EPA-450/3-76-034
October 1976
DETECTION
AND INTERPRETATION
OF TRENDS
IN OXIDANT AIR QUALITY
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/3-76-034
DETECTION AND INTERPRETATION
OF TRENDS
IN OXIDANT AIR QUALITY
by
Lowell Wayne, Katherine Wilson, and Clarence Boyd
Pacific Environmental Services, Inc.
1930 Fourteenth St.
Santa Monica, California 90404
Contract No. 68-02-1890
Project No. TO No. 1
EPA Project Officer: Edwin L. Meyer, Jr.
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
October 1976
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35), Research Triangle Park, North Carolina
27711; or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency
by the Pacific Environmental Services, Inc. , Santa Monica, California,
in fulfillment of Contract No. 68-02-1890, Project No. TO #1. The
contents of this report are reproduced herein as received from the
Pacific Environmental Services> Inc. The opinions, findings, and
conclusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency. Mention of company
or product riames is not to be considered as an endorsement by the
Environmental Protection Agency.
Publication No. EPA-450/3-76-034
11
-------
EXECUTIVE SUMMARY
This document presents a study of trends in oxidant air
qual.ity and their relation to trends in the emissions of oxidant
precursors (namely, hydrocarbons and oxides of nitrogen), utilizing
information from locations for which relevant data were available,
representing five or more years of essentially continuous air
monitoring.
Locations reviewed include five in California (not including
the South Coast Air Basin), one in Colorado and one in Pennsylvania.
Statistically significant trends in oxidant air quality were detected
from the data only for the San Francisco Bay Area (California) and
for the Metropolitan Denver Air Quality Control Region (Colorado).
For both of these areas, the data yielded convincing evidence of
trends toward lower oxidant concentrations over periods of five
years or more. For the other locations reviewed, indications of
trends were found, but the evidence was not unequivocal. An
auxiliary analysis showed that oxidant air quality is not uniform
throughout the San Francisco Bay Area Air Basin and that there
are important differences in air quality related to the geographi-
cal distribution of stations with the Basin.
Comparing oxidant trends to trends in emissions of the
precursors (hydrocarbons and oxides of nitrogen), the evidence
showed that a nearly steady reduction of oxidant levels in the
San Francisco Bay Area during the years from 1962-1974 was
associated with a substantial decline of hydrocarbon emissions
and a simultaneous major increase in the emissions of oxides of
nitrogen, both consequences mainly of changes in emissions from
mobile sources resulting from motor vehicle pollution control
programs.
-------
It was not possible to interpret the reasons for the
significant improvement of air quality in the Denver area in
terms of trends in the emissions of precursors, partly because
of limitations of the data base for precursor emissions.
Within the single meteorological system which includes
the San Francisco Bay Area and the Central Valley of California,
oxidant problems were found to be most severe at inland valley
locations and least severe in populous, industrial Bayside
communities. These conditions seem to be related to two princi-
pal factors: the superior ventilation of communities along the
main waterfronts, and the local oxidant-reducing effect (in such
communities) of the oxides of nitrogen emitted by the more
concentrated industry and motor vehicle traffic. Both of these
protective circumstances tend to fade as the marine air moves
downwind to inland valley locations.
To recapitulate, the main conclusions resulting from this
project are:
1. Long-term trends in oxidant air quality in the San
Francisco Bay Area are significantly related to trends in hydro-
carbon emissions.
2. Oxidant concentrations in the San Francisco Bay Area
are inversely affected by oxides of nitrogen emissions; thus, re-
duction of these emissions tends to increase measured oxidant
levels, on the whole.
3^ In the San Francisco Bay Area, oxidant problems tend
to be more severe in inland valley locations than in industrial,
Bayside locations.
4. The oxidant air quality of communities in the Central
Valley of California is appreciably affected by the flow of pol-
luted marine air from the adjacent San Francisco Bay Area.
11
-------
5. Monitoring data for a five-year period is not, in
general, sufficient to establish statistically significant oxi-
dant trends for individual monitoring stations.
6. A significant reduction of oxidant levels occurred in
Denver during the period 1968-1973. This cannot be attributed
with confidence to observed trends in emissions.
7. The second highest daily high-hour oxidant value for
a single year at a single station is a statistically useful mea-
sure of oxidant air quality.
m
-------
ACKNOWLEDGEMENTS
Pacific Environmental Services wishes to acknowledge, with
gratitude, the generous assistance of the persons listed below,
in furnishing data and answering inquiries essential to the suc-
cess of this project.
California Air Resources Board:
. H. Linnard, R. Tate, R. McMullen, R. Grewal,
R. Kuhlman, S. Duckworth, J. Paskind
Bay Area Air Pollution Control District:
W. Crouse, R. Villanueva, T. Storey, W. Siu,
J. S. Sandberg
Shasta County Air Pollution Control District
R. Nevins, B. Hill
Stanislaus County Air Pollution Control District:
M. Boise, W. Morgan
San J.oaquin County Air Pollution Control District:
C. Duns ing, J. Spano .,
Monterey Bay Unified Air Pollution Control District:
J. Maloney, J. Abercrombie
Southern California Air Pollution Control District:
W. Zwiacher, A. Davidson
Philadelphia Department of Public Health, Air Management
Services:
M. Goldner, R. Ostrowski
Delaware Valley Regional Planning Commission:
A. Toizer
Colorado Department of Health, Air Pollution Control
Division:
. S. Arnold, J. Spiegel, A. Stewart
IV
-------
TABLE OF CONTENTS
Chapter Page
EXECUTIVE SUMMARY i
ACKNOWLEDGMENTS iv
I. INTRODUCTION 1-1
REFERENCES 1-3
II. LOCATIONS SELECTED FOR STUDY II-l
A. INTRODUCTION II-l
B. THE SAN FRANCISCO BAY AREA II-4
1. PHYSICAL DESCRIPTION II-4
2. MONITORING STATIONS 11-8
3. POPULATION AND GROWTH FACTORS 11-13
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-15
C. REDDING (SACRAMENTO VALLEY AIR BASIN) 11-17
1. PHYSICAL DESCRIPTION 11-17
2. AIR MONITORING 11-18
3. POPULATION AND GROWTH FACTORS 11-18
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-18
D. STOCKTON AND MODESTO (SAN JOAQUIN VALLEY
AIR BASIN) 11-20
1. PHYSICAL DESCRIPTION 11-20
2. AIR MONITORING 11-20
3. POPULATION AND GROWTH FACTORS 11-21
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-22
E. SALINAS (NORTH CENTRAL COAST AIR BASIN) 11-23
1. PHYSICAL DESCRIPTION 11-23
2. AIR MONITORING 11-25
3. POPULATION AND GROWTH FACTORS 11-25
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-25
F. LANCASTER (SOUTHEAST DESERT AIR BASIN) 11-27
1. PHYSICAL DESCRIPTION 11-27
2. AIR MONITORING 11-28
3. POPULATION AND GROWTH FACTORS 11-28
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-28
G. DENVER, COLORADO 11-30
1. PHYSICAL DESCRIPTION 11-30
2. AIR MONITORING 11-30
3. POPULATION AND GROWTH FACTORS 11-32
4. INDUSTRY AND OTHER EMITTING ACTIVITIES.. 11-33
-------
TABLE OF CONTENTS (continued)
Chapter Page
H. PHILADELPHIA, PENNSYLVANIA 11-34
1. PHYSICAL DESCRIPTION 11-34
2. AIR MONITORING 11-34
3. POPULATION AND GROWTH FACTORS 11-34
REFERENCES 11-36
III. OXIDANT TREND ANALYSIS III-l
A. INTRODUCTION III-l
B. SECOND HIGHEST MAXIMUM III-2
C. AVERAGE MAXIMUM III-2
D. THIRD QUARTER AVERAGE MAXIMUM III-5
E. FREQUENCY EXCEEDING NAAQS LEVEL 111-7
F_ FREQUENCY EXCEEDING TWICE NAAQS LEVEL III-7
G. RELATIONS BETWEEN THE INDICES III-7
H. REVIEW AND CONCLUSIONS... 111-14
REFERENCES. 111-18
IV. OXIDANT:PRECURSOR RELATIONSHIPS IV-1
A. SOUTH COAST AIR BASIN, CALIFORNIA IV-1
B. REVIEW OF EMISSIONS AND TRENDS,
1970-1974 IV-3
1. BAY AREA AIR POLLUTION CONTROL DISTRICT... IV-3
2. REDDING (SACRAMENTO VALLEY AIR BASIN) IV-10
3. STOCKTON AND MODESTO (SAN JOAQUIN
VALLEY AIR BASIN) IV-12
4. SALINAS (NORTH CENTRAL COAST AIR BASIN)... IV-18
5. LANCASTER (LOS ANGELES COUNTY PORTION
OF SOUTHEAST DESERT AIR BASIN) IV-20
6. DENVER, COLORADO (METROPOLITAN DENVER
AIR QUALITY CONTROL REGION) IV-24
7. PHILADELPHIA IV-24
C. COMPARISON BETWEEN OXIDANT TRENDS AND
EMISSION TRENDS, 1970-1974 IV-28
1. SAN FRANCISCO BAY AREA IV-28
2. DENVER IV-29
3. OTHER AREAS INCLUDED IN STUDY IV-30
VI
-------
TABLE OF CONTENTS (continued)
Chapter
D. REVIEW OF LONG TERM TRENDS,
SAN FRANCISCO BAY AREA IV-33
REFERENCES IV-41
V. INTERACTIONS WITH WEATHER AND OTHER FACTORS. V-l
A. SAN FRANCISCO BAY AREA, WEATHER
VARIABLES V-l
B. GEOGRAPHICAL EFFECTS, BAY AREA AND
CENTRAL VALLEY V-5
REFERENCES V-l 0
VI. SUMMARY AND CONCLUSIONS VI-1
LIST OF FIGURES
Figure Page
II-l. CALIFORNIA AIR BASINS II-3
II-2. BAY AREA AIR BASIN II-5
II-3. MAP OF THE SAN FRANCISCO BAY AREA II-6
II-4. (FIGURE DELETED) 11-12
11-5. LARGE STATIONARY SOURCES, BAY AREA AIR BASIN H-16
II-6. NORTH CENTRAL COAST AIR BASIN 11-24
II-7. COLORADO AIR QUALITY CONTROL REGIONS 11-31
IV-1. SMALLEST VALUE OF CORRELATION COEFFICIENT
SIGNIFICANTLY DIFFERENT FROM ZERO IV-36
V-l. NORMAL FLOW PATTERN IN THE SUMMERTIME
MARINE LAYER V-6
Vll
-------
TABLE OF CONTENTS (continued)
LIST OF TABLES
Table Page
II-l. LOCATIONS OF AIR QUALITY MONITORING STATIONS
PROVIDING DATA USED IN STUDY 11-9
II-2. POPULATION AND BUILDING ACTIVITY BY YEAR,
1970-1974 11-14
III-l. OXIDANT LEVEL, 2ND HIGH BY YEAR,
AND TREND DIRECTION ITI-3
II1-2. TOTAL OXIDANT AVERAGE
MAXIMA AND TREND DIRECTIONS.. III-4
III-3. OXIDANT 3RD QUARTER AVERAGE
MAXIMA AND TREND DIRECTIONS III-6
III-4. FREQUENCIES OF EXCEEDING 0.08 ppm,
AND TREND DIRECTIONS.. III-8
III.-5. FREQUENCIES OF EXCEEDING 0.16 ppm
HOURLY AVERAGE OXIDANT I-11-9-
III-6. TREND DIRECTIONS FOR VARIOUS OXIDANT INDICES,
AS SHOWN AT VARIOUS MONITORING STATIONS 111-10
III-7. RANKINGS OF 12 CALIFORNIA MONITORING STATIONS
ACCORDING TO VARIOUS OXIDANT AIR QUALITY INDICES.. 111-13
III-8. ANALYSIS OF VARIANCE OF SECOND HIGH OXIDANT
LEVELS FOR 9 BAY AREA (MONITORING STATIONS,
FIVE YEARS (1970-1974) 111-15
IV-1. POLLUTANT EMISSIONS BY YEAR, BAY AREA AIR
POLLUTION CONTROL DISTRICT IV-4
IV-2. BAY AREA AIR POLLUTION CONTROL DISTRICT
HYDROCARBON EMISSIONS BY COUNTY, 1974 IV-5
IV-3. BAY AREA AIR POLLUTION CONTROL DISTRICT
OXIDES OF NITROGEN EMISSIONS BY COUNTY, 1974 IV-7
IV-4. MOBILE SOURCE EMISSION INDICES IV-9
IV-5. POLLUTANT EMISSIONS BY YEAR, SHASTA COUNTY
PORTION OF SACRAMENTO VALLEY AIR BASIN IV-11
IV-6. MOBILE SOURCE EMISSION INDICES, REDDING IV-13
IV-7. POLLUTANT EMISSIONS BY YEAR, SAN JOAQUIN COUNTY... IV-14
IV-8. POLLUTANT EMISSIONS BY YEAR, STANISLAUS COUNTY.... IV-15
vi 11
-------
TABLE OF CONTENTS (continued)
LIST OF TABLES (continued)
Table Page
IV-9. MOBILE SOURCE EMISSION INDICES, TWO CITIES
IN SAN JOAQUIN VALLEY AIR BASIN IV-17
IV-10. POLLUTANT EMISSIONS BY YEAR, MONTEREY COUNTY IV-19
IV-11. MOBILE SOURCE EMISSION INDICES, SALINAS IV-21
IV-12. POLLUTANT EMISSIONS BY YEAR, LOS ANGELES COUNTY
PORTION OF SOUTHEAST DESERT AIR BASIN IV-22
IV-13. MOBILE SOURCE EMISSIONS INDICES, LANCASTER IV-23
IV-14. HYDROCARBON EMISSIONS BY YEAR,
PHILADELPHIA, PA IV-25
IV-15. NITROGEN OXIDE EMISSIONS BY YEAR,
PHILADELPHIA, PA IV-26
IV-16. TREND DIRECTIONS FOR VARIOUS EMISSIONS INDICES
AND VARIOUS AIR QUALITY INDICES, FOR FIVE
CALIFORNIA LOCATIONS AND PHILADELPHIA IV-31
IV-17. AVERAGE HIGH-HOUR OXIDANT CONCENTRATIONS FOR
DAYS WITH COMPARABLE TEMPERATURE AND INVERSION
CONDITIONS (APRIL THROUGH OCTOBER SMOG SEASONS,
1962-1974) IV-35
IV-18. TREND DIRECTIONS FOR VARIOUS OXIDANT INDICES
SHOWN AT VARIOUS MONITORING STATIONS, SAN
FRANCISCO BAY AREA IV-38
IV-19. ESTIMATED EMISSIONS BY YEAR, SAN FRANCISCO
BAY AREA AIR POLLUTION CONTROL DISTRICT,
1962-1974, AND COMBINED EMISSION INDICES, FOR
HYDROCARBONS AND OXIDES OF NITROGEN IV-39
V-l. RANK CORRELATION COEFFICIENTS BETWEEN VARIOUS
ANNUAL OXIDANT INDICES AND AVERAGE NUMBER OF
SMOG-CONDUCIVE DAYS, FOR YEARS 1970 TO 1974 V-2
V-2. RANK CORRELATION COEFFICIENTS BETWEEN VARIOUS
ANNUAL OXIDANT INDICES AND ANNUAL AVERAGE
MAXIMUM TEMPERATURE IN DOWNTOWN SAN FRANCISCO,
FOR YEARS 1970 TO 1974 V-4
ix
-------
I. INTRODUCTION
As part of its mission to protect and enhance the Nation's
air resources, the U.S. Environmental -Protection Agency has es-
tablished a program of regular review and reporting, through the
Monitoring and Data Analysis Division, Office of Air Quality
Planning and Standards, on trends in air quality and emissions
on both a national and a regional basis.
The first such report, published in 1973, recognized in-
adequacies in the data available for interpretation at that time
and emphasized that a trends review could only portray a parti-
cular cross section of an evolving process rather than a final
resuTt.
In particular, in regard to trends in oxidant air quality,
the 1973 Trends report concluded, "The limitations in these data
make it impossible to reach a meaningful conclusion."
Since 1973 much work has been done and much progress made
in evaluating trends in oxidant air quality. To further this
task of evaluation and interpretation, the Monitoring and Data
Analysis Division contracted with Pacific Environmental Services
to identify, review and interpret oxidant air quality data and
oxidant precursor emissions data for a variety of locations
satisfying certain minimum criteria for relevance to the problem
of quantifying the relations between air quality and emissions
trends.
Under the contract (Task Order No. 1 of Basic Ordering
Agreement No. 68-02-1890) PES was directed to search for loca-
tions which could furnish five or more years of ambient air data
from oxidant or ozone monitoring, to develop trend lines for
various indices of oxidant air quality, and to investigate the
1-1
-------
relationship of these trend lines to the history of precursor
emission control programs. The observed trends were to be in-
terpreted with respect to such factors as:
• meteorological differences
• growth patterns and geographical differences
• NO emission trends or ambient data
/\
• hydrocarbon reductions
• monitoring site locations
• information from photochemical smog studies
• theoretical chemical dynamics of smog
• other pertinent information
This document embodies the final report of the project
carried out under the subject contract. Chapter II describes the
locations which were selected for more detailed consideration on
the basis of preliminary inquiries as to the availability of
relevant data. Chapter III reviews the oxidant air quality data
in detail and discusses the methodology and results of trend
evaluation. Chapter IV describes the available data concerning
emissions of hydrocarbons and oxides of nitrogen, and explores
the relations between the oxidant air quality trends and the
emissions data. Chapter V takes up the possible relation of
oxidant air quality trends to other factors, especially meteor-
ological and geographical differences. Finally, Chapter VI pre-
sents the summary and conclusions.
1-2
-------
REFERENCES
1. The National Air Monitoring Program: Air Quality and Emis-
sions Trends. Annual Report, August 1973. U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina 27711.
1-3
-------
II. LOCATIONS SELECTED FOR STUDY
A. INTRODUCTION
Minimum requirements for data availability, as specified in
the Statement of Work for Task Order #1, were five or more years
of ambient air oxidant or ozone monitoring records, with con-
current information on the history of hydrocarbon emission con-
trol programs. Up to eight locations were to be reviewed in detail,
but the South Coast Air Basin of California was excluded by the
project officer because a report of the desired scope had recent-
ly been published. Locations outside of California were to be re-
viewed and included, if sufficient data were available.
After review of information supplied by NEDS, SAROAD, and the
California Air Resources Board, the following locations were se-
lected for more detailed study:
• The San Francisco Bay Area, California
• Redding, in Shasta County, within the Sacramento
Valley Air Basin, California
• The northern San Joaquin Valley, California, rep-
resented by Stockton, in San Joaquin County, and
Modesto, in Stanislaus County
• The Salinas Valley, in Monterey County, within the
North Central Coast Air Basin, California
• Lancaster, in Los Angeles County, within the Ante-
lope Valley, a part of the Southeast Desert Air
Basin, California
• Denver City and County, part of the Metropolitan
Denver Air Quality Control Region, Colorado
• Philadelphia City and County, part of the Metro-
politan Philadelphia Air Quality Control Region,
Pennsylvania
• Whiteface Mountain, near Wilmington, Essex County,
New York
II-l
-------
Of the locations selected, five were in California, three
in other states, namely, Colorado, Pennsylvania and New York.
Three (San Francisco, Denver, Philadelphia) were in metropolitan
areas, four (Redding, San Joaquin Valley, Salinas and Lancaster)
were in rural agricultural areas, one (Whiteface) was in an iso-
lated location away from population centers.
Because of the relative abundance of ozone and oxidant data
in California, more than half the locations selected for study were
in that state. Figure II-l shows the outlines of California Air
Basins superposed on a physiographic map of the state. The loca-
tions selected represent a wide variety of types, from a large
metropolitan area with a population of almost 5 million persons,
to a somewhat agricultural desert community with a low-density pop-
ulation of about 50,000. Two of the areas are coastal, three in-
land.
Of the three inland locations, two are in the rich agricul-
tural Central Valley, which comprises the Sacramento Valley in
the north and the San Joaquin Valley in the South. Redding is at
the northernmost extreme of the Central Valley, while Stockton
and Modesto are near the geographical center of the Central Valley
and within the relatively near vicinity of the San Francisco metro-
politan area.
Expected information from the Whiteface Mountain location was,
unfortunately, never received, so detailed analysis was confined to
the remaining seven locations. The principal relevant character-
istics of these seven regions and their air monitoring stations are
reviewed in the following sections of this chapter.
II-2
-------
CALIFORNI A
AIR BASINS
SOUTH
CENTRAL
COAST
FIGURE II-l, CALIFORNIA AI« BASINS
II-3
-------
B. THE SAN FRANCISCO BAY AREA
1. Physical Description
The Bay Area Air Pollution Control District (BAAPCD) covers
covers all of seven counties and the southern portions of two
other counties, for a total area of about 6,000 square miles. Of
twelve monitoring stations operated by BAAPCD in 1970, nine were
chosen to represent the urban region (see map, Figure II-2).
This urban region includes the following significant topo-
graphic features possibly relevant to the generation and trans-
port of ozone. To the west of the San Francisco peninsula lies
the Pacific Ocean, to the east various bays and rivers. The
land area includes the Santa Cruz and the Diablo mountain ranges,
the former a part of the San Francisco peninsula and the latter
located to the southeast across the Bay. Other land areas in-
clude Various lowlands, •foothills, canyons and valleys. Largest
of the inland valleys is the Santa Clara Valley, south and east
of the Bay. The Livermore Valley is a smaller, isolated valley
due east of the Bay. In the northeast, the delta of the Sacra-
mento and San Joaquin Rivers forms a passage to the great Central
Valley, which is outside of the Bay Area District. The general
complexity of the terrain is indicated by the contours shown in
Figure II-3.
Meteorologically, the San Francisco Bay Area and associated
valleys constitute a well-defined zone somewhat broken into sub-
parts in terms of wind climatology. Rather low hills, the in-
fluence of the large water areas of the bays, and a large influx
of maritime air produce several well-defined wind patterns in
the area.
Except during late September and October, and during hot
spells in April, May or June, wind movement provides consistent
ventilation in much of the Bay Area.
II-4
-------
CONTRA COSTA CO.
SF
RC
SJ
FR
SL
OK
RI
PT
LV
STATIONS
San Francisco
Redwood City
San Jose
Fremont
San Leandro
Oakland
Richmond
Pittsburd
Livermore
FIGURE II-2. BAY AREA AIR BASIN
II-5
-------
fe-^'
- . ,
!^ ' ',' ; 'i " ' >-,... » «»Ii
-|^, —. ri-( eowco^o o T~V;' r.':~vr~-
r '
i^.'v?-'~ ',}''• r« "•!•. .---'• \\K5yC2n', j£v •^••'•r-
.Vfet> ^ -4 A^^^%'^
i •• %\W-^Kct-"-- ^^^r^A.
'••^W?^%.--f^^t^^
"^ ,x»«»j»«f SU'^"* ^ ! V^\ <^'^*;t0'RP'' '•"'IliiniH1'^' \ ^-«i
U^i""''^ ! ,,0-sM ^^Ki'i-^, ^^C^i-£lr4^-:.;
•^/V-V^; fi-'vC if^'"?ivL^0' ^. ^-sr^vji^ '•*
"&^J Wxv. '\^^v^ ^^TSjx:
"/^'vS ,• csC^=> 'S.V?X^i'sr;^s; x,> <^
A '^^.V^^'-rit^'-v
\'---, ^/V'T^^/i'} >r^<^n(i
-------
During much of the year the winds from the sea divide to flow
northward into the Sonoma and Napa valleys, eastward through the
Carquinez Strait, and southward into the Santa Clara valley.
This divisiion of airflows makes the opposite ends of the Bay Area
meteorological subparts of the region. The large flow of marine air
through the Carquinez Strait has a marked influence upon the climate
in portions of Solano and San Joaquin Counties.
As in other coastal areas, the subsidence inversion is domi-
mant over this area most of the year. It varies seasonally and daily
between 1,000 and 3,000 feet elevation. Due to solar heating, the
inversion may be destroyed over the extreme ends of the Sonoma and
Santa Clara valleys. Wide variations in vertical mixing occur over
the extreme ends of these valleys.
3
Lovill and Miller (1968) found that the maximum ozone in
vertical profiles occurred above the inversion base and that it
coincided with a wind jet in the inversion layer. Subsequently,
Miller et al (1970) , concluded that the ozone within inversion
layer comes from urban pollution by 2 routes: (1) by upward mixing
from the crests of waves in the inversion base; (2) by vertical
mixing in zones where the temperature inversion at its inland edge
is destroyed by heating of the mixing layer, leaving ozone aloft when
a new inversion is subsequently formed. At this inland edge the ver-
tical temperature profile becomes unstable, and ozone concentrations
have been observed to increase at altitudes as high as 2,000 meters.
From very recent work (San Jose State University, February,
c
1976) Miller describes the inversion as having a mean base of 300 m
ASL at Mt. Sutro (in San Francisco), with daily oscillations as much
as + 60m. The minimum height is reached in the late afternoon and the
mixing occurs in predawn hours. The amplitude of oscillation at
Oakland, on the east side of the bay, appears to be double that found
at the Mt. Sutro tower. Ozone concentrations were invariably 5-10
times higher within the inversion layer than in the marine air below
the inversion. Concentrations as high as 30 pphm have been observed
II-7
-------
within the inversion layer when levels in the marine air were only
1-3 pphm. The inversion may extend Westward over the Pacific ocean
as far as 100 Km.
Within the inversion layer at Mt. Sutro, Miller et al observed
strong turbulence with rapid temperature and wind changes. The slope
of the inversion is generally downward from west to east,
2. Monitoring Stations
For a fair representation of the air quality of the San Fran-
cisco Bay Area no single monitoring station would suffice, in view of
the complexities of terrain and wind regimes discussed in the previous
section. As a basis for statistical comparisons, PES selected for study
the nine locations shown in Figure II-2. Coordinates and characteris-
tics of the monitoring stations representing these communities are
listed in Table II-l.
These locations are mainly in the more densely populated part
of the Bay Area, along the southern and eastern shores of the Bay.
Highest population densities are represented by San Francisco, Oakland,
and San Jose. Relatively more suburban areas are Redwood City and
Fremont; relatively more industrial areas are Richmond and San Leandro.
Livermore represents a small, inland city in an isolated valley, while
Pittsburg, another inland city, is situated near the rather flat Delta
country, along the main channel between Suisun Bay and the Sacramento
River.
Not all of the nine counties included in the Bay Area Air Pollu-
tion Control District are represented by this selection. In fact, four
of the selected locations -- Oakland, San Leandro, Fremont, and Liver-
more — are in one county, Alameda. Other counties represented are
Contra Costa with two stations, Richmond and Pittsburg; San Francisco;
San Mateo, with Redwood City: and Santa Clara, with San Jose. The
unrepresented counties — Marin, Sonoma, Napa and Solano -- are less
II-8
-------
Table II-l. Locations of Air Quality Monitoring Stations Providing
Data Used in Study
a. San Francisco (SF) -
Address: 939 Ellis St. - sample period 1/70-12/74
UTM Northing: 4181810
Easting: 550980
Central City - Commercial
Above Ground 104' above MSL 229'
Near Civic Center, close to a heavily traveled
Boulevard.
Station Type:
Elevation:
Comment:
b. Redwood City (RC)
Address: 897 Barroo.-Ave
UTM Northing: 4148560
Easting: 570549
Station Type
Elevation:
Comment: In
- sample period 1/70-12/74
Suburban - Commercial
Above Ground 14' above MSL 29'
a residential and light industrial section
south of the city. (Entire city is near a busy
freeway.)
San Jose (SJ)
Address: 301 So. 5th Str. - sample period 1/70-5/70
UTM Northing:
Easting:
Station Type:
Elevation:
4132150
599120
Center City - Commercial
Above Ground 20' above MSL 109'
Address: 104 W. Alma - sample period 6/70=7/72
UTM Northing: 413027
599770
Center City - Commercial
Above Ground 28' above MSL 136'
About two km from previous location.
Easting:
Station Type:
Elevation:
Comment:
Address: 1203 N. 4th St. - sample period 7/72-12/74
UTM Northing: 4132920
Easting: 598580
Station Type: Center City - Commercial
Elevation: Above Ground 28' above MSL 108'
Comment: About 4km from previous location, about 1 km
from original location.
II-9
-------
Table II-l (cont) Locations of Air Quality Monitoring Stations
Providing Data Used in Study
d. Fremont (FR)
Address: 3965 Union St. - sample period 1/70-12/70
UTM Northing: 4154310
Easting: 592060
Station Type: Center City - Commercial
Elevation: Above ground 20' above MSL 71'
Address: 40733 Chapel Way - sample period 1/71-12/74
UTM Northing: 4154620
Easting: 591880
Station Type: Suburban - Residential
Elevation: Above Ground 18' above MSL 78'
Comment: Farther from freeway than previous location.
e. San Leandro (SL)
Address: 850 Thornton - sample period: 1/70-12/74
UTM Northing: 4174690
Easting: 573960
Station Type: . Center City - Industrial
Elevation: Above GroundJ30' above MSL 80'
Comment: Near schools, parks, freeway
f. Oakland (OK)
Address: 1111 Jackson St., - sample period: 1/70-12/74
UTM Northing: 4183650
Easting: 564580
Station Type: Center City - Commercial
Elevation: Above Ground 118' above MSL 154'
Comment: Near Lake Merritt and city center
g. Richmond (RI)
Address: 1430 Nevin Ave. - sample period: 1/70-3/72
UTM Northing: 4198850
Easting: 556730
Station Type: Center City - Commercial
Elevation: Above Ground msg. above MSL msg.
Address: 1144 - 13th St., - sample period 3/72-12/74
UTM Northing: Latitude N 37° 57' 00"
Easting: Longitude W 122° 2V 21"
Station Type: Suburban - Industrial, Residential & Commercial
Elevation: Above ground 22.5' above MSL 76.5'
11-10
-------
Table II-l (cont) Locations of Air Quality Monitoring Stations
Providing Data Used in Study
h. Pittsburg (PT)
Address: 583 W. Tenth St. - sample period: 1/70-12/74
UTM Northing: 4209500
597100
Center City - Commercial
Above Ground 25' above MSL 30'
Easting:
Station Type:
Elevation:
Comment: Not far from main river channel
Livermore (LV)
Address: 2500
UTM Northing:
Easting:
Station Type:
Elevation:
Comment: Well
Railroad Ave. - sample period: 1/70-12/74
4171230
608850
Center City - Commercial
Above Ground 20' above MSL 504'
removed from freeway
11-11
-------
FIGURE DELETED
11-12
-------
populous and have not been provided with monitoring stations with the
exception of one in San Rafael, Marin County.
In some of these communities, the monitoring facilities were
relocated during the period of interest, 1970 to 1974. In San Jose
there were two relocations, one after the first five months of 1970,
and the second in July 1972, half way through the five-year period.
In Fremont and in Richmond, one relocation occurred. To the extent
that these moves may have changed the relation between the monitoring
station and nearby contaminant sources such as freeways or busy
intersections, they could introduce some bias into the determination
of air quality trends. However, the number of such changes, for a
nine station network, has been minimal.
3. Population and Growth Factors
Population in the Bay Area is centered around the San Fran-
cisco Bay with the greatest densities near the cities of San Fran-
cisco, Oakland and San Jose. The northern counties, Marin, Sonoma,
Napa and Solano are relatively sparsely populated. Between 1970 and
1974 little or no population growth occurred in San Francisco city/
county or in Alameda and San Mateo counties; in fact a 5% decrease
has taken place in San Francisco. Contra Costa county increased 4%
(20,900) with the increase spread over a dozen cities rather than
concentrated in one location. Marin county population grew 3%, but
this is equivalent to only 4500 persons. The greatest growth occurred
in Santa Clara county (9% or 97,000) almost entirely in San Jose and
nearby Sunnyvale.
Population figures are given in Table II-2. In addition, the
numbers of building permits issued in each city are listed. In general,
these figures parallel the population growth. In summary, there has
been little growth in the Bay Area between 1970 and 1974 except at the
southern tip of San Francisco Bay in the general araa of San Jose, and
even here the growth has amounted to less than 10% over this 5-year
period.
11-13
-------
County
Table 11-2
Population* and Building Activity by Year, 1970 - 1974
Measure of Growth
1970
1971
1972
1973
1974
5an Francisco
tarin
/ontra Costa
Mameda
Santa Clara
San Mateo
TOTAL APCD
City/county population
City building permits
County population
San Rafael population
San Rafael building permits
County population :
Richmond population
Richmond building permits
Pittsburgh population
Plttsburg building permits
County population
Oakland population
Oakland building permits
Fremont population
Fremont building permits
Livermore population
Llvermore building permits
County population
San Jose population
San Jose building permits
County population
Redwood City population
Redwood City building permits
POPULATION WITHIN AREA
712,100
1771
207,000
38,949
124
557,400 ;
79,088 :
218
20,651
261
1,072,700
361,613 .
1969 ;
100,875
2191
37 ,703
857
1,072,400
446,504
9516
557,200
55,539
128
4,526,000
709,000
3476
209,200
716
562,900
484
639:
1,088,100
1480 :
3470 :
1756
1,093,600
10,645
559,900
803
4,624,000
695,800
3188
211,500
1051
567,600
271
1149
1 ,094 ,400
1829
2730
1401
1,122,000
8014
560,900
249
4,667,000
692,800
4148
214,100
523
573,600
328
119
1,089,100
1170
• 979
508
1,146,900
7428
565,500
119
4,706,000
679,200
1370
211,500
44,850
220
578,300
74,700
130
24,500
274
1,087,300
339,600
493
115,600
592
47,400
101
1,169,400
528,900
5097
568,900
54,000
155
4,729,000
* Data obtained from U. S. Bureau of the Census and from the California State Population Research Unit.
11-14
-------
4. Industry and Other Emitting Activities
Most major industrial activity in the Bay Area is located on
the eastern shore of San Francisco Bay in Alameda County, or along
the shoreline of Contra Costa County, including the shores of Carquinez
Strait and Suisun Bay, east of Richmond.
Oil refineries, chemical plants, and power plants are the major
point sources of hydrocarbons and NO in Contra Costa county. Two
/\
auto assembly plants near the south end of the bay are large hydro-
carbon emitters, and power plants on the bay near San Francisco emit
large quantities of nitrogen oxides. These are shown on the map in
Figure II-5. Historical records are not maintained on emissions from
stationary sources, but no major changes have occurred either as a
result of changes in emissions regulations or as a result plant shut-
downs or openings. Regulations promulgated by BAAPCD for control of
emissions from stationary sources remained essentially uniform through-
out the period, 1970-1974. Open burning has been effectively curtailed
since a control program was implemented in the period 1960 to 1965.
There have, however, been numerous minor changes in the APCD regulations.
Regulation 2, which deals with incineration, heat transfer and general
fuel combustion sources, was first adopted in 1960; its current version
is a twelfth revision, adopted in February 1975 and amended in March
1975. Regulation 3, limiting emissions of organic gases was adopted
in 1967 and revised late in 1974. Regulation 4, requiring crankcase
emission control on certain motor vehicles, went into effect in mid-
1971. A regulation requiring recovery of vapors from gasoline marketing
operations was adopted in 1973, but did not become effective until 1976.
Mobile source emissions trends are discussed below (Chapter IV).
In general, hydrocarbon emissions from such sources decreased less than
10% in most areas with increases and decreases both being recorded.:
In San Jose, where growth was rapid, hydrocarbons increased about 20%
between 1970 and 1974. Nitrogen oxides usually stayed constant or
increased by no more than 20% except in San Jose where the increase
11-15
-------
\ / \
C3I7TRA COS-A CO.
Large stationary
sources in the Bay
Area Air Basin
o Hydrocarbons
x Oxides of Nitrogen
S^ -i
Figure I1-5. LARGE STATIONARY SOURCES, BAY AREA AIR BASIN
11-16
-------
amounted to 35% in 1973 and dropped to a 26% increase in 1974.
The opening of the Bay Area Rapid Transit system during the period
had little or no effect on traffic volumes.
C. REDDING (SACRAMENTO VALLEY AIR BASIN)
1. Physical Description
The Sacramento Valley Air Basin, covering an area of nearly
36,000 sq. km. (14,000 sq. miles), includes eight entire counties
and portions of two more, viz., Shasta and Solano. The City of
Redding is the largest city in Shasta County and lies at the extreme
northern end of the Sacramento Valley at an elevation of 190m
(580 feet). The air basin'takes in the northern third of the great
Central Valley as well as parts of the surrounding mountain ranges;
its location relative to other California Air Basins is shown in
in Figure II-T.
The valley floor slopes up gradually from Sacramento north-
ward to Redding, a distance of about 300 km (190 miles), averaging
about 80 km in width. Wind patterns reflect the channeling effect
of the mountain ranges on three sides of the air basin. In summer,
marine air from San Francisco Bay enters the Sacramento Valley through
the Carquinez Straits and the Cordelia Gap in the Coast Range, at the
southeastern corner of the air basin. Because of intense heating
of the valley floor and the usual presence of a well-developed high
pressure cell off the California coast, marine air pushes far up the
valley and often reaches the extreme northern end at Redding. In
October and November, however, reduced surface heating shuts off the
influx of marine air and the prevailing up-valley winds give way to
diurnal down-valley flow.
Inversions occur with great frequency in all seasons at Red
Bluff, about 50 km south of Redding, which is the nearest location
where vertical soundings have been made regularly. From soundings
taken at about 0500 PST daily, it appears that there is a surface
11-17
-------
inversion on about five days of every six, and less than 3 percent
of the soundings show no inversion at all. The atmosphere is
especially stable in early fall, predisposing the entire valley
to stagnation conditions. In summer, however,'the marine inversions
are generally accompanied by brisk afternoon winds which .provide good
ventilation.
2. Air Monitoring
The monitoring station established in Redding by the California
Air Resources Board became operational in June, 1970. SAROAD infor-
mation is as follows:
Address: 1348 Market Street
UTM Northing: 4492770
Easting: 551650
Station Type: Center City - Commercial
Elevation: Above Ground 30' above MSL 612'
The address is in-the main city center. Monitoring was dis-
continued in July, 1975.
3. Population and Growth Factors
The population within the area covered by the Sacramento
Valley Air Basin was, in 1972, 1.3 million, of whom about 6 percent
(80,000) resided in Shasta County, including less than 2 percent
(18,000) in Redding. During the period 1970 to 1974, County popu-
lation increased at about 2.5 percent per year. It is believed that
the population of Redding, however, grew less rapidly, perhaps at
about 1 percent per year. Building permits issued by the city
between 1971 and 1974 varied from a little more than 200 to a little
less than 300 annually.
4. Industry and Other Emitting Activities
All industry in Shasta County is basically oriented to the
production of lumber and other wood products. There was no indication
of any substantial changes in this production during the period under
study, although one plywood plant was subsequently destroyed by fire
(March, 1976).
11-18
-------
Automobile traffic in Redding increased more rapidly than
did population, presumably facilitated by the opening of new free-
way links in 1970 and in 1974.
11-19
-------
D. STOCKTON AND MODESTO (SAN JOAQUIN VALLEY AIR BASIN)
1. Physical Description
The San Joaquin Valley Air Basin covers the southern part
of California's great Central Valley, including seven entire
counties and the major part of an eighth. Its area is nearly 65,000
sq. km. (25,000 sq. miles), its length nearly 500 km., along a
northwest-southeast orientation. Stockton and Modesto are the largest
cities in the two northern-most counties, respectively San Joaquin
and Stanislaus. They lie on the San Joaquin River at distances of
approximately 30 and 80 km. from its confluence with the Sacramento
with the Sacramento River in the delta region. Its northern
boundary also marks a meteorological separation from the Sacramento
Valley. (See Figure II-l).
In summer marine air enters the valley, primarily through the gap
at San Francisco Bay. During the late fall and the winter, cold air
draining from mountain slopes flows northward in the San Joaquin
Valley and southward in the Sacramento Valley, creating a zone of con-
vergence that fluctuates to the north and south of the delta area.
The general circulation, characterized by summertime up-valley and
wintertime down-valley winds, permits the transport of pollution over
long distances along the axis of the valley. Stockton and Modesto,
however, are relatively near the source of the marine air, the low-
level gap comprising the delta lowlands and the Carquinez Strait.
Modesto, being the more southerly of the two cities, also presumably
receives more air from the southern end of the San Francisco Bay Area
via the Hayward gap and the Livermore Valley.
2. Air Monitoring
SAROAD information for the monitoring stations at Stockton and
Modesto is as follows:
11-20
-------
a. Stockton (ST)
Address: 1601 E. Hazleton-sample period: 1/70-12/74
Northing: 4201560
Easting: 676608
Station Type: Center city - Commercial
Elevation: Above ground 88' above MSL 175'
b. Modesto (MO)
Address: 1024 J St. - sample period: 6/70-12/74
UTM Northing: 4167650
Easting: 676608
Station Type: Center City - Commercial
Elevation: Above ground 88' above MSL 175'
The Stockton site is appreciably removed from the city center,
while the Modesto site is downtown, close to the city center. These
stations were placed in operation late in 1970, so that the data
available cover only four full years of the study period, 1970 to 1974.
3. Population and Growth Factors
Population of the area included in the San Joaquin Air Basin
was estimated to be about 1.6 million in 1970; this, however, included
four counties with a total population of about 54,000 which were sub-
sequently assigned to a different air basin. The rate of growth of
population in this twelve-county area following 1970 was about 2
percent per year.
The primary centers of population in the San Joaquin Valley are,
in order of decreasing size, Fresno (Fresno County), Stockton (San
Joaquin County), Bakersfield (Kern County) and Modesto (Stanislaus
County). (Fresno is about 200 km. southeast of Modesto, while Bakers-
field is nearly 400 km in the same direction). Population in San Joaquin
County, about 300,000 in 1974, increased by less than one percent
annually during the study period (1970-1974), with about half the
increase occurring in Stockton, where about forty percent of the
11-21
-------
population now lives. Population in Stanislaus County, about
200,000 in 1972, increased somewhat more rapidly -- about 1.6 percent
annually, with most of the increase occurring in Modesto. Collectively,
about half a million people, thirty percent of those in San Joaquin
Valley, live in these two counties.
4. Industry and Other Emitting Activities
Agriculture and associated businesses are the dominant industry
in the entire San Joaquin Valley Air Basin. Other major sources of
basic income in the air basin are mineral extraction, forestry, and
various manufacturing industries. In San Joaquin County there is
some chemical industry and one power plant; in Stanislaus County,
a large fiberboard plant.
Air pollution control regulations for this two-county area are
of relatively recent origin, In late 196;9, open burning and the use of
single-chamber incinerators were banned. Late in 1971, burning of
agricultural wastes was restricted to favorable days only. In 1972,
new rules were promulgated to govern the use of solvents and archi-
tectural coatings. In 1973 a system requiring permits for con-
struction and operation of newsources was undertaken; in 1974 a new
rule banned unacceptable orchard heaters. Enforcement of these
regulations received no attention until late in 1974, when personnel
were assigned to that duty. Emissions from the fiberboard plant
apparently began to decrease at about the same time.
Motor vehicle traffic appears to have remained steady in
Stanislaus County throughout the five-year period. In San Joaquin
County, on the other hand, a new freeway link was opened in 1972,
causing an increase in traffic and associated emissions.
11-22
-------
E. SALINAS CNORTH CENTRAL COAST AIR BASIN)
1. Physical Description
The North Central Coast Air Basin comprises ..three
California counties, Santa Cruz, Monterey and San Bern'to; its
topography is indicated in Figure II-6. The Salinas Valley
is a steep-sloped coastal valley which opens out on Monterey
Plain and extends southeastward from the bay with mountain ranges
of two to three thousand feet on either side. Salinas lies on the
Salinas River, about 20 km. southeast of its mouth at Monterey
Bay.
The climate at Salinas is typically coastal. During the
summer month:;, stratus and fog occur frequently, primarily at night
and early morning, breaking up under insolation in late morning
or early afternoon. In late summer, persistence of the fog and
stratus lessens. During the warm season there is a general
onshore flow of air; the velocity of the Seabreeze is minimal around
sunrise, and increases during the day, reaching a maximum in the
afternoon. The marine air is heated as it moves inland in the
Salinas Valley with a temperature increase of nearly 10 degrees C
in the average maximum surface temperature between Salinas and King
City. Mean morning mixing heights, i.e., the approximate height
of the inversion base, varied from 341 meters in the winter to 529
meters in the summer, according to aircraft soundings carried out
for the California Air Resources Board in the period, October 1971
to April 1974.
11-23
-------
GENERALIZED TOPOGRAPHY
NORTH CENTRAL COAST
AIR BASIN
I. SAN BENITO VALLEY
2. SANTA CLARA VALLEY
3: SALINAS VALLEY
4. MONTEREY BAY
5. CARMEL VALLEY
CRUZ • ^
SAHTA ClWfr-
\ /-*.
CATSOHVILLE
\~ wuisreff\
/- ^ GONZALES Vff
5 ^ ^ ., SOLEDAO
MONTEREY ^*
\
cirr
\
j
!
Figure II-6
10
10 JO 30
SCALE IN MILES
11-24
-------
2. Air Monitoring
SAROAD information regarding the Salinas air monitoring
site is as follows:
Address: 312 E. Alisal St. - sample period: 1/70-12/74
UTM Northing: 4059400
Easting: 621230
Station type: Suburban - Commercial
Elevation: Above ground 31' above MSL 76'
3. Population and Growth Factors
The population of Monterey County increased from 248,000 in
1970 to about 267,000 in 1975, an annual rate of about 1.4'percent.
The city of Salinas, with about one fourth of the county's population,
was apparently receiving more than four-fifths of the new growth
between 1974 and 1975.
4. Industry and Other Emitting Activities
Industry in the North Central Coast Air Basin is primarily
related to agriculture, though a power plant and a producing petroleum
field are in operation. In the immediate vicinity of Salinas, the
principal industry is food processing. To the northeast at Moss
Landing on Monterey Bay, are canneries and power generation; farther
up the Salinas Valley at San Ardo is the only major stationary source
of hydrocarbons in the air basin, the San Ardo oil field, where steam
injection enhances oil recovery.
Until 1971, there were no regulations applicable to control
of emissions from stationary sources in the air basin. In 1971,
rules were passed to prohibit burning waste in outdoor fires or in
single-chamber incinerators, to regulate agricultural burning and to
limit emissions of oxides of nitrogen from combustion sources. In
1972, limits were placed on discharge of organic vapors from the
11-25
-------
use of solvents or architectural coatings and acceptable types of
orchard heaters were required. In 1973 a permit system for con-
struction and operation of new facilities was instituted. In 1974,
a new air pollution control agency was created, the Monterey Bay
Unified Air Pollution Control District, having authority in all
three counties, and having (for the first time) a staff of enforce-
ment officers.
Motor vehicle traffic in the air basin was light, increasing
only slowly during the five-year study period.
11-26
-------
F. LANCASTER (SOUTHEAST DESERT AIR BASIN)
1. hPhy<;ical Description
The Southeast Desert Air Basin includes the hottest and
driest portions of California. It comprises Imperial County,
eastern portions of San Bernardino, Riverside, Kern and San Diego
Counties, and the northeastern corner of Los Angeles County, inclu-
ding the city of Palmdale and the unincorporated community of Lancaster.
Its entire area is-about 88,000 sq. km., of which about 4 percent
is in Los Angeles County. This portion is in the extreme western end
of the Mojave Desert which constitutes the northern part of the air
basin. It is sheltered from maritime influences and gross inter-
basin transport of polluted air by mountain barriers to the north and
south, which join to the west. Limited transport of air from the
South Coast Air Basin and from the San Joaquin Valley Air Basin may
follow passeis in these ranges -- respectively, Soledad Canyon and
Tehachapi Pa.ss.
During the summer, when the Pacific High (a semi-permanent
cell of high atmospheric pressure) is well developed to the west of
California, a thermal trough overlies the desert. The dry, clear
conditions result in intense solar radiation, which is highly con-
ducive to photochemical smog formation. Prevailing winds are pre-
dominantly from the south and west, so the Lancaster area is
relatively little affected by air from the other parts of the air
basin and has a greater frequency of calm conditions than some other
desert population centers, viz., Victorville and El Centre.
Subsidence inversions over the southeast desert area are usually
quite high, at least 2,000 meters above the desert floor, providing
little hindrance to vertical dispersion of pollutants. Radiation
inversions are prevalent at night throughout the year. In the summer
they are usually destroyed early in the day, but in winter they some-
times persist well into the day, limiting mixing to a layer which may
be as ;low as; 100 meters.
11-27
-------
2. Air Monitoring
SAROAD information for the Lancaster air monitoring station
is as follows:
Address: 45547 North Beech St. - sample period 11/70-12/74
UTM Northing: 3841540
Easting: 396540
Station type: Center City - Commercial
Elevation: Above ground 18' above MSL 2347'
The site is characterized as a "center city" location; it
lies within a rather diffuse community, as the street number, 45547,
may suggest.
3. Population and Growth Factors
An unincorporated area of about 110 sq. km., Lancaster increased
in population from about 41,000 in 1972 to 44,000 in 1974, an annual
rate of nearly 4 percent. The nearby city of Palmdale, with population
approaching 10,000, experienced a similar growth rate. A few other
communities in the Los Angeles County portion of the air basin are
much smaller than these.
4. Industry and Other Emitting Activities
Lancaster is a desert community with some agriculture and virtually
no other industry. Two large asphalt plants are located 22 km. to the
southeast, at Littlerock. A large Air Force base, Edwards, is located
nearby to the northeast, with headquarters about 30 km. from Lancaster.
In spite of the increase of population noted above, motor vehicle
traffic in the Lancaster area is reported to have decreased during the
study period (1970 to 1974).
Since Lancaster is in Los Angeles County, it has for many years
been subject to the relatively strict air pollution control regulations
11-28
-------
which have been enforced in the heavily populated and industrialized
sections of the county. No unusual variations in either stationary
or mobile sources are known to have occurred during the study
period. In 1971, there were changes in the regulations regarding solvents
being made subject to control. Other rule changes affected emissions
of sulfur oxides, particulates and carbon monoxide, but not hydrocarbons
or oxides of nitrogen. New regulations on gasoline storage were
announced in 1973, but not made effective until 1975.
11-29
-------
G. DENVER, COLORADO
1. Physical Description
The City and County of Denver is part of the Metropolitan
Denver Air Quality Control Region, which comprises Denver and
seven surrounding counties as shown in Figure II-7. This AQCR
roughly coincides with the drainage basin of the South Platte
River; it extends from the Continental Divide eastward to the
Colorado plains, covering about 20,000 sq. km., with Denver
approximately in its center.
Denver is the highest major metropolitan area in the United
States, with altitude about 1600 meters. It therefore receives,
in clear weather, more ultraviolet irradiation from the sun than
cities near sea level. This relatively intense radiation tends
to accelerate the photochemical reactions which generate oxidants
.1
in urban atmospheres; consequently, Denver may be more subject
to oxidant smog than other cities at a similar latitude.
During late fall, the Denver area is subject to frequent
atmospheric inversions which prevent dispersion of pollutants
and cause the formation of a "brown cloud," with restricted
visibility. This phenomenon has been the subject of at least
two special investigations, primarily because of its noticeabil-
ity. The topography and meteorology of the area tend to concen-
trate the urban plume over a well-defined area where it is in
sharp visual contrast to surrounding clean air. These condi-
tions are, apparently, not prevalent in the summer months when
photochemical oxidant is most often a problem, and when westerly
winds are prevalent.
2. Air Monitoring
SAROAD information regarding the CAMP air monitoring
station in Denver is as follows.
11-30
-------
PAWNEE I
"GRANDlMESA
COMANCHE
jp
SANjI SABEL
S A N SL U I S
FOUR CORNERS
GO
Figure II-7.
COLORADO AIR QUALITY CONTROL REGIONS
(REGION 2 SHADED)
-------
Address: 2105 Broadway--sample period 1/68-12/73
UTM Northing: 4399920
Easting: 501100
Station Type: Center City--Commercial
Elevation: Above ground 9' above MSL 5220'
This is the only Denver station for which five years oxi-
*
dant data were available for this study. Five additional sta-
tions have, more recently, been equipped for oxidant monitoring,
and data for 1974 indicate that the NAAQS level for ozone, 0.08
ppm, was exceeded more frequently at four of these stations than
at the CAMP station. For this reason, the CAMP station is now
considered to be "a very poor indicator of the ozone situation
in the Denver AQCR, underestimating the problem for the suburban
areas."
3. Population and Growth Factors
The Metropolitan Denver Air Pollution Control Region is
the most populous region of Colorado, with a population in 1970
of 1,050,000, nearly half the population of Colorado. In the
City and County of Denver, the population in the same year was
515,000, nearly half that of the region.
The population of Denver is quite stable, the estimated
increase between 1970 and 1974 being only slightly more than 1
percent, about 6,000 people. However, growth in the Metropoli-
tan Denver Region, which includes the suburban counties, has
been rapid, continuing the trend established with a 26 percent
increase of population in Colorado between 1960 and 1970. The
population in Denver Standard Metropolitan Statistical Area also
grew rapidly, from 929,000 in 1960 to 1,228,000 in 1970, an in-
crease of 32 percent.
Oxidant data for the summer months of 1969, 1971 and 1974
were incomplete or unavailable.
11-32
-------
4. Industry and Other Emitting Activities
The Denver Air Quality Control Region is heavily metropoli-
tan, but not strongly industrial. Stationary sources include
three major power plants, two oil refineries and various other
large or medium-sized industrial sources, mainly involved in
processing foods and extraction and processing of minerals.
Motor vehicle traffic appears to have remained fairly steady, as
traffic counts taken in 1971 and 1974 near the air quality moni-
toring stations showed little change.
Attempts to control or reduce hydrocarbon emissions seem
to have been focused mainly on mass transportation programs and
reduction of vehicle miles travelled, by such means as encourag-
ing formation of car pools for commuters.
11-33
-------
H. PHILADELPHIA. PENNSYLVANIA
1. Physical Description
Philadelphia occupies a section of the Atlantic Coastal
Plain lying about 100 km inland from the Atlantic Ocean, south
of the Delaware River Valley, at the confluence of the Schuylkill
River with the Delaware. The western portion of the city rises
into the Piedmont plateau and lies about 80 km southeast of the
Appalachian Mountains. Elevations within the city range up to
about 100 meters. The climate is typical of that of the
Atlantic Coastal Plain.
2. Air Monitoring
Information obtained from SAROAD regarding the Philadelphia
monitoring station is as follows:
Address: 2031 Race Street—sample period 1/69-12/74
UTM Northing: 4422830
Easting: 485150
Station Type: Center City—Mobile (near major traffic
arteries)
Elevation: Above ground 11" above MSL 36'
The Race Street station is the only Philadelphia station
for which oxidant data have been accumulated for a five-year
period, but samples collected at the Air Management Services
Laboratory from 1969 through 1972 indicated a much larger in-
cidence of samples exceeding the NAAQS level for oxidant than
did the values obtained at the Race Street station. Detailed
data for 1974 were not available.
3. Population and Growth Factors
Philadelphia is a city of nearly two million people and
the center of a metropolitan area of more than five million,
the Delaware Valley Region. In the decade 1960-1970, the
11-34
-------
Philadelphia Standard Metropolitan Statistical Area, consisting
of five counties in Pennsylvania and three in New Jersey, in-
creased in population by 11 percent, or about 1 percent annually.
The rate of increase for the Delaware Valley Region, covering
nine countieis, for the period 1969 to 1974, seems to be only
about half a,s large. The City of Philadelphia, on the other
hand, has beien losing population since 1960. There was a net
decrease of -2.7 percent (about 50,000 people) in the decade
1960-1970, which appears to be accelerating in the subsequent
years—from 1,950,000 in 1969 to 1,875,000 in 1974.
11-35
-------
REFERENCES
1. Trijonis, Peng, McRae and Lees, "Emissions and Air Quality
Trends in the South Coast Air Basin," EQL Memorandum No.
16, Environmental Quality Laboratory, California Institute
of Technology, Pasadena, California, January 1976.
2. Miller, A., "Wind Profiles in West Coast Temperature Inver-
sions," Report No. 4, San Jose State College, Department of
Meteorology, March 1968.
3. Lovill, J. E., and Miller, A., "The Vertical Distribution of
Ozone over the San Francisco Bay Area," J_. Geophysical Re-
search. 73, 5073 (1968).
4. Miller, et. al., "Ozone Within and Below the West Coast
Temperature Inversion," Tellus. 22, 328 (1970).
5. Miller, A., "Wave Properties in the West Coast Inversion,"
San Jose State University, Department of Meteorology,
February 1976.
6. Ferman, Eisinger and Monson, "Characterization of Denver
Air Quality." Research Publication GMR-1847, General
Motor Corp., Warren, Michigan. March 24, 1975.
7. "Assessment of Air Quality, Metro-Denver Region." Air
Pollution Control Division, Colorado Department of Health,
Denver, Colorado. August 1975.
11-36
-------
III. OXIDANT TREND ANALYSIS
A. INTRODUCTION
A complication in the analysis of trends in oxidant air quality
arises from the fact that different indices of oxidant air quality may
exhibit different trends. In the present study, this has been seen
especially in the analysis of data from the Bay Area Air Pollution
Control District. Previously, Trijonis ert a_P ' reported a similar
finding, namely, that the frequency of concentrations exceeding high
criterion levels (20 to 30 pphm) in the South Coast Air Basin was
reduced significantly more in the decade, 1965 to 1974, than the
frequency exceeding the federal standard.
In this study, various air quality indices have been compiled
and examined for evidence of trends, separately and in combination.
The indices which have been systematically examined are:
•> 2nd high: the second highest daily high-hour
oxidant value observed during a full year of
observations at a single monitoring location.
• Average max.: the average of the highest
hourly average oxidant values observed on all
days during a full year of monitoring at a
single location.
•> 3rd quarter average max: the average of the
highest average oxidant values observed on all
days during the months of July, August and
September at a single location.
• Frequency >.08: the frequency, in percent of
all days monitored, of days with oxidant values
exceeding 0.8 ppm (8 pphm; the NAAQS level) during a
full year of monitoring at a single location.
• Frequency >.16: the frequency, in percent of all
days monitored, of days with oxidant values exceed-
ing 0.16 ppm (16 pphm) during a full year of moni-
toring at a single location.
Finding:; of the study in relation to these various indices are
discussed in detail in following sections. Relations between the
indices themselves, as shown by the data gathered, are discussed in
Section G> with summary and conclusions in Section :H.
III-l
-------
B. SECOND HIGHEST MAXIMUM
Table 111-1 shows the value of this index for each monitoring
station for each year of the pentad, 1970-74.
Individual values range from 0.06 ppm (San Francisco, 1974)
to 0.29 ppm (Livermore, 1970), both extremes occurring in the Bay
Area. Only nine of the values fail to exceed the NAAQS value, 0.08
ppm; these values occurred at two Bay Area stations, at Salinas, CA.,
and at Denver.
Trend directions, evaluated from these data, are indicated in
the final column of Table III-l. Trend magnitudes are not shown,
because (with only five years of data) they are not statistically
significant. Nevertheless, the fact that the trend was downward for
eight of the nine Bay Area stations is significant (p< .05) and indi-
cates a downward trend for that district as a whole. The magnitude
of this trend, separately evaluated, amounts to about 25 percent over
the four-year interval. For the San Joaquin Valley, California, trends
are in opposite directions at the two locations; the overall trend, not
significant, is a decrease of about 6 percent over the four-year
interval.
C. AVERAGE MAXIMUM
Table 111-2 shows the value of this index for each monitoring
station for each year of the pentad, 1970-1974.
Individual values range from 0.011 ppm (Philadelphia, 1971) to
0.082 ppm (Modesto, 1970, an incomplete record) or 0.072 ppm (Liver-
more, 1974). Averages for the pentad, also shown in Table III-2,
range from 0.015 (Philadelphia) to 0.060 (Modesto) or 0.057 (Livermore).
Trend directions, indicated in the last column of Table III-2,
are predominantly positive, although only the Denver trend was significant.
For the Bay Area as a whole, separately evaluated, the trend amounts to
I.e., the sign of the coefficient of correlation, index vs. year
(1970 to 1974). For explanation of trend evaluation, see p. 111-18.
III-2
-------
TABLE III - 1
OXIDANT LEVEL, 2ND HIGH BY YEAR, AND TREND DIRECTION
(Second highest daily maximum hourly average, ppm)
'70
'71
'72
'73
'74
Trend
Bay Area
SF
RC
SJ
FR
SL
OK
RI
PT
LV
Sacramento Valley
Redd i ng
San Joaquin Viilley
Stockton
Modesto
No. Central Coast
Salinas
S. E. Desert
Lancaster
'68
Denver 0.11
Philadelphia -
Whiteface Mtn,. Not
.14
.16
.16
.27
.21
.20
.18
.18
.29
(,13)a
.17
(.16)a
(.07)b
—
'69
0.08
0.11
Received
.15
.16
.14
.24
.24
.21
.14
.18
.21
.13
.14
.22
.09
.13
'70
0.08
0.12
.07
.17
.17
.29
.14
.09
.10
.18
.18
.14
.13
.16
.08
.14
'71
0.10
0.13
.08
.12
.18
.18
.20
.14
.10
.10
.20
.13
.18
.17
.13
.20
'72
0.09
0.12
.06
.16
.26 +
.21
.17
.09
.08
.10
.25
.14 (0)C
.15 +
•16 (-)C
.10 (+)C
.14 (+)C
1
'73 '74 Trend
0.08 - (-)c
0.19 - ( + )c
a. Only 7 months of monitoring.
b. Only 9 months of monitoring.
c. Parenthes;es indicate data not complete over the period 1970-1974.
III-3
-------
TOTAL OXIDANT
(Yearly average
'70
TABLE III - 2
AVERAGE MAXIMA AND TREND DIRECTIONS
of daily maximum hourly average, ppm)
'71 '72 '73 '74 Ave.
Trend
Bay
Area
SF
RC
SJ
FR
SL
OK
RI
PT
LV
.028
.037
.055
.052
.049
.036
.042
.052
.063
.017
.041
.038
.047
.042
.022
.032
.042
.050
.022
.043
.036
.056
.032
.028
.032
.046
.045
.018
.032
.047
.047
.048
.026
.032
.036
.055
.026
.048
.065
.062
.059
.034
.028
.044
.072
.022
.040 +
.048 +
.053 +
.046 +
.029 +
.033
.044
.057 +
Sacramento Valley
San
No.
S.
Redding
Joaquin
Stockton
Modesto
Central Coast
Salinas
E. Desert
Lancaster
(.057)a
.059
(.082)a
(.036)b
.025C
.050
.035
.061
.038
.032
.049
.036
.048
.038
.063
.052
.056
.050
.041
.072
.062
.056
.062
.044
.050
(.054) (+)d
.048 +
(.060) (-)d
(.039) (+)d
(.055) (+)d
Denver
Philadelphia
Whiteface Mtn.
'68
.032
—
Not
'69
.028
.018
Received
'70
.023
.014
'71
.022
.012
'72
.022
.011
'73
.013
.015
'74
.021
Ave.
.015
Trend
(-)"*
+
a. Only 7 months of monitoring.
b. Only 9 months of monitoring.
c. Only 2 months of monitoring
d. Parentheses indicate data not complete over the period 1970-1974.
e. Significant, p<0.'01; trend, - 0.0033 ppm/yr.
III-4
-------
an increase of about 4 percent in the four-year interval. For the
sixteen stations nationwide, eleven positive trends are indicated,
while for the nine stations of the Bay Area, six positive trends
are indicated; the proportion is not high enough to be statistically
significant in either case. Trends for the two monitoring locations
in the San Joaquin Valley are in opposite directions; the over-all
trend, not significant, is a decrease of about 12 percent over the
four-year interval.
Data for the period 1968 to 1973 for Denver showed a down-
ward trend, significant at the one percent level, of 3.3 ppb (0.0033
ppm) per year. This is more than ten percent of the 1968 level; over
the six year period, the total decline amounted to 59 percent. The
trend in the annual 2nd high (Table III-l) was in the same direction,
but smaller and not significant.
D. THIRD QUARTER AVERAGE MAXIMUM
Table .111-3 shows the value of this index for each monitoring
station for each year of the pentad, 1970-1974.
Individual values range from 0.010 ppm (Philadelphia, 1972)
to 0.11 (Modesto, 1971; Lancaster, 1973; San Jose and Livermore, 1974).
Averages for the five summers, also shown in Table III-3, range from
0.013 ppm (Philadelphia, based on incomplete information) or 0.027
ppm (San Francisco) to 0.096 (Modesto).
Trend directions, indicated in Table III-3, are about evenly
divided between positive and negative; none of these trends is statis-
tically significant. For the Bay Area as a whole, separately evaluated,
the composite trend amounts to a decrease of about ten percent durirtg
the four-year interval. For the two monitoring stations in the San
Joaquin Valley, trends are in opposite directions; the over-all trend,
not significant, is a decrease of about 7 percent over the four-year
interval.
III-5
-------
TABLE III - 3
OXIDANT 3RD QUARTER AVERAGE MAXIMA AND TREND DIRECTIONS
(Three-month3 average of daily maximum hourly average, ppm)
'70
'71
'72
'73
'74
Ave.
Trend
Bay Area
SF
RC
SO
FR
SL
OK
RI
PT
LV
.050
.047
.087
.080
.057
.047
.057
.077
.090
.023
.053
.053
.080
.057
.043
.037
.070
.080
.027
.057
.057
.080
.050
.030
.037
.073
.063
.013
.033
.063
.053
.050
.027
.033
.043
.083
.023
.053
.11
.080
.066
.043
.040
.050
.11
.027
.049
.074
.075
.056
.038
.041
.063
.085
—
—
+
—
+
—
—
—
+
Sacramento Valley
San
No.
S.
Redding
Joaquin Valley
Stockton
Modesto
Central Coast
Salinas
E. Desert
Lancaster
.083
.083
.10
.037
NA
.087
.032
.11
.037
NA
.077
.060
.087
.037
.080
.083
,083
.083
.043
.11
.097
.087
.10
.043
.087
.085
.069
.096
.039
(•09)b
+
+
—
+
(+)b
'68
'69
'70
'71
'72
'73
'74
Ave. Trend
Denver
Philadelphia
Whiteface Mtn.
0.037
Not
NA
.018
Received
0.027
.014
NA
.010
0.033
.011
0.
0.
017
018
(.026)
(.013)
b (-)b
b (+)b
a. Averages for July, August and September
b. Parentheses indicate data not complete for these three months in all five
years.
III-6
-------
E. FREQUENCY EXCEEDING NAAQS LEVEL
Table 111-4 shows the value of this index for each monitoring
station for each year of the pentad, 1970-74,
Individual values range from zero (which occurred four times
each at San Francisco and Salinas, twice each at Oakland, Richmond.
and Pittsburg, and once at Lancaster, with only two months of obser-
vations) to 22.7 percent (Lancaster, 1973). For the entire pentad,
values ranged from 0.3 percent (San Francisco) to about 16 percent
(Modesto, 5 months information missing) or 13 percent (Livermore).
Trend directions, as shown, were almost equally divided
between positive and negative. None of these trends were statistically
significant. For the Bay Area, seven of the nine stations showed
negative trends; the composite trend was a reduction of the frequency
value by about 25 percent during the four-year interval. For the two
stations in the San Joaquin Valley, the trends seemed to be in opposite
directions, with net effect near zero.
F. FREQUENCY EXCEEDING TWICE NAAQS LEVEL
Table III-5 shows the value of this index for each monitoring
station for each year of the pentad, 1970-74.
It is of interest to note that days in this category were so
rare that only a handful of non-zero entries appear. The chosen
criterion level, .16 ppm, was exceeded more than once at only four
stations (namely, Fremont, Livermore, San Jose, and San Leandro) in
the Bay Area, and in Denver.
G. RELATIONS BETWEEN THE INDICES
Tables III-l to III-5 provided indications that the various
indices of air quality may be differently affected by changes in air
quality. Table III-6 presents, for ready comparison, the trend
direction information from the earlier tables. It appears from this
III-7
-------
TABLE III - 4
FREQUENCIES OF EXCEEDING 0.08 ppm, AND TREND DIRECTIONS
(Days with hourly average oxidant levels exceeding
0.08 ppm, shown as percentage of days monitored)
'70 '71 '72 '73 '74 Ave.
Trend
Bay Area
SF 1.6 0 0
RC 3.4 3.7 3.4
SJ 11.0 8.7 3.8
FR 9.4 9.2 10.0
SL 7.3 4.7 3.4
OK 3.2 2.2 0
RI 5.6 1.8 0
PT 8.4 4.4 4.4
LV 16.9 9.2 5.1
Sacramento Valley
Redding (12.8)a 9.0 8.4
San Joaquin
Stockton 8.4 3.4 2.4
Modesto (27.3)a 18.9 11.9
No. Central Coast
Salinas (0)b 0 0
S. E. Desert
Lancaster (0)c 1.7 1.4
'68 '69 '70
1 Denver 14.9 4.7 8.1
Philadelphia6 0.5 0.5
, Whiteface Mtn. Not Received
a. Only seven months of monitoring.
b. Only nine months of monitoring.
c. Only two months of monitoring.
0 0
2.7 2.7
8.4 17.1
5.9 6.9
6.7 3.8
1.8 0
1.4 0
0 0
14.1 18.9
11.9 11.9
11.9 6.9
11.9 16.9
2.4 0
22.7 11.9
71 '72
5.6 18.2
0.2 0.3
d. Parentheses indicate data not complete over the period
e. Data for Philadelphia are given in terms
which the criterion level is exceeded.
on which such levels were reached may be
of percent of
The frequency
much higher.
0.3
3.2
7.8 +
8.3
5.2
1.4
1.8
3.4
12.8 +
HO) ( + )
6.6 +
H6) H
(0.5) (+)
(-9) (+)
1
'73 '74 Trend
5.3 - (+)
2.8 (+)
1970-1974.
all hours during
in terms of days
III-8
-------
TABLE III - 5
FREQUENCIES OF EXCEEDING 0.16 ppm HOURLY AVERAGE OXIDANT
(Days with hourly average oxidant levels exceeding
0.16 ppm, shown as percentage of days monitored)
'69
'70
'71
'72
'73
'74
Bay Area
SF
RC
SJ
FR
SL
OK
RI
PT
LV
Sacramento Valley
Redding
San Joaquin
Stockton
Modesto
North Central Coast
Salinas
S. E. Desert
Lancaster
Denver
Philadelphia
Whiteface Mountain
0
0
0
1.8
0.3
0
0
0
1.3
0
0
0
0
0
0 0
0.0 0.0
Not
0
0.3
0
1.5
0.3
0.6
0.3
0
0
0
0
0
0
0
0.5
0.0
Received
0
0
0
1.6
0
0
0
0
0
0
0
0
0
0
0.3
0.0
0
0
0.3
0
1.0
0
0
0
0
0
0
0
0
0
0.4
0.0
0
0
1.8
o
0
0
0
0
1.7
0
0
0
0
0
III-9
-------
TABLE III - 6
TREND DIRECTIONS FOR VARIOUS OXIDANT INDICES,
AS SHOWN AT VARIOUS MONITORING STATIONS
SF RC SJ FR- SL OK RI PT LV RD ST MO SA LN DV PH
Second-high __ + ___
Average Max. _ + + + + +
3rd Qtr. Ave. Max. -- + - + _
Frequency >0.08 __ + ___
111-10
-------
presentation that there must be a substantial degree of independence
between the second-high statistic and the yearly average maximum,
since there are 5 disagreements among sixteen comparisons for these
indices. Perhaps more important, the second-high statistic shows a
statistically significant trend for the Bay Area (when evaluated for
the niine stations selected for this study), while the remaining
statistics do not.
To explore these relations, several auxiliary analyses were
undertaken. First, correlation coefficients were computed, using
paired values of different indices for individual stations year by
year, and for all five years. Second, coefficients of concordance
were computed for 12 California stations as ranked according to
different indices in various combinations. Third, a two-way analysis
of variance was performed to determine whether the error variance
associated with the use of the second-high index was consistent
with statistically sound use of this index in the detection of trends
in the present study. Results of these analyses are discussed in the
three following subsections:
a. Correlation between Oxidant Indices
Using data from the nine Bay Area stations, the
2nd-high statistic and the yearly average maximum
were compared at 45 data points (five years for
each station), yielding a correlation coefficient
of 0.728. As anticipated, this correlation is
far from perfect, although it is, indeed, statis-
tically convincing. When the average maximum is
taken as an independent variable, the regression
coefficient is 3.3* and a standard error of esti-
mate is 0.031 ppm, corresponding to a rather
large coefficient of variation, about twenty per-
cent. Thus, it is reasonable to conclude that
the 2nd-high statistic may have, in relation to
oxidant data analysis, advantages that will bear
further investigation.
Correlation coefficients between these two indices
for the individual years showed very wide variation;
values for successive years starting in 1970
*This means that a unit increase in the average maximum leads to
an expected increase of 3.3 units in the 2nd-high value.
III-ll
-------
were .67, .42, .95, .88, and .96 respectively.
Combining the three later years yielded a co-
efficient of 0.866, indicating much less inde-
pendence between these indices. Since this
coefficient, also, is highly significant, the
finding suggests that physical factors which
cannot be detected by this analysis may have
caused the change.
Since the correlation between 2nd-high and
annual average maximum indices was especially
low for 1971, it was of interest to determine
whether the 3rd-quarter average maximum would
provide a better correlation. This might be
expected, in view of the fact that there is a
higher frequency of high-oxidant days in summer
than in other seasons, especially on the West
Coast. A test of this correlation (i.e., 2nd-
high and 3rd quarter ave. max.) for 1971 yielded
a coefficient of 0.61; higher than the figure
(0.42) for the yearly average maximum, but still
lower than the other coefficients cited in the
previous paragraph.
b. Concordance between Oxidant Indices
If different oxidant indices reflect different,
incompletely correlated, aspects of air quality,
it should, perhaps, be expected that a set of
monitoring stations would rank differently when
ranked in terms of different indices. This
hypothesis was tested by ranking twelve northern
California stations (nine Bay Area stations plus
Redding, Stockton and Modesto) by four indices
(2nd-high oxidant, yearly average maximum, third
quarter average maximum and frequency equal to or
exceeding 0.08 PRm), then computing coefficients
of concordance.^/(This coefficient measures the
extent of agreement among several sets of ranks,
ranging from zero for no agreement to unity for
full agreement.) When the set of four indices was
used to rank the twelve stations according to five-
year average values, the coefficient of concordance
was 0.83. When the 2nd-high statistic was excluded
and the remaining three indices used, the coefficient
was 0.99. (Both significant at 1% level.)
The rankings obtained in this procedure are shown
in Table III-7. Examination of this table
suggested that the main differences between the
111-12
-------
TABLE III - 7
RANKINGS OF 12 CALIFORNIA MONITORING STATIONS
ACCORDING TO VARIOUS OXIDANT AIR QUALITY INDICES
Index (Syr, ave.) FR LV SL SJ MO ST PT RC OK RD RI SF
Second-high
Yearly Ave. Max.
3rd Qtr. Ave. Max.
Frequency >.08
Frequency >.16
1
4
4
4
1
2
2
2h
2
2
3
7
8
7
4
4
6
5
5
3
5
1
1
1
6
5
6
6
7
8
7
8
8
9
9
9
9
11
11
11
10
3
2%
3
11
10
10
10
12
12
12
12
111-13
-------
ranking for the 2nd-high index and the other
three could be due to differences at four stations:
Fremont, San Leandro, Modesto and Redding. In
fact, when the procedure was repeated with these
four stations deleted, the coefficient of con-
cordance for all four indices was 0.98.
The results confirm that very good agreement between
these indices can be obtained under restricted
circumstances and with carefully selected monitoring
sites. More divergence may be expected, however,
when observations are collected from monitoring
stations representing different types of terrain
or a greater variety of urban and suburban or rural sites,
c. Analysis of Variance, Second-High Statistic
To test the significance of yearly changes and
differences between stations, in view of the
expected rather large variance of the extreme-value
statistic called 2nd-high, an analysis of variance
was performed, using data for 9 stations of the
Bay Area monitoring network. Results are presented
in Table III-8.
Year averages ranged from 0.144 to 0.199 ppm, and
the yearly change effect was significant at the
5 percent level. Station averages ranged from
0.100 to 0.238 ppm and the differences between
stations were significant at the 1 percent level.
Residual variance was 0.00144 ppm 2} corresponding
to an error standard deviation of 0.038 ppm --
large, but clearly not disabling for data bases
as large as the one treated here.
H. REVIEW AND CONCLUSIONS
Oxidant air quality data from five air quality control regions
in California and from Denver, Colorado, and Philadelphia, Pennsylvania,
have been tabulated and their trends evaluated over the five-year period
1970 to 1974. Various air quality indices have been applied in this
analysis, namely:
111-14
-------
TABLE III - 8
ANALYSIS OF VARIANCE OF SECOND-HIGH OXIDANT LEVELS
FOR 9 BAY AREA MONITORING STATIONS, FIVE YEARS (1970-1974)
Station
SF
RC
SJ
FR
SL
OK
RI
PT
LV
Total
Average
Year
70 71 72 73 74
14 15 7 8 6
16 16 17 12 16
16 14 17 18 26
27 24 29 18 21
21 24 14 20 17
20 21 9 14 9
18 14 10 10 8
18 18 18 10 10
29 21 18 20 25
179 167 139 130 138
19.9 18.6 15.4 14.4 15.3
Total
50
77
91
119
96
73
60
74
113
753
Average
10.0
15.4
18.2
23.8
19.2
14.6
12.0
14.8
22.6
1
I 16.7
Source of
Variance
Years
Stations
Residual
Total
Sum of Squared
Deviations
199
852
460
1511
Degrees of
Freedom
4
8
32
44
Variance"
Estimate
50
107
14.4
34
F
Ratio
3.5
7.4
P
<5%
<1%
a. Values coded, for convenience, as pphm.
111-15
-------
• The second highest daily high-hour oxidant level
in each calendar year;
• The yearly average daily high-hour oxidant
level;
• The average daily high-hour oxidant for three
summer months (July, August, September) of each
year;
• The percentage frequency of days with high-hour
average exceeding the NAAQS (0.08 ppm).
The percentage frequency of days with high-hour averages
exceeding 0.16 ppm was also tabulated, but the number of non-zero
values was too small for useful statistical comparisons.
With a single exception, the data for 1970 to 1974 which were
compiled (Tables III-l to III-5) revealed no trends which were
statistically significant at the five percent level. The exception
was the significant decrease in oxidant levels in the Bay Area,
when expressed in terms of the second highest hourly average and
utilizing the pooled information from 9 Bay Area monitoring stations.
This trend amounts to a decrease of about 25 percent over the four-
year interval from 1970 to 1974, giving an average of about 6 percent
per year for that period.
Data for the period 1968 to 1973 for Denver, in terms of yearly
average high-hour oxidant, also showed a significant downward trend,
amounting to more than ten percent per year. This trend was not
reflected in the statistics for frequency of violations of the ambient
air standard during the same years.
Various analyses were performed to explore the indicated rela-
tions between the air quality indices described above. By several
tests, it appears that the other indices follow the second-high
statistic in a general way, but not closely.
An analysis of variance of the five years' data from the Bay
Area confirms that both :the observed yearly trend and the observed
111-16
-------
differences in air quality at various monitoring stations are
statistically significant and, therefore, not attributable to random
variations in oxidant measurements or to unsystematic variations
in external factors which may cause short-term fluctuations in oxi-
dant levels. In short, these effects are too large to be accounted
for by accidental variations; they therefore require explanation in
terms of the major variables intrinsic to the determination of air
quality: emissions and weather.
111-17
-------
REFERENCES AND NOTES
1. Trijonis, Peng, McRae and Lees, "Emissions and Air Quality
Trends in the South Coast Air Basin," EQL Memorandum
No. 16 (January 1976). Calif. Inst. of Technology,
Pasadena, Ca.
2. Kendall, M.G., "Rank Correlation Methods." Griffin,
London, 1948.
The linear trend is evaluated as the coefficient of regression,
where the air quality index is taken as the dependent variable
and the independent variable is the year. The trend is statis-
tically significant at a level £ if the correlation between the
two variables is significant at that level. Using the conven-
tional significance level, p = 0.05, trends calculated for a
5-year data base are significant only if the coefficient of
correlation equals or exceeds 0.88 in absolute value, independent
of sign.
The trend parameter is the coefficient of regression, which
has units of parts per million per year.
The coefficient of concordance is defined as
W = 12S/[m2(n3-n)],
where m is the number of indices to be compared, n_ is the number
of items to be ranked, and S^ is the sum of the squares of the
deviations of the r^ rank totals from their mean, which is given
by m(n +l)/2.
The significance of the coefficient of concordance can be
tested by the use of the ^-distribution, calculating £ as
(m-1)W/(1-W) and entering the _F tables with n-l-2/m degrees of
freedom for the greater estimate and (m-1)(n-1-2/m) degrees of
freedom for the lesser estimate. As an example, if m_= 4
and H = 12, then W^ is significant at the 5 percent level if it
equals or exceeds 0.42.
111-18
-------
IV. OXIDANT:PRECURSOR RELATIONSHIPS
A. SOUTH COAST AIR BASIN, CALIFORNIA
The South Coast Air Basin, which includes the County of Los
Angeles and all or part of five surrounding counties, was not
selected for study as part of this project. The omission was oc-
casioned not by lack of appropriate information but by the recent
publication of a report, "Emissions and Air Quality Trends in the
South Coast Air Basin," which substantially achieved the objec-
tives of the PES project, insofar as the South Coast Air Basin
might be concerned. Since the scope of that report--EQL Memoran-
dum No. 16, by John C. Trijonis, Ted K. Peng, Gregory J. McRae
and Lester Lees, January 1976--was directly comparable with the
work reported here, it is felt appropriate to cite its findings
here for the convenience of interested readers.
The paper documents the historical trends of pollutant emis-
sions and ambient air quality for the period 1965-1974. Emission
trends are developed for nitrogen oxides and reactive hydrocarbons,
as well as carbon monoxide. Ambient air concentrations of total
oxides of nitrogen, nitrogen dioxide, oxidant and carbon monoxide
are reviewed and compared to the changes in emissions, and various
air quality indices are presented for each pollutant.
With respect to the oxidant air quality, the authors report
a Basin-wide improvement of 19 percent over the decade, as the
average improvement for a network of thirteen stations. The es-
timate of Basin-wide reductions in emissions of reactive hydro-
carbons, 18 percent, agrees well with this oxidant reduction.
The geographical trends in oxidant agree, qualitatively, with the
geographical trends in reactive hydrocarbon emissions, especially
for Los Angeles County. The oxidant air quality improved much
IV-1
-------
more in the western areas than in the downwind eastern parts of
the Basin, leading to the postulate that an increase in oxides of
nitrogen emissions improved oxidant air quality near "source-in-
tensive" areas, but caused deterioration in areas farther down-
wi nd.
Histograms of the number of hours that oxidant concentra-
tions exceeded various levels show a much greater improvement in
dosages at high levels (20 to 30 pphm) than in dosages at low
levels, 8 pphm. Thus, it appears that present control strategies
have been more effective in reducing the higher ambient oxidant
concentrations than the lower. This finding may have implications
regarding the feasibility of achieving the NAAQS level in the South
Coast Air Basin.
IV-2
-------
B. REVIEW OF EMISSIONS AND TRENDS. 1970-1974
1. Bay Area Afr Pollution Control District
Emission inventory estimates for the Bay Area Air Pollution
Control District were reviewed in APCD bulletins dated late 1975
and early 1976. Estimates for the years 1970 to 1974, for hydro-
carbons and oxides of nitrogen, are shown in Table IV-1.
The estimates indicate that, during the five year period,
hydrocarbon emissions declined, although not steadily, Evaluated
by a least squares fit, the series shows a trend of -27 tons per
day per year, or about 2.5 percent decline per year. During the
same interval, oxides of nitrogen emissions increased, then
decreased during the last year, arriving at a level perhaps 3 per-
cent lower than the 1970 estimate. Mobile sources accounted for
a little more than half the hydrocarbon emissions and a little less
than two thirds of the oxides of nitrogen, decreasing slightly for
both, over the four year interval. The ratio of hydrocarbon
emissions to oxides of nitrogen emissions seems to have decreased
by at least ten percent.
These emissions are, of course, distributed unevenly with
respect to both geographic subdivisions and source-type categories.
Hydrocarbon emissions estimated by county for 1974 are shown in Table
IV-2, together with the estimated totals for stationary and mobile
sources and the ratio, stationary/mobile. The total ranges from 18
tons per day, for Napa County, to 241 tons per day, for Santa Clara
County. The contribution of stationary sources is greatest in Contra
Costs County, where there are a series of large industrial sources
along an extensive shoreline. The contribution of mobile sources is
greatest in Santa Clara County, which has experienced rapid population
growth during the past decade; relative to the stationary sources,
however, the contribution of mobile sources is greatest in Marin
County, a mountainous suburban county having little industry.
IV-3
-------
TABLE IV-1
POLLUTANT EMISSIONS BY YEAR, BAY AREA AIR
POLLUTION CONTROL DISTRICT3
tons per day
1970 1971 1972 1973 1974
Hydrocarbons, Total 1140 1113 1135 1069 1028
Stationary sources 510 510 550 510 510
Mobile sources 630 603 585 559 518
Percent mobile sources 55 53 51 52 51
Oxides of Nitrogen, Total 740 740 790 820 720
Stationary sources 277 267 296 315 276
Mobile sources 463 473 494 505 444
Percent mobile sources 63 64 63 62 62
Ratio, HC(Total)/NOx(Total)1.54 1.50 1.44 1.30 1.43
a
Source: "Emissions Trends Report, Base Year 1974", Bay Area
Air Pollution Control District, February 6, 1976.
IV-4
-------
TABLE IV-2
BAY AREA AIR POLLUTION CONTROL DISTRICT HYDROCARBON
EMISSIONS BY COUNTY, 1974
(tons per day)
County Total
San Francisco 108
San Mateo 111
Santa Clara 241
Alameda 226
Contra Costa 172
Solano 53
Napa 18
Sonoma 55
Marin 45
Total 1,028
Stationary
54
51
109
103
118
23
9
20
16
504
Mobile
54
60
131
123
55
30
9
35
29
524
Ratio
1.0
0,9
0,8
0.8
2,2
0.8
1.0
0.7
0,6
1.0
IV-5
-------
Similar information regarding emissions of oxides of nitrogen
is presented in Table IV-3. Here the total emissions range from 10
tons per day (in Napa Co.) to 172 tons per day (in Contra Costa).
Emissions from stationary sources range from only 1 ton per day to 121;
from mobile sources, from 9 to 124. Contra Costa county has, by far,
the highest ratio of stationary source emissions to mobile source
emissions, for oxides of nitrogen as for hydrocarbons. It is of interest
to note, however, that Contra Costa County has the lowest ratio
of hydrocarbon emissions to oxides of nitrogen emissions, namely
about 1.0, as compared to ratios up to about 1.6 for other counties.
Trends in mobile-source emissions may be substantially different
from trends in stationary-source emissions, since control programs for
the two types of sources are substantially independent of each other,
both technologically and administratively. To estimate these trends
for areas constituting traffic centers, emissions are usualy computed
by estimating traffic volume (in terms of vehicle miles travelled) and
multiplying by appropriate emission factors. Emission factors have
been subject to a downward trend during the last several years because
of the replacement of old automobiles with newer vehicles equipped
with emission control systems. On the other hand, traffic volumes
have increased in many areas, so that the net trend may be an increase
in some places, despite vehicle emission controls.
An independent estimate of trends in mobile source emissions
was made, using traffic counts for state highways, published annually
by the California Department of Transportation. For this purpose it
was assumed that trends in highway traffic volumes would be accompanied
by similar trends in traffic on local streets. Relative traffic volume
for each locality was thus estimated from traffic counts on the most
heavily travelled highways -- usually, two roads intersecting in or
near the city of interest. These traffic volumes for various years
were multiplied by appropriate emission factors, derived by the Cali-
fornia Air Resources Board, representing the California vehicle
IV-6
-------
TABLE IV-3
BAY AREA AIR POLLUTION CONTROL DISTRICT OXIDES OF NITROGEN EMISSIONS
BY COUNTY, 1974 (tons per day)
County Total
San Francisco 64
San Mateo 68
Santa Clara 156
Alameda 144
Contra Costa 172
Solano 43
Napa 10
Sonoma 34
Marin 31
722
Stationary
15
6
37
20
121
14
1
1
2
217
Mobile
49
62
119
124
51
29
9
33
29
505
Ratio
0.3
0.1
0,3
0,2
2,4
0.5
0.1
0.03
0.07
0.43
IV-7
-------
inventory and the California vehicle emission standards. These
emissions factors were adjusted for an average speed of 45 miles
per hour for freeways and 40 mph for limited access highways except
that for 1974, when the California speed limit was lowered, 40 mph
was assumed for freeways also. The products of traffic counts by
emission factors were normalized to a value of 1.00 for the year, 1970;
Table IV-4 shows the values thus calculated for various Bay Area
localities.
It is of interest to note, from Table IV-4, that the indicated
trend in hydrocarbon emissions is downward in each of the older
major population centers except San Jose. On the other hand, Fremont,
Livermore and San Rafael, outlying communities which like San Jose
have recently experienced substantial population growth, like San Jose
also experienced growth in hydrocarbon emissions from motor vehicles
(although not to the same degree as San Jose).
Emissions of oxides of nitrogen from mobile sources, according
to these estimates, increased after 1970 in all the listed locations
except Richmond (northwestern Contra Costa County). In most cases
they reached a maximum in 1972 or 1973, then underwent a noticeable
decline in 1974. The largest increase was registered in San Jose,
amounting to 35 percent in 1973, then declining to 26 percent in
1974. Due to the reduction in oxides of nitrogen emissions experienced
in 1974, 1974 emissions from mobile sources were lower than 1970
emissions in three localities: San Francisco, Richmond, and Redwood
City.
The BAAPCD Emissions Inventory Summary Report for Base Year
1974 lists eighty-four major point sources of various pollutants,
where "major" implies emissions of at least 200 pounds per day.
The list includes eight facilities estimated to emit at least 5 tons
of hydrocarbons per day and six which emit at least five tons of
oxides of nitrogen per day. Collectively, these plants emit more
than half the oxides of nitrogen attributed to stationary sources
IV-8
-------
TABLE IV-4
MOBILE SOURCE EMISSION INDICES*
County
San Francisco San Francisco
Marin
San Mateo
San Rafael
Contra Costa Richmond
Pittsburg
Alameda Oakland
Fremont
Livermore
Santa Clara San Jose
Redwood City
1970
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOX
Hydrocarbon
NOV
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
1971
0.
1.
0.
1.
0.
0.
0.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0.
1.
97
03
99
07
91
98
95
03
01
09
03
12
13
22
09
18
93
01
1972
0.
1.
1.
1.
0.
1.
1.
1.
0.
1.
1.
1.
1.
1.
1.
1.
0.
1.
93
04
01
13
89
00
00
12
99
11
08
21
07
20
15
29
94
05
1973
0
1
1
1
0
1
0
1
0
1
1
1
1
1
1
1
0
1
.90
.02
.00
.16
.86
.00
.97
.12
.96
.11
.06
.23
.07
.23
.17
.35
.89
.03
1974
0.
0.
1.
1.
0.
0.
0.
1.
0.
1.
1.
1.
1.
1.
1.
1.
0.
0.
87
96
07
12
86
91
98
02
96
01
06
11
07
11
21
26
89
94
*Normalized to 1970
IV-9
-------
and about one fifth of the hydrocarbons from stationary sources.
Figure II-5 shows the geographical distribution of these large
sources: the largest number of them are situated in Contra Costa
County, on the shore of one of the bays or connecting straits.
Other large facilities exist, as shown, in Solano, Alameda, and Santa
Clara Counties. Except for two electric power generating plants in
the eastern part of San Francisco, all sources emitting more than
one ton per day of either hydrocarbons or oxides of nitrogen are
also distributed in a pattern similar to that shown in Figure II-5,
although with more smaller emitters along the shore of Alameda County.
The largest emitters of hydrocarbons are petroleum refineries,
chemical manufacturing plants, and automobile assembly plants.
Principal emitters of oxides of nitrogen include the refineries and
chemical plants and, in addition, electric power generating plants
and a large cement plant.
2. Redding (Sacramento Valley Air Basin)
Emission inventory information for Shasta County has been
obtained from the Shasta Air Pollution Control District and the
California Air Resources Board. Table IV-5 shows estimates for the
years 1970 to 1974, except 1971, for hydrocarbons and oxides of
nitrogen.
The estimates indicate that hydrocarbon emissions, on the
whole, decreased by about 45 percent between 1970 and 1974 while
oxides of nitrogen emissions decreased by about 20 percent between
1970 and 1972, thereafter undergoing even less variation. For
hydrocarbons, the fraction of emissions attributable to motor vehicles
and other mobile sources decreased strongly in the period 1970 to
1972, going from 69 percent to 41 percent; subsequent changes in the
proportion of the contributions from motor vehicles were relatively
inconsequential. For oxides of nitrogen, the proportion attributable
to mobile sources remained at seventy to eighty percent.
IV-10
-------
TABLE IV-5
POLLUTANT EMISSIONS BY YEAR, SHASTA COUNTY PORTION
OF SACRAMENTO VALLEY AIR BASIN
1970°
tons per day
1972L
1973
1974
Hydrocarbons, Total
Stationary sources
Mobile sources
Percent mobile sources
Oxides of nitrogen, Total
Stationary sources
Mobile sources
Percent mobile sources
45
15
31
69
22
7
15
69
39
22
16
41
17
3
14
81
28
17
11
39
19
5
14
73
25
13
11
44
18
4
14
77
Ratio, HC(Total)/NOx(Total)
a
b
c
2.0
2.3
1.5
1.4
Source: California Air Resources Board
Source: California Air Resources Board, preliminary data
Source: Shasta County Emission Inventory 1974, supplied by
Shasta County Air Pollution Control District.
IV-11
-------
If attention be focused on the City of Redding which has
about one fifth of the population of the entire county, a somewhat
different picture emerges. Table IV-6 shows estimates of mobile
source emission indices (normalized to unity for 1970), derived as
explained above (Section IV-B1). By this measure, hydrocarbon
emissions from traffic in the immediate vicinity of Redding increased
by 12 percent between 1970 and 1974, while nitrogen oxides showed an
increase of 22 percent by 1973, falling off in 1974 to 16 percent
above the 1970 value. These increases apparently stemmed from a
rather rapid increase of automobile traffic in the Redding vicinity.
There was no sign of any striking change in emissions from stationary
sources during these years.
Thus, while the county-wide estimates indicate a substantial
decrease in total hydrocarbon emissions and a somewhat smaller
decrease in emission of oxides of nitrogen, there may have been
increases in emissions of both pollutants within the immediate
vicinity of Redding during the period 1970-1974.
3. Stockton and Modesto (San Joaquin Valley Air Basin)
The San Joaquin Valley Air Basin is represented in this review
by two cities, both a few miles east of the adjacent San Francisco
Bay Area Air Basin: Stockton, in San Joaquin County, and Modesto,
in Stanislaus County.
Emission inventory estimates for these two counties have been
provided by the California Air Resources Board for each year of the
pentad, 1970 to 1974, except 1971. These estimates are shown in
Table IV-7 for San Joaquin County, and in Table IV-8 for Stanislaus
County.
These counties are both primarily agricultural and they show
quite similar emission patterns, with San Joaquin County emissions
on the whole about 50 percent larger than those of Stanislaus County.
IV-12
-------
TABLE IV-6
MOBILE SOURCE EMISSION INDICES, REDDING
(SHASTA COUNTY, SACRAMENTO VALLEY AIR BASIN)
tons per day
1970a 1971 1972 1973 1974
Hydrocarbons 1.00 1.01 1.01 1.05 1.11
Oxides of Nitrogen 1.00 1.10 1.13 1.22 1.16
3 Normalized to 1970 = 1.00
IV-13
-------
TABLE IV-7
POLLUTANT EMISSIONS BY YEAR, SAN JOAQUIN COUNTY
(SAN JOAQUIN VALLEY AIR BASIN)
1970°
tons per day
1972
1973"
1974
Hydrocarbons, Total
Stationary sources
Mobile sources
Percent mobile sources
Oxides of Nitrogen, Total
Stationary sources
Mobile sources
Percent mobile sources
117
41
71
63
46
8
38
82
83
30
53
64
49
7
42
85
72
37
35
49
50
9
42
83
77
42
35
46
51
9
42
82
Ratio, HC(Total)/NOx(Total)
a
b
c
2.4
1.7
1.4
1.5
Source, California Air Resources Board
Source, California Air Resources Board, Preliminary Data
Source, San Joaquin County Emission Inventory
IV-14
-------
TABLE IV-8
POLLUTANT EMISSIONS BY YEAR, STANISLAUS COUNTY
(SAN JOAQUIN VALLEY AIR BASIN)
1970
tons per day
1972 1973
1974
Hydrocarbons, Total
Stationary sources
Mobile sources
Percent mobile sources
Oxides of Nitrogen, Total
Stationary sources
Mobile sources
Percent mobile sources
70
21
49
70
30
4
26
85
63
25
38
60
34
4
30
88
54
29
26
48
36
5
31
86
50
24
26
52
35
5
29
85
Ratio, HC(Total)/NOX(Total)
2.3
1.8
1.5
1.4
IV-15
-------
Hydrocarbon emissions from all sources decreased by about 30 percent
in each county, between 1970 and 1974; the hydrocarbon contribution
from mobile sources decreased by about 50 percent in each county during
that time, while the stationary source emissions showed no consistent
trend in either county. As a consequence, the fraction of hydrocarbons
attributable to mobile sources declined appreciably in both counties.
Emissions of oxides of nitrogen increased slightly in both San
Joaquin County and Stanislaus County between 1970 and 1972, then
levelled out, with very little change between 1972 and 1974. The
fraction attributable to mobile sources was between 82 and 85 percent
for San Joaquin County and slightly higher — between 85 and 88 percent --
for Stanislaus County. Ratios of hydrocarbon emissions to oxides of
nitrogen emissions decreased in a similar fashion in both counties,
falling from about 2.4 in 1970 to about 1,4 in 1974.
Estimated mobile sourcee indices for the cities of Stockton and
Modesto, calculated as explained in Section IV-B1, are shown in Table
IV-9. Here, the differences between the two localities are a little
more obvious. Motor vehicle traffic in the Stockton vicinity increased
sharply in 1972 after a new freeway link was opened, causing increases
in mobile source emissions, which were not paralleled in Modesto.
Accordingly, over the four-year interval, hydrocarbon emissions from
motor vehicles decreased by 15 percent in Modesto, but by only 2
percent in Stockton.
Emissions of oxides of nitrogen showed even greater divergence,
as the table indicates. In Stockton, the emissions rose sharply to a
peak in 1973, 25 percent above 1970; in Modesto, emissions rose
slightly until 1972, then decreased more sharply to a new low in 1974
about 8 percent less than in 1970.
Neither city appears to have been much affected by changes in
emissions from stationary sources during this period. In the Modesto
area, a single large manufacturing plant emitting solvent vapors at
IV-16
-------
TABLE IV-9
MOBILE SOURCE EMISSION INDICES, TWO CITIES IN
SAN JOAQUIN VALLEY AIR BASIN
1970a 1971 1972 1973 1974
Hydrocarbons, Stockton 1.00 0.95 0.98 1.01 0.98
Modesto 1.00 0.97 0.94 0.90 0.85
Nitrogen Oxides, Stockton 1.00 1.04 1.12 1.25 1.15
Modesto 1.00 1.05 1.05 1.03 0.92
a normalized to 1970 = 1.00
IV-U
-------
the rate of about 2 tons per day was subjected to emission controls,
which became operational only late in 1974.
Each city is credited with a population of about 35 to 40 percent
of the entire county population. San Joaquin County had about 300,000,
Stanislaus County about 200,000 residents during the period discussed
here. Aside from these population differences and the roughly pro-
portional differences in emissions, the two areas can be considered
as having similar activities and emission patterns.
4. Salinas (North Central Coast Air Basin)
Emission inventory estimates for Monterey County have been
obtained from the Monterey Bay Unified Air Pollution Control District
and from the California Air Resources Board. Table IV-10 shows
these estimates for the years 1970 to 1974 except 1971, for hydrocar-
bons and oxides of nitrogen.
The estimates indicate that hydrocarbon emissions, on the whole,
increased dramatically during the period under study -- from 81 tons
per day in 1970 to 178 tons per day in 1974. These estimates may not be
taken at face value, however, because the basis for estimation changed
after 1972, to reflect newly recognized emissions from a producing oil
field in the Salinas Valley. In fact, it is probable that the
emissions of hydrocarbons from this source were of the same order of
magnitude before they came to the attention of the control agencies as
afterwards, and that the actual changes in emissions from year to year
did not exceed about 20 tons per day, the apparent increase in hydro-
carbons from recognized stationary sources between 1970 and 1972.
Emissions of oxides of nitrogen appear to have increased by about
30 percent between 1970 and 1972, and to have changed only slightly
thereafter. Between 40 and 50 percent of these emissions seemed to be
attributable to mobile sources in the county. Mobile source emissions
in the immediate vicinity of Salinas have been estimated in terms of
the mobile source index, derived as explained above (Section IVB-1).
IV-18
-------
TABLE IV-10
POLLUTANT EMISSIONS BY YEAR, MONTEREY COUNTY
(NORTH CENTRAL COAST AIR BASIN)
Hydrocarbons, Total
Stationary sources
Mobile sources
Percent mobile sources
Oxides of nitrogen, Total
Stationary sources
Mobile sources
Percent mobile sources
Ratio, HC(Total)/NOX(Total)
tons per day
1970a 1972a 1973 1974
163 178
135 138
27 40
16 23
77 72
44 40
32 32
42 45
(1.4) (1-2) 2.1 2.5
(81)
(28)
53
(65)
59
30
29
49
(91)
(48)
43
(47)
76
40
36
48
During the years prior to 1973, the inventories did not include
emissions of hydrocarbons from a producing oil field in which
steam injection was employed as an aid to oil recovery. In
1973 and 1974, the emissions from this source alone were esti-
mated to be more than 100 tons per day. Parentheses in these
columns indicate that the figures presented are probably erron-
eous .
IV-19
-------
Table IV-11 shows the estimated values of this index, normalized
to unity for 1970 emissions. The table indicates that neither hydro-
carbon emissions nor nitrogen oxides emissions deviated by more than
10 percent from the 1970 value during the next four years.
On the whole, the impression is gained that emissions of oxidant
precursors were fairly constant during the study period, both in the
immediate vicinity of Salinas and in Monterey County generally.
5. Lancaster (Los Angeles County Portion of Southeast Desert
Air Basin)
Emission inventory information for the Los Angeles County portion
of the Southeast Desert Air Basin has been obtained from the California
Air Resources Board and from the Southern California Air Pollution
Control District. Table IV-11 shows estimates for the years 1970,
1972, and 1973, with somewhat less detailed information for 1974.
The estimates indicate that hydrocarbon emissions decreased
rather abruptly from 28 to 22 tons per day between 1972 and 1973,
due to an even sharper drop in hydrocarbons from mobile sources. This
decrease in mobile source hydrocarbons was apparently not accompanied
by a corresponding drop in emissions of oxides of nitrogen from mobile
sources. Except for 1970, the information on which Table IV-12 is
based was furnished with the warning (from ARB) that it was
unpublished, preliminary information subject to revision. Further,
major organizational changes in the cognizant local agencies have
recently been made. For these and other reasons, it seems likely that
some of the entries listed in Table IV-12 are of relatively low
reliability.
Mobile source emissions in the immediate vicinity of Lancaster
have been estimated in terms of the mobile source index, derived as
explained in Section IV B-l. Table IV-13 shows these values, normalized
to unity for 1970 emissions, for both hydrocarbons and oxides of
nitrogen. The table indicates that emissions of both pollutants from
IV-20
-------
TABLE IV-11
MOBILE SOURCE EMISSION INDICES, SALINAS
(MONTEREY COUNTY, NORTH CENTRAL COAST AIR BASIN)
tons per day
19703 1971 1972 1973 1974
Hydrocarbons 1.00 0.94 0.90 0.92 0.94
Oxides of Nitrogen 1.00 1.02 1.01 1.07 0.98
3 Normalized to 1970 = 1.00
IV-21
-------
TABLE IV-12
POLLUTANT EMISSIONS BY YEAR, LOS ANGELES COUNTY PORTION
OF SOUTHEAST DESERT AIR BASIN (LANCASTER AREA)
tons per day
1970 1972 1973
1974
Hydrocarbons, Total
Stationary sources
Mobile sources
Percent mobile sources
Oxides of Nitrogen, Total
Stationary sources
Mobile sources
Percent mobile sources
27
5
22
81
11
2
9
86
28
6
22
78
12
1
12
96
22
10
11
52
13
1
12
91
20
11
Ratio, HC(Total)/NO,.(Total)
2.6
2.3
1.7
1.8
IV-22
-------
TABLE IV-13
MOBILE SOURCE EMISSIONS INDICES, LANCASTER
(SOUTHEAST DESERT AIR BASIN)
tons per day
1970a 1971 1972 1973 1974
Hydrocarbons 1.00 0.89 0.73 0.73 0.72
Oxides of Nitrogen 1.00 0.96 0.82 0.83 0.78
a Normalized to 1970 = 1.00
IV-23
-------
mobile sources decreased rather markedly; hydrocarbons were down
by 11 percent by 1971, 27 percent by 1972, and oxides of nitrogen
decreased by 18 percent, mainly between 1971 and 1972. Since there
are no known large stationary sources of these contaminants in the
Lancaster vicinity, it seems likely that the trend of emissions in the
Lancaster area was downward for the entire period under study. (This
conclusion is also consistent with the information given in Table IV-12,
aside from the question of its reliability.)
6. Denver, Colorado (Metropolitan Denver Air Quality Control
Region).
Information about emissions in the Denver area is, at best,
fragmentary. Annual reports provided by the Colorado Air Pollution
Control Commission indicate that an emission inventory for the seven
counties in the AQCR was carried out in the summer of 1971 and that
supplementary information suitable for updating was obtained in 1974.
Hydrocarbon emissions were thus assessed at 463 tons per day in 1971
(applicable to 1970) and at 412 tons per day three years later.
Corresponding tonnages of nitrogen oxides were 277 and 247, respectively.
For Denver County alone, estimates obtained from the Environmental
Health Service put hydrocarbon emissions at 180 tons per day in 1969,
201 in 1970, and 194 in 1971 and subsequently.
Taken at face value, these estimates suggest that emissions
of hydrocarbons were rising until about 1970, but decreased gradually
from then to 1973. Emissions of oxides of nitrogen apparently decreased
similarly from 1970 to 1973, but no information is available as to their
status before 1970.
7. Philadelphia
Information about emissions of hydrocarbons and oxides of
nitrogen for the Philadelphia area was furnished by the Philadelphia
Department of Public Health, Air Management Services. The estimates
are shown in Tables IV-14 and IV-15 for the years 1969 through 1973.
IV-24
-------
TABLE IV-14
HYDROCARBON EMISSIONS BY YEAR, PHILADELPHIA, PA.
tons per day
1969 1970 1971 1972 1973
From Industrial Processes 129 124 118 172 158
From Space Heating 21 22 20 20 18
From Power Generation 22332
From Refuse Incineration 31 31 31 5 5
From Transportation 351 337 329 278 232
From Gasoline Handling3 (18) (18) (18) 18 18
From Non-industrial Solvents (7) (7) (7) 7 7_
Total 559 541 526 503 440
a
Entries for these categories in years 1969 through 1971 were
not given in original inventories, but supplied by PES to
maintain comparable totals.
IV-25
-------
TABLE IV-15
NITROGEN OXIDE EMISSIONS BY YEAR, PHILADELPHIA, PA.
tons per day
1969 1970 1971 1972 1973
From Industrial Processes 24 24 22 40 37
From Space Heating 98 107 99 98 89
From Power Generation 98 98 119 103 89
From Refuse Incineration 33333
From Transportation 90 93 96 111 112
Total 313 325 339 355 330
IV-26
-------
For hydrocarbons, Table IV-14 indicates a generally declining
emission rate. The inventory basis was altered slightly, beginning
with the 1972 accounting, when 26 tons per day of emissions from gaso-
line handling and from non-industrial use of solvents were recognized
and added in. Assuming similar contributions from these categories
in the earlier years, the successive annual totals give a uniformly
decreasing sequence, with a substantially greater drop between 1972 and
1973 thaln earlier years. .
Nitrogen oxides emissions, in contrast, showed a generally
increasing emission rate from 1969 to 1972, falling off slightly in
1973.
Regarding the comparisons of these totals from year to year,
the Air Management Services report warns, "Due to variations in ...
methods and response...it would be unwise to draw fine-point dis-
tinctions from emission estimate differences. Each revised emission
factor publication usually contains more accurate values as more
information and experience are obtained. Consequently, changes
between inventories may be due to revisions in emission factors which
may be greater than changes in actual emissions...unqualified inter-
pretations are not recommended particularly between emissions and air
quality data."
Despite the Air Management Services' conscientious explanations,
the data presented in Tables IV-14 and IV-15 are the only data
available for estimating the trends and they are, in all probability,
no less accurate than data from the other jurisdictions represented
in this report. Accordingly, it is reasonable to take these data at
face value for the present purpose, as indicating the existence of a
negative trend in hydrocarbon emissions and a positive trend in nitrogen
oxide emissions over the period 1969 through 1973.
IV-27
-------
C. COMPARISON BETWEEN OXIDANT TRENDS AND EMISSION TRENDS, 1970-1974
1. San Francisco Bay Area
As noted in Section III-B, photochemical oxidant levels in the
San Francisco Bay Area exhibited a significantly downward trend when
data from nine stations, expressed in terms of yearly second highest
daily maximum, were pooled. Hydrocarbon emissions also declined
substantially over the pentad, 1970 to 1974, as discussed in Section
IVB-1. The correlation coefficient between these quantities, pooled
second-high oxidant level and regionwide hydrocarbon emissions, was
0.58, positive but not significantly greater than zero in view of the
small number (5) of data points.
This low degree of correlation is not very surprising in view of
the expected effects of various other factors (besides hydrocarbon
emissions) on the development of photochemical oxidants. Chief among
these factors are emissions of oxides of nitrogen, weather variations,
and effects of geography and terrain.
Since oxides of nitrogen (along with hydrocarbons) are pre-
cursors of oxidants, it was of interest to explore correlations of
the oxidant levels with emissions indices which might combine the
effects of hydrocarbons and oxides of nitrogen. When the index used
was the total emissions of both pollutants, the correlation coefficient
was 0.60 -- practically unchanged. When the ratio of total hydro-
carbons to total oxides of nitrogen emissions was tested, the corre-
lation coefficient was found to be 0.88, a value significant at the
5 percent level. A correlation coefficient of 0.87, also significant,
was obtained when the emissions index was taken as the difference
between hydrocarbon emissions and oxides of nitrogen emissions.
These findings are especially interesting in view of the
known chemical effects of nitric oxide on generation of oxidant in
irradiation chamber experiments, where it is generally found that an
excess of nitric oxide tends to retard the accumulation of ozone,
IV-28
-------
even though it may lead eventually to higher ozone levels. If this
retardant effect were predominant in the atmosphere, a positive
correlation between oxidant levels and the HC/NO ratio would be
/\
expected. Also, if this were the case, it would imply that
decreasing emissions of oxides of nitrogen should be expected to
have little effect in reducing oxidant levels, and might even tend
to increase them. The correlation between oxidant level and nitrogen
oxides emissions was -0.55, negative but not significant.
In any event, these results show that a large traction of the
variability in oxidant air quality from year to year can be
accounted for in terms of the effects of varying rates of emissions,
where both hydrocarbons and nitrogen oxides emissions are recognized
as contributing factors. Emissions data are not readily available
in sufficient .detail to permit testing these correlations for the
various monitored localities within the Bay Area Air Pollution .Control
District but it seems unlikely that stronger associations wowld be
found, because of the known tendency for the primary contaminants, once
emitted, to travel for many miles before oxidant peaks are generated.
This aspect of the system is discussed in a later section of this
report.
2. Denver
The information reported in Section III-C indicates that oxidant
air quality at one station in Denver, expressed in terms of the average
daily maximum oxidant concentration for each year, decreased steadily
between 1968 and 1973, at an average rate of 0.0033 ppm per year.
Available information on emissions-of hydrocarbons and oxides of
nitrogen is not sufficiently detailed to establish whether a comparable
reduction in emissions took place at the same time.
In the Metropolitan Denver Air Quality Control Region, hydro-
carbon emissions are reported to have decreased from 463 tons per day
in 1971 to 412 tons per day in 1974, an average of about four percent
per year. Oxides of nitrogen emissions are said to have decreased from
IV-29
-------
277 to 247 tons per day in the same period, an average of almost four
percent per year. For Denver County alone, emissions have been
reported as increasing from 180 tons per day in 1969 to 194 tons per
day in 1971, an average increase of four percent per year. These
figures show no apparent relation to the oxidant air quality trend
indicated above.
During the interval from 1971 to 1974, average daily traffic
at the intersection nearest to the monitoring station increased by
only six percent, or two percent per year. This stability could
account, in part, for the observed oxidant trend, if the hydrocarbon
emission factors for motor vehicles were decreasing during the same
period. A simultaneous increase in motor vehicle emission factors
for oxides of nitrogen could enhance this effect. However, these
conditions were probably not achieved until after 1971, whereas the
observed oxidant trend covers a longer period, starting three years
earlier. Thus, although plausible ad hoc explanations for the oxidant
trend may exist, the available data base is insufficient to confirm
any of them.
3. Other Areas Included in Study
As shown in Chapter III, none of the individual oxidant trends
were found to be significant except in the San Francisco Bay Area and
the Denver area, already discussed in previous sections. Since these
individual trends were not significant, a detailed comparison with
emissions trends would not be especially useful. However, if the trend
direction for air quality agrees with the trend direction for emissions
in a large enough proportion of the cases studied, this would constitute
additional support for the hypothesis that reduction of emissions should
lead to improved oxidant air quality.
Table IV-16 shows the matrix of trend directions for various
emissions indices and air quality indices for the six locations not
already discussed more fully. Cursory inspection shows that there is
IV-30
-------
TABLE IV-16
TREND DIRECTIONS FOR VARIOUS EMISSIONS INDICES AND
VARIOUS AIR QUALITY INDICES, FOR FIVE
CALIFORNIA LOCATIONS AND PHILADELPHIA
location
RD ST MO SA LN PH
Hydrocarbon emissions _ _ _ 0 _ _
Mobile Source Index + 0
Nitrogen Oxide emissions - + + + -
Yearly second high Ox 0 + - + + +
Average maximum Ox + + - + + +
3rd Qtr Average max. Ox + + - + + +
Frequency, Ox>0.08ppm + + - + + -
Locations: RD, Redding, CA
ST, Stockton, CA
MO, Modesto, CA
SA, Salinas, CA
LN, Lancaster, CA
PH, Philadelphia, PA
IV-31
-------
no general agreement as to trend direction, between any of the
emissions indices and any of the air quality indices as listed.
Thus the matrix in toto furnishes no more support for the hypo-
thesis of a connection between emissions and air quality than can
be found by an examination of the individual cases.
IV-32
-------
D. REVIEW OF LONG-TERM TRENDS. SAN FRANCISCO BAY AREA
Assessment of a trend in air quality in any given region is
likely to be complicated by the variations in weather conditions
from year to year, which may be expected to cause fluctuations in
the air quality indices, even in the absence of any change in
emission inventories. Even trends which have been evaluated as
statistically significant are not necessarily caused by reductions
in emissions, since weather trends might conceivably have the same
effect. The longer the period of observation is, the less is the
likelihood of a sustained trend due to weather factors alone. A
five-year period, however, is not long enough to rule out the
appearance of air quality trends unrelated to abatement of pollutant
emissions.
In relation to photochemical oxidant levels, the main weather
factors expected to cause important effects are wind, sunshine,
and atmospheric stability -- the existence and height of a mixing
layer. In any given situation, the development of high concentrations
of oxidants is favored by low wind speeds, bright sunshine, and low
mixing heights (.i.e., low inversions).
In studies of oxidant levels using a data base going back to
1954, the Technical Services Division of the Sain Francisco Bay Area
Air Pollution Control District has observed that daily maximum tem-
peratures tend to reflect both sunshine intensity and limited venti-
lation. Thus the days with weather most conducive to oxidant forma-
tion are also days on which the maximum temperature falls within some
readily discernible range. The Division has made use of this obser-
vation in formulating criteria for selecting days of comparable
weather to refine the detection of weather-independent oxidant trends.
To further sharpen the definition of comparable days, the
Division added a criterion reflecting vertical dilution, namely,
an inversion base within a certain height limit at a given time of
IV-33
-------
day. They then investigated the oxidant levels attained on all
days meeting these criteria over the 13-year period, 1962 through
1974.
Table IV-17 shows the results of this analysis, for one
station in each of the original six counties of the District.
The index for the six stations combined declined from values of 0.12
and 0.13 ppm in the period 1962 to 1965 to values of 0.08 and 0.09
in the period 1970 through 1974. Although the trends at the individual
stations were not completely parallel to the general trend, all
reflected the general trend to an appreciable extent.
Specifically for the period 1970 to 1974, a downward trend is
evident for all of the original six stations except San Jose, which
registered a new high for the pentad of 0.16 ppm in 1974, Table
IV-18 shows how the trends indicated by the Division's analysis
compare with those identified in Section III-G (Table III-6) for the
stations common to both analyses. The Division's index, for these
five stations, agrees with the index based on frequency of oxidant
values exceeding 0.08 ppm in showing positive trends at San Jose and
Livermore and negative trends at the other three locations.
As reported above (Section III-H), trend magnitudes for the
Bay Area for the last five available years were, for the most part,
found to be not statistically significant. The test involves the
Student t_ statistic, as explained by Moroney/2' with only 5 points
to establish a trend, a high correlation coefficient, 0.88 or larger,
is needed. With more years of data, trends may be significant even
though accompanied by much lower correlation coefficients. Figure
IV-1 illustrates the relation between the number of points available
and the correlation coefficient needed for significance at the 5
percent level. With eight points, coefficients above 0.71 are
significant; with thirteen points, coefficients above 0.55 are
significant.
IV-34
-------
TABLE IV-17
AVERAGE HIGH-HOUR OXIDANT CONCENTRATIONS FOR DAYS WITH COMPARABLE TEMPERATURE AND
INVERSION CONDITIONS. (APRIL THROUGH OCTOBER OXIDANT SMOG SEASONS, 1962-1974)
(Source: Information Bulletin 3-25-75: A Study of Oxidant Concentration Trends: Technical
Services Division, Bay Area Air Pollution Control District.)
Monitoring
Station
San Francisco
San Leandro
San Jose
Redwood City
Walnut Creek
San Rafael
BAAPCD
Average*
Li vermore**
'62
.14
.13
.11
.13
,10
.08
.12
—
'63
.12
.16
.17
.10
.11
.09
.12
--
'64
.15
.19
.14
.10
.10
.07
.13
--
Average High-Hour Oxidant Concentration
(KI parts per million)
'65
.09
.19
.16
.14
.11
.08
.13
--
'66
.08
.14
.11
.10
.10
.07
.10
--
'67
.08
.12
.13
.09
.13
.07
.10
.13
'68
.05
.11
.13
.08
.10
.06
.09
.18
'69
.04
.12
.13
.09
.13
.07
.10
.18
'70
.07
.12
.12
.08
.09
.08
.09
.13
'71
.05
.11
.08
.07
.09
.07
.08
.11
'72
.03
.10
.10
.08
.09
.05
.08
.09
'73
.04
.11
.11
.07
.08
.05
.08
.12
'74
.05
.10
.16
.07
.08
.06
.09
.13
13-yr
.08
.13
.13
.09
.10
.07
.10
.13
* For benchmark stations above, with 13 years of record.
** Station with 8 years of record.
CO
en
-------
C
0)
o
4-
0)
O
0
O
(O
r—
Ol
S-
O
O
1.0
0.9
0.8
0.7
0.6
0.5
p = 0.05
10 11
12
13
14
15
Number of points N
Figure IV-1. Smallest value of correlation coefficient
.significantly different from 0 at p = 0.05,0.01,0.001
(two-way test), for a data base of N points.
(Number of degrees of freedom = N-2)
CO
CTi
-------
Using the data shown in Table IV-17, tests for significance
were made on various subsets of the data. When the data for all six
stations having a 13-year record were pooled, the correlation
coefficient was -0.880, which is significant at the 0.1 percent level;
thus, there is no question that oxidant air quality for the Bay Area
District has improved since 1962. Again, when the data for all seven
stations having an 8-year record were pooled, the correlation
coefficient was -0.741, significant at the five percent level. Over
periods ending in 1974 but shorter than eight years, significant
correlations were not found, mainly because of the sharp upturn
observed at San Jose in 1974. (With San Jose omitted, the remaining
six stations showed a significant trend for the six-year period,
1969 through 1974.)
Since it was clear that hydrocarbon emissions had been on the
decline in the Bay Area District for several years, it was of interest
to test correlation of oxidant air quality levels against emissions
for the corresponding years. Accordingly, estimates of hydrocarbon
emissions were taken from a recent tabulation (BAAPCD Emission Trends
Report, Base Year 1974, issued Feb. 6, 1976) and used in a correlation
analysis. When these emissions data (Table IV-19) were paired with
pooled oxidant air quality data for six stations (Table IV-17) for
13 years, the correlation coefficient was +0.767, significant at the
5 percent level and higher than the value (0.58) found for correlation
between hydrocarbon emissions and yearly second-high oxidant, 1970 to
1974 (Section IV-C-1).
Again, indices combining the effects of emissions of hydro-
carbons and nitrogen oxides were tested. When the index used was
the total of both pollutants, there was no appreciable correlation
(r = -0.16). When the ratio of hydrocarbon emissions to nitrogen
oxide emissions was tested, the coefficient was found to be 0.92,
significant at the 0.1 percent level. An index in which the oxides
of nitrogen emissions were subtracted from hydrocarbon emissions (on
IV-37
-------
TABLE IV-18
TREND DIRECTIONS FOR VARIOUS OXIDANT INDICES
SHOWN AT VARIOUS MONITORING STATIONS,
SAN FRANCISCO BAY AREA.
SF RC SJ SL LV
Average max on comparable days - - + - +
Yearly second high maximum +
Yearly average maximum - + + + +
Third quarter average maximum - - + + +
Frequency> 0.08ppm - - + - +
Locations: SF, San Francisco
RC, Redwood City
SJ, San Jose
SL, San Leandro
LV, Livermore
IV-38
-------
TABLE IV-19
ESTIMATED EMISSIONS BY YEAR, SAN FRANCISCO BAY
AREA AIR POLLUTION CONTROL DISTRICT, 1962-1974,
AND COMBINED EMISSION INDICES, FOR
HYDROCARBONS AND OXIDES OF NITROGEN.
tons per day
Year HC
1962 1300
1963 1300
1964 1400
1965 1300
1966 1300
1967 1300
1968 1300
1969 1300
1970 1140
1971 1110
1972 1130
1973 1070
1974 1030
NOX
490
500
550
550
620
620
670
700
740
740
790
820
720
(HC+NOX)
1790
1800
1950
1850
1920
1920
1970
2000
1880
1850
1920
1890
1750
(HC/NOX)
2.65
2.60
2.55
2.36
2.10
2.10
1.94
1.86
1.54
1.50
1.43
1.30
1.43
(HC-NOX)
810
800
850
750
680
680
630
600
400
370
340
250
310
IV-39
-------
a ton-for-ton basis) gave correlation coefficient 0.892, also
significant at the 0.1 percent level. The coefficient of
correlation between oxidant level and nitrogen oxides emissions
alone was, in this analysis, -0.91; this is nearly as large as
the coefficient cited above in which hydrocarbons were included,
and is equally significant.
These results reinforce those discussed in Section IV-C-1
in a very satisfactory way. The correlation coefficients obtained
with the use of a relatively weather-independent oxidant index turn
out to be of the same order of magnitude as, but slightly higher
than, those found for the yearly second-high, an index more
sensitive to weather. More important, the trends shown to be
marginally significant when assessing data for five years (1970
to 1974) are found to be very convincing when data for 13 years
are available. The fact that the correlation coefficient is
larger for oxides of nitrogen emissions than for hydrocarbon
emissions may be due to the larger relative range of the oxides
of nitrogen emissions during the period under study. It should
not be construed as proving that the influence of hydrocarbons
on oxidant levels is necessarily small relative to the influence
of oxides of nitrogen, since there is certainly a substantial
degree of confounding between these two nominally independent
variables—the correlation between them is -0.79, significant
at the 0.01 level.
IV-40
-------
REFERENCES
1. A. Ranzieri, Personal Communication
2. M. J. Moroney, "Facts From Figures," 3rd Edition, p. 311
Penguin Books. Baltimore, Maryland, 1956.
IV-41
-------
V. INTERACTIONS WITH WEATHER AND OTHER FACTORS
A. SAN FRANCISCO BAY AREA, WEATHER VARIABLES
Although the correlations described in Sections IV-C and
IV-D have demonstrated the existence of a statistically significant
relationship between oxidant air quality and precursor emissions,
on an annual basis, it is nevertheless of interest to investigate
whether there may also be a demonstrable relationship between
oxidant air quality and available weather variables, year by year.
A convenient statistical tool for this purpose is rank correlation.
If changes in the weather from year to year can have an appreciable
influence on air quality, as is commonly assumed, then it should be
possible to show some degree of agreement between measured air quality
indices and observed weather parameters, despite the relatively long-
term trends in contaminant emissions. From the BAAPCD Information
Bulletin of 3-25-75, we have records of the number of days judged
conducive to smog in the District's study of oxidant concentration
trends. These may evidently serve as a rough indication of the joint
effect of relevant weather parameters.
For the seven stations treated in the Bulletin, the average
numbers of smog-conducive days reported in the successive years
1970 to 1974 were 46.3, 45.0, 36.7, 42.0 and 44.3, respectively. Thus
the ranking of these years with respect to smog conduciveness was
1, 2, 5, 4, 3. The various annual oxidant indices were tested for
agreement with this order for each of the nine air monitoring
stations included in this study. The results, in terms of rank
correlation coefficient, are shown in Table V-l.
The correlation coefficients found range from + 0.8 to -0.5.
Individually, none of these is statistically significant at the five
percent level, since the minimum value for significance with N=5 is
0.87. Cumulatively, however, these results show convincing significance
V-l
-------
TABLE V-l. RANK CORRELATION COEFFICIENTS BETWEEN VARIOUS ANNUAL
OXIDANT INDICES AND AVERAGE NUMBER OF SMOG-CONDUCIVE
DAYS, FOR YEARS 1970 TO 1974
INDEX STATION
SF RC SJ FR SL OK RI PT LV AVG.
Second High .6 .5 -.5 0 .8 .7 .7 .4 .7 .4
Average Maximum .3 -.2 .5 -.2 .5 .3 .5 .3 .5 .3
3rd Quarter Average .4 -.3 .2 .5 .6 .8 .7 .4 .5 .4
Frequency >0.08 .7 .4 .4 -.1 .7 .8 .5 .6 .5. .5
V-2
-------
for each oxidant index, as the number of values greater than zero
is seven or eight out of nine in each case. (Tables of the Binomial
Distribution indicate that the probability of eight positive values
being obtained by chance is only 0.02; the probability of seven,
0.09). Thus, it is statistically reasonable to assert that annual
oxidant air quality in the Bay Area is sensitive to differences in
weather from year to year, as well as to changes in pollutant
emissions.
A further point of interest is to what extent oxidant air
quality may be related specifically to temperature differences from
year to year, in view of the fact that photochemical smog is generally
associated with sunny, warm weather. Table V-2 shows the results
of rank correlation, based on average maximum temperatures recorded
at the San Francisco station. (Temperatures from the Oakland Airport
also give the same results, since the ranking of years is the same
for the two locations, namely, 2, 1, 3, 5, 4. The rank correlation
between these temperatures and the number of smog-conducive days,
however, is only 0.5.)
The coefficients for these correlations range from + 0.9 to -0.9.
There is, again, a significant tendency toward positive correlations,
although it is not as consistent as for the correlations with the smog-
conducive days. In particular, the average maximum index (i.e., the
average daily high-hour value for the entire year) shows no detectable
tendency to be correlated with these maximum temperatures. These
results suggest that, on the whole, the index of smog-conduciveness
developed by the Bay Area Air Pollution Control District is a sub-
stantially better smog-predicting variable than is temperature alone.
They tend to confirm the utility of the District's approach, in which
a criterion of vertical dilution is added to the temperature criterion
for defining smog-conducive days. According to the District's Infor-
mation Bulletin (loc. cit.). "The addition of the inversion criterion
V-3
-------
TABLE V-2. RANK CORRELATION COEFFICIENTS BETWEEN VARIOUS ANNUAL
OXIDANT INDICES AND ANNUAL AVERAGE MAXIMUM TEMPERA-
TURE IN DOWNTOWN SAN FRANCISCO, FOR YEARS 1970 TO 1974
INDEX STATION
SF RC SJ FR SL OK RI PT LV AVG.
Second High .7 .5 -.9 .6 .6 .7 .7 .9 .3 .5
Average Maximum -.1 .1 -.3 -.2 -.3 -.1 .5 .4 -.3 0
3rd Quarter Average .6 .4 -.5 .7 .3 .7 .4 .7 -.3 .3
Frequency >.08 .2 .9 -.3 -.3 .2 .6 .5 .8 -.3 .3
V-4
-------
eliminates some of the anomalous cases in which heating actually
destroys the inversion and results in low oxidant maxima because
of vertical dilution."
Scrutiny of Table V-2 reveals that two stations showing
most of the negative correlations are San Jose and Livermore.
These stations also yielded anomalous results in the BAAPCD study
(loc. cit.), leading the District to explore the relevance of local
wind speed and direction as variables which might further improve
oxidant prediction. This exploration yielded the conclusion that
the local wind factor could account for a significant part of the
increase in oxidant level observed at San Jose in 1974.
B. GEOGRAPHICAL EFFECTS. BAY AREA AND CENTRAL VALLEY
The geographical variation in oxidant air quality in three
of the air basins discussed in this report are most easily dis-
cussed and interpreted in terms of a single meteorological system,
whose general nature is seen in Figure V-l.
During much of the year, especially during the summer months
when high oxidant levels are common, the prevalent flow of air is
onshore, into the Bay Area through the Golden Gate (the entrance
to San Francisco Bay) and eastward into the Central Valley via
San Pablo Bay, Carquinez Strait, Suisun Bay and the delta of the
Sacramento and San Joaquin Rivers. Within the Bay Area, part of
this flow is diverted northward into the Sonoma and Napa Valleys
and southward into the Santa Clara and Livermore Valleys. In
the Central Valley the main flow is bifurcated, ventilating the
Sacramento Valley to the north and the San Joaquin Valley to the
south.
As in other California Coastal areas, the subsidence inversion
dominates this wind field for a large part of the year, varying
seasonally between about 300 and 1,000 meters elevation. Due to
V-5
-------
O^ARYSVi :_•.:. .. . :
;H
w
• . ' <.
• :'-X
QFRESNOV '..••&*'
&•£'
0 20 40 80 80 100
SCALE IN MILES
FIGURE V-l. NORMAL FLOW PATTERN IN THE
SUMMERTIME MARINE LAYER
(Source: S. J. Savage, 1967)
V-6
-------
solar heating the inversion is sometimes destroyed over the down-
wind, up-valley extremes of these valleys, and wide variations in
the extent of vertical mixing characterize these locations even
when the inversion persists.
The meteorology and climate of the Central Valley are es-
pecially conducive to air pollution. Photochemical smog in the
summer and early fall is enhanced by an almost unbroken succession
of warm, sunny days. In the Bay Area, summer sun is often mod-
erated by the presence of stratus, reducing insolation until
late in the morning. This effect is greatest near the main chan-
nel of flow of the marine air, whereas isolated valleys such as
the Napa, Livermore, and Diablo Valleys tend to be warmer and
sunnier.
This study has related to oxidant air quality as measured
at nine monitoring stations in the Bay Area, one in the Sacramen-
to Valley, and two in the San Joaquin Valley, as described in
Chapter II. Rankings of the twelve stations relative to four
different indices of air quality are shown in Table III-7, from
which it can be seen that the oxidant problems in inland valley
locations are, on the whole, more severe than those in the pop-
ulous industrial areas fronting the Bay.
Dividing the twelve stations into these two categories, we
find the inland-type stations best represented by Fremont, San
Jose, and Livermore in the Bay Area and the Central Valley lo-
cations, Redding, Stockton and Modesto. The remaining stations,
San Francisco, Redwood City, Oakland, San Leandro, Richmond and
Pittsburg, are predominantly industrial, bayside communities.
Using the annual second-high oxidant as the air quality index,
five of the valley locations appear to be among the six most
affected; using any of the other indices, the valley locations
are the six most affected.
V-7
-------
Two stations, Redding and Modesto, rank much lower by the
second-high statistic than by the other three indices. These
stations are both far downwind of the Bay Area; Redding is 300
km. to the north, Modesto about 100 km. to the south. Especially
in Redding, there is little local industry and a sparse popula-
tion. It is striking, therefore, that this area has relatively
high oxidant levels and high frequency of'exceeding the* NAAQS.
In all probability, these levels predominantly result from trans-
port of the polluted marine air along the track indicated in
Figure V-l. (Note that Redding is some 50 km. north of Red
Bluff, the city shown on the map.)
On the other hand, two stations, Fremont and San Leandro,
rank much higher by the second-high statistic than by the other
indices. San Leandro is an industrial area near the Bay, and
Fremont a nearby residential area only about 15 km. farther in-
'land; their high-oxidant episodes probably occur mainly in less
typical air flow situations when the ventilation of the east
Bay Area by the sea breeze is relatively weak.
Stockton and Pittsburg represent intermediate cases. Al-
though fairly well inland, Stockton is only a few kilometers
from the delta area and is well ventilated by the up-valley wind
from the Bay Area. Pittsburg is on the waterfront and in the
marine air stream which later ventilates the valley. Stockton
and Pittsburg are therefore, to an appreciable extent, protected
from both the long-distance transport that delivers oxidants to
Modesto and Redding and from the local stagnation episodes that
occasionally afflict Fremont and San Leandro.
The relatively low rankings of the remaining locations—
San Francisco, Oakland, Richmond and Redwood City—provide un-
mistakable evidence that the maximum effect of oxidant precursors
is not to be found in the area where the contaminants are emitted,
V-8
-------
but at various downwind locations. One principal reason for this
state of affairs is the dual role of nitric oxide as both a pre-
cursor and an inhibitor of the oxidant-forming process. The in-
dustrial and vehicular sources of this precursor furnish a
necessary constituent for oxidant formation, but one which en-
sures that the air will have long since moved to another loca-
tion by the time its oxidant potential becomes manifest.
V-9
-------
REFERENCES
1. S. J. Savage, "Summertime Marine Air Penetration to the City
of Sacramento." M.A. Thesis, San Jose State College, 1967.
V-10
-------
VI. SUMMARY AND CONCLUSIONS
A study of the relation between oxidant air quality and
emissions of oxidant precursors (hydrocarbons and oxides of ni-
trogen) has been made, utilizing data from five or more recent
years of ambient air monitoring. Areas investigated were:
• in California,
1. the San Francisco Bay Area, represented by monitor-
ing sites at San Francisco, Redwood City, San Jose,
Fremont, San Leandro, Oakland, Richmond, Pittsburg,
and Livermore
2. the northern Sacramento Valley Air Basin, represented
by Redding, Shasta County
3. the northern San Joaquin Valley Air Basin, repre-
sented by Stockton, San Joaquin County, and Modes-
to, Stanislaus County
4. the North Central Coast Air Basin, represented by
Salinas, Monterey County
5. the Southeast Desert Air Basin, represented by Lan-
caster, Los Angeles County.
• in Colorado, Denver City and County, part of the Metro-
politan Denver Air Quality Control Region.
• in Pennsylvania, Philadelphia City and County, part of
the Metropolitan Philadelphia Interstate Air Quality
Control Region.
These areas ranged in demographic character from sparsely
populated agricultural and desert communities to densely popu-
lated metropolitan areas. Geographically, they represented the
Atlantic Seaboard, the Rocky Mountain area, and the Pacific
Coast, with attention to coastal and interior locations in Cali-
fornia. Industrial activity ranges from negligible to fairly in-
tensive.
Meteorologically, the California vicinities are subject to
the influence of the Pacific High, which causes a semi-permanent
VI-1
-------
subsidence inversion over most of the state. The Colorado and
Pennsylvania cities reflect continental and seaboard meteorology.
Various air quality indices have been compiled and examined
for evidence of trends. These include:
• 2nd high: the second highest daily high-hour oxi-
dant value observed during a full year of observa-
tions at a single monitoring location.
• Average maximum: the average daily high-hour oxi-
dant value for a year of observations.
0 3rd quarter average maximum: the average daily
high-hour oxidant value for the months of July,
August and September in any one year.
• Frequency >.08: the frequency, in percent of all
days monitored, of days with oxidant values ex-
ceeding the NAAQS level (0.08 ppm) during a full
year of monitoring at a single location.
• Frequency >.16: the frequency, in percent of all
days monitored, of days with oxidant values ex-
ceeding 0.16 ppm during' a full year of monitoring
at a single location.
Most of the trends for individual stations, as evaluated for
the five-year period, were not statistically significant at the
5 percent level. Forty-five pairs of corresponding values of
these two indices (i.e., values for nine stations for five years)
yielded a correlation coefficient of 0.73, highly significant.
This trend was not paralleled by significant decreases in the
other indices for the Denver area.
A substantial degree of independence between the 2nd-high
statistic and the average maximum was indicated by comparisons of
the trends at various monitoring stations. To explore the rela-
tions between the indices, correlation coefficients were computed
for the Bay Area stations. For the entire data base, a highly
VI-2
-------
significant correlation was found. Taking the average maximum
index as the independent variable, the regression coefficient was
3.3 and the standard error of estimate about 20 percent. These
results tend to substantiate the utility of the 2nd-high index
as a measure of oxidant air quality.
Calculation of coefficients of concordance among the four
indices of oxidant air quality (frequency above 0.16 excluded)
showed substantial agreement in the statistical sense (C=0.83),
but also showed that the degree of agreement could be radically
improved (C=0.99) by either excluding the 2nd-high index from
the set tested or by excluding four of the twelve northern Cali-
fornia monitoring stations (namely, Fremont, San Leandro, Modesto
and Redding). The results confirm that excellent agreement be-
tween all indices, for any extensive set of monitoring locations,
is not to be expected.
An analysis of variance of the 2nd-high statistics for the
Bay Area stations showed that the yearly change effect was signi-
ficant at the 5 percent level and that the differences between
stations were significant at the 1 percent level. Error standard
deviation was evaluated as 0.038 ppm. Because these differences
are statistically significant, the effects cannot be accounted
for as accidental variations, but require explanation in terms
of the major variables intrinsic to the determination of air
quality: emissions and weather.
Emissions and emission trends have been reviewed in detail
for each of the selected study locations. For both hydrocarbons
and oxides of nitrogen, the estimated emissions have been par-
titioned into those from stationary sources and those from mobile
sources, and the year-to-year trends examined. In most locations,
the predominant changes in hydrocarbon emissions were in the emis-
VI-3
-------
sions from mobile sources; especially in the rural areas, where
mobile sources constitute the major part of the total. For Cali-
fornia, mobile sources are discussed in terms of a "mobile source
emission index," based on combining motor vehicle traffic in-
formation with appropriate motor vehicle emission factors obtained
from the California Air Resources Board.
For the San Francisco Bay Area, using as the annual oxidant
index the pooled second-high statistics for the nine monitoring
stations, correlations with annual total emissions were investi-
gated. The correlation with total hydrocarbon emissions was posi-
tive (0.58) but not statistically significant at the 5 percent
level. A similar result was obtained (0.60) using the total emis-
sions of both hydrocarbons and oxides of nitrogen. However, there
was high and significant correlation when the emissions of oxides
of nitrogen were introduced as an inverse factor (0.88) or as
a negative contribution (0.87).
These results show that a large fraction of the variability
in oxidant air quality from year-to-year (as determined by the
pooled second-high statistics for the San Francisco Bay Area)can
be accounted for in terms of the effects of varying rates of
emissions, where both hydrocarbons and oxides of nitrogen emissions
are recognized as contributing factors.
In the Denver area, a confused picture on the evaluation of
appropriate emissions trends prevents an adequate statistical test
of the relation of the observed significant decrease of average
maximum oxidant levels to emission totals. Hydrocarbon emissions
have been increasing in Denver County, but decreasing in the Metro-
politan Denver Air Quality Control Region. In the vicinity of the
single monitoring station, motor vehicle traffic increased, while
emission factors were generally decreasing. Plausible ad hoc
explanations for the oxidant trend may exist, but the available
data base is insufficient to confirm them.
VI-4
-------
For the other areas included in the study, a matrix of trend
directions for various emissions indices and air quality indices
showed no general agreements as to trend directions.
Using data on long-term oxidant trends from a special study
carried out by the Bay Area Air Pollution Control District, the
correlation of a weather-stratified oxidant index with annual
emissions was tested, for the thirteen-year period 1962-1974.
When the data for six stations having a thirteen-year record were
pooled, the oxidant trend was significant at the 0.1 percent level.
When these data were paired with hydrocarbon emissions for the Bay
Area for the same period, the correlation coefficient was 0.767,
significant at the 5 percent level and higher than the value (0.58)
found in studying the 1970-1974 period (v.s.). Again, considera-
tion of oxides of nitrogen emissions increased the degree and sig-
nificance of the correlation. When these emissions were included
as an inverse factor, the correlation coefficient was 0.92, sig-
nificant at the 0.1 percent level. When the oxides of nitrogen
emissions were taken as a negative contribution, the correlation
coefficient was 0.89, also significant at the 0.1 percent level.
These results, therefore, reinforce those obtained in studying
the five-year data, discussed above.
Rank correlation studies showed that, in the Bay Area, the
various annual indices of oxidant air quality significantly re-
flected changes in annual weather conditions as measured by the
number of smog-conducive days per year (by Bay Area Air Pollution
Control District criteria). The oxidant indices also reflected
annual average maximum temperatures, but not as strongly or con-
sistently as they reflected the frequency of smog-conducive
weather.
Reviewing the relation of oxidant air quality to geographical
factors within the single meteorological system comprising the
VI-5
-------
San Francisco Bay Area and the Central Valley, it is seen that
oxidant problems are most severe at inland valley locations and
least severe in populous, industrial Bayside communities. These
conditions seem to be related to two principal factors: the
superior ventilation of communities along the main waterfronts;
and the inhibitory effect of the oxides of nitrogen emitted by
industry and motor vehicle traffic. Both of these protective
circumstances tend to fade as the marine air moves downwind to
the inland valley locations.
To recapitulate, the main conclusions resulting from this
project are:
1. Long-term trends in oxidant air quality in the San
Francisco Bay Areas are significantly related to rends in hydro-
carbon emissions.
2. Oxidant concentrations in the San Francisco Bay Area
are inversely affected by oxides of nitrogen emissions; thus, re-
duction of these emissions tends to increase measured oxidant
levels, on the whole.
3. In the San Francisco Bay Area, oxidant problems tend
to be more severe in inland valley locations than in industrial,
Bayside locations.
4. The oxidant air quality of communities in the Central
Valley of Caifornia is appreciably affected by the flow of pol-
luted marine air from the adjacent San Francisco Bay Area.
5. Monitoring data for a five-year period is not, in
general, sufficient to establish statistically significant oxi-
dant trends for individual monitoring stations.
6. A significant reduction of oxidant levels occurred in
Denver during the period 1968-1973. This cannot be attributed
with confidence to observed trends in emissions.
7. The second highest daily high-hour oxidant value for
a single year at a single station is a statistically useful mea-
sure of oxidant air quality as shown by analysis of variance.
VI-6
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-76-034
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Detection and Interpretation of Trends in Oxidant
Air Quality
5. REPORT DATE
October 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Lowell G. Wayne, Katherine W. Wilson,
Clarence L. Boyd
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Pacific Environmental Services, Inc.
1930 Fourteenth St.
Santa Monica, CA 90404
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-1890, TO 1
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning & Standards, MDAD
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Trends in ambient levels of oxidants over periods of five years or more are
reviewed for seven locations in the United States. Statistically significant
downward trends were identified in the San Francisco Bay Area and in Denver.
Trends in emissions of organic and NOx precursors over comparable periods
are also reviewed. Only in San Francisco were the data sufficient to note a
statistically significant relationship between precursor emission trends and
the trend in ambient oxidant. Downward oxidant trends are most pronounced in
the western part of the Bay area. Indications are that a five year period may
not be sufficient to identify statistically significant trends in many areas.
Downward trends in the Bay Area are much more pronounced when 13 years of
records are examined.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Photochemical Air Pollutants
Oxidants
Precursors
Organic Pollutants
Nitrogen Oxides
Trends
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Trends in
Oxidant
Trends in
Precursor
Ambient
Oxidant
Emissions
Civil Engineering
(1302)
Air Pollution
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAC.
129
20. SECURITY CLAS.S (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |