EPA-600/3-77-117
November 1977
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part V. The Issue of Oxidant Transport
D.H. Pack
Consulting Meteorologist
McLean, Virginia
Contract No. DA-7-1935A
E. Robinson
Washington State University
Pullman, Washington
Contract No. DA-7-2085A
F. Vukovich
Research Triangle Institute
Research Triangle Park, North Carolina
Contract No. DA-7-2170H
Project Officer
Basil Dimitriades
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
In general, the texts of papers included in this report have been repro-
duced in the form submitted by the authors.
-------
ABSTRACT
In recognition of the important and somewhat controversial nature of the
oxidant control problem, the U.S. Environmental Protection Agency (EPA)
organized and conducted a 5-day International Conference in September 1976.
The more than one hundred presentations and discussions at the Conference
revealed the existence of several issues and prompted the EPA to sponsor a
followup review/analysis effort. The followup effort was designed to review
carefully and impartially, to analyze relevant evidence and viewpoints re-
ported at the International Conference (and elsewhere), and to attempt to
resolve some of the oxidant-related scientific issues. The review/analysis
was conducted by experts (who did not work for the EPA or for industry) of
widely recognized competence and experience in the area of photochemical
pollution occurrence and control.
Part V includes discussions on the issue of oxidant transport written by
Donald H. Pack, Consulting Meteorologist, McLean, Virginia; Elmer Robinson,
Washington State University, Pullman, Washington; and Fred M. Vukovich,
Research Triangle Institute, Research Triangle Park, North Carolina. The
authors deal with the phenomena of urban plume formation and transport,
measurement and tracking, and oxidants and precursor ranges, and recommend
future studies.
iii
-------
-------
CONTENTS
ABSTRACT iii
FIGURES vi
TABLES vi
INTRODUCTION 1
B. Dimitriades and A.P. Altshuller
THE ISSUE OF OXIDANT TRANSPORT 3
B. Dimitriades and A.P. Altshuller
REVIEW AND ANALYSIS 9
D.H. Pack
Summary 9
Introductory Comments 10
Information Assessment 11
REVIEW AND ANALYSIS 21
E. Robinson
Introduction , . 21
Natural Versus Anthropogenic Ozone ... 22
Transport from Upwind Sources and Local Problems 35
Ozone and Precursor Transport Distance 36
Control Strategies Recognizing Oxidant Transport 37
REVIEW AND ANALYSIS 39
F.M. Vukovich
Introduction 39
Chemistry of Ozone in Remote Regions 39
Mesoscale Transport 43
Synoptic-Scale Transport of Ozone 48
Summary and Conclusion 54
Recommendations for Further Research 55
REFERENCES 57
BIBLIOGRAPHY 67
-------
REVIEW ^UV- *-l!\i?iLYSIb
Number Page
1 Diurnal variation of rn.:-, vertical distribution
of ozone at Wilmir>cton. Ohio on 1 Aucmst 1974 ..... 45
2 The "'critical Iistributi-?'~ ,f ozone over Indianapolis . . . SO
TABLES
REVIEW AND ANALYSIS - D.H. Pack
Number Page
1 Ozone Concentrate -. ... . . . . , , . . . . i':
Average Ozr - Jun<-'ent:.:cj''-io''- r-jm the
; tc : 000 f- /^csus '! -.'= of PH- Us:
f' r "";: I
VI
-------
ACKNOWLEDGMENTS
These contracts were jointly funded by the Office of Research and Devel-
opment (Environmental Sciences Research Laboratory) and the Office of Air
Quality Planning and Standards.
The assistance of the technical editorial staff of Northrop Services,
Inc. (under contract 68-02-2566) in preparing these reports is gratefully
acknowledged.
VI1
-------
INTRODUCTION
Basil Dimitriades and A. Paul Altshuller
In recognition of the important and somewhat controversial nature of the
oxidant control problem, the U.S. Environmental Protection Agency (EPA)
organized and conducted a 5-day International Conference in September 1976.
The one hundred or so presentations and discussions at the Conference revealed
the existence of several issues and prompted EPA to sponsor a followup review/
analysis effort. Specifically, this followup effort is to review carefully
and impartially and analyze relevant evidence and viewpoints reported at the
International Conference (and elsewhere) and to attempt to resolve some of the
oxidant-related scientific issues. This review/analysis effort has been
contracted out by EPA to scientists (who do not work for EPA or industry) with
extensive experience and expertise in the area of photochemical pollution
occurrence and control. The first part of the overall effort, performed by
the EPA Project Officer and reported in a scientific journal (1), was an
explanatory analysis of the problem and definition of key issues, as viewed
within the research component of EPA. The reports of the contractor expert/
reviewer groups offering either resolutions of those issues or recommendations
for additional research needed to achieve such resolutions are presented in
the volumes composing this series.
This report presents the reviews/analyses prepared by the contractor
experts on the issue of oxidant transport. In the interest of completeness
the report will include also an introductory discussion of the issue, taken
from Part I. The reviews/analyses prepared by the contractor experts follow.
-------
THE ISSUE OF OXIDANT TRANSPORT
Basil Dimitriades and A. Paul Altshuller
It is to be remembered that this review-analysis effort is concerned only
with existing issues, that is with questions that have been answered but
conflictingly. If such a definition of issues is to be observed strictly,
then while there are numerous unanswered questions there may be no issue
related to oxidant transport, at least, none other than those already defined
and discussed in connection with the stratospheric ozone and the natural
emissions. To explain, one major question relevant to the oxidant problem is
on the relative strengths of the natural and the anthropogenic sources in a
given region or area. This question has not been answered unequivocally and
quantitatively because anthropogenic pollutant transport makes it difficult to
assess the strength of the natural sources. The question has been answered
qualitatively, a consensus being that pollutant transport does occur and
contributes to oxidant buildup in areas far from the sources (2-4). Quanti-
tative answers, however, have not been agreed upon, and this disagreement
constitutes the issues already presented in the preceding two sections of this
analysis.
Aside from its connection to the natural vs anthropogenic sources ques-
tion, the phenomenon of oxidant transport is of interest for yet another
extremely important reason. This reason is the strong possibility a fact,
to some investigators that oxidant and/or oxidant precursors transported
from upwind sources obscure the role and impact of local emissions to a degree
that local control requirements cannot be estimated with confidence. In fact,
this obscuring effect is a problem of much broader nature, affecting both main
components of the oxidant control strategy, namely:
-------
(a) the source-receptor relationship, that is, the geographical defini-
tion of the area within which emission control must be applied in
order to reduce the oxidant levels observed in a (given) locality,
and
(b) the quantitative relationship between oxidant-related air quality and
precursor emission rates.
This connection between oxidant transport and oxidant control strategy is
a conclusion arrived at as a result of numerous recent field studies (2-4).
Specifically, these studies established that the emission-dispersion and
photochemical reaction processes do not have a simple and "local" nature as
was assumed in the designing of the first and current oxidant control
strategy. The phenomena of urban oxidant plume formation and movement, rural
oxidant occurrence (at problem levels), "Sunday-weekday effect," and nighttime
oxidant occurrence, previously either unnoticed or thought to be "odd," are
now believed to be manifestations of an extremely complex emission/pollutant
dispersion process. Such complexitd.es, for example, are introduced by hori-
zontal and/or vertical transport of oxidants and/or of precursor mixtures to
long distances without excessive dilution.
It is this connection between oxidant control strategy and oxidant trans-
port, and, within this context, the specific areas of nature, extent of, and
consequences from pollutant transport, in which several questions exist and
need to be answered. Some of these questions have been given conflicting
answers, but the supporting evidence was in almost every case either scant or
none. For this latter reason, these questions should perhaps be considered as
"unanswered" questions rather than as questions at issue. Nevertheless, for
important and urgent reasons these questions will be included in this review/
analysis with the understanding that the need here is either for answers or
for specific recommendations for research that would provide answers. These
questions and related explanations/discussion are as follows:
1. What is the maximum range of ozone transport?
-------
The question pertains specifically to ozone, and is concerned with the maximum
distance downwind that ozone can travel without excessive destruction or
dilution (e.g., no more destruction or dilution than 80 percent). Answers have
been offered, but are nonspecific and vary by several tens of kilometers.
Thus, from direct observations upon an urban oxidant plume it cannot always be
ascertained whether the transported oxidant constitutes a fraction of the
concentration at the point of origin or is fresh oxidant formed during trans-
port. Most likely, both types of oxidant exist but in unknown, and not easy
to determine, proportions. The question involves considerations of chemistry
and meteorology (dispersion) and could perhaps be answered in parts. For
example, it would be relevant and useful to answer the following questions:
(a) What is the photochemistry-related lifetime of ozone?
Answers have been calculated for an ideal atmospheric system containing
no HC and NO pollutants, except for methane (and CO) at their global back-
X
ground levels (5). It is conceivable, however, that in the presence of
trace-levels of HC and NO levels such that their potential for 0 formation
X -J
is either negligible or predictable that the lifetime of O may be quite
different.
(b) What is the range of ozone lifetimes related to atmospheric (at
ground level) on surface destruction processes?
(c) For an inert pollutant, what is the longest lifetime related to
the atmospheric dilution process?
2. What is the maximum range of oxidant-precursor transport?
The question pertains to HC and NO pollutants as a mixture not to the
X
individual precursors and is concerned with the maximum distance downwind
that a HC-NO mixture can travel without excessive loss of its potential for
oxidant formation, (e.g., no more loss of oxidant potential than 80 percent).
The question is far more complex than the preceding one on ozone; neverthe-
less, answers have been offered, although again nonspecific and diverse. For
-------
example, analysis of aerometric data provides evidence suggestive of "long-
range" transport but does not identify corresponding source and receptor
areas. Also, based on meteorological modeling techniques, it has been calcu-
lated that the residence time of air parcels and their pollutants in high
pressure systems can be as long as several days (6); but this does not neces-
sarily mean that the resident pollutants preserved a significant potential for
oxidant formation that long. In general, conclusive evidence is insufficient
because the aerometric data available do not distinguish transported precur-
sors from fresh ones, and situations in which precursors travel above entirely
source-free areas (e.g., over sea or lakes) are not easy to study. Laboratory
and modeling evidence also exists but with the usual uncertainity problems
arising from the indirect nature of such evidence. It is the composite of all
this evidence that needs to be considered before an answer and/or recommenda-
tions can be arrived at.
3. Is it possible to measure reliably the fraction of oxidant caused by
local (urban) emissions, and, if so, how?
This question is an extremely important one although it is only a gross sim-
plification of the main question of relating air quality to emissions. To
explain, the role of local emissions is in many cases obscured by interfer-
ences which, in appearances as least, have extraneous origin. Thus, if the
no-interference case is the one in which the local oxidant originates entirely
from the day's local emissions, then interferences could be expected to be
caused by
(a) precursors originating from upwind sources,
(b) oxidant originating from upwind sources,
(c) precursors originating from local sources but from previous days, and
(d) oxidant originating from local sources but from previous days.
Of these, only cases (a) and (b) pertain to extraneous influences and, hence,
are of interest insofar as this question is concerned, and merit further
-------
examination. Such an examination is made and offered here in the form of a
proposed rather than established picture-, comments, therefore, are solicited.
Precursors from upwind sources may cause an oxidant problem in the source-
free rural areas but not in urban areas, since such precursors cannot be
expected to be important relative to those emitted within the urban areas; if
they are, then the urban area affected should perhaps be redefined to include
these upwind sources. For this reason, case (a) can be disregarded as being
irrelevant to the question asked here.
Oxidant from upwind sources can affect an urban atmosphere in two con-
ceivable ways: (a) by directly mixing into the urban atmosphere, and (b) by
interacting with the photochemistry of local emissions. Obviously, upwind
oxidant can be measured directly, and if found absent throughout the day, then
it can be concluded that the local problem is caused entirely by local emis-
sions. If upwind oxidant does exist at significant levels, then its main
effect is expected to be a direct contribution to oxidant buildup and/or the
conversion of an equivalent amount of NO into NO . This latter conversion
should have an enhancing effect upon oxidant accumulation. The magnitude of
these effects must be known if the role and impact of the local emissions is
to be correctly assessed.
In conclusion, the question presented here can be reduced or more narrow-
ly defined as follows:
Is it possible to measure reliably the effect of upwind oxidant upon the
locally observed peak oxidant, and, if so, how?"
-------
REVIEW AND ANALYSIS
Donald H. Pack
SUMMARY
1. Long distance transport of anthropogenically produced ozone has been
convincingly demonstrated to occur over distances up to about 1000 km.
Transport over the longest ranges requires a minimized near-surface destruc-
tion rate because of either the protective "barrier" of a surface inversion, a
slower surface contact destruction rate (e.g., sea surfaces), or the absence
of significant concentrations of ozone destroying compounds.
2. Ozone "lifetimes" may be as long as 2-3 days under the above circum-
stances; the data are insufficient to set a precise upper limit. The "life-
times" may also be as short as a few hours when the circumstances described
above are not evident.
3. There is indirect but rather ample evidence that the highest episodic
levels of ozone are related to anticycIonic weather systems moving over
significant emission areas. These weather systems maximize solar radiation,
and slow ventilation and nocturnal surface radiation inversions. All of these
combine to permit an accumulation of ozone and, apparently, ozone precursors.
The time scale of these events is on the order of 2-3 days at a fixed point
and is directly related to the speed of movement of the anticyclone system.
The total effect of the (moving) system may be two to three times as long as
it successively affects different areas.
4. There is insufficient data to separate the effects of "fossil" ozone from
fresh ozone formed from unreacted ozone precursors on the second and succeed-
-------
ing days of high ozone episodes. Indeed, such precursors are not yet identi-
fied, but the scanty evidence favors such a phenomenon.
5. An experimental configuration has been suggested to permit following and
measuring a "parcel" of air (the Lagrangian approach) as the most direct and
feasible way to determine atmospheric chemistry production and destruction
rate and to study the (probably variable) inventories of ozone-related com-
pounds. Such an approach would facilitate understanding the mechanisms of,
especially, ozone destruction and vertical mixing by the atmosphere.
INTRODUCTORY COMMENTS
The article by Dimitriades and Altshuller (1) seems to define the issues
as though there is only a single set of physical events: emission-oxidant
formation-oxidant destruction unique in space and time. From the data review-
ed it appears that while there is a dominant theme, there is enough variation
in the physical systems that connect the anthropogenic and natural emissions
to the observed oxidant (measured as ozone) levels to raise the possibility of
qualitatively different circumstances leading to the creation of high ozone
levels. Such items as the variation in the amount and ratios of NMHC and NO ,
X
atmospheric dilution, "early" vs "late" photochemistry, gas phase and surface
contact destruction rates may give rise to quite different end results.
For example, none of the field data were taken in areas where there is
the possibility of large natural emissions only and the meteorological condi-
tions were optimum for ozone formation. (The negative aspect of the preceding,
i.e., "I haven't measured it, but it could exist." is recognized.) However,
the experimental data have illuminated the most urgent problem, the urban-to-
rural transport/formation of high ozone concentrations. The role of natural
emissions was masked by the large anthropogenic emissions. The time seems
ripe to perform a similar evaluation over areas with significant natural
emissions but (almost) free of anthropogenic sources.
It is important to recognize the possibility that variations in the
combinations of the physical mechanisms might, and I emphasize the uncertainty,
10
-------
lead to different "average" cause-effect relationships geographically and over
time.
INFORMATION ASSESSMENT
Question 1: What is the maximum range of ozone transport (dilution
and/or destruction equal to or less than 80%)?
None of the data permit a precise evaluation of the question. Elevated
levels of ozone, greater than 80 ppb (160 g m ), created from anthropogenic
precursors have been observed after travel distances of at least 700 km (7)
and over travel times of 48 hours or more. The anthropogenic character of the
source is supported by the concurrent high levels of ozone and CC1 F, the
latter being strictly anthropogenic. This study postulated that the slower
contact destruction rate between ozone and the sea surface, together with the
minimum dilution characteristic of a large area source permitted long distance
transport at sustained high concentrations. It appears that relatively slow
vertical dilution and, more importantly, minimum destruction rates were the
controlling phenomena. This is supported by the concurrent and contrasting
behavior of ozone at land locations where the combination of contact and gas-
phase destruction, together with a surface-based inversion that inhibited
replenishment of ozone from sources aloft, reduced surface level concentra-
tions from more than 100 ppb to less than 50 ppb in a few hours.
A study of the dilution of CC1 F and CC1. moving from Europe to Ireland
(8) suggests that a travel time of about 120 hours without additional emissions
into the volume is sufficient to dilute inert pollutants from a large area
source to values near background levels.
None of the field data come close to the continuous surveillance of a
moving parcel of ozone and precursor-rich air over sufficiently large time-
space coordinates and free of additional emissions. Such data are required to
provide accurate answers to the question.
11
-------
References presenting field data are consistent in showing that area-
wide, nonurban, high ozone episodes require significant anthropogenic emissions,
some level of UV radiation flux (never specified), and not too vigorous
atmospheric dilution. References 6, 9-15 emphasize the large area episodic
situation characterized by anticyclones moving, not too rapidly, across multi-
ple emission sources and the cumulative buildup of high ozone concentrations
as new emissions add to "fossil" ozone (either ozone per se or from late
reacting precursors).
The ozone "lifetime" at a point then becomes much more the speed of
motion of the synoptic systems. Climatological studies indicate that these
situations tend to be more prevalent in the eastern U.S. and more likely in
the fall and early winter, not in summer at the time of the maximum solar
radiation. These studies also show that "episodes" of 2-days' duration at a
particular location should be fairly common, but that duration of an episode,
at a point, decreases markedly to about one per year north of 35° N Lat. and
near two per year in the southeastern states.
No data were available for this area of the country, which appears to be
particularly vulnerable to the episode phenomenon providing the emissions are
sufficiently large.
It seems clear from the data presented that: (a) large area buildups in
high ozone concentrations involve air parcel residence times of at least 2
days to accumulate sufficient source material (6); (b) "fossil" ozone carried
aloft by daytime convection (but not too far aloft) must be prevented from
mixing downwards to the surface at night, usually by a surface-based inver-
sion, to reduce the near-surface destruction if it is to survive in sufficient
quantity to influence the following day's ozone concentration. This is shown
most clearly by the "DaVinci" data (16).
In contrast several papers (16-27) demonstrate directly augmented ozone
levels resulting from transport over 6-8 hours and distances from 100 to 300
kilometers. In these instances ozone concentrations continue to increase with
12
-------
time and distance from the primary source area (e.g., New York to Boston). It
appears (19,22,28) that this condition is due to not only delayed ozone
formation downwind from the primary source but also to the fact that the pro-
duction rate exceeds the combined gas-phase/contact destruction rates even
though vertical mixing remains moderately active.
With this preamble one can estimate answers to:
What is the photochemistry-related lifetime of ozone?
Under circumstances of minimal surface contact destruction and a major
reduction in the concentration of compounds that cause gas-phase destruction
(e.g., NO), the demonstrated lifetime is at least 48 hours. This has been
shown to occur for transport over the sea and (for somewhat shorter times)
over land at altitudes above 100 to 300 meters in the presence of a surface
inversion.
Under circumstances permitting contact destruction and continued emission
of gas phase destruction compounds, the lifetime is less than 6 hours, prob-
ably nearer to 2 hours. For example there are a few data that can be used to
make crude estimates of this latter effect. The following ozone concentration
data were extracted from Figure 6 in Decker et al. (16).
TABLE 1. OZONE CONCENTRATION (ug m )
Maximum Minimum
Surface 285 (1800LST) 59 (2300LST)
Aloft(800m) 291 (1830LST) 236 (2315LST)
The reduction in ozone aloft can be postulated to be due to only O -NO
J -3
gas-phase destruction. The reduction in the ozone concentration, 55 pg m ,
occurs in 4.75 hours. If we assume that this decrease in ozone also repre-
sents the rate of reduction in NO, and if we further assume an exponential
form for the decrease, this results in an NO "half-life" of 0.82 hour. Assump-
tion of a linear decrease gives the obvious value of 12 yg m per hour.
13
-------
If these values are applied to the surface data that show a decrease of
ozone of 226 yg m in 5.0 hours, the exponential NO decay model would account
for almost 99% of the reduction, leaving very little to surface contact
removal. (The linear removal rate would account for only 58 ug m with the
remainder to be destroyed by surface contact. However, since this requires
the exhaustion of all the NO as soon as the 58 ug m of ozone is destroyed
with no replenishment of NO, this seems an illogical model.)
The actual reduction of surface ozone, 285 to 59 yg m , if the reduction
rate is of exponential form, suggests a half-life of somewhat more than 2
hours.
This arithmetical exercise suggests that the gas-phase, removal mechanism
reduces ozone by 50% each 1 to 2 hours with the surface contact removal rates
over land somewhat slower, on the order of 2-4 hours when couched as a "half-
life." However, the crude assumptions and lack of data suggest treating these
numbers with considerable caution!
What is the range of ozone lifetimes related to atmospheric (at ground
level) on-surface destruction processes?
One attempt to estimate such numbers has been given above. However, this
issue cannot yet be satisfactorily disentangled from the gas-phase destruction.
Flux measurements (e.g., 1 x 10 molecules cm s ) can be questioned unless
ancillary measurements of possible ozone-destroying compounds are also report-
ed. Estimates of global residence times refer to the total troposphere, not
to the contact layer. These estimates of about 4 months (29) are much too
long for this smaller volume.
Assumption of a deposition velocity approach using v values similar to
-1 ^ -1
other reactive gases (e.g., iodine-v = 2.5 cm s ; SO -v = 0.85 cm s )
would deplete the contact surface layer of ozone in a few minutes, unless
replaced by transport from above. This may be too rapid a removal rate for a
layer of finite depth of, say, 10 to 100 meters. Here the rate-limiting
14
-------
process is not the actual destruction but the downward turbulent flux of
ozone. At night this may be slower than the "deposition."
Thus with the data now at hand we must again resort to pure empiricism
and guess at values. Considering a surface layer of 10-100 meters thick,
values seem to range from near 2 hours overland to a few days for overwater
trajectories.
For an inert pollutant, what is the longest lifetime related to the
atmospheric dilution process?
The question as stated, even for "no more destruction or dilution than
80%," is not meaningful unless one specifies the source configuration, size,
location, and intensity of emissions. To take an extreme example, natural
particulates with their hemispheric-wide source distribution are estimated to
have a lifetime of about 30 days. Anthropogenic halocarbons may reach levels
of near 1000 ppt(v) on the outskirts of cities, but the global background is
still near 15% of these values even after circling the globe.
In contrast, a single point source, even in modest dilution conditions,
-4
is reduced by about 10 within a kilometer of travel.
On a more relevant air pollution scale, if the source region is of
sufficient lateral extent, say 2000 x 2000 kilometers, atmospheric dilution is
primarily confined to the vertical dimension, with this determined by the day-
to-day variation in the maximum mixing depth (MD). Day-to-day changes in the
MD are usually rather small, especially in the anticyclone conditions so
frequently responsible for area-wide high ozone episodes. Examination of the
combined MD/wind speed data of Holzworth (30) show day-to-day variations in
the afternoon MD of 10-20% together with light and chaotic wind conditions.
The day-to-night variation is much larger. Typical values for summer would be
about 1800 meters in the afternoon and near 400 meters at night, or rather the
following early morning. Wind speeds averaged over the afternoon depth are
about 5 meters per second and 3-4 meters per second over the morning MD.
15
-------
A crude model of dilution can be constructed from these data. Assume a
uniform distribution of an inert material throughout the 1800 MD. With the
onset of stability the upper 1400 meters (1800 - 400) are isolated from
surface sources and carried away at the difference in wind speeds between the
two layers, 2 meters per second. However, double this to 4 meters per second
to account for the higher winds aloft at night. In 12 hours this volume moves
about 170 kilometers. Then let the MD of the next day increase to 2700 meters
(a very large change). This reduces the concentration in this isolated volume
to about 50% of its original value. Wind direction shear with height would
result in some additional dilution but probably less than an additional factor
of two.
Under such conditions and for large area sources, the reduction in
concentration due to atmospheric dilution alone is unlikely to be more than
75% in a 24-hour period, probably less.
An alternate approach is suggested by Decker et al. (9) with the concept
of "residence time" within an anticyclone. These calculations suggest a
residence time of 1-3 days within the system. Again vertical mixing is the
primary atmospheric dilution mechanism. Since these systems do not show large
variations in vertical mixing, dilution on the order of 50% per day is reason-
able leading to periods of 2-3 days required to reduce concentrations to about
20% of the original value.
Note that both of these crude estimates are compatible with the interes-
ting trajectory/statistical analysis of Ludwig et al. (31). The space scales
for high ozone episodes were near 450 kilometers for 12 hours and 1150 kilo-
meters for 36 hours in this study.
Finally a few measurements of the near inert gases, acetylene and propane
^o> ind"*""*4"'^ nit*~l~ 0""" ^" cli"'u"*~ "* ^*i Across a d^ ^tci^cc c^ ^.crs "*~ha*^ 1600 ki^o
meters (Des Moines, Iowa Albany, New York). Since emissions are added from
below, these data cannot be used to directly calculate dilution, but the lack
of variation of more than a few tenths of a percent confirms the absence of
rapid dilution between significant source inputs.
16
-------
Question 2: What is the maximum range of oxidant-precursor transport?
I was unable to locate any quantitative information from which direct
answers to this question could be derived. The weight of suggestive evidence
is that the more highly reactive precursors are "used up" in the first day's
photochemistry and that "maximum" range must be defined in terms of "highly,"
"intermediately," and "slowly" reacting classes of materials. Ludwig et al.
show a correlation between high ozone levels and NO emissions 24-36 hours
* x
earlier (31). This in turn relates to distances of transport of 7QO and 1150
kilometers. The difficulty with this information is that it cannot separate
the contribution of "fossil" ozone from any fresTi ozone created from previous-
ly unused precursors. (See Ref. 18, for example.)
The mathematical models presented seem equivocal and too simplified to
presently contribute much to the quantitative answers.
However, the chamber experiments of RTI (32) indicate a capability for
ozone formation is retained for 2-3 days without additions of more source
compounds. If true for the outdoor atmosphere, this translates into distances
of 1000 kilometers or more.
Question 3 and the final, unnumbered question are also related to Ques-
tion 2.
Question 3: Is it possible to measure reliably the fraction of oxidant
caused by local (urban) emissions, and, if so, how?
Is it possible to measure reliably the effect of upwind oxidant upon the
locally observed peak oxidant, and, if so, how?
All of these questions involve solutions to a multidimensional problem
that involves at least the following:
17
-------
Emissions - O , NMHC, NO
x x
UV Activation energy levels
O Formation/destruction mechanisms and rates
X
NO Formation/destruction rates
NMHC Changes in composition by "classes"
Atmospheric transport
Atmospheric dilution in x, y, z space
Time
Since the data now at hand do not appear to be adequate to resolve these
issues, I believe that the most promising approach is to design field studies
to consider the preceding parameters in a Lagrangian framework with Eulerian
supporting data. The DaVinci experiment suggests that this would be a very
profitable approach.
Concurrently mathematical photochemistry models tailored to Lagrangian
type of data should be prepared. The models can also assist in the design of
the necessary measurements, both the compounds ("classes?") and the observa-
tional frequency requirements. I am not qualified to comment on model design.
Also, I cannot comment on the reliability aspects of the experimental
configuration to be suggested since this aspect seems, to me at least, to
depend mainly on air quality instrumentation its sensitivity, precision,
accuracy, and the capability of rapid repetition rates, at least for most of
the compounds.
Assuming, however, that the requisite instrumentation can be assembled,
the following experimental configuration is suggested.
Lagrangian Photochemistry Experiment
Platforms:
Free floating balloon(s): Equipped to measure Ov THC, NMHC (by
18
-------
"classes?"), NO, NO , CC1 F, nephelometer UV radiation intensity,
temperature, relative humidity, and height (pressure).
Surface platform(s), car or boat: Equipped as above.
Aircraft: Equipped to measure O , NO, NO2/ THC, CC13F, nephelometer,
temperature, relative humidity, and height.
Upper wind systems
Temperature profile systems
Small constant volume balloons (tetroons); radar and/or visual tracking.
The free balloon would measure in situ time changes. Comparison of its
data with the data from the surface platform and aircraft air quality profiles
would permit estimates of vertical fluxes and of destruction rates. Horizon-
tal traverses by the aircraft would provide data on x-y gradients (inhomoge-
neities). Vertical wind and temperature profile data are essential, espe-
cially the latter, to study the rarely measured time-change in the mixing depth
and the upward extent of the near-surface nighttime stable layer.
The crucial aspect of this experiment is the selection of the experi-
mental site(s). An essential experiment is the tracking of an urban emission
volume (e.g., the emissions-from 0800-0900) for 24 to 36 hours to study ozone
and precursor behavior and interaction, to measure quantitatively the "fossil"
vs "fresh" ozone. It is necessary to have complete assurance that there are
zero emissions along the track of the free balloon to eliminate the problems
posed by the present mixed data.
One possibility is to launch in the plume from a relatively isolated
coastal city (Charleston, S.C.?) with a trajectory moving offshore 10-20 miles
then, hopefully, paralleling the coast but entirely over water. The feasi-
bility of this from the standpoint of the safety of the balloon crew may be
19
-------
marginal at best. The possibility of automated instrumentation on a radio-
controlled, unmanned large balloon should be investigated-
A second possibility is a site near Lake Michigan or Lake Superior;
Chicago, Milwaukee, or Duluth might serve as the city source.
A third possibility is to stage the experiment in the land-sea breeze
regime of the Los Angeles Basin through the use of the offshore islands as
launch points. The stringent air traffic controls and crew safety may create
difficulties.
Lastly, and less desirable, would be the selection of an isolated smaller
city (SpokaneT Washington, Portland, Maine?) where the "no additional emission"
criterion might be approached.
(Note: Portland, Maine, was suggested, which raises the issue of the
role of natural emissions. While the data available do not seem to show any
large natural influences, one can also infer that in the areas where the data
were collected and at the times of the experiments the anthropogenic emissions
were too overwhelming to permit detection of the effect of natural emissions.
It seems desirable to collect some preliminary data to evaluate the need for
a "DaVinci" type of experiment over, for example, rural North and South Caro-
lina during slow dilution, high radiation intensity conditions.)
Use of the Lagrangian approach for data to answer the final (unnumbered)
question may not be practical for air traffic and safety reasons since it
would involve deliberate free ballooning over a city. However, it should be
possible to use this approach to measure the pollutant from upwind to the edge
of the city, then shift to a helicopter platform following a tetroon over the
city. This technique was used very successfully in the LARRP. Support of a
ground air quality network would be essential. St. Louis, Missouri, seems a
logical site choice.
20
-------
REVIEW AND ANALYSIS
Elmer Robinson
INTRODUCTION
One of the very urgent problems in the Federal Government's drive to
achieve cleaner air is to obtain some degree of improvement in the area of
photochemical air pollution. This initially meant bringing oxidant and ozone
concentrations observed in urban areas down so that the prescribed ambient air
quality standards were met in the nation's urban areas. This ozone/oxidant
program has been proceeding slowly through an urban source control program
directed at the control of the oxidant precursors of nitrogen oxides and
reactive hydrocarbons. Transportation sources were major contributors in
these areas and were the target for a variety of control plans and actions.
The control program for obtaining control over photochemical oxidant* was
directed at sources in the general vicinity of the observed excessive concen-
trations or over surrounding regions having an obvious strong impact on the
area where the excessive O levels were measured. Thus this was a "local
source-local control" program, and the various EPA regulations were estab-
lished on the premise that excessive O would be countered by a program of
source control in the general area of the excessive O observations.
This program of local source-local control began to come under attack
when excessive ozone concentrations were measured in rural areas where there
were few obvious local sources. One of the first of these ozone-affected
rural areas was in the mountains of western Maryland. Even if very restric-
tive local soxirce control options were followed in these local areas, it was
obvious that the 0, levels would persist unaffected. This would occur because
*Ozone or O will be used in the rest of this text to indicate photochemical
oxidant.
21
-------
it was recognized that large-scale atmospheric transport of O. was capable, in
certain situations, of advecting air with high concentrations of urban pollut-
ants, o , and ozone-precursors into the remote areas. If ozone advection was
a general rather than just a peculiar condition, then some revision in the
source control strategies directed toward photochemical oxidant would be
called for. At the time this report is being prepared (early 1977), there are
still many confusing aspects present in the urban-rural photochemical oxidant
situation. This report is directed toward an examination of various meteoro-
logical transport features and problems that are associated with photochemical
oxidant in rural areas.
Some of these O transport problem areas have been outlined by Dimitriades
and Altshuller (1) to include the following areas:
The importance of natural versus anthropogenic 0_ in rural areas;
The possibility that O transported from upwind sources could obscure
local problems, e.g., the areas where emission controls should be
established or perhaps the local relationship between O_ and precursors;
What is the maximum range of significant O transport?
What is the maximum range of significant O precursor transport?
Can sources for both advected and local O be quantified when O is
advected into an area?
What control strategies can be proposed that will account for some
expectation of success in lowering observed O levels?
The following discussion will consider these various problems from the
basis of a critical examination of the available current literature.
NATURAL VERSUS ANTHROPOGENIC OZONE
Introduction
Ozone is a natural constituent of the atmosphere and has been reported by
various authors as having a ground surface concentration up to about 50
22
-------
ppb (33). The source of this natural ozone has typically been attributed to
the injection of ozone into the troposphere from source regions in the strato-
sphere. There is no doubt that ozone of stratospheric origin exists at the
surface and that this is a major source for the ozone found in many remote
areas.*
The current literature on rural O and O transport shows both similar-
ities and differences in various regions of the country. Thus there may not
be a single model or explanation that can be fitted to all regions of the
country or, indeed, to a single region for all types of seasonal episodes or
circumstances. The discussion below will consider three major areas of the
U.S. where the rural 0 problem seems to pose regulatory problems at the
present time. These are the gulf coast, the upper Midwest, and the Northeast.
California problems will not be considered in detail because no nonanthro-
pogenic explanation of the O patterns in this area makes any sense. The
California situations can, however, provide data on and some support to models
explaining O conditions in other regions.
It should be recognized that what were probably the first situations
where natural ozone was suspected as being a significant air pollution factor
occurred in southern California. The first was the initial characterization
of the oxidant in the Los Angeles photochemical smog as being caused by local,
urban precursor sources. A second event that is not recalled often is the
explanation of the high oxidant concentrations, over 300 ppb oxidant in some
cases, accompanying the "clean" onshore sea breeze in rural northern San Diego
County in the later 1950"s. The cause of these very high, by Midwest and east
coast standards even now, oxidant levels was described by Bell who showed that
overwater trajectories of smog from the Los Angeles basin could be identified
with the high oxidant incidents in San Diego County (34). In these meteoro-
*This author would like to define the two terms "remote" and "rural" to have
the following specific meaning: "remote" is defined as being removed from
any possible identifiable anthropogenic source, e.g., an Arctic tundra station;
"rural" is a nonurban location that from time to time may be influenced by
man's activities at an identifiable source area, although the mechanism by
which the pollutants affect the given location may not be readily recognized,
e.g., an upstate New York station.
23
-------
logical situations, characterized by offshore flow in the Los Angeles area and
onshore winds in the San Diego area, the ozone and/or its precursors persisted
overnight and into the next day. In addition the pollutants traveled about
100 miles until the smog parcels were observed as high oxidant incidents in an
area of minor emission rates. Before Bell made his careful study, it might
have been concluded that this was natural ozone because it was carried by a
sea breeze coming off a source-free area.
Most investigators in this area of natural versus anthropogenic ozone in
rural (not remote) areas believe that their results indicate anthropogenic
influences as being the cause of O levels above the 80 ppb National Ambient
Air Quality Standard (NAAQS). The study by Decker et al. at Research Triangle
Institute (RTI) is one of the more definitive research programs in this regard
(16). Two large geographic areas were covered by RTI in extensive aerial and
ground monitoring programs.
Gulf Coast and Texas Areas
One of the RTI study areas included the Texas-Louisiana gulf coast and
extended over both land and water areas in the summer and fall of 1975. This
gulf coast study area is of particular importance because earlier assessments
of Texas 0 levels had led authors to conclude that widespread O concentra-
tions could not be explained by anthropogenic factors because O3 was frequent-
ly high with onshore winds at stations such as Corpus Christi. Thus it was
concluded from these early local studies that the O must represent the impact
of natural sources.
In a subsequent assessment of the "Texas problem," Price came to the con-
clusion that high O days in Texas (days having a maximum 1-hour average 0
greater than 150 ppb) were normally the result of the combined effects of old
photochemical pollutants and fresh new pollutants (35). The accumulation of
sufficient ambient concentrations was favored by weak winds corresponding to
weak pressure gradients associated with the synoptic weather accompanying the
high 0 periods. Price concluded that local sources were the major problem in
Texas although some transport over larger areas could have been involved.
24
-------
Price determined that correlations of maximum 0 levels with other cumulative
pollutants or pollutant effects such as CO, nonmethane hydrocarbons, N0_, or
visibility were not significant or especially informative. This result is in
contrast to the higher and apparently significant 0 general pollutant correla-
tions reported by Husar et al. (13) and may be indicative of geographic
differences in this large scale or "air mass" O3 problem.
The conclusion that natural tropospheric O was not a factor in the high
O events in Texas was also supported by Price's review of the prevailing
synoptic weather systems. It was shown that none of the high 0 days was
associated with frontal passages which would be generally indicative of wide-
area vertical motion and deep tropospheric mixing.
In contrast to the study of Price, which was based on a careful assess-
ment of available data from a number of sources, the RTI gulf coast study (16)
was a comprehensive ground and aerial sampling research program that included,
in addition to 0 measurements, supporting data on nitrogen oxides and hydro-
carbons. Both aircraft profiles and ozonesonde measurements were made to
provide O data on the upper layers of the atmosphere.
The analysis of the RTI aircraft data showed that there were four identi-
fiable aerial distributions of 0 in summer and fall seasons in the gulf coast
area:
Area-wide low concentrations, i.e., O up to 35 or 50 ppb.
Localized plumes downwind of major source areas.
General, larger than a plume, regions of elevated 0 where some
values exceeded 80 ppb, and where the concentrations in various
situations tended to increase from west to east.
Large area (e.g., Louisiana to North Carolina) 0 concentrations
generally exceeding 80 ppb.
Some of the more important facts about the gulf coast 0 situation were con-
cerned with the meteorology of the high (and low) 0. events.
25
-------
As Price had noted, the RTI study found that slow moving air (i.e.,
weather systems with weak winds and weak pressure gradient) moving over large
hydrocarbon sources was associated with the higher concentration O_ events at
both urban and rural locations. The wind trajectories associated with these
high concentration situations characteristically show anticyclonic curvature
indicative of flow in a high pressure system. By way of contrast, situations
with stronger winds and less of a tendency for anticyclonic curvature showed
lower 0 levels. Trajectories with long overwater fetches were also associ-
ated with lower 0 concentrations. In those situations when elevated O was
measured over the Gulf, trajectory analyses usually were able to show that the
air stream had a relatively recent (within 24 hours) history showing passage
over land areas and often over concentrated source areas. The documentation
of this type of an event is obviously the answer to the much earlier Texas
examples of high O in the onshore sea breeze at places like Corpus Christi.
The finding is obviously similar to Bell's California result.
Trajectories in the Texas area also documentated situations of inter-city
sport, and one sue!
80 ppb at Austin (16).
transport, and one such situation may have been the cause of O in excess of
When area-wide high concentrations were observed, a stable layer that
could limit vertical mixing was usually observed below 2 km. This would also
be characteristic of a large, high pressure system. The observation would
also be consistent with the data from the ozonesonde observations showing that
although mid- to upper-level ozone concentrations could change by 50% during a
day, these changes did not reach to the lower layers of the atmosphere and
affect ground-level O concentrations.
Thus the conclusion reached by both RTI and by Price that anthropogenic
emissions are the main cause for O, values exceeding 80 ppb in the Texas gulf
coast seems well founded.
Upper Midwest Regions
In the area of the upper Midwest there are two groups that have carried
26
-------
out extensive 0 studies that are relative to both 0 transport and to urban/
rural relationships. The two main groups that will be cited in this analysis
are RTI (16) and the Washington University group (13,36).
In these major studies in the upper Midwest, the RTI program attempted to
follow developing high pressure systems as they moved eastward from Montana
(16). Tracking was done by aircraft, a string of ozone monitors on the ground,
and an ozonesonde program. The program of Washington University was directed
toward assessments of the urban plume from St. Louis that typically stretched
to the northeast of the city (13). This group also analyzed some air mass
transport systems on the basis of areas and periods characterized by low, non-
fog visibility.
The RTI upper Midwest study showed that high O concentrations in the
summer months across these states were associated with the passage of high
pressure systems. Within the anticyclone, the lowest 0 concentrations are
found in the leading or eastern portion of the anticyclone, and the highest
concentrations are found in the trailing or western portion of the cell. This
pattern has been recognized for a number of years and has been described by
different authors on the basis of independent investigations (31,37). The
change in 0 in the anticyclone is attributed to the fact that the air parcels
in the leading portion of the system have had the shortest residence time in
the anticyclonic regime and thus the least exposure to anthropogenic pollutant
sources under synoptic conditions favorable for pollutant accumulation. This
is in contrast to the fact that in the trailing or western portion of the high
pressure area the parcels would in general have accumulated the largest
exposure time to anthropogenic sources and to weather conditions favoring
pollutant accumulation. In terms of travel across the multistate area the
daily maximum concentration or the total hours exceeding 80 ppb O increased
in the high pressure system as it moved from west to east.
In North Dakota, near the origin region for the systems followed in this
migratory anticyclone study, a series of ozonesonde flights in 1975 failed to
show that O3 intrusions from the stratosphere were important in the surface
layer chemistry. Thus it was concluded that natural O3 was not an important
27
-------
factor in the subsequent development of excessive ozone in these moving
weather systems.
These moving, high pressure areas gradually accumulated concentrations of
other pollutants as well as O, as the system moved along. This was shown by
hydrocarbon and halocarbon measurements as well as by the sulfate and nitrate
content of the total suspended particles in the atmospheric samples.
Comparative analyses of changes in the average weather patterns for 1973,
1974, and 1975 were made. It was determined that, of these several years,
1975 had fewer hours of high pressure, less high O , and lower average O
during the passage of high pressure systems. This phase of the study points
out clearly that year-to-year changes do occur, and thus such facts must be
accounted for in programs dealing with photochemical oxidant.
The vertical motions that are usually postulated for a high pressure
system do not explain the higher O levels on the back side of the system.
Subsidence and surface divergence are usually most pronounced in the center
and in the leading edge of the high. This should be accentuated by the
vertical motions related to the preceding trough and frontal system as well.
The RTI report thus argues that the increase at the surface was due to an
increase in O synthesis in the surface layers. Likewise, it is concluded
that there is an increase in O synthesis in the eastern areas of the U.S. as
compared to regions in the west.
The RTI authors made some detailed model calculations to estimate the
likely residence time within a high pressure system of air parcels having an
origin in various parts of the system (16). It was determined that parcels
initially in the northeast quadrant had the longest potential resident time,
about 6 days, and that parcels on the western side of the circulation pattern
had been in the system the longest time. This latter conclusion is in general
agreement with O measurements in pollutant source areas such as have been
described by Westberg et al. (37). A very important point is made by the RTI
authors (p. 161): an air parcel does not move across the country as a perma-
nent part of a high pressure system. Rather, it is a case of a parcel moving
28
-------
in on one side of the system and out on the other as the pressure pattern
moves across the area. Only under a weather pattern with a unique series of
coordinated systems would there be a direct net transport of air from area to
area continually under the influence of a given high pressure system. With
this in mind, the RTI findings that O patterns in the west and the east must
be explained by different chemical relationships between precursors and de-
structive agents points up the rather obvious fact that emission character-
istics are different in the two areas. The fact that both areas observe
higher concentrations of O with the transit of a high pressure system is
probably an example of concentration effects rather than a result of similar
chemistry.
With regard to transport, the set of model calculations by RTI would
appear to show that even though the continuous travel of air pollution effects
(e.g., lowered visibility) can be mapped across the country, perhaps for 200
km, the individual air parcels and their entrained pollutants had much shorter
residence times within the anticyclonic weather system. The haze "blobs"
studied by Husar et al. (13) are a case in point. In this case hazy condi-
tions as a result of entrained air pollutants were tracked from June 25 until
July 5, 1975, a period of 10 days. The "blob" of low visibility conditions
was initially observed in the Midwest near St. Louis on June 25, and after a
tortuous path, was determined to be affecting Atlanta, Georgia, on July 5. As
indicated by the RTI model, single parcels and given masses of pollutant would
not persist in the system for such an extended period. What does persist,
however, are the atmospheric conditions that favor the accumulation for periods
of one to a few days of pollutants from a succession of more local sources
at relatively high concentrations. Since some considerable amount of wind
movement is necessary to produce a long-distance transport system and at the
same time low winds are necessary to produce a high concentration of pollut-
ants, there must be some sort of delicate balance among the factors of source
strength, dilution by wind, and transport flow.
The studies in the St. Louis area have included a large number of groups
and these have produced an even larger number of papers, many of which have
dealt with 0 transport from the St. Louis urban area (13,36,38). These
29
-------
papers as well as many others show a reasonably consistent picture: an urban
plume is found stretching downwind, typically toward the northeast, from St.
Louis for distances of up to 160 km (36). At times White et al. report O
concentrations occurring in excess of 300 ppb. Weather conditions obviously
have to be favorable for long distance transport and minimum dilution during
such travel. These are typically found in the anticyclonic systems that
traverse the area in the summer. This results in a situation where the urban
plume is imbedded in an already dirty air mass. For example, White et al.
note that on July 18, 1975, when the St. Louis urban plume was detected by
aircraft sampling out to 160 km, the air mass O concentrations outside the
plume were at levels up to 70 ppb at the 160 km distance. Thus the whole air
mass affecting the region was accumulating a significant background of O even
while the major urban plumes were making significant contributions. The fact
that the St. Louis plume was so persistent under conditions when the whole air
mass had above normal concentrations of ozone is considered by this author to
be due to meteorological factors being present in the high pressure area and
not to a chemical synergistic system between the air mass and the urban plume.
The intensity of the analysis of the St. Louis Plume has led to attempts
at chemical modeling of O reactions in the plume system. White et al.
conclude that the apparent 0 yield on the basis of hydrocarbon emission rates
is close to the stoichiometric upper limit (36). Some arguments are made that
this apparent high yield may be due to the influences of unaccounted-for
hydrocarbons brought into the area as part of background concentrations, as
mentioned above. No data are brought in to support this argument. However,
the air masses where such enhanced O formation apparently occurs has sup-
ported considerable O- formation already, as indicated by the observed 70 ppb
"background," the St. Louis plume may be gaining some additional O from a
revitalization of spent urban emissions present in the air mass as it moves
across St. Louis. This level of anthropogenic emissions should not be con-
fused with natural or biogenic emissions.
In transport situations it is important to differentiate between a bio-
genic emissions background and an anthropogenic background. The biogenic air
mass background, such as was observed by RTI in Montana anticyclones, has been
30
-------
shown by Grimsrud and Rasmussen to be poor in NO and to have a strong ozone-
formation potential after the addition of NO (39). In contrast, an anthro-
pogenic air mass background as is characteristic of rural areas in the Midwest
shows little 0 response to an increase in NO . In fact in many cases the
J *»
response of experimental air parcels is a decrease in O formation compared to
an air parcel that contains only the ambient air NO concentration. This has
X
also been shown by Grimsrud and Rasmussen.
Thus across the northern part of the U.S., from Montana to the east
coast, there is strong evidence for the existence of a west to east gradient
of photochemical O production. Because single air parcels do not complete
this total track, the buildup and change in the 0_ content of anticyclones
should be attributed to changing characteristics of regional sources, espe-
cially in the density of sources, and perhaps to an increasing intensity of
the moving anticyclonic systems. In the Midwest and probably on the east
coast, air mass background levels are more characteristic of a spent urban
pollutant mixture and may show a negative potential for producing O with an
increase of NO . Thus when an urban plume is dispersed into a Midwest high
pressure system, it must possess all of the O precursors and not just the NO
levels as might be the case for a clean biogenic background. Because of the
higher levels of pollutants in an anthropogenic background situation, and the
likely presence of O scavengers, the impact of an urban plume would probably
be less than would occur in a cleaner biogenic background. Along the gulf
coast it has already been mentioned that urban plumes marked with excessive
ozone were readily detected. This may be indicative of a lower anthropogenic
background in this region. However, the background measurements of RTI always
showed some levels of anthropogenic emissions, so the gulf coast was not an
example of a biogenic background area (16).
Northeast U.S. Regions
The northeastern part of the U.S., including the area generally north and
east of Philadelphia, contains a dense population and probably has the largest,
most congested, commercial-industrial area in the U.S. in the area of New York
and New Jersey surrounding New York City. In addition, the long coast line
31
-------
complicates the regional wind patterns by introducing local land and sea
breeze effects.
Oxidant patterns over the Northeast have received considerable attention
because high concentrations have been measured almost routinely in rural and
distant suburban areas for a number of years. One group that has led in the
regional analysis of this area has been the Bell Laboratory group under the
leadership of W.S. Cleveland and T. Graedel. Their work has emphasized the
interpretation of the air quality data, particularly ozone, by the various
agencies in the area rather than in mounting a field measurement program.
The result of the Bell Laboratory work has been to show clearly the urban
plume effects in this area and the displacement of maximum O concentrations
downwind of major source areas such as New York City and Philadelphia. The
regional transport in this area that is most important for O control strategy
is the travel to the northeast of the New York City - New Jersey urban plume
with apparent .effects in Connecticut and Massachusetts (40). This transport
is characterized by a number of features that are not unexpected, given our
present understanding of the urban photochemical reaction system. These are:
The central urban areas show low O. levels because the transport of
effluents out of the city moderates the pollutant concentrations, the
time required to form O levels, and the high concentrations of NO
scavengers in the local emissions.
The highest O concentrations occur outside the main urban area along
the wind trajectory at a distance determined by 1-2 hours of midday
wind travel.
These peak 0 concentrations occur during the early afternoon at the
time of maximum solar radiation.
Beyond the area of peak O concentrations the average time of the
peak O is progressively later in the day and is considered indic-
ative of the movement with the regional wind of the peak O3 formed
earlier in the day.
32
-------
Although there are undoubtedly 0 impacts from the larger local urban
areas along the track of the New York City plume, these are frequent-
ly overshadowed by the transport from the New York City source area.
These previous statements are based generally on the statistical analysis by
the Bell Laboratories group of summer season 0 data files (40).
The assessment of individual days and 0. events in the Northeast has been
carried out by a number of groups including Washington State University (37),
Battelle-Columbus (41), and EPA-Las Vegas (28). These analyses resulted from
the extensive 1975 Northeast Oxidant Survey involving these three research
groups. In general the results are in agreement with the results from the
Bell Laboratory research and the importance of O advection across the area
has been clearly established. One feature that the daily research data show
clearly is the importance of the urban plume from New York City moving east-
ward along the southern shore of Long Island. This situation is not well
defined by the Bell Laboratory results. This plume south of Long Island forms
a source for 0 concentrations occurring along the Connecticut and Rhode
Island coasts under typical sea breeze patterns. These coastal areas can
receive significant advection of 0 from the New York City urban plume under
wind conditions that do not carry the plume directly over land areas (37).
Spicer was able to develop a simple regression expression to relate
ambient O and fluorocarbon levels at two rural Connecticut locations. Since
the fluorocarbons are unique, stable anthropogenic emissions, these regres-
sions provide strong evidence for the impact of the urban plume at Simsbury
and Groton, Connecticut. These results obviously would apply to other areas
as well.
Weather patterns seem to play a lesser role in establishing O pollutant
levels in the Northeast than is seen either in the Midwest or gulf coast
areas. It is, of course, recognized that the general pollutant loads reflect
the general weather patterns, but it seems that, on a subjective basis, there
is a much broader set of weather conditions that can accompany the occurrences
of high 0 levels in this area and less sensitivity, for example, to tae
33
-------
location or intensity of an anticyclonic system. This could be a result of
the population density over the area and the relative dominance of urban
plumes over the anthropogenic background levels in the air mass.
Local circulations in the surface layers play a dominant role along the
coastal areas in the Northeast as noted by Westberg (37). This dominance was
not as apparent in the Midwest and gulf coast areas, although important local
systems have been noted in the Great Lakes area particularly around Chicago.
Discussion of Transport Relative to Natural Ozone Sources
In general, this writer is of the opinion that the dominant effect
represented by rural 0 concentrations is the transport of anthropogenic
pollutants. It is recognized, however, that there are areas where natural O_
and, in particular, stratospheric 0 may occur in important concentrations.
As previously mentioned in the gulf coast discussion, stratospheric intrusions
of 0 are related to strong cyclonic storms that affect a deep layer and
result in a breach of the tropopause at the polar jet stream. Danielsen and
Mohnen have assessed tropopause folding as one specific mechanism for the
injection of stratospheric ozone into the troposphere (42).
The question of the importance of natural O3, in what would probably be
described as an anthropogenic background situation, comes up in the several
papers reporting measurements on Whiteface Mountain, New York, by Mohnen and
Coffey and their associates (42-44). In these papers the fact that O con-
centrations measured at the top of Whiteface Mountain (altitude 4860 ft) show
patterns that are out of phase with urban and rural stations at lower alti-
tudes is used to support the importance of a stratospheric 0 source in up-
state New York. The analysis is not convincing in that when downward trans-
port of 0 is argued the pattern could also be explained by a transport of
destructive or scavenging factors out of the nearby urban areas and a longer
residence time of photochemical O. at the elevation of the mountain top under
less pressure from scavenging reactions. The increased persistence of O, in
upper layers of the atmosphere has been documented in detail in southern
California. This would seem to be a better explanation of the Whiteface
observations than is stratospheric subsidence, except when situations of
34
-------
tropospheric folding are likely in the vicinity. These would be mostly pre-
sent in the spring. There is no reason that a single cause has to be given
for all situations at this location.
Ozone reported at the ground in remote locations may well be due to
stratospheric injection (45) or to photochemical reactions involving natural
hydrocarbons (46). Chatfield and Harrison show that O measurements in a
remote coastal area of the Pacific Northwest may range up to about 70 ppb with
the mode for monthly data between 40 and 55 ppb. When high natural ozone
occurs there are generally indications of more vigorous weather systems, and
under such conditions there seems to be little chance for confusion with the
summertime anticyclonic type of high ozone incident. The question of natural
biogenic hydrocarbon production of O is harder to resolve and separate from
the anthropogenic system because the photochemical production would be a
maximum under the same conditions that the anthropogenic cycle would. However,
the biogenic concentrations would be lower, and at present there seems to be
no evidence of excessive (i.e., concentration of 80 ppb or higher ) 0_
production from just biogenic emissions.
TRANSPORT FROM UPWIND SOURCES AND LOCAL PROBLEMS
Ozone transport from one urban area into a less dense or smaller urban
area can be a major factor in the O pattern observed in the smaller area.
Such situations are observed clearly in the Northeast because of the massive
nature of the New York City urban plume. Along the coast of Connecticut, it
is usually immaterial whether the emissions from small towns such as Groton
are included in the daily O pattern or not since the O concentrations are
dominated by the New York City urban plume carried by the sea breeze. The
obvious question in the Northeast is how big does the downwind city have to be
before less than 50% of its O would come from the New York City urban plume?
This cannot be determined in any fixed manner, but a review of the results
given by Cleveland et al. seems to indicate that the Hartford, Connecticut,
area may be in the size range where 50% or less of its ozone problem can be
due to advected O (40). The 1970 population of the Hartford metropolitan
area was about 700,000 with the city of Hartford having a population of about
160,000.
35
-------
The question of upwind emission of 0 precursors and the subsequent
formation of photochemical O over the downwind receptor area is another
variation of 0 advection. The separation of 0 advection or transport from
O J
precursor advection is a difficult question to resolve. One way that could
probably lead to a qualitative answer is the area-wide statistical analysis of
O data as done by Cleveland et al. The longer term relationships developed
in such a study could show an O pattern more closely linked to the large
upwind urban area. Careful local studies of air chemistry could also be
carried out such as was done by Westberg (37) ; however an unequivocal answer
would be hard to obtain and expensive.
In situations where the upwind area is smaller than the receptor area,
the evaluation of the impact on a larger population center in terms of either
O or precursors would be hard to measure to any degree of accuracy.
In all of these cases, calculated impacts derived from meteorological and
chemical models can be obtained. The results would be questionable becaxise,
as yet, there are only crude photochemical models available. However, consid-
ering the problems that are encountered in field programs the modeled esti-
mates may be as useful as any but the most comprehensive field study programs.
OZONE AND PRECURSOR TRANSPORT DISTANCES
Identifiable transport of O over distances of 100 miles or more has been
observed on a number of occasions, and the data reported by Siple et al.
indicate that the New York City urban plume could be identified as an O plume
for 400 km (250 mi) out over the Atlantic (28). This is about a 24-hour or 1-
day trajectory and may be about the limit for the identification of a parti-
cular plume. Beyond a 1-day transport distance, it would be expected that
local meteorological conditions, such as turbulence and wind shear, would act
to break up the identifiable plume; the pollutants it contained then would be
more evenly mixed and included in the anthropogenic background levels. Cox et
al. identified O situations in England and southern Ireland that apparently
had trajectories of over 600 km from European source areas (7),
36
-------
In many situations an O plume may persist in layers above the ground,
but emissions of NO will completely scavenge the surface layers. This is
commonly seen in city areas at night where 0 will go to zero at the surface
but may still be present aloft. This is a typical situation in southern Cali-
fornia and was clearly observed over Canton, Ohio, by WSU in 1974. When this
layer aloft is present at high concentrations, its mixing to the surface may
make a major contribution to the next day's O problem. These situations can
be identified by field measurements using aircraft data taken during stable
night and early morning hours. Fumigation situations might also be identified
by a careful analysis of the O time series and the local meteorology. It
would probably be necessary to have such models verified by measurements aloft
using aircraft.
The transport of O precursors rather than O doubtless occurs because 0,
seems to develop in the urban plumes downwind from the city. White et al.
reported such an incident for the St. Louis plume (36). The unique identifi-
cation of precursors in such plumes has generally been much less successful.
Decker et al. tried to show NO buildup in the migratory anticyclones, but the
X
levels were within the noise of their instruments (16). Detailed resolution
of hydrocarbon species with regard to O precursors also seems to be doubtful
at this time. Thus the transport of O precursors probably cannot be demon-
strated at this time to occur over as long a distance as the transport of 0
can be shown to exist. This may be more of an analytical problem than an
indication that the phenomenon is not an important one.
CONTROL STRATEGIES RECOGNIZING OXIDANT TRANSPORT
Ozone or photochemical oxidant control strategies must be developed that
take into account the extended transport that can occur between source areas
and receptor regions. Thus the present concept that excessive concentrations
at a given location must be countered by some control of emissions in the
vicinity of the station must be expanded so that transport situations invol-
ving distant sources can be recognized and covered by special regulatory
procedures. For example, it is obvious that O concentrations along the
Connecticut coast around Groton will never change significantly until the New
37
-------
York City urban plume is reduced. Local control, even to the extent of moving
the population out of the area, would probably not even be observable in local
O records. Other suburban areas in other parts of the U.S. could also be
found where a similar situation prevails.
A control approach that seems reasonable to this investigator is one that
is based first on control of the vehicle to the maximum practical extent
without regard to its area of operation, and second the initiation of urban
area vehicle travel control programs in selected major urban areas. The
selection of those areas to be subject to area-wide transportation control
plans would be based on two possible monitoring situations:
The present concept of local adverse O concentrations, but sub-
stantiated by good evidence that local sources are sufficient to
produce the observed effect.
The transport modified criteria where the relevance of upwind sources
is judged and where meteorological and air chemistry data substant-
iate the regulatory decision. Here transportation control steps are
prescribed in the upwind area rather than in the area where the
excessive 0 levels were measured.
There are many obvious problems in the implementation of such a control
scheme, but compared to the inadequacies of the present system the usefulness
of a change seems well worth the effort.
Probably one of the most serious regulatory problems is the adoption of
apparently unnecessary controls. It seems that this problem would be less
likely to occur in the scheme of regulations described above than under the
present local control system. Where atmospheric transport is taken into
account, agency control efforts could be concentrated in the larger and more
congested urban centers, such as New York City, because these are the source
areas that form significant urban plumes. This would seem to be a more
publicly acceptable program.
38
-------
REVIEW AND ANALYSIS
P.M. Vukovich
INTRODUCTION
The literature suggests that there are two predominant scales of trans-
port that markedly affect the ozone concentration at the surface of the earth.
These scales are: (a) the mesoscale transport that encompasses the effect of
the urban heat island circulation, land and sea breeze circulations, and air
pollution plumes downwind of major cities; and (b) synoptic-scale transport
that appears to be closely associated with high pressure systems. These
scales of transport do not designate necessarily the mechanism responsible for
high ozone.
There is considerable controversy as to the principal mechanism respon-
sible for levels of high ozone observed in locations remote from sources. Two
conceptual models have been proposed. One suggests that direct transport of
ozone is the principal mechanism. The ozone originates in urban areas or the
stratosphere. The other model suggests that ozone precursors are transported
to remote locations where they combine with naturally-emitted precursors to
produce high ozone.
The credibility of these hypotheses is examined in this paper. The
available information is employed to identify the strengths and weaknesses in
each. From this analysis, a conclusion as to the dominant mechanism is
suggested.
CHEMISTRY OF OZONE IN REMOTE REGIONS
Before considering oxidant transport, it is important to review recent
39
-------
concepts on ozone chemistry in the rural environment. Recent outdoor smog
chamber experiments by Sickles et al. indicate that in a partially spent
simulated urban photochemical system, ozone was generated in concentrations
above the NAAQS with low concentrations of NO and high NMHC/NO ratios (32).
The smog chamber results indicate that dilution of the urban system results in
more efficient generation of ozone per molecule of NO. It was further noted
that at the end of a diurnal cycle a high minimum ozone concentration existed.
These data strongly suggest that higher concentrations of ozone should be
found in diluted plumes downwind of urban regions than in the urban regions.
It will be shown later that this phenomenon is substantiated by observations
in urban plumes.
The presence of large minimum concentrations in a dilute environment also
suggests that the potential exists for producing high ozone concentrations on
the following day without requiring either high generation or transport. This
phenomenon will be substantiated.
Bufalini et al. have shown that all atmospheric organic compounds are
capable of producing ozone (5). Larger molecules of organic compounds tend to
produce larger concentrations of ozone. The reaction times of organic com-
pounds in producing ozone vary substantially, ranging from a period for 99
percent reaction of one-half hour (trans-2-butene) to 150 days (methane).
Most organic compounds have 99 percent reaction times of less than 12 hours.
Carbon monoxide and acetylene have reaction times on the order of 8 1/2 days,
whereas the reaction time of ethane is on the order of 4 days. These data
suggest that the most reactive organic compounds are removed rapidly through
ozone production. These reactions should take place either immediately at the
source (e.g., the urban region) or immediately downwind of the source. The
less reactive compounds would stay in the air parcel and accumulate. These
would react at some later time (1, 2, or up to 8 days later). They could
combine with more reactive compounds that may be injected into the system to
produce high levels of ozone, particularly if a relatively high minimum
concentration already exists.
40
-------
Vukovich et al. have produced evidence that suggests that the less reac-
tive hydrocarbons are important in producing high ozone levels (6), Data
indicate that ozone concentrations began to decrease after a frontal passage,
reaching a minimum somewhere between the frontal passage and the center of a
following high pressure system. Following the ozone minimum, the ozone levels
increase to a maximum on the back side of the high pressure system. Calcula-
tions show that on the front side of the high, an air parcel has been in the
system usually less than a day. However, on the back side of the high, the
air has been in the system for at least 2 and as many as 6 days. It is impor-
tant to note that the high pressure system offers the proper environmental
conditions (high solar radiation, low wind speed) for photochemical production
of high ozone (2). The correlation between high air-parcel residence time and
high ozone suggests that sufficient time exists for the legs reactive hydro-
carbons to accumulate and to react.
Vukovich et al. further showed that when high ozone is observed in a
rural environment, it is usually in the presence of a high diurnal minimum
from the previous day's diurnal cycle. These data showed that an extremely
large generation or transport of ozone was not required to produce high ozone.
The conclusions of Vukovich et al. are in contrast to Lonneman's (47).
Lonneman gave evidence that he said eliminated photochemical production of
ozone in the rural environment as the mechanism for producing high ozone
levels. He concluded that transport of ozone from upwind sources must be
responsible for the high ozone levels. Lonneman cited measurements of ozone,
hydrocarbon, and NO in the rural environment near Wilmington, Ohio, during
X
August of 1974. Of the 14 observations presented, only six had diurnal maxi-
mum ozone concentrations (hourly average) greater than 80 ppb (the NAAQS).
Using Dodges' model (48) and the observed NO and hydrocarbon concentrations,
X
he showed that in only two of the cases were there sufficient NO and hydro-
x *
carbon to photochemically produce ozone concentrations greater than 80 ppb.
His data indicate that in neither of these cases did the ozone actually exceed
80 ppb. It was on this basis that Lonneman concluded that photochemical
production of ozone could not be responsible for the high ozone observed.
41
-------
However, this fails to take into account the fact that, normally in rural
areas, the nocturnal minimum ozone concentration is relatively large. For the
summer of 1974, the average diurnal minimum at Wilmington, Ohio, was 20 ppb
(2). Vukovich et al. have further shown that prior to the occurrence of high
ozone in the rural environment, the diurnal minimum increases substantially.
Therefore, it was only necessary that less than 60 ppb ozone be generated by
photochemical processes to produce ozone greater than or equal to 80 ppb at
Wilmington, Ohio. The data presented by Lonneman using Dodge's model suggests
that 35 to 60 ppb ozone could have been generated easily by photochemical
processes in the other cases.
Vukovich et al. show, using a simplified model, that synthesis and
destruction of ozone are greater in the eastern part of the United States than
in the western Great Plains region (6). This leads to the hypothesis that
high ozone is correlated with high population density (16). The implication
is, of course, that population density is proportional to the anthropogenic
emissions of ozone precursors. The population density in the Great Plains
area is on the order of 13 people per square mile, whereas, in the eastern
part of the United States, it is on the order of 213 people per square mile.
This hypothesis further suggests that each urban area, large or small, acts as
a point source for ozone precursors. These are diluted in the air mass, but
are sufficient for the genesis of high concentrations of ozone. The origin of
the precursor in any single air parcel becomes obscure due to the scale of the
transport. Though strong circumstantial evidence exists that supports this
hypothesis (2,6,16,31,46), the results are not conclusive.
The above model implies that synthesis of ozone occurs in the rural
boundary layer and that some of the precursors of ozone found in the rural
boundary layer have urban origins. Although the evidence implies that this is
the case, it is circumstantial. In the following section, further evidence to
support this model is presented.
42
-------
MESOSCALE TRANSPORT
There exists sufficient evidence to show that high ozone exists in plumes
emitted from urban areas (38,49-51). These plumes extend anywhere from 60 to
3000 km downstream. In some cases such as the New York plume, large areas are
affected. However, in most cases, such as the St. Louis plume, a very local-
ized area downwind of the city is affected. Apparently, the affected area is
a function of the area of the source (the area of the city).
An important point to note is that the maximum ozone level is found in
the plume downwind of the city rather than in the city. Two potential proces-
ses may produce such a maximum. These are mass convergence and photochemical
generation. Mass convergence requires some meteorological anomaly that would
cause convergence of mass within the plume. Since the downwind maximum is
found in all cases reported, this suggests that if it is 3 meteorological
anomaly, the anomaly must be directly associated with the plume. However/
there is no known forcing function associated with urban plumes that would
cause a meteorological perturbation capable of producing mass convergence. It
is therefore concluded that the principal mechanism for producing high ozone
in the urban plume downwind of the city is photochemical generation. It is
further suggested that the dilution effect discussed by Sickles et al. accounts
for the downwind maximum (32).
Decker et al. have presented data from the DaVinci II experiment to show
that during the daytime hours, the ozone is high both at the surface and aloft
in the plume downwind of the city of St. Louis (9). However, at night the
ozone concentrations aloft remain high, whereas the ozone concentrations at
the surface diminish significantly. In the morning, the ozone aloft still is
high, whereas the ozone at the surface begins to increase. Their observa-
tional period ended before the end of the diurnal cycle for the following day.
Based on these data, Ripperton et al. have established the following
hypothesis (15). The high ozone observed during the day downwind of an urban
region is a result of photochemical processes associated with precursors
43
-------
transported in the urban plume. At night, a low level of inversion is devel-
oped by radiation processes. The inversion prevents mixing with the air
aloft. Surface-based, ozone-destructive agents destroy the ozone trapped in
the surface layer. Because mixing with the layers aloft is inhibited, ozone-
destructive agents do not reach the air aloft, and the ozone there remains
relatively unchanged. When, on the following day, solar insolation heats the
surface layer, mixing brings considerable ozone to the surface. This hypoth-
esis is supported by the observations of Coffey et al., Chatfield and Rasmussen,
and Jerskey et al. (11,46,50).
The notion that vertical mixing of ozone is a principal mechanism leading
to high ozone concentrations in the surface layer appears to be controversial,
since much of the data that indicates mixing to be the principal cause for
high ozone is from point observations. For instance, the data of Decker et
al. consist of a point observation made aloft by DaVinci II and observations
made at the surface by a mobile van. Coffey et al. report a point observation
made at Whiteface Mountain that is compared to point measurements made at
stations at lower levels, as far as 60 km from the Whiteface Mountain site
(11). In none of these cases are there any real computations of vertical
mixing, only the implication that since ozone is high aloft the flux must be
downward and that this produces the observed high ozone at the surface later
in the day.
This highlights an important shortcoming of ozone studies to date. There
are no significant data on the diurnal variation of the vertical profile of
ozone through the boundary layer. Only three such profiles have been reported
(2,52).
These three profiles, though taken in different places, show similar
diurnal variations. Before sunrise, relatively low concentrations of ozone
are found in the surface layer compared to that aloft (see Figure 1). The
concentrations of ozone aloft exceed the standards. At midday, the ozone
concentration from the surface to the top of the boundary layer is approxi-
mately constant indicating a well-mixed boundary layer. In the late afternoon
44
-------
11000 -i
100
i r
150 200
Ozone (yg/m3)
Figure 1. Diurnal variation of the vertical distribution of ozone
at Wilmington, Ohio on 1 August 1974. These profiles have
been smoothed using both the ascent and descent portions of
aircraft flights where possible. The indicated times are the
mean time (EOT) of each profile. Altitude is measured from
the surface. (Unsmoothed data are in reference 2.)
45
-------
around the time of maximum ozone at the surface, these profiles indicate that
the ozone in the surface layer is significantly greater than the ozone aloft.
The importance of this fact will be examined using Table 1.
TABLE 1. INTEGRATED AVERAGE OZONE CONCENTRATION FROM THE
SURFACE TO 5,000 FT VERSUS TIME OF DAY USING DATA
IN FIGURE 1.
Time (LOT)
-3
Average Ozone (\ig m )
0720
155
1350
185
1725
225
The data in Table 1 show the integrated average ozone concentrations,
where the integration was accomplished from the surface to 5,000 ft using the
data in Figure 1. Mixing redistributes mass in a vertical column such that
the resultant concentration of mass at any level within the column equals the
average mass concentration in the column. In this particular case, the
average mass concentration early in the morning is 155 yg m . Mixing could
potentially increase the surface concentration from 80 yg m (Figure 1) to
155 yg m where the additional mass (approximately 75 yg m ) is transported
from layers above 1,000 ft to approximately the top of the mixing layer at
5,000 ft. If mixing alone produced the increase of ozone at the surface, then
the mean concentration in the layer from the surface to 5,000 ft should have
remained constant over time.
However, the data in the table show that, between 0720 LDT and 1350 LOT,
there was an average increase of approximately 30 yg m in the ozone concen-
tration in the column. The total change of ozone at the surface in that time
period was approximately 104 yg m of which 75 yg m or 72 percent of the
change could have been accounted for by vertical mixing (but not necessarily).
However, the total increase in ozone in the column in the time period relative
to the profile one would get through mixing alone (i.e., a constant value of
-3 -3
ozone at approximately 155 yg m through the column) is a positive 185 yg m
Since mixing can only redistribute mass, it cannot account for this increase
in mass in the column. Furthermore, it is noted between the time period 1350
46
-------
LDT and 1725 LDT, that there was a further increase of approximately 45 yg m
in the column. A mass flux from the upper levels cannot be responsible for
this mass increase since the gradient of ozone indicates that the flux should
be from the surface to the upper levels.
These data suggest that, although, in the period after sunrise and
before noon, the observed increase in ozone at the surface may be substantially
influenced by vertical transport of ozone, there is a significantly more
important process influencing the increase of ozone in the boundary layer.
Besides the urban plume, there are other mesoscale transport character-
istics markedly affecting the ozone distribution. Hyde and Hawke discuss the
occurrence of high ozone associated with morning drainage flow and afternoon
sea breezes (20). They indicated that drainage flow brings precursors of
ozone from inland sources to the Sydney basin, and that the sea breeze brings
pollution air (in which significant photochemical processes have occurred)
back to the Sydney basin in the afternoon.
Kauper and Niemann discuss the influence of the upper-level flow assoc-
iated with the Los Angeles land breeze on the transport of high ozone to
Ventura County (18). They indicated that ozone-rich layers exist aloft over
Los Angeles, and that these are transported by the land breeze to Ventura
County where mixing brings the aged photochemical pollution cloud to the
surface. The authors do not offer concrete evidence that the air pollution
aloft comes to the surface in Ventura County. They only suggest this as a
possible mechanism for the high ozone in Ventura County. The conclusion that
the upper-level land breeze influenced the horizontal transport of ozone
aloft, however, was well substantiated.
Decker et al. studied the relationship of transport to high ozone obser-
vations in the gulf coast regions of the United States (16). The study was
concentrated in the two-state area of Louisiana and Texas. On this scale, it
was found that the higher concentrations of ozone were associated with slow-
moving air that passed over high precursor emission areas and arrived from a
non-prevailing wind direction. The lower concentrations were associated with
47
-------
faster-moving air having a long fetch over water. In those cases where high
ozone was associated with flow from the Gulf of Mexico, it was found that the
air had a previous history over the high precursor emission regions of Louisiana
and Texas (southerly flow) and that changes in wind direction (northerly flow)
brought this air back to the coast. Though the evidence is far from being
conclusive, it indicates that, on the average, air parcels from the source
region of the Gulf of Mexico contained low concentrations of ozone. This is
important because it infers that oceanic areas are sinks for ozone.
SYNOPTIC-SCALE TRANSPORT OF OZONE
Considerable evidence exists in the literature relating high ozone and
synoptic-scale high pressure systems (6,11,12,14,15,16,31,46). There is also
substantial evidence that the high ozone is found on the western side (back
side) of an eastward moving high pressure system. Though the relationship
between high ozone and high pressure appears to be well substantiated, there
exists considerable controversy as to the mechanism that produces the high
ozone in the high pressure system. Three mechanisms are discussed in the
literature. These are: (a) the direct transport of ozone; (b) the direct
transport of ozone precursors from the southern portions of the high pressure
system and subsequent photochemical production; and (c) the synthesis of high
ozone in air parcels that have large residence times in the high pressure
system. Though the last two appear to be similar (i.e., the mechanism is
synthesis), there is a fundamental difference between the basic character-
istics (the transport) of the system that allows the synthesis of ozone.
The conclusions reached by Coffey et al. and Dobbins et al. support the
notion that direct transport, either vertically or horizontally, of ozone is a
principal mechanism producing high ozone over a large area (11,17). Both
investigators suggested that above the nocturnal inversion, high ozone per-
sists throughout the night and serves to replenish the surface ozone concen-
tration during the daylight hours. The mechanism was discussed in some detail
in the previous section; however, there are further aspects concerning that
question.
48
-------
Both of the investigators conclude that vertical mixing is the primary
mechanism for producing high ozone at the surface. If this is true, then one
must ask why is high ozone associated with high pressure systems (a point that
Coffey et al. also noted). More significant vertical mixing occurs in low
pressure systems, yet no statistical correlation has been found between high
ozone and low pressure systems. Admittedly, cases of high ozone in low pres-
sure systems have been found, but these have been isolated cases.
Potential sources of ozone are the stratosphere and urban regions.
Transport of ozone from the stratosphere could be a mechanism for high ozone
at the surface. Reiter has shown that downward transport of ozone from the
lower stratosphere accompanies tropopause folding (53). Evidence exists that
in the springtime there is increased transport of ozone from the stratosphere
to troposphere (43,52). Though these data suggest the potential for ozone
from the stratosphere to contribute to the ozone found at the surface, there
is no concrete evidence to date that demonstrates that this ozone is respon-
sible for the large summertime high ozone concentrations found at the surface.
Furthermore, many of the more recent vertical profiles of ozone through the
troposphere in high ozone situations indicate the presence of a minimum con-
centration for ozone in the midtroposphere with ozone increasing above and
below that point reaching a maximum in the stratosphere and near or at the
surface (Figure 2). This suggests two distinct regions of ozone activity:
the upper layer having its source in the stratosphere, and the lower layer
having its source at the surface. This further indicates that the high ozone
in the lower troposphere is the result of some mechanism other than transport
of ozone from the stratosphere.
Sufficient data exist that show a preference for high ozone to occur on
the back side of a migratory high pressure system. In the high pressure
systems, mixing through a deep portion of the atmosphere is severely inhibited
by the presence of a subsidence inversion (12). It may be that tb«? subsidence
inversion is responsible for the local minimum at midtroposphere in vertical
profiles of ozone in high ozone conditions.
49
-------
30,000 r
20,000
-o
10.000
Date: 8/28/75
Time: 5:00 p.m.
Ground Level
0 0.05 0.10 0.15 0.20
Ozone Concentration, ppm
Figure 2. The vertical distribution of ozone over Indianapolis (52)
50
-------
Coffey et al. pointed out that urban areas have less ozone than rural
areas (11). This factor was also pointed out by Ripperton et al., who used it
to show that it is unlikely that direct transport of ozone from urban areas to
rural areas is the principal mechanism producing high ozone in rural areas on
a synoptic scale (15). If the urban region were the source of ozone and
direct transport were the mechanism, there could not be higher ozone in the
rural areas than in the urban areas. This is because the urban areas con-
stitute about one-fourth the volume of the United States air mass (the urban
area multiplied by the height of the daytime boundary layer) compared with
three-fourths for the rural area. Thus, volumetric mixing would yield smaller
concentrations of ozone in rural regions than in urban regions. Since ava,il-
able evidence indicates that rural regions have, on the average, larger con-
centrations of ozone than urban regions, this signifies that another mechanism
prevails.
Wolff et al. and Shenfield indicate that synthesis of ozone is most
probably the mechanism that produces high ozone on a synoptic scale in the
rural boundary layer (14,54). High ozone is concentrated on the back side of
the high where the flow is generally from the south to the north. This
suggested that there is generally transport of ozone precursors from the south
to the north part of the back side of the high where the high ozone is pro-
duced. The fact that long distance transport (_>_ 1000 km) of ozone precursors
occurs has been discussed by Cox et al. and Grennfelt (7,21). Flow from the
southern portions of a high pressure system to the northern portions of a high
pressure system on the back side only occurs when that high pressure system is
stationary. Under such conditions, it has been shown that high ozone is
concentrated over most of the area encompassed by the high pressure system.
Under these conditions, the highest ozone has been found to be located around
the center of the high pressure system where transport is minimized (2) .
Theoretical calculations in moving high pressure systems have suggested that
the motion of an air parcel through a moving high pressure system is basically
from the west through the east (9). For a high pressure system moving at 10
meters per second, the total travel distance of an air parcel in the high
pressure system will be about 350 nautical miles. About 60 percent of that
51
-------
travel distance will be in a west to east direction. Long-distance transport
(distances much greater than 1,000 km) in a south to north or north to south
direction does not occur in most of the moving high pressure systems.
As pointed out earlier, Vukovich et al. have shown a correlation between
the residence time of air parcels and high ozone in moving high pressure
systems (6). In such systems, high residence time and high ozone are both
found on the back side of a high pressure system. It was found that the
source region of air with large residence times was the northeastern quadrant
of the high pressure system. The data also suggest that vertical transport of
ozone is not responsible for the high ozone. Therefore, photochemical genera-
tion is concluded to be the principal mechanism.
The fact that the high residence time and high ozone correlated suggests
that a period of time is required so that sufficient or critical concentra-
tions of ozone precursors are injected into the air parcels while they remain
in the high pressure system where proper environmental conditions exist for
ozone production (i.e., high solar radiation, low wind speed). Decker et al.
have inferred that it is the less reactive hydrocarbons that -accumulate or
reach critical concentrations (9). After a certain time, these begin to react
and produce ozone. This mechanism is augmented by daily injection of the very
reactive hydrocarbons from anthropogenic and natural sources to produce the
high ozone in the rural boundary layer.
Decker et al. have also shown that, in the regions from the Rocky Moun-
tains to the east coast of the United States, there exists a gradient of ozone
such that the ozone is low in the west and high in the east (9). Furthermore,
Vukovich et al. have shown that, as a high pressure system moves out of Canada
to the western portions of the above-mentioned regions and, subsequently, to
the eastern portion of that region, the ozone concentrations in the high
pressure system were larger when the system was in the east (6). Sandhu has
_3
shown that 20 to 60 yg m is a typical background concentration for ozone in
the rural Alberta, Canada, environment (55). Alberta is a major source region
for continental high pressure systems that move into the United States east of
the Rocky Mountains. In the eastern portions of the United States, ozone
52
-------
concentrations in the high pressure system exceed the United States standards
for ozone. It was further shown that, though the ozone was higher in the east
than in the west and both photochemical production and destruction are la,rger
in the east than in the west in these migratory high pressure systems, the
amplitude of ozone (the maximum concentration of ozone minus the minimum
concentration of ozone) was not significantly different in the east from that
in the west. The real difference was in the minimum concentration for ozone.
In the east, the minimum concentration was three times greater than that in
the west. These factors have led Decker et al. and Ripperton et al. to
conclude that high ozone is related to high population density, which, of
course, is related to high anthropogenic emissions of ozone precursors (15,16).
Husar et al. noted a correlation between elevated ozone concentrations
and atmospheric visibility or haziness (13). A similar correlation was found
with high levels of sulfate. This is an important finding in that it suggests
a correlation between atmospheric haziness and air pollution, not specifically
ozone, which may define the dimensions of the problem. Though the correlation
seems to exist, the significance of the correlation is somewhat masked by the
fact that the visibility observations were made using data from NOAA weather
stations. These are normally located in major urban areas. It is impossible
to determine what effects local perturbations produced by the urban air pollu-
tion had on the visibility relative to the interpolation of the analyses into
rural regions. Husar et al. imply that the synoptic-scale zone of haziness
moves with the air mass, which then suggests the air pollutants move with the
air mass. The haziness was associated with a high pressure system in the
east. These systems move as a result of changes in the columnar divergence,
whereas haziness that is a result of suspended aerosols must move with the
wind. These movements are normally independent of each other. It is sug-
gested that the changes in haziness may be the result of the formation of haze
and the dilution of aerosols.
5.3
-------
SUMMARY AND CONCLUSIONS
There are two major scales of transport related to high surface ozone
concentrations: mesoscale transport and synoptic-scale transport. On the
mesoscale, the urban plume is most significant. The nature and scale of the
urban plume is dependent on the nature and scale of the source, i.e., the
urban area. Assuredly, direct transport of ozone occurs in the urban plume.
However, the fact that the observations show that there is a maximum ozone
concentration downwind of the urban region indicates that photochemical pro-
duction of ozone must occur since a meteorologically-induced mass convergence
is unlikely. The photochemical production is most probably a result of down-
wind transport of ozone precursors in the urban plume, combined with the
effect of dilution on the ozone chemistry.
Widespread regions of high ozone are found in the rural boundary layer
associated with migratory high pressure systems. The high ozone is generally
found on the back side of the high and only when the high pressure system is
in the eastern portions of the United States. Available vertical profiles
through the troposphere show that when the ozone is high at the surface, a
minimum ozone concentration is found in the mid-troposphere. This suggests
strongly that there are two regimes of ozone behavior: that above the minimum
which is dominated by the stratosphere; and that below the minimum which seems
to be dominated by activity at the surface. Though existing vertical profiles
through the boundary layer suggest that mixing may be important in the morning
immediately after sunrise, they show that mixing is not the predominant
mechanism producing diurnal variations of ozone at the surface.
Comparisons of rural and urban data show that on the average the rural
ozone concentrations are larger than the urban concentrations. This factor,
plus the fact that volumetric dilution would produce the opposite effect,
indicates that direct transport of ozone from urban sources is not responsible
for the widespread high ozone episodes found in the rural boundary layer.
By process of elimination based on existing evidence, this leaves photo-
chemical generation as the remaining mechanism that could be responsible for
54
-------
the high ozone in the rural boundary layer. It is suggested that the urban
mix of very reactive and less reactive ozone precursors is transported into
the rural boundary layer. The very reactive precursors produce ozone immedi-
ately. This accounts for the downwind high ozone in urban plumes. The less
reactive precursors remain in the air parcel. The air parcel continues its
motion collecting very reactive and less reactive anthropogenic and natural
ozone precursors in its traverse through the boundary layer. Daily, it loses
the very reactive precursors, but increases the concentration of the less
reactive precursors. Finally, with sufficient sunlight, less reactive pre-
cursors begin to produce ozone. This, combined with the production from more
reactive precursors collected daily, allows the air parcel to contain large
concentrations of ozone.
Sufficient sunlight is available in the relatively cloud-free high
pressure systems. Evidently, it takes somewhere between 2 and 6 days for the
less reactive precursors to become active according to theoretical computa-
tions of air parcel residence times in high pressure systems. This is con-
sistent with chemical evidence. At the time high ozone occurs in the air
parcels, identification of specific sources that have contributed to the ozone
precursor concentration is not possible. The transport of precursors in this
case is on the order of 1000 km or greater.
The basic conclusion of this analysis is that the most important mech-
anism for producing high ozone in the rural boundary layer on any scale is the
transport of ozone precursors from anthropogenic and natural sources combined
with photochemical generation. This factor is most important on the synoptic
scale in the eastern portions of the United States and on the mesoscale
immediately downwind of any urban area.
RECOMMENDATIONS FOR FURTHER RESEARCH
There are two major areas where additional data are required to remove
uncertainties in the model. First, additional measurements are needed to
understand the role of mixing in the boundary layer on the diurnal variation
55
-------
of ozone, particularly under high ozone conditions. This would require that
vertical profiles of ozone be obtained diurnally through the boundary layer
with both high and low ozone concentrations at the surface. This study
should be conducted in a number of locations in the continental United States.
Each measurement series should extend over a 4- to 6-week period. Measure-
ments should include vertical profiles of wind and temperature through the
boundary layer, which are useful in computations of mixing.
The second study needed involves the transport of ozone from the strato-
sphere to the surface and the relationship of that transport to high ozone
episodes. In this case, daily vertical profiles of ozone through the tropo-
sphere would be the data base. Along with these data, basic meteorological
information should be obtained through the troposphere. The profiles should
be obtained during the time that the diurnal maximum at the surface is reached.
This study should be undertaken in a 3- to 4-month period, encompassing mid-
spring to midsummer. During that period, special diurnal vertical profiles
through the troposphere should also be obtained for very specific situations
(e.g., during periods of tropopause folding).
56
-------
REFERENCES
1. Dimitriades, B., and A.P. Altshuller. International Conference on Oxi-
dant Problems: Analysis of the Evidence/Viewpoints Presented. Part I:
Definition of Key Issues. JAPCA, 27(4):299-307, 1977.
2. Research Triangle Institute. Investigation of Rural Oxidant Levels as
Related to Urban Hydrocarbon Control Strategies. EPA-450/3-75-036,
Environmental Protection Agency, Research Triangle Park, N.C., 1975.
359 pp.
3. Cleveland, W.S., B. Kleiner, J.E. McRae, and R.E. Pasceri. The Analysis
of Ground-Level Ozone Data from New Jersey, New York, Connecticut, and
Massachusetts: Data Quality Assessment and Temporal and Geographical
Properties. International Conference on Photochemical Oxidant Pollution
and Its Control, Proceedings. 1:185-196. EPA-600/3-77-001a, Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
4. Research Triangle Institute. Ambient Monitoring Aloft of Ozone and Pre-
cursors in the Vicinity and Downwind of a Major City. Interim Report to
U.S. Environmental Protection Agency. Research Triangle Institute Proj-
ect No. 43U-1272. Research Triangle Institute, Research Triangle Park,
1976.
5. Bufalini, J.J., T.A. Walter, and M.M. Bufalini. Ozone Formation Poten-
tial of Organic Compounds. Environmental Science Technology, 10:908-
912, 1976.
57
-------
6. Vukovich, T.M., W.D. Bach, B.W. Crissman, and W.J. King. On the Rela-
tionship between High Ozone in the Rural Boundary Layer and High Pressure
Systems. (To be published in Environ. Sci. Technol.), 1977.
7. Cox, R.A. , A.E.J. Eggleton, R.G. Derwent, J.E. Lovelock, and D.H. Pack.
Long-Range Transport of Photochemical Ozone in North-Western Europe.
Nature, 255(5504):118-121, 1975.
8. Pack, D.H., J.E. Lovelock, G. Cotton, and C. Curthoys. Halocarbon Behav-
ior from a Long Time Series. Atmos. Environ., 11(4):329-344, 1977.
9. Decker, C.E., W.D. Bach, L.A. Ripperton, and J.J.B. Worth. Research
Triangle Institute Interim Report on Project No. 434-1272 (EPA Contract
No. 68-02-2931). July 1976.
10. Karl, T.R., and G.A. DeMarrais. Meterological Conditions Conducive to
High Levels of Ozone. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:75-88. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
11. Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban Areas
of New York State. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:89-96. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
12. Bach, W.D., Jr., J.E. Sickles, III, R. Denyszyn, and W.C. Eaton. Chemical
and Meteorological Analysis of the Mesoscale Variability of Ozone Concen-
trations over a Six-Day Period. International Conference on Photochemi-
cal Oxidant Pollution and Its Control, Proceedings. 1:197-210. EPA-
600/3-77-OOla, Environmental Protection Agency, Research Triangle Park,
N.C., 1977.
58
-------
13. Husar, R.B., D.E. Patterson, C.C. Paley, and N.V. Gillani. Ozone in Hazy
Air Masses, international Conference on Photochemical Oxidant Pollution
and Its Control, Proceedings. 1:275-282. EPA-600/3-77-001a, Environmen-
tal Protection Agency, Research Triangle Park, N.C., 1977.
14. Wolff, G.T., P.J. Lioy, G.D. Wight, R.E. Meyers, and R.T. Cederwall. An
Investigation of Long-Range Transport of Ozone across the Midwestern and
Eastern United States. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:307-317. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
15. Ripperton, L.A., J.J.B. Worth, P.M. Vukovich, and C.E. Decker. Research
Triangle Institute Studies of High Ozone Concentrations in Non-Urban
Areas. International Conference on Photochemical Oxidant Pollution and
Its Control, Proceedings. 1:413-424. EPA-600/3-77-00la, Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
16. Decker, C.E., L.A. Ripperton, J.J.B. Worth, P.M. Vukovich, W.D. Bach,
J.B. Tommerdahl, F. Smith, and D.E. Wagoner. Formation and Transport of
Oxidants along Gulf Coast in Northern U.S. EPA-450/3-76-033, Environmen-
tal Protection Agency, Research Triangle Park, N.C., 1976.
17. Dobbins, R.A., J.L. Nolan, J.P. Okolowicz, and A.J. Gilbert. Meteoro-
logical Factors Controlling Photochemical Pollutants in Southeastern New
England. International Conference on Photochemical Oxidant Pollution and
Its Control, Proceedings. 1:109-118. EPA-600/3-77-001a, Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
18. Kauper, E.K., and B.L. Niemann. Transport of Ozone by Upper-Level Land
Breeze An Example of a City's Polluted Wake Upwind from Its Center.
International Conference en Photochemical Oxidant Pollution and Its
Control, Proceedings. 1:227-236. EPA-600/3-77-001a, Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
59
-------
19. White, W.H., D.L. Blumenthal, J.A. Anderson, R.B. Husar, and W.E. Wilson,
Jr. Ozone Formation in the St. Louis Urban Plume. International Confer-
ence on Photochemical Oxidant Pollution and Its Control, Proceedings.
1:237-247. EPA-600/3-77-001a, Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
20. Hyde, R., and G.S. Hawke. The Transport of Photochemical Smog across the
Sydney Basin. International Conference on Photochemical Oxidant Pollu-
tion and Its Control, Proceedings. 1:285-297. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
21. Grennfelt, P. Ozone Episodes on the Swedish West Coast. International
Conference on Photochemical Oxidant Pollution and Its Control, Proceed-
ings. 1:329-337. EPA-600/3-77-001a, Environmental Protection Agency,
Research Triangle Park, N.C., 1977.
22. Cleveland, W.S., B. Kleiner, J.E. McRae, and J.L. Warner. Photochemical
Air Pollution: Transport from the New York City Area into Connecticut
and Massachusetts. Science, 191(4223):179-181, 1976.
23. Environment Agency. A Mesoscale Diffusion Experiment for the Photochemi-
cal Air Pollution. Presented at the U.S.-Japan Conference on Air Pollu-
tion Related Meteorology, Tokyo, Japan, Nov. 17-21, 1975.
24. Uchida, E. Sea-Breeze and Air Pollution. Presented at the U.S.-Japan
Conference on Air Pollution Related Meteorology, Tokyo, Japan, Nov. 17-
21, 1975.
25. Takeuchi, K. Recent Studies on Air Pollution Diffusion Model and Rele-
vant Meteorological Observation in Japan. Presented at the U.S.-Japan
Conf. on Air Pollution Related Meteorology, Tokyo, Japa,n, Nov. 17-21,
1975.
60
-------
26. Kawamura, T. Local Wind Systems and Photochemical Oxidants in Kanto
Plain. Presented at the U.S.-Japan Conf. on Air Pollution Related Mete-
orology, Tokyo, Japan, Nov. 17-21, 1975.
27. Sasaki, R. Current Status of Air Pollution Meteorological Forecasting in
Japan. Presented at the U.S.-Japan Conf. on Air Pollution Related
Meteorology, Tokyo, Japan, Nov. 17-21, 1975.
28. Siple, G.W., C.K. Fitzsimmons, K.F. Zeller, and R.B. Evans. Long Range
Airborne Measurements of Ozone off the Coast of the Northeastern United
States. International Conference on Photochemical Oxidant Pollution and
Its Control, Proceedings. 1:249-258. EPA-600/3-77-001a, Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
29. Junge, C.E. The Cycle of Atmospheric Gases Natural and Man-made.
Quart. J. Royal Met. Soc., 98(418):711-729, 1972.
30. Holzworth, G.C. Meteorological Episodes of Slowest Dilution in Contig-
uous United States. EPA-650/4-74-002, Environmental Protection Agency,
Research Triangle Park, N.C., 1974.
31. Ludwig, F.L., W.B. Johnson, R.E. Ruff, and H.B. Singh. Important Factors
Affecting Rural Ozone Concentration. International Conference on Photo-
chemical Oxidant Pollution and Its Control, Proceedings. 1:425-438.
EPA-600/3-77-001a, Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
32. Sickles, J.E., II, L.A. Ripperton, and W.C. Eaton. Oxidant and Precursor
Transport Simulation Studies in the Research Triangle Institute Smog
Chambers. International Conference on Photochemical Oxidant Pollution
and Its Control, Proceedings. 1:319-328. EPA-600/3-77-001a, Environmen-
tal Protection Agency, Research Triangle Park, N.C., 1977.
61
-------
33. Junge, C.E. Air Chemistry and Radioactivity. Academic Press, New York,
1963. 382 pp.
34. Bell, G.B. Meteorological Conditions during Oxidant Episodes in Coastal
San Diego County in October and November, 1959. California State Depart-
ment of Public Health, Berkeley, California, 1960.
35. Price, J.H. A Study of Factors Associated with High Urban Ozone Concen-
trations in Texas. Ozone/Oxidants Interactions with the Total Environ-
ment. APCA Specialty Conference (Southwest Section), Proceedings. Air
Pollution Control Association, Pittsburgh, Pennsylvania, 1976.
pp. 282-292.
36. White, W.H., J.A. Anderson, D.L. Blumenthal. R.B. Husar, N.V. Gellani,
J.D. Husar, and W.E. Wilson, Jr. Formation and Transport of Secondary
Air Pollutants: Ozone and Aerosols in the St. Louis Urban Plume. Science
194(4261):187-189, 1976.
37. Westberg, H. Studies of Oxidant Transport beyond Urban Areas. Final
Report on EPA Contract No. 68-02-2239. EPA-600/3-77-055. Washington
State University, Pullman, Washington, 1977.
38. Hester, N.E., R.B. Evans, F.G. Johnson, and E.L. Martinez. Airborne
Measurements of Primary and Secondary Pollutant Concentrations in the St.
Louis Urban Plume. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:259-274. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C,, 1977.
39. Grimsrud, E., and R.A. Rasmussen. Photochemical Ozone Production from
Captured Rural Air Masses of Ohio and Idaho. Final Report on EPA Grant
No. 800670. Washington State University, Pullman, Washington, 1974.
62
-------
40. Cleveland, W.S., R. Guarino, B. Kleiner, J.E. McRae, and J. L, Warner.
The Analysis of the Ozone Problem in the Northeast United States. Ozone/
Oxidants Interactions with the Total Environment. APCA Specialty Con-
ference (Southwest Section), Proceedings. Air Pollution Control Asso-
ciation, Pittsburgh, Pennsylvania, 1976. pp. 109-120.
41. Spicer, C.W., J.L. Gamma, and P.R. Sticksel. Ozone: Transport Consid-
erations and Predictive Models for Southern New England. Final Report on
EPA Contract No. 68-02-2241. Battelie-Columbus, Columbus, Ohio, 1977.
42. Danielsen, E.J. and V.A. Mohnen. Ozone Measurements and Meteorological
Analysis of Tropopause Folding. Presented at the International Symposium
on Ozone. Dresden, Germany, August 1976.
43. Mohnen, V.A. Ozone Measurements in Rural Areas. The Non-Urban Tropo-
sheric Composition, Proceedings. Hollywood, Florida, 1976.
44. Coffey, P.E., and W.N. Stasiuk. Evidence of Atmospheric Transport of
Ozone into Urban Areas. Environmental Science Technology, 9(1):59-62,
1975.
45. Chatfield, R., and H. Harrison. Ozone in the Remote Troposphere: Higher
Levels? Ozone/Oxidants Interactions with the Total Environment. APCA
Specialty Conference (Southwest Section), Proceedings. Air Pollution
Control Association, Pittsburgh, Pennsylvania, 1976. pp. 77-83.
46. Chatfield, R. , and R.A. Rasmussen. An Assessment of the Continental
Lower Troposphere Ozone Budget. International Conference on Photochemi-
cal Oxidant Pollution and Its Control, Proceedings. 1:121-136. EPA-
600/3-77-OOla, Environmental Protection Agency, Research Triangle Park,
N.C., 1977.
63
-------
47. Lonneman, W.A. Ozone and Hydrocarbon Measurements in Recent Oxidant
Iransport Studies. International Conference on Photochemical Oxidant
Pollution and Its Control. 1:211-223, EPA-600/3-77-001a, Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
48. Dodge, M.C., Combined Use of Modeling Techniques and Smog Chamber Data to
Derive Ozone-Precursor Relationships. International Conference on
Photochemical Oxidant Pollution and Its Control, Proceedings. 2:881-889.
EPA-600/3-77-001b, Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
49. Cleveland, W.S., B. Kleiner, J.E. McCrae, and R.E. Pasceri. The Analysis
of Ground-Level Ozone Data from New Jersey, New York, Connecticut and
Massachusetts: Data Quality Assessment and Temporal and Geographical
Properties. International Conference on Photochemical Oxidant Pollution
and Its Control, Proceedings. 1:185-196. EPA-600/3-77-001a, Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
50. Jerskey, T.N., T.B. Smith, S.L. Marsh, and W.H. White. The Effect of
Ozone Layers Aloft on Surface Concentrations. International Conference
on Photochemical Oxidant Pollution and Its Control, Proceedings. 1:145-
153. EPA-600/3-77-OOla, Environmental Protection Agency, Research Tri-
angle Park, N.C., 1977.
51. white, W.H., D.L. Blumenthal, J.A. Anderson, R.B. Husar, and W.E. Wilson,
Jr. Ozone Formation in the St. Louis Plume. International Conference on
Photochemical Oxidant Pollution and Its Control, Proceedings. 1:237-248.
EPA-600/3-77-001a, Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
52. Huffman, G.D., G.W. Haering, R.C. Bourke, P.P. Cooke, and M.P. Sillars.
Ozone Observations in and Around a Midwestern Metropolitan Area. Inter-
national Conference on Photochemical Oxidant Pollution and Its Control,
Proceedings. 1:341-352. EPA-600/3-77-001, Environmental Protection
Agency, Research Triangle Park, N.C., 1977.
64
-------
53. Reiter, E.R. StratosphericTropospheric Exchange Processes. Rev. of
Geophys. and Space Physics, 13(4):459-474, 1975.
54. Shenfeld, L. Report on Oxidants and Their Precursors in Canada. Inter-
national Conference on Photochemical Oxidant Pollution and Its Control,
Proceedings. 2:917-926. EPA-600/3-77-001b, Environmental Protection
Agency, Research Triangle Park, N.C., 1977.
55. Sandhu, H.S. Oxidant Levels in Alberta Airsheds. International Con-
ference on Photochemical Oxidant Pollution and Its Control, Proceedings.
1:299-305. EPA-600/3-77-001a, Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
65
-------
Alt-rans. ~ ' , and -.J. Komii-.r.
Res., *j 30, :C-"5S9~53M, 1976,
^iivLI, V " Mecbi .ism Accounting ±cr tne Production of Ozone in Rural
Polluted Atmospheres, Inter aationai ~onferfcr.ce on fnutocnemicaj. Oxiaa.n -
Pollution and Its Control, Proceeding^. 1:^^,9-450. EPA-600/3-"?7-001a,
ent.!*.!. Protection Agenry, Research Triangle Park, N.C, . 1977,
breeding, k.J., h,,rr. Klonis. j.P. L.C-^MS. Jr., J.B. Pane, D.C, Sheesley ".R.
rntjlp-rt, aiiC o.F. --!-,ars, ;-'^as,iremer.t-& of An~ospheric P'-llutancs in tne r -. ,
ijOu -f A.-, to. - err- ,<-; .-..' "iron. , : f '" "~ '< - " -"- . "C-
Cruczen, F.J. Photochemical Reactions Initiated by and Influencing Ozone in
-re ''Jrn-.ollui.cd '"roposphere ^ir, '-''elius, XXVI ^ I-2, 47--7, IS/4,
I)fc..Ktr,, C.ii.., W D. Sach, anu ?_.£.. Kipperton. Ambient Monitoring Aloft of
O.rone and Precursors Near and Downwind of St. Louis. EPA-450/3-77-009^
Environmental Protection Agency, Research Triangle Park, N.C., 1977. 69 pp.
67
-------
Drivas, P.J. and F.H. Shair. A Tracer Study of Pollutant Transport and Dis-
persion in the Los Angeles Area. Atmos. Environ., 8(11):1155-1163, 1974.
Environmental Protection Agency. Control of Photochemical Oxidants Techni-
cal Basis and Implications of Recent Findings. EPA-450/2-75-005, Environmen-
tal Protection Agency, Research Triangle Park, N.C., 1975. 42 pp.
Falconer, P.O., and J.D. Holdeman. Measurements of Atmospheric Ozone Made
from a GASP-Equipped 747 Airliner: Mid-March 1975. Geophys. Res. Lett.,
3(2):101-104, 1976.
Fankhauser, R.K. Ozone Levels in the Vicinity of 33 Cities. JAPCA, 26(8):
Ill-Ill, 1976.
Farrow, L.A., T.E. Graedel, and T.A. Weber. Urban Kinetic Chemistry under
Altered Source Conditions. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:137-144. EPA-600/3-77-001a,
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
Khrgian, A.Kh. The Physics of Atmospheric Ozone. (Originally published by
Gidrometeoizdat, Leningrad, 1973. 291 pp.) John Wiley and Sons, New York,
1975. 268 pp.
Leighton, P.A. Photochemistry of Air Pollution. Academic Press, New York,
1961. 300 pp.
Mohnen, V.A., A. Hogan, E. Danielsen, and P. Coffey. Ozone Measurement and
Meteorological Analysis of Tropopause Folding. International Conference on
Photochemical Oxidant Pollution and Its Control, Proceedings. 1:97-108. EPA-
600/ J-7 7-GOIa, Euvirouiaental Ti^tection Agency, Research TH arg!0 O=,T-V. N-<"".,
1977.
68
-------
Rubino, R.A., L. Bruckman, and J. Magyar. Ozone Transport. JAPCA, 26(10):
972-975, 1976.
Westberg, H., K.J. Allwine, and D. Elias. Vertical Ozone Distribution above
Several Urban and Adjacent Rural Areas across the United States. Ozone/
Oxidants Interactions with the Total Environment. APCA Specialty Conference
(Southwest Section), Proceedings. Air Pollution Control Assoc., Pittsburgh,
Pennsylvania, 1976. pp. 84-95.
69
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
; REPORT NO
EPA-600/3-77-117
4 TITLE AND SUBTITLE
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part V. The Issue of Oxidant Transport
3. RECIPIENT'S ACCESSIOIVNO.
5. REPORT DATE
NOVEMBER 1977
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
1. Donald H. Pack
2. Elmer Robinson
Fred Vukovich
8. PERFORMING ORGANIZATION REPORT NO.
9, PERFORMING ORGANIZATION NAME AND ADDRESS
1. Consulting Meteorologist, McLean, VA
2. Washington State Univ., Pullman, WA
3. Research Triangle Institute, RTP, NC
10. PROGRAM ELEMENT NO.
1AA603 AJ-13 (FY-76)
11. CONTRACT/GRANT NO.
1. DA-7-1935A
2. DA-7-2085A
i, DA-7-217QH
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
Partially funded by the Office of Air Quality Planning and Standards.
16. ABSTRACT
In recognition of the important and somewhat controversial nature of the
oxidant control problem, the U.S. Environmental Protection Agency (EPA) organized
and conducted a 5-day International Conference in September 1976. The more than one
hundred presentations and discussions at the Conference revealed the existence of
several issues and prompted the EPA to sponsor a followup review/analysis effort.
The followup effort was designed to review carefully and impartially, to analyze
relevant evidence and viewpoints reported at the International Conference (and
elsewhere), and to attempt to resolve some of the oxidant-related scientific issues.
The review/analysis was conducted by experts (who did not work for the EPA or for
industry) of widely recognized competence and experience in the area of photochemi-
cal pollution occurrence and control.
Part V includes discussions on the issue of oxidant transport written
by Donald H. Pack, Consulting Meteorologist, McLean, Virginia; Elmer Robinson,
Washington State University, Pullman, Washington; and Fred M. Vukovich, Research
Triangle Institute, Research Triangle Park, North Carolina. The authors deal with
the phenomena of urban plume formation and transport, measurement and tracking, and
oxidants and precursor ranges, and recommend future studies.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
* Air pollution
* Ozone
Transport properties
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
13B
07B
13. -'iTR.BUTION STATEMENT
RELEASE TO PUBLIC
I
19. SECUR'TY CLASS (This Report!
UNCLASSIFIED
21. NO. OF PAGES
78
20. SECURITY CLASS (THispage)
UNCLASSIFIED
22. PRtCE
EPA Form 2220-1 (9-73)
70
------- An error occurred while trying to OCR this image.
------- |