EPA 901/9-78-001
Final Report September 1978
ATMOSPHERIC PROCESSES AFFECTING OZONE
CONCENTRATIONS IN NORTHERN NEW ENGLAND
By:
F. L. Ludwig
Rosemary Maughan
Prepared For:
Environmental Protection Agency
Region 1
J.F. Kennedy Federal Office Building
Boston, Massachusetts
Contract No. 68-02-2548 (Requisition No. F74673)
SRI International Project 6908
-------
EPA 901/9-78-001
Final Report September 1978
ATMOSPHERIC PROCESSES AFFECTING OZONE
CONCENTRATIONS IN NORTHERN NEW ENGLAND
By:
F. L. Ludwig
Rosemary Maughan
Prepared For:
Environmental Protection Agency
Region 1
J.F. Kennedy Federal Office Building
Boston, Massachusetts
Contract No. 68-02-2548 (Requisition No. F74673)
SRI International Project 6908
Approved By:
R.T.H. Collis, Director
Atmospheric Sciences Laboratory
Earle Jones, Executive Director
Advanced Developments Division
-------
ABSTRACT
Readily available meteorological and air quality data were analyzed to determine the extent to which
ozone concentrations in the Northern New England states of Maine, New Hampshire, and Vermont are
influenced by causes external to those states. It is concluded on the basis of air trajectory and wind
analysis that ozone generated from precursor emissions to the southwest or west is transported into the
southern parts of Vermont, New Hampshire, and Maine. In the northern parts of New Hampshire and
Vermont, violations of the ozone standard are more frequently associated with air that has come from the
areas around Lakes Erie and Ontario. Although the Northern New England states are influenced by
ozone transported from elsewhere, some control measures might still be required within the area even if
the external sources were controlled and concentrations entering the region were reduced to levels near
the tropospheric background.
iii
-------
CONTENTS
Abstract iii
List of Illustrations vii
List of Tables ix
Acknowledgments xi
I. INTRODUCTION 1
II. DATA 3
A. General 3
B. Regular Surface Observations 3
1. Pollution Data 3
2. Meteorological Data 3
C. Special Studies 8
III. BACKGROUND 11
A. The Nature of the Problem 11
B. Possible Origins of Oxidant Concentrations in Excess of the Standard 15
1. Natural Sources and Sinks of Tropospheric Ozone 16
a. Tropospheric Synthesis 16
b. Transport from the Stratosphere 16
2. Anthropogenic Sources of Tropospheric Ozone 20
3. Combined Natural and Anthropogenic Effects 22
IV. ANALYSIS OF THE ORIGINS OF THE AIR AS RELATED*TO OZONE CONCENTRATIONS. 25
A. Governing Factors 25
B. Formation and Transport in the Lower Atmosphere 25
1. Winds and General Weather Patterns Associated with Oxidant Concentrations
Above the NAAQS 25
a. Winds and Trajectories 25
b. Weather Patterns 33
2. Effects of Weather Fronts 35
a. 18-20 April 1976 35
b. 11-12 July 1976 38
c. 17-18 June 1977 44
3. Effects of Stagnation 48
a. 20-22 April 1977 48
b. 20-24 May 1977 50
4. Special Meteorological Situations 50
a. Nighttime Effects 50
i. 28-29 June 1977 53
ii. 15-16 June 1976 58
b. Sea Breeze Effects 62
c. Diurnal Patterns of NAAQS Violations 65
C. Weekday Versus Weekend Ozone Concentrations 67
1. General 67
2. Observations 67
V. SUMMARY AND CONCLUSIONS....r 77
A. Local Versus Imported Oxidants 77
B. Implications for Control Strategies 78
1. General 78
2. The Extent of Urban Influences on Oxides of Nitrogen Concentration 78
3. The Relative Importance of Transported and Locally Generated Ozone 80
4. An Example of How the EKMA Might Be Used to Devise Control Strategies in Northern
New England 84
C. Final Remarks and Recommendations 85
APPENDICES:
A. MONITORING SITE CHARACTERISTICS 87
B. SUMMARY OF HOURS WHEN THE NATIONAL AMBIENT AIR QUALITY
STANDARD FOR OZONE WAS VIOLATED 91
REFERENCES 103
-------
ILLUSTRATIONS
1 Locations of SAROAD Ozone Monitoring Sites in New England 4
2 Example of Surface Weather Map 6
3 Example of an Upper Air Chart (850 mb) 7
4 Example of National Weather Service Hourly Weather Observation Records ... 9
5 Percentage of Hours When Ozone Concentrations Equalled or Exceeded
NAAQS 12
6 Diurnal Variation of Ozone Concentrations of 80 ppb or Greater for Selected
Stations 14
7 Schematic Representation of Ozone Variations at the Surface and in the Free
Troposphere 17
8 Idealized Ozone Variations at Remote Locations , 19
9 Schematic Diagram of the Transport of Precursors, Ozone Production, and
Ozone Destruction Downwind of a City 21
10 Diurnal Distribution of Daily 1-hr 0, Maximum Occurrence at White Face
Mountain, N.Y. for 1974, 1975, and 1976 24
11 Number of Hours when Ozone Concentrations Violated the NAAQS for
Oxidants as a Function of 850-mb Wind Direction--!976 27
12 Number of Hours when Ozone Concentrations Violated the NAAQS for
Oxidants as a Function of 850-mb Wind Direction-1977 28
2 1
13 Counties with Average Annual NO Emissions Greater than 75 t m" yr" . . . . 29
14 Locations of Air that Arrived at Manchester in 12 Hours with an Ozone
Concentration of 70 ppb or Greater Upon Arrival 31
15 Locations of Air that Arrived at Burlington in 12 Hours with an Ozone Con-
centration of 70 ppb or Greater Upon Arrival 32
16 Weather Maps for 18-20 April 1976, 0800 EOT 36
17 Frontal Positions in Northern New England, 18-20 April 1976 37
18 Time Histories of Ozone Concentrations in Northern New England, 18-20. . . 39
April 1976
19 Weather Maps for 11-12 July 1976 40
20 Frontal Positions in Northern New England, 11 July 1976 41
21 Time Histories of Winds at Selected Northern New England Sites, 11-12 July
1976 42
22 Time Histories of Ozone Concentrations in Northern New England, 11-12
July 1976 43
-------
23 Frontal Positions in Northern New England, 17-18 June 1977 45
24 Time Histories of Winds at Selected Northern New England Sites, 17-18
June 1977 46
25 Time Histories of Ozone Concentrations in Northern New England, 11-18
June 1977 47
26 Time Histories of Ozone Concentrations in Northern New England, 20-22
April 1977 49
27 Time Histories of Winds at Burlington, Vermont and Concord, New
Hampshire, 20-22 April 1977 51
28 Time Histories of Ozone Concentrations at Four Northern New England
Sites, 20-23 May 1977 52
29 Time Histories of Ozone Concentration in Northern New England, 28-29
June 1977 54
30 Weather Maps for 28-29 June 1977 55
31 Frontal Positions in Northern New England, 28-29 June 1977 56
32 Time Histories of Winds, 28-29 June 1977 57
33 Time Histories of Ozone Concentrations at Northern New England Sites,
15-16 June 1976 .59
34 Weather Maps for 15-17 June 1976 60
35 Time Histories of Winds, 15-16 June 1976 61
36 Schematic Depiction of Summer Sea Breeze Circulation 63
37 Ozone Concentrations at an Altitude of 300m, 24 July 1975 64
38 Relative Frequencies of NAAQS Violations at Different Times of Day for
Selected Coastal and Inland Sites 66
39 Diurnal Variations of Average Ozone Concentrations for Each Day of the
Week at Manchester, N.H 68
40 Diurnal Variations of Average Ozone Concentrations for Each Day of the
Week at Nashua, N.H 69
41 Diurnal Variations of Average Ozone Concentrations for Each Day of the
Week at Burlington, Vermont 70
42 Sensitivity of Maximum Afternoon Concentrations to Morning Precursor
Levels Measured Upwind 79
43 Estimated Radius at Which NO Concentrations Fall Below 7 ppb as a Func-
tion of Metropolitan Population .81
44 Areas Appropriate for Hydrocarbon Emission Controls According to OAQPS
(1978) 82
viii
-------
TABLES
Months for Which Ozone Data Were Available for Sites in Ver-
mont, Maine, and New Hampshire, During 1976 and 1977. . . .
2 Frequency of Ozone Concentrations of 80 ppb or Greater for Vary-
ing Periods of Time 13
3 Winds Reported on Morning Weather Map in Areas Where Peak-
Hour Ozone Exceeded 80 ppb During the Day 26
4 Time Periods for Which Air Trajectories Were Calculated 30
5 Meteorological Features Associated with Violations of the NAAQS
for Ozone During 1974 in the Eastern United States 33
6 Meteorological Features Associated With Observed Ozone Concen-
trations of 80 ppb or Greater in the Three Northern New England
States 34
7 Daily Mean Ozone Concentrations—Weekdays and Weekends ... .71
8 Average Daily Maximum Ozone Concentrations—Weekdays and
Weekends 73
9 Qualitative Impact of Various Factors on the Additivity of Tran-
sported Urban Ozone to Maximum Ozone Concentrations in Urban
Areas .' 83
A-l Ranking of Local Urban Influence at Monitoring Sites 89
ix
-------
ACKNOWLEDGMENTS
Barbara Ikalainen, the EPA Project Officer, and Norman Beloin, also of EPA, have given indispens-
able assistance in providing data, background information, and guidance for which we are very grateful.
We are indebted to Mr. David Dixon, Mrs. Norma Gordon and the staff of the Maine Bureau of Air
Quality for the many suggestions that they made concerning this report. Mr. Dale Coventry of EPA pro-
vided valuable assistance. In the preparation of this report we have received much help from Linda
Jones, Josette Louvigny, Joyce Kealoha, and Grace Tsai of SRI International.
xi
-------
I INTRODUCTION
Observations in the Northern New England States of Maine, New Hampshire, and Ver-
mont have shown that the National Ambient Air Quality Standard (NAAQS) for ozone of 80
ppb is violated on occasion. National policy requires states and local agencies to take corrective
action to prevent such violations of the standard in the future. The intelligent planning of
corrective measures requires that the causes of the violations be understood. It is this fact that
has motivated the research described in this report. We have attempted to identify, under-
stand, and quantify, where possible, the causes of the oxidant problems in Northern New Eng-
land. Understanding the oxidant problem is an intermediate goal; the ultimate goal has been to
determine if an oxidant control strategy is actually required for the Northern New England
states.
At first it seems paradoxical to question the need for a strategy to control the problem
when the existence of the problem has already been admitted. However, in the case of atmos-
pheric oxidants the causes may not be fully under the control of persons or organizations in the
areas where the problems are known to occur. If the ingredients of the problem are generated
locally, then the problem is controllable locally, but if they are brought in from elsewhere or if
the ozone is the result of natural causes, then control strategies imposed only in the problem
areas will be futile.
There have been a few constraints to this study. The most important of these is that only
readily available data have been used. In general, this has not been a particularly severe prob-
lem, but there are instances where we have used the results of studies from other areas as basis
for estimating Northern New England effects by analogy. Otherwise, the approach to the
research has been quite straightforward. The severity of the problem was first defined by
analysis of existing monitoring data. The possible causes, both natural and anthropogenic, were
next examined. It was then possible to use the existing data and past studies to try to find evi-
dence for the different processes thought to be operating to produce the ozone problem in
Northern New England.
In general, the evidence that was sought for the possible origins of the Northern New
England ozone was related to natural versus anthropogenic causes, and to transport into the
area from outside versus local origin. The search for the evidence has led to the examination
of some specific phenomena—e.g., the effects of sea breeze circulations and the differences
between weekdays and weekends, but in each case the central problem has been the origin of
the ozone, especially remote versus local. Finally, the methods used in this study have been
chosen to match available data and the questions being addressed.
This report is organized along the same lines as the research. It begins with the statement
of the problem. This is followed by discussion of the evidence that is available, and that dis-
cussion is followed in turn by analysis of the evidence. The report concludes with a section
that summarizes the known facts regarding the origins of high concentrations of ozone in the
Northern New England states. Most importantly, this final section relates those known facts to
possible corrective measures that might be taken to avoid future violations of the NAAQS for
oxidant in the Northern New England area.
-------
II DATA
A. General
The work reported here has been confined to the analysis and interpretation of readily
available data. No field measurement programs were mounted to collect new data. While the
use of the data collected for other purposes does restrict the study somewhat, it is still possible
to extract considerable information of use in this study.
Data obtained by routine monitoring of air quality at several sites in the Northern New
England states have provided a valuable source of information. Other sources of routinely col-
lected data are the numerous weather stations. The meteorological data have proven very valu-
able to the interpretation of the air quality data. Finally, some special studies have been con-
ducted in New England in the past. The reports from those studies have also provided infor-
mation valuable to the project. In the following sections, these data sources are discussed in
more detail.
B. Regular Surface Observations
1. Pollution Data
Hourly ozone concentrations for a number of sites in New England were obtained
from the EPA SAROAD data base. The data for sites in Vermont, Maine, and New Hampshire
are of principal interest in this report. Figure 1 shows the location of these sites. The periods
for which ozone data were available at each site in the three Northern New England states dur-
ing the period January 1976 to June 1977 are listed in Table 1. Only Nashua and Manchester
in New Hampshire have complete data sets; Burlington and Berlin each have only one month
missing. The disparity among the data periods of the other sites has made spatial variation
analysis difficult.
The EPA reviewed the sites and the monitoring of ozone, NO~ and HC in New
England between April and August 1977. Their comments on the sites in Vermont, Maine,
and New Hampshire are included in Appendix A of this report. While all sites are described as
adequate, and most are thought likely to experience some ozone depression due to urban loca-
tions, it should be noted that Manchester and Nashua in New Hampshire would "experience
significant ozone depression" due to downtown locations. This, of course, has implications in
the assessment of violations of the NAAQS for oxidant; the severity of the problem in such
areas may well be greater than the measured concentrations would indicate.
2. Meteorological Data
Extensive use has been made of routinely collected National Weather Service data.
Surface weather maps at 3-hour intervals and twice-daily upper ah- maps were referenced from
microfilm data sets acquired from the National Climatic Center. The Daily Weather Map
Series, produced by the National Oceanic and Atmospheric Administration (NOAA), showing
surface station information and pressure patterns for the whole United States at 0700 EST
(0800 EDT), were also used. Examples of these data sources are shown in Figures 2 and 3.
More detailed meteorological data were taken from copies of National Weather Service WBAN
form 10A, listing hourly data for sky cover, visibility, temperature, humidity, wind speed, and
direction. Three-hourly data listings were also used. Data from Burlington, Vermont; Concord,
-------
BURLINGTON
WHITE RIVER
JUNCTION/ GRAFTON
COUNTY
DEERFIELD
{^PORTSMOUTH
MANCHESTER
f i i
NASHUA.
MASSACHUSETTS
100
150
200
250 MilM
0 60 100 150 200 250 300 350 Kilometer*
FIGURE 1 LOCATION OF SAROAD OZONE MONITORING SITES IN NEW ENGLAND
4
-------
TABLE 1
Months for Which Ozone Data Were Available
For Sites in Vermont, Maine, and New Hampshire During 1976 and 1977
State
Maine
Vermont
New Hampshire
Site
Portland
Burlington
White River Junction
Berlin
Deerfield
Grafton Country
Keene
Manchester
Nashua
Portsmouth
1976
JFMAMJJASOND
XXXXXXXXX
XXXXXXX XXXX
XXXXXXXXXXXX
X X
X X X X X
*
XXXXXXXXXXXX
XXXXXXXXXXXX
1977
J F M A M J i
!
X X X X X X
X X X X X X
XXX XX
XXX
X X X X X X
X X X X X X
XXXX
Ul
-------
TUESDAY, JUNE 15, 1976
; I \U A1HI K MAP .
' AND STATICS ACA1IIIR >
[ AT 7 00 A M C <> 1
FIGURE 2 EXAMPLE OF SURFACE WEATHER MAP
-------
850MB ANALYSIS HEIGHTS/TEMPERATURE
FIGURE 3 EXAMPLE OF UPPER AIR CHART (850 mb)
-------
New Hampshire; Portland, Maine; Boston, Massachusetts; and Albany, New York were used in
the study of selected high ozone periods. An example of National Weather Service Records is
shown in Figure 4.
C. Special Studies
Very few special studies of photochemical pollutants were identified for the Northern New
England States. One such study involved the measurement of hydrocarbons, oxides of nitro-
gen, ozone and meteorological parameters at three locations in the vicinity of Portland, Maine
during the summer of 1974 (Londergan and Polgar, 1975). The report of this study also
included a limited discussion of ozone measurements made aloft from an aircraft.
Other special studies that made use of aircraft were centered over the Southern New Eng-
land States (see, e.g. Ludwig and Shelar, 1977), and contributed little to the understanding of
conditions in Northern New England. An exception was a report by Spicer, Gemma, and Stick-
sel (1977) that contained some analyses extending to the offshore areas of Maine and New
Hampshire. Finally, most analyses of existing data (e.g., Wishinki, 1977) were generally based
on fewer data than the study reported here.
8
-------
SURFACE WEATHER OBSERVATIONS
£ tty
TpjN^MASS.
SKY ANO CCILIMC
ANO
OBSTRUCTIONS
TO VISION
ALTrx
ET(M
(IT.
TlMC
REMARKS AHO SUPPttMEMTAi. CODED DATA
\n
51 XL
O/ j 10 \
7
12.
'££>.
/!&/->
»2
SLfL
5|
Mr
-amove. i>
Hi
rtjc-j f-»-\
*" ~f^ """\ «/ 'I
E
(52
\w, SS1-*1
ov
g.m ti3o
•m
Ho XV
.K \
Cp
EiS
ocs
offS!
\S
.nts
'
-------
Ill BACKGROUND
A. The Nature of the Problem
The NAAQS for oxidant states that an hourly average of 80 ppb should not be exceeded
more than once in any year. With the exception of Berlin, New Hampshire, no single site
measuring oxidant in Maine, New Hampshire, or Vermont came close to meeting that standard
in either 1976 or 1977. The greatest numbers of times when the concentrations equaled or
exceeded the NAAQS were experienced at Nashua in both years (168 hours in 1976, and 190
hours in 1977) while both Portsmouth and Manchester showed well over 100 such hours in
1977 (150 and 112, hours respectively). Even Burlington and White River Junction in Ver-
mont, seemingly remote from possible anthropogenic sources of oxidant from major urban
areas to the south and west, both experienced nearly 100 hours of 80 ppb or more in the first
six months of 1977.
Figure 5 shows that the frequencies of high concentrations exhibited marked seasonal
cycles, tending to be above the standard for the greatest number of hours in May, June, July,
and August. There were no violations between October and February. Differences in meteorol-
ogy from year to year can cause substantial differences in the frequency of standards violations,
especially in the months at the beginning and end of the photochemical "season". For example,
one major difference between the two years studied was the greater number of violations in
March, April and May 1977 than during the same months in 1976. This, together with the
large number of hours at all sites in May 1977 when the standard was equaled or violated,
meant that there were a greater number of such cases in the three states during the first six
months of 1977 that were available for analysis than there were during the whole of the year
1976.
The duration of periods when concentrations were 80 ppb or more is shown hi Table 2.
At most sites, 50% of these periods persisted for up to 4 hours, and 10% up to 6 hours. There
were two periods, 17-18 May 1977 and 28-29 June 1977 that persisted for 16 to 22 hours. The
most prolonged period of high ozone concentrations occurred at Nashua, on 26-27 May 1977,
where levels equaled or exceeded 80 ppb for 25 hours without interuption. If one ignores the
single hour recording 75 ppb, linking two periods, the full extent was 37 hours.
The observed seasonal patterns confirm what is known of the seasonal variations in both
natural and anthropogenic sources of ozone. Although large amounts of ozone are produced in
the stratosphere, the contribution to ground level ozone concentrations is generally small (see
Section III-B-1-b). However, there are occasions during the late winter and spring when much
higher concentraions of stratospheric ozone can be mixed well down in the troposphere. This,
together with the start of photochemical activity and subsequent anthropogenic production of
ozone, mean that the spring months are likely to mark the onset of relatively high levels of oxi-
dant, and of consequent violations of the standard. As the year progresses, violations are more
likely to reflect photochemical activity alone. As this ceases in the fall, oxidant levels,
reflecting the lower autumn background concentrations and decreasing photochemical activity,
are less likely to violate the standard.
The diurnal variations in the frequency of occurrence of ozone concentrations of 80 ppb
or more that are displayed in Figure 6 show a general pattern that is clear and common to all
sites: A maximum number of violations occurred during daylight hours, peaking in the main
between 1400 and 1600 hours EDT, but significant frequencies were observed at some sites,
e.g. Portland, into early evening. There were also secondary peaks occurring during night
11
-------
0
10
5
in
1 n
i J
i
£
3 0
g 20
15
10
0
PORTLAND I 3|
18 1 I 12
• I • 1 • l£lBlllllllll i«i«i«l i i .
1 1 ' * 1 f * iii | | | | j |
PORTSMOUTH
Ul
i i i i i i i i i 1 J i i i i i
MANCHESTER 5
|l7
lie 11 V
i i iii ililil t J,l,
"NASHUA
shown ft the top of each bar
— • Denotes "no data"
~ 62
45
I 36
-
|8 „ I
lllll 1 1 1 1 lllll
I
1
36
LiM.
3
30
|
ai __
-
20
s
0
10
5
0
8 m
s.
i
> _
K 5
O
0
in
5
o
O
i 1 1 I 1 1 1 1 i 1 1 1 1 ! 1 1 1
DEER FIELD
a
| | 1 | j i i i i i i | i i i i i
KEENE
38
1 1 1 1151 1 i ' i 1 1 1 1 1 1 1 1
118 GRAFTON COUNTY
WHITE RIVER JUNCTION 64
27 1 —
||
1
BURLINGTON
63
39 1 30
1 f
7 1 8 9 ill
I I 1 lllllll-lli lllll , lllll
| | | | | ! 1 1 1 1 1 , -,— | 1 — ,
BERLIN
i i i i i i i i i i i i i i i • il I1
J FMAMJ JASONDJ FMAMJ
1976 1977
J FMAMJ JASONDJFMAMJ
1976 1977
FIGURE 5 PERCENTAGE OF HOURS WHEN OZONE CONCENTRATIONS EQUALLED
OR EXCEEDED NAAQS
12
-------
Table 2
FREQUENCY OF OZONE CONCENTRATIONS OF 80 ppb
OR GREATER FOR VARYING PERIODS OF TIME
(Number of Cases)
.Consecutive
iurs
Station^XDuration
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Berlin
Deerfield
Graf ton Co.
Keene
Burlington
White River Junction
Portland
Portsmouth
Manchester
Nashua
4311
864431111 1 1 1 1 1 1
666442221 1 1 1 1 1 1 1
33 26 22 17 15 10 7 7 5 5 3 3 3 2 1 1 1 1 1 1
21 14 13 12 10 7 3 3 2 1 1 1 1 ,.
25 18 14 9 8 7 4 4 3
30 24 19 17 14 12 9 7 7 4 4 2 2 1 1 1
43 32 24 18 10 8633 2 2 2 2 1 1 1 1 1 1
70 56 40 32 27 23 22 20 14 14 9 5 4 3 3 3 2 2 2 2 2 1 1 1
-------
BURLINGTON
\
MANCHESTER
NASHUA
— percent
LU
0
HI
OC
OC
D
U.
O
z
LU
2
8
™
12
8
4
0
: /"
>•
~~'~^.^ ^-x
PORTLAND
/
: /
/
— ' i \ f ' i
X\
"\
****** *i*_^
• ^
\ :
\ -
, ,
UJ
OC
PORTSMOUTH
l£.
8
4
0
: /
'"\_.^
v\
WHITE RIVER JUNCTION
12
8
4
n
— ^\
: 1 '"V :
\ / \ :
1 *\ 1 1 S 1 1 1 1
4 8 12 16
EASTERN DAYLIGHT TIME - hour
20
23
FIGURE 6 DIURNAL VARIATION OF OZONE CONCENTRATIONS OF 80 ppb OR GREATER
FOR SELECTED STATIONS
14
-------
hours--around 2200 or 2300 EOT for Portsmouth, Nashua, and Manchester, and 0100 to 0300
hours for White River Junction, Portland, and most pronounced of all, Burlington. The time
least likely to experience violations was between 0400 and 0800.
As with seasonal variations, the diurnal patterns of high ozone concentration are to be
expected from a consideration of diurnal cycles of the precursors to ozone formation, especially
nitric oxide (NO). Emissions of NO and HC during early morning hours will scavenge oxidant
present during this time (oxidant probably of natural origin, or left from the previous day).
With little photochemical activity at this time, concentrations will be low, and violations excep-
tional. Photochemical activity will be most intense around noon, and the chance of violations
greatest during early afternoon hours. Concentrations will fall as photochemical production of
ozone decreases, and finally ceases after sunset. It is often the case that ozone produced in the
urban plume during the day becomes isolated in or above a stable layer formed during night
hours. If vertical mixing occurs at night, this ozone can be mixed to ground level. Advection
of an urban plume would mean that nighttime violations can occur well downwind of the
source. It may be that the difference in nightime patterns of violations for Nashua and Bur-
lington reflect this point.
Thus, although daytime violations of the standard may Veil occur as the result of normal
diurnal cycles in precursors and meteorological variables, nighttime violations perhaps more
often reflect particular synoptic situations that can cause ozone aloft to be mixed down to
ground level. Examination of the weather situation associated with days when ozone levels
exceeded the standard shows that 50% of these occasions occurred either with a warm air mass
near a cold front, or in the warm sector of a frontal wave (see Section IV-B-l-b, Table 5).
B. Possible Origins of Oxidant Concentrations in Excess of the Standard
The preceding section has shown that violations of the NAAQS occur in all parts of
Northern New England, including those that are well away from any highly urbanized area.
When violations occur near urban areas, with large emissions of known precursors to ozone,
then the natural assumption has been to link the observed high ozone concentrations causally
with the nearby emitters. This assumption underlies the philosophy behind the NAAQS. It has
been tacitly assumed that air quality could be improved by locally instituted abatement pro-
grams. The originally published EPA procedures for determining the necessary precursor reduc-
tions (Federal Register, 1971) applied to local areas where measureable oxidant problems exist.
At the time that the procedures were devised, it was believed that ozone was only a local urban
problem and that diffusive and destructive processes reduced the oxidant concentrations in rural
areas to insignificant levels. However, by the early 1970s it had been observed that high oxi-
dant concentrations were quite widespread within rural areas. The situation described in the
preceding section is not unusual.
The sources of the high concentrations of ozone outside urban areas are subject to some
controversy, especially with regard to their relative magnitude. The importance of the sources
to the problem of formulating control strategies is obvious. If the major sources are anthropo-
genic, then a control strategy can presumably be formulated, but it is absolutely necessary that
the controls be applied in the proper place. If the observed ozone concentrations are the result
of transport over long distances, then the controls must be applied at suitable places, possibly
far upwind. If major sources are natural, then control measures are unlikely to be of any use
and the considerable cost of their implementation should be avoided.
The several possible processes that might cause high non-urban ozone concentrations are
15
-------
discussed in the following sections. The major natural possibilities are transport from the stra-
tosphere and photochemical production from natural precursors within the troposphere. The
two basic anthropogenic possibilities are generation from locally emitted precursors, or transport
over long distances. Much of the discussion of natural processes that follows has been
extracted from a report by Singh et al. (1977). The discussion is intended to provide some
background for the understanding and interpretation of the available data.
1. Natural Sources and Sinks of Tropospheric Ozone
a. Tropospheric Synthesis
Went (1960) suggested that tropospheric ozone might be synthesized photo-
chemically from natural terpenes and natural NC^. Ripperton et al. (1971) tested this
hypothesis under controlled conditions and confirmed that terpene and NO can result in ozone
formation processes similar to those in polluted atmospheres. Although terpenoid compounds
and NO2 have been measured at relatively remote locations, their involvement in the tropos-
pheric balance of ozone is uncertain. This uncertainty arises from inadequate data bases, the
inhomogeneous distribution of trees, the extreme reactivity of the terpenes, the the temporal
and seasonal variations of natural emissions, and the inability to measure NO accurately at less
than part-per-billion concentrations.
Crutzen (1971) hypothesized that methane oxidation chains should also lead to
the net production of ozone in the troposphere. This is an important proposition because
methane is ubiquitous and occurs at fairly high concentrations, about 1.4 ppm. There is no
consensus on the effectiveness of methane oxidation chains in producing ozone. In fact, they
can either produce or destroy ozone, depending on the chosen NO levels (see, for example,
Chameides and Stedman, 1976; Fishman and Crutzen, 1976; and Wemstock and Chang, 1976).
Although natural reactive hydrocarbons (e.g., terpenes) and less reactive hydrocarbons (e.g.,
methane) are widespread in the atmosphere, their relation to the production of ozone appears
to be controlled critically by the availabilty of oxides of nitrogen.
b. Transport from the Stratosphere
Large amounts of ozone are known to be produced in the stratosphere. Some
of this stratospheric ozone gets transferred to the troposphere by various meteorological
processes. There are seasonal variations in the rate at which ozone is transferred to the tropo-
sphere. The greatest rates occur in the late winter and spring. It appears that background
ozone concentrations in the lower troposphere tend to have a phase lag of one or two months
behind the injection cycle from the stratosphere to the troposphere. The major sink for the tro-
pospheric ozone is the destruction that takes place at the surface. There is some uncertainty
about the amount of ozone in the troposphere that can be attributed to stratospheric sources.
Reiter (1977) provides an estimate of 10 to 15 ppb as the average contribution of stratospheric
ozone to the background at ground level.- Singh et al. (1978) and Mohnen (1977) have
estimated the yearly mean tropospheric background ozone concentration to be about 30 ppb,
and that nearly all of this can be attributed to a stratospheric source. In the springtime, concen-
trations are likely to be higher, than the annual average value, and in the fall, lower. In any
event, there appears to be a natural background of ozone in the troposphere at a level of a few
tens of parts per billion. This represents an appreciable fraction of the NAAQS for oxidant.
It is worth examining how this natural ozone behaves, because it is quite easy
to mistake the natural ozone for ozone of anthropogenic origin. Singh et al. (1977) have pro-
vided an idealized picture of the behavior of natural ozone in remote locations. Figure 7 is a
16
-------
I 3
I
UJ
T.
r T
FREE TROPOSPHERE
TOP OF THE AFTERNOON MIXED LAYER
OZONE
AFTERNOON VA -™H°°
03 WITHOUT M 23,™™r
POLLUTION /J
O3 DIURNAL PROFILE
150
100
g
n
O
FREE TROPOSPHERE
50
25
50
OZONE (ppb)
75 0 5 10 15 20
LOCAL TIME (hrs)
25
FIGURE 7 SCHEMATIC REPRESENTATION OF OZONE VARIATIONS AT THE SURFACE
AND IN THE FREE TROPOSPHERE
-------
schematic representation based on a figure from their report. It shows a large reservoir of ozone
aloft. That reservoir is relatively unaffected by daily short-term changes. Above the mixed
layer, there is little diurnal variation of ozone concentration. However, as shown in the right-
hand side of the figure, there is considerable variation of the concentration at the surface dur-
ing the day. This diurnal variation of surface concentration occurs because the ozone that is
within, or below, the nocturnal inversion is destroyed and is not replaced by ozone from aloft.
As the sun rises and the warming of the ground progresses, there will be convective mixing
that brings ozone from the reservoir aloft down to ground level to replace that which is des-
troyed. Hence, the ground level concentrations will increase during the times of day when there
is mixing. Although a diurnal profile at the surface such as that shown in Figure 7 is much the
same as that produced by photochemical processes, it can arise solely from the changes in diur-
nal mixing. Figure 7 also shows how the vertical profile of ozone might look in the afternoon if
the natural ozone were supplemented by ozone produced from nearby emissions through photo-
chemical processes.
Singh et al. (1978), have also provided a picture of the annual variations in the
natural tropospheric ozone burden. These are shown schematically in Figure 8. At- very
remote sites, unaffected by anthropogenic emissions, the ozone concentrations reach their max-
imum in the early spring. In general, the natural ozone falls somewhere in the shaded area
marked A in the figure. Natural concentrations reach their minimum in the late fall or early
winter. The decline of ozone concentrations in these remote locations is in part due to the
decrease in stratospheric injection into the troposphere; it is also possible that photochemical
processes destroy the natural ozone when no NO is present. If oxides of nitrogen are present,
either from natural or anthropogenic sources, then the situation is quite different and photo-
chemicl reactions will cause a net increase in ozone. We will return the discussion of these
effects in the next section.
To this point, the stratospheric contribution has been discussed only in terms
of averages. An important question is whether stratospheric air, rich in ozone, ever reaches
ground level before ozone concentrations have been diluted below the NAAQS. Danielsen
(1964) has proposed a mechanism involving the folding of the tropopause that brings relatively
undiluted stratospheric air deep into the troposphere, perhaps down to levels of 3000 m or so.
Danielsen and Mohnen (1976) have presented data documenting such events, but the ozone
concentrations are generally diluted before reaching altitudes near sea level. This does not
mean that stratospheric ozone never causes violations of the NAAQS standards for oxidants at
ground level. Reiter (1977) has attributed one instance where concentrations of nearly 200 ppb
were observed at the 3000 m peak, Zugspitze, in Germany to an intrusion of stratospheric air.
Singh et al. (1977) present another example from Mauna Loa in Hawaii when concentrations of
nearly 100 ppb were observed.
Both the above examples were observed at rather high altitudes, but Lamb
(1976) has conducted a detailed analysis of an incident near sea level in Santa Rosa, California.
In this incident hour-averaged ozone concentrations of about 220 ppb were observed during the
early morning hours of 19 November 1972. Lamb's analysis suggested that these high concen-
trations were the product of a rather unusual sequence of events. The ozone was brought into
the troposphere from the stratosphere by rather large-scale circulations associated with the
advance of the frontal zone. However the ultimate transport to the ground resulted from
smaller scale air circulations around a shower cloud. Although the events produced very high
concentrations, those concentrations were short-lived and affected only a relatively small area
within a few tens of kilometers of the observation site.
18
-------
120
100 —
O
N
O
FROM URBAN CENTERS
LOCAL OZONE
SYNTHESIS
20
NOV
JAN
MAR
MAY
JUL
SEPT
NOV
FIGURE 8 IDEALIZED OZONE VARIATIONS AT REMOTE LOCATIONS
Although it appears that violations of the air quality standards due solely to the
introduction of stratospheric ozone into the lower atmosphere are rather rare, and we were
unable to identify such events in the data that we analyzed, they should not be dismissed alto-
gether. Occurrences of violations induced by stratospheric air are likely to be associated with
meteorological conditions that are different from those that produce photochemical ozone; and
any control strategies based on these anomalous conditions are not likely to be very effective.
Strong outbreaks of cold air during cyclogenesis in the late winter or spring are candidate events
for intrusion of stratospheric air, but are unlikely to be associated with the photochemical pro-
duction of ozone. For purposes of policy formulation, it is extremely important to differentiate
between ozone of natural origin and that produced photochemically from anthropogenic emis-
sions.
19
-------
2. Anthropogenic Sources of Tropospheric Ozone
Intelligent formulation of control policy requires not only the differentiation between
natural an anthropogenic causes of standards violations, but also the differentiation among
several different kinds of anthropogenic causes. As noted before, anthropogenic causes of
ozone standard violations can be classified in three categories. The first involves those cases
where the precursor emissions.and the standards violations take place reasonably close to each
other~i.e., the problem is localized. The second category involves cases where problems are
almost solely of anthropogenic origin, but the ozone is transported over long distances so that
the problem is widely separated from its cause. The third category includes those cases where
appreciable amounts of ozone are advected into an area and the standard is violated when the
advected ozone is supplemented by that produced from local emissions.
Ludwig and Shelar (1978a) examined the problem of determining where the greatest
amounts of locally produced ozone are found relative to a city. They concluded that such areas
are apt to be where the air will be after it has traveled 5 to 7 hours from the upwind side of
the city. When conditions are favorable to the formation of photochemical oxidants~i.e., warm,
sunny days with light winds. They note that for a smaller city (as is the rule in Northern New
England) the maximum effects are likely to be observed closer to the city. In any event, max-
imum concentrations of locally produced ozone are apt to occur within a few tens of kilometers
of the source region, but outside the area of NO emissions because the short-term effects of
NO are to reduce ozone concentrations.
Although the maximum effects occur within a few tens of kilometers of the city, a
detectable effect can extend much farther downwind. Ludwig and Shelar (1977) have con-
cluded that appreciable effects could be observed for more than 100 km downwind of some
large cities in the northeastern United States. This is in general agreement with findings of
many others (e.g., Cleveland et al., 1976; Zeller et al., 1977) the latter authors have reported
observing the plume of ozone generated from emissions in the Boston area over the ocean 200
km downwind of Boston. Another example is discussed later.
There are likely to be differences in the diurnal patterns of ozone concentrations
between the locally generated ozone and that which is advected from cities long distances
away. Figure 9 is a schematic illustration of what one might expect to observe in the plume of
emissions from a city. The top part of the figure shows the plume of hydrocarbons and oxides
of nitrogen emitted during the morning rush hour. These emissions are reasonably well mixed
and have traveled some distance downwind of the city. At the greater downwind distances the
densities are much less, showing the dilution that has taken place and the effects of the much
smaller nighttime emissions. There is some ozone scattered among the oxides of nitrogen and
the hydrocarbons. At this morning hour, some of this ozone is likely to have been produced by
photochemical activity, but most is apt to have been advected or to be of natural origin. The
second panel of the figure shows the situation at about midday. The plume is dense with
oxides of nitrogen and hydrocarbons for some distance downwind. Of considerable significance
is the fact that much ozone has been produced by photochemical processes during the day. This
ozone is distributed nearly uniformly in the vertical because of the strong daytime mixing'. In
the horizontal, we see the concentrations increasing downwind of the city and then falling off.
The third panel of Figure 9 shows the situation in the late afternoon or early even-
ing, when the oxides of nitrogen and hydrocarbons are uniformly mixed in the vertical and are
seen to be fairly dense for considerable distances downwind of the city. However, there are few
ozone symbols scattered among emissions that occurred later in the day (and are still close to
20
-------
HOUR OF j
THE DAY I N°x
HC
HC
0900
HC N°x
NO
; HC
no
.HC
HC
WO
HC
NO
HC
MO
HC
MOX
HC
NOX
NO
NO
NO HC
N0x
,,^
HC 3
HC
He NOX HC O3
°3 NOX QHi
* °3 HC 3 ,
, 3 HQ HC
HC NOX
O
f'°* O, NOx*C°3NOx°3NOrj
—-"•* *•* ^-— ——
1800
I
! HC NO,
HC
0 03 HC 0
^X ° 3
NO I
HC
* NO HC t»O O N0)( NOX
IP N0 „ HC
JlMn HC N° N° N° "
'^•Majuia
NO
O ,
3
O O -MOv°3HC
°3 HC NO 3 ^°30
,,r. NOXHC x-
HC!
2300
HC
NOX
NO. „ HC
. NO^S !
HC
FIGURE 9 SCHEMATIC DIAGRAM OF THE TRANSPORT OF PRECURSORS.
OZONE PRODUCTION, AND OZONE DESTRUCTION DOWNWIND
OF A CITY
21
-------
the city), because photochemical processes are rather ineffective in producing ozone in the late
afternoon. Another important characteristic of this diagram is the fact that ozone is confined to
the upper parts of the plums at the greater downwind distances; by the late afternoon or early
evening a stable layer is likely to have formed at ground level and this prevents mixing of the
ozone throughout the entire depth of the plume. The ozone that was in the lowest layers is des-
troyed at the ground and is not. replaced by mixing from above.
The final panel of Figure 9 is an extension of the situation shown in the the preced-
ing panel. The concentrations of oxides of nitrogen and hydrocarbons begin to drop as the
emission rates decline later at night. No ozone is produced after sunset, and only that which
occurs naturally or has been advected from elsewhere, will be found in the part of the plume
that is close to the city. Farther downwind, in the part of the plume that was released earlier in
the day and in which photochemical activity has taken place, one sees the ozone that was pro-
duced during the afternoon. As before, this ozone is confined to the upper parts of the the
plume; the ozone in the lower parts has been destroyed at the surface.
Figure 9 shows that a large part of the transport of ozone may take place in elevated
layers. Hence it will be very difficult to estimate the contribution of advected ozone to
observed concentrations unless data from aloft are available. Some estimates can be made
when vertical mixing is good and the data are free of urban influences. Since the extent of
urban influence is very difficult to quantify and virtually no aircraft data were available, it has
been nearly impossible to determine the exact magnitude of the concentrations of ozone
imported into the northern New England states.
If the vertical mixing associated with a weather system were to occur at night, then
the ozone isolated aloft would be mixed to ground level and high concentrations would be
observed. Looking at the last panel of Figure 9, it is evident that the highest concentrations
would occur far downwind of the source area if the vertical mixing were a widespread
phenomenon. Ludwig and Shelar (1977) have shown two examples of this kind of nighttime
behavior in the area downwind of New York City. One of the implications of the behavior
shown in Figure 9 is that the emissions causing high concentrations of ozone at night probably
occurred many hours earlier. Even under very light wind conditions, where the wind speed
averaged through the mixing layer may be only 5-15 km per hour, the high nighttime ozone
concentrations may be found 100 to 200 km from the source area. In fact, it appears that high
concentrations of ozone are more likely to occur at night at great distances from source areas
than close to them.
3. Combined Natural and A nthropogenic Effects
Figures 7 and 8, which were shown earlier to illustrate the diurnal and annual cycles
of ozone concentrations in the lower troposphere, also include a schematic representation of the
changes in concentrations that occur when anthropogenic effects are introduced. Figure 7
shows that one might expect larger diurnal variations hi ozone concentrations when there is
photochemical production from emissions near the monitoring site. As was discussed before, if
there is appreciable vertical mixing during the evening, then night and day ozone concentra-
tions at ground level will not differ by very much when the only source of ozone is the tropos-
pheric background. When there is local photochemical production of ozone, then afternoon
values are likely to be much higher than nighttime values, even in the presence of nighttime
mixing.
22
-------
If anthropogenic ozone is transported over long distances before reaching the sta-
tion, then it is quite possible to have nighttime ozone concentrations higher than those that
occurred during the afternoon. This is very unlikely to happen when the sources are natural
background or local photochemical production. In fact, if the station is located on a hill or a
mountain, away from the destructive surface processes, and it is subject to receiving ozone that
has been transported over long distances, then it may well be that the highest concentration will
quite regularly occur during the night. The monitoring site at White Face Mountain is an
example of such a site. There are few local anthropogenic sources of ozone precursors, and the
site is on a mountain top. It is frequently downwind of several of the urban centers in the
east--e.g., Pittsburgh, Cleveland, Toronto, Montreal, and New York. Figure 10 shows the fre-
quency with which the maximum daily ozone concentrations occurred during different hours of
the day for three recent years at White Face Mountain. When the highest hour-averaged con-
centration occurred more than once during a day, these occurrences were distributed among the
several hours in which that maximum occurred. For example, if the maximum occurred during
two hours, each of those hours was credited with half an occurrence for that day.
In 2 of the 3 years shown in Figure 10, the maximum was most likely to occur
between 10 p.m. and 4 a.m. EST. In the third year shown, there were also an appreciable
number of maxima during the afternoon hours.
The principal reason for the preceding discussion is to provide some information
about what one might look for in the data in order to recognize the nature of the sources pro-
ducing high ozone concentrations. Unfortunately, the connections between the causes and the
observed diurnal and annual cycles of ozone are not always unique. The changes from season
to season and from hour to hour provide a useful, but not infallible, guide to the understanding
of what is happening.
23
-------
60
CO
40
ui
g 20
(a)
YEAR 1974
60
CO
O 40
u_
O
O 20
cc
ui
a.
(b)
YEAR 1975
6U
CO
0
u.
O
1-
z
g 20
UJ
o.
0
. (O
.
YEAR 1976 "
I
i , I . ' . I
•
5 10 15 20
EASTERN STANDARD TIME - hours
25
FIGURE 10 DIURNAL DISTRIBUTION OF DAILY 1-HR O3 MAXIMUM OCCURRENCE AT
WHITE FACE MOUNTAINS, N.Y. FOR 1974, 1975, AND 1976 / /
24
-------
IV ANALYSIS OF THE ORIGINS OF THE AIR AS RELATED TO OZONE CONCENTRATIONS
A. Governing Factors
It should be evident from preceding discussions that two very basic requirements have to
be met before high oxidant concentrations will be observeed. First, the oxidant must be pro-
duced, and second, it must be-brought to the place of observation without too much destruc-
tion or dilution. The major production process of concern here is photochemical, operating on
precursor materials (hydrocarbons and oxides of nitrogen) emitted into the lower atmosphere.
To be operative, this production mechanism has some well defined requirements-sunshine,
warm temperatures, and an accumulation of precursors, Some of the meteorological situations
that are accompanied by conditions conducive to oxidant formation are examined in this sec-
tion.
In particular, examples of the stagnation conditions conducive to the accumulation of high
concentrations of precursors are discussed, as are cases involving the change from warm, sunny
air masses conducive to photochemical activity, to cooler air masses. The change from one air
mass type to another is generally marked by the passage of a weather front. Several of the
cases discussed on the following pages involve the behavior of ozone concentrations during
frontal passages.
The transport of ozone in the troposphere is also dependent on the prevailng meteorologi-
cal conditions. The following sections discuss the transport of ozone in the Northern New Eng-
land states. Transport must be considered not only as a means by which ozone is moved from
one place to another, but also in terms of its effect on the destruction and dilution of preexist-
ing ozone.
B. Formation and Transport in the Lower Atmosphere
1. Winds and General Weather Patterns Associated with
Oxidant Concentrations above the NAAQS
a. Winds and Trajectories
Ludwig et al. (1977a) classified days on which the ozone standards were
violated according to the wind conditions and the large-scale meteorological patterns that pre-
vailed in different parts of the eastern United States during 1974. Table 3 from that report
summarizes the results that they obtained when the days were classified according to the winds.
It is apparent from Table 3 that the heavy preponderance of high concentrations in New Eng-
land occurred with winds from the southwest quadrant. The data from Northern New England
have been examined in a simlilar fashion.
Figure 11 shows schematically the number of hours when observed ozone con-
centrations in excess of 80 ppb at different sites were associated with winds from different
directions. The winds at the 850-mb level, about 1500 m above sea level, were estimated from
U.S. Weather Service analyses. The lengths of the sector radii in Figure 11 are proportional to
the numbers of hours that the air was moving from the corresponding directions. The 850-mb
winds were used because they are less influenced by local topography and tend to give a truer
picture of general air motions through the region. The importance of air motions from the
southwest is evident, but westerly winds in general seem to be about equally important. Similar
25
-------
Table 3
WINDS REPORTED ON MORNING WEATHER MAP IN AREAS
WHERE PEAK-HOUR OZONE EXCEEDED 80 ppb DURING THE DAY
(Number of days from June through August, 1974)
Region
Florida Peninsula
Texas -Louisiana
Gulf Coast
New York -New
England
Western Oklahoma,
Kansas , Nebraska
SE of Lakes Erie
and Ontario
Washington -Phil-
adelphia Corridor
S or SW shore of
Lake Michigan
Ohio River Valley
& Surroundings
Calm
11
17
3
1
20
9
7
21
Surface Winds
> 2 m/s
N to E
5
10
4
7
1
5
0
2
E to S
4
0
7
13
10
7
5
1
S to W
2
7
26
36
6
7
6
7
W to N
0
1
2
11
1
4
1
0
26
-------
76 76
OTTAWA
!
UONTRtAl
/T
/
sT
PORTLAND
GRAFTON
I
DEERFIELD
Hf'
MANCHESTER
0 10 20 30 40 50 hours
M 100 iso ;oo no •.-,.
50 100 ISO 300 J50 XO ISO K.iom.1.,
FIGURE 11 NUMBER OF HOURS WHEN OZONE CONCENTRATIONS VIOLATED THE IMAAQS FOR OXIDANTS
AS A FUNCTION OF 850-MB WIND DIRECTION—1976
-------
(a)
o u
-------
i52 ' -^^' ^^f^f^f ^
rr'^^ <\ .^^^
_ JV I S& .' -T-^I--. ; ' \ 7. ^/
^1C ^^^^«^k:.^f^
^UH8T' -- <— - " —•
- fl ,*J _ .,, " "u_ l^"^J^\* -4|
- -s^s^i^^rvj^" ^^li^-^^-^^y
- -^.: \.^&^-£^^^'':''^~-^^&
- £-- rS ~*- -i i f* ' >>TJ -^"iiJ-i-- - -~x*-• ^^r^'^iSw
;;W:,^^V:;^a^i^'
^^iS^i^Msferff.
.
--^s • -H.' "-."-^S;-~^^'-&-^S'-^'*--~^-3&'> '
;.-v^^-_,^tei^^f^^ .
.^^^ "%ij^^f^^^^
x';^f^i^S^j^y t
^^bx-c^u/
_jv.^ --^ :JT^H;-^VC
FIGURE 13 COUNTIES WITH AVERAGE ANNUAL NOX EMISSIONS
GREATER THAN 75 TON M -2 YR ~1
displays, based on the 1977 data, are shown in Figure 12. In the 1977 data set, winds from the
northwest assumed greater importance than for 1976.
It should be remembered that the results shown in the figures reflect three fac-
tors: first is the size of the data set (see Table 1), second is the overall frequency of winds from
the various directions, and the third is the frequency with which winds from a given direction
are associated with conditions suitable for oxidant concentrations in excess of 80 ppb. The two
figures also seem to display evidence of a bias against wind directions other than the eight
major directions; there seems to have been a preference for labeling winds W or SW rather
WSW, for example. This may reflect a real physical effect, because the emissions of oxides of
nitrogen tend to be greater toward the southwest and the west for most of the stations. Figure
13 summarizes county-wide NO emissions in the northeastern United States for 1974. Greater
emissions are found to the west or the southwest than to the west-southwest for many of the
stations.
-\
The cases associated wih northwesterly winds in 1977 were nearly all connected
with a very few episodes in April and May. It appears that the air arriving from the northwest
29
-------
was usually part of a high pressure circulation that had persisted long enough for there to be an
accumulation of emissions from many sources in the northeastern United States. The accumu-
lated emissions could have moved to the north and then have entered the northern New Eng-
land area from the northwest. This serves to illustrate the fact that wind directions over a lim-
ited time interval are not always accurate indicators of the origins of the air.
Air trajectories were calculated from upper air wind data for selected cases in
order to provide a better estimate of the origins of the air. The calculations were prepared by
Mr. Dale Coventry of the EPA in Research Triangle Park, North Carolina. The calculations are
based on the computer program described by Heffter and Taylor (1975). The application of this
program to problems of this type has been discussed by Ludwig et al. (1977b). Briefly, the
winds are interpolated horizontally and averaged vertically through the mixing layer, The air
motions are calculated from these averaged, interpolated winds for the periods between obser-
vations. The program provides calculated air positions at 6 hour intervals.
It was beyond the scope of this project to calculate trajectories for every day
and for every ozone monitoring site. Trajectories terminating at Burlington, Vermont and Man-
chester, New Hampshire were calculated for all the days shown in Table 4. The time periods
were selected to include one or more instances of widespread violations of the oxidant standard
in northern New England or some period that was being considered for special analysis because
of the prevailing meteorology or high nighttime ozone concentrations.
Table 4
TIME PERIODS FOR WHICH AIR TRAJECTORIES
WERE CALCULATED
17 April 1976 to
27 May 1976 to
13 June 1976 to
8 July 1976 to
24 August 1976 to
8 September 1976 to
9 March 1977 to
18 April 1977 to
29 April 1977 to
12 June 1977 to
20 April 1976
30 May 1976
17 June 1976
12 July 1976
29 August 1976
11 September 1976
12 March 1977
25 April 1977
24 May 1977
29 June 1977
30
-------
.,.118
FIGURE 14 LOCATIONS OF AIR THAT ARRIVED AT MANCHESTER IN 12 HOURS
WITH AN OZONE CONCENTRATION OF 70 PPB OR GREATER UPON ARRIVAL
Ozone concentrations are shown for time of arrival. Underlined figures show arrival
at 1900; others, at 1300
The ozone concentration at the terminus of each trajectory (Burlington or
Manchester) was plotted at the location of the air 12 hours earlier. This was done for every
available trajectory that terminated at 1300 or 1900. Figures 14 and 15 show those cases where
the air arrived at the target location with an ozone concentration in excess of 70 ppb. For
example the numeral 78 plotted near Boston in Figure 14 means that the air at that location
arrived in Manchester 12 hours later (at 1300 EST) and the ozone concentration observed in
Manchester at the time of arrival was 78 ppb.
It is evident from Figure 14 that the higher ozone concentrations measured at
Manchester were frequently associated with air that had passed over the high population density
areas of New Jersey, Connecticut, and southeastern New York during the preceding 12 hours.
Another favored corridor stretched over the Buffalo, Toronto, and Rochester areas. It is some-
what surprising that the nearby Boston area was not involved in very many instances of higher
ozone concentrations at Manchester.
31
-------
-«-82
72
I
FIGURE 15 LOCATIONS OF AIR THAT ARRIVED AT BURLINGTON IN 12 HOURS
WITH AN OZONE CONCENTRATION OF 70 PPB OR GREATER UPON ARRIVAL
Ozone concentrations are shown for time of arrival. Underlined figures show arrival
at 1900; others, at 1300
It should be noted that Boston may well influence some areas in northern New
England especially those to the northeast, such as Portland, Maine. For example, Figure 11
shows that high ozone concentrations at Portland are frequently associated with winds from the
direction of Boston. In that same figure, Manchester shows no occurrences from that directon,
but consistent with the trajectory results, Manchester shows frequent occurrences with winds
from the southwest and west. The subject of transport toward Portland along a path parallel to
the coast will be discussed again later.
Figure IS shows that even Burlington is influenced on occasion by air arriving
from the direction of New York and New Jersey. However, it is more common that the higher
afternoon and early evening ozone concentrations are associated with air that has passed, over
the areas southeast of Lakes Erie and Ontario. This swath includes metropolitan areas such as
Pittsburgh, Cleveland, and Buffalo. There were three instances in this sample when concentra-
tions above 70 ppb were connected with air that had been in the vicinity of Montreal twelve
hours earlier. It appears that when there are light north-northwesterly winds the influence of
Montreal just extends to Burlington.
32
-------
Those cases in which air coming from the area north of Lake Huron produced
high ozone concentrations are somewhat puzzling because of the long travel distances involved
The long distance of travel implies very high wind speeds, which in turn indicates very little
time for emissions to accumulate and reach the high concentrations that are usually presumed
to be needed for the formation of high ozone concentrations. The time and data available for
this project have not been adequate to determine whether the observed ozone concentrations
are connected with smelting or other industrial activities north of Lake Huron If more
comprehensive trajectory analysis showed that air from that area is regularly associated with
high ozone concentrations, then special measurement programs to clarify the effects might be
warranted.
b. Weather Patterns
Ludwig et al. (1977b) prepared analyses of ozone concentrations in the eastern
United States for each day of 1974. The areas where the NAAQS for oxidant were violated
were classified according to the weather type in the same area. Table 5 from Ludwig et al
(1977a) shows the weather patterns that were most frequently associated with high ozone con-
centrations. A similar analysis has been prepared for the Northern New England States.
Table 5
METEOROLOGICAL FEATURES ASSOCIATED WITH VIOLATIONS OF THE
NAAQS FOR OZONE DURING 1974 IN THE EASTERN UNITED STATES
(Number of Cases per Month)
Warm air mass
near front
Warm sector of
frontal wave
West side of
anticyclone
Center or east
of anticyclone
Squall line
Behind strong
cold front
Other
January
1
0
0
0
0
1
0
February
0
1
I
n
1
2
0
0
0
March
6
1
2
2
0
2
1
1-1
•t-j
&
<
5
0
12
4
0
2
6
I
15
4
11
14
1
1
10
0)
c
>->
8
4
22
15
1
6
15
>,
1-1
3
•n
19
8
35
44
0
7
15
4J
W
§>
3
23
4
30
20
0
10
21
September
9
2
18
11
0
0
7
October
8
2
14
10
0
2
2
November
2
1
0
2
0
2
0
December
0
0
2
1
0
0
0
33
-------
Weather maps from the "Daily Weather Map" series (National Oceanic and
Atmospheric Administration; 1976, 1977) were subjectively classified for those days during
which the NAAQS were observed to be violated somewhere in the three Northern New Eng-
land States during 1976 and the first half of 1977. Table 6 summarizes the number of days of
violation associated with each of several different weather types. The weather pattern
classifications are the same as those used by Ludwig et al. (1977a).
Table 6 shows that the northwest quadrant of a high-pressure system, or anti-
cyclone, is the most common location for high ozone concentrations in the Northern New Eng-
land States, as it is elsewhere in the eastern United States according to Table 5. The extreme
western boundaries of high-pressure systems are often marked by approaching frontal systems.
The warm air mass ahead of the front is often amenable to ozone formation, especially in areas
where the air moving from the southwest passes over an extended emissions area. Table 6
shows that the Northern New England States have an appreciable number of days when the
NAAQS are violated and the weather pattern fits into this category.
Table 6
METEOROLOGICAL FEATURES ASSOCIATED WITH OBSERVED OZONE CONCENTRATIONS
OF 80 ppb OR GREATER IN THE THREE NORTHERN NEW ENGLAND STATES
(Number of Cases)
Weather Feature
Number of
Days
Warm air mass
near front
Warm sector of a
frontal wave
West or northwest side
of an anticyclone
Center or east side of
an anticyclone
Other
16
11
25
12
34
-------
Another weather category of importance is the warm air region of the wave
between a cold front and a warm front. Eleven violations of the NAAQS were found in such a
region. Six days of violations for ozone occurred when the central or eastern parts of a high-
pressure system were over the states of Maine, New Hampshire, or Vermont. As shown in
Table 6, 12 days could not be categorized.
The following text presents some specific cases when the high ozone concen-
trations were associated with light winds or stagnation in a high- pressure system or with a
weather front and its associated air circulations.
2. The Effects of Weather Fronts
Weather fronts mark the boundary between two bodies of air of different properties
(air masses). In meteorology the emphasis is on the contrast in temperature and humidity
across the weather front. However, there will also be contrasts in the amounts of pollutants in
the air and the origins of those pollutants. Thus, the changes in pollutant concentration during
a frontal passage or the variation of concentrations in the vicinity of a front are indicators of the
relative importance of the sources that contributed to the concentrations on either side of the
front. Ludwig and Shelar (1977, 1978b) have presented some dramatic examples of changes in
ozone concentration during frontal passages over Southern New England. In those cases, warm
air laden with the emissions from the New York and New Jersey areas was contrasted with
cooler, relatively cleaner air from the north. The results of the trajectory analysis suggest that
similar events might occur farther to the north.
Several days when frontal passages occurred have been chosen for detailed analysis.
One of the objectives of the analysis is to obtain an estimate of the relative importance to the
production of oxidants of the sources in the two different air masses. One of the problems that
arises in such an analysis is that the concentrations of ozone are controlled by more than just
precursor concentrations. In particular, photochemical production of ozone varies with the time
of day, and it is not always easy to separate the diurnal variations of ozone production, destruc-
tion, and dilution from the variations caused by differences between the air masses on either
side of a front as it passes. We have been fortunate to find some examples where a front
moved back and forth through the area, providing more than one sample of the differences
between the air masses. These cases are discussed on the following pages.
a. 18-20 April 1976
This was an interesting period for two reasons. First, it happened relatively
early in the year for violations of the oxidant NAAQS, and second, the behavior of the warm
front that affected the area on the 19th was unusual. Although violations were recorded only at
Nashua and Portland, concentrations approaching the 80 ppb standard for ozone were recorded
throughout the area.
This was an unusually warm period for mid-April, with temperatures reaching
the upper 80's (F) and low 90's on the 18th and 19th. Surface wind speeds were generally low.
The daily weather map for 0800 EOT on the 18th [Figure 16(a)] shows that Northern New
England was influenced by high pressure to the south, and a weak warm front to the northeast.
By 0800 EOT on the 19th [Figure 16(b)l the area was lying in the warm sector of a more well
defined depression. Figure 17(b) shows the movement of the warm front during the morning
hours of the 19th. As the afternoon progressed, the front behaved in a rather unusual manner,
35
-------
r/«v I«*vctf£w
W^f*-^'
$<&*&&¥
&T'\V%?$£
M^-^'^^fe
(a) 18 APRIL 1976
(b) 19 APRIL 1976
-------
OJ
(d) 20 APRIL 1976 , 0200 EOT
(c) 19/20 APRIL 1976 \
Jl
FIGURE 17 FRONTAL POSITIONS IN NORTHERN NEW ENGLAND, 18-20 APRIL 1976
-------
looping back over the southern part of the area [Figure 17(c)] to affect sites in Southern New
Hampshire, and Maine. The cold front of the system moved southeast across the area on the
20th, clearing Portsmouth and Portland by 0800 EDT [Figure 16(c)].
The variations in ozone concentrations during this time are shown in Figure
18. There is a marked contrast between the smooth diurnal cycle of the 18th (especially at
Manchester and Nashua) and the much more variable cycle on the 19th. The low wind speeds
and clear skies together with the high temperatures provided near ideal conditions for the pho-
tochemical production of ozone on the 18th. All sites show rises in concentration between
0600 and 0800 EDT on the 18th. On the 19th, however, the initial rise in concentrations hap-
pened earlier (0300, 0400 at Manchester and Nashua) and was followed by a series of increases
and decreases throughout the day and into the night hours. These variations can be ascribed to
the influence of the warm and cold sectors of the depression as the warm front moved back and
forth over the area. The estimated times of influence of each air mass have been marked on
the graphs of Figure 18.
The relationship between the air masses and the ozone concentration was most
pronounced at Nashua, but all the sites showed the effect to some extent The obvious ten-
dency is for ozone concentrations to increase as the sites come under the influence of the warm
air and decrease when the cold air mass arrives. The warm air was richer in ozone, probably for
two reasons. First, the meteorological conditions were more suitable for photochemical
activity, and second, the warmer mass had been exposed to more precursor emissions from the
heavily populated areas to the south than had the cold air.
b. 11, 12 July 1976
There were widespread violations of the standard in Northern New England
during daylight hours on 11 July 1976. The area was influenced by the development of a low-
pressure system just to the north of New Hampshire. Highest concentrations were recorded
around 1400 EDT on the llth; the period was broken up by the passage of the cold front across
the area on the 12th.
The surface weather map (Figure 19) for the llth shows extremely weak pres-
sure gradients over the Northeast. At 0800 EDT there was a weak front lying to the west of
the New England Area. Figure 20 shows the movement of this front during the llth. The
whole area was within the warm air mass by 2000 EDT. The surface wind speeds (Figure 21)
were low until 0900; wind directions were variable. As the wind speed increased around 0900,
the wind directions became steadier, and from the south.
The observed ozone concentrations for individual sites are shown in Figure 22.
The initial sharp rise in concentrations around 0600 EDT at Portland, Manchester, and
Deerfield cannot be attributed to photochemistry, because of the early hour and consequent low
level of insolation. The rise does correspond with the invasion of warm ah- from the south.
The observed ozone had probably been formed in the warm air mass the previous day and then
was isolated aloft by the stable layer associated with surface cooling; finally, it was transported
northward. There was probably some increase in vertical mixing associated with the area near
the front and the onset of surface heating at sunrise. This vertical mixing brought ozone stored
aloft in the warm air down to ground level and caused the observed rise in concentrations.
Burlington concentrations remained at about SO ppb through the night, .but also
began to rise somewhat later, at around 0800 EDT. Burlington seems to have remained in the
cold air mass until early afternoon, so the morning rise reflects the photochemical production of
38
-------
80
BURLINGTON
40
COLD AIR MASS WARM AIR MASS COLD AIR MASS
••••••••••••*^^^""»i^^™B^^^^^MBBHBBBB»» ••*•••••**••••
V\A J\ f
• \ I V
;
v—x
80
MANCHESTER
• ••••• aMi^^n •••••••!
40 -
I
O
EC
W
O
1 x
/
/
/
^ /
,
— \__I_ 1
u
1
A
^ i V\ A
j v v \.
1 1
• — y
/•••
i §
120
80
40
NASHUA
40 -
PORTLAND
80 -
40 -
-SOppb•
\
-.' \
•-. \
T-J
\ . v
. \ A .
j—*" • i
/
— * ,
' /7/1 ' -
i , , " i , , "\,
"f\
i
,\s
J
i
• i •
/• /\ ouppb • •
-N /V ' /-A
V \ / \
Vs-s / \ A .-^^
\ / V
i i . i V.-j / i i
0 8 16 0 8 16 0 8 16 0
18 APRIL 1976 19 APRIL 1976 20 APRIL 1976
EASTERN DAYLIGHT TIME - hour
FIGURE 18 TIME HISTORIES OF OZONE CONCENTRATIONS IN NORTHERN NEW ENGLAND.
18-20 APRIL 1976
39
-------
\
^'^'•'
. - -
(a) 11 JULY 1976
(b) 12 JULY 1976
FIGURE 19 WEATHER MAPS FOR 11-12 JULY 1976
-------
1100 EOT
1400 EDT
2000 EDT
0 K 100 IM TOO »0 300 190
FIGURE 20 FRONTAL POSITIONS IN NORTHERN NEW ENGLAND, 11 JULY 1976
-------
360
to
i
13
11 JULY 1976
12 JULY 1976
EASTERN DAYLIGHT TIME - hour
FIGURE 21 TIME HISTORIES OF WINDS AT SELECTED NORTHERN NEW ENGLAND SITES, 11-12 JULY 1976
-------
DEERFIELD
4
a
1
O
H
ZONE CONCENTR»
u
40
0
12O
80
40
0
140
120
80
40
o
i i | • i
A
r v.
- / V/N- ,—
~s — / .iii,
MANCHESTER
1 /\ 1 1 1 1
«*** \
/
BURLINGTON
i i | i i
/•J1.-.^*X1 AOanh -
m \f ^
0 8 16 0 8 16
11 JULY 1976 12 JULY 1976
<
i i | i •
i - 1 i , X""~
BERLIN
i i | i i
A
/'" VM
PORTLAND
1 i | i <
A....
i \
/ \ / *^' ~\ S"
) 8 16 0 8 16 0
11 JULY 1976 12 JULY 1976
EASTERN DAYLIGHT TIME - hour
FIGURE 22 TIME HISTORIES OF OZONE CONCENTRATIONS IN NORTHERN NEW ENGLAND, 11-12 JULY 1976
-------
ozone from precursors within that air mass. The transition from the cold air ozone concentra-
tions to the warm air concentrations that takes place in the early afternoon is quite smooth.
This suggests that the ozone in the warm air mass had been diluted to the point where it was
about the same as cool air values by the time it reached this northerly station. This gives some
indication of how rapidly the ozone concentrations decrease (by dilution and destruction) with
distance from the areas where precursor emissions originate.
The graphs for Grafton County (Fraconia Notch) and Berlin show a much
more gradual increase, with Grafton never rising much above 40 ppb. Berlin gradually
approached the standard by 1600 EOT. As discussed in Appendix A, both these sites are rather
anomalous.
Conditions were extremely suitable for production of ozone. It is likely that
concentrations at all sites were reinforced by remoter sources of ozone upwind moving from
the urban complexes to the south.
Concentrations were considerably less at all sites on the 12th than they had
been on 11 July 1976. The surface wind data (Figure 21) showed a gradual increase in speed
around 0900 EDT, and a change in wind direction to the northwest. This is consistent with the
development and movement of the low pressure system, and the passage of the cold front
across the area (Figure 19). The air behind the front had different origins than that ahead of it
on the preceding day. The postfrontal air did not have the recent exposure to heavy precursor
emissions from the conurbation to the south. Therefore it is not surprising that the photo-
chemical reactions on 12 July did not produce concentrations that were as great as they had
been the day before. To some extent the difference between the two days reflects the influence
of sources to the south and southwest of the Northern New England states. However, it should
be remembered that the differences also reflect differences in the meteorolological conditions
on the two days.
c. 17-18 June 1977
These two days were a time of late afternoon and early evening violations asso-
ciated with the movement of a depression across Northern New England. The surface weather
charts indicated that the warm front of a slow moving low-pressure system entered the area
from the southwest around noon on the 17th (Figure 23). However, the surface wind data at
sites through the area (Figure 24) show that the front was weak, hardly affecting speed or direc-
tion. At 1400 EDT, Portsmouth and Portland were still in the cold air according to the
National Weather Service analysis. The warm front of the system passed over these coastal
locations by 1700. The cold front of the system had a much more definite impact on wind
speeds and directions, particularly at Burlington (Figure 24). The front passed Burlington
around 2300 EDT on the 17th, shifting wind directions to the northwest. It probably crossed
the White River Junction area some 3 to 7 hours later, and cleared southern New Hampshire
by 1400. Portsmouth was still in the warm sector at this time. Another system began to
develop, and toward the end of the 18th the area was once again in a warm sector.
Ozone concentrations for sites with data at this time are shown in Figure 125.
Violations were recorded between 1600 and about 2300 on the 17th at all but Portsmouth which
showed a pronounced peak of violations around 1700 EDT on the 18th at a time when Man-
chester and Nashua experienced a sharp drop in concentrations.
The observed variations can be explained in part by the movement of the
depression across the area. Concentrations began rising steadily at all sites except Burlington
44
-------
(a) 17 JUNE 1977
(b) 17-18 JUNE 1977
(c) 18 JUNE 1977
100 50 0 100 200 300 400 500
SCALE - km
FIGURE 23 FRONTAL POSITIONS IN NORTHERN NEW ENGLAND, 17-18 JUNE 1977
-------
BURLINGTON
PORTLAND
CONCORD
360
270
180
90
20
10
I
O
H
EC
S
Q
Z
i
i
Q
ui
ui
i
f
i i . i
\AAA...
1
A
I
t /
II 1 '
/Vv" ~\
T \uv 'V ^
i , ) ,"
i i i i i
/ -i
/vA/ K .A A
.AA/ . rA/x/ ':
i i i i i
A
/
A i , .if. :
1 9 17
17 JUNE 1977
9 17 1 1 9 17 1
18 JUNE 1977 17 JUNE 1977
9 17 1 1
18 JUNE 1077
5 17 1
17 JUNE 1977
9 17 1
18 JUNE 1977
EASTERN DAYLIGHT TIME - hour
FIGURE 24 TIME HISTORIES OF WINDS AT SELECTED NORTHERN NEW ENGLAND SITES, 17-18 JUNE 1977
-------
COLD AIR MASS WARM AIR MASS COLD AIR MASS
BHH^^^^^H«*"""*"*"
120
80
40
0
4
, 120
O
5
OZONE CONCENTR
g §
o
120
80
40
0
- 1
KEENE
r~\
•- /' V.-" A-\
. r/ ~\j v.
— '--^.4 . I , i L
WHITE RIVER •-.
JUNCTION / \
/ s_..-. — aoppb
- / NV^\,/-
/ . w
.-•III '1
BURLINGTON
.,.'N_/\
/ ^ 80 ppb
/ \
:"-\ J \
' '"^ \ -A.. /"•
. . y-< . v ,
1 9 17 1 9 17 1
17 JUNE 1977 18 JUNE 1977
PORTSMOUTH A
80 ppb ,. / ,
A/N/V-- \f \A
- / v
/
I J- . L
NASHUA
.A
/ - — 80 ppb ~
r \
v\ /"^'\. r\
y \./ \ /
/ \j
/"
, , I , , __L
MANCHESTER / \
/ \ SOpph
/ \ r\t\
• / v v -
./:.,..
1 9 17 1 9 17 1
17 JUNE 1977 18 JUNE 1977
EASTERN DAYLIGHT TIME - hour
FIGURE 25 TIME HISTORIES OF OZONE CONCENTRATIONS IN NORTHERN NEW ENGLAND,
11-18 JUNE 1977
-------
around 1900 EDT on the 17th. Conditions were not exactly favorable to the formation of pho-
tochemical ozone; skies were almost totally overcast, and wind speeds were in excess of 10
knots for most of the day. Figure 24 shows that wind speeds at Burlington had been relatively
high since 0400 EDT, indicating vertical mixing that was transfering momentum groundward.
The higher concentrations observed at this time may reflect this nighttime vertical mixing that
was not present at the other sites. Any ozone aloft would have been brought down to replace
that destroyed at the surface. The late occurrence of violations suggests that the ozone may
have been transported into the area from the large urban complexes to the south and arrived
later at this site than elsewhere. Wind directions remained steadily from the south during
much of this time. The concentrations at Portsmouth approached, but did not reach 80 ppb.
Portsmouth may have been away from the center of the polluted plume.
Wind speeds decreased during the evening of 18 June 1977, and ozone con-
centrations also fell gradually. All sites showed a minimum in concentration around 0500 EDT
on the 18th. The one experience of violations at Portsmouth on the 18th seems to be from the
movement of the cold front. The other sites in Southern New Hampshire seem to have had
the warm polluted air replaced by cooler, cleaner air behind the front sometime during the early
afternoon. Later, they were again overrun by the warm air according to the National Weather
Service analyses. The period in the cool cleaner air corresponds to the concentration minima at
Manchester and Nashua. Portsmouth probably remained in the warm sector during this period
and the violations were the result of transport of materials from the south. This case shows the
effect of transport from areas to the south at a time of day when local production of ozone was
minimal.
3. The Effects of Stagnation
If air remains in the same area with little or no motion for prolonged periods of
time, the emissions tend to accumulate, and if weather conditons are right, copious amounts of
ozone may be formed. Light winds tend to be most often associated with high-pressure cells,
and high-pressure cells are also noted for sunny conditons so the weather accompanying stagna-
tion conditions is probably conducive to ozone formation. True stagnation for extended periods
of time is highly unusual, but it has been possible to identify two instances where the air that
arrived in Northern New England had traveled only about 300 km or less in two days. These
two cases are discussed below. It was not possible to identify any cases where the air had
remained in Northern New England for more than a day.
a. 20-22 April 1977
This period provides an example of an instance when the Northern New Eng-
land states remain in the northwest quadrant of a high-pressure cell for an extended period of
time. Eventually, on April 22, a cold front passed through the region, bringing an end to the
period of light southwesterly winds that had prevailed. Trajectory calculations show that the air
arriving in the area of interest about midday on 21 April had been traveling very slowly from
the southwest at an average speed of only about 4 m/s for the preceding 24 hours.
Figure 26 shows the ozone concentrations for these days at the six sites with
data. A striking feature of the figure is the marked diurnal cycle at all sites, peaking around
1200 to 1600 EDT, with a minimum around 0000 to 0400. The violations occurred at Man-
chester and Portsmouth on the 21st and 22nd, on all three days at White River Junction, and
for an extended period between the 21st and 22nd at Burlington where, unfortunately, data
were missing on 20 April 1977. The regular diurnal cycle shows the effects of photochemical
production with its midday to early afternoon peak. The fact that the concentrations were
48
-------
KEENE
WHITE RIVER JUNCTION
I
I
o
I-
cc
z
iu
o
at
o
N
O
1 1 1 1 I I I I
MANCHESTER
80
40
NASHUA
40 -
PORTSMOUTH
80
40
0
17 1 9 17 1 9
EASTERN DAYLIGHT TIME - hour
17
FIGURE 26 TIME HISTORIES OF OZONE CONCENTRATIONS IN NORTHERN
NEW ENGLAND. 20-22 APRIL 1977
49
-------
rather low, in spite of the near-ideal opportunity for the accumulation of precursors from major
urban areas, is probably because skies were cloudy in the area.
The minima observed in the early evening at several of the sites may be effects
of local emissions of NO during the evening hours. These were weekdays, so an increase in
evening traffic is a plausible explanation. Wind speeds (Figure 27) increased during the late
evening and early morning hours at Concord and Burlington, suggesting that some vertical
mixing may have been going on that would have brought ozone formed earlier down to ground
level.
b. 20-24 May 1977
This is a second example of violations associated with the prolonged presence
of a high-pressure system, in this case generally to the southeast of Northern New England.
Although regular diurnal cycles were present in the ozone concentrations at each site (Figure
28), they were less distinct than those of the April example. NAAQS were violated for
extended periods during daylight hours on the 21st, 22nd and 23rd. To illustrate how slow the
air motion was during the period, calculated trajectories ending at Manchester showed that the
air that arrived at the beginning of the period had been about 250 km to the north-northeast
two days earlier. Air arriving at midday on the 21st had been about 1300 km to the west two
days before. However, the intervening trajectory brought it just north of the New York City
area. The air arriving at Manchester at midday on 22 April and on 23 April had been near Phi-
ladelphia the day before in both cases.
Although the average air motion was less for air arriving at Manchester on the
21st, the ozone concentrations were lower than on subsequent days when the air had moved
faster, but had passed over the high emissions areas toward the southwest. Obviously, stagna-
tion without emissions is not sufficient to be an effective producer of ozone. The alignment of
airflow with the elongated axis of the East coast emissions area seems to offset the high ventila-
tion rates even in distinctly non-stagnant conditions like those ahead of a front.
4. Special Meteorological Situations
a. Nighttime Effects
As has been noted before, the commonly observed diurnal cycle in ozone con-
centrations at ground level is largely the result of two mechanisms. The first is the production
of ozone during daylight hours from the photochemical reactions of its precursors, such as HC
and NO. Such production requires certain meteorological conditions, primarily a sufficient
intensity of sunlight, and therefore is confined to hours with light winds between sunrise and
sunset. Production is generally greatest around midday, with peaks in ozone concentrations
during early to midafternoon hours. The second factor is the vertical temperature structure of
the atmosphere. This will tend to promote vertical mixing during daylight hours when the sur-
face is warmer than the air above and mix down to ground level both locally produced ozone
and any ozone from natural or more remote anthropgenic sources that has been advected into
the area. During night hours, vertical mixing is usually inhibited because the surface cools
more rapidly than the surrounding air. This difference of cooling rates can lead to the forma-
tion of a temperature inversion near ground level, and prevent the mixing of any ozone trapped
above the lowest layers down to the ground. Any ozone that has remained below the inversion
after sunset will be destroyed by the scavenging action of HC, NO, and contact with surfaces,
but that at higher altitudes will be isolated from the surface destruction.
50
-------
360
270
O 180
I
cc
S
90
g 17
20 APRIL 1977
9 17
21 APRIL 1977
EASTERN DAYLIGHT TIME - hour
9 17
22 APRIL 1977
FIGURE 27 TIME HISTORIES OF WINDS AT BURLINGTON, VERMONT AND CONCORD, NEW HAMPSHIRE
20-22 APRIL 1977
-------
120
80
40
n
i i i i i
- PORTSMOUTH
/ \
/ \
• •
J \
^ 1 ^•'l ill'
I 1 i I i
— SOppb
•1 /~~ , , ,
(III
X. /*•». . . /
\i , ,
i ' / \ ' ' ' ' ' ' I
/•''~"N.' "% ~
/ \ / -""\ -
' \ / V
V
a
a
1
O
120
80
40
n
_ MANCHESTER
"
: / v^
i i i
i , /
../-
/
/
/
i i
\
\ .^r
v,^ /
111111
>»-»»^* \ y \
"% A
\ / "
, , i ,N--/. ,
_
—
• 7*
V:
ro
UJ
o
UJ
I
120
80
40
0
120
80
40
0
.11111
_ WHITE RIVER JUNCTION
1 ' ' ' u y\ ' ' ' ' '
SODDb / ^- / ^ -•' "1
BUPPD / N / " \
\,y \_
1
I-'
^
-.-•• " \
- BURLINGTON
: /,./\
_ . / -\
y\,<_< \
111111 /"\' ' '<"-— --\ '
nnnnK ,.•-«-• /\ / V y \r N
s
1
|
i i i i i
"N. f'*' ^». *~-S \ /
\l / * \t
^_/
I 1 1 1 1
8 16
20 MAY 1977
8 16
21 MAY 1977
8 16
22 MAY 1977
8 16
23 MAY 1977
EASTERN DAYLIGHT TIME - hour
FIGURE 28 TIME HISTORIES OF OZONE CONCENTRATIONS AT FOUR NORTHERN NEW ENGLAND SITES,
20-23 MAY 1977
-------
The occurrence of NAAQS violations during night hours therefore suggests
two prerequisites: (1) an initially sufficient source of ozone above ground level, which in turn
requires that conditions have been conducive to the formation and transport of ozone from
urban areas upwind for a period of time beforehand; and (2) the presence of vertical mixing in
order to bring the elevated ozone source to ground level. Usually the onset of vertical mixing
at night will be accompanied by certain symptomatic meteorological conditions. Symptoms
include cloudy skies, rises in temperature, or medium to strong winds. Increased temperatures
and winds mark the downward flux of heat and momentum, and cloudy skies will inhibit the
cooling at the surface that is required for the formation of temperature inversions.
A feature in the variation of ozone concentrations that follows from this rea-
soning is the decrease in concentrations as mixing continues. This results because the night-
time ozone source represents the daylight production from urban areas upwind, and as such it
would eventually become diluted or exhausted. There is no photochemical activity at night to
replace the ozone destroyed at the surface.
Referring back to Figure 6, the only site with frequent occurrences of viola-
tions during night hours was Burlington. Periods when more than just the odd hour or so
exceeded the standards were not common at most sites, as can be seen by reference to Appen-
dix B. However, two examples will now be discussed that serve to illustrate the points outlined
above.
i. 28-29 June 1977
This a period of violations of the standards at Nashua, Manchester, Burlington,
and Keene extending from about 1000 EOT on the 28th to as late as 0500 on the 29th at Bur-
lington. At all the sites, violations were recorded until about midnight on the 28th (see Figure
29). It should be noted that no data were available from Berlin, Portsmouth, and White River
Junction for this time.
•
The daily weather maps for these days (Figure 30) show a rather unusual situa-
tion with a warm front lying within the warm sector of a larger frontal system. By 0800 EDT
on the 29th, this front was straddling the area on roughly a NNW-SSE line. The movement of
the front for 9 hours previous is shown in Figure 31.
The surface wind data for these days (Figure 32) shows that there were con-
sistent southerly winds for most of the 28th and the morning hours of the 29th until the pas-
sage of the cold front, when wind directions shifted to the W and NW. Wind speeds, after an
initial low on the 28th, remained relativelyi high at Burlington and Albany (10 to 15 knots), and
a little lower at Portland and Concord (5 to 10 knots).
Ozone concentrations at all sites (Figure 29) showed a sharp increase at around
0700 on the 28th, probably the result of increased mixing. The increased wind speeds indicate
a downward transfer of momentum along with ozone from aloft as well. Photochemical activity
was probably under-way by this hour. Standards were exceeded by 0900 EDT at Keene, by
1000 at Burlington, and by 1300 at Manchester and Nashua. The timing of the violations is con-
sistent with the advection of warm dirty air behind the warm front.
The continuation of levels above 80 ppb well into night hours suggests that the
mechanisms outlined earlier were in operation. The area between the warm and cold fronts was
probably unstable so that vertical mixing was encouraged, even after sunset. Wind speeds had
remained high during night hours throughout the area, suggesting that vertical mixing was
53
-------
Ui
160
120
80
"
O
<
ai
s
O
NASHUA
A
X * X""t
T~~^~,
V ^
7
/
120
80
BURLINGTON ' S^s''\. '
/-"" V\
1 1 !__.. II 1 1 1 1 1 1
1 1 1 l.<\ 1 | T-
MANCHESTER / \
v,/ \-A
f \
\f>.i>r- ^ > — i — i — i — i «
— i 1 r r
SOppb
• f\
r1 \
v , , \
m
i
1 5 9 13 17 21 1
28 JUNE 1977
5 9 13 17 21 15 9 13 17 21 1
29 JUNE 1977 28 JUNE 1977
EASTERN DAYLIGHT TIME - hour
5 9 13 17 21 1
29 JUNE 1977
FIGURE 29 TIME HISTORIES OF OZONE CONCENTRATIONS IN NORTHERN NEW ENGLAND, 28-29 JUNE 1977
-------
U1
Ol
*l^?BBi /*'**&/ -3SUW
^f^^^^^^/»^^-*^o<«^^^
^rft^-JlBSV^-^SsSK
5>^i ^c-W-m^tV^ (]£*
S'^ii^^^fe^^
Saft^JSflSSBS^u^
(a) 28 JUNE 1977
(b) 29 JUNE 1977
FIGURE 30 WEATHER MAPS FOR 28-29 JUNE 1977
-------
MONTREAL
•
OTTAWA | 29 JUNE 1977
0200 I
so IPO ISP «n
0 9) 100 150 200 ISO 300 350
-------
360
270
•8
i
o
UJ
cc
5
o
90
- \
BURLINGTON
ALBANY
r
i
20
I
2 10
ui
a.
9 17
28 JUNE 1977
9 17
29 JUNE 1977
PORTLAND
9 17
28 JUNE 1977
9 17 1
29 JUNE 1977
EASTERN DAYLIGHT TIME - hour
FIGURE 32 TIME HISTORIES OF WINDS, 28-29 JUNE 1977
-------
taking place. Certainly, conditions for the previous 12 hours had been conducive to formation
and transport of ozone from the large metropolitan areas to the south. Concentrations began to
fall below the standard during the early morning hours of 29 June 1977, which probably meant
that the supply of ozone aloft had been depleted. The depletion of ozone could have come
about from the accumulated effects of destruction or because the ozone produced during the
day had been advected beyond the study area. National Weather Service analyses show that the
cold front did not pass the Burlington area until about 1200 EOT on the 29th, and Concord and
Portland until about 1700 EOT.
The increased afternoon concentrations associated with the usual diurnal cycle
were considerably suppressed on the 29th, but Keene did experience a short period of viola-
tions around 1500 on the 29th. However, both local and remote sources of ozone were likely
to have been reduced once the front had passed, since it was accompanied by an increase in
cloud cover and a shift in wind direction well away from source areas to the south.
All the features discussed earlier appeared to be present in this example. With
the reasonably high wind speeds that were observed, it is probable that most of the ozone con-
centrations were the product of precursors advected from remote sources, rather than local pro-
duction.
ii 15-16 June 1976
This was a period when violations of the standard began around noon (IS June
1976) and remained high until about 0400 EOT the next morning. Figure 33 shows the ozone
concentrations at sites with data during this period; the notable exception to the nighttime vio-
lations was Portland.
The daily weather maps (Figure 34) show the changes of meteorological
influence, beginning with the south-southeasterly winds caused by a high-pressure system to the
southeast, and continuing on the 16th in the warm sector ahead of the advancing cold front that
passed through the area early on the 17th. The surface wind data in Figure 35 show fairly
steady southerly wind directions at Burlington throughout the 15th and 16th. Directions at
Concord and Portland were a little more variable, tending more to the SW on the 15th and
more to the S on the 16th. Wind speeds were reasonably high (8 to 15 knots) on the 15th,
dropping slightly during the early morning hours of the 16th, but then picking up sharply again
toward noon.
Again the prime conditions required for high nighttime ozone concentrations
were met; on 15 June 1976 there was a period of time suitable for the production and transport
of ozone, and vertical mixing was present, as indicated by wind speeds that remained fairly high
at night. Concentrations decreased after midnight at all sites but Burlington and Portland. The
extended period of low concentrations at Portland, on the 15th was probably related to the
offshore wind direction. As the direction became more southerly (and developed an onshore
component) on the 16th, concentrations began to rise, and there was a period when concentra-
tions exceeded 80 ppb around 1400 EOT. The importance of offshore and onshore winds ^ill
be discussed later in greater detail in connection with the sea breeze. The decrease in concen-
trations from 1900 on the 16th was consistent between all the sites and is probably the result of
the formation of a stable layer at the ground in the early evening.
As with the previous example, the observations point to more remote sources
of ozone than to any ozone that was locally produced.
58
-------
100
50
BURLINGTON
\
SOppb
100 -
50 -
•a
Q.
I
z
2
oc
LU
o
I
Ul
1
O
_ MANCHESTER
'""v~\ /-•
i I i i .
..... /"~\ ,-, A -
\m-y - v v
x^' \
100
50
NASHUA
. ,
'
\
V
\
100
50
100
50 -
PORTLAND
-SOppb-
A
-^•^'- -"'' \/
I 1 1 1
GRAFTON COUNTV .- >,
1 1 1
1 1
.-N. S^*~ ~~ \
""*" '^.'' \
1 '
8 16 0 8 16
15 JUNE 1976 16 JUNE 1976
EASTERN DAYLIGHT TIME - hour
FIGURE 33 TIME HISTORIES OF OZONE CONCENTRATIONS AT NORTHERN
NEW ENGLAND SITES. 15-16 JUNE 1976
59
-------
ON
' o
(a) 15 JUNE 1976
(b) 16 JUNE 1976
(c) 17 JUNE 1976
FIGURE 34 WEATHER MAPS FOR 15-17 JUNE 1976
-------
360
270
I 180
UJ
cc
5
§ 90
PORTLAND
,S'\ / '-' \"—^.
•^ /^ v \ ,-^:
CONCORD
20
a is
e
g 10
UJ
a.
C/9
a
2 5
5
,
v y x^i r.nttr*r\or\
•\^_. -: /\r-
\/-\ _. / A
13
17
21
15 JUNE 1976
5 9
16 JUNE 1976
13
17
EASTERN DAYLIGHT TIME - hour
FIGURE 35 TIME HISTORIES OF WINDS, 15-16 JUNE 1976
-------
b. Sea-Breeze Effects
Many of the processes discussed in the preceding sections have been of a fairly large
scale, often involving transport over distances of 100 km or more. In this section, one of the
smaller-scale phenomena will be discussed-- the effects of the sea breeze circulation at coastal
locations. Unfortunately, data suitable for a comprehensive analysis of the effects as they are
observed in the Northern New England States are few. However, the phenomenon has been
studied in detail in other shore locations and those studies can serve by analogy to identify the
important factors that may be operative in Northern New England.
Figure 36 (from Lyons and Cole, 1976) describes schematically the behavior of pol-
lutants in the vicinity of a shorelne. The top half of the figure shows the behavior at about the
time of the morning rush hour. The precursor pollutants are carried out over the nearby ocean,
but have been separated from the surface by a very stable layer. This stable layer effectively
inhibits mixing and prevents the precursors, or the ozone formed from them, from being
mixed to the surface. The insolation and photochemical processes continue and the concentra-
tions of ozone build. At the same time, the land is heating relative to the water. This produces
a thermally induced landward circulation.
The air returning to the land in the afternoon is laden wih precursors and the ozone
that has formed from them. According to Lyons and Cole (1976) the air is intercepted by a
thermal internal boundary layer (TIBL) as it moves inland. More and more ozone is mixed
downward as the air moves inland; eventually all the ozone is mixed into the lower layer. This
process causes the highest concentrations to be found some distance inland from the shore.
Lyons and Cole (1976) suggest that the maximum occurs a few kilometers inland. The effects
of increased ozone entrainment will be modified by ozone destruction when the city is at the
shore. The city's emissions of NO will remove much of the ozone near the shore and perhaps
for a limited distance farther inland. This will accentuate the inland maximum.
It should be understood that the description above and especially the schematic
diagrams in Figure 36 deal only with the components at right angles to the shoreline. There
may also be motion parallel to the shore, so that the precursor emissions may leave the city,
pass over the water, and return onshore at some other place. Lyons and Cole explained some
observed ozone concentrations in southern Wisconsin as having been derived from emissions
originating in Chicago.
The question arises, does the mechanism above, which was used to describe condi-
tions on the western shore of Lake Michigan, also apply on the Northern New England Coast.
Zeller et al. (1977) observed similar effects in the Boston region. Another important finding of
their research was that the ozone "plume" from Boston was observable for long downwind dis-
tances. Concentrations in excess of 80 ppb were found aloft as far as 200 km downwind. In
principle, this plume might be brought onshore somewhere downwind and mixed groundward if
the conditions were right. Ludwig and Shelar (1977) have pointed out that winds from the
southwest are frequently associated with high ozone concentrations in southern New, England,
so transport to the north-northeast along the coast, with subsequent onshore sea breeze flow,
could produce high ozone concentrations at locations near the coast. It needs to be shown that
high ozone concentrations are associated with onshore flow, especially when the air is moving
from a southerly direction. It also needs to be demonstrated that high ozone concentrations
can be transported to the coastal areas of New Hampshire and Maine.
We have examined ozone observations and wind direction measurements from a
special study conducted during the latter part of July and most of August, 1974 (Londergan and
-62
-------
MORNING RUSH HOUR ~8am
DEEPENING
MIXED
LAYER
MID AFTERNOON ~ 2pm
°3 CONCENTRATIO
RAPID O3
\
DESTRUCTION\
LITTLE 03
FUMIGATION ZONE
Source: Lyons and Cole (1976)
FIGURE 36 SCHEMATIC DEPICTION OF SUMMER SEA BREEZE CIRCULATION
Polgar 1975), and while there is some evidence of a sea breeze effect, it is not wholly con-
clusive On five days during the month of operations, ozone concentrations in excess of 80 ppb
were observed at one or more of three stations that were monitoring ozone in the vicinity of
Portland Maine. A total of nine hours were involved. During eight of those hours, an
onshore wind with a southerly component was observed. In each of the cases there had also
been a period during the preceding morning when offshore winds were observed.
A similar analysis was performed using the routine Portland wind and ozone data for
1976 There were 16 afternoons or early evenings during which ozone concentrations were
greater than 80 ppb. In all but one of the cases, the winds were onshore (from 40 degrees to
200 degrees) with a southerly component during most of the period of violation In fact about
85 percent of the 80 hours involved fell in the onshore flow category Of the 15 days when the
afternoon ozone violations occurred with onshore flow, 14 had had offshore flow earlier in the
day.
Spicer et al (1977) have presented data showing conclusively that ozone at high
concentrations can be transported to an area where the sea breeze could advect it to Oara. Fig-
ure 37 was taken from their report. It shows the ozone concentrations during the afternoon of
24 July 1975 at an altitude of about 300 m (1000 feet). According to the authors, the origins
of the high ozone concentraions observed off the coast of Maine were emissions from Philadel-
63
-------
Source: Spicer et al. (1977)
FIGURE 37 OZONE CONCENTRATIONS AT AN ALTITUDE OF 300 m,
24 JULY 1975
phia and New York on the preceding day. These earlier emissions and the ozone that arose
from them were augmented by later emissions and more ozone from the Boston area on the
morning of the day the observations were made.
It has not been possible to draw detailed wind fields and ozone concentration pat-
terns for any specific cases, due to the lack of data. Therefore, it is not possible to show the
sea breeze effects as conclusively as Lyons and Cole (1976) and Zeller et al. (1977) were able
to do. However, it can be said with confidence that the winds at Portland are such that a major-
ity of oxidant standards violations could involve sea breeze effects, but this does not necessarily
mean that they actually do. It has also been shown that it is possible to transport large amounts
of ozone at low altitudes over water to the coast of Maine where it would be in a position to be
moved inland. Again, only the potential has been demonstrated. Although unproven, it
appears quite probable that transport over water and sea breeze effects account for at least some
violations in coastal Maine and New Hampshire.
64
-------
c. Diurnal Patterns of NAAQS Violations
The mechanisms associated with high nighttime ozone concentrations and with sea
breeze effects have at least one very important thing in common. They both involve ozone that
is isolated from the surface in or above a stable layer. In the nighttime cases, the stable layer is
produced by the radiative cooling of the surface after sundown. Sea breeze effects involve pas-
sage over cold water where the air also becomes very stable by being cooled from below.
The trapping of ozone aloft in a stable layer is very important for two reasons. The
first of these is the isolation from surface destruction, which serves to "hermetically" seal off
the ozone and preserve it. This effect could be offset if the ozone-rich air were mixed with
cleaner air from above so that the concentrations were substantially diluted. However, the inhi-
bition of vertical mixing in the stable layer that isolated the ozone from surface destruction can
also reduce dilution by vertical mixing with air above.
' *
The two phenomena differ in the time of day during which they are most pro-
nounced. The sea breeze effect is a daytime phenomenon—the sea will be cooler than the
overlying air, and for closely related reasons the air will flow onshore almost exclusively during
the daytime hours. When the air remains over land it can be stabilized by radiative cooling of
the underlying surface, which occurs only during the late afternoon and nighttime hours. Thus,
we would expect high ozone concentraions to be more frequent at night at inland sites than at
coastal locations. Figure 38 shows that such is the case.
Figure 38 displays the relative frequency of occurrence of ozone concentrations of
80 ppb or greater at different times of day. The data from three inland sites-Burlington, Man-
chester, and Nashua-were grouped together as were those from two coastal sites-Portland and
Portsmouth. As discussed in Appendix A, these sites have all been ranked as having compar-
able levels of local urban influence, so the differences are believed to represent the coastal
versus inland effects. Both sets of sites show the greatest frequency of high ozone concentra-
tions during the afternoon hours. However, the frequency drops more rapidly and to appreci-
ably lower levels with the onset of night at the coastal sites as compared to those inland.
As one might expect, the diurnal effect is more pronounced at the coastal sites
where the photochemical processes, the typical mixing patterns, and the sea breeze effects all
work together to emphasize afternoon ozone maxima. At inland sites the sea breeze is missing
and the occurrences of high nighttime ozone concentrations due to long-range transport and
nighttime mixing assume greater relative importance.
It should be noted that an island site or a site on a sharp point of land may behave
much differently than ordinary coastal sites. The monitor on a point or island will often be in
the midst of the general flow and will not have to depend on the sea breeze to bring transported
ozone onshore. Furthermore, the sea might be warmer than the overlying air at night, so there
would be greater vertical mixing than over land. High concentrations at night might be more
frequent on a point than at an onshore site and the hypothesized daytime stable layer at the
surface of the sea would reduce the afternoon frequency of high concentrations over the ocean.
According to Mr. David Dixon of the Maine Department of Environmental Protection, Cape
Elizabeth observed nighttime concentrations in excess of 80 ppb more frequently in 1977 than
are shown for the coastal sites in Figure 38 and afternoon frequencies were less at the Cape
Elizabeth site than the onshore sites. These more recent data are consistent with the
hypothesized over water transport mechanism.
65
-------
LU
a
in
oc
li.
HI
1
ai
{£
EASTERN STANDARD TIME - hour
FIGURE 38 RELATIVE FREQUENCIES OF NAAQS VIOLATIONS AT DIFFERENT TIMES OF DAY
FOR SELECTED COASTAL AND INLAND SITES
66
-------
C. Weekday Versus Weekend Ozone Concentrations
1. General
Although nature has its cycles, those cycles may differ enough from the cycles of
human activity so that they can be used to differentiate between natural and anthropogenic
causes of high ozone. Daily and annual cycles would seem to be poor choices because both
man and nature exhibit patterns of behavior with 24 hour and 365 day periods. However,
man's habit of reduced weekend activity is not imitated by nature, so differences among the
average patterns of pollutant behavior among weekdays, Saturdays, and Sundays are likely to
reflect man's influence.
In the case of the photochemical oxidants, the effects of man's activities are some-
what paradoxical. Oxidant concentrations can be decreased or increased by anthropogenic emis-
sions, especially NO. The immediate effect of NO is to destroy ozone, so the commonly
observed "weekend effect" (see, e.g., Cleveland et al., 1974) in cities is to produce significantly
lower ozone concentrations on weekdays than on weekends. Eventually, the same emissions
that initially reduce ozone concentrations will result in increased concentrations somewhere
downwind.
The behavior described in the preceding paragraph suggests that areas without
important emissions might exhibit a reverse weekend effect if they were appreciably affected by
ozone produced from imported anthropogenic emissions. However, such effects might be
difficult to demonstrate statistically. The "weekend effect" at a site surrounded by emissions is
independent of wind direction. However, the only cases that will contribute to the differences
between weekdays and weekends at a remote site are those that occur when the monitor is in
the urban plume. One is faced with the problem that the population of cases that demonstrate
the effect is embedded in a larger, more homogeneous population of cases with no urban
influence. Thus, it is much more difficult to use weekday/weekend differences to show that the
major influences at a site are from some remote area than it is to use the effect to show that
there are important local sources.
2. Observations
Mean hourly ozone concentrations were plotted for each day of the week at selected
sites, for broad seasonal periods. These are shown in Figures 39 through 41. Marked
differences are apparent at Manchester and Burlington. The differences show that Saturdays and
Sundays had somewhat higher concentrations than Mondays through Fridays for most of the
24-hour period. Other sites, for example Nashua, tend to show the reverse to be true. The ini-
tial observations lend some weight to the arguments outlined earlier. Areas having relatively
large local emissions of HC and NO the precursors to photochemical production of ozone,
seem to experience higher ozone concentrations on weekends than weekdays. This would be
due to the scavenging effect HC and NOX have on ozone.
The observed differences between weekdays and weekends have been tested for sta-
tistical significance using the t-distribution. Differences between (1) Sundays versus Mondays
through Fridays (2) Saturdays versus Mondays through Fridays, and (3) Sundays versus Satur-
days were tested at each site for two main periods- summer (March through September) and
winter (October through February). The results of the tests for both the daily mean and the
daily maximum of the hourly-averaged ozone concentrations are shown in Tables 7 and 8.
Although statistical significance was rare, the patterns of the results, particularly the tendencies
67
-------
OV
00
50
40
30
20
10
5 60
iu
u
8 50
ui
§ 40
30
20
10
(a) MARCH - MAY 1976
Saturday
Friday
Tuesday
8 12
TIME*- hour
16
(b) JUNE - SEPTEMBER 1976
Sunday
Saturday
(e) MARCH - MAY 1977 /
Sunday //ff I Monday
8 12
16
TIME - hour
* (a), (b): Eastern daylight
(c): Eastern standard
20 24
20 24
FIGURE 39 DIURNAL VARIATIONS OF AVERAGE OZONE CONCENTRATIONS FOR EACH DAY
OF THE WEEK AT MANCHESTER, N. H.
-------
VO
60
50 -
I 1 1
(b) JUNE - SEPTEMBER 1976
(a) MARCH - MAY 1978
TIME - hour
-------
-d
O
I 1 1 1 1
- (•) MARCH - MAY 1976
h (b) JUNE - SEPTEMBER
1976
Sunday - V //.
* (a), (b): Eastern daylight
(c): Eastern standard
20 24
TIME - hour
FIGURE 41 DIURNAL VARIATIONS OF AVERAGE OZONE CONCENTRATIONS FOR EACH DAY
OF THE WEEK AT BURLINGTON, VERMONT
-------
Table 7
DAILY MEAN OZONE CONCENTRATIONS—WEEKDAYS AND WEEKENDS
Site
Manchester
Nashua
Burlington
Portland
Berlin
Day of
Week
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
"Winter"
Mean Difference Significance
6.5 0.8
5.7
6.3 0.6
5.7
6.5 0.2
6.3
6.7 -0.1
6.8
9.6 2.8
6.8
6.7 -2.9
9.6
19.1 2.0
17.1
17.8 0.7
17.1
19.1 1.3
17.8
13.5 3.1 5X
10.4
11.1 .7
10.4
13.5 2.4
11.1
6.9 -0.6
7.5
6.5 -1.0
7.5
6.9 0.4
6.5
"Sunnier"
Mean Difference Significance
25.8 5.4 1%
20.4
25.0 4.6 IX
20.4
25.8 0.8
25.0
22.0 -1.7
23.7
23.2 -0.5
23.7
22.0 -1.2
23.2
34.8 5.5 n
29.3
31.7 2.t
29.3
34.8 3.1
31.7
22.6 1.3
21.3
24.1 2.8
21.3
22.6 -1.5
24.1
7.7 0.1
7.6
7.0 -0.6
7.6
7.7 0.7
7.0
March - May 1976
Mean Difference Significance
25.6 11.6 0.12
14.0 -
22.1 8.1 O.U
14.0
25.6 3.5
22.1
26.2 8.9 0.1Z
17.3
24.6 7.3 12
17.3
26.2 1.6
24.6
31.9 8.7 5*
23.2
29.4 6.2
23.2
31.9 2.5
29.4
31.3 10.7 „ 0.1X
20.6
26.4 5.8
20.6
31.3 4.9
26.4
9.1 2.4
6.7
7.1 0.4
6.7
9.1 2.0
7.1
March - May 1977
Mean Difference Significance
30.6 2.3
28.3
32.4 4.1
28.3
30.6 2.3
28.3
17.5 -10.9
28.4
18.0 -10.4
28.4
17.5 -0.5
18.0
38.7 0.4
38.3
34.9 -3.4
38.3
38.7 3.8
34.9
31.1 0.4
30.7
27.5 -3.2
30.7
31.1 3.6
27.5
-------
Table 7 (Concluded)'
Site
Grafton
County
Deerfield
Keene
Portsmouth
White River
Junction
Day of
Week
Sunday
Weekday
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
"Winter"
Mean Difference Significance
21.8 -5.0
26.8
23.4 -3. A
26.8
21.8 -1.6
23.4
"Summer"
Mean Difference Significance
13.4 -1.6
15.0
11.3 -3.7
15.0
13.4 2.1
11.3
21.6 1.9
19.7
22.1 2.4
19.7
21.6 -0.5
22.1
26.9 -.2
27.1
31.2 4.1
27.1
26.9 -4.3
31.2
33.4 0.6
32.8
30.8 -2.0
32.8
33.4 2.6
30.8
32.8 3.2
29.6
30.5 0.9
29.6
32.8 2.3
30.5
March - May 1976
Mean Difference Significance
March - May 1977
Mean Difference Significance
33.7 -0.1
33.8
28.3 -5.5
33.8
33.7 5.4
28.3
36.7 4.3
32.2
31.4 -0.8
32.2
36.7 5.3
31.4
ho
-------
Table 8
AVERAGE DAILY MAXIMUM OZONE CONCENTRATION—WEEKDAYS AND WEEKENDS
Site
Manchester
Nashua
Berlin
Foreland
Burlington
Day of
Week
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
"Winter"
Mean Difference Significance
17.8 2.6
15.2
17.1 1.9
15.2
17.8 0.7
17.1
15.1 -1.5
16.6
22.8 6.2
16.6
15.1 -7.7
22.8
14.0 -3.2
17.2
13.7 -3.5
17.2
14.0 0.3
13.7
31.5 9.2 0.1%
22.3
24.1 1.8
22.3
31.5 7.5 1%
24.1
30.3 3.3
27.0
28.6 1.6
27.0
30.3 1.7
28.6
"Summer"
Mean Difference Significance
46.2 4.4
41.8
46.0 4.2
41.8
46.2 0.2
46.0
40.6 -5.4
46.0
39.0 -7.0
46.0
40.6 1.6
39.0
14.7 2.3
12.4
15.5 3.1
12.4
14.7 -0.8
15.5
37.0 -4.2
41.2
45.0 3.8
41.2
37.0 -8.0
45.0
51.0 4.4
46.6
49.0 2.4
46.6
51.0 2.0
49.0
March - May 1976
Mean Difference Significance
39.6 11.5
28.1
38.3 10.2
28.1
39.6 1.3
38.3
38.0 6.8
31.2
39.0 7.8
31.2
38.0 -1.0
39.0
16.1 0.6
15.5
17.7 1.6
15.5
16.1 -1.6
17.7
43.9 8.1 ,
35.8
44.8 9.0
35.8
43.9 -0.9
44.8
44.0 8.4
35.6
44.7 9.1
35.6
44.0 -0.7
44.7
March - May 1977
Mean Difference Significance
53.0 -0.2
55.2
55.4 2.2
53.2
53.0 -2.4
55.4
41.0 -14.0
55.0
28.9 -26.1
55.0
41.0 12.1
28.9
47.6 -3.0
50.6
42.1 -8.5
50.6
47.6 5.5
42.1
58.3 2.0
56.3
51.1 -5.2
56.3
58.3 7.2
51.1
U)
-------
Table 8 (Concluded)
Site
White River
Junction
Grafton
County
Keene
Portsmouth
Deerfield
Week
Sunday
Weekdays '
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Sunday
Weekdays
Saturday
Weekdays
Sunday
Saturday
Mean Difference Significance
35.1 -4.8
39.9
38.7 -1.2
39.9
35.1 -3.6
38.7
Mean Difference Significance
51.7 -3.4
55.1
53.2 -1.9
55.1
51.7 -1.5
53.2
23.2 -4.1
27.4
24.5 -2.9
27.4
23.2 -1.2
24.5
44.0 -2.9
46.9
48.4 1.5
46.9
44.0 -4.4
48.4
57.3 -0.4
57.7
55.9 -1.8
57.7
57.3 1.4
55.9
34.8 -3.0
37.8
38.6 0.8
37.8
34.8 -3.8
38.6
1
March - Hay 1976
Mean Difference Significance
March - Hay 1977
Mean Difference Significance
56.4 -2.2
58.6
53.8 -4.8
58.6
56.4 2.6
53.8
54.5 -3.3
57.8
50.4 -7.4
57.8
54.5 4.1
50.4
-------
for Saturdays and Sundays to have greater or lesser ozone concentrations than Mondays
through Fridays, seems to be at least consistent with much of the reasoning above. The follow-
ing points may be extracted from Tables 7 and 8:
(1) Statistically significant differences were found at sites in the main urban
centers-Burlington, Manchester, Portland. During the summer period, differences between
mean daily values of Sunday and those of Monday through Friday were significant for Manches-
ter and Burlington; also, the difference between means for Saturday versus Monday through
Friday was significant at Manchester.
(2) Portland was the only site to show significant differences of both mean and max-
imum values during the winter period.
(3) t-test probabilities for most of the other sites were high, particularly for the
summer months. Though differences were small, Sundays and Saturdays tended to have lower
mean concentrations than Mondays through Fridays. The differertees were most pronounced at
Nashua, Berlin, and the Grafton County sites.
(4) The differences between t-test probabilities for daily mean and maximum
hourly concentrations at each site are difficult to generalize. However, the significance of max-
imum hourly differences was greater than that of daily mean values during the winter months.
In addition to the broad summer/winter analyses, differences were tested for the
March through May period of both 1976 and 1977. This is the period of year when natural
affects should be a maximum and when anthropogenic effects should be increasing. The results
for these months (also given in Tables 7 and 8) reflect the effects of reducing the sample size,
especially for the period March through May 1976, which was characterized by a number of
high ozone periods produced by unusual synoptic conditions, that were unevenly distributed
among the days of the week. Tables 7 and 8 show that at all the sites with data during the
period March through May 1976, daily mean and maximum hourly ozone concentrations on
Saturdays and Sundays were greater than those for Mondays through Fridays. The effects were
widespread. The differences observed at both Berlin and, more spectacularly, at Nashua were
quite different from those observed during other time periods.
The March through May 1977 results shown in Tables 7 and 8 are more in line with
the results for the longer-period analyses— i.e., Burlington and Manchester show a "weekend"
effect (though not significant) but Nashua and Berlin show a tendency toward the reverse of the
usual urban weekend effect.
It appears that in some of the small subsamples of data, that certain synoptic
meteorological patterns were not distributed randomly among the days of the week. Given the
influence that synoptic events can have on ozone concentration, this could easily mask any 7-
day cycles associated with human activity. The longer the period available for analysis, the
better "average" conditions may be approximated. Thus the results for sites with small data sets
(Deny, Keene, Portsmouth) should be interpreted with great caution.
75
-------
V SUMMARY AND CONCLUSIONS
A. Local Versus Imported Oxidants
The evidence seems to leave little doubt that ozone produced from emissions generated
outside the three Northern New England states, and subsequently transported into those states,
represents a very significant part of their air quality problems. The general areas responsible for
the contributed emissions are identifiable, but solid quantitative estimates of the local versus
the imported contributions cannot yet be made. However, even if quantitative estimates were
available, their usefulness in the formulation of control strategies would be quite limited by
uncertainties about the future actions of neighbor states and the evolving status of photochem-
ical air quality models.
The trajectory analyses and the wind analyses both suggest that the southern parts of Ver-
mont, New Hampshire, and possibly Maine are often affected by emissions from the large
urban areas toward the southwest. Conditions associated with photochemical production of
ozone often include winds blowing from the southwest. Another general area from which
emissions seem to come during periods when the NAAQS are violated is the region more to
the west of the Northern New England States- i.e., the industrial areas around Lakes Ontario
and Erie. The northern parts of Vermont and New Hampshire are affected by these areas more
frequently during periods of high ozone concentrations than they are by the more southerly
source areas. Trajectory analysis of air arriving at Burlington, Vermont suggests that Montreal
and perhaps the area north of Lake Huron can also contribute.
There were no ozone data from the northern parts of Maine, so it was not possible to
identify periods of high concentration and to construct trajectories. The analysis of the wind
directions and patterns associated with ozone concentrations in other parts of the Northeastern
United States suggests that Montreal and the industrial areas around Lakes Ontario and Erie are
the most likely areas to contribute to any air quality problems that might be found in northern
Maine, if that region actually has any problems. This association is made because the data
show that the westerly and southwesterly winds are most commonly associated with violations
of the NAAQS for oxidant in other parts of the three Northern New England States.
The analysis of sea breeze effects suggested that coastal waters of Northern New England
may serve as an effective avenue for the transport of ozone. The air is cooled from below as it
passes over the sea surface, causing it to become more stable in the lower layers. This, in turn,
inhibits dilution and isolates the ozone from destructive processes at the surface. During the
afternoon, the preserved ozone that has been transported along the coast from its areas of ori-
gin is brought inland by the sea breeze and mixed to the surface by the turbulence within the
thermally induced internal boundary layer over land areas near the coast. The data have shown
this mechanism to be possible, but the data are not adequate to determine its importance.
However, considering what is known about the winds and weather patterns that go with high
oxidant concentrations in the Northeast, it seems likely that over-water transport and sea breeze
advection are very important factors in the determination of the air quality of the coastal areas
of Northern New England.
Some local effects are evident in the data from a few of the sites. The average weekend
ozone concentrations at some of the sites are significantly greater than average values during
the week. This is the result of the scavenging of ozone by local emissions of NO. The site
descriptions given in Appendix A indicate that this localized effect is consistent with the degree
of urban influence on the stations. However, there is little to suggest that these urban
77
-------
influences will extend very far. In the next section, the extent of the urban influence is
explored, because of its impact on control policies.
B. Implications for Control Strategies
1. General
A control strategy requires the specification of what is to be controlled, how much it
should be controlled, and where and when it should be controlled. What is to be controlled
and how much it should be controlled are dependent on what pollutants are already present. A
highly simplified modeling approach has recently been introduced by the OAQPS (1977b, 1978)
to describe the behavior of photochemical pollutants. The Empirical Kinetic Modeling
Approach (EKMA) provides a useful tool for evaluating the impact of possible control stra-
tegies in the Northern parts of New England. It is especially useful to the analysis of local con-
tributions in the more urbanized areas such as Portland, Maine and the Southern New
Hampshire counties where the concentrations of hydrocarbon and NO emitters are the
greatest.
The basic tool of the EKMA is an isopleth chart relating the sensitivity of ozone
concentrations to changes in precursor concentrations. The report (OAQPS, 1977b) recom-
mends that a separate isopleth chart be devised for each urban area of interest so that local
differences in hydrocarbon mix and meteorological factors are incorporated into the system.
However, when the appropriate measurements are not available, as they are not for locations in
Northern New England, then a standardized graph like that in Figure 42 is used. Examination
of that graph shows that concentrations of ozone produced from a mixture of hydrocarbons and
oxides of nitrogen are sensitive to the relative amounts of those two classes of pollutants that
are present. It is apparent that hydrocarbon controls are likely to be more effective for those
mixtures of hydrocarbons and NO represented by the upper-left half of the diagram. When
NO concentrations are low, and the lower-right half of the figure is used, then NO controls
are likely to be much more effective than hydrocarbon controls. The next section examines the
question of where high NO concentrations are likely to be found, and hence where hydrocar-
bon control strategies will be most effective.
2. The Extent of Urban Influence on Oxides of Nitrogen Concentration
Some studies have indicated that the ozone concentrations in rural regions are more
influenced by the upwind anthropogenic emissions of oxides of nitrogen than by hydrocarbon
emissions (e.g., Meyer et al., 1976; Ludwig et al., 1977a). Singh et al. (1977) suggested that
the photochemical production of ozone in remote areas was limited by the availability of NOX>
The empirical evidence and the relationships displayed in Figure 42 underly this suggested
explanation. Figure 8, which was discussed earlier, shows schematically how the concentrations
of ozone might increase in remote areas if some oxides of nitrogen were introduced. If tlie
assessment of these studies is accurate, then the extent to which NOX from urban areas reaches
into the surrounding countryside might serve as a guideline for estimating the extent of urban
influence on ozone concentrations.
The EPA's Office of Air Quality Planning and Standards (OAQPS, 1977a) has recog-
nized the importance of NO to the photochemical production of ozone in non-urban areas.
They have examined the aerial extent to which urban areas cause measurable concentrations
(defined as 7 ppb) of oxides of nitrogen. Figure 43, based on an illustration in the OAQPS
78
-------
to
0.4 0.6
0.8 1.0 1.2
NMHC - ppmC
1.4
1.6 1.8
Source: OAQPS, (1978)
FIGURE 42 SENSITIVITY OF MAXIMUM AFTERNOON CONCENTRATIONS TO MORNING PRECURSOR LEVELS
MEASURED UPWIND
-------
(1977a) report, relates population of an urban area to its radius of influence. Within the radius
defined by Figure 43, widespread hydrocarbon control strategies are likely to be effective in the
reduction of ozone concentrations. Figure 44, also derived from the OAQPS (1977a) report,
shows the areas in the U.S. that are influenced by urban NO emissions. It is evident that,
based on the criteria used by OAQPS, there are not any large areas outside the cities and towns
of northern New England where NO concentrations are likely to be appreciable.
A
Some caution should be exercised in the interpretation of Figures 43 and 44. They
provide only an estimate of the regions within which hydrocarbon emission controls might be
most effective in reducing photochemical oxidant production. The areas do not really have
sharp boundaries. The relationship between population and concentration is uncertain and the
the 7 ppb limit on NO concentration is arbitrary. Finally, the ozone that is produced from
emissions within the designated areas may be transported well beyond those areas. This is an
important point. It means that ozone standard violations can occur in the unshaded areas, but
the control of such violations will have to be carefully considered. It may require limitations of
hydrocarbon emissions within one or more of the shaded areas or of oxides of nitrogen and
hydrocarbon emissions outside those areas.
3. The Relative Importance of Transported and Locally Generated Ozone
Calculations have been made with the EKMA that indicate that locally generated
ozone concentrations are not increased by imported ozone in a linear fashion. In general, the
amount of increase attributable to imported ozone will be half the imported concentrations or
less, but the degree to which the transported ozone contributes depends on what meteorological
and air quality conditions prevail. Table 9, from the OAQAPS (1977b) report, shows qualita-
tively how the additivity of the transported ozone depends on other factors. The various factors
in the table are listed in decreasing order of importance.
The first factor in Table 9 is a meteorological one and may vary from locale to locale
within Northern New England, but in general the dilution rates should be fairly high. Holz-
worth (1972) shows that summer afternoon mixing depths tend to be three or four times those
in the morning for the Northern New England states. If the change from the morning height
to the afternoon height occurs over a period of about 9 hours, then the average dilution rate is
13 percent per hour for a total change by a factor of 3. Thus, relatively high additivity would
be expected in Northern New England on that basis.
Even the largest urban areas of Northern New England are relatively small, so the
second factor listed in Table 9 also points toward greater influence from imported ozone. This
is also true for the fourth factor listed in the table because air parcels leave small cities very
quickly except under very stagnant conditions. Once the air parcel has left the city, the contri-
bution from later emissions decreases dramatically.
A few data are available for Portland, Maine showing the average ratios of
NMHC/NO for local sources and for the air transported into the area. According to the table
given by OAQPS (1978), based on data from Londergan and Polgar (1975), the average ratio
was 11.1 when the monitor was downwind of the source and about 31 when it was upwind.
This factor should minimize the effects of transported ozone, unless of course severe hydrocar-
bon control measures reduced the ratio. A reduction of the ratio would tend to make the
imported ozone fraction more significant.
In summary, it appears that in the vicinity of the urban areas of Northern New Eng-
land (e.g., Portland and Portsmouth) about half the transported ozone would be additive to any
80
-------
4000
3500
3000
•8
£ 2500
O
F
a.
O
a.
ui
IT
N
Z
ea
a:
2000
1800
1000
500
200
For urban areas with populations
greater than 4 million, radius of
influence is about 140 km
20
60
80
100
120
140
ESTIMATED RADIUS (tan) WHERE NOX < 7 ppb
Source: OAQPS, 1977 a
FIGURE 43 ESTIMATED RADIUS AT WHICH NOX CONCENTRATIONS FALL
BELOW 7 ppb, AS A FUNCTION OF METROPOLITAN POPULATION
81
-------
oo
ro
\ •
FIGURE 44 AREAS APPROPRIATE FOR HYDROCARBON EMISSION CONTROLS ACCORDING TO OAQPS (1977a)
-------
Table 9
QUALITATIVE IMPACT OF VARIOUS FACTORS ON THE ADDITIVITY OF
TRANSPORTED OZONE TO MAXIMUM OZONE CONCENTRATIONS IN URBAN AREAS*
Factor
Factor Value
Additivity
Dilution Rate (i.e., the
extent and rate at which
the diurnal mixing depth
increases)
Quantities of locally
emitted precursors
NMHC/NO.. Ratio
Importance of post 9 a.m.
emissions (This reflects
both diurnal emission patterns
and the larger atmospheric
dilution capacity which
generally occurs during the
mid-morning and afternoon.)
Relatively High
(e.g., >13%/hour)
Relatively Low
(e.g., small city
^200,000)
Relatively Low
(e.g., ' 6:1)
Relatively High
(e.g., significant
NO emissions in the
afternoon such as
would occur if an
air parcel remained
within the city
limits in the after-
noon during a stag-
nation period)
Relatively high
(> 0.45)
Relatively high
(> 0.45)
Relatively high
(> 0.45)
Relatively low
(< 0.45)
Source: OAQAPS (1977b).
83
-------
that might be generated locally. In the more remote regions such as areas in Vermont, the con-
tribution would be relatively greater. Presumably, in an area free of local influences, nearly the
entire transported amount would be observed, although some reduction would take place
because of dilution and destruction at the surface.
4. An Example of How the EKMA Might be Used to Devise
Control Strategies in Northern New England
Data summaries that were received as this final report was being written indicate that
the second highest concentration in the southeastern parts of the region of interest was around
200 ppb--225 ppb at Cape Elizabeth near Portland, Maine and at Portsmouth, N.H.; and 160
ppb at Portland itself. For purposes of illustration we might consider the case of Portland with
its 160 ppb concentration. If we assume that imported ozone amounted to about what was
observed on the day illustrated earlier in Figure 37—say, 140 ppb. Only half of this--70 ppb—
would be contributing to the 160 ppb observed in the area, leaving about 90 ppb from local
causes.
As noted earlier, the average NMHC/NO ratio that has been observed in the Port-
land area is about 11. It should be noted that the value of 11 for the NMHNO ratio is based
on very few observations and better information would be required to provide a more reliable
basis for the development of appropriate control strategies. According to Figure 42, morning
NMHC and NO concentrations of about 220 ppbC and 20 ppb, respectively, would have an
NMHC/NO ratio of 11 and would lead to the production of about 90 ppb ozone concentra-
tions.
How much would the morning NMHC concentrations have to be reduced to achieve
standards? Since the contribution from upwind sources would be greater than the 80 ppb
NAAQS even if all local emissions were stopped, it must be assumed that the controls applied
in upwind areas can reduce the ozone concentrations received in the Northern New England
states to some value nearer the general background of about 40 ppb. Let us assume that
upwind controls reduce imported ozone to 50 ppb, of which 25 ppb is additive to the locally
generated ozone. This means that the locally generated ozone must be reduced from 90 ppb to
about 55 ppb. If this were done through hydrocarbon controls alone, it would require a 55 per-
cent reduction in hydrocarbon emissions, to about 45 percent of the current amounts. If hydro-
carbons and NO emissions were reduced in the same proportions, keeping the NMHC/NOX
ratio the same, then each would have to be at about 55 to 60 percent of its current level.
If we had used the 225 ppb ozone value that was observed in Portsmouth (and at
Cape Elizabeth), the requirements for reduced emissions would have been more severe.
Assuming that Portsmouth also has an NMHC/ NOX emissions ratio of 11 and that imported
ozone might be slightly more concentrated, say 180 ppb (of which 90 ppb is additive) because
Portsmouth is closer to urban source areas, then the local production of ozone would be about
135 ppb. According to Figure 42, the corresponding NMHC and NO concentrations (assum-
ing the same NMHC/ NOX ratio of 11) would be about 400 ppbC and 36 ppb, respectively. If
we again assume that imported ozone can be reduced to 50 ppb, of which 25 ppb would be
additive, then the reduction of the local contribution from 135 ppb to 55 ppb (as would be
required to meet an 80 ppb standard) would mean a reduction of NMHC emissions to about 20
to 25 percent of current levels. If the standard were raised to 100 ppb as has been considered,
then NMHC emissions would still have to be reduced to about 30 percent of current levels.
The preceding cases are intended to serve as examples that illustrate how the EKMA
might be applied and to show that much of the information that is required for even so simple
84
-------
an approach as the EKMA is still not available. Although the examples use what are believed to
be reasonable numbers, they do not provide an adequate basis for control strategy development
Control strategies deserve more than "educated estimates" for their formulation. The most
important missing elements are: (1) ozone measurements free of local interferences, (2) reli-
able measurements of the imported ozone contributions, (3) good NMHC and NO measure-
ments during episode conditions and, (4) reliable forecasts of the control measures that will be
apphed elsewhere and their effectiveness in reducing the imported ozone contributions.
How the reductions in emissions might be achieved is beyond the scope of this
report. The approaches usually considered include mobile source controls (directly through
exhaust gas limitations or indirectly through reduced traffic), vapor recovery systems at gas sta-
tions, controls of vapor loss at bulk storage and processing plants, and so forth. To give some
idea of the magnitude of NMHC reductions available, the Federal motor vehicle emission con-
trol program should reduce the total emissions in Vermont by 41 percent by 1987 according to
Wishmski (1977). Control of oxides of nitrogen is much more difficult with presently available
technology.
It should be noted that in the rural areas the NMHC/NO ratios are apt to be very
high, and any increases in NOx emissions could lead to rather large increases in ozone concen-
trations, as indicated by Figures 8 and 42. Thus, while controls may not be necessary, the most
serious consideration should be given to the possible consequences of any introduction of new
sources of oxides of nitrogen.
C. Final Remarks and Recommendations
When the research reported here was begun, the use of EKMA was not foreseen. It cer-
tainly was not anticipated when the data that have been used in this report were collected. Had
it been, there could have been more emphasis on measurements that would define the relative
contributions of those urban areas within the three northern New England states.
Now that the data requirements for control strategy formulation are better defined, future
data collections can be made to suit the recognized needs. For example, the existence now of
monitoring stations at both Portland and Cape Elizabeth should aid future data analysts in
defining the local contributions of Portland. According to Tudor (1978), much more extensive
monitorng of ozone will be conducted throughout Maine during the summer of 1978. This will
help define conditions in those areas that are farthest from the major sources. It may help
define the extent to which transport of ozone to and from Canada is a problem. A good net-
work in Maine would also provide the coverage necessary to describe the effects of weather
fronts and nighttime transport in better detail.
In our opinion the most pressing remaining needs are for data that could be used to quan-
tify the over-water transport mechanism and for better descriptions of the sea breeze effects.
These would require another airborne monitoring program, but it would not have to be nearly
as large as that conducted in Southern New England during 1975, if the effort focused on an
investigation of coastal transport and sea breeze effects.
The existing data have provided the basis for reasonable explanations of many of the
observed phenomena. They have also provided hypotheses than can be tested with the data
soon to be collected. Finally, they have helped to define where shortcomings are likely to
remain.
85
-------
APPENDIX A
MONITORING SITE CHARACTERISTICS
87
-------
APPENDIX A
MONITORING SITE CHARACTERISTICS
Introduction
The descriptions of monitoring stations that follow were provided by Norman Beloin of
EPA Region I. During the course of these studies, we attempted to rank the various stations
according to the degree to which they were influenced by urban emissions. Four criteria were
used: population, nearness to the downtown area, nearness to a street, and traffic on the
nearest street. The rankings for the individual criteria and the overall rankings are summarized
in Table A-1.
Table A-l
RANKING OF LOCAL URBAN
INFLUENCE AT MONITORING SITES
Portland
Burlington
Portsmouth
Manchester
Nashua
Berlin
Keene
White River Jet.
Deerfield
Fraconia Notch
Population
2
4
5
1
3
7
6
8
9
10
Distance
from Road
3
1
4
.5%
2
5k
7
8
9
10
Traffic
1
2%
5
5
7
2k
5
8
9
10
Distance from
Town Center
2
4
2
5
8
6%
6%
2
9
10
|
Overall
1
2
3
4
5
6
7
8
9
10
89
-------
Station Descriptions
BURLINGTON, VERMONT. The probe is located 15 feet above the ground at a mobile
trailer site within 20 feet of a two lane surface street. The traffic is two-way, with moderate to
heavy travel. The site is within a downtown area and 60 to 70 feet away from a stop light at
which there is substantial queuing during commuter hours.
MANCHESTER, NEW HAMPSHIRE. The probe is located on the roof of a fire station,
(three stories high), which is one block from the central business district. The nearest road is
80 to 90 feet away and is a two way street with moderate traffic.
NASHUA, NEW HAMPSHIRE. The probe is located on the roof of a four-story building
that is located one and a half blocks away from the central business district. The surrounding
area is mostly residential, with some commerical buildings and only light traffic.
PORTLAND, MAINE. The probe is located on a 4-story building 10 feet from the main
street of the central business district (Congress Street). This is very much of a "downtown site"
with a heavily travelled roadway network. Furthermore, Portland is a coastal city located on a
hill and this site is located on the top of the hill.
BERLIN, N.H. The monitoring is done from a trailer on the bank of the river at the bot-
tom of a narrow valley, 250 to 500 m deep. The largest pulp mill in New England is located
about 800m up the valley. A paper mill is down the valley from this site.
PORTSMOUTH, N.H. The site is being relocated. It has been in a trailer in the center
of town, about 4 to 5 m high and about 15m from the road.
DEERFIELD, N.H. This site was used for special studies in 1976. The site is rural,
located behind a school building with a parking lot.
KEENE, N.H. This is a downtown site about a block from the center of town. It is on
the roof of a two-story building. It is about 25m back from a road with moderate traffic.
FRACONIA NOTCH (CANNON MOUNTAIN ~ GRAFTON COUNTY), VT. This site
is about 1200 m above sea level in the summit house at the top of the Gondola lift. It is a
rural mountain site. It has suffered moisture problems.
WHITE RIVER JUNCTION, VT. this is an acceptable site whose center-city location
may cause a reduction in measured ozone concentrations.
90
-------
APPENDIX B
SUMMARY OF HOURS WHEN NATIONAL
AMBIENT AIR QUALITY STANDARD FOR OZONE WAS VIOLATED
91
-------
APPENDIX B
SUMMARY OF HOURS WHEN NATIONAL
AMBIENT AIR QUALITY STANDARD FOR OZONE WAS VIOLATED
This appendix lists all the hours when ozone concentrations above 80 ppb were observed
at one of the Northern New England stations, the observed concentrations are given in ppb.
The data were obtained from the SAROAD tapes provided by EPA Region 1. No data were
available for dates subsquent to 30 June 1977 or before 1 January 1976. As noted in the text
of the report, there were substantial gaps in the data sets from several of the stations. It should
be noted that 1977 New Hampshire data were reported using Eastern Standard Time. Other
states used Eastern Daylight Time (EOT), as did New Hampshire in 1976.
93
-------
Time (EOT)
Site
Foreland, Maine
Date
11 Hay 76
14 Hay 76
29 May 76
10 June 76
16 June 76
29 June 76
6 July 76
7 July 76
11 July 76
16 July 76
4 August 76
5 August 76
6 August 76
12 August 76
26 August 76
27 August 76
1 September 76
19 September 76
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
82
82
88 86
92 100 112 118 170 146 114 92
'104 122 98
82 102 94 94
100 108 124 102 84 84 84 84 84
96 108 136 120 84
106 120 122 108 108 108 108 102 86
92 84
88 90
94 120 140 94 88 82
82
66 82 161 120 110 115 103 98 85
104 108 115 114 100 100 92 98 84
94 86 82
92 92 100 90 90 92
94 92
VO
-------
Time"
VO
Ul
Site
Manchester, N.H.
'
Date
15 June 76
16 June 76
20 June 76
28 June 76
30 June 76
7 July 76
8 July 76
11 July 76
23 July 76
5 August 76
12 August 76
22 August 76
26 August 76
11 March 77
12 March 77
30 March 77
12 April 77
13 April 77
21 April 77
22 April 77
1 May 77
2 May 77
17 May 77
18 May 77
21 Hay 77
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
88 90 87 85 104 110 108 101
92 81 82 92 88 84 111 89
126 88
82
140 125 150 112 100
92 95 95
88
91 95 100 118 112 107 102 90
88 86
84
88
82 95
101
90 98 99 89
89 107 116 109
83 88
82 81
H
86 85
81 83 81
81 81
83 82 88 90 81 100 92 95 96 91 89
84
88 88 87 100 100 102 101 90 102 101 92 93 108
92 100 94 100
85 110 130 127 140 140 105
1976 datea uie Eastern Daylight Time (EDT)j 1977 dates use Eaetern Standard Time (EST).
-------
Time (EST)
VO
o\
Site
Manchester,
N.H.
(cont.)
Date
22 May 1977
24 May 1977
28 May 1977
14 June 1977
17 June 1977
24 June 1977
25 June 1977
28 June 1977
01 02 03 04 OS 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
86 95 90 90 93 108 96 82 82 95 90
92 91 86
86
98 112 94
98 120 120 96 82
88 87 91 112 121 101 85
81
81 85 88 95 140 116 119 103 89 90 97 88
-------
Time (EDT)
VO
•xl
Site
Nashua, N.H.
Date
19 Apr 76
14 Hay 76
10 June 76
11 June 76
IS June 76
16 June 76
18 June 76
28 June 76
29 June 76
6 July 76
7 July 76
16 July 76
20 July 76
23 July 76
27 July 76
S August 76
12 August 76
13 August 76
22 August 76
26 August 76
1 September 76
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
85 120 85 100 100 90
81 85 95 100 100 110 110 100 90 90
85
85 85 85 85 85
85 95 100 110 125 120 120 120 110 120 125 125 130 120
100 100 100 100 105 100 95 90 100
95 105 110 125 110 90
90 105 105 105 110 125 125 120 90
95 95 90 90
85 85
90 85 90 125 105 85
90 95 95 95 90
85 85 85 100 110 135 135 120 90
85
85 90 90 105 105 105 110 105 100 95
85 90 95 85 100 90 90 85 85
115 95 125 145 125 105
85 90 90
95
85 110 115
95 100 95 90 90
-------
Time (ES?)
Site
Date
01 02 03 04 05 06 07 08 09 10 11 12 13 1ft IS 16 17 18 19 20 21 22 23 24
Nashua, H.H.
VO
00
12 March 77
12 April 77
13 April 77
25 April 77
1 Hay 77
2 Hay 77
4 Hay 77
5 May 77
6 May 77
16 Hay 77
17 Hay 77
18 Hay 77
20 Hay 77
24 Hay 77
25 Hay 77
26 May 77
27 May 77
28 May 77
30 May 77
31 May 77
17 June 77
24 June 77
28 June 77
85
100 105 100 90
95 100 100 100 100 95 90 85
85 95 90 85
85
90
100 95 100 100 100 95 90
85 95 100 105 100 100 100 90
85 115 115 110 105 95 90 100 125 115 90
85
85 95 105 105 110 110 110 120 130 125 120 120 115 115 110 110 115
125 120 120 90 90 95 95
85
105 110 105 100 100 105 95 85
85 85
85 90 90 90 90 90 95 85 85 90 110
100 105 100 95 95 100 120 130 135 140 140 150 140 140 135 135 130 130 145 120 105 90
90 130 170 170 170 185 160 150 145 100
85 85 85
85
90 95
85 90 100 90
85 85 95 100 95 85 85
-------
Time"
VO
vo
Sice
Portsmouth, N.H.
Berlin, N.H,
Deerfield, N.H.
Date
29 March 77
30 March 77
12 April 77
13 April 77
21 April 77
22 April 77
1 May 77
17 May 77
18 May 77
20 May 77
22 May 77
23 May 77
24 May 77
25 May 77
31 May 77
10 June 77
14 June 77
IB June 77
24 June 77
30 June 77
24 May 77
il July 76
20 July 76
22 August 76
01 02 03 04 OS 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
85
85 90 90 85 85
85 85
85 90
85 95 100 100 95
100 100 100 100 95 85
85 90 95 100 95 85 90 85
95 100 105 110 115 120 115 105 85 95 90 90
90 85
90 110 110 100 85
95 110 120 125 125 115 120 140 120 115 100 85 85 85
100 110 105 105 105 100 105 110 110
105 120 135 135 150 200 230 230 180 165 110 85
85 .105 110 110 110 105 100 95 90
105 110 105
90 90 85 35
90 95 95
85 85 85 85 1R 140 110
85 85 150 130
85 90 98 95 95 100 100 95 95 90
82
90 82
91 81
90
*
1976 dates use EOT; 1977 dates use EST
-------
Time
Site
Crafton County
(Fraconia Notch)
N.H.
Keene, N.H.
White River Jet. ,
Vt.
" x-'
--
ate
9 May 76
June 76
June 76
0 June 76
5 June 76
22 August 76
22 May 77
17 June 77
25 June 77
28 June 77
12 April 77
13 April 77
20 April 77
21 April 77
22 April 77
5 May 77
6 May 77
17 May 77
18 May 77
21 May 77
22 May 77
23 May 77
24 May 77
17 June 77
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
82 94 96 90 84 90 95 96 92 91 86 87 85 81 81
85 85
82
83 81 85 85 85
85 85 85
83
82 83 81
107 102 101 98 84
81 85 94 102 95 84 84 82 85 102 121 109 93
89 87 90 87 84 92 103 127 133 135 154 147 118 105 120 110 93
90 94 94 88 86 84
87 100 102 102 92 83 81
85 87 89 89
95 105 95 94 92 88
91
85 81
81 87 89 82 84 83 '
102 119 129 128 125 123 99 87
88 87 88 89 89
93 97 98 101 104 102 96 108 128 118 108 94
82 84 94 96 83 81 87 86 83
83
82 93 99 98 96
91 100 108 108 87 81
Keene, N.H. data are EST; others use EOT.
-------
Tine (EOT)
Site
Burlington, Vt.
Date
29 May 76
10 June 76
11 June 76
13 June 76
14 June 76
15 June 76
16 June 76
19 June 76
7 July 76
11 July 76
14 September 76
15 September 76
21 April 77
22 April 77
1 May 77
2 May 77
17 May 77
18 May 77
21 May 77
22 May 77
23 May 77
24 May 77
17 June 77
26 June 77
28 June 77
29 June 77
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
86 99 97 89 85 90
82 88 92
90 87 86 86 84
83 82 81
82 83 88 87 85
82 89 95 98 100 98 87 83 83
85 85 86 87 87 82 83 89 90 87
84 100 82
95 98 89
83 86 86 85
84 87 90 83 88 88 94
«9
100 102 104 102 96 98 102 95 94 88 94
85 81
84 86 88 83 88 98
96 91 83
85 92 94 82 88 96 88 93 102 104
98 102 95 85
82 84 82 90 90
110 113 104 91 84 85 91 100 103 104 104 101 104 108 94 94 94
84 84 82
84
85 90 92 88 87 94 102 83
82
85 92 97 95 98 100 105 112 110 107 112 119 108 89 105
101 92 91 83
-------
REFERENCES
Chameides, W.L., and D.H. Stedman, 1976: Ozone Formation from NO in "Clean Air" Env.
Sci. and Tech., 10, 150-154. x
Cleveland, W. S., et al., 1974: Sunday and Workday Variations in Photochemical Air Pollutants
in New Jersey and New York. Science, 186, 1037-1038.
Cleveland, W. S., B. Kleiner, J. E. McRae and J. L. Warner, 1976: Photochemical Air Pollu-
tion: Transport from the New York City Area into Connecticut and Massachusetts.
Science, 191, 179-181.
Crutzen, P. J., 1971: Ozone Production Rates in an Oxygen-Hydrogen-Nitrogen Oxide Atmo-
sphere. J. Geophys. Res., 76, 7311-7327.
Danielsen, E. F. and 1964: Report on Project Springfield. Report DASA-1517, Defense
Atomic Support Agency, Washington, D.C.
Danielsen, E. F. and V. A. Mohnen, 1976: Project Dust Storm Report: Ozone Measurements
and Meteorological Analyses of Tropopause Folding. Proc. the Non-urban Tropos-
pheric Composition Symposium, Hollywood, Florida, November 10-12, pp 17.1-
17.13.
Federal Register, 1971: Volume 36, Part 50-National Primary and Secondary Ambient Air
Quality Standards, pp. 22384-22397.
Fishman, J. and D. J. Crutzen, 1976: A Numerical Investigation of Tropospheric Photochemis-
try Using a One-Dimensional Model. Proc. The Non-Urban Tropospheric Composi-
tion Symposium, Hollywood, Florida, November 10-12, pp. 23.1-23.15.
Heffter, J. L. and A. D. Taylor, 1975: A Regional-Continental Scale Transport, Diffusion and
Deposition Model, Part I: Trajectory Model. NOAA Tech. Memo ERL ARL-50, pp.
1-16.
Holzworth, G. C., 1972: Mixing Heights, Wind Speeds, and Potential for Urban Air Pollution
through the Contiguous United States, Pub No. AP-101, EPA Office of Air Pro-
grams.
Lamb, R., 1976: A Case of Study of Stratospheric Ozone Affecting Ground-Level Oxidant
Concentrations (unpublished manuscript; SAI, San Rafael, California)
Londergan, R. J. and L. G. Polgar, 1975: Measurement Program for Ambient 03, NO and
NMHC at Portland, Maine-Summer, 1974. Report, EPA Contract 68-02-1382,
Task Order 2, The Research Corp. of New England, Wethersfield, Connecticut.
Ludwig, F. L., E. Reiter, E. Shelar, and W. B. Johnson, 1977a: The Relation of Oxidant Levels
to Precursor Emissions and Meteorological Features, Part 1: Analysis of Findings.
EPA Report No. 450/3-77-022a, Final Report, EPA Contract 68-02-2089, SRI Inter-
national, Menlo Park, CA.
Ludwig, F. L., P. B. Simmon, R. L. Mancuso, and J. H. S. Kealoha, 1977b: The Relation of
Oxidant Levels to Precursor Emissions and Meteorological Features, Part 3:
103
-------
Appendices. EPA Report No. 450/3-77-022, Final Report, Contract 68-02-2084.
SRI International, Menlo Park, CA.
Ludwig, F. L., and E. Shelar, 1977: Ozone in the Northeastern United States. EPA Report No.
901/9-76-007, Final Report, EPA Contract 68-02-2352. Stanford Research Institute,
Menlo Park, CA.
Ludwig, F. L. and E. Shelar, 1978a: Site Selection for the Monitoring of Photochemical Air Pol-
lutants, EPA Report No. 450/3-78-013, Final Report, EPA Contract 68-02-2028,
SRI International, Menlo Park, CA.
Ludwig, F. L. and E. Shelar, 1978b: Effects of Weather Fronts on Ozone Transport, in "Air
Quality and Atmospheric Ozone", edited by A. L. Morris and R.C. Barras, American
Soc. Test, and Materials Doc. No. STP 653, 389-406.
Lyons, W. A. and H. S. Cole, 1976: Photochemical Oxidant Transport: Mesoscale Lake Breeze
and Synoptic-Scale Aspects, J. Appl. Meteorol., 15, 733-743.
Meyer, E. L., W. P. Freas, J. E. Summerhays, and P. L. Youngblood, 1976: The Use of Trajec-
tory Analysis for Determining Empirical Relationships Among Ambient Ozone Lev-
els and Meteorological and Emissions Variables. Proc. of Int. Conf. on Photochem.
Oxidant Pollution and its Control, Raleigh, NC, September 12-17, 1976.
Mohnen, V. A., 1977: International Conference on Oxidants 1976-Analysis of Evidence and
Viewpoints. Part III. The Issue of Stratospheric Ozone Intrusion. EPA Report No.
600/3-77-115, pp. 7-66.
National Oceanic and Atmospheric Administration, 1976, 1977: Daily Weather Map Series.
OAQPS, 1977a: Effectiveness of Hydrocarbon Emission Control Programs as a Function of
Geographic Location, Unpublished Report. U.S. EPA Office of Air Quality Planning
and Standards.
OAQPS, 1977. Uses, Limitations and Technical Basis of Procedures for Quantifying Relation-
ships Between Photochemical Oxidants and Precursors. EPA Report No. 450/2-77-
021a. U.S. EPA Office of Air Quality Planning and Standards.
OAQPS, 1978: Procedures for Quantifying Relationships Between Photochemical Oxidants and
Precursors. Supporting Documentation, EPA Report No. 450/2-77-021 b, U.S. EPA
Office of Air Qualty Planning and Standards.
Reiter, E. R., 1975: Stratospheric-Tropospheric Exchange Processes. Rev of Geophys. and
Space Phys., 13, 459-474.
Reiter, E., 1977: International Conference on Oxidants, 1976-Analysis of Evidence and
Viewpoints, Part III, The Issue of Stratospheric Ozone Intrusion. EPA Report No.
600/3-77-115.
Ripperton, L. A., H. Jeffries, and J. B. Worth, 1971: Natural Synthesis of Ozone in the Tropo-
sphere. Env. Sci. and Tech. 5, 246-248.
104
-------
Singh, H. B., F. L. Ludwig and W. B. Johnson, 1977: Ozone in Clean Remote Atmospheres,
Final Report for Coordinating Research Council, Stanford Research Institute, Menlo
Park, CA.
Singh, H. B., F. L. Ludwig and W. B. Johnson, 1978: Tropospheric Ozone: Concentrations and
Variations in Clean Remote Atmospheres. Atmos. Environ., in press.
Spicer, C. W., J. L. Gemma and P. R. Sticksel, 1977: Ozone: Transport Considerations and
Predictive Models for Southern New England. Draft Report, EPA Contract 68-02-
2241, Battelle, Columbus, Ohio,
Tudor, D., 1978: Status Report on Maine's Air Quality. Presented at Workshop on Air Pollu-
tion in Maine; Augusta, Maine June 1, 1978. Sponsored by Maine League of
Women Voters.
i
Weinstock, B., and T. Y. Chang, 1976: Methane and Nonurban Ozone. Presented at the
APCA meeting in Portland, Oregon, June 29.
Went, F. W., 1960: Organic Matter in the Atmosphere and its Possible Relation to Petroleum
Formations. Proc. National Academy of Science, Vol. 46, 211.
Wishinski, P., 1977: Report Establishing a Data Base for Photochemical Oxidants (0^) and
Preliminary Analysis of Data. State of Vermont Air Program.
Zeller, K., R. B. Evans, C. K. Fitzsimmons, and G. W. Siple, 1977: Mesoscale Analysis of
Ozone Measurements in the Boston Environs. J. Geophys. Res., 82, 5879-5888.
105
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO. 2.
EPA 901/9-78-001
4. TITLE AND SUBTITLE
Atmospheric Processes Affecting Ozone
Concentrations in Northern New England
7. AUTHOR(S)
F. L. Ludwig and Rosemary Maughan
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SRI International
333 Ravenswood Avenue
Menlo Park, California 94025
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Region 1, Air Branch
Room 2113, J. F. Kennedy Federal Building
Boston, Massachusetts 02203
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
September 1978
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2548
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Readily available meteorological and air quality data were analyzed to determine the extent to which
ozone concentrations in the Northern New England states of Maine, New Hampshire, and Vermont are
influenced by causes external to those states. It is concluded on the basis of air trajectory and wind
analysis that ozone generated from precursor emissions to the southwest or west is transported into the
southern parts of Vermont, New Hampshire, and Maine. In the northern parts of New Hampshire and
Vermont, violations of the ozone standard are more frequently associated with air that has come from the
areas around Lakes Erie and Ontario. Although the Northern New England states are influenced by
ozone transported from elsewhere, some control measures might still be required within the area even if
the external sources were controlled and concentrations entering the region were reduced to levels near
the tropospheric background.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Ozone
Atmospheric Transport
Photochemistry
Air Quality
Maine, New Hampshire, Vermont
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport}
Unclassified
21. NO. OF PAGES
118
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 ,(Rev. 4-77) PREVIOUS EDITION is OBSOLETE
106
------- |