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

  
-------
 (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

-------