TT 66 6219
U.S.S.R. LITERATURE ON AIR POLLUTION
AND RELATED OCCUPATIONAL DISEASES. VOL.1
PARTS I&2
B.S. Lev!ne
U.S. Department of Commerce
Spr i ngf ield, Virginia
1966
DISTRIBUTED BY:
CLEARINGHOUSE
FOR FEDERAL SCIENTIFIC AND TECHNICAL INFORMATION
U. S. DEPARTMENT OF COMMERCE,/ NATIONAL BUREAU OF STANDARDS / INSTITUTE FOR APPLIED TECHNOLOGY
-------
REPORT
selection aids
Pinpointing R&D reports for industry
GOVERNMENT-WIDE INDEX—semimonthly guide to Federal scientific and
technical reports. Annual subscription $22.00 ($27.50 foreign mailing).
Single copy $1.75. Order from Clearinghouse, Springfield, Va. 22151.
U. S. GOVERNMENT RESEARCH AND DEVELOPMENT REPORTS
— semimonthly journal announcing R&D reports and listing selected current
R&D projects . Annual subscription $30.00 ($37.50 foreign mailing). Single
copy $2.25. Order from Clearinghouse, Springfield, Va. 2215L.
TECHNICAL TRANSLATIONS—semimonthly foreign technical literature
announcement journal listing Government and private sources . Annual
subscription $12, add $4 for foreign mailing, single copy 60 cents. Order
from Superintendent of Documents , U. S. Government Printing Office.
FAST ANNOUNCEMENT SERVICE—summaries of selected R&D reports
mailed by industrial categories, $5.00 annual subscription.
BIBLIOGRAPHIES—more than 200 available at $1.00 each. Write for a list.
Some examples:
SB-475 (Suppl. 2) Information, Storage and Retrieval
SB-541 Fluid Amplifiers
SB-543 Numerical Control Tool Systems
SB-544 Life Testing of Materials
SB-546 Space Environmental Conditions (Materials)
SB-547 Cryogenics
PACKAGED REVIEWS—resumes, abstracts and bibliographies $1.00 each.
Write for a list.
Some examples:
OTR-116 Numerical Controls
OTR-128 Plastic coatings
OTR-129 Brazing Technology and Application
OTR-132 High Temperature Adhesives above 500°F
OTR-139 Flexible, Transparent Packaging Materials
Clearinghouse for Federal Scientific and Technical Information
5285 Port Royal Road, Springfield, Va. 22151
-------
U.S.S.R. LITERATURE ON AIR POLLUTION
AND RELATED OCCUPATIONAL
DISEASES
Volume 13
A SURVEY
by B. S. Levine, Ph. D.
Part I - Atmospheric Ozone. Results of U.S.S.R. International Geophysical
Year Studies Presented at the 28-31 October 1959 Conference
Reports and Resolutions
Part II - Atmospheric Ozone. Data Presented at the 21-23 May 1963
Conference on Atmospheric Ozone
Processed by
CLEARINGHOUSE FOR FEDERAL SCIENTIFIC AND
TECHNICAL INFORMATION
U. S. DEPARTMENT OF COMMERCE
Springfield, Virginia
-------
- -- ""i"--
i \- .
i "
I
:! '
U. S. S. R. LITERATURE ON AIR
ANDRELATED~OCCUPATIONAL
POLLUTION
DISE AS'ES
. .
.i
VOLUME 13
Part I .- Atmospheric Ozone. Results of U. S. S. R. International Geophysica.l
. Year Studies Presented at the 28 - 31 October 1959Conterence
Reports and Resolutions' . .
Published in 1961 by the M. V. Lomonosov Moscow State Univer.sity'
The present English edition is a part of a
survey conducted by B. S. Levine, Ph. D.
of Washingto~, D. C., supported by PHS
Research Grant AP - 00176 awarded by the
. Division of Air Pollution of the U. S. P. H. Service
'Part II - Atmospheric Ozone. Data Presented at the 21 - 23 May 1963'
" Conference on Atmospheric Ozone . .
Published in Leningrad by the Hydrometeoro1ogica1 Publishers in 1965
The present English edition is a part of a
survey conducted by B. S. Levine, Ph. D.
of Washington, D. C., supported by PHS
Research Grant AP - 00176 awarded by the
Division of Air Pollution of the U. S. P. H. Service
/
.. ~.
Transliterated Russian Titles:.
P art I
. ." .
. '.
. ..
- Atmosfernyi ozon. Rezul'taty rabot mezhdunarodnogo
geofizicheskogo goda v SSSR. Konferentsiya 28 - 31
. Oktyabrya 1959 goda
Izdate1'stvo Moskovskogo Universiteta, 1961
Part II - Atmosfernyi ozon. Materialy III Mezhduvedomstvennogo
soveshchaniya po atmosfernomu ozonu 21 - 23 Maya 1963 g.
Pod redaktsiei G. P. Gushcina .
; .
Gidrometeoro1ogicheskoe izdate1'stvo, L~ningrad, 1965
..,
-------
INTRODUCTION
Ozone is perhaps the most intriguing constituent of the earth's
atmosphere.
In some of its manifestations it is protective, in
others destructive to biota.
Created by ultraviolet solar radiation
in the high atmosphere it protects living substances at the sur'face
by filtering out harmful wave lengths from the sun.
But there is
a1so considerable evidence that there are, in addition, natural
sources of ozone at the surface.
I
To these modern civilization has
I .
~ added the photochemically produced ozone from automobile exhaust
I
gas es .
This smog-born ozone is not only an irritant to humans but
toxic to plants and destructive to rubber.
Some have even suspected
I
I ozone as the elusive link between weather and disease.
spheric ozone may soon be an environmental hazard when supersonic
;
I
I transport aircraft soar to great heights.
The strato-
Knowledge about atmospheric ozone has accumulated only slowly.
Aside from some local analyses prompted by air pollution problems,
a major impetus to worldwide observations and analysis has been the
International Geophysical Year (IGY) and the subsequent years of
I International Geophysical Cooperation (IGC) and International Year
of the Quiet Sun (IQSY).
Nearly all nations participated with im-
proved procedures, and scientific results accelerated.
For wes tern
workers in this and related fields it is therefore much to be wel-
comed that Dr. B. S. Levine has undertaken the great effort to
-------
----,
-\',
"
translate two of the principal Soviet symposia dealing with atmospheric
ozone obs~rvations during part of these international intervals.
The two parts, reflecting conferences originally held in 1959
and 1963, contain 29 scientific papers.
These cover a wide variety of
subjects:
New instrumentation and chemical analysis systems; vertical
distribution of ozone; seasonal variation of ozone at various 10ca1-
I.
I'
ities, with some particularly valuable data from the arctic zone;
relation of OZone to atmospheric circulation; some theoretical work
. on the s,truc ture of. the OZone molecule; and a review on .relations
between ozone concentration and radioactive fall-out, based on western
and Soviet data.
Ozone observations at Soviet stations for the year 1963 are
appended to the report of the Second Conference.
Such local sources
of information are, of course, now superseded by the worldwide
publication, under auspices of the World Meteorological Organization
(WHO), through the Canadian Meteorological Service.
These translations permit an assessment of the Soviet work in
this field and add to the useful information on an important element
of the environment.
September 1966
H. \:'. Landsberg, .Director
Environmental Data Service
Environmental Science Services
Administration
I .
I
-------
I .
CONTENTS
Part I - pages 1 - 140 .
Contemporary Problems of Atmospheric Ozone Studies.
A. Kh. Khrgian
R,ocketborne Measurements of Vertical Atmospheric
: . Ozone Distribution. L. A. Kudryavaseva
Photoelectric Spectrophotometer for Atmospheric Ozone
. Observations. V. A. Iozenas and A. P. Kuznetsov
Chemical Ozone Content Determination, A. S. Britayev
Soine Results of 1958 Arctic Ozone Observations.
G. U. Karimova
I
Time-Dependent Variations in Total Atmospheric Ozone over
Dixon Is. and its Correlation with Meteorological Elements.
T. S. Gol'm
1. Instruments, Methods, and Experimental Errors
2. Time-Dependent Variation in Total Ozone Content
Observed at Dixon Is.
3. Relationship Between Total Atmospheric Ozone
Variations and Basic Upper Troposphere and Lower
Stratosphere Meteorological Elements
Moscow Vertical Ozone Distribution Observations, A.P. Kuznet-
sov, V. A. Iozenas, A. S. Britaev
Total Ozone Fluctuations in Abastumani Between July 1957
and June 1959. Sh; M. Chkhaidze
Systematic Errors in Filter,;,Equipped Ozonometers. L. G.
Bol'shakova, A. L. Osherovich, 1. V. Peisakhson
Some Photoelectric Ozonomet~r Types, A. L. Osherovich
and S. F. Radionov. .
Filter-Equipped Ozonometer.
Three-Channel Ozonograph with Diffraction Gratings
- 1 -
1
3
9
11
22
28
28
31
33
3E;
42.
44
50'
51
53
-------
. !'
I.
O:zone
, 1.
2.
3.
4.
and General Atmosphere Circulation. G. 1. Kuznetsov
Introductory Remarks
Daily Ozone Fluctuations
Circulation Index
General Relationships Between Ozone Concentration
and Atmospheric Circulation
Speculations Concerning the Mechanism of Connection
between Atmospheric Ozone and Circulation
Southern Hemisphere Characteristics
5.
6.
,Aer.osol Origin of Atmospheric Ozone (A hypothesis)
V. D. Reshetov
1. Introduction
2. Formation of Free OH Radicals Above the Surface of
a Wet Aerosol as a Result of Selective Sorption at the
Water-Air Boundary
'3,. Ozone Formation in a Wet Aerosol Initiated by Free
i OH Radicals
4:. Characteristic,s lof Atmospheri:c Ozone Phenomena
Atmospheric Ozone Temperature Regime According to
Spectroscopic Ground Observations. R. S. Steblova
I: .
A Method for Computing Total Atmospheric Ozone Measurements
Made With Light Filter Equipped Instruments. G. P. Gush-
chin
Regularities in Horizontal Distribution of and Seasonal Changes
in Atmospheric Ozone. G. P. Gushchin
Introduction
Seasonal and Latitudinal Atmospheric Ozone Changes
Seasonal and Latitudinal Total Atmospheric Ozone
Patterns
Some Conclusions Arrived at From the Equations for
Calculating Seasonal and Latitudinal Ozone Fluctuations
Instantaneous Horizontal Distribution of Total Atmospheric
Ozone
1
Connection Between Atmospheric Ozone and Meteorological
Conditions. A. S. Britaev and A. P. Kuznetsov
Two Important Features of Ozonometric Instruments.
. G~ P. Gushchin
- 11 -
57
57
5H
60
,63 .
6S'
71
73
73
73
76
80
86
101.
107
107
107
III
117
1: 18
123
126
-------
II
!
I
Diurnal Course of Atmospheric Ozone.
I .
G. 1. Kuznetsov
A. Kh. Khrgian and
Comparison. of Ozonometric Instruments made at the Main
A. T. Voeikovo Geophysical Observatory. A. A. Znamenski
Resolutions of the Conference on Atmospheric Ozone Held in
Moscow, on 28-31 October, 1959
, I
I
Part Il - pages 142 - 275
Foreword
I
Causes of Rapid Winter Temperature Variations in the
Arctic Stratosphere, G. P. Gushchin
Some Results of Ozone Observations Made 15 February 1961
During a Total Solar Eclipse. A. Kh. Khrgian and
G. 1. Kuznetsov
Effect of Circulation Conditions on the Distribution of
. Total Ozone in the Arctic. 1. M. Dolgin and G. U.
Karimova
Characteristics of Winter and Summer Air Circulation in the
Northern Hemisphere Stratosphere. Kh. P. Pogosyan
and A. A. Pav10vskaya
A Comparative Analysis of Observed Planetary Distributions
of Ozone and Certain Radioisotopes in the Atmosphere.
1. L. Karol'
1. Introduction
2. A Brief Review of Data on Wor1d- Wide Atmospheric
Ozone Distribution and Seasonal Changes in its
Concentrations
3. Data on the Worldwide Distribution of Some Radioisotopes
in the Atmosphere and on Seasonal Variations in their
Concentrations
4. A Comparison of Meridional Ozone Distribution and
Seasonal Changes with Strontium-90 Concentrations
in the Atmosphere
5. Resume
6. Conclusions
- iii -
132
134
139
142
143
164
170 \
177
185
185
185
188
195
202
203
-------
I
. I .
. Results of 1962 Atmospheric Ozone Observations in Om~k
. in Juxtaposition with some Meterological Elements;
L. A. Govorushkin . . :
jl
~hemical DEitermination of Ground Layer Ozone
I:'. :If. Svistov .
I
Meth~ds' for the Calibration of Zenithal and Lunar
Ozopometer Assemblies. K. 1. Romashkina
at V oeikovo.
I
!
I
Universal
Structur.al Ozone Molecule Models.
O. M. Rozental'
Atmospheric Ozone and its Effect on Some Vegetation Species
G. P. Gushchin
.1
:Resolutions of the Third Interdepartmental Scientific Co~erence
. .onAtmospheric Ozone held on 21 May, 1963 at the qGO,
I Leningrad . .,
I !.
. I
[Soviet St'ation Data on Total Atmospheric Ozone for 1963. .
G. P. Gushchin, 1. 1. Romashkina, and O.N. Chemyakina
of the State Geophysical Observatory
I ..
I
. I
!
,
;
,
,
, !
i .
- iv -
205
213
220
225
229
233
i
. I
235
-------
CONTEMPORARY PROBLEMS OF ATMOSPHERIC OZONE STUDIES
A. Kh. Khrgian
The atmospheric ozone problem has become an important phase in
. the study of atmospheric physics. Numerous observatories for ozone
strtdies have been built iri the USSR and abroad whiCh conduct their investi-
gations within the scope of the International Geophysical Year (IGY); .
dif;ferent scientific institutes study the properties of atmospheric 'ozoI1e
and methods for 'quantIfcifive-;de.termination; its synoptic and dynamic
mJteorology specialists investigate the relationship between ozone and at-
m6spheric currents, temperature, etc. Ozone can be regarded as a prob-
lerh of the upper atmosphere layers, since short-wave solar radiation
wliich fails to reach the Earth is a factor in ozone formation. However,
ozone changes also vary with the horizontal arid vertical (and also turbu-
lent) atmospheric currents generated in. the lower atmospheric layers.
Therefore, the study of ozone extends over the most important problems of
. present day atmospheric physics and engages a wide circle of scientists
interested in the properties of atmospheric ozone.
The Ministry of Higher Education, the Main Administration of the
Hydrometeorological Service, and the USSR Academy of Sciences called
for a conference on atmospheric ozone which convened in October 1959.
The. organizing committee consisted of: Chairman A. Kh. Khrgian from
the Ministry of Higher: Education, and Secretary G. 1. Kuznetsov, both of,
Moscow State University (MGU). S. F. :.Radionov of Leningrad St~te Uni-
vet-sity (LGU), Ye. S. Selezneva, and C. P. Gushchin, of the Main Geo-
physical Observatory (GGO)j and F. F. Yudalevich, colaborator at the In-;-
st~tute of Atmospheric Physics, representing the Committee of Atmos-
pheric :r>hysics of the USSR Academy of Sciences. Reports were pre-
sented by 1) A. Kh. Khrgian, Ye. S. Kuznetsov, A. P. Kuznetsov, G. 1.
Kuznetsov, and V. A. Iozenas from The Moscow State University, 2) S. F.
Radionov, and A. L. Osherovich from The Leningrad State University,
3) G. P. Gushchin, A. A. Znamenskii, R. G. Romanova, K. 1. Romash-
kina, and B. Ye.. Shneyerov from The Main Geophysical Observatory,
4) A. S. 'Britayev, L. A. Kudryavtseva, and V. D. Reshotov from The
Central Aerological Observatory, 5) R. G. Karimova of The Artic Institute,
6) T. S. Holm of the Dickson Island Observatory, and ~. M. Chkhaidze of
,The Abastumani Astrophysical Observatory. All Soviet institutes engaged
in atmospheric ozone research w:ere:':repr~sented. . Another participant in
the. conference was Yu. M. Ye'mel'yanov, Chairman of the Chemistry
Department of Moscow State University and organizi~g committee member
of the All-Union Conference on Ozone Chemistry held in 1960~
- 1 -
I
I
~l
.i
,;'
-------
Other conference representatives were associated with the Geogra.phy
Department of Moscow State University, Institute of Weather Forecasting
State Optical Institute, Institute of Atmospheric Physics of the USSR
AcaderY?-Y of Sciences, Alma-Ata Observatory, NIFI LGU, NIIAK, and
Institute of Applied Geophysics of the USSR Academy of Sciences.
The wide and varied representation resulted in the reporting of
. a broad scope of information and in a multilateral discus sion' and evalua-
tion of reports, particularly of suggested methods, data interpretation,
etc. The subjects covered by the reports can be classed into the. follow-
ing three groups:
1) methods for the quantitative determination' of ozone and its ver-
tical distribution; 2) physics of formation processes and evolution of ozone
layers; and 3) mechanism of ozone variations associated with synoptic anc'"
meteorological conditions. Most reports dealt with important experimental
data, much of which was compiled during the IGY and filed at the Inter- .
national IGY Data Center. The general feeling was that the scientists were
. far from having attained theIGY goals, including atmospheric ozone prob-
lems; still the reports presented at the Conference indicated that consider-
able progress was made during 1957-1959. . The conference also pointed
out the large volume of compiled and classified data, the nature of pre-
limina~y data processing began by Soviet and overseas stations, and,
primarily, the usefulness of IGY data to future studies.
Reports read at the conference emphasized the value of observa-'
tions m'ade from a network of Soviet ozonometric stations in cooperation
with observatories in other countries: many new conclusions related to
geographic distribution, ozone transfer, carryover, etc., were made
po~sible only by such network cooperation. It was announced at the con-
ference .that International Geophysical Year investigations may now be
continu~d on a basis of international cooperation. This announcement
was received as welcome news. Accordingly, 'the conference requested
~hat Soviet ozono.metrists at the southern and far northern stations should
~ontinue their studies unabated during 1960-1961, so that efforts made and
conclusions projected during the IGY could be fully utilized. This stand
was made a part of the resolution to be conveyed without delay to the USSR
Ac'ademy of Scienc~s and other organizations. Institutes and other o~-
ganizations engaged in making ozone observations ~ndconducting scientific
research in this field are numerous. Therefore, the conference resolved
thci.t a special atmospheric ozone committee be created. to coordinate and
. plan. such vrrork, and keep ~n touch with international organizations eng~ged
in OZqne studie s. . ,.
, .
In summary it can. be said that the atmospheric ozone conference
proved fruitful as a ~esult of complete information exchange concerning
the status and methods of ozonmetric investigations, detailed evalua-
-- _......-.-J.--
- 2 -
-------
. tion of reports and resolutions. The conference bro'ught out the fact that
. many USSR ozonometric investigations were mutually complementary and
supplementary with regard to extent and depth of attained progress. Many
re~orts'delivered at the conference were useful contributions to scienc~.
and the field covered was considerably broader than 3-4 years ago; it is
hoped that the conference created possibilities for future expansion of
oz6nometric studies. Therefore, it was resolved that another conference
. on atmospheric ozone be convened in the next 2-3 years.
J'
ROCKET BORNE MEAS UREMENTS OF VERTICAL A TMOS PHERIC
OZONE DISTRIBUTION
L; A. fK~d~y-a~a~eva----".'''' -,
Vertical atmospheric ozone distribution was measured in 1958 by
recording ultraviolet solar radiation spectrographically as the instrument,
mounted on a meteorological rocket, penetrated into the higher atmosphere.
The spectrograph was equipped with an automatic electronic exposure
switch and a servo system which continuously directed the slit towards the
sun. The rocket and spectrograph were sent up 1 October 1958 at a 19°
solar elevation. Solar elevation was selected so as to secure optimal
spectrograph and servo tracking system operating conditions along the as-
cending portion of the rocket trajectory. Unfortunately, actual photogra-
phy of solar spectra during the rocket flight continued only up to an altitude
of 24 km, although the instrument continued to operate past that point.
Failure to record the spectra may have been caused by 1) rocket deviatio~
from the vertical by angles considerably greater th'an those for which the
spectrograph tracking system was designed, or 2) disturbance in the proper
operation of' the tracking system due to temperature variation effect on the
photoresistors which regulated the horizontal tracking channel operation.
Figure 1 shows reproductions of the solar spectra obtained during
the rocket flight. The right hand side of the figure shows the clock dials
photographed simultaneously with the spectra which were photographed
at different exposure intervals. One cycle consisted of the following four
consecutive time exposures: 2.40,1. 04, 0.29, and 0.07 sec. After three
such cycles, two additional 2. ~ s~c. exposures appear. Clock dial records
appear once per cycle. Numbers near the clock dials indicate altitudes at
which the solar spectra were photographed. The weak spectral lines at
altitudes of 6 to 11 km appearing even at high exposures may have been
caused by the same phenomenon as the complete disappearance of spectra
at 24 km, and also by the variability in the exposure times.
---_._--
- 3 -
-------
Fig. 1
')
..,
------------- -_.._~--
SOLAR SPECTRA PHOTOGRAPHED i
FROM FLYING ROCKET:
Spectrograms were processed by an MF-4 re.
cording microphotometer. Spectral line densities were
translated into relative intensities by means of a char-
-' acteristic film curve plotted along the density marks
,
recorded by a PRK-2 mercury lamp. Blackened metal-
. lic netline screens served as attenuators. Character-
is tic curves were plotted on the basis of mercury lin'~s
in the ultraviolet spectral range. Because no spectral
distribution of solar radiation was obtained during the
rockets flight outside the ozone layer, and because the
blackening of spectra due to the above shown unsatis-
. factory aiming of the spectrograph slit at the sun and
. the variable exposure times in 'flight, it was necessary
to use slightly modified methods of computing th~ ver-
tical ozone distribution. These m.ethods were suggest-
ed by V. N. Pokrovskii (1) for similar cases.
Assum.e that light attenuation in the observed
spectral region of solar radiation occurred only due to
ozone absorption and to light scattering by air mole'"
cules. Now, use two equations of atmospheric extinc - .
tion for two wavelengths 1..1 and 1..2
-_._----~--- ----
I.
.
--.J £z41,~.m,
., . .
11=10110 - .
'. .' --'1 £zd',-P.m.
I. = 10.10
(2)
(I)
where 101,102,11 and 12 denote intensities of solar radiation at wavelengths
1..1 and 1..2 outside the atmosphere at altitudes h1 above the earth IS
surface I
(Xl and O!a symbolize decimal coefficients of radiation absorption by ozone
wavelengths 1..1 and 1..2 in cml ; I
E stands for ozone concentration at altitude Zj :
h represents the solar ray path to the instrument at altitude h1;
. ai and ~2 denote the coefficients of molecular atmosphere scattering
at -radiation waveTengfus 1..1 and 1..2 per unit mass of atmosphere;
ml represents the atmospheric mass at altitude hl,
- ------._-----.-- ...--.-
. I } pz dl
. ml = --:- . - .
: H. Pt
,-.~ -~. . I .
-.------ --
where H is the homogenous atmosphere altitude; Po and p represent air
densities at the earth IS surface and at altitude z. ~espectively
-------
- 4 -
-------
-Udl--~-U- ---[R+-z)dz-: - - ._~-~ --.
1 Y (R + z)t - (R + h.)ICOSIIl
where R is the earth's radius and e is the angular altitude of the sun.
.! Divide equation (1) by equation (2)
I
---_.--._- _0____._- ~ -.-- .--- '~------iD------- -- .
6a f EJlI, i- 6>nt.
(!.!.) = 101 10 ;., .
. II fa,. In
H~RE... Aa == ai-at H ~--,::;,..ft.l.=jh
USE A SIMilAR EXPRESSION FOR ALTITUDE H2
...
6. r £~dl.+6pm. .
(!.!-) =!.!L 10 A. . .'
I. A. 10, 0
(3)
(4)
---------~-----------~---
FROM (3) AND (4) OBTAIN
- - .
S Ez dl1 - S Ez dl. =~ [Ig (ll) - Ig (!l) - A~ (ml- ml») . (5)
~CI II fa, II fa,
~ fa, . .
,'-.- EX.PRESS THE -FI~STINT'£-~~L--IN.THE-LEFT-PARToOF (5) _. -- ----"-~
IN THE FORM OF A SUM OF TWO INTEGRALS
. ".'--'-- 1""'1-.1." ...... -- ~
... fa,.-
S Ez dll = J £zdll + r £zdl1.
h, ., i.
--'---,-~~ --'-----------'--'-~--
----------' -- - -
IF INTEGRAL OF HEIGHTS FROM HI TO H2 IS SUCH THAT ITS OZONE
CONCENTRATION CHANGED BUT SLIGHT~Y WITH HEIGHT THEN
fa, . fa,
J £z ~/l =Ez -( d11.
II. . ~I
- --------_._------------~--~-
BY MANIPULATING EXPRESSION (5) OBTAIN
Ez = ~[(~lg2L_~ --1- OOfpzdl) -..
'0' . ~a I. ~a HOp.,. .
. d!a. "
h'( \. ..) ] I It, .
1 II ~~ 1
- -Ig--;----Spzdl +
~a 'I ~a HoPo
" Ai.
+ -,d- (j £z dl.- S £z dll) .
f~1 fa, fa, .
;., .
(6)
.
Since the last term on the right part of (6) is considerably smaller
than the first, the equation can be used in computing the vertical ozone dis-
tribution by the method of successive approximations. Assume a certa.in .
atmospheric ozone distribution E and substitute it into the right side of
equation (6), and compute the firJ1 approximation.E . Subsequently,
substitute the value obtained into equation (6) and ob1ain the next approxima-
tion. In the spring of 1958 A. P. Kuznetsov and A. S. Britaev of the Cen-
tral Aerological Observatory obtained ground measurements by the inversion
method (2). Results of their measurements were used as the first approxi-
mation of vertical ozone distribution.
- 5 -
-------
...
i J'" 0 d/
Preliminary computations indicated that the term ,z
, h
~,~c,::e.ased with altitude and cal! be neglected' above 25 km.
rapidly,
The term
,0
S Ezdl gradually decreased with altitude up to the end of the ozone maximU~Tl
,'6'-,=-:--' Having passed the main ozone mass. the integral begins to de-
crease rapidly and becomes negligible above 25 km. Correction factor
values for different layers are shown in column 2, Table 1
,
Table 1 VERTICAL OZONE DISTRICUTION ACCORDING TO DATA
OF ROCKET-BORNE DETERMINATIONS
--------,
HEIGHT a.IO' C,Il/KM £-.10' cMIKM Av. QUAORA31C
IN KM ERROR 10
'- cM/KM
0-5 0,31 1,9 :1:0,2
5-10 0,31 4,1 0,5
10-13 0,30 5,8 0,3
I 13-16 '0,28 7,6 0,6
16-18 0,27 1 1,1 0,5
18-20' 0.22 15,6 0,7
i ' 20-21 0,20 16,8 0,6
!I . 21-23 0,18 17,5 0,6
I ,[ 23-24 , 0,17 10,8 0,7
i':
, i
I
I
i
! ' ' ,
.y~iE.:~si ,(){ 19 III for different pairs of wavelengths in the 3030 - 3350 region
, I I 2 "
A '~er~ recorded'ona film as, the rocket ascended into the atmosphere.
Wa~e llengths were selected so that ~;z. O. 5 cm~l . The Vigroux (3) absorp-
tion coefficients and Penndorf (4) optical'densities of absolutely pure atmos-
phe:re ~ere used. "
. ~ - I:~ G E.dl. -= ~ ~~ dl.) 'O>DIFFi:'~'m's- - .
i Ta'Qle 1 shows average vertical ozone distribution computed for all
pairs 6f wavelengths using equation (6) and listed in, the third column. If
the co:trection factor value in the first approximation be taken into account,
then itiwill be noted that above 13 km the value is less than 2% of the oz6ne
'con'cenitration, and can be disregarded in subsequent approximations.
, !
I The computed ozone distribution indicated one maximum in the
20-30 km layer with concentrations 6f 17.5 x 10-3 cm/km.' Above 23 km
a shar~ 'decrease was observed in the ozone concentration, although it must
, 1 ", ," ,I I ' ,
be borne in ~ind thafthe obtained data. were not en.tirely reliable due to, '
errors: in the last heavily blackened photometJ:~yspectrum recorded at 24 km.
!
; Figure 2 shows a comparison between the'rocketborne and ground
vertical ozone distribution measurements (2). According to the rocket
---.-_____-n ----
- 6 -
I,'
-------
data ozone concentration in the 0-5 km layer was slightly below that re-
corded on the ground; irround me-asurements macie o'n May 27 in the 5 - -13
km layer yielded results approximately identical with those obtained dur-
ing the rocket ascent, and on June 1 results were sligh1y higher; rocket
measurements of ozone concentrations above 13 km yielded greater values.
The same Figure shows measurements of vertical ozone distribution re ~
corded during a rocket ascent in New Mexico in 1959 (5). -
Fig. 2 -------------------------------- ---______n-----
I Kftf I
" -,
- -- -- ---- ---
ID
, I I
r----,-~-'
,
"
5
I
I -,
I ,-
10581 J6 58 ,1.6 51
I ,
--'---_J
- \
I I
I I
I I
I I
I I
o
5
t
/0
!5
..
- - ---_._~--------------- -- ~------ -. --. ~. - .
COMPARISON OF ROCKET-eORNE AND ABOVE-GROUND OZOYE
OETERMI NAT! ONS
Table 2 shows the total ozone content in the 0 -24 km layer as re-
corded October 1 by rocketborne measurements, and ground measuremen~s
recorded in the Spring of 1958. It can be seen from Table 2 that the total
ozone content in the 0 -24 km layer, according to rocketborne measure - -
ments, was higher than the ground values recorded on 25 and 27 May, and-
coincided with the values determined on 12 May and 1 June. Apparently
greater ozone content during the autumn can be explained by the intrusion
of cold air observed on the At-500 and At-300 charts for October 1.
- 7 -
-------
Table 2.
DATA
TOTAL OZO/IE IN
0.0 -
24. 0 KM
--- ---- _. ---
--
12 MAY
25 MA Y
27 MA Y
I JUNE
5 JUNE
6 JUNE
1 OCTOBER
0,192 eM
0,150 eM
0,152 eM
0,198 eM
0,135 eM
0,148 eM
o ,181 eM
Cqnclusions
. 1. Recorded ozone data were secured by means of a spectrograph.
with a tra.cking system mounted on a meterological rocket; using the
method of successive approximations in processing such data the vertical
atmospheric ozone distribution was determined up to an altitude of 24 km.
Results pointed to a single concentration maximum from 21 to"23 km. .
. I
2. In computing ozone concentrations by equation .(6) it was es-
tablished that the facto17 which allowed for molecular scattering decreC!;sed
rapidly with the altitude and could be disregarded above 35 km. The cpr-
rection factor in the first approximation could be disregarded abqve 24 km;
it did not exceed 2% of the ozone content at 13 km. ..
3. Comparison of rocket recorded and ground made measurements
indicated that ozone concentration values were of the same order for both.
Bibliography
- - ~.~ - ,--~-- --~._--- -- ------ --- -_.-._-_._-~------ -- ---- -------~ u_-...
- - ------
I. II" kp 0 B (" " H ii B. H. TpY.:1b1 Ll.AO, Dbln. 16, 1956.
2. Ky 3 H e 11 0.. A. n.. 6 pH T a e B A. C. TpY;J.bI Ll.AO. Bllm. 32. 1959.
:1. V i g r 0 u x E. Annales de Phys.. 8. 709. 1953.
I. Pen n do r f R. Journ. Opt. Soc:. Amer.. 47. No.2, 1957.
5. John~on. Pun:ell. Tou~ey. Journ.Geoph.Res..56.No.4. 584.
-_... --_u.
- 8 -
. i
-------
PHOTOELECTRIC SPECTR.OPHOTOMETER FOR
I
ATMOSPHERIC OZONE OBSERVATIONS
V. A. Iozenas and A. P. Kuzne ts ov
A spectrophotometer was designed and built at Moscow State Uni-
versity based on a domestic double quartz monochromator (DMR-l), for
measuring spectral intensity of zenith scattered solar light. The instru-
ment operates basically on the same principle as Dobson's spectrophoto-
meter (I). A special feature of the instrument is the use of a narrow band
. amplifier and a vibrating-reed converter. In addition to themonochroma-
tqr which separates two narrow wavelength regions in solar light spectrum;
t~e instrument is equipped with a synchronous detector consisting of a
v~brating-reed converter, a FEU -19 radiation detector, and a DC amplifif.:r
with an output null-microammeter. The instrument's optical system is
. illustrated schematically in Figure 1 where P1 is a rotating prism which
reflected light incident on the output slit from the zenith; Sl is the mono-
chromator input slit; S2 and S3 are the monochromator middle slits; S4 an.d
S6 are the monochromator output slits; D1 and D2 are the dispersive quartz
prisms j M2 and M4. are the collimator parabolic mirrors j M1 and Ms are
plane mirrors; P2 and Ps are the rotating prisms; V is a vibrating-reed
converter; W is the optical (step) wedge; L is a lens which projected an
image from prism D2 onto photocathode FEU -19.
Fig. 1 -
---------~- '¥;{~ ~---- -- ..--
.' . ~~2;i~~~tf~O~~1EC9
M.4~ -r't!_:~.,.
. .., V\..../=:;~:'~.
. M~----------___--:S3~~8>-m-
. :a t"~:-:---:.-:_-------:- ~:.- ""f-':UP1
,-._=-:.-:.=--:------~s . s-;'
--:: =-VI- .
.. .- L\ -~-=-~-----
OPTICAL PLAN OF THE PHOTOELECTRIC SPECTR~PHOTOMETER
----------
The relative instrument aperture is 1:5. The instrument dispersion
is fourfold that of a single prism. In the 3200 A spectral region the back:-
scattering was' 25)'/mm. The width of the input slit .Sl is 0.2 mm. Slh
Sa emanated a spectral region of approximately :3114 A, and slit"53 of
approximately 3324 A. Widths of slits S2 and S3 were correspondingly 0.2
mm and 0.5 mm.
----~----
- 9 -
-------
The emanating spectral intervai wavelength 3114 A vias io A Wave-:
length shifts can be realized by turning the reference drum connected with
the dispersive prisms by a swivel mechanism. Atmospheric transparency
.. .
in a vapor of 3324 A wavelength can be controlled at 356D A. The instru-
ment's electrical system is presented schematically in Figure 2.. The
photomultiplier signal was fed to the input of an AC amplifier. - Owing to
a selective T-filter negative feedback, the amplifier exhibited reasonance'
properties. The maximum gain at 90 cps was approxi~~tely 104. The
vibrating-reed converte'r frequency was v'ery stable enabling a narrowing'
of the amplifier frequency band to 5 cps, thereby enhancing the signal-to-
noise ratio. .
Fig. 2
-.---_._--
- - -----..-- - ----.--
--. - ~--- --- -. --
IIJ~"
.JlXJ'
-"\
-,
76011 .
---- ----------'------"".'---"'---- ----- -. .--_.... .
ELECT~ICAL PLAN OF THE PHOTOELECTRIC SPECTROPHOTOMETER
Ratio of two wavelength intensities was measured by the null method.
Light flu'F balancing was done using an optical (step) wedge W . Balance was
controlled by means of a null-microammeter with a 3JJ. scale placed at
the amplifier output. The optical wedge was made of 55:-4 optical glass on
a quartz backing with a 2-unit density differential. The synchronous detec-
tor consisted of a vibrating-reed converter, the armature of which support-
I .
ed a shutter for passing light beams emerging from the slits.
In experiments with direct solar light, the sun's image was foc~s6d'
on an etched quartz plate placed in front of the input slit. Power was s'up-
plied from dry batteries and a storage battery. The instrument and the
power supply we.re mounted on a trolley. To determine th~ ratio of two'
wavelength intensities the instrument's constant had to be known,which
depended on the dispersion, width of slits, instrument trans mission coeffi-
dent, arid the photomultiplier sensitivity for the corresponding wave-
lengths. For the given spectrophotometer the instrument constant .was de-
termined using the known spectral intensity ratio for a standard bulb; it
equalled 0.174. .
.----.- ---~ ~
- 10 -
-------
I
The inten~ity ratio at a 60° zenith was measured within approximate-
1y 0.5%. The instrument was calibrated at different temperatures to allow
for temperature effects. During measurements in field conditions, corre/;-
.~~on.s- were made by changing the wavelength by means of the reference drum.
Minimum light fluxes registered by the instrument around 3100 Awere iO-14
w/cm2 sec.
The spectrophotometer can measure variations in the ratio of spec-
tral intensities of solar light scattered from the zenith in a cloudless sky
in daytime, 10 minutes before sunrise, and 10 minutes after sunset.
Bibliography
---------------- -~-- -----------
- ~~. - - - .
1. Do b son G. M. B.. Phys. Soc. Proc.. 43. 324. 1931.
2. Do b son G. M. Boo Ann. of the InternaL Geophys. Year. Pub!. by Per.
gamon Press. London - New York - Paris. 5 part. No.1. 1957.
CHEMICAL OZONE CONTENT DETERMINATION
A. S. Britayev
Direct chemical ozone measurement in the lower atmospheric
layers enabled the exact formulation of boundary conditions for vertical
ozone distribution and, in particular, refinement of ozone concentration
calculations in the upper atmospheric layers based on ground spectro-
metric data. Furthermore, it was possible to use tropospheric ozone ob-
servations in computing vertical wind motion, turbulence, transfer coeffi-
cients, to study relationship between ozone, cloudiness, and precipitation,
etc. Development and C'onstruction of the electrochemical ozonometer and
chemical ozone radiosonde (6, 7, 51) stimulated and advanced the use of
chemical and electrochemical methods.
Atmospheric ozone studies were first conducted by chemical methods.
Regular ozone measurements in the ground layer were conducted one hundred
years ago in Germany by means of Schoenbein KI starch paper. First at-
tempts' to explain the origin of ozone in the earth's atmosphere were made
about the same time. Thus, .in 1854;'Dr. Harold Akkerman, a. local physi-
cian on the Island of Silt, made and reported regular ozone measure ments
in that area: "9zone is formed by electrical sparks on upper cloud sur-
faces during precipitation formation. II
. u ._--
- 11 -
-------
I '
}" I
Ozone measurements were made by the above method at the
Zwickau meteorological station in 1863 -1870, in Leipzig in 1866 -1879, in
Greitz in 1868-1879, at Fort Konigstein in 1872-1879, and from 1879 to
1908 Lepape and Colange (23) made systematic quantitative ozone measure-
ments in France at the Montsouic Observatory by oxidizing differentsolu-
tions including one of potassium iodide. In 1933 -1935 Deauvillier used
ozone to oxidize a titrated sodium arsenite solution while making ozpne '
measurements in Abisco and Scorsby-Sound (13, 14). Ozone was de~er-"
, mined quantitatively by the residual unoxidized solution dete rmined by
subsequent titration with iodine solution.'
In 1881, Schuller noted a weak luminescence in ozonized water, and
in 1910 Beger demonstrated that this luminescence was caused by oZfme
conversion into oxygen. Following this many oxy1uminescence reactions
were discove'red for the reduction, of ozone. Ozone caused luminescence
in contact with some chemical elements such as iodine, sulfur, sodium,
and thallium, mineral compounds, such as sulfides, chlorides, nitric
oxide, etc., sea water, milk, and chlorophyll. A particularly bright'
luminescence was observed when ozone reacted with luminol in alka~ine
solution. Based on the intensity of oxyluminescence, Bernanoze and Rene
, I
(5) recently developed a method for the determination of ozoneconcentra-
tions under laboratory conditions. An interesting chemical-optical tnethod,
, I
was developed in 1934 by M. A. Konstantinova-Shlezinger (4). Sheob-
served that dihydroacridine oxidation by ozone imparted a luminescence to
fluorescent acridine. Ozone measurements were made on this basiJ in the
,El'brus Mts. in 1934-1935, and air samples collected during Soviet ~ub-'
" m . .. \
stratospheric ;biilIoon-'nfghts were --analyze-ciTziT936-'and 1937~-- -- -- i
I
In 1940 -42 Emert (15 -18) proposed an electrochemical method !for
measuring ozone concentrations based on variations in the conductivity of
ozone oxidized potassium iodide solutions. Almost simultaneously Pa.ne!:h
and Caiickauf (26) developed a method for measuring ozone based on ~epo1ari-
zation of platinum electrodes by iodine liberated during the oxidatio~ of
potassium iodide solutions by ozone. In those years Emert used the: electro-
chemical method for airborne ozone determinations over Germany at al-
titudes up to 9 km. Regener of the US improved Emert's' method and used
it in investigating turbulence ~nd certain problems of mi~rometeorology
(31, 33-36), and in 1952-1953 Key (22) used it in making airborne measure-
ment of ozone concentrations at altitudes up to 12.5 km. Higher altitudes
I
were recently attained by Margetroyd (25), who made chemical ozone dis-
I
tribution determinations over England up to an altitude of 15 km while fly-
ing a Canberra aircraft. Vasey 'of France and collaborators uS,ed, automat~c
rec'ording equip'tnent in making continubus electrochemical atmospheric:' - '.
ozone measuremen~s. They conducted a series of inve~!i~at!.ons aimedat
clarifying the relationship between ozone content and weather conditions
(40, 45-47).
- -----
- 12 -
-------
I I
Regular ozone measurements are being made in Oxford (21), Hawaii
(27), Little America Antarctic Station (49), and other locations. Initial
data on the ozonosphere structure above the Antarctic were obtained with
the aid of c'hemical ozone radiosondes (24, 49) during the International Geo-
physical Year. Most contemporary chemical and electrochemical methods
used in ozonometry are based on ozone oxi.dation of an aqueous potassium
iodide solution as shown by the following equation:
'02 + 2KI :I:: HaO = 12 +02 + 2KOH
(1)
, ,-, -T-he'reaction between ozone and potassium iodide is hi~ly
-selective. ,,' According to Emert (16), Carbene and Vassy (9) potas-
sium iodide reacted poorly, if at all, with nitrogen oxides, sulfur trioxide,
and some other oxidizers which may be present in the atmosphere together
with ozone.
Equation (1) indicates that the quantity of reacted ozone can be de-
termined from the residual potas sium iodide in the solution or from th~
amount of potassium hydroxide or iodine vapor and oxygen reaction pro-
ducts. For practical purposes, the problem can be reduced to the quanti-
tative determination of the residual potassium iodide in solution, or the
iodine released in the reaction. Ozone concentration in air is normally
within the limits of' 10-7 - 10-9 g ozone li air; therefore, highly sensitive
reaction indicators should be used in the quaD:titative analysis. The amount
of reaction products is usually determined at a giyen stage of reaction by
air indicator color change (colorimetric method), or by change in the solu-
tion conductivity (conductometric method)., The colorimetric method of
ozone measurements is the simplest. The one proposed by Cauer (10 -12)
will be examined first. This method requires the passage of approximate-
1y 100 li of air containing a certain quantity of ozone through 3ml of a 2%
potassium iodide solution. The iodine vapor liberated during the reaction
is removed by the air current; the iodine remaining in the solution is then
determined at the end of the reaction colorimetrically with the aid of a
ch~~E.9fo~n: _c:l~_op i_~- a ~od~~m- acetate buffered solution. Development of an
intense' color in the chlorofo.:rm drop indicates a low ozone content in the air.
Bussy's method (8) calls for the passage of 20-30 li of air through
3 ml of a 2% potassium iodide solution containing 2 -3 drops of diluted starch
which imparts to the solution a blue color. The solution is then titrated
with a 0.001 N solution of thiosulfate to the point of complete color disap-
pearance. Ozone is determined quantitatively from the amount of thiosul-
fate used in binding the iodine and decolorizing the starch. In this way,
'the reaction' is enhanced by the thios~lfate iodine binding reaction, indicated
in equation (2) ,
12:+ 2NaaS20g = 2NaI + Na~408
( 2)
-.-
- 13 -
-------
By this method ozope concentrations above' 20 y 1m3 can be determined with
an accuracy of S -100/0. The lowest detectable ozone'concentration is 3 y m3'
The shortcoming of this method is the loss of iodine before titration which
can be accounted for partly in the treatment of experimental data. Iodine
loss can be avoided by passing the air through a potassium iodide solution
to which 96 mm3 of 0.001 N thiosulfate solution has been added, and subse-
quent electrolysis of the solution is carried out by passing a 76 \.La current
for Z min. The amount of iodine liberated as a result of electrolysis is '
exactly equivalent to the amount of iodine bound by the thiosulfate. Tq.e
amount of free iodin,e remaining in the solution (and consequently the amount
of ozone) is determined as described in the previous case.
Indigocarmine oxidation by ozone is tliesimplest colorimetric reac-
tion. It was used in many instances for qualitative analysis of comparative-
1y high ozone concentrations in the air. Air was passed through 10 ml of
a slightly colored indigocarmine solution of known dye content until the dark
- or light-blue ,color disappeared. Complete color disappearance indicated
complete trans~tion :of the brightly-:-colored indigocarmine into a colorless
isatin. The ozone concentration is calculated stoichiometrically i. e..
according to the chemical reaction equation and to the amount of air pass-
ed through the solution. The methpd IS accuracy is within 10 -lS% at ozone
air concentrations not less than 30 y 1m3 .
Oxidatibn of potassium iodide and other iodide soiutions by ozone
is accompanied by an increase in the effective concentration of hydrogen
ions in the solution (an increase in pH). This makes p6ssible the use of
color indicators, such as phenol, bromothymol, nitrophenol, phenol-
phthalein, and several other 'similar agents for' qualitative dete rminat~on
of high ozone concentrations in air. -Abs-oiute ozone concentration can be
obtained by comparing the reaction of each indicator under given analyti-
cal conditions with an absolute quantitative ozone determination metho'd.
Colorimetric methods for quantitative ozone determination though sim~ple
, in theiir procedures, possess the disadvantages of subjectivity in evaldating
color change or judging the actual moment of color disappearance.
Spectrophotometers used for regulating the transparency of a solu-
tion facilitate the application of colorimetric methods in ozonometry and
enhance their accuracy up to 2-fold. However, it shoul'd be added that
increased photometric accuracy is impeded by spontaneous changes in
solution color upon completion of the reaction with ozone. Moreover, the
rate ,of color change depended on the nature of the indicator, reaction rate,
illumination, and qther unci:mtrolrab~ experimental conditions which limit
the possible applic~tion of photo'metric control of color reactions. Ac-
curate quantitative ozone determinations are made by electrochemical
methods in which the degradation of ozone is controlled by variation of
the electrical properties of the ozone oxidized solution~ ArrlO_ng the
electrochemical methods employed in ozonometry'Eme-~t"s tIS-IS) and
- 14 -
-------
Teicherf"s (41-44)'i1re th~ most widely used. They will be discussed brief'-
ly in'the following paragraphs.
In Emert's method the ozone-containing air is passed through 3 ml
of a 2% potassium iodide solution containing a small amount of thiosulfate.
As in reactions (1) and (2), this amount of thiosulfate is used for binding
iodine liberated during the oxidation of potassium iodide by ozone. In or-'
der to determine the moment of thiosulfate disappearance, the container
with the test solution is equipped with four platimum electrodes. If less
than the voltage required for electrolysis, say 0.18 v, is applied to two
electrodes, then so-called reversible iodine electrodes a.re formed with
i. polarizing voltage which is equal to the applied voltage. The current
~trength measured in the given two-electrode network drops to zero.
Such a condition persists until the entire thiosulfate is used up and free
i:odine appears in the solution. The free iodine, fed to the reversible
electrodes by mechanical solution stirring elicits depolarization and in-
duces a current in the electrical network proportional to the agitation in-
tensity. Only a certain portion of thiosulfate contained in the solution
is normally neutralized as a result of the reaction with ozone. The re-
maining thiosulfate is neutralized by the iodine generated in the electroly-
sis as the current passes through the other pair of electrodes. The in-
strument's sensitivity permits determinations with 10 li of air; de-
termination can be made automatically. The instrumental error in de-.
termining ozone concentration does not exceed 5%. According to Wadelyn
(48) similar results can be attained by amperometric back-titration of
Na:aS:a 03 excess with 0.01 N KI03 solution.
Teichert's method determines the minimum conductivity value of
the potassium iodide solution remaining after reaction with ozone, and
titrated with 0.005 N solution of silver nitrate. During the titration in-
soluble silver iodide is precipitated according to the following reaction:
KI + AgN03 = AgI + KN03
(3)
..Mobility of nitrate ions is less than that of iodine ions, which causes the
solution to reduce at a slow rate in the course of the titration until the re-
action with potassium iodide is completed. Addition of silver nitrate in-
creases the effective ion concentration in the solution, subsequently.en-
hancing the solution's electrocondu'ctivity. Thus, from the viewpoint of
equivalence the residual potassium iodide and th.e ozone concentration in
the solution can be computed. Potassium iodide dissolved in a buffered
solution of pH = 7, consisting of 61% solution of 11.876 g Na2HP04. 2H20 in
lli of water, and 39% solution of 9.078 g KH2P04 in 1 1i of water. Electro-
conductivity measurements can be made with an accuracy up to' 0.1° /100 by
.:;-a'-bddge---:fed by a 2000-cps alternating current, equipped with an amplifier.
Since each measuring cycle lasted approximately one, hour, the measured
----- ---.- -.-
- 15 -
-------
i
i
I.:
vallies 6f ozone concentration were averaged for the given period.
The ozonometric group at the Central Aerological Observatory
built a laboratory device for the study of electrochemical reactions and
selectiop of optimal'measuring conditions. The device is presented
schematically in Fig. 1.
Fig. 1
F''''''--- h-_.
._-~._- ~.-,.----- ---- ~ ~-------------- -- - - -- ._--~..
,.
i'
.,
."
.. ,
i
I
I
! ------..-.'-'
PLAN OF THE OZONE ANALYZER
I-ELECTRICAL OZONIZER; 2-FILGERS; 3-HIGH VOLTAGE INDue-
r TOR; 4-MEASURING RECEIVER; 5-GASOMETER; 6-ELECT1ICAl AS-
I PIUTOR; "7-BRIOGE; 8--AMPLIFIER; 9-0SCILlOGRAPH; ro-pH .
I 'I ~~TER; II-STAN~ARD..!)!~N~~_~.Ei!.~,~~!~~..
.i I ; , ,
I I .~atural'air or air ozonized by an electrostatic ozonizer (1) is freed
from aerosols by passing through filters (2); then through v~ssel (4) co,n- ,
ta~ning the test solution through gasometer (5), and electrical aspirator (6).
:The measuring:or analytical vessel (4) consists of a 10-ml glass cylinder.
~q~ipped with two 1 cm2 platinum electrodes. The electrodes can be switch-
ed[ to br~dge (7) orto ielectronic pH indicator (10). AC voltCl.'ge from a" .
~d.ndar~ signal generator (11) is aRplied to one of the bridge arms (7).
When the bridge is unbalanced the 'electrical signals from theothe r arrr.
arle amplified by a wide. band amplifier (8) and are displayed on the CRO (9).
T~e bridge balancing corresponds toaminimum in the amplitude of sinusoi-
da~ oscillations. An electrochemical ozone analyzer was used to examine
pOrtassium iodide ~olution oxidation by ozone at different dilution degre~s in
distilled: water and in buffered solutions (2).
i" ,
I . A typic~l cur~e variation for a 0.001 N solution is shown in Fig. 2
(curve 1). The portion of curve 1 which runs parallel to ,the abscissa
. c~rresp~mds to total potassium iodide consumption. The pH factor in-
creased'in the course of oxidation, and attained a constant value upon com\'
plletion 9f th~ reaction. The solution turnsyellow by th'e'separation of .
frlee iodine; addition. of any of the above -considered color indicators causes
I 1
a IsharPlchange in the solution color. Curve 2 in' Fig. 2 illustrates the
change in electroconductivity'during titration of a partially ozone oxidized
! [ . - =]X ~ 0,
I I
I !
--~-- - .__._----~
, "
'1
-------
potassium iodide solution with. silver nitrate. The inflection point indi-
cates completion of the substitution reaction (3). The reaction course can
be followed by changes in the pH reverse variation values which are of op-
posite or reverse significance.
. Fi g . 2
1-
Cf
Curve 3 in Fig. 2 illustrates electroconductivity
-----..._----;--variations in a potassium iodide solution mixed
with a 0.0010/0 thiosulfate solution. The amount
of thiosulfate in the solution is sufficient to bind
10 y of ozone. It can be seen from Fig. 2 that
solution resistance increased slowly until the
thiosulfate becomes completely consumed. Fur-
ther oxidation caused an increase in the test
. solution alkalinity and a decrease in the resist-
ance until the reaction becomes completed.
I
- --------_._-
----------------" V,It The solution conductivity was determined
E"ECTROCONDUCTIVITY CHANGES IN-- With the aid of an alternating current which re-.
MEASURING SOLUTIONS ." .
I-OXIDATION REACTION OF THE KI duced the electrolytic effects. Appl1cation of an
SO!.N; 2-SILVER NITRATE TITRA- alternating electrical field elicited complex re-
TION OF PARTIALLY OXIDIZED KI versible and irreversible reactions in the solu-
SOLN; 3-0XI DATI ON REACTION OF KI .. .,. ,
SOLN. IN THE PRESENCE OF HYPOSUL- hon. The solute Ions oscIllated around fIxed.
FITE . equilibrium centers of the AC frequency. These
oscillations can be damped by electrostatical
forces which bind the ion with the ion-polarized solution portion. Accord-
ing to the theory of strong electrolytes, binding forces are determined by
reaction time", which is a function of shade diffusion coefficient D, the
damping factor T1, and solution concentration n:
-----'----------
(e =.2,7J8). .
- - T.D
. - S;:elll .
(4)
The interacting forces between the ion and its ionic atmosphere and,
co.nsequently, the solution electrical resistance, are minimal at fi.-equen"cy
1/,.. -Experimental functional relationship between effective- resistance of
different solution concentrati.ons and AC frequency accorded qualitati~ely
with results of the electrolyte theory. The present au~or1s experimental
data showed that~e-s-istance of test solutions decreased approaching a fixed
limit as frequency increased from 0 to 2000 cps . The AC electric field at
. the electrode surface causes irreversible phenomena associated with a
nonuniform rate of anode and cathode processes. The intensity of irrever-
sible processes rose with increase in the amplitude of electrical oscilla-
tions and de-creased with an increas:e. in AC frequency. Thus, the most suc-
cessful electrocon'ductivity measurements were made by a compensating
method at 1600-2000 cps frequency.
Potassium iodide solutions with sodium thiosulfate had been used
~-'-----.
- 17 -
-------
. .
since January 1959 in making electrochemical ozone measurements in at-
mospheric ground layers in D6lgoprudny, 20 km N of Moscow. Ozone con~
centrations at the 2 m level w~s compared with its total content, with varia-
tions in meteorological elements near the earth's surface,' and also with
vertical currents at 850 and 700 mb levels determined from wind velocity
divergency. Total ozone was determined by a Dobson-type photoelectric
spectrophotometer using methods described in (1). Typi'cal variations in
ozone concentration and in meteorological characteristics of a medium are
shown in Fig. 3, where ozone concentration variations in the ground layer
generally coincide with variations in the total atmospheric ozone content.
Fig. 3
-._------ ._-
811atJ aJO ,30
-.-, -2 --------3 ----.
-"-5 --, .
H r T
II/1(/(} D.511 CD
"" 001lD.3D
, ,
----..----.--------------.----.-.-..-.- ---------_._.~- -----.-----
I-COURSE OF OZONE CONCNTRN.. AT THE 2M LEVEL; 2-0Z,JfIE TOTAL.
CONTENT; 3-PRESSURE; ~HUMIDITY; 5-VERTICAL FLOWS AT 850 Me;
- ~OITTO AT 700 MB (RELATIVE UNITS) AND ATMOSPHERIC FRONTS
, OUR I NG FEBRUARY, L959
It is also evident that catabatic currents were frequently associated
with an increase in ozone content through the entire thickness of the a~mos-
. phere and at the Earth's surface. High ozone concentrations have been ob-
served predomin~ntly at the north and northwest winds, ,and in certain
cases following precipitation. The relatively low value of ozone concen-,
tration in the Moscow region compared with data. obtained for Western
Europe is of particular interest. This difference can be explained on the
basis of possible geographical factors. Diurnal ozone variations at the
2 m level have a distinctly wide maximum close to the meridional hours,
and a rhinimum at night and in the morning. Sharp ozone concentration
variations from 0 to 5'0 y/m3 and up can be observed during' unstable wea-
ther. Ozone variations do not always agree with air advection which, aPt'
parently, supports the opinions (19. 20, 33; 36-39) concerning t~e possible
existence of an ozone source near the,earth's surface. .
-------- ------_u_----. .
-, 18 -
,'.,
, I
-------
No fixed point of view exists concerning ozone origin in the tropo-
s.phere. Early concepts regarding ozone formation in the earth's atmos-
phere exclusively by the effect of ultra-violet solar radiation at altitudes
above 20 km have been challenged basically by results of recent investi-
g~tions. The first impact to the stratospheric theory of ozone origin see ms
to have come from results of California poisonous fog investigations. Ac-
cording to V. Regener (33), who measured ozone concentr'ations near Los
Angeles at ground level and in the air by using an aircraft, ozone was:
. formed in the' Los Angeles Valley directly from materials contained in th.':
su~inversion aerosol layer adjoining the ground surface. Regener regards
the high ozone maximum observed in this area during' meridional hours as
the result of hydrogen peroxide evaporation from hygroscopic aerosol par-
ticles which dissociated photochemically immediately thereafter with the
formation of ozone. According to Stevens (39) ozone was formed during
nitrogen dioxide decomposition unde r the effect of ultraviolet radiation!
which entered the air with industrial smoke. The attendant generated i
nitrogen oxide emanated from the reaction sphere in the course of inter-
action with organic' substances and formation of peroxynitrite. The latter
dissociated liberating nitro'gen dioxide, etc., thereby completing the di~
urnal ozone' variation cycle. Investigations of quantitative ozone varia-
tions in the Los Angeles area frequently produced divergent results (see
37); however, all investigators agree that local ozone format~(:m was not
connected with air transfer from the stratosphere to the 'ground iayer.
Ozone-forming factors in the lower atmospheric layers were con-
sidered within a more general proposal by Frenkiel (19, 20), who suggest-
ed, three possible sources of tropospheric ozone: (1) ozone was transferred'
from the stratosphere to the troposphere by turbulent diffusion, (2) an '
amount of ozone was formed in the troposphere through the photochemical
dissociation of diatomic oxygen under the effect of ultraviolet rays, and (3)
. 10 / 2
approximately 3 x 10 cm' sec. of ozone molecules occurr~d in the 30 to
60° northern latitude region when radiation with 3660 .A wavelength dis-
sociated nitrogen dioxide which permeated into the atmosphere from human
physiological products. The amount of nitrogen dioxide noted on a planetary
scale,' and consequently also of ozone, originated during volcanic eruptions.
forest fires and in the course of organic compounds decomposition. Ozone
was also formed during corona and electric spark discharges in the tropo-
sphere during St. Elmo's Fire, snow storms, etc., as a result of natural
soil' and air radioactivity, and during experiments with nuclear devices (38).
The question of ozone origin was considered recently by V. D.
Reshetov (3) (see Reshetov's report in this issue). - In discussing the
ae'r'osoi theo'ry of.ozone origin one must bear in mind thepossihilityof
atmospheric ozone redistribution by aerosols the solubility of which'in
water varied with the temperature. At 0° C, 0'.49 cm3 of ozone dissolved
in water. Ozone solubility in water decreased "with temperature increase
and under given conditions can lead either to partial sepa.ration of pre-
- 19 -
. I
-------
. .
I;
viously dissolved ozone or to its full liberation after comple'te waterevapor-
. ation. .
The final solution to the problem of ozone origin in the earth's at-
mosphere. and in partfcular in the troposphere. can be found only by .or-
ganizing further quantitative studies including chemical methods for the
determination of ozone concentrations in the air.
.....
I
I .
)
/
BIBLIOGRAPHY
. ----..-- - --,_._._------~._----------'----'-'-_.
I'
ii
I,
1. 6 p II .T a e B A.C.. K y 3 H e U 0 B A. n. BepTHKaJlbHoe pllcnpeAe-
MHlle 030118. Tpy;1bl U,AO. Bbln. 32. 23-35. 1959.
2. B II H 0 r p a A 0 B a E. H. MeTo;tbl onpe;te.1eJlJul KOIIueJlTpaUHH BOJJ.OpoJ('
'lIblX HOIIOB. J.h.o.-BO Mry. 1950.
3. Pew e TO B' B. ll. A3po30.1bHaR runoTe33 aT-MOCclJepttol'O 030H8. Kolfll>.
. no aTM. 030IIY. H3A'BO Mry. 1961.
4. K 0 II eTa II T II II 0 B a . ill .II e 3 II II r e p M. A. TpYJlbl 3J1b6pYCCKOA
3KcneAIIUHH 1934-1939 rr. IhD.-BO AH CCCP. 1936.
5. B ern a nos e A. S.. Ken e M. G. Oxyluminescence of a few floures-
. cent I compounds of ozone. Ozone Chern and T echnoJ.. Washingto,,: D. C.
Amer.. Chern. Soc.. 7-12. 1959.
6. B r ewe r A. W.. Mil for d S. R.. G rig g sM. An electrochemi-
, mical ozonesonde. lL'GG WMO Ozone Symposium. Oxford. July. 1959.
7. B r ewe r A. W., Mil for d S. R. Vertical measurements of ozone
distribution. IUCC WMO Ozone Symposium. Oxford, July. 1959.
~. Bus s Y W. Uber eine handliche und hinreichend genanae kolorimetrische.
... .~~t~~~~!~~~_onm~s.:s~.~~e~.~e~eo~~I~_~u_"-d~hau. N~._7:_.~4_--J.~~_.I~_,-
9 Car b e nay S.. Vas s y A. Dosage continu de I'ozone atmospherique.
, Ann. Geophys.. No.9. 200-308. 1953.' .
10.' C au e r H. Ergebnisse chemisch - meteorologischen.. Forschungen.. Arch.
Met. Geoph. Siokl.; Ser. B. 221.' 1949. .' . .
11.1 C 3 U e r H. Chemisch - bioklimatische Studien in Konigstein im Taunu s.
, Arch. physik. Ther.. I, '208. 1951. . .
12. 'C a u e r H. Ergebnisse yon Ozonuntersuchungen in der Sachsen. 1949':"-
. '1950. Arch. phys. Ther.., 3. 208. 1951. .
13 D a u v i I lie r A. Recherches sur l'ozone atmospheriQue effectuees au
. Scorsby' - Sund Journ. de Phys. ct de Radium. 7.455. 1934.
14., D a u villi erA. Compt.. Rend.. 201. 679. 1935. ..
15. E h mer t A. Uber den Ozongchalt der unteren atmosphare bei winter-
! 'lichem Hochdnick - wetter nach Messungen 1m Flugzeug. Fersch. u. Err
, Brr. d. RWD. Berlin. 13. 1941. .
J6., E h mer t A. Ein einfaches Verfahren zur absoluten Messung des Ozon-
; gehalts von Luft. Meleorol. Rur.dschau. Nr. 2. 64. 1951.
17., E h mer t A. Gleichzeitige Messung des Ozongehalts bodennaher Luft
:an mehreren stationen mit eincn einfachen absoluten Verfahren. Journ.
. : atmo~pheric terrestr. Physics. No.2. 189. 1959. .
18. E h mer t A. Uber ortliche Einfhi'sse auf den Ozongehalt der Luft. Ber.
I Dt. Wetterd.. 38. 283. 1952. . .
19., F r e n k i elF. N. On Vertical ozone profiles in the troposphere. Publi-
I. . cation IAMAP. No. 118. 90..1958. , .
" 20. F r e n k i elF. N. Troposphenc ozone. lUOO WMO. Ozone Symposium.
Oxford. July. 1959. . . '
I 21. H a r r i son F: R. Surface ozone concentration measurements at Oxford.
\LiGG WMO. Ozone Symposium. Oxford. July. 1959.
22. Kay P. H3~lepelluR c ca~IO.1eTa BepTIIKa.lblloro pacnpe.J.e.lclIlIlI 030lla AO
BbiCOTbi 12 /U. XIIMII'IeCKml ~leTO.J.O~1. C6. cPaKeTHble JlCc.le:toBaHHII. 8epxHeil
aTMocljlepblJ. no,iI pe)l. 601ba n. JI. II CIIToHa M. ll. MJI. 1957. crp.' 236-239.
. 2,3. l ~ pap e. A:. Co.1 an g e \. C. R. Acad. Sci.. 189.53. 1929. . '
.24.: M a.c D 0 w a I I S. Soundings of the Antarctic Ozonsphere 1 UOO WMO.
. Ozone Symposium. Oxford. July. 1959. '.
25. M u r g a' t r 0 y Ii R. S., Some' recent measurements ofoZO:le concentra'
tions from a Ca:1berra aircraft up to 15 km. lUGG WMO. Ozone Symposium.
. Oxfold. July. 1959. .
26. Pan e t h F.. G 1 u'c k auf E. ~\easurements 01 atmospheric ozone
by a quick electrochemical method. Nature. 147. 614. 1941.
----.---
- 20 -
-------
,
I
I.
I
I
, !
. I
i/
27. P,r ice S.. P a I e s '5. 'Some observations' cir ozone a Mauna loa oDser.
vatory.' Havaii IUGG WMO. Ozone Symposium. Oxford. July. 1959.
28. R !l s s 0 0 I S. Relation entre les epaisseur~ reduites de I'ozone atmospbe.
nque et sa concentration au niveau du sol. GeoHs. purs e applt'. No.3.
54-60. 1956. .
29. Reg e n e~. E. Uber die Schwankugen des Orons in der Tropospha're urid
Stratosphare. J. atmospher. terrestr. Physics. No.2. 173. 1952.
30. Reg e n erE. Neues yon Ozon in der Erdatmosphare. Naturwissen.
Rundschau..7. Nr. I. 8. 1954. .
31. Regen'er E.. Paetzold H..' Ehmert A. .lI.aJlbHeilwHe HCCJle-
. J1CBaHHSI 030HHoro CJlOSI. C6. cPaKeTHbie HCCJIeilOBaHIlSi BepxHeil aritoalJe-
pblt. nOJ\ pe.!1. 501i.l1a P. 11. II CIITOHS M. .Ll. MJI. 1957. crp. 230-235.
32. Reg e n e r F.. Z u c k e O. Uber die meteorologischen Bedingungen der
Ozonschicht in Bodennaher Luft. Abhandl. MHD der DDR, 2. 13. 1953.
33. Reg e n e r V. H. New experimental results on atmospheric ozone. Sci
proc. of IAMAP. London. 1956. . .
34. Reg e n e r V. H. The daily -va~jation Of atmospher'ic' ozone near the sur-
face. JUGG WMO. Ozone Symposium. Oxford. July. 1959.
. 35. We g e n e r V. H. The vertical flux of atmospheric ozone. Joum. Geo-
phys. Res. No.2. 221-228. 1957.
36. R e ~ e n e r V. H. Ozone concentration profiJe.<. Exploring the atmosphe-
re 'first'smile. 1957. pp. 188-198.
37. Saltzman B.. Gi Ibert N. Ozone reaction with I-hexene
Industr and Engin. Chem.. 51. No. II. 1415-1420. 1959. .
38. S c h rei d e r 0 v A. J. T!Jermounuc1ear expksions and the ozonosphere
Puhlication. IAMAP. No. lib. 94. 1958. . .
39.' S t e p h an s F.. Air Pollution .ab. reports from. Los-Angeles. Joum.
Franklin. Inst.. 4. 349-359. 1957.
40. Tan a e v sky 0.. Vas s y . A. Sur Ie dosage C'himique de I'ozone at.
mospherique. C. R. Acad. Sci.. 244. 924-925. 1957. .
41. 1 e I c her I F. Erfahrungen mit chemischen Ozonmessmethoden Z. f.
Meleorol.. 6. 132. 1952.
42. T e i c her t F. Verglei~hende Messung des Ozongehalts des I.uft an Erd-
boden und in 80 m Holie Z. f. Meteero1.. 9. 21-27; 1955.
43. T e i c her I F.. War m b t W. Ozonuntersuchungen am Meteorologl-
schen Observatorium Wahnsdorf. Abhantd1. MHD der DDR.. 5. Bbln. 34
67. 1955.
44. T e i c her t F.. War m b t W. Ozonmes5ungen Dresden - Wahnsdorf
. unci Fichlelbcrg ]91'411955. Zeitschr. Meleoro1.. 9. 264~277. 1956. .
45. Vas s y A. Appareil enregistreur donnanl la concentratoin de !'ozone
dans I'air. Geofis pura e appl.. I. 164-173. 1958. .
46. Vas 5 y A. Cun('ntralion de I'air en ozone a la slation sCientifique du
.iungfraujoch. C. R. Acad.. Sci.. 25. 2409-241 \. 1958.
47. Vas 5 y A. Ozor:e concenlrahon at ground level - IUGG WMO Ozone
Symposium. Oxlord. July. 1959. .
48. \V a del i n C. Determin:ltion of ozone and other oxida.nts in air. Analyt.
Chern.. 3. 441-442. 1957. . .
49. War m b I'W.. R('gionale Untersuchungen des bodennahen Ozon. Arch.
Phys. Therap.. 4. 328-336. 1958. Measuring ozone in the slratosphere.
New Sc~~ntist.. 2, 32. 1957.' .
50. \V e x I e r' H.. M 0 rei and W.. H a r I i n B.. \\' e y ant W. A pre-
liminary report on ozone obserntions at Little America. Antarctica. IUGG
WMO. Ozone Symposium. Oxford. July. 1959.
I.
!.
Ii.
.
I
I
I.
i
I !
I I.
I
I
I i
,
I
I, r
i i
I
"'
.
- 21
-------
SOME RESULTS OF 1958 ARCTIC OZONE OBSERVATIONS
G. U. Karimova
In connection with the International Geophysical Year (IGY). the
Arctic and Antarctic Scientific -Research Institute working in cooperation
with the Voeikov Main Geophysical Observatory conducted ozonometric
observations on Dixon Is. in 1957. In addition the Institute organized
supplemental studies at two other points: Heiss Is. and SP-6 (North Po1e-
6. drifting). At the present time observations are being made on the drift-
ing SP-8 station. As a result of these observations data had been secured
for the first time on total ozone content at high latitudes of the Soviet Arc-
tic (Fig. 1).
. .. -. ---
- ----.-- .-- ----. - "--.-
Fig. 1
DISTRIBUTION OF OZONOMETRIC POINTS IN THE ARCTIC
Equipment at the above stations differed. Ozone measurements
were made on Dixon Is. during daylight hours using spectrophotometer
SFD -1 equipped with a diffraction grating; OFE -3 equipment was used in'
making night measurements. Heiss Is. observations were made with a '
GGO-constructed ozonometer. The same ozonometer was used on the
SP-6 drifting station during daylight hours only. since no equipment was
available for nocturnal observations.
Ozonometer GGO consisted of an e1ectrophotometer with two light
fi1ters~ each made up of the folloWing two glass fUters: I'
1.
ZhS -3 + UFS -2 with a passband from 3040 to 3240 A.
2. SZS -18 + UFS -2 with ~Ras-=,_band from 3640 to 3800 A.
- 22 -
I
I
-------
i
To'tal ozone determinations were made at all the points by the optical me1:h-
of based on atte'nuation of direct ultraviolet solar radiation by atmospheric
ozone~. To eliminate aerosol effect observations were carried out in two
spectral regions. Increased ozone content in the atmosphere elicited shaT'p
ultraviolet attenuation in the first spectral region, which coincided with the
right edge of Hartley's band; direct solar radiation was attenuated only
negligibly by ozone in the other regions. Total atmospheric ozone was de-
termined in centimeters based on degree of ultraviolet attenuation. A
magnesium photoelement, nonsensitive in the red and infrared regions,
was used as the receiver during daytime and an FEU -18 photomultiplier
, was used at night. Observations were made during daylight hours on the
. basis of direct solar radiationj thus, observations were made every 15 min.
on Heiss Is., and.hourly on SP-6 drifting station in clear weather and dur-
ing breaks between clouds on cloudy days.
Total ozone determinations were made for the sun '8 elevation in
. the 12-35°' range. It should be noted that during spring days, when the
Arctic frequently experienced clear weather, the sun appeared low above
the horizon; during the summer the sky was frequently hidden behind low
clouds considerably limiting the chances for making ozone determinations.
The total number of ozone measurements made on Heiss Is. during March
-I September 1958 was 1870; ozone determinations made on the SP-6 drift-
i,hg station from April to August totaled only 220, due to the fact that the
S:P-6 drifting station ozone determinations were made by the station's
afrologist, who was basically responsible for carrying out the aerological
program.
Ozone determination with the aid of spectrophotometer SFD-l was
reduced to the solution of the following equation:
I
---- --- --~._--------
Lo-L-0.II0m
.1"= .
0,865:.1
Whe re
La denotes the ratio of the direct solar radiation intensities 10 and
'-TI)- -outside the atmosphere,
L denotes the ratio of I and I I direct solar radiation intensities at
two wavelengths A and A I, respectively,
#J. symbolizes optical ozone mass which depends on angular sun
elevation, -
.m siand~ for optical atmospheric mass.
A different method for ozone.calculation was used in the case of GGO
ozonometers. Data obtained with the aid of these instruments were pro-
ces~ed using a nomogram constructed on the basis of Buger's formula !or
20-A intervals at different solar elevation~ for all ozone values x, (from
x = O. 20 to x = 0.52 cm). With this nomogram.it was possible to determine
graphically the ozone content from a known ratio of intensities L at a given
--
- 23 -
-------
solar elevation with an accuracy up to 0.005 cm.
150 the error increased due to a strong bunching
the nomogram. : .
At solar elevations below
of the ozone isolines on
Prior to departing for the Arctic all instruments were calibrated
against the Dobson spectrophotometer at GGO, where it was demonstrated
that the 'error in determinations made with the aid of ozonometers GGO did
not exceed 5%. Upon return from the Arctic, the ozonometer used at the
SP-6 drifting station was recalibrated, and it was found that, during the.
five months of operation under Arctic conditions the calibration coefficient
of the iIistrumen,t changed by 6% of the initial value. The above considera-
tions, suggested that data obtained by the three stations we:re adequately
uniform and could be used for comparative purposes. .
~ !
Fig. 2 Results of total ozone determina-
.- -- _.- ._u- .. -.---' ----- --..--. ----- - t~ons made at three stations in the Arctic
Xt:.IIl II, (~ix:on Is., Heiss Is., 'and SP-6 drifting
,.!: station during March to,September are
'"~L. ........ lis t~d. in Table 1 togethe r with me an frac-
-----";::... . ! I '
, ""',~ tfonal values of the total ozone measure-
~.
".' --.....;: ments and in Fig. 2 above where the mean
~""h___.. I .
""........-.:'~.::::---_. rr;'-0nthly values are. plotted.' For the pur-
_.:.~, ppse. of comparison, graEhs inFig. 2 ,
::=.-:J' show data collected around Voeikovo dur-
~. . '
III 'lY Y If JlDI II lng the same period. The graph shows'
. i. MoriTfiS~ clearly th~t total ozone co~tent in the Arc-
------______L__. --~------'------- --------.:~-- tic decreased considerably between spring
, AVERAG~ MONTHLY VALUES OF TOTAL OZONE IN and fall. Between April and August 1958,
I . THE ARCTIC IN 1953 .
I-SP6;; ~HEISS Is; 3-Dlxoll Is; 4-VOEUOVO ozone content dropped by 440/0 on Sp-6 .
, [, t, :' . ! drifting station, by 40% on Heiss Is., and
'by 30% 'on Dixon Is., while hi the vicinity of Voeikovoozone concentration
I falllm:ountedto 21%. Similar decreases (""50%) were observed by Johansen
at T~ori.so. The sharp droplin th~ curve's amplitude at high latitudes c6in-
,cided ~th notable variations in ~olar radiation intensity between winter
II . .
and summer. I
I
: : The amount of ozone wis greater in the -Arctic Reg~on than in: Vaei -
kovo, during the ,month of April; the ratio became reversed in July; - less
ozone was found in the Arctic Region than in Voeikovo, which accorded with
the well-known conclusion concerning the occurrence of shift in the rn~xi-
. I 0" I
mum ozone conte~t during summer to the 60 latitude. . I
. ' .
0.200
: "
-j
: -r..
. ,
I
, .
; I
- I.
I. . . . . .
,Latitudihal ozone variation was tnostclea:rly in .eVidence during
April ~hen m6re o~01ie was observed bn the SP-6 'driftin'g station and
Hei~s' Is. than 6n Dixon Is. and at Voeikovo, indicating thafthe spring '.-
ozone maximum distribution shifted m~ch c1bs.er to the Pole ~--(bey:ond '800~) .
thanl'~~s t~ought'heretofore. Despite seasonal' total ozone variations
~--- -- -~---
. I"': . - 24 -
.1
I,
-------
i.
I
. .
. I
associated with changes in the solar radiation influx, short-period varia-
tions in the total ozone had been observed and attributed to atmospheric
circulation. .
Table 1 -
. A'IERASE VALUE OF TOTAL'. OZONe' .tII,Cft-.,a'T8E ARCrt C FOR 1958
I.'DIXO~ 16. HEISS 18 8P6
I: -
I
MA1ICH I r
II
III 0,651
APR IL' I (1,510
II U.4I5 0,462 0,539
III 0,422 0,458 0,548
..
MAY ' I 0,475 0,455 0,526
II 0,452 0,479 0,490'
m 0..432
JUNE i I ' 0,330 0,389
II 0,370 0,321 0,345
i III I 0,302 0.328 0,354
'
JULY: I I 0.314 0,348 0,308
I
II I 0,223 0,309 0,325
III I 0.185 0,284 0,334
AUGUST: 1 I 0,222 0,301 L 0,302'
'. '. II 0,295
. ,,fll 0,267
Although processes of ozone formation and dissociation proceeded at slow
rates in the layers below 25 km, ozone: ~n;those layers can be regarded as
a conserv~tive (minor?) air admixture~Ozone moving with an air mass
can be regarded in many instances as a specific property of a given air
mass. Thus, it appeared possible to correlate sharp changes in total
ozone with temperature changes. With this. in mind, it seemed desirable
to analyze obtained ozonometric data in the hope that a relationship could
be found between total ozone in the Arctic 'atmosphere and the aerosynoptic
conditions. April and July were selected for a detailed analysis of daily
data, since both months were most distinctly representative of the spring
and summer seasons in the Arctic. Total ozon,e was compared daily with
mean thermal conditions of the troposphere, stratosphere, and tropopause,
The 500-300 mb layer was used for ttpper tro.posphere temperature analy-
sisJ the height.and temperature of the tropopause were then examined and
the temperature recorded at the 15-km level was taken as the indicator
of temperature conditions prevailing in the lower stratosphere. The l5-km
level corresponded approximately to the maximal ascent of radiosondes dur-
ing the given months.'
- 25 -
> .
-------
Fig. 3
.\
,.
o
. -. -/.:
, i
--------~--- -------
, .
I
I"
"j" :..:
.' .
I
Fig. 4
~--------<-..-_~-:---- ----- - ~ _.. -...,----' --.- .
- -_.- --- __".0 -.- -.
"- I'
". ..,
. '.
o
I,~"'M
-/,t
. I:
III
IJ,/Jt ..,,.
o
-0.02..,1/1
'-
.- .
,u
\.. - .
"-----'r,.
I
I
'il
I!'
I
f
2 J "5 G ? , 9 /811/2/.1 '" /,1/&/7" II zg Z11UJlftSlI /lNII
_.u__~_'2 '!..":.../51617181ItDJ/221UtZJll ANII
H DEVIATI pNSFROM AVERAGE -OALU;---'---
500-300 MBIHTENP AND T'SKM DURING APRIL
'I t958 AT HEI~ Is. .
Comparative results are listed [in
Figs. 3,4, and 5, which include deviations
from mean monthly values in total ozone
content ( Ax. ), thicknes s of the SOp -300 mb
, layer (MIS00-300 ), height and temperature'
of the tropopause (AH and N ), and
temperature at 15 km tp ( AT1S) tp for ~ :
Heiss Is., for April and July, and foI:' SP-6
drifting station during April, 1956 (the SP-6
data for July are. insufficient). The coeffi-
cients of correlation between the total ozone
(x), temperatu~e at the 6-kmer~),9-k~ (Ts),
" 4H(S8I-JII) and l5-km (T1s)1eve1s, tropopause height -
f (H), are shown in Table 2. Data in Table
. ):. '. .2. tP., indicate that the type of gorrelation
I/I/j:Gll/8191Dtll1ZJ1tlSlll'l- . factors in April was the sp,me for both sta-
---lfE\lIATIOii'sFRoM AVE~A~~ x VA;UES . --tions. The coefficients of correlation be - '
H500-3OO I1B: HTEMP TTEHP AID TI5 FOR APRIL tween ozone and temperature at the 6-km
U 1958 AT SP006 . level r (x, Ta km) had large negative values
I , --------- --
:', ' - 26 '-
j
I '.
..A~---.- -- +--- -. -_:... __;m ----,
~'II , '
4' 15-
IF'
, 19. 5
-- ,__-_"m -
: i
" ,,",
. "
. --~--------- ~--'--- ------_._--- --_...:. --_._--,._u_._..__...._--_.~-_.---
DEVIATION FROM AVERAGE X VALUES
H~o-300 MB HTEMP TTEMP AND, TI5 KM'DURING JU'lY
OF 1958 AT HEISS Is.
4X
---'
.,...."
-------
I ..
/
which decreased toward ~e' 9-km leve1.,and at the 15 km level became and
remained negative and small. ' Coefficients of correlation between
total ozone and the height of the tropopause r(x, H ) were negative d\.1ri~g
spring andsum~r, and hi J:uly rose to 0.87 on tp Heiss Is. It was ob-
served on Heiss 18. 'that'the coefficientS cot correlation between the 'oi)ooe
an4 all 'elements had larger values in July than in Apr~l. and between the
9 a'nd 15-km levels the sign ,0£ the, correlation coefficient became reverl5~d..
, I, ,
An increase in the correlation coefficient related to the ,15-km layer d\.1r - ,
ing summer as compared With spring must be explained by a substantial
".. ,--_._.~~cr~~~.:_~,~~~~r r~di~~ion in the polar region during S\.1mmer. ,--______n___-
I
Table 2 In order to account for the effect of
I
CORRELATION COEFFICIENT: synoptic conditions gro\.1nd-surface weather'
I maps and baric topog' raphy maps AT 50'0: and "
X. T.. HTp. T.. Tn KM &
, f:ATaoo were analyzed. Temperature adv'ection --
I' HEISS 18.; aJ-6 , >n the troposphere and stratosphere were evalu-
APRIL! APRIL' : ated qualitatively, and shifts and oscillations
. rex. T.)=O.49 r(X. T.)=-;J 61 i. b i f ti i t iti . ti t d'
,(;C. T,)=-O.07 ,(X. T.)=-O:OI i in ar c orma on n ens es were lnv~s ga e. ,
,,(x. HTp)=-O.47 ,r(X.HTp)='-O.70! Apparently, atmospheric processes in; Arct~c re --,--
r(X. TI6)=-O.19 r(X.TI6)=-O 23
" JULY _,I ,;' I . 'gions adjacent to the North Atlantic and Pacific
I . ' ,Oceans, wer, of greater intensities and varia-
fiX. T.)=~.82 'bility than in the Central Arctic. This c0\.11d be
,(X. T,)=-O.72 " '"
,(X. HTp)=...:....o.87 demonstrated by comparmg synoptic conditions
rex. Tu)=+O.82 . prevailing in April on Heiss Is. and SP-6 , '
(Figs. 3 and 5). It is evident that during April conditions in Heiss Is.' ,'----
region were characterized by frequent changes in the baric formations--
sign, and by a considerable troposphere heat advection. Thus, on April
22, the earth's air temperature increased by 10°' during a 12 hour period
and, continued t,o increase during days following. Moreover', total ozone
decreased by 9% remaining stationary d'uring the ensuing days. The rela-
tive "activity" of' Heiss Is. in April was further confirmed by the wide,
range of tropopause heig~t and temperature variations .6H ... 4.6 km and
AT' ... 140 . ' tp
, tp .
At the same time, drifting stationSP-6'located in a more stationary
cyclonic field, tracked up to the 200-millibar surface, with negligible vari-
ations in Hand T (.6H -0.8 km and' & -S°,) and a practicaily constant
tp tp' tp tp
te~p.erature at 15 km. Total ozone values on SP-6
were highly stable in April and its deviations from the mean monthly value
fell, generally within the range of experimental error.
'CONCLUSIONS
1. Sharper seasonal variations in,tota1 ozone had been observed
in the Arctic than at the intermediate latitudes.
- 21 - .
,
'j'
1
-------
I.
I,
i: ,2.' Latitudinal ozone variations were also clearly in ~vidence at
, least d;p to 80o'N; more ozone was found on SP-6 and Heiss Is.. than on
Dixon Is.' and at Voiekovo.,
II
I'
"
, : '3. Contrary to results obtained by Johansen for Tromso, coeffi-
cients :of correlation between C)zone and temperature during April we re
small and remained negative up to the l5-km.level; in July they rose, and
I '
the coefficient of correlation between ozone and temperature at the 15 km
i' '
, level became positive '.
!
~-- .~ ~ .' , -..--"
lHBLlocfRAPHY
, "
~ -.- - ~----
---- ,---_.~- ---.----.-----...-------.-.---.-----. --- . --- .
-------.-- ---
n
I .,'
ii
I Ii
"
I i !:
i I '
I ,: ,I
I "
'1, 11 06 c 0 H , I) P 10 e p. K 8 a A JI 0 H r. MeTe.Jpo.'Y.IrHII, HKIKKHX CJIOeB,
aTNOctflepbl. YcnexH cpH311'IeCKKX HaYK. 31.,Nt I. 1917. . ' ,
2. n po K 0 cp b e 8 a H. A. AnaoccpepHIoIH 03OH. H]~.BO AH CCCP. 1951.
3. X ,8 0 C T II K 0 8 H.A. OJ:)~ 8 ar...o;~~pe.. Ycn~XH cjI,UHQeCKRX HaYK.
49. Nt 2. 1956.' '
4. .K Y 3 H e U 08 r. H. npo6.nelola 03OHa. MeTeOpo.l0rMII H rK;1po.11orHII. Nt 4. '
1959. ;.
5. J 0 h a n sen. Variations in the total ,amount of OZOile over TromsO rnd
their correlation witli other meteorological elements. Geofys. Publ.. XIX,
No.5; 1955. I, .
6. N 9 r man d C h. Atmospheric ozone and upper air conditions. Quart.
, . Jo~.~.~.~,.~~.m I~.
7. M ~ e t ~ a m. The correlation of the arpount of ozone with other characte-
I ristics:of the atmosphere. Quart. Journf Roy. Met. Sac.. 63. No. 271. 1937.
I
I
I'!
I '
, '
, I
!
I :
i .1
; I
!
!
I
I :
i
i
TIME-DEPENDENT VARIATIONS IN TOTAL ATMOSPHERIC OZONE
OVER DIXON IS. AND ITS CORRELATION WITH
. I
METEOROLOGICAL ELEMENTS
"
"
;,
T. S. Go1'm
I
i ' i
, :This article ;presents a sh~rt prlliminary analysis of data collected
1 ! I i
from observa,tions of total atmospheric lozone at Dixon Is. observato~y dur-:
'ing 195~! - 1959 and compared with some; aerologica1 data.
: Ii, , ,
1. : INSTRUMENTS, METHODS. A~D EXPERIMENTAL ERRORS
':, j' i
, I '
~otal atmospheric ozone observations began at Dixon Is. Observa-
tory Ju\y 1, '1957and were made at first: with t~e aid of ozonograph OFET-3
propos~
-------
I
sodium tetraborate. Spectral transmissivity of the instrument gradually
declined and in the spring of 1958 was reduced practically to zero. This
forced the preslent author to give up the analysis and: publication of data,
secured with aid of the instrument. Accordingly, it should be borne in mind.
that, systematic observations of total ozone in the aqnosphe're began in
April 1958, when spectrophotometer SFD-l eqUipped .,',n,th a quartz optical
syst~.~ ~a~ _d~li~!.red to the Dixon Is. Observatory. It should be noted alsl)
.that' total ozone deter.minations in the atmosphere were made at the Dixon
Is. Observatory using direct solar light or full moon lunar light.
Direct solar light was beamed at a monochromator by means of a
heliostat equipped with a trihedral quartz prism rotating mirror, and a
lens which focused the light beam on the monochromator input slit. Opti-
cal system of the instrument is schematically illustrated in Fig. 1. The
parallel light beam was separated in the monochromator by differential
grating replica having 600 lines per mm, a ruled 60 x 60 mm surface, and
maximum energy concentration in the first order spectrum.
-----_._------------~----- -- -- -..
Fig. 1
s
-- ---- ----- ----
- -.------.---. --. ... --.
OPTICAL PLAN oFSFD-1
I-QUARTZ PLATELET; 2-INPUT SLIT; 3-REPLICA OF DIFFRACTION
SCREEN; 4-EXTRA-AXIAL OBJECTIVE PARABOLA; 5-0UTGOING SLIT;
6-lENS
The monochromator operated within the 2200 - 11,000 A range. Disper-
I .
sion was constant throughout the spectrum and was equal to 32 A/mm. The
light beam was filtered at the monochromator output by passing through a
.
combination of UFS-2-ZhS-3 glass filters adjusted to the 3100-3400 A re-
gion. The monochromatic beam was then passed through a Sb-Cs photo-
element having a thin ultraviolet-transmitting glass window. The instru-
mentis photoelectric setup is schematically illustrated in Fig. 2; the in-
strument operates as follows: photocurrent passes through a 2.2 kilome-
ga~hm resistance connecting the grid and the cathode of the first amplifier
tube. The amplified voltage drop across the resistance is fed to a meter-
i,ng device operated as -~ compensating network. The sp~.c~rophotometer .
potentiometer 'circuit was d-esigned to record only the intensity ratios be-
tween two light beams. This is somewhat disadvantageous. since the in-
strument can record direct data on so~ar radiation intensity in the ultra-
yiolet range. A properly trained operatdr can make a single instrumental
reading in not more than 15 -20 sec.
..~
- - --- . --_.~
- 29 -
-------
i Fig. 2
- .. ~ -------_.._--~- .'<-"'.
-.. -----._---- - -~-~-_._-----_.. -- _.---- - ---..- . ...
...--...---
r-I--------------- ----~=--T~iE'-l
I . I
I I
I =J, I
I I
I I
- --1
I
I
I.
I
I
I
I
I
I
I
I
I
---____~_l
--------_.
.--.------..
i:1'
PHOTGELECTRICAL PLAN OF SFo.r
~.'
i Obs.ervations were made with a 0.3-0. 8-mm wide monochromator
output slit, i. e., the spectral light beam interval did not exceed 25 A. .
The monochromator scale readings or the instrument calibration were
checked regularly every 7-10 days using a mercury-helium arc spectrum.
It is significant that the scale correction was totally independent of the'
instrutr;lent temperature, although it varied within a 4-5 A range. Varia-
tions wbre accounted for while making the determinations~ . The functiorial
relatio~ship between correction and the' instrument temperature for 1959 .
, '.
is illustrated in Fig. 3. The mean correction value displacement from
. ! . .
20 to 6;0 A, occurred during a small ;mechanical instrument vibtation.
.,./ . . , . ,
Fig. 3. Total atmospheric ozone deter'-
"
" minations were made using SFp-l, and
-.--.------_! C9R;;CTI-;»I-~-"---~' selecting two pairs of wavelengths most
6 A I suitable for operations under Arctic con-
il . . t{l . . . ditions. The first pair was A1 = 3114 A
,i:~ and A2 = 3324 1;1 ~he second pair was
. .11' 18 .. A1 = 3088 A and 'A~. = 3350 A. The first
. '; .. 20 .... pair of wavelengths was used at solar
i '11 elevation less than 30°, the second - at,
. . ~JO..ii Ii.?" '/8.' ". !8 JI t' Pe from 23 to 40°. Calculations were'
~-.-i~-; .;..._.~:~~.~~/...:...._. made according to a knQvm formula for two
I-CORRECTION FROM 19/VI TO I~IX 19:1}; wavelengths. Ozone absorption coeffi-
a.CORRE~r ON FRO 22/111 TO '~VI 19:1} dents we:re taken from Vigroux's recent
.j
,
- 30 -
:I.
:1
,
.1
.!
-------
i'
10
---- - - -----
data for temperature of -440 . 0 Lo= Ig (~!-) for each formula was deter-
mined graphically from the L (J.L}curve°s-, 0 where L = 19 ~) based on
a two-year observation series. The observations were I made on
o days with prevailing constant meteorological conditions, checked by analy-
sis of synoptic maps. Lo value for the first pair of wavelengths was taken
as the mean of 56 curves, and for the second pair of wavelengths of 25
curves. The difference between data obtained by using both wavelengths
in concurrent observations lies within the range of experimental error.
Generally speaking, the present author encountered difficulties in deter-
~ining the experimental error due to lack of data concerning the error
introduced by the spectrophotometer circuit, and to being unable to deteOr-
mine it under conditions prevailing on Dixon Is. However, calculations
had shown that the relative error was approximately 6% and depended
almost entirely on the accuracy of the readings. This relative error can
be reduced to 2 -3% by properly-trained operators.
I 0 The instrument rating plate indicated that the scattered light in
the instrument did not exceed 1% when measurements were taken in the
2:200-2400 A and 10,000-11,000 A regions; it was negligible in other regions.
~his was confirmed by the fact that total ozone remained constant when
measured up to he = 10-110 , whereas, elementary reasons postulate that
I
t~e presence of large scattered light quantities in an instrument would lead
tb an improbable overestimate of the total ozone in the atmosphere for
100 1
hG = 15. Comparison of SFD- with a standard Dobson spectrophoto-
meter was conducted over 20 spring days in 1958 at Voeikovo, which
yielded satisfact ory results. Data yielded by concurrent observations
using both instruments were nearly identical. 0 0
I Direct lunar light observations were ~made us~ng -instr-umenfOF£; -3.
However, due to many reasons data obtained during winter months could
not be used in the analysis. It should be added that instrument OFET-3
had some structural characteristics which made it unsuitable for use under
Arctic conditions.
2.
TIME-DEPENDENT VARIATION IN TOTAL OZONE
CONTENT OBSERVED AT DIXON IS.
The total number of atmospheric ozone content observations is re~
corded in Table 1. Upper figures in the top section represent the number
of days, lower figures in the same section represent number of observa-
tions.
Fig. 4 shows a smoothed-out curve of annual total ozone variations
in the atmosphere. The solid curve represents - mean data collected over
two years observation, dotted parts of the curve represent results of one
year observation. Variations ~n total ozone from April to October were
identical for both years. ..-----
- 31 -
-------
i
'Table :1
!
i
I,"
I
'" j
,
i
"
. -.. - .--- - - .-- -----..-------.- -. --- ---
---- -- -----.--
,-
-
-----:-~..;- -;--.~ - .---
MONTH FOR
. 'VE'Air I II III IV V VI VII VIII IX X XI XII THE
YEAR
--...
No. Of"
DAYS' - - - 7 10 6 14 7 - 5 7 8 64
195L-~
No. OF
CASES - - - - 151 173 142 298 76 - 22~ 447 214 1729
N~ 'j
DAYS 2 20 18 14 13 10 13 2 9S
3 - - -
19~-
No. OF 29 557 1511 481 260 133 150 30 2219
CASES 68 - - -
i
Fig. 41 '
-'X-fO-:J;;'---~--;------------------------~~--------
It should be noted that
according to the ozone curve
the ozone minimum occu~red not
in the fall, buta.t the end of July
followed by another smaller, min-
imum in December. This phe-
nomenon may ha've been asso-
ciated with polar night. Maxi-
o .-- ,.
mum atmosi:.heric ozone was
. ,
noted during March. Ozone
I '
,concentration varied sha.r'ply
200 -~_:-----~-----------,--- -------a:-------r--------- from day to da'y during April :
> W = = M
:;; ~ I- ~ ::; ~ ~ and ay at times to the extent
<=> (/)WIDEE at '
~ ~ ~ Ii: ~ ~ ~ of 25-30/0 of mean ozone con-
e &LI 0 &a.I (3 0 &LI . ,
~ l1.. ...: II) 0 Z Q centration in the atmosphere;
moreorer, totaL at~ospheric ozone remained relatively high. A rapid 'drop
in atmospheric ozone concentration occurred at the end of May; and ampli-
tudes 6f ~bove mentioned variations became sharply reduced. 'I ' .
II Daily, or more ex~ctly diurnal, variations in total atmosphe~ic '
ozone exhibited the following peculiarity: in most instances atmosphe~ic '
ozone honcentration 'ro~e ~lightly between 10-12 o'clock local time (6~8 .
, o'cloc~ Moscow time) and remained approximately constant during the rest
'of the aa:y. Curves in Fig., 5 illustrate the cause of diu~nal ozone concen-:
tratiod variations 'during June 22, 1959 and July 18, 1959. . \
I I ' , ,
I' ' . , ' , ,i ,I
'I . ' I ,
:i N~w, consider in greater detail the rapid short p~riod of vari~tiOIis
iri"tota,i,atmospheric ozone during the. day:. Analysis of data col1ected:dur-
ing bdght days of summer 1959- showed that, generally, the amplitud'e of
sudde~, varia.tions changed daily from 1. 5 to 9% of the mean d~urnal oz~ne
,conceritration, and that these variations were closely related to baric,
condit~ons prevailin,g at Dixon Is. Analysis of the latter were based dn ~ap
At-30d. Maximumamplit~del variat,ions-amounting to 5-9,% of the mean
I
i
I ,"~ "
!
! '
I
I\..
i/ '\
" 1'\
,I
, ! /" --- "' ,
,,' \.. '\
, V '\
I
!
500
i 1100
300
:r ...
u - I> UJ >
a:: a:: z ...
< "- ,co: => =>
::E ...:, ,r. "") -;)
-1
, I
, I
I', '
,-1
- ,: ':1
i;
~ ! '
';.1
I .
--=-32-
I
I '
I
-------
diurnal level were observed in the anticyclone center and in cyclonic troughs,
1. e. in areas where aerial currents of different origins alternated. In a
diffused baric field and on the outer periphery of baric systems, under con-
ditions of slightly perturbed atmosphere, rapid variations never exceeded
2-3%, 1. e., they remained within the range" of experimental accuracy. The
variations were not associated with the time of the year, and were indepen-
dent of atmospheric total ozone. This is illustrated in Table 2, in which
data for June 20-22, 1959 are. of particular interest; on June 20 and 21 an
anticyclone center 'was observed over Dixon Is. which was moving gradual-
ly in the north-westerly direction at the 300 mb surface level. On June 22
the anticyclone passed beyond Dixon Is. Short period ozone concentration
fluctuations amounted to 9.6% on june 20, dropped to 5.6% on June 21, and
completely abated on June 22. Total ozone decreased accordingly from
0.324 cm on June 20 to 0.284 cm on June 22, which enhanced the probabil-
ity of Reed's hypothesis concerning the effect of descending currents in the
lower stratosphere on total atmospheric ozone increase (13).
---.-------- --------
- ---,.-.-----.--------------- .
3.
RELATIONSHIP BETWEEN TOTAL ATMOSPHERIC OZONE VARIATIONS
AND BASIC UPPER TROPOSPHERE AND LOWER STRA TOSPHERE
METEOROLOGICAL ELEMENTS
In calculating coefficients of correlation between total atmospheric
ozone and variations in meteorological elements at different stratosphere
levels this author used data obtained by aerological atmosphere sounding on
Dixon Is. from April to September, 1958-1959. Similar to the case of ozon",
data, aerological tables were examined in detail and questionable data wert:'
excluded. The following parameters were selected for comparative pur-
poses: tropopause height and temperature and temperature and pressure
at 6, 7, 10, 12, and 14 kn.
Fig. 5
--- _.~-----_.._--'------- -~_.._-----
X"/O"JCM
350
.300
250
22. VI. Sge.
o
300
250
200
18. VIl59z
una'" -
-.., ~.
I 2 J 4 5 6 7 8 9 to!!__iJ.!! i~ Iff iG. ;7 ~~---
MOSCOW TIME IN HOURS
- ._.._--- "-.-
- 33 -
-------
Table 2
----------.---
..--.. .'-.
DATA
-------
TOTAL' !
OZurfE '
AV~- Af.tP':~- L- -------
OF RAPIO 'I SVqOPTIC SU.RFACE CONDITIONS
FLUCT'S ' AT THE 300 MB LEVEL
I
4jIV-1959 r.
I7fIV-1959 r.
2O!IV-1959 r.
41V-1959 r.'
20tVI-1959 r.
2iJV1-1959 r.
22!VI-1959 r. I
27jVI-1959 r.
2jVII-f959 r.
18/VII-1959 r. I
31jVII-1959 r.
. 2jVIII-1959 r I
10jVIII-1959 r.
I4/IX-:!9:;g r. I
0.392
0,491
0,503
0,385
0,318
0.313
0,286 i
0.313 !
0.322 :
0,240 I[
0,277
0,267'
0,275
'0,288
- -----
..------
2,1
2,5
5,7
2.1
9.6
5.3 -
2.0
3,7
8,5
3,1
5.1
4,3
1,5
PEAK AC
DIFFUSED BARIC ZONE
FRONTAL PART C
DIF"USED BARIC ZONE
CENTER AC
CENTER AC
CENTRAL SECTION AC
DIFFUSED BARIC ZONE
TROUGH C
DIFFUSED PEAK AC
CENTRAL SECTION OF
ST AT! ONARY AG
SOUTH-EASTERN PER I PHERY C
EXTREME WESTER N
PER I PHER)',
DIFFUSED BARIC ZONE
2,9
,
The mean tropopause height dur~ng the considered period was 9.3 km.,
Data of 5-day periods pertaining to ozone and the remaining elements were
I
average. The number of sunny days on Dixon Is. was regretfully small,
limiting the averaging period to!1-2 days. In such instances aerological
sounding data were taken only on days when ozone had been observed. All
, ! I '
5 day average data were collected into three groups of two months each,
, and correlation coefficients werle computed for each group individually.
Grouping of the ex~sting data ap:peared natural since each group differed
characteristically from the otheir groups with respect to total atmosphe ric
ozone variations. High total ozone and sudden daily fluctuations were tYFi-
c~l of the first group (April - May); less pronounced fluctuations and sharp
reduction in total ozone characterized the second g~oup (June-July), and a
moderate incr,ease in total atmospheric ozone characterized the third
group (August-September). Correlation coefficients were computed from
5-day mean value's of deviation from the smoothed adjusted mean ann~al
curve for each 7lement. Results thus obtained are listed in Table 3. ,D.ata
in the Table show that correlation coefficients varied substantially from'
those obtained by Johansen for Tromso (9) and Meetham for Oxford (11). .
It should be, not~d first that highest values for coefficients of correlat~on
between ozone and temperature and pressure occurred not during th~ "
spring but during the summer months. Low correlation coefficients oc-
curred in April and May, which: completely abated in August-September.
It is worth noti~g that the'negatlve correlatiO"n coefficient with temperature
. ".. '.
rose during April -: May with th~ altitude, the highest value occurring at
Mkm. ' ' .
..
The correlation coefficient for ozone concentration and pressure
was hi~he_st at an altitude of 12 km. Even at that height, it was only 0.47.
-! '
- 34 -
-------
I
According to Johansen's data (9L coefficients of correlation between ozone
concentration and tropopause altitude and ozone concentration, 'temperatui-e,
and pressure at altitudes of 6 and 12 km averaged 0.56 (on the absolute
scale); the temperature correlation was negative in the upper tropospher'€:
. and positive in the lower strat0!5phe!e; the coefficient was negative for the
remaining elements. Data 'of the second gz:oup agreed satisfactorily with .
the data for TromsOI however, the present author found a low corre-1atio~
. coefficient between total ozone and 'pressu~e at the 6. and 8-'km levels.
Table 3
. -- -------. -----------_._-----_._- ----
- --- -- -----"- -----
_..- - - ~
- ---_.- --..-
MONTHS APRIL -MAY JUNE-JULY AUGUST~SEPTEMBER
COR- I ERROR ,OOMPARAT. ~ L-ERR--i COMPARATo, COR- ~--' CO-
CORREl'N RECT, REcr '.: RROR . REC ERROR OMPA~A
ELEMENTS C CORRECT'H/ REGR. COEF. ,CORRECT'~ REGR. I C T CORRECT 'N REGR.
OEF. OEF. '
hr -0.,14 :1:0.,16 ~hr=-Q,2Q Ax -0.,58 :1:0.,10. ~hr=-O,~Q ~x I -0,09 - -
Ax=-Q,Q9 ~hr ~x=-Q.23 6hr
Tr -0.,21 :1:0.,15 .1x=-D,32 A Tr +0.,43 :1:0.,13 dr=+D,44'\ Tr -
ATr=-Q,1I6x 6T=+Q,41 Ax
T, -0.,36 :1:0.,14 ~x=-o, 72 6 T -0,56 :i:Q,1O ~x=-Q,55 6 T -0.,0.7 - -
AT=-Q,186x ~ T=-Q.64 ~x
T, -0.,23 :1:0.,15 Ax=-Q,45 6 T -0,46 :tQ,1I I\xo-=.-O ,49 6 T - - -
AT=-<',II Ax .:\ T=-Q,43 ~x
T10 -0.,28 :1:0.,15 ~x=-Q,34 A T +0.,17 :1:0.,15 ~x=+Q,21 ~ T - - -
AT=-Q,226x ~T=+O,14 .6x
1'.. -0,36 :1:0.,14 ~x=-O,49 hT +0.,56 :1:0.,10 hx=+Q,43 ~T +0.,11
~ T=-Q,25 ~x A T=+Q,73 /:1%
T,. -0,42 :1:0.,13 [.x=-Q,46 AT. ..j-Q,47 :1:0.,11 6x=+Q,60 6 T - - -
AT=-o,37 Ax AT=+Q,41 Ax
I' -0,0.7 - - -0,39 :tQ,13 Ax=-o,22 t'1P I
'
AP=-O,67 Ax +0.,06 .-' -
P, -0,15 :1:0.,16 Ax=-Q,II ~p -0,39 :1:0.,13 Ax=-Q,23 t'1P - - -
AP=-Q,17 .b t'1P=-Q,67 6x
['a. -0,20. :tG,IS Ax=-o,26 foP -:0.,50 :tG,l1 Ax=-Q,37 I\P - - -
AP--o,17 Ax AP=-G.G7 fu
. PII -0,47' :tG,12 Ax=-o,96 AP -0,53 :1:0.,10. Ax=-o,7Q AP +0.,0.1 - -
AP--o.1G Ax t'1P=-{),4G t'1x
P" -o,OQ .- -0,40. :to,13 Ax=-o,fi3 AP - - -
.. Af'.-.-Q ,16 AJr
- '
.
T.
Mean 2-year variations in total ozone during April to September
are presented graphically in Fig. 6 alongside the 5 -day data and the cor-
responding mean variations in tropopause height. The graph indicates
unmistakably that, generally, variations in total ozone and tropopause
altitudes were of an inversely proportional. character. This indicated
that total ozone decreased during the period under consideration while the
tropopause altitude rose, although individual fluctuations in the variations
of both elements approximately coincided 'on1y during June and July.
Undoubtedly, the number of observations on total atmospheric ozone
made over Dix(;m Is. thus far is not as great as might be desired. Never-,
theless, it is evident that disagreements between the present author's data'
and the data obtained at Tromso and Oxford, can not be explained on the
_. -..------
- 35 -
-------
basis. of obse~v~tion volume.' I.t is known that in computing values of cor-
relatIon coeffIcIents for total ozone and variations in elements for 0 fo d
M. () x r,
eetham 1 had only 31 cases of simultaneous observations during April.
Nevertheless, his resul,ts agreed well with those obtained by Johansen.
Fig. 6
An attempt was made by
the present author to find an ex- .
planation for the disagreement be-
"lit. tween his data and those of Johan-
r - . I
. sen by making a synoptic analysis.
- If the Dixon Is. . characte ristic .
I
,synoptic conditions were consider-
ed broadly, then it can be stated
. .firstly that circulation over Dixon
.Is. had the explicit characteristics
of a monsoon; secondly that no regu-
7iPRTL MAY JUNE JULY AUGUST SEPTEMBER lar ground or altitude 5-day pres-
--'----~7-----C---_. _._._-----==-~-_-:::-:,,-=-'-zl_-__-- sure wave alterationswereobser-
, - CURVE 0;= OZONE CONTENT 2 - CURVE OF TROPOPAUSE ved at Dixon Is. similar to. those
Al TI TUDE. noted by authors (9, 10 and 11) in
Western Europe. Apparently, this is related to the fact that cyclones pass-
ing over the south-eastern region of Kara Sea generally were old abatib.g
baric meteorological formations occluded by secondary front chains, while
cyclones which passed over Western Europe were young cyclones whifh
followed one another with great regularity. A more detailed analysis bf .
synoptic conditions as compared with fluctuations in total atmosphe.rip lozone,
calls for additional ozone concentration observations during cloudy day~, or
prolonged observations of direct solar radiation. Limited ozone datJ pre.~
vented this. author from making a more detailed study of this probleml and, \
in particular, from examining. the tropopause fluctuations in re lation to jvari-
ations in atmospheric ozone concentrations. I' ;1
: ,In conclusion, let us pause over the following interesting proDl~!m;
authors who investigated the relationship between total atmospheric oizQne
and synoptic conditions noted that an increase in total atmospheric ozdn~
was normally associated with .cold air invasions from the polar region,s;1 and,-
conversely, intrusion of air from the South caused a decrease in total bzone.
In conducting the synoptic analysis the present authors devoted consider~ble .
attention to this problem. Wh~n total .ozone variations were compared:/with
nightly direction of winds at heights of 6 to 18 km, and, particularly, at .
11 km during April and May 1958 and 1959, it was brought out that increase
in total.atmospheric ozone over DiXQn Is. occurred predominantly during
southerly winds, and that invasion of/air from the Central Polar Basin had
little effect on atmospheric ozone copcentration. For se'verai reasons and,
particularly,' due to limited data, it was not possible to conduct quantitative
analyses and to pr~sent corresponding correlation coefficients, as was done
by Tonsberg and Olsen for Tromso (12).
- -
- 36 -
X 10'3(10/
"(1,1.): -0.(11
"(',/1)' - OJlJ
r(. ",'-0.81
500
I I I I r h
! lit' I ((
I)
- .J t.' 10
..It .1 , {! \
g.
I
.. if I B
-.
400
JOO
200
-------
CONCLUSIONS
Brief analysis of data on total atmospheric ozone was performed in
the Dixon Is. Observatory with the following results:
1. Annual variation in total atmospheric ozone over Dixon Is. ac-
c,orded generally with the mean annual variations for high latitudes (maxi-
mum-during spring, minimum-during autumn).
f Z. In most cases, an appreciable increase. in total atmospheric
ozone had been observed during noon hours, although, at times, total ozone
vlariations persbted during the day.
I 3. Sudden, short-period fluctuations in total ozone were related
directly to baric conditions over the observation point; sharpest fluctua-
t~ons occurred when baric centers passed over Dixon Is. (at the 300 mb
s~rfa.celevel) and disappeared in the presence 'of a weak baric field in the
atmos phe re.
4. Correlation of total atmospheric ozone with tropopause height,
tempe rature, and pressure in the upper troposphere and lower stratosphere
was most distinct (r = ::1:0.56) during Summer months, in June -July, and ve ry
weak during 5 pring and autumn.
5. No relationship was established between annual variations in
total atmospheric ozone and changes in basic types of atmospheric circu-
lation, previously reported by G. Ya. Vangenheim.
BIBLIOGRAPHY
-..--.----- ._-
.- -- -- -- .--- --------_.- -------.-
I. n po J( 0 41 b e B a J.i. A. ATMOCtpepHblii 03OH. H3;1.-BO AH C..ccP. 1951.
2. M H T pac:. K. BepxHRR aTMoctpepa. rJl. 4. c03OHoctpepa', Hn. M.. 1955.
3. ) Be p e B A. C. CIIIIOnTII'IecKaR MeTeopoJlOrHR. rHApoMeTH3AaT. 1957.
4. r v m H H r. n. K BOnpocy 06 \l3MepeHIIII oOmero cOAep>KallllR 030IIa
Ii ero BepTIIKaJlbHOrO pacnpe.:leJJeIlIlR. ABropetPepaT AllccepTalllUI. rro
11M. BoeiiKoBa.
5 5 e 3 B e p x II II ii lll. A. 0 HeKoropblx OC06eHIIOCTIIX y.~bTpa
-------
- 'f
i .'
MOSCOW VERTICAL OZONE DISTRIBUTION OBSERVATIONS
I .
!
A. P. Kuznetsov, V. A. Iozenas, A. S. Britaev
I
I' . . .
I Information on vertical ozone distributi.on has become significant
and va:~uable. The wid~ly used method for collecting such information.
depen1s upon the study of dispersed ultraviolet zenith light during cloud-
less drYs. Such observations are usually limited to relative quantitative
deter'finations, specifically to intensity ratios of dispersed ultraviolet
light of two wavelengths, one of which is absorbed strongly and the other
I .
only wleakly by atmospheric ozone. A spectrophotometer which recorded
such i~tensity ratios was constructed first by Dobson (1), and the, method
I
for de~ermining vertical ozone distribution in the earth's atmosphere,.
based Ion these measurements, was described by Goetz. The principle of
Goetz's method rests on the "rotation effect" he discovered which caused
I' . ,
:qim toll postulate tha;t as the sun's zenith distance increasedt ratios of I
two wavelength intensities decreased to a minimal value of ,approximately
a . ! ~
C = 85 i' after which the ratios began to increase again. ' I : :
I ' '. I '
i During the 1959 IGY conference a discussion was rafsed on the sub-
ject of the physical nature of the rotation effect which revealed that the,
final sp1ution tol this' problem was yet to be found. However; the present
auth.ors think th~t independently of the physical nature of the rotation ef-
fect, tp,e method for analyzing vertiCal ozone distribution proposed bYI : .
Goetz and Dobson had not lost its value. Despite its apparent crudeness,
. . I
this method yields sufficiently accurate results which can be confirm~d
'11 .'
by dir~ct observations with the aid of rockets and large balloons. DUr to
its adaptivity, and in the absence of other simple methods, Goetz's method
found ~de application abroad. Goetz, Dobson, an'd Meetham used this
methoq. ~or many years in measuring vertical ozone distribution, and ,so:
did ma~y authors since. Using the principle of Dobson's photoe1ectri~ '
spectrpphotometer with certain modifications, the following spectrophoto-
meter il was built at th~ Moscow State University (MGU). The double quartz
monochronometer, used in the instrument, was manufactured in the Soviet
. .
Union.' The instrument was adapted to conducting regular ,ozone measure-
mentsibased on the absorption of the ultraviolet section of the solar spec,'
trum.1 . . 1
. I I
: I :
, The present ~uthor made observations at MGU in J...enin Hills (3)
in 1955-195.6. Beginning with 1957, observations have been made on the
grounds o{ the Central Aerologica1 Observatory .of the GUQMS at Dcilgp-
prudnd, 20 km from Moscow. . A relatively 1argenumber:of observations
have been made of dispersed light in the zenith 'of the cloudless sky, as
well as ot direct solar radiation measurements. Computations of vertical
ozone distribution were made up to September 1957 using methods proposed
by Goetz, Dobson; and Tonsberg. Data were processed by method "A": (4).
- 38 -
~ . ~ i '
-------
In the numerical solution of the problem the present authors used ozone
absorption coefficients derived by Ni Ten-Ze and Chung Shi-Piu, and air
dispersion coefficients derived with the aid of Ray1e~gh's formula (5); air
density values up to approximately 25 km were taken from the Central Aero-
logical Observatory files, and for greater heights, from standard atmos-
. pheric tables.
The atmosphere was divided, as usual, into five layers:
0-5, 5-20, 20-35, and 35-50 km and above 50 km. Moreover, it was'
assumed that total ozone and ozone concentration in the lower 5-km layer
was 10/0 per kilometer height, and that no ozone existed above the 50 km
level.. The following two assumptions were made hi solving the problem:
1) total ozone was constant during the day, and 2) only primary light dis-
persion was taken into consideration. The International Committee of a
Conference which met in Rome recommended that the E. K. Vigroux
ozone absorption coefficients should be used in computing atmospheric
ozone. Subsequently, the Committee advised all ozonometric stations
officially to use the new absorption coefficients in computing atmospheric
ozone beginning with 1957. Accordingly, the present authors began to use.
the new absorption coefficients from the beginning of the IG Y. In compari-
son with pre~<>.~~- coefficients, the new coefficients yielded an "-'30% in-
crease in :tot,ii ozone. In computing vertical ozone distribution the presen::
authors also used the new method of dividing the atmosphere into (Walton
(7) proposed layers shown in the following table.
.------ --- ---_._------- - ------....--.
-----<4-LTI TUDE BOUNDARY
LAVER NO.1 IN KH
o 54-00
1 48-54
42-18
36-42
24-36
12-24
0-12
--' . - .
2
3
4
ASSUMED OZONE VOL.
IN KH
o
0.057 %1
0,204 Xl %1
0,739 %1
.1,
%-.1'1-%'-"
U .
It has been assumed that the quantity of ozone in the atmosphere
up to 12 km above the earth's surface amounted to 8.5% of the ozone total,
and that it had a uniform vertical distribution. This assumption was based
on the mean results of chemical airborne determinations made by Key
in 1954. It has been assumed also that ozone density decreased exponen-
tially with height above 36 km. This exponential decrease is now.regarded
as an established characteristic of vertical ozone distribution which ac-
corded with data obtained by Johnson -=.! ~ in 1952 by means of a rocket-
, borne device. Total ozone has beenregarded as a fixed quantity, and it
, was assumed that above 54 km the atmosphere contained no ozone. .
The method of solving the basic integral equation for vertical
ozone concentrations at given altitudes was based on the substitution of
the integrals I summations of a small number of terms only. By this method
- 39 -
-'I
-------
,
I
. i
the integral equation could be replaced by a system of equations containing
two unknowns. Subsequently, the numerical values of ozone. concentrations
. in the Xl and X2 layers were derived by means of Newton's successive ap-
proximation method. Knowing Ozone concentrations in all layers made
pOHsible the determination of ozone concentration per 1 km in each atmos-
pheric layer.
. Computed ozone concentration data per I km in different atmos,-
pheric' layers during August 2, 1955 and April 5 and 16, 1958 are shown
in Figs. 1 and 2. (See pg. 41 for Fig~ z)
- .-
---------~~_._----.-------'--'-~--.----- .--.- - .--. ---..-
Fig. 1
50-
4S
I,.{)
%: 35
~.
:' SO
~' 25
::>.
1-,
j: 20
..
-< 15
10
5
'.
u
-O~OII~-'I~-I'O-3 '~M~
I .
I 2 J . 56 78.9 m U U M n a
OZONEiiSTRI1JUTION-AT ALTIT.UDES ACCORDI-NG TO -:OBSERVATION
g t ~~ _~lv- ~Jr~_~~~~C!~~~~~D ~!.~_!.~:~~~!~~;~~~s.~-~ ~~~T ~~~~ER~
i
I !
The following should be noted in conclusion:
I '
I
I !
. 1. Introduction of Vigroux's ozone absorption coefficients in-I i
creased the comp'uted ozone total value by approximately 33.3%. Thi,s:
indicated that ozone total in the atmosphere varied within 0.2-0.6 cm :1imits
without affecting adversely the relationship between total 9zone and meteoro-
logical elements. i
i I
2. In cO,mputing ozone concentrations in atmospheric layers as-
cendingly the nh-mber of cases in which solution of equations yielded i~k
definite results,:increased.' '. 11
I I
I
3. The present authors noted the altitude of atmospheric layers.
containing maxi,mal ozone concentrations was higher than reported in pre ~
! . . I
vious investiga~ion's; this may be due to the new arbitrary' dividing of ;t}:le
atmosphere into five layers. , I '
Refinement of the method for the determination of vertical ozone
- 40 -
-------
I
distribution. in the atmosphere remains an urgent problem.
Fig. ~----_.
5. '.
._---- -- --- --- --- - -. -
~
...
Q
. ~ 2"
to-
...
4:
12
'. !.
r;
o.OOJ
0.006
0.008
0.01: 0.015 0.0/8
. OzoNE-IN avKM'
511
1/8
--- 112
~ JG
z
...
Q
. ~ 211
to-
...
«
12
I. 1-.--'-
0.0/2 ---. o.O/~ .- _0.0:8
OZONE IN cM/KM
OZONE DISTRIBUTION AT ALTITUDES ACCORDING TO OBSERVATIONS MADE AT-
THE CENTRAL AEROLOGICAL OBSERVAJOQV TERRIJOjV I~ THE ~lIV OF DOL-
GOPRUDNVI 20 KM FROM Moscow ON {A} 5 AND {B 16/IV, 1958- A - X =
0.410; B - X = 0.425. COf1PUTED BV THE WALTON METHO~
/J.:'OJ
,
aUUG
I ,
0.0U9
o
BIBLIOGRAPHY
TDo b s:O'n~-M.-~.-Phys..-SOC:'-Proc.:-4i.324~193'I-.- .--- -- . H
2. Got z F. W. P., Me e t ham A. K. and Do b son G. M. B., Proc.
Roy. Soc. (A). 145. 416. 1934.
3. K y 3 H e u 08 A. n. IhBeCTHIi AH CCCP. cep. reocpH3.. JI& 9. 1154-1163.
.1957. .
4.. Ton s b erg E. and 0 I s c n K... Investigations on atmospheric ozone
. at NordIyobservatoriet..TromsO. Geophys. Publ.. Oslo. 13, No. 12, .,19.14.
5. n po K 0 II> b e.8 a H. A. ATMOCljJepllblil 030H. H3;J.-BO AH aIP. M.-:I..
1951.
6. Vi g r 0 u x E. K.. Annales de Physique. n~ 8. iOJ, 1953.
7. Wa I ton I. F. Ann. of the internat Geophys. Year. Pub!. by Pergamon
Press.. London - New York - Paris. 5 part., No.1. 1957.
- 41 -
-------
I ,
TOTAL OZONE FLUCTUATIONS IN ABASTUMANI
BETWEEN JULY 1957 AND JUNE 1959
Sh. M. Chkhaidze
Regular ozonometric observations have. been conducted by the
A?astu~ani Astrophysical Observatory of the Georgian SSR Academy-~f
Sciences at elevation h = 1600 m, beginning with the International Geo-
physical Year IGY since July 1, 1957. The instrument used is the OFET-3
three-channel photoelectric ozonometer built by the experimental produc-
tion shops of the A. A. Zhdanov Scientific Physical Research Iristitute,of
Leningrad University. The ozonometer was originally calibrated at the
GGO by comparing its response with Dobson's spectrophotometer; it was
field tested in July 1958 against ozonometer GGO No.3. On the basis of
these 'comparative tests working formulas were evolved for the d.et~r-
mina~on of total ozone from data obtained by the present author with the
aid ofl the ozonometer at his disposal; ,
! I
"
-------_._--~----_.----,.--- --- -~. -------'-'-- ---.
XI-2 = - 0.030- Ig (II/I.) - 0,205,
. 0,484",
X:L2 = -0.400-lg(/./I,) - 0,241
0,415",
for filters 1 and 2, and filters 3 and 4 respectively.
. Reg-ular observations' were made us~ng filters 2 and 3
were used as controls:~" -
.\
I .
while filters .1 ,and 2
I
Passbands of the filters are shown in -form of curve in Fig. l.
I
-_..- ----- --
- _..~. --- .' ._- -----
. --.------ ___n
Observations were conducted
in direct sunlight only, and since poor
weather conditions prevailed, the .num-
ber of observations made was limited.
In the two years from July 1957 to July
1959 there were 320 days during which
it was possible to make observa~ions.
However the 320 days were uniformly
distributed according to seasons,' so
that it was possible to estimate annual
I
total ozone variations in the_atmbs-
phere over Abastumani. On the basis
. of data ,collected during the two years',
a graph was plotted of the .annual total
ozone variations in the atmosp~ere over
Abast~mani, (See Fig. 2).
- 42 -
Fi g ~ 1
) . 20
~
z
. -
\L.
3500 3700
z
o
. !< 10 '
a::
to-
....
-
-'. .
".. ..
- -~ - --
-------
Fig. 2
-- ~ .--
0."00
z:
u
z;
- . MOO
UJ '
Z
o'
Ie;. -
; "200
"
\ I II OJ lY Y Yl W 'Illl ~-~~_l1!
. MONTHS
-------
ANNUAL CURVES OF TOTAL OZONE AOVER ABASTUMAN'
Curve in Fig. 2 shows mean
monthly values of total ozone. The
curves also show that maximum total
ozone was observed in the atmosphere
during spring (March - April) and a
minimum during autumn (September).
It was als 0 noted that the mean monthly
values for September through April fol-
lowed a sinusoidal curve, and that sub-
stantial deviations from this curve were
noted during May through August and a
secondary maximum in total ozone oc-
curred during June - August as shown
by the curve in Fig. 2.
The immediate reaction was to questio~ the values recorded through
August and to look for errors introduced by the instruments. However, by
comparing the data with corresponding data of other observation posts,
published in a monograph by R. M. Gudi (1), it became apparent that most
of the stations observed a similar effect; on the basis of the aforementioned
Gudi stated: "There are indications that a secondary maximum occurred
, in late summer". * Apparently, considerable disturbances developed in
the ozonosphere during the annual warm seasons which disrupted the annual
cycle of the atmospheric total ozone.
It has been known that the relationship between total ozone varia-
tions and temperature in the upper atmospheric layers was of particular
interest from the meteorological viewpoint. This relationship was investi-
gated by many authors, who obtained a satisfactory correlation between
these two quantities (2 and othersJ.-' In checking the same the present
authors used temperature data for altitudes from 6 to 16 km obtained in
the course of aerologic observations made over Tbilisi.
The straight-line distance between Tbilisi and Abastumani is 150 km;
it was assumed that at the altitudes of the two cities a distance of 150 km
would have a negligible effect on temperature. Correlation coefficients f0r --
the months from July through December 1957 were arrived at by computa-
tion. Results are listed in the Table ;'on'page- 44.
Data in the Table show that at heights up to 12 km negative correla-
tion coefficients predominated (83% of all cases), while at heights above
13 km, po~itive correlation coefficients -predomina~ed, though to a i~sser '
degree (57% of all ca:ses). However, not one of the correlation coefficients
under consideration could be regarded as reliable. Therefore, the calcula-
*G. I.. Kuznetsov established that the secondary summer maximum was ob-
'observed 'o~er Reykjavik, Ukkel, Voeikovo, and Alma - Ata (Ed.).
, - 43 -
-------
tion of correlation coefficients'.was discontinued.
~.._.. - --------..-__h - .. >- --- . - ..--. .._---_. - - - ~ - . - -. -
1957 VII I Vl/I I IX ./ X XI XII
6 1o:.v .-0,32 I -0,22 -0,31 -0,31 -0,42 -0,52
8 KM -0,23 ' -0,18 -0,27' -0,26 -0,37 -0,52
10 1o:M ~,48 -0,06 -0,33 -0,21 -0,65 -t 0,14.
II K.tI -0,43 -0,11 -0,35 +0,28 -0,79 +0,35
12 KM -0,13 -0,12 -0,14 . +0,12 -0,48 +0,43'
13 KM -0,20 +0,01 -0,19 +0,12 -0,29 +0.45
14 1o:M -'-0,16 +0,33 -0,52 +0.08 -0.27 +0.41
15 KM ...0,44 .~O,32 -0,28 +0,01 -0,45 +0,39
161o:M +O.fJO -!-O,37 -0,10 -0,20 -0,23 +0,27
I There appeared no reason to question the positive results of other
auth9rs and, as far as. results of the present study were concerned, it ap-
peared. that the distante of 150 km (Abastumani - Tbilisi) actually ~was of
consi:derable significance. This problem will be investigated more-- .
thoro,ughly in the near future using the temperature data of two aerological
stations Tbilisi and Sukhumi between which Abastumani is located.
! .
BIB LIOGRAPHY
----_.------------- ---
_.~~....- ---- .-------- -.-.---- -. .---.---.-.----.- ____n.". -~- -.
I j
I i
I :
! ':
1. r y .Il H P. M. ctIII311Ka .CTpaTOajJepbl. rll.IlpoMeTH3J18T, n., 1958.
2. 5 e 3 B e p x H HAW. A. '030HOMeTpH'IecKHe AaHHble no AnNa-Ate B. co-
lIOCTas.1eHHH C HeKoropblNH Ne-reopoJlorH'IecKHMH 4JaKTOpaNH. TpYAJII
Ka38XcK. HHrMH. Bwn. 5, 1955. .
I I
; II
1 !; ,
'Si~STEMATIG ERRORS IN FILTER
I j :1
,I :;
. I .
i I :
. I
: I
. II'Different types of electrophotometers equipped with light filter:;
are of~Em used in measuring spectral transmittance or optical mediumi den-
sity. , The instruments detect mean radiation intensity within a certam I
., i! . I
finite' spectral range. Therefore, accurate interpretation of experimental
I .
data a9:quired in this manner cannot be made without examining the effect
of this::bandwidth on the results .
i
- EQUIPPED OZONOMETERS
I
L. G. Bol'shakova, A. L. Osherovich, 1. V. Peisakhson
I
I
I I
, I
. ( .
I Let 10 (A) - denote the spectral-energy distribution of a light source.
,.
. ex ().) - stand for attentuation factor of the medium under study,
'I (A) - symbolize spectral transmittance of a light filter,
S (A) - stand for spectral characteristic of a radiation detector.
- 44 -
-------
1 - denote optical density of the medium under. study.
Two problems may be formulated:
1.
Determination of a (A.) when 10 (A), 'I" (A), S(A), and 1 are known,
and
2.
Determination of 1 when 10 (A), 'I"()\.), SeA), and a(A) a~e known.
Measurement errors corresponding to the first problem, are.
analyzed in [lJ and [2J.
An attempt was made in the present report to estimate errors in
problems of type 1 occuring in determining 1. In doing that it was as-
surned that the relationship between electrophotometer output signal and
the luminous flux undergoing measurement was linear. Now, examine one
possible method of solving this problem in a specific case, -e. g. determin-
ation of total ozone in atmosphere (x) by the optical method. In principle,
the method consists in measuring the spectral intensity*' IA of any extra-
terrestial light source radiating in the ozone absorption range. .
It is known that
- -'-- ,..-
.h. = loAIO-(~A.J:!'+PAm+t).m+'IAM).
(1)
. where 10 A - denotes intensity of a monochromatic radiation from an extra-
terrestial light source on the atmosphere boundary,
aA - denotes fractional coefficient of ozone absorption per ml (**)
f3A - denotes coefficient of Rayleigh dispersion,
')fA - denotes absorption coefficients of 02,.H20. CO2, and other'
gases, contained in the atmosphere, .
1] - denotes attenuation factor due to aerosol dispersion,
\.10. m anlM, are the so-called masses of atmospheric ozone. at the atmos-
phere, and of the aerosols, respectively, in the direction of extraterrestial
light source. Spectral intensity values for a strictly monochromatic ligh.t
can not be obtained through the determination of LA. by any ..spectrophotome~ric
device, since such a procedure will yield only the mean intensity values.
within spectral intervals of the instrument's sensitivity. Only .very- com-
plex spectral systems can separate spectral intervals a few angstroms wide.
In this case, the measured intensity (allowing for system losses) differed
little from the monochromatic, and all factors which affected the absorp-
tion coefficient corre-sponded suff-iciently accurately to the. median of the
..h spect.ra.l i~terval._~,:a_n~t~_e_s- ~A.'. f3>.. a~~'YA_~~n not be regarded as true.
-*Ene-r-gy--incidentori a- unit-wavelength l'ange-pe! unit time.
**The author intended presumably J~m-l" (Tra~"slator)
- 45 -
-------
constants in an optical system, e. g. a light filter which discriminates,
~ spectral interval of the order of 100 A and higher, and radiation intensity
I registered by a filter equipped photometer is to a considerable degree dif-
ferent' from monochromatic intensity. However, in this case total absorp-
tion can be represented by formula (la) similar to formula (1)
-..-- -... .".--'-"
--- ---- -_.....~--
- - K
" = '010- ,
(la)
where K - represents integral coefficient of absorption (3),
I and 10- are determined by (4) and (5).
!
Assuming a finite spectral bandwidth discriminated' by the light!
filter, estimate the error in determining value of x by formula (la) in a
case where ozone' absorption coefficient, a component of integral coeffi-
cient of absorption K, is replaced by the monochromatic value O:'A corres-
ponding to a wavelength at which a photometric system exhibited m
maxim.um spec:tral sensitivity. '
Effective spectral sensitivity of a system "light source + filter +,'
detector", may be characterized the the following function:
+--. .. -- -- -----.-- -
-~.---_.~ - - ----------..----------..----
... ----.- - -..- n - - - ----_. -
.. .-..-------
- -_.- _..~ -- -..------- . ----.
h_'_"-' -------.1'- (~L~Jo (i.) -: (A) £ (A).
(2)
which is normalized, so that
....,
. ' , J (I.) tfl. = 1,. ,(3)
- - ~! - ---~... ------- -_.._----_.._---_J~!_---,._--_._-----_._.--_._,._- -- --~--_..- ....+-'. .-- -. -
where Al and A2 define the limits of photometer sensitivity. The mean.
radiation intensity detected by;he photometer is determined by the following
. ' I
ratio: '
A,
, i 1 (A) -: (i.) E (k) fA
/- i,
, - A,
I ": l.A) E (I.) d)'
-- --- -- .. - - . -- .- --'------' -- ---+- -----~--- - ----:.I
, and the mean value of 10 .
! .
(4)
---- -- -- ---
- "-- ~ -- -
10 =
.. '
f 10 (k) ": (A) E(~) tf).
it
...
J ": (>.) E (k) di-
A,
"
(5)'
-- - - - - - --- ---- --.-----
I,
The~ using(Z) ~nd(3) d~rive (6)
---- - .n
--.----- -- _._---~---- -:-----~--_._--_._~---
J- = \" 1 (A) c1J(I.)dA.
10-.. .. I,(A.)
A,
(6)
- 46 -
-------
A:::~sume 'as the first approximation that
t~- - ------
-- -- --~-----------
no - .--- -..---- -- --
1 (') -- ~ (I p.. - i I )
(, I. -.. -
. /!A .lA.'
(7)
i. e.. effective sensitivity distribution is represented graphically by:
1) a triangle with the apex at A =A . and 2) a half-width equal to &.
This approximation can be regar~d as sufficiently accurate for filter
equipped photometers to be used in ozonometry. In this case, as shown
by one of the authors (4), the following approximate relationship prevails:
---------.-- ._h - ---_._~'---'--- ---- - -- ---------~'- - - -- ._- .- .
- -- - --
I'
I,
7 I (Am) iJI [ 11] AI.'
==-+- - -+... (~)
. '0 I. (/'m). 0,.' I. (I.) A-1111_.~__.- -- -.- . -- ------- -'-
-. -.- - '-Whe~'-u'nd;;~-'such '~o-~ditions I (A)-i~--~xpressed by equation (1)
by equation (la) then
and I
,J
:)
I
I (I.)
-In -- = k(/.),
I. (1.\ .
(9)
whe re
k(i.) = 2,3 [I1I,XI-L + p,m + -;..m + "tj,Mj
( 10)
-. "T -- -..----
and
- In Z- . = K. ( II)
"'-.----'-'---- ,.-.---' -'------'-;:-":'(-----'" .. '. -
Based on (8). (9), and (11), it follows L4J, that in the first approximation
K = k(i, )_{Ik'(i))'-k"(')} /),.,.'. (12) .
m 12 1-1.
~)
__--on -- As"s.ti-m;;--ili;;'t-in' the passband.<:>f a_~_~~gl~.filt~!_JA'-:.:::-100~'150-A). -..---'
Assume ~A., ')fA, and ji}/t (quantities which vary with wavelength much slower
than (aA) to be constant in the 100 -150 A.sing1e filter passband for A = Am..
. Assume further that. ,
~, m +- TA m + 'f,.. M = B
1ft . .
( 13)
then differentiating (10) with respect to A ; derive
--- --- -______n______------ - .------------.--.-.--.--
K = 2,3 (111 X~ + B) -
III
12.3' (2' {Ah-1 %!,-JI- 2.3(2". (Ah-.. %!'-J}
- I.m III /),.).1.
I 12
---~---- - _A -.-- -~---~~~_.__.._-,------------_._._----_._--_._- .-.-
Now convert (11) to the common logarithm; introduce
and obtain
(14)
a new notat~.on
J( T
P = 2:3 = - Ig7;'
- -A;;~~i~-ith~{(iaf h.~d~ fo~-the monoch'romatic
coefficient of ozo~e absorption is a.Am we then have
P = 111 Xu'" + B.
.
radiation. and that
(15)
~---
- 47 -
-------
Values of xo computed from (15) differ
x by an amount Ax.
from the actual ozone content
. _. ~ --
. - - -_.,.--u.. ~----- -- .....
ouU ---' -
and
% = %0 + .lx.
P-B
%0=--.
/'4J.
.
(16)
(17)
--F~o-~Tf4j;- (16 r -~;d-{J. 7)it-~an beseen-thatUthe- -errorTil-- de-te rmi~i ng -
x with the aid of (17), up to terms of the second order, is .
~X = Xo
2.3 (2~ )II'X. - 2: .
1ft .. ~;..t.
121.
I.
1ft
. (18)
.---- --- -U-i~- th-e-3-00-0-~-3500~A-;;-~~-g~~th;- o~-;~e--fu~-cif~; a -(ir~~n-be -~~proxi-
mated '. ,
------_u_._____n_____u_--_u_._----~g:x (I.) = C1 - Ct (i. u- 3000), --------- ----- (1~L -- __n
. in which the wavel~ngth is expressed in Angstrom units. For example,
in the 3100 - 3200 A range, the mean curve is reproduced sufficiently ac~.
curatelyfor Cl = 0.696, and in the 3200 - 3400 A range for Cl = O. 738. .
.' -3 0-1
It can be assumed that in both instances C2 = 6. 36 x 10 A ~ Using equa-
tion (19), and calculating the derivatives rl (A.) and all deriv::e .
;1% ~= O,44xo (2,3:1;. XuJ-L - J) (C,!J.).".
lit .
(20)
.i
It can be seen from form:ula (20) that magnitude and .s~gn oferro'rAx.
---"cfeperided essentIally: onmeasu-rement conditions (XO, fJ.,AA., and Q\ )'. When
2.3 a~ . XOfJ. < 1, Ax.'< 0, i. e. equation (17) yielded an overestimateWtotal
ozone I?I' value; and when 2.3 a :eofJ.> 1,- then ~he value was unde.restimated.
. Minimal errors are encountere~ln using equabon (17) when magnItude of
xo~ was the order of 1/2. 3a).m' Calculation of errors with. the aid of (20),
when xo fJ. varied from 0.1 to 3:"5 (v.rhich corresponded approximately t'o ob,-
servatiori conditions) and &.. = 125 A and A. = 3160 A, showed that the ex- I .
m - - ---- ." --
treme relative errors were in the 30% to +40% range for xo fJ. =' 0.1 and XO fJ.
.-- --=3. S;'respectlvely, -and- d-ecrea'sedt6 1% -for Xc ~= o. 9~Equation 20 shows'
. error magnitudes were directly proportional to the square of the filter
~. -----."" - .. - - ,---~-'--~' '. . .
;lpassband) half-width. Thus, at A.A. = 70 A extreme -errors- varied from
. . ,
-8. to +20%, and fo.r AA. = 40 A. from -3 to +7%/ . ,
. 'I .
Two filters have ~een used generally in ozonometry to reduce errors
due to attenuating aerosol properties. ,Their sensitivity maxima corres-
pond to A.1. and )'2' and the half -widths to A.A.1 and AA.:;!. The formula for
determining total oz'one froni. measurements made with the aid. of two filters,
. 'I - ,
is of. the fol1ow~ng form:, . .' . . .
. ~ . . .
--1------------;----'---- -'--'---------'-- -- ---
I
.50-5 -c
xo=
IJ. (:I;.. - "..)
(21)
---"- - "'- n -
- 48 -
, .
-------
I i
in which
. --L -- -;---- -
I' S ~ Jg ~ S =.~ Ja /(A,) C 1
o 70(/,)'. ~ 1(>.,)' = <. " - :1).,)m.
-------~-
~-_.__._-- - -
In "this case error 6x is expressed as follows:
A.. 0 44 ".,2 (2,31). xol-' - I) '1;. .:ai.~ - (2.3xo:''1. - I) '1. .:ai.22
lJt,A. == . XoL,2 I I ,.~ ,.~.
". -C!.
I.. 1...,
(22)
J Galculation of Ax values by formula (22) had shown that for equal
half-widths, 6x was of the same order of magnitudes as when measuring
x with a single filte r .
The same problem can be solved by a different approach. In
calculating x from equation (17), the mean adsorption coefficient value fo;
.~ ,~_~,:,:en interval can be used instead of aAm as presented -..""
. i .
:( '1 (A) cD (i., di.
i..
a = .
'. .
J cD (I.) di.
;
(23)
However, in this case x:o value also differed from the actual ozone
concentration x. It can be shown with the same degree of accuracy that
the calculation error in this case is equal to .
...0 -n.' &X~'2~~~ .r~~~ (t 2S'if .... . .'
{24l--.'
Thus, underestimated values of x are always obtained when using a in
equation (17). As can be seen from equation (24), 6x value in this case
depended also on the observation conditions. The order of magnitude re-
mained in the same range as in the previous case. Only the ratio of error
to x:ol1 was different.
All results in this paper are approximations; nevertheless. they
are of considerable value in determining the order of magnitude of systemat-
ic errors introduced by the coefficient of ozone absorption when using the
indicated methods for the determination of x values with the aid of filter-
equipped photometers. It was previously noted that the error may at
times amount to 40% or more. especially when values of x:o 11 and bA were
large. In such cases, determination of x by the indicated methods. even
allowing for the systematic error, leads to false results, since the error
is estimated in the first approximation. However, it is extremely difficult
to estimate the errors accurately. It appears more rational in similar
cases to directly integrate equation (la) over the entire filter passband.
However, in this case, accuracy of the results will depend on the filte r
- 49 -
-------
bandwidth. Estimating the possible errors led the present authors to con-
clude that most accurC!-te results in determining atmospheric ozone conc~n-
tration, allowing for possible systematic errors, were obtained using
filters the passband h~1f -widths of which did not exceed 100 A. and that the
3100 -3300 A was the most suitable for such observations. .
BIBLIOGRAPHY
- . . ---- --. .-. .
CA~~');-I-u'~-E~-~~~rl \\;-u'-nrl'-c-;:li'~-h-~i~-Optik~li-~;. '11. 5(.3-512.1955'
2. H an 5 e II G. Optik 12. Nr. II. .197-502. 1955.
3. Ii II K 0 II 0 II B. I). DIO.1.1fTrllb .-\6aeTYMalieKoH aeTpo<\JlI311'1ecKoit 06cep.
DaropHH. 14. 1953.
4. n e it c a x eo H H. B. BeeTlIII" .'lellllllrpa.leKoro )'IIIIBepeIiTeTa. eep. $H3.. .
Xv .5. 129,-143. 1955.
--
I
I .
SOME; PHOTOELECTRIC OZONOMETER TYPES
i.
A. L. Osherovich and S. F. Radionov
I . .
. Most ozonometers used at the present time, including complex,
spectral systems and simple filter-equipped reticular ins_trur:nents, are
main1YI photoelectric in principle. Yet, none have reached th~ point of
perfection. Attainment of higher accuracy in ozonometry depends upon
improved accuracy, simplification, and applicability of the corresponding
I - ..n n-
instruxpents. In this paper the, authors present results of research
. -iii-thiiY'dii-ectlon;n and propose two type's -Of ozonometers which con-
sist of:three basic units:
I I
;1.
,intervals
!
A, system for di~c:riminating comparatively narrow spectral
in the 3100-3300 A and 4000-5000 A regions in some cases.
.12.
An e1ectrophotometer.
3.
A device for aimirig the ozonometers at an extraterrestrial
!
light source.
In general, photoelectric ozonometers can be divided conditionally
into two groups:
I'
. a) instruments! used in extensive station networks
simple optical syste'rri (e. g. light filters), and
, ..
I -50'- '\\
equipped with a
-------
b) observatory-type instruments having a complex system with
double spectral discrimination.
In using photoelectric ozonometers, instrumental errors, intro-
duced into the metJlOd of determining total ozone x, depend basically on
1). degree of monochromatization of the measured radiation, 2) the extra-
terrestrial source, and 3) the photometric evaluation of line intensity IA
(or the ratio 1A1 /IA2)'
FILTER-EQUIPPED OZONOMETER
A simple integral elecf:rophotometer equipped with light filters was
used in ozonometry long ago [1, zJ. However, due to technological
shortcomings of electrophotometers and filters existing at that time in-
tegral ozonometers were replaced by spectrographs and by the Dobson
spectrophotometer. The modern filter-equipped ozonometer was developed
and first used in the USSR by the present authors in 1948 [3J. An advanced
version of that instrument is described in this report. It should be pointed
out that the type of proposed photocell, photomultiplier, and amplifier were
not the only possible ones and that in principle the instrument remains
unchanged even when individual components were replaced by analogous
parts. The line block-diagram of the new instrument remains the same as
_0e_ol~_one[_9). . Multiple-slit cathode photocells.a~d the STsV-9 antimony-
cesium sensors wei~ used as the solar radiation detectors. Figure 1 shows
their spectral response curves in absolute units.
Fig. 1
--..------- -----
---.--. ----. -
18.
~ ta
CT
~ 0.1
o .
a:
l-
e.>
...
if. '0.15
>
I-
-
>
;: 0.1
en
:z:
...
en
; D.D!
a:
l-
e.>
...
11-
(/)
z
,.. ,...
: ",,' t,.....--
t ..., '......
I "
",
"-
2000
JOOO
----- - -
-----
SPECTRAL CHARACTERISTICS OF PHOTOCATHODES USED
IN THE OZONOMETERS .
}.ANTIMONY-CESIUM; 2-MULTISLIT
The present authors used either an
FEU -11 photomultiplier, featuring a
multiple -slit semitransparent cathode
on the ultraviolet-transmitting glaS!i or
an FEU -18 photomultiplier in making
lunar observations.
Zero drift was reduced by
amplifying the photo current in the
instrument by means of a differential
cathode follower [5J instead of a con-
ventional D. C. amplifier [4J. The
line diagram of an integral ozonome:te r
used in making diurnal and nocturnal
observations is shown in Fig. 2.
Zero-drift at a current gain of 5 x 104
did not exceed 2 IJa over 3 hours
continuous operation.
---'-. ._--
- 51 -
-------
Fig. 2
- --_.- -- ----------.---:--.---/----- .,-------'''-
..._- --.-'.---
0-<..'84'
----.-. .--- -----
-!fA$IC-'PLANhOj!-tlIT-IliTEGRU--OZONOMETER FOR DAY AND NIGHT 09- .
SERVATIONSe U--OPERATING-TENSION OF- THE PHOTOMULTIPLIER;
l, AND ~ PENfODE5---o---PHOTOELEMENT; ---PHOTOAMPLIFIER;
P - SENSITIVITY SWtTCH; -- MICROAMPEROMETER;
Rl = R2 = 100 mO; R3 = 25 mQj ~ = 5 mO;
R6 =- 1 mOj Re' = R7 = 51 kOj Ra = 500 kOj
Rg = R10 = 6.2 kO; Rll = l5kO; Cl = C; =
\.,O.\J.F; C3 = O.lIJ.F; C" = 0.5 IJ.F; C5 = 1 IJ.F
The siet of filters in the new instrument represnets combination of colored
glass,C6J and dielectrically-coated interference filters [7, 8J. Transmis-
sion ~urves of a set of interference filters are shown in Fig. 3 in which
percentage transmission prepared and described by T. N. Krylova and
R. S. ~Sokolova [7J is plotted along the ordinate. Interference filters with
a still~narrower transmission band in the 2900-8000 A region were de-'
velopE!:ci by F. A. Koroiev [8J. Figure 4 shows the spectral response of
\ 0
the t,6 Korolev's filters in the 3350 -4300 A region.
i Fig. 3 Fig. 4
." -_. .-'.' -. -- .--- --- ---
;1' f'A
'fA
,i:\ ':.Q 20
I -, ~
I:'
,.
. - '*- ~ f5
(5
z :z
-- -
:z :z
0 0
1-- 1/1, in
< f,o en 10 ,.
a: ~
1-'
~ en
Ii - z
I.L. <
rr
....
0.5 5
o
j!JOO
'0"0- ---- - --~-- - -0-
WA VE LENGTH' A
WAVE LENGTH A
----------------_. -- -------.-. .-------.. ---
.-._------_..-._- -_..--- --- ---- ..~ -.........- - -> ------.
TRANSMISSION CURVES OF KRYLOVA AND
SOKOLOVA INTERFERENCE LIGHT FILTERS
TRANSMISSION CURVES OF F. A. KRYLOVA
INTERFERENCE LIGHT FI LTERS
- 52 -
-------
Curves in Figs. 3 and 4 show that the proposed filte rs exhibited
considerable advantage, with respect to the transmission band width over
the ones used earlier [9, 10J.
The second advantage of the above mentioned interference filters
is the possibility to shift the transmission maximum position to the de-
sired ozonometric region, a procedure difficult to achieve with other
filter types. In using such filters, or any other metal or film single -layer
filters, it becomes necessary .to consider their mechanical integrity. A
slight perforation in a short-wave filter can admit longer wavelengths and,
as a result of a sharp rise of the solar spectral response in the direction.
of longwave, can seriously distort measurement's. Such exposure can be_~
..n _--~~~i~=d by .us~Ilg_a system of two identical, but separate, filters, mounted' i
one on top of the othe r.
Research is now being conducted in the photometric laboratory of
NIFI LGU with the aim of further narrowing the filter passband in the 3100-
3300 .A region with a 15 -20 A passband and,. A = 3 -4%. In this connection'
S. B. Ioffe determined that au}_~..mm <:ii~~~!e_l"fi!ter with the required para-
meters will be 30 mm thick. Such a filter can be used not only in ozonometry
but also in making other spectrometric observations.
Temperature effect on the position of a polarizing filter trans mis-
sion maximum may require a certain precalibration of the spectral system.
The proposed filter-equipped ozonometer can be used within a broad ozono-
metric system. The instrument differs from similar existing instruments
in having filters with higher optical quality and higher amplifier unit stabil-
ity and having a negligible zero-drift. The interference filters and differ-
ential cathode followers can be used to modernize the OFET -3 reticular
ozonometer. In fact, this procedure was carried out at the photometric
laboratory with satisfactory results.
THREE-CHANNEL OZONOGRAPH WITH DIFFRACTION GRA TINGS
In building an ozonometer with a high spectral resolution and sensi-
tivity, low inertia of the system, and automatic recording, A. L. Oshero-
vich and B.A. Kiselev developed an observatory-type instrument consisting
of the .following units:
1) Coelostat for automatic guidance of the sunfs'image.
-2) . Double monochromator with diffraction gratings and fixed slits
for separating three spectral inte rvals. -
3) Electrophotometer with an automatic recorder.
The plan of the instrument is schematically presented in Fig. 5.
- 53 '-
-------
Fig. 5
. .. ._u -_.- . - --.
.. _0. -- - -._-.. - - ..
_.0 .. ..-- .. --_u
r. u
~e
I
GENE RUPL A N -OFY-RIPLE=-CHANNEl OZONOGRAPHWI TH-Di'FRACT -. oli-sCRE ENS;
C - CELOSTAT; StN-iNC. SLIT; 3., 32~. 33'~4~S~ERICA MIRRORS; DG-
DIFRACTION SCREEN; SM - MID SLlTS; 'r].. ~. ~ - LIGHT FILTERS;
SOVT- OUTGo SLITS; Sc - SCREEN; PHEM - PHOTOELECTRIC MULTIPLIER;
Yt' Y2' Y3 - UECTHC CURRENT AMPLIFIERS; TPA "3" -.TRIPUNCTATE
AUTORECORDER.
. The present authors used parallax guide OFFET-3 [lOJ, in which
the mirror rotated at the rate of one revolution per 48 hours as a coelo-
I
stat. Other devices for automatic guiding, such as Badinov's photo~
electric guide, can also be used. The instrument's optical system Idis-
. crim~nated three spectral intervals ~f the following wavelengths: Ai =
3100 A, Aa = 3300 A, and As = 4358 A, . The instrument can displace
o
. working spectral intervals by :!:: 50 A. The system was designed for dif-
: fradion grati~gs. or replica gratings, with 600 lines/mm and on effec-
.' . ~-._--_. ;---....---'. '--,--'----' . . .. I
! tive 62 x 51 mm optical area'. The linear dlsperslOn at the spectr:lal
I system output was as follows: .
I
.-..------- ----.---- ---------------. .--
.------ ----- ._-_.-
j. DISPERSIOII SPECTRUM
J. AI- ORDER
3100 12,3 3
3300 7,8 2
4358 7,3 2
The working s lit width is
300 mm, and it's aperture ratio
effec'tive grating area.
0.3 mm. the collimating lens focus
was determined from the above shdwn
Figure 6 is a photograph of the uncovered instrument; it measures
75 x 30 x 50 cm. .In constructing the instrument use was made of a bank
of three identical-type electrophotometers with FEU -18 and FEU -17
- 54 -
-------
photomultipliers. Amplifier output currents were recorded by a microam-
meter or by EPP-09 three-channel recorder with a-lO-mv sensitivity over
the entire instrument scale. FEU -18 and FEU -17 photomultipliers were
properly chosen to help reduce recovery time of the photo electric circuit
to 30-40 min. Photometer stability, linearity, and gain at various grid
loads were checked- under laborat~ry conditions. The instrument's high
spectral resolution sensitivity, fideli~y. and low inertia are essentially
new features which enable investigators to conduct concurrently typical
ozonometric observations and precise spectrophotometric measurements
of low radiation under conditions rapidly changing with time.
Fig. 6
r ._- - ---
~
- -- ---- ---
- --.- -- -~ --- -- -
.":t:~-
--- -. ---.
----------
---- -------
PHoro-OZONOGRAPH WITH THE LID OFF
This observation category includes ozone and aerosol measurements
during the sun's and moon's rising and setting, during eclipses, and,
in general, during any rapid radiation processes. The structural analy-
sis of a measured spectrum for the determination of temperature of the
upper atmosphere layers and the like also belongs to the same category.
The new instruments were checked under field conditions in the summer
of 1959 in Crimea and Karadag. Typical results are depicted graphical-
ly in Fig. 7 in which values of x determined by the known three -wave-
lengths method (13) on 25 August, 1959 are indicated along the ordinate.
Cross and solid black circular plats in Fig. 7 clearly show that the ozono-
metric data agreed closely with the spectral ozonograph data. Linear
curves of log lA. as a function of sec. z. which can be used in computing
x, are shown in Fig. 8 for a particular observation day. That day was
not included in the sum of X values .due to the anomalous slope of the
.
strai~ht lines (slope of A. = 3095 A -line was not as great as slope of A.2 =
3295 A instead of showing an inverse ratio).
The unusual behavior of the straight lines in Fig. 8 may have been
caused by an increased amount of atmospheric aerosols on 18 August, 1959.
- 55 -
-------
----
~-H_--_- ---------------- -
Fig. 7
Fig. 8
'IJ.
1.11
. ,
~ .
. .11 .
~ .... . . x. x.- ...
.. . X8 IItX. .
.,," ...
..
'.J
2' J
1#
s
8
7
3;C/If
0.3
0.2
/Q
0,'
. ....,.
IlIe ICIe Z
. O,j
o
I
I
I
I 88cZ
z
J
II
6
i8
o
-------
------~'-------
-~---------------
- . ,
FU~NCTiU~Ai-R~EiAT-iONSj:iIPBETWE-EN LOG ,-,'
I AND SEe z IS/VI I" 1959 AS OBSERVED IN -e
. KARADAGALCJll.M£.L- --- -.. -- --.
).,=-3"095 A; ).,2 = 3295 Ai ).,3 =
. -- -"0 --.
4358 A
COURSE OF TOTAL ATMOSPHERIC OZONE AT DAYLIGHT'
OF 2~VII', 1959 AS OBSERVED IN KAEADAGA, CRI-
MEA. I - OZONOGRAPH . 2 - OZONOMETER
The selective attenuation caused by aerosols in the ultraviolet region
and its effect on ozonometric data had been investigated by one of the authors
[14J, who emphasized on numerous occasions the importance of taking into
consideration the aerosol component in atmospheric transparency. Att~rition
should be directed also to the necessity of using in ozonometers light filters,
which exhibited the na,rrowest passband. The use of wide-passband filters
in ozonometric studies can lead to errors in determining total atmospheric
ozone. This possibility is not given sufficient consideration by some in-
vestigators.
I
The authors express their gratitude to S. Brukman and N. Shpakov,
students at the Physics Department, Leningrad State University, for thdr
help in making measurements. '
BIBLIOGRAPHY
-------.----- ------.--- --"---
-------_.__.~-
1. f e T 'u n. ATMOC4lepllblii 03011. fTTH. 1934.'
2. R its c heN. Z. Meteor.. 13 okt.. 1947.
3. PO,!!,IIOIl9B C. CP.. OWepOBIIQ A. JI. H P.:tYJlToBcKa8
E. B. llAH. 66. Ni 3. 381. 1949.
4. B 0 H Q - B P Y e B II Q A. M. npliMelieime 3.1eKTpoHHblX ,13Mn B 3KcnepH- -
MeHT3.1bIIOii .
-------
,!, ----
---~ po 3 B. TI--:-A)i(. 33. -X; 5. 1f7:-19s6.' ' - . . ,,- _.
6. K;lTa.~or uueTiforo CTeK.13. O6opollrll3. M.. 1956.
7. Co K C .1 0 B a P. C. K K p bI .1 0 BaT. H. OnTIlKa II ~~neKTpocKonH..
6. ]li9 6. 788. 1959. . I
. 8. K 0 pO.1 e B 4>. A. BecTIIIIK MocKOBcKoro YIIIIBe~CKTeTa. cep. eIIK3.. HI 3.
97. 1953. I .
9. 5 e 3 B e p x II II Ii ill. A.. 0 ill e po B II 'I A. n. K Po;J. ~ 0 II 0 B C.~.
H3BecTliR AH CCCP. cep. re;)(jIl\3.. H9 3. 93. 1952.
10. 5 e ,3 B e p x II II H W. A.. 0 w e po B II 'I A. .Tl II PO..'1 II 0 II 0 B C. ~.
OJOllorpaK3waTII3,UT.
M.. 1958.
13. n po K 0 . H3BeCTIIR AH CCCP. cep. reoct>"3. Nt 19. 334. 1950.
OZONE AND GENERAL ATMOSPHERE CIRCULATION
G. 1. Kuznetsov
1.
Introduc tory
Remarks
The problem of ozone dependence on large-scale atmospheric
processes and motions is an important one. It has been known that ozone
distribution in the atmosphere can not be explained without taking atmos-
pheric motion into account. On the other hand, ozone itself significantly
affected upper temperature and, consequently, baric and wind fields.
Therefore, it is difficult at times to separate cause from effect in a prob-
lem which is occasionally referred to as "ozone -weather". The probable
role of ozone as an intermediator between the sun and earth, its effect on
radiation balance of the strato-and troposphere, the "hothouse" effect of
an ozone layer, strict correlation between ozone and temperature of the
lower stratosphere, and the direct relationship between high temperatures
at the 35 -50 km level" all indicate the great importance of ozone in the
"ozone-weather" problem. However, in large-scale atmospheric pro-
cesses which also occur in the "conservative" ozone region at altitudes
up to 25-27 km, i. e., in the ozone dynamics problems -the'dominant role
is played by synoptic processes.
It is to be regreted that present day synoptic meteorologists ap-
proach the detailed study of atmospheric motion processes above the 20
km level only.' The numerous existing studies are concerned with the
relationship between ozone content and its variations and certain synoptic
processes occurring mainly in the troposphere and lower stratosphere,
namely: the propagation of cyclones, the origins and the type of air masses,
..-- -.. - _..-
- 57 -
-------
the temperature and pressure at different altitudes, etc. These studies
have shown that there existed a well-defined relationship between ozone
and such elements of the gen~ra1 atmosphere circulation as cyclones, air
masses, and others. Ozone changes can be represented by mean annual
variations or by short-term daily ozone content fluctuations in relation
to the time scale.
2. DAILY OZONE FLUCTUATIONS
Rapid and substantial daily changes in total atmospheric ozone
appear clearly evident against the background of comparatively smobth
annual variations. The daily variations tend to increase during the winter
season and at the beginning of spring, and to abate during the summer and
autumn seasons" although new increased variations had been observed at
certain stations during summer [12J. Apparently seasonal and daily varia-
tions in ozone concentration "are caused by different reasons. The former
ozone concentration changes are determined by large-scale processes re-
. sulting from seasonal changes in solar radiation, photoch;emical process-
es, general advection, turbulence, and slow vertical currents; the latter
are induced by corresponding short-term variations within a system of
gelferal atmospheric circulation. Based on direct measurements of at-
mospheric ozone distribution, Paetzold was able to show [lJ that annual
variations in ozone concentration above the 30 km level were in good
agreement with the. photochemical theory, and exhibited a maximum dur-
ing summer and an amplitude of'" 0.02 em, 1. e. of the order of 5% of.
the ozone total at intermediate latitudes. Seasonal variations in the 25-
30 km layer had not been observed generally. The same investigations
showed that daily ozone variations above the 25 km level were quite amaH.
.Annual fluctuation amplitudes in ozone total at intermediate lati-
tudes attained 40% of the annual mean (see Table 1). In other words,
basic ozone changes occurred in a layer below 25 km. The greatest
monthly difference, even during periods of sharpest seasonal changes
of mid-winter and late autumn, did not exceed 8% of the mean annual
ozone content. At the same time, short-term ozone fluctuations during
a single day or several days, frequently attained 40-50% during spring.
For example total ozone at Reykjavik on 6 and H February, 1958 was
0.326 and 0.57 em, respectively, for an annual mean of O. 350 em, and
at Vigna-di- Valle (Italy) on 29 January and 2 and 6 February, 1958 it
was 0.424, 0.280, and 0.406 em, respectively. The processed IGY
. materials show that short-term ozone fluctuations at intermediate lati~
tudes' attained 25~30%of xo, even during mo,nth"s of weak seasonal varia-
.' I'., .
hons. .
Larger variations in total ozone can be caused by ozone transfer.
during such dynamic processes in the lower stratosphere as interlati-
tudinal transfer, turbulent mixing, ozone accumulation near the air
-:;8 -
-------
I .
. ~'I
-.0 .
I
I'
,
I
- -
._.~ ---~----_.----- -~_._. --..-
Table 1
. .. I
-2 I
AVERAGE MONT.HLY TOTAL OZONE VALUES IN 10 CM .
-~--_..._-. . ....~. _.__..-...~._._--_._-----_.._--_..- ---- _0_.._-
.1957 r. 1958 r. 1959 '"
(/) '"
..... ..'"
4 Villi IX I X I XI I~ ..QI 4"
III I 1111 IV V I VI I VII VIII I Ix I XII III 1111 I IV I I VI =-c "'-
. z a: z..
I IX XI I V Z... ZD.
<> <:I:
< <
\425 - -1- '
DIXON 382 - - - - 525 574 445 336 308 - 342 - - - - - - - - 408 2fi6
RUKJAYIK, 1322 .314 293 303 -, ~1 345 419 427 407 400 347 331 326 302 - 272 - - . - - - - - 347 155
VOEIKOYO. 357 340 327 307 2951 - - 459 467 434 389 365 357 344 302 280 268 - 387 - -. -' - - 365 199
UKKf. JOt 297 305 260 3331 - - - 350 390 378 340 345 - 271 281 245 248 - - - - - - 315 145
ARIBU - - - - - - 364 386 371 - - - - - - - - - - - - - -
BISf1ARK. .296 288 280 253' 262 306 336 375 388 403 ;:156 355 - - - - - - - - - - - - 325 ISO
XL'BRUG'. - - - - 2i2 250 292 I 330 365 401 339 305 252 2:\3 231 244 244 281 305 - - - - - 293 170
LHWTA - - - - .213 233 280: 304 289 281 286 2~2 258 240 247 252 247 286 - - - - - - 271 71
VLAOIYOST1 - 270 278 268 301 - - - - 257 322 263 257 285 269 al9 292 288 - - - - - - - -
V I GNAo-O I
VALLE. 326 314 305 292 302 330 360 371 412 419 366 - - - - - - - - - - - - - 345 127
~BA6TUMANI 278 271 258 272 280, 296 314 358 366 352 30:> 311 287 287 266 260 276 300 315 - - - - - 307 106
loMAS 279 '304 291 294 295 302 322 328 334 364 336 - - - - - - - - - - - - - 314 86
WASH I NGTON 325 321 292 304 289 300 330 372 360 366 366 337 324 - - - - - - - - - - - 330 71
MASS I NA 267 329 314 308 318 331 356 377 379 ~3 348 - - - - - - "- - - - - - - ~ 110
M",...loA I - - - -.262 258 256 251 2iO 284 288 284 270 271 268 - - - -" - - - - - 259 37
BRISBANE - - - - - - - - - - 292 303 315 - 347 330. .319 326 289 284 288 296 293 286 309 61.
"ASPENDALE' - - - - - - - - - - - 377 - 398 386 345 "328 297 304 300 282 310 ~23 335 II'
MACKUOR I -
fS . r 356 353 373 428 381 - 334 299 332 342 337 395 378 412 414 435 392 365" 330 327 - - - - 370 134
I ~
: I
\
I
I
i
i
I
-------
. "barriers" associated with general atmospheric circulation. Total ozone
and origin of air masses which transfer it, the type of circulation, and the
wind direction at heights were compared in several studies [2, 3, 4J,
which showed that inflow of aIr masses from the north increased ozone
total at a given station at certain heights. Maiyake and Kawamura [4J of
Japan calculated the coefficients of correlation between total ozone above
the Tateno Aerological ()b_~~!.~atory and the wind direction at various
heights. ,'.-.' they: found that total ozone increased with northe'rn winds
(Table 2). The highest correlation coefficient was associated with heights
of 5 km above the tropopause. It should be noted that greatest seasonal.
and daily changes in total ozone had been actually observed in the 10 -20
km layer [1]. It has been known also that a secondary ozone maximum has
been sometimes observed at heights of 15-16 km for a given advection [lJ.
The above indicate that greatest changes in total atmospheric ozone were
associated with winds in the lower stratosphere.
Table 2
_... __~.__4- --------. -'------------~'--"- - .. - U. -
'--Cir:-FFI.CIENTO';-CORR ELATI ON -;ET~E-;O~ON~'CONTE-ri';' .----
AND WIND DIRECTIOn AT DIFFERENT ALTITUDES.
~:I~H.~.-~~"KM I N~~5~~ I ~~~:E~:: I--;~MA;KS" -_.-
IFF! 01 Ern
-_._-_._~_._----_. ---.-.-
6
6
10
15
_____17 ---
5 KM ABOVE
TROPOPAUS~
114
78
78
67
36
55
+0.03
+0,20
+0,28
+0,53
+0,46
+0,60
FOR .T.HE YEAR :-
...6.~1 NG-.SUMMER- .
FOR THE EN!.~.fI.~
It It It
It It It YEAR
It It
It
3.
CIRCULATION INDEX
In studying ozone concentration dependence on motion of air I
masses, a question may be posed regarding the relationship between
ozone and quantitative atmos'pheric circulation characteristic of circu-
lati~:m which constituted the circulation index. The regional circulation
index, 1. e. the ratio between the wind velocity gradient 'and the earth's
rotation rate at a given latitude, can be computed for different layers
in the atmosphere. The circulation index for the region of 40 to 65() ],~t;-
tude N, which encompasses the entire northern hemisphere, has been
used by The Me~e6rological Service. In this connection the following:
question arises first: to what extent can the circulation index computed'
for the entire hemisphere be correlated with processe's occurring in t,he
region -under investigation, . in general, and in Europe ,in particular?
This problem was subject of special investigation by ~. L. Dzerdzeyev-
skii and A. S. Monin [5J, who compared circulation indices, computed
for the 500 -mb level for individual areas, with the general index. The'(
.. ,
- 60 -
-------
I 6 0 I 0 °
icomputed lh for the 40- 5 N region in the area from 15 W to 90 E,
covering Europe and a part of Siberta, (X2 for the area from 1050 to 2100 E
covering Far East and the Pacific Oc'~an, and a3 for the area from 135 tft
300W covering North America. A comparison with' the general index a
showed that of 122 test days CJ. and CJ.l had different signs during 14 days,
a and as - 13 days, and a and as - 17 days. It was natural to expect a
still-smaller number of deviations for higher levels, where the' baric
chart became even simpler. Thus, it can be assumed that comparatively
great changes in the circulation index corresponded to processes occur-
ring in the atmosphere over the entire hemisphere.
Fig. 1
- --..----
--- )Ie;..... --~
...
o
z
. ----------'----..---------;------- --.---
z
o
-
l-
e
~
::J
U
,'=
. '0
-500Ml
-50M6
--JIM"'"
6~!...IlJML-- -
C-. AT GROUND
,g57z
-CCOURS'E- OF'r-EAiivMOtlTHLY AVERAGE CIRCULATION INDEX'-a-
ACCORDING TO LEVELS
Values of indices for 1957-1958 were obtained from the Dynamic
Meteorology Division of the Central Institute of Forecasting. Figs. 1-3
show annual variations in circulation indices according to months for
~-2-fferent levels, and .efmp.les of daily variations in the normal and mean
a" = (an-I + 2a" +an+l)f- Ilnd1ces. .
.4-
What was the cause of circulation index fluctuations and what did
they represent physically? Temperature differences between equatorial
and polar atmospheres caused a regional motion of the atmosphere. Velo-
city differences of individual regions resulting from the' Coriolis force
gradient and the local anomalies led to the 'generation of wave -like f1uctu~-
tions in the general current and the, so talied, -lonrfbaric waves. .
These converted into eddys and gave rise to a sys~em of cyclones and
anticyclones, the energy of which was derived from an ordered motion of
the regional circulation. Moreover, regional circulation weakened and
meridional mass transfer increased. Dzerdzeievskii and Monin had
shown that an increase in the interlatitudinal tr.ansfer corresponded to a
period during which a decreased. The cyclones gradually lost their energy
- 61 -
-------
Fig~ 2
100
80
.-. 100111'
~ 200M'
--- 306M'
----80
x
III
C>
:z:
:z: 70
o
t-
<
...
=>
(;)
a: 60
-
<.:>
IQ
110
~t-
j(!
.
\/'\
. \
.VV
. \
.
\
J~--~ }O DAYS
------- - JANUARY..
DAILY C.'RCULAT-ION INDE-XC;;~~;~----
(JANUARY, L9'.:B)
Fig. 3
and abated. The thermal inhomogeneity,
created by a nonuniform influx of solar
energy, began to grow and the circulation
index increased again. According to [sJ,
an increase in Ol corresponded to the prev-
a1ance of regional circulation. Thus, its
short-time changes were associated with
the :mo.f£on of large air mas ses, and it is
in connection with this dynamic aut9-
fluctuating character of the general, atmo-
sphere that short-term variations in the
ozone content had been examined. Another
special characteristic of the circu1~tion
index is illustrated in Fig.. It will be seen
that the curve of mean daily variati,ons in
the circulation index for different levels
shows that appearance of maxima and min-
ima at higher levels (50 mb, 100 mb) are
delayed by 1-2 days. This may hav~ been
caused by the fact that the ene rgy of at- ,
mospheric motions was derived ultiJmate-
1y from the lower atmosphere -troposphere
layers. I .
,.
n.. -- n__- -~ n_- -----.- - ._------..-
-----.---.......- ---
100
50
'10
jo 1
III
20
.-. '0: v~
-Q2;'~...t
o cJ:J",J
. -
,0
39
'J
- 62 -
-------
4. GENERAL RELATIONSHIPS BETWEEN OZONE CONCENTRATION
AND ATMOSPHERIC CIRCULATION
Ozone content at different stations was compared with circulation
,conditions by computing first general correlation coefficients between
mean diurnal ozone content (x) at a given sta,tion and the index a for dif-
ferent altitudes and seasons during 1957 -1958 (spring: February - May;
o summer: June -August; autumn: September - November; wint~r: Decem-
ber - January). Graphical comparison was also made between three-
points system mean values of (a) and (:~) at some stations during given
seasons. Where the number of days suitable for observations was suffi-
cient, the correlation coefficient between ozone concentration and cir-
,cu1ation index was computed by the "running" averages method to deter-
imine the relationships between fluctuations of several days. Ozone
concentration at some stations was compared with synoptic conditions
!as found in corresponding baric topographic charts.
I
iTab1e 3
,
-.- --- .---. . -_. - -.-. -_.
-.- -. -~. .---
- -
-"
--- ~ ""
u ----
- 0
> > "" en
z .. < 0 a: =>
0 '" ~ "" < a:
en ILl 0 0<: - -' :E: '"
C > >< - ILl ::0:: en .
'" '" -' au 0 ::0:: - ...
(f) :....1 c a: > ::::>
J.
o
J,
z
(D
- z
z 0
< I- <
:E: (D z
=> z -
I- (/) x (f)
(f) ~ en
< en '"
'" ... < ~
< UJ 3
C
l-
I
:E:
...
<
.-
'>
- -0,201
- -0,51
- -0,52
- -0,39
-0,66
-0,60
-0,69
-0,02
-{) ,50
-_.~------- -0,261_0,14 -0,11 +0,76 - -0,321-0.17 -0,04 I
SPR I NG 500 -0,27 +0,19 +0,22
1958 300 -0.32 +0,07 -0,24 +0,63 - -0,341-0,30 -0,05 -0,34 +0,21 +0,15
II-V 103 -0,37 +0,16 +0,52 +0,24 +0,33 +0,08.+0,05 -0,06 +0,47 +0,05 +0,15 +0,08
50 -0,32 +0,14 +0,57 +0,77 - +0,141+0,30 -0,02 +0,50 -10,09
-SUMMER -- 500 -0,47 -0 02 -0,14 -0,14 -0,1)21-0,37' -0,08
1958 300 -0,52 -0,17 -0,22 -0,59.-0,39 -0,2\
VI-VIII 100 -0,64 -0,18 -0,34 -0,671-0,55 -0,44
50 -0,46 -0,231-0,39 -0,64 -0,27 -0,29
----.-
FALL 500 -0,31 -0.30 -0,04 +0.46 +0,28 - +0,17
1958 300 -0,14 -0,16 - +0,26 +0,23 - +0,12
IX-XI 100 -0,17 +0,09 +0,02 +0,14 - +0,02
50 -0,23 -0,35 +0,11 0,00 - .,0,02
Seasonal correlation coefficient values (4), shown in Table 3, re-
flect a relationship between ozone concentration at European stations and
circulation indexes at different levels; (r) was found positive in! 37 cases
and negative in 68 cases. Thus, it appeared clear that ozone concentra-
I
I
- 63 -
-------
tion increased when circulation decreased over the 40-650N region. A.
more detailed analysis of this relationship will be presented in the fol-
lowing paragraphs. The negative relationship appeared particularly'
clear during winter and summer months when seasonal variations were
weak in both quantities. Highest correlation coefficients were associated,
as a rule; with high levels - (100- and 50-mb-) confirming the dominant
role of ,these heights in ozone variations. Positive correlation between
the two phenomena occurred at some stations during early spring and
autumn 1. e., during periods of rapid changes in the circulation index,
when general annual variations partly suppressed the short-time fluctua-
tions in the circulation index, thereby concealing their relationship with
ozone' fluctuations.
.Fig. 4
- .h_.._-- ~--.~--_._-----
-- ---..--"- - ---
. ... ,.----- n_-':----'_-~---~_._-------'-" __m_~--_~ -~ --..
--.- - --- _'0"_'.'
'0
SOD
200
z,
2:'
(f"Jt,;)'650
'0
)C
III
JO ~
-
:: , ODfJ
...
:2:
III
1;; 350
o
t,;)
III
~ 300
N
o
'-
:z
o
''0 ;::
4C
, ...
::>
t,;)
:.iO ::;
o.
, 6'
25U 1
I " . IlII . , yr . I
'0 20 2/1 10 20 JO '0 20
.-.1 ~2
JOI ~.. ._.!n__-.!.~_m JQ ! 7'
DAYS OF MONTHS
I 6(1
"
'00
-----_._._-~----~
CI1CULATION INDEX AT 100 MB AND OZONE CONTENT AT STATION VIGNA-DI-VALLE DURING
JANUARY-I'1fiY, ',958
t~IRCUtATION JNDEX AT 100 MB 2-OZ0NE CONTENT
The relationship between atmospheric circulation and changes in
ozone concentration is clearly illustrated by curves in Fig. 4. The curves
show variations in the smoothed-out values (see above) at lOO-mb altitude
and ozone concentration at the Vigna-di - Valle station of italy for the peri-
, \ . I
od January to May" 1958. For the sake of clear representation, the index,
. (a) scale was inverted. A significant agreementwas noted between the
general fluctuations and the extremes of both curves. The negative re-
1ationship between the short-time changes in ozone concentration and the
circulation index during that period was c1earcut; the, relationship became
discontinuous for short periods of time only, e.g. 28 March~-3 April, 14
- 64 -
-------
April--16 April.
A more detailed analysis of changes in ozone concentr~tion and cir-
culation index at the 100 mb level at Vigna-di-Valle (see Fig. 4) had shown
from 1 January to 31 May changes in the circulation index reciprocal and in
ozone concentration counted 16 clearcut extreme .and 20 clearcut minima
and maxima in the course of (a). Moreover, the ozone and reciprocal in-
dex maxima coincided in 15 cases, and their minima- -in 12 cases. A simi-
1ar coincidence of short-time fluctuations in (a) and (x) was also noted at
other southern stations, in particular at Abastumani and El'brus. How-
. ever, detailed graphical comparison was more difficult there due to a
smaller number of direct solar observations. Some characteristic rela-
tionship between ozone concentration changes and changes in the circula-
tion index are indicated by the spectra in Fig. 5, where changes in (x)
and (a) are shown with respect to the 100-mb surface. Both spectra are
similar pointing to a clear relationship between short-time fluctuations
in ozone concentration and the circulation index at the 100-mb level.
Fig. 5
Data in Table 3 shows that the re-
1ationship between ozone and atmospheric
circulation at other levels appeared weak-
er. Apparently, atmospheric motion at
the 100 -mb level (,...,,16 km), which lies in
the lower portion of the main ozone layer,
exhibited a weak effect on the ozone
changes. In accordance with the above
shown relationship between the meridional
and regional circulations it can be assumed
that ozone concentration over a station
situated in the southern portion of the in-
ves tigated region (40 - -650 N) inc rea sed
, during periods of intensified meridional
Z -~~~_3.~,!~I~-
circulation, and vice versa. Such de-
. DAYS -
pendence was observed in Japan and at
-PER'OD-,-I: FREQUENCY OF OZONE CONTENT -;;.5E
AND FALL (I) AND CIRCULATION INDEX AT . locations elsewhere. Intensity of meri-
100 MB LEVEL (II) AT STATION VIGNA-D' .dional circulation and recurrence of a
VALL~ DURING ~ANUARY-MAY,,9.X3 northern wind at any height mus t not be
equated. However, either way, meridional circulation impeded transfer
of air masses from north to south and, accordingly, southern masses
were propagated in the northerly direction.
---- ---. .- --.
.. - -- .". . -.. -- --
,
5
1
"
,-,.,
18
7 8 9 IQ
- DAY5---
11
Italian stations are particu.larly conspicuous in this respect. A
warm ridge is observed over Western Europe and on mean baric topo-
graphic charts, with its axis directly along the western continental bound-
aries approximately parallel to the Gulf Stream. A cold trough extended
to the east of this ridge towards the Mediterranean Sea. The ridge and
trough were most pronounced during the winter half-year, although both
- 65 -
-------
,-
were also clearly observed during the summer months. This baric situa.-
tion above Western Europe was accompanied by considerable meridional
circulation, causing northern air masses to flow along the isohypses into
the regions of Southern Europe. These flows became intensified as the
general circulation index decreased. The indicated connection was there - ,
_n"-hibrouglit info sharp"focus :-n,
It was previously indicated that seasonal changes in ozone concen~
tration and in circulation indexes concealed the correlation of their short-
time fluctuations. In trying to bring out the correlation' of short-time
I -u- " ,n ," , ,
fluctuations at -station. Vigna,..di- Valle for the same spring period in 1958,
correlation coefficient between (x) and (a) was computed by the method of
"runningll ave-rages using the' following formula:
.-. --_._-~--,-,-,--'--~-~-'-------~----'~-----------_._--.._~- -----
~ ft
~ (XI - xu) (g, ...:... Yo) - E (XOI - xu) (g01 - go)
. I-I '~I
, - , , .
- .. /'[,£ (X,- XO)I- E (Xu/- X~)2] [ i (YI- Yo)t- ,i (Yo/- Yo)l]
V 1-1 1-1 1-1 , I, I
I
, , I
where x. and y. denote running values of ozone concentration and circu1a-
l' 1 '
tion index, respectively, Xc and Yo denote seasonal values of these quan-
tities, and x , and y . denote values of x and y, respectively, taken
from the cur~~ whicR1 was smoothed in such a way as to exclude short-
period fluctuations in given values. The new correlation coefficient (r*)
was expected1y higher than the previous one and equalled 0.50. This
value reemphasized the existence of a sufficiently close inverse re1ation- .
ship between the short-period ozone fluctuations and the circulation index.
It will be noted, upon reexamining Fig. 4, that the general parallel course
of reve,rse (a) in relation to ozone content at 100 mb level at station Vigna
I '
di Valle shows some clearly defined deviations from this general relation-
ship, particularly from 28 March to 3 April, 14 April to 16 April, and 15
May to 20 May, 1958. In explanation of these anomalies, local gradients
of pre~sur~ (a1) between 40 and 600N were computed from the baric t6po-
graphic charts of the 200-mb surface at the station's meridian which ~s
the same meridian as Rome. To some extent the gradients repres~nt ~
regional. current intensity in a given area. Comparison of local gradients'
with general index (a) showed that, as a rule, changes in ,the former <:or-
responded to changes in the latter. At the same time, well-defined di-
vergences had been observed in changes of both indices. For examp1~,
during 28 to 30 March (0') was on the increase, (a1) was decreasing and
t4e amount of ozone in the Rome area increased correspondingly,' showing
a maximum on 30,..31 March; during the 'same period total regional eir,cu-
1ation over the ,entire hemisphere attained a. general maximum intensity.
A similar sharp'rise in (x) during 14to 17 April at station Vigna-di-Valle
which occurred, contrary to changes in the region circulation index was
in sati~factory agreement with the drop in local index (a~) values during
- 66 -
~
-------
14 - 16 May. However, a significant ozone anomaly which was noted dur-
ing 14-17 April and which could not be explained simply, was not in agree-
ment with either the general or local circulation indices. It should be
noted that during 14 to 16 April a substa.ntial intensification of circulation
occurred over the entire northern hemisphere and highest velocities had
been observed south of 40° N. Individual disturbances in total regional
flow were observed at intermediate latitudes at the same time. It can be
assumed that in regions further down south a subtropical jet stream
acted as a "barrier" during the given period. A jet stream originating
in the polar region was characteristic for the western European area dur-
ing that period. It was traced distinctly on the ZOO -mb surface charts,
and it transported air masses from the Greenland region across Spain
and further along the Mediterranean Sea into the Bl~ck Sea region. Un-
fortunately the E1 'brus and Abastumani stations had no systematically
collected ozone concentration data during that period. Howeve r, ob-
servations at the El'brus station on 15 April showed that the ozone con-
tent at that station exceeded the monthly avera.ge.
It is of significance that the above shown deviations in (x) and (~)
had been observed during periods of strongly increased zonality. It has
been known that the antiphase nature of zonality and meridiona1ity were
disturbed more often during periods of inc reased zonality than during de-
creased zonality (6). Accordingly, zonality intensification during these
periods may be accompanied by meridional motions intensification result-
ing in an ozop.e content increase over the southern stations. Such an
ozone content increase had been observed over station Vigna-di - Valle
when the above shown relationship between the ozone content and circu1a.-
tion index had been disturbed. .
Unlike the case at the southern stations, where most favorable
inverse relationships between ozone content fluctuations and circulation
indices existed at the 100 mb level, such a relationship was closer and
more direct at the ZOO-300-mb level. Fig. 6 presents an example of
variations in the smoothed-out values of ozone content and circulation in-
dex at the 300-mb level and non-inverted scale from 1 January to 31 May,
1958. Despite the fact that the station was situated at the very edge of
the region for which the circulation index was computed, direct relation-
ship between these quantities was clearly indicated. This appears to be
contrary to the inverse relationship for station Vigna-di- Valle which lies
\
generally south of the planetary frontal region. The above difference
can be easily explained: Reykjavik is situated north of the main frontal
region where ozone content is generally increased. Abatement in th~
cir-culation along this region and appearance of air motion meridional com-
ponents augment the southward ozone flow. Subsequently, ozone content
in the north decreased and increased in the area south of the frontal --
region.
_. - ~--
- 67 -
-------
Fig. 6
- 10,'
~. IS/}
('I') ,
I
o
-,
i :f
~ '.uti '
o.
N
o
(,0
, SSO
, , '~ j/\. /:~V~' ,/, . . ",~.
r f'J' (J \ VA\. :. .. ~
1\ ~'\ / " V)J( r \. \)~ t D
V \ \J "V V . 10: ,
8lD 12 y" I
20 JO 10 20 28 10 20 30 IQ 20 J!1._- H- -_'..0..._- 20 - -~IJ- - 131
0-. I 2 DAYS ,OF THE MONTH, II
-----.- -~--- -.--- ~- ----~--- . ,
. ---- _._--~---_._---,--- ------ -~. --. ---~-_._~~-------
CIRCULATION INDEX AT 300 HB LEVEL AND OZONE CONTENT AT STATION REIKJAVJK DURING
--------- m . "- - _~~_NUA~Y-MAY.I958 - -.- - - - '
I-CIRCULATION INDEX AT 300 MB LEVEL 2-0zoNE CON1ENT .
><
80 ~
z
lID
111
106
Some giant peaks in ozone content observed at sta:tion Reikya~ik q.t
the end of January and in February 1958. are of interest: "fromm the analytical
viewpoint pertaining to ozone origins and ozone transfer ,mechanism. 'Sharp
jumps'in (x) from 0.310 cm on 23 January to 0.470 cm and 0.540 cm were
observed on 26 arid 30 Ja~uary, 1958 respectively at Reikyavik duringithat
period after a prolonged depression when ozone content in January stood at
the 0.200 -0.300 cm level. After a drop to 0.400 cm during early February
ozone content again increased sharply to 0.570 cm on 11 Feb-ruary, 19-78.
In the course of the year this increase in ozone concentration coincided with
a period of rapid temperature rise in the polar stratosphere observed) as is
known, in 1-953 (7). For example during 23 January to 1 February 1958 the
temperature over station S P-7 increased by 21, 33 and 35 degrees at the
. I
100, 50, and 30 mb levels, respectively (8). A substantial increase in the
ozone content can not be explained in terms of upper radiation. . Some lin-
direct data point to an increase in solar activity during that period, such as
intensifiCation in polar luminescence, sharp decrease in critical F2 layer'
frequencies observed at some ionospheric stations in general. and over
Moscow, in particular.
, '. . I .
Generated ozone can be used as a source of additional stratosphere
radiation heating. It should be pointed out that parallel with the systematic
and enhanced temperature rise of the arctic stratosphere at the end of winter,
sharp increases in ozone content at high and intermediate latitudes had been
frequently observed. 'For example, all northern European stations recorded
- .
- 68 -
-------
a sharp increase in ozone content during the first ten days of February
1953 by 0.100 cm,according to old coefficients. Apart from the 1953 in-
crease in ozone content noted in Oxfc;>rd, sharp increases were also ob-
served at the end of January and beginning of February of 1951, 1952, and
1954 (9).
5.
SPECULA TIONS CONCERNING THE MECHANISM OF CONNECTION
BETWEEN A TMOS PHERIC OZONE AND CIRCULA TION
On the basis of the broad scope of available data it is possible to
construct the following system of short period fluctuations in ozone con-
tent in different parts of a temperate latitude region during seasons of
large latitudinal ozone gradients. Abatement in regional circulation leads
to an increase in interlatitudinal transfer, resulting in ozone flows south
from ozone -rich northern regions and increasing its content there. Cor-
responding intensification of regional circulation and abatement of meridio-
nal transfer impede ozone transfer to the south causing an ozone accumula-
tion below latitudes 60-70°. Thus, intensification of regional circulation
can be regarded as a unique barrier which impeded the propagation of ozone
from the ozone -rich northern regions to the southern regions. Since a
southward gradient in the ozone content had been noted all the way to the
Pole during the spring months, the closest direct relationship bet ween
ozone content in the northern regions and regional circulation indexes is
indicated during those months. Subsequently, an increase in the inter-
latitudinal transfer attenuated the latitudinal ozone gradient. This pertains
mainly to short period fluctuations in ozone content and atmospheric circu-
lation. However, this concept of mechanism related ozone with circulation
may be manifested also in certain dependences of ozone on seasons and
latitude.
It is conceivable that these special circulation fea.tures can explain
the encountered divergence of latitudinal gradients recorded in IGY materi.
als which are smaller in areas of strong cyclonic activity, according to
data obtained by ~eikyavik and other Western European stations, and con-
siderably greater deep inside the continent, as shown by data obtained at
Dixon Is. and Alma-Ata stations [12J. On the other hand, it is well known
that the interlatitudinal transfer of air masses was much more pronounced
over Western Europe than in the heart of the Eurasian continent. Proceed..
ing from the above shown relationship between the ozone content and atmos-
pheric circulation, gradient divergence can be adequately explained by the
attentuation of interlatitudinal transfer below the continent. Furthe r verifi-
cation of this conclusion calls for establishment of ozonometric stations in
the middle of the continental portion .of the USSR, 1;>etwe.en 50 and 60° N. It
has been known that a sharp discontinuity occurred in the mean meridional
ozone distribution soufh of 30-35°N, where ozone content sharply decreased,
i. e. in the northern hemisphere. A powerful regional circulation system,
the so-called, subtropical jet stream developed in this belt at the 300-200-mb
- 6 9- ~
-------
-.--
- "- _._~._----
level. Its existence reflected a considerable temperature contrast be- ,
tween the tropical ,region and the temperate region, in which a continuous
air mass transfer to and from the arctic regions took place. In accordance
with this, ozone transported by the northern air masses, failed to pene - ,
trate further south, resulting in a sharp ozone gradient at these latitudes.
According to Kh. P. Pogosyan [10, p. l35J a difference in the geo-
potential between the consecutive ridges and troughs in the 500-mb sur-
face, characteristic of the interlatitudinal transfer intensity, is 60-l!;>9 at
, latitudes from 40 to 60°, and only 14 -28 at latitudes from 20 to 30°. 'A
, similar difference probably existed als 0 in the lower s_t}"atosphere. This
phase of the circulation explained the sharp ozone radiant observed in the
30 -35° N belt, attributed earlier to a change in the height of the tropopause
at, these latitudes, and to the powerful effects of tropical troposphere on
ozone decomposition. However, it is not possible to consider as tropical
the troposphere over Kabul or Northern India, where considerably less
ozone was observed than at the southern stations of the -USSR, or Italy.
In addition, recent airborne ozone measurements made by chemical'meth-
ods failed to show any sharp gradients in ozone content related to the tropo-
pause [11J. This means that the process of ~zone decomposition was not
as accelerated under the tropopause as it was under the stratosphere.
The interesting phase of seasonal changes in ozone over the northern
hemisphere during 1957-l958'was the occurrence of a secondary summer
maximum, observed at a number of stations, such as Alma-Ata, Abastu-
mani, Voieikovo, Ukkel, Reikyavik, and others. The mean monthly ozone
'content at these statlons increased during June-July by 0.015-0.020 cm
with respect to total annual ozone changes during the pre'ceding and fol-
, lowing months. A similar summer maximum was frequently observed
during Augus~ in 9:#,ord and in other places, showing a clear connection
with latitude. It appeared;at southern stations, such as Alma-Ata, in June,
and by August, moved north towards Reikyavik. According to Paetzold,
seasonal changes in the photochemical equilibrium gave rise to ozone
fluctuations with a maximum in the summer and an amplitude not exceed-
ing 0'-02 cm. Consequently, the ozone maximum should rise in July by
0.004 cm above the maxima of the adjacent months. The actual observed
summer maximum was, as can be seen, consid~rably higher than the
above value. It can be assumed to be advective in origin and associated "
with the increasing' meridional component of the general atmospheric
circulation during summer. In fact, regional circulation decreased
toward the midd-le of summer and, simultaneously, conditions were
created favorable to the intensification of the meridional circulation in-
cident to the rise in temperature gradients of continents -oceans.
IGY oz~ne- data [12J showed that a secondary maxiI?um in the
daily ozone variations was observed at all stations after the winter-spring
period which was particularly clear during June-August at Southern sta-
- 70 -
-------
tions of the intermediate latitude. An increase in ozone 'changes was defin-
itely caused by an increase in the interlatitudinal transfer. A similar
secondary maximum in ozone content changes over Oxford in June and
over Aroze in July was indicated by Normand in 1954 [9J. IGY data also
indicated that on the background of a gradual drop in the meridional ozon.e
content gradients in the interval between spring and fall the re was also
noted a shallow dip during the summer months of June and July (See Table
4). The latitudinal gradient decrease during summer apparently pointed
to the advective nature of the secondary summer maximum in the southern.
portion of the temperature belt. Preliminary conclusions drawn. from the
processed IGY ozone data suggest a relationship between the summer ozone
maxima and the meridional circulation intensifications.
Table 4.
+------ .-.---- ~ -------. ---. - - -
-3
INTERlADITUDINAL OZONE CONTENT GRADIENTS IN 10 CM
STATION
-------- II I III I IV I I VII VII'IVIIi/ IX I I
MONTHS V X XI
~
REIKJAVIK . +77 + 17 +22 -i-7 -14 +18 +:\11 .;-6! -:-27
UKKL
VOEIKOVO +101 ~IOI +82 -: 80+54 +70 -;-57 +36\.2°1--8
EL'BRUS
VOEIKOVO
ALMwn +155 +17li-153 -!-103 +83 -i!J9 +1041-:.551 T~81 ;-20
6. SOUTHERN HEMISPHERE CHARACTERISTICS
.It has been known that circulation conditions in the northern and
southern hemispheres varied considerably. The more uniform surface
J of the southern hemisphere and the presence of the high and cold Antarc-.
tica gener~ted regional circulation in the southern hemisphere which was
twice as intense as in the northern hemisphere. Any relationship between
ozone content and circulation should manifest itself as a recordable differ-
ence in the condition of ozone over the northern and southern hemispheres.
Data in Table 1 show that the ozone quantity was considerably greater in
the southern hemisphere than at similar latitudes of the northern hemi-
sphere* This is also shown by the large latitudinal ozone gradient exist-
ing in the southern hemisphere. This gradient can be explained by the
fact that a considerably stronger regional circulation in the southern
hemisphere impede9 ozone penetration from the temperate into the tropi-
cal1atitudes, enhancingits accumulation at the higher and intermediate
- . I
*Further in the text identical seasons of both hemispheres, for example,
December-January at the northern hemisphere and June-July at the south-
ern hemisphere, etc.
- 71 -
-------
latitudes. It is conceivable that a smaller annual amplitude of ozone con-
tent at temperate southern latitudes was associated, if only partially,
with a lesser variability in the circulation over the southern hemisphere
than over the northern hemisphere.
The secondary ozone maximum observed in the northern hemis-
phere during summer appeared less pronounced in the southern hemi-
sphere. This was illustrated by Goetz' earlier diagram and by the obl- .
servation data at Aspendale and Makkuori. This shows that ozone could
be found during short intervals in the summer months in the northern
hemisphere in greater quantities than in the southern hemisphere. Since
the photochemical ozone equilibrium conditions' were almost identicai in
both hemispheres, the difference in their maxima may be a reflection of
their advective origins, and may indicate that the intensification of meri-
dional circulation during summer was expressed more clearly in the
northern hemisphere. The foregoing is another manifestation of the I
close relationship which probably existed between ozone quantity and'
. general atmospheric circulation. .
I
Conclusions arrived at by this author concerning this relationship
are not final. Accumulated data which are still being processed will pe
critically reexamined. It is hoped that these conclusions may suggest
new courses Jor future investigations of relationships between atmos-
pheric ozone. and weather conditions.
BIBLIOGRAPHY
i. i> a e t z~ 1 nl.--K-- -~e\\'-experinlentaT -and-ttiro~~ti~al in\"estigat;o~s
of the atmospheric ozone layer. Scient. Proc, Int. Ass. Meteor.. Rom~.
195-1; London. 1956. .
2. !\ (J r m 3 n d. Atmospheric ozone and upper air conditions. Quart. Journ.
Roy. Met. Soc.. 79. N! 339. 1953. .
3. I) e 311 e p x II II i. W. A. OJOII0x,eTpll'leCKlie .lallHLle 00 AJlMa.ATe. 8 co-
O()CTall.leHJII~ c lIeKOTOpbl\1I1 \leTcopo.l0tll'lCCKIIXIII TaKTOpa\llI. TpY.lbl Ka-
33\0:- 1i111.\\11. IILlO. 5, 1955.
4. .\\ i yak e \'. "a w a'l11 u r a K.Studieson atmosperic ozone at Tokyo.
. Joutn. Met. Soc. Japan. Ser. 2,.32. N1 4. 1954.
. 5. .13 e p 3 e e II C K II il B. .1.. .\\01111 H A: C. TllnOAbie cxexlbI oOwer. UIIPKY'
.1Rilllll aT\")C~>epLi II ceUepllO\1 no.1Y III a pll II II JlIl.leKC UllpKy.1RUIIiI. H3Bl'CTHIi
.\H CCCP. cep. reo1j>1I3.. N! 6. 1954.
..- ... ----
..~~- ---- -
~._--_.-
- 72 -
-------
AEROSOL ORIGIN OF ATMOSPHERIC OZONE
(A hypothesis)
V. D. Reshetov
1.
Introduction
.
. The modern theory of photochemical atmospheric ozone origin fails
to explain some facts and ozone properties. For instance, in the south,
where ultraviolet radiation is greater the amount of ozone is smaller than
in northern regions, where radiation arrival is considerably less, and
during the polar night is nonexistent. It also fails to explain annual
changes in ozone content with a maximum at the end of winter and begin- ,
ning of spring, and high ozone content in the lower portion of the strato-
sphere and troposphere,. since that portion of ultraviolet radiation, whicb
generates atomic oxygen, is absorbed totally above the 20-km level. It
is also difficult to explain the appearance of a special ozone maximum ob-
served sometimes in the tropopause, or the considerable fluctuations in
ozone content associated with weather conditions. The latter causes an
increase in the ozone content during a foehn birth of an anticyclone, dis-
integration of a fog after passage of fronts, and during some other atmo-
. spheric phenomena. [1.. 2J It is conceivable that atmospheric ozone is
also gene,rated under the effect of other factors, which, in a number of
cases, may be highly important. A hypothesis is presented below on the
generation of atmospheric ozone by atmospheric aerosols from water and
atmospheric oxygen during selective sorption and desorption at the aerosol
surface.
2. FORMATION OF FREE OH RADICALS ABOVE THE SURFACE OF
A WET AEROSOL AS .(\ RESULT OF SELECTIVE SORPTION AT THE
WATER -AIR BOUNDARY
The atmosphere always contains so-called condensation nuclei
surrounded by a film of moisture and condensation products in t1~e form
of incipient drops. The number of condensation nuclei in the air of 10-
20 km visibility is 10-20 x 103/cm3 close to the ground; it decreases rapid-
ly with height, although at 5-6 km it still remains' of the order of 102 Icm3.
Considerable condensation nuclei and products are collected under the re-
straining layers and inversions in the troposphere, indicating that atmo-
spheric aerosol concentration increased sharply in these regions. When
radii' of atmospheric aerosol particles in "clean" air are of the order of
llJ and in clouds and fogs approximately 10 1J, and their concentration ap-
proximately 103 Icm3, then the total surface over which sorption phenom-
ena occur is of the order of 102 and 103 km2 for every km3 of "clean" air,
- ---- - - -- ~-- -- -.---- --
- 73 -
-------
-~.
and 104 kmz for every km3 of ~,louds or fog. Accordingly"
face of an atmospheric aerosol is in the order of 106 times
area of the Pacific Ocean.
the sorbing sur-
greate r than the
Examine the nature of sorption at .the water-air int,erphase appli-
cable to atmospheric aerosol. Assume that water was always partially
dissociated into Wand 011 ions (distilled water pH=7) and that water ad-
sorption potentials were different with respect to different ions. Under
such conditions some ions may be drawn into the inner portion of the sur-
face layer, and some will appear closer to the free surface and to the,
so-called, double electric layer creating a corresponding electrokinetic
, potential at the water -air interphase. These phenomena with respect to
the atmospheric aerosol were studied insufficiently in the past. Aiken [3J
studied the behavior of some aqueous solutions and conceived 'of the pro-
cess as an ejection of large ions by the cohesive forces of water molecules
upward to the water surface. In the case of water, such ions are the 011
ions. ' Bach and Gilman [4J studied electrophoresis of air bubbles in dis-
tilled water and demonstrated that air bubbles became charged negatively
with respect to the positively charged water. Accordingly, the 011 ions
were distributed predominantly closer to the inner water surface, which
facilitated their separation and escape together with the vapor molecules
under the thermal motion of the molecules.
Coehn and Neumann [5J also studied electrophoresis of air bubbles
and noted that in water acidified with high concentration of hydrogen ions
air bubbles became positively charged while in alkaline water with an in-
creased content of 011 ions the bubbles became charged. Apparently,
the difference in the hydroxyl and hydrogen ion concentrations in the outer
and inner regions of the surface, layer of distilled water was relatively
small and could be bridged by a general increase in the hydrogen ions con-
centration during acidification, etc. The present author studied the uni-
polar charge of artifit:ial aerosol drops as a function of pH value of an
atomized solution and found [6J that neutral aerosol existed only at pH ~ 5.
At pH < 5 the aer'osol drops became charged negatively with respect to
air. At pH > 5, including pH = 7, the aerosol drops became charged pos-,
itively with respect to air. Electrophoretic observations of aerosol
particles in a fog, chamber (Fig. 1) showed that when pH of an atomized
solution was Ie s s than 5, particle s settled predo minantl y at the anode,
and for pH> 5 and pH = 7, predominantly at the cathode,. 1. e., the par-
ticles became charged positively with respect to ai,r.
Apparently, there occured a selective sorption transfer of ions
between the aqueous solution and the air. At pH = 7, hydroxyll0ns con-
c'entrated (ad~orbed) predominantly at the outer portion of the surface
layer. If hydroxyl ions concentrated on the surface of aerosol particles
in the atmosphere, then, in accordance with the general' ~aws of sorption,
their balanced concentration there should be greater, the, ,lower the tempe r-
- 74 -
-------
I -
t~re. Thus, when an aerosol deviated from an equilibrium state due to
i~crease in temperature or' reduction in relative humidity, desorption oc-
ctirred in the course of which water molecules and OHt- ions were stripped
from the surfac~ of the aerosol particles. At the same time the pH of the
stmospheric aerosol water drops decreased, the water acquired acidic
properties, and the aerosol acquired a positive charge [7J.
The sediment which "washed out" the atmo-
spheric aerosol, is generally of an acid re-
--~ - - ------------ ----"
action. The data on the pH value of atmo-
spheric precipitation observed at almost 20
stations during the winter months --January
and February during 1954 - 1958 - and regu-
larly published in the Tellus journal [8J in-
dicated that pH of atmospheric precipitation
was less than 7, i. e-., aqueous atmospheric
aerosol was acidic. It was also found that
the hydroxyl desorption from the aerosol
surface occurred with greater intensity in
the more southerly warm regions; toward
the North, where the air was colder, de-
sorption was less intense; the pH value of
the precipitation was higher there but re-
mained below 7, a fact which indirectly
----.----- ---" -- -- confirmed the postulated low-temperature
effect. Taking into consideration the fact
that the principal source of atmospheric
aerosol index was the Pacific Ocean surface
the water of which was alkaline [pH'" 9J
then the intensity of this process became even more apparent.
Fig. 1
'""
m/f
5 0
"
0
3
2
o
, 2 J "
5 r: 7
AAf10 BETWEEN AEROSOL PRECIPITATION
GRAVIMETRIC RESIDUALS ON ANODE MHr
AND CATKODE Mr AT DIFFERENT SOLUTION
pH .
~-;--
N. N. Semenov [9] had shown that the change of hydroxyl ions 011
into free OH radicals could take place by one electron transfer to the elec-:
trode or to another particle. An aerosol droplet (particle) could serve as
an electrode, retaining an electron from one of the hydroxyls ejected to-
. gether with vapor molecules. Directly above the aerosol particle there
- -- --appeared - a number of free OH radicals. These could also be generated
in the surface layer of the droplet by the removal of electrons from
hydroxyl ions at the surface, approaching or recoiling back to the gas and
vapor molecules. In this manner, the free radical formation at the sur-
face of aerosol particles and in air proceeded through two stages
. --c-- ~-----H~o+-M;~-H~+6j{- + M~-
r
I
..- - -- ..1.
. ,(I)
and
OH- + Mz :: OH -;- M2"'
: (2)
- 75 -
-------
,
-~..-'
3.
O'ZONE FORMATION IN A WET AEROSOL INITIATED BY FREE
OH RADICAL<;
The appearance of free OH radicals on the surface and beneath
the surface of aerosol droplets opens the way to a number of chemical re-
actions. A possible reaction is one of the hydroxyl with water molecule!!
(vapor) and oxygen to form ozone 03 and hydroxonium H30 according to
the general equation: .'
------_._~----~------------_._--_.~-
OH + H20,:... Ol~' M ~ 03'~~ H.O + M.
. This reaction can apparently develop in the following manner:
-------------_._~
------------
OH -:- 01 ~ O. + H
H -~- H20 + M -.. HHzO + M,
~3)
.. -----.---.--. -
or:
OH ,:- O2 ~ OIH + 0, I
o + O2 . f- M ~ 0. + M,
02H .c. H20 ~ HH20 + 01. .
(3")
This reaction should be accompanied by the liberation of 35 kcal/mole.
The amount of generated ozone must be proportional to the number of ap-
pearing OH molecules, that is, it should increase with increasing water
dispersion, since this was accompanied by an increasing droplet surface.
The generated ozone amount should also increase in the course of aerosol
heating and vap?rization, since this increased the emission of hydroxyl
ions and radicals into the air. Another reaction was also possible which
involved the formation of hydrogen peroxide in the aerosol droplet par-
ticles: .
._--~--~------~-_.'-------'._._-'------'--'- --- .
OH + OH + M --+ HIO, + M.
(4)
I
;
Such association of free radicals was accompanied by an energy rel~ase
of approximately 50 kcal/mole. It is well known that hydrogen peroxide
was an unstable compound which decomposed at elevated temperatures.
Consequently, a larger amount of hydrogen peroxide 'can be formed and
accumulated in an atmospheric aerosol at lower temperatures. Hydrogen
peroxide may exist also in a given equilibrated amount of air over the
aerosol pfLrticles' surface.
, "
In studying rainwater,-' snow and fog; Ivanov [lo/J detected the
presence of hydrogen peroxide. In ordinary rainwate'r it amounted to
approximateiy 0.004 mg/liter; it may, reach 1 mg/li in thunderstorm
water; snow, ice and fog contain approximately O~ 05 mg/liter' of hydrogen
, peroxide:', tfnder the influence of heat or light, especially ultraviolet light,
-:-76::-'
-------
hydrogen peroxide decomposed to form free radicals according to the fol-
lowing chain reaction [9J:
~ - -'---.,.-..-- -.. --... -- -_. - . -.
OH -+ H,O, - HO, + H,O.
HO, + OH - H,O. + 0;-,
O2 + H20, -+ OH- + OH + 0'0
(5)
Thus, above the aerosol particles I surface hydroxy1s appeared in the form
of chemically active free OH radicals and as OIr ions' and hydronium
H30+ ions. This leads to the formation of appreciable ozone amountsac-
cording to the above indicated reaction (3).
It should be noted that reaction [5J was brought about in the labor-
atory and in industrial! practlce- by "heating and by photolysis involving the-
use of powerful pulse sources [9J. In the course of a flash tube, pulse
lasting several microseconds, free radicals were generated sufficient
to initiate a chain reaction of hydrogen peroxide decomposition, which
persisted for a considerable length of time. Since hydrogen peroxide
molecules were larger than oxygen and ozone molecules and were charac-
terized by low energy of the 0-0 bond (54 kcal/gmole) they dissociated
not only under the action of short-wave ultraviolet, which also produced.
the decomposition of molecular oxygen and ozone, but also under the ac-
tion of the relatively long-wave region of the ultraviolet s_pectrum, which
penetrated below 20 km into the stratosphere and even into the troposphere.
Hydrogen peroxide also decomposed under the effect of short-wave visible
- spectrum. It will be shown later that these conditions played an important
part in the characteristic atmospheric ozone distribution. In addition to
the above described possible ~~c::ha.~isl!ls of atmospheric ozone formation,
ozone may also form' at lower temperatures in the aerosol water itself.
It may be assumed" that water dissociated partly into hydrogen and atomic
- oxy"gen, as shown below:
---~----_.._-_.
-----~ -- -.----- --_. ---------
- ~_. - _. --
H O..l-- M -. H+ .!- OH-...L M -. 2H+ +0- + M-.
2 I A ~ . . +-
(6)
Even if the number-of nascent-oxygen atoms in water was small,
they may still lead to the accumulation of a certain amount of ozone in the
aerosol water according to the following equation:
- - - --- . -- .-
0, + 0- + M = 0, + M-.
(7)
-,
Since ozone solubility in water is by a whole order of magnitude more
- soluble than oxygen, ozone should gradually accumulate in aerosol water,
especially at lower temperatur~s. It is sufficient, however, to heat a
cold aerosol to produce immediate desorption of ;the dissolved ozone into
air, since with rising temperature the solubility (adsorption) of ozone in
- 77 -
-------
~.
water sharply decreased. This occurred also when aeros..ol elements
partly or completely evaporated with air drying or with a fall in relative
humidity. In comparing the probability of reactidns (1-5) and (6-7), it
can be assumed that the reaction course indicated in(1-2-4-3) will pre-
dominate, that is, a preliminary formation of hydrogen peroxide will takf;
place. The course of reactions (6-7) in water is hindered by the low
probability of oxygen atoms recombination with 92' Reactions (6 -7) re-
quire that an appreciable energy barrier (219 kcal/gmole) be overcome,
and are connected with the solvation energy of highly charged ion-mole-
cules~. At the same time reactions (1-4) require that a much lower
energy barrier should be overcome of no more than 118 kcal/gmole. Fur-
thermore, each stage of the reaction required no more than 30- 60 kcal/
gmole, 1. e. no more than 1. 5-3 ev of the kinetic energy of. colliding mole-
cules. All above possible modes of ozone formation in atmospheric air
were contingent upon the concerted action of dissociation (association)
and selective sorption brought about by thermal or radiant energy. They
all lead to an increase in air ozone content in the course of heating or
vaporizing an atmospheric aerosol. This occurred when ozone, hyd,roxyls
and hydrogen peroxide were desorbed from the aerosol surface. Hydrogen
peroxide, in decomposing under the influence of. heat or light, formed.
cJ:iemically active free OH radicals which produced ozone. according to [3J.
,The following experiments were performed to illustra.te the action
of the above ozone formation modes. A moist linen cloth 1. 5 x 2.5 m
simulating the aerosol particle surface was suspended outdoors in th~
shade, early in the morning when the temperature was approximately
5° C. After one hour exposure the cloth was brought into the laboratory,
where the ozone concentration in air was previously 4.5 Y (1 Y = 10~ g of
ozone per 1 m3 of air) (see the article by Britayev in this collection).
The cloth was brought in from outdoors into the laboratory; this increased
the amount of ozone in the room to 7.8 Y (Fig. 2, curve 2) (each ozone
determination required approximately one hour). After one hour, when
the cloth was almost dry, the ozone content was still 5.6 Y (curve 3-;~Fig. 2).
The cloth was removed, and after two hours the oz.one concentration, in
the room dropped to 3.8 Y (Fig. 2, curve 4). In this manner the intro-
duction of a wet cloth brought into the warm room from the "cold" out-
r .. I
doors increased the ozone concentration by 100 -150%. This apparently
confirmed the existence of ozone formation processes which accompanied
the desorption of a .cooled atmospheric aerosol on heating and vaporiza-
tion. In another experiment a somewhat smaller cloth (l.x 1. 5 m) was
moistened with a 3% hydrogen peroxide solution directly in the room. It
can be seen from Fig. 3 that the amount of ozone in the roort;1 (curve 2,
Fig. 3) increased by approximately 2000/0 compared with the initial con-
centration o'r with that after removal of the cloth.
I
It may be concluded from this experiment that hydrogen peroxide
desorption into air led to ozone generation.
- 78 -
-------
Fig. 2
Fig. 3
--_.. ..- ~.
.------
. ---..--'-
----_._~--- ---~----- --. - ---- .- -
--. ----
0] 6,.
DJ 6,.
G
5
Ii
--.,;L'1/0Z0NATOR B
rz----;\
I I I
, ~
6/0: J 1
I L_____-'!!!"
IJ
-----
G
5
Ii
J
iOZONATOR PB
r--f-,
I 2 I
I I
. 1_"':--___---
: : 22 OCTOBER, I~
I I
, I
LL
J/u
, J
2
I
()
-----
2
,
~/o
22..MAY, 1959
o
2
j
4 . 5-
---.---
HOURS
Ii ,5
-. _n_-
HOURS'
'-OiON-E-GENERATED~' -1-riDOO~S -iiYA SHEET
, MOISTENED WITH HYDROGEN PEROx.-
IDE OLUTlol1
I & 3 OZONE CONTENT PRIOR TO SUS-
PENDING MOISTENED SHEET AND TWO Hitfi.
LATER CORRESPONDINGLY; 2 - OZONE CON-
TENT 0.5 M FROM THE SUSPENDED wE.T
SHEET. SHEET MEASURED 1.0 X .l.5 M;
I NDOOR ROOM CAPACITY :Qrt'
2
J
OZONE GENERATED INDOORS BY A WET SHEET
BROUGHT IN FROM OUTDOORS
1 &4- MEASURED OZONE CONTENT INDOORS
PRIOR TO BRINGING IN THE WET SHEET AND
2 HRS LATER CORRESPONDINGLY; 2 & 3-TWO
SUCCESSIVE OZONE DETE~MINATIONS MADE I
M FROM THE WET SHEET \EACH DETERMINATlaN
CONSUMED 40-60 M-' N.) I HDOOR I EMP. +20 ,
OUTDOOR TEMP. +!f!. W!T SHEET MEASURiD
.!.!2.~~~ INDOOR ROOM CAPACITY :Q ~.,- ,
. ---- -. - - -- -- ----- ---
It should be remembered that hydrogen peroxide existence in water ob-
tained from an atmospheric aerosol had been established previously [lOJ.'
The one objection that can be made is that hydrogen peroxide decomposed
faster in the presence of organic compounds. In dealing with an atmosphe::,-
ic i.aeros'ol it musf be remembe'red'that It contained an appreciable amount
of organic compounds.
Another experiment consisted 9f outdoor ozone measurements in
front of an open window in a warmed room, so that the aerosol entering
from the outside was heated and partly vaporized, forming ozone. This
experiment lasted two days. Fig. 4 shows that the amount of ozone in
the room prior to opening the window was substantially lower than outdoor.:;.
After opening the window it ~ncreased and reached a level ~gher than that.
outdoors. T~~s ~~_nfi~~ed ~.e a~sum1?!~()X:,s made concerning ozone gene~a-
--- Hc)n 'when-atmospheric aerosol was heated 'and vaporized. Assuming that
the aerosol simply desorbed the ozone adsorbed outdoors, then the amount
of ozone in the room with the opened window could not exceed the amount
of ozone outdoors, especially when ozone rapidly decomposed at the walls
of the room.
In 1952 Regner measured ozone in the ground layers of air in the
Los Angeles area (11) and found increased ozone concentrations, exceeding
those 'observed in the free atmosphere. The sharp increase in ozone in
this case was observed from 8:00 to 12:00 A.M., 1. e. when the ground
-79 -
-------
layer of the aerosol was heated and vaporized. Regner suggested that
apparently "adsorption of ozone on hygroscopic aerosols occurred, which
led to a temporary formation of hydrogen peroxide. This substance was
liberated in the morning hours, when the fall in relative humidity pro-
duced vaporization of the aerosol. It was followe4 by photochemical dis-
sociation' which could again lead to ozone formation. "
Fig. 4
-- - -- -. -----
T -
6
.-------..-------------- ----_._- -- - -----
- -- -----
..
.J l'
5
4
J
2'
'\n" " ,
o
-'-. ~
--;X:~-:-"'"
, ',-
,t, '...
, .0, ----
/ .
-'"
o
, -
.. '" III '"
..
'" .:
. \ .
..",\7-"
10 12 JIL/5_. .18_20_22 2" OLllLOLDR_ID.'2, (II
. 24 OCTOBER 25 OCTOBER. ' TIME
.;/
-..- ---~~- --~---------r---
. OZONE GENERATED BY FLOWING IN OUTDOOR AIR'
: DOTTED LINE CURVE - INDOOR OZONE CONTENT I M rROM AN OPEN'
WINDOW; SOLID LINE CURVE - STREET AIR OZONE CONTENT. OUT-
OOQJl ANQ ANOOOR TEMPERATURE DIFFERENCE WITHIN LlM,TS O,F, 5-
., . --'-~-~-~tI!~.?OR- AIR AEROS~- OF SMOKY ORIGIN.H,,
It can be seen from the above that the trend, of sci~ntific thought
, dealing with tropospheric ozone genesis proceeded in parallel ~aths in
different countries, even though serious disagreements existed. Regner
, believed that "ozone formed directly from substances contained in aerosols
accumulated under descending inversions". He considered an atmospheric
aerosol to be a storage depository rather than the generator of atmospher:'.c
ozone, and believed that it was impossible to dispense with photochemical.
. ,
dissociation which can onlylnfe'nsHy the process. If ozone were not form-
ed in the troposphere, it would be rapidly destroyed by the presence ~f or-
ganic substances in the atmospheric aerosol and on the ground, since it
was a strong oxidizer. Existing ozone concentration in the troposphere
and at the ground was determined by equilibrium between the actual mechan-
ism of ozone formation and the factors leading to its decomposition. '
4.
CHARAC TERIS TICS OF A TMOS PHERIC OZONE PHENOMENA
An attempt will be made to explain the knoWn facts concerning at-
mospheric ozone from the viewpoint ,of the above hypothesis. For ex-
a.mple, the appre'ciable increase in ozone during a foehn [lJ or during
-~~!~~xclogenbsis [2J, du:-in~ ,~catt~ring~,f.a fog or mist in the morning
- 80 -
-------
[llJ, during dissipation (vaporization) of low ground fog [2J can be ex-
plained by heating of descending air and, partial or full, vaporization
of ground level aerosol. Desorption of 03, H:a02 and OH from aerosol sur-
face leads to a .sharp increase of ozone in air. Table 1 presents an ex-
ample of sharp ozone increase close to the ground during cloud dispersion
(a rising fog which had existed previously under an anticyclone inversion
for 2-3 days.) The investigation was conducted by Teichert and Warmot
[2J.
Table 1
----- .- ---_. ---- - -- --..-----
~----_.._-- . -.
OZONE CONTENT ON 18 MARCH 1954
TIME
OZONE
o HRS. 6 HRS.
8 9
HEAVY FOG
9 HRS.
21
12 HRS.
o
15 HRS. 18 HRS.
30
CLEAR WEATHER
25 HR S.
3
As soon as the stratified cloudiness disappeared, the amount of ozone
increased by a factor of 2 -1/3 and then again decreased, supporting the
. above hypothesis. This viewpoint can also explain the increased ozone
content near waterfalls and in the area of ocean spray [12]. In general,
an increased ozone concentration can be expected above those sections
of the ocean where air tempera'ture is higher than the water tempera-
. ture, and since the latter was alkaline (pH approximately 9), the desorp-
tion of OH radicals during the evaporation of ocean spray led to ozone
formation. Ozone amount should increase during the daytime breeze,
especially in the first half of the summer, when the water temperature
is lower'than the air temperature.
Teichert and Warmbt [2J detected an increase in ozone after the
passage of a front, and M. A. Konstantinova after a. snowfall [13]. To-
gether with a high hydrogen pe roxide content in rainstorm water. (approxi-
mately 1 mg/li) and in snow water [lOJ, this confirmed the dissociation
of hydrogen peroxide and hydroxyl radicals; and since thunderstorm rain
was formed in high atmospheric layers from rather cold upper cumulus
cloud parts, it also indicated a vigorous desorption of hydrogen peroxide
and hydroxyl radicals from the drops ~t higher temperatures followed by
ozone formation.
Now, examine ozone distribution, especially its secondary maxi-
mum, at an altitude of 11 km from the viewpoint of its relationship to at-
mospheric aerosol (Fig. S). The ozone maximum at an altitude of 11 km
can not be explained by the action of ultraviolet solar radiation. It may
be explained better. by the accumulation of condensation products under.
the tropopause. Cirrus clouds are often encountered here, 1. e. a cold
atmospheric aerosol was present. Turbulent diffusion carried the cold
aerosol upwards into the region of tropopause inversion and into the
lower stratosphere, leading to a partial or appreciable vaporization of
- 81 -
-------
- ~ ~ ~ - -~-
moisture from aerosol particles and to increasing ozone concentration.
similar phenomenon could'be expected in inversions above stratus and
stratocumulus clouds in the atmosphere.
Fig. 5
Z8"...
...
...
..
...
...
"'...
"'"
30 "...
"'''........
75t?J:~~ )
"I'
"
I,
t-/ :
, I
, I
, I
, I
, I
, 2~
" I
,
20
10
,/ TROPOPAUSE
,/ ----- .
, --- - -
, .'
, ,
, '
,
I '
,
o
C
o.OIZ C'M/":M
0.01< 0.08
A
Actually, simila! considerations. to-
, gether with those, of photochemi~al equilibri\;lm,.
~ couid be used to explain the main. ozone maxi-
mum at 25-30 km. It is here that significant
temperature inversion begins and continues up
to 50 km [14J. The main aerosol vaporization
must take place here, accompanied by signifi-
cant ozone. hydroxyl and hydrogen peroxide
desorption with a subsequent. increase in ozone
concentration. This is particularly true in the
lower part of the zone. i. e. at 25-35 km. It
should be remembered that Driving [15J examined
the combined data on ozone and nacreous clouds
of northern Norway and noted a connection be-
tween maximum ozone altitude distribution
and the maximum repeatab~lity of nacreous
doud~. Recent (after 1957) determi.nations
nave frequently pointed to a lower position of
the ozone maximum than had been previously
assumed, and to the frequent repeatability of
the maxi~um in lower layers of the strato-
sphere directly above the tropopause. One of
such examples of Central Aerological Observa-
. tory measurments during the IGY is hown'in Fig.
2. The present author's explanation of such
phenomena was presented ab9ve~
- -
VERTICAL OZONE DISTRIBUTION II
NACREOUS CLOUDS ACCORDING TO
DRIVING (15) AND AN EXAMPLE OF
HIGH OZONE CONTENT BELOW 20 KM
ACCORDING TO MGG AND CAO DATA
I-VERTICAL OZONE DISTRIBUTION
DURING DAYS Or HIGH TOTAL OZONE;
3-DURING DAYS OF VERY HIGH TOTAL
OZONE; 3-REPEATABILITY OF NACRE--
DUS CLOUDS; 4-0ZDtlE DISTRIBUTION The Presence of hydroxyls discovered in
OVER MOSCOW ON I 7/1 v, 1958 AFTER
CAO DATA PROCESSED BY A. P. Kuz- recent years at an altitude of approximately
- -- NETSOV ' 80 km, may be the consequence of selective
------- desorption of OH radicals from the moisture
of noctiluc~nt clouds, as described above. It had' been assumed that
notilucent c"1ouds consisted of minute water elements in liquid.and solid
states. Shklovskii [11J assumed that water may be formed here from the
discovered hydroxyl ozone, as shown in the following equation:
I'
20H + 0, = HIO + 20'0
In this reaction, however, the equality sign may be replaced by the
equilibrium sign, i. e.:
--
H20 + 201 ;:. 20H + 0..
.,
- 82 -
---- ----
-------
since the reaction may proceed in either direction depending upon external
conditions. Therefore, the appearance of hydroxyls at these altitudes may
be the consequence of the above selective desorption of OH radicals from
the elements of notilucent clouds when the latter are found in the tempera-
.ture inversion located above, in accordance with the theoretical explana-
tion of hydroxyl and ozone formation.
Examine ozone distribution with latitude and its annual migration
in the northern hemisphere. These exhibit two important characteristics:
an increase in general ozone content in the atmosphere toward the north
with a maximum near 700N, most sharply expressed toward the end of the
. Arctic night, and a general increase in ozone in the early sp:i:ing" March
and April, over the entire hemisphere. It has been known that the greater
part of ozone was found in the stratosphere. A comparison of these facts
with the meridional temperature profile along the 80° meridian according
to Kokhanskii [18J, indicated that in April there existed near the area of
700N in the stratosphere a well-pronounced source of warm air, warmer
than either to the north or to the south of it. At the same time, rather
low temperatures were observed in the troposphere of this region. The
course of Arctic cyclones passed through this area close to the earth eject-
ing into the upper troposphere and into the lower stratosphere large. .
amounts of a rather cold aerosol. When the aerosol entered into the upper,
relatively warmer, part of the stratosphere, it vaporized, increasing the
amount of atmospheric ozone. In February and March the ozone maximum
near 700N coincided with the end of the Arctic night. At this time of the
year the sun appeared for brief periods above the horizon and penetrated
the gloom of the Arctic night. Taking into account the sun's low altitude
and the large number of "penetrated atmospheres", it appears unlikely
that the'ultraviolet rays which dissociated oxygen penetrated below alti-
tudes of 40-50 km. The assumption that an appreciable amount of hydrogen
peroxide accumulated in the cold aerosol of the Arctic air as a result of
hydroxyl association by aerosol particles at low temperatures, leads to the
following conclusion: the short-pe riod illumination of the polar atmosphere
by solar rays, which contained only visible or relatively long-wave ultra-
violet components, acted similar to a flash tube and produced a chain of
hydrogen peroxide decomposition into free radicals. Due to their high
chemical activity the latter resulted in ozone formation according to equa-
tion (3).
General ozone increase in the atmosphere of the northern hemi-
sphere in the spring, particularly in moderate and high latitudes, can be
explained on the basis of aerosol vaporization and of the snow cover dis-
appearance, which is accompanied by hydrogen peroxide liberation into the
atmosphere [lOJ.. In the .opinion of the present author the aerosol origin
of tropospheric ozone, including the secondary ozone maximum near 11 km,
appeared-quite credible. This is further supported by the fact that any
direct action of ultraviolet light was apparently excluded here. . At the
- 83 -
-------
same time, it can be said that the above views in no way detracted from the
significant results of ozonometry based on observations of ultraviolet ozone
absorption. Similarly this faUs to change the views pertaining to ozone
contribution to the thermal regime of the stratosphere held by this writer,
since ozone absorptivity did not depend on the mechanism of its formation"
The present author's hypothesis of the aerosol origin of atmospheric
oxygen called for some comments. It is known, for example, that highest
aerosol concentraion has been observed often in the lower atmospheric'
layers from which it follows that highest ozone concentration should also
be' observed there. However, to be able to create the condition it becomes
necessary to cool the aerosol for a long time, so as to permit accumula-
tion of hydrogen peroxide and its subsequent heating, illumination and
vaporization. These conditions can be realized near the ground only rare-
ly. Conditions above the tropopause are much more favorable in this
respect. In addition, factors contributing to ozone destruction were more
intensive near the ground, e. g. oxidation. Mention was made above of
difficulties connected with the recombination of free OH radicals in water,
I '
as cdmpared with air. Above a certain optimum air density ozone genera-
ti on is inhibited by the restricted motion of free hydroxyl radicals. This
explanation can be. refined by applying the concept of dependence of a re-
action chain on the total pressure of reacting component mixtures. '
Theoretical kinetics of chain reactions show [9J that there existed upper
and lpwer pressure limits outside of which the reaction did not proceed
d ----. -_u- -. -- - - -.... '--'--'- ___'d ------ . .. - o-
at al~ or did so considerably slower. For example, abov.e 500 C auto-
igniti'on of hydrogen and oxygen can take place only at pressures between
5 and 25 mm. The spontaneous appearance of a cold flame in a mixture
of air and carbon disulfide (0.03%) at 2000 C can take place at pressures
between 5 and 170 mm, etc. Similarly, the region of spontaneous ozone
formation in an atmospheric aerosol from moisture and oxygenn:l~ls~_hav~
upper and lower pressure limits, if it is initiated, as ass;umed, by fr~e Of':
radicals. By analogy with the above processes, it can be assumed thi'lt \
ozone formation occurred in the regi0n of 5-200 mbars: The see-and ques-
. - -. ..' - --' - -.- - - . -" . .. .. - .' --' -. - - I' .
. tion which arises relates to the mO,de by which the above processes can take
place in an aerosol at -500 of -600 C. 1. e. considerably below the freezing
point of water. It is known that at these temperatures light-beam pro1?es
detected pre -cloud drops with radii on the order of microns. In addition,
the above mechan~sm may function In an ice aerosol. Since the ice crystal
lattice is not as strong as that of true crystals, it is always possible to
displace OH- ions from the lattice and consequently to form free OH radi-
cals. It is also probable that the selective sorption of water at the air'
boundary was retained with respect to H+. OH~ and oxygen, and was only
weakened by freezing. - This author's' concepts of aerosol generation also
explains the type of fog formed after sunrise, most frequently in the
spring or fall and in relatively cool weather, which is found when the
ground layers' stratification was stable and the relative hu~idity was high.
It may be assumed that in the course of the night hydrogen peroxide ac-
- 84 -
-------
I
.. I
.cumulated in the cooled aerosol, on the condensation nuclei.' With sun-
rise and under the action of the sun rays and with incipient warming hydro-
gen peroxide begins to decompose according to (5). A large ~umber of
hydronium ions H30+, hydroxyl ions otr, and oxygen ions O2- are formed.
A large number of active condensation centers (ions) was formed in the
air leading to the creation of a stable fine fog persisting .for hours.
BIBLIOGRAPHY
.---.-----
--_._------- ----- ---- --~----,-- . ---- . --.
- -_.-.--_.-
1. r y 11 H P. M. ct>H3I1Ka CTpaTOC~rbl. rll:lpo~~eTII311aT. JI.. 1958. CTp. 111.
2. T e i C her t F.. \V arm b t \\. Ozonuntt'rsuchungen am mt'teorologi.
schen Observalorium Wahnsdorr. Abh. d. '\\el. Hydro!. Dienst. d. DDR.
4. Nr. 34. Berlin.
3. 3 II K e H A. Kypc XIIMH'ieCKOil tt>113I1KH. T. H. 1933.
4. D a x H. II r H JI b M aHA. 3.1eKTpoKIIHeTH'IeCKHii nOTeHUHaJI Ha rpa.
Hllue ra3-paCT80p. )KXcJ>. XII. Bbln. 2-3. 1938. . .
5. C () e h n A.. N e u man n H. Zeit. Physik. 20. Nr. 54. 1923.
6. Pew e TO B B.,[l. Hcc.1e.lOBalllle YHllno.1RpHblX 3apRJlOB 33po30JleA.
TpYAbi UeHTpaJlbHOH a.po.10rJl'IeCKOii 06cepBaTOpHH. Bbln. 30. 1958.
7. Pew e T 0 B B..J. npo6.1e~ia aTMoccpeplloro ..1eKTpH'IeCTBa H a,po30nb.
TpY.J.bI UeHTpaJlbllOil a.po.10rll'lCCKOIl 06cepBaTOpHH. Bbln. 30. 1958.
8. Tel Ius. 9. No.3. 423. 1957. .
9. C e M e HOB H. H. 0 HeKoTOpLix npo6.1eMax XHMH'IeCl(oli KHHeTHKH H pe.
aKulloHHOii cnoco6110CTH. IhJl'BQ AH CCCP. M.. 1958.
10. H B a II 0 B n. M. nepeKIICb l!O:lopoAa B CHe)l(lIOIi BO;le. Me-reopoJlOrHR
II rH;J.po,10rHR. N2 9. 1958.
II. Reg e n e r V. New Experimental results on atmospheric ozone. Scient.
Proc. Inl. Ass. Meteor. Rome. 1954; London. 1956.
12. JI e II 6 e it 3011. OCHOBbi Ta.1.1aCO'Tepanllll. DlloMe;lTII3. 1936. .
13. K 0 H C T a H T II HOB a . ill JI e 3 H H r e p M. A. H3BeCTHR AK3Jle!\lHH
HaYK. cep. cpI13.. Nv 2.213. 1937.
14. A JI e K C e e B n. n.. De C " 11 0 B C K H Ii E. A.. f 0.11 bI weB f. H.
H 3 a K 0 B M. A.. K a caT K H H A. M.. K 0 K 11 H r. A..JI H 8'
IU HuH. C.. Mac a HOB a H. ,[l.. ill B H 11 K 08 C K H Ii E. r. Pa.
KeTHbie HCCJ\e;1oBaHHR aTMoccpepbl. ."'eTeoPOJlOrHR H rHApo,10rHR. HI 8.
1957.
----------- -
15,.u!lp-;-~-;;-H-rTSI. 0 nep.'a~'YTpoBbiX 0l5.13K3X. H3BecTJ1J1 AH CCCP.
cep. rcocpIl3.. Nt 3. 1959. ,
16. K pac 0 B C K H ii B. H. 0 HO'IHOM H3nY'leHHII Hroa B HHcppaKpaCHOA 06.
JlaCTII cneKTpa. ,[lAH CCCP. 66. Nt 1. 53. 1949. .
17. X P r H aHA. X. cJ>1I311Ka aTMoc~pbl. rnTTJI~ M.. 1958.-
18. K 0 C h a n ski A. Cross section 01 the mean zonal flow and temperature
along EO' \\'. Journ. Meteor.. 12. No.2. 1955.
.. 85 -
-------
A TMOSPHERIC OZONE TEMPERATURE REGIME
ACCORDING TO SPECTROSCOPIC GROUND OBSERVATIONS
R. S. Steblava
The number af investigatians devated to. the study af atmaspheric
azane temperature is small. Many prablems have came up in the caurse
af atmasphe ric azane investigatians canducted at the Institute af Terres-
trial Magnetism. Of greatest interest are prablems related to. azane
layer temperature and to. the c~nnectian between factars causing changes
in azanasphere praperties including temperature and vertical distributian
changes, in relatian to. lawer atmaspheric layers and to. salar activity. .
Same reparts mentian such prablems (1-4), ather reparts refute their
impartance. The present paper reparts an studies aimed~t finding the
carrect answer to. such prablems.
Methads fa1; azane cantent determinatian reached a high develap-
ment level; in cantrast azane layer temperature determinatian had been
canducted in a nansystematic manner using different abservatian methads.
No. arganized ab!>ervatians had been previausly canducted in the USSR;
therefare, the study methads will be cansidered first. The methads are
essentially based an the principle af functianal relatianship between tem-
perature and absarptiancaefficients:
Temperature effect an azane absarptian caefficients in the ultra-
vialet regian was established experimentally in the caurse af many in-
vestigatians [7 -10, .14J. Labaratary measurements wer~ made using a
light saurce having a cantinuaus spectrum, e. g. a hydragen tube. Re-
sults shawed that successive maxima and minima were superpased an the
curve representing averall variatians in the azane absarptian caefficient
in the ultravialet regian (Fig. 1). A camparisan was mad~ between azane
absarptian caefficients recarde~~Il~.~ _labaratary in the Huggin's band re-
gian, and azane absarptian;.-c-oefficients abtained fram atmaspheric mea-
surements, using the same spectragraphs; results shawed, in bath cases,
. a caincidence af absarptian caefficients at the maxima, while at the minima
the caefficients measured in the atmasphere, were systematically belaw
thase abtained in the labaratary [7 -14J. The lack af caincidence at the
minima cauld be explained by temperature arpressure effects and pas-
sibly bath. At the 1929 Paris Canference an Ozane [16J, the questian was
raised whether azane absarptian caefficients in the ultravialet regian de-
pended an pressure. This questian has been' salved experimentally [15J.
Light fram a hydragen lamp was passed through an azane -fille~--t.~~~ _~~~ced -
in a thermastat. Absarptian caefficients were camputed.taCf1rst under nar-
.malc-anditians of 20° temperature and 760 mm pressure; J subsequently,
azane pressure was gradually reduced until it equalled the prevailing pres..
sure af 50 m. Mo.reaver, azane absa'rptian caefficients remained un-
- 86 -
-------
changed in the lfltraviolet region. Consequently, absorption coefficient
changes dependjd entirely on temperature fluctuations..
Fig. 1.
- -- ------ -- ------~-
.--..-. --- _. -- -~- - -
ec
R.1
Q.G
0.5
0.11
o.J
0.2
41
00
JI50
J200
J250
A
- ------------------------------
-----
I CURVE OF OZONE ABSORPTION COEFFICIENT IN THE HEGGINS BAND
Experimental data related to temperature effect on ozone absorption
. coefficients are contradictory. Wulf and Melvin [7J showed that coefficients
corresponding to absorption minima decreased with temperature decrease,
while coefficients corresponding to maxima remained practically unchanged.
E. and A. Vassy [10-14J ~ound that in the zone of minimum absorption the
.functional relationship between the coefficients and temperature within the
limits of '':;10 to -700 followed the course of a straight line. According to
Barbier and Chalonge [8. 9J absorption maxima and minima also varied
and contrast increased with decreasing temperature. Thus. the problem
becomes reduced to a numerical functional relationship between (k )
and temperature. Basic principles controlling absorption maxi~a m
behavior are not clearly understood. The method for temperature evalua-
tion unde r conside ration consists of two important factors:
1) laboratory data on absorption coefficient values as a te.mperature func-
don. and Z) method for converting and expressing measured quantities in
terms of temperature. Selection of a conversion method is closely asso~
ciated with experimental dat.a pertaining to the behavior of k = k(tO), whi~h
some investigators regard as basic. E. and A. Vassy assumed that a .
-----. ---
- 87 - /
-------
linear relationship existed between temperature and absorption coefficients
at the minima and that kA' the absorption coefficient, could be expressed
in the form kA =a + f3t, where (a) and (t3) are constants and (t) is tempera-
ture. Accordingly .
-----~._---_._-----'-_.-_.--_._----_.~._-'----- -
.____'H--'- _0. -----
k;20 -' ki.t = ~ (20-'- t),
-.. ..-.. --. .-. ".,- . -. 0" - - .. '-'--" -- . --.. ..,
, ;'ass1,lme t = - 80 and eliminate- ~ arid get
k,.20 - kAt = (k;.20 -.:.... k}.-so) 20 - t .
100 '
(1)
(2)
Determine absorption coefficients proper at the minima as. follows:
Plot a graph in which the measured optical density (D) is shown as a func '.
tion of the absorption coefficients at the maxima k , assuming that th~
, max
latter are independent of temperature.
...-.-. _._-'-'------'-~-----'-"-
D. =(k.mnx + ~o, '
where (x) is the ozone thickness and t30 is the coefficient of molecular
dispersion. Plotted points follow a straight line with slope (x), the re-
duced ozone mass, remaining the same at a given time for all wavelengths
at absorption maxima and minima. To determine absorption coefficients
at the minima, superpose their optical density on the graph's ordinate and
draw lines parallel to the abscissa to a point of loc:us intersection. The'
corresponding values on the abscissa will represent the looked for absorp-
tion coefficients at the minima. '
Barbier and Chalonge conducted laboratory experiments and found
that the difference between absorption coefficients, and not the coefficients
proper stood in linear relationship to temperature. This finding formed
the starting point of their calculations. Contrast va~ues could be deter-
mined 'with the aid of the following equation:
. -- -_._--- -- ..---~._- ._._--- _. -- ..------------.--
". .. -. ."-._-... - -- __d'.
T = KT'- kr -'- (K20 - k20),
(3)
where K , and k - symbolized absorption coefficients at the maxima and
minima ~especti~ely and were obtained from spectograms; Kao and kg)
- absorption coefficients for the same two wavelengths which were obtained
by making laboratory measurements at t iIt 20° C. The ~a~ue of'Y c.an be
determined with the aid of equation (4) obtained by COmblnlng equations (2)
and (3);
--- ------_.._---._-----_.~-_.- ----~---_.. ----
,,- . --'-----"-'2o--='i (k20 (1-- a) ~ K20 (1 - A)J. (4)
1 ~. 100
whe re
k.-so'
a---
- k10 .
A= K-so.
K10
- ---..- - -...:.
- 88 -
-------
Contrast difference can be determined -for six pairs of wavelengths and
,averages obtained using the following equation:
, I
---.----
._-------_._~
1~., 20 - T I ~ 20 T
r = - ) r = - - -- I k(1 - a) - K (1 - A) J = -=-;-
, 6 .8...J - 100 6 ' 100 .
or
r
...
20-T
100
-
The value of r is found from observations, Xl - is computed from labor-
atory data. It can be seen from the brief description of the two methods'
",tha£ "both are based on the following assumption: in the first case, ab-
sorption coefficients proper are dependent linearly on temperature and in
the second case the linear functional relationship is the property of their
difference. Subsequent studies conducted by Vigroux on temperature
effect on absorption coefficients failed to confirm the linear dependence
hypothesis. Vigroux's data were placed at the disposal of the International
Committee on Ozone for use in processing observations made during the
IGY, and were used as a reference in this investigation.
Now; examine the ozone absorption spe'ctrum in the ultraviolet
region in order that the structure of the absorption maxima and minima,
with which the temperature is evaluated, could be defined more accurately.
Methods proposed by Barbier, Chalonge, the Vassys and others
actually fail to make allowances for the fine structure of the vibro-rota-
tional bands., It has been known that total molecular energy nearly
equalled the sum of three energy components:
E=E'+E +E
el vi br sp
or, in other terms
T=T +G+F
el
Thus, according to Bohr's frequency condition, frequency of a spectral
line absorbed or emitted by a molecule can also be expressed as a sum
of three components
--- -----.- ._-----'-'._~-----.~---'-_._-- -
- ~_.- --_..----
v = T' - T" = (T~., - T:~) + (G' - G") + (F' - F") =
= v.., + '11"0.1 + Yep = '110 + Yep'
For a given vibrational transition, the value of 'Vo = vel + v ib is constant
while v varied with the corresponding values of the rota'honal quantum
rot I , .
in the upper and lower states. All the possible transitions between rota-
; f
- 89 -
-------
donal levels at -,p = const form a single band.
The number of molecules (N..) at a rotational level (j) of a lowe r Vl-
brational state ,and temperature (T) J is proportional to: '
------ _._--~. ---- _. - -~-, - -"-'--------- - .-
" -F(/) A
Nj~(2j + l)e ItT
or
\ -Bj(j+I) ':
Ni~(2j +' I)e It'
where 'If::-~~--Ti---, is a rotational c<5nstant;
, 8J1'q-tr' I' I , ; ,
I = llr2 "-:"1 is' moment of molecular inertia.
" '
. ' I';'
number at which a maximum of mdiecules
the basis of temperature and consdt.'dt B.
I is, the quantum
max
occurs. and is determined on
For higher vibrational states
------------~-
'. . , . - (G ~ F) ~~,
"'.' liT
Nj = (2J + I)e '
'-G!!!.. '
However, fora given vibrational transition, e ,kT is constant. The ac-
tual number of molecules in rotati6h:a1 state is obtained by multiplying the
above constant by the total number of molecules (N) and dividing by the
sum with reference to states. which: can be substituted by the integral
--~---,
.s Nidj.'
o
Accordingly
-.",th
'. -BiC/-I'J)"IlT ' ,II
(21 + ~ t ,:, = N :.ria (2j + 1) e -BI(/~ I) 'k7-.
f~~ '
0'
N.=N
J
(5)
Sin~e (j) appears in equation (5) as a. multiplier in the negative index, which
inc reased with j, the number of mdi~cules at different rotational levels,
j. j + 1, j + 2, . .'. iricreased at first with increase in th,e order of the level,
up to a maximum, and then, decreased when the multiplier became nega-
tive. Rotational level j at th~ ~reatest numbe r of molecules and at a
given temperature (T) ,~aJC:, can H~ found by setting dN/dJ = 0:
- -- - ~- -~
.+_._---_.~--~--_.' .------------- --..- -
.. - -.--- ._- . -- - .
a.v = N Bch ti3i(j+~) :~ [2 - Beh (2' + 1)1] = 0
oj kT kT I ,
whe nee
(6)
r-
..' , ' kT' I
lniai =)/ 2&1a -2'
(7)
Spectral line intensity is
\.
determined by'the
--~-~-'--- ,
.!.70 -
number of molecules at a level
-------
I
i
, i
,
from which the transition originated using statistical weights
and lower states; and emission frequency .
I
!
of the upper
(2(- 117(2r + I) _. (j' + ~, , I)
2 - J T ,
(8)
~ . D-h -BTW+I) ~
I N.... (" I .., + J) tT
ABSPN.== -.r-Y I Tie . .
(8')
For a given rotational-vibrational band. 'V = canst., and intensity of the
absorption line at a given temperature is expressed, in the final form,
as follows: .
'r--- = c .J- (/" + J.,' " J) e -B"i"W+I) 4;.-
ABSPN. T ..
(9)
Thus/it follows from equations (7 -9) that an increase in temperature
shifted the position of the intensity maximum in direction of the' increased
rotational quantum number and, furthermore, the entire band broadened,
while the intensity maximum flattened out.
Curves in Fig. 2 illustrate theoretical intensity distribution, and
were obtained by plotting quantum numbers along the abcissa and intensi-
ties along the ordinate. The curves show the schematic distribution of
1 b b for two temperature~ (T1 and T:;r> Td.
a sor. .
Fig. 2
---- -----.-.. --- -
It has been known that the bands can form
[17, 18J. Let F' (j') and FII (j") be the
rotational terms in the upper and lower
states respectively. The general formula
of a line intensity is 'V = 'Vo + F' (j ') -FII (jll),
and the selection rule for the rotational
quantum number
:J
,
- ---'-- ----- .
, ' QUANT. NUMBER
.1
j ~j = j' - j" = 0, :!:: 1
i
I iiei(f~- th~"ee-k~(;w~ branches
Ry = Yo + F' (j' + 1) - F" (j"),
Q', = Yo + F' (j') - F" (j"). .
p., = Yo + F' (j'. - 1) - F' (j").
Assuming- F (j) = Bj (j + 1) and disregarding higher order terms, it is found
,that the branch wave -numbers can be expressed by the following formulas
. ,
R', = Yo + 28' + (38' - 8") i -+- (B' - B")jl,
Qy = Yo + (B' - B")/ + (B' - 8")f.
p., = "0 + (B' - B1i +-(B' -Bjjl.
--':91' ~
-------
Due to the quadratic term~ one of the branches turns and forms an edge.
An edge forms in the, R branch when the coefficient of ja has a negative
sign, i. e., B' ,;" B". < 0, wher: B' < B". This occurs when the distance
between nuclei in th~. upper vibrational state is greater than in the lower
state. In this case.the edge' lies on the short-wave side. and the band ex-
hibits multiple darkeDing. When the value of B' < B" is not too s~all,
the edge will lie close to, the beginning .of the band.
. , ,I .~ \
~ig. 3 . :. / I
Fig. 4
'----;~---1~-:----:-----
.-------_.. --._- -- --------.-- -.--. - --
~
.If
OUANTUM NUMBER
If
. '. OUAtn'UM NUMB ER '
~--------
. .
,.
. Let Figs. 3 ahd 4 represent schematic intensity distribution in a
band at a given tenip~ ra:ture~ and in the respective absence and presence
of an edge. 'Line M marks the position of a virtual intens.ity maximum.
If the fine structure remains unresolved, the lines will blend in the vicin-
ity of the edge. Somewhere near, or on, the edg~, a maximum intensity.
interval existed the position of which need not correspond to the true maxi-
. mum of intensity distribution within the band. 1£ the band shifts toward
. the region of red, ~e apparent intensity maximum must lie on the long-
wave side with respect to the edge, and the intensity minimum must lie
at the band's end. Such a system of bands sets up successive maxima
and minima in the a.b~orption spectrum. V. N. Kondrat'yev and A. V.
Yakovleva showed that gaseous 03 generated four absorption spectra asso-
ciated with different electron transitions in an 03 molecule. The spectrum
. 0
in the 2900-3500 A region consisted of a system of bands with sharply-
-------
The ve rtical dotted ~ines indicate edge positions according to Y akovleva' 5
and Kondrat'yev's data. Clearly~ the edges lie basically on the short-
wave side with respect to the absorption maximum. '
Fig. 5
"
----;------- _._-----------.--- .------------- ---;---- ----=---- -_.. -- - - -....--------. --_.-
, '
:f I I I
, I I
,
0.7
0.8,
0.5
'ail
0.3
0.2
0.1
0.0 31:;1}
I
I_~~_I
3200
WAVE LENGTH
CURVE OF ABSORPTION COEFFICIENTS IN HEGGINS BANDS AND DISTRiBUTION OF EDGES
"
, I
.
It should be noted that for the sake of simplicity this author consider-
ed transitions in a diatomic molecule, although the ozone molecule is'tri-
atomic. Theoretical computation of an ozone molecule, absorption spectrum
. appeared impossible, since the very structure of such a molec'ule had not
be~n studied sufficiently. It had not been established up to the present
whether the shape of the ozone molecule was :Unear or triangular, a~_d-'_u~f- -
triangular, what was the apex' angle. According to Yakovleva and '
Kondrat'yev the molecule was linear [19].' Adel and Dennison [20] claim-
ed that it was triangular with a 1340 apex angle. Recent studies ,indicated
tha~ the_ap~~ angle was lZf>. :.-The_~ot_~_~~nal co~sta!it;=--a!l~~-co~s~e_~~~~-i~e'
\ fi~~ s tr~~~ur~~f the_r_c:>ta!!:!at~_o~~!.~aI!ds~__~~ul~.:.. ~?~ !:>e__d~t~~~~E-e~----
without knowing t1:le molecular shape and th:e i.t:lternucl~aI." distances~ "
A change in temperature should cause a. shift in the position of the absorp-
tion maxima and minima. However, absorption coefficients were measured
at one temperature only in the majority of experiments, the., position of
maxima and minima was observed at the same temperature and, subse-
quently,' a selected wave absorption was measured at other temperatures.
Under such experimental conditions, a change in absorption coefficients,
------- .
- 93 -
-------
for a' given wavelength was noted instead of a shift in the maxima and
minima. Observation of changes required ;high dispersion and resolution.
r---C5~ the basis of above consideration~, it may be coricluded that
successive maxima and minima in the ultraviolet portion of an ozone ab-
sorption spectrum actually.described the absorption intensity distribution
., I
in a vibrational-rotational band. The absorption intensity distribution' is
a nonlinear function of temperature. It was correctly stated at the Paris
Conference on Ozone that absorption was independent of pressure.
The present author I s computations of ozone absorption coefficient5
by means of spectrograms and the subsequent temperature evaluations
will now be discussed in the following paragraph. It was mentioned earlier
that the most accurate tables of ozone absorption coefficients, as functions
temperature, had been obtained by Vigroux [19 J under laboratory condi-
Hons [14J. In principle, one can determine spectrographically the absorp-
tion coefficient for a wavelength in the Huggins band region, for which, '
tabulated data exist, and, subsequently, determine the temperature fr,orr.~
these data. In orde~ to improve measuring accuracy, the difference qf
absorption coefficients for two waves was ~ound, 1. e., the so-called c'Dn-
trast. On the one hand, this was done to exclude systematic observation
errors. On the other hand, 'if the aerosol absorption was poorly seledtive,
I
contrast for a small wavelength region could be disregarded. Furthermore,
scattered light in our instrument was only partially filtered out. The ef-
fect of scattered light on the intensity difference of two similar wavelengths
was considerably weaker than the absolute Ivalue of intensity of one wave-
, I
length. Above considerations indicate that, the difference !n wavelengt;hs
for which contrasts are meas~red should not exceed 5-10 A. The followiHg
five p~irs of wavel~ngths were soelected for t~mperature evalu~tion: 3~3(j:-
3135 A, 3167-3176 A, 3130 ~3200 A, 3216 -3l20 A, and 3216 - 3!48 A, or pre-
cisely the same as tho~e examined by Bar8ier and Chalonge, the Vassys,
and Vigroux. In this case, the larger wavelength occurred at an absorp-
tion maximum, and the smaller- at an absorption minimum, which means
I
that these waves pertained to different vibrational-rotational bands, since
the latter are shifted in direction of the red, and, as the above considera;-
tions suggested, an absorption maximum should lie in the band's short-, I
wave region. It would have been proper to select inversely related pairs,
so that they would lie in the same 'band; however, in this case, distances
between waves would be 20 -30 A, unsuitable from the viewpoint of making
allowance for aerosol components and light scattering. Contrasts, as
functions of temperature, were plotted for selected pairs on the basis of
Vigroux's tables, the results.of whichlare shown in Fig. 6~ Contrasts
were found for each experimental pair. and corresponding temperatures
were determined from the curve. Subsequently, temperature values found
for each of the five pairs were averaged.
- 94 - '
: I
-------
Fig. 6
I
K
. 1.108
.'.IJS
OJJ
. 'OO.C -80
-GO
-qO
-20
00
20
qg
---'--
SO 80.C t.
----- .-._--
FUNCTIONAL RELATIONSHIP BETWEEN TEMPERATURE AND CONTRAST ACCORDING
TO VIGROUX
Now, examine the procedure for the determination of adsorption
coefficients~ Obtain observed data in the form of ultraviolet solar radia-
tion spectrograms. Determine spectral energy distribution on the basis
of many factors depending, first, on the solar spectrum radiation distri-
bution and, subsequently, on total ozone content. Ozone absorption coeffi-
cients depend, in turn, on the medium's temperature; determine energy
distribution by the instrument's parameters: its dispersion, resolving
power, and film. 1£ all measurements had been made by the same instru-
ment and were recorded on identical film, then the last three parameters;
should remain constant for a given wavelength pair. The thickness, 1. e.,
total ozone content, is a variable quantity which is always determined i
from the ratio of two wavelength intensities, so selected as to reduce terrlper-
ature effect on the absorption coefficients.
I
(kx + I' Make computations on the basis of the Lambert-Bouger law 1 = 10 e'j .
o Jm, where (k) is the absorption coefficient; (x) - ozone thickness;
(1'0) - molecular scattering coefficient; (m) - air mass; (10) - radiation
intensity at the outer atmosphere limit; (1) - radiation intensity after pas-,
sage through an absorbing layer; [k = k(t)] - absorption coefficient depent
dent on temperature. Each spectrogram measured the absorption density;
at selected wavelengths; it is then compared with the absorption density for
a wavelength lying outside the Huggins band. For this purpose wavelength.
o ( I
(A,2 = 3398..N can be used. Hence, for another wavelength, e. g. A = 3130
A) the following can be written: .
'-'Ig '/1' ,: )~(lol!/o2) --(kl - ks)x + (~I -PI)] m,
/1 . .
L = Ig(/1!J2). Lo = Ig(/o1i/os)
~ ...
assuming
- -'
- 95 -
-------
and having plotted values of (m).along the a.bscis sa and (L) along the ordin-
ate measured by spectrograms.: obtain a number of points which lie on a'
straight line. The slope of the line represents optical thicknes s differ-
ence for two wavelengths [(kl - k2) x ~ (~l - f3:a) J. and the y -axis inter-
cepts represent 1..0. On a day when, ozone thickness (x) is practically
c'onstant the dispersal of points will depend on absorption coefficient
changes only, the occurrence of which was due to temperature changes.
The same extrapolation can be applied to the other wavelengths of the
selected pair (Aa) and' (A3) (say, A3 = 3135 A)
. --- "- ._--~.--
19!1 ~= Ig(lu3/lo2) - [(k3 - k2)x -T (~03 - ~02))m
.n."' I. .
from which find
~ = Ig(loa102).
~ ------------- ._------..
Accordingly, the difference
-.....-'.. ~ -.--
Lo-~ = Ig(101iI0l)
will exclude the intermediate pa'rameter 19 loa. The value [lg (h /I3)J at a
,given i,nstant is known and. havin~ determi,ned (x) independently. .the fo1-,
lowing can be computed
. k1~k3 ~ 19(1odlo;)-1£(llil')'_(~I-~8) .
mx x
Above extrapolations for wavelengths 3130 -3B5 A are shown in Fig. 7 for
19 March 1959. (page 97)
Sir].i.ilar' extrapolations are carried out for several days and L() values
averaged out. Absorption coefficients are determined in a like manner
for other wavelength pairs. and the temperature found from the corres-
ponding curves.
. Determination of the desired contrast for different wavelength
pairs is complicated by the dissimilar number of Fraunhofer lines in
individual spectral intervals. Figure 8 (page 97) shows chang~s '
in a continuous spectrum and absorption coefficients in the Huggins' band'
region. The course of curves shows that the 3167 and 3176 ~ pair accords
well with the continuous spectrum. while the 3130 and 3135 A wavelengths
are shifted by several tenths of an angstrom and only partially coincide
with Fraunhofer lines. In 'measuring ;intensities for identical wavelengths
optical density changes were compared over the entire spectral interval'
which spanned a given waveleng.th pair (Fig. 9). (page 99) .
Spectroscopic atmospheric ozone observations were made in two stages:
1) during 1951 - 1952. and 2) during the IGY. to the"end of 1959. ISP-22
quartz spectrograph was used 'during the first period to obtain solar spec-
trograms. The sun's image was projected on the spectr,ograph slit by a
- 96 -
-------
-+--------- .
-.. - ---
- --" --- --.--- --.-
.,---+ --- - .------ --. . ---.
Fig. 7
3/3
ao
1.0
1.5
2,S
.0,5
/fO
1.5
'2.0
2.5 '
",
+ ---------
---- - ------ ----- --- -.
-+ - ----- ._--
EXTRAPOLATION BEYOND OUTER LIMITS OF EARTH'S ATMOSPHERE
1- 3190-3398 A; 2 - 3135-3398 A
Fig.,_~ -
0.6
H167A --; --- ,
"
"
,
I
,It
/ 10
/
fl8
Q8
0.5
, II
II
I H178A I
'I ~
I I
~/"""V,/I,
;W87 A /1 V--
O,lJ
U.2
-I ------2
.
J8.A
3130 .JlliO 3150 3188 317Q "8(1
0.3
~ - - - .-- - --.-.-- . - ---- --- -' 0
:OZONE ABS!?!'~!I~~ CU.RV.E AT _~~.!!E - ~A 3110-3180 A;
.~, - - - - COURSE OF CONTINUOUS SPECTRUM
! ' /
I
j.
- 97 -
-------
-~
I
mirror system in a horizontal, solar telescope and by a cylindrical quartz
lens. Using spectral plates with anti-aureole substrate and, subsequently,
diapositive plates, exposed for 5 min. and 30 sec~, respectively, normal
illumination was obtained in the 3020-3300 A and 3010-3400 A regions during
. summer and winter, respectively.
Fi g. 9
e
tg eo
~
e
&9 to
\
1,1
, :\3130;
...
...
...
,
...
~o
o,g
:\3157;
~ JI\70; /
, I
... I
... , I
1,3
. ~2
1,1
. .
. I .,
Furthermore, normal illumination over the entire wavelength interval,
where solar radiation intensity sharply decreased. was obtained bya graded
platinum attenuator. Data were processed with the aid of MF-2 microphoto-
meter. Observations were made systematically on clear days, and through
cloud' 'windows " on cloudy days.
- 98 -
-------
I .
The scop~ of observations was expanded during the second period.
Direct solar radiation spectrograms were obtained daily using the same
ISP-22 quartz spectrograph. Scattered light was eliminated by placing
USF-1 filter in front of the slit. Nocturnal observations were made dur-
I ,
ing and around full moon at the end of 1958 using SP-49 A spectrograph.
The latter possessed high aperture ratio, 1:1. 2 relative camera aperture,
which enabled it to obtain densities lying on the straight portion of the
characteristic curve in the 3100-3400 A region, for a slit width equal to
that of five normal [widths] and a 30 mm exposure. Unfortunately this in-
strument has small dispersion. The dispersion of. the ISP-22 spectrograph
in th~ working range is 15-26 A mm, whereas that of the SP-49 A spectro-
. 0
graph is SO A mm. These instruments have different parameters. There-
fore, all additional calculations, particularly those of 1.0 and 1..5, .were car-
ried out independently for each instrument. Diurnal obs ervation control
was done with the aid of 'instrument OFET -11. Approximately 3,000 solar
and 200 lunar (during 30 nights) spectrograms were obtained during both
periods. Changes in ozone thickness and in temperature during 19 Ma.-rch
1959 and 21 July 1959 are shown in Figs. 10 and 11, respectively.
Fig. 10
.[ t.
".-.....,. ..,..~-----..
. -- - + -...
20 MARCH
en- +1 ~
r
------
QJ O. t .
. ~~------------
Q.Z .100.'
10 II /1
~ HUM n M m M n n n H I
+ -------.-.-..u
HOURS
Fig. 11
------.- - --
.- ---- -
r t.
,. ./00.'
21 JULY, 1959
0. J ~:'-~--- . . .
....._._._------_::.::.----~
8.1 -100.' --- qJ.1m-iI
, ;
I 1
1 +
5 I
7 6 9 /0 11 12 /3 If /S 16 n
HOURS
. Mean ozone thickness reached 0.430 cm during the day,
cm at night of March 19. Respective mean temperatures were
-4SoC. Corresponding ozone values for July 21 wereO. 320 cm
cm and the temperature averaged -10 C and -20 C respectively.
and 0.440
-45° C and
and 0.295
Curves in
- 99 -
-------
-,~----- --
Figs. 10 and 11 show that both the ozone thickness and temperature ex-
hibited greater changes during the day, reaching up to 300/0 of the mean
diurnal value. At night, both the temperature and thickness remained
constant within the range of measuring error. The ozone thickness in-
creased characteristically during spring and was accompanied by a
decrease in temperature. Data collected during two days observation of-
fer no basis on which to judge the course of annual temperature variations
except perhaps, in a qualitative sense. With regard to ozone thickness,
similar ratios between intermediate latitudes had been confirmed by many
investigations. In general, temperature changes and ozone thickness were
reflections of each nther. A relationship has been noted between ozone '.
content and temperature changes and solar activity. Beginning with 13-15
July, 1952 an active region consisting of two symmetrically located spots
with an area of Sp = 1000 units passed across the central solar meridian
zone. The active region was surrounded by bright flare fields and flocculi,
and chromospheric flares. Moreover, ozone thickness values changed
from 0.247 to 0.446 em, and temperatures decreased by'300. On July 28,
1952 a similar, but smaller (600 units), group passed across the sun's
meridian, changing the ozone thickness from 0.260 to 0.384 cm, while the
temperature decreased by 40°. On April 19, 1954 an active region with a
large group of. spots and an 1870 unit area was observed at the central meri-
dian. The very instant an absorption flare was noted by an ionospheric
station, ozone thickness increased from 0.400 to 0.480 cm, while temper-
ature dropped by 28°. Externally it appeared that ozone thicknes s increas-
ed during geoactive flares, and that it was accompanied by a sharp tem-
perature drop.
This report presents in brief preliminary experimental observa '-
tion results intended basically for the description of observation methods.
Therefore, other important problems concerning the nature of measured
temperature and its connection with kinetic temperature have not been
touched upon in the present report. Apparently, laboratory measure-
. ments of [k{t)] were made at a. given, specifically kinetic, tempe rature.
Here the following question arises: is this tempe rature identical with
the. one measured spectrographically? This question can be further ex-
amined by studying the fine structure intensity of vibrational-rotational
bands. If such temperature is real, several other questions remain un-
clear: 1) what is its relationship to the t'emperature at different levels,
2) to what level does the maximum temperature change refer, and 3) to
what extent is the obs erved cooling during flares real? , Before the above
questions can be answered, the observed phenomena must be supported
first by a greater volume of data, and secondly, it requires the inclusion
'... .
of such additional data as vertical ozone distribution, direct temperature
measurements by means of radiosondes, and others.
- 100 -
-------
I
i.
BIBLIOGRAPHY
1-. Mal ~ r -k a r S. L. - Geo;;;~gn~tic va~iaiiori~ and diurnal range 0' atmos-
pheric ozone. Ann. Geophys.. 7. No. 2. 2~-213. 1954.
2. G I a Vi ion H. and G;; t z. F. W. P. Gerlands Seitr Geoph)'s 50
3EO, 1938. . ..'
3. r" e 8 iii W e 8 M. H.. A)I(. %5. 103. 1948.
4. n po K 0 ell b e B a H. A. 610.1. KO~IIIC'CIIII no IIC'C.'. Co.'tlua. N! I. 30. 1949.
5. K. ran d i k a r R. V. Proc. Indian Acad. Soc.. lA). 28. 63. 1948.
6. Per r y S I., La r r i e y H. C. Proc. Florida Acad. Soc.. 2. 89. 1931.
7. W u I' Q. R. and Me I \' i n E. H- Physical Review.. 38. 330. 1931-
8. Bar b i e t. D., C h a Ion g e D. Recherches sur I'ozone atmospherique
J. de Physique. 16. nO 3. 113. 1939.
9. Bar b i e rD., C h a Ion g e D. Annal. de Physique. 17. 272. 1942.
10. Vas s y E. Comptes Rendus. 203. 1363. 1936. .
II. Vas 5.y E.. Comptes Rendus. 202. 1426. 1936.
12. Va 5 5 Y E. Comptes Rendus. 214. 219. 1942.
13. Va 5 s y E. J. de Physique. 10. 2~0. 366. -459. 1939.
14. V i If r 0 U J[ E. Comptes Rendus. 230. No. 25. 2170. 2277. 1950.
15. Vas 5 y. These de docloraL Paris. 1937.
16. Qu~rtely Journal of the Royal .\\eteorological Society. Supplement. 62.
15-55. 1936.
17. r e p u 6 e p r r. CneKTpbl II CTpoellHe .:IByxaTOMHblx !oIG.1eKY.,. 1-1:1. 1949.
18. r e p u 6 e p r r. Ko.1e6aTe.'bllbie H BpaWaTe.1Lllbie cneKTpbl "Horoaro...
HIiIJ[ wo.,eK)'.,. "J1. 1949. '.
19. K 0 HAp a T b eBB. H 51" 0 B .1 e BaA. IKypHa., 'KcnepHMeHTanbHoA
H TeOperH'IecxoJ1 eIIH3JIKH. I J. Bbln. I. 1932. .
20. Ad e 1 A.. Den n i son D. J. Chem. Phys.. 14. 319. 1946.
A METHOD FOR COMPUTING TOTAL A TMOSPHERIC OZONE
MEASUREMENTS MADE WITH LIGHT FILTER
- _n. --- -' .
. EQUIPPED INSTRUMENTS
G. P. Gushchin
The well-known Dobson [4t 5J method for the determination of
ozone is based on the use of spectral intervals 15-25 A. wide discriminated
by a spectrophotometer. The lightt within these intervals can be con-
sidered monochrom'atic with sufficient accuracYt thereby enabling the use
of Boug~er's formula for the computation of atmospheric ozone [3J with,
relative ease cmd simplicity. The above problem must be approached' dJ,.f-
ferently for the use of filter-equipped ozonometric instruments. The light
o
filters I passband used in ozonometry is normally from 50 to 250 A. In
this case~ a direct use of Bouguer's law fails, since the coefficient of
ozone absorption changes several times within a filter-separated interval.
To illustrate thist Fig. 1 shows the variation in the fractional ozone ab-
sorption coefficient (curve C) in cm-l according to Vigroux [8J. .
- 101 -
-------
Clearly, a method for computing ozone, applicable to instruments capable
of discriminating narrow spectral regions is unsuitable for instruments.
with conventional light filters. Naturally, the possibility of using suffi-
ciently narrow ultraviolet filters in the. future is not excluded.
Fig. I In computing total ozone, measurl~d
-- ---- - by means of a filter-equipped instrument,.
~! -=--- Stair [6J used a method in which allow-
~: IE ance was made for the filter width and th.e
ILl: .4.0 u
> :; spectral curve of the solar radiation out-
...
~, !z side the atmosphere. However, Stair
~: J.D ~ used single spectral region measurements
z, I
-, 1.2. ::: and dis re garded solar light attenuation
>- - -lIGHT FILTER I, ....
::: 1.0 2.0 g caused by aerosols. The use of several
>
;:: 0.8 ~ filters did not improve the situation, s~nce
-.
jg 0.6 1.0 t: ozone content computations were made'
:: 0.4 LIGHT FILTER 11--- ~ individually for each filter by the samel
~'0'2 - 8: method. This resulted in considerab1ej
~ JZCD.___J-"llQ J60o....J6DD ~ dispersion of ozone-content values eXl
~----~- --_\i!!_~_I,E_tl_GJ:H A °ceeding many times dispersal of va1lfesi
SPECTRAL SENSI.TlVITY OF THE UNIVERSAL obtained by Dobson's spectrophotomete~.
OZONOMETER Bezverknii, Osherovich and Radionov [lJ
(A.& B) AND OZONE ABSORPTION COEFFI-
CIENT (c) AS A FUNCTION OF WAVE lowered the effect of aerosols on mea-!
LENGTH sured ozone values by using Dobson's'
method in connection with the filter-equipped instruments. They alsp de-
termined the effective wavelength of a spectral system which depended on
the filter passband, spectral response of the photoelement, and the spec-
tral curve of solar radiation. In contrast to Stair's method, two or three
I. ,
filters were used concurrently; readout ratios were found for different
filters. and ozone calculations were made by the. same formulas as those:
used by Dobson. The use of readout ratios for two, or three, filters
whose passband, in one case, coincided with the right wing of the oZbne
absorption band and, in another case was essentially outside the abs,orp-
tion band, almost precluded the effect of aerosols on the measured quan-
tity, sInce light attenuation by aerosols depended considerably less .on .
wavelength than did ozone absorption. Repeated use of same ozone ~b-
sorption coefficient and attenuation values caused by pure air scattering
for a given pair of light filters, presupposes constant solar energy curves
at the earth's surface. However, due to the variable light attentuation by
air, the above postulate is invalid in practice. Therefore, use of an in-
strument with an unchanged effective wavelength curtailed ozone measure..
ments in a wide range of solar heights. The effective wavelength of a
filter-equipped instrument is a function of the sun's height. Therefore,
atmospheric ozone content can be determined only by taking into consider-
ation changes in absorption coefficients and in scattering, with the sun's
height. In view of this, the present author proposed a new method of ozone
computation which is relatively free from above disadvantages. In develop-
- 102 -
-------
ing the method, this author tried to simplify the process of computing
ozone content. The proposed method is based on the following known
formula:
/ 1.'/ IO-lp"'l"mh+m,~-,1
;, =- n 1 1. 0 ' .
(I)
v.:-here KA - symbolizes spectral response of the instrument in relative
units; IA - denotes direct solar radiation intensity with wavelength A;
Ii - stands for direct solar radiation intensity outside the atmosphere;
I1I!.o -symbolizes ozone mass; x - total ozone; a - ozone absorption coeffi-
cient; m - air mas s; I3A - coefficient of attenuaiion due to scattering in a
p~re air; ml - aerosol mass; 0A - aerosol attenuation coefficient.
Equation (I) holds for plane -parallel light fluxes. Due to the
presence in the atmosphere of scattered light. particularly ultraviolet,
a~d, due to small sun or moon heights equation (I) is applicable to
sjmall-aperture instruments only. Therefore, the proposed method of
measuring ozone applies only to ozonometers with "small solid angles";
ih this cas~ the equivalent ozonometer angle was 30. Curves (a) and (b)
ift Fig. 1 represent spectral response curve (K ) of the GGO-constructed
olzonometers and apply to the first and second Alters, respectively. The
dzonometer spectral response was determined experimentally by a vacuum
thermocouple and a photoelectrooptical amplifier. The monochromatic
radiation was provided by SF-4 spectrophotometer with VSFU -3 hydrogen
vapor lamp. Spectral regions.' corresponding to curves (a) and (b) in Fig.
1, were separated into narrow ~ZO A) intervals, and, subsequently, equa..,'
Hon (1) was applied to each interval.
Values of I were taken from the solar spectrum tables. com-'
piled by Johnson '\,0 L7 J and reproduced in LZJ. Values of (m) and (11)
as functions of the sun's height e and, also the values of (at) and (I') were
taken from tables published in L 2J. At any given instant, the out~ut
microammeter readout for the first filter [1] is a linear function of the
sum of readouts from each spectral interval,' i. e.,
. - - .-----...---.
_.- --- . -.- . --- -
/ = K 1 ,-, h.
...
Similarly, the second filter readout II is
/' = Kl L /~ .
(2)
(3)
where Kl = const.
The readout ratio 1/1' for the two given filters and a photoelement ~s J
on the basis of (1).- (2), and (3), a function of the sun's height (e) and the
total ozone content (x), i. e., .
- 103 -
-------
,
- =--, F(& X).
I' '
(4)
Function F l8, x) is as follows in the integral form:
1,
J '1 tI).
- F (6. x) = "
['1-tI).
(5)
where 1..1 and 1..2 are spectral interval limits of the first filter and ""3 and
'\4 are similar limits of the second filter.
A graph of F (8, x) as a function of 8 can be plotted with a chosen
value of x = Xl. A number of such graphs can be plotted for different (x)
values on the basis of which a nomogram can be constructed for the de-
termination of total atmospheric ozone; the nomogram curves will also
represent loci of identical ozone content. In constructing the ozone nomo-
gram based on equations (1) and (4), no allowance was made for aerosol
attentuation coefficient 01..' on the as sumption that t\. = u. This is valid,
since: a) filter-discriminated spectral intervals were almost identical
(Fig. 1); b) (, ln the 3000 - 4000 .A region varied much less with wave-
length than dilcoeffidents 0\ and SA; c) and I and l' in equation (4) de-
pended on ,\.
Fi g . 2
1.7
1.6
1.5
1.1,
----1,3
1.2
~'" 1I
- .
"")
co 1.0
o
;: O,J
c
a: 0.8
:: 0.7
o
~ 0.6
0:
0.5
0.4
0.3
0.2
0.1
5
0./6
. o.ZO
0.24
0.28
~32
36
_40
a«
0.48
o,SZ
I
30 35 1,0 ItS 50 55 60 65
- - -+-----
DEGREES OF SOLAR ELEVATION
-------------
----------------- -- ---------
NOMOGRAM FOR THE DETERMINATION OF TOTAL ATMO-
SPHERIC OZONE
- 104 -
A nomogram for computing
total ozone which appeared in Fig.
2 was plotted according to the
above described method. Absds-
sa in Fig. 2 represents Solar
height in degrees, and the ordin-
ate represents readout ratios
11/12 for two filters. Ozone values
in cm, appear next to the respec-
tive curves. Atmospheric ozone
content can be determined easily
from the observed data with the
aid of the nomogram in Fig. 2.
To do this find a point on the nomo-
gram determined by the sun's
height and readout ratio multiplied
by the graduation scale factor, and
determine total ozone in cm by in-
te rpolation.
The proposed method for
computing total atmospheric
ozone can be used in association
-------
with instruments of comparatively wide spectral intervals. Analysis of
errors introduced by this method showed that the more the first filter
response maximum shifted in the direction of shorter wavelengths, the
smaller was the error; in other words, the greater were the ozone ab-
sorption coefficients, the smaller were the errors. The filter response "
maximum must not be shifted below 3000 ~ where the solar spectrum ob-
served at the earth's surface te rminated.
. The nomogram for calculating total atmospheric ozone can be con-
str\1cted for different heights above sea level up to the upper boundary of
the ozone layer. In so doing, coefficients ~A in equation (1) are multiplied
by a ratio P/Po, where P - is the pressure at a selected height, and Po
is the pressure at sea level. The proposed method, thus determined
thickness of overlying ozone layers at any level. If a filter-equipped ozon-
ometer could be lifted from the earth's surface to the upper boundary of an
ozone layer and, if along the ascent, the ratio of intensities of the direct
solar radiation within two spectral regions can be measured at given inter-
vals (say, 2 km), then vertical ozone distribution could be plotted from the
obtained ratios.
Due to filte r inhomogeneities, unidentical filter thickness photo-
element input slits and photomultipliers, an ozonometer batch may be pro-
duced having similar optical characteristics; in such an eventuality it be-
comes expedient to calibrate such instruments against a standard rather
than determine the spectral response of each instrument. Furthermore,
spectral response of one particular ozonometer can be used as the standard.
The calibration is done by making parallel ozone measurements by the
standard and all other ozonometers. The standard determines .total atmos-
pheric ozone content, at a given temperature corresponding to the two-fil-
ter readout ratios. The sun's height is logged during the measurements.
Readout ratios are found from the data on oz.one content and sun's height
using nomogram shown in Fig. 2; the found value is divided by the readout
ratios obtained experimentally. The quotient represents the ozonometer
calibration factor at a given temperature. Calibration factors are deter-
mined at different temperatures and are affixed to each instrument in a
tabular form. Subsequently readout ratios are multiplied by corresponding
calibration factors. Calibration factor dependence on temperature is ex-
plained in terms of filter transmission coefficient dependence on tempera-
ture. An example of this is shown in Fig. 3
U sing the method and instruments especially constructed at the GGO
ozone measurements were made at ~ifferent points in the USSR (Voieykovo,
SP-b, SP-8, Heiss Island, Antarctica) during the IGY and IGC. Compari-
son of the GGO-constructed ozonometers with Dobson's spectrophotometer,
conducted during six months in 1959, showed that the divergence. of mean
diurnal ozone values, obtained by means of these instruments did not ex-
ceed 80/0 of the absolute value.
- 105 -
-------
Fig. 3
-... -- . ----_. ---- ------ ----._--
'-~_._-- --0__-
.---'------.
~ (II
a:
(!)
"""""" IoQ.
-.q"
"'I..
The following conclusions can be
:0;; 1.2
...
-
made:
~ 1,0
II.
::; 1. The developed method for com-
g 0
--.-..1 -_JJ.- fL_- _21! __25 JO J6 . puting total atmospheric ozone can be us ed
ApPARATUS TEMPERATUR-E--INDEGRE-eS- in association with instruments equipped
--O~~;iOMETER---GRADUATIONCOEFF-IC-IENT-INRE:':-- with glass and other light filters.
LATION TO AIR TEMPERATURE
2.
The proposed method eliminates
aerosol introduced errors.
3. The desc ribed and proposed
ozone nomogram simplifies and facilitates computations.
BIBLIOGRAPHY
"
1': 'M-al u' ik a r S. L' Geomagnctic va'iTations-'an!f diurnal range of atmos; '''.
pheric orone. Ann. Geophys., 7. No.2. 2r.9-213. 1954.
2. G 1 a w ion H. and G (; t z. F. W. P. Gerlands Beitr. Geophys.. . 50,
3EO. 1938.
3. r He B bI weB M. H., A)I(. 25. 10). 1948.
4. n po K 0 IfI b e B a H. A. 610.1. KO~'lIccIlU no IICC.1. C..o.1uua. N~ I. 30. 1949.
5. K a ran d i k a r R. V. Proc Indian Acad. Soc.. (A). 28. 63. 1948.
6. Per r y S C, La r r i e y H. C. Proc. Florida Acad. Soc., 2. 89. 1937.
7. W u If Q. R. and Me I \. in E. H. Physical Review. 38. 330. 1931.
8. Bar b i e r. D., C h a Ion f! e D. Recherches sur I'ozone atmosphe~ique
J. de Physique. 16. nO 3. 113. 1939.
9. Bar b i e rD., C h a Ion g e D. Annal. de Physique. 17. 272. 1942.
10. Vas s y E. Cornptes Rendus. 203. 1363. 1936. .
11. Vas s y E. Comptes Rendus. 202, 1426. 1936.
12. Vas s y E. Cornpte.~ Rendus. 214. 219. 1942.
13. Vas s y E. J. de Physique. 10. 2!:O. 366. 459. 1939.
14. Vi g r 0 U x E. Comptes Rendus. 230. No. 25. 2170. 2277. 1950.
15. Vas s y. These de doctor at.. Paris. 1937.
16. Quartely Journal of Ihe Royal MeleorologicalSocicly. Supplement. 62.
15-55. 1936.
17. r e p u 6 e p r r. ClleKTpbl II CTpoellHe .J,ByxaToMHblx MO.1CKy.1. 1-1.:1. 1949.
18. r e p u 6 e p r r. Ko.1roare.1bllble, II BpawaTe.1Lllbie cnelapUi MHOTOaTON'
IIblX Mo.1eK)'.1. 1-1.1, 1949. '.
19. K 0 II A paT b eBB. II s:I" 0 R.1 e BaA. i!\ypHa., 3KcnepHMeHTaJlbilOA
\I TeopeTlI'IeCKOli IfIH3I1KH. 11. Bbln. 1. 1932.
20. Adel A., Dennison D. J. Chern. Phys.. 14.379.1946.
- 106 -
-------
I
I
I
REGULARITIES IN HORIZONTAL DISTRIBUTION OF AND
SEA~ONAL CHANGES IN ATMOSPHERIC OZONE
G. P. Gushchin
INT RODUC TION
I . I~ investig~ti~g the connection between total ozon~ and meteorol?gi - .
cal conditions, it 1S 1mportant to study the principles Wh1Ch govern mean
and instantaneous atmospheric ozone distribution. The mean total ozone
distribution pattern is characterized by mean ozone values at different
latitudes over a period of years; it is determined by slow ozone changes,.
il. e. by seasonal and latitudinal ozone changes. Instantaneous horizontal
dzone distribution is characterized by hourly and at times by mean diu~nal
ozone values at different latitudes. and depended on rapid ozone change~.
It can be assumed that slow and rapid ozone changes were governed by ~
different factors.
i The author:,'s purpose was firstly to analyze the basic principles ~If
Horizontal ozone distribution and its seasonal changes from observed data
a:nd, secondly; to study basic factors and their interactions which deter'-
I
mined the mean and instantaneous horizontal ozone distributions. I
SEASONAL AND LATITUDINAL ATMOSPHERIC OZONE CHANGES i
Figure 1 shows the seasonal changes in total atmosphe ric ozone (x)
during 1958, according to data obtained by Soviet ozonometric stations.
All (x) values were reduced to the new Vigroux scale.
Fig. 1
- -... -- - --
, a iIJQ
:z:
u
z 10.588
-'
7
8
...!
z; O,WJO
w,
I-
z'
g' O.JDO
w,
z,
0' .
~. O,ZOO
~,N.,FEBR.,MARCH,APR., MAY, JUNE, JULY, AUGo, SEPT.,
OCTo, NoVo,.DECo .. -.. ._------
'AVERAGE .MONTHlY TOTAL ATMOSPHERIC OZONE CONTENT FOR 1958
I-ABSTUMANI; 2-AlMA-ATA; ~VlADIVOSTOK; 4-VOEIKoVO; 5-DI~
ON 16; 6-CRo PAKHRA; ~HEI6S ISo; 8-NORTH POLE; 9-EL'BRUS
- 107 -
-------
. .
Figure I indicates that the maximum ozone quantity was observed at all
stations during spring and the minimal during autumn. The greatest differ-
ence between spring and autumn ozone values was observed at the northern
and the least - at the southern stations. A rapid drop in total ozone from
spring to summer was observed at the polar stations: Dixon, Heiss, and
SP-6. - Figure 2 shows the latitudinal ozone changes at different times of the
year. Curves in Fig. 2 were plotted from data collected at stations located
near 40° longitude E. Apparently, a positive ozone gradient existed between
latitudes from 60° to 40° during the March-October period. The maximum
ozone content during that period did not always occur above ,the Pole, but
appeared to be distributed along the meridian. In May, 1958, the ozone
maximum was observed at,..., 80°, and in June - August - at,..., 60° latitude N.
Observations at Indian stations, at low latitudes, indicated that annual ozone
amplitude changes in the tropics were not of lesse r magnitudes than those
at other latitudes,' and that ozone maximum and minimum were observed
there during summer and winter, respectively [7J. ,Data at hand have also
disclosed numerous irregularities in seasonal vertical ozone distribution
changes. The height of maximum ozone concentration at intermediate and
high latitudes -increased from spring to autumn, and decreased from autumn
to spring. Furthermore, in the first case, ozone content decreased in the
-lower portion of the ozone layer, and in the second case -- it increased.
The height of ozone concentration maxi-
mum changed only slightly at equatorial
latitudes. Changes in ozone concen-
tration at intermediate and low latitudes
seldom occurred duri~g the year above
- the 30 km level. - Seasonal variations
in vertical ozone distribution are shown
in Fig. 3, where curves of vertical
ozone distribution above Flagstaff,
Arizona were plotted from Adel's data\
[12J for the period from February to
- liD 50 GO 70 80 gO I
------------- --- --------------- August 1955. Vertical ozone distribu-
DEGREES OF NORTHERN LATITUDE
tion measurements were made there
from the ground simu1t~neously in the
ultraviolet and infrared regions. Figure
3 shows that the height of maximum
ozone condensation incr~ased from 16
to 28 km fr~m February to August, and that in the layer below 20 km it
decreased considerably during the same period.
Fig. 2-
---.---- - -- - ~~ ~
- - _.. _n- .
___-0.800
J:
u D.5qo
z
1;; 0,1i00
UJ
...
Z
o
u 0.300
UJ
""
o
N
a 0.200
---------.--- - --_._--~ .----- -------------+_.
LATITUDINAL COURSE OF IvTAL ATMOSPHERIC OZONE
IN 1958 ACCORDING TO DATE OF USSR OZONOMETRIC
STATIONS
I-MARCH; 2-MAV; 3-JUNE; 4-AUGUST; 5-SEPTEMBER
Theoretic~Lcalcula.tions of vertical atmospheric ozone distribution
made by many authors [9J based on photochemical theory showed accord-
ing to observations, that ozone concentration above 30 km was practically
independent of local seasons and latitude. However, according to the
above theory the greatest balanced ozone concentration below 30 km should
- .108
-------
be observed as seasonal changes during summer rather than during spring.
and in the equatorial rather than the polar regions. The d\sagreement be-
tween the theory and observations can be explained by the fact that photo-
chemic'a1 equilibrium of atmospheric ozone occurred only at heights ex-
ceeding'" 25 km during summer and ",,- 30km dl!ring wint~r.' Below t:hese
levels, photochemical ozone equilibrium were practically never attained.
and, therefore, seasonal and latitudinal ozone variations the re were not de-
termined by the photochemical theor'y, b~t depended on different factors.
Fi g . 3
-.---.--.-..-
- -.,-. ---.------
-.. --- -_.
.~ KM AM
IJOESJEBRO ,:~"~ l3?L I
JJ Jt' J
20 .. I '
I" !c 10
o 5 :17 'a 5 10 r. 5 10
10 'I C-I'.
~~ ~« ~ ~
"~ j'~JU" 'p rVAu"~: I
,,1: - ..': ..t~. J.
:J z. ~ h
10 It 10 I
\' 5 10 0 5 10 " 5 ,'0 0 ~ 10 15
'--,-- -----OZ~N~~~I!!iCE~i~_~Tf~_R_~_- , - - ~O-1 ~/IIM
VERTICAL ~TMOSPHERIC OZONE DISTRIBUTION DURING'
FEBRUARY-AUGUST 1955 IN FLAGSrAFF
This case is illustrated in Table
1, taken from Deutsch [llJ,
where the halftime necessary
for establishing the photochemi-
ca1 ozone equlibrium at differ-
ent latitudes and heights, was
adduced on the basis of theoreti..
ca1 calculations. Below 30 km
the time exceeded 12 days every-
where.
In demonstrating the season
a1 and latitudinal total ozone
changes it has been assumed tha,t
ozone concentration depended on
the temperature of a given atmospheric layer. The higher the layer tem-
perature the less ozone should be present in it, other conditions (radiation,
circulation and contents of the atmosphere) being equal. Available facts'
fail to confirm this assumption. The mean air temperature at low latitudes
is lower the year aroundat 15-20 km than at higher latitudes. with the ex-
ception of a small circumpolar region during wintertime.
Table 1
-------
TIME OF OZONE SEMIREOUCTION AT OIFFERENT LATITUDES AT DIFFER-
ENT ANNUAL SEASONS
IN SECONDS
ALTITUDE ~EQ;;;:TOI1 4sD 600" ~
IN KM 21 MARCH SUMMER SUMMER 'WINTER
50 3,1.103/3,5,1031 3,9.103
45 6.5.103 7,9.10~ 8.7.103
40 1 ,6. 104 I 2,4. 104 2,4 . 104
37,5 3.4.10414,6.104 4,9.101
~5 7,5.104 9,6.104 1,0.10&
32.5 1,8.1052.3.1052,4.10&
30 4,6. 105 5,8.10&, 5.7. 10&
27,5 1,2.10' 1.3.106 1.1.10'
25,3,5.10' 3,3.10" 3,5.10"
22,5 1,2.)07 8,5.10" 9,0.10"
20 5.1.107 2.2.107 2.2.J07
15 5,7,10",1.8,108 12.1.108
- 109 -
IN DAYS
600
WINTER
600
SUM-
I
MER
--- _.
60°
WIN-
TER
3 9.103
1 :2.10'
5,3.10'
1.2.105
3,1.10&
8.2.105
2,5.10'
1.7.107
60.10'
2> . 10'
1 6.10&
4' 6. iO&
1:1.10'
3.2.108
8,6.10"
3.0.107,
1510
1.0.108
4,3.108
1.2.\08
6.3.108
0,058 0.301
0,13 1,4
0,36 8,1
0.74 21
1.6 52
3.5: 160
8.3' 434
17
53
136
324 5890
3230' 31 600
-------
-.-
--.--- -
Regardless of the above, the mean ozone content at these levels was gener..
ally smaller at low latitudes than at high latitudes through the year. This
fact can not be explained by the photochemical the'ory which postulates
that at 15 to 20 km. more ozone existed at low latitudes than at high lati-
tudes [9J. Coefficients .of correlation between tota.l ozone and tempera-
ture at heights of 15 and 20 km were positive during summer and autumn,
indicating that total ozone increased with an increase in temperature of a
given layer. Thus. in the case ofVoeikovo in 1958 indicated correlation
coefficients at 15 km were + 0.62 during September -October. In the case
of Vladivostok, the coefficient of correlation between total ozone and
temperature at the 15-km level was + O. 53 i~ September 1957, and + 0.58
in May 1958. The same coefficients at the 20-km level at Vladivostok
were correspondingly 0.47 and 0.28. At the same time these coeffi-
cients had negative values at 3.6 and 9.0 km in Voeikovo and in Vladivostok.
The 15- and 20-km levels were chosen because, as previously men-
tioned, greatest seasonal ozone changes occurred there, and were parallel
to the seasonal changes in total ozone. Aerological data for layers above
20 kmwere not sufficient. Summer and autumn se'asonswere selected
because total ozone latitudinal gradient was minimal at that time, and,
consequently, the advection effect on ozone was lowest"which indicated
that air temperature did not affect the seasonal and latitudinal ozone
change s .
Occasionally. seasonal and latitudinal atmospheric ozone changes
were explained in terms of water vapor [7J. Apparently, water vapor de-
composed ozone. particularly in the presence of ultraviolet radiation [7J.
It has also been known that seasonal and latitudinal changes in the. atmos-
pheric water vapor content were considerable. UI?-fortunately no systemat..
ic proof has been presented of the water vapor content in the stratosphere;
making a comparison between water vapor and ozone impossible. The com-
parison of seasonal changes in ground [surfaceJ ~zone and in tropospheric
water vapor led to the conclusion that the two were not correlated. At in-
termediate latitudes, maximum ground-ozone concentrations had beenob-
served during summer [15J; accordingly, maximum water vapor concen-
tration in the troposphere should also be observed duri~g summer, and
not during winter. It can be speculated that due to turbulence ground ozone
dropped into the lower atmosphere layers;from the stratosphere, and pene-
trated the entire troposphere which is relatively rich in water vapor. Ultr-r-
violet solar radiation in the troposphere attained a maximum during sum- i
mer, and ozone concentration in the lower stratosphere at intermediate
latitudes during winter.and summer differed comparatively littie from each
other. If water" vapor affected the quantity of ozone, then the -latter, hav- :
ing penetrated from the stratosphere through the tropo.sphere onto the .
earth's surface should be caught there in an amount considerably less than
during winter. However, an intensified turbulence in the lower tropos-
phere during summertime appeared to be more responsible for concentrat-
. ---~-.
- 110 -
-------
ing ozone than ;the decomposition of ozone by water vapor. On the other
hand, the stratosphere contained considerably less water vapor than the
troposphere. Consequently, it can not be assumed that water vapor acted
as a factor controlling the seasonal and latitudin.al changes of total ozone.
SEASONAL AND LA TITUDINAL TOTAL ATMOSPHERIC
OZONE PATTERNS
According to observations seasonal and latitudinal-differences in
total ozone occurred basically below the 20 km level. Ultraviolet radia-.
tion which elicited photodissociation of molecular oxygen, practically.
speaking did not penetrate the atmosphere below the 20 km level, and con-
sequently, elicited no ozone formation. The ultraviolet wavelength which
dissociated oxygen was less than 2420 A; oxygen dissociation begins to pro-
ceed at a noticeable rate only at wavelengths less than 2025 A [8J. Due to
absorption by ozone and oxygen and to molecular air scattering, radiation
at the above wavelengths is blocked by atmosphere layers above 20 km [5].
Radiation of wavelengths greater than 2800.A penetrated the atmosphere
below the 20-km level, causing ozone decomposition. Ozone decomposition
in the 0-20 km layer occurred due to radiation in a bandwidth range of
2800 to 11,340 A [9J. A photochemical equilibrium is ~aintained in the
Jpper portion of the ozone layer at 25-60 km, and seasonal ozone changes
1
there are small. Solar radiation which penetrated into the lower portion
of the ozone layer should cause ozone decomposition there and, conse-
quently, a continuous ozone depreciation. However, ozone depreciation
Has not been noted. Therefore, another process must operate which ac-
domplished depreciation. It is the opinion of the present write r that such
Jrocess was. the transfer of ozone from the upper portion of the ozone
lk.yer into the lower due to turbulent diffusion. More'over, ozone depreci.'a-
1 '
tion in the upper portion of the layer occurred continuously due to photo-
chemical reactions. ,j,
'I I
' I
Ozone transfer in the atmosphere from top ~o bottom due to turbp.-
lent agitation or diffusion, can be explained in te r9s of vertical ozone di7-
tribution. Mean ozone concentration curves exhibit maxima at 23 km [14J.
Mean curves below this level indicated a decrease! in ozone concentration
from top to bottom.* Turbulent atmosphe ric diffusion equalized ozone con-
c'entration at all heights, and, in so doing, initiat'ed ozone transfer from
top to bottom. Figure 4 shows curves which illus trate the above process.
The upper curves represent vertical ozone distribution during October.
an:d April. The large arrows denote solar radiation which entered the I
ozone layer. The horizontal line at the 27-km level separated that layer
into two parts, - upper and lower. The upper portion received the ozone-
__~ormi~:!L J~~24~Q A)-~n~ o.zo~~_=~~~~mp~sing. (A
-------
lower portion received only the ozone-decomposing radiation designated f.
. The small arrows indicate the direction'of ozone flow. (u) from top to bot-
tom layers as a result of turbulent diffusion~ Curves of-the 10v~t~r po~tion
of Fig. 4 's-how the annual changes in ozone-decomposing solar radiation
(solid curve), and .the seasonal ozone variations (dotted .curve) ,During
the period of we'ak solar radiation, from October to April, the ozone con:...
tent in the lower portion of the iayer increased and a maximum was ob7
s,erved in April. The converse occurred from April to October when'
'solar radiation showed high mean-monthly totals; then the ozone conte*
in the lower portion of the layer decreased. The above explains why maxi-
mum seasonal ozone changes were noted during spring and minimum I
changes during autumn (in October). The above a1so indicated that annual
I
ozone changes were weak at low latitudes, where seasonal changes in to-
. 'I
tal solar radiation were small. Conversely, at high latitudes, where s;ea-
sona1 changes in the solar radiation were high, seasonal ozone changes,
were appreciable. This manner of increase' in the ozone amplitude with
latitude has been observed in actuality. . ,
. , . I
Figure 4 also shows the basicphotochenii':"
ca1 reactions which lead to the formation and de ~
, composition of atmospheric ozone, taken into con-
OZONE GENERAT- . P . L.. 9 J . 1 1 i h i
ING RADIATION. side ration by etzo1d in ca cu at ng t evert -
OZONE GENERAT. , .
INS RADfATIONca1 ozone distribution curve in the atmosphere at 10
OZONE. to 50 km. '
.. DIFFUSION .
'/ ------
, ~"')
~,
.,.- A PR I L I
.' -
/'
Fig. 4
SCHEMA ILLUSTRATION TOTAL AN-
N~AL ATMOSPHERIC OZONE FLUC-
TUATIONS EFFECTED BY SOLAR, Rh-
- DIATION
-620"' FORI'IATION
O,+hv=O+O' 0.<2424 AI
O+O:+,\\=O.+M.
, OZONE BREAKDOWN' - "
O,+h,.=O,+O CA<11340A'):
0,.+0=20, :
,
The following denotations have been intro-
duced for the quantitative description of seasonal
changes in atmospheric ozone at different latitudes:
u(t) -ozone flow per cm2 per sec from top to
bottom of theozcme layer at time t,
f{t)-intensity of ozone -decomposing solar
radiation in the lower portion of the ozone layer at
time t, .
xl-ozone content in the upper portion of 'the,
ozone layer (assumed constant), . .
'Xc -initial ozone content in the lower portion,
of the ozone layer, '
, x{ t) -to~l ozone content in the atmos phe re
during time interv.al (t). .
. - I
Ozone intake into the lower portion of the layer during a time 1',j at
a latitude cp is . '
- 112-
"
---~-
-------
-_..- ---. - - --.-- _. ----- -- - -- -._-
- --- --- - - ---
t
fu(t)dl.
(,
. D~ring th~-'S';:~'~I ti~~-i-~t;~-;';:l -~-~i~-~-;';;'di;.ti~; ;ill d~compo~e an amount
of>'l-?zone in the 19wer portion of the layer proportional to its content
,:. x'(ft - x I .
'.'-,.,,,1 C J Ix (I) --x')f(/)dt. (2)
.-t"t~7'''' -----~--- -~ - ---_._----_....J_------- -- p-.
where C - certain coefficient. The total ozone content in the atmosphere
at a time" will be
. ~ t
x(:).. x' + Xo + f u (I) dt - C i Ix(t) - x'Jf(t) dt. (3)
b i . . .
'A-~s'u~i~g that tot~l ozone Iil--th-e- atmosphere at a latitude remained on the
average unchanged froni year to year, set up an equation for the annual
ozone balance
(I)
T T
f U (t) dt .-- C IJx (I) - x'J f (I) dl ,,;, O.
. i 0 .
where t-.:. time iritervaT -equa.lt"o one year: -- .
Many problems can be solved with the aid of equa.tions (3) and (4),
.' in the field of atmospheric ozone. Now consider the following two methods
. of simplifying equations (3) and (4): .' .
1. Ozone flow into the lower portion of the ozone layer remained
c'onstant in time: u (1)'~ u = canst. . (5)
.(4)
- '2. Ozone flow into the lower portion of the ozone
Lii.r', to the index of zonal atmospheric circulation, 1. e.,
qiiilm during winter, and a minimum during summer
.j~ -- - - . ---, ~',-~' U(t)=al(l-bSin ;-t),-- _n
where (a) and (b) - -~'onst~~-t-sand 0 < -i --i T- --
A ~omparison of equation (6) with the curve bf annual changes in
zonal circulation index (A. L. Katz [4J showed that b = O. 35. For the
sake of convenience, let .
layer changed simi-
it attained a maxi-
(6)
t
R(:) = i Ix (t) - x'Jf(t) dt.
b
[(7)
~
- _. -----.----- - -------_.,------
----.---------- --.
In the first case, on the basis of equations (5) and (7), equation
(1) can be transformed to:
---.-.-----
xC:) = x' + Xo + u-::-CR('t).
--- -. .-----------.- -------~---
in equation (8) can be determined from equations (4)
i
(8)
\
and (5)
Constant C
C=~.
- R{T)
(9)
------:--futhe -seconci- ca~;, -using equati-~~~ (6f and (7), equation (3) can
be transformed to:
xC:) ~ x' + Xo + a:- - abT (I -cas .!:.. 't)-C1R(-::).
- 1: T
(10)
-- ~-n3 -
-------
Constant C in equation (10) can be determined from equations (4)
and (5)
- - -
-----... ...------..----.
To'
aT-2ab -'
C. ~ It
R(TJ
(11)
Knowing the constants x', Xo, u, and C, seasona.l changes in total ozone
can be computed from equation (8). Similarly, knowing constants Xl, Xo,
a, and Cl, the same can be determined from equation (10). However,
-these computations become considerably complex, since equations (8) and
(10) contain a function R(T) which, itself, depended on the ozon.e content
x(t) .
Equations (8) and (10) can be reduced to a form convenient for com-
putation. In computing mean monthly ozone values, replace integral (7)
with the following sum
...-------.-----------..-...-
, j
Ri == ~ (xm - x')! m ~t,
. m-I
(12)
whe reo i = 1, 2, 3, ,.., 12, and x and t represent respectively the mean
monthly values of ozone and of sdrar racfl1.tion, in the lower portion of the
ozone layer, and At - time interval equal to one month. It is evident that
integral (7) can be substituted by a sum of other smaller intervals, and the
problem can be solved with greater accuracy, although ~t = 1 month is a
convenient unit for comparing computed and experimental results. Let us
express equation (8) for i of the 1st and i-th months as a system of two
equations
-.. - - ..--..-.__._.~--..._-----_. ----.-----. .,--..
.\"i-I = x' + Xu -+- (i - I)u -CR,- h
Xi = x' + x" + iu - CRi.
(13)
Subtracting the lower equation from the upper and noting that according'
to (12)
....~--, -.- -....--- ---..-----..--.....- ....
R: -Ri-J = M(Xi.-X')/..
(14)
obtain
X, =
"'i"":l -+- u + C).t;c'I,
I + C).tfi
. (15)
where C from (9) is
c=~.
Ru
, (]6)
- 114 -
-------
I
I :
i
i
I
,
Equation (15) be~omes a ~ecurrence formula for computing seasonal ozone
changes. -Use of equation (16) requires a knowledge of constants u, C, X I,
and variable fi' :also the mean ozone value for any month, assuming it to be
the first. Similarly, from the two equations, (10) for the l-st and i-th
months, and (14) get
Xi-I - a -
12ab ["It "It ]
- cos -i -cos - (i - 1) + C1~tx'f,
:: 12 12 -
I + C1::.tf,
(17)
Xi=
whe re
C1=
12a(I-~)
. r. I
Ria
(18)
Equation (15) is the second recurrence formula for calculating seasonal
ozone changes. Its use requires a knowledge of constants a, b, Xl (b =
O. 35), the variable fi' and the mean monthly value of total ozone at the
- beginning.
Examination of recorded vertical atmospheric ozone distribution
curves (8) showed, that ozone content in upper portions of ozone layers
at intermediate and low latitudes was, on the average, x' = 0.150 em
(Vigroux's scale). The above value was used in the present calculations
for all latitudes. No published data were found on the intensity of ozone-
decomposing radiation f. in the lower ozone laye r portions. For the pres-
ent study it suffices to ~ow the magnitude of f. in relative units. For the
sake of simplicity, it can be assumed that f. w~s proportional to the possi-
ble diurnal solar radiation total incident ol a horizontal surface. The
present author used values of these totals as recorded in monographs [5J
and [bJ.Constants (u) and (C) in equations (8) and (15), and constants (a)
and (C1) in equations (10) and (17). were arrived at by substituting mean
monthly values of ozone totals (x) [2,8], and mean monthly values of possi-
ble solar radiation totals f. into equations (8) and (10). Coefficients (u).
(C), (a), and (Cl) were deiived by substituting two pairs of values x ('f) and
f ('f) into equations (8) and (10). The following mean values of above coeffi-
cients were obtained for latitudes from 0 to 60° with dispersion-less than
70/0:
u ~-7-,O.1-O-.eAlO. ;
eym
C =--, 9,7.10--' ~a__.;
- CAl.
a = 9,7.10-' eM 'J. ;
cym
- - I
C1 = 10,3.10-1 c~ -
CAL.
- 115 -
-------
It should be remembered that values of coefficients C and C1 here re-
ferred not to actual intensities of ozone-decomposing radiation, but to
values of expected totals of direct solar radiation onto horizontal surfaces
at sea level. It was assumed that the underlying surface albedo at 80° was
700/0. Coefficient values were in this case as follows:
u = '9,0". io~~ 'cM O. ;
. cym
a coI2,7.lOu..cMO.;
cym
. . l"MI
C=9.7.~".e=.; .
. CAL.
. . ;
C. = 10,7.10-' ~.~~".
. CAL.
Seasonal change curves for (x) at latitudes 0°, 30°, 45°, 60°, and 80° were
calculated and plotted as shown in Figure 5 with the aid of the computed
constants and equations (15) and (17).
Fig. 5
E
u
z: 0,f45Q
UJ
Z
o
N
o 0.J50
...
..,
I-
o
I-
n.ZJ/J
-.---- --- - .
n.
.. . 1- .
za: c:c: > z.... CJ a.. 1-
em -ca.. -c~ ::);:)1&.1(.)
-;JUJ r.c r.,.., "'.c II) <:>
~- t ------2 'C'3 c.,.
. .
>. (.)
o UJ
Z 0
0-__'- --
-..--. ___._0-
SEASONAL COURSE OF TOTAL OZONE IN ATMOSPHERE AT DIFFERENT
LATITU!)ES .
I-COI1,.UTED ACCORDING TO FORMULA (15)~COMPUTED .ACCORDING
TO FORMULA (17) 3-VOEIKOVO 4-ASASTUMANI
The dotted curves were obtained using equation (15), and the solid curves
with the aid of equation (17). All above curves exhibited maxima during
. . I
spring, and a minima during autumn, which is in agreement with observed
data. An increase in latitude caused an increase in ozone content, the
greatest latitudinal ozone gradient occurring during spring, and the least
during autumn. Amplitudes of annual ozone fluctuations also increased
with la.titude. For comparison, mean monthly ozone values at Voeikovo
and Abastumanifor 1955-1959 are indicated in Fig. 5 by circles and squares,
respectively. Agreement between theoretical curves and experimental data
at latitudes under consideration is adequately high. Curves computed from
equation (17) are in better agreement with eXIJerimental data than those
computed by equation (15).
- 116 -
-------
I !
SOME CONCLUSIONS ARRIVED AT FROM THE EQUA TIONS FOR
CALCULA TING SEASONAL AND LATITUDINAL OZONE
FLUCTUATIONS
; Many problems related to ozone can be solved using the derived
cloefficients u, C, a, C1, supplemented by some assumptions. Constants
(u) and (a) can be used to evaluate the turbulent ozone transfer from the
u!pper portion of the ozone layer into the lower. The established value of
uI= 7.0 X 1O-4cm 03 /day represents the mean vertical ozone flow through
1 cmz per day. The ozone flow amounted to 0.021 cm per month~ and
0.252 cm per year. The vertical ozone flow per year for the entire earth
is '" 3 "X 109 m. On the basis of different considerations Petzold [9J ob-
t~ined for the above a value of the same order of magnitude, namely 109 m.
Constant (a) also yielded a closely approximate value for mean annual
I - ,
v;lertica1 ozone flow. The amount of solar radiation required to decompose
a given amount of ozone in the lower portion of the ozone layer, can be
a!pproximated with the aid of Constants C and Cl, and equation (2). By vir-
t~e of the ozone layer balance (equation (2) ), the amount of ozone decom-
posed annually in the lower portion of the ozone layer is equal to the ozone
ih.flow from top to bottom. On the assumption that the ozone-decomposing
r!adiation which becomes absorbed in the lowe-r portion of the ozone layer
amounted to 1% of the possible radiation total, it is possible to compute
the mean value of solar radiation required to decompose, for example, -
0-.001 cm 03 in the lower portion of the ozone layer. Calculations, by
- means of equation (2), indicate that ~ 5 ca1/cmz were required to decom-
p'ose U. OU1 cm of ozone. The above amount of radiation. absorbed in the
10-20-km layer raised the tempera ture of this layer by 0.10. Concurrent.-
ly, cooling of the same layer occu~red basically due to infrared losses in
water vapor and carbon dioxide, so that resulting temperature changes can
- be smaller or negative. It must be borne in mind that at different latitudes
and different seasons of the year, the above amount of radiation was ab-
sorbed in the lower layer in different time intervals. The interval will
also be shorter in the South during summer and longer in the North dur~ng
winter. The rate at which the lower ozone layer becomes decomposed at-
different latitudes and seas ons of the year can be calculated by means of
equation (2). Results indicated that, at 60° latitude 0.00004 cm, 0.0016 cm,
and 0.0014 cm of ozone were decomposed per day during January, May, and
June, respectively.
Knowledge of the vertical turbulent ozone flow from top to bottom
and of vertical ozone distribution, makes possible the computation of
- transfer (A) and the atmospheric turbulence coefficients (K) at 14 to 19 km.
Coefficients (A) and (K) are related to vertical ozone concentration gradi-
ents dx/dz in.the following manner:
. -. - -- .. ---
--------- -..--
dx dx
u=-A- =-Kp-.
dz dz
(19)
- 117 -
-------
where z denotes elevation above sea level, and p - air de~sity in the con-
sidered layer. Relationship (l9) applies when ozone concentration remain-
ed unchanged with sea$ons due ~o reasons other than turbulence. Above
conditions are sufficiently satisfied for ozone at high and intermedi~te
latitudes during polar nights and winter respectively. If these conditions
were neglected, then values qf (A) and (K) will be overstated due to a
relatively faster ozone decomposition by.radiation in the upper portion
of the l4-l9-km layer, as shown by equation (19). Computed values of
coefficients (A) and (K) in g/cm/sec and m2 see, respectively, for heights.
from 14 to 19 km are shown in Table 2, without corrections for ozone de-
composition. Minimum A and K values in Table 2 with such corrections
a.re, apparently, closer to actual values than their maxima. More de-
tailed proof of the vertical ozone distribution will, probably, result in
more accurate future determinations of turbulence characteristics in the
lower stratosphere and upper troposphere.
Table 2
---'-
1
"
!
VAlUES--oTEX~iiANGE COEF-FICIENT A IN G/CM/SEc., AriD TURBULENCE COEFFI-
CIENT K IN M ISEC. AT 14-19 KM ALTITUDE, COMPUTED ON THE BASIS OF AT-
~--------__m__~OSPHERIC OZONE DATA --------
~~ 2' METHOD USED IN DE-
PLACE OZONE 'DETEn.:. _A,e/cl"If.: K,M l TERMININe VERTICAL
-----_.-- HI NATION DATES, see. . SEC. atoNE DISTRIBUTION
. 1-
DEI.HI 364 187 RUTATION EFFECT
. NEW MEXICO 2/1V 1948 r. 104 68 ROCKET
NEW MEXICO 25/11 1950 r. '3O 24 OZONE SONDE
WEISENAU 9,'XII 1953 r. 55 55 n
70 50 n
WE ISENAU 43 40 n
WEISENAU JI/V 1955 r. 379 271 . DIRECT ILLUMINATION
. LA I ITUDE 60050.1. 14/1X 1912 r. 73 30 ' LUNAR
364. EClI PSE
TROEMSE 187 ROTATION EFFECT
------------------ .----------------
11.
_._.._-_.~._._---,.......-~---_._----_._-------- ~-- _. ---------
INSTANTANEOUS HORIZONTAL D1STRIBUTION OF TOTAL
ATMOSPHERIC OZONE
Studies of instantaneous horizontal total atmospheric ozone distri-
bution conducted at the Main Voeikovo Geophysical. Observatory (GGO)
with the aid of aeroplanes [3J, supplemented by ozonometric station data
indicated that there occurred a frequent complex geographical ozone dis-
tiibution. Different ozone concentrations were observed frequently at
identical latitudes. Cases were observed. when total ozone content de-
creased with latitude in some regions. Baric topographic charts of
- 118 -
-._-.-... _____0.____-
-------
different levels were drawn in search for the cause of such differences.
Ozone values from 10 Soviet and 10 foreign ozonometric station's were
plotted on daily baric charts of 300 - and 200 -mb surfaces. Foreign sta-
tion data were obtained with the cooperation of the world's IGY Data Cent~r
inMoscow. Ozone data for 1957-1959 obtained from direct solar observa-
tions were used in the analysis. No ozone data obtained from observing
the zenith ceiling were used for lack of reliability, and jet stream regions
were determined from baric charts. Studies were conducted on days when
jet streams occurred in the ozonometric station areas on the one hand and.
when ozone measurements were being made on the other hand. As a re-
sult it was possible to establish the occurrence of increased horizontal
gradients of total ozone in jet stream zones. Ozone gradients in such
zones were normally oriented in the direction of high pressure, where
300- or 200-mb charts were used, so that areas with increased ozone con-
tent were situated to the left of the jet stream axis, and an area with a
reduced ozone content was located to the right of the same axis. Ozone
gradient magnitudes in jet stream regions were several times and oc-
casionally more than 10 times greater than the mean latitudinal ozone gradi-
ent for the same latitude.
Fig. 6
- - ---- -.
B
Tr.'~- ~~ Ji'-SHARK
w'tSH' NGl'ON tU
;// I
-,
-z
AJ
.' -
BAR IC TOPOGRAPHY CHARTS. OF 300 ~~ LEVEL --- .
A - 30/11 1 J 1958; B - 29/IV, 1958; 1 - ISOHYPSES; 2 - FLOWING.
CURRENT; 3 - OtOIOHETR'C STATIO' AND OZOIE VALUES II 10-3 c"-
Figure 6 shows baric topography charts of the 300-mb surface for
3-o'clock Moscow time on 30 March, 1958 (a). and 29 April, 1958 (b).
Thin lines indicate contour lines (isohypses) while thick lines with arrows
indicate jet streams. The North Pole is indicated by NP, ozonometric
stations are indicated by triangles, and the adjoining numbers represent
total ozone in 10-3 cm. An examination of Fig. 6 discloses that ozone con-
tent to the left of the jet stream axis, looking in the direction of the flow,
increased, and to the right of the axis -decreased. A similar pattern can
- 119 -
-------
be seen in Fig. 7, where baric .'charts of the 300 -mb surface ~re shown
for 6 November, 1957 (a) and 21 May, 1958 (b). Ozone content at the
American (Bismarck, Washington, Caribou) and Italian (Vigna di Valle,
Messina, Elmas) stations depended essentially on the position of the jet
stream axis at a given moment. For example, on 6 November; 1957 there
was 21% more ozone in Washington than in Bismarck (Fig. 7); on 29 April,
1958 there was 270/0 less ozone in Washington than in Bismarck (Fig. 6),
while on 21 May, 1958 ozone content at the same stations differed only by
6% (Fig. 7). In the first case, 'Washington was situated to the left of the
jet stream axis, and Bismarck was located to the right; in the second case,
the situations were reversed. In the third, both stations were located to
the right of the axis.
Fig. 7
,_A
-,
-2
.&3
-BARIC TOPOGRAPHY CHARTS AT 300 "8 LEVEL
A - 61XI, 1957; B - 21/V, 19'33 1 - ISOHYPSES 2 - FLOW~NG CUR-
RENTS 3 - OZONOMETRIC STArloNS A~OZO"E VALUES IN Icr eM._-
---------- -'-- -- -- - - -.------
-------------------
In cases where no jet streams occurred in the station area, no
sudden differences in ozone content over the neighboring stations had been,
observed normally. This is illustrated in Fig. 7b where three Italian
stations (Vagna di Valle, Messina and Elmas) are shown outside the jet
stream zone. Ozone content at the above stations was at that time 0.340
em, 0.327 em, and 0.340 em respectively, 1. e., all three figures were
nearly identical. Figure 7a illustrated a case when less ozone was ob-
served at Voeikovo, the northern-most station in Europe, during autumn
than the more southerly station, Ukkel. Similarly, less ozone was ob-
served at the'same time at Bismarck U.S.A. than in Washington. In-
dependently of the st~tion latitude, the observed high ozone values (Ukkel,
- 120 -
-------
i
Washington) were located in both cases to the left of the jet stream axis;
stations re,cording low ozone value (Voeikovo, Bismarck) were situated
to the right of the same axis. .
Dependence of horizontal ozone gradients in jet stream zones on
the time of the year was also investigated. It was demonstrated that the
above gradients remained small in subtropical and extratropical jet
stream zones during spring. Such a case was apparently associated with
the presence .of maximal total latitudinal ozone gradients at that time.
In the subtropical jet stream zones horizontal ozone gradients remain,
generally, increased through the year. Increased gradients are seldom
qbserved during summer in the extratropical jet stream zones, a fact
vi,hich can be explained by the latitudinal ozone distribution pecularities
during that period. It was previously mentioned that a maximum ozone
total can be observed at intermediate latitudes during summer (Fig. 2),
as a result of which meridional stratosphere advection caused no ozone
I - -- - . .-.
increase in the extratropical jet :' stream zone during this period. It has
I . .
1)een ~nown that jet streams had been discovered frequently at the southern
~Ieriphery of high-altitude troughs, and at a northern periphery of high-alti-
I
tude crests at the 200- and 300-mb levels. In this case, the principal
portions of the high-altitude trough and crest were situated to the left and
right of the jet stream axes respectively. Thus, an increased ozone con-
tent should be observed at high-altitude troughs, and a decreased ozone
content at high-altitude crests. Such a condition was observed in actuality.
The vertical structure of jet streams was examined in an attempt
to explain the obtained facts. It has been known that flows which generated
subtropical jet streams attained heights of 18 km during summer, and 20
km and higher during winter. Since rapid increase in ozone concentration
with height began directly above the tropopause [10], it can be assumed.
that jet streams affected horizontal total atmospheric ozone distribution.
Aerial masses in jet streams propogated at high speeds in the direction
df jets, and mean transverse motion components are small; the' .
. jet streams exhibited a barrier effect on ozone, found to the left of the
jet, preventing the passage of large ozone quantities through the jet zones. .
Air masses of small ozone content were situated generally to the
right of the jets, in accordance with the gene ral latitudinal ozone dis tribu-
don. In view of this a horizontal total ozone gradient was noted in jet
stream. zones. It is conceivable that jet barrier effects were intensified
as a result of ageostrophic wind changes in direction of low pressure oc-
'curring due to turbulence -created friction. As a result, air iil the jet
zones moved slowly across the contour lines (isohypses) in the direction
of decreased pressure, preventing an increased quantity of ozone, on the
left hand side of the jet, from moving through the jet stream.
-
- 121 -
-------
In conclusion, it should be stated that the mean horizontal distribu-
tion of total atmospheric ozone can be defined by two basic factors: solar
radiation and turbulent diffusion. The pattern of instantaneous horizontal
atmospheric total ozone distribution was associated closely with high-alti-
tude baric fields and, in particular, with jet streams. Naturally, princi-
ples which governed latitudinal jet stream distribution can also affect the
average pattern of horizontal ozone distribution.
BIBLIOGRAPHY
ni. fy W ~I H r. n. W3~;('pcHlle 030Ha c ca~lO.leT3. Tp~:.:i~ no. lIb/n. 93. 1959.
2. r Y WII H r. n. npe.:i1l3pIlTe.lbHble pe3y.lbT3TbI 113~epeHllii 06wero COJ1~P'
>l'aHIISI 030Ha 110 Bpe~ISI .\\rr II CCCP. TpY.:ibl rro. IIbln. 10:;. 19<.0.
::\. f Y W H H r. n.. Po M a H 0 II a P. r.. Po M a III K II H a K. H. Ihw~'
peHlle 06mero CO.1ep)f\aIlHSI aTMoc4>er""ro 030Ha IkI IIpe,.,1I ropllJOHTanbllbix
nO.leT08. TpY..1.bI frO. lIb/n. 103. 19(0.
4. 3 II e p e B A. C. CilHOnTIIlleCK3S1 "'eToopo.l0rIlSl. fll.:ipo,.,erH3.'13T. J1.. 1957.
5. K O.H a paT b e B K. 51. JIYlIlICTaR ~Hepr\t)1 CO.1H11a. fll,.'tpoMerH3J.3T. Jl..
1954. .
6. M II .1 a 'H K 0 II II II M. MaTeM3T1ll1ecK311 K.lI1M3ro.l0TIIII II aCTpoHoIolH'IecKaR
TOOpliSi Kone6alillH KJlHMaTa. fOCTeXIIJ)1aT. 1939. .
7. MilT paC. K. BepxlISlII aTMoc4>epa. HJl. M.. 1955.
1:1. n po K 0 ct> b e B a H. A. ArMOCct>ePHblH 03OH.' H3,J,'BO AH CCCP. M.- JI..
1951. .
9. X B 0 C T II K 0 B H. A. 030H B cTparocct>epe. ~'cnexH ct>H3H'IecKHX KaYK.
59. Bbln. 2. 1956. .
10. B r ewe r A. \\'. Ozone concentration measurements from an airceraft.
Quart. Jouv. Roy. Met. Soc.. 83. No. 356. 1957:
11. D'u t s C II H. 0. Das atmospha'rische. Ozon als Indikator fUr Stromungen
in,der. Slratosphare. Arch. Meleor. Geoph. Siok!. Ser.. A.. Nr. I. 1966.
12. Epstein r... Ostcrbeig'Ch~. Adc'r .\. A new method for tile
. determination of the verlical rlistribition 01 "wne from the gTound. Journ.
. .\\eleor.. 13. No.4. 1956. .
J:~. J 0 h a n s c n If. On the relation behn'en meteorological conditions and
Ihe total amount 01 ozone. Polar Atmo~ph. Symposium. Pt. I. London.
1958.
IL Reg e 11l' r E. Neues \'omOlOn in der ErdatmMpliilfc. NatuTwi5!i. Rundsch.,
7. Nr. I. 195-1.
15 \r a r OJ h t W. Unterwchung des hodcnnahcn Ozon. Arch. Phys. Thera-
pie. 10. ~r. 4. 1958.
1'. ,
- 122 -
-------
!"
I
CONNECTION BETWEEN ATMOSPHERIC OZONE
I AND METEOROLOGICAL CONDITIONS
A. S. Britaev and A. P. Kuznetsov
Knowledge of atmospheric ozone and its relation to atmospheric
physical processes can be used in investigating basic factors which form-
.. ed weather. and governed the sun's effect on our planet. The most exten-
sive study thus far pertains to the relationship between ozone and hori-
zontal air advection. Dobson, and investigators who followed him investi-
gated different geographical regions, and showed that most of the atmos ,:
~heric ozone was" contained normally in air, flowing into observation points
(rom higher latitudes. This pointed to a correlation existing between
changes in ozone content and pressure, tropopause height, temperature,
visibility, and some other atmosphe ric properties" Similar statistical
relationships were demonstrated earlie r for Central Asia by Sh. A ~ Bez-
v:erkhnii [lJ, and for Moscow and Lower Volga regions by other authors
[IIZ,3J 'The relatt'onship between total atmospheric ozone and tropopause
h!eight was clearly indicated by the fact that s mall amounts of ozone corres-
~ond to high tropopause, and vice versa. According to P. F. Zaichikov,
aind E. M. Orlova [4,5J, the! correlation coefficient of quantitative ozone
changes and the tropopause height attained a value of 0.5. It is interest-
ing to note that the extreme total ozone values compared with tropopause
height lagged by approximately one day [zJ, indicating that ozone changes
could be attributed to circulation processes, and, consequently, could not
be used by weather forecasters outside their relationship to changes in
o,ther atmospheric characteristics.
Total ozone increase is accompanied by upper troposphere cooling
and lower stratosphere heating. The correlation coefficient (r) between '.
changes in total ozone (x) and mean temperature of the 3 -km laye r below
the tropopause tl r(x, tl) = 0.54 on Moscow region and coincides with the
order of magnitude as value r(x, h) = 0.56, computed by Johallsen for Nor-
way [7J. The temperature correlation coefficient for the 4-km layer
above the tropopause was r(x,~) = 0.48, which is somewhat smaller than
the one found by Meetham for Oxford (r = 0.53 [8J and the one determined
on the basis of temperature at.13-km height for Alma Ata (r =0.88) [lJ.
According to I. A. Khvostikov [6J, similar relationships can be explained
by properties. of dynamic processes which occurred when lowering the
tropopause level corresponded to raising the temperature of the lower
stratosphere and lowering it in the upper troposphere.
The horizontal air transfer alone can not explain all c:;:hanges in
total ozone; therefore, many authors assumed that vertical air currents
effected ozone fluctuations. Photochemical processes swiftly com-
pensated ozone decrease in the upper atmosphere occurring in descending
- lZ3 -
-------
air currents by increased total. ozone in vertical air columns. The basic
difficulty in evaluating the role of vertical currents is presented by the
complexity of their measurements or calculations at heights up to 20-30
km. In evaluating vertical currents from basic topography charts, Reed
[llJ found an actual increase in total ozone in descending air currents at
10 to 16 km height. Miyake and Kawamura compared ozone content with
approximate values of the vertical wind. velocities at the 500-mb surface
, level over Tokyo [9J and arrived at similar conclusions. The possible
relation between total ozone and the sign and magnitude of the v:ertical
current was also studied in lower atmospheric layers. As an example,
Fig. 1 shows ozone deviations from the m'ean value and the vertical w~nds
at different levels over Moscow during the period of 31 July to 6 August,
1959. Vertical currents at 1. 5- and 3-km levels were computed from the
divergence of horizontal wind velocities [sJ, and at heights of 3 to 28 km
from the impeller rotation rate of radiosondes [4Jlaunched by the Central
Aerological Observatory four times daily for the purpose of aerological
atmosphere sounding.
Fig. 1
. ~- . ~. --: H:~ ~;..._._-_._------ --~-'-----. - .------.--------. . ---... - . . .._. ---- - -_....-- - .
:~:2 .. II \~ ! I \l f\.
0.II5.01S.& 1\ r"'---l \ !' )
0.2 11.0 /4.0 l / I .J. - " \., (r.
OC 2.D t2.fJ ;': i ~~,-~~---' ,.
. ./' ~ ' . ,.) . . -;--, )
~z 0 ~o ,P. I. ~
. . / i./'; 'J
0.11 2.D 8.0 . '\~- )1 . /
0.5 4,0 5.0 I \'
0.8 G~ 11.0 ) k ~ ~ ~ j I
::: .~l I
31 J Z J II 5 G 0 AYS
'JuL y' "'AUGUST ' .
-- . . -_._._~-_._._.._-- -_._~----~----_:.......__. -
CONNECTION BETWEEN OZONE CHANGES AND VERTICAL CURRENTS;
HEAVY SOLID LINE - OZONE CHANGES; FINE SOLID LINE - VER-
TICAL WIND CURRENTS AT 850 MB LEVEL IN RELATIVE UNITS;
LINE AND DASH - VERTICAL WIND CURRENT AT 700 M8 LEVEL IN
RELATIVE UNITS; DASH AND DOT LINE - VERTICAL WIND CURRENT
ACCORDING TO RADIOSONDE DETERMINATIONS IN RELATIVE UNITS.
POINTS TO THE LEFT FROM THE MIDDLE LINE CORRESPOND TO THE
DESCENDING CURRENT, POINTS TO THE RIGH CORRESPOND WITH THE
. ASCENDING CURRENT
Best agreement between changes in two given values was noted,
as expected, at 16 to 20 km and higher. However, ozone fluctuations at
times also agreed well with the sign of vertical currents in the troposphere.
The reason for this lies apparently in the fact that in determining a synopti-
---_.--
- 124 -
-------
cal situation, f~r instance, in cyclonic and anticyclonic regions. vertical
currents of the same sign covered the greater portion of the troposphere
and extended into the stratosphere. eventually reaching a layer having an
increased ozon~ concentration.
Fig.. 2
Figure 2 shows the correlation
coefficient. (ozone content vs. vertical
wind) as a function of height over Moscow
for February, May, and August 1959. In
addition to the sharply defined relation-
ship between positive ozone changes and
descending currents above the tropo-
pause, correlation clearly depended on
time of the year. A change in the cor-
relation sign occurred most frequently
in the troposphere, which can be explain-
ed, most likely, by intensifica~ion of wind
variability in the lower atmosphere. A
positive relation between ozone changes
and descending wind with r'" + 0.1 -pre-
dominated during Februaryand Mardi.
The amplitude of changes in the correla-,
a'8.5 fJ -O.f .0.5 a -0.5 r tion curve around zero was lowest dur-
- .. ---. ing August and highest during May. The
CONNECTlOii-il-ETWEEN-'quANTITAr I VE" oZONE DEv IAT IONS ab solute correlation coefficient value as
FROM AVERAGE MONTHLY VALUES AND DETERMINATIONS OF
VERTICAL WIND VELOCITIES OVER Moscow WITH RESPECT an average of all altitudes had a reverse
TO ALTITuDE direction during the months under con-
I - FEBRUARY, 2 - MAY, 3 - AUGUST, 19S1 sideration.
".1tM
1.1
z.
22
28
III
3
f8
.16
12
10
8
6
- II
2
Fig. 3
r
0.'6
0.11 0
D.08 o.
0.011
~D
-D.DII
SPRING
SUMMER
SEASON~
FALL-
Curves in Fig. 3 show ozone
variations as functions of vertical wind
change through the year. Summer and
winter fluctuations in total ozone are
associated more closely with vertical
air movements than with horizontal
transfers, and, conversely, during the
transition periods, particularly in spring,
the effects of advection and large -scale
turbulence predominated. It can be as-
sumed that the appearance of a maximum
ozone total in spring was related to advec-
tion, because the vertical air motion, at
that time was almost independent of
ozone fluctuations.
. ~ - - -- - - ..
. wfiiTER
------_._--~-_._----~--_.- -
SEASONAL COURSE OF OZONE FLUCTUATIONS COR-
RELATION COEFFICIENT IN RELATION TO VARIA-
TIONS IN VERTICAL FLOW ACCORDING TO 1959
. DATA
Thus, vertical c'urrents constituted a basic factor which regulated
the amount of atmospheric ozone. This conclusion is compatible with the
- 125 -
'/
-------
data obtained by Paetzold [lOJ, who assumed that advection was responsi-
ble for ozone fluctuations during spring only; Paetzold also explained ozone
changes during the rest of the year in terms of vertical transfers. The
material presented.above leads to the conclusion that the relationship be-
. tween ozone fluctuations and meteorological conditions can be expres sed
primarily in terms of horizontal and ve rtical atmospheric currents.
BIBLIOGRAPHY
------------ ---- --------- - - ---" ---- ._._-~~--_. -. --
I. 5 e 3 B e p x II II if W. A. 0301l0~leTpll'leCKlle .1allHble no .AJJNa.Are B CO-
I10CTaBJlellHII C. HeKoropblMII MeTeopo.lofll'leCKIINH ~aKropaNH. Tpyp.J.I
Ka3axcK. Hl1f,\\I1. Bbln., 5. 89-100. 1955. .
2. 5 p 11 T ae 8 '\. C.. 11 0 3e 11 a c B. A.. K y 3 Hell. 0 B A. n. K 80-
. IlpoCY 0 CB5I311 o~Ulero CO,JepiK3HII5I 03OH3 C MeTeopo.l0rH'IfCKHIIIH yc..l0-
BII5IMII. MeTeopo.l0rll51 II rIlJlpo.l0rl!SI. X9 10. 24-29. 1958.
3. I) P II T a e 8 A. C.. K y 3 H e u 0 II A. n. BepTHKa.lbHoe pacnpue.leHHe
0301Hi. TpY.1b1 U,\O. 8b1n. 32. 28-35. 1959.
4. 3 a il.' II K 0 B n. CPo K 8Onpocy 06 113MeHellllll BepTIIKa.lbllbix p.BlliKe-
Hllil B03.i1}'xa. TPY;lbl U,\O. Bbln. 10. 1953. .
.5. 0 p .'\ 0 II 3 E. M. K BOnpocv o~ paC'IeTe BepTIIKa.lbHJ.lX cKopocreil.
MeTeopoJl. II rH.1p.. N9 I. 1955. .
6. X B 0 C T II K 0 B H. A. 0:1011 B CTparoccpepe. ~.'cnexII ~1I311'1fCKHX HaYK.
. 59. Bbln..2. 229-324. 1956. '.. ..
7. J 0 h an s e n H. Variations in the total amount 01 ozone:o\'er Tromso and
. their. correlations with other meteorological elements. Geofys. Pub!.. XIX,
5. 1955. .
8. Me e t ham A. The correlation of the amount .01 ozone with other charac-
teristics of the atmo~phere. Quart. Jour~. Roy. Met. Soc.. N~. 271. 1937.
9. M i yak e\'. K 3 W' a rn u r a K. StudIes 01 the atrnosperlc ozone at
Tokyo. Journ. Met. Soc. Japan. No.4. 1954.' ".
10. P a e t z old H. K. Die \'ertikale Verleilung des atmosphanschen Olons.
Zeit. f. Naturlorsch.. lOa. H. I. 1955.
II. R e e d R. The role 01 vertical motions in ozone-weather relationships.
Journ. ."'etoor. 7. 263. 1950.
TWO IMPORTANT FEA TURES OF OZONOMETRIC INSTRUMENTS.
G. P. Gushchin
Five years experience with different ozonometric instruments at
the Voeikovo Main Geophysical Observatory (GGO), placed into focus two
factors which sub~tantial1y affected measured ozone quantity, namely, 'the
s~ze of an ozonometric instrument's solid angle and the inst!u.ment's tem-
perature fluctuations. No attention has been given until recently to the
magnitude of the ozonometer's solid angle. This was due partly to the
fact that ozone measurements frequently yielded solar radiation intensity
ratios for two closely-spaced spectral intervals at the same solid angle.
On the other hand, it has been known that ratios of scattered to direct
solar radiation, at the same solar elevation was considerably higher in
- 126 -
-------
the ultraviolet region in which ozone was measured, than in the visible
spectral region. Two experiments were conducted for the purpose of ex-
plaining the effect of the instrument's solid angle on the measured ozone
quantity. The purpose of the first experiment was to determine the func..
tional relation of the ratio between two filter readings to the solid angle d
the universal ozonometer at different solar elevations. The unive rsal
ozonometer has one light inflow opening and its light filters can be re-
volved between the light exit diaphragm and the photocell, or photomulti-
plier, by means of a turret mechanism. Two light filters were used:
the maximum response of the first and second light filters was at 312
and 372 'm~, respectively. . The ozonometer was oriented at the sun by
means of an optical sight and the measurements were made during clear
sky weather. Direct and atmosphere-scattered solar radiation was in-
cident on the instrument within the limits of its solid angle. It is known
that in order to determine ozone content, it is sufficient to know the react-
ing ratio of two light filters and the sun's elevation at the time of measure-
~ent making.
The solid angle was varied by changing the light inflow diaphragms.
Four different diaphragms were used, subtending four different solid
angles equal" to 2.0°, 6.2°, 11. 'f, and 30. So. For the sake of brevity
these are expressed from here on in two-dimensional coordinates. As an
example,' Fig. 1 shows results of measured ratios of reading made at
different solid angles under clear skies at Voeikovo. The reading ratio
from two light filters 11/12 is shown in Fig. 1 as a function of solar eleva-
tion 0°. Numbers from 1 to 4 indicate solid angle values (1-2.0°, 2-6.2°,
3 -11. "t" and 4 - 3 o. SO ).
Fi g. 1
r'>''!
\IS 0.3
-
l-
e
II:
'"
z:
~ 0.2
c
&II
II:
11/,
1,2
0-1
"'-Z
o-J
.-.
0.1
10 '12 III 1$ /8 11
- .4_- . - - ---- - -----
DEGREES OF SOLAR ELEVATIO~
READING 'iiATlO-l7f;INR'E-l'AT-1ON TOSOLAf
HEIGHT AT DIFFERENT SOLID ANGLES (I - 20 .
2 - 6.2°, 3 - 7.11°,4 - 30.SO)
It follows from Fig. 1 that the
greater the solid angle, the greater is
the reading ratio. The divergence be-
tween readings at different solid angles
increased with increase in the solar ele-
vation; it becomes negligible for eleva-
tions about 20° and solid angles less
than 11. 'f on clear sky days. These re-
results indicate directly that reading
ratios obtained by an ozonometer were
functions of the ozohomete r' s solid angle.
Different ozone values may be obtained
subsequently if measurements were made
using variable light inflow diaphragms. ,
In addition, calculations will yield lower
ozone values in instances of greater
solid angles. Results thus obtained can
be inte rpre ted as follows: as wave-
127 -
-------
o '
lengths 'decreased a rapid increase in atmosphere-scattered light occurred
in the ultraviolet region in which measurements were made. This caused
the intensity of scatte red radiation to become relatively greate r within the
first filter interval than within the second filter interval. In the case of a
greater solid angle additional scattered light becomes comparable to direct
solar light. Since scattered light intensity in comparison with the direct
sola! light increased with decrease in solar elevation particularly in the
case of short wavelengths, the solid angle effect at low sun appears more
pronounced and is manifested by a considerable divergence of the reading
ratio 11/12 ,1"S indicated in Fig.!.
. Figure 1 indicates that for solid angles of 2.0 and 6.2°. the reading
ratio 11/12 remained the same within the 10-20° interval of solar elevation.
This means that under clear s Ides and solar elevation above 10° , ozone
. ° .
measurements can be made at a solid angle of 6. However, in the "
presence of clouds in the sun's proximity smaller solid angles should be
used in making, ozone measurements. To illustrate this situation" Fig. 2
shows results ~f :readirig ratio, ]1/12 measurements' made at two (3 and 10°)
solid'angles at Vbeikovo. .
1.200 Z,
0,100 10
Measurements at solar elevatioJ of
less than 25° were made in clear weather.
. . °
At elevations above 25 light cirrus clouds
were seen in the sky near the sun. ,Hori-
zontal visibility exceeded 20 km. The read-
ing ratio 1/1 at a 3° solid angle (curve 1) and
o n
solar elevation up to 40 , was negligibly
smaller than the same ratio at a 10° solid
angle (curve 2). Curve 3 in Fig. 2 indicat€:s
percent deviation between c'urves 1 and 2 at
different solar elevations, L e.,
Fig. 2
r,1'1 61
11 ,
\
0.610 61
MO' 6'
D. "', ..
0.316 II
q
II
,z,
tI
.,"
- ~ . - -
RBADING RATIOS 1 /12 FOR OZONOMETER
, LIGHT FILTERS INIRELATION TO SOLAR
, ELEVATION e (VOEIKOVO 16/V",i~)
. I - 50(10 ANGLE OF APPARATUS - 30. '
2 - SOLID ANGLE OF APPAR}TUS - 106; Deviation cr increased rapidly with
3 - CHANGES 1M RATIO 11/12 IN % decrease in solar elevation indicating that
(SRACKETED EXPRESSION). DETERMINED " . ,
AT VOEIKOVO 16/VII, 195). TOTAL 0.. -effect of the solid a~gle on read1ng ratl9
ZONE DURING DETERMINATION TESTS EQUAL- 11/12 was most pronounced at a low sun.
LED 0.332 CM ,
, !
The purp~se of the second experiment was to determine the func-
tional relationship of reading ratio 11 /12 to altitude above sea levelifl
using instruments having diffe'rent solid angles at same solar elevation.
For this purpose an automatic ozonometer [1] was lifted in the Voeikovo
region by helicopter to a height of 2.1 km. Measurements were made
through a speciaily constructed open window, using instruments of 3 and
10° solid angles. Measurement results are shown in Fig. 3, where reading
--~~--- ,_. -
- - /1//' (10°) -/111. PO) 100.
J-' /1//,(3<)
(I)
d
- 128 -
-------
ratio P = Ii /12 and the percentage deviation (J are shown as a function of
elevation above sea level in km [1].
Curves land 2 refer to the 3° and 100
solid angles, respectively. Light cirrus
clouds were seen around the sun during
measurement making; solar elevation was 49° .
A haze of nonuniform density was clearly seen
from the helicopte r in all directions. Hori-
zontal visibility at the earth's surface exceed-
ed 20 km. Analysis of plots in Fig. 3 indi-
cated that reading ratio Ii /12 increased with
elevation above sea l~vel (curves 1 and 2).
This can be explained by the fact that short-
wave radiation in the atmosphere increased
with increase in elevation more rapidly than
did long-wave radiation. At 2.1 km the read.-
ing ratio Ii /12 is the same at the 3° and 10°
solid angles. In this case it was possible to
observe ozone subsequently at heights of 2 km
and higher using a 10° soUd angle. The ef-
fect of aerosols on reading ratio Ii /12 was clearly evident in the lower two-
kilometer layer. Curve 3 in Fig. 3 shows that the deviation between read-
i~g ratios Ii /la, measured with instruments *of 3° a'n\:i 10° s oUd angles was
11% on the ground, 5% at l-km level and 0.2% at 2-km level. Total ozone
content in the atmosphere was 0.317 cm at the time measurements were
made with the helicopter bourne instrument. Calculations showed that
changes in the Ii /12 ratio which resulted from 11% and 50/0 changes in the
solid angle amounted to 21% and 10% apparent changes in the ozone content
respectively.
Fig. 3
3% P
15 0.&5
J
10 010
2 '....,
5 0.55 ~
0,58
'. 0
.. . - - - __2.
ALTITUDE IN KH
CHANGES IN READING RA~~ up ;-1-';1-2-
IN RELATION TO ALTITUDE AND
DEVIATION VALUES C1 ': (plCj P3)1.P3"
I-PWITH'SOLIOANGlE .
100; 2 - C1 IN %. HELICOPTER DE-
TERMINATION AT VOEIKOVO 18 JULY,
1959 AT 490 SOLAR ELEVATION
Analysis of the above three examples showed that the solid angle of
an ozonometric instrument constituted an important factor which exerted
a. significant effect on the measured ozone value. The accuracy of ozone'
measurements increased with solar elevation, with reduction in the instru-
I
m'~nt's solid angle, and with increase in elevation above sea level.
The second important factor which affected measured ozone values
was the ozonometric instrument's temperature; changes in the ozonometric
instruments temperature, distorted the measured ozone value due to'ther-
mal expansion of some other factors. For example, at a constant ozone
concentration in the atmosphere, the Dobson spectrophotometer begins to
indicate decreasing ozone values with increase in temperature. To com-
. pensate for the temperature error, Dobson's spectrophotometer has been
equipped with one or three pIano-parallel quartz plates; when rotated by
a small angle the plates shifted the solar spectrum to the previous posi-
tion [3 J.
- 12 9 -
-------
The pre$ent author's recent investigations showed that fi1ter~quip-
ped ozonometric instruments distorted the measured ozone value as the re.-
suIt of temperature oscillation. Iri search for an explanation of the above
effect, the present author inves'tigated the GGO universal ozonometer. It
was found that temperature error in the universal ozonometer was caused-
mainly by a change in the transmission coefficient of the second filter.
Figure 4 shows the changes in transmission c--.--.---.-.. .-----.__"n_' -____._n.' ture variations, while the transmiJsior-
- '-' ~~....~~PARAT~S..!.E~!E~!_~URE IN ~E.~REES , I-
COEFFICIENT OF OZONOI1ETER LIGHT FILTER coefficient of the second filter decreased
- I
TRANSMISSIBILITY I~. RELATION TO APPAR- with increase in temperature, attaining
ATUS TEMPERATURE. TESTS "ADE AT VuEI- ° i
KOVO 6/VI, 1959 IN DIRECT ~OLAR RADIA- a 20% reduction over a 12 temperature
TION AT NOON WITH e". 51.4°. increase. The transmission coe£filcient
I - FIRST LIGHT FILTER; 2 - SECOND of a glass filter chang~d little in thlle ultra-
LIGHT FILTER
.violet region with temperature changes
[2]. For instance, the transmission maxima of the UFS -2 and ZhS -3 glass
filters dropped by 6.4% and 4% respectively when their temperatures' has
been raised by 1000 .
In the present author's study the first filter represented combina-
tion of UFS -2 and ZhS -3 glass filters the transmission curve limits 6£
which did not c.qincide with the filter passband. Thus, a 12° change in
temperature had 'practically no effect on the filter transmissivity. How-
ever, the seco~d. filte r, constituting a combination of U:F.S -2 and SZS -18
glasses was different. In this case, the right limit of ,the UFS -2 glass
passband and the left limit of the SZS-18 ,glass passband coincided with the
, \
filter passbandj in other words, they actually formed the filter passband~
'The left bound of SZS-18 glass passband shifted to the tight upon heating. \
At the same time, the ieft bound of SZS -18 glas~; passband was extremely
steep. Thus, temperature increase led to a reduction in the passband
of the second filter, a process which is illustrated in Fig. 4 as a decrease
in the transmission coefficient. The temperature-induced errors in filter-
- 130 -
-------
equipped ozonometers is compensated by introducing a calibration factor
constituting a function of temperature. The calibration factor is obtained
by comparing a filter-equipped ozonometer with a standard ozonometric
instrument.
Recent models of universal ozonometers are equipped with the SZS-
9 glass instead of the SZS-18 glass. The passband of the SZS-9 glass
has a more gradually sloped left edge. This results in a lower dependence
of the second filter's transmission coefficient on temperature than in pre-
vious cases.
The following conclusions can be drawn from the statements pre-
sented above:
1. The solid angle of an optical ozonometric instrument must be
s mall. Its magnitude should be less than 6 degrees at least in two-dimen-
sional coordinates. A solid angle in excess of 10 degrees introduces an
additional error in ozone measurements which is associated with atmo-
spheric light scattering. The larger the solid angle and the lower the solar
elevation at the time of the measurements. the greater the error.
.2. Ozonometric instruments should be provided with means for
compensating errors due to temperature. Temperature errors in filter-
equipped ozonometers reflected the dependence of filter transmissivity"
on temperature.
BIBLIOGRAPHY
I. r y W H II r. n. H3Mep~He. o~~a.ccaM~~eTl. TpyJU,/" rro. IIhln. 93, I~
2. 11 0 p 41 o. n. Te""nepaTypllble xaplKTepllCTIlKH cBero4lH.IIbTpOB JIJIR Y.llb'P8.
. cjIHo.'eToeoll OOnaCTIl cneKTpa. CaeTOTeJlIIHKI. Nt 3. 1957.
3. Do b son G. M. B. Observers'handbook for the ozone spectrophotometer
Ann. Internat. Geophys. Gear., 5. No. 1-3. 1957.
- 131 -
-------
DIURNAL COURSE OF ATMOSPHERIC OZONE
A. Kh. Khrgian and G. 1. Kuznetsov
It will be shown in the following paragraphs that diurnal atmos-
pheric ozone variations amounted to 3 -4% of its total content. Nonetheless,
it reflects to a degree changes in air temperature at heights of 40 to 50 .....
km. A study of diurnal atmospheric ozone variations touches upon the
: processes of ozone origin and transfer by vertical and horizontal atmos-
pheric movements. It is well known that relaxation of photochemical pro-
cesses generating ozone in the upper portion of ozone layers above the
32 -35 km level is of the order of several hours. This means that diurnal
solar radiation variations occurring at those levels can cause delayed di.-
urnal ozone variations. Amplitudes of the latter were determined theoreti-
cally on that basis by Dutsch [lJ in 1954, who obtained values up toO. 0027
cm, and by Horiuchi [2J in 1956 (up to O~ 005 cm). Ozone maxima should
exist between the hours of 14:00 and 16:00. The present author tried to
determine magnitudes of diurnal ozone changes from observation data of
several IG Y observatories, such as the Reikjavik (Iceland), Vigna -di- Valle
(Italy), Elmas (Sardinia). The most reliable and comprehensive data were
obtained by the Vigna-di-Valle observatory, and the less detailed, although.
still reliable, by the Reikjavik observatory.
The frequently occurring considerable nonperiodic. diurnal ozone
concentration changes should be attributed primarily to atmospheric move-
ments, and possibly to observation errors. Nevertheless, calculatioJ,'ls of
mean values (x) for individual hours during any season enable the investi-
gator to distinguish systematic diurnal ozone content vari'!ltions. It was
determined with the aid of such calculations that ozone content in Vigna-di-
Valle increased by 0.005 cm during the period from July to August and be-
tween the hours of 09:00 and 16:00, (approxitnate time); the ozone co~tent
increased by 0.006 cm from September to November, between the hours
of 10:00 and 16:00. During the July-September period and between the.
hours of 09:00 and 16:00 (local time) (x) increased by 0.011 cm, and dur-
ing October-December between the hours of 08:00 and 16:00 it increas~d
by 0.016 cm. THe diurnal increase in (x) was less regular in Reikjavik.
- -- --._---- ----
- - - -------.---,.-
------....-. .- -
1957 r.
1958 r.
IAVERAGE~
. MONTHS: Vll-VIII IX-X
. Ho-uns' 9-16 9--' 16
z -: 0,002 +0.015
Ill-V
10-14
+0.001
VI- VIII
10--16
0,021
IX-X
9-14 9-15 . .
-0,002 +0.005
- 132 -
-------
Based on the averages of three stations, it can be as sumed that di-
urnal increase in (x) between the hours of 09:00 and 16:00 was approximate..
ly Ax. = 0.008 cm.- This value is somewhat higher than the one reported by
Horiuchi in Japan (6= 0.003 cm). The divergence between the two appears
even greater if it is considered that the present calculations were also based
on the far north _
-------
----
- -'.
COMPARISON OF OZONOMETRIC INSTRUMENTS MADE AT THE
MAIN A. T. VOEIKOVO GEOPHYSICAL OBSERVATORY
A. A. Znamenskii
Ozone measurements are conducted in the Soviet Union at the Main
Geophysical Observatory (GGO), at the Arctic and Antarctic Institut~, at
Moscow State ,University (MaU). and at other scientific -research establifh-
ments.' The importance of measurii"lg total atmospheric ozone has been
on the upgrade at these establishments, and the scope of the problem,
broaclenedeach year. Thus, an ozonometric network consisting of [six
up-to-date points has been organized by the USSRHydrometeorological
Service; provisions have been made for early future expansion to seventeen
. I
points distributed over the entire USSR territory.
Total atmospheric ozone determinations have been made with' differ-
ent ozone measuring instruments and the comparability of the data th,us ob-
tained is now challenging the attention of investigators. In addition, es- .
tablishment of an ozonometric station network demands that operatio~ally-
. suitable and reliable instruments be built of comparable high degree: of
, accuracy.
For the above reasons a comparative study of the, different ozono- .
'metric instruments was conducted at Voeikovo, the scientific-experimental
base GGO near Leningrad, in the early summer of 1959.. The followirig in-
struments were compared: Dobson spectrophotometer, MGU -bun~sEec-tro-
photometer, SF-4 spectrophotometer with a GGO attachment, Gushchin's--
universal ozonometer, the OFET -3 by Radionov, Bezverkhnii, and Oshero-
vich. . As a preliminary step a wavelength check was made for the Dobson
and MGU spectrophotometers. It was established that the two' shortwave
slits of the Dobson spectrophotometer discriminated two intervals with
centers at 3114 and 3323 A. Concurrently with the wavelength check, the in-
struments I temperature compensation was checked; it was found that the
. I
compensation tolerance was + 0.7 and + O. 3 units, respectively. Th~se
values were regarded as fully acceptable. At to = + 10° (the MGU -bui~t'
spectrophoto~et~:r discriminated two spectral interv'als ,with. centers ;at
3114 and 3326 A. j . ' ;
The SF-4 spectrophotometer used in this comparative study was
equipped with a set of ZhS -3 and UFS -2 filters. The latter filter was' high-
lyopaque, and its transmission coefficient at A = 3i2 mIL was 12% instead of
80-90%. It was ~ecided not to replace the UFS -2 light filter by a spare
high transmissi6n coefficient filter, since spectromete'r SF-4 was cali-
brated as a monochromatic light source equipped with an old UFS -2 filter.
Therefore, in making measurements with spectrophotometer SF-41 its
slit was increased to 1 m m. The sensitivity of spectrophotometer SF -4 :
- 134 -
-------
was inadequate for making ozone measurements with the selected slit due
to the ins ufficient filter transparency. For this reason the comparis on re-
fleeted only measurements made at solar elevations above 25° .
The ozqnometric instruments comparison was conducted by making
ozone determinations simultaneously with all the above mentioned direct-
radiation instruments at 10 min. intervals. Ozone measurements were
made at solar elevations from 15 to 52 degrees.
Simultaneous measurements were made during cloudy days on 2,
4, 5, and 7 June, making 62, 64, 13 and 9 concurrent respective readings
amounting to a total of 148 readings. Maximum ti me dis pe rsion during
measurements made with all the instruments stayed within the range of
:i: 2 min. Measurements were made by the following observers:
A. S. Britaev and A. V. Shestakova (Central Aerological Observatory-
TsAO), R. G. Romanova, K. 1. Romashkina, V. B. Aleksandrovich, A.M.
Shalamyanskii (GGO), and G. P. Gushchin, (using the Dobson and SF-4
spectrophotometers). Observers were changed around periodically so
'that 2 -3 observers had the bppportunity to operate each instrument. Solar
elevation was determined with the aid of ShT theodolite with an accuracy up
to O. 1° .
Total atmospheric ozone was computed on the basis of data ob-
tained with the Dobson spectrophotometer using the following formula:
.. -X' ~ 'L~':"'-L-O.lrm. The formula was used since the beginning of the
0.86511 IGY and, as in following formulas (X) represented
total atmospheric ozone in em, L denoted the logarithm of the ratio of
direct solar radiation intensities at two wavelengths (with an accuracy up
to the constant term); m and IJo represented ozone and air masses, respec-
tively. The value of constant ~ = 0.07 was obtained prior to the experi-
ments from the L (1Jo) curves plotted in 1957 and 1958.
Computation of total ozone from data obtained with the MGU -built
spectrophotometer was made allowing for changes in the monochromator
wav~length due to temperature (Table 1), and change in the optical wedge
transmission coefficient.
Based on test results formulas were derived for MGU and SF-4
spectrophotometers independently of the Dobson spectrophotometer;
in this connection ~ values were obtained from the graphs of L (1Jo) which
were plotted from-data secured using spectrophotometer MGU for June
4, 5, and 7, 1959, and spectrophotometer SF-4 for June 2, 4, and 5, 1959.
Mean values were computed on the basis of three ~ readings for each in-
strument; and the cor'responding formulas were as follows:
a) Table 1 for spectrophotometer MGU
-" -...- -- .
--.-----.-
b) X=
2.458 - L -O.II.m
0.86511
for spectrophotometer SF -4.
- 135 -
-------
.-.:. _.T.a.-QJ.~_L__--..,,- .----------.--. - --. .
WAVE LENGTH AT DIFFERENT APPARATUS TEMPERATURE AND C1.., C1..1
.' 'tEMPERATURE - 44°C
._-------_..~~
AND ~, ~1
VALUES AT OZONE LAYER
).
).'
"
12'
I
0,462 I
0,346
3108 +30 0,931 0,460 0,864 0,113 LO-L -0,131
3320 .0.0665 0,347 JC=
. 0,864 f1
3111 +20 I 0,921 0,458 0,864 0,110 Lo-L -0,127
3323 0,0568 0,348 JC= ,O,I!65 f1
3114 +10 0,912 . I 0,456 0,107 :X.... Lo -L -0,124
3326 0,0466 I 0,349 0,865 0,865 f1
; Lo-L .
3117 00. .0,890 0,454 0,850 0,104... -0,122
3329 0,0400 0,350 JC= 0,850 f1
3120 1='0.- 0,859 0,452 0,897 0,101 l.e-L -0,113
3332 0,0417 0,3514 "... 0.897 f1
-----~. --- --_._-~, '
---- -- ----------.. . ----- ----.--- --- "-.- ------------ .~-- -
I
3105 I
3317
IAPPAi\'A'Tud
, TEMP../
I +40. I
I ~-p' I
I
\12'-12'
I 0,888
. ~'
WOR KING F oiiHU"L-A .
~
0,116 ILL
JC= 0- -0.131
o 888,...
0,962 ! 0,0735 I
NOTE: COEFFICIENTS C1.., C1..1, 13, 131 WERE AVERAGE ON THE BASIS OF 8 AND 14 A
. - INTERPOLATION VALUES
Total atmospheric ozone based on data obtained with the aid of a
universal ozonometer was determined from a special ozone nomogram
shown in Fig. 1. Instrument temperature correction was' made by multi-
plying obtained data by a coefficient deduced from a comparison of the
universal ozonometer and the Dobson spectrophotometer during February
- May 1959.
Fig. 1
?,
3z
Z.O
1,5
1,0
0.5
o
Z'l..
30 '
40
50
-- - - ...~
50
70
10
DEGREES OF SOLAR ELEVATION
- NOMOGRAM"OF-TOTAl-ATMOSPHER-IC OZ-ONE-DETERMi NATION. GGO No.3
OZONOMETER USED AT 625 MM PRESSURE
- 136 -
-------
Temperature coefficient values of the
universal ozonometer are shown in
Table 2.
----. .. - - - --
- ------ .--.u
~- .-. - -+- .-
Prior to using the available spec
trophotometer OFET-3 it has undergone
a basic overhauling and check calibration
against the Dobson spectrophotometer,
as a result of which the following formula
was obtained for (X)
-.UN-'V ERS~L -~~~~OH~~; ~RA~U~~-I~~~-O~FF -.:---
CIENT VALUES AT DIFFERENT TEMPERATURE
r/k1r/k/r/
.
o 11,35! 12 /1,32 23 1,18 x = 0,555-L_O,19.
~ I: :~~ :~ I::~ ~~ :::: 0,525:~
3 1,35 15 1,29 26 1.12 . Thus. four of five available in-
4 1,35161,2827 1,10
5 1,34 17 1.27 28 1.08 struments were checked against each
~ ::~: :~ ::~~ ~ ::~ other. and one (OFET-3) against the
8 1,33 20 1,23 31 1,03 Dobson spectrophotometer. Results of
9 1,33 21 1,21
10 1,33 22 1,20 total atmos pheric ozone calculations
11 1,32 bas ed on readings obtained with the five
. . instruments are shown in a graphic for.en.
in Fig:; 2, whe:te total atmospheric ozone in cm is shown as a function of
Moscow time. There was considerable disagreement between readings re-
corded by the Dobson and MGU -built spectrophotometers on June 2. This
: called for an additional adjustment of the MGU instrument by way of im-
proving the operation of a modulator. No further adjustments were made
: after June 2.
Fig. 2
---.------. ---
~--_._--- -- - --- - -- . -~ ~-
-.- --- -"
z:
(.) 0.100
z
&II a~oo
z
0
S .0.300
-'
c
I-
0 6 ,
I-
10
12
2 JUNE
/1,
/6
/8
6 8 /0 /2 II, -16
-- -- ~~~~ruBi_--.- .
HOURS OF Moscow TIME -
-----2 _..- -J ._.no..-of
18 6 8
5 JUNE
6
10 12
.7 JUNE
-I
----- ,
------'.-.--'-
----- -- .
.-_..- .-.
..----------' -----. - .
COMPARISON AND TEST CURVES OF OZONOHETRIC DEVICES: I - DOBSON SPECTROPHOTOMETER;
2 - MGU SPECTROPHOTOMETER; 3 - SPECTROPHOTOMETER SF-4; 4 - UNIVERSAL OZONOMETER;
~OFET-3 DEVICE.
Curves in Fig. 2 show that, except for the MGU spectrophotometric
. data of June 2, maximum deviations of total atmospheric ozone values' ob-
tained by t}1e five instruments did not exceed 23% of the values obtained by
the Dobson spectrophotometer. Average daily deviations were consider-
ably smaller. Average daily deviations in total atmospheric ozone ac-
cording to results obtained by the Dobson spectrophotometer are listed in
- 137 -
-------
-----.-----
Table 3.
: DA IL Y AVERAGES OF SIMULTANEOUS TOTAL ATMOSPHER I C' OZONE DETERMI NAT IONS
MADE WITH DIFFERENT DEVICES AND PER CENT DEVIATIONS FROM RESULTS YIELD-
ED BY 'THE DOBSON SPECTROPHOTOMETER
N
--------- OZONE -" ,--
OZONE
AVERAGES~ AVERAGES DEVIATIONS
- ------- ACCORDING
AME OF APPARATUS [JATES-OF' 'TESTS BY BY
TESTED DOBSON TO
DEVICE APPARATUS DOBSON
"
UNIVERSAL
OZONOMETER
2/VI 1959
4/VI 1959
5/VI 1959
7/VI 1959 r
r. 0,394 0,373 +5,6
r. 0.357 0,390 -8,5
r. 0,388 0,384 +1,0
0.372 0,406 --8,4
".---------- ---
AVERAGE -2,6.
r. 0,365 0,373 -2,1
r. 0.353 0.390 -9,5
r. 0,328 0,384 -14,6
-- ------- .~u- -
SPECTR OPHOTOMETER 2/VI 1959
SF0.4 4/VI 195Q
S/VI 1959
. AVERAGE
.
-8,7
, '-MGY
SPECTROPHOTOMETER
2/VI 1959 r.
4/VI 1959 r.
S/VI 1959 r.
7jVI 1959 r.
.0,465
0,393
0,464
0,390
0,384
0,410
+19,2
+2,3
+13,2
AVERAGE
+11,6
. Data in Table 3 show that the mean deViation from the Dobson
spectrophotometer through the measurement period was + 11.6% for the
MGU spectrophotometer, -8.7% for the SF-4, and -2.6% for the universal
instrument. Mea,n daily deviations for the entire period 'of simultaneous
observations, except for June 4, did not exceed lS'ro. Readings of spectro
photometers MGU and SF-4 showed a comparative greater mean deViation
from the standard instrument. This may be explained by the fact that
the magnitude of Lo for these instruments was determined 1) from a l,imited
three series of observations and, 2) by the instrument's temperature' ef-
fectwhich in the case of the SF-4 was totally disregarded and only partial
. allowance was made in the case of spectrophotometer MGU.
CONCLUSIONS
1. A comparative stu<,iy of ozonometric instruments showed that
values obtained in determining total atmospheric ozone I!'imultaneously by
different instruments deViated from those recorded by the Dobson spectro-
photometer by aSrrluch as 23%.
, 2. Total atmospheric ozone curves plotted from ,data obtained by
different instruments exhibited a general parallelism with an occasional
departure.
- 138 -
-------
I !
.3. The comparative study established that a universal ozonometer
was the most suitable instrument for measuring total atmospheric ozone at
. all points of a station network. .
4. All stations of an ozonometric network should be equipped with
identical instruments, preferably of the above described universal type.
5. Parameters of calibrated instruments should be left unchanged,
so that the same instruments can be used as the standard in future compara-
tive studies of the instruments r stability with time
6. Amore detailed investigation of all ozonometric instruments for
linearity, dispersion, wedge, effect of temperature on readings, etc. ap-
pears mandatory. .
7. It is recommended that nomograms be used by all stations of
. an ozonometric network for simplification and acceleration of data pro-
cessing.
! 8. Standardization of ozonometric instruments should be done in
I
. regions where clear air and favorable weather conditions predominated.
This should be accomplished in 1961.
., RESOLUTIONS OF THE CONFERENCE ON ATMOSPHERIC OZONE
HELD IN MOSCOW, ON 28-31 OCTOBER, 1959
. 1. The Conference notes with pleasure that USSR investigations in
the field of ozonometry, which had been conducted previously as partial,
occasional, and uncoordinated efforts, have been organized in recent years
on the basis of IGY into a coordinated common working program and that
the number of problems and participating scientists have been rapidly in-
creasing. Reports presented at the Conference were of a great scientific
and practical value.
2. To enhance the effectiveness of scientific investigations in at-
mospheric ozone, the Conference intends to form a joint committee on at-
. rriospheric ozone studies which would provide communications, more de-
tailed information, and would coordinate the ozonometric achievements
attained by different organizations. The committee should also establish
permanent communication channels with foreign institutions which are en-
gaged in the study of atmospheric ozone problems. In developing a unified
plan of investigations, the committee should insure a maximum complexity
- 139 -
-------
of all achievements, and the participation of all possible institutes in the
org~nization of ozonometric studies, and b) development of ozonometric
equipment for networks and observatories. and the standardization and
approval of applicable instruments. Furthermore, one of the priority
tasks of such a committee in 1962 should be the standardization of all
ozonometric instrument systems used in the USSR, including the Dobson
instrument and the one used at ,Moscow State Unive rsity.
3. In the field of observation methods improvement and analysis
of ozone layer physical regularities the Conference recommended that a
special study be made of the physical nature of the, so-called, circulation
effect, necessary for the purpose of establishing definite methods of com~
puting vertical ozone distribution. The Conference deemed it essential to
develop a more accurate method for determining allowance for light scat-:-
tering due to atmospheric aerosols. . .
4. Many significant conclusions have been arrived at regarding
sea~ona1, latitudinal, and longitudinal fluctuations in'atr~lOspheric ozone
content. These were determined by observations ma'deduring the IGY at
USSR stations established in sections of different geographical conditions,
i~ the i\.rctic Abastumani, in the Far Eastern VladivQstok, etc. In addi-
tion, tne GUGMS resolved to extend ozonometric station networks in 1960
, , .
- 1961. Based on the above, the Conference appealed to other/organiza-
tions to take action which would insure the continuation:9f the following
initiated problems: -
a) the IPG AN SSSR at the E1'brus Station of the
institute at Tersko1
b) the Arctic and Antarctic NIl at Dixon and Heiss Islands
c) Academy of Sciences, Gruz SSR at the Abastumani Observatory.
5. Since extensive and significant atmospheric ozone investigation~
have been initiated and a considerable volume of original observation mate:"
rial was accumulated each year on the basis of which conclusions have been
drawn, th,e Conference found it desirable to convene similar conferences of,
specialists on the atmospheric ozone every 2-3 years.
6. The :Conference recommended urgently that its proceedings be
published promptly, and requested all contributors to present their manu-
. . . I /
scripts to the editor not later than 31 December, 1959. /
- 140 -
-------
I
I !
I :
I i
Part II
-------
/'"
FOREWORD
. - .-. Development of studies in tIle _fiel~ of atmospheric ozone necessi-
. tated coordination of the results. The first attempt ip. that direction I
- was made by the A. 1. Voeikov Main Geophysical Qbservatory (IGY) at
. the First Inferdepartmental Scientific Conference on Atmosph.eric Ozohe
which convened in Leningrad 4-5 April 1957. This conference, the firlst
in the Soviet Union, was devoted to the organization of the International
Geophysical Year (IGY) which was to be initiated in July, 1957. The S~cond
. Interdepartmental Scientific Conference on Atmospheric Ozone was held
-.-G;Mo.s.c-ow.. at --MoscowStateUniver-sity (M5U) d-uring 28-31 October,. i
- - . I
1959, when some results of ozone investigations conducted during the IGY .
I
were presented. The Third Interdepartmental Scientific Conference on:
Atmospheric Ozone was held in Leningrad at the Main Geophysical Obs~rva-
tory during 21-2.3 May, 1963. The Conference was attended by represehta-
tives of the following institutions: Main Geophysical Observatory, M~s-
cow State University~~C.eriFraI: Aerological Observatory, Central Fore~
casting Institute, Leningrad State University, Arctic and.Antarctic Sden-
I
tific-Research Institute, Institute of Terestial Magnetism and Radiowaive
Distribution of the USSR Academy of Science. The Third Conference was
convened on .the eve of the International Quiet Sun Year (IQSY) in 1964-1965)
. and was devoted to the preliminary preparation for this important undertak-
ing.
This report contains papers delivered at the Third Conference.
The remaining conference papers, not contained in this collection, appeared
earlier in other publications, or had not been completed at the publication
deadline. G. P. Gushchin's papers entitled lIThe Recent State of the At-
mospheric Ozone Problem" and "The Turbulence-Photochemical Theory
of Atmospheric Ozone" were incorporated in his monographs under the
titles of "Atmospheric Ozone Research" (Issledovaniye atmosfernogo ozona,
Gidrometeoizdat, L., 1963), and "Ozone and Aerosynoptic Conditions in
the Atmosphere" (Ozon i aerosinoptickieskiye usloviya v atmosfere,
Gidrometeoizdat, ~I964). Kh. A. Khrgian's confere-;;'ce paper entitled
"Some Observation Results- of the World Ozonometi"ic Network During the
IGY -IGC" was published in "Meteorology and Hydrology" No.1, 1964) is
I
part of an article entitled "Atmospheric Ozone, - Some Results of IGY \
Studies)". . K. 1. Romashkina's conference paper entitled "Ozone and Wind
in the Atmos phere" was incorporated basically in a paper published in
Trudy (Proceeding-s) GGO, no. 154, 1964. R.S. Steblova's conference paper;
entitled "A Model of a Nonstationary Ozonosphere" was incorporated in her
article published in "Geomagnetism and Aeronomy", No.2, 1963.
- 142 -
-------
I .
The present volume also includes atmosphe ric ozone observation
data recorded at Soviet Stations for 1963 which were compiled for publi-
cation at the Ozonometric Section of the Governmental Geophysical Ob-
s~rvatory (GGO). The value and importance of scientific conferences
. are frequently found in the discussions raised by the presented reports.
Reports dealing with the following problems elicited prolonged discussions
at the Third Interdepartmental Scientific Conference on ozone: the role of
vertical movements in ozone oscillations, ozone changes during solar
eclipses, recovery time of the photochemical ozone balance. and explana-
tion of the "anomalous transparency" effect. The discussions showed a
necessity of further compilation of facts, particularly those which per-
tained to ve rtical ozone distribution. and- furthe r development of theoretical
aspects of many ozone problems. It was established at the Confe rence that
ozone played an, important role in many meteorological processes. particu-
larly in the formation of the thermobaric field of the stratosphere and
mesosphere.
A program of atmospheric ozone investigations in the USSR during
IQS Y was reviewed and indorsed at the closing session (see "Informatsi-
onnyi sbornik GUGMS". no. 1, 1963). Reports of the Conference's or-
ganizing committee dealing with the atmospheric ozone terminology, units,
and symbols were taken into consideration. It was recommended particu-
larly that units used in recording vertical ozone qistribution be referred to
as the density or partial pressure units rather than concentrations.
The Conference adopted a resolution formulated in this collection.
CA USES OF RAPID WINTER TEMPERA TURE VARIATIONS
IN THE ARCTIC STRA TOSPHERE
G. P. Gushchin
New and important properties of the atmosphere's structure and
dynamics had been discovered recently as a result of the radiosonde climb
ceiling elevation. Rapid and significant temperature inc::reases' had been
discovered in the stratosphere at high and intermediate latitudes amounting
at time s to 20 degrees per day a:t altitudes of 20 to 30 km. It soon becamE
apparent that this effect represented a large-scale process in the strato-
sphere associated with a high-altitude baric field rearrangement and the
shifting of great air masses. The rapid winter temperature increases in
the Arctic stratosphere had been previously described in several reports
(4, 10. 12, 13). Cases of rapid stratosphere temperature increases at'
inte rmediate latitudes were described even earlier (26, 27). Howeve r,
reasons offered for rapid stratosphere temperature increases differed con-
I
- 143 -
-------
siderably. Scherhag (26) was of the opinion that increase in solar activity
played an important role in the phenomenon. Warnecke (30) believed
that air mass circulation induced advection in the stratosphere causing the
noted temperature rise. Veksler (3) believed that the stratospheric te.mper-
ature rises were independent of the solar activity and that they resulted from
adiabatic heating of air which descended from the upper layers. This was in
agreement with the opinion previously expressed by Teweles and Finger (29).
G. D. Zubyan (10) believed that the rise in the stratospheric temper-
ature observed by him in January of 1958 was caused by the transfer of air
masses from one latitudinal zone into another. Kh. P. Pogosyan (13)
and A., A. Pavlovskaya (12) believed that temperature increases were :the
result of advection and vertical movements and that the role of the former
was more basic. S. S. Gaygerov (4) was of the opinion that the intensive
meridional exchange caused stratospheric temperature increases.
, Many investigators pointed to the possible existence of a relationship
between rises in stratospheric temperature and atmospheric ozone. For
instance, Teweles and Finger (29) noted that the January 1958.,strato-
spheric temperature increases in Greenland in the Great Lakes area had
been accompanied by a significant increase in total ozone (up to 0.48-
0.49 cm 03 at the Alert and Resolute Stations): . Diit sch (18) and London (22)
believed to have noted a close relationship between atmospheric ozone and
stratospheric temperature increases in Europe and America in January -
February, 1959. S. S. Gaigerov (4) advanced a hypothesis that strato., ,
spheric temperature increases were associated with ozone transfer from
the Arctic and Subarctic. The Arctic stratospheric air masses, ri'ch
in ozone, first moved into lower and subsequently into higher latitudes,
warming up along the way and causing rapid increases in temperature of
the stratosphere at high latitudes. In Refs. (6. 8) the author assumed
. that the rapid temperature increases in the stratosphere over the Arctic
were caused by an intrusion of warm stratosphe ric masses with an in-
creased ozone content. These masses formed over the northern portions
of the Pacific and Atlantic Oceans as the result of a relatively low water
- _uu~~-~fac:_~ atb~doas compared with dry land, and due to increased turbulence
in these regions. The purpose of the present article is to substantiate this
as s umption.
It has been known that vast areas of the Asian continent north of
500N became covered with snow beginning with the month of October.
Snow ~an be also seen in the northet'n portions of the Amer~can continent
at such time. The water surface of the northern portion of the Pacific
Ocean extended between these two continents. The northern portion of
the Atlantic Ocean. with a water surface smaller than that of the Pacific,
also lies between two continents, namely - North America and Europe.
The snow and water surface albedos are essentially different.
- 144 -
-------
.'
According to data published in K. Ya. Kondrat'ev's book (11) it can be
assumed that under average conditions the snow-cove'r albedo amounted to
I
70%. while that of the water surface amounted to only 100/0. In this con-
\. nection conditions for the irradiation of ozone layers over dry land and
over water are different during winter and to a lesser degree during sum-
mer. The ozone layer received an additional amount of reflected radi~-
"tion over dry land which. as was indicated by the author in Ref. (8). breaks
". "the ozone down. The reflected radiation was basically in the visible range;
this made it effective in the part of the ozone layer where the photochemical
balance had not been established. Moreover. turbulence -photochemical
balance is generally established at a considerably greater rate in the, upper
" portion of the ozone layer. i. e.. at heights from 20 to 40 km. Ozone break-
down occurred at a slow rate in the lower layers. and the above mentioned
balance was practically never attained. Consequently. additional oi,one
breakdown takes place in the 20-40-km layer over dry land due' to i-eflected
radiation. "
Decrease in ozone density over dry land is noticeable only when the
westerly transfer at heights of 20 to 40 km leading to the appearance of
air masses intermittently over the ocean"and over dry land, is either
small or totally nonexistent. A study of baric topography charts of 50-, ?O-,
and 10-mb surfaces indicated that the westerly transfer at such levels was.
small during autumn. At the beginning of autumn a rearrangement of the
high"-altitude baric field occurred at these heights. whereby a circumpolar
"anticyclone was replaced by a circumpolar cyclone. and "the easterly winds
became westerly. Moreover. numerous observations indicated that a
minimum wind velocity occurred at heights from 20 to 30 km at inter-
mediate latitudes. Under the effect of ground-reflected solar radiation
ozone density at latitudes from 40 to 65° N and at heights of 20 to 40 km de-
creased toward the end of autumn over continents as compared with oceans.
The powerful turbulence zones which existed in some coastal re-
gions of the Pacific, and to a lesser degree of the Atlantic. and which ex-
tended to considerable heights in the atmosphere were factors of consider-
able importance. Of particular significance are the winter and spring
turbulence zones over the eastern coast of the Eurasian continent. During
those seasons. strong jet streams prevailed. the upper boundary of which
reached the 50-mb level and higher. The author of monograph (8) ob-
served a considerably increased ozone density at the 300 -50-mb levels
in jet stream regio'ns. These were the results of measurements taken at
the..European stations. It is conceivable that even a greater increase in
ozone density'in the eastern coastal region of the Pacific Ocean may be ob-
served at greater heights since the turbulence zones in those regions ex-
tended to greater heights. A similar, though less pronounced case was
observed over the eastern coast of North America and over England. The
- 145 -
-------
).,
presence of jet streams and, consequently, of turbulence zones in the
coas,tline regions can conceivably explain the difference in the albedo of
. continents and oceans. On the other hand, numerous observations show
that the greater the amount of ozone in the stratosphere, the higher is the
air temperature. Thus, in accordance with Dutsch's data (18) the coeffi-
cient of correlation between the ozone density and air temperature fori
Arose. Switzerland at the IO-mb level in January, 1958, was 0.82. '
It is believed that the increased temperature in air layers with
higher ozone density could be regarded as the result of additional solar
radiation absorption by the excess (with respect to the surrounding area)
of ozone. Author of monograph (8) had shown that the rate of air heating
due to solar radiation absorption by the ozone was directly proportional
to the ozone density in a given air mass. Thus, the rate of stratospheric
heating in the northern hemisphere due to absorption by ozone of the solar
radiation was greater over oce<;tns than over dry land during autumn and
winter. This fact leads to a redistribution of the stratospheric tempera-
ture field, so that thermal regions which formed during wintertime over
the northern areas of the Pacific, and to a lesser degree of the Atlantic,
, I
were warmer than those which formed over dry land. Moreover, forrria-
Hon of ' thermal regions in the stratosphere over oceans created closed
anticycl6nic circulation and isolated the ozone-rich air masses which aug-
, mented the stratosphere heating.
However, the above general arguments must be substantiated by
proper computatiqns. The observed horizontal and vertical ozone dis-
tribution in the earth's atmosphere were explained in (8). The manner
in which the ground-surface albedo and the turbulence affected the vertical
ozone distribution will be examined in the light of that explanation. Calcu-
lation of vertical ozone distribution in the 0-30-km level was presenteCl in
(8), allowing for the effects of turbulence and of photochemical reactions.
However, the author's calculations were not precise with respect to an
air density change with height. In calculating the vertical ozone distribu-
tion in a turbulent atmosphere the present author made allowance for
changes in air density with height. Moreover, a different method was used
for taking into consideration the effect of solar radiation on o~one. \
The differential equation of the turbulent diffusion ,of atmospheric
admixture is (15) as follows:,
a'S. - - 1 a (as ) 1 a (as) .
at = - Vvs + Tax- pk.r ax +p: iJy pky ay +
1 a ( k iJs)
+ p iJz p z iJz '
(1)
- 146 -
-------
,
where s - denotes average amount of admixture under consideration
per unit air mass, or the mean ratio (by mass) of admixture to air.
If ozone is considered as an admixture, and if an ozone source exist-
ed, equation (1) can be converted into the following:
, "
ar; - -- 'I a ( ar)
at = - Vv ',3 + p iJx pkx iJ~ +
+ ~ ~ (k 3~) - ~ ~ ( ara) -
,', p iJy P Y iJy r p iJz pkz (fZ +.W3,
(2)
where r3 - denotes average mass concentration of ozone ra,ti,o of ozone
to air in the atmosphere which depended on x. y, z and t; V - symbol-
izes mean wind velocity vector;
o 0 0
'il = i- + j- + k- - del; r. J. k -
Ox oy oz
-.-----.- -~ _._----- --,
unit coordinate vectors: V 'ilr3 - scaler product of the V and Vrs vectors
which corresponded to the advective changes ln r3; P - air density; k . k k
x y, z
- :turbulence coefficients along the x, y, z axes, respectively; ';'3 - mean
. ra~e of change in ozone concentration due to the ozone source.
It should be noted that in equation (2) the ozone concentration rs
and not its density P3 is the unknown variable.
In the case of atmosphere the density of which remains unchanged
with height (due to lack of gravitational field}, latitude, and longitude,
qu~ntity r3 in (2) is replaced by P3.
In fact, in such a case consider that
P3
'3=-
P
(3)
. -. -. - __.0.. -
and that p = const. and get from (2)
~.=-vvPa+ :x (kx ~~)+ ~(ky ~~)+',
+ :z (kz :~) + W3 .
(4)
Equation (4) is obviously, applicable in cases of comparatively thin
atmospheric layers.
- 147 -
, .
i .
II
-------
tions:
Solve equation (2) on the basis of the following simplifying assump""'
1) advective ozone transfer in the atmosphere at heights
from 10 to 30 km is zero, 1. e., V Vr3 = 0;
2) turbulent ozone transfer along the x and y axes at heights 10 - .
30 km is zero, 1. e., the second and third terms on the rhs of equation (2)
are ze ro: .
3) air density varies with height exponentially
-..-' - . -- -- .-_. --, "un
z
-71
P = Po e
(5)
!
where Po - air density at sea level, H - height of homogenous atmosphere
(H = 7996 m); . i'.
4) turbulence coefficient in thelO-30-km layer k = k = const;
z
5) under constant conditions,
1. e.,
Pr .
3. - O.
- - I
pt
6) decrease in ozone concentration occurred in the 10-30 km under,
the effect of 'solar radiation, and rate of decrease - two cases are con-!
sidered - is:
. - . --------------- .
"----.. .~_._.
(6)'
a) w3=-bDr3,
.' ,
where b = const. D - average intensity of solar radiation; 'for the
sake of simplicity it is assumed that D was independent of z;
.. _-n.__- '--.
6)w =-w e-b, (30-z)
3 3,30 ,
(7)
whe re W3.30
bl = const.
- average rate of decrease of ozone concentration at 30 km,
I '
i . ~
; The above assumptions reflected in a measure average conditions
of ozone distribution in the 10-30-km layer for.a sufficiently long period
of time (e. g., one year). ,.
1£ the above simplifications are applied to equation (2)' and co~ -
side ring that
- 148 -
-------
. -
~ ( k Or3) = k [~ or3 + o~ r3 ]
OZ P OZ . OZ OZ P Oz2
. . - -. -. - . - -- .. - -_...-------"--
and that on the basis of (5)
op - p
dz -:- - H'
(8)
(9)
we get in the
'. . .
first case (assumption 6) .
o2r3 1 o~D
dz2 -HdZ -bTra=Q
(10)
---------- --
and in the second case
02 r3 1 or3 W3,30 -6 (~O-Z)
di2-Hd"Z-lle' =0.
(11)
Now, solve equation (10) as follows: Let
and
~ 1
P-'-.H'
D
q . -bT.
(12)
(13)
Equation (10) can then be written as follows:
r; -+ pr~ + qr 3 = o.
(14)
The general solution of a second-order nonlinear differential
equation (14) has the following solution:
where
r = c eg, z + c -1(. Z
3 1 2 t:- ,
(15)
gl=-1-+ V(+Y-q,
.g2=-~-V(~r-q
(16)
(17) -
and gl > O. g2 < O.
The above two inequalities follow from equations
and the evaluation of p and q (Table 1). (1) Values of gl
relation to q (km- 2 .
(12 and (13),
( -1
and ga km it:l
- 149 -
-------
~
\
Table 1
--_. --, ----.----~._---- -----
( -I IN RELATION TO q (KM-2)
Ql VALUES OF GI AND G2 KM )
q. . -0,001 --0,002 --0,004 --0,008 --0,016
11 ( ~ r -:q- 0,0700 0,0768 0,0889 0,1091 0,141
gl . . 0,132 0,139 0,151 0,172 0,203
g2 . --0,0075 -0,0143 --0,0206 --0,0466 --0,0785
Find the solution of equation (14) under the following limiting condi-
tions:
--.- ----- .------...
-. ---..
Z=ZT'
"<:3 = '3, T ;
(18)
(19)
Z = 30 KM, '3='3,30'
whe re Z
'I
-, height of tropopause.
I
The limiting conditions (18) and (19) indicate that a given ozone con-
centration was maintained at 30 km at all times, due to photochemical re-
actions, since a photochemical equilibrium existed at that level, excluding
from conside-ration the seasonal ozone concentration oscillations at that
, height.
,Substitute equation (18) and (19) into general equation (15) and get: ,
r iOg, - , eltlg,
3, T ~,~O
Cl = elOg, i'g, - iOg, elilg, ..
"-- '3, T iOg, + r3,30 elOg,
el:1g, iOg, - e3~'g., e\IJg, .
(20)
C2=
(21)
In the above equations it was assumed that Z= iO"'km.
, ,.
The equations can be simplified, provided the values of all terms
puted (Table 1). As a result of such simplifications obtain;
,
wer~com-\
and
-30g,
Cl ='3,30 e ,
- e-10g, -, e- (20g, 'HOg,)
C2 - '3. T 3.30 .
(22)
(23)
tions
Solve the genera11inear differential equa~ion by substituting equa~
(22) and (23) into (15) and get the folloWing performance formula:
- -. : . ,
, -r-'e~nO-Z)g2+, e-(3J-Z)g.,-, e-(20g,+log.-zg,) (24)
3 3. T 3.30 3, 30 , .
- 150 -
-------
The value of r3 was calculated for different values of z and q.
Furthermore, in accordance with existing data on ozone concentration
(8), it was assumed that
;- - 0-5.10-6 - r03 --
3, T, r - ,fir;
(A)
and
r330= 127.10-6 r03_-
, , r Ai,.
(B)
. The results of calculating gl, ga and r3 using formulas (16),
(17), and (24), are shown respectively in tables 1 and 2.
Values of r3 (10-3 rO.3
g alr
in relation to z and q
Table 2
------- ~~~ OF R - ( ; 0.:51103 )-
3 G AIR
. .. ." -0,00' j -0,"
- -.--..-
'" RELATION TO Z ANO Q
q x..-2
-0,004'
-0,008
-0,016 -
10 0,5 0,5. 0,5 0,5 0,5
15 1,39 1,30 1,21 1,03 0,78
20 3,10 2,92 2,72 2,34 1,79
25 6,21 6,12 5,88 5,42 4,86
, 30 12,7 12,7 12,7 12,7 12,7
Ozone density curves P3 obtained by recalculating data in Table 2
are shown in Fig. l-a as functions of altitude. The calculation of P3 was
made by the following formula:
P3 ( 10-3 CM 03) = 16,3 ' 10-6r3 ( ~_9~ ) P (M-O)
KM \r AIR T CK)
(25)
for the average values of pressure P and temperature T at the investigateq
!
heights [7J.
The curve of ozone density above the 30 -km level, shown in Fig. 1,
was calculated according to -the- photochemical theory [17J.
:Fig.') shows that maximum ozone density was observed at 15 to
25 km., and that maximum density height decreased with an inc rease in
ozone density in the 15-20-km layer. In addition, an increase in parameter
- 151 -
'.1
-------
q gave rise to two maxima observed on density curves 1 and 2..
ditated cases are confirmed by obse rvation.
Fig. 1
I(/tf
50 a)
\
40 \
......
......
30
....
eI
=>
...
;: 20
-'
40(
10
""
,/
""
I
o
5 10 15
----- .~---- - --.-"--.
Ozone
T he in-
\
'-
o 5
.------ .----. -
DenRity
10 .15
10 "3C'" OJ
X14
---.-VERTI-CAL-OZONE --OISTRI-a-UTloN- -- -
A-ACCORDING TO FORMULA (24) .
- - 2 -2
I) q- -0,016 KM-2. 2) q- -0,008 KM- . 3) q- -:.0,004 KM .
-----&:~~~O~~'~~G 1i~:~~ofM~~A~t~(..-":,- .
1) C=8. 10-12, 2) C-6' 10-12, 3) . C=4. 10-12,
4) C-2. 10-12, 5) C= 1,0. 10-12.
It follows from Figure I-a that ozone density increased in the 10 -30-
km layer with a decrease in absolute value of q. Hence, bearing in mind
equation (13), it follows that constant ozone density in a given layer in-
creased with increase in the turbulence coefficient k, e. g., in jet streams,
and with a decrease in radiation intensity D, e. g., with increase in lati-
tude. The above ozone distribution law can be actually observed.
Equation ( 11) for the second case was solved under the following
limiting conditions:
r3(0) - 0,
r3 (30). r3.::/
(c)
Solution of equation (11) yielded the following working
- - -' 0"2"41- .( 0,125z 1)
r3 == 0, r3.30 e - -:-
- 0,946. lO~J 0,0241 (eO,l25Z - 1) - e-1.093 (30 - Z)] .
formula:
(26)
- 152 -
-------
This formula was used in computing vertical ozone concentration
distribution in the 0 --30 -km atmospheric layer as a function of paramete r
c: -
w' -
C = 3.30,-
k .
(27)
Subsequently, values of r3 were recalculated by formula (25) to
obtain ozone density values P3.
Results of P3 computed by equations (26) and (25) are shown in
Fig. I-b.
It should be noted that the magnitude of W3 in equation (27) was of
the first orde r of approximation, directly proportional to solar radiation
intensity. It follows from Fig. l-b that the law of the stationary verti-
cal ozone distribution in the second case was similar to the first case
(Fig~ I-a) except that in the first case the dependence of P3 on the parameter
q was ,more pronounced that the dependence of P3 on the parameter C in the
second case. The above discussed laws governed the theoretic'al ozone
distribution at a certain height and were qualitatively similar to laws ob-
tained by simplifying the methods described in the author's monograph
[8J. Results obtained in this paper differed quantitatively from those in
[8J with respect to ozone density dependence on parameter C, or a similar
parameter where the dependence was not as sharply expressed (Fig. 1).
On the basis of the obtained results consider the manner in which
stationary ozone density varied at different heights as a function of fluctua-
tions in solar radiation intensity and turbulent diffusion. The f6regoing
indicated that the presented comparison could be applied only to average
ozone density values, radiation intensity, and turbulence. The compari-
son was made by using mean annual data on ozone density at different lati-
tudes from [25 J, and the mean annual values of solar radiation incident
on the upper limit of the atmosphere at different latitudes in ref. ! [lJ.
, Consider~ for example, a change in ozone density at the 20-km
level with latitude as a function of latitudinal change in radiation and
,turbulence. For the sake of brevity consider latitudes from 10 to 70° .
The mean annual radiation values at these latitudes are 840 and 390 cal/cm2
per day, respectively [1]. The mea_~~~ual_?~one density values at the
20 -km level at these latitudes are 'lO-3cm03 using the
, 6.0 and 10.5 x km
Vigreoux scale [25 J . For simplicity, assume that the' mean annual value
of the stratospheric turbulence coefficient along the ve~tical at 103 was
only one -half of that at 70°. Assume that D and W3 were proportional to
the mean annual radiation values at ,the corresponding latitudes and find
153 -
-------
. that both paramete rs, q and C, changed with a change in latitude from 10°
700 840 x 2
to 390 x 1
= 4.3 times.
This variation in parameters corresponded
to ozone density changes at the 20 -km .level from 7.4 to 11.2 and from 7.5
1 -3
13 0 O' cm03
to. k
. m
in the first (Fig. I-a) and second (Fig. I-b) cases,
respectively. The theoretically obtained ozone density changes we re simi-
1ar to the above shown actual changes with latitude [25J. It can be con-
cluded from the above that the presence in the atmosphere of two factors, -
. .
photochemical reactions, associated with solar radiation, and turbulence
, were sufficient to ensure stationary vertical ozone distribution, correspond-
ing to the actual mean-annual distribution.
Now, consider the question of numerical values of changes in
ozone density at certain heights occurring due to: 1) oscillations in the
'ground-surface -reflected solar radiation and 2) atmospheric turbulence.
Change in radiation incident per unit volume in the atmosphere occurred,
among other reC\.sons, due to changes in ground surface albedo. Such radia-
tion change for the considered albedo cases 100/0 (water) and 700/0 (snow),
'amounted to 1. 7 + 1.1 = 1. 54.
Changes in the atmospheric turbulent state occurred within con-
side rably greater limits. Thus, considering regions with and without
jet streams, the turbulence coefficient in the upper troposphere and lower
stratosphere varied according to the data at'hand, from several units in
the first case to hundreds! of m3/sec in the second case [8J. Apparently,
the coefficient of turbulence in the entire column of the lower stratosphere
can have an average ten-fold variation. Accordingly, either q in equation
(13) or ,C in equation (27) decreased'during winter in the transition from a
continent with calm atmosphere to a sea with a turbulence region
q continent.. =' ~.-'54 'x 10 '= 15.4
q sea
(D)
I-a, the stationary
10-3 cm03'
density increased at the 25-km level from 8.8 to 11.7 km
. , 10-3 cm03
. and at the 20-km level from 7.4 to 12.8 km
Under such conditions and in accordance with Fig.
ozone
1.
e ., 1. 3 and 1. 7
times, respectively. In accordance with Fig. l-b the above ozone density
increase occurred to a still greater degree.
"
- 154 -
-------
Table 3
DEGREES PER DAY OF AIR WARMING UP AT DIFFERENT STRATOSPHERE
ALTITUDES AS THE RESULT OF DIRECT SOLAR RADIATION ABSORPTION
BY OZONE ACCORDING TO DATA OF DIFFERENT AUTHORS.
ALTITUDE IN KM
~
-
0,0 0,1 0,3 0,5
0,0 0,1 0,4 . 1,0
0,1 0,3 0,8 1,5-
.OR I NG (2~OCTOBEfi. 6o.:70oNO~.LA;:.
BIYUKKOVA & KASTROYA (2),
SPRING EQUINOX
GUSCH'N (8), .SOLAR ELEVATION 30°,
ACCOUNTING FOR DISPERSED RADIATION
It was previously mentioned that the rate of air heating in the
. stratosphere depended on ozone density present in the stratosphere.
T~ble 3 shows data available to the author on the rate of air heating due to
the direct solar radiation absorption by ozone at 15 to 35 km. Ohring's
data [25J included also less intensive heating - about 1/4 of the ozone heat-
ing - due to solar radiation absorption by water vapor in the stratosphere.
Increased ozone densities over the oceans cause the rate of stratospher-
ic heating to be higher over oceans than over continents. Assuming that
the rate of stratospheric heating over dry land at a latitude 500N corres-
ponded to data shown in the first line of Table 4 compute the heating of air
over- a sea, Assume also that all the conditions, except increase in ozone
density, remain unchanged. Assume, for the sake of simplicity, that the
heating rate of 20 to 25 km was directly proportional to ozone density,
and that an increase in the latter in the stratosphere occurred in the manner
indicated above. Under such conditions the rate of air heating will corres-
pond to values indicated in the second line of Table 4. The difference be-
tween values in the first and second lines in Table 4 yields a value of air
heating surplus rate - (third line of Table 4) caused by increased ozone
density over the ocean.
Table 4
----
--RATEOF ATMOSPHER I C-A~W~RMINGUPATDlFFi~E--;r-AL T.i~-~D E~
. LATION TO SUBJACENT SURFACE
ALTITUDE IN KM . . . . . . . 0 . . . . . . . .. 20
DAILY AIR TEMPERATURE RISE IN DEGREES OVER DRY
LAND CAUSED BY OZONE SOLAR RADIATION ABSORP-
TION (FALL 5)0 No. lAT.) . . . . . . . . . . . .0,1
01TTO OVER OCEAN. . . 0 . . . . . . . . . . . . 0 17
ENHANCED WARMING IN DEGREES PER 24 HRS EFFECT- '
ED BY INCREASED OZONE DENSITY
OVER OCEAN. . . . . . . . . . . . . . . . .
D . 0,07
ITTO, DEGREES PER 100 DAYS. . . . . . . . . .
7,0
IN RE-
25
0,3
0,4
0,1
10,0
- 155 -
-------
It can be seen from line 4 in Table 4 that the surplus heating rate
for 100 days was 7° at the 20 -km level and 10° at the 25 -km level. In each
case it amounted to an appreciable value. Moreover, anticyclonic circula-
tion, . which was developing in the stratosphere over the Pacific Ocean in
the fall, winter, and partly in the spring, contributed further "to a greater
surplus air heating at 20 to 30 km, so that during the circulation air mass-
. es propagated over regions, situated at lower latitudes and exposed to a
more intensive solar radiation. Note also that the limitations imposed .on
the solution of equations (10) and (11) in the form of the limiting conditions
(19) postulated a constant ozone density at the 30-km level. Such limita-
tions can be avoided assuming, for instance, that a constant ozone density
existed at the 35- or 40-km levels. Under such conditions the solution of
equations (10) and (11) will reflect a considerable air heati~g surplus also
at the 30-km level.
. Now; assume that the rate of stratosphere cooling due to radiation
losses by carbon dioxide and water vapor was approximately equal over
continents and oceans at identical latitudes. Under such conditions the
rate of the stratos phe re heating, calculated above, should be adequat'e
to insure the appearance in January of an appreciable thermal region over
the northern portion of the Pacific Ocean, or, to a lesser degree, Atlantic
Ocean. Evidently, the thermal region in the stratosphere will.be less
distinctly expressed at lower latitudes, 0 -- 300N since the continents in
that area remained clear of the snow. On the other hand, at higher lati-
tudes, 70 -90° N, unpenetrable by the solar radiation during winter,
formation of a t;h.ermal region, in a manner indicated above, will be im- .
possible.
Examine actual data on thermal stratospheric regions and determine
to what extent such regions conformed to the proposed scheme. According
to Dubentsov's average data [9J for January, a thermal stratospheric re-
gion over the northern portion of the Pacific Ocean was most distinct at.
the 50-, 30-, and 20-mb levels, which roughly corresponded to 20, ~4,
and 26 km, and, less clearly expressed, at the 200-, 100-, and lO-mb:
levels. The center of this region corresponded approxirI:lately to a degree
with coordinates of 50° Nand 170° W. The closed -450 isotherm at the.
. 20-mb level extended approximately between 40° and 60° N. The mean
monthly January temperature at the 50 -20 -mb levels over the northe rn
portion of the Pacific Ocean was 10-15° higher than in other areas.
Table 5, organized on the basis of Kh. P. Pogosyan's data [13J,
shows the mean stratospheric air temperature during January and July
over dry lands from SOoN and 10-30° E,' and over the ocean from 1700E
to 1700 W. The Table also presents differences in air tempe rature at
different heights and longitudes.
- 156 -
-------
Table 5
I .
" AYERAGEj" STRATiis'p HERIC-AIR TEHPERAT UiE-INJ'"iiUI.RY AND J-ULY AT "3J0 NORTHERN
LATITUDE AT DIFFERENT LONGITUDES AND ALTITUDES ACCORDING TO DATA OF KH.P.
, POGOSYAN (13) !
m, ". LON" ,,, E ~ T .- - -, DEG-R-EES LONG i nio'E
AlT'-tuOE EI1P. AltITUDE TEMP.
Me 10-30 E.!11i:I-'E;lO- DIFFER. Me (0-30 E.II70 E.LO~ ' DIFFER.
(KM) La. 170 WIILO - (KI1) Lo. 'nOW.LO.
.------..--- -- ..' .
JANU"RY JULY
200 (12) -57 -52 5 200 (12) -52 -48 4
100 (16) -59 -51 ' 8 100 (16) -52 -52 0
50 (20) -58 -47 11 50 (20) -49 -49 0
30 (24) -57 -43 14 30 (24) -48 '--48 0
,15 (28) -57 -43 14 15 (28) -44 -43 1
Table 5 shows that greatest temperature differences were observed
during winter at altitudes of 24 to 28 km. During summer, these differe-
I 1 .
ences \Vere a most n6nexlstent.
A thermal region was observed regularly during winter in the strato-
spHere over the northern portion of the Pacific Ocean. This region closely
coihcided with the stratosphere region of increased pressure [9J, although
the I latter was particularly well defined on the mean monthly charts for
much higher levels (10mb). This indicated that the increased pressure'
region was created by the isobaric surface rise in the warm stratosphe ric
air. At still lower levels -50 and 30 -mb- the increased pressure region
over the Pacific Ocean was displaced to the south by a thermal region.
This was due to the fact that the effect of the underlying high-pressure
region situated to the south was noteworthy.
Some tempt:rature increase was observed in the stratosphere over
the Pacific Ocean in the summer, although it was considerably less
pronounced than during wintertime. This can be explained by the fact
that the principal thermal region in the summer stratosphere was found
at high latitudes, and the diffe rence in the albedo of continents and oceans
was considerably less during summertime. A thermal region was also ob~
served over the Atlantic Ocean in the winter stratosphere, although, as
in dicated by the average charts, with less clarity than in the same region
over the Pacific Ocean. . The daily baric high-altitude topography charts
clearly indicated a thermal region in the Atlantic Ocean area. The fact
,that this region was relatively indistinguishable on the average chart can be
explained by the considerable mobility of this region during that month \
and also by the smaller size of the Atlantic Ocean. .
The above facts concerning high altitude temperatui-e are not in
variance with the mechanism of stratospheric heating advanced in the
present report. It is believed that the available data on ozone constant
- 157
-------
in the Pacific Ocean area were insufficient and were principally from
c oas tal stations.
Kulkarni, Andreji, and Ramanathan noted [20J that the Japanese
station Tateno, at 36°N, presented consistently higher total ozone values
than did the Indian Station S rinagar situated at the approximately identical,
34° N, latitude. The differences were particularly great during winters
and springs. Considerably more ozone was observed at Chinese sta-
tionSi-ka-wei at 31N, than atSrinagar, despite the fact that the former
was situated further south thanS rinagar. Data collected over a number
of years and averaged for February indicated O. 345 cm 03 for stations
Tateno and Si-ka":wei, and 0.277 cm 03 for Srinagar. Muramatsu [24J
noted that a considerably higher total ozone than was recorded at Tateno
w:as observed at Japanese station Sapporo at 43°N, i.e., 7° further north,
the observed increase exceeded the. mean latitudinal ozone variation. Ac-
cording to mean-monthly data, the maximal total ozone occurred at Sapporo
earlier than at other Japanese S~ther stations.
. Monograph [8J presents to tal ozone observation data at Soviet
. stations Vladivostok and Yakutsk for 1961-1962. The two-stations are
situated close to- the Pacific Ocean. Noticeable increase in the ozone
value was observed there during winter and spring, and in compari- .
son with continental stations at identical latitudes. Thus, in Vladivostok,
the mean monthly ozone value was 0.456 cm in January, 0.:482 cm in
February and 0.548 cm in March, while in Yakutsk, in February and March
the corresponding values were 0.576 cm and 0.548 cm. Meanwhile,
. according to Godson's diagram [19J, the mean ozone value at the Vladivostok
latitude was 0.310 cm in January, and 0.340 cm in February, and at the.
Yakutsk latitude - o. 380-cm in February and O. 41Ocm.in March.
. The ozone data compiled directly by the research ships in the
Pacific Ocean area are of considerable interest.
I
'/
- 158 -
~
I '
-------
Fig. 2
Fi g . 3
CM
0.5
3V
UJ
Z
o
N
o 0,3
....
<
t-
o
....,
0.2
0,1
"O'C/U. 30
20
20
30
40' IO.IJJ.
10
o
10
-~ --+----_._- ----.-
. --- -- ----_. - -.-.
TOTAL ATMOSPHERIC ~ZONE IN RELATION TO LATITUDE ACCORDING
TO "SHOKILtSKII DETERMINATlorlS
I - PACIFIC OCEAN OBSERVATIONS, 25 DECEMBER 1951 - 22 JANUARY 1962 CLOSE
TO MERIDIAN 1800 EAST. LON.; 2 - DITTO 5-22 MAY 1962 CLOSE TO MERIDIAN
1700 EAST LOll.
\
I .
0,(10 "h'J
500
400
-60
-70
IX
XI
n
x
XII
I
/1/
IV V
-~-- _._--_.~---- .'. - -- . +---- .-- - ---- - ---
-- - -- --.--
AVERAGE DECADE TEMPERATURE VALUES AT 100 MB LEVEL
AND AVERAGE OZONE CONTENT OUR I NG WI NTERS OF L~
. 58 OVER SCANDINAVIAN STATIONS (19)
I-TROMSE; ~UPSALA
- 159 -
-------
.,
i',
-- Fig. 4
MO A
19,5 ----1
39 2
~ -. ----3
76 ~':. B ,.,5 .
~" ... . -. . '" It
156 '\~ f' 10
"
3/2 )1. .I, 2S
. / .. I \
UJ 625 ,/"-: . \ 50
C> /r ....., \. I
=
...
... /250 t ..- \ \ TOO
'"' / I ,.'
.0: I ',' i..\
2500 F
I ,\ 250
5000 1:1
I ~\
MKM~- J~_. lq_~___1f.q
PARTIAL PRESSUE
-00. - (;0 - ~O -30.
TEMPERATURE
-Exr;EMES--OF.OZONE -YAl-UES- AN.D-OF--YEATICAL-.iE~-;~~-~T~R E-D I~ -
T"18UTION DURING RAPID HEAT FLOW IN STRATOSPHERE OVER AROSE
IN JANUARY AND FEBRUARY OF 19513
-------
the onset of a sharp increase in total ozone. Similq.r results were ob-
tained by Godson for other European and Canadian stations.
- - I I .
Dutsch [18J 'and London !i2Jdetermined the trajectory of the thermal
stratosphere region center and the trajecto-ry of the center of maximum
pbsitive deviations in total ozone during January-February 1958 over
E!urope, the Atlantic Ocean, and America. Both investigators observed
ajnoteworthy proximity of the two trajectories. Dutsch [18J also in-
vestigated a change in the vertical ozone distribution over Arose and
Switzerland, in January-February 1958 during the passage of a thermal
region over that station. The ve rtical ozone distribution measurement
was made by the inversion method. Figure 4 shows changes in a) the
- v-erti~al ozonedistributlon and b) temperature changes over Arose
during three days in January and February, 1958. Figure 4 indicates
- that during the rapid temperature increase on January 23, partial pressure
of the stratospheric ozone also increased. Facts adduced with respect to
ozone distribution over the continents and oceans were not at variance with
the mechanism of stratospheric heating discussed above.
- It should be pointed out that the term "rapid stratosphere warming-
up" is somewhat of a misnomer. Actually, instead of a rapid strato-
spheric ozone heating there occurred at the considered point a rapid re-
placement of relatively cold air by warmer air. On the whole, this pro-
cess probably occurred in the following manner. Simultaneous relative
increases in stratospheric ozone content and in temperature occurred
during autumn and winter at the intermediate latitudes over oceans and
certain coastal regions; for reasons still unknown noticeable intensifica-
tion in the Pacific anticyclone and in its migration to the Arctic were some-
times observed during winter. Moreover, the polar whirlwind was forced
back frequently into the northern portion of the Eurasian Continent. In
this case, a sharp increase, up to 20-40°, in the stratospheric tempera-
ture was observed at the Arctic stations during polar night. A relative-
ly ozone-lean and cold stratospheric air in the polar whirlwind system was
replaced in the polar region by a warm, ozone -rich air from the Pacific
anticyclone. Such gigantic stratospheric air mass transfer occurred
in January of 1958.
The warm stratospheric air from the Atlantic Ocean, 1. e.,
stratospheric anticyclone in many instances migrated over Europe and
Northern Africa, driving back the cooler air. - When this occurred, a
sharp increase in the stratospheric temperature was observed over
certain regions of Europe. Appa'rently,such warriling-up was observed
by Scherhag over Berlin in 1952 and by Dutsch ovelr Arose in 1958.
In this cd~nection it should be pointed out that a rapid increase in the
stratospheric temperature over the Antarctic during winter has not been
I
- 161 -
-------
observed [3]. Considerable warm-ups were observed there during spring,
in October, rather than during winter. This author is of the opinion that
this fact favorably supported the above -proposed rapid stratosphe ric heat-
ingmechariism. The Southern Hemisphere is basically, an oceanic hemi-
sphere. The effect of continents on ozone layers appeared considerably
weaker in the Southern Hemisphere than in the Northern Hemisphere. The
Southern Hemisphere afforded no conditions for the formation of individual
warm stratospheric anticyclones similar to the Pacific anticyclone. Accord..
ing to the data at hand, the mean annual total ozone maximum in the Southern
Hemisphere was observed at a latitude of approximately 50° and not at 90° ,
as was the case in the Northern Hemisphere [8].
The observed zonal circulation in the stratosphere was more regu-
.lar in the Southern Hemisphere than in the Northern Hemisphere. There-
fore, intrusions of warm ozone-rich stratospheric air masses into the polar.
region were unlikely in the Southern Hemisphere during winter, indicating
that the Antarctic received ozone from lower latitudes to a lesser degree;
than did the Arctic, thereby affecting the mean ozone distributionmeridion-
ally; less ozone was found over the South Pole than over the North Pole. It
should also be noted that in the above -compiled calculations no allowance
. wa.s made for cloudiness and for ice cover. Generally, the foregoing
indicated that the radiative and turbulence proce sses and, in particular,
atmospheric ozone played an important role in the complex phenomenon
of stratospheric heating. However, these processes fail to explain the.
total complexity of the discussed phenomenon; problems of its dynamic~
and the possibility of solar activity effect on tp,ese proc~sses still remain
unexplaine d.
Bibliography
I. Be p ,1 S! H)J. T. f. Pacnpe;1e.1CHlle CO.1HC4HOii pa}wall.lIlI Ha KOHTHIICIlTax.
flupOMeTeOH3}1.aT, .rI., 1961.
2. Blip JO K 0 B a .r1. A., K aCT p 0 B B. f. 0 CYT04HO~' XO.1e Te~lncpaTypbl
B CTpaToccpcpe. MeTcopo.l0rJIS! II rH.llpOJlOrHS!, N~ 8, 1961. .'
3. Be K C ,1 e p r. KOJle6aHIIS! Te~lnepaTypbl B anlOccpcpc Ha.ll AHTapKTIIKoH.
. MeTeopo'lofHS! II rll,'lpo.l0rJIS!, N~ 3, 1959.
4. rail r e po B C. C. BOnpOCbJ a3po.lorJ\4ccKoro CTpOCIIIIS!, lI.lIpKy.1S!lI.llII
II K.1HMaTa' CB060}l.HOii anlocq,epbl Ll.cHTpa.lbHoii ApKTlIKII II AHTapKTlIKH.
YI3.n.-BO AH CCCP, M., 1962.
5. r Y III II H r. n. 03011 II HCKOTOpblC OC06CHHOCTII. anlOcq,cpHoii U.llp~y.1S!UHH.
Tpy.1J.bI frO, Bbln. 134, ]962. .' .
6. f Y III II H r. n. OC06CHHOCTII ropll30HTa.lbHoro pacnpe.n.c.leHHS! 030Ha no
MaTCpllaJiaM Mrr H MrC. Tpy.1J.bl BHMC, T. V. rH)J,po~leTeoH3.n.aT, J1.,
1962. .' J1
7. f Y III II H f. n. I1cCJlCiJ.OBaHlle anlOccpcpHoro 030Ha. fH;1POMCTcoIl3JJ.aT, .,
1963.' '.
8. r y Ill.!! H f. n. 030H H a3pocHHonT1I4ccKIIC YC.l0BII!I B anlOccpcpe. fIlAPO-
MeTeOI\3)J.aT, J1., 1964.
. 9. lI. Y 6 e H U 0 B B. P. B03.1YWHbIC TC'ICHIIS! B cTpaToccpcpe. MCTeopo.lorH!I
H m.apo,10fHS!, .N'~ II, 1959. .
10. 3 y 6 !I H f. lI.. a MC:iKWHpOTHOM 06MCHe TCnJlbJX II XO.1JO)J.HblX Macc B03-
. .llyxa B CTpaToccpepe 31BlOii. MeTeopo.1JOfH!I II rH.n.po.l0rHS!, N~ I, 1959.
II. K 0 H .a paT b e B K. 51. JIY'IIICTa!l 3HCpfH!I Co.1Hua. fll)J.pOMeTCOH3)J.aT, J1.,
. 1954.
- 162 .-
-------
I
,
]2. naB .~:o B C K aSIA. A. J Jorell.~eHIHI B crparoc. .llHHaMHlleCKa!! H
-------
, .'
I .
SOME RESULTS OF OZONE OBSERVATIONS MADE 15 FEBRUARY 1961
DURING A TOTAL SOLAR ECLIPSE
A. Kh. Khrgian and G. I. Kuznetsov
" Ozone changes occurring during a solar eclipse throw light on the
actual rate of photoche mical reactions taking place in the ozone la ye rand
on their stability. Total ozone measurements were conducted on board a
search-plane LI-2 over the Azov Sea area near a point defined by coor-
dinates CP = 46.2° N and A = 35. "f W. The point was located along the path
of total solar eclipse. The search plane executed 20 -min forward and.
reverse flights along the total solar eclipse belt on a base of approximately
70 km and at a 3000 m height. Observations were made through open
illuminators on the port and starboard sides of the plane. A filter-equipped
!J.niversal GGQ (Government Geophysical Observatory) ozonomete r was used.
,The filter band centers we re at: 1- ~5l20A and II --3 700A. The instrument
was calibrated at the GGO by means of a Dobson spectrophotometer. Ozone
content was determined from. nomograms computed for the corresponding
, flight altitudes.
I
The eclipse in the observation area was first contacted at 09:56
'Moscow time; the second contact was at 11:09; the third --11:11. The
eclipse ended (the fourth contact) at 12:29. Ozone observations began 43.
min. before the first contact and ended 31 min. after the fourth contact.
During that time, 68 total ozone measurements were made above the
flight level, of which 11 were made before, 44 during, and 13 after the
eclipse. General weather conditions during the observations were as
follows: Dense clouds spread below the search plane; a 10-point St .
layer at the 0.1-0, 3-km level; a 10-point Sc translucent layer ~xtended
. above the 0.9-1. 2-km level, under a strong inversion; between 1.18 and
1. 34 km the temperature rose from- 7.2° to 2.5° and the humidity fell
from 92 to 50%. The upper cloud boundary was flat. The atmosphere
above the clouds was clear, the sky around the sun was clear during the
entire period of measurements, except near the horizon in the north,
where traces of Ci clouds appeared toward the end of the observation
period. The sky was light-blue at the onset of the eclipse and darkened
with the appearance of the stars as the eclipse was approaching its mid~le
phase. Temperature and humidity outside the plane at the 3-km level
fell to -9° and 12%, respectively. The air above inversions in cases of
low cloudiness was normally pure. '
For-purposes of control, training, and comparison with the synop-
tic conditions, ozone content observations were made from the plane also,
on February 14, 16 and 18 at flying altitutdes of 5.2, 3.2, 4.2, and 2.0 km,
res pectively.
164 -
-------
Table 1
-~. ---- - -- -. -
DATES x.10' CM n hyp T,,,,,
14 II 346 4 9,7 -51
15 II 319 68 10,3 -54
16 II . 28Q 36 11,1 -59
17 II 10,3 -52
18 II 364 13 10,3 -57
Table and Fig. 1 show the following: total ozone (x), number of
observations on a given. day (n), height of tropopause (h ), and tempera-
ture of the 100 -mb level (Tl00) on observation days. It s1fould be noted that
all'-ozone content data were reduced to the 3-km level. Moreover, it was as-
sumed that the lower l2-km layer, contained 8.5% of the total ozone [1], and
that its density in that layer was identical. In view of this, ozone density
in the troposphere was 0.002 cm/km on the day of the eclipse.
. During the period under consideration total ozone decreased from
x = 0.346 cm on February 14 to 0.280 cmon February 16; it rose to 0.364
cm on February 18. Table 1 and Fig. 1 manifest a regular relationship
during that period between changes in x, on the one hand, and changes in
hand Tl00' on the other hand. Actually, high ozone content normally
c~;'responded to a low tropopause and a warm stratosphere, and vice versa.
The first impression was that decrease in ozone content on 15th and, in
particular, 16th February 1961 had no direct connection with the eclipse.
However, a deep high-altitude trol:1gh existed above the Crimea from 13
to 15 February which could be clearly identified on the baric topography
charts of the 100-mb surface. Strong northern winds, up to 60-80 km/hr,
prevailed over the European USSR at the 100 -mb level during that period.
Such synoptic conditions lead normally to an increase in the ozone content
at the southern stations in the moderate belt toward the end of winter.
This phenomenon was not observed in the present case. Thus, the reason
for a considerable drop in the ozone content on the day of the eclipse and
on the following day remained unexplained.
Now, proceed directly with the results of x observations during
the eclipse. Mean values of ozone content x during different periods of
eclipse are shown in Table 2 with the number of observations indicated
in parentheses. The same table also shows the computed values of the
mean square deviation (J for the corresponding periods.
Data in Table 2 show that ozone variability before the eclipse was
a = 0.010 cm, increasing up to 0.020 cm and 0.015 cm respectively during
and after the eclipse. .
- 165 -
-------
Fig. 1
I1.mp KM
:r CM.tO J
11
T,oo
10
300
-50
200
-60
1~
JS_---.!~---_-11.. -- .18
~ FEBRUARY 1961
,
. ------ _--_--:;_0____..'------- ------.--
CHANGES IN OZONE CONTENT \X}, TROPOPAUSE ALTITUDE
(Hn), AND TEMPERATURE AT 100 MB LEVEL('rIOO) DUR-
ING 14-15 FEBRUARY 1961
Table 2
- ~- -..-.---.
x.lO' CM
0.\0-'
BEFORE THE ECLIPSE. .
D UR I NG THE ECL IPSE. .
AFTER THE ECLIPSE
338 (11)
32.') (44)
283 (13)
0,010
0,020
0,015
. Curve "1 in Figure 2 illustrates th~ variation of x during the eclipse
on 15 February 1961 in the Azov Sea region. The curve significantly point!;
to some interesting features in ozone changes during the eclipse. Moreover,
only changes with amplitude exceeding the mean square deviation during \.
a given eclipse period will be recorded.
1: A general drop occurred in ozone content from x = 0.338 cm
recorded at the onset of the eclipse to x = 0.283 cm recorded at the end;
the net change amounted toO. 555 cm, or 17%
2. Some decrease in x was observed between the first and second
contacts, and an increase in x toward the beginning of the fourth contact.
3. The above decrease during phases O. 4~1. 0 was followed by an
increase as the phase progressed up to and after the full phase.
4. A clearly discernable "peak" was observed approximately dur-
ing the c.entral phase with a, clear . shift to the right, 1. e., lagging with
respect to the eclipse. The impression was that an increase in x occurred
immediately after the end of the third contact. This peak exceeded the
- 166 -
-------
mean value of x (0.325 cm) by 0.065 cm (200/0) and the minimum value of:<
(0.295 cm) by 0.095 cm (30%) during the eclipse.
I
'I
Fi g. 2
1"--0'
x CM" 103
100
600
500
xcwlO 3
400
300
300
t-
9
10
11
12
13 'lac
-------- COURSE-OF OlONE CO/frE-NT CHANGES-OURING.TH-E- ECLlPSEc--
I-OBSERVATIONS OVER SEA OF -A£o.v;-iA: Oi-T1:0 IN- RELATION TO SOi:AR EDGE -
ECliPSE; 2 - ROSTOV OBSERVATIONS (3). VERTICAL LINES MARK ECLIPSE CO~
TACTS; 3 - TIME OF DAY
A sharply increased ozone variability was also observed before the fourth
contact, followed by a considerable immediate drop in ozone. A similar,
although less pronounced, drop was also observed before the first con-
tact.
Considerable ozone accumulation may appear around the central
phase noted in connection with a darkening effect extending toward the edgE;
of the sun's disk. This darkening, 1. e., radiation attenuation, is more ~
pronounced at shorter wavelengths. As a result, the recorded ratio of
intensities I(A = 3120)I(A = 3700) in the case of greater eclipse phases
became reduced befor-e the solar radiation penetration into the ozone layer. -
This was pointed out by Svenson in 1957 [2J who also performed correspond-
ing calculations of the apparent increase in the ozone content during an
eclipse observed with the aid of a spectrometer. R. S. Steblova [3J
computed the effective ozone increase during the eclipse on 15 February
1961 using an ISP-22 spectrograph. She found that for the 3112 and 3320A
wavelengths the increase did not exceed 0.038 cm for the central phase,
-becoming smaller on both sides of that phase, and almost negligible
at 2/3 of the total eclipse. The instrument used in the present study
was not spectrally broad, but had adequate broadband filters, rendering
the use of Svenson's method difficult. In view of this, the present author
evaluated only the apparent ozone increase for the central phase using
- 167 -
-------
Wal'dmayer's data [4J on the speCtral dependence of reddening toward the
sun's disk.
It was found that in the case of the present author's instrument the
apparent ozone increase during the central phase did not exceed 0.060 cm.
Hence, assuming the above value and the nature of changes in the apparent
ozone inc rease as a function of the eclipse stage calculated in [3 J, the
curve (Curve I-a, Fig. 2) of ozone content was plotted at the time of the
eclipse, taking into account the "darkening effect". Allowance for the
.latter did not interfere with the es s~ntial properties of changes in x during
the eclipse,. except for the fact that a partial decrease in x afte r the first
contact may continue up to full eclipse. Tot~.1 ozone content during the
.eclipse of 15 February 1961 was also measured over the Rostov-on-Don
areal by R. S. Steblova [3 J. The measurements were made onboard a
plane using an ISP-22 spectrograph. Results of measurements made over
Rostov, taken from [3J, are shown for comparison in curve 2 of Fig. 2.
When the two curves are compared it should be borne in mind that curve 2
. was shifted 8 minutes backward in order to achieve coincidence of the
eclipse contacts.
If one were .to depart from criterion of absolute total ozone values
and its variability during eclipse, according to [3J, then the mean value
and variability during eclipse would be respectively x'= 500 cm and CJ =
0.120 cm; in the opinion of this author these values appear high2). the curve
comparison would reveal a certain likeness between them. . Data secured in
this investigation also indicated that during the measurements over Rostov,
ozone content decreased with time from 0.500 cm during eclipse to 0.420
cm after the eclipse. Both curves exhibit maxima ne~r full eclipse and
are identically shifted to the right from if. 'The nature of intensified os-
cillations in x was also similar during the period before the fourth con-
tact.
It should be noted that She A. Bezverkhnii [5J noted a sharp increase,
in x immediately after an eclipse. The same effect was described earlier
by Kawabata [6], who also noted a decrease in the ozone content after the
first contact. A general drop in x during eclipse was also pointed out
by Ollson [7J. The foregoing confirms the reality of the above shown fea-
., tur'es 'of total ozone changes as sodated with energy redistribution of the.
.L Ozone measurements during the eclipse of 19 February 1961 were conducted
also in Karaganda, Crimea. Results of these were published in G. P. Gush-
chin's monograph "lssledovanie atmosfernogo ozona ", Gidrometeoizdat,
1963. [Ed. ] .
2The instrument which R.S.Steblova used in the ozo~e measurements was not
calibrated against Dobson's standard spectrophotometer available at'the GGO.
I
[Ed. ] - 168 -. I
-------
"1
solar spectrum during an eclipse. Interpretation of tfle observed data is
. extremely difficult because the actual spectral energy; redistribution in the
. ultraviolet region during the eclipse outside the atmo~ phere remained un-
I .
. known. Based on the darkening effect one could expect a decrease in total
ozone symmetrical with respect to the central phase of eclipse. It should
be elicited by a sharper attenuation of the oxygen-disintegrating radiation
. component as compared with the ozone -disintegrating component. The
. magnitude of the above decrease should be a function of the build-up time
of the photo-chemical equilibrium in the upper ozone layer portion. In
the light of existing cr values, the above decrease is small. However, it
is conceivable that cr values may have to be reexamined in the light of
current ozone layer studies conducted under conditions of varying solar
radiation [3 J.
A similar decrease in total ozone was observed in the course of
the present investigation, at least during the first half of the eclipse.
. Tl:1e sharp increase in ozone immediately after the central phase 'was more
difficult to explain. It can only be noted that the observed changes in ozone
. were compatible with those theoretically computed by Dutsch [9J for a
period during a sunset, i. e., under conditions of a sharp change in the
radiation regime of a high atmosphere.
Bibliography
I. Key B. N., B r ewe r. Do b son G. M. B. Some measurements of vertical
distribution of ozone. Sci. Proc. Int. Ass. met., Rome. 1954, p. 189.
2. S v ens 0 n B. Observations of amount of ozone during the solar eclipse.
Arkiv. geophys., B-2, No 28, p. 573, 1957.
3. C T C 6 n 0 B a P. C. Ha6mo1\eHHJI 3a aTMoc!jJepJlblM OJOJlOM 80 8peMJI con.
He'lHOrO 3aTMeHllll 15 II 1961 r. J\hmeTll3M II a3pOJlOMIIJI, T. II, ,t{2 I, 1962.
4. Ban b 1\ M a He p M. PeJynbTaTbI H np06neMbI JlccneJl.OOallHJI ConHua, YlJ1,
M., 1950.
5. Be 38 e p x H II if Ill. A., 0 ill e pOD II 'I A. JI., Po IJ. II 0 II 0 8 C. <1>. 3neKTpo-
!jJoToMeTpll'leCKHe IICCnelJ.OOaHIIJI aTMoc!jJepbl 80 OpeMJI 3aTMeHJlJI ConHlla
25 II 1952 r. H 30 VI 1954 r. .l{AH CCCP, T. 106, N2 4, CTp. 651, 1956.
6. K a w a bat a 1. Spectrographic observation of the atmospheric ozone at the
total solar eclipse in Juny 19, 1936. Japanes Journ. Astronomy and Geoph.,
14, No 1-3, Tokyo, 1936-37. .
7. J e r low N., 0 II son H., S c h ii p peN. Measurements of solar radiation ot
Lovanger in Sweden dur(ng the tolal eclipse 1945. Tellus, No 6, No I,
1954. . .
8. Fo urn i e r d'A I beE. Some observations on nocturnal ozone. Sci. Proc. Int.
Ass. met., Rome, 1954. " .
9. D ii t s c h H. Sci. Proc. I nt. Ass. met., Rome, 1954.
- 169 -
-------
---=----,
EFFECT OF CIRCULATION CONDITIONS ON THE DISTRIBUTION
. OF TOTAL O~ONE IN THE ARCTIC
I.M. Dolgin and G. U. Karimova
. . This article deals with general observations made on total ozone
in the Arctic during 1958 - 1962j an attempt was made to establish a connec-
tion between total ozone fluctuations and changes in atmospheric oircu1a-
tion conditions. Results of total ozone measurements made during the
daylight period of the year (March - September) at.the Heiss Is. station
(80° 3PN and 58° 03 tE) and the drifting stations "North Pole" in the area
bounded by 78-82° Nand 140 1.190o E are shown in Table 1, where paren-
thetic data indicate the number of observation days. The number of days
suitable for observations was 117 on Heissls. and 66 on the drifting sta-
tions due to a heavy recurrence of clouds in the lower cloud leveL' Due'
to'the insufficient number. of simultaneous observations made along a I
. . I
. clouded sky zenith and the sun, it was not possible, at the time, to I .
establish a rational correlation between these observations.
Table I
TOTAL OZONE OBSERVATION DATA FROM ARCTIC STATION RECORDED DURING SOLAR
TIME PERIOD
HEISS Is.
. . . . .
\ III \ IV \ V
I 0,551' 0,469 0,441 0,298 0;290 0,244 0,255
(5) (17) (21) (21) (27) (20) (6)
\ VI EEE'.
STATION
DRIFTINGls~--r
NORTHERN POt.E
0,542 0,494 0,325 0,309 0,210
(9) (10)(25) (10) (12)
I
--~~-~.-
. . I
Comparison of data for the summer periods 1958-1959 and 196q-
1962 showed that total ozone values during 1960-1962 were somewhat lower
. I
than those during the preceding years. A sharp decrease in the amou~t of
ozone in the Arctic from spring to summer was observed during all the
years. A comparison.oitota1 ozone values at Heiss Is. and the "North Pole"
during identical summer days of 1962 showed that they were sufficiently'
close. .
Aerial observations of total ozone were made from an IL-l4 plane
[6] simultaneously with ground observations at Heiss Is. and the "North
Pole" stations during August -September 1960. The flight~s ~ere-p~~'- .
grammed for a height of 2000 -3000 m, along' 1250 E in a sector from 79
- 170 -
-------
to 83°N.
Results of observations are shown in Fig. 1
Fig. 1
CM
q30
w
z
o
~ 025
,
....
<
...
a
...
25 I//If
.......
........
........
,
........
"'"'-
..........--.-.-.----.......
0.20
79
.80
81
SIX
/
/
-
83-
82
.oioNE CHANGES -WITHLATITUDE--
The August 25 and 27 flights were made along a dike between two baric
systems; total ozone values observed on the two flights were sufficiently
close. The September 5 flight was made in a diffused baric field; ozone
content in this case was lower than during the preceding two flights, probably
due to seasonal changes in total ozone. A continuous cloudiness prevailed
over Dixon Is. on the scheduled flight days, making it impossible to com-
pare results of the aerial and ground observations. In contrast to the
sporadic character of observations made at the drifting stations, and to
the small volume of work at Heiss Is., where observations were made
. interruptedly during 1960-1962, observations on Dixon Is. were made
during a five-year period on a systematic uninterrupted basis. Total ozone
observations made in the Arctic as of June 1960 had been conducted with
the aid of the Universal ozonometer GGO. Figure 2 shows a normalized
curve of the annual variations in the total ozone content ove r Dixon Is. wr-J.ch
was plotted on the basis of data collected during 1958-1962. The solid and'
dash portions of the curve indicate data from direct solar and lunar light
obse rvations, respectively.
Observations of recent years confirmed the presence of well-de-
fined annual ozone changes over Dixon Is. noticed during the first observa,-
tion years. Therefore, changes in the ozone content curve for many year~
are analogous to changes obtained during the first observation years. .
. According to re~ults of 5-year observations maximum ozone content
occurr-ed generally during the summer months (March-May) and amounted
to 0.390 -0.410 em, and the minimum was observed in September and amount-
ed to 0.240 em. Thus, the amplitude of annual total ozone fluctuations ove:r
Dixon Is. equalled 0.170 em. Similar amplitudes were observed also over'
- 171 -
-------
1--
foreign Arctic stations at Tromso and Alert [1. 9J.
Fig. 2
CN
0.'-0
0,35
"
/' .
"
"
I
I
I
/.
.---- ,/
....
z
o
~ 0,30
...
<
...
::. o. 25
0,20
II
III
IV
v
VI
VII
VIII IX
x
XI
XII
ANNUAL COURSE OF TOTAL OZONE CHANGES OVER DIXON Is. IN
. 1958-1962
U nUke in the A rdie. the amplitude of annual ozone n\etuations was e ori -
siderably smaller at intermediate latitudes. Thus. according to the mean
I
multi-year data the annual amplitude at Voeikovd (600 N) was 0.125 crri..
. The absolute extrema of the mean daily ozone value,s over Dixon Is. we!re
0.511 cm and 0.170 cm in April 1959 and September 1961. respectively.
_'\The~nnua,l variations curve indicated that the sharpest drop in total ozone
-occurred during May to June. and that the ,amount of ozone decreased by
300/0 'during June. Thus. t.otal ozone decreased during a 6 months period
from spring to autumn by 600/0. with one half of the decrease (300/0)oc -i
curring during May to June. '
O.M. Kuzne,tsov. a student at the Leningrad Hydrometerological
Institute. broadly investigated ozone content in the Arctic and Ant~rctic
and compared ozone content data collected by Soviet and foreign Arctic sta-
tions. He found that annual ozone changes over Dixon Is. and over the
Alert and Resolute stations were similar. which complemented eV,idence
obtained by the present author 'on ozone content in the Arctic. and con-
firmed the accuracy of the results. A comparison of annual ozone con:"
tent changes. observed at stations located at different latitudes showed
[1. 7J that the ozone maximum shifted to the intermediate latitudes dur-;
ing the s pring to autumn pe riod. Thus. when the s pring ozone maximuql
prevailed over the Arctic. then. beginning with June. the ozone content:
at intermediate latitudes was greater than over the Arctic.
It has been known that ozone content was affected by circulation
conditions, which normally changed from year to year, eliciting changes
in total ozone. A comparison of total ozone values over individual years
- 172 -
-------
'hiid clearly indicated thiit they fluct.uated considerably, especially in spring,
and were least pronounced during autumn. Maximal total ozone deviations
from the multiyear mean were observed during May, 1958 (+0.050 cm) and
May, 1961 (-0.072 cm). Analysis of the baric topography charts of the up-
per troposphere and lower stratosphere for May in those years showed the
following: In the first case, high-altitude closed baric systems existed
over Dixon Is., on southern periphery of which a regional transfer of
aerial masses occurred at intermediate latitudes. In the second case,
. an intense' interlatitudina1 transfer occurred over Dixon Is. region asso-
ciated with the existence of a deep cyclone over the European portion of
the USSR. Furthermore, aerial mass migration from south to north was
also taking place in the area under consideration. Therefore, annual
changes in the circulation conditions during the same month changed the
total ozone by 0.122 cm.
In spite of changes in ozone content and in its annual fluctuations,
daily ozone fluctuations up to 25% of the mean diurnal ozone content during
spring, are extremely important. It has been known [2J that ozone oscilla-
tions connected with weathe r conditions occurred as1 the result of continuous
steady spatial aerial mass redistribution in', the atmosphere with different
vertical and horizontal ozone contents. A sharp dec rease in ozone which
occurred over Dixon Is. during March 18-20, 1961 and which lowered
ozone content by 0.115 cm is an examplary illustration of that.
Analysis of the A T300 charts for the above days showed, that on
18 March Dixon Is. was in the sphere of the eastern edge of an extensive
deep-seated cyclone (Fig. 3-a), which encompassed almost the entire
European USSR area and extended into the 100-mb level. .Moreov\~r, the
Dixon Is. region was being invaded by aerial masses from the Arctic basin,
rich (during spring) in ozone the content of which was 0.430 cm on that day.
On 20 March (Fig. 3-b), Dixon Is. was close to the axis of a high-aLtitude
crest, seen on the AT300 chart, at the periphery of which an intense c.l.erial
mass migration proceeded from lower latitudes, having a lower ozone con-
tent. Moreover, the amount of ozone over Dixon Is. sharply decreased' to
0.315 em. It is significant that on the same day total ozone over Voeiko'vo,
located in the rear portion of the huge cyclone where the transfer of aerL:l.1
masses from the Arctic was taking place, was rather high, 0.540 em, 1. e.,
it exceeded the ozone content over Dixon Is. by 0.225 em. The refore, re-'
moval of aerial masses from the north during spring led to an ozone con-
tent increase over a given point, and, conversely, removal from the south
decreased the ozone content. This convincingly demonstrated the existence
of a close relationship between ozone content and circulation conditions,
and strengthened the hypothesis postulated by A. Kh. Khrgian in explana-
tion of the presence of an ozone maximum during-spring in the Arctic as
the result of specific circulation conditions prevailing there at the time.
- 173 -
-------
,~
. Many reports [3,6, and 8J deal with the relationship betwen ozone
content and gene'ral circulation conditions. Khrgian and Kuznetsov demon-
strated [3 J the relationship between ozone content and a type of circula-
ti on characterized by the distrib'htion of high-aititude crests and troughs
. over the 500-mb surface. It was found that during the transfer from one
type of circulation to another, O. 016-cm and O. 009-cm changes in the ozone
content occurred during the cold and warm half-years respectively. The
W, C, and E atmospheric circulation forms proposed at the AANII by
~. Ya. Vaingenheim, were used in explaining the dependence of total
ozone on circulation conditions in the A rctic during summertime.
W is the regional phase and E and C - the meridional phases of the
general atmospheric circulation. Figure 4, drawn according to Girs
[5 J, shows schematically the distribution of the main high-altitude crests
and troughs at the 500-mb level, observed most frequently in processes
.of the C-, E-, and W-type. The figure clearly indicates that the E and C
forms were characterized by a well-developed meridional circulation and,
in c'onnection with this, exhibited essentialand basic differences caused by
the different positions of the main crests and troughs.
,
, ,
In the case of Dixon Is. area, this '. difference is reduced to the
fact that in relation to the C -type circulation Dixon Is. was situated unde r
the NW portion of the crest spreading over Central Siberia, which created
intense interlatitudinal transfer involving removal of aerial masse.s from
the more southerly latitudes. In the case of E-type circulation, Dixon Is.
. lies under the eastern portion of the crest which extended over the Euro-
pean USSR, while aerial mass transfer moved from more northerly latitude~.
Total ozone determinati~ns wer,e_lTlade using air samples collected
in 147 days over Dixon Is. during the summers'oCi958~62. Results were
analyzed statistically. Daily total ozone determinations were evaluated
in the light of a given form of atmospheric circulation picked from
. Vaingenheim's catalogue j total ozone deviations from the mean under differ-
ent types of atmospheric circulation were then computed. Results showed
that greatest deviations in total ozone from the mean, + 0.015 cm, was
noted in the C -type circulation. Regions with richer ozone content generally
. mo~ed from :the Arctic 'into intermediate latitudes during summers [lJ;
acqn:dingly, a slight ozone increase over Dixon Is.~ in the case of C -type
circulation could be attributed to migration of mass'es with a higher ozone
content from the south to the Dixon Is. area. In the case of the W -type cir-
'culation, the deviation in the ozone content from the mean was only -0. '005 cm.
'This is understandable when flow regionality is taken into consideration.
, Total ozone deviation from the mean in the case of E -type circulation was
only -0.005 c'm. The axis direction of the European high-altitude crest
was frequently from SW to NE (toward Taymyr) during summer, which
caused flow zonality to predominate over the southern Kara Sea and the
deviation in the ozone content to be similar to that of W -type circulation.
,- 174 -
::/: '"
-------
I
I-'
-.J
VI
<0
~o
60
10
110
80
BARIC TOPOGRAPHIC CHARTS AT 300 MB LEVEL DURING 18 .\
AND 20 MARCH 1961
.1
. I
.... Fig. 3.
Fig. 4
./
i
I
!
\
\
,
I
I
I
I
. !
.,
I
i
-f!
- --c :
SCHEMATIC ILLUSTRATION Or ALTITUDINAL (500 MS) CRESTS AND
TROUGHS IN THE CULMINATING STAGE OF PROCESSES FORMS W, E,
AND C.
-------
The W -, C -, and E - circulation types indicated above are simplifi-
cations of a great variety of Arctic synoptic processes. Each circulation
type naturally corresponded to a given set of synoptic conditions over a
given region. Total ozone deviation over Dixon Is. appeared as a function
of synoptic process related to specific types of atmospheric circulation
as defined by L. A. Dydina, who examined six groups of synoptic processE:s
(A, B. C, D, E, and G) in relation to Arctic baric field distribution [4J.
In the case of southern Kara Sea region cyclonic conditions normally pre-
vailed in the A, C and E processes, and the anticyclonic conditions in the
B. D, and F processes.
Table 2
---- -~--
TOTAL OZONE DEVIATIONS IN THE USE OF
DIFFERENT TYPES OF SYNOPTIC PROCESSES
OVER DIXON Is.
CYCLONIC ANTICYCLONIC
ATMOSPHERIC CONDITIONS CONDITIONS
CIRcULATION (PROCESS (PROCESS
FORMS GROUPS A,B,E .GROUPS B,G~~)
W +0,002 CM -0,005 nf
(10) (35)
C +0,026 CM -0,001 CM
(23) (15)
E +0,010 CM -0,012 CM
(21) (35)
Calculations of total ozone devia-
tions from the mean for these groups help-
ed to es tablish the fol1o~ing relationships:
Greatest deviations in the ozone content
(+ 0.026 cm) occurred unde r cyclonic con-
ditions over the Di~ori Is. . region, resulting
from the C -type circulation (Table 2). Unde::
, .
similar synoptic conditions, induced by
W -type circulation, ozone content deviations
amounted to a mere + 0.002 cm; deviation in
ozone content was int,ermediate in the case
of E-type circulation. OZ0ne content devia-
tion was small for all circulation types and
did not exceed - O. 012 c m for the E -type
circulation under anticyclonic conditions
even in the Dixon Is. Region.
The present authors' conclusions were based on a limited volume of
materials for a summer period when synoptic processes in the Arctic were
less intense. Therefore, the results must be considered as of preliminary
nature, and incomplete for the solution of the problem. However, the ob-
servations in conjunction with similar observations by others point to the
need of continued efforts in this direction. :
Bibliography
1. r Y III II II r. n. Hcc.'e.ll.OBaHllf 'aT~loc
-------
6. K 3 P II MOB 3 f. Y. HeKOTOpblepc3y.%T3TbI H36J1IOJ1eHllii 33 CO.Qep>K3HlleM
039H3 B ApKTHKe. M3TepH3.~bl KOHcjJepeHUHii no IITor3M Mff (1960 r.)
II ~leTeOpOJlOnI'JeCKOrO 113Y'Jellll!l AIIT3pKTH.QbI (1959 r.). fll.QpoMeTeOH3JJ.3T,
M., 1961.
7. X B 0 C T II ~ 0 B M. A. 4>IIJHK3 0301l0ccjJepbl II HOHoccjJepbl. Pe3YJlbT3TbI IIccJle-
.QOB311I1H no nporp3MMe Mff. 110HoccjJepHble IICCJle.QOB311I1!1, .N'~ II. 113.Q-BO
AH CCCP, M., 1963.
8. 6 e 3 B e p x H II if ill. A. OJOHOMeTpH'JeCKlle }J.3HHble-no AJlM3-ATe B conOCT3B-
JleHlI1I C neKOTopblMH MeTeOpOJlOrH'JeCKIiMII cjJ3KTOp3MH. TpY.ll.bI K33HI1fMI1,
Bbln. 5, 1955.
CHARACTERISTICS OF WINTER AND SUMMER AIR CIRCULATION
IN THE NORTHERN HEMISPHERE STRATOSPHERE
Kh. P. Pogosyan and A. A. Pavlovskaya
I The structure, energetics, and dynamics of the atmosphere, the
. diurnal and seasonal variations in the meteorological paramete rs at high
altitudes, including ozone, are of the utmost interest to meteorologists
around the world. These can be studied best in the stratosphe re which is
outstanding for its complex circulation conditions. Different seasonal
radiation conditions in the stratosphere induce different circulation systems
during cold and warm periods of the year, such as predominance of western
and eastern transfers during winter and summer, caused by the formation.
of respective polar cyclones and anticyclones.
Polar stratospheric anticyclones which predominated in the summer,
begin to attenuate during the second half of July. This process originated
in the lower and progressed into the upper levels. A decrease in the geo-
potential during July-August affected the circulation system only slightly,
and the easterly transfer remained well expressed. A decrease in the geo-
~otential and, generally, attenuation of the anticyclone, augmented the cir-.
culation of rapidly-developing tropospheric whirlwinds which affected the
wind conditions in the stratosphere. Deep cyclones are associated with
the occurrence of individual troughs at the 25 -30 -km levels oriented from
south to north. Separation of a polar stratospheric anticyclone center and
departure of two newly-formed centers to the south, mark the beginning
of a transition into winter circulation. Such a process occurred in the mid-
dle of August at the 30-mb level, and toward the end of that month at the .
lO-mb level. Traces of cyclones were observed at times at still lower
levels in June, when the stratospheric anticyclone attained the highest
intensity.
- 177 -
-------
Subsequent development, of the proces s was limited to cyclone intensi-
, fication and its displacement towar,d the Pole. A well-defined western
transfer of air was established in the stratosphere at the high and, partly,
intermediate latitudes as early as the end of September., This process'
becomes intensified due to transition to hibernal radiation conditions.
Thus, toward the end of November, the height of the darkened atmosphere
over the Pole exceeded 300 km, and the radiative cooling of air amounted
to approximately 0.3° / day at heights of 20 to 30 km, leading to a further
decrease in geopotential and in the intensification of the hibernal stratospher-
ic polar cyclones and western circulation. Significantly, the first half of
,the cold half:-year was characterized by a comparatively calm circulation
regime. The second half was characterized by frequent and, at time,s,' !'
, sharp ,circulation disturbances accompanied by meridional reorganization
of the baric field and large temJ:>erature changes up to 30-40°.' ,An'almost-
stable western transfer became, established in the northern hemisphere as
early as October. Simultaneously, in the southern (warm) hemisphere,
above the 16 -20 -km level, the circulation became predominantly eastern
encompassing the entire stratosphere. These circu'lation conditions pre-
vailed through the entire winter and the first half of spring. The type of
, circulation in both hemispheres changed basically toward ,June; the' northe'rn
and southern hemispheres appeared to exchange positions. The south was
characterized by the westerly circulation which propagated over the entire
strato- and mesosphere, and in t~e north, above the 16-20-km level, a
stable easterly process became established.
Obse rvations indicated that the easterly stratosphe re circulation'
was more stable during warm periods than the westerly circulation during
winter. Thorough studies conducted in the northern hemisphere indicated
that sharp meridional transfers of the ,thermal and baric fields, charac-
teristic for winter, caused disturbances in the westerly transfer over a
considerable portion of the hemisphere. In some ca:ses, easterly transfer
became established over individual regions.
The general nature of mean seasonal temperature distribution and
the predominant winter and summer circulations can be judged by the
vertical profile of the atmosphere up to 90-100 km, plotte,d on the
basis of the radiosonde and rocket soundings during 1960 [2J, including
refinements introduced by the author as shown in Fig. 1. The Figure shows
1) ~ well-defined {roposphere with substantial. te'mperature gradients and j'et
streams, 2) stratosphere with a characteristic increase in temperature with
height, with western winds during winters and the eastern winds during sum-
mers, 3) decrease in temperature and in wind velocities with height,
(western during winters and the eastern during summers).
HoJever, as was previously pointed out, air circulation in the
stratosphere frequently' experienced considerable changes, particularly
, - 178 -
-------
. Fig. 1
Do
.....
-.J
'"
WINTER:
SUMMER
C
In
ca. .
C> .
o .
co
CI
c
GO
<:)
Q
....
Q
co
Q
ALTITUDE, KM
---- 2
-..-6
-3
---4
--...- 5
-I
AVERAGE. VELOCITIES OF PREVAILING WESTERN AND EASTERN WINDS BETWEEN THE EARTH'S SURFACE AND 30 KM LEVEL DURING
WINTER AND SUMMER
1-lsOTACHS; 3-ISOTHERMS; ~TROPOPAUSE; 4-FRONTS; 5-ST~ATPAUSE; 5-MESAPAUSE
::s I» [)Q 1-"
o ::s (\) ::s
Ii p-. 0 c;-
3 (') 't:J ::s
I» '< g. {/]
..... ~ (\) (\)
{/] 0 ::s 0-
(\) ::s ::t ~
I» (\) I» Ii
{/] ~ ..... 1-'0
g 0 ~~
I» (')
..... (') E?~.
(') ~ (\)-::s
~ ~ ....
I» .... (') (\)
::s ::s (\) Ii
OQOQg.1»
~ I» (\) ::s
{/] Ii 0-
I» 0 {/]
'""'t:J
Ii .... Ii
(\) ~ 1-"
{/] (\) ::s
e. ~~
.... ....
c" 1-"
o (\) ::s
'"" Ii ....
{/] ::s ::r
::r I» (\)
I» ..... '""
Ii {/] 0
't:J .... Ii
(') Ii 3
.... I»
Ii 0 0
(') {/] '""
~ 't:J I»
Pi' ::r 't:J
""(\)'t:J
5' ::t Ii
::s (') (\)
(')-
(') 1-"
0- '< I»
(\) (') c"
;5. 0' ro
~ ::s (')
1-'0 (\) ::r
o III III
P ::s ::s
::( 0- ~
o.{/] {/]
3 ~ 1-"
3 ::s
....
::r 3 ....
eb (\) ~
Ii
-------
H- ro -< -'PJ ....-<()
~~~OOoro ~tg"
(1) I-+, ~-J I-+, ::s
PJ 3 '"d .... " 00
lid OI1QOro
00 ro::! (;0 300
CIJgO::!N PJ
~31-+,;:t.() H-
I-+, p., PJ P"' .....
I-+, 0" .... I-" P"'
rorol-+, PJ ro
~~(?--"
.. ~... PJ
p., .. ()
I I
H DKM
J/92[
"12 r
J03J
I
2952
I
2872 r
I
2792
r
I
I
VIII I
Fig. 2
IX
XII
III
IV
! ,
~ ;.,"
v
VI
x
/I
-0-1960-61
XI
1958-59
-"-1961-62 -' -'1962-63
- - -1957-58
1959-50
DAILY GEOPOTENTIAL VALUES IN THE CENTER OF THE STRATOSPHERIC CYCLONE AT THE 10 MB LEVEL DURING THE PERIOD OF
SEPTEMBER - MAY 1958 - 1963
.-..--- ~---
-------
exceptionally sharp changes. Most marked changes in the system of a
polar cyclonic whirlwind in the 1958-1963 period occurred at a period be-
ginning with the latter part of January and ending withl the first part of
. February, 1958. A normal, and as later observed, f~equent1y-recurring,
process of the polar cyclone center separation began January 23. On that
day, one of the centers was over eastern Asia, the other over Greenland.
The. removal of warm air from the moderate into the high latitudes, in a
system of a stratospheric frontal region which enveloped the leading edge
of cyclones, was accompanied by the adiabatic heating which led to a con-
siderable increase in the geopotential of both cyclones [lJ. Priming of"
the latte~ was sointens-e, that, as early as February and the first half of
March after the process ended, the values of geopotential at the center of
cyclone exceeded those observed during the identical months of the other
years by 150-190 dkm in February and 40-100 dkm in March.
Fig. 2, illustrates that priming of a hibernal polar stratospheric
cyclone during spring differed from that during autumn. It began early
in 1959 and 1961 (approximately 25 February) and ended comparatively
readily in 1959 on 10 April and in 1961 on 18 April; during other years it
extended over longer periods. For example, in 1962 the process lasted
nearly 90 days, and in 1959 - only 45 days. With the polar day onset
the stratospheric air temperature subsequently rose, the isobaric sur-
faces became elevated, and the polar stratospheric cyclone became
gradually charged-up in all cases. This process was closely connected
with the disappearance of the north-Pacific hibernal stratospheric anti-
cyclone. Final transition to the summer regime, when stable anti-cyclones
. formed in the stratosphere over the Pole and the easterly circulation be-
came established over the entire hemisphere, came to an end normally
in mid-June. The formation of an anticyclone did not occur uniformly
every year. Four year's data indicated that the beginning of its stable
formation differed over a wide range, which can not be the result of
radiation conditions.
Graph in Fig. 3, plotted by M. Ts. Shabel'nikova depicts geo-
potential variations at the center of a summer stratospheric anticyclone,
and indicates that the summer stratospheric anticyclone was characterized
by a calmer regime. Curves of the H values for different years are very
close, particularly on the RHS of the- graph which reflected a decrease in
the absolute geopotential and anticyclone disintegration dl.lring August-
September. On the LHS of the graph, which shows the H values at the
anticyclone center during its formation~ curves differed considerably
from the autumn curves at the time of cyclone disintegration.
Because atmospheric processes are closely related, dates of I
summer anticyclone appearance at h~gh latitudes must have some connec -.
tion with the dates of hibernal stratospheric whirlwinds disintegration.
- 181 -
-------
Fig. 3
mo~
3220 [
3200 ~
I
....
3180 t
, .
,."j.J
,;
,....jJ'
'/
"
,-',:-
J"---'
3160
3100
IV
v
---1958
VI
VII
VIII
IX
1950 - - -1961 -. - .1962
-".U- --~ - --..
DAILY GEOPOTENTIAL VALUES IN THE CENTER OF THE POLAR STRATOSPHERIC ANTICYCLONE AT THE 10 HB LEVEL
DURING SUMMER OF I~ AND OF t960-62
-~ - ---<. --
As was previously pointed out, the latter can vary just as the former do. i
For example, narth of 80° the stratospheric anticyclone center appeared,
in April in 1960, and at the end of May in 1962. Accordingly, Figs. 2 and
3 indicate that summer anticyclone disintegration during August - September,
and the formation of a hibernal stratosphere cyclone during September -
November, proceeded more uniformly during different years than did the
cyclone to anticyclone transition during January - May.
This gives rise to the following question: How can these basic I
differences in the processes during the spring months and their uniformity
during autumn and pre-winter months over various years be explained?
It can be assumed that the special air circulation characteristics in the
stratosphere were due basically to different ozone concentrations in
the polar stratosphere during spring and autumn. However, under the
conditions of a generally-inc1"'eased ozone concentration during spring,
, atmospheric processes develop differently in different years. Moreover,
periods of considerable geopotential changes were comparatIvely short
and, consequently, radiation could not be the decisive factor in the intensifi-
cation and attenuation of stratospheric polar whirlwinds during winter and
I
I
I.
summer.
- 182 -
-------
I
!
I
,
. .
The belie!,f prevails that cyclones and' anticyclo'nes, which occurred ill
the troposphere,' fail to reach the stratosphere; the hypothetical assump-
tiem that the tropospheric processes effect a change in the temperature field
and [2] in the stratosphere circulation has not been generally or fully
. refOgnized. On the other hand, experience with daily charts of high-altitude
su.rfaces pointed to the validity of this hypothesis. Actually, intense' baric
tr?POsPhere formations were frequently well defined in the lower stratosphere
at Ithe 20 -25 -km levels ~ Cyclones which developed with great intensity
,
over the continents, frequently extended to 28-30 km, and, during summer
. I
to 20-22 km.
Cyclones and anticyclones propagated during winter and summer
di~ferently for the following reasons:.It has been known that the summer
direction of the horizontal stratosphere tempe rature gradient, which be-
caine established over the tropopause, held for heights up to 25-30 km,
and, judged by the nature of the seasonal stratosphere temperature field
apparently even higher, up to the 50-60 - km level.' However the winter
. direction of the horizontal temperature gradient above the l8-20-km
level at the intermediate and high latitudes, again became identical with
that of the troposphere. Thus, it can be seen that the seasonal nature
of the stratosphere temperature gradient direct during winter and summer,
caused by the radiating heat transfer, particularly at high latitudes under
the conditions of the polar night and day, produced different results.
During winter, between the intermediate and high strato~l.phere
latitudes, the horizontal temperature gradient increased with hei'ght,
the temperature contrast increased, winds became intensified, ar\d, in
the 50-700. region at heights from 25 to 35 km, attained velocities of.
40-50 m/sec. Wind velocities above the tropopause decreased du-ring sum-
mer becoming easterly with height; at 25 to 35 km in the same region
(50 -700) the winds normally did not exceed 10 -15 m/sec. Consequerltly,
horizontal and vertical air circulation intensity in the lower stratosphere
increased with height during winter and decreased during summ,er.
The above is a summary of the basic reasons for differences in.
the nature of stratospheric processes during winter and summer. Thexe-
fore, stratospheric processes which occurred during winter and summer
exhibited an incomparably higher activity than those which occurred during
summer and fall. It also explains the frequent temperature increases in
the stratosphere during winter and at the beginning of spring, which were
completely unrelated to the radiative air heating.
. It sho~ld be noted that efforts have als 0 been made to explain the
intraseasonal anomalies in the stratosphere temperature field in terms of
ozone concent~ation changes. The latter became enhanced by the relation-
- 183 -
-------
ship established between changes in the ozone content and changes in air
temperature. It is almost certain that this relationship existed. Dobson
and others have previously pointed out changes in total ozone in a system
of expanding baric formations. These changes are connected with air
mass advection from the north and south. Naturally, the amount of ozone
increased with tP.e air inflowing' from the north (at the rear of a cyclone),
and decreased with the air inflowing from the south (leading edge of a
cyclone). .
. Measurement of ozone content at different heights had been made
in recent years. Results showed that vertical ozone distribution was
frequently changed and that increase and decrease in ozone content we,re
accompanied by rise and fall in temperature at the identical levels. How-
ever, an increasing number of investigations led to the conclusion that
changes in ozone content and in air temperature we re both the res ults of the
same cause. Changes in ozone content and temperature indicated particu1ar-
'ly that the two phenomena were controlled mainly by the vertical air move-
ments [5J. In the stage of desc~nding movements in the 10-25 km 1ayejr,
air descended from the upper levels and contained more ozone than was
normally available at a lower level. In view of this, air tempe rature in-
creased adiabatically. This can be used in explaining the excellent co:tre1a-
tion manifested between the temperature increases and ozone content in the
lower stratospheric layers. i
I
i
In the horizontal transfer stage, air from the north was norma~ly
ozone-rich, and the air from the south-ozone-poor. However, it was:
difficult to establish in either case a close relationship between the i
temperature increase and ozone content, or the converse for the vertical
transfer. . This, apparently, can be explained by a change in the tempe:rature
distribution in the stratosphere, caused by atmospheric processes. I~
addition, an increasing number of investigators began to lean to the con-
clusion that the same advective and dynamic processes in the stratosphere
as those in the troposphere, played the leading role in temperature field
change's and in air circulation. Moreover, the effect of tropospheric
processes on changes in wind and temperature fields in the stratosphere,
was an e s s entia1 fac tor.
Bibliography
I. naB J\ 0 B C K a!1 A. A. I3l1yrpllcc301lllLle 113MCIICIIIIH 3rMOC(pCPIlOii IlllpKy.~H'
1!l1II B IJcpXllcil rpOlloc
-------
I
!
I
4. n a II JJ 0 II C K a II A. A.. nOJJC TC~I.'lcpaTypLI II CIICTCMC pa31!lWa;OUlllxcH oaplf-
4CCKIIX 00pa3013allllll B fllDKIICII cTpaToc<;1I.iJ.PO.10rlBl .n. 'McTeopo.10rllR, N~ 2, CO(~IIR. 1963.
6. Dally IOO-mlilibar and 50-nllllibar Three funcs ""\onthly 3D-millibar synoptic
- weather mars. or the IGY pcriod VII 1957-VI 1958. Washington. 1961.
I. Metcorol. Abhandlugcn. Bd X, XII. XIII, XV-XVI.I. XIX, XXV. XXVI.
. I
i
A COMPARATIVE ANALYSIS OF OBSERVED PLANETARY
DISTRIBUTIONS OF OZONE AND CERTAIN
RADIOISOTOPES IN THE A'TMOSPHERE
1. L. Karol'
1. Introduction
1. 1. The relationship between observed me ridional distribution of
and seasonal changes in total atmospheric ozone and the world-wide dis:"
tribution of radioactive fallout due to nuclear explosions was noted in 1956
by Stewart et al [19J and by others [13, 8, l6J. Records of such observa-
tions contain no detailed analysis or comparison of existing ozone data with
radioactive debris in the atmosphere. The present study represents an .
attempt to analyze and compare data published on meridional and vertical
distributions of mean monthly ozone concentrations in relation to some radio-
isotopes released by giant nuclear explosions in the troposphere and lower'
stratosphere of the northern hemisphere. Results of the analysis should
be helpful in arriving at some preliminary qualitative conclusions regarding
the nature of world-wide ozone and fallout distribution in the atmosphere'
and on the special characteristics of universal atmospheric movements.
Such studies constitute a part of a recent rapidly-developing trend in
meteorology, namely to determine atmospheric movements and processes
with the aid of tracer substances [3J.
2.
A Brief Review of Data on World-Wide Atmospheric Ozone
Distribution and Seasonal Changes in its Concentrations
2.1: It has been established [5, 6J that atmospheric ozone was
generated and disintegrated mainly in the stratosphere under the effect of
ultraviolet radiation, and that above the 25-30-km level it coexisted in a
state of photochemical equilibrium with atmospheric ozone, 1. e.. the
number of ozone molecules forming and disintegrating .per unit volume
- 185 -
-------
per unit time was constant. Below the 25 -30 -km level ozone formed a
. practically non-decomposing gas admixture transported by atmospheric
movements into the lower stratosphere and troposphere. It has been
assumed that ozone idsintegrated in the lower troposphere and at the con-
tact phase with the underlying surface; however, quantitatively speaking,
this process had been studied little thus far.
2.2. A considerable volume of data on total ozone had been accumu-
lated over a fairly heavy network of observation checkpoints by spectro-
photometric methods, particularly during the IGY -IGC period (1957 -1959)
[1, 4, 16J. Considerably less data are available on vertical ozone con-
centration distributionl measured by the inve rsion method, and on ozone
concentrations in the troposphere and by chemical methods near the ground
[2-9J. The accumulated data are in satisfactory agreement and can be
used in discerning the following basic characteristics of meridional dis-
tribution of and seasonal variation in mean monthly ozone concentration
in the atmosphere over the northern hemisphere.
l} Total ozone content and its concentration at different levels
below 25 km exhibited well defined annual fluctuations with a minimum
over the equatorial region during summer, one outside that region toward
the end of winter and also during spring, and a corresponding minimum
..
during winter and autumn.
2} The amplitude of annual fluctuations and the meridional dis-
tribution of total ozone and its concentration at individual levels during
periods of maxima and minima decreased toward the south. During the
intermediate months, meridional distribution exhibited at times a weakly-
defined maximum in the moderate belt [5, 6, 9, 14J.
3} Vertical ozone concentration distribution exhibited an almost
constant maximum at the lower boundary of the photochemical equilib,rium
layer (25-30-km), and, at times, toward the end of winter and during spring
a second broad maximum in the lower stratosphere (in the 12-l4-km layer)
at the high and moderate latitudes. Accordingly, seasonal ozone con-
centration variations attained 'at these latitudes maximum amplitudes in
the 12 -13 -km laye r.
4} Phase displacement was observed (Fig. l) at the beginning of
spring total ozone maximum (lag in the case of displacement to the south)
and ozone concentration at a given level (lag with a decrease in the level
height). .
1 In th{s article~ oy ozone concentration is
or its partial pressure in mk/mb (Ed. )
- 186 -
meant either its density in cm/km,
-------
Fig. 2 -+ 1 MAY
KM
40 2 2
~
30 60
12
12
20
2
50 JULY
1
40 2 2
~ ----: "
6
30 10 8
UJ 10 10
o ~
::> 8
!: 20
I- 6 "
-'
< 2
10 4
Fig. 1
-- -----.-.
MKH8 ;
----..- H -.-- . - .
ANNUAL COURSE OF OZONE CONCENTRATION
CHANGES AT DIFFERENT ATMOSPHERIC LE~
ELS OVER AROSE (4~ No. LAT.) Acco~o-
ING TO DETERMINATIONS DURING 1955-
~ (9)
5) It has been established
that total ozone sharply increased
from the subtropical and polar jet
stream axes toward the north, par-
ticularly during periods of spring
ozone maximum (Fig. 2) [I, 8, l7J.
50 .
NOVEMBER
:: i=t~{
8 ~
20
2
"
10
2
70
100N
50
LATITUDE
30
AVERAGE HERIDIONAl OZOIlE CON'~ENTi\ATlOH DISTRIBUTION
ACCDI1DI NG TO DETERMIIIATI O~S MADE BY THE UH1(EHR
HETHOD BEFOR E 1 ~ (, 7)
6) Total ozone magnitude
agreed well with its amount in the
l2-24-km layer, i. e., ozone con-
tent in the lower stratosphere at the intermediate latitudes, below the, 50-
called, critical level of photochemical ozone equilibrium; a particularly
clear correlation was discerned during winter and spring. For all practi-
cal purposes total ozone magnitude represented its amount in the lower
stratosphere (at the l7-km level in Fig. 1) and reflected air mass movements
in that layer [1, 5, 9, 15-l7J.
7) Difficulties encountered in obtaining representative data on ozone
cpncentrations iti the troposphere and near the ground resulted in the present -
day lack of a clear concept relative to the meridional ozone concentration
distribution near the ground and in the lower troposphere. Me,asurements
taken at individual, especially mountain, stations at the moderate and sub-
- 187 -
-------
. ,
tropical latitudes confirmed the ,existence of seasonal changes in ground
ozone concentrations. with a maximum in May and minimum in December.
[13J. Direct measurements of ozone flow toward the ground st,irface, on
,the basis of which total ozone removed from the atmosphere by the under-
lying surface [13J has been estimated. are practically unavailable.
3.
Data on the Worldwide Distribution of Some Radioisotopes in the
Atmosphere and on Seasonal Variations in their Concentrations
3. 1. Radioactive debris from nuclear explosions. such as uranium
and plutonium fission fragments and isotopes generated by neutro'n 'Iluxes
in the atoms of a surrounding medium penetrated the troposphere as
aerosols and gases, together with radioactive clouds generated by the
ground-, air- and. sometimes. underground and underwater kiloton-size
nuclear explosions. The radioactive cloud of a stJ;'ong megaton thermonu....
clear explosion, which contained almost the entire fallout as submicron
aerosols and gases. penetrated into the lower and. sometimes. middle
stratosphere. Accordingly, two types of global fallout. th~ tropo- and
stratospheric. can be distinguished. The tropospheric fallout occurred
mainly during 1-2 months following the explosion in the same zonal beit
as the point of explosion. The stratosphe ric fallout occurred with a greater
. .
or lesser intensity over almost the entire earth's surface during a period
of one to several Y'~.1:"S after t:he explosion. Moreover, a pool of radioac-
tive'aerosols and gases formed in the stratosphere which, as a result of air
mass transfer, gradually penetrated the troposphere through the tropopause.
Radioactive aerosols are readily removed from the troposphere during
1-2 months, reaching the earth's surface mainly with precipitation.
3.2. Determinations have been made of fallout and some natural
. . .
radioisotope concentrations found only on atmosphe ric and on atmospheric
precipitation aerosols, their ground run-off. or their content in soils and.
to a lesser degree, in fresh and salt waters at different geographic re-
gions. Other similar determinations have been intensively conducted in
many countries during recent years following nuclear weapons tests.
Such measurements are usually made by a two-stage method:
first by collecting samples of atmospheric aerosol. rain-, snow-, river-,.
seawater, etc. and second by making actual radioactive content determina-
tion. .
The collection of samples is comparatively simple. In the' case of
atmospheric aerosols, a large air volume (up to tens of thousands of
m3/day) is aspirated close to the ground through special filters by spe-
cial aspirators, or by using dynamic air pressure ~hen the samples are
collected onboard planes. The air can also be passed through special
electric filte rs. Subsequently, the ae ros ol-retaining filters are calcined
- 188 - .
-"".-.1/
-------
I
I
. and prepared f~r final determination.
Rain, ~now. and aerosol samples which became adsorbed to the
underlying surface are collected into a container or an adhesive-covered
flat table exposed in the open for a day, week, month. Subsequently,
the flat tables are calcined, liquid samples are evaporated to a dry resi-
due, or passed through ion-exchange rosin filters and calcined. Samples
. of fresh- and sea -water we re treated similarly, and soil samples with
vegetation, collected from areas which satisfy given requirements, are
calcined and pressed.1 Determination of individual radioactive material
concentrations in a sample proved to be more complex. All samples of
the external medium contained a mixture of diverse natural and artificial
radioisotopes. Determination of total beta - or gamma- activity in all
radioactive materials contained in a sample required the use of complex
radioelectronic equipment, and activity determination of a given radio-
isotope frequently constituted a serious research problem which could be
solved only by resorting to the latest advances in radiochemistry, nuclear
physics, and radioelectron!cs.:3 However, this did not i.mpede, but seems
to have stimulated the accumulation of measurement results on total beta
radioactive fallout in ground-air aerosols, in atmospheric precipitation,
in river water, and in the soil over large territories.
Considerably fewer determinations were made of individual radio-
isotope concentrations in the same objects, and only a limited number of
determinations are available for the troposphere and stratosphere. As a
rule, quantitative determinations are made of the following radiologically
important isotopes: uranium and plutonium fission products: - strontium-
90 (halflife Tt = 28 years); cesium-137 (Tt = 30 years), and also the
following short-lived isotopes: ce rium-l44 T~ = 285 days), cerium-141 ,
(T~ :: 33 days), strontium-89(Tt = 50 days), and barium-140 (T~ = 13 d!:tys).
3.3 Radiosotopes contained in the atmosphere and in other migratory
geophysical media can serve as excellent "tracers" in studying pattern
movements in the atmosphere and its reaction with other media. Mass-
wise3 radioisotopes represented a negligible but quantitatively me asur-
able passive admixture, and the simple law of their radioactive decay
makes possible the determination of the processes time scale labeled by
the isotopes. The determination of total individual isotopes is more suit-.
able for such studies than the determination of cumulative aerosols activity
in precipitation, soil, or water. It was mentioned that only few determin~-
tions have been made of individual atmospheric isotopes; therefore, the I
1 &:3Discussedin greater detail in [3, 4,7, and 19].
3 Approximately 40 kg of strontium-90 was generated by all the world's nu-
clear explosions since 1961.
- 189 -
-------
study of atmospheric movements must be paralleled by a comparison of
isotopes witl?- other tracers and, above all, with ozone.
Sources of atmospheric ozone and of radio~ctive debris which
comprise the global stratospheric fallout, are distributed at approximately
the same level in the stratosphere. Both impurities propagate into the
lower atmospheric layers, and are removed from the latter in the lower
tr~posphere. The basic physical difference between ozone and fallout is
found in the fact that the former is a gas, while the latter is principally,
adsorbed, by aerosols.' This difference should affect, above all, the natu~e
of their removal from the lower troposphere; aerosols, unlike gases, ar~
easily removed by clouds and precipitation. The transfer of ozone and of
,radioactive aerosols by ,air masses in the upper troposphere and strato-
sphere should occur almost identically. Therefore, in the study of down-
ward labeled mass air movements from the stratosphere, data on the
quantitative determination of ozone and radioactive debris seem to be
mutually complementary. ,There exists a large volume of data on total
stratospheric ozone in layers the radioactive debris concentration of which
was determined inadequately. On the other hand, there are only few deteI'-
minationsof radioactive debris concentrations in lower troposphere layers
with inadequate and insufficiently reliable ozone determinations [13J.
Therefore, co~parison and simultaneous use of the observed world-wide
ozone and radioactive debris distribution can be of value in the propagation
study of both atmospheric admixtures by the tracer method.
, In addition to a parallel comparative study of radioactive deb:z:is
with' ozone, a comparative study should be made of the observed global quan-
titative distribution of some natural atmospheric radioisotopes, such as
isotopes resulting from the decay of radon, thoron, carbon-14, berillium-7,
etc., although their sources in the atmosphere differ from the atmospheric
ozone sources to a greater degree than the stratospheric reservoir of
radioactive debris [3J. However, this problem lies outside the scope of
the present study.
3.4. As in ozonometry. the use of accurate and standardized
determination methods over a wide station network is of utmost value in
comparing results of determinations made at different locations and at
different times. The refore. the following published measurement data on
atmospheric debris concentration in the northern hemisphere were select-
ed fch analysis:
a) data on mean monthly strontium-90 concentration in ground air
at a network of stations, located approximately along the 80W meridian~
c-ollected between January 1960 and September 1961. The following stations
are located in the northern hemisphere: Thule (76N), Musorii (SIN), Wash-
ington, D. C. (39N), Miami (28N), and Miraflores, Canal Zone (9N).
- 190 -
-------
.,
. I
The data were obtained by monthly or bimonthly radiochemical analysis of
aspiration filters which were changed every other day; results are illus-
trated in Figs. 3 and 4 [10, 14]:
Fig. 3
10 - 2 D I S';;;;I No 113
I . "
2.0
~,
'.I \
'/ \
J \
-I
-.-.2
-..-..3
-4
---5
I,S
1.0
0.5
/0. -'---.--.
..-".",,' """',
"" '
.....
I II III IV V VI VII VIII IX X XI XII I II III IV V VI VII VIII IX
1960 1961
.. . -----_.- .- - - --------. ..--
SEASONAL AVERA~E MONTHLY CHANSES IN STRONTIUM-9Q CONCENTRATIONS IN
GROUND AIR LAYERS AT STATIONS LOCATED ALONG MERIDIAN 80° W. LAT. IN
. 1~51. (1~14).
I-GREENLAND~TuLE 75035' N. LAT~ 68D3StIW." LONe.J-~IRAFLORES 9°00'
No. LAT. 79"35' W. LONG.J 3-CANADA 51016' N. LA To 80° 39' W. LONGo;
4-WASHINGTON 38050' N. LAT. 7506?!.J~/. LONG.;.5-FLORIDA, MIAMI 2S049~
. No LA To 80017' W. lONG.
b) data on mean monthly concentrations of same isotope in the
free atmosphere at the following levels: 4'-6,-.7.6,12.2,15.3.18.3 and
20 km over Alaska (70N), USA (35N), . tropical region of the Pacific
(10-15N), Argentina and Australia (408). during May and November, 1960
and May, 1961, obtained by averaging results of radiochemical analyses
of airplane carried filters. collected in 2-3 flights made during the
indicated months, and shown in Figs. 5-8 [11; 12].
The January 1960 to August 1961 period was chosen for the follow-
ing reason: by January 1960 more than a year had elapsed since the 1958
powerful nuclear tests had been halted and the fall-out from the test series
conducted at different latitutdes interdispersed into the stratosphere
forming a commo.n reservoir; this was clearly indicated by the results
of determinations made as described in [14]. This reservoir can be
logically oompared with the ozone reservoir in the stratosphere.
Strontium-90 data have been used because ,quantitative determinations
of this long-lived isotope, the rate of radioactive decay of "v.hich during.
the period under consideration can be neglected, have been conducted ex-
tensively.
- 191 -
-------
;0 .1, 0 ISTR/MI No H3
1,8
1.6
o -- I I I J------1 I I
N 90. 80 70 60 50 40 30 20 10 0
LATITUDE
. - - . - .
MERIDIONAL DISTRIBUTION OF MONTHLY AVERAGE CONCENTRATIONS OF STRONTIUM-90
IN 'rHE GROUND AIR LAVER ALONG 800 W. LONG. DURING SOME MONTHS OF 1960-51
(IO,14J - . .
1950-- I-JANUARY; ~~ARCH;-3-MAY; 4-JULV; 5-SEPTEHBER; 6-NOVEMBER
95r - -JANUARV-FEBRUARV; d-MARCH-APRIL; 3-MAV-JUNE; 4-JULV-AUGUST
Fig. 4
'.4
0.8
0,6
0.4
. 0.2
0.6
0,4
0,2
.'
7.2
- -,
I
~t\
II .
/.' \
/
4
6-65
1960 .
x I
"
2
- --J
-,-,-4
1.0
o
06
o
"
1,4
1.2
1961.
,,'
x 1
0.8
1.0
--2
---J
-.-.-4
'-'- /\
~ X!\X~
./ '\X
\.-.
I I
1 I
30 40
50 60.$
Concentration determinations of another long-lived isotope, cesium-137,
presented a pattern similar to strontium-90 at a constant 5:3 ratio in kn
instances discussed in this report. .
- 192 -
-------
Fig. 5
I
----~-------"---'- --
D ISTII./MI No KG
10' -
8
6
5
4
J
2
I
8
6
5
~
3
2
10"/
8
6
5
"
J
2
10'2.
8 ..
6
5
"
J
2
"
.'0"3
. 8
6
5
"
3
" -
"
"-
" - /
"
\ /
\ /
\ /
\ /
V
.
. -
.
. .
"
)( I
---2
o
0-3
OF AIR
19. 8 HM
15.25HM
/
.... I I. t t
10 N 80° 10 60 50 40 30 20 10
iUupomo
....... ...4
l> 115
6
a '07
2 . '.~' -'- 8
I
o
;0 io jo 40°$
o
- -_4 ------
MERIDIONAL DISTRIBUTION OF MONTHLY AVERAGE. CONCENTRATIONS OF t)IRvNTlUK-
90 AT DIFFERENT ATMOSPHERIC LEVELS IN MAY (I- 3~ 5. 7) AND NOVEMBER (2.
4, 5. 8) 1960 (Ir,12)
- 193 -
-------
Fig. 6 Fig. 7
/(N
29 a)
27
25
23
I
101 i
20 .~ ..... 21 !
...... I
" 19
18 / :
,/ 17
16 . /
<. --- 1S
, - -
14 . ,...
KN
I 20 6)
~ 12 18
...0
+=-
I 10
8
-1 -1
-- -2 -.-.2
~.
" ~_. -.-.-3' ---3
-
'-
, I
. ., .,... -,' . i
0.04 0.08 0./2 0.16 0.20 0.24 0.28 0.32 0.36 0.40 0.44 0..48 0.~2 I
. ' iHSTRIB./MIN.. ~ I
VERTIC~L PROFILES OF MONTHLY AVERAGE VOL. CONCENTRAT'ioNSOF 'STRONTlU~
90 OVER ALASKA (700 NO. LATo) IN MAY (t) AND NOVEMBER (2) 1950 AND IN I
. MAY (3) 1951 (11,12) .
0.06 0.1 all, 0./8 0.22 0.26 0.30. 0.34 0.38 0.42 a~6 0.50.
. . , "., ,D ISTItIB.It:!Ufo M3 : !
VERTICAL PROFILES OF MONTHLY AVERAGE STRONTI U,M-90 CO NCE,NTRATIO,NS OVER I
USA IN MAY (I) AND NOVEMBER (2) 1950, AND MAY (3) I~I' ACCOROI NG TO AIR-
BORNE DETERMI NAT! ONS MADE AT 3~ No. LAT. (A) AND AEROSTATlO,.D£T.ERMI NAt-
.. ,.-.. ---TIONSOVER '3'-O"No.--i-AT~- (a) (11;- j'2r--' , \:
..
- -.",......-------
------ - _..-
_.--_.- -- .--..--- - -. -- _.- .-...-- '-..--'
..--. -~. --- - - _.
",- ---.---....
.----..-- "-'--"'------'-'
-- .-... ---..- .----.-.'.---- ....--..-.
----- ---- -.. -~-- '-.
~>
.
-------
Fig. 8
t
-------
Table 1
ABU
TAT
Mus
. Au ITUDE, KM"
I I ----
STAT'O-N' GROUNiJ TOTAL REMARKS SOURCES
LEVEL 6 12,5 17 21,S CONTENT
AROSE, 4-,0 2,0 2,0 2,0 2,0 1,5 1,5 MAXIM I N. FIG. I;
No. LAT. 1,9 1,5 1,5 1,3 1,2 1,3 MAV/NOVEMBR FIG. I
.
LAYER LI MITS, KM TOTAL I
STATION 0-616-12\12-18118-24 CO~I REMARKS
TENT
- 24.5 No~ LAT. j 1,0 1,3 1,6 1,3 1,2 MARCH!NoVEM
ENO, 360 No. LAT 2,1 2,8 2,1 1,7 1,5 MARCH!NoVEM
ONI. 510 No. LAT 1,1 1,4' 2,8 1,9 1,6 FEBR.!NoVEM
._--~
SOURCES
B.
B.
90
[ 17)
[17 J
[ 15)
- ~ ..-- - ...
IACC~RDI~G TO DATA IN REF. (13)
I
Table 1 shows ozone . concentrations ratio during seasonal maxima
and minima at different levels over some 'stations in the northern hemilsphere.
T able 2
OETE
M
.. ,"""oojou.o
ALTITUDE, KM
4'6/7'6l '. I
ADE AT LEVEL 12 15 I 18 I REMA~KS
20
.0 -
70 No. lAT. 4,5 2,8 1,5 1,8 0,77 1,1 0,95 AT GROUND LEVEL
STATION rULE
o 4,3 7,2 700 No. LAT.
No. LAT. 8,9 4,8 1,5 0,80 1,2 GROUND LEVEL
WASHINGTON
C OCEAN 39° No. LAT~
o No. LAT. 14,5 5,5 1,7 1,1 0,68 0,88 0,95
AT GROUND lEVEL
MIRAFLORES
90 No. LAT.
\
ALASKA
USA 35
PAC'FI
10-15
Table 2 shows similar ratios of strontium-90 volume concentrations
during May and November at approximately the same levels and latitud~s Cj.s
in Table 1. Effect of the systematic decrease in atmospheric strontiu~-90
content on the value of this ratio caused by the radioactive decay and fall-
out on the latter can be accounted for with the aid of Table 2 and .othe r i
I
Tables which list average values of mean monthly strontium concentrations
during May, 1960 and 1961. Comparison of data in Ta1:>les 1 and 2 show~
that ozone concentration ratios varied little with height and' exhibited
sllIilll m,ilxin\d in the lower stratosphere above the tropopause. Strontium
conc\:.~ntration ratios exceeded the corresponding ozone ratios in the tropo-
sphere, and approached unity in the stratosphe re, indicating that strontium
, - 196 -
-------
stratosphere cohcentrations are, on the average, not much smaller in May
than in November, Yet, this is no proof of the absence of seas onal stronti-
I
um concentration changes in the lower stratosphere. since data exist which
indicate that strontium and ozone spring maxima in the stratosphere oc-
cuirred at the end of February or in March. i. e.. prior to May. when the
determinations were made [18J. However. the amplitudes of annual stronti-
urll concentration changes in the lower stratosphere appeared smaller than
th~ corresponding amplitude for the ozone concentration changes ," This can
bel best explained by the fact that the stratospheric ozone source intensity
was of a periodic nature. particularly at extratropicallatitudes. and that
it Iwas related to changes in the sun's position. while the stratospheric
sOlurce of the radioactive debris was a fixed gradually-depleting strato-
spheric reservoir.
Table 3
.- - --
AT LAtER LIMITS, KM TOTAL OZONE
OZONE
RATIO MONTH GROUND I I 12-181 110 (17) I
LEVEL 0-6 6-12 18-24 110 (4)
Y- III 2,9 3,2 3,2 2,2 1,7 1,4
T XI 1,2 1 ,1 1,8 1,5 1,2 1,3
n III 1,3 1,4 1,3 1,2 1 ,1 1,3
- XI 1,7 2,5 1,2 1,4 1,2 1,2
Y
Note: Close to the ground data are according to (13): y IT .- ratio between
ozone in the temperate zone and ozone in the tropical zone; ply -
ratio between ozone in the polar zone and ozone in the temperate
zone.
Table 4
STRO
ColiC
TIO
--
ALTI TUDE, KM
mfuE: AT I I I I I
ENTHA-- MONTI! GROI/IID 4,6 7,6 12 15 18 20
N RATIO LEVEL
.-
Y V 4,7 10 3,7 230 3,5 3,2 2,1
T XI 8,8 7,9 2,3 50 24 4,0 1,7
n V 0,5 1,5 2,6 2,9 2,7 1,4 0,7
- XI 0,6 3,8 15 7,6 5,6 1,3 1,0
Y
Note:
Close to the ground station data are as follows: Mira£lores at the
tropics, Washington at the moderate zone, and Tulla at the tropical
zone; ply also for the polar and temperate zones.
.- 197 -
-------
~ -------;-
. 4.2. Tables 3 and 4 present relative values which reflected the
meridional distribution at different atmos phe ric ozone concentration
levels during March and November, and of str~mtium concentrations
during May and November. These values were, in ~act, two ozone
strontium concentration ratios at the moderate latitudes (30 -40N) and at
the tropical belt, and their concentration ratios in the polar region and at
moderate latitudes. Strontium concentrations for May 1960 and 1961 were
averaged as in the case of Table 2.
A comparison of data in Tables 3 and 4 with graphs in Figs. 2 and
5 shows that the meridional ozone and strontium concentration dis tributions
were practically identical in the tropo- and stratosphere up to the 18-km
level. In the. tropo- and stratosphere distributions the concentration
decrease was from the pole to the equator; moreover, steepest hori-
zontal gradients we re obse rved in the subtropical belt in the 7 - and 15 -km
layer, 1. e., in the distribution region of the subtropical jet stream. . The
relative strontium concentration' gradients exceeded the corres ponding
relative ozone concentration gradients at almost all levels below.18 km,
particularly in the latter region. It was shown in 4.1 that the relative ampli-
tude of annual strontium concentration changes in the troposphere also ex-
ceed the corresponding amplitude of ozone concentration changes. It fol-
lows from this that both the meridional and time ozone. concentration dis-
tributlons in a layer in which transfer occurred from its source to its
sink (surface troposphere) were of a smoother/nature than'similar strontium
concentration distributions. . .
Data in tables 3 and 4 indicate that both ratios changed from spring
to autumn in opposite phases up to the l8-km level in the case of strontium,
and in the entire layer under consideration in the case of ozone; 1. e.,
ratio of concentrations at the moderate and tropical latitudes decreased
between spring and autumn, while the same ratio increased at .the polar
and moderate latitudes.' This phenomenon is most likely related to the
seasonal displacement during summer of the steep horizontal ozone and
strontium concentration gradientsfrom< the subtropical region toward north.
The nature of the meridional ozone concentration distribution remain-
ed unaltered above the l8-km level, and the strontium distribution became
smoothed out, its maximum shifting toward the moderate latitudes (Fig. 5).
. This phenomenon may be caused by the fact that the height at. which strbntium
and other radioac~ive debri attained their vertical maximum volume distri-
bution increased with a drop in latitude. Figures 6..7 and 8 indicate t~at
at 70Nthis maximum was notable at the 16-18-km layer, although aero-
plane studies failed to obse rve it at other latitudes up to the .20 ~krn level.
However, results of strontium determinations made by aeroplanes in the
USA at latitude 31N up to the 27 -28 -km level during the period under in-
vestigation and presented in Fig. 7, indicate that the maximum rested
- 198 -
-------
in the 19-21-km 'layer.l Itis probable that the maximum volume concen-
tration of radio~ctive debris was higher; however, since no determinations
of radioactivity of the tropical stratosphere were made above the 20-km
level during the period in question, this assumption cannot be verified.
It can only be pointed out that a similar meridional altitudinal strato-
s};>heric maximum concentration distribution was observed in 1958-1959 in
connection with tungsten-185 (T~ = 74 days) which was deposited in the lower
equatorial stratosphere in the summer of 1958 by nuclear explosions con-
ducted in the Pacific Ocean [l1J.
The similarity between meridional ozone and strontium concentra-
tion distributions in the upper troposphere and lower stratosphere above
the tropopause indicated that propagation of both admixtures through the
tropopause occurred at all latitudes with an approximately-equal average
intensity. Moreover', the singular role of the subtropical tropopause
discontinuity as the principal channel for air mass diffusion between the
tropo- and stratosphere, widely interpreted in many works [8, 9, 13, 16,
- and 18 J, can not be distinguished. ,
4.3. An appreciable difference in the meridional ozone and radio-
active debris concentration distributions was observed near the ground
surface. It was shown above that comparatively few representative atmo-
spheric ozone concentr.ation determinations had been made near the ground sur-
face, which had been reviewed recently by Junge [13J. In accordance with
Junge's report ground ozone concentrations were practically identical at
all latitudes, although systematic data are available only for Arose and
several other stations in the subtropical and tropical regions; ground ozone
concentrations also exhibited seasonal maximum and minimum changes oc-
curring in May and November-December, respectively.
Figures 3 and 4, which illustrate me ridional distribution and
seasonal changes in ground surface strontium concentrations along the
80W meridian, show that the spring maximum was observed in March in
Greenland and at the Canal Zone, and in May at the remaining hemisphere
latitudes. An early spring maximum noted at Miraflores was related,
above all, to the beginning of the rainy season, which lowered sharply
the air aerosol concentration.
The concentration maximum was observed in the 20-40N region
1 As shown in [12J, absolute v<3;lues of results of the airborne measurement.
'of the radioactive debris concentration can not be regarded-as itiUy reliable- .',
due to the difficulties in determining the effectiveness of' filters and filter-
ed amounts of air in the stratosphere. However, the relative distribution of
strontium concentration used here is hardly affected by these difficulties.
- 199 -
I
-------
through the entire period. The second maximum at Thule, noted in
January, 1960, remained unconfi:rmed by determinations conducted
at other polar stations and by data obtained on strontium-90 fallout
. on the ground surface. . ..
Fig. 9
. -.-. -. - ---------
-. ---- -- ------,- .--. .--~-
- -. - ._- -. - -- ___nn
1.8
. 1800
1.6
1600
~ 1400
1.4
1-.
=
o
~
:: 0.8
...
1200 ~
I-
=
o
I
1000 j
41:
...
C\I
::I:
~ 1.2
=
o
::I:
z
- 1.0
8
l: 0.6
=
...
800 .~
z
z:
41:
...
600 ;:
o
I-
I-
.z
~ 0.4
. I- .
(/)
200
10
o.
.---,.----.----" --- -- - -.
--.----. --- -..------.--- ------------.--------.---
MERIDIONAL DISTRIBUTION OF SEASONAL AVERAGE SR90 FALL-OUT W1TH
PRECIPITATION BETWEEN MAY 1960 AND APRIL 1961, AND ANNUAL SEA-
. SONAL AVERAGE OF CLIMATIC PRECIPITATION (II, 12)
STREAKED PARTS OF BLOCKS INDICATE PREciPITATION.
Blocks in Figure 9 illustrate the meridional mean regional fallout
distribution of strontitim through precipitation for the year of May 1960'
to May 1961 and the mean regional precipitation distribution for one year.
The figure shows that the fallout and annual precipitation maxima occurred
at the extra tropical latitudes in the 30-50N belt, and that the fallout varue :
decreased noticeably toward the pole and the equator. The difference be-',
tween the meridional ground-surface. ozone and radioactive debris dis-
tributions may have been due to the fact that the latter was carried by :
aerosols washed out from the lower troposphere by <::louds and precipita-
tion, while ozone wa~.practically an unwashab1e gas. HoWever, the
meridional ground surface distribution of strontium can not be determined
entirely by precipitation distribution, otherwise a minimum would be
noted in the extratropical maximum precipitation region (30 ~50N), which
. - 200 -
-------
~ '
is at variance y.rith results of the determinations. In this connection it is
interesting to note that even at the 4.6 km level the meridional strontium
distribution differed significantly from the ground-surface distribution
and exhibited a maximum in the polar region. Table 3 shows that strontium
, concentration in the 0 -6 -km layer also inc reased toward the north, to a
lesser degree. '
It follows from the foregoing that an intensive removal of radio-
active aerosols from the middle troposphere occurred in the lower
, troposphere at moderate latitudes with the following results: it produced
1) a ground-surface concentration maximum without precipitation participa-
tion, and 2) a radioactive fallout maximum at moderate latitudes with
precipitation participation. The radioactive aerosol removal can be caused
by several atmosphe'ric processes: a) descending currents in the subtropical
anticyclone region, b) macro-turbulent ve rtical' transfer elicited by fronts,
cyclones, and anticyclones at intermediate latitudes', and c) ve rtical tu'rbulent
transfer which is more intensive at intermediate than high latitudes and
which was definitely convective during spring and summer.
, 4.4. As shown in [13J, Junge compared 1) the onset time of mean
monthly ground surface ozone concentrations 'in the spring of 1950-1952 (May),
2) the mean monthly total specific beta activity of the radioactive debris
in the ground surface air at a station network along the 80W (80N'?)
meridian in April of 1961, and 3) the mean monthly berillium-7 concentra-
tion in the ground-surface air, a radioisotope generated in the general
atmosphere under the effect of cosmic radiation [13J for several stations
at intermediate latitudes in April of 1960-1961. On the basis of this com-
parison it can be concluded that the spring ground ozone maximum occurred
one month following the radioactive aerosol'maximum. Assuming further
that the spring ozone content and radioactive aerosols maxima occurred
simultaneously at the tropopause level, and taking into consideration some
supplemental facts, Junge concluded that the two months mean time during
which ozone remained in the troposphere exceeded by one month the mean
time T during which radioactive aerosols remained in the atmosphere [3,
feiJ: -_. a ,
Junge assumed that ozone and radioactive aerosols entered the tropo-
sphere mainly at the intermediate latitudes and that the subtropical tropo-
pause discontinuity played an important role as a path for their ingress.
For an average of two months it remained in the troposphere at which time
ozone blended throughout the general hemisphere and produced the homo-
geneous ground-surface concentration observed by Junge. Th,e average
one month during which the radioactive' aerosols remained in the tropo-
sphere proved insufficient for their complete blending which produced their
maximum ground-surface concentration observed at intermediate latitudes.
- 201 -
-------
Junge' 5 concept about To being greater than T a appeared rational
because ozone, which was not washed out by precipitation, remained in
the troposphere, probably, longer than the aerosols, although the proof
and evaluation presented in [13J were not sufficiently convincing. . Graphs
.in Fig. 3 show that the spring maximum of the ground-surface strontium
concentration occurred at intermediate latitudes alc:>ng- the ~_OW ,meridian
in May, 1. e., at the time when the ground-surface ozone concentration
was at its maximum... The spring increase in berillium-7 concentration
in the surface .air layer may not be associated with the arrival of addi-
tionai quantities of this isotope from the stratosphere [3J. Data on
general ground-surface ozone concentration are meagre and onset time
radioactive debris maxima contained in the ground-surface air at differ-
ent stations are widely dispersed. Therefore, it is still impossible to
find a well defined .mean phase shift in the ground-surface ozone and radio-
active debris concentrations which could be used in determining To. The
preceding discussion presented fn 4.3 clearly brought out the groundlessness
of JUhge's assumption concerning the dominant ozone and radioactive aero-
sols penetration into the troposphere at ihtermediate latitudes.
I
Junge correctly pointed to the need for organizing a. meridional
station network similar to the network along the 80W meridian [13J ozone
determinations. A meridional network should also be organized for
making o'zone concentration determinations at different troposphere le'fels
to cp.eck the similarity of meridional ozone concentration distribution in
the troposphere and lower stratosphere, and to determine the phase slUft
. I
of its seasonal changes at different levels. 1
I
. 5., Resume
5.1. The scope of the present work limited the review of exist~ng
hypotheses on the causes and mechanisms of observed meridional distribu-
tion and of seasonal changes in ozone and radioisotopes concentrations at
different atmospheric levels [1, 8, 9, 16-18J. The proposed hypotheses
can be grouped into two basic classes. Class 1 includes hypotheses w~ich
postulated that observed meridional ozone and radioactive debris distribu-
. I
Hon resulted from ordered me ridional flows in the tropo - and stratosphere
in general, and from the Brewer-Dobson transfer in the lower stratosphere
from the equatorial belt t~ward the poles, in particular [8, 1.7-l9J. 1jhe .
other hypotheses ascribe a principal role to macroturbulent ozone and iradio-
active debris diffusion from their sources in the stratosphere [9, 16].
The hypothesis of G. P. Guschin merits special consideration. It defines
observed me ridional ozone distribution and its seasonal changes in terms' of
. I
latitudinal and seasonal solar radiation changes which decompose ozone be-;
low the 20 -km level [1]. ,,' ,,-- I
- 202 -
-------
All the above hypotheses explain the results of observations only
qualitatively. The quantitative proof of assumptions postulated by in-
. I
divldual hypotheses and based on observed available data, is, at the
'present time, possible and necessary. In this connection, Newell's work
[16J provides daily data on total ozone and wind at the AT100 and AT60
levels for the IGY period, for a station network situated at intermediate
latitudes such data can be and are used to calculate mean quarterly
hori~ontal macroturbulent ozone flow at the above levels. This flow
appears to be directed toward north at intermediate latitudes; it ex-
hibited a maximum during winter and spring, when it comprised 90% of
the total ozone transfer across the 50N circle. In his studies of relation-
ships governing macroturbulent stratosphere ozone diffusion, Newell used
data, previously mentioned in 4.2, on the propagation of tungsten-185 in
the lower stratosphere. Using supplemental arguments, Newell con-
cluded that ozone was propagated from its principal source (tropical strato,-
sphere) mainly by the macroturbulent ozone-rich air mass diffusion toward
the poles. There they descended simultaneously below the ozone photo-
chemical equilibrium level at the intermediate and high latitudes, particu-
larly during winter. The foregoing makes possible the establishment,of
an observed meridional ozone maximum in the lower stratosphere at high
latitudes during spring.
Similar quantitative analyses of existing qualitative ideas about
propagation processes in the stratosphere and the transfer of radioactive
debris into the troposphere, can be made only having a large volume of
data on the concentration of the debris in free atmosphere. The latter can
be determined with the aid of complex and expensive equipment, the use of
which is economically prohibitive in prolonged systematic daily determina-
tions over a wide station network. Therefore, relationships which govern
the global propagation of radioactive debris and the effect of different
meteorological factors on the latter. particularly on a scale smaller than
global should be studied by making systematic daily ozone concentration
determinations. over a wide and dense network of stations. However. data
thus used are of value only as complementary to a more detailed comparative
study of data on ozone and radioactive debris. derived from a more exten-
sive and detailed information complex than the one presented in this report.
It is also advisable to conduct simultaneous ozone and some radioisotope
concentration determinations at the same point under different meteoro-
logical and ge ographical conditions.
6.
C onclus ions
1. Data presented in this report indicated that meridional ozone and
radioactive debris distribution in spring and autumn in the middle and upper
tropospheres, and in the lower stratosphere layer adjacent to the tropopause
were in close agreement. Seasonal concentration changes in the above
- 203 -
-------
substances in the same layers s'howed similar agreement.
for higher stt:'atospheric layers are meagre to form a basis
conclusion. .
Exis ting data
for any reliable
2. Similarity between meridional ozone and radioactive debris
distributions in the atmospheric layers adjacent to the tropopause, indi-
cated that the air mass exchange through the tropopause occurred at an
approximately identical intensity at all extratropicallatltudes.
3. Differences in the physico-chemical properties of ozone (ga,s)
and radioactive aerosols reflected their different behavior in the lower,
I
troposphere layer which is washed by precipitation. However, data in-
,sufficiency on the tropospheric ozone co~centrations prevents the formation
of a reliable basis for a detailed explanation of the difference.
4. The difference observed in the meridional radioactive fission
products concentration distribution in spring and autumn in the middle
troposphere (maximum at high latitudes) and near the grol\nd (maximum
at ~ntermediate latitudes) reflected an intensive removal of radioactive
aerosols from the lower troposphere at the intermediate latitudes, which
occurred probably as a result of vertical turbulent transfer and air
descent in the subtropical anticyclone region.
5. Verification of the concept related to an increase in the alti~ude of
maximum vertical radioactive debris concentration distribution in a sta-
tionary stratospheric reservoir, with a decrease in latitude, must be
made by determining tropical stratospheric radioactivity at a height ex-
ceeding the 20-km level during a period free from nuclear weapons testing.
Bibliogr:.'1.phy
I .
I. f Y II( II II f. n. 3aKoiloMcpHOC1'II ropll30ll1'aJlbllOrO pacnpC.11I311Ka anlOc1I3~la1'rn3, M., 1958.
7. IDaJl.IIOaK1'lIlJllble 3arpSl311cIIIIII IJlleUI/ICll CpCALI. 1l0.1l pCl1. B. n. ill 13 C;1 0 II a
II C. 11. III II po K 0 B a. rOca1'O~11I3J1.a1', /I\., 1962.
8. B r e \V erA Thc transfcr of atmospheric ozonc into thc 1ropo~ph.erc ",\ICle-
orological factors influcncig thc transport and rcnlOw'a] of r;Hlioadi\'c
dcbris. WMO Tech. note N~ 43, \VJ\IO J\1! III. TP. 49. Gl'nC\\'a, 19GI.
- 204 -
-------
fl. D ii tls:c h H: J\\iltchwrte und ~\"clterhafle Schwankungen des almosphari-
schen Ozongehaltcs in \'crschiedenen Hohen iiber 1\rosa. 1\rchiv 1I\eteorol.
Geophys. Bioclimat., ser. A, Bd. 13, h 2. 1962.
10. Fallout program. Quart. summary rep., January I 1963, US 1\EC Health &
Safety Lad., HASL-132. .lI.oKY~leHT OOH. AlAC. 82/G/L.
II. fallout programm. Quart. summary rep., October I 1961, US AEC HASL-
115. .lI.OKYMeIlT OOH. AlAC. 82/G/L. 679.
12. fallout program. Quart. summary rep., December 30 1961, US AEC HASL-
117. .lI.oKplellT OOH. AlAC. 82/G/L. 737.
13. J u n gee. E. Global ozone budget and exchange between stratosphere and
troposphere: Tellus, vol. 14, No 4, 1962. .
14. L 0 c k h art L., Pat t e r son R., Sa u n d e r s A., B I a c k S. Fission pro-
duct radio activity in the air along the 80 th meridian (west) during 1960.
U. S. Naval Res. Lab. NRL Report. 5962, 1961. .lI.oKYMeIlT OOH. AlAC.
82/G/L. 736.
15. 11\ ate erG., God son \V. The vertical distribution of atmospheric ozone
over Canadian stations from umkehr observations. Quart. j. Roy. met. soc.,
vol. 86, NQ 370, 1960.
16. Newell R. The transport of trace substances in the atmosphere and their
implications from the general circulation of the stratosphere. Geofis. pura
e appl., vol. 49 (1961/11).
17. Ram a n at h an K., K u I k a r n i R. Mean meridional distributions of ozone
in different seasons calculated from umkehr observations and probable
vertical transport mechanisms. Quart. j. Roy. met. soc., vol. 86, NQ 368,
. 1960. . . .
18. S t ebb ins A. (ed) Second special report on high altitude sampling pro-
gram. US AEC report DASA-539 B, 1961, AlAC. 82/G/L. 741.
19. Stewart N., Osmond R., Crooks R., Fisher E. M..The world-wide
desposition of long-lived fission products from .nuclear test explosions,
Atom. Energy Res. Establ. Harwell, A. E. R. E. HP/R 2354, 1957. .lI.OKY'
MeHT OOH. AlAe. 82/G/R. 143.
RESULTS OF 1962 ATMOSPHERIC OZONE OBSERVATIONS IN OMSK
IN JUXTAPOSITION WITH SOME METEROLOGICAL ELEMENTS
L. A. Govorushkin
Recognizing the value of ozone investigations in the study of the
atmosphere and meteorological processes which occurred in it, the USSR
expanded in 1960, 1961 the network of ozonometric stations equipped with
monotypic::tl instruments. Regular total atmospheric ozone determina-
tions were initiated in Omsk in December of 1961, using the Gushchin
universal ozonometer. which consisted essentially of a filter-equipped
e1ectrophotometer. During December of 1961 and May of 1963 observations
the instrument was calibrated twice against Dobson's standard spectro-
photometer in Leningrad: once before shipment to Omsk in September of
1961, and the second time during June-July of 1962. II). both instances the
standardization lasted 15 days. The calibration factor changed only slight-
ly during operation (on the average by 3-4%) and.' therefore, ozone data
were regarded as sufficiently accurate for statistical analysis. Ozone ob-
servations based on the direct solar light sightings wel"e conducted hou:dy
- 205 -
-------
during one year over an interval of sun's heights from 10 to 58°. The d2.ta
were processed by a special ozone nomogram which made possible total
ozone calculation from 0.12 to 0.56 cm in terms of 1) solar ultraviolet
light intensities of two wavelengths A and Aland 2) sun's height.
The number of observation days was almost uniformly distributed
seasonally:
winter 24
s-p-iing 38--- -
summer 27
autumn 25
I
The number of monthly observation days avel'aged 12, although
the direct November solar light observations were conducted on only 4 days
due to 'unfavorab1e weather conditions. Ozone concentrations were deter-
mined only during the first 10 days of June, because the instrument was
sent to Leningrad for calibration. No ozone determinations were made
during the dark hours due to lack of night equipment.
Determination Results
-- ....------...--- - - - ---- - .-
- - - ~--- ----- - - .- -~..~.-
Table 1 presents mean monthly values of total atmospheric ozone
over Omsk, and graphs in Figure 1 show the cour se of annnal ozone con-
centration changes during 1962 and January-May of 1963.
Table 1
- -_. - --~------ .------------------..--.------
MONTHLY OZONe CONTENT OVER OHAK CITY IN eM IN 1962
. JAN. - . 0--. . . . 0.357 . -JiILy--;-;--; . -. . . 0.280'
FEBR. . 0 . 0 . . 0.390 AUGUST. 0 0 . . . 0.290
MAR CH . . . . 0 . 0.390 ::iEe'TEMB. . . 0 . . 0,272
APRIL. . . 0 .. .0,410 O'.;TOBER. . . . .0.305
MAY. '. . . . . 0,346 Nov. . . . . 0 . . 0.300 .
JUliE. . . .. 0.308 DEC. . . . . . . . 0.315
The .maximum ozone concentration occurred in April (spring) and
. the minimum in S.~ptember. A sharp decrease in atmospheric ozone con-
centration was noted in the beginning of April; during April to September
ozone concentration decreased by 340/0. The annual amplitude of total ozone
change's,; averaged 400/0 for one year. Of particular significance was the
fact that during January-May of 1963 ozone changes failed to corresf>ond
with ozone changes of the identical period during 1962 (Fig. 1). A signifi-
cantly large increase in ozone concentration was observed in.1963. and the
mean ozone maximum shifted from April to March (x = 0.410 and 0.:540 em,
correspondingly). The correct definition of the difference in mean annual,
ozone values and their evaluation of their systematic or well-defined charac-'
ter can be made only on the basis of data collected over a number of years.
- 206 -
I
-------
Fig. 1
'.1
- -- .- .--- - .-"-
- - ---- -- -- ---
CM
/ '
I \
I '
\
I ~
- "
-
/ /' \
~ .'
---
'- i--- ~ -'" V
- 0.50
:0.45
...'
Z
III
t; ; 0.40
o
0:
... I
~ ' 0.35
N' .
0.
, 0.30
0.25 I
II
1/1
IV
v
'.II
VII VIII IX
X
XI X II
-- - -- -
COURSE OF OZONE CONTENT OVER OMSK IN 1952 AND 1953
- ..-. .. - ---- -.- .--,..... -.----- - -.
Particularly interesting were the nonperiodic ozone concentration fluctua-
tions which appearl~d dul"ing one or several days (Fig. 2) on the general
background of annual ozone changes. Sharp ozone concentration changes
were noted in April - May during a period of maximum atmospheric ozone
concentration. The deviation (.6x) attained .30..40% of the mean annual
value.
Fig. 2
CM
0.45
"
"
:;;; 0.40
...
...
z
o
u 0.35
..,
'"
o
N
o. 0.30
/
/
0.2$
°
4
8
12
16
20
24
28
- . --- - - ..-
. -- -- - -- -----
COURSE OF OZONE CONTENT DURING APRil (IJ AND MAY (2)
A decrease in total atmospheric ozone in the summer and,pai-ticu1ar1y,
in autumn (September) signified a sharp reduction in the daily fluctuations
which amounted to only 10-15%. Thus. in 1962, ozone concentration over
Omsk was 0.457 cm on April 14, 0.365 cm on April 16, and 0.435 cm on
----- ---- -. - - -
- 207 -
-------
April 20. The mean annual ozone concentration was O. 330 cm; on May 1 it
was 0.328 cm; May 2 -0,428 cm; May 5 - 0.295 cm; September 23 - 0.296
c m; September 24-0.255 cm; a11,d September 25 - 0.285 cm. Sharp changes
in synoptic conditions are general1y typical at the intermediate latitudes of
the northern hemisphere, and it is quite .possible that there existed a defin-
ite relationship between the rapid daily ozone changes and the short-period
changes in the atmospheric circulation system. Using data of annual ob-
servations ozone concentration was compared with the synoptic conditions
consulting baric topography charts of the 300 -mb level. Despite the limited
volume of analyzed materials, the obtained results confirme4 the conclu-
sions of many authors concerning the above relationship. A particularly
high ozone content was noted west of the cyclone center, and the lowe!?t ozone
content was observed west of the anticyclone center. On 1 May of 1962
Omsk was situated under the western periphery of a high-altitude crest;
2 May - under the extreme eastern periphery of the crest; 16 April - under
the south-eastern portion of a high-altitude trough; and on 23 April - under.
the western portion of the trough. The corresponding ozone contents dur-
ing that time were: 0.328, 0.428, 0.365, and 0.449 cm,' respectively.
An identical ozone distribution in a cyclone and an anticyclone was mentioned
I
by Dobson in 1928.
The unexpected increase ~n ozone content coincided with an
occurrence of high-altitude cyclones and troughs, while decrease in ozone
content was observed at high-altitude crests and anticyclones. Thus, on 1
March of 1962 Omsk was under the central trough portion of a high-altitude
. cyclone where total ozone was 0.449 cm; on 22 March, Omsk.was under
the central portion of a high-altitude crest, and the ozone conent was 0.364
cm, A similar relationship between the baric field and ozone distribution
was disce rned quite clearly on several occasions, although deviations from
it may be encountered. This may be due largely to the type of synoptic
materials (charts).
In considering the relationship between short-term changes in
total ozone and wind direction in the lower stratosphere, some author~
expressed the opinion that ozone content increased with north winds.
For ve rificatlon purposes wind directions at different altitudes. and
particularly at the 100-mb level (approximately 16 km). where the air i
motion exerted a considerable effect on ozone changes were. invf.:stiga~ed
in juxtaposition w:ith the ozone contents during April. May. August .ind
September. The first two months were characterized by a sharp sea-
sonal change, while. September showed a slight change. Fig. 3 shows
changes in total ozone and in wind direction at ,the 100 -mb level for each.
day in.September. The mean daily wind direction was obtained from data
of aerological obser.vations. As can be seen f~om Fig. 3, the relation-
ship between wind direction at the 100-mb level over the observation point
- 208 -
-------
,
and changes in the ozone content was somewhat indefinite. The rise and
fall in the ozone content paralleled the north and south winds. Apart from
this, conclusion~ concerning ozone rise in the lower stratosphere with
ndrthern winds failed to agree with ozone distribution in the high-altitude
crests and troughs.
Fig. 3
---
x '10 -3CIII
400
-- - _W1rro_Jl1B£CT_LQt!.L.JtE_~~ AS F_O~LO~~ -- - -- - -- --
s s s s S S II N N ~ N N N I 'N N N N N N N N S S S S S S S
,- . EEl E E E E E
I
\
300
- .--
/~ t'-..
\ ~ :--.
J "--
",. ['.. r--- ......V -
'\~ ,;' ,;'
-- ,;'
200
I
5
10
15
20
25
. 30
- - -~ - .. - --
- - -- - _n.. - ----- -- --~ .
GRAPHIC COMPARISON OF OZONE CHANGES AND WIND DIRECTION AT 100 HB LEVEL
, Curve in Figure 3 shows that a southern wind was observed at the
lOO-mb level over Omsk on 6 September, 1962. At the samE: time ozone in-
creased from 0.252 cm on 3 September to 0.355 cm on 6 September, since
Omsk was under the eastern periphery of a high-altitude trough. Therefore,
the: conclusion that the northern wind in the stratosphere was accompanied
by ,an increase in total ozone appeared to be inaccurate. Daily changes in
oZbne content are of considerable significance. Daily, or more precisely,
diurnal changes in total atmospheric ozone in Omsk undergo changes of
great significance. Some increase in ozone content was observed during
morning and evening hours as compared with the midday values. The in-
crease amounted to an average of 0.020 cm with a maximum up, to 0.040 cm.
These changes were attributed at first to errors made in the determinations.
However, a detailed statistical analysis of all available materials indicated
that diurnal ozone variability was a true phenomenon resulting from insuffi-
cient allowance for the effect of atmospheric aerosol. However, a notice-
able change in ozone content values in the course of a day occurred in prac-
tice preventing the determination of its changes with sufficient accuracy.
~here ozone determinations were made by using the universal ozonometer,
total atmospheric ozone was calculated with the aid of a special ozone
nomogram built on the basis of the following well-known formula:
- -- -_.--_.-
--- ~- .---- --.
I - kl 10 - [ IU''). + m~). + ml a). I
).- ).,0' .
(1)
- 209 -
-------
v:here ~l - denotes aerosol mas's, and 0A - (coefficient of direct solar
l1ght wlth wavelength A) denotes attenuation by aerosols. Coefficient 0
was not taken into consideration when final calculations were made A
with the aid of the ozone nomogram. Allowance for aer,osol transparency
component should be made where filter-equipped instruments are used
which discriminated broad spectral intervals. '
The study of atmospheric ozone relationship to meteorological con-
ditions constitutes an important phase of the total problem. The present
author correlated total ozone with the following aerological data:
temperature, pressure, and tropopause height. Results are listed in
Table 2. Coefficients of correlation between ozone, on the one hand, and
temperature and pressure, on the other, were determined for altitudes
. of 6, 8, 10, 12, 14, and 16 km. Coefficient values, shown in parentheses?
were calculated for a small number of ozone data. Due to only few sunny
days (4 days) in November, correlation coefficients for that month were
not calcula.ted. Calculation results showed that close relationship existed
between ozone content and meteorological elements at different altitudes;
moreover, the correlation coefficient exhibited a pattern of well.:.defined
changes during the year. The most distinct correlation between ozone and
temperature and pressure W<;i.S observed in the summer and autumn months.
The distribution of negative ozone correlation with tropopause
height was more uniform, although a maximum correlation coefficient
r = 0.87) was observed also in the summer, in August. The relationship
between ozone and temperature and pressure was expressed weakly in
the spring. The foregoing indicated that the correlation coefficient ex-
hibited seasonal changes. An extremely close negative relationship was
observed between ozone and temperature at heights up to 10 km (r = 0.81).
The temperature correlation was positive in the higher layers of the lower
stratosphere. Moreover, the sign of the ozone-temperature correlation
coefficient changed at different seasons. The ozone-temperature correla-
tion wa's negative at all heights during May - June; it should be noted,
however, that an anomalous relationship was observed in the stratosphere
where the correlation coefficient changed sign frequently. The ozone-
pressure correlation was negative, and the general value was small.
The correlation coefficient at heights of 6 to 10 kmattainedonly - 0.06
in March. The highest negative ozone -pressure correlation (r = -0. 78)
was observed during autumn at the lO-km level.
In order to compare total ozone with pressure in greater det~il,
changes in ozone content were compared with pressure at the ground.
surface. A detailed analysis was made of the d1ata on ozone and pressure
changes at the' observation point during March, April, May, August, and
October, 1. e., for periods with well-defined types of correlation.
- 210 -
-------
Table 2
! COEFFICIENTS OF Ozons-TEMPERATURE, OZONS-P~ESSURE AND OZONE-TROPOPAUSE ALTITUDE CORRELATIONS AT
OMsK, 1962
I II III IV V VI VII VIII IX X XI XII
T6 -0,77 -0,67 -0,13 -0,13 -0,42 (-0,08) -0,42 -0,48 -0,66 - - -0,17
T8 -0,14 -:0,64 . -0,12 -:-0,21 -0,65 (-0,26) -0,42 ~O,61 -0,60 -0,81 - -0,19
T10 +0,13 .:...{),35 +0,22 --':0,35 -0,30 (-0,03) +0,.17 -0,31 -0,54 +0,63 - -0,04
T12 +0,54 0 +0,12 +0,48 -0,02 -0,34 '(-0,46) +0,28 +0,82 +0,51 +0,81 - +0,81
. TI4 +0,51 +0,04 +0,50 +0,31 ..:..0,57 .(-0,81) +0,29 +0,75 +0,60. +0,83 - +0,37
,0.TI6 +0,48 -0,15 +0,59 +0,49 -0,25 (-0,64) -0,09 +0,70 . +0,55 +0,74 - +0,20
,
.. -0,05 -0,05 :""0,41 -0,77 -0,44 -0,72 -0,23
P6 -0,31 -0,18 +0,31 (-0,19) -
Pa -0,01 -0,05 -0,06 -0,14 +0,18 (-0,62) -0,40 -0,65 -0,58 -0,80 - -0,21
Plo -0,13 -0,05 . -0,06 -0,20 +0,02 (-0,18) -0,40 -0,73 -0,57 -0,78 - -0,20
P12 -0,09 -0,04 -0,17 -0,15 -0,08 (-0,18) -0,38 -0,65 -0,58 -0,70 - -0,0
PI4 +0,16 -0,05 -0,42 -0,21 -0,19 (-0,50) -0,55 -0,39 -0.,46 -0,31 - -0,03
. PI6 +0,32 -0,03 -0,19 -0,04 -0,42 (-0,12) -0,33 -0,17 -0,36 -0,40 - -0,11
-. .0 ...
I
I HTp -0,56 -0,22 ..-0,56 . -o,3l +0,16 (-0,39) -0,30 -O,~9 -0,34 -0,86 - -0,56
f 0
- ------ - -
I-t
III
~ s::
..-4 ~ ~ J.1
U ~ ~ ,. s:: ~
'tjs::II)~O~
~J.14)~ON4)
'"' ::1
ro .... ,.0 .c: '"'
~'tjs::uEbO.
p.. lIS'''' III
p..]u-EI-t~1I)
III .... .... ... 2 .c: '"'
~II)HS::I~~
s:::> 0"" >
o ? ..... J.1 . I I
NO....~o~~
o..d 1I)..-4..d '"' u
II) ::1 ~ II) ::s '>
..-4 bO '"' II)
1II~::1,",]II)'tj
'0 « 0 ..... ~ s::
.... bO'tj us::'"' III
s:: .... s:: ..-4 ~ p......
.... ~ lIS III ~ s::
- '"' :> ~ ~
v, ..-4 ~ ? U ....
~ ."" s:: .... III s::
bO'tj,",~~,+--....II)~..-4
up..'"''"'..d I S::..-4
s::E~.8~]~N
o 0 II) U 0 ::s 0
~u~~:D0~
,",~bOlI)lII'"'..d
::1 '"' s:: .... ...... bO....
II) ~ lIS .... ~
II) :>.c: .«1 '"' ~ II)
., ? U II) .c: III
... ~ .... ~
I-t II) ~ III > s::
p.. '"' '"' .... .... -.-4 '"'
o :;j"''''' ~
~ .... II) ~ III II) ..-4
UUII)::::bO~..-4
III III " ,.0 ~ bO III
';::: '+-< ~.... s:: t:: E
::1~ p..~ ~~ II)
~o'tj~£(4)
'tj ..0 s:: II) . ..d
s:: '+-< III ~~]....
::1 0 .... > I-t '" ~
o II) s:: I-t III ~
'"' ~ ~ ::1 ~ II) '"'
bO>""U..-4t::::1
'+-< '"' s:: U 0 II)
o ::18o£i'tj'tl ~
.... U 0 ~ '" '"'
us::~.DII):;jp..
~ ~ s:: .... II) U
'+-< .. 0 '" ~ ... .~
'+-<""N..dl-t-'..c
~ ~o.... p..4:: ~
-------
----iii'
The distribution of the extremal
interesting relationship. Ozone
behind the ~ressureextrefu.a.
ozone and pressure values shows a very
content fluctuations lagg'ed by 2-3 days
1
'- ,
Fig. 4
-- -,-.- -- ~ - ~ - . --
- ._-- .----
- ----.-._-
r 'IO'~CM
Pnp 500'
300
~
I' '
f \ 1 \ '
I "\
2 I
\ f-L - J
t
'" I
" v
"
1020' 400
1000
"
, \
980 200
, ~
8 12' 16 20 t", 28 '
.A,nTr:- 1962
, I . I I j .. I I I \, I~. I I .
~ 8 12' 16 2024, 28
," 'AU~US1" 1962:,
\
---.clJffvii'r,rOZONe"coN:rENTi CU;'iE (2f'p~ESSUR'E
The above-inferences'should not be taken as final conclusions;
nevertheless,. they offer a basis for further studies on the relationship be-
I ' ' .
tween ozone and some meterdrological elements. Results of calculated'
correlation coefficients agreed well with data obtained at other stations
(Voeikovo. Tromso). The data coincidence refle<;:ts the existence ofa '
regular relationship between ozone and meteorological elements.
Bibliography
1. r y tq H H f. n. ° .D.BYX Ba>KHblX oc06eUHOCTliX 0301l0MerpllIJeCKIIX nplI60poB.
ATMoclpepllblA 030H. 113.D.. MfY, 1961.
2. f Y tq II H f. n. MeTO.D.' p3CIJeT3 06IUero cO.D.ep>KaHIIS! anlOclpepHoro 030H8
.D.JIS! npll60poB co cBerolpHJlbTpaMII. ATMoclpepHblH 030H. 113.D.. MfY, 1961.
3. f Y IU H H f. n. npe.D.B3pllTeJlbHble pe3YJlbT3TbI. 113MepeHIIH 06IUero co.o.ep>Ka-
HilS! 030H3 BO BpeMS! Mrr B CCCP. TPYAbi rro, Bbln. 105, ,1960.
4. f e T Ll, n. A TMoclpepHblii 03011. Y cneXH reolpH3HKH" 1934.
- 212 -
I ' ,
-------
CHEMICAL DETERMINATION OF GROUND LAYER OZONE
AT VOEIKOVO
P. F. S .nstov
Ozorie density determinations in the ground layer are based on the
following reaction:
(1) 2KI + 03 + HaO = Ia + 2KOH :i: Oa
(A)
The iodine is liberated into the air, the potassium iodide concentration is
reduced, and the pH is increased. Treadwell and Annaler [2J found that
the amount of iodine was equivalent to the amount of ozone only in neutral
solution. Oxides of nitrogen and other atmospheric oxidizers did not react
with such a solution of potassium iodide.
Several methods of chemical ozone determination in atmospheric
air were tested O,2J. In one methoc;1 the amount of ozone was calculated
on the basis of conductometric potassium iodide titration results before and
after air aspiration [3J.. .
,
The second method. which will be described in greater detail.
was based. on ozone density determination by difference between iodine
content in the potassium iodide before and after air aspiration. The iodine
is extracted with chloroform and its color intensity determined photocolor:-
imetrically. .
The following reagents are required for analysis:
1) 3 N sulfuric acid;
2) chloroform;
3) potas'sium iodide solution containing 0.1 mg iodine per mI.
prepared by dissolving 0.1307 of KI in 1 liter of water;
------------- -~ ---.--- - -----_.- -- -- --
-------- '-4Tph~;ph;te buffer (9.078 g KHa P04 / li, and 11.876 g
NaaHP04 . 2Ha 0). The pH is 6.98, if 40 ml of the KHa P04
and 60 ml of the NaaHP04. 2HaO .... are taken;
5) Fresh nitrososulfuric acid. prepared by dissolving 100. mg of
sodium nitrite in 20 ml 6f 3 N sulfuric add. The solution is
read"y for use after the liberation of nitrogen oxides is completed.
- 213 -
-------
Collection of air samples: a~pirate the air through an aspirator equipped
with a porous Schott filter #1; containing 15 m1 of. the potassium iodide
solution" and 2 m1 of the phosphate buffer at the rate of 5 li/min for 60 min.
Analys-is-. Place 10 m1 of the absorber fluid into a 50 ml cylinder, add
1 ml of the nitrososulfuric acid and 20 ml of chloroform; mix thoroughly.
After the chloroform had separated decant it into a colorimetric cup and
determine photocolorimetrically.
Construct a calibration curve for quantitative determip.ations.
Prepare a standard scale using the initial potassium iodide solution
(0.1307 g KI/li); place successively 2,' 5, 7, 10, .12 and 15 ml into a series
of test tubes and add to each tube 1 m1 of the nHrososulfuric acid, .0.5-2 m1
of the phosphate buffe r, and 2,0 ml of chloroform; mix vigorously, but care-
fully, decant the chloroform after it had completely separated and make
colorimetric determinations. . Final calculations "should be based on re-
suIts of duplicate or triplicate tests, usingf6rmula
--.--- -,._------,,- -.- ~~-... -----
D = Ig io - Ig i1 ,
(B)
where D is the optical density of the solution;
io is the galvanometer photocurrent reading of the
standard potassium iodide solution (more precisely)
1 m1 of 0.1307 g/li KIsolution;
i1 is the galvanometer photocurrent reading of the
standard scale solutions
According to the Lambert-Beer law a quantitative relationship existed
between the intensity of the inc'ldent light, trans mitted light, the substance
concentration, &IJd the solution layer thickness, expressed by the equa-
tion h = jo .10- , where h is the intensity of light transmitted through
the solution; jo ,is the intensity of incident light; € is the extinction coeffi-
cient which depended on the incident light wavelength, the nature of the
solute, and the solution temperature, and is independent of the concentra-
tion; c - is the concentration of the solute; 1- is the solution layer thick-
ness. The" photocurrent is proportional to the intensity of light incident
upon the photocell Jo =~. Substituting in the equation of optical density
Jo h
photocurrents proportional to the illumination, obtain"D = 19 10 -lgi1 =
€c1r 1. e. for a constant layer thickness of the absorbing solution 1 obtain
D = €C. . Accordingly, the difference in the logarithms of the phbtocurrent
strengths before and after light absorption is d~rectly proportional to the
concentration of the absorbing substance; this is the fundamental equation
of photocolorimetry. .
- 214 -
-------
The optical density data and the corresponding iodine concentrations
in standard solu~ions are plotted on graph paper. The amount of ozone,
equivalent to the iodine concentration in the' standards~(the iodine concen-
tration in standard solutions is multiplied by 0.189) is plotted' along the
ordinate. A line is drawn through the plotted points. A linear functional
relationship will thus be obtained within the concentration limits ~ccording
to Beer's law. The ozone content in the air aspirated through the solution
. 3
can be determined using the calibration curve, and later converted to 'Y /m .
However, the calibration curve can be used only when every step of the
procedure 'was conducted under identical standard conditions. . Light ab-
- sorption measurement must be made in cells identical with those used in
measuring light absorption by standard solutions,. usipg the same photo-
colorimeter.
In the third method ozone density determinations are made on the
basis of pH or pOH changes in a neutral potassium iodide solution.
As was pointed out previously, th'e reaction of ozone with potassium
iodide elicited a rise in the pH value in the absence of a buffer. In practice
it is inconvenient to express the acidity of solutions by numbers having
negative exponents, such as 10- 5, 10-12 g. ion/H. Therefore, the reaction
of aqueous solutions, indicating their acidity or alkalinity, is more co'nven-
tionally expressed not in terms of H or OH concentrations, but as their
logarithms to the base 10, with reversed signs. These carry the designa-
tions pH and pOH.
Thus pH = -lgaH+
and pOH = -lgaaIr
For very dilute electrolyte solutions, when aH+::: [H+ ] and aOIr '"
[OIrJ it may be considered that pH R: - log [H+ ]
-- -- .'
pH~-lg [H+],
pOH = - 19 [OH-] .
(G)
It has been demonstrated that in dilute neutral solutions, and in pure wate r
[H+ ] [aIr] = 10-3. 4 g. ion/HI (220 C); therefore, pH ,+ pOH = 14.00. The
lower'-the pH value; the highe:r the .activity of hydrogen ions and conse-
quently the higher th'e acidity of the solution.
A laboratory apparatus was assembled for the determination of pH
changes in the potassium iodide s'olution consisting of two polyethylene ab-
sorbers with perchloroviny1 filters and a glass electrode potentiometric pH
meter. The first absorber is used to remove the dust from the air sample
and to re~ove the more active gases. The second absorber is filled with 50
- 215 -
-------
m1 of a potassium iodide solution (5 x 1(j4N) and 5 ml of a potassium
hydroxide solution (5 x .l(r4~., The pH of the first solution is approximate-
ly 5.50, and of the second 6.85-6.90. .
The quantitative effect of pH changes on ozone concentration is
determined experimentally. Fifty m1 of the potassium iodide solution,
(5 xlO-4N), is taken and to it 2 -15 m1 of a potassium hydroxide solution
(5 x 1O-4N) is added gradually in 0.5 m1 portions. After each addition
the pH of the potassium iodide solution is determined. The concentration
of caustic potash must be reduced proportionately (e. g. , 50 ml)for all
additions to avoid dilution effects.
According to the above reaction 48 ozone' units cor:responded to 34
units of hydroxyl ions. Therefore, having determined the amount of
hydroxyl ions in 2, 3, 4, . ~. 15- m1 of the KOH for 50 ml of potassium
iodide and having determined the pH, it becomes possible to construct a
calibaration curve, with the amount of ozone in the analyzed volume of
air correspondtwg to tM ~ir volume shown on the ordinate, and the pH of
the potassium iodide sow.tion shown 'on the abscissa (Fig. 1). It can be
seen from the curve that the accuracy with which the zo~e will be de-
termined in the ground air layer depended on the accuracty of pH deter-
mination in the potassium i(jdi~e solution. The accuracy of domestic pH
meters does not ~xceed pH :I: 0.01, making the error in the potentiometric
ozone determination amOUl'lt f6 :I: 2 ;' 1m3. This method has been described
briefly, since it is still in the ~eve10pmental stage. .
Air mixtures conta4ning s mall amounts of ozone are difficult to
prepare; for this reason parallel ozone determinations by the three above
methods were conducted. Experimental results are shown in Table 1.
As can be seen from the Table, parallel determinations agreed satis-
factorily. Only samples 7, 8 and ',' showed low ozone 'density values.
This may have been due to the effect of city air, which under certain con-
ditions may become heavily polluted by gases.
Results of 128 determinations showed that average ozone density
in Voeikovo was 33 ,,1m3, .th'€ minimum 3-4 and the maximum 86 ;,/m3.
The data are available for the months of June, July, September and October.
Comparison of these ozone density values with the average diurnal tempera-
tures and atmospheric pressures did not indicate any connection between
them. The same 'data indicated that ozone density in the ground layer was
somewhat higher in the summer than in the fall, and higher during dr~
I
summers.. ,
Wind patterns in relation to ,ozone densities indicated acertain
density increase during westerly winds. This was observed during hot sum-
mer days at relative humidities not exceeding 70%, and wind velocities up
- 216 -
-------
to 2 m/sec. A diurnal ozone density during the same days with a weak
westerly wind fluctuation with a maximum at 10-11:00 A. M. were observed.
-- --..-.-------- ..~--._~. -. - -. ---
Fig. 1 OJy/",J
70
60
50
40
30
'20 I
10
6.5
/
,/
7,0
7.5pH '
--- -- ----------- ---- --- --~.u
CALIBRATED CURVE FOR OZONE DETERMINATION BY
THE pH 10NOMETRIC METHOD
Table 1
---- .
- ----~
-----------~------_. --- - ---
OZONE DENSITY DETERMINATIONS AT GROUND LEVEL BY DIFFERENT METHODS
. p:::'. ~-I METIIOD -
--_.~ -------- -. ...----- -- -.---
COif- COif- PHoT6-- ~._---'>-----
TEST TEST
DUCT~. COLORI- pH 10N~ DUCT~ COLORI- pH I ON~
No. METRIC METRIC No. METRIC
HETR I C HETRIC HETRIC
1 36 33 28 9 23 20 (15)
2 38 36 40 10 25 27 26
3 28 26 25 11 53 50 47
4 41 '43 39 12 43 40 45
5 '44 41 - 13 39 37 -
6 40' 38 - 14 42 40 -
7 19 21 (10) 15 34 36 37
8 17 19 (12) 16 18 19 19
This unusual increase in ozone may be connected with the urban effect, or
could be connected with. a westerly transfer. The latter is unlikely, since
in the -fall 'a weak westerly transfer and a stable atmospheric stratification
produced minimal ozone densities (Table 2). .
- 217 -
-------
Table 2
--------'----- _._- ..-- "-.---- --..
SOME OZONE DENSITY DATA AND HYDROGEN IN~EX CHANGES 8N THE GROUNO LEVEL I
AIR LAYER IN VILLAGE VOEIKOVO
- ------- u ----
OZONE CHANGES IN -----.-- ~--'---- .--. - ----.
,
DEN,TY I HYDROGEN SAMPLE TA'" WIND DIRECTION CLOUDINESS UNITS
y/J DON DNDEX ING TIME AND VELOCITY, AND FORMS
LlpH tV'SEC. .
48
60
50
48
76
63
81
43
42
44
68
64
57
62
80
20
53
69
17
34
47
42
12
34
26
23
27
70
24
76
35
32
31
72
22
5
67
36
21
7
23
16
44
14
23
16
6
8
4
10
13
+0,02
-0,10
+0,07
+0,10
+0,25
-0,08
+0,28
-0 15.
-0:08
+0,15
+0,11
+0,10
-0,21
+0,25
-0,13
+0,13
-003
+0:10
-o~05
+0,03
+0,03
-013
+0:17
+0,13
+0,15
+0,05
+0,07
+0,06
+0,14
0,0
-0,01
+0,84
+0,28
+0,18
+0,07
+0,15
-0 14
.. _.-.J
-0,84
+0,07
+0,01
+0,43
+0,14
+0,33
+0,11
+0,20
+0,17
-----------
S UtU1ER 1953
10-11 . -,-I
9-10 VI . ~,5
10--11 S --6-7
11-12 SE --7
13-14 SE -7
9-10 1
.12-13 ,'WSI'I -1
13-14 '11 --0 5
14-15 11 -1'
10-11 NW -5-6
11-12 NW -7
10-11 SE -3
12-13 ~-~ - -4
9-10 ,--I 5
. 11-}2 HI -5~
12--13 Sill" -3--4
10-11 \!I - -1
11-12 SE-'-2
10-11 HE -3
12-13 US\! -1
13-14 . rI -0 5
14-15 VI -0'5-1
16-16 8--- -2'
12~13 -sw -3
11-12 51! -:-2
12--13 SI:
13-14 VI; -2
14-15 517 --4
14--15 . S -2
- -- - --_.---+------.
'. FALL 1963
13-14 /iN'E -4 .
9--10 SE -6-7
10-11 . SW--5
11-12 ~E -4--5
12-13 8r-5
13-14 SE --5
14-15 SE -5
9--10 SE -7
10-11 SE --6
12-13 .S--5
13--14~ssw -4
14-15 .,.: SSW -4
- -- JQ-:-':11_- -- .' 'II -6-:-7
11-12- 3+7
13--14 3--5-6
14--15 3-3-4
9--10 3-1
10--11 3-1
11-12 3-1
12--13 3--1
13--14 3-1
14-15 3-1
- 218 -
Ac -1--2
o
Cb -5, STORMY
To 'DiticL.
Cb --7, STORMY
o
Cs -1-2 -
Cc-l
o
Cu-l--?
Cb -5, STORI1Y
CU -1-2'
Cu -2--3
o
Cu -4-5
Sc-7
Cc-I-2
Sc --4-5
Sc -3-4
o
Ci -1-2
Cc -1-2
C u --'2
Cu--3
Cu-2
Cu-3
Cu-l
Cc-4
Cu-I-2
Cu-l
Ac--7
As --10
Cu -Sc--lO.
Ns -10, RilN
Ns-lO,
Ns --10,
Ns -10
Ns --10
Ns -10
Cu fr.-lO
As-7-8
Sc-lO..
Sc-lO
Cu-7-8
As-4--5
As -10
As -10, FOG
As -10,
As -10,
As --'10,
As-lO,
-------
Ozone determinations were paralleled by measurements of atmo-
spheric air OH. After air samples were aspirated through distilled
water, the latter acquired a degree of acidity. Aerosols gave an alkaline
reaction, while gaseous air impurities, such as sulfur dioxide, hydrogen
chloride,' nitrogen oxides, etc., produced an acid reaction, except for
ammonia. The solubility of gaseous impurities in distilled water varied
and was difficult to control when absorption of vapors and gases by aerosols
took place in the atmosphere. This problem was further complicated by the
presence in atmospheric air of droplets of various sizes and chemical
composition. The droplets alone could have degrees of acidity, depending
on their composition and the solubility of their nucleating particles, as
well as ,on the acidity of the condensing moisture. Undoubtedly, an im-
portant part was played by meteorological conditions leading to condensation,
coagulation, aerosol capture, and adsorption of vapors and gases by liquid
and - solid aerosols. No direct connection was observed between atmo - .
. spheric acidity, and ozone density, though some relationship existed between
ozone density and some atmospheric state (Fig. 2)
Eig... 2.-
~pH '
/. . ,
.
. / . /
0.4 . 2
.
0.3
.
0.2. /.. .
3.
0.1 .I"~. . . .,,:
.... .
. .
...
o .
. 20 40 60 800JYINI
.
-0.' .
. .
.
.
-0.2
--------.-------- -- -
- ---_.----
RATIOS BETWEEN ATMOSPHERIC pH AND OZONE DENSITY
UNDER DIFfERENT METEOROLOGICAL CONDITIONS IN
. 1960-63 .
I-LIGHT WESTERLY WIND; STABilE ATMOSPHERIC
STRATifiCATION (FALL Of 1963) 2-LAYER FORM-
ED CLOUDINESS; 3-CUMULUS CLOUDS OF GOOD WE..
THER
In conclusion Table 2 is pre-
sented which contains data on ozone
densities and atmospheric pH changes
in the ground layer of village Voeikovo.
It is evident from the Table that some
negligible changes in ozone density
were observed prior to a storm, during
a storm, and sometimes in the beginning
of rainfall in the fall, which originated
from altostratus and cumulostratus clouds.
(Bibliography next page)
- 219 -
-------
Bibliography
. - - ----
+- -~,-- - -~.. __0.__- +
-.-- - -----
1. T e i c her t F., War m b t W. Ozonuntersuchungen am meteorologischen
Observatorium Wahnsdorf. Abhand!. d. Meteoro!. und Hydro!. Dienstens,
d. D.D.R., N234, Bd V, 1955. .
2. 3 B e p e B K. C. Me-rO.1.IlKa onpe.l!e.~eHIUI. 030Ha 8 aTMoc. 2\\eTO.l!IIKa Onpe.l!eMHIlH KOHQCHTpaQIIII 030Ha 8 npll3eM-
110M B03AYXC. TpYJl.bI ffO, Bbln. 134, 1962.
4. AnlOcHlfe aTMoc
-------
However, the a.uthors
for low or high ozone
infrequent.
indicated that it was difficult to construct curves
values because days with such ozone contents were
The. above method for the calibration of zenithal observations was
modified by the GGO Ozonometric Group in its application to total ozone
determinations by the universal ozonometer. Total ozone determined by
the universal ozonometer at full solar illumina~ion is computed finally.
with the aid. of a special nomogram [1]. The nomogram is constructed
according to Bouguer's law for direct solar light with an allo"\\9.nce for
spectral response of the instrument and photocell. Therefore, it appear-
ed more expedient to .determine ozone from zenithal observations by tind-
ing the calibration factor K for zenithal observations, than to construct
. . ------..- - z
new nomograms.
Logic indicates that the empirically determined K reflected the
z.. .
difference between zenith and directly scattered light witnln the spectral
observation interval, where solar and zenithal observations were con-
ducted over a short period of time when total ozone content and instrument
parameters remained constant. The refore, K is in effect the ratio of
ratios of the first and second filters used in the s51ar and zenithal obse rva-
tions.
K - (h/Ja)sol
z - (h /Ja grnd
(1)
The ratio J1/J2 varied nonuniformly with changes in the suns height,
so that K = f (e). In order to determine K -, it is only necessary to con-
duct a nufuber of paired observations for dl1ferent solar heights. I:Jowever,
possibilities of conducting such observations, particularly during a clouded
zenith sky, are greatly limited. Therefore, it is assumed that total ozon'e
remained constant during the day, and Kz was determined from the daily
direct and zenithal o])servations made several hours apart. . Insofar as
Kz determinations are made on the basis of observations conducted at dif-
ferent instrument temperatures, equation (1) should include the instrument
calibration factor KT' which is a function of the instrument temperature,
in order to eliminate temperature -dependent variations in the trans mis sion
coefficient of the light filters [1J.
K
z
= (J1/J2K,.)sol
(J1/J2 K,.rg;nd
(2).
In the present instance it appeared rational to take as the paired
solar observation any mean ratio which corresponded to the mean diurnal
ozone content and the sun's height at which zenithal observations were
- 221 -
-------
carried out, rather than to take the ratio shown in the numerator in
Taken from the nomogram, such a mean ratio can be (Jl /J2 )nom.
equation (2) can be written as follows:
(J1 /J2 )nom
~
J2 grnd
(2)
Thus,
~.
K =
z
. '(3)
Thus, if the solar assembly of the universal ozonometer was cali-
. brated against a standard, 1. e., if the instrument calibration factor KT
was determined as a function of temperature, the zenithal assembly sh.ould
be calibrated at a station as a function of the station's latitude and weathe r
conditions over 3-6 months. The calibration procedure of the zenithal
assembly at an ozonometric station was as follows: Three ozone deter-
minations were made first, following which hourly zenith sky observations
were made whether the zenith sky was clear or cloudy. When direct ob-
servations could not be made zenithal observations only were carried out.
The.sun's height at the time of zenithal observations was determined from
Nabokov's nomogram. Errors introduced by the assumption that total
ozone remained constant during-the day were evaluated by calculating Kz
with the aid of mean hourly a.nd daily ozone values. If time of zenithal
and solar observations differed by not more than 15 minutes,K in (3)
. . z
was determined fro.m the mean hourly ozone value and the sun's zenith.
In the case of observations with intervals greater than 15 min. ('-;lP to
several hours), mean diurnal total ozone was used. Calibration factor K
. was determined separately for three graduations of cloudiness (lower,' z
intermediate, and upper) and the clear zenith. Values of Kz are pre-
sented in Fig. 1 as functions of solar height for a) clear zenith, b) upper
cloudiness, and c) lower cloudines.s. These were compiled from observa-
tions made at the Kiev ozonometric station during June-December, 1963.
It can be seen from Fig. 1, that dispersion of .Kz values deter-
mined from mean hourly and diurnal ozone values, was identical.
Maximum deviation from the mean was ::1:15% and ::1:10% for a cloudy and
clear zenith, respectively, and the accuracy in determining th.e mean
K value depended on the number of observations. Considerable data
. dGpersion for a clouded zenith indicated that accuracy in determining
Kz depended, (apart from all other factors which determined Kz for a
clear zenith) on the type of cloud gradation of each zenithal observation.
This error will obviously be reduced in the future, as more experience
with zenithal observations is accumulated.
The lunar assembly of the universal ozonometer was c'alibrated
by the same method as the zenithal assembly. It was assumed that total
ozone remained constant during the day when the solar and lunar observa-
tions were made.
- 222 -
-------
Fig. 1
- . -- . - .
0.8
0,6
O,~
Q2
10
a)
..".... '.' ,',
.
" ..
.. " ..
...-
6)
... '. .~r ,
.. ..
"~ " .:
-=--.
8) 1
. 2
.
..":' ..' .. ~ ... -
.. .',".
.~ ", ..."
-
..
15
55
60
;!
0.4
0.2
0.6
0,4
0.2
20 }~--}{J_.. _J~--, _40. ""~.. Sp
Solar Elevation
STANO-AR-D-I'ZAT I ONe GRADiiATI"ONT COEFFIC-IENTS"OF- THE U"NIVERSA'L ~ZONOMETER
No.4 ZENITHAL INSTALLATION AT STATION KIEV.
A - FOR CLEAR ZENITH; 8 - FOR "IGH CLOUDINESS; C - FOR LOW CLOUDINESS
. I - FOR COMPUTED AVERAGE HOURLY OZONE VALUES; 2 - FOR COMPUTED AVER-
AGE DAILY OZONE VALUES
The ratio (Jl /J2) was determined from the mean daily ozone content.
and the moon's hJ\gWt for each lunar observation by the ozone nomogram.
The ratio (Jl) was determined directly from lunar observations
(J2 K,.) lunar
As in the case of zenithal observations. the calibration factor of the lunar
assembly was as follows:
K
z
=
(Jl)
(J;fnom
(J 1. 'K )
~ ")lunar
(4)
The lunar assembly calibration and the subsequent total ozone de-
termination from lunar observations were made by the same solar ozone
nomogram.
Lunar light differed from solar light with respect to its spectral
composition. a factor which had a bearing on the value of Kl' The differ-
ence is a function, of the lunar wavelength reflection coefficient. Lunar
light measurements were made by a photomultiplier. Laboratory deter-
minations of spectral response of instruments equipped with a 1) photocell
and 2) photomultiplier. showed the following: Divergence of relative spec-
- 223 -
-------
tral response curves of an instrument was negligible in either case; this
was understandable, since the operational photocell and photomultiplier
had identical cesium-antimonide photocathodes, and identical light filters
, were used in the solar and lunar equipment. Therefore, the second factor,
which determined Kl was represented by the small difference in the rela-
tive spectral responses of the photocell and photomultiplier.
Fig. 2
--_._-----.~._-- -.- _._---~---
. - ...- ---- -- ----" -.
. -.-------.--.---.- -- - -
- --,'. ----
Kn
1,2 .
...
. .
1,0 ~.
. .
0,8
20
. ..:--. ~. e.
. . . . .~. :. o. . I . . .: .
. .. ... ~ .... . .~. --..-~ ~..-: eo ., :
~.-.-. . .
25 30 35
-- ---- ------.--..------.--. --.
, LUNAR ELEVATION I
- STANDARDIZAT I O'N-(GRADUATI O;'-,-COEFFICIENY-S-OF-TH Euiii VERSAL -OZONOMETER'
No.4 LUNAR INSTALLATION AT STATION KIEV.
40
45.
The foregoing indicates that the two factors which determined Kl were'in-
dependent of the moon IS height. Consequently Kl was also independent of
the moonts height. Figure 2 shows Kl of No.4 universal ozonometer. The
calibration was carried out by the staff of the Kiev ozonometric station
during June-December, 1963. Figure 2 clearly indicates that Kl was in-
dependent of the moon's height. The mean value of K1, which can be used
in the subsequent calculation of total ozone, can be determined from either,
the graph (the straight line in Fig. 2 corresponds to the mean value of Kl)
. or from a table containing calibration data. In this case, the maximum
point dispersion was :1::15%, and the accuracy of determining the mean value
of K depended on a number of observations, which was also true of zenithal
assembly calibration. Zenithal and lunar observation data were processed
in the same m~nner as were data of solar observations [1J except that in,
the first case ozone was determined from the solar height ,or the moon
height by a nomogram, and from the readout ratio of the first and second
filters multiplied by KT and Kz or K1, respectively.
The proposed method of calibrating zenithal and lunar observa-
tions is essentially an empirical one. Some of the assu,mptions on which
it is based lack verification. Therefore, the accuracy of determining,
. total ozone determination on the basis of zenithal ,and lunar obse rvations
is below the accuracy of solar observations. The use of these data must
be approached more cautious,ly, especially in research studies., Never-
theless, this method can be used advantageously in condticting continuous
observations of total atmospheric ozone. '
- 224 -
-------
Bibliography.
--. - -.;------ --------.._--~------ -- ------ - - - ,....
. .
1. f Y II! H H f. H. I1CCJIe,llOBaHHe aTMoc4>epHoro 030Ha. fHJlpOMeTeOH3JlaT, JI.,
1963. .
2. Do b son G. M. B. and Nor man d C. Dete~mination of constant etc used
in the calculation of the amount of ozone from spectrophotometr measure-
ments and an analysis of the accuracy of the results. Annals of the IGY.
Pergamon Press. 1958.
STRUCTURAL OZONE MOLECULE MODELS
O. M. Rozental'
A clear concept of the ozone molecule structure might advance the
solution of the atmospheric ozone problem. the different aspects of which
are under discussion at the present conference.
Four configurations of atomic centers had been proposed in the
past for the ozone molecule, each of a different symmetrical type. The
inconsistencies of each proposed model are brought out in this report and
proof is offered in favor of an absolutely nonsymmetric model, the kind of
which had not been considered before.
The linear type of symmetrical model exhibited two basic auto
vibration frequencies. However, ozone spectrum shows the presence
of some binary harmonies of the basic activity frequency which is in-
consistent with the concept of a linear symmetrical model.
V. N. Kondrat'ev and A. V. Yakovleva proposed a linear nonsym-
metric model. The spectrf,l analysis performed by them disclosed two
longitudinal vibrations of,," 3/1 frequency. The rotational structure of
basic frequency was in contradiction with such a molecular model. Experi-
ence in electron diffraction also supported the concept of curved molecular
ozone configuration even though the quantitative evaluation of its parameters
was inconclusive.
In the case of an equilateral. triangle (symmetrical vertex) only one
norrpal.auto vibration could be manifest in: the infrared spectral region.
However. the concept of infrared ozone spectrum can not be based on the
above assumption only.
- 225 -
-------
. The isosceles triangle model is characterized by three fundamental
. frequencies, which, according to Sutherland and Penney, are as follows:
V1 = .1043 cm-l (doublet); V'J; = 710 cm-l (doublet); and V:3 = 1740 cm-l
(doublet), where Vl, and Va, and V:3 are the frequencies. However, it is
not clear in this case, why the intensity of a line with a V:3 frequency should
be of considerably lesser value than of a line with frequency Vl,. A converse
relationship would be the natural one.
Reactions of the type 03 = Oa + 0, where O. Oa. and 03 are oxygen
atoms, oxygen molecules, and ozone molecules, respectively occurred easily,
which calls for an assumption that the vertex angle of an isosceles triangle
must be acute and that the third atom only distorted the oxygen molecule.
However" experiments with rotational levels and transitions of a non-
linear triatomic molecule (asymmetric top) yielded a ve rtex angle of 12 ~ .
A similar angle value was yielded by an X-ray structural, analysis.
OI3}
The low ozone absorption co-
efficient in the infrared and optical
, ,
regions also supported the inconsist-
ency of nonlinear symmetrical models.:
Fig. 1
- --- - - ..~- - . --.
s,
-."---- -- --- . --.
: MOLECUI.E OF 03
O/2}
Normal oscillations of a non-
symmetrical molecule (Fig. 1) are cal-
culated as shown below. Calculations
were made within a system of intrinsic
coordinates: variations of valence
lengths and angles q. Potential energy
V can be expressed in terms of normal
force constances k.. as follows:
lJ
o III
v=+ ~k/jqlqj'
I}
(1)
Determination of the dynamic constants in terms ,of normal oscilld-
tions must take into consideration kinematic interaction.
- ~ - --- -- - -
'1 .
T=T ~lljqIJ'
I}
(2)
where T - kentic energy; liJ' - coeffiCients; q..
, .. ~
with respect to time. and the secular equation,
following form must be solved:
coordinat,e value derivatives
which iri this case has the
- 226 -
-------
kll A - all
k'Jt ). - a21
, k3.L)~ - a31
(>.. WAVE LENGTH)
kI2), - a12
k22A - a22
kJ2)' - a32
k13)' - al3
k23A - a23
k33A - a33
(3)
where>.. is a wavelength.
I Cross and Van Fleck used such a method in determining normal
vibrations of ethyl chloride (C2H6Cl). They considered an imaginary system
I '
of:three material points which simulated the CH3 J CH:a J and Cl groups.
The problem was solved on the basis of assumed valence forces, 1. e. J a
substantial part of nonvalence interactions was disregarded.
The present author utilized a more highly developed r:nethod proposed
byl Soviet scientists for the direct determination of kinematic interactions.
The kinematic coefficients were determined from Yel'yashevich's tables
for inertial interactions of bonds and angles.
Where, for the sake of simplicity, use was made of the hypothesis
of potential forces, and assuming that
;~-~---'--- ' --.- - .-
" V =--," \ k..q.q. np" l=)
2 .~ 'I I 1 '
; . .ii .
(4)
the following motion equations can be derived:
--q'l --=--~ [(2kl + k4) ~I +' (k2 + ks) q2 cas 03 + (k3 + k6) q3 cas 02],
q2 . - ~ [ (2k2 + ks) q2 + (k1 + k4)ql coS 03 + (k3 + k6) q3 COSOI] , .
- 1 [
q3=-m (2k3+k6)q3+(kl+k4)qlCOS02+
+ (k2 + ks) q2 cas °1] J.
q4 = - .~ [i2 sin 01 k2 q2' + ;3 sin 01 k3 q3 +
+ (S24S3 - S22S3 cas "(1) k4 q4] ,
qs = - ~ [-Sl sin 02 k1 q 1 + -J. sin 02 k3 q3 +
mi' .4)3
+ (.SI4S3 - SI2S3 cas (2) k5 95] ,
- 1[1 ;ak +1 .ak +'
q6 = - - -s sm v3 I ql -s sm va 2 q3
,m I 2 .
+ ( SI4S2 - Sl2S2 COS (3) k6Q6]'
(5)
'.
- 227 -
-------
where' ql, q2, and q3 - variations of the valence lengths Sl, S2' and S3'
respectively; q4' q5, and qs - variations of the valence angles e1, 82
and 83, respectively; kl, ka, and k3 - valence bond force constants;
1~, k6. and ks - valence angle force constants. The system of equations
takes into consideration a considerable part of interactions. The remain-
- ing terms had to be discarded for the sake of simplicity and ease of calcu-
lation. .
The solutions assume the following forms for normal vibrations:
ql
- 2. -
- -w ql, q2 ~
3
- W q2; .. . . .; qa
- 2
= '- w qs
1) trigonometric relationships in a triangle;
2) if kl is the only coefficient in a homopolar ozone molecule,
it follows from chemical expressions that:
k1 » k2, ~ and 1<4 » k 6, 5 [sic] *
3) on the basis of an assumed interaction between an almost
undistorted symmetrical ozone molecule and the third ozone atom, it
was calculated that the third atom became equalized on th-e circumference
- ..-.- --.
2-+ 2 .s)
Xl X2=T'
(6)
where Xl and x:a - cartesian coordinates of the third atom in the molecule's
plane with the origin at the center of symmetry of the oxygen molecule.
Fig. 2
- -.- --. ---" -- .-- .
~.
NORMAL ViBRATIONS OF AN ASYMMETRIC MOLECULE
The interpretation offered by Sutherland and ~enney holds for the'
three fundamental active frequencies of a linear nonsymmetrical model.
This results in reasonable values of force constants and exempts the model
*should read 1<4 » k6, S
- 228 -
-------
. from the previous contradictions. The low probability and instability of
such a configuration determine ozone ability to yield the third 'atom
under excitation, when the system r s configuration was likely to resemble
an isosceles triangle.
Instead of undertaking a cumbersome solution of a system of motion
equations, the results are here shown graphically in Fig. 2 in the actual
form of normal stretching vibrations of an ozone molecule, and the table
below indicates its parameters:
P1A~.~.. . ~-' . . . . . . . 1,50+0,1
61 DEGREES. . ., 90+5
63 DEGREES' . . . . . . . 120:£7
Further proof of the acceptability of a nonsymmetric model amounts
to a strictly numerical calculation procedure which involves 1) analysis of
the infrared ozone spectrum, 2) study of the energy levels in the case of
an assumed motion of the third atom, and 3) accurate determination of
dynamic coefficients.
A TMOSPHERIC OZONE AND ITS EFFECT ON SOME
VEGETATION SPECIES
G. P. Gushchin
It has been established that low ozone concentrations in the air were
beneficial to the human organism as a mild stimulator of oxidation process-
es. In many instances, ozone serves as a destroyer of many harmful
bacteria suspended in the atmosphere and in other media. On the other
hand, ozone is harmful to man in high concentrations.
oio~e--~-~~~~~tr1.tion~~ise~l1ari)iy ~ith altitu'de, and attain a maxi-
~um at approximately 40-45 km. Accordingly, harmful ozone coocentra-
tions can be expected in the atmosphere at certain heights. This is of
- 229 -
-------
particular importance to present-day and future aviation, in view of the
fact that the flight ceiling of passenger planes continues to rise and that
the external air is used for the passengers and the crew. .
The purpose of this article is to show at what altitude s in the
atmosphere ozone concentrations become unsafe to man.
. The maximum safe ozone concentration by volume for man and.
certain materials was determined as 0.1-1. 0\ cm303 /M3. It was reported
. '3/' 3 3 3
that ozone was perceived by odor at 0.002 cm . 03/M. At 300 cm 03/M
ozone caused irritation of the eye and of respiratory organs. Death oc-
curred after 2-h;r-s. exposure,to 400-10,000 cm30s/M3.
No data on ve rtica1 ozone distribution in terms of units of volume
. ,
concentration were found in the scientific literature on this subject. Nor-
mally, such data are expressed in density units (cmOs/km) or as units. of
. partial pressure (~mb).
The following formula is used for the conversion of ozone density
Ps into ozone mass concentration.rs [1]:
-- u_. .._.m- . .- -"" ".. - .-" ,,- 3-- .. ). '...
(10- CM 03
i r3=(-M~G03_)=006147 P3( '; T("K).,
. G " . ( M9 )
(1 )
The following formula [lJ is used for the conversion of mass ozone
concentration into volume concentration'
.>-......~-_. ~ .-_.
N (01303) - 0 603 ("KG 03)
3 M3 -. r3 r .
(2)
In trying to determine ozone volume concentration at different at-
mospheric altitudes the present a,uthor used average data on vertical ozone
distribution over Canada as reported by Meteer and Godson [5J.' Having'
converted their data into ozone volume concentration (cm3/m3) for two
seasons (winte;'and summer) of 1957 - 1959 'this writ~r prepared the
following table: m_.~=~-. ,_._._~::..-:;:=-------:.=:.=::::":,===:=:=c'-::'('~M303). :':." .CC'. , .-' .,,'
Table 1 . ATMOSPHERIC OZONE VOLUME CONCENTRATION. --;a:-, OVER CANADA
. IN 1957 - 1959
- '. - --
. .
-LAYER IN KM
----- .. 0-616-12 12-18118-24124-30 30-361.36-42'
SEASON 42-48148-54
.. .- j
WI NTER 0,043 0,16 1,4 3,1 6,6 9,3 11,0 13,3 11,1
_._-- --- - :
SUMMER 0,058 0,13 0,60 2,0 5,5 10,0 9,3 10,6 7,0
- 230 -
-------
Data in Table 1 show that the lower limit of maximum safe ozone
concentration (0.1 cm" /m3) appeared in the 6-1Z-km layer, and the upper
limit was found in the 1Z-l8-km layer during winter and in the 18-Z4-km,
layer during summer. Ozone concentrations ten times in excess of the
safe maximum were found in the 30-36-km layer. In some observed in-
stknces ozone volume concentrations were Z-3 times in excess of the
average at 1O-Z0 km. The foregoing indicates that at 10 km and higher
it becomes necessary to apply special measures for the protection of the
passengers and crew from harmful ozone effects, particularly during pro-
longed flights. Such passenger and crew protection can be attained by
briefly heating the taken-in external air to 300 -400° ~ by appropriate
m~ans, since ozone as such decomposed at 150-300°. The air can then
bel cooled and supplied into the cabin without danger to the passengers
. or: crew. It should be pointed out, however~ that it is necessary to continue
further research on the determination of maximum safe ozone concentra-
I
tiqns in the atmosphere. All above considerations concerning harmful
oz:one concentrations are based on data presented in [3 J1 .
Based on results of many ozone concentration determinations [zJ it can
be concluded that unsafe ozone concentrations are not normally encountered
in the lower and middle tropo~pheres.
It has been found that ozone had a decomposing effect on rubber .[ 4].
Accordingly, it can be expected that rubber covers of radiosondes, rubber
sea1s~ rubber tires of high-altitude aircraft, and other rubber materials
should undergo gradual decomposition by ozone. It was shown in [4J that
the height of rubber covered radiosondes ascent depended on atmospheric
content. The height at which the. rubber casings rupture was on the aver-
age, maximal during autumn and minimal during spring, which is at
variance with seasonal ozone variations.
1 In L. C. McCabets article which was published in a volume entitled
"Pollution of Atmospheric Air" (World Health Organization, Geneva,
196Z), data are adduced for the critically-safe ozone concentrations in
Los Angeles. As indicated by the article, unsafe ozone concentrations oc-
curred when 1. 5 parts of Os per million are present. Assuming that we'
are speaking in terms of mass ozone concentration, we see, after con- .
verting it to volume concentration that it corresponded to 0.9 cm./m3 [sic J*.
This value is similar to the one shown in. [3].
V. N. Uzhovts monograph entitled "Dust Prevention in Industry"
(Goskhimizdat, M., 196Z) provides a table of the critically-safe concen-
trations of toxic gases and vapors in the air in industrial installations. For
ozone, this value is 0.000 mg/l or O. 05 cm~ /m3, i.'e'., somewhat below
the value shown in [3 J. ' .
*should read O. 9 ch:1~ /m~ [B.S. L. J
- Z31 -
,I
I
if
!
-------
The mean height atwhich the rubber covers broke down decreased with.
higher latitudes', ~hich is also in converse correspondence with the
. latitudinal variation in total ozone. .
However, .it should be pointed out that various rub'Der types were
affected differently by ozone. There exist relatively "ozone resistant"
types of rubber used predominantly in locations with an incre.ased ozone
concentration. I~elatively high ozone concentrations unfavQxably affect-
ed some vegetation forms [6 J.
. .,
Under fixed meteorological and topographical conditions ozone.
concentration in the atmospheric boundary layer can occasionally in-
crease over certain locations 10-20 fold. The con~entration increase oc-
. curred basically as the effect of local ozone sources generated by chemi-
cal interaction among different industrial contaminants, such as auto-
mobile exhausts, factory smoke" oil and gas combustion residues, etc.
Such sharp ozone concentration increases damaged some materials
considerably; cer~al grain and other plants were heavily da'maged, atmo-
spheric visibility decreased, and harmful physiological effects were in-
tensified [6 J . .
Some type~;of tobacco leaves were sensitive .to the effects of
. ozone and could serve as indicators of ozone presence. Tobacco leaves
exposed t6 ozone effect suffer damage in the form of clearly noticeable
spots. Damages caused to tobacco leaves by ozone was recorded at the
Agricultural Mechanization StatiOl'i near Washington, D. C. during 1952-1961
. [6J. Damage which some tobacco leaves sustained in t~e field was re- .
~ 3/ 3
produced under laboratory conditions by 6'2 hours exposure. to -.25 cm m
ozone concentratiop.s.. The problem of atmospheric ozone effect on man,
.animals, and plants had not been fully investigated. Th~ study of this
problem should b~ .continued on a more intensive and broader research o.
basis. .
Bibliography
. ------....
.--.. - ... -- - .-- - .--"-
.._-~--- ------ - -'---- ------.
I. r y 111:." H r. n. J1cc.'e)10BaHlIe aTMOc
-------
RESOLUTIONS OF THE THIRD INTERDEPARTMENTAL SCIENTIFIC
CONFERENCE ON ATMOSPHERIC OZONE HELD ON 2l-MAY. 1963
AT THE GGO. LENINGRAD
. The:,e;;nference :recorded considerable progress in atmospheric'
ozone investigations since the Second Interdepartmental Conference of 1959.
The USSR ozonometric network was expanded from 11 to 20 stations. A
considerable volume of important data on atmospheric ozone has been col..
lected at the Soviet and global ozonometric station networks. Materials
collected in the USSR by means of aeroplanes and ocean ships proved of
particular value. Work of this type was enhanced by the development of
more efficient aeroplane -borne monotypica1 network instruments. Analysis
of ozone data showed that atmospheric ozone was one of the important mete-
0rological factors which determined atmospheric temperature conditions and
circulation.' It was established that horizontal and vertical ozone distribu-
tion and turbulence wer-e closely related. For this reason, ozone data
yielded valuable information on stratospheric turbulence. Authors of
conference reports advanced the concept that ozone distribution .was a I
function of vertical atmospheric motion.
Results of many USSR research investigations had shown that there
existed a close relationship between atmospheric ozone and jet streams
which included turbulent regions of aircraft bumping. Analysis of the
,Soviet network data established that a noticeable increase in total ozone
occurred in the eastern regions of the USSR during winter and spring
periods, suggesting the existence of the longitudinal ozone changes over,
the USSR territory. Data we re obtained on the relationship between me:d-
dional distribution of and annual variations in mean monthly ozone con-
tent and certain radioisotope products of nuclear explosions in the upper
troposphere and lower stratosphere.
, Lack of systematic data on vertical ozone distribution constituted a
shortcoming' of atmospheric ozone investigations in the USSR and abroad. '
Determination of vertical ozone distribution by the available instruments,
yielded considerable errors; nevertheless data on vertical ozone distribu-
tion are extremely valuable in investigating atmospheric turbulence, heat
balance, and circulation, which pointed to the acute need to develop re-
liable monotype network instruments for the determination of vertical
ozone distribution in the USSR. Atmospheric ozone investigations are
greatly enhanced by artificial earth satellites which make possible to
obtain data over extensive ocean regions where ozone observations are not
being conducted at the present time.
'. The Conference resolved to ask the GUGMS,r-) to organize ozone ob-
servations in the USSR with the aid of artificial earth' satellites.
(1) Gla~no~uT.ip;a~l;nie Gidrom~teorologicheskoi Sluzhby, SSSR
(Main Administration of the Hydrometeorological Service of USSR)
- 233 -
-------
Suchobservations were conducted in the USA, and plans are presently
in progress to conduct similar observations on a large scale in the USSR
and in England.
The Conference expressed the opinion that the~ Sovietozonometric
network should work in intimate cooperation with the global network.
In this connection, the GUGMS was instructed to send a Soviet representa',
tive with an ozonometer to Arose (SWitzerland) or to Oxford (England) whe re
. standard ozone spectrophotometers were available.
. In view of the reorganized scientific program at the Hydrometeoro-
logical SerVice Institutes, it was recommended that provision be made for
Government Geophysical Observatory and TsAO to include in their scientif-
ic-research programs special studies on atmospheric ozone during 1964-
1970. It should he noted in this connection that up to the present no ozono-
metric laboratories existed in'the USSR,. seriously inhibiting the deve1op-
. ment of atmospheric ozone research.
. .
. . ".
The Conference heard approximately 20 reports on different aspects
. of atmospheric ozone. Members of the following organizations and insti-
tutes attended the Conference: GGO, TsAO, MGU, LGU, lPG, IZMIR, AN
SSSR, AANll, and TsIP.
. It was resolved that the next conference on atmospheric ozone <;:on-
vene in 1965 for th~ purpose of summing up results of theIQSY.
. The Conferenc"e requested the GUGMS to permit the publication of
the proceedings of the 1964 Conference.
The Conference instructed A. Kh. Khrgian and G. I. Kuznetsov
. to report on its meetings in "Uspekhi Fizikocheskikh Nauk~', and G. P.
Gushchin to do the same in JIMeteoro1ogiyai Gidro1ogiya".
i.
,
. i. 1 .
. ~ .
~ . ~
- 234 -
,."
\ ,
-------
SOVIET STATION DATA ON TOTAL ATMOSPHERIC OZONE FOR 1963
Total atmospheric ozone data recorded by 24 Soviet stations during
1963 are listed below. Ozone data collected at 24 USSR stations for the
period of 1957-59 were published as a collection in [4J; simila1." data for
1960..62 were published in monograph form rere referred to as [2J. De-
tailed information regarding all USSR ozonometric stations the data of which
have been discussed i11 the prese~t report is also presented in [2]. Unlike
the data presented in [4J and [2J, data presented in this r'~port represe"nt
ozone results by employing the same ozonometric apparatus, namely, the
universal ozonometer [3J; with the exception of the Leningrad Voeikovo
data which - were obtained by the Dobson spectrophotometer.
Comparing the earliest ozone data [4J with data following later, it
should be noted that due to imperfect OFET-3 apparatuses used, the data
represented only satisfactory approximations. In or de r to obviate the
Forbs effect it was necessary to disregard ozone data obtained at low solar
elevation. ~ut even this failed to completely obviate the introdudion of
systematic error~; 50 that this fact must be taken into consideration in
analyzing ozone data obtained during 1957-59. This applies particularly
'to ozone minima and maxima determined during that period. Recent ozono-
metric instruments have been considerably improved mainly as regards
methods of ozone -measurements. The Forbs effect and temperature fluc.tua-
tion have practically 110 effect on present day ozone determination results. -
However, vibrations in the aerosol layer sti11 affect ozone measurement
results. Ali published results were obtained in direct sunlight. All uni-
versal ozonometers used at the USSR stations were attached to Dobson
spectrophotometers kept at Main Geophysical A. 1. Voeikovobservatory.
Ozone data were sent to the State Geophysical Observatory for critical
checking and analysis.
Final preparation and publication of the
G. P. Gu.shchin, 1. 1. Romashkina, and O. N.
Geophysical Observatory.
data w,?re accomplished br
Chemyakina of the State
Bibliography
------~-- -._-- ~-_._-_.._----------------- -.- - .------.
1. r y W II H r. n. I1Cc.~e.iWBaHIiJI aTMoc1>epHoro 030Ha. rll.iJ.pOMeTe01l3JlaT, JI.,
. 1963. '-.
2. r Y W II H r. n. 030H II a3pocIIHOnTII'IeCKlle YCJlOBIIJI B anlOc1>epe. rIlJlPO-
MeTeOIl3.'laT, JI.. 1964.
3. r Y W II H r. n. ABTopcKoe CBII!\eTeJlbCTBO CCCP. ,Ng 160877. 030HoMeTp.
EIO.~.:JeTeHb 11306peTeHllii, ,Ng 5, 1964.
- 4. llaHHble no XII~III'IeCKmIY cocTaBY aTMoc
-------
- ..._-.~-
--_. -'-'=~-":"=:.:'-'::":::::="':.: -. -
TOTAL ATMOSPHERIC OZONE FOR 1953 AT USSR STATrONS
--~ -- -- - - h - -
OBSERVA- No. OF TOTAL OBSERVA- No. OF TOTAL
TI ON I OBSER- OZONE TiOll OB5ER- OZONE
DATA VATIONS I03CM DATA VATIONS I03CH
. h
ABASTUMANI (41045' N~ LAT., 420 ~, E. LONG.)
JULY . 19 27 226
20 4 218
20 9 228 23 23 . 253
21 27 207 24 6 ""229
22 2 242 25 17 216
25 15 232 26 8 195
26 - 19 364 27 27 216
27 21 202 28 30 208
28 28 218 29 ;. ___2~--- - 198
30 10 236 . AVERAGE ' 224
31 -. --_.~- 262
AVERAGE . 243 OCTOBER.
.,
-- -~ ---.---'-'--.-- I 9 236
AUGUST 2 9 249
I 8 232 3 18 214
3 18 234 4 12 228
6 10 228 5 10 256
7 48 228 7 23 226
8 9 227 8 23 204
13 5 241 9 20 216
14 5 220 .10 27 207
16 44 224 II 12 2I1
27 6 236 . ]2 8 214
28 3 285 14 21 225
29 18 250 15 17 205
30 31 240 16 6 230
3'1 h ...29._- 226 18 20 215
AVERAGE . 236 19 q d' 7. 222
-- _.--.-- _.~-- --~- AVERAGE . 222
SEPTEMBER --NO.VEMBER -
I 22 204 5 20 201
2 10 257 6 18 209
. 3 6 . 225 7 18 220
4 29 246 8 15 233-
7 32 247 10 15 244
8 23 219 13 12 240
9 21 207 18 20 228
10 27 209 19 18 210
II 25 2I1 20 II 208
12 30 236 21 13 207
13 23 255 22 ---- .8 - 217
14 16 240 A_v.~~AG_L~- _,., ...220
15 10 226
16 20 211 DECEMBER
17 22 228 3 2 255
18 26 225 4 22 263
- 236 -
~ --'~---'4_- -
-------
. .--
.. - - -. ~- -- -. .
_. - ----"- -----
o'sERT':l-N"o~ iii d. ----
OeSERYA- No. OF TOTAL TOTAL
TION OBSER- OZ~NE . . TION OBSER- Ot~NE
DATA VAT I OMS 10 CM DATA VAT I ONS 10 Cf1
11 3 210 26 17 245 .
12 20 259 27 15 233
20 10 239 28 8 234
21 12 238 ---.- . 239
AVERAGE.
23 20 210 I I
1 .
10
12
17
20
22
23
24
25
26
28
"
.A~"~TA (430-i-:V- N:--UT.; 760 56' E~-L-ON~)..- .
--.-----.....-
2
3
7
11
12
14
16
17
18
20
21
23
24
25
26
27
28
29
JANUARY
6
15
4
9
3
9.
3
6
5
12
3
15
18
13
8
8
3
15
-~.__.~-
AVERAGE.
-------
FEBRUARY
11
15
3
6.
16
9
13
15
9
12
____3__-
AVERAGE
---
MARCH
5
6
10
6
21
12
319
328
480
406
342
372
385
324
311
333
383
356
329
358
360
336
342
326
.355
309
415
405
367
419
339
354
293
309
301
351
.352
405
338
424
- 237 -
11
12
13
14
15
16
17
19
24
25
26
27
29
31
21
9
15
9
12
11
15
9
J
9
21
17
. 3
---- J2 -
AYERAGE .
-----_. _.-
A1'RIL -
7
11
18
19
11
15
10
12
21
18
17
12
18
3
6
15
3
15
---~.-
AvtRAGE
.- _u. - MA Y - -
21
21
4
5
6
7
8
9
11
16
17
18
20
21
22
23
24
25
28
29 -
30
1
2
402
361
390
397
433
385
391
345
442
345
322
312
432
289
.377
434
389
364
285.
266 .
305
320
358
322
329
358
406
3511
296
365
352
292
311
289
.338.
I
258
263
-------
---..----- ----.---
n - .- - . - - -. '-
-"- ---.- .-.. _._--~--
- - -- - - . .- -~. - -. - . --.- - ---
--- --. _.__.~ ... .---- - -- --
OBSERVA- ! No. OF - :TOTAL OBSERV~ No. OF TOTAL
TlON ' OBSER- - ' ~ OZ~HE , TIOR OBSER- OZ~NE
DATA 'VATIONS -10 CM. DATA VATI ONS 10 CM
3 18 246 16 18 270
4 13 243 17 9 241'
5 4 237 18 17 236
7 12 252 19 21 237
8' 3 288 20 10 236
- 11' 18 329 21 3' 277
12 18 279 .22 21 262
16 21 341 23 15 248
17 9 320 24 9 263
18 ._-- - _6- --- 336 25 18 256
, AVERAGE . 283 26 2 260
27 21 255
----~-- ----- 28 12 277
JULY 29 21 263
5 17 241 30 21 ' 259
7 15 263 31 .._____~1_- 263
8 18 223 AVERAGE. . 257
.9 3 230 --.-----.+. .~._- --
10 5 247 SEPTEMBER .
11 6 286
12 18 312 1 21 211
13 - -- 9 290 2 17 247
14 6 273 3- 18 250
15 11 264 4 21 265
16 15 273 5 21 250
17 20 252 6 17 260
18 9 265 7 3 268
19 21 276 8 3 - 248
20 20 257 9 6 ' 226
22 6 318 12 -9 258
23 11 282 13 8 271
24 21 267 14 9 282
25 21 233 15 21 283
26 20 209 16 21 250
27 3. 197 17 21 248
28 13 224 18 20 220
29 3 241 19 21 234
30 5 276 21 21 277
31 - -~-- -- 9.- 263 .22 21 220
AVERAGE, . 259 23 20 202
v/ 24 9 230
------. --- .-- 25 12 288
AUGUST
28 1&. 300
I 6 273 , . AVERAGE . 252
2 21 258
7 21 251 - - . - ---..
OCTOBER
8 12 253
9 21 286 2 21 239
12 21 267 3 11 245
-13 21 252 6 21 252
14 - 21 259 7 6 277
15 . -, 6 241.. 10' 21 279
,
- 238 -
, .
-------
-. -... --.. .. . u -
BSERVA-, No. OF ---TOT Af--- ------ -- --------
OBSERVA- No. OF TOTAL
TION I OBSER- OZONE TION OBSER- OZONE
DATA VATI ONS l03cM DATA VATI ONS l03cM
11 18 258 15 10 251
12 21 281 II} 13 266
13 15 279 19 6 276
14 15 269 21 20 213
15 17 246 22 5 198
17 15 244 23 7 195
19 3 257 25 7 231
20 6 281 28 - __J2.__- 242
22 3 257 AVERAGE ' 243 .
31 6 305 .- --- ----
------ .265
AYERAGE DECEMBER
--.---- - 10 3 267
NOVEMBER 14 11 303
1 15 , 260 17 3 332
3 7 316 20 14 282
7 3 205 21 3 190
8 11 ~ 220 23 5 209
11 9 302 -----.--.----- .264
AGE
o
AVER
. --~;';;~~BA;(370-~~- N. -LAT~:- 580 20' i: - LO;~:)-"
--.--------
JANUARY
1
2
3
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
3
5
18
4
12
21
5
21
15
21
21
21
21,
3
4
21
21
21
21
, '16
15
18
21
_._5~
~!.E~~
FEBRUARY'
1
6
12
6
302
338
303
337
324
267
330
350
292
235
, 211
224
256
297
311
324
240
233
261
240
. 230
202
246
285
. 281
'I
296
322
- 239 -
. 10 21
11 18
12 14
13 2
14 2
16 9
17 24
18 18
. 19 24
20 21
21 24
22 24
23 3
, 24 19
25 6 '
26 \ , 6
---- ~-- -
AVERAGE
--~ -
3
4
5
8
9,
10
11
12
14
15 '
16
MARCH
10
2
23
26
21
21
21
27
12
27
21
I
\
"
'II.
352'
351
310
272
240
297
248
316
331
254
289
270-
235
274
. 283
298
. 291
381
35~
295
449 '
362
349
312'
344
339
299
284
"
--- ~~
-------
- .- - - -
- ... _. .~---- -
--- - .----- ~--
--_.- -.-.
--. -.. --.' .. -. ------ ----- -- - "--...---- . -.--. . ---'--~. '"
OBSEIIVA- tlO. of TOTAL -OBSERVA-- No. OF TOTAL-
nON OBSER.. OZ~NE nON OBSER- O~NE
DATA VAT'ONS 10 CM DATA VATIONS 1 CM
17 6 292 9 29 302
18 18 318 10 32 288
19 6 276 13 5 313
20 9 296 14 33 346
21 -5 366 15 33 311
22 8 377 16 14 268
23 21 376 17 -.- - 18 - - 269
24 21 285 AVERAGE . 284
25 27 270 - ---
29 26 312 ---
JUNE
30 15 272 35 182
31 - 6 .278 20
AviRAG.£-:. . 325 21 33 169
22 . 32 168
UIIIl]~ 23 18 217
24 14 264
3 24 300 25 33 274
4 24 373 26 33 228
5 24 286 27 30 191
6 9 272 28 14 201 -
7 4 291 29 30 234
8 5 296 30 7 279
9 17 269 --.._----~._._- .219
11 18 240 AVERAGE
13 6 272 .------- ---
JULY
15 6 260
16 12 274 I 27 278
17 18 311 2 30 224
18 21.-- 288 3 24 192
19 12 280 4 33 - - 187
20 6 319 5 32 178
'22 - 6 16 185-
16 330
23 21 325 - 7 21 190
24 18 269 8 17 185-
25 6 288 9 18 192
26 3 293 10 17 185-
27 24 279 11 9 14~
28 21 269 12 32 186
29 21 236 13 21 198
30 6 240 - 14 21 212
'-AVERA'GE . . 286 15 24 201
16 7 230-
-- - -. 17 21 182
HAY 18 21 - 167
1 24 224 19 21 168
2 3 248 20 21 170
3 15 262 21 21 185
4 4 281 22 21 182
5 30 283 23 21 . 182
,. 6 11 284 24 '21 168
7 9 292 25 20 155
8 27 287 26 5 159
8*
- 240 -
-------
,~
-----_..- .- - - - -- -- -----
OBSERVA- No. OF TOTAL OBSERVA- No. OF TOTAL
TlON OBSEII- OZgHE TION I OBSEII- OZ~NE
DATA VATIONS 10 CM DATA : VAT! ONS 10 CM
27 9 173 10 15 240
29 21 178 11 21 219
30 20 181 12 21 198
31 ____13_. 216 13 21 224
AVERAGE . 188 14 21 216
15 21 212
-_.. --- 16 20 205
AUGUST 17 21 192
1 21 218 18 21 194
I 2 14 203 19 17 198
3 21 210 20 21 195
4 18 206 21 . 20 179
5 20 190 22 4 205
'6 21 189 23 21 224
7 21 172 24 14 204
.8 21 163 25 21 189
9 20 174 26 21 187
1.0 18 180 27 21 185
11 21 184 28 - 20 179
12 21 173 29 21 197
13 21 166 30 21 187
14 21 173 -~-- .-. . .195
15 21 171 AVERAGE.
16 16 175 ----- --- -
17 18 182 OCTOBER
18 . 21 179 1 21 189
19 21 188 2 21 190
20 17 195 3 19 193
21 _21 - 178 4 19 186
22 17 199 5 18 196
23 21 191 6 - 20 198
24 18 201 1 15 230
25 11 205 8 20 234
26 21 199 9 21 235
27 21 196 10 . 21 188
28 21 188 11 21 184
29 -21 203 12 21 176
30 21 188 13 21 177
21 172. 14 - 13 203
31
---- .- . .i87 15 21 203
AVERAGE. 16 17 186.
----- - - ---- 17 12 197
SEPTEMBER 19 15 232
1 21 178 20 21 242
2 21 165 21 21 176
3 21 170 22 15 182
4 21 173 23 14 169
5 21 167 24 21 193
6 21 168 25 15 202
7 21 179 26 20 221
8 12 199 28 9 . 234
9 2 212 -29 17 233
I I
I ! ~
.
I
I
I
I
I!
, I
'II
i'
, i
- 241 -
I :,
-------
. .- ..-. -_._.. ----- -------------.-------~------.-.-~-...---.- -.-.
'.- ..- ---...-
-- q - - No. OF TOTAL OBSERV".. -- -- --- - .-
OBSERVA- NO;; OF TOTAL
_TION OBSER- OZONE TION OBSER- OZONE
DATA VAT IONS , IQ3CM DATA VAT IONS IU3cM
30 21 223 19 21 241
31 - _____--2L - 201 20 18 218
I - AVERAGE. . 203 21 21 212
22 21 206
----.---'- ---.....--- 23 12 204
NOVEMBER 24 11 196
3 5 248 27 21 273
4 2 225 28 21 246
5 5 269 30 -----..15- 361
6 21 245 AVERAGE . 241
7 21 218 DECEM3ER - :
8 20 233
9 21 242 1 20 294
10 21 232 7 21 255
11 21 230 26 20 406
13 21 265 27 12 364
14 21 236 29 20 373
15 21 214 30 15 324
16 18 204 31 9 315
18 21 319 - - AVEiiAGE~ . . 333
16
20
21
23
24
26
30
9
11
15
16
19
21
28 -
.--- ----"u -..-.-"--
----.-.----------..--------.-----.-------- ._~--_.._---
SEPTEMBER,
BL'SHAYA ELAN' (46055' N. LAT., 142044 E. LONG.)
-2
5
7
6
12
9
-- --_--_--_3_-
AVERAGE. .
_._-.--- .-._--
- OCTONER ,
12
8
12
3
- 6
12
--------Q --
AvERAGE'
264
235
215
240 -
251
230
234
. .238
246
255
271
309
292
264
231
. . . 267
--------- .
4
5
13
19
22
25
28
NOVEMBER
6
10
6
9
6
6
-- -- ___3_-
AVERAGE.. .
DECEM9ER-
6
3
8
3
3
8
6
I __6__.
AVERAGE
, 323
. 264
269
254
229
293
253
. . 269
270
297
310
326
347
271
311
366
. . .312
- --- - . - ._-
- -.-.. ---- -.---.-- ---'---~------_. -.. - -.-.--. ...---:--.
10
12
14
16
19
20
26
27
VLADIVOSTOK (4300~ N. LAT., 131054' E. LONG.)
JANUARY 3 19
10 I 428 4 11
2 495 5 15
- 242 -
1
2
488
496
448
-------
------- --.---.--- --- -. -- . -.. - _.- -OBSERVA-- --
OBSERYA- No. OF TOTAL No. OF TOTAL
TION i OBSERYA- . OZ~WE TI ON OBSER- I OZgNE
DATA 1I0NS 10 CM DATA VAT! OtiS 10 CM .
,
7 18 472 27 26 438
8 7 390 28 --~- 393
9 18 427 AYbAGE . . 410
10 16 487
13 18 '.522 ----
14 12 540 MARCH
15 12 530 2 27 381
16 13 516 3 27 419
17 17 534 4 32 480
18 12 532 5 27 431
19 5 505 6 28 451
20 6 476 7 29 440
21 . II 570 8 30 392
22 6 517 9 30 382
24 16 567 10 30 337
25 18 556 II 30 350
26 18 484 12 30 . 372
27 9 495 13 30 355
28 2 478 14 9 343
29 12 438 15 18 417
30 4 . 441 16 29 420
31 ___II 447 17 26 435
AVERAGE . 492 18 14 420
-.----- . 19 12 374
FEBR UARY , - 20 23 371
21 31 421
I 8 452 22 22 431
2 3 524 23 29 444
3 17 471 24 29 408
4 22 466 25 26 . 423
5 24 426. 26 9 382
6 24 394 27 6 432
7 24 357 28 3 388
8 24 355 29 30 417
9 24 374 30 3 385
10 24 387 31 1___-,,-. I 402
II 9 441 AVERAGE . 403
12 18 426
13 23 378 -. --- --. -
14 15 347 AI"R I L
15 24 386 I '8 496
16 24 350 .2 33 379
17 19 446 3 25 337
18 24 384 4 23 360
19 24 431 ,6 16 428
20 24 390 7 29 . 436
21 24 389 9 5 510
22 24 382 10 18 456
23 8 411 11 10 371
24 11 435 12 32 316
25 26 422 13 3 302
26 II 44~ 15 6. 345
- 243
/~I .
-------
-- .-- -- -------------
- ---.-- -~---_. ---- ---_.. ----_._._._----~ --".- - - -. ,-
---- - -- -------- .- -.
--------- - -------
OBS£RVt.- 1'10. OF TOT AL OBSERVA.- No. OF TOTAL
TION aBSEil- oz~rlE TlON OBSEII- OZ~NE
DATA VATIONS i 10 a1 DATA VATIORS I 10 CM
16 18 327 6 12 273
17 21 306 9 23 309
18 15 276 10 39 307
19 31 350 11 30 306
20 31 354 12 18 282
21 33 403 14 6 278
22 18 320 20 6 217
24 18 346 22 16 224
25 10 468 23 15 229
26 24 423 27 6 236
27 3. 445 30 2 465
2'9 30 339 31 -__,.._18_-- -. 243
30 13 388 AVERAGE 0 00 284
.. -- _n- - - 0378
AVERAGE --.---- ---
- .----- -----_...- AUGU~T
HAY
I 1 9 235
1 9 430 3 24 236
2 33 437 4 26 271
3 13 403 5 2 215
6 3 372 6 12 263
7 14 415 8. 14 246
8 27 412 6 -- 249
10
9 44 387 15 6 239
10 35 431 20 17 245
11 34 387 23 19 246
12 27 346 24 8 301
13 26 307 25 30 283
14 3 380 26 22 255
15 36' 352 " 28 9 252
16 32 315 30 18 262
17 30 294 31 30 252
- 18 --- --__6 366 - '-AVERAG! . '. . 253
,--
AVERAGE 0 376
_.__._---~-_._.-
-- ------. -- SEP'fEMBER
JUNE
11 17 374 2 4 , 288
12 38 308 3 16 283
13 '39 316 4 24 268
14 3 307 5 7 298
15 21 323 6 19 270 '
16 38 32t 7 33 256 '
17 30 267 8 24 259
,21 12 282 '9 17 290
28 3 300 10 5 290
30 23 243 11 24 270
12 33 274
AVERAGE. . 304 13 33 271
JULY 14 11 274
16 7 236
1 I 31 I 291 17 21 280
2 19 308 18 12 270
- 244 -
, I
i
-------
. .. .--.-----.--------'-'-
-~---- -.--'--- -- ---. -
o
..
BSERYA- i No. OF TOTAL - rro;-oF TOTAL'
OBSERVA-
TION OBSERo- OZ~NE TIOff OBSEA- OZ~NE
DATA 'lATIONS i 10 CM DATA . 'lATIONS i , 10 CM
19 33 275 6 9 317
20 19 266 9 21 322
22 3 300 10 16 302
23 32 279 12 18 319
25 9 282 13 20 336
26 9 329 14 15 290
27 13 288 15 17 330
28 25 297 16 8 368
29 ___J2.- 309 17 II 331
AVERAGE . 281 18 18 316
20 21 302
------- 21 21 3ll
OCTOBER i 23 6 354
333 24 3 305
I 4 26 19 359
2 27 286 28 21 396
3 5 297 29 13 358
4 3 285 30 21 301
5 8 302 ._---- - .. 325
6 19 318 AVERAGE
7 18 333 --..--~_._----
8 21 310 ,OECEMBE~ .
9 2 298
10 3 332 I 19 3ll
II 26 324 2 21 361
12 21 282 3 18 317
13 6 259 4 6 340
14 27 294 6 10 388
15 27 302 8 21 342
17 21 344 9 21 344
18 21 292 10' , 13 357
19 24 315 II 17 325
20 27 301 12 19 378
21 27 310 13 18 323
22 27 297 14 4 353
23 22 280 15 21 326
24 19 262 16 15 338
26 22 315 17 18 302
27 26 292 18 19 389
28 6 314 19 19 351
30 18 276 20 12 41I
31 . ---~--- 290 21 17 394
AVERAGE .301 22 ' 17 389
23 14 366
.---.-- 25 20 396
NOVEMBER 26 15 442
27 20 427
I 24 274 28 3 392
2 18 316 29 3 " 394
3 15 332 30 20 377
4 20 350 31 __M_- 399
5 21 270 AVERAGE. . 366
'- 245
'\;
"
-------
-+-_... - ._- --. ~.-
TEM8E.R .
13 6 296 5 5 250
14 6 273 6 14 240
15 7 207 9 3 233
22 9 275 10 20 278
23 II 305 12 3 293
24 23 332 13 4 310
--+ - .-.- .288 20 9 . 277
AVERAGE
23 14 255
-'-J\UGusr- 25 9 285
29 7 290
14 7 232 -." .260
15 9 240 AVERAGE.
19 9 217 OCTOBER;
21 16 226
26 14 226 I 5 305
27. 15 225 +-- <-------- .305
AVERAGE
.----... ---- --- - -- ------ --+---------_... -----.------ ~- -----.-- --.---
DUSHANBE (380 3~ . 0
N. LAT., 68 37' E. LONG.)
-JUL Y-- 27 21 197
28 17 172
3 18 243 29 18 189
4 21 237 30 15 185
5 21 220 31 ----~ 21 .. 206
6 16 204 AYERA9E ', .210
8 20 203
9 19 208 AUGUST
11 6 179
12 21 205 2 24 194
13 24 234 3 9 218
14 21 221 4 18 191
15 21 229 5 IS" 181
16 19 236 6 15 185
17 24 225 7 17 182
18 24 206 8 21 188
19 22 210 .9 12 185
20 7 181 lO 12 182
21 18 221 11 22' 197
22 24 212 12 . 20 186
23 24 222 13 12 184
24 22 214 14 21 190
25 22 202 15 11 187
26 18 188 '6 13 116
- 246 -
-------
- - -"'----.-----------. -- --.---------.-.-.-- - -.
--- TOTAL -- - ----
oaSER'''' No. OF OaSERv" No. OF .TOTAL
1108 I OaSER- i OZONE 1 TIOH OaSER- ozorlE :
. r03cM I
DATA I 'ATIONS ~ j. DATA 'AT I ONSu' 1 03C/1 I
17 14 186 3 24 196
18 13 176 4 15 194
19 12 '189 5 20 191
20 14 174 6 27 207
21 11 193 7 27 218
22 19 168 8 24 215
23 27 196 9 27 224
24 22 182 10 '. 20 214
25 24 178 11 21 204
26 '16 197 12 24 .191
27 21 188 13 15 173
28 24 . 192 14 18 191
29 15 177 15 14 183
30 12 188 16 21 , 199
31 _._~- 188 17 17 190
A~E!lAGE .. . '186 21 21 200
22 15 170
. SEPTEMBER 23 23 190
24 18 186
1 13 187 25 20 197
2 12 196 26 13 . 211
3 14 182 27 21 199
4 18 188 28 . 4 200
, 5 14 197 29 24 213
6 11 .184 30 24 233
7 12 188 31 12 201
8 15 186 . --- - . 199
9 19 187 AVERAGE
I 10 13 168
~ 24 200 ----'---~ - --
12 NOVEMBER
13 24 212
14 23 205 211
15 18 206 I 21
16 . 24 206 2 3 195
17 '24 214 4. .3. 225
18 24 201 7 21 225
19 23 193 8 - 11 215 ,
20 12. 182 9 20 263
21 20. 209 10 . 21 25i
22 16 197 11 6 281
23. 12 214 12. 13 207
14 172 15 21 269 '.
25.
26 15 '180 16 21 201
27 17 197 17 21 247
28. 20 178 19 18 313
29 20 169 20 20 259
30 _..~ 221 21 21 233
AVERAGE . 187 22 21 234
23' 9 209
---- 24 3 289
oCTOBE~ 28 18 273
1 20 192
2' 18 194 AVERAGE ./. . 242
. I'
- 247 - I
t...
- .
-------
- --. -- 7 - -_n+_- +. -~~~~--- -I h -+ --- -- -No. --
06SERVA-i No. OF OBSERYA- OF TOTAL
TION OBSER- TION OB8i!11- OZOtH!
- I .I03CM VATION8 I 103011
DATA, VATIONS DA1A
'--- 16 17 292
DECEMBER 17 9 296
1 2 324 18 21 293
2 15 291 -19 17 289
7 12 279 20 17 305
8 18- 239 21 17 314
9 21 226 22 15 295
10 5 231 23 6 325
11 10 254 - 24 2 335
13 16 350 30 9 361
14 8 316 31 ______21-- 299
15 20 331 G .291
AVERA E
------- -------7- 1040~-21-;--E~--~~~~Y-
---- ------. ~ -
IRKUTSK (52010'N. LAT.~
, -----.- -- -++------
JANUARV APRI L
1 6 376
14 I 3 517 2 10 526
15 L _u_- J 481 3 5 505
25 538 4 4 525
AVERAGE .512 5 8 473
8 21 407
-FEBRUARY- 9 11 420
10 3 380
12 9 413
2 6 560 13 13 - 504-
15 6 532 . 15 1 559
16 7 479 17 1 460
20 13 515 19 6 484
~ 3 553 . 23 18 505
11 468 24 10 444
26 _--_IL- 485 21 4 555 .
AVERAGE. . 513 28 - --_15 - 529
AVE:lAGE . .475
. ..------------ ----
MARCH MAY ,
5 16 520 3 20 . 460
6 3 467 4 14 442
7 6 535 6 8 496
8 13 477 8 16 311-
9 12 475 9 ------- _6 362-
13 14 415 AVERAGE . 429
14 12 420 -------' I
19 14 615 JUNE
20 18 556 22 8 342
21 15 534 24 5 276
22 19 467 25 12 295
23 9 452 26 18 306
27 12 466 27 19 378'
30 9 529 29 16 29~
- -iV-ERAGE . 495 -- - AVERAGE ,-31
.. 248 -
-------
- -.- -_. - -. - .
I O'S,".,,- - - - -..~ .. . - .
OeSER VA- No. OF TOTAL No. OF TOTAL
TION oeSER- OZ~IIE 1I0H t 03SER- O%OI/E
OATA VATlOIISl 10' CM DATA VATI ONS 103CM
h
Jut.'/' 27 I 17 I 248
1 27 295 28 , 1~ ' 291
3 8 309 29 257
5 26 272 ' AVERAGE -. .235
6 23 251 '------
SEPTEMBER,
1 17 274 3 ' 13 263
8 6 260
9 21 282 4 ' 14 234
10 24 265 5 11 263
11 22 302 10 9 289
12 25 288 11 9 255
15. 20 270 - 14 14 260
16 2l- 274 16 1 237
20 12' 250' 17 9 271
24 16 196 19 12 ' 279
, 26 21 213 20 21 245
. 27 18 206 21 6 308
29 4 232 24 1 248
, 25 2 267
31 3 237
AVERAGE . 260 26 6 340
28 15 277 I
---AUGust 300 19 ,310
--- - . 272
". ,AVERAGE
i 11 248 -OCTO-BEA --
2 12 211 5 8 266
3 20 I 238 '1 16 284
5 21 206 8 16 302
6 15 210 16 10 321 ,
8 6 187 19 15 280
9 12 149 24 8 316
12 8 197 '28 8 262.
13 11 210 29 11 240
14 8 273 30 _u- ..JO- - 260
16 13 250 AVERAGE . 282
19 13 284 ." -'
20 ' 25 227 NOVEMBER
21 18 228 1 J 14 I 292
22 22 246 2 11 284
23 6 250 14 9 354
26 6 299 AVERAGE. . 310
-~._- 24
NOVEMBER 25
11 12 . 222 28
12 15 230 29
23 13 205 30
.-
----K~;~~-N;~'-(49048;--N. L~T..-,- 73°08' E~-~~IIG.) .
9
12
14
5
9
240
212
281
333
361
- AVEiiA~'E . . . . 26"
- 249
-------
_.,- ---- - _._- - -. ----..- -----_. u_--- ---- -O~SE~V;:: -.. '-' . .
o BS_E'RV"'" i No. OF TOTAL No. OF TOTAL
'TION OBSER- OZONE : TIOH ' OBSER- ala""
:
DATA VATIONS I ro3CM ' OATA VATIONS lo3cM
-I
..:..-...- ------- 13 4 273
DECEMBER
15 10 318
1 4 I 346 18 12 359
5 2 235- 19 6 349
6 3 251 20 8 259
7, 6 350 22 12 336
8 10 278 23 12 271
9 2 244 31 -4 330
----..-..
11- 7 220 AVERAGE-- 295
. -
:------KIE~-{:iJ°i4;UN.--lA~~:-3ch7' -E:-~~;~~-) --
-"J.uiliARY;
"
I.
9:
11
12
II}
Hi
I~'
20
21-
22
23
26:
29;,-
30:
1
7
9"
11 ;
12
13
23
24 '
25
27- '
28
}
4
5
6-
",
14
2
1
2
6
: 3
14
: 15
9
9
9
6
12
. 3
------------
AVERAGE
----"- .-..-----
FE9RUARY
15
15
3
18
21
9
22
17
21
21
17
--------~----_.-
AVERAGE'
----pfARCH--
15
, 14
, 21
17
311
245
360
269
458
413
436
283
416
458
403
382
508
347
- 385
386
399
392
358
400
412
454
460
422
587
550
. , . 430
453
464
492
395
9
10
11
18
19,
20
23
26
29
5
6
7
8
9
18
19
20
21
22
23--
24
25
26
27
28-
29
2
3
4
5
-. Zr,O -
21
18
6
21
17
19
9
15
. 9
~----AVERAGE
---u~-APRTC '
23
28
26
18
11
6
15
14
12
12
12
9
28
9
7
28
. 20
- ---------- --
AVERAGE
---.------------...-
Mr.Y
6
30
11
9
367
439 .
451 .
502
427 -'
541. .
453
385
387
" 443
456-
438
470'
439 -
425
368
404
395
332
371
344
382
380
373
, 391
395
412 :
:.399 .
358;
322 :
353
352
-------
_0 ------ ------------- --
_._--~--_._- .
. -- -- - - .~U- -
-- - .---
OBSERV" No. OF TOTAL OBSERVA- No. OF TOTAL
TION OBSER- OZONE TION OBSER- OZ~NE
OATA VATIORS 103cH DATA VATIONS I
10 CH
. 6 14 338 23 21 244
7 27 342 24 21 2~
8 20 375 25 21 237
9 24 297 26 21 254
10 27 378 27 21 263
11 30 321 . 28 18 370
12 29 336 29 6 352
13 30 317 30 21 323
14 10 423; 31 21 274
AVERAGE .347 -~ERAGE. ' 288
-JUliE, ---_._- +
AUGUST
10 ' 15 344 I 21 255
11 24 366 2 21 255
12 7 378 3 18 i . 293
13 9 354 5 9 334
14 25 328 6 12 353
15 30 ' 312 8 20 279
16 3 407 9 . 21 268
. 18 12 378 10 6 288
19 19 303 12 18 288
21 16 406 13 .6 278
22' 9 389 14 ~ 8 331
23. 5 354 15 18 310
24 22 304 17 15 278
25 27 267 19 5 280
26 5 296 20 21 296
27. 28 278 21 19 262
28 21 303 22 21 276
29 . . 25 242 26 12 297
-~ AVERAGE' .334 27 19 247
28 21 237
~iiLY-- 29 21 264
30 29 252
I 19 246 31 27 211
2 15 282 -----..-- - - - - . ; . 280
.3 21 285 AVERAGE' '.
4 16 279 --SEPTEMBER
5 9 350 ' '.
6 18 323 1 26 249
8. 21 325 2' 27 272.
9 20 319 3 27 261
II 21 283 4 21 246'
12 21 283 5 II 313
13 12 285 ' 6 12 227
15 3 278 8 4 199
17 18 334 9 6 211
18 . 6 260 10 27 213
t9 14 I 285 II 14 276
20 4 I 255" 12 9 204
22 21 243 14 20 286
- 251 -
/ «I '
.: t,' r
.' . }I!
I: : :.
-------
.'OBSERvl;
TlON
DATA
15'
16 .
17
18 .
19
20
21
22
- 23
24
25
27
28
29
2
3
4
6
7
8
9
13
14
15
16.
17
18.
19
24
31
-NO.-OF'-
aBSEil-
. VATI ONS
, 24
24
- 27
I 26
27
12
27
26
21
24
21
'rl
12
8
._--_._--~ -----
AVERAGE
--OCT~~
17
21
6
6
21
15
12
21
9
3
10
15
6
18
7
20
------ifVERAGE- .
TOTAL
OZONE
lo3CM
228
220
201
216
221
232
235
249
240
285
264
'245
368
318'
. 249
290
280
227
216
287
262
259
205
270
258
332
237
283
360
261
254
. 268
. --.-.-----_._-- --- ...-----
OBSERYA-
. TION
DATA
3.
4
6
8
9
10
14
15
18
23
24
25
29
-----~~-
NO. OF
OBS.ER-
VATIOIIS
I
2
4
9
10
12
14
15
16
19
20
22
23
24
25
'rl
28
29
3.0
--NOv'E-m-ER-
16
14
-3 -
5
15
,5
6
5
4
6
3
,3
6
3
13
5
7
11
-_.J_~ -
AVERAGE
------- ~_.~ -----~
DECEMBER'
4
12.
6
2
.3 I
3.
15
9
3
7
I 611
14
.-'--:"'_-Q..u i
AVERAGE
K~~ ;;~~E~-(53015;-N~--I.~;~-~-~027t'E~-L~~G'.-- .
-----~---_.-
TOTAL
OZONE
lo3cH
231
249
247
277
247
221' '
292
236
242'
243
243
247
258
.240
273
234
274
'rl2
242
. 25-1
219
224
.' 223"
240
358'
190.
232
244'
, 195
241
283
212
223
.238
: ~--_._.,- -" ----
JANuARY FEBRUARY.
I
4 3 317 I 15 I 416
I
18 3 503 2 11 . . 399
21 8 411 4 13 425
22 6 462 5 15 391
23 7 352 6 15 371:
24 10 381 .7 6 .418
27' 9 317 8 15 40f
-'- - AVERAGE . 392 9 11 376
! ~
- 252 -
I.
-------
. . - +--~- ---._----- ---+_._--- -----< --'<-- --.--.-
-- -- -.. -, ,
OBSERV"', NO. 0' ,TOTAL OBSERVA- No. OF TOTAL
TlON OBSER- . OZ~NE TION .OBSER- '
V~TIONS ~ I VATIONS! DZ~NE
DATA : 10 CM DATA 10 CM
10 9 365 13 28 299
'11 9 444 14 21 ; 313
12 18 407 15 11 337
14 6 378 16 : 19 317
,
15 18 394 17 21 315
16 12 382 18 13 312
21'--' ,12,- .-...: . 384 19 9. : 306,
24 18 381 20 18 275 \
25 15 451 21 1 283
26 10 489 22' 13 . ; 268
27 8 492 23 18 259,
28 10 497 24 18 276
------- . 413 25 19 286
AVERAGE
26 11 ' 286
-MARcH-- 27 21 343
'j 16 461 28 18 315
5. 8 494 29 20 290
11' 12 468 30 21 257
:~ ,-~~t - 424 31 ,---,21,_- 250
450 AVERAGE . 304
445 <--.--- -- - ,
AVERAGE ' 457 AUGUST
,, -juNi-- 1 3 314
17-- 6 255 2 13 288
18 16 256 3 18 278
19 18 279 4 19 ,286
20 9 279 5 21 276
21 21 315 7. 11, 302'
22' 11, 332 8' 14 292
23 17 286 9 18 377
24 6 302 10 20 299
11 20, 314
25 13 290 13 16 ; 276
27 9 279 14 ' 21 2TT
. 28 ' 11 287
29, 5 326 15 ' 18 - 275
, 30 :.._,,gl_~ 247 16 18' 318
17 ' 21 ," 294
AVERAGE . 287 19 16 278
--- ' 20 20 : 262
JULY 21 8 262
I 17 288 22 18 . 267
4, 4 291 23 14 254
5 23 337 24 21 290,
6, 20 325 25 21" 263
7 22, 296 26 10 316
8 24 318 27 10 304,
9 22 296 28:', ', 14' 262
10 17 385 29 21 255
II' 20 325 30 18 288
'12 ' 10 305 31 --- -_J~-'- 292.
AVERAGE ; .--. .287
, .
- 253 -
I: '
/.'.-' :
I '
"
.. i' ~
" !,
-------
~---h___- -- . - --- ---- -- ._-~-_.-
u.-- -----.----.---.--..-.-. _n.._-~ --.
... . .. . u ,. .,. -~ -~. ~ .. . .,
OBSERVA- NO. OF TOTAL OBSEflVA- No. OF TOTAL
nON OBSEII- OZOHE TIOH i OBSER- OZOI/E
DATA VATIONS l03cM DATA VATIONS . I03CM
..-------~- 212
. SEPTEMBER' 14 4
.1 . 21 271 15 16 267
16:' 8 281
2 20 223 17 3 273
3 16 282 19 5 284
4 . 20 251 20 17 246
6 14 288 25 12 185
7 20 297 27 10 191
8 ; 20 275 28 6 229
9' 3. 252 29 12 237
In. 17 242 '
: 30 3 238
11 20 ' 221 31, . 5 235
14 I 6, 233 . AVERAGE . 247.
15 9. 237
16" ' 3. 258 _._-~~-~--_._.._-
18 9 244 NOVEMBER
;
19: , 21 269 I 4 233
20 . 5 236 3 14 214
2t;' 16 . 254 4 ! 8 240
23 19 '. 273 . 8 14 196
24 . 21 260 13' 15' 214
25. 21 271 18 i 6; 202
26" 21 ~. 280 23, 10 202
27 ! 12: ; 279 26' 4 252
28 20 276 27 3 251
29: -.-- .._i. _,:L 247 29 11 255
AVERAGE.. . 259 30 8 266 .'
~ --" ----- --. . 23()
----- ---- ~--~_._--_. AVERAGE .'
OCTOBER .'
._=----_..-...----+--
1 18 . 263 DECEMBER '.
2 2 329 l' 11 275
3':, 9 289 4 10 226
:4, 6; 296 ' 5 7 185
5:' 1 307 13 9 247
6 I 21 192 16 5 237
7: 21 219 19 3 243
8: 19 218 20 7 234.
g',' 18 . 224 23,. --.-'- . 6.~u 303.
10 9 212 , AVERAGE .244
. 26 _u 1. -_._~.. I
AVERAGE. .
-_nLE"N-I'~G~~~'(VOEI~~~oj-~O~;- N~-~'~;:..' 300; 9t E. LONGo}
. - JA~IUARV' 20 7 .
22 4
25 3
26 . 6
AVERAGE
MARCH
6
16
289
. . 289
391'
402
475
. 464
. .,.412
---------------------
13
18
. 19
I
FEBR UA.RY
12
9
II
I
357
410
-- 389
1
, 14
I :
343
430
---. - .--- ------
- ~--~------------
- 254 -
-------
-- +.+------.---
-- ---_.&..._- - ' - .. --
OBSERVA- NO. OF TOTAL OBSERVA- NO. OF TOT.\\.
TION .OBSER- OZONE TION OBSER- OZ~NE
!)ATA VATI OilS ,03CH OATA VATlOIIS ' 10 1:\1
"
15 17 407 30 I 17 375
16 3 402 31.__I_-.,}L 350
18 18 : 385 AVERAGE . 329
19 18 4ffi
20 4 371 ,-.- JUNE -
21 9, 407 . - ;
I' 8: 314
22 3 410 2 56 351
23 5 478 3 21 . 349
25 11 399. 4 33 . ' 324
27 14 355 5 44 302
28 12 377 6:. 32 .. 325
29 . 17 361 7: 26 1. 311,
30 . 12 335 8 ': '14. 295
--------- . 391
AVERAGE . 9 8' 338
10. 23 322
-...--.--- II ; 7 400
APRIL
12 . 21 343.
2 12 343 13. 27 365
3 14 341 14 36 416.
4 11 387 15.:. 13 i 349
5 2 425 17 6. 316
8 8 397 - 18 10: 309:
9 6 394. 19:.: 18 316.
11 14 369 20,'.. 14 309
12 5 347 25 6 281
13 9 317 26 10 ' . 273,
17 6 374 28 9 310
20 10 332 -.-- + --_._--- -- . . , 328
AVERAGE "
22 5 366 -
28 8 323 ~-_.__u-
JULY
29 8 321 1 .5 . 306
_.~---- . .359
AVERAGE 2 9 267
" :
----r-1iY- 4 8' 289
5 10 278
296 6' 9 287
3 6 16 3 315
10 6 305 17 10 : 279
11 .9 312 19 3 281 ;
13 3 341 20 3 279
16 9 313 23 24 268
17 38 329 24 14 245
20 42 305 25 12. 255
21 12 343 31 - ---- H 235
22 24 315 AVERAGE . 276
23 11 292. '
24 34 360 ------ -'
25 39 355 AUGUST
26 20 323 I 14 255
27 47 347 2 10 260
28 68 317 5. 9 241
29 5 335 6' 13 243
,
- 255 -
I
I
oJ
-------
--...------ ~.~- -. --
-_. _n-
- .u__. -- ~------
,
I
i
i
I
--.-.-- .---- - -.--------- _ .- ----..-. " No-:oF --- -. ----.., -.
OBSERVA-- No. OF TOTAL OBSERVA-- TOTAL
TION OBSER- OZONE TI. ON OBSER- OZONE
DATA J VATIOIIS' I03Cli DATA VATIONS I03ci'l
7 11 239 17 15 216
8, 10 267 18 6 210
9 6 230 19 12 254
16 16 266 20 3 267
18 9 230 21 7 237
27 '11 249 23 11 247
30 12 241 25 15 248
31 12 . 235 26 -- ... 18 255
-.-- _._-----~-- . 246 AVERAGE ' 238
AVERAGE
- ~_.__.n. -- -OCTODEi-'
SEPTEMBER ,
2 11' 234 2 2 ' 275
:3 9 239 ' 9. 8 233
4 11 231 16 10 261.
6. 3' 248 23 12 236
9 8 235 26. 18 259
10 - .17' 216 30, 4 233
.11 7 23P ---------.----..-.--- : 250
AVERAGE
-- -.. .- ._._~_. -"-0;;;:'-'--- . _.m .---. - ._~----_..-----.: . --.c -- .-
-.- MoscoW (:6 45' N. LAT., 3r-=,5' E. LONG..
~N~R'i'- . .--APRIL~"-
2~__L__J- l 242 1 13 430
337 5 12 420
AVERAGE. .290 9. 13 450
10 \ 20 490
---'FE8RUAR y' 11 20 . 490
6. r 16 440 12 21 440
11 10 . 425 14 13 450
13. . 11 455 19 14 414
19. . . 6. . 366 20 13 440
20 . 17' 450 21 17 438
23. '. 11. 470 23' ..14 . 470-
. --- ..---.----.--'- .' . 433 .24 10, 470'
. . AVERAGE 25 20 520
------MAiICH- 26 11 495
2 . 5 348 27. 11 415
28 15 432
4. 17 . 500 29 5 406
5, 11 460 30 14 431
10 13 495 I .- . 450
II 25 415 AVERAGE
15 22 480 V....,
16 18 525
19 21 . 450 3 17 346
20 18 540 4 13 409
21 20" 520 ,5 11 368
23 16 525 6 4 390
24 15 500 7 14 390
30 : . 19 440 8 9 340
-. ._---.-_.~-.-.._- . 476 9 15 313
AVERAGE
9.
- 256 -
'I
-------
, '
. - -- -- -- -- ------------.---_. ~.._--
" --. -- -
OBSERVA- NO. OF TOTAL OBSERVA-' No. OF TOTAL
TION OBSER- O~NE TlO/1 OBSER- OZOllE
DATA VATIO/IS I CM DATA VATtONS . I 03CM
10 16 325 5 15 255
11 5 380 6 21 23~
12 16 420 7 48 263.
13 .9 465 9. 16 341
14 9 . 483 11 -. 19' : 276
15 . 8 425 12' . . 17 209:
16' 7 445 13:" 17 240'
18 12 396 14 10 .' 301'
19 ,8 355 15 -: 21 . 270
------ ,390 16 24 245
AVERAGE
18 20 279
JUNE-' 19 20 246
21." 24:, 237
26 4 ; 295 22", 22 303'
27 . 13 256 23,: , 24 . 268
28 15 340' 24.'" 17 220
29 11 290, 25 . ~ 19 248
., AVrnBE ' 295 27. 21 229
'. ~
28 19 242
----- 29 19 229
JULY 30 20 243
5 4 240 31 " 17 269
6 15 284 --AVERAGE-:." " 251
7 12 290
8' 16 290 ---- -"-
11 18 263 SEPTEMBER ; -.
.'
12 15 300 1 17 .. - - 240
13 . 12 357 2 16 237
14 16 280
15' 17 308 3 19 243
16 18 . 248 4. 27 215
5" 28 312
17 10 267 6,' 24 20a
18 14 286 7" 22 210
19 '10 .260 9 29 205,
20 12 294. 10 .. 19 200
22 8 237
23 21 .' 220 11 'II 219
14 II 212
24 19 260- 15 . 25 190
25 27 . 300 16 17 205
26 21 287 17 17 -- 170
27- 15 314 18 21 206
28 16 320 20 14 : 208
29 21 260 21 - 17 228
30 17 224 22-. 9 247
31 18 200 23 20 205
- -AVERAGE -. .275 '
24 12 . 230
25 20 300
AUGUST 26 17 306
.- 23 242 27. 13 224-
2 21 242 29 12 236.
3 17 254 30 9 270-
4 21 304 ----~---- . 229-
AVERAGE ;
.
- 257 -
-------
"
.. _. --. --.. -------.-.--. --------- -
OOSERVA-- No. OF TOTAL OOSE!lVA-o No. OF TOTAL.
nON oeSER- O~NE , nON OBSER- O~NE
DATA VATIONS 'CM OATA ' VATIOUS I 011
,
/ ---+
NOVEMBER
- OCTOBER , 2 11 154
" 11 13 115
I' 8 235 18 10 305
22 5 115
2 18 225 25 16 220
3 16 226 27 11 210
4 18 190 29 12 248
6 25 203 --" AVERAGE . 195
8 14 276
9 23 227 ---~~ -.- - -+.-
11 15 244 DECEMBER .
31 11 214 11 ~ L '.11 I 257
,- -'----.... -m . 13 . --' ___8._. 303
AVERAGE .227 AVERAGE ., .280
--'MURMiiiSK-f6e°:et N. LAT.', 33°03' -------- --.- ~ -+ ..
E. LOr4G,,)
_n____n____~-- "
FEBRUARY --'----.
HAY
25 '-.: ---.1..-. I 455 I 3 352 .
27 354 2 21 411
AVERAGE .405 4 . 7 419
-------. ----- 5 26 320
MARCH '6 34 339
7 33 370
3 1-__L 333 8 3 341
6 341 9 4 460
11 490' 10 15 456
13 450 11 16 419
AVERAGE . 404 12 30 387
-------- 13 30 418
APRIL 14 36 390
11 20 465 ' 15 30 332
. 16 18 ' 416
12 3 445 - -----.-.- - .389
15. 24 455 AVERAGE . ,
16' 24 395
18 ' 3 466 JUNE
19 4. 471 20 15 333
20 21 465 26 29 294
21 '5 484 27 27 308
22 32 473 28 32 242
23 26 469 29 32 293
24 33 418 30 10 316
25 33 376, ---~ . 298
26 12 470 AVERAGE
,:
27 13 360 -. --~ - ----
28 12 356 JULY
29" 11 390 1 7 'j 281
3"0.., " 27 345 2 7 277.
. ", """"" --'----'- - .430 3 12 273
AVERAGE'
~ ;.
- 2';8 -
i,
-------
----- -- --- ---.-
__4 - -- ---- -._--~._--- --. --.
No. OF - - OBSEflVA-o - - - - -
OBSEflVa- TOTAL No. OF TOTAL
TION OBSER- O~NE nON OBSER- ozorlE
DATA VATIONS 1 CM DA1A VATIONS: 103cM
4 5 308 20 11 288
8 13 326 21 6 219
10 24 302. 22 ..6 360
11 32 277 23 15 297
12 11 309 28 6 303
14 4 318 29 27 257
16 12 315 31 - - -. 25__. 246
17 3 295 AVERAGE . 280
18 11 275
19 23 271 'SEP-TEMBE~1-'.
20 31 261 1 I 17 251 .
21 6 270 2 18 240
24 8 292 6 12 312
26 15 246
27 14 259 7 10 292
29 12 240 10 9 243
30 ______9__- 258 12 16 . 243
13 3 245
AVERAGE, .283 '14 5 235
---- --.---- - -- 21 6 250
AUGUST .22 3 198
1 17 310. 23 6 228
2 34 275 24 11 236
3 26 283 28 8 300
29 16 225
6 16 265 ---- - . 250
7 22 247 AVERAGE'
8 11 261 --------______4.- -
9 16 259 OCTOBER
10 3 293 6 I 14 I 237 . -
11 3 309 10 4. 231
12 12 284 .--__-___4 - 4 . 234
AVERAGE
----'----------~n______---~--- -----~-_. ..._----- - u.---.
NAGAEVO (59~5' N. LAT., 15:)°47' E. LONG.)
------~---- -------- -~_...-
~
FEBRUARV HARCH
1 18 456
1 7 370 2 15 460
6 8 -A33 4 21 484
7 9 419 5 20 509
8 11 435 6 24 481
11 9 462 7 21 499
18 14 401 8 16 461
20 18 407 9 21 463
21 . 13 478 11 17 495
23 2 500 12 19 530
25 4 499 13 19 527
26 10 521 14 19 502
27 12 501 15 21 494
28 _. --- - 1:t.. 480 16 19 470
AVERAGE . 454 18 21 506
- 259 -
-------
j,
_.- .-- .- - .-
-+. .-. .--+ . . - h_.- - . , --'--
01l3ERV'- No. OF TOTAL OBSERV/t. No. OF TOTI\L
T.qN . OBSER- OZ~NE T'ON OBSER- OZ~E
DATA VAT IONS ' 10 CM OA1A VATIOrlS 10 M
,
19, 27 425 'MAY-- .
20 12 382 2 458
21. 3 444 2
22 22 529 6 23 432
23' 10 522 7 '16 461
24 . 12 515 8 9 420
25' 20 500. 9 21 411
27 .' 5 445 10 20 422
28. 3 456 II 9 421
30 _.._.~--- 347 12 4 415
i AVERAGE .477 AVERAGE-' . 430
I -------
j . . n.. --ApRIL" . j SEPTEMBER .,
I' 9 449 16 L_~-.i~ I 300
2 12 . 517 28 238
3 2 504 AVERAGE . 270
4 8 520 I 'OOfolflflf .'
5 7 550
7 12 556 30 I ..., '.~ I 362
8 8 553 31 396
9 7 565 . AVERAGE .379
10 30 526
.11 25 562 ---.----.,. ._--.-
12 24 541 NOVEMBER ,
13 20 508 ' I 13 359
18 25 446 2 6 349
19 5 409 5 9 408
21 15 409 '6 6 372
22 27 424 9 5 384
23 18 444 II 5 388
24 20 456 i2 5 346
25 6 392 27 3 296
26 9 394 29 2 302
28 --._,- __2.6_. 447 30 .-- - - 4 306
. AVERAGE . 484 AVERAGE . . 351
I
2
6
7
10'
. 15
18
19
21
27
. --. oo~~~ (46029,'N.
-JAiiuARV .
15 232
9 210
5 239
12 225
5 235
12 289
13 322
15 299
5 230
6 352
n, AVERAGE- . . . . 262
LAT.,-3Q038tE-;;LOrIG~ )
.----.--. -
FEBRUARV
4
5
6
20
22
23
24
25
26.
5
15
7
18
4,
15
18
21
21
182
212
188
268
259
250
299
315
306
. . . 253
---+--
AVERAGE
- 260 -
-------
. - - -_.- ----- _. --. -- - - -~--- - -.---- - --.----.---- --~----
-- .h___.- --- .
-----~- -- ------ -----.
OOSERVA.- No. OF TOUL OBSERV,..' No. OF TOTAL
TlON OtlSERo-, OZ~E TION ! OBSER- OZONE
DATA VATIONS ~ 10 OM DATA VATIONS l03cI1
-M.\RCH- 3 21 304
1 18 312 4 14 319
5 19 311
5 21 300 6 18, 326
7 6 289 7 24 323
8 9 324 8 \ 22 343
9 21 319 9 124 345
13 3' 269 10 3 276
18 '3 350 11 5 ' I 290
25' 23 274 12 14 274
26 24 262 13 18 328
27 -_.- _____9 - 301 14 15 346
AVERAGE . 291 16 '3 313
---------- 17 ,i 12 333
APRI!. 18 18 294
4 3 316 19 ' " 15 307
5 II 391 20 16 276
6 24 311 21 ' 18 239
7 24 342 22 18 262
9 9' 363 23 24 276
II 9 302 24 24 269
12 3 286 25:' 24 305
17 9 242 26 24 280
18 24 287 27 24 349
19 15 296 28 24 258
20 21 289 29 18 319
22 18 273 30 21 303
23' 15 288 31 15 - 316
24 24 297 AVERAGE 0' .302
25' 21 308 I
26 . 18 " 311 -AUGUSrr-
27, 13 343 1. 15 265
~' 24 328
30 -~--L 320 2: 24 269
---- 3 24 282
AVERAGE . 310 4 24 274
---- 5 24 289
MAY 6 24 326
1 23 , 274, 7 21 320
2 24 318 8 24 301
3 24 296 9 21 297
5' 24 283 ' 10 18 270
6 6 324 12 24 279
7 24 292 13 24 268
8' 24 318 14 15 318
9 18 297 15 24 282
10 24 266 16 24 245
AVERAGE .296 17 ' 21 261
----- 18 15 261
JULY 19 24 203
1 I. 18 j i, 309 20 24 245
2 15 274 21 24 232
!, '
" ,
- 261 ...
-------
..-- .. ..-.
- --._u .-.- --- -.
o
_._.~ - -~.. 1--------. ._--. ---- -- -.-----.... --.-
BSEnVA-- No.. OF "01' AL OB:;;ERV~ No. OF TOTAL
TION 09SER- OZONE TION OBSER- OZONE
DATA VATI ONS .03cM DATA ; VATIONS i '03CH
22 24 ' 235 3 21 205
23 15 260 4 21 207
24 16 290 6 21 181
25 18 260 7 21 187
26 9 273 9 10 ' 203
27 24 233 10 6 270
28 24 239 12 15 192
29 21 255 13 15 179
30 24 238 14 9 , 211
31 21 244 15: 15 193
- -AVERAGi:" , 267 16 12 195
17 13 200
I 18 12 168
~._'---'- 19, 6 287
SEP'rEMBER 20 6 226
22 9 '255
1 24 2'J:l 27 3 233
2 21 244 31 -- - .....!~- 183
3 24 234 AVERAGE .211
4 24 230
5 12 257 -,
6 18 - 231 . ---- -~ _d__'
1 21 219 . - NOVEMBER
8 24 - 164
9 -- 24 194 1 18 189
10 24 - 207 3 . 3 172
11 15 242 , . 6:: 12 174
12 21 231 -- 8 3 187
13 24 202 9 3 189
14 6 233 10 6 184
15 24 , 231 11 6 173
16 15 232 12 9 167
17 24 196 13 3 179
18 24 193 14 3 202
19 21 217 : 15 12 225
20, 24 194 17 6 170
21 24 188 19 9 184
22 24 205 20 8 164
23 21 213 22 8 173
24 24 202 24 ' ,6 194
25 24 212 29 6 205
26 15 " 223 30 9 .' 176
28 3 182 --- ~ .-.--- .- - . . 184
AVERAGE
30 -.. l~n -253
AvERAGE, .. ,216
-~--.. --.-- -- :
DICEMBER
.
-.-------
OCTOBER 2 uJ._---- i I : 177
3 178
1" I 15 I 207 25 176
2 16 233 AVERAGE"-, ,. -,,177
--
- 262 - -,
',!
, -
-------
. --. ...---- - ~----- -.---
-_no - - n_-
I -- - - - f--' -. - --- ~ --
OBS1!RVA- No. . OF TOTAL OBSEflVA- No. OF TOTAL
TION i OBSER- OZgNE Tloll OBSER" OZ~NE
I' 10 eM. DATA " VAT IONS 10 eM
OATA VATI uNS I
OMsK {54°!Xi' NOo LAT.,! 73°24' E~ LONGf-
JAHU~ 24 1 4 592'
4 9 434 27 3 657
---.-' .520
8 6 327 AVERAGE
9 6 452 --- -. - .--'--
14. 9 408 APRIL
18 11 403 2. 15 537 '
19 10 406 6 12 348 '
21. 12 451 '7 16 320
25 6 408 8 2 348
26 3 399 10 11 505'
27 15 382 11 12 474.
28 13 378 12 21 494
30 . 5 425 14 21 528
-"'VEil AGE- .' . 406 15 3 468 -
17 7 498
- FEBRUm- 18 21 455
4 11 470 20 . 21 443
21 1 496
6 15 450 22 10 467
9 16 364 23 21 . 421
10 18 359 24 17 453
11 12 501 25 11 437
12 7 516 26 -15 434
13 6 446 27 21 404
14 3 437 28 20 494
15 3 431 29 I, - 470
16 3 443 30 - ----~.!--- 343
17 8 428 .445
18 14 433 AVERAGE
20 3 360 .---......:.>.--
22 . 11 462 MAY
24 L 12' 320 2 15 327
25 ~~-- 426 3 9 425'.
26 429 5 21 361
AVERAGE . 428. 6 4 404
7 2 365
---- 8 .3 413
MARCH 10 21 478
'1 18 567 11, 15 417
3 10 435 AVERAGE. . 396
4 3 452
8 9 597 JUNE
10 9 506
13 7 497 20 12 238
14 18 502 21 21 290
19 12 411 22 21 294
20 3 430 23 21 233
23 7 595 24 21 227
;.
- 263 -
,
-------
'"
- --.--- - ---~. ~---.- --"
, .
--_._-~_._---- .
-. --_.u_- - ---- ---. -.+ _..- --. -
----n_- -- --
OBSERVA-o No.. OF, TOTAL OBSERV"",, No.. OF TOTAl.
,TION OBSER- OZ~NE ' TION OOSEII-" OZONE ,
OATA VATIONS 10 C~ I DATA ' VAT IONS '03cM !
25 18 ' 237 '30 4 365
26 ' 13 253 31 -, ------~--- 232
27 21 ,245 AVERAGE . 290
28 19 272
29 21 257 __._".n
30 ---_J!.Q-- 228 SEP1f!MBEk
AVERAGE . 277
1 3 267
2 17 222
-. ~--_.__u- 3 11 285
JULY 4" 12 241
5 12 284
1 21 . 228 7 8 262
2 21 242 8 , 12, 252
3 21 257 9 13 242
4, 16 280 10 20 230
5, 21 258 14 17 245
6 7 " 248 15 ' 20 277
11 11 274 16 - 19 ' 264
12 13 320 17 ' 21 - 258
13 27 312 18 4 294
15 8 289 20 ' 11 308
16 18 256 22 3 258
, 17 18 248 23 5 240
18 6 275 24 6 282
19 20 270 26 10 242
20 18 304 27 7 275
21 ' 16 293 28 5 292
22 16 248 30 -'~-- 17 259
23 ' 15 261 AVERAGE . 262
25 12 253
29 5 309
31 4 250 ; ..---- _nO ---..
uAVERA-;E- .270 OCTOBER.
1 9 265
---AUGUST---- 2 21 246
3 21 256
,- 4,' 21 257
I 21 264 5 '12 282
2 6 335 10 6 282
5 11 335 13 I 270
6 10 326 15 II 254
1 21 262 16 9 251,
8 4 281 17 -- ' - -_.-1_~-- 244
12 3 297 AVe/lAGE ' 260
13 7 340
14 11 289 ---.--.--- ._- -
15 21 254 NOVEMBER
16 ' 21 284
24 11 255 13 -.I ~-- -- ?_- I 264
29 6 242 AVERAGE, ; ' 264
,
- 264
-------
___0 -- - -- -, - ..-_0 .. ----- . - . . , "-
-JBSERV&- No. OF TOTAL OBSERYA- N~ OF TOTAL
TION OBSER- O~NE TlON OaSER- OI%NE
DATA VATIONS ' 10 CM 041.\ VAft ONS 10 CM
------ ----~ - .~--
PETROPAVLOVSK- KAMCHATKA (52°58' N. LAT., 158045' E. LONG.)
-----
: OCTOBER'
, 95
10 6
11 19
12 12
14 21
16 3
19 2
20 13
2\ 11
22 13
23 11
24 12
25 17
28 . 9
29 10
30 - -- - ,---~--
AVERAGE'
251
241
239
239
228
289
24:1
263
247
, 237
259
241
245
253
241
256
. . . 249
NOVEMBER
3 10 254
4 3 264
6 ' 13 273
II 11 258
12 4 258
15 6 275
18 3 246
19 8 251
20" - _~L-' 290
" AVERAGE.. . ..263
- --- - ..-
DECEMBER.
1" \ 2
~ : ~ ----~ --
AVERAGE.
292
322
331
. . . 315
;~~;RA (6?iO:"" 'N~A;~; 57000"£. lONG) ,
-----
JULY 26 I 21 244,
18 1 10 253 27 , 15 255.
19 18 252 28 ' -- --_?l . 24~,
20 6 247 AVERAGE '. .244
22 21 277
23 - ~_! 293 -----
26 249 ' SEPTEMBER
27 218
AVERAGE. . 255 6 3 256
--.-- 7 12 273
AUGUST 11' 15 247
19 7 250 '
,
23 6 225 AVERAGE . 25~
---;;~0$058' N: --~~;.-,-240M' E:-~~~)
--JANUARY .25 L~-- ~- 409
29 435
8 6 337 AVERAGE . 398
9 3 398 FEBRUARY
13 2 395
16 4 353 6 J 4 I 425
21 2 458 9 5 423
- 265 ..
-...
iv.'
-------
- ~_.- -'. -, .- ---
. -.. . - - -_.- ~_.. - --.
--- ..- __n.. --...-- .h ..-. - . .-. - . -
-O~5ER.VW;; -NO;-- OF-
TION : DOSER->
DATA 'lATIONS
----"-OTAi- -OBSERVA-:
OZ~NE '. TION J
10 CM DATA
--
_h-.___-
-.-..-- -.------
NO. OF
OBSen~
'ATIoJJ8
TOTAL
OZONE
103Ct1
11 9 327 . 19 21" 350
12 15 394 20 21. 315
19 4 449 21 8 359
2'1 3 425 22 16 404
22 6 363 23 12 370
23 11 373 26 21 367
25 2 480 27 21 354'
26 ---- -_.:._~. 468 28 14 375 :
AVE'1AGE . 412 ' 29 18 370
30 ---. ___.2. 414
-"--------- - AVERAGE . 387
MARCil :
1. 15 372 ---..--.-
2 16 331 MA"!'
4 3 404 3 3 3is --
. ,
9 19 365 4 6 346
10 18 360 '\ 5 13 328
11. 17 ' 466 6 '20 " 308
14 18 336 7 13'. 354
;
15 21 419 8 12 352
. 16 12 443 9 21 359
17 17 - 414 10 21 - 389
18 20 436 11 14 381
19 12 418 12 20 376
20 15 467 13 19 401
21 15 . 461 14 21 36()
22 18 411 15 7 441
23 13 463 16 21 383
21 6 483 7 ---_.L 414
28 17 472' AVERAGE: . 368
29 12 " 496 - ----- ~--_..
30 21 443 - JUNE
31 ..-' ._:.._.lQ. 457 19 . 21 334
. 425 :
AVERAGE 21 8 408
-_._--- 22 18 370
APRIL 23 ' 4 367
1 16 469 '24 6 367
2 5. 382 25 12 309
3 21 414 - 27 15 354
5 11 510 28 12 362
6 21 437 29 13 319
7 21 394 30 - '-- ._~- 307
8 .21 374 AVERAGE" 350
9 21 404 --------- --- -. -
10 21 397 JULY.
11 15 333 .. 17 285
12 18 317 2 21 242
14 16 335 3 21 269 .
15 20 384 4 21 258
16 ,", 10 418 5 21 262
18 15 428 6 21 276
.
- 266 -
.'
-------
. ._-- --. --- ---_. - _.-.- -- 9_.- --_...~ . ---- -.--- ..-
OBSERV,," NO. OF TOTAL OBSERV,," -NO. OF TOTAL
TlON OBSER- ~NE TION OBSER- OZONE
DATA VATIONSI. 1 eM DATA VATIONS 100eM
7 16 332 5 9 317
11 17 375 7 14 279
12 19 329 8 8 300
13 9 343 9 21 257
14 14 365 10 10 244
15 7 305 12 3 233
16 9 298 13 15 294
]7 9 301 14 17 242
18 .10 391 15 14 214
19 21 310 16 17 224
20 6 376 17 15 210
.21 20 263 18 13 313
22 21 231 19 21 224
23 20 232 20 20 227
24 14 . 334 21 21 234
25 20 401 22 21 . 237
26 12 339 23 9 346
28 12 276 24 9 284
29 10 374 25 21 260.
30 10 264 26 5 314
31 21 335 28 2 288
--- . 310 29 . 6 283
AVERAGE - --------
AVERAGE . 266
AUGUST --OCfcfBE"iI
1 21 335 ] ]5 275
2 14 . 360 2 .3 307
3 17 288 5 4 282
4 12 . 363 .6 7 228
5 21 . 367 8 7 293
6 8 334 9 17 224
7 4 249 10 10 214'
8 20 276 11 ]7 268
10 17 278 13 2 294
.11 3 269 16 6 262
12 4 311
13 7 277 20 16 220
14 10 337 26 10 235
15 13 292 28 4 226
17 10 304 30 11. 266
18 12 304 31 ----~-- 230
21 13 295 AVERAGE. .255
--..---.-
24 16 302 , NOVEMBER
26 12 288 I . 16 256
29 8 267 10 4 238
30 2 233 14 6 286..
31 . 15 240 15 2 294
--- -_. - - _.-- .298
AVERAGE. 17 3 263
------ SEP1EMBER -- 21 8 273 .
24 3 269
1 I 21 229 29 5 .273
2 10 322 30 L____5- 257
3 21 275 . AVERAGE . .268:
- 267 -
I'
.1
.'
-------
.- ~._-~_.~._.~_.... ---.-.-- ..-_..'" -----..-...--- ----....---
-... -" --~ -
OBSERV,...! ---- ------- -'------. _._---~._- - - ----
I NO. OF TOTAL .OIl51!1f'1 tv. NO. OF f.o-'Al
TION , OBt>EII- OZ~NE TIOll OBSER- .~Z~NE
DATA ' VAT! ONS I 10 Ct1 DAtA 'Ad OilS' 10 CM
'SveRDIPVAR (55°48' No -----~.- ~
LAT~, 6O~t' E. LONG.)
.JANUARY 5 12 465
6 24 513
3 9. 232 7 18 476
4 3 355 8 6. 375
5 6 286 9 13 405
7 6 243 11 17 517
8 6 222 12 27 . 424
9 9 263 13 27 . 420
14 9 326 I 14 12 351
15 5 310 17 9 336 : I
17 4 350 19 19 . 443
21 9 395 22. 6 521
22 9 399 23 21 515
23 6 306 24 13 558
24 12 319 26 16 586
27 12 241 27 27 553
28 6 260 28 21 502
30 15 262 29 15 472 .
31 5 256 31 _.__27 ~- 499
---~--- .295 AfERAGE .: . 464
. AVERAGE.
-----.----- ------
FEBRUARY; APRIL
1 6 314 475
4 15 293 1 27
5 3 316 2 27 485
7 15 245 3 27 436
8 18 241 7 15 331
10 14 385 9 19 407
11 18 368 10 30 463
12 18 . 324 Il 28 446
13 18 283 '12 30 432 ,
14 .19 302 13 30 461
15 21 346 14 24 394
16 21 345 15 30 368
18 9 387 16 30 359
22 12 346 17 30 376
23 14 285 18 27 417
24 9 351 19 30 433
25 ' . 21 415 20 26 430
26 21 '368 21 22 4B6
27 .- _._-~J_. - 348 22 12 432
AVERAGE. .330 24 30 417
25 30 450
~__n.- 26 30 415
MARCH 27 30 390
28 27 416
I I 23 I 434 29 18 397
3 21 414 30 .:.._-~_...- 372
4 i 21 437 ' AVERAGE . 420
- 268-
-------
. - -- ... -,~ --- -'.' ----- _.-- _.- .._- - --------- - -
- .._~. -- - - TOTAL-- --.- -'--'-- --..---.
oaSERVA-' No. OF , oaSERVA- No. OF TOT AL
TION I OBSER- i OZ~NE TIOO OBSER- OZOII£
DATA VATIONS 10 CM DATA VATIONS' ,1Q3CM
HAY 19 26 295
4 15 405 20 27 325
5 30 337 21 27 434
22 27 327
6 12 383 23 30 .- 324
7 30 329 25 12 322
8 28 385 26 9 350
9 27 411 27 14 350
10 30 387 28 27 359
11 19 469 29 20 289
12 5. 529, 30 24 . i 312
13 30 422 .31 . 3 352
14 30. 381 -~_.__._._--- .. .; . 345
15. 28 443 AVERAGE
16 22 454 -----.
'J.7- -12,-- 416 . AUGUST
AVERAGE .411 1 3 . 324
I ------. 3 4 358
JUNE 4 8 322
20 27 280 5 27 326
I 21 15 272 6 6 366
22 18 267, 7 2 310
23 19 280 8, 5 314
24 27 234 10 2 398
25 27 262 11 9 328
26 :,15 357 12 27 324
27 3 319 13 " 15 342
28 27 278 14 21 331
29 27 , 265 ' 15 23 288
~O 27 258 16 6' 334
AVERAGE .280 19 8 305
. 20 17 291
-----~
"ULY 21 23 303
22 8 251
1 27 272 23' 22 276
2 27 280 24 17 294
3 13 309 25 9 343
4. 16 314 26 27 261
5 22 346 27 27 262
6 21 344 '28 27 293,
7. 21 . . 354 ,29 27 325
8, 27 335' 30 27 267
9. 21 388 31 27 240
10 24 353 AVERAGE ' 310
11 27 445
12 18 453 _._--~. - _.-
13 18 386 SEPTEMBER .
14 2 373 1 27 293
15 5 349 2 27 293
16 21 348 3 27 324
17 18 376 4' 23 261
18 '27 3fJ7 6 7 318
- 26,9
-------
onSERVA- No. Of" TOTAl. OBSERVAoa No. OF, TOTAL
TI'JN OBSSR- OZOIIE TION OBsen- OZOIIE
DATA : VATI ONS! '03CI1 I).\TA VATIONS l03cI1
8 25 274 15 9 304
9 3 277. 16 12 231
11 2 244 17 3 236
12 27 255 19 3 293
13 27 274 20 3 290
14 27 266 21 9 280
15 8 243 22 3 252
16 18 242 24 12 252
17 16 267 27 12 210
19 I 282 29 3 256
24 7 303 31 9 269
25 12 253 AVERAGE . 2ti5
26 14 239
27 19 335. NOVEMBER
28 27 304 I 10 299
29 27 272
30 25 235 4 5 213
. 275 6 8 214
~VERAGE 7 10 209
8 10 246
., OCTOBER 14 3 266
1 24 223 19 3 217
2 27 277 22 3 207
3 27 294 28 4' 293 .
4 12 322 AVERAGE . 240
5 5 330
7 24 234 DECEMBER
8 24 227 I ,6 245
9 9 248 2 ,3 219
10 4 273 ' 7 3 173
11 12 272 19 .3 228
14 8 247 AVERAGE . 217
FEODOSIYA (45°02' N. LAT., isf23' E. LONG.)
JANUARY
MARCt!
6 15 444
7 8 409
9 15 391
12 12 -. 426
18 15 424
,20 6 476
21 6 448
23 ' ,9 362
25 ., 9 ,419
29 12 ','332
AVERAGE . 413
I
3
19
6
9
,f'.
A'VeRAGE
294
364
461
. . . .376
FEBRUARY
"
5 11
7 15
26. 15
28': 9
: AVERAGE
314
310
429
439
. . . .373'
'.
- 270 -
. "
, I
i
I
..
':'."
-------
-=
036ERVA- No. OF TOTAL 06GERVA- No. OF TOT,!.
TION OaGE\!- OZ~NE TlON OfJSEiI- 0% 0 r/!,
aATA V:\TIONS 10 C01 DATA VAT IONS 1'13C;1
. '. 241
APR I', 12 12
1 9 341 13 . 24 243
14 24 267
3 6 359 15 ' 24 301
7 12 430 16 3 287
9 21 347 17 6 282
22 12 307 18 6 272
23 24 306 19 21 241
24 18 311 20 24 219
25 24 320 21 15 200
26 6 353 22 15 229
29 6 346 23 12 264
30 12 348 26 24 206
- . 343
AVERAGE 27 15 325
28 21 319
HA'f . 29 24 311
2 I 24 342 30 18 244
5 21 279 31 24 241
6 20 292 AVERAGE. . 265
7 24 307 -
8 24 312 AUGUSr
9 15 297 1 24 235
10 24 286 2 24. 205
11 21 289 3 24 198
12 15 322 4 24 221
13 24 329 5 24 223
14 24 294 6 24 255
15 15 302 7 24 324
AVERAGE' . 304 8 24 311
9 24 287
JUNe 10 24 220
20 20 298 II 9 247
23 19 255 12 4 236
24 12 289 13 24 243
25 23 250 14 24 263
26 24 243 15 18 279
27 . 24 265 16 24 246
28 24 226 17. 15 257
29 24 213 18 21 205
30 .. 24 203 19 24 191
AVERAGE . 249 20 24 . 223
21 24 188
JULY 22 24 190
2 21 261 23 3 231
3 11 266 24 9 277
5 10 280 25 14 ',/.')7
6 9 286 26 3 :WI
7 21 31 I 1- 27 21 24(i
8 18 307 28 24 244
9 24 250 29 15 257
10 15 256 31 24 225
11 21 I 254 . 245
. 271 -
AVERAGE
r
-------
..' .. .. --- ---" -. .. .. - --- - . -_. ..
OBSERVA-.: No. OF " TOTAL OaSERVA";' No.. OF TOTAL
TioN ' OaSER- O~NE . T'ON OBSER';' . OZ~NE
.DATA VATIONS I . CM .' DAn VATIO/IS .10 CM
.. --_.-. .-.-..----.
SEPTEMBER ". 21 21 . 328
2 24 " '. 22 3 284
261 23 3 260
3 5, 259 25 3 255
6 21 223 26 " 12 241
7 24 210 27 i2 272
'8 24 180 28 3 248
.' 9 24 170
.10 24 187 ,29 14 264
Il 13 208 30 15 246
12 io 229 31 9 312
- --_. _. -. , 253
'-13 24 228 AV!'RAGE
17 ,.. 18 220
-~----_.. -
18 24 205 NOVEMBER
19 ., Il 250 2 24 270
20 6 252
22 .'. 21 267 5 14 205
23 . 24 251 7 12 227
.8 ' 15 212
24 9 267
25 24 254 9 . 9 244
26 13 262 11 9 251
','27 10 246 12. 10 233
28 ' 12 235 13 . 9 216
30 . 9 296 15 .6 295
. 18 12 218
. - -. . - ..n'- -" . , 233 19 8 227
AVERAGE 20 15 222
. --.- -- _.:.-- - 22 6 279
OCTOBER .24 9 276
1 17 304 26 .3 263
2 23 240 27 3 276
3 24 225 28 12 288
4 24 246, 29 15 335
5 21 219 30 ' . 12 319
6 12 ' . 209 ,256
AVERA'3E
8 12 199 '
9 18 238 ; . .--- . --- - ;
12 20 269 DECEMBER
13 14 246 4 12 292
14 24 218 6' 12 273
15 6 225 9 6 320
16 9 240 12 6 281
. 17 24 242 14 12 283
18 9 2..39 2..5 12 y~
20 18 3Q5 't'2'i A,.", ' tl:
: --: .
I
I
I
"
... "
"
I
I'
I
. .
I "I
I
!
. . A
. H~-ISS 1S:(80~ ~' LAT., $60" Ea'lO'lG,)
MARCH
\
, "
28 '
29
30
I :. 1~
. AVERAGE
3
4
5
11
442
447
397
, , " , 429
10.
'-
- 272 -
APRI!.
12
15 .
6
6
I
, '
430
366
328
434
-------
I
. -.. _.- ._~. --~_._-------_._._._-- ._,--_._--_._-~--_.'--,--'--_._---,-._~-
'---, ~O~ ~ - -,.- . --- -...- .
OB6ERVA.. I 0' TOTAL : OBSERY"," ; NO. OF TOTAL
rlON . I 'OBSER-- OZONE ' rlOH ; OBSER- OlONE
DATA! . YAr I OIlS l03cl4 DATA VATIONS IO'3CH
., "
12 15 468 25 24 380
13 12 410 26 21 365
15 11 425 28 15 412.
18 21 405 30 27 407
22 15 385 AnRAGE . 401
. .
1 .
2
3
4
5
6
7
8
9
Il
12
13
}4
Ifii'
13>
zr
22
23
25
26
27
28
29
-~---
: FEBRUARY;
. . 0 )
YAKUTSK' (62°~ N. LAT., 129 45' E. LOnGo
-----=-----
f
12
15
16. .
17
18
19
20
21 .
23
25
28
3
'9
12
12
12
4
9
9
14
17 .
18 .
AveRAGE I'
lURCH
6
9
6
18
12
12
21
15
18
15
. 17
, 21
Ui
di
(Ii
- 15
18.
12
3
9
9
18
2
AVERAeE
,600
590
550
451
'469
508
. 491
430
431
468
418
.' . 487-. ,.
556
571
580
568
591
566
557
530
452
493
522
til1
~
m
56(}1
559'
557
542
550
589
. 573
515
555
.550 :
I
4
7
8
.9
10
Il
12
13
15
16 .
17
18
19
20
21
22
23
24
26
27
28
6
7
8
9
11
12
- 273
A PR I L
18
11
; 18
2
15
21
12
19
15
12
15 .
33
21
21
9
6
12
9
21
21
21
8
AYERAGE
t _-:--
.. HAY
3 21
4 15
.5 ; J5
~ ,I I M!
~ ...1 .' 1~
A\i!t#',(@
-----.
Julie
21
8
12
3
15 .
3
602
627
649
652
615
573
537
555
515
526
520
514
543
537
542
529.
411
455
'432
507
487
,474
. 539
; 544
529
I
. 52if'
290
350
302
345 .
299
317
-------
1~-"-.lI~.._~-----:
"
,. '
..
,. ", ~ " - .-- -.... _. .-~_.--.- -,
;OBSERVA-e NOa OF 'TOTAL', oaSERVA- No. OF TOUt.
TIOH OBSER.. OZ~NE TION oaSER.. oZ3r1e
DAn VATIONS 10 CM DATA ' 'lATIONS:' 10 CM
.-- .
13 9 302 AUGUST
14 12 274 7 18 246,
15 12 319 11 12 .. 303
17 21 293 12 21 240
18 18 252 13 ,9 267
19 3 273 14 12 233
, 20 .18 268 17 6 350
21 21 257 21 12 234
22 12 274 22 3 240
24 21 234 23 .3 237
25 18 ' ' 228 27 21 201
,26 15 230 ' 28 21 215
27 18 215 29 ' 6 218
28 15 198 30 " 18 212
29 6 233 ..,
.. , AVERAGE .246
- --- -.
I\VERAGE .274 ~ - - -
SEPTEi1BER "
-, i 12 298
JULY 3 18 245
4 21 242
5 10 293
1 15 232 6 6 263 '
2 6 " 270 ,9 18 257
3 12 227 10 18 261
4 12' 269 11 12 275
5 21 263 12 9 '-. 226
6 9 255 13 9 231
8 15 280 14 . 18 245
9 12 248 15 18 247
10 6 253 '20 18 249,
12 21 245 21 15 265
13 15 234 22 18 257
14 12 229 24 3 248
15 18 ' 248 27 21 215
16 12 222 28 6 248
, 17 21 ' 214 29 3 280
18 15 218
19 18 219 AVERAGE . 255
20 6 254 ...-
21 15 260 OCT\JBER
22 15 200
23 15 284 2 12 338
25 21 201 3 18 352
26 21 228 .. 4 18 347
28 12 233 5 11 326
29 18 207 6 13 296
30 18 208 7 7 347
31 15 279 8 12 .306
AYf!RAGE . 240 ,(
274 -
-------
" .
. .
" . -
'.
"~
.~
"" ..' -
OBSERVAo- No. OF . TOTAL 08SERVA- NO. Ot TOT AL
OBSER- O~III; TION GaSER- '
n:>:r O!~~IE
IJATA VATIOI/S I Cl1 . DATA VATIOIIS 10 eM
9 9 300 26 12 310
15 6 290 AVERAGE . 293
17" " 12 276
18 8 370 NOV~MBE!I
21 15 274
24 12 287 2 I.. 2 1 320
25 6 .298 5 9 374
AVERAGE .347
275 -
------- |