EPA-600/3-77-120
November 1977
Ecological Research Series
iTinun nnurrnrunr rm n
/ (
... '• 7::;^ii^llp|^pM^pr^:|
-
''* ''*'"** -is -* .'" 'f L;i>'« ''* *"-
* ' '
ttvf\:,^,-: *- »t ; - -,
%S ^ t; :>*T *' VM
aS».»&.-^4A tr *4»*%.. v!*,**' \ -, 1 „->!.. iiftfct **':-•• 4 »s "•• > * , * ,
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are
1 Environmental Health Effects Research
2. Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5 Socioeconorruc Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8 "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials Problems are assessed for their long- and short-term influ-
ences Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/3-77-120
December 1977
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976 -
ANALYSIS OF THE EVIDENCE AND VIEWPOINTS
Part VIII. The Issue of Optimum Oxidant Control Strategy
W. Bonta and J. Paisie
Department of Health and
Mental Hygiene
State of Maryland
Baltimore, Maryland
Contract No. DA-7-1934A
P. Koziar and B. Becker
Department of Natural Resources
State of Wisconsin
Madison, Wisconsin '
Contract No. DA-7-2175A
L. Jager
Department of Natural Resources
State of Michigan
Lansing, Michigan
Contract No. DA-7-2044A
G. Wolff
Interstate Sanitation Commission
(New York, New Jersey, and
Connecticut)
New York, New York
Contract No. DA-7-2005H
F. Spuhler and K. Waid
Texas Air Control Board
State of Texas
Austin, Texas
. 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
reproduced in the form submitted by the authors.
ii
-------
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 pol-
lution occurrence and control.
Officials representing the states of Maryland, Wisconsin, Texas, Michigan,
and New York reviewed papers presented at the 1976 International Conference
on Photochemical Oxidant Pollution and Its Control concerning the issue of
optimum oxidant control strategy and presented their views on the current
status and resolution of the issue and on the needs for additional research.
iii
-------
CONTENTS
ABSTRACT iii
FIGURES vi
TABLES vi
INTRODUCTION 1
B. Dimitriades and A. P. Altshuller
THE ISSUE OF OPTIMUM OXIDANT CONTROL STRATEGY 3
B. Dimitriades and A. P. Altshuller
REVIEW AND ANALYSIS 11
W. K. Bonta and J. W. Paisie
Introduction 11
Qualitative Assessment of Current Control Strategy 13
Source-Receptor Relationship for Oxidant Control Strategies .23
Conclusion and Recommendations 27
REVIEW AND ANALYSIS 29
P. Koziar and B. Becker
Summary 29
Directional Impacts of Control - Qualitative Evaluation. . . 30
Source-Receptor Relationships 34
Quantitative Evaluation 48
REVIEW AND ANALYSIS 61
F. Spuhler and K. Waid
REVIEW AND ANALYSIS 67
L. E. Jager
REVIEW AND ANALYSIS 73
G. T. Wolff
Introduction 73
The Case for Anthropogenic Precursor Controls 74
The Case for Nonmethane Hydrocarbon Controls 80
The Case for NO Controls 81
x
Development of a Cause and Effect Oxidant Reduction Strategy 89
Summary and Conclusions 92
REFERENCES 95
-------
FIGURES
REVIEW AND ANALYSIS - P. Koziar and B. Becker
Number Page
1 Maximum hourly ozone concentrations recorded at monitoring
sites in Southeastern Wisconsin on August 20, 1976 52
REVIEW AND ANALYSIS - F. Spuhler and K. Waid
Number
1 Schematic relationships among oxidant and precursors under
low, moderate, and high NMHC/NO ratios (52) 62
REVIEW AND ANALYSIS - G. Wolff
Number Page
1 Hydrocarbon emissions 75
2 Surface ozone distribution 76
3 Surface ozone distributions 77
4 Surface ozone distributions 78
5 Surface ozone distributions 79
6 Greenwich, Conn. 1976, April-September ozone data Q-Q plot of
maximum hourly values, weekends vs. weekdays 86
7 Trenton, New Jersey 87
TABLES
REVIEW AND ANALYSIS - G. Wolff
Number Page
1 Ambient NMHC, NO , NMHC/NO Levels at Selected Monitoring
sites ? ? 83
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.
vii
-------
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 optimum oxidant control strategy. 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 OP OPTIMUM OXIDANT CONTROL STRATEGY
Basil Dimitriades and A. Paul Altshuller
As discussed in the introduction paragraphs preceding the first six issue
sections, there are several questions or problems that must be resolved if
oxidant control measures are to be developed and applied. Some of these
questions, by virtue of their importance and basic nature, and the consid-
erable contradictory attention paid to them, have attained an issue status and
have been treated individually in the preceding sections. Resolution of those
issues will definitely and considerably advance the understanding of the
oxidant problem, but will still leave the primary issue of optimum oxidant
control strategy somewhat open. There are still several questions more
directly and specifically addressed to the subject of oxidant control strategy
that need to be given definitive answers. These questions are to be dealt
with in this section.
Departure points in this discussion/analysis will be two facts: The
existence of an oxidant control strategy since 1971, and the generation, since
1971, of considerable new scientific evidence pertaining to the oxidant con-
trol problem. The immediate and obvious question that arises from these facts
is whether the new evidence supports or invalidates, partly or wholly, the
first, existing oxidant control strategy. If the strategy is shown to be
invalidated in some respects, then the next obvious question is what strategy
revisions are dictated or can be justified by the new evidence. To explore
these questions or possibilities, it is necessary that the new scientific
evidence, or, more directly, the changes in understanding brought about by the
new evidence be specified.
The most important change in understanding that has been brought about by
the new evidence concerns the source-receptor relationship, that is, the
-------
geographical or spatial relationship between areas in which emissions are
discharged and the areas where the air quality is impacted by these emissions.
Much of the discussion needed here about this relationship has already been
presented as part of the discussion on the Oxidant Transport issue. The
control strategy implications, however, of the source-receptor relationships,
as now understood, need to be further expanded.
The first implication is that long-range pollutant transport introduces a
link between the urban oxidant jDroblem and the rural oxidant problem. This
means that in many areas, urban emissions and oxidant significantly affect
rural air quality and, conversely, oxidant-carrying rural air upwind from a
city significantly affects the city's air quality. Thus, from a control
standpoint, the urban and rural problems are not entirely disassociated, and
therefore, respective optimum control strategies should not be pursued en-
tirely independent of each other.
A second implication is the one arising from the vertical mixing patterns
observed and associated with radiation and subsidence inversion phenomena.
Such mixing patterns suggest that local emissions may have significant carry-
over effects upon next day's local air quality. This and the previous im-
plication depict a new picture of the photochemical processes responsible for
a city's oxidant problem. According to this picture, the local oxidant pro-
blem is the composite of contributions originating from:
(a) local, fresh emissions,
(b) local but previous day's emissions,
(c) extraneous emissions (probably from previous days), and
(d) natural sources.
The third implication relates to the relative roles of the HC and NO
X
emissions in the oxidant problem. The existing oxidant control strategy
formally recognizes only a HC role; no oxidant-related controls are imposed
upon NO emissions. Such roles of the precursors, however, are now thought to
be incorrect quantitatively and,, perhaps, qualitatively also. Thus, the
oxidant-to-HC dependence is not independent of the NO factor. Also, and more
X
-------
importantly, the effects of local HC and NO emission controls on oxidant are
X
expected to vary depending on whether the oxidant results from local fresh
emissions or from previous day's local emissions or from extraneous emissions.
In the light of these implications of the recent scientific findings, the
specific questions that need to be answered here are as follows:
1. Is the qualitative basis of the existing oxidant control strategy
still sound? That is, is hydrocarbon emission control an optimum
approach to urban oxidant reduction?
Because of the link between the urban and the rural oxidant problems, the
answer must be based on considerations of both types of problem. Considera-
tion must also be given to the situation in which local emissions have multi-
day carryover effects upon local oxidant. Finally, consideration should be
given to the relative importances of the anthropogenic and the natural sources.
In essence then, the first question asked here can be reworded as follows:
Considering the four origins of urban oxidant [cases (a) - (d)
described in preceding paragraphs], is control of local (urban) HC
emissions expected to be an effective means to local oxidant reduction?
Relevant evidence consisting of both laboratory and field data has been
reported at the International Conference and elsewhere, and is conflicting.
Thus some field studies showed HC control to have a strong beneficial effect,
others did not show a detectable effect, and others showed effects varying
with attendant NO emission change. Smog chamber studies showed that for the
portion of oxidant formed from the day's local emissions, control of HC is
beneficial except for atmospheres with extremely high — ordinarily not ob-
served — hydrocarbon-to-NO ratios; for such atmospheres HC control, unless
drastic, will have very little effect. For the portion of oxidant formed
through multiday irradiation of emissions, the evidence is considered scant
and inconclusive by some investigators but conclusive by others. Those who
feel that the evidence is conclusive claim that HC control will have small
effects upon multiday, irradiated air masses, relative to the effect upon
single-day, irradiated air. Control of HC emissions upwind, again, has little
5
-------
effect according to some investigators, undetermined effect by others. While
it is certain that the effectiveness of the HC control approach is different
for different localities, the judgment called for here is whether this ap-
proach should be retained or replaced by another approach .
2. Insofar as the urban oxidant problem is concerned, is NO emission
X
control imperative? desirable? tolerable? undesirable? intoler-
able?
Again, for the questions to be answered properly, consideration must be
given to the various sources of urban oxidant, namely, pollutant transport,
local/fresh emissions, local/aged emissions, and natural sources. Evidence
from field studies is conflicting in that it shows higher NO emission rates
to be associated with lower oxidant concentrations in some cases, and no such
association in others. Smog chamber data exist only for the situation in
which the urban oxidant forms from the day's local emissions. For this
situation, NO control has varying effects depending on the hydrocarbon-to-NO
X X
ratio of the reacting emissions. The effect of control of the upwind emis-
sions of NO also may be in dispute. Again, as with the preceding questioi ,
X
the judgment called for here is whether the NO emission factor should con-
X
tinue to be nearly ignored — as is the case with the existing strategy — or
should be considered, and how.
The preceding paragraphs dealt with the qualitative bases of an optimum
oxidant control strategy, that is, with the questions pertaining to the direc-
tional impacts of the HC and NO emission controls. The questions that need
X
to be defined next deal with the quantitative bases of an optimum strategy,
that is, with the quantitative impacts of controls. Such quantitative bases
consist of the following two components:
The quantitative relationship between ambient oxidant concentra-
tions and emission rates, and
The definitive relationship between source area and receptor
area, meaning the definition of the geographical area where the
required control — as calculated from the oxidant-emission
relationship — must be applied to solve the oxidant problem
observed in a given locality.
-------
The quantitative relationship between oxidant and emissions is the
subject of the issue on the replacement of the Appendix-J method, and will
not be treated here. Nevertheless, it might be helpful to mention here that
the smog chamber method proposed as a replacement of the Appendix-J method
provides a cause-effect relationship between oxidant and precursors, and
unlike the Appendix-J method, does not take any specific source-receptor
relationship for granted. Thus the smog chamber method does not prescribe, as
the Appendix-J method does, that control be confined within the urban area
where the oxidant problem was observed.
Assuming that the replacement of the Appendix-J method will be a method
based on a cause-effect relationship between oxidant and emissions, it will be
necessary that "cause" and "effect" be identified, respectively, with a
"source area" and a corresponding "receptor area," the latter being the area
where the air quality is impacted by the source area.
Defining the source-receptor relationship consistent with the cause-
effect nature can be approached, in theory at least, in several different
ways. By one, first approach, "effect" is identified with the maximum oxidant
observed in the "receptor" area, and "ca,use" is identified with the emissions
from all sources — local and upwind — that impact the receptor area. This
approach requires that all source areas impacting the receptor area be iden-
tified, a requirement that is extremely difficult — if at all possible — to
fulfill.
By the second approach, "effect" is identified with the fraction of the
observed maximum oxidant concentration attributable to the local emissions,
and "cause" is identified with the local emissions. This approach has at
least two problems:
• The determination of the oxidant fraction associated with the
local emissions, and
• The requirement that another source-receptor relationship be
defined for the fraction of oxidant associated with the extraneous,
upwind emissions.
-------
The first problem is not an insolvable one; e.g., a rough estimate of the
oxidant fractions associated with the local and the extraneous emissions could
be obtained from oxidant measurements upwind and downwind from the receptor
area. The second problem, however, is extremely difficult — if at all pos-
sible — to solve.
A third approach could be conceived as a simplified compromise between
the two preceding ones. By this third approach, "effect" is identified with
the maximum oxidant concentration observed in the receptor area, and "cause"
is identified with the local emissions. The assumption is made here that the
local emissions are the sole and whole cause of the oxidant problem. This
assumption - it is well recognized - is not valid in oxidant transport situa-
tions in which the local emissions cause only part of the problem. Neverthe-
less, the assumption is justified on grounds that in the very same (oxidant
transport) situations, the local emissions almost surely contribute to or
cause additional problems to downwind areas. It might appear at first glance
that application of controls calculated by this approach upon the receptor
area as well as upon the upwind areas will result in over-control. In actu-
ality, however, this will not necessarily be the case because the sum total of
the "local" and "transported-in" contributions to oxidant may exceed the 0.08-
ppm standard even though the individual contributions are each less than 0.08
ppm.
Finally, a fourth approach could be conceived as a more-stringent-control
version of the second approach. By the fourth approach, "effect" is iden-
tified with the fraction of the maximum oxidant concentrations attributable to
the local emissions. However, calculating control requirements so as to
reduce a fraction of the oxidant down to 0.08 ppm, obviously, will not solve
the oxidant problem. To solve the problem, control requirements should be
calculated so as to reduce the "local" oxidant fraction below 0.08 ppm, that
is, to a level such that the total oxidant will be, if possible, at or below
0.08 ppm. Thus, by the fourth approach, the oxidant contribution from the
extraneous (upwind) sources is partly or wholly offset by imposing increased
control of local emissions.
-------
The preceding analysis identified four conceivable approaches to formu-
lating an optimum strategy for oxidant control. Obviously, there may be others,
The question to be asked here is:
Which of these four, or any other approaches, is the one to be
recommended?
-------
REVIEW AND ANALYSIS
William K. Bonta and Joseph W. Paisie
INTRODUCTION
We find that answering the question on the "issue of the optimum oxidant
control strategy" is considerably more difficult a chore than first antici-
pated. Our goal in Maryland has been to maintain a reasonable grasp of the
state-of-the-art photochemistry and related modeling so that our oxidant
abatement program is kept progressing on a relatively informed basis. Our
program has for the most part consisted of following the Federal, new car
program and the California stationary source control regulations. In con-
templating the question, we ask ourselves if there were not abatement programs
now in existence, where would we start? The answer becomes more difficult as
our insight broadens.
Since the issue is whether or not to change that National strategy
adopted previously, we first look toward the basis for existing policy. To do
this from a state perspective is very difficult and we hope the following is
not too inaccurate.
Although we were not directly involved in the original National policy
development, it appears from references to the advisory panel and Congres-
sional proceedings that some of these questions were discussed at that time.
At any rate, the policy adopted in 1967, merely 10 years ago, was to control
hydrocarbons only. Again, though we were in a peripheral position in 1970, it
is our understanding that the control of NO was added to the program by the
X
Congress for the sake of the NO impacts alone and not because of the in-
X
fluence on photochemical oxidant production.
11
-------
We feel fairly confident that the present EPA oxidant control strategy
was based upon three premises contained in the photochemical oxidant criteria
document (2). The first premise was that photochemical oxidants were only an
urban air pollution problem. The second premise was that, in urban atmo-
spheres, there were very large emissions of both organics and oxides of
nitrogen. The available evidence from smog chambers indicated that the
optimum photochemical oxidant control approach in such a situation would be to
control organic emissions. The final premise was the presumption that the
local photochemical oxidant levels were the result of local precursor emis-
sions. The role of rural precursors was thought to have little impact on the
urban situation; the levels of oxidant measured in rural areas at that time
were less than 0.07 ppm. Any ozone transported to the urban areas was thought
to react with morning NO emissions in the suburban fringes, and thus, the zero
background assumption was thought to be justified.
The mechanistic picture of the photochemical oxidant problem has been
altered as a result of EPA studies (3-7) which discovered rural oxidant levels
2 to 3 times the national ambient standard. These discoveries have changed
this early concept in several ways. First, the literature has suggested that
oxidant and/or precursors can be transported in significant quantities from
urban areas to rural areas and vice versa. The implication is that the closed
system (smog chamber) may not accurately simulate all situations of elevated
oxidant as the early premise assumed. Secondly, the problem of photochemical
oxidants has been demonstrated to be more widespread than originally thought
and definitely not restricted to urban areas. Finally, the roles of hydro-
carbons and oxides of nitrogen may be different than perceived earlier; i.e.,
hydrocarbon may not always be the limiting reactant as was assumed earlier.
This especially applies to rural areas.
The impact of this new information is one of raising questions concerning
the current EPA control strategy. This review will consider the recent infor-
mation concerning the state-of-the-art oxidant control strategy and recommend
changes based on this new information.
12
-------
QUALITATIVE ASSESSMENT OF CURRENT CONTROL STRATEGY
The fundamental chemistry between organics and oxides of nitrogen in a
photochemical atmosphere must be defined in a closed system. This requirement
dictates the use of a smog chamber. The results of these experiments (Dimi-
triades (8), Jeffries et al. (9) and others) has identified this fundamental
qualitative relationship for the lower concentrations typical of urban atmo-
spheres. Quantitatively, the results have not agreed as well. However, it is
unarguable to say that these experimental results have not been contradicted.
The results of these experiments have been extended by Dodge (10) through the
use of chemical kinetic models to conditions which were not experimentally
practicable, i.e., very high and very low ratios of nonmethane hydrocarbons to
oxides of nitrogen, i.e., greater than 12/1 and less than 2/1. Certainly,
there is a degree of risk involved in extrapolating into these areas through
the use of a model; however, the results of this extrapolation seem reasonable
and are usable. Therefore, the results of these experiments determine the
basic role of organics and oxides of nitrogen in the photochemical oxidant
generation process. Qualitatively, these results do not differ from the
earlier results produced in smog chambers (11). Therefore, the fundamental
roles of organics and oxides of nitrogen have not changed.
The change that has prompted a reevaluation of the oxidant control
program is the perception of the meteorological phenomena affecting the
production of photochemical oxidants. The early smog chamber work was an
adequate simulation of a "first day" urban situation, i.e., the situation in
which fresh urban emissions are released into a relatively clean atmosphere.
It is our opinion that this is an overly simplistic picture of the urban
situation. The urban problem is the result of many factors influencing the
production of oxidants. Included in these phenomena are elevated stable
layers of ozone and precursors which may be the result of the urban area it-
self or transport from other areas. Other factors are the effects of natural
phenomena such as the stratospheric intrusion of natural ozone and natural
precursor emissions. The evaluation of the control strategy in urban areas is
complicated by these factors.
13
-------
The paper by Sickles et al. (12) demonstrates that relatively small
amounts of precursor, which are the leftover products of a first day irradia-
tion, can generate significant amounts of ozone. These results support the
diffusion modeling analysis of Dabberdt and Singh (13). These authors modeled
stringent control strategies which reduced both organic emissions and oxides
of nitrogen and were initiated after 2 or 3 days of an episode situation. The
results indicated that a very small reduction in ambient oxidant levels could
be expected as a result of the control strategies. Certainly, the quantita-
tive conclusions drawn by this modeling analysis may be erroneous since the
reliability of the results of photochemical diffusion models is questionable
at this time. However, the qualitative results of this analysis appear to be
correct. These results may be further supported by the work of Cleveland et
al. (14), which demonstrates that Sunday levels of oxidants, with reduced
emissions of both organic and oxides of nitrogen emissions, do not necessarily
result in a corresponding decrease in photochemical oxidant level. It is our
opinion that these results demonstrate that when considering a short-term
situation, the effect of organic and oxides of nitrogen reduction may not
result in a concomitant reduction in oxidant concentrations.
The long-term analysis of this situation indicates somewhat different
results. The use of trend analysis is very helpful in the evaluation of the
long-term effects of the EPA control strategy. The work of Trijonis et al.
(15), in considering the trends in oxidant and precursor trends in an urban
area (Los Angeles), is very helpful. This work indicates that the basinwide
emission trend of organics is down by 18% and the emission of oxides of
nitrogen is up by 36%, resulting in a basinwide reduction in oxidant of 19%.
These results correspond qualitatively with the prediction that would be made
using the smog-chamber diagram and a ratio of nonmethane hydrocarbons to
oxides of nitrogen that would be appropriate to an urban area. Therefore, the
qualitative behavior of the Los Angeles atmosphere is expected. This study
cannot by itself demonstrate the quantitative relationships between the de-
pendent and the two independent variables.
The study by Trijonis et al. is further supported by the study of Martinez
et al. (16). The analysis performed in this study, which was mainly concerned
14
-------
with San Francisco, demonstrates that oxidant reduction in the Bay Area was
correlated with organic emission reduction. However, the correlation between
the NMHC/NO ratio and the oxidant reduction is more significant. Again, this
X
is an expected result considering the existing model for the behavior of a
"typical" urban atmosphere; i.e., reducing levels of organics and raising
levels of oxides of nitrogen will result in lower levels of oxidant. We
hesitate to use the term "typical" here because of the obvious differences
between the California basin situation and those of other U. S. cities.
The final report relating the behavior of photochemical oxidants with EPA
control strategy of organic emission reduction was conducted by Altshuller
(17). This study, which was conducted using the CAMP station data in six
urban areas, demonstrates a reduction in oxidant levels. It was shown that
the trend in emissions in these metropolitan areas is similar to that found in
Los Angeles and San Francisco, i.e., a reduction in oxidant concurrent with a
reduction in nonmethane hydrocarbon levels. The study showed that oxides of
nitrogen levels did not increase in three of the data sets, whether this
observation eliminates our concern over the relative split between the effects
of the precursor changes (that is, that oxidant reduction occurs with hydro-
carbon reduction while keeping NO constant) may be an anomaly generated by
X
large gaps in the NO data.
X
The general body of observational evidence indicates that, qualitatively,
the role of organic emissions in the formation of urban oxidants has not
changed in the past years. This conclusion is based only on the results of
analysis of urban situations. The trend information has indicated that, in
spite of elevated ozone layers, carryover of oxidant and precursor, and
natural sources of precursor, urban oxidant levels have decreased. The major
problem with these analyses is that for most of the urban areas under con-
sideration, it is difficult to separate the effect of the decrease in organic
emissions from an increase in oxides of nitrogen emissions, we conclude that
is not possible to attribute the reduction in urban oxidant levels to reduc-
tion in organic emissions alone.
15
-------
The role of organic precursor emissions in the production of rural oxi-
dants is not as clear as the urban situation. The available information
indicates that a number of processes are available that could conceivably
contribute to levels of rural oxidant. Among the processes are the transport
of mature ozone from urban areas, multiday irradiation of transported urban
precursor, stratospheric intrusion of ozone, irradiation of natural precur-
sors, irradiation of transported urban precursors, and irradiation of both
transported urban precursors and fresh rural emissions. Each of these factors
could contribute to the formation of rural ozone. The available information
does not indicate what quantitative relationships are applicable. In addition
to this, the NMHC/NO ratio in rural areas is considerably higher than the
corresponding ratio in urban areas. With this increase in ratio, the rela-
tionship between organic precursor emission and resulting oxidant may change.
According to the extrapolated smog chamber diagrams, which we believe repre-
sents the chemistry of the rural atmosphere, at high ratios (>30/1) of NMHC/
NO , the level of resultant oxidant does not appear to be significantly re-
X
lated to the level of organics; i.e., reduction of organics will not have an
effect on oxidant levels until great reductions on organic levels are achieved.
The information presented at the International Conference which would present
real world confirmation of the rural generation theorum is somewhat confusing.
The paper presented by Hathorn and Walker (18) attributes the primary
source of rural ozone to stratospheric intrusion of ozone-rich air. The
hypothesis that stratospheric intrusion does occur has been advanced by others
(19) and the phenomenon is sufficiently well documented to accept its occur-
rence at least occasionally. However, the data presented by Hathorn and
Walker does not conclusively demonstrate their case. Tracer information,
which could demonstrate where the ozone originated, was not presented. This
makes interpretation of the data very difficult. Additionally, the lack of
supporting data, both meteorological and chemical, makes the case very weak.
They do not quantify the fraction of ozone due to stratospheric intrusion. We
are of the opinion that attributing stratospheric intrusion of ozone as the
primary rural source is a "leap of faith" that is not supportable using this
data. The hypothesis is interesting, however, and we wonder if a sophis-
ticated statistical analysis would not disprove the theory.
16
-------
Another potential source of rural ozone is transported urban ozone and
precursor. The work of Sickles et al. (12) utilizing smog chamber simulation
of rural atmospheres is useful in demonstrating this phenomenon. This work
demonstrates the effects of dilution and second and third day irradiation of
urban atmospheres on the production of photochemical oxidants. The experi-
mental data demonstrates that the "dilution effect" occurs and can be a major
contributor to rural ozone. However, the half-life calculations indicate that
previous day's ozone can exist for only 20-30 hours. This implies that high
rural oxidant levels lasting for more than 2 days could not occur if it were
only the result of re-irradiation of diluted urban air masses. It follows
that fresh precursor emissions are necessary to the process, if continuing
high levels of rural ozone are expected to occur. The question of which
precursor is necessary to the process is not clear. The critical parameter to
be considered is the NMHC/NO ratio. If this ratio is greater than 75/1,
using the data of Sickles et al. (12), the effect of additional organics is
not important and NO becomes the crucial factor. If the ratio of NMHC/NO is
X X
less than 75/1, the crucial factor is the effect of additional organics. This
information indicates that rural oxidant can be the result of transported
urban precursor and oxidant plus fresh precursor from the rural areas. Which
precursor is necessary would depend on the NMHC/NO ratio. The problem with
this analysis is that no quantification is presented concerning the impact of
urban precursor and oxidant or rural oxidant.
A third body of information concerning the origin of rural oxidant is the
study of Meyer et al. (20). The use of trajectory analysis, as was done in
the study, is useful in determining the relationship between meteorological
and emissions parameters and resulting levels of photochemical oxidants. The
process used is probably not quantitatively useful because of inaccuracies in
input data. The results indicate that meteorological parameters are more
highly correlated with oxidant levels than emission characteristics. This is
consistent with other studies. This study demonstrates that both upwind and
local emissions of organics are positively correlated with observed levels of
photochemical oxidants in rural areas. An interesting result, however, is the
positive correlation of NO emissions of the previous day.
X
17
-------
Based upon all the information, the role of organic emissions in rural
oxidants is dependent upon the NMHC/NO ratio. Smog chamber diagrams indicate
that at sufficiently high NMHC/NO ratios (>30/1) the effect of organic emis-
sion control or oxidant production diminishes. There is a hint that atmo-
spheres of this sort may exist in Meyer's trajectory analysis. The problem is
that at the present time not enough monitoring data is available to determine
where areas with a ratio greater than this occur, if they occur at all. Until
such time as more data is available and the proper analyses can be made, it is
not possible to determine those areas where it is certain that organic emis-
sion control in rural areas is beneficial. In addition to this, the impact of
natural emissions of organics, which may produce significant rural oxidants at
ratios of approximately 15/1 and 25/1, has not been assessed. The entire
rural situation needs greater study and a more complete quantification before
a definitive answer concerning the effect of organic emission control can be
formulated.
Based upon the above discussion, we are of the opinion that the role of
organic precursors in rural atmospheres is not clear. The determination of
the role of organics must be based upon an adequate knowledge of the spatial
distribution of NMHC/NO ratio in rural areas. The information available is
x
insufficient to adequately address this question.
The next important topic in this assessment of the current oxidant con-
trol strategy concerns the role of oxides of nitrogen. The current strategy
does not recognize the role of NO in the production of photochemical oxi-
dants. The recent information presented at the international Conference on
Photochemical Oxidant Pollution and Its Control may require that this judgment
be reconsidered.
The first task in the reconsideration of the role of oxides of nitrogen
in the photochemical oxidant production process involves a discussion of the
urban situation. Again, it is necessary to rely on the work done in somg
chambers (8,9) to determine the qualitative chemical relationship between
oxides of nitrogen and photochemical oxidant production. The work that has
been done using concentrations of organics and oxides of nitrogen that can be
18
-------
considered typical of our urban atmosphere and a correspondingly typical ratio
of NMHC/NO has demonstrated that increasing NO , at constant organic concen-
X X
tration, results in reduced levels of oxidants. The oxide of nitrogen added
is NO, a well known scavenger of ozone. This phenomenon has been noted many
times, especially in the placing of air monitoring stations near heavily
traveled roads.
Ambient studies on the effects of changes in concentration of oxides of
nitrogen on urban oxidants have demonstrated this behavior. The work of
Trijonis et al. (15) in Los Angeles, discussed earlier, demonstrated that
increasing concentrations of NO and decreasing concentrations of organics has
resulted in a reduction in oxidant. This qualitatively agrees with the pre-
dictions of smog chamber. However, the work does not demonstrate the quanti-
tative relationship involved.
The work of Martinez et al. (16) has also demonstrated that urban levels
of oxidant are most sensitive to the NMHC/NO ratio. Specifically, the work
X
has shown that decreasing the NMHC/NO ratio, which can be accomplished by
X
reducing NMHC at constant NO or increasing NO at constant NMHC is related to
X X
decreasing levels of photochemical oxidants. This evidence is a further
indication that the qualitative predictions of the smog chamber are sub-
stantiated in an urban atmosphere.
The only ambient study that would, at face value, tend to contradict
these results is the study of Van Ham and Niebor (21). This study shows that
addition of NO to an air sample results in increasing photochemical oxidant
X,
production. This is a difficult study to interpret since a complete NMHC/NO
ratio is not reported. If the ratio had been reported, it would be poM.siblo
to assess more adequately the results. If the ratio of NMHC/NO were suf-
X
ficiently high, i.e., a system that is NO deficient, the addition of NO
x x
would increase oxidant production, if the sample were organic deficient,
i.e., a low NMHC/NO ratio, the results of this study would be contrary to the
results of other investigators and a more in-depth analysis of the procedures
and the results would be necessary. Without the NMHC/NO ratio, the study
19
-------
must be considered inconclusive. A repetition of the study with a more
complete data set would be in order.
Based upon these data and results, the control of oxides of nitrogen in
urban areas to control urban oxidant is unnecessary and would be counter-
productive. However, before reaching a final judgment on the control of
oxides of nitrogen, one must consider the rural area problem where the role of
NO is not as clear.
x
Ambient monitoring in rural atmospheres (3-7) indicates that, in general,
rural atmospheres are characterized by high NMHC/NO ratios and levels of NO
X X
that are at or near the minimum detectable limit of the instrument, i.e.,
0.005 ppm for an hour. The NO present in rural atmospheres is almost totally
X
NO2, which is an ozone precursor (as opposed to NO, which is an ozone scav-
enger) . Therefore, the effect of additional NO in rural atmospheres would
X
depend upon which form of NO is added to the rural atmosphere and where the
X
effect is to be measured. If a rural source of NO is to be considered, the
X
emission from fuel burning will be mostly NO. Therefore, in the neighborhood
of the source, one could expect a decrease in ozone because of the NO-ozone
scavenging reaction. However, farther away from the source, one could expect
that the NO will be converted to NO?, and the concentration of NO« will be
decreased by dilution. This could result in an increase in oxidant produc-
tion, similarly, one could expect that increased NO emissions in urban areas
X
could be transported to rural areas as NO . The increase in urban NO emis-
^ X
sions could be expected to expand the region of the rural area affected by
urban oxidant and precursors.
This picture of the rural oxidant situation is supported by the work of
Meyer et al. (20). While the trajectory analysis approach used in the study
has many limitations and only 40% of the observed variance is explained by
the correlations performed, the qualitative picture of rural oxidant de-
lineated in the study is similar to that described previously. The major
finding in this study with respect to NO may be the confirmation in situa-
X
tions other than the California basins that NO is probably a scavenger of
ozone and NO may be an ozone generator. The results indicate that the
20
-------
previous day's emissions of oxides of nitrogen were positively correlated
with observed levels of photochemical oxidant. Emissions of NO on the da
the oxidant is observed are negatively correlated with levels of ozone.
Another study concerning the role of oxides of nitrogen in the rural
atmosphere is that of Sickles et al. (12). This study demonstrates the
effect of changes in NMHC/NO ratio on ozone maxima and ozone generation. This
X
study has determined that a maximum occurs in the relationship between the
NMHC/NO ratio and the amount of ozone generated. This maximum occurs at
X
NMHC/NO ratios equal to approximately 75/1. At ratios lower than 75/1,
X
decreasing the ratio decreases ozone. At ratios greater than this, dec-
reasing the ratio will result in increased ozone production. These results
are qualitatively consistent with other smog chamber experiments (8,9) re-
lating the effect of the NMHC/NO ratio on ozone production. However, the
X
critical ratio in these studies ranges from 15/1 to 30/1. The important fact
is that this phenomenon has been demonstrated.
The foregoing qualitative discussion of the current EPA control strategy
has addressed the role of the precursor pollutants in a segmented fashion;
i.e., it has discussed the urban situations separate from the rural situation
and has discussed the role or organic emission control separate from oxides
of nitrogen control. This is not the proper way to analyze photochemical
oxidants. The EPA studies concerning transport have clearly demonstrated that
the urban problem and the rural problem are not inseparable. Therefore, a
discussion of the coordination between urban and rural phenomena is necessary.
To discuss this aspect of the current control strategy, it is necessary to
first outline a few more observations.
The current EPA control strategy with respect to automobile emissions has
been aimed at the two types of automotive emissions, i.e., exhaust and evapora-
tive. The trend in automobile emissions from Black's (22) study shows lower
mass amounts of emissions as well as a change in composition. It also shows
that evaporative emissions are becoming larger than exhaust emissions. The
addition of catalysts has effected a change in composition of exhaust emis-
21
-------
sions with the trend toward a greater percent of paraffinics. Additionally,
the composition of evaporative emissions from automobiles is mostly paraffinic
with n-butane and isopentane being the dominant constituents.
The trend in stationary source emission control is outlined by Walsh
(23). Basically, the past strategy for stationary sources was based on Los
Angeles' Rule 66 controls. This strategy was based upon reactivity and
substitution of slower reacting solvents for rapid reacting solvents. This
strategy resulted in solvent reformulation and substitution. However, it did
not result in a great reduction in the tonnage of emissions from stationary
sources. The future strategy will be based on positive organic emission
control.
The present control program for oxides of nitrogen relies on the control
of automotive emissions. The new-car control program is the only aggressive
program for control of this pollutant presently in force. Stationary source
NO control has not been initiated except in California. With these observa-
X
tions in mind, the next step is to evaluate the impact these existing stra-
tegies will have on ambient levels of oxidant.
Obviously, the present strategy will reduce levels of organic compounds
in urban or source areas. This, coupled with the fact that the automobile
control program, specifically the catalytic converter, will result in emis-
sions of organic compounds requiring longer irradiation to reach maximum
oxidant concentrations, will result in lower levels of ozone. This will be
aided by the fact that oxides of nitrogen emission controls have been delayed.
Consequently, it can be expected that the urban ratio of NMHC/NO will con-
X
tinue to decrease and will result in lower urban oxidant levels. The result
of this action in the urban areas may be an increasing carryover of urban
emissions into rural areas. This will definitely be true for oxides of
nitrogen and may be true for organic emissions. This can be expected to
deliver more precursor to rural areas where further irradiation can generate
oxidant. Indications from the study of Sickles et al. (12) are that urban
oxidant cannot last more than 20-30 hours in a rural area, based upon half-
life measurements both in the chamber and in the ambient atmosphere. There-
22
-------
fore, since generation of oxidant is already occurring in rural areas, the
expanded carryover of urban precursor into rural areas can be expected to
accelerate this process. Additionally, reduction in urban organic precursor
without a concomitant reduction, or at least holding of the line, in emissions
of urban oxides of nitrogen could result in an expansion of the current rural
oxidant problem since most rural atmospheres appear to be NO deficient. Thus
X
it appears that positive control of urban organic emissions is necessary and
must be continued. Control of urban oxides of nitrogen emissions is necessary
for the benefit of rural areas. However, the control of urban oxides of
nitrogen is counterproductive when considering urban oxidant. Therefore, the
best policy for the interim is to maintain urban oxides of nitrogen at their
current level. If abatement of oxides of nitrogen is to be initiated in urban
areas, the control of organic emissions must be coordinated with this program
to ensure that the NMHC/NO ratio continues to decrease.
The precursor control program in rural areas is not as clear. The
atmospheric levels of organics and oxides of nitrogen in rural areas have not
been adequately defined. Therefore, any recommendations for rural areas must
necessarily be speculative. However, it can generally be said that rural
atmospheres would probably tend to be NO poor. The ratios of NMHC/NO can
X X
be expected to be high relative to the urban area. The question is how much
higher. If rural atmospheres have ratios greater than 75/1, as per Sickles et
al. (12), the control of oxides of nitrogen appears to be the best method for
control. If ratios are less than this, organic emission control can be ex-
pected to be effective. Until more data is available, this is pure specula-
tion.
SOURCE-RECEPTOR RELATIONSHIP FOR OXIDANT CONTROL STRATEGIES
The preceding section dealt with the fundamental chemical relationship
between oxidant precursor and oxidant. This section will deal with the con-
sequences of meteorological transport on the oxidant production process and
how it bears on the source-receptor relationship in photochemical oxidant
control strategies.
23
-------
The current EPA control strategy is based upon the following source-
receptor relationship. Precursor concentrations of organic compounds in the
6-9 a.m. time period at a given location are associated with the photochemical
oxidant concentrations measured near that location later in the same day.
This view of the oxidant production process has been seriously questioned as a
result of the EPA transport studies and measured levels of rural oxidants. We
feel that this concept of the oxidant production process is incorrect and must
be modified. A number of alternatives have been suggested. Each of the
alternatives attempts to define the correct source-receptor relationship,
i.e., a cause-effect relation.
The first alternative identifies the effect with the maximum oxidant
observed in the "receptor" area. The cause is associated with all sources
both local and upwind. This approach is the most complicated approach to be
suggested since it requires that a complete tropospheric ozone budget be
developed for every urban area experiencing oxidant levels elevated above the
standard. This type of information is not currently available since many
factors in the tropospheric ozone generation process are still inadequately
quantified, e.g., natural emissions, stratospheric ozone intrusion, and
natural scavenging mechanisms. Therefore, this approach, which may be the
most realistic characterization of the source-receptor relationship, must be
relegated to the distant future since it is simply too complex to be useful at
the present time. As more data becomes available and the understanding of all
the factors affecting tropospheric ozone is improved, this technique should be
reconsidered.
The second alternative approach is to identify the fraction of observed
maximum oxidant concentration attributable to the local emissions assuming
that the local emissions cause this oxidant. Certainly, this method would
allocate control properly, i.e., local sources would only be required to con-
trol in order to reduce the local contribution. The first problem with this
approach would be the quantification of the contribution that loca.1 sources
made to the local problem. This would require the use of a deterministic
model which is not available for many areas in the country. The notable
exceptions would be Los Angeles, San Francisco, Denver, and St. Louis. We
24
-------
believe that, at this time, no other cities in the country have the necessary
instrumentation or the clearly defined airshed in which one could perform this
task. Application of this method to other cities in the United States is not
insurmountable; however, it would require time and more resources than will
likely be available in the near future. The second problem involves the
determination of the source of the extraneous oxidant entering an urban area.
This is the most difficult problem to solve. The transport studies of EPA
show high levels of ozone associated with the back edges of high pressure
systems. It would seem that these air parcels can have residence times as
long as 4 days. Therefore, it seems highly unlikely that one can specifically
identify the source of emissions that contributes to those elevated oxidant
levels. This approach, therefore, is also too complicated to be used at this
time. The combined need of a deterministic, quantitative oxidant model and a
complete knowledge of the meteorological transport patterns occurring over
synoptic scale areas of the United States makes this approach unworkable.
The third approach to be considered is based upon the concept that
maximum oxidant observed locally is the result of only locally produced emis-
sions of precursors. This assumption is not correct since it is well known
that local emissions are not the sole cause of local maximum oxidant. How-
ever, it is known that the local emissions can be transported and affect a
different receptor area at a later time. Therefore, this approach is one of
requiring that each source area be concerned with its own problem and ignore
the contributions from other areas. The major problem involved in this ap-
proach is that the possibility exists that over-control of local sources could
result. Whether or not this will actually occur is a moot point at the pre-
sent time since the type of information necessary to confirm or deny this is
not available at the present time. Specifically, the complex qualitative
relationship of transported and locally generated oxidant is not available at
the present. However, from an administrative viewpoint, this approach is
probably the most acceptable alternative. Each urban source area would be
concerned with reducing its own emissions. EPA would have the responsibility
to ensure that each urban area do the most to reduce its own level of emis-
sions to ensure that the standard is not violated as a result of local emis-
sions. This is a fairly conservative approach that could be justified as
25
-------
being a reasonable method that allows states to develop their own implemen-
tation plans and keeps the EPA role to one of surveying the plans developed.
In addition to this, the current data available in most urban areas would be
sufficient to support this approach.
The final approach is similar to approach number two in that a fraction
of local maximum oxidant concentrations is associated with local emissions.
However, when calculating control requirements, the local fraction of the
maximum oxidant is reduced to the extent that the national standard of 0.08
ppm is not exceeded. This means that the contributions from upwind sources is
completely balanced by increased reductions by local sources. This approach
requires that an agency know the contribution of local sources to the maximum
oxidant concentration. This also requires a deterministic quantification of
the local contribution to maximum oxidant; this is still a considerable pro-
blem. Additionally, it assumes that transported oxidant can never be greater
than the standard. This may or may not be the case. Washington, D. C., and
Baltimore are two large metropolitan areas separated by only 40 miles. The
possibility of transport of large amounts of ozone between these two metro-
politan areas is quite likely. Therefore, Baltimore may never be able to
control sufficiently to attain the standard. A policy would need to be
developed relative to this situation since it may be repeated many times on
the east coast. Finally, the possibility of over-control exists under this
approach. As upwind sources reduce their contribution to an urban area, the
need for the restrictive controls on local sources will diminish.
Based upon the foregoing discussions, the third approach appears to be
the most amenable to usage in controlling photochemical oxidant. The other
approaches may be more scientifically accurate, but they are not usable with
the data base presently available in most urban areas. Certainly, as more
data becomes available, reconsideration of the source-receptor relationship
will be necessary.
26
-------
CONCLUSION AND RECOMMENDATIONS
Based upon the above discussion, there are four recommendations we would
like to make concerning the issue of the optimum oxidant control strategy.
The first two recommendations concern the existing control program. The other
two recommendations concern further work that could be done to improve the
oxidant control program.
The first recommendation concerns the definition of the optimum oxidant
control strategy. In the urban areas, positive control of organic emissions
is necessary and must be continued. Control of urban emissions of oxides of
nitrogen is necessary for the benefit of the rural areas. However, the con-
trol of urban oxides of nitrogen is counterproductive when considering urban
oxidant. Therefore, the best policy for the interim is to maintain urban
oxides of nitrogen at their current level. If abatement of oxides of nitrogen
is to be initiated in urban areas, the control program must be coordinated
with the organic emission control program to ensure that the NMHC/NO ratio
X
continues to decrease. In the rural areas, the precursor control program is
not as clear. Sufficient data is not presently available to adequately assess
the problem. Therefore, any control strategy for reducing rural precursor for
the benefit of rural area oxidant would be very speculative. Until more
information is available, the control program for rural precursors should be
held in abeyance.
The second recommendation concerns the source-receptor relationship for
use in oxidant control programs. The approach that is most usable at the
present time is based upon the assumption that urban area emissions are solely
responsible for the observed oxidant maximum in that urban area. This is a
simplification of more complex precursor-oxidant relationships which are more
scientifically valid but are too complicated to be of value at the present
time. The available data would support the use of this method. This technique
should be viewed as an interim approach that can be changed as the more valid
relationships become usable.
27
-------
The third recommendation concerns the consolidation of all the informa-
tion relating to the phenomenon of photochemical oxidant pollution. Specifi-
cally, work needs to be done to compile a quantitative model of photochemical
oxidant pollution. This model should include urban-rural interactions, the
role of natural precursor, and the role of stratospheric intrusion. At the
present time, these phenomena have been considered in piecemeal fashion in the
literature. Work needs to be done to integrate all the current information
into a comprehensive quantitative assessment of photochemical oxidant pol-
lution. Without this type of work, control strategy development will remain
speculative and qualitative. We believe this is an enormous task; however, as
the process is initiated, it will crystallize the available knowledge and will
enphasize those aspects of the problem that need further research.
Finally, as pointed out earlier to EPA (24), we are of the opinion that
there is a need for a different administrative mechanism for coordinating
whatever multistate abatement strategies are adopted. Though there is provi-
sion in existing law that the states must provide in the SIP for interstate
coordination of the abatement programs, our own observations have demonstrated
that such an indirect approach has been somewhat ineffective applied to the
traditional pollutants in the past. When the same argument is applied to
photochemical oxidants in an area that may be comprised of several EPA Re-
gions, it is clear to us that the current governmental machinery needs work.
Whether or not new amendments to the Act must be proposed remains to be seen.
Perhaps an interpretive ruling on the regulations that require this interstate
mechanism will get the necessary attention to the matter. We feel assured
that this type of problem will be similar for the sulfate problem that EPA may
be facing shortly.
28
-------
REVIEW AND ANALYSIS
Paul Koziar and Brooks Becker
SUMMARY
At the request of the United States Environmental Protection Agency a
review of the recent scientific literature pertaining to the photochemical
oxidant problem was conducted. The literature reviewed was obtained from
technical journals and from papers presented at the International Conference
on Photochemical Oxidant Pollution and Its Control and at the Specialty
Conference on Ozone/Oxidant Interactions with the Total Environment. This
review was prepared to evaluate the qualitative basis for the existing oxidant
control strategy and to recommend an optimum urban oxidant control strategy
identifying the direction that hydrocarbon and nitrogen oxide control should
take.
The four quantitative approaches for an optimum urban oxidant strategy
suggested by EPA were reviewed in order to recommend one or to suggest al-
ternatives .
Field and lab data were reviewed both independently and compared with one
another to evaluate the scientific evidence presented. After studying the
evidence, it is our opinion that the control of hydrocarbon emissions is the
optimum urban oxidant control strategy.
On the basis of scientific evidence, the control of nitrogen oxides is
judged to be tolerable, but from the implementation standpoint, it is un-
desirable at this time.
29
-------
Of the four quantitative control approaches suggested, the one judged
best is Approach IV. This approach utilizes the fraction of the maximum
oxidant concentration as a means of determining local control but implies that
further local control may be necessary to offset extraneous emissions. Al-
though Approach IV is the best of those reviewed, modifications to it are
suggested, and these are described in the section on Quantitative Evaluation,
the Comparison of Approaches subsection.
DIRECTIONAL IMPACTS OF CONTROL — QUALITATIVE EVALUATION
Natural Source of Oxidants
Natural ozone occurs in the troposphere as a result of both meteoro-
logical phenomena and the photooxidation of natural emissions of hydrocarbons
and nitrogen oxides. The degree to which each of these mechanisms contributes
to ozone concentrations at the ground level will be discussed below.
Ozone formed in the stratosphere may be transported to the lower layers
of the troposphere in situations involving strong vertical mixing. Meteoro-
logical phenomena which may produce strong vertical mixing are:
• Active upper-level weather systems
• Intense squall lines ahead of cold fronts
• Large thunderstorm cells.
The first two meteorological conditions are usually associated with the
same mesoscale weather system, whereas thunderstorms may occur independently
as a result of local convective activity. The injection of ozone into the
troposphere from upper-level low pressure systems or intense squall lines
occurs as the tropopause (stratosphere-troposphere boundary) is brought closer
to the earth's surface, thereby facilitating mass exchange of the stratosphere
and the troposphere.
Large thunderstorms also may penetrate the tropopause and transport ozone
to the surface in strong downdrafts.
30
-------
Stratospheric transport of ozone, associated with active upper-level
weather systems, has been investigated by numerous people. Using ozonesonde
data gathered by the Air Force in the 1960's, Carney (25) concluded that most
high tropospheric ozone (0 ) concentrations are the result of stratospheric
transport. Sticksel (26), using the same data, estimated that mean monthly
averages of ozone in the lower layers of the troposphere may be as high as
0.065 ppm, and daily concentrations may be as high as 0.12 ppm. A reexamina-
tion of this data by Chatfield and Harrison (27) indicated that ozone may
range from 0.026 to 0.072 ppm in the middle troposphere. Chatfield and
Rasmussen (28) , using aircraft data, concluded that ozone ranges from 55 to 60
ppb across the United States, above the mixing layer.
The results of a study of ozonesonde data by Bach (29), which was gathered
by the U.S. Environmental Protection Agency (30), indicates that strato-
spheric transport of ozone into the upper troposphere occurs, but changes in
ozone concentrations below 3 km are primarily controlled by processes existing
below the boundary layer. He also concluded that mid- to upper-tropospheric
ozone can change by 50 percent during the day, but that these observed changes
do not exert any direct control upon ground-level ozone concentrations.
Surface ozone measurements have also been used to estimate the impact of
stratospheric ozone. Hathorn and Walker (18) analyzed surface ozone data
during a widespread episode, and attributed high ground-level ozone to strato-
spheric transport of ozone. Chatfield ad Harrison (27) concluded that surface
ozone from stratospheric transport ranges from 0.040 to 0.055 ppm. Their
conclusions were reached from analyzing ground-based ozone data recorded in
the northwest corner of Washington. An analysis of surface ozone data re-
corded in a remote area of western Colorado by Jones and LaHue (31) shows that
hourly averages of ozone approached 0.08 ppm, but no one mechanism is suggested
as the source.
Another meteorological phenomenon that has been suggested as a possible
mechanism for stratospheric transport of ozone in the troposphere is vertical
mixing associated with squall lines and thunderstorms. Davis and Jensen (32)
conducted a comprehensive analysis of the effect of weather systems on low-
31
-------
level ozone concentrations. They concluded that most ozone concentration
rises, which occurred in association with squall lines on thunderstorms, were
of short duration.
The photooxidation of natural emissions of hydrocarbons and nitrogen
oxides also can contribute to ozone in the troposphere. Earlier studies by
Rasmussen (33) indicated that the rates of organic emissions from vegetation,
in the form of terpene-type compounds, are higher than from man-made emissions
when compared on a global basis. Although the rate of organic emissions is
high, Robinson and Rasmussen (34) further reported that the photochemical
processes associated with natural emissions of hydrocarbons (HC) or nitrogen
oxides (NO ) account for about 0.02 ppm ozone. Their estimates were based
X
upon the irradiation of rural air in Tedlar bags.
Gay and Arnts (35) also investigated the potential for oxidant formation
from naturally emitted hydrocarbons. Measurements of natural hydrocarbon con-
centrations were made in and above a forest canopy. Concentrations of terpenes
in the canopy were highest in the nighttime hours, but above the canopy, con-
centrations were very low. Irradiation of these terpene-type compounds yielded
the following results:
• Terpenes react rapidly when irradiated with UV in the presence of
NO .
x
• The ozone produced by irradiation is a function of the HC/NO
ratio.
• Ozone reacts rapidly with terpenes.
• In rural areas, terpenes participate in aerosol formation, but
the significance of their oxidant formation is low.
The scientific data in the recent literature presents conflicting evi-
dence on the fraction of oxidants contributed from natural processes, es-
pecially the evidence presented for stratospheric transport of ozone to the
surface layers. Although most investigators agree that the contributions from
natural sources of oxidants exist, they disagree on the magnitude of the
contribution.
32
-------
Based upon our judgment of the conflicting evidence regarding the con-
tribution of natural ozone sources to the surface tropospheric ozone budget,
we have reached the following conclusions:
Stratospheric transport of ozone is significant and may be the
prime contributor to exceeding the ozone standard (0.08 ppm)
in some cases.
Ozone contributed from stratospheric transport alone cannot
explain the frequency and duration of oxidant concentrations
exceeding the standards or the maximum oxidants recorded at
urban sites.
The overall contribution of ozone from natural hydrocarbon
emissions alone is judged to be generally smaller than that of
the stratosphere, but may be higher in areas where specific
combinations of plant communities emitting terpenes exist.
In geographic areas where the additive effect of ozone from
the stratosphere and photooxidation of natural emissions can
occur, the frequency of exceeding the NAAQS is greater than
from the individual sources alone.
Ozone contributed from both stratospheric transport and
natural emissions cannot explain the frequency and duration
of oxidants exceeding the standards or the maximum oxidants
recorded at urban sites.
The conclusions reached here regarding the contribution of natural emissions
have some anthropogenic precursor control implications:
That precursor emissions from anthropogenic sources are the
prime contributors of maximum oxidant concentrations recorded
at the surface.
That the control of anthropogenic precursors is necessary to
reduce oxidant concentrations in urban areas.
That the extent to which the control of precursor emissions
can attain the standards is dependent upon the localized
influence of natural emissions.
Anthropogenic Sources of Oxidants
In the previous section, natural sources of oxidants were discussed to
evaluate their contribution to the control of photochemical oxidants. To
33
-------
evaluate the effect and required control of anthropogenic precursor emissions,
relationships between the source of precursor emissions and the maximum oxi-
dants recorded at the receptor must be quantified.
Prior to 1970, the relationships of oxidants to hydrocarbons and nitrogen
oxides were derived from data recorded only in large urban areas (2). Recently
though, monitoring in rural areas has identified that the maximum oxidants
recorded in rural and urban areas may exhibit different source-receptor
relationships.
Recent field studies by the EPA in the Ohio Valley (5) and in the north-
ern United States and Gulf Coast (30) have acquired data both in rural and
urban areas. The studies show that the frequency of ozone concentrations
exceeding the standards and the maximum ozone recorded in some rural areas is
comparable to what has been found in urban areas, despite the fact that the
local precursor emission base is relatively small. The transport of pollutant
emissions from urban areas to rural areas is hypothesized as a link between
urban and rural oxidant problems. Therefore, two different source-receptor
relationships need to be developed to understand the effectiveness of the
control of precursor emissions.
The scientific literature reviewed in the following sections will focus
upon the source-receptor relationship between urban oxidants and urban pre-
cursors. The review will also look at long-range pollutant transport because
the urban oxidant problem and the rural oxidant problem appear to be related.
SOURCE-RECEPTOR RELATIONSHIPS (URBAN AND NEAR URBAN AREAS)
Field Data
Investigations into the specific relationships between urban hydrocarbon
and nitrogen oxide emissions and/or concentrations, and urban oxidants have
been conducted extensively in recent years utilizing field data. Different
field data gathering techniques were used to determine source-receptor re-
lationships and are:
34
-------
• Upwind and downwind monitoring of urban areas for O_, hydrocarbons,
and nitrogen oxides by fixed ground stations or airborne measure-
ments ;
• Trajectory analysis;
• Oxidant and emission trends.
Ground-based ozone monitoring data have been used to qualify and quantify
the relationship between ozone, hydrocarbons, and nitrogen oxides. The national
EPA strategy for the control of hydrocarbon emissions (Appendix J) was based
on the hydrocarbon and ozone data recorded at CAMP monitoring sites across the
United States (2). Recently, extensive use of both ground-based and airborne
ozone data gathered by local, state, and Federal air pollution control agencies,
and by the private sector, have been used to qualitatively assess the contri-
bution of specific urban areas and their associated emissions on downwind
ozone concentrations.
The effect of urban complexes on downwind ground-based ozone monitors was
investigated in New York City by Rubino et al. (36) and Cleveland et al. (37),
in Camden-Philadelphia by Cleveland and Kleiner (38), and in Chicago by Lyons
and Cole (39). In all cases, elevated ozone concentrations were reported
downwind of the urban complex. Cleveland et al. (37) also found that
there was a shift of photochemical activity to later hours of the day with
increasing distance downwind of New York. This determination was based upon
the fact that the time of maximum ozone concentration progressed with in-
creasing distance downwind of New York City.
A study conducted in Ohio by the Environmental Protection Agency (5)
utilized both ground-level and airborne ozone measurements to assess the con-
tribution of the emissions of specific urban areas on downwind ozone con-
centrations. Of the urban areas investigated, elevated ozone concentrations
were recorded by ground stations and by airborne flights downwind. White et
al. (40) also found elevated ozone concentrations above the ground, downwind
of St. Louis, using data taken by aircraft.
35
-------
Data contained in the reports reviewed above suggest a qualitative
relationship between emissions of hydrocarbons and nitrogen oxides and ozone.
In all cases, the concentrations of ozone recorded downwind of the urban
complexes exceeded the upwind concentrations. In many cases, upwind con-
centrations of ozone were above 0.08 ppm.
Hydrocarbon and nitrogen oxide data recorded at monitoring sites simul-
taneously with ozone were used in the past, especially in the Los Angeles area
(2), to relate ozone concentrations to hydrocarbon and nitrogen oxide levels.
With recent evidence accumulating that broad geographic areas, especially east
of the Mississippi, experience widespread high ozone concentrations, the
thrust of recent field studies has been concentrated on the source-receptor
relationship of monitoring sites in urban and rural areas. Field studies, such
as the ones conducted in the Ohio Valley in 1974 (5) and in the northern
United States and Gulf Coast in 1975 (30), have concluded that different
source-receptor relationships exist between rural and urban areas. In urban
areas, the concentrations of both ozone and precursor emissions are high,
whereas in rural areas, the concentrations of ozone may be high but precursor
emissions are low. In these EPA studies, both elevated ozone and hydrocarbon
concentrations were found downwind of cities. The maximum concentration of
ozone and hydrocarbons occurred at different distances downwind.
Studies completed in Texas by Walker (41) and Neal et al. (42) used 6
a.m. to 9 a.m. nonmethane hydrocarbons and recorded the maximum ozone con-
centrations to establish source-receptor relationships. Similarly, Huffman et
al. (43) used 6 a.m. to 9 a.m. nonmethane hydrocarbons and nitrogen dioxide
concentrations for Indianapolis. For these studies, results indicate a weak
relationship between precursor emissions and maximum ozone concentrations.
Trajectory analysis has also been used to establish relationships between
precursor emissions and maximum ozone concentrations. The EPA field studies
reviewed earlier (5,30) utilized trajectory analysis and determined that the
highest ozone concentrations measured in the studies can be attributed to
principal cities or areas of high ozone precursor emissions. Meyer et al.
36
-------
(20) performed a study using ozone data recorded at rural and urban locations.
The correlation between local precursor emissions is significant and positive
for the urban locations, and there exists a high degree of intercorrelation
between HC and NO emissions. No apparent relationship between local, manmade
X
precursor emissions and ozone is apparent at rural sites.
Data recorded from ambient ozone monitors over long periods of time has
also been analyzed, where control strategies have been implemented for nitrogen
oxides and hydrocarbons, to establish source-receptor relationships. In one
such study, Trijonis et al. (15) compared the historical trends in oxidant
concentrations in Los Angeles with trends in hydrocarbon and nitrogen con-
centrations and emissions. He concluded that improvements in oxidant air
quality agreed well with the reductions in hydrocarbon emissions, and that NO
X
emission increases may have contributed to the reduction by decreasing the
HC/NO ratio.
From field data reviewed in this section, it appears that a qualitative
relationship does exist between precursor emissions and the maximum oxidant
concentrations recorded at urban sites. The correlation of hydrocarbons to
maximum oxidant concentrations is better supported than the relationships
between oxidants and nitrogen oxides. It appears, though, that there may be
inherent limitations in correlating ozone data at specific monitoring sites
with primary pollutants that may undergo complex photochemical reactions.
Based upon the evidence presented from the field data and considering
that there are limitations to its use for control strategy development, the
optimum strategy for oxidant reduction is the control of hydrocarbon emis-
sions.
Lab Data
Earlier Smog Chamber Studies—
Before 1970, it was reported in the literature that several studies of
37
-------
ultraviolet-irradiated hydrocarbon and nitrogen oxide mixtures had been con-
ducted to simulate the production of ozone in the ambient air. The results of
these studies, as reported in the criteria document for photochemical oxidants
(2), show a dependence of oxidants on hydrocarbon concentrations in spite of
the many differences in choice of initial reactants and experimental design.
Initial hydrocarbon concentrations used in those studies ranged from 1.0 ppm C
to 12.0 ppm C, and the composition of the hydrocarbon mixture was either
automobile exhaust or propylene. A similar dependence of oxidants on nitrogen
oxides was demonstrated in these smog chamber studies. Concentrations of
nitrogen oxides ranged from 0.5 ppm to 1.0 ppm. These laboratory photooxida-
tion experiments also suggest the existence of an optimum HC to NO ratio with
regard to the maximum attainable oxidant, in addition to the dependence on the
absolute concentrations of hydrocarbons and nitrogen oxides.
The maximum ozone produced in these laboratory studies (2) was within the
range of the maximum hourly ozone concentration recorded at the CAMP moni-
toring sites located across the United States from 1964-1967. Similarly, the
concentrations of the initial reactants irradiated were comparable to those
concentrations of hydrocarbons (44) and nitrogen oxides (11) recorded in major
cities in the United states. Although some of the initial reactant concentra-
tions used in the smog chamber studies were within the range of ambient
monitored levels, the majority fell outside of the range of ambient monitored
values.
The application of the results of smog chamber studies to determine the
control of photochemical oxidants was tested using the oxidant isopleths
derived from laboratory experiments that showed the effect of varying initial
precursor hydrocarbon and nitrogen oxide concentrations. In all cases, at
every HC-NO concentration, reductions in hydrocarbons without NO reduction
X X
resulted in reduction of oxidants, whereas NO reduction without HC reduction
X
did not always lead to a reduction of oxidants.
Recent Smog Chamber Studies—
Recent indoor smog chamber studies (45-47) have focused their attention
38
-------
on specific interrelationships of the initial reactants of the photooxidi^ed
mixture, the duration and light intensity of the irradiation source, and the
relative reactivity of the hydrocarbon mixtures. In all cases, the results
have still yielded the dependence of oxidants on both hydrocarbon and nitrogen
oxide concentrations. The results of smog chamber studies completed by
Dimitriades (45) in which automobile exhaust was irradiated for a 6-hour
period, indicate that the amount of ozone produced is dependent upon the
HC/NO ratio of the irradiated mixture, and at higher HC/NO ratios the ozone
X X
produced in the chamber is maintained for a longer duration.
Powers (48) assessed the oxidant-formation potential of both catalyst-
equipped and unequipped automobile exhaust mixtures irradiated over a 10-hour
period. He found that the maximum ozone produced occurred using a hydrocarbon
mixture representative of a vehicle without a catalyst. The results of the
experiment also showed that hydrocarbon reactivity was a valid concept under
experimental conditions, other smog chamber studies by Glasson and Wendschuh
(46) and Pitts et al. (47) have confirmed that hydrocarbon reactivity is
important because hydrocarbon compounds, when irradiated, will show differ-
ences in the amount of ozone that can be produced and the time necessary to
reach the maximum ozone.
To simulate the effect of atmospheric transport conditions, Glasson and
Wendschuh (46) irradiated the initial reactant mixture under alternating 6-
hour periods of light and darkness. Under these conditions, automobile
exhaust produced the highest multiday maximum of ozone in a 3-day period.
Similarly, the effects of atmospheric dilution were tested by Powers (48). The
results of the dilution experiment indicate that reactivity differences are
still discernable under the prolonged experimental dilution.
These smog chamber studies are representative of actual atmospheric
conditions only to the extent to which the following conditions are satisfied:
The interdependence of initial reactants and resultant oxidant
levels are the same in the smog chamber as in the atmosphere.
39
-------
The composition of the initial hydrocarbons in the smog chamber
are the same as in the atmosphere.
The hydrocarbon to nitrogen oxide ratios of the initial reactants
are the same as in the atmosphere.
Also, an important consideration is "dirty" chamber effects associated with
man-made enclosures. Although the conditions in the atmosphere cannot be
duplicated exactly in the smog chamber, these results can be used in a gross
way to test the effectiveness of precursor emission control strategies for
oxidants.
The smog chamber studies of Dimitriades (45,49) indicate that the control
of precursor emissions is dependent upon the HC/NO ratio and that the control
X
of hydrocarbons without NO control, no matter what the HC/NO ratio is, will
X X
result in an oxidant reduction. On the other hand, at certain HC/NO ratios,
X
NO reduction does not always lead to a reduction in ozone without HC re-
X
duction. Dodge (10) developed a photochemical model and simulated the results
of Dimitriades with good correlation. Extrapolation of these results re-
garding control strategy implications was tested. The results were similar to
those of Dimitriades (45).
Most recently, outdoor smog chamber studies have been employed to simulate
conditions of photochemical oxidant formation in the ambient atmosphere. The
results of these studies show a similar dependence of oxidant concentrations
on hydrocarbons, nitrogen oxides, and the HC/NO ratio as 'reported in the in-
X
door smog chamber. In addition, experiments on the effects of multiday ir-
radiation, dilution, and light intensity were tested.
In an outdoor smog chamber study performed by Sickles et al. (12),
initial chamber reactants were irradiated for daylight periods and subjected
to a period of dilution to simulate the movement of air from an urban to a
nonurban area. The results of this study indicate that, in partially spent,
simulated urban photochemical systems, ozone could be generated in concen-
trations above the NAAQS with low absolute NO concentrations and at high
HC/NO ratios. The data also suggest that an ozone carryover exists, and that
40
-------
this carryover may provide a high minimum on which to build a high maximum the
next day. The carryover or halflife of ozone concentrations ranged from 20 to
30 hours.
The effect of light intensity was studied by Jeffries et al. (50), and
results show that fluctuations in solar radiation, provided they are small,
tend to accelerate a photochemical system giving higher rates of NO and 0
formation compared to results of experiments done when either constant or
gradual changes in light intensity were applied.
The results obtained from indoor smog chamber studies are representative
of the actual atmosphere only to the extent to which those conditions speci-
fied in the subsection on Recent Smog Chamber Studies are satisfied. Concen-
trations of the initial chamber reactants in these studies ranged from 1.0 ppm
C to 7.0 ppm C of hydrocarbons and 0.10 ppm to 1.0 ppm of NO . Propylene was
used as the hydrocarbon irradiated in the experiments done by Jeffries et al.
(50), and a surrogate urban mixture of hydrocarbons was used by Sickles et al.
(12). The fact that outdoor smog chambers are irradiated in natural sunlight
and are subject to natural variations of the light intensity profile makes
outdoor smog chamber studies inherently more closely related to the simulation
of air parcels in the urban environment. In addition, it has been reported
(50) that the heterogenous reactions (dirty chamber effects) are less sig-
nificant in these experiments due to the large Teflon chambers.
The results of these outdoor smog chamber experiments have introduced
some important implications for control of precursor emissions. These results
are:
The half-life of ozone being 20 to 30 hours, the ozone carryover
may produce a high minimum on which to build a high maximum. This
carryover may occur locally, or it could be transported downwind to
provide a high minimum for another urban area.
A mechanism has been provided to explain one source that may
contribute to high ozone in rural areas. This mechanism suggests
that the emission carryover of a photochemical oxidant system could
cause the production of ozone above the NAAQS with low absolute NO
concentrations and at high HC/NO ratios. x
X
41
-------
These results imply that the amount of precursor emissions to be controlled
must go beyond the amount that is necessary to alleviate the local urban
oxidant problem. Both locally generated emissions and oxidants may provide a
significant carryover that can either provide the local downwind urban area
with a high ozone minimum or a sufficient amount of emissions on which to
build a higher maximum ozone concentration or both.
Comparison of Outdoor-Indoor Smog Chamber Studies—
The results of those indoor and outdoor smog chamber studies reviewed in
the subsection on Recent Smog Chamber Studies show a dependence of oxidants on
hydrocarbon concentrations in spite of the many differences in experimental
design. In addition, indoor (46,47) and outdoor (9) studies have shown a
dependence of oxidant concentrations on individual hydrocarbon compounds. The
composition of the initial hydrocarbon mixture used in both indoor and outdoor
smog chamber studies varied, with propylene and auto exhaust being the prime
initial smog chamber reactants. Similarly, the dependence of oxidant concen-
trations on nitrogen oxide has been demonstrated in both chamber environments,
although the rates of NO formation have been reported (50) to be higher due
to natural light intensity effects in outdoor smog chambers.
Dimitriades (45) demonstrated in an indoor smog chamber that ozone was
dependent upon the HC/NO ratio of the initial chamber reactants and that
ozone produced in the chamber is maintained for a longer duration. Also, in
an outdoor smog chamber study done by Jeffries et al. (9) in which a synthetic
urban mixture of acetylene, paraffins, and olefins were used, the maximum
ozone produced increased proportionately to the increase in the initial HC/NO
X
ratio of the chamber.
The maximum ozone produced in both chamber environs compare favorably,
although the experiments of Jeffries et al. (50) indicate that at higher HC
concentrations, the maximum ozone produced in outdoor smog chambers exceeded
the amount formed in those indoor studies done by Dimitriades (45). Jeffries
et al. (50) say that the effects of natural light intensity are the cause of
the differences observed.
42
-------
In comparing the results of indoor and outdoor smog chamber studies as to
their representativeness of air parcels in the urban environment, it appears
that outdoor smog chamber studies simulate atmospheric conditions more closely
for the following reasons:
As reported by Jeffries et al. (50), the use of large Teflon chambers
appears to reduce dirty chamber effects, which may be a significant
factor in indoor chamber studies.
The irradiation of HC-NO mixtures in the presence of natural
X
sunlight results in a better approximation of the photooxidation
process in the ambient air.
It appears that outdoor smog chamber studies provide a better under-
standing of photochemical systems and allow a better extrapolation of their
results to real world situation. However, extrapolation of indoor and outdoor
smog chamber results to the control of precursor emissions do not yield sig-
nificantly different results. Only a few of the papers reviewed deal directly
with the quantitative implications of the control of precursor emissions on
reduction of photochemical oxidants. The data implies that the reduction of
hydrocarbons in all cases will reduce urban oxidant concentrations, whereas
the reduction in nitrogen oxides may have an effect depending upon the HC/NO
X
ratio. Therefore, from the results obtained by smog chamber data, the optimum
control of urban oxidants can be achieved by a reduction in hydrocarbon emis-
sions. Of course, the degree to which urban oxidant reduction can be achieved
by the control of local hydrocarbon emissions, as determined by both indoor
and outdoor smog chamber data, is dependent upon whether ozone concentrations
in an area are significantly affected by nonlocal sources.
Lab and Field Data/Oxidant — HC/NO Relationships
X
The field data and lab data have been reviewed to assess whether quanti-
tative and qualitative relationships exist between anthropogenic emissions of
hydrocarbons and nitrogen oxides and oxidant concentrations in the ambient
air. The review was conducted to evaluate whether the relative evidence pre-
sented by the field and lab data support the original hypothesis that hydro-
carbon reduction is the optimum urban oxidant control strategy.
43
-------
As discussed in the subsection on Field Data, the field data does show
evidence that a qualitative relationship exists between hydrocarbons and oxi-
dant concentrations particularly in urban areas. This qualitative relation-
ship is supported by the fact that there exist elevated ozone concentrations
downwind of major urban complexes which exceed upwind concentrations. Al-
though downwind monitoring does suggest a positive relationship, the correla-
tion of ozone concentrations with hydrocarbon and nitrogen oxide data recorded
at the same monitoring site on a diurnal time frame has been marginal, at
best. Correlations of this type, though, may be limited because of the com-
plexities associated with relating primary and secondary pollutants under
dynamic atmospheric conditions in a short time frame.
Long-term trends of oxidant, both in the Los Angeles area (15) and in
major cities of the United States (16), have been significantly related to
trends in hydrocarbon emissions. In addition, Martinez et al. (16) found the
combined HC-NO emission trends to be more significantly correlated than HC
alone.
The evidence presented by the field data tends to support the conclusions
that the control of hydrocarbons alone is the optimum urban oxidant control
strategy. The selection of the control of hydrocarbons only is based upon the
fact that more scientific evidence presented in field data supports hydro-
carbon reduction, whereas the effects of NO reduction may cause as yet
x
undetermined and inconclusive effects upon oxidants.
Lab data, particularly that derived from both indoor and outdoor smog
chamber studies, has qualitatively and quantitatively established relation-
ships between hydrocarbons, nitrogen oxides, the ratios of HC to NO , and
X
oxidants, in spite of the many design differences associated with these
experiments. Indoor smog chamber studies, particularly those of Dimitriades
(45), have been used to postulate relationships between the control of oxidants
and reductions in hydrocarbon and nitrogen oxide emissions. These data
suggest that the control of hydrocarbons at any combination of HC-NO con-
x
centrations will cause a reduction in oxidant concentration, whereas the
44
-------
effect of NO control is dependent upon the HC/NO ratio and, therefore, is
X X
highly speculative.
Although the evidence strongly suggests that control of hydrocarbons
alone may be the optimum urban oxidant control strategy, Dimitriades (45)
cautions that the extrapolation of his results may be valid only so far as the
chamber environment represents ambient conditions. The effects of the chamber
itself, plus the fact that most indoor chamber experiments have been conducted
using both hydrocarbon concentrations, individual hydrocarbon compounds, and
nitrogen oxide concentrations which may be unrepresentative of ambient air,
introduces an element of speculation into the extrapolation of chamber results
to real world situations. Outdoor smog chamber studies have duplicated the
results obtained from indoor studies and in addition, as Jeffries et al. (50)
report, chamber effects have been minimized.
Considering the fact that both outdoor and indoor smog chambers can only
approximate the real world, the results still indicate qualitatively that
control of hydrocarbons is the optimum urban oxidant strategy. Although the
effect of NO in the chamber can be demonstrated, its effect on oxidant can be
different depending upon the chemical balance (HC/NO ) of the ambient air.
X
The extrapolation of NO effects in the chamber to the ambient air at this
x
time would be inappropriate without further studies to support NO effects
X
over all possible cases to be expected in the ambient air.
The precursor control implications, as determined from field and lab
data, support the fact that hydrocarbon control is the optimum urban oxidant
control strategy. The selection of this strategy, though, is better supported
by lab data than field data at this time.
Recommendations — Oxidant Control Strategies
Based upon our judgment of the evidence presented in the review of field
and lab data, the optimum strategy for urban oxidant reduction is hydrocarbon
emission control. As far as the urban oxidant problem is concerned, it is our
recommendation that NO control is tolerable when scientific evidence is
x
considered but undesirable from an implementation standpoint.
45
-------
The discussions below are intended both to qualify the recommendations
made and to touch upon other topics that were considered in arriving at the
recommendation. Also included are general suggestions of research that may
help to explain further the complex photochemical relationships associated
with the urban oxidant problem.
Hydrocarbon Control—
The most recent scientific evidence tends to support continuation of
hydrocarbon reduction as an effective strategy. The most promising reports
are those of apparent oxidant reduction paralleling the HC reduction in Los
Angeles (15). The word "apparent" is used because long term urban sampling
may have been subject to changes in instruments, methods, site locations,
instrument height, probe materials, local changes in alkenes and NO , seasonal
changes in temperature and solar radiation from year to year, increases in
data quality control efforts, and recently discovered problems in calibration.
Smog chamber data support the trends detected in Los Angeles; i.e.,
control of hydrocarbons should, in general, reduce oxidants. The agreement
between the field and lab data in this situation, though, must be qualified
because, to a certain degree, the meteorological regime existing in Los
Angeles approximates the simulation of a confined air mass that exists in a
laboratory smog chamber. The Los Angeles area, though, is geographically
unique and is quite different from the vast areas to the east of the Mis-
sissippi.
Field studies in the Northern United States and Gulf Coast (30), the Ohio
Valley (5), and the Great Lakes (39) have identified different macroscale and
mesoscale meteorological situations under which high ozone episodes may occur.
The extrapolation of the lab and field data, which represent relatively con-
fined air mass situations, to these areas may not be quantitatively representa-
tive because of the complex source-receptor relationships. Although the
quantitative relationship may be different under different meteorological
regimes, the data still supports the fact that hydrocarbon control is the
optimum urban oxidant control strategy.
46
-------
In addition to the field aiul lab drtt.-* p« c^s^ut nA, hy*1t vu'at K>i\ , >>u
-------
Research—
The review of both lab and field studies has indicated that there are
still deficiencies in applying the data gathered to the quantification of
oxidant reduction to be achieved by EC and NO control. Of the EPA field
x
studies completed to date, the main emphasis has been directed toward charac-
terizing the ozone problems in broad geographic areas. The results of these
studies have identified that both rural and urban areas have different source-
receptor relationships. Since national EPA policy is leaning towards the
control of urban precursors as the first step in attaining the NAAQS for
oxidants, additional research is recommended that deals directly with the
complex source-receptor relationship of individual cities.
Specific recommendations for additional research axe:
Further investigate the feasibility of simulating continental
ozone problems through chamber experiments using appropriate dilution
and reradiation experiments so as to simulate transport situations.
Using chamber experiments, determine the ozone yields of ambient
air containing representative concentrations of hydrocarbons and
nitrogen oxides.
Where different meteorological regimes impact the source-receptor
relationship of broad geographic areas, intensively study represen-
tative cities within that area to determine relationships among
ozone, hydrocarbons, hydrocarbon compounds, and nitrogen oxides.
Chamber studies could possibly be conducted simultaneously with the
data acquisition effort.
These specific recommendations for additional research may help to clarify
the source-receptor relationship in urban areas and ultimately provide ad-
ditional data to aid in the design of State-level control strategies for
the control of oxidants.
QUANTITATIVE EVALUATION
In the preceding sections, lab and field data were discussed in terms of
the directional impacts of oxidant control. It was recommended, after a
48
-------
review of the pertinent scientific literature, that the optimum strategy
for urban oxidant reduction was the control of hydrocarbon emissions. In
the following sections, four different approaches will be discussed evalu-
ating the quantitative impacts of hydrocarbon control in terms of the re-
lationship between ambient oxidant concentrations and emission control,
and the relationship between the source area and receptor area. Each approach
will be discussed individually and collectively in terms of its relative
advantages and disadvantages. The discussion will be focused upon a "cause"
and "effect" relationship with a "source area" and "receptor area," respectively,
defined as the area where hydrocarbons are emitted and the area where air
quality is being impacted.
Approach I
The approach proposed for consideration here is identified with the cause
of the maximum oxidant concentration in the "receptor area," due to both local
and upwind emissions in the "source" area. This approach requires that all
the source areas impacting upon the receptor area be identified and that the
impact of upwind emissions be quantitatively evaluated to determine the amount
of oxidants being contributed from upwind sources. For large urban complexes
which are isolated from upwind influences, such as Los Angeles, this approach
may be feasible but for an area such as the northeastern United States where
many cities may have influence upon the air quality of others, this approach
could involve a major research effort. It is not, however, impossible.
Upwind influences of emissions on worst case oxidant days could be
accomplished by identifying the source areas by trajectory analyses and
quantifying their influence by applying one of the state-of-the-art photo-
chemical models. The results obtained from this exercise, though, could be
quite speculative considering the fact that smog chamber data (12,46) and
field studies (5,30) have indicated that both emission and ozone carryover may
be a significant factor in contributing to second-day high minimums on which a
high maximum can be produced. The effect of carryover from one urban area to
the next may underestimate the effect of other urban areas on the local urban
oxidant problem and, hence, cause more local control to be imposed.
49
-------
Considering the time and effort that may be invested in this work and the
fact that the results may be speculative, this approach is not very advan-
tageous .
Approach II
This approach considers the fraction of the observed maximum oxidant
concentration as the effect. The cause is identified with the local emis-
sions. Application of this method to determine the degree of local emissions
control requires that ozone be obtained from an upwind and a downwind receptor
so that a maximum oxidant fraction can be estimated. The assumptions that are
made when determining the maximum oxidant fraction are:
• That the upwind and downwind receptors are situated so that
chemical and physical interferences are minimal. For example,
if the upwind monitor is being influenced in some way by local
NO emissions, the maximum oxidant fraction calculated would be
Xi
erroneously high. In other words, the fraction of oxidant due to
local emissions would be high, and more local control of hydrocarbons
would be required.
• That the fraction of the maximum oxidant estimated does not
contain any oxidant produced by extraneous upwind emissions; i.e.,
the fraction of maximum oxidant estimated is only due to local
emissions.
The problem of site monitoring referred to in the first assumption can be
solved partially by careful placement of monitors so as to minimize physical
and chemical interferences. The second assumption, though, presents some
complications.
The fraction of the maximum oxidant concentration across the urban area,
estimated from measurements at the two receptors, may be the result of both
extraneous and local emissions. The upwind and downwind monitors should be
sited so that the maximum oxidant fraction across the urban area is being
measured. Also, the downwind monitoring site should be placed at a location
where local emissions will produce their maximum oxidant concentration. Even
if the siting problem could be solved, the problem of quantifying the indi-
50
-------
vidual contribution of oxidants from local and extraneous emissions still
exists.
Although there are some serious disadvantages to this method, especially
for large cities grouped together, it could be applied to urban areas that are
not being substantially affected by extraneous emissions. From recent field
studies, urban plumes of major cities have been tracked out to 100 km. A
criterion for application of this method might be that the urban area would
have to be outside of a 100-km radius of another major urban area.
Approach III
The third approach is identified with the "effect" being the maximum
oxidant concentration observed in the receptor area and "cause" being the
"local" emissions. This method is another reiteration of the Appendix-J
method, since it specifies that control be confined within the urban area
where the oxidant problem was observed. The assumption that this method
agrees with is that local emissions are the sole contributors to the local
oxidant problem. It also neglects the importance of emission and/or oxidant
transport.
The obvious disadvantage of this method is that application of controls
calculated by this approach will result in over-control in the source area. It
has been argued, though, that the disadvantage in actuality is not necessarily
a problem because the sum total of the "local" and "transported" contributions
may exceed the 0.08 ppm standard, even though individual contributions are
less than 0.08 ppm. In addition, it has been argued that local emissions
almost surely contribute to or add additional problems to downwind areas.
To clarify the differences in opinion offered in the evaluation of this
approach, an example is provided that will illustrate the use of this approach
in Wisconsin.
Figure 1 is a graphical layout of the urban complex associated with the
areas of southeast Wisconsin, northeast Illinois, and northwest Indiana.
51
-------
GRAFTON
(.255 ppm)
MILWAUKEE
(.216 ppm)
KENOSHA
(.18H ppm)
CHICAGO
OZONE
MONITORING
SITES
Figure 1. Maximum hourly ozone concentrations recorded at monitoring sites
in Southeastern Wisconsin on August 20, 1976.
52
-------
Plotted on the map are the maximum oxidant values recorded at three monitors
in southeast Wisconsin for August 20, 1976. Meteorological conditions asso-
ciated with these high oxidant values were characterized by sunny skies,
moderate wind speeds, and wind direction from southwest through southeast.
For discussion purposes here, let us assume that these concentrations
were the maximum oxidants measured that year. Applying the methodology of
Approach III, the maximum concentration of 0.255 ppm would be used to cal-
culate the necessary oxidant reduction for southeast Wisconsin, i.e. , the
Milwaukee urban complex. In this case, the amount of hydrocarbon control
would be based upon eliminating a fraction of 0,175 ppm oxidants even though
the sites upwind of Milwaukee are approximately 0.20 ppm. As is the case when
most ozone concentrations recorded in southeast Wisconsin exceed 0.08 ppm,
transported ozone dominates the local urban situation. Therefore, it would
not be applicable to apply this control method to situations where the trans-
port dominates the local problem, as reported in northeast states by Rubino et
al. (36) and Cleveland et al. (37). Although the calculation of hydrocarbon
by this method may not be valid in transport-dominated areas, the argument
that local emissions cause additional problems in downwind areas is valid and
cannot be neglected.
Approach IV
In this approach, the fraction of the maximum oxidant concentration is
defined as the "effect" and local emissions as "the cause." The methodology
proposed here is similar to that of the second approach but acknowledges the
fact that increased control of local emissions may be necessary to offset a
contribution from upwind sources. In other words, each urban complex would be
required not only to achieve 0.08 ppm, but to control its emissions further to
offset extraneous upwind emissions.
No method for estimating the necessary oxidant fraction to offset extraneous
sources is proposed in addition to the arguments already expressed in the
subsection on Approach II. Therefore, some criteria must be developed to
account for the offset required. This additional amount of control should
53
-------
not, of course, be based on the values recorded at the upwind monitoring site
but on a prescribed offset, for instance 0.02 ppm. The prescribed offset
eliminates any possible problems that may exist, especially those that may
arise from the interstate transport of pollutants, since transported air from
each urban complex would be cleaner than the standards.
Comparison of Approaches
The control strategies reviewed in the previous sections can be separated
into two general categories. Approaches I and III relate the maximum oxidant
concentration to the "effect" and either local and/or upwind emissions to
"cause." Approaches II and IV relate the fraction of the maximum oxidant
concentration to the "effect" and local emissions to the "cause." The ad-
vantages and disadvantages of each approach have been discussed individually.
This discussion, therefore, will focus on whether any of these approaches will
be recommended as a strategy for urban oxidant reduction.
The goal of any air pollution control strategy is to control pollutant
emissions in the most equitable manner possible, i.e., source reduction
equated to source contribution. The quantitative evaluation of source contri-
bution is the crux of the entire photochemical oxidant control issue. The
four different approaches that have been reviewed propose ways in which local
emission control can be determined in terms of how locally measured oxidants
relate to both extraneous and local emissions.
All of the approaches except III focus on ways of evaluating the contri-
bution of local and extraneous emissions to the local urban oxidant problem
and obtaining local control of emissions. Approach I requires that all
source areas affecting the receptor be identified, and Approach II separates
all source areas into two categories, local and extraneous. In a sense,
Approach II is the same as I because all source areas other than local sources
are considered collectively and identified as extraneous. Although Approach I
uses the maximum oxidant observed in the receptor as the "cause," it still
proposes to evaluate the individual contributions of each source; therefore,
54
-------
the fraction of maximum oxidant observed is implied as the determination for
local control, as in Approach II.
The fourth approach utilizes the fraction of the maximum oxidant concen-
tration as a means of determining local control but implies that further local
control may be necessary to offset extraneous emissions. This approach dif-
fers from I and IT only in terms of the degree local control must be imple-
mented. It addresses the situation where the local receptor may be above 0.08
ppm, even after local control has been implemented. In terms of the basic
design of determining the degree of local emission control, Approaches I, II,
and IV are similar except that Approach IV recommends additional local control
to offset the effect of extraneous emissions.
Approach III is very different from all other approaches proposed because
it treats extraneous emissions in a qualitative manner. The maximum oxidant
observed in the receptor area is used to determine local control of emissions
to achieve the standard. The argument presented supporting this approach is
that although extraneous emissions are recognized as a factor, local emissions
still contribute to a downwind problem. Even considering the fact that local
emissions do add to downwind concentrations, Approach III cannot be considered
as the optimum oxidant strategy for the following reasons:
• Recent scientific evidence has clearly demonstrated that extraneous
emissions are an important factor in control strategy development.
• The quantitative effect of extraneous emissions is eliminated as
a decision-making tool to local hydrocarbon control.
• Although local emission control may have to go beyond the amount
necessary for control to attain 0.08 ppm, the quantitative effect
of local emissions is eliminated as a decision-making tool to
local hydrocarbon control.
• In most cases, use of this method will result in an inequitable
degree of local emission control.
Of the three control strategies remaining, Approach IV provides the best
method for quantifying the degree of local hydrocarbon control necessary be-
55
-------
cause both Approach I and II have obvious disadvantages. Approach I relies on
many theoretical assumptions to identify the effects of individual source
areas on the maximum oxidant. These assumptions introduce too many variables
on which to quantify local hydrocarbon reduction. The second approach, al-
though basically sound, only considers the control of local emissions to
reduce the fraction of maximum oxidant observed to 0.08 ppm. Calculating
control requirements to reduce the fraction of the oxidant down to 0.08 ppm
obviously will not solve the problem.
The control strategy proposed in Approach IV has been selected for the
following reasons:
The effect of extraneous emissions and local emissions using
actual ozone monitoring data is quantified.
The maximum observed fractional oxidant concentration is used to
quantify the local hydrocarbon reduction necessary.
Control of local emissions beyond that required to reduce the
local oxidant fraction is recognized.
Although Approach IV is considered the best of the four reviewed, it is
not recommended as the optimum oxidant strategy because problems associated
with quantifying local hydrocarbon control still exist. The issue of esti-
mating the fractional contributions of ozone from local and extraneous emis-
sions is unresolved, and there is no quantitative method to determine the
additional fraction of hydrocarbon control necessary to create an offset.
Therefore, certain modifications are proposed for consideration to be
included in Approach IV as the recommended optimum urban oxidant control
strategy. The modifications are:
Hydrocarbon monitoring should be considered at locations upwind and
downwind of urban complexes so as to relate increases of ozone
downwind of cities to increases in overall hydrocarbons downwind of
cities. This monitoring may clarify whether the elevated ozone
concentrations downwind of cities are due to local emissions
or to a photochemical peaking of extraneous emissions.
56
-------
• A prescribed oxidant goal of 0.06 ppm be set in urban areas so
that advected air arriving at the upwind urban monitor of an
adjacent urban area will be cleaner than standards.
• No local hydrocarbon control for an urban area be required if the
fraction of oxidants due to local emissions is determined to be
less than 25% of the oxidant standard.
The decision to include these modifications to Approach IV was based upon
evidence presented both in the field and lab data reviewed and also upon our
judgment of how an equitable allocation of hydrocarbon emission reduction can
be achieved. The proposed hydrocarbon monitoring was recommended after ex-
amining the hydrocarbon and ozone data obtained from surface ground measure-
ments and aircraft upwind and downwind of Columbus, Ohio, in the 1974 summer
ozone study (5). The data indicates that increases in hydrocarbon concen-
trations downwind of the city can be qualitatively correlated with the in-
creases in ozone. Of particular importance is the fact that the ozone peak
occurred farther downwind than the hydrocarbon peak. The small number of
experiments of this type, though, limits the quantitative relationship that
could be derived. Longer term evaluation of this type of data might be
obtained from the monitoring program suggested for inclusion in Approach IV.
The data obtained may be used to evaluate the individual contributions to the
fraction of maximum oxidant measured across the city. A suggested approach
would be to quantify the oxidant and hydrocarbon fraction across the urban
area.
If both the hydrocarbon and ozone data show significant increases down-
wind of the urban area, then the local emissions may be the predominant con-
tributor. In contrast, if ozone increases occur downwind of the city and
hydrocarbons show no significant increases, the oxidants measured downwind are
probably due to peaking of the extraneous upwind emissions. Using criteria
developed from the monitoring program, the term "significant" can be quanti-
fied so as to establish meaningful interpretations of the data.
The second and third modifications suggested are very closely interrelated.
It has been demonstrated to some degree by the scientific evidence presented
that extraneous emissions from urban areas may affect the air quality of
57
-------
others. The lab data particularly has implied that either carryover ozone, or
emissions, or both may help to provide areas downwind with high minimum con-
centrations on which local emissions could build high maximums. This implies
that the control of oxidants by individual urban areas to obtain the standard
may prevent the standards from being attained because of the additive effects
that may occur when urban areas are situated close to each other. Therefore,
an oxidant goal of 0.06 ppm is suggested so that advected air cleaner than the
standard arrives at the upwind urban site. This offset then would allow a
smaller city, as alluded to in the third approach, to contribute 0.02 ppm and
still prevent the oxidant standard from being violated. One may argue, though,
that the air advected from the smaller city is providing air to another urban
area at the standard. Is it, though, an equitable allocation of emission
reduction to require the smaller city to reduce its local emissions totally to
prevent an air quality violation, whereas the larger urban area is contri-
buting a larger share of the problem? The requirement for no local control
below 0.02 ppm was also suggested to account for the fact that there do exist
limitations in the monitoring of ozone itself.
The modifications suggested in Approach IV are recommended to accomplish
two purposes:
• To better quantify the contribution of local emissions so as to
calculate local hydrocarbon reduction.
• To achieve an equitable allocation of hydrocarbon emission reduction.
The optimum urban oxidant strategy recommended is, therefore, Approach IV
including the modifications suggested.
Recommendations
As discussed in the preceding section, Approach IV and the suggested
modifications to it are recommended as the optimum urban oxidant control
strategy. These proposed modifications recommended that hydrocarbon data be
obtained from monitoring sites upwind and downwind of cities, that a pre-
scribed oxidant offset be quantified to account for extraneous upwind emis-
58
-------
sions, and that no local control of emissions be required if the local oxidant
fraction is less than 25% of the national standard or 0.02 ppm.
Of the three modifications suggested, the proposal to monitor hydro-
carbons may impose some difficulties at both the state and national level. No
long-term hydrocarbon data are available for the majority of urban complexes
in the United States. It would, therefore, be at least 2 years before hydro-
carbon data could be utilized in a meaningful program of quantifying local
emission reduction. In the interim period, though, quantifying the local
oxidant fraction could be done with the upwind and downwind oxidant receptors.
The recent scientific evidence presented in lab and field studies, which were
reviewed in this report, qualitatively support the urban oxidant strategy that
was recommended. The scientific basis for the support of this strategy or any
other strategy, though, is very marginal. Additional research in the forth-
coming years should be focused upon accumulating data that can be practically
applied to the development and support of state and national level oxidant
strategies. As recommended in the subsection on Recommendations — Research,
the likely support and acceptance of the strategies by state and local air
pollution control agencies is dependent upon strong and accepted scientific
evidence, particularly where the application of controls is not very cost-
effective.
59
-------
REVIEW AND ANALYSIS
Frank Spuhler and Ken Waid
An optimum oxidant control strategy must recognize the photochemical
mechanisms that produce ozone since ozone is the pollutant used to determine
compliance with the national ambient air quality standard. First, it is
important to note that the input reactants (precursors) into the system
include: hydrocarbons (HC) (i.e., all organics, derivatives and specie as
intermediate products), nitrogen oxides (NO ), carbon monoxide (CO), water
X
vapor, and sulfur dioxide (SO ). For the most part focus has been only on HC
(organics), leaving out the role of the other precursors. Carbon monoxide
and nitrogen oxides produce ozone even when HCs are not present.
Of greater importance is recognition of the basic NO -NO-O cycle that
results from photolysis and establishes an equilibrium concentration of ozone
in the absence of any other precursor emissions. This is stated as follows:
[NO ]
[03] - (0.021 ppm) —-
Figure 1 illustrates this cycle. The dominant factor in these systems is the
ratio of nitrogen dioxide to nitric oxide. The challenge, then, is to explain
the conversion of nitric oxide to nitrogen dioxide. Once that is done, the
ozone concentration will follow the [NO_]/[NO] ratio (51).
The oxidizable pollutants — such as hydrocarbons, aldehydes,
and carbon monoxide — serve the function of regenerating free radi-
cals that will react with the oxygen in the air to form alkylperoxy
and hydroperoxy. Thus, these oxidizable pollutants can be thought
of as pumping the nitric oxide to nitrogen dioxide. In the process,
they become degraded to other compounds, some of which are still
reactive (e.g., formaldehyde, CH2O, and others). The amount of
61
-------
No HC
Basic Cycle
NO,/*—
E*'(^o.
NO
0 *-0 + O_ + M-
Eq
:\
[O ] in all systems in
toward the equilibrium
condition
(Accumulation)
(a) Low NMHC/NO Ratios
T2 —^ Eq.(l)
*-O + M (Accumulation)
(b) Moderately High NMHC/NO Ratios
Eq.(9) HC
I
NO.
(competes with
reaction (2))
Eq.(4)
Transport HC
2
Transport
,.,/3
(Accumulation limited
by NO /NO ratio)
Eq.(9)
(c) Very_ High NMHC/NO Ratios
.—AXVX — • iw x"
(competes with
reaction (2))
-* -Eq. (4)—-NO
Eq.(1)
HC
j Eq.(10)
RH-»-Oxidation
Product
(Accumulation stopped - and
possibly even reversed
by available NO and excess
organics which scavenge O )
Figure 1.
Schematic relationships among oxidant and precursors under low,
moderate, and high NMHC/NO ratios (52).
62
-------
pumping that can be done, and thus the amount of photochemical oxi-
dant formed, depends on both the reactivity of the oxidizable pol-
lutant and its concentration in a nonlinear way. As the oxidant concen-
tration builds up, the probability of ozone's reacting with hydro-
carbons and various free radicals increases, and the rate of ozone
accumulation decreases. Therefore, the major primary pollutants of
importance to oxidant formation are nitric oxide, hydrocarbons, alde-
hydes, and carbon monoxide. A few free radicals are formed by photo-
lysis of aldehydes and nitrous acid by sunlight or by the reaction
of traces of ozone with reactive hydrocarbons. These free radicals
initiate chain reactions involving hydroperoxy and alkylperoxy radicals.
During these chain reactions, the nitric oxide is converted to nitrogen
dioxide, and the hydrocarbons and aldehydes are degraded. The photo-
lysis of nitrogen dioxide by sunlight forms a free oxygen atom, which
combines with an oxygen molecule to form ozone. Because of the NO-NO -O
cycle, the ozone concentration is determined primarily by the ratio
[N0_]/[NO] and so does not become large until most of the nitric oxide
£1
has been converted to nitrogen dioxide. The total amount of oxidant
formed depends, in a nonlinear fashion, on the amount of hydrocarbons
available to continue pumping the nitric oxide to nitrogen dioxide.
Aldehydes and even carbon monoxide can also serve this pumping function.
When some of the peroxy radicals recombine or react with the nitrogen
oxides, many secondary products, such as hydrogen peroxide and PAN,
are formed (52).
From these considerations, one concludes that present oxidant control stra-
tegies have not done a very good job in following the basic chemical laws
involved in this dynamic mechanism. We have not included all of the reactants
(precursors) and have oversimplified the applications that might indicate
where the harmful dosage level of oxidant concentrations can be expected. EPA
has stated that reductions of carbon compound emissions have reduced ozone
emissions; however, in several of these cases there has been an increase of
nitrogen oxides. The nitrogen oxide emission increases were not considered.
It is, therefore, necessary to reexamine the entire oxidant issue to determine
a realistic control strategy.
Although the photochemical oxidant pollution problem ha.s been recognized
for several decades, the nature, extent, and magnitude of the problem have
been poorly defined. The data base is meager. We treat only part of the
precursors. We leave out transport, mixing, and all that they imply. We have
only spotty sampling data. We are not rigorous in our measurements and in
measurements definition. Our knowledge of true health effects is superficial
63
-------
and our correlations are sloppy. Reviews in all areas recommend more research
and better problem definition.
The evidence presented at the International Conference on Oxidant Pollution
and Its Control indicates that hydrocarbon emission control alone is not an
optimum approach to solving the urban oxidant problem.
Some field studies showed HC control to have a strong beneficial effect,
others did not show a detectable effect, and, others showed effects varying
with attendant NO emission change. Smog chamber studies showed that for that
X.
portion of oxidant formed from the day's local emissions, control of HC is
beneficial except for atmospheres with extremely high — ordinarily not ob-
served — hydrocarbon-to-NO ratios; for such atmospheres HC control, unless
X
drastic, will have very little effect. For that portion of oxidant formed
through multiday irradiation of emissions, the evidence is considered scant
and inconclusive by some investigators but conclusive by others. Those who
feel that the evidence is conclusive claim that HC control will have small
effects upon multiday, irradiated air masses, relative to the effect upon
single day, irradiated air. Control of HC emissions upwind, again, has little
effect according to some investigators, undetermined effect by others. While
it is certain that the effectiveness of the HC control approach is different
for different localities, the judgment called for is to recognize that dif-
ferent areas have different concentrations of the precursors for oxidant
formation, and the same control strategy applied everywhere will not give
consistent results.
NO is a precursor to oxidant formation and should be controlled. The
data base and the present research are meager, and control strategies must ad-
dress all precursors to oxidant formation.
The optimum control strategy for the oxidant problem should be application
of Reasonably Available Control Technology (RACT) to both carbon compound
emissions and NO emissions. The control strategy should state that there are
no techniques at the present time to quantify the effects of the reductions.
64
-------
The control strategy should not require any attempt to show a quantitative
relationship between ambient oxidant concentrations and emission rates or a
definitive relationship between source area and receptor area. The overall
conclusion of all research on the oxidant problem today is that the data
necessary to quantify the relationships between sources and receptors does not
exist.
65
-------
REVIEW AND ANALYSIS
Lee E. Jager
Reference is made to the paper entitled "International Conference on
Oxidant Problems: An Analysis of the Evidence/Viewpoints Presented. Part I:
Definition of Key Issues (1)." That paper identifies as one of the key issues
"The Issue of the Optimum Oxidant Control Strategy" and within that issue asks
several specific questions. My discussion will begin by responding to those
specific questions; however, I do not believe that the optimum control strategy
can be identified merely by answering those questions. Therefore, an additional
discussion is also provided.
The first question asked is: Is the qualitative basis of the oxidant
control strategy still sound? That is, is hydrocarbon emission control an
optimum approach to urban oxidant reduction?"
It appears there is adequate reason to believe that in an urban situation
the ambient level of oxidant is a function of the hydrocarbon emissions that
occur from relatively nearby locations. There is also a great deal of evidence
that oxidants from natural causes are significant; in fact, some claim (with
seemingly good reason) that the air quality standards in certain areas may not
be achievable because of naturally occurring oxidants. However, in most urban
situations, it appears logical to conclude that, whatever the level of natu-
rally occurring oxidants may be, local emissions of hydrocarbons are at least
partially responsible for the local oxidant situation. It certainly is no-
where suggested that reduction in hydrocarbons can have anything but a bene-
ficial effect in terms of reducing oxidants — the only disagreement being to
what extent oxidants can be improved.
67
-------
The second questions asks: "Insofar as the urban oxidant problem is
concerned, is NO emission control imperative? desirable? tolerable? un-
X
desirable? intolerable?"
The issue paper notes that "NO control has varying effects depending on
the hydrocarbon to NO ratio of the reacting emissions." This statement ap-
X
pears adequately supported by the data in the various papers given at the
International Conference and would seem to give rise to the argument that, in
at least some situations, NO should not be controlled since it may result in
X
less scavenging of oxidants and, therefore, higher observed ambient concentra-
tions. In fact, I suppose that argument could be expanded to promote the
introduction of additional NO into an urban area to scavenge oxidants.
It appears amply demonstrated that, in addition to hydrocarbons, NO
X
emissions are a substantial cause of observed ambient concentrations of photo-
chemical oxidants. Although there may be theoretical arguments that some
would make to defend against controlling NO , it appears much more reasonable
X
to conclude that control of NO emissions should improve the oxidant situation
X
as looked at from a broader perspective than just the immediate urban environ-
ment where the NO emission is occurring. There appears good reason to be-
X
lieve that the control of the rural oxidant problem will only be achieved
through control of NO emissions that occur, not only in the rural environ-
X
ment, but also in the urban environment. Therefore, if it is necessary to
select one of the suggested answers to the basic question, I would select
"desirable."
The issue paper then addressed four approaches at defining the source-
receptor relationship between oxidants and emissions. Without discussing the
relative merits of any one of these four approaches, or for that matter,
suggesting any other approach, it would appear to me that with all of the
disagreement, uncertainties, and confusion surrounding this issue, that the
optimum control strategy should be developed based on the qualitative aspects
of the source-receptor relationship rather than the quantitative aspects. If
so, given the answer to the first question, maximum reasonable control of
hydrocarbon emissions seems indicated.
f
' 68
-------
It appears that nearly all investigators have at this moment rejected
Appendix J. The heir apparent to the accepted method appears to be the
isopleth method posed by Dimitriades; however, there is sufficient controversy
surrounding that approach not to use it without further documentation.
In my opinion, the current Clean Air Act requires the impossible to be
done. The country will not attain the ambient air quality standard for photo-
chemical oxidants by mid-1977, and there is absolutely nothing EPA nor anyone
else can do to cause it to happen. It is time to tell that to Congress.
It would not be a service to the Congress nor to the American public to
pretend to develop control strategies which will achieve the standards because
of some fictitious new method of describing quantitatively the source-receptor
relationship. Let's be honest with everyone, including ourselves, and admit
that we are dealing with a very difficult problem, that we are going to con-
tinue to attempt to develop the relationship (assuming we ever have the in-
telligence to do so), but meanwhile make every reasonable effort to control
those things we have good cause to believe are involved in the problem and
which we have the national will to control.
The first step in a control strategy should be to obtain relief from the
Congress in amendments to the Clean Air Act to remove the imperatives of
meeting this standard in a specified time. The Clean Air Act could, perhaps,
even be so drastically amended as to reflect the fact that we may be dealing
with such complex relationships that mortal man does not know the answer. In
fact, man may not even be causing some of the problems. Since the majority of
the scientific community appears willing to accept the fact that reducing
hydrocarbons is a positive step in oxidant control, let us then begin with
applying reasonably available control technology to significant existing
hydrocarbon sources. This can be done in a couple different ways. One way is
to rely upon the States to do it; however, because of the transport nature of
the precursor compounds and of oxidants themselves, it is quite likely that
the states in which the problems exist do not have control over the sources of
contaminants causing the problem. Certainly States should be encouraged to
address those aspects of the problem that they are capable of addressing. EPA
69
-------
guidance and encouragement in this area are important. The current Federal
plan to develop Reasonably Available Control Technology (RACT) documents to
give the States guidance as to what type of controls are available for certain
industrial processes will provide many States with the basis for getting into
the business of controlling hydrocarbon sources. There are those, however,
that will not.
If meaningful progress is to be made, EPA must develop a national plan to
deal with hydrocarbon emissions. The first step should be the development of
new source performance standards dealing with all significant stationary
hydrocarbon sources. EPA may also wish to look at Section III(d) of the Clean
Air Act to require States to then adopt controls for existing hydrocarbon
sources (it appears to me that hydrocarbons meet the definiton of a "designated
pollutant").
When EPA feels it has justification to define regional problems, EPA has
the authority to develop regional plans and promulgate regulations affecting
sources within those regions.
EPA should develop a national strategy for transportation controls. Many
of the controls that have been suggested are highly socially disruptive; also,
many come with incredibly high price tags. Components of transportation con-
trol strategies that are "reasonable" should be determined at the Federal
level, preferably by act of Congress. It is not realistic to expect a State
to enact stringent controls affecting its citizens when the benefits to be
derived from those controls are practically impossible to quantify and may not
even be realized within that State. If EPA is correct about the transboundary
nature of the problem (and there appears ample reason to believe that they
are), the national policy should not be to deal with the problem on a State-
by-State basis.
In summary, I believe the following would be a reasonable approach to an
oxidant control strategy until a more rational determination can be made of
the source-receptor relationship of hydrocarbons, NO , and oxidants:
X
70
-------
• Reasonably available control technology for stationary sources
should be defined by the Federal Government and should be implemented
by those States required to do so because of State Implementation
Plan (SIP) deficiencies. EPA should promulgate RACT regulations
for those areas with SIP deficiencies, if the States fail to do so.
• In States where EPA has not determined SIP deficiencies, EPA should
encourage the States to implement RACT requirements when it is
reasonable to conclude that emissions from sources in that State
result in problems outside of the State.
• Where the need for regional (multistate) plans can be documented,
EPA should promulgate such plans with the cooperation of the af-
fected States. This is especially true where large metropolitan
areas are upwind of nonattainment areas, but are not themselves in
a nonattainment area.
• EPA should expedite the promulgation of new source performance
standards for all significant stationary sources of hydrocarbons
and NO .
x
• EPA should investigate the applicability of Section III(d) to
hydrocarbon emissions.
A national policy should be developed on transportation control
strategies. This national policy should consist of a decision,
by Congress, on mandatory components of an approvable Transportation
Control Plan (TCP). Any additional restraints should be left to
State option.
The present effort to develop a cause/effect relationship of emissions
and ambient levels of oxidants should be continued and, in fact,
increased. However, excessive control programs should not be demanded
at this time because of the amount of hydrocarbon control predicted as
necessary by Appendix J or by any of the alternatives yet proposed.
71
-------
REVIEW AND ANALYSIS
George T. Wolff
INTRODUCTION
Since 1971, when the U.S. EPA formulated and legislated into use the
first oxidant control strategy, it has come under sharp criticism from State
air pollution officials, scientific and university groups. The primary
reasons for these criticisms were: a definitive quantitative relationship
between hydrocarbons and oxidants was not established, the role of nitrogen
oxides was not included, transport was not considered, and nonanthropogenic
contributions were ignored.
As a result of these questions, the U.S. EPA, State agencies, industry,
and research institutions undertook extensive research activities within the
past 4 years to reexamine the entire oxidant control problem. The purpose of
this report is to: review these recent studies that have dealt with deter-
mining oxidant-precursor relationships, and attempt to define an optimum
control strategy for oxidants.
The questions that have been addressed are listed below.
Are anthropogenic precursor controls even warranted?
Is a hydrocarbon control strategy by itself sufficient to reduce
oxidants?
Do we also need to apply emission controls on sources that produce
oxides of nitrogen?
In terms of a cause-effect relationship, which cause (source) do
we control: local, upwind, or both?
73
-------
In terms of a cause-effect relationship, which effect do we design
our strategy to eliminate: the local net oxidant production, the
upwind or fossil oxidant, or both?
THE CASE FOR ANTHROPOGENIC PRECURSOR CONTROLS
The oxidant problem in the Eastern United States is associated during the
summertime with high pressure systems which typically originate in Centra.1
Canada, move southeastward over the Midwestern States, and then eastward to
the Atlantic Ocean. Data gathered for the entire 1976 oxidant season by the
"Moodus Data Analysis Task Force" (53) indicate that oxidant (ozone) concen-
trations within these high pressure systems, as they first come out of Canada,
are generally between 30-50 ppb. By the time the high has moved off the
Atlantic coast and the Northeastern States, which are on the backside of the
high, levels approaching 300 ppb are observed.
The increase in the ozone is undoubtedly due to a combination of ozone
synthesis from both anthropogenic and natural precursors as well as strato-
spheric injection. An examination of ozone isopleth maps covering the 20-
state Moodus study area on the initial days of an ozone episode (initial day
is defined as the first day the Canadian high is located in the United States
and contains ozone levels exceeding the NAAQS) indicate that anthropogenic
precursor emissions are the principal cause of the elevated ozone. According
to the U.S. EPA's hydrocarbon emission inventory map (Figure 1), two areas
with extremely high hydrocarbon emission densities can be discerned in the
central and eastern Midwestern States and in the Washington, D. C., to Boston,
Mass., corridor. Figures 2 and 3 show typical initial days of an ozone epi-
sode. It is quite obvious that the areas of elevated ozone originated in the
two high-density emission areas. As the high persists over the United States
for several days, the ozone area in the Midwest generally builds eastward
until the two areas merge. Figures 4 and 5 show typical days after the merging
has taken place. The relationship between the areas of the high ozone and the
high-emission density areas is still evident though.
Recently, Hathorn and Walker (18) proposed that the source of the additional
ozone required to cause an episode in the backside of a high is stratospheric.
74
-------
in
o
•H
w
to
-a
§
.8
S-l
IT)
O
2
T!
PM
75
-------
c
o
•H
4J
3
X)
•H
M
4J
tn
c
o
N
O
o
(0
CM
Q)
76
-------
.1?
1111)
lllll I
'l1
'ill'
1,1,
77
-------
to
o
•H
-P
+J
0)
-H
T)
(U
O
N
O
0)
O
3
to
<*
0)
78
-------
tn
o
•H
4J
XI
•H
•H
T)
(1)
C
O
N
O
Q)
O
W)
in"
0)
M
tn
•H
Cn
79
-------
The Moodus data, which does not discount a stratospheric contribution, indicate
that man-made pollution sources are capable of providing the additional ozone
associated with an episode in the Midwestern and Northeastern United States,
THE CASE FOR NONMETHANE HYDROCARBON CONTROLS
The most significant supporting evidence for a hydrocarbon reduction
strategy as a means of reducing oxidants was shown by Dimitriades (49,54) and
is based largely on the chamber studies and mathematical simulations of Dodge
(10). From these studies, a series of ozone isopleths as a function of initial
NMHC and NO has been developed.
Supporting evidence has been presented by a number of investigators.
Meyer et al. (20) found a significant correlation between locally emitted
hydrocarbons and afternoon ozone values at urban sites and between the pre-
vious morning's hydrocarbon emissions and afternoon's ozone at nonurban and
rural sites. Using a numerical photochemical simulation model, Dabberdt and
Singh (13) concluded that a reduction in hydrocarbons on the first day of an
oxidant episode in the South Coast Air Basin would result in lower peak oxi-
dant values. Data presented by Decker et al. (30) and Wolff et al. (55) have
shown evidence of significant oxidant production downwind of petroleum refineries.
Ambient oxidant trends have also been compared with hydrocarbon emission
trends. In both the South Coast Air Basin (Metropolitan Los Angeles) and the
Bay Area Air Basin (Metropolitan San Francisco), substantial reductions in
hydrocarbon emissions have been achieved since 1964. In the South Coast Basin
where hydrocarbon emissions decreased 18% from 1964-1974, Trijonis et al. (15)
found that the basinwide oxidants decreased 19% over the same period. Simil-
arly, Martinez et al. (16) found significant reductions in the second highest
annual ozone maximum, the average ozone maximum, the third-quarter average
maximum, and the frequency of emissions greater than 0.08 ppm at most sites in
the Bay Area Basin where hydrocarbon emissions were reduced 24% from 1964 to
1974.
80
-------
The evidence presented by Dimitriades (49,54), Dodge (10), Meyer et al.
(20), Dabberdt and Singh (13), Decker et al. (30), and Wolff et al, (55)
indicate the need for hydrocarbon controls. Although the trend analysis
conducted by Trijonis et al. (15) and Martinez et al. (16) indicates decreas-
ing oxidant levels with decreasing hydrocarbon emissions, the results are far
from conclusive because, during the same period (1964-1974), NO emissions
X
increased in the South Coast Air Basin by 36% and in the Bay Area Air Basin by
31%. The possible effects of these increases are discussed in the next section.
THE CASE FOR NO CONTROLS
x
While most investigators are in agreement concerning the need for non-
methane hydrocarbon controls, they are divided on the need for NO controls.
x
The conflicting views have been attributed by Dimitriades (54,56) to the
dual role of nitrogen oxides in the photochemical smog complex. At high
ratios of NMHC/NO (<8-10), the addition of NO increases the maximum observed
x x
O in the chamber studies. At low NMHC/NO ratios (<8-10), an increase in the
•^ X
NO will lower the maximum observed ozone. This is a result of the rapid NO +
X
0 -> NO + O reaction, which becomes extremely important at these low ratios.
As a result, Dimitriades (56) predicts that a reduction in urban NO levels
x
will increase urban ozone, some of which may be transported into downwind
areas and hence increase the downwind or rural oxidant burden. Dabberdt and
Singh (13), on the other hand, feel that any emission reductions, both NO^and
NMHC, may result in a decrease in the total oxidant formation potential that
may not be realized for several days after the reductions have been initiated.
The evidence against NO controls (above those necessary to achieve the
NAAQS for NO ) are summarized.
Meyer et al. (20) found a negative correlation between locally emitted
NO emissions and maximum O_ levels in urban areas. The classical case of
x 3
decreasing NO levels resulting in increasing ozone levels in urban areas is
X
the "Sunday Effect" reported by Cleveland et al. (14) and modeled by Farrow et
al. (57). At several highly urbanized sites in northeastern New Jersey,
81
-------
Cleveland et al. (14) observed that on Sundays, when ambient levels of NO were
significantly lower than on weekdays, the ozone values tended to be slightly
higher. This system was successfully modeled by Farrow et al. (57), who
attributed it to the higher rate of O scavenging by NO and N0? on weekdays.
In their analysis of an air pollution episode in Los Angeles, Dabberdt and
Singh's model (13) predicted a significant increase in ozone downwind of a
power plant, if NO emission controls were used by the power plant.
X
The preceding discussion indicates that NO controls will result in an
X
increase in the ozone levels in many urban areas. An exception to this was
presented by Walker (41) who examined data from urban sites in Texas. His
analysis indicates that the highest ozone values are observed in Texas when
the early morning NMHC are between 5-10 ppm. At values above and below this,
lower ozone maximums were observed. This suggests that NO may be the con-
X,
trolling parameter in Texas. He also states that NMHC/NO ratios in Texas
average from 17.5/1 to 70/1. With these ratios, according to the O -NMHC-NO
relationships developed by Dodge (10) and Dimitriades (54), NO controls would
X
be the most effective approach.
The most important evidence for the use of NO controls appears in the
X
data presented by Dimitriades (54) and Dodge (10). As mentioned above, it is
apparent from their data that maximum ozone is obtained at a NMHC/NO ratio
between 8/1 and 10/1. This is also the transition point where the role re-
versal of NO occurs. At ratio's greater than 8/1-10/1, the effect of NO
X X
reductions on ozone reductions become greater than a proportional reduction in
NMHC reductions. In a recent report (52), EPA summarizes ambient NMHC/NO
_ __ _ "
ratios obtained by various contractors. (This is shown in Table 1.) Of the 10
urban sites where these measurements have been made, 8 out of 10 have ratios
greater than or equal to 8/1. Ten out of 10 suburban sites and 6 out of 6
rural sites also have ratios greater than 8/1. As a result, most of the sites
would experience a greater ozone reduction if NO controls were employed. For
X
those sites close to 8/1, the maximum benefit would be achieved with both NO
X
and HC controls. The same EPA report (52) shows the NMHC/NO -0 relationships
obtained from chamber studies using "aged precursors." These studies indicate
82
------- An error occurred while trying to OCR this image.
-------
that the critical ratio for NO controls to be proportionately more effective
X
than NMHC controls is as high as 30/1.
Additional support for NO controls have been presented by Meyer et al.
(20), Van Ham and Niebor (21), Farrow et al. (57) and Wight (unpublished
data).
In their empirical correlation analysis, Meyer et al. (20) found a
significant positive correlation between early morning upwind NO emissions on
X
the previous day and afternoon ozone values at rural sites. Although a
similar relationship was found with the previous day's hydrocarbon emissions,
the extremely high NMHC/NO ratio observed at rural sites (>100/1) would
X
preclude any benefit of implementing only NMHC controls in the upwind urban
areas.
The experimental work of Van Ham and Niebor (21) also support the
implementation of NO controls. Spiking ambient air samples collected in bags
X
and exposing them to sunlight with NO , they found that increasing the NO not
X X
only increases the maximum ozone, but it also increases the time of irradiation
required to reach the ozone maximum. Initially, the NO acts as an ozone
X
scavenger, but at some point in time, which is a function of the NO con-
X
centration and the NMHC/NO ratio, the role of NO is reversed (as Dimitriades
XL „ . _. . _ X
suggested). At this point, the NO , which is mostly NO, at this time, begins
X ^
to become an increasingly effective supplier of the odd oxygen atom that is
required for ozone production. In terms of the real atmosphere, increasing
the NO concentration delays the formation of the ozone maximum (this explains
X
why urban 0 values decrease with increasing local NO emissions) but in-
J X
creases the ozone maximum, which may not be observed for many hours or many
miles downwind. This is consistent with the observations of Meyer et al. (20)
and the suggestion of Dabberdt and Singh (13) that reduction of all precursors
will reduce the air pollution potential.
According to the model of Farrow et al. (57), the net increase in NO
X
flux due to higher emissions on weekdays far outweigh the additional O flux
caused by decreased emissions on Sundays. If this is in fact true, the
84
-------
"Sunday Effect" should not be observed at sites downwind of northeastern New
Jersey. Unpublished data gathered by Wight using the same computational
techniques of Cleveland et al. (14) show that Greenwich, Conn., does not
experience a Sunday Effect (see Figure 6). In fact, the weekday values appear
to be substantially higher.
There is some evidence that the zone of ozone depletion near large
sources of NO emissions is quite limited, and ozone detectors located near
X
such sources would differ widely from monitors located a few blocks away.
Since most urban monitors are located in high density traffic areas (this is
especially true in northeastern New Jersey where the "Sunday Effect" origi-
nated) , they may only record a relatively small scale phenomenon. In 1974,
Wolff et al. (58) obtained ozone measurements at 1,000 ft over several ozone
monitoring sites and compared them with the surface levels in mid-afternoon.
At 1,000 ft, there was no noticeable O depression over the urban centers. In
fact, the 1,000-ft ozone levels were extremely similar over most of northern
New Jersey. At rural sites and at urban sites where the monitor was located
away from a high traffic density area, there was extremely good agreement
between the surface and 1,000-ft ozone measurements. Over the high traffic
density area monitors, which incidentally were the sites used by Cleveland et
al. (14) to demonstrate the "Sunday Effect," surface ozone values were between
40-50% of the ozone observed at the 1,000-ft level.
Additional evidence of the relatively small-scale impact of ozone
depletion caused by local NO emissions was presented by Wolff (59). The New
X
Jersey Department of Environmental Protection operated an ozone monitor on
State Street in Trenton, New Jersey, during 1973 and 1974. In 1973, State
Street was a heavily traveled vehicular thoroughfare. In early 1974, the
street was blocked off to vehicular traffic and made into a pedestrian mall.
There is no evidence that this decreased traffic in Trenton — it merely re-
distributed it to surrounding streets. Although no NO measurements were
X
made, CO measurements, which should represent the relative NO changes, were
made. Figure 7 shows that the average daily maximum CO values decreased 38%,
while the average daily maximum ozone values increased 60%. This increase in
ozone could not be explained by meteorological conditions since most New
85
-------
0.16
0.14
"I
< "0.12
X
<
M
O
0.10
0.08
I? 0.06
-------
CO IN AUGUST
C9
4
ec
Ul
0
0.08
0.07
0.06
0.05
0.04
03 IN AUGUST
o
u
C9
cc
> 0.03
0.02
0.01
1234567
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
DST
Figure 7. Trenton, New Jersey.
87
-------
Jersey sites experienced lower 0 in August 1974 than in August 1973. As a
result, it is felt that the impact of ozone destruction by NO (role #1) is
X
very limited in geographic scale and that the more important effect is to
increase the ozone values short distances away from the NO sources.
x
As mentioned previously, the trend analyses data presented by Martinez et
al. (16) and Trijonis et al. (15) are inconclusive with respect to the cause
of the oxidant reductions in both the South Coast Air Basin and the Bay Area
Air Basin. In the two areas, ozone reductions corresponded to hydrocarbon
reductions of 18% and 24% and NO increases of 36% and 31%, respectively.
X
According to the preceding discussion, these NO increases alone should be
X
enough to account for the observed ozone decreases at the highly urbanized
sites. To test this hypothesis, if the urban ozone reductions were due to NO
X
increases, then according to the previous discussions, the ozone levels should
have increased in downwind areas. On close examination of the data of
Trijonis et al. (15), it is apparent that the sites which did record ozone
reductions were confined to the densely populated counties of Los Angeles and
Orange. The lesser densely populated counties of San Bernadino and Riverside,
which are located downwind (to the east) of Los Angeles and Orange, do in fact
report an average ozone increase of 8% over the same period (1964-1974).
Similarly, a close examination of the data of Martinez et al. (16) indicates
that the ozone reductions in the Bay Area Air Basin are generally confined to
the eastern sections (the sites within and closest to San Francisco and
Oakland), while increases in ozone levels have been observed at two out of the
three easternmost (downwind) sites. Martinez et al. (16) also examined the
trends at four other sites located in less densely populated areas of Califor-
nia. Three out of four sites showed an increasing ozone concentration trend.
Based on the discussions in this section, it is concluded that reductions
in NO emissions in urban areas will increase the ozone values in a relatively
small area in the proximity of and downwind of the reduced sources. These
same reductions will decrease the total ozone-forming potential of the air and
result in decreased ozone levels at downwind and rural sites. It has also
been suggested by Farrow et al. (57) that a reduction in NO emissions will
X.
88
-------
reduce the organic and inorganic nitrates and reduce the potential of the
atmosphere to cause lachrymation.
DEVELOPMENT OF A CAUSE AND EFFECT OXIDANT REDUCTION STRATEGY
It is widely accepted now that ozone is capable of being transported long
distances (60). Urban plumes are distinguishable for nearly 100 miles down-
wind. Beyond this point, urban plumes merge and disperse, and this results in
a relatively uniform ozone concentration extending for hundreds of miles
within a particular air mass. During the day and at night above the nocturnal
inversion layer, the ozone is somewhat conserved. Based on vertical ozone
profile data, Sickles et al. (12) estimated the half-life to be 29.5 hr. Data
from the DaVinci II experiment suggested a half-life between 16 and 34 hr. An
extrapolation of Stasiuk and Coffey's (19) mean diurnal ozone profile at
Whiteface Mt. yields a half-life of the order of 23 hr.
As a result of this relatively long half-life of ozone, local control
strategies must deal not only with local emissions but also ozone transported
into the area and the impact that the ozone produced by their local precursor
emissions has on downwind cities. Transported ozone entering an urban area
has been observed to be as high as 0.120 ppm in the Northeastern United States
(61). Huffman et al. (43) observed ozone levels as high as 0.18 ppm being
transported into a major Midwestern city. It appears that in the Eastern
United states, high levels of ozone produced by one area's emissions are
conserved and become some other downwind city's problem the next day. Con-
sequently, control strategies must be designed to reduce the total ozone-
forming potential.
An analogous situation exists in Southern California. In this case,
however, the ozone is generally confined to the South Coast Air Basin because
of its unique topographical features. The first day's spent emissions and
generated ozone are conserved aloft, and these add to the second day's problem
when the shallow nocturnal inversion is broken. Evidence supporting this has
been presented by Dabberdt and Singh (13) and Johnson and Singh (62). Con-
89
-------
sequently, control strategies in Southern California must also be designed to
reduce the total ozone-forming potential.
Very limited data have been obtained to evaluate the effect of NMHC and
NO concentrations on ozone production during multiday periods. As mentioned
X
previously, Meyer et al. (20) found a positive relationship (significant at
the 0.10 significance level) between rural 0. levels and NO emissions 24
j «C
hours upwind. A weaker relationship was found for rural 03 and NMHC emissions
30 hours upwind.
A few multiday smog chamber studies have been conducted, and they have
been discussed by Dimitriades (56) and Sickles et al. (12). Although the data
base is extremely limited, several observations relating to this discussion
have been made.
In an attempt to simulate the real atmosphere, outdoor chambers were
used. Sunlight was the only source of photons, and clean air was introduced
at a constant rate to simulate atmospheric dispersion processes. The results
clearly indicated that the dilution of ozone is not proportional to the dilu-
tion of the initial chamber air. The dilution of the ozone is substantially
less (12,56). This supports the previously cited conclusion that at low NO
concentrations, each NO molecule becomes more efficient in producing ozone
X
(12).
A few multiday irradiations were conducted with different initial NMHC
and NO concentrations. In one set of experiments, initial NMHC were depleted
X
approximately 65%, while NO remained the same. During day 1, in the chamber
X
with the reduced NMHC, ozone levels were reduced approximately 75%. On day 2,
in the chamber with the uncontrolled NMHC, the ozone reached about the same
maximum as it did on day 1. In the controlled chamber, ozone values nearly
quadrupled from day 1 and reached a maximum of about 90% of ozone in the un-
controlled chamber. It appears that the effect of decreasing the NMHC and the
NMHC/NOx ratio (from about 7.5 to 3) was to delay the formation of the ozone.
Unfortunately, since this experiment ended on the second day, no conclusions
90
-------
can be made concerning the effect of the NMHC reduction on the total ozone-
forming potential.
Additional 3-day experiments conducted by Sickles et al. (12) with
different initial precursor concentrations have indicated that there appears
to be a general trend for lower ozone maximum of all 3 days when both initial
NMHC and NO concentrations are reduced. In addition, day 2 and day 3 ozone
X
maxima decrease with decreasing early morning NO and NMHC on subsequent days.
X
The data obtained in these multiday chamber studies appear to be con-
sistent with the observations of Meyer et al. (20) using ambient data and with
the data presented in the following subsection of this report. The most ef-
fective way of reducing ozone is to reduce both NMHC and NO . Unfortunately,
X
the corresponding reduction in ozone will be substantially less than propor-
tional.
Other chamber studies relating to oxidant control strategies were
conducted by Jeffries et al. (9), who studied the effect of continuous emis-
sions on oxidant precursor relationships. Two simultaneous experiments were
conducted. In one chamber, all of the NMHC and NO were injected into the
X
chamber at approximately 6 a.m. In the second chamber, the same total mass
of precursors was added but ramp-injected over the 0600- to 1800-hr, period.
The results showed that the ramp side usually forms 0 faster and is higher in
O until the afternoon when the static side usually overtakes and slightly
surpasses the ramped side. The total dosage of O during the day in both
chambers was nearly identical. This indicates that the 6 a.m.-9 a.m. emis-
sions, which control strategies have been exclusively directed toward, are not
the only emissions important in O formation. On the contrary, the results
indicate that controls should be applied to emissions throughout the day. In
addition, this study indicates that the additive effect of nonurban emissions
could be important and they should also be included in any control strategy.
91
-------
SUMMARY AND CONCLUSIONS
Anthropogenic emissions of nonmethane hydrocarbons (NMHC) and nitrogen
oxides (NO ) are responsible for a substantial portion of the high oxidant
(ozone) concentrations observed at both urban and rural locations in the
United States.
Studies indicate that hydrocarbon reductions will result in greater
oxidant reductions in urban than in rural areas. This is a result of a de-
crease in the NMHC/NO ratio, which would make NO a more effective O
X X -3
scavenger in urban areas.
Reductions in NO in urban areas will result in local increases in ozone
within the urban area, but the effect will be to reduce the ozone in downwind
and rural areas.
It has been shown that ozone levels are generally lower on weekdays at
sites with high weekday traffic densities. On Sundays with lower NO con-
X
centrations, the rate of ozone scavenging is reduced, and consequently the
measured ozone is higher. Downwind on Sundays, however, lower ozone levels
are observed.
Because of long-range transport, it is not enough for any urban area to
develop a control strategy designed to achieve the NAAQS for oxidants in their
area without any concern for those areas downwind. Controls must be placed on
upwind sources to assure that the upwind source area meets the NAAQS and to
allow the downwind areas to meet the NAAQS by applying their own reasonable
local controls. Multiday chamber studies to date have not provided any
quantitative evidence demonstrating the degree of control required for this
approach. It is evident, however, that severe reductions of both NO and NMHC
X
are required. Consequently, until such quantitative information is available,
the most feasible short-term approach appears to be a modification of that
adopted by the "Moodus" States (63). Their approach is to control hydro-
carbons with the best available control technology which incorporates advances
.' 92
-------
in the state of the art. The required modification is to extend the "Moodus"
approach to include the control of nitrogen oxides.
Finally, evidence has been presented that indicates that the control of
only 6 a.m.-9 a.m. precursor emissions is inadequate. Instead, control of all
emissions throughout the day is needed. This applies to urban as well as
nonurban sources.
93
-------
REFERENCES
1. Dimitriades, B., and A.P. Altshuller. International Conference on
Oxidant Problems: Analysis of the Evidence/Viewpoints Presented.
Part I: Definition of Key Issues. JAPCA, 27(4):229-307, 1977.
2. Air Quality Criteria for Photochemical Oxidants. AP-63. U.S.
Public Health Service, Washington, D.C., 1970.
3. Research Triangle Institute. Investigation of Ozone and Ozone
Precursor Concentrations at Nonurban Locations in the Eastern
United States. EPA-450/3-74-034. Environmental Protection Agency,
Research Triangle Park, N.C., 1974.
4. Meteorology Research, Inc. Determination of the Feasibility of the
Long-Range Transport of Ozone or Ozone Precursors. EPA-450/3-74-
061. Environmental Protection Agency, Research Triangle Park,
N.C., 1974.
5. 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. 385 pp.
6. Spicer, C.W., J.L. Gemma, D.W. Joseph, P.R. Sticksel, and G.F.
Ward. The Transport of Oxidants Beyond Urban Areas. EPA-600/3-76-
018. Environmental Protection Agency, Research Triangle Park,
N.C., 1976.
95
-------
7. Research Triangle Institute. Formation and Transport of Oxidants
along Gulf Coast and Northern U.S. EPA-450/3-76-033. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1976.
8. Dimitriades, B. On the Function of Hydrocarbon and Nitrogen Oxides
in Photochemical-Smog Formation. Bureau of Mines Report of Investi-
gations 7433. U.S. Department of Interior, Washington, D.C., 1970.
37 pp.
9. Jeffries, H.E., R. Kamens, D.L. Fox, and B. Dimitriades. Outdoor
Smog Chamber Studies: Effect of Diurnal Light, Dilution, and
Continuous Emission on Oxidant Precursor Relationships. Int. Conf.
Ox. Poll., Proc. 2:891-902. EPA-600/3-77-001b. Environmental Pro-
tection Agency, Research Triangle Park, N.C., 1977.
10. Dodge, M.C. Combined Use of Modeling Techniques and Smog Chamber
Data to Derive Ozone-Precursor Relationships. Int. Conf. Ox.
Poll., Proc. 2:881-889. EPA-600/3-77-001b. Environmental Pro-
tection Agency, Research Triangle Park, N.C., 1977.
11. Air Quality Criteria for Nitrogen Oxides. AP-84. Environmental
Protection Agency, Washington, D.C. , 1971.
12. Sickles, J.E., II, L.A. Ripperton, and W.C. Eaton. Oxidant and
Precursor Transport Simulation Studies in the Research Triangle
Institute Smog Chambers. Int. Conf. Ox. Poll., Proc. 1:319-328.
EPA-600/3-77-OOla. Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
13. Dabberdt, W.F., and H.B. Singh. A Preliminary Investigation of the
Effectiveness of Air Pollution Emergency Plans. Int. Conf. Ox.
Poll., Proc. 2:837-848. EPA-600/3-77-001b. Environmental Pro-
tection Agency, Research Triangle Park, N.C., 1977.
96
-------
14. Cleveland, W.S., T.E. Graedel, B. Kleiner, and J.L. Warner. Sunday
and Workday Variations in Photochemical Air Pollutants in New
Jersey and New York. Science, 186:1037-1038, 1974.
15. Trijonis, J., T. Peng, G. McRae, and L. Lees. Oxidants and Pre-
cursor Trends in the Metropolitan Los Angeles Region. Int. Conf.
Ox. Poll., Proc. 2:1077-1093. EPA-600/3-77-001b. Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
16. Martinez. E.L., N.C. Possiel, E.L. Meyer, L.G. Wayne, K.W. Wilson,
and C.L. Boyd. Trends in Ambient Levels of Oxidant and Their
Possible Underlying Explanations. Int. Conf. Ox. Poll., Proc.
2:1103-1112. EPA-600/3-77001b. Environmental Protection Agency,
Research Triangle Park, N.C., 1977.
17. Altshuller, A.P. Evaluation of Oxidant Results at CAMP Sites in
the United States. JAPCA, 25(l):19-24, 1975.
18. Hathorn, J.W., III, and H.M. Walker. A "Texas-Size" Ozone Episode
Tracked to Its Source. Int. Conf. Ox. Poll., Proc. 1:353-380.
EPA-600/3-77-001a. Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
19. Stasiuk, W.N., Jr., and P.E. Coffey. Rural and Urban Ozone Rela-
tionships in New York State. JAPCA, 24(6):564-568, 1974.
20. Meyer, E.L., Jr., W.D. Freas, III, J.E. Summerhays, and P.L.
Youngblood. Use of Trajectory Analysis for Determining Empirical
Relationships Among Ambient Ozone Levels and Meteorological and
Emission Variables. Int. Conf. Ox. Poll., Proc. 2:903-913. EPA-
600/3-77-OOlb. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
97
-------
21. Van Ham, J., and H. Niebor. Smog Potential of Ambient Air Sampled
at Delft, Netherlands: The Effect of Increasing NO Concentrations.
X
Int. Conf. Ox. Poll., Proc. 2:943-953. EPA-600/3-77-001b. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
22. Black, F. The Impact of Emissions Control Technology on Passenger
Car Hydrocarbon Emissions Rates and Patterns. Int. Conf. Ox.
Poll., Proc. 2:1053-1067. EPA-600/3-77-001b. Environmental Pro-
tection Agency, Research Triangle Park, N.C., 1977.
23. Walsh, R.T. Control Regulations for Stationary Sources of Hydro-
carbons in the United States. Int. Conf. Ox. Poll., Proc. 2:1155-
1165. EPA-600/3-77-001b. Environmental Protection Agency, Re-
search Triangle Park, 1977.
24. Bonta, W.K., and J.W. Paisie. Problems with Converting State-of-
the-Art Photochemistry to State Level Control Strategies. Int.
Conf. Ox. Poll., Proc. 2:1123-1133. EPA-600/3-77-001b. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
25. Carney, T.B. Evidence of the Role of Stratospheric Transport in
the Distribution of Tropospheric Ozone. In: Ozone/Oxidants -
Interactions with the Total Environment. APCA Specialty Conference
(Southwest Section), Proceedings. Air Pollution Control Asso-
ciation, Pittsburgh, Pa., 1976. pp. 234-241.
26. Sticksel, P.R. Occurrence and Movement of Tropospheric Ozone
Maxima. In: Ozone/Oxidants - Interactions with the Total En-
vironment. APCA Speciality Conference (Southwest Section), Pro-
ceedings. Air Pollution Control Association, Pittsburgh, Pa.,
1976. pp. 252-267.
98
-------
27. chatfield, R., and H. Harrison. Ozone in the Remote Troposphere:
Higher Levels? In: Ozone/Oxidants - Interactions with the Total
Environment. APCA Speciality Conference (Southwest Section),
Proceedings. Air Pollution Control Association, Pittsburgh, Pa.,
1976. pp. 77-83.
28. Chatfield, R., and R.A. Rasmussen. An Assessment of the Conti-
nental Lower Troposheric Ozone Budget. Int. Conf. Ox. Poll., Proc.
1:121-136. EPA-600/3-77-001a. Environmental Protection Agency,
Research Triangle Park, N.C., 1977.
29. Bach, W.D., Jr. Analysis and Interpretation of Serial Ozonesonde
Releases. In: Ozone/Oxidants - Interactions with the Total Environ-
ment. APCA Specialty Conference (Southwest Section), Proceedings.
Air Polution Control Association, Pittsburgh. Pa., 1976. pp. 96-
108.
30. Decker, C.E., L.A. Ripperton, J.J.B. Worth, F.M. Vukovich, W.D.
Bach, Jr., J.B. Tommerdahl, F. Smith, and D.E. Wagoner. Formation
and Transport of Oxidants along Gulf Coast and Northern U.S. EPA-
450/3-76-033. Environmental Protection Agency, Research Triangle
Park, N.C., 1976.
31. Jones, D.C., and M. LaHue. Background Ozone Concentration in
Western Colorado. In: Ozone/Oxidants - Interactions with the
Total Environment. APCA Speciality Conference (Southwest Section),
Proceedings. Air Pollution Control Association, Pittsburgh, Pa.,
1976. pp. 121-132.
32. Davis, D.R., and R.E. Jensen. Low Level Ozone and Weather Systems.
In: Ozone/Oxidants - Interactions with the Total Environment.
APCA Speciality Conference (Southwest Section), Proceedings. Air
Pollution Control Association, Pittsburgh, Pa., 1976. pp. 242-251.
99
-------
33. Rasmussen, R. What Do the Hydrocarbons from Trees Contribute to
Air Pollution? JAPCA, 22(7):537-543, 1972.
34. Robinson, E., and R.A. Rasmussen. Identification of Natural and
Anthropogenic Rural Ozone for Control Purposes. In: Ozone/Oxi-
dants - Interaction with the Total Environment. APCA Specialty
Conference (Southwest Section), Proceedings. Air Pollution Control
Association, Pittsburgh, Pa., 1976. pp. 3-13.
35. Gay, B., Jr., and R. Arnts. The Chemistry of Naturally Emitted
Hydrocarbons. Int. Conf. Ox. Poll., Proc. 2:745-751. EPA-600/377-
OOlb. Environmental Protection Agency, Research Triangle Park,
N.C., 1977.
36. Rubino, R.A., L. Bruckman, and J. Magyar. Ozone Transport. JAPCA,
26(10):972-975, 1976.
37. Cleveland, W.S., B. Kleiner, J.E. McRae, and J.L. Warner. Photo-
chemical Air Pollution Transport from the New York City Area into
Connecticut and Massachusetts. Science, 191:179-181, 1976.
38. Cleveland, W.S., and B. Kleiner. Transport of Photochemical Air
Pollution from Camden-Philadephia Urban Complex. Environ. Sci.
Technol., 9(9):869-872, 1975.
39. Lyons, W.A., and H.S. Cole. Photochemical Oxidant Transport:
Mesoscale Lake Breeze and Synoptic-Scale Aspects. J. Appl. Me-
teorol., 15(7):733-743, 1976.
40. 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.
100
-------
41. Walker, H.M. A "J" Relationship for Texas. Int. Conf. Ox. Poll.,
Proc. 2:851-869. EPA-600/3-77-001b. Environmental Protection
Agency, Research Triangle Park, N.C., 1977.
42. Neal, R., R. Severs, L. Wenzel, and K. MacKenzie. Simultaneous
Chemiluminescent Ozone and KI Oxidant Measurements in Houston,
Texas, 1975. In: Ozone/Oxidants - Interaction with the Total En-
vironment. APCA Specialty Conference (Southwest Section), Proceedings.
Air Pollution Control Association. Pittsburgh, Pa., 1976. pp.
180-188.
43. 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. Int. Conf. Ox. Poll., Proc. 1:341-352. EPA-600/3-77-001a.
Environmental Protection Agency, Research Triangle Park, N.C.,
1977.
44. Air Quality Criteria for Hydrocarbons. AP-64. U.S. Dept. of
Health, Education, and Welfare. Public Health Service, Environ-
mental Health Service, National Air Pollution Control Administra-
tion, Washington, D.C., 1970.
45. Dimitriades, B. Effects of Hydrocarbon and Nitrogen Oxides on
Photochemical Smog Formation. Environ. Sci. Technol., 6(3):253-
260, 1972.
46. Glasson, W.A., and P.H. Wendschuh. Multiday Irradiation of NO
Jx
Organic Mixtures. Int. Conf. Ox. Poll., Proc. 2:677-685. EPA-
600/3-77-OOlb. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
101
-------
47. Pitts, J.N,, Jr., A.M. Winer, K.R. Darnall, A.C. Lloyd, and G.J.
Doyle. Hydrocarbon Reactivity and the Role of Hydrocarbons, Oxides
of Nitrogen, and Aged Smog in the Production of Photochemical
Oxidants. Int. Conf. Ox. Poll., Proc. 2:687-704. EPA-600/3-77-
OOlb. Environmental Protection Agency, Research Triangle Park,
N.C., 1977.
48. Powers, T.R. The Oxidant Formation Potential of Emissions From
Catalyst-Equipped Vehicles. In: Ozone/Oxidants - Interactions
with the Total Environment. APCA Specialty Conference (Southwest
Section), Proceedings. Air Pollution Control Association, Pitts-
burgh, Pa., 1976. pp. 189-200.
49. Dimitriades, B. Oxidant Control Strategies, Part I: Urban Oxidant
Control Strategy Derived from Existing Smog Chamber Data. Environ.
Sci. Technol., ll(l):80-88, 1977.
50. Jeffries, H., D. Fox, and R. Kamens. Outdoor Smog Chamber Studies:
Light Effects Relative to Indoor Chambers. Environ. Sci. Technol.,
10(10):1006-1011, 1976.
51. National Research Council. Ozone and Other Photochemical Oxidants.
National Academy of Sciences, Washington, D.C., 1976.
52. Office of Air Quality Planning and Standards. Effectiveness of
Organic Emission Control Program as a Function of Geographic Loca-
tion. Environmental Protection Agency, Research Triangle Park,
N.C., 1977. Draft.
53. Wolff, G.T., P.J. Lioy, and G.D. Wight. An overview of the Current
Ozone Problem in the Midwestern and Northeastern U.S. In: Pro-
ceedings of the Specialty Meeting of the Air Pollution Control
Association on Hydrocarbon Control Feasibility, Air Pollution
Control Association, Pittsburgh, Pa., 1977. p. 98.
102
-------
54. Dimitriades, B. An Alternative to the Appendix-J Method for Cal-
culating Oxidant - and NO - Related Control Requirements. Int.
Conf. Ox. Poll., Proc. 2:871-879. EPA-600/3-77-001b. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
55. Wolff, A.T., P.J. Lioy, G.D. Wight, and R.E. Pacseri. Aerial In-
vestigation of the Ozone Plume Phenomenon. JAPCA, 27(5):460-463,
1977.
56. Dimitriades, B. Oxidant Control Strategy: Recent Developments.
Int. Conf. Ox. Poll., Proc. 2:1143-1154. EPA-600/3-77-001b.
Environmental Protection Agency, Research Triangle Park, N.C.,
1977.
57. Farrow, L.A., T.E. Graedel, and T.A. Weber. Urban Kinetic Chem-
istry under Altered Source Conditions. Int. Conf. Ox. Poll., Proc.
1:137-144. EPA-600/3-77-001a. Environmental Protection Agency,
Research Triangle Park, N.C., 1977.
58. Wolff, G.T., W.N. Stasiuk, Jr., P.E. Coffey, and R.E. Pasceri.
Aerial Ozone Measurements over New Jersey, New York, and Connecti-
cut. In: Proceedings of 68th APCA Annual Meeting, Vol. 5, Paper
75-58.6. Air Pollution Control Association, Pittsburgh, Pa., 1975.
59. Wolff, G.T. Local Effects of Nitric Oxide Emissions. 1974 Ozone
Study in New York, New Jersey, and Connecticut. Interstate Sani-
tation Commission, New York, 1975.
60. Environmental Protection Agency. Control of Photochemical Oxidants
Technical Basis and Implications of Recent Findings. Environmental
Protection Agency, Research Triangle Park, N.C., 1975. 35 pp.
103
-------
61. Wolff, G.T., P.J. Lioy, R.E. Meyers, R.T. Cederwall, G.D. Wight,
R.E. Pasceri, and R.S. Taylor. Anatomy of Two Ozone Transport
Episodes in the Washington, D.C. - Boston, Massachusetts Corridor.
Environ. Sci. Technol., 11(5):506-510, 1977.
62. Johnson, W., and H. Singh. The Effect of Ozone Layers Aloft on
Ground Level Concentrations. (To be published in Geophys. Res.
Letters, 1977.)
63. Conference on Hydrocarbon Emissions Control Policy: Moodus II.
Held in New York, Sept. 28, 1976. (Published by Connecticut De-
partment of Environmental Protection, Hartford, Conn.)
104
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO.
EPA-600/3-77-120
3. RECIPIENT'S ACCESSION-NO.
4. TITLE ANDSUBTITLE
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part VIII. The Issue of Optimum Oxidant Control Strat
5. REPORT DATE
December 1977
6. PERFORMING ORGANIZATION CODE
gy
7. AUTHOR(S)
1. W. Bonta 3. P. Koziar 5. F. Spuhler 7. L. Jagir
2. J. Paisie 4. B. Becker 6. K. Waid 8. G. Wolff
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
1. & 2. Dept. of Health & Mental Hygiene, Baltimore,
3. & 4. Dept. of Natural Resources, Madison, WI
5. & 6. Texas Air Control Board, Austin, TX
7. Dept. of Natural Resources. Lansing, MI
8. Interstate Sanitation Commission, New York, NY
Mi)
10. PROGRAM ELEMENT NO.
1AA603 AJ-13 (FY-76)
11. CONTRACT/GRANT NO.
1-2 DA-7-1934A 7. DA-7-2044A
3-4 DA-7-2175A 8. DA-7-2005H
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, NC 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 work for the EPA or for indus-
try) of widely recognized competence and experience in the area of photochemical
pollution occurrence and control.
Officials representing the states of Maryland, Wisconsin, Texas, Michigan,
and New York reviewed the papers presented at the 1976 International Conference
on Oxidants on the issue of optimum oxidant control strategy, and in Part VIII
offered their viewpoints in regards to the current status and resolution of the
issue and to the needs for additional research.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
* Air Pollution
* Ozone
* Control
* Strategy
13B
07B
18 DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
113
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
105
------- An error occurred while trying to OCR this image.
------- |