vxEPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
Research and Development EPA-600/D-84-144 Aug. 1984
ENVIRONMENTAL
RESEARCH BRIEF
Assessment of the Contribution of Stratospheric Ozone to
Ground-Level Ozone Concentrations
A. P. Altshuller
Introduction
Several assessments of technical issues related to ozone
and other photochemically generated products formed in
the atmosphere have been requested by the Office of Air
Quality Planning and Standards. This assessment is
concerned with evaluating the contribution of Oa trans-
ported from the stratosphere to ground level Oa concentra-
tions. The contributions of stratospheric O3 during short
episodic periods as well as to long term average Oa
concentrations is of interest. The episodic intrusions of
stratospheric air have the potential to cause the 03
concentrations at ground level to exceed the National
Ambient Air Standard for ozone. Longer time-averaged
concentrations of O3 reaching the boundary layer from the
stratosphere are of interest for several reasons. ThisOa(l)
contributes to the total Oa aloft which is available to be
entrained downward into urban areas, (2) when entrained
into rural areas upwind of urban areas it can contribute to
the Oa subsequently advected into urban areas, (3) when
entrained downward in rural areas can contribute to
adverse effects on field crops and forests.
Stratospheric O3 intrusions were reviewed as an issue in
1977.1 Since 1977, a number of additional investigations of
Oa concentrations aloft and at rural ground level locations
have appeared which associate the O3 measured with
stratospheric intrusions. In addition, several tropospheric
models have been developed which are concerned with the
origin of 03 in the troposphere. Therefore, the Office of
Research and Development arranged that a scientific
review on this subject be prepared.2 This review updates
earlier work and provides new insights into the problem.
This assessment is based on studies discussed in this
review2 as well as papers published after the review was
completed.
Discussion of Issues
Several issues have been identified for consideration in this
assessment. Included are those issues concerning strato-
spheric sources of Oa. However, it often is difficult, if not
impossible, to ascertain the origins of the Oa measured
aloft. Therefore, issues related to other sources of ozone
contributing to the ozone measured aloft must be con-
sidered also.
Issue 1. To what extent are direct rapid intrusions of
stratospheric Oa down through the troposphere and down
to ground level likely to result in Oa concentrations in
excess of the National Ambient Air Quality Standard for 03?
Large amounts of Oa are formed in the stratosphere.3
Atomic oxygen, 0, formed by the photolysis of oxygen
molecules by the ultraviolet radiation penetrating the
stratosphere combine with 02 to form ozone, O3. Ozone
destruction also can occur by photolysis of 03 by ultraviolet
radiation of higher wave lengths. The resulting equilibrium
for Oa in the stratosphere depends on temperature and on
altitude.3
Substantial evidence exists that O3 observed in the upper
troposphere has been transported down from the strato-
sphere as a result of stratospheric-tropospheric exchange
processes.1"3 The processes involved in such exchanges
include (a) seasonal adjustments of tropopause level, (b)
mean meridional circulation, (c) large scale eddies (tropo-
pause folding events) and, (d) small scale eddies.1'2 Over
half of the transfer has been attributed to processes (a) and
(b).' Such an evaluation was based on the estimate that, on
average, less than one tropopause folding event occurs
each day in the northern hemisphere. Subsequent work
indicates that about four tropopause folding events occur,
-------
on average, each day in the northern hemisphere. ' This
higher estimate results in the mechanism of transport
associated with large scale eddies, tropopause folding
events, being the most important mechanism for transport
of 03 from the lower stratosphere into the upper tropo-
sphere at mid-latitudes.2
Tropopause folding events occur near the mid-latitude jet
stream under conditions having the potential for rapid and
relatively undiluted transport of stratospheric air containing
Os down into the upper troposphere.2 Tropopause folding
events are associated with the positions of 500-mb low
pressure troughs and jet streams. As a result, the frequency
of tropopause folding events will vary substantially with the
month of the year, and with latitude and longitude. For
example, no 500-mb low pressure troughs are likely to
occur during the summer months between 20°N and 30°N
latitude over the United States.2'4
The downward transport of Oa in air originating in the
stratosphere has been followed by instrumented aircraft
subsequent to a number of tropopause folding events.2'5
These aircraft flights provide Os prof iles through portions of
the troposphere. It has been observed that the air originating
in the stratosphere frequently becomes horizontal in the
upper troposphere although stratospheric air at times
becomes a discrete horizontal layer as low as several
kilometers above ground level.6 The Oa concentrations at
the lower extremities of such layers usually are in the 60 to
80 ppb concentration level at or somewhat above the 3 km
altitude level.5 Stratospheric air has not been followed
down to ground level.
Three-dimensional air parcel trajectories can be used to
trace movements from the tropopause down to the 700-mb
level, 10,000 ft, if appropriate meteorological observations
are available.5~6The extent of the geographical area having
the potential to be impacted can be estimated. The total
integrated ozone flux from the 700-mb level downward
over such geographical areas can be calculated.6 This
technique provides no information on the detailed move-
ments of the air parcels within the planetary boundary layer
or the losses of Oa which may occur.
Four hypothetical mechanisms have been suggested to
account for the ultimate fate of stratospheric 63 injected
into the troposphere.5
1. Dissipation of the intrusion by general mixing and
diffusion into the troposphere above the planetary
boundary layer, i.e., the free troposphere.
2. Persistence of the intrusion down to the planetary
boundary layer where the lower portion of the
intrusion is mixed down to the ground by turbulent
eddies and convection at the top of the boundary
layer.
3. Coupling of the intrusion to the frontal zone associ-
ated with a cold front with direct transport to the
ground by frontal downdrafts.
4. The tropospheric air becomes entrained in organized
frontal and prefrontal convection, which then trans-
ports it to the ground in connection with rainshowers
or thunderstorm downdrafts.
Some theoretical and observational results support the
existence of mechanisms 3 and 4. However, only limited
field data is available relevant to the quantitative signifi-
cance of these mechanisms.5
The discussion to this point has been concerned with
information related to the downward movement of strato-
spheric Oa in intrusion events. The other approach is to
attempt to identify episodes of elevated Os concentrations
at ground level which can be associated with stratospheric
intrusions. Ten case studies have been identified in a
compilation of the literature of episodic events ranging from
1964 to 1978.2 The length of the records examined were
highly variable. Records of several years or more were
examined in five of the case studies. In one other case
study, stratospheric intrusion has been claimed to impact
on a site in the western United States over an extended
period. This study will be considered in more detail in
discussion of a subsequent issue.
The length of the reported episodic events varied from one
hour or less to a few days. While several of the episodes
were reported during the summer, most of the episodes
identified occurred between November and March. Six of
the ten episodes, involving ten days, were reported at
locations within the United States. Three of the case
studies reported in the United States involved measure-
ments at elevated locations at or above 1.5 km.
The experimental evidence supporting claims that the
elevated Os concentrations were associated with strato-'
spheric intrusions reaching the surface vary considerably
among case studies. The reliability of several of the case
studies has been questioned.2 The questioned reports
involve episodes in which ground level (^concentrations in
excess of 100 ppb were associated with stratospheric Oa
intrusions. These ground level Oa concentrations were
higher than expected according to the O3 profiles obtained
aloft during the actual tracking of stratospheric intrusions
through the troposphere.
During only two of the episodes associated with strato-
spheric intrusions at ground level locations within the
United States did the Oa concentrations exceed 120 ppb.
These two episodes occurred at Quincy, FL on March 3,
1964 and Santa Rosa, CA on November 19, 1972. The 03
concentration at Santa Rosa reached 200-230 ppb for one
hour during passage of a thunderstorm.7 The elevated Oa
concentration was highly localized with no similar eleva-
tions of Oa concentrations observed at the other monitoring
sites in adjacent areas of northern California. The evidence
presented to support this episode was based on isentropic
trajectory analysis, the low water vapor mixing ratio, and
the high potential vorticity along with an assumption that a
growing, precipitating cloud entered a layer of stratospheric
air aloft. The O3 in the core of the downdraft was said to
have arrived at ground level largely undiluted. This episode
appears to be an example of mechanism 4 discussed above.
Tracer techniques involving isotopic species such as 90Sr
and 7Be known to be present in the stratosphere have been
utilized to estimate stratospheric contributions to surface
Oa concentrations. These techniques are based on the
assumption that the aerosol particles containing the 90Sr
-------
and 7Be behavior is similar enough to that of Os to serve as
an adequate tracer for the transport of the 03. This
assumption has been questioned; and the possible limita-
tions to these techniques will be discussed after some of
the results obtained from their use are considered.
The concentrations of 90Sr measured at ground level after
nuclear weapons testing in the early 1960's has been used
to estimate stratospheric contributions to ground level 03
concentrations.1 Maximum 24-hour average 90Sr concen-
trations and 90Sr to 03 ratios measured during 1963 and
1964 indicated that the corresponding 24-hour average 03
concentrations were in the range of 15 to 30 ppb. Maximum
1-hr Oa concentrations associated with stratospheric
intrusions could not be adequately estimated.
In more recent studies, a 7Be tracer technique has been
used.8"10 The 7Be is formed in the stratosphere through the
collision of cosmic ray protons and neutrons with oxygen
and nitrogen. The 7Be isotope also can be formed (a) in the
upper troposphere south of 30°N latitude and (b) in the
upper troposphere on the warm-air (anticyclonic) side of the
mid-latitude jet stream.2 Because no mechanisms for 7Be
formation exist in the lower troposphere, a substantial
downward gradient in 7Be concentrations occurs from the
tropopause to ground level.
The radioactive lifetime of 7Be is 53 days,2'11 while the
residence-time in the free troposphere of the aerosols
containing 7Be has been estimated at 35±15 days. Ozone
has been estimated to have average lifetimes at the same
altitudes ranging from 1 to 4 months with its losses related
to photochemical processes and deposition.2'11 As a result,
the concurrent presence and transport of 7Be and O3
through the troposphere to ground level was considered to
have occurred.
Measurements of 7Be and of 03at the summit of Whiteface,
Mt., NY have been reported for both July 1975s and June
and July 1977.9 Of the 30 days of measurement in July
1975 at Whiteface Mt., stratospheric origin was claimed for
the air parcels on 7 days based on isentropic trajectory
analysis and potential vorticities. The 7Be concentrations
on these seven days were in the 100 to 200 fCim"3 range.
In a subsequent study9 at the same location, quantitative
estimates of the contribution of stratospheric 03 concen-
trations was made for the 11 days out of a total of 26 days on
which the 24-hour average 7Be concentrations exceeded
200fCim~3. Isentropic trajectory analyses were not reported
during this latter study. On these 11 days, the estimated
stratospheric contributions to the observed O3 concentra-
tions averaged 28 ppb, and ranged from 19 to 47 ppb. These
contributions averaged 51 %, and on individual days ranged
from 26 to 94% of the observed ozone concentrations.
These estimates are based on a procedure utilizing an
average ratio of 7Be to 03 of 10.8 fCim"3 per ppb measured
in samples collected above the tropopause near 40°N
latitude during April to June 1978.9
If the 10.8 fCim"3 per ppb ratio9 is used with the results for
the 7 days in July 1975,8 the estimated stratospheric
contributions to the observed 03 concentrations averged 14
ppb, and ranged from 10 to 18 ppb. The estimated strato-
spheric contributions averaged 41 % of the observed ozone
concentrations, and oh individual days ranged from 24 to
59%.
In a later study,10 both 6-hour and 24-hour average 7Be
samples were obtained at sites in Houston, TX and San
Antonio, TX between June and October 1978. Four strato-
spheric intrusion events were claimed to have occurred
either in the San Antonio or Houston area, but not in both
areas concurrently. These episodes were based on the 7Be
concentrations exceeding 400 fCinrT3 on either a 6-hour or
24-hour average. During the 6-hour periods of 7Be measure-
ment, 40 and 60 ppb of 03 was attributed to stratospheric
intrusions. However, in one of the episodes the observed 03
concentration, 40 ppb, actually was less than the calculated
O3 attributed to stratospheric intrusion. Isentropic trajectory
analyses were not reported in this study,10 but are seen as
essential in view of the absence of 500-mb troughs at these
latitudes during the study period.2'4
Reservations have been expressed as to the 7Be-to-03 ratio
technique.11 The reasons for these reservations are as
follows: (a) 7Be and 03 are subject to different removal
processes, (b) evidence is lacking that short-term intrusions
routinely penetrate to ground level, (c) use of isentropic
trajectory analysis is essential to verify that the air parcels
reaching the surface came from the stratosphere rather
than the upper troposphere, (d) no generally accepted 7Be-
to-O3 ratios exist for the lower stratosphere and a fixed ratio
is unlikely to exist in this region of the atmosphere, and (e)
short sampling times for 7Be are needed to properly resolve
stratospheric contributions. These limitations must be
taken into consideration in judging the validity of the
quantitative estimates discussed above.
Summary—Well-authenticated episodes of stratospheric
intrusions reaching ground level are rare. Episodes in
which stratospheric intrusions contributed to O3 concen-
trations at mountain locations2'8'9 are more frequent than
those reaching ground level. However, very occasional
episodes involving intrusions reaching ground level as a
result of mechanisms 3 and 4 discussed above cannot be
discounted. Episodes in which stratospheric intrusions
reaching ground level would contribute sufficient 03to lead
to observed 03 concentrations in excess of 120 ppb appear
to be extremely infrequent.
Issue 2. To what extent does stratospheric 03 contribute
to longer time-averaged 03 concentrations at ground level
in non-urban areas?
Longer time-averaged O3 concentrations in non-urban
areas are of interest for several reasons: (a) in estimating
typical inflows into individual urban areas, (b) to provide
boundary conditions for regional scale 03 models, (c) in
estimating exposures of field crops and forested areas to
03. In addition to improving our basic scientific under-
standing, identification and quantitation of individual
sources of background 03are useful in assessing the extent
to which the background 03 concentrations may be
controllable.
In early work on the 03 background during the 1960's, it
was assumed that the 03 concentrations observed in non-
-------
urban areas could be uniquely associated with stratospheric
03 transport to ground level.3 This conclusion was based on
the recognition, at that time, of only one other source for 03
production. The source was photochemical formation of 03
within large urban areas. With the recognition during the
1970's that long-range transport of 03 and its precursors
could occur from anthropogenic sources, the determination
of the contribution of stratospheric 03 to the 03 measured in
non-urban areas became more complex.
Compilations of 03 concentration levels in the free tropo-
sphere are available.2 Based on ozonesonde measurements
obtained between 1962 and 1975 at 5 km altitude, the
seasonal average daily maximum ozone concentrations
between 40°N and 60°N latitude in ppb were as follows:
winter, 40; spring, 40 to 50; summer, 40 to 60; fall, 40.
Similarly, ozonesonde measurements obtained between
1962 and 1965 at altitudes between 3 and 7 km have been
considered separately to obtain mean annual 03 concen-
trations over the eastern United States of 60 to 65 ppb, and
over the western United States of 55 to 60 ppb. Aircraft
measurements in 1978 of 03 profiles over the central
United States provided the following 03 concentrations in
ppb: May, 1.5 km, 55; 4 km, 72; October, 1.5 km, 33; 4 km,
46. Because of the multiple sources mentioned above, the
O3 concentrations in the free troposphere above the
planetary boundary layer should not be assumed to
necessarily equal the stratospheric 03 contribution.
Observations of the average of daily maximum 03 concen-
trations and the average of hourly O3 concentrations at
sites between ground level and 3 km in the northern and
southern hemisphere have been compiled/'12"1'4 At non-
urban sites in the United States between 1976-1978, 03
concentrations in ppb by quarter of the year were observed
over the following ranges: 1st quarter, 30 to 44; 2nd
quarter, 33 to 52; 3rd quarter, 29 to 45; 4th quarter, 22 to
30. Because of the complexities in the origins of 03, these
observed 03 concentrations should not be assumed to be
predominantly stratospheric 03.
Less than half of the variance in O3 concentrations is
explained by a linear relation with 7Be.4'15 Consequently, a
detailed analysis of the meteorological conditions as related
to the 7Be and O3 concentrations was undertaken.4
Three-dimensional air parcel trajectories4 were calculated
during the study period.'5 It was found that southwesterly
air flow resulted in increasing 03, while northwesterly air
flow resulted in decreasing O3. However, the higher 7Be
concentrations were observed with both southwesterly and
northwesterly air flows. During southwesterly flows, no
500-mb low pressure troughs occurred.4 It was postulated
that the higher 7Be concentrations from the'southwest
involved transport from the troposphere by subsiding
circulation associated with high pressure systems.4 Three
weak stratospheric intrusion events associated with 500
mb low pressure troughs were identified in association
with northwesterly air flow. During these weak intrusion
events, the 24-hour average O3 concentrations ranged
from 32 to 34 ppb. During these intrusion events the 24-
hour average 7Be concentrations ranged from 125 to 175
fCim"3.4 If the stratospheric 7Be-to-03 ratio of 10.8 fCirrf3
per ppb9 can be applied to these events, the predicted
stratospheric 03concentrations would range from 12 to 16
ppb. Half of 7Be concentrations measured during the study
were in the range between 40fCim"3 and 125fCim"3. Such
low 7Be concentrations suggest long residence times with
dilution by air present in the lower free troposphere or the
planetary boundary layer.
The 90Sr ground level concentration measurements have
been used to estimate long time-averaged 03 concentra-
tions.1 The mean 90Sr concentrations between January and
June 1963 outside of the nuclear weapons test area,
ranged from 5 pCim"3 to 9 pCim"3, corresponding to mean
03 concentration from 7 to 12 ppb. Based on these
measurements and other evidence, the background associ-
ated with stratospheric O3 was assumed to be approxi-
mately 15 ppb.1
A relatively complete set of air-quality measurements were
obtained during an eight week period in July to September
1978 at a relatively remote location 40 km west-northwest
of Pierre, SD.16 Hourly average 03 concentrations ranged
from 10 to 56 ppb with a mean of maximum hourly 03
values of 41 ppb. While the authors concluded that the
dominant source of the O3 appeared to be the stratosphere,
they also pointed out that their measurements could not
distinguish between 03 from the stratosphere and O3 from
the upper troposphere. Based on (1) remoteness of the
location, (2) the low concentrations measured of several
species, (3) the small standard deviations of the concentra-
tions of these species, and (4) the interpretation of 7Be
measurements, the authors considered that anthropogenic
sources of 03 within the boundary layer were not contrib-
uting significantly to the 03 measured at this site. The
highest O3 concentrations were observed in the continental
tropical air masses which usually arrived on the backside of
the high pressure systems passing the site during the study
period. An increase in 7Be as well as 03 was observed on
the backside compared to the f rontside of the high pressure
systems.
Summary— Based on the results discussed above, it appears
that only a part of the observed 03 concentrations even at
relatively remote locations can be associated with strato-
spheric 03 contributions.
Issue 3. Does in-situ O3 production in the free troposphere
contribute significantly to the 03from aloft reaching ground
level?
In-situ photochemical formation of O3 within the free
troposphere can occur as the result of reactions involving
methane and carbon monoxide with nitrogen oxides.16 On a
global basis, anthropogenic emissions of CO have been
estimated to contribute from one-sixth of the total CO
present in the free troposphere.17"20 In the northern
hemisphere, a range of contributions from anthropogenic
sources of 31 to 54% has been estimated.20 In the northern
hemisphere, a contribution of about 10% to the free
troposphere has been associated with anthropogenic
sources of CH4.21 Over the United States, a recent estimate
associates over 95% of the NO, emissions with anthropo-
genic sources.22
-------
Production of O3 by photochemical processes in the free
troposphere has been predicted to occur at significant rates
for NOX concentrations above 0.03 ppb.16 The vertical
profile of NOX through the troposphere is critical to photo-
chemical modeling in this region.2 Over remote areas, a
vertical profile has been observed with NO* concentrations
increasing from 0.01 ppb near ground level to 1 ppb at the
tropopause.23 However, the concentrations of NOX near the
surface in populated non-urban areas on continents are in
the 1 to 10 ppb range.24 Over these areas, a vertical profile
of NO, should decrease with increasing altitude. Several
recent tropospheric models predict in-situ tropospheric
formation of 03.25'27
In a study26 using a one-dimensional model, the observed
latitudinal features for O3 could not be reproduced unless
free tropospheric photochemical production of 03 were
included. A net photochemical source of Oa was predicted
from 25°N to 65°N with very high rates of O3 production in
the free troposphere between 40°N and 60°N.
A two-dimensional tropospheric model has been tested
using a number of assumptions as to sources of O3.26 The
assumption of a stratospheric source of O3—along with
photochemical generation of Oa in the upper troposphere,
with NOX sources from lightning and the stratosphere—
reasonably well simulated profiles of 03 over remote
maritime regions. Inclusion of ground level sources of NO,
as necessary to obtain predicted 03 profiles consistent with
observed profiles at mid-latitudes in the northern hemi-
sphere. Inclusion of ground level pollution sources of NO,
more than doubled the amount of 03 in the lower tropo-
sphere in the summer at mid-latitudes in the northern
hemisphere.
In a third modeling study, one- and two-dimensional tropo-
spheric models were utilized.27 When the one-dimensional
model was applied to summertime conditions at 40°N, the
03 concentrations in ppb predicted at several latitudes
without and with ground level NO, sources were as follows:
1 km, 15, 37; 4 km, 26, 50; 7 km, 43, 67 and 10 km, 63, 87.
In the lower troposphere between 1 and 4 km the model
predicts that half or more of the 03 produced was associated
with the ground-level NO, emissions. The total O3 concen-
trations predicted with inclusion of ground level NO,
emissions corresponded satisfactorily to the observed 03
concentrations in the troposphere at mid-latitudes in the
northern hemisphere in summertime. When the two-
dimensional model was applied to summertime conditions
in the northern hemisphere at mid-latitudes, the inclusion
of upward transport of ground level emissions of CO and
NO, resulted in an increase of more than 20 ppb of 03 in the
total O3 predicted. This result is in good agreement with the
differences of 21 to 24 ppb in 03 with altitude predicted for
the same conditions with the two-dimensional model.
Additional evidence for tropospheric 03formation has been
obtained from measurements of the small scale vertical
variability of O3 and CO in the troposphere.27 A large scale
region of positive correlation was observed between 20°N
and 45°N from altitudes of 4 to 10 km. This strong positive
correlation was associated with in-situ photochemical O3
production in this same region of the troposphere.
Summary—A number of modeling studies predict substan-
tial increments of 03associated with in-situ photochemical
Oa production in the free troposphere at mid-latitudes in the
northern hemisphere, especially in the summertime. Two
such studies predict a doubling in 03 concentrations in the
lower portions of the troposphere as a result of upward
transport of ground level NO* emissions.20'21 A substantial
portion of the precursors participating in in-situ tropo-
spheric 03 production are emitted by anthropogenic sources
at the earth's surface.
General Discussion
There is little documented evidence that high episodic O3
concentrations occur at ground level in urban or non-urban
sites in the United States as a result of direct stratospheric
intrusions of 03. Only two episodes have been reported in
which stratospheric 03 intrusions have caused ground level
Oa concentrations to exceed 120 ppb.2 In a number of other
episodes during which both elevated 7Be concentrations
and Oa were measured, application of the 7Be-to-03 ratio
technique indicates that, on average, about half of the Oa
measured actually originated in the stratosphere.
Several different processes contribute to the longer time-
averaged Oa concentrations at relatively remote locations.
A part of the O3 measured is likely to be of stratospheric
origin. In-situ O3 formation in the free troposphere also
contributes to the 03 concentrations at remote locations.
There is evidence for local in-situ formation of 03 even at
relatively remote sites. Aircraft'measurements of 03 over
ranchland and irrigated fields in rural northeastern Colorado
indicated, that during some daytime periods, the entire rate
of increase with time of 03 in the lower half of the mixed
layer could be accounted for by local photochemical 03
production.28 At Zugspitze Mountain at 2964 m MSL, in a
clean remote area of the Bavarian Alps, the 03 concentra-
tion was lower than that over an adjacent valley site at 740
m MSL2 The incremental monthly average daily maximum
Oa concentrations of 8 to 15 ppb on sunny days between
March and September have been attributed either to local
03 formation or to long range transport.2 In contrast, over
pine forests north of Houston, TX, net destruction of Oa was
indicated in measurements above the forest canopy.29 This
result is consistent with a recent review which concluded
that biogenic hydrocarbons would not make a substantial
contribution to 03 formation.30 Photochemical 03 produc-
tion also has been observed in traverses taken well out over
the Gulf of Mexico29,'although at a slower rate than over
northeastern Colorado.28 Measurements of Oa over the
Pacific have been considered consistent with local photo-
chemical Oaover the remote marine boundary layer.31
A fourth process which could contribute 03 to remote sites
is transport upwards from the planetary boundary layer into
the free troposphere transport over long distances in the
free troposphere followed by downward transport back
down into the boundary layer. Cloud venting through a
well-developed convective, strato cumulus-topped layer
can effectively transport O3 and other species from the
boundary layer into the free troposphere.31 Thunderstorm
systems can cause the exchange of air containing 03 and
other species from the boundary layer with air from the free
troposphere.32
-------
At non-urban locations in the eastern United States,
additional processes contribute to Oa measured. These
processes involve Oa formation and transport within the
boundary layer in individual urban,33"39 power plant,
industrial,46 and petroleum refinery47"49plumes. During the
movement of high pressure systems over the eastern
United States the contributions of Oaand precursors can contribute
over multiday periods.50'51 Most of the episodes of elevated
Os concentrations studied in the eastern United States
have been associated with these anthropogenic pro-
cesses.12'13'50'52
An additional complication results from the varying lifetime
of Oa with season, and in different regions of the atmos-
phere. The lifetime of Oa in the free troposphere due to
photochemical destruction has been estimated to vary from
45daysinthesummerto 105daysinthe winter.11 Within
the planetary boundary layer, the lifetime of Oa is shorter.
Aloft at night, within the planetary boundary layer, a
lifetime of about 30 hours was estimated from vertical
profiles of Oa53'54 and from 1 6 to 34 hours was estimated
from 03 measurement during a balloon flight. If transport
from the top of the planetary boundary layer to ground level
is slow, only a portion of the Oa will reach ground level.
Conclusions
1. Only a small number of short-term direct strato-
spheric ozone intrusions to the surface have been
adequately documented. Of these episodes, only a
few have been associated with low-altitude ground
level locations in the United States. Such intrusions
to the surface appear to be rare and are usually
observed at rural sites. Direct stratospheric ozone
intrusions to the surface which result in ozone
concentrations exceeding the National Ambient Air
Quality Standard for 03 of 120 ppb appear to be
extremely infrequent.
2. Episodes of substantially elevated ozone concentra-
tions in rural areas in the eastern United States
predominately have been associated with anthropo-
genic sources of ozone.
3. Stratospheric ozone contributes to longer term
average background ozone concentrations at ground
level sites. Substantial contributions also can be
associated with ozone formed in the free troposphere,
and with ozone formed from a number of types of
sources within the boundary layer. The average
background ozone is likely to be the result of contri-
butionsfrom these processes which can vary diurnal-
ly, synoptically and seasonally.
4. Based on tropospheric model calculations along with
7Be-to-03 and 90Sr-to-Oa ratios, it is estimated that
during the summer months 15 ppb or less of the
background ozone concentrations may originate in
the stratosphere. In the early spring, the stratospheric
ozone contribution to the background ozone is more
significant
5 The contribution of stratospheric ozone plus the free
tropospheric ozone to average seasonal ozone con-
centrations should be substantial, particularly over
many of the rural areas in which major field crops are
grown.
6. Tracer substances can be useful in identifying the
origins of ozone measured in rural areas. However, a
tracer such as 7Be formed in the stratosphere and
upper troposphere is only an indicator of a contri-
bution of air from well aloft. The sources and removal
processes for ozone can be much different than the
aerosols to which 7Be is attached.
Recommendations
1. Better estimates should be made of the contribution
of various sources of ozone to the growing season
average ozone concentration in representative crop
areas.
2. Since a substantial fraction of the precursors to free
troposphere ozone formation are anthropogenic in
origin, hemispheric scale models should be utilized in
developing control strategies, if needed, for the
reduction of average growing season ozone concen-
trations.
3. A more complete evaluation of the limitations and
applications of 7Be as a tracer of stratospheric and
free tropospheric ozone would be useful. A number of
other tracer substances such as particle sulfur and
lead from major sources of ozone should be utilized to
aid in quantifying contributions to background ozone
concentrations.
References
1. Mohnen, V.A. and Reiter, E.R. (1977) International
conference on oxidants—1976. Analysis of evidence
and view points. Part III. The issue of stratospheric
ozone intrusion. EPA-600/3-77-115.
2. Viezee W. and Singh, H.B. (1982) Contribution of
stratospheric ozone to ground-level ozone concentra-
tions—A scientific review of existing evidence. SRI
Project 3643. Prepared for Environmental Sciences
Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711.
3. Junge, C.E. (1963) Air Chemistry and Radioactivity.
International Geophysical Series Vol. 4, pp. 37-59.
Academic Press, NY.
4. Viezee, W., Singh, H.B., and Shigeishi, H. (1982)The
impact of stratospheric ozone on tropospheric air
quality—Implications from an analysis of existing
field data. Final Report to Coordinating Research
Council, Inc., 219 Perimeter Center Parkway, Atlanta,
GA 30346.
5. Johnson, W.B. and Viezee, W. (1981) Stratospheric
ozone in the lower troposphere—I. Presentation and
interpretation of aircraft measurements. Atmos.
Environ. 15, 1309-1323.
-------
6. Viezee, W. and Singh, H.B. (1982) A method to predict
a stratospheric impact on boundary-layer and ground-
level ozone concentrations. Draft Report. SRI Project
3643. Prepared for Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711.
7. Lamb, R.G.(1977)Acasestudyof stratospheric ozone
affecting ground-level oxidant concentrations. J.
Appl. Meteorol. 16, 780-794.
8. Husain, L, Coffey, P.E., Meyers, R.E., and Cederwall,
R.T. (1977) Ozone transport from stratosphere to
troposphere. Geophys. Res. Lett. 4, 363-365.
9. Dutkiewicz, V.A. and Husain, L. (1979) Determination
of stratospheric ozone at ground level using 7Be/
ozone ratios. Geophys. Res. Lett. 6, 171-174.
10. Husain, L., Rusheed, A., and Dutkiewicz, V.A. (1979)
Sources of natural ozone: Stratospheric intrusions in
Texas. In an Air Pollution Control Association speci-
alty conference on: Ozone/oxidants. Interactions with
the total environment. (Ed., E.R. Fredrick). Houston,
TX.
11. Singh, H.B., Viezee, W., Johnson, W.B., and Ludwig,
F.L. (1979) The impact of stratospheric ozone on
tropospheric air quality. In an Air Pollution Control
Association specialty conference on: Ozone/oxidants.
Interactions with the total environment. (Ed., E.R.
Fredrick). Houston, TX.
12. Evans, G., Finkelstein, P., Martin, B., Possiel, N., and
Graves, M. (1983) Ozone measurements from a
network of remote sites. J. Air Poll. ControlAssoc. 33,
291-296.
13. Evans, G., Finkelstein, P., Martin, B., Possiel, N., and
Graves M. (1982) The National Air Pollution Back-
ground Network 1976-1980. Environmental Monitor-
ing Systems Laboratory Report, U.S. Environmental
Protection Agency, Research Triangle Park, NC
27711.
14. Computed from a tabulation of ozone meaurements
at Whiteface Mt., NY provided by Dr. Volker Mohnen
1982.
15. Kelly, N.A., Wolff, G.T., and Ferman, M.A. (1982)
Background pollutant measurements in air masses
affecting the eastern half of the United States—I. Air
masses arriving from the northwest. Atmos. Environ.
16, 1077-1088.
16. Fishman, J., Solomon, S., and Crutzen, P.J. (1979)
Observational and theoretical evidence in support of
a significant in-situ photochemical source of tropo-
spheric ozone. Tellus 31, 432-446.
17. Seller, W. (1974) The cycle of atmospheric CO, Tellus
26, 117-135.
18. Zimmerman, P.R., Chatfield, R.B., Fishman, J.,
Crutzen, P.J., and Hanst, P.L. (1978) Estimations on
the production of CO and H2 from the oxidation of
hydrocarbon emissions from vegetation. Geophys.
Res. Lett. 5,679-681.
19. Kitada,T. and Peters, LK. (1982)Athree-dimensional
transform-chemistry analysis of CO and CH4 in the
troposphere. In 2nd symposium: Composition of the
nonurban atmosphere. May 25-28, 1982, Williams-
burg, VA, Amer. Meteor. Soc., 45 Beacon St., Boston,
MA 02108.
20. Hanst, P.L, Spence, J.W., and Edney, E.O. (1980)
Carbon monoxide production in photooxidation of
organic molecules in the air. Atmos. Environ. 14,
1077-1088.
21. Khalil, M.A.K. and Rasmussen, R.A. (1983) Sources,
sinks and seasonal cycles of atmospheric methane. J.
Geophys. Res. 88, 5131-5144.
22. Robinson, E. (1983) Natural emissions sources.
Chapter A-2, Section 2.2, In: The acidic deposition
phenomenon and its effects: Critical assessment
review papers (Altshuller, A.P. and Linthurst, R.A.,
eds.) Vol. I. Public Review Draft.
23. Kley, D., Drummond, J.W., McFarland, M., and Liu, S.
C. (1981) Tropospheric profiles of NOX. J. Geophys.
Res. 86, 3153-3161.
24. Altshuller, A.P. (1983) Atmospheric concentrations
and distributions of chemical substances. Chapter
A-5, Section 5.3. In: The acidic deposition phenom-
enon and its effects: Critical assessment review
papers (Altshuller, A.P. and Linthurst, R.A., eds.) Vol. I
Public Review Draft.
25. Hameed, S. and Steward, R.W. (1983) Latitudinal
variation of tropospheric ozone in a photochemical
model. J. Geophys. Res. 88, 5153-5162.
26. Gidel, L.T. and Crutzen, P.J. (1982)The role of various
natural and anthropogenic NOX emissions of atmos-
pheric ozone in a two dimensional model. In 2nd
symposium: Composition of the nonurban atmos-
phere. May 25-28, 1982. Williamsburg, VA, Amer.
Meteorol. Soc., 45 Beacon St., Boston, MA 02108.
27. Fishman, J. and Seller, W. (1983) Correlative nature
of ozone and carbon monoxide in the troposphere:
Implications for the tropospheric ozone budget. J.
Geophys. Res. 88, 3662-3670.
28. Lenschow, D.H., Pearson, R., Jr., and Stankov, B.B.
(1981) Estimating the ozone budget in the boundary
layer by use of aircraft measurements of ozone eddy
flux and mean concentration. J. Geophys. Res., 86,
7291-7297.
29. Lenschow, D.H., Pearson, R., Jr., and Stankov, B.B.
(1982) Measurements of ozone vertical flux to ocean
and forest. J. Geophys. Res. 87, 8833-8837.
-------
30. Altshuller, A.P (1983) Review: Natural volatile or- 42.
ganic substances and their effect on air quality in the
United States. Atmos. Environ. 17, 21 31 -21 65.
31. Gidel, L.T. (1983) On the maintenance of the marine 43.
boundary layer-free tropospheric gradient of transient
tracers. J. Geophys. Res. 88, 5145-5152.
32. Lyons, W.A. and Calby, R.H. (1983) Impact of meso- 44.
scale convective precipitation systems on regional
visibility and ozone distributions. Report to Environ-
mental Sciences Research Laboratory, U.S. Environ-
mental Protection Agency, Research Triangle Park,
NC 27711.
33. Blumenthal, D.L, White, W.H., and Smith, T.B. (1978)
Anatomy of a Los Angeles smog episode: Pollutant
transport in the daytime sea breeze regime. Atmos.
Environ. 12, 893-907.
34. Spicer, C.W., Joseph, D.W., Sticksel, P.R., and Ward,
G.F. (1979) Ozone sources and transport in the
northeastern United States. Environ. Sci. Techno/.
13, 975-985.
35. Wolff, G.T., Lioy, P.J., Wight, G.D., Pasceri, R.E.
(1977) Aerial investigation of the ozone plume
phenomenon. 7. Air Poll. Control Assoc. 27, 460-463.
36. Spicer, C.W. and Sverdrup, G.M. (1 981) Trace nitro-
gen chemistry during the Philadelphia oxidant en-
hancement study (1979). Report to Office of Air
Quality Planning and Standards, U.S. Environmental
Protection Agency, Research Triangle Park, NC
27711.
37. Spicer, C.W., Koetz, J.R., Keigley, G.W., Sverdrup,
G.M., and Ward, G.F. (1 982) Nitrogen oxides reactions
within urban plumes transported over the ocean.
Report to Environmental Sciences Research Labora-
tory, U.S. Environmental Protection Agency, Research
Triangle Park, NC 27711.
38. White, W.H., Anderson, J.A., Blumenthal, D.L., Husar,
R.B., Gillani, N.V., Husar, J.D., and Wilson, W.E., Jr.
(1976) Formation and transport of secondary air
pollutants: Ozone and aerosols in the St. Louis urban
plume. Science 194, 187.
45.
46.
47.
48.
49.
50.
51.
52.
39. Spicer, C.W., Joseph, D.W., and Sticksel, P.R. (1 982)
An investigation of the ozone plume fro ma small city.
J. Air Poll. Control Assoc. 32, 278-281. 53.
40. Davis, D.D., Smith, G., and Klauber, G. (1974) Trace
gas analysis of power plant plumes via aircraft
measurement. Os, NO, and S02 chemistry. Science 54
186, 733-736
41 Gillani, N V., Husar, R B., Husar, J.D., Patterson, D.E.,
and Wilson, W E , Jr (1978) Project MISTT: Kinetics
of paniculate sulfur formation in a power plant plume
out to 300 km Atmos. Environ. 12, 589-598
Gillani, N.V. and Wilson, W.E. (1980) Formation of
ozone and aerosols in power plant plumes. Ann. New
YorkAcad. Sci. 338, 276-296.
Miller, D.F., Alkezweeny, A.J., Hales, J.M., and Lee,
R.N. (1978) Ozone formation related to power plant
emissions. Science 202, 1186-1188.
Hegg, D.A., Hobbs, P.V., and Radke, LF. (1976)
Reactions of nitrogen oxides, ozone and sulfur in
power plant plumes. EPRI EA-270. Prepared for
Electric Power Research Institute, 3412 Hillview
Ave., Palo Alto, CA 94304.
Ogren, J.A., Blumenthal, D.L., and Vanderpol, A.H.
(1977) Oxidant measurements in western power
plant plumes. Vol. 1. Technical analysis EPRI-421.
Prepared for Electric Power Research Institute, 341 2
Hillview Ave., Palo Alto, CA 94304.
Sexton, K. and Westberg, H. (1980) Ambient hydro-
carbon and ozone measurements downwind of a
large automotive painting plant. Environ. Sci.
Techno/. 14, 329-332.
Westberg, H.H., Allwine, K.J., and Robinson, E. (1 977)
Ambient hydrocarbon and ozone concentrations near
a refinery: Lawrenceville, Illinois—1974. EPA-600/
7-77-049.
Sexton, K. and Westberg, H. (1980). Ambient hydro-
carbon and ozone concentrations near a refinery.
Benicia, California 1975. EPA-600/7-80-028.
Sexton, K. and Westberg, H. (1983) Photochemical
ozone formation from petroleum refinery emissions.
Atmos. Environ. 17, 467-475.
Vukovich, F.M., Bach, W.D., Jr., Crissman, B.W., and
King, W.V. (1 977) On the relationship between high
ozone in the rural surface layer and high pressure
systems. Atmos. Environ. 77,967-983.
Wolff, G.T. and Lioy, P.J. (1980) Development of an
ozone river associated with synoptic scale episodes in
the eastern United States. Environ. Sci. Techno/. 14,
1257-1260.
Martinez, J.R. and Singh, H.B. (1979) Survey of the
role of NO, in nonurban ozone formation. SRI Project
6780-8. Report to Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711.
Research Triangle Institute (1975) Investigations of
rural oxidant levels as related to urban hydrocarbon
control strategies. EPA-450/3-74-036.
Sickles, J.E., II, Ripperton, L.A., and Eaton, W.C.
(1977) Oxidant and precursor transport simulation
studies in the Research Triangle Institute Smog
Chambers. In: Proceedings of the International Con-
ference on Photochemical Pollution and Its Control.
pp. 319-328. Vol. I. (Dimitriades, B., ed.) EPA-600/3-
77-012.
8
*USGPO: 1984-759-102-10637
------- |