EPA-600/3-77-115
December 1977
Ecological Research Series
wi*~W^3*< *+'£**'*• * *J?>?: *v>:% > -^* 1 QIC
N^
^tD ^
rJtk^ % * V *«' ,"i?i s V< % * - ";%' v * * >. - -
glf|c:;s^V'< :>;<^|^''t%<- *x;'^, M% *." - . r
i%^^«->^§^:'v*-'- n* •% .;^'-f %.' , -
KWvv:tK^^K; *** -'?-:- *\.;,
*Jf^K41*f^;£>/,":>'/r ,,;* *,: '*' '*v// % :,- V-
-------
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 Socioeconomic 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-600/3-77-115
December 1977
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976 —
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part III. The Issue of Stratospheric Ozone Intrusion
V.A. Mohnen
State University of New York
Albany, New York
Contract No. DA-7-1936A
E.R. Reiter
Colorado State University
Fort Collins, Colorado
Contract No. DA-7-1305J
Project Officer
Basil Dimitriades
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORAORY
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.
11
-------
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.
In Part III V.A. Mohnen and E.R. Reiter discuss the issue of strato-
spheric ozone intrusion, i.e., whether ozone of stratospheric origin con-
tributes significantly to ground-level ozone buildup. The literature on
the subject of ozone intrusion is discussed and suggestions for further
research to resolve some of the questions raised are made.
111
-------
CONTENTS
ABSTRACT iii
FIGURES vi
TABLES x
INTRODUCTION 1
B. Dimitriades and A.P. Altshuller
THE ISSUE OF STRATOSPHERIC OZONE INTRUSION 3
B. Dimitriades and A.P. Altshuller
REVIEW AND ANALYSIS 7
V.A. Mohnen
Introduction 7
Ozone Intrusion into the Troposphere and Subsequent
Downward Transport to the Sink Region near and at
the Earth's Surface 14
Measurements Focusing on Stratospheric-Troposperic
Exchange Processes 20
Measurements of the Representative Ground-Level Ozone
Concentrations 33
Summary 63
Acknowledgements 64
Comments by Elmar Reiter 64
REVIEW AND ANALYSIS 67
E.R. Reiter
Abstract 67
Introduction 67
The Stratospheric Reservoir 68
Stratospheric-Tropospheric 73
The Life History of Stratospheric Intrusion Episodes 99
Conflicting Evidence from Direct Ozone Measurements 103
Conclusion 115
Suggestions for Future Research 115
Comments by Volker Mohnen 118
REFERENCES 125
-------
FIGURES
REVIEW AND ANALYSIS - V.A. Mohnen
Number Page
1 Tentative model of ozone fluxes produced by the
general circulation 8
2 Comparison between year-to-year variations in different
layers and of the total amount over Switzerland
1967-1972 9
3 Variation of integrated vertical ozone column with season
at tropical, temperate, and polar zones 10
4 North American ozonesonde network 11
5 Two-year average transport of ozone by transient eddies
from December to May 1962-1964. Transient eddy ozone
transport, 1962-1964 12
6 Average ozone mass mixing ratio distribution for spring
1963-1964 14
7 Time series of comparison between cyclone indes and shorter
period fallout fluctuations 17
8 Average tropopause heights for the period 1946-1956 at Swan
Island (17°N), Phoenix (33°N), North Platte (41°N), and
International Falls (49°N) 19
9 Vertical profiles of ozone mixing ratio (thin line) and
potential vorticity (heavy line) derived from Figure 12
for Bedford, Mass., and Tallahassee, Fla., 1200 GMT 24
April 1963 21
-10
10 Potential vorticity (contoured at intervals of 100 x 10 cm
sec (°K gm"-'-) computed from Figure 2 and 8 activity of
SR90 (dpm/KSCF) 22
11 Vertical cross-section of potential vorticity in units of
lO"10 cm sec deg g'1 23
12 Trajectories on the 300°K isentropic surface 24
VI
-------
Number Page
13 Heavy line is monthly mean variation of GASP ozone at
11-12 km, 36-42°N from March, 1975, through March,
1976 28
14 Solid lines are "zonal" means of ozone (ppbv) at 10°
latitude belts for combined March data (1975 and 1976) .
15 Vertical profiles of the global vertical ozone transport . . 32
16 Annual variation of monthly average values of daily maximum
ozone concentration at various places 37
17 Probability of the occurrence of the daily maximum of ozone
near the ground over hourly intervals 41
18 Correlation of mean ozone values measured within the Prandtl
layer (2.5 - 10 m) with ozone concentration of the layer
between 97.5 and 105 m 42
19 Monthly means of tropospheric ozone, derived from near
surface registrations of project TROZ 43
20 Variation of ozone mixing ratio with latitude for March and
May 44
21 Mean residence time of ozone in the troposphere assuming
destruction at the surface to be the only sink 47
22 Frequency distribution of ozone concentration at Whiteface
Mt. during 1974 51
23 Ozone trend and potential temperature trend (850 mb) at
Whiteface Mt 53
24 Isentropic analysis and ozone concentration - Whiteface Mt.-
July 1975 54
25 Correlation of tropospheric ozone mixing ratio for a day
T with mixing ratio for day T-l 58
26 Correlation of tropospheric ozone mixing ratio for day T
with mixing ratio for T-2 59
27 Secular variation of tropospheric ozone concentration over
North America (ng/g) 61
28 Mean, monthly ozone mixing ratios obtained in the upper
troposphere by participating GASP airliners 62
-------
FIGURES
REVIEW AND ANALYSIS - E.R. Reiter
Number Page
1 Monthly mean total ozone amounts at Arosa, Switzerland
(46.5°N, 9.4°E) 69
2 Comparison pf the smoothed sunspot number with the total-
ozone variation in regions with the longest records
and most stations 70
3 Temporal variation in total ozone in West Europe by season . 71
4 Worldwide total ozone as a function of season and latitude . 72
QO
5 Stratospheric inventory of Sr 75
6 Pole-to-pole cross-section of vertical ozone distribution
in nb (1 nb = 0.1 N/m2) 77
7 Vertical cross-section of potential vorticity (10-10 cm
sec deg/g), Zr^5 activities (dpm/scf), and ozone mixing
ratios (10~7 g/g) 78
8 Mean meridional circulation (mass flow in units of 10^ g
_ 1 j
sec ) for the four seasons 80
9 Schematic three-dimensional view of mass flow from strato-
sphere to troposphere near a jet stream 85
10 Time series of comparison between cyclone index, C, and
shorter period fallout fluctuations 87
11 Daily values of energy by mode and for the total of available
potential plus kinetic energy from July 1967 through
June 1968 88
12 Mean ozone distribution for March-April (solid lines, slanting
numbers, yg/g) and Sr^ distribution for May-August 1963
(dashed lines, vertical numbers, dpm/1000 SCF) 90
13 Sr90/0 ratios as a function of time 91
14 Mean fallout, January-June 1963, pCi/m3 94
15 Number of days with fallout >^ 10 pCi/m3, 1963 95
16 Maximum 24-hour fallout 1963, pCi/m3 96
VJ.11
-------
Number Page
17 Percent frequency distribution of ozone concentrations
(ppb) observed at Zugspitze, Germany, from August
1973 to October 1975 97
17a Mean seasonal variation of tropospheric ozone from airplane
measurements for the areas 0° - 25°S and 0° - 25°N ... 98
18 Trajectories on the 300°K isentropic surface 101
19 Difference between maximum hourly ozone concentrations (ppb)
and daily mean concentrations as a function of daily
mean concentrations, observed at Zugspitze (3000 m
above MSL) between August 1973 and February 1976 .... 105
20 100-mb map, 7 January 1975, 00 GMT 107
21 Average monthly ozone concentrations recorded at summit of
Mount Whiteface 108
22 Ozone concentrations at Whiteface and Pack Forest from
August 6, 1973 to August 17, 1973 109
23 Hourly ozone averages at Whiteface, Pack Forest, and Glens
Fall sites for July, 1973 109
24 Comparison of ozone concentration at Whiteface site with that
at Utsayantha and Syracuse sites for the first 17 days
of August, 1973 110
25 The mean maximum mixing depth for January and July 112
26 Measurements of ozone and other meteorological problems at
Hohenpeissenberg (977 MSL) , Germany 113
IX
-------
TABLES
REVIEW AND ANALYSIS - V.A. Mohnen
Number Page
1 Summary of GASP Data 26
2 Ozone Mixing Ratio 30
3 List of Stations of Project TROZ 39
4 Average Mean a , Amplitude a and Phase a , of
Annual Wave and Their Mean Statistical Error 45
5 Calculation of Global Sink 48
6 Classification of Critical Ozonagrams by Vertical
Distributions (> 0.10/ig-g -1 below 500 mb) 56
REVIEW AND ANALYSIS - E.R. Reiter
Number Page
1 Annual Mass Flux from Stratosphere, in Percent
of Mass of One Hemispheric Stratosphere 73
2 Mass Flux from Stratosphere to Troposphere
Accomplished by Hadley Cell Circulation 76
3 Percentage Contribution of Particular Nuclides in
Total Monthly Radioactive Debris Measured
in Rainfall at Westwood, New Jersey 92
4 Decrease of P Along 12-Hour Trajectory Segments 100
5 Hourly Ozone Concentrations (ppb), Zugspitze (Germany)
on January 8 and 9, 1975 106
-------
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.
XI
-------
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 non-EPA, non-industry scientists with extensive ex-
perience 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 of-
fering 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 stratospheric ozone intrusion. 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, along with the experts' comments on each other's reports.
-------
THE ISSUE OF STRATOSPHERIC OZONE INTRUSION
Basil Dimitriades and A. Paul Altshuller
In a broad sense, the question at issue here is whether ozone of strato-
spheric origin contributes significantly to the ozone problems observed in
urban and rural areas. Aside from the possibility that stratospheric-tropo-
spheric exchange contributes directly and significantly to ground-level ozone
buildup, stratospheric ozone has also been proposed to have a "reaction-
trigger" function that accelerates and enhances photochemical oxidant forma-
tion from hydrocarbon-NO precursors. The stratospheric ozone intrusion ques-
X
tion is part of the broader question regarding the magnitude and extent of the
ozone problem caused by natural causes, which in turn is a part of the issue
of achievability of the National Air Quality Standard for Oxidants (NAQS-O ).
X
To further explain the interest in the stratospheric ozone question, it should
be clarified and stressed here that this question, in fact the entire issue of
achievability of the NAQS-O , has no bearing whatever upon the justification
of the NAQS-O ; such justification is based strictly on health effects con-
X
siderations. The stratospheric ozone question needs to be answered only for
the purpose of more accurately estimating the benefits to be derived from
anthropogenic emission reduction.
Evidence interpreted to show accumulation of stratospheric ozone within
the troposphere varies widely in type and degree of directness. Thus, high
levels of ozone were measured in the upper troposphere near tropopause dis-
continuity points (2), evidence that attests to stratospheric origin most
directly. On the other extreme, ground-level oxidant buildup in some in-
stances was attributed by investigators to stratospheric intrusion only be-
cause these investigators did not have or would not accept any other explana-
tions (3). Overall, direct, unequivocal evidence on the impact of strato-
spheric ozone intrusion upon tropospheric air quality is lacking, and for this
-------
reason it may be expected that the viewpoints and interpretations of evidence
expressed to date reflect to some — perhaps substantial — degree a subjective
judgment.
At present, a realistic assessment would suggest that the extent (by
area), intensity (by concentration), and frequency of occurrence of strato-
spheric ozone buildup at ground level, all vary widely so that single answers
and answers to all of the questions that constitute the issue cannot be given.
It would, therefore, be more productive to define and offer as the subject of
this review only those that are most relevant to the oxidant control strategy
issue and receive substantial research attention. These questions are pro-
posed here to be as follows:
1. Accepting that intensive stratosphere-troposphere exchanges do
occur at tropopause discontinuity points, what is the extent,
frequency, duration, and spatial/temporal predictability of such
occurrences?
The terms "extent" and "spatial" here refer to areas at high altitudes, that
is, near the tropopause, not at ground level. While quantitative answers are
not expected, at least, a judgment should be made whether such exchanges are
sporadic, unpredictable incidents causing local ozone accumulations or are
significantly extensive and predictable. Main interest, of course, is in
occurrences within the U.S.
2. Accepting that localized high concentrations of stratospheric
ozone can occur in the upper troposphere, what fraction of such
ozone is expected to reach ground level
(a) under meteorological conditions most conducive to downward
transport, and
(b) under meteorological conditions most conducive to photochemical
oxidant formation?
The questions asked here, in essence, are again whether or not stratospheric
ozone excursions to ground level are sporadic, unpredictable incidents causing
-------
only local, short-lived (e.g., a few hours) ozone accumulations, and whether
or not such excursions are likely to occur during smog episode periods.
The question concerning the possible "reaction-trigger" function of strato-
spheric ozone is not raised here because, thus far at least, it has been a sub-
ject of speculation only; no relevant evidence apparently exists, except for
a few as yet unreported smog chamber experiments. Nevertheless, comments from
the reviewers on this question are welcome.
-------
REVIEW AND ANALYSIS
Volker A. Mohnen
INTRODUCTION
The fundamental aspects of the "classical hypothesis of the ozone cycle"
are:
• Photochemical production in the stratosphere, mainly at low
latitudes (ozone source region, above 20-km altitude) ;
• Poleward flux through stratospheric general circulation;
• Intrusion into the troposphere through stratospheric-tropospheric
exchange processes;
• Mixing in the troposphere;
• Ozone destruction within the planetary boundary layer and at the
earth's surface (sink region).
These mechanisms are shown diagrammatically in Figure 1 (4) . Processes 1 and
2 are not the subject of this limited literature assessment.
A reasonable hypothesis of the three-dimensional ozone distribution in
the stratosphere and its variation with time presently exists. With a few
exceptions, this knowledge is not yet accurate enough to definitely describe
ozone fluxes, and the flux-producing mechanisms. Hence, we cannot accurately
predict the total ozone* content (mainly stratospheric ozone) as a function of
latitude and season.
* The total ozone content of the atmosphere over a fixed point is given as the thickness of the pure ozone layer that would
be obtained if all ozone in a 1 cm2 vertical colvunn were concentrated at normal temperature and pressure (NTP) . It i»
usually expressed in units of one thousandth (10~3) of a cm and is equal to 2.687 x 1016 molecules/cm2 column. Thi« unit
is also called the Dobson unit (DU) . Vortical ozone distribution data iray be presented in terms of ozone density P3
measured in pg/ra3. The conversions to other unite and ozone parameters are as follows:
molecules -m~3 « 1.255 x Id16 x PS (pg-nT3)^
partial pressure in nanobars = 1.732 x 10 3 x T("K) x PS (pg-m 3)
mixing ratio in pg-g'^12.871 x icf 3x p3_(pg-m"3) x T(°K) )/P(mb)
" ~
_
m atra - cm km"1 = 4.67 x 10~2 x pa tug-in 3)
7
-------
.'IVph...
V.r!tJ(l,tl
^"
V
6
' *v
X / "'-•/
• K'oin
\?\ O.ont^-v* S,,0)OJ he,i(.
v^ ^/
> Region /
<.,'>/. >
l.v
Tropopousa
Ozont destruction near
90- €0" 30" O- SO" 60- 90"
Figure 1. Tentative model of ozone fluxes
produced by the general circulation.
Source: "The Ozone Distribution in the Atmosphere,'
H.U. Dutsch. Can. J. Chem., 52, 1974.
The best available long-term record of total ozone (beginning in 1933) is
from Arosa, Switzerland 46.5°N, 9.4°E (see Figure 2). Of particular interest
in Figure 2 is the total ozone amount in the 125-250 mb layer since this layer
constitutes a potential "ozone reservoir" for the troposphere. The systematic
natural variation of total ozone with season and with latitude (compiled
mainly from ozonesonde data) is given in Figure 3. The seasonal variation is
clearly dominated by transport processes leading to the late winter-spring
maximum. Results from the ozonesonde network, established by the Air Force
Cambridge Research Laboratories in January 1963 (see Figure 4), indicate that
the northward transport of ozone across middle latitudes over North America
occurs predominately in the lower stratosphere and just above the tropopause.
The average flux strength observed in the summer and fall seasons is less than
one-third of the average winter and spring transport. Figure 5 shows the 2-
year average values calculated for the winter and spring seasons for three
stations near 40°N. The indicated eddy flux diminishes sharply above 16 km to
rather small values. Also shown in Figure 5 is the 2-year average temporal
eddy flux for the winter and summer half years. The overall flux strength is
much weaker during the less vigorous and less disturbed circulation regime in
the summer and fall seasons. If the indicated transports are indeed representa-
tive of the average hemispheric flux, the total northward ozone transport
8
-------
across middle latitudes would be 1.4 x 10 tons for the cold half year and 0.4
9
x 10 tons for the warm half year (5).
I3S8 I9C9 1970
1 r
A /3I.2 -62,5mb
?1C
Figure 2. Comparison between year-to-year variations
in different layers and of the total amount
over Switzerland 1967-1972. (From Ref. 14)
Source: "The Ozone Distribution in the Atmosphere,"
H.U. Dutsch. Can J. Chem., 52, 1974.
-------
0.5
2
o
2
ID
DC
UJ
UJ
2
O
fvj
O
MAR JUNE SEPT
DEC
Figure 3. Variation of integrated vertical ozone column
with season at tropical, temperate, and polar
zones.
Source: Environmental Impact of Stratospheric Flight.
Biological and Climatic Effects of Aircraft Emissions
in the Stratosphere. National Academy of Sciences,
Washington, D.C., 1975.
10
-------
Figure 4. North American ozonesonde network.
Source: "Ozone and atmospheric transport processes,"
W.S. Hering. Tellus XVIII (1966), 2.
11
-------
3aniij.iv
in
0)
tn
•rl
•o
•0
(U
-P
c
0)
•H
M
c
10
4J
^1
.Q
OJ
C
o
N
0
M-t
o
-P
^
a
en
C
id
J^
-P
0)
&>
id
tt)
id
^_|
rtf
(1)
^i
1
o
EH
H
(Tt
1
O
H
C
-H
0)
V-l
id
en
CD
3
iH
id
•
^*
VD
cn
rH
1
CTi
rH
>1
id
S
O
M
0)
•i
0)
o
a
6
o
MH
S
CT*
, — 1
1
CN
0
a
CO
c
^
4J
i
Tl
•d
0)
4J
c
(U
•H
en
C
id
^
EH
t
CN
U
rH
O
0)
en
•
CN
i
g
U
rH
1
U
cu
en
c
&
T
0
•-H
C
•H
.C
^|
d)
ffi
•
w
•
&
=
cn
CD
cn
en
<1J
o
0
Pj
^_J
^1
0
cn
C
id
4J
•
O CN
•rl
0) ~
cu ^o
en CT>
0 i-H
id H
H
•O H
id x
Q) en
O rH
N rH
O Q)
= EH
a)
u
o
to
12
-------
The average seasonal distribution of ozone mixing ratio along a cross-
section extending from the Canal Zone, Panama, to Thule, Greenland, is shown
in Figure 6 as derived from a total of 421 ozonesonde ascents obtained from
the AFCRL in 1963 and 1964. The average tropospheric ozone amount is around
0.05 yg/g with a small but significant maximum at 30° to 40°N. Another im-
portant result derived from the AFCRL ozonesonde network is that the potential
vorticity* computed from monthly-mean cross-sections is positively correlated
throughout the lower stratosphere with the bimonthly mean values of ozone
mixing ratio (6).
These general conclusions can be drawn from this rather limited data:
In the latitude range 40-50°N (i.e., the area covering the U.S.
continent), the northward transport of ozone occurs mainly in the
lower part of the stratosphere.
The average flux strength for "winter-spring" is about three times
that for "summer-fall" in the latitude range 40-50°N.
The average meridional flux strength and the depth of the total
ozone column exhibit a sinusoidal behavior during the year.
The mean and the amplitude of both are latitude dependent.
Potential vorticity and ozone concentration exhibit a positive
correlation in the stratosphere.
* Potential vorticity (P ) is a quasi-conservative scalar whose large values
are generated in the middle stratosphere by diabatic radiative processes.
P = -Q — [cm*s«deg-g *] with Q the vertical component of absolute vor-
0 Z oP Z
ticity; Q the potential temperature; and P the pressure. Potential vorticity
is conserved in adiabatic flow. Values of P > +100 x 10 10 cm-s*deg*g * are
characteristic of stratospheric air. Potential vorticity is destroyed in
the troposphere by diabatic mixing, overturning, and frictional dissipation
at the ground.
13
-------
MEAN OZONE MIXING RATIO
FOR MARCH.APR1L.MAY (1963, 1964)
30-
28-
26 -
24-
22-
20-
14-
12-
10-
8-
6-
4-
2-
0
SO'
80-
THULE
TO1
60' | 50'
GOOSE BAY
| 40*
BEDFORD
J3O*
20*
•0*
TALLAHASSEE GRAND Tl«K CANAL ZONE
Figure 6. Average ozone mass mixing ratio distribution for spring 1963-1964.
Values are in yg/g or ppm.
Source: "Ozone and atmospheric transport processes," W.S. Bering. Tellus
XVIII (1966), 2.
OZONE INTRUSION INTO THE TROPOSPHERE AND SUBSEQUENT DOWNWARD TRANSPORT TO THE
SINK REGION NEAR AND AT THE EARTH'S SURFACE
The third, fourth, and fifth processes, in the initial hypothesis, will
now be discussed in some detail. Existing literature references to these
processes can be classified into two main categories:
Those describing measurements of the stratospheric-tropospheric
exchange processes with emphasis on identifying and quantifying
these processes. Stratosphere to troposphere ozone fluxes are
then calculated from these measured values.
Measurements of the representative ground-level ozone concentra-
tion ("tropospheric background"), its diurnal, seasonal, and
secular variation, and of the ozone destruction rates near and
at the earth's surface. Stratospheric-tropospheric ozone fluxes
can then be calculated from the rates of change in concentration.
14
-------
Since ozone transport is achieved by means of mass transfer between
stratosphere and troposphere, it is imperative to have some quantitative mass
transfer estimates in conjunction with the ozone mixing ratio for these strato-
spheric air mass intrusions. The following processes are responsible for
stratospheric-tropospheric mass exchange (7):
A - Large-scale eddy transports, mainly in the jet-stream region;
B - Seasonal adjustments in the height of the mean tropopause level;
C - Mesoscale and small-scale eddy transport across the tropopause;
D - Organized large-scale quasi-horizontal and vertical motions
expressed by the mean meridional circulation.
Referring to process A, transport occurs when the boundary between the
stratosphere and the troposphere deforms (in this case the tropopause is de-
fined on the basis of potential vorticity), becomes vertical in the core of
the jet stream, and then passes beneath the jet core. Reed and Danielsen (8)
showed that the folded structure could be identified by its large values
(stratospheric values) of potential vorticity and used the term "tropopause
folding" to describe the process. Mass transports from the stratosphere were
computed for several case studies of large-scale cyclogenesis in which the
three-dimensional trajectories were determined from isentropic analysis.
Danielsen (9) concluded, from these cases, that tropopause folding was an
integral part of cyclogenesis. Therefore, the net seasonal and annual trans-
port of mass from the stratosphere could be estimated by multiplying the mass
transport per cyclogenesis by the number of cyclogenetic events. Danielsen
calculated a mass transport value from the stratosphere to the troposphere of
(4.0 ± 0.5) x 10 g during a 36-hour period over North America from January 2
to 3, 1958. A similar calculation for a cyclone of average intensity over
North America from November 22 to 23, 1962, involved a total transfer of 6 x
10 g from the stratosphere (10).
The amount of mass transported during these two events is in relatively
good agreement. [Note: the total stratospheric mass in the northern hemi-
20
sphere is of the order of (4.5 ±0.5) x 10 g.] Danielsen1s estimate of 4.3
20 -i
x 10 g*year for the total northern hemispheric outflow due to large-scale
15
-------
eddy transport implies that a mass comparable to the entire northern hemispheric
stratosphere is exchanged in 1 year. Reiter based his estimate for this
annual outflow on a simple cyclone index proposed by Mahlman (11) (see Figure
7). During 1963 and 1964, the estimate of cyclonic activity yielded 22 and 23
respectively in the sector 70°-180°W. Reiter computed an annual transport of
17 19
6 x 10 x 22.5 x 3 = 4.5 x 10 g for the polar front jet-stream belt, at
40°-60°N in the northern hemisphere. Reiter assumes twice that amount, of
exchange 800 x 10 g, as a reasonable value for large-scale eddy transport
from the stratosphere to the troposphere on an annual basis for the entire
northern hemisphere. This is approximately 20% of the entire northern hemi-
spheric stratosphere.
The numerical discrepancy (^ 90% according to Danielsen versus ^ 20%
according to Reiter) is a consequence of inferring all this from a few events
(well characterized and analyzed cyclogenetic case studies) and the global
frequency, intensity, and duration of cyclogenesis. On the basis of strontium-
90 concentrations measured in the lower stratosphere during 1959-1960 and
1962-1963, Danielsen (12) has deduced a mass outflow rate from the entire
northern stratosphere:
= a, + a cos Q (t - a.) (Eq. 1)
Qu , — - _L •& T j
outflow
where a : annual mean outflow rate [g yr ]
a : amplitude of annual mean [g yr ]
a : phase function
T: 1 year
t: varies from 0 to 1 year
The numerical values are
(~) = [3.6 + 1.8 cos 2T (t- |^-)] x 102° (g-yr'1) (Eq. 2)
outflow
16
-------
a>
31 s:
a
•8
4
71
3 8
<,- -^
. V i
'u
L ^-^
T
.•.-...,
{-
;3
i^^
a a § o „ u
H tH -H
r-
0)
3
Cn
•H
0)
u
3
O
17
-------
The maximum outflow in mid-April is three times the minimum in mid-October.
The annual outflow is about -r1-^- = 0.80, i.e., about 80% of the total northern
4* • -J
hemisphere's stratospheric mass. (Danielsen's estimate is consistent with the
90
total budget of atmospheric Sr as determined during project Springfield: the
total amount of bomb debris injected into the stratosphere, the total amount
present in the stratosphere, and the total amount deposited at the earth's
surface are balanced by his analysis.)
On the basis of limited data available, there still remains large un-
certainties on the total mass flow associated with large-scale eddy transport.
As a consequence, it will be rather difficult to obtain reliable seasonal and
annual estimates on the amount of ozone transported into the troposphere as a
result of cyclonic activity. The foregoing analysis indicates, however, that
this process plays a main role in transporting ozpne from the stratosphere into
the troposphere, as will be discussed later.
Relative to process B, Reiter (7) calculated the net decrease in strato-
spheric mass from winter to summer as 400 x 10 g, which is roughly 10% of
the mass of the northern hemispheric stratosphere. His estimate is based
mainly on seasonal variation of average tropopause heights as function of
latitude (see Figure 8). There are no estimates available in literature on
the amount of ozone entrained into the troposphere due to the seasonal ad-
justments in tropopause levels. (Some new evidence is presented later.)
As an estimate of the magnitude of process C, the mean meridional circula—
tion transport has been estimated by Reiter (7) to amount annually to 1633 x
10 g, which is 38% of the mass of the stratosphere in the northern hemisphe
This meridional transport is mainly Hadley cell circulation.
The tropical segment of the Hadley cell is apparently very effective in
introducing large amounts of tropospheric (ozone poor) air into the strato-
sphere. Continuity is retained as roughly the same amounts of stratospheric
air will return into the troposphere in middle and high latitudes (7). Ex-
perimental evidence on the ozone intrusion into the troposphere attributable
to the descending branches (for example the subtropical branch) of the mean
18
-------
meridional circulation is unavailable. Unlike the clear-cut tropopause folding
events, which can be analyzed from routine radiosonde data, the mean meridional
"circulation" is a parameterized concept of a very complex circulation system.
Examining process D, the mesoscale and small-scale eddy transport across
the tropopause has been calculated by Reiter (7) to account for less than 1 to
5% of the total flux between stratosphere and troposphere (assuming a vertical
2 2-1
eddy diffusivity value of 10 cm «s ). Ozone trai
eddies is negligible, as will be discussed later.
2 2-1
eddy diffusivity value of 10 cm «s ). Ozone transport due to these sub-grid
£• 40
r, 35
JFMAMJJASOND
MONTH
Figure 8. Average tropopause heights for the period 1946-1956 at
Swan Island (17°N), Phoenix (33°N), North Platte (41,0N) ,
and International Falls (49°N) [Staley, 1962].
Source: "Stratospheric-Tropospheric Exchange Processes," E.R. Reiter.
Reviews of Geophysics and Space Physics, 13, 4, Aug. 1975.
19
-------
MEASUREMENTS FOCUSING ON STRATOSPHERIC-TROPOSPHERIC EXCHANGE PROCESSES
Measurements reported in the literature will be discussed as they relate
to stratospheric ozone intrusions or as they relate to information on the
stratospheric-tropospheric ozone flux.
Danielsen has presented overwhelming evidence that the mean potential
vorticity, radioactivity, and ozone mixing ratios can be used as tracers for
stratospheric air intrusions into the troposphere. Although each tracer has
a different mid-stratospheric source, they develop a positive correlation in
the lower stratosphere as a result of active adiabatic mixing (see Figures 9
and 10). Additional examples of fully analyzed tropopause folding events are
shown in Figure 11 and Figure 12. The most recent study of three examples of
tropopause folding (made in Project Duststorm during April 1975) (13) reinforced
the conclusion (as did the flight data obtained during the operation of Project
Springfield), that ozone-rich air is transported into the troposphere with
each major cyclonic development. Ozone mixing ratios have been measured in
Project Duststorm down to 20,000 feet, but no attempts have been made to
follow the intruded stratospheric air as it mixes downward and (eventually)
impacts on the ground. Such direct experimental evidence has not been re-
ported in the literature.
17
Assuming a stratospheric mass outflow of (4 to 6) x 10 g per cyclonic
event and an ozone mass mixing ratio of 1.3 x 10 g/g in the lower strato-
sphere would yield an amount of (5.2 to 7.8) x 10 g ozone transported into
the troposphere during each cyclonic event. If one follows Reiter's argument
in arriving at the northern hemispheric mass flux due to large-scale eddy
transport as outlined under A above, the annual outflow of ozone would amount
to: 1.3 x 105 (g'g1) x 6 x 10 (g) x 22.5 (cyclonic events for 40°-60°N,
14
70°-180°W) x 3 (for full circumference) x 2 (for all latitudes) = 1.05 x 10
g of ozone resulting from process A, and 1.3 x 10 (g-g ) x 1633 x 10 g =
14
2.12 x 10 g of ozone resulting from transport process B outlined above.
20
-------
POTENTIAL VOKTICITY
10 25 50 100 250 500 1000 2500 500O
10 25 50 IOO gSO 5OO IOOO 2500 50OO
L BEDFORD
24 APRIL 1953
1224 GMT
LTALLAHASSEE
24 APRIL I9S3
U2I7CMT
IOOO
aOI GD2 COS 0.1 O2 OS
5 10 20 OOI DOE OO5 01 0.2
OZONE MIXING RATIO (I0~ss«™/sram)
O5 I 2 5 10 2O
Figure 9. Vertical profiles of ozone mixing ratio (thin line) and potential
vorticity (heavy line) derived front Figure 12 for Bedford, Mass.,
and Tallahassee,. Fla., 1200 GMT 24 April 1963.
Source: "Stratospheric-Tropospheric Exchange Based on Radioactivity, Ozone
and Potential Vorticity," by Edwin F. Danielsen. J. of the Atmo-
spheric Sciences, Vol. 25, pp. 502-518.
21
-------
41
4*
M
M
M
U
M
H
II
H
M
M
II
H
14
II
W
APRIL 1963 002
•;*V9*CftviTf
41
40
M
II
14
It
M
II
II
14
II
l«
II
II
14
II
Fifura 10, Potantial vortioity (contoured at intarvali of 100 x 10 em
(dpm/KBCF)
••o (*x gm" ) computed from Figur* 2 and 0 activity of Sr
Sourott "Stratoiphario-Tropoiphtrio Bxahanga Baitd on Radioactivity,
Oiona and Potential Vortioity," B, Danialatn, J, of tht Atmoiphario
Soitnoai, 25.
-------
"V
tOQ-i
POTENTIAL VOD.TiCITY
OZONE MIXING RATIO
• ZIRCONIUM 95 GAMMA ACTIVITY
Figure 11. Vertical cross-section of potential vorticity in units of 10
era sec deg g . Zirconium 95 activities in units of dpm (SCF)
are plotted above the sampling points, with additional isoline of
ozone mixing ratio in units of 10 g g (dashed lines).
Source: "Observed Distribution of Radioactivity, Ozone,'and Potential
Vorticity Associated with Tropopause Folding," E. Danielsen, et al.
J. of Geophysical Research, 75, 12, April 20, 1970.
23
-------
PLOTTING MOBIL!
POTINTIAL VONTICITV
MONTOOMMVITKIAM FUNCTION
PMIIUAI
TMICKNIII OF ITAILI LAVIN
Figurt 12. Trajtotorist on tht 300'K iatntropio aurfaot, Daahtd linta art
Novtmbtr 22, 1962, 0000 to 1200 UT» full linta with arrowa art
Novtmbtr 22, 1200 UT, to November 23, 0000 UT, Valuta of pettn-
tial vortioity (units of 10* on • dtg/g), of Montgomtry atrtam
function (unite of 10' am2/!2), of prtaaurt of tht 300»K aurfaot
(millibara), and of thiokntu of tht itablt laytr (milliban) art
tnttrtd according to tht plotting modtlr slanting nutnbtrt art for
Kovtmbtr 22, 0000 UT, and vtrtieal nutnbtra art for othtr map
timti. Tht otnttri of tht hatehtd band mark boundariti on
Novtmbtr 22, 1200 UT, and Novtmbtr 23, 0000 UT, of itratoiphtrio
air from tht tropopauat Itvtl that rtaohti tht ground ovtr tht
•outhtrn Unittd Btatti [Rtittr and Mahlman, 1964],
Sourcti "Stratoiphtrio-Tropoiphtric Bxehangt Prootiati," S.R, iuiittr.
Rtvitwi of Qtophyaiot and Spaot Phyaioa, 1^, 4, Aug. 1975.
24
-------
Hering and Borden (14) have derived an expression for the ozone mass
mixing ratio in the lower stratosphere:
X = [1.3 = 0.3 cos — (t - jj-)] x 10 g'g (Eq. 3)
The product of Danielsen's mass outflow rate expressed in Equation 2 and
14
Equation 3 predicts an annual outflow of 4.7 x 10 g of ozone for the
northern hemisphere.
Summation of all these arguments produces these estimates for the annual
outflow of ozone from the stratosphere over the northern hemisphere:
14
• Danielsen (large scale cyclogenesis): 4.7 x 10 g of ozone
• Reiter (large scale cyclogenesis and Hartley circulation):
14
3.2 x 10 g of ozone
Note that neither of these values of vertical ozone transport across the
tropopause boundary were measured directly. Note also that the assumed ozone
mass mixing ratio of 1.3 x 10 g*g for the lower stratosphere enters criti-
cally in these annual flux calculations. Taking all uncertainties into account,
14
the true annual flux is probably in the range of 2 to 8 x 10 g of ozone, a
significant portion of this being introduced in the latitude belt 30° to 60°N.
Ozone measurements taken from commercial airliners (NASA Global Atmo-
spheric Sampling Program, GASP) offer an opportunity for establishing an ozone
climatology along major flight corridors. Table 1 shows pertinent data
statistics for the sampling period March 1975 through March 1976. Nastrom
(15) has presented a comprehensive analysis of these GASP data (archived on
tapes VL0001-VL0004). The bulk of the flights were within the continental
United States, from the mainland to Hawaii, and from the United States to
Europe. The in situ ozone mixing ratio, measured by an ultraviolet absorption
photometer, is reported every 5 minutes (i.e., about every 75 km, but every
other observation is missed because the instrument is in a calibration mode).
25
-------
Most observations were taken between 10-12 km altitude. Flight level atmo-
spheric pressure, temperature, wind direction and speed, and an indication
from the aircraft accelerometer of turbulence occurrence, are reported with
each ozone observation. Supplementary parameters were computed for each ozone
observation from the NMC northern hemisphere grids of isobaric height fields
and tropopause pressure fields. Tropopause separation pressure (P
•^ tropopause
P • ffc)t geostrophic winds and vorticity, potential vorticity, and the
algebraic sign of the vertical velocity were also computed.
TABLE 1. SUMMARY OF GASP DATA
Month
Mar 1975
Apr
May
Jun
Jul
Aug
Sep
Oct
Dec
Jan 1976
Feb
Mar
Total
Flights
57
26
66
35
3
16
23
25
10
36
54
39
Total
Obs.
1263
554
1625
908
78
434
579
716
326
1119
1435
1057
Latitude
Range
9N-61N
19N-47N
23S-47N
19N-45N
21N-41N
21N-47N
21N-43N
21N-43N
21N-45N
9N-61N
33S-43N
9N-57N
Longitude
Range
180E-180W
75W-159W
45W-114E
84W-159W
84W-159W
84W-159W
75W-159W
75W-159W
72W-156W
180E-180W
72W-114E
180E-180W
Source: "Variability and Transport of Ozone at the Tropopause from the First
Year of GASP Data," by G. D. Nastrom, Control Data Corp., Research
Report No.4, Feb. 22, 1977, Contract NAS 2-7807 for NASA-Lewis
Research Center.
26
-------
Nastrom compared mean ozone values from GASP flights with those from
North American ozonesondes and found them compatible, establishing confidence
for the NASA GASP data (see Figure 13). Monthly mean values of ozone and
potential vorticity are compared in Figure 14. As Danielsen, Hering, and
others have found, there exists a close correspondence of the two fields
(correlation coefficient = 0.95). Particularly notable are the apparent in-
trusions of ozone and potential vorticity below the tropopause near 40°N.
Nastrom has also calculated the vertical flux of ozone: all ozone observa-
tions taken within 50 hPa of the tropopause and north of 30 °N were sorted
according to the sign of the associated vertical motion. To estimate the
magnitude of the flux, the mean magnitude of the vertical velocity at the
tropopause has to be assumed, for which a value of |w| = 0.5 em's is
adapted. With the assumed |w| value, the average annual flux across the
~12 -2 ~1
tropopause amounts to 6.2 x 10 (g-cm 'S )
X 1.656 ,
cn
or 7.8 x 10 molecules cm -s . This vertical ozone transport estimate
reflects only the transport by motions whose wavelength is longer than about
700 km (large-scale eddy transport, Process A) .
The annual flux of 7.8 x 10 molecules -cm s compares very well with
the results of Fabian and Pruchniewicz (16) derived on the basis of surface
ozone measurements (see Table 2).
The transport of ozone by disturbances smaller than about 700 km (small-
scale eddy transport, Process C) was estimated by Nastrom by assuming the flux
is the product of an eddy diffusion coefficient and the gradient of ozone
across the tropopause. The diffusion coefficient at the tropopause used by
Cunnold et al. (17) — K = 3 x 10 cm s" — was adopted, and the gradient of
Z
ozone was estimated by finite differences of mean values of layers 50 hPa
thick and centered 25 hPa above and below the tropopause. The resulting
estimates of the diffusive flux (Process C) are only about 3% as large as the
corresponding fluxes by large-scale motions (Process A) (see Table 2). Similar
27
-------
-o
-U.
rQ
O
O)
i
CM
I
4->
a
8
> m
r-
rH
•d ~-
(U X
-H ~
£
8 -38
a!
1*
^
CO MH
^H
z
o
0)
>, ro
S g
as 5
QJ
^
S"
&S
g a
0) M
A MH
(
(0 2
nj o
^3 o
'S1
tn
to
C 4J 0)
•H «J t-H
X 3
4J 01 O
0 C
O *
o
-p
o
MH
r-
o
00
r~
|s
id
M =
EH •
*2
(0 Q
>i ft
•P CO
rH O
3
M
t5
e m
O -H
^ f-jT.
CO 0)
g5
I
O MH
Q
CM
•
A
pr, Ij
(1)
«• 4J
-a- c
.8
fi«
o
l_t Jj
SH ifl
ad)
to
O 0)
OS Bl
0)
O
I
H
MH a
o a
to
in
01
H r^
id m
ro
O H
•H I
•a «
o u
-H I
0) CO
- s m
C C CM
n O ro
in
z
d
S
01
-H
Q
4J 4-> fl
CO M rt!
id
w
• to
O 4-)
U n) M
^J ll I
-H Q)
sccu
O rH
• N A
a! O id
00
o
CVJ
CVJ
28
-------
KM
11.5-
10.5-
9,5-
8,5-
400
200 100
300 NA UK s 50
55 45 35 25
15
Figure 14. Solid lines are "zonal" means of ozone (ppbv) at 10° latitude
belts for combined March data (1975 and 1976). The dashed line
is mean tropopause location, and the dotted lines are "zonal"
means of potential vorticity (10~6 deg hPa~l s"1), for each
belt.
Source: "Variability and Transport of Ozone at the Tropopause from the
First Year of GASP Data," G.D. Nastrom. Control Data Corp.,
Research Report No. 4, Feb. 22, 1977, Contract WAS 2-7807 for
NASA-Lewis Research Center.
29
-------
O
H
a
H
x
H
s
I
o
CN
CN
I
(U
•H
in X
g
U
o
(U
H
335
MH fa o
•rl rH
Q O
to
"8 6rH~
in u i
id tn
XI m CN
X o
3 II
rH O
far— Q)
5 I-H
4->^ O
2 0 O
H
o
>i
4->
&>
C in
id c
CD CD i
g Q O
j?
Q)
C
O
N
O T3 C
M O
C Id -rl
id S 4-1
l££
CN
d
m
CN
m
Is-
ro
ro
o
ro
ro
ro
m
0)
4-J
CO • -H
•rH •
tn in rH
C i
U
4-) V-l
(U ro
tn in o
C XI H
•H 0
•C X
t i ll i
o o ro
D
O H II
Id Q)
T3 S
cu 3 c
4J C 0
M
0 0) TJ
W .5 0)
EH 0)
Id
id xt
pH •
,C <1> TO
M -H
O (U
ft T3 rH
O Q) MH
M 0)
EH 3 0)
>
0 0) -H
4-> M 10
id 3
id MH
0< 2: MH
fi o -H
O TJ
O ro
in 4->
SH id C
T3 CD
Cr> H
c >i id
•H rH £L(
X C
•HOC
g -H
N MH -rl
O O U>
1
CO
J^J
*• 13
ff 0
0 MH
g U
O r-
M Id O
MH 4-> CO
id r-
(1) Q 1
tn CN
3 rH
id o to
ft ^ <
o 4-1 2:
ft C
O O 4J
H U 0
EH (d
Q) • 4J
X! g C
"" 2 8
4-1 4-1
id in ^
id r^
0) 2 r-
C CM
O • rH
CJ CN
in CN
0 >i
4J rQ
M = (U
a^ fa
id
W 4-1 - •
C id ^3* i-t
Id Q 0)
M • 4-1
EH CM 0 C
tO !3 tD
'O ^ U
C d 4J
id ri ,c
MH O U
>i O ft M
-P CD IB
-H H P^ CD
rH nj tn
•H H o a;
-H q a)
£ r^ PI i— i
••
0)
o
3
O
LO
30
-------
results for the total mass transport across the tropopause have been obtained
2 2
by Reiter, as pointed out earlier (although Reiter used a K value of 10 cm
Nastrom's results are the first direct estimates of the ozone flux across
the tropopause (the main assumption being the representative |w| value).
The detailed mechanism whereby this flux occurs, such as done by Danielsen and
Reiter, cannot be derived directly from the GASP data because the NMC grid is
too coarse to resolve folds in the tropopause. This detailed transfer mech-
anism is not critical if the downward transport of ozone is associated with
large-scale motions and can be parameterized by a representative |w| value.
If one assumes that the ozone flux of 7.8 x 10 molecules*cm 's
(deduced from direct ozone observations at 10-12 km altitude, 30°-60°N and
mainly over the U.S.) is a representative average for the northern hemisphere,
14
it would yield an annual ozone mass flow into the troposphere of 4.9 x 10 g.
There exists some discrepancy in literature with respect to the relative
importance of ozone transport into the troposphere via large-scale eddies
(Process A, mainly cyclonic activity) versus mean meridional circulation
(Process D), which cannot be resolved at this time on the basis of available
experimental data. Some information is available, however, from the U.C.L.A.
General Circulation Model of Mintz and Schlesinger (18). This is a three-
dimensional, global atmospheric general-circulation model that extends from
the earth's surface to the stratopause. Its circulation, radiational heating,
and ozone photochemistry are fully coupled and interactive. Their results of
computations to date are:
In the zonal average, the largest vertical transport of ozone is
in the descending branches of the mean meridional circulation, but
in the global average, the vertical eddy transport of ozone is several
times larger than the vertical transport by the mean meridional
circulation (see Figure 15).
31
-------
.o
E
UJ
tr
:D
(f)
CO
UJ
IT
a.
MERIDIONAL
CIRCULATION
NET
1000
-60 -40
-20 0
TRANSPORT
20 40
(tons/sec)
60
Figure 15.
Vertical profiles of the global vertical
ozone transport. Positive values are
downward.
Source: "Ozone production and transport with the U.C.L.A. general circulation
model," Y. Mintz and M. Schlesinger. Proceedings of the 4th
Conference on the Climatic Impact Assessment Program. Edited by
T. Hard and A.J. Broderick. DOT-TST-75-38, 201-222, February
(1975).
32
-------
• The vertically-integrated, latitudinal transport of ozone in the
tropics consists almost entirely of a large-scale transport from
the summer to the winter hemisphere by the mean meridional circula-
tion, but in the middle and higher latitudes it is the small dif-
ference between a large poleward transport by the eddies and a
large equatorward transport_by the mean meridional circulation.
• The vertically integrated latitudinal transport by the eddies is,
itself, the small difference between a large eddy transport toward
the poles in the middle and lower stratosphere and a large eddy
transport toward the equator in the upper and middle stratosphere.
[Note: Mintz and Schlesinger1s model also shows the baroclinic waves for the
500 mb geopotential height field which move slowly eastward. By superimposing
the trough and ridge lines of these waves on the total ozone field, the model
shows that the maxima and minima of total ozone coincide, respectively, with
the trough and ridge lines of the 500 mb waves. Both the amplitude of this
zonal variation of total ozone and its phase relationship to the baroclinic
disturbances at 500 mb (and at 300 mb) are precisely what is observed in the
real atmosphere, as shown for example by Normand (19).]
MEASUREMENTS OF THE REPRESENTATIVE GROUND-LEVEL OZONE CONCENTRATIONS
Junge (20) presented a comprehensive assessment and analysis of existing
ozone data and arrived at a global ozone budget based on ground-level ozone
observations, on ozone surface destruction rates, and on estimations of
stratospheric-tropospheric exchange processes.
Junge's analysis is based on the assumption that tropospheric ozone in
the northern hemisphere and its seasonal variation can be approximated by an
expression of the form:
9 = a + a sin[co(t - a )] (tons) (Eq. 4)
33
-------
where a : annual mean for 0, the total tropospheric ozone content
a : amplitude
a : phase function
u: 2ir/T and T = 1 year
t: 0-1 year
To determine the parameters a and a« in Equation 4, it is imperative to
obtain representative values for the tropospheric ozone and its seasonal
variation.
Junge points out that data obtained at or near the earth ' s surface are
usually affected by ozone destruction at or near the ground. Average con-
centrations of ozone near the ground are thus of little value for general
considerations because they more or less reflect the microclimatological
conditions of the sampling site. The daily maximum values of ozone usually
occur around noon when vertical mixing is strongest and when the influence
of the surface destruction layer is minimized. It can be expected that these
values are approximately representative for ozone in the undisturbed tropo-
sphere in areas which are free of air pollution. In polluted air, ozone
destruction is of considerable magnitude within the air. In reducing atmo-
spheres, such as in the planetary boundary layer over central Europe, the
maximum ozone values are usually lower than in the layers aloft. Junge 's
careful search for proper ozone measuring sites to obtain tropospheric back-
ground ozone concentration was reinforced later by Dutsch (4) :
As the ground is thought to be the main sink for stratosphere-
borne ozone, it is hoped that the measurements of ozone at the
surface could give additional useful information on the atmospheric
ozone budget. This is, however, only the case if it is possible to
obtain from such continuous registration inference on the ozone
concentration of the lower troposphere above the planetary boundary
layer. Very careful selection of sites free from any pollution and
from influences of inversions is needed. Mountain stations would seem
especially useful but even there, disturbances by local circulation
effects are indicated. Careful comparison with available sounding
series may lead to useful concepts for finally evaluating destruction
near the surface in connection with surface structure on one side and
with stratospheric- tropospheric ozone transfer on the other side.
34
-------
Junge selected for his analysis three mountain stations:
Arosa (47°N, 1860 MSL)
Mauna Loa (20°N, 3000 MSL)
Srinagar, North India (34°N, 1700 MSL)
He calculated monthly average values of daily maximum ozone concentration
(see Figure 16). The annual mean concentration for these stations is taken
as 50 pg*m , which fixes the annual average for the total tropopause ozone
g
content a = 1.30 x 10 tons ozone. (Based on an average tropopause height of
12 km, and an ozone mixing ratio increasing linearly by a factor of two as
height increases from 1 km to 12 km.) The amplitude of the yearly variation
of ozone for the representative stations (Figure 16) is about ± 15 yg-m
8
which results in a. = 0.39 x 10 tons ozone. Junge observes from total ozone
column measurements (most of the mass of which is in the stratosphere) at
Arosa and Mauna Loa and from the calculated total northern hemispheric total
ozone (see Figure 16) that the phase of the surface ozone is somewhat delayed
against that of total (mainly stratospheric) ozone. He interprets this time
delay as the result of a limited lifetime of tropospheric ozone. Junge formu-
lates a budget equation with the following assumptions: ozone destruction
occurs primarily at or near the ground; the rate of ozone destruction, D, is
approximately proportional to the ozone concentration and thus proportional
to the total tropospheric ozone content, 0, as presented in Equation 4.
D = - C e (tons*yr )
o
(tropospheric lifetime T = C )
(Eq. 5)
From Equations 4 and 5 it follows that the rate of injection, I, of strato-
spheric ozone into the troposphere must have the form
I = C + C sin (tot) (tons«yr~ ) (Eq. 6)
where t = 0 is not identical to the beginning of the calendar year. The phase
35
-------
difference between I and 9 is taken into account by a in Equation 4. Combining
Equations 5 and 6 results in the budget equation
d6
— = I + D = C. +C_ sin (uit) - C 9 (Eq. 7)
at 12 o
with the solution
C C C , C j.
6 = — + — [(—)2 + 1] sinfoot - arc sin (— + 1) ] (Eq. 8)
C a) a)
o
Direct comparison of Equations 8 and 4 yields
C,
~ (Eq. 9)
o
C C
a = — (— + 1)~% (Eq. 10)
£ y
OJ W
C ,
a, = - arc sin (— + I)'* (Eq. 11)
3 o
o) or
from which C , C, and C_ can be calculated.
o 1 2.
Junge recognized that the delay time a (phase) is the most uncertain
value. It can be taken from observations if one assumes that the injection
rate I is in phase with the total ozone burden of the stratosphere, which is
only a very rough approximation. By evaluating the gross fission product
concentration in surface air averaged for the northern hemisphere according
to Equations 8 through 11 and taking for C the experimentally established
-1 °
value of 12 yr (the tropospheric lifetime of radioactive aerosol is of the
order of 1 month) yields a time delay of a = 0.9 month between I (fission
products) and 8 (fission products).
36
-------
20-
O AHOSA, SV.!7ZtPL.".D, 47 N, 1351
L * MA'JNA LOA, MAV>Ai;, I35N. I3M
| • SRIflAGAR, I\DU. 3" H, 1957-1360
I J F M A M J J A S 0 M D
I J F M A M J J A S 0 N D
f+ LITTLE 4^E»iCA. A>
r • I I L
I I I J_
J FMAMJJASONO
030
I i L _ L—I—J —J 1 1 —•—i—-h
J > M" A M" J J A S 0 N D
:r==rr 020
Figure 16. Annual variation of monthly average values of daily maximum
ozone concentration at various places, (a) Values for Arosa
1800 m above sea level (Gotz & Volz, 1951), Mauna Loa 3000 m
above sea level (Price & Pales, 1959, 1961) and Srinagar 1700 m
above sea level (Ramanathan, 1961). The value in April at
Mauna Loa is based on a few days only, (b) Values for three
consecutive years in Arosa, 1951 (Gotz & Volz, 1951) and 1952
and 1953 (Volz & Perl, 1961). For comparison the average
monthly concentration of long lived fission products in surface
air for the northern hemisphere obtained from the 80th Meridian
Network data (Lockhart, 1961). (c) Values for Ahmedabad 50 m
above sea level (Dave, 1961) and Little America 50 m above sea
level (Wexler et al., 1960). (d) Total ozone for Arosa (Gotz
& Volz, 1951) Mauna Loa (Price & Pales, 1959, 1961) and the
integrated average value based on data by Dutsch (1946) after
correction for newer absorbtion data (Dutsch, 1959).
Source: Junge, C.E.: "Global ozone budget and exchange between stratosphere
and troposphere," Tellus V XIV, 4 (1962).
37
-------
Since there appeared to be another time delay of 1 month between the
spring maximum of radioactive fallout products and ozone (see Figure 16),
Junge took the a value for ozone as 2 months. The numerical values for
the ozone budget are therefore (northern hemisphere):
Stratospheric ozone injection rate
I = [4.7 = 2.8 sin (cot)] x 1014 (g-yr"1)
ozone flux = 7.5 x 10 molecules*cm s
annual ozone mass flow into troposphere
14
= 4.7 x 10 g of ozone
average tropospheric residence time for ozone
T = C =3.3 months
t o
Junge was first to present a thorough analysis of the ozone budget based
on a carefully selected and representative tropospheric ozone data base.
Earlier estimates, also based on tropospheric ozone observations, are compatible
with Junge's results:
Kroening and Ney (21): ozone flux = 6 x 10 molecules cm s
Paetzold (22): ozone flux = (2.5 - 5) x 10 molecules cm s
Junge (20) proposed the establishment of a meridional network of repre-
sentative surface ozone recording stations, which should be maintained for
several years. This would provide better data for a (phase) and may also
show small systematic differences of the delay time, which can provide informa-
tion on the injection rate as a function of latitude and, perhaps, longitude.
This network was established in 1969 under Project "Tropospharisches
Ozon" by Fabian and Pruchniewicz (16). Permanent registration of ozone at
ground level was obtained from 16 stations, located on a meridional chain
between 69.5°N and 34°s, centered between 10 and 20 degrees longitude east
(see Table 3). The instruments were installed at sites with relatively clean
38
-------
N
O
OS
EH
EH
L)
W
1-3
O
C<
CU
Pn
O
CO
Z
O
H
EH
H
CO
Pn
O
EH
CO
H
J
•
ro
H
tJ]
CQ
H
c
O
-H
(fl
0)
o
MH
0
o
>r_l
EH
C
O
-H
; l
rfl
4J
in
tn
c
•H
fi
rd
4-1
C
•H
tfl
g
C
o
•H
4-1
3
4-1
•H
4-1
in
C
H
O
Tl
4J
-rH
4->
rd
r-3
fi
O
• H
rd
4-1
to
O
r-
4J cn
C rH
0)
in
CD C
SH rO
ft ^
1 1
O 00
r> 10
en en
rH rH
, .
en u
3 CD
< 0
X>
to
e
o g
EH 0
Xi
SH o
O O
rd w
$>
SH MH
CU 0
in
XI >i
O 4->
•H
rH in
rd SH
M eu
0 >
SH -H
3 C
< D
2 2
0 O
m
•
en ro
VO VO
c
0)
rH
:0 :tfl
in X
e T)
O CD
rH SH
B CQ
CN
f^.
01 4-1
rH C
CD
in
> CD
O SH
2 ft
1 1
O rH
[ — [ — '
en en
rH rH
• ^1
en
c -g
e w
•H O
0) ^
ft •
X 4J
H rd
g
C -rH
CD rH
1 1 t^J
rd 0
4-1 -H
CO CQ
Z Z
o o
O in
vo in
T)
C
tfl
H
SH
cu
CD 4->
in in
•H 0)
M J5
4J 4J
C C
in in
CD 4)
ft ft
1 1
en en
vO vO
en en
rH rH
^1 *
rH C
3 rd
-P
in
c
4-1 H
CD Cu
Z Z
O O
in
•
ro CM
in in
i^i
CD
C
CU rd
r^ C
0 -H
2 J
CNJ
t —
0i
H
>i
rrj
s
1
en
vO
cn
rH
•
a
(0
J_|
O
x:
o
u
4J
3
2
•b
e
3
•H
J^
O
4-1
(C -P
> w
^_| Cj
CD CD
CO -rH
Q TJ
O SH
0)
• -P
4J -P
CD CD
2 3
2
O
00
^
tn
i_i
_Q
•M
S
CD
cn
cn
•H
CL)
ft
C
.8
0
»
4-14-14-1
CH CH f^
CD CU CU
in in in
CD CD CD
Vj SH SH
ft ft ft
1 1 1
O
cn r^ o
vo cn r^
en rH cn
rH rH
•
• 4J
C ft >
id CD o
^ to 2
r~\
0 0)
tn < -H
-rl S W
g W -rH
(fl O §
O EH
0) rO
£ M rH
O a 6
rH ^a)
O >
[/] ^ ^J
a ffl cu
z z z
0 0 O
in
•
r^ oo vo
^* m ro
O
N -H C
4-) SH rd
-H rfl 3
ft -H p
w H SH
tj! Ol -H
rj ifl rfl
NUN;
•*
r-
cn
rH
CD
C
\-j
1
O
cn
rH
>
4J
ft
(D
CO
f-£
Z
U
w
CO
2
O
CM
rH
rrj
g S
^ d
i i *m
rH t3
0 S
^
en
rH
4J 1
C T^
0) -H
in E
0)
SH 0
U C
o -^
rO
i-t
O
CTI
c
ie3j
CD
Tl
O
U
•H
tn
o
rH
0
SH
O
CD
4-1
CU
S
0
o
•H
^
^
C T3 C
CU -H CD
ing en
CD CU
SH CD M
ft U ft
c
-rH
i in i
4-1
U
O (0 O
t — 4-i r —
en C ffi
rH 0 rH
U
4-1 O >
0 C O
0-^2
Ifl
rH
o
Oi C
C 0
ft -H
4-1
CD id
T) 4-1
CO
,
O,
1 1
. ,_! *^
D> S
W1 «5<
o *•
n ^
H rJ3
ri tfl
4J CD
JJ ^H
^ ^j
SCO
O SH
O CU
•H X!
> 4J
SH rO
CU dl
to s
CO CO
O O
in ro
•H CN
rfl
j-i
• rH
0)
•a
c
ifl ^
PQ CD
o
ifl X!
Tl Tl
C
Ifl -H
CO S
^
r~~ 4j
en C
rH CD
in
>i CU
to SH
S ft
I I
o
o r^
r— cn
01 rH
rH
•
4-1
> ft
O eu
2 to
C
O
•H
•P PH
rd H
•P to
W U
• ^
-M r^
(D 0
s S
*• ^
3 SH
rfl (D
cu in
SH .0
3 O
CQ
O
SH -H
0) 4J
X! CD
-P C
(fl tn
0) rfl
S S
to to
0 O
00 *J
l>1
rfl
CQ
^
CD in
Tl 3
C C
rO rfl
X E
CD SH
rH 0)
I<
»
rH
1 CM
' 1
VO
r-
l
0
o
rH
1
1
H
CU
S
•
N
O •
•H CD
S -H
CD g
•H O
C C
x; o
U SH
3 cu
Cu
• 3
O Cn
Cu -P
3
C -rl
rfl 4-1
cn
c c
(0 H
•H 1
X! M
(fl O
Pn C
rfl
• rH
Cu Cu
|
s X
> rd
C S
0
N >
O -P
MH
in rd
CD X!
x: o
u w
w c
•H -rl
SH cu
rfl g
Xi to
ft cn
in cn
a cn
r^
0 3
SH X!
H 0
: in
SH
4-1 O
CJ Pn
CD
•n CU
0 Xi
SH u
Cu n
4-1
C 3
0 CD
Q
4J
aS
CU VO
K 1
H PM
rfl
C •
•H O
&4 £H
4J
•• O
CD rfl
CJ SH
r*4 | *
3 C
0 O
CO O
39
-------
air conditions. Several general conclusions can be drawn from the long-term
ozone observations (16) which confirm Junge's earlier observations and which
are very relevant to recent discussions focusing on "photochemical production
of ozone versus ozone transport."
a. The diurnal variations of ozone concentrations exhibit regular
features characteristic of the particular stations. Low ozone values
near the ground indicate weak or stagnant vertical exchange of the
surface layer with the free adiabatic troposphere, while the daily
maxima appear when ozone-rich air from the free troposphere reaches
the ground station. Typical continental stations with convective
mixing during the day (Fort Lamy, Sa da Bandeira) show the daily
maximum around noon, when this convection is strongest (see Figure
17). At mountain stations (Zugspitze) the maximum occurs during the
night due to the small-scale mountain circulation pattern (Katabatic
wind flow). Coastline stations governed by land-sea circulation
(Westerland, Norderney, Alexander Bay) show the daily maximum during
evening and night.
b. The ozone daily maximum values from ground stations are representa-
tive of the ozone concentration of the free troposphere. This was
shown by studying the "Austausch-coefficient" of the near surface
layer using a 120 m tower (Tsumeb, S.W. Africa). The area surrounding
the tower is covered with a homogeneous covering of brushy vegetation
for thousands of square kilometers. Correlation coefficients of 0.40
between [O ] at 5 m and [O ] at 100 m above surface were obtained
under stable conditions (Ri [Richardson number]> 0) versus 0.95 for
unstable (Ri < 0) conditions (see Figure 18).
Several investigators (23-26) have interpreted part of all of the ozone
variation they observed on the basis of the phenomena described in (a) and (b)
above. The comprehensive and quantitative analysis of Fabian and Pruchniewicz
of the impact of surfaces and of "reducing atmospheres" on the ozone concen-
tration within the planetary boundary layer (acting as an ozone sink) implies
that photochemical production is not the only mechanism to explain the com-
40
-------
'.H, )MV
, , IAN MA*() , .'AN MfHi ,-1 1ih.iiU«*»«~.~»>.»*-
n FORT-LAMY 1
» 1 °NCUIC ^ i
,0
0 n r£kj
„ SA DA BANOEIRA
" APRIL - MAY 1971 1
20 f 53 DAYS 1
^Kf ',*. Al f N
i - 1 pi i4t •:.
f.i. ; />is
.v^l :j-f?t /.'/
.'• )' r ,rC I, '
• , ".* 1 1- 1 •
!
4. •^-iw"-«'/1-'-
! 7iiO'if"T/E
«PW,L - JUNF I'j7u
1 }•' DAY«i
r
1
[ '
|
I ;
1
L -i"t;l;,,!kii
! KAIROUAN
1 JU V • SEPT 1971
1 66 DAYS
i ,nj"i
h^^-nLJn
uUANOA
1 JAN • MARCH 1971
T 82 DAYS
A
^ ,.< Kn.
WlNrnOEK
! JULY ">tP1 1971
f 86 DArS i"i.
i
(
of!
j
i
j
j
t^t 1
!
j
Figure 17: Probability of the occurrence of the daily maximum of ozone near the
ground over hourly intervals, n denotes the number of days out
of the total given in the headlines when the daily maximum occurs
at given hourly intervals.
Source: Final Report on Project "Tropospharisches Ozon," P. Fabian and
P.G. Pruchniewicz. Contract No. FA-62/1 Deutsche Forschungsgameinschaft,
Max-Planck-Institut Fur Aeronomie. MPAE-W-100-76-21.
41
-------
,
Ippbl
so-- ["STABLE T 123
• ! Ri>0 ! PROFILES
t" ' X
w4
r
t CORRCOEF r^O.40
F
30 f /'y
1- ."V/
'%j»\
?0 •[•
r . ' •
(
I
1
[_
.-, 4 • .- i - - • • •• J- i- i .. i 1 4. -
:
,
/ ,
/
r ' '
;•••
V H
' ' ' . -\
-'
•
1
I opt:
i' - 4 . i -. . +. .
\
ipptol _
[UNSTABLE! 137 ~j
"- I Ri<0 ! PROFILES !
K L ~ ,_-l - i , '
y
" •! *
I CORRCOCF .- = 0.95 .•&.
•iff*
-jr*
^
? ••"
f
.
'- 0|A
..A *.i , . . * - * * i , i . .. -.» i . .- -
iO ?0 30 40 50
OZONE (••)7.Sm-iOsjOn-!
10
10
Figure 18. Correlation of mean ozone values measured within the Prandtl
layer (2.5-10m) with ozone concentration of the layer between
97.5 and 105 m.
Source: Final Report on Project "Tropospharisches Ozon," P. Fabian and
P.G. Pruchniewicz. Contract No. FA 62/1 Deutsche Forschungsgame-
inschaft, Max-Planck-Institut Fur Aeronomie. MPAE-W-100-76-21.
monly observed diurnal variation in ozone concentration. The monthly means of
tropospheric ozone, computed from the surface ozone daily maxima for Fabian
and Pruchniewicz's meridional network are shown in Figure 19. The annual
variations are fairly regular, and nearly all stations exhibit a pronounced
annual sinusoidal wave pattern with characteristic mean, amplitude, and phase
throughout the time of observation (see Table 4, a harmonic analysis of the
annual variation was performed). The average meridional distribution of ozone
in the troposphere varies by a factor of about 2 between the lowest values in
the tropics and the highest values observed in the subtropical regions and
northern mid-latitudes. This decreasing trend in ozone concentration towards
the equator is especially pronounced at high altitudes (11 km) as Holdeman and
Humenik (27) showed from the analysis of NASA-GASP data (see Figure 20). The
phase a of the annual maximum shows a clear variation with latitude (see
Table 4). The annual maximum which occurs during April/May in high northern
latitudes is shifted towards July in northern mid-latitudes and it appears
42
-------
I 1975 5u
•1 t
•2 d'lj
V v
12 ' 6 17
'973 , V
• r'x 'i
!*4i LUANDA
^ 1970
-*A Q* BANOtfA
>A/
W ~- "V
s •; b i?" " e
WNtWOCK
i-i7-
191?
^>^
•973 'y?^
-•'^•••">.v
ij 6 tf 6 '2' 13"
•972 , 1973
1971
1974
'9-rt,
/"•~ ;V/ ^x^y^. '
t .;• t I/' 6 ' ir !,'••«• «' •»' >'C
A,t ifiNUCt' &*r
'971
W^
V'
Figure 19. Monthly means of tropospheric ozone, derived from near surface
registrations of project TROZ.
Source: Final Report on Project "Tropospharisches Ozon," P. Fabian and
P.G. Pruchniewicz. Contract No. FA 62/1 Deutsche Forschungsgame-
inschaft, Max-Planck-Institut Fur Aeronomie. MPAE-W-100-76-21.
43
-------
—A-
600
400
200
JQ
Q.
CL
GASP (1975-1976)
(1963-196U) Hering et al.
(1963-1971) Wilcox et al.
Tropopause locations at indicated altitude /
I/
.5 11.0 11.5
km
a) March 1975, 1976
Figure 20.
20 30
Latitude, °N
b) May 1975
Variation of ozone mixing ratio with latitude for March and May.
GASP Data for altitudes 10.5 - 11.5 km. Hering and Wilcox Data
interpolated to 11 km.
Source: Holdeman, J.D. and F.M. Humenik: "NASA-GASP Data Report for Tape
VL0005, NASA TMX-73608, Lewis Res. Center, Cleveland, OH, 32 pp.
Feb. 1977.
Wilcox, R.W., G.D. Nastrom and A.D. Belmont: "Periodic analysis
of total ozone and its vertical distribution," (RR-3, Control Data
Corp.; NAS2-7807), NASA CR-137737 (1975).
Hering, W.S. and T.R. Borden: "Mean distributions of ozone density
over North America, 1963-1964," AFCRL-65-913, Air Force Cambridge
Research Labs. (AD-629989) (1965).
44
-------
TABLE 4. AVERAGE MEAN a , AMPLITUDE a AND PHASE a , OF ANNUAL WAVE AND
THEIR MEAN STATISTICAL ERROR
Station
Tromso
Bredkalen
Rise
Westerland
Norderney
Lindau
Hohen-
peissenberg
Zugspitze
Cagliari
Kairouan
Port-Lamy
Luanda
Sa da
Bandeira
Windhoek
Alexander
Bay
Hermanus
mean a\
[yg/m3]
40.4+3.3
45.7+11.4
44.8+0.2
47.1+9.8
41.9+15.4
34.4+13.3
64.8+16.5
45.1+5.4
41.0+14.2
56.1±15.6
33.416.0
28.8±3.0
34.1±10.4
50.6114.6
36.8±9.6
44.3±6.1
amplitude a
[yg/m3]
6.2+4.9
13.8
12.5+10.0
10.3+_3.0
15.7+6.5
11 .7+5.1
34.4+8.8
15.2+4.7
8.0+6.0
14.3±3.2
6.9±5.0
12.6±3.3
20.3±2.5
11.7+3.7
13.6±4.4
14.616.0
phase a of
annual maximum
May 7+52 days
April 20
May 15 +_ 53 days
July 1 +_ 17 days
July 3 +_ 47 days
June 6+27 days
June 22 +_ 17 days
June 23+19 days
June 25+48 days
July 11 +_ 15 days
June 15 + 50 days
July 14 + 16 days
Sept. 2+8 days
Oct. 20 +_ 17 days
Sept. 15 +_ 18 days
Sept. 3+24 days
a2/Sl
0.15
0.30
0.28
0.22
0.35
0.34
0.53
0.34
0.20
0.25
0.21
0.44
0.59
0.23
0.37
0.33
Source:
Final Report on Project "Tropospharisches Ozon", P. Fabian and
P.G. Pruchniewicz. Contract No. FA 62/1 Deutsche Forschungs-
gameinschaft, Max-Planck-Institut Fur Aeronomie. MPAE-W-100-
76-21.
45
-------
during May/June in low northern latitudes. In the southern tropics, a very
strong phase variation with latitude is observed.
Using the extensive measurement records from the meridional network
(Table 4), Fabian and Pruchniewicz (16) have calculated mean injection rates
for ozone as function of latitude following the analytical concept derived by
Junge (Equations 4 through 11). The unknown parameter C was obtained experi-
mentally. Disregarding ozone destruction within the troposphere and assuming
that ozone is decomposed at the surface, the ozone destruction rate F can be
written as:
F + e-q [031 (Eq. 12)
where : surface parameter o < e _< 1 (depends on the conditions of the
boundary layer, such as horizontal wind velocity and roughness
length)
[0_] : ozone concentration near the surface
3 o
q: specific ozone destruction rate of the particular surface type
q(sea surface) =0.1 cm*s (28)
qdand surface) = 1.0 cm*s~ (29)
_ -j
q(ice and snow) = 0.02 cm*s (30)
From Equations 12 and 5 follows T = C , i.e., the tropospheric
residence time for ozone as function of latitude (see Figure 21).
The final results of Fabian and Pruchniewicz's ozone injection and surface
destruction rates are shown in Table 5. Their values are very close to those
obtained by
Danielsen: 7 x 10 molecules-cm *s (from analysis of tropopause
folding events, cyclonic activity)
and Nastrom: 7.8 x 10 molecules-cm -s (from analysis of GASP data)
46
-------
However, the two investigators used very different experimental approaches and
assumptions in arriving at the above quoted stratospheric-tropospheric ozone
fluxes.
150 r
100 -
o
Q
50-
90°
60° 30°
Soutn
Latitude
30° 60°
North
90°
Figure 21. Mean residence time of ozone in the troposphere assuming
destruction at the surface to be the only sink. The uncertainty
range is due to uncertainties and variation range of wind velocity
and roughness length (see Fabian and Junge [1970]).
Source: Final Report on Project "Tropospharisches Ozon," P. Fabian and
P.G. Pruchniewicz. Contract No. FA 62/1 Deutsche Forschungsgame-
inschaft, Max-Planck-Institut Fur Aeronomie. MPAE-W-100-76-21.
All the evidence accumulated so far on the original of the tropospheric
ozone supports the "classical" concept of ozone intrusion from the stratosphere
and destruction at or near the earth's surface as being the dominant source and
sink for the tropospheric ozone. A substantial net production of ozone in the
unperturbed troposphere due to photochemical processes involving the
NO -HO-CH-CL-0 -system (without anthropogenic contributions) (31,32) is
X *£ 4 £ J
incompatible with representative tropospheric ozone measurements and budget
calculations, and radio nuclide measurements, but the existence of this
chemical process cannot be ruled out.
As was shown earlier (see Figure 16), simultaneous measurements of ozone
90
and radioactive fallout (Sr ) at a clean air station reveal a similar secular
pattern except that there exists a phase shift difference (i.e., tropospheric
lifetimes) for the two trace substances due to different removal processes.
Reiter et al. (33) found a statistically significant correlation coefficient
of 0.55 for the ozone/Be and of 0.45 for the ozone/fallout pair of variates.
47
-------
^
5J
H
CO
^
PQ
s
0
p4
O
*y
0
H
f~l
3
^
Q)
J
(rf
u
•
in
a
9
EH
s
w
0)
M1
•rl
Q
<0
fi
G
^j
O
2
M
01
*c
A
to
-rH
£2
0)
.c
c
!H
o>
JM
4J
O
^
C
•H
to
0>
o
N
O
C
O
•H
4J
O
01
g
•rH
{^
n]
0)
6
x
c
•H
10
N
o
c
o
•rH
4J
01
C
•rH
a
frt
0)
g
^-^
CN
O
i-H
•^-*
O
(D
to
v^
rH
O
j^
CN
O
__|
n
^^^
o
0)
to
CN
g
o
r-t
g
O
rH
O
rH
^-^
"o
0)
to
\^
rH
Q
I^B
CN
o
i-H
^^
0
0)
to
CN
o
rH
£
0
rH
O
i-H
^-*
id
0)
rrt
HJ
0)
TJ
4J
•H
4->
a
in co co ^f ^" in O
r*- en oS ^* oo vo r^*
r~ oo CN r- co r- ^
r-H rH CN rH H
o ^j1 c^- oo ^ o in
^3* ^* in ^* "^ co CN
O 00 CO in VO CN rH
in en co ^f ^* in co
•* CN O VO H in 00
^ ^ ^* ro CO CN rH
O O 0 O O 0 O
O O O O O O O
I-H CN co ^r in vo t~»
0 0 0 O O 0 0
O 0 0 O 0 O O
rH CN PO ^J1 in VO
pv \£
^ r^
CT> VO O
r^ CN to
\
rH
g
00
CN
O
rH
X
0^
,
00
tp ^Q
*^
rH
(4
O
co r~
co r-
O CO
I1 rH 0)
(0
rH
C
CO
CN
O
H
X
ff\
in m
in m o
CO (O
to
rH
ID C
•H 3
id o
r£ ^1
a o
to CM
O
a o) •
0 £ rH
rH O CN
E-< 01 1
3 r-
•P 0) 1
O Q O
0) O
•n I-H I-H
O \ 1
rH CN S
ft vO 1
M
G ril f^
O PTI QJ
§
-P •
M O
O*5 •
Qi 0>
fl\ t j ,_j
U/ -M *i"1
« o e
* o
H SH C
id 4-1 O
ecu
•rl O 0)
t< u «;
fll
UJ
3
o
en
48
-------
With the available test volume of 290 pairs of variates (clean air station at
Zugspitze, Germany during 1973-74), the highest accidental value of the cor-
relation coefficient would be 0.178. since Be is a unique tracer of strato-
spheric origin, the "background ozone" observed at this station should have
originated predominantly in the stratosphere.
The representative annual mean concentration of ozone for the unperturbed
("remote") troposphere has been subject to considerable debate. Most "rural"
stations show an ozone diurnal indicative of substantial ozone destruction
during times of limited vertical exchange, or of local or regional photo-
chemical synthesis involving anthropogenic precursor gases or a combination of
both. Mountain stations such as Whiteface Mt. (1600m MSL), N.Y., (26) or
Hurricane Ridge (1700m MSL), Olympic Mt., Washington, (23), located over 100
miles from significant pollution sources, are suitable for obtaining representa-
tive tropospheric ozone concentrations. For Whiteface Mt., the secular varia-
tion of ozone concentration (derived from the 1974 continuous ozone measure-
ments) is described by
X = 30 + 10 sin [2ir(t - 0.29)] (ppbv) (30 ppbv = 64 pg-m"3) (Eq. 13)
1 12
(t = — to — representing Jan.-Dec.)
The annual mean of 64 yg«m compares well with the mountain stations
Hohenpeissenberg, Germany and Zugspitze, for which values of 64.8 and 45 yg*m
respectively have been reported (see Table 4). The diurnal variation of ozone
at Whiteface Mt. is 12%. (The daily maximum occurs at 22:00 on an annual
basis, as is to be expected from a mountain station.)
On the basis of a few stations with long-term continuous ozone measure-
ments free of local or regional anthropogenic emissions (including ozone pre-
cursor gases), a range of x(annual mean) = 30 ± ppbv constitute a representa-
o
tive tropospheric ozone level for 35° to 50° N latitude. Note, however, that
the tropospheric background level increases up to MO ppbv during spring and
early summer and declines to ^20 ppbv in the late winter due to the seasonal
variation in stratospheric injection rate.
49
-------
Transport of ozone from the stratospheric source region to the ground is
an integral part of the "classical" ozone cycle. The transport across the
tropopause barrier occurs, for example, during cyclonic activity (30°-50° N
latitude) through tropopause folds (see Figure 11). The ozone extruded in the
folded tropopause, therefore, has the potential of occasionally causing
elevated regional ground-level ozone concentration. To what extent it will
exceed the seasonal or annual mean of the tropospheric background ozone con-
centration depends upon the dilution caused by mixing during the air's descent.
Measurements of radioactivity made in Project Springfield (12) show that the
dilution is concentrated along both the upper and lower boundaries, i.e., the
ozone-rich zone or layer is maintained by a convergent inflow normal to the
boundaries which counteracts the turbulent diffusion. In the center the
dilution is usually less than a factor of 5. Therefore, if the layer descends
close to the ground, ozone concentration of the order of 0.26 yg*g (1.3 x 10
g*g ) from Equation 3 divided by dilution factor 5), or 150 ppbv, a,re con-
ceivable. The ozone-rich surface pattern is strongly asymmetrical due to the
narrowness of the folded structure and the strong deformations in the de-
scending air. Some local regions may be influenced by this ozone-rich layer
for just 2 or 3 hours, others for 1 to 3 days. There is no meteorological
reason that the intruded ozone-rich layers must descend to the ground for each
and every tropopause folding event. The cyclonic index of 22.5 per year (11)
for the contiguous United States (each cyclonic event lasting up to 4 days and
associated with at least one tropopause folding event somewhere over the
United States) appears to be sufficiently large so that excessively high
regional ozone ground-level concentrations associated with baroclinic dis-
turbances would be a common phenomena ("stratospheric ozone episodes") - if
indeed it would be associated with each event.
The frequency distribution of ozone concentration at Whiteface Mt. shows
for the year 1974 (see Figure 22):
Total number of hourly ozone averages
(978 hours of missing data) 7782 hours
Number of hourly ozone averages exceeding 80 ppbv 69 hours
50
-------
0)
u
a)
+J
•H
4J
«J
o
•H
-P
0)
O
0
u
0)
c
0
N
0
14-1
0
§
•H
-P
3
4J
CO
-H
T)
O
Q)
CN
(N
01
0)
O
0)
c
o
•H
+J
(0
4J
CO
T)
rH
0)
•H
CM
4J
a
0)
o
<0
M-l
(1)
I
•o
0) .
A C
(0 0)
3JS
0)
o
M
§
sjnoi) |0|0| ^o % -
51
-------
This statistic shows that in only 0.9% did the tropospheric ozone concentra-
tion within the clean planetary boundary layer exceed 80 ppbv. This percentage
figure may vary from year to year since the meteorological pattern at a measur-
ing site will not remain identical year after year. To what degree, if any,
the observed excessive ozone concentration at Whiteface Mt. is a result of
long-range transport of ozone and/or ozone precursor gases from metropolitan
complexes (Chicago/Detroit, Boston, New York City, Montreal) cannot be assessed
with any degree of certainty. It is shown, however, that the trend of ozone
and the trend of potential temperature at the 850 mb level (see Figure 23) up
to 250 mbs (see Figure 24) are alike. This very strongly suggests that the
source of ozone is aloft, as does the night-peaking diurnal variation of mean
ozone concentrations, and that ozone is brought down with the descending air
masses. The case study for Whiteface Mt. covering the period July 21-August
1, 1975 (see Figure 24) showed hourly ozone concentrations in excess of 80
ppbv on July 31 associated with descending air masses.
Reiter (34) has examined the AFCRL ozonesonde network data from 1962
through 1965, a data base of some 1500 observations. From this sample, only
2% of the soundings exceeded 80 ppbv at ground level. Reiter chose to elim-
inate 28 of these cases as being suspect of tropospheric sources. This then
places a lower limit of 0.2% on the 80 ppbv exceedance level, which is in
reasonable agreement with the results from Whiteface Mt. It should be noted,
however, that ground-level ozone concentrations deduced from ozonesondes are
not representative means. The overwhelming majority of the ozonesonde ascents
that Hering and Borden report were made close to 12:00 Universal time, or just
after dawn local time, when ozone concentrations below the planetary boundary
layer are minimal.
Reiter (34) also examined hourly ozone data from the Zugspitze, Germany,
study for the period August 1973 through February 1976. Of the 529 days of
observations, ozone mixing ratios exceeding 80 ppbv were encountered only 0.2%
of the time. For these two events, isentropic back trajectories identified
the ozone source in the lower stratosphere.
-------
dWBl lOd
o
o
O
0s
(\J
in
LU
CC
I-
LJ
O
M
O
00
z:
o
LP
00
i i i
oo
oo
00
r-
oo
oo
LT
00
oo
oo
oo
LP
LU
O
o
o
LT
o
IT)
CO
Tf
c
0)
3
4J
(0
0) (U
n o
4J 10
IW
Q) Q)
fi -p
0 -rl
n
CN
(U
9
rl 4-1
0) C
0) ft us
030513
3 ft -rH
f8 X
ft >l
O (U
H M-l
B M-t
0
tW U
O
W *
>i C
H Q)
tC 03
i§ 0)
-H
RJ rd
O Q
•H
(C f-
U ^1
•H I
g CN
O •
O -P
-p ft
O (U
JZ W
ft
g J
•
0) !2
O
C
(1)
6 w a)
en
-H
-p
rH
-------
r—r
"TiIl
TTTT'-ri T
L
o
o
ro
O
O
o
o
m
o
o
_J L..J J__l_
000
000
oo en o
t
_l_L_L_LLJ.i 1..
O O C5 O O
oo to vr cvj
Od/A Wdd 3NOZO
r>
o
ro
CM
CO
CM
£
C\!
CM
rQ
"
^r
CO
OJ
CvJ
CM
OJ
1
•
4J
g
0)
o
i
(U rH
co 3
H 1-3
*3"
CN
0)
^
tn
•H
t,
*
•H
•O
rH
0
CM
a)
CO
3
fl5
a
0
EH
t4_|
0
en
-iH
CO
"rd
r<
(^
f^
It
U
•H
O
rH
0
0
CU
+j
CU
2
•d
c
rd
4J
C
g
0)
SH
3
CO
rd
0)
g
0)
d
0
N
0
C1J
D
3
O
CO
4J
ftf
•d
en
+J
H
•d
HlJ
c
0
•rH
JJ
3
rH
, — |
0
4J
C
rd
•d
•H
^
0
rH
U
•H
e
0)
.c
o
o
+J
o
ft
c
0
(U
U
c
0)
^_J
0)
M-f
c
o
U
rH
rd
c
o
•H
.p
rd
c
0)
-p
c
H
&
f^
en
rH
<^
r-
rH
1
,— I
•
+J
ft
-------
Carney (35) examined also the ozonesonde data from eight stations within
the contiguous United States in the network operated by the Air Force Cambridge
Research Laboratory (36). Ozonagrams from April 3, 1963, to December 12,
1965, were used for Carney's study. Carney tested Reiter's (37) hypothesis
describing a transport mechanism that correlated high low-level ozone amounts
at mid-latitudes to the seasonal fluctuations of an "ozone reservoir" in the
lower stratosphere and to synoptic scale transport in the troposphere. Carney
assumes that vertical distribution of ozone can be used to discern the prin-
cipal source of low-level ozone. For convenience, a level of significance
was chosen to be an ozone mixing ratio of 0.10 yg*g (or 0.06 ppbv at 20°C).
Soundings were considered critical when the mixing ratio exceeded this level
below 500 mb. The premise is that if ozone is transported downward from above
the selected 500 mb level, the sounding should show no gradient or a positive
ozone concentration gradient with ascent. (This assumes a continuous downward
transport instead of massive short-term transports that may occur in the
vicinity of large convective cells or squall lines.) Carney classified these
ozone soundings as "transport type." Distributions that suggest a possible
low-level source of ozone were referred to as "photochemical type." The
results of Carney's investigation follow.
Critical soundings at Point Mugu, California generally had steep gradients
near the surface. Lea (38) has demonstrated a correlation between ozone
maxima and easterly winds bringing air from the Los Angeles basin to the
launch site. Otherwise the overwhelming majority of the critical soundings
were of the "transport type," with constant or increasing mixing ratios with
height. It appears therefore that most high tropospheric ozone concentrations
(>0.10 ug-g below 500 mb) w<
conceivably the stratosphere.
(>0.10 ug*g below 500 mb) were largely the result of transport from aloft,
The frequency of stratospheric ozone intrusion is sporadic due to the
sporadic occurrence of cyclonic events or other exchange mechanisms. After
injection, the ozone-rich layers or zones disperse in time, depending on the
atmospheric stability structure. It is, therefore, not surprising to occa-
sionally find ozone layers of variable thickness imbedded in the troposphere,
as has been reported in literature (see, for example, reference 39). It is
55
-------
TABLE 6. CLASSIFICATION OF CRITICAL OZONAGRAMS BY VERTICAL
DISTRIBUTIONS (> 0.10 pg-g"1 below 500 mb)
Total
Soundings A
Albuquerque , NM
Bedford, MA
Fort Collins, CO
Green Bay, WI
Madison, WI
Pt. Mugu, CA
Seattle, WA
Tallahassee, FL
Totals
200
185
152
53
28
18
139
129
904
17
34
21
4
11
6
29
33
155
(100%)
Percent of Percent of
Critical Total
Soundings* B Soundings* C
8.5%
18.4%
13.8%
7.5%
39.3%
33.3%
20.9%
25.6%
17.1%
13
12
18
4
6
0
21
23
102
(65.8%)
6.5%
9.2%
11.8%
7.5%
21.4%
0.0%
15.1%
17.8%
11.3%
4
13
1
0
3
6
2
5
34
(21.9%)
D
0
4
2
0
2
0
6
5
19
(12.3%)
A - Number of critical soundings
B - Number of "transport" soundings
C - Number of "photochemical" soundings
D - Number of unclassified soundings
* Total soundings equals 100%.
Source: Carney, T.B. Evidence of the Role of Stratospheric Transport
in the Distribution of Tropospheric Ozone. Ozone/Oxidants -
Interactions with the Total Environment. APCA Specialty
Conference (Southwest Section), Proceedings. Air Pollution
Control Association, Pittsburgh, Pa., 1976. pp. 234-241.
56
-------
also not surprising to find the ozone mixing ratio in the chemically unper-
turbed troposphere fluctuating from day to day. The concept of a well mixed
tropospheric ozone content can only be considered as a first rough approxima-
tion. Prunchniewicz (40) has analyzed the ozonagrams obtained from 1963-1966
by Dutsch (41). He compared the ozone mixing ratios at the 400 mb and 500 mb
layer for a certain day T with the corresponding value for the previous day T-
1 or T-2. His results (shown in Figures 25 and 26) demonstrate the great
variability of the ozone in the mid-troposphere, although the intrusion
(tropopause) and destruction region (below planetary boundary layer) are quite
some distances above and below the selected observation layer. The mean ozone
mixing ratio in that layer was found to be 0.08 yg*g (400-500 mb), with a
mean deviation of 50% for the (T, T-l) and 25% for the (T, T2) data pairs.
Pruchniewicz (40) attempted, furthermore, to correlate the relative changes of
ozone mixing ratios at the 300 and 500 mb layer with those at the 200 mb layer
("stratospheric ozone reservoir"). He found correlation coefficients between
0.05 and 0.35, i.e., there exists only weak correlation between the strato-
spheric and tropospheric ozone content within a vertical column above a fixed
geographic location. This means that the ozone measured in the mid-tropo-
sphere did not, in most cases, originate from the stratospheric reservoir
directly above, but was injected into the troposphere at any point "upwind."
The importance of tropospheric ozone transport and mixing processes have thus
been clearly demonstrated.
All ground-level ozone observations exhibit a definite "spring rise
phenomena" which is attributed to the increased stratospheric-tropospheric
transport of ozone-rich air (Figures 2, 3, 13, 16, and 19). The magnitude and
time of the year of the resulting primary ozone maximum, as shown earlier,
depends on the geogrpahic latitude of the ground-level measuring site. There
are indications of secondary ozone maxima occurring in early summer and/or
broad maxima extending from April into August. Fabian and Pruchniewicz (16)
have shown through harmonic analysis of the annual variation of ozone observed
in the meridional network (Figure 19) that only the annual wave is statisti-
cally significant. Higher harmonies are insignificant, without exceptions,
for all stations in his network.
57
-------
120
113
lit
112
110
g|-a»
102
100
98
96
94
92
90
88
84
84
•c —0,15
78
74
72
70
»3
64
64
42
40
93
94
52-0.10
90
46
44
4*
40
36
34
32
30
78
26
- 0,05
22
20
18
14
14
12
10
a
*
1
11 1
1 1 11
1111
11 1
1 1 111
1 211/112 111
I I 2 Yil I
1 4 11 1/21 1
111222 /I 11 11
11 /I 111
111/ 121211
II /14U1
1 2/ 1
12/5 It
1 I/ 2 11
0,05
I
0,10
I
I
20
I
60
0.15
I
I
80
0,20 fjg/g
I
100
I
123
1
1-0
Figure 25. Correlation of tropospheric ozone mixing ratio for a day T with
mixing ratio for day T-l. Each data point consists of a pair
of ozone mixing ratios for day T and T-l. This graph is made
up of a total of 161 ascents, with ozone mixing ratios from the
400 and 500 mb layer over Boulder, Colorado (ozonesondes by
Dutsch). Seven data pairs (4.35%) are in excess of 0.15 yg-g
Source: P.G. Pruchniewicz, "Uber ein Ozon-Registriergerat und Untersuchung
der zeitlichen und raumlichen Variationen des Tropospharischen
Ozons auf der Nordhalbkugel der Erde." Mitteilungen, Max-Planck-
Institut fur Aeronomie, No. 42, 70 pp (1970). Springer-Verlag,
Berlin-Heidelberg-New York.
58
-------
r»G r
120
113
116
114
112
110
"3-0,20
104
102
100
98
92
90
83
86
84
80—0,15
ta
76
74
72
TO
ta
66
64
62
60
38
56
52-0,10
50
46
44
42
40
38
36
34
32
30
28
"-0-05
22
20
18
16
14
12
10
a
6
1 1
1
1
1 11
1 1 11 1
1 12 1 I ,
12 11
1 12
1 1111 I ll/ 11
1 1 112
1U/3 1
1 11 223 11
1 11 1 JU 1
I 11 12Z/UU1
11 111/ 11 1 1 1
1 222 L2 1 11 11
1 12/ 1 211 11
111/ 111
2/21 1 1
I/ I
11 /I 1 2 I
I/ I 11
1
1
0,05
I
0,10
I
20
I
40
I
60
0,15
I
I
10
0,20
I
ico
I
123
Figure 26. Correlation of tropospheric ozone mixing ratio for day T with
mixing ratio for T-2. Each data point consists of a pair of
ozone mixing ratios for day T and T-2. This graph is made up
of a total of 174 ascents, with ozone mixing ratios from the
400 and 500 mb layer over Boulder, Colorado (ozonesondes by Dutsch)
Five data points (2.9%) are in excess of 0.15 yg«g-1.
Source: P.G. Pruchniewicz, "Uber ein Ozon-Registriergerat und Untersuchung
der zeitlichen und raumlichen Variationen des Tropospharischen
Ozons auf der Nordhalbkugel der Erde." Mitteilungen, Max-Planck-
Institut fur Aeronomie, No. 42, 70 pp (1970). Springer-Verlag,
Berlin-Heidelberg-New York.
59
-------
Pruchniewicz (40) has reanalyzed the ozonesonde data of the AFCRL network
(Figure 4) and calculated the mean ozone mixing ratio between 1 km and 9 km
altitude for all 13 stations as a function of season. His results are shown
in Figure 27. The averaged "total" tropospheric ozone does indicate a per-
sistence of high tropospheric ozone values into the summer period. Falconer
(42) has demonstrated on the basis of available GASP data (1975/76) that this
biomodality for the tropospheric ozone concentration does also exist at levels
just below the troposphere (see Figure 28). His explanation for the statisti-
cally significant secondary ozone maximum is based on the fact that the tropo-
pause rises rapidly in early summer within the latitude belt 30-60°N. Hence,
previously stratospheric air (and "stratospheric" ozone) will be incorporated
into the troposphere. The resulting ozone intrusion will eventually be
transported down to the ground and give rise to increased ground-level ozone
concentrations leading conceivably to a secondary ozone maximum and/or broad-
ening effect of the primary spring maximum. On the basis of available data,
one cannot exclude at this time that the tropospheric methane oxidation cycle
could also contribute to and enhance the "delayed decline" of tropospheric
ozone concentration from the spring maximum to the winter minimum. The
observations of Falconer (42) (GASP data) presented in Figure 28 are, however,
incompatible with a major ozone source strength located within the unperturbed
troposphere. It is generally agreed that no net production of ozone occurs
above 6 km altitude involving the methane oxidation cycle. Hence, meteoro-
logical exchange and transport processes (cyclonic activity and tropopause
lifting) must again be mainly responsible for the observed ozone concentra-
tion. Further indication for the validity of the "meteorological" concept is
seen in Figure 27. The ozonesonde data from Albrook (Balboa), located near
the equator, show a spring maximum but no indications of a summer peak (note
that there is no tropopause rise at low latitudes; see Figure 8, "Swan Is-
land") although the solar intensity would favor highest photochemical activity.
Global mesoscale and microscale meteorology-climatology plays a dominant
role in determining the daily, seasonal, and secular characteristics of the
tropospheric ozone concentration.
60
-------
\—
(:
i
W 1 F
j— —
C
•
W 1 F
j)
S
S
1 .
H
H
W I F I S I H
W
1
1
©
1
F ' S
H
w
©
F ! S ' H
W
I
I
|
®
i
F ! S
H
W I F ' S H
F I S I H
ED =O.
1 W I F I S I H
W 1 F I S i H
W= Winter
F = Fruhjabr
5 = Sommer
H = Herbst
1 Thule
2 Alaska
3 Fort Churchill
4 Goose Bay
5 Seattle Washington
6 Green Bay
76,5°N
64,8°N
58,8°N
53,8°N
47,4°N
44,5°N
7 Bedford 42,5°N
8 Ft. Collins/Colorado 40,6°N
9 Albuquerque/ New Mexico 35,0°N
10 Tallahasse/Florida 30,4°N
11 Grand Turk 21,5°N
12 Albrook (Balboa) 9,0°N
Figure 27. Secular variation of tropospheric ozone concentration over
North America (yg/g). Mean ozone mixing ratio between 1 km and
9 km altitude extracted from ozonesonde data 1963, 1964, and
1965.
Source: P.G. Pruchniewicz, "Uber ein Ozon-Registriergerat und Untersuchung
der zeitlichen und raumlichen Variationen des Tropospharischen
Ozons auf der Nordhalbkugel der Erde." Mitteilungen, Max-Planck-
Institut fur Aeronomie, No. 42, 70 pp (1970). Springer-Verlag,
Berlin-Heidelberg-New York.
61
-------
LTJ
•d,
fl
•-3
•S
•<
•s
• u.
•-D
•Q
•Z
-O
r
CD
CM
CO
r^-
en
I
a
G
-H •
in
T! rl
0) Q)
C C
•rl -H
18 r-l
•P H
•8T3
W O<
o en
•rl <
•P O
(8
C -P
•rl 18
X ft
•H -rl
e u
c u
0 ITS
N ft
e a
0)
- o
S3 &
i
.p
•H •
rH O
33
Oi U
c
-H
rH
a -P
g
3
O PQ
•H
H 0)
o) c
J5 O
ft N
in O
e o
•P -H
< S-l
a)
•rl
<
fi
o
13
.8
MH
O r^
O ft
EH U)
Cn
G
•• -H
H £
a> in
G -H
O rH
U X)
rH (8 O
re -P e
Cn in : -P
W . sC
• (8
Q H -P t3
• O <8 G
CM M-i O <8
0)
O
H
3
O
10
CD
CO
CD
CO
/\qdd 'OIJDJ BUIXILU suozo
62
-------
As pointed out earlier, stratospheric ozone can occasionally reach tne
ground, causing excessively high concentration levels of regional extent.
There are a few reports in the literature of those ozone episodes. Walker and
Hathorn (3) described such a possible case for Texas, with ozone levels ex-
ceeding 80 ppb for several days (September 25 - October 1, 1975). Huffman et
al. (25) reported on a high regional ozone incidence that occurred on February
24, 25, and 26, 1976. A typical tropopause folding event over the midwest was
identified as the source of ozone.
There are no statistics available at this time for the frequency of
occurrence of "stratospheric ozone episodes" of regional extent. They may be
infrequent/ but cannot be defined from commonly available surface ozone data,
most of which is influenced by local or regional pollution.
SUMMARY
The ozone concentration in the unperturbed troposphere is governed mainly
by stratospheric ozone intrusion and by ozone destruction at the surface
and/or within the planetary boundary layer. The contributions of the various
stratospheric-tropospheric transfer processes to the total amount of ozone
transported into the troposphere are not yet well established. The resulting
annual mean concentration representative for the unperturbed lower troposphere
(near ground) is of the order of 30 ± ppbv for the 30° - 50°N latitude belt.
o
Ground-level ozone concentrations (hourly averages) exceeding 80 ppbv occur at
a frequency of less than 1% on an annual basis. "Stratospheric ozone episodes"
are rare events. A frequency of occurrence cannot be established at this
time. The methane oxidation cycle does not constitute a dominant source of
tropospheric ozone. It must be pointed out that considerably more research
efforts are required to quantify the various meteorological and chemical
processes governing the natural ozone cycle. The lack of long-term measure-
ment series from ground-level stations representative for the unperturbed
natural troposphere is a very serious limitation for refining our knowledge of
the "natural ozone."
63
-------
ACKNOWLEDGMENTS
The author is very grateful to Dr. Edwin Danielsen of the National Center
for Atmospheric Research, and to Dr. Austin Hogan and Phillip Falconer of the
Atmospheric Sciences Research Center for many clarifying discussions.
COMMENTS BY ELMAR REITER
In summary, I should like to say that Mohnen's paper is excellent. Its
overall conclusions agree with those I have reached in my paper. There are,
however, some internal inconsistencies in the use and interpretation of data
which might, in a few places of his discussion, lead the reader to expect very
high influxes of stratospheric ozone. My following discussion points out these
inconsistencies. If Dr. Mohnen could reconsider the use of some of these values,
all aspects of his paper could be brought into total agreement with his and my
final conclusions.
In Mohnen's report a major difficulty exists on pp. 15-19. On p. 15
20
he quotes Danielsen's estimate of 4.3 x 10 g/year (i.e., 95% to 100% of one
hemispheric stratosphere) due to eddy transport. I don't know where this
quotation comes from. It is not in the Danielsen paper referenced on p. 15.
This number is either wrong, or attributed to the wrong transport processes.
If one were to add to this eddy transport the other transport processes quoted
in my report (see Table 1 of my report) one would end up with 179% of one
hemispheric stratosphere exchanged per year. This would lead to a mean residence
time (according to my Eq. 1) of 0.56 years or 6.7 months, which is clearly in
90
disagreement with observed Sr residence times in the stratosphere.
The last paragraph on p. 16 in Mohnen's paper ascribes the discrepancy
between Danielsen and myself to wrong inferences from just one case study.
Actually, there were several such studies in good agreement with each other,
and my value has been compared with a total stratospheric mass balance (see
Table 1 in my report). To my knowledge this has not been done with Danielsen's
number. Again I have to state, that I don't know how he derived his number.
64
-------
The estimate of 80% mass exchange rate quoted on p. 18 of Mohnen's report,
on the other hand, refers to the total outflow rate from the stratosphere
(including effects of the mean meridional circulation) and is in reasonable
agreement with my estimates (average of 89 and 73%, see p. 74 of my report).
Mohnen apparently takes this correct estimate of all transport processes in
support of the wrong statement of 90% due to eddy transports (p. 16 of his
report).
On p. 20 there is another questionable use of data. Mohnen assumes an
ozone mass mixing ratio in the lower stratosphere of 1.3 x 10 g/g. This value
occurs at levels above the tropopause which usually are not involved directly
in the intrusion processes (see for instance Mohnen's Figure 11 and my Figure 7).
Very recent measurements, to be published by Mel Shapiro of NCAR, suggest that a
value of 0.5 to 0.7 x 10 g/g would be a more appropriate O mixing ratio to use
in such ozone transfer estimates into the troposphere. Thus Mohnen's outflow
rates are overestimated by a factor of 2. Note: The relatively low mean
ozone concentration in the layers above the tropopause apparently comes about
by the fact that tropospheric air is mixed into the stratosphere near
7 32
jet streams. From Be and P observations, E. Reiter, and R. Reiter et al.,
(33,43) have concluded that air in the lower stratosphere, to an appreciable extent,
is of tropospheric origin.
The outflow rates computed by Mohnen, using Equations 2 and 3 are subject
to the same overestimate of low-tropospheric 0 concentrations. When all is
considered (Mohnen, p. 25) the annual outflow rate should be of the order of
14 14
2 x 10 g or less; 8 x 10 g would be an overestimate by a considerable factor.
If the estimates of vertical velocity, w, are to be extended over the
whole hemisphere and not just in the jet-stream region, a value of 0.5 cm/sec
is an overestimate by at least a factor of 2 (see my paper, p. 82).
On p. 46 the mass flux of 7 x 10 molecules*cm sec should be attributed
to all processes, and not to eddy processes and tropopause folding alone.
65
-------
A mean tropospheric background level of 30 ppb quoted on p. 49 agrees
excellently with my estimates.
Mohnen's estimate (p. 52) of 0.9% of hours exceeding the Federal standard,
does not disagree with my value of 0.2% of local exceedance probability, which
refers to days during which the 1-hour standard is exceeded.
Last sentence in first paragraph on p. 52: Change to "...ozone concentrations
in the lower part of the planetary boundary layer are minimal."
66
-------
REVIEW AND ANALYSIS
Elmar R. Reiter
ABSTRACT
High ozone concentrations in rural areas far away from possible indus-
trial sources have been blamed, at least in part, on stratospheric air intru-
sions into the troposphere. The assessment of the actual impact of such
intrusions on ground-level ozone concentrations requires sophisticated analy-
sis techniques which are not without difficulties. This paper explores var-
ious estimates of the impact of stratospheric ozone on ground-based ozone (0 )
concentration measurements. Measurements on mountain peaks have to be treated
with special care because, due to local circulation systems and to mountain-
generated turbulence, they could be contaminated by air masses of low-tropo-
spheric origin, whereas the closest radiosonde ascent might not reveal such
contamination. Circumstantial evidence is presented in this paper for a low-
tropospheric "ozone climatology."
INTRODUCTION
The search for the origin of high oxidant concentrations frequently
encountered in rural areas still is faced with conflicting evidence. Whereas
there is no doubt that photochemical reactions of certain pollutants in the
low troposphere, especially within the planetary boundary layer (into which
most of the anthropogenic contaminants are emitted), play a major role in
generating high oxidant concentrations observed close to the ground, a con-
troversy still exists about the impact of stratospheric air intrusions on low-
tropospheric ozone "background" and "peak" concentration values (1).
67
-------
This paper tries to provide an assessment of the processes involved in
generating high ozone concentrations of possible stratospheric origin near the
ground. It weighs some of the conflicting observational evidence, points out
possible sources of error and suggests approaches for additional investigations.
THE STRATOSPHERIC RESERVOIR
Photochemical production and destruction mechanisms of 0 in the strato-
sphere have come under close scrutiny during the Climatic Impact Assessment
Program (CIAP) sponsored by the U.S. Department of Transportation (33). The
effects of anthropogenic halocarbons have been included recently in the long
list of possible photochemical reactions (see for example, reference 44), and
the complexity of chemical interactions in the stratosphere recognized as
having possible effects on the O reservoir is increasing steadily. Inter-
actions of 0 with NO are still considered as being of major importance in
.3 X
controlling the stratospheric 0 distribution, together with the "Chapman
mechanism" of direct production and destruction of 0 by UV absorption.
Recent studies (45) have addressed themselves to cataclysmic increases of
stratospheric NO from
of volcanic eruptions.
mechanism" of direct production and destruction of 0 by UV absorption.
;a
stratospheric NO from a nuclear holocast, but also to the more subtle effects
X
The magnitude of long-term anthropogenic effects on the stratospheric
ozone reservoir is still very much under dispute. It could amount to anywhere
from 5 to 20 percent. A long-term variability of similar magnitude can be
attributed to changes in the effectivenss of stratospheric transport processes.
The variability of these transport processes is tied to climatic variations.
Colder climates would show significantly lower stratospheric ozone concentra-
tions in middle and high latitudes than would warmer climates (46).
In an assessment of the effects of stratospheric ozone on low-tropo-
spheric oxidant concentrations, can the long-term (secular) variability of the
stratospheric ozone reservoir be ignored? A number of stations (as exempli-
fied by Arosa, Switzerland) have shown a gradual increase in total atmospheric
ozone, especially during the spring months of recent years. According to
68
-------
Figure 1 the annual spring peak of total ozone over Arosa increased by approxi-
mately 20 percent between 1967 and 1970 (47). Angell and Korshover (48)
pursued the problem of long-term variability of global ozone in more detail.
Figure 2 depicts this variability in several geographic sectors, together with
a plot of smoothed sunspot numbers. It is very tempting to conclude that the
sharp increase in total ozone during the late 1960's coincided with an increase
in sunspot activity, and the decrease in O observed during the early seventies,
likewise, parallels a decrease in sunspot numbers. There is some doubt about
the accuracy of the Russian filter ozonometer. If one considers, therefore,
only the European and American measurements, an amplitude of the secular
variability of total ozone of about 10 percent appears to be a reasonable
assumption.
300
n 35°
2 300
1 I I I I I I
. A/wvvsi
1935 1933 1934 1935 1936 1937
§250 i.—- * .—.. * — i nn.— * i .1 i i
1938 1939 1940 1941 1942 1943 1944 1945 1946
1947 1948 1S49 I960 1951 195? 195] 1954 1955
1956 1957 1958 1959 1960 1961 1962 1963 1964
250
1964 1965 1966 1967 1968 1969 1970
Figure 1.
Monthly mean total ozone amounts at Arosa, Switzerland (46.5°N,
9.4°E). [Top part of diagram from J.M. Wallace and R.E. Newell,
Quarterly Journal of the Royal Meteorological Society, 92; 487
(1966); bottom part of diagram: data courtesy of Dr. J. London.]
Source: Reiter, E.R., and B.C. MacDonald. Quasi-Biennial Variations in the
Wintertime Circulations of High Latitudes. Arch. Meteorol. Geophys.
Bioklim., Series A, 22(1):145-167, 1973.
69
-------
-------
-4
1955 1958 1961 1964 1967 1970 1973
Figure 3. Temporal variation in total ozone in West Europe by season. A 1-
2-1 smoothing has been applied to the successive yearly values for
each season.
Source: Angell, J.K., and J. Korshover. Global Analysis of Recent Total
Ozone Fluctuations. Monthly Weather Review, 104(1):63-75, 1976.
71
-------
12
10 II
12
Figure 4. Worldwide total ozone as a function of season and latitude [From
Diitsch (1971), based upon data by London (1963) and Sticksel
(1970H . The numbers are total amounts in the conventional units
of 10~3 atm-cm STP.
Source: Reiter, R., H.J. Kanter, R. Sladkovic, and K. Potzl. Measurement of
Airborne Radioactivity and Its Meteorological Application. Part V.
Annual Report 1 April 1973 - 31 July 1974. Institut Fur Atmo-
spharische Umweltforshung. ERDA Document No. NYO-3425-12. 1976.
72
-------
In conclusion, we can state that the stratospheric reservoir of ozone is
subject to a strong seasonal modulation (38%), with a weaker (10%) secular
trend superimposed. The seasonal variation of stratospheric ozone certainly
will have to be taken into account in an assessment of the impact of strato-
spheric ozone on ground-level oxidant concentrations. The effects of a secular
trend of the magnitude indicated above can, perhaps, be ignored for the time
being, because other processes described in the subsequent sections are beset
with much larger "error bars" than 10 percent. If, however, anthropogenic
effects on stratospheric ozone should increase its secular variability signif-
icantly beyond this value, the natural background of ground-level oxidants
will have to be adjusted accordingly.
STRATOSPHERIC-TROPOSPHERIC TRANSPORT PROCESSES
The interchange of stratospheric and troposphere air masses proceeds by
various mechanisms, each of which is characterized by short- and long-term
variability with time. The following mean magnitudes of mass flux from a
"box" encompassing one hemispheric stratosphere, in percent of the mass equiv-
alent to that in this "box," have been estimated from evidence presented in
the literature (for references see Reiter (7,50,51); Singh et al. (52)).
TABLE 1. ANNUAL MASS FLUX PROM STRATOSPHERE,
IN PERCENT OF MASS OF ONE HEMISPHERIC STRATOSPHERE
Seasonal adjustment of tropopause level 10%
Mean meridional circulation 43%
Stratospheric exchange between hemispheres 16%
Large-scale eddies (jet streams) 20%*
Small-scale eddies negligible
TOTAL 89%
* An average mass flow of 4 x 1017g per cyclogenetic event (9,10,53) times
31 cyclonic disturbances per year between 40° and 60°N and 70° to 180°W
(11) would yield 12.4 x 1018g in that sector per year, or approximately
3.72 x lO^g per year to the whole northern hemisphere in a deliberate
overestimate. Mohnen's et al. (26) estimate of an eddy flux of 4 x 1020
g/year is an overestimate of at least a factor of five and is in disagree-
ment with other evidence of mass fluxes from the stratosphere.
73
-------
Whereas there is some uncertainty as to the exact contribution of each of
these processes to the total mass exchange across the boundaries of one
hemispheric stratosphere, the total flux estimate is in excellent agreement
90
with the observed depletion of Sr in the northern and southern hemisphere
stratospheres after the nuclear test ban treaty went into effect and before
the resumption of atmospheric nuclear testing by France and China.
90
For the northern-hemisphere stratosphere, where Sr was injected prior
to 1963, the depletion of this relatively long-lived radionuclide (T
radio-
,. » T . ; T are characteristic e-folding times) can be
active atmospheric residence
formulated as follows:
$ = - | N (Eq. 1)
T is the e-folding stratospheric residence time, N. is taken to be 100%, AN is
89% according to the estimates listed above, At is 1 year. From these values
90
one arrives at T = 1.12 years or 13.4 months, in agreement with Sr data
(Figure 5). For the southern-hemisphere stratosphere we have to write
H= - |N + S (Eq. 2)
where S = 16% is the "source" consisting of the import from the northern
hemisphere, and AN = 89% - 16% = 73%. Again assuming N = 100% and At = 1
year, we arrive at T = 1.75 years or 21 months, which is in excellent agree-
ment with the data presented in Figure 5.
90
Unfortunately O is not a passive tracer like Sr which, once injected
into the stratosphere, is subject only to transport processes. There are
continuous, but time-variable, photochemical sources and sinks of O acting in
both hemispheres. Also the transport processes which remove O from the
stratosphere are variable with short-term, seasonal, and long-term fluctuations.
To assess the impact of transport processes across the tropopause on low-
tropospheric ozone concentrations we can limit ourselves to O concentrations
observed in the vicinity of, or slightly above, the mid-latitude tropopause
level, and to the downward directed mass fluxes in these latitudes. The
upward flux of tropospheric air into the stratosphere, which takes place
74
-------
mainly in low latitudes, can be ignored in our estimates, because it does not
contribute positively to the tropospheric ozone burden.
TOTAL STRATOSPHERE
NORTHERN HEMISPHERE
SOUTHERN HEMISPHERE
LARGE ATMOSPHERIC TESTS I
RESPECTIVE HEMISPHERE
Figure 5. Stratospheric inventory of 90Sr.
Source: Krey, P.W., M. Shonberg, and L. Toonkel. Updating Stratospheric In-
ventions to January 1973. Report HASL-281. I-130-I-142. U.S. Atomic
Energy Commission, Washington, D.C., 1974.
75
-------
The following evidence can be called upon: cross-sections provided by
DUtsch (49) (see reference 33) and reproduced in Figure 6 show average ozone
concentrations of 80 nanobars (nb) near the 300-mb level in middle and high
latitudes of the northern hemisphere during the cold season. Conversion of
these concentration units into ozone mixing ratios, r , is accomplished by the
relationship
2.871 x 10 p (nanobars)
r3(vg/g) = r (Eq. 3)
1.732 x 10 p(mb)
where p is the atmospheric pressure. Hence 80 nb at 300 mb corresponds to
0.44 pg/g of O mixing ratios. Inspection of individual ozonesonde ascents
(14,33,36,37,55-57) reveals that 0.5 yg/g sets a convenient upper limit of
average ozone concentrations a short distance above the tropopause. Actually,
ozone concentrations at tropopause level in mid-latitudes usually are somewhat
lower, as is evident from Figure 7, which depicts an intrusion case of strato-
spheric ozone (58).
In the following table (50) we have listed mass fluxes across the tropo-
pause during different seasons as computed by J.F. Louis (see reference 33),
and pertaining to the mean-meridional circulation contribution.
TABLE 2. MASS FLUX FROM STRATOSPHERE TO TROPOSPHERE
ACCOMPLISHED BY HADLEY CELL CIRCULATION
Season
Dec.
March
June
Sept.
- Feb.
- May
- Aug.
- Nov.
Mass Flow
12
(units 10 g/sec)
10
4
7.5
7
Total Flux in
Three Months
(1017g)
788
311
583
544
Contribution in
Northern
(10
622
272
389
560
Hemisphere
17 .
g)
Annual flux, northern hemisphere 1843 x 10 g
76
-------
10
1000
9O*S 80 70 6O 5O 4O 3O HO IO 0 10 2O 30 4O 50 60 70
LATITUDE
(a)
March/April
IOOO
80 9O"N
a:
CO
co
id
a:
a.
10
20
30
40
50
70
hJ 100
a:
IU
-
CL
CO
f-
200
3OO
40O
IOOO
II I I ! I I I I
I II 1
I I
9CTS 80 70 60 30 40 3O 20 10 0 10 20 30 4O 50 60 70
LATITUDE
(b)
October/November
Figure 6. Pole-to-pole cross-section of vertical ozone distribution
nb = 0.1 N/m2).
Source: DUtsch, H.U. Photochemistry of Atmospheric Ozone. Advances
Geophysics, 15:219-322, 1971.
77
IO
20
30
40
50
70
IOO
200
30O
400
IOOO
80 90*N
in nb (1
in
-------
H E0l '1HOI3H
to
0)
•H
-P
•rH
•H
-p
U
(d
iH
N
LO
en
tn
-S
O
0)
to
•P Cn
•H \
o en
•rH
Ml
o o
id to
•H O
•P -H
C -P
0) «)
-M M
o
ft en
o x
•H
§g
•H <1J
•P C
U O
Q) N
10 O
W C
O (d
U •>
rH C
SH ft
• -p
J* -H
o 5
o
rH T3
rH O
(0 r-
K rH rji
(d rH
• -rl
P| _|_1 *!.
C rH
••
tn -P ro
M 0 CN
5 T3 if)
&H C f'O
(d id CM
S
• C CN
KC O rH
N ^
«• O IT)
> -P
O -H
M
S
T) -H K
(!) ^J
^! O •
CO Id 10
O >i
• -H A
*T3 Pi
• 51
o>
a
u HH
0) O •
rH J"3
m §
. -H •
Pi -P 01
3 C
- X! -H
• -rH -O
h> JJ O
H 0) Fn
•rH
•> O fl)
C to
0) T3 3
to a) id
rH > ft
Q) M O
•H 0) ft
fi tfl 0
3<§£
••H
fri
(1)
O
78
-------
Strongest cross-tropopause fluxes by a mean meridional circulation,
according to this table, are encountered during winter and amount to 10 x
10 g/sec of air or to 5 x 10 g of ozone per second. The annual total flux in
17 7
the northern hemisphere would be 1843 x 10 g of air or 9.215 x 10 tons of
ozone by the action of the mean meridional circulation. This value compares
to estimates of 40 x 10 g tons per year on a hemispheric basis by Junge (20),
(see also Mohnen et al. (26)) and of 65 x 10 tons per year by Aldaz (30) (see
Reiter (37)). Pruchniewicz (59) estimates a global destruction rate at the
surface of 75 x 10 tons per year, 3/4 of which is thought to occur in the
northern hemisphere. His estimate was derived from world-wide surface measure-
ments, aircraft measurements, and balloon soundings. Obviously there is quite
a discrepancy between these estimates. Aldaz' values involve the measurement
of vertical ozone concentration gradients over land and water. However, the
ozone concentrations near the ground used by Aldaz and by Pruchniewicz most
likely were contaminated by tropospheric sources, hence the computed fluxes
should not be ascribed entirely to stratospheric sources.
We will have to take into account that the major downward transport
during most of the year, accomplished by the mean meridional circulation,
occurs in middle latitudes and not over the entire hemisphere (Figure 8a,b,c,d)
A flux of 6 x 10 g/sec of ozone during the period December-February through
the tropopause of middle latitudes can be translated into a flux of 0.4 x 10
g/m sec, if this downward transport by the mean meridional circulation is
confined to only one-half of the area of the northern hemisphere. This number
-7 2
compares to a flux of 0.5 x 10 g/m sec estimated by Junge (20) (see Reiter,
(37)) as an annual average value. An annual flux of ozone of 9.215 x 10 tons
-7 2
translates into a flux of 0.23 x 10 g/m sec, if concentrated over half the
area of one hemisphere. This number does not yet include the effects of eddy
transport processes in the jet-stream region, which also have a preponderance
in middle latitudes and, on the average, might increase the above estimate by
50%. Such an increase, however, still leaves us short of the estimates by
Junge and Aldaz, probably because of anthropogenic contamination of some of
their data.
79
-------
/T-.-l-».--j
^rfs^XK ^ThZ'
; J-» ^ _ I ««..
<~ -i
i ,MEAN STRATOPAUV
.J.jt-.\ 4. ,
90 80 /O
r K. ?.. 30
March-Ma\ _
Figure 8. Mean meridional circulation (mass flow in units of 1012 g s l) for
the four seasons [from Louis].
Source: Reiter, R., H.J. Kanter, R. Sladkovic, and K. Potzl. Measurement of
Airborne Radioactivity and Its Meteorological Application. Part V.
Annual Report 1 April 1973 - 31 July 1974. Institut Fur Atmo-
spharische Umweltforshung. ERDA Document No. NYO-3425-12. 1976.
80
-------
_ _L_._i...- I..
.-••JY-5--4 < /
•-^i- ^r ) -— K — ;- i i ;
»0 80 70 60 50 40 30 20 10 0
40 50
°South
June-August
0 10 20 30 40 SO 60 70 80 90
/O 60 50 40 30 20 10
90 80
September-November.
Figure 8. (Continued)
81
-------
It is logical to assume that the mean meridional circulation, as depicted
in Figure 8, provides a certain background level of ozone of stratospheric
origin in the lower troposphere. To arrive at these concentrations we can
2
divide the flux estimates provided above in units of g/m sec by the mean
vertical velocities in m/sec prevailing in the troposphere. The result of
such computations will be concentrations of ozone in g/m .
Junge (see Reiter, (37)) estimates the mean tropospheric residence time
of ozone to be of the order of 3.3 months. An average tropopause height of 12
km thus would yield a mean transit velocity of 1.4 x 10 m/sec. Together with
-7 2
Junge1s flux value of 0.5 x 10 g/m sec this yields a concentration of 35.6
yg/m or approximately 16.8 nanobars of ozone.
On the other hand, we can estimate vertical velocities from the mass
fluxes shown in Figure 8, using the expression
,
2irR p cos
where iji is the mass-weighted stream function depicted in Figure 8 in units of
12
10 g/sec, R is the earth's radius, p is air density, and 3 is the latitude
difference in radians. Using the December-February data between 10°N and 40°N
and at a height of approximately 8 km, one arrives at w = 2.74 x 10 m/sec.
-7 2
Together with a flux of 0.4 x 10 g/m sec estimated earlier this vertical
velocity will result in a mean background concentration of O of 14.6 yg/m ,
less than half of what Junge 's data would yield. For comparison, the present
Federal standards, specifying maximum concentrations not to be exceeded for
longer than 1 hour, are 160 ymg/m .
From a comparison of these values we can estimate that the natural mean
background of O of stratospheric origin in the mid-latitude troposphere
produced by the mean meridional circulation should be of the order of 10 to 20
percent of the Federal 1-hour standard. Actually, the lower one of these two
percent values appears to be more appropriate as a conservative estimate of
•nean background concentrations, because the assumption of an ozone mixing
ratio of 0.5 yg/g at tropopause level, which entered into the above estimates,
82
-------
is a conservative overestimation. Sticksel (39) on the other hand obtains a
"tropospheric background" of approximately 80 percent of the 1-hour Federal
standard from the ozonesonde observations of the 1960's. These data are,
however, suspect of anthropogenic contamination (Reiter, (60)), and do not
consider further dilution within the planetary boundary layer.
Using a different approach, Reiter (7) estimated the mean background of
stratospheric 0 in the troposphere to be approximately 5 percent of the
Federal standard 1-hour maximum concentration. This number was arrived at by
90
estimated Sr /O ratios in the lower stratosphere and, using these ratios as
a conservative quantity, inferring the ground-level ozone concentrations from
radioactive fallout measurements as compiled by the U.S. Public Health Service
Radiation Surveillance Network. As will be pointed out later, this value of
5% may have been an underestimate.
We have to expect some interannual variability of the mean meridional
circulation and its effects on background levels of O in the troposphere.
Only a few studies exist that point out that the interannual variability of
the atmospheric general circulation is significant, without testing the spe-
cifics of transport processes (61-65). Mean meridional circulation patterns
in the stratosphere published by Vincent (66) indicate stronger downward
fluxes in the lower stratosphere of middle latitudes during January, March,
and May 1965 than during the same months of 1964. This is in agreement with
the observations in Figure 1, suggesting that the variability in the various
transport mechanisms may have as much to do with ozone variability as the
sunspot cycle. (A number of literature references explore the possibility of
correlations between atmospheric circulation patterns and the sunspot cycle.)
Our mass-budget of stratospheric-tropospheric interchange ascribes
approximately 20% of the equivalent of one hemispheric stratosphere to enter
the troposphere within 1 year through large-scale eddy mixing, mainly in the
jet-stream region. Even though this effect is less than one-half of that of
the mean meridional circulation, it will influence low-tropospheric ozone
concentrations more profoundly. Whereas vertical velocities ascribed to the
83
-------
mean meridional circulation are typically of the order of a few millimeters
per second, subsidence associated with synoptic-scale jet-stream systems is of
the order of a few cm/sec (see for example, reference 53). Such velocities
would entail a transit time of stratospheric air intrusions through the depth
of the troposphere of a few days (order of magnitude 3 to 5 days, on the
average). Less mixing will occur with these shorter transit times than with
the slow-acting mean meridional circulation. Quantitative estimates of these
mixing processes will be made in the subsequent section.
It has been demonstrated in numerous studies (see, for example, references
9,10,43,53,67-70) that stratospheric air intrusions into the troposphere on a
massive scale occur only through the "tropopause break," i.e., a discontinuity
in the tropopause level observed on the anticyIonic side of the jet-stream
systems of subtropical and middle latitudes. Especially the polar-front jet
stream with its associated frontal system offers a convenient "sliding surface"
along which stratospheric air masses can move into the lower troposphere
without excessive mixing and in a matter of a few days. For the lack of a
frontal system the "subtropical jet stream" offers less opportunity for a
rapid air mass transit through the depth of the troposphere. Therefore, this
jet-stream system is of only secondary concern when studying the effects of
stratospheric air intrusions on high ozone concentration levels near the
ground. The subtropical jet, however, still is important in maintaining the
"natural background" of ozone of stratospheric origin observed in the tropo-
sphere and discussed earlier. Wide-spread subsidence motion underneath the
subtropical jet-stream belt of winter actually contributes to the mean merid-
ional motion effects described before.
The vertical turnover and mixing of the atmosphere is more violent in
the polar-front jet-stream region than in the subtropics. Stratospheric air-
mass intrusions into the troposphere coincide with a tropopause-folding
process (8). The stratospheric air masses start their descent in the rear
left quadrant of a jet maximum (Figure 9) in the course of the development of
a cyclonic disturbance of the westerly flow (68,71). Mahlman (11) has shown
that the magnitude, or amplitude of such a disturbance — expressed in terms of
a cyclone index that measures the meridionality of wind directions in mid-
84
-------
latitudes and in the upper troposphere — is positively correlated with the
amount of stratospheric aJr extruding into the troposphere and/or with the
rapidity with whirv_ it traverses the depth of the troposphere.
STRA TOSPHCKC
TROPOPAUSE
iSENTRCPIC
J SURFACES
DEBRIS
TRAJECTOHV
\ FRONTAL ZONE
Figure 9. Schematic three-dimensional view of mass flow from stratosphere to
troposphere near a jet stream. Isentropic surfaces are indicated
by thin lines. Surfaces of constant wind speed, boundaries of the
frontal zone, and the tropopause are marked by heavy lines.
Source: Reiter, E.R. A Case Study of Radioactive Fallout. J. Appl. Meteorol.,
2(6):691-705, 1963.
Unfortunately, at this time we have only a very poor statistical grasp of
the effects of the intensity of cyclonic disturbances on the amount of strato-
spheric air injected into the troposphere. Our only estimates refer to a few
case studies which were of strong enough intensity to produce such fallout
(9,10,53) and one which did not (70). The intrusions which produced radio-
active fallout were massive and quick enough to reach the lower troposphere in
12
a relatively undiluted state and contained approximately 0.4 to 0.6 x 10
metric tons of stratospheric air. The intrusion studied by Reiter and Mahlman,
which did not reach the ground, hence contributed only to the general and
diluted radioactivity and ozone burden of the troposphere involved only about
12
0.25 x 10 metric tons of air. There probably is a cut-off value, not only
in terms of Mahlman1s cyclone index, but also in terms of tonnage of strato-
spheric air intrusions which must be exceeded before relatively undiluted
stratospheric air comes to rest within a subsidence inversion capping the
planetary boundary layer or mixing layer above the earth's surface. We do not
know precisely where these cut-off values lie, but we could guess that they
85
-------
are near 2.4 for the cyclone index (for its definition, see reference 71} and
12
near 0.4 x 10 tons for the mass of air involved in the intrusion.
Mahlman (11) counted 31 cyclonic disturbances over the North American
sector (70°W to 180°N, and 40°N to 60°N) during 1963, 23 of which met or
exceeded the cyclone index value of 2.4 (Figure 10). During 1964, again, 31
disturbances were counted, 23 of which met or exceeded the "critical" value,
suggesting little or no interannual variability of the large-scale eddy trans-
port processes in the jet-stream region between these 2 years. The same
conclusion can be reached from McGuirk's (72) analysis of eddy available
potential (A ) and kinetic (K ) energies computed on a hemispheric basis.
Mahlman1s analysis (Figure 10) shows little, if any, seasonal variability
of the number of cyclonic disturbances passing the North American sector, nor
does it reveal a significant seasonal march of the intensity of these episodes.
McGuirk (72) showed from a 10-year data base that the mean seasonal double-
amplitude (the difference between maximum and minimum values) for A is about
65% of the mean value (Figure 11). It is approximately the same for K,.
Maxima in both parameters occur in January, suggesting, on the average, a
tendency of stronger stratospheric air intrusions during winter than during
summer. According to Figure 4, highest stratospheric ozone concentrations in
the (lower) stratosphere occur in April. Because of the vertical velocities
characteristic of the mean meridional circulation, one should expect a delay
by 1 month of the appearance of the ozone maximum of stratospheric origin at
the ground, moving it into May (see also, reference 73). The jet-stream
transport effects, superimposed upon the mean background and modulating it
with the passage of cyclonic and anticyIonic disturbances, should yield — on
the average — maximum stratospheric ozone effects near the ground in mid-
latitudes during March and April. As we can see from Figure 10, the radio-
active fallout distribution over the United States, and presumably also the O
ground-level concentrations due to stratospheric import, peak in April-May.
The average mass transport from the stratosphere to the lower troposphere in
mid-latitudes, therefore, is more strongly influenced by the seasonal varia-
bility of the stratospheric reservoir of radioactivity or ozone, hence by the
86
-------
SI
»<
U
*
4
7t
1 i i
<— ' ' ^
. > 1
*>
L ^Jv
•v— U
T
.•t*..
! ~.
It-
3
r T
•O • 4-> >i
0 U) U O
•H 0) Id
W -H g
ft 0) -P M
W 0) M-l
0)
-P
0) n5
C 0)
O 4-1 H
i—I •<] Cn
U \
>i -» 0)
U (N S-l
u id
-o
•H I U)
ft in
id •—< id i
^-t O ,0
[fl
0) X
4J O
O C O -H
x: -H ^ H
Cn
Q) O
4-1 O
W
Q) rH
tl C T) id
C 0 0) O
(d H 4-> -H
U W 4-1
O O -r-i 0»
* ^ id
X 0) I
fl) .C (1)
T) 4J Cn
C 0 id d)
•HO 3
CO
o> co
0 to
H -H
U
O
C CT> I
d) id
0) -H
id
4->
id o
TJ rH
I rH
CO O
(d co
T) 0)
p
in rH
id
d >
•H
,C M
•H id
0)
in !H
0
id o
0) rH
x; a)
xi
0)
> 0)
0 ^
•8 *
CQ
U] ^
M id
(U ^3
a fi
3 -H
2 43
4->
.Q 0 CO 4->
(1) C
S f C i
0 M -H g
CQ (d rH
•H ft C
M T3 id
id M o u
ft a; 4J e
g ft O
O ft 0) CO
o p g a)
c >
m o -H
o • o CT
co
co c c 0)
0) 0 -rl C
•H -H £ -H
M 4-> 4-> rH
0) id
W 3 Si
4-1 " -P
(1) 0
CO rH
id o)
i
id J5 o IH
4-1 g o
D 2 U 0)
o (d 3
M 0) rH
0) • id
ft CO Cn >
0) -
en id
-H (1)
c
•H rH
!M id
3 U
O 4J
0) Q) T3 -H
43 13 0) H
4-1 4-) O
V '"t
l"l r-*
ftW CO-H U4->
O
rH
0)
Cn
•H
4-) 0)
3 >
O -H
•H C
rH D
id
PM 0)
4J
0) id
> 4J
•H CO
O O
O id
•H ^
T3 O
Id rH
K O
O
CO O
0) r~
£ •
id o
43 2
U M
X 0)
•8 &
C ft
(U
0)
U
o
to
8 s1 a
X30M
87
-------
M.O
M.O
OL
UJ
UJ
10.0
30.0
20.0
GIO.O
re
PE
10.0
30.0
cc
UJ
UJ
M.O
30.0
20.0
(K
UJ
KZ
KE
-1 1-
100.0
90.0
80.0
.0
60.0
i
"50.0
G'0.0
.0
in
O
acr
68
68
Figure 11. Daily values of energy by mode and for the total of available
potential plus kinetic energy from July 1967 through June 1968.
The smoothed line represents the annual energy obtained from
averaging nine individually smoothed annual energy series.
Source: McQuirk, J.P. Fluctuations in the Atmosphere's Energy Cycle. En-
vironmental Research Paper No. 6, Colorado State University, Fort
Collins, Colo., 1977.
88
-------
mean meridional circulation in the stratosphere, than by the seasonal behavior
of jet-stream and cyclogenetic activities. This statement is corroborated by
Table 1.
Future investigations will have to establish a correlation between the
intensities of cyclonic disturbances in mid-latitudes and the amount of
stratospheric air transferred into the troposphere by these disturbances.
Also a better frequency count of these disturbances around the whole hemi-
sphere would be quite useful.
The impact of large-scale eddy motions in the jet-stream region of mid-
latitudes on surface ozone concentrations has been estimated by Reiter (50,74).
Average O concentrations in the lower stratosphere as established by Dutsch
90
(49) for the appropriate season were compared to Sr distributions measured
by aircraft or balloon (75). An example for 1963 is shown in Figure 12.
Corrections were made for the depletion of radioactivity from the stratosphere
and for seasonal O variability to bring different observational periods into
agreement with each other (Figure 13).
Surface radioactivity measurements (reported as total activity in pCi/m )
were available from the U.S. Public Health Service Radiation Surveillance
90
Network. The assumption was made that the Sr /O concentration ratios in-
ferred for the lower stratosphere would remain constant during rapidly proceed-
90
ing stratospheric air intrusions into the lower troposphere. Sr and 0 of
stratospheric origin would be subject to the same transport and mixing proc-
90
esses. Hence, if the Sr fallout concentration levels at the ground were
known, the stratospheric contribution to concentrations of O associated with
the same air mass could be derived.
Since surface radioactivity was given in terms of total activity, esti-
90
mates had to be made of the specific Sr contribution towards this total
activity. Assuming that during spring of 1963 the bulk of the radioactive
debris in the stratosphere was less than 1 year old, Reiter (74) derived a
90
contribution of Sr towards total radioactivity of 1 percent from fission
yield and radioactive decay curves. Table 3 (11) shows that such a percent
89
-------
o
o
C\J
o o o o o
to ^- m r- o
CM
T3
(fl O
O
— o
to -
s-i to
CU t-t
A CU
C rH
•H nJ
4-> U
fi -H
(d 4J
rH H
CO OJ
w •«
(1) W
C 0)
•H C
rH -H
i-H
•d
•H T)
rH 0)
0 ^3
W CO
^ nj
T)
S-i ro
CO
0 I
>W >i
§ s
•H r)
-P O
3 IH
••3 C
r) O
•P -H
(0 -P
•H 3
•d J3
•H
(]) rl
C -P
o w
N -H
0 T!
Co X
njon PM
0) rl U
CN
rH
CU
fc
3
&i
•H
CU
N
Id rt
CQ W
-•H
M O
43 -P
0)
Q ti
0) Q
> a
•H
C O
(d 0) 2
M M
EH CU -
^ 0)
0) ft -P
£ 0) 3
EH 0 -P
-P -H
(0 -P
• !H w
« -P C
• W H
w
CU X!
» A O
M -P r<
CU «J
•P 6 CU
•rl 0 W
-------
c
X
1
t/
o
cr>
I
t
>
f
'
/
,*'
/
HI
/'
/7
/
/
i 'i
/
'
s
369 12 369 12 36 9 i;
1963 1964 1965
90
Figure 13. Sr/0 ratios as a function of time (see text for
explanation) .
Source: Reiter, E.R. The Transport of Radioactive Debris and Ozone
from the Stratosphere to the Ground. Report to Stanford
Research Institute. Nov. 22, 1975. 36 pp.
f~\ o O ^ O O O O O
? j2 CD m ^ ro co -
5
91
-------
CO
H
Pi
9
W
CO
W ^
5 rij
H Q
U 2
•a; H
O
H W
Q Q
f£ H
K
E"i K
2 U
o «;
& H
1 "1 ftj
Si o
S W
VJ m
EH fa
H
2 r-1
H 1 •
fa CO
co a w
O EC W
1 i jw
i_^ f_i
U • 2
CO <
« « ft
<5 H
3 to 2
D H
O IS
H W 2
OH ^
< •• H
ft Q O
^^
fa O CO
O 5 H
r~4
2 co
O W
M !S
r_l
3 EH
ffl rt!
IH
02 iJ
£ 3
O fa
U 2
M
a <
O 05
c_| ]2j
23 H
W
U Q
« W
H «
£L| ^
CO
ro §
I—I
§
H
W
Q
H
l_^
CJ
P
2
fa
o
O
H
EH
D
m
3
EH
£j
O
U
W
u
K
W
ft
rH
^3* -*^
rH tf
H
1 ,
1 •
ro
0 ro
U ^
«^^
r- T3
ro o
rH ^
rH
^4
Ut rH
O ~
in -—
en -c
o
m
^4 VO
CS3 ~--
^
^* x—*
rH Tl
in
1 00
CN
u
en ~
00 T3
m
o
>H m
CO ^
s~*
O TJ
en o
CN
1 rH
r4 O
CO rH
*~^
Of
rH o r~ rr CN o
m in o ^ o m
rH rH
#
^VDrOCOCNrHCN^f'sl'^J1
HrHCNCNrO^l'invOVOlO
df
OOCNOOOOCNCNVDOrrtN
^•^roincovoinoinro
^•rOrOCNCNCNCNCNrHrH
dP
in CN co r^ r^ vo VD CN ^ ^0
rH^vDen^rHinincncN
CN oo ro ^* ^* in in vo vo r^
*
cNoootN^rHvoinencn
^DrocomrHocnin'd'ro
CN CN rH rH rH i-H
encMocNcooocncn
OrHrHcNCNCSifrororo
ro
^O
en
rH M
>1 0
>1 r4 43 r)
M rd 4-> g 0
id 3 42 rH W 0 XI
3MU-H 0>,34JO
CXlV4>H>iCirHC7lft4->
rd0rdftid3330U
i-3faS1
•i -i s
003
> U C
0 0 rd
2 Q I'D
•P
O
i-l
rH
id
fa 0
o> 4->
•rH CO
4J
0 O
rd *O
o id
•a o
Id rH
OS O
u
o
4J
•
to o
0 r--
en
§ '
X! 2
0
)H en
X 0 o
(1) ft rH
'O Id 1
C ft *?
H CO
H
rH Id •
0 O ft
t> -rt ft
1 O •
0 0 ro
CO EH VD
3 en
Id 0 rH
ft 0
o a -
ft 0 •
O -H 0
H U rH
P-i rn rt
t1 UJ W
u
«tH U
O -H *
M 10
C 0 C
O 45 -H
•H ft rH
4-1 W rH
id o o
rH g U
0 4J
OS rfj 4-1
O
• 'fa
Q to
• C •>
i-D 0 ><
•H 4J
•>+>•*
c id to
id 3 in
g 4J 0
rH 0 >
43 3 -H
Id rH fi
S fa D
0
U
^|
3
O
CO
92
-------
value would, indeed, have been correct for January and February 1963. For the
first half year of 1963, however, a cont:
have been a more appropriate assumption.
90
first half year of 1963, however, a contribution of 2 percent by Sr would
Figure 14 shows the mean fallout distribution for the period January-July
1963 in units of pCi/m . For the period March-April 1963 the lower-strato-
90 90
spheric Sr /O ratio was determined to be 620, with Sr concentrations given
in dpm/1000 SCF (disintegrations per minute per 1000 standard cubic foot) and
O concentrations in yg/g. Assigning a value of 2 percent to the contribution
90 3
of Sr to total fallout, 1 pCi/m in Figure 14 would correspond to approxi-
mately 0.002 yg/g of o .
Disregarding the local fallout pattern around the Nevada test site,
maximum mean fallout (measured over sampling periods of 24 hours) in Figure 14
of 9 pCi/m corresponded to roughly 0.018 yg/g of 0 . By comparison, the
Federal standard 1-hour maximum concentration is 0.1238 yg/g. Thus, the
distribution in Figure 14 indicates that the background concentration of
stratospheric O near the ground in the mid-latitude jet-stream region might
rise to around 15 percent of the Federal 1-hour standard. This value is close
to the one ascribed earlier to the effects of the mean meridional circulation.
Figure 15 shows that over the southeastern and eastern United States, but
also along the Rocky Mountain Front Range, there is a significant number of
days per year during which background levels of 0 of stratospheric ozone
measure higher than 0.02 yg/g. Again, we have to disregard the local effects
of the Nevada test site. Maximum 24-hour concentrations of radioactivity
observed during 1963 would suggest ground-level 0 concentrations (averaged
over 24 hours) of 0.05 yg/g or slightly less than one-half of the Federal 1-
hour standard (Figure 16). Similar maximum 24-hour 0 concentrations of
stratospheric origin at the earth's surface could be inferred from the 1964
90
maximum fallout distribution, allowing for a higher Sr contribution to total
fallout during that year.
Mean 24-hour O concentrations measured at Zugspitze Mountain, Germany
(3000 m above MSL), are shown in Figure 17 (51). The mode value of observa-
93
-------
•H
U
ft
n
vo
en
hJ
&
4J
3
o
iH
i-i
(0
Iti
-
2
3
T3 U-t
0)
o
0
w
94
-------
cu
o
N
O
(d id
-P
a to
-H
H 0
J3 -P
(!)
Q -P
Q)
>
0
Ck
ft
O U5
(fl ro
O •
as .
rd 3 in
K 0 r~
H <^
u-i O H
0
(!) «
+J .C (N
^ -P OJ
ft O •
W -P >
C 0
(d 0) 2
)-i H
EH fl) ••
^1 0)
(U ft -P
X W 3
E-i 0 -P
•P -H
td -P
M tn
•P C
en H
a) .c
tf
•
W
" A
H -P
0)
.p
rd
0)
•HOW
0)
O
O
cn
95
-------
•H
U
ft
(S
in
3
O
X
nl
S
cu
M
3
tn
•H
fe
a
°8
•a ft ft
•H (U ft
o
g .
^•g
VD
4-
C
m
r~
en
a) -
U3 CN
-P CN
(-1
ft 0 •
to -P >
c o
(0 0) 2
M M
^ (U ^
-------
•/.FREQUENCY
80
70
60
30
20
10
50
40
30
10
40
10
50
30
20
10
DEC JAN FEB
89
MAR APR MAY
27
JUN JUU AUG
101
SEP OCT
177
NOV
0 10 20 30 40 !>Q l>0 70 ; ppb 0,
o
Figure 17. Percent frequency distribution of ozone concentrations (ppb)
observed at Zugspitze, Germany, from August 1973 to October 1975.
Number of observations is indicated in each histogram.
Source: Reiter, E.R. On the correlation between cosmogenic Be-7 and ozone.
Report to Stanford Research Institute, 23 December 1975.
97
-------
tions appears to lie close to one-half of the Federal 1-hour standard value
(indicated by the dashed-dotted line in Figure 17). Allowing for additional
dilution between 3000 m and the general terrain level of the eastern United
States, the estimates of 0.05 ug/g of 24-hour averaged maximum ozone concen-
trations made above appear to be reasonable. Aircraft measurements below the
tropopause near 10 to 12 km reported by Pruchniewicz et al. (76) yield similar
mean concentrations (Figure 17a).
u.u •
010 -
008 -
1"006 -
<
-?
^ 004 -
O
M
° 002 -
o>
i_2j
<
o009-
2£
* 006 -
OOi -
002 -
nnn -
25° 5 • 0°
]
i i I i i i i i i i i
I I 1 1 1 1 1 I 1 1 1
0° - 25° N
1 1 1 1 1 1 1 1 1 1 1
MAR. APR MAY JUN JUL AUGSEPIOC1 NOV DEC JAN F6B
Figure 17a. Mean seasonal variation of tropospheric ozone from
airplane measurements for the areas 0° -25°S and 0°
-25°N.
Source: Pruchniewicz, P.G., H. Tiefenau, P. Fabian, P. Wilbrandt,
and W. Jessen. The Distribution of Tropospheric Ozone from
Worldwide Surface and Aircraft Observations. In: Proceedings
of the International Conference on Structure, Composition
and General Circulation of the Upper and Lower Atmospheres
and Possible Anthropogenic Perturbations, Vol. 1. Melbourne,
Australia, 1974. pp. 439-451.
98
-------
From these circumstantial, and heavily smoothed, data it appears that the
background level in the lower troposphere of 0 of stratospheric origin, at
times, will reach significant fractions of the Federal 1-hour standard value.
It should not be surprising, therefore, if, on occasion, this value would be
exceeded. This possibility will be investigated more closely in the subse-
quent section.
In Table 1 the effects of seasonal adjustments in tropopause height have
been estimated to contribute roughly 10 percent to the stratospheric mass
budget. These adjustments in tropopause height occur mainly in the transition
season and in the jet-stream region of mid-latitudes. It would be difficult,
therefore, to separate this effect from the ones already described above. The
radioactive fallout distribution discussed earlier already includes this
seasonal behavior of the tropopause level which is, to a large extent, caused
by the dynamics of jet-stream systems. Therefore, we will not attempt to
separate estimates of its contribution to surface 0 concentrations. Small-
scale diffusion near tropopause level is also thought to contribute to the 0
problem in the lower troposphere only at the noise level.
THE LIFE HISTORY OF STRATOSPHERIC INTRUSION EPISODES
Several case studies of intrusion of stratospheric air into the tropo-
sphere have been reported in the literature (9,10,53,70). From these studies
the following general pattern emerges.
The "tropopause folding" process injects stratospheric air into the upper
part of the frontal zone that is connected to the cyclogenetically active jet-
stream system (see Figures 7 and 9). The, extent of this stratospheric air
intrusion can be delineated very well by potential vorticity
ao
where Qn is the vertical component of absolute vorticity on an isentropic
surface (on which potential temperature, 0, is constant by definition), g is
the acceleration due to gravity, 80/9p is the vertical potential temperature
lapse rate in a pressure coordinate system and provides a measure of thermal
99
-------
stability. P is a conservative quantity for adiabatic transport processes.
Non-adiabatic processes, which include the effects of radiative warming or
cooling of the air mass under consideration and of turbulent mixing with other
air masses, render P not conservative.
Because of the high stability, -g r—, of stratospheric air, and because
dp
of the cyclonic vorticity (QQ > O) prevailing in the jet-stream region where
the stratospheric air intrusions into the troposphere tend to occur, P assumes
relatively large values within the intruding air mass (order of magnitude 100
-9
x 10 cm sec deg/g). By comparison, the tropospheric air outside these intru-
-9
sions has typical potential vorticity values of only a few units times 10 cm
sec deg/g. Mixing of stratospheric air with the surrounding tropospheric air
will gradually reduce the potential vorticity within the intruding stable air
layer.
If this air layer subsiding from the stratosphere is cooled uniformly by
radiative heat losses as it descends through the troposphere, its stability,
80
- g -jj—, should not be affected. We can, therefore, in a first approximation,
ascribe decreases of P along an isentropic trajectory of this air mass to the
effects of turbulent mixing.
Figure 18 shows a stratospheric air mass intrusion in its final stages as
it settles into a "subsidence inversion" above the anticyclonic system behind
a cold front. We note the following decreases of P during several 12-hour
trajectory segments:
TABLE 4. DECREASE OF P ALONG 12-HOUR TRAJECTORY SEGMENTS
-9
From 16 to 11 x 10 cm sec deg/g = 31% decrease
11 to 11 = 0%
16 to 13 = 19%
22 to 10 = 55%
24 to 16 = 33%
54 to 50 = 7%
100
-------
PLOTTING MODEL
POTENTIAL VORTICITY
MONTGOMERY STREAM FUNCTION
PRESSURE
THICKNESS.OF STABLE LAYER
II
30369
Figure 18. Trajectories on the 300°K isentropic surface. Dashed lines, Nov.
22, 1962, 0000 to 1200 GMT; full lines with arrows, Nov. 22, 1200
GMT, to Nov. 23, 0000 GMT. Values of potential vorticity (units
of 10 9 cm sec deg/g), of Montgomery stream function (units of 107
cm2/sec2), of pressure (mb) of the 300°K surface, and of thickness
(mb) of the stable layer are entered according to the plotting
model; slanting numbers of Nov. 22, 0000 GMT, and vertical numbers
for other map times. The centers of the hatched bands mark bound-
aries on Nov. 22, 1200 GMT, and Nov. 23, 0000 GMT, of strato-
spheric air from the tropopause level that reaches the ground over
the southern United. States.
Source: Reiter, E.R., and J.D. Mahlman. Atmospheric Transport Processes
Leading to Radioactive Fallout over the United States in November
1962. In: Radioactive Fallout From Nuclear Weapons Test, Proceed-
ings of 2nd Conference, AEC Symposium Series 5, 1965. pp. 450-463.
Among these widely scattering values a 20 percent decrease of P due to
mixing appears to be a reasonably conservative assumption.
An ozone concentration of 80 nanobars at 300 mb, corresponding to 0.44
> should, therefore, be reduced to approximately 64 nanobars, or 0.35
f by the time it reaches the lower troposphere in a massive stratospheric
air intrusion. This is approximately three times the Federal 1-hour maximum
level. For a specific case of radioactive fallout over the southern United
101
-------
States, Reiter and Mahlman (16,53) estimated that air within the "subsidence
inversion" of stratospheric origin was mixed with air from the planetary
boundary layer underneath with a ratio of 1 part to 3 parts. With the esti-
mates made above, such mixing still could produce 0 concentrations somewhat
in excess of the 1-hour Federal maximum value.
If one were to attempt an estimate of the frequency of such excessively
strong intrusions per year, one is confronted with rather shaky evidence at
this time. More detailed investigations, relating perhaps potential vorticity
values in a wide range of such intrusions with the intensity of cyclogenetic
activity associated with these intrusions, would have to be conducted before
one could arrive at probability estimates of exceedence of the Federal stand-
ards by stratospheric ozone with some degree of reliability.
If, at this time, we were to ascribe rather arbitrarily a cyclone index
of 4.0 as the cut-off value for such massive air intrusions into the tropo-
sphere , we can estimate from Figure 10 that perhaps 10 to 12 occasions per
year would arise over the United States, in which the Federal 1-hour maximum
standard is approached or exceeded. The area covered by such intrusions could
be "Texas-sized" (3) in its horizontal dimensions, covering perhaps 1/20 of
the United States. We have to realize that the southeastern United States are
more prone to be influenced by such stratospheric air mass intrusions than the
southwestern or northern tier of states. Realizing this geographic bias in
radioactive fallout, hence in stratospheric ozone intrusions, one might cau-
tiously peg the probability of local exceedance of the Federal 1-hour maximum
at one case per 1 to 2 years. From different evidence, presented by sporadic
ozonesonde observations, Reiter (60) estimated this probability to be 0.2
percent, measured in days of observations on an annual basis at a given loca-
tion. This estimate agrees well with the one derived above. (Reiter's proba-
bility estimate from ozonesonde observations considers only ozone peaks observed
in the planetary boundary layer below 750 or 800 mb, depending on location.
Carney (35) arrives at higher probability estimates of approximately 11 per-
cent by considering ozone peaks up to the 500-mb level and a "critical" con-
centration of 0.10 ug/g instead of 0.124 yg/g. Considerable dilution should
102
-------
be expected of ozone peaks observed above the planetary boundary layer during
their transit to short distances above the ground.
The conclusion that can be reached so far, therefore, is that ozone
concentrations of stratospheric origin observed near the ground can, on occa-
sion, exceed the Federal maximum 1-hour concentration. Such cases, however,
are rare and, on the average, far between. They are expected to occur perhaps
once a year over the southern and eastern United States.
Careful insentropic trajectory analysis techniques provide a relatively
reliable tool to diagnose such events during which ozone of stratospheric
origin can reach relatively high concentration levels near the ground (see,
for example, 77). Forecasts of such events 24 to 48 hours ahead of time could
be made with reasonable reliability. Such forecasts, however, would require
computational work of a specialized nature which, most likely, would be con-
sidered burdensome by the National Weather Service.
CONFLICTING EVIDENCE FROM DIRECT OZONE MEASUREMENTS
The scientific literature of recent years is full of claims and counter-
claims as to the importance of stratospheric air intrusions for ground-level
ozone concentrations. We will examine here several of these claims and point
out possible sources of errors which might have an effect on the conclusions
derived from these measurements. Some of the discrepancies appearing in the
published data cannot be resolved easily and will have to await clarification
by additional, well-planned and well-executed field experiments.
Measurements of O concentrations on Zugspitze Mountain (3000 m above
MSL) Germany, on the one hand, and from Whiteface Mountain (1518 m MSL), New
York, and from Hohenpeissenberg (977 m), Germany, on the other hand can be
cited as proponents of opposing evidence.
Reiter (34) analyzed daily 1-hour maximum 0 concentrations observed on
Zugspitze and correlated them with the daily mean values of 0 concentrations.
103
-------
Figure 19 shows the results of 529 days of data analyzed in this fashion.
Federal maximum 1-hour standards were exceeded on 2 days during several
consecutive hours comprising one continuing episode (Table 5). (In Figure 19,
the limiting Federal standard value is indicated by the hatched line.)
Observed maximum exceedance was by more than a factor of two.
The episode of these excessively high O concentrations observed on
Zupspitze coincided with a rather anomalous stratospheric circulation pattern
(Figure 20) with a pronounced trough over eastern Europe. This trough situa-
tion could have produced an enrichment of O in the lower-stratospheric reser-
voir prior to the tapping of this reservoir by a tropospheric cyclonic disturbance.
The relative infrequency of exceedance of Federal standards on Zugspitze
Mountain is in good agreement with our earlier probability estimate of 0.2
percent of local occurrence in mid-latitudes. Nevertheless, the question of
instrument inaccuracies at Zugspitze has been raised (Dr. W. Attmannspacher,
Director of the German Weather Service Observatory at Hohenpeissenberg; oral
communication made at the Dresden Meeting, July 1976) and refuted by recent
calibrations (Dr. Reinhold Reiter, Director of the observatories in Garmisch-
Partenkirchen and on Zugspitze Mountain; letter communication, 1976). Whereas
for reasons of agreement with deductions on expected O concentrations pre-
sented earlier in this paper, there should be no quarrel with the Zugspitze
measurements, a closer check on the reliability of various measurement tech-
niques may be called for.
Measurements on Whiteface Mountain reveal a frequent exceedance • of the
Federal standard value for 1-hour maximum concentrations. Even average monthly
concentrations approach this value dangerously close (Figure 21), closer than
the frequency distribution for Zugspitze shown in Figure 17 would indicate
(78) . Both the Whiteface Mountain and Zugspitze data indicate relatively high
average O concentrations during summer rather than spring. We should keep in
mind that during the spring season the stratospheric reservoir shows the
highest O concentrations. Figures 17 and 21 would indicate, therefore, at
least in a preliminary way, that tropospheric photochemistry and O generation
cannot be excluded, not even on Zugspitze mountain.
104
-------
ppb (MAX.HOURLY)-(DAILY MEAN)
50
120.40 141.16
I I
10
2 12 32 39 61 95 117 75 62 21 85 529
10 20 30 40 50 60 ppb DAILY MEAN
Figure 19. Difference between maximum hourly ozone concentrations (ppb) and daily mean con-
centrations as a function of daily mean concentrations, observed at Zugspitze (3000
m above MSL) between August 1.973 and February 1976. The frequency distribution by
5-ppb classes of the daily mean and 2-ppb classes of hourly maximum minus daily mean
is given by the numbers in the diagram. The dashed line indicates the limits of the
present data distribution. To the right of the shaded line the federal maximum
value of 74.7 ppb would be exceeded. Dots indicate the mean values of (max. hourly -
daily mean) in each class of daily mean values. The solid line gives an approximate
best fit to these dots. The cross marks the mean value of both distributions, that
of (max. hourly - daily mean) and that of the daily mean values. The dashed-dotted
line approximates the position of the mode values in each class of daily mean concen-
trations. Note that two observations fall outside the plotted distribution.
Source: Reiter, E.R. The Role of Stratospheric Import on Tropospheric Ozone Concentrations.
Int. Conf. Ox. Poll., Proc. 1:393-410. EPA-600/3-77-001a. Environmental Protection
Agency, Research Trianyle Park, N.C., 1977.
105
-------
TABLE 5. HOURLY OZONE CONCENTRATIONS (PPB), ZUGSPITZE (GERMANY)
ON JANUARY 8 AND 9, 1975
Time I
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
January 8
hr. 25.82 ppb
27.28
27.94
30.08
31.30
30.93
28.07
27.37
26.91
27.69
37.78
47.88
42.88
50.81
62.21
52.92
52.06
54.83
57.03
62.80
79.85
114.10
157.74
197.55
January 9
158.89 ppb
170.36
171.44
159.78
81.33
28.50
25.33
27.79
30.31
30.71
31.43
31.82
32.12
32.22
25.61
25.18
24.88
22.37
22.39
19.84
19.10
17.92
16.89
17.17
Source: Reiter, E.R. The Role of Stratospheric Import on Tropospheric Ozone
Concentrations. Int. Conf. Ox. Poll., Proc. 1:393-410. EPA-600/3-77-
OOla. Environmental Protection Agency, Research Triangle Park, N.C.,
1977.
106
-------
.p
45
U
•H
ft H
43 H CJi
M (3
0 • -H
ft ° .H
EH I 45
M O
C CTi H
O ro flj
.. 0)
4J H 01
0 •
•H
b
ft O
0 C
-H - 0)
fl) r4 ,<
45 r-H
ft O C
0> ft O
O -H
4J . +J
m x u
M O (!)
^j ^j
w • o
M-l M
1] [ C* ft.
0 0
U rH
0) (0
H • 4->
O -P C
tf C g
(i) 2
45 O
EH • ^
to -H
C >
• 0 C
at -H a •
• -p 1^
W id r-
M • CTi
- -P (0 H
H C H
-------
70-1
SO
40
N 30
20
10-
JFMAMJJASONDJFMAM
1973 | 1974
Figure 21. Average monthly ozone concentrations recorded at summit of Mount
Whiteface.
Source: Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban
Areas of New York State. Int. Conf. Ox. Poll., Proc. 1:89-96. EPA-
600/3-77-OOla. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
The interpretation of Figures 22, 23 and 24 by Coffey et al. (78) is
correct: Since the Whiteface Mountain curve of hourly O concentration
measurements envelops the other curves obtained from lower elevations, the
major source of O should be sought aloft and not near the ground. The eai
surface has to be considered as the major ozone sink, as witnessed by the
strong diurnal modulations of 0 with
(Figure 22) and Syracuse (Figure 24).
strong diurnal modulations of 0 with nighttime minima observed at Pack Forest
How, then, can we reconcile our earlier conclusion that ozone from
stratospheric sources rarely exceeds Federal 1-hour standards near the earth's
surface with Coffey's conclusion that the major ozone source influencing rural
areas in the (eastern) U.S. has to be sought aloft? We can present the follow-
ing arguments.
108
-------
.16
.14
.12
.06
04
02
WHITE FACE
PACK FOREST
8/6/73 7
NOON
10
13
14
15
16
17
DAYS
Figure 22. Ozone concentrations at Whiteface and Pack Forest from August 6,
1973 to August 17, 1973.
Source: Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban
Areas of New York State. Int. Conf. Ox. Poll., Proc. 1:89-96. EPA-
600/3-77-OOla. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
HOURLY AVERAGES FOR JULY 73
WHITEFACE
PACK FOREST
GLENS FALLS
<0 12 14 16 18 20 22 24
HOURS
Figure 23. Hourly ozone averages at Whiteface, Pack Forest, and Glens Fall
sites for July, 1973.
Source: Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban
Areas of New York State. Int. Conf. Ox. Poll., Proc. 1:89-96. EPA-
600/3-77-OOla. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
109
-------
14-
.12
.10
.08
.06
.04
.OZ-
. WHITE FACE
SYRACUSE
WHITEFACE
UTSAYANTHA
.12
.10
.Oi-
.06
.04-
OZ-
0
1 12 Z
NOON
10
12
13
14
15
16
DAYS
Figure 24. Comparison of ozone concentration at Whiteface site with that at
Utsayantha and Syracuse sites for the first 17 days of August,
1973.
Source: Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban
Areas of New York State. Int. Conf. Ox. Poll., Proc. 1:89-96. EPA-
600/3-77-OOla. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
The depth of the mixing layer (ML), as defined by Holzworth (79), and its
diurnal variation controls to a large extent the pollution burden of the
planetary boundary layer (PEL) (see, for example, reference 80). For the
purpose of long-range transport studies we have to acknowledge the fact that,
whenever the low-level inversion breaks up, pollutants escape from the PEL and
are mixed into a deeper layer of the troposphere. Under most of the anticy-
clonic conditions, when cloudfree skies permit warming of the PEL during the
morning hours and a breakup of the inversion before noon, this additional
mixing will not proceed all the way to the tropopause, encompassing the whole
depth of the troposphere. If it did, the ascending thermal plumes would
result in the development of cumuli congest! and of thunderstorms, which are
110
-------
very efficient removers of pollutants. Since such is not the case under most
anticyclonic weather patterns, we have to conclude that a second mixing height
is reached during the daytime somewhere in the lower to middle troposphere,
perhaps between the 1500 and 3000 m levels. (Adiabatic lifting of air by 3000
m with less than 20% relative humidity would not yet produce clouds.) Holz-
worth's (79) chart of maximum mixing layer heights for July (Figure 25)
provides a first guess of the effectiveness of daytime mixing.
Pollution precursors for photochemical ozone generation in the upper part
of this mixing layer, which may encompass the lowest 2000 to 3000 m of the
troposphere (see, for example, reference 81), will travel long distances
relatively undisturbed because they are not in direct contact with the ground.
Little is known at the present about the fate of pollutants in the layer
between the nighttime inversion and the daytime maximum mixing height. Ongo-
ing efforts (80) are geared toward determining the transport patterns and
diffusion mechanisms controlling the pollutants in this layer.
Measurements made at the Hohenpeissenberg Observatory indicate frequent
high 0 concentrations similar to the ones observed on Whiteface Mountain
(Figure 26) (Attmannspacher, oral communication). Occurrence of high 0
values prior to thunderstorms need not necessarily herald the advent of strato-
spheric O . It could simply mean the advection of O from regions in the
upper reaches of the daytime mixing layer that have not yet lost their ozone
by contact with the ground. By virtue of the long-range, "undisturbed" trans-
port processes acting in the upper part of the daytime mixing layer, the
source of this ozone, or of its precursors, may have to be sought far upstream
and many days before its appearance over the measurement site. It appears
that Mohnen et al. (26) jumped to conclusions when they tried to link high
ozone concentrations in the upper region of the daytime mixing layer to strato-
spheric air intrusions.
In the light of the foregoing discussion we might accept as real the
disagreement between ozone measurements at Zugspitze and at Hohenpeissenberg.
Hohenpeissenberg is well within the range of the daytime maximum mixing layer
111
-------
20 18 IS J4 ..40.
Figure 25. The mean maximum mixing depth for January and July. These data
were computed from atmospheric temperature soundings obtained at
45 points in the United States.
Source: Holzworth, G.C. Estimates of Mean Maximum Mixing Depth in the Con-
tinguous United States. Monthly Weather Review, 92(5):235-242, 1964.
112
-------
19"
18"
-3.0KV/ni
Bodenozon
Wmdrichtung
Niederschlagssumme
100
Luftelektrische Feldstarke
— Relative Feuchte
— Lufttemperatur
T
200
300
l^J
,J_«—-" L_
IQO
-1.5KV/m
•1.5KV/m
W
•C-5
_500r
.30KV/m
RR.eine Stufe 0.09mm msges 3.17mm
RF' g'o loo*/.
DWO
MetObs.Hp
Extrem hohe Bodenozonwerte in Synchrondarstellung
mit anderen met. Messgrbflen 26.2.1971
Aero/0]
1971/3
— ,.Surface"Ozone Air Electricfield Strength
Wmddirection Relative Humidity
j-f Precipitation-Amount Air Temperature unOi/ri^Air
0 200 (.00 600 800 1000
13°°-
.
\
> ('
;. f
S-
5 ( )
s: i. *
^-- \ 1—
"^•^*— - »
_x~ *
r-
"' I:
"-30'KV/m ' '-15KV/ni 6 ' .1 SKV/iri ' ' '.30kv/m
» E S W N °C -5 "6 RR- 1 Step =
0.09mm, total 3.17mm
RH 90 1007.
DWD Analog Registration of ..Surface" Ozone and other Met. Parameters Aero/0]
Met.Obs.Hp 26.2.1971 1971 /2
Figure 26. Measurements of ozone and other meteorological problems at Hohen-
peissenberg (977 MSL), Germany. Measurements of 23 June 1975 were
made under the leading edge of a well developed cumulonimbus.
Measurements on 26 February 1971 and 19 April 1973 were associated
with strong snow showers and a sequence of cold fronts.
Source: Courtesy of Dr. W. Attmannspacher.
113
-------
"HI
a.
?J
50 100 150 200 250 300
i _j_ 1 1 1 1 1
^^~==^SSS
— ^wwHs»Tsz!sr^7^:"r";; 'Z
'J^^-'— '-—'---•-*-"*"•"• {TRy**"^-- J— '
1
1
L ' ' -6'.6 ' ' ' -3'3 ' ' ' t ' ' ' .13 '
-6 ' -1 ' -2 ' U ' .2
OWO Analog-Registrierung des bodennahen
MetObsHp und anderer meteorologischer Grbflen
Bodenoion Turm I, 6 Stock
T°C
350 1,00 1.50 500 [nb]
-J 1
1
i
\
i
i
\
=*=— | — ,
"]
.6.6 -mo [KVAn]
.1. .6 Skalentle.
-3 -2 -1 0 .1 .
Ozons
19. k. 73
2 [T'C]
Aero/03
1973/1
Y-*omp X-Komp
00
H OOH
70'/. '-lOV. 100 V.
Ost
Sud.
60
80
DWD
Met.Obs.HP
Analog-Reg, meteorologischer Grtiflen
Hohenpeiflenberg. 23.6.1975
Aero/03
1975/7
Figure 26. (Continued)
114
-------
height, whereas Zugspitze Mountain is not, hence the much lower peak ozone
concentrations there. Local mountain and valley wind circulation systems are
expected to produce some air exchange along the mountain slopes, even up to
the 3000-m level. This may be the reason for the appearance of relatively
higher 0 concentrations at Zugspitze in summer and not in spring revealing a
similar seasonal trend as on Whiteface Mountain. With such local circulation
systems even mountain observatories located as high as the Zugspitze would
still be exposed to some tropospherically generated ozone. More research on
local circulation systems and on different chemical trace substances is needed
to settle this question.
CONCLUSION
In the light of the foregoing discussion, we have to conclude that
stratospheric O most of the time only plays a minor direct contribution to
high ozone concentrations observed in rural regions. The natural background
of stratospheric 0 can be assumed with approximately 15 ppm. On occasion
(approximate local probability 0.2 percent of all days), stratospheric O can
exceed the Federal 1-hour standard.
Most of the observed 0 in rural areas appears to come from the upper
layers of that region of the lower troposphere that is mixed by thermal
convection during daytime hours. Anticyclonic systems are most conducive to
carrying high amounts of such ozone of tropospheric (natural or anthropogenic)
origin. It will be difficult to separate this tropospherically generated 0
from the stratospheric contribution, unless other stratospheric and tropo-
spheric trace constituents are examined at the same time.
SUGGESTIONS FOR FUTURE RESEARCH
The numerical estimates of the effects of stratospheric ozone on ground-
level ozone concentrations made in the preceding chapters certainly are not
without errors. Unfortunately, we are lacking an accurate data base that
would allow a more precise formulation of these estimates at this time. The
following improvements in our knowledge could be perceived.
115
-------
(a) From a number of detailed case studies involving isentropic tra-
jectory analysis of stratospheric air intrusions, one should be able
to derive a correlation between the intensity of cyclogenetic events
in the upper troposphere and the mass of air involved in stratosphere-
troposphere exchange processes. These case studies would be rather
cumbersome and time-consuming. They are needed, however, to provide
a calibration for the statistical assessments of the frequency with
which cyclogenetic events of different intensity occur in various
sectors of a hemisphere. Airborne measurements, such as those obtain-
ed from the NASA Global Atmospheric Sampling Program (GASP) (2)
should be used for an assessment of ozone concentrations near the
tropopause gap during the incipient stages of stratospheric air
intrusions.
(b) The amount of mixing, as a function of time, between air intruding
from the stratosphere and surrounding tropospheric air should be
established more closely than could be done in the preceding sections.
Isentropic analysis of the path of intruding air masses and their
potential vorticity budget would offer one avenue of approach.
Another, which preferably should proceed simultaneously, would be to
probe these stratospheric air intrusions and their tropospheric
surroundings by airborne sampling devices for the detailed distribu-
tion of trace constitutents of stratospheric or tropospheric origin.
7 32 =
Be , P , SO , and SO are constituents which should be prime targets
for such investigations. Hydrocarbons and NO might also prove
X
helpful in "tagging" certain air masses.
(c) The fate of pollutants in the upper reaches of the layer capped by
the maximum mixing height achieved during daytime still is poorly
understood. What is their residence time in the atmosphere, allowing
for varying synoptic weather conditions and for possible photochemi-
cal reactions? Are there effective diffusion mechanisms that spread
these pollutants even beyond this maximum mixing height? Detailed
case studies can be advocated that involve careful synoptic analyses,
116
-------
and the measurement of vertical profiles of O , (see, for example,
O
reference 25), SCL, NO , and hydrocarbons over several locations,
^- X
distributed over a reasonably large region and over a period of time
commensurate to the life history of a distinct air mass.
(d) Even if a stable layer of stratospheric origin impinges upon a moun-
tain range — such as the Alps or the Adirondacks — one will have to
exercise great care in interpreting observed trace-constitutent
concentrations. The fact that such a stable layer is intercepted by
a mountain usually is established from radiosonde ascents, launched
at considerable distance from this mountain, and from the temperature
record at the mountain observatory. This evidence, however, is
remiss in giving us information about the mixing between that stable
layer of stratospheric origin and the underlying tropospheric air.
Such mixing could be enhanced by local mountain-valley circulation
systems as mentioned earlier, and by turbulence generated by the
terrain. If an up-slope wind component is observed underneath the
stable layer of stratospheric origin, one would have to assume that
some of the air of tropospheric origin is forced across the mountain
range together with the air in the stable layer, and a certain amount
of mixing would be expected between the two air masses. If the
tropospheric air were laden with O , above-normal 0 concentrations
at the mountain observatory could then be misinterpreted as having
come from the stratosphere.
Again, judicious sampling of other trace contaminants, especially of
those characteristic for tropospheric sources (e.g., SO , SO , and
hydrocarbons) would help in establishing the origin, stratospheric or
tropospheric, of the observed ozone.
(e) To arrive at a reasonable control strategy of anthropogenic oxidants,
one will have to test forecasting schemes which predict the amount of
stratospheric 0 involved in pollution episodes over different regions
and at different elevations. The development of such prediction
schemes should not stray too far from data sources routinely avail-
117
-------
able from the National Weather Service. One could envision a scheme
that uses the correlation between the intensity of cyclogenesis and
the amount of stratospheric air injected into the troposphere (see
(a) above), the amount of mixing to be expected during the transition
of this intruding air mass through the troposphere and the likely
course such an intrusion will take ((b) above), and the prognostic
charts provided by the U.S. Weather Bureau that give upper-flow
configurations and surface frontal systems. One should allow for
testing periods during which the predicted progress of stratospheric
air intrusions is monitored by airborne and ground-based data sampling.
COMMENTS BY VOLKER MOHNEN
Introduction
Reiter presents a comprehensive assessment of the stratospheric ozone
issue. Considering the limited data base available at this time one is more
or less forced to rely on circumstantial evidence, estimates, data interpreta-
tion, etc. It is, therefore, not surprising to find somewhat differing
conclusions drawn by different authors even if the same data basis is used.
There are several topics in Reiter's paper that deserve further, more elaborate
discussion.
Stratospheric — Tropospheric Transport Processes
Reiter1s separation of annual mass flux from stratosphere into
• seasonal adjustment of the tropopause height,
• mean meridional circulation,
• stratospheric exchange between hemispheres,
• large-scale eddies,
• small-scale eddies,
has the inherent danger of "double counting" fluxes. The other alternative
118
-------
would be to treat the exchange process on the basis of only two contributors,
namely:
• mean meridional circulation
• eddies (large- and small-scale)
The manner in which the mean and eddy horizontal and vertical ozone
fluxes contribute to the total transport has been discussed by London and Park
(82,83). London and Park employed the 12-layer General Circulation Model
developed by NCAR. The photochemical calculations for ozone have been carried
out as time-dependent in a three-dimensional, 0-H-N system. Their calcula-
tions confirm the dominant importance of large-scale eddy motions in trans-
porting ozone poleward and downward in mid-latitudes. Hadley circulation in
the Northern hemisphere is clearly shown in the ozone intrusion through the
subtropical tropopause region. However, comparison with observations indicate
that the downward branch of the Hadley circulation in the Northern hemisphere
as computed by the General Circulation Model is probably too large. This is a
result of too strong a return (downward) flow computed by the GCM. The com-
puted ozone flux* is a maximum of 7 x 10 molecules cm -s at latitude 60°N
decreasing to 3.5 x 10 cm *s at the equator. This calculated flux (and
its latitudinal variation) compares very favorable with that reported for the
Northern hemisphere by
Danielsen (58)
(mainly due to large scale
eddies)
Nastrom (15)
(derived from GASP data,
mainly 30° - 60°N)
Junge (20)
(derived mainly from
surface observations)
7 x 10 molecules cm «s
7.8 x 10 molecules cm *s
7.5 x 10 molecules cm *s
* Flux determined at the lower boundary, i.e., the earth's surface.
119
-------
Fabian and Pruchniewicz
(derived from surface
observation, hemispheric
average)
(16)
r r ,,,10 , -2-1
6.6 x 10 molecules cm *s
Reiter estimates this annual flux to be of the order of 1.44 x 10
-2 -1 7
molecules cm *s (9.215 x 10 tons ozone/year for the Northern hemisphere
(accomplished by Hadley cell circulation only)). Reiter states, "We will have
to take into account that the major downward transport during most of the
year, accomplished by the mean meridional circulation, occurs in middle lati-
tudes and not over the entire hemisphere. An annual flux of ozone of 9.215 x
7 10 —2 —1
10 tons translates into a flux of 2.88 x 10 molecules cm *s (or 0.23 x
-7 -2 -1
10 g*m «s ) if concentrated over half the area of one hemisphere. This
number does not yet include the effects of eddy transport processes in the
jet-stream region, which also have a preponderance in middle latitudes and on
the average, might increase the above estimate by 50%." We can conclude,
therefore that Reiter's best estimate yields:
Reiter (34)
(estimated from stratospheric-
tropospheric mass exchange, Hadley
cell circulation, and large-scale
eddies)
4.32 x 10 molecules cm *s
Reiter attributes the discrepancy between his ozone flux estimate and the
surface ozone fluxes published by others to "anthropogenic contamination ot
some of the data." However, an equally valid reason for the different flux
estimates is the fact that various authors used different ozone mass mixing
ratios above the tropopause:
Reiter (34)
Danielsen (58)
r = 0.5 x 10 g-g
-6 -1
r = 1.3 x 10 g*g
Reiter states, "0.5 x 10 g'g sets a convenient upper limit of average
ozone concentration a short distance above the tropopause." One can argue
that during tropopause folding events (large-scale eddies), stratospheric air
is intruded from higher levels above the tropopause. Therefore, the mass
120
-------
-6 -1
mixing ratio applicable to these events (latitude >30°N) is <1.3 x 10 g'g
used by Danielsen. For exchange processes at lower latitudes, the intruded
air originates from lower levels above the troposphere, hence the 0.5 x 10
g-g~ mass mixing ratio is more applicable. In any event, the ozone mass
mixing ratio enters critically in any calculation of ozone fluxes. It is
mainly for this reason that Danielsen's and Reiter's calculated ozone fluxes
differ. It must be reitinerated here, that Reiter's flux estimate is one of
the lowest reported in literature.
Mean Vertical Velocities Prevailing in the Troposphere
Reiter calculated a mean vertical velocity of w = 2.74 x 10 m*s
December-February data, 10°N to 40°N, at a height of 8 km). At higher lati-
tudes, this value might be different. Other mean vertical velocity calcula-
tions range from (1-7) x 10 m-s (Lance Bosart, personal communication,
1977) with strong seasonal variation. Average annual values for w of 5 x 10
-1 -3 -1
iri'S (15) or 4.7 x 10 m«s (Danielsen, personal communication, 1977) for
mid-latitudes have been used. Following, nevertheless, Reiter's circumstan-
tial evidence, one would arrive at a mean background concentration of ozone
for the 10°N-40°N latitude belt of 14.6 yg«m~ . In a strict sense, this
derived mean concentration would be representative only at the 8-km level.
Measured annual mean concentrations for this latitude belt and altitude region
_o
have been recently shown to be of the order of >40 ug*m as derived from GASP
data (Falconer, personal communication, 1977). This discrepancy cannot be
resolved at this time.
Reiter used Junge's (20) estimate for the mean tropospheric residence
time of ozone (3.3 months) to arrive at a mean transit velocity of 1.4 x 10~3
m»s which then yields, together with Junge's flux value of 0.5 x 10~7 g-m~2
•s , a mean tropospheric ozone concentration of 35.7 yg«m~ . This is con-
trary to Junge's (1962) concept of flux derivation. Junge (20) assumed a
priori (from carefully selected tropospheric background measurement) an annual
mean concentration of 50 yg-m , which fixes the annual average for the total
Q
tropospheric ozone content to 1.3 x 10 tons (based on an average tropopause
height of 12 km, and an ozone mixing ratio increasing linearly by a factor of
121
-------
two as height increases from 1 km to 12 km). With this basic assumption,
Junge eventually arrives at a flux and a mean tropospheric residence time.
The concept of estimating ozone background concentration for the entire tropo-
sphere from mean vertical velocities is, therefore, at least, debatable.
Zugspitze Mountain Measurements
Reiter states "The mode value of observations appears to lie close to
one-half of the Federal 1-hour standard value. Allowing for additional
dilution between 3000 m and the general terrain level of the eastern United
States, the estimates of 0.05 x 10 g*g of 24-hour averages maximum ozone
concentrations made above appear to be reasonable. Aircraft measurements
below the tropopause near 10 to 12 km reported by Pruchniewicz et al. (76)
yield similar mean concentrations." This circumstantial evidence is accept-
able for ground-level concentrations (measurements in the Adirondacks, White-
face Mtn., N.Y. , and Olympic Mountains, Washington; see Coffey et al. (24),
Chatfield et al. (23)), but debatable for upper tropospheric concentrations:
Pruchniewicz et al. aircraft data are only reported for 25°N - 0° and 0° -
25°S. It is well known that the tropospheric ozone concentration is lower in
low latitudes, i.e., from 30°N to 30°S. GASP measurements at mid-latitudes
show higher values (15).
The Troposphere as a Source or Sink for Ozone
Reiter raises at several occasions the issue of "ozone of tropospheric
(natural or anthropogenic) origin." The existence of photochemical ozone
episodes around major metropolitan complexes is well established. The geogra-
phic extent of these regions of high ozone concentrations is not yet esta-
blished. It is therefore conceivable that rural or remote locations can
suffer from elevated ozone concentration as a result of "long-range transport
of ozone and/or ozone precursor gases." Furthermore, the question of tropo-
spheric ozone production and destruction from natural precursor gases has not
been resolved yet. Two opposite views are presented by, for example, Chameides
and Stedman (84), and Fishman and Crutzen (85). Chameides and Stedman find:
"It is interesting that as we now understand the tropospheric ozone budget,
122
-------
the major role of photochemistry in the ambient atmosphere, where nitrogen
oxide densities are low, is to act as a net sink for ozone. Thus the calcu-
lated ozone abundance with photochemistry and transport is lower than the
abundance calculated with transport alone." On the other hand, Fishman and
Crutzen's numerical investigations show that "it becomes difficult to explain
the observed tropospheric ozone profiles. We suspect, therefore, that cata-
lytic ozone-producing mechanisms are operative in the troposphere in addition
to those we have considered in this study." It is obvious that better infor-
mation of the tropospheric NO-NO concentration and on critical rate constants
(e.g., for NO + HO ) is required before this controversy can be resolved.
Therefore, the origin of ozone observed at rural-remote stations still
remains open to debate. Reiter's circumstantial evidence for a low-tropo-
spheric "ozone climatology" can, therefore, not be disputed on the basis of
existing data.
123
-------
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):299-307, 1977.
2. Holderman, J.D., and E.A. Lezberg. NASA Global Atmospheric Sampling Program
(GASP). Data Report for Tape VL0001. NASA Technical Memorandum, NASA
TM x-71905, NASA, Lewis Research Center, Cleveland, Ohio, 1976.
3. 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.
4. Dutsch, H.U. The Ozone Distribution in the Atmosphere. Can. J. Chem.,
52(8):1491-1504, 1974.
5. Bering, W.S. Ozone and Atmospheric Transport Processes. Tellus XVIII,
(2):329-336, 1966.
6. Bering, W.S. Ozone Measurements for Diagnostic Studies of Atmospheric
Circulation. American Institute of Aeronautics and Astronautics. Paper
No. 65-462, 1965. 17 pp.
7. Reiter, E.R. Stratospheric-Tropospheric Exchange Processes. Rev. of
Geophys. and Space Physics, 13(4):459-474, 1975.
8. Reed, R.J., and E.F. Danielsen. Fronts in the Vicinity of the Tropopause.
Arch. Meteor. Geophys. Biokl., Series A, Bll:l-17, 1959.
125
-------
9. Danielsen, E.F. A Determination of the Mass Transported from Stratosphere
to Troposphere over North America during a Thirty-Six Hour Interval. Mittei-
lungen des Deutschen Wetterdienstes, 20(3):10-11, 1959.
10. Reiter, E.R., and J.D. Mahlman. Atmospheric Transport Processes Leading
to Radioactive Fallout over the United States in November 1962. In: Radio
active Fallout From Nuclear Weapons Test, Proceedings of 2nd Conference,
AEC Symposium Series 5, 1965. pp. 450-463.
11. Mahlman, J.D. Relation of Tropopause-Level Index Changes to Radioactive
Fallout Fluctuations.1 Atmospheric Science Technical Paper No. 70. Colorado
State Univeristy, Fort Collins, Colo., 1963. pp. 84-109.
12. Danielsen, E.F. Report on Project Springfield Report. Defense Atomic
Support Agency Report. DASA1517, (DDC), (Contract DA-49-146-X2-079).
Washington, D.C., 1964. 97 pp.
13. Danielsen, E., and V. Mohnen. Project Duststorm Report: Ozone Measurements
and Meteorological Analyses of Tropopause Folding. Abstract In: Proceedings
of the Joint IAOC/ICACGP Symposium on Atmospheric Ozone and Related Topics,
Dresden, German Democratic Republic, Aug. 9-17, 1976.
14. Bering, W.S., and T.R. Borden, Jr. Mean Distribution of Ozone Density over
North America, 1963-1964. AFCRL 65-913. Environmental Research Papers,
No. 162. USAF, L.G. Hanscom Field, Bedford, Mass., 1965. 28 pp.
15. Nastrom, G.D. Variability and Transport of Ozone at the Tropopause from
the First Year of GASP Data. Research Report No. 4. Contract NAS-2-7807,
NASA-Lewis Research Center, Cleveland, Ohio, 1977. 28 pp.
16. Fabian, P., and P.G. Pruchniewicz. Tropospharisches Ozon. Final Report
on Contract No. Fa 62/1. Deutsche Forschungogemeinschaft. MPAE-W-100-
76-21, 1976. 28 pp.
126
-------
17. Cunnold, D., F. Alyea, N. Philips, and R. Prinn. A Three-Dimensional
Dynamical-Chemical Model of Atmospheric Ozone. J. Atmos. Sci.,
32(1):170-194, 1975.
•
18. Mintz, Y., and M. Schlesinger. Ozone Production and Transport with the
UCLA General-Circulation Model. In: Proceedings of the 4th Conference
on the Climatic Impact Assessment Program, T. Hard and A.J. Broderick,
eds. DOT-TSC-OST-75-38, 1975. pp. 201-222.
19. Normand, C. Atmospheric Ozone and the Upper-Air Conditions. Quart. J.
Roy. Met. Soc., 79:39-50, 1953.
20. Junge, C.E. Global Ozone Budget and Exchange between Stratosphere and
Troposphere. Tellus XIV, (4):363-377, 1962.
21. Kroening, J.L., and E.P. Ney. Atmospheric Ozone. J. Geophys. Res.,
67(5):1867-1875, 1962.
22. Paetzold, H.K. New Experimental and Theoretical Investigations on the
Atmospheric Ozone Layer. J. Atmosph. Terrestrial Physics, 7(3):128-140,
1955.
23. Chatfield, R., and H. Harrison. Ozone in the Remote Troposphere: Mixing
versus Photochemistry. J. Geophys. Res., 81(3):421-423, 1976.
24. Coffey, P.E., and W.N. Stasiuk. Evidence of Atmospheric Transport of
Ozone into Urban Areas. Environ. Sci. Technol., 9(1):59-62, 1975.
25. 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.
127
-------
26. Mohnen, V.A., A. Hogan, E. Danielsen, and P. Coffey. Ozone Measurement
and Meteorological Analysis of Tropopause Folding. Int. Conf. Ox. Poll.,
Proc. 1:97-108. EPA-600/3-77-001a. Environmental Protection Agency,
Research Triangle Park, N.C., 1977.
27. Holdeman, J.D., and P.M. Humenik. NASA-GASP Data Report for Tape
VL0005. NASA TMX-73608. Lewis Research Center, Cleveland Ohio, 1977.
45 pp.
28. Tiefenau, H.K. The Specific Ozone Destruction Rate of the Ocean Surface and
Its Dependence on Horizontal Wind Velocity. Pure and Applied Geophys.
106-108:1116-1123, 1973.
29. Wilbrandt, P.J. Bestitmnung der spezifischen ozonerstorungsrate uber
buschsteppe und des ozonflusses in diese oberflache mit hilfe von ozon-
und temperaturprofilmessungen and unem 120 m mast in tsumeb/S.W.A.
Mitteilungen, Max-Plank-Institut ur der Aeronomie, No. 54. Springer-
Verlag, Berlin, W. Germany, 1975. 36 pp.
30. Aldaz, L. Flux Measurements of Atmospheric Ozone over Land and Water.
J. Geophys. Res., 74(28):6943-6946, 1969.
31. Crutzen, P.J. A Discussion of the Chemistry of Some Minor Constituents
in the Stratosphere and Troposphere. Report AP-12, Institute of Meteorology,
University of Stockholm, Stockholm, Sweden, 1973.
32. Chameides, W., and J.C.G. Walker. A Fine Dependent Photochemical Model
For Ozone Near the Ground. J. Geophys. Res., 81(3):413-420, 1976.
33. Reiter, R., H.J. Ranter, R. Sladkovic, and K. Potzl. Measurement of
Airborne Radioactivity and Its Meteorological Application. Part V.
Annual Report 1 April 1973 - 31 July 1974. Institut Fur Atmospharische
Umweltforshung. ERDA Document No. NYO-3425-12. 1976.
128
-------
34. Reiter, E.R. The Role of Stratospheric Import on Tropospheric Ozone
Concentrations. Int. Conf. Ox. Poll., Proc. 1:393-410. EPA-600/3-77-
OOla. Environmental Protection Agency, Research Triangle Park, N.C.,
1977.
35. Carney, T.B. Evidence of the Role of Stratospheric Transport in the
Distribution of Tropospheric Ozone. Ozone/Oxidants - Interactions
with the Total Environment. APCA Specialty Conference (Southwest Section),
Proceedings. Air Pollution Control Association, Pittsburgh,
Pa., 1976. pp. 234-241.
36. Bering, W.S., and T.R. Borden, Jr. Ozonesonde Observations over North
America, Vol. 3. Environmental Research Paper No. 133. Report AFCRL-
64-30(111), Air Force Cambridge Research Laboratories, 1965.
37. Reiter, E.R. Atmospheric Transfer Processes, Part II. Chemical Tracers.
U.S. Atomic Energy Commission Report, Atmospheric Science Paper 133.
TID-25314, 1971. 382 pp.
38. Lea, D.A. Vertical Ozone Distribution in the Lower Troposphere near an
Urban Pollution Complex. J. Appl. Meteorol., 7(2):252-267, 1968.
39. Sticksel, P.R. Occurrence and Movement of Tropospheric Ozone Maxima.
Ozone/Oxidants - Interactions with the Total Environment. APCA Specialty
Conference (Southwest Section), Proceedings. Air Pollution
Control Association, Pittsburgh, Pa., 1976. pp. 252-267.
40. Pruchniewicz, P.G. Uber ein Ozon-Registriergerat und Untersuchung der
zeitlichen un raumlic Variationen des Troposphanschen Ozon auf der
Nordhalbkugel der Erde. Mitteilungen, Max-Plank-Institut fur der
Aeronomie, No. 42. Springer-Verlag, Berlin, Germany, 1970. 70 pp.
41. Dutsch, H.U. Two Years of Regular Ozone Soundings over Boulder, Colo.
NCAR Technical Notes #10, Boulder, Colo., 1966. 444 pp.
129
-------
42. Falconer, P.D. The Global Atmospheric Sampling Program: The Prospects
for Establishing a Tropospheric Ozone Budget from Commercial Aircraft
Data. Presented at the ASTM Conference on Air Quality Meteorology and
Atmospheric Ozone, Boulder, Colo., 1977.
43. Reiter, E.R. The Behavior of Jet Streams in Potential Fallout Situations.
Archiv. Meteorol. Geophys. Biokl., Series A, 17(1):8-16, 1968.
44. Crutzen, P.J. Estimates of Possible Variations in Total Ozone due to
Natural Cause and Human Activities. Ambio, 3(6):201-210, 1974.
45. Nier, A.O.C., J.P. Friend, L.H. Hempelmann, J.F. McCormick, D.R. Parker,
E.R. Reiter, A.H. Seymour, and P.E. Waggoner. Long-Term Worldwide Effects
Of Multiple Nuclear-Weapons Detonations. National Academy of Sciences,
Washington, D.C., 1975. 213 pp.
46. Baker - Blocker, A. Stratospheric Ozone Fluctuations and the Evolution
of Skin Pigmentation in Man. Archiv, Meteorol. Geophys. Biokl., Series
B, 25(1):91-100, 1977.
47. Reiter, E.R., and B.C. MacDonald. Quasi-Biennial Variations in the
Wintertime Circulations of High Latitudes. Arch. Meteorol. Geophys.
Bioklim., Series A, 22(1):145-167, 1973.
48. Angell, J.K., and J. Korshover. Global Analysis of Recent Total Ozone
Fluctuations. Monthly Weather Review, 104(1):63-75, 1976.
49. Dutsch, H.U. Photochemistry of Atmospheric Ozone. Advances in Geophysics,
15:219-322, 1971.
50. Reiter, E.R. Stratospheric - Tropospheric Transport Processes and Their
Relation to the Influx of Ozone into the Troposphere. Report to Stanford
Research Institute, Sept. 26, 1975. 12 pp.
130
-------
51. Reiter, E.R. Significance of Stratospheric Ozone for Ground-Level
Ozone Concentrations. Report to Stanford Research Institute, April
7, 1976. 15 pp.
52. Singh, H.B., W.B. Johnson, and E.R. Reiter. The Relation of Oxidant
Levels to Meteorochemical Processes: A Review of Available Research
Results and Monitoring Data. Interim Report. Stanford Research Institute,
Menlo Park, Calif., 1975. 119 pp.
53. Reiter, E.R., and J.D. Mahlman. Heavy Radioactive Fallout over
the Southern United States, November 1962. J. Geophys. Res,,
70(18) .-4501-4520, 1965.
54. Bauer, E., and E.R. Reiter. The Natural Stratosphere of 1974. In:
Report of Findings. The Effects of Stratospheric Pollution by Aircraft,
A.J. Grobecker, S.C. Coronite, and R.H. Cannon, Jr., eds. CIAP Monograph 1,
DOT-TST-75-50, Appendix C, 1974.
55. Hering, W.S. Ozonesonde Observations over North America, Vol. 1.
Research Report AFCRL-64-30(I). Air Force Cambridge Research Labor-
atories, 1964. 528 pp..
56. Bering, W.S., and T.R. Borden, Jr. Ozonesonde Observations over North
America, Vol 2. Environmental Research Paper No. 38. Report AFCRL-64-
30(11). Air Force Cambridge Research Laboratories, 1964.
57. Bering, W.S., and T.R. Borden, Jr. Ozonesonde Observations over North
America, Vol. 4. Environmental Research Paper No. 279. Report
AFCRL-64-30(IV). Air Force Cambridge Research Laboratories, 1967.
58. Danielsen, E.F., R. Bleck, J. Shedlovsky, A. Wartburg, P. Haagenson,
and W. Pollock. Observed Distribution of Radioactivity, Ozone and
Potential Vorticity Associated with Tropopause Folding. J. Geophys.
Res., 75(12):2353-2361, 1970.
131
-------
59. Pruchniewicz, P.G. A Study of the Tropospheric Ozone Budget Based
on Interhemispheric Mass Exchange. In: Proceedings of the International
Conference on Structure, Composition and General Circulation of the
Upper and Lower Atmosphere and Possible Anthropogenic Perturbations,
Vol. I. Melbourne, Australia, 1974. pp. 429-438.
60. Reiter, E.R. Ozone Concentrations in the Lower Troposphere as Revealed
by Ozonesonde Observations. Report to Stanford Research Institute,
March 24, 1976. 106 pp.
61. Reiter, E.R. Atmospheric Transport Processes. Part I: Energy
Transfers and Transformations. NTIS-TID-24868. U.S. Atomic Energy
Commission, Washington, D.C., 1968. 253 pp.
62. Oort, A.H., and E.M. Rasmussen. Atmospheric Circulation Statistics.
NOAA Professional Paper 5, 1971. 323 pp.
63. Oort, A.H. Year-to-Year Variations in the Energy Balance of the Artie
Atmosphere. In: Climate of the Artie. Proceedings of Alaska Science
Conference 24th, August 15-17, 1973, G. Weller and S.A. Bowling, eds.
Geophysical Institute, University of Alaska, Fairbanks, Alaska, 1975.
pp. 68-75.
64. Reiter, E.R. Atmospheric Variability and Climatic Modeling. In: Climate
of the Arctic. Proceedings of Alaska Science Conference 24th, August 15-
17, 1973, G. Weller and S.H. Bowling, eds. Geophysical Institute, University of
Alaska, Fairbanks, Alaska, 1975. pp. 125-131.
65. McQuirk, J.P., and E.R. Reiter. A Vacillation in Atmospheric Energy
Parameters. J. Atmos. Sci., 33(11):2079-2093, 1976.
66. Vincent, D.G. Mean Meridional Circulation in the Northern Hemisphere
Lower Stratosphere During 1964 and 1965. Quart. J. Royal Meteorol.
Soc., 94(401):333-349, 1968.
132
-------
67. Danielsen, E.F. Trajectories: Isobaric, Isentropic and Actual. J.
Meteorol., 18(4):479-486, 1961.
68. Reiter, E.R. A Case Study of Radioactive Fallout. J. Appl. Meteorol.,
2(6):691-705, 1963.
69. Reiter, E.R. Comments on Paper by S. Penn and E.A. Martell: An
Analysis of Radioactive Fallout over North America in Late September
1961. J. Geophys. Res., 69(4):786-788, 1964.
70. Reiter, E.R. and J.D. Mahlman. A Case Study of Mass Transport from
Stratosphere to Troposphere not Associated with Surface Fallout.
Atmospheric Science Technical Paper No. 70, Colorado State
University, Fort Collins, Colo., 1965.
71. Reiter, E.R. Atmospheric Transport Processes. Part 3. Hvdrodvnamic
Tracers. U.S. Atomic Energy Commission Report. TIY25731, 1972. 212pp.
72. McQuirk, J.P. Fluctuations in the Atmosphere's Energy Cycle. En-
vironmental Research Paper No. 6, Colorado State University, Fort
Collins, Colo., 1977.
73. Newell, R.E., H.W. Brandli, and D.A. Widen. Concentration of Ozone
in Surface Air over Greater Boston in 1965. J. Appl. Meteorol., 5(5):
740-741, 1966.
74. Seitz, H., B. Davidson, J.P. Friend, and H.W. Feely. Numerical Models
of Transport, Diffusion and Fallout of Stratospheric Material. Report
No. NYO-3654-4, Isotopes, inc., Westwood, New Jersey, 1968. 103 pp.
75. Reiter, E.R. The Transport of Radioactive Debris and Ozone from the
Stratosphere to the Ground. Report to Stanford Research Institute,
Nov. 22, 1975. 36 pp.
133
-------
76. Pruchniewicz, P.G., H. Tiefenau/ P. Fabian, P. Wilbrandt, and W. Jessen.
The Distribution of Tropospheric Ozone from Worldwide Surface and Air-
craft Observations. In: Proceedings of the International Conference
on Structure, Composition and General Circulation of the Upper and
Lower Atmospheres and Possible Anthropogenic Perturbations, Vol. 1.
Melbourne, Australia, 1974. pp. 439-451.
77. Sticksel, P.R. Application of 1960's Ozone Sounding Information to
1970's Surface Ozone Studies. Int. Conf. Ox. Poll., Proc. 1:381-391.
EPA-600/3-77-001a. Environmental Protection Agency, Research Triangle
Park, N.C., 1973.
78. Coffey, P.E., W.N. Stasiuk, V.A. Mohnen. Ozone in Rural and Urban
Areas of New York State. Int. Conf. Ox. Poll.., Proc. 1:89-96.
EPA-600/3-77-001a. Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
79. Holzworth, G.C. Estimates of Mean Maximum Mixing Depth in the Contiguous
United States. Monthly Weather Review, 92(5):235-242, 1964.
80. Reiter, E.R., and T. Henmi. Residence Time of Atmospheric Pollutants
and Long-Range Transport. Annual Report to EPA on Contract R303685-01-0,
1976.
81. Bach, W.D., Jr. Analysis and Interpretation of Serial Ozonesonde Releases.
In: Ozone/Oxidants - Interactions with the Total Environment. Proceedings
APCA Specialty Conference (Southwest Section), Proceedings. Air Pollution
Control Association, Pittsburgh, Pa., 1976. pp. 96-108.
82. London, J., and J. Park. Application of General Circulation Models to
the Study of Stratospheric Ozone. Pure and Applied Geophysics, 106-108:
1611-1617, 1973.
134
-------
83. London, J., and J. Park. The Interaction of Ozone Photochemistry and
Dynamics in the Stratosphere. A Three-Dimensional Atmospheric Model.
Can. J. Chem., 52(812):1599-1609, 1974.
84. Chameides, W.L., and D.H. Stedman. Tropospheric Ozone: Coupling
Transport and Photochemistry. J. Geophys. Res., 82(12):1787-1794, 1977.
85. Fishman, J., and P.J. Crutzen. A Numerical Investigation of Tropospheric
Photochemistry Using a One-Dimensional Model. Symposium on Non-Urban
Tropospheric Composition, Miami, Fla., Nov. 10-12, 1976.
135
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-77-115
2.
4. TITLE ANDSUBTITLE
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976 -
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part III. The Issue of Stratospheric Ozone Intrusion
5. REPORT DATE
December 1977
6. PERFORMING ORGANIZATION CODE
3. RECIPIENT'S ACCESSION-NO.
7. AUTHOR(S)
1. V.A. Mohnen
2. E.R. Reiter
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
1. State Univ. of N.Y., Albany, NY
2. Colorado St. Univ., Fort Collins, CO
10. PROGRAM ELEMENT NO.
1AA603 AJ-13 (FY-76)
11. CONTRACT/GRANT NO.
1. DA-7-1936A
2. DA-7-1305J
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTF, 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 not work for the EPA or for industry) of widely recognized
competence and experience in the area of photochemical pollution occurrence
and control.
In Part III V.A. Mohnen and E.R. Reiter discuss the issue of stratospheric
ozone intrusion, i.e., whether ozone of stratospheric origin contributes signifi-
cantly to ground-level ozone buildup. The literature on the subject of ozone
intrusion is discussed and suggestions for further research to resolve some of
the questions raised are made.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
* Air pollution
* Ozone
* Stratosphere
13 DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b.lDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATI Field/Group
13B
07B
04A
21. NO, OF PAGES
148
22. PRICE
EPA Form 2220-1 (9-73)
136
-------
«>'
fn
-•. Cl>
3 ^
c o
08-*
-
to e->
O- )
'/j-i "^
^GE,^
AN EQUAL
O
iPPORTUNI
H
<
m
2
TJ
r
o
-<
rn
;g
PENALTY
T|
0
3
TJ
3
<
>
H
m
C
/A
u)
m
^
u
o
0
0
Tl
•n
O
>
r
CD
C
U)
z
m
0)
0)
o
3
o
5'
D
d)
O
3-
o
*.
en
rv-\
<\J
O)
CXl
i
D
o
CO
— h
O
—1
3
CO
(-+•
6
3
C/)
CO
— h
—i^
q
o'
(D
O
— *>
33
esearch an<
l_L
D
CD
<
CD
6"
7~1
**•*
(D
^3
S. ENVIRON
^.
m
z
H
>
f
-o
33
o
H
m
o
H
0
z
>
0
m
z
o
CD
0
O
5T
(fl
T3
CD
O
T1
O
C
a
0)
!•+
(D
01
c
0)
m
z
o o
z
2
m
Z
H
Cfl
H
>
O
m
> >
r r z
0) T) O
^ T
0 m
H m
m en
§ J
i 5
>
G>
m
Z
n
------- |