|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 66
8
Theoretical Projections of
Stratospheric Change Due to
Increasing Greenhouse Gases and
Changing Ozone Concentrations
JERRY D. MAHLMAN
Geophysical Fluid Dynamics Laboratory
National Oceanic and Atmospheric Administration
This talk discusses what has happened in the stratosphere and
what may happen there in the future. ~ will first review the ensemble
of gases present in the stratosphere and their effects:
1. ChIorofluorocarbon (CFC-ll and CFC-12) increase. The
increase of CFCs already appears to be causing ozone loss through
the action of chlorine. CFCs also act as greenhouse gases in the
troposphere.
2. Methane (CH4) increase. We have heard in a previous pre-
sentation that methane is a tropospheric greenhouse gas. The ap-
proximately 1 percent per year increase in methane also has implica-
tions for long-term increases of the water vapor amount in the middle
and upper stratosphere. However, the methane chemistry opposes
the chlorine catalysis chemistry in the lower stratosphere. Methane
increases also play a role in reducing the amount of hydroxy! (OH)
in the troposphere.
3. Nitrous oxide (N2O) increase. Nitrous oxide is increasing
by what appears to be the comparatively modest amount of about
0.2 percent per year. However, the N2O anthropogenic source is
about one-thircI of the natural source and thus is not negligible on
long time scales. Nitrous oxide is a tropospheric greenhouse gas, but
its small annual increases are contributing little to current increaser
in infrared radiative forcing. Reaction of N2O with excited atomic
66
OCR for page 67
THEORETICAL PROJECTIONS
67
oxygen in the rniddIe stratosphere produces reactive nitrogen (NO2),
which provides the major natural catalytic loss of ozone. Oddly,
NOX provides an important negative feedback against the growing
attack on ozone by reactive chlorine (CITY. This occurs through the
formation of chlorine nitrate (ClONO2), thus inhibiting both C1x and
NOX catalytic ozone destruction cycles.
4. Stratospheric carbon dioxide (CO2) increase. The increases
in CO2 will lead to a strong cooling trend in the stratosphere. To
some extent, this cooling effect acts to moderate the expected mid-
stratospheric ozone decreases.
5. Stratospheric ozone decrease. Large ozone decreases will
also result in large stratospheric cooling. There is also strong column
"self-healing" of ozone, which ~ will discuss later.
6. Stratospheric water vapor increase. An increase would re-
sult in increased downward infrared radiative flux, complex chemical
changes if stratospheric ice clouds form, and an increased ozone loss
in the 30- to 50-km layer.
I will next discuss what the NASA-WMO Ozone Trends Pane!
(Watson et al., 1988) has learned about recent (1979 to 1985) ozone
trends in the stratosphere. SAGE satellite data suggest a decrease
of about 3 percent near 40 km altitude, on an average worldwide,
over the 6-year period. The ground-based Umkehr data, on the other
hand, suggest a decrease of about 9 percent at 40 km. Mode! calcula-
tions predict a 4 to 9 percent decrease in response to increased trace
gases, primarily CFMs, and a 1 to 3 percent decrease in response to
declining solar activity, for a total decrease of 5 to 12 percent. Given
this range of uncertainty, the observed ozone changes near 40 km
are not inconsistent with theory. The credibility of the theoretical
results is enhanced by the observation that stratospheric tempera-
tures have decreased globally, between 25 and 55 km altitude, by
1.7°C since 1979. This decrease is consistent with decreases in up-
per stratospheric ozone of up to (but not larger than) 10 percent.
The vertical profile of ozone change also is in fair agreement with
theoretical predictions.
~ will now turn to a discussion of future trends. Atmospheric
Ozone 1985 (WMO-NASA, 1986) contains estimates for equilibrium
(infinite elapsed time) changes using one-dimensional chemical mod-
els and assuming that CFM emissions are held constant at the 1980
rate. At 10 ppb reactive chlorine species, equilibrium total column
loss is estimated at between 5 and 9 percent. At 15 ppb chlorine
OCR for page 68
l
68
JERRY D. MAHLA~4N
species, the estimated loss ~ 10 to 20 percent. Thus the equilib-
rium column loss doubles, for a 50 percent increase in odd chlorine.
This nonlinearity results from the progressively greater scavenging
of nitrogen oxides by chlorine, leaving the excess chlorine to destroy
increasingly more ozone. On the other hand, if carbon dioxide is dou-
bled (with no other atmospheric changes), equilibrium total column
ozone will increase by 2 to 3 percent.
However, the changes at the 4~km level predicted by the same
mode! are much more drastic. At a concentration of 10 ppb chlorine
species, the equilibrium prediction is an ozone decrease of 60 to 80
percent, and at a concentration of 15 ppb chlorine species, a loss of
70 to 85 percent would result. For the gIobal-mean vertical profile
of ozone mixing ratio, the equilibrium prediction is for an increase
in ozone in the lower stratosphere, between 10 and 25 km, of a few
percent, and a decrease at higher altitudes, with the maximum de-
crease at about 40 km. (The feedback of higher temperatures in the
lower stratosphere and lower temperatures in the higher stratosphere
would reduce slightly the magnitudes of both the increase and de-
crease.) The predicted increase in ozone below about 25 km would
result from more ultraviolet (UV) radiation penetrating to lower alti-
tudes and creating more ozone in a kind of negative feedback process
that tends to limit the depletion of total column ozone. Thus the
change in the total column ozone is a comparatively small difference
between two large numbers, given that the atmospheric mass drops
off nearly exponentially with increased altitude.
Figure 8-1 shows the two-dimensional mode} prediction of
percent ozone decrease by Atmospheric and Environmental Re-
search, Inc. (AER) as reported in WMO-NASA (1986) at 8.2 ppb
C1x species. This is about equal to the equilibrium CI2 for 1980
emission rates. The resulting stratospheric distribution shows maxi-
mum ozone losses at about 40 to 45 km poleward of about 50°N and
50°S latitudes. In these regions, the predicted ozone loss is greater
than 50 percent. On the other hand, ozone is predicted to increase
by about 20 percent in the lower stratosphere (15 to 20 km) near
the equator. In low latitudes, the total column self-healing effect
is comparatively strong because the incident solar UV ~ strong all
year, whereas the effect is much weaker at the high latitudes because
of weaker UV and the presence of large downward mean advection
of ozone at these latitudes.
A question that is fair to ask is, what is the credibility of such
tw - dimensional models? This depends, of course, on which effects
OCR for page 69
THEORETICAL PROJECTIONS
OI I 1 ~r 1 ~,- -I ,
· r 1 ~, , [---_
APRIL _ - 60
_
l
-
E
-
10
100
1 n
-50
-40
In '`
1 \ ~
/ 1 ~
10 J o ~
1 ~1
1 , 1 1 . 1 1 1 1
69
-
By
LO
40
30
A:
X
- 20 cr
CL
- 1 0
-
1 000 _
- 90- 60 - 30 ED 30 60 90
S
LATITUDE
N
FIGURE 8-1 Two-dimensional model prediction of the percent change in ozone
for 8.2 ppbv of total reactive chlorine (Clay. Model results from the chemical-
transport model of Atmospheric and Environmental Research, Inc. (Reprinted
from WMO, 1986.)
are included and which are left out. The models predict large changes
in ozone concentrations; therefore, one would expect significant tem-
perature and circulation changes to occur as well. It should be
pointed out that tw>dimensional models used for ozone assessment
all make a similar assumption about change in circulation. They
implicitly assume that the stratospheric circulation does not change
as absorbers are changed. This corresponds to the "fixed dynamical
heating" limit; that is, the net heating of local air by dynamical
processes does not change with changing absorbers, even though the
temperature itself is free to change. However, when a real climate
system such as the stratosphere is perturbed, the changed distri-
bution of absorbers may or may not lead to a changed temperature
OCR for page 70
70
JERRY D. MAILMAN
distribution in order to maintain what we call "climate balance." The
sources of temperature change are not just radiative but may also
be dynamical. Dynamical heating may result from advection, diffu-
sion, and adiabatic compression or expansion. The radiative heating
is due to short-wave solar radiation (almost independent of atmo-
spheric temperature) and long-wave radiation (strongly a function
of temperature). The assumption is routinely made that the strato-
sphere is in radiative equilibrium, resulting in a very simple system.
But we know that this is a gross oversimplification, especially in the
higher latitudes, where dynamic effects are very important. Actual
temperatures in the polar latitudes sometimes differ by as much as
50 to 60 K from those given by radiative equilibrium models.
Thus if the distribution of absorbers is changed, the dynam-
ical heating, as well as the radiative heating, is likely to change.
If dyna~rucal heating is important, three-dimensional models that
include dynamical response should be used in preference to the
two-dimensional models that implicitly invoke the fixed dynamical
heating assumption. Such a two-dimensional model uses a latitude-
altitude framework that has mean mass circulation directed upward
in low latitudes, downward in high latitudes, and from equator to pole
at stratospheric altitudes. Superunposed on the mean meridional
motion is a meridional eddy diffusion of particles that is driven by
upward-propagating tropospheric disturbances. Both of these trans-
port mechanisms arise in the troposphere; thus the stratospheric
latitude-altitude motions (and the degree of departure from radia-
tive equilibrium) are driven by the troposphere and would decay to
near zero in the absence of dynamical forcing from the troposphere.
The question now is, how does the stratospheric circulation re-
spond when the distribution of absorbers is changed? Calculations
of this type were done at the Geophysical Fluid Dynamics I,abora-
tory (GFDL) in 1980 by Fels et al. In one numerical experiment,
carbon dioxide was doubled but sea temperatures were held fixed.
In this case, Figure ~2 shows that the three-dimensional dynamic
equilibrium mode} results agree fairly closely, to within 1°C, with
those from a fixed dynamical heating radiative model. The dynamic
mode} predicts cooling of 11°C at 45 km, compared to the radiative
model's prediction of about 10°C. Both models show a cooling of
about 2 to 4°C between 20 and 25 km. Figure 8-3 shows another
experiment, in which an arbitrary uniform reduction of ozone by
50 percent was assumed. (This experiment ignored the column self-
healing effect.) Again, the radiative mode! agrees fairly closely with
OCR for page 71
THEORETICAL PROJECTIONS
71
the dynamic mode! in terms of magnitude of cooling, but some of
the gradient details are radically different, particularly in the tropics.
Cooling of 6 to 8°C is predicted in the high latitudes between 15 and
45 km by both models. At the equatorial tropopause, a 12°C cooling
is predicted. If we more realistically assume that the ozone at the
equatorial tropopause increases as in the AER model, then we may
infer that a warming of about 4°C will tend to occur there.
Recently, we have learned much more about how poor a job the
older models did of simulating the dynamics of the stratosphere. As
the horizontal resolution of models has increased, predicted strato-
spheric temperatures have increased and come into better agreement
with observations. Predicted temperatures using a 1°-latitude reso-
lution dynamic mode} agree closely with temperatures observed at
62°N latitude in December, whereas results for a 9° resolution dy-
namic mode! were fairly close to those obtained with the radiative
mode! but from 15 to 45°C colder than either the observed tempera-
tures or results for the high-resolution dynamic model (see MahIman
and Umscheid, 1987~. Thus, the traditional higher-latitude strato-
spheric cold bias of general circulation models was not the fault of
radiative transfer but rather the fault of oversimplifying the dynam-
ics in the models. At high resolution, the models are considerably
more dynamical in the stratosphere and produce temperatures and
motions that look much more like those in the real stratosphere. It
appears that the tropospheric dynamical processes are strong enough
to push the stratosphere some distance away from its radiative equi-
librium condition.
Thus dynamic modeling is essential for predicting changes cor-
rectly in the stratosphere. The good news is that a mode] with suffi-
ciently high resolution is capable of useful predictions. The bad news
is that such models, if run to equilibrium conditions, require sum
stantial computer resources. Possibly, some of the resolution can be
traded for carefully devised parameterizations that are self-consistent
and appropriately sensitive to climate changes. In a stratosphere
that is dynamically driven, the interannual variability is quite large,
thereby increasing the overall computational problem.
Based on our experience thus far, ~ would like to speculate on
what ~ consider to be the stratosphere climate Issues that we will have
to face beyond what we already know. We think we know that the
upper stratosphere will coo} by 20 to 25°C, perhaps more; this makes
the stratosphere a candidate for inclusion in a full climate system
model. Also, there will presumably be large ozone decreases in the
OCR for page 72
72
not
.02
.03
.05
.10
.20
.30
.50
1.0
20
Dd
LO
5.0
10
20 _
50 _
100
200
300
500
1000 _
JERRY D. MAILMAN
T2Xco2-TcoNTRoL (FDH)
1 1 1 ~ 1 \1 1 1 1 1 1 1 1 1/ 1 1 \ 1 1 1
-8
Am\
6
8
0
11 v'
2 >
3
4
5
6
7
8
9
20
21
22
23
24
25
26
27
28
29
30
31
8 .5 76 5 67.5 58 5 49.5 dC.S 3! 5 22.5 13 5 4 5 -~.5 -~3.5 -22 5 -3' 5 -40 5 -~.5 -SH.S -67.5 -76.5 -8 40
LATITUDE
\\ J
\ ~
_10-
~-
Rae
-
_ 55
_ 50
_ 45 3:
_ 40 I
_ 35 ~
Z
_ ~^ ~
_ 5
_ O
._
70
65
60
45
40
35
30
25
20
15
10
FIGURE 8-2 GFDL "SKYHI" model equilibrium temperature changes due to
a doubling of carbon dioxide as determined from an annual mean model with
prescribed sea-surface temperatures. On the left is a fixed dynamical heating
upper stratosphere that will be partially compensated for in the lower
stratosphere. But, the unpredicted antarctic ozone situation warns
us to be wary of things that may be missing from current model
calculations. Speculation follows on some of the things that may
need to be considered in future models.
The first speculation concerns the possibility of reduced strato-
spheric transport circulation, that is, reduced efficiency of the merid-
ional circulation. The reason is related to Robert Dickinson's pre-
sentation, which follows. The projected greenhouse gas warming in
OCR for page 73
THEORETICAL PROJECTIONS
oft
.02
.03
.os
.10 _
.20
.30
.so _
1.0 _
2.0
10
20
an
'00 _
200 ~_ M
300 _
soo _
1000 1 1 11 1 1 1 1 1 id_
85.5N 76.5 67.5 58.5 49.5 40.5 31.5 22.5 13.5 4.5 -4.5 -13.5 -22.5 -31.5 -40.5 -49.5 -58.5 -67.5 -76.5 -85.5s
LATITUDE
1 1
(
T2XCO2- TCONTROL (GCM)
1 1 1 1 1 1 1 1 1 1 /1 1 1 \1 1
~ ) Cot
-~-8-
7
11 ~
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Off
73
1 1 1 1 1 1 1 1 -
Rn
7S
70
60
55
50
-
45 I
V
UJ
40 ~
35 ~
z
A:
30 ~
25
IS
10
(FDH) calculation; on the right is the full general circulation model (GCM)
result. (Reprinted, by permission, from Fels et al., 1980. Copyright Hi) 1980 by
The American Meteorological Society.)
the troposphere should result in a weaker meridional temperature
gradient, thus weakening tropospheric circulation and decreasing the
flux of wave activity to the stratosphere. The stratospheric transport
circulation would in turn be reduced.
Another speculation is that water vapor will increase in the
lower stratosphere (as well as in the upper stratosphere because of
methane). With the column feedback process resulting in increaser]
ozone in the lower stratosphere, it ~ possible that a significant heating
of the equatorial tropopause may occur. A heating of up to 4°C
OCR for page 74
74
JERRY D. MAILMAN
n1
no
.03
.05
an
.20
.50
1.0
2.0
50L
10
50
100
200
300
soot
1000 L
I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ,_
85 SN 76 5 67 5 58 5 49 5 ".5 31.5 22 5 13.5 4 5 -4 5 -13 5 -22 5 -31.5 -I 5 -49 5 -58 5 -67 5 -76 5 -85 55
LATITUDE
so
75
70
~5
.s5
50 E
45 I
40 ~
a:
35 ~
Z
tan
on
IS
In
FIGURE 8-3 GFDL "SKYHIn model equilibrium temperature changes due to
a uniform 50 percent ozone reduction. On the left is a fixed dynamical heating
(FDH) calculation; on the right is the full general circulation model (GCM)
increases the saturation vapor pressure by up to a factor of two. This
temperature effect, leading to a water vapor increase in the 15- to
2~km region, may be much greater than that due to methane at
these altitudes.
We are beginning to understand the influence of the antarctic
ozone seasonal depletion on the ozone climatology of the Southern
Hemisphere. It appears that dilution is occurring, causing significant
ozone decreases throughout the Southern Hemisphere, as pointed out
in the presentations by Robert Watson and F. Sherwood Rowland.
OCR for page 75
THEORETICAL PROJECTIONS
~01
02
.03
.05
in
n
~ n
s.o
1O
20
sot
20C
30
50(
1000 it- -- 1 1 1 1 1 1
85.5N 76.5 67.5 58.5 49.5 40.5 31.5 22.5 13.5
75
HERO
__
75
70
65
60
'1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
_ - _ 31
1 1 1 1 1 1 1 1 1 1 = 40
4.5 -4.5 -13.5 -22.5 -31.5 ~40.5 -49.5 -58.5 -67.5 -76.5 -85.5S
LATITUDE
_55
_ O
_
50 E
-
45 ~
V
_
40 ~
C)
_ 35 ~
aS
_ 30-
25
20
15
10
result. (Reprinted, by permission, from Fels et al., 1980. Copyright ~ 1980 by
the American Meteorological Society.)
We currently know much less about the situation in the Northern
Hemisphere. However, the northern high-latitude region probably
contains more tote] amount of chlorine species at any particular
time in winter than does the southern region in winter, because the
Northern Hemisphere region is dynamically more active. Strato-
spheric cooling due to the combined effects of greenhouse gases and
reduced ozone will likely be accented in higher latitudes. We al-
ready know of the existence of some polar stratospheric clouds in the
northern polar vortex, although they do not form nearly as efficiently
there as in the Southern Hemisphere. The reason is not chemical but
OCR for page 76
76
JERRY D. MAILMAN
dynamical; Northern Hemisphere dynamical mixing is greater than
that in the Southern Hemisphere. But as the stratosphere cools
due to changed greenhouse gases, polar stratospheric clouds should
become more common and widespread in the north. Also, reduced
transport circulation in the Northern Hemisphere could lead to a
further decrease of temperatures at high latitudes. Thus the kind
of ozone depletion due to heterogeneous chemistry that is now well
documented in the Southern Hemisphere may become common in
the Northern Hemisphere as well. ~ think that much work needs to
be done, not only in field observations, but also in theoretical studies
and modeling, to quantify this change and the other possible changes
that ~ have mentioned.
(In answer to a question): Only GFDL and NCAR have looked
in detail at dynamical modeling of stratospheric climate change. We
have learned that the stratosphere has a peculiar nonlinear "switch."
This has been noted for many years in sudden warming events,
which can temporarily push the stratosphere far out of radiative
equilibrium. When this occurs, it seems to be much easier for the
system to keep the new configuration, despite the large radiative
imbalance. The reason goes back to wave propagation theory, in that
if the winds are very strong and the system ~ very cold, planetary
waves are refracted toward the equator with great efficiency. If zone]
winds decrease, then the effects of tropospheric forcing are more likely
to focus toward higher latitudes. In the climate-modeling case, the
models had a cold bias that made model stratospheric receptivity
to planetary waves too weak, which in turn accentuated the cold
bias. However, if warning is introduced from another cause, then
the "switch" may be thrown, rapidly putting the stratosphere into
another, quite different, quasi-equilibrium state. Such a switch could
also work in the opposite direction. The antarctic polar region, for
example, is strongly resistant to any wave forcing because it is on the
cold, high-wind-speed side of this implied "dynamical limit." The
arctic region, on the other hand, is presently not so constrained by
this dynamical limit because it experiences relatively high levels of
dynamical forcing.
(In answer to another question): Rowland, in his talk, and others
have speculated that the antarctic ozone hole-generating process may
already be operating to some extent in the Northern Hemisphere
polar region. However, the ejects should be much less noticeable
OCR for page 77
THEORETICAL PROJECTIONS
77
because of the greater dynamic variability there as well as likely
smaller levels of ozone chemical destruction.
(In answer to a question about mode} resolution and mode] ac-
curacy): Fifteen years ago, we were constrained to work with 9° reso-
lution because of computer and other limitations, but we knew from
comparison with observations that we were in big trouble without
knowing why. About ~ years ago, we progressed to finer resolution,
and results began to reflect much more the real stratosphere. We also
found that there were several other modeling problems to contend
with. We learned that low-resolution models cannot resolve gravity
waves. We also learned that we could sometimes get the right answer
even though the radiation was incorrectly specified. (One way to
"fix" the cold bias in the low-resolution models was to put in a bad
radiation code that allowed us to get closer to the observed temper-
ature.) Only recently have we learned that at least the winter half
of the year is dynamical in a fundamental way. There was a tacit
assumption that the stratosphere was in radiative equilibrium in the
wintertime except during sudden warming events. However, we have
learned that tropospheric forcing of the stratosphere does not permit
radiative equilibrium to be established, a fact that was first theorized
by Dickinson (1975~.
To put together a stratospheric mode! that does not ~cheat," in
the sense of forcing a lower boundary condition or including a bogus
radiation parameterization, requires an extremely long and strong
commitment of an interdisciplinary team convinced that spending a
decade on the problem is worth it. As a result, there have been few
sustained participants in comprehensive stratospheric modeling.
We knew, even 15 years ago, that stratospheric modeling in-
volved more dynamics ant} three-dimensionality than we were able
to represent at the time. My chemical colleagues have challenged
me with the following question: What can be said, on the basis of
three-dimensional dynamical models, about the viability of one- and
two-dimensional models? One thing we did learn is that one- and
two-dimensional models do have a theoretically defensible fundamen-
tal basis. In both cases, we have learned that the basis is trickier than
had been assumed. For example, in one dimension, the eddy diffu-
sion coefficients are a function of the chemistry. Two-dimensional
models can capture much of what a three-dimensional mode} does in
a self-consistent way as long as they stick to prescribed transport,
but when dynamical adjustment of the stratosphere is a dominant
factor, the two-dimensional models are completely inept. Even so,
OCR for page 78
78
JERRY D. MAILMAN
this mode} hierarchy has great value provided we are also aware of
the limitations of each type of model.
Even the 1° dynamical mode! gives polar temperatures in the
lower stratosphere that are a bit too cold. The lower stratosphere is
still not dynamical enough, probably because gravity waves are not
properly represented. We may have to resort to parameterizing the
effects of these gravity waves. However, when a parameterization is
introduced to "fix" a model, one is not really justified in perturb-
ing the mode} climate unless the parameterization is also perturbed
accordingly. Here the modeler is faced with a dilemma, since the na-
ture of most pararneterizations is that their variation under changed
climatic conditions is unknown.
REFERENCES
Dickinson, R.E. 1975. Energetics of the stratosphere. J. Atmos. Terr. Phys.
37:855-864.
Fels, S.B., J.D. Mahlm an, M.D. Schwarzkopf, and R.W. Sinclair. 1980. Strato-
sphere sensitivity to perturbations in ozone and carbon dioxide: Radiative
and dynamical response. J. Atmos. Sci. 37:2265-2297.
Mahlman, J.D., and L.J. Umscheid. 1987. Comprehensive modeling of the
middle atmosphere: The influence of horizontal resolution. Transport
Processes in the Middle Atmosphere, G. Visconti and R. Garcia teds.),
NATO ASI Series C: Mathematical and Physical Sciences, Vol. 213, D.
Reidel Publishing Co., 251-266.
Watson, R.T., M.J. Prather, and M.J. Kurylo. 1988. Present State of Knowl-
edge of the Upper Atmosphere 1988: An Assessment Report. NASA
Reference Publication No. 1208, National Aeronautics and Space Admin-
istration, Washington, D.C.
World Meteorological Organization-National Aeronautics and Space Adminis-
tration (WMO-NASA). 1986. Atmospheric Ozone 1985: Assessment of
Our Understanding of the Processes Controlling Its Present Distribution
and Change. Global Ozone Research and Monitoring Project, Report No.
16, 3 vole., WMO, Geneva.
Representative terms from entire chapter:
theoretical projections
