C-l
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-91-023
November 1991
AIR
National Air Quality and
Emissions Trends Report,
1990
0 3 - CO P M I 0 0 3 - C 0 0 3 - P M Ifl te C 0 - P M I
03-CO-PM10B03-CO-PM10-N02 + S 0 2 - Pb
Counties with Non-Attainment Areas
-------
EPA-450/4-91-023
National Air Quality and
Emissions Trends Report,
1990
Technical Support Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
November 1991
U.S. Environment:; "^' "on
Region 5, Library ("'
77 West Jackson P-.
Chicago, IL 60604-...
-------
DISCLAIMER
This report has been reviewed by the Office of Air Quality Planning and Standards,
U. S. Environmental Protection Agency, and has been approved for publication.
Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use.
About the Cover: The map displays those counties within the contiguous U.S. that contain areas not
meeting ozone, carbon monoxide and/or paniculate matter National Ambient Air
Quality Standards (NAAQS). See Section 4 for information on these areas.
11
-------
PREFACE
This is the eighteenth annual report of air pollution trends issued by the U. S.
Environmental Protection Agency. The report is prepared by the Technical Support
Division and is directed toward both the technical air pollution audience and the
interested general public. The Division solicits comments on this report and welcomes
suggestions on our trend techniques, interpretations, conclusions, and methods of
presentation. Please forward any response to Dr. Thomas C. Curran, (MD-14) U. S.
Environmental Protection Agency, Technical Support Division, Research Triangle Park,
North Carolina 27711.
111
-------
CONTENTS
LIST OF FIGURES vii
LIST OF TABLES xi
1. EXECUTIVE SUMMARY 1-1
1.1 INTRODUCTION 1-1
1.2 SOME PERSPECTIVE 1-2
1.3 MAJOR FINDINGS 1-4
Particulate Matter 1-4
Sulfur Dioxide 1-6
Carbon Monoxide 1-8
Nitrogen Dioxide 1-10
Ozone 1-12
Lead 1-14
1.4 REFERENCES 1-16
2. INTRODUCTION 2-1
2.1 AIR QUALITY DATA BASE 2-3
2.2 TREND STATISTICS 2-4
2.3 REFERENCES 2-7
3. NATIONAL AND REGIONAL TRENDS IN NAAQS POLLUTANTS ... 3-1
3.1 TRENDS IN PARTICULATE MATTER 3-2
3.1.1 Long-term TSP Trends: 1981-90 3-3
3.1.2 Total Particulate Emission Trends 3-3
3.1.3 Recent TSP Trends: 1988-90 3-5
3.1.4 Recent PM-10 Air Quality: 1988-90 3-5
3.1.5 PM-10 Emission Trends 3-8
3.1.6 Visibility Trends 3-9
3.2 TRENDS IN SULFUR DIOXIDE 3-10
3.2.1 Long-term SO2 Trends: 1981-90 3-10
3.2.2 Recent SO2 Trends: 1988-90 3-15
3.3 TRENDS IN CARBON MONOXIDE 3-16
3.3.1 Long-term CO Trends: 1981-90 3-16
3.3.2 Recent CO Trends: 1988-1990 3-19
3.4 TRENDS IN NITROGEN DIOXIDE 3-20
3.4.1 Long-term NO2 Trends: 1981-90 3-20
3.4.2 Recent NO2 Trends: 1988-1990 3-23
3.5 TRENDS IN OZONE 3-24
3.5.1 Long-term O3 Trends: 1981-90 3-25
3.5.2 Recent O3 Trends: 1988-1990 3-28
3.6 TRENDS IN LEAD 3-29
-------
3.6.1 Long-term Pb Trends: 1981-90 3-29
3.6.2 Recent Pb Trends: 1988-90 3-33
3.7 REFERENCES 3-35
4. AIR QUALITY STATUS OF METROPOLITAN AREAS, 1990 4-1
4.1 Nonattainment Areas 4-1
4.2 Population Estimates For Counties Not Meeting NAAQS, 1990 ... 4-7
4.3 Air Quality Levels in Metropolitan Statistical Areas 4-10
4.3.1 Metropolitan Statistical Area Air Quality Maps, 1990 4-10
4.3.2 Metropolitan Statistical Area Air Quality Summary, 1990 . 4-11
4.4 REFERENCES 4-11
5. SELECTED METROPOLITAN AREA TRENDS 5-1
5.1 The Pollutant Standards Index 5-1
5.2 Summary of PSI Analyses 5-2
5.3 Description of Graphics 5-6
Atlanta, GA 5-8
Boston, MA 5-10
Chicago, IL 5-12
Dallas, TX 5-14
Denver, CO 5-16
Detroit, MI 5-18
Houston, TX 5-20
Kansas City, MO-KS 5-22
Los Angeles, CA 5-24
New York, NY 5-26
Philadelphia, PA 5-28
Pittsburgh, PA 5-30
San Francisco, CA 5-32
Seattle, WA 5-34
Washington, DC-MD-VA 5-36
VI
-------
LIST OF FIGURES
2-1. Sample illustration of use of confidence intervals to determine
statistically significant change 2-5
2-2. Illustration of plotting convention of boxplots 2-5
2-3. Ten Regions of the U.S. Environmental Protection Agency 2-6
3-1. Comparison of 1970 and 1990 emissions 3-1
3-2. National trend in the number of TSP and PM-10 monitoring locations,
1981-1990 3-2
3-3. Boxplot comparisons of trends in annual geometric mean total
suspended particulate concentrations at 1265 sites, 1981-1990 3-3
3-4. National trend in total particulate emissions, 1981-1990 3-4
3-5. Regional comparisons of the 1988,1989 and 1990 composite averages
of the geometric mean total suspended particulate concentrations 3-5
3-6. Boxplot comparisons of 1988, 1989,1990 PM-10 concentrations at 339
sites with 1990 PM-10 air quality at 979 sites 3-6
3-7. Boxplot comparisons of 24-hour PM-10 peak value statistics for 1990 at
979 sites 3-6
3-8. Regional comparisons of annual mean and 90th percentile of 24-hour
PM-10 concentrations for 1990 3-7
3-9. Regional changes in annual average and 90th percentile of 24-hour
PM-10 concentrations, 1988-1990 3-7
3-10. National trend in annual average sulfur dioxide concentration at both
NAMS and all sites with 95 percent confidence intervals, 1981-1990 . . . 3-10
3-11. National trend in the second-highest 24-hour sulfur dioxide
concentration at both NAMS and all sites with 95 percent confidence
intervals, 1981-1990 3-11
3-12. National trend in the estimated number of exceedances of the 24-hour
sulfur dioxide NAAQS at both NAMS and all sites with 95 percent
confidence intervals, 1981-1990 3-11
3-13. Boxplot comparisons of trends in annual mean sulfur dioxide
concentrations at 457 sites, 1981-1990 3-12
3-14. Boxplot comparisons of trends in second highest 24-hour average
sulfur dioxide concentrations at 452 sites, 1981-1990 3-12
3-15. National trend in sulfur oxides emissions, 1981-1990 3-13
3-16. Location of the 200 largest power plant emitters of sulfur oxides 3-14
3-17. Regional comparisons of the 1988,1989,1990 composite averages of
the annual average sulfur dioxide concentrations 3-15
3-18. National trend in the composite average of the second highest
nonoverlapping 8-hour average carbon monoxide concentration at
both NAMS and all sites with 95 percent confidence intervals,
1981-1990 3-17
3-19. Boxplot comparisons of trends in second highest nonoverlapping
8-hour average carbon monoxide concentrations at 301 sites, 1981-1990. . 3-17
vii
-------
3-20. National trend in the composite average of the estimated number of
exceedances of the 8-hour carbon monoxide NAAQS, at both NAMS
and all sites with 95 percent confidence intervals, 1981-1990 3-17
3-21. National trend in carbon monoxide emissions, 1981-1990 3-18
3-22. Comparison of trends in total national vehicle miles traveled and
national highway vehicle emissions, 1981-1990 3-19
3-23. Regional comparisons of 1988,1989,1990 composite averages of the
second highest nonoverlapping 8-hour average carbon monoxide
concentrations 3-19
3-24. National trend in the composite annual average nitrogen dioxide
concentration at both NAMS and all sites with 95 percent confidence
intervals, 1981-1990 3-20
3-25. Boxplot comparisons of trends in annual mean nitrogen dioxide
concentrations at 166 sites, 1981-1990 3-21
3-26. National trend in nitrogen oxides emissions, 1981-1990 3-22
3-27. Regional comparisons of 1988,1989,1990 composite averages of the
annual mean nitrogen dioxide concentrations 3-23
3-28. National trend in the composite average of the second highest
maximum 1-hour ozone concentration at both NAMS and all sites
with 95 percent confidence intervals, 1981-1990 3-24
3-29. Boxplot comparisons of trends in annual second highest daily
maximum 1-hour ozone concentration at 471 sites, 1981-1990 3-25
3-30. National trend in the estimated number of daily exceedances of the
ozone NAAQS in the ozone season at both NAMS and all sites with
95 percent confidence intervals, 1981-1990 3-26
3-31. National trend in volatile organic compound emissions, 1981-1990 3-27
3-32. Regional comparisons of the 1988,1989,1990 composite averages of
the second-highest daily 1-hour ozone concentrations 3-28
3-33. National trend in the composite average of the maximum quarterly
average lead concentration at both NAMS and all sites with 95 percent
confidence intervals, 1981-1990 3-30
3-34. Comparison of national trend in the composite average of the
maximum quarterly average lead concentrations at urban and point-
source oriented sites, 1981-1990 3-30
3-35. Boxplot comparisons of trends in maximum quarterly average lead
concentrations at 202 sites, 1981-1990. 3-31
3-36. National trend in lead emissions, 1981-1990 3-32
3-37. National trend in emissions of lead excluding transportation sources,
1981-1990 3-33
3-38. Regional comparison of the 1988,1989,1990 composite average of the
maximum quarterly average lead concentrations 3-33
4-1. Nonattainment areas for ozone 4-2
4-2. Nonattainment areas for carbon monoxide 4-3
4-3. Nonattainment areas for particulate matter 4-4
viii
-------
4-4. Nonattainment areas for sulfur dioxide 4-5
4-5. Nonattainment areas for lead 4-6
4-6. Number of persons living in counties with air quality levels above the
primary national ambient air quality standards in 1990 (based on 1987
population data) 4-7
4-7. Midwest region on July 17,1987 4-8
4-8. Northeast region on July 7,1988 4-8
4-9. Northeast region at 1400 EST on July 8,1988 4-9
4-10. Northeast region at 2300 EST on July 8,1988 4-9
4-11. United States map of the highest second maximum 24-hour average
PM-10 concentration by MSA, 1990 4-12
4-12. United States map of the highest annual arithmetic mean PM-10
concentration by MSA, 1990 4-13
4-13. United States map of the highest annual arithmetic mean sulfur
dioxide concentration by MSA, 1990 4-14
4-14. United States map of the highest second maximum 24-hour average
sulfur dioxide concentration by MSA, 1990 4-15
4-15. United States map of the highest second maximum nonoverlapping 8-
hour average carbon monoxide concentration by MSA, 1990 4-16
4-16. United States map of the highest annual arithmetic mean nitrogen
dioxide concentration by MSA, 1990 4-17
4-17. United States map of the highest second daily maximum 1-hour
average ozone concentration by MSA, 1990 4-18
4-18. United States map of the highest maximum quarterly average lead
concentration by MSA, 1990 4-19
5-1. PSI days > 100 in 1988,1989 and 1990 using all sites 5-5
IX
-------
LIST OF TABLES
2-1. National Ambient Air Quality Standards (NAAQS) in Effect in
1990 2-2
2-2. Number of Air Quality Trend Sites, 1981-90 and 1988-90 2-4
3-1. National Total Participate Emission Estimates, 1981-1990 3-4
3-2. National PM-10 Emission Estimates, 1985-1990 3-8
3-3. National PM-10 Fugitive Emission Estimates, 1985-1990 3-9
3-4. National Sulfur Oxides Emission Estimates, 1981-1990 3-13
3-5. National Carbon Monoxide Emission Estimates, 1981-1990 3-18
3-6. National Nitrogen Oxides Emission Estimates, 1981-1990 3-22
3-7. National Volatile Organic Compound Emission Estimates, 1981-1990 .. . 3-27
3-8. National Lead Emission Estimates, 1981-1990 3-32
4-1. Nonattainment Areas for NAAQS Pollutants as of October 1991 4-1
4-2. Population Distribution of Metropolitan Statistical Areas Based on
1987 Population Estimates 4-10
4-3. 1990 Metropolitan Statistical Area Air Quality Factbook Peak Statistics
for Selected Pollutants by MSA 4-20
5-1. PSI Categories and Health Effect Descriptor Words 5-1
5-2. Number of PSI Days Greater than 100 at Trend Sites, 1981-90, and All
Sites in 1990 5-3
5-3. (Ozone Only) Number of PSI Days Greater Than Trend Sites, 1981-90,
and All Sites in 1990 5-4
5-4. Number of Trend Monitoring Sites for the 15 Urban Area Analyses .... 5-6
XI
-------
NATIONAL AIR QUALITY AND EMISSIONS TRENDS REPORT. 1990
1. EXECUTIVE SUMMARY
1.1 INTRODUCTION
This is the eighteenth annual report1"17 documenting air pollution trends in the
United States for those pollutants for which the U.S. Environmental Protection Agency
(EPA) has established National Ambient Air Quality Standards (NAAQS). EPA set
these standards to protect public health and welfare. There are two types of NAAQS,
primary and secondary. Primary standards are designed to protect public health, while
secondary standards protect public welfare, such as effects of air pollution on
vegetation, materials and visibility.
This report focuses on comparisons with the primary standards in effect in 1990
to examine changes in air pollution levels over time, and to summarize current air
pollution status. There are six pollutants that have NAAQS: particulate matter (formerly
as total suspended particulate (TSP) and now as PM-10 which emphasizes the smaller
particles), sulfur dioxide (SO2), carbon monoxide (CO), nitrogen dioxide (NO2), ozone
(O3) and lead (Pb). It is important to note that the discussions of ozone in this report
refer to ground level, or tropospheric, ozone and not to stratospheric ozone. Ozone in
the stratosphere, miles above the earth, is a beneficial screen from the sun's ultraviolet
rays. Ozone at ground level, in the air we breathe, is a health and environmental
concern and is the primary ingredient of what is commonly called smog.
The report tracks two kinds of trends: air concentrations, based on actual direct
measurements of pollutant concentrations at selected sites throughout the country; and
emissions, which are based upon the best available engineering calculations. It also
provides estimates of the total tonnage of these pollutants released into the air annually.
Chapter 4 of this report includes a detailed listing of selected 1990 air quality summary
statistics for every metropolitan statistical area (MSA) in the nation and maps
highlighting the largest MSAs. Chapter 5 presents 1981-90 trends for 15 cities and
includes maps highlighting the locations of the monitoring networks.
A landmark event for air pollution control in the United States occurred in
November 1990, with the passage of the Clean Air Act Amendments. While it is much
too early for this Act to have influenced air pollution trends, some provisions are
discussed briefly in this report because of the major role that the Act will play in
dictating future air quality and emission trends in the U.S.
1-1
-------
1.2 SOME PERSPECTIVE
A 10-year time period is convenient for considering ambient air pollution trends
because monitoring networks underwent many changes around 1980. However, it is
important not to overlook some of the earlier control efforts in the air pollution field. Emission
estimates are useful in examining longer term trends. Between 1970 and 1990, lead clearly
shows the most impressive decrease (-97 percent) but improvements are also seen for total
paniculate (-59 percent), sulfur oxides (-25 percent), carbon monoxide (-41 percent), and
volatile organic compounds (-31 percent). Only nitrogen oxides did not show improvement
with emissions estimated to have increased 6 percent, due primarily to increased fuel
combustion by stationary sources and motor vehicles. It is also important to realize that many
of these reductions occurred even in the face of growth of emissions sources. More detailed
information is contained in a companion report.18
120
COMPARISON OF 1970 AND 1990 EMISSIONS
MILLION METRIC TONS/YEAR
THOUSAND
METRIC TONS/YEAR
100 -
80 -
60 -
40 -
20 A
250
200
150
100
50
NOx
VOC
LEAD
1970 D1990
While it is important to recognize that progress has been made, it is also important not
to lose sight of the magnitude of the air pollution problem that still remains. About 74 million
people in the U.S. reside in counties which did not meet at least one air quality standard
based upon data for 1990. The 63 million people living in counties that exceeded the ozone
standard in 1990 is 4 million fewer than in 1989. The 1990 estimates for carbon monoxide
and PM-10 are substantially lower than population totals for 1989. These statistics, and
associated qualifiers and limitations, are discussed in Chapter 4. These population estimates
are based only upon a single year of data, 1990, and only consider counties with monitoring
1-2
-------
data for that pollutant. As noted in Chapter 4, there are other approaches that would yield
different numbers. For example, it is estimated that 140 million people live in ozone
nonattainment areas based upon EPA's October 1991 designations. This is because ozone
nonattainment decisions are based upon three years of data, rather than just one, to reflect a
broader range of meteorological conditions. Also, nonattainment boundaries may consider
other air quality related information, such as emission inventories and modeling, and may
extend beyond those counties with monitoring data to more fully characterize the ozone
problem and to facilitate the development of an adequate control strategy.
People in counties with 1990 air quality above
primary National Ambient Air Quality Standards
pollutant
PM10
0 20
Note: Based on 1987 county population data.
40 60
millions of people
80
100
Finally, it should be recognized that this report focuses on those six pollutants that
have National Ambient Air Quality Standards. There are other pollutants of concern.
According to industry estimates, more than 2.4 billion pounds of toxic pollutants were emitted
into the atmosphere in 1988. They are chemicals known or suspected of causing cancer or
other serious health effects (e.g. reproductive effects). Control programs for the NAAQS
pollutants can be expected to reduce these air toxic emissions by controlling particulates,
volatile organic compounds and nitrogen oxides. However, Title 111 of the Clean Air Act
Amendments of 1990 provided specific new tools to address routine and accidental releases
of these pollutants. The statute established an initial list of 190 hazardous air pollutants.
Using this list, EPA will publish a list of the source categories for which emission standards
will be developed. EPA will issue standards for each listed source category, requiring the
maximum degree of emissions reduction that has been demonstrated to be achievable.
These are commonly referred to as maximum achievable control technology (MACT)
standards. The Act also includes programs to help prevent the accidental release of
hazardous chemicals.
1-3
-------
1.3 MAJOR FINDINGS
PARTIQULATi MATTER
AIR CONCENTRATIONS: Total Suspended Particulates (TSP)
and PM-10
1982-90*: 3 percent decrease TSP (based on geometric mean at 1265 sites)
* 1981 data affected by a change in filters
1989-90: 3 percent decrease TSP (based on geometric mean at 734 sites)
8 percent decrease PM-10 (based on arithmetic mean at 339 sites)
EMISSIONS : Total Particulates (TP) and PM-10
1981-90: 6 percent decrease (TP)
(Note: 9-year 1982-90 change was 6 percent increase)
1989-90: 4 percent increase (TP); 5 percent increase (PM-10)
OVERVIEW
Trends TP emissions from historically inventoried sources have been reduced 59
percent since 1970. During the 1980's, TSP air quality levels improved 3 percent. In
1987, EPA replaced the earlier TSP standard with a PM-10 standard. (PM-10 focuses
on the smaller particles likely to be responsible for adverse health effects because of
their ability to reach the lower regions of the respiratory tract.) Ambient monitoring
networks have recently been revised to measure PM-10 rather than TSP. Although
PM-10 trends data are limited, ambient levels decreased 11 percent between 1988 and
1990. The PM-10 portion of TP emissions is estimated to have increased 7 percent
since 1985 due to increases from transportation sources and forest fires. Nationally,
fugitive sources provide 6-8 times more tonnage of PM-10 emissions than historically
inventoried sources.
Status In October 1991, EPA designated 70 areas as nonattainrnent for PM-10.
National average TSP levels in 1990 were the lowest of the past decade. Comparing
1989 and 1990, most of the country experienced an increase in precipitation and a
decrease in TSP and PM-10.
1990 Clean Air Act The Act focuses attention on nonattainrnent of PM-10 health
based standards. The Acid Rain provisions of the Act address visibility impairment
caused by fine (<2.5 micrometer) particles.
1-4
-------
TSP TREND, 1981-1990
(ANNUAL GEOMETRIC MEAN)
TP EMISSIONS TREND
(1981 vs. 1990)
100-
CONCENTRATION, UG/MJ
80-
60-
40-
20-
1265 SITES
FORMER NAAQS
90% ot sites have lower
.^etf Meafi 'Concentrations
than 1his line ;
10% ot sites have lower
Geo Mean concentrations
than this line
10
MILLION METRIC TONS PER YEAR
7.5
I I I I I T I I
81 82 83 84 85 86 87 88 89 90
1981
^Transportation
THIndustrial
Processes
1990
Fuel
Combustion
Solid Waste
&Misc.
PM EFFECTS
on studies of human populations exposed to high concentrations of
particles (often in the presence; of sulfur dioxide), and laboratory studies of
animals and humans, the major effects of concern for human health include
effects on breathing and respiratory symptoms, aggravation of existing
respiratory and cardiovascular disease, alterations in the body's defense
Sysbms against foreign materials, damage to lung tissue, carcinogenesis and
premature mortality, The major subgroups of the population that appear likely
to be most sensitive to the effects of particulate matter include individuals with
chronic obstructive pulmonary or cardiovascular disease, individuals with ;
influenza, asthmatics, the elderly and children. Particuiate matter causes
damage to materials, soiling and is a major cause of substantial visibility
impairment in many parts of the U.S.
1-5
-------
SULFUR DIOXIDE (SO2)
AIR CONCENTRATIONS
1981-90: 24 percent decrease (arithmetic mean at 457 sites)
30 percent decrease (24-hour second high at 452 sites)
87 percent decrease (24-hour exceedances at 452 sites)
1989-90: 7 percent decrease (arithmetic mean at 552 sites)
EMISSIONS: SOX
1981-90: 6 percent decrease
1989-90: 2 percent increase
OVERVIEW
Trends SOX emissions decreased 25 percent since 1970. During the 1980's,
emissions improved 6 percent while average air quality improved by 24 percent. This
difference occurs because the historical ambient monitoring networks were population-
oriented while the major emission sources tend to be in less populated areas. The
exceedance trend is dominated by source oriented sites. The 1981-90 decrease in
emissions reflects reductions at coal-fired power plants. The 1989-90 emissions
increase is due to increases from fuel combustion.
Status Almost all monitors in U.S. urban areas meet EPA's ambient SO2 standards.
Dispersion models are commonly used to assess ambient SO2 problems around point
sources because it is frequently impractical to operate enough monitors to provide a
complete air quality assessment. Currently, there are 50 areas designated
nonattainment for SO2. Current concerns focus on major emitters, total atmospheric
loadings and the possible need for a shorter-term (i.e. 1-hour) standard. Seventy
percent of all national SOX emissions are generated by electric utilities (92% of which
come from coal fired power plants).
1990 Clean Air Act The Acid Rain provisions include a goal of reducing SOX
emissions by 10 million tons relative to 1980 levels. The focus in this control program
is innovative market-based emission allowances which will provide affected sources
flexibility in meeting the mandated emission reductions. This is EPA's first large-scale
regulatory use of market-based incentives. These reductions will improve visibility in
the East by substantially reducing SOX emissions. These emissions are transformed
into fine acid sulfate aerosol, the main cause of regional visibility impairment.
1-6
-------
SO2 TREND, 1981-1990
(ANNUAL ARITHMETIC MEAN)
CONCENTRATION, PPM
SOX EMISSIONS TREND
(1981 vs. 1990)
MILLION METRIC TONS PER YEAR
0.04
0.03-
0.02-
0.01-
0.00
457 SITES
NAAQS
90% of sites have tower
Arith Mean concentrations
than this line
Average tor aS sites
10%irf$iles have tower;
Arith Mean concentrations
than this line
1981
\\ \ \ I I I I
81 82 83 84 85 86 87 88 89 90
[Transportation
Industrial
Processes
1990
Fuel
Combustion
Solid Waste
&Misc.
SO2 EFFECTS
The major health effects of concern associated with high exposures to sulfur
dioxide include effects on breathing, respiratory illness and symptoms,
alterations in the lung's defenses, aggravation of existing respiratory and
cardiovascular disease, and mortality. The major subgroups of the population
most sensitive to sulfur dioxide Include asthmatics and individuals with chronic
lung disease (such as bronchitis or emphysema) or cardiovascular disease.
Children and the efderly may also be sensitive. Sulfur dioxide produces foliar
damage on trees and agricultural crops. It and nitrogen oxides are major
precursors to acidic deposition (acid rain), which is associated with a number of
effects including acidification of lakes and streams, accelerated corrosion of
buildings and monuments and visibility impairment.
1-7
-------
CARBON MONOXIDE (00)
AIR CONCENTRATIONS
1981-90: 29 percent decrease (8-hour second high at 301 sites)
87 percent decrease (8-hour exceedances at 301 sites)
1989-90: 8 percent decrease (8-hour second high at 359 sites)
EMISSIONS
1981-90: 22 percent decrease
1989-90: less than 1 percent decrease
OVERVIEW
Trends Carbon monoxide emissions decreased 41 percent since 1970. Progress
continued through the 1980's with 29 percent improvement in air quality levels and a
22 percent reduction in total emissions. This progress occurred despite continued
growth in miles of travel in the U.S. Transportation sources account for approximately
two-thirds of the nation's CO emissions. Emissions from highway vehicles decreased
37 percent during the 1981-90 period, despite a 37 percent increase in vehicle miles of
travel. Estimated nationwide CO emissions decreased less than 1 percent between
1989 and 1990, with forest fire activity in 1990 offsetting the 7 percent decrease in CO
emissions from highway vehicles.
Status In October 1991, EPA designated 42 areas as nonattainment for CO.
1990 Clean Air Act The remaining CO nonattainment areas have specific planning
and implementation requirements specified in Title I of the Act that vary depending
upon the magnitude of the CO problem. In addition, Title II of the Act, which deals
with mobile sources, includes a variety of provisions to help reduce CO levels including
a winter time oxygenated fuels program for CO nonattainment areas, increased
application of vehicle inspection and maintenance programs, and a tailpipe standard
for CO under cold temperature conditions.
1-8
-------
15'
CO TREND, 1981-1990
(ANNUAL 2ND MAX 8-HR AVG)
CONCENTRATION, PPM
ID-
301 SITES
90% of sites have lower
2nd max 8-hr concentrations
than this line
10% of sites have lower
2nd max 8-hr concentrations
than this line
CO EMISSIONS TREND
(1981 vs. 1990)
MILLION METRIC TONS PER YEAR
100
I I I I I I I I
81 82 83 84 85 86 87 88 89 90
1981
[Transportation
Industrial
Processes
1990
Fuel
Combustion
Solid Waste
&Misc.
CO EFFECTS:
Cparbon monoxide enters the bloodstream and reduces the delivery of oxygen to
the body's organs and tissues. The health threat from carbon monoxide is
most serious for those who suffer from cardiovascular disease, particularly
those! with angina or peripheral vascular disease. Healthy individuals also are
Effected but only at higher levels. Exposure to elevated carbon monoxide
levels is associated with impairment of visual perception, work capacity, manual
dexterity, learning ability and performance of complex tasks.
1-9
-------
NITROGEN DIOXIDE (NO2)
AIR CONCENTRATIONS
1981-90: 8 percent decrease (annual mean at 166 sites)
1989-90: 6 percent decrease (annual mean at 211 sites)
EMISSIONS: NOX
1981-90: 6 percent decrease
1989-90: 1 percent decrease
OVERVIEW
Trends Nitrogen oxide emissions increased 6 percent since 1970 but both emissions
(-6 percent) and nitrogen dioxide air quality (-8 percent) showed improvement during
the 1980's. The national trend in annual mean NO2 concentrations was flat for most of
the 1980's, however, annual mean NO2 levels have declined during the past two years.
The two primary source categories of nitrogen oxide emissions, and their contribution
in 1990, are fuel combustion (57 percent) and transportation (38 percent). The
transportation category has decreased 24 percent while fuel combustion emissions are
estimated to have increased by 12 percent.
Status In October 1991, EPA designated only one area as nonattainment for NO2.
Los Angeles, CA, which reported an annual mean of 0.056 ppm in 1990, is the only
urban area that has recorded violations of the annual NO2 NAAQS of 0.053 ppm during
the past 10 years.
1990 Clean Air Act Although Los Angeles is the only nonattainment area for nitrogen
dioxide, the Clean Air Act Amendments of 1990 recognized the need for nitrogen oxide
controls due to its contributing role in other problems including ozone (smog) and acid
rain. EPA has already issued final tighter tailpipe standards for NOX as required under
the new amendments. Future ozone (smog) control plans will address further NOX
controls and the Acid Rain provisions of the Act calls for a 2 million ton NOX reduction
from affected utilities.
1-10
-------
NO2 TREND, 1981-1990
(ANNUAL ARITHMETIC MEAN)
0.07
CONCENTRATION, PPM
NOX EMISSIONS TREND
(1981 vs. 1990)
MILLION METRIC TONS PER YEAR
0.06-
0.04-
0.03-
0.02-
0.01-
166 SITES
NAAQS
90% of sites have lower
Arith Mean concentrations
than this line
Average for ail sites
10% of sites have tower
Arith Mean concentrations
than this line
1981
o-oo-iIiiIiIir~
81 82 83 84 85 86 87 88 89 90
[Transportation
Industrial
Processes
1990
Fuel
Combustion
Solid Waste
& Misc.
NO2 EFFECTS ' \ '. ;, ;' : -- ' - |i ".,:-:;'-"' :.-; = . :
Nitrogen dioxide can Irritate the lungs and lower resistance |o respiratory;:;'' :,
infection {such as irifiuertza), ;The effects ol short-term exposure are stt| i!
: unclear but continue?! or frequent exposure to concentration^ higher than thpse
normally found in th6 ambient air may cause increased incidence of acute: .
respiratory disease in children, ; Nitrogen oxides are ah important precursor:
both to ozone and to acidic precipitation and may affect both terrestriaj and i
: aquatic ecosystems. Atmospheric deposition of NO* is a potentially significant:
contributor to ecosystem effects Including alga) blooms in certain estuaries such
as the Chesapeake Bay. In some western, area?, NO* is an important
'''''''''''""
1-11
-------
AIR CONCENTRATIONS
1981-90: 10 percent decrease (second highest daily max 1-hour at 471 sites)
51 percent decrease (exceedance days at 471 sites)
1989-90: 1 percent decrease (second highest daily max 1-hour at 590 sites)
EMISSIONS: VOC
1981-90: 12 percent decrease
1989-90: 1 percent increase
OVERVIEW
Trends Ground level ozone, the primary constituent of smog, has been a pervasive
pollution problem for the U.S. Ambient trends during the 1980's were influenced by
varying meteorological conditions. Relatively high 1983 and 1988 ozone levels are
likely attributed in part to hot, dry, stagnant conditions in some areas of the country.
Both 1989 and 1990 levels showed improvement but the complexity of the ozone
problem warrants caution in interpreting the data. There have been recent control
measures, such as lower Reid Vapor Pressure (RVP) for gasoline resulting in lower
fuel volatility and lower NOX and VOC emissions from tailpipes. Emission estimates for
volatile organic compounds (VOCs), which contribute to ozone formation, are estimated
to have improved by 31 percent since 1970 and 12 percent since 1981. However,
these volatile organic compound (VOC) emission estimates represent annual totals.
NOX emissions, the other major precursor factor in ozone formation, decreased 6
percent between 1981 and 1990. While these annual emission totals are the best
national numbers now available, ozone is predominantly a warm weather problem and
seasonal emission trends would be preferable.
Status In October 1991, EPA designated 98 areas as nonattainment for O3.
1990 Clean Air Act The Act expanded the framework for designating areas as
attainment or nonattainment for ozone by further classifying areas based upon the
magnitude of their problem. Ozone nonattainment areas are now classified as
marginal, moderate, serious, severe or extreme. This allows more flexibility in the
required control program. The Act includes a variety of new requirements for cars and
other sources of ozone precursors, including the introduction of cleaner (reformulated)
gasoline beginning in 1995 into the nine U.S. cities with the worst ozone problems.
1-12
-------
OZONE TREND, 1981 -1990 VOC EMISSIONS TREND
(ANNUAL 2ND DAILY MAX HOUR) (1981 VS. 1990)
0.30
0.25i
0.20-
0.15-
0.10-
0.05-
CONCENTRATION, PPM
MILLION METRIC TONS PER YEAR
0.00
471 SITES
90% of sites have lower
2nd max 1-hr concentrations
than this line
NAAOS '::;:
10% of sites have lower
2nd max 1-hr concentrations
than this line
25
20
15
10
213
187
1981
\ I I I I I 1 1
81 82 83 84 85 86 87 88 89 90
Transportation
Industrial
Processes
1990
I Fuel
Combustion
ISolid Waste
j& Misc.
63 EFFECTS
The reactivity of ozone causes health problems because it damages biological
tissues and cells. Recent scientific evidence indicates that ambient levels of
ozone not onJy affect people with impaired respiratory systems, such as
asthmatics, but healthy adults and children; as welL Exposure to ozone for 6 -
7 hours at relatively low concentrations (i.e. 0.08 ppm) has been found to
significantly reduce lung function in normal, healthy people during periods of
moderate exercise. This decrease in lung function often is accompanied by
such symptoms as chest pain, coughing, nausea and pulmonary congestion.
Though less well established in humans, animal studies have demonstrated
that repeated exposure to ozone for months to years can produce permanent
structural damage in the lungs and accelerate the rate of lung function loss and
aging of tne lungs. Ozone is responsible each year for agricultural crop yield
losjs in the U-S, of several billion dollars and causes noticeable foliar damage In
many crops and species of trees. Forest and ecosystem studies indicate that
damage is resulting from current ambient ozone levels.
1-13
-------
AIR CONCENTRATIONS
1981-90: 85 percent decrease (maximum quarterly average at 202 sites)
1989-90: 12 percent decrease (maximum quarterly average at 229 sites)
EMISSIONS
1981-90: 87 percent decrease in total lead emissions
(95 percent decrease in lead emissions from transportation sources)
1989-90: 1 percent decrease in total lead emissions
(no change in lead emissions from transportation sources)
OVERVIEW
Trends Total lead emissions have dropped 97 percent since 1970 due principally to
reductions in ambient lead levels from automotive sources. Ambient lead (Pb)
concentrations in urban areas throughout the country have decreased 85 percent since
1981 while emissions decreased by 87 percent. The drop in Pb consumption and
subsequent Pb emissions was brought about by the increased use of unleaded
gasoline in catalyst-equipped cars (89 percent of the total gasoline market in 1990)
and the reduced Pb content in leaded gasoline.
Status In 1990, the reduction of exposure to lead became a top priority objective for
the Agency. Among other things, EPA identified 29 stationary sources with potential
problems. An assessment of these sources' compliance status, ambient monitoring
availability and State implementation plan (SIP) adequacy was completed.
1990 Clean Air Act The Amendments, for the first time, authorize EPA to designate
areas nonattainment, attainment or unclassifiable for the lead NAAQS. As such, EPA
has designated as nonattainment 12 areas which have recently recorded violations of
the lead NAAQS. EPA has also designated as unclassifiable 9 areas for which
existing air quality data are insufficient at this time to designate as either attainment or
nonattainment. As States submit designation requests and as ambient monitoring data
become available, EPA will proceed to designate additional lead areas as appropriate.
Once an area is designated nonattainment for the lead NAAQS, States must submit
revised pollution control plans within 18 months of the area's nonattainment
designation.
1-14
-------
PB TREND, 1981-1990
(ANNUAL MAX QUARTERLY AVG)
PB EMISSIONS TREND
(1981 vs. 1990)
CONCENTRATION, UG/M"
THOUSAND METRIC TONS PER YEAR
1.5-
1-
0.5-
202 SITES
NAAQS
90% of sites have tower
Max Quarterly Means
than this line
10% bf shes h&re tower
Max Quarterly Means
than this line
81 82 83 84 85 86 87 88 89 90
1981
Transportation
iljllndustrial
_1J Processes
1990
Fuel
Combustion
Solid Waste
&Mlsc.
PB EFFECTS
Exposure to lead can occur through multiple pathways, including Inhalation of
afr, diet and ingestion of lead in food, water, soil or dust. Lead accumulates In
the body in blood, bone and soft tissue. Because it is not readily excretejd, lead;
also affects the kidneys, liver, nervous system and blood-forming organs. '
Excessive exposure to lead may cause neurological impairments such as
seizures, mental retardation and/or behavioral disorders. Even at low doses,
lead exposure ts associated with changes in fundamental enzymatic, energy
transfer and homeostatic mechanisms in the body. Fetuses, infants and;
childrervare especially susceptible to low doses of lead, often suffering central
nervous: system damage. Recent studies have also shown that lead may be a
factor in high blood pressure and subsequent heart disease in middle-aged
white males.
1-15
-------
1.4 REFERENCES
1. The National Air Monitoring Program: Air Quality and Emissions Trends - Annual
Report. EPA-450/1-73-001 a and b, U. S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle Park, NC 27711, July 1973.
2. Monitoring and Air Quality Trends Report. 1972. EPA-450/1-73-004, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, December 1973.
3. Monitoring and Air Quality Trends Report. 1973. EPA-450/1-74-007, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, October 1974.
4. Monitoring and Air Quality Trends Report. 1974. EPA-450/1-76-001, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, February 1976.
5. National Air Quality and Emissions Trends Report, 1975, EPA-450/1-76-002, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, November 1976.
6. National Air Quality and Emissions Trends Report. 1976. EPA-450/1-77-002, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, December 1977.
7. National Air Quality and Emissions Trends Report. 1977. EPA-450/2-78-052, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, December 1978.
8. 1980 Ambient Assessment - Air Portion. EPA-450/4-81-014, U. S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC
27711, February 1981.
9. National Air Quality and Emissions Trends Report. 1981. EPA-450/4-83-011, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, April 1983.
10. National Air Quality and Emissions Trends Report. 1982. EPA-450/4-84-002, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, March 1984.
11. National Air Quality and Emissions Trends Report. 1983, EPA-450/4-84-029, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, April 1985.
1-16
-------
12. National Air Quality and Emissions Trends Report. 1984. EPA-450/4-86-001, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, April 1986.
13. National Air Quality and Emissions Trends Report. 1985. EPA-450/4-87-001, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, February 1987.
14. National Air Quality and Emissions Trends Report. 1986. EPA-450/4-88-001, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, February 1988.
15. National Air Quality and Emissions Trends Report. 1987. EPA-450/4-89-001, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, March 1989.
16. National Air Quality and Emissions Trends Report. 1988. EPA-450/4-90-002, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, March 1990.
17. National Air Quality and Emissions Trends Report, 1989. EPA-450/4-91-003, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, February 1991.
18. National Air Pollutant Emission Estimates. 1940-1990. EPA-450/4-91-026, U. S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711, November 1991.
1-17
-------
2. INTRODUCTION
This report focuses on 10-year (1981-90)
national air quality trends for each of the major
pollutants for which National Ambient Air Quality
Standards (NAAQS) have been established. This
section presents many of the technical details
involved in these analyses; readers familiar with
previous reports may prefer initially to proceed
directly to the remaining sections. The national
analyses are complemented in Section 5 with air
quality trends in 15 metropolitan areas for the
period 1981 through 1990.
The national air quality trends are based on the
results of actual air pollution measurements at air
monitoring sites located throughout the U.S. The
National Air Monitoring Station (NAMS) sites were
established through monitoring regulations
promulgated in May 19791. The NAMS sites
provide accurate and timely data to the U.S.
Environmental Protection Agency (EPA) from a
national air monitoring network. The NAMS are
located in areas with higher pollutant
concentrations and high population exposure.
These stations meet uniform criteria for siting,
quality assurance, equivalent analytical
methodology, sampling intervals and instrument
selection to assure consistent data reporting among
the States. Other sites operated by the State and
local air pollution control agencies, such as the
State and Local Air Monitoring Stations (SLAMS)
and Special Purpose Monitors (SPM), in general,
also meet the same rigid criteria. However, in
addition to being located in the area of highest
concentration and high population exposure, these
sites are located in other areas as well.
Air quality status may be determined by
comparing the ambient air pollution levels with the
appropriate primary and secondary NAAQS for
each of the pollutants (Table 2-1). Primary
standards protect the public health; secondary
standards protect the public welfare as measured
by effects of pollution on vegetation, materials and
visibility. The standards are further categorized for
different averaging times. Long-term standards
specify an annual or quarterly mean that may not
be exceeded; short-term standards specify upper
limit values for 1-, 3-, 8- or 24-hour averages.
Except for the pollutants ozone and PM-10, the
short-term standards are not to be exceeded more
than once per year. The ozone standard requires
that the expected number of days per calendar year
with daily maximum hourly concentrations
exceeding 0.12 parts per million (ppm) be less than
or equal to one. The 24-hour PM-10 standard also
allows one expected exceedance per year.
Trends are also presented for annual
nationwide emissions. These are estimates of the
amount and kinds of pollution being emitted by
automobiles, factories and other sources, based
upon the best available engineering calculations for
a given time period. Five source categories of
direct emissions have been historically inventoried:
tail pipe emissions from transportation sources, fuel
combustion from powerplants and residential
sources, other stationary sources resulting from
industrial processes, solid waste and
miscellaneous. The latter largely consists of
emissions resulting from forest fires. The 1990
emission estimates are preliminary and may be
revised in the next annual report. The emission
trends are taken from the EPA publication, National
Air Pollutant Emission Estimates, 1940-19902. The
reader is referred to this publication for more
detailed information. For particulates, emission
estimates are presented for both total particulates,
without any distinction of particle sizes, as well as
for PM-10, which refers to "inhalable" particles with
aerodynamic diameter less that 10 microns.
Trends in sources of fugitive dust emissions for
PM-10 are included for 1985-1990. These fugitive
emissions are estimated to amount to a
considerable portion of paniculate emissions.
Fugitive sources surveyed include vehicular traffic
on paved and unpaved roads, wind erosion,
construction activity and agriculture tilling.
Section 4 of this report, "Air Quality Status of
Metropolitan Areas, 1990", provides greatly
simplified air pollution information. Air quality
statistics are presented for each of the pollutants
for all Metropolitan Statistical Areas (MSAs)
reporting monitoring data to EPA for 1990.
2-1
-------
TABLE 2-1. National Ambient Air Quality Standards (NAAQS) in Effect in 1990.
POLLUTANT PRIMARY (HEALTH RELATED) SECONDARY (WELFARE RELATED)
Averaging Time
Standard Level
Concentration* Averaging Time
Standard Level
Concenlration
PM-10
Annual
Arithmetic
Mean6
24-hourb
50 u.g/m3
150 u.g/m3
Same as Primary
Same as Primary
SO,
Annual
Arithmetic
Mean
(0.03 ppm)
80 u.g/m3
3-hour0
1300 (ig/m3
(0.50 ppm)
24-hour0
(0.14 ppm)
365 u.g/m3
CO
8-hour0
9 ppm
(10ug/m3)
No Secondary Standard
1-hour0
35 ppm
(40 u.g/m3)
No Secondary Standard
NO,
Annual
Arithmetic
Mean
0.053 ppm
(100 u.g/m3)
Same as Primary
Maximum Daily
1-hour
Averaged
0.12 ppm
(235
Same as Primary
Pb
Maximum
Quarterly
Average
1.5 u.g/m3
Same as Primary
Parenthetical value is an approximately equivalent concentration.
TSP was the indicator pollutant for the original paniculate matter (PM) standards. This
standard has been replaced with the new PM-10 standard and it is no longer in effect. New
PM standards were promulgated in 1987, using PM-10 (particles less than 10u, in diameter) as
the new indicator pollutant. The annual standard is attained when the expected annual
arithmetic mean concentration is less than or equal to 50 (ig/m3; the 24-hour standard is
attained when the expected number of days per calendar year above 150 u,g/m3 is equal to or
less than 1; as determined according to Appendix K of the PM NAAQS.
Not to be exceeded more than once per year.
The standard is attained when the expected number of days per calendar year with maximum
hourly average concentrations above 0.12 ppm is equal to or less than 1, as determined
according to Appendix H of the Ozone NAAQS.
2-2
-------
2.1 AIR QUALITY DATA BASE
The ambient air quality data used in this report
were obtained from EPA's Aerometric Information
Retrieval System (AIRS). Air quality data are
submitted to AIRS by both State and local
governments, as well as federal agencies.
Presently, there are about 500 million air pollution
measurements on AIRS. The vast majority of
these measurements represent the more heavily
populated urban areas of the nation.
For a monitoring site to have been included in
the national 10-year trend analysis, the site had to
contain complete data for at least 8 of the 10 years
1981 to 1990. For the regional comparisons, the
site had to report data in each of the last three
years to be included in the analysis. Table 2-2
displays the number of sites meeting the
completeness criteria for both data bases. For
PM-10, whose monitoring network has just been
initiated over the last few years, analyses are
based on 339 sites with data in 1988 through 1990.
Data for each year had to satisfy annual data
completeness criteria appropriate to pollutant and
measurement methodology. The air quality data
are divided into two major groupings - 24-hour
measurements and continuous 1-hour
measurements. The 24-hour measurements are
obtained from monitoring instruments that produce
one measurement per 24-hour period and are
typically operated on a systematic sampling
schedule of once every 6 days, or 61 samples per
year. Such instruments are used to measure TSP,
PM-10, SO2, NO2 and Pb. For PM-10, more
frequent sampling of every other day or everyday
is now also common. Data collected only as
24-hour measurements were not used in the SO2
and NO2 trends analyses because these methods
have essentially been phased out of the monitoring
network. Total suspended paniculate and PM-10
data were judged adequate for trends analysis if
there were at least 48 samples for the year. Both
24-hour and composite data were used in the Pb
trends analyses. The 24-hour Pb data had to have
at least six samples per quarter in at least 3 of the
4 calendar quarters. Monthly composite Pb data
were used if at least two monthly samples were
available for at least 3 of the 4 calendar quarters.
The 1-hour data are obtained from monitoring
instruments that operate continuously, producing a
measurement every hour for a possible total of
8760 hourly measurements in a year. For
continuous hourly data, a valid annual mean for
SO2 and NO2 trends requires at least 4380 hourly
observations. This same annual data
completeness, of at least 4380 hourly values, was
required for the CO standard related statistics - the
second maximum nonoverlapping 8-hour average
and the estimated number of exceedances of the
8-hour average CO standard. A slightly different
criterion was used for the SO2 standard related
daily statistics - the second daily maximum 24-hour
average and the estimated number of daily
exceedances of the SO2 standard. Instead of
requiring 4380 or more hourly values, 183 or more
daily values were required. A valid day is defined
as one consisting of at least 18 hourly
observations. Because of the different selection
criteria, the number of sites used to produce the
daily SO2 statistics is slightly different than the
number of sites used to produce the annual SO2
statistics.
Finally, because of the seasonal nature of
ozone, both the second daily maximum 1-hour
value and the estimated number of exceedances of
the O3 NAAQS were calculated for the ozone
season, which typically varies by State.3 For
example, in California, the ozone season is defined
as 12 months, January through December, while in
New Jersey it is defined as 7 months, April through
October. For a site to be included, at least 50
percent of the daily data had to be available for the
ozone season.
The use of a moving 10-year window for trends
yields a data base that is more consistent with the
current monitoring network and reflects the period
following promulgation of uniform monitoring
requirements. In addition, this procedure increases
the total number of trend sites (except for TSP) for
the 10-year period relative to the data bases used
in the last annual report.4 As shown in Section 3,
the size of the TSP monitoring network has been
declining, especially since promulgation of the
PM-10 standard.
2-3
-------
TABLE 2-2, Number of Air Quality Trend Sites, 1981-90 and 1988-90.
NUMBER OF SITES NUMBER OF
POLLUTANT REPORTING IN 1990 1981-90
Total Suspended Paniculate (TSP)
Particulate (PM-10)
Sulfur Dioxide (SO2)
Carbon Monoxide (CO)
Nitrogen Dioxide (NO2)
Ozone (O3)
Lead (Pb)
TOTAL
1,061
1,279
741
491
330
812
406
5,120
1,265
N/A
457
301
166
471
202
2,862
TREND SITES
1988-90
734
339
552
359
211
590
229
3,014
2.2 TREND STATISTICS
The air quality analyses presented in this report
comply with the recommendations of the
Intra-Agency Task Force on Air Quality Indicators.5
The air quality statistics used in these
pollutant-specific trend analyses relate to the
appropriate NAAQSs. Two types of
standard-related statistics are used - peak statistics
(the second maximum 24-hour SO2 average, the
second maximum nonoverlapping 8-hour CO
average, and the second daily maximum 1-hour O3
average) and long-term averages (the annual
geometric mean for TSP, the annual arithmetic
means for PM-10, SO2 and NO2, and the quarterly
arithmetic mean for Pb). For the peak statistics,
the second maximum value is used, because this
is the value that traditionally has been used to
determine whether or not a site has or has not met
an air quality standard in a particular year. For
PM-10, with its variable sampling frequency, the
90th percentile of 24-hour concentrations is used to
examine changes in peak values. A composite
average of each of these statistics is used in the
graphical presentations that follow. All sites were
weighted equally in calculating the composite
average trend statistic. Missing annual summary
statistics for the second through ninth years for a
site are estimated by linear interpolation from the
surrounding years. Missing end points are replaced
with the nearest valid year of data. This procedure
results in a statistically balanced data set to which
simple statistical procedures and graphics can be
applied. The procedure is also conservative,
because end-point rates of change are dampened
by the interpolated estimates.
This report presents statistical confidence
intervals to facilitate a better understanding of
measured changes in air quality. Confidence
intervals are placed around composite averages,
based on sites that satisfy annual data
completeness requirements. The confidence
intervals can be used to make comparisons
between years; if the confidence intervals for any 2
years do not overlap, then the composite averages
2-4
-------
of the 2 years are significantly different (Figure
2-1). Ninety-five percent confidence intervals for
composite averages of annual means (arithmetic
and geometric) and second maxima were
calculated from a two-way analysis of variance
followed by an application of the Tukey Studentized
Range.6 The confidence intervals for composite
averages of estimated exceedances were
calculated by fitting Poisson distributions7 to the
exceedances each year and then applying the
Bonferroni multiple comparisons procedure.8 The
utilization of these procedures is explained in
publications by Pollack, Hunt and Curran9 and
Pollack and Hunt.10
Boxplots11 are used to present air quality trends
because they have the advantage of displaying,
simultaneously, several features of the data.
Figure 2-2 illustrates the use of this technique in
presenting the 5th, 10th, 25th, 50th (median), 75th,
90th and 95th percentiles of the data, as well as
the composite average. The 5th, 10th and 25th
percentiles depict the "cleaner" sites. The 75th,
90th and 95th depict the "higher" sites, and the
median and average describe the "typical" sites.
For example, 90 percent of the sites would have
concentrations equal to or lower than the 90th
percentile. Although the average and median both
-95thPERCENTILE
-90th PERCENTILE
-75th PERCENTILE
- COMPOSITE AVERAGE
-MEDIAN
-25th PERCENTILE
-10th PERCENTILE
-5th PERCENTILE
Figure 2-2. Illustration of plotting
convention of boxplots.
COMPOSITE MEAN
RELATIONSHIPS (MULTIPLE COMPARISONS):
YEARS 1 AND 2 ARE NOT SIGNIFICANTLY
DIFFERENT.
YEARS 2 AND 3 ARE NOT SIGNIFICANTLY
DIFFERENT.
YEARS 1 AND 3 ARE SIGNIFICANTLY
DIFFERENT.
YEAR 4 IS SIGNIFICANTLY DIFFERENT FROM
ALL OTHERS.
95% CONFIDENCE
INTERVAL ABOUT
COMPOSITE MEAN
YEAR1
YEAR 2
YEARS
YEAR 4
Figure 2-1. Sample illustration of use of confidence intervals to
determine statistically significant change.
2-5
-------
characterize typical behavior, the median has the
advantage of not being affected by a few extremely
high observations. The use of the boxplots allows
us simultaneously to compare trends in the
"cleaner", "typical" and "higher" sites.
Bar graphs are introduced for the Regional
comparisons with the 3-year trend data base.
These comparisons are based on the ten EPA
Regions (Figure 2-3). The composite averages of
the appropriate air quality statistic of the years
1988, 1989 and 1990 are presented. The
approach is simple, and it allows the reader at a
glance to compare the short-term changes in all ten
EPA Regions.
In addition to concentration related statistics,
other statistics are used, when appropriate, to
clarify further the observed air quality trends.
Particular attention is given to the estimated
number of exceedances of the short-term NAAQSs.
The estimated number of exceedances is the
measured number of exceedances adjusted to
account for incomplete sampling. Trends in
exceedances tend to be more variable than in the
other concentration related statistics, particularly on
a percentage basis. For example, a site may show
a 50 percent decrease in annual exceedances,
from 2 to 1 per year, and yet record less than a 5
percent decrease in average concentration levels.
The change in concentration levels is likely to be
more indicative of changes in emission levels.
Trends are also presented for annual
nationwide emissions. These emissions data are
estimated using the best available engineering
calculations. The emissions data are reported as
teragrams (one million metric tons) emitted to the
atmosphere per year, except for lead emissions,
which are reported as gigagrams (one thousand
metric tons).2 These are estimates of the amount
and kinds of pollution being generated by
automobiles, factories and other sources.
Estimates for earlier years are recomputed using
current methodology so that these estimates are
comparable over time.
9 {>
Figure 2-3. Ten Regions of the U.S. Environmental Protection Agency.
2-6
-------
2.3 REFERENCES
1. Ambient Air Quality Surveillance, 44 FR
27558, May 10, 1979.
2. National Air Pollutant Emission Estimates.
1940-1990. EPA-450/4-91-026, U. S. Environmental
Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC,
November 1991.
3. Ambient Air Quality Surveillance.
51 FR 9597, March 19, 1986.
4. National Air Quality and Emissions Trends
Report. 1989. EPA-450/4-91-003, U. S.
Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle
Park, NC, February 1991.
5. U.S. Environmental Protection Agency
Intra-Agency Task Force Report on Air Quality
Indicators. EPA-450/4-81-015, U. S. Environmental
Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC,
February 1981.
6. B. J. Winer, Statistical Principles in
Experimental Design. McGraw-Hill, NY, 1971.
7. N. L. Johnson and S. Kotz, Discrete
Distributions, Wiley, NY, 1969.
8. R. G. Miller, Jr., Simultaneous Statistical
Inference, Springer-Verlag, NY, 1981.
9. A. Pollack, W. F. Hunt, Jr., and T. C.
Curran, "Analysis of Variance Applied to National
Ozone Air Quality Trends", presented at the 77th
Annual Meeting of the Air Pollution Control
Association, San Francisco, CA, June 1984.
10. A. Pollack and W. Hunt, "Analysis of Trends
and Variability in Extreme and Annual Average
Sulfur Dioxide Concentrations", presented at the Air
Pollution Control Association, American Society for
Quality Control Specialty Conference on Quality
Assurance in Air Pollution Measurements, Boulder,
CO, 1985.
11. J. W. Tukey, Exploratory Data Analysis.
Addison-Wesley Publishing Company, Reading,
MA, 1977.
2-7
-------
3. NATIONAL AND REGIONAL TRENDS IN NAAQS POLLUTANTS
EPA has set National Ambient Air Quality
Standards (NAAQS) for six pollutants considered
harmful to public health: paniculate matter [formerly
as total suspended particulates (TSP), now as
particulates less than 10 microns in diameter
(PM-10)], sulfur dioxide (SO2), carbon monoxide
(CO), nitrogen dioxide (NO2), ozone (O3) and lead
(Pb). This chapter focuses on both 10-year (1981-
°3) trends and recent changes in air quality and
emissions for these six pollutants. Changes since
1988, and comparisons between all the trend sites
and the subset of National Air Monitoring Stations
(NAMS) are highlighted. Trends are examined for
both the nation and the ten EPA Regions.
As in previous reports, the air quality trends are
presented using trend lines, confidence intervals,
boxplots and bar graphs. The reader is referred to
Section 2.2 for a detailed description of the
confidence interval and boxplot procedures. The
plotting conventions for the confidence intervals
and boxplots are shown in Figures 2-1 and 2-2,
respectively. Boxplots of all trend sites are
presented for each year in the 10-year trend.
Recent changes are presented using the 3-year
data base, 1988 through 1990. The recent 3-year
period is presented to take advantage of the larger
number of sites for all but particulates, and of sites
that have operated continuously during the last
three years.
Trends are also presented for annual
nationwide emissions of paniculate
matter, sulfur oxides (SOX), carbon
monoxide (CO), nitrogen oxides (NOJ,
volatile organic compounds (VOC) and
lead (Pb). These emissions data are
estimated using best available
engineering calculations. The reader is
referred to a companion report for a
detailed description of emission trends,
source categories and estimation
procedures.1 For particulates, emission
estimates are presented both in terms of
total paniculate (TP), which includes all
particles regardless of size, and for
PM-10. This report presents short-term
paniculate matter trends relating to
PM-10 air quality and emissions data.
While the ambient data trends and the emission
trends can be viewed as independent assessments
that lend added credence to the results, the
emission estimates can also be used to provide
information on trends over longer time periods.
Because of changes that have occurred in ambient
monitoring measurement methodology and the
change over time in the geographical distribution of
monitors, it is difficult to provide ambient trends
going back to 1970, other than for TSP, and yet it
is important not to lose sight of some of the earlier
progress that was made in air pollution control.
Emission estimates can provide some insight in this
area. Figure 3-1 depicts long-term change in
emission estimates. Lead clearly shows the most
impressive decrease of 97 percent but
improvements are also seen for TP (-59 percent),
SOX (-25 percent), CO (-41 percent) and VOC (-31
percent). Only NO, has not shown improvement
with emissions estimated to have increased 6
percent from 1970 levels, due primarily to increased
fuel combustion by stationary sources. However,
the 1990 NOX emissions estimate is 6 percent lower
than the estimate for 1981. Los Angeles is the only
metropolitan area that currently does not meet the
NO2 NAAQS.
MILLION METRIC TONS/YEAR
METRIC TONS/YEAR
LEAD
Figure 3-1. Comparison of 1970 and 1990 emissions.
3-1
-------
3.1 TRENDS IN PARTICULATE MATTER
Air pollutants called paniculate matter include
dust, dirt, soot, smoke and liquid droplets directly
emitted into the air by sources such as factories,
power plants, cars, construction activity, fires and
natural windblown dust as well as particles formed
in the atmosphere by condensation or
transformation of emitted gases such as sulfur
dioxide and volatile organic compounds.
Annual and 24-hour National Ambient Air
Quality Standards (NAAQS) for paniculate matter
were first set in 1971. Total suspended paniculate
(TSP) was the indicator used to represent
suspended particles in the ambient air. TSP is
measured using a high volume sampler (Hi-Vol)
which collects suspended particles ranging up to
approximately 45 micrometers in diameter.
On July 1,1987 EPA promulgated new annual
and 24-hour standards for paniculate matter, using
a new indicator, PM-10, that includes only those
particles with aerodynamic diameter smaller than
10 micrometers. These smaller particles are likely
responsible for most adverse health effects of
paniculate because of their ability to reach the
thoracic or lower regions of the
respiratory tract. The original (TSP)
standards were an annual geometric
mean of 75 u,g/m3, not to be exceeded,
and a 24-hour concentration of 260
u,g/m3, not to be exceeded more than
once per year. The new (PM-10)
standards specify an expected annual
arithmetic mean not to exceed 50 (ig/m3
and an expected number of 24-hour
concentrations greater than 150 u.g/m3
per year not to exceed one.
less that 1100 for TSP. In 1981 there were
approximately 4000 TSP monitoring locations.
There are basically two types ot reference
instruments currently used to sample PM-10. The
first is essentially a Hi-Vol, like the one used for
TSP, but with a different size selective inlet (SSI).
This sampler uses an inert quartz filter. The other
type of instrument is a "dichotomous" sampler. It
uses a different PM-10 inlet, operates at a slower
flow rate, and produces two separate samples: 2.5
to 10 microns and less than 2.5 microns, each
collected on a teflon filter.
With the new PM-10 standards, more emphasis
is being placed on detection of peak 24-hour
concentrations. Unlike monitoring regulations for
TSP which only required once in 6-day sampling,
new specifications for PM-10 now dictate more
frequent sampling. Approximately 15 percent of all
PM-10 sampling sites operate either every other
day or everyday. In contrast, only 5 percent of TSP
Hi-Vols had been operating more frequently than
once in 6 days.
# SITES
4,000
3,000 -
2,000 -
1,000 -
With the change from TSP to PM-10
as the indicator for paniculate matter, the
number of TSP monitors has been
steadily declining and a network of
locations to monitor PM-10 has evolved.
Figure 3-2 shows the 10-year decline of
the number of TSP monitors nationally,
contrasted with the developing PM-10
network. Approximately 1300 PM-10
sites were active in 1990, compared with
1981
1982 1983 1984 1985 1986
YEAR
1987 1988 1989 1990
TSP sites
PM10 sites
Figure 3-2. National trend in the number of TSP
and PM-10 monitoring locations, 1981-1990.
3-2
-------
Although some monitoring for PM-10 was
initiated prior to promulgation of the new standards,
most networks did not produce data with approved
reference samplers until mid-1987 or 1988. Thus,
only a limited data base is currently available to
examine trends in PM-10 air quality and
longer-term trends in paniculate matter can only be
based on TSP. Both 10-year trends and recent
3-year changes in TSP are presented in terms of
average air quality (annual geometric mean).
Available information on PM-10 air quality will be
used to report the 1988-1990 changes in PM-10
concentration levels. Two PM-10 statistics are
presented. The annual arithmetic mean
concentration is used to reflect average air quality,
and the 90th percentile of 24-hour concentrations
is used to represent the behavior of peak
concentrations. Because PM-10 sampling
frequency varies among sites and may have
changed during the 3-year period, the 90th
percentile is used. This statistic is less sensitive to
changes in sampling frequency than the peak
values. Finally, cross sectional PM-10 data are
included forthe more comprehensive data available
for calendar year 1990.
3.1.1 Long-term TSP Trends: 1981-90
The 10-year trend in national average
TSP levels, 1981 through 1990, is shown
in Figure 3-3 for 1265 sites
geographically distributed throughout the
Nation. In addition, the entire distribution
of geometric mean concentrations among
all locations are depicted with box-plots.
decrease between 1978 and 1982 has been well
documented).6 However, since the exact
magnitude of the 1981-1982 change is uncertain,
the longer-term change in total paniculate
concentrations is best described in terms of the
9-year period 1982-1990.
Nationally, the composite average TSP levels
declined 3 percent from 1982 to 1990. Upon close
inspection, some changes in composite means
since 1982 are evident. Although the levels over
the last 9 years are relatively stable, the national
TSP levels in 1990 are statistically lower than those
produced in 1987,1988 and all years prior to 1985.
In fact, they are the lowest national numbers
reported in EPA's trends reports. The recent
changes in total suspended particulate matter will
be discussed in more detail in Section 3.1.3.
3.1.2 Total Particulate Emission Trends
Nationwide Total Particulate (TP) emission
trends from historical inventoried sources show an
overall decrease of 6 percent from 1981 to 1990.
(See Table 3-1 and Figure 3-4). The general
10-year emission pattern has similarity to that of
composite average air quality. Both showed a
CONCENTRATION, UG/M3
Measured TSP concentrations
appear to have declined about 15
percent between 1981 and 1982 and are
relatively stable during the last 9 years.
However, the data collected in 1981 (as
well as 1979 and 1980) may have been
affected by the type of filters used to
collect the TSP.2"5 For this reason, the
portion of Figure 3-3 showing the data
for 1981 is shaded to indicate the
uncertainty in these TSP measurements.
Despite this uncertainty, some of the
observed decrease in ambient particulate
matter between 1981 and 1982 is
thought to be real (a 20 percent
1265 SITES
Former NAAQS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-3. Boxplot comparisons of trends in annual
geometric mean total suspended particulate
concentrations at 1265 sites, 1981-1990.
3-3
-------
15
10 -
relatively large decline between 1981
and 1982. This similarity appears to
reduce some of the uncertainty in the
large reported decrease in ambient
concentrations between 1981 and 1982.
The emissions also increased in 1988
and again in 1990, both times due to
atypically large forest fires. In 1990, the
fires occurred in Alaska, but since there
are relatively few TSP monitors in
Alaska, the ambient trends did not follow.
The 4 percent increase in TP emissions
between 1989 and 1990 are partly
responsible for the relatively lower
10-year emission trends than those
presented in last year's report. In fact,
the emissions from 1982 to 1990
increased 6 percent. In any case, the
trend in TP emissions is normally not
expected to agree precisely with the
trend in ambient TSP levels due to unaccounted for
natural paniculate matter background and
uninventoried emission sources such as unpaved
roads and construction activity. Such fugitive
emissions are not considered in estimates of the
annual nationwide total and could be significant in
populated areas. Information on these sources is
TP EMISSIONS, 106 METRIC TONSAEAR
SOURCE CATEGORY
TRANSPORTATION
m FUEL
COMBUSTION
i INDUSTRIAL PROCESSES
I SOLID WASTE & MISC
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-4. National trend in total particulate
emissions, 1981-1990,
presented in terms of the PM-10 portion of
paniculate matter in Section 3.1.5. Total particulate
emission estimates also exclude significant
contributions from gas phase particulate matter
precursors (principally sulfur oxides and nitrogen
oxides). The 10-year reductions in inventoried total
particulate emissions occurred primarily in the fuel
TABLE 3-1. National Total Particulate Emission Estimates, 1981-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
1.3
2.3
3.0
0.4
0.9
8.0
1982
1.3
2.2
2.6
0.3
0.7
7.1
1983
1.3
2.0
2.4
0.3
1.1
7.1
1984
1.3
2.1
2.8
0.3
0.9
7.4
1985
1.4
1.8
2.8
0.3
1.0
7.2
1986
1.4
1.8
2.6
0.3
0.8
6.7
1987
1.4
1.8
2.5
0.3
0.9
6.9
1988
1.5
1.7
2.7
0.3
1.3
7.5
1989
1.5
1.8
2.7
0.3
0.9
7.2
1990
1.5
1.7
2.8
0.3
1.2
7.5
NOT£: The sums of sub-categories may not equal total due to rounding.
3-4
-------
80
70
60
50
40
30
20
10
combustion category. This is attributed to
installation of control equipment by electric
utilities, despite an increase in fuel
consumed. Some of the reduced
emissions is also due to reduced activity
in some industries, such as iron and
steel.1
3.13 Recent TSP Trends: 1988-90
The TSP trends for the 3-year period
1988-1990 are presented in terms of 734
sites which produced data in each of the
last 3 years. The group of sites
qualifying for this analysis is smaller than
the group used to analyze long-term
trends, reflecting the revisions to TSP
SLAMS networks and the shift of
particulate monitoring to PM-10.
Nationally, a small but statistically
significant decrease occurred between
1988 and 1990. Average concentrations
decreased 9 percent between these 2
years. This 3-year decline occurred in
each geographic Region of the country. The same
3-year pattern can also be seen in Figure 3-3 for
the larger 10-year data base.
Figure 3-5 focuses on the last 3 years with a
bar chart of Regional average TSP. Relatively
large declines occurred in almost all Regions. This
pattern is likely attributed, at least in part, to the
effects of the weather.
Rainfall has the effect of reducing reentrainment
of particles and of washing particles out of the air.
Generally drier conditions are also associated with
an increase in forest fires.
During 1988, most of the nation experienced an
extreme drought. Nationally, this year was the
driest since 1956 and the second driest in the last
50 years. Most of the Nation returned to more
normal annual rainfall in 1989 and to slightly above
average rainfall in 1990. The drought, however,
continues in some western states, most notably,
Montana and California.
Comparing 1990 with 1989, most of the country
had experienced an increase in precipitation.
Nationally, TSP decreased 3 percent between
CONCENTRATION, UG/M3
COMPOSITE AVERAGE
1983 ^ 1989 CD 19!
EPA REGION I
NO. OF SITES 54
II
42
III IV
62 221
V
130
VI
32
VII
73
VIII
40
IX
58
X
22
Figure 3-5. Regional comparisons of the 1988,1989,
1990 composite averages of the geometric mean total
suspended particulate concentrations.
these two years. Region IV and Mid-Atlantic States
were exceptions which had more than 10 percent
decreases in total precipitation.7 Correspondingly,
only Region IV showed a 2-year (1989-1990)
increase in total suspended particulate.
3.1.4 Recent PM-10 Air Quality: 1988-90
The 1988 to 1990 change in the PM-10 portion
of total particulate concentrations is examined at
339 monitoring locations which produced data in all
three years. A more comprehensive national
sample of 979 sites is also presented to provide a
more representative current picture of PM-10 air
quality produced by reference PM-10 samplers.
The sample of 339 "trend" sites reveals an 11
percent decrease in average PM-10 concentrations
between 1988 and 1990. This is consistent with
the 9 percent decrease in total particulates
described earlier. Most of the recent decrease in
PM-10 occurred over the last two years
(1989-1990); average PM-10 concentrations
decreased 8 percent. Similarly, peak 24-hour
PM-10 concentrations similarly decreased 11
percent since 1988 and 8 percent since 1989.
3-5
-------
110
100 -
90 -
80 -
70 -
60 -
50
40 -
30 -
20 -
10 -
0
339
Trend Sites
1988-1990
979 Sites
for 1990
ANNUAL
2..-MEAN
NAAQS
Annual Arithmetic
Means
90th %-tile
of 24-hr concentrations
1988 1989 1990 1990
1988 1989 1990
1990
Figure 3-6. Boxplot comparisons of 1988,1989 and 1990 PM-10
concentrations at 339 sites with 1990 PM-10 air quality at 979 sites.
Change in peak concentrations was examined in
terms of the average of the 90th percentiles of
24-hour concentrations among sampling locations.
Figure 3-6 displays boxplots of the
concentration distribution for the two PM-10 trend
statistics - annual arithmetic mean and 90th
percentile of 24-hour concentrations. The 1988
and 1989 national distributions are very similar for
both annual average and 90th percentile of 24-hour
PM-10 concentrations. The distributions for 1990
are lower for all percentiles than those for the
preceding two years. Figure 3-6 also displays the
concentration distributions for the larger sample of
979 sites. While the larger group of sites is 5
percent lower in the composite average annual
arithmetic mean, it also has a slightly higher
percentage of high concentrations sites.
The larger sample offered by the 1990
concentration distribution of annual arithmetic
means also provides a basis for direct comparison
to the annual standard of 50 u,g/m3. Approximately
200
CONCENTRATION, UG/Mv
175 ~
150
125 -
100 -
75 -
50 -
25 -
24-HOUR NAAQS
90TH
%-TILE
2ND
MAX
MAX
Figure 3-7. Boxplot comparisons of 24-
hour PM-10 peak value statistics for
1990 at 979 sites.
3-6
-------
3 percent of all monitoring stations
reported averages above the annual
standard in this year.
Although the 90th percentile is a
reasonable peak concentration indicator
for temporal comparisons, it does not
directly relate to the 150 u.g/m3 level of
the 24-hour PM-10 standard. Since this
standard permits one expected
exceedance per year, the maximum and
second maximum 24-hour concentrations
provide a more direct indication of
attainment status. A comparison of the
90th percentile of 24-hour concentrations
to these other indicators of peak
concentrations is presented in Figure 3-7
using boxplots of the 1990 national
concentration distribution. Although the
90th percentile concentrations are well
below 150 u.g/m3, maximum
concentrations exceed the standard at 9
percent of the reporting locations while
the second maximum concentrations
exceed at 4 percent.
Figure 3-8 presents the Regional
distribution of PM-10 concentrations for
both average and 90th percentile
concentrations among the 979 stations
producing reference measurements in
1990. On the average, the highest
annual mean and peak 24-hour
concentrations are found in Region IX.
The 90th percentile of 24-hour
concentrations has been used as the
indicator of peak concentrations because
of differences in sampling frequency
among PM-10 sampling locations. Note
that average sampling frequency varies
among Regions, with samplers in
Regions VIII and X operating at
approximately twice the frequency of
samplers in, say, Region II and Region
IX. The monitoring regulations permit
such differences in sampling frequency.
The regulations specify that areas that
are close to the 24-hour standard must
sample more frequently.
CONCENTRATION, UG/M '
REGION
# SITES 105 59
Figure 3-8. Regional comparisons of annual mean
and 90th percentile of 24-hour PM-10 concentrations
for 1990.
150
130 -
110 -
90 -
70 -
50 -
30 -
10 -
CONGELATION,
1988-90 1988-90
MEANS CZD 90th-%tiles
n ru
EPA REGION I II III IV V VI VII VIII IX X
NO. OF SITES 26 18 31 40 80 26 26 44 16 32
Figure 3-9. Regional changes in annual average and
90th percentile of 24-hour PM-10 concentrations,
1988-1990.
3-7
-------
Figure 3-9 presents the 1988 to 1990 changes
in annual average and 90th percentile PM-10
concentrations by EPA Region. The 3-year
national decrease is evident in all Regions. Most of
this decrease occurred everywhere between 1989
and 1990. On the other hand, average PM-10
concentrations in Region IV displayed a minor
increase between 1989 and 1990. Recalling that
Region IV was the only Region with an increase in
total particulates, this outcome may be related to
drier conditions throughout the southeast in 1990.
3.1.5 PM-10 Emission Trends
Trends in the PM-10 portion of historically
inventoried particulate matter emissions are
presented for the 6-year period, 1985-1990 in
Table 3-2. Comparing Tables 3-1 and 3-2, PM-10
appears to represent essentially all of the
particulate emissions from transportation and
industrial sources and most of the emissions in the
other source categories. As was the case for TP,
higher emissions occurred in 1988 and 1990 due to
forest fires. Total PM-10 emissions increased 5
percent since 1989, 2 percent since 1988 and 7
percent since 1985.
National estimates are also provided for PM-10
fugitive emissions for 1985-1990, in Table 3-3.
These estimates provide a good indication of the
relative impacts of major contributors to particulate
matter air quality. In total, these fugitive emissions
are 6 to 8 times more than the historically
inventoried particulate matter sources categories.
Note that PM-10 estimates are not included for
contributions from gas phase particulate matter
precursors, principally sulfur oxides and nitrogen
oxides.
Construction activity and unpaved roads are
consistently the major contributors over time for
most Regions. Nationally, roadway particulate
matter emissions are estimated to have increased
due to increased vehicle traffic. Among road types,
emissions from unpaved and paved roads are
estimated to have increased 6 percent and 22
percent, respectively, since 1985. Emissions from
unpaved roads are highest in Regions which cover
large geographic areas. Emissions due to
construction are estimated to have decreased over
21 percent since 1985 due to reduced activity in
this industry.
TABLE 3-2. National PM-10 Emission Estimates, 1985-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1985
1.3
1.1
2.7
0.3
0.7
6.0
1986
1.3
1.1
2.5
0.2
0.5
5.6
1987
1.3
1.1
2.4
0.2
0.7
5.8
1988
1.4
1.1
2.6
0.2
1.0
6.3
1989
1.5
1.1
2.6
0.2
0.7
6.1
1990
1.5
1.1
2.7
0.2
0.9
6.4
NOTE: The sums of sub-categories may not equal total
due to rounding.
3-8
-------
Agricultural activity is a smaller contributor to
the national total, but estimated to the major source
in specific Regions. Tilling is estimated to be a big
contributor in Regions V, VII, VIII and X, but has
not shown much change over the 6-year period.
Wind erosion particulate emissions are estimated to
be extremely variable from year to year and can
also be a major contributor in some Regions.
Particulate emissions due to wind erosion are very
sensitive to Regional soil conditions and
year-to-year changes in total precipitation.
Accordingly, estimated emissions from wind erosion
were extremely high for the drought year of 1988,
particularly for Regions VI and VII. Finally, among
all fugitive categories surveyed, mining and
quarrying is estimated to be a relatively small
contributor to total fugitive particulate matter
emissions.
3.1.6 VisibUity Trends
Many parts of the nation have experienced
long-term impairment in visibility due to build-up of
emissions around urban areas and from long range
transport of small particles (< 2.5 microns) across
broad regions of the country. This increase in haze
has occurred in the summer season across the
Eastern U.S., although there has been
improvement in the winter. In the Eastern and
Southwestern U.S., regional visibility is mostly
attributed to sulfates formed by release of sulfur
oxides. In the Northwestern U.S., carbon particles
play an important role in the degradation. The
Clean Air Act Amendments of 1990 will address
regional haze in the East through the acid rain
program which will substantially reduce sulfur
oxides emissions. To address regional haze in the
West, the new Act has strengthened the work
already started on protection of visibility in national
park and wilderness areas. Required research will
focus on transport mechanisms and atmospheric
conditions which contribute to hazes.
TABLE 3-3. National PM-10 Fugitive Emission Estimates, 1985-1990
(million metric tons/year)
SOURCE
CATEGORY
Agricultural
Tilling
Construction
Mining and
Quarrying
Paved Roads
Unpaved
Roads
Wind Erosion
TOTAL
1985
6.2
11.5
0.3
5.9
13.3
3.2
40.5
1986
6.3
10.7
0.3
6.1
13.3
8.5
45.3
1987
6.4
11.0
0.3
6.5
12.7
1.3
38.1
1988
6.4
10.6
0.3
6.9
14.2
15.9
54.3
1989
6.3
10.2
0.3
7.0
13.9
10.7
48.5
1990
6.3
9.1
0.3
7.2
14.1
3.8
40.8
NOTE: The sums of sub-categories may not equal total due to
rounding.
3-9
-------
3.2 TRENDS IN SULFUR DIOXIDE
Ambient sulfur dioxide (SO2) results largely
from stationary source coal and oil combustion,
refineries, pulp and paper mills and from
nonferrous smelters. There are three NAAQS for
SO2: an annual arithmetic mean of 0.03 ppm (80
u.g/m3), a 24-hour level of 0.14 ppm (365 u,g/m3)
and a 3-hour level of 0.50 ppm (1300 jig/m3). The
first two standards are primary (health-related)
standards, while the 3-hour NAAQS is a secondary
(welfare-related) standard. The annual mean
standard is not to be exceeded, while the
short-term standards are not to be exceeded more
than once per year. The trend analyses which
follow are for the primary standards. It should be
noted that EPA is currently evaluating the need for
a new shorter-term 1-hour standard.8
High concentrations of SO2 affect breathing and
may aggravate existing respiratory and
cardiovascular disease. Sensitive populations
include asthmatics, individuals with bronchitis or
emphysema, children and the elderly. Although
this report does not directly address trends in acid
deposition, of which SO2 is a major contributor, it
does include information on total nationwide
emissions which is a measure relating to total
CONCENTRATION, PPM
atmospheric loadings. SO2 also produces foliar
damage on trees and agricultural crops.
The trends in ambient concentrations are
derived from continuous monitoring instruments
which can measure as many as 8760 hourly values
per year. The SO2 measurements reported in this
section are summarized into a variety of summary
statistics which relate to the SO2 NAAQS. The
statistics on which ambient trends will be reported
are the annual arithmetic mean concentration, the
second highest annual 24-hour average
(summarized midnight to midnight), and the
expected annual number of 24-hour exceedances
of the 24-hour standard of 0.14 ppm.
3.2.1 Long-term SO2 Trends: 1981-90
The long-term trend in ambient SO2, 1981
through 1990, is graphically presented in Figures
3-10 through 3-12. In each figure, the trend at the
NAMS is contrasted with the trend at all sites. For
each of the statistics presented, a 10-year
downward trend is evident, although the rate of
decline has slowed over the last 3 years.
Nationally, the annual mean SO2, examined at 457
n mo
0.025 -
0.020 -
0.015 -
0.010 -
0.005 -
n nnn
IMA AO°
FM/VAUO
» ^ * * i 3 i j_
*"~ * -*-.r^7*
ALL SITES (457 J NAMS SITES (1331
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-10. National trend in annual average sulfur dioxide
concentration at both NAMS and all sites with 95 percent confidence
intervals, 1981-1990.
3-10
-------
sites, decreased at a median rate of
approximately 2 percent per year; this
resulted in an overall change of about
24 percent (Figure 3-10). The subset of
133 NAMS recorded higher average
concentrations but declined at a median
rate of 3 percent per year, with a net
change of 29 percent for the 10-year
period.
The annual second highest 24-hour
values displayed a similar improvement
between 1981 and 1990. Nationally,
among 452 stations with adequate trend
data, the median rate of change was 3
percent per year, with an overall decline
of 30 percent (Figure 3-11). The 134
NAMS exhibited an overall decrease of
33 percent. The estimated number of
exceedances also showed declines for
the NAMS as well as for the composite
of all sites (Figure 3-12). The national
composite estimated number of
exceedances decreased 87 percent
from 1981 to 1990. However, the vast
majority of SO2 sites do not show any
exceedances of the 24-hour NAAQS.
Most of the exceedances, as well as the
bulk of the improvements, occurred at
source-oriented sites.
The statistical significance of these
long-term trends is graphically illustrated
in Figures 3-10 to 3-12 with the 95
percent confidence intervals. These
figures show that the 1990 composite
average and composite second
maximum 24-hour SO2 levels are the
lowest reported in EPA trends reports.
The 1990 composite annual mean is
statistically lower than all previous
years. Similarly, the composite 1990
peak values are statistically different
than all years except for 1987.
The inter-site variability for annual
mean and annual second highest
24-hour SO2 concentrations is
graphically displayed in Figures 3-13
and 3-14. These figures show that
0.16
CONCENTRATION, PPM
0.14
0.12 -
0.10 -
008 -
0.06 -
0.04 -
0.02 -
0.00
NAAQS
NAMS SITES (134)
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-11. National trend in the second-highest
24-hour sulfur dioxide concentration at both NAMS
and all sites with 95 percent confidence intervals,
1981-1990.
1.5
ESTIMATED EXCEEDANCES
1 -
0.5 -
ALL SITES £452 _)
NAMS SITES (134)
*
1 T T P IT iT^
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-12. National trend in the estimated
number of exceedances of the 24-hour sulfur
dioxide NAAQS at both NAMS and all sites with 95
percent confidence intervals, 1981-1990.
3-11
-------
higher concentration sites decreased
more rapidly and that the concentration
range among sites has also diminished
during the 1980's.
Nationally, sulfur oxides (SO,)
emissions decreased 6 percent from
1981 to 1990 (Figure 3-15 and Table
3-4). This decrease is attributable to
three general changes.1 First, the
decrease is attributable to the
installation of flue gas desulfurization
controls at new coal-fired electric
generating stations and a reduction in
the average sulfur content of fuels
consumed over the 10-year period.
Second, emissions from industrial
processes have declined, primarily as
the result of controls implemented to
reduce emissions from nonferrous
smelters and sulfuric acid manufacturing
plants, as well as shutdowns of some
large smelters. Finally, emissions from
other stationary source fuel combustion
sectors also declined, mainly due to
decreased combustion of coal by these
consumers. The 2 percent increase in
sulfur oxides emissions between 1989
and 1990 is attributed to a projected
increase in electric generation.
The disparity between 10-year
trends and 2-year changes in SO2 air
quality and SOX emissions can be
attributed to several factors. SO2
monitors with sufficient historical data
for trends are mostly urban
population-oriented. They do not
monitor many of the major emitters
which tend to be located in more rural
areas {e.g. large power plants).
Although most of the trend sites are
categorized as population-oriented, the
majority of SOX emissions are
dominated by large point sources.
Seventy percent of all national SOX
emissions are generated by electric
utilities (92 percent of which come from
coal fired power plants). The majority of
0.040
CONCENTRATION, PPM
0.035 -
0.030
0.025 -
0.020 -
0.015 -
0.010 -
0.005 -
0.000
457 SITES
NAAQS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-13. Boxplot comparisons of trends in
annual mean sulfur dioxide concentrations at 457
sites, 1981-1990.
0.20
CONCENTRATION, PPM
0.15 -
0.10 -
0.05 -
0.00
452 SITES
NAACS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-14. Boxplot comparisons of trends in
second highest 24-hour average sulfur dioxide
concentrations at 452 sites, 1981-1990.
3-12
-------
these emissions, however, are produced
by a small number of facilities. Fifty
individual plants in 15 states account for
approximately one-half of all power
plant emissions. In addition, the 200
highest SOX emitters account for more
than 86 percent of all SOX power plant
emissions. These 200 plants shown in
Figure 3-16 account for 60 percent of all
SOX emissions nationally.9
Title IV of the Clean Air Act
Amendments of 1990 addresses the
control of pollutants associated with acid
deposition and includes a goal of
reducing sulfur oxide emissions by 10
million tons relative to 1980 levels. The
focus in this control program is an
innovative market-based emission
allowance program which will provide
affected sources flexibility in meeting
the mandated emission reductions.
This is the first large scale regulatory
use of market-based incentives.
10
SOX EMISSIONS, 106 METRIC TONS/YEA"
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-15. National trend in sulfur oxides
emissions, 1981-1990.
TABLE 3-4. National Sulfur Oxides Emission Estimates, 1981-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
0.9
17.8
3.8
0.0
0.0
22.5
1982
0.8
17.3
3.1
0.0
0.0
21.2
1983
0.8
16.7
3.1
0.0
0.0
20.6
1984
0.8
17.4
3.2
0.0
0.0
21.5
1985
0.9
17.0
3.2
0.0
0.0
21.1
1986
0.9
16.9
3.2
0.0
0.0
20.9
1987
0.9
16.6
3.0
0.0
0.0
20.5
1988
0.9
16.6
3.1
0.0
0.0
20.6
1989
1.0
16.8
3.0
0.0
0.0
20.8
1990
0.9
17.1
3.1
0.0
0.0
21.2
NOTE; The sums of sub-categories may not equal total due to rounding.
3-13
-------
Plant Emission?
50 Largest Plants
Next 150 Largest Plants
if 1990 CAA Phase I Plants
Figure 3-16. Location of the 200 largest power plant emitters of sulfur oxides.
3-14
-------
Figure 3-16 shows 110 power plants which are
required by the 1990 Clean Air Act Amendments to
reduce emissions to specified allowable tonnage by
January 1, 1995. This will accomplish Phase I of
the legislated ten million ton reductions.
Another factor which may account for
differences in SOX emissions and ambient air
quality is stack height. At large utilities and
smelters, SOX is generally released into the
atmosphere through tall stacks. Although sources
are not permitted to increase emissions through
increased dispersion from tall stacks, measured
ground level concentrations in the vicinity of these
existing sources may not reflect local emissions.
Total atmospheric loading impacts also arise, in
part, as a consequence of tall stacks.
3.2.2 Recent SO2 Trends: 1988-90
Nationally, SO2 showed improvement over the
last three years in both average and peak 24-hour
concentrations. Composite annual mean
concentrations consistently decreased for a total of
11 percent between 1988 and 1990. Over the last
2 years, the average annual mean SO2 decrease
was 7 percent. Composite 24-hour SO2
concentrations declined 15 percent since 1988 and
11 percent since 1989.
Figure 3-17 presents the Regional changes in
composite annual average SO2 concentrations for
the last 3 years, 1988-1990. All Regions except for
the Northwest (Region X) follow the national pattern
of change in annual mean SO2. Although not
presented here in graphical format, every Region of
the country reported 3-year declines in peak
24-hour SO2 concentrations.
0.016
0.014
0.012
CONCENTRATION, PPM
COMPOSITE AVERAGE
1988 m 1989 CH 1990
EPA REGION I II
NO. OF SITES 68 41
63
IV
85
V
143
VI
40
VII
33
VIII
25
IX
44
X
10
Figure 3-17. Regional comparisons of the 1988,1989,
1990 composite averages of the annual average
sulfur dioxide concentrations.
3-15
-------
3.3 TRENDS IN CARBON MONOXIDE
Carbon monoxide (CO) is a colorless, odorless
and poisonous gas produced by incomplete burning
of carbon in fuels. Two-thirds of the nationwide
CO emissions are from transportation sources, with
the largest contribution coming from highway motor
vehicles. The NAAQS for ambient CO specify
upper limits for both 1-hour and 8-hour averages
that are not to be exceeded more than once per
year. The 1-hour level is 35 ppm, and the 8-hour
level is 9 ppm. This trends analysis focuses on the
8-hour average results because the 8-hour
standard is generally the more restrictive limit.
Nationally, during 1990, only two exceedances of
the CO 1-hour NAAQS were recorded at a site
which is impacted by a localized, non-mobile
source, and in each case the 8-hour NAAQS was
still the controlling standard.
Carbon monoxide enters the bloodstream and
reduces the delivery of oxygen to the body's organs
and tissues. The health threat is most serious for
those who suffer from cardiovascular disease,
particularly those with angina or peripheral vascular
disease. Exposure to elevated carbon monoxide
levels is associated with impairment of visual
perception, manual dexterity, learning ability and
performance of complex tasks.
Trends sites were selected using the
procedures presented in Section 2.1 which yielded
a data base of 301 sites for the 10-year period
1981-90 and a data base of 359 sites for the
3-year 1988-90 period. There were 92 NAMS sites
included in the 10-year data base and 104 NAMS
sites in the 3-year data base.
3.3.1 Long-term CO Trends: 1981-90
The 1981-90 composite national average trend
is shown in Figure 3-18 for the second highest
non-overlapping 8-hour CO concentration for the
301 long-term trend sites and the subset of 92
NAMS sites. During this 10-year period, the
national composite average decreased by 29
percent and the subset of NAMS decreased by 32
percent. Both curves show similar trends for the
NAMS and the larger group of long-term trend
sites. The median rate of improvement for this time
period is more than 3 percent per year. Except for
a small upturn between 1985 and 1986, composite
average levels have shown a steady decline
throughout this period. Long-term improvement
was seen in each EPA Region with median rates of
improvement varying from 2 to 5 percent per year.
The 1990 composite average is significantly lower
than the composite means for 1986 and earlier
years. This same trend is shown in Figure 3-19 by
a boxplot presentation which provides more
information on the year-to-year distribution of
ambient CO levels at the long-term trend sites.
While there is some year to year fluctuation in
certain percentiles, the general long-term
improvement in ambient CO levels is clear.
Figure 3-20 displays the 10-year trend in the
composite average of the estimated number of
exceedances of the 8-hour CO NAAQS. This
exceedance rate was adjusted to account for
incomplete sampling. The trend in exceedances
shows long-term improvement but the rates are
much higher than those for the second maximums.
The composite average of estimated exceedances
decreased 87 percent between 1981 and 1990 for
the 301 long-term trend sites, while the subset of
92 NAMS showed an 86 percent decrease. These
percentage changes for exceedances are typically
much larger than those found for peak
concentrations. The trend in annual second
maximum 8-hour value is more likely to reflect the
change in emission levels. For both curves, the
1990 composite average of the estimated
exceedances is significantly lower than levels for
1986 and earlier years.
3-16
-------
Figure 3-18. National trend in the
composite average of the second
highest nonoverlapping 8-hour
average carbon monoxide
concentration at both NAMS and
all sites with 95 percent
confidence intervals, 1981-1990.
12
CONCENTRATION, PPM
10 -
8 -
6 -
4 -
2 -
* ALL SJTES _(3_01 j
NAMS SITES (92]
I I I I I I I I I T
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-19. Boxplot comparisons
of trends in second highest
nonoverlapping 8-hour average
carbon monoxide concentrations
at 301 sites, 1981-1990.
20
CONCENTRATION, PPM
15 -
10 -
5 -
301 SITES
NAAQS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-20. National trend in the
composite average of the
estimated number of exceedances
of the 8-hour carbon monoxide
NAAQS, at both NAMS and all
sites with 95 percent confidence
intervals, 1981-1990.
15
EST. 8-HR EXCEEDANCES
10 -
5 -
" ALL .SITES J30_1_).
NAMS SITES (92)
* * * ^
I I 1 i i i i i i i
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
3-17
-------
The 10-year 1981-90 trend in
national carbon monoxide emission
estimates is shown in Figure 3-21 and in
Table 3-5. The estimates for emissions
from forest fires for the years 1985
through 1989 have been revised
downward about 9 percent, from the
levels reported in last year's report.6
However, this decrease in forest fire
emissions yields only a 1 percent change
in total emissions for 1985-1989. These
estimates show a 22 percent decrease in
total emissions between 1981 and 1990.
Transportation sources accounted for
approximately 71 percent of the total in
1981 and decreased to 63 percent of
total emissions in 1990. Emissions from
highway vehicles decreased 37 percent
during the 1981-90 period, despite a 37
percent increase in vehicle miles of
travel.1 Figure 3-22 contrasts the 10-
year increasing trend in vehicle miles
traveled (VMT) with the declining trend in
carbon monoxide emissions from
highway vehicles. This indicates that the
Federal Motor Vehicle Control
120
CO EMISSIONS, 106 METRIC TONS/YEAR
100 -
80 -
60 -
40 -
SOURCE CATEGORY
TRANSPORTATION
FUEL
COMBUSTION
888 INDUSTRIAL PROCESSES
SOLID WASTE & MISC
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-21. National trend in carbon monoxide
emissions, 1981-1990.
TABLE 3-5. National Carbon Monoxide Emission Estimates, 1981-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
55.4
7.7
5.9
2.1
6.4
77.5
1982
52.9
8.2
4.4
2.0
4.9
72.5
1983
52.4
8.2
4.3
1.9
7.8
74.5
1984
50.6
8.3
4.7
1.9
6.4
71.9
1985
47.9
7.5
4.4
1.9
7.1
68.7
1986
44.6
7.5
4.2
1.8
5.1
63.2
1987
43.3
7.6
4.3
1.8
6.4
63.4
1988
41.2
7.6
4.6
1.7
9.5
64.7
1989
40.0
7.8
4.6
1.7
6.3
60.4
1990
37.6
7.5
4.7
1.7
8.6
60.1
NOTE: The sums of sub-categories may not equal total due to rounding.
3-18
-------
Program (FMVCP) has been effective on
the national scale, with controls more
than offsetting growth during this period.
While there is general agreement
between changes in air quality and
emissions over this 10-year period, it is
worth noting that the emission changes
reflect estimated national totals, while
ambient CO monitors are frequently
located to identify local problems. The
mix of vehicles and the change in vehicle
miles of travel in the area around a
specific CO monitoring site may differ
from the national averages.
3.3.2 Recent CO Trends:
1990
1988-
This section examines ambient CO
changes during the last 3 years, 1988-90
at sites that recorded data in all three
years. Between 1989 and 1990, the
composite average of the second highest
non-overlapping 8-hour average
concentration at 359 sites decreased by
8 percent and by 9 percent at the 104
NAMS sites. The composite average of
the estimated number of exceedances of
the 8-hour CO NAAQS decreased by 43
percent between 1989 and 1990 at these
359 sites. Estimated nationwide CO
emissions decreased less than one
percent between 1989 and 1990. The 7
percent reduction in CO emissions from
highway vehicles was offset by the
increase in forest fire emissions in
Alaska.
Figure 3-23 shows the composite
Regional averages for the 1988-90 time
period. Every Region, except Region I,
has 1990 composite mean levels less
than the corresponding 1988 and 1989
values. These Regional graphs are
primarily intended to depict relative
change. Because the mix of monitoring
sites may vary from one area to another,
this graph is not intended to indicate
Regional differences in concentration
levels.
160
% of 1981 level
140 -
120 -
Highway Vehicles
I CO Emissions VMT
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-22. Comparison of trends in total national
vehicle miles traveled and national highway vehicle
emissions, 1981-1990.
14
12
10
CONCENTRATION, PPM
COMPOSITE AVERAGE
1983 CM 1989 CD 1990
EPA REGION I II IN IV V VI VII VIII IX X
NO. OF SITES 14 28 42 58 55 31 23 16 76 16
Figure 3-23. Regional comparisons of 1988,1989,
1990 composite averages of the second highest
nonoverlapping 8-hour average carbon monoxide
concentrations.
3-19
-------
3.4 TRENDS IN NITROGEN DIOXIDE
Nitrogen dioxide (NO2) is a yellowish brown,
highly reactive gas which is present in urban
atmospheres. The major mechanism for the
formation of NO2 in the atmosphere is the oxidation
of the primary air pollutant, nitric oxide (NO).
Nitrogen oxides play a major role, together with
volatile organic compounds, in the atmospheric
reactions that produce ozone. Nitrogen oxides
form when fuel is burned at high temperatures.
The two major emissions sources are transportation
and stationary fuel combustion sources such as
electric utility and industrial boilers.
Nitrogen dioxide can irritate the lungs, cause
bronchitis and pneumonia, and lower resistance to
respiratory infections. Nitrogen oxides are an
important precursor both to ozone and acidic
precipitation and may affect both terrestrial and
aquatic ecosystems. Los Angeles, CA is the only
urban area that has recorded violations of the
annual average NO2 standard of 0.053 ppm during
the past 10 years.
NO2 is measured using a continuous monitoring
instrument which can collect as many as 8760
hourly observations per year. Only annual means
based on at least 4380 hourly observations were
considered in the trends analyses which follow. A
total of 166 sites were selected for the 10-year
period and 211 sites were selected for the 3-year
data base.
3.4.1 Long-term NO2 Trends 1981-90
The composite average long-term trend for the
nitrogen dioxide mean concentrations at the 166
trend sites and the 42 NAMS sites, is shown in
Figure 3-24. The 95 percent confidence intervals
about the composite means reveal that the 1981 -89
NO2 levels are statistically indistinguishable. The
1990 composite average NO2 level is 8 percent
lower than the 1981 level, and the difference is
statistically significant. A similar trend is seen for
the NAMS sites which, for NO2, are located only in
large urban areas with populations of one million or
0.06
CONCENTRATION, PPM
0.05 -
0.04 -
0.03 -
0.02 -
0.01 -
0.00
NAAQS
A.LLSJTES_(1_66)
NAMS SITES (42)
i i i i i I ( i i
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-24. National trend in the composite annual average nitrogen
dioxide concentration at both NAMS and all sites with 95 percent
confidence intervals, 1981-1990.
3-20
-------
greater. As expected, the composite averages of
the NAMS are higher than those of all sites,
however, they also recorded a statistically
significant decrease of 8 percent during this period.
Long-term trends in NO2 annual average
concentrations are also displayed in Figure 3-25
with the use of boxplots. The middle quartiles for
the years 1981 through 1989 are similar, while a
decrease in levels can be seen in 1990. The upper
percentiles, which generally reflect NO2 annual
mean levels in the Los Angeles metropolitan area,
also show improvement during the last three years.
The lower percentiles show little change, however.
Last year's report presented long-term NO2 annual
mean trends among metropolitan areas of varying
population size. The level of the N02 composite
means varied by metropolitan area size, with the
larger areas recording the higher concentration
levels.
0.07
CONCENTRATION, PPM
0.00
0.06 H
0.05
0.04 -
0.03 -
0.02 -
0.01 -
166 SITES
NAAQS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-25. Boxplot comparisons of trends in annual mean nitrogen
dioxide concentrations at 166 sites, 1981-1990.
3-21
-------
Table 3-6 presents the trend in
estimated nationwide emissions of
nitrogen oxides (NO,). The decreasing
trend in NOX emissions from 1981
through 1983 was reversed in 1984. The
decline in NOX nationwide emissions
between 1985 and 1986 has been
followed by increased NOX emissions in
1987 and 1988. However, total 1990
nitrogen oxides emissions are 6 percent
less than 1981 emissions. Highway
vehicle emissions decreased by 30
percent during this period, while fuel
combustion emissions have recorded
yearly increases during the last 4 years.
Most of the decreases in mobile source
emissions occurred in urban areas, while
much of the increases in stationary
source emissions occurred at facilities
located outside these urbanized areas.
Figure 3-26 shows that the two primary
source categories of nitrogen oxides
emissions are fuel combustion and
transportation, composing 57 percent
and 38 percent, respectively, of total
1990 nitrogen oxides emissions.
30
NO EMISSIONS, 106 METRIC TONS/YEAR
25 -
20 -
15 -
SOURCE CATEGORY
TRANSPORTATION
FUEL COMBUSTION
i INDUSTRIAL PROCESSES
I SOLID WASTE & MISC
1981 1982 1983 1984 1985 1986 1987 1988 1989 1989
Figure 3-26. National trend in nitrogen oxides
emissions, 1981-1990.
TABLE 3-6. National Nitrogen Oxides Emission Estimates, 1981-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
9.9
10.0
0.6
0.1
0.2
20.9
1982
9.4
9.8
0.5
0.1
0.1
20.0
1983
8.9
9.6
0.5
0.1
0.2
19.4
1984
8.8
10.2
0.6
0.1
0.2
19.8
1985
8.9
10.2
0.6
0.1
0.2
19.9
1986
8.3
10.0
0.6
0.1
0.2
19.1
1987
8.1
10.5
0.6
0.1
0.2
19.4
1988
8.1
10.9
0.6
0.1
0.3
20.0
1989
7.9
11.1
0.6
0.1
0.2
19.8
1990
7.5
11.2
0.6
0.1
0.3
19.6
NOTE: The sums of sub-categories may not equal total due to rounding.
3-22
-------
3.4L2 Recent NO2 Trends: 1988-1990
Between 1989 and 1990, the
composite annual mean NO2
concentration at 211 sites, with complete
data during the last three years,
decreased by 6 percent, the largest
decline in recent years. At the subset of
47 NAMS, the composite mean
concentration decreased 5 percent
between 1989 and 1990. Nationwide
emissions of nitrogen oxides are
estimated to have decreased 1 percent
between 1989 and 1990.
Regional trends in the composite
average NO2 concentrations for the years
1988-90 are displayed in Figure 3-27
with bar graphs. Region X, which did not
have any NO2 sites meeting the 3-year
data completeness and continuity criteria,
is not shown. All of the remaining nine
Regions have 1990 composite average
NO2 annual mean concentrations that are
lower than the previous two years 1988
and 1989. Seven of the nine Regions
have 1989 composite mean
concentrations which are lower than the
corresponding 1988 levels. These
Regional graphs are primarily intended to
depict relative change. Because the mix
of monitoring sites may vary from one
area to another, this graph is not
intended to indicate Regional differences
in absolute concentration levels.
CONCENTRATION, PPM
0035
0.030
0020
0010
COMPOSITE AVERAGE
1968 B 1999 CD 1990
EPA REGION
NO OF SITES
15
13
38
IV
17
V
25
VI
22
VII
12
VIII
9
IX
60
Figure 3-27. Regional comparisons of 1988,1989,
1990 composite averages of the annual mean
nitrogen dioxide concentrations.
3-23
-------
3.5 TRENDS IN OZONE
Ozone (03) is a photochemical oxidant and the
major component of smog. While ozone in the
upper atmosphere is beneficial to life by shielding
the earth from harmful ultraviolet radiation from the
sun, high concentrations of ozone at ground level
are a major health and environmental concern.
Ozone is not emitted directly into the air but is
formed through complex chemical reactions
between precursor emissions of volatile organic
compounds and nitrogen oxides in the presence of
sunlight. These reactions are stimulated by
sunlight and temperature so that peak ozone levels
occur typically during the warmer times of the year.
Both volatile organic compounds and nitrogen
oxides are emitted by transportation and industrial
sources. Volatile organic compounds are emitted
from sources as diverse as autos, chemical
manufacturing, and dry cleaners, paint shops and
other sources using solvents. Nitrogen oxides
emissions were discussed in Section 3.4.
The reactivity of ozone causes health problems
because it tends to break down biological tissues
and cells. Recent scientific evidence indicates that
ambient levels of ozone not only affect people with
impaired respiratory systems, such as asthmatics,
but healthy adults and children, as well. Exposure
to ozone for several hours at relatively low
concentrations has been found to significantly
reduce lung function in normal, healthy people
during exercise. This decrease in lung function
generally is accompanied by symptoms including
chest pain, coughing, sneezing and pulmonary
congestion.
The O3 NAAQS is defined in terms of the daily
maximum, that is, the highest hourly average for
the day, and it specifies that the expected number
of days per year with values greater than 0.12 ppm
should not be greater than one. Both the annual
second highest daily maximum and the number of
daily exceedances during the ozone season are
considered in this analysis. The strong seasonally
of ozone levels makes it possible for areas to limit
their ozone monitoring to a certain portion of the
year, termed the ozone season. The length of the
ozone season varies from one area of the country
to another. May through October is typical but
States in the south and southwest may monitor the
entire year. Northern States would have shorter
ozone seasons such as May through September for
0.18
CONCENTRATION, PPM
0.16 -
0.14 -
0.12
0.10 -
0.08 -
0.06 -
0.04 -
0.02
0.00
=**
NAAQS
SITES (471J
NAMS SITES (194)
I I I I I I I I I I
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-28. National trend in the composite average of the second
highest maximum 1-hour ozone concentration at both NAMS and all
sites with 95 percent confidence intervals, 1981-1990.
3-24
-------
North Dakota. This analysis uses these ozone
seasons to ensure that the data completeness
requirements apply to the relevant portions of the
year.
The trends site selection process, discussed in
Section 2.1, resulted in 471 sites being selected for
the 1981-90 period, an increase of 40 sites (or 9%)
from the 1980-89 trends data base. A total of 590
sites are included in the 1988-90 data base. The
NAMS compose 194 of the long-term trends sites
and 206 of the sites in the 3-year data base.
3.5.1 Long-term O3 Trends: 1981-90
Figure 3-28 displays the 10-year composite
average trend for the second highest day during
the ozone season for the 471 trends sites and the
subset of 194 NAMS sites. The 1990 composite
average for the 471 trend sites is 10 percent lower
than the 1981 average and 9 percent lower for the
subset of 194 NAMS. These 1990 values are the
lowest composite averages of the past ten years.
The 1990 composite average is significantly less
than the 1988 composite mean, which is the
second highest average (1983 was the highest)
during this 10-year period. The relatively high
ozone concentrations in both 1983 and 1988 are
likely attributed in part to hot, dry, stagnant
conditions in some areas of the country that were
more conducive to ozone formation than other
years. Peak ozone concentrations typically occur
during hot, dry, stagnant summertime conditions
(high temperature and strong solar insolation).10'11
Previous reports have compared the regional
variability in meteorological parameters such as
maximum daily temperature and precipitation with
the variability in peak ozone concentrations.6
The interpretation of recent ozone trends is
difficult due to the confounding factors of
meteorology and emission changes. Just as the
increase in 1988 is attributed in part to
meteorological conditions, the 1989 decrease is
0.30
CONCENTRATION, PPM
0.25 -1
0.20 -
0.15
0.10
0.05
0.00
471 SITES
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-29. Boxplot comparisons of trends in annual second highest
daily maximum 1-hour ozone concentration at 471 sites, 1981-1990.
3-25
-------
likely due, in part, to meteorological conditions
being less favorable for ozone formation in 1989
than in 1988. This pattern was followed by
summer 1990 which nationally was warmer and
drier than the long-term climatological means.
Also, precursor emissions of nitrogen oxides and
volatile organic compound emissions from highway
vehicles have decreased in urban areas. The
volatility of gasoline was reduced by new
regulations which lowered national average
summertime Reid Vapor Pressure (RVP) in regular
unleaded gasoline from 10.0 to 8.9 pounds per
square inch (psi) between 1988 and 1989.12-13'14
RVP was reduced an additional 3 percent between
1989 and 1990.15
The inter-site variability of the annual second
highest daily maximum concentrations for the 471
site data base is displayed in Figure 3-29. The
years 1983 and 1988 values are similarly high,
while the remaining years in the 1981-90 period are
generally lower, with 1990 being the lowest, on
average. The distribution of second daily maximum
1-hour concentrations in 1990 is similar to that
recorded in 1989 and 1986. Figure 3-30 depicts
the 1981-90 trend for the composite average
number of ozone exceedances. This statistic is
adjusted for missing data, and it reflects the
number of days that the ozone standard is
exceeded during the ozone season. Since 1981,
the expected number of exceedances decreased 51
percent for both the 471 long-term trend sites and
the subset of 194 NAMS. As with the second
maximum, the 1983 and 1988 values are higher
than the other years in the 1981-90 period. The
1989 and 1990 levels are significantly lower than all
the previous years.
Table 3-7 and Figure 3-31 display the 1981-90
emission trends for volatile organic compounds
(VOC) which, along with nitrogen oxides, are
involved in the atmospheric chemical and physical
processes that result in the formation of O3. Total
VOC emissions are estimated to have decreased
12 percent between 1981 and 1990. Between
1981 and 1990, VOC emissions from highway
vehicles decreased 34 percent, despite a 37
percent increase in vehicle miles of travel during
this time period (see Figure 3-21). Previously, VOC
emissions from highway vehicles were estimated
using nationwide annual temperatures and
nationwide average RVP. Starting with last year's
report, these VOC estimates for the 10-year period
are now based on statewide average monthly
temperatures and statewide average RVP.
15
NO. OF EXCEEDANCES
10 -
5 -
ALL SJT_ES__(4_71__i_
NAMS SITES (194
u i r i i i i i i i i
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-30. National trend in the estimated number of daily
exceedances of the ozone NAAQS in the ozone season at both NAMS
and all sites with 95 percent confidence intervals, 1981-1990.
3-26
-------
The highway vehicle emission estimates
for the years 1989 and 1990 were
calculated using 1988 RVP data, the last
year for which statewide figures were
available. Thus, the reductions in
average summertime RVP levels that
have occurred since 1988 are not
reflected in the emissions totals for
transportation sources. The increase in
VOC emissions between 1989 and 1990
is due to VOC emissions from forest fires
in Alaska, which are not likely to have
contributed to ozone formation in urban
areas. These VOC emissions estimates
represent annual totals. While these are
the best national numbers now available,
ozone is predominately a warm weather
problem and seasonal emission trends
would be preferable.
VOC EMISSIONS, 10s METRIC TONS/TEAR
JU
25 -
20 -
l"-\
SOURCE CATEGORY
TRANSPORTATION S88 INDUSTRIAL PROCESSES
I FUEL COMBUSTION SOLID WASTE & MISC
^ -_
' __
I
10
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-31. National trend in volatile organic
compound emissions, 1981-1990.
TABLE 3-7. National Volatile Organic Compound Emission Estimates, 1981-1990
(million metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
8.9
1.0
8.3
0.7
2.4
21.3
1982
8.3
1.0
7.5
0.6
2.1
19.6
1983
8.2
1.0
7.9
0.6
2.7
20.4
1984
8.1
1.0
8.9
0.6
2.6
21.2
1985
7.6
0.9
8.5
0.6
2.5
20.1
1986
7.2
0.9
8.0
0.6
2.2
19.0
1987
7.1
0.9
8.3
0.6
2.4
19.3
1988
6.9
0.9
8.1
0.6
2.9
19.4
1989
6.4
0.9
8.1
0.6
2.5
18.5
1990
6.4
0.9
8.1
0.6
2.7
18.7
NOTE: The sums of sub-categor
es may not equal total due to rounding.
3-27
-------
3.5.2 Recent O3 Trends: 1988-1990
This section discusses ambient O3 changes
during the 3-year time period 1988-90. Using this
3-year period permits the use of a larger data base
of 590 sites, compared to 471 for the 10-year
period.
Nationally, 1988 was the third hottest summer
since 1931, with hot, dry meteorological conditions
experienced in much of the Eastern U.S. during the
summer.16 In the East, the period from January
through July 1989 was among the wettest on
record in nine states.17 Summer 1990 temperature
averaged across the nation was above the
long-term mean and ranks as the 15th warmest
summer on record since 1895.18 Spatially
averaged 1990 precipitation was below the
long-term mean and ranks as the 29th driest
summer. Regionally, the Central and East North
Central had average summer temperatures, with
other regions above normal. During the summer of
1990, the South and Southeast were unusually dry,
while the Northeast, East, North Central and
Northwest Regions had above average
precipitation.18 Also, 1990 average RVP decreased
3 percent from summer 1989 levels, and 1989 was
11 percent lower than 1988 average
RVP.15
indicates, the largest decreases were recorded in
the northeastern states, composing EPA Regions I
through III. Except for the northwest (Region X)
when 1990 was the peak year, every region
recorded its highest composite mean for this 3-year
period in 1988. Five Regions had 1990 composite
averages lower than 1989 levels, while the
remaining five Regions had 1990 composite
averages lower than 1988 and 1989.
These Regional graphs are primarily intended to
depict relative change. Because the mix of
monitoring sites may vary from one area to another,
this graph is not intended to indicate Regional
differences in absolute concentration levels.
Previous reports have presented a preliminary
estimate of the trend in the composite average of
the second highest daily maximum 1 -hour ozone
concentration. These estimates were based on
preliminary, unvalidated data that were reported to
EPA about 3 months ahead of the schedule
typically required for quality assurance and data
submittal. The accelerated printing schedule for
this year's report precluded an advanced estimate
for 1991, because sufficient 1991 data were not
available as the report went to press.
Between 1989 and 1990, composite
mean ozone concentrations decreased 1
percent at the 590 sites and by less than
1 percent at the subset of 206 NAMS.
The 1990 composite average is 17
percent lower than the 1988 composite
mean for these 590 sites. Between 1989
and 1990, the composite average of the
number of estimated exceedances of the
ozone standard decreased by 17 percent
at the 590 sites, and 14 percent at the
206 NAMS. Because of forest fires in
Alaska, nationwide VOC emissions
increased 1 percent between 1989 and
1990. There was a 4 percent decrease
between 1988 and 1990.
The composite average of the
second daily maximum concentrations
decreased in every Region of the nation
between 1988 and 1989. As Figure 3-32
0.20
0.16
0.12
0.08
0.04
CONCENTRATION, PPM
COMPOSITE AVERAGE
1988 EM 1989 I 1 1990
EPA REGION
NO. OF SITES
I
38
32
72
IV
91
V
131
VI
58
VII
31
VIII
16
IX
',14
Figure 3-32. Regional comparisons of the 1988,1989,
1990 composite averages of the second-highest daily
1-hour ozone concentrations.
3-28
-------
3.6 TRENDS IN LEAD
Lead (Pb) gasoline additives, nonferrous
smelters and battery plants are the most significant
contributors to atmospheric Pb emissions.
Transportation sources in 1990 contributed 31
percent of the annual emissions, down substantially
from 73 percent in 1985. Total lead emissions
from all sources dropped from 20.1 x 103 metric
tons in 1985 to 7.2 x 103 and 7.1 x 103 metric tons,
respectively in 1989 and 1990. The decrease in
lead emissions from highway vehicles accounts for
essentially all of this drop. The reasons for this
drop are noted below.
Two air pollution control programs implemented
by EPA before promulgation of the Pb standard in
October 197819 have resulted in lower ambient Pb
levels. First, regulations issued in the early 1970s
required gradual reduction of the Pb content of all
gasoline over a period of many years. More
recently, the Pb content of the leaded gasoline pool
was reduced from an average of 1.0 grams/gallon
to 0.5 grams/gallon on July 1, 1985 and still further
to 0.1 grams/gallon on January 1,1986. Second,
as part of EPA's overall automotive emission
control program, unleaded gasoline was introduced
in 1975 for use in automobiles equipped with
catalytic control devices. These devices reduce
emissions of carbon monoxide, volatile organics
and nitrogen oxides. In 1990, unleaded gasoline
sales accounted for 89 percent of the total gasoline
market. These programs have essentially
eliminated violations of the lead standard in urban
areas, except in those areas with lead point
sources. Programs are also in place to control Pb
emissions from stationary point sources. Pb
emissions from stationary sources have been
substantially reduced by control programs oriented
toward attainment of the paniculate matter and Pb
ambient standards, however, significant ambient
problems still remain around some lead point
sources. Lead emissions in 1990 from industrial
sources, e.g. primary and secondary lead smelters,
dropped by more than one-half from levels reported
in the late 70s. Emissions of lead from solid waste
disposal are down 45 percent since the late 70s.
In 1990, emissions from solid waste disposal and
industrial processes and transportation were each
estimated to be 2.2 x 103 metric tons. The overall
effect of these three control programs has been a
major reduction in the amount of Pb in the ambient
air. In addition to the above Pb pollution reduction
activities, additional reductions in Pb are anticipated
as result of the Agency's Multi-media Lead Strategy
issued in February, 1991.20 The goal of the
Agency's Lead Strategy is to reduce Pb exposures
to the fullest extent practicable.
Exposure to lead can occur through multiple
pathways, including inhalation of air, diet and
ingestion of lead in food, water, soil or dust.
Excessive lead exposure can cause seizures,
mental retardation and/or behavioral disorders.
Fetuses, infants and children are especially
susceptible to low doses of lead, resulting in central
nervous system damage. Recent studies have also
shown that lead may be a factor in high blood
pressure and subsequent heart disease in
middle-aged white males.
3.6.1 Long-term Pb Trends: 1981-90
Early trend analyses of ambient Pb data21'22
were based almost exclusively on National Air
Surveillance Network (NASN) sites. These sites
were established in the 1960's to monitor ambient
air quality levels of TSP and associated trace
metals, including Pb. The sites were predominantly
located in the central business districts of larger
American cities. In September 1981, ambient Pb
monitoring regulations were promulgated.23 The
siting criteria in the regulations resulted in finding
many of the old historic TSP monitoring sites
unsuitable for the measurement of ambient Pb
concentrations and many of the earlier sites were
moved or discontinued.
As with the other pollutants, the sites selected
for the long-term trend analysis had to satisfy
annual data completeness criteria of at least 8 out
of 10 years of data in the 1981 to 1990 period. A
year was included as "valid" if at least 3 of the 4
quarterly averages were available. As in last year's
report, composite lead data, i.e., individual 24-hour
observations are composited together by month or
quarter and a single analysis made, are being used
in the trend analysis. Fifteen sites qualified for the
10-year trend because of the addition of composite
3-29
-------
data. A total of 202 urban-oriented sites,
from 38 States and Puerto Rico, met the
data completeness criteria. Sixty-four of
these sites were NAMS, the largest
number of lead NAMS sites to qualify for
the 10-year trends. Thirty-two (16
percent) of the 202 trend sites were
located in the State of California, thus
this state is over-represented in the
sample of sites satisfying the long-term
trend criteria. However, the lead trend at
the California sites was almost identical
to the trend at the non-California sites;
so that these sites did not distort the
overall trends. Other states with 10 or
more trend sites included: Illinois (23),
Kansas (16), Pennsylvania (11), West
Virginia (10) and Texas (10). Again, the
Pb trend in each of these states was
very similar to the national trend. Sites
that were located near lead point sources
such as primary and secondary lead
smelters were excluded from the urban
trend analysis, because the magnitude of
the levels at these sources could mask
the underlying urban trends. Trends at
lead point source oriented sites will be
discussed separately in the next section.
The means of the composite
maximum quarterly averages and their
respective 95 percent confidence
intervals are shown in Figure 3-33 for
both the 202 urban sites and 64 NAMS
sites (1981-1990). There was an 85
percent (1981-90) decrease in the
average for the 202 urban sites. Lead
emissions over this 10-year period also
decreased. There was an 87 percent
decrease in total lead emissions and a
95 percent decrease in lead emissions
from transportation sources. The
confidence intervals for all sites indicate
that the 1985-90 averages are
significantly less than all averages from
preceding years. Because of the smaller
number (64) of NAMS sites with at least
8 years of data, the confidence intervals
are wider. However, the 1985-90 NAMS
averages are still significantly different
from all NAMS averages before 1985. It
1.8
CONCENTRATION, UG/M3
1.2 -
1 -
0.8 -
0.6 -
0.4 -
0.2 -
0
NAAQS
ALL SJTESJ202:)_
MAMS SITES (64 )
i i i i i i i i i i
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-33. National trend in the composite
average of the maximum quarterly average lead
concentration at both NAMS and all sites with 95
percent confidence intervals, 1981-1990.
CONCENTRATION, UG/M3
2.5 -
2 -
1.5
1 -
0.5 -
' POINT SOURCE SITES (33) a URBAN SITES (202)
-NAAQS
1981
i I l i I l \ \ I
1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-34. Comparison of national trend in the
composite average of the maximum quarterly
average lead concentrations at urban and point-
source oriented sites, 1981-1990.
3-30
-------
is interesting to note that the composite average
lead concentration at the NAMS sites in 1990 is
essentially the same (0.068 u.g/m3) as the "all sites"
average; whereas in the early 1980's the averages
of the NAMS sites were significantly higher. Figure
3-34 shows the trend in average lead
concentrations for the urban-oriented sites and for
33 point-source oriented sites which met the 10-
year data completeness criteria. Composite
average ambient lead concentrations at the
point-source oriented sites, located near industrial
sources of lead, e.g. smelters, battery plants,
improved 72%, compared to 85% at the urban
oriented sites. The average at the point-source
oriented sites dropped in magnitude from 2.2 to 0.6
u,g/m3, a 1.6 u.g/m3 difference; whereas, the
average at the urban sites dropped only from 0.5 to
0.1 u.g/m3. This improvement at the point-source
oriented sites reflects both industrial and
automotive lead emission controls, but in some
cases, the industrial source reductions are because
of plant shutdowns. However, there are still
several urban areas where significant Pb problems
persist. The 10 MSAs shown in Table 4-3 that are
above the lead NAAQS in 1990 are all due to lead
point sources. These MSAs are Birmingham, AL;
Columbus, GA-AL; Dallas, TX; Indianapolis, IN;
Memphis, TN; Minneapolis, MN; Nashville, TN;
CONCENTRATION, UG/M3
Omaha, NE-IA; Philadelphia, PA; and St Louis,
MO-IL. Figure 3-35 shows boxplot comparisons of
the maximum quarterly average Pb concentrations
at the 202 urban-oriented Pb trend sites (1981-90).
This figure shows the dramatic improvement in
ambient Pb concentrations over the entire
distribution of trend sites. As with the composite
average concentration since 1981, most of the
percentiles also show a monotonically decreasing
pattern. The 202 urban-oriented sites that qualified
for the 1981-90 period, when compared to the 189
sites for 1980-89 and the 139 sites for 1979-88
period,6'16 indicate a substantial expansion of the
10-year trends data base.
The trend in total lead emissions is shown in
Figure 3-36. Table 3-8 summarizes the Pb
emissions data as well. The 1981-90 drop in total
Pb emissions was 87 percent. Lead emissions in
the transportation category account for most of this
drop. The trend in Pb emissions from non-
transportation sources is shown in Figure 3-37.
This figure shows the trend in two categories:
industrial and the total of all non-transportation
sources. Lead emissions from both of these
categories show a drop early in the time period with
a leveling off thereafter. The drop in the non-
transportation emissions is due to decreases in
1.5
1 -
0.5 -
202 SITES
NAAQS
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-35. Boxplot comparisons of trends in maximum quarterly
average lead concentrations at 202 sites, 1981-1990.
3-31
-------
lead from all categories as shown in
Table 3-8. This compares with the 85
percent decrease (1981-90) in ambient
lead concentrations. The drop in Pb
consumption and subsequent Pb
emissions since 1981 was brought about
by the increased use of unleaded
gasoline in catalyst-equipped cars and
the reduced Pb content in leaded
gasoline. The results of these actions in
1990 amounted to a 65 percent reduction
nationwide in total Pb emissions from
1985 levels. As noted previously,
unleaded gasoline represented 89
percent of 1990 total gasoline sales.
Although the good agreement among the
trend in lead consumption, emissions
and ambient levels is based upon a
limited geographical sample, it does
show that ambient urban Pb levels are
responding to the drop in lead emissions.
80
LEAD EMISSIONS, 103 METRIC TONS/VEAR
60 -
40 -
20 -
SOURCE CATEGORY
TRANSPORTATION
BB FUEL
COMBUSTION
i INDUSTRIAL PROCESSES
I SOLID WASTE
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Figure 3-36. National trend in lead emissions, 1981-
1990.
TABLE 3-8. National Lead Emission Estimates, 1981-1990
(thousand metric tons/year)
SOURCE
CATEGORY
Transportation
Fuel
Combustion
Industrial
Processes
Solid Waste
Miscellaneous
TOTAL
1981
46.5
2.8
3.0
3.7
0.0
56.0
1982
47.0
1.7
2.7
3.1
0.0
54.5
1983
40.8
0.6
2.4
2.7
0.0
46.6
1984
34.7
0.5
2.3
2.7
0.0
40.2
1985
14.7
0.5
2.3
2.6
0.0
20.1
1986
3.5
0.5
1.9
2.6
0.0
8.4
1987
3.0
0.5
1.9
2.6
0.0
8.0
1988
2.6
0.5
2.0
2.5
0.0
7.6
1989
2.2
0.5
2.3
2.3
0.0
7.2
1990
2.2
0.5
2.2
2.2
0.0
7.1
NOTE: The sums of sub-categories may not equal total due to rounding.
3-32
-------
In Canada a very similar trend in ambient
lead concentrations has been observed.
Declines in composite average lead
concentrations of 86 percent were found
for the 1980-89 time period.24 Also,
average ambient Pb concentrations in
Tokyo, Japan25 have dropped from around
1.0 ng/m3 in 1967 to approximately 0.1
(ig/m3 in 1985 - a 90% improvement.
3.6.2 Recent Pb Trends: 1988-90
Ambient Pb trends were also studied
over the shorter period 1988-90. A total
of 229 urban sites from 37 States and
Puerto Rico met the data requirement
that a site have all 3 years with data. In
recent years, the number of lead sites
has dropped because of the elimination
of some TSP monitors from state and
local air monitoring programs. Some
monitors were eliminated due to the
change in the paniculate matter standard
from TSP to PM-10 while others were
discontinued because of the very low
lead concentrations measured in many
urban locations. Although some further
attrition may occur, the core network of
NAMS lead sites together with
supplementary State and local sites
should be sufficient to access national
ambient lead trends. The 3-year data
base (1988-90) showed an improvement
of 26 percent in composite average
urban Pb concentrations. The 1988 and
1990 lead averages respectively were
0.087 and 0.064 u.g/m3, a 26 percent
improvement. This corresponds to
reductions in total Pb emissions of 7
percent and a reduction of 15 percent in
lead emissions from transportation
sources. Most of this decrease in total
nationwide Pb emissions was due once
again to the decrease in automotive Pb
emissions. Even this larger group of
sites was disproportionately weighted by
sites in California, Illinois, Kansas and
Texas. These States had about 35
percent of the 229 sites represented.
However, the percent changes in
Tons Gigagrams
11,000-
8,800-
6,600-
4,400-
2,200-
0-
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
Year
| Solid Waste Disposal | Industrial Processes Q Industrial Boilers [J Electric Utilities
Figure 3-37. National trend in emissions of lead
excluding transportation sources, 1981 - 1990.
CONCENTRATION, UG/M3
1.4
1.2
1
0.8
0.6
0.4
0.2
COMPOSITE AVERAGE
1986 B 1989 CH 1990
EPA REGION I II III IV V VI VII VIII IX X
NO OF SITES 12 15 29 32 41 29 24 7 33 7
Figure 3-38. Regional comparison of the 1988,1989,
1990 composite average of the maximum quarterly
average lead concentrations.
3-33
-------
1988-90 average Pb concentrations for these four
States were very similar to the percent change for
the remaining sites, thus the contributions of these
sites did not distort the national trends. Although
urban lead concentrations continue to decline
consistently, there are indications that the rate of
the decline has slowed down. Clearly in some
areas, urban lead levels are so low, that further
improvements have become difficult.
Indeed, as will be shown later, all sections of
the country are showing declines in average lead
concentrations. Sixty-three (63) point source
oriented sites showed an average drop of 36
percent over the 1988-90 time period. Thus, this
decrease in ambient lead concentrations near lead
point sources has been slightly more pronounced
than in urban areas. The average lead levels at
these sites are much higher here than at the urban
sites. The 1989 and 1990 lead point source
averages were 0.79 and 0.80 u.g/m3 respectively.
The larger sample of sites represented in the
3-year trends (1988-90) will be used to compare
the most recent individual yearly averages.
However, for the 10-year time period the largest
single year drop in average lead concentrations, 42
percent, occurs as expected between 1985 and
1986, because of the shift of the lead content in
leaded gasoline. The 1990 composite average
lead concentrations show the more modest decline
of 12 percent from 1989 levels. The 10-year data
base showed a 5 percent decrease in average lead
concentrations from 1989 to 1990. There has been
no change in estimated Pb emissions for the
transportation category between 1989 and 1990.
Although, VMT increased 1 percent between 1989
and 1990. The Pb emissions trend is expected to
continue downward, but at a slower rate, primarily
because the leaded gasoline market will continue
to shrink. Between 1989 and 1990, total lead
emissions decreased 1 percent, while emissions
from transportation sources remained unchanged.
Some major petroleum companies have
discontinued refining leaded gasoline because of
the dwindling market, so that in the future the
consumer will find it more difficult to purchase
regular leaded gasoline. Figure 3-38 shows
1988, 1989 and 1990 composite average Pb
concentrations, by EPA Region. Once again the
larger more representative 3-year data base of 229
sites was used for this comparison. The number of
sites varies dramatically by Region from 7 in
Regions VIII and X to 41 in Region V. In all
Regions, except Region IV, there is a decrease in
average Pb urban concentrations between 1988
and 1990. These results confirm that average Pb
concentrations in urban areas are continuing to
decrease throughout the country, which is exactly
what is to be expected because of the national air
pollution control program in place for Pb.
3-34
-------
3.7 REFERENCES
1. National Air Pollutant Emission Estimates.
1940-1990. EPA-450/4-91-026, U. S. Environmental
Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC,
November 1991.
2. National Air Quality and Emissions Trends
Report. 1983. EPA-450/4-84-029,
U. S. Environmental Protection Agency, Office of
Air Quality Planning and Standards, Research
Triangle Park, NC, April 1985.
3. National Air Quality and Emissions Trends
Report. 1985. EPA-450/4-87-001,
U. S. Environmental Protection Agency, Office of
Air Quality Planning and Standards, Research
Triangle Park, NC, February 1987.
4. N. H. Frank, "Nationwide Trends in Total
Suspended Paniculate Matter and Associated
Changes in the Measurement Process", presented
at the Air Pollution Control Association, American
Society for Quality Control Specialty Conference
on Quality Assurance in Air Pollution
Measurements, Boulder, CO, October 1984.
5. Written communication from Thomas R.
Mauser, Environmental Monitoring Systems
Laboratory, U. S. Environmental Protection Agency,
Research Triangle Park, NC, to Richard G.
Rhoads, Monitoring and Data Analysis Division, U.
S. Environmental Protection Agency, Research
Triangle Park, NC, January 11, 1984.
6. National Air Quality and Emissions Trends
Report. 1989. EPA-450/4-91-003, U. S.
Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle
Park, NC, February 1991.
7. E. Olaguer, "Precipitation Data in Support of
EPA 1990 Trends Report", EPA Contract No.
68D00119, IT Air Quality Services, Durham, NC,
August 1991.
8. Proposed Decision Not to Revise the
National Ambient Air Quality Standards for Sulfur
Oxides (Sulfur Dioxide). 53 FR 14926, April 26,
1988.
9. Aerometric Information Retrieval System
(AIRS), AIRS Facility Subsystem, U. S.
Environmental Protection Agency, Research
Triangle Park, NC, September 1991.
10. D.J. Kolaz and R.L. Swinford, "How to
Remove the influence of meteorology from the
Chicago Areas Ozone Trend," presented at the
83rd Annual AWMA Meeting, Pittsburgh, PA, June
1990.
11. Use of Meteorological Data in Air Quality
Trend Analysis. EPA-450/3-78-024, U.S.
Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle
Park, NC, May 1978.
12. Volatility Regulations for Gasoline and
Alcohol Blends Sold in Calendar Years 1989 and
Beyond. 54 FR 11868, March 22, 1989.
13. National Fuel Survey: Motor Gasoline -
Summer 1988. Motor Vehicle Manufacturers
Association, Washington, D.C., 1988.
14. National Fuel Survey: Gasoline and Diesel
Fuel -Summer 1989. Motor Vehicle Manufacturers
Association, Washington, D.C., 1989.
15. National Fuel Survey: Motor Gasoline -
Summer 1990. Motor Vehicle Manufacturers
Association, Washington, D.C., 1990.
16. National Air Quality and Emissions Trends
Report. 1988. EPA-450/4-90-002, U. S.
Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle
Park, NC, March 1990.
17. R. H. Heim, Jr., "United States July Climate
in Historical Perspective", National Climatic Data
Center, NOAA, Asheville, NC, August 1989.
3-35
-------
18. R. H. Heim, Jr., "United States Summer
Climate in Historical Perspective", National Climatic
Data Center, NOAA, Asheville, NC, August 1990.
19. National Primary and Secondary Ambient Air
Quality Standards for Lead. 43 FR 46246, October
5, 1978.
20. Memorandum. Joseph S. Carra to Office
Directors Lead Committee. Final Agency Lead
Strategy. February 26, 1991.
21. R. B. Faoro and T. B. McMullen, National
Trends in Trace Metals Ambient Air. 1965-1974,
EPA-450/I-77-003, U. S. Environmental Protection
Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC, February
1977.
22. W. Hunt, "Experimental Design in Air Quality
Management," Andrews Memorial Technical
Supplement, American Society for Quality Control,
Milwaukee, Wl, 1984.
23. Ambient Air Quality Surveillance, 46 FR
44159, Septembers, 1981.
24. T. Furmanczyk, Environment Canada,
personal communication to R. Faoro, U.S.
Environmental Protection Agency, Nov. 6, 1990.
25. Hazardous Air Pollutants Project Country
Report of Japan, Organization For Economic Co-
operation and Development, Paris, France, March,
1991.
3-36
-------
4. AIR QUALITY STATUS OF METROPOLITAN AREAS, 1990
This chapter provides general information
on the current air quality status of metropolitan
areas1 within the United States. Four different
summaries are presented in the following
sections. First, maps depicting the areas
designated nonattainment for the National
Ambient Air Quality Standards (NAAQS) for
particulate matter (PM-10), sulfur dioxide (S02),
carbon monoxide (CO), ozone (O3) and lead
(Pb) are presented. Next, an estimate is
provided of the number of people living in
counties which did not meet the NAAQS based
on only 1990 air quality data. (Note that
nonattainment designations typically involve
multi-year periods.) Third, pollutant-specific
maps are presented to provide the reader with
a geographical view of how peak 1990 air
quality levels varied throughout the 90 largest
Metropolitan Statistical Areas (MSAs) in the
continental United States. Finally, the peak
pollutant-specific statistics are listed for each
MSA with 1990 air quality monitoring data.
Table 4-1. Nonattainment Areas
for NAAQS Pollutants as of
October 1991
Pollutant
Particulate Matter (PM-10)
Sulfur Dioxide (SO2)
Carbon Monoxide (CO)
Nitrogen Dioxide (NO2)
Ozone (O3)
Lead (Pb)
Number of
Nonattainment
Areas
70
50
42*
1
98*
12
4.1 Nonattainment Areas
This section presents maps indicating the
nonattainment areas for each of the six NAAQS
pollutants, except nitrogen dioxide. Because
Los Angeles, CA is the only area currently not
meeting the N02 standard, a map is not
presented for this pollutant. The nonattainment
designation is the result of a formal process but,
for the purposes of this section, may be viewed
as simply indicating areas that do not meet a
specific air quality standard. The Clean Air Act
Amendments (CAAA) of 1990 further classify
ozone and carbon monoxide nonattainment
areas based upon the magnitude of the
problem. Depending on their particular
nonattainment classification, the area must
adopt, at a minimum, certain air pollution
reduction measures. The classification of an
area also determines when the area must reach
attainment. The technical details underlying
these classifications are discussed in Part 81 of
the Code of Federal Regulations (Federal
Register, November 6, 1991).2 Table 4-1
displays the number of nonattainment areas for
each pollutant.
Figures 4-1 and 4-2 display the
nonattainment areas for ozone and carbon
monoxide, respectively. These maps also
indicate the CAAA classifications which are
based upon the design value, a concentration
indicating the magnitude of the problem. To
facilitate the identification of sub-county CO
nonattainment areas, the county boundaries of
these areas are highlighted in light-blue. States
containing nonattainment areas are shown in
yellow. Unclassified areas and transitional
ozone areas are not displayed on the 03 and
CO maps. Figures 4-3 through 4-5 show the
nonattainment areas for PM-10, SO2 and Pb,
respectively. States containing nonattainment
areas are highlighted with solid color shading.
Unclassified areas and transitional
areas are not included in the totals.
4-1
-------
0)
c
o
N
O
0)
CO
c
o
T3
O
"5
O)
'55
o>
0
(0
CO
o
4-2
-------
0
x
o
c
o
o
-Q
c5
o
0)
c
5
OB
c
o
08
co
0)
o
(0
(0
0)
4-3
-------
4-4
-------
CM
O
DC
O
LU
O
UJ
g
0)
LU
O
CO
UJ
DC
s
CO
tr
2
CO
35
01
Q
111
CO
LU
Q
4-5
-------
Q
<
HI
cc
o
Li.
H
LU
^^^H
<
<
O
UJ
UJ
Q
CO
UJ
DC
o
<
LLJ
_J
CC
2
LU
CC
UJ
p
Q
III
p
to
111
I-
2
4»
4-6
-------
4L2 Population Estimates For Counties Not Meeting NAAQS, 1990
Figure 4-6 provides an estimate of the number of
people living in counties in which the levels of the
pollutant-specific primary health NAAQS were not
met by measured air quality in 1990. These
estimates use a single-year interpretation of the
NAAQS to indicate the current extent of the problem
for each pollutant. Table 2-1 lists the selected air
quality statistics and their associated NAAQS.
Figure 4-6 clearly demonstrates that O3 was the
most pervasive air pollution problem in 1990 for the
United States with an estimated 62.9 million people
living in counties which did not meet the 03
standard. This estimate is slightly lower than last
year's 1989 estimate of 66.7 million people.
However, the population estimates for the past 2
years are substantially lower than the 112 million
people living in areas which did not meet the ozone
NAAQS in 1988. This large decrease is likely due
in part to meteorological conditions in 1988 being
more conducive to ozone formation than recent
years (recall the hot, dry summer in the eastern
U.S.), and to new and ongoing emission control
programs. Between 1988 and 1989, implementation
of gasoline volatility regulations lowered the average
Reid Vapor Pressure (RVP) of regular unleaded
gasoline from 10.0 to 8.9 pounds per square inch
(psi). RVP was reduced an additional 3 percent
between 1989 and 1990. Carbon monoxide follows
with 21.7 million people; PM-10 with 18.8 million
people; NO2 with 8.5 million people; Pb with 5.3
million people and SO2 with 1.4 million people. A
total of 74 million persons resided in counties not
meeting at least one air quality standard during 1990
(out of a total 1987 population of 243 million).
Future reports will incorporate the 1990 Census
population estimates.
These population estimates are intended to
provide a relative measure of the extent of the
problem for each pollutant. The limitations of this
indicator should be recognized. An individual living
in a county that violates an air quality standard may
not actually be exposed to unhealthy air. For
example, if CO violations were confined to a traffic-
congested center city location during evening rush
hours in the winter, it is possible that an individual
may never be in that area, or may be there only at
other times of the day or during other seasons.
However, it is worth noting that ozone, which
appears to be the most pervasive pollution problem
by this measure, is also the pollutant most likely to
have fairly uniform concentrations throughout an
area.
pollutant
PM10
SO2
NO2
Ozone
Lead
Any NAAQS
0 20
Note: Based on 1987 county population data.
40 60
millions of people
100
Figure 4-6. Number of persons living in counties with air quality
levels above the primary national ambient air quality standards in
1990 (based on 1987 population data).
4-7
-------
Any population estimates depend upon the
assumptions and methodology used. In some cases
there can be a wide swing in the estimate. For
example, while there are an estimated 63 million
people living in counties that had 1990 ozone data
not meeting the ozone NAAQS, there are an
estimated 140 million people living in ozone
nonattainment areas. Although these numbers are
properly qualified, with such a large difference, it is
important to highlight some of the factors involved in
these estimates.
The estimate of 63 million people only considers
data from the single year, 1990 and only considers
counties with ozone monitoring data. In contrast,
ozone nonattainment areas are typically based upon
three years of data to ensure a broader
representation of possible meteorological conditions.
This use of multiple years of data, rather than a
single year, is intended to ensure that the attainment
decision is not simply the result of data from a single
year and should provide more assurance that the
NAAQS is met over all years, not simply in favorable
years.
Another difference is that the estimate of 63
million people living in counties with air quality levels
not meeting the ozone NAAQS only considers
counties that had ozone monitoring data for 1990.
As shown in Table 2-2, there were only 812 ozone
monitors reporting in 1990. These monitors were
located in 467 counties, which clearly falls far short
of the 3186 counties in the U.S. This shortfall is not
as bad as it may initially appear because it is often
possible to take advantage of other air quality
considerations in interpreting the monitoring data.
This, in fact, is why other factors are considered in
determining nonattainment areas. Ozone tends to
be an area-wide problem with fairly similar levels
occurring across broad regions. Because ozone is
not simply a localized hot-spot problem, effective
ozone control strategies have to incorporate a broad
view of the problem. Nonattainment boundaries may
consider other air quality related information, such as
emission inventories and modeling, and may extend
beyond those counties with monitoring data to more
fully characterize the ozone problem and to facilitate
the development of an adequate control strategy.
Since the early 1970's, there has been a growing
awareness that ozone and ozone precursors are
transported beyond the political jurisdiction of source
areas and affect air quality levels at considerable
distances downwind. To address this aspect of the
problem, the 1990 Clean Air Act Amendments
establishes a transport commission for the
Northeast, and allows the establishment of
commissions in other parts of the country. Generally
speaking, an entire transport region, including its
rural areas, is subject to the same requirements as
moderate areas.
JLL 17, 1987 18 EST
JUL 07, 1988 IB EST
Figure 4-7. Midwest region on
July 17,1987.
Figure 4-8. Northeast region on
July 7,1988.
4-8
-------
The transport of ozone concentrations generated
from urban manmade emissions of precursors in
numerous areas to locations further downwind can
result in rather widespread areas of elevated levels
of ozone across regional spatial scales. Visualizing
the regional nature of ozone transport is difficult
using monitoring data because ambient
measurements are available from only a discrete
number of individual sites which are generally
concentrated in or near major cities. In this sense,
modeling data, which produce three-dimensional
predictions of air quality, can provide added insight
into transport patterns. Several examples of urban
ozone plumes and their interaction on a regional
scale are shown from predictions of the Regional
Oxidant Model (ROM) for the Midwest and Northeast
U.S. in Figures 4-7 through 4-10. Each figure
displays ROM ozone predictions for a single hour
extracted from simulations which encompass multi-
day episodes. On July 17,1987 (Figure 4-7), urban
ozone plumes extend from the major Midwest cities
downwind with the southerly wind flow on this day.
Note the high ozone predicted along the western
shore of Lake Michigan and offshore over the Lake
resulting from urban areas along the western shore.
Also, ozone plumes from cities along the Ohio River
extend northward as far as Detroit.
In the Northeast, predictions for July 7, 1988
(Figure 4-8) reveal a continuous "river" of moderately
high ozone extending from eastern Ohio
southeastward to Richmond, VA then northeastward
along the Northeast Corridor from Washington, DC
to coastal Maine. Ozone plumes with concentrations
exceeding the level of the NAAQS are embedded
within this area over and downwind of major ozone-
precursor emissions areas. In fact, a continuous
area with ozone levels at or above .12 ppm are
predicted from southeastern Pennsylvania into
Rhode Island. Finally, comparing ozone patterns for
1400 EST and 2300 EST on July 8, 1988, (Figures
4-9 and 4-10) shows the transport of ozone
exceeding .15 ppm to the Boston area from sources
further to the Southwest during this time period.
These isopleth maps illustrate the spatial
patterns associated with ozone. They also serve to
indicate how a broader view based upon a variety of
air quality considerations could differ from an
interpretation based only upon a single year of
monitoring data from a limited number of sites. The
nonattainment area approach reflects a broader
range of meteorological conditions and incorporates
control strategy considerations. The single year
population estimate, using only monitoring data,
provides a convenient snapshot that emphasizes the
most recent status.
JUL 08, 1988 23 EST
JUL 08, 1988 14 EST
Figure 4-9. Northeast region at 1400
EST on July 8,1988.
Figure 4-10. Northeast region at 2300
EST on July 8,1988.
4-9
-------
4.3 Air Quality Levels in Metropolitan Statistical Areas
This section provides information on 1990 air
quality levels in each Metropolitan Statistical Area
(MSA) in the United States for general air pollution
audiences. For those large MSAs with populations
greater than 500,000, the 1990 annual air quality
statistics are also displayed geographically on
three-dimensional maps.
The general concept of a metropolitan area is
one of a large population center, with adjacent
communities which have a high degree of
economic and social integration with the urban
center. Metropolitan Statistical Areas contain a
central county(ies), and any adjacent counties with
at least 50 percent of their population in the
urbanized area.1 Although MSAs compose only 16
percent of the land area in the U.S., they account
for 78 percent of the total population of 243 million.
Table 4-2 displays the population distribution of the
341 MSAs, based on 1987 population estimates.1
The New York, NY MSA is the nation's largest
metropolitan area with a 1987 population in excess
of 8 million. The smallest MSA is Enid, OK with a
population of 60,000.
4.3.1 Metropolitan Statistical Area Air
Quality Maps, 1990
Figures 4-11 through 4-18 introduce air quality
maps of the United States that show at a glance
how air quality varies among the largest MSAs
within the contiguous United States. To enable the
reader to distinguish individual urban areas, only
the 90 MSAs within the continental U.S. having
populations greater than 500,000 are shown. Two
large MSAs, Honolulu, HI and San Juan, PR are
not shown. However, neither area has exceeded
any of the NAAQS during 1990. In each map, a
spike is plotted at the city location on the map
surface. This represents the highest pollutant
concentration recorded in 1990, corresponding to
the appropriate air quality standard. Each spike is
projected onto a back-drop for comparison with the
level of the standard. The backdrop also provides
an east-west profile of concentration variability
throughout the country.
TABLE 4-2. Population Distribution of Metropolitan Statistical Areas Based on 1987
Population Estimates
POPULATION RANGE
< 100,000
100,000 < population < 250,000
250,000 < population < 500,000
500,000 < population < 1,000,000
1,000,000 < population < 2,000,000
population > 2,000,000
NUMBER OF
MSA'S
28
148
73
48
26
18
POPULATION
2,367,600
23,513,000
25,218,000
34,367,000
38,685,000
65,747,000
MS A TOTAL 341 189,897,600
4-10
-------
4.3.2 Metropolitan Statistical Area Air Quality Summary, 1990
Table 4-3 presents a summary of 1990 air quality
for each Metropolitan Statistical Area (MSA) in the
United States. The air quality levels reported for
each metropolitan area are the highest levels
measured from all available sites within the MSA.
The MSAs are listed alphabetically, with the 1987
population estimate and air quality statistics for each
pollutant. Concentrations above the level of the
respective NAAQS are shown in bold type.
In the case of O3, the problem is pervasive, and
the high values associated with the pollutant can
reflect a large part of the MSA. However in many
cases, peak ozone concentrations occur downwind
of major urban areas, e.g., peak ozone levels
attributed to the Chicago metropolitan area are
recorded in and near Kenosha, Wisconsin. In
contrast, high CO values generally are highly
localized and reflect areas with heavy traffic. The
scale of measurement for the pollutants - PM-10,
SO2 and NO2 - falls somewhere in between. Finally,
while Pb measurements generally reflect Pb
concentrations near roadways in the MSA, if a
monitor is located near a point source of lead
emissions it can produce readings substantially
higher. Such is the case in several MSAs. Pb
monitors located near a point source are footnoted
accordingly in Table 4-3.
The pollutant-specific statistics reported in this
section are for a single year of data. For example,
if an MSA has three ozone monitors in 1990 with
second highest daily hourly maxima of 0.15 ppm,
0.14 ppm and 0.12 ppm, the highest of these, 0.15
ppm, would be reported for that MSA. The
associated primary NAAQS concentrations for each
pollutant are summarized in Table 2-1.
The same annual data completeness criteria
used in the air quality trends data base for
continuous data was used here for the calculation of
annual means, (i.e., 50 percent of the required
samples for SO2 and NO2). If some data have been
collected at one or more sites, but none of these
sites meet the annual data completeness criteria,
then the reader will be advised that there are
insufficient data to calculate the annual mean. With
respect to the summary statistics on air quality levels
with averaging times less than or equal to 24-hours,
all sites are included, even if they do not meet the
annual data completeness requirement.
For PM-10 and Pb, the arithmetic mean statistics
are based on 24-hour measurements, which are
typically obtained from a systematic sampling
schedule. In contrast to the trends analyses in
Section 3 which used a more relaxed indicator, only
maximum quarterly average Pb concentrations and
weighted PM-10 annual means meeting the AIRS
validity criteria are displayed in Table 4-3.
This summary provides the reader with
information on how air quality varied among the
nation's metropolitan areas in 1990. The highest air
quality levels measured in each MSA are
summarized for each pollutant monitored in 1990.
Individual MSAs are listed to provide more extensive
spatial coverage for large metropolitan complexes.
The reader Is cautioned that this
summary is not adequate In itself to
numerically rank MSAs according to their
air quality. To rank properly the air
pollution severity among different MSAs,
data on population characteristics, daily
population mobility, transportation
patterns, industrial composition,
emission inventories, meteorological
factors and, most important, the spatial
representativeness of the monitoring
sites would also be needed.
4.4 REFERENCES
1. Statistical Abstract of the United States^
1989, U. S. Department of Commerce, U. S.
Bureau of the Census, Appendix II.
2. 40CFR, PART 81 (Federal Register,
November 6, 1991).
4-11
-------
PM10
2ND MAX 24-HR AVG
Figure 4-11. United States map of the highest second maximum 24-hour
average PM-10 concentration by MSA, 1990.
The map for PM-10 shows the 1990 highest second maximum 24-hour average
PM-10 concentration in metropolitan areas greater than 500,000 population.
Concentrations above the level of the 24-hour PM-10 standard of 150 |o.g/m3 are
found in 9 of these metropolitan areas.
4-12
-------
PM10
ANNUAL ARITHMETIC MEAN
Figure 4-12. United States map of the highest annual arithmetic mean PM-10
concentration by MSA, 1990.
The map for PM-10 shows the 1990 maximum annual arithmetic means in
metropolitan areas greater than 500,000 population. Concentrations above the
level of the annual mean PM-10 standard of 50 ng/m3 are found in 8 of these
metropolitan areas.
4-13
-------
'a*
SULFUR DIOXIDE
ANNUAL ARITHMETIC MEAN
Figure 4-13. United States map of the highest annual arithmetic mean sulfur
dioxide concentration by MSA, 1990.
The map for sulfur dioxide shows maximum annual mean concentrations in 1990.
Among these large metropolitan areas, the higher concentrations are found in the
heavily populated Midwest and Northeast and near point sources in the west. All
these large urban areas have ambient air quality concentrations lower than the
current annual standard of 80 M-g/m3 (0.03 ppm). Because this map only
represents areas with population greater than one half million, it does not reflect
air quality in the vicinity of smelters or large power plants in rural areas.
4-14
-------
SULFUR DIOXIDE
2ND MAX 24-HR AVG
Figure 4-14. United States map of the highest second maximum 24-hour
average sulfur dioxide concentration by MSA, 1990.
The map for sulfur dioxide shows the highest second highest 24-hour average
sulfur dioxide concentration by MSA in 1990. Pittsburgh, PA is the only large
urban area which had ambient concentrations above the 24-hour NAAQS of 365
u,g/m3 (0.14 ppm).
4-15
-------
CARBON MONOXIDE
2ND MAX 8-HR AVG
Figure 4-15. United States map of the highest second maximum
nonoverlapping 8-hour average carbon monoxide concentration
by MSA, 1990.
The map for carbon monoxide shows the highest second highest 8-hour value
recorded in 1990. Twelve of these urban areas have air quality exceeding the 9
ppm level of the standard. The highest concentration recorded in 1990 is found
in Los Angeles, CA.
4-16
-------
NITROGEN DIOXIDE
ANNUAL ARITHMETIC MEAN
Figure 4-16. United States map of the highest annual arithmetic mean
nitrogen dioxide concentration by MSA, 1990.
The map for nitrogen dioxide displays the maximum annual mean measured in the
nation's largest metropolitan areas during 1990. Los Angeles, California, with an
annual NO2 mean of 0.056 ppm is the only area in the country exceeding the NO2
air quality standard of 0.053 ppm.
4-17
-------
OZONE
2ND DAILY MAX 1-HR AVG
Figure 4-17. United States map of the highest second daily maximum 1-hour
average ozone concentration by MSA, 1990.
The ozone map shows the second highest daily maximum 1-hour concentration
in the 90 largest metropolitan areas in the Continental U.S. As shown, 39 of these
areas did not meet the 0.12 ppm standard in 1990. The highest concentrations
are observed in Southern California, but high levels also persist in the Texas Gulf
Coast, Northeast Corridor and other heavily populated regions.
4-18
-------
LEAD
MAX QUARTERLY MEAN
Figure 4-18. United States map of the highest maximum quarterly average
lead concentration by MSA, 1990.
The map for Pb displays maximum quarterly average concentrations in the nation's
largest metropolitan areas. Exceedances of the Pb NAAQS are found in nine
areas in the vicinity of nonferrous smelters or other point sources of lead.
Because of the switch to unleaded gasoline, areas primarily affected by automotive
lead emissions show levels below the current standard of 1.5 |o.g/m3.
4-19
-------
O
O
m
O
?
o><
O» I
^ w
m
tu;
CM '
o:
OfC*
8311
v> ^
ill
iuL*
g
.
o'o
o o o o'oooo o
ooooo^o^oot^
zzzzzszgzzg
do d
zzzzzzzz:
, Q Q I
o oo o
Q Q r; Q t Q i
zzgzgz:
oo ooo o ooo ooo o
O O
o . . o ^. " "
o* d odd d o o o O o
cj
>
O
EC
I
O
£
"5"S
IU=J°.
ZQ Z
iiJ
ff
: ? * < ^ k
-------
dd
^^ \J ^J \J %*J
d odd
zzz zzz zzzz z z z z zzz
dddddd
do
ddddddd d
88888
°, °. °. P. P.
88888
ooooo
8888888888888888
oooooooooooooooo
T- r-
«JO>cM
CM o T-
mmCQCDODajtOCDmCOCQCQODmBQCDOOOOOOOOOOOOOO
u_ i
8!
o> --
5 .9
E
T
RI
A
ST
S
ST
ST
E
HIG
HIG
INDIC
NDICA
1 CO £D
O Q.
<0 ";
I f
-------
P
.< ai
§i
li
LU |-
§£
^ 03
EQ
CD
p?
Pi
^*? jf% ~**~
55
§
I
ET
d o d T* o v* do od
z g z z 5
d d
z z
pzzzzzzzzzg:
d d
zz zzz
SZZZ^ZgZZZZ
d .do
zzzzz
z z
o o ooooodo
oo
oo
z z:
; CD Q CM Q |
O O
do" dododdd odd do odd do do
§
Ot <\J ^ i CM i
o:
uf:
^ w
^ o
lilxi1
*8J
uluf ~t
_
OO
_
OODQQQQQOOOQQQQQUJLUUJUJUJUJUlliJU.lj-U.U-U,U,ttU.U.
-------
dod
dddd
oo
Z O O i
d d d > d o d d d d d d d d d o d
Q- Q
o do
O
d
d d
ZZZZ
ZZ Z
Z ZZZ Z Z
odd do do odd odd d do
O <> o
3 88
o o o Q o o o o ^ o o o o o o o
odd d odd ddddd do
ZO
T
|||8
8 8 £ S 9 S' S fe ^ S ^" S ^" S!
:8888888888
^o_oo_o_o_o_o_ooo
" ?>-Q ^
3 E 0} o
s -Lit
gwli*i2t
2^ c ° "" Z m Z *".
5^CO/2OlO O)1^"
Z ^^rfi °-S.S2
a>o°*frrcn "CO
1 « CO 1 z CO 1< y
=2^co 2 -9 LU 2 S-z co
^9 < J z ^ 5 b
z.sJ'O.si! fc
Z a -9 Tr § b =5. >-
?!ijz8i^! ^
< = SO I o-z 2
2 g-Sh- co i-UJ -, S
fetcls^zQ ^
= HIGHEST SECOND MAXIMUM 24-HOUR CONCE
IGHEST ARITHMETIC MEAN CONCENTRATION
HIGHEST ARITHMETIC MEAN CONCENTRATION
IGHEST SECOND MAXIMUM 24-HOUR CONCENT
HIGHEST SECOND MAXIMUM NON-OVERLAPPIN
= HIGHEST ARITHMETIC MEAN CONCENTRATION
HIGHEST SECOND DAILY MAXIMUM 1-HOUR CO
HIGHEST QUARTERLY MAXIMUM CONCENTRAT
INDICATES DATA NOT AVAILABLE
NDICATES INSUFFICIENT DATA TO CALCULATE
o I " I n " « H u
2 " O " O O eo CO Q -,
Q- CO O Z O CL Z 5
act from an industrial source.
Q.
«
1
CO
d
en
i*
8
CO
1
Q
C
'&
10
(/}
a>
x:
D>
I
^
>^
i
o
_c
s
C
8
-------
O
8
ID w
O2
LU t
oc 3
80
p
< LU
b w
££
LU
fe
O) h-
^ OT
a
O JB j
n\
s
EC
i
I
g
SQ Q N- Q
Z2OZ
do d
d ddcfddd
dd
dd
dodo dd odd
O '
d
Z Z
ZZZZ ZZ ZZZ ZZZ Z
dddd
d
ZZZ Z Z
dddd odd dddd d ddddd
dddddd odd
d dd d ddddd d
l
l
lii
iii
illiliiii
g
s
LU -
x
ZR
xzg
Osd ,
-------
o o o ooo
Q Q
zz
Q 55
Z Q
Q i
Z ,
odd
O O i- !
pop
odd
o
o o
p p
d do
°. d °.
1 » 'V -7 TS in
*-*-*- T>
888888 88 8888888
ooooooooooooooo
8 8 & 8 IS S? & 5f * s" °" s" 3' s"
88 8§88
oooooo
" "
8
o
CO
00
d
5^
.w
LU
5
O < < £2
Z QC CO 3 ^ ffl
333333-^
8Sp.grf^o-
CO f- Q ^ Q. . p
UJI Oj Z ^ < UJ| CO
_i _i
CO
CO
0 _
"
O Q.
!>
o
in
z
.2
Is
ex =3. :
^I^C
i°^|:
NCENTRATION (Applicable NAAQS
ION (Applicable NAAQS is 50 ug/m3)
TION (Applicable NAAQS is 0.03 pp
CENTRATION (Applicable NAAQS is
PPING 8-HOUR CONCENTRATIO
TION (Applicable NAAQS is 0.053
RA
NC
LA
RA
is! j J i
\s ^f CO CD Q -9
O Z O CL 2 ?
-------
5^
i^S
£o
^a
is
I
LU K-
O H
* (0
a
5
CO
sss-
i- z r- z r- z
; z n z :
: z °. z
T T T T- O T*»- >r»
o
ZZZZ
ZZ Z
dodd
Z Z
o
ZZZZZ
od
Z Z
d dddddddd d d d od dddddd d
OOO oooooooo o o o ooo ooooooo O
^^ to ^^ ^~ O) Oft ^1 ti?" n* evi»5it«.«<2
§§
§
CJ
₯ 5g 8"8 5
'8f8"8"8T
T- -I- 7- CO
§§
8
_l
0»S §
CO CO flf> T^
<# »-" W
<
' LJ_I
o o
isiii
0003D
-t-u.:i>JZ=iu.imrf_jrf-=i-i^HVuJuu
-------
o o o o
do
o o o
odd
o o
do
zz
5 O> if) Q> j- i
CO * CM Cv ^s '
O O O O O '
o d d d d
d d d d
OOOOO OOOO
oocjpp^p po.
ddddd d do
888°
oooooooo
oooooooo
ooooooo
'
CM CM T- CJ
UJ
DC
111
CO CO O t^
! CM CO * T-
&5
LU
Of
tr
UJ
<>
UJ
S° fc
59uJtt=>
s|lg|
Q -> LU
t-1 O LU
i^jy
m I O
Jiii2^io"
Q.D.o.SIoIac£c£a:cc
trcc
a> w
a:
LJJ
o_,ozEOX5
gQ|gib||
^|§i|§l
|PS|S!S52
QZr?SEOf>^r>
LU
§
o
QC
111
KM
fc<£
Sfeg
SSi
n^1
o* "
i "S
a. co
2
5a
II
O w
UJ W
CO h-
1- W
W^
LU I
iS
01
i ii
8
to
2Q
E o
< m
HW
l|
?i
8
8
s^y
^
CO
ia
_
a>
"S
-------
D CO
02
*
||
uJF
SCO
O h-
«<
O) |
3
«3ti
8 I
* ^^
\-s u* <
°5i
KV
8:
'§:
i
*4-
2
S
SO Q Q O> f*l f^- W CO ^ f) Q Q Q CO Q Q Q CM Q Q Q Q Q
cc'r do
T" ^ O O
o
o
Z *~ ZZ Z Z
OQOOOOOOO o o
c> dd dddddd d
o o^ o o^ o * T * o o * o -C?
d dddd d do dd
o H H
00^0000500000050:
qo^qopq^cso. 3°. d°. P'
od dddd o o o o o
q^-qqoq^q^-qq^-oo
d dddd d dd dd
Ct ?O ^5 ^? C5 ^? C5 CD CD CD CD CD CJ (3 CD Ct CD ^3 CD X3
~CD CD CD
n<
s
UJ H
2tz"
of 11
UJSg
feJiJO
uj Q >-
B^§
<-> in '*
o
8
g
fe
HI
cc
!
01
^
II
I
o$2
ws
-
S-S^t
isll
^ait
-------
o> w Q o T- |
r» o -y T-
d ci ci
.
o odo
.
c> d c> o o c! d
z
dd ddd doo
o
zzz zz z zzz zz z zz;
d odd
P d °. °.'
d do
o o oo ooo oo o o
Q Q
do d
o o o
^J VJ ^^ *>J ^^ t? ^^
do d d
8
o
o o
80
o
i o
;8
o o
o o
o o
R
£ 8 52 2 S S 8 £ 8
OH
KS
, NJ
AZ
N
,
EDO,
EKA,
NTO
SO
O«c
TOL
TOP
TRE
TUC
" ^£~£t
co ^i'E'CI-E
0
"g-
O
UJ Z K z Z o P
o^,ozEo85
SRiBisi
'^ofStQuj
-
CO
1
3
fcs
^£5
£fe£
I CO T
?si
XC2X
o^:
i "S
Q. CO
;8^i^
tu«
5
in
«
o>
S>
I
o>
I
-------
5. SELECTED METROPOLITAN AREA TRENDS
This chapter discusses 1981-90 air quality
trends in fifteen major urban areas: the ten EPA
Regional Offices (Boston, New York, Philadelphia,
Atlanta, Chicago, Dallas, Kansas City, Denver, San
Francisco and Seattle) and five additional cities
(Detroit, Houston, Los Angeles, Pittsburgh and
Washington, DC.)
The presentation of urban area trends includes
maps of the urban area showing major roadways
and rivers, county boundaries and the urbanized
area. These maps show the location of the current
air quality monitoring networks and indicate the
sites used in the trend analysis. To complement
the map and show the general orientation of the
ambient monitoring network with respect to wind
flow patterns, a wind rose is presented. The wind
rose shows the direction the winds came from
during June, July and August, emphasizing the
ozone season. Also, three graphical displays are
used to depict urban air quality trends. One graph
uses the Pollutant Standards Index (PSI) as the
measure of air quality. The other two graphs
include temperature and display average levels for
CO and O3.
The air quality data used for the trend statistics
were obtained from the EPA Aerometric Information
Retrieval System (AIRS). This year's report
presents trends in the PSI, used locally in many
areas to characterize and publicly report air quality.
The new PSI analyses are based on daily
maximum statistics from selected monitoring sites.
The urban area trends for CO and O3 use the
same annual validity and site selection criteria that
were used for the national trends. It should be
noted that no interpolation is used in this chapter;
this corresponds with typical PSI reporting.
5.1 The Pollutant Standards Index
The PSI is used in this section as an air quality
indicator for describing urban area trends. Only CO
and O3 monitoring sites had to satisfy the trends
selection criteria discussed in Section 2.1 to be
included in these PSI trend analyses. Data for
other pollutants were used without applying this
historical trends criterion, except for SO2 in
Pittsburgh because this pollutant contributed a
significant number of days in the high PSI range.
Results for individual years could be somewhat
different if data from all monitoring sites and all
pollutants were considered in an area. This is
illustrated for 1990, where the number of PSI days
from all monitoring sites is compared to the results
for the subset of trend sites.
The PSI has found widespread use in the air
pollution field to report daily air quality to the
general public. The index integrates information
from many pollutants across an entire monitoring
network into a single number that represents the
worst daily air quality experienced in the urban
area. The PSI is computed for PM-10, SO2, CO,
O3 and NO2 based on their short-term National
Ambient Air Quality Standards (NAAQS), Federal
Episode Criteria and Significant Harm Levels. Lead
is the only criteria pollutant not included in the
Table 5-1. PSI Categories and Health Effect Descriptor Words
INDEX RANGE
OtoSO
51 to 100
101 to 199
200 to 299
300 and Above
DESCRIPTOR WORDS
Good
Moderate
Unhealthful
Very Unhealthful
Hazardous
5-1
-------
index because it does not have a short-term
NAAQS, a Federal Episode Criteria or a Significant
Harm Level.
The PSI converts daily monitoring information
into a single measure of air quality by first
computing a separate sub-index for each pollutant
with data for the day. The PSI index value used in
this analysis represents the highest of the pollutant
sub-index values for all sites selected for the MSA.
Local agencies may use only selected monitoring
sites to determine the PSI value so that differences
are possible between the PSI values reported here
and those done by the local agencies.
The PSI simplifies the presentation of air quality
data by producing a single dimensionless number
ranging from 0 to 500. The PSI uses data from all
selected sites in the MSA and combines different
air pollutants with different averaging times,
different units of concentration, and more
importantly, with different NAAQS, Federal Episode
Criteria and Significant Harm Levels. Table 5-1
shows the 5 PSI categories and health effect
descriptor words. The PSI is primarily used to
report the daily air quality of a large urban area as
a single number or descriptor word. Frequently,
the index is reported as a regular feature on local
TV or radio news programs or in newspapers.
Throughout this section, emphasis is placed on
CO and O3 which cause most of the NAAQS
violations in urban areas.
5.2 Summary of PSI Analyses
Table 5-2 shows the trend in the number of
PSI days greater than 100 (unhealthful or worse
days). The impact of the very hot and dry
summers in 1983 and 1988 in the eastern United
States on O3 concentrations can clearly be seen.
Pittsburgh is the only city where a significant
number of PSI days greater than 100 are due to
pollutants other than CO or 03. For Pittsburgh,
SO2 and PM-10 account for the additional days.
The two right most columns show the number of
currently active monitoring sites and the
corresponding total number of PSI days > 100,
using these sites. Note that for all urban areas
except Houston, New York and Seattle there is
close agreement between both statistics for 1990.
The differences are attributed to currently active
sites without sufficient historical data to be used for
trends.
For all practical purposes CO, O3> PM-10 and
SO2 are the only pollutants that contribute to the
PSI in these analyses. NO2 rarely is a factor
because it does not have a short-term NAAQS and
can only be included when concentrations exceed
one of the Federal Episode Criteria or Significant
Harm levels. TSP is not included in the index
because the revised particulate matter NAAQS is
for PM-10, not TSP. As noted above, lead is the
only criteria pollutant not included in the index
because it does not have a short-term NAAQS or
Federal Episode Criteria and Significant Harm
Levels.
Table 5-3 shows the trend in the number of PSI
days greater than 100 (unhealthful or worse) due to
O3. The only 3 areas where O3 did not account for
most of these days were: Denver, New York City
and Pittsburgh. In Denver and New York City, CO
accounted for a larger number of these days.
However, because of the overall improvement in
CO levels (see Section 3.3 in this report), CO
accounts for far less of these days in the latter half
of the 10-year period. Overall, 72% of the PSI
greater than 100 days were due to O3. In
Pittsburgh SO2 and PM-10 contribute a significant
number of these days with PSI greater than 100.
Figure 5-1 is a bar chart: showing the number of
PSI days above 100 in 1988, 1989 and 1990 for
fourteen of the cities being studied. To permit
better scaling, Los Angeles is not shown on the
graph but the values were 228, 213 and 163 for
1988, 1989 and 1990 respectively. This
comparison uses all the monitoring sites available
in an area for the 3 years;. The use of all sites
explains why these figures may not agree with
Table 5-2, where only the CO and O3 sites that met
the trend criteria were used. In most cases, there
Note: Urban lead concentrations have dropped dramatically over
the past 15 or so years (See Chapter 3). As a result, only 10
urban areas violated the lead NAAQS based upon 1990 data
only. Dallas and Philadelphia are the only two of the 15 urban
areas that have a 1990 lead violation. In Dallas, the problem
occurred near a smelter located outside of Dallas County, in
adjoining Collin County. In Philadelphia, the problem occurred
near a smelting and a materials handling operation.
5-2
-------
« 0>CO
>§?
ts o"t 8
CB * £ 0>
Mr
w
a? A 8
(0
-
CM
^
8
CO
9
in
O)
r.
IV
CO
CM
9
£
eg
eg
s
CM
a
3
-
8
CM
8
T-
m
CO
CM
in
CM
CO
CO
0
CO
3
.a
C/3
8
(0
o
8
T~
C
t! oc
1 ft
1 "
u
CB
0
E
o
1
3
cn
S
*~
S
o>
cn
T"
CO
s
T-
O
i
1
CM
oo
o>
5
en
c S
Q.
CO
co
m
o>
IV
cn
co
S3
m
o>
CO
| ATLANTA
-
-
in
o
eg
CO
co
in
CM
CO
| BOSTON
CO
CM
oo
o>
^
CO
oo
s
CO
CO
CD
1 CHICAGO
in
co
co
CO
in
CM
o
£
-
CM
CO
| DALLAS
-
^
oo
CO
CO
CO
s
CD
CM
in
in
*
1 DENVER
CO
eg
-
O)
co
CM
h>
00
T
CO
oo
cn
1 DETROIT
in
CO
o>
CO
CO
oo
8
o
co
9
s
s
in
1 HOUSTON
eg
CM
CO
in
00
-
CM
*
O
^
IV
1 KANSAS Cm
§
CM
eg
CD
IV
oo
o
eg
co
o>
oo
CM
oo
$
oo
CO
W
| LOSANGELE
o
en
in
CO
CO
CO
CM
8
$
8
o
o
oo
DC
i
-
CD
S
CO
CO
CM
S
co
co
in
?
0,
in
1 PHILADELPH
CM
^
8
in
en
CD
^
CO
CO
*
£
CM
T
1 PITTSBURGH
i
O
-
-
^
in
CM
*
CM
^_
CO
ISAN
FRANCISCO
CM
oo
CO
oo
co
CM
^.
cn
cn
CM
^
1 SEATTLE
in
CO
S3
T-
m
o
5?
in
eg
CO
2
»
| WASHINGTOf
CO
CM
in
CO
i
o>
o
o
^J"
CD
O)
CO
s
O)
CO
1
CO
eg
1 TOTAL
§
I
3!
V
i
-------
.3
8g>|
S1 = 8
HI**
§E!
2Ǥ
|8|
i>»
*
T-
m
o
co
in
<*
0
ID
co
o>
?
co
CM
CO
N.
T-
8
CO
(0
-
o
o
o>
-
*
*
t
to
CM
r^
SI
in
CM
(0
V)
o
£
c
5
o
o
C OC
5 2
H >
1
i
0
0)
ffl
Q
5
E
Z
8
o>
T-
!
§
T-
Y-
^
o>
i
"
CM
00
cn
T-
1
^
£ "35
«
CO
co
in
o>
£
01
oo
g}
co
o>
CM
1 ATLANTA
-
-
-
o
CM
co
o
CO
o
-
Z
o
-
in
O)
CM
CO
CO
CM
CO
CO
in
1 CHICAGO
in
CO
CO
co
m
-
o
f-
*-
CM
CM
1 DALLAS
o
0
CO
"J-
-
o
T-
r-
^f
'
CM
1 DENVER
CO
o
CO
CO
co
-
*
CD
1^
CM
oo
Q
in
CO
o>
co
0
8
o
co
CO
5?
CM
B
*
| HOUSTON
CM
-
CO
CM
co
co
*-
^
o
-
in
1 KANSAS CITY
CO
fs.
co
s
co
ij-
^"
s
s
s
*"
w
^
CO
CO
CM
in
CO
1 LOS ANGELES
CO
in
8
,,.
CO
C<3
CO
CO
CM
CM
in
oc
g
i
-
t
*
CM
CO
O)
s
cvl
CM
CO
a
0
1 PHILADELPHIA
0
-
CM
^
CM
CM
0
U)
^t
cn
in
1 PITTSBURGH
o
0
o
0
o
V-
o
CM
o
-
CM
II
SAN
1 FRANCISCO
CM
0
-
0
'-
o
o
o
o
-
*-
1 SEATTLE
in
-
CO
CO
oo
cn
CM
CM
00
Ol
CM
-
|| WASHINGTON
s
in
8
CO
CM
CO
s
CM
O
CM
00
"*
£!
CM
m
CO
I TOTAL
§
I
s
s
I
-------
FIGURE 5-1. PSI DAYS>100 IN 1988,1989 AND 1990 USING ALL SITES
ATLANTA
BOSTON
CHICAGO
DALLAS
DENVER
DETROIT
HOUSTON
KANSAS CITY
NEW YORK CITY
PHILADELPHIA
PITTSBURGH
SEATTLE
SAN FRANCISCO
WASHINGTON DC
0
10
20
30 40
DAYS
50
60
70
1988
* NOTE: Los Angeles not shown because of scaling problem.
See Table 5-2 for the PSMOO days in Los Angeles.
1989
1990
Figure 5-1. PSI days > 100 in 1988,1989 and 1990 using all sites.
has been a reduction in the frequency of these
days between 1988 and 1990. For the eastern and
mid-western cities, meteorological conditions during
the summer of 1988 were very favorable for O3
formation. Nationally, the summer of 1988 was the
third hottest on record since 1931. Even some of
the western cities, e.g. Los Angeles, Denver and
Seattle, follow this pattern too.
The pollutant having the highest sub-index
value, from all the monitoring sites considered in an
MSA, becomes the PSI value used for that day.
PSI estimates depend upon the number of
pollutants monitored and the number of monitoring
sites collecting data. The more pollutants and sites
that are available in an area, the better the
estimate of the maximum PSI for that day is likely
to be. Ozone accounts for most of the days with a
PSI above 100 and O3 air quality is relatively
uniform over large areas so that a small number of
sites can still estimate maximum pollutant
concentrations. All of the included cities had at
least one CO trend site and one O3 trend site.
Table 5-4 separately shows the number of CO and
O3 trend sites used in each of the MSA's. In
addition, 8 SO2 trend sites were used in Pittsburgh
because SO2 accounted for a sizeable number of
days when the PSI was greater than 100. In Table
5-4, the months corresponding to the O3 season in
the 15 areas are also provided. The PSI trend
analyses are presented for the Primary MSA
(PMSA) in each city studied, not the larger
Consolidated Metropolitan Statistical Area (CMSA).
Using the principal PMSA limits the geographical
area studied and emphasizes the area having the
highest population density. The PMSA monitors
are in the core of the urban area; there are typically
additional sites in surrounding areas. For example,
while there are 21 active monitoring sites in the Los
Angeles PMSA in 1990, there are more than 30
monitors for ozone alone in the larger metropolitan
area.
5-5
-------
Table 5-4. Number of Trend Monitoring Sites for the 15 Urban Area Analyses
Primary Metropolitan
Statistical Area (PMSA)
Atlanta, GA
Boston, MA
Chicago, IL
Dallas, TX
Denver, CO
Detroit, Ml
Houston, TX
Kansas City, MO-KS
Los Angeles, CA
New York, NY
Philadelphia, PA
Pittsburgh, PA
San Francisco, CA
Seattle, WA
Washington, DC-MD-VA
CO Sites
1
2
3
1
4
6
3
3
11
3
9
3
3
6
10
O3 Sites
2
1
5
2
2
8
4
5
13
5
10
5
2
1
11
O3 Season
MAR - NOV
APR - OCT
APR - OCT
MAR - OCT
MAR - SEP
APR - OCT
JAN - DEC
APR - OCT
JAN - DEC
APR - OCT
APR - OCT
APR - OCT
JAN - DEC
APR - OCT
APR - OCT
There are several assumptions that are implicit
in the PSI analysis. Probably the most important is
that the monitoring data available for a given area
provide a reasonable estimate of maximum short-
term concentration levels. The PSI procedure uses
the maximum concentration which may not
represent the air pollution exposure for the entire
area. If the downwind maximum concentration site
for ozone is outside the PMSA, these data are not
used in this analysis. Finally, the PSI assumes that
synergism does not exist between pollutants. Each
pollutant is examined independently. Combining
pollutant concentrations is not possible at this time
because the synergistic effects are not known.
5.3 Description of Graphics
Each of the fifteen cities has all of the graphics
and explanatory text presented on facing pages.
The first page includes a map of the area
highlighting the location of the current ambient
monitoring network within the PMSA as well as
other important features like rivers, lakes and major
highways. At each site, the shaded pie wedges of
a circle identify the pollutants monitored in 1990.
Circles with four tick marks indicate trend sites.
Below each map is an inset showing the
location of each area, a legend describing the sites
and a wind rose. The legend identifies the shaded
5-6
-------
wedges corresponding to particular pollutants. The
pollutants (CO, O3 and PM-10) appear on the
upper half of the circle, while the other pollutants
(lead, SO2 and NO2) are on the lower half. The
wind rose shows the frequency of hourly wind
direction measurements for June, July and August
of calendar year 1990. This corresponds to the
principal part of the ozone season. The wind
direction refers to the direction the wind is blowing
from. The wind data comes from one of the
National Oceanographic and Atmospheric
Administration (NOAA) meteorological observation
stations in the area, usually located at the principal
airport.
The accompanying graphs are based on the
PSI methodology described earlier. The PSI
graphs feature a bar chart which shows the number
of PSI days in four PSI categories: 0-50, 51-100,
101-199 and >200. Table 5-1 shows the PSI
descriptor words associated with these categories.
The last 2 PSI categories (very unhealthful and
hazardous) were combined because there were so
few hazardous days reported. The total number of
unhealthful, very unhealthful and hazardous days is
used to indicate trends. These days are
sometimes referred to as the days when the PSI is
greater than 100. It is important to note that a PSI
of 100 means that the pollutant with the highest
sub-index value is at the level of its NAAQS.
Because of numerical rounding, the number of
days with PSI > 100 does not necessarily
correspond exactly to the number of NAAQS
exceedances.
CO and O3 trends are shown on separate plots
incorporating information on temperature. CO
trends are displayed in terms of the daily maximum
8-hour average data. These averages are also
shown for two different categories: days with
minimum temperatures greater than 40 degrees
and those less than or equal to 40 degrees
Fahrenheit. Maximum daily temperatures are used
for O3. The 03 plots show the trend in average
daily maximum 1-hour concentrations for three
categories during the O3 season: 1.) the ten highest
O3 concentration days, 2.) the days when the
maximum temperature was 80° F or more and 3.)
for all days. The average maximum temperature
on the days with the ten highest ozone values are
shown as bars in the background of these graphs.
The O3 season for each of these areas is shown in
Table 5-4. These plots are an attempt to indicate
the impact of temperature, an important
meteorological variable. Ozone levels are highest
in the summer, especially on very hot stagnant
days, while CO is highest usually in the winter
months. The New York MSA is one exception; it
appears that higher temperatures are associated
with higher CO levels. The winter, spring, summer
and fall seasons that are referred to correspond
respectively to the following months: December-
February, March-May, June-August and
September-November.
A simple nonparametric test was used to
determine the statistical significance of the trends.
This test correlated the ranks of the pollution
variable, either the number of days that the PSI
was above 100 or the average of CO or O3 in
various temperature categories, with the
corresponding rank of year. The magnitude of the
observed correlation, known as the Spearman
correlation coefficientRs), indicates the strength of
the trend. Coefficients near 1 signify a close
agreement between the ranks; whereas,
coefficients near 0 signify no agreement. When a
trend is said to be significant, it is understood to be
significant at the 0.10 level. Correlations that are
referred to as significant mean that the correlation
is significantly different from 0. The strength of the
correlation between average temperature and O3
levels on the ten highest O3 days was also
examined. The following sections present the
metropolitan areas analyses.
5-7
-------
l-l_7-l«J7-27
i i^^
Kind Speed (Knot!
5-8
-------
YE
81
82
83
84
85
86
87
88
89
90
C
CO
2.5
2
1.5
1
0.5
0
A
o,
0
01
0.
0.0
Number of Days in PSI Categories
AR
» , ,.\ - * i
^.* . f
> : x :..i HH
* -- >...,.. '" : 1, .
-"'. .VZ>V T.'.VV.x .'.'." "il P
- - - - 1 ra
» "- -- " ; \ i sA
»..: ' * -v : *. ".. - i HI
H "...I". 1 I
- j i . »
100 200 300
DAYS
[^jijijlGood f I Moderate ^^H Unhealthful ^^H ., __,
Avg Daily Max 8-hr CO by Temperature
ppm
"--. ...-%
^x1 ~ v\ y^"
.^^^£^
81 82 83 84 85 86 87 88 89
YEAR
Tomp_>_40°F Temp_..!..., .,?. i- ....»,,...» i :: i
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Days.-> 80 F Ten High Q3 Days Avg
AII.D.ays Avg Tamp of Ten High 03 days
400
*
90
J
DO
5
D
5
3
5
3
Atlanta, GA
The Atlanta PMSA consists of 18 counties,
with most of the people living in Fulton,
De Kalb and Cobb Counties. The estimated
1987 population was 2.7 million. Its size and
summertime meteorology contribute to the
area's air pollution potential. The Bermuda
High has a dominant effect on Atlanta's air
quality, especially during the summer when
the hot stagnant days are conducive to O3
formation. The map shows 12 currently
active monitoring sites.
The PSI trend for Atlanta is based on 3 sites:
1 for CO and 2 for O3. The CO site is a
population exposure site located in De Kalb
County. The O3 sites are a maximum
concentration site in Rockdale County and a
population exposure site in De Kalb County.
Ozone is the pollutant in Atlanta that
accounts for all but 2 of the unhealthful days.
The number of days with an unhealthful or
worse rating varied from 23 days in 1983 to
a low of 3 days in 1989. In 1990 there were
16 of these days, all due to O3. A trend test
did not show a significant trend in these
days. In the 10-year period only 1 day (in
1987) had a value in the very unhealthful
range; no hazardous days were reported.
Average CO concentrations have dropped
slightly in both temperature classes. Since
1987 or 1988, CO levels have increased but
still meet EPA's standard. The upturn in
1989 in the higher temperature category
results from higher CO levels in the August-
December period. CO levels average 3
times higher in this period compared to the
other months in 1989. Levels of CO
averaged 19% higher on the colder days.
The O3 trend in the ten highest O3 days, the
days with temperatures greater or equal to
80° F and for all days all show little change.
A trend test was not significant for any of the
O3 averages. Unlike many eastern cities,
1988 average daily maximum temperatures
on the ten highest O3 days were not
unusually high. The average daily maximum
temperature and O3 levels on these days
tracked well and were significantly correlated.
5-9
-------
I8BU8M
fc w
5-10
-------
Number of Days in PSI Categories
YEAR
85
87
Good
D
Moderate I
200
DAYS
B Unhealthful
Very Unheallhful
land Hazardous
Avg Daily Max 8-hr CO by Temperature
COppm
5
81
82
83
84
85 86
YEAR
TemP.V.40°F
87
88
89
90
All Days
Avg Daily Max 1-hr Ozone by Temperature
O3 ppm
0.16
o.u
0.12
0.1
0.08
0.06
0.04
0.02
0
TEMPERATURE
100
81
82
83
84
Avg on Days_-> 80 F
AIIJ}ays
85 86
YEAR
87
90
Ten High Q3 Days Avg
Avg Temp ol Ten High O3 days
Boston, MA
The Boston PMSA consists of Suffolk County
and parts of 6 other counties. The estimated
1987 population was 2.8 million. Its size and
location as a part of the eastern seaboard
megalopolis contribute to the area's air
pollution potential. There are 26 currently
active monitoring sites in this PMSA.
The PSI trend for Boston is based on 3 sites:
2 for CO and 1 for O3. The CO sites are a
maximum concentration site located at
Kenmore Square and a NAMS neighborhood
scale site located in east Boston. The O3
site is a maximum concentration site in
Sudbury (Middlesex County). The number of
PSI days > 100 fluctuates primarily due to
O3. This can be seen particularly in 1988
when the very hot summer caused an
increase in these days. Thirty-eight(78%) of
the 49 unhealthful or worse days reported
are due to O3. In 1990 there was one day
with a PSI value above 100. In the entire 10-
year period, only 1 day (in 1983) had a value
in the very unhealthful range. There were no
hazardous days reported.
Average CO levels declined in both
temperature categories. The trend is
significant in both of these cases and for all
days as well. Average CO levels over the
10-year period are 15% higher on the colder
days.
Average O3 concentrations show no clear
long-term trend for the ten highest O3 days,
the days with maximum temperature equal or
greater than 80° F and all days during the O3
season. Trend tests were not significant for
any of these averages. The impact of the
hot and dry 1983 and 1988 summers are
apparent in the O3 plot. Average daily
maximum temperature and O3 levels on the
ten highest O3 days do not relate well. The
correlation between these two variables was
not significant. However, the highest
average daily maximum temperature and
second highest O3 average occurred in 1988.
5-11
-------
0 10 20
Percent Frequency
5-12
-------
YE
81
82
83
84
85
86
87
88
89
90
0
Number of Days in PSI Categories
AR
*.. :. " | |
- . ^--" - ,. 1 J
* x4S.r \
> o.t r.J
- «"^x \ I
« T 1 1
£« V-* -.( g
«" 'J&&& i HI
" i i
- ^ ' 1 1
100 200 300
DAYS
400
Avg Daily Max 8-hr CO by Temperature
COppm
6 -
5 -i
4
4 -
3 -
2 -
1 -
,'""--.
£^ ~^v\
^^V^f^
81 82 83 84 85 86 87 88 89
YEAR
Temp_>_40°F Temp_<-_40°F All Days
A
o:
0.
0.1
0.
0.0
vg Daily Max 1-hr Ozone by Temperature
3 ppm TEMPERATUR
I 4 ' ", :.,,'", ' -^ ^. : ^^ ~ fl
5 - "".;.;. ' :::.£ ,.. ,V"1' .-...'.. ' . ; '
0 -v. i : : »:::: t i - ?. : . .'. i'::-:-: - 80 F Ten High Q3, Days Avg
Always Avg Temp of Ten High O3 days
*4
90
1
00
5
0
5
0
5
D
Chicago, IL
The Chicago PMSA consists of Cook,
Du Page and McHenry Counties. The
estimated 1987 population was 6.2 million
with 85% living in Cook County. Its size and
heavy industry contribute to the area's air
pollution potential. There are 42 currently
active monitoring sites located on the map.
The PSI trend for Chicago is based on 6
sites: 2 sites that monitored for both CO and
O3, plus 1 other CO site and 3 additional O3
sites. The CO sites are population oriented
sites in Cook County. The O3 sites include a
maximum concentration site in northwest
McHenry County and 4 population exposure
oriented sites. The number of days with PSI
> 100 is relatively stable except for 1983 and
1988. The impact of the very hot summers
of 1983 and 1988 on O3 levels is clear in the
PSI display. Ozone accounted for 27 of the
32 days having a PSI value above 100 for
these 2 years. CO accounts for most of the
PSI days > 100 in the first 5 years, while O3
is the main contributor in the last 5 years. In
1989 and 1990, there were 2 and 3 days in
the unhealthful or worse category. In the
entire 10-year period, only 2 days (both in
1988) were in the very unhealthful range.
There were no hazardous days reported.
There has been a significant decrease in
average CO concentrations. This occurred in
both temperature categories and for the
overall average as well. Unlike most cities
CO levels are higher on the warmer days,
averaging 16% higher.
Average O3 levels do not show a significant
long-term trend for any of the categories.
The impact of the very hot 1983 and 1988
summers can be seen in the O3 averages.
The 1990 O3 averages are the lowest
reported over the ten year period. Average
daily maximum temperature and O3 are
correlated significantly for the ten highest O3
days.
5-13
-------
5-14
-------
Number of Days in PSI Categories
YEAR
90
»
i
100
200
DAYS
300
400
v!
^Band
uunh±lhlul
Hazardous
Avg Daily Max 8-hr CO by Temperature
COppm
3.5
3
2.5
2
1.5
1
0.5
82
83
84
85 86
YEAR
Temp_<-_40°F
87
88
89
90
All Days
Avg Daily Max 1-hr Ozone by Temperature
O3ppm
0.16
0.12
0.08
0.06
0.04
0.02
TEMPERATURE
100
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on D£yg.-> 80 F Ten High QJ Days Avg
Avg Temp of Ten High O3 days
Dallas, TX
The Dallas PMSA consists of 6 counties with
75 percent of the population living in Dallas
County. The estimated 1987 population was
2.5 million. Its size and summertime
meteorology contribute to the area's air
pollution potential. The map shows 19
currently active monitoring sites for the area.
The PSI trend for Dallas is based on 3 sites:
1 for CO and 2 for O3. The CO site is a
maximum concentration site located in Dallas
County. The O3 sites are a maximum
concentration site in Denton County and a
population exposure oriented site in Dallas
County. The number of PSI>100 days have
declined over the 10-year period. This trend
was significant. The number of these days
averaged 12.4 in 1981-85 and 4.4 in 1986-
90. In Dallas, all but 1 of the PSI>100 days
are due to O3. As expected most (77%) of
these days occurred in the summer (June -
August). The fall months (September -
November) accounted for most of the rest of
these days (15%). In 1990 there were 5
days above a PSI of 100. In the entire 10-
year period only 1 day (in 1982) was in the
very unhealthful range. There were no
hazardous days reported.
Average CO levels have decreased in both
temperature categories and for the combined
days. The trend was significant in all three
cases. Average CO levels are 29% higher
on the colder days.
There has also been a significant decline in
average O3 levels on the ten highest days.
However, this amounted to only a 9% decline
in average O3 levels on the ten highest O3
days over the 10-year period. The other O3
categories did not show significant trends.
The correlation between the average daily
maximum temperature and O3 levels on the
ten highest O3 days was not significant.
5-15
-------
5-16
-------
Number of Days in PSI Categories
YEAR
100
o
Good Moderate
200
DAYS
lunhealthful
Very Unhoallhful
land Hazardous
Avg Daily Max 8-hr CO by Temperature
CO ppm
82
63
84
TertiD_>4 80 F Ten High QJ Days Avg
AllJiays Avg Temp of Ten High O3 days
Denver, CO
The Denver PMSA consists of 5 counties;
Denver is the most populated with 30 percent
of the area's 1.6 million residents, based
upon estimates for 1987. The area's size
and altitude contribute to its air pollution
potential. Seventeen monitoring sites are
currently active and are shown on the area
map.
The PSI trend for Denver is based on 4 sites:
2 where both CO and O3 are monitored plus
2 other CO sites. The CO sites are a
maximum concentration site located in
downtown Denver and 3 population oriented
sites in Arapahoe, Denver and Jefferson
Counties. The O3 sites are a maximum
concentration site in Arapahoe Co. and a
population exposure site in Jefferson Co.
The number of unhealthful or worse days
have dropped, especially in recent years.
This decrease was significant. Ninety-two
percent of the PSI days>100 were due to CO
and most (90%) occurred in the fall or winter.
There has been a significant decline in the
days>100 due to CO. Overall, the lowest
number of unhealthful days (7) occurred in
1990. For the second straight year (1990),
there were no very unhealthful days reported.
Three days in the 10-year period were
judged to be hazardous, the last occurring in
1985.
Average CO levels have declined significantly
for both temperature categories and for all
days. Average CO levels are 61% higher on
the colder days.
Average O3 levels have declined significantly
on the ten highest O3 days but not for the
other averages. The O3 averages on the ten
highest days ranged from a low of 0.10 ppm
in 1985, 1987 and 1990 to a high of 0.13
ppm in 1983. The trend in the ten highest O3
days is down despite the average
temperature being higher in the 1987-90
period. Average daily maximum temperature
and O3 levels on the ten highest days were
not correlated significantly.
5-17
-------
1-6 7-16 17-27 >=2B
Wind Speed (Knots)
0 10 20
Percent Frequency
5-18
-------
YE
81
82
83
84
85
86
87
88
89
90
C
CO
3
2.5
2
1.5
1
0.5
0
Number of Days in PSI Categories
AR
» l-X y- s SX J H
*. . - ...">--",,.. -- t HHI
* .. . -. j JBH
- r .:..: i i t
u
U
- "?>:..;";* -^ .-.! i
'""" ^ 1 ' '
* -. . ^ ^ - -.". 40°F Temp_<-_400F All Days
g Daily Max 1-hr Ozone by Temperature
>pm TEMPERATUR
- 9
cf^*"" TT."*-^ «»»ii._ :<:^^!*'":^""""'*T>'V«N,^
:v'i:..-.;?;^.:'li:::..,i; 5s$iL.-x.!'Vx:..:.Jv^ ,.:-: ''''-. ..-":.''. - ^
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Dgy^.> 80 F Ten High 23 Days Avg
Alleys Avg Temp of Ten High O3 days
^
90
1
00
5
0
5
0
5
D
Detroit, MI
The Detroit PMSA consists of 7 counties with
73% of the population living in Wayne and
Oakland counties. The estimated 1987
population was 4.4 million. It's size and
industry contribute to the area's air pollution
potential. A total of 21 monitoring sites are
currently active in the PMSA - 18 of these
sites are located in that portion of the PMSA
shown on the map.
The PSI trend is based on 5 sites where both
CO and O3 are monitored, plus 1 additional
site for CO and 3 other sites for O3. Four of
the CO sites are located in Wayne County,
while Macomb and Oakland Counties each
have 1 site. All of these CO sites are
identified as population exposure oriented.
There are 2 maximum concentration oriented
O3 sites among the trend sites - 1 each in
Macomb and Wayne Counties. The number
of PSI days greater than 100 were highest in
1981, 1982, 1983, 1988 and 1989 -
averaging 17 days for these years. The
lowest number of these days (2) occurred in
1985; while, the second lowest (3) occurred
in 1990. There was not a significant trend in
these PSI days. Eighty percent of these
days occurred in the summer and fall.
Seventy-nine percent of these days over the
10-year period were due to O3. The only
very unhealthful day occurred in 1988. There
were no hazardous days reported.
Average CO levels showed a significant
decline in the lower temperature category
and for all days. Average CO levels are
higher (21%) on the colder days. Average
O3 levels did not show a significant trend for
the three averages presented. The average
daily maximum temperature on the ten
highest O3 days was highest in 1988 (95° F)
and lowest in 1982 (81° F). These two years
had the highest average O3 levels on the ten
highest days. In 1982, the highest average
O3 was associated with the lowest average
daily maximum temperature. Average daily
maximum temperature and O3 levels on the
ten highest O3 days were not significantly
correlated.
5-19
-------
-6 7-16 17-27 >=28
.Wind Speed (Knots)
0 10 20
Percent Frequency
5-20
-------
YE
81
82
83
84
85
86
87
88
89
90
C
CO
3.5
3
2.5
2
1.5
1
0.5
0
Number of Days in PSI Categories
AR
.
- : > s i HUB
^ <." } Htti
:. .>. ; .l.:,-^. ..>£ 1 ,
V
".......;;.. j.'.;...^;; >.... \ . IB!
'" '- " " " - i HP
" \ s .--- 5-- i TB&
""- = VV-^₯ -s\-} jpll
- -<- -| imi
..,.,,..,, i uei
1 HKH
* . i . i
100 200 300
DAYS
400
Avg Daily Max 8-hr CO by Temperature
ppm
.*
81
Av
O3|
0.25
0.2
0.15
0.1
0.05
0
/ -"'
'/^==^=^ ^-^__^-
82 83 84 85 86 87 88 89
YEAR
Tem^>_40°F Temp^-^'F All Days
g Daily Max 1-hr Ozone by Temperature
>pm TEMPERATUR!
- 8
- 7
. i ' : ' , .₯ '' ' . .-." fr'- l .... 1 ...'- 'i .'*"-' 1 :l I
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Days.-> 80 F Ten High Q3, Days Avg
AllJiays Avg Temp of Ten High O3 days
r-
90
I
DO
5
3
5
3
3
Houston, TX
The Houston PMSA consists of the principal
county of Harris and 4 other counties. The
estimated 1987 population was 3.2 million
with 86 percent living in Harris County. Its
size and industry, mainly petroleum
refineries, contribute to the area's air
pollution potential. Its high temperatures and
proximity to the Texas City-Galveston area
are also factors which contribute to its air
pollution potential. There are 17 currently
active monitoring sites located on the map.
The PSI trend for Houston is based on 5
sites: 2 sites where both CO and O3 were
monitored, 1 additional site for CO and 2
additional sites for O3. All of these sites are
located in Harris County and include a
maximum concentration site for each of these
pollutants. The other sites are population
exposure oriented. The number of PSI days
> 100 is stable over the period. The lowest
number of unhealthful days (19) occurred in
1989; while, the highest (43) occurred in
1983. The very unhealthful days range from
a low of 0 in 1987 to a high of 8 in 1981 and
1983. During the 10-year period, 37 (12%)
of the 306 days of PSI>100 were very
unhealthful. In Houston, 97 percent of the
PSI days > 100 are due to O3. There were
no hazardous days reported.
Average CO levels are stable over the
10-year period and do not show a significant
trend. The 10 year CO average on the days
with a minimum temperature below 40° F
was 51 percent higher than on the warmer
days.
Average O3 levels were also stable for all the
O3-temperature categories and did not show
a statistically significant trend. The lowest
average O3 concentration (0.17 ppm) on the
ten highest O3 days occurred in 1987.
Average daily maximum temperature on the
ten highest O3 days ranged from a low of 83°
F in 1985 to a high of 92° F in 1986.
Average daily maximum temperature and O3
on the ten highest days were not significantly
correlated.
5-21
-------
Kind Spud (Knoti
0 10 20
P« rein t Friquincy
5-22
-------
YE
81
82
83
84
85
86
87
88
89
90
C
CO
4
3.5
3
2.5
2
1.5
1
0.5
0
A
0,
0.1
0.1
0.
0.0
0.0
0.0
0.0
Number of Days in PSI Categories
AR
»'-.''{ I
'."". > --... * - I 1
., -. -. I 1
- 5-^x...^1* m. ' - ..I., il H
- ^r:::::::-" *:::::. :;i i
- "" " 1 '
" ' - i
M > 1 |
H " v " J |
^;:;:;:;:;j |
100 200 300
DAYS
Qc3ood [^Moderate Qjunhealthful B^H.^*!^1
Avg Daily Max 8-hr CO by Temperature
ppm
^_
^^*1^t-~:;-^
^^rr^r:
81 82 83 84 85 86 87 88 89
YEAR
Tem£»40°F Tempi-<-_40'>F All Days
ivg Daily Max 1-hr Ozone by Temperature
3 ppm TEMPERATUR
1 ~ ' -...
' .. - 9
" - 8
4
2 - ' ?
0 .-..» * < '-> '- -i----- -' -.« - r..'.:'....t '...i.!.: . i -
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Days..* 80 F Ten High Q3. Days Avg
AllJJays Avg Temp of Ten High O3 days
400
»
90
1
00
5
0
5
0
5
0
Kansas City, MO-KS
The Kansas City MSA consists of 6 counties
in Missouri and 4 counties in Kansas. The
estimated 1987 population was 1.5 million.
Its size and summertime meteorology
contribute to the area's air pollution potential.
Its chief air quality problems occur during the
summer when the hot, stagnant days are
conducive to O3 formation. The map shows
21 currently active monitoring sites in this
PMSA.
The PSI trend for Kansas City is based on 7
monitoring sites: 1 site where both CO and
O3 are monitored plus 2 other CO and 4
other O3 sites. These CO sites are all
population exposure oriented, while the O3
sites include 1 maximum concentration site
and 4 population exposure sites. The largest
number(12) of unhealthful days occurred in
1984. The last 2 years had 2 unhealthful
days each. O3 accounted for 33 (70%) of the
47 unhealthful days during the 10-year
period. There were 9 unhealthful days
attributed to PM-10 - the last occurred in
1989. Sixty-four percent of these days
occurred during the summer (June-August).
There is a significant downward trend in the
number of these days. In the 10-year period
there were no days in the very unhealthful or
worse ranges.
Average CO levels have dropped significantly
over the 10-year period in both temperature
categories and overall. CO levels are higher
on the colder days especially in the latter half
of the period.
Average O3 levels show no clear long-term
trend over the 10-year period for all
temperature categories. Average daily
maximum temperature on the ten highest O3
days varied from a low of 86° F in 1982 to a
high of 98° F in 1983. Visually the average
daily maximum temperature and O3 on the
ten highest days appear to track together.
This was confirmed when these variables
had a significant positive correlation.
5-23
-------
5-24
-------
Number of Days in PSI Categories
YEAR
400
iGoodl
Moderate
I Unhealthful
Kery Unhealthful
nd Hazardous
Avg Daily Max 8-hr CO by Temperature
COppm
81
82
83
84
85 86
YEAR
87
88
89
90
Avg Daily Max 1-hr Ozone by Temperature
O3 ppm TEMPERATURE
0.4
0.3
0.2
0.1
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Dayg.-> 80 F Ten High QJ Days Avg
AllJJays Avg Temp of Ten High O3 days
100
95
90
85
80
75
70
Los Angeles, CA
The Los Angeles PMSA consists of Los
Angeles County, where an estimated 8.5
million people lived in 1987. The Los
Angeles "basin" is bounded by the Pacific
Ocean on the west and south and several
mountain ranges on the north and east. Its
complex meteorology is characterized by a
land-sea-breeze circulation, frequent
inversions and a high incidence of sunlight.
There are 21 currently active monitoring sites
located on the map.
The PSI trend for Los Angeles is based on
13 sites: 11 where both CO and 03 are
monitored plus 2 other O3 sites. For each
pollutant, there is a NAMS maximum
concentration site; the others are all
population oriented SLAMS sites. Los
Angeles has the largest number of PSI days
> 100 of any urban area - averaging 201 per
year for the 10-year period. The trend in
these days is essentially flat; however, there
has been a 60 percent reduction in the very
unhealthful days over the 10-year period.
The number (37) of very unhealthful days in
1990 was the lowest reported for the 10-year
period. The smallest number (163) of days
with a PSI > 100 occurred in 1990. In 1990,
these 163 days were broken down into 126
unhealthful and 37 very unhealthful days. In
the Los Angeles PMSA, 73 percent of the
PSI days > 100 are due to O3. In the 10-
year period, only 1 day (1982) was in the
hazardous category.
There is a significant downward trend in
average CO levels on all days. There were
insufficient days in the lower temperature
category so the temperature category
averages are omitted.
Average O3 levels showed a significant
downward trend over the period for the ten
highest O3 days and for days with
temperatures of 80° F or above. The lowest
O3 average occurred in 1990 for both the ten
highest O3 days and for all days. The
temperature data were taken at the Los
Angeles Civic Center in downtown LA.
5-25
-------
, Wind Speed (Knots)
-------
Number of Days in PSI Categories
YEAR
f.
100
200
DAYS
300
Avg Daily Max 8-hr CO by Temperature
COppm
10
81 82 83 84 85 86 87 88 89 90
YEAR
TemD_>_40° F
All Days
Avg Daily Max 1-hr Ozone by Temperature
O3 ppm
0.2
0.15
0.1
0.05
TEMPERATURE
100
: "i ..: T' . - I '
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Days.-> 80 F Ten High QJ Days Avg
Avg Temp of Ten High O3 days
New York, NY
The New York PMSA consists of 8 counties
with 81 percent of the 8.5 million population
in Bronx, Kings, New York and Queens
Counties. Its size and location as a part of
the eastern seaboard megalopolis contribute
to the area's air pollution potential. Twenty-
four monitoring sites are operating in the
PMSA - twenty-two of which are located in
the portion of the PMSA shown on the map.
The PSI trend for New York is based on data
from 8 sites: 3 for CO and 5 for O3. The CO
sites are 2 maximum concentration sites
located in Manhattan (New York County) and
a population exposure site in Kings County.
The O3 sites include a maximum
concentration site in Westchester County.
Unlike most areas, PSI days > 100 are due
more to CO (58%) than O3 over the 10-year
period; however, O3 has contributed more
PSI > 100 days since 1987. In 1983 and
particularly 1988 the impact of the very hot
summers is seen with the increase in the
number of these days due to O3. The most
dramatic improvement has been in CO where
the number of these days declined from 79 in
1981 to 2 in 1990. The trend in the number
of unhealthful or worse days was significant
for all days and for days when CO was the
responsible pollutant. There was a total of 9
very unhealthful days reported, but none in
1989 or 1990. No hazardous days were
reported.
Average CO levels showed a significant drop
over the period for both temperature
categories and for all days. Unlike most of
the other urban areas studied, CO levels are
higher on the warmer days. The CO average
on the warmer days was 12% higher.
Average O3 levels also decreased
significantly for all days and for days with
temperatures of 80° F or higher. Average
daily maximum temperature for the ten
highest O3 days varied from a low of 87° F in
1982 to a high of 96° F in 1988. The
correlation between average daily maximum
temperature and O3 on the ten highest O3
days was low and not significant.
5-27
-------
5-28
-------
Number of Days in PSI Categories
YEAR
90 -
_±
100
200
DAYS
oood
u
Hazardous
Avg Daily Max 8-hr CO by Temperature
COppm
81
82
83 84 85 86
YEAR
87
TemD.>40°F
Temp.<-.40°F
88 89
All Days
90
Avg Daily Max 1-hr Ozone by Temperature
O3ppm
0.2
0.15
0.05
TEMPERATURE
100
. :*.;;:...
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Days.-> 80 F Ten High Q£ Days Avg
AllJJays Avg Temp of Ten High O3 days
Philadelphia, PA
The Philadelphia PMSA consists of 8
counties, 5 in Pennsylvania and 3 in New
Jersey. The most populated county is
Philadelphia which accounts for 33 percent of
the total population. The estimated 1987
population was 4.9 million. Its size and
location as a part of the eastern megalopolis
contribute to the area's air pollution potential.
There are 29 currently active air monitoring
sites shown on the map.
The PSI trend for Philadelphia is based on
data from 15 sites: 4 sites where both CO
and O3 are monitored, 5 additional CO sites
and another 6 O3 sites. The CO sites include
two maximum concentration sites located in
Philadelphia and Burlington Co., New Jersey.
The O3 sites include maximum concentration
sites in New Jersey for Burlington,
Gloucester and Camden counties. The trend
in the PSI days > 100 did not show a
significant trend over the 10-year period but
the number of these days declined in 1989-
90. The total of 11 of these days in 1990
was the lowest reported. The next lowest
was 19 in 1989. The number of unhealthful
or worse days due to O3 dropped from a high
of 52 in 1983 to a low of 11 in 1990. Eighty-
seven percent of the PSI days > 100 are due
to O3. In the 10 years, 12 days were in the
very unhealthful range. No days in the
hazardous range were reported.
Average CO levels have declined significantly
over the 10-year period in both temperature
categories and for all days. CO levels are
11% higher on the colder days.
Average O3 levels do not show a significant
trend; although, for all three averages
presented the two lowest averages occurred
in 1989 and 1990. Average daily maximum
temperature for the ten highest O3 days
ranged from a low of 85° F in 1982 to a high
of 96° F in 1988. Average daily maximum
temperature and O3 levels on the ten highest
O3 days were not significantly correlated.
5-29
-------
0 10 20
Percent Frequency
5-30
-------
Number of Days in PSI Categories
YEAR
j.
100
llGood
D
Moderate
200
DAYS
lunhealthful
300
400
ery Unhoalthtul
nd Hazardous
Avg Daily Max 8-hr CO by Temperature
COppm
81
83
84
85 86
YEAR
Temp_<-_40°F
87
89
90
All Days
Avg Daily Max 1-hr Ozone by Temperature
O3 ppm
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
TEMPERATURE
100
81
82 83
84
Avg on Days.» 80 F
Always
85 86
YEAR
87
88 89
90
Ten High Q3 Days Avg
Avg Temp of Ten High 03 days
Pittsburgh, PA
The Pittsburgh PMSA consists of 4 counties,
with 65 percent of the population living in
Allegheny County. The estimated 1987
population for the entire area was 2.1 million.
Its size and heavy industry contribute to the
area's air pollution potential. There are 36
currently active monitoring sites shown on
the map.
The PSI trend is based on data from 12
sites: 1 where CO, O3 and SO2 are all
monitored, 3 where both O3 and S02 are
measured plus 8 other monitoring sites (2
CO, 1 O3 and 5 SO2). Each of these
pollutants had a NAMS maximum
concentration site in Allegheny County. The
other sites were all population exposure
oriented. The number of unhealthful or
worse days varied from a low of 6 in 1985 to
36 in 1983. Ozone accounted for most of
these days in 1988 when the meteorology
was especially conducive for O3 formation. A
trend test was not significant on the number
of these days over the 10 years. Thirty-
seven percent of all PSI days > 100 are due
each to O3 and SO2. The remainder of high
PSI days are due to CO (14 percent) and
PM-10 (12 percent). In the 10-year period, 2
days (the last in 1985) fell in the very
unhealthful category. No hazardous days
were reported.
Average CO levels increased from 1981 to
1983 and then decreased but this mixed
pattern resulted in no significant overall trend.
Average CO levels are slightly higher on the
warmer days.
Average O3 levels show no clear long-term
trend, with peaks occurring in 1983 and
1988. However, the lowest average for each
03 category occurred in 1990. Average daily
maximum temperature for the ten highest O3
days varied from a low of 83° F in 1982 to
96° F in 1988. The correlation was not
significant between the average daily
maximum temperature and O3 on these days.
5-31
-------
5-32
-------
Y
81
82
83
84
85
86
87
88
89
90
Number of Days in PSI Categories
EAR
.,*,::, v.-'- *» :--- -. " ' -- I
* s \ & vXW1-* ,-,\V \\ v^vtvnstttHHMSv.^ f\ 1
t* -I -. \ %" v. .** "» \vt\\y.'' \ sssss \ |
.... , ^ ^ .,. . % . ... j
;:"; ' -s-- v o-;- ,^x « ' |
*"" -- \ - - ?' ' -.- -- j
« ' " %« \^^ >% ' % ' %' {
M,'^ ' ' ' ' V~% "^ " "\« ;« v "" J
--- -^ - >.>:<::.:::...> i
-^ ^-^ :. J" " ,, 1
I.I.I.
0 100 200 300
DAYS
Avg Daily Max 8-hr CO by Temperature
CO ppm
5
4
2
1
/' %
..../ \
^^ "^^^>J'^
81 82 83 84 85 86 87 88 89
YEAR
Temp_>40°F Temp<<>.._40'>F All Days
/
0
0.
c
0.
0.
0.
Wg Daily Max 1-hr Ozone by Temperature
3 ppm TEMPERATUR
.1 - / ^""><1»»^t-^ - 9
08 - ~ ^ '' ' :::;:;.: \(/' ^**^»v - 9
06 '. ::-..- &;;'.:'£:.. ^$J~~'' 1 '**."* V^ - 8
;;;.;.;;::;.;:. , ,.f -::'- :~X~-/; Tff" ifV^- ! \ '
1 M./^Hl^f'" .:;P..:-\-is-xC." '*" :ix,-. ..".'.' ' !*"" '
02 -,,4v: -.-M^iV" ' ::i;y ^'.-i'^vs'. :>'.-'-,....^r; :'..... \ , -7
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on Dgys_-> 80 F Ten High Q3 Days Avg
Always Avg Temp of Ten High O3 days
400
90
j
00
5
0
5
0
5
0
San Francisco, CA
The San Francisco PMSA consists of Marin,
San Francisco and San Mateo Counties.
The estimated 1987 population was 1.6
million. Its urban area size contributes to the
area's air pollution potential. There are 5
currently active monitoring sites located on
the map.
The PSI trend for San Francisco is based on
data from 3 sites: 2 monitoring both CO and
O3 and additional one site for CO. The CO
sites are a NAMS maximum concentration
site located in San Francisco County and 2
population exposure sites in Marin and San
Mateo Counties. The O3 data are from 2
population exposure sites in Marin and San
Mateo Counties. The number of PSI days >
100 average slightly more than 2 days per
year. The largest number of these days (5)
occurred in 1985. In 1990, there was one
day reported. A trend test did not show a
significant trend in these days over the 10-
year period. In San Francisco, 76% of the
PSI days > 100 are due to CO. In the entire
10-year period, only 1 day (in 1985) was in
the very unheatthful range. No hazardous
days were reported.
Average CO levels did not show a significant
trend for any of the averages. CO averages
are 29% higher on the colder days.
Average O3 levels are also stable - not
showing a significant trend even though the
1990 averages are the lowest reported. The
highest average O3 levels for the ten highest
O3 days and for the days with temperatures
of 80° F and higher occurred in 1983. The
highest O3 average for all days occurred in
1984. Average daily maximum temperature
for the ten highest O3 days ranged from a
low of 76° F in 1990 to a high of 92° F in
1984. The average daily maximum
temperature and O3 levels correlated well on
these days. The correlation (-0.82) was
highly significant.
5-33
-------
(I 10 20
Percent Frequency
5-34
-------
YE
81
82
83
84
85
86
87
68
89
90
C
Number of Days in PSI Categories
AR
'* i ' " ' ««
- ..: i M
..,.-.- , 1
''":-":-: T-"^-^ . 1
" '" - " - - i H9B
T iV" -"'"I P|
" ""-1-- \
*. "-. ' >. v.^» " >" 1
«- -- - ^- s- 1
, : *m i
i.i.i.
100 200 300
DAYS
FWlGood Fixate Bunhearthful *'* ""'"I*1'"'
t:xf:i 1 1 BH and Hazardous
400
Avg Daily Max 8-hr CO by Temperature
COppm
<
6 -|
4 -
2 -
\.,-"\
^^IZ^rrr^
^^^^
81 82 83 84 85 86 87 88 89
YEAR
Temgi>400F Temp_F All Days
A
o:
0.1
0.
0.0
0.0
0.0-
0.0
vg Daily Max 1-hr Ozone by Temperature
) ppm TEMPERATURE
^"'' \X""~"" ' "*x S
5 ' ... " . NX - 8
2 I."- '- .. ."" ...-. .- . - 7
. :i :-:':-. x T..V:.'. 1 '.: !:'? I::..-:. ..-.I... .- ₯ t '. 1 : . I
81 62 83 84 85 86 67 88 69 90
YEAR
Avg on Dgyj.-* 80 F Ten High 2J Days Avg
Always Avg Temp of Ten High O3 days
»
?*
90
I
DO
5
3
3
)
Seattle, WA
The Seattle PMSA consists of King and
Snohomish Counties. Seventy-seven percent
of its population lives in King County. The
estimated 1987 population was 1.8 million.
Twenty-three currently active monitoring sites
are shown on the map.
The PSI trend for Seattle is based on 7 sites:
6 for CO and 1 for O3, all located in King
County. There are 2 maximum concentration
CO sites and 4 population exposure sites.
The O3 site is a population exposure site.
The number of PSI days > 100 are
dominated by CO, which accounts for 141
(91%) of these days over the 10-year period.
There has been a significant improvement in
these days. In 1990, for the first time, CO
did not account for any PSI > 100 days at
these trend sites. However, one of the
maximum concentration CO sites did not
report data in 1990. The 2 PSI > 100 days
reported in 1990 were from O3. Seventy-two
percent of the unhealthful or worse days
occurred in the winter. The 2 very
unhealthful days occurred in 1981. CO was
responsible for both of these days. No
hazardous days were reported.
Average CO levels showed a significant
decrease over the 10-year period for both
temperature categories and for all days. CO
levels are 28% higher on the colder days.
Average O3 levels are stable for the 3
averages presented. However, the 1990 O3
averages increased along with the average
daily maximum temperature on the ten
highest days, which was the highest average
daily maximum temperature reported (89.5°
F). The highest average O3 levels for the ten
highest O3 days and for days with
temperatures of 80° F or higher occurred in
1981; while, the lowest O3 averages for these
categories occurred in 1989. The lowest
average daily maximum temperature (78° F)
on the ten highest days occurred in 1983.
The correlation between average daily
maximum temperature and O3 levels was
positive and significant.
5-35
-------
5-36
-------
Y[
81
82
83
84
85
86
87
88
89
90
Number of Days in PSI Categories
EAR
* ^ "' ' I HH
.« .&. ....;. ..^..r H&a
.,«*.>: \..>.j . ... .... iilliil
* $-^^-....^.- i ... ... . wss
-.^^v.x.w v i p
T" ; T5V", 'Uj J . BH
- '««-:'':' *'.'. I mesa
K v, % '. ts^v.%' I 1
' ' < , i
0 100 200 300
DAYS
Avg Daily Max 8-hr CO by Temperature
COppm
5
4
3
2
1
400
f. x.
*"
81 82 83 84 85 86 87 88 89
YEAR
Temp_>40° F Temp_**''** ^-""""vis. - 8
.** >*"*"*»* X-.,* :
)5 -
-.-. x .-7
Q --» » t 1 1 1 » t i i _{
81 82 83 84 85 86 87 88 89 90
YEAR
Avg on D|yj.-> 80 F Ten High QJ Days Avg
All JJays Avg Temp of Ten High 03 days
90
J
X
5
5
5
)
Washington, DC-MD-VA
The Washington MSA consists of 10
counties, the District of Columbia (DC) and 5
independent cities. The principal population
centers are DC, Fairfax County in Virginia
and Montgomery and Prince Georges
Counties in Maryland. The estimated 1988
population was 3.6 million. Its size and
location as a part of the eastern seaboard
megalopolis contribute to the area's air
pollution potential. A total of thirty-two
currently active monitoring sites are operating
in the PMSA - 29 of these sites are located
in that portion of the PMSA shown on the
map.
The Washington PSI trend is based on data
from 14 sites: 3 CO, 4 O3 and 7 where both
pollutants were monitored. Both CO and O3
had 1 maximum concentration site reporting
data. The maximum concentration O3 site is
located in Prince Georges County, Maryland,
while the maximum concentration site for CO
is located in DC. The number of unhealthfu!
or worse days varied from a high of 53 in
1983 to a low of 5 in 1990. The number of
days attributed to CO declined significantly,
averaging 13 days for the first 4 years and 2
for the last 4 years. Ozone accounted for all
but 1 of the 34 unhealthful or worse days in
the very hot summer of 1988. In the 10-year
period, 8 days fell in the very unhealthful
category; the last occurred in 1987. No
hazardous days were reported.
Average CO levels showed a significant
decline over the ten years for all 3 averages
presented. CO levels were 19% higher on
the colder days. Average O3 levels show no
clear long-term trend over the 10 years for
any of the averages. Once again the effect
of the very hot and dry summers of 1983 and
1988 can be seen in average O3
concentrations. The average daily maximum
temperature for the ten highest O3 days
varied from a low of 89° F in 1983 to a high
of 94° F in 1987 and 1988. The correlation
was not significant between the temperature
and O3 levels on the ten highest days.
5-37
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA 450/4-91-023
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
National Air Quality and Emissions Trends
Report, 1990
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T. Curran, R. Faoro, T. Fitz-Simons, N. Frank,
W. Freas, B. Beard, W. Frietsche, M. Stewart, and W. F. Hunt, Jr.
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
The computer graphics were prepared by W. Freas, B.Beard and T. Fitz-Simons
and the typing by H. Hinton. Tom Rosendahl and Barry Gilbert prepared the nonattainment maps:
16. ABSTRAC7
This report presents national and regional trends in air quality from 1981
through 1990 for paniculate matter, sulfur dioxide, carbon monoxide, nitrogen dioxide,
ozone and lead. Air quality trends are also presented for 15 metropolitan areas. Both
national and regional trends in each of these pollutants are examined. National air quality
trends are also presented for both the National Air Monitoring Sites (N AMS) and other site
categories. In addition to ambient air quality, trends are also presented for annual
nationwide emissions. These emissions are estimated using the best available
engineering calculations; the ambient levels presented are averages of direct
measurements.
This report also includes a section, Air Quality Levels in Metropolitan Statistical Areas
(MSAs). Its purpose is to provide interested members of the air pollution control
community, the private sector and the general public with greatly simplified air pollution
information. Air quality statistics are presented for each of the pollutants for all MSAs with
data in 1990.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution Trends
Emission Trends
Carbon Monoxide
Nitrogen Dioxide
Ozone
Sulfur Dioxide
Total Suspended Particulates
Particulate Matter
Lead
Air Pollution
Air Quality Standards
National Air Monitoring
Stations (NAMS)
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (TinsReport)
21. NO. OF PAGES
142
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (R«y. 4-77) PREVIOUS EDITION is OBSOLETE
*US GOVERNMENT PRINTING OFFICE 1992-62T-T95
------- |