
5 Another important feature of the current study is that it assesses the benefits and costs of all air
6 pollution control measures implemented from 1970 to 1990 regardless of the motivation for their
/ adoption. This approach makes it unnecessary to speculate about which actions were taken by states and
i local governments in response to Clean Air Act mandates and which actions were taken for other
» reasons.
10 By allowing for acceptable limitations on the disaggregation of results by government sector,,
u industrial sector, pollutant, and location, the study could be designed and data and models could be
n selected which allowed EPA to measure the total benefits and costs of historical air pollution controls. In
u fact, even with these limitations, the present study represents the most extensive evaluation ever
14 completed of the Clean Air Act
is . The study derived the benefit and cost estimates by examining the differences in economic, human
16 health, and environmental outcomes under two alternative scenarios: a "control scenario" and a "no-
n control scenario." The control scenario reflects actual historical implementation of clean air programs
i» and is based largely on historical data. The no-control scenario ia a hypothetical scenario which reflects
19 the assumption that no air pollution controls were established beyond those in place prior to passage of
20 the 1970 Amendments. Each of the two scenarios were then evaluated by a sequence of economic,
n emissions, air quality, physical effect,economic valuation, and uncertainty models to yield the
22 differences between the scenarios in economic, human health, and environmental outcomes. Details of
23 this analytical sequence are presented in Chapter 1 and are summarized in Figure 4 of mat chapter.
24
25
Summary of Results
"" "
Direct Costs"
26 To comply with the Clean Air Act, businesses, consumers, and government entities all incurred
27 higher costs for many goods and services. The costs of providing goods and services to the economy
a were higher primarily due to requirements to install, operate, and maintain pollution abatement
29 equipment In addition, costs were incurred to design and implement governmental regulation, monitor
jo and report regulatory compliance, and invest in research and development Ultimately, these higher
31 costs of production were borne by stockholders, business owners, consumers, and taxpayers.
n To tally total compliance costs of the Clean Air Act over the period from 1970 to 1990, actual dollars
33 spent La a given year were adjusted to reflect their value in 1 990. The purpose of this adjustment was to
u correct for the effects of inflation so that expenditure levels in different years could be compared. In
15 addition, an adjustment was made to the raw expenditure data to take account of the fact that some
36 expenditures in any given year were for control equipment which provided benefits for more than one
37 year. This "annualization" adjustment, which is similar to calculating the value of a year's worth of
3s mortgage payments for a house, allowed for a comparison of costs in a given year with the value of the
39 emissions reductions achieved in that same year. Figure 1 summarizes the historical data on Clean Air
Vlll
-------
Executive Summary
2
3
4
5
6
7
a
9
ia
11
u
13
14
IS
16
Act compliance costs, adjusted both for
inflation and for the value of long-term
investments in equipment
These direct cost data show that, even on
an inflation-adjusted basis, direct costs to
meet requirements of the Clean Air Act
increased steadily over the period of the
analysis at a rate of increase equal to a little
under four percent per year. This compares
to a growth rate in Real Gross Domestic.
Product (GDP) over the same period of about
3'/4 percent. The implication of this
comparison with GDP is that air pollution
control expenditures during the study period
increased at a slightly faster rate than total
output by the U.S. economy.
Figure 1. Total Direct Costs of the CAA (in billions of
inflation-adjusted 1990 dollars.)
1973
1990
17
IS
19
21
22
23
24
25
26
27
21
29
30
31
32
33
34
35
36
37
3S
39
Indirect Economic Effects
As part of the process of developing input data to model emissions under the control and no-control
scenarios, a macroeconomic model was run to estimate changes in key economic variables such as total
electricity sales. In addition to providing economic indicators needed for the emissions modeling, the
macroeconomic model provided estimates of other, indirect economic effects of the Clean Air Act For
example, the macroeconomiQ model juns for the control and no-control scenario provided estimates of
overall economic production, prices fix goods and services from various sectors, levels and patterns of
investment, and patterns of employment
-""..- '" - -5i>-^f-'~
While these indirect effect estimatesjare reported in Chapter 2 and Appendix A, EPA concluded that
these projections do not provide a complete and satisfactory indication of the overall economic effects of
the Clean Air Act and are not presented as primary results of this analysis. The principal limitation of
the macroeconomic results istfaat^while they do reflect the indirect economic costs of air pollution
control expenditures, they do not reflect the indirect economic benefits of the pollution control achieved.
For example, the macroeconomic model results suggest mere was a decrease in total U.S. production due
to displacement of capital investment caused by pollution control expenditures; however, the
macroeconomic model fails to capture the increase in total U.S. production achieved by reductions in air
pollution-related worker illness and absenteeism. The macroeconomic model also fails to capture
expenditure reductions achieved by the Clean Air Act, such as reductions in medical expenses for air
pollution-related illness and disease, and apply those savings to offset the capital displacement effect of
compliance expenditures. This unbalanced treatment of the effects of the Clean Air Act led EPA to
conclude that the most appropriate comparisons in the context of this study are between the direct costs
and the direct benefits of the Clean Air Act, a conclusion supported by the Science Advisory Board
review panel charged with providing scientific and economic review of the study.
IX
-------
Executive Summary
2
3
4
5
6
7
I
9
10
11
12
13
14
75
16
17
18
19
20
21
22
23
24
25
26
27
21
29
30
31
32
33
34
Emissions
Emissions were substantially lower by
1990 under the control scenario than under
the no-control scenario, as shown in Figure 2.
Sulfur dioxide (SOj) emissions were 40
percent lower, primarily due to utilities
installing scrubbers and/or switching to lower
sulfur fuels. Nitrogen oxides (NOJ
emissions were 30 percent lower by 1990,
mostly because of the installation of catalytic
converters on highway vehicles. Volatile
organic compound (VOC) emissions were
also lower primarily due to motor vehicle
controls, with 1990 VOC levels 45 percent
lower under the control scenario than under
the no-control scenario. Motor vehicle
controls were also primarily responsible for a
SO percent reduction in 1990 carbon
monoxide (CO) emissions under the control
scenario. - s*>
7igure 2. 1990 Control and No-control Scenario Emissions
in millions of short tons).
g No-control
• Control
00
For particulate matter, it is important to recognize the distinction between reductions in directly
emitted particulate matter and reductions hi ambient concentrations of particulate matter in the
atmosphere. As discussed further io the next section^ changes in particulate matter air quality depend
both on changes in emissions of total pispended particulates (TSP) and on changes in emissions of
gaseous pollutants, such as sulfur djoiude^nd nitrogen oxides, which are later converted to particulate
matter through chemical transformatibiLitt the atmosphere. Emissions of primary, or directly emitted,
total suspended partiGutaiejEJwere 75 pcrejit lower under the control scenario by 1990 than under the no-
control scenario. This jubsiililial difference is primarily due to vigorous efforts in the 1970s to reduce
visible emissions from utiUt^iaidiiidustrial smokestacks.
The vast majority of the difference in lead emissions under the two scenarios is attributable to
reductions in the use of leaded gasoline. By 1990, reductions in highway vehicle emissions accounted for
221 thousand of the total=234 thousand ton reduction hi total airborne lead emissions under the control
3_ jjT ' -
scenario. The reduction in total airborne lead emissions from all sources under the control scenario was
almost99percent.p
36
37
X
39
40
41
Air Quality
The substantial reductions in air pollutant emissions achieved by the Clean Air Act translate into
significantly unproved air quality throughout the U.S. For sulfur dioxide, nitrogen oxides, and carbon
monoxide, the improvements in air quality under the control scenario were treated as proportional to the
estimated reduction in emissions. This is because, for these pollutants, changes in ambient
concentrations in a particular area are strongly related to changes in emissions in that area. While the
• differences in control and no-control scenario air quality for each of these pollutants do vary from place
-------
Executive Summary
to place because of local variability in emissions reductions, the overall national average improvements
2 in air quality for these pollutants achieved by 1990 were: 40 percent improvement for sulfur dioxides, 30
3 percent improvement for nitrogen oxides, and 50 percent improvement for carbon monoxide.
* The differences in ambient concentrations of ground-level ozone and particulate matte* under the
s two scenarios, however, are not so straightforward. Long-range transport through the atmosphere, non-
6 linear formation mechanisms, and other complexities meant that air quality models had to be used to
7 translate changes in emissions of precursors of these pollutants to changes in ambient concentrations of
s ozone or particulate matter. Estimating differences in acid deposition and visibility in the Eastern U.S.
9 also required the use of air quality models since they are not necessarily directly proportional to changes
10 in emissions of pollutants which cause these effects.
/; Reductions in ground-level ozone were achieved through reductions in its precursor pollutants,
12 particularly volatile organic compounds and nitrogen oxides. Tie differences in ambient ozone
n concentrations estimated under the control scenario varied significantly from one location to another,
H primarily because of local differences in the relative proportion of VOC and NO,, in meteorology and in
is precursor emissions reductions. On a national average basis, however, ozone concentrations in 1990
16 under the control scenario were about 1S percent lower man projected under the no-control scenario. For
n several reasons, this overall reduction in ozone is significantly less than the estimated 30 percent
is reduction in precursor nitrogen oxides and the estimated 45 percent reduction in precursor volatile
19 organic compounds. First, significant natural (i.e., biogenic) sources of VOCs limit the level of ozone
20 reduction achieved by reductions in man-made (i.e., anthropogenic) VOCs. Second, current knowledge
•>i of atmospheric photochemistry suggests that ozone reductions will tend to be proportionally smaller than
reductions in precursor emissions. Finally, the plume model system used to estimate changes in urban
23 . ozone for this study is incapable of handling long-range transport of ozone from upwind areas, complex
24 flow phenomena, and multi-day pollution events in a realistic manner. While regional scale ozone
2i models address these phenomena more effectively, resource limitations precluded using these more
u sophisticated but costly models for this study.
*" -"-a- 3L.
27 There are many pollutants which contribute to ambient concentrations of particulate matter. The
2> relative contributions of these individual pollutant species to ambient particulate matter concentrations
29 vary from one region of the country to the next, and from urban areas to rural areas. The most important
30 particle species, from a humaaheaith standpoint, are the fine particles which can be respired deep into
31 the lungs. While some fine particles are directly emitted by sources, the most important fine particle
32 species are actually formed in the atmosphere through chemical conversion of gaseous air pollutants.
33 These species are referred to as secondary particles. The three most important secondary particles are (1)
34 sulfates, which derive primarily from sulfur dioxide emissions; (2) nitrates, which derive primarily from
33 nitrogen oxide emissions; and (3) organic aerosols, which derive primarily from volatile organic
36 compound emissions. This highlights an important and unique feature of particulate matter as an
37 ambient pollutant: more than any other pollutant, reductions hi particulate matter are actually achieved
3t through reductions in a wide variety of air pollutants. In other words, controlling particulate matter
39 means controlling "air pollution" in a very broad sense.
40 Because of its comprehensive coverage of pollutants contributing to ambient particulate matter, the
41 present study is uniquely configured to illustrate the broader range of benefits achieved by historical
42 reductions in "air pollution." The results of this analysis indicate that the reductions hi sulfur dioxide,
nitrogen oxides, volatile organic compounds, and directly-emitted particles achieved by the Clean Air
XI
-------
Executive Summary
i Act resulted in an overall, national average reduction in total ambient particulate matter of about 45
2. percent by 1990.
3 Reductions in sulfur dioxide and nitrogen oxide also translate into reductions in formation, transport,
4 and deposition of secondarily formed acidic compounds such as sulfate and nitric acid. These are the
s principal pollutants responsible for acid precipitation, or "acid rain." Regional acid deposition modeling
6 of the control and no-control scenarios indicates mat, by 1990, sulfur and nitrogen deposition were
7 significantly higher under the no-control scenario throughout the 31 Eastern Steles covered by the model,
« Percentage increases in sulfur deposition under the no-control scenario ranged up to more than 40
9, percent in the upper Great Lakes and Florida-Southeast Atlantic Coast areas and were due, primarily, to
10 significant projected increases in the use of high-siulfur fuels by utilities in the upper Great Lakes and
11 Gulf Coast states. Nitrogen deposition also was significantly Higher under the no-control scenario, with
12 percentage increases reaching levels of 25 percent or higher along file Eastern Seaboard. This higher.
13 level of nitrogen deposition can be attributed primarily to higher projected emissions of nitrogen oxides
u from motor vehicles. ---*-.> =.: t r
15 Finally, decreases in ambient concentrations of light-scattering pollutants^ such as sulfates and
a nitrates, under the control scenario were estimated^ lead to perceptible improvements in visibility
n throughout the Eastern states and Southwestern urban areas modeled for mis study.
IS
Physical Effects
19 The lower ambient concentrations of sulfur dioxide, nitrogen oxides, particulate matter, carbon
20 monoxide, ozone and lead projected under the ctiNBb^l scenario yielded a substantial variety of human
21 health, welfare and ecological benefits. For a ntSber of these benefit categories, valid and reliable
22 quantitative functions were avallabielrom the scientific literature which allowed estimation of the
23 reduction in incidence of advera effects* Examples of these categories include the human mortality and
24 morbidity effects of a number of poOtitaBts^ changes in neurobehavioral effects among children caused
25 by exposure to lead and changes in visibility impairment
"
26 A number of benefit categajii^however, could not be quantified and/or monetized for a variety of
27 reasons. In some cases, strong iclentific evidence of an effect existed, but data were still too limited to
21 support quantitative estimates of incidence reduction (e.g., changes in lung function associated with
29 long-term exposure to ozone). In other cases, substantial scientific uncertainties prevailed regarding the
x existence and magnitude of adverse effect (e.g., contribution of ozone to air pollution-related mortality).
31 Finally, there were effects for which mere was sufficient information to estimate incidence reduction, but
32 for which there were no available economic value measures; thus reductions in adverse effects could not
33 be expressed in monetary terms. Examples of mis latter category include pulmonary function
34 decrements (Caused by acute exposures to ozone and reduced time to onset of angina pain caused by
3s carbon monoxide exposure.
x Table 1 provides a summary of some of the key differences in estimated human health outcomes
37 under the control and no-control scenarios. Results are presented as thousands or millions of cases
x avoided per year due to control of the pollutants listed in the table and reflect reductions estimated for
39 the entire U.S. population living hi the 48 continental states.
Xll
-------
Executive Summary
2
3
4
5
6
7
g
9
10
11
12
13
14
15
16
17
IS
19
20
21
22
23
23
26
27
21
29
30
31
32
33
34
3S
36
37
3»
39
40
41
42
43
Table 1. Selected Health Benefits of the CAA, 1970-1990 (fa
thousands of eases reduced per year, except as noted).
******
1985 1990
MortoBty
high
okl
tow
M
2ft
II
54
39
124 146
40
4$
HcftrtAtfeefc*
0**
i
i
low
9
7
9
14
10
24
18
Siroite*
(thousands)
i
r
i
5
4
10
to
f
Itapiratery tymptomt
M
165
15
Numerous scientific studies were
used as sources of the concentration-
response functions applied to
estimate differences in mortality,
morbidity, and other outcomes under
the two scenarios. The "high"
estimates presented in Table 1 reflect
the "best estimate" of the
concentration-response function with
the largest predicted effect The
"low" estimates reflect the "best
estimate" of the concentration-
response function with the smallest
predicted effect The "mid" value is
the average of the "best estimates" of
all the concentration-response
functions used to estimate each
effect1 It is important to note that
these results do not reflect the
uncertainty surrounding the "best
estimates," rather, each reported
value should be recognized as the
average value in a range of possible j
values. ae.r:
Adverse human health effects of
the Clean Air Act "criteria /I;. "/";£":-.= •
pollutants" sulfur dioxide, nitrogen ' ,HBBHMHHHBHBBBIIIBHIiBIIBIIIIBBIBIBHHBMH|^iMHHHB^
oxide, ozone, participate matter, % ''-':^-_f
carbon monoxide, and lead dominate &e quantitative estimates presented in this report in part because
knowledge of physical relationships is greatest for these pollutants. The Clean Air Act yielded other
benefits, however, which are ffijpoitant even though they may have been uncertain and/or difficult to
quantify. The other principal fan«fit categories which, for a variety of reasons, could not be
satisfactorily quantified include ill benefits accruing from reductions in hazardous air pollutants (also
referred to as air toxics); reductions in damage to cultural resources, buildings, and other materials;
reductions in adverse effects on wetland, forest, and aquatic ecosystems; and a variety of additional
human health and welfare effects of criteria pollutants. A complete list of the major benefits which
could not be monetized is provided in Table 2. .
In addition to controlling the six criteria pollutants, the 1970 and 1977 Clean Air Act Amendments
led to reductions in emissions of a small number of hazardous air pollutants. Control of these pollutants
resulted both from regulatory standards set specifically to control hazardous air pollutants and from
incidental reductions achieved through programs aimed at controlling criteria pollutants. Due to
resource and data constraints, however, reliable quantitative estimates of reductions in adverse effects of
hazardous air pollutants could not be developed for this report.
high
low
s
4
12
10
8
16
1 Results for lead (Pb) are handled differently. The lead analysis was conducted using two alternative baselines. The "high" and "low"
values repotted here reflect the results implied by the two baselines; the "mid" value is the average of the "high" and "low" values.
Xlll
-------
Executive Summary
i
2
}
4
5
6
. 7
a
9
10
n
12
13
14
IS
16
17
IS
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
31
39
40
41
42
43
Table 2. Major Nonmonetized Benefits of the Clean Air Act,
Poltotant
Benefits
Matter
Otfxat Oatsafc
IHaases
Ofcerorgaa systems)
Befcttsfe^BJ&ctt
Reductions in both hazardous air
pollutants and criteria pollutants
likely yielded widespread
improvements in the health and
quality of aquatic and terrestrial
ecosystems. In addition to any
intrinsic value to be attributed to
these ecological systems, human
welfare is enhanced through
improvements in a variety of
ecological services. For example,
protection of freshwater ecosystems
achieved through reductions in
deposition of acidic air pollutants
may have improved commercial and
recreational fishing. Other potential
ecological benefits of reduced acid
deposition include improved wildlife
viewing, maintenance of biodiversity,
and nutrient cycling. Increased
growth and productivity of U.S.
forests may have resulted from \ -
reduced emissions of ozone-formings
precursors, especially volatile - ;:
organic compounds and nitrogen-isi
oxides. Healthier, more vigorous ^
forest ecosystems m turn yield * l^
variety of benefits, including '
increased timber production;
improved forest aesthetics for people
enjoying outdoor activities
hunting, fishing, and
improvements in ecological services
such as nutrient cycling and
sequestration of global wanning
gases. Again, due to resource and
data limitations these improvements
in ecological conditions have not
been quantified in mis assessment
In considering the quantitative ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
results of this analysis, it is important
to remember the potential
significance of the benefits which could not be quantified in physical terms. This caveat is even more
significant for consideration of the monetary benefit results presented below. The range of benefits
included in the economic assessment is further reduced because some effects, such as carbon monoxide-
14*4'
ABHaman Health Effects
XIV
-------
Executive Summary
related reduction in time to onset of angina attack, could be quantified in physical terms but not valued in
economic terms.
.< Economic Valuation
•4 Estimated reductions in the incidence of adverse effects is an important, and to some people a
5 sufficient, basis for characterizing the benefits of air pollution control. However, to aggregate the full
6 range of benefits and compare them to costs, a common unit of measure must be applied. The metric
7 most commonly used for cost-benefit analysis is dollars. Assigning monetary values to all adverse
« effects which are amenable to economic valuation permits a summation of monetized benefits and
9 subsequent comparison to the costs of the Clean Air Act
w Before proceeding through this step, it is important to recognize the substantial controversies and
11 uncertainties which pervade attempts to characterize adverse human health and ecological effects of
11 pollution in dollar terms. To many, assigning dollar values to outcomes such as loss of human life, pain
13 and suffering, or ecological degradation requires value judgments which should not be left to the
14 economic analyst, or indeed to any one individual or group who would make such judgments on behalf
is of all Americans. Typically, cost-benefit analysts sidestep such value judgments by adopting the most
it technically defensible valuation estimates and leave the moral dimensions to those who must decide
n whether, and how, to use cost-benefit results in making public policy decisions. This is the paradigm
is adopted in the present study. Given the Congressional mandate to perform a cost-benefit study of the
19 Clean Air Act, the Project Team has endeavored to apply widely-recognized, customary techniques to
perform this cost-benefit analysis. Because of the implied value judgments and substantial uncertainties
21 associated with assigning dollar values to health and environmental effects, EPA strongly encourages
22 readers to look beyond the simple calculus of economic value of the Clean Air Act and consider how
23 they would value the reductions in adverse health and environmental effects estimated in this study.
24 Even when monetary values can be estimated for a given effect, such estimates are typically fraught
25 with uncertainties. Potential errors in the estimates arise for a variety of reasons. One important
26 example is the imcertamty caused[by application of monetary values derived from one circumstance
27 (e.g., wage compensation demanded for accepting riskier jobs) to a different circumstance (e.g., the value
2s of reductions in risk from exposure to air pollution). Substantial uncertainties also apply hi cases where
2» survey-based approaches, such as "contingent valuation," are used to estimate the monetary value of
x reductions in risk of an adverse effect
31 For this report, unit valuation estimates were derived from the economic literature and reported in
32 dollars per case reduced for health effects, and dollars per unit of avoided damage for welfare effects.
33 Similar to estimates of physical effects provided by health studies, the monetary values of benefits for
34 this study are reported bom in terms of mean value and, later hi the report, as a range of estimates. The
33 mean values are summarized in Table 3 for illustrative purposes, but it must be emphasized that these
36 values only reflect the central tendencies of the ranges of values used for a given effect in this analysis.
xv
-------
Executive Summary
i
2
3
4
5
6
7
I
9
10
n
12
13
14
IS
16
17
II
19
20
21
22
23
24
2!
26
27
21
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Monetized Benefits
The total monetized economic
benefit attributable to the Clean Air Act
was derived by applying the unit values
(or ranges of values) to the stream of
monetized physical effects calculated for
the 1970 to 1990 period. In developing
these estimates, steps were taken to
avoid double-counting of benefits. In
particular, it was assumed that studies
providing different measures of a given
effect were providing alternative
estimates of the same effect For
example, the Project Team had access to
several studies estimating air pollution-
related "restricted activity days" of
various degrees of severity, as well as
studies which estimated "lost work
days." To avoid double-counting, the •"-"
various "restricted activity" health ^
effects were considered to be alternative
measures for a single benefit category.
Using this approach, "restricted activity
day" benefits and **lost work day*' r *~
benefits were not counted in this r C;
analysis as separate, duplicative " • • *
outcomes. --" =^*
Table 3. Central Estimates of Economic Value per Unit of
Avoided Effect (in 1990 dollars),
Sttoto*
$4,800,000 per ewe
4587,000 p«
-------
Executive Summary
2
3
4
S
6
7
S
9
10
n
12
13
14
15
16
n
u
19
20
23
24
25
26
27
21
29
30
31
32
33
34
35
36
37
31
39
40
41
42
uncertainties associated with earlier steps in the analytical sequence would have required multiple runs
of the macroeconomic, emissions, and/or air quality models. This was far beyond the scope of the
resources available for this project. However, prior experience in regulatory analysis has demonstrated
repeatedly that the physical effects estimation and economic valuation steps contribute much more to
overall uncertainty than these earlier steps. ,
By adding up the ranges of values for each effect obtained by the procedure described above, a range
of possible economic values for the estimated differences in physical outcomes under the control and no-
control scenarios could be derived. Again, each of the possible outcomes in this range carries with it a
certain probability of occurring. Based on this probability information, and adding up the results
obtained for the entire 20 year period of the study, the following statement can be made:
The Project Team found that 90 percent of the credible estimates of the total monetized
benefits of the Clean Air Act realized daring the period from 1970 to 1990 were within the
range of 2.7 to 14.6 trillion 1990-valne dollars, with a central estimate of 6.8 trillion dollars.
However, uncertainties in compliance cost estimates, macroeconomic effects estimates,
emissions estimates, and air quality modeling resolte would be expected to broaden the
range of possible outcomes. Furthermore, inclusion of nnquantified and nonmonetized
benefits would likely raise the lower and upper bounds of the range, perhaps by highly
significant amounts. More important* the central estimate of 6.8 trillion dollars in benefits
may also be significantly underestimated due to the exclusion of large numbers of benefits
from the monetized benefit estimate. This range of estimated benefits compares to direct
costs of compliance equal to approximately 436 billion dollars in 1990-value dollars over
the same 20 year period. Subtracting total costs from total benefits, this analysis indicates
that net benefits of 1970 to 1990 Clean Air Act controls probably were within a 90 percent
credible interval of 2 J and 14 J trillion 1990-value dollars, with a central estimate of 6.4
trillion dollars. ; -
Figure 3 provides a graphical --;.-_.^^-_
representation of the estimated range of total
monetary benefits and compares mis range to
estimated compliance costs. Cfeariy, even
the lower bound estimate of monetized
benefits substantially exceeds the costs of the
historical Clean Air Act. As shown in the
more detailed, year to year results presented
later in this report, monetized benefits
consistently and substantially exceeded costs
throughout the 1970 to 1990 period.
Dividing the maximum likelihood
estimate of 1990 monetized benefits by 1990
estimated direct costs indicates that
Americans received approximately 20 dollars
of value in reduced risks of death, illness, and
other adverse effects for every dollar spent to
control air pollution. It is important to
remember that even this level of benefit does
Figure 3. Total Direct Costs and Monetized Benefits of
the Clean Air Act, 1970 to 1990 (in trillions of 1990
dollars).
Oasts
XV11
-------
Executive Summary
i not capture the additional value of reductions in hazardous air pollutants, protection of ecosystems, and a '
2 variety of other health and environmental effects which could not be quantified or monetized.
Conclusions and Futuro Directions
3 - - ...
4 First and foremost, the clear implication of these results is that Ae benefits of the Clean Air Act and
5 associated control programs substantially exceeded costs. Even considering the large number of
6 important uncertainties permeating each step of the analysis, it is extremely unlikely that the converse
7 could be true. .
« A second important implication of this study is that a large proportion of the monetized benefits of
9 the historical Clean Air Act derive from eliminating two pollutants: lead and particulate matter. Some
10 readers may argue that, while programs to control these two pollutants may have been worthwhile, many
/; other historical Clean Air Act programs would not pass a benefit-cost test when considered in isolation.
12 While mis may or may not be true, this analysis provides no evidence to support or reject such
13 conjectures. On the cost side, the historical expenditure data used in mis analysis are not structured in
14 ways which allow attribution of control costs to specific programs or standards. On the benefit side,
is most control programs yielded a variety of benefits, many of which included reductions hi other
16 pollutants such as ambient particulate matter. For example, new source performance standards for sulfur
n dioxide emissions from coal-fired utility plants yielded benefits beyond those associated with reducing
is exposures to gaseous sulfur dioxide. The reductions in sulfur dioxide emissions also led to reductions in
19 ambient fine particle sulfates, yielding human health, ecological, and visibility benefits. Even so, his
20 likely that some specific historical programs or standards may not have yielded monetized benefits in
21 excess of costs. It is inevitable mat in any large-scale public policy program, some particular outcomes
22 may yield troubling anecdotes or unfavorable outcomes. These outlier events should clearly be reviewed
23 and mitigated where possible. However, the overriding conclusion to be drawn from this analysis is mat
24 the level of investment made under Clean Air Act programs during the 1970 to 1990 period yielded huge
25 dividends.
--- --
2t Because benefits continued to exceed costs by more than an order of magnitude in 1990, it is likely
27 that substantial additional cost-beneficial protection remained to be achieved. In fact, a recent peer-
as reviewed study examining sulfate-related health benefits alone of the 1990 Amendments pertaining to
29 acid rain control found benefits well in excess of the total estimated costs of the 1990 Amendments.2
jo Examining this implication of substantial unfinished business to be addressed by the Clean Air Act
31 Amendments of 1990 is the central purpose of the first Section 812 Prospective Study now hi progress.
32 As the first in an ongoing series of benefit-cost studies of future Clean Air Act programs, this study will
33 evaluate the specific, incremental costs and benefits of new control programs initiated subsequent to the
34 1990 Amendments. As for the Retrospective Study, the first Prospective Study is designed to
31 complement, not substitute for, the regulatory analyses developed for individual programs or standards.
36 Nevertheless, the first Prospective Study will differ from the Retrospective Study in that h will be
37 stru ured to allow comparisons of the costs and benefits of major program areas, rather than just the
* tota / of the 1990 Amendments. This will be feasible largely because, unlike the Retrospective Study,
39 subs untial high-quality, program-specific benefit and cost data have already been developed which are
1 U.S. Environmental Protection Agency, Human Health Benefits from Stilfate Reductions Under Title IV of the 1990 Clean Air Act
Amendments, Office of Air and Radiation / Office of Atmospheric Programs / Acid Rain Division, November 1995.
xviii
-------
Executive Summary
. consistent with the specific control and baseline scenarios to be analyzed. Examples include the Title IV
2 sulfate health study mentioned above and the emissions inventories being developed by the Ozone
3 Transport Assessment Group.
4 Another important element of the first Prospective Study is that the Project Team will seek,
5 resources-permitting, to expand the range and depth of the benefits analysis. In particular, efforts will be
6 made to develop valid and reliable estimates of the benefits of future reductions in hazardous air
7 pollutants, one of the key program areas addressed more fully by the 1990 Amendments than by the 1970
a and 1977 Amendments. In addition, efforts will be made to provide a more thorough and informative
» assessment of the ecological effects of reductions in both criteria pollutants and hazardous air pollutants.
w EPA's view is that ecological benefits of air pollution controls were substantial, even if they could not be
// quantified for this study. In addition, ecological effects have social and political significance far in
n excess of any apparent dollar-based, economic measures of value. As such, the Project Team places a
13 high priority on improving and expanding assessment of ecological effects of air pollution controls. To
14 this end, steps have already been taken by EPA to improve coordination between the research and
is assessment programs of the Agency. -
16 Finally, the results of this Retrospective Study provide useful lessons with respect to the value and
17 the limitations of cost-benefit analysis as a toot lor evaluating environmental programs. Cost-benefit
a analysis can provide a valuable framework for organizing and evaluating information on the effects of
19 environmental programs. When used properly, cost-benefit analysis can help illuminate important
20 effects of changes in policy and can help set priorities for closing information gaps and reducing
21 uncertainty. Such proper use, however^ requires that sufficient levels of time and resources be provided
to permit careful, thorough, and technically and scientifically sound data-gathering and analysis. When
23 adequate resources are not provided, however, or when cost-benefit analyses are presented without
24 effective characterization of the uncertainties associated with the results, cost-benefit studies can be used
25 in highly misleading and damaging ways. Given the substantial uncertainties which permeate cost-
26 benefit assessment of environmental programs, as demonstrated by the broad range of estimated benefits
27 presented in this study, cost-benefit analysis is best used to inform, but not dictate, decisions related to
28 environmental protection policies, programs, and research.
XIX
-------
[This page intentionally blank]
XX
-------
2 Table 1. Selected Health Benefits of the CAA, 1970-1990 (in thousands of cases reduced per year,
3 except as noted). , , xiii
4 Table 2. Major Nonmonetized Benefits of the Clean Air Act, xiv
5 Table 3. Central Estimates of Economic Value per Unit of Avoided Effect (in 1990 dollars) xvi
& Table 4. Estimated Annual Direct CAA Compliance Costs (SbilUons). . ..;...,,,,..•. y
• 7 Table 5. Summary of Sector-Specific Emission Modeling Approaches ;...,..... -17
a Table 6. Uncertainties Associated with Emissions Modeling.«. «-..: ;.,..., »... 20
9 Table 7. Key Uncertainties Associated with Air Quality Modeling. ., * ...,.,.. 30
10 TableS. Human Health Effects of Criteria Pollutants ..... ,. 33
;/ Table9. Selected Welfare Effects of Criteria Pollutants ;.„.»„ 34
n Table 10. Percent of Population .(of the Continental US) within 50km of a monitor (or in a County with
u PMmonitors), 1970-1990 ... „ 36
H Table 11. Selected Health Benefits of the CAA, 1970-1990, for population in lower 48 States (in
is thousands of cases reduced per year, except as noted). 38
16 Table 12. Health and Welfare Effects of Hazardous Air Pollutants 40
n Table 13. Uncertainties Associated with Physical Effects Modeling, „.. 42
is Table 14. Monetized Health and Welfare Effects (1990 dollars). ,...,. 44
19 Table IS. Sensitivity of 1970 to 1990 All 48 State Population Monetized Benefits Mid-Point Estimate
20 to Inclusion of Cohort Mortality Studies (hi billions of 1990-value dollars) 54
Table 16. Estimated Reductions in Incidence of Chronic Bronchitis under the Control Scenario (in
thousands of new cases per year); ,...... * 54
23 Table 17. Sensitivity of 1970 to 1990 All 48 State Population Monetized Benefits Mid-Point Estimate
24 to Inclusion of Chronic Bronchitis (in billions of 1990-value dollars). 55
25 Table 18. Key Distinguishing Characteristics of the Jorgenson-Wilcoxen Model 60
26 Table 19. Definitions of Industries Within the J/W Model 61
27 Table 20. Estimated Capital and O^^Esi^ditures for Stetionary Soun» Air Pollution Control
2s (millions of current dollars). .....-„,-j"~ 69
29 Table 21. Estimated Recovered Costs for Stationary Source Air Pollution Control (millions of current
30 dollars) .1,,:,,,»..., 70
31 Table 22. Estimated Capital and Operation and Maintenance Expenditures for Mobile Source Air
32 Pollution Control (millions of current dollars) 71
33 Table 23. O&M Costs and Credits (millions of current dollars) 72
34 Table24. Other Air Pollution Control Expenditures 74
35 Table 25. Potential Soirees of Error and Their Effect on Total Costs of Compliance 75
36 Table26. BEAEstimates of Mobile SourceCosts 76
37 Table 27. Estimated Annual Direct CAA Compliance Expenditures (Smillions) 77
38 Table 28. Annualized Costs and Annual Compliance Expenditures, 1973-1990 ($1990 millions) 78
39 Table 29. Differences in Gross National Product Between the Control and No-control Scenarios. ... 79
40 Table 30. Difference in Personal Consumption Between the Control and No-Control Scenarios 80
4i Table 31. Percentage Difference in Energy Prices Between the Control and No-control Scenarios. .. 81
42 Table 32. Correspondence Between Process Emissions Categories Used by MSCET, Trends, and J/W
43 Industrial Sectors and Identifier Codes 91
44 Table 32 (continued). Correspondence Between Process Emissions Categories Used by MSCET,
Trends, and J/W Industrial Sectors and Identifier Codes 92
46 Table 33. Fuel Use Changes Between Control and No-control Scenarios 96
xxi
-------
. Tables
i Table 34. Difference in Control and No-control Scenario Off-Highway Mobile Source Emissions. . 100
2 Table 35. Sources of Data for Transportation Sector Control Scenario Activity Projection 107
3 Table 36. Distribution of Households by Demographic Attributes for Control Scenario 108
4 Table 37. Economic and Vehicle Usage Data for Vehicle Ownership Projection -
s Control Scenario .......; ; 109
6 Table 38. Control Scenario Personal Characteristics. 110
7 Table 39. Distribution of Households by Income Class for No-control Scenario.,..'. Ill
a Table 40. Economic and Vehicle Usage Data for Vehicle Ownership Projection-
9 No-control Scenario , ..V....... 112
10 Table 41. Percent Changes in Key Vehicle Characteristics Between the Control and No-control
11 Scenarios. ,,.«.;..,.. 113
12 Table 42. J/W Estimates of Percentage Increases in National Electricity Generation Under No-control
n Scenario. • **..*-» .121
u Table 43. Trends Source Categories and (1975 to 1985) Scaling-Factors for TSP and CO 126
is Table 44. Percentage Change in Real Energy Demand by Households from Control to No-control
i6 Scenario *„.*„... 128
'n Table 45. Percentage Change in Commercial Energy Demand from Control toHo-control Scenario. 129
is Table 46. JW Percent Differential in Economic Variables Used in CRESS. 129
a Table 47. TSP Emissions Under the Control and No-control Scenarios by Target Year (in thousands of
20 short tons) ,. .., , »-„".;. ,,=„ 130
21 Table 48. SO, Emissions Under the Control and No-control Scenarios by Target Year (in thousands of
22 short tons) ..,, ;. 130
23 Table 49. NOX Emissions Under the Control and No-control Scenarios by Target Year (in thousands of
24 short tons) »*-. -v*. 131
25 Table 50. VOC Emissions Under tfw Control and No-control Scenarios by Target Year (hi thousands of
26 short tons) ...;., .* v*f£* ;-;•;**• 131
27 Table 51. CO Emissions Under me Control and No-control Scenarios by Target Year (in thousands of
21 short tons). »...,. *..»»/4%, • • 132
29 Table 52. Lead (Pb) Emissions Under tite Control and No-control Scenarios by Target Year (in
x thousands of short tons). , Sip 132
31 Table 53. Summary of CO Monitoring Data. 138
32 Table 54. Format of Air Qualfry Profile Databases 140
33 Table 55. Summary of SC^sMonitoring Data. 143
34 Table 56. Summary of NOj Monitoring Data. 145
35 Table 57. Summary of NO Monitoring Data. 145
36 Table 58. Summary of TSP Monitoring Data. .. 155
37 Table 59. Summary of PM,0 Monitoring Data. .155
m Table <50. Fine Particle (PMU) Chemical Composition by U.S. Region 156
39 Table €1. Coarse Particle (PMU to PMIO) Chemical Composition by U.S. Region 157
« Table 62. PM Control Scenario Air Quality Profile Filenames 158
41 Table 63. PM No-Control Scenario Air Quality Profile Filenames 158
42 Table64. Urban Areas Modeled with OZ3PM4. 161
u Table 65. Summary of Ozone Monitoring Data. 163
u Table 66. Apportionment of Emissions Inventories for SAQM Runs. 164
45 Table 67. 1990 Control Scenario. Visibility Conditions for 30 Southwestern U.S. Cities. 170
46 Table 68. 1990 No-control Scenario Visibility Conditions for 30 Southwestern U.S. Cities. 171
47 Table 69: Summary of Relative Change in Visual Range and DeciView Between 1990 Control and No-
4t control Scenario Visibility Conditions for 30 Southwestern U.S. Cities 174
xxii
-------
•••«*
•f
Tables
Table 70. Summary of Dose Response Functions for Ozone .... 139
2 Table 71. Summary of Dose Response Functions for Particulate Matter 198
3 Table 72. Summary of Dose Response Functions for NO2 207
4 Table 73. Summary of Dose Response Functions for Carbon Monoxide . 209
s Table 74. Summary of Dose Response Functions for Sulfur Dioxide ; , 211
6 Table 75. Summary of Concentration-Response Functions for Visibility* 213
r Table 76. Criteria Air Pollutant Monitors in the U.S., 1970 -1990 .;... 217
s Table 77. Population Coverage in the "Within 50 km" Model Runs (percent of continental U.S.
9 population) , ,.»,..,.. 218
10 Table 78. Population Coverage for "Extrapolated to All U.S." Model Runs (percent of continental U."S.
;; population) , , ....,,.««,. 218
n Table 79. Human Health Effects of Criteria Pollutants ,,.,, 220
u Table 80. Quantified Benefits Which Could Not Be Monetized - Extrapolated to the Entire 48 State
u Population. «,..„, 222
is Table 81. Criteria Pollutants Health Effects - Population Within 50 KM of a Monitor Modeled (cases
16 peryear) «->._. 223
n Table 82. Criteria Pollutants Health Effects - Extrapolated to Entire U.S. Population (cases per year).
it ;...,,.„.., ..224
19 Table 83. Results from Long-Term Exposure Studies (cases peryear). .., 225
» Table 84. Summary of Biological Changes with Surface Water Acidification 233
21 Table 85. Comparison of Population of Acidic National Surface Water Survey (NSWS) by Chemical
22 Category .„-";;- 235
v Table 86. Results from Benefits Assessments of Aquatic Ecosystem Use Values from Acid Deposition
Avoidance ^ -^ . „.. 238
is Table 87. Agriculture Exposure-Response Functions. ,. 260
26 Table 88. Relative No-control to Control Percent Yield Change (harvested acres) for the Minimum
27 • Scenario. .:.„».,..»•*« 262
2« Table 89. Relative No-control to Control Percent Yield Change (harvested acres) for the Maximum
29 Scenario. .,,»^iys*-, -*,_,§•?*5*?- .• 263
x Table 90. Change in FannlProgram Payments, Net Crop Income, Consumer Surplus, and Net Surplus
31 Due to the CAA (millions nominal $). .265
32 Table 9L AGSIM Output: Barley Acreage 266
33 Table 92. AGSIM Output- Barley Production, Supply, and Stocks 267
34 Tabte93. AGSIM Output; Barley Prices 268
15 Table 94. AGSIM Output: Corn Acreage (Minimum Scenario) 269
36 Table 95. AGSIM Output Corn Acreage (Maximum Scenario) 270
37 Table 96. AGSIM Output: Corn Production, Supply, and Stocks (Minimum Scenario) 271
a Table 97. AGSIM Output: Corn Production, Supply, and Stocks (Maximum Scenario) 272
39 Table 98. AGSIM Output: Corn Prices (Minimum Scenario). 273
40 Tabte99k AGSIM Output: Corn Prices (Maximum Scenario) 274
4i Table 100. AGSIM Output: Cotton Acreage (Minimum Scenario) 275
42 Table 101. AGSIM Output: Cotton Acreage (Maximum Scenario) 276
43 Table 102. AGSIM Output: Cotton Lint Production, Supply, and Stocks (Minimum Scenario). ... 277
44 Table 103. AGSIM Output: Cotton Lint Production, Supply, and Stocks (Maximum Scenario). ... 278
45 Table 104. AGSIM Output: Cotton Lint Prices (Minimum Scenario). 279
46 Table 105. AGSIMOutput: Cotton Lint Prices (Maximum Scenario). 280
Table 106. AGSIM Output: Cottonseed Production, Supply, and Stocks (Minimum Scenario) 281
« Table 107. AGSIM Output: Cottonseed Production, Supply, and Stocks (Maximum Scenario). ... 282
XXlll
-------
. _ . _ _ Tables
i Table 108. AGSIM, Output: Cottonseed Prices (Minimum Scenario) ....... ............. . ____ 283
2 Table 109. AGSIM Output: Cottonseed Prices (Maximum Scenario) ...... -. ....... ...... . ____ 284
3 Table 110. AGSIM Output: Peanut Acreage. ...'... .......... ..... .................... ... 285
4 Table 111. AGSIM Output: Peanuts Production, Supply, and Stocks ........... ..... , , ________ 286
j Table 112. AGSIMOutput: Peanut Prices. . .............................. ; ____ \ ........ 287
6 Table 1 13, AGSIM Output: Soybean Acreage (Minimum Scenario). ...... '.. ............... . . 288
7 Table 1 14. AGSIM Output: Soybean Acreage (Maximum Scenario) ......... , . . ; ..... ....... 289
* Table 1 IS. AGSIM Output: Soybeans Production, Supply, and Stocks (Minimum Scenario) ...... 290
9 Table 1 16. AGSIM Output: Soybeans Production, Supply, and Stocks (Maximum Scenario) ...... 291
to Table 117. AGSIM Output: Soybean Prices (Minimum Scenario) ....... .- ..... !..",«. .„,., ..... 292
11 Table 1 18. AGSIM Output: Soybean Prices (Maximum Scenario). . . . . . .......... , . ..... .-, . . 293
12 Table 119. AGSIM Output: Sorghum Acreage ......... »»-, ",:,:, . ^ ............... ;„.-*.".... 294
n Table 120. AGSIMOutput: Sorghum Production, Supply, and Stocks, ................ ....... 295
14 Table 121. AGSIMOutput: Sorghum Prices ................ ,......-.,. ................... . . 296
is Table 122. AGSIM Output: Wheat Acreage (Minimum Scenario). , *-,..,,. .: ................... 297
16 . Table 123. AGSIMOutput: Wheat Acreage (Maximum Scenario). ..... ---- .:.. ............... 298
i? Table 124. AGSIM Output Wheat Production, Supply, and Stocks (Minimum Scenario) ......... 299
is Table 125. AGSIM Output Wheat Production, Supply, and Stocks (Maximum Scenario) ...... . . 300
19 Table 126. AGSIM Output: Wheat Prices (Minimum Scenario), ... . . .„. .................... 301
20 Table 127. AGSIM Output: Wheat Prices (Maximum Scenario)* i,,;. ..................... 302
21 Table 128. Quantified and Unquantified Health Effects of Lead. . . . . . ....................... 305
22 Table 129. Elements of Piecewise Linear Function for Estimating Probability of IQ < 70 as a Function
u of Blood Lead (PbB) Range. ...^Jrf~ ---- ±ff ....... .-/. ........................... .. 312
i4 Table 130. Air Modeling Parameters. . ...... ..gi ..... ., .]* ........... . ................... 334
25 Table 131. Estimated Indirect Intake Slopes: fncmoent of Blood Lead Concentration (in ug/dL) per
26 Unit of AirLead Concentration (ug/m3). ^^-^'. ............. . . . . .................. 337
27 Table 1 32. Estimated Lead Emissions from Electric Utilities, Industrial Processes, and Industrial
2t Combustion (in Tons). . . . .-^^^j^^; .............................. ............... 340
29 Table 133. Yearly Differences m Number tif Health Effects Between the Controlled and Uncontrolled
x Scenarios: Industrial Processes, Boilei^ and Electric UtilMes (Holding Omer I^ead Sources at
31 ConstantaS^Ol^^tyjIr, ---- /:f ........................... ..... ... ............ 341
n Table 134. Yearly Diffe^c^ m|Number of Health Effects Between the Controlled and Uncontrolled
33 Scenarios: Industrial Processes, Boilers, and Electric Utilities (Holding Other Lead Sources at
34 Constant 1990 Levels)-^, .^f ................................. . . ........... . -------- 342
15 Table 135. Lead Burned in Gasoline (in tons) ............. . ............ ...... .......... . . 346
36 Table 136. Yearly Differences hi Number of Health Effects Between the Controlled and Uncontrolled
37 Scenarios: Lead in Gasoline only (Holding Other Lead Sources at Constant 1970 Levels) ...... 347
3s Table 137. YearlyJMfferences in Number of Health Effects Between the Controlled and Uncontrolled
39 Scenarios: Lead hi Gasoline only (Holding Other Lead Sources at Constant 1990 Levels). .... 348
40 Table 138. Health and Welfare Effects of Hazardous Air Pollutants ................. . ........ 356
4i Table 139. Cancer Incidence Reductions and Monetized Benefits for NESHAPs ................ 360
42 Table 140. Unit Values Used for Economically Valuing Endpoints .......... ........... ..... 376
43 Table 141 . Monte Carlo Simulation Model Results for Target Years, Plus Present Value hi 1990 Terms
44 of Total Monetized Benefits for Entire 1970 to 1990 Period (hi billions of 1990-value dollars). . 385
45 Table 142. Criteria Pollutants Health and Welfare Benefits - Population Within 50KM of a Monitor
46 Modeled (in millions of 1990 dollars per year) .................. . . ..... ---- ........... 386
47 Table 143. Criteria Pollutants Heajth and Welfare Benefits - Extrapolated to Entire 48 State Population
48 (in millions of 1990 dollars per year) ......................... ..... .............. .... 3S"7
xxiv
-------
2 Figure 1. Total Direct Costs of the CAA (in billions of inflation-adjusted 1990 dollars.) ..,., ix
3 Figure 2. 1990 Control and No-control Scenario Emissions (in millions of short tons). x
4 Figure 3. Total Direct Costs and Monetized Benefits of the Clean Air Act, 1970 to 1990 (in trillions of
3 " . 1990dollars) „ ........ ;^ xvii
6 Figure 4. Summary of Analytical Sequence and Modeled versus Historical Data Basis. 5
7 Figure 5. Control and No-control Scenario Total SO, Emission Estimates ,,, 18
« Figure 6. Control and No-control Scenario Total NO, Emission Estimates. 18
9 Figure 7. Control and No-control Scenario Total VOC Emission Estimates 18
10 Figure 8. Control and No-control Scenario Total CO Emission Estimates 19
// Figure 9. Control and No-control Scenario Total TSP Emission Estimates 19
n Figure 10. Control and No-control Scenario Total Pb Emission Estimates. 19
13 Figure 11. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
14 Average CO Concentrations, by Monitor ,,. *.. ...„,,,.,* 24
is Figure 12. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
a Average SO2 Concentrations, by Monitor. .,*«,..."«,.,•„» 25
n Figure 13. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
is Average NO2 Concentrations, by Monitor. •.. „ i--i:,.,, 26
19 Figure 14. Distribution of 1990 County-Level Annual Mean TSP CAA to No-CAA Ratios 26
20 Figure 15. Distribution of 1990 Control to No-control OZIPM4 Simulated Peak Ozone Ratios 27
Figure 16. Distribution of 1990 Control to No-control RADM Simulated Ozone Ratios 27
Figure 17. Distribution of 1990 Control to Ncncontrol SAQM Simulated Ozone Ratios 27
23 Figure 18. RADM-Predicted Percent Increase in Total Sulfur Deposition (Wet + Dry) Under the No-
.24 control Scenario »;;t;,^>f», ;,.";.«.. .;..... 28
25 Figure 19. RADM-Predicted Percentlncrease in Total Nitrogen Deposition (Wet + Dry) Under the No-
2& control Scenario, -*, ^^iv.Vii,^. . 28
27 Figure 20. RADM-Bredicted Increase m Visibility Degradation, Expressed mDeciVi^^
28 Visibility Conditions (90th Percentile) Under the No-control Scenario . 29
29 Figure 21. Aggregate Dii^ Costs and Monetized Benefits of the Clean Air Act (in billions of 1990
30 dollars). »* ^f-ffae^^e • • • < 48
31 Figure 22. Comparison of Total 1970 to 1990 Clean Air Act Monetized Benefits and Direct Costs (in
32 billions of 1990 dollars and in 1990 Present Value terms). 49
33 Figure 23. Distribution of 1990 Monetized Benefits of CAA (in billions of 1990 dollars) 52
34 Figure 24. Percent Difference in Real Investment Between Control and No-control Scenarios. 82
33 Figure 25. Percent Difference in Price of Output by Sector Between Control and No-control Scenario for
36 1990,-...... ^, 83
37 Figure 26. Percent Difference in Quantity of Output by Sector Between Control and No-control Scenario
38 for 1990 ; 83
39 Figure 27. Percent Difference in Employment by Sector Between Control and No-control Scenario for
w 1990 84
4i Figure 28. Comparison of Control, No-control, and Trends SO, Emission Estimates. 87
42 Figure 29. Comparison of Control, No-control, and Trends NOX Emission Estimates. 87
43 Figure 30. Comparison of Control, No-control, and Trends VOC Emission Estimates. 88
44 Figure 31. Comparison of Control, No-control, and Trends CO Emission Estimates 88
Figure 32. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
46 Average CO Concentrations, by Monitor. 141
XXV
-------
Figures
i Figure 33. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
2 Average SO2 Concentrations, by Monitor. .. 144
3 Figure 34. Frequency Distribution for 1990 Control to No-control Scenario 95th Percentile 1-Hour
4 Average NO2 Concentrations, by Monitor ,... 146
5 Figure 35. Location of the High Resolution RADM20-km Grid Nested Inside the 80-kmRADM
6 Domain '. ;, 148
7 Figure 36. RADM-Predicted 1990 Total Sulfur Deposition (Wet + Dry; in kg/ha) Under the Control
i Scenario ,-.'. ;. 149
9 Figure 37. RADM-Predicted 1990 Total Nitrogen Deposition (Wet + Dry; in kg/ha) Under the Control
10 Scenario. .=.. ,"*...... 150
n Figure 38. RADM-Predicted 1990 Total Sulfur Deposition (Wet + Dry; in kg/ha) Underthe No-control
a Scenario ,*,,,,...., .-, ^ . 150
n Figure 39. RADM-Predicted 1990 Total Nitrogen Deposition{Wet+Dry; in kg/ha) Under the No-
14 control Scenario. .".»-.-., 151
is Figure 40. RADM-Predicted Percent Increase in Total Sulfur Deposition (Wet + Dry; in kg/ha) Under
16 the No-control Scenario ,.„.,.,„** 151
17 Figure 41. RADM-Predicted Percent Increase in Total Nitrogen Deposition (Wet + Dry; in kg/ha) Under
IB the No-control Scenario „., ;- 152
19 Figure 42. Counties with Annual Mean TSP Concentrations > 120 ug/m3, Expressed as a Percent of the
20 Number of Counties with TSP Monitors. ,. ., „„„£: 159
21 Figure 43. Counties with 2nd High 24-Hour TSP Concentrations >260 ug/m3, Expressed as a Percent of
22 the Number of Counties with TSP Monitors. -.. 159
23 Figure 44. Distribution of County-Lewd Annual Mean TSP CAA to No-CAA Ratios 159*
24 Figure 45. RADM and SAQM Modeling Domains, with Rural Ozone Monitor Locations 162
21 Figure 46. Distribution of 1990 Control to No-control QZJPM4 Simulated Peak Ozone Ratios 165
26 Figure 47. Distribution of 1990 Control to No-control RADM-Simulated Rural Ozone Ratios 166
27 Figure 48, Distribution of 1990 Control to No-control SAQM-Simulated Ozone Ratios 166
2s Figure 49. RADM-Predicted Visibility Degradation, Expressed in Annual Average DeciView, for Poor
29 Visibility additions (90th Perceatile) Under the Control Scenario 169
» Figure 50. RADM-Predicted Visibility Degradation, Expressed in Annual Average DeciView, for Poor
31 Visibility Conditions (90tfi Percentile) Under the No-control Scenario 172
12 Figure 51. RADM-Predicted Increase in Visibility Degradation, Expressed in Annual Average
33 DeciView, for Poor VislblppCbnditions (90th Percentile) Under the No-control Scenario. .... 173
34 Figure 52. PES Estimated Reductions in HAP-Related Cancer Cases. 363
33 Figure 53. ICF Estimated Reductions in Total HAP-Related Cancer Cases Using Upper Bound Asbestos
36 Incidence and Lower Bound Non-Asbestos HAP Incidence 365
37 Figure 54. ICF Estimated Reduction in Total HAP-Related Cancer Cases Using Upper Bound Incidence
a forAUHAPs.*>'. 365
39 Figure 55. National Annual Average Motor Vehicle HAP Exposures (ug/m3) 368
XXVI
-------
Acronyms and Abbreviations
2
3
4
S
6
7
a
9
10
12
12
13
14
IS
16
17
19
20
21
22
23
**
2S
26
27
28
29
30
31
32
33
34
33
CERL
CEUM
CHD
micro equivalents per liter
micrograms per cubic meter
micrograms
micrometers, also referred to as microns
AGricultural Simulation Model
EPA Aerometric Information Retrieval System
aluminum
acid neutralizing capacity
Argonne National Laboratories
Argonne Power Plant Inventory
Air Quality Control Region
Argonne Utility Simulation Model
Acid Stress Index
Air Toxic Exposure and Risk Information System
Aggregate Timberland Assessment System
Advanced Utility Simulation Model
Bureau of Economic Analysis
total light extinction „ .
Block Group/Enumeration District
atherothrombotic brain infarction
Background Information Document
faJood pressure >_^-
British Thermal Unit
confidence interval
cerebrovascular accident
Clean Air Act
Clean Air Act Amendments of 1990
SAB Clean Air Act Compliance Analysis Council
SAB CAACAC Physical Effects Subcommittee
EPA's Criteria Air Pollutant Modeling System
California Air Resources Board
SAB Clean Air Scientific Advisory Committee
Centers for Disease Control (now CDCP, Centers for Disease Control and
Prevention)
EPA/ORD Corvallis Environmental Research Laboratory (old name; see NERL)
ICF Coal and Electric Utility Model
coronary heart disease
xxvii
-------
Acronyms and Abbreviations
/
2
3
4 •
i
6
7
a
9
10
11
12
13
.1*
15
16
17
11
19
20
21
22
23
24
25
27
28
29
30
31
32
33
34
33
36
37
31
if
40
CIPP
CO
CQ2
COH
i
COHb
COPD
CPUE
CR
CRESS
.CSTM
CTG
CV
CVM
D.C.
DBF
DDE
DDT
DFEV,
dL
DOC
DOE
DOI
DRI
dV
n\/QAX>f
U V d^uVl
EC
J^ViS -
EDB
EDC
EFI
El
EIA
EKMA
ELI
EOL
EPA
EPRI
ESEERCO
ESP
FERC
FEV,
changes in production processes
carbon monoxide
carbon dioxide
coefficient of haze
blood level of carboxyhemoglobin
chronic obstructive pulmonary disease
catch per unit effort .
concentration-response ' .
Commercial and Residential Simulation System model
Coal Supply and Transportation Model
Control Techniques Guidelines
contingent valuation
contingent valuation method . -
District of Columbia
diastolic blood pressure
dichlorodiphenyldichloroemylene
dichlorodipheriyltrichloroethane
decrement of forced expiratory volume (in one second)
deciliter
Department of Commerce X,\
Department of Energy V
Department of Interior --_
~ :- Data Resources^^biCQrporated
= - . _~_-^ i_ -- - -- - : -- * • •
tteaR(iew Hazeindwc^
-- T%f"- - " - i ir "i.-*"™t •°i<» t * 1 1 »• »» i- i
jjisaggregate vemcie^btocK Allocation Model
ftTfriiicHnBr cneflfi r i«nt
ethyleneswibttsnide
emylene dienloride
Electronic Fuel Injection
Electronic Ignition
Energy Information Administration
Empirical Kinetics Modeling Approach
Environmental Law Institute
end-of-line
Environmental Protection Agency
Electric Power Research Institute
Empire State Electric Energy Research Corporation
electrostatic precipitator
Federal Energy Regulatory Commission
forced expiratory volume (in one second)
XXV111
-------
Acronyms and Abbreviations
2
3
4
5
6
7
I
9
10
11
12
13
14
15
16
17
IS
19
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
40
FGD flue gas desulfurization
FHWA Federal Highway Administration
FIFRA Federal Insecticide, Fungicide, and Rodenticide Act
FIP Federal Information Processing System
FR Federal Register
FRP Forest Response Program
GDP gross domestic product
GEMS Graphical Exposure Modeling System •
GM geometric mean / ,
GNP Gross National Product
GSD geometric standard deviation
H2SO4 sulfuricacid
ha hectares
HAP Hazardous Air Pollutant .
HAPEM-MS Hazardous Air Pollutant Exposure Model-Mobile Source
HNO3 nitric acid
hp horsepower . : ,:---•
HTCM HedcmicTravel-C^Model ,
ICARUS Investigation of Costs and Reliability in Utility Systems
ICD-9 International Classification of Diseases, Ninth Version (1975 Revision)
ICE Industrial Combustion Emissions model
lEc Industrial Economics, Incorporated
IEUBK . EPA's Integrated Exposure Uptake Biokinetic model
IMS . Integrated Model Sef%,
IFF iterative proportional fitting
IQ intelligence quotient
ISCLT Indiistrial Source Complex Long Term air quality model
J/W Jorgenson/Wilcoxen .
kg kilograms
km kilometers
£~
Ibs pounds
LRI lower respiratory illness
m/s meters per second
m meters
m3 cubic meters
Mm megameters
MMBTU million BTU
MOBILESa EPA's mobile source emission factor model
mpg miles per gallon
Mpls Minneapolis
xxix
-------
Acronyms and Abbreviations
;
2
3
4
S
6
7
S
9
10
11
n
13
14
IS
16
17
11
19 • -
20
21
22
23
24
23
26
27
21
29
30
31
32
33
34
33
36
37
3S
39
40
MRAD
MSCET
MTD
MVATS
MVMA
Mwe
,N
NA
NAAQS
NAPAP
NARSTO
NATICH
NCLAN
NBA
NERA
NERC
NERL
NESHAP
NHANES
NHANES H
NIPA
NMOCs
NO
NO2
NO3-
NOX
NPTS
NSPS
NSWS
O&M
PJ
OAQPS
OAR
QMS
OPAR
OPPE
ORD
OZIPM4
PACE
PAN
minor restricted activity day
Month and State Current Emission Trends
metric tons per day
EPA's Motor Vehicle-Related Air Toxics Study
Motor Vehicle Manufacturers Association
megawatt equivalent .
nitrogen
not available -
National Ambient Air Quality Standard .
•VT .• « A • J »» • •. i- « „-!•»
National Acid rrecipitation Assessment rrogram
North American Research Strategy for Tropospheric Ozone
National Air Toxics Information Clearinghouse;
National Crop Loss Assessment Network
National Energy Accounts
National Economic Research Associates
North American Electric Reliability Council
EPA/ORD National Exposure Research Laboratory (new name for CERL)
National Emission Standard for Hazardous Air Pollutants
First National HeaHh and Nutrition Examination Survey
Second National Health and Nutrition Examination Survey
National Income and Product Accounts
nonmethane organic compounds
nitric oxides _^^_
nitrogen dioxide v .
nitrate km -: -
nitrogen oxides
Nationwide Personal Transportation Survey
New Source Performance Standards
S
National Surface Water Survey
operating and maintenance
•a- a"
ozone
EPA/OAR Office of Air Quality Planning and Standards
EPA Office of Air and Radiation
EPA/OAR Office of Mobile Sources
EPA/OAR Office of Policy Analysis and Review
EPA Office of Policy Planning and Evaluation
EPA Office of Research and Development
Ozone Isopleth Plotting with Optional Mechanism-IV
Pollution Abatement Costs and Expenditures survey
peroxyacetyl nitrate
XXX
-------
Acronyms and Abbreviations
2
3
4
!
6
7
S
9
10
11
12
13
14
15
16
17
IS
'9
11
22
23
24
23
26
27
28
29
X
31
32
33
34
35
36
37
40
PAPE
Pb
PbB
PCS
PES
pH
PIC
PM10
POP
ppb
PPH
pphm
ppm
PPRG
PRYL
PURHAPS
pVC
r2
RAD
RADM
RADM/EM
RAMC
RIA
ROM
RRAD
RUM
s,e.
SAB
SAI
SAQM
SARA
SARMAP
SCC
SEDS
SIC
SEP
SJVAQS
Pollution Abatement Plant and Equipment survey
lead
blood lead level
polychlorinated biphenyl
Pacific Environmental Services
the logarithm of the reciprocal of hydrogen ion concentration, a measure of acidity
product of incomplete combustion
particulates less than or equal to 10 microns in aerometric diameter
particulates less than or equal to 2.5 microns in aerometric diameter
population
exposed population of exercising mild asthmatics
exposed population of exercising moderate asthmatics
parts per billion
people per household
parts per hundred million
parts per million '
Pooling Project Research Group
percentage relative yield loss
PURchased Heat And Power -
polyvinyl chlojlle
statistical correlation coefficient, squared
restricted activity day
Regional Acid Deposition Model
Resource Allocation and Mine Costing model
Regulatory Impact Analysis
Regional Oxidant Model
respiratory; restricted activity day
Random Utility Model
standard error
Science Advisory Board
Systems Applications International
SARMAP Air QuaUty Model
Superfund Amendment Reauthorization Act
SJVAQS/AUSPEX Regional Modeling Adaptation Project
Source Classification Code
State Energy Data System
'Standard Industrial Classification
State Implementation Plan
San Joaquin Valley Air Quality Study
xxxi
-------
Acronyms and Abbreviations
i
2 '
3
4
1
6
7
t
9
10
11
12
13
14
11
16
17
IS
19
20
21
22
23
SMSA Standard Metropolitan Statistical Area
SO2 sulfur dioxide
SO^ sulfateion
SOS/T State of Science and Technology (refers to a series of NAPAP reports)
SRaw Specific Airway Resistance
STAR Stability Array weather database
TAMM90 Timber Assessment Market Model (revised version)
TEEMS Transportation Energy and Emissions Modeling System
TIUS Truck Inventory and Use Surveys
TRI . Toxic Release Inventory
TSP total suspended particulate
U.S. United States
UAM Urban Airshed Model
URI upper respiratory illness
USDA Department of Agriculture
VC vinyl chloride .
VMT vehicle miles traveled T .
VOC volatile organic compounds
VOP Vehicle Ownership Projection
VR visual range
W126 mdex of peak weighted average of cumulative ozone concentrations
WLD Work Loss Day
WtP willingness to pay
xxxu
-------
1
Introduction
Background and Purpose
4. As part of the Clean Air Act Amendments of 1990, Congress established a requirement under
s Section 812 that EPA develop periodic Reports to Congress estimating the benefits and costs of the
6 Clean Air Act itself. The first such report was to be a retrospective analysis, with a series of prospective
7 analyses to follow every two years thereafter. This report represents the retrospective study, covering the
s period beginning with passage of the watershed Clean Air Act Amendments of 1970, until 1990 when
9 Congress enacted the most recent comprehensive amendments to the Act
'« Since the legislative history associated with section 812 is sparse, there is considerable uncertainty
regarding Congressional intent behind the requirement for periodic cost-benefit evaluations of the Clean
12 Air Act (CAA). However, EPA believes the principal goal of these amendments was that EPA should
13 develop, and periodically exercise, the ability to provide Congress and the public with up-to-date,
14 comprehensive information about the economic costs, economic benefits, and health, welfare, and
is ecological effects of CAA programs. The results of such analyses might then provide useful information
16 for refmement of CAA programs during future reauthorizations of the Act
17 The retrospective analysis presented in mis Report to Congress has been designed to provide ah
is unprecedented examination of ^he overall costs and benefits of the historical Clean Air Act. Many other
19 analyses have attempted to identify die isolated effects of individual standards or programs, but no
20 analysis with the present degree of validity, breadth and integration has ever been successfully
21 developed. Despite data limitations, considerable scientific uncertainties, and severe resource
22 constraints; the EPA Project Team was able to develop a nearly complete picture of the costs and
23 benefits associated with the major CAA programs of the 1970 to 1990 period. Beyond the statutory
24 goals of Section 812, EPA intends to use the results of this study to help support decisions on future
25 investments in air pollution research. Finally, many of the methodologies and modeling systems
26 developed for the Retrospective Study may be applied in the future to the ongoing series of section 812
27 prospective studies.
-------
Chapter 1: Introduction
Clean Air Act Requirements, 197O to 199O
2 The Clean Air Act establishes a framework for the attainment and maintenance of clean and
3 healthful air quality levels. The modern Clean Air Act was enacted in 1970 and amended twice ~ in
4
27
1977 and most recently in 1990.
s The 1970 Clean Air Act contained a number of key provisions. First, EPA was directed to establish
6 national ambient air quality standards for the major criteria air pollutants. The states were required to
7 develop implementation plans describing how they would control emission limits from individual
« sources to meet and maintain the national standards. Second, the 1970 CAA contained deadlines and
9 strengthened enforcement of emission limitations and state plans with measures involving both the states
10 and the federal government Third, the 1970 Act forced new sources to meet standards based on the best
11 available technology. Finally, the Clean Air Act of 1970 addressed hazardous pollutants and automobile
n exhausts.
a The 1977 Clean Air Act Amendments also set new requirements on clean areas already in attainment
14 with the national ambient air quality standards. In addition, the 1977 Amendments set out provisions to
is help areas that failed to comply with deadlines for achievement of th« national ambient air quality
16 standards. For example, pennits for new major sources and modifications were required.
n The 1990 Clean Air Act Amendments considerably strengthened the earlier versions of the Act
is With respect to nonattainment, the Act set forth a detailed and graduated program, reflecting the fact that
19 problems in some areas are mare difficult and complex man others. The 1990 Act also established a list
20 of 189 regulated hazardous air pollotants and ft multi-step program for controlling emissions of these
21 toxic air poUutmte Signified ^
22 precursors and stratospheric ozon&4epieting chemicals. The biggest regulatory procedural change hi the
23 Act is the new permit program whem^&major sources are now required to obtain an operating permit
24 Finally, the amendments considerably expanded the enforcement provisions of the Clean Air Act with
25 increased administrative penalties and die role of citizen suits.
. .:.:.. :;::::.- *- I !_,,-„
26 Section O12 of the Clean Air Aot Amendment*
Of 199O
n Section 812 of (he Clean Air Act Amendments of 1990 requires the EPA to perform a "retrospective"
29 analysis which will assess the costs and benefits to the public health, economy and the environment of
» legislation before me 1990 amendments. Section 812 directs that EPA shall measure me effects on
31 "employment, productivity, cost of living, economic growth, and the overall economy of the United
32 States" of the Clean Air Act Section 812 also requires that EPA consider all of the economic, public
33 health, and environmental benefits of efforts to comply with air pollution standards. Finally, section 812
34 requires EPA to evaluate the prospective costs and benefits of the Clean Air Act every two years.
-------
Chapter 1: Introduction
Analytical Design and Review
i
2 Target Variable
-3 The retrospective analysis was designed to answer the following question:
4 "How do the overall health, welfare, ecological, and economic benefits of Clean Air
s Act programs compare to the costs of these programs?"
6 By examining the overall effects of the Clean Air Act, this analysis complements the Regulatory
7 Impact Analyses (RIAs) developed by EPA over the years to evaluate individual regulations. Resources
s were used more efficiently by recognizing that these RIAs, and other EPA analyses, provide complete
9 information about the costs and benefits of specific rules. _. > "
10 Focusing on the broader target variables of "overall costs" and "overall benefits" of the Clean Air
n Act, the EPA Project Team adopted an approach based onxxmstruction and comparison of two distinct
a scenarios: a "no-control scenario" and a "control scenario." The no-control scenario essentially freezes
13 federal, state, and local air pollution controls at the levels of stringency and effectiveness which
14 prevailed in 1970. The control scenario assumes that all federal, state, and local rules promulgated
15 pursuant to, or in support of, the CAA during 1970 to 1990 were enacted.3 This analysis men estimates
•I the differences between the economic and environmental outcomes associated with these two scenarios.
17 Key Assumptions ;>
is Two key assumptions were made during the scenario design process to avoid miring the analytical
19 process in endless speculation. First, the %o-control" scenario was defined to reflect the assumption that
20 no additional air pollution controls were imposed by any level of government after 1970. Second, it is
21 assumed that the geograpli&distribution of population and economic activity remains the same between
22 the two scenarios.
23 The first assumption is an obvious simplification. In the absence of the CAA, one would expect to
24 see some air pollution abatement activity, either voluntary or due to state or local regulation. It is
25 conceivable that state ami local regulation would have required air pollution abatement equal to -or even
26 greater than- that required by the CAA; particularly since some states, most notably California, have
27 done so. If one we« to assume mat state and local regulations would have been equivalent to CAA
28 standards, men a cost-benefit analysis of the CAA would be a meaningless exercise since both costs and
29 benefits would equal zero. Any attempt to predict how states' and localities' regulations would have
x differed from the CAA would be too speculative to support the credibility of the ensuing analysis.
31 Instead, the no-control scenario has been structured to include the assumption that states and localities
32 would not have set up air pollution control programs in the absence of the federal CAA. That is, this
1 While the configuration of these scenarios nay appear straightforward, the need to model at least one hypothetical scenario raises a
number of issues which are discussed to detail in the main text For example, the level of state and/or local controls which would have
prevailed in the absence of a federal CAA is highly speculative. So while some may argue mat states and localities might have provided
protections comparable to those required by the federal CAA, adopting that assumption yields a finding of zero costs and zero benefits of the
federal Act; a result which is neither informative nor responsive to Congressional and EPA objectives.
-------
Chapter 1: Introduction
i analysis accounts for all costs and benefits of air pollution control from 1970 to 1990. Speculation about
2 the fraction of costs and benefits attributable exclusively to the federal CAA is left to others.
3 The second assumption concerns changing demographic patterns in response to air pollution. In the
4 hypothetical no-control world, air quality is worse than mat hi the historical "control" world particularly
5 in urban industrial areas. It is possible that in the no-control case more people, relative to the control
6 case, would move away from the most heavily polluted areas. Rather than speculate on the scale of
7 population movement, the analysis assumes no differences in demographic patterns between the two
» scenarios. Similarly, the analysis assumes no changes hi the spatial pattern of economic activity. For
9 example: if, in the no-control case, an industry is expected to produce greater output man it did in the
10 control case, that increased output is produced by actual historical plants, avoiding the need to speculate
;/ about the location or other characteristics of new plants providing additional productive capacity..
12 Analytical Sequence
The analysis was designed and implemented in a sequential manner following seven basic steps
14
which are summarized below and described hi detail later in this report The seven major steps were:
is • direct cost estimation ., " - ,
n • macroeconomic modeling ,-. :
i? • emissions modeling -=-f . ,
H • air quality modeling "\=~ _~*
19 • health and environmental effects estimation _:~:
20 • results aggregation and imceitamty estimation >r
21 By necessity, each of these components had to be completed in a sequential manner. The emissions
22 modeling effort had to be completed entirely before the air quality models could be configured and run;
23 the air quality modeling results had to be completed before the health and environmental consequences
24 of air quality changes could be derived^ and so on. The analytical sequence, and the modeled versus
25 actual data basis for each analytical component, are summarized in Figure 4 and described in the
26 remainder of this section. ; > .Vr-
."-,"- "-j^s^ ' . .
27 The first step of the analysis was to estimate the total direct costs incurred by public and private
is entities to comply with po^t- 1970 CAA requirements. These data were obtained directly from Census
29 Bureau and Bureau of Economic Analysis (BEA) data on compliance expenditures reported by sources,
30 and from EPA regulatory and budget analyses. These direct cost data were then adopted as inputs to the
31 macroeconomic model used to project economic conditions -such as production levels, prices,
32 employment patterns, and other economic indicators- under the two scenarios. To ensure a consistent
33 basis for scenario comparison, the analysis applied the same macroeconomic modeling system to
M estimate control and no-control scenario economic conditions.4 First, a control scenario was constructed
31 by running the macroeconomic model using actual historical data for input factors such as
36 economicgrowth rates during the 1970 to 1990 period. The model was then re-run for the no-control
4 It is important to emphasize that using modeled economic conditions for both scenarios has both advantages and disadvantages. The
principal disadvantage is that historical economic conditions "predicted" by a macroeconomic model will not precisely duplicate actual
historical events and conditions. However, this disadvantage is outweighed by the avoidance of distortions and biases which would result from
comparing a modeled no-control scenario with actual historical conditions. By using the same macroeconomic model for bom scenarios, model
enrois and biases essentially cancel out, yielding more robust estimates of scenario difGsrences, which are what thb analysis seeks to evaluate.
-------
Chapter 1: Introduction
Figure 4. Summary of Analytical Sequence and Modeled versus Historical Data Basis.
Control Scenario
No-control Scenario
Compile historical compliance expenditure
data
Develop modeled macroeconomic scenario
based on actual historical economic data
Develop modeled macroeconomk scenario
by rerunning control scenario with
compliance expenditures added bac& to the
economy
Project emissions by year, pollutant, and
sector using control scenario
macroeconomic projection as input to
sector-specific emissions models
Re-run sector-specific emissions models
using no-control scenario macroeconomic
projection
Develop statistical profiles of historical air
quality for each pollutant based on
historical monitoring data (plus
extrapolations to cover unmonitored areas)
Derive no-control air quality profiles by
adjusting control scenario profiles based on
differences in air quality modeling of
control scenario and no-control scenario
emissions inventories
Estimate physical effects based on
application of concentration-response --- •-
functions to historical air quality profiles
Estimate physical effects based on
application of concentration-response •
functions to no-control scenario air quality
profiles
Calculate differences in physical outcomes
between control and no-control scenario
Estimate economic value of differences in
physical outcomes between the two
scenarios*
Compare historical, direct compliance costs
with estimated economic value of
monetized benefits, considering additional
benefits which could not be quantified
and/or monetized
* In some cases, economic value is derived directly from physical effects modeling (e.g., agricultural yield loss).
-------
Chapter 1: Introduction
i scenario by, in essence, returning all post-1970 CAA compliance expenditures to the economy. With
2 thes additional resources available for capital formation, personal consumption, and other purposes,
3 overall economic conditions under the no-control scenario were different In addition to providing
4 estimates of the difference in overall economic growth and other outcomes under the two scenarios, these
5 first two analytical steps were used to define specific economic conditions used as inputs to the '
6 em) sions modeling effort, the first step hi the estimation of CAA benefits.3
7 Using appropriate economic indicators from the macroeconomic model results as inputs, a variety of
« emissions models were run to estimate emissions levels under the two scenarios. These emissions
9 mof is provided estimates of emissions of six major pollutants* from each of six key emitting sectors:
10 utii. es, industrial processes, industrial combustion, on-highway vehicles, off-highway vehicles, and
n oorc mercial/residential sources. The resulting emissions profiles reflect state-wide total emissions from
12 eacr pollutant-sector combination for the years 1975,1980,1985, and 1990.7
u ' n next step toward estimation of benefits involved translating these emissions inventories into
u estir -.1, js of air quality conditions under each scenario. Given the staggering complexity, data
is requ. nents, and operating costs of state-of-the-art air quality mocsls -and the afore-mentioned
16 resc constraints- the EPA Project Team adopted simplified, linear scaling approaches for a number
17 ofp ants. However, for ozone and other pollutants or air quality conditions which involve
is subs. aal non-linear formation effects and/or long-range atmospheric transport and transformation, the
19 EP/ Project Team invested the time and resources needed to usemore sophisticated modeling systems.
20 For example, urban area-specific ozone modeling was conducted for 147 urban areas throughout the 48
21 contiguous states.
22 Up to this point of the anarysis^both the control and no-control scenario were based on modeled
23 conditions and outcomes. HiOTweveri.at the air quality modeling step, the analysis returns to a foundation
24 based on actual historical conditicns and data. Specifically, actual historical air quality monitoring data
25 from 1970 to 1990 anyised to del^Jhejcontrol scenario. Air quality conditions under the no-control
26 scenario are then derived by scalrng ti^ Jiistorical data adopted for the control scenario by the ratio of the
27 modeled control and no-control scenariosjtir quality.* This approach takes advantage of the richness of
21 the historical data on air qoal^ provide!! a realistic grounding for the benefit measures, and yet retains
29 the analytical consistency coi^x«4 by using the same modeling approach for both scenarios. The
1 For example, themacroeconomk model projected different electricity sales levels under the two scenarios, and these sales levels were
used as key input assumptions by the utility sector emissions model.
These six pollutants are total suspended paniculates (TSP), sulfur dioxide (SO]), nitrogen oxides (NOJ, carbon monoxide (CO), volatile
organic compounds (VOCs), and lead (Pb). The other CAA criteria pollutant, ozone (O,), is formed in the atmosphere through the interaction
of sun'ight and ozone precursor pollutants such as NO, and VOCs.
By definition, 1970 emissions under the t» scenarios are identical.
A vastly oversimplified example clarifies the concept say the actual year-long average 1980 sulfur dioxide concentration at a particular
monitor in upstate New York was 0.08 parts per million (ppm). Assume further that the air quality modeling performed for this analysis
. estimated that the.New York state-wide average 1980 concentration were 0.10 ppm under the control scenario and 0.16 ppm under the no-
control scenario. In this case, the 0.08 ppm historical figure is adopted as the 1980 control scenario value for that monitor. The value adopted
for the 1980 no-control scenario at that monitor would be 0.014 ppm, reflecting the 0.08 ppm control scenario value plus the 0.06 ppm
difference between the modeled scenario outcomes.
-------
Chapter 1: Introduction
outputs of this step pf the analysis are statistical profiles for each pollutant characterizing air quality
2 conditions at each monitoring site in the lower 48 states.9
3 The control and no-control scenario air quality profiles were then used as inputs to a modeling
4 system which translates air quality to physical outcomes -such as mortality, emergency room visits, or
j crop yield losses-through the use of concentration-response functions. These concentration-response
« functions are in turn derived from studies found in the scientific literature on the health and ecological
7 effects of air pollutants. At this point, estimates were derived of the differences between the two
s scenarios in terms of incidence rates for a broad range of human health and other effects of air pollution
9 by year, by pollutant, and by monitor.10
10 In the next step, economic valuation models or coefficients were^used to estimate the economic value
/; of the reduction in incidence of those adverse effects which were amenable to such monetization. For
n example, unit values derived from the economic literature were used to estimate the value of reductions
n in mortality risk associated with exposure to particulate matter. In addition, benefits which could not be
14 expressed in economic terms were compiled and are presented herein. la some cases, quantitative
is estimates of scenario differences in the incidence of a nonmonetized effect were calculated.11 In many
16 other cases, available data and techniques were insufficient to support anything more than a qualitative
i? characterization of the change in effects.
is Finally, the costs and monetized benefits were combined to provide a range of estimates for the
19 partial, net economic benefit of the CAAwith the range reflecting quantified uncertainties associated
20 with the physical effects and economic valuation steps.12 The term "partial" is emphasized because the
costs of historical CAA programs were readily obtained but only a subset of the total potential benefits of
22 the CAA could be represented in economic terms
-------
Chapter 1: Introduction
Review Process
i
2 The CAA requires EPA to consult with an outside panel of experts -referred to statutonly as the
3 Advisory Council on Clean Air Act Compliance Analysis (CAACAC)- in developing the section 812
4 ' analyses: In addition, EPA is required to consult with the Department of Labor and the Department of
s Commerce.
6 The CAACAC was organized hi 1991 under the auspices and procedures of EPA's Science Advisory
7 Board (SAB). Organizing the review committee under the SAB ensured that review of the section 812
i studies would be conducted by highly qualified experts in an objective, rigorous, and publicly open
9 manner. The SAB CAACAC has met several times during the development of the retrospective study to
10 review methodologies and interim results. While the full CAACAC retains overall review responsibility
n for the section 812 studies, some specific issues concerning physical effects and air quality modeling
12 have been referred to subcommittees comprised of both CAACAC members and members of other SAB
13 committees. The CAACAC Physical Effects Subcommittee has met several times and has provided its
H own review findings to the full CAACAC. Similarly, the CAACAC Ah- Quality Subcommittee has held
is several teleconference meetings to review methodology proposals: r-
16 With respect to the interagency review process, the EPA has expanded the list of consulted agencies
n to include the Department of Energy, the Council on Environmental Quality, the National Acid
11 Precipitation Assessment Program, The Office of Management and Budget, and the Council of Economic
19 Advisors. While several meetings have been heM with the Interagency Review Group during the course
20 of the retrospective study, me EPAiProject Team now provides opportunity for integrated interagency,
21 expert panel, and public (x>nsultatloj) through the public review meetings conducted by the SAB
22 CAACAC;T::±, "'" •• . •
Report Organization
23 .
- -Jw
24 TTieremamder of me ma^teittof mis report sunm^
25 retrospective study, organized by chapter according to the seven major steps of the analysis. The direct
26 cost estimation and macroeconomic modeling steps are presented hi Chapter 2. The emissions modeling
27 is summarized in Chapter^. Chapter 4 presents the air quality modeling methodology and sample
is results. Chapter 5 describes the approaches used and principal results obtained through the physical
29 effects estimation process. Economic valuation, methodologies and initial results for the aggregation of
30 monetized benefits of the historical CAA are presented in Chapter .6. Chapter 1 describes the uncertainty
31 analysis and associated re-aggregation of the overall cost and benefit estimates.
n Additional details regarding the methodologies and results are presented hi the appendices and in the
33 referenced supporting documents. Appendix A covers the direct cost and macroeconomic modeling.
34 Appendix B provides additional detail on the sector-specific emissions modeling effort Details of the
35 air quality models used and results obtained are presented or referenced in Appendix C. The effects of
je the CAA on human health and visibility; aquatic, wetland, and forest ecosystems; and agriculture are
37 presented in Appendices D, E, and F, respectively. Appendix G presents details of the lead (Pb) benefits
33 analysis. Air toxics reduction benefits are discussed in Appendix H. The methods and assumptions used
39 to value quantified effects of the CAA in economic terms are described in Appendix I. Appendix J
8
-------
Cliapter 1: Introduction
discusses some of the preliminary research implications of this study and describes plans for future
2 Section 812 analyses.
-------
[This page intentionally blank]
10
-------
2
Cost and Macrceccncmic Effects
3
4
S
6
7
S
9
10
11
12
13
16
17
IS
19
20
21
22
23
24
25
26
27
2S
29
The costs of complying with Clean Air Act (CAA)
requirements through the 1970 to 1990 period affected
patterns of industrial production, capital investment,
productivity, consumption, employment, and overall
economic growth. The purpose of thee analyses summarized
in this chapter was to estimate those direct costs and the
magnitude and significance of resulting changes to the overall
economy. This was accomplished by comparing economic
indicators under two alternative scenarios: a control scenario
serving as the historical benchmark, including the historical
CAA as implemented; and a no-control scenario which
assumes historical CAA programs did not exist.„ The
estimated economic consequences of the historical CAA were
taken as the difference between these two scenarios.
The remainder of this chapter summarizes the basic
methods and results of the direct compliance cost and
macroeconomic analyses. Further detail regarding the
modeling methods and assumptions employed, as well as
additional analytical results, are presented in Appendix A.
Direct Compliance Costs
Compliance with the CAA imposed direct costs on
businesses, consumers, and governmental units; and triggered
other expenditures such as governmental regulation and
monitoring costs and expenditures for research and
development by both government and industry. Although
expenditures unadjusted for inflation —that is, expenditures
denominated in "current dollars"— increased steadily over
Table 4. Estanated Annual Direct CAA
S>*
AmnBtBzed
Year Sam-art St990 S1990
#73 74
19*5
iu 2*4
195T7
I97?f
W
im
&&
24,8
144
15J?
I7JF
1983
1984
j«;t 21*8
!«4 294
18,0 224
t*J
23J
i^r
21.4
18,9 20.6
I9#
1999
204
Ab«MWM««l Comrol Expendiwrw
SEA; 'Cb^aiiBeaefifii of ReduciagljMd fowl
11
-------
Chapter 2: Cost and Macroeconomic Effects
i the 1973 to 1990 period,14 annual direct CAA compliance expenditures adjusted for inflation were )
2 remarkably stable, ranging from $20 billion to $25 billion in 1990 dollars (see Table 4). This is equal to
3 approximately 1/> of one percent of total domestic output during that period, with the percentage falling
4 from Vi of one percent of total output in 1973 to l/4 of one percent in 1990.
3 Although useful for many purposes, a summary of direct annual expenditures is not the best cost
6 measure to use when comparing costs to benefits. Capital expenditures are investments, generating a
7 stream of benefits and opportunity cost15 over the life of the investment. The appropriate accounting
i technique to use for capital expenditures in a cost/benefit analysis is to antiimligp the expenditure. This
9 technique, analogous to calculating the monthly payment associated with a home mortgage, involves
w spreading the cost of the capital equipment over the useful life of the equipment using a discount rate to
;/ account for the time value of money.
12 For this cost/benefit analysis, all capital expenditures were amnialized at a 3 percent, inflation-
n adjusted rate of interest Therefore, "annualized" costs reported for any given year are equal to O&M
14 expenditures — including R&D and other similarly recurring expenditures— .plus amortized capital costs
is . (i.e., depreciation plus interest costs associated with the capital s/odt) for that year. Stationary source air
16 pollution control capital costs were amortized over 20 years; mobile source air pollution control costs
n were amortized over 1 0 years. Table 1 1 summarizes costs annualized at 3 percent, as well as annual
is expenditures. f - = : ' ^fi
- - ""-"
it
Indirect Effects cf the CAA
_--
x Through changmgprodurtiba costs, CAAimplem
21 and miismmesizeandcompositioi^of economic bu^ut The Project Team used a general equilibrium
22 macroeconomic model to assess the extent of such second-order effects. This type of model is useful
23 because h can capture the feedback effects of an action. In the Section 812 macroeconomic modeling
24 exercise, the feedback effects arising from expenditure changes were captured, but the analogous effects
25 arising from improvements in human health were not captured by the model. Consequently, the
26 macroeconomic mcKlelmgexefcJiD provides limited and incomplete information on the type and
27 potential scale of mdirecte«JttMiuc effects.
28 The effects estimated by the macroeconomic model can be grouped into two broad classes: sectoral
29 impacts (i.e., changes in the composition of economic output), and aggregate effects (i.e., changes in the
30 degree of output or of some measure of human welfare). The predicted sectoral effects were used as
14 Due to data limitations, the cost analysis for this CAA retrospective stirts in 1973, missing costs incurred in 1970-72. Cost of Clean
(1990), however, which is the tource for annualized capital costs, includes capital expenditures for 1972. Therefore, amortized costs arising
from 1972 capital bwestments ore fecfaferf in the 1973-1990 amualized costs, even trxxigh 1972 «jsts«er^otheni^ included in the ,
analysis. This limitation is not debilitating in the context of this analysis, however, because relatively litfe in the way of compliance with the
"new" provisions of the 1970 CAA was requited in the first two years following passage. It should also be noted that, conversely, only a
portion of, for example, the 1990 capital expenditures are reflected in the 1990 annualized «>»ts - the remainder of the costs are spread through
the following two decades, which tall outside of the scope of this study. Similarly and consistently, benefits arising from emission reductions
realized after 1990 as a result of capital investments made during the 1970 to ic^ period of muaiialysis are lutfaiducledm the estimates of
benefits included in this report
11 In mis context, "opportunity cost" is defined as the value of alternative investments or other uses of funds foregone as a result of the
investment.
12 ' -
-------
Chapter 2: Cost and Macroeconomic Effects
inputs to the emissions models as discussed in Chapter 2. In general, the estimated second-order
macroeconomic effects were small relative to the size of the U.S. economy.
3 Sectoral Impacts
'4 The CAA had variable compliance impacts across economic sectors. The greatest effects were on the
5 largest energy producers and consumers, particularly those sectors which relied most heavily on
6 consumption of fossil fuels (or energy generated from fossil fuels). In addition, production costs
7 increased more for capital-intensive industries than for less capital-intensive industries under the control
s scenario due to a projected increase in interest rates. The interest rate Increase, which resulted in an
9 increase in the cost of capital, occurred under the control scenario because CAA-mandated investment in
10 pollution abatement reduced the level of resources available for other uses, including capital formation.
;/ Generally, the estimated difference in cost impacts under the control and no-control scenarios for a
12 particular economic sector was a function of the relative energy-intensity and capital-intensity of that
13 sector. Increased production costs in energy- and capital-intensive sectors under the control scenario
14 were reflected in higher consumer prices, which resulted in reductions in the quantity of consumer
is purchases of goods and services produced by those sectors. This reduction in consumer demand under
16 the control scenario led, ultimately, to reductions hi output and employment in those sectors. The sectors
i? most affected by the CAA were motor vehicles, petroleum refining, and electricity generation. The
is electricity generation sector, for example, incurred a two to four percent increase in consumer prices by
19 1990, resulting in a three to five and a half percent reduction ill output. Many other manufacturing
sectors saw an output effect in the one percent range.
f _ ~-~ -_•= - 3 ~_ _"j "-
11 Some other sectors, however, were projected id increase output iinder the control scenario. Apart
22 from the pollution central equipment industry, which was not separately identified and captured in the
23 macroeconomic modeling performed for mis study, two example sectors for which output was higher and
24 prices were lower under me control scenarioare food and furniture. These two sectors showed •
25 production cost and consumer price reductions of one to two percent relative to other industries under the
26 control scenario, i
27
M As noted above, the control and no-control scenarios yield different estimated mixes of investment
29 In particular, the control scenario was associated with more pollution control capital expenditure and less
30 consumer commodity capital expenditure. As a result, the growth pattern of the economy under the
31 control scenario differed from the no-control scenario. Under the control scenario, the macroeconomic
32 model projected a rate of long-run GNP growth equal to about one twentieth of one percent per year
33 lower than under the no-control scenario. Aggregating these slower growth effects of the control
34 scenario over the entire 1970 to 1990 period of this study results, by 1990, in a level of GNP one percent
15 lower than projected under the no-control scenario. This lower level of GNP is equivalent to a reduction
36 of approximately $50 billion in consumer products produced in the United States. However, it is
37 important to emphasize that this analysis addressed only the negative economic impacts of the CAA and
did not capture the beneficial economic effects of air pollution control, such as improved worker
productivity and reduced medical expenditures.
13
-------
,-.,s^
19
Chapter 2: Cost and Macroeconomic Effects
; It is also important to emphasize that changes in GNP do not necessarily provide a good indication of
2 changes in social welfare. Social welfare is not unproved, for example, by major oil tanker spills even
3 though measured GNP is increased by the "economic production" associated with clean-up activities.
4 Nevertheless, the effects of the CAA on long-term economic growth would be expected to have had
5 some effect on economic welfare. That is, changes in production costs and consumer prices should result
6 in decreased consumption by consumers (disregarding the increase in consumption of environmental
7 amenities).
s To gauge the magnitude of this effect, the Project Team estimated the monetary compensation which
9 would.be required to offset the losses in consumption associated with the control scenario, The resulting
10 estimates indicate that the required monetary compensation is about 120 to 160 percent of CAA
11 compliance cost. That is, if the indirect costs to consumers captured by this analysis are added to direct
12 compliance costs, the resulting total cost would be about 20 to 60 percent greater than direct compliance
13 cost alone. As noted above, however, these additional indirect costs do not reflect the indirect economic
14 benefits associated with improvements in human health and environment .Because of this limitation, the
is cost-benefit comparisons presented in this report reflect only the differences in direct costs and direct
16 benefits associated with the control and no-control scenarios. If both indirect costs and indirect benefits
n could be effectively captured, however, the differences between the two scenarios for both total costs and
a total benefits would likely be higher than presented in this repeat. ,
Conclusions
w The results of this assessment of the costs anil economic effects of the CAA must be carefully.
21 interpreted. It is importantto remember that the cost and macroeconomic simulation modeling does not
22 incorporate many health, environmental, and economic benefits of the CAA. For example,
23 improvements in worker productivity and reductions in medical expenditures due to the CAA were not
24 incorporated in the macroeconomic mode^ The direct and indirect benefits of these CAA-related
2s productivity benefits and expenditure reductions are not reflected in the estimates of changes hi GNP,
26 consumption, prices, outpi& and employment presented in this study.
27 The estimates of compliance-cost and economy-wide effects presented in this chapter do not by
2* themselves inform the question Of whether the net social and economic benefits of the CAA are positive.
29 This question requires careful and balanced consideration of the benefits of the CAA. The remainder of
30 this report focuses on estimation of these benefits, and the integration and interpretation of the overall
31 costs and benefits of the CAA are presented in subsequent chapters. Additional details about the cost
32 and macroeconomic modeling performed for this assessment are presented in Appendix A.
33 Differences between control and no-control scenario outcomes were observed at both economy-wide
34 and industry levels. The CAA imposed direct costs of 20 to 25 billion dollars per year on an inflation-
is adjusted basis. Furthermore, the level of GNP was estimated by the macroeconomic model to be one
36 percent lower in 1990 under the control scenario, and personal consumption was also one percent lower
37 by 1990. The greatest industry-level difference between the two scenarios is realized by the motor
3s vehicle sector, which showed 5.3 percent lower output and 3.8 percent higher price levels under the
39 control scenario. Output effects are also seen in other capital-intensive and energy-intensive industries,
40 including petroleum refining, electric utilities, transportation equipment, primary metals, and metal
41 mining. •
14
-------
3
3 • This chapter presents estimates of emissions reductions due to the Clean Air Act (CAA) for six
4 criteria air pollutants. Reductions are calculated by estimating, on a sector-by-sector basis, the
s differences in emissions between the control and no-control scenarios. While the relevant years in this
6 analysis are 1970 through 1990, full reporting of emissions was only made for the 1975 to 1990 period
7 since 1970 emission levels are, by assumption, identical for the two scenarios. The criteria pollutants for
s which emissions are reported in this analysis are: total suspended particulates(TSP),16 carbon monoxide
9 (CO), volatile organic compounds (VOC), sulfur dioxide (SOj), nitrogen oxides (NOJ, and Lead (Pb).
10 The purpose of the present study is to estimate the differences In economic and environmental
// . conditions between a scenario reflecting implementation of historical CAA controls and a scenario which
n assumes that no additional CAA-related control programs were introduced after 1970. Because of the
•' focus on differences in -rather than absolute levels of- emissions between the scenarios, the various
sector-specific emission models were used to estimate both the control and no-control scenario emission
75 inventories. This approach ensures that differences between me scenarios are not distorted by
16 differences between modeled and actual historical emission estimates.17
n ' - ' - ~ ~~ - -: - ". ".
n Despite the use of models to estimate control scenario emission inventories, the models used were
19 configured and/or calibrated using historical emissions estimates. The control scenario utility emissions
20 estimates, for example, werCibased on ttte ICF CEUM model which was calibrated using historical
21 emissions inventory data.11 fcother cases, such as the EPA Emissions Trends Report (Trends)
22 methodology19 used to estimate indiistrial pro<»ss emissions, m'stori<^ data were used as the basis for
23 control scenario emissions with little or no subsequent modification. Nevertheless, differences in model
24 selection, model configuration, 4nd macroeconomic input data20 result in unavoidable, but in this case
* IB 1987, EPA replaced Ac earlier TSP standard who a standard for paniculate matter of 10 microns or smaller (PM10).
n By necessity, emission iw>deU must be used to estimate the hvpothe^ If actual historical emissions data were
adopted for Jhe control scenario, differences between the monitoring data and/or models used to develop historical emission inventories and the
models used to develop no-control scenario emission estimates would bias the estimates of the differences between the scenarios.
" See ICF Resources, Inc., "Results of Retrospective Electric Utility Clean Air Act Analysis -1980,1985 and 1990," September 30,1992,
Appendix C.
" EPA, 1994a: U.S. Environmental Protection Agency, "National Air Pollutant Emission Trends, 1900-1993," EPA-454/R-94-027,
Office of Air Quality Planning and Standards, Research Triangle Park, NC, October 1994.
" The Jorgenson/Wilcoxen macroeconomic model outputs were used to configure both the control and no-control scenario emission
model runs. While this satisfies the primary objective of avoiding "across model" bias between the scenarios, the macroeconomic conditions
, associated with the control scenario would not be expected to match actual historical economic events and conditions. To the extent actual
historical economic conditions are used to estimate official historical emission inventories, conformity between these historical emissions
15
-------
Chapter 3: Emissions
i Justifiable, differences between national total historical emission estimates and national total control
2 scenario emission estimates for each pollutant. Comparisons between no-control, control, and official
3 EPA Trends Report historical emissions inventories are presented in Appendix B.21
4 To estimate no-control scenario emissions, sector-specific historical emissions are adjusted based on
5 changes in the following two factors: (1) growth by sector predicted to occur under the no-control
6 scenario; and (2) the exclusion of controls attributable to specific provisions of theCAA.
7 To adjust emissions for economic changes under the no-control s^nario, activity levels that affect
« emissions from each sector were identified. These activity levels include, lor example, fuel use,
9 industrial activity, and vehicle miles traveled (VMT). The Jorgenson-Wilcoxen (J/W) general
10 equilibrium model was used to estimate changes in general economic conditions, as well as sector-
/; specific economic outcomes used as inputs to the individual sector emission models.22
12 The specific outputs from the J/W model used in this analysis are tile percentage changes in gross
13 national product (GNP), personal consumption, and output for variouTeconomic sectors under the
14 control and no-control scenario for the years 1975,1980,1985, and 1990*?. The sectors for which the
is results of the J/W model are used include: industrial processes, electric utilities, highway vehicles,
16 industrial boilers, and the commercial/residential sector. For the off-highway sector, economic growth
17 was not taken into account since there was no direct correspondence between J/W sectors and the off-
ii highway vehicle source category activity.. ' ;;
19 In addition to adjusting for economic activity changes, any CAA-related control efficiencies that
20 were applied to calculate control scenario emissions were removed for the no-control scenario. In most
21 instances, emissions were recalculated based on 1970 control levels.
f -=^p"
23 The approaches used to calculate emissions for each sector vary based on the complexity of .
24 estimating emissions in mejuSsencejof CAA controls, taking economic activity levels and CAA
25 regulations into account I^l|e off-highway vehicle and industrial process sectors, a relatively simple
26 methodology was developed lite approaches used for the highway vehicles, electric utilities, industrial
27 boilers, and commercial/residential sectors were more complex because the J/W model does not address
11 In grant, these comparisons show dose correspondence between oxarol scenario tod Fr««iestim«es with the latest
occurring for VOC and CO emissions. The Trtnds report VOC estimates we generally higher than the control scenario estimates due primarily
to the inclusion of Waste Disposal and Recycling ai a VOC source in the TVvndt report This inconsistency is of no consequence since Waste
Disposal and Recycling sources were essentially uncontrolled by the historical C\A and theiefijre do not appear as a difference between the
control and no-control scenarios. The higher CO emission estimates in the Trtntb Report are primarily associated with higher off-highway
vehicle emissions estimates. Again, since off-highway emissions do not change between the control and no-control scenario in the present
analysis, this inconsistency 'u of no consequence.
a For example^ the change in distribution of households by income class predicted by the J/W model was used as input to me
transportation sector model system. Changes in household income resulted in changes in vehicle ownership and usage patterns which, in turn,
influence VMT and emissions. (See Pechan, 1995, p. 43) . .
23 For details regarding the data linkages between the J/W model and the various emission sector models, see Pechan (1995).
16
-------
Chapter 3: Emissions
all of the determinants of economic activity in these sectors that might have changed in the absence of
regulation. The approaches by sector used to estimate emissions for the two scenarios are summarized in
Table 5, and are described in more detail in Appendix B.
Table 5. Sommary of Sector-Specific EmissJCT Modeling Approaches.
Sector
Modeling Approach
On-Higbway Vehicles
UQIStntflQll
MOBILES* emissiflttfectoa wa» used to calculate emissions.
fts&aatgd by A& Associates based
QD hiswrtcal garatiae sates Add die lead co«i^o| leaded gasdlmpmtadi
target year. % '
elridtes
w^; inade »
Electric
ICF's Coal and Electric Utility Model (CEUM) used to asses* SO^, NO,, and
TlwArgonne Utility Siasdatwn Model (ARGUS) provided CO and VOC
li iftfttf lfflflff Wftff iCtflCTfaft^ feiXHjdl W 'ftPiftffiy' fifffflffldBMtfeHI: ^fa ffflfli
Industrial Combustion
CPffllHiBt^iOft IjffiitiB^fl'Bii ffCf^ flBfxfeJIi
fftomladustrMboliettwbB
Treads metbods; recalculated using 1970 coottol efficiencies.
*
gyf ^ffitfyasi ift'ihi.'TtHatTyffyffif.'t-^
4ouatedlt»ind^«atproM!«e8«i^
,and
OojBxnerciaJ /
Jlesldential
AHt-'s Coauneicial aodl^deatial Simaiatfatt,Syj«eitQ)
unodeliira*
17
-------
Chapter 3: Emissions
Summary of Results
2
3
.4
5
6
7
S
9
10
11
12
13
14
IS
16
17
IS
19
20
21
22
23
24
IS
26
27
n
29
30
31
32
33
34
35
36
37
3t
39
40
41
42
Figure S compares the total
estimated sulfur dioxide emission from
all sectors under the control and *
no-control scenarios over the period
from 1975 to 1990. Figures 6,7, 8,9,
and 10 provide similar comparisons for
NO,, VOCs, CO, TSP, and Lead (Pb)
respectively.
Additional tables presented in
Appendix B provide further breakdown
of die emissions estimates by
individual sector, though the essential
results are characterized below. For
most sectors, emission levels under the
control scenario were substantially
lower than levels projected under the
no-control scenario. .
The CAA controls that affected M?
SO2 emitting sources had the greatest
proportional effect on industdat i|
process emissions, which were^Sj^
percent lower in 1990 than they W^fldS
have been under the no-control
Figure S. Control and No-control Scenario Total SO, Emission
Estimates. • -
40
30
* I 20
S I
I
10
1975
1980 1915
Yew
1990
j-E"_%a. _ -
scenario. SO2 emissions from elecdfcllS
utilities and industrial boifcg|%pre
each nearly 40 percent 1
a result of CAA controls, "fill
absolute tons of emission reductions,
controls on electric utilities account for
over 10 million of the total 16 million
ton difference between the 1990 control
and no-control scenario SO, emission
estimates.
CAA regulation of the highway
vehicles sector led to the greatest
percent reductions in VOC and NO,.
Control scenario emissions of these
pollutants in 1990 were 66 percent and
47 percent lower, respectively, than the
levels estimated under the no-control
scenario. In absolute terms, highway
?igure 6. Control and No-control Scenario Total NO, Emission
?«timates " .= .--"
*:-*.
40
30
Jl 20
a S
10
(Control I
..No-Control
1975
19SO
1985
1990
Yew
Figure 7. Control and No-control Scenario Total VOC Emission
.9
40
30
20
10
t Control I
•.No-Control
1975
1980
1985
1990
Yew
i.*
18
-------
Chapter 3: Emissions
3
4
S
6
7
S
9
10
II
12
13
14
IS
16
17
IS
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
3S
39
40
41
42
43
44
vehicle VOC controls account for over
IS million of the roughly 17 million ton
difference in control and no-control
scenario emissions.
Differences between control and
no-control scenario CO emissions are
also most significant for highway
vehicles. In percentage terms, highway
vehicle CO emissions were 56 percent
lower in 1990 under the control
scenario than under the no-control
scenario. Industrial process CO
emission estimates under the control
scenario were about half the levels
projected under the no-control scenario.
Of the roughly 89 million ton
difference in CO emissions between the
two scenarios, 84 million tons are
•attributable to highway vehicle controls
and the rest is associated with
reductions from industrial process
emissions. ;:
x/
For TSP, the highest level of ,
reductions on a percentage basis wen
achieved in the electric utilitiesi secilf .,,
TSP emissions from electric utilities :
were 93 percent lower in 1990 undejB:^;:
the control scenario than projected '""'-- -:
under the no-control scenario, TSP *--
'igure 8. Control and No-control Scenario Total CO Emission
Estimates.
200
§
H
150
.s 100
50
.g Control I
^. No-Control
1975
19SO
1985
1990
Yew
'igure 9. Control and No-control Scenario Total TSP Emission
Estimates. •
30
20
10
OUL
v Control I
.4. No-Control
1975
1980 1985
Year
1990
were also significantly lowwolii^
percentage basis under the control
scenario, with me differential reaching
76 percent by 1990. This is not an
unexpected result as ak pollution
control regulations Ifttiie 1970's
focused on solving the visible
particulate problems from large fuel
combustors. In terms of absolute tons,
electric utilities account for over 5
million of the 16 million ton difference
between the two scenarios and
industrial processes account for almost
10 million tons.
'igure 10. Control and No-control Scenario Total Pb Emission
200
150
100
50
^ Control I
^.No-Control
1975 1980 1985 1990
Yetr
19
-------
Chapter 3: Emissions
The vast majority of the difference in lead emissions under the two scenarios is attributable to
reductions in burning of leaded gasoline. By 1990, reductions in highway vehicle emissions account for
221 thousand of the total 234 thousand ton difference in lead emissions. As shown in more detail in
Appendix B, airborne lead emissions from all sectors were virtually eliminated by 1990.
As described in the following chapter and in Appendix C, these emissions inventories were used as
inputs to a series of air quality models. These air quality models were used to estimate air quality
conditions under the control and no-control scenarios. •
Potential errors in the emissions modeling, and their significance relative to overall uncertainty in the
monetary benefit estimate, are presented in Table 6.
Potential
Directfoa of Potential
Bia»itt Estimate of
Use of -I07& ototor veawle eajisswn &$&* "
for
Ito
anrf
Bfectranio
a
rather ami CEUM.
Negligible.
inventories remam fb^ betweea «»e contRd
•: -i • • j .,
-, "" • •> "• ^ 1'^''v ^ 'Wd- •%
i- .-.> ^
05W
-------
4
Air Quality
3 Air quality modeling is the crucial analytical step which links emissions to changes in atmospheric
4 concentrations of pollutants which affect human health and the environmental. It is also one of the more
5 complex and resource-intensive steps, and contributes significantly to overall uncertainty in the bottom-
6 line estimate of net benefits of air pollution control programs.
7 The key challenges faced by air quality modelers attempting to translate emission inventories into air
s quality measures involve modeling of pollutant dispersion and atmospheric transport, and modeling of
9 atmospheric chemistry and pollutant transformation. These challenges are particularly acute for those
10 pollutants which, rather than being directly emitted, are formed through secondary formation processes.
// Ozone is the paramount example since it is formed in the atmosphere through complex, nonlinear
12 chemical interactions of precursor pollutants, particularly volatile organic compounds (VOCs) and
* nitrogen oxides (NOJ. In addition, atmospheric transport and transformation of gaseous sulfur dioxide
and nitrogen oxides to particulate suliates and nitrates, respectively, contributes significantly to ambient
11 concentrations of fine particulate matter. In addition to managing the complex atmospheric chemistry
16 relevant for some pollutants, air quality modelers also must deal with uncertainties associated with
n variable meteorology and me spatial and temporal distribution of emissions.
is Given its comprehensive nature, the present analysis entails all of the aforementioned challenges,
19 and mvolves addhlonal con^Ucations as well. For many pollutants which cause a variety~of human
20 health and eavironmentaJucp^ct^ the concentration-response functions which have been developed to
21 estimate mose effects reqtore, 18*11^ For example, adverse human
22 health effects of paiticulate paper are primarily associated with the fine particle fraction;24 whereas
23 household soiling is a function of total suspended particulates, especially coarse particles. It is not
24 enough, therefore, to simply provide a single measure of particulate matter air quality. Even for
25 pollutants for which particle size and other characteristics are not an issue, different air quality indicators
26 are needed which reflect different periods of cumulative exposure (i.e., "averaging periods"). For
27 example, 3-month growing season averages are needed to estimate effects of ozone on yields of some
2s agricultural crops, whereas adverse human health effect estimates require ozone concentration profiles
29 based on a variety of short-term averaging periods.23
jo Fortunately, in responding to the need for scientifically valid and reliable estimation of air quality
31 changes, air quality modelers and researchers have developed a number of highly sophisticated
32 atmospheric dispersion and transformation models. These models have been successfully employed for
M Particles with an acrometric diameter of less than or equal to 2.5 microns.
23 For example, ozone concentntton-response data exists for effects associated with 1-hour, 2.5-hour, and 6.6-hour exposures.
21
-------
Chapter 4: Air Quality
i years supporting the development of overall federal clean air programs, national assessment studies,
2 State Implementation Plans (SIPs), and individual air toxic source risk assessments. Some of these
3 models, however, require massive amounts of computing power. For example, completing the 160 runs
4 of the Regional Acid Deposition Model (RADM) required for the present study required approximately
s 1,080 hours of CPU time on a Cray-YMP supercomputer at EPA's Bay Chy Supercomputing Center.
6 Given the resource-intensity of many state-of-the-art models, me Project Team was forced to make
7 difficult choices regarding where to invest the limited resources available for tax quality modeling. With
s a mandate to analyze all of the key pollutants affected by historical Clean Air Act programs, to estimai.
9 all of the significant endpoints associated with those pollutants, and to do so for a 20 year period
10 covering the entire continental U.S., it was necessary to use simplified approaches for most of the
n pollutants to be analyzed. In several cases related to primary emissions—particularly sulfur dioxide
n (SOz), nitrogen oxides (NOJ, and carbon monoxide (CO)— simple "roll-up model" strategies were
13 adopted based on the expectation mat changes hi emissions of these pollutants would be highly
14 correlated with subsequent changes hi air quality.26 Significant pollutants involving secondary
is atmospheric formation, nonlinear formation mechanisms, and/or long-range, transport were analyzed
16 using the best ah- quality model which was affordable given time and resowre limitations. These
n models, discussed in detail in Appendix C, included the Ozone Isopleth Plotting with Optional
a Mechanism-rV (OZIPM4) model for urban ozone; various forms o£the above-referenced RADM model
19 for background ozone, acid deposition, sulfate, nitrate, and visibility effects in the eastern U.S.; and the
20 SJVAQS/AUSPEX Regional Modeling Adaptation Project (SARMAP) Air Quality Model (SAQM) for
21 rural ozone in California agricultural areas. In addition, a linear scaling approach was developed and
22 implemented to estimate visibility changes in large southwestern U.S. urban areas.
=. •" -"" = •
"'s.V " -" * _
23 By adopting simplified approaches for some pollutants, me air quality modeling step adds to the
24 overall uncertainties and limitations of the present analysis. The limited expanse and density of the U.S.
25 air quality monitoring network and the limited coverage by available air quality models of major
26 geograpnic areas" rurfter constramliwa^ Under these
27 circumstances, it is important to remember {he extent and significance of gaps hi geographic coverage
21 for key pollutants when cons^ering the overaU results of this analysis. Key uncertainties are
29 siimmarized at the end of this chapter ui Table 7. More extensive discussion of the caveats and
» uncertainties associated wlu die air jquah^modelmg step is presented in Appendix C. In addition,
31 information regarding the specific w quality models used, the characteristics of the historical monitoring
32 data used as the basis for the control scenario profiles, pollutant-specific modeling strategies and
33 assumptions, references to key supporting documents, and important caveats and uncertainties are also
34 presented in Appendix Cs
* It is important to emphasize that the correlation expected is between change* in emissions and changes in air quality. Direct
correlations between the absolute emissions estimates and empirical air quality measurements used in the present analysis may not be strong
due to expected inconsistencies between the geographically local, monitor-proximate emissions densities affecting air quality data.
v For example, the regional oxidant models available for the present study do not cover some key Midwestern states, where human
health, agricultural crop, and other effects from ozone may be significant
22
-------
Chapter 4: Air Quality
General Methodology
2 The general methodological approach taken in this analysis starts with the assumption that actual
3 historical air quality will be taken to represent the control scenario. This may seem somewhat
4 inconsistent with the approach taken in earlier steps of the analysis, which used modeled macroeconomic .
5 conditions as the basis for estimating macroeconomic effects and emissions. However, the central focus
& of the overall analysis is to estimate the difference in cost and benefit outcomes between the control and
7 no-control scenarios. It is consistent with this central paradigm to use actual historical air quality data as
* the basis for estimating how air quality might have changed in the absence of the Clean Air Act
9 The initial step, then, for each of the five non-lead (Pb) criteria pollutants21 was to compile
w comprehensive air quality profiles covering the entire analytical period from 1970 to 1990. The source
n for these data was EPA's Aerometric Information Retrieval System (AIRS), which is a publicly
n accessible database of historical air quality data. The vast number of air quality observations occurring
n over this twenty year period from the thousands of monitors In the U.S. indicates the need to represent
14 these observations by statistical distributions. As documented in detail in the supporting documents
is covering SOj, NO0 CO, and ozone,29 both lognormal and gamma distributional forms were tested against
16 actual data to determine the form which provided the best fit te the historical data.30 Based on these
n tests, one or the other statistical distribution was adopted for the ak quality profiles developed for each
is pollutant. In addition to reducing the air quality data to a manageable form, this approach facilitated
19 transformations of air quality profiles from one averaging period basis to another.
24
26
20 Once the control scenario profiles based on historical data were developed, no-control scenarios were
21 derived based on the results of the various air qualitylmodeling efforts. Again, the focus of the overall
22 analysis is to isolate the difference in outcomes between me control and no-control scenarios. Theno-
23 control scenario air quality profiles were therefore derived by adjusting the control scenario profiles
upward (or downward) based on an appropriate measure of the difference in modeled air quality
outcomes. To illustrate this approach, consider a simplified example where the modeled concentration of
Pollutant A under the no-control scenario is 0.12 ppm, compared to a modeled concentration under the
27 control scenario of 0.10 pjKt|& appropriate measure of the difference between these outcomes,
is whether it is the 0.02 ppm difference hi concentration or the 20 percent percentage differential, is then
29 used to ratchet up the control case profile to derive the no-control case profile. Generally, the modeled
jo differential is applied acrossithe entire control case profile to derive the no-control case profile. As
31 described below in the individual sections covering paniculate matter and ozone, however, more refined
32 approaches are used where necessary to take account of differential outcomes for component species
33 (i.e., particulate matter), long-range transport, and background levels of pollutants.
" Lead (Pb), the sixth criteria pollutant, is analyzed separately. The ability to correlate emissions directly with blood lead levels obviates
the need for using air quality modeling as an intermediate step toward estimation of exposure.
29 See SAI SO* NOM and CO Report (1994) and SAI Ozone Report (1995).
10 The statistical tests used to determine goodness of fit are described in UK SAI reports.
23
-------
Chapter 4: Air Quality
Sample Results
2
3
4
5
6
7
I
9
10
11
12
The results of the air quality modeling effort include a vast array of monitor-specific air quality
profiles for fine particulates (PMIO), total suspended particulates (TSP),31 SOj, NO* NO, CO, and ozone;
RADM grid cell-based estimates of sulfur and nitrogen deposition; and estimates of visibility
degradation for eastern U.S. RADM grid cells and southwestern U.S. urban areas. All of these data were
transferred to the effects modelers for use in configuring the human health, welfare, and ecosystem
physical effects models. Given the massive quantity and intermediate nature of the air quality data, they
are not exhaustively reported herein.31 To provide the reader with some sense of the magnitude of the
difference hi modeled air quality conditions under the control and no-control scenarios, some illustrative
results for 1990 are presented hi this chapter and hi Appendix C. In addition, maps depicting absolute
levels of control and no-control scenario acid deposition and visibility are presented to avoid potential
confusion which might arise through examination of percent change maps done.33
13
14
IS
16
17
IS
19
20
21
22
23
24
25
26
27
2S
29
30
31
32
33
Carbon Monoxide
Figure 1 1 provides an illustrative
comparison of 1990 control versus no-control
scenario CO concentrations, expressed a» a
frequency distribution of the ratios of 1990
control to no-control scenario 95th percentile
1-hour average concentrations at individual
CO monitors. (Consistent withlhe«mission
changes underlying these air quality insults,
CO concentrations under the ccffita!^ scenario
tend to be about half those projected Jinderv
the no-control scenario, with most individual
monitor ratios rangmg from about 0.40 to
0.60 percent, and a few wftteraiiosjin the 0.60
to 0.80 range. //I®?
Figure 11. Frequeoey Distribution for 1990 Control to No-
xmtrdl Scenario 95uYPercentile 1-Hour Average CO
Concentrations, by Monitor.
300
0.05 055 0.45 0.65 0.85 1.05 1.25
Ratio of CAAtto-CAA 93th Percentile 1-Hour Average
In considering these results, his
important to note that CO is essentially a "hot
spot" pollutant, meaning that higher concentrations tend to be observed hi localized areas of relatively
high emissions. Examples of such areas include major highways, major intersections, and tunnels. Since
CO monitors tend to be located in order to monitor the high CO concentrations observed in such
locations, one might suspect that using state-wide emissions changes to scale air quality concentration
11PM data are reported as county-wide values for counties with PM monitors and a sufficient number of monitor observations.
" The actual air quality profiles, however, are available on disk from EPA. See Appendix C for further information.
" Large percentage changes can result from even modest absolute changes when they occur in areas with good initial (e.g., control
scenario) air quality. Considering percentage changes alone might create false impressions regarding absolute changes in air quality in some
areas. For example, Appendix C discusses in detail two such cases: the Upper Great Lakes and Florida-Southeast Atlantic Coast areas, which
show high percentage changes in sulfur deposition and visibility. .
24
-------
Chapter 4: Air Quality
estimates at strategically located monitors might create some bias in the estimates. However, the vast
majority of ambient CO is contributed from cm-highway vehicles. In addition, the vast majority of the
change in CO emissions between the control and no-control scenario occurs due to catalyst controls on
highway vehicles. Since CO hot spots result primarily from highway vehicles emissions, controlling
such vehicles would mean CO concentrations would be commensurately lowered at CO monitors. While
variability in monitor location relative to actual hot spots and other factors raise legitimate concerns
about assuming ambient concentrations are correlated with emission changes at any given monitor, the
Project Team believes that the results observed provide a reasonable characterization of the aggregate
change in ambient CO concentrations between the two scenarios.
10
11
12
13
14
IS
16
17
IS
19
W
22
13
24
25
26
27
21
29
X
31
32
33
34
3S
36
37
Sulfur Dioxide
As for CO, no-control scenario SO2
concentrations were derived by scaling
control scenario air quality profiles based on
the difference in emissions predicted under
the two scenarios. Unlike CO, SO2 is
predominantly emitted from industrial and
utility sources. This means that emissions,
and the changes in emissions predicted under
the two scenarios, will tend .to be :
concentrated in the vicinity of major point
sources. Again, while state-wide emissions
changes are used to scale SOj concentrations
between the scenarios, these state-wide
emission changes reflect the controls placed
on these individual point sources. TJierefbre,
the Project Team again considers me
distribution of control to no-control
Figure 12. Frequency Distribution for 1990 Control to No-
control Scenario 95th Percentile 1-Hour Average SO2
Concentrations, by Monitor,
300
0.03 0.25 0.45 0.65 0.85 1.05 125
Ratio of CAA3*o-CAA 95th PereentUe 1 -Hour Avenge
be a reasonable characterization of the aggregate results despite the uncertainties associated with
estimation of changes at individual monitors. ',
Figures 12 provides a histogram of the predicted control to no-control ratios for SO2 which is similar
to the one presented for CO. The results indicate that, on an overall basis, SO2 concentrations were
reduced by about one-thud. The histogram also shows a much wider distribution of control to no-control
ratios for individual monitors than was projected for CO. This result reflects the greater state to state
variability in SOj emission changes projected in this analysis. This greater state to state variability in
turn is a function of the variable responses of SO2 point sources to historical CAA control requirements.
This source-specific variability was not observed for CO because controls were applied relatively
uniformly on highway vehicles.
41
Nitrogen Dioxide
»
Results for NO2 are presented in Figure 13. These results are similar to the results observed for CO,
and for a similar reason: the vast majority of change in NO2 emissions between the two scenarios is
related to control of highway vehicle emissions. While baseline emissions of NO2 from stationary
25
-------
Chapter 4: Air Quality
i sources may be significant, these sources
2 were jubject to minimal controls during the
3 historical period of this analysis. On an
4 aggregated basis, overall NO2 concentrations
5 are estimated to be roughly one-third lower
6 under the control scenario than under the no-
7 control scenario.
* Rartlculato Matter
9 An indication of the difference in
w outcomes for TSP between the two scenarios
n is provided by Figure 14. This graph shows
12 the distribution of control to no-control ratios
13 for annual mean TSP in 1990 for those
14 counties which both had paniculate monitors
is and a sufficient number of observations from
16 those monitors.34 While the distribution of
n results is relatively wide, reflecting
it significant county to county variability in
a ambient concentration, on a national . ~
20 aggregate basis particulate matter
21 concentrations under the control scenario
22 were just over half the level projected under
23 the no-control scenario. The significant
24 county to county variability observed in this
2i case reflects point source-specific controls on
26 particulate matter precursors, especially SOj,
27 and the effects of long-range transport and*
2« transformation. :; -
Figure 13. Frequency Distribution for 1990 Control to No-
il Scenario 95th Peicentile 1-Hour Average NO2
, by Monitor.
300
100 .
0.05 0.23 0.45 0.65 O.S5 1.05 1.25
Ratio of CAAJto-CAA 95th Percentiie 1-Hour Average
Figure 14. Distribution of 1990 County-Level Annual
Mean TSP CAA to No-CAA Ratios.
0.00 030 0.40 0.60 0.80 1.00
Ratio of CAANo-CAA Annutl Mean TSP (interval midpoint)
29
31
32
33
34
35
36
Urban Ozone
Figure IS presents a summary of the results of the 1990 OZIPM4 ozone results for all 147 of the
modeled urban areas. In this case, the graph depicts the distribution of ratios of peak ozone
concentrations estimated for the control and no-control scenarios. While the vast majority of simulated
peak ozone concentration ratios fall below 1.00, eight urban areas show lower simulated peak ozone for
the no-control scenario than for the control scenario. For these eight urban areas, emissions of precursors
were higher under the no-control scenario; however, the high proportion of ambient NOX compared to
14 Given the relative importance of particulate matter changes to the bottom line estimate of CAA benefits, and the fact that a substantial
portion of the population lives in unmonitored counties, a methodology was developed to allow estimation of particulate matter benefits for
these unmonitored counties. This methodology was based on the use of regional air quality modeling to interpolate between monitored
counties. It is summarized in Appendix C and described in detail in the SAIPM Interpolation Report (1996).
26
-------
Chapter 4: Air Quality
6
7
S
9
10
n
12
13
14
IS
16
17
19
20
21
22
23
24
25
26
27
2»
29
30
31
32
33
34
3S
36
37
ambient non-methane organic compounds
(NMOCs) in these areas results in a decrease
in net ozone production in the vicinity of the
monitor when NO, emissions increase.33
Rural Ozone
"figures 16 and 17 present frequency
distributions for control to no-control ratios
of average ozone-season daytime ozone
concentrations at rural monitors as simulated
by RADM and SAQM, respectively.
-Both the RADM and SAQM results
indicate relatively little overall change in
rural-ozone concentrations. This is primarily
because reductions in ozone precursor
emissions were concentrated in populated
areas.
Add Deposition t
Figure 18 is a contour map showing the
estimated percent increase in.sutfur
deposition under the no-control scenario
relative to the control scenario for 1990,
Figure 19 provides comparable information
for nitrogen deposition.
These results show thai i
rates increase significantly i
control scenario, particulai^iti;tfjl Atlantic
Coast area and in the vicinity of states for
which relatively large increases in emissions
are projected under the no-control scenario
(i.e., Kentucky, Florida, Michigan,
Mississippi, Connecticut, and Florida).
Figure IS. Distribution of 1990 Control to No-control
K3ZIPM4 Simulated Peak Ozone Ratios.
In the areas associated with large
increases in sulfur dioxide emissions, rates of
sulfur deposition increase to greater than or
equal to 40 percent The high proportional
increase hi these areas reflects bom the
significant increase in acid deposition
30
10
0.00 0.20 0.40 0.60 0.80 1.00 1.20
RatiDofCAAtte-CAA PeakOnae(tatervtlmUpotat)
'igure 16. Distribution of 1990 Control to No-control
RADM Simulated Ozone Ratios.
200
ISO
100
0.00 0.20 0.40 0.60 O.JO 1.00 1.20
fatfeofCAA Jfo-CAA Ome-SeuoBDiTttae Avenie Ome (kmivalmUpota)
Figure IT. Distribution of 1990 Control to No-control
SAQM Simulated Ozone Ratios.
10
0.00 0.20 0.40 0.60 OJO 1.00 1.20
bib of CA A tto-CA A Ome-Seura D»yttae Avenge Ome (tale nrtl midpoint)
" Over m unbounded geographic atea, NO, reductions generally decrease net ozone production.
27
-------
Chapter 4: Air Quality
10
n
12
13
14
IS
16
17
II
19
20
21
22
23.
24
25
26
27
21
29
30
precursor emissions in upwind areas and the
relatively low deposition rates observed under the
control scenario.36
Along the Atlantic Coast, 1990 nitrogen
deposition rates increase by greater than or equal
to 25 percent under the no-control scenario. This
is primarily due to the significant increase in
mobile source nitrogen oxide emissions along the
ma urban corridors of the eastern seaboard.
Visibility
The difference in modeled 1990 control and
no-control scenario visibility conditions projected
by the RADM/EM for the eastern U.S. is depicted
by the contour map presented in Figure 20. This
figure shows the increase in modeled annual
average visibility degradation, in DeciView37.
terms, for 1990 when moving from the control to
the no-control scenario. Since the DeciView
metric is based on perceptible changes In -c
visibility, these results indicate noticeable
deterioration of visibility in the eastern U.S. under
the no-control scenario. ...^ -..-
Figure 18. RADM-Predicted Percent Increase in
[Total Sulfur Deposition (Wet + Dry) Under the No-
control Scenario.
.
Visibility changes in 30 southwestern U.S.
urban areas were also estimated using emissions
scaling techniques, Tbiranalysis also {Quad
significant, perceptiblfrchaagesgn visibility
f tills
between the two scenarios,
analysis, including the specifiiiitttcomes for the
30 individual urban areas, tie presented hi
Appendix C. :;-'
'igure 19. RADM-Predicted Percent Increase in
Fotal Nitrogen Deposition (Wet + Dry) Under die
^o-conttol Scenario.
* Even small changes in absolute deposition can yield large percentage changes when initial absolute deposition is low. See Appendix C
for further discussion of this issue.
17 The DeciView Haze Index (dV) is a relatively new visibility indicator ainKd at roeasiirmgYisibaity changes in terms of human
perception. It is described in detail in Appendix C.
28
-------
Chapter 4: Air Quality
Figure 20. RADM-Predicted Increase in Visibility
Degradation, Expressed in DeciViews, for Poor
Visibility Conditions (90th Percentile) Under the No-
XJntrol Scenario.
29
-------
Chapter 4: Air Qualify
Table 7. Key Uncertainties AMOciated with Air Quality Modetog> •
Potential Source of Error
Direction**
Potential Bfaw
to Estimate
of Air
Benefit*
Significance Relative to Key Uncertainties
, in Overall Monetary Benefit Estimate
Jf
coHlrf not be
«
-------
16
25
5
Human Health and Welfare Effects Modeling
s This chapter identifies and, where possible, estimates the principal health and welfare benefits
6 enjoyed by Americans due to unproved air quality resulting from the CAA* Health benefits have
7 resulted from avoidance of air pollution-related health effects, such as premature mortality, respiratory
s illness, and heart disease. Welfare benefits accrued where improved air quality averted damage to
9 measurable resources, including agricultural production and visibility! The analysis of physical effects
10 required a combination of three components: air quality, population, and health or welfare effects. As
structured in this study, the 3-step process involved (1) estimating changes in air quality between the
control and no-control scenarios, (^estimating the humanpopulations and natural resources exposed to
n these changed air quality conditions; and (3) applying a series of concentration-response equations which
i4 translated changes in air quality to diangesm^
'u populations. -. ~~ V^x ^V
Air Quality
17 The Project Team first estimated changes in concentrations of criteria air pollutants between the
is control scenario, which at lids step was based on historical air quality, and the no-control scenario. Air
19 quality improvements resulting from the Act were evaluated in terms of both their temporal distribution
20 from 1970 to 1990 and their spatial distribution across the 48 conterminous United States. Generally, air
21 pollution monitoring data provided baseline ambient air quality levels for the control scenario. Air
22 quality modeling was used to generate estimated ambient concentrations for the no-control scenario, A
23 variety of modeling techniques were applied, depending on the pollutant modeled. These modeling
24 approaches and results are summarized in Chapter 4 and presented in detail in Appendix C.
Population
26 Health and some welfare benefits resulting from air quality improvements were distributed to
27 individuals in proportion to the reduction in exposure. Predicting individual exposures, then, was a
necessary step in estimating health effects. Evaluating exposure changes for the present analysis
required not only an understanding of where air quality improved as a result of the CAA, but also how
jo many individuals were affected by varying levels of air quality improvements. Thus, a critical
31
-------
Chester 5.- Physical Effects
i component of the benefits analysis required that the distribution of the U.S. population nationwide be
2 established.
3 Three years of U.S. Census data were used to represent the geographical distribution of U.S.
4 residents: 1970,1980, and 1990. Population data was supplied at the census block group level, with
5 approximately 290,000 block groups nationwide. Allocating air quality improvements to the population
6 for the other target years of this study -1975 and 1985- necessitated interpolation of the three years of
7 population data. Linear interpolation was accomplished for each block group in order to maintain the
s variability in growth rates throughout the country. . .
9 Health and Welfare Effects
10 Benefits attributable to the CAA were measured in terms of the avoided incidence of physical health
n effects and measured welfare effects. To quantify such benefits, it was necessary to identify
12 concentration-response relationships for each effect being considered. As detailed in Appendix D, such
u relationships were derived from the published science literature. In the case of health effects,
M concentration-response functions combined the air quality improvement and population distribution data
15 with estimates of the number of fewer individuals that suffer an adverse health effect per unit change in
16 air quality. By evaluating each concentration-response function fcff every monitored location throughout
i? the country, and aggregating me resulting pcidence estimates, ftiwas possible to generate national
is estimates of incidence under the control and no-control scenarios.
19 In performing this step of the analysis, the Project Team discovered mat it was impossible to
20 estimate all of the health and welfare benefits whidi have resuhed from the Clean Air Act While
21 scientific information was avfiUbleJo support estimation of some effects, many other important health
22 and welfare effects could not be fiftpiiM Furthermore, even though some physical effects could be
23 quantified, the state of the science did not support assessment of the economic value of all of these
24 effects. Table 8. shows the health effects |br which quantitative analysis was prepared, as well a£ some of
25 the health effects which could not be quantified in the analysis. Table 9 provides similar information for
26 selected welfare effects. ol „,;-;..,
- - ..
27 While the 3-step analytical process described above was applied for most pollutants, health effects
21 for lead were evaluated using a different methodology. Gasoline as a source of lead exposure was
29 addressed separately from conventional point sources. Instead of using ambient concentrations of lead
x resulting from use of leaded gasoline, the concentration-response functions linked changes in lead
31 releases directly to changes in the population's mean blood lead level. The amount of leaded gasoline
n used each year wa&directly related to mean blood lead levels using a relationship described in the 1 985
33 Lead Regulatory Impact Analysis (U.S. EPA, 1985). Health effects resulting from exposure to point
34 sources of atmospheric lead, such as industrial facilities, were considered using the air concentration
15 distributions modeled around these point sources. Concentration-response functions were then used to
36 estimate changes in blood lead levels in nearby populations.
37 Most welfare effects were analyzed using the same basic 3-step process used to analyze health
3a effects, with one major difference in the concentration-response functions used. Instead of quantifying
39 the relationship between a given air quality change and the number of cases of a physical outcome,
40 welfare effects were measured in terms of the avoided resource losses. An example is the reduction in
41 agricultural crop losses resulting from lower ambient ozone concentrations under the control scenario.
, i
32
-------
Chapter 5: Physical Effects
Tables. Human Hftaltfa Effects of Criteria Pollutants,
Effects
Ozone
idputeiooary
Decreased i
Mttter/TSP/
drams feroochifis
lower respinft»yalwaw
Chett
Mi
.-
-------
Chapter 5: Physical Effects
Table 9. Selected Welfare Effects of Criteria Pollutants.
Pollutant
Ozone
itofloriate Matter/
Housefeold softmg
Otta
SoIftirDtaride
Materiali damage doe to wad deposition
4
5
6
7
I
9
10
Welfare effects quantified using this general approach include household soiling damage by PM10 and
agricultural benefits measured in terms of net economic surplus.
Another important wel&xe efi^ assessed herein 1$ the benefits accruing from improvements in
visibility under die control scenario^ Again, a slightly different methodological approach was used to
evaluate visibility uiiprovemen^|nBibili1y changes were a direct output of the models used to estimate
changes in air quality,3! Hie models provided estimates of changes in light extinction, which were then
translated mathematically into various specific measures of perceived visibility change.39 These
visibility change measures were then combined with population data to estimate the economic value of
the visibility changes. Therestdts pf the welfare effects analysis are found in Chapter 6 and in
Appendices D and F.
n
12
13
14
Because of a lack of available concentration-response functions (or a lack of information concerning
affected populations), ecological effects were not quantified for mis analysis. However, Appendix E
provides extensive discussion of many of .the important ecological benefits which may have accrued due
to historical implementation of the CAA.
M These models, and the specific visibility changes estimated by these models, are described in summaiy ftshion in flie previous cluster
and are discussed in detail in Appendix C.
" These visibility measures ate described in Appendix C.
34
-------
Chapter 5: Physical Effects
Key Analytical Assumptions
2 Several important analytical assumptions affect the confidence which can be placed in the results of
3 the physical effects analysis. The most obvious concerns the choices which have to be made among
4 competing scientific studies. The Project Team relied on the most recent available, published scientific
5 literature to ascertain the relationship between air pollution and hitman heafth and welfare effects. The
6 choice of studies, and the uncertainties underlying those studies, also created uncertainties in the results.
7 For example, to the extent the published literature may collectively overstate die effects of pollution,
s EPA's analysis will overstate the effects of the CAA. Such outcomes may occur because scientific
9 research which fails to find significant relationships is less likely to be published man research with
10 positive results. On the other hand, history has shown that it is highly likely that scientific understanding
/; of the effects of air pollution will improve hi the future, resulting la discovery of previously unknown
12 effects. Important examples of this phenomenon are the substantial expected health and welfare benefits
13 of reductions in lead and ambient paniculate matter, both of which have been shown hi recent studies to
14 impose more severe effects than scientists previously believed. To the extent the present analysis misses
15 effects of air pollution that have not yet been subject to adequate scientific inquiry, the analysis may
it understate the effects of the CAA. _:_ -:/:>\
17 Because these resultant uncertainties were caused by the inadequacy of currently available scientific
is infonnation, mere is no compelh^g reason to believe that me results of me present analysis are bias^
19 a particular direction. Some significant uncertainties, however, may have arisen from interpretation of
model results, underlying data, and supporting scientific studies. To the maximum extent practical, these
-------
Chapter 5: Physical Effects
i the CAA. However, a significant portion of the ^^•"^•^••^•^•^•"••••^•^•^••^•^•^^•^•^••w
2 omitted population for a given year or pollutant TaMe 18. Pawtatdf Boptu^fJett<«f^ ^^ 39fc» o£«m however, a more sophisticated approach was used
21 which should provide more reliable extrapolation results for this important indicator pollutant Rather
22 than extrapolating concentration measurements from the nearest monitored county, the grid cell-specific
23 sulfate particle concentrations predicted by the RADM model were used to provide a scaled interpolation
24 between monitored counties in the eastern 31 states.^ For counties outside the RADM domain, an
25 alternative method based on state-wide average concentrations was used. While less certain than the
26 results obtained in the 31 eastern states, using state-wide averages provided better estimates man those
27 obtained through simple extrapolation. Wjjth mis supplemental analysis, estimates were developed of the
21 health effects of the CAA_on amic>st Ac entire continental U.S. population.41 It should be noted,
29 however, that compliance costs incurred in Alaska and Hawaii were included in mis study but the
so benefits of historical air pojution reductions were not In addition, the CAA yielded benefits to Mexico
31 and Canada which were not^captufed in this study.
32 There was a substantial difference between the results of the "50 km" and "all population" analyses.
33 For NOj, there was a 64 percent increase hi the population modeled for 1990 (i.e., 61 percent of the
34 population lives within 50 km of an NO2 monitor while 100 percent of the continental U.S. population
35 were captured using the extrapolation method). Incidence of increased respiratory illness among adults
36 in 1990 caused by NO2 exposure increased by 41 percent in the "all population" scenario relative to the
37 "50 km" scenario, A 35 percent increase in 1990 population exposure for ozone (i.e., from 74 percent
33 coverage under the 50 km limit to 100 percent using the extrapolation method) resulted in a predicted 27
39 percent increase in ozone-related asthma attacks. A 39 percent increase in 1990 population exposure for
40 SO2 (i.e., from 71 percent under the 50 km case to 98 percent for the all population case) resulted in a
4i predicted 42 percent increase in SO2-related respiratory symptoms among adults. Finally, the "all-
40 The specific methodology is described in detail in Appendix C.
41 While this alternative captures the vast majority of the U.S. population, it does not model exposure for everyone. Toirr.prove
computational efficiency, those grid cells with populations less than 1,000 were not modeled; the analysis covered more than 95 percent of the
population. •
36
-------
Chapter 5: Physical Effects
population" approach covers 45 percent more persons in 1990 than does die "monitored counties only"
2 approach for PM, resulting in a roughly equivalent increase in PM-related health effects.
3 Clearly, the "50 km" approach severely understated aggregate health effects because it failed to
4 cover 20 to 40 percent of the population. The "all-population" approach covers more of the population,
5 but projections for persons living great distances from monitors were significantly more uncertain than
6 were those for persons living close to monitors, and may misrepresent the effects on those persons. As
7 noted previously, the results for PM effects in unmonitored eastern U.S. counties were more accurate
a than the simple extrapolation results used for other pollutants given the use of the regional model results
9 to scale the interpolation to unmonitored counties. In addition, the extrapolated results for ozone should
w be more accurate than for the other non-PM pollutants because ozone is more of a regional pollutant.
n Finally, it is important to note that the "all-population" approach still fails to cover up to 5 percent of the
12 population,42 and the health benefits for that residual portion of the population were not reflected hi the
13 quantitative results of this study. Appendix D provides detailed results of both the "50 km" and "all-
14 population" analyses.
IS
Choice of Study
16 As noted above, the Project Team estimated health and welfare effects of unproved air quality
n through the use of concentration-response (CR) functions derived from the peer-reviewed scientific
it literature. For some health endpoints, however, several functions were available, and significant
19 differences were found among the CR functions in implied health effects. For example, 14 CR functions
were available correlating excess rjfejhature mortality to exposure to particulate matter (PM). It is not
21 unusual for two equally-reputable studies to differ b^afector of three or four hi implied health impact
22 The difference in implied healllt efi|cts across studies can be considered an indication of the degree of
23 scientific uncertainty associatodjf^measuremeht of that health effect
24 Where more than oneacceptable studyji»s available, me Project Team used CR functions from all
25 relevant studies to infer health effects.^Iftifis, the health effect implied by each study is reported, and a
26 range of reported results f^«|iarticular health endpoint can be interpreted as a measure of the
27 uncertainty of the estimate.
2, Variance Within Studios
_-j
29 Even where only; one CR function was available for use, the uncertainty associated with application
» of that function to estimate physical outcomes can be evaluated quantitatively. Health effects studies
31 provided ""best estimates" of the relationship between air quality changes and health effects, and a
32 measure of the statistical uncertainty of the relationship. In this analysis, the Project Team used
33 simulation modeling techniques to evaluate the overall uncertainty of the results given uncertainties
34 within individual studies, across studies examining a given endpoint, and in the economic valuation
x coefficients applied to each endpoint The analysis estimating aggregate quantitative uncertainty is
36 presented hi Chapter 7. For clarity, the results reported hi this chapter and hi Chapter 6 reflect only the
42 Ibid, footnote 41.
37
-------
Chapter 5: Physical Effects
"best estimates" of the underlying studies. In considering these "best estimate" results, however, it was
important to bear in mind mat there is a distribution of possible results around the "best estimates."
Health Effects Modeling Results
4
i
f
7
I
9
10
11
12
13
14
IS
If
17
IS
19
20
21
22
23
24
23
26
27
21
29
30
31
32
This section provides a summary of the differences in health effects estimated under the control and
no-control scenarios. Because the differences in air quality between the two scenarios generally
increased from 1970 to 1990, and the affected population grew larger during that period, the beneficial
health effects of the CAA increased steadily during the 1970 to 1990 period. More detailed results,
including results for all quantified health endpoints, are presented in Appendix D.
Table 11 provides a summary of
a subset of the health effect modeling
results. Results are expressed as
cases reduced per year as a result of
the CAA, and the pollutants
associated with these health effect
. estimates are listed in the table.
Several studies were used to =
derive alternative estimates for some --.
of the endpoints: the "high" estimate
reflects the "best estimate" of the C&
function with the greatest predicted!
effect, the "low" estimate reflects the
"best estimate" of the CR function : J
with the smallest predicted effect, T, ri
and the "mid" value is the average of -;-J
the "best estimates" of all the CR
functions for mortality.41
does not reflect the uncertainty
surrounding the 'best estimates.?
These "mid" value results should
therefore be interpreted as average
values in a much wider distribution
of possible results.- «-~
Table It Selected Heaftb Benefits of the CAA, 1970-1990, for
population ta lower 4$ States (w ttwusaacis of cases re^aced per
JidtorttBty
J&tttAttttfcf
low
mid
low
(thousands)
Mb*««
tow
1975 198ft IMS
38
20
31
,97
54
$6
7
5
6
5
124
70
40
14
,36
6
165
22
Id
8
79
4$
24
15
10
7
15
16
IS
" Results for lead (Pb) were handled differently. The Pb analysis was conducted using two baselines. The "high" and "low" values
reported here reflect the results implied by the two baselines; die "mid" value is the average of the "high" and "low" values.
38
-------
Chapter 5: Physical-Effects
30
Of her Physical Effects
2 Human health impacts of criteria pollutants dominate quantitative analyses of the effects of the CAA,
3 in part because the scientific bases for quantifying air quality and physical effect relationships are most
4 advanced for health effects. The CAA yielded other benefits, however, which are important even though
5 they were sometimes difficult or impossible to quantify fully given currently available scientific and
6 applied economic information.
7 Ecological Effects ,
s Beyond the intrinsic value of preserving natural aquatic and terrestrial ecosystems and the life they
9 support, protection of ecosystems from the adverse effects of air pollution can yield significant benefits
10 to human welfare. The historical reductions in air pollution achieved under the CAA probably led to
n significant improvements in the health of ecosystems and the myriad ecological services they provide.
n For example, improvements hi water quality stemming from a reduction hi acid deposition-related air
n pollutants such as SOX and NO, likely yielded significant benefits to human welfare through
14 enhancements hi commercial and recreational fishing, wildlife viewing, maintenance of biodiversity, and
is nutrient cycling. Increased growth and productivity of U.S. forests were likely enhanced due to reduced
16 concentrations of ambient ozone. Protejtion of forest resounds results in benefits associated with
n increased timber production, improved btttdoor recreation (e.g., hunting, camping), and ecological
services such as nutrient cycling and carbon sequestration. These improvements hi ecological conditions
19 have not been quantified hi thfeassfssment. F^ajullei^iscussion of the possible ecological effects of
20 the CAA, see Appendix E,
21
Quantified Agricultural Effects
22 Quantification of thedeffec|SNof the CAA. on agriculture was limited to effects on seven crops.
23 Changes^n crop yields betsifjifs^jlfywo scenarios were less man one percent hi 1990 for barley, corn,
24 soybeans, and sorghum.44 l€3ianjes hi peanut and cotton yields were estimated hi the four to seven
2s percent range, and winter wheat yield changes were estimated at one-half of one percent to eight percent.
26 These ranges reflect usage of alternate exposure-response functions. Exposure-response relationships
27 and cultivar mix reflect historical patterns and do not account for possible substitution of more ozone-
21 resistant cultivars in the no-control scenario. Thus, these are likely overestimates of the actual effect of
29 me(
Effects at Air Toxics
n In addition to control of criteria pollutants, the Clean Air Act resulted hi control of some air toxics —
31 defined as non-criteria pollutants which can cause adverse effects to human health and the environment.
33 Control of these pollutants resulted both from incidental control due to criteria pollutant programs and
" In fact, the effect on barley was several oidera of magnitude less than 1 percent
39
-------
Chapter 5: Physical Effects
specific controls targeted at air toxics through the National Emission Standards for Hazardous Air
Pollutants (NESHAPs) under Section 112 of the Act. .
Air toxics are capable of producing a wide variety of effects. Table 12 presents the range of
potential human health and ecological effects which can occur due to air toxics exposure. For several
years, the primary focus of risk assessments and control programs designed to reduce air toxics has been
cancer. According to present EPA criteria, there are over 100 known or suspected carcinogens. EPA's
1990 Cancer Risk study indicated that as many as 1,000 to 3,000 cancers annually may be attributable to
the air toxics for which assessments were available (virtually all of mis estimate came from assessments
of about a dozen well-studied pollutants).45
taible 12x
!E«fecte
Otter Boss0)Ie Effects
Cancer Mortality
Cancer Mortality
swaee
-mohflesource
Btoian Welfare
Decreased Income xnt
jecttatkiaapiwr&aifle*
fest&ittg fifcatt decteased
Ecoiogleal
Uwofldologieal
Other Weffiu*
70
72
In addition to cancer, these pollutants can cause a wide variety of health effects, ranging from
respiratory problems to reproductive and developmental effects. There has been considerably less work
done to assess the magnitude of non-cancer effects from air toxics, but one survey study has shown that
41 U.S. EPA, Cancer Risk from Outdoor Exposure to Air Toxics. EPA-450/1-90-004C Prepared by EPA/OAR/OAQPS.
40
-------
5
S
7
8
12
13
14
15
16
17
IS
19
1.1
22
23
24
Chapter 5: Physical Effect*
'tip.
41
-------
Chapter 5: Physical Effects
Table 13. Uncertainties Associated with Physical Effects Modeling,
Potential Source of Error
Direction of Potential
Bias 111 Physical Eftfecte
Estimate
Significance Relative to Key
Uncertainties to Overall Monetary
Extrapolatioa of health e8ec& to
distant ihTOBstoutors
<
-------
6
'conomic Valuation
3
4
3
6
7
I
9
10
11
14
15
16
17
It
19
20
21
22
23
24
2!
26
27
2t
29
Estimating the reduced incidence of physical effects represents a valuable measure of health benefits
for individual endpoints; however, to compare or aggregate benefits across endpoints, the benefits must
be monetized. Assigning a monetary value to avoided incidences of ia individual effect permits a
summation, in terms of dollars, of monetized benefits realized as a resultTof the CAA, and allows that
summation to be compared to the cost of the CAA. -~- --^:.-:-^ - i_ '_
For the Section 812 analysis of health benefits, vduation estinjtates were obtained from the economic
literature, and are reported in dollars per case-reduced fir health effects and dollars per unit of avoided
damage for welfare effects.4* Similar to estimates of physical effects provided by health studies, the
monetary values of benefits for this study;are reported bom hi terms of mean values as well as a
probability distribution of estimates* pliis permits an evaluation of some aspects of the uncertainty
associated with the point estitnates^IThe distribM
-------
Chapter 6: Economic Valuation
i
2
3
4
S
6
7
8
9
10
11
12
13
14
15
16
17
II
19
20
2J .
22
23
24
25
26
27
21
29
30
31
32
33
34
35
36
37
38
an environmental effect. For example, the value of an avoided respiratory symptom would be a person's
WtP to avoid that symptom.49
For most goods, WtP can be observed by examining actual market transactions. For example, if a
gallon of bottled drinking water sells for one dollar, it can be observed that at least some persons are
willing to pay one dollar for such water. For goods that are not exchanged in the market, such as most
environmental "goods," valuation is not so straightforward. However, value may be inferred by
observed behavior, such as through estimation of the WtP for mortality risk reductions based on
observed sales and prices of safety devices such as smoke detectors. Alternatively, surveys may be used
in an attempt to elicit directly WtP for an environmental improvement
Wherever possible, this analysis used estimates of the WtP among the U.S. population to avoid an.
environmental effect as the value of avoiding that effect. In some cases, such estimates were not
available, and the cost of .
mitigating or avoiding the mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm*
effect was used as a rough
estimate of the value of
avoiding the effect For
example, if an effect results
in hospitalization, the
avoided hospitalization '
costs were considered as a
possible estimate of the
value of avoiding the effect
Finally, where even the
"avoided cost" estimate
was not available, the
analysis relied on otter
Table 14. Monetized Htdth and WetferaESectt (1990 dollan).
available methods to ::
provide a rough • ~ °-->jfi
approximation of WtP, A& :
noted above, this analysis ^
used a range of values for :;':
most environmental effects,
or endpoints. Table 14
provides a summary of the
mid-range values used for
per unit valuations; The
full range of values can be
found in Appendix I.
Heart Attack*
n
**
O6f-CIIME
$« per cm
«**
$45
'
Exttactton
Direct Economic
" The "compensation principle" is used implicitly for cost calculations as well. In the case of the CAA, compliance cost represents the
compensation required to elicit voluntary provision of labor and capital services, supplies, etc., to meet CAA requirements.
44
-------
Chapter 6: Economic Valuation
i Wllllngnoss-to-Ray Estimates
2 Mortality
3 Some forms of air pollution increase the probability that individuals will die prematurely. The
4 concentration-response functions for mortality used in this analysis report this increase in mortality risk
5 as cases of "excess premature mortality" per time period (e.g., per year). It is not possible to compensate
s the victims; thus, if s not possible to "value" the lives of individual victims in a benefit-cost sense. One
7 can, however, determine the WtP of individuals to accept relatively small reductions in mortality risk
< (conversely, one can determine the compensation required for individuals to accept small increases in
9 mortality risk). Since air pollution affects mortality rates by changing the mortality probability (per time
16 period) for many individuals, it is standard practice in cost-benefit analyses to value this effect of air
n pollution by examining the WtP for small reductions m morta^ risk. For expository purposes, this
12 valuation is expressed as "dollars per mortality avoided," even moug&1he actual valuation is based on
13 small changes in mortality risk.10 For this analysis, the Project Team surveyed me economics literature
14 to determine an appropriate valuation for avoided mortality, and used a range of values with an
is arithmetic average of $4.8 million per mortality avoided; * : j -
16 The concentration-response functions used to estimate mortality effects of air pollution examined the
n effects of high-pollution incidents on mortality rates during and immediately following the incidents.
it This approach is not ideal for the purpoSfjaf benefit-cost analysis:
It may overstate the mortality ef&ct of air pollution insofar as it includes as "excess premature
20 mortalities" some individuals wjio were vep||&4nd would have died within a few days or
21 weeks even without the air fbtfution evenfc§|l||i!piluation method used in this analysis was
22 based on SpiHfor reducedinbfilpi^risk of individuals who are, generally, healthy and with an
23 expectation of inaityyearaoil^^ Applying that valuation to extremely ill
24 individuals may overstate beneffisl "
--.^--;--ft ff ' --1
23 --- * . -or-
26 It may understate the motjliyeffect of air pollution insofar as it fails to address years of life
27 lost due to long-term exposure to air pollution. Air pollution can have both acute and chronic
2s effects. The CR functions used address cflty me "acute," or short tenn, effects of air pollution on
29 mortality rates. It isjwssible mat longer term effects exist, causing "premature mortality" that
» does not necessarily occur during or immediately following a high-pollution event
31 Ideally, instead oifvaluing changes in short-term mortality risk, one would value changes in the risk of
32 incurring a health effect which imposes a specific reduction in life expectancy. Unfortunately, the state
33 of health science is not advanced enough to provide reliable "life-shortening" estimates, so the Project
34 Team estimated and valued me expected number of "excess premature mortalities" caused by air
33 pollution.
50 For example, if the WtP for a one percent reduction in mortality risk is $20,000, then the "benefit per life saved" would be [$20,000 x
100-] $2 million.
45
-------
Chapter 6: Economic Valuation
i Survey-Based Values . )
2 The use of survey methods (such as the "contingent valuation" method) to determine willingness-to-
3 pay for environmental improvement is controversial within the economics profession. In general,
4 economists prefer to infer WtP from observed behavior. There are times when such inferences are
5 impossible, however, and some type of survey technique may be the only means of eliciting WtP.
6 Economists' beliefs regarding the reliability of such survey-based data cover a broad spectrum, from
7 unqualified acceptances of the results of properly-conducted surveys to outright rejections of all survey-
s based valuations. ,
9 For this analysis, unit valuations based on the. contingent valuation method have been used for
10 respiratory illnesses and symptoms, restricted activity days, and asthma attacks. The Project Team
n reviewed the relevant economic literature to define a range of values for those health endpoints, and
12 calculated "mid-range" estimates. Those mid-range estimates are summarized in Table 14, above.
n Although a large fraction of the monetized health endpoints rely on survey-based valuations, the
14 aggregate benefit calculations are insensitive to these values.31
15
Avoided Cost Estimate*
is For some health effects, WtP estimates were not available, and the Project Team instead used "costs
n avoided" as a best estimate of WtP. Presumably, willingness-to-pay to avoid a health effect that would
is require hospitalization would be equal to the cost of hospitalization plus the value of avoided pain and
19 suffering, lost leisure time, etc. Consequently, the cost of hospitalization would be a lower-bound
20 estimate of one's WtP to avoid a health effect In a society where the cost of hospitalization is not paid
21 directly by the patient, however, one's WtP to avoid a health effect may be completely independent of
22 the hospitalization cost avoided; tiius^ it is unclear whether "cost avoided" understates or overstates WtP
23 to avoid the health effect32 :; ?
24 This analysis used Avoided costs* to monetize benefits for hospital admissions (the analysis uses
25 three categories of hospital admissions) and "Changes in the Incidence of IQ<70" as a result of
26 diminished cognitive devetopmeaidp young children due to Pb exposure (unit values are summarized in
27 Table 14). In both cases, "avoMed costs" must be recognized as a very rough approximation of WtP to
21 avoid those health effects. In me IQ<70 case, the "avoided cost" is the estimated incremental cost of
29 part-time special education for children in regular classrooms as opposed to in special education
x programs for twelve years (i.e., grades one through twelve).
91 That is, changing the unit values derived through the contingent valuation method has little effect on the overall level of benefits found
in this analysis. The health endpoints that are monetized through the UK of CVM studies account for approximately 1 percent of all monetized
benefits in this cost-benefit analysis. • ,
52 If it could be calculated, one would ideally count as'a benefit of environmental improvement WtP to avoid pain and suffering, lost
leisure time, etc. Avoided hospitalization costs, whether paid directly by the patient or communally (e.g., through an insurance pool), would be
counted as a negative cost (i.e., a reduction in the cost of environmental controls, just as reduced automobile maintenance costs are counted as a
"negative cost" of the CAA). •
46
-------
Chapter 6: Economic Valuation
i Other Valuation Estimates
2 Heart Attacks and Strokes
3 In its review of the economics literature, the Project Team found no studies that estimated WtP to
4 avoid the risk of non-fatal heart attacks and strokes. Although it would be possible to estimate the
5 avoided hospitalization costs due to decreased incidence of heart attack and stroke, such an approach
6 would almost certainly provide a gross underestimate of the WtP to avoid those conditions. Instead, the
7 Project Team used a valuation derived for a different medical condition (chronic bronchitis), assuming
s that it would provide a rough approximation for WtP to avoid heart attacks and strokes. Based on a
9 review of the literature, the Project Team believes mat a reasonable estimate for WtP to avoid the risk of
10 chronic bronchitis is 587,500 dollars. The Project Team has therefore assumed that the WtP to avoid the
;; risk of a case of heart attack or stroke is also 587,500 dollars, / v :='
12 Work Loss Days : ::- -r=S=?
13 For this analysis, it was assumed mat the median daily 1990 wage income of 83 dollars was a
14 reasonable approximation of WtP to avoid a day of lost work. Alhough a work loss day may or may not
is affect the income of the worker, depending on the terms of employment it does affect economic output
16 and is thus a cost to society. Conversely, avoiding the work loss day is a benefit
n Hypertension ,' :s -iv-
11 As with heart attacks and strokes, the Project Team found no studies which estimate WtP to avoid
19 hypertension (high blood pressurep^tudies do exist, however, which estimate the medical costs and
20. work loss o^ys associated wftlihyiie^eDsion. Although mere are further effects that would affect an
21 individual's WtP to avoid hypertMti^||i,g., the value of pain and suffering, the value placed on dietary
22 changes), mis analyst assumes thatpeisiiiii of medical costs (per year) and work loss costs per year
23 (valued as discussed abc«fe||| a reasona|]|fapproximation of the WtP per year to avoid a case of
24 hypertension. ™
32
as Lost IQ Points
26 One of the major effects of Pb exposure is permanently impaired cognitive development in children.
27 No ready estimates of society's WtP for improved cognitive ability are currently available. However,
28 estimates of the relationship between IQ and lifetime earnings as well as estimates of the IQ effects of Pb
29 exposure were available. Although mis is a very crude estimate (and likely underestimate) of WtP for
x improved cognitive development in children, this study assumed that changes in lifetime earnings equal
31 to $5,551 per IQ point change were a reasonable approximation of WtP.
Aggregation of Benefits
33 The total monetized economic benefit attributable to the CAA was derived by applying the unit
valuations discussed above to the stream of physical effects calculated for the 1970 to 1990 period.
,., Some aggregation difficulties arose, however, which required modification to the physical effects and/or
47
-------
Chapter 6: Economic Valuation
i
2
3
4
S
6
7
S
9
10
11
12
13
14
15
16
valuation data to avoid double-counting of benefits. For example, some studies that were used as sources for
physical effects estimates provided cases of hospital admissions for pneumonia, while other studies provided
cases of hospital admissions for all respiratory problems. In this analysis, double-counting of effects was
avoided by assuming that the two sets of studies provided alternative estimates of the same health impact
(i.e., they were not treated as additive). Similarly, the Project Team had access tojgeveral studies estimating
the effect of air pollution on "restricted activity days" of various degrees of severity as wett as studies that
estimated lost work days. To avoid double-counting, the various "restricted activity* health effects were
combined into a single benefit category.
Figure 21 summarizes the results of the aggregation of monetized benefits for the eatira population of the
lower 48 States in each of the target years of this analysis and compares these results with the total direct
compliance costs incurred in those years. Each of the bars representing total monetized benefits in the figure
include a high, mid, and low estimate. These alternative estimates reflect die fact that, for some health
endpoints, several studies were available for use in estimating physical e&ecte* In addition, for some other
endpoints more than one method of calculation was used. For those endpointa where more than one estimate
was available, the lowest estimate, highest estimate, and average of all available estimates were considered.
The "low" lines in Figure 21 reflect the sum of the benefits estimated by using lie low estimate" physical
{Figure 21. Aggre
900-
800-
V JVAA
Ctttt / BcaeOtt (bUU* t)
i i i i i i
.tuv
100 -
0.
NOTE:
criteria]
(e.g. ion
gate Direct Costs and Monetized Benefits of theCletB Air Act (at billions of 1990 dollars).
1 - •
_ . ftMtfi
— •
1975
ift flMj
- -
•M
tar __^_
19M
nlbtanta. Ako omitted an non-ipuntifiabk
IB avoided advene ecological eflbetf ).
MMl
bend
"»*F>
•M
( 1
ftMM
MkMM
nM
tar
1 1
1985 199C
n mnn^W.kU K-«-4V. .»«« 4mm r^Mtvnl nf rril
bad*
«M
-
tar
»
other than the ta
avpoQutanta
48
-------
Chapter 6: Economic Valuation,
9
10.
11
12
13
14
15
16
effect studies; the "mid" lines reflect the numerical average of results based on all available effect studies,
and the "high" lines reflect the sum of benefits using the "high estimate" studies.
It is important to note that these estimates do not reflect the combined uncertainties in physical effects
and economic valuation of those effects. Rather, the results presented in this chapter use mid-range estimates
for economic valuation of effects. Therefore the ranges presented reflect only the uncertainties associated
with the selection of physical effect studies. Broader ranges of total monetized benefit estimates which also
reflect uncertainties in economic valuation are presented in Chapter 7.
The total costs and benefits accumulated over the entire 20 year period of this study are presented in
Figure 22. The summary estimates for both costs and benefits are derived by summing the yearfy cost and
benefit estimates adjusting those yearly values to reflect their equivalent value in 1990.1 This summary figure
also clarifies the substantial number of benefits which could not be included in the total monetized benefit
estimate. Including these and other nonmonetized benefits would; increase the high, mid-range, and low
benefit estimates by unknown degrees. In addition, as noted in the preceding paragraph, these total benefit
estimates do not reflect uncertainties in economic valuation of effects. More detailed results, including those
which indicate the relative contribution of individual endjpoints to the total monetized benefit estimate, are
presented in Appendix I. /
Figure 22. Comparison of Total 1970 to 1990 Clean Air ActMonetizedUenefits and Direct Costs (in billions of 1990
dollars and in 1990 Present Value terms). J
ailiool
12,000
10,000
8000
6000
4000
2000
0
S423 tfflun
-------
(This page intentionally blank]
50
-------
7
Uncertainty in the Benefits
4 Some degree of uncertainty was associated with every step of this analysis, from cost estimation
5 through economic valuation of benefits. The results were most sensitive, however, to uncertainties in the
6 physical effects estimation and economic valuation steps. The Project Team evaluated these two key
7 areas of uncertainty through a quantitative analysis of the distribution of likely results, and a discussion
s of some of the health effects excluded from the analysis.
9 Modeling Uncertainty of Health Effects and Valuation
••> Different published results reported in the scientific literature on the adverse health effects associated
with elevated levels of air pollution typically do not report identical findings. In some instances the
12 differences are substantial, and considerable controversy exists about the magnitude of virtually all of the
n suspected effects. Given the level of uncertainty associated with the estimates of any of the
14 concentration-response relationships, it was difficult to identify a single relationship that best represents
is the nationwide impact of pollution on human health. Rather than try to identify a single "best estimate,"
16 two different approaches were used in this assessment to estimate the magnitude of the effects, and to
n fairly represent the uncertainty in current scientific opinion.
is The first approach identified a set of concentration-response relationships drawn from recent
19 scientific literature which included the range of estimated effect magnitude for each known or suspected
20 healtib endpoint. This approach produced multiple estimates of the avoided incidence of health effects.
21 . The set of point estimates-which make up this range of estimates, both in terms of the incidence of the
22 health effect and in monetary terms, indicate not only the range of credible estimates, but the extent to
23 which there is a clustering of estimates hi some portion of the range. This analytical approach was used
24 to derive the ranges of results presented hi Chapters S and 6.
25 The second approach used in this assessment relied on a Monte Carlo method to evaluate the effect
26 of combining the uncertainties in the concentration-response function and the economic value for each
27 endpoint. The first step in the Monte Carlo technique used here was to randomly select a study from the
2s set of available studies for a particular health effect Each study provided, when selected by the Monte
29 Carlo model, an estimate of the concentration-response function and its standard deviation. In the
so second step, a normal statistical distribution was derived based on the mean and variance of the
coefficient associated with the selected study. A specific value for the coefficient was then randomly
' selected from the derived normal distribution- This two-step procedure was repeated to produce a large
33 number of estimates. Because the value of the coefficient varied from iteration to iteration due to the
51
-------
2
3
4
5
6
7
8
9
10
11
12
two random selection procedures described above, the predicted health effects varied as well. The Monte
Carlo method thus produced a distribution of predicted health effect outcomes under the control and no-
control scenarios.
Concurrently, this two-stage Monte Carlo technique was used to estimate the range of monetary
benefits associated with the avoided health effects. The monetary values assigned to each health or
welfare endpoint were also measured with considerable uncertainty, and a distribution was used to reflect
the uncertainty of monetary values. Each iteration of the Monte Carlo assessment picked a specific level
for the monetary value of the endpoint from the assigned distribution of values. The estimate of the
monetary benefits for changes in the health effect incidence for that Monte Carlo iteration was thus
composed of a random draw of the estimated change in incidence multiplied times an independent
random draw of the monetary value of the health effect Figure 23 illustrates the results of the Monte
Carlo analysis of benefits for the year 1990.
figure 23. Distribution of 1990 Monetized Benefits of CAA (in billions of 1990 dollars).
5th percentile Monte Carlo » $216B
50th percentile Monte Carlo * $489B
95th percentile Monte Carlo - $1,020B
8 §
§
V
5th percentile 50th percentile
t-
I
V
I
V
*-
O
o
-------
/ range captures the true value for total 1990 monetized benefits. The 90 percent credible range for total
2 1990 monetized benefits estimated in this analysis was from 216 billion to 1.02 trillion dollars.
i Similar 50th percentile and 90 percent credible interval estimates were also derived for the other
4 target years of the study: 1975,1980, and 1985. To estimate the 50th percentile and 90 percent credible
s interval for total monetized benefits accruing over the entire 1970 to 1990 period, the Project Team
6 developed estimates for intervening years through linear interpolation between target years. For.
7 example, results for 1971 through 1974 were derived by linearly interpolating between the 1975 results
« and the 1970 result (which equaled zero in this case since the control and no-control scenarios were
9 assumed to be equal at the start of the study period). Finally, the results for each year were adjusted to
jo their equivalent value in 1990 and summed. The 50th percentile estimate for total 1970 to 1990
n monetized benefit of the Clean Air Act derived by this method was 6.8 trillion 1990-value dollars. The
n 90 percent credible interval was 2.7 to 14.6 trillion 1990-value dollars.
13
Stud/09 and Health Effects Excluded from the Analysis
u This assessment was designed to be as comprehensive as possible, and to include "best estimates" of
is air pollution effects even when there were significant uncertainties surrounding those estimates.
16 Generally, the assessment quantified all physical effects where concentration-response relationships were
i? found in the published literature. Two categories of effects for which concentration-response
it relationships were found, however, were excluded because of inconsistencies in the characteristics of the
19 effects studied in the literature and the effects estimated in the context of the present analysis. These two
20 effects were (1) premature mortality predicted through "cohort" studies and (2) chronic bronchitis.
_"="jjr "~ • -. -ife.
Avoided Premature Mortality - Cohort Studies .- -V" -
22 As noted in Chapter 6, tte preferred basis for valuing avoided premature mortality in the context of
23 cost-benefit analysts would belne}iil}ingness-to-pay for some probability of an extension of life years,
24 which would then be applied to a measure of "life-years lost" due to long-term, rather than episodic,
25 exposure to air pollution. Some research Hasbeen published which addresses the mortality impacts of
26 long-term exposure to waf pollution, n|pai|icular due to exposure to ambient particulate matter. Use of
27 the concentration-response relationships found in these studies (referred to herein as the "cohort
2< studies") would allow estimation o| average life-years lost across die population as a whole.
29 The Project Team, howevef> was not satisfied by any of the methods proposed for converting the
x concentration-response relationships defined by the cohort studies to forms which could be used for the
31 present study. To evaluate the sensitivity of the estimates of total monetized benefit to this exclusion, an
32 attempt was made to derive some estimate of the value of estimated reductions in mortality consistent
33 with the inferences of the cohort studies. Specifically, concentration response relationships were inferred
34 from the cohort studies and the "life years lost" results were converted into "premature mortalities
35 avoided." The resulting estimate of premature mortalities avoided were then valued using the same
36 valuation methods used for the primary results presented herein.
37 The sensitivity analysis was conducted by, first, estimating total 1970 to 1990 monetized benefits in
3i 1990-value dollars absent all particulate matter-related mortality and reflecting numerical averages of
39 alternative outcomes based on alternative studies of a particular effect (Note that this estimate differs
40 from the 50th percentile Monte Carlo result presented as the expected benefit estimated by this study,
, although they are very close.) The second estimate was identical except that particulate matter mortality
was added by applying equal weights to the alternative mortality outcomes predicted by the fourteen
43 "episodic" particulate matter mortality studies. The third estimate was identical to the second estimate
. 53
-------
1
2
3
4
1
6
7
t
9
10
11
12
13
14
IS
If
17
IS
19
21
22
23
24
25
26
27
a
29
30
31
32
33
34
33
36
37
31
39
40
41
42
43
44
except that the two cohort mortality studies
for particulate matter were included with the
episodic studies. The two cohort studies
were given weights equal to the fourteen
episodic studies in this third estimate, which
significantly dampened the effect of
including the cohort studies. The fourth
estimate derived was identical to the second
estimate except that the episodic studies were
dropped and only the two cohort studies were
used to estimate particulate matter mortality.
These four estimates are presented in Table
15.
TabielS. Sensitivity of 1970 to 1990 Ail 48 Stote
Bs^
value dollars}*
fetinute t
ftofr
It fflffurnlir ffrodftrt wtfift TdiHff fltttittit
The clear implication of the sensitivity
analysis was that the inclusion of the cohort studies would significantly increase the estimated total
monetized benefits of the Clean Air Act However, further research is needed to improve the potential
interpretation and incorporation of the results of cohort studies for benefit-cost analyses. Based on
currently available information, the Project Team was not confident enough in its interpretation method
to include the results of the cohort studies in the primary results of this assessment
20 Chronic Bronchitis
J2tt ..JM.
565
495
afefc«rtfaate *»$
Evidence indicates there may be a relationship
between exposure to PM10 and development of
chronic bronchitis. As with other health endpoints
studied for this study, the Project Team identified
the published scientific titeraturj addressing PM»~ j
and chronic bronchitis, and mfen^ concentration-
response relationsh^ from mosestud^ As with
the other health endpoints studied^ nuige of
estimated mcidence oiutcomes associated jvrim the
control and no-controlscoiirios were developed.
The differences between the« scenarios in me
estimated outcomes for chronic1}|qnchitis, ^^^^^^^^^^^^^^^^^^^^^^
reflecting low, high, and mid estimates similar to
those derived for other endpoints in Chapter 5, are
summarized in Table 16.
The Project Team also developed a unit value for avoidance of a case of chronic bronchitis. Based
on a review of the economics literature and recommendations from the Science Advisory Board
Advisory Council on Clean Air Compliance Analysis, the Project team used a value of $587,500 per
case avoided to translate avoided incidences into estimated monetized benefit Using these values, a
sensitivity analysis was performed to evaluate the potential significance of including chronic bronchitis
reduction benefits in the total 1970 to 1990 monetized benefit estimate. The first estimate is identical to
Estimate 2 presented in Table 15. The second estimate is identical to this estimate with the total 1970 to
1990 estimated monetized value of chronic bronchitis reductions added. The results of this sensitivity
analysis are summarized in Table 17. ,
54
-------
4
s
6
7
a
9
10
11
n
13
14
is
Table 17. Sensitivityof 1970 to 1990 All 48 State
Inclusion of Chronic Bronchitis (in billions of 1990-vaIue
dollars),
WOMtl;
Upon further review, it became apparent
that the "chronic bronchitis" valued in the
economics literature is far more severe than
the "chronic bronchitis" predicted in the
health effects literature used by in the present
study.54 There is almost certainly a chronic
bronchitis health effect which could,
theoretically, be valued. At this time,
however, the Project Team considers the
valuation procedure so uncertain that it
borders on speculative. Consequently, the iiiillBBiBI—BBi—BB-IB-ipiBB--BiBB|^^
Project Team did not include benefits from -
avoided cases of chronic bronchitis in the primary estimate of .benefits presented in this study. As for the
particulate matter-related cohort mortality studies, additional research is needed to improve monetary
characterizations of these benefits and facilitate their incorporation in benefit-cost studies.
54 With an average of one-half million new cases of chronic bronchitis each year for over a decade, the total number of persons with
chronic bronchitis would approach two to three percent of the U.S. population by 1990. It was considered unlikely that two percent of the
population could develop a health condition of the severity of the chronic bronchitis valued in the economics literature:
55
-------
[This page intentionally blank]
56
-------
Appendix As Cost and
Macroeconomic Modeling
3 The purpose of this appendix is to describe in detail the estimation of direct compliance costs
4 associated with the CAA and the effect of those expenditures on U.S. economic conditions from 1970 to
5 1990. The first section of this appendix describes the dynamic, general equilibrium macroeconomic
6 model used to examine economy-wide effects. Two broad categories of models were considered for use
7 in the assessment: Macroeconomic forecasting models (e.g., the Data Resources Inc. model of the U.S.
« economy), and general equilibrium models (e.g., Hazilla and Kopp [1990], and Jorgenson and Wilcoxen
9 [ 1990a]). The project team selected the Jorgenson-Wilcoxen (J/W) general equilibrium model of the
10 United States for this analysis (Jorgenson and Wilcoxen [1990a]). There are two main reasons for
11 choosing a dynamic general equilibrium approach: To capture both the direct and indirect economic
12 effects of environmental regulation, and to capture the long-run dynamics of the adjustment of the
n economy. The general equilibrium framework enabled the project team to assess shifts in economic
u activity between industries, including changes in distributions of labor, capital, and other production
is factors within the economy, and changes hi the distribution of goods and services.
a The second section describes the data sources for direct compliance expenditures and presents
n estimates of historical air pollution control expenditures. These estimates are derived primarily from
is EPA's 1990 report entitled "Environmental Investments: The Cost of a Clean Environment"13 (hereafter
referred to as Cost of Clean). Specific adjustments to the Cost of Clean stationary source and mobile
source O&M data needed to adapt these data for use in the present study are also described. These
21 adjusted expenditure estimates represent the compliance cost data used as inputs to the J/W model to
22 determine macroeconomic effects.
~ s" -"- -. ~~~ ~~~ ~
23 The final section presents a summary of the direct expenditure data, presents direct costs in a form
24 that can be compared to me benefits estimates found elsewhere hi the study, and discusses indirect
25 effects estimated by the maicroeconomfc model. The effects reported by the model are indirect costs and
26 sectoral impacts; the model hap not estimated the indirect benefits attributable to improvements in
27 environmental quality, or avoitoW of environment^ or natural resource degradation. That is, the
2s benefits that result from a cleaner environment are not captured and allowed to propagate through the
29 economy as are compliance costs. For example, air quality improvements leading to improvements in
x human health translate into reduced medical expenditures. These savings hi medical expenditure would
31 then flow into alternative economic activities. A fully integrated model would contain environmental-
32 economic relationships such as this so there would be a simultaneous accounting for changes hi costs and
33 benefits. However, no existing macroeconomic model integrates all market and non-market costs and
34 benefits of environmental programs.
" Environmental Investments: The Cost of a Clean Environment, Report of the Administrator of the Environmental Protection Agency to
the Congress of the United States, EPA-230-11-9WH3, November 1990.
57
-------
Appendix A: Cost and Macroeconomic Modeling
Macroeconomlc Modeling
i
i EPA analyses of the costs of environmental regulations typically quantify the direct costs of
3 pollution abatement equipment and related operating and maintenance expenses. However, this
4 approach does not fully account for all of the broader economic consequences of reallocating resources
5 to the production and use of pollution abatement equipment A general equilibrium, macroeconomic
6 model could, in theory, capture the complex interactions between sectors in the economy and assess the
7 full economic cost of air pollution control. This would be particularly useful for assessing regulations
» that may produce significant interaction effects between markets. Another advantage of m general
9 equilibrium, macroeconomic framework is that it is internally consistent. The consistency of sectoral
10 forecasts with realistic projections of U.S. economic growth is ensured since they are estimated within
n the context of a single model.36 This contrasts with typical EPA analyses that compile cost estimates
n from disparate sectoral and partial equilibrium models. ~'.:-f;^-.
13 The economic effects of the CAA may be over- or underestimated, if general equilibrium effects are
14 ignored, to the extent that sectors not directly regulated are affected. For example, it is well known that
is the CAA imposed significant direct costs on the energy industry. Economic sectors not directly
16 regulated will nonetheless be affected by changes in energy prices. However, an examination of the
n broader effects of the CAA on the entire economy might reveal mat the CAA also led to more rapid
is technological development and market penetration of environmentally "clean" renewable sources of
19 energy (e.g., photovoltaics). These effects would partially offset adverse effects on the energy industry,
20 and lead to a different estimate of the total economic cost to society of the CAA.
-5.-. 4-, - .---V..-: i=7 "i^,
21 The significance of general equilibrium effects in the context of any particular analysis is an
22 empirical question. Kokoski and Smith (1987) used a computable general equilibrium model to
23 demonstrate that partial-equilibrium welfare measures can offer reasonable approximations of the true
welfare changes f^ large exogenous chaa^ss. In contrast, the results of Jorgenson and Wilcoxen
(1990a) and Hazflla and Kfigip (1990) suggest that total pollution abatement in the U.S. has been a major
claimant on productive resources, and die effect on long-run economic growth may be significant
27 Again, such conclusions imtstl^ppnsidered in light of the limitations of general equilibrium models.
23 One significant limitation o^current macroeconomic modeling tools is mat they only capture those
29 effects that manifest themselves in the market Furthermore, they may not even capture all potentially
24
25
26
significant market effects/7 For example, although the prices and quantities of residential housing stocks
are included in the J/W model, changes in prices caused by changes in air pollution-related visibility in
residential areas ansjaot captured. Also, some of the economic benefits of air pollution control are not
fully characterized by the J/W model. For example, since there is no distinct sector representing the
34 pollution, control industry, increases in production, export sales, and employment in this particular
* In the present study, both benefits and costs are driven off of the same macroeconomic projections from the Jorgenson/WUcoxen model,
to ensure that me estimates are based on a consistent set of economic assumptions.
57 See March 24.1993 letter from the SAB CAACAC review panel to the EPA Administrator, po2-3, which states: 'hiprinciple, effects
such as decreases in worker productivity and increases In medical care e^>enditure that woidd occur in ^ no control scenario should affect
Gross Domestic Product (GDP). We do not believe, however, that it is lUcefy to be productive to attempt formalfy to'close the loop'by
incorporating such ejects in the general equilibrium modeling. As we understand the Jorgenson/Wilcoxen (J/W) model, it may be also
importble for it to show bnpacts on the pollution control industry, narrowly
detailed composition of investment goods produced."
58
-------
Appendix A: Cost and Macroeconomic Modeling
industry created by air pollution control programs are not reported by the J/W model.1* The limited
2 capabilities of macroeconomic models are even more acute for non-market effects, some of which may
3 have highly significant consequences for social and economic welfare. For example, the J/W model does
4 not capture pain and suffering caused by air pollution-related respiratory disease, the effects of increased
5 deposition in lakes of airborne heavy metals and subsequent uptake by fish caught by individuals for
6 personal consumption, adverse effects of air pollution on wildlife and sensitive ecosystems, and many
7 other health and environmental effects.
« In general, the value of the investment in using a general equilibrium model in any given analysis
9 must be weighed against alternative investments which could reduce other sources of uncertainty and
10 bias in the overall analysis. Although the project team chose to perform macroeconomic modeling for
n the present study, significant questions regarding the benefits of investing scarce project tone and
12 resources in macroeconomic modeling have led to the conclusion that a sector-level approach will be
13 more appropriate for the first §812 Prospective Study of the costs and benefits of the CAA. This
H conclusion is reinforced by the recommendation of the CAACAC review panel.39
is Choice of MacroGConomlo Model
---"" s~~ - - -~'~
u The adequacy of any model or modeling approach must be judged hi light of the policy questions
n being asked. One goal of the present study is to assess the effects of clean air regulations on
it macroeconomic activity. Two broad categories of macroeconornic models were considered for use hi the
19 assessment: short run, Keynesian models and long-run, general equilibrium models.
20 Recognizing that structural differences exist between the models, one needs to focus in on the
21 particular questions that should be answered with any particular model. The Congressional Budget
22 Office (1990) noted: /.";-"-"!-=.
23 "Both the [Data Resources Incorporated) DRI and the IPCAEO models show relatively limited
24 possibilities for increasing energy efficiency and substituting other goods for energy hi the short
25 run... Both models focus |>nmarily on short-term responses to higher energy prices, and neither is
26 very good at exammmgjjjOWJ^stnKture of the economy could change in response to changing
27 energy prices. The [Jorgenson-Wilcoxen] model completes this part of the picture..."60
21 One strategy for assessing the macroeconomic effects of the CAA would be to use a DRI-type model
29 in conjunction with the Jorgenson-Wilcoxen model to assess both the long-term effects and the short-run
jo transitions, in much the same way that the Congressional Budget Office used these models to assess the
31 effects of carbon taxes. However, because of significant difficulties hi trying to implement the DRI
32 model in a meaningful way, the project team chose to focus on the long-run effects of the CAA.
*/W(4 footnote 57.
" See May 14,1993 letter from SAB CAACAC to the EPA Administrator, p. 6, which states: "... the Aemcv should learn from the
^nw/on/ft ff i
modelinv or literatim review efforts. * (authors' emphasis)
* The Congressional Budget Office report (1990) refers to an older (1981) version of the Jorgenson model, not the current (1988) version.
The approach to long-run dynamics differs between the two models. The newer Jorgenson-Wilcoxen model contains bom the capital
accumulation equation and the capital asset pricing equation. The 1981 version of the model contained only the capital accumulation equation.
59
-------
Appendix A: Cost and Macroeconomic Modeling
Structural changes (e.g., changes in employment in the coal sector due to the CAA) can be identified
with the Jorgenson-Wilcoxen model.
Overview of tho Jorgonson-WIIcoxen Modol
4
S
6
7
I
9
10'.
It
12
13
14
11
It
17
II
19
20
21
22
23
24
21
26
27
U
29'
30
31
32 '
33
34
31
36
37
38
39
40
41
42
The discussion below focuses on those
characteristics of the Jorgenson-Wilcoxen model
that have important implications for its use in the
assessment of environmental regulations (see Table
18). The J/W model is a detailed dynamic general
equilibrium model of the U.S. economy designed for
medium run analysis of regulatory and tax policy
(Jorgenson and Wilcoxen [1990a]). It provides •
projections of key macroeconomic variables, such as
GNP and aggregate consumption, as well as energy
flows between economic sectors. As a result, the ,
model is particularly useful for examining how thep:=-
structure of the economy could change in response
to changes in resource prices. For the purpose of
this study, it has .five key features: a detailed
treatment of production and consumptio^l
parameters estimated econometricalfy form
historical data, an endogenous model of technical
change, a rigorous representation ojf^saving an&
investment, and iree mobility (Kjflablar and capiL* .^v
« -i" m~" s*~ - ^"aJrs :-as-»aJs. ""_«=- •=•-
between industries^
table 18.
.. ••••'J$^^'&$!^
f .ttl)KfffiftftlBfflwTyiPftCff:i'''^?ff roMf"^ftttff"dif
. ^^i^i^^^m^'^m^^
detail and econometric estimation, allow the model to
in time for given levels of technology and the size of the
The first two features, Jndui
capture the effects of jtte^pAA at eaclbPj
economy's capital stocli^ iyd|i|^ed treatment of production and consumption is important because the
principal effects of the Clla&lirf||t fell most heavily on a handful of industries. The J/W model
divides total U.S. pnxlucti0fr!i|pS industries which allows the primary economic effects of the CAA to
be captured. Econometric estimation is equally important because it ensures that the behavior of
households and firms in till model is consistent with the historical record.
'The model's second two features —its representations of technical change and capital
accumulation— complement the model's intratemporal features by providing specific information on
how the Act affected technical change and the accumulation of capital. Many analyses of environmental
regulations overlook or ignore intertemporal effects but these effects can be very important Jorgenson
and Wilcoxen (1990a) suggests that the largest cost of all U.S. environmental regulations together was
that the regulations reduced me rate of capital accumulation.
The model's last feature, free mobility of a single type of capital and a single type of labor, is
important because it limits the model's ability to measure the short run costs of changes in policy. J/W is
a full-employment model that describes the long-run dynamics of transitions from one equilibrium to
another. Capital and labor are both assumed to be freely mobile between sectors (that is, they can be
moved from one industry to anothei1 at zero cost) and to be fully used at all times. Over the medium to
60
-------
Appendix A: Cost and Macroeconomic Modeling
2 .
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
IS
19
20
21
23
24
25
26
27
U
29
30
31
32
33
34
35
36
37
3t
39
40
41
long run, this is a reasonable assumption, but in the
short run it is too optimistic. In particular, the
model will understate the short run costs of a change
in policy because it does not capture unemployment,
underemployment, or the costs of moving capital
from one industry to another. A single rate of return
on capital exists that efficiently allocates the capital
in each period among sectors. Similarly, a single
equilibrium wage rate allocates labor throughout the
economy.
Structure of the Jorgenson-Wilcoxen Model
The J/W model assesses a broad array of
economic effects of environmental regulations.
Direct costs are captured as increased expenditures
on factors of production —capital, labor, energy and
materials— that the various industries must make to
comply with the regulations, as well as additional
out-of-pocket expenditures that consumers must
make. Indirect costs are captured as general
equilibrium effects that occur throughout He
economy as the prices of factors of production , :
change (e.g., energy prices). Also, the rate of ' ;/L
technological change can respond to changes in ihe
prices of factors of production, causing changes m,
productivity (Jprgenson andFraumeni* 1981).
:the
Table 19* Befiaitioas of Industries Wifcia fte J/W
Model. ' -
ladHStty • • , \ -
Mnmtier Description
, and
The model is divided into four
business, household, government, and«iSJt«bf-the-
world sectors. The business sector is farther
subdivided into 35 rndustrie*(see T^ble 19)." Each sector
produces a primary product and some produce
secondary products. These outputs serve as inputs
to the production processes of me other industries,
are used for investment, satisfy final demands by the
household and government sectors, and are
exported. The model also allows for imports from
the rest of the world.
The Business Sector
J
8
9
10
n
12
Metal anab
trofetuu; and 0aiurat KSS
Tobacco
Fttfjoititte
The model of producer behavior allocates the |H|||||M|BB|MHH|||^H|||||||M|HHHBBM|BHMH||||
value of output of each industry among the inputs of
the 35 commodity groups, capital services, labor services, and noncompeting imports. Output supply
and factor demands of each sector are modeled as the results of choices made by wealth maximizing,
" The 35 industries roughly correspond to • two-digit SIC code classification scheme.
61
-------
Appendix A: Cost and Macroeconomic Modeling
i price taking firms which are subject to technological constraints. Firms have perfect foresight of all
2 future prices and interest rates. Production technologies are represented by econometrically estimated
3 cost functions that fully capture factor substitution possibilities and industry-level biased technological
4 change.
i " , -
5 Capital and energy are specified separately hi the factor demand functions of each industry. The
6 ability of the model to estimate the degree of substitutability between factor inputs facilitates the
7 assessment of the effect of environmental regulations. A high degree of substmxtalttiiry between inputs
i implies that the cost of environmental regulation is low, while a low degree of substitDtability implies .
9 high costs of environmental regulation. Also, different types of regulations lead to different responses on
w the part of producers. Some regulations require the use of specific typesof equipment Others
/; regulations restrict the use of particular factor inputs; for example, through restrictions onlhe
n combustion of certain types of fuels. Bom of mese effects can change the rate of productivity growth in
n an industry through changes in factor prices. --;«.;%.
14 ' The Household Sector , ,_ »1?i
15 In the model of consumer behavior, consumer choices between labor and leisure and between
16 consumption and saving are determined. A system of mdlvidi^demographically defined household
n demand functions are also econometrically estimated. Household CQBSvanption is modeled as a three
it stage optimization process. In the first stage households allocate lifeiime wealth to full consumption in
19 current and future time periods to maximize intertemporal utility. Lifetime wealth includes financial
20 wealth, discounted labor income, and Ibl imputed value of leisure. Households have perfect foresight of
21 future prices and interest rates. In tne: second stiige* for eachMme period full consumption is allocated
22 hgtween grxvjs and services qqtf-jepiire to maxfiFlfoff JB^Iieinpnral utility This yields an allocation of
23 household's time endowmeMhet^sen the labor fflar]ce| (giving rise to labor supply and labor income)
24 and leisure time and demandlf|§^^ In me third stage, personal consumption
25 expenditures are laUoqated amoi^£^||||Bbor, noncompeting imports and me outputs of the 3 5
26 production sectoreto maximize a subuljl&fwction for gcKxis consumption. As with the business
Jr . - „-&_ -.-__ - -__ -__ ^--gn-^^^jjjjjfHjjisppr ^ f
27 sector, siihstitiitipn;r*9ffi^ittftif^ exii^iiiajiyg™'ptinn Hecisinns. The model's flexibility enables it to
28 capture the subs6tut«ffi^lA^utuig^p%ducts for polluting ones mat may be induced by
29 environmental regulation^. ^||i§|||di| this end, purchases of energy and capital services by households are
x specified separately withinl|iiMiimer demand functions for individual commodities.
n It is important to be dear regarding the notions of labor supply and demand within the J/W model,
n and what is meant by "employment" throughout this report Labor demands and supplies are represented
33 as quality-adjusted hours denominated in constant dollars. The labor market clears in each period; the
34 quantity of labor sejvices offered by households is absorbed fully by the economy's producing sectors.
15 However, inferences regarding the number of persons employed require information on labor quality and
36 work-hours per person over time and across simulations. Neither of these are explicitly modeled.
37 The Government Sector
3» The behavior of government is constrained by exogenously specified budget deficits. Government
39 tax revenues are determined by exogenously specified tax rates applied to appropriate transactions in the
« business and household sectors. Levels of economic activity in these sectors are endogenously
41 determined. Capital income from government enterprises (determined endogenously), and nontax
62
-------
Appendix A: Cost and Maeroeconomic Modeling
receipts (given exogenously), are added to tax revenues to obtain total government revenues.
2 Government expenditures adjust to satisfy the exogenous budget deficit constraint.
3 The Rest-of-the-World Sector
4 The current account balance is exogenous, limiting the usefulness of the model to assess trade
i competitiveness effects. Imports are treated as imperfect substitutes for similar domestic commodities
6 and compete on price. Export demands are functions of foreign incomes and ratios of commodity prices
7 in U.S. currency to me exchange rate. Import prices, foreign incomes, and tariff policies are
« exogenously specified. Foreign prices of U.S. exports are determined endogenously by domestic prices
9 and the exchange rate. The exchange rate adjusts to satisfy the-exogenous constraint on net exports.
10 Environmental Regulation, Investment, and Capital Formation
11 Environmental regulations have several important effects on capital formation. At the most obvious
12 level, regulations often require investment in specific pieces of pollution abatement equipment. If the
n economy's pool of savings were essentially fixed, the need to invest in abatement equipment would
14 reduce, or crowd out, investment in other kinds of capital on a dollar for dollar basis. On the other hand,
a if the supply of savings were very elastic then abatement investments might not crowd out other
16 investment at all. In the J/W model, both the current account and government budget deficits are fixed
i? exogenously so any change in the supply of funds for domestic investment must come from a change in
a domestic savings. Because households choose consumption, and hence savings, to maximize a lifetime
19 utility function, domestic savings will be somewhat elastic. Tims, abatement investment will crowd out
other investment, although not on a dollar for dollar basis. "
- --IF
- .If ~ "- J - " - " 1 -?--•
21 The J/W assumption that me culrrent account dpes not change as a result of environmental regulation
22 is probably unrealistic, but ftjs not at aU clear that mis biases me crowdm^
23 direction. By itself, ihe need to invest in abatement capital would tend to raise U.S. interest rates and
24 draw in foreign savings. To the extent this occurred, crowding out would be reduced. At the same time,
25 however, regulation reduces the profitability of domestic firms. This effect would tend to lower the
26 return on domestic assets, leading to a reduced supply of foreign savings which would exacerbate
27 crowding out Which efi^tipnaiiiates is an empirical question beyond the scope of this study.
""**„-- -"_J" ~- ~ I*.-"8 - *
a In additional to crowding but ordinary investment, environmental regulation also has a more subtle
29 effect on the rate of capital formation. Regulations raise the prices of intermediate goods used to
jo produce new capital. This leads to a reduction hi the number of capital goods which can be purchased
31 with a given pool of sayings. This is not crowding out hi the usual sense of the term, but it is. an
32 important means bjpwhich regulation reduces capital formation.62
33 The General Equilibrium
34 The J/W framework contains intertemporal and intratemporal models (Jorgenson and Wilcoxen
35 [1990c]). In any particular time period, all markets clear. This market clearing process occurs in
36 response to any changes hi the levels of variables that are specified exogenously to the model. The
0 Wilcoxen (1988) suggests that environmental regulation may actually lead to * "crowding in" phenomenon. Wilcoxen examined the
effects of regulation at the firm level, and introduced costs into the model related to the installation of capital. He found that when firms shut
down their plants to install environmental capital, they take account of the adjustment costs and often concurrently replace other older capital
equipment This effect, however, a not captured in the current version of the Jorgenson-Wilcoxen model.
/
63
-------
Appendix A: Cost and Macroeconomic Modeling
i interactions among sectors determine, for each period, aggregate domestic output, capital accumulation,
2 employment, the composition of output, the allocation of output across different household types, and
3 other variables.
4 ' The model also produces an intertemporal equilibrium path from the initial conditions at the start of
5 the simulation to the stationary state. (A stationary solution for the model is obtained by merging the
« intertemporal and intratemporal models.) The dynamics of the J/W model have two elements: An
7 accumulation equation for capital, and a capital asset pricing equation. Changes in exogenous variables
« cause several adjustments to occur within the model. Fust, the single stock of capital is efficiently
9 allocated among all sectors, including the household sector. Capital is assumed to be perfectly malleable
10 and mobile among sectors, so that the price of capital services in each sector is proportional to asingle
11 capital service price for the economy as a whole. The value of capital services is equal to capital income.
12 The supply of capital available in each period is the result of past investment, i.e., capital at the end of
13 each period is a function of investment during the period and capital at the beginning of the period. This
14 capital accumulation equation is backward-looking and captures the effect of investments in all past
15 periods on the capital available in the current period. . ; ;-
i6 The capital asset pricing equation specifies the price of capital services in terms of the price of
17 investment goods at the beginning and end of each period,, the rate of return to capital for the economy as
a a whole, the rate of depreciation, and variables describing the tax structure for income from capital. The
19 current price of investment goods incorporates an assumption of perfect foresight or rational
20 expectations. Under this assumption, me. pice of investment goods in every period is based on
21 expectations of future capital service prices and discount rates mat are fulfilled by the solution of the
22 model. This equation for the investment goods price in each time period is forward-looking.63
23 One way to characterize the J?f| model -^oiaijpi>ther neoclassical growth model— is that the short-
24 run supply of capital is perfect^ iaeltistic, since ft is completely determined by past investment.
25 However, the supply of capital ispejr&otly elastic in the long run. The capital stock adjusts to the time
26 endowment, while die rate of retunLdefia&pnly on the intertemporal preferences of the household
27 sector.
_
M A predetermined amdiin||i|iidbnical progress also takes place mat serves to lower the cost of
29 sectoral production. Finalj^||||^jg|iility of labor is enhanced, giving rise to higher productivity and
x lowercosts of production. ,J1
-f _f . '
31 Given all of these changes, the model solves for a new price vector and attains a new general
a equilibrium. Across all time periods, the model solves for the time paths of the capital stock, household
33 consumption, and prices. The outcomes represent a general equilibrium in all time periods and in all
34 markets covered By the J/W model.
0 The price of capital assets is also equal to the cost of production, so that changes in the rate of capital accumulation result in an increase
in the cost of producing investment goods. This has to be equilibrated with the discounted value of fiiture rentals in order to produce an
intertemporal equilibrium. The ristag cost of producing investaeirtuawm of adjusting to a new iHter^^
64
-------
Appendix A: Cost and Macroeconomic Modeling
/ Configuration of the No-control Scenario
2 One of the difficulties in describing the no-control scenario is ascertaining how much environmental
3 regulation would have been initiated by state and local governments in the absence of a federal program.
4 It may reasonably be argued that many state and local.governments would have initiated their own
5 control programs in the absence of a federal role. This view is further supported by the fact that many
6 states and localities have, in fact, issued rules and ordinances which are significantly more stringent and
7 encompassing than federal minimum requirements. However, it may also be argued mat the federal .
« CAA has motivated a substantial number of stringent state and local control programs. /
9 Specifying the range and stringency of state and local programs that would have occurred in the
10 absence of the federal CAA would be almost entirely speculative. For example, factors which would
;/ complicate developing assumptions about stringency and scope of unilateral state and local programs
n include: (I) the significance of federal funding to support state and local program development; (ii) the
n influence of more severe air pollution episodes which might be expected In the absence of federalry-
14 mandated controls; (iii) the potential emergence of pollution havens, as well as anti-pollution havens,
a motivated by local political and economic conditions; (iv) the influence of federally-sponsored research
16 on the development of pollution effects information and control technologies; and (v) the need to make
i? specific assumptions about individual state and local control levels for individual pollutants to allow
is estimation of incremental reductions attributable to federal control programs.
•« Another complication associated with the no-control scenario is the treatment of air pollution control
requirements among the major trading partners of the U.S. 'Real-world manifestation of a no-control
21 scenario would imply that public health and environmental goals were not deemed sufficiently
22 compelling by U.S. policymakersv Under these conditions, major trading partners of the U.S. in Japan,
23 Europe, and Canada may welt reach similar policy conclusions. Simply put, if the U.S. saw no need for
24 air pollution controls, there is little nason to assume other developed industrial countries would have
25 either. In this case, some of the estimated (Economic benefits of reducing or eliminating air pollution
26 controls in the U.S. would not materialize[because U.S. manufacturers would not necessarily gain a
27 production cost advantage over foreign competitors. However, like the question of state and local
2s programs in the absence of a federal program, foreign government policies under a no-control scenario
29 would be highly speculative?':-'":>^
_===:"-"
jo Given the severity of these confounding factors, the only analytically feasible assumptions with
31 respect to the no-control scenario are mat (a) no new control programs would have been initiated after
32 1970 by the states or local governments in the absence of a federal role, and (b) environmental policies
33 ofU.S. trading partners remain constant regardless of U.S. policy.
34
35
Elimination of Compliance Costs In the No-Control
36 Industries that are affected by environmental regulations can generally respond in three ways: (I)
37 with process changes (e.g., fluidized bed combustion); (ii) through input substitution (e.g., switching
from high sulfur coal to low sulfur coal); and (iii) end-of-pipe abatement (e.g., the use of electrostatic
65
-------
Appendix A: Cost and Macroeconomic Modeling
i precipitation to reduce the emissions of participates by combustion equipment).64 Clean air regulations
2 have typically led to the latter two responses, especially in the short run. End-of-pipe abatement is
3 usually the method of choice for existing facilities, since modifying existing production processes can be
4 costly. This approach is also encouraged by EPA's setting of standards based on the notion of "best
s available technology "(Freeman, 1978).
6 All three possible responses may lead to: (I) unanticipated losses to equity owners; (ii) changes in
7 current output; and (iii) changes in long-run profitability. If firms Were initially maximizing profits, then
« any of the above three responses will increase its costs. Fixed costs of investment will fee capitalized
9 immediately. This will result in a loss to owners of equity when regulations are introduced. As far as
10 firms are concerned, this is just like a lump sum tax on sunk capital. Such effects will not affect growth
/; or efficiency. However, regulations could also change marginal costs and therefore current: output. In
n addition, they could change profits (i.e., the earnings of capital), and thus affect investment. Both of
ij these effects will reduce the measured output of the economy.
14 On the consumption side, environmental regulations change consumers' expectations of their lifetime
a wealth. In the no-control scenario of mis assessment, lifetime wealth increases. This causes an increase
16 in consumption. In fact, with perfect foresight, consumption rises more in earlier time periods. This also
n results in a change in savings. .^ ;i>v-- ?'.-
it Capital Costs - Stationary Sources _-.- ; . A: V
is To appropriately model investment jn pollution control requires a recognition that the CAA had two
20 different effects on capital markets, Tfrst, CAA regulations led to the retrofitting of existing capital stock
21 hi order to meet environmental standards. In theJnoHxmtibl scenario, these expenditures do not occur.
22 Instead, the resources that were invested in pollution abatement equipment to retrofit existing sources are
23 available to go to other competing^in^estments. Thus, at each point in time, these resources might go to
24 investments m capital jin the regidaled industry, or may go into investments in other industries,
25 depending upon relatives rates of return oa those investments. This will affect the processes of capital
26 formation and deepening^
27 Second, tne CAA pla^^ptions on new sources of emissions. When making investment
2t decisions, firms take into acetpnl^tn-e additional cost of pollution abatement equipment Effectively, the
29 "pricef of investment goodtis higher because more units of capital are required to produce the same
» amount of output In the no-control scenario, there are no restrictions on new sources and hence no
a requirements for pollution control expenditures. Effectively, the "price" of investment goods is lower.
32 Thus, at each point in time, investors are faced with a lower price of investment goods. This results in a
33 different profile ^investment over time.
.".""' -~V--*~
u Operating and Maintenance Costs - Stationary Sources
15 In addition to purchasing pollution abatement equipment, firms incurred costs to run and maintain
36 the pollution abatement equipment In the no-control scenario, resources used to pay for these operating
37 and maintenance (O&M) costs are freed up for other uses. The model assumes that the resources
3i required to run and maintain pollution control equipment are in the same proportions as the factor inputs
39 used in the underlying production technology. For example, if 1 unit of labor and 2 units of materials are
M Regulation may also affect the rate of investment, and change the rate of capital accumulation.
66
-------
Appendix A: Cost and Macroeconomic Modeling
used to produce 1 unit of output, then one-ithird of pollution control O&M costs are allocated to labor and
2 two-thirds are allocated to materials. These adjustments were introduced at the sector level. O&M
3 expenditures are exclusive of depreciation charges and offset by any recovered costs.
4 Capital Costs - Mobile Sources
5 Capital costs associated with pollution control equipment were represented by changing costs for
6 motors vehicles (sector 24) and other transportation equipment (sector 26). Prices (unit-costs) were
7 reduced in proportion to the value of the pollution control devices contained in cars, trucks, motorcycles,
s and aircraft. -
_ A
9 Operating and Maintenance - Mobile Sources
w Prices for refined petroleum products (sector 16) were changed to reflect the resource costs
11 associated with producing unleaded and reduced lead gasoline (fuel pike penalty), the change in fuel
n economy for vehicles equipped with pollution control devices (fuel economy penalty), and the change in
u fuel economy due to the increased fuel density of lower leaded and no lead gasoline (fuel economy
H credit). Third, inspection and maintenance costs and a maintenance credit associated with the use of
a unleaded and lower leaded (i.e., unleaded and lower leaded gasoline is less corrosive, and therefore
16 results in fewer muffler replacements, less spark plug corrosion, and less degradation of engine oil) were
n represented as changes in prices for other services (sector 34). = *, -----
Direct Compliance Expenditures
19
20 Cost data for mis study are derivedprimarily from the 1990 Cost of Clean report EPA publishes
21 cost data in response to vejuitements of the Clean Air and Clean Water Acts. The following subsections
22 describe Cost of Clean dafa m dftail, as well as adjustments made to the data and data from other
23 sources. ". .ii^r/""
24 Cost of Clean Data -./
25 EPA is required to compile and publish public and private costs resulting from enactment of the
26 Clean Air Act and the Clean Water Act The 1990 Cost of Clean report presents estimates of historical
27 pollution control expenditures for the years 1972 through 1988 and projected future costs for the years
2s 1989 through 2000. This includes federal, state, and local governments as well as the private sector.
29 Estimates of capital costs, operation and maintenance (O&M) costs, and total annualized costs for five
x categories of environmental media, including air, water, land, chemical, and multi-media, are presented.
31 It should be noted that these estimates represent direct regulatory implementation and compliance costs
32 rather than social costs. The Cost of Clean relied on data from two governmental sources, the EPA and
33 the U.S. Department of Commerce (Commerce).
67
-------
Appendix A: Cost and Macroeconomic Modeling
/ EPA Data
2 EPA expenditures were estimated from EPA budget justification documents.63 Estimates of capital
3 and operating costs resulting from new and forthcoming regulations were derived from EPA's Regulatory
4 Impact Analyses (RIAs). RIAs have been prepared prior to the issuance of all major regulations since
5 1981. Finally, special analyses conducted by EPA program offices or contractors were used when other
& data sources did not provide adequate or reliable data.
7 Commerce Data
» Data collected by Commerce were used extensively in the Cost of Clean for estimates of historical
9 pollution control expenditures made by government agencies other than EPA and by the private sector.
10 Two Commerce agencies, the Bureau of Economic Analysis (BEA) and the Bureau of the Census
u (Census), have collected capital and operating costs for compliance wim environmental regulations since
12 the early 1970's. Commerce is, in fact, the primary source of original survey data for environmental
13 regulation compliance costs. Commerce publishes a number of documents that report responses to
H surveys and comprise most of the current domain of known pollution abatement and control costs in the
is United States, including: - .
16 > A series of articles entitled "Pollution Abatement and Control Expenditures'1 published annually in
17 the Survey of Current Business by BEA (BEA articles); -^^
». ri.1ir~
u > A series of documents entitled "Pollution Abatement Costs and Expenditures" published annually in
19 faeCitrrent Industrial Reports-^Cewv&(]^^
20 * A series of documents entitled^
21 Finances). 4 ^;"- \l|i
-- *.--_-- =,-. - ^-SA"t-£-^Sf.
22 BEA articles contain data derived from a number of sources, including two key agency surveys —the
23 "Pollution Abatement Costs and Expenditures Survey" (PACE Survey) and the "Pollution Abatement
24 Plant and Equipment Su«|5p (|,,APE Survey)—- which are conducted annually by Census for BEA. Data
25 have been reported for 19p|lB5QUg| 1988.66 .
26 PACE reports have been published annually since 1973 with the exception of 1987. Figures for 1987
27 were estimated on the basis of historical shares within total manufacturing. These reports contain
21 expenditure estimates derived from surveys of about 20,000 manufacturing establishments. Pollution
» abatement expenditures for air, water and solid waste are reported by state and Standard Industrial Code
30 (SIC) at the four-dipt level. According to Census, surveys conducted since 1976 have not included
31 establishments with fewer than 20 employees because early surveys showed that they contributed only
3i .about 2 percent to the pollution estimates while constituting more than 10 percent of the sample size.
*
33 Each year Census conducts a survey of state, local, and county governments; and survey results are
34 published in Government Finances. Census asks government units to report revenue and expenditures,
3i including expenditures for pollution control and abatement.
a The main source of data for EPA expenditures is the Justification ofApprcpriatomEainK^JbrCoiHmttteonApproprtations.
M The most recent BEA article used in the Cost of Clean was "Pollution Abatement and Control Expenditures, 1985-88" in Survey of
Current Business, November 1990.
.68
-------
Appendix A: Cost and Macroeconomic Modeling
Non-EPA Federal expenditures were estimated from surveys completed by federal agencies detailing
their pollution control expenditures, which are submitted to BEA. Private sector air pollution control
expenditures, as well as state and local government air pollution expenditures, were taken from BEA
articles. .
s
9
10
11
12
13
14
a
16
17
It
20
21
22
23
24
25
26
27
2S
29
30
31
32
33
34
31
Stationary Source Cost
Capital Expenditures Data
Capital expenditures for stationary air
pollution control are made by factories and electric
utilities for plant and equipment that abate
pollutants through end-of-line (EOL) techniques or
that reduce or eliminate the generation of pollutants
through changes in production processes (CIPP).
For the purposes of this report EOL and CIPP
expenditures are aggregated.67 Table 20
summarizes capital expenditures for stationary air
pollution control, categorized as "nonfann j
business" or "government enterprise" expenditures.
Nonfarm business capital expenditures consist
of plant and equipment expenditures made by 1}
manufacturing companies and 2) privately and '/* k _
cooperatively owned electricinilMesi arid 3) other
nonmanufacturing companies. -^^isp^
"Government enterprise" is, acconiiag to EEA, an
agency of the government ifbose operating costs, to
a substantial extent,^ arei«p|||i|;jby the sale of
goods and services. Here^^g0v%$nffient enterprise
means specifically goveranleitliiterprise electric
utilities. Government enterprise capital
expenditures are pollution abatement expenditures
made by publicly owned electric utilities.6'
Operation and Maintenance Expenditures Data
Stationary source O&M expenditures art made
by manufacturing establishments, private and public
electric utilities, and other nonmanufacturing
Table 20, Estimated Capi^aM QAM
Noofttnai
1973
197*
1975
1976
2,968 Ir407
1978
1999
1981
3,798,
3J977
,$m
104
102
156
197
205
1983
1984
1985
198$
5,085
4,155
451
509
6,690
4,141
416
328
7,469
om
,\»
45
58
60
72
106
. 148
135
141
, 143
147
189
< 140
130
|&>B
4,267
4,760
243
235
in
" Survey respondents to die Census annual Pollution Abatement Surveys report the difference between expenditures fat CIPP and what
they would have spent for comparable plant and equipment without poUution abatement features. Disaggregated capital expenditures by
private manufacturing establishments can be found in annual issues of Census reports.
<* BEA calculates these expenditures using numbers obtained from Energy Information Agency (EIA) Form 767 on steam-electric plant
air quality control.
69
-------
Appendix A: Cost and Macroeconomic Modeling
businesses to operate air pollution abatement equipment O&M expenditures for electric utilities are
made up of two parts: 1) expenditures for operating air pollution equipment and 2) the additional
expenditures associated with switching to alternative fuels that have lower sulfur content (fuel
differential). Expenditures to operate air pollution abatement equipment are for the collection and
disposal of flyash, bottom ash, sulfur and sulfur products, and other products from flue gills.69 O&M
expenditures are net of depreciation and payments to governmental units, and are summarized in Table
20. O&M data were disaggregated to the two digit SIC level for use in the macroeconomic model.
Recovered Costs " . -- „ - "
9
10
II
12
13
14
IS
If
17
II
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
3S
"Recovered costs" are costs recovered (i.e.,
revenues realized) by private manufacturing
establishments through abatement activities.
According to instructions provided to survey
participants by Census, recovered costs consist of 1)
the value of materials or energy reclaimed through
abatement activities that were reused in production
and 2) revenue that was obtained from the sale of
materials or energy reclaimed through abatement
activities. Estimates of recovered costs were obtained
from the PACE reports and are summarized in Table 4
21. In this analysis, recovered costs were removed
from total stationary source air pollutifD icbntrol:£f
O&M costs — that is, net O&M cost II any year
would be O&M expenditures (see Table:
recovered costs. Recovered cost data were
disaggregated tome two dig&S!(| Jmpi for use in the
macroeconomic'modeL. ~ -- -'^Vr^'—.
Mobile Source
Costs of controlling poIlotioO^missions from
motor vehicles were estimated by calculating the
purchase price and O&M cost premiums associated
wjth vehicles equipped with pollution abatement
controls over the costs for vehicles not equipped with
such controls. Tnew costs were derived using EPA
analyses, including EPA RIAs, the Cost of Clean, and
other EPA reports.70
Xfitt
PACE*
w*
m
389
ISfi®
1 '#*
1,000
822
HWF
870
768
* Ait ce«j
I(t FACE
ri^fe^
" Faiber, Kit D. and Gaiy L. Rutledge, *PoUutkm Abatement and Control Exponlitures: Methods and Sources for Cunwrt-DoUar
Estimates," Unpublished paper, Bureau of Economic Analysis, U.S. Department of Commerce, October 1989.
70 A complete listing of sources used in calculating mobile source capital and operating expenditures can be found in Environmental
Investments: Tht Cost of a Clean Environment, keport of the Administrator of the Environmental Protection Agency to the Congress of the
United State, EPA-230-11-90-083, November 1990. ,
70
-------
Appendix A: Cost and Macroeconomic Modeling
2
3
• 4
1
6
7
8
9
70
11
12
13
14
IS
16
17
IS
19
20
22
23
&
23
26
27
28
29
30
31
32
33
34
31
36
37
3»
39
40
41
T*bfe22. Estimated C*j>ftat attd
Source AfrPoUatiaa Control {mfllM» of
Caaftai
1973
1975
'**$?
1978
1,961
2^48
2,282
zm
1,786
908
1980
1981
1985
-155
-326
Capital Expenditures Data
Capital expenditures for mobile source emission
control are associated primarily with pollution abatement
equipment on passenger cars, which comprise the bulk of
all mobile sources of pollution. These capital costs
reflect increasingly stringent regulatory requirements
and improvements in pollution control technologies over
time. Each of the following devices have been used at
one time or another dating back to the Clean Air Act
Amendments of 1965: air pumps, exhaust-gas
recirculation valves, high altitude controls, evaporative
emissions controls, and catalysts. The cost estimates for
each component were computed on a per-vehicle basis
by engineering cost analyses commissioned by EPA.
The resulting per-vehicle capital costs were multiplied
by vehicle production estimates to determine annual _: -
capital costs. Table 22 summarizes mobile source ; --.•
capital costs. j:--
Operation and Maintenance Expenditures Data >
Costs for operation and maintenanjl of emission
abatement devices include the costsoJf mamtainffig
—- -.J- -u.=Tlsfrj: jj
pollution control equipment plus th? cost of ve|^|etr,A:^
inspection/maintenance programsj^OpeiatingCHOStl per
vehicle were multiplied by total vehl^es in use to B "
determine annual cost* Mobile source <1&M costs are
, - 3,- - -™isiHB^Js£S«ST!«_.
made up of three factofs: 1) fuel pria^np^, 2) fuel
economy penalty, and 3) ip^ection aitf ipintenance
program costs as described Aijjfpr. These costs are
mitigated by cost savings intfoel|bfm of maintenance
economy and fuel density tagftjjjj? Table 23
summarizes mobile source Q&M expenditures and cost
savings by categories, with net O&M costs summarized ^^^^^^^^^^^^^^^^^^^^^^
above in Table 22. The following sections describe the
components of the mobile source O&M cost estimates.
Fuel Price Penalty
Historically, the price of unleaded fuel has been several cents per gallon higher than the price of
leaded fuel. CAA costs were calculated as the difference between the cost of making unleaded gasoline
and leaded gasoline with lower lead levels and the cost of making only leaded gasoline with a lead
content set at pre-regulatory levels. These cost estimates were developed using a linear programming
model of the refinery industry. Prices of crude oil and other unfinished oils, along with the prices of
refinery outputs, were adjusted annually according to price indices for imported crude oil over the period
of analysis. The relative shares of leaded and unleaded gasoline and the average lead content in leaded
1986
1987
-1,636
71
-------
Appendix A: Cost and Macroeconomic Modeling
i
2
3
4
3
6
7
S
9
10
11
12
13
14
IS
16
17
II
19
20
21
22
23
24
23
26
21
21
29
30
31
32
33
34
33
36
gasoline also were adjusted annually according to the
historical record.
These estimates may tend to understate costs due
to a number of biases inherent in the analysis process.
For example, the refinery model was allowed to
optimize process capacities in each year. This
procedure is likely to understate costs because
regulatory requirements and market developments
cannot be perfectly anticipated over time. This
procedure resulted in estimates that are about ten
percent less than estimates in other EPA reports.71
However, new process technologies that were
developed in the mid-1980s were not reflected in
either die base case or regulatory case runs. It is
reasonable to expect that regulatory requirements
would have encouraged development of technologies
at a faster rate than would have occurred otherwise.
Fuel Economy Penalty
The fuel economy penalty benefit is the cost
associated with the increased/decreased amount of
fuel used by automobiles with air pollution control
devices (all el
catalytic nveiters, decrease iu^ap|bile fuel
efficiency.71 If this Istsumption isl^
control devices ineM^the total f^jjc^:||^
consumers. An allranH^ll^sumptiomf"^"1
of catalytic converters J
This increase has been aUi|b^j^|iyarge measure to
-way
Table 23. OAW Costs and Credits (millions of
V
fttee
Penalty TWnrty
244
-26
•#
-289
im
1978
**»
198(1
1981
1983
1983
1984
1985
2305
*»'
- 2106
* 568 ', 1956
m * Wfr 4527
1187 1868 -1826
1912 1998 -2120
tttt ^im: w
2M **"9O* -2542
Caste
1765
2351
2060
1786
908
1229
1790
1389
555
3922.
'**?
-265J
im
-4126
1990
-4794
-5089
' -326
'* 337
-1394
-1302
4575
, 4636
-1816
jSfflttw<3)^*<
. Office ^I
assumption, the decrease in total fuel cost to consumers is considered a
the feedback mechanism
catalytic converters.73 Undjr
benefit of the program. f
For the purposes of this study, sensitivity analyses were performed using data presented in the Cost
of Clean report These analyses were conducted to evaluate the significance of assumptions about the
relationship between mile per gallon (MPG) values for controlled automobiles and MPG values for
uncontrolled cars. Based on results of these and other analyses, fuel economy was assumed to be equal
71 Costs and Benefits of Reducing Lead in Gasoline: Final Regulatory Impact Analysis, U.S. Environmental Protection Agency, Office of
Policy Analysis, EPA-230-05-85-006, February 1985.
71 Memo fiom Joel Schwartz (EPA/OPPE) to Joe Somers and Jim DeMocker dated DecenAer 12,1991, and entitled "Fuel Economy
Benefits.* Schwartz states that since mis awlysisu relative to mi» Clean Air Act baseliM,iwt a
relevant In the absence of regulation, tuning of engines for maximum economy would presumably be optimal in the base case as well.
71 Memo fiom Joseph H. Somers, EPA Office of Mobile Sources, to Anne Grambsch (EPA/OPPE) and Joel Schwartz (EPA/OPPE)
entitled "Fuel Economy Penalties for Section 812 Report," December 23,1991.
72
-------
Appendix A: Cost and Macroeconomic Modeling
. for controlled and uncontrolled vehicles from 1976 onward. This may bias the cost estimates although in
2 an unknown direction.
3 Inspection and Maintenance Programs
4 Inspection and maintenance programs are administered by a number of states. Although these
5 programs are required by the Clean Air Act, the details of administration were left to the discretion of
6 state or local officials. The primary purpose of inspection and maintenance programs is to identify cars
7 that require maintenance —including cars that 1) have had poor maintenance, 2) have been deliberately
* tampered with or had pollution control devices removed, or 3) have used leaded gasoline when unleaded
9 is required— and force the owners of those cars to make necessary repairs or adjustments.7*
10 Expenditures for inspection and maintenance were taken from the Cost of Clean.
"""_'=.
// Beneficial effects of the mobile source control program associated with maintenance and fuel density
n were also identified. These cost savings were included hi this study as credits to be attributed to the
n mobile source control program. Credits were estimated based on an EPA study,7? where more detailed
i4 explanations may be found. ;_.--. ;.i""
15
27
Maintenance Credits
16 Catalytic converters require the use of unleaded fuel, which is less corrosive than leaded gasoline.
i? On the basis of fleet trials, the use of unleaded or lower leaded gasoline results in fewer muffler
is replacements, less spark plug corrosion, and less degradation of engine oil, thus reducing maintenance
costs. Maintenance credits account for the majority of the direct (non-health) economic benefits of
jo reducing the lead concentration in gasoline. '"••'"'°~
""-"_." ^-i" 5-
22 The process of refining unleaded gasoffiie. increases its density. The result is a gasoline that has
23 higher energy content Furthermore, unleaded gasoline generates more deposits hi engine combustion
24 chambers, resulting in slightly increased compression and engine efficiency. Higher energy content of
25 unleaded gasoline and rncxeas^ ejogjne efficiency from the used of unleaded gasoline yield greater fuel
26 economy and therefore savings ia refining, distribution, and retailing costs.
Other Direct Cos* Data
2» The Ctof0/C2«« report includes several other categories of cost that are not easily classified as
29 either stationary source or mobile source expenditures. Federal and state governments incur air
M pollution abatement costs; additionally, federal and state governments incur costs to develop and
si enforce CAA regulations. Research and development expenditures by the federal government, state
32 and local governments, and (especially) the private sector can be attributed to the CAA. These data are
33 summarized by year hi Table 24.
74 Walsh, Michael P., 'Motor Vehicles and Fuels: The Problem,' EPA Journal, Vol. 17, No. 1, January/February 1991, p. 12.
15 Schwartz, ).,etaL Costs and Benefits of Reducing Lead In Gasoline: Final Regulatory Impact Analysis, U.S. Environmental
Protection Agency, Economic Analysis Division, Office of Policy Analysis, February 1985.
73
-------
Appendix A: Cost and Macroeconomic Modeling
i
2
3
4
i
6
7
8
9
10
11
12
13
14
IS
16
17
II
Unlike the other
expenditure data used for
this analysis, the survey
data used as a source for
private sector R&D
expenditures cannot be
disaggregated into industry-
specifie expenditure totals.
Consequently, private sector
R&D expenditures are
omitted from the
macroeconomic modeling
exercise (the macro model
is industry-specific). The
R&D expenditures are,
however, included hi
aggregate cost totals used in
the benefit-cost analysis.
Table 24,
1976
1978
19>?9
1980
1981
1982
im
1984
1985
1986
.^frfrtenffiflfl
Federal State
47 0
S6 d
105; i
Iflf 'I
9ft 0
103 0,
,^B ' &
85 0
$7 0
136 4
115 14
98 12
J» 35
Regalatfons
^^l^ffi^ffff)^ -
Stated
'Federal Local
50 115v
:. 5J> 131
, «S ,139
', «r m
-' '«? »t^
93 181
100 200
m 207
93 2§ft
88 239
101 250
State &
Private Federal Local
451 126.
492 100
466 108
654 144
789 146
924 105
869 130
852 131
1,315
1,359 165
347
1988
100
110, 300
120 """"'"
198* - ,#,
m,.
1,574 209
^ Tttfi ^ ""^SB-
£,/j5 ..*sy
1,820 231
6
7 S38
9 ffMrO*
6 990
7 1,153
'§ 1,309
5 1,'428
$ 1,921
4 2,008
^ 2,140
4 2,214
2 2,281
1 2,388
2 2,522
2 2,613
"-- .= *
19
20
21
22
23
24
21
26
27
2S
29.
30
31
Uncertalntlos la the Direct Cost Data
~ ~*""—_
Potential Sources of Error
Because of the importance of me Cos/ of Clean data for this assessment, the project team
investigated potential sources of error due to the use of industry's self-reported costs of compliance with
air pollution abatement requirements. Concerns about the accuracy of responses include (1) misreporting
by firms in response tp federal agency surveys, and (2) omission of important categories of compliance
cost from the datajspilected or reported by these federal agencies.76 Table 25 contains a summary of the
results of the analysis. .
The major finding from this analysis indicates that total O&M costs are likely to be under-reported
due to exclusion of private research and development (R&D) expenditures. Note, however, that although
these costs were excluded from those used for the macroeconomic modeling, they were included in the
overall direct cost estimate of the CAA; see "Other Direct Costs," above. Both BEA and Census instruct
respondents to exclude research and development costs from their estimates of pollution abatement
76 Memorandum from Industrial Economics, Incorporated to Jim DeMocker (EPA/OAR) dated 10/16/91 and entitled "Sources of Error in
Reported Costs of Compliance with Air Pollution Abatement Requirements."
74
-------
Appendix A: Cost and Macroeconomic Modeling
Table 25. V&e^SfXst<&<£Tfatt^
SourceofBrror
Effect on Capital Costs
Effect on O&M Costs
Lack of Data^tf Firm level
ftetcoa Unkaowa
GveMBpotted
W® Bjqwwc*
E^dusion of Rjxxrrered Costs /\
to
1T5S
1 to25t
MHgp&te&ty'1 ft* 2$
in
Com of C^
, ,. ..
wife Air Polhflion Abatement RcquimoaJte/ October 16, Wil,
exp«iditures. Private R&D averages about 18 percent of all private, non-capital pollution control outlays
over the period 1972-89. The range is from 12 to 25 percent with the vast majority (12 of 18) of the
annual observations clustered in the 17 to 18 percent range. For example, it has been estimated that air
pollution abatement R&D spending by private parties in 1987 totaled $1.574 billion, or 17 percent of all
current account air pollution abatement spending by private business.77 These costs are excluded from
the macromodeling because they cannot be disaggregated by industry and, more importantly, because
there is no information on what was purchased or obtained as a result of these expenditures.
71 Bratton, David M. and Gary L. Rutledgp. "Pollution Abatement and Control Expenditures, 1982-1988," Survey of Current Business,
November 1990. •
75
-------
Appendix A: Cost and Macroeconomic Modeling
6
7
8
9
10
11
12
13
14
IS
16
17
It
19
20
21
22
23
24
25
26
27
2S
29
30
31
32
33
34
35
36
37
31
39
40
41
42
43
44
It should be noted that based on die need indicated by the ffic review, modifications to die BEA data
were made to remedy some of the biases noted above. In particular, recovered costs for stationary source
air pollution, e.g. sulfur removed using scrubbers that is then sold in the chemical market, have been
accounted for in the data set used in the model runs.
Mobile Source Costs
For the Section 812 analysis, EPA used the
best available information on the estimated cost
of mobile source air pollution control. Several
other sources of cost estimates exist, however,
including a cost series produced by the
Department of Commerce Bureau of Economic
Analysis (BEA), which is summarized in Table
26. The BEA estimates differ significantly from
EPA estimates, particularly with respect to
estimates of capital costs and the "fuel price
penalty" associated with the use of unleaded
gasoline. .
EPA's capital cost estimates are based on
estimates of the cost of equipment required by
mobile source regulations. BEA's estimates are
based on survey data from the Bureau of Labor
Statistics (BLS) that measures&e increase
per.
economy changes for that model feji|
difference in approach ||£ignific
annual capital cost estate* exceed falWjs oy a
factor «tf flYMigfttyy tam* 'ffijfijljppftay underestimate
costs to the extent mat en^lps||||cpst estimates
of components exclude desipiiatt development
costs for those components|TTfie BLS estimates
add the incremental annual costs to all past costs
to derive total current-year costs. Such an
approach overestimate*costs to the extent mat it
falls to account for cost savings due to changes in
component mixes over time.
TabfeSS. BEABsfiaiateAof Mobife^aitceC^tt.
y«ar
1975
1974
1975
3976
1977
3978<
1***
1418
2,131
2,802
3371
1,380
4^34
- 5
97
309
> 763
1,209
931, 1,636
3990
1981
3m
1983
3984
1985
7,663
9,52$
14.368
13,725;,
1988
1989
1990
15340
14,321
2,996
^flO ^'^EiEft
TV«F yuf'xn ^
274 4,235
tii - *,42£ •
165 4,995
4,522
3,672
3,73«-
im
697
1,180
3^44
1,363
3,408.
1,397
3,79J
2,320
2,252
1,31®
3,133,
658
183
itadm
Source:
Some mobile source pollution control devices required the use of unleaded fuel. Unleaded gasoline is
more costly to produce than is leaded gasoline, and generally has a greater retail price, thus imposing a
cost on consumers. EPA estimated the "fuel price penalty" by using a petroleum refinery cost model to
determine the expected difference in production cost between leaded and unleaded gasoline. BEA's
"fuel price penalty" was the difference between the retail price of unleaded gasoline and that of leaded
gasoline. For the 1970 to mid 1980s period, the BEA's estimated fuel price penalty exceeded EPA's by a
factor of three to four. In the late 1980s, the two series converge, and mere is no substantial difference
between the fuel price penalties estimated by the two methods.
76
-------
Appendix A: Cost and Macroeconomic Modeling
3
4
3
6
7
t
9
10
11
12
13
14
11
16
!7
19
20
21
22
23
14
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Direct Expenditures and Costs
Compliance with the CAA imposed direct costs on businesses, consumers, and governmental units,
and triggered other expenditures such as governmental regulation and monitoring costs and expenditures
for research and development by both t . .
government and industry. As shown in Table r" ' -
27, annual direct CAA compliance
Table 27.
GDP
O&Mtad , Prk«
Year Capital "Other" Total
1973 < 3325 , 3*797 .- 7>t23^
lf?4< W&* ^JW«t' '9#a£ 44,9
1975 5,586 5,009 10^595 49J
Total
19,524
21,334
1977
1978
6,695
5324 12,019 60.3
24,099
24,019
24,821
24,443
6363 16,108 83.8
expenditures —including R&D, etc.- over the
period from 1973 to 1990 were remarkably
stable, ranging from $20 billion to $25 billion
in inflation-adjusted 1990 dollars. This is
equal to approximately V& of one percent of
total-domestic output during that period, with
the percentage falling from '/z of one percent
of total output hi 1973 to Vi of one percent in
1990.
f-
Although useful for many purposes, a
summary of direct annual expenditures is not
the best cost measure to use when comparing
costs to benefits. Capital expenditures are
investments, generating a stream of benefits
(and opportunity cost^ over me lifetime
investment The appropriate accounting 7V >
technique to use for capital expenditures in a
cost/benefit analysis i&iooariaalize the ;r
expenditure- i.e., spread dw cost oyer the
useful ||fe of the mvestmen^iapplymg a
discount rate to account fbfsihe time value of
money. _-;-'
For, this cost/benefit analysis, all capital
expenditures have been annualized at a 3
percent (real) rate of interest (and, for
illustrative purposes, a 7 percent rate).
Therefore, "annualized" costs reported for , , - - -
any given year are equal to O&M "••"i••••^^g^^^^^uHmmiiimmmmmimmmmmim
expenditures (plus R&D, etc., expenditures)
plus amortized capital costs (i.e., depreciation plus interest costs associated with the capital stock) for
that year. Stationary source air pollution control capital costs are amortized over 20 years; mobile source
air pollution control costs are amortized over 10 years. Table 28 summarizes annual expenditures and
costs annualized at both 3 percent and 7 percent.
1989 *
1981
1982, 9445,
1983 8,908
1984 10,377
1985 10,856
1986 11,288
1987 < 11,30?
1988 1'1,716
1989 12,048
1990 11,707
18,026 91.0
20,958
22,424
23,684
21,608
74*1' NMW
18,896 , 103.9
8383 20,090 113.2
average expendftHre, 19734990^1990) *
20,587
20,583
20,090
22,446
77
-------
Appendix A: Cost and Macroeconomic Modeling
i
2
3
4
1.
6
7
S
9
10
11
12
13
14
15
16
17
IS
19
20
21
22
23
24
23
27
21
29
30
31
32
33
34
35
36
37
3t
39
40
Due to data limitations, the cost analysis for-this
CAA retrospective starts in 1973, missing costs incurred
in 1970-72. Cost of Clean (1990), however, which is is
the source for annualized capital costs, includes capital
expenditures for 1972. Therefore, amortized costs
arising from 1972 capital investments are included in the
1973-1990 annualized costs, even though 1972 costs are
not otherwise included in the analysis. Conversely, only
a portion of the (e.g.) 1990 capital expenditures are
reflected in the 1990 annualized costs — the remainder of
the costs are spread through the following two decades,
which fall outside of the scope of this study (similarly,
benefits arising from emission reductions in, e.g., 1995
caused by 1990 capital investments are not captured by
the benefits analysis).
Indirect Effects of tit* CAA
In addition to imposing direct compliance costs on
the economy, the CAA induced indirect economic 'j:
effects, primarily by changing the size anj£compositton
of consumption and investment flows^ Although this
analysis does not add these indirect jgtiects to t
costs and include mem in the jQ§aiparison 1
they flic ttr^ portsjiiT to i^oTCji" "'fSiflrffiftC'lliOFi ffliy^fl^jfflrtrffiMrsfts
most important effects of melMA|iisi estimated by the
"--—J-s---^
Tabfe 28. Aiittifti&feil Costs aaft Apia*! * -
jqpemfiiares,, 1973-1990 01999
8). '
J/W macroecononuc simulation.
GNP and Penonal ConniBiption
—™ •«_- « ^B"Jt
Under me no-control:|op
control case (see Table 2!
id Costs
JSSBC IfojiMML at 3% at ?%-
1973 19,524 11,462 11,866
$74 2133* 13,734 14,373
1976 24,099
1977 &M9
*m awe*
im
l*m 16,154
IT.93?
3&£%&
^,7«i
1981
1982
1983
1984 22,424
IM»^
20,146
21,426
23,469
22,670
23^97
1Z5497
2?,SQ0
26.349
1988 20^87 23,092 127,863
1990 , 20,090 25,094 30,305
level of GNP increases by one percent in 1990 relative to the
[the period 1973-1990, the percent change in real GNP rises
monotonically from 026 percent to 1.0 percent The increase in the level of GNP is attributable to a
rapid accumulation of capflal, which is driven by changes in the price of investment goods. The capital
accumulation effect is augmented by a decline in energy prices relative to the base case. Lower energy
prices that correspond to a world with no CAA regulations decreases costs and increases real household
income, mus increasing consumption.
---,": ^jt?"~ . ' .
Removing the pollution control component of new capital is equivalent to lowering the marginal
price of investment goods. Combining mis with the windfall gain of not having to bring existing capital
into compliance leads to an initial surge in the economy's rate of return, raising the level of real
investment The investment effects are summarized in Figure 24, which is found at the end of this
appendix. More rapid (ordinary) capital accumulation leads to a decline in the rental price of capital
services which, in turn, stimulates the demand for capital services by producers and consumers. The
capital rental price reductions also serve to lower the prices of goods and services and, so, the overall
78
-------
Appendix A: Cost and Macroeconomic Modeling
2
3
4
5
6
, 7
S
9
10
11
12
13
14
IS
16
11
It
19
20
21
24
23
26
27
21
29
30
31
32
33
34
33
36
37
3S
39
T*bJe2*. Differences
1974
197$
-0.09
4JD
0,27
0.44
1978
IW
1980
0.54
0.56
0.63
1982
1983
-0.14
-0*19
0.73
0,74
0,98
XL12
0.95
0,98
price level. Obviously, the more capital intensive sectors
exhibit larger price reductions.7* The price effects from
investment changes are compounded by the cost
reductions associated with releasing resources from the
operation and maintenance of pollution control
equipment and by the elimination of higher prices due to
regulations on mobile sources.
To households, no-control scenario conditions are
manifest as an increase in permanent future real earnings
which supports an increase in real consumption in all
periods and, generally, an increase in the demand for
leisure (see Table 30). Households marginally reduce
their offer of labor services as the income effects of
higher real earnings dominate the substitution effects of
lower goods prices. The increase in consumption is
dampened by an increase in the rate of return that
produces greater investment (and personal savings). , ,;;
Finally, technical change is a very important aspect
of the supply-side adjustments under the no-control a:t
scenario. Lower factor prices increase the endogenous
rates of technical change in those industries that are
factor-using. Lower rental prices for capital benefit the
capital-using sectors, lower materials prices benefit the -f
materials-using sectors, and lower energy prices benefit
the energy-usmg sectore. (^baknce, a significant ^^^^^^^
portion of the increase in economic/ growth is attributable
to accelerated productivity growth. Under tte no-control scenario, economic growth averages 0:05
percentage points higher over the interval 1973- 1990. The increased availability of capital accounts for
60 percent of this increase while faster productivity growth accounts for the remaining 40 percent. Thus,
the principal effect arismgfrom=the Boosts associated with CAA initiatives is to slow the economy's rates
of capital accumulation and p>ON|uctivity growth.
Prices
A principal direct consequence of the Clean Air Act is that it changes prices. The largest price
reductions accrue to the most heavily regulated industries which are the large energy producers and
consumers (see Table 31). But these are also the most capital intensive sectors and it is the investment
effects that are the dominant influences in altering the course of the economy. Focusing on energy
prices, under the no-control scenario the price of coal in 1990 declines by 1.3 percent, refined petroleum
declines by 3.03 percent, electricity from electric utilities declines by 2.75 percent, and the price of
natural gas from gas utilities declines by 1.2 percent The declining price of fossil fuels induces
mt
1990
1.09
0.99
1.00
71 Not surprisingly, at the industry level, the principal beneficiaries in the long ran of eliminating the costs associated with air pollution
abatement are the most heavily regulated industries. The largest changes in industry prices and outputs occur in the motor vehicles industry.
Other industries mat benefit significantly from the elimination of environmental controls are refined petroleum products, electric utilities, and
other transportation equipment Turning to mani'iftcturing industries, metal mining and the primary metals have the largest gains in output
from elimination of air pollution controls. ,
79
-------
Appendix A: Cost and Macroeconomic Modeling
10
11
12
13
14
a
If
17
II
19
20
21
n
23
24
21
26
27
21
29
30
31
32
33
34
35
36
37
a
39
40
41
substitution toward fossil fuel energy sources and toward
energy in general. Total Btu consumption also increases.
Sectoral Effects: Changes in Prices and Output by
Industry,
At the commodity level, the effect of the CAA varies
considerably. Figure 25, found at the end of this appendix,
shows the changes in the supply price of the 35 commodities
measured as changes between the no-control case and the ^
control-case for 1990. -
In 1990, the largest change occurs in the price of motor
vehicles (commodity 24), which declines by 3.8 percent in the
no-control case. Other prices showing significant effects are
those for refined petroleum products (commodity 16) which
declines by 3.0 percent, and electricity (commodity 30) which
declines 2.7 percent Eight of the remaining industries have : ?
decreases in prices of 1.0 to 1.4 percent under the no-contral?V-
scenario. The rest are largely unaffected by environmental ''"'"
regulations, exhibiting price decreases between 0.3 apd 0.8
percent
To assess the intertemporal
consider the model's dynamic
prices between If 75 and 1
effect is on me price of output!
(sector 16), which decpnes by
price of petroleum
scenario levels. Iti contra
levels in
TaWe30. Dififawceia Personal
RealSt
-0.02
im
0,24
0.39
0.54
1980
-S43
1982
0,74
0,81
0.85
jm~v
m&',
"..&&<' ^.i&
0,88
0.94
IS89
1.04
1.01
x>leum refining
But by 1990, the
below control
The price changes
vehicles (sector 24) is about 2.4 percent below baseline
percent below baseline levels hi 1990.
bdrty demands, which in turn determine how industry outputs are
affected. Figure 26, found at UteTend of mis appendix, shows percentage changes in quantities produced
by the 35 industries for thei years 1975,1980,1985 and 1990. As noted earlier, the principal
beneficiaries under the no-control scenario are the most heavily regulated industries: motor vehicles,
petroleum refining, and electric utilities.
In 1990, Jthcjaotor vehicle sector (sector 24) shows the largest change in output, partly due to the fact
that die demand for motor vehicles is price elastic. Recall mat the largest increase hi prices also occurred
in the motor vehicles sector. The 3.8 percent reduction in prices produces an increase in output of 5.3
percent relative to the base case.
Significant output effects are also seen hi the petroleum refining sector (sector 16) with a 3.2 percent
increase, hi electricity (sector 30) with a 3.0 percent increase, and hi other transportation equipment
(sector 25) with a 1.6 percent increase. The large gains in output for these industries are mostly due to
the decline in then* prices. In manufacturing, the sectors exhibiting the most significant output effects are
metal mining (sector 2) with a 2.0 percent increase, and primary metals (sector 20) with a 1.8 percent
80
-------
Appendix A: Cost and Macroeconomic Modeling
2-
3
4
3
6
7
S
9
10
11
12
13
14
15
16
17
It
19
20
11
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
43
Electric1'
1974 -0.47 -4.84
,-4MB'. '"*4M
-0.44
-0JI
1978 -0.86 -3.28
-2,92
im
1980
1981
1982 -0.98
J98J ' 4.0?
1984 / 4.J1
198£ , 4,21,
1986 , -L27
im xut
1988
-2J9 -0.68
4&T 4&
<&n
-2.26
-3.45,
-74*
1990
increase. Twenty of the remaining industries
exhibit increase hi output of less man 0.9
percent after pollution controls are removed.
While most sectors increase output under
the no-control scenario, a few sectors decline
in size in the absence of air pollution
controls. The most notable of these are food
and kindred products (sector 7) which decline
by 0.5 percent, furniture and fixtures (sector
12) which decline by 0.6 percent, and rubber
and plastic products (sector 17) which decline
by 0.3 percent. These sectors are among the
least capital intensive, so the fall hi the rental
price of capital services has little effect on
the prices of outputs. Buyers of the
commodities produced by these industries
face higher relative prices and substitute
other commodities in both intermediate and
final demand. The rest of the sectors are
largely unaffected by environmental .
regulations. : '
Changes in Employment Across Industries
The effect of the CAA on employment mmmmmmmmmmmmmmmmmmmmHfmmmmmmmmm
presents a much more complicated pic^ire.
Although Jorgenson-Wilcoxen is a full-employment model and cannot be used to simulate
unemployment effects, it is useful for joining insights about changes hi the patterns of employment
across industries. Percentage changes in employment by sector for 1990 are presented hi Figure 27,
which is located at me end of mis appendix.
For 1990, the most significant changes hi the level of employment relative to the control scenario
occur in motor vehicles (sectof24) which increases 12 percent, other transportation equipment (sector
25) which increases 0.8 percent, electric utilities (sector 30) which increases 0.7, percent, and primary
metals (sector 20) which increases 0.6 percent The level of employment is higher relative to the control
case in 10 other industries.
For a few sectors, the no-control scenario results in changes hi real wages which cause reductions in
employment. The most notable reductions in employment under the no-control scenario occur in
tobacco manufacturing (sector 8) which declines 1.2 percent, furniture and fixtures (sector 12) which
declines 0.8 percent, rubber and plastic products (sector 17) which declines 0.8 percent, food and kindred
products (sector 7) which declines 0.7 percent, stone, clay and glass products (sector 19) which declines
0.6 percent, and instruments (sector 26) which declines 0.6 percent These sectors are generally those in
which the level of output was lower hi 1990 relative to the control scenario, since they are among the
least capital intensive and the fall in the rental price of capital services has little effect on the prices of
outputs. Buyers of the commodities produced by these industries face higher relative prices and
substitute other commodities hi both intermediate and final demand. It is interesting to note that several
81
-------
Appendix A: Cost and Macroecononac Modeling
i
2
3
4
S
6
7
S
9
10
of the least capital intensive sectors experience insignificant employment effects in the short run (1975)
under the no-control scenario, but increasingly adverse effects over the 20-year period of analysis. .These
include food and kindred products, furniture and fixtures, rubber and plastic products, stone, clay and
glass products, and instruments.
Examination of the transition of employment in the economy from the initial equilibrium to 1990
reveals that the employment effects of the CAA on motor vehicles, transpoiliaiQnjiquipment, electric
utilities, and primary metals persist over me entire period of analyst* Emp1<%ment varies from: an
increase of 1.7 percent in 1975 to 12 percent in 1990 in motoffwhicles; from an increase of 0.7 in 1975
to 0.8 percent in 1990 in transportation equipment; from an ifrease of 1J percent in 1975 to 0.7 percent
in 1990 in electric utilities; and from an increase of 0.8 percent in 1975 to 0.6 percent in 1990. /"
[Figure 24. Percent Difference in Real Investment Between Control and No-eontrokScenarios.
u. :~'*'•'* {"
82
-------
Appendix A: Cost and Macroeconomic Modeling
figure 25. Percent Difference in Price of Output by Sector Between Control and No-control Scenario for 1990.
•11 W U
---•=-
> «• tl ttttJM *» I
Figure 26. Percent Difference in Quantily of Output by Sector Between Control and No-control Scenario for
1990. ^f;
1 1
Jl
I
ll
11 a u u » t> 17 11 » >t 11 n a
Swtor
83
-------
Appendix A: Cost and Macroeconomic Modeling
'igure 27. Percent Difference in Employment by Sector Between Control and No-control Scenario for 1990.
IP
I
w »*<
rilMH>ia»MM
-.-jsr
*•:«"
-==
Cost and Macroeconomlo Modeling References
" ,._ '_n _TP sinn
Programs:
Environm
J ,.
6
7
S
9
10
11
12
13
14
a
16
17
^ipEhe. Macroeconomic Impacts of Federal Pollution Control
for the Council on Environmental Quality and the
Congressional Budget
Fossil Fuels, Washington
Charges as a Response to Global Warming: The Effects of Taxing
r.S. Government Printing Office, August 1990.
Data Resources, Inc., "TheMacroeconomic Impacts of Federal Pollution Control Programs: 1978
Assessment," Report prepared for the Environmental Protection Agency and the Council on
Environmental Quality, 1979. .
Data Resources,, In6^ "The Macroeconomic Impact of Federal Pollution Control Programs: 1981
Assessment," Report prepared for the Environmental Protection Agency, July 17,1981.
Freeman, A.M., "Air and Water Pollution Policy," in P.R. Portney (ed.), Current Issues in U.S.
Environmental Policy, Johns Hopkins University Press, Baltimore, 1978.
Hazilla, M., and R.J. Kopp, "Social Cost of Environmental Quality Regulations: A General Equilibrium
Analysis," Journal of Political Economy, Vol. 98, No. 4, August 1990.
Industrial Economics, Incorporated, "Sources of Error in Reported Costs of Compliance with Air
Pollution Abatement Requirements," memorandum to Jim DeMocker, EPA/OAR; October 16,1991.
84
-------
Appendix A: Cost and Macroeconomic Modeling
Jorgenson, Dale W. and Peter J. Wilcoxen, "Environmental Regulation and U.S. Economic Growth,"
2 RAND Journal of Economics, Vol. 21, No. 2, Summer 1990(a), 314-340.
3 Jorgenson, Dale W., and Peter J. Wilcoxen, "Energy, the Environment and Economic Growth," in
4 Handbook of Natural Resource and Energy Economics, Allen V. Kneese and James L. Sweeney,
5
eds., Volume 3, Chapter 27, North-Holland, Amsterdam, forthcoming, 1993.
6 Jorgenson, Dale W., and Peter J. Wilcoxen, "Intertemporal General£quilibrium Modelmg of U.S.
7 Environmental Regulation," Journal of Policy Modeling, Vol. 12, No. 4, Winter 1990(c), 715-744.
'„="": -_-." 5"~ ~- '" 3 ~~ - •>' - " "-- - - "
s Jorgenson, Dale W., and Barbara M. Fraumeni, "Relative Prices and Technical Change," in E. Bemdt
9 and B. Field, eds., Modeling and Measuring Natural Resource Substitution, NOT Press, Cambridge,
10 MA, 1981. : : -.
•- ,. ~_5
11 Jorgenson, Dale W., and Barbara M. Fraumeni, "The Accumulation of Human and Nonhuman Capital,
n 1948-1984," in R.E. Lipsey and H.S. Tice, eds., The Measurement of Saying, Investment, and
13 Wealth, University of Chicago Press, Chicago, 11,1989.
14 Kokoski, Mary F., and V. Kerry Smith, "A General Equilibrium AWysis of Partial-Equilibrium Welfare
is Measures: The Case of Climate Change," American Economic Review, Vol. 77, No. 3, June 1987,
M 331-341. --.- - > -^
n U.S. Environmental Protection Agev^JEnvironmshtal Investments: The Cost of a Clean Environment^
Report to the Congress, Office of Policy, Planning and Evaluation, EPA-230-12-90-084, December
19 1990. vv. C""" ^:>:^~'
20 Verleger, Philip K., Jr., "Clean Air ^Regulation and the LA. Riots," The Wall Street Journal, Tuesday,
21 May 19,1992* A14-
22 Wilcoxen, Peter Jtp- 3fJte Effects of Environmental Regulation and Energy Prices on U.S. Economic
23 Performance, Dctitoi^thesis presented to the Department of Economics at Harvard University,
24 Cambridge, MA, December 1988.
85
-------
[This page intentionally blink] f
86
-------
Appendix B: Emissions Modeling
2
3
4
5
6
7
t
9
10
11
12
13
14
15
16
17
II
19
21
22
23
24
25
26
27
2S
29
30
31
32
33
34
35
36
37
39
39
40
41
This appendix provides additional detail regarding the methodologies used to estimate control and
no-control scenario emissions from each of the six principal emission sectors; industrial combustion,
industrial processes, electric utilities, on-highway vehicles, off-highway vehicles, and
commercial/residential sources. This appendix is adapted from the draft report "The Impact of the Clean
Air Act on 1970 to 1990 'Emissions; Section 812 Retrospective Analysis," March 1,.1995 by Pechan
Associates^ which distills the , V:: .;-."-•-.
methodologies and results associated
with the emission modeling efforts
by Argonne National Laboratory
(ANL), ICF Resources Incorporated
(ICF), Abt Associates (Abt), and the
Environmental Law Institute (ELI).
. The control scenario emission
' results are similar, but not identical,
to Official EPA historical emission
estimates provided by the EPA
National Air Pollutant Emission
Trends Reports.79 Comparisons
between the current estimates and the
Trends data for SO* NO,, VQe^andT
CO are presented in Figures 28, 29,1'
30, and 31, respectively. No/;: :*3™
comparison between the particulate
matter (PM) emission estimates •
developed for this analysis and the
estimates generated for % 1BA.
Trends Report are
igure 28. Comparison of Control, No-control, and Irends SO,
Emission Estimates. , - - •
40
1 30
H
. 20
2
I '
10
1975 1980 1985
Year
1990
'igure 29. Comparison of Control, No-control, and Trends NO,
jEmission Estimates.
however, since Trends estimates pw
presented for only PM10, a subset of
the TSP measures developed for this
analysis. More detailed tables
providing emission estimates by
sector and by target year for TSP,
SO* NO,, VQGi CO, and Lead are
presented in Tables 47, 48, 49, SO,
5 1 , and 52, respectively, at the end of
this appendix.
40
30
20
10
1975
1980
1985
1990
Though the EPA Trends and the present study emission profiles are similar to each other, they
should not be expected to match precisely. This is because the emission estimates developed for the
present study are based on modeled macroeconomic and emission sector conditions. Even though the
" EPA/OAQPS, "National Air Pollutant Emission Trends 1900 -1994,7 EPA^54/R-95-011, October 1995.
87
-------
Appendix B: Emissions Modeling
i
2
3
4
f
6
7
I
9
10
11
a
13
14
IS
16
17
It
19
20
21
22
23
24
25
26
27
U
29
30
31
32
33,
34
33
36
37
31
39
40
41
42
macroeconomic and sector models themselves are constructed and calibrated using historical data,
modeled replications of historical trends would not be expected to precisely capture actual historical
events and conditions which affect emissions. Relying on modeled historical scenarios is considered
reasonable for the present analysis since its purpose is to estimate the differences between conditions
with and without the CAA. , . ;!-'
Comparing actual historical
emissions with modeled no-control
emissions would lead to an
inconsistent basis for comparisons
between scenarios. Using models for
both scenarios allows potential model
biases to essentially cancel out
In general, however, these
comparisons show close
correspondence between control
scenario and Trends estimates with
the largest differences occurring for
VOC and CO emissions. The Trends
report VOC estimates are generally
higher than the control scenario
estimates due to the inclusion of
Waste Disposal and Recycling as a ,
VOC source in the Trends report
This inconsistency is of no
consequence llnce Waste DJ
Figure 30. Comparison of Control, No-control, and Trends VOC
• Estimates. *a±;* °n£i~~~~
30
I 20
10
1975
1980
1985
1990
ij, =~ -_.
4 ,"__^T"J^?5
the ^m
essentially uncontrolled by the
historical CAA and Jibusrefbre do
_ -T3?"-«---•' '*--**£
appear as a differences
control and noMControJ
The higherwCb emission c
the Trends Report are
associated with higher off-
vehicle emissions estimate|: Again,
since off-highway emisskms do not
change between me control and no-
cootrol scenario iapB present
consequence.
Figure |l||Compwfaon of Control, No-control, and Trends CO
"
200
ISO
100
50
^ Control
^.No-Control
^. TRENDS
1975
1980
1985
1990
The following sections of this appendix summarize the methodologies used to model control and no-
control scenario emissions for each of the six major emission sectors. Additional details can be found in
the supporting documents listed in the References section of this appendix.
-------
Appendix B: Emissions Modeling
Industrial Boilers and Processes
i - . • • '
2 For the purposes of the retrospective analysis, the industrial sector was divided into two components:
3 (1) boilers; and (2) industrial processes and process heaters. The factors affecting emissions from these
• 4 •" two source types are different, and, as a result, separate methods were used to calculate control and
5 no-control scenario emissions in each of the target years. To analyze the change in emissions from
« industrial boilers, ANL used the ICE model (Hogan, 1988). This model was developed under the
7 auspices of NAPAP to forecast State-level fuel choice and emissions from conventional, steam raising,
a industrial boilers. For the retrospective analysis of industrial processes and fuel use emissions from
9 process heaters, ELI used the EPA Trends methods and the ANL MSCET data base (EPA,. 1991; Kohout,
10 1990). The Trends report contains estimates of national emissions, for a variety of industrial sources for
11 the time period of interest The MSCET data base provided the spatial distribution used to calculate
12 State-level emissions. . '---_--?
a The distinction between industrial boilers and non-boiter industrial processes was necessitated by the
14 structure of the CAA regulations and by the factors affecting emission levels from these two source
is types. Boilers are regulated differently from processes and process heaters. Emissions from industrial
it processes are primarily a function of levels of industrial activity. The emissions from fuel combustion,
17 however, are a function of energy use and fuel choice as well as industrial activity. Fossil fuel emissions
is in the absence of the CAA are not proportional to industrial output, since the level of energy use is a
19 decision variable for the firm in its production process. Therefore, in the ICE model simulations used to
estimate no-control scenario boiler emissions, the level (and type) of energy use were determined first,
21 and then the effects of emission regulation were tal^a into account.
22 Ovorvlew of Approach
, - " ~-f-"- -'?:-=- :f_.j*""s-
-" - 1= '—«. " -%--•=-,,=-.>..-'-
21 Industrial Boiler! _." fr"-?**. ^/W
2< ICE model inputs mcJodf liie||dces, total fossil boiler fuel demand by industry type, and
25 environmental control coste||||i|i6utouts of die ICE model were SO2, NO^ and TSP emissions by State,
26 industry, and boiler size class. The model runs in 5-year increments and has a current base year of 1985.
27 The boiler demand input data were required at the State level. The following seven industry types
a were included in the IGE model: Standard Industrial Classification (SIC) codes 20,22,26,28,29,33,
29 and "other manufacturing." ANL's approach assumed that industrial boiler fuel use occurs only in the
x manufacturing sector. Fuel price data in each of the target years was required at the Federal Region
31 level. Prices by grade of coal and petroleum product, such as sulfur content and heating value, were used
32 by the model. These prices were used to determine the cost of compliance and to determine emissions
33 when the regulations are not binding.
34 Control costs were computed by engineering subroutines in the model. These costs were used by the
33 ICE model's fuel choice component to determine the effect of CAA-related costs on the market share of a
•« particular fuel. This fuel choice decision only applies to new industrial boilers, since the cost of existing
emission controls are not in the ICE data base and fuel choice is not re-evaluated for existing boilers.
89
-------
Appendix B: Emissions Modeling
2
3
4
S
6
7
I
9
10
11
12
13
14
a
16
17
II
20
21
22
23
24
25
26
27
21
29
30
31
32
33
34
35
Industrial Processes and In-Process Fuel Combustion
The calculation of historical emissions from industrial processes uses EPA Trends methods to
estimate national emissions for the analysis years, then allocates these emissions to States using the State
shares from the MSCET data base.
industrial sectors.
to allocate
istribution
it the MSCET
ions
MSCET uses a variety of methods to estimate historical emi
For industrial process emissions, MSCET is based on histori
emissions based on the[State level distribution of the polluting^i^tivities. Tbj§
benchmark is based on the 1985 NAPAP Inventory (EPA, 198?). This
data corresponds directly to the 1985 NAPAP Inventory, and mat, for any State, the sum
from Source Classification Codes (SCCs) that comprise the Mpspfindustry sector are
MSCET data for that State and sector. Data from Trends are used tprMSCET to provide information on
changes in the aggregate level of control for years other than the 1985 benchmark. Since no direct
correspondence existed between the Trends data and MSCET, a reMplhlpwas developed to link
MSCET sectors to Trends industry categories to industry categories UTth^ifWjaodel, which was used to
change activity levels for the no-control scenario. -:-:-~
Table 32 shows the relationship between the sector d
model. The mapping from MSCET to J/W and Trends &use4|i5
activity and emission control for the calculation of n&control seengtio emissions
', 7>endj,andtheJ/W
changes in aggregate
19 Establishment of Control Scenario ipdis
Energy use and
industrial processes. Thelat$
comi
issions
fgpry incli
between boilers and non-boiler
kilns, internal combustion engines (e.g.,
emissions, which ^ycre jsubject1a|(
(Emissions from some types of i
sources was tar
this study, ANL assumed:
boiler portion of
may be affected indirectly
of process heat The focus of this analysis is on boiler
Oiingly stringent regulations over the 1970 to 1990 period.
were also regulated, but regulation of non-boiler
industrial process itself, not on its fuel combustion) For
boilefluel use is affected by emission regulations. The non-steam
not directly affected by the CAA. This portion of the emissions
in industry activity level and fuel consumption. The emissions
fromtton-boUerffldustrialpixx^
.Control Scenario Boiler Emissions
' Ckratrol sceiunfaboilCT ^ The
MSCET data base provided an estimate of historical emissions for total fossil fuel combustion by
industry.,Since MSCET does not identify the two required components of boiler and non-boiler
emissions, ANL defined the residual of the ICE model control scenario and MSCET as the non-boiler or
in-process fuel use emissions. For the relevant study period, MSCET provided a control scenario
90
-------
v •" .A °"
r f f ' X'/\
'\ - '" ' - -
' I i
i;
I
1
I-
«*
f •'*'
I ,'
-------
*,.;.. $».;-%:$ ft Wfaf » Cxi
- , •• ,VC "-ft-; "' '': ^ >' *]r - 'I
-------
Appendix B: Emissions Modeling
i estimate of total boiler and non-boiler emissions, which was used to calculate the control scenario State-
2 level boiler emissions based on a special run of the ICE model.10
i In order to use ICE to model the historical emissions path, it was necessary to construct a new ICE
4 model base year file and new user input file so that the model could begin its calculations from 1975
s conditions. Construction of the base year file is discussed first The construction was completed in two
6 stages, using two different data sources. The user input file has several elements, including energy prices
7 and historical boiler fuel use. The description of me construction of me control scenario user input file
« follows the discussion of the ICE base year file. The model base year file provided the energy use in
9 boilers and corresponding emission control regulations (State implementation Plans -SDPs- for example)
ID by several categories. These categories include: LJT.
-* - _.'- -~r
n > State;
12 * Industry group (one of seven);
n *• Fuel type (natural gas, distillate or residual fuel oil, and coal); j, ;
14 >• Boiler size class (MMBTU/hr, one of eight categories);
is > Utilization rate (one of five categories); and .--.=-_-_-„;
16 > Air quality control region (AQCR). '."- , ^
77 For the purposes of ANL's analysis, onjy the first three categories were assumed to vary. In other
is words, for each State, industry, and fuel^ype combination, the distribution of boiler size, utilization rate,
19 and AQCR was assumed to be constant, -Sver time^ however, changes in the aggregate composition of
State, industry, and fuel type would cause corresponding changes in the aggregate composition of the
21 other three characteristics. As nientfoned previous^, the cunent base year fUe was 1^ The
22 retrospective analysis requireds 1975 base year/lBeeause of data limitations, the approach to construct a
23 new base year was achievedin tiie fallowing two Hips: the. construction of a 1980 interim base year file
24 from the 1985 file, and then meMnstwctipn of the 1975 file from the interim 1980 file.
2s Estimates of boiler fossil fuel conaiiiypbn in 1980 for each State and major fuel type were provided
26 by Hogan (Hogan, 1988^ TJ^e estima^ are based on me assum
27 utilization, and AQCR distribiitiofi within a State are constant Through assuming this relationship, the
2« 1985 ICE base year was scaledlto match the data for 1980, thus forming the 1980 interim base year data.
29 To construct the 1975 base year file, the assumption of a constant industry mix for a State and fuel
x type was no longer necessary, since detailed data on each industry for 1980 and 1975 were available
3, from PURchased Heat And Power (PURHAPS) model data files (Werbos, 1983). These PURHAPS data
32 files were denyed^firom the Annual Survey of Manufactures: Fuels and Electric Energy Purchased for
33 Heat and Power (DOC, 1991). The available data in these files were for total fuel use not boiler fuel use.
34 To make use of these data, it was necessary to assume that the fraction of fuel used in boilers, for any
31 given State and industry, remained constant from 1975 to 1980. To the extent that the fraction of boilers'
36 heat versus process heat applications is a function of the specific industrial production process, mis
37 assumption is reasonable.
"MSCET does not provide State-level estimates of TSP, while ICE does. To estimate total regional TSP from fuel combustion, the
Trends model was employed. These national emissions estimates were allocated to the States based on the State-level shares of TSP from the
NAPAP inventory.
93
-------
Appendix B: Emissions Modeling
•v
/ Based on the assumption of constant boiler fuel fraction of total fuel use, the ratio of 1975 to 1980
2 energy use for each State, industry, and fuel type was applied to the corresponding record of the 1980
3 interim base year file to produce 1975 base year files.
4 Control Scenario Industrial Process Emissions r;^
5 To estimate boiler emissions of sulfur oxides (SOJ, NO,, and _¥OC fix>m indiisffial processes, data
6 from Trends were used. The percentage change in national emissions by T^n^caligory was applied to
7 the appropriate sector from MSCET to obtain State-level emissions. In some casejsf there are several
s categories in Trends mat match directly with MSCET categories (see tabie32). In these cases, the
» Trends sectors were aggregated and the percentage change wis computed. It was assumed that th* level
jo of control in each industry sector implied by Trends was uniform across States. The changes in::
11 emissions in each State are not equal to those at the national level, since the industry composition hi each
12 State varies. : " :. .
« Development of Economic Driver Data for the Control
H Scenario - Industrial Boilers and Processes
is The results of the J/W model were the primary source of activityWflie ICE model driver data.
16 These results were also used by ELI to produce the national results ibl industrial processes from Trends.
17 Bom ICE and Trends use the forecasted change uxiadustrial activity mat results under the no-control
it scenario. These data were in the foin|0frndustryjpecific changes in energy consumption and industrial
a output, for boilers and industrial processes. ":-:.;i /
"" - ~~ ~
20 Econoirfc Driver Data forli&^
21 Using the 1975 base year filers a sauting point, the ICE model estimated fuel choice and emissions*
22 basedonauserinjNitfi]^ The 1975,
23 interim 1980, and origmal i^5 base year files contained the required information on energy demand for
24 each industry group andSid||ii^Jhe da& in these three files were aggregated across fuel type, and other
25 boiler characteristics (for«c^Bi^||g%ize). These aggregated data provided the energy demand for three of
26 the target years. Since 1990 SJi^Ievel data on energy use by industry group were not available at the
27 time of the study, the NAEAP base case forecast for the ICE model for 1990 was used to provide the
2s demand data for this year, s
29 Its user input file^br ICE also requires a price input for each target year. These prices were input
» by Federal Regioa^for distillate oil, 4 grades of residual oil (by sulfur content), natural gas, and 1 1 grades
31 of coal (fay sulfur content and coal rank, i.e., bituminous and sub-bituminous). Prices for 1985 and 1990
32 were obtained from the NAPAP base case user input file. The prices for 1975 and 1980 are from U.S.
33 Department of Energy (DOE) data on State-level industrial energy prices (DOE, 1990). Regional prices
34 of natural gas, distillate oil, steam coal, and residual oil were constructed by aggregating expenditures
•35 across States within each region and dividing by total British thermal unit (BTU) consumption for the
36 years 1975, 1980, and 1985. Since prices by sulfur content grade are not reported by mis DOE source,
37 ANL assumed that the sulfur premium implied by the 1985 ICE model input file was proportional to the
> average price. Based on this assumption, the ratio of the regional coal and residual oil price hi 1975 and
39 . 1980 to the 1985 price was applied to the 1985 price in the ICE model base case file for each grade of
94
-------
3
4
5
6
7
8
9
70
11
13
14
ss^^^^^^^CTsr-
*f
IS
16
17
It
20
21
22
23
24
23
26
27
21
29
95
-------
•^
Appendix B: Emissions Modeling
advantageous to buy this local coal, which raises the price back to an equilibrium level near to mat of the
control scenario. This is even more likely to be true of industrial delivered prices, since industrial prices
are more affected by transportation costs than are the utility prices. No additional ICE modeling was
performed. =
Economic Driver Data for the Industrial Process Approach A :%
The J/W model was also used to account for activity level changes hi the calculation of industrial
process emissions under the no-control scenario. The correspondence between fiends, MSCET, and die
J/W model was used to apply changes hi industrial activity in each target year to each industrial process.
No-control Scenario Emissions
10
11
12
13
14
a
16
17
II
19
21
22
23
24
25
26
27
2S
29
30
'31
32
33
34
35
36
37
38
Industrial Boiler Emissions of SO,, NO,, and TSP ' ---': \".>
The CAA imposed different regulations, SIPs, and New Source Performance Standards (NSPS) that
apply to industrial boilers of varying size. The primary effect of CAA regulations on industrial boilers
was simulated by defining the Air Quality Control Region (AQCRX the resulting SIPs, and subsequent
NSPS for boilers. The industrial boiler SIP regulations were included in me ICE base year file discussed
hi the previous section. Since the ICE model estimates new boiler emissions for each target year, the
ooiler NSPS are mput through the ICE^sfflf files. Industrial NSPS were implemented hi two phases. The
1971 regulations are imposed for meygnfy years 1975 and 1980. The 1984 NSPS revisions are imposed
hi the study years 1985 and 1990. For the no-control scenario, ANL set the SIPs and NSPS to a flag that
indicated "no regulation." ,:
to Industrial Boiler Emissions of CO and VOC
Two of the criteria pollutants emit
industrial fuel (X>nibusidti^|i
included as ou^>uts of%|IC|B|iuXiel. Therefore, CO
and VOC emissions wereaniiyj06d§separately using
Trends methods. Control s&ija^CQ and VOC
emissions were taken directly from Trends.
To estimate CO andrlTOC emissions from
industrial rombustionjbr the no-control scenario, fuel
use for industrial mimufacturing was adjusted,
reflecting fuel consumption changes estimated by the
J/W model. These changes hi the level of fuel
consumption by industrial combustion were also used
hi ANL's ICE boiler model. Changes hi industrial
combustion fuel use by manufacturing hi the
no-control scenario are reported in Table 33. These
estimates represent an average of several sectors,
which were developed by ANL as part of the
modeling process for ICE.
ISEF
1980
1985
oil
GO
Gas
Coat
m
C3as
Coal
•.<•
Ott
V*
-.0061
96
-------
Appendix B: Emissions Modeling
No-control scenario emissions were computed using 1970 emission factors. Since there were no
i add-on controls for industrial combustion VOC and CO emissions, it was not necessary to adjust the no-
i control scenario for changes in control efficiency.
4 Emission estimates were regionalized using State-level emissions data fix>m industrial boilers
5 recorded hi MSCET. For the control scenario estimates, VOCs were regionalized using the MSCET
6 State-level shares for industrial fuel combustion. In the no-control scenario,^ JStafe- level shares were
7 held constant. The control scenario emissions of CO were regionalized usmg mi coptrol scenario NOZ
« emissions from the ICE model. This approach assumes that CQKemissions are confisterit with NOX -•'
9 emissions. The no-control scenario CO emission estimates from industrial i»mbuil|pii;aiprioes were?-
10 regionalized using no-control NOX emission estimates from industrial combustion sources.
11
16
Industrial Process Emissions
12 A wide range of controls were imposed on industrial processes. Thes« emission limits are embodied
13 in the assumptions of control efficiencies in the Trends model. Data on national no-control scenario
14 emissions from industrial processes were provided by EPA. These data weri combined with MSCET to
is produce regional-level results. ,
•i Estimates of lead emissions fromJaiustrial boilers and industrial processes were completed by Abt
Associates. The methods used for calculating lead emissions from industrial processes and industrial
19 boilers were similar. The starting point was the TJSI, which provides air toxics emissions data for
20 manufacturingiacilities with^Eno|elkan 10 employees. To estimate lead emissions from industrial
21 boilers and processes, l9901Si^§^^l lead emissions data were extracted from the TRI. These data
22 were then adjusted to Create estmiafM|||kad emissions from industrial sources under the control and
23 no-control scenarios lor each of thetajgjfttiplrs. For the control scenario, lead emissions for 1975,1980,
24 and 1985 were obtained bj jpctractrng angniission factor and a control efficiency for each lead-emitting
25 industrial process m the p^^dato base: These emission factors and control efficiencies were
26 multiplied by the eo>noniicasilyi^jdata for each year for each process as reported in Trends to yield
27 estimated control scenario emissions by industrial process. Each industrial process was assigned a code
28 to correspond with energy consumption data by industrial process compiled in the National Energy
29 Accounts (NBA) by the Bureau of Economic Analysis, and emissions were summed over all processes to
x obtain a total for each target year.
__%•,"" ' • .
31 For consistency with the other emission estimates in this analysis; industrial process no-control
32 scenario lead emissions were adjusted for changes in industrial output, and for changes in emissions per
33 unit of output due to control technology applications. Changes in industrial output were accounted for
34 using results from the J/W model. Lead-emitting industrial processes in the Trends data base were
15 assigned to a J/W sector. For each sector, the percentage change in economic output was used to adjust
36 the economic activity data for that process from the Trends data base. These adjusted economic output
37 figures were used with the 1970 emission factors and control efficiencies to derive the estimated
3s no-control scenario lead emissions for each industrial process in each target year. The process-level
-> emissions were then aggregated to the NEA-eode level as in the control scenario.
97
-------
Appendix B: Emissions Modeling
i The lead emission estimates from industrial processes, by NBA code, were used to derive percentage
2 changes in emissions under the control and no-control scenarios by NBA code for application to the TRI
3 emissions data. Since TRI data are reported by SIC code, NEA codes were "mapped" to the appropriate
4 SIC codes, and then the percentage change for each NEA code was used to represent the percentage
5 change for all SIC codes covered by that NEA code. :i"~
7 ~- -~
. ~ - --'
6 To calculate lead emissions from industrial boilers, Abt Associates developed estimates .of lead
? emissions from industrial combustion under the CAA for each GjfiyMa^ It-ends data base
« contains national aggregate industrial fuel consumption data byluel type. For each fuel type, the fuel
» consumption estimate was disaggregated by the share of that fiiel used by each NEA industrial category.
10 The Trends data base also contains emission factors for industrial fuel use, by fuel type, as weB as
// control efficiencies. The lead emissions from industrial combustkm for each NEA category were derived
12 by multiplying the fuel-specific combustion estimate for each NBA category by the emission factor and
u control efficiency for that fuel type. The result was emissions of lead by NEA code and by fuel type.
» Emissions from all fuel types were then summed by NEA code. The NEA data were used to
a disaggregate the industrial fuel consumption figures, based on the assumption mat the ICE are the same
u among all industries covered by a given NEA code. :' ^ ^°-'f
29
34
i? To estimate no-control scenario lead emissions, the macroeconomic effect of the CAA and the
IB change in emissions per unit of output that resulted from specific pollotion control mandates of the CAA
19 were both taken into account As in the eqntrol scenario, the national aggregate industrial fuel
20 consumption estimate by fuel type was disaggregated by the share of that fuel used by each NEA
21 industrial category. The fuel use wastipifadjustiBidln two ways: some NEA codes were specifically
22 modeled by the ICE model, and for tlie remaining |JEA codj», J/W percentage changes in fuel use were
23 applied. These fuel use estimates were men c|iil|^ flitii the 1970 emission factors and control
24 efficiencies for industrial combustion by fuel ty|ie-ih»&lfae Trends data base to obtain no-control
a scenario combustion-related teadc|n||sic>ns from industrial j>oilers by NEA code. These estimates of
26 total lead emissions by NEA codej||ie»B matched to SIC codes, and men to the data in the TRI data base.
27 This approach assumed that an average fnijssion value was assigned to all reporting TRI facilities in a
2s given SIC code.
X The off-highway vehkle sector includes all transportation sources that are not counted as highway
31 vehicles. Therefore, this sector includes marine vessels, railroads, aircraft, and off-road internal
32 combustion engines and vehicles. As a whole, off-highway vehicle emissions are a relatively small
33 fraction of total national anthropogenic emissions.
Overv/eiv of Approach
35 The process used by ELI to determine the national level of emissions from the off- highway
36 transportation sector is similar to the procedure outlined above for industrial processes. To estimate the
37 emissions of criteria air pollutants from these sources under the no-control scenario, the historical
3t activity levels were held constant, rather than attempting to calculate a new no-control scenario level of
39 off-highway vehicle activity. This assumption was necessary since the off-highway activity indicators
98
-------
Appendix B: Emissions Modeling
(amount of fuel consumed, and landing and take-off cycles for aircraft) do not .have direct
2 correspondence with a given J/W category. The national no-control scenario emissions of criteria air
3 pollutants from these sources were simply derived by recalculating emissions using 1970 emission
4 factors.
5 Development of Control Scenario
6 To estimate control scenario emissions, the analysis relied o&Frends memb^ using historical
7 activity indicators, emission factors, and control efficiencies. Essentially, the estimates of off-highway
a emissions under the control scenario represent the historical estimates from the rra%& data base.
9 No-control Scenario Emissions Estimates
10 The calculation of off-highway emissions for the no-control scenario required the Trends data to be
11 adjusted to reflect changes in controls and economic activity in each of the target years. Linking source
n activity changes with economic activity for this section is not straightforward. The economic activity
u data for off-highway engines and vehicles are expressed either in terms of amount of fuel consumed, or
14 in terms of landing and take-off cycles for aircraft. Neither of these off-highway activity indicators has a
is direct correspondence with a given J/W sector, making the sort of direct linkage between Trends
is categories and J/W sectoral outputs that was used for industrial preKassses mappropriate.
j"g£T - .-' ~ . _!?-"
17 In the absence of a link between theje^nomic factors that are determinants of emissions from this
is sector and the available economic ac|n|iyibrecasp the no-control scenario emissions of criteria air
pollutants from off-highway mobil&iburces wefe estimated based on the same historical activity levels
20 used for the control scenario. AltfiCfgh there $/^ changes in sectoral output and personal income that
21 might have had an effect onjoAil^vray vehiclB asage,'these changes were deemed to be small and not
22 likely to have ainajor effect^||e|pissions from this sector.
23 Emission factoid fb^rch of uliip||^|way sources were also held constant at 1970 levels to
24 calculate no-contt^jieiiEli^emissioTMyieach target year. The national emissions of criteria air
25 pollutants fromtiiese5Oj|ti||e|i||re thenHcalculated using 1970 emission factors.
26
27
National and Stato-Lovol Off-Highway Emission
21 Table 34 summarises national-level emission estimates for off-highway sources. The emission
29 estimates derivediiom using the methodology discussed above yielded results that seem counter-
30 intuitive. The emissions from off-highway sources, in particular the emissions from aircraft, are lower in
31 the no-control scenario than those projected for the control scenario for most pollutants. This is a result
n of calculating emissions using 1970 emission factors, since the 1970 emission factors for aircraft are
33 lower than me aircraft emission factors in later years.
34 ELI identified several potential sources of uncertainty in the emission estimates for this sector. First,
15 the assumption that the total level of off-highway vehicle fuel consumption is constant between the two
>« scenarios may be flawed. Second, the use of 1970 emission factors in the no-control scenario may fail to
capture significant changes in technology These technological changes are implicitly captured in the
99
-------
. Appendix B: Emissions Modeling
TabteM,
•- *L' ,"*i°;^«^
Sctaar
-No-Caaao!
1986
-, ~|* I.
•\~*y >
:£
. ;^ y "-; -' v,,-^^
- ^ ^ ,* •."•^>: - -Si**- -
2,042.7
406.4
• 4*
<, ; ''•^ •.^'-••.\'f ," ^*%\"^% *''$'
control scenario and it is possible tiiat these technological changes may also have occurred under a
One possible response to the biases created by the use of 1970 emission factors for all years in the
no-control scenario is to test how results might differ if the emission factors used for the control
scenario, which would include technological change, were also used for the no-control scenario.
However, using this treatment of emission factors, the emissions projections from the adopted
methodology from non-highway sources in the no-control scenario would be identical to the emissions
projections under the control scenario. The reason for this is that the economic activity levels were not
adjusted for the calculation of emissions under the no-control scenario.
100
-------
Appendix B: Emissions Modeling
In order to disaggregate the national data to a State level, the methodology used the MSCET data
2 base, which is described starting on page 89. Emissions of VOC, SOD and NO, were regionalized using
3 the State-level shares from the MSCET methodology. The emissions of TSP were regionalized by using
4 the State-level shares for SO, reported by MSCET, and the emissions of CO were regionalized using the
5 State-level shares for NO« also reported by MSCET. The potential bias that this introduces is likely to
0 be small, due to the relative homogeneity of off-highway vehicle emission sources. As with
7 regionalization of industrial process emissions, the State-level shares are held constant between the two
« scenarios. To the extent that the distribution of economic activity between States was not constant over, 1
9 the period of the analysis, holding State-level emission shares constant may bias the results, although me
10 direction and magnitude of the potential bias is unknown. ^ , ~- - * ".-„ 4>,
On-Hlghway "";S\f
12 This section addresses the highway vehicle portion of the transportation jHCtor. Highway vehicle
13 emissions depend on fuel type, vehicle type, technology, and extent of travel Emissions from these
14 vehicles have been regulated through Federal emission standards and enforced through in-use
is compliance programs, such as. State-run emission inspection programs. Vehicle activity levels are
16 related to changes in economic conditions, fuel prices, cost of reflations, and population characteristics.
i? Emissions are a function of vehicle activity/levels and emission rates per unit activity.
is TEEMS was employed by ANL iojiinalyze the transportation sector. The modeling system links
several models, disaggregate and aggregate, to produce State-level estimates of criteria pollutants. The
20 system is subdivided into two modules: an activity/energy module and an emissions module. Each
u module contains multiple models, TEEMS has be^dociimentedm several reports and papers (Mintz
22 and Vyas, IS^ljVyas and Saric*$ 1986; Saricks; 1985). It has been used for several policy analyses and
u assessment studies iotpOE and 1|^^ This section presents an overview of the approach used to
24
-------
Appendix B: Emissions Modeling
i
i
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
IS
19
X
21
23
24
25
26
27
21
29
30
31
32
33
34
35
36
37
3t
and based on the VMT of each household type. The basis for household transportation activity
projection has been empirically established through analysis of the 1983-84 Nationwide Personal
Transportation Survey (NPTS) (FHWA, 1986; Mintz and Vyas, 1991). VMT are projected using mis
empirical relationship, and estimates of the elasticity of VMT to,vehicle operating cost are then made.
Energy consumption was estimated in each target year using VMT, shares of VMTT>y vehicle type, and
exogenously developed vehicle characteristics.
The following three models and an acco
personal travel activity projections:
year
1. The first model projected the target year distribution of households by their attributes. This
model employed an iterative proportional fitting (IFF) technique and projected the number of
households in each cell of the household matrix - each of which is defined by various
categories within six household attributes.
2.
The second model projected changes in vehicle ownership resuitmg fnom changes in income
and cost of vehicle operation. The model applied estima^ ownenihfp changes to each
target year household matrix such that the control values within each of the household
attributes, excepting vehicle ownership, remained unchanged, jr
3. The third model estimated the composition of household vehicle fleet by type (cars and
trucks), size, technology, and
4. An accounting procedure
of household attril
size, and fuel.
Each of these
Iterative
VMT|% vehicle to vehicle ownership in each combination
ion were accumulated by vehicle type,
in the following subsections.
This H*F model modij^^|iptrol scenario matrix of household counts. A household matrix was
developed from the 1983 'and upgraded to the year 198S using published aggregate data. The
procedure used in constructing^ib 1985 household matrix has been documented elsewhere (Appendix B
of Mintz and Vyas, 199iy3rhe matrix is defined by six attributes: (1) residential location (central city,
suburb, rural); (2) household income; (3) age of householder, (4) household size; (5) number of drivers;
and (3) number of vehicles. The household matrix has 3,072 cells, some of which are illogical (such as 1
person, 2 driversXilUbgical cells were replaced with zeros.
- r —r- 4^="^^*-
Household shares within each attribute in each target year were developed exogenously using data
from the Bureau of the Census and selected macroeconomic model runs. The projected total of
households and shares of households in each category of an attribute were supplied to the IFF model.
The model modified the control scenario household matrix to match the specified shares and total
number of households.
The IFF model treated household distribution within each attribute as a set of vectors. These vectors
were scaled to match the specified shares and household total. Following the initial scaling, a gradual
scaling technique was used to move hi the direction of the target shares. The scaling process was
102
-------
Appendix B: Emissions Modeling
repeated until closure was achieved for all attribute classes. Since vehicle ownership levels were
2 estimated by the vehicle ownership model (described in the next section), shares within the sixth
3 household attribute (number of vehicles held) were not specified, leaving it uncontrolled. This flexibility
4 of an uncontrolled attribute helped to facilitate the model operation. The number of households in each
5 class of vehicle ownership within the output matrix represents distribution of households using the
6 control scenario (1985) relationship of vehicle ownership to other household attributes^
7 Vehicle Ownership Projection (VOP)
s The VOP model projected the changes in vehicle ownership resulting from changes in the number of
9 licensed drivers, disposable personal income, and annual fuel cost of vehicle operation. The model is
10 based on historical household ownership rates. A target per-driver ownership rate was computed using
n disposable income and fuel cost This target rate represented desired ownership if income and fuel cost
12 were the only determinants. A parameter representing owneR^tpp^^Uities such as acquisition
13 effort, disposal effort, parking requirements, and other indirect aspects was applied to adjust this target
14 The new ownership rate was used to estimate the number of household vehicles.
75 The household matrix created by the IPF model was revised to match the projected household
16 vehicle ownership. Household shares within the first five attributes remain constant while those within
77 the sixth attribute (i.e., number of vehicles) were variable. A deviation; measure was defined and its
a value for each class within the first five attributes was minimized^ A let of simultaneous equations was
19 solved using Lagrangian multipliers. t.?tf ^ :i?
Projection of Vehicle Fleet Composition J>; il"
21 The composition of householdjvehicles was projected for each household matrix cell using a vehicle
22 choice model called the Disaggregate Vehicle Stock Allocation Model (DVSAM). Vehicles are defined
23 by type (auto, liglht truck), size (sm|U, jmid-size, full-size auto; small pickup, small utility/minivan,
24 standard pickup, ^gftj^^/stan^^^ji^e^any other size classification), fuel (gasoline, diesel,
25 methanol, ethanqt 4ffcomnjipssed naturj] jas), and technology (stratified charge, direct injection,
26 electric, fuel cell, or p^-*-^* ""^
27 The model computed veljclfiomposition based on an individual vehicle's utility to households and
2s household needs. A menupfviaicles classified by me previously mentioned vehicle attributes was
29 supplied to the model. The menu specified characteristics of each vehicle available to households.
» Vehicles were characterized by price, operating cost, seating capacity, curb weight, and horsepower.
31 These variables formed the basis for computing "utility" (analogous to consumer satisfaction). The
32 household matrix plovided demographic and economic attributes which, when combined with vehicle
33 usage m mites, define household needs. Vehicle usage (VMT) was computed as a function of income,
u number of drivers, and number of vehicles. A logk model was applied to compute vehicle ownership
15 shares. Several model enhancements facilitated modeling of limited range vehicles, and representation
36 of supply constraints and/or regulated market penetration.
37 Activity/Energy Computation
v An accounting procedure was applied to compute personal travel activity in terms of VMT by
vehicle type. Control scenario VMT per vehicle estimates for each cell in the household matrix were
40 developed from the 1983 NPTS. These rates were adjusted within the procedure on the basis of changes
103
-------
^Appendix B: Emissions Modeling
34
i in average vehicle operating cost per mile for each cell. The vehicle composition projection model
2 computes ownership shares and share-weighted change in vehicle operating cost Elasticity values were
3 applied to this change.
4 ANL assumed that VMT per vehicle remained nearly unchanged for a household matrix cell over
s time (with the exception of the effect of changes in vehicle operating cost). In other words, variation of
e VMT across household types is far greater than within household types. VMT peritbusehold vehicle
7 remained stable during the period from 1977 to 1984 (Klinger and IJizmva£,^86% Some increases
8 were observed hi recent years, which were attributed to lower fttel prices and increased household
9 income (DOC, 1991; FHWA, 1992). (A portion of the increase could be attributed to me method of
10 computing average VMT per vehicle.) The assumption that VMT per vehicle for each cell remainai
// nearly constant and was elastic relative to vehicle operating cost is reasonable. As households move
12 from one cell of the matrix to another, they "acquire" the VMT per vehicle rate of that cell. Thus, this
n approach accounted for changes in VMT per vehicle due to increased household affluence, increased rate
14 of driver licensing, changes in fuel price, and changes in vehicle technology^
-Ii = ~ --
is Goods Movement '---.- - -r- -
- -f - ~V-H>- r. "-ff
16 Energy and activity demand resulting from movement of 24 aggregate categories of commodities is
17 estimated by this subcomponent of the TEEMS activity module. Changes in commodity
is demand/production were provided by growth indexes by two-digit SIC generated by a macro model. A
19 model that projects shifts in mode shares junong truck, rail, marine, air, and pipeline modes was used,
20 followed by a procedure to compute tommies of travel for each mode, VMT by fuel type for trucks, and
21 energy consumption by operation type for ncro-highway modes. The model used 1 985 control scenario
22 data, which were compiled from railroad waybjl SffiQ^t| and publications, waterbome commerce
23 pubUc^ations,tianspoitation statisti^ and ornersbiar^es; The procedure used in developing the 1985
24 control scenario freight data has beea documented in an ANL report (Appendix A of Mintz and Vyas,
* 1991). """'V '-^ ' • :
= ,- -- j ;<•_ -~- --~-'-~~ -?•-
26 This goods movementaiodel was not^used for mis retrospective analysis because of funding and
27 time constraints. A procedure to estunafc track V^^ Published
21 historical VMT values (^HWi^9M; 1992) were used along with VMT shares by fuel and truck type
» from Truck Inventory and Use Surveys (TIUS) (DOC, 1981; 1984; 1990).
" "Ji8
™ " - _ ~-= " ' ' '
x Other Transportation Activities
- -- " • ?
31 The activity/energy module also has other models for developing activity and energy use projections
n for air, fleet automobiles, and bus modes. Fleet automobile activity estimates from an earlier study
n (Mmtz and Vyasj 1991) were used while other modes were not analyzed.
L0aet Emission*
35 Estimates of lead emissions in the transportation sector were developed by Abt Associates based on
36 changes in reductions of lead in gasoline. This estimation required the estimates of lead in gasoline
37 consumed over the period from 1970 to 1990 and the amount of lead content in gasoline that would have
39 been consumed in the absence of the CAA. These values were calculated using the quantity of both
39 leaded and unleaded gasoline sold each year and the lead concentration hi leaded gasoline in each target
104
-------
Appendix B: Emissions Modeling
/ year. Data on annual gasoline sales were taken from a report by ANL mat presented gasoline sales for
2 each State in each target year. For the control scenario, data on the fraction of gasoline sales represented
3 by leaded gasoline were used. For the no-control scenario, all of the gasoline sold was assumed to be
4 leaded. Data on the lead content of gasoline was obtained from ANL for 1975 through 1990. For 1970
i through 1975, the analysis assumed mat the 1974 lead content was used. _ --^f
,*i x^wjdtn
6 Estimation of No-control Scenario Emissions
' . ' \rr- ,.^??*l^€i.
7 TEEMS emissions projections were carried out by ANL in the following three st^LJ 2> „„„ ..._._
T ' ••' ""„ ~ "~ ~ " - -- _=">•"
s 1. Development of emission factors; "-<'' ° :f" 7 -"- ^.
9 2. Allocation of highway activity to States; and .': ;
10 3. Development of highway pollutant estimates. :.?:„„
// The following subsections describe the procedures used for computmg jhjghway vehicle emissions.
n Development of Emission Factors ''''r": • "i ' ^
.~ ' -" .: "-.. I/-; .-Htf
13 EPA's MOBILESa Mobile Source Emission Factor model was nse&to provide all of the highway
14 vehicle emission factors used to estimate 1975 to 1990 emission rates (EPA, 1994b). Documentation of
is the MOBILESa model is found in the Uaeis Guide for the MOBILES model.*1
Although the actual emission factors used by ANL are not documented in either the original ANL
n TEEMS model report or hi the Pechan siimmaiy rjepjor^ me Project Team provided direction that defined
a the emission factors to be use& I%r|he ccmtrol scenario, ANL was directed to use the official EPA
19 emission factors prevailing at the tinie for each target year. For example, the official EPA emission
20 factor being used in 19£0 for on^tighwayi vehicle NO. was to be used to estimate 1980 control scenario
"* "•" "~ j ~i ""^i-" ^— "=iLr **~ __
a on-highway vehicle NQ^emissioiiSilof scenario, the official EPA emission factors used
22 to estimate emissions in 1970 were tcvbe used throughout the 1970 to 1990 period.
.--~v -vri^*^ '. vrf"
23 It is important to note ffiitU|ini the 1970 on-highway vehicle emission factors to estimate no-control
24 scenario emissions for me eltiiefll970 to 1990 period may bias scenario emission differentials upward.
25 This is because it is possible mat technological changes to on-highway vehicles unrelated to CAA
26 compliance strategies may have yielded incidental reductions in emissions. However, EPA Office of
27 Mobile Sources (EPA/OMS) experts indicate that the two major technological changes in vehicles
22 occurring during the,period of the analysis -electronic ignition and electronic fuel injection- would have
29 yielded negligiblejemission reductions in the absence of catalytic converters.*3
x Another potential bias is introduced by assuming the CAA had no substantial effect on vehicle
n turnover. However, two factors render mis potential bias negligible. First and foremost, under the no-
* EPA/OAR/OMS, "User'f Guide to MOBILES,' EPA-AA-AQAB-94-01, M«y 1994; see also 58 FR 29409, May 20,1993.
0 Telephone convenatkm betwcdt Jim DeMocker, EPA/OAR and EPA/OMS/Aim Arbor Laboratory staff (date unknown). Nevertheless,
the Project Team did consider reviewing emission fetors fbr European wrtonwbUes to atteinpt to estimate no
-------
Appendix B: Emissions Modeling
i control scenario retired vehicles would be replaced by new but equally uncontrolled vehicles. Second,
2 no-control scenario vehicle use is greater in terms of VMT per year. This means no-control scenario
3 vehicles would reach the end of their service lives earlier, offsetting to some extent the alleged incentive
4 to retire vehicles earlier due to costs imposed by CAA control requirements.
5 Allocation of Highway Activity to States
j__"
6 TEEMS* activity module generated national activity and energy estimates. These activity totals were
7 allocated to States through a regionalization algorithm mat used time series date on historical highway
s activity shares by State. A trend extrapolation methodology nwas used mat stabUizes shifts after 5 years
9 in the future. For the retrospective analysis, historical highway activity snares for eachtarget year were
10 developed using data published by the Federal Highway Administration (FHWA) (FHWA, 1988; 1992).
u Development of Highway Pollutant Estimates ..
n Highway emission estimates were calculated in both scenarios for each target year using VMT
13 estimates generated by TEEMS and emission factors from MQBILESa. Control scenario activity levels
H were adjusted for the no-control scenario using economic forecasts and historical data.
is Control Scenario Emissions Calculation :
- -^ ~ ~ ~=*
u Control scenario data for the traiispoitation sector were compiled from several sources. Household
n counts and shares of households by saacjattributes were obtained from various editions of the Statistical
u Abstracts of the United States. Household income information was obtained from the control scenario
» runofrne J/W model. Fuel prie^*yereobtainedfr^
20 vehicle fuel economy and aggrejite VMT per vehicle were obtained from Highway Statistics (FHWA,
21 1988; 1992^ 7Tatte35 listed
~~"=?:
-'-*Siic-:i..'iK--"i
22 Table 36 shomhoiisehold shai^fWep^ped for the IFF model. The total number of households
23 increased from 63;4niiiJUu^in 1970 to^JJ million hi 1990. A gradual shift from rural to urban was
24 observed with inoveme^iois^^
23 1980 was an increase in shai(i||frliie lowest income category; more households moved to the highest
26 income group from 1970 to|^i(|2while the lower middle income group share expanded and the upper
27 middle income share declinld; line rate of household formation was high during the 1970's, which
21 resulted in increases in smaller and younger households. The trend hi younger households reversed after
29 1980 as household formation slowed. Average household size dropped from 3.2 hi 1970 to 2.67 in 1990.
x The number of licensed drivers increased throughout the analysis period as more and more young people
a were licensed to dii»B.
- ; :^:.f • •
32 Data lor the VOP model included disposable income per capita, fuel price, overall personal vehicle
33 fuel economy, and annual usage in terms of VMT. Table 37 shows these data for each year in the
34 analysis period.
is Data preparation for the model that projected household vehicle composition was limited to
36 characterization of existing technology vehicles. Seven vehicle size and type combinations were
37 characterized for 1975 and 1980 while one vehicle, minivan/small utility, was added for 1985 and 1990.
38 Control scenario vehicle characteristics are tabulated in Table 38. TEEMS'activity and energy
106
-------
Appendix B: Emissions Modeling
'
W&8X&
Date Item ..;,, ;*\
fo
> *%^v:
fet>^
, '<•<$ - **' 's '
•x'v ' 'WSB %%
•-' ^;,¥F , *
• ^4 *• s ^' **
»*Vj°AT' -. '
J \v J^f - * ^ ,
* \
. v*4^ '*i',^ ^v%
.V ,v.**f ^ ^f," 5
:^. .. .r \ . ..^*..t
DVSAM
Pod Prices
OVSAM
^--.
lnforjB«ioii A
Automotive New, Market D«* Book 1985.
\-v ^ ^«.% f ff vX- A< ^v.«'y^->.A"'%^'>v'' > ^. ^ ' ** *
-H m -
Vffl*
DVSAM
, -
''< , " ^ i '< X'< <*«' *'
j A f vf *'f AAv? ffyt. X-
» ---S'"' 7"V
<. % '
5
6
7
8
9
10
11
computation procedure was
estimates.
to produce personal vehicle travel and energy consumption
. Gommercial truck travel was not modeled but, historical data published by the FHWA (FHWA,
1987; 1991) were useds^FHWA publishes
-------
Appendix B: Emissions Modeling
0^*'^%^'"'
$-4
36,0
;-«;
31.6
31.4
313
30.3
, s ' 23U2
\<- *1 14 J
22.1
29.0
21,5
- , - itf
--; - ^ ^4
24.fi
32.2
- 334
10,4
14-
,4*J'
15.0 ;
52*5 *
6.6
26.0
53,5
108
-------
Appendix B: Emissions Modeling
». .-• x.--. ,m, -,-" - ; m. ".*..-'.. i;-:-*i" r
|^;q;,:'vif. ;•/;• ,_^MV X^;. =v» I
109
-------
Appendix B: Emissions Modeling
Tabte 38, - Ctafcol Sefeawia Personal
"• < •. * .V
Ctowaistfe*^ > ' V '•" ~ -
U«i* track
110
-------
Appendix B: Emissions Modeling
2
3
•4
5
6
7
S
9
10
11
12
13
14
IS
16
17
It
19
TO
22
23
24
2}
20
27
2i
29
30
31
32
33
No-control Scenario Emissions
The control scenario data were modified to reflect no-control scenario emissions using economic
changes predicted by the J/W model, EPA, and ANL. The J/W model predicted a sUghtllls of
employment and drop in GNP in terms of nominal dollars. However, me lower rate oflnflation
coincided with a real GNP rise. ANL's information from the model did not |ncln|te InV indexes for
converting nominal income to real income. ANL assumed real income changes |^|e similar to those of
real GNP and modified household shares by income classes accordingly. Tb«;modelia1io predicted a
slight drop in refined petroleum price beginning in 1973. Thttipredicted drop was the largjast(5.35 jf^
percent) in 1973, reached the lowest level (2.16 percent) in 1984, men-increased to a sec<^]p(ia|;(3.44
percent) in 1988,. and dropped again from 1989 to 1990. Since these changes were inconsistent with
historical patterns of leaded and unleaded gasoline price change, ANL developed an estimate of changes
in fuel price resulting from the cost of removal of lead from gasoline and other infrastructure costs
involved with distributing a new grade of fuel. Subsequently, EPA provided a set of fuel costs for use in
the analysis. Both ANL and EPA fuel prices followed a similar pattern, although their magnitudes
differed. The no-control scenario was analyzed with EPA fuel prices. ANL also established a
relationship with cost of regulation/emission control technology, and the effect of costs on vehicle price
and fuel economy directly from the EPA publication Cost of Clean Environment (EPA, 1990). These
changes were used in the analysis. „" " '_ _ ] ":F
The IFF model was executed for -~~
target years 1975,1980,1985, and 1990
using a set of revised household shares
by income clajs. Table 39 shcjws the
revised shams.; €omparing^aile 3i|
no-control scenario;
Table 36 fortiie control i
seems to be a slight shift away frcml|
travel by the lowest iiji
toward the middle i
The> vehicle ownershi]
model was executed for thipHii four
target years using the data listed in Table
4Q. Changes in fleet characteristics are
summarized hi Table 41.
floaachoM Shared), by Year
im im m$ tm
*»«/
13.6
;»a,
•^j&&
-*wi
•^ &*
25 J
14,3
Ill
-------
Appendix B: Emissions Modeling
Disposable
FaeUTfce ' Mltetf
YMT/Vtfrfele
197*:''
1973 <
J974,',
7,998*
8,463
: las
, , •'
,«M
IS 4
J»J - / ^ r
"" "
1978
, \ o " ;,
*'\
1980
1981
V-" •*
I9M-
1983
X»V^;
1981 /.>5- ^:|»^W>>^ xui.,,
> •.* v4.s^-'%AV ^\ •>••'•
A;> •&.' S y-h^ •* x^% ' VC \ s * *''•
j9kk;^,K;^^'^:j0ftv
- •* *~ "s» i$ff^ ^'•' tt ,'"•• ' , S'^'' !•
-J913- , ^' ',;\«- ;^^-r;;v^' x\ '^Jg?^
- v, >%;irsi -j -^ 154,;,^^ '
fcMhVv *'%
-------
Appendix B: Emissions Modeling
towage* iDK^y^cfeCIwtwrtwwdcs $«iwce8$tt
^^i^ittrfKd^aMtte^^eBfflifts, - -- , /„- ^\ -
Vehkfe
HP
•f,
<.
* +
HP
113
-------
Appendix B: Emissions Modeling
29
Utilities
i •
i ' •
2 The electric utility industry retrospective analysis was prepared using two different utility simulation
3 models. ICF utilized its CEUM to estimate control and no-control scenario emissions for SOj, PM, and
4 NOS in each of the target years. ANL's ARGUS model was used to estmiate^iciic^utility CO and VOC
5 emissions for the same period. This mix of modeling approachesnps used'licin^^tile CEUM was
6 determined to be a better tool for examining fuel shifts mat were affected by|ii ^irflian ARGUS, the
7 CEUM model was not initially set-up to evaluate CO or V(X^;^issions.yjymou]^ll^Mcan be (and
a eventually was) configured to provide emission estimates for pollutants other than SO^KO,, and FM,
9 ARGUS was already configured to provide VOC and CO emissiom However, it should also be noted
10 that VOC and CO emissions from utilities are quite low, as efficient fuel combustion reduces both
n pollutants. Thus, for this sector, the presence or absence of the C!A% would not produce any different
n VOC or CO control techniques. VOC and CO emission rates for this sector differ primarily based on the
13 fuel and boiler type. Therefore, a simpler modeling approach was judged to bje acceptable and
14 appropriate for these two pollutants. This chapter presents the methodotogy used to estimate utility
is emissions under the control and no-control scenario usmg me^^UM and ARGUS models. The method
16 used by Abt Associates to estimate lead emissions from utilities is also presented.
,:-^ .---' ''
.' ;4^ ;-4 =
n Overview of Approach 4
is The CEUM model uses industtyjeapachy date and specific unit-by-unit characteristics, operating
19 costs data, electricity demandejftimates undei||i|jSiitipftd no-control scenario, and historical fuel
20 prices to estimate SO* TSP^ aa J|^ emissions|980, 1985, and 1990. Changes in electric utility
u emissions, iposfe* ibri regi^^
22 historical elecMiNijjii^ The ARGUS model, which was used by ANL
23 to estimate utiUty VOG aad CO en^lom^ jt^driven by operating costs, industry capacity and generation
24 data, demand for cold, a^out-levef operating characteristics. The J/W model is used to incorporate
25 predicted changes m^e^cl^iid^and under the no-control scenario. Finally, Abt Associates relied
26 upon energy use data, me^€>pffli||||ta base, and the Interim 1990 Inventory to calculate utility lead
27 emissions based on coal consuiipcm. The approaches used by each of these three contractors are
21 discussed individually in thi following sections.
Establishment of Control Scenario Emissions
x A. common fegpre of me approaches taken by ICF and ANL was to identify conditions that are
31 inputs to ihe CEDM and ARGUS models, respectively, hi the control scenario. Later hi the analysis,
n these variables were revised to reflect no-control scenario conditions. The next section discusses the
33 specific assumptions used in the CEUM analysis.
34 Key Assumptions in the Development of the ICF Analysis
33 At EPA's direction, ICF made several assumptions in conducting mis analysis for purposes of
36 consistency with other ongoing EPA efforts assessing the effects of the CAA. These include the
37 macroeconomic assumptions regarding the effects of the CAA on economic growth, or more specifically,
3i electricity demand, developed from other EPA commissioned efforts. Each is described briefly below.
114
-------
^Appendix B: Emissions Modeling
Pollution Control Equipment Costs • •
i Only limited actual data were available for this analysis on the historical capital and operating costs
3 of pollution control equipment Accordingly, for this analysis, the actual capital and operating costs of
4 scrubbers were estimated using EPA scrubber cost assumptions adjusted to reflect actual data from a
5 survey of scrubbed power plants with scrubbers installed during the 1970s and«arly 1980s. For those
6 power plants with actual survey data, actual capital costs were used. For other pre*1985 scrubbers, ICF
7 relied on the average costs from me survey data. For particulate control equipment (primarily
i electrostatic precipitators, or ESPs), costs were estimated based on limited actual data^ find a 1980
9 Electric Power Research Institute (EPRI) study of ESP and baghouse costs. Based on this information,
10 ESPs were estimated to cost an average of $50 per kilowatt (in 1991$), 3Tie development of mate
n detailed data on actual power plant pollution control costs was beyond the scope of ICFs analysis. ICF
12 concluded that such an effort would not significantly change the national or regional cost estimates
u developed by its approach. , r:>
u Electricity Demand end Fuel Prices ''
u Consistent with other EPA ongoing analyses, ICF assumed mat the CAA resulted in a reduction in
is electricity demand of 3.27 percent hi 1980,2.77percent in 1985, and 2.97 percent in 1990. Also
17 consistent with these studies, ICF assumed that natural gas prices and oil prices would not be affected by
is the CAA. CoaJ prices were estimated to change in line with increases and decreases in demand for
19 specific coal supplies (and consistent wiJfelCFs detailed modeling of coal supply and demand). The
20 average prices of all residual oils consumed were also estimated to change due to a greater use of more
expensive lower sulfur residual oilsunder the GAIL. '*:
Coed, Nuclear, Hydro,
23 At EPA's direction, ICFs app|||d|was based on the assumption that no changes in the amount of
24 nuclear, coal, hy^ro^bil/gas strel%o|||^bined cycle capacity would be built or in place in- 198Q,
25 1985, or 1990. Given tbitiEhf drivingisctdtrs associated with the actual decisions to build new baseload
2t capacity werenbt base4so3p^;Gp economics but entailed financial, regulatory, and political factors as
27 well, the actual effect of Sjpllfiliojj these build decisions is very uncertain. To the extent that more
a coal-fixed power plants w<|it||ftiu]it and fewer oil/gas-fired power plants constructed, the actual
29 emission reductions associflpMiih the CAA would be greater than those estimated by ICF, while the
x estimated costs of the CAA would be greater (because fewer, lower-cost, coal-fired power plants would
31 be on line under the CAA) However, the CAA had virtually no effect on the costs of constructing new
n coal-fired power plants that came on line prior to about 1975 and a relatively moderate cost effect on
33 coal-fired power pjajnts mat came on line through the early 1980s (since these power plants were not .
34 required to install icrubbers). Since a large majority of coal-fired power plant capacity came on line
is prior to 1975, ICF concluded that the effect of the CAA on the amount of total coal-fired capacity was
36 not expected to be very large.
37 Natural Gas Consumption
u The analysis assumed that the amount of natural gas consumed under the no-control scenario could
w not exceed the actual amount of consumption hi 1980,1985, and 1990. In part, because of natural gas
price regulation and the oil price shocks of the 1970s, natural gas was often unavailable to electric
4i utilities in the early 1980s. Since the CAA is relatively unrelated to the questions of supply availability
i
115
-------
Appendix B: Emissions Modeling
i and price regulation of natural gas, ICF assumed that no additional gas supplies would be available if the -
2 CAA had never been adopted. It is possible, however, that in the absence of the CAA, industrial and
3 commercial users of natural gas would have used more oil or coal. To the extent that this would have
4 occurred, there would have been more natural gas supplies available to the electric utility sector. This
5 increase hi supply would have resulted hi an increase hi the estimated costs of the CAA, and a
6 corresponding decrease in the estimated emission reductions. ICF concluded, however, that this effect
7 would not be very significant :
_-_--" " j?s— ' -' -l' - ' '
s State and Local Environmental Regulations
9 At EPA's direction, ICF assumed that there would be no State and local emission limits or other
w emission control requirements under the no-control scenario. Accordingly, ICF assumed that there
11 would be no SO* NOV or TSP emission limits under the no-control scenario and mat all scrubbers, NOX
n controls, and ESPs/baghouses (at coal-fired power plants) were installed as a result of the CAA. (The
n more limited amount of particulate control equipment installed at oil-fired plants was assumed to have
14 been installed prior to the passage of the CAA.) In the case of particulate control equipment, some ESPs
is and other equipment were installed at coal plants prior to the 1970 CAA. To the extent that this is the
16 case, the estimates of the costs of meeting the CAA have been overstated. ICF concluded, however, that
i? the amount of such capacity was not substantial, • : %.
' „ 5- . "" -_ "* ~-.ff" if
n Retirement Age AT ;:V:3iF
Vf~ -- '•- ' ••
19 The analysis assumed that unit retirement age was constant between the control and no-controls
20 scenarios. Adoption of this assumption might bias the emission reduction estimates upward to the extent
21 tirniover rates of older (and presumably Wg^
22 scenarios, because more significant CAA contiol requirements focused on new units. However the vast
23 majority of existing coal and oil capacity was built after 1950 and h is generally acknowledged mat a
24 relatively short technical plant lifeti^f^ld be about 40 years. As such, even if the no-control
23 scenarios resulted in aoldfe-extensiofi actiyi|y, there would be virtually no effect over the 1970 to 1990
x timeframe o ^i
-. -ifj&:
27 ICF 1975 Control ScenO^t Emissions
„!>„' S
a The 1975 emissions uodbnboth scenarios were calculated differently than emissions in 1980,1985,
29 and 1990. In calculating or estimating 1975 SO2 emissions for the control scenario (i.e., "actual" 1975),
so the weighted average emission rates at the State level, hi the year 1975 were estimated, based on plant
31 level average sulfur content of fuel deliveries from Federal Energy Regulatory Commission (FERC)
32 Form 423 and assumed AP-42 sulfur retention in ash. These weighted average emission rates were then
33 applied to actual State-level electric utility fuel consumption hi the year 1975 (DOE, 1991). In the case
34 of NO, emissions, first, an estimate of Statewide NO, emissions hi the year 1975 was derived based on
35 the use of the same NO, emission rates, by fuel type, as developed for the 1980 no-control scenario
36 . modeling runs. These emission rates were specific to the fuel type (coal, oil, or natural gas). These
37 Statewide NO, emission rates or factors were then applied to actual fuel consumed by electric utilities hi
33 the year 1975, hi order to obtain estimated "actual" 1975 emissions. As before, the fuel consumption at a
39 State level was derived from the State Energy Data Report (DOE, 1991). ICF calculated the weighted
40 average heat content (BTU/lb) by State from the 1975 FERC Form 423 data and used these figures with
the TSP emission factors (Ibs/ton) to derive emission rates by State (Ibs/MMBTU). These emission rates
116
-------
Appendix B: Emissions Modeling
7
a
9
10
II
12
13
14
13
16
17
IS.
20
21
22
23
24
26
27
21
29
30
31
a
33
34
35
36
37
39
were men applied to 1975 fuel consumption estimates obtained from the State Energy Data Report. For
the control scenario 1975 estimates, ICF used the 1975 factors.
For the remaining target years, ICF used the results of CEUM runs mat provided fuel consumption
figures in 1980,1985, and 1990, respectively. Emissions were then calculated using the appropriate
emission factors for each year.
ARGUS Modeling Assumptions
The portion of the electric utility sector analysis conducted by ANL witii the.
described in this subsection. ARGUS contains four major coniponents; JBUILD, DIS
Emissions and Cost Model, and the Coal Supply and Transportafiffli^fbdel (CSTM). An ovepiew of
ARGUS can be found in Veselka et al (1990). Only the DISPATCH and CSTM modules were used for
the present analysis. A brief description of the ARGUS cxrapojieiitf iufc^m this analysis is found in the
following subsections. , -^3|iSh_,
DISPATCH Module - \^ ffifc^
The DISPATCH module contains a probabilistic production-^^ modjel caUed the Investigation of
Costs and Reliability in Utility Systems (ICAjtUS). This moa^J|l|i&s reliability and cost
information for a utility system. ICARUS represents detailed, uiut-^-unit operating characteristics such
as fuel cost, forced outage rate, scheduj^|aamten|pce, heat rateC and fixed and variable operating and
maintenance (O&M) costs. These comjicments are used to efficiently compute system reliability (such as
loss-of-load probability and unserved energy) and production costs,
The input data i^uirecyif|^p|US mcludpiionthly load duration curves, annual peak demands,
and, for botfr aew and exisrli^jilfeSpit sizes, capital costs, fixed and variable O&M costs, fuel types
and costs,
ICARUS inc
system.
and equivalent forced outage rates. The output from
, generation, cost, and reliability for the entire generating
die least-cost combination, on a per BTU basis, of coal supply
t for each demand source. First, it estimates coal market prices based on
regional demands for coj|%om all economic sectors. To generate market prices, CSTM estimates
regional coal productijf patterns and coal transportation routes. The CSTM input data are grouped into
threejnajor oategfp? demand, supply, and transportation. CSTM uses supply curves from the
and Mine Costing (RAMC) Model (DOE, 1982). Every region has a separate curve
for one orinore of the 60 different coal types mat may be produced in that region. CSTM modifies the
original RAMC supply curve' by dividing the single RAMC curve into two curves, one representing deep
mines and the other representing surface mines, but still uses the same ranges for heating values and
mine prices that define me supply curves in RAMC. Prices fluctuate as a result of different mining
methods, size of mining operations, reserve characteristics, and depletion effects.
The transportation data defines the network that connects 32 coal supply origins with 48 demand
centers. Transportation cost is affected by distance, terrain, congestion, variable fuel costs, cost
escalators for fuels and facility upgrades, and competition. CSTM first computes the production cost for
117
-------
. ; . Appendix B: Emissions Modeling
i each coal supply region and coal type. It then matches supply sources with transportation routes to find
2 the lowest delivered costs.
3 Coal demand for a particular region is based on die amount, geographic region, economic sector, and
4 range of coal types. There are 44 domestic demand regions. CSTM allows demand to be met by one, or
5 a combination of, different supply regions. .^ . ••
6 The ARGUS input data for existing units are based on the Argonne Power Plant Inventory (APPI).
7 APPI is a data base of operating and planned generating units in the United States that was current
s through 1988 at the time of ANL's analysis. This data base is updated annually based on information in
9 the regional North American Electric Reliability Council (NERC) reports, reports from the Energy
w Information Administration (EIA), and other sources. Unit operating characteristics (fixedQ&M,
n variable O&M, heat rate, forced outage rate, and scheduled maintenance) are based on regional data as
n defined in the EPRI report on regional systems and other historic data (EPRI, 1981).
13 ANL used the 1988 inventory to generate a 1990 inventory. The 1990 inventory was then used to
14 generate a separate unit inventory for the target years 1975,1980 and 1985. "The target year inventories
is were generated by removing units whose on-line year was greater than the target year, from their
16 respective inventory. The regional capacity totalsiin theseiprelirainary inventories were tabulated by
n major fuel category (nuclear, coal, oil and gas steam) and compared to tite regional historic NERC totals.
n This review identified capacity differences* especiallyjn 1975 and 1980 inventories. The original plan
19 was to add phantom units to match the regional historic totals. Jfowever, based on the need for State-
20 level emissions, it was decided that a more thorough review of the unit inventories was required.
• ^f -if. , -„/ •
- -~_-*r5" • -sT!:s •="" jUs* , "-""i
21 ANL's detailed review included in examiniti^ of tiK&iclear and coal units greater man 100
22 megawatt equivalent (MWe|mff||target year/ Blgphg units, with the appropriate unit size and State
a code, were added so mat mej«|Jo|uRtltotals were comparable. The availability of coal units was based on
24 the on-line year of the unit as repfflftedlfi the EIA report Inventory of Power Plants in the United States
25 (IX)E, 1986). TtacoalA^
26 verify the existence of flue gaisdesulfuri^bn^GD)systenism each of the target years. The nuclear
27 unit inventories were verified with the EIA report An Analysis of Nuclear Power Plant Operating Costs
21 (DOE, IIP). The review alsoMshided oil and gas steam units greater man 100 MWe. The total
29 capacity of me oU and gas steam units were com
x oil to gas during the relevant time period. The oil and gas units were compared to historic inventories
31 based on information provided by Applied Economic Research. In addition to thermal generation, the
32 hydro and exchange energy was reviewed. For each target year, the hydro generation and firm purchase
33 and sate capacity data was adjusted to reflect the historic levels. These two components, hydro and firm
34 purchase and satayue accounted for first in the loading order. If these variables are overestimated,
is there will be less generation from coal units. Likewise, if they are underestimated, there will be too
36 much coal generation. The hydro and firm purchases and sales can vary significantly from year to year
37 because of weather conditions and other variables. Therefore, it was important mat they be accurately
x represented.
39 No-control Scenario Emissions
40 In order to calculate utility emissions under the no-control scenario, inputs to both the CEUM and
41 ARGUS models were adjusted to reflect no-control scenario conditions. The changes made to each
42 model's base year input files are discussed separately in the following sections.
t
118
-------
^Appendix B: Emissions Modeling
$
9
ia
n
a
13
14
a
16
n
it
a
to
a
a
24
a
x
27
a
29
X
31
32
33
34
33
36
37
3t
40
ICF Estimates of SOj, TSP, and NO, Emissions in the No-control Scenario
As described on page 116, ICF utilized a different methodology to calculate 1975 emission
estimates. Rather man relying on the use of detailed modeling runs, ICF based the 1975 emission
estimation on historic fuel consumption and sulfur content data in 197S. Thi&subsecticff first outlines
the process used to calculate no-control scenario emissions in 1975 and men pnssents the methods used
for the remaining target years.
1975 Utility SO* NO, and TSP Emissions
To develop State-level no-control scenario utility SO2
SO2 emission rates. A reasonable surrogate for these emi
implementation of the SIPs under the CAA. ICF developed
available for FERC Form 423) and compared these with 197
1975 rates was used in the calculation of SO2 emissions in tfaif
level no-control scenario SO2 emissions, no-control scenario fuel
assumed that the demand for electricity hi 1975
in 1975. This assumption is identical to the ncMXintrotscpari
from the J/W projections. For the purpose of thtianarysuy
demand would have been met hi 1975 from tip oil and coil
consumption of these fuels was assumed taie in the same
energy mix for electricity generation Wtt State. Jfwas assi
ICF developed no-control scenario
rates just prior tOKthe
on the earliest year
State, the greater of 1972 or
To develop State-
were needed. ICF
the actual energy sales
jections derived
increment in
each State. The increase in
their share in the 1975 total
mat the generation of nuclear, gas-
fired, and other electricity generationjpuld not change. A sejdlitivity case without an assumed
electricity demand change was alsq^|llculated.||ihe seiuitijiy analysis results are presented later in this
appendix, starting on page I2lj '
For HI
would have
1975
BTU using me
calc
For 1980,1985, and
figures from the <
h was also assumed that the 1975 electricity sales
the case in 1975. No-control scenario TSP emissions hi •
ibers from EPA that were converted to pounds per million
in each State. No-control scenario TSP emissions were
(Braine, Kohli, and Kim, 1993).
ions
i, ICF calculated no-control scenario emissions based on fuel consumption
is, and 1970 emission factors from EPA.
emission estimates are approximately 10 million tons (or about 38 percent) lower
1 scenario man under the no-control scenario. Most of this estimated difference
resufeiitan the imposition of emission limits at existing power plants through the SIPs under me 1970
CAA. Most of these SIPs were effective by 1980 (with some not fully effective until 1985). Most of the
additional reductions mat occurred during the 1980s were the result of the electric utility NSPS, which
required the installation of 70 to 90 percent SC^ removal control equipment
By contrast, electric utility NO, emission estimates under the control scenario are only about 1.2
million tons, or 14 percent, lower man under the no-control scenario by 1990. This occurs because,
under the implementation of the 1970 CAA, only a few existing power plants were subject to NO,
emission limits. Virtually all of the estimated reductions are the result of NO, NSPS, which generally
119
-------
Appendix B: Emissions Modeling
i
2
3
4
S
6
7
a
9
10
11
12
13
14
IS
16
17
18
19
20
21
32
23
24
23
27
28
29
30
31
32
33
34
3S
required moderate reductions at power plants relative to uncontrolled levels. In addition, electricity
demand is estimated to be about 3 percent lower under the control scenario. This decrease reduces the
utilization of existing power plants and also contributes to lower NOX emissions (add other pollutants as
well).
" ' ' " . -y ''
Electric utility annualized costs (levelized capital, fuel, and O&M) are estimated to be $0.2 billion
lower in 1980, $1.5 billion higher in 1985, and $1.9 billion higher in 1990 un^thecontrol scenario.
Note, however, mat this reflects the effects of two offsetting faofalp (1) mei^iierJ^l^r compliance
costs associated with using lower sulfur fuels, and the increased O&M and cajHta1|coit8 Issociated with
scrubbers and particulate control equipment; and (2) lower utifity generetin|fcostsi(fi^
capital costs) associated with lower electricity demand requirements. la 1980, the mcreaseinilel^osts
due to higher generation requirements (under the no-coi
capital and O&M costs and thus yielded a cost increase over me control case.
However, lower electricity demand for the utility sectoEWould trf
sectors (as electricity substitutes are used). This effect was captured tb|
macroeconomic modeling conducted for the present asilysij|%?
into higher costs in other
by the original J/W
Average levelized U.S. electricity rate estimates are
scenario during the 1980s. Note that year byyear, electric utility^
expenditures (not estimated by ICF)
particularly in the 1970s and early 19
were brought into the rate base.
Significant shifts in regi
no-control scenarios. High
3 percent higher under the control
requirements and capital
Midwest/Central West are
sulfur coal producing regions suep
production.
estimated to have jjicfeased by a greater percentage
incremental capital expenditures for scrubbers and ESPs
fear" 'Spas'
4" 3£ j?S
S4
26 ARGUS No-control Scenario
-
piled to have occurred between the control and
such as Northern Appalachia and the
have lower production under the control scenario, while lower
and Southern Appalachia are estimated to have higher coal
ir
Regional fuel prices, ffegfiiifiermal units, were based on historic information from the EIA Form
423 data for the year 1977,1980 and 1985. The 1977 data was used for 1975. Fixed and variable O&M
costs were adjusted from the 1988 level, and all cost data were converted to 1985 dollars.
V ' , ----,• . .
* •
The load data were based on regional historic NERC data for each of the target years. The shapes of
the monthly load duration curves are the result of modifications based on the data in the EPRI report on
regional systems (EPRI, 1981). The shapes were modified to match the projected 1988 monthly load
factors for the NERC regions. These load shapes were held constant for all years.
The actual peak-loads were selected from historic information and used with the existing load
duration curves. The system was dispatched so mat the calculated generation could be compared with
14 At EPA'» direction, ICFs analysis did not estimate the effect of shifts in noiwitility cod consumptkm <» regkxid cod production, nor
did it consider the possibility that fewer new coal powoplmts might hive been buUt due to the CAA as discussed earlier. Both of these frctors
could result in a greater estimated change in total U.S. coal production than estimated hereto although the differaxx is hot ^
significant
120
-------
Appendix B: Emissions Modeling
2'
3
4
S
6
7
I
9
10
11
12
13
14
IS
16
17
II
20
21
22
23
24
25
26
27
2t
29
30
historic data. Discrepancies were resolved by adjusting the
peak load so that the annual generation was on target This
procedure was repeated for each of the target years.
The electric Anilities were expected to have an increase in
generation as identified by the J/W data. Table 42 identifies
the increase in national level generation by year. The
national level increase in generation was applied to each
power pool. j
In addition to load changes, coal units with FGD
equipment were modified. These units had their FGD IS
equipment removed along with a 3 percent decrease in heat ^
rate, a 2 percentage point decrease in forced Outage rate, and a
SO percent decrease in their fixed and variable O&M costs.f#::'
These changes were incorporated into the ARGUS model for
each of the target years. Model runs were
arrive at estimates of VOC and CO emissions in
no-control scenario.
'-'•*•/;
;?. -•"•/'
/-^•:
ft/maf/on of Load
In order to estimate Pb emi
different sources were used.
the nationatopaluse
The Trends database
identified utility characteristics
control and
Inventory and
for lead and control
calculated^ These
control on coal-burning
control level in 1970.
each of the target years, data from three
'use data and no-control scenarios were obtained from
for theSe!|oir812 analysis by ICF (Braine and Kim, 1993).
factors and control efficiencies, and the Interim 1990 Inventory
bases provided the amount of coal consumed for both the
target years. A correspondence between the Interim
through the plant name variable. Using emission factors
utilities, estimates of lead emissions per plant per year were
from the Trends data base. It was assumed mat pollution
under die no-control scenario would be the same as the pollution
i^me control efficiency from 1970 is used as the basis for the no-control
31
32
33
34
3S
36
37
cei/M sensitivity
~.~ . ~- £ - ™ ™.n:r ^
In addition to comparing actual (control scenario) historical costs and emissions with the higher
electricity demand under the no-control scenario, ICF also evaluated emissions in a sensitivity case
without the C AA or no-control scenario with the same electricity demand (versus the no-control
scenarios with higher demand). The purpose of this sensitivity analysis was to isolate the incremental
electric utility compliance costs and reductions in emissions associated with the CAA from the lower
resulting generation costs and emissions due to lower estimated electricity demand under the CAA. The
incremental effects of the CAA when compared with this case indicate:
121
-------
75
25
Appendix B: Emissions Modeling
V
7 » Estimated reductions in emissions due to the CAA are somewhat lower if measured against the
2 sensitivity case without the CAA with the same electricity demand than the emissions without the
3 CAA with lower demand. This occurs because lower electricity demand under the no-control
* scenario sensitivity results in lower utilization of existing coal and oil plants which, in turn, results in
5 lower emissions. As noted above, in some sense, the changes in emissions represent fib effects of
' " ~jp
6 electric utility compliance actions under the CAA, absent the effect of lower resulltarit demand for
7 electricity. => • _>:
s *• When measured against the sensitivity case without the CAA. (with the same electricity demand),:'.-/
9 electric utility annualized costs are estimated to have increased by aboutSS to $£ biUitim during me
10 1980 to 1990 period. This reflects the following cost factors: (1) higher annualized capital costs
n associated primarily with scrubbers and ESPs installed bj? electrip utilities to comply^rilh me CAA;
12 (2) higher O&M costs associated with the additional air pollution control equipment; and (3) higher
13 fuel costs associated with using lower sulfur coaland oil moiilert& meet the emission limit
u requirements of the CAA.
Commerclal/Reslcientlat
u The Commercial and Residential Simulation System (CRESS) model was developed by ANL as part
17 of the Emissions and Control Costs mtegrated Model Set and usld in me NAPAP assessment (Methods
a for Modeling Futw-e Emissions and Goj^l Costji fa
» (McDonald and South, 1984). CRESS is desigo||to project: emissions for five pollutants: SO,, NO0
20 VOC, TSP, and CO. The CRESS output is tggiijjijt^ Residential and commercial subsectors related
21 to both eccmomic activity and fiiel^ise. The iAttipM|jfl^ material provided in this appendix about
22 CRESS describes the base year asbpng 1985. It appears in this way because CRESS was originally
23 developed to operate using the l|^l!^^ Emission Inventory as its base year data set For the five .
24 pollutants reported by CRESS, emission esiaaates are provided for the following sectors:
• r:-=-*"j:-,fiii'!fe- s'—ls(tilS*r '
26 • ooafJmcludmgpc|li|^Kt categories of anthracite and bituminous boilers;
27 • 'liquid fuel, rncludijgjiii|i!(Fand space heating uses of residuaU distillate, LPG, and other fuels;
u »." naiiu^ gas boilers, spaw heater and mtenial combustion
29 » wood used in boilers and space heaters; and
X * other mixed or unclassified fuel use.
31 *• Residential _.—:£{?~~~
32 » coal, incli^mg area sources of anthracite and bitimiinous;
33 * liquid fuel, composed of distillate and residual oil;
34 • natural gas; and
i5 • wood.
36 *• Miscellaneous
37 • waste disposal, incineration, and open burning; and
3t • other, including forest fires, managed and agricultural burning, structural fires, cutback
39 asphalt paving, and internal combustion engine testing.
122
-------
Appendix B: Emissions Modeling
9
10
11
12
13
14
IS
16
17
18
19
20
21
23
24
23
26
27
28
In addition, VOC emissions are projected for these source categories:
*• Service stations and gasoline marketing;
> Dry-cleaning point and area sources; and
> Other solvents, including architectural surface coating, auto-body re:
consumer/commercial solvent use.
this section describes the use of CRESS to estimate coi
the commercial/residential sector.
Control Scenario Emissions
For the NAPAP assessment, 1985 CRESS output corresponded to ttteiSSfBAPAP Inventory (EPA,
1989), which served as the benchmark for any projections. "Hie design ofOUEISS is such mat emissions
by NAPAP SCC are input for each State, then projei^ed toiutui|^y by scaling them to economic data
such as energy demand. In estimating emissions, differences %iemissipii3cbntrols associated with new,
replacement, and existing equipment are taken into account where,f|i|| differences are considered
significant. The basic modeling approach|s shown in the f -" —*"- ——'---
E=
D*
(this takes into
technology
driver data indi
category bin the base year, or fora subcategoryj subject to controls in year t
in emission rates mat may occur as a result bf emission regulations or
levels in tin base and future years
fraction of total activity in year t differentially affected by emission controls
jv- s-a "
f carried out in two subroutines, one for SO* NO0 TSP and CO, and one for VOC.
Typically SO* NOV TSP, and CO emissions are projected by multiplying the 1985 NAPAP SCC
data or base year data by the ratio of the driver data (activity level) value in the projection year to its
value in the base year. Because there are few controls on SOZ or NO, emissions from the sources
covered by CRESS, projected emissions for most sectors are proportional to the expected activity levels.
Thus,
123
-------
Appendix B: Emissions Modeling
i There are a few source types, such as commercial/institutional boilers, for which emission controls
2 are mandated. These are modeled by multiplying the 1985 emission data by the ratio of the controlled
3 emission factor to the base-year emission factor. Emission factors for each source type are weighted by
4 the proportion of base year activity in each subsector to which controls are expected to apply.
5 • • . ' , .sjilijp"
6 where: -:| /-*' " " . •
7 g = the fraction of base-year activity accounted for by existii^ source b, replacement source r.cff new
i source n in year t . „ *"---_
i " "~°i% *
» The effective emission factor (B,n) for the sector is cateulatedty weighing the portions of sectoral
10 emissions subject to NSPS controls and those likely to continue at existing levels. An appropriate
n Internal Revenue Service-based rate at which new equipment replaces existing sources is applied to each
12 sector in the model. This is done to estimate how emissions might change as older sources are retired
13 and replaced by new sources that emit at lower rates. " -
14 The SCVNO/TSP/CO subroutine varies, in new and replacementeinission-source fractions subject to
is NSPS controls. These fractions are applied to the emission-source replacement rates. In addition, ratios
16 for new source emission factors are vailed by State. However, emission ratios for any pollutant/source
n type combination do not vary over fhe projection period. jf
- i" ^~ ~ *
is The VOC estimation methodology, is similar* but allows variation in emission factors over time.
19 Emission ratios are calculated ihapn||les of replacement and existing source emission factors weighted
20 by me replacementrate for each SCC|OT and new source factors by State. These are input for each 5-year
21 projection interval For most source categories, VOC controls are not envisioned, and the 1985 NAPAP
22 emissions for the qategoiy are simply scaled proportionally to changes in the driver (activity level) data.
21 For sources to wMch«oi]lral| apply, a variation on the following equation is employed:
24 , ~™
23 In equation 6,Jflie emission factors for new and existing sources are effectively weighted by the
26 proportion of total activity in year t to which controls apply.
27 In using CRESS for the CAA retrospective analysis, the base year was 1975. CRESS requires
28 emissions information by State and NAPAP source category as input Since detailed information on
29 emission levels for 1975 by NAPAP source category were not available, the data were developed from a
x combination of sources. The procedure for calculating 1975 emissions based on the 1985 NAPAP
31 inventory is described below. The emissions module uses these initial values in conjunction with activity
32 estimates to project control and no-control scenario emissions.
124
-------
Appendix B: Emissions Modeling
Emissions Data
2 Since the starting point for the analysis was 1975, emissions data by State and SCC for SO* NOD
3 VOC, TSP, and CO were required. Available emissions, information for this year was not at the level of
4 detail needed by CRESS. The 198S NAPAP Inventory, which contains the necessary levefof detail, in
s conjunction with information from EPA's National Air Pollutant Emission Esiimates,3940-1990
6 (Trends) and ANL's MSCET, was used to construct an emissions inventory jjetjjstljfc The model then
7 uses these emissions as a benchmark for the analysis.
« The method for constructing the 1975 emissions data basewas consistent for all pollutants; however,
9 two different sources of emissions data were necessary in order to obtain time series information on all
10 pollutants. MSCET contains monthly State-level emission estimate? from 1975 to 1985 1>y emission
/; source group for SO* NO* and VOC. Therefore, MSCET inibni^o& was used for SO* NO,, and
12 VOC, while Trends data were used for TSP and CO. Emission sip]ii^|^9ups from MSCET were
13 matched with 1985 NAPAP Inventory SCCs. The MSCET methodology is benchmarked to the 1985
u NAPAP Inventory and uses time series information from Trends in conjunction with activity information
a to estimate State-level emissions for SO* NO0 and VOC. Although the level of detail contained in the
16 NAPAP Inventory could not be preserved because of Ae aggregation needed to match with MSCET
n emissions sources, MSCET provided the State-level spatial detail required by CRESS.
is . Once the 1985 emissions by SCC and State from the 1985 NAPAP nventory were matched with
19 emission source groups and States front the MSCET data base, an estimate of 1 975 emissions was
* computed by multiplying the 1985 NAPAP Inventory emissions value by the ratio of 1975 MSCET
emissions to 1985 MSCET emissions/Ratios were computed and applied for each combination of State,
-= ~~ - _-3._.--r~i?— *«•" "H •
22 pollutant, and MSCET emission source group,, _J=?vr>^Ef|ji
23 This method of constructing an emissions inventory for 1975 utilizes the State estimates from
24 MSCET, thus capturing the spatial s(hife^at occurred over the analysis period. It is assumed mat
25 NAPAP provides the most reliable poprtan|.area source information in terms of the level of 1985
26 emissions (which is also |he assumption gf lie MSCET methodology). Note mat if mere were a 1 -to-1
27 correspondence betweenMS^CET and NSPAP, mis method would be equivalent to using the MSCET
28 methodology directly for
-------
Appendix B: Emissions Modeling
i
2
3
4
5
6
7
3
9
10
11
12
13
14
IS
IS
17
IS
19
20
21
22
23
24
25
26
27
a
29
30
31
32
33
34
35
36
37
31
39
40
41
42
43
' •£_ •.•.'< .•> s" % x>: *<*•£*#&*•>'** ' <•
- -*B*-~-. fr.-..^ .• MfMM^J^.. --A^^ -> ^
+ .'*'*^ ••?.„,. ''''^' ' '.,' '''/
, , ,<-*
CAA regulation of
commercial/residential emissions was limited
and largely confined to fuel combustion
sources (SO* NO,, TSP), gasoline marketing
(VOC), dry cleaning (VOC), and surface
coating (VOC). NSPS regulations of small
(over 29 MW capacity) fuel combustors were
promulgated in 1984 and 1986. For purposes
of emissions calculations, the stipulated
NSPS for SO* NO« and TSP were
incorporated into the control scenario for
1985 and 1990. Emission rates for source
categories subject to VOC regulation were
similarly adjusted.
Energy Data
Nearly 75 percent of the source
categories in CRESS use energy consumption
by State and sector as the driver for the
emissions calculation. State-level
consumption statistics are publi
in State Energy Data Report, Consvjjjption
Estimates, 1960-1989, and are electronically
available asrpart of the S
System (SEDS) (DOE, 1
data base contains annual
consumption estimates
various end-use
commercial, industrial
and electric utilities.
Seven fuel-type categorieslire used in
CRESS: coal, distillate oil^ residual oil,
natural gas, liquid petrclfum gas, wood, and
electricity. The model assumes zero
consumption of residual fuel oil in the
residential sectot^md zero consumption of
wood in the commercial sector. Energy ^^^^^^^^^^^^^^^^^^^^^^^^^^^
consumption for each fuel-type was
expressed in BTUs for purposes of model
calculations. With the exception of wood consumption, all of the energy consumption statistics used in
CRESS were obtained from SEDS.
Residential wood consumption estimates were derived from two data sources. State-level residential
sector wood consumption estimates for 1975 and 1980 were obtained from Estimates of U.S. Wood
Energy Consumption from 1949 to 1981 (EIA, 1982). State-level wood consumption, however, was not
-* - \r?.^/v*. -^1^12
u**iA j'Mlwtr ttttfmm* - Ijgft ,' •• |J%
lJN..I^lll|. K IlllPr- 111 INUIIIIJ^^ ^ TV1"""--, •• -.^^ K «-»^^-
''",}'''>>' 4 <• ' ''' "•*$•. , -
^ISFiew^wto
t^^
126
-------
Appendix B: Emissions Modeling
available for 198S and 1990, therefore, regional information from an alternative publication, Estimates of
2 U.S. Eiofuels Consumption 1990 (EIA, 1990), was used to derive State-level residential wood use
3 figures. Regional 1985 and 1990 wood consumption was distributed among States using 1981 State
4 shares. All wood consumption figures were converted to BTLTs using an average value of 172 million
s BTU per short ton. f<
- L" ~.=
6 Economic/Demographic Data ___ ': • ? :
- -==%."" "" "^"-^^f^i^ - '
7 Emissions from slightly more than 25 percent of the CRESS source categories follow State-level
< economic and demographic activity variables. The demographic variables used by CRESS include State- .
9 level population, rural population, and forest acreage. State population is the activity indicator for six
10 emissions source categories for SO* NO,, TSP, and CO, andlSVQG source categories. State
11 population data were assembled from the SEDS data base. Rural population, which is the indicator of
12 residential open burning activity, is computed as a fraction of total State population. Forest wildfires and
n managed open burning activity are related to 1977 State-level forest acreage. The demographic
u information is assumed to be invariant to CAA regulations and thus is the same in the control and
is no-control scenario scenarios.
16 . • • _. -"„-;.- '
17 Car stock (or vehicle population), the driver variable for the auto body jrefinishing, is approximated
is by State motor vehicle registrations. Highway Statistics* an annual publication by the FHWA, was the
19 source for data on State motor vehicle registrations. The three source categories connected with gasoline
20 marketing are driven by State-level gasoline sales in gallons. Sjpte gasoline consumption was obtained
v from the SEDS data base. Housing starts and 10 percent of the^existing housing stock were combined to
form the activity indicator for architdlural surface coating emissions. Housing data compiled by the
23 U.S. Bureau of the Census werejavailable hi ^Statistical Abstract of the United States (DOC, 1975;
24 1977; 1982; 1983; 1987; 1993). Regional- level data for 1975 was allocated to the States based on the
2s 1980 State distribution. ^-1^%^ - '
' "
26
31
-- k
No-control Scenario Emission
27 Adjustments to contnl«|iripyemissions in each of the target years to reflect conditions under the
x no-control scenario were aohle||i3 through emission factors, energy input data, and
29 economic/demographic date; The adjustments made to each of these variables to generate no-control
jo scenario emissions are discussed individually in the following subsections.
Emissions Data
32 CAA regulation of the commercial/residential sector was minimal. For regulated source categories,
33 emission factors were revised to reflect pre-regulation emission rates. Six commercial/residential source
34 categories were regulated for VOC emissions: Service Stations Stage I Emissions, Service Stations
15 Stage n Emissions, Dry Cleaning (perchloroethylene), Gasoline Marketed, Dry Cleaning (solvent), and
36 Cutback Asphalt Paving. Commercial-Institutional boilers were regulated for SO2 and TSP and internal
37 combustion sources were regulated for NO, emissions. All NSPS were removed for these sources to
39 estimate no-control scenario emissions levels.
127
-------
Appendix B: Emissions Modeling
2
3
' 4
3
6
7
I
9
10
n
n
13
14
IS
16
17
II
19
20
a
22
23
24
25
26
27
2S
,29
30
31
32
33
34
3S
36
37
38
39
40
41
42
*" ^ f
iitttsc&Q]ii$
Energy Data
' *
State-level energy demand for the residential and commercial sectors for the no-control scenario was
estimated from the J/W model forecast Final energy demand estimates for the household jector were
calculated by an EPA contractor for the purposes of the no-control scenario analysis. Sta|i%llocationof
the national-level estimates was based on historic State shares, i.e., this assumgi that there is no change
in the distribution of energy demand across States as a result of removing regiSt^f In addition, the
J/W model estimates an
aggregate refined petroleum
category and does not
distinguish among liquid
petroleum gas, distillate oil, and
residual oil. The relative shares
among these three categories of
petroleum products remained
constant between the control
and no-control scenarios. The
information on percentage
change in energy demand by
fuel type as provided by the J/W
model is listed .hi Table 44.
ag&N MT
j-
was calculated from the combination of
Finance, Insurance,
Other Services; and
Government
~ •
ces.
The differential for commercial sector final
four intermediate product flow categories from&gLJ/W forecast The National Income and Product
Accounts (NIPA) for the commercial sector correspond to J/W SIC categories 32 through 35:
(32)
(33)
(34)
(35)
'_ _- -. - -, - " "- ,-- -^w-jEj I§~ji
Percentage change hi|Bipi^l§,from the J/W forecast for energy cost shares, value of output, and
energy prices was used to caJ^iJPme differential in commercial sector energy demand for the
no-control scenario. The ejprgjrcost share is defined as the cost of energy input divided by the value of
the output In order to calculate the percentage change hi commercial sector energy demand, the change
hi energy price was subtracted from the percentage change hi energy cost, and added to the change in the
value of output Each of these variables was available from the J/W model results. This calculation was
performed for eachpf the four energy types, and each of the four NIPA categories. The change in
cx>nimercial seetolr energy demand was obtained by taking the weighted average of the four NIPA
categories. Since data on relative energy demand for NIPA categories were not readily available, square
footage was used as a proxy for calculating the weights. These data were taken from the Nonresidential
Buildings Energy Consumption Survey, Commercial Buildings Consumption and Expenditure 1986 (EIA,
1989). The resulting estimate for commercial sector changes hi energy demand is provided hi Table 45.
The national-level change in commercial sector energy demand was allocated to the States using
historic shares. Implicit is the assumption that removal of CAA regulations does not alter the State
distribution of energy use.
128
-------
Appendix B: Emissions Modeling
2
3
4
5
6
7
I
9
10
11
12
13
14
IS
16
17
IS
19
20
aw
Economic/Demographic Data
State population was
assumed not to vary as a result
of CAA regulations, thus only
the economic variables were
revised for the no-control
scenario. No-control scenario
housing starts and car stock
were derived from J/W forecast
information on construction and
motor vehicles. The differential ^^^^V__^
for categories 6 (construction) •mllil^^^^^m
and 24 (motor vehicles and
equipment) was applied to control scenario values to
obtain no-control scenario levels. The percentage
change from the J/W forecast is given in Table 46.
State-level gasoline sales is one of the activities
forecasted by the transportation sector model. The ,_
percentage change hi gasoline sales calculated by the
TEEMS model was used in the no-control scenario as
a CRESS model input ^ =' ;S
ft&rtawl
•%.
,s*
Table 46. JWPere«ttDiffe«Qti»i in Economic
Varables Used m CRESS,
Motor
Vehtete
129
-------
Appendix B: Emissions Modeling
^^^s^^^^w*^
Seeto-
TOTAL*
i
-m
• , aupK 2410
,- - s -- -* ^^ %s* £<0 J^ V v*' ^
-. _. *" v >, /#^\<* ^^•^^•f* '* * f X-".
**«*
1%
US
^ > f
•"v •. % -S.^ ^ •• *y*¥^*- •'''•• < ^f''' '•v?:' **•"• •" s '•
^*2^9KIA •* '• jlJMQ'
55jjjjp:.. ^ylffifl
•" >^%^ '' W&
'•Juitmrfo
lt!y«towM«^
todn»l^»a(fi«k)wiB Tte^firrotiOTrtp^i»l»te(wp»^
130
-------
Appendix B: Emissions Modeling
HW -Bit
4,090
«» KO
, 2,060 1*
*
830
22,060
890 956 <2«)
31.MO
Table m
lJ
^^1«** " """ . ";^' iliiii>ti»%" ^J^ ^r
am
IndufCiM Bctitof
1WAL*
v ' .. , - ,
131
-------
tons).
Ytfckk*
Appendix B: Emissions Modeling
\ , •• •"• " H <•> ,' ^ ' ' ',$?"?'• ,{
- - ,-. -.. y'"- < ~i£&#fM"
?., J. " % --V ' - ^ •.-"-. ->Vv
>:I- IJlMteCtef
7,5»
J8» , ' "S^' v 'f?8<
*,*» , 5,140
^->»-5«|ft^v ^Itsr'^K.m/'' vf*
, *% J v' •• '0%S^ ^V ^T- ^i'%*-'<-^X
^» ' , " -vj'^f* -*< "',''" \?<" "XJ '/v / ^
•• - '', - '^ -^*'$?",- \K^ ^s, *' 'vi'.
, -4. «^%^*> x^x«xf^««jo »^5(f . » •> *
9
<-v'-^
• .- v-.-X-
iitiWiKllpflft'lffil'infltlljf ffrt"TJiii riflflTlftrtr ni ti Irtufjftift tfctfiflj|[
132
-------
Appendix B: Emissions Modeling
Emissions Modeling References
2 Abt, 1995: Abt Associates Inc., "the Impact of the Clean Air Act on Lead Pollution; Emissions
3 Reductions, Health Effects, and Economic Benefits from 1970 to 1990," Final Report, Bethesda,
/ MD, October 1995. .S^ ^s-^v- .
/;-: j-'-VV'd^""'':. - '"
s ANL, 1990: Argonne National Laboratory, "Current Emission Trends for Itftrogen Oxides^ Sulfur ./•"-
6 Dioxide, and Volatile Organic Chemicals by Month and State: Methodology and Results^*l"Argonne,
7 IL, August 1990. "*-:17- " - .; .".-.-"."" .
t ANL, 1992: Argonne National Laboratory, "Retrospective Clean Air Act Analysis: Sectoral Impact on
9 Emissions from 1975 to 1990," (Draft), Argonne, EL, July 1992,- ' ; : • - *,.
'- i~J -f; *•;•,!,
jo Braine, Bruce, and P. Kim, "Fuel Consumption and Emission Estimates by State," ICF Resources, Inc.,
;; . Fairfax, VA, memorandum to Jim DeMocker, EPA, April 21, 1993.
n Braine, Kohli, and Kim, 1993: Braine, Bruce, S. Kohli, and P, Kfi^lj?75 Emission Estimates with and
13 without the Clean Air Act," ICF Resources, Inc., Fairfax, VA, memorandum to Jim DeMocker, EPA,
u April 15, 1993. __.,;.
DOC, 1975: U.S. Department of Commerce, Bureau of the Census, "Statistical Abstract of the United
16 States: 1975(96mEdition),nWashmgton,DQ September 1975.
,„_ .-- - •-
n DOC, 1977r U.S. Departmeirtof C^Mmiierce, Bureau of the Census, "Statistical Abstract of the United
is States: 1977 (98th Edition)," Washington, DC, September 1977.
19 DOC, 1981: U.S. Department of Commerce, Bureau of the Census, "1977 Truck Inventory and Use
20 Survey," TC~Tr-T~v^&a,T&, Aagost 19&1.
21 DOC, 1982: U.S. Department <>f Commerce, Bureau of the Census, "Statistical Abstract of the United
22 States: 1982-1983 (103rd Edition)," Washington, DC, December 1982.
4f • •
23 DOC, 1983: U.S. Department of Commerce, Bureau of the Census, "Statistical Abstract of the United
24 States: 1984 (1044 Edition)," Washington, DC, December 1983.
25 DOC> 1984: UiS^Department of Commerce, Bureau of the Census, "1982 Truck Inventory and Use
26 Survey,"TC-82-T-52, Washington, DC, August 1984.
27 DOC, 1987: U.S. Department of Commerce, Bureau of the Census, Statistical Abstract of the United
zs States: 1988 (108th Edition), Washington, DC, December 1987.
29 DOC, 1990: U.S. Department of Commerce, Bureau of the Census, "1987 Truck Inventory and Use
x Survey," TC87-T-52, Washington, DC, August 1990.
a DOC, 1991: U.S. Department of Commerce, "Annual Survey of Manufactures: Purchased Fuels and
32 Electric Energy Used for Heat and Power by Industry Group," M87(AS>1, Washington, DC, 1991.
133
-------
21
Appendix B: Emissions Modeling
\
i- DOC, 1993: U.S. Department of Commerce, Bureau of the Census, "Statistical Abstract of the United
2 States: 1993 (113th Edition)," Washington, DC, 1993.
3 DOE, 1982: U.S. Department of Energy, Energy Information Administration, "Documentation of the
4 Resource Allocation and Mine Costing (RAMC) Model," DOE/NBB-0200,1982. ^
-*
5 DOE, 1986: U.S. Department of Energy, Energy Information Administration Tnventory of Power
6 Plants in the United States 1985," DOE/EIA-0095(85), Washington, DC, August 1986.
7 DOE, 1988: U.S. Department of Energy, "An Analysis of Nuclear Power Plant Operating Costs," Energy
s Information Administration, DOE/EIA-0511(88), 1988.% =- ? L
9 DOE, 1990: U.S. Department of Energy, Energy Information Administration, "State Energy Price and
10 ExpenditureReport 1988," DOE/EIA-0376<88), Washington, DC, September 1990.
-= .% r -
u DOE, 1991: U.S. Department of Energy, Energy Information Administration,?State Energy Data
n Report: Consumption Estimates -1960-1989," DOE/EIA-0214(89)," Washington, DC, May 1991.
• • ,."....-'.... ""•"=*'
u DOE, 1992: U.S. Department of Energy, Energy Information Administration, "Annual Energy Review
u 1991," DOE/EIA-0384(91), Washington, DC, 1992.
a EIA, 1982: Energy Information AdminMration, "Estimates of His. Wood Energy Consumption from
u 1949 to 1981," U.S. Department of Energy, D0E/EIA-0341, August 1982.
"
77 EIA, 1985: Energy InfonnatiOTAministratio|^
» U.S. Department of Energy, DOE^EIA-0091(85X 1985.
- iL-v:"^ . '.-s^SSrKiii^t. „ :;"-„*•"- • -'•'-. • .
19 EIA, 1989: Energy B^rmation A|pfi|i||tration, "Nonresidential Buildings Energy Consumption Survey:
20 CommercMIMJiings Consumption and Expenditures 1986," U.S. Department of Energy,
22 EIA, 1990: Energy Information Administration, "Estimates of U.S. Biofuels Consumption 1990," U.S.
23 Department of Energy,i|||||pd\-0548(90), October 1990.
24 EPA, 1985: U.S. Environmental Protection Agency, "Compilation of Air Pollutant Emission Factors,
25 .--.Vohmel: Stationary Point and Area Sources," AP-42, Fourth Edition, GPO No. 055-000-00251-7,
26 Research TrianglaPark, NC, September 1985. .
27 EPA, 1989; UjS.iEnvironmental Protection Agency, "The 1985 NAPAP Emissions Inventory," EPA-
21 600/7?89-012a, Research Triangle Park, NC, November 1989.
29 EPA, 1990: U.S. Environmental Protection Agency, "The Cost of a Clean Environment," EPA-230-11 -
30 90-083, November 1990.
31 EPA, 1991: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
32 "National Air Pollutant Emissions Estimates, 1940-1990," EPA-450/4-91-026, Research Triangle
33 Park, NC, November 1991.
134
-------
^Appendix B: Emissions Modeling
EPA, 1992: U.S. Environmental Protection Agency, "1990 Toxics Release Inventory," EPA-700-S-92-
2 002, Washington, DC, 1992. '
3 EPA, 1994a: U.S. Environmental Protection Agency, "National Air Pollutant Emission Trends, 1900-
4- 1993," EPA-454/R-94-027, Office of Air Quality Planning and Standards, Research Triangle Park,
, NC, October 1994.
/
6 EPA, 1994b: U.S. Environmental Protection Agency, Office of Mobile Guide to
7 MOBILES (Mobile Source Emission Factor Model)," EPA-AA-AQAB-94-01, Ann Arbor, MI, May
• ' 1994. . '. • -:*"
L£j -j _- ^
9 EPRI, 1981: Electric Power Research Institute, "The EPRI RepeiinaliSystems," EPRI-P-1950^ Palo
Alto, CA, 1981. ' • v
. '
n FHWA, 1986: Federal Highway Administration, U.S. Dep^rtoentolpailpqrtation, "1983-1984
n Nationwide Personal Transportation Survey," Washington, DC,
n FHWA, 1988: Federal Highway Administration, U.S. J>epartment of Transportation, "Highway
M Statistics 1987," PB89-127369, Washington, DC, 19fc^fo||^ ^'
u . FHWA, 1992: Federal Highway Administration, U.S. Department of Transportation, "Highway
M Statistics 1991," FHWA-PI^92-02Jg(yiingtom, DC, 1990? .
- ~£_--~ iff r'feT
Gschwandtner, 1989: Gschwandtne^ Gerfiaixl, fjpjipcedui^^l^
it Air Pollutant Emissions liendsReport," E^JPechan^ Associates, Inc., Durham, NC, December
n 1989.
20 Hogan, 1988: Hogan, Tim, "Indus^^C^nbustion Emissions Model (Version 6.0) Users Manual," U.S.
21 Envirorimenta]PJp^onAgei2lP400/8-88-007a> 1988.
--,-;:-»». .
22 ICF, 1992: ICF ResouKJes^Isfe "Results of Retrospective Electric Utility Clean Air Act Analysis -
23 1980,1985 and 1990," September 30,1992.
24 Jorgenson and Wilcoxen, 1989: ? Jorgenson, D.W., and P. Wilcoxen, "Environmental Regulation and U.S.
25 Economic Growth," Harvard University Press, Cambridge, MA, 1989.
26 Klinger and Kuzmyak^ 1986: Klinger, D., and JJt Kuzmyak, "Personal Travel in the United States, Vol.
27 , I: 1983-84 Nationwide Personal Transportation Study," U.S. Department of Transportation, Federal
2s HighwayAdministration, Washington, DC, August 1986.
29 Kohout et al., 1990: Kohout, Ed, "Current Emission Trends for Nitrogen Oxides, Sulfur Dioxide, and
x Volatile Organic Compounds by Month and State: Methodology and Results," Argonne National
31 Laboratory, ANL/EAIS/TM-25, Argonne, IL, 1990.
32 Lockhart, 1992: Lockhart, Jim, "Projecting with and without Clean Air Act Emissions for the Section
33 812 Retrospective Analysis: A Methodology Based Upon the Projection System used in me Office
of Air Quality Planning and Standards National Air Pollutant Emission Estimate Reports," (Draft
35 Report), Environmental Law Institute, November 16,1992.
135
-------
Appendix B: Emissions Modeling,
i McDonald and South, 1984: McDonald, J.F., and D.W. South, "The Commercial and Residential Energy
2 Use and Emissions Simulation System (CRESS): • Selection Process, Structure, and Capabilities;"
3 Argonne National Laboratory, ANL/EAIS/TM-12, Argonne, IL, October 1984. _
4 Mintz and Vyas, 1991: Mintz, MM., and AD. Vyas: "Forecast of Transportation Energy Demand
5 through the Year 2010," Argonne National l^aboratoiy,ANL/ESD-9,Argonae,ILj, April 1991.
e Pechan Associates. 1995. The Impact of me Clean Air Act on 1970 to 1990 Emissions-.Section 812
7 R^o,
-------
Appendix Cs Air Quality Modeling
Introduction . % /
3 • • - _,j* . . -
3 This appendix describes in greater detail the various methodologies used to translate differences in
4 control and no-control scenario emission estimates into changes in air quality conditions. Summary ,
s characterizations of the results of the air quality modeling efforts for 1990 are provided here and in the
6 main text Further details and discussion of key analytical and modeling issues can be found in a number
7 of supporting documents. These documents, which provide the analytical basis for the results presented
a herein, are: lll^:Jr
9 »• ICF Kaiser/Systems Applications International, "Retrospective Analysis of Ozone Air Quality in the United
it States", Final Report, May 1995. (Hereafter referred to as "SAI Ozone Report (1995).")
11 •• ICF Kaiser/Systems Applications International, "Retrospective Analysis of Paniculate Matter Air Quality in the
12 United States", Draft Report, September 1992. (Hereafter referred to as "SAI PM Report (1992).")
. > :,-, " ,: ;&$jj^f'"
13 »• ICF Kaiser/Systems Applications International, "Retrospective Analysis of Particvlate Matter Air Quality in the
14 United States", Final Report, AprU 19j^^jpereafle«efeired to ^SAIPM Report (1995).")
» ICF Kaiser/Systems Applications International, >¥& Interpolation Methodology for the §812 Retrospective
u Analysis ", Memorandum from J. Lahgstaff to J.DeModoer, March 1996. (Hereafter referred to as "SAI PM
n Interpolation Memo (1996)^ :: 1°> J.-*%=r:"
it > ICF Kaiser/Systems Applications Intemational, "Retrospective Analysis of SO,, NO, and CO Air Quality in the
19 United States"* Final Report, Noww^ 1994. (Hereanw refen^d to as "SAI SOj, NOX and CO Report
20 (1994).") _ ~ -'."/^Wc. T=r^§4Sp .
- - _ - ; =-—•-•=- f "-
21 > ICF Kaisei^ystems Applications International, "Retrospective Analysis of the Impact of the Clean Air Act on
22 Urban Visibility in the SovOaaaaem United States", Final Report, October 1994. (Hereafter referred to as "SAI
21 SWVisibility Report (19fJ!$$$£~ •'
24 > Dennis, Robin L., US EPA, ORD/NERL, "Estimation of Regional Air Quality and Deposition Changes Under
23 Alternative 812 Emissions Scenarios Predicted by the Regional Acid Deposition Model, RADM", Draft Report,
26 October 1995. (Hereafter referred to as "Dennis, R. RADM Report (1995).*)
„ Ji" -~
\
27 Hie remainder of this appendix describes, for each pollutant or air quality effect of concern, (a) the
a basis for development of the control scenario air quality profiles; (b) the air quality modeling approach
29 used to estimate differences in air quality outcomes for the control and no-control scenario and the
30 application of those results to the derivation of the no-control scenario air quality profiles; O the key
31 assumptions, caveats, analytical issues, and limitations associated with the modeling approach used; and
a (d) a summary characterization of the differences in estimated air quality outcomes for the control and
33 no-control scenarios.
137
-------
Appendix C: Air Quality Modeling
Oar ban Monoxide
\ •
3
4
5.
6
7
8
9
10
11
a
13
14
IS
16
17
IS
19
20
21
22
Control scenario carbon monoxide profit
1-,
As described in the preceding general methodology sectioo^^ startin
elopment of
control scenario air quality profiles was EPA's AIRS database| ^Hourly CO apqol^ nionitoring data
were compiled for all monitors in the 48 contiguous states fbrjhe study target yeare|KfJi9f|, 1975, 1980,
1985, and 1990. Although the CO monitoring network was sparse in 1970, by 1990 506 monitors in 244
counties provided monitoring coverage for 55 percent of the popjiatidn in the conterminous U.S. Table
53 summarizes the CO monitoring data derived from AIRS. AddlHiEial data regarding the EPA Region
location, land use category, location-setting category, and objectfoicittejMy of the monitors providing
these data are described in the SAI SO* NOV and CO Report (1994^11:; i
•*• ">-. t f ~" '• * %
v"-"^ ' ^ rX ' ^ •'> " ', ' '
' •• , /* -C
-.' • •• '.*'*••',,,.•,:• ,•• '•*
Year
1985
SBft'
522
••y."
••*'•
50%
The next step in constructing me control scenario air quality profiles was to calculate moving
averages, for a variety of time periods, of the hourly CO data for each monitor. For CO, moving
averages of 1,3,5, -7,8,12, and 24 hours were calculated. Daily maximum concentrations observed at
each monitor for each of these averaging periods were then calculated. Finally, profiles were developed
to reflect the average and maximum concentrations for each of the seven averaging periods. However,
profiles were only developed for a given monitor when at least 10 percent of its theoretically available
samples were actually available. The purpose of applying this cutoff was to avoid inclusion of monitors
for which available sample sizes were too small to provide a reliable indication of historical air quality.
As discussed in the air quality modeling chapter of the main text, development of representative
distributions for these profiles was then necessary to provide a manageable characterization of air quality
conditions. Initially, two-parameter lognormal distributions were fitted to the profiles based on
substantial evidence that such distributions are appropriate for modeling air quality data. However,
138
-------
Appendix C: Air Quality Modeling
given the relative importance of accurately modeling higher percentile observations (i.e., 90th percentile
2 and higher), a three-parameter modeling approach was used to isolate the effect of observations equal, or
3 very close, to zero. In this approach one parameter defines the proportion of data below a cutoff close
4 to zero and the remaining two parameters describe the distribution of data above the cutoff value.
5 Several other studies have already demonstrated good fit
6 parameter gamma distribution, and bom lognormal and gamma distributions using a three-parameter
7 approach were developed for the present study. As documented in the SAI SO^lfO*; and CO Report
» (1994), a cutoff of 0.05 ppm was applied and bom the mree-panpel^logntKedill^hree-parameter .'.
9 gamma distributions provided a good fit to the empirical data, for CO, the gamma distribution provided
w the best fit ^ r~~ - ' -; ~~*"i
i ~", - .
;; The control scenario air quality profiles are available on diskette!. The filename for the CO Control
n Scenario profile database is COCAA.DAT, and adopts the format presented in Table 54.
13
No-control scenario carbon monoxldo profiles
14 . To derive comparably configured profiles representing (^ air .quality in the no-ntrol scenari
is control scenario profile means and variances were adjusted m proportion to the difference in emissions
i6 ' estimated under the two scenarios. Specifically, for all control sceltiiriipiir quality observations
17. predicted by the three-parameter distributions falling above the "near-zero" cutoff level, comparable
is no-control estimates were derived by the following equation: :
19 where XHCAA ** ate quality measareffieBt for the non-CAA scenario,
20 XCAA -- aff
-------
Appendix C: Air Quality Modeling
**-»
86-95
136-445
*< ,
Real
lest
Rwl
••, ,% ^ ,-• •
'i,
pffig^ .
jftfflffi Bflfffrffifrlfflfftffp^fflf '1
%
' '''''''
^ } ^ -. '%""> ^ :• *>V^J_^-*\
J •• \, •» I' %••* ^ •• ^ v S % •*
•ili)i^M
-------
Appendix C: Air Quality Modeling
t
9
10
11
12
13
14
IS
If
17
18
'9
21
22
23
24
25
26
27
21
29
30
31
32
33
34
35
36
37
what drive the difference in the control and no-control air quality profiles for CO. In other words, the
ENCM t° ECM ratios used to derive the no-control profiles according the Equation (1) above are essentially
based on state-level emissions estimates for CO.
As for the control scenario air quality profiles, the no-control scenario air quality profiles are
available on diskette, the filename for the CO No-control Scenario profile database is CONCAA.DAT.
The same data format described in Table 54 is adopted.
Figure 32. Freqi
control Scenario 95
Concentrations, by Monitor,:
1990 Control to No-
i-Hour Average CO
Summary differences In carbon monoxide afc quality
While the control and no-control scenario
air quality profiles are too extensive to
present in their entirety hi this report, a
summary indication of the difference in
control and no-control scenario CO
concentrations is useful. Figure 32 provides
this summary characterization. Specifically,
the air quality indicator provided is the 95th
percentile observation of 1990 CO
concentrations averaged over a 1-hour period.
The graph shows the number of monitors for
which the ratio of 1990 control to no-control
scenario 95th percentile 1-hour average
concentrations falls within a particular range. I
The x-axis values in the graph represent the
midpoint of each bin. The results indicate
that, by 1990, CO concentrations under:ano-
control scenario would have been
0.05 0.25 0.45 0.0 . O.S5 1.05 1.25
Ratio of CAAJJo-CAA 95th Percentile 1-Hour Avenge
dramatically higher than control scenario concentrations.
Key caveats and uncertainties for cartoon monoxide
A number of important uncertainties should be noted regarding the CO air quality estimates used in
this analysis. First and foremost, CO is a highly localized, "hot spot" pollutant As such, CO monitors
are often located near heavily-used highways and intersections to capture the peak concentrations
associated with mobile sources. Since this analysis relies on state-level aggregate changes in CO
emissions from all sources, the representativeness and accuracy of the predicted CO air quality changes
are uncertain. There is no basis, however, for assuming any systematic bias which would lead to over- or
under-estimation of air quality conditions due to reliance on state-wide emission estimates.
A second source of uncertainty is the extent to which the three-parameter distributions adequately
characterize air quality indicators of concern. Appendix C of the SAI SO* NO,, and CO Report (1994)
presents a number of graphs comparing the fitted versus empirical data for one-hour and 12-hour
averaging periods. 'In the case of CO, the gamma distribution appears to provide a very reasonable fit,
though clearly some uncertainty remains.
141
-------
Appendix C: Air Quality Modeling
i
2
3
4
5
6
7
s
9
w
Finally, as noted in footnote 26, a central premise of this analysis is that changes in CO emissions
should be well-correlated with changes in CO air quality. Strong correlation between the state-level
emissions estimates used in this analysis and empirical air quality measurements would not be expected
due to inconsistencies between the state-level scale of modeled emissions versus the monitor-level scale
of the air quality data, and between the modeled control scenario emissions inventories aip actual
historical air quality measurements. Under these circumstances, it is partkula|||^inip^!rtottofocuson
the primary objective of the current analysis, which is to estimate the differf^ppfjity quality outcomes
between scenarios which assume the absence or presence of histopal air pc9tt|brols. In the
process of taking differences, some of the uncertainties are expected to
in the overall analysis to predict historical air quality, or hyjxjihetical air
Clean Air Act, in absolute terms. "•"
is made
of the
/
Sulfur Dioxide
>Jd=*
12
13
14
16
is
20
21
22
23
24
25
26
27
23
29
30
31
32
Sulfur dioxide (SO}) emissions lead to several air quality effects, i
ondary formation of
fine particle sulfates, long range transport and depodtionof sulluric acid, and localized concentrations of
gaseous sulfur dioxide. The first two effects are addressed later in this appendix, under the particulate
matter and acid deposition sections. The focus of this section is estimation of changes hi local
concentrations of sulfur dioxide. ; :n^:
The methodology applied to estimation of local sulfur dioxide air quality is essentially identical to
the one applied for carbon monoxides As such, this section does not repeat the "roll-up" modeling
methodological description presented in the CO section, but instead simply highlights those elements of
the sulfur dio^demodelmg which djfler from ca
Control scenario
dioxide prof lies
-— --_=i.- ., ~f
Unlike the CO monitoring network, me number of monitors as well as the population coverage of the
SO2 monitoring network sfaim&^iUnng the 1980's. Table 55 summarizes the SO2 monitoring data used as
the basis for development ofthe^cbhtrol scenario air quality profiles.
As for CO, air quality profiles reflecting average values and dairy maxima for 1,3,5, 7, 8, 12, and 24
hour averages were compiled from AIRS for monitors in the lower 48 states which had at least 10
percent of their potential samples available. Applying a cutoff of 0.1 ppb to isolate the zero and near-
zero obsereationsjghree-parameterlognormal and gamma distributions were fitted to these empirical
profiles, fin the case of SO* the three-parameter lognormal distribution was found to provide the best fit
The control scenario SO2 air quality profiles are available on diskette, contained in a file named
SO2CAA.DAT. The same data format described in Table 54 is adopted.
142
-------
Appendix C: Air Quality Modeling
Table 55,
1975
1980
1985
1990
926
440
401
a/*
50%
6,602,615
NamberoT
No-control scenario sulfur dloxld& profiles
The no-control air quality profiles for SO, are derived using the same equation -equation 7 en page
139- that was applied for CO. for SOj, the baclqaprcttind concentration was assumed to be zero.
Although anthropogenic emissions contribute only small amounts to total global atmospheric, sulfur,
measured background concentrations for tiie continental U.S. range from only 0.1 to 1.3 ppb.
Background SO2 is discussed in more detail in the supporting document SAI SO* NOD and CO Report
(1994).15 . 7-; •*':--.*,.!. l"'\-?
The no-control scenario SQz air quality profiles are available on diskette, contained in a file named
SO2NCAA.DAT. The data format "is described in Table 54.
10
11
12
13
14
IS
16
Summary differences In sulfur dioxide sir quality
As for CO, reporting differences in control and no-control scenario air quality projections for each
monitor covered in the analysis is impractical due to the large amount of data involved. However, Figure
33 provides an illustration of scenario differences similar to the one provided for CO. Specifically, the
graph shows the distribution of 1990 control to no-control scenario 95th percentile 1-hour average
concentrations ratios at SO2 monitors. By 1990, SO2 concentrations under the no-control scenario were
substantially higher than those associated with the control scenario.
* SAI SOj, NOx, and CO Report (1994), page 4-9.
143
-------
^Appendix C: Air Quality Modeling
2
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
IS
19
20
21
22
23
14
23
26
27
21
29
X
31
32
33
34
33
36
37
39
39
Key
uncertainty
dloxldo
IffCf
for sulfur
Figure 33. Frequency Distribution for 1990 Control to No-
control Scenario 95th Percentile 1-Hour Average SO2
Concentrations, by Monitor. .:;
0.05 0.25 0.45 0.65 0.«5 1.05 1.25
Ratio of CAAtto-CAA 95th Percentile 1-Hour Avenp
The height of stacks used to vent flue
gases from utility and industrial fossil fuel-
fired boilers has a significant effect on the
dispersion of sulfur dioxide and on the
formation and long-range transport of
secondary products such as participate
sulfates. Under a no-control scenario, it is
conceivable mat some sources might have
built taller stacks to allow higher emission
rates without creating extremely high ground-
level concentrations of flue gases. On the
other hand, it is also conceivable that, in the '-
absence of post-1970 air pollution control programs, sources might have built shorter stacks to avoid
incurring the higher costs associated with building and maintaining taller stacks. To the extent facilities
would have adopted different stack height Configurations under a no-control scenario, both local
exposures to sulfur dioxides (and other emissions fjom fossil fiiel combustion) and long-range transport,
deposition, and exposure associated With secondary formatiojLproducts may have been different
However, this analysis assumes that both the locption of individual facilities and the height and
configuration of emission stacfi are constant between die two scenarios. If, in fact, stack heights were
raised unctalltfe^^ increases hi local SO2 concentrations under
the no-control scenario may be overlsAmated. However, this same assumption may at the same time
lead to underestimation under mi lioicoivtreil scenario of long-range transport and formation of secondary
particulates associa^^tb taller staclcsl for stacks built lower under a no-control scenario, local SO2
exposures would have beejnjljigher and long-range effects lower. Finally, die comments on uncertainties
for carbon monoxide on^i^ apply as well to SO2.
^
Nitrogen Oxides
- - ~K=£r- • • •
I-gl-T ' '
' " " f-£r°~~
Similarly-to sulfur dioxide, emissions of nitrogen oxides (NO,) -including nitrogen dioxide (NOj)
and nitrous oxidV (NO)- lead to several air quality effects. These effects include secondary formation of
fine particle nitrates, formation of ground-level ozone, long range transport and deposition of nitric acid,
and localized concentrations of bom NO2 and NO. The first three effects are addressed later hi this
appendix, under the particulate matter, ozone, and acid deposition sections. The focus of this section is
estimation of changes in local concentrations of NO2 and NO.
The methodology applied to estimation of local nitrogen oxides air quality is essentially identical to
the one applied for carbon monoxide and sulfur dioxide. As such, this section does not repeat the "roll-
jp" modeling methodological description presented in the CO section, but instead simply highlights
those elements of the nitrogen oxides modeling which differ from carbon monoxide.
144
-------
Appendix C: Air Quality Modeling
Control scenario nitrogen oxides profiles
After peaking around 1980, the number of NO2 and NO monitors, their county coverage, and then-
population coverage shrank between 1980 and 1990. Tables 56 and Table 57 summarize, respectively,
the NO2 and NO monitoring data used as the basis for development of the control scenario air quality
profiles. ~ _ *)"
Number of
Mbaftot* ;
Number of '
Mtooet •
Number of
Monitor
1*70
45
m
115
SOT
1980
1985
fcfo
2,142,606
1990
**
Table57.
V •-•^
, \ *
CoonUes
Number of
Moattor
n/a
1,023,834
1985
n/a
6,881
1990
145
-------
Appendix C: Air Quality Modeling
As for CO and SO* air quality profiles reflecting average values and maxima for 1,3,5,7,8,12, and
24 hour NO2 and NO averages were compiled from AIRS for monitors in the lower 48 states which had
at least 10 percent of their potential samples available. Applying a cutoff of 0.5 ppb to both NO2 and NO
to isolate the zero and near-zero observations, three-parameter lognormal and gamma distributions were
fitted to these empirical profiles. For NQ2 and NO, the three-parameter gamma distribution was found to
provide the best fit i;--.. „,.-
,T -" _ ~J!
The control scenario NO2 and NO air quality profiles are available on diskette, contained in files
named NO2CAADAT and NOCAA.DAT, respectively. The same data format described in Table 54 is
adopted. _-\ . . <. • - ": -> _
n
12
13
14
is
16
17
It
19
30
21
22
23
24
21
26
27
2S
29
30
No-control scenario nitrogen oxides prof lies
The no-control air quality profiles for NO2 and NO are derived using the same equation -equation 7
on page 139- that was applied for CO and SO2. As discussed hi detail in the SAl SOj, NO,, and CO
Report (1994),16 nitrogen oxides are emitted almost entirely from anthropogenic sources and they do not
have long atmospheric residence times. Therefore^ global background concentrations are very low, on
the order of 0.1 or 0.2 ppb. For the present analysis, background concentrations of NO2 and NO were
assumed to be zero. * hV> M ;*-'
The no-control scenario NO2 and NCfcJur quality profiles are available on diskette, contained in files
named NO2NCAA.DAT and NONCAA.DAT, respectively. The data format is described in Table 54.
Summary differences In nitrogen oxide* air quality
- ~ ™ . "- ~_'~~ ~ - _~ _~; ~- "—"^a^
Figure 34 provides a suniniar^ nidkation
of the differences jncontrol and 1
scenario air quality |^|||^. As:
SO* the graph shows^i|lri!bution i
control to no-control scenario 95th perciotUe
1-hour average ncentratio|(iil§iat NO,
monitors. These ratios indicate that, by 1990,
no-control scenario NO2 c
-------
Appendix C: Air Quality Modeling
2
Key caveats ancf uncertainties for nitrogen oxldos
A number of caveats and uncertainties specific to modeling NO, should be noted. First, stack height
3 and stack height control strategies likely to have influenced local concentrations of SOj may also have
4 influenced local concentrations of NO2 and NO. (For. a fuller discussion of the stack heights issue, refer
5 to the section "Key caveats and uncertainties for SO2" on page 144) In addition, the discussion on page
6 142 of uncertainties resulting from the use of state-level emissions and the cancellation of uncertainties
7 resulting from analyzing only differences or relative changes also applies to NO^. •
Acid Deposition l ?>
9 The focus of air quality modeling efforts described above for carbon monoxide, sulfur dioxide, and
10 nitrogen oxides was to estimate the change, in ambient concentrations of those pollutants as a result of
n changes in emissions. Particularly since the emissions modeling was driven by modeled macroeconomic
12 conditions, rather than actual historical economic activity patterns, neither the emissions inventories nor
13 the resultant air quality conditions developed for mis analysis would he expected to match historical
14 outcomes. The need to focus on relative changes, rather than absolute predictions, becomes even more
is acute for estimating air quality outcomes for pollutants subject to long-range transport, chemical
16 transformation, and atmospheric deposition. The complexity of the relationships between emissions, air
•7 concentrations, and deposition is well-described in the following paragraph from the RADM report
document developed by Robin Dennis of US EPA^si National Exposure Research Laboratory in support
19 of the present analysis:*7 ; . -'"- ~~ "_''.-'
20 "Sulfur, nitrogen, and oxidani species in the atmosphere can be transported hundreds to
21 thousands qfMometers by meteorological forces. During transport the primary emissions, SOj,
22 NO, and volatile organic emissions (FOC) are oxidized in the air or in cloud-water to form new,
23 secondary compounds* which are actiiK, particularly sulfate and nitric acid, or -which add to or
24 subtract from the ambiextt levels ofoxidants, such as ozone. The oxidizers, such as the hydroxyl
2s radical, hydrogen perox^a^ ozone are produced by reactions of VOC and NOf
26 and nitrogen pollutants are deposited to the earth through either wet or dry deposition creating
27 a food of pollutants to die earth's surface... However, the atmosphere is partly cleansed of
21 oxidants through a number of physical processes including deposition (e.g., ozone is removed by
29 wet and dry deposition). Dry deposition occurs when particles settle out of the air onto the earth
x or when gaseous or jfme particle species directly impact land, plants, or water or when plant
a stomata take ^gaseous species, such as SO* In wet deposition, pollutants are removed from
32 the atmosphere by either rain or snow. In addition, fine particles or secondary aerosols formed
33 by the gas- and aqueous-phase transformation processes scatter or absorb visible light and thus
34 contribute to impairment of visibility."
35 The complexity and nonlinearity of me relationships between localized emissions of precursors, such
36 as SO2 and VOC, and subsequent regional scale air quality and deposition effects are so substantial that
'/ the simple "roll-up" modeling methodology used for estimating local ambient concentrations
17 Dennis, R. RADM Report (1995), p. 1.
147
-------
Appendix C: Air Quality Modeling
i
2-
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
IS
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
and CO is inadequate,
even for a broad-scale,
aggregate assessment such as
die present study. For sulfur
deposition, and for a number
of other effects addressed in
subsequent sections of this
appendix, a regional air
quality model was required.
After careful review of the
capabilities, geographic
coverage, computing
intensity, and resource
requirements associated with
available regional air quality
models, EPA decided to use.
various forms of the
Regional Acid Deposition
Model (RADM) to estimate ,
these effects." Figure 35
shows the geographic domain
of the RADM.
Figure 35. Location of the High Resolution RADM 20-km Grid Nested
i the 80-km RADM Domain.
Control
deposition prof Horn
The derivation of control sceaario emission inventory inputs to the RADM model is succinctly
described in this excerpt from me Dennis, R. RADM Report (1995):
-f- ' • '
The RADM model requires a very defatted emissions inventory in boto tune and space. The
^emissions fields are also day-specific to account for the temperature effects on the volatile organics
and toe winded temperature effects on toe phone rise of toe major point sources. At toe time of
toe S12 Retrospective Study RADM runs, these inventories had been developed for 1985, using toe
1985 NAPAP (National Acid Precipitation Assessment Program) inventory, and adjusted for point
source emissions to 1988 for toe Eulerian Model Evaluation Field Study funded by NAPAP. These
RADM emissions inventories had county-level and detailed SCC and species-level information
incorporated into them to provide toe 80- and 20-km detail. The 812 Study emissions are
principally computed at toe state level. While toe 1985 812 Study emissions are close to toe
NAPAP inventory, they do not exactly match, nor do they have toe spatial, nor economic sector,
nor species detail within a state needed to run RADM. To connect toe 812 Study emissions to toe
11 For a detailed description of the various forms of the RADM and Us evaluation history, see the Dennis, R. RADM Report (1995).
148
-------
Appendix C: Air Quality Modeling
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
11
IS
19
20
21
n
24
25
26
27
28
29
30
31 '
32
33
34
35
36
37
38
KADM emissions, the following approach was followed: An industry/commercial-level
disaggregation (including mobile sources) was developed for the 812 emissions to allow different
sectors in a state to change their emissions across time -without being in lock step and the detailed
NAPAP emissions for every 80- and 204cm RADM grid-cell were grouped by state tojhe same level
of industry/commercial aggregation for an exact correspondence: Then it was assumed that the 812
Study 1985 control emissions were effectively the same as the 1985 NAPAP emissions. Relative
changes in emissions between the 8121985 control and any other scenario fag., 1985 no-control,
or 1990 control, or 1980 no-control, etc.) were then applied to the 1985 NAPAP state-level
industry/commercial groups in the appropriate 80- and 20-km grid cells, Thus* state-level
emissions for each group would retain the same state-level geographic pattern in the different '•
scenarios years, but the mix across groups could change with time. In this way, the more detailed
emissions required by RADM were modeled for each scenario year using the 812 Study emissions
data sets. .--
Although the focus of the present
analysis is to estimate the differences
between the control and norcontrol
scenarios, it is useful to illustrate the
absolute levels of acid deposition
associated with the two scenarios. It
is particularly important to
demonstrate the initial deposition
conditions to preclude possible
misinterpretations of the maps
showing percent change in . -
deposition. A relatively high ^
percentage change in a particular :•;
region, for example, may occur when
initial deposition is low^ even when J
the change in deposition!;! also
modest The RADM-modeled 1990
control scenario wet and dry ail^u
deposition pattern is shown in JpigUPB
36. A comparable map for .nitrogen
deposition is presented inJpgure 37.
Maps of the RADM-predicted 1990
No-control Scenario sulfur and
nitrogen deposition:are presented in
Figures 38 and 39; respectively.
-igure 36. RADM-Predicted 1990 Total Sulfur Deposition (Wet +
Dry; in kg/ha) Under the Control ^cenano.
39
40
41
44
No-control scenario mold deposition profit
Configuration of the RADM model for the present analysis —including allocation of emission
inventories to model grid cells, design of meteorological cases, treatment of biogenic versus
anthropogenic emissions, and temporal, spatial, and species allocation of emissions— are described in
detail in the Dennis, R. RADM Report (1995). The remainder of mis section provides a summary
description of the acid deposition modeling effort
149
-------
Appendix C: Air Quality Modeling
i
2
3
4
5
6
7
*
9
10
11
13
14
IS
16
17
IS
19
20
21
22
23
24
25
26
27
21
29
30
31
32
33
34
33
36
37
38
39
40
41
. 42
For sulfur deposition, the RADM
Engineering Model (RADM/EM),
which focuses on sulfur compounds,
was used to derive annual average
total (wet plus dry) deposition of
sulfur in kilograms sulfur per hectare
(kg-S/ha) under both the control and
no-control scenarios. The relative
changes in annual average total sulfur
deposition for each of the 80-km
RADM/EM grid cells for 1975,1980,
1985, and 1990 were then compiled.
Nitrogen deposition was
calculated in a different manner.
Since nitrogen effects are not
included in the computationally fast
RADM/EM, nitrogen deposition had
to be derived from the full-scale,
15-layer RADM runs. Because of the
cost and computational intensity of
the 15-layer RADM, nitrogen
deposition estimates were only J
developed for 1980 and 1990. As fjaj/
sulfur deposition, the relative
changes in annual average 4
plus dry) ntoog^ deposition*!
expressed as kg-N/hSj were
calculated for each 80*km grid ceDlf
and for each 6f the two scenarios. It
is important to note that ammoM
deposition contributes signifi|iDif$
total nitrogen deposition.
the activities of sources associated
with formation and deposition of
ammonia, such as livestock farming
and wildlife, were essentially
unaffected by Clean An- Act-related
control programs during the 1970 to
1990 period of mis analysis.
Therefore, ammonia deposition is
held constant between the two
scenarios.
igure 37. RADM-Predicted 1990 Total Nitrogen Deposition (Wet
+ Dry; in kg/ha) Under the Control Scenario.
d*:
'igure 31. RADM-Predicted 1990 Total Sulfur Deposition (Wet +
; in kg/ha) Under the No-control Scenario.
150
-------
Appendix C: Air Quality Modeling,
i
2
3
4
J
6
7
I
9
10
11
12
13
' 14
a
16
17
IS
22
23
24
25
26
27
28
»
JO
31
32
33
34
35
36
37
38
39
40
41
42
Summary
differences In add
t/t/on
Figure 39. RADM-Predicted 1990 Total Nitrogen Deposition (Wet
1+ Dry; in kg/ha) Under the No-control Scenario.
Figure 40 is a contour map
showing the estimated percent
increase in sulfur deposition under
the no-control scenario relative to the
control scenario for 1990. Figure 41
provides comparable information for
nitrogen deposition. These maps
indicate that by 1990 acid deposition
would have been significantly higher
across the RADM domain under the
•no-control scenario.
Examination of the percent
change sulfur deposition map
indicates relatively large percentage
changes in the upper Great Lakes and
•the Florida-Southeast Atlantic Coast
areas. This result may appear
somewhat surprising to readers ,
familiar with the historical patterns =_
of acid deposition. However, a -.y^
review of the emission data and the ;
control scenario sulfur deposition --=?
map reveal the reasons for this result.
First, Figure 36 shows that control
scenario deposition rates arelMV
relatively low. As described above,
even a small absolute increase in
deposition leads to a large percentage
increase in areas with low Initial rates
of deposition. Second, the scenario
differences in SO, emission rates for
these areas were substantial. For
example, 1990 no-control scenario
total SOX emissions for Michigan
were approximately 1.8 million tods
but control scenario emissions for the
same year were less than 600,000
tons; a reduction of over two-thirds.
Similarly, 1990 no-control scenario
emissions for Florida were over 2.3
million ir is, compared to
approximately 800,000 tons under
Figure 40. RADM-Predicted Percent Increase in Total Sulfur
t (Wet + Dry; in kg/ha) Under the No-control Scenario.
151
-------
Appendix C: Air Quality Modeling
i
2
3
4
S
6
7
8
9
10
11
12
If
14
IS
16
17
.11
19
20
21
22
23
24
"23
26
27
21
29
30
31
32
33
34
35
36
37
3S
39
40
41
the control scenario; also a reduction
of about two-thirds. Almost 1
million tons of the Michigan
reduction and approximately 1.3
million tons of the Florida reduction
were associated with utilities.
Emission reductions of these
magnitudes would be expected to
yield significant reductions in rates of
acid deposition.
Figure 41. RADM-Predicted Percent Increase in Total Nitrogen
Deposition (Wet + Dry; in kg/ha) Under the No-control Scenario.
Key caveats and
uncertainties for
acid deposition
Regional-scale oxidant and
deposition modeling involves
substantial uncertainty. This
uncertainty arises from uncertainties
in modeling atmospheric chemistry,
incomplete meteorological data,
normal seasonal and temporal
fluctuations in atmospheric ; --^'\-^..^.'^
conditions, temporal and spatial variability in emiaBions, and many other factors. Uncertainties specific
to the RADM model, and this particular exercise* tie discussed in detail in the Dennis, R. RADM Report
(1995). Itkimportant, however, 4® highlight some of the potential^ources of modeling uncertainty
unique to this analysis! —'s^-l" ' . •
' '
The first soiuxe of mcertainty specific to mis analysis is associated with the spatial and geographic
disaggregation of emissions ditip. As discussed in the Dennis, R. RADM report, the RADM model
requires emission rnventoi^inpi^ivhich are highly disaggregated over both time and space. The ideal
emissions inventory fed intolBf HADM model includes day-specific emissions to account for
temperature effects on VOGs and the significance of localized meteorological conditions around major
point sources. Given me broad-scale, comprehensive nature of the present study, such detailed emissions
inventories were not available. However, the industry/commercial-level disaggregation approach
developed for the present analysis would not be expected to introduce any systematic bias, and the
contribution of mis disaggregation of emissions would not be expected to contribute significantly to the
overall uncertajnty of the larger analysis.
= ' - f
The acid deposition estimates included in the present analysis are limited in that only the eastern 31
of the 48 coterminous states are covered. Although acid deposition is a problem primarily for the eastern
U.S., acid deposition does occur in states west of the RADM domain. The magnitude of the benefits of
reducing acid deposition in these western states is likely to be small, however, relative to the overall
benefits of the historical Clean Air Act
152
-------
Appendix C: Air Quality Modelmz
Rartlculate Matter
i •
2 Developing air quality profiles for particulate matter is significantly complicated by ihe fact that
3 "particulate matter" is actually an aggregation of different pollutants with varying chemical and
4 aerodynamic properties. Particulate species include chemically inert substances,' such as wind-blown
5 sand, as well as toxic substances such as acid aerosols; and mchide coarse partkles implicated in
6 household soiling as well as fine particles which contribute to human respiratoiy eifects. In addition,
7 emissions of both primary particulate matter and precursors of sendarify~fonned particulates are
» generated by a wide variety of mobile and stationary sources, further complicating specification of
9 particulate air quality models. Finally, particulate air quality models must take account of potentially
10 significant background concentrations of atmospheric particles. .."..-_
11 Modeling multiple species and emission sources, however, is not the only major challenge related to
12 particulate matter which is faced in the present study. Over the 1970 to 1990 period being analyzed,
13 understanding of the relative significance of fine versus coarse particles evolved significantly. Up until
14 the mid- 1 980's, particulate air quality data were collected as Total SuspendedParticulates (TSP).
is However, during the 1980's, health scientists concluded mat fine particles, especially those with an
16 aerodynamic diameter of less than or equal to 1 6 microns (PMJOX tvefe the component of particulate
n matter primarily responsible for adverse human health effects. As «f 1987, federal health-based ambient
it air quality standards for particulate master were revised to be expressed in terms of PM 10 rather than TSP.
19 Starting in the mid-1980's, therefore^the U.S. began shifting away from TSP monitors toward PM10
monitors. As a result, neither TSP norPM10 are fully represented by historical air quality data over the
21 1970 to 1990 period of mis analysis. Furthermore, a large number of U.S. counties have no historical
22 PM monitoring data at all, making ft difficult to estimate changes in ambient concentrations of mis
23 significant pollutant for areas containing roughly 30 percent of the U.S. population.
24 Given me retainre^^
25 historical Clean Air Act, It was important to develop methodologies to meet each of these challenges.
26 The methodologies developed and data lied are described primarily in the two supporting documents
27 SAI PM Report (1992) and SAI PM Report (1995).*9 To summarize the overall approach, historical TSP
21 data were broken down mto principal component species, including primary particulates, sulfates,
29 nitrates, organic particulates, and background particulates. Historical data were used for the control
jo scenario. To derive the no-control profiles, the four non-background components were scaled up based
31 on corresponding no-control to control ratios of emissions and/or modeled atmospheric concentrations.
32 Specifically, the primary particulate component was scaled up by the ratio of no-control to control
33 emissions of PM. Organic constituents were scaled up by the ratio of no-control to control VOC
34 emissions. In the eastern 3 1 states where RADM sulfate and nitrate data were available, values for SO4
15 and NO3 from an appropriate RADM grid cell were assigned to the relevant county and used to scale
36 these components of PM. For the western states not covered by RADM, sulfates were scaled up by the
37 change in SO2 emissions and nitrates were scaled up the change in NO, emissions. No-control profiles
ss were then constructed by adding these scaled components to background concentrations.
* In addition, SAI memoranda and reports which supplement the results and methodologies used in this analysis are included in the
references*
153
-------
Appendix C: Air Quality Modeling
i To resolve the problem of variable records of TSP and PM10 data, both TSP and PMIO profiles were /
2 generated for the entire 20 year period. Missing early- year data for PM10 were derived by applying
3 region-specific, land use category-specific PM10 to TSP ratios to the historical TSP data. Missing recent
4 year TSP data were derived for those areas where PM10 monitors replaced TSP monitors byapplying the
5 reciprocal of the relevant PM10 to TSP ratio. The methodology is described in detail in the SAIPM
6 Report (1995). % ,r^
• ' • «. ~^-tt*?$~ • . •
• "••"•, --r"Jf- •1-5%
7 In addition, to increase estimates of air quality on a ^eogi^M^ oasisfattliiieipolation
i methodology90 was developed to predict air quality for the control scenario bi^coiiin&a without
9 measured data. PM concentrations were estimated by first «ginuiting the ^nipon^
w sulfate, nitrate, and organic particulate, and primary particulate). The methodology for developing the
a concentrations of components within a county differed dependmg upon whether the county was within
13 or outside the RADM domain.
29
13 For those counties within the RADM domain, the RADM modeled ceieentrations for 1980 and
14 1990 were used to predict sulfate air quality. Relationships based on linear regressions that related
a 1980 and 1990 RADM sulfate concentrations to estimated sulfate paniculate concentrations were
i6 caloilated for counties wife AIRS data. Sulfate^^
n counties in the domain by applying the regression results to theR^M|ri^
is over the county center. Statewide average nitrate, VOC, and prin^| particulate concentrations were
19 calculated from measured ambient TSP and PMIO to describe fteaiionstituents in counties without data.
20 Control PM profiles were developed by adding the RADM-estpated sulfate particulate levels with the
21 statewide average nitrate, VOC, andpjciniary paiticulate levels, and background.
22 For counties outside the RAJDIrf domain, in alternate procedure was used. Using the primary and
23 secondary pafticulate estimates fo|i«punties wife date, statewide average sulfate, nitrate, VOC, and
i4 primary p^rfiailate concentratio^gjere determined. Control PMIO was predicted by adding the
25 statewide averages of all primary aal secondary particulate, and background. Usmg this method, all
26 counties that did noth«^ ^om^rs i^ are'in me same state are assigned the same PM concentration
27 profiles. These iaterpojated results af&skprry less certain than results based on actual historical
is monitoring data and''aiiidipii&re presented separately.
Control scenario partlculato matter prof/torn
x The number of TSP monitors peaked in 1977 and declined throughout the 1980's. Table 58
31 summarizes the daily $ie., 24-hour average) TSP monitoring data used as the basis for development of
32 the contnH^cenarKiiiir quality profiles. Most of the TSP and PM,0 monitors collected samples every six
33 days (i^e., 61 samples per year).
34 Daily PM10 data were also collected for each year between 1983 and 1990. Table 59 summarizes the
n daily PM,0 monitoring data used for the control scenario air quality profiles.
36 . Further speciation of TSP and PM,0 air quality data serves two purposes in the present analysis.
37 First, speciation of TSP into PM10 and other fractions allows derivation of PM10:TSP ratios. Such ratios
90 The interpolation methodology is described in detail in SAI, 19% Memo from J. Langsttff to J. DeMocker. PM Interpolation
Methodology for the §812 Retrospective Analysis. March 1996.
154
-------
-I
1
3
3
4
5
6
7
S
9
10
11
12
13
14
15
16
17
IS
19
20
23
24
21
26
27
23
29
X
31
32
33
34
35
36
37
38
39
40
41
Appendix C: Air Quality Modeling
can then be used to
estimate historical PM10 for
those years and monitors
which had TSP data but no
PM10data. The reciprocal
ratio is also applied in this
analysis to expand 1985
and 1990 TSP data to cover
those areas which
monitored PM10 but not
TSP. The second purpose
served by speciation of
particulate data is, as
described on page 153, to
provide a basis for scaling
up concentrations of each
species to derive
no-control scenario TSP
and PM,0 profiles.
Table 58.
Year
am
1975
1980
1990
Nttttberof
410
56,804
Monitor
64
Table 59. "Sommaiy of PMl&Momtoring Data,
-191$'
NnmAwrof
303
NumlKa-of
Mean
Number of
Monitor
To break the TSP and
PM to data down into
component species,
speciation factors were
applied to thePM fractions
with aerodynamic
diameters below 2.5 ^
microns (PMjj) andfiom
2.5 to 10 microns (PMW).
The PMjj speciatioa
factors wjpe drawn from ah ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
National Acid Precipitation; ^jj!!!1^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^""
Assessment Program ^ * ' •
(NAPAP) report on visibility which reviewed and consolidated speciation data from a number of
studies." These factors are presented in Table 60. In the table, fine particle concentrations are based on
particle mass measured after equilibrating to a relative humidity of 40 to 50 percent; and organics
include fine organic carbon.
~-1r
To develop speciation factors for coarser particles (i.e., in the PMU to PM10 range), a review was
performed by SAI of the available literature, including Conner et al. (1991), Wolff and Korsog (1989),
Lewis and Macias (1980), Wolff et al. (1983), Wolff et al. (1991), and Chow et al. (1994).92 These
speciation factors are summarized in Table 61. Data were too limited to allow differentiation between
urban and rural locations for coarser particles.
"J. Trijonis, "VisibUity: Existing and Historical Conditions-Causes and Effects," NAPAP Report 24,1990.
" This literature review, and complete citations of the underlying studies, are presented in the SAI PM Report (1995), pp. 4-2 to 4-6 and
pp. R-l to R-2, respectively.
155
-------
Appendix C: Air Quality Modeling
,.%.
V »'
•."'•. '•:•
URBAN EASt ,
Anunoniammttate ;
$£^-f\? y*f~ ;
Aoawatom i
'•>'• ^ $'• V
Data Source*: SAIPM
156
-------
Appendix C: Air Quality Modeling
Table 61. C<«r»Partide(PMw to PM^ Chemical Composi^
"Coanepartacte
B&ate
WEST
'OMHUfl psiticte Bftotrftntrfftf'ii'ft'
of Data
18
18
•.
18
14
7.? -36.7
8.41-25.81
The TSP andPMft control scoiario profiles developed based on this methodology are available on
diskette, under the filenames listed in Table 62.
No-control scenario partlculate manor profiles
4 To derive the no-control TSP and PM,0 air quality profiles, individual component species were
5 adjusted to reflect the relative change in emissions or, in the case of sulfates and nitrates in the eastern
6 U.S., the relative change in modeled ambient concentration. The following excerpt from the SAIPM
7 Report<1995) describes the specific algorithm used:93
_*"-'" f
a "For the retrospective analysis, the no-CAA scenario TSP andPM,0 air quality was estimated by
9 means of the following algorithm:
w »• Apportion CAA scenario TSP and PMIO to size categories and species;
n > Adjust for background concentrations;
SAI PM Report (1995), p. 5-1.
157
-------
Appendix C: Air Quality Modeling
i
2
3
4
S
f
7
S
9
10
11
12
a
14
IS
If
17
II
19
20
21
22
23
24
25
26
27
2t
29
30
31
32
33
34
» Use a linear scaling to
adjust the rum- Tabled. PM Control Scenario Air Quality Profile
background portions of
primary particulates,
sulfate, nitrate, and
organic components
based on emissions
ratios ofPM, SO* NO,
andVOC, and Regional
Acid Deposition Model
(BADAQ annual
aggregation results for
SO 4 and NO j;
PMiOCMEA^PAT
Add up the scaled +* - <
components to estimate ^» "CXJ« w^
the no-CAA scenario
concentrations.'
The specific procedures and
values used for the linear
rollback, speciation, fine to
coarse particle ratio, scaling, and
background adjustment steps are
described in detail in the SAI .
PM report (l^Sj^Jlable 63 "
lists the names of fte electronic
data files containing
and PM10 profiles for the
no-control scenario.
PMIONCOQ.&AT
Summary differences in pmrtlculate matter air quality
Figure 42 provides one indication of the estimated change in TSP air quality between the control and
no-control scenarios. Specifically, the graph provides data on the percentage of counties having TSP
monitors with estimated annual mean TSP concentrations above 120 micrograms per cubic meter
(ug/m3). Figure 43 provides similar county data for estimated 2nd High 24-Hour TSP concentrations
M SAI PM Report (1995), pp. 5-2 to 5-15.
158
-------
Appendix C: Air Quality Modeling
i
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
IB
19
21
22
23
24
25
26
27
28
29
30
31
32
33
34
33
36
37
38
39
40
43
above 260 ng/m3. Another indication of the
difference in outcomes between the two
scenarios is provided by Figure 44. This
graph shows the distribution of annual mean
TSP control to no-control ratios for 1990.
The X-axis values represent the mid-point of
the ratio interval bin, and the Y-axis provides
the number of counties falling into each bin.
Figures 42 and 43 indicate that by 1990 a
significantly greater proportion of monitored
counties would have observed high PM
concentrations under the no-control scenario
than under the control scenario. Figure 44
further indicates that annual average PM
concentrations would also have been
substantially higher in monitored counties
under the no-control scenario.
Key caveats and
uncertainties for ,
partlculate matter
There are several important caveats and
uncertainties associated wife the TSP and
PM10 air quality profiles developed iorthis
study. Although further reductions infuse
uncertainties were not possible for mis study
given time and resource limitations, the J?'
relative importance of particulate matter
reduction contributions towarfsiatel benefits
of the Clean Air Act highHgWsSiie
importance of these uncertainties.
A number of uncertainties were
introduced in the process of speciating and
rolling up individual components of
particulate matter. First, temporal and spatial
variability in the size and chemical properties
of particulate emissions are substantial.
These characteristics change from day to day
at any given location. Second, using changes
in proxy pollutant emissions, such as using
SO2 as a surrogate for SO4 in the western
states, to roll up individual PM components
may introduce significant uncertainty. Third,
even assuming a satisfactorily high degree of
Figure 42. Counties with Annual Mean TSP Concentrations
> 120 pg/m3. Expressed as a Percent of the Number of
bounties with TSP Monitors.
I Control
I No-control
1970 1975 1980 1983 1990
Year
7igure 43. Counties with 2nd High 24-Hour TSP
Ztacentratioas >26Q pg/m3, Expressed as a Percent of the
slumber of Counties with TSP Monitors.
1970 1975 1980 19S5 1990
Figure 44. Distribution of County-Level Annual Mean TSP
:AA to No-CAA Ratios.
0.00 0.20 0.40 0.60 0.80 1.00
Ratio of CAAJfo-CAA Annul Man TSP (interval (nit^dit)
159
-------
Appendix C: Air Quality Modeling
i correlation between target and surrogate pollutants, relying on predicted Changes in emissions at the state
2 level further compounds the uncertainty. Finally, and perhaps most important, using PM10 to TSP ratios
3 derived from late 1980's monitoring data may lead to significant underestimation of reductions in fine
4 particulars achieved in earlier years. This is because historical Clean Ah* Act programs focused
j extensively on controlling combustion sources of fine particulates. As a result, the sharejjif TSP
6 represented by PMIO observed hi the late 1980's would be lower due to miplementatipjaof'controls on
7 combustion sources. This would lead, hi turn, to underestimation of baselhiel|^^acentrations, as a
> share of TSP, hi the 1970's and early 1980's. If baseline PM10 concentrations in &ese early years are
9 underestimated, the reductions in PM,0 estimated by luiear scaling would also be^
10
n Nonlinear formation processes, long-range atmospheric transport, multiple precursors, complex
12 atmospheric chemistry, and acute sensitivity to meteorological conditions:combtrie to pose substantial
u difficulties in estimating ah*, quality profiles for ozone. Even in the context of |a aggregated, national
14 study such as this, the location-specific factors controlltng ozone formation preclude the use of roll-up
is modeling based on proxy pollutants or application of state-wide of jjation-wide average conditions. Such
u simplifications would yield virtually meaningless results for
77 Ideally, large-scale oxidant models—such as the Regional Oxidant Model (ROM)— would be
is combined with Eulerian photochemical grid models —such as the Urban Airshed Model (UAM>— to
19 develop control and no-control scenario estimates for ozone concentrations in rural and urban areas.
20 However, the substantial computing time and date lnput|«quirements for both ROM and UAM
21 precluded their use for mis stnd^||fj|Instead, tra^ separate modeling efforts were conducted to provide
22 urban and rural ozone profiles ib||iio^ areas of the lower 48 states for which ozone effects may be most
23 ' important. / :v:.~^ %~~ f£t.-i3fSs:i, ' •
24 First, for urban areas the Ozone Isopleth Plotting with Optional Mechanisms-IV (OZIPM4) model
25 was run for 14? urban *reasi|pabie 64 lists the urban areas modeled with OZIPM4. Although it requires
26 substantially less mput data£r^||AM, me OZIPM4 model provides reasonable evaluations of the
27 relative reactivity of ozone ps&upors and ozone formation mechanisms associated with urban air
28 masses.97 Three to five meteorological episodes were modeled for each of the 147 urban areas; and for
29 each of these, four model runs were performed to simulate the 1980 and 1990 control and no-control
x scenarios. The outputs ofmese model runs were peak ozone concentrations for each of me target
31 year-scenario combinations. The differentials between the control and no-control scenario outputs were
32 averaged over meteorological episodes and men applied to scale up historical ah- quality at individual
33 monitors to obtain no-control case profiles. As for the other pollutants, the control scenario profiles were
34 derived by fitting statistical distributions to actual historical data for individual monitors.
" See SAIPM Report (1995), p. 5-9.
"For a description of. * rtensive data inputs required to operate UAM, see SAI Ozone Report (1995), p. 1-1.
" See SAI Ozone Report (1995), p. 1-1.
160 '
-------
Table 64. Urban Aim Modeled with OZIPM4,
Appendix C: Air Quality Modeling
, 1S(M
, Alfeotown, PA-NI
Aftooai, PA
Atlanta, GA
BatanRouge, LA
< Bestunont, TX
WA
MT
Birmingham, AL,
Boalder, CO
Cedar Rapds, LA
IL
Charleston, SC
OwdestoB, WV
Oiarioae, NC
OH
Columbus, GA-AL
Dallas, TX
Decatur, IL
Denver, CO
OH
Fayctfevfife,
Flint, MI
FortCoBim.CO
WI '
, NC
SC:,
WV-KY
lajoesvilfe Rodk Co, WI
J
MS
Hfflade^hia, FA ^
Portanoufli,
ttMbttwW Mi"
JKaxGigB* PR-.
leading, PA
Re^HV ••
Rieamoo
-------
Appendix C: Air Quality Modeling
t
9
10
11
a
Second, the 15-layer RADM runs for 1980 and 1990 were used to estimate the relative change in rural
ozone distributions for the eastern 31 states. In addition, a limited number of 20-km grid cell high-resolution
RADM runs were conducted to benchmark the 15-layer, 80-km RADM median ozone response and to
estimate high ozone response. The relative changes in modeled median and 90th perccntile rural ozone were
then assumed to be proportional to the changes in, respectively, the median aid 9$hpeicejiiiie ozone
concentrations. The domain of the high-resolution RADM is shown in Figure 3S' and the general RADM
domain is shown in Figure 45. ,A
Finally, the SARMAP Air Quality Model (SAQNf) was runibr EPA by t
Bcwd(CARB) to gauge the differajcesm peak ozcraccm^
1980 and 1990. No-control profiles were developed for ozondataitors jfk these areas by
relative change in peak ozone concentration also applies to the snedfantif the ozone distribution. The domain
of the SAQM is shown in Figure 45. "
Figure 45. RADM and SAQM Modeling Domains, with Rural Ozone Monitor ffl^fflfif.
162
-------
Appendix C: Air Quality Modeling
2
3
4
5
6
7
g
9
10
11
12
13
14
IS
16
17
IS
•o
21
22
23'
24
25
26
27
28
29
30
31
Control scenario ozone profiles .
For ozone, air quality profiles were developed from historical AIRS data and calculated for
individual monitors based on 1,2,6,12, and 24 hour averaging times. Profiles based entile daily
maximum concentrations for these averaging times were also calculated. Givenlhe significance of
seasonal and diurnal ozone formation, twelve separate profiles of hourly ozone distributions were also
developed for six 2-month periods and for daytime and nighttime hours. The 2*montiLperiods are
January-February, March-April, and so form. The diurnal/nocturnal profiles are divided it 7 A.M. and 7
P.M. Local Standard Tune. All of these profiles are based on constructing 1, 2,6,12, and 24-hour
moving average profiles from the hourly ozone data from each monitor.* A two-parameter gamma
distribution is then fitted to characterize each of these air quality profiles.99 The functional form of the
gamma distribution, the basis for deriving the monitor-specific values for mean and variance, and an
analysis of the goodness of fit to the data are presented in the SAI Ozone Report (1995).
Table 65 summarizes the ozone monitoring data
used as the basis for the control scenario profiles. The
distribution of these monitors among urban, suburban,
and rural locations is presented in Table C-2 of the
SAI Ozone Report (1995). :"
» . - *'"
Given the substantial number of alternative air
quality profiles for ozone, approximately 20 high-
density disks are required to hold the profiles, even in
compressed data format Resource limitations ' ~_
therefore preclude general distribution of the actual =•-'
profiles. As discussed in the caveats and uncertainties
subsection below, however, me substantial
uncertainties associated with mcriel results JOT any
given area preclude application of these profiles in
contexts other than broad-sea^ aggregated
assessments such as the present study. The historical
ozone monitoring data used as Ihe basis for mis study
are,nevertheless, available through EPA's Aerometric
Information Retrieval System (AIRS).
Hooter
-------
Appendix C: Air Quality Modeling
i
2
3
4
3
6
7
I
9
10
11
12
13
14
IS
16
17
IS
19
20
21
22
23
24
Assumptions and modeling procedures not otherwise described in the SAI report were conducted in
accordance with standard EPA guidance.100 •
/ •
Similarly, the RADM modeling methodology used to estimate changes in day-time rural ozone
distributions in the eastern 31 states are described in detail in the Dennis, R. RADM Report (1995). The
referenced report also provides complete citations of the literature associated with development, standard
application procedures, and evaluation of RADM by the National Acid Precipitation Assessment
Program (NAPAP). ;-*-}? "„--"-" '
To derive the no-control scenario results for key California agricultural areas, lite California Air
Resources Board and US EPA's Region 9 office agreed to conduct three runs of the SAQM. For the
1990 control scenario, the 1990 SARMAP base case scenario adopted for California State
Implementation Plan modeling was adopted.101 Derivation of 1990 no<-control and 1980 control and
no-control scenarios was based on adjusting the aggregate mobile, point and area source VOC and NOX
emissions associated with each of these cases. For example, the 1980 no^control results were derived by,
first, multiplying the 1990 SARMAP base case mobile source VOC emissions by the ratio of 1980
no-control scenario to 1990 control scenario mobile source VOC emissions derived for the present study.
Similar adjustments were made for point and area sources, and for NOZ. The SAQM was then re-run
holding fixed all other conditions associated with the 1990 SARMAP basfccase, including meteorology,
activity patterns, and other conditions. The specific emission ratios used to modify the 1990 SARMAP
base case are presented in Table 66. The ratios themselves were derived by adding on-highway and
off-highway emissions to represent the mobile source category; adding utility, industrial process, and
industrial combustion emissions to represent point sources; and using commercial/residential emissions
to represent area sources. The no-ccptrol scenarios were then-derived by adjusting the peak and median
of the control scenario ozone distribution based on the ratioof SARMAP-predicted peak ozone
concentrations under, the control a&jd no-control scenarios.
> .... ^ ,. s A
Table $6*' Apportfosromart <^&jBfaffa»lg»cBteet»lte SAQM Bprc,
% v
VOC
no.
Mobile
"AfiMt,
AM*
Poto
'. wff^W^ '•^•WW^^M- '
to 199* Control Ratio
" ' f Jit,
IJWNoControIto
J990CoBtrd Ratio
tsss
100 US EPA, Office of Air Quality Wanning and Standards, "Procedures for Applying CUy-Sptctflc EKMA," EPA-450/4-89-012,1989.
101 Documentation of tbe SARMAP Air Quality Model and the SARMAP 1990 base case can be found in the SAQM references listed at
the end of this appendix.
164
-------
Appendix C: Air Quality Modeling
i
2
3
4
5
6
7
8
9
10
II
12
13
14
IS
16
17
It
19
20
21
23
24
25
The relative results of the control and no-control scenario runs of the OZIPM4, RADM, and SAQM
models were then used to derive the no-control case air quality profiles. For the urban monitors relying
on OZEPM4 results, only ozone-season daytime concentrations could be calculated directly from
OZEPM4 results. This is because OZIPM4 provides only the maximum hourly ozone concentration.
However, to estimate all the various physical consequences of changes in ambient ozone concentrations,
the current study requires estimation of the shift in the entire distribution ofozone concentrations. Since
it is daytime ozone season concentrations which are most sensitive to changes in VOC and NO,
emissions, the predicted shifts in the most important component of the ozone concentration distribution
are reasonably well-founded. The method adopted for this analysis involved applying the no-control to
control peak concentration ratio to all concentrations in the distribution down to a level of 0.04 ppm.
The 0.04 ppm level is considered at the high end of hypothetical ambient ozone concentrations in the
absence of all anthropogenic ozone precursor emissions. A ratio of 1.0 is used for ozone concentrations
at or near zero. The methodology is described in more detail in the SAI Ozone Report (1995) on page
4-6.. ' ' . . • --^-''"y.
Estimating changes in rural ozone concentrations is required primarily for estimating effects on
agricultural crops, trees, and other vegetation. For this reason, only the differences in daytime, growing
season ozone concentrations are derived for the present study. As described in detail in the SAI Ozone
Report (1995) on page 4-7, the no-control rural ozone profiles are calculated by, first, taking the ratio of
the average daytime growing season ozone concentrations simulated by RADM or SAQM (whichever is
relevant for that monitor). The ratio of no-control to control scenario average ozone concentration is
then applied to all the hourly concentrations from that monitor. Profiles based on 1,2,6,12, and 24-hour
averages are then calculated for the control case; and averages for daytime hours are calculated for the
no-control case.102 Even though the control and no-control scenario off-season profiles are held constant,
profiles for the no-control scenario are developed for all months of the year since the ozone season varies
throughoutAecountry. , H:_
26
27
21
29
x
31
32
33
34
35
36
37
3t
39
40
Summary dlfforoncoB In oxono air quality
Figure 46 presents a suiiiiiQary^ the
results of the 1990 OZn»M4 results "for all
1 47 of the modeled urban areast
Specifically, the graph depicts a frequency
distribution of the ratio of control to
no-control scenario peat ozone. While the
vast majority of simulated peak ozone
concentration ratios fall below 1 .00, eight
urban areas show lower simulated peak ozone
for the no-control scenario than for the
control scenario. For these eight urban areas,
emissions of precursors were higher under
the no-control scenario; however, the high
proportion of ambient NO, compared to
ambient non-methane organic compounds
Figure 46. Distribution of 1990 Control to No-control
DZIPM4 Simulated Peak Ozone Ratios.
30
20
•s
10
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Ritb of CA A Ho-CA A P«»k Oione (taurral • Upotat)
'? The no-control scenario nighttime profiles are assumed to be the same as the control scenario profiles.
165
-------
Appendix C: Air Quality Modeling
i
2
3
4
J
6
7
S
9
10
II
12
13
14
IS
16
17
a
19
20
21
22
23
24
25
26
27
21
29
30
31
32
33
34
3S
36
37
38
39
(NMOCs) in these areas results in a decrease
in net ozone production when NOZ emissions
increase. Figures 47 and 48 present
frequency distributions for control to
no-control ratios of average ozone-season
daytime ozone concentrations at rural
monitors as simulated by RADM and SAQM,
respectively.
These figures indicate that, by 1990, no-
control scenario ozone concentrations in the
modeled areas would have been generally
higher in both urban and rural areas. Rural
area concentrations differences are not as
great as urban area differences due to (a) the
differentially greater effect of CAA emission
controls in high population density areas, and
(b) potential differences hi the models used
for urban and rural areas.
Ozone reductions in both rural and urban
areas projected in mis analysis are not«iIP'
proportionally large as the estimatedpl*^
reductions in emissions of ozone precursors
for at least four reasons. Firsts funpt
knowledge of atmospheric photocfiknistiy
suggests that ozone Auctions i^esiiiibg from
emissions changes will be proportionally-
smaller than the emissions reductions; |^v=
Second, biogenic emissions
important ozone precuraor, ar^y|ignific
and are held constant forfiuBy
?igure47. Distribution of 1990 Control to No-control
lADM-Simulated Rural Ozone Ratios.
200
130
too
0.00 0.20 0.40 0.« OJO 1.00 IM
RattoofCAAtto-CAA Onw-Seuoi Dmyttae Avertfe (tan* (bttmlnidpofat)
•ignw 4«, Distribution of 19*|eontrol to No-control
SAQM-S&nulated Ozone Ratios.
10
0.00 OJO. 0.40 OM 0.80 1.00 1.20
btto ft CAA *h*CA A Oa>H-S*uei Diyttat Arenge OXOM (hteivtlmkipotot)
and no-
control scenarios of this anij^sppnogenic emissions are important because they contribute roughly half
of the total (manmade plus Batumi) VOG emissions nationwide. Due to this abundance of VQC loading
and the inherent nonlinearify of the ozone-precursor response system,103 historical reductions in
anthropogenic VOC emissions can yield minimal reductions hi ozone, especially in rural environments.
Third, this rural effect also influences urban areas receiving substantial ozone transported in from
surrounding areas? Consequently, the effect of emission controls placed in urban areas often is reduced
since mucilof the urban area ozone is imported. Thus, the problem is truly regionalized given the
importance of transport, biogenic emissions and associated urban-rural interactions; all contributing
toward a relatively non-responsive atmospheric system.104 Finally, physical process characterizations
"" Nonlinear systems are those where • reduction in precursors can result in a wde range of response m secondary poUutants such *s
ozone. Ozone response often is-flafornonresponsive to reductions of VOCs in many niral areas with significant natural VOC emissions.
Also; ozone can increase in response to increases in NO, emissions in certain localized urban areas.
m Both the 1990 CAA and EPA's and the National Academy of Sciew*^ Section 185B Report to Congre. ^.»gnized the
consequences of biogenks, transport and the need to conduct regk^izedassessnxfflts, as reflected in oi^aniz«tk)nal structures such as the
Ozone Transport Commission and the North American Research Strategy for Tropospheric Ozone (NARSTO).
166
-------
Appendix C: Air Quality Modeling
within OZIPM4 are severely limited and incapable of handling transport, complex flow phenomena, and
2 multi-day pollution events in a physically realistic manner. Consequently, ft is possible that the OZEPM4
3 method used herein produces negative bias tendencies in control estimations. Additional discussion of
4 uncertainties in the ozone air quality modeling is presented in the following section.
5 Key caveats and uncertainties for ozone
6 There are a number of uncertainties in the overall analytical results of the present study contributed
7 by the ozone air quality modeling in addition to the potential systematic downward bias discussed above.
« First, there are substantial uncertainties inherent in any effort to model ozone formation^aiid dispersion.
9 These uncertainties are compounded hi the present study by the need to perform city-specific air quality
10 modeling using OZIPM4, which is less sophisticated than an Eulerian model such as the Urban Airshed
u Model. However, while the absolute ozone predictions for any given urban area provided by OZIPM4
a may be quite uncertain, the process of aggregating results for a number of cities and meteorological
13 episodes should significantly reduce mis uncertainty. Urban areas for which ozone changes may be
14 overpredicted are offset to some degree by urban areas for which the change in ozone concentrations
is may be underpredicted. In weighing the significance of this source of uncertainty, it is important to
16 consider the central purpose of the present study, which is to develop a reasonable estimate of the overall
n costs and benefits of all historical Clean Air, Act programs. All analyses are based on relative modeled
is results, and ratios of the model predictions for the control and no-control scenarios, rather than the
19 absolute predictions. As a result of this, ||e effect of any bias in the model predictions is greatly
•"> reduced due to partial cancellation.
21 Additional uncertainty is contributed by othef MtDitations of the models, the supporting data, and the
22 scope of the present analysis. Inlying on linear Intefpolation between 1970 and modeled 1980 results to
u derive results for 1975, and between modeled results for 1980 and 1990 to derive results for 1985, clearly
24 adds to the umxrtaiiily associated ffi^ Assuming that changes
25 in peak concentration predicted by €@Ei|4ind SAQM can be applied to scale hourly ozone values
26 throughout the wogeiMi^ Resource and model limitations
27 dsorequiredihatniglift^^ This leads to
21 an underestimation of mewghj|tilpMX)mponent of ozone transport 'Finally, changes in rural ozone in
29 areas net covered by RADSi^|£QM could not be estimated. As a resuh, potentially significant
x changes in ambient ozone ip omer major agricultural areas, such as in the mid-west, could not be
31 developed for this analysis/ The Project Team considered using an emissions scaling (i.e., a roll-back)
32 modeling strategy to develop crude estimates of the potential change hi rural ozone concentrations hi
33 monitored areas outside the RADM and SAQM domains. However, me Project Team concluded that
34 such estimates, would be unreliable due to the nonlinear effect on ozone of precursor emission changes.
15 Furthermore, the Team concluded that baseline levels of ozone and changes in precursor emissions in
36 these areas are relatively low. The decision not to spend scarce project resources on estimating ozone
37 changes hi these rural areas is further supported by the relatively modest change in rural ozone
si concentrations estimated within the RADM and SAQM domains.
167
-------
Appendix C: Air Quality Modeling
' ' • 'Y
Visibility
2 Two separate modeling approaches were used to estimate changes hi visibility degradation in the
3 eastern and southwestern U.S. These are the two regions of the coterminous U.S. foe which Clean Air
4 Act programs were expected to have yielded the most significant reductions in;psibtlity degradation.
5 Visibility changes in the eastern 31 states were estimated based oj|fte RADM^itesults for sulfates;
6 and changes in visibility in 30 southwestern U.S. urban areas were calculated uio^ ft linear emissions g
7 scaling approach. Despite the potential significance of Clean Air Act-related visibll^ changes in ^
» southwestern U.S. Class I areas, such as National Parks, resource limitations precluded ii
9 of the analysis planned for these areas. :|
10 The RADM/EM system includes a post-processor which compi^^arious measures of visibility .
n degradation associated with changes hi sulfate aerosols.105 The basic iff|pch is to allocate the light
12 extinction budget for the eastern U.S. among various aerosols, mcludn^|frtpukte sulfates, nitrates, and
13 organics. The change in light extinction from sulfates is provided directiy^^
14 the complex formation and transport mechanisms associated with mis most significant contributor to
a light extinction in the eastern U.S. Nitrates are adit estimaledJBiil^ by ^ADM^ Instead, RADM-
u estimated concentrations of nitric acid are used as a surrogate topiiHpkime basis for estimating changes
17 in the particulate nitrate contribution to light extinction. The organic fractions were held constant
is between the two scenarios. Standard outouts include day light distribution of light extinction, visual
19. range, and DeciViews106 for each of RADM's 80-km grid celkuFor me present study, the RADM
20 visibility post-processor was configured to provide the 90th percentile for light extinction and the 10th
21 percentile for visual range to nsgreajnt worst casjasi and tip 50ft percentile for both of ftese to represent
22 average cases. More detailed jdoisimientation of iie RADM/EM system and the assumptions used to
23 configure $»j|sj^^
24 To estimate diiierepces hi contfc^idjBi^Bontrol scenario visibility in southwestern U.S. urban
25 areas, a modified Mncpr ^nhjck appnp^lt-was developed and applied to 30 major urban areas with
26 population grejtertl^ IQ|y||D^ For each of me 30 urban centers, seasonal average 1990 air quality
27 data was compiled for ke^|jn||i$|D|s contributing to visibility degradation in southwestern U.S. coastal
21 and inland cities, includingi^^pnd PMu. PM,0 was then speciated into its key components using city-
29 specific annual average PMjjppfQfile data. After adjusting for regional—and for some species
x city-specific—backgroundilevels, concentrations of individual light-attenuating species were scaled
31 linearly based on changes? In emissions of that pollutant or a proxy pollutant101 Using the same approach
32 used fin* the 1993 EPAJleport to Congress on effects of the 1990 Clean Air Act Amendments on
33 . visibility in Classipreas, light extinction coefficients for each of ftese species were then multiplied by
«« A complete discunkn, including appropriate references to other
-------
Appendix C: Air Quality Modeling
their respective concentrations to derive a city-specific light extinction budget109 This process was
repeated for pre-1990 control and all no-control scenarios by scaling 1990 results by the relative change
in annual county-level emissions of SO0 NO,, and PM. Based on the city-specific light extinction
budget calculations, measures for total extinction, visual range, and DeciView were calculated for each
scenario and target year. . :^=-:"
Control scenario visibility
9
10
11
12
13
14
IS
16
17
It
19
*•»
22
23
24
35
26
27
28
29
30
31
Unlike the other air quality
conditions addressed in the present
study, modeled visibility conditions
are used as the basis for the control
scenario rather than actual historical
conditions. However, like the other
.air quality benefits of the historical
Clean Air Act, it is the differences
between modeled visibility outcomes
for the control and no-control
scenarios which are used to estimate
visibility benefits. Nevertheless,
1990 absolute levels of eastern U.S.
visibility predicted by RADM under
the control scenario are presented in
Figure 49 to provide a sense of initial
visibility conditions. / - --•--"- =
For the!
1990 control scenario annual average
light extinction budge^ visual range,
and DeciView conditions are listed in
Table 67. These 1990 results are ^
presented to give the reader a sense
of the initial visibility conditions in
absolute, albeit approximate, terms.
igure49. RADM-Predicted Viiffiffliry Degradation,
\nmial Average DeciView, for Poor Visibility Conditions (90th
Percentile) Under the Coattol Scenario. .
32
33
34
3S
No-control scenario visibility
The no-control scenario visibility results for the eastern U.S. area covered by RADM are presented
in Figure 50. No-control scenario 1990 outcomes for the 30 southwestern U.S. urban areas are presented
in Table 68.
m The term "light extinction budget" refers to the apportionment of total light attenuation in an area to the relevant pollutant species.
169
-------
Appendix C: Air Quality Modeling
Table 67. 1990 Control Scenario
^
SinBiejo.CA
****&/ "
imi
' ftttf
*»
4U9R5
•"'/,«*""
36.0
JBRf-
_._.
If
170
-------
Appendix C: Air Quality Modeling
Table 68, 1990 N
-------
, Appendix C:Air Quality Modeling
Figure 50. RADM-Predicted Visibility Degradation, Expressed
t Annual Average DeciView, for Poor Visibility Conditions
|(90th Percentile) Under the No-control Scenario.
Sum,
In visibility
Decrview Haze lnd«
The DeciView H^ Index £dV) has recently been proposed as an indicator of the clarity of the
atmosphere that is more closet^ lifted to human perception man visual range (VR) or total extinction
(bgj (Pitchford and Malm, MllMplt is defined by the equation:
9
10
11
n
13-
where: "r" / ' • • •
h,,,, = total extinction in inverse megameters (Mm'1)
This index has the value of approximately 0 when the extinction coefficient is equal to the scattering
coefficient for particle-free air (Rayleigh scattering) and increases in value by approximately one unit for
each 10 percent increase in b^. Since the apparent change in visibility is related to the percent change hi
b^ (Pitchford et al.; 1990), equal changes in dV correspond to approximately equally perceptible
c' u ges in visibility. Recent research indicates that, for most observers, a "just noticeable change" in
visibility corresponds to an increase or decrease of about one to two dV units.
172
-------
•1
!
Appendix C: Air Quality Modeling
9
10
11
12
13
14
IS
16
17
II
19
20
21.
23
24
25
26
27
28
29
30
31
32
33
34
3S
36
37
3f
39
Both VR and dV are measures of the value of b^ at one location in the atmosphere. Both are
unaffected by the actual variability of the compositions and illumination of the atmosphere, so neither is
closely linked to the human perception of an actual scene. The isolation of these parameters from site-
specific variations and temporal fluctuations of the atmospheric illumination increases their usefulness
for comparing the effects of air quality on visibility across a range of geographic locations for a range of
time periods. Each parameter attempts to scale the b^ data so that changes in air quality can be used to
provide an indication of changes hi the human perception of a scene;
Modeling Results! • "---'_ - "--', i-
™-s =' "
The differences in modeled 1990 control and no-control scenario visibility conditions projected by
the RADM/EM for the eastern U.S. are presented in Figure 51. -The map shows the percent increase in
modeled annual average visibility
degradation under poor conditions for
1990 when moving from the control to
the no-control scenario. The results
indicate perceptible differences in
visibility between the control and no-
control scenario throughout the RADM
domain. The relatively large increase in
visibility impairment in the Gulf Coast
area is a reflection of the significant
increases in 1990 sulfate concentrations
associated with the no-control scenario.
(For related discussion of effects in Ihis
region, see page 149.) -;~;• "^
Jigure51. RADM-PredietedIncrease in Visibility Degradation,
Expressed in Annual Average DeciView, for Poor Visibility
Conditions (90th Percentile) Under the No-control Scenario.
The differences in modeled:
control and no-control scenario'
conditions in the 30 soBtippsteni
urban areas projected by itt
modeling are j
When reviewing these vis
degradation differentials for the 30
southwestern U.S. urban alias, it is
important to consider that while
estimated differences in visual range
were in many cases-very large, changes
in the DeciView Haze Index (dV) may
be relatively small. This is because the perception of visibility degradation measured by dV may be
small when baseline visibility is high.110 Even so, the results indicate that, by 1990, visibility in
southwestern U.S. urban areas would be noticeably worse under the no-control scenario.
110 See SAISW Visibility Report (1994), page 5-3.
173
-------
Appendix C: Air Quality Modeling
Tabled
Scenario Visibility €oafitk>n»fbr30
Southwestern U£. Cities.
4
V«naa»,CA
306
-4
W
in
240
153
Tttcwn,AZ
219
243
•»
SAISW VisibiJity Report (1994).,
174
-------
Appendix C: Air Quality Modeling
i Key caveats and uncertainties for visibility
2 There are several sources of uncertainty in the RADM and southwestern U.S. linear scaling model
3 analyses. For RADM, the use of nitric acid as a surrogatefor estimating changes in lighfeattenuating
4 nitrate particles ignores the interaction effects of nitrates, sulfates, and ammonia. Asa result, increases
5 in nitrates may be overestimated by the model when both sulfates/iind nitric acid uicrease. However, the
6 significance of this potential overestimation is mitigated to some extent by the relative insignificance of
7 nitrate-related visibility degradation relative to sulfates which prevails in the eastern U.&
29
s ' Several important uncertainties in the southwestern U.S. urban area visibility analysis are described
» in detail in the SAI SW Visibility Report (1994). First, the need to use seasonal average conditions leads
10 to underestimation of extreme visibility impairment episodes associated with high humidity, since
n particle growth due to water absorption is highly nonlinear. Second, although the use of city-specific
n light extinction and PM speciation data is significantly better than reliance on regional averages,
13 uncertainties hi city-specific data may contribute to overall uncertainty in &e estimates. However,
14 overall uncertainty associated with these factors will be reduced to some extent since overestimation of
is visibility degradation hi some cities will be offset by underestimations hi other cities. Finally, the linear
is scaling used to estimate the pre-1990 control scenarios and the no-control Scenarios was based on
n changes in county-wide or ah- basin emissions. Uncertainties associated with apportionment of state-
11 wide emission changes to individual counties or air basins may contribute significantly to overall
19 uncertainty in the visibility change estimates. Such apportionment is particularly difficult for SO,
'•> emission changes, since emission reductions achieved by the Clean Air Act tended to be at relatively
remote utility and smelter plants. However, suliates are a relatively minor source of light attenuation in
22 western urban areas. . - - -.."_,..; ~- v- ~~-'~
23 An important overall limitation ofthe visibility analysis conducted for the present study is that only
24 southwestern urban areas and the eastemSl states were included. The Clean Air Act may have
25 contributed toward significant reductions IB visibility degradation hi other areas. For example, Clean Air
26 Act programs to reduce ambient particulate matter may have motivated reductions hi silvicultural
27 burning hi some northwesterfa states. Perhaps the greatest deficiency in geographic coverage by the
28 present study is the omission of inability changes in Class I areas in the west
Mr Quality Modeling References
x Chang. "SARMAP Air Quality Model (SAQM)." Final report to San Joaquin Valley wide Air Pollution
31 Study Agency, 1995. .
32 DaMassa, Tanrikulu, and Ranzier. "Photochemical Modeling of August 3-6,1990, Ozone Episode in
33 Central California Using the SARMAP Air Quality Model. Part II: Sensitivity and Diagnostic
34 Testing." Preprints, Ninth Joint Conference on the Applications of Ah- Pollution Meteorology with
35 Air Waste Management Association. January 28 - February 2,1996, Atlanta, Georgia.
% Dennis, US EPA, ORD/NERL. "Estimation of Regional Air Quality and Deposition Changes Under
Alternative 812 Emissions Scenarios Predicted by the Regional Acid Deposition Model, RADM."
33 Draft Report. October 1995.
175
-------
Appendix C: Air Quality Modeling
i ICF Kaiser/Systems Applications International. "Retrospective Analysis of Ozone Air Quality in die
2 United States." Final Report May 1995.
3 ICF Kaiser/Systems Applications International. "Retrospective Analysis of Particulate Matter Air
4 Quality in the United States." Draft Report. September 1992. - = ^
3 ICF Kaiser/Systems Applications International. "Retrospective Analysis of^lnrticitliaite Matter Air
6 Quality in the United States." Final Report April 1995. -*•—---
ICF Kaiser/Systems Applications International. Retrospective Analysis^SOi, NO, and CO Air
in the United States." Final Report November 11*94. .f' X":=-^fe
9 ICF Kaiser/Systems Applications International. "Retrospective Analysis of the Impact of the Clean Air
10 Act on Urban Visibility in the Southwestern United States," Finalilepprt. October 1994.
ICF Kaiser/Science Applications International. Memo from J. Langstaffto^DDeMocker. PM
Interpolation Methodology for the §812 Retrospective Analysis. March 1996.
13 ICF Resources Incorporated. "Results of Retrospective Elelctefel^^
u 1985, and 1990." September 30,1992.^
is Seaman and Stauffer. "Development a0§]Design Testing of theiSARMAP Meteorological Model."
is Fhial report to San Joaqum VaUey^le Air P^ution Study Agency. 1995.
17 Seaman, Stauffer, and Lario-Gibbs|f"A Midti-|c|l|^FjfipIKniensional Data Assimilation System
u Applied in the San Jo^||^^|^ DuringPp^^' Part I: M0^®^ Design and Basic
a PerWmaiice Qborateristics^ of Applied Meteorology. Volume 34. In press. 1995.
20 Tanrikulu, DaMassa^ and Ranzieri^ ^PJhotcil^mical Modeling of August 3-6, 1990 Ozone Episode in
21 Central California Using the SARMAIlllLir Quality Model. Part I: Model Formulation, Description
22 and Basic Perfbrmani%i |ft^rints;illinth Joint Conference on the Application of Air Pollution
23 Meteorology with Air lpiik|jilanagement Association. January 28 - February 2, 1996. Atlanta,
24 Georgia.
25 Trijonis. "Visibility: Existing and Historical Conditions-Causes and Effects." NAPAP Report 24.
* 1990. :"
27 US EPA, Office of Air Quality Planning ^nd Standards. "Procedures for Applying City-Specific
a EKMA." EE^450/4-89-012. 1989.
176
-------
Appendix Ds Human Health
Visibility Effects of Criteria
Pollutants
ana
3
6
7
I
10
II
12
13
16
17
II
19
20
21
22
23
24
25
26
27
2»
29
X
31
32
33
introduction and Overview
"-
Principles tor the §Bt& Bon&flts
177
-------
_ Appendix D: Human Health Ejffects and Visibility Effects of Criteria Pollutants
i Efficient yse of Previous Research Results: Significant research effort has been spent to
2 understand and quantify the complex relationships between air pollution and human health. The §812
3 assessment has relied as much as possible on available research results, making adjustments as necessary
4 to apply the existing results to the current analysis.
5 Incorporate Uncertainty: To properly convey the results of any benefits assessment, it is
6 important to include an evaluation of how much confidence the analysts haveinthe estimates. Ideally
7 this would include a formal quantitative assessment of the potential for error, and what are the sources
> and directions of the likely biases and omissions. A method for considering and reporting uncertainty
9 must be built into the fundamental structure of the research design for the assessment , '
10
Health Effects Studl&i
n Scientific research about air pollution's adverse impacts uses a broad array of methods and
12 procedures. The research methods used to investigate the health effects of air pollution have become
13 considerably more sophisticated over time, and will continue to evolve hi tine future. This progress is the
14 result of better available research techniques and date, and the ability to focus further research more
is sharply on key remaining issues based on the contributions 0f earlier work.
16 The available health effects studies that could potentially be used as the basis of the §812
n assessment are divided into Epidemiology Studies and Human Clinical Studies. Epidemiological
» research hi air pollution mvestigates^assc>ciatioji between exposure to air pollution and observed
19 health effects in the study population. Human cUpical studies involve examination of human responses
20 to controlled conditions in a laboratory setting.(E^ hasiconducted research on health effects from
21 exposure, to pollution using each approach, and stufUes using these techniques have been considered in
22 various fcWalreguktoiypnxseedingj^ Each type of study (as it is iis^
23 described below, and the relative strengths and weaknesses for the purposes of the §812 assessment are •
24 examined. .'-/-.^'^ "^S-.rSj^
- ' --kM-- "'J " ' '
25 Epidemiolofi^cal Studied
26 Epidemiological studies evaluate the relationship between ambient exposures to and health
27 effects in the human population, typically in a "natural" setting. Statistical techniques (typically variants
2» of multivariate regression analysis) are used to estimate quantitative concentration-response (or
29 exposure-response) relationships between pollution levels and health effects.
x Epidemiology studies can examine many of the types of health effects mat are difficult to study
3i using a clinical approach. Epidemiological results are well-suited for quantitative benefit analyses
n because mey provide a means to estimate die incidence of health effects related to varying levels of
33 ambient air pollution without extensive further modeling effort These estimated relationships implicitly
34 take into account at least some of the complex real-world human activity patterns, spatial and temporal
•35 distributions of air pollution, synergistic effects of multiple pollutants and other risk factors, and
36 compensating or mitigating behavior by the subject population. Suspected relationships between air
37 pollution and the effects of bom long-term and short-term exposure can be investigated using an
33 epidemic-logical approach. In addition, observable health endpoints are measured, unlike clinical studies
39' .. which often monitor endpoints that do not result in observable health effects (e.g. forced expiratory
40 volume). Thus, from the point of view of conducting a benefits analysis, the results of epidemiological
178
-------
• Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
studies, combined with measures of ambient pollution levels and the size of the relevant population,
2 provide all of the essential components. •
3 Two types of epidemiological studies are considered for dose-response modeling: cohort and
4 longitudinal studies. Cohort-based studies are population based studies where initially disease free
5 individuals are followed over a certain period of time, with periodic reporting of jhe health status from
6 the individuals. Studies about relatively rare events such as c^cermcidenc» tfiioriatity can require
7 tracking the individuals over a long period of time, while more common eventsl(e||4 respiratory
s , symptoms) occur with sufficient frequency to evaluate the relationship over a much shorter time period.
9 An important feature of cohort studies is information is known about each individual including other
10 potential variables related to the disease being studied. These variables, called confounders, are %:-
a important to identify becauseiif theyare not accounted for in the stody and they are associated with air
n pollution levels, a false association may be associated with air pollution. The exposure information can
13 range from data from a personal exposure monitor carried by the participants for a few weeks in a study
14 concerning minor symptoms, to monitoring data of ambient air concentrations.
• _ v--^j±ifr::%^
is A: second type of study used in. this analysis is IcraghMinal or time*seiie$ studies. The
16 relationship between population-wide health uiformationsa^ as counts for daily mortality, hospital
n admissions, or emergency room visits (e.g. heajttattack, nudtalUy, acute asthma attack or chronic
is bronchitis) and ambient levels of air pollution are evaluated. Oneadviffitage of this study design is it
19 allows "the population to serve as its own control". Changes in such factors as tobacco, alcohol and
20 illicit drug use, access to health care, employment, wd nutrition inay have a pronounced impact on the
v health of an individual, and, if they vary with the pollution variable of interest, need to be included hi a
cohort study. However, such potential confounding factors are unlikely to vary with pollution levels and
23 these variables do not need to jbje, evaluated mthcanarysp
24 Soine of &e drawbacks ol^
25 exposure, measurement errors m the explanatory variables, the influence of unmeasured variables, and
26 correlations between the pollution varialrfesaf concern and both the included and omitted variables.
-"""*_ _"„ _= r - ~£ «
27 These can potentially lead ito spurious conclusions. However, epidemiological studies involve a large
a number of people and domot suffer extrapolation problems like clinical studies.
~-~_"-~"T ." ''*- •
29 Human Clinical Studies \ *y
30 Clinical studies of air pollution involve exposing human subjects to various levels of air
31 pollution in a carefully controlled and monitored laboratory situation. The physical condition of the
32 subjects are measured before, during and after the pollution exposure. The measured physical condition
33 can include general biomedical information (e.g., pulse rate and blood pressure), physiological effects
34 specifically affected by the pollutant (e.g., lung function), the onset of symptoms (e.g., wheezing or chest
35 pain), or the ability of the individual to perform specific physical or cognitive tasks (e.g., maximum
36 sustainable speed on a treadmill). These studies often involve exposing the individuals to pollutants
37 while exercising, increasing the amount of pollutants that are actually introduced into the lungs.
33 Clinical studies can isolate cause-effect relationships between pollutants and certain human
39 health effects. Repeated experiments altering the pollutant level, exercise regime duration and types of
40 participants can potentially identify effect thresholds, the impact of recovery (rest) periods, and the
differences in response among population groups. While cost considerations tend to limit the number of
42 participants and experimental variants examined in a single study, clinical studies can follow rigorous
43 laboratory scientific protocols, such as the use of placebos (clean air) to establish a baseline level of
44 effects and precise measurement of certain health effects of concern.
179
-------
-Appendix D: Human Health Effects and Visibility Effects of 'Criteria Pollutants
i There are drawbacks to using clinical studies as the basisTor a comprehensive benefits analysis.
2 Clinical studies are appropriate for examining acute symptoms caused by short-term exposure to a
3 pollutant While mis permits examination of some important health effects from ah- pollution, such as
4 bronchoconstrictibn hi asthmatic individuals caused by sulfur dioxide, it excludes studying more severe
5 effects or effects caused by long term exposure. Another drawback is health effects measured hi some
6 well-designed clinical studies are selected on the basis of the ability to measur&precisery the effect rather
/ than a larger symptom, for example forced expiratory volume. The impact ofiome clinically measurable
s health effects such as lung function on future medical rendition or lifestyle ^
9 understood.
10 Ethical limits on experiments involving humans also impose important limits tolhe potential
n scope of clinical research. Chronic effects cannot be investigated because people cannot^ekeptin
12 controlled conditions for an extended period of time, and because these effects are generally irreversible.
13 Participation is generally restricted to healthy subjects, or at least to exchjde people with substantial
14 health conditions that compromise then- safe inclusion hi the study. This can cause clinical studies to
is avoid providing direct evidence about populations of most concern, such as people who already have
16 serious respiratory diseases. Ethical considerations also limit the exposures to relatively modest
17 exposure levels, and to examining only mild health effects that do no permanent damage. Obviously for
if ethical reasons human clinical evidence cannot be obtained on the possible relationship between
19 pollution and mortality, heart attack or stroke, or cancer,,
u The very precision about expospecond^tiom possible mcUnical studies also creates an obstacle
21 to using clinical dose-response functions for benefit analysis. K is difficult to extrapolate results from
22 clinical settings to daily exposures fated by the w|nle popuption. For example, many clinical studies
u evaluate effects on exercising individual. Only & small portion of the population engages hi strenuous
24 activity (manual labor or exercise) at any time. Reflecting these fundamental differences between the
2s laboratory setting and the Seal woxid? impo
» information about human activittjiitttim^ exercise levels, and poUution levels.
• - - - - • •
27 issue* in Selecting Studies To Estimate Health Effects
a A number of issues aria|]when selecting and linking the individual components of a
29 comprehensive benefits analysis: The appropriate procedure for handling each issue must be decided
jo within the context of the current analytical needs, considering the broader analytical framework. While
31 more sophisticated or robust studies may be available hi some circumstances, the potential impact on the
n overall analysis maydpake using a simpler, more tractable approach the pragmatic choice. In
33 considering the0V|£all impact of selecting a study for use hi the §812 assessment, important factors to
34 consider inchideile likely «"agnfr"Hg the decision will have on the overall analysis, the balance between
35 the overall level of analytical rigor and comprehensiveness hi separate pieces of the analysis, the effect
36 on the scientific defensibility of the overall project, and the potential impact on the schedule and EPA's
37 available resources.
33 This section discusses seven critical issues hi selecting health information for use in the §8 12
39 assessment types of studies considered, confounding variables, concentration-response function
40 thresholds, exposure modeling, meta-analyses, uncertainty and the links to other portions of the §812
4i assessment The previous discussion about the types of research methods available for the health
a information alluded to some of these issues, as they are potentially important factors hi selecting between
180
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
studies using different methods. Other issues address how scientific research will be used in the overall
2 analytical framework, and have not been previously discussed.
3 Types of Studies Considered
, ; '• • .-.' •-.. - ' • . • • ..&>• •
4 Given the major advantages of epidemiological studies - exposures do jtot need to be modeled
5 and health effects are observed in a large, more heterogeneous population, epideniiological studies are
6 used as the basis for determining both health effects and dose response curv^^£Jink^ studies are used
7 if there are health effects observed in clinical studies not observed in epidemH^togl^sbidies.
« Confounding Factors -••';• . • _ J
• • • '--is^lT^f^ "^f^jf^y'
9 Confounding can .occur when the real cause of disease is associated with the disease and the
10 variable being evaluated. If only me variable oemg evaluated Is iisediilh^
n can occur. For example, in longitudinal epidemiology studies of air imitation* it is important to take into
12 account weather conditions, because weather is assciciated wim air Dolfotiona^ health outcomes. If
u only air pollution is evaluated, a false associated could occur. Potential confounders include weather-
14 related variables, age and gender mix of the subject population, and the type and distribution of the
11 pollution emissions. Studies, that include a broad range ojT likely cojofounders can offer a more robust
16 conclusion about an individual pollutant, even if the statistical confidence interval is larger due to
n . including more variables in the analysis. , rijA-f^
it Thresholds / : ~;:
/-"-•? .>-r
19 Exposure-response relationships are conceptualizex! as either exhibiting a threshold of exposure
20 below which adverse effects are not expected to occur^^r as having no response threshold, where any
21 exposure level theoretically poses a non-zero risk of response. The methods employed by health
22 researchers to characterize exposure-response relationships may or may not explicitly analyze the data
23 for the existence of a threshold. Studies may analyze relationships between health and air pollution
24 without considering a threshold. However/extrapolating the risk relationship beyond levels hi the study
25 may overestimate benefits, if 4 threshold for population risk exists but is not identified by researchers,
26 then Clean Air Act benefits wxHildb? underestimated, because non-existent risks would be estimated at
27 low levels of air pollution and tubJracted from the risks associated with high exposure levels. On the
21 contrary, if a threshold is artificially imposed where one does not exist, the relative benefits of the Clean
29 Air Act may be overestimated, because the population risk associated with levels of pollution below the
x threshold would be zero* In general, those studies that explicitly consider the question of a threshold
31 (whether a threshold is identified or not) provide stronger evidence and will be considered a positive
32 feature when selecting studies for this analysis.
33 Exposure Modeling
34 One potential obstacle to using dose-response information from clinical research methods in a
3i benefits assessment is the need for an exposure model. This requirement adds an additional step in the
36 analytical process, introducing another source of uncertainty and possible error. The dose-response
37 functions developed from clinical research are specific to the population participating in the study and
33 the exposure conditions used in the laboratory setting. For air pollution research, important factors
affecting exposures include concentration, duration of exposure, recovery times in cleaner conditions,
4o and exercise (i.e., ventilation) levels.
181
-------
'• Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i To apply the clinical results to model the general population, two decisions must be made. First,
2 how far to expand the conditions in the clinical selling? For example, if the subjects in the clinical study
3 were healthy male college students, should the results be applied to the entire population, including
4 children? Second, how many people in the general population are exposed to conditions similar to those
5 used in the clinical setting Frequently clinical studies are conducted at relatively high exercise levels
t (increasing the dose, or the quantity of pollutants actually delivered to the lungs). In the general
/ population few people experience these conditions very often, and people do not reach these exercise
i levels with equal frequencies during the day and night In addition to die exen^ level question, the
9 analyst must also determine the number of people mat are exposed to the levels of ambient conditions
10 seen in the laboratory. Air quality varies throughout a city and is typically reported % data from
n monitors located at various places throughout the city. However, people are not exposed to the f ;
a conditions at any one monitor all day. As people move around in the city, they are exposed to ambient
n air quality conditions represented by different monitors at different times during the day. To further
14 compound the problem, air quality also varies between indoors and outdoors, within a car or garage, and
u by such factors as proximity to a roadway or major pollution source (or sink). The exposure model must
is account for the ambient conditions in the "microenvironments" that Depopulation actually experiences.
n The issues of study subjects, exercise and nucroenvaxMiments can influence the choice of clinical
a studies selected for the §812 assessment Clinical studies that use exposure regimes and exercise levels
n more similar to what larger groups of the population see are easier tor apply in a benefits model than are
20 more narrow studies. Similarly, studies mat use a diverse group 0f subjects are easier to apply to the
21 general population than are more narrow studies. -->" -'
. ifSv-jn^p _>,;
-------
Appendix D: Human Health Effects and Visibility Effects of 'Criteria Pollutants
the macroeconomic, emissions, dispersion, and exposure models because of the diverse origins of the
2 models. Therefore, instead of a complete formal uncertainty analysis, the §812 assessment includes a
3 less rigorous analysis of the inherent uncertainties in the modeling effort The uncertainty analysis
4 combines quantitative and qualitative elements designed to sufficiently describe the implications of the
5 uncertainties. A primary goal of the sensitivity/uncertainty analysis is to identify the health effects that
6 make a sizable contribution to the overall assessment of the monetary benefits;' There may be situations
7 where there are significant differencesin the available mfonnation used to predict tiie incidence of a
s particular health effect (i.e., the uncertainty bounds are large). It is Important to alert the reader to
9 situations where using the lower incidence estimates may portray the health effect as only modestly
10 contributing to the overall total benefits, but using reasonable alternative Uglier estimated incidence*
n figures (or higher monetized values) would substantially impact not only the monetized value of the
12 . individual health effect, but actually make a noticeable difference in the total benefits assessment.
n Consideration of the overall uncertainties inherent in the §812 assessment has several important
14 implications on health study selection. It is important to carefully examine the pragmatic balance
a between the level of uncertainties in the analysis and the need for comprehensive coverage of all benefit
i6 categories. There are frequent situations where there is a direct tradeoff between more comprehensive
17 coverage and the restriction of the analysis to mote certain information. The relationship between the
a overall uncertainty in other parts of the analysis and the uncertainty choices concerning each particular
19 health effect will be carefully considered. In addition, a balance must also be maintained between
20 comprehensiveness of an assessment and the allocation of limited analytical resources.
21. Links to Air Quality Modeling :, - :
.. -„=-
22 Selection of health studies for the §812 assessment must consider the need to. match the health
23 information tome air quality modeling conducted for assessment For example, information on the
24 health effects from short tens (five minute) exposure to sulfur
25 infonnation on average daUy suffittj|^ade levels. There are two key components of matching the air
26 quality information, wMch are actua% intimately related: averaging times and spatial resolution.
27 Matching the health studie^ the exposuiminodeling (if needed) and the air quality modeling is a case of
a searching for a "lowest cwmiHi denominator" between the components of the assessment
-= - --"'a £- -i ~-£~fetT~ .
. -• '"--£*%&k
» Target Population }_ -> ;r
't- . • ' •
"_•*
30 Many of me studies relevant to quantify ing the benefits of air pollution reductions have focused
31 on specific sensitive subpopulations suspected to be most susceptible to the effects of the pollutant.
32 Some of these effects may be relevant only for the studied subpopulation; effects on other individuals are
33 either unknqw^ or aot expected to occur. For such studies, the challenge of analysis is to identify the
34 size and characteristics of the subpopulation and match its occurrence to exposure. Other studies have
15 examined specific cohorts who may be less susceptible than the general population to health effects from
36 air pollution (e.g., healthy workers), or who differ in age, gender, race, ethnicity or other relevant
37 characteristics from the target population of the benefits analysis. Extrapolating results from studies on
38 nonrepresentative subpopulations to the general population introduces uncertainties to the analysis, but
39 the magnitude of the uncertainty and its direction is unknown. Because of these uncertainties, analyses
« often limit the application of the dose-response functions only to those subpopulations with the
4i characteristics of the study population. While this approach has merit in minimi/ing uncertainty in the
analysis, it can also severely underestimate benefits if hi fact similar effects are likely to occur in other
43 populations. For these reasons, studies that examine broad, representative populations are preferable to
44 studies with narrower scope because they allow application of the functions to larger numbers of persons
45 without introducing additional uncertainty.
183
-------
Appendix D: Human Health /Effects and Visibility Effects of Criteria Pollutants
i Exposure Ranges
2 One component of Ae §812 analysis will estimate the air pollution levels that would have
3 occurred in the absence of the Glean Air Act These estimates will certainly be much larger man
4 currently observed levels of US air pollution, and perhaps even levels currently observed elsewhere in
5 the world. This aspect of me analysis poses difficulties wim me application of concentration-response
6 functions that have been based on exposures at much lower pollution levels. The shape of the
7 concentration-response function much above observed exposures^jevels is unkaown^ It is possible that
i biological mechanisms affecting response that are unimportantltflow levels of exposure may dominate
9 the form of response at higher levels, introducing nc^mearity to me niathematical relatkmsiiip. In >
10 general, studies mat include exposure levels spanning the raf^ofmtei?estmme§812aKessineotafe
n , preferable to studies at levels outside of the range, or that only include a narrow part of the linge, A
a possible drawback to this approach is that studies which fit mis criterion have often been conducted
13 outside the US The application of foreign studies to US populations introduces additional uncertainties
14 regarding the representativeness of the exposed population and the relativecomposition of the air
75 pollution mix for which the single pollutant is an indicator. These difficult issues must be balanced by
16 the analyst when selecting studies for the benefits analysis. ;[, • • f
i? Peer Reviewed Research ^v-"" . . ._"' "i*.i¥;i>. _^~
a Whenever possible, peer reviewed research rather than unpiibHshed information has been relied
19 upon. Research that has been reviewed by the EPA's own peer review processes, such as review by the
20 Clean Air Science Advisory Commit|B§{eASAC>6f the Science Advisory Board (SAB) has been used
2/ whenever possible. Research reviewid by other public scientific peer review processes such as the
22 National Academy of Science^ AeJNational A|i|jc Decapitation Assessment Program is also be
23 . included in this category. ""
24 Res«^^pidN^edmpeJii^|wBd journals but not reviewed by CASAC has also been
25 considered for use in tf»J812 asseslxffi^%|d has been used if his determined to be the most-
26 appropriate available tMn||^|Resea]^iO$ipted for publication .by peer reviewed journals ("in press") has
27 been considered to h«i^|^||^lished:£^idications that EPA intends to submit research to the CASAC
23 (such as inclusion in a drl(i^H|%pocument or Staff Paper) provide further evidence mat the journal-
29 published research
, -•
x •..,; Air pollution healtil research is a very active field of scientific inquiry, and new results are being
31 produced constantly. Mftiry research findings are first released in University Working Papers,
32 dissertations, govenmpot reports, non-reviewed journals and conference proceedings. Some research is
33 published in abstra^form in journals, which does not constitute peer review. In order to use the most
34 receot research finlings and be as comprehensive as possible, unpublished research has been examined
15 for possible use in the §812 assessment Any unpublished research used will be carefully identified in
36 the report, and treated as having a higher degree of uncertainty man published results. The peer review
37 of the §812 assessment by the Advisory Council on Clean Air Compliance Analysis will provide one
33 review process for not only all components of the assessment, but also the way in which the components
39 have been used. .
184
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Application Of Health Science Research And Air
Quality Results To Estimate Health Effects
j The general modeling approach used in the §812 assessment is a "reduced form" or "embedded
¥ model" approach."2 The concept of a reduced form is to use simplified versions of previously
5 •constructed complex models to characterize the impact of a series of linked physical and socioeconomic
6 .processes. The results of three major types of models have been used in the integrated reduced form
7 model, but the models themselves remain exogenous to the reduced form model. Those three models .are
s the macroeconomic model, the emissions model, and the air quality dispersion models for the criteria
9 pollutants.
10 The reduced form model differs primarily from the previous health effects benefits models
// prepared by EPA for criteria pollutant analysis in two ways. First; a. single integrated model is used for
n all criteria pollutants. This increases the consistency in the integrated assessment, and also facilitates
n sensitivity analysis of vital components.
T \-
14 . , The second major difference is that the level of detail used in the exposure model component is
is limited to a level of detail that matches the air quality mcKlelingresults prepared for the assessment. For
16 example, EPA's previous models for conducting ozone health analysis require spatially dispersed ozone
n data on an hourly basis. Because that level of time-disaggregated data is not available for the §812
is assessment (particularly for the "no-control" scenario), the §812 reduced form model has not been run at
o that level of disaggregation. This creates difficulties in estimating health effects with dose-response
functions based on short-term exposures. To mkigate this problem, a procedure was developed for
21 estimating the appropriate daily distribution of pollutant levels (e.g., one, two or seven hour data) from
22 available data from the §812 air quality modeling, A simplified version of the human activity pattern
23 portion of me exposure model was also developed. While these simplifications increase the uncertainty
24 levels, the overall effect is small. Considering the uncertainties in the other components of the analysis
25 (especially the steps leading to the "too-conHoT air quality modeling), the pragmatic advantages of a
26 simplified integrated model are a valuable tradeoff for less detailed modeling.
27 Some reanalysis and manipulation has been done on information used in various portions of the
2» analytical structure to link all the diverse pieces appropriately. In general, the approach favored
29 adjusting the exposure components of the analysis rather than the dose-response (or concentration-
» response) information. ,'.'
31 One difficult issue that confronts any health benefits estimation project of air pollution is
32 aggregating results* An important objective of any integrated analytical approach is to avoid double
33 counting of effects while also avoiding inadvertently omitting some effects altogether. Two main facets
u of this objective are aggregating effects across symptoms (or diseases) and across pollutants. Many of
35 the criteria air pollutants are closely related, and considerable effort has gone into health effects research
36 to isolate the effects of each pollutant This is especially difficult in the case of particulate matter,
37 because that category includes all of another criteria pollutant (lead), and a huge and important portion of
39 two others (sulfur oxides). In addition, all the criteria pollutants can and do occur simultaneously,
39 making it difficult to isolate impacts.
112 This approach draws heavily from the proposed modeling approach in "Acid Rain Benefit Assessment: Draft
Plan for the [§812] 1992 Assessment," Acid Rain Division, U.S. EPA, 1991.
185
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Concentration-Response Functions
i •.'-•• • • •
2 Issues Common to Several Pollutants
3 . ' • '-*'..
'4 The concentration-response functions describe the relationships between pollutant concentrations
5 and a variety of health effects. The following discussion includes general issues common to several or
6 all pollutants regarding the selection of health endpoints and studies, the form of the concentration-
7 response functions, choice of values used for baseline incidence, and the choice of a£fected populations
a modeled. Additional pollutant-specific information is included in sections pertaining to individual
9 pollutants. Each pollutant section also includes a table that presents the dose-response functions, specific
10 information needed to apply these functions for the current analysis* and references for the information.
/; For each pollutant, the choice of the health effects modeled was as inclusive as possible. For
12 those pollutants associated with, a variety of endpoints, the health effects may overlap with each other, as
13 described hi sections below. Despite this overlap, the current analysis presents separate estimates of
14 incidence for all health effects. ; > >:
•"••• • ' •- ^f*^^
is The analysis includes as many studies related to a given health effect as possible, except for
16 studies inapplicable to the current analysis. For some endpoints, me group of adequate studies yielded
i? mixed results, with some showing statistically significant responses to pollutant concentrations and
ii others with insignificant associationa^Mnless study methods have been judged inadequate, dose-
a response functions with both statistically significant and insignificant coefficients have been included to
20 characterize the possible range; of risk estimates. .Hfwever, 'ft should be noted that some studies that
21 found insignificant effects for a pp|(atant could not be used because they did not report the insignificant
22 coefficient values. " -
23 The analysis includes modeling of the health effects that have been associated with exposure to a
24 number of pollutants,TJttt$pn cases^vhefe the health effect is being modeled for the several correlated
25 pollutants of mterest, n^K^sup models-mat included multiple pollutant exposures were chosen in
26 preference to single poUul^ iacilete to avoid over counting. Using separate regressions for each
27 pollutant (if the pollutants ajf^Jaftelated) may overstate the number of effects caused by each pollutant
21 alone. As an example, this^naiysis models incidence of premature mortality for bom participate matter
29 and ozone. Thus, coefficients for bom these pollutants come from the regression model that considers
x the effects of bom pollutants simultaneously.
j?*' ' '
31 -A few- studies have identified pollutant concentration thresholds below which health effects have
32 been shown not to occur. However, for the current analysis, all concentration-response functions assume
33 no threshold effects. Thus, in certain cases, the incidence of health effects may be overstated at low
34 concentrations.
is Many studies reported only a relative risk value (defined as the ratio of the incidence of disease
36 hi two groups exposed to two different exposure levels). The analysis required conversion of these
37 values to then* corresponding regression coefficients when the coefficients were not reported. When
33 converting the relative risk to a coefficient value, the analysis used the functional form of the regression
39 Aquation reported by the authors of the study.
186
-------
Appendix D: Human Health Effects and Visibility Effects Of Criteria Pollutants
The coefficients from a number of studies measured the change in the number of health effects
2 for the study population rather than a change per individual. These coefficients were divided by the size
3 of the study population to obtain an estimate of change per individual. The coefficient could then be
4 multiplied by the size of the population modeled in the current analysis to determine total incidence of
5 health effects.
< . " ' ; • '
7 Certain dose-response functions (those expressed as a change relative to baseline conditions)
a required baseline incidence data associated with ambient levels.of pollutants; Incidence data necessary
9 for the calculation of risk and benefits were obtained from national sources whenever possible, because
10 these data are most applicable to a national assessment of benefits. The National Center for Health .
11 Statistics provided much of the information on national incidence rates. However, for some studies, the
a only available incidence information come from the studies themselves; in these cases, incidence in the
13 study population is assumed to represent typical incidence nationally.
« Many studies focus on a particular age cohort of the population far the identification of health
a effects. The choice of age group is often a matter of convenience (e.g., extensive Medicare data may be
16 available for the elderly population) and not because the effects are, hi reality, restricted to the specific
n age group (even though men- incidence may vary considerably over the life span). Thus, the application
is of the concentration-response functions was, in most cases, canned out for tfie entire population using
19 incidence values appropriate to the population at large. Additional calculations were also carried out for
20 the subset of the population specifically studied in the report which served as the source of a
21 concentration-response function. In other pses, studies were performed on individuals with specific
22 occupations, activity patterns, or medical conditions because these traits relate to the likelihood of effect.
In these cases, application of the dose>response function has been restricted to populations of individuals
24 with these same characteristics. v:. V -'-
25 Pollutants Using Doso-r&sponse Information from
26 Epldemlolofflcal Studies
27
Ozone
,.^^-j -TIP,
"-:>= ~ _>> ~
~ -_=• --
2< .Ozone has been associated with a variety of health effects. Table 70 includes the concentration-
29 response functions used in ibis analysis as well as information needed to apply these functions. This
x section addresses several specific issues concerning selection and use of ozone dose-response functions.
31 - • Choice of concentration-response functions for each health effect. For each health effect the
32 analysis includes concentration-response functions identified in the peer-reviewed literature. The current
33 analysis excludes some studies based on the quality of the study or inapplicability of the study to this
34 analysis (e.g., use of exposures adjusted for indoor and outdoor areas near individuals rather than
35 monitored ambient exposures).
36 Ozone concentration thresholds. Several studies have found statistically significant relationships
37 between ozone and health effects at ozone concentrations below the current hourly standard of 0.12 ppm.
3» Although all functions in the current analysis assume no thresholds of ozone concentration below which
*> health effects are assumed not to occur, one study listed in Table 70 does suggest a threshold of 0.025
ppm.
187
-------
_ Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i Conversion of coefficients dependent on symptom status during the previous day. Krupnicket
2 al. (1990) employed a Markov process to determine the probability of symptoms that were dependent on
3 symptom status of the previous day. The current analysis adjusts the regression coefficients produced by
4 the model in order to eliminate this dependence on previous day's symptom status.
s Within-study meta-analvsis. Ostro and Rothschild (1989) report six separate regression
6 coefficients that correspond to regression models run for six separate years. To estimate a single
7 coefficient, this analysis performed a fixed coefficient meta-ana^ps on the six yesis>f data.
" "
= ~~'~ * —
s N^rmfflizalioji of ft coefficients by population. For several studies of
9 admissions, the current analysis required normalizing the P coefficient by me population
10 geographical area of study. For example, Thurston et al. ( 1 994) reported the dependent variable as the
u number of admissions/day hi the Toronto, Canada study area. Thus, the P coefficient was divided by .the
12 population of the study area to determine admissions/person/day, . •
.- -~ • -->=--=- -=* --•._
13 Issues related to the applied populations. In general, the analyses apply; to the entire population,
14 with the exception that some studies applied only to the particular cohort studied in the research papers.
is For example, Crocker and Horst ( 1 98 1) has been applied only to individuals who work in occupations
16 requiring heavy outdoor physical labor. Whittemore and Koin {1980) has been used only for the
17 population of asthmatics. In addition, Ostro and Rothschild (1989J measured effects on working age
is adults, thus their findings were applied only to those It to 65 years of age.
- • ^'vy*- "&•' .a- '
19
fifjfcQtSi Severalftealth endpoints for ozone overlap with
20 each other. These relate primarily to studies with]^^
21 studies of the presence of any of 1ft j^ptom$;p;il|pi^i»f al., 1990) and studies of restrictions in daily
22 activity (Ostro and Rothschild, 1989; Portney antfituliahy, 1986). All endpoints are included in the
23 current analysis. The possibility of ebuble-counting overlapping effects was considered in aggregating
24 results across endpoints. ViL.
188
-------
,2
I
(w
O
.2
o
'I
!
I
8.
a
r
1
£
«•
i
I
1
li
11!
8.'
i »
I <*
M
.s
ts
oo
§
9
ii
ea.
I
.s
-
•rf
ts
I"?
o
II
o
1.
I
'•a
I
*
o .
n
CO.
3.
.s
'a
t*.
o\
oo
'.S
-------
I
•n "S
a- *
s
O
n ,
<=> 's —
i »
X
g
§
o
II
s
o
1
3
«
o
I
f
•*a
S
1 .a
00
r
r
A »o
R A o •-:
iiti
?-
iitii»
if
4?
'•• -Ms
1-
8 a a a 5 a a-
fe a a. 8
a a a
-------
4
r
8
o
u
«
eo_
8.
S
ea
g.
£
a
£
t
i
3
i
a
r
r
IJU
.13*
A " *-C ^^
lf
Wt <• K «
s = a
-------
I
I
•8
I
t
I
i
ff
S
2
•-•s
•«{ x
A
«if-
2
- 51
I
! *
»
,* l '
••-*- .
•.f
s
I
a
£
i
•g-
t*
=3^
15
• 2
U X
MVO
•§•
- «2
s>^l
i V)
S
^2-5
i'S
1>
i
111 I
1
1
ni
!^«
i:-
8
I I
I I
I-!*
?I?
!^r
t
it*
r f;
s B a s 3
-------
I
I
o
M
i
•*
.S .S
•§ 4
o .S
ffrf
*S
P»
ll
! ii ii
, j^U ca^i
is":
!9J
' N .
! C--
S °
3 5 S ^ II
«' 1^1
s iJltil
11
•s
-x'-'S
=>d<5'S
^ >>
as
H g
H i»!
> ^rfl
I
•.s
•r.
!•
I
lg.
•i1
M
•a-
11*1
•S
•a
It
-s
-------
.8
e
8 s
df i
«'•;
"s
N
£
-------
I
*
I
.
a u
o
"8
If
^5
83
Q
•s
•a
1$
i
00
» n
§>
S&.
u ^^ "
Is-
£
1
.a
N*
?:
y
* H"
<&
*
.5
HN*-
1-i
9 9 o
r K n n
j a jo
O O <3 X
II II II -g
« «r «r *.'
ON
r1|Jg
till
P «•*
-------
«n
-------
_ Appendix D: Human Health Effects and Visibility Effects of Criteria Polliaants
Particulate Matter •
2 Similarly to ozone, participate matter has been associated with a variety of health effects,
j ranging from acute respiratory symptoms to mortality. Although detailed discussion of each health
4 endpoint is not provided, this section addresses several specific issues concerning the dose-response
5 functions. These functions pertain to PM-10.
""•="-.
« Selection of dose-response functions. The analysis has Mentified many studies that include
7 dose-response functions fpr paniculate matter. For example, some studies included multiple analyses,
» yielding both significant and non-significant results, depending on the nature of the Input parameters
9 (e.g., for different lag periods or concurrent exposures). la these cases, only significant results iwere
10 included. Studies were excluded if health endpoints could not he defined in the U.S. population. For
// example, in Pope and Dockery (1992) the authors developed a unique definition of symptomatic
12 children in Utah which has no correlation in the incidence data bases which were available (e.g., from
13 National Center for Health Statistics, Centers for Disease Control); consequently, the results could not
14 be applied to the general population. Table 71 includes the paniculate matter concentration-response
is functions used in this analysis. In general, the chosen functions come from fleer-reviewed literature.
16 Conversion oj coefficients dependent ojn symptom sjajug djlfJQg Jhe. previous day. Krupnick et
n al. (1990) employed a Markov process to determine the probability of symptoms that were dependent
is on symptom status of the previous day. The current analysis adjusts the regression coefficients
19 produced by the model in order to eliminate this dependence on previous day's symptom status.
Within-study ^etaanaiysfc, Ostro and Rofhschild (1989) report six separate regression
21 coefficients that correspond to regression models run for '&*. separate years. To estimate a single
22 coefficient, this analysis performed a fixed coefficient metaanalysis on the six years of data.
*_
23 Conversion of exposure measures. Some studies modeled concentration of total suspended
24 particulates (TSP)/fineparticulates,or a,coffif^ This analysis converted these measures to
25 PM-10 based on location-specific information whenever possible.
2« |sjues related tp. ihfrjqpplfad, population. As noted earlier, the application of the concentration-
27 response functions was general!^ carried out for the entire population and the particular age cohort of
28 the study. For paniculate matter, the current analysis includes several exceptions to using the
29 concentration-response functions for the entire population. The Schwartz hospital admission studies
jo (1994a, 1994b, 1994c, and 1995) found effects for individuals over 65 years of age; thus the current
31 analysis uses the concentration-response function for this cohort only. Pope et al. (1991) measured
n respiratory effects in 10-12 year old children. These effects are extrapolated to those under 18 years
33 old, nationwide. Another paper, Ostro et al. (1991), applies only to asthmatics, while Ostro and
34 Rothschild (1989) applies only to working age adults.
t • '
15 Issues related to the overlap of health effects. As with ozone, several of the health endpoints
3t identified in the paniculate matter literature may overlap with each other. This occurrence is apparent
37 especially with studies that model various respiratory symptoms. For example, one study models the
3» presence of any of 19 respiratory symptoms (Krupnick et al. , 1990), while another models lower or
19 upper respiratory symptom responses (Pope et al. , 1991) which may include the same symptoms
modeled hi Krupnick et al. (1990). Still other studies include dose-response functions for acute
41 respiratory symptoms that result in minor restrictions in daily activity or both minor and more severe
42 restrictions in daily activity (such as days of lost work) (Ostro and Rothschild, 1989; Ostro, 1987).
43 This analysis models incidence of all endpoints. .
197
-------
a .
] '
. 1
••g
s
CO
r-
4>
I
i ^^
rl
8
8
O
I
i:
ai
JS
I
3
I
i
T3
t$
e
I
o
8
o
i
o
H
•a
v
2
1
u
o
II
.3
t
•a
ts
o
I
t
I
vi
D
•5
u
I
.S'.
1"
§:
•>
II
.a
=5
fj?
i
!*»>
P
•s
-------
8S.
*
•8
O
II
o
It
o
II
u
o
II
M
O
S
o
II
8
o
2 .S
e>
00
H
.5
i
;4
CO
»
•s
I
"?
D
1*8
i
i
!!
i
!«*?
!»
i
£i
i
: «0
co
co
sea
-------
sens . B vs. s . s s s s *
II
p
S
i
P
ii
""
I if
» *
311
2,1
1 ?•
.§111
'*•?&*
JL§
*S
X S
ers>
Iff
2.5
I *~
1 I
f"
05
t
s.8t
oP
8
'S
la eg « r'
Pp Pi?
I
f.
s
-------
CO
f
!
.
^* 5
•s
CO
-
— "X
I
« D
a
5 1
« D
S
n
e
n
§
n
i-.i
s -
s *
I o
.J "i
U
S s
§
O
H
u
O
1
g
O
I
8
O
d
CO.
I
1
.S
.s
1
i2
"
8 co
bo
.a
I to
a
8
in
*|5
till
I-
t s s » a a a * a
s s
-------
>
1
•*.
§ '
I1
i
1
t
!•".
\9
i
e
*
I
i
1*>
n
p
S
s\
a
E.
ft
e.
-------
o
II
o
II
«
£
*
a
i
•&,
n :
t1
.?
. -4
4
•S. %
*'
*
r t
, »
^M-,
A
***! S
ii-f
J
I
£
1
S
8
«s
ts
? I
- I
1
•8
i'S
i'S
!o 5.
M
1
I2
!^|1
^f"8
* r
ii
A A
§
1
pit
tt
g.
lSg.8,
•a *-
£
•s
l-r
v
:
-------
-a «•«•"•
Ills
if-'
g? I
I;
=• *•
09
i
•S'
Cg
«J 5.
.311,
l*i
'»» I
*rv
?! I
B.
o
« B «
it «
§
i|i
^p ™
' 1
.1
i
I
S
E
• " £
I!
a
P-
o1
8-
B. B.
i!
.si
I
2,
II
r
B • ' *
11
I
|h! If
ai
'I
90
8!
11
i
ill III
|s
a..1
£
8
I
i
I
if
-------
£8
~
J3
ill
!ll 1
.s
•8
-a
•3)=5
ftl 1
1|| J
in Z ^J£
_, • — •
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants.
i Nitrogen Oxides
2 Nitrogen dioxide (NOj) is the primary focus of health studies on the nitrogen oxides and serves
3 as the basis for this analysis. The primary pathophysiology of NO2 in humans involves the respiratory
'4 system and the concentration-response function identified for NO2 describes the relationships between
s measures of NO2 and respiratory illness. - .
6 A number of epidemiological studies are available; however, most have either confounded
7 exposures (with other pollutants) or insufficient exposure quantification (e.g., exposure assessment
* indicates only absence or presence of a gas stove). Most studies consider NO2 generated by gas stoves or
9 other combustion sources in homes. Most studies are not directly usable in concentration-response
10 functions. However, studies by Melia et al, 1980 and Hasselblad et al, 1992 provide a reasonable basis
11 for development of a concentration response function. Table 72 presents the function obtained from their
12 work. The function relates NQ2 to respiratory illness in children. -„-
13 Applied population. The only concentration-response function readily adaptable to this analysis
14 measured respiratory effects in children aged six and seven, .The current analysis applies the
15 concentration-response function to all ages, based on the likelihood that adults are also at risk.
u Epidemiological studies demonstrate the susceptibility of adolls to other pollutants (see PM and ozone
17 analyses) and adult respiratory systems do not differ substantially from pose of children. In addition,
is adults, especially the elderly, have illnesses which may make mem more susceptible to effects of
a respiratory damage. Consequently, if adults were not included in this analysis, serious undercounting
20 would be likely to occur. On the otherihand, many of the studies which evaluated very young children in
21 depth have obtained results indicating that the youngest age cohort are not at risk (Melia, et al, 1982 and
22 1983; Sametetal, 1993); me specific age c;ut-o^¥jry^
23 studies report positive associations in young chfldte& (Ware et al, 1984). Thus, the most inclusive
24 alternative mvoJves^valuatmg risks to the entire population, applying the notion that all individuals,
25 even very young children and older adu)^ may be susceptible to nitrogen oxides.
.- :;;,^:=i§L
"' -/-
206
-------
§"
<§
o
^
en
PJ
u
I
j
lil
i en
p
o
I
•f
1
.s
I
$
2* s
P *
I*
§£
o
!-a
«
1M
s
•S g a
If. I
1 !
^
ii i
[i -s
I1
!!s?
-------
lm
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
/ Pollutants Using Dose-response information from
2 Laboratory Studios
3 Carbon Monoxide .
4 Three dose-response relationships are available for estimating the health effects of carbon
5 monoxide. The first relates ambient CO levels with hospital admissions for congestive heart failure
6 (Morris et al., 1995). The second equation (Allied et al., 1989a,b, 1991) relates the CO level in the
7 bloodstream to the relative change hi time of onset to angina pain upon exertion. The third relates the
« CO level in the bloodstream to the relative change in time of onset to sBent ischemia. Doe Id the lack of
9 quantitative information relating silent ischemia to a meaningful physical health effect, this analysis uses
10 only the first two dose-response functions. :
11 Table 73 presents these functions. The Morris et at (1995) function represents a straightforward
n application of a linear function. The relationship presented in Allred et al, (1989a,b,; 1991) requires a
13 relationship between CO exposure (inppm) to %COHb (Hie measure of CO M&e bloodstream, used by
14 Allred et al.). Ihe document yi/rg^f^Crifer&jjgr^
u graphically (U.S. EPA, 1991). Relationships for 1-hour «^l|Jy;»«xpjoiure times are available for two
16 . levels of exertion, at rest (alveolar ventilation rate of 10L/min^ and wim light exercise (20 Umui).
n These functions assume an initial COHb ofA5 percent and standard physical characteristics for the
a exposed individual. The current analyfisjpresents the linear functions relating 1-hour CO exposure
19 during light exercise and 8-hour CQ exposure at rest to COHbJevels.
20 The analysis requires the fjdlowing addMpaal dato in order to characterize the health effects of
*"
22 natibnvnde pofMilation of i%iis patients,
23 frequency of jangma attacks :ftf||i^|dual patients, and
24 baselmetimefi&iDiii^^
25 TJte nationwide pOf|a||t|6g^ angina patients was 3,080,000 in 1989 (the American Heart
26 Association, 1991). Data oimeifiiquency of angina attacks or baseline time to onset of angina could not
27 be located; therefore, data p Allred et al. (1991) were used. Allred et al. (1991) give the mean frequency
u of angina attacks for the study population as 4.6 per week with a range of 0 to 63. The baseline time to
29 onset of angina on beginning of the treadmill test was 515. seconds at a COHb level of 0.63 percent This
30 number was obtaineduby taking the mean of the mean times to onset for four treadmill tests: three pre-
31 exposuretests andJene post-exposure test in which the exposure was to background laboratory air in a
32 control chambor; The treadmill test used a modified Naughton protocol in which individuals were
33 subjected to an increase of 1 basal metabolic equivalent every two minutes (Naughton and Haider, as
34 cited by Allied etal., 1991).
15 The study population of Alhsd et al. (1989a,b, 1991) was 63 males, age 35-75, with stable
36 angina. Only nonsmokers who had quit at least three months before the study were included, but most of
37 the individuals had a history of smoking. Because active smokers have chronically high COHb levels
38 (Allred et al., 1989a), the dose-response relationship used in this analysis can only be applied to
39' nonsmokers. This analysis assumes that all angina patients are nonsmokers, although they may have had
40 a history of smoking. In addition,'the analysis assumes mat the dose-response relationship is accurate for
41 all male and female angina patients.
208
-------
w *
•> ON
"L
J
a
•I
1 §•
CO
u
1
•as;
I
I
£
I
s
1
.a
131
I
812
.9
a a
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
10
11
12
1J
14
IS
16
17.
Sulfur Dioxide
This analysis estimated one concentration-response function based on an epidemiologies! study
of premature mortality (Moolgavkar et al., 199S). A second function was estimated using the clinical
data on the responses of exercising asthmatics to SO* as measured by the occurrence of respiratory
symptoms. The table mat follows summarizes the derivation of these functions^ The procedure for
estimating a concentration-response function for respiratory symptoms is detailed below.
Symptoms reported for individual subjects are available from two studies (BjOgeret al., 1985 and
Linn et al., 1987). Using the individual subject data, the percentage of subjects experiencing at least
moderate symptoms as a result of exposure to SO2 while exercising cajq, be estimated. ~ '
>-*«-
All types of symptoms {chest tightness, shortness of bream^or wheeze) were combined; that is, if
a subject had any of these symptoms, he/she was recorded as havrng f iymptom response; if the subject
had more than one of these responses, the highest of the symptom sco^|»|siused hi the analysis.
A multiple regression was performed with thejk^odfs of experiefisigtf least noticeable
symptoms as the dependent variable, and SOj exposjiKlilel.ilm |tpm and ai^ina status as the
independent variables. A similar regression wpperfbnn^f
moderate symptoms. To calculate cases frongpme log odds regsWlWHiiiF'to8 °dds ate fi*8* transformed
back to probability:
• < « s V\ f „ ..XvWvV-v-'
. < x ' \ ,f &-&A.,\
\.
! •• •W™-!"'5
^.^-J^^
K1•P^m.^^
*>^5"
II
19
20
The probability for mild i
a value of 0 for ariMiaiid 1 fon
by the size of die applicable;
latics isi calculated separately using this equation, and using
• the status variable. Finally, these probabilities are multiplied
21
22
23
24
25
26
27
2S
29
30
31
32
33
whexi Popaju = exposed;
exercising moderate as
i of exercising mild asthmatics, Pop^j = exposed population of
tics and Cases = number of occurrences of at least moderate symptoms.
--4-~ The size of the relevant exposed population is also needed. The number of asthmatics in the
population is estimja|il to be four percent (US EPA, 1994a). The percent of asmmatics exercising at a
giveatimeisuniceltain. US EPA (1994a) presents several statistics of studies on bom the general
population and specific asthmatic populations regarding the percent of time devoted to exercise. The
value used hi the US EPA (1994a) analysis was 1.7 percent of waking hours, based on a study of the
general population. Estimates from studies specifically on asmmatics range from 0.2 to 3.3 percent We
use the point estimate of 1.7 percent and the range of values as the probability of a person hi the
asthmatic population exercising during a given hour. An additional piece of information is needed
regarding the relative proportions of mild and moderate asthmatics hi the population. This analysis
assumes that two-thirds of the asthmatic population has a mild form of the disease, and one-third are
moderate asthmatics.
210
-------
I
CO
I
I
8
5
S
I
* 3
Wl
ta
>- *
- i
> 1
- s
«
, I
I-1. i
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i
2
3
4
5
6
7
a
9
10
11
12
.13
14
IS
16
17
II
19
20
Visibility: Selection and Use of Concentration-
Response Functions
This analysis estimates benefits due to unproved visibility in two ways. First, die change in
visibility, measured in deciview, captures the physical change in atmospheric conditions attributable to
emissions reductions achieved under the CAA. Second, the direct economic valuation of monetary
benefits translates the estimated visibility improvement to dollars based on a willmgnes^ to pay estimate
for a unit change in visibility. ,1--- • «/7 /-., 5&?:_
Deciview represents a visibility measure useful
visibility across a range of geographic locations
other common visibility measures, visual range (measured in tan)
'); however, it characterizes visibility hi terms of perceptible changesrii
As noted in Table 75, two separate sources provided visibility data,
the southwestern U.S. In the east, the Regional Acid Deposition
coefficient estimates for each of 1,330 grid cells in the ItADM domain (<
*^ „.._- f~ = ^ - =1.3-= *
country). A conversion was necessary to translatejie
western U.S., visibility conditions, measured in deciview il
model and a conventional extinction budget approach (SAX 1
Physical benefits in terms of
weighted deciview change. These
benefits are likewise computed for
on
It is directly related to two
extinction (measured hi km*
ndent of baseline conditions.
-eastern U.S., and one for
generated extinction
the eastern half of the
deciview terms. In the
using a linear rollback
measured in terms of a population-
Is are computed sepfiitely for the east and west Monetary
212
-------
•8
r
I
S
I
i
on
!l
1
.S c*
» «
aS
, *•'
*
> •'
> --'
a
4f! I
&>> s
5*5 S.
il -
I f
' '
.a
.s
3
S
1
s
09
I
.a
.9 o •?
_, en 2
1-1 •
.a |.a
1-^5
irr
tn
5 a
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Health Effects Model
2
3
4
5
6
7
S
9
10
11
12
13
14
13
16
17
IS
19
20
21
22
23
24
25
26
27
2S
29
30
31
32
Section 812 of the 1990 Clean Air Act Amendments charged EPA with assessing the costs and
benefits attributable to the original 1970 Clean Air Act (CAA) and its subsequent amendments. In order
to prepare such an assessment, EPA identified and estimated the quantifiable health and welfare benefits
enjoyed by Americans due to improved air quality resulting from me CAA. Health benefits resulted
from an avoidance of air pollution-related health effects, such as mortality, respiratoryillness, and heart
disease. Welfare benefits accrued where improved air qualhyaverted damage to measurable resources,
for example agricultural production (these are described in Appendix F) and visibility. This section .
describes the analysis of health effects in terms of its three major components:
• Air Quality - estimated changes in exposure concentrations of criteria air pollutants between the
factual scenario with the CAA controls and the hypothetical"4no-con!rol"icenario without the CAA. Air
quality improvements resulting from the Act were evaluated hi terms of tx>& their temporal distribution
from 1970 to 1990 and as their spatial distribution across the United States, =
*_ J ~ "-•;__
• Population Distribution - determined the population exposed to the different levels of air quality
improvement This represents a necessary step towards quantifying ^ulpiange in health effects
attributable to CAA control.
Health and Welfare Effects •
tified the relationship between exposure to a given level of air
> associated with such exposure.
Results of most of theja||pses were geii|i§iel,'ui terms of bom point estimates of benefits
(avoided incidences and conejS|Kai|l|jlg monetary values) as well as a probability distribution of these
estimates based onMpnte Qirlbiu|ft(ia|aty analysis. The following sections describe the three elements
of the benefits analysispynore
Air
.
:X The first step towaiSs quantifying benefits attributable to the CAA involved characterizing air
quality improvements for the following pollutants:
ite matter, less than 10 microns in diameter (PM,0)
(03)
Nitrogen dioxide (NOj)
Sulfur dioxide (SOj)
Carbon monoxide (CO)
Light extinction113
Lead(Pb)
113 While light extinction is not a criteria air pollutant, it provides an important measure of a significant welfare
effect resulting from air pollution, visibility degradation. Light extinction results from light scattered by fine
particles in the atmosphere, especially sulfates and ammonium nitrates. As .atmospheric concentrations of such
particles increases, light is attenuated and visible conditions decrease.
214
-------
_ Appendix D: Human Health Effects and Visibility Effects of "Criteria Pollutants
i Generally, air pollution monitoring data provided baseline ambient air quality levels for the
2 factual control scenario. Air quality modeling was used to generate estimated ambient concentrations for
3 the hypothetical (no-control) scenario. With the exception of lead data, air quality modeling data were
4 provided by EPA' s Office of Air and Radiation. A variety of modeling techniques were applied,
5 depending on the pollutant modeled, requiring different applications of the air quality data to the benefits
e . analysis, as described below. ; ' '
7 Data Coverage .
9. PM10 air quality estimates were provided at the county level, synthesizing historical ambient
w measurement data collected throughout each monitored county. Annual data for each of the two
/; scenarios (control and no-control) were provided from 1970 to 1990*: These air quality estimates were
12 comprised of daily averages and annual means for each monitored county (measured in ug/m3). In lieu
13 of modeling daily air quality data for each monitor nationwide over two decades, annual concentration
u profiles were reduced to frequency distributions. That is, the daily averages for * given year were
is reported as a single distribution, which, for PM,te was summarized using the pollutant levels at every
le fifth percentile of that distribution. While these descriptive summaries limited analysis of air quality
n changes to the single day level (i.e., it was not possible to determine pollutant concentrations for
u sequences of days), they eased the computational burden and data storage requirements tremendously.
-•1§I .":/' ^
» Two different measures of ambient concentrations of particulate matter were used in the United
3 States during the period 1970 to 1990; Prior to .1387, the indicator for the National Ambient Air Quality
21 Standard for PM was total suspended particulalesjflOS^ jfa 1987, the indicator was changed to PM10
22 (particles less than 10 uM mjjji^afisr). Widespnpj PM10 monitoring did not begin until 1 985; prior to
23 that only TSP data is available, Jtoia|j,se die majority of the PM concentration response functions used in
24 this analysis are in terms of PM^i^tt^ta is preferred, if available. Where only TSP is available, PM10
25 concentrations were estimated using^iifti^TSP ratios that vary by area of the country and for urban/rural
26 adjustments.
27 In order to compute an- iguality improvements relative to PM10, differences were taken between
2» the distributions for the cootrjal|i^ no-control scenarios at the 20 intervals provided. The result was an
29 approximation of the distnTstrtkM of PM,0 concentration changes at the given monitor, reported as 20
jo concentrations along the probability density function.
=.
r->2P '
31 In addition to JpodelmgPM10 concentrations in counties for which ambient monitoring data was
32 available, EPA alapnodeled concentrations in unmonitored counties using large-scale regional models.
u Extendmg the coverage of PM,0 monitors to counties beyond the monitor network decreased the
34 confidence associated with the estimates but at the same time extended PMIO exposure estimates to a
15 significantly greater proportion of the U.S. population. This issue, is discussed further hi below.
*>
37 Ozone, NO2, SOj, and CO
>
33 Ozone, NOj, SOj, and CO concentrations for both the controlled and uncontrolled scenarios
39 were provided for the locations of air quality monitors. Throughout the U.S., measured ambient
3 concentrations and modeled no-control concentrations were reported for a variety of averaging times
4i ranging from one-hour to yearly. With the exception of ozone, estimates of pollutant concentrations for
42 these monitor-level pollutants were provided for 1970, 1975, 1980, 1985, and 1990. Ambient ozone
43 concentrations for both scenarios were provided at each monitor annually from 1970 through 1990.
215
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i
2
3
4
3
6
7
S
9
10
n
12
13
14
IS
16
17
IS
19
20
21
23
24
25
26
27
29
30
31
32
33
34
35
»
37
38
39
40
Similar to PMIO, profiles of measured and modeled concentrations at each monitor were reduced
to descriptive statistical summaries of the concentration distributions. From the original concentration
profile for a given monitor during a given year, distributions were generated for a variety of useful
averaging times (e.g., 1-hour, 8-hour, 12-hour, daily). For each averaging time, a frequency distribution
of pollutant concentrations was developed. A statistical function was fit to the distributigJB%hich was
then summarized by two or three fitting parameters (typically the moments ofjhf distribution). For the
pollutants described here, the distributional form used was almost always theja|oma (one was log
normal), which is commonly used to approximate air quality observations, l^jptimeters are used to
describe the gamma distribution, reducing the amount of data required to describe a year's worth of hour
concentrations to two values. From these parameters, the approximate distributions can be rebuilt to c
estimate the changes in health or welfare effects.
For most analyses, translating the reported fitting panuneteia^uito air quality changes attributable
to the CAA required a transformation of the profile panunetereffc eaib^Mthe two scenarios (control and
no-control) back into probability density functions. Similar to PMp^||psrence between the two
distributions was evaluated at five percent intervals, yielding a disn-ibatf^^&su:entration changes
resulting directly from enactment of the CAA. For tite agriCWtimil welfal|§|||ysis, an inverse
cumulative density function was used to calculate jcffiM^^
froyii the ffflmmfl distnDutioiis. From these u8tflj«^fr lAnff^ftffHi xsctniift itnocx^nft^ C&D DC used m
plant response was calculated. Because ozone conciliations
determine crop yield loss, and the growingjieasons Yfffby cropi
used separate ozone profiles in two mo|||plocks thj6ughout
Light Extinction
• BaiHipdata
Regional
at five year
scenarios are
growing season is used to
the agricultural analysis
for 60 60 kilometer gridcells modeled by EPA's
Extinction coefficients, measured in I/meters, were reported
Only annual mean extinction values for each of the two
on changes in visual range.
Determining the iii||i||pPihe. CAA on atmospheric lead concentrations was handled differently
than the other pollutants. || cellrast to using ambient measurements for the control case and modeled
source of lead emissionsjfemissions from lead in gasoline (essentially an area source due to its former
widespread use in mo|pr fuels) were treated differently man point sources of atmospheric lead. EPA
estimated the leaditent of gasoline, by far the dominant source of lead emissions, for the period 1974
m mis case emissions estimates (based oh gasoline content) rather than air quality
data were used to measure the impact of CAA control. Leaded gasoline was assumed to have remained
in use in the no-control scenario. Lead emissions from point sources (combustion sources, electric
utilities, and industrial processes and boilers) were obtained from a variety of EPA sources for 1990.
Growth and emissions factors were used to estimate emissions back to earlier years. Air dispersion
modeling of each point source was used to quantify ambient concentrations hi the vicinity of these
sources over the 20 year period of the study,
216
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
With respect to the distribution of air quality
data across the two decades considered, it should be
noted that both the number and location of monitors
tracking air quality changed over time. Table 76
depicts the number of monitors for each pollutant
across the period of this analysis. The number of .
monitors generally increased throughout the 1970s
and leveled off or declined at varying points during
the 1980s, depending on the pollutant
Table 76. ' Criteria Air Pollutant Monitor* la tbe
Pollutant
aa*
y»
10
u
.12
13
14
IS
16
17
19
20
21
22
23
24
26
27
U
29
30
31
32
33
.34
Population Distribution
Health and some welfare benefits resulting from air qualityimprovements are distributed to
individuals in proportion to the reduction in exposure each enjoys. Predicting individual exposures, then,
is a necessary step in estimating health effects. Doing so for the Section 812 analysis required not only
an understanding of where air quality improved as a result of the CAA, but also how many individuals
were affected by varying levels of air quality ^prov^mients: ^lui^a critical component of the benefits
analysis required that the distribution of the U.S. population nation
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i
2
3
4
J
6
7
S
9
10
U
12
13
14
13
It
17
IS
19
20
21
22
23
24
23
26
27
21
29
30
31
32
33
34
33
36
(1) Air quality improvements (difference between control and no-control scenarios) were applied to
individuals living in the vicinity of air quality monitors. For pollutants with monitor-level data, it was
assumed that the individuals in a gridcell were exposed to air quality changes estimated at the nearest
monitor if it was within SO kilometers. Likewise, for PM,0 (for which data was available^ the county
level) the population of each monitored county was assumed to be exposed to the air quality changes
reported for that county.114 The remainder of the population was excluded from the analysis.
Unfortunately, by limiting the quantitative
analysis to populations within SO km (or within a
monitored county, for PM), a significant portion
of the U.S. population was left out of the analysis
(see Table 77). For most pollutants in most years
(excepting lead), less than % of the population •
lived within SO km of a monitor (or within a PM-
monitored county). Clearly, an analysis mat
excluded 25 percent of the population from the
benefits calculations (thus implicitly assuming
that the CAA had no impact on mat population)*
would understate the physical effects of me CAA.
Conversely, ascribing air pollution reduction
benefits to persons living great distances fiom air
quality monitors is a somewhat speculative
exercise, and could overstate
if. -
JtiMT
70.4*
,712*
».**
too*
*»*
*M»ttC % 4Juk^s
tOO* luftUt
(2) Air qiiality improvementsj^m'japplied to ____„_.,..
ahnost all iodtvMuals nati6a^|&§P|^^ie monitor::™
data was not ayailalde within 50ptetrl^«rs, data
from the closest monitor, regardless ""
was used. SunilarJI^|ii^^^icentn
extrapolated using i^ioiudjdr1;quality models to
all countks (even those fbf^i|^^monitoring data
was unavailable) and applied ililie populations of
those counties. -•
Although subject*) less certain air quality
data, the second alternative extrapolates pollutant
exposuw estinuaeflo ahnost the entire population
usingithe dosesftmonitoring data available (see
*
^^, %^ JJBt. ••' Vtftft HOT.
toom
114 Since the lead (Fb) analysis, winch was handled sq»mtelyfiom that of *e otfaer criteria poUutmts, did not itquiie air quality
modeling data, the Une of proximity to monitors is irrelevant The Pb analysis exteuled to 100 percent of the population.
111 While this alternative captures the yast majority of the U.S. population, it does not model exposure for everyone. To improve
computational efficiency, those gridcells with populations less than 1,000 were not modeled; these cells account for less than three percent of
the U.S. population.
218
-------
2
3
4
3
6
7
I
9
10
11
12
13
14
IS
16
17
II
'9
0
21
22
23
24
25
26
27
21
29
30
31
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Health And Welfare Effects
Benefits attributable to the CAA were measured in terms of the avoided incidence of physical
health effects and measured welfare effects. In order to quantify such benefits, it was necessary to
identify concentration-response relationships for each effect being considered. As detailed above, such
relationships were derived from the published science literature. lathe case ^healm effects,
concentration-response functions combined the air quality improvement and pf^jation distribution data
with estimates of the number of fewer individuals that contracts disease per unit Change in air quality.
By evaluating each concentration response function for every gridceU throughout the country, and
aggregating the resulting incidence estimates, it was possibteto ^generate national estimates of avoided
incidence. ' . . ' K "!"-_ '
' It is impossible to estimate all of the physical effects that would J»ve occurred without the Clean
Air Act While scientific information is available that makes it possible to estimate certain effects, many
other, potentially very important, health and welfare effects cannot be estimated at this time. Other
physical effects can be quantified, but it is impossible to assess the economic^vame of those endpoints.
Table 79 shows the health and welfare effects for which [quantitative analysis has been prepared, as well
as some of the health effects that have not l>een quantifiedmttoe aa^
Health effects for lead were treatedjslightly differently. Jkgam, gasoline as a source was
addressed separately from conventional|>dint sources. Instead-If using ambient concentrations of lead
* * JJsMjnfe.- --=.;=- -~- ^
resulting from gasoline, the o>n(^ntntsure to pomt sources of
atmospheric lead were (X7nsidered^agme air concentration distributions modeled around point
sources. Conoentration responsely||ion& were used to relate changes in blood lead levels hi nearby
populations.
Welfare
the concentration-response
in a similar manner to health effects, with one major difference in
used. Instead of quantifying the relationship between a given air
quality change and the num||r Ifcases of a physical outcome, welfare effects were measured in terms of
the avoided losses, such asieduction hi crop losses due to unproved ozone conditions. Welfare effects
quantified as part of this analysis included household soiling damage by PM10, visibility degradation
measured by light extinction, and agricultural benefits measured hi terms of net surplus.
219
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Ifcfefe m.
-.-- %
OtBer
Matter/
finis tttttetaat v
s
Uttro. "•
IQ tow effott(oe lifetime
' ..*.' ft.T
220
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
10
u
12
13
14
15 •
It
17
II
19
21
22
27.
Health Effects Motto/ Results
Tables 81 and 82 summarize the results of the health (and visibility) effects modeling. For some
health endpoints, more than one CR function was used, with each being based on a different health effects
study. The alternative CR functions provide differing measures of the health effect. These can be used to
derive a range of possible results. In the case of lead (Pb), alternative CR functions were not used; rather,
two analytical procedures were used (labeled the "backward-looking" and "forward looking" analyses),
giving a range of results for most Pb endpoints. '•*- ^ -~-::~=IJ> -T". '
• -*" -5P" " '''''-^--'--:=' -
Table 81 presents the results of the "50 km" model run. Observeiiat, for moseiteaith endpoints
where the pollutant of concern is PMIO only, incidence declines between 1985 and 1990. Only two factors,
besides the CR function itself, should affect the model's output; Tlie difference hi ambient air quality
between the two scenarios, and the affected population. Between J985^ad 1990, difference in air quality
between the scenarios increased, and the U.S. population increased. One should see, men, an
unambiguous increase in incidence between 1985 and 1990, The decrease in incidence observed here is
due exclusively to the decline in population residing in counties with PM10 monitors (i.e., due to the
decrease in the number of monitors). ' ,. \--~- --" . :;4.
Table 78 presents the results of the "all U.S. population" mpdetpih (although, with the exception
of Pb, not all of the population is modeled, with up to 5 percent being: excluded hi a given year), and
provides a more accurate depiction ofti&jjMern of health effects incidence across years. The accuracy of
the scale of incidence is less certain, /ffiese resultsjjire almost certainly more accurate than the "50 km"
results, but rely on the assumption tjjfr(for a pcff|pn of the population) distant air quality monitors provide
a reasonable estimate of local j^gaJBty conditij^^Eho^ me results presented here are somewhat
speculative. Ills likely that^fa^icted healt^ejle«ts are overstated for mat population group (20 to 30
23 percent of tola! population in
24 an implied zero health impact
fiie|£PM18) for which distant monitors are used. Conversely, mere is
of the population (3 to 4 percent hi the case of PM,0)
excluded from the aaal^altogeme^jpfplrstatement of health impacts for mat group.
for several health effects which could be
quantified but not monetized fo| flu* analysis.
* - ":H-"r~i~_— J^jci^^lgliig-e »
221
-------
^.
C-l
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
s\ '* s ,v», JfV'4^-
% \Mt.-J* A IV. X jS*.
w,•^^ '^ •: '""•v "«" "*s. •»• f -. x ^vsjsy% ^rsr •vcfvSyy»w T*T"VWA vj %^ -*. •>.< • s
, -\
^'Y, ^mt'^'^m^^^*^" <'">-'
223
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
tablet*
«*»*<»
JiBtt V
mm-
30JSSF
454*7
4M1*
Ufa"
J*
1%
JSr
&,w &m
,^,«ws^ " tssstt-
1W»
M»
•Jft^ /s -wars-.
U.
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Other Result*
Additional analyses were conducted in an effort to quantify effects resulting from long-term
exposure air pollution: premature mortality due to chronic exposure to PM,0 (where the CR function was
inferred from "cohort" mortality studies), chronic bronchitis from long-term PM^ exposure, and chronic
asthma due to ozone exposure. The results of these analyses are found in Table 83 As with the health, '
effects presented above, results for both the "50 km" model run and the "all U.S." model run are
presented, *_ . * , ;--
Suit**
ms
tittfitt
JMJM
..mm.^ mm
.-»**
•fc-fcfci dtt 1'tta
<•','< 'V ,',|?, ^-.',v '•?•*" x •*, :
Health Modeling Reference*
9
10
11
n
13
14
Abbey, DJE.,, LebjppNz, MJ)., Mills, P JL, Peterson, F.F., Beeson, W.L., and RJ. Burchette. 199Sa.
Long-term ambient concentrations of particulates and oxidants and development of chronic
disease in a cohort of nonsmoking California residents. Inhalation Toxicology. 7:19-34.
Abbey, DJE, Petersen, F., Mills, P.K., and W.L. Beeson. 1993. Long-term ambient concentrations of
total suspended particulates, ozone, and sulfur dioxide and respiratory symptoms in a
nonsmoking population. Archives of Environmental Health. 48(1): 33-46.
225
-------
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i Allied EN, Bleecker ER, Chaitman BR, Dahms TE, Gottlieb SO, Hackney JD, Pagano M, Selvester RH,
2 Walden SM, Warren J. 1989a. Short-term effects of carbon monoxide exposure on the exercise
i performance of subjects with coronary heart disease. New England Journal of Medicine
4 321:1426-1432.
5 Allred EN, Bleecker ER, Chaitman BR, Dahms TE, Gottlieb SO, Hackney JD, Hayes D, Pagano M,
6 Selvester RH, Walden SM, Warren J. 1989b. Acute effects of carbon monoxide exposure on
7 individuals with coronary heart disease. Cambridge, MA: Health Effects Institute, research
« ' report no 25. _;"" .-."-.""'
9 AUred EN, Bleecker ER, Chaitman BR, Dahms TE, Gottlieb SO, Hackney JD, Pagano M, Selvester RH,
10 Walden SM, Warren J. 1991. Effects of carbon monoxide on myocardial ischemia.
u Environmental Health Perspectives 91:89-132. i- ->.
12 American Heart Association. 1991. 1992 Heart and Stroke Facts.
""-"^'"'^ftf*^ '
13 Braun-Fahrlander, C., U. Ackermann-Liebiich, J. Schwartz, HJP. Gnehm, M Rutishauser, and H.U.
14 Wanner. 1992. Air pollution and respiratory symptom
a Dis. 145:42-47. , Effect of Ozone Associated wim
20 Summertime Photochern]cJ!|Smog on |fii|||i|uency of Asthma visits to Hospital Emergency
21
22 Crocker T.D., HiQiil-i^Jr. 198ltl|||, ^work, labor productivity, and environmental conditions: a
23 case stud)?; f-I^Rjfjnr of Ecacs and Statistics 63:361-368.
24 Dockery, D.W., F.E. Speii^K^||trani, JJL Ware, J.D. Spongier, and B.G. Ferris, Jr. 1989. Effects
2i of inhalable particliiii^piratory health of children. Am. Rev. Respir. Dis. 139: 587-594.
"!-""K^ IPsra "T
=5 * . , •
26 Dqckery, D.W., J. Schwartz, and JJ>. Spengler. 1992. Air pollution and dairy mortality: associations
27 : with particulates and acid aerosols. Environ Res. 59:362-373.
-- ' • •
21 DockeryrD.W,,etiiL 1993. An association between air pollution and mortality in six U.S. cities. The
29 New England Journal of Medicine. 329 (24): \ltt-\l 59.
x Empire State Electric Energy Research Corporation (ESEERCO). 1994. New York State Environmental
31 Externalities Cost Study. Report 2: Methodology. Prepared by: RCG/Hagler, Bailly, Inc.,
a November.
u Fairley, D. 1990. The relationship of daily mortality to suspended particulates in Santa Clara County,
34 1980-1986. Environ. Health Perspect. 89:159-168.
226
-------
_^__ Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
.1 Hasselblad, V. Eddy, D.M., Kotchmar, D.J. 1992. Synthesis of Environmental Evidence: Nitrogen
2 Dioxide Epidemiology Studies, J. Air Waste Mgmt. Assoc. v. 42,662-671.
3 Ito, K., P. Kinney, and G.D. Thurston. 1995. Variations in PM,0 concentrations within twjimetropolitan
4 areas and their implications to health effects analyses. Inhalation Toxicol., in press.
s Kinney, P.L, H. Ozkaynak. 1991. Associations of daily mortality and air pollution in Los Angeles
6 County. Environ Res. 54:99-120. , =-> ->-s
7 Kinney, Pi, H. Ozkaynak. 1992. Associations between ozone and daily mortality m Los Angeles and
« New York City. Am Rev Respir Dis. 145: A95. :: : :-::;%':;
o .•• - _ J__ r =_'-„..••"
---"_" If -=: *" — .. ~-~J- -"i--"-
9 Kinney, P.L., K. Ito, and G.D. Thurston. 1995. A sensitivity analysis of mortality/PM10 associations in
10 Los Angeles. Inhalation Toxicology 7: 59-69. ; .; :=.
•* '-"-"__ ~- ~ ~ —„
// Krupnick A J., Harrington W., Ostro B. 1990. Ambient Ozone and Acute Health Effects: Evidence
12 from Daily Data. Journal of Environmental Economics and Management 18:1-18.
' - -""""„ ' i_~jr~
13 Manuel, E.H., Jr., Horst, RJL., Jr., Brennan, KJ^,l^me%^O€iJDu^ M^, and J.K. Tapiero. 1982.
14 Benefits analysis of alternative secondary national amb^t air quality standards for sulfur
15 dioxide and total suspended paitkntotes, Volumes I-IV. prepared by U.S. Environmental
16 Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC.
i? [Cited rnESEERCO, 1994].sfSir ^ A?
H McClelland, G., Schulze, W., Waldman, DL, Irwin, J., Schenk, D., Stewart, T., Deck, L., and M. Thayer.
19 199Jl£|i^Jumg East^ii^^ Prepared for
.20 l£|g|ii|^ D.C. [Cited in ESEERCO, 1994.]
21 Melia R.J.W, Florey C du V, B J)., Brooks A.G J., John H.H., Clark D:, Craighead
22 IS., aDdXj0^jjOL^1980. Tiii»iation between indoor air pollution from nitrogen dioxide and
23 respirgtoiyTIli^ 16:7P-8P.
24 Melia RJ.W., Florey C du V, Morris R.W., Goldstein B.D., John H.H., Clark D., Craighead I.B., and J.C.
25 ;;: Mackinlay. 1982. ^hiidhood respiratory ilhiess and the home environment: IT. association
26 jf between respiratory illness and nitrogen dioxide, temperature, and relative humidity.
27 "> International Jownal of Epidemiology 11:164-169.
is Mel^ RJ.]^^ FJU»pifC du V, and Y. Sittampalarm. 1983. The relationship between respiratory illness
29 minlantsluid gas cooking in the UK: A preliminary report, in Air Quality Vth World Congress:
30 Proceedings of the International Union of Air Pollution Prevention Associations. SEPIC
31 (APPA), Paris France, pp. 263-269.
32 Ostro, B.D. 1987. Air pollution and morbidity revisited: a specification test J. Environ. Econ. Manage.
33 14:87-98.
227
-------
3
4
S
6
7
I
9
10
11
12
13
14
IS
16
n
It
19
20
21
22
23
24
25
26
27
21
29
30
31
Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
Ostro BD, Rothschild S. 1989. Air Pollution and Acute Respiratory Morbidity: An Observational Study
of Multiple Pollutants. Environmental Research 50:238-247.
Ostro, BD., MJ. Lipsett, M.B. Wiener, and J.C. Seiner. 1991. Asthmatic responses to airborne acid
aerosols. American Journal of Public Health 81:694-702. -;^'
Ostro, BD., MJ. Lipsett, J.K. Mann, A. Krupnick, and W. Harrington. 1993p Afcpollution and
respiratory morbidity among adults in Southern Califoi^^ American ^turnal of Epidemiology
137:691-700.
Ostro, BD., J.M. Sanchez, C. Aranda, and G.S. Eskeland. 1995. Air pollution and
study of Santiago, Chile. J. Exposure Anal. Environ, £ffi(^idl., submitted.
Ozkaynak, H., J. Xue, P. Severance, R. Burnett, and M. Raizenni;SW|iy| jAssociations between daily
mortality, ozone and particulate matter air pollution fa Torontp^£inada. Presented at
. • " • . . • - -"-_"- =—-aj- _-_ _„.
Colloquium on particulate air pollution and human mortality andjmoi$iidiry; January, Irvine, CA:
University of California at Irvine, Air Pollution Health Effects Laboratory; report no. 94-02.
of a
Pope, C A., m. 1991. Respiratory hospital admissions
Lake, and Cache Valleys. Arch. Environ. Health.
Pope, C.A., m, and D.W. Dockery. 1
asymptomatic children.
Pope, C .A., m D.W.
pollution; a daily
10 pollution in Utah, Salt
Pope, C A., in, Schwartz, J., a
Environ. Health V; 211-
Pope,
Acute health effectslif PM,0 pollution on symptomatic and
145: llil-1128.
i. 1991. Respiratory health and PM10
ir. Dis. 144:668-674.
som. 1992. Dairy mortality and PM10 in Utah valley. Arch
. Partii
adults. Am.J.
D.W. Dockery, J.S. Evans, F.E. Speizer, and C.W. Heath,
as a predictor of mortality in a prospective study of U.S.
Med 151; 669-674.
Portney P.R., Mullahy J. =1986. Urban Air Quality and Acute Respiratory Illness. Journal of Urban
' JT Economics 20:2
PortBsy»PJL aod|||Mullahy. 1990. Urban air quality and chronic respiratory disease. Regional Science
and Urban'iconomics. 20:407-418.
Samet, J.M., Lambert, W.E., Skipper, B J., Gushing, A.H., Hunt, W.C., Young, S A., McLaren, L.C.,
Schwab, M., and J.D. Spengler. 1993. Nitrogen dioxide and respiratory illnesses hi infants, Am
Rev. Respir. Dis. Vol. 148, pp. 1258-1265.
32 Schwartz, J. 1991. Particulate air pollution and daily mortality in Detroit Environ. Res. 56:204-213.
228
-------
. Appendix D: Human Health Effects and Visibility Effects of Criteria Pollutants
i Schwartz, J. 1991/1992. Particulate air pollution and daily mortality: a synthesis. Public Health Rev. 19:
, 39-60.
3 Schwartz, J. 1993a.. Air pollution and daily mortality in Birmingham, Alabama. Am. J. Epidemiol. 137:
•4 1136-1147. . r
j Schwartz,!. 1993b. Particukte air pollution and chronic respirato/y disease. Environmental Research.
£f\ *7 1 ** =E i" --_ = -- -a~ — ,-."_-
6 . 62:7-13. J---&* '-'-'•_- -" •
? Schwartz, J. 1994a. Air Pollution and Hospital Admissions in Elderly Patients in Birmingham, Alabama.
i American Journal of Epidemiology 139:589-98. -1. -f~ - 5
" -' — ^^. _ "»_"-" ?
9 Schwartz, J. 1994b. Air pollution and hospital admissions for the elderly in Detroit, Michigan.
10 American Journal of Respiratory Care Med 150:648-55*- - "•
20
a Schwartz, J. 1994c. PMIO> Ozone and Hospital Admissions for the Elderty in Minneapolis-St. Paul,
u Minnesota. Archives of Environmental Health 49^:366-374. '^fi^s?
- " " ~S "" "-'
_r_
u Schwartz,!. In press. Short term fluctuations Jn;airrx>llv^onai^ hospital admissions of the
14 - respiratory disease. . ,4" ."-" ""-Ir^^^S^7
" ~
75 Schwartz, J., and D.W. Dockery. 1992a1 farticula&air pollution and daily mortality in Steubenville,
u Ohio. Am. J.Epidemiol. 135;M9. J*f „?
n Schwartz, J., and D.W. Dockery. 1992b. Inci!6l|liCiii||ij% in Philadelphia associated with daily air
78 poltoticto concentration^ Rev. Re^KJ^, 145:600-604.
19 Schwartz, J.,! and & Morris. i9^yir jppllution and cardiovascular hospital admissions. Am. J.
21 Schwartz, J^ P^/BJi|i^^', Pier»id«, T. and J.Koenig. 1993. Particulate Air Pollution and
22 Hospital Emergency Room Visits for Asthma in Seattle. Am Rev. Respir Dis 147: 826-831.
23 Suny^-,J.,M. Saez, C. Mu||lo|t; Castellsague, F. Martinez, and J.M. Ant6. 1993. Air pollution and
24 '73' emergency room admissions for chronic obstructive puhnonary disease: a 5-year study. Am. J.
2s =:, § Epidemiol. 137jpl-705.
26 Thurston, G., Kinirtyj P., Ito, K., and M. Lippmann. 1992. Daily Respiratory Hospital Admissions and
27 : SummorHaze Air Pollution hi Several New York Metropolitan Areas, Journal of Exposure
21 Analysis and Environmental Epidemiology. 2 (4):429-450.
29 Thurston, G. Ito, K., Hayes, C., Bates, D. and M. Lippmann. 1994. Respiratory Hospital Admission and
M Summertime Haze Air Pollution in Toronto, Ontario: Consideration of the Role of Acid
31 Aerosols. Environmental Research 65:271-290.
229
-------
J
6
7
S
9
10
11
12
13
14
IS
16
17
IS
a
20
21
22
23
24
23
26
27
U
29
30
31
32
33
34
3S
36
r=id££- - .__, *• :-""= - - a" B..^"r-E - ~
U.S. Environmental Protection
Oz-One Computer Model (Version
|»| and Standards. Prepared by: Mamtech, Inc., under EPA
Zv^Auoiict
Wan JJL,
cooking, and res\
Disease 129:366-1'
TT^l^**"18-0-**"**• 1984- Passive smoking, gas
of children living in six cities. American Review of Respiratory
-' '- -;- - - - :-v*|j^
"^^^t^^^^^^^^^^-
230
-------
Appendix Eg Ecological Effects of
Criteria Pollutants
introduction
4
3
6
t
t
9
10
11
11
13
14
IS
13
19
20
21
23
23
24
23
26
Benefits to human welfare from air pollution reductions achieved voider the CAA can be r ;
expected to arise from likely improvements in the health of aquatic and terrestrial ecosystems and the
myriad of ecological services they provide. For example, improvements in water quality stemming from
a reduction in acid deposition-related air pollutants (e.g., SOX and NO^ could conceivably benefit human
welfare through enhancements in certain consumptive services such as commercial and recreational
fishing, hi addition to non-consumptive services such as wildlife viewmg^mamtenance of biodiversity,
and nutrient cycling. Increased growth and productivity of U.S. forests could oticur from reduced
emissions of ozone forming precursors, particularly V(JP and NO^ and thus may result in benefits from
increased timber production; greater opportunities fOTrecreaiiojial services such as hunting, camping,
wildlife observation; and nonuse benefits such as nutrient: c^l^Q^-jpHpiestration, and existence
value.
In mis Appendix, the
ifits from CAA pollutant controls are discussed in
the context of three types of ecosystems: aquatic ^and wetland ecosystems and forests. In describing the
potential ecological benefits ofjthe CAA, it is cfc|% recognized that mis discussion is far from being
comprehensive in terms of thetyicf and mag^ie^^blogical benefits mat may actually occur from
the implementation of CAA^Raihfirf this disctttoefeflects current limitations hi understanding and
quantifying tibe linkages whu& cs^ bjtween air qualhy and ecological services, hi addition to
limitations in the subsequent valuali<&a|;|uese services hi monetary terms. This discussion also does
not cover potential benefits from improviapots in other ecological services, namely agriculture and
visibility, which are djacwwed,and quattpbed hi other sections of mis report This appendix is dedicated
to a qualitative evaluat£qiL|^^y^gicai benefits. However, where possible, the existing body of
nan attempt to provide hisights to the possible magnitude of benefits
iprovements of selected ecological services.
scientific literature is drawn
that may have resulted frdnj
27
2*
29
30
31
32
33
34
\nc& of Damages to
tilts From Avolda
Aquatic Ecosystems
Aquatic ecosystems (lakes, streams, rivers, estuaries, coastal areas) provide a diverse range of
services that benefit the welfare of the human population. Commercially, aquatic ecosystems provide a
valuable food source to humans (e.g., commercial fish and shellfish harvesting), are used for the
transportation of goods and services, serve as important drinking water sources, and are used extensively
for irrigation and industrial process (e.g., cooling water, electrical generation). Recreationally, water
bodies provide important services mat include recreational fishing, boating, swimming, and wildlife
231
-------
Appendix E: Ecological Effects of Criteria Pollutants
viewing. They also provide numerous indirect services such as nutrient cycling, and the maintenance of
biological diversity.
Clearly, these and other services of aquatic ecosystems would not be expected to be equally
responsive to changes in air pollution resulting from the implementation of meBCAA. The available
scientific information suggests that the CAA-regulated pollutants that can be most clearly linked to
effects on aquatic resources include SOZ and NOX (through acid deposition and increases in trace element
bioavailability), NOZ (through eutrophication of nitrogen-limited water bodies), and mercury (through
changes in atmospheric deposition). Potential benefits from each of these processes (acid deposition,
eutrophication, mercury accumulation in fish) are described separately in me following sections. ;-_
10
11
12
13
14
15
16
17
II
19
20
21
22
23
24
25
26
27
21
29
30
31
32
33
Add Deposition
Background : -
Acid deposition refers to the depositing of acids (e,g., I^SO4, HNC5jJfitjm the atmosphere to the
earth's surface. Acid deposition can occur in the wet or dry form and can adversely affect aquatic
resources through the acidification of water bodies and waterah1e^|^iification of aquatic ecosystems
is of primary- concern because of the adverse effects of low pHan^lisociated high aluminum levels on
fish and other aquatic organisms. Low pHpan produce direct effasoi on organisms, through
physiological stress and toxichy procesfHiand indirect effects/piediated by population and community
changes within aquatic ecosystems, jlpification can affect many different aquatic organisms and
communities. As pH decreases to ||f species j|||Dess in the phytoplankton, zooplankton, and benthic
invertebrate communities decjB&iieif1' AddMliilPlicmtses in pH affect species richness more
biological*
due to
and communities:
Researchers have
Table 84 presents descriptions of the
pH levels. In evaluating the severity of biological changes
changes is an important consideration; biological populations
improved water quality under certain circumstances.
its through many different experimental protocols,
including laboratory'1»c^||pp^rticuliirly concerning pH, aluminum, and calcium; manipulative whole-
system acidification studi|sM^|Ml|Bld; and comparative, nonmanipulative field studies.
» J. Baker etaL. NAPAP SOS/T 13,199a
•» J. Baker etal., NAPAP SOS/T 13,1990,
232
-------
Appendix E: Ecological Effects of Criteria Pollutants
^ vVw''
"
.. ° % '
+ <
^^ftm^llMllMtlUfrl HHpfi/f/ff/ffiftti * •' -
m»m*tmfo-* '"
«^ ^
'%- % ' ' * • ''\ibL*iAi^^' v""v' ^i
" ' %
v *
sy* '% "* ^
•OWSttWm!' ]ft wMV* (ttttft-
233
-------
Appendix E: Ecological Effects of Criteria Pollutants
14
IS
16
17
II
19
JO
a
22
U
24
a
26
27
U
29
30
31
32
33
Population losses occur frequently because of recruitment failure,121 specifically due to increased
mortality of early life stages.127 Changes at other trophic levels may affect fish populations by altering
food availability.123 Fish in poorly buffered, low pH water bodies may accumulate higher levels of
mercury, a toxic metal, than in less acidic water bodies, due to increased mercury bioavailability. The
primary consequence of mercury accumulation appears to be hazardous levels to humans and wildlife
who consume fish, rather than direct harm to aquatic organisms (discussed further below).
The CAA-regulated pollutants that are likely to have the greatest affect on aquatic ecosystems
through acid deposition and acidification are SO2 and NO,. In me atmosphere, SOa and NO, react to
form sulfate and nitrate particulates, which may be dry-deposited; also the pollutants may react with
water and be wet-deposited as dilute sulfuric and nitric acids. SO2 is considered the primary cause of
acidic deposition, contributing 75 to 95 percent of the acidity in rainfall in the eastern United States
10
11.
12 (NAPAP, 1991).
13 Current Impacts of Acid Deposition
Effects on Water Chemistry
The effects of acid deposition and resulting acidification!^ was intensively studied
as part of a 10-year, congressionally-mandated study of acid raurpipbjim in me United States.124 Based
on the NAPAP study, it is estimated mat 4Jprcent ofjfo lakes apiipercent of tne streams in acid-
sensitive regions of the U.S. are chronically acidic djte to naturafand anthropogenic causes. NAPAP
defines acidic conditions as occumngyiieh the acid neutraliz|ig capacity125 (ANC) is below 0 ueq/L.
is and lakes hi these regions are considered to be
and slightly more than half show some
Furthermore, approximately 20 percept of the
extremely susceptible to acidil^diSned as
susceptibUity to acidification (defined as AN<
In terms ofJhe role of
bodies, it is estima^ Jhajt 75 percel
studied under NAPAP receSvj} their T
Table 85). On a
to result from regional
individual watersheds. F<
by add deposition include|iel
it I _!_ •___... . _ *»Fh!f;= • «*
t as a causal mechanism for the acidification of water
acidic lakes and 47 percent of the 4,668 streams
; source of acid anions from atmospheric deposition (see
importance of acid deposition varies considerably, which is believed
SO, and NO, emissions and differences hi the biogeochemistry of
:es (ANC <0), die regions mat appear most likely to be influenced
;ks and Mid-Atlantic Highland region, with acid deposition cited
as the dominant source ofjfjbidity in 100 percent of me acidic lakes studied (Table 85). This is in stark
contrast to the West reginti, where none of the acidic lakes studied were dominated by acid deposition
(notably, the sample size of lakes for mis region was small to begin with). For acidic streams, the Mid-
Atlantk HigUand region contains the greatest proportion of streams whose acidic inputs are dominated
U1 RoMdmd. 1986, u cited by J.BOeret«l.. NAPAP SOSH-13,1990.
m J. Biker etiL, NAPAP SOS/T 13.1990.
m Mffli et •!., 1987. u cittd by J. Baker tt «!.; NAPAP SOS/T13.1990.
m NAPAP, 1991.
125 ANC Is expressed in units of micro equivalents per Hter (jieq/L), where m equivtlajtb the capacity to neuttalize one mole of H*iow.
Generally, waters with u ANC < 0 have cotresponding pH values of leu than 5.5 (L. Baker et al., NAPAP SOS/T 9,1990).
234
-------
Appendix E: Ecological Effects of Criteria Pollutants
by acid deposition (56 percent). This contrasts with acidic streams of Florida, where the vast majority
(79 percent) are acidic primarily due to organic acids, rather than acid deposition.
4
1
6
7
S
9
10
Table 8& Comparison of Population of A£id^
top*
Dta*Mlat
AcadMia* ***&*
'****"*
**
tfrrtfc,
4
»
' *'
SIW5AMS
,
-------
Appendix E: Ecological Effects of Criteria Pollutants
7
8
9
10
.11
12
13
14
U
16
17
IS
19
JO
21
22
23
24
25
26
27
21
29
30
31
32
33
34
3S
36
37
minnows. About 2 percent and 6 percent of the lakes in the New England region are estimated to be
unsuitable for acid-tolerant and acid-sensitive fish species, respectively. A greater proportion of streams
in the Mid-Atlantic Highland region are estimated to be unsuitable for acid-tolerant and acid-resistant
fish species (18 percent and 30 percent, respectively); however, about 44 percent of streams surveyed in
this region are thought to be heavily influenced by acid mine drainage (Table 85).
Economic Damages to Recreational Fishing
I . ' ' „ ^, .
In an effort to assess some of the impacts from existinghvels of acid deposition to public
welfare, NAPAP investigated the current economic damages associated with acid deposition to trout
anglers of New York, Maine, Vermont, and New Hampshire^rThe general approach used consisted of
linking the catch per unit effort (CPUE) for four species of trout at individual lakes (estimated using
participation survey data) to the relevant water quality conditions at Knese lakes (namely, the acid stress
index or AST). Using historical water quality data, critical water quality conditions (i.e., the ASI values)
were estimated for lakes in the absence of acid deposition and compared conent conditions reflecting the
presence of acid deposition. Using two types of travel cost models, me Jtandp^Utility Model (RUM)
and Hedonic travel-cost model (HTCM), estimates of the willingness to payj[Sp^) per trip of sampled
trout anglers were obtained. Aggregate estimates of die ^pppiere obtamed across the populations of
trout anglers using statistical weighting factors^ ^Finally, the^dipeff&ice injptal WTP between the current
(acid deposition) scenario and the historical (acid o^sition-Iree| jcenarios was determined.
_ __-3 : -==,5^1
The resulting estimates of
relatively small. Specifically, d
dollars) for the hedonic travel
estimates can be considered to muferCiSt imate
limitations precluded me de||^|p|nt of
be a siimiflcant^omponent^i&filhinK in
damages to trout anglers in the four state region are
- - from |p million to $1.8 million (in 1989
modejC respectively. By many accounts, these
ates
exclusion of a^laige population oft
economic damage equates were limj
economic damages to anglers fishing
NAPAP analysis was pertonned in the 1
important welfare imp
non-usevalues of lakes in j
to anglers in these states. First, data
estimates for brook trout anglers, which may
Second, resource constraints necessitated
(i.e., those residing in New York CHy). Third, the
put anglers, thus excluding potentially similar if mot greater
• coldwater or warmwater fish species. In addition, the
; recreational fishing in lakes, thereby excluding potentially
al fishing in streams. Finally, these estimates do not address
Benefits From Acid Deposition Avoidance Under the CAA .
- * "-" "- -=
^L1--. - .Jr ' • '
. It is cunenti^estimated mat in the absence of pollution reductions achieved under me Clean Air
Acl, total sulfur emissions to the atmosphere would have increased by nearly 16 million tons by 1990, a.
40 percent increase above 1990 levels estimated with CAA controls remaining in place.127 Based on
atmospheric transport and deposition modeling, this increase in sulfur emissions corresponds to an
approximate 25 to 35 percent increase in total sulfur deposition (wet & dry) in large portions of the
northeastern portion of the United States.12* Given sulfur emission and deposition changes of this
07 U.S. EPA, 1995; Tabte B-2.
"* U.S. EPA 1995, pages 3-10.
236
-------
Appendix E: Ecological Effects of Criteria Pollutants
9
10
11
12
13
14
IS
16
17
IS
19
21
22
23
24
25
26
27
21
29
30
magnitude, and the importance of sulfur emissions in contributing to acid deposition; one would expect
some benefits to human welfare to be achieved as a result of unproved quality of aquatic ecosystems. To
date, however, no formal benefits assessment of CAA-avoided acid deposition impacts has been
conducted for aquatic ecosystems. Nevertheless, past benefit assessments involving acid deposition
impacts on aquatic ecosystems provide some opportunity to gain insights into the relative magnitude of
certain aquatic-based benefits that may be achieved through pollution reductions under the CAA.129
Recreational Fishing ;'_;.? ':-/.•
NAPAP evaluated the impact of changes in acid deposition on use values of aquatic ecosystems
(i.e., recreational fishing).130 In their integrated assessment, NAPAP valued the impacts of three different
sulfur-induced acid deposition scenarios to trout anglers from NY, W, NH and ME.131 The three
scenarios evaluated were: ~-_
1. No change hi acid deposition, ' ,/-. v"
2. a SO percent reduction in acid deposition, and " - r5- \ -. -, '.- _
3. a 30 percent increase in acid deposition. """"-=,-
As described above, equations were developed by NAPAP to estimate the catch per hour for species at
each lake as a function of the ASI value for each kike and of the technique of the fishers. Baseline and
predicted changes in CPUE were evaluatedibr all kikes modeled in the region. Willingness-to-pay
estimates for CPUE per trip were derived for the baseline and sulfur emission scenarios using two travel-
cost models, a random utility model and a hedonic travel cost model. These willingness-to-pay estimates
were then combined with the results 0f a participation model mat predicted the total number of trips
taken by trout anglers. Total welfare changes were determined over a SO year period (from 1990 to
2040). .'
At current:Ife||Jls of acid depjiftion, NAPAP estimates that trout anglers in these four states will
experience amiual lk)ss^|y the yearllO|Q of $5.3 or $27.5 million (in 1989 dollars) for the random
utility model and hed6fflc||i^e^ cost mfldel, respectively (see Table 86). If acid deposition increases by
30 percent, which roug1i]^g|^px>ndst^ me 25 to 35 percent increase predicted for the northeast U.S. in
the absence of CAA sulfuljC|aflte|||t"2 the resulting economic losses to trout anglers hi 2030 would range
from $10 million to nearlyi||^|irMon annually (in 1989 dollars) for the RUM and HTCM,
respectively. If deposition decreases by SO percent, annual benefits to recreational anglers are estimated
to be $14.7 million (RUM) or $4.2 million (HTCM).
"> See, for example, NAPAP, 1991.
110 NAPAP, 1991.
111 NAPAP, 1991; page 383-384.
111 U.S. EPA 1995.
237
-------
Appendix E: Ecological Effects of Criteria Pollutants
i-
2
3
4
S
6
7
»
9
10
11
12
13
14
15
16
17
It
19
20
21.
22
23
24
23
26
27
21
29
30
31
32
33
34
33
36
Table 86.
8fl^8fiMttt-ll$6
*»*4*»ifwl**i*-;,V' li-"*"* *
:«^ttiffi«
"'J^IK
-'" HUGH
HTCM
f
While an estimation
of CAA-related benefits to
trout anglers based on the 30
percent increase in acid
deposition scenario has some
appeal, a strict transfer of
these benefits to the 812
retrospective analysis is
hindered by several factors.
First, the NAPAP benefits
estimates are projected for
• future conditions (the year
2030). Therefore, the extent
to which the NAPAP benefits
reflect conditions and benefits
in 1990 (the focus of the 812
retrospective assessment) is
unclear. Second, the NAPAP
and CAA 812 analyses
operate from different baselines (1990 for the NAPAP study versus 1970-1990 for the 812 study).
However, the NAPAP estimates of annual benefits of=$10 to $10|iduiion provide a rough benchmark for
assessing the likely magnitude of the avoided damages to an important and sensitive recreational fishery
hi a four state area most impacted by^spflce water acidificat^ form atmospheric deposition.
Eutrophlcatlon
Eutrophicaiiettis me
most common nutriotts JBVolved i
species). When
occur from excessi
biomass. Under highly
resuh in subsequent loss
")
aquatic systems respond to nutrient enrichment The
tioh are nitrogen and phosphorous (and related chemical
amounts of nutrients, adverse impacts on men- health can
reduction in dissolved oxygen caused by decaying algal
excessive nutrients can cause depleted oxygen levels that
important benthic organisms (shellfish), fish kills, and changes
hi phytoplankton, zooplan||bii|ind fish community composition.113 Nuisance algal blooms can have
numerous economic and bjllogical costs, including water quality deterioration affecting biological
resources, toxicity to vertebrates and higher invertebrates, and decreased recreational and aesthetic value
of waters.134 Although severe eutrophication is likely to adversely affect organisms, especially fish, a
moderate increasepa nutrient levels may also increase fish stocks, by increasing productivity hi the food
'" Paai, 1993.
04 Paeri, 1988. •
>» Hansson and Rudsttm, 1990; Rosenberg et •!., 1990; Plot, 1993.
238
-------
Appendix E: Ecological Effects of Criteria Pollutants
i Atmospheric Deposition and Entrophication '
2 The deposition of NO, in aquatic systems and their watersheds is one source of nitrogen that may
3 contribute to eutrophication. The relative importance of NOX deposition as a contributor to aquatic
eutrophication depends on the extent to which the productivity of an aquatic ecosystem is limited by
nitrogen availability and the relative importance of nitrogen deposition compared toother internal and
external sources of nitrogen to the aquatic ecosystem. Furthermore, the vulnerability of aquatic
ecosystems to eutrophication is known to vary seasonally and spatially, although these systems are
« affected by nutrient deposition throughout the year. In generaj, freshwater ecosystems appear to be more
9 often limited by phosphorus, rather than nitrogen, and are notas likely to be heavily impacted by
10 nitrogen deposition compared to some estuarine and coastal ecosystems,136 In contrast to acidification of
n streams and lakes, eutrophication from atmospheric deposition Of nitrogen is more commonly found in
12 coastal and estuarine ecosystems, which are more frequently nitrogen-limited.137
23
IS
n Unfortunately, there is limited information with regard to the relative importance of atmospheric
14 deposition as a nitrogen source in many estuarine and marine ecosystems. Estimates of the importance
is of atmospheric nitrogen deposition are difficult to make because of uncertainties in estimating
16 deposition, especially dry deposition, as well as watershed nitrogen retention.Bf Paerl (1993) reviews the
77 importance of atmospheric nitrogen deposition as a ntributor to eutrophication of coastal ecosystems;
IB he concludes that 10 to SO percent of the totajbnitrogen loadmglp coastal waters is from direct and
indirect atmospheric deposition. Estimate? of the nitrogen budget ibf the economically important
Chesapeake Bay indicate that about 25 pejcent of the nitrogen Ipadings to the bay occur via atmospheric
deposition.139 Hinga et al. (1991) est|i:iite that anthropogeni&deposition provides 11 percent of total
spSSr • M " • j5i|r f~~ *" "* " •""™
anthropogenic inputs of nitrogen iixlprraganse^^ 10 percent
for New York Bay. Fisher andjiOp||nheimer mat atmospheric nitrogen provides 23
24 percent of to||l nkrogen l(^^^^anglslail||^^^^^'23 pereent to the lower Neuse River in North-
21 C^olma. Information on lie icipioli^
siiiiirhSjSippiiiiinKiSSiiBSi, r Of
26 ecosystems is ap|?a||able m ml;li||pp5. Episodic atmospheric inputs of nitrogen may be an
27 • important soui^ Of iiit|t!iigen to nuf^|pp||p||giiarine ecosystems, such as the Norm Atlantic near
29 Valuing Potential Benei^||oiBtr]|ntrophicatioB Avoidance Under the CAA
30 It is currently estiinatei) that in the absence of pollution reductions achieved under me Clean Air
n Acktotal nitrogen emissions to the atmosphere would increase by nearly 90 million tons by 1990, a 2-
32 foMtocrease above 1990Jevels estimated with CAA controls remaining in place.141 However, the ability
u to determine the potential economic benefit from such a reduction in nitrogen emissions is heavily
34 constrained by gajpiin our current biological and economic knowledge base of aquatic ecosystems.
"* Hecky and Killum, 1988; Vttovwk and Howtrth, 1991.
'" U.S. EPA, 1993t; Paerl, 1993.
'" U.S.EPA,1993a.
wU.S.EPA,1993a.
'* Owen* etal^ 1992.
141 VS. EPA, 1995; Table B-3.
239
-------
20
Appendix E: Ecological Effects of Criteria Pollutants
i One water body that has received much study in the area of nitrogen induced eutrophication is
2 Chesapeake Bay. As previously discussed, it is estimated that atmospheric deposition of nitrogen
3 contributes approximately 25 percent to the total nitrogen loadings to the bay.142 In deposition terms, an
4 . estimated 15 to more than 25 percent increase in total nitrogen deposition has been forecast in the
5 Chesapeake Bay watershed by 1990 in the absence of CAA pollution controls.143 These results are based
6 on an estimated 40,000 tons of atmospherically deposited nitrogen (as nitrate and ammonia) to
7 Chesapeake Bay hi 1985,144 which means a 20 percent increase injgmosphe|p^ would amount
« to approximately 8,000 additional tons. ' ^
9- ' One indirect method available to gage the potential fliibmic rcU^^
10 atmospheric nitrogen loadings to Chesapeake Bay is througf||| avoidj^
u However, such an assessment is difficult due to me site, faci^
12 treatment costs. For example, Camacho (1993; as cited by Nl^^^^| reviewed nhrogen'treatment
13 costs for chemical treatment of water from important point so|{{p^fKi^^|nibu^ owned treatment-
14 works) and found that costs ranged from $9,600 to $20,600 per ton (annual "costs, 1990 dollars),
a depending on the facility evaluated. Biological treatment of nitrogen :frl|i:|i||t!sources was far more
16 expensive, varying from $4,000 to $36,000 per ton. J^i|fl||^ of noit|^p^^^ loading, vames of
17 nitrogen removal practices ranged from $1,000 tOj|f|^^ Tak|||^chemical addition as one
u possible example, the avoided costs of treatment bf 8,OolllliMMlB)igenliiiould range from about $75
* * .:Es;ii™:" .iilililr Hln:jl5H!!r:lll:L;!l!:iiiiLa^i^sbi!^~s=B«!:s! **
a million to about $170 million annually (in 19JO dollars)!
Morcury
21 Mercury, in the form of methyl mercujrjf ^|;||pbtoxin of concern and can accumulate in tissue
22 of fish to levels that are haaynliK^humans aid |§iatic-feeding wildlife in the U.S. In relation to the
23 812 CAA retrospective anarysis^n^^ First, potential benefits to human
24 welfare may Mve occurred as a fesaft|a:fliCTCury emission controls implemented under EPA's National •
a Emission Standai&l&i^ Second, experimental and observational
26 evidence suggests that acidification of water bodies enhances mercury accumulation in fish tissues.146
*•*«* =- -— v „_- -= _ -s^jg-SEn-rffiFJrt-. -£i«s? *
27 Therefore, GAA-manQ^edledjprions in sulfur and nitrogen oxide emissions and subsequent acid
21 deposition may have resuAte^lppdirect benefits from a reduction in mercury accumulation in fish and
29 subsequent improvements ||fviSiii health and welfare.
x Li The accumulation^ mercury to hazardous levels in fish has become a pervasive problem in the
31 U.S. and Canada. A rapid increase in advisories occurred during the 1980s, including a blanket advisory
32 affecting 11,000 lakesJn Michigan.147 The Ontario Ministries of Environment and Natural Resources
33 (1990, aseited byjpry and Wiener, 1991) recommend fish consumption restrictions for 90 percent of the
141 U.S. EPA, 1993«.
m U.S. EPA 1995, Figure C-«.
144 NERA, 1994.
141 Shyter, 1992; is died by NERA, 1994.
146 Bloom etal., 1991; Wains md Bloom, 1992; Miskjmmin et aL, 1992; Spry and Wiener, 1991; Wiener etil., 1990.
147 Watrasetal., 1994. .
240
-------
Appendix E: Ecological Effects of Criteria Pollutants
i
2
3
4
5
6
7
8
9
to
II
12
13
14
walleye populations, 80 percent of smallmouth bass populations, and 60 percent of lake trout populations
in 1,218 Ontario lakes because of mercury accumulation. In many instances, mercury has accumulated
to hazardous levels in fish in highly remote water bodies that are free from direct water discharges of
mercury.14* Mass balance studies have shown that atmospheric deposition of mercury can account for
the accumulation of mercury in fish to high levels in lakes of these remote regions.149 Th&potential
impacts of mercury on the health of humans and fish-eating (piscivorous) wildlife has lead EPA to
recently establish water quality criteria to protect piscivorous species in the Great Lakes.150
Although mercury accumulation in fish via atmospheric deposition is now widely recognized as
a potential hazard to human health and certain wildlife specie^ studies establishing quantitative linkages
between sources of mercury emissions, atmospheric deposition of mercury, and subsequent accumulation
in fish are lacking. Thus at the present time, we are unable to quantify potential benefits from CAA-
avoided mercury accumulation in fish of U.S. water bodies. Given the pervasiveness of the mercury
problem' with CAA-pollution controls, potential benefits to human health and welfare from avoidance of
further mercury related damages to aquatic ecosystems could be substantial.
15
If
Benefits from Avoided Damages to Wetland
IS
19
20
21
22
23
24
IS
26
27
21
29
Introduction
This review addresses
acidification and nutrient
water q
suggests mat air pollutants may most
nitrogen
of airpolUpids on wetland ecosystems; the focus is on
le sen?iceJflbws of wetland ecosystems include flood control,
habitat, and biodiversity. The limited scientific evidence
.biodiversity, in particular because of nutrient loading through
Wetlands are biiQi3ig|cJ|pfacterized as transitional areas between terrestrial and aquatic systems
in which (he water table isWiaf liirthe surface or the land is periodically covered by shallow water.151
Jr~_—'^ — -"-jhili -=;: "ar - * ,
Types of wetlands mclude«(|v&|tips (forested wetlands), marshes (herbaceous vegetation), and peatlands,
which are wetlands that accumulate partially decayed vegetative matter due to limited decomposition.152
Peatiands include bogs and fens. Bogs receive water solely from precipitation, are generally dominated
\sySphagmon moss, and are low in nutrients. Fens receive water from groundwater and precipitation,
contain more marsh-like vegetation, and have higher pH and nutrient levels man bogs.153 Most of the
ia Glass et al., 1990; Sorenson et al.. 1990; Grieb et al. 1990; Schofidd et al. 1994.
'* Fitzgerald etal. 1991.
130 U.S. EPA, 1995.
151 Cowwdinettl., 1979.
10 MHsch and Gosselink, 1986.
10 Mitsch and Gosseiink, 1986. '
241
-------
AppendixE: Ecological Effects of Criteria Pollutants
i
2
3
4
5
6
7
I
9
10
limited work on the effects of atmospheric deposition on wetlands has been done in peatlands,
specifically in Europe, where levels of atmospheric deposition are generally much higher than in the U.S.
The air pollutants of greatest concern with respect to effects on wetland ecosystems are oxides of
nitrogen (NOJ and oxides of sulfijr (SOJ, primarily sulfur dioxide (SOj). Air pollutants may affect
wetland ecosystems by acidification of vulnerable wetlands and by increasing nutrient levels.
Acidification in vulnerable wetlands may affect vegetation adversely, as ar^eais tohave occurred in
Europe. In wetlands where nitrogen levels are low, increased nitrogen depospjainay alter the
dynamics of competition between plant species. Species adapted to low-rutrogen teveb, irwluding
many endangered species, may decrease in abundance. "* --* . ;--"' v;l>>?%ft x::
11
12
13
14
IS
16
17
IS
19
20
21
22
• 23
24
23
26
27
21
29
30
31
32
Effects at Acidification -
Limited evidence suggests mat acidic deposition and decreased pHlmay harm certain wetland
plants, alter competitive relations between wetland plants and cause changes in wetland drainage and
water retention. • - .= "":\~--- ~""
--«-_=» -,- - ----•*». T"---
Work concerning the possible acidification of peatlands u inconclusive. Acidic deposition is
unlikely to result in displacement of base cations from cation excJ|upgf 'sites in bogs, and therefore it
will not cause a drop in pH.135 Peatiand sediments are low in 4Sj^NK> mobilization of toxic aluminum
is not a concern as it is hi forest soils^a^aquatic ecosystems.|f Acidification might affect certain fen
ecosystems. Gorham et al. (1984, as Cited by Turner et al., NAPAP SOS/T 10, 1990) have
hypothesized mat acidic deposition obuld leachlbju cations from mineral-poor fens and decrease pH
levels. This fpuld result in ajsandion to bog ^g^iapp such as Sphagnum and away from sedge
meadow vegetiBton. At th^^e|^|s remainti|^bmesis; however, pH did not decrease in a
mmei^-pojorQn|irip fen dura^|i4p&year period in which researchers experimentally increased
acidic i
levels of deposition for many years, acidic deposition
Roelofs (1986) reports that acidification of heath pools in the
composition with Sphagnum and rushes replacing the
leant declines in Sphagnum in British bogs have occurred in areas
pollution, including nitrogen deposition.138 It is unclear how such
has serious^ Affected
Netherlands has caused
original vegetation.
affected by 200 years of
changes have affected wetland service flows apart from the effects on biodiversity; however, water
retention has decreased Jnd significant erosion has occurred in seriously perturbed British bogs near
Manchester and
154 U.S. EPA, 1993*.
'» Ooduunettl., 1984, ucited by Turner et»l., NAPAP SOS/T 10.1990.
'* Tinner et •!., NAPAP SOS/T 10.1990.
157 Rochefortettl., 1990.
1M Lee et •!.. 1986, u cted by U.S. EPA. 1993a.
09 Lee et •!.. 1986, «s cited by U.S. EPA, 1993a.
242
-------
Appendix E: Ecological Effects of Criteria Pollutants
, Effects oi Nutrl&nt Loading
2 Atmospheric deposition may affect wetlands by increasing the level of nutrients,
3 particularly nitrogen, in wetlands. Sulfur is not a limiting nutrient in peatlands,160 but nitrogen
4 commonly limits plant growth.161 The effects of increased nitrogen levels in wetlands Include an
5 increased threat to endangered plant species and possible large-scale changes in plant populations and
6 community structure. Endangered and threatened plant specie&are common in wetlands, with wetland
7 species representing 17 percent of the endangered plant species in the U.S. (U.S, EPA, 1993a). These
s plants are often specifically adapted to low nitrogen levels; examples include isoetids1*2 and
9 insectivorous plants.163 In eastern Canadian wetlands, nationally rare species are most common in
10 infertile sites.164 When nitrogen levels increase, other species adapted to higher levels of nitrogen may.
n competitively displace these species. Thus, NO, emissions mat increase nitrogen levels in nitrogen-
12 poor wetlands may increase the danger to threatened and endangered species.
13 By changing competitive relations between plant species, increased nitrogen deposition may
14 broadly affect community structure in certain wetlands. Common species that thrive in nitrogen-poor
a wetlands may become less abundant. Many nitrogen-poor bogs in the northern U.S. are dominated by
is Sphagnum species. These species capture low levels of nitrogen from precipitation. Increased nitrogen
17 levels may directly harm Sphagnum and cape increased nitrogen to be available to vascular plants that
a may out compete Sphagnum.165 Studies in Great Britain have docomented large declines in
a moss because of atmospheric poUution^^nitrogenloadingmaypky an miportant role in these
f) declines. However, Rochefort et aL-||990) document limited effects of fertilization from
experimentally-increased NO3" andSC^2- deposition on an Ontario mineral-poor fen over a four-year
22 period, apart from initially ii
23 Iflcsej^iiinitrogett^
24 are extremely nitrogen-limited and that receive small amounts of nitrogen naturally Since bogs,
25 including Sphagnum bogs, receive^^sjwJKe water runoff, they get most of their nutrient and water
26 loadings mroughprec^)|tatioa. These l^f may receive a total of approximately 10 kg nitrogen per
n hectare per year (N/ha^Jpi^^ is one to two orders of magnitude less nitrogen than other freshwater
2t wetlands and saltmarshes -Jisi^^ As atmospheric deposition of nitrogen has been estimated to be at
29 least 5.5 to 11.7 kg N/ha/^^p&iges in NO, emissions would most likely affect these bogs. The
x results of a model by Logofet and Alexandrov (1984) suggest that a treeless, nutrient-poor bog may
31 undergo succession to a forested bog because of the input of greater than 7 kg N/ha/yr.
*" Turoeretak. KAPAP SOS/T10.1990:
m UJL EPA; i«Bt.
10 Boston, 1986, u cited by U.S. EPA, 1993*.
1(3 Moore etil., 1989.
144 Moore et al., 1989; Wiiheu and Keddy, 1989, u cited by U.S. EPA, 1993a.
1
-------
Appendix E: Ecological Effects of Criteria Pollutants
6
7
t
9
10
11
12
13
14
U
16
17
II
19
20
21
22
23
24
23
26
27
21
29
30
As in freshwater wetlands, significantly increased nitrogen deposition to coastal wetlands will
increase productivity and alter the competitive relationships between species (U.S. EPA, 1993a).
However, studies showing this increased productivity have used 100 to 3000 kg N/ha/yr.169 Therefore,
limited changes in NO, emissions may not affect coastal wetland productivity.
Summary of Wetland Ecosystem
The effects of air pollutants on wetlands have received little attention, Mcontiast to the large;
body of work on the effects of acid rain on aquatic and fbrcstecc>svstem«J^itde evidence exists that
would suggest that acidification due to atmospheric deposition is a major threat to wetlands, Sin
particular, peatlands are naturally acidic, although niineral^ipfep inay be at risk from acidification.
Nitrogen loading may alter community composition in wetlands natuxallv low in nutrients, such as
bogs. Nitrogen loading may threatenrare species adapted to ib^Wtei^^ levels. In Britain and the
Netherlands, heavy atmospheric deposition over a long
in peatlands.
Air pollutants appear to threaten most
community composition hi wetlands,
declines in Sphagnum
|ndangerjp ipecies, biodiversity, and
* -* difficult to associate with
Air pollutants may
changes in economic value; even the qualitative aaturefif the^
not significantly affect such important wetifnd servM flows a&Jfoof control, water quality protection,
and wildlife habitat in most wetlands.iiolhe impacjB on the economic value of wetlands may be limited.
Benefits from Avoid*
introdlk
|ofthe land mass in the U.S. (some 738 million acres) and provide a
wealth of services to the Notable services provided by forests include timber
production, recreational such as hunting, camping, hiking, wildlife observation,
maintenance of water quaijp', nutrient removal and cycling, flood control, erosion control, carbon
sequestration, preservatiojiof diversity and existence values. In 1991, hunting participation alone
accounted for 236 niilUip recreatioa days that included 214 million person trips with estimated
expenditures valued at $12.3 billion.170
Air Act-regulated pollutants of greatest concern with respect to effects on forest
ecosystems are oxides of sulfur (SOJ, primarily sulfur dioxide (SOz), oxides of nitrogen (NOJ, and
volatile organic compounds (VOCs). While extremely high ambient concentrations of SO2 and NO, may
'« U.S. EPA. 1993a.
"°UJS.IXM,I9M.
244
-------
Appendix E: Ecological Effects of Criteria Pollutants
i directly affect vegetation, such effects are uncommon in the U.S.;171 the indirect effects of these
2 pollutants are of greater concern. Specifically, emissions of SO2 and NOZ are known to contribute to acid
3 deposition in portions of the United States, with SO2 contributing 75 to 95 percent of the acidity in
4 . rainfall in the eastern U.S.m Acid deposition is of concern to forests primarily from the acidification of
5 soils (i.e., by reducing seed germination, altering nutrient availability). Direct foliar damage can occur
« from precipitation with extremely low pH levels (i.e., 3.0-3.6 and below), although these levels are lower
7 than ambient levels in the U.S.113 VOCs and NOX are important precursors to ozone formation, which
* can affect leaf photosynthesis and senescence and decrease cold;hardiness, thereby causing deleterious
9 impacts on tree growth, survival and reproduction. Deposition of NOT .may also alter the nutrient balance
M of forest soils, which in turn might alter the competitive relationships between tree species and affect •
11 species composition and diversity.174 • _t
n Current Air Pollutant Eff&cts on Forosts
.13 Acid Deposition Impacts ^ " - -_. - --;
u in 1985,. NAPAP organized the Forest Response Program (FRP) in order to evaluate the
is significance of forest damage caused by acidic deposition, me causal relationships between air pollutants
is and forest damage, and the dynamics of these relationships regionally^ Research was focussed on four
17 forest regions: Eastern Spruce-Fir, Southern Commercial Forest^ Eastern Hanlwoods, and West^
11 Conifers. With the exception of highrfle^ition spruce-fir forest^ the available evidence suggests that
19 acidic deposition does not currently affectthese fbfests and Apt observed declines in sugar maple and
j_i _? A. j A_ »_*• i^arl: _•*_• l(WL.==! "iHT*
3 southern pines are not due to acidic desition
21 Circumstantial evid|D|$|^gests that«c|^deposition may affect high-elevation spruce-fir
22 forests in me northeastern UJS^^IIIu»i;;&>rests have extensive contact with acidic cloud water.176
23 Experimental evidence suggests |[pi||u|k deposition may affect cold hardiness in red spruce, an
24 important component of the spiuce^f|iy||r|i^^ Significant declines in red spruce growth and in its
23 importance in the j^^fli&^ip^uiTed^lfew York and northern New England. The proximate cause of
26 dearnofi^sjmicemi^rej^
27 biological stresses and wfiltivi^^ Ozone may
a also play a role in red spnic$diiifpe in mis region.177 Available evidence suggests that soil aluminum
29 and soil pH levels have not affected red spruce adversely.I7t
m arineret«L,NAPAP SOSVT18,1990.
™ WffAP, 1991.
ra Shriner et al., NAPAP SOSfT 18,1990.
174 U.S. EPA, 1993«.
171 Barnard et al., NAPAP SOOT 16,1990; NAPAP 1991.
'* Barnard et al., NAPAP SOSfT 16,1990.
177 Shriner et al., NAPAP SOSfT 18,1990.
171 Barnard et al., NAPAP SOOT 16,1990.
245
-------
Appendix E: Ecological Effects of Criteria Pollutants
i Ozone Impacts • .
2 Experimental Evidence
3 For practical reasons, the majority of experimental evidence Unking ozone exposure to damage
4 to tree species has been derived from studies of individual plants, especially seedling and branch
5 studies.179 Results from these studies suggests mat ozone exposure can reduce photo&ynmesis and
6 increase senescence in leaves. Subsequently, such effects from ozone may alter tie carbohydrate
7 allocation to plant tissues such as roots, which may affect plant growth and cold hardiness. Decreases in
i cold tolerance may be particularly important for trees in northern latitudes and high elevations. Recent
9 work on quantifying the relationship between ozone exposure and plant responses suggest that seedlings
w of aspen, ponderosa pine, black cherry, tulip poplar, sugar maple, and eastern white pine seedlings may
u experience biomass reductions of approximately 10 percent at or near ambient ozone exposures.110
12 Although indicative of short-term relative response to ozone exposure, results from these
13 experiments are unable to provide reliable information on the long-term efjfects of ozone on forests. This
14 Imitation arises because the effects of ozone on forests will depend on both me response of individual
is plants to ozone exposure and the response of populations of plants, which interact with their
16 environment Population response will altered by Ae varying mtraspecific genetic susceptibility to
n ozone. Individual plant response will also be affected by many envinnmiental factors, including insect
a pests, pathogens, plant symbionts, competing plants, moisture, temperature, light, and other pollutants.
19 Consistent evidence on the interaction of ozone witk other environmental factors is lacking.
20 Furthermore, most experimental studitiahave only studied exposure for one growing season; effects on
21 forest species may occur over decades;111 Therefore, considerable uncertainties occur in scaling across
22 individuals of different ages, foam individuals to pji^latijcms and communities, and across time..
. " „-=."! - " - ~:Si"!;' IB! "~4F-"!I!a1— !SJ"-s':n?i:s'li:"
13
_-
Observational E
24 Studies ofthe forests of me^lii^piaardino Mountains provide the strongest case for linking
25 ozone exposure todaina|^Ei|to an ento!f||pest ecosystem. These forests have been exposed to extremely
26 high ambient ozone le¥«l§.jii||r|Ae past Id years due to their proximity to the Los Angeles area. The
27 area has been extensivel]iiii||i^^garding the effects of ozone, as described in U.S. EPA (1993b). The
a ecosystem has been serioiHp^fi&d by ozone pollution, with the cUmax-dominant, but ozone-sensitive
29 ponderosa pine and Jeffreyjrihf declining in abundance at the expense of more ozone-tolerant species.
30 These sensitive species hawe experienced decreased growth, survival, reproduction, and susceptibility to
31 insects. The effects of ozone on these species have resulted in other ecosystem effects, including the
32 buildup of a large litter layer, due to increased needle senescence. The decline of the fire-tolerant
33 ponderosa^nd Jefj|ey pines may seriously affect the fire ecology of the ecosystem, with fire-sensitive
34 species b^cxmung more common. Ozone concentrations have been declining in recent decades, and
35 crown injury of ponderosa and Jeffrey pine has decreased. However, the two species have continued to
I7»U.S.EPA,1993b.
110 Hogiett et it. 1993; as cited by U.S. EPA, 1993b; Table 5-25.
111 U.S. EPA, 1993b.
246
-------
Appendix E: Ecological Effects of Criteria Pollutants
i
2
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
II
19
20
22
23
24
25
26
27
IS
29
JO
31
decline in abundance, as measured by total basal area, compared with other species over the period 1974
to!988.112
Limited field studies have been completed hi other forest ecosystems. Peterson etal. (1991, as
cited by U.S. EPA, 1993b) reports foliar damage and decreased radial growth in ponderosa-and Jeffrey
pine in the Sierra Nevada Mountains of California. Work in the southern Appalachian Mountains shows
that foliar injury from ozone is particularly common in white pine and black cherry (and occurs in other
species), and that ozone can reduce the growth of tree species and herbaceous vegetation.113 Evidence of
population and community changes caused by ozone has not been found hi these areas.~J--
.' - _"* ~- ' ~ - - ' - '-
Monitoring by the USDA Forest Service shows that growth rates: of yellow pine in the Southeast
have been decreasing over the past two decades in natural stands but not hi pine plantations.1*4 Solid
evidence linking this growth reduction to air pollutants is lacking, although ozone, in particular, may be a
factor.1*5 Ambient ozone levels in the region are high enough to damage sensitive tree species, including
pine seedlings during experimental exposure.116 Due to the commercial importance of yellow pine, the
economic impacts of ozone on forest ecosystems in this area could be significant if ozone is affecting
growth. '---'-_ :viV>*
Although the ecosystem effects occurring in the San Bernardino forest ecosystem have occurred
at very high ozone exposures, lower ozone exposure elsewhere inflw;UJ."may still affect forests. Even
small changes in the health of ozone-sensitive species may affect competition between sensitive and
tolerant species, changing forest stand dyjiunics.1*7 Depending on the sensitivities of individual '
competing species, this could affect tijnEiber production either positively or negatively, and affect
community composition and, possi["
-=_
—"_-=•
Valuation of
Background
From CAA-Avold&d Damages to
In order to qua
the economic benefits of avoided damages of relevant CAA
link estimated changes in air pollution to measures of forest health
quantified in economic terms. For commercial timber production, this
relationship between atmospheric deposition and measures of forest
pollutants to forests, it is
and conditions that can be,
would require quantifying
productivity such as timber yield. For assessing recreational benefits, linkages would have to be drawn
between air pollutiopiad vulnerable factors that mfluence forest-based recreation (e.g., sfte-
canopy density, type of tree species, degree of visible tree damage, etc.). While
m MfflerctiL, 1989 nd Miller*•!., 1991. is cited by U.S. EPA. 1993b.
10 U.S.EPA,1993b.
114 NAPAP, 1991.
m NAPAP, 1991.
'« NAPAP, 1991.
117 U.S.EPA,1993b.
247
-------
Appendix E: Ecological Effects of Criteria Pollutants
/
2
3
4
S
6
7
S
9
10
11
11
13
14
IS
16
17
It
20
21
a
23
24
33
26
27
a
29
30
31
32
33
34
15
36
37
important strides have been made in establishing these linkages (e.g., NAPAP modeling of air pollution
effects on forest soil chemistry and tree branch physiology), critical gaps in our ability to predict whole
tree and forest responses to air pollution changes have precluded the establishment of such quantitative
linkages.1" Critical knowledge gaps exist in our ability to extrapolate experimental results from seedling
and branch studies to whole tree and forest responses, to account for key growth processel of mature
trees, to integrate various mechanisms by which air pollution can affect trees (fig., so&acidification,
nutrification, and direct foliar damage, winter stress, etc.), and to^cottntfc^^^^cthmofomeT
stressors on forest health and dynamics (susceptibility to insert d|mage, drli|||||||||pe, fire, nutrient ,
and light competition, etc.). . _ f
Despite these constraints to quantifying economic benefits frojglir pollutionQl^^^^rtirest
ecosystems, relevant studies that have attempted to value aiijpp|tiojfimages on forests |rt reviewed
and summarized below. In some cases, the relationship betwcffff9^
estimated using expert judgement (e.g., for NAPAP assessmen|irHi||bus growth scenarios). In other
cases, damage estimates reflect current impacts of air pollution on :fof|||f p^the dose-response
relationship is absent In the aggregate, this summary provides some n||||||||||i|)ossible CAA-rehited
benefits from avoided damages to a select and nam)wj|;Jf|(||ssed group of fip|st2iervices, but, because
of severe date constraints, is not able to provide e|^ll^^^^iiate of f^^eiAllnaige of forest-
based benefits possible under me CAA.
n Commercial Timber Harvesting
The economic impact o
(both hardwood and softwood
ranging from 5 to 10
assumed to occur as a resultofi
by deSteigner and Pye
market rcspons^tejhese
revised version of m$ Timber
leductioji in northeastern and southeastern trees
the United States. Econpra|il
changes in revenue to
consumer and producer sui
(in 1957 dollars). Simul
was mtensively studied under NAPAP.1*9 Growth reductions
10 yev|itipi^^|^ending on the species and location, were
of air^p^UUpirbased on expert opinion derived from a survey
Haynes and Kaiser, NAPAP SOS/T 27 part B, 1990), Timber
declines were modeled until me year 2040 using a
Model (TAMM90) and the Aggregate Timberland
"mMlftt* t*1"^ inve«lt""«^ *** privatq timWlanH in
included changes in consumer and producer surplus and
owners. Results indicate that annualized reductions in
total $0.5 billion by the year 2000 and $3 billion by the year 2040
on stumpage owners' revenues were minimal ($10 to $20 million).
: In an attempt ^estimate the net economic damages from ozone effects on selected U.S. forests,
NAPAP studied the ejpct of various owiwieJ reductions in growth rates of commercial southeastern
pine forests j(bx^|pural and planted).190 For bom planted and natural plus planted pines, the following
changes m growflif fates were assiimed to occiu^ a 2 pen»nt mcrease, no change, a 2 percent decrease, a 5
percent decrease, and a 10 percent decrease. The 2 to 5 percent growth reductions were considered as .
possible Outcomes from current ozone induced damage to southeastern forests, although no quantitative
•" NAPAP, 1991.
'" Haynet mad Kaiser, NAPAP SOS/T 27 p«t B, 1990.
"° NAPAP, 1991.
248
-------
Appendix E: Ecological Effects of Criteria Pollutants
i
2
3
4
5
6
7
I
9
10
II
a
13
14
a
16
17
H
19
JO
21
22
24
25
26
27
21
29
30
31
32
33
34
35
36
37
linkage between ozone exposure and damages, was established. The 10 percent growth reduction
scenario was primarily included for evaluating model sensitivity to severe changes in growth and was
considered out of the range of likely ozone damage estimates. The TAMM and ATLAS models were
again used to simulate timber market responses under baseline and hypothesized growth change
scenarios from 1985 to 2040. Results indicate that annual changes in total economic surplus (i.e., the
sum of consumer and producer surplus and timber owner revenues in 1989 dollars) would range from an
increase of $40 million (for the 2 percent increase in growth scenario) to a decrease of $110 million (for
the 10 percent decrease in growth scenario) planted and natural pine model simulations.
In the context of estimated benefits from avoidance of other damages in the absence of the Clean
Air Act from 1970 to 1990,191 the magnitude of economic damages estimated to the commercial timber
industry are comparatively small. For example, economic damage estimates range up to $3 billion
annually for 5 to 10 percent growth rate reductions in northeast and southeast forests, and just $110
million for southeastern pines. However, in the context of damages to forest-based services as a whole,
the NAPAP-derived commercial timber damage estimates should be viewed as representing a lower
bound estimate for a variety of reasons. First, these damage estimates exclude other categories of
possible forest-based benefits, including recreational and non-use values. Second, even within the
context of timber-related damages, the NAPAP forest-damage studies focused on a portion of U.S.
forests (northeastern and southeastern U.S.); a much greater geographic range of forests could become
susceptible to timber-related damages in the absence of CAA cootrqls^ ;Fmally, the NAPAP damage
estimates consider only two types of tree species: planted and naturally grown pines, although these
species are economically important Dam|ges to other commercially harvested tree species, such as
mixed pine and hardwood forests, art ilherefore excluded. 3?
* = --s-ic- .;;..=-_4. j ,ti
23 Non-marketed Forest Service!
.
Inaaeff^toaddressj
damages to non?i
review of the econoittic literature i
NAPAP could not identi
recreational use) on a
was the absence of a quan
characteristics which
benefits resulting from avoidance of acid deposition-induced
(e.g., recreation use, existence value), an extensive
ted under the auspices of NAPAP.1*2 From their review,
model mat could be reliably used to quantify economic
Used damages to non-marketed forest services (such as
1 basis. The primary limitation in many of the studies reviewed
ige between the value of a recreational user day and important site
to air pollution effects. In addition, most studies were narrowly
footled geographically to sjraific sites and did not attempt to value system-wide (larger scale) damages
that epuld result from acjji deposition over an entire region. Since the availability of nearby substitution
sites will affect the recreational value for given site, the benefits from such site-specific studies may not
reflect actual eco^oinic damages incurred from wide-scale air pollution impacts on forests. The inability
of studies to consider additional crowding at unaffected sites in addition to changes in recreational
participation rates as a function of air pollution damages was also recognized as an important limitation.
Most notably avoided In
altfa effects, which ace estimated on the order of $300 to $800 billkm annually.
ln Rosentnal, NAPAP SOS/T 27 Part B, 1990.
249
-------
Appendix E: Ecological Effects of Criteria Pollutants
i Despite not being able to quantitatively assess the benefits from avoided acid deposition-induced
i damages to nonmaiket forest services, several important concepts emerge from NAPAP's review of
3 recreational benefits, that bear relevance to the 812 retrospective analysis. First, several studies were
4 . identified that established a relationship between key forest site characteristics and the value of
5 recreational participation. For example, Brown et al. (1989, as cited in Rosenthal, NAPAP SOS/T 27
6 Part B, 1990) used contingent valuation to evaluate the relationship between scenic beauty ratings and
7 willingness of recreationalists to pay at pictured sites. Based on their interviews with over 1400
i recreationalists at 10 different sites in Arizona, positive correlations were established between scenic
9 beauty rankings determined from one group of recreationalists andwillingness to pay to recreate
10 determined by a separate group of recreationalists (r2 ranged fipm 027 to 0.98 depending on ranking). In
n another study, Walsh et al. (1989, as cited in Rosenthal, NAPAP SOS/T 27 Part B, 1990) developed a
12 functional relationship between reduction of recreational benefits and tree density changes that reflected
n varying levels of msect damage at sk<»mpgrounds in the Front Range of me Colorado RTC By
u using both contingent valuatioirand travel cost models, Walsh et al. (1989) were able to show that 10
is percent, 20 percent, and 30 percent decreases in tree densities reduces the total recreational benefits at
i6 their sites by 7 percent, 15 percent and 24 percent, respectively. Although results from these studies are
17 limited to the sites from which they were derived, they do confirm intuition mat the degree of visible
is damage to forests is to some extent correlated with me magnitude of damages to forest-based recreation
19 expected. This finding supports the notion that Jheavoidance of damages to forest ecosystems from
20 CAA-induced pollution controls (albeit currently unquantified) have likely benefited forest-based
21 recreation in the U.S. * -- ->*,.*=-r
•--'?f' -j? '-»
°3s-~>*£ ,T-^ "--TJ^
22 In addition to establishing relationships between recreational value and visible damage to forest
23 sites, there is evidence Unking air pollution (ozoofl effects on forests to economic damages to non-use
24 values of forests. For example, Peterson et alv{|9l7^ af iited in Rosenthal, NAPAP SOS/T 27 Part B,
25 1990) valued ozone-included datnapes to foresMpiipJliding the Los Angeles area. Using contingent
26 valuation meliKxls^Petei^n^ |||[|^7) surveyed recreationalists (a random survey of households in
27 the San Bernardino, jLos Angles anil <%pge counties) and residence (a sample of property owners within •
2s the San Bernardino and Angeles nitiiond fopests) for their willingness to pay to prevent forest scenes
29 from degrading one step on^i "forest qujdjlf ladder" depicting various levels of ozone-induced damages.
x The mean wiiyagnesft%^i^r|ptect1in-mer degradation was $37.61 and $119.48 per household for
31 * recreationalists and residents! «ffitpegtively. Annual damages to Los Angeles area residences from a one-
32 step drop on the forest qual^la^iir were estimated between $27 million and $147 million.
jj ; Although these estimates cannot be directly translated into a rough estimate of the potential non-
34 use values of avoided forest damages, they do provide evidence that the recreational and non-use benefits
3i may substantially exceed the commercial timber values.
250
-------
Appendix E: Ecological Effects of Criteria Pollutants
2
3
4
S
6
7
t
9
10
11
12
13
14
16
7
11
19
20
22
23
24
23
Baker, LP., D.P. Bernard, S.W. Christensen, MJ. Sale, J. Freda, K. Heltcher, D. Marmorek, L. Rowe, P.
Scanlon, G. Suter, W. Warren-Hicks and P. Welbourn. 1990. Biological effects of changes in
surface water acid-base chemistry. NAPAP SOS/T Report 13, In: Acidic Deposition: State of
Science and Technology, Volume n, National Acid Precipitation Assessment Program, 722
Jackson PlaceNW, Washington, D.C. 20503. _//' -J?*-;~:\-^"-.-A-
-=a _ff-<< "'""'iV"!."'- -"'*"..
Barnard, J.E., AA. Lucier, R.T. Brooks, A.H. Johnson, PH. Dunn and D.F. Karnosky. 19PO, <3ianges
in forest health and productivity in the United States and Canada. NAPAP SOS/T Report 16, In:
Acidic Deposition: State of Science and Technology, Volume m, NationaT Acid Precipitation
Assessment Program, 722 Jackson Place NW, Washmgton, JXC» 2J)503.
Bloom N.S., C J. Watras and JJP. Hurley. 1991. Impact of acidification^ ttemethylmercury cycle of
remote seepage lakes. Water AfrSoU Pollution 56:477-491.
Fitzgerald, W.F., R.P. Mason, and G.M. Vandal. 1991. Atmosph
Exchange of
56:745-767.
Glass, G. E., JA. Sorenson, K.W. ScJ
and Air-Water
Water, Soil, Air & Soil Poll.
t, and GJL Rapp, JrJI990. New Source Identification of
Mercury Contamination rn the Great Lal^. Enviroa* Sci. Technol. 24:1059-1069.
Grieb, T.M, C.T,
affecting mercury a
•21 Hansson, S. and LXs.
CX. Schofield, G.L. Bowie and D.B. Porcella. 1990. Factors
in fishTiOTSe upper Michigan peninsula. Environ. Toxicol.
Haynes, R.W. and DM:
forests. IQ: The
Ion and Baltic fish communities. Ambio 19:123-125.
Assessing economic impacts of air pollution damage to U.S.
ipact of Air Pollution on Timber Markets, J.E. deSteigner, Ed.
U.S. Department ofAgi&ulture, Southeastern Forest Experiment Station; general technical
report no. SE-75. [Cited in U.S. EPA (1993b).] >
26
27
21
29
30
31
32
Hecky,RJi. andP.Kflham. 1988. Nutrient limitation of phytoplankton in freshwater and marine
: . eayironments: a review of recent evidence on the. effects of enrichment Limnology and
Oceanography 33:796-822.
Hinga, K.R., AA. Keller and C A. Oviatt 1991. Atmospheric deposition and nitrogen inputs to coastal
waters. Ambio 20:256-260.
Hogsett, W.E., A. Heretrom,J A. Laurence, J.Weber, E.H. Lee and D.T.Tingey. 1993. Ecosystem
exposure assessment: ozone risks to forests. In: Comparative Risk Analysis and Priority Setting
251
-------
Appendix E: Ecological Effects of Criteria Pollutants
i of Air Pollution Issues. Air and Waste Management Association, Pittsburgh; PA. [Cited in US
2 EPA(1993b).]
3 Miller, P.R., J.R. McBride, SX. Schilling and A.P. Gomez. 1989. Trend of ozone damage to conifer
4 forests between 1974 and .1988 hi the San Bernardino mountains of southern California. In:
5 Effects of Air Pollution on Western Forests. R.K. Olson and A.S. Lefohn, Eds. Pittsburgh, PA:
< Air and Waste Management Association, Pittsburgh, PA, pp. 309-324 (Transaction series no.
7 16). [Cited in U.S. EPA (1993b).] -.-=>jf -;>? ^
s Miller, P.R., J.R. McBride and SX. Schilling. 1991. Chronicozone injury and associated stresses affect
9 relative competitive capacity of species comprising me California mixed conifer forest type. In:
10 Memories del Primer Simposial National; Agriculture Sostenible: Una Option para Desarollo
n sin Deterioro Ambiental. Comision de Estudios Ambientales, Colegio de Postgraduados,
12 Montecillo,Edo. Mexico, Mexico, pp. 161-172. [Cited in U.S. EPA (1993b).]
13 Miskimmin, B M., J.WJM. Rudd and CA. Kelly. 1992. Influence of dissolved organic carbon, pH, and
14 microbial respiration rates on mercury methylation and demethylation in lake water. Can. J.
a - Fish. AquatSci. 49:17-22 : - ??
i6 National Acid Precipitation Assessment Program (NAPAP). 1991, i99^Integrated assessment report
n National Acid Precipitation Assessment Program, 722 Jackson Place NW, Washington, D.C.
« 20503. . &-'.* > $:
^ ;_7
19 National Economics Research Associate (NERA^Inc.. 1994 The benefits of Reducing Emissions of
20 Nfrttgen Oxides unde*£h^
21 Owens, N^^f^%GaUo^ 1992. Episodic atmospheric nitrogen deposition to
22 oligotoeplbic oceans. Nature^ 357^97-399.
- " J?i- -E-, "a3Js - "-~™ ~^^^^~jy~~=--S.
*-_ -.-. _-
23 Paerl, H.W. 1988. |>^isaaM>hvtoplaa^^ Limnol.
" "
25 Paerl, BLW. 1993. Emei^giiig|i^^
26 . ; biogeochemical and tropic perspectives. Can. J. Fish. Aquat Sci. 50:2254-2269.
~ - - ^ '•
2? Rpsoiberg,R.,R.Ehngren,S.Fleischer,P.Jonsson,G.PerssonandRDahlin. 1990. Marine
2s eutropbicationcase studies in Sweden. Ambio 19:102-108.
29 ScAojfeId,jfeii|Stff. Driscoll, R. K. Munson, C. Yan, and J.G. Holsapple. 1994. The Mercury Cycle and
30 fish in the Adirondack Lakes. Environ. Science & Tech. 28:3:136-143.
31 Shriner D.S., W.W. Heck, S&. McLaughlin, D.W. Johnson, P.M. Irving, JD. Joslin, and C.E. Peterson.
32 1990. Response of vegetation to atmospheric deposition and air pollution. NAPAP SOS/T
33 Report 18, In: Acidic Deposition: State of Science and Technology, Volume ffl, National Acid
34 Precipitation Assessment Program, 722 Jackson Place NW, Washington, D.C. 20503.
252
-------
Appendix E: Ecological Effects of Criteria Pollutants
i Sorenson, JA., Glass, G. E.,, K.W. Schmidt, and G.R. Rapp, Jr. 1990. Airborne Mercury
2 Deposition and Watershed Characteristics in Relation to Mercury Concentrations in Water,
3 Sediment, Plankton and Fish of Eighty Northern Minnesota Lakes. Environ. Sci. Technol. 24:
4 1716-1727. . ;
s Spry, DJ. and J.G. Wiener. 1991. Metal bioavailability and toxicity to fish in tow-alkalinity lakes: a
6 critical review. Environmental Pollution 71:243-304. .... .-'-
7 U.S. DOI, 1993. Fish and Wildlife Service and U.S. Department of Commerce, 1991 National Survey of
« Fishing, Hunting, and Wildlife-Associated Recreation, U.S. Government Printing Office,
9 Washington D.C. ' _" ~ ; v
10 U.S. EPA. 1993a. AkQxiality Criteria for Oxides of Nitrogen. Office of Health and Environmental
u Assessment, Environmental Criteria and Assessment Office, Research triangle Park, NC; EPA
n report no. EPA600/8-91/049bF. 3v. V.-V ^..
" ' - .""'\--Sjl-~-;VTr,
14 U.S. EPA. 1993b. External Draft, Air Quality Criteria for Ozone and Related Photochemical Oxidants.
is Volume n. Office of Health and Environmental Assessment, Environmental Criteria and
16 Assessment Office, Research Tiiangle Park, NC; EPA report no. EPA/600/AP-93/ 3v.
U.S. EPA. 1995. The Benefits and Costs of me Clean Air Act 1970 to 1990 _ Report to Congress.
n Vitousek, P.M. and R.W. Howarth. 1591: Nitrogen limitation on land and in me sea: How can it occur?
- _grs •**• .:: -5 I
Biogeochemistry 13:87-115i?* ' __^.{"
20 Watras, CJ. and N.S. Bloomy 19^2| Mercury and meffiyhnercury in individual zooplankton:
21 implications for bioaccuffli|^n. Limnology and Oceanography37:1313-1318.
22 Watras, C.J., N.S, Bloom, RJ.M. Hidsofb^^herini, JL Munson, S A. Claas, KA. Morrison, J. Hurley,
u J.G. Wiener, WJF. Fitzgerald, R. Mason, G. Vandal, D. Powell, R. Rada, L. Rislov, M. Winfrey,
24 J. Elder,D.Krabbeah6fl;,A.W.Andren,C.Babiarz,D.B.Porc«llaandJ.W. Huckabee. 1994.
25 Sources and fates ^lipi^y and methyhnercury in Wisconsin lakes. In: Mercury Pollution:
26 -Jmtegration and Synil||pej. Watras and J.W. Huckabee, Eds. Lewis Publishers, Boca Raton,
27 Florida, pp. 153-180. '*^F • •
21 Wiener, J.G., R.E. Martini, T.B. Shefiy and G.E. Glass. 1990. Factors influencing mercury
29 . concentrations.^ walleyes m northern Wisconsin lakes. Trans. Am. Fish. Soc. 119:862-870.
x [Cited in Spry and Wiener (1991).]
253
-------
[This page intentionally blank] _i:~~
254
-------
4
Appendix Fs Effects of Criteria
Roiiutants on Agriculture
Introduction
One potential impact of air pollutants on economic welfare is their effect on agricultural crops,
5 including annual and perennial species. Pollutants may affect processes within individual plants mat
6 affect growth and reproduction, thereby affecting yields of agricultural crops. Possible physiological
? effects of pollutants include the following: decreased photosynthesis; changes in carbohydrate allocation;
i increased foliar leaching; decreased nutrient uptake; increased sensitivity to, climatic stress, pests, and
9 pathogens; decreased competitive ability; and decreased reproductive efficiency . These physiological
to effects, in conjunction with environmental factors and intraspecies differences in susceptibility, may
// affect crop yields. ' " ~_"j:
12 Air pollutants that might damage plants include SO^NO^ peroxyacetyl nitrate (PAN), and
13 volatile organic compounds (VOCs). Thes&pollutants may have direct effects on crops, or they may
H damage crops indirectly by contributing Jtolropospheric (ground-level) ozone and/or acid deposition,
is both of which damage plants. Tropospkfric ozone is formed by photochemical reactions involving '
< VOCs and NO^ while SO2 and NOX cause acidic deposition.
•W x^:.. -^-
n While all of these air pollutants may inflict incremental stresses on crop plants, in most cases air
is pollutants other than ozone ar«not-asignificantdanger to crops. Based primarily on EPA's National
19 Acid Precipitation Assessment Program (NAPAP) conclusions,193 mis analysis considers ozone to be the
20 primary pollutaflt affecting agricultural production. .
11 This analysis estimates the economic value of the difference in agricultural production that has
22 resulted due to the existence of the CAA since 1970. The analysis is restricted to a subset of agricultural
23 commodities, and excludes tho^ commodity crops for which ozone response data are not available.
24 Fruits, vegetables, ornamental^ and specialty crops are also excluded from this analysis. To estimate the
a economic value of ozone reductions under the CAA, agricultural production levels expected from control
26 scenario ozone conditions are first compared with those expected to be associated with ozone levels
27 predicted under the no-control scenario. Estimated changes in economic welfare are then calculated
a based on a comparison of estimated economic benefits associated with each level of production.
Shriner et •!., 1990; NAPAP, 1991.
255
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
Ozone Concentration Data
2
3
4
5
6
7
I
9
10
11
12
13
14
IS
16.
17
II
19
20
21
22
23
24
25
26
27 :
21
29
30
To estimate the nationwide crop damages as a result of ozone exposure, the first step is to
estimate the nationwide ozone concentrations under the control and no-control scenarios. This section
describes the methodology used to estimate ozone concentrations for each county in each of these two
scenarios.
First, historical ozone concentration data at the monitor level were compiled Iron^ EPA's AIRS
system. Differences between me modeled control and no-control scenarjb ozone concentration were
•then used to scale historical data to derive no-control scenario oapie air quality profiles.^* Ifext, the
ozone index used in the exposure response evaluation was cakailated and applied at the monitor level.
For this analysis, the W126 index, a peak-weighted average of cumulative ozone concentrations, was
selected to conform with the index currently being used by EPA hi ozone MAAQS benefits analysis. The
W126 index is one of several cumulative statistics, and may correlatemore closely to crop damage than
do unweighted indices.191 EPA has not yet made a final determination of ttie appropriate index to use in
agricultural benefits analysis; thus this analysis shoi^bj| viemsd only as an indicator of me magnitude
of potential benefits.
-a
ff
The third step hi ozone concentratipiestimation mvolve||ii3ilse of triangulation and planar
interpolation to arrive at a W126 statistpji the county, rather thjih at the monitor, level. For each county
centroid, me closest surrounding triai^ oYmon^ils is located and the W126 is calculated for mat
county using a distance-weighted average of mejiane concentration at each of these monitors.
Control and No-control Scenario Ozono Concentration
The initial
performed by Systems
ozone datalfrom the EPA'
these data by fitting gam
in the control and no-control scenarios was
Intelrnational (SAI). To create the control scenario, SAI compiled
Information and Retrieval System (AIRS).19* SAI summarized
Etions to them and providing the alpha and the beta parameters to
these distributions. Each ofjthe&e distributions describes a set of ozone concentration levels, and the
distributions are categorizfd by year, season, and averaging time. SAI defines six distinct "seasons,"
each composed of a two month period in the year. This analysis uses .those distributions which describe
1-hour average ozortgponcentrations taken from 7 AM to 7 PM and separated into seasons. The analysis
utilizes only those monitor records that were modeled hi both the control and no-control scenarios.
1M Derivation of these ozone air quality profiles fcr the control and no-control scenario is 9
described in detail in Appendix C.
'* Lefohnetil., 1988.
'* SAI,ICFKiiier, 1995.
rized in the following section and
256
-------
AppendixF: Effects of Criteria Pollutants on Agriculture
; To determine the ozone concentrations for the no-control scenario, SAI utilized the Ozone
2 Isopleth Plotting with Optional Mechanisms-IV (OZIPM4) model. The input data required for OZIPM4
3 includes air quality data, surface and upper-air meteorological data, and estimates of anthropogenic and
4 biogenic emissions of volatile organic compounds, NO, and CO.197 To create these inputs, SAI used
s (among other sources) outputs from the Regional Acid Deposition Model (RADM) and the
6 SJVAQS/AUSPEX Regional Modeling Adaptation project (SARMAP). Additional detail concerning the
7 development of ozone concentration data is available in Appendix C and in the SAI report to EPA.19t
< Calculation of tho W12O Statistic
9 Using the SAI ozone concentration distributions, we calculated a sigmoidally weighted ozone
10 index for each monitor. The generalized sigmoidal weighting function used in calculating such indices is
n presented in Lefohn and Runeckles (1987) as: , * :
12 where: w, = weighting factor for concentration I (
13 c^p concentration I (ppm) - "'
14 . "_W* an arbitrary constant
is >i = an arbitrary constant
-'^S. <- -
_ -t
16 The constants M and A are chosen to give different weights to higher or lower concentrations. The index
17 used in this analysis is the W1M statistic, wWcb is calculated as follows:199
it and -
79 Missing values are accounted for by multiplying the resulting W126 statistic by the ratio of the number
20 of potential observations to the number of actual observations (i.e., total hours in period/hours of data in
21 period). : -^
22 --•..
'" SAI, ICF Kaiser, 1995.
"• SAI, ICF Kaiser, 199$.
'" Lefohn etal, 1988.
257
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
To calculate W126 indices from the monitor level gamma distributions, we used an inverse
cumulative density function to calculate a separate representative air concentration for each hour in the
two month season. These values are then used in the above equation to obtain a monitor-level W126
statistic. . . • •
To ensure that the interpretation of the gamma distributions in this manner does not generate
errors, we tested our gamma-derived control-scenario W126s against W126s calculated directly from the
AIRS database. We found mat insignificant error resulted from the utilization of the gamma
distributions to create W126 statistics. : ; ;
10
n
12
13
14
IS
16
17
IF
19
30
21
22
23
14
25
26
27
21
29
30
31
32
Aggregating Ozone Data to the County Level :
Because crop production data are available at the county level, the lowest level of aggregation
that could be used for ozone indices is also the county level. Therefore, monitor level data needed to be
aggregated to a county level. For each county, we first located the monitors Iropjvhich we would be
interpolating data. To identify these monitors, we searched for die three monitors" which formed the
. J»JMB ""- --^"J8a- " "" "-"H !=~si5_='r:r
closest triangle around the centtoid of the county^ The.cjtoses||nangle was defined as that triangle in
which the sum of the distances from the three monitors tol££ii|ffi|||^pipid was the least The
algorithm stopped searching for closest triangles of monitors w^M§tfsearched all monitors within
500 km of a given county centroid (an arbitrary distance, selected® reduce computational requirements).
For coastal counties and
centroid do not exist We assignedfKe W126
counties. Approximately 15
W126s in this manner.
of all
in some fears, monitor triangles around the county
from tite monitor closest to the centroid to these
of 248,880 records) were assigned
the closest triangle of monitors was found, a "planar
at that county for that year. One way to picture this
as a point in space. For each monitor, the x axis represents
aid the z axis represents the W126 statistic. A plane can men be
Furthermore, using the equation for die plane, and given the x
for the county centroid, the county centroid's z value (W126
:, mis procedure calculates a distance-weighted average of three
interpolation was
process is to plot cacjh of
longitude, they axis
drawn through these
and y values (latitude and
statistic) can be <^culated|lln
monitors'W126 index vafiils and assiens this value to the county centroid,
V s- -a--"- ~
-""". -—^ . ' '
_ "j ''" - ispF
X=.o The final resuj* of this data manipulation is a monthly W126 statistic for each county in the
continental Upted|States for the years 1971-1990. From these data, yield change estimates were
generated, and economic impacts were estimated.
"° The vast majority of moniton had Utitude and longitude data available through AIRS. U28 of 1,536 monitors were located in this
manner. For the remaining 8 nwniton, if to a given year of monitor date toother monitor exists in te
discarded the unlocated monitor's data. Otherwise, we located that monitor at the county's centroid. We located 5 of the remaining 8 monitors
in this fashion.
258
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
Yield Change Estimates
2 There are several steps involved in generating yield change estimates. The first is the selection
3 of relevant ozone exposure-response functions (minimum and maximum) for each crop in the analysis.
4 Ozone data, triangulated to the county level, are transformed into an index suitable for use in the selected
j function(s) to estimate county level predicted yield losses for both me control and no-control scenarios.
6 In the next step, the proportion of each county to the national production of each crop is calculated to
7 permit national aggregation of estimated yield losses. Finally^ the control scenario percentage relative
a yield loss (PRYL) is compared to the minimum and maximum PRYL tor the no-control scenario. Each
9. step is discussed in more detail below. _>. > - " -~-L''<
Exposure-Rosponso Functions
10
u To estimate yield impacts from ozone, exposure-response functions are required for each crop to
12 be analyzed. This analysis was restricted to estimating changes in yields for those commodity crops for
13 which consistent exposure-response functions JMW available and that are included hi national agricultural
14 sector models. To maintain consistency with tile current ozone NAAlQS benefits analysis being
is conducted by OAQPS, NCLAN-based exposure-response functions using a Weibull functional form and
16 a 12-hour W126 ozone index were \
7 Several crops included in thelJCLAN iraearch program were not evaluated hi mis analysis.
11 Non-commodity crops that ami|ot iaodeled in\|i|||^ ajricultural sector models were not. included hi
19 mis analysis: lettuce, tomato^|||atpes, alM&^jjjUsco, turnips, and kidney beans. In addition, one
20 commodity crop, spring whe9^ pa^ccihided because the NCLAN exposure-response function was only
21 developed for winter wheat, f
Miiumain/MaxiiaiiiiiExDOfiire-R
" -_ _ - -!^j: i - ~ *r- "-cT-aTji
23 Estimated responsro0|c||^a given crop to ozone varies within the NGLAN data. This range of
24 response is partially explaiij|i||||pne program's evaluation of several cultivars for some crops; ozone
2s sensitivity varies across cuMvam In addition, the conditions for different experiments varied due to
26 variations in location, year/and additional treatments included hi some experiments. No one exposure-
27 response function can be assumed to be representative of all cultivars hi use, or of all environmental
2t rooditipiis for crop induction. To develop a nmge of benefits estimates that reflects this variation in
responsiveness.
Minimum responsiveness and a maximum responsiveness function were selected for
each crop. Mutuality, a number of different cultivars are planted by producers, and so ozone response
31 will be a weighted average of the responsiveness of each cultivar to its ozone condition and its
32 proportion of total acreage. It is important to note that these values do not necessarily bound the
jj analysis, since the number of cultivars evaluated by NCLAN is small relative to the number grown for
34 many crops.
15 For die crops usedin this study, CERL conducted an analysis to identify the ozone concentration
36 required to reduce yields by 10 percent for each crop cultivar using its 12-hour W126 exposure-response
259
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
function. For each crop, the function demonstrating the lowest ozone concentration at a 10 percent yield
loss represents the maximum response, and the function with the highest concentration at 10 percent
yield loss represents the minimum response. Table 87 reports the minimum and maximum exposure-
response functions for each crop. Two crops, peanuts and sorghum, did not have multiple NCLAN
experiments on which to base a comparison of the responsiveness of different cultivars or the variation in
response with different experimental conditions. : ^__ :
.- -• .• t. .- -V
^ , V;/;3| ;*„ ,1 "
''r ""s" %
Cont-Picid
PAG397
Conrf'idd
S3
12$
9
10
11
11
13
14
a
16
Ca/cu/at/on of Ozone
Each NCLAN ozone
ise experiment exposed each studied crop over a portion of
the crop's growing season,Hie^duration of the NCLAN experiments was provided by CERL and was
rounded to the nearest month. The W126 index is cumulative, and so is sensitive bom to the duration
over which it is calculated and to the specific month(s) within a growing season mat are included in it
Because cropping sejppns vary across the U.S., the ozone index used to calculate county-level changes hi
yjejd duetoozonejoiust reflect the local season for each crop. To determine which portion of the
growing season irparticular exposure period pertains to (in order to calculate an exposure index), we
developed state-specific growing seasons based on planting and harvesting data developed by USDA.201
The W126 index was calculated using the county level ozone data developed hi .the prior section,
*' USDA, 1984. Some ftata did not tarn explicit growing leasom reported fbr certain crops
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
summed for the number of months of NCLAN experimental duration, with the exposure period anchored
j
.2 on the usual harvest month for each crop.202
3 Calculations of County
4 • • . - -•""-"."" . . •
s Because the benefits analysis did not require a regional level of disaggregatibn and to minimize
6 computational burdens the economic analysis was conducted at a national level. Ozone data and
7 estimated yield responses, however, were developed at a county level. To conduct a national analysis*
i the county level yield change estimates were weighted to develop a single national percent relative yield
9 loss for each crop relative to me control scenario, for both the mm
10 'responses. As a part of calculating a percent change hi vield at the national level, weights for each
n county and crop were created for 1975,1980,1985, and 1990. Tlie weights for these four years were
11 used to represent the year itself and the four years immediately following it (e.g., 1975 weights were also
u used for 1976,1977,1978, and 1979). Although weather and other conditions may change the
14 proportion of counties' production to the total national production in eachyear*five year weights should
is reflect stable periods of agricultural policy between eflch Farm Bill, and ari sitSf cient for the level of
16 precision needed for this analysis. The weights were calculated by dividing the production level of a
17 crop in a county203 by the sum of all states' reported production fb^lhat crop.204 These county weights
/s were applied to the percent relative yield loss results for^h county, as discussed below.
Calculation of Porcorit Chango In Ylold
20 Ozoneexposiu^respOB|| junctions arei|p||g!dln terms of percent relative yield loss (PRYL);
21 the ozone level hemgaiialy^iii^^ To calculate
22 a percent change in yield betweegM^^control and no-control scenarios, we first calculate a PRYL based
23 on the county-level control scenMdI|]i& ozone index, and a PRYL based on me no-control scenario
24 W126 index. Nex^tfaecojpty weigi||an|fjpplied to the PRYLs. The change in yield, measured relative
25 to the hypometieaLge^^
26 To obtain the change in terms of our (non-zero) baseline yield, we divide by that yield, and get:
302 This analysis required "rounding" some months: if a hwest
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
f ^ .•"£ v ', > v " , w% A •"•"/
X "-AV +%
»Al.*ri4 *•'' ^ ' s $
To create the national percent change in yield for each crop, the results of this equation are
summed for each scenario (maximum and minimum) and for each year. Tables 88 and 89 present the
resulting percent yield changes that were used as inputs to the economic model. '
Table 88.
Year
Barley
Catttat
Crop
Winter Wheat
1*75
1976
WJ
1979
CM*.
1982
1983
1984
1985
1986
1988
1989
•* j
1990
-0.000013
^.000013
-0.000019
•0,011936
-0.017505
^0.013114
^.000291
^J.000024
-0.000024
-0.006635
-0,024048
-0.015150
^0,017606
-Ot013067
-0.001166
-0.002171
-0,000717
-0.001841
-0.001118
-0.002480
•'
-0.017295
-0.014269
•G.014200
-0.004841
-0.005464
-0.005894
-0.003964
'^Mawrss
%s •« •••• .••*..
•QMS**
-OXW3854
-0.047395
-0^01564
-0.001812
-0^)07316
-0.019873
-0.007605
-0.001567
262
-------
1
3
4
;
6
7
S
9
10
11
12
13
Appendix F: Effects of Criteria Pollutants on Agriculture
^#^toteWHdmm9mj^l "
- -£_-}.. vrjd-
I. . ""^t^i.
-~ -^T-. ~:™~- ssfe^l^p":
1 ~ = - *- XJ-'JT --*>
--f^^ '^Jfllil'
t ,e o
into an economic model
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
I
2
3
4
S
6
7
I
9
10
11
.12
13
14
IS
If
17
II
19
20
21
22
23
24
23
26
27
21
29
30
31
32
33
Taylor, modified AGSIM for this analysis to reflect production conditions and policies as they changed
through the 20-year span of the Clean Air Act, from 1970 to 1990. During this period, U.S. farm policy
parameters changed every five years with the enactment of each Farm Bill, and producer participation
varied significantly over the period. Over this time, due to policy, weather, technological development,
and other variations, production levels and prices have varied, as have production technologies, costs of
production, and relevant cultivars. To reflect these changes, Dr. Taylor re-estimated demand
relationships for three periods (1975-79; 1980-84; and 1985-89) based on the farm policies in effect in
each period, and structured the model to run on a national level rather man alpeglpil^vel. The period,
from 1970-1975 was not modeled because of data limitations and tecause there wirlimited impact from
the CAA on ozone levels during that period. J7 *-y~
The AGSIM baseline production and price data serve aflhe control scenario I
relative yield losses (PRYLs) between the control and no-control scenarios are the relevant Input
parameter for this analysis, from which AGSIM calculates new yield per planted acre values. Based on
these values (as well as on lagged price data, ending stocks fittm the|^ivions year, and other variables),
AGSIM predicts acreage, production, supply, and price parameters for e^clkcfpj for each year, as well as
calculating yield per harvested acre. From these results andthe demand relationships embedded hi the
model, AGSIM calculates the utilization of each croj>{L
well as the change in consumer surplus, net cnpincomi
support payments. Net siuplus is calculated^ net crop
payments. The first year of results is 1976 because AGSIM mi
Table 90 presents the net <
cumulative present value (discounttiptt 3,
The positive surpluses exMbhe|JiViImost all j
lower ozone levels man those predicted to V
feed usejother domestic use, etc.), as
and other government
ief surplus, less government
one year (1975) of lagged data.
lie surpluses in bom nominal terms and as a
• the period 1976-1990 due to the Clean Ah* Act
t of the increase in yields associated with
: no-control scenario. The present value of the
estimated
ural
in me:m
represents the
response:
response fimctioa. The!
crop responses, and it is
uniform response to
likely magnitude of agriculjfir
notthe precise value of thole benefits.
L range? Between $4.6 billion in the minimum response case to
[response case. It should be reiterated mat this range
yjfthe acreage planted to a given crop had an ozone
theipinpium available response function or the maximum available
response functions do not necessarily bracket the true range of potential
anticipate mat all acreage will be planted in cultivars with a
I These considerations notwithstanding, these values do indicate the
fits associated with control of ozone precursors under the CAA, but
-- 3- *;..:
264
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
TabJeSO.
ragrant
friinEaum
• Caaoge
..NMCMjp-
hitBiom
ftffaiBMHfft Axaxin^Qn
CSangete
MtodmBni
ISW77
0
*
23
1979/80
1989/81
361
I9S2A3
m
4»
"»,
,-s^
" 5
^59
/ 47
a®*
33Q*
199
458
H8-
202
4H»
852
-283
Ifi?
2»
id
202
1188
19*3/84
1984/85
198S/BS
243
1515^
1373
1475
24!',,
1987/88
«»;»
198SWW
et*
494
5W
344
384:-
2089
1715
1593
•==_
Conclusions
9
10
11
12
V; Agricultural benefits associated with control of ozone precursors under the Clean Air Act are
likely to be fairly large} "Because it is possible that over tune producers have adopted more ozone-
resistont cjultivarSjiiftniay be appropriate to consider the lower end of the range of predicted benefits to be
mote indicative of the likely total benefits. The estimates developed in mis analysis, however, do not
represent all of the likely benefits accruing to agriculture, in that many high-value and/or ozone sensitive
crops could not be included in the analysis due to either exposure-response data limitations or
agricultural sector modeling limitations. The second consideration implies that benefits will likely be
larger than estimated. The minimum case may be the most appropriate starting point, however, due to
the first consideration: the current crop mix is probably biased toward lower ozone responsiveness.
Therefore, we anticipate that cumulative total agricultural benefits from the Clean Air Act are on the
order of under ten billion dollars (real terms).
265
-------
'•>+ 1
-------
o ., -* v- 5-v- ; -->,- ,^v ^ j - ;-> a
It
II
r
:i
-------
^ •*• x* ** % v
% A
n*ir
v'0 ^| ,V v >,> v»i^x ^^'
fl
V*«. .s,.v' s V&- V*j*'' '''Kf ^''w *^ ""x'f 1"
^Wfefe'^' < }*^-
r
i
\'
i * •. -••,, "' w+ •V'^
< A s ^ * *,
•. ^ v-. •" -^^ X s ' %*
-------
I
I;
o\
-------
-------
r
"" /
tltt
*f
If
II
-------
ii
,
-'< '^-'\ * - " - r
~- A>*
K H i & fe ^ I p'^'I^S^l i
[I
-------
-------
.
' '
> tit
s*
1
^; * '=-
fte
iti
%, ><,•• v '
' '•••• •>
!*^:
II:
I
f
-------
I
i
-i •> ' % -uP* 'V v V ^ ••
, ^ ' ' "V-iM^-s. /-a^X^f
' '"" J *• ^".Zf'?*"'^^
ilS^PJli;
^^ f*+ <^ TW 1M ^ "v^ s"^" ^7 "
-------
-------
•«" Jw. *" «• t,s '•<*'• f "•
"*f '•'"'" % *" -r.
5tifc ttft U^ ~^V lifr ift1 M^** ^ff '^fr >ft -^fr J
illlSflPII*
rf_». A^. ^Z v^ AV jjJ j^Z •*+ ^J1 j.1 -uZ ^C i^
/ ^ -^, ^ ^ X -^,^V
-------
00
sssssssss
II
-------
I
$ ?*
14(1
*•->••.,<<
%
"
ill
I
IS
-------
n
;ft!
*r
I1IIJ
ilill^mft
f
>
1
-------
£S
j 1 ^ '
.-' '}'•
•»•- '$.>'*'<•.'
I
f tr> Jt
-------
N)
li
^
- r T
IsJ ?**: S'"S
»* #
2*
li
1
-------
ts
-------
- X*
,'v''' '' £» ^ '',
'1
E
I
I
I
-------
lS ****'***«'a
1
„
, •*
-------
Ii
I
II
l|ll:|ll I'll il!
^
ii
-t
in
-', r
*>: A. $1
-------
i
i;
g
i
li
i
;***
v <
"***>
-------
1
If
;
.
* •**
X
I
-------
,- / j~ ':"-: v; <-
I
* \' ,',^XS^
, ' A* % •» S <
•?'•' ^ •• /^
Os
00
-------
-------
*« sf f Sif
„,- ' ' -,'\'V < -'•=•- V/* ']v
'\** / ^
I
' V ™%% '/* /v*s' *> »*,. 'I
•" ^ , ^^ ^ !•"'%'
^ >' •* f ,"••*•• "• '*" A v -V
v , ^» i ••' ; 4 >
'i
-------
5 a is
il
f
II
i
-------
i!
,1
j
i
I!
I;
II
i
•• v*j£ x-~
-x * ,
S?i
• ^
' &S 2&'iS
-------
..„.. . .f ... ^
^s : v>._ t t J V^v^ •/•• ^ f '•
v - ;&**/•#> i-^^-V j
s '
1
I
t**£****
-------
I.
W irsr*
IISIPSI
m r**
-------
-------
il
P
w .^
'V
-^ 8
-------
oo
-------
I
aiaffa.i
I*
ii
,g *'•»
'f-T-.*
-------
if
II
fc*l
i 1-1II
:*..
5 & i* ^ >5» S
« ift IS ®'-|i ii _.
t"** * ** *
11
i
-------
i
1
i
!l
act
-IA %' .v*\.
"SO A * '
O
CO
-------
S
--xv\ ^ x, ;>?
V •- -f *X^ W-- +fr-.^-
•• \\^%'*'> i.^^""
7v
I
l!
III;
**
*
i
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
2
3
4
5
6
1
I
9
10
li
12
13
14
a
6
17
U
19
Agricultural Effects References
Lefohn, Allen S. et al. 1988. A comparison of indices that describe .the relationship between exposure to
ozone and reduction in the yield of agricultural crops. Atmospheric Environment 22: 1229-1240.
Lee, E. Henry et al. 1994. Attainment and effects issues regarding alternative secondly ozone air
quality standards. J. Environ. Qual. 23:1129-1140. . '
National Acid Precipitation Assessment Program (NAPAP), 199 1 . 1 990 Integrated assessment report.
National Acid Precipitation Assessment Program, 722 Jackson Place NW, Washington, D.C.
20503. yv;,_
SAI, ICF Kaiser. 1995. Retrospective Analysis of ozone air quality in the United States: final report
Prepared by Systems Applications International under contract 68-D4-0103. U.S. EPA, Office
of Policy Analysis and Review. =.-:V"--- -:-- .-
Shriner, D.S., W.W. Heck, S.B. Mclaughlin, Wf. JohiiM^^JN&lrvJng^Jj. Joslin and C.E. Peterson.
1990. Response of vegetation to atmo^hericderwsition and air pollution. NAPAP SOS/T
Report 1 8, In: Acidic Deposition: State of Science and Technology, Volume m, National Acid
Precipitation Assessment Program^ 722 Jackson Place NW, Washington, D.C. 20503.
USDA1984. Usual Planting and Harvesting Da^es for U.SjField Crops. Statistical Reporting Service
W ''
Statistics Service Dataset (Electronic File)
USDA 199|;
93100Aand93100B.
- - -, -;<_ifi'ri,
303
-------
Appendix F: Effects of Criteria Pollutants on Agriculture
flTiis page intentionally bEHflt)
304
-------
Appendix Gs Lead Benefits Analysis
Methods Used to Measure and Value Health
4
5
6
7
S
9
10
11
12
13
14
IS
16
The scientific understanding of the relationship between lead and human health Is rapidly
expanding. This expansion is documented in numerous EPA studies on the health effects associated with
lead exposure. In a pioneering study, Schwartz et al. (U.S. EPA, 1985) quantified a number of health
benefits that would result from reductions in the lead content of gasoline. The work was extended by
EPA's analysis of lead in drinking water (U.S. EPA, 1986a) and by an EPA-funded study of alternative
lead National Ambient Air Quality Standards (U.S. EPA, 1987).
Despite this substantial research, much uncertainty remains. While die health effects of very
high levels of Blood Lead (PbB) are quite severe (including convulsions, coma and death from lead
toxicity) and have been known for many years, the effects of lower lead doses continue to be the subject
of intensive scientific investigation. Dose-response functions are available for only a handful of health
endpoints associated with elevated blood lead levels. -Other known or strongly suspected health
endpoints cannot be quantified due to a lack of information on the relationship between dose and effect
Table 128 presents the health effects that are quantified hi this analysis, as well as important known
health effects that are not quantified. - ;. - ' '
Table 128.
-s ' '.."' •. 'i ,
,,,'2 ' "-• ,,,,,->," '-'
t«iqim>fe>&*hi£m
II
19
Some of the health effects mat are quantified in this analysis have not been estimated hi previous
EPA analyses. This is largely .due to more recent information about the dose-response functions that
305
-------
Appendix G: Lead Benefits Analysis
i
2
3
4
5
6
7
makes it possible to expand the health effect coverage beyond what was done previously. Recent
information is available for previously unquantified health effects, and new information on previously
estimated dose-response functions is also available.
The following sections present relevant dose-response relationships for three population groups:
children, men, and women. These sections also discuss data sources used for the dose-response
relationships, although an extensive review of the literature is not given.205 In addition, each section
includes the methods used to value the changes in health effects determined using these dose-response
relationships. .
10
12
13
IS
16
II
19
20
21
a
23
24
25
26
27
28
29
30
31
33
Health Benefits to Children
Changes in IQ
Elevated PbB levels may induce a number of efi||cj| on the human nervous system. Generally,
these neurobehavioral effects are more serious for children than for adults because of children's rapid rate
of development It is believed that neurooehavicn^derlcitem children may result from both pre-natal
and early post-natal exposure. These nervous system ef£^^^^^^iiypm^\ityt behavioral and
attentional difficulties, delayed mental development, and motor and ercetual skill deficits.
Quantification of certain manifestations ofjmese effects is possible because sufficient data exist to
estimate a dose-response relationship andlQ loss. The relationship used in the analysis is discussed
below. • i-def'"
Quantirying the RelatioMhip Between
research
linear regressioacoejQBcieQt for
wei
Leveb and IQ
IQ decrements has been estimated by a meta-analysis of seven
.^ each study were used to determine a weighted average
between lead and IQ. Each regression coefficient was
estimate. To determine an overall coefficient, the
used natural logarithms of lead as the exposure index were
t was linearized in the blood lead range of 10 to 20 ug/dL.
1991), 70 percent of me data was below 10 ug/dL; thus, the
5 to 15 ug/dL range. For the studies that did not transform lead
concentrations, the regression coefficients were used directiy. Given the typical uncertainty within
individual studies, the variation in the regression coefficients among studies was not more man would be
expected. The relationship determined by Schwartz (1993) suggests mat for a 1 ug/dL increase in lead, a
decrease of 0^5 IQ,points can be expected. The p-value (< 0.0001) indicates that mis relationship is
highly signific
linearized. In general,
However, hi one study
Bellinger data was linearized
105 For • detailed review of this literature see U.S. Environmental Protection Agency, (1986) Air Quality Criteria Document far Lead, and
1989 Addendum. Environmental Criteria and Assessment Office. Office of Research and Development, March.
"• Schwartz, 1993.
306
-------
Appendix G: Lead Benefits Analysis
To obtain the total change in number of IQ points for a population of children, the 0.25 points
lost per ug/dL change in blood lead is multiplied by the average blood lead level for that population. The
average blood lead level modeled in this analysis is a geometric mean, not an arithmetic mean. To adjust
for this, we use a relationship between the expected value and the geometric mean of a lognormally
distributed random variable:
* +
70
11
-where E(X) is the expected value (mean) of the distribution, GM Is the geometric mean, and GSD is the
geometric standard deviation. Taking the natural logarithm of Equation 39 and rearranging gives the
ratio between the expected value and the GM: * ' i
(40)
For a GSD
and GM is 1.1 17. This
The total lost IQ points for each group was estimated as:
D of children's blood lead levels207), the resulting ratio between E(X)
n equation 43.
i.m
72
13
14
where (PopX represents the number of children (up to age six) around a given industrial source (in the
case of estimating benefits from reduced industrial emissions) or the total U.S. population of children (in
the case of estimating benefits from reductions in gasoline lead emissions).
m Suggested value for sub-populations provided by IEUBK guidance manual. (EPA, 1994)
307
-------
Appendix G: Lead Benefits Analysis
i
2
3
4
5
6
7
S
9
10
11
12
13
14
IS
16
17
U
19
20
21
22
23
24
25
26
27
2S
29
30
31
As shown in equation 43, the population of children up to age six is divided by seven to avoid
double counting. If we assume that children are evenly distributed by age, this division applies this
equation to only children age 0-1. If we did not divide, this equation would count a child who is age zero
in the first year of the analysis and count that same child 6 more times in successive years. Dividing by
seven does creates some undercounting because in the first year of the analysis children from age 1 to 6
are not accounted for, while presumably they are affected by the lead exposure. ;L
The current model assumes a permanent loss of IQ based on blood Iead10t^ls estimated for
children six years and younger. EPA, 1986b (Vol IV) states that ill a series of studies on a group of lead-
exposed children, "IQ deficits... were no longer evident at the five year follow up." More recent
studies,201 though, provide concrete evidence of long-term effects from childhood lead exposure. Future
exploration of this assumption is probably warranted. ^ ':&-1-f '-_-. w
- ~S.3-ii if"
Valuing Changes in Children's Intelligence
Available economic research provides little empirical data fbrsocietv^s^willingness to pay (WTP)
to avoid a decrease in an infant's IQ. As an alternative measure, it was assumed Ifiat IQ deficits incurred
through lead exposure will persist throughout the exposed in&ofs lifetime. Jfwo consequences of this IQ
decrement, representing a portion of society's £pwiUmgnessjD|^ are then considered: the decreased
present value of expected lifetime earnings forthe infam^ andlhi|||is|sed educational resources
expended for a infant who becomes mentally handicapped or is mli^ of compensatory education as a
consequence of lead exposure. The value $f foregone earnings j| addressed in mis section.
,4F -if
The reduction in IQ has a i
straightforward — lower IQS <
reduced educational attaim
these effects ffttiaxnjngs i
effects separately^
Direct
Henry Aaron, ZvP
relationship between IQ and
landinc
teffecton earnings. The direct effect is
Reduced IQ also results in
Imings and labor force participation. Note that
the studies used for mis analysis have controlled for these
and Paul Taubman have reviewed the literature examining the
earnings.209 They find mat the direct effect, (schooling held
constant) of IQ on wage rates ringed from 0.2 percent to 0.75 percent Perhaps the best of these studies
is Griliches (1977).210 He finds mat the direct effect of IQ on wage rates to be slightly more than 0.5
percent per IQ point Because mis is roughly the median estimate of the USEPA review of the literature,
misestimate is
** For example, Bellinger 1992.
*» USEPA 1984.
110 Griliches used a structural equations model to estimate the impact of multiple variables on an outcome of interest This method has
conceptual advantages over other empirical estimates used in me literature because hsuccessfuUy controls for me many confounding variables
that can affect earnings.
308
-------
Appendix G: Lead Benefits Analysis
i Indirect Effect oflQ on Earnings
2 From Needleman et al. (1990) it is possible to estimate the change in years of schooling attained
3 per one IQ point change. Their regression coefficients for the effect of tooth lead on achieved grade
4 provide an estimate of current grade achieved. However, many of these children were in college at the
5 time and are expected to achieve a higher grade level, Following Schwartz (1990), after adjusting the
6 published results for the fact that a higher percentage of children with low tooth lead were attending
/ college, a 0.59 year difference in expected maximum grade achieved between the high and low exposure
s groups was estimated. It is assumed that educational attainment relates with blood lead levels in
9 proportion to IQ. The difference in IQ score between the high and low exposure group was 4.5 points.
10 Dividing 0.59/4.5 = 0.131 suggests that the increase in lead exposure which reduces IQ by one point may
n also reduce years of schooling by 0.131 years.211
12 Studies that estimate the relationship between educational attainment and wage rates (while
n controlling for IQ and other factors) are less common. Chamberlain and Griliches (1977) estimate that a
H one year increase in schooling would increase wages by 6.4 percent. In a longitudinal study of 799
is subjects over 8 years, Ashenfelter and Ham (1979) reported that an extra year of education increased the
16 average wage rate over the period by 8.8 percent. Conservatively, we use a lower bound by assuming
n one year of additional schooling increases the wage rate by 6 percent.212 To arrive at the indirect effect
is . of increased schooling, increased wages per IQ point is calculated using: (6 percent wage increase/school
19 year) x (0.131 school years/IQ) = 0.786 percent increase in wages per IQ point.
20 There is one final indirect effect on earnings. Changes hi IQ affect labor force participation.
21 Failure to graduate high school, for example, correlates with participation in the labor force, principally
22 through higher unemployment rates and earlier retirement ages. Lead is also a strong correlate with
23 attention span deficits, which likely reduce labor force participation. The results of Needleman et al.
24 (1990) relating lead to failure to graduate high school can be used to estimate changes in earnings due to
25 labor force participation. Using the odds ratio from Needleman et al., it was estimated that a one IQ
26 point deficit would also result in a 4.5 percent increase in the risk of failing to graduate. Krupnick and
27 Cropper (1989) provide estimates of labor force participation between high school graduates and non-
2s graduates, controlling for age, marital status, children, race, region, and other socioeconomic status
29 factors. Based on their data, average participation in the labor force is reduced by 10.6 percent for
30 persons failing to graduate from high school. Because labor force participation is only one component of
31 lifetime earnings (i.e., earnings = wage rate X years of work), this indirect effect of schooling is additive
32 to the effect on wage rates. Combining this estimate with the Needleman result of 4.5 percent increase in
33 the risk of failing to graduate high school per IQ point, indicates that the mean impact of one IQ point
34 loss is a (10.6 percent x 4.5 percent) = 0.477 percent decrease in expected earnings from reduced labor
35 force participation.
"' Following Schwartz (1990s), this analysis uses the Needleman (1990) to quantify the change in grade achievement from lead
exposure.
212 A conservative rather than a best estimate (e.g., such as 7.6 percent, which is the average of the two cited studies) is used herein to
attempt to account for mitigating factors that we may have missed. For example, some have suggested that the cost of additional schooling
should be netted out of the benefits calculation. The average cost of one year of schooling in the U.S. in 1990 ($5,000 per year to $6,000 per
year) was approximately '/> the difference in changed lifetime earnings implied by using 6 percent rather than 7.6 percent
309
-------
Appendix G: Lead Benefits Analysis
i Combining the direct effect of 0.5 percent with the tw6 indirect effects (0.786 percent for less
2 schooling and 0.477 percent for reduced labor force participation) yields a total of 1.76 percent decrease
j in earnings for every loss of one IQ point.
4 Value of Foregone Earnings
' s ' In the next step to monetize intelligence effects, the percent earnings loss estimate must be
6 combined with an estimate of the present value of expected lifetime earnings. Data on expected lifetime
7 earnings as a function of educational attainment and sex was reported for the U.S. population in 1979 by
» the Bureau of the Census.213 Given the distribution of the 1979 population with respect to age,
9 educational attainment, and sex, the Census used age specific employment rates and average wage rates'
w to estimate annual earnings as a function of age, sex and education. Assuming various rates of real wage
/; growth (productivity effect) and discount factors, the annual earnings stream from age 18 to age 64 was
12 collapsed to a series of estimates of the present value of lifetime earnings using an assumption of one
13 percent real wage growth and a three percent discount rate. Men tend to earn more than women because
14 of higher wage rates and higher labor force participation. However, for both men and women, expected
is lifetime earnings increase greatly with education.
16 The Census estimates were expressed in 1981 dollars and assumed that the age/education
17 specific employment and average wage rates would remain constant over time. A number of issues must
is therefore be addressed in updating the Census estimates to 1990 dollars. First, educational attainment
19 has changed since 1979, with a greater proportion of the population attending college (especially a
20 greater proportion of women). Second, wage rates have increased both due to productivity effects (real
21 wage growth) and inflation. Third, age-specific employment rates may have changed. Women, in
22 particular, are likely to have higher rates of labor force participation than in 1979.
23 In revising the Census estimates, the first issue was addressed by using more recent data on
24 education. USDOC (1992) provides data on educational attainment for the 1991 population. For this
25 analysis, data on the population over age 25 were used in order to remove the influence of those
26 individuals too young to have completed schooling. The population data were used as weighting factors
27 to derive a sex and education weighted average of expected lifetime earnings. So constructed, the
28 weighted average adjusts the estimate based on 1979 data to current levels of educational attainment.
29 The present value of lifetime earnings were thus calculated to be $349,000 (discounted at three percent)
30 for the average work force participant. The next step in adjusting the earnings estimate is to apply an
31 adjustment for wage growth to update from 1981 to 1990 dollars. The Bureau of Labor Statistic's
32 Employment Cost Index rose from a level of 67.2 in 1981 to 105.4 in 1990, an increase of about 57
33 percent. Updating the average lifetime earnings to 1990 dollars yields a revised estimate of $547,000.
While more recent age, sex, and education-specific employment rates could be used to re-
cMimaie labor force participation, a complete analysis requires steps to dampen the effects of cyclical
unemployment. Such an exercise would require considerable effort and is beyond the scope of this
analysis. To the extent that labor force participation has increased for specific groups since 1979, the
adjusted value presented here underestimates the true expected lifetime earnings. For example, if the
34
35 estimate
36
37
3S
213 USDOC 1983.
310
-------
Appendix G: Lead Benefits Analysis
i percentage of female children eventually joining the permanent workforce is greater than the percentage
2 of women over age 25 that worked in 1979, the expected lifetime earnings of female children would be
3 greater than estimated in this analysis.
4 Note that use of earnings is an incomplete measure of an individual's value to society. Those
5 individuals who choose not to participate in the labor force for all of their working years must be
« accounted for, since the lost value of their productive services may not be accurately measured by wage
7 rates. The largest group are those who remain at home doing housework and child rearing. Also,
« volunteer work contributes significantly to social welfare and rates of volunteerism tend to increase with
9 educational attainment and income.214 If the opportunity cost of non-wage compensated work is assumed
10 to be the average wage earned by persons of the same sex, age, and education, the average lifetime
11 earnings estimates would be significantly higher and could be approximated by recalculating the tables
n using full employment rates for all age/sex/education groups. To be conservative, only the value of lost
13 wages is considered in this analysis.
14
IS
16
17
The adjusted value of expected lifetime earnings obtained above is a present value for an
individual entering the labor force at age 18 and working until age 64. Because a lead-induced IQ
decrement occurs in childhood, the $547,000 figure must be further discounted to the specific age at
which the health effect is measured and adjusted for the probability that the infant would survive to age
is 18.215 For an infant less than one year old, the present value of lifetime earnings discounted at three
19 percent from age 18 and adjusted for survival would be $315,370. Combining this value with the
20 estimate of percent wage loss per IQ point yields: $315,370 x 1.76 percent = $5,551 per lost IQ point.
21 Children with IQs Less Than 70
22 Quantifying the Number of Children with IQs Less than 70
23 Unlike the dose-response function used for IQ point loss, the dose-response function used to
24 estimate the lead-induced incidence of IQ values below 70 is not a constant function across all blood lead
25 levels. The dose-response function used in this analysis is derived from Wallsten and Whitfleld (1986).
26 Based on encoded expert opinion, Wallsten and Whitfleld provide estimates of expected percentages of
27 children with IQs below 70 for specified levels of blood lead. The opinions of the experts were averaged
2a to develop a dose-response relationship for the risk of IQ falling below a threshold of 70 points. The
29 expert judgements relating blood lead levels to mean IQ decrements were averaged to come up with a
30 mean (over all experts) response rate of children having IQ below 70 points for discrete PbB levels
31 ranging from 5 to 50 ug/dL (the specified levels varied by expert). Next, the PbB distribution was
32 divided into 200 intervals from 0 to 50 ug/dL (i.e., 0.25, 0.5, 0.75, etc.). By interpolation of the mean
33 values obtained from Wallsten and Whitfleld, a mean response rate was associated with each of the 200
34 PbB intervals. Then, within specific ranges of the PbB distribution, all the data points were used to
35 estimate a linear regression equation relating PbB value to mean response rate. The result is a piecewise-
36 linear function of mean response rate for IQ below 70 points (see Table 129).
214 Statistical Abstract of the United States, 1986. Table No. 651, p. 383.
215 Special education costs for children who do not survive to age 18 are not counted, which results in some underestimation of benefits.
However, most child mortality occurs before the age of 7, when the special education begins, so this under-counting is not substantive.
311
-------
Appendix G: Lead Benefits Analysis
i
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
IS
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Table 129. Elements of Kecewise Linear
Function for Estimating Probability of IQ < 70 i
a Function of Blood Lead (PbB) Range.
(M/dD
0.5,0
5,1-7.5
7.6-10*
10.1 - 12^5
12.fr- 1J.O
15.1 . 175
22.6 -25J)
>2«J
Slope
.2.04*10*
. 4.88*10'
U&xlO*
t.04*xtO»
9.76 ^ 1O*
U6X10*
15*2X10.
1.464X 10*
te««*
3J»*W*
2.1S x 10*
-2.17x10*
«!.» x 10*
-1.08 x 1O*
-5.*3 x 1O*
4.112 x 10*
-9.42 x 10*
This function quantifies only the change in
the number of children who pass below the IQ=70
threshold. Any other changes in children's IQ are
quantified using the IQ point loss function. Using
these two endpoints additively does not result in
double counting, because the value associated with
the IQ point loss function is the change in worker
productivity while the value associated with IQs less
than 70 is the increased educational costs for the
individual.
The piecewise linear function is used in
conjunction with the PbB distribution of a population
to estimate the response rate associated with each
percentile group. The PbB distribution is developed
for the population given geometric mean PbB, the
assumed GSD. Once the PbB value is calculated, the
slope and intercept coefficients for the appropriate
linear segment in Exhibit 2-2 are used to calculate the
mean IQ response rate for that PbB percentile. This ^^^^^^^^^^^^^^^^^^^^^^^^
response rate is multiplied by the exposed population
in each PbB percentile grouping to estimate the
proportion of infants in that group having IQ less than 70 points. Summing the number of infants below
70 points for each PbB percentile group yields the total number of cases for the exposed population.
As in the IQ point loss equation, the results of this function are applied to children age 0-6 and
divided by seven to avoid double counting. (See discussion under equation 43).
Valuing the Reduction in Number of Children with IQs less than 70
To value the reduction in the number of children with IQs less than 70, the reduction in
education costs were measured - a clear underestimate of the total benefits.216 Kakalik et al. (1981),
using data from a study prepared for the Department of Education's Office of Special Education
Programs, estimated that part-time special education costs for children who remained in regular
classrooms cost $3,064 extra per child per year in 1978. Adjusting for inflation and real income growth
using the GNP price deflator yields an estimate of $6,318 per child in 1990 dollars. For the calculations,
this incremental estimate of the cost of part-tune special education was used to estimate the cost per year
per child needing special education as a result of impacts of lead on mental development. Costs would
be incurred from grades one through twelve. Discounting future expenses at a rate of three percent
yields an expected present value cost of approximately $52,700 per infant (assuming compensatory
education begins at age 7 and continues through age 18). Note that this underestimates the cost, since
216 The largest part of this benefit is the parents' willingness to pay to avoid having their child become mentally handicapped, above and
beyond the increased educational costs.
312
-------
Appendix G: Lead Benefits Analysis
i Kakalik et al. measured the increased cost to educate children attending regular school ~ not a special
2 education program.
3 Changes in Neonatal Mortality
4 Quantifying the relationship between PbB levels and neonatal mortality
5 U.S. EPA (1990) cites a number of studies linking fetal exposure to lead (via in utero exposure
6 from maternal intake of lead) to several adverse health effects. These effects include decreased
7 gestational age, reduced birth weight, late fetal death, and increases in infant mortality. The Centers for
s Disease Control (CDC, 1991 a) presents a method to estimate changes in infant mortality due to changes
9 in maternal blood lead levels during pregnancy.217 The analysis links two relationships. The first
w relationship, between maternal blood lead level and gestational age of the newborn, was estimated by
// Dietrich et al. (1987). CDC then estimated infant mortality as a function of gestational age, using data
n from the Linked Birth and Infant Death Record Project from the National Center for Health Statistics.
n The resulting association is a decreased risk of infant mortality of 10"4 (or 0.0001) for each 1 ug/dL
14 decrease in maternal blood lead level during pregnancy. This is the relationship used in the current
is analysis.
16
Valuing changes in neonatal mortality
n Fisher et al. (1989) conducted a review of studies quantifying individuals' willingness to pay to
is avoid risks to life. In the reviewed studies, value estimates were obtained from hedonic wage studies,
19 consumer market studies or contingent valuation studies. Based on that review, a range of $ 1.6 million
20 to $8.5 million (1986 dollars) was recommended when valuing statistical deaths for policy evaluation.
21 The Fisher et al. range has frequently been used by the U.S. EPA for a wide variety of policy analyses.
22 The above values are expressed in 1986 dollars and must be adjusted to be consistent with other
23 benefits estimates, which are expressed in 1990 dollars. One potential concern in simply adjusting the
24 above estimates for inflation is that society's willingness to pay is likely to increase as national income
25 increases. Since risk reductions are a normal good, willingness to pay is likely to increase with national
26 income. This is clearly an empirical issue but casual examination of the literature suggests that the larger
27 willingness to pay estimates come from studies in later years when incomes were higher. Since the
28 regulatory costs were expressed in 1990 dollars, the statistical life Values above are adjusted using the
29 relative change in nominal, per capita GNP from 1986 to 1990 (approximately a 25 percent increase).
SB Using the change in nominal GNP accounts for changes in income as well as inflation. The use of per
31 capita GNP eliminates the influence of population growth on national income. The adjustment results in
32 a range of $2 million to $11 million as the value of a statistical life.
33 The Fisher et al. range, adjusted to 1990 dollars, is consistent with other recent reviews of the
34 range of value of a statistical life. Viscusi (1992) reviews the literature on value of life estimates,
117 The estimated change in infant mortality due to change in birthweight was not modeled because the data relating prenatal lead
exposure to birth weight is not as strong as data relating lead exposure and gestational age.
313
-------
Appendix G: Lead Benefits Analysis
i including all of the research included in Fisher et al. as well as additional studies published through "~"N
2 1991. He concludes that-
3 "most of the reasonable estimates of the value of life are clustered in the S3 to $7 million
4 range (in 1990 dollars). Moreover, these estimates are for the population of exposed
s . workers, who generally have lower incomes than the individuals being protected by
6 broadly based risk regulations. Recognition of the income elasticity of the value of life
7 will lead to the use of different value of life depending on the population being
« protected." (p.74) . '
9 Viscusi thus recommends a narrower range that is wholly contained within the adjusted Fisher et al.
10 range, and supports the nature of the adj ustment to 1990 dollars described for Fisher et al.
;/ The value of avoiding a statistical death used in this analysis is $4.8 million. This is the
12 proposed value by Industrial Economics Incorporated in their Section 812 valuation work.2" There is
u some EPA precedent for using this figure as well. The U.S. EPA Office of Indoor Air (1994) identified
14 26 acceptable "value of life" studies and calculated the geometric mean estimated willingness to pay to
a avoid a "statistical death" as $4.8 million (1990 dollars). This same estimated value is used for all
16 premature mortality calculations in this report regardless of age, gender, racial or other demographic
17 characteristics of the population at risk.
I8 Health Benefits to Men
19 The health effects in adults that are quantified and included in the benefits analysis are all related
20 to the effects of lead on blood pressure.219 These quantified health effects include increased incidence of
21 hypertension, heart attack, stroke and mortality. Both men and women suffer from these lead-induced
22 health effects, but the estimated relationships differ for men and women. This section describes the
21 quantified health effects for men, and Section 2.4 describes the health effects for women.
24 Hypertension
25 Quantifying the relationship between PbB levels and hypertension
26 Elevated blood lead has been linked to elevated blood pressure (BP) in adult males, especially
27 men aged 40-59 years 22° Further studies have demonstrated a dose-response relationship for
28 hypertension (defined as diastolic blood pressure above 90 mm Hg for this model) in males aged 20-74
29 years.221 This relationship is:
211 IEc 1992, lEc 1993.
219 Citing laboratory studies with rodents, U.S. EPA (1990) also presents evidence of the genotoxicity and/or carcinogenicity of lead
compounds. While such animal lexicological evidence suggests that human cancer effects are possible, dose-response relationships are not
currently available.
220 Pirkleetal., 1985.
221 Schwartz, 1988.
314
-------
Appendix G: Lead Benefits Analysis
2.744 - >793 » 2.744 - 0.79J'(la J>W,)
(44)
; where: . . •
2 APr(HYP) = the change in the probability of hypertension;
3 PbB, = blood lead level in the control scenario; and
4 PbB2 = blood lead level hi the no-control scenario.
s Valuing reductions in hypertension
6 The best measure of the social costs of hypertension, society's willingness to pay to avoid the
7 condition, cannot be quantified without basic research well beyond the scope of this project. Ideally, the
s measure would include all the medical costs associated with treating hypertension, the individual's
9 willingness to pay to avoid the worry that hypertension could lead to a stroke or heart attack, and the
w individual's willingness to pay to avoid changes in behavior that may be required to reduce that
n probability that hypertension leads to a stroke or heart attack. Medical costs of hypertension can be
12 divided into four categories: physician charges, medication costs, hospitalization costs and lost work
13 time.
14 This analysis uses recent research results to quantify two components of this benefit category.
is Cropper and Krupnick (1989), using data from the National Medical Care Expenditure Survey, have
16 estimated the medical costs of hypertension. These costs include physician care, drugs and
n hospitalization costs. In addition, hypertensives have more bed disability days and work loss days than
is others of their age and sex. Krupnick and Cropper estimated the increase in work loss days at 0.8 per
19 year, and these were valued at the mean daily wage rate. Adjusting the above costs to 1990 dollars gives
20 an estimate of the annual cost of each case of hypertension of $681. The estimate is likely to be an
21 underestimate of the true social benefit of avoiding a case of hypertension for several reasons. First, a
22 measure of the value of pain, suffering and stress associated with hypertension is not included. Second,
23 the direct costs (out-of-pocket expenses) of diet and behavior modification (e.g., salt-free diets, etc.) are
24 not valued. These costs are likely to be significant, since modifications are typically severe. Third, the
2s loss of satisfaction associated with the diet and behavior modifications are ignored. Finally, the
26 medication for hypertension produces side effects including drowsiness, nausea, vomiting, anemia,
27 impotence, cancer, and depression. The benefits of avoiding these side effects are not included in this
28 estimate.
29 Changes In Coronary Heart Disease
30 Quantifying the relationship between blood lead and blood pressure
31 Because blood pressure has been identified as a risk factor in a number of cardiovascular
32 illnesses,222 it is useful to quantify the effect of changes in blood lead levels on changes in blood pressure
Shurtleff, 1974; McGee and Gordon, 1976; Pooling Project, 1978.
315
-------
Appendix G: Lead Benefits Analysis
i for reasons other than predicting the probability of hypertension. A meta-analysis of several studies
2 estimated that a 1.4 mm Hg change in blood pressure is expected for a one ng/dL change in PbB.223 The
3 benefits analysis uses the following equation based on the Schwartz relationship:
* 1.4 x (PbB, - PbBJ (45)
4 where: .
5. ADBPmcn = the change in men's diastolic blood pressure expected from a change in
< PbB;
/ PbB, . = blood lead level in the control scenario; and
« ' PbB2 = blood lead level in the no-control scenario.
9 This blood lead to blood pressure relationship is used to estimate the incidence of heart attacks (initial
10 coronary heart disease), strokes (atherothrombotic brain infarctions and initial cerebrovascular accidents)
;/ and mortality in men.
n Quantifying the relationship between blood pressure and coronary heart disease
is The relationship between blood pressure and other health effects can be used to predict increased
14 probabilities of the initial occurrence of heart attack and stroke.224 Increased blood pressure would also
is increase the probability of reoccurrences of heart attacks and strokes, but these quantified relationships
i6 are not available. First-time heart attacks (coronary heart disease events) in men can be predicted using
17 an equation with different coefficients for each of three age groups. For men between 40 and 59 years
is old, information from a 1978 study by the Pooling Project Research Group (PPRG) is used. PPRG
19 (1978) presents a multivariate model (controlling for smoking and serum cholesterol) that relates the
20 probability of coronary heart disease (CHD) to blood pressure. The model used data from five different
21 epidemiological studies. From this study, the equation for the change in 10-year probability of
22 occurrence of CHD is:
. 4.996 -
+'e I + «
23 .
24 where:
25 APKCHD^,,) = change in 10-year probability of occurrence of CHD event for men
26 between 40-59 years old,
27 DBP, = mean diastolic blood pressure in the control scenario; and
2s DBP2 = mean diastolic blood pressure in the no-control scenario.
29 The relationship between BP and first time heart attacks in older men was determined from
30 information presented in Shurtleff (1974). This study also uses data from the Framingham Study to
31 estimate univariate relationships between BP and a variety of health effects by sex and for each of the
223 Schwartz, 1992a.
224 U.S. EPA, 1987.
316
-------
Appendix G: Lead Benefits Analysis
i following age ranges: 45-54, 55-64, and 65-74 years. Single composite analyses for ages 45-74 were
2 also performed for each sex. For every equation, t-statistics on the variable blood pressure are
3 significant at the 99th percent confidence interval. For men aged 60 to 64 years old, first-tune heart
4 attacks can be predicted from the following equation:
- 5.19*76 - 0.02351 *DSP, 5. 1M7« - 8.0235 1
5
6 where:
,) = change in 2 yea- probability of occurrence of CHD event for men from 60 to 64
s years old;
9 DBF, = mean diastolic blood pressure in the control scenario; and
10 DBP2 = mean diastolic blood pressure in the no-control scenario.
;/ For men aged 65 to 74 years old, the following equation uses data from Shurtleff (1974) to predict the
12 probability of first-time heart attacks:
1 t
. - 0.0203HJ>M>, } 4.90M3 - 20J1
13
14 where:
is APr(CHD65.74) = change in 2 year probability of occurrence of CHD event for
16 men from 65 to 74 years old;
17 DBF, = mean diastolic blood pressure in the control scenario; and
is DBP2 = mean diastolic blood pressure in the no-control scenario.
19 Valuing reductions in CHD events
20 There are two categories of benefits associated with reductions in CHD events: the medical costs
21 and foregone earnings associated with the treatment of and recovery from a heart attack, and the
22 additional willingness to pay to avoid the pain, discomfort, and potentially reduced quality of life after
23 suffering a CHD event. While estimates of the medical costs and foregone earnings associated with
24 CHD events are given in USEPA (1987), these values are likely a small component of the total value of
25 avoiding a statistical case of CHD. There are no studies that explicitly measure people's willingness to
26 pay to avoid a CHD event. However, the EPA economic research program funded a series of studies to
27 estimate the willingness to pay to avoid health risks.
28 While a WtP value for avoiding a CHD event is not available, several studies estimate the WtP
29 to avoid a statistical case of chronic bronchitis. The present study uses the valuation for chronic
so bronchitis as a surrogate for CHD based on the assumption that people believe that the pain, suffering
n and activity restrictions associated with a non-fatal CHD event are at least as severe as those associated
n with chronic bronchitis. The results of the chronic bronchitis study valuation studies are summarized on
33 a consistent basis, and the estimated values are reported in 1990 dollars in Viscusi (1992). Viscusi et al.
34 (1991) and Krupnick and Cropper (1992) estimated the WtP for chronic bronchitis avoidance in terms of
35 willingness to trade the risk of chronic bronchitis for the risk of a fatal auto accident (a risk-risk
317
-------
Appendix G: Lead Benefits Analysis
i tradeoff). Four central estimates result from the studies: (1) a mean value of $883,000 (Viscusi etal.,
2 1991); (2) a median value of $45 7,000 which captures the skewness of the response distribution (Viscusi
3 et al., 1991; (3) $210,000 recommended by Rowe et al based on Viscusi et al. (1991) with an adjustment
4 for the severity of the chronic bronchitis case; and (4) a value of $800,000 derived from the two-step
s approach of Viscusi et al. (1991) but adjusting the skewness of auto death using the median value of auto
6 death. To derive a single point estimate of the value of an avoided case of lead exposure-induced CHD,
7. the present study uses the mean of the four estimates outlined above, which equals"$587,500, as the WtP
s for avoidance of CHD.
9 The uncertainty surrounding the point estimate is also derived from the four valuations given
w above. It is assumed that individuals are WtP one of the four valuation estimates to avoid CHD (equated
/; to chronic bronchitis). Without further information, it is not possible to weight the proportion of
12 individuals that would select each alternative value. Therefore, four discrete values, each with equal
n probability of selection, describe the .uncertainty distribution associated with this endpoint.
14 The valuation for hypertension is additive with the valuation for CHD despite the fact that the
is conditions often occur together because the two values represent different costs associated with the
16 conditions. The valuation for hypertension is based on loss of work days as a result of hypertension and
n some of the medical costs associated with treating hypertension. Alternatively, the valuation for a heart
is attack is based on the willingness to pay to avoid the pain and suffering of the heart attack itself.
19 Therefore, these two valuations can be separated and added together.
20 Changes in Initial Cerebrovascular Accidents and Initial Atherothrombotic Brain Infarctions
21 Quantifying the relationship between blood pressure and first-time stroke
22 Two types of health events are categorized as strokes: initial cerebrovascular accidents (CA) and
23 initial atherothrombotic brain infarctions (BI). The risk has been quantified for the male population
n between 45 and 74 years old.225 For initial cerebrovascular accidents, the logistic equation is:
I.5IU9 - 0.04066 *£JW>, V*")
• I +
25 where:
26 APitCA^J = change in 2 year probability of cerebrovascular accident in men;
27 DBF, = mean diastolic blood pressure in the control scenario; and
28 DBP2 = mean diastolic blood pressure in the no-control scenario.
123 Shurtleff, 1974.
318
-------
Appendix G: Lead Benefits Analysis
For initial atherothrombotic brain infarctions, the logistic equation is:
1
- 0.0«40»/«P
7 where: .
.) APrCBImen) = change jn 2 year probability of brain infarction in men;
4 DBF, = mean diastolic blood pressure in the control scenario; and
5 DBPj = mean diastolic blood pressure in the no-control scenario. .
6 ' Valuing reductions in strokes
7 As with hypertension, very few data exist for estimating society's WtP to avoid a lead exposure-
« induced stroke. Data on medical costs are available, but a value for pain, suffering, and reduced quality
9 of life has not been rigorously quantified. Thus, as in the case of CHD, the best option is to use the
w chronic bronchitis value ($587,500 per case avoided) as a surrogate. Again, the activity restrictions
// associated with strokes are assumed to be at least as severe as those for chronic bronchitis. Since strokes
12 often include lingering paralysis, actual WtP is likely to be quite substantial.
n Although the foregone medical costs of averted strokes are not explicitly included in the measure
14 of benefits, it is worth discussing the available data. Hartunian et al. (1981 ) estimated the medical
is expenses and foregone earnings associated with three types of strokes: hemorrhagic, infarctive, and
i6 transient ischemic attacks. Weighing these cost estimates by the distribution of these types of strokes by
17 age yields an average estimate of $48,000. Given that these treatment costs are based on old data, and
/* represent only a fraction of the WtP to avoid pain, suffering and reduced quality of life, they are not
19 included in the estimate of benefits per case avoided used in this analysis. Given that the chronic
20 bronchitis valuation is used as a surrogate, the uncertainty surrounding the estimate is the same as that
21 presented for CHD events above.
22 Changes in Premature Mortality
23 Quantifying the relationship between blood pressure and premature mortality
24 Information also exists to predict the increased probability of premature death from all causes as
25 a function of elevated blood pressure. U.S. EPA (1987) used population mean values for serum
26 cholesterol and smoking to reduce results from a 12 year follow-up of men aged 40-54 in the
27 Framingham Study (McGee and Gordon, 1976) to an equation in one explanatory variable:
$4 - - - -
5.3151 - 0.035 !*•/»?, . 5,3151 - 0.0351«*flW>,
319
-------
23
Appendix G: Lead Benefits Analysis
where:
APr(MORT«o.54) = the change in 12 year probability of death for men aged 40-54;
DBF, = mean diastolic blood pressure in the control scenario; and
DBP2 = mean diastolic blood pressure in the no-control scenario.
Information from Shurtleff (1974) can be used to estimate the probability of premature death in
men older than 54 years old. This study has a 2 year follow up period, so a 2 year probability is
estimated. For men aged 55 to 64 years old, mortality can be predicted by the following equation:
rt »-S
. i 4.USW - O.OWW «OW,
* where: ;
9 . APrCMORTJ5_64)= the change in 2 year probability of death in men aged 55-64;
w DBF, = mean diastolic blood pressure in the control scenario; and
11 DBPj = mean diastolic blood pressure in the no-control scenario.
12 For men aged 65 to 74 years old, premature mortality can be predicted by the following
n equation:
3,05723 - 6MSt7*DBFt . 3J1572J -
14 where:
15 APr(MORT$5.74) = the change in 2 year probability of death in men aged 55-64;
i6 DBF, = mean diastolic blood pressure in the control scenario; and
n DBP2 = mean diastolic blood pressure in the no-control scenario.
is Valuing reductions in premature mortality
19 As discussed above, all premature mortality is valued at $4.8 million in this report Because this
20 valuation is based on the willingness to pay to avoid death itself, and the CHD valuation is based on the
21. willingness to pay to avoid the pain and suffering of a heart attack, these two endpoints are additive as
22 well.
Health Benefits to Women
24 Available evidence suggests the possibility of health benefits from reducing women's exposure
25 to lead. Recent expanded analysis of data from the second National Health and Nutrition Examination
320
-------
Appendix G: Lead Benefits Analysis
i Survey226 (NHANESII) by Schwartz (1990) indicates a significant association between blood pressure
2 and blood lead in women. Another study, by Rabinowitz et al. (1987), found a small but demonstrable
3 association between maternal blood lead and pregnancy hypertension and blood pressure at time of
4 delivery.
5 A quantitative estimate of female blood pressure related to PbB can be estimated from a recent
6 review of ten published studies.227 All of the reviewed studies included data for men, and some included
7 data for women. A concordance procedure was used to combine data from each study to predict the
s decrease in diastolic BP associated with a decrease .from 10 ug/dL to 5 ug/dL PbB. The results suggest
9 that this decrease in PbB would decrease diastolic BP by 1 mm Hg in adult males, and about 0.6 mm Hg
in adult females. Thus, lead's effect on BP in women is estimated to be 60 percent of the effect seen in
men. Applying this value to Equation '2-2 for men, the resulting equation is:
10
(54)
12
'13
14
IS
16
17
IS
19
20
21
22
where:
ADBPW
PbB,
PbBj
the change in women's diastolic blood pressure expected from a change
in PbB;
blood lead level in the control scenario; and
blood lead level in the no-control scenario.
Although women are at risk of having lead-induced hypertension, there is not a dose-response
function for hypertension in women available at this time. Omitting the hypertension benefits for women
creates an underestimate of the total benefits, but the impact on the total benefits estimation will likely be
small. Lead raises blood pressure in women less than in men, so the probability of causing hypertension
is likely to be less than in men, and the total value of hypertension in men is a small portion of the
overall estimated benefits.
26
27
28
29
Changes in Coronary Heart Disease
Quantifying the relationship between blood pressure and coronary heart disease
Elevated blood pressure in women results in the same effects as for men (the occurrence of heart
attack, two types of stroke, and premature death). However, the general relationships between BP and
these health effects are not identical to the dose-response functions estimated for men. All relationships
presented here have been estimated for women aged 45 to 74 years old using information from Shurtleff
(1974). First-time heart attacks in women can be estimated from the following equation:
226 The Second National Health and Nutrition Examination Survey (NHANES II) was conducted by the U.S. Department of Health and
Human Services from 1976 to 1980 and provides researchers with a comprehensive set of nutritional, demographic and health data for the U.S.
population.
227 Schwartz, 1992b.
321
-------
Appendix G: Lead Benefits Analysis
r^
where:
= change in 2 year probability of occurrence of CHD event for women aged 45-
4. . 74;
s DBF, = mean diastolic blood pressure in the control scenario; and
6 DBP2 = mean diastolic blood pressure in the no-control scenario.
7 Valuing reductions in CHD events
s Values of reducing CHD events for women are assumed to be equal those calculated for men
9 (above): $587,500 per CHD event.
10 Changes in Atherothrombotic Brain Infarctions and Initial Cerebrovascular Accidents
n Quantifying the relationship between blood pressure and first-time stroke
12 For initial atherothrombotic brain infarctions in women, the logistic equation is:
rV womm) ~ ~ 1&.6716 - 0.0544 »DW, . ^ 10.6716 - O.OJ44«OWj (56)
13 where:
14 APr^B^a,) = change in 2 year probability of brain infarction in women aged 45-74;
is DBF, = mean diastolic blood pressure in the control scenario; and
16 DBP2 = mean diastolic blood pressure in the no-control scenario.
n The relationship between BP and initial cerebrovascular accidents can be predicted by the following
is logistic equation:
) ~
-worn***
?J)77J7 - V.Mlt7*DBPt . ».«797 - 0.0*297 *D8Ft
19 where:
20 APrCCA^^eJ = change in 2 year probability of cerebrovascular accident in women aged 45-74;
21 DBF, = mean diastolic blood pressure in the control scenario; and
22 DBP2 = mean diastolic blood pressure in the no-control scenario.
322
-------
Appendix G: Lead Benefits Analysis
i Valuing reductions in strokes
2 Values of reducing stroke events for women are assumed to be equal those calculated for men
3 (above): $587,500 per BI and CA event.
4 Changes in Premature Mortality
s Quantifying the relationship between blood pressure and premature mortality
6 The risk of premature mortality in women can be estimated by the following equation:
15
5.40374 - O.OtSll'&Bf. t J.40374 - fr.01SH*£MU>
e ' I * e . l
7 where:
« APrCMORT^a,) = the change in 2 year probability of death for women aged 45-74;
9 DBF, = mean diastolic blood pressure in the control scenario; and
10 DBP2 = mean diastolic blood pressure in the no-control scenario.
11 Valuing reductions in premature mortality
12 Values of reducing premature mortality for women are assumed to be equal those calculated for
u all premature mortality (above): $4.8 million per premature mortality.228
Industrial Processes and Boilers and Electric
Utilities
16 This section describes the methods and data sources used to estimate changes in blood lead
n levels due to changes in lead emissions from industrial processes and boilers between 1970 and 1990 and
is. from electric utilities between 1975 and 1990. It also presents estimates of the resulting changes in
19 health effects and the monetary value of these effects.
IEc 1992, IEc 1993.
323
-------
Appendix G: Lead Benefits Analysis
12
i Methods Used to Determine Changes In Lead
2 Emissions from industrial Processes from 197O to
199O
4 This analysis used several sources to determine the changes in facility-specific emissions of lead
5 from industrial processes. To summarize, the analysis extracted 1990 facility-specific lead emissions
« data from the Toxics Release Inventory (TRI), which provides recent emissions data for over 20,000 U.S.
7 manufacturing facilities. This study then adjusted these data by the relative changes in lead emissions
« between 1970 and 1990; these relative changes were derived from several data sources described below.
9 This method yielded facility-specific emissions for five year intervals between 1970 and 1990 for both
10 the controlled and uncontrolled scenarios. The five-year values were interpolated to derive annual
;/ changes for each year between 1970 and 1990. Specific details on this approach are given below.
TRI Data
13 The Toxics Release Inventory (TRI) is mandated by the Superfund Amendment Reauthorization
H Act (SARA) Title III Section 313 and requires that U.S. manufacturing facilities with more than 10
is employees file annual reports documenting multimedia environmental releases and off-site transfers for
i6 over 300 chemicals. Facilities report both stack and fugitive releases to air.' Reported releases are
17 generally estimates rather than precise quantifications. Emissions data can be presented as numerical
is point estimates, or, if releases are below 1,000 pounds, as an estimated range of emissions.
19 TRI data are available for the year 1988-1990. From the TRI data base, this analysis extracted
20 data from the reporting year 1990 for all facilities reporting emissions of lead to air, as either stack or
21 fugitive emissions. Data were reported as annual emissions (in pounds per year). Where emissions are
22 reported as a range, this analysis used the upper bound of the range to represent the emissions.229 TRI
23 facilities also report their location by latitude and longitude. In order to later match facilities emitting
24 lead with Census data on surrounding exposed populations, this analysis uses the latitudes and longitudes
25 of lead-emitting facilities.
26 Derivation of Industrial Process Emissions Differentials 1970-1990
27 The TRI database is the Agency's single best source of consistently reported release data;
2s however, the database does not include information for most of the years modeled in this analysis.
29 Furthermore, this analysis required estimates of hypothetical emissions in the absence of the CAA.
30 Therefore, estimates were created for the emissions of lead from industrial sources under the CAA, and,
31 in the absence of the CAA, for the years 1970,1975,1980,1985, and 1990. The percent cJiaogss, or
32 differentials, reflected by these estimates were then applied to the 1990 TRI data to obtain facility-level
33 release estimates for the years of interest for the controlled and uncontrolled scenarios.
u' Ranges are infrequently reported and are either reported as 0-500 Ibs. or 500-1000 Ibs. The infrequency of the incidence of a facility
reporting a range and the relatively small quantities of lead released by those facilities means any overestimation of benefits that results from
using the upper limit of the range is extremely minor.
324
-------
Appendix G: Lead Benefits Analysis
i The method for creating these differentials captured the two potential causes of the differences
2 between emissions from industrial sources regulated by the CAA and those emissions in the absence of
j the CAA. The first cause of the difference in emissions is a change in overall industrial output, resulting
4 from the macroeconomic impact of the CAA. The second element is a change in emissions per Unit of
5 output, which results from the adoption of cleaner processes and the application of emissions control
d technology mandated by the CAA. The methods used to project the effects of these two causes,
7 described below, were designed to be as consistent as possible with other emissions projection methods
* for other segments of the CAA retrospective analysis.
9 Data sources
w Data for the differentials estimates were taken from the following sources:
/; • the Jorgenson/Wilcoxen (J/W) model projections, conducted as part of the Section 812
12 analysis. This data source addresses the first cause of changes in emissions: the
13 macroeconomic changes that resulted from the implementation of the 1970 CAA. The
14 J/W model calculated the change in economic output for each of thirty-five industrial
is sectors, roughly analogous to 2-digit SIC codes, that resulted from the CAA's
16 implementation. The specific output used from the J/W model in this analysis was the
n percentage change in economic output for the various industrial sectors, rather than any
is absolute measure of economic activity.
19 • the 1991 OAQPS Trends database. This data base is an emissions projection system that
20 was used to produce the report, "The National Air Pollutant Emission Estimates, 1940-
21 1990." It contains information on economic activity, national level emissions and
22 emission controls, by industrial process, from 1970 through 1990. Three different
23 elements were extracted from the Trends database: the emissions of lead per unit
24 economic output for various industrial processes for the years 1970-1990; annual
25 economic output data for these industrial processes; and the emission calculation
26 formula.
27 • the National Energy Accounts (NEA), compiled by the Bureau of Economic Analysis.
28 This database records the historical levels of industrial energy consumption,
29 disaggregated by fuel type at the approximately three-digit SIC code level.
30 The manner in which these data were combined to derive lead emissions estimates is described below.
31 Estimates of industrial process emissions in the controlled scenario
32 Emissions data for industrial processes were estimated for the years 1970,1975,1980,1985, and
33 1990. For each of these years, this analysis extracted an emission factor and a control efficiency for each
34 lead-emitting industrial process in the Trends database. Emissions factors are expressed as amount of
35 lead emitted per unit of economic activity, and control efficiencies are reported as the percent that
36 emissions are reduced through the application of pollution control technology to the process. The year-
37 specific emission factors and control efficiencies were multiplied by the economic activity data for that
325
-------
Appendix G: Lead Benefits Analysis
i year, for that process, as reported in the Trends database, using the following equation found in the
2 Trends report:
Emissions = (Economic Activity) * (Emission Factor) * (1 - Control Efficiency) (59)
3 . This calculation yielded the estimated controlled scenario emissions, by industrial process.
4 Industrial processes were then assigned to an NEA code. Finally, all processes assigned to a given NBA
j code were summed to give a total emissions estimate for that NEA code.
6 Estimates of industrial process emissions in the uncontrolled scenario
7 The results from the J/W model were used to estimate process emissions in the uncontrolled
« scenario. As stated above, the J/W model provides percent changes in economic outputs by industrial
9 sector. To use these values, lead-emitting industrial processes (in the Trends database) were assigned to
10 a J/W sector. The percent change for that sector from the J/W model was then used to adjust the
/; economic activity data for that process from the Trends database. These adjusted economic output
12 figures were used together with 1970 emission factors and control efficiencies to derive the estimated
n lead emissions for each industrial process in the uncontrolled scenario. The 1970 emission factors and
14 control efficiencies were used for all years in the analysis (1970,1975,1980,1985 and 1990) in the
is uncontrolled scenario; this assumes that emissions per unit economic output and control efficiencies
16 would have been constant over time in the absence of the CAA. This is the same approach that was used
n to project the changes in emissions from industrial processes for other criteria pollutants in other portions
is of the CAA retrospective analysis. The process-level emissions were then aggregated to the NEA-code
19 level, as in the controlled scenario.
20 Matching TRI Data to Industrial Process Emissions Differentials
21 The methods described in the preceding section yielded emissions estimates from industrial
22 processes in the controlled and uncontrolled scenarios, by NEA code. We used these estimates to derive
23 percent changes in emissions between controlled and uncontrolled scenarios, by NEA code, for
24 application to the TRI emissions data. However,, since TRI data are reported by SIC code, we first
25 "mapped" NEA codes to the appropriate SIC codes, and used the percent change for each NEA code to
26 represent the percent change for all SIC codes covered by that NEA code.
27 It should be noted that the Trends data base covers only the most important sources of lead in air,
28 not all sources; as a result, not all SIC codes reporting lead emissions in TRI correspond to an NEA code
29 for which emission differentials have been estimated. However, we assume that the TRI emissions
30 sources that have a match are the most important sources of lead air emissions. In fact, although only 48
31 out of 519 legitimate SIC codes reporting lead emissions in TRI have matching differentials, these SIC
32 codes account for over 69 percent of the lead emissions reported in TRI. The remaining 31 percent of
33 the emissions are distributed relatively evenly among the remaining 471 SIC codes, each of which
34 contributes a small amount to total emissions.
35 For the 31 percent of the emissions without differentials, this analysis has no information
36 regarding the change in the lead emissions over time or between the controlled and uncontrolled
326
-------
Appendix G: Lead Benefits Analysis
i scenarios; therefore, we are unable to predict benefits attributable to the CAA for these emission sources.
2 Although excluding these sources may lead us to underestimate total benefits, we believe these sources
3 are unlikely to contribute significantly to the difference between controlled and uncontrolled scenarios.
4 The Trends data focuses on the point sources of lead emissions of greatest concern to the Project Team
5 and of greatest regulatory activity. If a process within ah SIC code does not appear in the Trends, it is
e • unlikely to have had specific CAA controls instituted over the past 20 years. A lack of control
7 efficiencies for smaller sources prevents them from being included.
« It should also be noted that the total industrial process emissions of lead estimated in the 1990
9 Trends report actually exceeds the reported lead emissions in TRI, despite the fact that TRI covers more
10 SIC codes. This is probably attributable in part to the fact that TRI covers only a subset of the facilities
// contributing to economic output in an SIC code. TRI reporting rules only require facilities with greater
12 than 10 employees and who use certain amounts of lead in their processes to submit information to TRI,
is while the Trends report attempted to estimate emissions from all sources contributing to the economic
14 output for the industrial sector, regardless of size. However, the components of the Trends data base
is used in this analysis (i.e., emissions factors, economic output data) represent typical conditions at
16 average facilities; they do not allow for the representation of the distribution of emissions across
i? particular facilities. In contrast, a major strength of the TRI is its match of emissions data with
is geographical information. Because the distribution of emissions geographically determines the size of
' 19 . exposed populations, this analysis used the TRI data, rather than Trends data, to characterize lead release
20 quantities, and used the Trends figures only to characterize relative emissions and changes over time,
21 rather than to estimate total quantities.
22 Because the Trends data are intended only as an estimate of emissions using typical conditions at
23 average facilities, and do not capture the differences in facility-level emissions, the data do not provide
24 sufficient information to make specific quantitative adjustments to the TRI-based benefits estimates to
25 account for the overall higher emissions estimates in Trends. However, since Trends does generally
26 suggest that there are many more sources than are accounted for by TRI, it is possible that our benefits
27 calculations may be underestimated.
is Some additional assumptions were necessary when matching the TRI lead release data and the
29 differentials from the Trends data. Ideally, we would know whether the facilities present at a given
30 location, as reported in the 1990 TRI, were present and operating in earlier years; whether facilities
31 operating in 1970 have ceased to operate; and whether new facilities would have been constructed in the
32 uncontrolled situation. Unfortunately, data do not exist in an accessible form at this level of detail for
33 the years 1970 through 1990. Therefore, for the purposes of this exercise, we have assumed that the
34 locations and numbers of the 1990 sources are the same as they were in 1970.
35 Methods Used to Determine Changes In Lead
36 Emissions from Industrial Boilers from 197O to 199O
37 Several sources were used to determine the change in lead emissions from industrial boilers.
ss TRI locational data, Trends database national fuel consumption levels and emissions factors, and NBA
39 and SIC codes were used to derive the emissions for the controlled and uncontrolled scenarios.
327
-------
Appendix G: Lead Benefits Analysis
i TRIData
2 The TRI does not appear generally to contain combustion emissions data. In general, the
3 emissions data is from process sources. We reached this conclusion based on two pieces of information:
4 (1) TRI reporting requirements: TRI has three reporting requirements: (a) the facility must fall
5. in SIC codes 20-39; (b) the facility must employ more than 10 persons; and (c) the" facility must
6 manufacture or process >25,000 Ibs. of a TRI chemical, or otherwise use > 10,000 Ibs. Firms must
7 submit reports only for the chemical that exceeds the thresholds given in item (c), but they must report
s all releases of that chemical, including releases from uses that would not qualify alone. If the TRI
9 chemical is part of a blended substance and the quantity of the TRI chemical in the blend exceeds the
10 threshold, it must be reported. In our case, if the amount of lead in fuel were to exceed the 10,000 Ibs.
n threshold, then the firm would be required to report all emissions of lead from combustion of fuel. There
n is an exemption, however, for ingredients present in small proportions. If the amount of lead in the oil
n were less than 0.1 percent (1000 ppm), then the firm is not required to report the emissions,
14 The conclusion from the above information is that most firms burning used oil are probably not
is reporting lead combustion emissions to TRI because these releases fall outside the TRI reporting
16 requirements. The concentration at which lead is typically found is used oil (100 ppm) (NRDC, 1991) is
n much less than the minimum concentration required for reporting (1000 ppm).
.is (2) Use data from the TRI data base: The hypothesis that firms do not report lead combustion
19 was confirmed by an analysis of the data submitted by the firms reporting lead use to TRI. On the TRI
20 submission forms, firms must indicate how the chemical is used. Our analysis of category codes
21 submitted by firms reporting lead emissions showed the following four use category reports: as a
22 formulation component; as a reactant; as an article component; and repackaging only. None of these
23 category codes suggest that the source of the reported lead release is combustion. Therefore, we may
24 conclude that all of the lead emissions reported in TRI are process emissions.
25 Based on these analyses, this analysis could not use the TRI release data to evaluate releases of
26 lead from industrial combustion. However, this study still used the TRI geographical information to
27 locate industrial facilities by longitude and latitude in order to combine combustion data with population
28 information. For combustion emissions, the calculations included all TRI reporting facilities, not just
29 those who report lead emissions. The assignment of combustion emissions to these facilities is described
30 below.
31
Derivation of Industrial Combustion Emissions 1970-1990
32 As with industrial process emissions, estimates were created for the emissions of lead from
33 industrial combustion under the CAA, and in the absence of the CAA, for the years 1970,1975, 1980,
34 1985, and 1990. These emissions estimates were used, in combination with the TRI data base
35 geographic information, to obtain facility-level release estimates for the years of interest for the
36 controlled and uncontrolled scenarios. The method for deriving these emissions estimates included both
37 the macroeconomic impact of the CAA and the change in emissions per unit of output that resulted from
3s specific pollution control mandates of the CAA. The same data sources were used to derive combustion
39 differentials as were used to derive process differentials. The particular data elements and the methods
328
-------
Appendix G: Lead Benefits Analysis
i by which these data were combined to derive lead emissions estimates from industrial combustion are
2 described below.
3 Estimates of combustion emissions in the controlled scenario
4 The Trends database contains a national aggregate industrial fuel consumption estimate, by fuel
s type (coal, natural gas, oil). For each fuel type, the fuel consumption estimate was disaggregated by the
6 share of that fuel used by each NBA.industrial category, using the NEA data base. It should be noted that
7 the NEA includes data only for the years 1970 through 1985. For 1990, the 1985 figures were used to
s disaggregate the national-level consumption figure into NEA industrial categories.
9 The Trends database also contains emissions factors for industrial fuel use, by fuel type, as well
w as control efficiencies. The lead emissions from industrial combustion for each NEA category was
11 derived by multiplying the fuel-specific combustion estimate for each NEA category by the emission
n factor and control efficiency for mat fuel type. The result was emissions of lead by NEA code and by
13 fuel type. Emissions from all fuel types were then summed by NEA code. By using the NEA data to
H disaggregate the industrial fuel consumption figures, the analysis assumes that the industrial combustion
is emissions are the same among all industries covered by a given NEA code, an assumption which may
16 bias the analysis.
17
Estimates of combustion emissions in the uncontrolled scenario
is As in the controlled scenario, the national aggregate industrial fuel consumption estimate, by
19 fuel type (coal, natural gas, oil), was disaggregated by the share of that fuel used by each NEA industrial
20 category. The fuel use was then adjusted by one of two factors: (1) seven of the NEA codes were
21 specifically modeled by the Industrial Combustion Emissions (ICE) model — for these sectors, the ICE
22 modeled percent changes were used instead of J/W percent changes; or (2) the remaining NEA codes
23 were matched to J/W sectors — the J/W percent changes were then applied to those matched NEA codes.
24 These fuel use estimates were then combined with the 1970 emission factors and control efficiencies for
2s industrial combustion by fuel type from the Trends database to obtain combustion-related lead emissions
26 from industrial boilers hi the uncontrolled scenario, by NEA code.
27 The process-specific data hi the Trends database, and the energy use data in the NEA, are much
28 more disaggregated than the J/W sectoral projections. For the purpose of the analysis, it was assumed
29 that all of the specific industrial processes in the Trends database and industrial categories in the NEA
so dataset assigned to a given J/W sector changed at the same rate as the entire J/W sector. For example, if
31 the economic activity hi the J/W Sector 20, "Primary Metals," changed by one percent between the
32 controlled and uncontrolled scenarios, then the analysis assumed that economic activity in each industrial
33 process assigned to the Primary Metals sector also increased by one percent. This approach assumes
34 that the economic activity of specific industries within a sector are equally affected by the imposition of
35 the CAA. This assumption is consistent with the projection of the change in emissions from industrial
36 processes for the other criteria air pollutants, which were calculated using a similar process.
329
-------
Appendix G: Lead Benefits Analysis
i Matching TRI Data to Industrial Combustion Emissions Data
2 Because of the structure of the TRI reporting requirements, it does not appear that TRI generally
3 contains releases from combustion sources. Although TRI may incidentally contain lead combustion
4 emissions, TRI would contain data on such releases only if the reporting facility also used more than
5 10,000 pounds of lead per year for manufacturing or processing. As a result, the combustion releases,
6 estimated using the methods described above, do not have corresponding data in the TRI data base.
/ Therefore, we devised a different method for estimating benefits from changes in combustion releases.
s The first step in the method was to divide the estimates of total releases of lead from industrial
9 combustion, by NEA code, by an estimate of the number of facilities in each NBA code. The number of
10 facilities in each NEA category was estimated using the 1987 Census of Manufactures. This Census,
;/ conducted by the U.S. Department of Commerce, tallies the number of facilities by 4 digit SIC code;
12 these SIC codes were matched to the NEA codes.
u Dividing total lead emissions emitted by number of facilities yielded the average yearly lead
14 emissions from industrial combustion for each SIC code. We men assigned this average value to all
is reporting TRI facilities in the SIC code. The consequence of this approach is that the modeling of
16 combustion from industrial facilities includes substantially more sources than the modeling of industrial
i? process emissions; combustion emissions are assigned to essentially all facilities reporting to TRI, while
is the process emissions are only evaluated for facilities actually reporting lead air emissions from
19 processes.
20 One unavoidable drawback to this approach is that it cannot capture differences in release
21 quantities among facilities within an SIC code. Furthermore, this approach does not capture all
22 combustion emissions because we assign average emissions only to facilities that report to TRI. TRI
23 facilities account for between two percent and 50 percent of all facilities listed in the Census of
24 Manufacturers, depending on the SIC code. Because of the inability to place the remaining facilities
25 geographically, this analysis excludes the consideration of emissions from non-TRI facilities.
26 Methods Used to Determine Changes In Lead!
27 Emissions from Electric utilities from 1975 to 199O
28 The estimation of lead emissions from electric utilities required data from three different
29 sources. Energy use data for the control and no-control cases was obtained from the national coal use
30 estimates prepared for the Section 812 analysis by ICF Incorporated. The OAQPS Trends Database
31 provided emissions factors and control efficiencies. Individual plant latitudes, longitudes, and stack
12 information were collected from the EPA Interim Emissions Inventory. This analysis combines these
33 three pieces of data and creates yearly lead emissions at the plant level for coal burning electric utilities
34 in the control and no-control scenarios. This section describes the sources and the methods used to
35 create the final dataset.
330
-------
Appendix G: Lead Benefits Analysis
i Coal-Use Data
2 The energy use data obtained provides plant level energy consumption data for 822 electric
3 utilities. The dataset was broken into four distinct sets for each of the years 1975, 1980, 1985, and 1990.
4 Each set of data provided the state where the plants are located, the plant names, and the amount of coal
s consumed, for both the control and no-control cases. The four data sets were combined into one
6 comprehensive set by matching the plants'names and states.
7 The EPA Interim Emissions Inventory
« The EPA Office of Air Quality Planning and Standards Technical Support Division provided the
9 1991 EPA Interim Emissions Inventory. The Interim Inventory contains data for all electric utility and
10 industrial plants in the United States including latitude, longitude, stack height, stack diameter, stack
11 velocity, and stack temperature. The additional stack parameter data allowed the use of plant-specific
n parameters in the air modeling for electric utilities rather than average parameters for all facilities as was
13 done for industrial emissions.
14 Matching the Coal-Use Data to the Interim Emissions Inventory
is The combination of the Interim Emissions Inventory and the coal-use data required two steps.
16 First, the Interim Emissions Inventory needed to be pared down to include only electric utility data and to
i? narrow down the information provided for each utility. Secondly, the two datasets needed to be
is combined. One difficulty in the combination of the two sets was the lack of a common data field that
19 would allow a quick and complete matching process between the two.
20 Electric utility plants were identified in the Interim Emissions Inventory by SIC code (code
21 4911). The associated stack information file, which lists every stack on every plant, was reduced to
22 include only the largest stack for each plant. This provides a reasonable estimate of the stack height at
23 which most emissions occur. The air modeling assumes that each electric utility releases its emissions
24 from the largest stack that exists at that plant.
25 Next, the procedure matched the abridged Interim Emissions Inventory file with the coal use
26 data. Due to the lack of a common data field between the two sets, this process required several phases.
27 Both data sets had name fields, but these fields utilized different naming conventions for the plants.
2s Therefore the name fields were matched directly, with individual words in the names, and then with
29 abridged words from the names. Abridged word matches were double checked by ensuring that the
30 names were indeed similar and by verifying that the state fields matched. Finally some matches were
31 made by hand.
32 Only 27 unmatched plants with positive coal use remained. There were 493 matched plants with
33 positive coal usage and these were included in the final dataset.230 To eliminate under-counting of
34 emissions, the emissions from the 27 unmatched plants were allocated to matched plants within the states
230 Plants with zero coal usage were not immediately excluded from the analysis due to the possibility of analyzing lead emissions from
oil combustion at these plants. However, OAQPS has suggested that oil combustion comprises under two percent of the total lead emitted from
electric utilities. For this reason, the electric utility analysis focused entirely on coal.
331
-------
Appendix G: Lead Benefits Analysis
i where the unmatched plants were located. Allocations were weighted according to the emission level for
2 each matched plant within that state in the year in which the allocation was being made.
j Emissions Factors and Control Efficiencies
4 At this stage, the electric utilities data set contained coal consumption by plant by year in the
5 control and no-control cases as well as air modeling parameters. Using emissions factors for lead and
6 control efficiencies for electric utilities, estimates of lead emissions per plant per year could now be
7 calculated. As in the industrial source analysis, the emissions factors and control efficiencies come from
s the 1991 OAQPS Trends database.
9 Control efficiencies are available for coal-fired electric utilities in each year between 1975 and
10 1990. As in the industrial source analysis, it is assumed that pollution control on coal-burning power
/; plants without the CAA would be the same as the pollution control level in 1970. Therefore, the control
12 efficiency from 1970 is used in the no-control analysis.
13 The emissions factor obtained from the Trends database is expressed in terms of lead emitted per
14 ton of coal burned (6050 grams per 1,000 tons).231 The combined dataset, though, contains quantity of
is coal burned per plant per year in energy units (trillions of Btus). To reconcile this difference, a
16 conversion factor was obtained from a 1992 DOE report titled Cost and Quality of Fuels for Electric
17 Utility Plants 1991. The conversion factor used (20.93 million Btus per ton of coal) is the average Btu
is per pound of coal burned for all domestic electric utility plants in 1990. Data for a small subset of other
19 years were also provided in the DOE report, but they did not differ significantly from the 1990 number.
20 Therefore, the 1990 conversion factor (637.3 pounds of lead per trillion BTU) is assumed valid over the
21 entire study period. The final equation for lead emissions looks quite similar to the equation used in the
22 industrial source analysis.232 The only change is that "Economic Activity" has been replaced by "Coal
23 Consumed" for this particular analysis:
24
Emissions = (Coal Consumed) * (Emission Factory x (1 - Control Efficiency) (60)
21 This equation produces estimates of the emissions per plant per year in both the control and the no-
26 control scenarios. •
27 Use of Air Dispersion Modeling to Estimate Ambient Air
28 Lead Levels
29 To link estimates of lead emissions to blood lead levels of populations living in the vicinity of
30 facility, the lead benefits model first uses air dispersion modeling to estimate air lead concentrations
31 surrounding facilities that emit lead into the ah-. The air concentrations are then linked to blood lead
32 levels.
211 The actual figure cited is 12.1 metric pounds per 1000 tons. A metric pound is one two-thousanth of a metric ton.
232 EPA 1991.
332
-------
Appendix G: Lead Benefits Analysis
This analysis uses the Industrial Source Complex Long Term (ISCLT) air dispersion model, a
steady-state Gaussian plume model, to estimate long-term lead concentrations downwind of a source.
The concentration is modeled as a function of site parameters (stack height, stack velocity).233 The
general form of the concentration equation from a point source at a distance r greater than one meter
away is shown in Equation 61:234
2K
6 where,
7 C,;, = concentration at distance r (ug/m3),
« Q = pollutant emission rate (g/sec),
9 '. f = frequency of occurrence of wind speed and direction,
10 9 = sector width (radians),
;/ S smoothing function used to smooth discontinuities at sector boundaries,
12 u = mean wind speed (m/sec),
n oz = standard deviation of vertical concentration distribution (m),
H V = vertical term (m),
is • K = scaling coefficient for unit agreement.
16 For each facility modeled in the lead benefits model, a 21 by 21 kilometer grid around the source is
i? specified. The model stores data in 1 km by 1 km cells and calculates the air lead concentrations for
is each of the 441 cells surrounding a given facility. Fugitive sources are modeled similarly, the only
19 difference being a modified form of Equation 333.
20 For facility-specific weather data, the model used Stability Array (STAR) data. The STAR data
21 contains information on typical wind speed and direction for thousands of weather stations in the U.S.
22 For each facility, the model accesses the STAR data for the weather station nearest the source. Standard
13 default assumptions are used for the other parameters because facility-specific data for these parameters
24 are not available (except for utilities). For these parameters, we instead use standard default parameters.
25 Table 130 lists default parameters for the ISCLT, and summarizes sources for other parameters.
26 Industrial process emissions were modeled as either point or fugitive sources, depending on how
27 they were reported in TRI. All industrial combustion emissions were modeled as "fugitive" emissions.
28 This is a more appropriate model scenario for boiler emissions than a 10 meter stack scenario. All
29 electric utility sources were modeled as point sources.
10 The model tracks all lead emissions to a given cell. That is, if the plumes of two or more sources
31 overlap in a given cell, the air concentration in the given cell is determined from the sum of all of the
32 contributing sources.
231 Ideally, reported stack and fugitive air releases would be modeled using site-specific data (such as source area or stack height).
However, since TRI does not contain such facility-specific information, default values are used to model TRI facilities.
234 This equation is from EPA < 1992). The equation is for a specific wind speed, direction, and category (/>*). Each facility has several
combinations of these that must be added to arrive at a total concentration at that point The equation for area sources is similar.
333 '
-------
Appendix G: Lead Benefits Analysis
Table 130. Air Modeling Parameters,
Parameter
Stack height
Exit Velocity
Stack diameter
£xit gBft tcmpefBiurft
Area Mttrce size
Anm 9U4urce height
Lead emistion rate
Frequency of wind speed and
direction
Sector width
Wind speed
Smoothing function
Vertical term
Industrial
Source
Value.
10m
6.01 ra/j
1m
2»*K
Wm1
3m
•t» • >»l ami Itjl
stt&^pecinc
site-specific
22;S* ,
site-specific
calculated
calculated
Electric
Utility
Vatae
lite-specific
Industrial - UJS, BPA (1S92) Utilities - U.S.' EPA
(!»!>
faduonal-U^. EPA (1992) UtQiiia-U.S.EPA
««!}
ta*jftriat-y.S.B3>A(1992> Utilities -U.S. EPA
(«»«>
U.S. EPA (1992)
U.S. EPA (1952)
Industrial— 1H1$ qhs/yr)
UtiMei-SAl&OAQPS(Ibs/yi)
STAR
-------
Appendix G: Lead Benefits Analysis
i exposure results in higher air lead to blood lead slopes. In both cases, the slope relationship is expressed
2 as the change hi blood lead (ug/dL) per change in air concentration (ug/m3).
3 In performing this analysis, the choice was made between the use of air lead:blood lead
4 relationships that account for inhalation exposure ("direct" slopes) and those that account for exposure
; through inhalation and for exposure to lead deposited from air onto soil and dust ("indirect" slopes). The
6 choice of which slopes to use considered both the effects on the estimate of benefits over time (from
7 1970 to 1990) as well as the estimate of the difference hi benefits between the controlled and
t uncontrolled scenarios. The indirect slope is more comprehensive in its coverage of the types of
9 exposures that will result from air releases, and thus captures more of the health effects predicted to
10 occur from lead exposures, especially to children. For this reason, indirect slopes are preferred to direct
;; slopes, especially when comparing the controlled to uncontrolled scenarios: using only the direct slope
12 would underestimate the benefits of avoiding deposition that controls confer. However, indirect slopes
u may capture effects from exposure to soil and dust lead deposited from both current air releases and
14 historic air releases. Since lead's dissipation from soil is slow relative to its removal from ah*, the
15 reservoir of lead in soil and dust is unlikely to change at the same rate as the reductions in air lead
16 concentrations. Therefore, using indirect slopes to represent a change in blood lead over time due to
i? reduced air emissions may overestimate the change in blood lead, and thus overestimate the benefits of
is reductions over time, to the extent that the indirect slope captures exposure to the total reservoir of soil
19 and dust lead, rather than only recently deposited lead.
20 Given that the focus of this analysis is the difference between the controlled and uncontrolled
21 scenarios, it is important to capture both the benefits from reduced lead deposition that result from the
22 CAA, and the direct benefits from reduced ah* concentrations. Therefore, this analysis modeled changes
23 in blood lead levels using indirect slopes. It should be kept in mind that this choice may overestimate
24 blood lead changes over time, for both the controlled and uncontrolled scenarios.
25
26 The relationship between concentrations of lead in ambient ah* and blood lead concentrations has
27 been evaluated by a variety of methods. These include experimental studies of adult volunteers, as well
28 as epidemiological studies of different populations of children and adults. The discussion below
29 describes the slopes used in this analysis for children and adults, and for individuals with blood lead
30 values greater than 30 ug/dl.
'31 Children
32 U.S. EPA (1986) reports that slopes which include both direct (inhalation) and indirect (via soil,
33 dust, etc.) air lead contributions vary widely, but typically range from three to five ug/dl increment in
34 children's blood lead per ug/m3 increment in air lead concentration (roughly double the slope due to
35 inhaled air lead alone). Since hand dust levels can play a significant role hi blood lead levels (U.S. EPA,
36 1986), this higher slope may be due to mouthing behavior of children that brings them into contact with
37 dust and soil.
is Specific values for estimating contribution of air lead to blood lead, including indirect pathways,
39 are cited in U.S. EPA (1986); slope values (ranging from -2.63 to 31.2) and data sources for these values
« are presented in Table 11-36 of U.S. EPA (1986). The median of these values is 4.0 jig/dl per ug/m3,
41 which matches the midpoint of the range of typical slope values. This analysis used this value to
335
-------
Appendix G: Lead Benefits Analysis
i represent the relationship between air lead concentrations and blood lead concentrations for children
•2 living in the vicinity of point sources of lead emissions.
3 The use of this slope assumes that "indirect exposure" principally measures indirect effects of
4 lead emissions to ah- (through deposition to dust and soil). However, it is possible that these slopes
5 include other exposures not related to air lead. In many cases authors have measured other possible
6 exposures, such as water and food, and have confirmed that the most significant contribution comes from
7 soil and dust lead, which is assumed to result from air deposition of lead. Those studies that measured
a lead in tapwater showed that mean levels were generally low or not significantly related to blood lead.
9 Landrigan et al. (1975) measured lead in pottery and food; lead in pottery was found in only 2.8 percent
10 of homes, and food and water made no more than a negligible contribution to lead uptake. Lead in paint
11 was measured in some studies.235 Landrigan and Baker (1981) measured lead in paint at levels greater
n than one percent in about one fourth to one third of the houses in each area studied. Brunekreef et al'.
u (1981) measured high levels of paint in some houses, but excluded these data points from the analysis.
14 Despite the possibility of confounding factors, this analysis uses the median value determined
is above (4.0 ug/dl per ug/m3) as the appropriate slope for children living within five kilometers of the
16 point source. Five kilometers is chosen as the cut off point because the data from most of the studies
n cited collected the majority of their data points near lead smelters.236 Furthermore, these slopes, although
is measured primarily in the vicinity of smelters, are assumed applicable to all- point sources that emit lead
19 into the air.
20
Adults
21 For adult males and females, the air lead/ blood lead slopes that include indirect effects due to
22 soil and dust differ very little from slopes that include only direct effects. This result is expected since
23 the higher indirect slope values estimated for children are assumed to be as a result of mouthing behavior
24 typical of young children.
25 U.S. EPA (1986) describes several population studies that estimate indirect slopes for men; these
26 slopes range from -0.1 to 3.1 jig/dl per ug/m3.237 Snee (1981) determined a weighted average of these
27 studies and one other study,238 The average slope, weighted by the inverse of each study's variance, is
2« 1.0 ng/dl per ug/m3. However, the Azar study measured the direct relationship between air lead and
29 blood lead. Excluding the Azar study form the weighted average, the average slope is 1.1 ug/m3.
30 Excluding the highest and lowest slopes from this group (from Goldsmith, 1974 and Tsuchiya et al.,
31 1975), both of which had difficulties,239 the resulting slope is 1.4 ng/dl per ug/m3.
235 Landrigan and Baker, 1981; Brunekreef et al., 1981.
236 EPA 1986, Table 11-36.
257 Johnson et al., 1976; Nordman, 197S; Goldsmith, 1974; Tsuchiya et al., 1975; Fugas et al., 1973.
251 Azar etal., 1975.
219 Goldsmith (1974) refrigerated (rather than froze) the blood samples, and did not analyze the samples until 8 or 9 months after they
were taken, and restricted the analysis to one determination for each blood sample. Tsuchiya et al. (1975) measured air lead concentrations
after blood samples were taken; blood was drawn in August and September of 1971, whereas air samples were taken during the 13 month
period from September 1971 to September 1972.
336
-------
Appendix G: Lead Benefits Analysis
10
11
12
13
14
15
16
17
IS
Slopes for females range from 0.6 to 2.4 for general atmospheric conditions.240 Snee determined
an average slope for women of 0.9 ug/dl per ug/m3, weighted by the inverse of the variances of the
studies. Excluding the slope for women from Goldsmith (1974), the resulting slope for women is 1.0
ug/dl per ug/m3.
These values are adjusted by a factor of 1.3 to account for the resorption of lead from bone tissue
(according to Chamberlain, 1983), thus deriving an adjusted slope estimate of 1.8 ug/dl blood lead per
ug/m3 increment in air lead concentration for men and 1.3 for women. These are the slope estimates
used in this report.
Individuals with initial blood lead levels of 30 ug/dl and greater
For individuals with high blood lead levels, the air lead to blood lead uptake slopes have been
shown to be much shallower, as described by U.S. EPA (1986). An appropriate change in blood lead per
change in air lead is 0.5 ug/dl per ug/m3 for individuals that have-initial blood lead levels in the range of
30 to 40 ug/dl. This value is based on cross-sectional and experimental studies.241 For individuals with
initial blood lead levels greater than 40 ug/dl, an appropriate range of slopes is 0.03 to 0.2, as determined
by occupational studies listed in Table 11-37 of U.S. EPA (1986). The median value of these studies is
0.07. These two slopes (0.5 for the population with blood lead levels between 30 and 40 ug/dl and 0.07
for blood lead levels greater than 40 Ug/dl) are used for both children and adults in this analysis. These
relationships are summarized in Table 131.
Table 131. Estimated Indirect Intake Slopes: Increment of Blood Lead Concentration (in ug/dL) per Unit of Air
Lead Concentration {ug/m*).
Adult Malts
Adult Female*
Children
tadirtduata with Wood Je*dlerd» <
*Wdl
1.8
1.3
4.0 -
IndMdud* with fete* kriferdc
3*-»«/dl
0,5
0.5
0.5
IndlrMual* wtth blood lead
l«reb>4««/dl
&.07
0.07
0.07
19
20
21
21
23
Estimates of Initial Blood Lead Concentrations
The benefits model requires an initial distribution of blood lead levels in the exposed populations
to model health benefits of reducing lead air emissions. The model estimates the new distribution of
blood lead levels that would exist after a given change in air concentrations using the slopes described
above. Finally, the model estimates the difference between the two distributions. This analysis begins
240 Tepper and Levin, 1975; Johnson et al., 1976; Nordman, 1975; Goldsmith, 1974; Daines et al., 1972.
241 U.S. EPA, 1986.
337
-------
Appendix G: Lead Benefits Analysis
i with an initial 1970, uncontrolled blood lead distribution from which all subsequent changes are
2 modeled. This approach requires an estimate of the blood lead distributions in the U.S. population in
3 1970. Unfortunately, there are no actual national blood lead distribution estimates for 1970. Although
4 the first NHANES study covered 1970, blood lead data were not collected in this study.242 Nonetheless, a
5 1970 distribution of blood lead was estimated using NHANES II data (from 1976-1980), combined with
6 estimates of typical changes in blood lead levels from 1970-1976 observed in localized screening studies.
7 A major drawback to this approach is the uncertainty in deriving the 1970 estimates. Another
s • drawback to beginning with the 1970 level and modeling changes from that point is the analysis only
9 represents changes in lead exposure from air; reductions from other sources of lead exposure are not
w accounted for. The purpose of this analysis is to identify changes attributable to the CAA mandates;
/; changes from other sources of lead exposure should not be considered. However, due to nonlinear nature
12 of the lead concentration-response functions (see above), the overall exposure context in which the air
a lead exposure reductions take place will influence the estimate of benefits from those reductions.
14 ' Specifically, at higher blood lead levels, the slope of the concentration-response curve is shallower than
is at lower levels. As a result, a given change in the mean blood lead level may result in a smaller change
16 in the health effect if the change occurs from a relatively high starting level. On the contrary, if one
i? accounts for the fact that other sources of lead exposure are reduced at the same time that the given air
is reductions occur, then those air emissions reductions may result in greater changes in health risk.
19 This issue is of concern even though the analysis focuses on the difference between the
20 controlled and uncontrolled states of the world, since the health benefit implications of the emissions
21 differentials between the controlled and uncontrolled scenarios will depend on the point on the blood
22 lead distribution curve at which the differences are considered. That is, a difference between a mean
23 blood lead of 25 ug/dl and one of 20 ug/dl may have different health implications than a difference
24 between 15 ug/dl and 10 ug/dl, even though the absolute value of the difference is the same (5 ug/dl).
2s An alternative method is to "start" with a 1990 blood lead level and to "backcalculate" benefits
26 by representing the differentials as increases over the 1990 levels, rather than decreases from 1970
27 levels. The advantage of this approach is that it accounts for reductions in lead exposure from other
2s sources, as represented by current blood lead levels. Its disadvantage is that it holds other sources
29 constant to (lower) 1990 levels, and thus the modeling may underestimate actual blood lead distributions
30 in earlier years, and thereby overestimate benefits from controls during those years. This analysis
31 presents the results of both approaches.
3: Combination of Air Concentration Estimates with
33 Population Data
34 The modeled air lead concentrations at various distances from the source were combined with
35 population data from the Census Bureau to arrive at an estimate of the number of cases of health effects
36 for each of the years from 1970 to 1990 in both the controlled and uncontrolled scenarios. The primary
37 census information was accessed from the Graphical Exposure Modeling System Database (GEMS), an
141 NCHS, 1993.
338
-------
Appendix G: Lead Benefits Analysis
i EPA mainframe database system. The following data were obtained from GEMS for the years 1970,
2 1980, and 1990: total population for each Block Group/Enumeration District (BG/ED); state and county
3 FIPS codes associated with each BG/ED; latitude and longitude of each BG/ED; and population of males
4 under 5, and females under 5 for each BG/ED. The intervening five year intervals (1975 and 1985) were
s estimated using the Intercensal County Estimates from the Census, which estimate annual populations on
6 a county by county basis. The decennial Census data and the Intercensal County Estimates data sets
7 were related by county FIPS codes; the population in each BG/ED was assumed to grow or shrink at the'
s same rate as .the county population as a whole. .
9 Since the concentration-response data are particular to specific sex and adult age groups,
w additional population data were also required to determine the sizes of affected subpopulations. For
;/ , 1990 age and sex, the U.S. Census, 1992 was used, with age groups tallied as necessary. For 1980 age
12 and sex, the U.S. Census, 1982 was used, with age groups also tallied as necessary. The 1970 age and
n sex breakdowns were obtained through personal communication with the Census Bureau.243 The age and
H sex percentages were interpolated for intervening years.
is Pregnant women are often a subpopulation of interest for lead effects. Although pregnant
16 women themselves may be harmed by exposure to lead, this analysis was concerned with pregnant
n women because of possible effects on their fetuses who will be born and evince effects as young
is children. To estimate the number of exposed fetuses who were born during the years of interest,244 birth
19 rates for 1970, 1980 and 1990 were obtained from the Census Bureau.243 These birth rates were used to
20 interpolate for years between 1970 and 1980, and for the years between 1980 and 1990.
21
Results
22 For both the controlled and uncontrolled scenarios, Table 132 shows estimated lead emissions
23 from industrial processes, industrial combustion and electric utilities. Tables 133 and 134 show the
24 differences in health impacts between the two scenarios (for industrial processes, industrial combustion
25 and electric utilities only) for the "forward-looking" and "backward-looking" analyses.
243 Karl Kucllmer, Abt Associates, and the Bureau of Census, Population, Age and Sex telephone staff, March, 1994.
244 Note that we do not record the number of pregnancies, since the valuation only applies if the child is bom and lives to exhibit the
effect. Neither are we concerned with whether the births are single or multiple births, since each fetus is at risk, whether a pregnant women
carries one or more fetuses.
245 Personal communication, Karl Kuellmer, Abt Associates and the Bureau of Census, Population, Fertility/Births telephone staff.
339
-------
Appendix G: Lead Benefits Analysis
Table 132. Estimated Lead Emissions from Electric Utilities, Industrial Processes, and Industrial Combustion
(in Tons).
BbcteteUtilttiec*
Controfitd Scenario
Ekctric UtilttlM
Uncontrolled Scenario
T t *»-
Uncontrolled Sccni
via
^
Controlled Scenario
Industrial Combution
UncontniuCo Scttuno
1970
7.78*
T,7»
4J2?
4,3»
1975
t^Jl
2409
3^17
. 7,124
44«
4,457
1MB
-------
(1}
-o
8
•s.
I
3
03
S
V5
I
•3
•fi
tf)
*o
Qt 8
SS —'
ts
S
o
'S
"8
1
o
o
c
T3
§
I
i
p\
"S
o\
.s
c«
i
-------
ta
i
"2
.s
i
a
60
v 9
s»
i
I
-------
Appendix G: Lead Benefits Analysis
Reduction In Health Effects Attributable to
Gasoline Lead Reductions
3 Estimating Changes in Amount of Lead in Gasoline
from 197O to 199O
s The relationship between the national mean blood lead level and lead in gasoline is calculated as
6. a function of the amount of lead in gasoline consumed. Thus, to calculate the health benefits from
7 gasoline lead reductions, necessary inputs are estimates of lead in gasoline consumed over the period
« 1970 to 1990 and the amount of lead in gasoline that would have been consumed in the absence of the
9 Clean Air Act. These values are calculated using the quantity of both leaded and unleaded gasoline sold
10 each year and the concentration of lead in leaded and unleaded gasoline for each year in the period of
;/ interest. For each year, the relationship is expressed as:
LEAD = x IFRA€P» *
12 where:
n LEAD = average lead per day in gasoline sold in a given year (metric tons/day),
14 SOLD = total quantity of gasoline sold (million gal/yr),
is FRACPb= fraction of total gasoline sales represented by leaded gasoline (dimensionless),
16 Pbindai ~ lead content of leaded gasoline (g/gal), and
17 Pbmitadtj= lead content of unleaded gasoline (g/gal).
18
19 Gasoline Sales (SOLD1): Data on annual gasoline sales were taken from a report by Argonne
20 National Laboratories (1993) which presented gasoline sales for each state in five year intervals over the
21 period 1970-1990. This analysis used linear interpolation to estimate the gasoline sales for years
22 between the reported years. These data were summed to obtain national sales figures.
23 Fraction of Sales that were Leaded and Gasoline (FRACPfr): For the controlled scenario, this
'24 analysis used information reported by Kolb and Longo (1991) for the fraction of the gasoline sales
is represented by leaded gasoline for the years 1970 through 1988. For 1989 and 1990, data were taken
26 from DOE (1990 and 1991, respectively). For the uncontrolled scenario, all of the gasoline sold was
27 assumed to be leaded for all years.
2s Lead Content of Gasoline (Pb|Ctdcd and Pb^,.,^): Argonne National Laboratory in Argonne,
29 Illinois was the source for the data on the lead content of leaded and unleaded gasoline for the period
30 1974-1990. Argonne compiled these data from historical sales data submitted to EPA, from Clean Air
3i Act regulations on lead content, and from recent MVMA surveys. For 1970 through 1973, this analysis
32 assumed the lead content of gasoline to be at the 1974 level. For the uncontrolled scenario, this analysis
33 used the 1974 lead content in leaded gasoline as the lead content in all gasoline for each year.
343
-------
-•1
Appendix G: Lead Benefits Analysis
; Estimating the Change In Blood Lead Levels from the
2 Change In the Amount of Lead In Gasoline
3 Several studies have found positive correlations between gasoline lead content and blood lead
4 levels.246 Data from the National Health and Nutrition Examination Survey (NHANES II) has been used
5 by other researchers who determined similar positive correlations between gasoline lead and blood lead
6 levels.247 :
7 The current analysis used a direct relationship between consumption of lead in gasoline and
s blood lead levels to estimate changes in blood lead levels resulting from Clean Air Act regulation of the
9 lead content of gasoline. This relationship was based on regression analyses of the reduction of leaded
10 gasoline presented in the 1985 Regulatory Impact Analysis (RIA).24* Several multiple regressions were
/; performed in the RIA to relate gasoline usage with individuals' blood lead levels, which were taken from
n NHANES II. These regressions of blood lead on gasoline usage controlled for such variables as age, sex,
13 degree of urbanization, alcohol consumption, smoking, occupational exposure, dietary factors, region of
14 the country, educational attainment, and income. The regressions suggested that a decrease of 100
is metric tons per day (MTD) of lead used in gasoline is associated with a decrease in mean blood lead
16 concentration of 2.14 ug/dL for whites and 2.04 fig/dL for blacks. In both of these regressions, gasoline
17 use was found to be a highly significant predictor of blood lead (p < 0.0001).249
is To determine a single gasoline usage-blood lead slope for the entire population of the U.S., this
19 analysis used the average of the slopes for blacks and for whites, weighted by the percentage of blacks
20 and whites in the U.S. during the time period of the analysis.230 The resulting relationship is 2.13 ug/dl
21 blood lead per 100 metric tons of lead in gasoline consumed per day. The same relationship was used to
22 model changes in both children's and adults' blood lead levels. The U.S. EPA (198S) analyzed data from
23 a study of black children in Chicago during the time period 1976 to 1980 and determined a slope of 2.08
24 ug/dl per 100 MTD. This slope for children is very similar to the one used in this analysis.
25 1970-Forward and 1990-Backward Approaches
26 As with the industrial processes and boilers analysis, this analysis used two different approaches
27 to determine mean blood lead levels based on changes in lead concentrations in gasoline. In the 1970-
2s forward approach, the calculations began with the estimated blood lead level for 1970. The change in
29 blood lead level from one year to the next was based upon the change in the amount of lead in gasoline
30 sold, as discussed above, for both the controlled and uncontrolled scenarios. For example, to calculate
M U.S. EPA, 1985; Billick et al., 1979; Billick et al., 1982.
247 Janney, 1982; Annest et al., 1983; Center for Disease Control, 1993; National Center for Health Statistics, 1993.
"* U.S. EPA, 1985.
"' U.S. EPA, 1985.
230 U.S. Department of Commerce, 1992. Although the percentages of blacks and whites changed slightly over this time period (1970-
1990), the change did not affect the value of the weighted slope.
344
-------
Appendix G: Lead Benefits Analysis
i the blood lead level for 1971, the calculated change in blood lead from 1970 to 1971 was added to the
2 1970 value. This process was repeated for each succeeding year up to 1990.
3 The 1990-backward approach began with a mean blood lead level in 1990 for the controlled
4 scenario. For the uncontrolled scenario, the starting blood lead was estimated from the 1990 level used
5 in the controlled scenario, plus an additional blood lead increment resulting from the difference between.
« the 1990 consumption of lead in gasoline in the controlled and uncontrolled scenarios. Again, the
7 difference in mean blood lead levels from one year to the next was based on the change in gasoline lead
s for the corresponding years. For example, the difference hi blood lead levels between 1990 and 1989
9 was subtracted from the 1990 level to determine the 1989 level. The process was continued for each year
10 back to 1970. •
u Relating Blood Lead Levels to Population Health
n Effects
n The mean blood lead levels calculated using the methods described above were used in the dose-
14 response functions for various health effects (e.g., hypertension, chronic heart disease, mortality). This
75 information was then combined with data on the resident population of the 48 conterminous states in
16 each year to determine the total incidence of these health effects attributable to lead in gasoline in the
n controlled and uncontrolled scenarios. A Department of Commerce Publication (1991) was used to
is obtain the total population in 1970, 1980, and 1983-1990, while a different publication was the source of
19 the 1975 population values.251 Linear interpolation was used to estimate the populations in years for
20 which specific data were not available.
27 For certain health effects, it was necessary to know the size of various age groups within the
22 population. Two different sources were used to estimate the proportions of the population in the age
23 groups of interest. A U.S. Census summary (Dept. of Commerce, 1990) was used for information for
24 1990 for children and adults and for 1980 for adults, and Census Telephone Staff (Dept. of Commerce,
25 1994) provided information for 1980 for children and 1970 for children and adults. The populations for
26 the intervening years were estimated by linear interpolation.
251 Dept. of Commerce, 1976.
345
-------
Appendix G: Lead Benefits Analysis
Changes in Leaded Gasoline Emissions and Resulting Decreased Blood Lead Levels and Health
Effects
Table 135 shows the estimated quantity of lead burned in gasoline in the five year intervals from
1970 to 1990. Tables 136 and 137 show the difference in health impacts between the two scenarios (for
lead in gasoline only) for the "forward-looking" and "backward-looking" analyses.
table 135. Lead Burned in Gasoline (la tons).
Controlled Scenario
Uncontrolled Scenario
1970
176,100
176,100
1975
179,200
202,600
1980
86,400
206,900
1985
22,000
214,400
1990
2,300
222,900
346
-------
-------
oo
£
.3
.5-
I
o
w
I
I
•fc
•S£
§*"
v
fif;
ta.
I
00
-------
Appendix G: Lead Benefits Analysis
Benefits Analysis References
2 Abt, 1995, The Impact of the Clean Air Act on Lead Pollution: Emissions Reductions, Health Effects,
3. • and Economic Benefits From 1970 to 1990. Draft January 19, 1995. Prepared for Economic
4
Analysis and Inn6vations( Division, Office of Policy Planning and Evaluation, US EPA.
s Annest, J.L., J.L: Pirkle, D. Makuc, J.W. Neese, D.D. Bayse, and M.G, Kovar. 1983. Chronological
6 trend in blood lead levels between 1976 and 1980. New England Journal of Medicine, 308:
7 1373-1377.
« Azar, R.D., et al. (1975). An epidemiologic approach to community air lead exposure using personal air
9 samplers. In: Griffin, 1*.B. and Knelson, J.H., eds. Lead. Stuttgart, West Germany: Georg
10 Thieme Publishers; pp.254-290. (Coulston, F. and Korte, F., eds. Environmental quality and
11 safety: supplement v. 2).
n Bellinger, D. Sloman, J., Leviton, A., Rabinowitz, M., Needleman, H.L., and Waternaux, C. 1991. Low-
13 level lead exposure and children's cognitive function in the preschool years. Pediatrics. Vol 87.
H No. 2:219-227.
is Bellinger, D.C., 1992, Lead Exposure, Intelligence and Academic Achievement. Pediatrics. Vol 90. No
i6 6:855.
n Billick, I.H., A.S. Curran, and D.R. Shier. 1979. Analysis of pediatric blood lead levels in New York
is City for 1970-1976. Environmental Health Perspectives. 31: 183-190.
19 Billick, I.H., et al. 1982. Predictions of pediatric blood lead levels from gasoline consumption. U.S.
20 Department of Housing and Urban Development. [Cited in U.S. EPA, 1985.]
21 Brunekreef, B.D., et al. (1981). The Arnhem lead study: 1. lead uptake by 1- to 3-year-old children
22 living in the vicinity of a secondary lead smelter in Arnhem, the Netherlands. Environ. Res. 25:
23 441-448.
24 Center for Disease Control (CDC). 1993. Personal communication between Abt Associates and Jim
25 Pirkle. November 16.
26 Centers for Disease Control (CDC). 1985. Preventing Lead Poisoning in Young Children. U.S.
27 Department of Health and Human Services, Public Health Service, Centers for Disease Control,
2s Atlanta, GA.
29 Centers for Disease Control (CDC). 199 la. Strategic Plan for Elimination of Childhood Lead
30 Poisoning. U.S. Department of Health and Human Services, Public Health Service, Centers for
31 Disease Control. February.
349
-------
Appendix G: Lead Benefits Analysis
i Centers for Disease Control (CDC). 1991b. Preventing Lead Poisoning in Young Children. U.S.
. 2 Department of Health and Human Services, Public Health Service, Centers for Disease Control,
3 Atlanta, GA. October.
4 Chamberlain, A.C.( 1983). Effect of airborne lead on blood lead. Atmos. Environ. 17:677-692.
5 Daines, R.H., et al. (1972). Air levels of lead inside and outside of homes. Ind. Med. Surg. 41: 26-28.
6 Dietrich, K.N., Krafft, K.M., Shukla, R., Bornschein, R.L., Succop, P.A. 1987. The Neurobehavioral
/ Effects of Prenatal and Early Postnatal Lead Exposure. In: Toxic Substances and Mental .
s Retardation: Neurobehavioral Toxicology and Teratology, S.R. Schroeder, Ed. American
9 Association of Mental Deficiency, Washington DC, pp. 71-95 (Monograph No. 8).
10 DOE, 1990. Petroleum Supply Annual, 1989, Volume 1. DOE publication number EIA-0340(89)/1
11 DOE, 1991. Petroleum Supply Annual, 1990, Volume 1. DOE publication number EIA-0340(90)/1
12
13 Environmental Law Institute (ELI) 1992, Projecting With and Without Clean Ah- Act Emissions for the
14 Section 812 Retrospective Analysis: A Methodology Based upon the Projection System Used hi
is the OAQPS "National Air Pollutant Emission Estimates: Reports. [Jorgenson/Wilcoxen Model
16 Projections], Jim Lockhart.
17 Fisher, A., L.G. Chestnut and D.M. Violette (1989). The value of reducing risks of death: a note on new
/g evidence. J. of Policy Analysis and Mgmt., 8(1):88-100.
19 Fugas, M., et al. (1973). Concentration levels and particle size distribution of lead in the air of an urban
20 and an industrial area as a basis for the calculation of population exposure. In: Barth, D., et al.
21 eds. Environmental health aspects of lead: proceedings, international symposium; October 1972;
22 Amsterdam, The Netherlands. Luxembourg: Commission of the European Communities, pp.
23 961-968.
24 .
25 Goldsmith, J.R. (1974). Food chain and health implications of airborne lead. Sacramento, CA: State of
26 California, Air Resources Board; report no. ARB-R-102-74-36. Available from NTIS,
27 Springfield, VA PB-248745.
2S
29 Industrial Economics, Inc. (lEc) (1992). Memorandum to Jim DeMocker, Office of Policy Analysis and
30 Review, from Robert E. Unsworth and James E. Neumann. "Review of Existing Value of Life
3i Estimates: Valuation Document." 6 November, 1992.
32 Industrial Economics, Inc. (lEc) (1993). Memorandum to Jim DeMocker, Office of Policy Analysis and
33 Review, from Robert E. Unsworth and James E. Neumann. "Revisions to the Proposed Value of
34 Life: Methodology for the Section 812 Retrospective." 3 May, 1993.
35 Janney, A. The relationship between gasoline lead emissions and blood poisoning in Americans.
36 Prepared for U.S. EPA, Office of Policy Analysis. [Cited in U.S. EPA, 1985.]
350
-------
Appendix G: Lead Benefits Analysis
i Johnson, D.E., etal. (1976). Base line levels of platinum and palladium in human tissue. Research
2 Triangle Park, N.C.: U.S. EPA, Health Effects Research Laboratory; EPA report no. EPA-600/1 -
3 76-019. available from: NTIS, Springfield, Va; PB-251885.
4 Kolb, J. and K. Longo. 1991. Memorandum to Joel Schwartz, U.S. EPA, Washington, DC, November 5.
5 .
6 Landrigan, P. J. and E.L. Baker (1981). Exposure of children to heavy metals from smelter:
7 epidemiology and toxic consequences. Environ. Res. 25: 204-224.
8 .
9 Landrigan, P.J., etal. (1975). Epidemic lead absorption near an ore smelter: the role of particulate lead.
to N. Engl. J. Med. 292: 123-129. •
a
12 McGee and Gordon. 1976. The Results of the Framingham Study Applied to Four Other U.S;-based
13 Epidemiologic Studies of Coronary Heart Disease. The Framingham Study: An Epidemiological
14 Investigation of Cardiovascular Disease. Section 31, April.
is National Center for Health Statistics (NCHS). 1993. Facsimile received by Abt Associates from
16 Margaret McDowell regarding the types of laboratory tests conducted during NHANES I.
n December 14.
is National Center for Health Statistics (NCHS). 1993. Personal communication between Abt Associates
19 and NCHS Public Information Specialist. November 3.
20 National Energy Accounts, Bureau of Economic Analysis.
21 NHANES U, National Health and Nutrition Examination Survey, 1976-1980.
22 NHANES, National Health and Nutrition Examination Survey.
23
24 Nordman, C.H. (1975). Environmental lead exposure in finland: a study on selected population groups
2s [dissertation]. Helsinki, Finland: University of Helsinki.
26 Oliver, T. 1911. Lead Poisoning and the Race. British MedicalJournal 1(2628): 1096-1098. [Cited in
27 USEPA (1990)!]
23 Piomelli et al. 1984. Management of childhood lead poisoning. Pediatrics; 4:105.
29 Pirkle, J.L., J. Schwartz, J.R. Landis, and W.R. Harlan. 1985. The relationship between blood lead
so levels and blood pressure and its cardiovascular risk implications. American Journal of
31 Epidemiology. 121:246-258.
32 Pirkle, J. L., et al. 1994. "Decline in Blood Lead Levels in the United States, the National Health and
33 Nutrition Examination Survey (NHANES). JAMA, July 27, 1994, v272 n4, p284.
351
-------
Appendix G: Lead Benefits Analysis
i Pooling Project Research Group. 1978. Relationship of blood pressure, serum cholesterol, smoking
2 habit, relative weight and ECG abnormalities to incidence of major coronary events: final report
s of the Pooling Project. Journal of Chronic Disease. Vol. 31.
4 Rabinowitz, M., Bellinger, D., Leviton, A., Needleman, H., and Schoenbaum, S. 1987. Pregnancy
j hypertension, blood pressure during labor, and blood lead levels. Hypertension. Vol 10., No. 4,
6 October. . .
7 Schwartz, J. 1988. The relationship between blood lead and blood pressure in the NHANESII Survey.
a Environmental Health Perspectives. Vol. 78:15-22.
9 Schwartz, J. 1990. Lead, blood pressure, and cardiovascular disease in men and women. Environmental
w Health Perspectives, in press.
;; Schwartz,!. 1992a. Blood lead and blood pressure: a meta-analysis. Presented at the Annual Meeting
12 of Collegium Ramazzini. November.
n Schwartz,!. 1992b. Chapter 13: Lead, Blood Pressure and Cardiovascular Disease. In: Human Lead
14 Exposure, H. L. Needleman, Ed. CRC Press.
is Schwartz,!. 1993. Beyond LOEL's, p values, and vote counting: methods for looking at the shapes and
16 strengths of associations. Neurotoxicology. Vol. 14. No. 2/3. October;
n Shurtleff, D. 1974. Some Characteristics Related to the Incidence of Cardiovascular Disease and Death.
is The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease. Section
19 30, February.
20 Silbergeld, E.K., Schwartz, J., and K. Mahaffey. 1988. Lead and osteoporosis: mobilization of lead
21 from bone in postmenopausal women. Environmental Research. 47, 79-94.
22 Snee, R.D. (1981). Evaluation of studies of the relationship between blood lead and air lead. Int. Arch.
n Occup. Environ. Health 48: 219-242.
24 Tepper, L.B. and L.S. Levin (1975). A survey of air and population lead levels in selected American
25 communities. In: Griffin, T.B.; Knelson, J.H., eds. Lead. Stuttgart, West Germany: Georg
26 Thieme Publishers; pp. 152-196. (Coulston, F.; Korte, f., eds. Environmental quality and
27 safety: supplement v. 2).
2s Tsuchiya, K., etal. (1975). Study of lead concentrations in atmosphere and population in Japan. In:
29 Griffin, T.B. and Knelson, J.H.. eds. Lead. Stuttgart, West Germany: Georg Thieme Publishers;
30 pp.95-145. (Coulston, F.; Korte, F., eds/ Environmental quality and safety: supplement v. 2)
31 U.S. Census (1982) United States Summary, General Population Characteristics, Table 41: Single Years
32 of Age by Race, Spanish Origin, and Sex: 1980.
352
-------
Appendix G: Lead Benefits Analysis
i U.S. Census (1992) United States Summary, General Population Characteristics, Table 13: Single Years
• 2 by Sex, Race, and Hispanic Origin: 1990.
3 U.S. Department of Commerce. 1990. U.S. Census, United States Summary, General Population
4 Characteristics. .
s U.S. Department of Commerce. 1980. U.S. Census, United States Summary, General Population
6 Characteristics.
7 U.S. Department of Commerce. 1993. Personal Communication between Bureau of Census, Population,
s Age and Sex Telephone Staff and Karl Kuellmer of Abt Associates on December 8,1993
9 U.S. Department of Commerce. 1994. Personal Communication between Bureau of Census, Population,
10 Age and Sex: Telephone Staff, and Karl Kuellmer of Abt Associates on February 7,1994.
// U.S. Department of Commerce. 1987 Census of Manufacturers.
12 U.S. Department of Commerce. 1992. Statistical Abstract of the United States: 112th Edition. Bureau
13 of the Census. Washington, DC.
u U.S. Department of Commerce. 1991. Statistical Abstract of the United States: 111th Edition. Bureau
is of the Census. Washington, DC.
16 U.S. Department of Commerce. 1976. Statistical Abstract of the United States: 95th Edition, Bureau of
n the Census. Washington, DC.
18
19 U.S. DOE, 1992, Cost and Quality of Fuels for Electric Utility Plants 1991. DOE/EIA-0191(91) Energy
20 Information Administration, August 1992.
21 U.S. Environmental Protection Agency. 1985. Costs and Benefits of Reducing Lead in Gasoline: Final
22 Regulatory Impact Analysis. Prepared by U.S. Environmental Protection Agency, Office of
23 Policy Analysis, Economic Analysis Division. February.
24 U.S. Environmental Protection Agency. 1986a. Reducing Lead in Drinking Water: A Benefit Analysis.
25 Prepared by U.S. Environmental Protection Agency, Office of Policy Planning and Evaluation,
26 Draft Final Report. December.
27 U.S. Environmental Protection Agency. 1986b. Air Quality Criteria for Lead: Volume III.
28 Environmental Criteria and Assessment Office, Research Triangle Park, NC. EPA-600/8-
29 83/028cF. June.
30 U.S. Environmental Protection Agency. 1987. Methodology for Valuing Health Risks of Ambient Lead
31 Exposure. Prepared by Mathtech, Inc. for U.S. Environmental Protection Agency, Office of Air
32 Quality Planning and Standards, Ambient Standards Branch, Contract No. 68-02-4323.
353
-------
Appendix G: Lead Benefits Analysis
i U.S. Environmental Protection Agency. 1990. Review of the National Ambient Air Quality Standards
2 for Lead: Assessment of Scientific and Technical Information. OAQPS Staff Paper, Air Quality
3 Management Division, Research Triangle Park, N.C. December.
4 U.S. Environmental Protection Agency. 1994. Guidance Manual for the Integrated Exposure Uptake
5 Biokinetic Model for Lead in Children, February 1994, Publication Number 9285.7-14-1, EPA
6 540-R-93-081.
7 U.S. Environmental Protection Agency (1994). Cost and Benefits of the Smoke Free Environment Act of
s 1993 [HR 3434]. Prepared for Congressman Henry Waxman by the Indoor Air Division, Office
9 of Air and Radiation,
10 U.S. Environmental Protection Agency (U.S. EPA). 1986. Air Quality Criteria for Lead: Volume HI,
11 Environmental Criteria and Assessment Office, Research Triangle Park, NC. EPA-600/8-
a 83/028cF. June.
a U.S. EPA (1990). Office of Air Quality Planning and Standards, Airs Facility Subsystem Source
14 Classification Codes and Emission Factor Listing for Criteria Air Pollutants, Publication Number
is EPA-450/4-90-003, US EPA, Research Triangle Park, March 1990..
u U.S. EPA (1990). National Air Pollutant Emission Estimates 1940-1988. U.S. EPA. Office of Air
n Quality Planning and Standards, Technical Support Division, National Air Data Branch.
u Research Triangle Park. EPA no. EPA-450/4-90-001.
19 U.S. EPA (1991). National Air Quality and Emissions Trends Report, 1989. U.S. EPA. Office of Air
20 Quality Planning and Standards. Research Triangle Park. EPA no. EPA-450/4-91-003.
21 U.S. EPA (1991 b). The Interim Emissions Inventory. U.S. EPA. Office of Air Quality Planning and
22 Standards, Technical Support Division, Source Receptor Analysis Branch. Research Triangle
23 Park.
24 U.S. EPA (1992). 1990 Toxics .Release Inventory. U.S. EPA. Office of Pollution Prevention and
25 Toxics. Washington, D.C. 20460. EPA no. EPA-700-S-92-002.
26
U.S. EPA database. Graphical Exposure Modeling System Database (GEMS).
27 Viscusi, W.K. (1992). Fatal Trade Offs: Public and Private Responsibilities for Risk. New York:
28 Oxford University Press.
29 Wallsten and Whitfield. 1986. Assessing the Risks to Young Children of Three Effects Associated with
30 Elevated Blood Lead Levels. Argonne National Laboratory. December.
354
-------
Appendix Hs Air Toxics
Introduction
3 Air toxics are defined as air pollutants other than those six criteria pollutants for which EPA sets
4 acceptable concentrations in ambient air. The SARA 3 13 Toxic Release Inventory (TRI), covering 328
5 of the approximately 3000 potentially hazardous compounds detected in air, estimated that
t. approximately 1 .2 millions tons of air toxics were released to the atmosphere hi 1987 from U.S.
7 stationary sources alone. While the TRI estimate tends to understate emissions of toxics for a number of
a reasons, it does show that large quantities of toxics are emitted into the atmosphere annually.
» Effects of air toxic emissions are divided into three categories for study and assessment: cancer;
w "noncancer" effects, e.g. a wide variety of serious health effects such as abnormal development, birth
11 defects, neurological impairment, or reproductive impairment, etc.; and ecological effects. Each year,
n these air toxic emissions contribute to significant adverse effects on human health, human welfare, and
is ecosystems. In EPA's 1987 Unfinished Business Report*2 cancer and noncancer air toxic risk estimates
14 were considered sufficiently high, relative to risks addressed by other EPA programs, that the air toxics
is program area was among the few rated "high risk".
Limited Scope of this Assessment
n The effects of air toxic emissions are difficult to quantify. Adverse health effects of toxics are
is often irreversible, not mitigated or eliminated by reduction in ongoing exposure, and involve particularly
19 painful and/or protracted disease. Therefore these effects are not readily studied and quantified in human
20 clinical studies, as contrasted with, for example, ambient ozone. In addition, epidemiological studies of
21 effects in exposed populations are often confounded by simultaneous exposure of subjects to a variety of
22 pollutants. Therefore, effects of these pollutants are often quantified by extrapolating data from animal
23 studies to human exposure and expressed as risk per unit of exposure. Incidence of noncancer effects,
24 for example, often are difficult to translate into monetized benefits.
25 Similarly, the quantification of ecological effects due to emissions of air toxics is hampered by
26 lack of sufficient information regarding contribution of sources to exposure, associations between
27 exposure to mixtures of toxics and various ecological endpoints, and economic valuation for ecological
2* endpoints.
252 U.S. EPA. Office of Policy Planning and Evaluation. Unfinished Business: A Comparative Assessment of Environmental Problems.
February 1987.
355
-------
Appendix H: Air Toxics
Table 138. Health and Welfare Effects of Hazardous Air Pollutants.
Effect Category
Human Health
Human Welfare
Ecological
Other Welfare
Quantified Effects
Cancer Mortality
-nonutility
stationary source
- mobile source
Unquantified Effects
Cancer Mortality
- utility source
- area, source
Noncancer effects
- neurological
-respiratory
- ^productive.
- hematopoetic
- developmental
- immunological
- organ toxicity
Decreased income and
recreation
opportunities due to
fish advisories
Odors,
Effects on wildlife
Effects on plants
Ecosystem effects
Loss of biological
diversity
Visibility
Materials Damage
Other Possible Effects
, .
Decreased income resulting
from decreased
physical performance
Effects on global climate
T
The air toxics portion of this study is, of necessity, separate and more qualitative in nature than
the benefit analysis conducted for the criteria air pollutants. Limitations in the quantitative analyses of
air toxic effects led the Project Team to decide to exclude the available quantitative results from the
primary analysis of CAA costs and benefits. Table 138 presents the range of potential human health and
ecological effects that can occur due to air toxics exposure. As indicated, this appendix presents
alternative quantitative estimates of benefits of CAA air toxics control for the cancer mortality endpoint
for only nonutility stationary and mobile source categories. Noncancer effects and ecological effects are
described qualitatively.
356
-------
Appendix H: Air Toxics
, History of Air Toxics Standards under the
Clean Air Act of 197O
2 . . ' .
3 The 1970 Clean Air Act required the EPA to list a chemical as a hazardous air pollutant if it met
4 the legislative definition provided:
5 "The term 'hozordousair pollutant'means cm air pollutant to \vhich no ambient air
6 quality standard is applicable and which in the judgment of the Administrator may
7 cause, or contribute to, an increase in mortality or an increase in serious irreversible, or
s incapacitating reversible, illness. "2"
9 Once a HAP was listed, the EPA was required to:
w "establish any such standard at the level which in his judgment provides an ample
11 margin of safety to protect the public health from such hazardous air pollutant. "2"
n In other words the EPA had to first determine that a chemical was a HAP, and then regulate the
n emissions of each HAP based solely on human health effects and with an ample margin of safety. This
14 regulatory mandate proved extremely difficult for EPA to fulfill, for reasons discussed below, and the
is result was that only seven HAPs were regulated over a period of 20 years.
16 Listing chemicals became a difficult task because of debates within and outside of the EPA
n surrounding issues of how much data are needed and which methodologies should be used to list a
is chemical as a HAP. An even more difficult issue was how to define the Congressional mandate to
19 provide an "ample margin of safety." For carcinogens, there is generally no threshold of exposure
20 considered to be without risk. What level of risk, then, is acceptable, and how should it be calculated?
21 The EPA struggled to provide answers to these questions, and was challenged in court. The end result
22 was a 1987 ruling by the D.C. Circuit Court that provided the EPA with a legal framework with which to
21 determine an "ample margin of safety." This framework was interpreted and used by the EPA in its 1989
24 benzene regulations.
25
26
Quantifiable Stationary Source Air Toxics
27 One might be tempted to presume that the few federal HAP standards set would have achieved
28 relatively substantial reductions in quantifiable risk. While some standards set under section 112 of the
29 Clean Air Act appear to have achieved significant reductions in cancer incidence, the coverage,
30 quantification, and monetization of the full range of potential adverse effects remains severely limited.
233 42U.S.C.§1857(a)(l).
254 42U.S.C. §1857(b).
357
-------
Appendix H: Air Toxics
i This fact serves to highlight the inadequacy of current methods of evaluating HAP control benefits. This
2 limited ability to estimate the total human health and ecological benefits of HAP reductions is an
3 important area for future research. Thus the quantifiable benefits for CAA air toxics control presented
4 here are limited in scope.
5 There are three sources of information that provide a picture of potential stationary source air
6 toxics benefits of the CAA. EPA's Cancer Risk studies attempted to broadly assess the magnitude and
7 nature of the air toxics problem by developing quantitative estimates of cancer risks posed by selected air
s toxics and their sources. Secondly, risk assessments conducted in conjunction with the promulgation of
9 National Emissions Standards for Hazardous Air Pollutants (NESHAPs) offer a snapshot of potential
10 monetized cancer mortality benefits. Finally, the Project Team attempted to estimate historical non-
77 utility stationary source HAP-related direct inhalation cancer incidence reductions. Results from each of
n these studies are presented below.
13 ERA Analyses of Cancer Risks from Selected Air Toxic
H Pollutants
is The Agency conducted two efforts to broadly assess the magnitude and nature of the air toxics
i6 problem. The 1985 report entitled, "The Air Toxics Problem in the United States: An Analysis of Cancer
17 Risks for Selected Pollutants"231 otherwise known as the "Six Month Study," was intended to serve as a
is "scoping" study to provide a quick assessment of the air toxics problem utilizing only readily available
19 data on compound potencies, emissions, and ambient pollutant concentrations. The Agency updated this
20 analysis of cancer risks in the 1990 ^report entitled "Cancer Risk from Outdoor Exposure to Air Toxics"
21
referred to here as the "1990 Cancer Risk study.
"256
22 For the pollutant and source categories examined, the 1990 Cancer Risk study estimated the
23 total nationwide cancer incidence due to outdoor concentrations of air toxics to range from as many as
24 1,700 - 2,700 excess cancer cases per year with 14 compounds accounting for approximately 95 percent
25 . of the annual cancer cases. Additionally, point sources contribute 25 percent of annual cases and area
26 sources contribute 75 percent of annual cases, with mobile sources accounting for 56 percent of the
27 nationwide total."7
28 The Six Month study indicates that the criteria air pollutant programs appear to have done more
29 to reduce air toxics levels during the 1970 to 1990 period man have regulatory actions aimed at specific
30 toxic compounds promulgated during the same period. Metals and polynuclear compounds usually are
31 emitted as particulate matter and most of the volatile organic compounds are ozone precursors. As such,
32 they are regulated under State Implementation Plan (SIP) and New Source Performance Standard (NSPS)
255 U.S.EPA. Office of Air Quality Planning and Standards. The Air Toxics Problem in the United States: An Analysis of Cancer Risks
for Selected Pollutants. May 1985. EPA-450/1-85-001.
236 U.S. EPA. Office of Air Quality Planning and Standards. Cancer Risk from Outdoor Exposure to Air Toxics. September 1990. EPA-
450/1-90-004a.
257 The 1990 Cancer Risk study reported approximately 500 - 900 more cancer cases per year than the Six Month Study due primarily to
the inclusion of more pollutants, better accounting of emissions sources, and, in some cases, increases in unit risk estimates.
358
-------
Appendix H: Air Toxics
i programs and Title II motor vehicle regulations. A number of reports cited indicate significant
2 reductions in air toxic emissions attributable to actions taken under SIP, NSPS and mobile source
3 programs. Additionally, EPA conducted a comparison of air quality and emissions data for 1970 with the
4 estimates of cancer incidence for 1980.258 Methods, assumptions and pollutants included were held
5 constant over the period. The analysis showed a significant decrease in incidence during the decade due
6 to improvements in air quality, presumably related to general regulatory programs. For the 16 pollutants
7 studied, estimated nationwide cancer incidence decreased from 3600 in 1970 to 1600 in 1980. The 1990
s Cancer Risk Study did not attempt t6 update this analysis. •
9 Although it is difficult to make quantitative conclusions from these two studies regarding the
w benefits of CAA air toxics control, it is apparent that the pollutant-specific and source category-specific
n NESHAPs were not structured to reduce significant air toxic emissions from area and mobile sources. In
12 fact, the 1990 Cancer Risk Study indicates that considerable cancer risk remained prior to passage of the
n 1990 CAA Amendments: as many as 2,700 excess cancer cases annually. However, some studies
14 indicate that the criteria ah* pollutant program played a critical role during the 1970 to 1990 period in
is achieving air toxic emission reductions and therefore decreasing cancer risk.
,6 Cancer Risk Estimates from NESHAR Risk
n Assessments
is In looking back at the estimated effects of the HAP standards, EPA found that the effects of the
19 NESHAPs were not quantified completely. These estimates occurred at a time when emission estimation
20 and risk assessment methodologies for HAPs were first being developed. One consequence is that
21 because emissions were not fully characterized, air toxic exposures could not be completely assessed.
22 Additionally, most assessments only focused on the specific HAP being listed under the CAA and did
23 not assess the reduction of other pollutants, which are currently considered HAPs. For example, while
24 the vinyl chloride standard reduces emissions of ethylene dichloride, these emission reductions were not
25 assessed in the risk assessment. In a different context, reductions of HAP may also achieve reductions of
26 VOCandPM. The benefits of such reductions generally were also not evaluated. In addition, EPA
27 generally did not assess the potential exposure to high, short-term concentrations of HAP and therefore
is did not know whether toxics effects from acute exposures would have been predicted and possibly
29 addressed by the HAP standards.
30. In addition, people living near emission sources of concern are often exposed to a mix of
31 pollutants at once. Some pollutants have been shown to act synergistically together to create a health
32 risk greater than the risk that would be expected by simply adding the two exposure levels together.
33 More research is needed to understand the effects of multiple-pollutant exposures. Finally, HAP risks
34 tend to be distributed unevenly across exposed populations, with particularly high exposures occurring
35 closest to emission sources. It should be noted that HAP exposure to specific populations may tend to
36 fall disproportionately among the poor and minorities, who are more likely to live in close proximity to
37 emitting facilities.
"* Hunt, W.F., Faoro, R.B. and Curran, T.C., "Estimation of Cancer Incidence Cases and Rates for Selected Toxic Air Pollutants Using
Ambient Air Pollution Data, 1970 vs. 1980". U.S. EPA. April 1985.
359
-------
Appendix H: Air Toxics
With the above caveats in mind, Table 139 provides information about maximum individual risk
taken from the Federal Register notices for the NESHAPs promulgated before the 1990 amendments to
Table 139. Cancer Incidence Reductions and Monetized Benefits for NESHAPs.
Pollutant
benzene
benzene
benzene
benzene
benzene
arsenic
arsenic
asbestos
vinyl
chloride
Source
Category
coke by-
product
storage
vessels
waste
operations
transfer
operations
primary
copper
olaes maraif
demolition
PVC
.production
Year
Promulgated
1985
1984
1982
1986
1987
1986
1986
1973
1975
Pre-Reg
Maiimum
Individual
Risk
LSxUr3
7x10*
4.5x10-*
2x10*
oxlCr3
1.3x10? to
5x10*
7x«r*to
Sxifr*
Post-Reg
AwitXIItlllRI
Individual
Risk
4,5x10*
2X19-4
3-xlff*
5xio*
4xlff*
1.2x10*
to3xlCr*
1.7XKT4
tooxlCr*
Reduction m
Cancer
IfKMfflKV
-------
Appendix H: Air Toxics
, Non-utility Stationary Source Cancer indolence
Reductions
2 ' • . . .
3 The Project Team commissioned two studies to estimate reductions in cancer incidence due to
/ pre-1990 NESHAPs: the PES Study and the ICF Re-analysis. The methodology used for most air
5 pollutant evaluations involved a "back calculation" for the estimation of incidence reductions. However,
6 the EPA has elected not to rely on the results of this analysis given critical methodological flaws. Despite
7 the Project Team's concerns, the methodology and results of the two studies are presented below in the
s interest of full disclosure and to guide efforts to develop a more valid and reliable analysis of the health-
» related benefits of HAP reductions in the upcoming section 812 Prospective studies.
w RES Study
11 Methodology
n The first attempt to estimate, for this study, historical non-utility stationary source HAP-related
n direct inhalation cancer incidence reductions was conducted by Pacific Environmental Services (PES).
14 The basic approach used in the PES study was to adjust the cancer incidence estimates developed for
is EPA's 1990 Cancer Risk study to reflect the changes in emissions of, and exposures to, 14 key HAPs:
i6 arsenic, asbestos, benzene, 1,3-butadiene, carbon tetrachloride, chloroform, hexavalent chromium,
17 dioxin, ethylene dichloride, ethylene dibromide, formaldehyde, gasoline vapors, products of incomplete
is combustion (PICs), and vinyl chloride.
]9 The first step was to compile baseline incidence levels, defined as cancer cases per million
20 population, for each of the 14 pollutants. The point estimates of incidence from the 1990 Cancer Risk
21 study were used for this purpose. For some source categories, the "best point estimate" from the 1990
22 Cancer Risk study was used, for others a mid-point was selected.259 These baseline incidence levels
23 were based on measured ambient concentrations of the pollutant, modeled concentrations, or both.
24 The second step involved allocating baseline incidence levels to the individual source categories
25 known to emit the relevant pollutant. In some cases, adjustments were made to reflect differences among
26 the vintages of source category-specific data.260 All baseline incidence estimates were ultimately
27 expressed relative to a 1985 base year.261 The assumption was then made that source-category incidence
2« rates were proportional to the level of emissions from that source category.
259 For some of the source categories, the original NESHAP/Air Toxic Exposure and Risk Information System (NESHAP/ATERIS)
estimates of incidence were not available, in which case the baseline incidence was obtained from the 1989 National Air Toxics Information
Clearinghouse( NATICH) Database Report. (See PES, "Draft Summary of Methodology Used for Cancer from Stationary Sources,"
memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA, March 22,1993, p. 2.).
260 For example, six discrete sources for vinyl chloride were identified in the Six-Month Study Update. Point estimate incidences for each
of these source categories came from separate references with databases corresponding to different years. (See PES, "Retrospective Analysis
for Section 812(a) Benefits Study," September 30,1992, p. 8.)
261 See PES, March 22,1993 memorandum, p. 3.
361
-------
Appendix H: Air Toxics
i Next, levels of control for each source category-specific incidence rate were estimated for each /
2 of the target years of the present analysis (i.e., 1970, 1975, 1980, 1985, and 1990).262 Source category-
3 specific activity level indicators were then established and linked to changes in corresponding activity
4 indicators provided by the J/W macroeconomic modeling results. Activity levels were estimated for
5 each source category, for each of the target years, and for each of the two scenarios.
6 Finally, source category/pollutant combination incidence levels for both the control and no-
7 control scenarios were developed. These incidence levels were developed based on the baseline
s incidence levels, the activity indicators, and the control levels for each year. Both of these latter two
9 factors varied between the control and no-control scenarios. The activity levels differed based on the
10 specific levels of related sector economic activity predicted by the J/W model for the control and no-
11 control scenario. The control levels prevailing in each of the target years were used for the control
12 scenario, and the 1970 control level was applied throughout the 1970 to 1990 period for the no-control
a scenario.263 The formula used for these calculations was as follows:264
r, "
l
21
14 where:
is I = cancer incidence for a source category-pollutant combination
16 A = activity level for a source category
n P = population
is C = control level for a source category-pollutant combination
19 ty target year (1970... 1990)
20 by = base year
Findings
22 . The PES analysis concluded that substantial reductions in HAP-related cancer cases were
23 achieved during the reference period of the present study. The vast majority of these estimated
24 reductions were attributable to reduced exposures to asbestos, particularly from manufacturing and
25 fabricating sources.265 In fact, roughly 75 percent of the total reduction in cancer cases averaged over the
262 Control level estimates were based on one of the following: control efficiencies for related criteria pollutants defined in the criteria
pollutant analysis, reference documents such as Control Technology Guidelines (CTGs) or Background Information Documents (BIDs),
preambles for related regulations, or EPA experts. (See PES, March 22,1993 memorandum, p. 3.)
243 More detailed descriptions of the methodology and associated uncertainties are provided in "Retrospective Analysis fat Section 812(a)
Benefits Study," a September 30,1992 memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA.
244 See PES, March 22,1993 memorandum, p. 4.
265 PES, "Cancer Risk Estimates from Stationary Sources," memorandum from Ken Meardon, PES to Vasu Kilaru, US EPA, March 5,
1993.
362
-------
Appendix H: Air Toxics
i
2
3
4
5
6
7
S
9
10
11
12
13
14
15
16
17
IS
19
20
21
22
1970 to 1990 period were attributed to
asbestos control.266 Figure 52 summarizes the
PES study overall cancer incidence
reductions and the relative contribution of
asbestos-related reductions over the study
period.
Figure 52. PES Estimated Reductions in HAP-Related
Cancer Cases.
•a
o
JS
1975
1980 1985
Year
1990
The Project Team had several
concerns about the PES results. First and
foremost, the reductions in asbestos-related
cancer cases appeared to be substantially
higher than expected, particularly in the
earlier target years. Second, the control
scenario activity level indicators for several
sources with which Project Team members
were familiar did not appear to be even
remotely consistent with actual historical
activity patterns.267 Finally, the level of documentation of the analytical methodologies, assumptions,
and results was insufficient to ascertain the validity and reliability of the results. Ultimately, the Project
Team determined that it was necessary to conduct a formal review and re-analysis of the cancer
incidence reductions associated with non-utility stationary source HAP controls. The results of the PES
analysis remain a relevant part of the record of the. present study, however, since they provided a
substantial basis for the subsequent re-analysis by ICF Incorporated.
23
25
26
27.
28
29
3.0
31
32
ICF Re-analysis
24 Methodology
The purposes of the ICF Re-analysis were to examine the methodology and results of the PES
study, particularly to address the aforementioned concerns of the Project Team, and to develop a revised
set of estimates. Due to significant constraints on the resources remaining for HAP analysis in the
section 812 study, however, only a few key HAPs could be investigated in depth and many important
issues could not be addressed.268 Furthermore, the effects of two early and potentially important HAP
standards -the Beryllium and Mercury NESHAPs- could not be evaluated. Nevertheless, the ICF Re-
analysis clarified some potential sources of uncertainty in the PES results and provided revised cancer
incidence reduction estimates for several HAPs.
266 ICF, "Direct Inhalation Incidence Benefits," Draft Report, November 11,1994, p. 10.
267 For example, the activity indicators for Municipal Waste Combustors (MWCs) incorporated in the PES analysis decline dramatically
throughout the 1975 to 1990 period. (See PES, March 5,1993 memorandum to Vasu Kilaru, p. 10). In reality, overall MWC capacity and
throughput increased significantly over this period.
261 For example, the Project Team sought to develop and apply a methodology for estimating a central tendency estimate for the total
carcinogenic risk imposed by all the HAPs examined. The intent was to address concerns about potential overestimation of aggregate risk
measures when combining upper bound risk estimates of multiple HAPs. Unfortunately, resources were insufficient to continue development
of this methodology.
363
-------
Appendix H: Air Toxics
i A key uncertainty in the PES results was associated with the use of a "back-calculation"
2 technique to estimate incidence reductions for some HAPs. The back-calculation technique estimates
3 uncontrolled incidence by dividing residual incidence by the assumed control efficiency. This approach
4 means uncontrolled incidence, and therefore incidence reductions, are highly sensitive to small changes
s in assumed control efficiency.269 In some cases, the PES analysis may have used control efficiencies
6 which were too high, resulting in overestimation of uncontrolled incidence and therefore incidence
7 reductions attributable to the CAA.270 The vinyl chloride incidence reduction estimates appear to be
a significantly influenced by the use of this back-calculation technique. Another important source of
9 uncertainty identified by ICF involved the potential overestimation of incidence totals when source
10 apportionment is based on measured ambient concentrations.271 ICF was unable, however, to perform an
n extensive evaluation of the activity level indicators used in the PES study.272
12 The first step undertaken in the re-analysis was to conduct a screening test to identify the HAPs
n which accounted for the most significant estimated incidence reductions. Based on this screening
14 analysis, ICF eliminated 1,3-butadiene, carbon tetrachloride, chloroform, gasoline vapors, chromium,
is formaldehyde, and PICs from the detailed re-analysis effort.
16 Detailed reviews were then conducted for the remaining HAPs: vinyl chloride, dioxins, ethylene
17 dibromide (EDB), ethylene dichloride (EDC), benzene, asbestos, and arsenic. In the re-analysis of these
is HAPs, ICF determined whether a forward- or back-calculation technique was used for the relevant
19 source categories of a given HAP, reviewed the regulatory history of the relevant source categories to re-
20 evaluate the assumed control efficiencies, and reviewed the upper-bound unit risk factor for each HAP.
21 Revised total incidence reduction estimates for each HAP and for each target year were then calculated
22 using the same basic calculation procedure used by PES. Finally, ICF identified a number of residual
23 deficiencies in the analysis which could only be addressed through additional research and analysis.273
24 Findings
a The ICF Re-analysis largely affirmed the original results obtained by PES; primarily because the
26 PES analysis itself served as the basis for the re-analysis and only minor adjustments were adopted for
27 many critical variables. In particular, most Project Team concerns regarding the PES methodology could
28 not be resolved, including uncertainties associated with activity levels, assumed control efficiencies, and
29 the unexpectedly high estimated incidence reductions associated with asbestos. In fact, the ICF Re-
30 analysis produced a revised upper bound estimate for vinyl chloride-related incidence reductions which
31 were even higher than the asbestos benefits.
269 An example of this back-calculation technique illustrating the sensitivity to the assumed control efficiency is presented on page 12 of
the draft ICF report
270 See ICF Draft Report, p. 12.
171 See ICF Draft Report, p. 9.
272 See ICF Draft Report, p. 13.
273 Additional details of the ICF Re-analysis methodology can be found in ICF, "Direct Inhalation Incidence Benefits," Draft Report,
November 11,1994. • .
364
-------
Appendix H: Air Toxics
i
2
3
4
5
6
7
8
9
10
11
11
13
14
15
16
17
IS
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Figure 53. ICF Estimated Reductions in Total HAP-Related
Cancer Cases Using Upper Bound Asbestos Incidence and
Lower Bound Non-Asbestos HAP Incidence.
6
Several sets of results were developed by ICF and presented in either the November 1994 draft
report or in briefing materials prepared for the Science Advisory Board Clean Air Act Compliance
Analysis Council Physical Effects Subcommittee (SAB/CAACACPERS) in May 1995. The first set of
results is based on the assumption of 100 percent source compliance with HAP control requirements. An
alternative set of results was developed
assuming an 80 percent compliance rate with
applicable standards. Given the linear effect
of changes in compliance rates, these results
were precisely 20 percent lower than the first
set of estimates. At the May 1995
CAACACPERS briefing, estimates based on
the 100 percent compliance estimates were
presented. For asbestos, the revised
incidence reductions were presented and
characterized as upper bound. The asbestos
estimates were then combined with upper and
lower bound estimates for vinyl chloride and
for "all other compounds." Figure 53
presents the total cancer incidence reductions
derived from the ICF Re-analysis, using the
asbestos estimates combined with the lower
bound estimates for non-asbestos HAPs.
Figure 54 presents a comparable compilation
reflecting the upper bound estimates for all
HAPs.
tnOther HAPs 1
•^Asbestos I
1975 1980 1985 1990
Year
The Project Team remains concerned
about these incidence reduction estimates,
particularly given the doubts raised by the
SAB/CAACACPERS at the May 1995
presentation of these results. For instance,
several critical assumptions are needed to
make this analysis valid when applied to
EPA's NESHAPs. The flaws in these
assumptions are described below.
Figure 54. ICF Estimated Reduction in Total HAP-Related
Cancer Cases Using Upper Bound Incidence for All HAPs.
12
'
I
oOtherHAPsI
•Asbestos I
1975 1980 1985 1990
Year
(1) The risk estimates described in
the 1990 Cancer Risk study, which served as
the baseline for determining risk reductions,
were accepted without question. There are
myriad uncertainties in these estimates that must be recognized, as the study was designed only to
generate rough order-of-magnitude estimates of the extent of the air toxics cancer problem.
(2) The percent control efficiency for emission reductions, which is calculated in each standard,
would have to apply across every facility. Typically, the emission reductions vary between facilities;
using a single average reduction could skew the results
365
-------
Appendix H: Air Toxics
i (3) There is a direct correlation between the number of tons of emissions reduced and incidence '
2 reduced by a specific regulation. Given the assumption of a linear, non-threshold dose-response curve
3 (as is typically done for cancer), this is theoretically correct.
4 (4) Finally, the back calculation approach assumes that there is 100 percent compliance with the
5 regulation.
6 EPA staff reviewed the "back calculation" approach for one of the more controversial aspects of
7 the vinyl chloride (VC) NESHAP. The PES study estimates benefits at 426 cases reduced in 1990. The
s ICF Re-analysis resulted in an even higher estimate, between 1,000 and 7,000 cases annually. An
9 analysis by EPA staff indicated that these vinyl chloride risk estimates are highly suspect given historical
10 cancer incidence data for hepatic angiocarcinoma, a specific cancer that has been linked to vinyl chloride
/; (Koppikar and Fegley, 1995). The following analysis demonstrates the inadequacies of the assumptions
n in the 1993 study.
u (1) In the actual standard, no control technology was required for emissions from
14 oxychlorination vents at ethylene dichloride (EDC)/VC plants.'Applying "back calculation" for these
a emissions is inappropriate.
16 (2) In 1985, there were an estimated 8,000 fabrication plants which processed resins produced by
n PVC plants, thus resulting hi VC emissions, which were exempt from the VC NESHAP. They emit very
is small quantities of VC and back calculation is not appropriate.
19 (3) The 1993 study uses a baseline estimate of 18 residual cases from the NESHAP/ATERIS
20 data base. There is no evidence that these cases resulted only from emissions from PVC and EDC/VC
21 plants.
22 (4) The risk analysis performed for the October 21,1976 final VC regulation projected an
23 incidence reduction of 11 cases per year.
i4 In contrast, the PES study, using the "back calculation" method derived the following annual
25 incidence reductions:
26 1980-250 cases
27 . 1985-360 cases
28 1990-430 cases
»
30 The subsequent back calculation conducted in the ICF Re-analysis resulted in incidence reductions as
3i much as an order of magnitude higher than these.
32 Even considering the slightly different industrial output assumptions imposed by macroeconomic
33 modeling, such a stark contrast is difficult to explain except for a critically flawed approach. Growth in
34 activity and population nor other factors explain the difference in these two estimates. Given that the
35 same general methodology was used for all of the air toxic pollutant assessments as was used for the VC
36 NESHAP evaluation, there is reason to believe that cancer incidence results for the other air toxic
37 pollutants are also flawed.
366
-------
Appendix H: Air Toxics
Mobile Source HAP Exposure Reductions
EPA's Cancer Risk report estimated that approximately 60 percent of the total carcinogenic risk
posed by HAPs was attributed to mobile sources, with stationary sources contributing 15 percent and
area sources contributing the remaining 25 percent.274 The relative importance of mobile sources to total
HAP exposure was a significant motivation behind EPA's subsequent effort to examine exposures and
risks from mobile source HAPs.273 Although available analytical resources were severely limited, the
Project Team nevertheless decided it was necessary to perform at least an initial screening analysis to
estimate the differences in mobile source HAP exposures between the control and no-control scenarios
configured for the present study.
w Methodology
11 The approach used by ICF/SAI in conducting the mobile source HAP analysis closely followed
n the approach used in the EPA Motor Vehicle-Related Air Toxics Study (MVATS).276 Recognizing the
13 dearth of HAP ambient concentration and exposure data, both studies use carbon monoxide (CO)
14 concentrations as the basis for estimating mobile source HAP concentrations and exposures. An
is important difference between the two studies, however, is that the ICF/SAI study adjusted the estimated
is change in ambient CO concentrations to take account of background277 and non-mobile source278 CO
n emissions. The HAP exposure function used in the ICF/SAI analysis is summarized by the following
is equation:
E = ({C x A) - B) x S * M * (FQC * HAP) {64}
CO ^ f
19 where:
20 E = exposure to motor vehicle-emitted HAP
21 C = annual ambient CO concentration to annual CO exposure concentration conversion factor
22 A = county-level annual average ambient CO concentration
23 B background CO concentration
24 S = no-control to control scenario CO concentration adjustment factor (equals 1 for the
25 control scenario)
2« M total CO exposure to mobile source CO exposure conversion factor
274 Cancer Risk report, Page ES-12.
275 See US EPA/OAR/OMS, "Motor Vehicle-Related Air Toxics Study," EPA 420-R-93-005. April 1993.
276 ICF/SAI, "Retrospective Analysis of Inhalation Exposure to Hazardous Air Pollutants from Motor Vehicles," October 199S, p. 4.
277 Background CO is produced by the oxidation of biogenic hydrocarbons. See ICF/SAI, p. 7.
278 The EPA MVATS attributed all measured CO to motor vehicles, resulting in an overestimation of motor-vehicle contributions to CO
concentration changes. See ICF/SAI, p. 8. The MVATS assumption would also lead to a subsequent overestimation of changes in HAP
exposures.
367
-------
Appendix H: Air Toxics
i
2 '
3
4
s
6
7
8
9
10
11
12
13
14
15
16
voc
HAP
CO
VOC emissions by year, county, aijd scenario
VOC speciation factor by mobile source HAP
CO emissions by year, county, and scenario
Details of the derivation of each of the variables applied in the above equation are provided in
the ICF/SAI report. However, in essence, the calculation involves the following basic steps.
First, annual average, county-level CO ambient monitoring data are compiled from the EPA
Aerometric Information Retrieval System (AIRS) database. After adjusting for background and non-
mobile source contributions, these annual average ambient CO concentrations are converted to annual
average CO exposure concentrations. As hi the EPA MVATS, this conversion is made based on the
Hazardous Air Pollutant Exposure Model - Mobile Sources (HAPEM-MS) population exposure model,
which takes account of time spent in five indoor and outdoor microenvironments: indoors at home, other
indoor, in-vehicle, outdoors near roadway, and other outdoor.279 After adjusting for CO exposures
attributable to non-mobile sources of CO, the CO exposures are converted to exposures for each of the
mobile source HAPs based on available VOC speciation data and the ratio of co-located VOC and CO
emissions.280 These calculations are repeated for the no-control scenario after adjusting for differences in
CO ambient concentrations for each target year and for differences in fuel composition.
n
Results
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
By 1990, CAA controls
resulted in significant reductions in
exposure to motor vehicle HAPs.
Figure 55 summarizes the nationwide
annual average exposure levels, in
micrograms per cubic meter, for each
of the five HAPs analyzed under the
control and no-control scenarios.
Additional detailed results, including
breakdown by urban versus rural
environments and comparisons with the
EPA MVATS estimates, are provided
in the ICF/SAI report.
Analytical resources to carry
forward these exposure estimates to
derive estimates of the changes in ^^^^^^^^^^^
motor vehicle HAP-related adverse
effects attributable to historical CAA programs were not available.
Figure 55. National Annual Average Motor Vehicle HAP
Exposures (ug/m3).
a Control
•No-Control
Benzene AceuUehyde DieteiPM
Formaldehyde U-Buudiene
279 See ICF/SAI, p. 3.
280 The same HAP emission fractions used in the EPA MVATS were used herein, except for diesel PM which is not proportional to VOC
emissions. Instead, diesel PM emission factors were developed using year-specific PARTS diesel PM emission factors and VMT estimates for
diesel-powered vehicles.
368
-------
Appendix H: Air Toxics
Non-Cancer Health Effects
2 Broad gaps exist in the current state of knowledge about the quantifiable effects of air toxics
3 exposure. This is particularly true for a wide range of health effects such as tumors, abnormal
•4 development, birth defects, neurological impairment, or reproductive impairment, etc. For example, the
s EPA's Non-Cancer Study281 found that ambient concentrations for a substantial number of monitored and
6 modeled HAPs exceeded one or more health benchmarks.282 However no accepted methodology exists to
7 quantify the effects of such exceedences. More data on health effects is needed for a broad range of
s chemicals.
10 Through the 1970s and 1980s, the adverse effects of toxic pollution on the Great Lakes became
// clear and undeniable. Over the same time period, scientists began collecting a convincing body of
n evidence that toxic chemicals released to the air can travel long distances and be deposited on land or
13 water far from the original sources. An example of this evidence is the presence of such contaminants as
14 PCBs, toxaphene, and other pesticides in fish in Lake Siskiwit, a lake on an island on upper Lake
is Superior, which has no waterborne sources of pollution. Toxaphene, a pesticide used primarily in the
16 southeastern U.S. cotton belt, has been found as far away as the Arctic, with a decreasing air
n concentration gradient from the southeast toward the Great Lakes and the north Atlantic regions.
is Similarly, a growing body of evidence showed that pollutants that were persistent (do not easily
19 break down) and bioaccumulating (not significantly eliminated from the body) were magnifying up the
20 food chain, such that levels in top predator fish contained levels up to millions of times greater than the
21 harmless levels in the water. As such, those who ate those large fish, such as humans, eagles, mink, and
22 beluga whales could receive very high exposures to the pollutants. Wildlife were beginning to show
23 adverse effects in the wild, that could be duplicated in the lab. In the Great Lakes, such chemicals as
24 PCBs, mercury, dieldrin, hexachlorobenzene, Lindane, lead compounds, cadmium compounds,
25 DDT/DDE, and others are of significant concern. In other places in the country, similar effects are being
26 experienced, especially with mercury, which is transported primarily by air, but exposure to which is
27 primarily through contaminated fish. It was this kind of information about DDT and toxaphene that led
28 to their being banned in the U.S. under FIFRA.
29 While ecological and economical sciences are not yet sufficiently advanced to support the kind
30 of comprehensive, quantitative evaluation of benefits needed for the present study, selected local and
31 regional scale adverse ecological effects of HAPs, and their adverse consequences for human health and
"' U.S. Environmental Protection Agency, "Toxic Air Pollutants and Noncancer Risks: Screening Studies," External Review Draft,
September, 1990.
282 Relevant benchmarks include Acceptable Daily Intake (ADI), the estimate of daily exposure at which adverse health effects are
unlikely; and Lowest Observed Actual Effect Level (LOAEL), which is the lowest exposure level at which significant adverse health effects are
observed.
369
-------
Appendix H: Air Toxics
i welfare, can and have been surveyed. In May 1994, the EPA issued its first "Report to Congress on
•2 Deposition of Air Pollutants to the Great Waters."283 The Great Waters Report examined the pollutants
3 contributing to adverse ecological effects, the potential significance of the contribution to pollutant
4 loadings from deposition of airborne pollutants, and the potential adverse effects associated with these
5 pollutant loadings. Key HAPs identified in the Great Waters Report include PCBs, mercury, dioxins,
6 and other heavy metals and toxic organics.
7 Of particular relevance to the present assessment, the Great Waters Report demonstrated the
a significance of transport and transformation of HAPs through food webs, leading to increased toxicity
. 9 and biomagnification. A prime example of adverse transport and transformation is mercury.
10 Transformation from inorganic to methylated forms significantly increases the toxic effects of mercury
11 in ecosystems. A prime example of biomagnification is PCBs. As noted in the Great Waters Report:
12 ; "Pollutants of concern [such as PCBs] accumulate in body tissues and magnify up the
13 food \veb, with each level accumulating the toxics from its diet and passing the burden
14 along to the animal in the next level of the food -web. Top consumers in the food web,
is usually consumers of large fish, may accumulate chemical concentrations many millions
16 of times greater than the concentrations present in the water...High risk groups—include
i? breast-feeding mothers because breast-fed babies continue to accumulate [pollutants]
is from their mothers after birth. For example, they can have PCS levels four times higher
19 than their mothers after six to nine months of breast feeding.ms4
20 Because of the risk of significant exposure to infants and other high-risk groups, such as "sport
21 anglers, Native Americans, and the urban poor,"2*5 a substantial number of fish consumption advisories
22 have been issued in recent years. Current fish advisories for the Great Lakes alone include widespread
23 advisories for PCB's, chlordane, mercury and others, cautioning that nursing mothers, pregnant women,
24 women who anticipate bearing children, female children of any age and male children age IS and under
25 not eat certain high-food chain fish species. It should be noted as well that 40 states have issued mercury
26 advisories in some freshwater bodies, and nine states have issued mercury advisories for every
27 freshwater waterbody in the state (these states are Maine, New Hampshire, Vermont, Massachusetts,
za New York, New Jersey, Missouri, Michigan, and Florida).
29 There is little evidence indicating that the CAA had much beneficial effect on air toxic
x deposition to water bodies. Since the early NESHAPs were based on direct inhalation, primarily cancer
31 effects close to a plant, they did not address the issue of cumulative effects of persistent pollutants far
32 from the source. It was for this reason that Section 112(m) was included in the 1990 CAA Amendments,
33 with requirements to study and document the atmospheric contribution of water pollutants, the adverse
34 human health and environmental effects resulting and the sources that should be controlled to prevent
3i adverse effects, and additionally, to promulgate regulations to prevent adverse effects.
aj USEPA/OAR/OAQPS, "Deposition of Air Pollutants to the Great Waters, First Report to Congress," EPA-153/R-93-055, May 1994.
M EPA-453/R-93-055, May 1994, p. ix.
215 EPA-453/R-93-055, May 1994, p. x.
370
-------
Appendix H: Air Toxics
21
22
23
24
25
26
Conclusions -- Research Needs
As has been demonstrated, there are broad gaps in the current state of knowledge about the
iable effects of air toxics exposure for a wide r
The following discussion outlines areas in w
adequately quantify the benefits of air toxics control.
3 quantifiable effects of air toxics exposure for a wide range of both human health and environmental
4 effects. The following discussion outlines areas in which further research is needed in order to
6 Health Effects
7 • Develop health effects data on pollutants for which limited or no data currently exists. Such
s studies should be focused on pollutants with a relatively high probability of exposure and/or
9 potential adverse health effects.
10 • Understand mechanism of action of pollutants, for example through pharmacokinetic modeling.
/; This will allow for a more accurate assessment of the effects of these pollutants on humans.
n • Conduct research on factors that affect variations in susceptibility of human populations and
u determine the distribution of these factors in the U.S.
14 • Conduct research to better understand interactive effects of multiple pollutant exposures.
is • Develop methodologies to derive alternative estimates of human cancer risk from existing upper-
16 bound methods.
n • Acquire data and develop dose-response relationships for critical noncancer effects such as
is developmental, neurotoxic, mutagenic, respiratory and other effects. In particular, design
19 methodology to quantify effects of exposures above health benchmarks.
20 • Acquire data and develop methods to estimate effects from acute exposure.
Exposure Assessment
• Expand data collection efforts: pre- and post-control emissions; HAP speciation; facilities
location; facility parameters (stack heights, distances from stacks to fencelines, etc.)
• Develop more comprehensive exposure models which incorporate activity patterns, indirect
exposures, total body burden, ratios of time spent indoors to outdoors.
• Continue to refine uncertainty analysis methods.
371
-------
Appendix H: Air Toxics
16
2 • Reliable estimates/measures of the levels of persistent bioaccumulating toxics in different media
3 (air, water column, soils and sediments)
4 • Work to correlate levels of persistent bioaccumulating toxics with exposures, biota
5 concentrations/accumulation, and adverse effects, especially subtle effects such as wasting,
6
behavioral effects, and developmental effects.
7 • Criteria for effects, such as a wildlife correlate to a RfD or dose-response curve. This work
a should be done to complement the mass balance efforts now being completed, which will model
9 source emissions to water column concentrations, then design research to predict effects on
10 living resources given those predicted levels.
// • Work to determine the effects of mixtures of persistent bioaccumulating toxic pollutants, and to
n determine cause-effect relationships of exposures over long periods of time.
n Studies to evaluate toxic effects in less well understood terrestrial systems such as: soil
14 organisms/invertebrates, food web effects, amphibian effects, effects on endangered species and
is phytotoxic effects.
Economic Valuation
n • Develop valuation estimates for endpoints for which inadequate estimates currently exist. These
is valuation estimates must be consistent with the kinds of damages expected.
19 • Initiate broad-scope economic valuation of air toxics program using survey techniques.
372
-------
Appendix Is Valuation of Human
Health and Welfare Effects of
Criteria Pollutants
4 For the Section 812 analysis of health benefits, valuation estimates were obtained from the
s literature and reported in dollars per case reduced for health effects, and dollars per unit of avoided
6 damage for welfare effects. Similar to estimates of physical effects provided by health studies, the
? monetary values of benefits were reported both in terms of mean values as well as a probability
s distribution of estimates. This permitted an evaluation of the uncertainty associated with the point
9 estimates. It is interesting to note that the distributions of benefit values varied by endpoint. For
10 example, while the estimate of the dollar value of a reduced mortality was lognormally distributed, the
// estimate for the value of a reduced case of acute bronchitis was uniformly distributed between a
12 minimum and maximum value.
13 For the welfare benefits analysis, the avoided losses were in many cases measured in monetary
14 terms directly. For agricultural benefits, however, the benefits associated with estimated changes in crop
a yields were evaluated with an agricultural sector model, AGSIM. This model incorporates agricultural
16 price, farm policy, and other data for each year. Based on expected yields, the model estimates the
n production levels for each crop, the economic benefits to consumers, and the economic benefits to
is producers associated with these production levels. To the extent that alternative exposure-response
19 relationships were available, a range of potential benefits was calculated.
Methods Used to Value Health Effects
20
21 This analysis used several sources to obtain dollar values (indicated in 1990$) for the morbidity
22 and mortality effects modeled in the 812 analysis. Memos from Industrial Economics (lEc) to EPA's
23 Office of Air and Radiation provide most of the values (Unsworth et al., 1992; Neumann and Unsworth,
24 1993; Unsworth and Neumann, 1993; Neumann et al., 1994). A few values, however, were obtained
25 from other sources of information. For example, work by Abt Associates provides cost of illness
26 estimates for several types of hospital admissions (Currier, 1995). In addition, a recent EPA draft report
27 provides unit values for several of the lead-induced health effects (USEPA, 1995).
' is When using information from the lEc memos, some estimates have been used directly, whereas
29 others have been combined in a variety of ways to relate to the health effects modeled by the dose-
30 response functions. For example, lEc has developed estimates for shortness of breath and mortality,
31 which directly relate to health effects modeled by dose-response functions. However, in several cases,
32 dose-response functions have been developed for a variety of groups of respiratory symptoms. In these
33 instances, the analysis makes judgements about how to combine individual economic estimates
34 recommended by lEc. In other cases, the dose-response study does not clearly define the modeled health
35 effect, and thus, the analysis makes a judgement about the valuations to apply to the study. In cases
373
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
i where the current analysis combined economic estimates from lEc or applied the estimates to health
effects other than those listed specifically in the lEc memos, the analysis generally parallels the methods
3 used by lEc. For example, the current analysis estimates midpoint estimates in a manner according to
4 lEc recommendations when central values are not given by lEc (i.e., the analysis uses the midpoint of the
5 high and low valuations to represent central estimates).
6 Several points should be made about the choice of economic valuations fora few of the
7 individual endpoints. First, although economic values for hospital admissions were closely matched to
s the type of hospital admissions included in the concentration-response studies (e.g., COPD, pneumonia),
9 some studies included age groups for which economic values were not available in Currier (1995). •
10 These cases are noted in the table below. Second, the low and high estimates for both upper and lower
/; respiratory illness are based on the assumption that at the low end, only one symptom would occur, and
12 at the high end, a group of relevant symptoms would occur. Values for each individual symptom are
n midpoint estimates recommended by lEc.
14 Third, Restricted Activity Days (RADs) and Respiratory Restricted Activity Days (RRADs)
is include days of restricted activity that are more severe than those for Minor Restricted Activity Days
i6 (MRADs). Further, RADs include Work Loss Days, creating the potential for double counting.
n MRADs, by the definition in the study providing the dose response function, do not include WLDs.
is Therefore, the current analysis develops two alternative composite measures of loss-of days:
19
WorkLossDays + MRADs (65)
WorkLossDays + (*MDflKt4tHet - WLD^^^) x MRAD unitvalue (66)
The valuations for WLD and MRADs are from Unsworth and Neumann, 1993.
20 Hasselblad et al. (1992), which provides the dose-response function for nitrous oxides, models
21 occurrences of "respiratory illness," which the authors do not further define. The current analysis uses
22 an economic value for cough and upper and lower respiratory illness developed by lEc to represent an
23 economic value for respiratory illness modeled by Hasselblad et al. (1992). Table 140 includes
24 additional notes specific to each of the valuations.
25 The 812 analysis uses the midpoint estimates listed in Table 140 for the point estimate of
26 benefits. For the sensitivity analysis, two types of sampling have been predominantly used when
27 endpoints have more than one valuation estimate. For example, in cases where lEc memoranda
28 recommended low, midpoint, and high values, the 812 analysis gave each of the recommended values
29 equal probabilities of occurring. When lEc memos recommended only high and low estimates, the 812
30 analysis assumed that all values between the high and low estimates would be equally likely. A few
31 exceptions to the use of these distributions include hospital admissions for pneumonia and mortality (as
n well as health endpoints that have values based on the mortality valuation). In the case of hospital
33 admissions for pneumonia, the analysis assumes that costs vary according to the normal "bell-shaped"
34 curve.
374
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Values for mortality vary according to a lognormal distribution. However, it was necessary to
truncate the upper end value so that estimates would not be extremely high. Thus, the highest possible
estimate of mortality to be used in the sensitivity analysis ($13.5 million) is the highest value from all
mortality-valuation studies evaluated by lEc. Several endpoints (e.g., hospital admissions for several
health effects) have only midpoint estimates of values, and thus, the analysis does not model uncertainty
and variability for these effects.
375
-------
8,
1
00
'3
"3
_y
o
o
a)
>
^^
'S
D
o
§1
i!
11.
u a
i
B
|
"n
'e
'B
J**
jjj
•S
•a
s
B
i
I
3
1 Midpoint Estimate |
8.
.1
i
1
u
«
.1
f
is
a
i
1
J
'S
•o
B
U
£
1
0
£
1
•
fi
£2
S s
2 o.
.e '
O -M O
• J 1 j= "i
•'1 -S §
E T 5 f
e
o
i
1 '
I
12
VO
eo
«n
VO
oo
5,986
Cost of Illness value for
hospital admissions for all
respiratory illnesses for all
ages (Currier, 1995)
^
i
^'5.
S S
^i
1?
3-8 8
111
.t
§ "B
llf
Da. D.
II
c §
II
f^
•2"§
||
n •§
•3 "2 ^->
•a a S
i£-i
>n
"V
r-"
in
"V
7,555
Cost of Illness value for
pneumonia hospital
admissions for all ages
(Currier, 1995)
,5
<*•*
v>
C
"i -S
'o. §
II
-------
Comments on
Unit Valuatio
Use for ozone,
PM-10 for studies
of admissions of
all individuals
•c
.a
a
g
I
ii
.
r
I
§
>
"c
a
Si
«
•n
u
—
o
I
_e
'S
a
•o
I
e
a
£
£
is
Erfij
o o
X U
-------
II
i!
<§ -
sill
• <-3
a- .« g
i
o
CL
,§
a.
1
3
.2
|
.a
Q
"g
•
&
.5
t
u
B
;=>
mate
.•a
B
2
*
Ed
c
'o
V
1
8.
73
i
•al
.f8"
3 M
jr i
21
Sx
f ?
i!
o
o~
oo
1 j£ S
S- oo *
Hi
o
o
.
.I §• 8 B 5 5-
•i*i
si S'*
•sl'"
i 1 o
6
.y
1
"S-
.s
a
.2
'S
.S
O (A
2 ll
B
1
•O
u
K
7!
e
^
1
5?^
'
•<2
ai
Hospital
congesti
issi
Hospital
ischemic
-------
9
>
•o
B
«
t
I
S
1
-S
•
§.
."O
S
a.
.9 i
I1
i*
1J
r-
55
o
.S
Si
tf -3
9 S
i -g
-a
•
i
TS
§
_
m o.
ro -a
a
si
O
en
s
•o
§
S.
-
o
CO
11
S8"
Is-
2.
00 ^
*^ •• *v
[
SJ fi
en
B
'S
•B
u
e
1
o •
I"
I
£
I
•5 a
-------
si
fl
*
o
i
£
.a
je H
00 r*
it
II
•O n
11
t
II
3.2
«
•g
|
.2
«
s
I
.•s
g
5
u
J
u
««
C
I
5
.2?
I
<«_
o
.S
-
.|
U
£
I
I.
e
-------
II
n
3
o
3 a.
s
i
D ft.
i
Jllll
> . •
• e
1
i
!§>
••ai
ti
ll.
•o •=*
if
8"
.s
ll
—1 u
fl
i|
ll
sM
a
00
Ic
ll-
•a
IS
at -r
00
8-S
s; P a
S sis
e
I
•o
U
e
£
M
t
.r
l
s
II
•s
-------
u "5
S >
!•-
U a
•'
^
g
e
el
"3
.•s
B
i
•c
^
B
|-
1
8
3
Distribution Type*
S
S
S
3
1
s
|
u
'S •
'5
u
B
'S
&
i
B
I
•o
B
fi
!s
£
e
i
}!!
1 ii i
•a- 8-
> .s a
!
t
>^
M
>
1
b.
<2
%
D
o
i
i
il
*"-
Discounted at 3%:
Discounted at 7%:
(USEPA, 1995)
Wj
e
1
a
1
<2
S
D
Midpoint estimate only
Soo
r~5
Lognormal distribution
(geometric mean = 3.6 1
and geometric standard
= 2.3 from Unsworth et
1992); distribution truni
the upper end value of 1
million
1
ft
!| 5
U 3 w
e El |
1 Ilil
^>
1
g
g
1
?
~
<2
8
D
Midpoint estimate only
1
681 /year/case
(USEPA, 1995)
S
E
.S
§
S
O
i
~"
<2
%
ta
§ u ?f S
•"• C3 2v
^3 O S *>
^ll-j-5 §
^ 59S 8>
-------
§ I
S a
E -
I
i
I
i
I
i
B
i
I
u
I
e
I
t
T
L.
e
-S B
i i
P
O
1
.1
*
3 §?
2 Si
I
i
t
oo
en
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
\
Results of Valuation of Health and Welfare
3 Tables 142 and 143 apply the unit valuations to the health and welfare effects presented in the
4 "Health Effects Appendix." As noted in the Health Effects Appendix, there are several estimates for
5 some health and welfare impacts, and thus there are several benefit estimates. These can be used to
6
derive a range of possible results.
7 Table 142 presents the results of the "50 km" model run. Observe that, for those health
s endpoints where the pollutant of concern is PM,0 only, annual benefits decline between 1985 and 1990.
9 Only two factors, besides the CR function itself, should affect the model's output: The difference in
10 ambient air quality between the two scenarios, and the affected population. Between 1985 and 1990,
/; difference in air quality between the scenarios increased, and the U.S. population increased. One should
n see, then, an unambiguous increase in economic benefits between 1985 and 1990. The decrease hi
13 benefits observed here is due exclusively to the decline in population residing hi counties with PMIO
14 monitors (i.e., due to the decrease hi the number of monitors).
is Table 143 presents the results of the "all U.S. population" model run (although, with the
16 exception of Pb, not all of the population is modeled, with 2 to 5 percent being excluded), and provides a
n more accurate depiction of {he pattern of economic benefits across years. The accuracy of the scale of
is incidence is less certain. These results are almost certainly more accurate than the "50 km" results, but
19 rely on the assumption that (for a portion of the population) distant air quality monitors provide a
20 reasonable estimate of local air quality conditions. Thus, the results presented here are somewhat
21 speculative. It is likely that the predicted economic benefits are overstated for that population group (20
22 to 30 percent of total population in the case of PM10) for which distant monitors are used. Conversely,
23 there is an implied zero economic benefit for that portion of the population (3 to 4 percent in the case of
24 PMjo) excluded from the analysis altogether, an understatement of economic impacts for that group.
25 The results for total monetized benefits were also calculated using a Monte Carlo simulation
26 model to capture the combined uncertainties in the selection of physical effects function and unit values
27 for effects reductions. The Monte Carlo methodology used in this analysis and the summary results are
28 presented hi Chapter 7. The results of the Monte Carlo simulations for each of the target years, plus the
29 total 1990 value of monetized benefits and costs derived from the Monte Carlo analysis, are presented in
so Table 141.
384
-------
Append ,: yalMion ofHumm Hea,,h ^ Welf
Total Benefits By Year
i?S^^
^^r^^
385
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Table 142. CriteriaPoUutwtfsHealthmdWelfenfBenefits-PopalatkmWithinSOKMofaMiHiitorM<3d(^ni
$10
$59
$598
Si
$12
$4
$61
$6,520
$6,510
$314
$353
$670
$461
$40
$373
$276 .
$1.779
$220
1980
$347,754
$299393
$285.191
$200,102
$185,925
$176!506
$155,212
$152^292
$149,180
$110319
$135,131
$107,507
$92,102
$48,652
$33^25
$2,960
$392
$372
$281
$170
$559
$209
$127
S14i
$27
$187
$1,437
$3
$52
$14
$U
$153
$440
$31,830
$31,821
$1>47
$4326
$2,864
S146
S625
$4301
$254
tl TflX
1985
$401,740
$336.858
$326,526
$228,433
$212,147
$207438
$201336.
$176,920
$173,574 -
$170,089
$125,561
$119,637
$113,829
$104.765
$94201
$59362
$10.970
$3347
S469
$421
$329
$202
$421
$319
$250
$145
iiiU
$31
$220
$1,653
$2
SliS
$17
$19,
$177
$123
$54333
$54323
$2,210
$2,788
$8,404
$$339
$294
$1,750
$716
$4,815
$424!
tl *71
1990
$383.234
$311,774
$309j378
$216,976
$201,587
$197324
$191364
$16l!?09
$119,554
$113,929
$111.475
$99,800
$118,460
$72,568
$14^34
$73«D
$7,470
$3,927
$460
$402
$319
$204
$397
$307
$248
$138
S^54
$29
$207
$1,599
$2
$181
$18
$25
$166
$492
$65,855
$65.844
$2,624
$3377
$10,597
$6,633
$346
$2.027
$960
$4,631
$384
386
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Table 143, Criteria Pollutants Health and Welfere Benefits - Extrapolated to Entire 48 State Population 0n millions of 1990 dollars per year).
Endpoint
MortaBty
Cor«a*ry Heart Disease* Strokes
Coronary Heart Disease
Strokes
Hospital Adnisswii
All Respiratory Admtssto&s
Pneumonia + COPD Admissions
Ischemic Heart Disease
Congestive Heart Failure
Respiratory fflness or Symptoaas
Caadren
Upper 4- Lower Resp. Hoes*
Acute Bronchitis'
A $292 J134
m
$1
$20 .
st
m
$209
$6310
$314 '
$353.
$70
$373
«*w
$32
, $239
S3
$73
$1*
$(4
$52*
$31^30
$31,821
$4326
$2.«64
it**
$«2$
im
$38 $42
$299 $££7
$2,076 . $2311
]
si«
3 $3
i $254
$22 $27
524 $31
1220 $240
«M4 • $705
$54333 - $65,855
$54323 $65^44
$2^?10 $2^24
$2,1*8 $33??
$8,484 $103*?
J5J39 $6.633
$364- $454
$1,75* $2#ff
$?t« - . $960
387
-------
Appendix I: Valuation of Human Health and Welfare Effects of Criteria Pollutants
Economic Valuation References
i
2 Currier, R. 1995. Abt Associates Inc. Memorandum to Jim Neumann, Industrial Economics, Inc. and
3. , Jim DeMocker, USEPA. Applicability of existing cost of illness values for use in the Section
4 812 benefits analysis and recommendations regarding valuing chronic cough using cost of illness
5 • • estimates. May 16.
6 Dickie, M. et al. 1991. Reconciling Averting Behavior and Contingent Valuation Benefit Estimates of
7 Reducing Symptoms of Ozone Exposure (draft), as cited in Neumann, J.E., Dickie, M.T., and
« R.E. Unsworth. 1994. Industrial Economics, Incorporated. Memorandum to Jim DeMocker,
9 U.S. EPA, Office of Air and JRadiation. March 31.
10 Jones-Lee, M.W., et al. "The Value of Safety: Result of a National Sample Survey." Economic Journal
// 95:49-72. March 1985.
n Neumann, J.E. and R.E. Unsworth. 1993. Industrial Economics, Inc. Memorandum to Jim DeMocker,
13 U.S. EPA, Office of Air and Radiation. Revisions to the proposed value of life methodology for
14 the Section 812 retrospective. May 3.
is Neumann, J.E., Dickie, M.T., and R.E. Unsworth. 1994. Industrial Economics, Incorporated.
16 Memorandum to Jim DeMocker, U.S. EPA, Office of Air and Radiation. Linkage between
n health effects estimation and morbidity valuation in the Section 812 analysis ~ draft valuation
is document. March 31.
19 U.S. Department of Commerce (USDOC), Economics and Statistics Administration. 1992. Statistical
20 Abstract of the United States, 1992: The National Data Book. 112th Edition, Washington, D.C.
21 U.S. Environmental Protection Agency (USEPA). 1995. The impact of the Clean Air Acton lead
22 pollution: emissions reductions, health effects, and economic benefits from 1970 to 1990 (draft).
23 Prepared by Abt Associates Inc. for Economic Analysis and Innovations Division, Office of
24 Policy, Planning and Evaluation, U.S. EPA. Contract No. 68-D2-0175, W.A. 3-05. January 19.
25 Unsworth, R.E., and J.E. Neumann. 1993. Industrial Economics, Inc. Memorandum to Jim DeMocker,
26 U.S. EPA, Office of Air and Radiation. Review of existing value of morbidity avoidance
27 estimates: draft valuation document. September 30.
28 Unsworth, R.E., Neumann, J., and W.E. Browne. 1992. Memorandum to Jim DeMocker, U.S. EPA,
29 Office of Air and Radiation. Review of existing value of life estimates: valuation document.
30 November 6.
31 Viscusi, Kip W. and W. Evans. "Utility Functions that are Dependent on One's Heath Status." American
32 Economic Review 1990.
388
-------
10
11
12
21
Appendix Js Future Directions
Research Implications
In virtually any benefit analysis of environmental issues, the state of scientific information limits
the degree of coverage possible and the confidence in benefit estimation. For most benefit categories,
further scientific research would allow for a better quantification of benefits. One of the major outcomes
of the retrospective analysis is a clear delineation of the major limitations in the scientific and
economics literature in carrying out an analysis of this scope. Often, a list of research needs is generated
in studies such as this, but there is no clear internal mechanism to address these needs. With this study
(and the ongoing Section 812 program), a process has been initiated where identified research needs are
to be integrated into EPA's overall extramural research grants program, administered by the Office of
Research and Development, particularly in the areas of air toxics and ecological effects. It is hoped that
the research projects that flow from this process will enable future analyses to be more comprehensive.
n In recent years, the EPA's extramural grants program has grown considerably. Several grants
14 that were awarded in fiscal year 1995 are directly relevant to future Section 812 analyses. Examples of
is topic areas include:
16 • Health effects of particulate matter
n • Biologically-based risk assessment models for cancer
is • Non-cancer risk assessment statistical methods
19 • Valuation of damages to forest ecosystems
20 • Valuation of mortality risk
Future Section B12 Analyses
22 This Retrospective Study of the benefits and costs of the Clean Air Act was developed pursuant
23 to Section 812 of the 1990 Clean Air Act Amendments. Section 812 also requires EPA to generate an
24 ongoing series of prospective studies of the benefits and costs of the Act, to be delivered as Reports to
25 Congress every two years.
26 Design of the first Section 812 Prospective Study commenced in 1993. The EPA Project Team
27 developed a list of key analytical design issues and a "strawman" analytical design reflecting notional
389
-------
Appendix J: Future Directions
i decisions with respect to each of these design issues.286 The analytical issues list and strawman design
2 was presented to the Science Advisory Board Clean Air Act Compliance Analysis Council (CAACAC),
3 the same SAB review group which has provided review of the Retrospective Study. Subsequently, the
4 EPA Project Team developed a preliminary design for the first Prospective Study. Due to resource
5 limitations, however, full-scale efforts to implement the first Prospective Study did not begin until 1995
6 when expenditures for Retrospective Study work began to decline as major components of that study
7 were completed.
a As for the Retrospective, the first Prospective Study is designed to contrast two alternative
9 scenarios; however, in the Prospective Study the comparison will be between a scenario which reflects
10 full implementation of the CAAA90 and a scenario which reflects continued implementation only of
// those air pollution control programs and standards which were in place as of passage of the CAAA90.
12 This means that the first Prospective Study will provide an estimate of the incremental benefits and costs
13 oftheCAAA90.
14 The first Prospective Study is being implemented in two phases. The first phase involves
is development of a screening study, and the second phase will involve a more detailed and refined analysis
16 which will culminate in the first Prospective Study Report to Congress. The screening study compiles
n currently available information on the costs and benefits of the implementation of CAAA90 programs,
is and is intended to assist the Project Team in the design of the more detailed analysis by providing
19 insights regarding the quality of available data sources and analytical models, and the relative importance
20 of specific program areas; emitting sectors; pollutants; health, welfare, and ecological endpoints; and
21 other important factors and variables.
22 In developing and implementing the Retrospective Study, the Project Team developed a number
23 of important modeling systems, analytical resources, and techniques which will be directly applicable
24 and useful for the ongoing series of section 812 Prospective Studies. Principal among these are the
25 Criteria Air Pollutant Modeling System (CAPMS) model developed to translate air quality profile data
26 into quantitative measures of physical outcomes; and the economic valuation models, coefficients, and
27 approaches developed to translate those physical outcomes to economic terms.
is The Project Team also learned valuable lessons regarding analytical approaches or methods
29 which were not as productive or useful. In particular, the Project Team plans not to perform
30 macroeconomic modeling as an integral part of the first Prospective Analysis. In fact, there are currently
a no plans to conduct a macroeconomic analysis at all. Essentially, the Project Team concluded, with
32 confirmation by the SAB CAACAC, that the substantial investment of time and resources necessary to
33 perform macroeconomic modeling would be better invested in developing high quality data on the likely
34 effects of the CAA on key emitting sectors, such as utilities, on-highway vehicles, refineries, etc. While
35 the intended products of a macroeconomic modeling exercise -such as overall effects on overall
36 productivity, aggregate employment effects, indirect economic effects- are of theoretical interest, the
37 practical results of such exercises in the context of evaluating environmental programs may be
38 disappointing.
216 Copies of the Prospective Study planning briefing materials are available from EPA.
390
-------
Appendix J: Future Directions
i First, the CAA has certainly had a significant affect on several industrial sectors. However, the
2 coarse structure of a model geared toward simulating effects across the entire economy requires crude
j and potentially inaccurate matching of these polluting sectors to macroeconomic model sectors. For
4 example, the J/W model used for the Retrospective Study has only 35 sectors, with Electric Utilities
s comprising a single sector. In reality, a well-structured analysis of the broader economic effects of the
6 CAA would provide for separate and distinct treatment of coal-fired utility plants, oil-fired plants, and so
/ on. Furthermore, the outputs of the macroeconomic model are too aggregated to provide useful and
s accurate input information for the sector-specific emission models used to project the emissions
9 consequences of CAA programs. Again, the critical flaw is the inability to project important details
10 about differential effects on utilities burning alternative fuels.
.n
12
13
14
The second critical problem with organizing a comprehensive analysis of the CAA around a
macroeconomic modeling approach is that the effect information produced by the macroeconomic model
is relatively unimportant with respect to answering the fundamental, target variable: "How do the overall
health, welfare, ecological, and economic benefits of Clean Air Act programs compare to the costs of
these programs? " The Project Team believes that any adverse effect, no matter how small in a global
context, should not be deemed "insignificant" if even one individual is seriously harmed. However, the
Retrospective Study results themselves have shown that, when analytical resources are limited, the
resources invested in the macroeconomic modeling would have been better spent to provide a more
complete and less uncertain assessment of the benefit side of the equation. Even on the cost side of the
equation, it is far more important to invest in developing accurate and reliable estimates of sector-
specific compliance strategies and the direct cost implications of those strategies. This will be even more
true in the Prospective Study context when the Project Team will be faced with forecasting compliance
strategies and costs rather than simply compiling survey data on actual, retrospective compliance
Avn^nHitiirAc
16
17
IS
19
20
21
22
23
24 expenditures.
25 The third and most important limitation of macroeconomic modeling analysis of environmental
26 programs is that, unlike the economic costs of protection programs, the economic benefits are not
27 allowed to propagate through the economy. For example, while productivity losses associated with
28 reduced capital investment due to environmental regulation are counted, the productivity gains resulting
29 from reduced pollution-related illness and absenteeism of workers are not counted. The resulting
30 imbalance in the treatment of regulatory consequences raises serious concerns about the value of the
31 macroeconomic modeling evaluation of environmental programs. In the future, macroeconomic models
32 which address this and other concerns may be developed; however, until such time EPA is likely to have
33 limited confidence in the value of macroeconomic analysis of even broad-scale environmental protection
34 programs.
35 Based on these findings and other factors, the design of the first* Prospective Study differs in
36 important ways from the Retrospective Study design. First, rather than relying on broad-scale,
37 hypothetical, macroeconomic model-based scenario development and analysis, the first Prospective
38 Study will make greater use of existing information from EPA and other analyses which assess
39 compliance strategies and costs, and the emission and air quality effects of those strategies. After
« developing as comprehensive a data set as possible of regulatory requirements, compliance strategies,
4i compliance costs, and emissions consequences, the data set will be reviewed, refined, and extended as
42 feasible and appropriate. In particular, a number of in-depth sector studies will be conducted to develop
391
-------
Appendix J: Future Directions
_ .
i up-to-date, detailed projections of the effects of new CAA requirements on key emitting sectors.
2 Potential sectors include, among others, utilities, refineries, and on-highway vehicles.
3 The first Prospective Study will also differ from the Retrospective Study in that analytical
4 resources will be directed toward development of a more complete assessment of benefits. The
5 deficiencies which prevailed in the Retrospective Study relating to assessment of the benefits of air
6 toxics control will be addressed. In addition, the Project Team will endeavor to provide a more complete
7 and effective assessment of the ecological effects of air pollution control.
392
-------
NEIL GOLDSCHMIDT
GOVERNOR
Department of Environmental Quality
811 SW SIXTH AVENUE, PORTLAND, OREGON 97204-1390 PHONE (503) 229-5696
Oddvar K. Aurdal, Chief
Grants Administration Section
U.S. Environmental Protection Agency
Region 10
1200 Sixth Avenue, MD-100
Seattle, WA 98101
Dear Mr. Aurdal:
September 29, 1989
CERTIFIED MAIL
Re: V-000332-01-C
Extension of
Project/Budget Periods
We are pleased to accept your approval extending the Project/Budget
Period to September 30, 1990.
As requested, we have reviewed the amendment and find it to be
satisfactory.
Enclosed are the signed original and two copies.
Sincerely,
Fred Hansen
Director
FH:c
C1057
Enclosures
cc: Kenneth Brooks, EPA
Michael Downs, BCD Administrator, DEQ
Judy Hatton, Business Office, DEQ
John Loewy, Acting MSD Administrator, DEQ
John Rist, Budget Office, DEQ
RECEIVED
GRA.VS ADMINISTRATION
-------
NEILGOLDSCHMIDT
GOVERNOR
Department of Environmental Quality
811 SW SIXTH AVENUE, PORTLAND, OREGON 97204-1390 PHONE (503) 229-5696
CERTIFIED MAIL NO. P 915 446 265
RETURN RECEIPT REQUESTED
Oddvar K. Aurdal, Chief
Grants Administration Section
U.S. Environmental Protection Agency
Region 10
1200 Sixth Avenue
Seattle, WA 98101
RE: PA/SI V-000332-01-B
SUPERFUND MULTISITE
Dear Mr. Aurdal:
Enclosed are a signed original and two copies of the amendment for the PA/SI
portion of the above referenced agreement.
Our records confirm the decrease of $11,651 is consistent with the Final
Financial Status Report (FSR) for the PA/SI portion of the award completed
03-31-89. It is understood the amendment applies only to the portion of the
award completed, and does not affect the remaining $360,596 in the
Superfund Multisite Agreement.
Sincerely,
Fred Hansen
Director
RN:m
PPD\SM2501
Enclosures
cc: John Rist: MSD, DEQ
Rich Nourse: ECD, DEQ
DEQ-1
-------
UNITE
ATES ENVIRONMENTAL PROTECTS
.lENCY
FEB 2 6 1990
Reply To
Attn Of:
MOrl
Fred Hansin, Director
Oregon Department Environmental Quality
811 Southwest Sixth Avenue
Portland. Oregon 97204
Re: V-000332-01-E
Superfund Multi-Site Agreement
Dear Mr. Hansen:
I am pleased to approve an Increase of
management assistance activities at the Jos
Previously we awarded $360,596; the revlsec
$390,596.
Please review the enclosed amendment.
please sign, date, and return an original <
mall to Grants Administration Section, MD-'
receipt.
If you wish to discuss this amendment ,
Robinson, EPA Project Officer at (206) 442-V^
i-i^o
f^
i'^5^7
EPA Form i32o.i (martial Larsep Robinson Millam
OFFICIAL FILE COPY
* U.S.GPO: 1988-0-206-471
-------
United States Region 10 • Alaska
Environmental Protection 1200 Sixth Avenue Idaho
Agency Seattle WA 98101 Oregon
Washington
July 31, 1990
MEMORANDUM
SUBJECT: Cooperative Agreement V-000332-01
FROM: Kirk Robinson
TO: Mel Rozier
The Oregon State Department of Environmental Quality has
requested that $8,820 be transferred from the Martin Marietta
component of the above referenced agreement to the Teledyne
component. A copy of their letter of request is attached. The
letter contains the proposed distribution of funds within the
Teledyne component of the agreement. The State has also
requested that the Martin Marietta component of the agreement be
administratively closed once the funds transfer is completed.
I have reviewed and approve the request.
Thank you for your attention to this matter.
-------
JUN25I990
Department of Environmental Quali^^^uNo BRANCH
811 SW SIXTH AVENUE, PORTLAND, OREGON 97204-1390 PHONE (503) 229-5696
June 22, 1990
Mr. Kirk Robinson, Program Officer
U. S. Environmental Protection Agency
Region X
1200 Sixth Avenue
Seattle, WA 98101
Re: Federal Cooperative Agreement
V-000332-01 - Martin Marietta
Dear Mr. Robinson:
The Oregon Department of Environmental Quality (DEQ) is currently
a signature party to the Consent Decree for the Martin Marietta
Superfund Project. The state is receiving direct reimbursement
from the company for its participation in project oversight, and
is no longer charging expenditures to the portion of the above
referenced cooperative agreement approved for Martin Marietta.
According to the most recent fiscal reports, there is an
unexpended balance of $8,820 remaining in the Martin Marietta
Management Assistance project fund, and no outstanding obliga-
tions. The Oregon DEQ requests the existing balance of $8,820
currently on deposit with Martin Marietta be transferred to the
Teledyne Wah Chang project which has the same activity under the
cooperative agreement between EPA and the State of Oregon. Please
transfer the Martin Marietta funds to the Teledyne project funds
in accordance with the following distribution.
PERSONNEL: $ 4, 700. 00
BENEFITS: 1,645.00
INDIRECT COSTS: 2.475.00
$ 8,820.00
Once the transfer of funds has been completed, please initiate the
administrative completion and closeout of the Martin Marietta
project.
-------
Kirk Robinson
Martin Marietta Company
June 22, 1990
Page 2
Thank you for your assistance and cooperation. If you have
questions, or require additional information, please contact Rich
Nourse at 503-229-6801.
Sincerely,
Jeff Christensen
Policy & Program Development
Environmental Cleanup Division
RN:m
PPD\SM3105
cc: Chip Humphrey, Oregon Operations, EPA
Judy Hatton, DEQ Business Office
-------
NEIL GOLDSCHMIDT
GOVERNOR
Environmental Quality Commission
811 SW SIXTH AVENUE, PORTLAND, OR 97204 PHONE (503) 229-5696
Oddvar K. Aurdal, Chief
Grants Administration Section
U.S. Environmental Protection Agency
Region 10
1200 Sixth Avenue, MD-100
Seattle, WA 98101
Dear Mr. Aurdal:
March 21, 1990
CERTIFIED MAIL
Re: V-000332-01-E
Superfund Multi-Site
Agreement
We are pleased to accept your increase for the above referenced
cooperative agreement, dated February 26, 1990, of $30,000 for
participation in the management assistance activities at the Joseph
Forest Products site.
As requested, we have reviewed the amendment and find it to be
satisfactory.
Enclosed are the signed original and two copies.
Sincerely,
DECEIVED
M.4R 2 ; V-?0
GRANTS ADMINiSi
Fred Hansen
Director
FH:c
C1119
Enclosures
cc: Kenneth Brooks, EPA
Kirk Robinson, Project Officer, EPA
Michael Downs, Environmental Cleanup Division, DEQ
Richard Nourse, Environmental Cleanup Division, DEQ
Judy Hatton, Business Office, DEQ
Peter Dalke, MSD Administrator, DEQ
John Rist, Budget Office, DEQ
Peggy Halferty, Budget Analyst, DEQ
DEQ-46
-------
United States
Environmental Protection
Agency
Region 10
1 200 Sixth Avenue
Seattle WA 981 01
Alaska
Idaho
Oregon
Washington
Reply To
Attn. Of:
MD-100
April 20, 1990
Judy Hatton, Accounting Manager
Oregon Department of Environmental Quality
811 SW Sixth Avenue
Portland, Oregon 97204
Dear Ms. Hatton:
According to our records, the unpaid balance for Superfund agreements
are as follows:
Unpaid Balances
thru No. 239
V-000332-01
Teledyne 7,350
Gould 27,050
United Chrome 12,100
Martin Marietta 8,820
Umatilla 7,833
Joseph Forest Products 30,000
Allied Plating 22,150
V-000350-02 CORE 0
V-000350-03 CORE 290,000
V-000363-01 SARA 50,583
V-000399-01 PA/SI 35,580
A final Financial Status Report FSR)
program is due by May 31, 1990.
Agreement
Expires
9/30/90
9/30/90
9/30/90
9/30/90
9/30/90
2/28/91
9/30/90
2/28/90
2/28/91
6/30/90
5/31/91
on the 2nd segment of the Core
An annual FSR is due by June 30, 1990 for the budget period ending
March 31, 1990 for all sites except Joseph Forest Products covered in the
multi-site agreement, the 3rd segment of the Core program, and the SARA and
PA/SI agreements.
Please contact Mel Rozier, of my staff, on (206) 442-2919 if our
records are not in agreement.
cc: Rich Nourse, ODEQ
Kirk Robinson, HW-113
Debbie Flood, HW-093
Sincerely,
Oddvar K. Aurdal, Chief
Grants Administration Section
------- |