JPRS ID: 9884 USSR REPORT METEORLOGY AND HYDROLOGY NO.3, MARCH 1981

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000400030065-3
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 164
• Document Creation Date: November 1, 2016
• Sequence Number: 65
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-40850R040400034065-3
FOR OFF[CIAL USE ONLY
JPRS L/9884
31 July 1981
USSR Report j
METEOROLOGY AND HYDROLOGY
No. 3, March 1981
FBIS FOREIGN BROADCAST INFORMATION SERVICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000400030065-3
NOTE
JPRS publications contain information primarily from foreign
newspapers, periodicals and books, but also from news agency
transmissions and broadcasts. Materials from toreign-language
sources are translated; those from English-language sources
are transcribed or reprinted, with the original phrasing and
other characteristics retained.
Headlines, editorial reports, and material enclosed in brackets
are supplied by JPRS. Processing indicators such as [Text]
or [Excerpt] in the first line of each item, or following the
last line of a brief, indicate how the original information was
processed. Where no processing indicator is given, the infor-
mation was summarized or extracted.
Unfamiliar names rendered phonetically or transliterated are
enclosed in parentheses. Words or names prtceded by a ques-
tion mark and enclosed in parentheses were not clear in the
original but have been supplied as appropriate in context.
Other unattributed parenthetical notes with in the body of an
item originate with the source. Times within items are as
given by source.
The contents of this publication in no way represent the poli-
cies, views or at.titudes of the U.S. Government.
COPYRIGHT LAWS AND REGULA.TIONS GOVERNING OWNERSHIP OF
MATERIALS REPRODUCED HEREIN REQUIRE THAT DISSEMINATION
OF THIS PUBLICATION BE RESTRICTED FOR OFFICIAL USE ONLY.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
JPRS L/9884
31 July 1981
. USSR REPORT
METEOROLOGY AND HYDROLOGY
No. 3, March 1981
Translation of the Russian-language monthly journal METEOROLOGIYA I
GIDROLOGIYA published in Moscow by uidrometeoizdat.
CONTENTS
Structure and Conditions of Development of Thunderstorm Clouds 1
Anthropogezic Changes in Atmospheric C02 Concentration During the Next Fifty
Years .........s... 19
Experimental Investigation of the Correlation Between the Meteorological
Range of Visibility and Altitude of the Lower Cloud Boundary 36
Running Control and Evaluation of Alternative Models 45
Determination of Turbulent Diffusion Coefficients 52
Computation of Transport of Substances Contaminating the Atmosphere 60
Two-Frequency Microwave Radiometric Method for Determining Wind Velocity
From a Satellite 65
Diagnostic Model of Water and Ice Circulation in the Arctic Basin...'........... . 76
Complex Prediction of the Interanr.�.:al Variability of Inflow of North Sea Waters
Into the Baltic According to Shore Observation Data........................... 86
Status of the Study of the Element-Salt Composition of Sea Waters Using
Nuclear-Physical Methods...................................................... 91
Use of Radar Data in a Hydrodynamic Model of Rainwater Runoff With Distributed
Parameters 99
Experimental Substantiation of Computations of the Rate of Water Flow Along the
Ciil.tivated Surface of Slopes................................................. 107
- a- [III - USSR - 33 S&T FOUO]
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Determination of Damage to Cotton Plants in Different Development Stages
Resulting From Hailfalls 112
Improvement in the Method for Predicting the Intenaity and Quantity of
Precipitation in a Warm Period 119
Possibility of Prediction of L ightning Discharges 125
Determination of Filtration Co efficients of Cohesive Soils in a Frozen State
Through Their Kinetic Speci_fic Surface 128
Evaluation of Applicability of Different Methods for Determining
Evaparation From a Water Surface in a Zone of Hummocked Swamps............ 134
Reconci.ling of Mesospheric Temperature Values Measured by Different Rocket
Sounding Systems 138
Radio Device of a System for Thermal Sounding of the Atmosphere by the
Radioacoustic Sounding Metho d 146
Review of Monograph by V. R. Alekseyev and B. L. Sokolov: PolevyyE
Issledovaniya Naledey (Field Investigations of Ice Encrustations),
Leningrad, Gidrometeoizdat, 1980 151
Fiftieth Anniversary of the Leningrad Hydrometeorological Institute....... 154
Notes From Abroad........................................................... 157
Obituary of Maksim Sawich Kulik (1907-1980) 159
- b -
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R044400030065-3
FOR OF'FICIAL USE ONLY
UDC 551.(594.21:576.1)
STRUCTURE AND CONDITIONS OF DEVELOPMENT OF THUNDERSTORM CLOUDS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 5-17
[Article by I. M. Imyanitov, Main Geophysical Observatory, manuscript received
2 3u1 80]
- [Text] Abstract: Investigations of recent years have demon-
strated that the electric properties of thunderstorm
clouds differ appreciably from those postulated only
10 years ago. It has been established that thunder-
storm phenomena occur in stratiform clouds, that there
- is a high level of electric losses in active thunder-
storm clouds, and that there are electric currents
exceeding by orders of magnitude those which were
surmised. It was postulated that the critical condi-
tions for the occurrence of a discharge include en-
ergy factors. Existing models were inadequate for
explaining the observed electric characteristics. Two
schemes are presented: for the process of electrifica-
tion of a part of the clouds, based on the contact el-
ectrification process, and macroelectrification of the
clouds, based on the effect of falling of charged pre-
cipitation. Both schemes make it possible to explain
the effecte discovered during the last decade.
Man's acquaintanceship with thunderstorms can be divided into four stages. In the
first stage man attempted to answer the question "what is this?"; in the second
stage "how?"; in the third stage "why?"; and finally, now, in the fourth stage
"how to control it?" However, the answers to the preceding two questions are by
no means complete.
A solution of the problem of how thunderstorm clouds are structured and why thun-
derstorm phenomena develop should also be the basis for predicting thunderstorms
and for foreseeing the local and global consequences of industrial activity and
for understanding the physics of a high-voltage discharge in aerosol clouds. A1-
though no other phenomenon is expressed so sharply in the atmosphere as a thun-
derstorm in relation to the dependence of man on natural forces with the develop-
ment of technology and industry, the principal electric characteristics of clouds
have become known, and indeed, only in a general way, only now. Accordingly, there
are mutually exclusive theories of thunderstorm electricity [22] because thunder-
storms develop under conditions which meteorologists not only do not,consider dan-
gerous for the development of thunderstorms, but even conditions which are hypo-
thetically impossible for their occurrence [4].
1
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
1. Electric Model of Thunderstorm Cloud
A s.imple model of the electric structure of a thunderstorm cloud as formulated by
Simpson (for example, see [9, 14, 17] was not capable of explaining the new facts
(Table 1). This was also true of the Kuettner model (for example, see [171).
' A new model was developed and an attempt has been made to examine its statistical
and dynamic aspects [9-27]. A model of a thunderstorm cloud in the maturity stage
is shown in Fig. 1. As long as the stage of a well-developed cumulus cloud pre-
vails the electric charge 3 is absent and the charges 1, 2 and 4 are several or-
ders of magnitude less than those existing in the maturity stage. In the period
- of transition from the Cu cong stage to the stage of mature Cb, conventionally
- designated Cu cong-Cb, the charge 2 is small in comparison with the charge 1 and
therefore the field strength over the cloud has a different direction than in the
maturity stage. The charge 3 is associated for the most part with the charge of
falling precipitation. In the decay stage the charge 3 is considerably decreased
and the charges 1 and 2 become less than in the preceding stage.
Fig. 1. Static model of thimderstorm cloud
in maturity stage. 1) main electric charge
in cloud; 2) charge arising under influ-
ence of atmospheric conductivity; 3) pre-
cipitation charge; 4) charge of electric
inhomogeneities in cloud; 5) column of
unit section.
99,
93,~
Si
Fig. 2. Probability 1- 4~(E) of oc-
currence of fields with the strength
E less than the stipulated level in
clouds of different species.
The characteristics of the electric field outside the cloud are determined for the
most gart by the charges l, 2, 3. The charqe+s in zones of inhomogeneity 4 can cre-
ate local field strengths exceeding by an order of magnitude those created by the
2
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
1- 2(f) ?00', AG .45 Ns rA

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-04850R000400030065-3
FOR OFFICIAL USE ONLY
~
0 ~
"
0
c~
a
i i a i~
0010 o~
a
o
o
o
N
~ o
~
~ ~ l~,~~i
w
m
uw
m 0
E,
~
w
w
w
N
~0
3
a~
o
�
~
wooa~

E~
c~
a
iw
ua P-1 ,a oo
a~
q
H o
~
y,~
~ > ~
~
m
y
0
t
d
.C
2
U
'C
N
1. f
.l
cd I
~ N
?
3
~
~
�
�
.
cd
~
O
~ P~
a
i
u: a a
a
i
s+
~n
oo
u
~
~d
-
z
a o w
0 0
a
Cn
r+ o i
o u
-
00 r-I 004
O.C
O
Z
0 O
t OM
v
H
fr' U0 ~.1
Gl V
Cl
M
U1 r-Iri
cn
u1
Ci
-ri ~
~
O i
O i i ~1 ~D ~C
O
O
O
'a
V-~ O O
~
cd
i
~
~ O
� O O
-i
i
O
000
-I
-1
V-I
N
rI
N lU W
O b0 11
t
7-1
U
-ri r-1
r
r
r-I r
r
N
C!
b0 c0
~
,p
1
7,
U
N .a
ry~
w
~
r-I G! Gl
U
r-1
O
c0 u 1+
t
Ol
v
~
O
?C 0
b0
'C!
0
a4%
0
~
t

r

c
0 LL tA
V
~
fi)
4-1
0
m
~
U
~
~
N
a

b0
r.
4
1
Lf1
'b
O Gl L+
cl! 'b
rl
Z
Ctl
U p cd
~
1
~
~
0
0
H
20
I
v0
1
~
O
~
A
N
-H U
~
~
-
H
I
W
U
M I
rl
j
I
N
0
~
~
~O
O
~
O
O
O
O
OI
,
C
~
l
-i
*
-i
O O
-1 c.� o�
r-i
i-1
c`�
N
!A
N
r
r
r
r
.r{
a.
1+
C)
'b
a
v
�
s
~
~
m
Q1
�

b
"H
u
ao
u
o
o
v
~
~
~ a

�
~
~
u
cx3
u
c~
.
p
,
,
.
,
u
v
q
-H
M
w
~ o0
u
a
D+
(n
~
N
.n
q
al
N.c
-H
.C
3
.C
3
uf
4)
o H
D
Cd ~
H
W
N
O
o
-H
~
o
r-i
D
~
3
41
U
~ ~ q .
C
~
i.1
C7
N ft
'H iJ
C'i
C3
41
Gl 4+" C C~
y
-W
c0
t1
J~ 01
N cd
ai
hi
q
b0 C! ~-I
U
cd
r-i
rl
r-i
C
f:
H~~ TJ
-H
~
+1
~
�1
~
q
a
~
~ o
~
~
~
~
a
o
o
0
0
q
00
H
'
~
01
u
CJ
u
N
-H
C~
~
~ N
Q oo
'U p C S
~ o
D N
0
N
C:
c
a) ,i ~
0) W U
G~
u
"4
C'
f~ ~{d
0 0 N
~
~
r
~
+
^4-1
4-1
~
~ e"I ~1 0
i.~
~
11
C~ ~ Ul
rl ~4J
fA G
~ C
~ JJ
,2I 4J iJ
u
a.' 0
v

a
-H
b
u,-i m
9:
-I !
~
~n
a
j
m
oo ,-i u
a! 0
0
cd
u O
~
00
~
~
t~

~ w
"4

~ b
a
[
~
~
~r
q
'H
~
m b
q v)
~r
'
~ 1'~
~1
N
'I"
"
cd
74
O
!
U'b ",a.t N

O o.-I
W
N 1+
)
M
b0 ul
C+ r-I
1
M
0 o N r1
a
10
.~c
4. ~t
u

u
0 3
H
.-i
w~
a) ~ 4+ ,H
i w
-i oIv
w a)
rn -H
.n a
~
a) ~
o ~e
~ a)
oo 4
w
~d 1+ ~d
v,
u
,
r
~
4-4
~
.
0)
w
U 4-1
ri rl
4-1
O
`rl
01
'3 ~-i w
u l0 oD
Cl C T'~ 00 OD G'+
N H " ti cd O
0o
fa
Gl
b0
~
.0
~
L'+ O 44 r-I

i-1 1~1
y t~
C! 'J 01
b0 3~+ V C1 rl
H N~
1
ct1
1-1
Ql
t0 1+
t!~ [A
1-
(1)
rl 'C7
O b0
"
~
N N ~ ~
3
'
'U
N
'
l
;
~
.
R)F.3
l
G
00
Ey
1
i
l
41 N
~ k3
D
U cJ
Cl
00
cd
aJ ~
U
.
,
1
l R)
:
a
J f:
N
l f: ~
-
r
J C
l
c11

tA G
.
r
U C7
-
r
cd i:

H g v
G)
-1
41
O
1
N
R
f
N
r
r
O
l
D O
r
rl
C
O`a' ~rl
.
rl
r
lS~
f
'
r
.
~
U
J
W N
V~
7U
c
vz
~
v a
~
wz cd+
3
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R400404030065-3
FOR OFFICIAL USE ONLY
N O1
~
w Cd
p rn
H ~
~
H
0
-W
cd
~
~
~
~
'b
~
O
~
U
~
r!
N
N
DO
p
cd
~
U
r-I
cd
a
v
q
~
H
a
w
0
~
a~
0
r-i
ca
>
Id
a~
~
b
o m
�
�
c{~ r-I U1
%D .7 N
N
U
~
u
U
q
~ c
1
1
1
rn x
�
�
u ,H u

~
~ w
(a C! 0$40 N tn ~"1 ~O V1
A p v ctl
'b
c~d
~
~
C1
00 r-I
N~
cb
16N
I
G1 ~
Sa N
U
'b
p cd
C)
~ y
a~
U 00
1~ cd
U U
N
~ T
a~ a g
p a ~E
m w �
~ Cd a;
~
a~
a~
00
w
b
w
~
~
~
~rl
JJ
ca
a
c''1 %t O% O
N N M ~Y
Ln o 0 0
M tl% M O
N N
z
0
~n
z
z
to
i
0
0
tn
U)
%O
Ln
cn
M
4
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
t1
FOR OFFICIAL USE ONLY
main charges. The value of the main charges 1 and 2 is different for different
clouds and has a tendency, on the average, to increase toward the equator (Table
2).
The probability of encountering field strength values less than the average value
in clouds of the temperate latitudes is shown in Fig. 2, taken from [9]. This
probability is approximated satisfactorily by a log-normal distribution. The
principal electric macrocharacteristics of clouds are given in Table 1. In the an-
alysis of the data in Table 1 it must be taken into account that it gives esti-
mated data since the precise parameterization of the properties of clouds, a dis-
crimination of the peculiarities introduced by the specifics of aerosynoptic pro-
cesses and the influence of characteristic physiographic conditions require the
carrying out of prolonged systematic measurements whlch foi the time being have
not been made.
In summarizing what has been stated above, it should be noted that recently essen-
tially new data have been obtained cn the meteorological conditions for the oc-
currence of thunderstorm phenomena, on the electric structure of thunderstorm
clouds and its changes, on the principal electric charActeristics of clouds and
on the conditions for the occurrence of lightning in clouds. It was found that
the theories and hypotheses formulated up to the late 1960's are incapable of
explaining the observed phenomena, and the question naturally arose: what process-
es cause them?
2. Thunderstorm Cloud as an Electric Generator
We will represent a thunderstorm cloud as an electric generator in which particles,
having different velocities of motion, can be charged in large part with different
signs and thereafter, due to the difference in velocities, large volumes are form-
ed in space which are charged primarily by electricity of one sign or another. The
charges of these volumes should create electric fields whose strength and extent
are sufficiently great for the appearance of lightning. The answers to the ques-
tions of how the processes of micro- and macroelectrification occur and what the
term 0�adequacy" ot strength and extent for the occurrence of lightning means will
provide a key to the understanding of the true electric activity of a cloud.
2.1. Charging of particles in clouds. Among the many difficulties facing investi-
gators of the processes of charging of particles in clouds the greatest complex-
ities are probably caused by three circumstances: the meas ured charges on p recip-
itation particles were considerably greater than those arising in laboratory ex-
periments (for example, see [9, 14, 17, 28]); the charges of particles of both
clouds and precipitation could be both positive and negative; as a rule, if the
particles of one sign are encozm tered in a greater quantity, the mean value of
their charges is less than the mean value of the charges of particles present in
a lesser quantity (for example, see j14, 28]); finally, the conditions for the
charging of particles in different stages of cloud development and in its differ-
ent parts are so dissimilar that it seemed that tens of elementary electrifica-
tion processes developed, the relationship of whose activities was extremely dif-
ficult to determine (for example, see [9, 14, 28]). Investigators of any of these
processes almost invariably attempted to put "their" process as the basis for the
5
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/42/49: CIA-RDP82-04850R400404030065-3
FOR OFFICIAL USE ONLY
next theory of thunderstorm electricity, but it was invariably found that the el-
ectroproductivity of the processes was inadequate. For example, induction charging
with the breaking of a contact in an electric field was inadequately effective
because measurements in clouds [5, 24] indicated that the real field strength in
clouds cannot ensure the necessary charges. Investigations of the Workman and Rey-
nolds mechanism and its modifications, on which great hopes wPre also laid, led to
the unexpected conclusion that although this effect creates great potentials the
charges created by them are less than the charges appearing with the breaking of
the contact (rupture of the film) of ice and water particles [6]. Accordingly, M.
Brook [22] even proposed that mDratoria be imposed on new and old investigations
of the processes of elementary electrification wltil there was a more precise
clarification of the characteristics of clouds in order to avoid the appearance of
noncomparable hypotheses, theories and models. A solution to this problem was found
when it was possible to reduce all the principal processes of electrification of
particles in clouds to two (for example, see [9]):
charge exchange between particles separated in space (after the contact of dif-
ferent particles, during destruction oF a particle), so-called contact electrif-
ication;
transfer of charges from an ionized gas medium to particles.
Both processes can be substantially influenced by external strong electric fields,
changing both the electrification at the contact and the ionization level and cre-
ating reverse positive and negative feedbacks having a significant effect on the
electrification of particles.
Contact electrification arises in every case when there is a break in the contact
between two particles, during the destruction of particles, detachment of a water
~film from solid particles, escape of splinters from a freezing drop and similar
effects (for example, see [91).
If a neutral particle collides with another neutral particle and the difference in
their eiectrochemicai POLCIIi.iaia i5 Y1 2, fiiL8i vic8i:iiag oi ~lic :,CI'i~uC~
charges +ql and -ql arise on the particies
Iql l- 901, 2 cl, 2, ~1)
where c11 2 is the mutual capacitance of the particles at the time of breaking of
the electrical contact between them [3].
The difference in the eZectrochemical potentials 5P1'2 is dependent on the chem-
ical composition of the particles, their phase stateand surface properties; the
capacitance c1,2 is dependent on the geometry of the particles, their elasticity
properties and on the concentration and mobility of the charge carriers in the
particles and also on temperature. Under these conditions virtually any breaking
of the contact of particles results in their elec;trification, which will be the
greater the more diverse the particles. A paradox appears: it is more difficult
to explain why the particles in an aerosol cloud are uncharged than to explain their
charging.
If a large particle collides with small particles, similar in their properties,
the charge q of the large particle will increase exponentially. We will assume as
a simplification that the large and small particles are spherical and their radii
6
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFE'ICIAL USE ONI.Y
are equal to R and r respectively, R9 r, and the difference in their electrochemical
potentials is T1,2. Then the charge acquired by the large particle is
__-Q = - : _ -
!R t
1,5 r ~j~ qI ( I- e (2)
The limiting charge Qlim arising on the large particle will be equal to [2, 31
-R_ a,
qlim _Zi .~aCi. 2-. z q 2a)
I r . (
In this case cl 2 N Ar, where A= 3-8N 5, and ql is the charge imparted to the par-
ticle at the time uf the first contact. The relaxation time of the process will be
equal to 'CN R2/r2N, where N is the number of collisions of large and small par-
ticles per unit time.
q�lUuKn
= M~G
� MgM~
0 o Zn�C
~ Ag-C
2a a Be-E
0
10
0 ~ yo � y 3BeV
0
x
h\
�10
Fig. 3. Dependence of charge with breaking of contact of two particles having an
electric potential difference dT.
As a result of this effect, in clouds during collisions of precipitation particles
and cloud droplets a charge can be formed on the.first which exceeds by a factor of
102-104 that obtained in a single collision and uaually observed in the laboratory.
In the experiments of Buser and Aufdermaur [18] it was deuqnstrated that the charge
- imparted in this case in actuality is proportional to the difference in electro-
chemical potentials (Fig. 3). Formulas (2) and (3) have been confirmed in numerous
experiments [2, 11, 18]. Thus, collective effects can play a major role in the el-
ectrification of particles in a cloud.
If 9p1 2 and the sizes of particles are random values, the charges accumulated on
small and large particles are imparted with some distributions [11, 121. Ttao ex-
treme cases are possible: if large and small particles (precipitation particles

7
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFFIC[AL USE ONLY
and cloud particles) with equal probability acquire both positive and negative
charges no organized electrification will occur in the cloud, although the par-
ticles in it will be charged, but if the large particles are charged with one el-
ectric sign and the small particles are charged with another, organized electrif-
ication of the cloud arises which is intensified with the falling of precipita-
tion particles. In a general case, when the total values of the charges of large
and small particles are displaced in different directions from the zero value,
organized electrification is combined with the charging of particles of any sizes
with different electrlc signs.
The magnitude of the charge imparted during contact is dependent on the strength
of the external field E. However, the role of the induction mechanism of charging
occurs only in very strong fields. The fact is that induction charging is compar-
able with contact charging and exceeds it only when
y Eb q~ i. 2,
where J is the gap between the particles 1 and 2 at the time of breaking of the
contact and is the coefficient of intensification of the external field in the
gap between the particles; the'Y value is close to A, that is, y~* 5-10. Since for
liquid particles the value b= 10'7-10'4 cm, whereas .for solid particles it is
close to j = 10'8-10-6 cm, in fields ;0 104 V/cm the induction mechanism begins to
prevail over the contact mechanism. This effect is clearly manifested in experi-
ments [18] and explains the noncorrespondence between the Sartor, Zhiv and Levin
and other madels to tlie observed results.
Superposed on the contact electrification process is the capture of ions from the
air, usually lessening the charge acquired with the breaking of the contacts [12]
_ and also the charging of particles by the capture of ions from the surrounding air
(for example, see [9, 14, 17, 28]). However, these processes, even in the modern
interpretation [34], do not make it possible to obtain the charges required in the
cloud during the time.of its development.
The reduction of the numerous processes of particle electrification in clouds to
two made it possible to simplify the problem of experimental and theoretical model-
ing of the process of electrification of particles in clouds. However, such model-
ing must include allowance for the coagulation of particles, the influence of elec-
tric forces on the interaction of particles, the influence of inedium conductivity,
the change in cloud ionization in strong electric fields and allowance for the pos-
itive and negative feedbacks caused by these effects.
2.2. Accumulation of space charges in clouds. The literature describes two models
of the process of accumulation of electric charges in clo uds. In the first of these,
representing a development of the concepts of E1'ster and Geytel' and Wilson (for
example, see [9, 14, 17, 28]), the principal reason for the separation of regions
charged with the same sign in clouds is the falling of particles of prec3.pitation
charged with the same electric sign relative to smaller particles charged with the
other sign. The energy already accumulated in the cloud as a result of ascent of
its water part in the gravity field in this model is realized in an electric cloud
generator during the falling of particles. This model initially seemed to contra-
dict the facts because the charges of precipitation particles measured at the
8
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
earth's surface did not confirm the models of predominant charging of precipita-
tion. Only after measurements aloft in clouds and directly under them was
it foimd that in clouds there are extremely extended zones in which virtually all
the precipitation particles are charged with the same sign [5, 261 and only at the
ground level, due to the charge exchange effect, will the observed + and - charging
be observed.
In the second model, developed by Green and Vonnegut (for example, see [9, 17, 28]),
the charging of clouds is created due to convective transfer of the space charge
accunulating ne-3r the earth into a cloud by vertical air mvements; this charge,
passing through the cloud, settles on its droplets. The charge accumulating in the
cloud with its field causes an outflow of charges of the other sign toward the
boundaries of the cloud, primarily toward its upper part, creating a dipole elec-
tric structure. In this model the cloud acts as a singular electric filter. Th.e en-
ergy realized in the cloud electric generator appears due to vertical air movements.
Vonnegut, developing the Green scheme, postulated that the corona discharge of
pointed objects or features on the ground under the influence of the cloud field
leads to an intensified charging of the latter, which in turn intensifies the en-
try of space charges into the cloud, etc. The possible positive correlation between
the rate of entry of the charge into the cloud and the intensity of its generation
made it possible to regard the cloud as an induction type electrostatic machine.
' At the last, Fifth International Conference on Atmospheric Electricity the prin-
cipal reports [29, 30] represented the points of view set forth above and the
main problems were concentrated in them. Without dwelling on the details of the
discussed problems [22], we note only that the supporters of the Green -Vonnegut
model have pointed out that the model of charging of clouds by precipitation does
not explain the high value of the observed charges of precipitation, the noncoin-
cidence of the zone of falling of precipitation and the zune of generation of
lightning, thunderstorm phenomena in warm clouds and the appearance of a space
charge layer (electrode effect) in the upper part of the cloud. 'i'he opponents of
the Green-Vonnegut theory noted, in particular, that the sign of the charge accum-
- ulated by the cloud in the initial stage of the process is opposite that predicted
by theory.
~ In the experimei.ts carried out for solution of the problem of the possibilities of
the convective mechanj.sm, carried out under our direction [9, 10], it was demon-
strated that in well-developed cimmulus clouds with their tops at a level of 6 and
even 7 km, there is no appreciable electric charge as predicted by the convection
theory, the magnitude of the cloud charges is virtually not dependent on its
thickness, and only with the appearance of large droplets in the cloud 100~1m)
do the electric processes begin to intensify. At the same time the data 3.n Table 1
show that the absence of appreciable electrification in the Cu cong stage does not
make it possible to accumulate the observed charges as a result of convection in
the few minutes separating this stage from the thunderstorm stage. It can there-
fore be assumed that the first and foremost reason for the charging of a cloud is
- the falling of precipitation, although in the stage Cu cong-rCb and Cb air move-
ments can play a substantial role, although differing from that predicted by con-
- vection theory.
9
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Since the process of electrification of clouds in general, the same as for indi-
vidual particles, transpires in a conducting medium, in all computations of an
_ electric cloud generator it is necessary to take into acc:otmt the relationship of
intensity of charging and discharging processes. Failure to take this circumstance
into account made it impossihle to use the schemes and models of electrification
of clouds, which are being developed even recently [14, 22], and required the de-
velopment of a new moAel.
We will attempt to convey the principal features of this model using a simple
scheme. A complication of the model (for example, see [13]) leads to more precise
data, but deprives the model of graphic features.
- Maw we will examine the change with time t of the charge Q of a imit coliunn 5(see
Fig. 1) of an initially neutral cloud present in a conducting medium due to the
falling of precipitation whose particles acquire an electric charge as a result of
- the contact mechanism [9, 27]
dQ/dt - Ich - 41tO`aup + aalow + Aeff)R, (3)
where Ich is the current created by falling charged precipitation, Aau and Aal
are the atmospheric conductivities at the upper and lower boundaries o? the clouaw,
a eff is the effective conductivity in the cloud.
The electric current in the precipitation is
Ich IIi9i 1>1 ni 9i ~i(4)
; l
where ni, Vi and qi are the concentration and velocity of particles carrying a
charge of the same sign and nj, V1� and q~ are the corresponding notations for par-
_ ticles carrying a charge of a different sign.
The effective conductivity is '
'Neff - a0 + Pk, (5)
where ap is conductivity within the cloud, k is the mean value of the coefficient
of turbulent exchange within the cloud, P is a coefficient dependent on the geo-
- metry of the distribution of charges and turbulence within the cloud.
The first term on the right-hand side of equation (3) describes the electric cur-
rents charging the clouds; the second term describes its discharge currents Idis
through the upper and lower boundaries.
The solution of equation (3) is given for a one-dimensional case on the assumption
of a constancy of the currents Ich and Idis With time in the form
[3= ch] Q=~:,e~ ll, ` ~6)
where Z is the relaxation time,
'G = �
� 'R .X aup aiow Teff)
10
FOR OFFICIAL USE ONLY
(7)
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Under the conditions of the problem which we adopted equation (6) can be repre-
sented in the form - -
Qcloud � IchS'G 1, S : 11 - e r ~ , � (8)
where S is the cross-sectional area of the active part of the cloud and Qcloud
is the charge 1 in Fig. 1.
The upper charge 2(Fig. 1) is created by a conductivity current bringing from the
atmpsphere a charge which settles on the particles of the upper part of the cloud.
Since usually ~1au ~ a a oW, the corresponding charging.exerts no appreciable ef-
fect in the lower ~art o~ the cloud and the charge 2 of the cloud in the maturity
stage can become close in absolute value to the charge 1.
The electric moment M of the cloud in the maturity stage can accordingly be rep-
resented by the expression
M = OG hQcloud' (9)
where h i. cloud thickness andlX is a dimensionless coefficient showing by how
many times the cloud thickness exceeds the distance between the centers of the
charges 1 and 2(Table 2).
The lower charge 3(Fig. 1) is created by the charge of precipitation and is equal
to
qch nr 9i n; 9;� (10)
It is important to note that the charges 1 and 2 accumulated in the cloud can exc
ceed the charge 3 by many times. If there are significant changes in wind velo-
city with height this can lead to an appreciable displacement of the region of
lightning generation relative to the precipitation zone (for exanple, see [23]).
The stage of decline of electric activity in the cloud is characterized, in par-
ticular, by the fact that the cloud charging current Ich begins to decrease with
time. Usually this moment is close to the onset of attenuation of precipitation.
In this case the solution of (8) is somewhat modified: it contains a cofactor char-
acterizing this process. If the precipitation ceases suddenly, the charge and the
electric moment of the cloud decrease at a rate determined by the relaxation time
'L .
If electric simulations are used; the model resembles a Van Graff generator situat-
ed in a conducting meditmm. The role of a"conveyor belt," carrying the charges, is
played by precipitation and the role of an electrode is played by the cloud itself.
The considered model shows that the cloud electrification changes:
proportionally to the increase in the electric current of the precipitation,
_ which in turn changes proportional.ly to: a) the change in the charges on the in-
dividual precipitation particles, b) change in the difference in the concentrations
of precipitation particles charged with different signs, c) the increase in the
relative rate of falling of precipitation particles charged with different signs;
proportionally to the horizontal and vertical extents of the acttve part of the
cloud;
11
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
inversely proportionally to the increase in electric losses of the cloud, in
particular to: a) the increase in cloud conductivity, b) the increase in turbu-
lence wirhin the cloud, c) the increase in atmospheric conductivity at the
upper boundary of the cloud, dependent primarily on the altitude of the upper
boundary.
,i 1~C? �:CC . .
~ d g m3
Fig. 4. Computed dependence of field strength E in cloud on thickness H:(1) and
liquid-water content w(2) of the latter.
The model leads to the unexpected conclusion that one electrification process en-
sures a three-charge cloud structure. The adopted scheme for the electrification
of individual particles and allowance for collective effects ensure electric cur-
rents corresponding to the data in Table 1. The measured relaxation time values
indicated (for example, see [9]) that the principal losses arise in a thunder-
storm cloud and are related to a considerable degree to the turbulence in it. In
clouds developing over water areas the mean 'E values exceed by a factor greater
than 3 the 'G values characteristic for continental clouds, that is, thunderstorm
phenomena over the oceans can appear in clouds whose thickness and liquid-water
content are much less than for clouds over the land. If 'C in clouds is very small,
weak electrification processes (for example, in stratonimbus clouds) can lead to
the appearance of thunderstorms [27]. Similarly, even charging processes in warm
clouds, attenuated due to small fGl 2 can, if the losses in them are small, as
is the case in breeze clouds, lead to the appearance of a thunderstorm.
Figure 4 shows the correlations between the electrification of a cloud and its
thickness [22] and liquid-water content [9], computed using a refined mode1 [13].
Without dwelling in detail on the mechanism of the appearance of zones of inhomo-
geneities in clouds (Fig. 1) we note only that their existence is closely associ-
ated with air currents in clouds, with the circumstance that cloud turbulence
12
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
f- e%r. V / cm ,

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
cn
a
0
v
r
~
41
~
~
H
U
d
~
~
4
1
~I
G~
cr1
~
V
N Gl
N Cl
N
d Cl rl
N
4
~
CJ
4-1
co cC
co co
co
Cd co
N cI� N C-�
1.i
O
34 ~C
i4 i+
14
C W .'4 ?4
~
C
00
M .C .C
41 b0
U
41 v 00
Cl d cd
N
OD 01 N
Rd cd
G! 04
~
d d~ 3 3
x
d x~ x
~
a
i
d d x
d d 3
d
x
cd

U
bo
A
r-I
00
r-0i
'b O!
U U
u
r-i
~
~
l
r
v 0
O O
a~i
c"n 4-1
w q,�-~ c~o
co cd
~3 g
~-H
.a
ca
G
i i ~
o
~
�
a~ ~ i�
ri)
4
1
,H -H o
u
a ,H o
41
Cu H
u
Cd N aV+
0
a v,
a a H
4+ W Cd
41
Co
P. o ,H "
(L)
-H -rq
-H T+ u
b
,H u 41
o
au cd
u
H H
u u
0 c~
aG ~n
~c
w q w
0

oo a) a) w
o o
,-i a)
b
c
o o p
o 0o
w (n
r~ w$+ o
r-+ Ln
U H
~
.
u
uaa
u0
(C cn
-r+aa
u c1,
(1) '-3 u
cd
oo ,H
cn ,-i
a) d
u a+
i
.n
~--i
~ u) w u
U
a~
,H w w ~
w
a~+ ~ w
U)
co
-H m o a)
a~ (L) -H
ae x
o o a
o I
~ o
b
u
u ca w
~41
u v
w (L) 41
w aJ b
0
-H
-H 4 a
.~ccow
~-H
o>,;), ~m
41 o
o ~r.
)
o
co
a41
u~ o
'o 4
~ " d
c: "
a
o
r-i
o
w -H w
-M
" 41
0 ,i to H
(1) i
r-i ,H
U
rl
O'o N O
C 'd N
GJ Ul tA ~ OH
El
~ N M 1-1
o
w a
4+ a)
~
10 b
r 0 > cd
ao w w ~ u
a fl
o
~ u ~
m., aj
w

~ K~ y
w
0 a
1
2
0
o
a) -H 41 N
b aoo
o o
a) 41 4++1 b-H
r-i a)
0 c
4 r.
a~
w
:3 -H a)
r-i H
b 0 qx :1 41
v'o
a) P v
oo r-+ cn
o a ca
u 0
+ r-i H o0 0p
y4+
41 u (1) ,H
u
a)
(1) w
r-i
a)
a)
0 a) p
~
,H
z
a o
U
a~>
A
cno
H
4-1
~
`o
r-i

~
~u
co
~
~
~
�
o
-I a
1
~
c
d c
~
`a
+1
w
a
IL)
c
i
c
o
a
a
44
,a
~
~
~
~
o
aui r-I

u
~
+
41
i
a
~
u
~

i
41 bo
w
ii
~
~
m
o
o
,
i
a
w
ro
V a
oo r-i
�

44
~u

0N
c
d
y
o
N
4-4
U
O
U
tA ~
~
N
O
p~Ol
~
u
o�) o ~
o.
:3 a) 41
u-H
.
41
N
~
~
~
u
w

~
~ w
c
d

�
d c
d
c
u
C''
a
~
Z
i
w
u
)
u a

~
t�
i v
ti-
i
~
i ~
a
o
,-a
i
a) v
a
i .
c
+
a ~
a~
U
W
V
N~~
H
~ ~
~
N
0

44
W
~7
H
J
r. c
7
U O

N
W
.
rl
.
N
. .
C+1 ~7
. .
u1 tC
13
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
EOR OFFICIAL USE ONLY
leads to its homogeneity, creating local, relatively short-lived inhomogeneities.
As was demonstrated in [9], the lifetime of these inhomogeneities is sufficiently
great so that the electric fields in them considerably increase the rate of
growth of particles and the latter is reflected in the growth of the electric
field in the inhomogeneity. Lightning discharges can begin in zones of inhomogen-
eities. It was recently found [7] that the zone in which the discharge begins
should not only create a field with a strength greater than the critical value,
but also store an adequate electric energy (see Table 1). Accordingly, the pre-
vailing (for example, see [201) concept that a lightning discharge can begin with
a discharge on a particle (or several particles) is incorrect. The criticality of
field strength and energy determines the minimum extent of the region from which
the discharge can begin.
In general, the considered model eliminates the objections advanced against the
- theory of cloud electrification as a result of the falling of charged precipita-
tion [30]. For conversion from the examined model to a model of a real cloud it is
necessary to develop two- and then three-dimensional models, introduce time depen-
dences for the charging current and losses in a cloud and introduce feedbacks tak--
ing into account the reciprocal influence of electric forces and microphysical
processes in clouds.
3. Correlation Between Electric and Other Meteorological Characteristics of Clouds
A short-range forecast and diagnosis of the possibility of the appearance of thun-
derstorm phenomena in a specific cloud or small region is based on the measurement
of characteristics only indirectly related to the electric characteristics of a
_ thi.m derstorm and therefore it is desirable to be able to evaluate the reliability
of these correlations. On the other hand, a direct comparison of the electric char-
acteristics of different clouds without allowance for the conditions of their de-
velopment in application to a specific cloud electrification model can lead to
seemingly contradictory results.
According to the data published by Few, et al. [23], for example, projections on-
to the earth of regions of the falling of intensive precipitation and the genera-
tion of lightning are considerably displaced relative to one another, whereas ac-
cording to data published by Proctor [31] and Carter and Kidder [19] they coincide.
Above we noted the fundamental importance of solution of the problem of the rela-
tive location of these regions. The arising contradiction disappears when one con-
siders the genesis of the investigated clouds. If the theory of charging by precip-
itation is correct, for thunderstorms developing in the atmosphere, where there is
no appreciable change in wind velocity with altitude (for example, warm thunder-
storms), characteristic for the region of investigations [31], the regions of
falling of precipitation and generation of lightning should coincide. When there
is a considerable wind shear with altitude (for example, frontal, rapidly moving
thunderstorms characteristic for [19, 23]) these regions can differ considerably.
The appearance of thunderstorm phenomena in stratiform clouds without the presence
of masked convective clouds was impossible because there were no meteorological
conditions corresponding to tlie indirect indicators of a thunderstorm [27]. The
vigorous processes of enlargement of particles and tlie falling of precipitation in
cumulonimbus clouds still seem inexplicable.
14
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Now we will examine the physical principles of the developing relationships be-
tween electric and other meteorological characteristics of clouds from the point
of view of the considered model. The appearance of large electric charges on par-
ticles is associated with high values of the difference in electrochemical poten-
tials between them and a substantial difference in the size of cloud and precipita-
tion particles (see formulas (1) and (2)). The greatest potential differences
usually arise between water and ice particles, that is, in mixed clouds. This cir-
cumstance explains the usually observed relationship between thunderstorm pheno-
mena and the vertical extent of clouds (size oc precipttation particles) and the
presence of a well-developed zone with negative temperatures in clouds. However,
the organized electrification.of precipitation particles can be oUserved both in
warm clouds and in clouds with a small supercooled thickness; despite the fact that
the electric losses in clouds are stnall the occurrence of tliunderstorm phenomena in
the latter is not uncommon.
In order for significant charging currents to appear there must also be a gr.eat in-
tensity of precipitation (see formula (4)), which in turn is associated with the
great liquid-water content of the clouds and their vertical development. A correl-
ation between the intensity of precipitation and thunderstorm activity was already
noted in the last century (for example, see [14, 17, 28]). However, in the case of
small electric losses thunderstorm phenomena appear when there is relatively light
precipitation [27], but in the case of great losses they do not arise even in heavy
showers (for example, see [191). Therefore, probably, although radar criteria for
thunderstorms during the summer over the continents [15, 16] structurally coincide
with formulas (8)-(9), they cease ta be valid when deternd.ning thunderstorm condi-
tions during the cold season of the year, over water areas, etc. On the basis of
the data in this article it can be postulated that in addition to the necessity for
making an allowance for turbulent conductivity in the correlation between the in-
tensity of precipitation and its electrification it is also necessary to take into
account the possibility of a strong but relatively symmetric electrification of
particles (for example, see [29]) without appreciable organized electrification.
Table 3 gives a summary of evaluations of the reliability of the correlation of dif-
ferent meteorological elements with electric elements in clouds. It must be remem-
bered that the correlation in Table 3 is given for some mean conditions. In indi-
vidual regions and and seasons these correlations may be strengthened or weakened
due to the fact that in some mountainous region a thunderstorm with a high degree
of reliahility can correlate with the appearance of a small cloud over a certain
peak in the morning hours, although similar correlations are absent in other areas.
When evaluating the role of electric forces in the specifics of development of a
cloud it must be rememhered that the interaction of particles under the influence
of these forces radically changes the effectiveness of collision of particles, es-
necially small particles. According to the computations in [25, 32], the relative
coefficient of collision of particles measuring microns and tens and hundreds of
microns can increase by hundreds and thousands of times (attaining the value 1)
with fields and charges really existing in clouds in comparison with the value of
this coefficient in the absence of fields. In very strong fields the region of
weak (forbidden) coagulati.on virtually disappears. As indicated experimentally, the
values of the relative collision coefficient can attain even 20-30 [33]. Due to the
change in the rate of motion of particles in the electric field the length of the
15
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
colLmml of small particles with which a large particle collides during a given time
interval can increase hy several times [25]. In other words, under the influence
of electric forces the process of enlargement of particles for the most part is
reduced to the coagulation process. The processes of colli.sion, amalgamation and
separation of particles tranapire as they would in clouds without the influence of
electric forces but having a liquid-water content of hundreds, thousands and even
- more g/m3, that is, the rate of development of thunderstorm clouds is determined to
a great extent by the electric forces in them.
In the life of each science there are latent ("autumn" and "winter") periods when
facts are accumulated and compared. This is followed by a period of generalization
and interpretation (the "spring" and "sinnmer" of science), leading in the long run
to a new level of comprehension. Our understanding of thunderstorm processes has
prohably entered into this- second s.tage.
BIBLIOGRAPHY
1. Danilov, Yu. I., Yevteyev. B. F., Kazak, R. R., Kaprans, A. A. and Selvikyan,
Ya. V., "Investigation of Electrification of Bodies in Water Flows," TRUDY GGO
(Trans:actions of the Main Geophysical Observatory), No 350, 1977.
2. Imyanitov, I. M., ELEKTRIZATSIYA SAMOLETOV V OBLAKAKH I OSADKAKH (Electrifica-
tion of Aircraft in Clouds and Precipitation), Leningrad, Gidrometeoizdat,
1970.
3. Imyanitov, I. M., "On the Problem of the Electrostatic Chaxging Mechanism,"
DOKLADY AN SSSR (Reports of the USSR Academy of Sciences), Vol 121, No 1, Issue
93, 1958.
4. Ituyanitov, I. M., Yevteyev, B. F. and Kamaldina, I. I., 0 PRICHIIIAKH, PRIVOD-
YASHCHIKH I: PORAZHENIYU SAMOLETOV MOLNIYArII V KHOLODNOYE VREMYA GODA (Factors
Leading to the Damage of Aircraft by Lightning During the Cold Season of the
Year), Leningrad-Moscow, Gidrometeoizdat, 1976.
5. Imyanitov, I. M. and Mikhaylovskaya, V. V., "Experience in Investigating the
Charging of Precipitation Particles in the Free Atmosphere," TRUDY GGO, No 97,
1960.
6. Zmyanitov, I. M. and Mordovina, L. S., "Reason for the Appearance of Large
Potentials in the Process of Freezing of Some Aqueous Solutions," DOKLADY AN
SSSR, Vol 190, No 3, 1970.
7. Imyanitov, I. M., Pavlova, G. P., Ponomarev, Yu. F. and Chubarina, Ye. V., "An-
alysis of Conditions for the Damage of an Aircraft by an Atmospheric Electric
Discharge Outside Cumulonimbus Clouds," TRUDY GGO, No 424, 1979.
8. Imyanitov, I. M., Chubarina, Ye. V. and Shvarts, Ya. M., "Effect of Electric
Forces on Cloud Development," TRUDY GGO, No 301, 1974.
9. Imyanitov, I. M., Shvarts, Ya. M. and Chubarina, Ye. V., ELEKTRICHESTVO OBLAKOV
(Cloud Electricity), Leningrad, Gidrometeoizdat, 1971.
16
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
10. Markchev, N. T. and Fedchenko, A. M., "Possibility of Damage to Aircraft by
Electric Discharges in Nonthunderstorm Convective Clouds,"'NAUCHNO-TEKHN.
REFERAT. SBORNIK. SISTEMY NAVIGATSII, POSADKI I UPRAVLENIYA VOZD. DVIZH.
(Collection of Scientific and Technical Abstracts. Systems for Navigation,
Landing and Control of Air Traffic), No 1, Moscow, 1977.
11. Mordovina, L.S., "Electrification in a Flow of Aerosols," TRUDY GGO (Trans-
actions of the Main Geophysical Observatory), No 277, 1972.
12. Mordovina, L. S., "Random Electrification of Particles During Collisions,"
TRUDY GGO, No 301, 1974.
13. Mordovina, L. S., "Electrification of Stratonimbus Clouds as a Result of In-
teraction of Particles With One Another," TRUDY GGO, No 323, 1974.
14. Muchnik, V. M., FIZIKA GROZY (Thunderstorm Physics), Leningrad, Gidrometeoiz-
dat, 1975.
15. Sal'man, Ye. M., Gashina, S. B. and Divinskaya, B. Sh., "Radar Parameters of
Separation of Thunderstorm and Shower Activity," METEOROLOGIYA I GIDROLOGIYA
(Meteorology and Hydrology), No 4, 1969.
16. Stepanenko, V. D., RADIOLOKATSIYA V METEOROLOGII (Radar in Meteorology), Len-
ingrad, Gidrometeoizdat, 1973.
17. Chalmers, G. A., ATMOSFERNOYE ELEKTRICHESTVO (Atmospheric Electricity), Len-
ingrad, Gidrometeoizdat, 1974.
18. Buser, 0. and Aufdermaur, "Electrification of Collisions of Ice Particles on
Ice or Metal Targets," ELECTRICAL PROCESSES IN ATMOSPHERE, Darmstadt, Dietrich
Steinkopf, 1977.
19. Carte, A. E. and Kidder, R. E., "Lightning in Relation to Precipitation,"
J. ATMOS. TERREST. PHYS., Vol 39, 1977.
20. Crabb, J. A., Griffiths, R. F. and Latham, J., "Triggering of Lightning by Cor-
ona From Ice Hydrometeors or Colliding Raindrops," ELECTRIGAL PROCESSES IN AT-
MOSPHERE, Darmstadt, Dietrich Steinkopf, 1977.
21. Dolezalek, H., "Report on the Fifth Conference," ELECTRICAL PROCESSES-IN ATMO-
SPHERE, Darmstadt, Dietrich Steinkopf, 1977.
22. ELECTRICAL PROCESSES IN ATMOSPHERE, Darmstadt, Dietrich Steinkopf, 1977.
23. Few, A. A., Teer, T. L. and MacGorman, D. R., ADVANCES IN A DECADE OF THUNDER
RESEARCH, In [22].
24. Gaskell, W., Illingworth, A. J. and Latham, J., "Airborne Studies of Electric
Fields and the Charge and Size of Precipitation Elements in Thunderstorms,"
QUART. J. ROY. METEOROL. SOC., Vol 104, 1978.
17
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R044400030065-3
FOR OFFICIAL USE ONLY
25. Grover, S. N., "A Numerical Determination nf the Effect of Electric Fields
and Charges on the Efficiency With W;Zich Cloud Dropa and Small Raindrops
Collide With Aerosol Particles," Pageopd 114, 1976.
26. Gunn, R., "The Electrical Charge on Precipitation at Various Altitudes and
Its Relation to Thunderstorms," PHYS. REV., Vol 71, 1947.
27. Imyanitov, I. M., Evteev, B. F. and Kamaldina, I. I., "The Thunderstorm
Cloud," PLANETARY ELECTRODYNA14ICS, Gordon and Breach, Vol 1, New York, 1969.
28. Israel, lI., ATTiOSPHERIC ELECTRICITY, Volumes 1 and 2, Jerusalem, 1971, 1973.
29. Magono, Ch., "Precipitation Electricity of Thunderclouds and Showerclouds,"
in [22].
30. Moore, C. B., "An Assessment of Thunderstorm Electrification Mechanism," in
[22].
31. Proctor, D. E., "VIiF Radio Pictures of Lightning," in [22].
32. Schlamp, R. J., Grover, S. N. and Pruppacher, H. R., "A Numerical Investiva-
tion of the Effect of Electric Charges and Vertical External Electric F3:-ld
on the Collision Efficiency of Cloud Drops," J. ATMOS. SCI., Vol 33, 1976.
33. Smith, M. H., "The Influence of Electric Forces Upon Droplet Collection Ef-
ficiencies," in [22].
18
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FOR OFFICIAL USE ONLY
UDC 551.(510.4+588.7)
ANTHROPOGENIC CHANGES IN ATMOSPHERIC C02 CONCENTRATION DURING THE NEXT'FIFTY YEARS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 18-31
[Article by E. K. Byutner, doctor of physical and mathematical sciences, 0. K. Zakh-
arova and I. Ye. Turchinovich, candidates of physical and mathematical sciences,
and A. G. Lapenis, State Hydrological Institute, manuscript received 17 Jun 801
[Text] Abstract: The percentage distribution of car-
bon content was computed for three reservoirs
(biomass of the land, ocean and atmosphere)
for the next 50-year period (to 2030) for
determining the most probable rate of indus-
trial C02 emission. For use in computation
of C02 absorption by the ocean a correlation
was derived between the partial pressure of
C02 in the atmosphere and its content in the
upper quasihomogeneous layer of the ocean
(UCjHL) in a hroad range of temperatures. The
computations were made for a fixed mean global
temperature of the UQHL equal to 16.8�C. A de-
pendence of the C02 concentration in the atmo-
sphere on time was derived for the period up
to 2n30.
Introduction. The combustion of fossil fuel (petroleum, coal and natural gas) is an
essential part of modern power production. The supplies of or.ganic carbon accumul-
ated in the course of hundreds of millions of years are oxidized and enter the atmo-
sphere in the form of carbon dioxide. Data from the monitoring of COZ initiated in
1958 indicate a constant increase in the content of atmospheric C02 (see Fig. 1).
Over a 20-year period, from 1958 to 1978, the quantity of C02 increased from 315
million'1 to 335 million 1 parts in volume, that is by 42.5 Gt of carbon. [The
quantity 1 million 1 C02 is equivalent to 2.124�10B g of carbon in the atmosphere;
in the text which follows the masses of the reservoirs in all cases are given in
gigatons,.that is, in 1015 g of carbon.] If this figure is compared with the quan-
tity of COZ entering the atmosphere during these same years as a result of indus-
trial activity, it is found that about half of the anthropogenic carbon has remain-
ed in the atmosphere. The remaining quantity was distributed among the other carbon
- reservoirs of the so-called "mobile store," that is, the biomass and the ocean.
19
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FOR OFFICIAL USE ONI,Y
The total planetary carbon cycle includes three interrelated cycles which differ,
in particiil.ar, with respect to the intensity of the characteristic values of the
exchange fluxes, and accordingly, the time scales as well. In the first cycle, the
so-called cycle of carbon in the "mobile reserve," which includes the processes of
photosynthesis, respiration and oxidation of organic matter, as well as gas ex-
change between the atmosphere and ocean, the magnitude of the exchange fluxes is
tens Gt/year (see Fig. 2). Within the limits of the second cycle there is weather-
ing, dissolving and precipitation of carbonates with an intensity of the corres-
ponding fluxes from 0.1 to 1 Gt/year. Finally, the third cycle�, geological, in-
cludes the processes of carbon entry into the atmosphere from the earth's deep
layers and the accumulation of calcareous rocks; the rates of these processes in
order of magnitude are 10'.2 Gt/year and the changes in carbon content associated
with them have a characteristic time scale of 10~ years in the different reservoirs
j5, 151. The characteristic times of change in the content of carbon in the atmo-
sphere, the upper quasihomogeneous layer (UQHL) of the ocean and in the biomass
caith impairments in equilibrium in the first cycle are about 10 years, the time
scale during which the combustion of reserves Woo of fossil fuel available for ex-
ploitation is estimated at 200-300 years, and the W= value is considered to fall
in the range from 8 to 16 times the C02 content in the atmosphere. Accordingly,
impairments of equilibrium due to the combustion of fuel arise, in particular, in
the course of the first cycle. In the prediction of changes in the carbon content
in different reservoirs which can occur in the course of the next 50 years it is
possible to limit ourselves to an examination of the processes transpiring within
the limits of only the first cycle because the second cycle is characterized by
far greater inertia: its scale is estimated at 103-104 years [32].
H/'Ht million 1
340
~
J20 . '
� ~ ~
JDO ' - ~ � f
~ ~ iI9ti16o ~ . ~ d .
r-
c60 860 10: 1300 19i0 "940 7?60 195P
Fig. 1. Results of observations of atmospheric C02 conte:.c. From 1860 to 1960 [20]:
1) reliable data, 2) doubtf ul data, 3) data from aircraft observations. The solid
curve to 1958 represents dependence (9) for A= 8.2�10-4 from 1958 to 1978 mean
monitoring data for stations in the northern hemisphere.
The problem of the distribution of anthropogenic carbon dioxide between the atmo-
sphere, ocean and biomass with a stipulated temporal variation of the rate of in-
dustrial effluent and the total quantity W o0 of fossil fuel which can be burned
has been formulated and solved during recent years by a whole series of authors
[20, 21, 26, 30, 31, 34, 37, 40]. All these computations naturally contain the
parameters characterizing gas exchange among the principal reservoirs, as shown
in Fig. 2. The values of all these parameters are now known with some degree
20
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R044400030065-3
FOR OFFICIAL USF ONLY - -
of error. The type of equations used in different studies for describing the C02
cycle is also somewhat different.
Atmosphere
fl�7o
.
Ocean
I
~
x,� 617
F!�70
z; �665 ~i�TiN
.
IT
P, � 56
G,6,1 Pe
Y-W
Land
r BoOas
OR'
b�,i91oM
~
~
I
r-- - - - =
Livi~g
gi �J90D
yo �;000
Humus Y d_
1
Fig. 2. Carbon content in three reservoirs, Gt in atmosphere, biomass and ocean,
and also exchange fluxes between them, Gt/year.
OW/dt fm/zoB Gt/year
60
/
SO ~
1
~
f0 / ~
JO / ~
?0 ~ ~
/
/
10
_ t
~ f9B0 1990 ?000 1010 ?020
Fig. 3. Rates of industrial d:Cscharge of carbon Gt/year as function of time [26].
1) r= 0.0653; 2) r= 0.0453; 3) r= 0.0253; 4) most probable rate of discharge
- and its discharge in 2025. n= 0.5 (solid curves), n- 1(dashed curves).
21
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFFICIAL USE ONLY
The purpose of our study was more limited: it sets forth the results of computa-
tions of changes in atmospheric C02 concentration which can occur during the
next 50 years. In such a relatively short time interval the principal interaction
occurs between the atmosphere, biomass and upper quasihomogeneous layer of the
ocean and the deep ocean can be regarded as an infinitely large reservoir absorb-
ing excess carbon dioxi3e from the upper quasihomogeneous layer. In addition, a
whole series of nonlinear effects can be neglected and a slmplified system of
equations for the cycling of the carbon of the mobile reserve affords a possibil-
ity for evaluatir.g to what degree the inaccuracy in determining different para-
r.ieters can exert an effect on the results of computations.
Rate of Industriai Discharge
Figure 3 shocas the prosnostic functions d(d/dt for different scenarios of energy de-
velopment in the future in the form in which they were used in [20]. The vertical
line drawn at the level of the abscissa t= 2025 years was taken from [36] and
shows the extent to which the xange of possible rates of industrial discharge had
narrowed as a result of an analysis of data on the development of world energy
production during recent years. The dot on this curve designates the most prob-
able dW/dt value for the year 2025. The principal results of our computations
are represented by curve 4, which corresponds to the rate af industrial growth
r = w at = 4.5%/year
in 1975. A decrease in this value due to the rinite nature of the reserves of
available fossil fuel and changeover to other types of energy sources will still
have a weak effect in the next 50 years, although a rate of industrial discharge
in 2025 which is 1.8 times less than the rate corresponding to the constant value
r= 0.045 year-1 is predicted.
Even at the present time the intensity of the anthropogenic carbon dioxide source
dW/dt is 5 Gt/year and by 2025 will attain 26 Gt/year; accordingly, it should re-
sult in strong changes in the cycling of carbon of the mobile reserve.
Principal Reservoirs of Carbon in the Mobile Reserve
a) Atmosphere. The overwhelming part of carbon present in the atmosphere is present
there in the form of carbon dioxide. For the interval 1958-1978 Figure 1 shows the
mean values of the C02 concentration according to data for a number of stations
situated in the northern hemisphere. The concer.trations observed in the southern
hemisphere are less than in the northern hemisphere by 3-4 million 1 but have a
similar dependence on time [29]. Until 1958 there had been no monitoring of the
COZ content and there are only data from individual measurements which differ from
one another primarily due to the imperfection of the measurement method. As a re-
sult of analysis of such data Callender [29] concluded that the carbon dioxide
concentration in the mid-19th century was close to 290 million'1. A similar con-
clusion was drawn in [23]. In most models of the carbon cycle this figure is used
as the initial equilibrium concentration up to the beginning of anthropogenic dis-
turbances. It corresponds to an xp value of the atmospheric carbon reservair equal
to 617 Gt. It goes without saying that this figure cannot be regarded as absolute-
ly precise, but its mean square error probably falls i.n the range 5-8 milliori 1.
22
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
The estimates obtained on the basis of an analysis of data on the ratio of the iso-
topes 13C/12C in tree rings, from which substantially lesser xo values follow, are
based on an inadequately complete statistical material and can scarcely be assign-
ed a great weight.
b) Ocean. In the ocean inorganic carbon is present in the form of C02 in both dis-
solved gas and also the ions HC03 and C03.
The upper quasihomogeneous layer (UQHL) of the ocean on the average under station-
ary conditions should be in chemical equilibrium with atmospheric carbon dioxide.
� The total carbon content z~ in all three forms, that is, C02, HC03 and C03 (the
quantity nf nondissociated H2C03 is negligible) is determined by the temperature,
salinity and alkalinity of the water. If the values for the UQHL [1] t= 16.8�C,
s= 34.70/00, Alk = 2.44 meq/liter and a depth h= 77 m are used as the mean global
values, then with xo = 290 milliori 1 the zo~value is equal to 665 Gt. The remaining
ocean as a whole is 6-7% supersaturated with carbon dioxide in comparison with the
- UQHL as a result of activity of the ocean biomass. The basic functioning of bio-
mass occurs in the upper 200-m layer, but the dead organic remainr, settle into the
deeper layers, being oxidized as they descend. As a result of this process, the
ocean in general is doubly undersaturated with oxygen. This is known from both ex-
perimental observations [35] and from theoretical computations [8, 9]. Such an 02
undersaturation is about 6000 Gt and should correspond to a C02 supersaturation
of about 6000 6 12/32), that is, 2200 Gt. Accordingly, the total quantity of inor-
banic carbon z2 in the deep ocean is evidently close to 39 000 Gt [21]. The exist-
- ing uncertainty in this figure will not be reflected in the results of the computa-
tions as long as the deep ocean can be considered an infinitely great reserw ir
for the "runoff" of anthropogenic carbon dioxide. There are carbonates in the sed-
iments of the ocean floor. For the time being the interaction of ocean waters with
precipitated carbonates need not be taken into account because this is an inertial
process with a characteristic time scale not less than 103 years and the alkalinity
Alk of water can be considered constant. According to the data in [1, 14], the mean
global value of the Alk/C1 ratio is 0.12 and therefore with a salinity S equal to
34.70/0o we have Alk = 2.44 meq/liter.
The exchange flux FO between the ocean and the atmosphere can be represented in the
form
F�=VLSko(Pa-P::) =Fj -"FT, (1)
where the VL value has the dimensionality of velocity and accordingly is called
the characteristic rate of gas exchange through the discontinuity atmosphere-UQHL,
S is the area of the world ocean, kp is the coefficient of solubility of C02 in
sea water. The product kOpW is the limiting value of the C02 concentration in water
and kppW is the concentration of dissolved C02 gas in the UQHL, since the "partial
pressure of C02 in water pW by definition is the ratio of its concentration c to
the k0 value. As demonstrated in [12, 13], the difference pp - p has a latitudinal
and seasonal dependence and for the most part due to change in tte pw value. The
characteristic values of this difference are 30-40 million 1. Its mean annual
value is close to zero within the limits of accuracy of observational data and
the corresponding hydrochemical computations. The state of dynamic equilibrium be-
tween the UQHL and the atmosphere comes about by the evening-out of two oppositely
23
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
directed fluxes FT and F~ . Each of these can he obtained if the mean global value
of the VL parameter is known. Such a value has now been obtained on the basis of
data from laboratory experiments for determining the dependence of the VL parameter
on wind velocity with subsequent statistical averaging of this dependence using the
Probability distribution functions.for different wind velocity values and for the
surface of the world ocean. As a result, with storm effects taken into account it
was found that the mean global value VL is equal to 39�10-4 cm/sec. The Ff and F~
fluxes ara 70 Gt/year. The amplitude of the seasonal values of the exchange flux
FD is almost an order of magnitude less [2].
c) Biomass. The principal reservoir of organic carbon is in the continental biomass.
The ma.gnitude of the living biomass of the land, according to the presently most
commonly published estimates [28], is 827 Gt. Its greater part is in forests. The
data in [3] give a close figure: 2420 for dry organic matter (the mean molecular
weight is considered equal to 30). A detailed analysis of the presently available
evaluations of the y~ parameter and the possible reasons for the discrepancies be-
tween different authors was presented in [28]. But the conclusion of the author of
[28] that the figure 827 can be exaggerated by a factor of 2 is probably too crit-
ical. There is far greater uncertainty in estimates of the content of soil humus
~ y~. As demonstrated in [22], the figure used until recently, equal to 710 Gt, was
taken from publications dated 1915.
The mean global value y~, according to the estimates of the author of [22], is
3000�-500 Gt; according to the data in [28] it is half as great 1500 Gt. In the
3000 figure the author of [22] included 800 Gt of peat which is oxidated very slow-
ly. Accordingly, we used a y0 value equal to 2000 Gt, the same as in [34].
_ The mean annual primary productivity of t:ie continental phytomass, according to
the estimates in [34], is Po = 56 Gt/year. The data obtained in [7], 1410 Gt/year
For dry organic matter, virtually coincide with this Pp value. In a state of equil-
ibrium the Pp value should be compensated by expenditures on oxidation transpiring
in vital functioning processes with the dying of rapidly oxidizing organic matter
aiid with the decomposition of humus. A mean global value [10] was obtained for the
iast part of so-called "soil respiration"; it was 0.37 Pp. Accordingly,'the charac-
t2ristic time of oxidation of continental ltumus is
P0 = 2000/(0.37�56) = 96 years.
The corresponding "turnover time" for the living bioma.ss is
~(,1 = 800/56 = 14.5 years.
The living biomass of the ocean is primarily in the form of zooplankton [19] and
is approximately 20 Gt, that is, almost two orders of magnitude less than the con-
tinental biomass. The humus content in the ocean is reckoned at 1380 Gt [171, which
is not much less than the y~ value. The mean annual productivity of the ocean is
about half the productivity of continental plants, but there are many investigations
[39] according to which ocean productivity is limited by the nutrients received from
the depths, and not by the carbon dioxide, of which there are considerable quan-
tities in the ocean. On the basis of these considerations only the continental bio-
mass was included in our compu*_ation scheme.
24
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICfAL USE ONLY
Oceanic Reaction to Anthropogenic Disturbances
The equation descrihing oceanic reaction to changes in the atmospheric content of
carbon dioxide is naturally written in the following form:
h V L.S/ru lpn S(C - fv)� (2)
df
Here the term situated on the left-hand side describes the change in the content of
total inorganic carbon in the UQHL and on the right-hand side there are terms de-
scribing gas flows through the upper and lower boundaries of the UQHL respectively.
The principal feature of this equation is that gas exchange for COZ through the
upper boundary of ttie UQHL is proportional to the difference in the partial pres-
sures of COZ in the air pa and in the water pw (the pw value is determined as the
quotient by division of the concentration of dissolved C02 in water by the solubil-
ity coefficient ko). The pw value with stipulated temperature, salinity and alka-
linity values for sea water is unambiguously related to the total concentration of
inorganic carbon in the UQHL, that is, to the c value. The exchange flux of the
UQHL with depth is proportional to the deviation of c from the equilibrium value
c0 which prevailed in the UQHL prior to the onset of anthropogenic disturbances.
Accordin.&ly, .for the solution of equation (2) it is necessary to compute the func-
tion pw(c). Such computations were made from the known kp values, and also the
first kl and second k2 constants of dissociation of carbonic acid cited in [14]
on the basis of data from Liman, and a series of other known hydrochemical con-
stants. The results of computations with Alk = 2.44 meq/liter, S= 350/oo�and dif-
ferent UQHL temperature values are cited in Fig. 4. It follows from the graphs that
chemical equilibrium with the atmosphere, in which the pa value varies greatly, is
attained with an extremely small change in the content of inorg,anic carbon in the
UQHL. An increase in pa by a factor of 2 causes an increase in c by several peri
" cent. This is a well known f act specifically for the reason that the UQHL in its
time was known as a"bottleneck" through which there is an exchange of the carbon
dioxide between the atmosphere and the ocean. The characteristic time for the sett-
ing-in of chemical equilibrium, that is, the eq.uality of the pa and pw values be-
tween the atmosphere and the UQHL, does not exceed 1.2-2 years [24] and the time
interval in comptiting the mean C02 content trend is equal to 10 years6. According-
ly, with the use of equation (2) for computing the absorption of anthropo$enic C02
by the ocean it can be assumed that pa = pW, but the total concentzation c(t) of
carhon in the UQHL in this case must be taken from the results of precise computa-
tions of the hydrochemical relationships cited in Fig. 4 without recourse to lin-
earizations.
The curves in Fig. 4 demonstrate the dual role of temperature in the process of ex-
change of C02 between the ocean and atmosphere: with an increase in temperature
the transmissivity of the UQIiL as a buffer layer between the atmosphere and the
main layer of the ocean increases (the relative values (c - cp)/cp increase), but
its inherent capacity decreases. The computations indicated that the resultant in-
- fluence of temperature of the UQHL on the rate of absorption of C02 by the ocean
is small. A warmer UQHL leads to more intensive transfer of excess carbon dioxide
from the atmosphere into the ocean; a temper-ature increase by 1�C increases the
quantity of absorbed carbon dioxide hy 2%. However, it must be noted that such an
effect is obtained when the charac.teristics of exchange between the UQHL and the
underlying layers are not dependent on the temperature of the UQHL. It is evident
25
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
- that this is not the case and the total temperature effect can be computed only
- after obtaining the corresponding dependences for the characteristic rate Vmd of
gas exchange with depth. We carried out all subsequent computations for a fixed
mean global temperature of the UQHL equal to 16.8�C [18].
p ,.o. _i�-1
soo
400
.900
Fig. 4. Absolute (a) and relative (b) concentrations of total inorganic carbon (x-
axis) in ocean as a function of the partial pressure of C02 (y-axis) at a tempera-
ture of 5�C (1), 10�C (2), 15�C (3), 20�C (4).
~ The flux of excess total inorganic carbon from the UQHL in depth in equation (2)
is considered to be proportional to the deviation of the concentration c(t) from
the equilibrium value c0 which prevails in the UQHL at the time t= 0. Such an as-
simmption is usually ma.de in box models for describing the cycling of carbon between
different reservoirs. The principal feature of box models is the hypothesis that
the flux from the box i into the box j at each particular moment in time is pro-
portional to the deviation of the content of the impurity in the i-th box ni from
the equilibrium content N9 in this same box: Iii = kjini = kj i(Ni - NOi). The very
same applies to the flux from the box j into the box i: Ijk = kjinj. For fluxes
associated with photosynthesis and the biological release of C02 during the oxida-
tion of humus such a hypothesis is natural. However, in describing the processes
of transfer through the discontinuity between the ocean and the atmosphere and ex-
change between the UQHL and the deeper layers it is necessary to take into account
the conditional character of the box models. The kij coefficients are conditional
parameters and in a general case do not satisfy the principle of reciprocity of
the kinetic coefficients: in order that at equilibrium the content of the impurity
remain constant in each box it is necessary to satisfy the condition
k,1r'kjl = NO%N;',
which is compatible with the eque.lity kij = kji only when 112 = tJ~ . However, if kij~
kji, the total exchange flux Iij - Iji in the last analysis is not fotmd to be
proportional to the difference in the concentrations of the impurity near the dis-
continuity.
- Equation (2) contains the real, rather than conditional diffusion parameters VL and
Vmd�
26
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R044400030065-3
FOR OFFICIAL USE ONLY
The assumption that there is a proportionality hetween the flux in depth to the c-c0
- value is related to the fact that, first of all, the ocean below the UQHL ie con-
sidered an infinitely large reservoir in which the concentration retains an equil-
ibritan value, and second, since the activity of the oceanic biomass is not taken
into accowzt in the computations, this equilibrium value should be equal to cp.
In actuality, since the settling of particles of organic carbon from the surface
into the deep layers was excluded from our computations, the presence of a small
compensatory flux of total inorganic carbon, directed upward and caused by the
supersaturation of deep waters is also excluded.
For Vmd, which represents a value inverse of the diffusion resistance of the block-
ing stratified layer directly adjoining the lower bnundary of the UQHL, at the pres-
ent time the mean global value has not yet been ascertained. It is known that the
turbulence characteristics in this layer are highly dependent on the temperature
and salinity drops and have a very great spatial and seasonal variability.
A Vmd evaluation can be obtained using data on the relationship between the content
of radiocarbon 14C in the UQHL and in the deeper layers of the ocean obtained in
numerous measurements prior to the beginning of nuclear tests in the 1960's, as a
result of which the natural distribution of 14C was disrupted. These data are usual-
ly used in evaluations either of the coefficients describing gas exchange with the
atmosphere or the effective thickness of the upper layer of the ocean [40]. Since
the effective rate of gas exchange with the atmosphere VL was deterwined in [2] on
the basis of independent considerations, and the natural thickness of the UQHL is
used as the thickness of the upper layer of the ocean, the 14C distribution can be
used for evaluating the Vmd parameter (and also for checking the VL value).
Radiocarbon was formed in the atmosphere from nitrogen 14N under the influence of
cosmic rays with a mean intensity Q from 1.8 to 2.5 atoms/(cm2�sec) or from 0.58�
104 to 1.05�104 g/year. Its relative content by tnass in the atmosphere prior to
nuclear shots was 14C/12C = 1.24�10-12. The decay constant for radiocarbon a is
8000 years. The relative content
~ 1 _ ( 14C/12C)UQHL
(14C/12C )atm
(3)
is somewhat less than unity and the corresponding ratio in the deeper layers b 2
is less than j 1. With maintenance of a stationary 14C content this ensures a
continuous radiocarbon flux from the atmosphere compensating its radioactive decay
in the ocean.
Accordingly, the following relationships must be satisfied
i (I - ) = 1 (z. z, (4)
r, . z,
Vmd h k 0 G, - A .
27
- FOR OFFICIAL USE ONLY
(5)
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
According to data in [351, S 1 is equal to 0.96, 'S2 = 0.84. Formula (4) gives a
VL value equal to 64�10-4 cm/sec. This is somewhat greater than the value 39�10-4
cm/aec which we used, but it is necessary to take into account that the error in
computing the VL value in accordance with formula (4) is very great: a change in
a 1 by �0.02 changes the result by �40y.
Using formula (S) it is possible to compute the Vmd value. It is equal to 4.75 m/
year or 1.5�10-5 cm/sec.
Similar estimates (1.25�10-5 cm/sec) of vertical velocities Vmd in some circulation
models of the ocean are given in [9]. In [34], on the basis of data on the prop-
agation of tritiun in the ocean, which had been formed in the atmosphere as a re-
sult of nuclear tests, it was found that Vmd = 2.5�10-5 cm/sec with a thickness h
equal to 100 m. Evidently the effective global mean value of the parameter Vmd
for gas exchange processes fell within these limits, that is, from 1 to 2.5�10'5
cm/sec.
Biomass Reaction
The continental biomass, like the ocean, probably should also be az absorber of
anthropogenic C02. An increase in the intensity of photosynthesis with an increase
in the C02 concentration under laboratory (greenhouse) conditions is a well-estab-
lished fact.
In the time interval during which the system did not change greatly the equations
for the carbon content in living yl and dead biomass (humus) can be written in the
following way: -
f'� (1 -r ~~~1 n C
- - ,
~ -t',i, ~ t1(6)
dvd _ 0.37Y - ya
Jr -1 -d
(7)
where 'Gl and Zd are the characteristic times of relaxatiun of living biomass
and humus respectively, )8 is a coefficient characterizing the reaction of the bio-
mass to a change in the carbon dioxide content in the atmosphere. Under natural
conditions the value of this coefficient cannot be judged on the basis of data
from laboratory experiments since the increment of biomass can be limited by a
whole series of factors such as moistening, the quantity of nutrients, etc. How-
_ ever, in general the continental phytomass must increase the increment with an
increase in the x(t) function, since the present-day content of atmospheric car-
bon dioxide is considerably below the optimum for the photosynthesis process.
The methods for describing the biomass reaction adopted in different models [21,
31, 34] differ rather significantly from one another.
The system of equations (6) and (7) adopted in our computations is close to that
used in [34] with the difference that we assume the mean rate of oxidation of
soil hiunus is known on the basis of estimates of the global mean values of "soil
respiration" [10]. The same as in [34], equations (5) and (7) contain the assump-
tion of an invariability of the photosynthesizing green mass since the plants
strive to have an optimum surface.
- 28
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFiCIAL USE ONLY
The breakdown of biomass into living and dead with.the choice of a characteristic
"turnover" time for each of them seems to us to be justified in a problem in which
the biota is considered only from the point of view of its. properties as a C02
source and user. It is evident that only living hiomass can be a user of C02. In
general, both living and dead biomass serves as a source, but since it is the mean
annual primary productivity, and not total assimilation, which serves as an index
of intensity of use, the principal C02 source is soil humus and the intensity of
this source should he proportional to yd. With such a formulation the times Z 1
and 'G d are determined somewhat more reliably than with the breakdown of biomass
into two parts with small and large characteristic "turnover" times, as was done,
for example, in [26].
Naturally, the yi value consists of a whole set of populatiuns, to each of which it
is possible to assign its characteristic "turnover" time, but an elementary analy-
sis of the equation for yl indicates that imde r the condition of retention of the
total yol value and productivity Pp the details of such a breakdown exert little
effect on the computation results.
In solving the problem of biomass reaction to any changes in atmospheric gas compo-
sition it is necessary to take into account the fundamental circumstance that what-
ever increase there may be in the carbon dioxide concentration the biomass cannot
continue to increase indefinitely. There are restrictions on its increase due to
internal factors whic~ must be reflected in the equation by the appearance therein
of the term yl(yl - y0 The authors of [16] simply imposed a limitation of the mag-
nitude of the biomass. Keeling and Bacastow [26] introduced the function )3(t),
cahereas the authors of [34] obtained a limited increase in yl only due to a limita-
tion on the W(t) function.
The introduction of a quadratic term into equation (6) leads to a natural limita-
tion on the biomass due to internal factors, but for the time interval during
which the increment is small this limiting term can be discarded. In this case sys-
tem (6)-(7) contains only one unknown parameter, for whose determination it is pos-
sible to use the results of C02 monitoring.
y Use of Monitoring Data for Determining the p Coefficient
If AW is used to denote the total industrial discharge of C02 into the atmosphere
during the time interval Q t, the following expression must be satisfied
I W=A x+.1y' ;.I Vd~ ~2,-} A z2. (8)
I ~k
Az, A Z, J z, I J z2
I
Inter-
~w
a
A.~
I
~
'
val
+

I Vmd = I,2i~.10'S ~ Vmj _`?.3.10 '
I 8
I
,4 I
i
53
I
I
I 5
I
I
I
1 I
9
91
~4
2.5
,4
6
14
6
28
29
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Data on the A W value for the years 1860-1958 (I) and 1958-1978 (II), in accord-
ance with [36], are given in the table. This table gives the corresponding d x
values taken from the graph (Fig. 1) with a value xp = 290 milliori 1. In comput-
ing the Ll y and A z values tiy use of equations (2), (6) and (7) during these same
time intervals the x(t) function was approximated by two segments of the exponen-
tial curve:
for interval I �r eA, (t- ISfiU) 812 ' 10' year-1; (9)
x~
x, = e (lo)
for interval II .Y A- (1 - B;~S)
-
,
wliere xl = 315 million 1, A2 = 3.08�10-3 year 1.
The system of equations (6)-(7) is such that tlie increment of biomass, both living
_ and dead, is proportional to A . An analysis of its solution witli functions x/xo
of the Cype (9) and (10) shows that variation of the parameters y~ and y~ in Lhe
range �25% exerts little effect on the results of computation of py (in the limits
10%). In addition to combustion of fossil"fuel, some additional COZ quantity will
c:nter the atmosphere as a result of cutting of forests.
The entry of C02 into the atmosphere accompanying the procurement of wood repre-
sents some part of the annihilated biomass. There is also the uncontrollable anni-
hilation of forests. According to estimates [31], the existence of uncontrollable
factors in human activity means that tlie real quantity of anthropogenic C02 enter-
ing the atmosphere will be LS 14(1 + W), where oC falls in the range 0.35-0.15. If it
is assumed that in interval I� I= 0.35, and in interval IIlXII = 0.15 then the
reduction of the balance in the table with a mean value Vmd will give ~I = 0.18
and AII = 0.40. It is evident that the error in determining P by this method is
very great; it is even difficult to estimate quantitatively because the parameters
06I and 01~II are very uncertain.
The value of the A parameter can also be determined Uy another independent method by
r,ionitoring the atmospheric content of the isotope 13C. The atmosphere and the bio-
sphere, as well as fossil fuel, contain different quantities of the nonradioactive
isope of carbon 13C (J 13Cat~3= '7�/00, d 13Cbio -j13~fue1 - -250/00, where
j 13C is the deviation of the C content in different reservoirs from the stan-
_ dard (1.237%)). Since fossil fuel and biomass contain 13C in a volume 180/00 less
than the atmosphere, the discharge of C02 in the combustion of fuel, and also any
change in the biosphere cause changes in the content of 13C in the atmosphere.
Tlie measurements made by Keeling, et al. [27] indicated that from 1956 through 1978
the atmospheric content of 13C decreased from -6.69 to -7.240/00, that is, by
0.550/00. During this period the atmosphere received 79 Gt of C02 due to the carbon
in fossil fuel. The total reservoir which received this excess C02 with a reduced
13C quantity is approximately 3.3 xo (without allowance for the fractionation of
isotopes). These data make it poss.ible to estimate the change in biomass in these
22 years. In actuality, if the atmospheric content of 13C changed only as a result
c,f industrial effluent, then during the 22 years the change in the content of 13C
Q(~' 13C atm) ~~ould be determined by the difference Q(~' 13Cfue1- S13C atm) -'18�/00,
and also by the effluent and the total reservoir:
30
FOR OFFICIAL USE ONI..Y
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
18�. oo) 3x
3~o - - 0'7~/0�'
According to data in [27], during these years the decrease in atmospIieric 13C cqn-
tent was only -0.550/00. This means that the biomass partially replenished the 13C
deficit due to its growth: -
- 180 ,'....1;s, 3 xo U,7 i),55)�r a,,, ly=17 Gt.
Using the results of computations for A y on the basis of equations (6)-(7), we as-
sume A y for these years to be equal to 24.6(3 + 29.8 p = 54.4 p. Then the estimate of
- h on the basis of the change in atmospheric 13C content gives 0.31.
,
y '
I
70 ;
, 90 60 ~
_ 11o, of o5 ~
10 2
- ~ J
1110 1000 2010 1020 ?OJO
Fig. 5. Distribution of anthropogenic carbon dioxide between atmosphere (1), contin-
ental biomass (2) and ocean ~3) with P = 0.3, V~ = 1.25�105 cm/sec. The curve 1'
corresponds to V~ = 2.5�10- cm/sec. The inset shows the Ax/,d W values in % for the
year 2030 as a function of - -
a~ (ck) MnM�' milliori 1 .
;00
a) ~
!
~
S00
. ~
n r i ; t
I
~
Gt
1990 1000 10f0 1010 10.iJ a~0 'i00
Fig. 6. Temporal variation of C02 concentration in atmosphere (a) and dependence of
concentration in 2030 on quantity of combusted foasil fuel within limits of range
defined in Fig. 3 (b).
31
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Results of Computations of Changes in Atmospheric C02 Concentration to 2030
Computations of the distribution of anthropogenic carbon dioxide between the atmo-
sphere, ocean and biomass were made on the assumption that the entry of C02 into
the atmosphere occurs at the rate dW/dt corresponding to the most probable curve
shown in Fig. 3. Accordingly, the total discharges Q W beginning in 1980 will be
as follows: in 1990 63 Gt, in 2000 155 Gt, in 2010 297 Gt, i.n 2020
494 Gt, in 2030 788 Gt. The basic computatior.s were made for aPvalue obtained
using data from the monitoring of 13C, that is, 0.3, and for two Vmd values: 1.25�
10'S cm/sec and 2.5�10'5 cm/sec. The results are given in Figures 5 and 6. Figure
5 illustrates the percentage distribution of A W among the three reservoirs. Its
characteristic feature is that in that time interval ivhen dW/dt is increasing the
Fraction of industrial COZ remaining in the atmosphere is increasing. This occurs
due to the great inertia of the ocean reaction to atmospheric disturbances. The
fraction from the biomass remains approximately constant. The ratio Q x/0 W, equal,
according to monitoring data, to 43.5% in 1970, agrees well with extrapolation of
computations for Vmd = 2.5�10-5.
Variants of calculations with Vmd = 1.25�10-5 and 0.3 and with Vmd = 2.5�10-5
and ,8 = 0.15 give values of the temporal variation of atmospheric C02 content
coinciding within the limits 5 million 1. This function is also shown in the last
graph (Fig. 6). This also illustrates the influence of the total quantity of in-
dustrial discharge in 2030 on the x values for this same year. The influence of in-
dependent variations of the 18 and Vmd parameters is illustrated by the inset in
Fig. S. The results of computations by the models of the carbon cycle for 2020-
2040 mentioned at the beginning of the article fit approximately into this same
L1 x/A W range.
In one of the recent studies of this subject [37] there is an analysis of limiting
variants of the development of energy during the next 50 years. In the first vari-
ant the rate dW/dt of industrial efzluent ceases to increase beginning in 1975 and
remains equal to approximately 5 Gt/year. In the second the mean annual increment
remains constant and is equal to 4.5% per year. In this case the dW/dt value by
2030 will become e2�25, that is, 9.5 times greater than at the present time, that
is, 50-60 Gt/year. The doubling of the C02 concentration in the atmosphere by 2030
which we obtained corresponds to an intermediate variant in which the rate of in-
dustrial effluent will increase by 2030 to 26-30 Gt/year. In this case the total
magnitude of the anthropogenic C02 effluent will be about 800 Gt, which is consid-
erably less than the reserves of fossil fuel available for use, the quantity of
which is estimated at f rom 5000 to 10 000 Gt.
The authors express their appreciation to M. I. Budyko for initiating this study.

32
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FOR OFFICIAL USE ONLY
BIBLIOGRAPHY
1. ATLAS OKEANOV. T 1. TIKHIY OKEAN; 7 2. ATLANTICHESKIY I INDIYSKIY OKEANY
(Atlas of the Oceans. Vol 1. Pacific Ocean; Vol 2. Atlantic and Indian Oceane),
VMF SSSR, 1974, 1977.
2. Ariel', N. Z., Bortkovskiy, R. S., Byutner, E. K. and Strokina, L. A., "Influ-
ence of Ocean Contamination by an Ocean Film on the Exchange of Oxygen With
the Atmosphere," METEOROLOGIYA I GIDROLOGIYA (Meteorology and Hydrology), No
2, 19 79 .
3. Bazilevich, N. I, Rodin, L. Ye. and Rozov, N. N., GEOGRAFICHESKIYE ASPEKTY IZ-
UCHENIYA BIOLOGICHESKOY PRODUKTIVNOSTI (Geographical Aspects of Study of Bio-
logica). Productivity), Izd. Geogr. Obshch-va SSSR, Moscow, 1970.
4. Budyko, M. I., PROBLEMA UGLEKISLOGO GAZA (Problem of Carbon Dioxide), Lenin-
grad, Gidrometeoizdat, 1979.
5. Budyko, M. I. and Ronov, A. B., "Evolution of the Chemical Composition of the
Atmosphere in the Phanerozoic," GEOKHIMIYA (Geochemistry), No 5, 1979.
6. Budyko, M. I. and Vinnikov, K. Ya., "Global Warming," METEOROLOGIYA I GIDRO-
LOGIYA, No 7, 1476.
7. Yefimova, N. A., RADIATSIONNYYE FAKTORY PRODUKTIVNOSTI RASTITEL'NOGO POKROVA
(Radiation Factors in the Productivity of the Plant Cover), Leningrad, Gidro-
meteoizdat., 1977.
- 8. Kagan, B. A. and Ryabchenko, V. A., "Mean Concentration and Budget of Dissolv-
ed Oxygen in the World Ocean," DOKLADY AN SSSR (Reports of the USSR Academy
of Sciences), Vol 233, No 4, 1977.
9. Kagan, B. A. and Ryabchenko, V. A., TRASSERY V trIIROVOM OKEANE (Tracers in the
World Ocean), Leningrad, Gidrometeoizdat, 1978.
10. Kobak, K. N., "Carbon Dioxide Supply for Forest Biogeocoenoses, PROBLEMY
EKOLOGII I FIZIOLOGII LESNYKH RASTE-NIY (Problems in the Ecology and Physiol-
ogy of Forest Plants), Leningrad, LTA, No 2, 1965.
11. Koblents-Mishke, 0. I., Volkovinskiy, V. V. and Kabanova, Yu. G., "New Data
on the Magnitude of Primary Production in the World Ocean," DOKLADY AN SSSR,
Vol 183, No 3, 1968.
12. Lyakhin, Yu. I. and Ivanenkov, V. N., "Elements of the Carbonate System in
Atlantic Ocean Waters," KHIMIKO-0KEANOGRAFIQiESKIYE ISSLEDOVANIYA MOREY I
OKEANOV (Chemical-Oceanographic Investigations of Seas and Oceans), Moscow,
Nauka, 1975.
13. Lyakhin, Yu. I., Aleksandrov, V. P. and Pal'shin, I. I., "Computation of the
Balance of C02 Exchange Between the Ocean and Atmosphere Over the Atlantic,
Pacific and Indian Oceans," TRUDY LGMI (Transactions of the Leningrad Hydro-
meteorological Institute), No 65, 1978.
33
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
14. MORSKAYA VODA (Sea Water), edited by A. S. Monin, Moscow, Nauka, 1979.
15. Ronov, A. B., "Volcanism, Carbonate Accumulation and Life," GEOKHIMIYA, No 8,
1976.
16. Svirezhev, Yu. M and Tarko, A. N., "Modeling of the Global Biogeochemical Cycle
of Carbon," VESTNIK AN SSSR (Herald of the USSR Academy of Sciences), No 12,
1979.
17. Skopintsev, B. A., "Oxygen Consumption in the Deep Waters of the Ocean," OKEAN-
OLOGIYA (Oceanology), 15, tdo. S, 1975.
18, Stepanov, V. N., "Temperature Field of Waters in the World Ocean," OKEANOLOGIYA,
Vol 20, No 1, 1980.
19. Suyetova, I. A., "Patterns of Quantitative Distribution of Biomass of the Land
and Ocean," VESTNIK MGU: SERIYA GEOGRAF. (Herald of Moscow State University:
Geography Series), No 6, 1973.
20. Bacastow, R. B. and Keeling, C. D., "Models�to Predict Future Atmospheric C02
Concentrations," WORKSHOP ON THE GLOBAL EFFECTS OF CARBON DIOXIDE FROM FOSSIL
FUELS, Miami Beach, Florida, 1977.
21. Bjorkstrom, A., "A Model of the C02 Interaction Between Atmosphere, Oceans and
Land Biota," SCOPE REPORT 13. THE GLOBAL CARBON CYCLE, Ch 15, 1979.
22. Bohn, H., "On Organic Soil Carbon and C02," TELLUS, Vol 30, 1978.
23. Bray, J. R., "An Analysis of the Possible Recent Change in Atmospherie C02 Con-
centration," TELLUS, Vol 11, 1959.
24. Broecker, W. S. and Peng, T. H., "Gas Exchange Between Air and Sea," TELLUS,
Vol 26, 1974.
25. Eriksson, E., "The Atmospheric Carbon Dioxide Problem," TELLUS, Vol 30, 1978.
26. Keeling, C. D. and Bacastow, R. B., "Impact of Industrial Gases on Climate,"
ENERGY AND CLIMATE, 1977.
27. Keeling, C. D., Mook, W. G. and Tans, P. P., "Recent Trends in the 13C/12C
Ratio of Atmospheric Carbon Dioxide," NATURE, Vol 277, No 5692, 1979.
- 28. Loomis, R. S., "C02 and Biosphere," CONF-770385. WORKSHOP ON THE GLOBAL EFFECTS
OF CARBON DIOXIDE FROM FOSSIL FUELS, 44-50.
29. Machta, L., "Atmospheric Measurements of Carbon Dioxide," CONF-770385. WORKSHOP
ON THE GLOBAL EFFECTS OF CARBON DIOXIDE FROM FOSSIL FUELS, 44-50.
30. Oeschger, H., Siegenthaler, U., Schotterer and Gugelmann, A., "A Box Diffusion
Model to Study the Carbon Dioxide Exchange in Nature," TELLUS, Vol 27, 1975.
34
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
31. Olson, J. S., CHANGES IN THE GLOBAL CARBON CYCLE AND THE BIOSPHERE, Oak Ridge
Piational Laboratory, 1978.
32. Pytkowicz, R. M., "Carbonate Cycle and the Buffer Mechanism of Recent Oceans,"
GEOCHIM. ET COSMOCHIM. ACTA, Vol 31, 1967.
33. Pytkowicz, R. M. and Small, L. F., "Fossil Fuel Problem and Carbon Dioxide: An
Overview," THE FATE OF FOSSIL FUEL C02 IN THE OCEAN, edited by N. Andersen and
A. Malahoff, 1977.
- 34. Revell, R. and Munlc, W., "The Carbon Dioxide Cycle and the Biosphere," ENERGY
AND CLIMATE, 1977.
35. Riley, J. P. and Skirrow, R. W., CHEMICAL OCEANOGRAPIiY, Vol 2, 1975.
36. Rotty, R. M., "Present and Future Production of C02 From Fossil Fuels. A Global
Appraisal," CONF-770385. WORKSHOP ON THE GLOBAL EFFECPS OF CARBON DIORIDE FROM
FOSSIL FUELS: 36-43.
37. Siegenthaler, U. and Oeshger, H., "Predicting Future Atmospheric Carbon Dioxide
Levels," SCIENCE, Vol 199, No 4327, 1978.
38. Takahashi, T., "Carbon Dioxide Chemistry on Ocean Water," 1977. CONF-770385.
WORKSHOP ON THE GLOBAL EFFECTS OF CARBON DIOXIDE rROM FOSSIL FUELS: 63-71.
39. Vooys, C. G. N., "The Carbonate System of the Ocean," Ch 10, SCOPE Report 13,
THE GLOBAL CARBON CYCLE, 1979.
40. Ziemen, K. E., Offerman, P. and Hartman, G., "Source Functions of C02 and Fu-
ture C02 Burden in the Atmosphere," Z. NATURFORSCH., No 12.
35
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FUR OFFICIAL USE ONLY
UDC 551.(591.2+576.4)
EXPERIMENTAL INVESTIGATION OF THE CORRELATION BETWEEN THE METEOROLOGICAL RANGE OF
ViSIBILITY AND ALTITUDE OF THE LOWER CLOUD BOUNDARY
Mosccw METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 32-38
[Article by Ye. R. Milyutin, candidate of tiechnical sciences, a:id Yu. I. Yeremenko,
Leningrad Electrotechnical Institute of. Communication, manuscript received
24 Jun SO]
[Text] Abstract: On the basis of an analysis of ex-
perimental data a study was made of analytical
representatinns of real probabilistic distribu-
tions of the altitude of the lower boundary and
tfie quantity of clouds. The correlation and re-
gression relationships between the meteorological
range of visibility and the altitude of the lower
cloud boundary are determined.
The results of ineasurements of atmospheric transparency in horizontal and vertical
directions are used extensively in a nLUnber of technical applications (optical com-
munication, aviation, etc.). A universal characteristic of transparency in a hori-
zontal direction is the meteorological range of visibility SM (MRV), the finding of
whose analytical form of the probabilistic distribution law was the subject of a
study by the authors [15].
In many cases (in optical sounding, in optical communication on slant paths, in avi-
ation, etc.) a knowledge of the altitude of the lower cloud boundary Hlow is also
of great practical interest. Accordingly, attempts at determining the statistical
_ correlation of SM, measured at many meteorological stations, and HloW, that is, in
other words, attempts to find the correlation between horizontal and vertical atmo-
spheric transparency in the presence of a cloud cover, are extremely timely. This
problem was examined in a number of investigations [2, 7, 8, 12] in which the in-
fluence of low-lying clouds on the MRV was noted, but the correlation between SM
and Hlow was evaluated only qualitatively.
In this study we will determine the empirical probabilistic aistribution functions
for Hlow and the quantity of ::louds N(in tenths) and a study is made of their sea-
sonal and annual changes. On the basis of an analysis of the experimental data a
study is made of analytical representations of the real distributions and the cor-
relation and regression relationships between the raridom SM and Hlow values are
found.
36
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
�M
I
I
91
B ~
1
6
.
,i~� 6onaei 4
y tenths. .
6) / ~ 1
i
B
i
7 i ; ~
~
SL
1
Fig. l. Variation of monthly changes ms and mN for 14 years (1) for first (2), sec-
ond (3) and third (4) periods.
As initial data we used the results of MRV observations for 14 years (1964-1977)
made at Leningrad meteorological station. Durx.lg the first period (1964-1968)
the observations were made visually and in the second period (1969-1972) both vis-
ually and by means of a range-of-visibility recorder. During a third period (1973-
1977) only a range-of-visibility recorder was employed. The total number of observ-
ations was ns = 122 736.
Curves of the monthly changes in the sample mean ms both for individual periods aad
for 14 years are shown in Fig. la, where the sequence number for the month of the
year is plotted along the x-axis. At the same time we processed observational data on the quantity of clouds (N) which
was registered eight times a day. 7.`he total number of observations was nN = 40 912.
The results of space and aerological observations in this case could not be used
because we were interested in the state of the cloud cover within a stipulated solid
angle of the celestial hemisphere viewed from a definite point on the earth's sur-
face.
'On the basis of data on N we computed the frequencies of occurrence of the quantity
of clouds in the range from 0/10 to 10/10 with a 1/10 interval for each month:and
as an average for the year, after which we constructed graphs of the accumulated
37
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
.1 .

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-44850R000400030065-3
FOR OFFICIAL USE ONLY
frequencies (Fig. 2a), representing statistical analogues of the dik-tribution func-
tions F(N) and making it possible to find the prohahility P that N will be greater
than some level. The empirical dis.tribution runctions shown in Fig. 2a at the points
N= 0 and N= 10 have discontinutties of the first kind, which makes their analytical
description quite difficult. These peculiarities in the behavior of the distribution
are attributable to the fact that in temperate and polar climates it is most coumon
to observe values N= 0 and N= 10 [23]. In addition, we computed sample mean mN
and upbiased competent evaluations of the dispersion DN. The results are given in
the Lahle and Fig. lb shows the curves illustrating the monthly changes mi,l.
An analysis of the results for 14 years indicated that the minimum quantity of
clauds is observed in June and the maximum quantity is observed in November; even
in June mN > 6, and P(N > 5)> 0.61, which confirms a considerable influence of cloud
cover on atmospheric transparency in a vertical direction. However, as indicated in
Fig. lb, there are also deviations from the mean values, for example, in the f irst
period the cloud cover maximum fell in December, which agrees with the conclusion
presented in [4] on the basis of the results of observations of clouds in approx-
imately these same years.
Then we determined the distribution function for the altitude of the lower cloud
houndary, the quantity of which exceeds 5/10. We processed data from monthly ob-
servations of IiloW during 1964-1968 made using a cloud-altitude measuring apparatus;
nH = 43 848. All IlloW in accordance with ICAO recbmmendations are put into 12 inter-
vals in order that in the subsequent computations it would be possible to use data
From meteorological stations at the international airports of the ICAO member coun-
tries.
The sequence for the further statistical Qrocessing of observations was the same
as in the preceding case. The determined mH values are given in the table, from
whic?z it follows that the minimtmm mean altitude of the lower cloud boundary was
observed in January and the maximim altitude was observed in June. In an analytical
description of the distribution function HloW we used the same set of distributions
as for SM [15].
The computations made on an electronic computer indicated that according to the
Kolmogorov consistency test the most acceptable approximation of the real dis-
tributions for the spring and autumn unnths, and also the mean annual distribution,
is a truncated normal distribution (Fig. 3), whereas for the summer and winter
months it is a Rayleigh distribution (Fig. 3b). The derived dependences make it
possible, using the data in [4] on the relationships between cloud altitude and
type, to evaluate the attenuation of optical radiation in the cloud layer.
By comparing the behavior of the curves of monthly changes of the parameters ms,
~ and mH during the first period (Fig. l,a,b and the table) it can be seen that
the maxima of the mS and mH curves and the minimum of the mN curve are observed
in June, whereas the ms and mH minima coincide in January and the uN maximum is
displaced to December.
For a quantitative evaluation of the degree of correlation between SM and Hlow We
compiled correlation tables with gradations of SM and Hlow, established earlier,
for each month and as an average for the year. Using the data in the table we
38
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
computed the corresponding correlation coefficients r and ths correlation ratios
Y~ using the known formulas [1]
14 il
~ntl SUtH�,-nHMSmy
_ r _ !_1 r_l ~
( ny )H
14 (2)
YI = ' rVd ni I-My (S- d - My I' r
5H P_t
_ - - -
where [HH = Hloucj
mN (SN,) _/~'i [ a~ = Sa, i, = nj
is the conventional mean HloW value with SN,I = SM i, HIoW 3 is the middle of the j-th
intervals of the H1ow values, nij are the frequencies of combinations of the values
_ SM = SM i with Hlow - Hlow ,j�
The results of computations using formulas (1) and (2) are given in the table.
The significance of the correlation was. determined in the following way. According
to [14], in the case of large volumes of the sample nH (nH> 80) the r value is con-
sidered considerahly different from zero if the following inequality is observed

r > f1~- "y 2l- -
~ (3)
L = J
where toc is the critical value of the Student t-distribution with (nH - 2) degrees
of freedom, corresponding to the selected significance level ac3 (o< a3 < 0.5).
The critical value ta = tnH - 2(W-3) is a solution of the equation
Sny-2(tQ) = 1 - 2 , (4)
where snH_2(tpO is the Student distribution function.
We note that with nH_2>80 it is possible to use the approximate equation
S�H _ 2tt, )=0,5 + (P- (tQ~, (5)
where 4f 0(ta ) is the tabulated integral of probabilities.
_ In addition, with large nH the r value approximately eorrespunds to a normal dis-
tribution with the dispersion D= 1/nH(1 - r2)2, which makes it possible to evalu-
ate the r value using the confi5ence coefficient 'Y[8],
- P 1r - ~ I~U,
If the turbulent diffusion coefficients are considered turbulence characteristics,
they must be represented in the form
53
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R044400030065-3
FOR OFFICIAL USE ONLY
I:ij (z) = v kij (zu*/ v
where kij are universal functions, 1~ is the molecular viscosity coefficient.
(4)
Within the limits of the logarithmic region of flow, that is, with zu*/v >1, the
dependence on v must disappear, that is
Kij (z) _ 'X ij u*z.
(5)
The value of the constant x33, determining vertical turbulent transfer, is best
known. This constant is determined through the turbulent Prandtl number Prt and
the Karman constant -Y-( Y-33 =Y-/Prt) and according to data from a large number of
laboratory measurements [11], X 33 = 0.45-0.47. Among the other constants only the
value of the constant JC,~3 can be considered at least approximately (with an accur-
acy to 15%) known [3, 12 x13 = 3�5. This value was obtained from measurements
of the horizontal turbulent flow of heat SX = u'T' and the mean temperature profile
T(Z) (K13 = - SX/ )T/d z).
Yamamoto and Shimanuki in [13] attempted to evaluate the transverse diffusion co-
efficient by a comparison of the empirical data on diffusion from point sources and
the results of numerical solution of the stationary diffusion equation
rt (z) dX = dy lK~: ~n n� ( K:::;I
(6)
xqith K22 values containing an undetermined parameter: K22(z) _)f-u*zOC(L'O), where
06(9 o) is an unknown fur.ction dependent on stratification. Then for the purpose of
checking the adopted assumption concerning K22(z) the determined 01(J'0) values were
used in theoretical computations of the distribution of the concentration along the
y-axis with some x and z values, the rgsults of which were again compared with the
experimental data. For a surface concentration along the axis of the cloud of impur-
ity with a neutral stratification and K22 = 13Y-u*z (that is, with X 22) the expres-
sion q^'x 1�78, agreeing well with the experimental data. T32
It is difficult to determine the value of the constant X11 against the background
of longitudinal advection. Indirect evaluations of the K11 value on the basis of
laboratory measurements of the characteristics of turbulence in the boundary layer
[8] lead to a value Y 11^14�5X33. In addition, there are evaluations of Y-11 based
on semiempirical hypotheses [6], leading to values X11/ X33"'30-40.
Data on the X31 values are entirely lacking. There are a number of expressions
derived on the basis of approximate physical or qualitative considerations of the
type
zi:i = Z:;i, Z,~ ~ z
:n = -T Zti�
(7)
Thus, from a brief review of the esti.mated values of the constants Zi� it follows
that evaluations of thes,: constants were obtained from a broad spectr of empir-
ical and semiempirical assumptions sometimes having an indirect relationship to the
process of scattering of an impurity in the SLA. The values of some constants are
, 54
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
virtually unknown; others are not very reliahle. Accordingly, it is entirely reas-
onable to derive expressions determining the hehavior of some characteristics of a.
diffusing cloud of impurity, which are relatively simply and reliably determined
experimentally. The constants x ij enter into the analytical expressions for these
characteristics. Thus, having experimental information on these diffusion character-
istics it is possible to obtain evaluations of the X-ij constants.
The behavior of the coordinates of the center of gravity of the cloud of impurity
from an instantaneous point source of impurity is determined by the following ex-
pressions [4, 5]:
X (t) - u=r (lfl r,,1 - 1 (8)
Y. 1 �
Z(t) =r.3:tr# t, (9)
co
where y=- S e X ln x dx w0.58 is the Euler constant.
0
Thus, from the behavior of the trajectory of the center of gravity of the cloud
of impurity it is possible to determine the values of the 1r33 and X13 constants.
The /X33 value has been determined precisely in this way in laboratory measurements.
The longitudinal and vertical dispersions (t) and Z 33 (t), obtained from solu-
tion of equation (3), have the form [4, 71
~ x T~ll '~:i3 l y 13 1 us t_'
(r1t x ~1~~
z 1~ - [(6 - / ~
J
33 (t) . Z;;3 u, t3. li
E' c~
Using the moments method, we similarly determine
- c, �
Zl^ A3:; u1 , t3. (12)
2: ~c37 (_T 1 ~31 xl3 ~ ll`
13 1' 2 . (13)
Thus, expressions (8)-(13) determine the behavior of the center of gravity of a
diffusing cloud of impurity and its dispersions, by measuring which it is possible
to determine the constants X-i without recourse to any additional considerations.
For example, in the case of aliagonal tensor of the diffusion coefficients, that
is, when X 13 = y- 31 = 0, we obtain
T T2 r~ (14)
`I 1 I r.' - ~ l u' ( ,
Z (~1
.13u. r I b ~ xi z~ - Z(t) ut t
\
1=, (15)
R.~ ~ ~
r ~Jt
55
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
_ Z (tl
~s~ - u* t u* (r)'
(16)
Obtaining the estimated values of the constants Alj from diffus.ion measurements,
it is pos.sihle to use tfiem for computtng thQ concentration field in the SLA by
use of equation (3). A comparison of this field with the experimental distribution
makes it possible to evaluate the reliability of the initial hypothesis (5) on the
functional type of the turbulent diftus.ion coefficients. In addition, the once-
determined values )6i1 can later be used als-o in evaluating the diffusion charac-
teristics of the cloud of impurity determined by expressions (8)-(13).
It must be noted that theoretical methods for determining the universal constants
and functions describing turbulent diffusion of an impurity in the SLA rely on
different nonrigorous semiempirical hypotheses and therefore are not precise. Ac-
cordingly, it is useful to carry out a comparison of the conclusions from differ-
ent semiempirical hypotheses. The comparison makes it possible to clarify in what
cases these conclusions vary little with a change in the initial hypotheses and
therefore can be considered entirely reliable and in what cases they are character-
ized by a considerable scatter and accordingly indicate that the degree of accuracy
of the used hypotheses must be checked using materiaT from additional experimental
measurements.
A similar examination of scattering of an impurity in the SLA on the basis of equa-
tion (8) with diffusion coefficients Kij = X.'i3.u*t, dependent only on time, leads
to the following system of equations determining the behavior of the trajectory of
the center of gravity of a diffusing cloud of impurity and its dispersions:
,
u 2
y .a
= r 7 l
X(t) _ ~ ~,In 1' 2 l~r.~.~3 ~~;a -r Z~i) u~ t~
(17)
, M' 2 I 33 �tiT t,
(18)
2 In 2( ~.i.i ~.3i ly ia t
t
u( 1 , (19)
i_11.V,
(20)
' (21)
, 2~
's.,112 12~
2
(r) = 1 [ ! ..-t In 0 e)
+ + ,
2
(22)
If the values of the constants x-iJ . and x~3 are determined from one and the same
mass of experimental data it is possible to draw some conclusions concerning the
functional dependence of Kij. In actuality, by tlie process of finding the distrib-
ution of the concentration with diffusion coefficients dependent only on z or
56
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
only on t, and making a comparis,on with the experimental distrihution of the con-
centration, in this case proceeding on the basis of the best correspondence with
the experimental data, it is possible to draw conclusions concerning the reliabil-
ity of the initial hypotheses concerning the form of Kii. In a case when it is
found that different hypotheses on the functional form of the coefficients of tur-
bulent diffusion lead to virtually identical results, by comparing the correspond-
ing expressions for the dispersions we find the correlation between the constants
)4 i� and )1~j; in particular, for the case of a diagonal tensor of the di�fusion
coeificients we have
(23)
A=_~ - ~'9: x391 (24)
it
X,13 - - _ 2 ~33' (25)
Thus, the derived systems of expressions (8)-(13) and (17)-(22), describing the
behavior of the coordinates of the center of gravity of a diffusing cloud of im-
purity and its dispersions., make it possible to express through these character-
istics the constants entering into the determination of the diffusion coefficients
(2), to find the correlation between constants of a different model representa-
tion of the diffusion coefficients and with the availability of information on
the field of concentration or other more complex characteristics of the cloud of-
impurity to evaluate the reliability of the initial hypotheses on the diffusion
coefficients. This method can be used successfully in the case of scattering of
an impurity in the SLA with a stratification different from neutral.
Scattering of an Impurity in a Horizontally Uniform Stationary ABL
The method described above in the example of diffusion of an impurity in the SLA
is especially useful in an analysis of scattering of an impurity in the ABL since,
having experimental information on the behavior of the dispersions and the measur-
ed characteristics of the ABL it is possible to obtain evaluations of the horizon-
tal coefficients of turbulent diffusion which in contrast to the vertical coeffic-
ient are not determined by the ABL model. In actuality, the system of equations
for the first two moments qo(z, t), qm(z, t), qm(z,t)(m = 1, 2), through which the
behavior of the coordinates of the center of gravity of the cloud of impurity and
its dispersions is determined, has the following form:
aQ"- _ (K:1a 0 9 ) _ 0,
(26)
a
d dm / = Z , ~27>
d - z K3tdz m (z) 9n (Z' t)+
da~Tj - d ( K73 0ds 1 - ? Kmm (f) 94 (Z, 1) -f- ? 'Um 9jt (Zr t)� ~28~
In the derivation of the equations in the system it was assumed that the compon-
ents of the wind profile vm are dependent only on z, the tensor of the diffusion
coefficients has a diagonal form and vm(z) and K33(z) are determined from the ABL
model, and K11 and K22 will be considered dependent only on diffusion time. Such
an assumption for the horizontal diffusion coefficients is entirely reasonable
57
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
(for example, see [IO], which gives. the experimental dependences of dispersions
on time for different meteorological condi.tions).
Integrating equation (28) for z, we obtain an equation for the dispers.ion ~~(t)
c2 ~
~ d Lrmm _ K.m (t) + r z~m (2) q" (Z, t) dz i i~ (x,n(29)
dl ~
where Xm(t) are the coordinates of the center of gravity of the cloud of impurity.
lience we obtain an evaluation for the horizontal coefficients of turbulent dif-
fusion in the ABL
~ d ~nrm
dr
~
qm (t) d q~+
Z,,, (Z) 9, t) dZ r q~ dr ( q� (30)
- ~ ~ m ~
0
where qp(z, t) and qi(z, t) are determined from solution of equations (26) and (27)
and the temporal behavior of the dispersions is determined from the experimental
data.
Several attempts were made to evaluate the transverse diffusion coefficient K22 (for
example, see [1], which gives the lates.t, most complete review of the results of
experimental investigations of transverse diffusion in the ABL). These evaluations
were based on a formula derived by Csanady [9] and describing the behavior of the
transverse dispersion of the s.urface field of concentration from an instantaneous
point surface source
K:�, t f: P.
(31)
where G is the velocity of the geostrophic wind, f is the Coriolis parameter, and
the function F describes the influence of the shear effect on the scattering pro-
cess (that is, the effect of interaction between the wind profile and vertical
transfer). But it must be taken into account that formul.a (31) was derived in an
examination of scattering of the impurity in the Ekman ABL with constant (vertical
and horizontal) diffusion coefficients. Accordingly, these evaluations of the
transverse diffusion coefficient have a preliminary character and at best determine
the order of magnitude of K22, since in not one of the experiments used for evaluat-
ing K22 is there adherence to the conditions for the applicability of formula (31).
The turbulent diffusion coefficients obtained in this study by the considered meth-
od make it possible to avoid such ineorrectness and henceforth can be used in com-
puting the concentration field on the basis of the semiempirical equation (1). Com-
parison with the experimental distribution makes it possible to evaluate the relia-
bility of the initial hypothesis that the horizontal diffusion coefficients are
f.unctions of time. The use of other model representations and hypotheses on ;:he
functional dependence of the horizontal diffusion coefficients and computations of
the field of concentration on the basis of the evaluations K11 and K22 obtained
for this type of coefficients, and also comparison with the measured distribution
of the field of concentration make it possible to make an optimum selection of a
model of scattering of an impurity in the ABL.
58
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
BIBLIOGRAPHY
1. Garger, Ye. K., "Trans.verse Diffus.ion in the Atmospheric Boundary Layer,"
TRUDY IEM (Transactions of the Institute of Experimental Meteorology),
No 15(60), 1977.
2. Yaglom, A. M., "Equations Witfi.Time-Dependent Coefficients Describing Dif-
= fusion in the Surface Layer of the Atmosphere," IZV. AN SSSR: FIZIKA ATMO-
_ SFERY I OKEANA (News of the USSR Academy of Sciences: Physics of the Atmo-
sphere and Ocean), Vol 11, No 11, 1975.
3. Yaglom, A. M., "Turhulent Diffusion in the Atmospheric Surface Layer," IZV.
AN SSSR: FIZIKA ATMOSFERY I OKEANA, Vol 8, No 6, 1972.
4. Yaglom, A. M., "Diffusion of an Impurity From an Instantaneous Point Source
in a Turhulent Boundary Layer," TITRBULENTNYYE TECHENIYA (Turbulent Currents),
Moscow, Nauka, 1974.
5. Batchelor, G. K., "Diffusion From Sources in a Turbulent Boundary Layer,"
ARCH. MECH. STOSOWANEJ., Vol 16, No 3, 1964.
6. Brutsaert, W., "On the Anisotropy of the Eddy Diffusivity," J. METEOROL. SOC.
JAPAN, Vol 48, No S, 1970.
7. Chatwin, P. C., "The Dis.persion of a Puff of Passive Contaminant in the Con-
stant Stress Region," QUART. J. ROY. METEOROL. SOC., Vol 94, No 401, 1968.
8. Corrsin, S., "Limitations of Gradient Transport Models in Random Walks and
in Turbulence," ADV. GEOPHYS., Vol 18A, 1974.
9. Csanady, G. T., "Diffusion in an Ekman Layer," J. ATMOS. SCI., Vol 26, No 3,
1969.
10. Hildebrand, P.. H., "A Radar Study of Turbulent Diffusion in the Lower Atmo-
sphere," J. APPL. METEOROL., Vol 16, No 5, 1977.
11. Kader, B. A. and Yaglom, A. M., "Heat and Mass Transfer Laws for Fully Turbu-
lent Wall Flows," INT. J. HEAT MASS TRANSFER, Vol 15, No 12, 1972.
59
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFF[CIAL USE ONLY
UDC 551.(509.58+510.42)
COMPUTATION OF TRANSPORT OF SUBSTANCES CONTAMINATING THE ATMOSPHERE
tfoscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, trlar 81 pp 54-58
[Article by U. AndrasYi, candidate of economic sciences, and R. Schenk, candidate of
technical sciences, Water Management Institute, Berlin, manuscript received
1 Jul 80]
[Text] Abstract: The article is devoted to a de-
scription of a model and computation scheme
for computing the transport of atmospheric
impurities in the field of a variable wind
at country and regional scales. A program
for computing the transport of substances
contaminating the air has been developed.
A special working group of the Berlin Water Management Institute in the German
Democratic Republic has carried out investigations of the transport of substances
contaminating the air over great distances. The investigations were formulated
in connection with the need for monitoring the transport of harmful substances
not only near industries and in cities, but also in individual regions having
linear dimensions of thousands of kilometers, including transport across national
Uoundaries.
The authors here present a"fractional steps" method for solution of the three-
dimensional equation for the transport of contaminating substances. It is known
that the transport of impurities at different spatial and temporal scales (intra-
regional, distant, etc.) is described by one and the same differential equation.
Only methods for taking into account convective transport, turbulent diffusion,
and chemical transformations of the transported substances are different, depend-
ing on the scale of transport. Accordingl~;~, the method cited in the article has
great universality.
Description of Transport Mechanism
The processes of transport of substances contaminating the air over short, inter-
mediate and great distances can be described by the following equation:
nr + X + v~ay + V= az =
(1)
a ac v ' a ac
a ~Kr a )~r dy (K,de ' dy ) + (K-, a Q.
60
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
where VX, Vy, VZ are the wind velocity components, KX, Ky, KZ are parameters de-
scribing turbulent transport, Q characterizes the volume of the emissions and the
processes of chemical transformations of impurities.
The modeling of transport using equation (1) is based on the proposition that the
impurities are transported by the air medium in which they are scattered and that
the change in the velocity of movement of the impurity in comparison with the velo-
city of the air itself is possihle only as a result of turbulent transport.
A description of the influence exerted on the characteristic motion of impurities
by partial pressure and different density is possible only by use of multicomponent
models. In actual practice it is better to use equation (1), taking into account
the factors enumerated above in the parameters describing turbulent diffusion.
The best-known method for solving equation (1) is a description of the process of
transfer by a Gaussian distribution. This method is coming into wide use and is a
convenient tool for practical workers. However, if in the propagation process for
such an impurity there is no symmetry and the vector field of velocities is a
function of time, the volume of the computations when using a Gaussian distribu-
tion is many times greater than with the direct solution of (1).
The analogy between the processes of transport of momentum, heat and mass makes it
possible, in solving equation (1), to use the same numerical methods used in solv-
ing problems in hydrodynamics. One of the most effective methods in this case is
the Yanenko "fractional steps" method [2] in the form of geometric splitting. On
the basis of this method it is. possible to propose a difference scheme for the
solution of equation (1), set forth in a study by Schenk [3].
We note that the solution (1) is based on meteorological data on the wind at dif-
ferent levels and a semiempirical representation of turbulent transport. Accord-
ing to the empirical expressions, the KX, Ky, KZ values can be assumed proportion-
al to velocity and distance in the coordinate.
Formulation of Problem
The processes of transport of contaminating substances correspond to the mixed
problem for equation (1). With equation (1) taken into account, it is possible to
write the following differential equation:
dc dK,, dc V_ O/i ~ dc ~r _ (2)
- or - ~V.r - d.Y ~ d + a~ ) dy ( ~'s dz , o:
_ k,.Y a-c + K a-c +K= a�=~ ,
dr- ay~ v:- t Q.
The boundary conditions are stipulated by the equations
dc = 0
an I r-k ~ ~3)
0
ov 108 ~
61
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
K d" 4-uC =0 Y dV ' D~
C =0 1 oe
(3)
and use the notations:
the neglecting of transport at the free houndaries FR;
total absorption at the upper boundary of the considered region of propagation;
allowance for precipitation and vertical diffusion at the earth's surface.
The initial propagation of the impurity is stipulated by the condition
C (x,y,z,t-o)=o1 I {.FR-B�
(4)
The rate of dry precipitation of the impurity onto the earth's surface is taken
into account by the parameter P .
Splitting Scheme
The following splitting scheme can be given for differential equation (2):
t oc
~1dt -Q=~+ -
(V.,-aT -KXT =0,
(s)
d' C= 0
Y
a sr ~y - dy dy - Ky dy~ ,
I ac ac a2c
4 or +(V~_dKz) d: -K=o:= =0.
Such a splitting scheme corresponds to geometric splitting. Each equation in the
system (5) is solved by the scalar three-point splitting method. The solution of
equations (S) gives a full description of the process of transport of impurities.
Determination of Transboundary F1ows of Impurity
The introduction of linear sources for determination of transboundary flows of im-
purities is extremely desirable (2). Having a solution of equation (3), the inten-
sity of the linear source can be determined using the integral
114D = S ~cv�dydr, (6)
Y,; a
where vn is the projection of velocity onto the normal to the surface element
dydr1; the 'L coordinate is situated on the integration surface bounding the space
of propagation of the impurity.
62
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFF1ClAL USE ONLY
The following expressions are correct for the intensity of the flows across the
boundary of the space of propagation of the impurity
r11 VD _j c V i d Ar.
1'A
1VJhID - `cV, d Ai,
/`%.1
1l'IRD - f cVi d A;,
RA
, MLD- ScV; dA,,
� LA
MUD =j K,, y I B d At.
(7)
The subscript i means that y, V and A are vector$. Using linear sources as func-
tions of !'t it is easy to determ;ne the interregional flows.
Computation of Reliable Mean Values
In order to monitor the level of concentration of substances contaminating the air,
including the prevention of dangerous situations, in principle it is necessary to
know the fields of concentrations at any stipulated moment in time. However, in
riany cases it is possible to limit ourselves to the mean values for the sutficient-
ly long time period T, for example, for a year. Solving equation (2) for the
meteorological conditions prevailing during the period T and computing the inte-
gra.'
M- T f NidT, (8)
we obtain the reliable mean value of the transboundary flows.
Similarly it is possible to determine the mean concentration in the propagation
region T
c (.C, y, z) _ -T1 cdt. (9)
u
These mean values are of considerable importance for many practical problems. The
choice of the time interval T is determined by the nature of the problem.
Use of Computers
The splitting scheme used makes it possible to employ modular programming. In par-
ticular, the program was developed for a BESM-6 electronic computer. The program
makes it possible to solve the equations in a three-dimensional grid with 12,750
points of intersection; it occupies 20 000 machine words in the operationa.l memory
so that there is a reserve for broadening the propagation space. Ninety seconds of
computer time is required for one iteration in the mentioned grid. Thus,
63
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFFICIAL USE ONLY
computations of the transport of an impurity over a period of 24 hours requires
72 m;.nutes of computer time (time interval of 30 minutes). Using a curve plotter
the isolines. of concentrations are constructed in a regular square grid. The
transboundary flows of impuriti:es: are prtnted out in the form of a table (in kg/
sec).
The program was used successfully in computing transport over short, intermediate
and great distances. In particular, the contamination of a residential quarters
by the emissions of an industrial enterprise situated near this microregion was
computed. Computations were made of the propagation of contaminating substances
over large industrial regions, as well as over the entire territory of the GDR.
liere, in addition to the fields.of concentrations, it was possible to establish
the magnitude of the flows between regions and across the boundaries of the coun-
try for specific meteorological conditions.
Improvement of Model and Algorithm
The results indicate the fundamental applicability of the developed model and solir
tion method. The following refinements of the model and allowance for additional
information are planned for a more adequate reflection of the true pattern:
transformation to curvilinear coordinates;
al'_owance for relief of the underlying surface;
allowance for additional meteorological parameters (changes in height of the
mixing layer in space and time, zones of precipitation and its intensity, and also
characteristics of the boundary layer);
determination of the rate of dry precipitation of impurities as a function of
space and time (settling on water surfaces, snow, forested areas, etc.);
more precise modeling of specific processes of chemical transformations of an
impurity, chemical and physical processes accompanying its washing out;
use of aircraft measurements for deteYmining the vertical distribution of an
impurity.
BIBLIOGRAPHY
1. Izrael, Lysak, Nazarov, Pressman and Rjaboshapko, "On National Observational
System and Evaluation of the I.ong-Range Transmission of Pollutants," INTER-
NATIONAL SYMPOSIUM ON GLOBAL INTEGRATED MONITORING OF ENVIRONMENTAL POLLUTION,
Riga, December 1978.
2. Janenko, DIE ZWISCHENSCHRITTMETHODE ZUR LOSUNG MEHRDIMENSIONALER PROBLEME DER
MATHEMATISCHEN PHYSIK, Springer Verlag, Berlin, Gottingen, Heidelberg, 1969.
3. Schenk, NLiMERISCHE BEHANDLUNG INSTATIONARER TRANSPORTVORGANGE, TH, "C. Schor-
lemmer" Leuna Merseburg, 1978.
4. Schenk, DREIDIMENSIONALER INSTATIONARER WARMETRANSPORT IN DURCHSTROMTIII ROHREN,
ZAM[d, 1977.
5. Schenk/Seidel, "Numeris che Simulation von Kurz- und i,ang-strecken transporten,"
BERICHT DER AG SCHADSTOFFAUSBREITUNG BEIM IfW, Berlin, 1978.
6. Schenk, "Instationarer, dreidimensionaler Schadstofftransport in freier Atmo-
sphare," LUFT- UND KALTETECHNIK, 1978/1 UND ANWENDUNG DER DIGITALGRAFIK...,
1978/2.
64
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE O1NLY
UDC 551.(466+521+55)
TWO-FREQUENCY MICROWAVE RADIOMETRIC METfIOD FOR DETERMINING WIND VELOCITY
- FROM A SATELLITE
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 59-67
[Article by L. M. Martsinkevich, candidate of physical and mathematical sciences,
State Scientific Research Center for the Study of Natural Resources, manuscript
received 26 Jun 80]
[Text] Abstract: The article is devoted to a deter-
mination of wind velocity and wave height
(with an infinite or known fetch length)
from the "wave-covered sea surface - atmo-
sphere" system thermal radioemission measured
from a satellite. The author gives examples of
computations of wind velocity using data from
aircraft microwave measurements made during
the period of the Soviet-American "SAMEX-76"
experiment in the Pacific Ocean.
Different aspects of the influence of sea waves on the thermal radioemission of
the sea surface have already been studied for 15 years by many scientists in our
country and abroad. Most of this work has been based on the use of data from sur-
face and aircraf t microwave measurements. The authors of [1] and [5] came closest
to an interpretation of the results of satellite measurements of the microwave
radiation of the wave-covered sea surface.
This article is devoted to a solution of the inverse problem: determination of wind
velocity and wave height (with an infinite or known f etch length) on the basis of
the thermal radioemission of the "wave-covered sea surface - atmosphere" system at
two wavelengths as measured from a satellite.
The fundamental premise ensuring the possibility of solving the problem is a theor-
etically and experimentally established law: when sighting at definite angles
(close to the nadir or equal to 40-50�) the contribution of the geometric compon-
ent of waves to thermal radioemission can be neglected and it is necessary to con-
sider only the contribution of the foam ccver forming with the destruction of waves
and related to the wind velocity which forms the waves.
We will write a solution for the atmospheric radiation transfer equation:
65
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
H N
ds H - ~ 7 ds
=q, Te � ~T (z) ; e -
;I
[Cy M = tot(al) ]
- I ~ dr
m '
X i T(z)�;e
u
dz + (1 - =eyJi:
N
- 1 Y dt
d2 e a ~
(1)
where T is the temperature of the sea surface, T(z) is air temperature, Yis the
coefficient of attenuation of radiation in atmospheric gases, H is the altitude at
which the radiation detector is situated, z is the vertical coordinate.
Modeling the sea surface by a smooth water surface, the -V-th part of which is cov-
ered by foam, we write the total emissivity of such a surface Ftot in the form
of the sum � totE, (1 - _V) +F- foam v, (2)
where 6 is the coefficient of emission of the sea surface free of foam, Efoam is
the coefficient of emission of a sea surface completely covered by foam.
For brevity introducing the notations
r~ N
- dz H - I ; dz
T, - Te T: T(z) ; P ' dz; (3)
�;ds - ~ ;di
79 - ~ r(; ) e dz e~t
and adhering to the calculations made in [4], we write
Tbr - T2 - T3 ' P, (Tl - T3) - v ( F, foam ' F-) (T1 - T3), (4)
where Tbr is the radiobrightness temperature measured from a satellite.
The expression on the right-hand side of the equation represents the increment of
radiobrightness temperature of the sea surface, the 't/-th part of which is cover-
ed by foam, relative to the radiobrightness temperature of a smooth, foam-free
surface (radiobrightness contribution of foam).
D. T. Matveyev, on the basis of data from the
Uy American aircraft data and with the use of
the radiobrightness contribution of foam by a
measurements at the nadir [6]:
A TI =3� 10-4 Ti.-`~..;
66
"Cosmos-243" satellite, supplemented
a dynamic model of foam, approximated
dependence on wind velocity with
-3) 2 cxp[-n(l'-lO) (5)
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
where '1 is the wave length, in cm, V is wind velocity in m/sec at a height of 20
m, a= 0 when V< 10 m/sec and a= 0.032 when V710, m/sec.
This approximation agrees well with data from experimental observations in the
range of wind velocities from 10 to 30 m/sec; roughness effects can exert a consid-
erable influence when V< 10 m/sec.
~ Writing equations (4) and (5) for the increment of the emission coefficient d 6 and
equating their right-hand sides, we obtain
[S!= br] Tw-T.,-7:,- z (7',-T,1 _3 �l0-�).-u.s(V-3)' exP1- n(L'-10)J. (6)
T, - 7:;
For solving a system of two such equations at two wavelengths it is necessary to in-
troduce additional equa.tions relating to one another the components of the solution
for the radiation transfer equation, taking atmospheric radiation into account.
Due to the impossibility of establishing analytical dependences, here it was neces-
sary to go the way of f inding the empirical re?..ationships.
In order to investigate the atmospheric components of solution of the radiation
transfer equation and establish the relationships between them we prepared a pro-
gram for an electronic computer and carried out computations of atmospheric thermal
radioemission from the sea surface to the upper boundary of the atmosphere with the
use of radiosonde data during the period of the Soviet-American experiment "SAMEX-
76" in the Pacific Ocean. An altitude 16 km was used in the computations as the
upper boundary of the atmosphere; above this level thermal radioemission can be
neglected.
The computations were made by the method described in [3]. In the computations an
allowance was made for the absorption of radiation in oxygen, water vapor and drop-
let-liquid clouds; cases with precipitation were not considered.
An analysis was made of the characteristic emission of the atmospheric layer at dif-
ferent wavelengths (from 0.8 to 3.2 cm) between the sea surface and the radiation
detector T2, which can arbitrarily be called "radiation upward" (the second term,in
the solution of the radiation transfer equation); the radiation flux from the en-
tire atmosphere, which can be called "radiation downward"; attenuation of atmosph-
eric radiation.
An analysis of the results of computations indicated that there are some rather
stable empirical relationships between the mentioned components of radiation at the
upper boundary of the atmosphere.
Thus, the atmospheric radiation fluxes upward and downward differ from one another
usually by not more than 2%. If we write the relationship between atmospheric radi-
tion upward T2 and its radiation downward T y in the form
_T~=cTzi (7)
the c coefficient is equal to unity with the third significant digit after the dot.
This coefficient is dependent on frequency and the angle of sighting and for meas-
urements at the nadir is determined using the following expressions:
67
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICtAL USE ONLY
for a wavelength 0.8 cm
T~=1,0037�O,U1)17 TZ;
for a wavelength 1.6 cm
Tj=1,0012�0,0012 T2.
(8)
(9)
The indicated scatter of T~ values with their substitution into the solution of
the transfer equation leads to insignificant errors in the Tbr values, amounting
to tenths of a Kelvin.
From the physical point of view there is obviously a correlation between the char-
acteristic radiation of the atmosphere T2 and the integral attEnuation of radia-
- tioii in the atmosphere exp 'G). There are indications of the existence of this
relationship in [2, 31. The dependence was sought in the form
H exp (-t) =1-b T2, (10)
where L= J Ydz is the integral coefficient of attenuation of radiation in atmo-
0
spheric gases and the b values for all frequencies in accordance with the computa-
tions fall in the range 0.0036-0.0038. This relationship has a very weak frequency and angular variation; with sighting at
angles from 0 to 40-45� the latter can be virtually neglected.
For measurements at the nadir using the indicated group of materials we found the
following values of the b coefficient:
for a wavelength 0.8 cm
b = 0,00371 �U,OU010; (11)
for a wavelength 1.6 cm
h = 0.00369 � 0,00009. (12)
An analysis of data from computations of thermal radioemission of the entire atmo-
sphere to its upper boundary for the regions of the "SAMEX-76" polygons indicated
that there is a rather srable frequency dependence for characteristic atmospheric
radiation. It is very difficult to describe analytically; however, if we exclude
from consideration a sufficiently narrow frequency band near the channel 1.35 cm
(where the radiation is unambiguously related to the radiation in the adjacent chan-
nels) it is possible to find a rather stable coefficient relating the T2 values in
fixed pairs of channels in the range from 0.8 cm to 3.2 cm. In other words, it can
be written approximately that
T2 - kT,) , (13)
~
68
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
where ~,2 > k, k is a dimensionless coefficient less than unity. For the pair of
wavelengths 0.8 cm and 1.6 cm k is equal to 0.402, for the pair of wavelengths 1.6
cm and 3.2 cm k is equal to 0.255. Here the k coefficient is identical for all
sighting angles in the range 0-45�. The derived expressions were checked on the basis of independent computations of
thermal radioemission of the atmosphere using radiosonde data (130 soundings)
fror weather ships in the North Atlantic (1965-1968). The results of the comput-
ations were furnished us through the courtesy of Ye. P. Dombkovskaya and V. V.
Ozerkina. Checking indicated a good agreement of atmospheric radiobrightness tem-
peratures at a wavelength of 1.6 cm, computed using radiosonde data and computed
from the temperature value at a wavelength of 0.8 cm in accordance with (13).
It was found that the relative error in determining T2 (1.6) on the basis of stip-
ulated T2 (0.8) in 40% of the cases is less than 5%, in 74% of the cases is less
than 10%, in 90% of the cases is less than 15% and in 98% of the cases is less
than 20%.
Checking of dependence (13) for pairs of more long-wave channels indicated that the
- relative error in determining T2 on the average increases somewhat, but the abso-
lute error is small. For example, whereas for the pair of channels 1.6 cm-3.2 cm
according to data from weather ship B the mean relative error is 15.7%, the mean
absolute error is equal to only 1.16 K, which falls within the limits of accuracy
of modern radiometric instrumentation. According to data for ship L, for which
the relative error for this pair of channels was small, the absolute error was only
0.38 K.
Thus, expression (13) can be used in computing characteristic atmospheric radio-
emission in one of the stipulated channels on the basis of the known radiation in
the other channel in the range of wavelengths from 0.8 cm to 3.2 cm, with the ex-
ception of the region near the channel 1.35 cm.
Checking of expression (10)�using this same independent material indicated that the
values of the b coefficient do not fall beyond the limits of the already mentioned
values.
It goes without saying that the derived expressions require refinement. However,
the good correspondence between the results of computations for the two groups af
independent materials ("SA14EX-76" expedition and the weather ships) indicated a rel-
ative stability of these relationships and the fundamental possibility of their
use in global problems. It is entirely reasonable to assume that variations of the
coefficients in these expressions will have a"regime" character and a rigorous
geographical localization; accordingly, it is desirable to analyze the statistical
material in sea polygons.
Taking expressions (7), (10), (13) into account, the components for the trans-
fer equation T1 and T, determined by formulas (3), for the stipulated pair of
wavelengths 2l1 and ~2 (here A2 >/'k1) can be written in the following way:

Tj,.'=T ` rl-b,T1k ~
69
FOR OFFICIAL USE ONLY
(14)
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFFICIAL USE ONLY
r -T(1-b:kr 1;
~ 2,,)(15)
T - T c, r 1- b, Tg
3A' 21' ~ (16)
T3 = T,, c, / 1- b,k T,
l z;., 1
(17)
uzhere the subscripts 1 and 2 correspond to the wavelengths /11 and )~2.
The system of two equations of the type (6) for two wavelengths after the substit-
ution of expressions (14)-(17) into them, is written in the following way
T - T2. [1 - `;.1 Tbi + ci (1 + T2i ci b, (I - '�,1)
~ A' ~
T- T2~ ( Tbl + c, T2k bi ci
l ,
- 3. 10_4 11 0.3 ( V- 3)' exi) V- 10)
(18)
rA~^-j;.1T-.aTZTb.,=-c.( lk'=71 c..b~, ( l
T- kT ( Th.: j� c.. k= 1,2 b, c..
l41 - 1;.1 (19)
=3 .10_'X;�3
(V-3)' exp (-a(V- 10)1.
Thus, we obtained a system of two transcendental equations with two unknowns: wind
velocity V and the characteristic radiation of the atmosphere at the wavelength
a 1- T2A1. This system can be reduced to one algebraic equation with one un-
known T2 T 1, finding the ratio of equations (18) and (19). We obtain
>.1 �'D( rtt;. T -TZz [I `ai Tbl-~c,(l -=;.i~~ - 1.~ ti bi(l
( ) ~ ' '
T- TZ 1 Tbl = ci) + T2. bi ci - (20)
j. i
T - kT2~l [l Tb~ + + k: T r2 6=
T- k T21 ( Tb1 c,) ki 7''l b-j C2 '
I i
The transformation of the equation leads to an unwieldy fourth-degree equation
which in a general form has no solution. However, the left- and right-hand sides
of equation (20) are different functions of x= T2 A1; therefore, the solution is
conveniently obtained graphically.
It is interesting that the right-hand side of the equation is the increment of the
radiation coefficient at the wavelength a 2; therefore, on the computation nomogram
it is immediately possible to read the two parameters T2~ 1 and 16�a2.
70
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Figure 1 shows a computation nomogram which is a realization of the graphic solu-
tion of equation (20) for wavelengths 0.8 cm and 1.6 cm. Here along the x-axis we
have plotted the increments of radiation at a wavelength of 1.6 cm due to the foam
cover and along the y-axis the values of characteristic atmospheric radiation at
a wavelength 0.8 cm. 'I'he series of solid curves represents a graphic realization
of the left-hand side of the equation with different T2 (0.8) values and with a
definite value of the measured radiobrightness temperature of the "ocean-atmo-
sphere" system at a wavelength 0.8, as indicated on the curve; the series of dash-
ed curves represents a similar realization of the right-hand side of the equation
with a definite value of the radiobrightness temperature measured at a wavelength
1.6 cm.
.
r~

r'
,
. ,
. ~
,
� ~ .
. ~ .
~
' \ ~ \ \
\ \
 ~ ',D \ ~
~ ~ \ \ \
Fig. 1. Nomogram for determining characteristic atmospheric radiation at a wave-
len gth 0.8 cm and the increment of the radiation coefficient due to waves at a
wavelength 1.6 cm on the basis of satellite measurements.
The point of intersection of the corresponding pairs of curves for wavelengths 0.8
and 1.6 cm determines the atmospheric radiobrightness temperature at a wavelength
0.8 cm, which is read on the y-axis (T2(0.8)), and the value of the increment of
the coefficient of radiation of the wave-covered sea surface at a wavelength 1.6
cm (L1S1.6), which is read on the x-axis.
With the determined value T2(0.8) can be entered into equation (18) or (19) and
it is possible to compute the wind velocity over the sea.
Working with the nomogram it can be shown that an error in measuring radiobright-
ness temperature by 1�K leads to an error in determining wind velocity of 3% and
- an error of 1.50K leads to an error of 6%.
71
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
: 0,01 70- 0,06 0,06 0,10 0,11 0,14dCrte

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OE'F[CIAL USE ONLY
hill M r-
VMIc V m/sec h,y~ ~
z zs - ~

15
T� K
f60 ~
Tr ( 0, 8k- ~
T~40
d0 110
� �
00
."vYSr.sK 91
0145 hours
T:f4e~:
ti.� �
f9 S1 S.~ SS
h~� M
f 6)
7 m/s ec,
vM1C
~
1 15
~
.
:
s r~orK
~i(1,6)
JO r130
JylSnun 71 13
0315 hours

~ XZ
~
-r-,
.

I `
A =br
3H = cons
21 ?j
Fig. 2. Spatial changes in radiobrightness temperature at wavelengths 0.8 cm and
1.6 cm and characteristics of state of the sea surface computed from it along
flight line of I1-18 aircraft. a) 11 September, b) 13 September 1976. 1) wind velo-
city computed from pressure gradient, 2) considerable wave height, computed from
these wind velocity values.
The determined wind velocity value can be used in computing wave height in a case
when fetch length can be considered infinite (completely developed waves) or
known.
Thus, it was possible in general form to solve the inverse problem: determine wind
velocity over the sea (and from it wave height) on the basis of the microwave
radiation measured from a satellite.
Due to the lack of satellite synchronous microwave measurements in the selected
channels the developed method for salving the inverse problem for determining
wind velocity was applied to data from aircraft microwave measurements from aboard
an I1-18 aircraft of the Main Geophysical Observatory during the period of the
Soviet-American experiment "SAMEX-76." In the applieation of the "satellite" al-
gorithm using materials from aircraft measurements the results of these measure-
ments were "raised" to the satellite flight altitude by taking into account (on the
basis of computations) atmospheric radiation in the layer 4 km (aircraft flight al-
titude) - 16 km (radiobrightness analogue of the atmospheric upper boundary) for
each day of observations and radiation above 16 km, which was estimated approxim-
ately, on the basis of ineasurements with an upward-directed radiometer from aboard
an American "Convair-990" aircraft during the period of the first Soviet-American
"Bering" microwave experiment [8].
Figure 2 shows examples of application of the method for determining wind velocity
on the basis of microwave data for specific runs of the I1-18 aircraft.
72
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-04850R000404030065-3
FOR OFFICIAL USE ONLY
Figure 2a shows the spatial variation of radiobrightness temperatures at wave-
lengths 0.8 and 1.6 cm and the values of characteristic atmospheric radiation at
a wavelength 0.8 cm, wind velocity along the flight line and the height of the
waves (considerable) for the aircraft run of 11 September 1976 from 0145 hours to
0155 hours at an altitude of 4000 m with a sighting angle of 40� computed on their
basis. The increments of the radiation coefficient, computed for an angle of 40�,
were reduced to increments for the nadir by use of the Stogrun angular dependence
[7]' 6�40� - 0.9365 ,A�flo� It was necessary to dc; this since the D. T. Matveyev
- approximation (5) relates to radiobrightness increments measured at the nadir.
Unfortunately, no instrumental measurements were made to determine wind velocity
from aboard an I1-18 aircraft: therefore, as a comparison with the computed velo-
- cities, Fig. 2 gives the wind velocities and wave heights for three points com-
puted from the pressure gradient on a surface synoptic chart; the heights were
computed with allowance for fetch length. The segments of the horizontal straight
� lines near the wind velocity values and wave heights determined from the pressure
gradient mean that it is impossible to obtain a more precise localization of the
point of determination from a small-scale map.
The errors in determining wind velocity on the run are 20%, 5.2% and 5.6% for the
_ run, provided that the wind velocity values, determined from the pressure gradi-
ent, are assumed to be true; the errors in determining the considerable height of
waves are equal to 31.8%, 10% and 9.7% respectively. Thus, the errors in determin-
- ing the characteristics of state of the sea surface in the second and third cases
are small. The greater error in the first case does not hinder determining wind
conditions as storm conditions.
In the segment of the run corresponding to the extreme right-hand side of Fig. 2a
the wind velocity and the wave height decrease. This tendency corresponds to an
objective evaluation of the hydrometeorological conditions with advance toward
the work region of the ship, which was situated 80 km from the represented segment
of the run. The wind velocity, measured from aboard the ship at the time of pas-
sage of the aircraft along the mentioned flight line, was already 13.5 m/sec and
the considerable height of the wind component of waves according to data from the
radar attachment to the "Don" radar set was 2.7 m(unfortunately, it was impos-
sible to obtain wave recorder measurements of waves on this day). The indicated
height agrees fairly well with the computed value 3.4 m, especially if it is taken
into account that the wave height is a more conservative characteristic than the
wind.
Figure 2b shows the spatial variation of radiobrightness temperatures at a wave-
length 0.8 and 1.6 cm and the values of characteristic atmospheric radiation at a
wavelength 0.8 cm, wind velocity along the route and considerable wave height, com-
puted on the basis of these radiobrightness temperatures, for another run of the
I1-18 aircraft which took place on 13 September 1976 (from 0315 to 0425 hours) at
an altitude of 4000 m with orientation of the antenna for the radiometers to the
nadir (8 = 0�). Here, as a comparison, we have also given the wind velocity val-
ues computed at two points on the basis of the pressure gradient and the corres-
ponding (considerable) wave heights.
73
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
The wind velocity on the segment of the run represented in the right part of the
figure is reduced, which, as in the first case, corresponds to the real situation:
the aircraft is approaching the center of a cyclone.
The relative error in determining wind velocity on the basis of microwave two-chan-
nel measurements for the first control point in the measurement scheme was 1.8%.
and for the second 6.7%; the error in determining wave height was 2.6% and 15% re-
spectively.
Due to the scatter in T~ and b values, determined by equations (8), (9), (11),
(12) the errors in determining wind velocity will deviate from the mentioned mean
errors by 1-2%, which exerts virtually no influence on the possibilities of the
method.
The described method for determinin; wind velocity from the measured radioemission
of the "sea surface-atmosphere" system, as indicated above, in essence can be ap-
plied with any pair of wavelengths in the range 0.8-3.2 cm, except for a narrow
frequency band near ;k = 1.35 cm. The choice of the optimum pair of wavelengths
can be the subject of a special investigation.
The principal difficulty in application of the method is that in the computations
it is necessary to know the absolute radiobrightness temperature values and this
requires very correct signal calibration.
Further work in the direction of improvement of the method should proceed along the
lines of a refinement of the relationships between the atmospheric components of
radiation, the careful choice of the statistics of such relationships in differ-
ent regions of the world ocean and refinement of the dependence of the increment
of radiobrightness temperature due to the foam cover on wind velocity.
It is also necessary to check the method on the basis of mass material from satel-
lite microwave measurements.
BIBLIdGRAPHY
1. Basharinov, A. Ye. and Shutko, A. M., ISSLEDOVANIYE VZAIMOSVYAZI KHARAKTER-
ISTIK POLYA TEPLOVOGO RADIOIZLUCHENIYA S SOSTOYANIYEM POVERKHNOSTI AKVATORIY
(Investigation of the Interrelationship of Characteristics of the Field of
Thermal Radioemission With the State of a Water Surface), PREPRINT IRE AN SSSR
(Preprint of the Institute of Radioelectronics USSR Academy of Sciences), No
63, 1971.
2. Gorelik, A. G., Raykova, L. S. and Frolov, Yu. A., "Micrometric Radiometer Meth-
ods for Measuring Humidity in the Lower Troposphere," METEOROLOGIYA I GIDROLOG-
IYA (Meteorology and Hydrology), No 5, 1975.
3. Dombkovskaya, Ye. P., "Determination of Sea Surface Temperature and Atmospheric
Moisture Content Using Measurements of Thermal Radioemission of the Earth-Atmo-
sphere System From Artificial Earth Satellites," TRUDY GIDROMETTSENTRA SSSR
(Transactions of the USSR Hydrometeorological Center), No 50, 1969.
74
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00854R004400030065-3
FOR OFFICIAL USE ONLY
4. Martsinkevich., L. M., "Analysis of the Characteristics of State of the Sea
Surface Exerting an Influence on Thermal Radioemission in the Work Area of
the Scientific Research Weather Ship 'Pribo;' in the Soviet-American 'Bering'
Experiment," TRUDY ZAKLYUCHITEL'NOGO SIMPOZIUMA PO ITOGAM SOVETSKO-AMERIKAN-
SKOGO EKSPEDITSII (Transactions of the Final Symposium on the Resulta of the
Soviet-American Expedition), Leningrad, 12-17 May 1974, Leningrad, Gidrometeo-
izdat, 1975.
5. Martsinkevich, L. M. and Matveyev, D. T., "Correlation Between OLtgoing Micro-
wave Radiation and the State of the Sea Surface (According to Data From the
'Cosmos-243' Satellite)," METEOROLOGIYA I GIDROLOGIYA, No 8, 1971.
6. Ma.tveyev, D. T., "Analysis of the Results of Radiothermal Sounding of the Sea
Surface During a Storm," METEOROLOGIYA I GIDROLOGIYA, No 4, 1978.
7. Stogrun, A., "The Emissivity of the Sea Foam at Microwave Frequencies," JGR,
Vol 77, No 4, 1972.
8. Wilheit, T. T., Towler, M. G., Stomback, G. and Gloersen, P., "Microwave Radio-
metric Determination of Atmospheric Parameters During the Bering Sea Experi-
ment," SOVETSKO-AMERIKANSKIY EKSPERIMENT "BERING" (Soviet-American "Bering"
Experiment), TRUDY ZAKLYUCHITEL'NOGO SIMPOZIUMA PO ITOGAM SOVETSKO-AMERIKANSKOY
EKSPEDITSIII, Leningrad, 12-17 May 1974, Leningrad, Gidrometeoizda~t, 1975.
75
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
UDC 551.(465.4+466.467)(268)
DIAGNOSTIC MODEL OF WATER AND ICE CIRCULATION IN THE ARCTIC BASIN
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, 24ar 81 pp 68-75
[Article by V. I. Ponomarev and L. A. Glazova, Main Geophysical Observatory, manu-
script received 20 Jun 80]
(Text) Abstract: A diagnostic variant of a joint
model of motion of ice and water under the
influence of wind and horizontal nonuniform-
ity of the field of masses in the ocean is
examined. The authors present a method for
computing the vertical component of velocity,
taking into accoimt inertial forces. The re-
sults of computations of ocean circulation
in the Arctic basin with a detailed analysis
of vertical movements are given.
Introduction. Concepts concerning the circulation of water and ice in the Arctic
basin were obtained on the basis of observations made from the time of the remark-
able drift of the "Fram" [13]. On the basis of data on the drift of ships and polar
stations it was established that the principal features of macroscale circulation of
ice and surface waters here are the Transarctic Current, crossing the Arctic Easin
from the Chukchi Sea to Fram Strait, and the anticyclonic circulation over the Can-
andian Basin (see [1, 12]). The direction and velocity of propagation of Atlantic
- waters were evaluated for the most part by indirect methods, by an analysis of the
temperature and salinity fields [8, 12]. The movement of deep waters in the Arctic
Basin was the least studied due to the low measurement accuracy. The nature of the
vertical circulation of waters in the entire thickness of the ocean also has remain-
ed virtually uninvestigated by empirical methods.
Theoretical investigations made it possible to obtain new concepts concerning the
circulation of water and ice in the Arctic Basin. Hydrodynamic models have been
the principal tool of these investigations. Using barotropic models [3, 10, 111 it
was possible to describe the integral mean-term circulation of water and ice. Us-
ing diagnostic methods and models [4, 9] a study was made of the vertical struc-
ture of currents. The prognostic model in [14] made it possible to describe the
process of formation of climatic fields of currents, temperature and salinity in
the Arctic Ocean.
All these models best describe currents in the upper layer of the ocean. However,
it is of interest to make detailed computations of currents in the layers of Atlan-
tic and deep waters, investigate the character of vertical movements and the
76
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
interannual variabilitv of circulation. For this purpose it is necessary to have
models more precise than modele wsed earlier, taking into account the peculiarit-
ies of the Arctic basin, the presence of an ice cover and the weak intensity of
currents (the velocities of macroscale currents are several centimeters per second-,
and in the deep layers fractions of centimeters per second).
First of all it is necessary to examine the most complete and sufficientl,y precise
diagnostic model, making it possible to compute the three components nf current
velocity in the entire thickness of the ocean on the basis of the stipulated
density field and this article is devoted to this subject. Since within the frame-
work of the "POLEKS-Sever" national program, beginning with 1973, regular observa-
tions have been made of the temp.erature and salinity fields, such a diagnostic
model can be used in computing circulation for each specific year and in investigat-
ing its variability.
Principal expressions of model. As in the diagnostic models used earlier (see [7]),
the initial expressions are the continuity equation f.or an incompressible fluid
and the equations of motion in the hydrostatic nwj " nesq approximations. The
derivation of the expressions for the horizontal of current velocity
is accomplished in the usual way (the assumption. q`uasistationarity of current
velocity a u/ a t =d v/ d t= 0 is made). The equations of motion, written taking
into account the equations of statics, are integrated vertically from the surface
to the ocean floor. At the ocean surface (z = 0) and at its bottom (z = H) an al-
lowance is made for the boundary conditions for frictional stresses:
t I ` � - - .N
.r)' `.t')' I s = N y .
Then a standard procedure is carried out: the stream function of total flows is in-
troduced and the level slopes W /0 x, d; /d y) are excluded from the stationary
equations of motion using the vertically integrated equations of motion. This gives
the following expressions for the fiorizontal components of current velocity (u, v):
l d- -O --N I d�: g ri d?
~y y (lL -
tt = - p,~l !H H dy Z ~
(1)
+ g � dP ds I v _L , ( "I ' v i"-, t ( ,v (v) dz:
Pot J dY ~ !H v.r ) l lH.
l d:.r rf 1 J~~ g ft d?
V - Z - lt1
p�! d: ~,,lN + H v.r c� l; f o.r
u
gdp AL A, 'J�.' 1
~ ltZ - - , tt ~ ( ) L - (11) ,
fil l, d.r ( ~N oy [
~
tl
1H (u ) dz,
u
r) l ~ d
y ~ " dz '
where N ~ ) = tt d. d-~� +z
77
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
If the initial equations of motion or expression (1) are reduced to dimension-
less form, wa obtain two small parameters: one before the inertial force ( E 1)
and the other before the force of horizontal turbulent exchange (S 2). Expanding
u and v into a series for a small parameter and taking into account only the firet
two terms of the series, we expressed the nonlinear inertial accelerations through
the f irst the linear approximations for the velocity components. (The terms con-
*_aining the product F1� 2 are al.so not talcen into account because they are of the
same order of magnitude as terms with ~ i). Thus, by u, v~n (1) we understand the
sum of the first f ive terms on the right-hand side of the expressions for u and v.
The expressions for current velocities (1) satisfy a necessary condition: in the
integration of the left- and right-hand sides of (1) from the surface to the bot-
tom of the ocean these expressions become equal to identity; SXNK SX, Sy Mr. Sy,
where SX, Sy are components of the total flow vector.
In determining the integral stream function (41) entering into (1) use is made of
the nonlinear vorticity equation, derived by cross-differentiation of the equa-
tions of motion integrated vertically from 0 to H.
The expression for vertical velocity is derived by the substitution of expressions
(1) into the continuity equation, integrated vertically from 0 to z. Under the con-
dition of a"hard top" on the ocean surface (wlz = 0= 0) the expression for w has
the following form:
Po rotZ -T - PJ I 1- H I rotz
(I-
1 1
H aH (2)
- r~ rots + H(u y ax +
l
dH 1 g N
~ d! s
vH dr j~ P,, 12 dx H f~Z - H) d ~ ay ltZ
u
~ a a~ z " r
- z) a~ dE) - ay (H,f t= - N) oz dz - f(E - s) dx dt
) +
u , \ ~ u
/
L rs -dQs = QZ 1
(-jQ, u ay )dy -iH u dz +v dy :dz-
o , � )
A : , ,
- ~ A~s dt - H a1,~~ ,
~
where 4. _ dv _ du
� dx oy '
78
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034465-3
FOR OFFICIAL USE ONLY
uH, vH are elements of the barotropic component of current velocity, equal to the
velocity at the upper boundary of the bottom friction layer (the sums of the sec-
ond, third, fourth, seventh and ninth terms on tlie right-hand sides of formulas
(1))�
In the derivation of expression (2) the inertial forces entering into (1) were tak-
en into account partially: only the principal terms associated with vorticity
advection (S2Z) were retained (the term containing aW/aX - aw/Oy is not taken
into account, since it is more than an order of magnitude less than vorticity ad-
vection). Thus, the derived expression for vertical velocity (2) takes into ac-
count all the principal factors responsible for macroscale circulation in the
ocean, but this expression is approximate. At the same time, the used approxima-
tior.s are entirely valid and correct for any region in the world ocean, except for
the equatorial zone, where the inertial forces evidently cannot be considered
small (it is impossible to expand u and v into a series of powers of F,1 for this
reason).
It must be noted that the differences between formulas (2) and those used earlier
involve not only an allowance for additional factors, but also the method for tak-
ing the principal forces into account. It follows from the formulas cited in [7]
that the term caused by the divergence of drift currents is not dependent on depth,
that is, in the zonal flow (for example, in the Antarctic Circumpolar Current) the
vertical velocity in the entire thickness of the ocean outside the friction layers
is equal to the sum of the Ekman and anemobaric components w(since dP /a,1 = 0,
dy~/d 0). In this case w does not become equal to zero at the ocean floor and
does not satisfy any other physical condition.
In contrast to the formula from [7] in expression (2) the term containing the Ekman
- vertical velocity has the factor (1 - z/H) and therefore decreases linearly with
depth from a value equal to wE with z= hE to zero at the ocean floor. The terms
associated with the A -effect and the anemobaric vertical velocity also become
equal to zero at the ocean floor. In general, w in (2) satisfies the streamline
flow condition when z= H:
wl _ u dK + ` ti!-I
If N r.r oy (3)
At the upper boundary of the bottom friction layer the vertical velocity is approx-
imately equal to the sum of the friction (wE) and orographic (wor) components.
dH ~ dN
-nerr rot: 1-~ tt rr r).r + (4)
Conditions (3) and (4), denoting the presence of both friction and streamline flow
; on the bottom, are contradictory. This contradiction can be eliminated by intro-
ducing a bolindary layer with allowance for curvature of the underlying surface of
- the bottom. However, in the first approximation for H it is possible to use the
upper boundary of the bottom friction layer and leave the condition (4).
For closing the model relative to frictional stress at the lower surface of the ice
(Z 0CO) we use an extremely simple method for parameterization of the steady
x' y
drift of ice examined in [6]. The ice drift velocities (uo, vo), the samz as
79
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
the characteristics of the planetary boundary layers over the ice and under the
ice, are considered quasistationary. In addition, the equations of motion for
ice are written in a linear approximation without allowance for the horizontal
exchange of momentum:
d'
!lo = 1 - 1 + -g q 0:
, h! Y ! tly " ph! ~ '.r ox ' (5)
that is, the drift velocity of ice can be represented in the form of the sum of
- the wind I
y �v
t' - l
and _ ,Q d :
( tt� ~ - -
gradient drifts [B = wind; f= gradient].
In such a case the frictional stresses at the lower (-LO,'C 0) and upper
surfaces of the ice are determined in the following wayX y x y
yj G, (U, cos x; - Vl sin a,),
= n;~ *A" I1 G; (V; cos :t; + Ui sin 7;1. (6)
, _
Here x is the Karman constant, xl, 4Ci are the geostrophic friction coefficient
and the angle of total rotation of the wind (i = 1) or the current (i = 0) in the
planetary boundary layer (PBL). According to [2], ~Land a are universal functions
of the Rossby number (Ro) and the stratification parameter /A Q(Ro = Gi/.~,zo, zo
is the roughness parameter). The functions 'G(Ro, ~e.o) and pC(Ro, Yp ) were obtain-
ed as a result of solution of a closed system of equations for the PBL and are
given in [2].
In formulas (6) with i= 1 U1, Vl, Gl are the components and modulus of velocity
of the geostrophic wind at the upper boundary oi the atmospheric PBL, with i= 0
Uo, Vo, Go are the corresponding differences between the drift velocities of ice
and gradient currents.
It is assumed that the influence of inertial forces and horizontal turbulent ex-
change on the velocity of macroscale currents is small, the velocities of the gra-
dient current at the surface approximately coincide with the velocities of gradi-
ent drift of ice
g i~' g dy
and the vertical velocity shear in the friction layer under the ice is related
only to the wind drift of ice.
It is known that the steady wind drift of ice is close to isobaric (to Zubov drift)
and deviates from it by some angle Y(see [1]). Substituting (6) into (5) and re-
ducing similar terns, it is easy to represent the wind drift of ice in the form
of the sum of isobaric drift and deviations from it (as was done in [10] with use
of an Ekman model for the PBL with a stipulated turbulence coefficient:
[B = Wli1d] !lu e= Ce (U, cOS ~B -4 V, Slfl vu e*= C. (V, COS iu - U, Sli7 Ju)� (7)
80
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-04850R000400030065-3
FOR OFFICIAL USE ONLY
The expresaions for the wind isobaric coefficient Cwind and the angle Yhave the
Eollowing form:
[B = wind] Cd=M, (1-{-M~ --2M� sin uo)-'
1 ~8~
~
arctg cos al + A1o sin (au - a,)
sio 7, -t- M,, coa (Qo- a,) J�
2
r i 'l.i ~
G
M~ - 'r. h! I i= 1, 2.
where �
.
Thus, the already cited formulation of the diagnostic model of circulation of water
and ice is a closed problem. It should be noted that formulas (8) are a special
case of the expressions for CWind and Y derived in [5] with allowance for the
lateral friction force in the ice cover (FX = -ru, F y =-rv); the expression for
w(2) was Axamined in [6], which gave computations of vertical circulation in the
Arctic basin on the basis of the stipulated density field from [9] (for 1955-1956).
t\N\ r l
t ~
}
~
I r ~ f } 1
f1 ~

~
~
1
.
~ /
~ /
~ ~

~
~ ~
.
.
~

� ~
l

t ~1
Fig. 1. Fields of velocities of gradient curxents at the ocean surface. a) comput-
ed on the basis of data on density for 1958-1959 (from [91); b) computed from den-
sity data for 1975 obtained on the "Sever-27" expedition.
' In this article we give computations of the three components of current velocity
and ice drift velocity on the basis of data on density obtained on the "Sever-27"
expedition (carried out in 1975). The results of these computations were compared
- with the current velocity fields obtained in [6J on the basis of the density field
si
~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
from [9], observed two decades prior to the "Sever-27" expedition.
Computations of u, v, w were made at standard horizons at the points of intersec-
tion of a uniform grid employed earlier in [3, 4, 6]; the grid interval used was
200 km. We also stipulated the discharges in straits on the basis of data from
Timofeyev (see [8, 10]), the bottom relief field cited in [3] and the field of at-
niospheric pressure. Computations were made both with a stipulated mean long-term
atmospheric pressure field, cited in [10], and with a pressure field averaged for
1974, that is, averaged for the period preceding density observations.
Results of computations. First we note that the general conclusions concerning com-
putations of horizontal circulation in the Arctic basin and comparisons of differ-
ent methods for computing currents are given in [4]. The influence of different
factors on vertical velocity with a detailed analysis of all the terms in the ex-
pression for w(2) was examined in [6]. Accordingly, in this study we note only
the principal conclusions pertaining to vertical circulation in the Arctic basin
and we will compare the results of computations of circulation for the density
fields obtained in 1958-1959 and in 1975. These results of computations differ ap-
preciably due to important changes in the salinity field occurring over the course
of 16 years.
a) ~ ~ 0 0
o I
i ~
i r -B'Z i
i ~
o
; a
i ~0
n
Fig. 2. Vertical velocity fields at 500-m horizon. a) computed using density data
for 1958-1959, b) computed using density data for 1976. Regions of ascending move-
mer.ts are shaded. The north pole is designated by a dot.
Figure 1 shows that the Transarctic Current, intersecting the basin, was best ex-
pressed in 1958-1959, whereas in 1975 ttiis current was not concentrated in a nar-
row zone, although everywhere in the Atlantic subbasin there was a drift with a
82
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFIC[AL USE ONLY
velocity of about 1-2 cm/sec in the direction from the Siberian shelf to the Can-
adian archipelago. The current was considerably intensified only in individual re-
gions: along the Mendeleyev Range and beyond the Lomonosov Ridge. In addition,
in 1958-1959 one anticyclonic circulation occupied both the region of the Pacific
Ocean subbasin and the regions adjacent to the Canadian archipelago. Only over the
Canadian basin in the central part of the main circulation was it possible to dis-
criminate two anticyclonic vortices of a lesser scale (see Fig. la). In 1975 the
anticyclonic circulation divided into two vortices; one was situated over the Can-
adian basin and the other., the more intense with respect to velocity and depth, but
smaller in extent, was situated over the Alpha Rise, adjacent to the Canadian arch-
ipelago (Fig. lb).
_ The fields of horizontal current velocity in the layer of Atlantic waters also have
some differences, but less than at the above-lying horizons. It is interesting to
note that in 1975 at the 500-m horizon the circulation over the Canadian basin be-
comes opposite, forming a cyclonic circulation, whereas over the Alpha Rise, as
before, there is an intensive anticyclonic vortex.
In addition, it must be noted that the change in circulation :.s most appreciable
and graphic in the vertical velocity field, which is the most sensitive of all the
hydrological characteristics which we examined. Figv.re 2 shows the vertical velocity
field for the 500-m horizon where it is governed for the most part by the )8 -effect
(fifth term in expression ( 2 ) , which we will denote wg P). In the above-lying lay-
ers of the ocean the fields of the term wgo have the same peculiarities as the
W500 field, but the total vertical velocity w with z
0)
�1
a
to
n
p
u
a
,
V/
a
i
a
i
O
'C N
~rl
N
~O
i"+
U
tA
41
i
O
.C 4
.a
N
>
4-+
O
Cl
(b
11 -H
N
�rl
1-1
,-1
O
r-I
al
Ul
rz f17
r.
(A
ctl
U
m
11
cd
~
W
�
3a
q
Zj ~

~
cd
44
cd
19
Q
~
U
Z
cn c
tl
d
-.4
0
W
y+
~4
Q,
88
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
02 ml/liter = 1.54 0.0876H, (5)
0' 2 ml/liter = 1.8 - 0.087,A H, (6)
where 02 is the mean annual content of dissolved oxygen at the horizon 85/90 m in
the Bornholm depressiony 012 is the same as an average for the horizons 85/90 m
in the Bornholm depression and for the horizon 100/105 m in the Gdansk depres-
sion during the course of the year;,A H is the same as in equation (1).
An evaluation of the quality of the proposed methods for computing different hydro-
logical-hydrochemical elements using equations (1)-(6) is given in Table 1. Its
data show that the proposed procedures for predicting interannual variations in
the intensity of entry of "fresh" North Sea waters through the Danish straits by
means of computations of the depth of the 100/0o isohaline, by use of the corres-
ponding prognostic equations, as an index of the volume of North Sea waters flaw-
ing into the Baltic Sea, computations of their mean salinity and the content of
- dissolved oxygen within the limits of these waters are characterized by a high
guaranteed probability and effectiveness. They can be used for annual correction
~)f the super-long-range background forecast of the anticipated changes in salin-
ity in the bottom waters of tY1,3 sea [5].
In addition, the complex hydrological-hydrochemical prediction of the anticipated
interannual changes in the intensity and character of entry of North Sea waters
is also Qf independent importance, especially for fishing science and practical
work. The regime of bottom and deep waters, their volume, frequency and intensity
of renewal, the absolute values of salinity and dissolved oxygen within their lim-
its are important abiotic factors to a considerable degree determining the condi-
tions for reproduction and survival vf important species of commercial fish in
the Baltic Sea, such as cod and plaice [l, 21.
The broad use of the underflow of North Sea waters as a predictor of sea level
indices in complex forecasting is attributable to the fact that its level is
characterized by a definite universality in the reflection of hydrometeorological
processes transpiring in the sea. In actuality, in characterizing the physical
essence of the latter prognostic relationships (1)-(6) it should be noted that the
principal predictors the sea level and its dynamics are some of the important
indices of most hydrometeorological processes transpiring in the sea. The level
seemingly integrates the consequences of such phenomena as continental runoff,
atmospheric pressure, precipitation and evaporation from the sea surface, which in
one way or another exert an influence on water exchange in the strait zone, and
accordingly, on the nature and intensity of entr.y of North Sea waters into the
Baltic Sea. Hence there is a clear relationship between 4ynamics of sea level and
the relative volumes of inflowing North Sea waters, their solinity and dissolved
oxygen within the limits of these zones.
Thus, the use of the indices of sea level dynamics (and in part the runoff of the
Neva River) as a predictor makes it possible, on the basis of expressions (1)-(6),
to obtain some idea concerning impending interannual changes in the underflow of
North Sea waters into the Baltic Sea through the Danish straits and with an ad-
vance time of about eight months to prepare a forecast of the anticipated changes
in the thickness of the layer of these waters, their mean salinity and the dis-
solved oxygen within their limits.
89
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
A noteworthy characteristic of the proposed procedure for the complex marine hy-
drological forecast is that this prediction is prepared using data from routine ob-
servations in the sea and river network of the State Committee on Hydrometeorology.
BIBLIOGRAPHY
1. Antonov, A. Ye., OKE.ANOLOGICHESKIYE OSNOVY RYBOPROMYSLOVYKH PROGNOZOV V YUZHNOY
CHASTI BALTIYSKOGO MORYA (Oceanological Principles of Fishing Forecasts in the
Southern Part of tne Baltic Sea), Kaliningrad, Izd-vo Gazety "Kaliningradskaya
Pravda," 1964.
2. Antonov, A. Ye., "Background Forecast of the Relative Yield of Some Commercial
Fish in the Baltic Sea," RYBNOYE KHOZYAYSTVO (Fishing Industry), No 11, 1962.
3. Antonov, A. Ye., "On the Problem of the Reasons for Variation of Salinity in
the Baltic Sea," RYBOKHOZYAYSTVENNYYE ISSLEDOVANIYA V BASSEYNE BALTIYSKOGO
MORYA (Fishery Investigations in the Bal.tic Sea Basin), No 3, 1967.
4. Antonov, A. Ye., "Prognostic Relationships Between Hydrometeorological Pheno-
mena in the Coastal Zone and in Open Regions of the Baltic Sea," TRUDY GOINa
(Transactic:-:s of the State Oceanographic Institute), No 115, 1972.
5. Antonov, A. Ye., "Procedures for the Super-Long-Range Prediction of Bottom Sal-
inity in the Baltic Sea," TRUDY GOINa, No 157, 1980.
i
90
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FOR OFFICIAL USE ONLY
UDC 551.464:621.039.8
STATUS OF THE STUDY OF THE ELEMENT-SALT COMPOSITION OF SEA WATERS USING NUCLEAR-
PHYSICAL METHODS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 80-85
[Article by Ye. M. Filippov, profesSOr, 14arine Hydrophysical Institute, manuscript
received 25 Jul 80]
[Text] Abstract: On the basis of computations and ex-
perimental investigations it is shawn that nuc-
lear-physical methods can be used in studying
the fundamental salt composition of sea waters
under natural conditions. The total salinity
of the waters can be determined on the basis
of the attenuation of soft gaimma radiation
emitted by sources with an initial energy of
100-300 KeV. The potassium in water is determin-
ed on the basis of natural radioactivity of its
isotope K40. The roentgenofluorescent method
can be used in separate determination of chlor-
ine, potassium, calcium and bromine and neutron
methods can be used in determining clilorine,
sodium and hromine. Automated systems can be cre-
ated on the basis of all these methods.
There is a change in the salt composition of sea water under the influence of nat-
ural factors and anthropogenic activity [1, 3, 4]. At the present time it is be-
ing studied for the most part under laboratory conditions using chemical and phys-
icochemical methods [5]. These methods have a high accuracy. However, they are ex-
tremely time consuming and do not make it possible to automate measurement pro-
cesses. This problem can be solved on the basis of nuclear-physical methods [6-11].
It was demonstrated in [7] that for determining the total salinity of sea water it
is possible to use a method based on the attenuation of soft gamma radiation in it
from sources with an energy of 100-300 KeV. We will evaluate the spectrum of scat-
tered gamma radiation on the basis of the diffusion-age equation [6], which for
convenience in the computations will be represented in the form .
(E) ~ (E)
[K = C] n~ ~ Qs j ~,X~~ 40(E)
(E) . IEI (r.9 (E) -
91
FOR OFF'ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-40850R040400034065-3
FOR OFFICIAL USF ONLY
Here Q is the yield of gamma radiation from the source, s is the effective area
of the detector, �(E) is detector efficiency, r is the distance between the cen-
ters of the s.ource and detector, 'GC(E) is the linear coefficient of attenuation
of gamma radiation as a result of Compton scattering, g(E) is the mean change in
the wavelength of gamma radiation as a result of scattering;

~ ('T d ?.T
e 3 J l..1]l '
Ay ~
'U
where Xy, = 511/E r is the wavelength of the gaimma quanta.
i.T .
().T) dkT :K ().T) E
,
This function takes into account the probability of absorption of gamna quanta on
hydrogen and oxygen atoms. The function 9P(E)S is similar to i{r, but takes into ac-
count the absorption of gamma radiation on atoms of mineral salts dissolved in the
water. The SO(E) value takes into account the dependence of gamma quanta on energy
and the S value determines the totai salinity of water in g/liter or pro mille
(�/oo).
Fig. l. Spectral distribution of scattered
~;amma radiation for source with initial
energy of 100 KeV. The figures on the
curves represent water salinity in pro
mille (g/liter).
e
7.
Fig. 2. Diagram of spectrometric sen-
sor for study of element-salt compo-
sition of sea water samples. 1) Si-Li
detector; 2) thermal shield; 3) detec-
tor housing; 4) capsule for annular
source; 5) source; 6) support for cas-
sette; 7) cassette with sample; 8)
water sample; 9) direction of primary
radiation; 10) direction of character-
is-tic radiation; 11) Be detector window;
12) polyethylene film on cassette bot-
tom.
92
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
~ J0 SO 70� Er KeB
z a ~y
~ ~

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00854R004400030065-3
FOR OFFICIAL USE ONLY
= The equilibrium spectrum of scattered gamma radiation, as. ;s well known [6], is
~ formed at distances not less, than 3-4 lengths, of the free path ~1= ~o-1, where Co
is the linear coefficient of attenuation of primary gamma r.adiation of the source.
The spectral distrihution of the scattered gamna radiation for a source with an in-
itial energy of 100 KeV in water with different salinity is shown in Fig. 1. This
figure shows that on the basis of regisLry of radiation in the energy range 30-70
KeV it is possible to judge the change in the salinity of sea water. With work
with a source of the type Eu155 and Gd153 With a yield of about 109 quanta/sec atid
a scintillation detector the salinity of sea water can be determined with an abso-
lute error 0.02-0.030/0o with a duration of the measurement of approximately 5 min-
utes.
pulsesAnn/c 2 62
sec
�"t0009 ~
'S.^90
F
I_

,
L
i-
Ga
J, 69
f l
1
I d EI
~
~
dE
pulses/sec
ti rMn/G Nunn/C
-JOC3lC 8r
)1,9
27:
1--~~
I;
fOG~
I
I i
sor
3 4 11 91 En:
V,
Fig. 3. Spectrum of characteristic emission of chlorine, potassium, calcium and bro-
mine.
It was demonstrated in [3] that in the study of water salinity in the coastal marine
zone the requirements on the error in measuring this parameter are reduced to
_ 0.10/00. It follows from this that in measurements of salinity with a given accuracy
by the considered method it was pos.sible either to decrease the activity of the
- source by a factor of 25 or to decrease the duration of the measurements to 15 sec,
all other conditions heing equal.
The change in the potassium content in sea water can be judged from its radioactive
" isotope K40. For example, on the basis of experimental investigations with a low-
background gamma detector of the "Limon" type supplied with an NaI(T1) crystal meas-
uring 150 x 100 mm with a duration of ineasurement of o~~e hour the K content can be de-
termined with an absolute error 0.0050/00 [9]. For these purposes it is also
93
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
possible to create apparatus with scintillation heta counters with a developed
surface.
For an express determination of other chemical elements of the fundamental salt com-
position it is possible to use roentgenofluorescent and neutron methods, the phys-
ical principJes of which were set forth in detail in [6].
The possibilities of using proportional counters for measuring the characteristic
emission of chemical elements of excited atoms of sea water were examined in [8].
There it was demonstrated that on the basis of the roentgenofluorescence (PPM) meth-
- od it was possible to create automated systems for the analysis of samples under
laboratory and natural conditions. In this way it is possible to determine chlor-
ine, calcium and bromine in sea water. However, the possibilities of PPM are con-
siderably expanded when making measurements with semiconductor silicon-lithium de-
tectors. The author of [2] describes such detectors with thermoelectric cooling
having a resolution for potassium and calcium of 243 and 245 eV respectively. On
the basis of such a detector it was possible to create a sensor for measuring the
element-salt composition of sea water directly aboard a ship. For the measure-
ments it is sufficient to use water samples with a mass of 2-3 g. The measurement
scheme is represented in Fig. 2. Computations for this geometry are made in ac-
cordance with [8]. We will assume that for the excitation of chlorine, potassium
and calcium atoms use is made of a titanium-tr.itium source (T = 12.3 g, E y= 4.5
i;eV) with a yield of 7.5�106 quanta/sec, and for the excitation of bromine moly-
tadenum-93 (T = 2300 years, E y= 17 KeV) with a yield 1.5�10$ quanta/sec. The area
of the detector, in accordance with Fig. 2, was assumed to be equal to 3.8 cm231
whereas the area of the entrance window was 2.54 cm2.
On the basis of the considered PPM method, as demonstrated in [8], it is also pos-
si.ble to create submergible sensors. Such sensors can be assembled on the basis of
semiconductor detectors used in field geophysical practice.
Computations of the spectral distribution of the characteristic emission of chlor-
ine, potassium and calcium for ocean water (S = 350/00) are illustrated in Fig. 3.
The counting rate at the photopeak of each element and the backgrotmd is indicated
in Table l. The absolute (,6 C) and relative (S C) errors were computed with allow-
ance for the background emission in accordance with [6]. The table shows that the
highest accuracies in the measurements are attained when determining chlorine and
calcium. When making measurements with a submergible sensor, as a result of the
absorption of emission in an additional beryllium window, sealing in the sensor,
the measurement errors are somewnat further increased and measurement accuracy is
reduced. With an increase in source activity by an order of magnitude the measure-
nient error can be reduced by a factor of 3.
The discussed PPM sensors can be used in a study of changes in the element-salt
composition of sea water at shallow depths since otherwise they would be crushed
by the hydrostatic pressure of a great thickness of water. Neutron methods can be
employed in a study of sea water in an extremely wide range of depths: the neutron
gamma method, based on the registry of gamma emission of the radiation capture of
neutYons (NGM) and the neutron activation method (NAM).
94
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
'i'able 1
Results of PPM Computations and Anticipated Absolute Measurement Errors
I
t. sec
t sec
Element
~
N
v
~
,
ulse
*
I00 I
400
p,,ls
s*
100 I 400
~
sec
(
se
Clilorine
19 j
4608
1808
0,037
0,02
1475
SiS ~ 0,066 0,033
Potassium
0,38
21.~
2136
0,012
0,006
122
1209 i 0,016 0,008
Calcium
0,40
773
2154
0,0037
0,002
:.08
1415I 0,0045 0,0022
Bromine
0,063
123
800
0.0014
0,0007
121
774 0,0022 ~ 0,0011
*Nbackground/pulses
Table 2
(
;
Results of NAM Computations and Anticipated Measurement Errors
I
Element
1
r
E ~ ;V Nback
I t1eV I pulses I pulses ~ pulse
~ i

%
Ca..A C Iroo
;hlorine
2.15
I09500
64700
350
1,6
71400
36400
850
0,32
9,0�0,03
>odium
3,85
177
52
2,76
,
47000
37780
1
38
74630
36100
1000
0,37
5,25�0,019
3romine
0,62
2252
1000
850
5,2
0,0325�0.0017
4agnesiinn
1,013
235
1000
20
0,675~-0,136
;alcium
4�68
8.32
3,36
54.6
0,2�0,I09
4,05
21.2
3,1
39,1
The NGM, as demonstrated on the basis of theoretical [9, 10] and experimental in-
vestigations [11], can be used in determining chlorine. The author has demonstrated
that it can be determined using a method with a californium saurce with a yield
,..1�108 neutrons/sec and a scintillation counter of the "Limon" type. With a meas-
urement duration of 2 sec and a distance between the source and detector of 20 cm
the chlorine in Black Sea water (C = 9.50/00) can be determined with an absolute
error of 0.090/00 or a relative error of 0.940/00. By increasing the duration of
the measurements to 8 sec the measurement errors can be cut in half.
The neutron activation method [9, 10] can be used in determining chlorine, sodium,
- bromine and other elements. In these studies expressions have baen derived for cal-
culating the counting rates N at the photopeaks for individual lines of activating
chemical elements.
95
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
pulses/channel .
� BMOIIrQNQp ~ ~
r ti
BDOO~ a ~ ~
J71I1 4 {-,i7ti+1
~
H
�
PV1SeS
= una
v .iw Ct �
a
~ ey w`
0
61
8000 No 215
~ 1,3d ~ 6
6000 N ~ '0
�000 A
8r N.
P000 a 62 i
. 
PO ~ i~ 4, 05 4,168
0`
N a 60 J,1 N n
Z 7s b~ a, es
' t0 ,
h ~Q
, m GQ
i 4,OS 4i 6B
0
~r 0,8 1,2 1,B 2,0 fM0 40 44 J,B 4,2 44 48
MeV
Fig. 4. Gamma spectrum of Black Sea water activated by thermal neutrons. a) for de-
tector with ideal resolution; b) for scintillatiQn detector of the "Limon" type.
Using these algorithms, the author, in collaboration with A. Kh. Degtyarev, made
computations of the spectra of gamma radiation at the output of a"Limon" detec-
tor. The computations were made by the Monte Carlo method for an idealized case:
it was assumed that all the gamma quanta hit at the center of the crystal. Due to
Compton scattering and the effect of pair formation the emission reckoned at the
photopeaks decreased from N to N1 (Table 2). The resulting spectral distribution of
emission is illustrated in Fig. 4. The conversion from curve a to curve b was ac-
complished using a Gaussian for the purpose of obtaining an "experimental" spectrum.
For the left part of the spectrum the interval along the x-axis was selected equal
to 20 KeV, and for the right part 100 KeV. The results of computations of the
measurement errors are given in Table 2. Comput.ations were made for the case of
activation of water under natural conditions over a period of 30 minutes and meas-
uremenCs of induced activity were also made for 30 minutes. The table shows that
when using the NAM method the highest accuracies are attained for chlorine and so-
dium and magnesium and calcium are determined less accurately.
4le note on the basis of these materials that a combination of neutron methods is
- possible. The entire outfit for the study of sea water can be carried aboard ships,
small boats, amphibian craft or rafts. A paraffin or water container with a mass of
1 ton is sufficient for storage of the neutron source. During the measurements the
sensor together with the container can be lowered over the side of the ship. When
the source is present in the container it is possible to determine the potassium
content in sea water and evaluate the gamma background. Chlorine in sea water can
be determined in the course of its irradiation in the course of a f ew seconds. Then
after irradiation of the water and removal of the source the induced activity of
the forming isotopes is measured in the container. In order for the current not to
distort the measurement results tlie irradiated volume of water must be recorded
96
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
prior to irradiation and measurement. For this purpose it is necessary to fabri-
cate two hemispherical containers fixing the subsurface water volume used in the
measurements.
We note in conclusion that all the operations for studying the salt-element compo-
sition of the water can be automated. On the basis of specially prepared programs
the data from specti�ometric measurements can be processed on shipboard electronic
computers, at their output giving the salinity values and the concentrations of
determined elements of interest to the researcher.
BIBLIQGRAPHY
1. Baydin, S. S., "Redistribution of River Runoff Among Sea Basins and its F.ole
in the Natural Complex of Seas and River Mouths," TRUDY GOIN (Transactions of
the State Oceanographic Institute), No 143, 1979.
2. Beda, A. G., "Germanium and Silicon-Lithium Spectrometers With High Resolution
for Practical Purposes," ATOMI3AYA TEKHNIKA ZA RUBEZHOM (Atomic Engineering
Abroad), No 9, 1979.
3. Ivanov, G. S. and Ovsyannikov, A. N., "Variability of Water SaYinity in the
Coastal Zone of the Sea," METEOROLOGIYA I GIDROLOGIYA (Meteorology and Hydrol-
ogy), No 9, 1979.
4. Kosarev, A. N., "Problems of the Southern Seas of the USSR," ZEMLYA I VSELENN-
AYA (Earth and Universe), No 3, 1975.
5. RUKOVODSTVO PO METODAM KHIMICHESKOGO ANALIZA MORSKIKH VOD (Manual on Methods
for Chemical Analysis of Sea ldater), edited by S. G. Oradovskiy, Leningrad,
Gidrometeoizdat, 1977.
6. Filippov, Ye. M., YADERNAYA GEOFIZIKA (Nuclear Geophysics), Volwnes 1, 2, Novo-
sibirsk, Nauka, 1973.
7. Filippov, Ye. M., "Possibilities for Determining the Salinity and Density of
Sea Water From the Attenuation of Gamma Radiation," MORSKIYE GIDROFIZICHESKIYE
ISSLEDOVANIYA (Marine Hydrophysical Investigations), Sevastopol', No 1(80),
1978.
8. Filippov, Ye. M., Possibilities of the Roentgenofluorescent Method for Deter-
mining the Salt Composition of Sea Water," Article deposited at the All-Union
Institute of Scientific and Technical Information, 12 November 1979, No 3842-
79DEP.
9. Filippov, Ye. M., "Use of Californium Neutron Sources for Determining the Main
Salt-Element Composition of Sea Water Under Natural Conditions," ATOMNAYA ENER-
GIYA (Atomi.c Energy), Vol 47, No 4, 1979.
97
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFF[CIAL USE ONLY
10. Filippov, Ye. M., "Prospects for the Use of the Neutron Activation Method
in Natural Oceanological Investigations," Article deposited at the All-Union
Institute of Scientific and Technical Information, 13 February 1980, No 853-
80DEP.
11. Wiggins, P. F. and Athow, K. Y., "Salt Water Chlorine Ion Determination by
Neutron Capture Gamma Rays," NAVAL ENGINEERS J., No 5, 1971.
98
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 556.(166.4+047+013):551.501.81
USE OF RADAR DATA IN A HYDRODYNAMIC MODEL OF RAINWAtER RUNOFF WITH DISTRIBUTED
PARAMETERS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 86-92
[Article by V. A. Rumyantsev, candidate of technical sciences, and S. A. Kondrat'-
yev, State Hydrological Institute, manuscript received 5 Sep 801
[Text] Abstract: In the example of the rainwater high-
a water hydrograph for the Polomet' River the
authors tested a model of formation of rainwater
_ runoff with distributed parameters which makes
it possible to take into account the topography
of the watershed, the spatial variability of the
hydrophysical properties of soils in the basin
and evaporation from the watershed surface. In
the computations use was made o� information on
the precipitation fields determined both by the
radar method and using data from the surface
precipitation-gaging network. The results of
the computations made it possible to detect the
inadequacy of these fields, which is attribut-
able to inadequately perfect radar set calibra-
tion.
During recent years more and rw re studies have appeared in the hydrological liter-
ature which are devoted to models of rain-induced runoff. These are based on the
equations of motion of a fluid known from hydrodynamics [7]. However, these models
have not yet been used in computations and predictions af rain-induced runoff be-
cause one of their basic advantages, the possibility of taking into.account the
spatial variability of the input parameters, can be realized fully only whsn
there is thorough information on the hpdrophysical properties of soil and ground
and information on the field of precipitation. The latter can be obtained using
the radar method *or measuring precipitation. Today individual studies are known
in which the materials from radar measurements are used primarily in determining
the mean precipitation quantities for the basin and the conversion to the runoff hy-
drograph is accomplished using graphic dependences [3] wiCh use of models of run-
off with concentrated parameters [4] and linear conceptual runoff models [4, 10,
1111. In contrast to those mentioned, in this study an attempt is made in the
example of the Polomet' River basin (Dvorets village) with a watershed area of
99
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
432 km2 to demonstrate the possibility of using data on the precipitation field,
obtained using the radar measurement method, in a hydrodynamic model of forma-
tion of rain-induced runoff.
An automated complex for the radar measuremer.t of precipitation was introduced in
tne territory of the Valdayskaya Scientific Research Hydrological Laboratory imeni
V. A. Uryvayev, situated at a distance of 10 km from the Polomet' River watershed,
in July 1973. It includes a meteorological radar (MRL-2), the special "Osadki"
apparatus, developed by the Central Aerological Observatory [2] and the ASVT
M-6000 controlling electronic computer. The automated,radar measuring complex
makes it possible to measure the instantaneous intensity of atmospheric precipita-
tion in a radius 100 km around Valday each 5 minutes and represent it at a real
time scale in the form of a punched card for the mean intensity of precipitation
for elementary grid squares with an area of 10 x 10 km and simultaneously in the
form of punched tapes with information for grid squares with an area of 3.3 x 3.3
km for the entire territory scanned by the radar. In addition, the hourly layer
of precipitation and the sum of preeipitation accwnulated in 6 hours are printed
out and punche3 out. The choice of the Polomet' River as the object of investiga-
tions is also attributable to the good study of the basin, which makes it possible
to stipulate the spatial distribution of watershed characteristics.
i ~
i
i
i
�
 ~
r ~
~

~
. i
.
I
~ =1 ---3
Fig. 1. Schematic representation of watershed and selected runoff directions. 1)
actual channel, 2) schematic channel, 3) boundary of watershed, 4) sectors of
watershed in which heavy clayey loams predominate, 5) light clayey loams and
sandy loams, 6) swampy deposits.
In the Polomet' River basin surface runoff is observed extremely rarely; here there
is a predominance of so-called "contact" [1] runoff. In this connection for comput-
ing the inflow into the channel network use was made of the following equation:
(m - w~c) oi + a i ash> = R- E (1>
ioo
FOR OF'FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
M-1
S ~6

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFriCIAL USE ONI.Y
with the boundary conditions h(0, t) = h(x, 0) = 0. Here h is the intensity of
"contact" runoff, m is ground porosity, Wmmc is minimum moisture capacity, k is
a coefficient taking into account the hydraulic resistance of "contact" runoff,
R and E are the intensities of precipitation and evaporation, x and t are the
space and time coordinates.
In computing the transformation of the high-water wave in the channel we usEd the
equation for a kinematic wave.
-
- dA d S7p qas
ct + dy N 1 = 9 (2)
[p = ch(annel)] ~
with the conditions A(0, t) = A(t), A(y,0) = A(y). Here A is the cross-sectional
area, Sch is channel slope, N is the Manning roughness coefficient, P is the wett-
ing parameter, q is lateral inflow into the channel, y is the space coordinate.
The channels of the Polomet' River and its principal tributary, the Sosninka River,
- were approximated by linear segmemts (Fig. 1). In order for such an approximation
not to exert an influence on the final result it is necessary to change the para-
meters of the schematic channel (slope and Manning roughness coefficient) in a
3efinite way in comparison with the actual values. The relationship of the para-
meters can be obtained from the condition of equality of the travel time for the
initial and schematj.c channels. For determining the.travel time trun We used a
_ formula from [7] ^ 3
~s
trun � ~ N t , (3)
11 where L is channel length. ~ SP 43
By stipulating the 3rop in elevations La by the parameters of the initial channel
N', Sch , L' and the length of the schematic channel by L, ir is possible to deter-
mine the slope S= A/L and then from the condition of equ3lity of the travel time
for both cases, taking inte account that q= q'L'/L, it is possible to obtain the
relationship between the roughness coefficients
13
N = N' (LL ) ti ' (4)
For the approx{mation adopted in this problem we obtained Sch = 1.5%, N= 4N'
The computations were made for the high water pasaing during the period 5-20 August
1979, caused by rains falling during the period 5-7 August.
In formulating the initial conditions it was assumed that the low pre-high-water
river runoff occurs at the expense of the gr.ound water supply, virtually not chang-
ing during the time of high watero The intensity of water supply ensuring a stip-
ulated pre-high-water discharge at the lowest-lying station was stipulated as the
initial condition for channel flow. This made it possible to obtain stable values
of the parameters characterizing the channel flow prior to the onset of high water
and after its ending.
The slope in the diroction of runoff was determined for computing the inflow into
the channel network on the basis of an analysis of the topography for each of 53
grid squares with an area of 3.3 x 3.3 km covering the Polomet' River watershed.
101
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-40850R040400034065-3
FOR OFFICIAL USE ONLY
Then a schematic mapping of soil types was carried out (Fig. 1). A determination
of the soil type for each grid square was made in accordance with the maximum
percentage of the area occupied by different soils. It was found that in 24 grid
squares there was a predominance of heavy clayey loams, in 27 light clayey
loaras and sandy loams, and in two swampy deposits. We note in passing that
there was not a single grid square in which sand predominated (in the watershed
as a whole it occupied about 4% of the area). The grid squares were similarly
classified as "field" and "wooded."
Depending on the type of soils, on the basis of observations made by N. I. Kapo-
tova and A. A. Kapotov, specialists at the All-Union Scientific Reaearch Hydro-
iogical Laboratory, and or: the basis of data in the literature, for each grid
square we established its porosity and minimum moisture capacity values: for
heavy clayey loams m= 0.33 and Wmmc = 0.27; for light clayey loams and sandy
loams m= 0.36 and Wmmc = 0.25; for swampy deposits m= 0.96 and Wmmc = 0.83. The
evaporation intensities were stipulated on the basis of the mean monthly evapora-
tion values for August as given by S. F. Fedorov [8]: for wooded sectors it was
assumed that E= 0.11 mm/hour, for field sectorg E= 0.096 mm/hour. The intradi-
urnal change in evaporation intensity was not taken into account. The initial
moistening Wp was stipulated on the basis of the moisture content in Tayezhnyy
ravine (an analog for wooded sectors in the Polomet' River basin) and Usad'yevsk-
iy ravine (an analogue for field sectors): W0 ood 0.709 (3 August) and Wp field -
0.25 (30 July). The initial moisture reserves ~or swampy sectors were determined
hy a method proposed by 0. I. Krestovskiy,151 on the basis of the pre-high-water
runoff of tlie Polomet' River. Accordtng to an estimate which was made, the losses
prior to the onset of runoff from swamps should be 75 mm. It was assumed that the
runoff begins only after the moisture content attains the minimum moisture cap-
acity values in ttie half-meter soil layer.
As pointed out by N. F. Befani [l], the rate of runoff of "contact" water is many
times greater than the rate of ground water movement. In this connection it can
be said that the k coefficient in formula (1), which when it was obtained was in-
rerpreted as the filtratton coefficient, lost its initial sense here and describes
the hydraulic resistance not only af intrasoil runoff, but also of the fine network
of rilla. Therefore, for different sectors it was stipulated in the form
ki = aik0,
here ko is a correction factor taking into account the runoff of water in the fine
network of rills, which was assumed to be identical for different soil types.
The differences in the filtration capacity of the soils were taken into account by
means of the coefficients ai, which were stipulated from the condition of mainten-
ance of thz relationships between the filtration coefficients. On the basis of
data in the literature we adopted the following values of the ai coefficients: for
heavy clayey loams ai = 0.1, for light clayey loams and sandy loams ai = 1, and for
swampy deposits ai = 5.
Thus, in the computations we took into account:
a) the spatial-temporal variability of the intensity of precipitation;
b) the spatial change in watershed surface slopes;
c) the spatial change in the hydrophysical characteristics of soils: porosity, min-
imum moisture capacity, filtration coefficient;
102
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
_ FOR OFFICIAL USE ONLY
d) spatial change in the intensity of evaporation in dependence on the nature of
vegetation;
e) epatial change in initial moieture supplies.
The channel roughness coefficient was taken from the tabulated data in [9] and was
- equal to 0.025. The k.o coefficient was determined inversely from the condition of
best agreement of the maxima of the computed and observed hydrographs [7]. A numer-
ical realization of equations (1) and (2) was realized on the basis of an explicit
finite-difference scheme for the approximation of differential operators, based on
an expansion of the solution for the continuity equation
ah +aQ=q
into a Taylor series [7]. If the h values are known at the points of intersection
of the i-th time layer of a selected uniform grid, the values at the points of in-
tersection of tfie (i+l)-st layer are computed using_the formula
hl+l = k' + Q%+i QI-I + QI i 0 t+ [ D1+t + Dt Q%+i - Qi _
J i ( 2 ~x ) 2Ax ( A x
qr I l D~_ Qr Q! t 1 (5)
l+l + ql D1 / t !_l Qi + 4r-I -F-
2 2ax ~z 2
o n
+
A t ] 2 '
where D= c?Q/ G h; L1 x, At are the spatial and temporal integration intervals; j
is an index for the grid poiuts along the spatial axis.
m3/ sec
E
~
r
,

. z
Augttst 1979 ,4eocm1979t �
� 7 s
Fig. 2. Observed runoff hydrograph (1) for Polomet' River (Dvorets village) from
rains of 5-7 August 1979; hydrographs computed using radar informa.tion on the pre-
cipitation fields (2) and data from surface precipitation-gaging network (3).
The runoff hydrograph, computed using radar information on the precipitation
fields, is shown in Fig. 2. The value of the ko parameter, obtained invergely, was
3.65 m/min. The distribution of the precipitation layer over the area of the
watershed, according to radar data, is shown in Fig. 3a. Comparing the constructed
103
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FQR (1FFIrIe r UcE ONLY
hydrograph with the observed hydrograph, it is easy to see that during the drop-
off the computed water discharge values somewhat exceed the observed values.
The reason for this is an exaggeration of the radar data for the inteneity of
precipitation in the right part of the watershed, as is indicated, in our opin-
ion, by a munotonic increase in this direction of the intensity of precipitation
at any moment in time and the layer of precipitation during the entire period of
falling of rain (Fig. 3a). As a confirma.tion of this we computed the same high
water witli tfie use of data from the sur.face precipitation gage network. The pres-
ence of 44 pluviographs within the Polomet' River watershed makes possible a suf-
ficiently good descri.ption of the spatial variability of precipitation.
6215
SS, 6
S4, ff
S0, 7
Sf, J i
fi; f
37,1
3J, 6
J8,7'
33,3
J519
Y2,7
J1;6
f1,0
41,5
48,1
91, 7
f.~ 't
#6, 9
S2, 4
~~-1.9
47,3
SP, B
54, 7 I
48,0
51,3
51,9
62,7
S*, 6
SJ.O
57,2
6,~8
Q)
41, 7 50,1
4d, 9 61, 4 94,
57,6 77,8 119,
$43 90,5121,
71,1 9Q 5115,
16, 2 97, 9 120,
87,9
6J,5
60,3
60,1
6)
16,7
64,7
$9.1
59,0
60,9
64,4
74, 6
60,1
61, 0
SB,B
60,2
66,J
17,1
9>, 8
747
645
61,6
SB,4
60,0
66,5
16,2
89,4
14,7
68,1
62,5
60,0
61,2
67,8
7'r,6
85,6
69,0
65,6
63,6
61,7
62,5
70,0
72,2
80,7
12,0
69,8
69.1
72,3
68,8
68,0
70,Z
e1, 3
60 3
76, B
B 1, ~
Fig. 3. Spatial distribution of precipitation layer during period 5-7 August 1979
according to radar data (a) and according to data from surface precipitation gage
network (b).
The hourly precipitation layers obtained on the basis of pluviographic data were
interpolated at the points of intersection in a uniform grid with an interval of
_ 1.1 km with subsequent averaging for each of the earlier defined grid squares.
The interpolation was based on an approximation of the precipitation field by a
second-degree polynomial of the coordinates and was accomplished in accordance
with a program prepared at the Computation Center of the State Hydralogical Inst-
itute by 0. M. Tolshina and N. A. Livanova. The distribution of the layer of pre-
cipitation according to surface data is shown in Fig. 3b. The hydrograph, computed
using surface data with the former value of the kp parameter, is shown in Fig. 2.
Some increase in the maximum of the computed hydrograph is attributable to the
fact that in the left and central parts of the watershed the data from the sur-
face precipitation gage network are greater than the radar data, as can be seen in
Fig. 3. The good correspondence of the forms of the computed and observed hydro-
graphs, and the coincidence of the time of onset of the maximum for both hydro-
graphs,indicates the adequacy of the model and the correctness of a priori stipul-
ation of the numerical values of the parameters. An improvement in the results
of computations for the dropoff of the hydrograph when using data from the surface
104
FOR OFFICIAL USE O1VLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R440400030065-3
FOR OFFICIAL USE ONLY
precipitation gage network confirms the assumption made that there is an exagger-
ation of the radar data, and accordingly, is evidence of a definite noncorrespond-
ence of the precipitation fields obtained using data from the surface network and
by the radar'method. The reason for this is that the radar was calibrated with
calibration coefficients constant over the area and using precipitation layers
averaged for tfie watershed [6J, which in both cases virtually coincide. In this
connection further work must be done for improving the method for radar measure-
ment of precipitation for the purpose of obtaining calibration coefficients vary-
ing over the area of the watershed in accordance with its local orography. In the
event of a successful solution of this problem 3.t can be hoped that the accuracy
of radar data on the precipitation fields will not be lower than the accuracy of
the data from the surface network and due to the routineness in collection of
radar information there will be a substantial broadening of the region of applic-
ation of tfie described hydrodynamic model with distributed parameters.
At the present time work is being carried out for refining the physical picture of
formation of rainwater runoff in the basin of the Polomet' River. Although work on
the model was directed for the most part to solution of inethodological problems,
it is proposed tfiat it be brought to the stage of introduction into practical, rou-
tine short-ranbe forecasts for ensuring support for an optimimm regime for the op-
eration of inelioiation systems in the lower part of the Polomet' River basino In
addition, the model may prove useful in simulation work for evaluating the influ-
ence of different kinds of agricultural and forestry melioration measures on the
formation of rairiwater runoff in the basin. When carrying out numerical experi-
ments an allowance far such measures can be made by stipulating the soil charac-
teristica of the watershed or its topography modified for the reasons indicated
above. In solving the formulated problems, in addition to improving radar calibra-
tion, it is also necessary to make an evaluation and choice of the optimum interval
with stipulation of the hydrophysical characteristics of soils and topography in
the watershed.
BIBLIOGRAPIiY
1. Befani, N. F., PROGNOZIROVANIYE DOZHDEVYKH PAVODKOV NA OSNOVE TERRITORIAL'NO
OBSHCHIKH ZAVISIMOSTEY (Prediction of Rain-Induced High Waters on the Basis
of Territorially General Dependences), Leningrad, Gidrometeoizdat, 1977.
2. Beryulev, G. P., Yevpryakov, V. A., Kostarev, V. V., Mel'nichuk, Yu. V. and
Chernikov, A. A., "Apparatus for Measuring the Quantity of Liguid Precipita-
tion in an Area Using a Single-Wavelength Meteorological Qadar," TRUDY TsAO
(Transactions of the Central Aerological Observatory), No 121, 1975.
3. Dzhordzhevich, N. T. and Miokovich, M., "Methods for Predicting Rain-Induced
High Waters With a Special Review of Experiments for Use of a Radar in Oper-
ational Forecasting," SBORNIK DOKLADOV X KONFERENTSII PRIDUNAYSKIKH STRAN
PO GIDROLOGICHESKIM PROGNOZAM (Collection of Reports at the Tenth Conference
of Danubian Countries on Hydrological Forecasts), Vienna, 1979.
4. Koren', V. I., "Oti Use of Radar Measurements of Precipitation for Predicting
Rain-Induced High Waters," TRUDY GIDROMETTSENTRA SSSR (Transactions of the
USSR Hydrometeorological Center), No 191, 1977.
5. Krestovskiy, 0. I., "Investigation of Runoff and the Water Balance of Water-
sheds," TRUDY GGI (Transactions of the State Hydrological Institute), No 176,
1464. '
105
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFF[CIAL USE ONLY
yev, A. P., "Cornputation of the Intensity of Pre-
6. Livanova, N. A. and Uryva7
cipitation Using Data From Radar and Pl:rriographic Measurements," TRUDY GGI,
No 258, 1979.
7. Rumyantsev, V. A. and Kondrat'yev, S. A., "Mathematical Modeling in Hydrology.
Kinematic-Wave Model of Slope Runoff," GIDROMETEOROLOGIYA. GIDROLOGIYA SUSHI.
- OBZORNAYA INFORMATSIYA (Hydrometeorology, Hydrology of the Land, Review In-
formation), Obninsk, No 1, 1975.
8. Fedorov, S. F., "Results of Investigation of the Hydrological Role of the For-
est," TRUDY GGI, No 176, 1969.
9. Chugayev, R. R., GIDRAVLIKA (Hydraulics), Leningrad, Energiya, 1975.
10. Anderl, B., Attmannspacher, W. and Schultz, G. A., "Accuracy of Reservoir
Inflow Forecasts Based on Radar Rainfall Measurements," WATER RESOUR. RES.,
Vol 12, No 2, 1976.
11. Schultz, G. A. and Klatt, P., "Use of Data From Remote Sensing Sources for Hy-
drological Forecasting," PROC. OF THE OXFORD SYMPOSIUM. IAHS-AISH PUBL. No
129, 1980.
106
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
UDC 556.164
EXPERIMENTAL SUBSTANTIATION OF COMPUTATIONS OF THE RATE OF WATER FLOW ALONG THE
CULTIVATED SURFACE OF SLOPES
~
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 93-96
(Article hy G. V. Kostsov, candidate of technical sciences, Voronezh Agrj.cultural
- Institute, manuscript received 9 Jun 801
[Text] Abstract: On the basis of experimental in-
vestigations of the influence of runoff and
slope on the rate of water flow along the
cultivated surface of different types of
soil and ground the author has derived a
formula for computing the rate of water
runoff in individual sectors of plowed slopes.
Investigations of the nature of change in the rate of water runoff along a loosen-
ed soil s urface as a function of runoff discharge volume and the degree of slope
were carried out experimentally in the laboratory at the Voronezh Agricultural
Institute.
A special hydraulic flume was used in making the investigations. The working length
of the flume was 2.0 m and its width was 0.4 m. At the head of this flume there was
- a device for inlet of regulated water volumes in the form of a water layer uniform-
ly distributed in the width of the flume. Ground or soil was loaded in the trough.
A screw-type hoisting mechanism made it possible to impart different sloDes to the
flume. A detailed description of the design of this apparatus was given in an
earlier published paper [1].
The water volumes fed into the upper part of the flume were varied in the range
from 0.05 to 0.26 liter/second and the s urface slope was varied from 0.01 to 0.16.
The number of different variations of the water quantity fed into the flume was 5
and the number of slope variations was from 6 to 8. Each experiment was repeated
twice. The duration of each experiment (with a fixed discharge and slope) was 10
minutes. After its completion the ground or soil s urface was restored. Theyrunoff
rate was determined in the interval from the third to the fifth minutE from the
beginning of each experiment by the introduction of a special dye into the water.
The runoff rate was computed using the formula
107
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034465-3
FOR OFFICIAL USE ONLY
t
y a L
where L is the length of the flume in meters; t is the time required for the dyed
fluid to pass from the flume, registered hy a stopwatch, in seconds.
The design of the hydraulic flume and the method adopted for carrying out the exper-
- iments ensured the formation of rates of water runoff which in absolute value were
as close as possible to the local rates observed on individual slope sectors.
The ground or soil loaded into the flume was first saturated for a rather long time
(not less than a day) with water to the p4int of total moisture capacity. The lumpy
surface of the soil present in the flume had a roughness extremely close to tha
roughness of a plowed field. Thus, when conducting the experiments there was a sort
of simulation of the conditions for the runoff of ineltwater along a thawing surface
of a plowed slope.
In the course of the formulated experiment we studied the nature of the change in
the rate of water runoff along the loosened surface of gray forest and chernozem
soils, as well as sandy ground. In addition, allowance was made for the sheet ero-
sion of these types of soil and ground. The method for making such an allowance was
set forth in studies published earlier jl, 2]. The corresponding values of the total
sheet erosion obtained with identical discharges and slopes are given in the table.
In carrying out the experiments it was noted that with the runoff of water along tlie
surface of the tested samples there is formation of a network of microjets. Their
number and configuration change with time, which causes corresponding variations of
different hydrodynanic characteristics. For example, the depths of the microjets,
the hydraulic radii and the slopes of the forming channels change somewhat. Such a
phenomenon is especially characteristic for easily eroded types of ground and soil.
The rates of water runoff observed in the course of the experiments were varied in
a rather wide range. The greatest rates were observed in the case of easily eroded
types of soil and ground. The minimum rates were observed with soil and ground types
which are not easily eroded. The limits of variation of these rates are indicated
in the table.
In virtually all cases (except for individual cases of the runoff of water in small
volumes and with minimum slopes) there was a turbulent regime of water movement
since the Reynolds number was more than 300.
On the basis of the results of a preliminary analysis, and also taking into account
already available investigations of this question, the considered dependence was
approximated in the following way:
v = a QmIn,
(2)
where v is the rate of water runoff; a is a parameter characterizing the roughness
of the underlying surface; Q is the runoff per unit width of slope; I is surface
slope; m and n are the corresponding exponents.
108
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
The values of the parameters of this dependence were computed by the mathematical
processing of data from 675 experiments carried out with a YeS-1020 electronic
computer.
The established values of the exponents m and n were varied in the following range:
m-- from 0.28 to 0.52; n-- f rom 0.36 to 0.47. Accordingly, taking into account
the relatively small ranges of their variation, henceforth with some approximation
we have used their mean arithmetical values, which in both cases were 0.4.
After such a transformation and the introduction of the correction factor k for-
mula (2) was reduced to the following form:
v ~ kaQo.4 10.4.
(3)
The introduction of the factor k is attributable to the circumstance that during
the period of runoff of water along the plowed slope the resistance of the under-
lying surface somewhat decreases with time and the rate of runoff increases.be-
caus_ there is a gradual straightening of the runoff channels and a smoothing of
roughness elements.
The value of the k factor varies in dependence on the duration of runoff and in this
stage in the study can be estimated only approximately. Thus, on the basis of the
materials from these investigations and generalizing already available data on
this question [3] it can be assumed equal to 1.10-1.15.
- In formula (3) the dimensionality of the rate of water runoff is expressed in meters
per second, runoff discharge in liters per second per 1.0 m of slope width and slope
in promille. The corresponding values of the a parameter are given in the table.
kg
Fig. 1. Graph of the dependence of the a parameter on the degree to which different
types of soils and ground are subject to sheet erosion.
A subsequent analysis indicated that the values of the a parameter are dependent on
the degree to which differen:. *_;pes of soils and ground are subject to sheet ero-
sion, that is, on their erosional properties. The graph characterizing this depend-
ence is represented by a straight line (see figure). The graph v2ry clearly indi-
cates the increase in the a parameter (decrease in roughness) with an increase in
total sheet erosion (~,'g). This is attributable to the circumstance that during
water runoff along a loosened surface of easily eroded ground and soil the water
velocity will be greater than when the water flows over soils which are not easily
eroded. In the latter case an influence will be exerted by the greater sinuosity
109
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
~
a~
0
H
~
co
~
~
1+
0
)
c+1 n O~O o0 u1 00 t~ n t.
c1
T7
y
o0M 0 C1 OOm ONQ~

,O
O O ri r-1 O r-I e-I O
N
~
p
.
. � . . . �
.
~
D
D
O
O O O O O O O
O
O
co
.r{
`o
�
a
r-
i
a
+
0
w
co

a
44
~
o
o
v
u
>
a~
~
r-I
ro

04 ~7
~
~
c
d G
J .C 4 r
-I
~
~
F3 0
0 O Tf 'b 'd ~-I ri ~ H
W 41 a) G1 N cd Cd 01
~
daa4 :j
c'd co c
H
a~a
�
aaHpov
~
cn
110
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFIC[AL USE ONLY
and roughness of the microchannels forming during water runoff along the loosened
surface of these soils.
The accuracy of the conputations of the rate of water runoff made using formula
(3) can be evaluated on the hasis of the mean relative deviations of the computed
velocity values (vcomp) from the observed velocities (vobs), the values of which
are given in the table.
In the above-mentioned form formula (3) can be used in computing the rate of runoff
of ineltwater on individual sectors of plowed slopes. The following values of the a
parameter can be used: for sandy ground 0.130; for sandy loam 0.050-0.065;
for clayey loam and clayey soils 0.035-0.060. In the latter case the highest
values of the a parameter must be used for easily eroded soils and the minimum
values must be used for soils more resistant to erosion.
BIBLIOGRAPHY
1. Kostsov, G. V., "Problems and Methods for Experimental Study of the Processes
of Sheet Erosion of Different Types of Soil and Ground," REGULIROVANIYE STOKA,
SEL'SKOKHOZYAYSTVENNAYA MELIORATSIYA I ZASHCHITA ZEMEL' OT VODNOY EROZII V
TSENTRAL'NO-CHERNOZEMNOY ZONE. NAUCHNYYE TRUDY VORONEZHSKOGO SKhI (Regulation
of Runoff, Agricultural Melioration and Protection of Soils Against Water Ero-
sion in the Central Chernozem Zone. Scientific Transactions of the Voronezh
Agricultural Institute), Vol 69, Voronezh, 1975.
2. Kostsov, G. V., "Some Results af Experimental Study of the Patterns of Sheet
Erosion from Slopes," ZEMEL'NYYE FONDY TSENTRAL'NO-CHERNOZEMNOY ZONY I VOPROSY
IKH RATSIONAL'NOGO ISPOL'ZOVANIYA. NAUCHNYYE TRUDY VORONEZHSKOGO SKhI (Land
~ Resources of the Central Chernozem Zone and Problems in Their Rational Use.
Scientific Transactions of the Voronezh Agricultural Institute), Vol 80, Voro-
nezh, 1976.
3. Shvebs, G. I., FORMIROVANIYE VODNOY EROZII, STOKA NANOSOV I IKH OTSENKA (Forma-
tion of Water Erosion, Runoff of Sediments and Their Evaluation), Leningrad,
Gidrometeoizdat, 1974.
111
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 551.578.7:633.51
DETERMINATION OF DAMAGE TO COTTON PLANTS IN DIFFERENT DEVELOPMENT STAGES RESULTING
FROM HAILFALLS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 97-102
[Article by K. Makhmudov, Administration of the Hydrometeorological Service of the
Uzbek SSR, manuscript received 8 Sep 80]
[Text] Abstract: On the basis of experimental in-
vestigations a study was made of different
aspects of the influence of inechanical damage
i.nflicted by hail on the growth, development
and yield of cotton plants. It is shown that
the greatest losses from hailfalls are observed
iafien tfiere is severe damage to the cotton plants
durtng the period of maturing of the bolls,
whereas the minimum damage occurs in the budd-
ing phase.
Introduction. F,esearch work for clarifying the influence of hailfalls on the growth
and development of cotton plants, insofar as is known to the author, was not carr-
ied out in our country or abroad up to the 1970's.
For the first time experiments for studying the influence of inechanical damage in-
flicted by hail on the growth and development of cotton plants under natural condi-
tions were carried out by B. A. Kamalov [2]. Later M. Makhmudov [3] and the author
[4, 5] studied the dynamics and biology of development of damaged cotton plants.
The results of these experiments indicated that hailfalls during the period of de-
velopment of cotton plants cause a considerable decrease in yield and a deteriora-
tion in the quality of the raw cotton.
Unfortunately, the mentioned studies were made on the basis of limited statistical
data and did not take in all the phases of cotton plant development.
In this article we give some results of investigations for clarifying the influence
of inechanical damage during the entire cotton plant growing season by means of sim--
ulation of hailfalls.
Method for producing artificial hailfalls. The artificial simulation of hailfalls
was used for the first time by the American researcfiers H. Laude and A. Pauli [7].
They inflicted mechanical damage on different varieties of winter wheat by means of
112
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
~
a~
~
~
H
a~
~
~
r-4
a
~
0
~
~
O
U
W
O
a~
N
a
0
~
~
a~
A
~
O
N
00
~
~
A
r-I
c11
V
M
~
t0
~
U
~
44
O
N
u
~
~
~
w
~
H
4+
-W
-W
~ i o ~n
w ~ q
~
.
o
!
�ol
oclc
l�
c
~
v U
~
b U
~
o,~
a
oo
-
v-I U
�r1 U
A O
A O
co ~4
m
~
V4 0
0
I i:
~I
~
rl~
~ ~
r.
u
4-1 w 4-1 `rl i
03
O
CkoI I 1
w
m
t
d
r1.C cd Nlr
i
i
0) �m
.
Cd
I�
�
=
ti Is
~
%
-~1,8
+17,5 ; 2:3.9
-23,9
~ --I6,2
1 -16,6
28
Damaged
1:39n
5.4 ~ i;, ;
25.2
:3
4
! 3.001 I:,:;
1
Undamaged
I4:t,~
6.0 i i.~c
.^1I,2
,
:~,4a .
:i.I,91
Difference
-4g
-0.6
-0,91
--0,5, j -u,:;S
~
%
- :,.3
-1o,1) i - I!,2
- 141.2
- 16,7 ~
-I:),U
. .
i_~ \'l
Damaged
1-438
'
5.8 I ti. i
24,4
4.24
2,79
Undamaged
I11m
6,6 ~ 5,i;
134,4
S, 12
I,G0
Difference
-0.8
-lu.U
-0.8S
-0.-1i' -O,-~I
,o
-2h.~i
-I'~,t I -29,1
-~~,1
- 17,2
-I~1. 1
27 V(
a[naged
1237
4.3 I 4.6
18,4
3,69
2.4,51 1.24
Undamaged
Iti:6
6.1 ~ 9.2
36.3
:,.,,2
:;,:j21 2.~~0
n fference
~
--`~~?i
-1.8 ~ -4.6
-i~c,4
-I.~s:3
-I,071 -U.76
o
-2:~,4
-29.5 ~ -:;U.O
-.iU.O
-33,2
-3U,4 ~S.U
13 v II
amaged
423
1.5 ~ I,+
5,ti
3.37
2.121 1.2-1
ndamaged
1599
S.8 ~ S.8
:J.5,2
-1,-52
:3.56' I.96
D
erence
-I176
-4,3 i --7,4
-29.6
-2.15
-I,44 -O,ii
W
--73,.5
-74.1 I 84.1
- $4,1
- 38,9
-40.4 -36.2
28 4'1I
llamaged
1q.;
03 j 012
O,S
2.S6
I.;s2 I.i)~1
Undamaged
1�?h`
{,g ~ I
24,4
4,34
:i.M 1, 71i
D~
fference
I1-4-1
--1.6 ; -5.9
-2:;.G
-I,9ti
-1.26 ---U,12
,
ro
- '.3.�3
-93,9 -~6,7
-9~i,i
- IU,~) i
-!U.'.~ -.10,;)
12 1'111
Damaged
53
0.3 ~ 0.2
0.8
2.961
l.,S~s I,04
Undamaged
1;J9~i
6.1 l
:i'1,4
:~.O~il
:~.24 I
N 5
'?k~ference
- I31:3
3
--31,i;
--2.13 ~
,
--1.:36 - 0,;"
-94.3
-95.1 I-97,:1)
-Qi,;i
-41.8 I
42,0 -41,i;
116
FOR OFF[CIAL USE ONLY
Table 2
r
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFIC[AL USE ONLY
Observations of the growth and development of damaged plants indicate that strong
mechanical damage in the early stage in development of the cotton plant consider-
ahly decreases the density of the plant stand. For example, at the time of inflict-
ing the first artificial hailfall on the experimental plot, where cotton plants
were damaged on 13 May 1978 in the phase of two-four true leaves, there was a total
of 284 plants, and a month after damage there were 188 remaining, that is, the den-
sity of the plant stand had decreased by 33.8%, which was not observed in the case
of later damage. Similar data for a definite phase in the development of the cotton
plant were obtained earlier during natural hailfa?ls [3, 4].
Results of yield analysis. With the opening and maturing of the cotton bolls the
cotton was harvested several times. The number of bolls in the experimental and
control plots was counted for all the experimental variants. The results of the
yield analysis are given in Table 2. The table shows that the mechanical damage
inflicted on the cotton plant by hail in all phases of its development causes a de-
crease in Che yield and a deterioration in the quality of the raw cotton. This oc-
curs both due to a decrease in the mean ninnber of bolls, scaled to one plant (by
15-95%), and due to a decrease in the mean weight of the raw cotton per boll (by
50-42%). It is noted that the later the mechanical damage to plants occurs, the
greater are the losses and the worst is the deterioration in quality of the yield.
A decrease in cotton yield of cotton plants subjected to strong damage in the early
stage of development for the most part occurs due to a decrease in the density of
the plant stand. But a decrease in the density of the plant stand during this per-
iod favors the accumulation of a great number of bolls, scaled to one plant, which
is not observed in other periods. As we see, the yield losses from strong mechan-
ical damage to the cotton plants are from 19 to 97%.
On the basis of the results it was possible to ascertain the dependence of the loss-
es to the cotton yield on the date of damage (Fig. 2). It can be seen that the max-
imum losses are observed when there is strong damage to the cotton plants during
the period of maturing of the bolls and the minimum.losses are observed when there
is strong damage in the budding phase. Suc:h a dependence,is confirmed by the re-
sults of experiments carried out during natural hailfalls on cotton fields (see
Fig. 2). It makes it possible to establish the magnitude of the postulated losses
from hailfalls during the entire growingseason for cotton plants.
Loss, %
100 i t
� ~ ~
p
y0
io 10R7~ ~ ~ G4
r
A=11 ~~JJ il.�JNb l'~MAD .I41"v111
. Apr May Jun Jul Aug ,
Fig. 2. Dependence of yield losses on date of hailfall. 1) results obtained during
natural hailfalls on cotton fields; 2) results obtained with artificial hailfall
on cotton fields; 3) losses which occur under production conditions due to resow-
ing of cotton plants.
117
FOR OFF'ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
Summary
- After analyzing the results of these experiments the following conclusions can be
- drawn:
1. In cases of a strong hailfall on cotton fields there is a lag in growth and de-
velopment of 10-70 or more days in dependence on the date of damage.
2. Hailfalls in the early stage of cotton plant development cause a decrease in the
yield for the most part due to a decrease in the density of the plant stand (by
24-44%).
3. Hailfalls exert a negative influence on the quality of the yield of cotton
plants and there is a decrease in the yield of raw cotton, seeds and fiber from
one boll (by 15-42%).
4. In general, the strong mechanical damage caused by hail reduces the yield of
cotton by 19-97% in dependence on the date of the hailfall.
BIBLIOGRAPHY
1. Dospekhov, B. A., METODIKA POLEVOGO OPYTA (Field Experimentation-Methods),
Moscow, Kolos, 1968.
2. Kamalov, B. A., "Economic Effectiveness of Antihail Work," TRUDY SARNIGMI
(Transactions of the Central Asian Regional Scientific Research Hydrometeor-
ological Institute), No 16(97), 1975.
3. Makhmudov, M., "Influence of Hailfalls on the Growth and Development of Cotton,"
TRUDY SARNIGMI, No 35(116), 1975.
4. Makhmudov, K., "Influence of Hailfalls on the Growth and Development of Cotton
Plants," TRUDY SARNIGMI, No 67(148), 1979.
5. Makhmudov, K., "Hailfalls and Cotton Yield," TRUDY SARNIGMI, No 90(171), 1980.
6. Chepovskaya, 0. I., "Preliminary Results of Investigation of Hail Distribution
Over the Earth's Surface," TRUDY VGI (Transactions of the High-M4untain Geophys-
ical Institute), No 3(5), 1966.
7. Laude, H. and Pauli, A., "Simulated Hail Injures Winter Wheat," AGRICULT. EX-
PERT. STAT. KANSAS STATE COLLEGE OF AGRICULT., Bull. 402, 1952.
118
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 551.509.324.2(574)
IMPROVEMENT IN THE METHOD FOR PREDICTING THE INTENSITY AND QUANTITY OF
PRECIPITATION IN A WARM PERIOD
t-ioscow METEOROLOGIYA I GIDROLOGIYA in Russian N!c 3, 114ar 81 pp 103-106
[Article by A. T. Kenzhibayev and I. A. Petrichenko, candidate of physical and
mathematical sciences, USSR Hydrometeorological Scientific Research Center,
manuscript received 2 Jul 80]
[Text] Abs.tract: The transformation of formulas for
the prediction of precipitation is presented.
�These formulas are based on allowance for data
on air humidity and the liquid-water content of
clouds for the purpose of use of information on
air humidity and temperature available under opera-
tional conditions. Analytical formulas are given
for predicting the inteneity, duration and quantity
of precipitation and also regression analysis for-
mulas for refining the computations applicable to
the territory of Kazakhstan in the warm period.
This investigation is an improvement of the method for predicting the intensity
and quantity of precipitation applicable to the territory of Kazakhstan with al-
lowance for change in humidity and liquid-water content of clouds proposed by
I. A. Petrichenko in [4]. Using this method the prediction of intensity Ih and
the quantity of precipitation Q is accomplished in the coordinate system x, y, z,
t using the formulas H N
~f N / d.M d:t4
~h ~ P dl J: ~.6f Jie.ry C cl: -`(V ~y ) d::
h h h
r H r N r HI. il,Ll d.tt ~
Q daq dt clf - ~ I :Wd11'.ry Ce/Z i:l -F tr ljV ) il1
-;f J � at
where P is air density, q is specific humidity at maximum saturation, C, U, V are
velocity and the components of the velocity of movement of solid and fluid ele-
ments, M is the water mass in the liquia and solid phases in a imit air volume.
Equations (1) show that the intensity and quantity of any type of precipitation are
dependent, first of all, on the decrease in specific humidity accompanying the ver-
:tical ascent of saturated air, and second, on the redistribution of water avail-
able in the clouds in the liquid and solid phases, and also on the structure of the
vertical and horizontal air currents.
119
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R440400030065-3
FOR OFFtCIAL USE ONLY
It should be noted that the peculiarities in the horizontal and vertical structure
of air currents can be partially attributed to the lack of precipitation from al-
' ready forming cloud cover over Kazakhstan in summer. Accordingly, for making com-
putations of precipitation in this region it is desirable that equations (1) be
used.
'I'he principal difficul.ty in the practical realization of equations (1) involves
the absence at the present time of information on the liquid-water content of
clouds.
On the basis of [2, 3, 71 we have proposed a possible method for using additionally
available information on air temperature and humidity instead of information on
the liquid-water content of clouds.
We will use the equation for computing the liquid-water content of clouds and the
continuity equation in the form
d:6l d diVl (2)
oz '
,ti1 d 1 %�.v v: c
where M is the liquid-water content of clouds g/m39 V is the coefficient of ver-
tical exchange within the clouds.
In concise form we will write the results of expansion of an individual derivative
of the liquid-water content of clouds, with three-dimensional divergence of the
velocity of their movement and with addition of the left- and right-hand sides of
equations (2) and (3) in the form
d;bf ditif C' dM V d.bf W d d,bf
vr - 7x ' dy + -c~r - o ' d: (4)
We will introduce the parameter of intensity of precipitation I equ=1 to the value
M lW 1�
Transferring the first three terms in equation (4) into the right-hand side, we ob-
- tair. the equation -
dl d dM dM dMV
(amu
oz r ay (5)
After integration of equation (5) from the condensation level h to the upper cloud
boundary H and the assumption that the intensity of precipitation at the upper
cloud cover boundary is equal to zero we have
H H H((--ix- dMU db9V 1
~iY1 dz " 0 s I 1 + dy ~ d:. (6)
1
We ada the f irst equation from (1) and equation (6). After shorteni.ng of such terms
the resultant intensity of precipitation at the condensation level will be deter-
mined by the equation
120
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
H 11 dM H
- F dq d:+~ dd~ a I.
(7)
In accordance with expression (7) the intensity of precipitation in the atmosphere
is determined not only by macroscale atmospheric processes (first term), but also
physical processes within the clouds (the last two terms).
In the absence of precipitation some of the atmospheric moisture content is ex-
pended on the formation and maintenance of cloud cover.
From equation (7), with some transformations taken into account, it is possible to
derive prognostic equations making it possible to use information
available from operational work.
In accordance with [1], the liquid-water content for cumulonimbus clouds can be ex-
pressed through the elasticity of water vapor saturation E in the following way:
JN = 0,57E.
(8)
The coeff icient 0.57 was determined empirically for the territory of Kazakhstan.
The second and third terms in equation (7), taking expression (8) and the Clausius-
Clayperon equation in the form dE LE I.p- q
uT -R�T=s T
into account can be represented in the following way:
- - - -
_
H H rf H
dM d~ 0, ~~7 f 'E dz = 0,57 ~ dE dT ds = 0,5,"� L ~ P9 dT ds,
` ar d dr a 7' dr (9)
dM ~ dE H- 0.57 dL� d T H~_ 0, 57 L ~ Q Tsa H `lo>
Y ds hI 0, 57 , dz I d T ds i T hl '
where L is the specific heat of vaporization (condensation), the remaining nota-
tions are generally known.
Taking (9) and (10) into account, from equation (7) we obtain the principal prognos-
tic equations which can be used in a numerical hydrodynamic forecast
of the intensity and quantity of precipitation on an electronic computer, in par-
ticular, for the territory of Kazakhstan: -
_ H H H �
lh=-~ p dj~ dz 0.57 L` JT dt ds-{-0,5? L'rT 7ea +
. r, ..I
!H rN
p=-~~ h~ F dt ds dt r 0, 57 f f PT O! dz dt + (11)
121
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY

r N
-F S 0,57 l. , 7ea l dt.
h
In the coordinate system x, y, p, t the prognostic equations have the form
� dq o,~~ c rP ar u.37 c�, Q p
in = g f ~t dv- B J T dt dp -
8T "�ea I.
Po Pn Pi)
e p r p
QI` J ~q dp d! ~.5i L `J T d1 Jp dt
J
g o P, 8 u po
r p
O,.il L'o 4 7ea
gT I dt,
i~
po
[Bd = mpist adiabatic]
(11')
where g is the acceleration of free falling; the remaining notations are well
known.
The operational numerical synoptic-hydrodynamic scheme for the forecasting of pre-
cipitation for a time up to 36 hours, developed in the synoptic research labora-
tory at the USSR Hydrometeorological Center [5, 6], can serve as a base for carry-
ing out additional computations of precipitation using formulas (11)-(11') the
second and third terms.
In computing the quantity of precipitation a factor of more than a little importance
is its duration.
The author of [3] gives a semi-em�pirical formula for determining the duration of
shower precipitation
"-ioo)-Ksi
(12)
where ~10~0 - 850 is the length of the trajectory of movement of cloud cover giv-
- ing precip tation, v700-500 is the rate of movement of the cloud cover,,X is a co-
efficient numerically equal to the ratio of the area of falling precipitation to
the area of precipitation-forming cloud cover. ,
If the coefficient Z is introduced into formula (12), it is possible to compute
the duration of both strang and weak precipitation:
t V. 1IOOU-RIA1
r v,00 _Sp
122
FOR OFFICIAL USE ONLY
(13)
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000440030065-3
FOR OFF[CIAL USE ONLY
I1ow we will clarify the physical sense of the 1Z coefficient. If in computing Ih
in equations (7) we limit ourselves only to the first term, the equation can be
represented in the form
_-A 1.
(14)
where Ih is the resultant intensity of precipitation in the atmosphere at the
condensation level, I is the intensity of precipitation caused by allowance only
for the first term in equation (7); 1'tis a dimensionless approximating coefficient
governed by the following integral expression in accordance with equations (7) and
(14):
fam dz_v 4m j (15)
d
- ~ p aq ds
It
It follows from expression (15) that in schemes for the prediction of precipita-
tion it is necessary to include computations of the layer-by-layer content of mois-
ture in the atmosphere. '
For computing the duration of continuous precipitation it is also possible to use
a formula frcm [5], for exa.mple:
[obn = cloud] 12 t (T-Td) ~ (16)
I'. .
where tcloud $50 is the duration of falling of precipitation from the 850-mb level,
t is the advance time of the forecast, (T - Td) is the dew point spread predicted
at this level at a point, taking into account the advective and transformat3onal
changes in T and Td,I'V lis the modulus of ordered ascending vertical velocity for
the predicted period at the isobaric surface 850 mb.
The duration of falling of precipitation at the earth's surface is determined by
the duration of precipitation at AT850 and AT700�
It therefore follows from the above that a determination of the intensity of precip-
itation in the atmosphere and its duration using formulas (11), (11'), (13) or (12)
makes it possible to predict the quantity of precipitation in different regions, in-
cluding with a continentally arid climate, in particular, in the territory of Kaz-
akhstan.
The prediction of precipitation for the territory of Kazakhstan on the basis of its
intensity and duration computed using formulas (11)-(13) can be improved with the
additional use of data from a regression analysis, making it possible to take local
regional conditions into account.
For example, far one of the stations in Kustanayskaya Oblast, with use of data from
automatic recorders on the intensity I, duration t and quantity Q of precipitation
for the warm period of the year on the basis of observational data for 1967-1976
(100 cases were examined) a regression equation was derived in the form
123
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2047/02109: CIA-RDP82-00850R040400030065-3
FOR OFFICIAL USE ONLY
Q = Cla t�, (17)
where the coefficients C= 0.794, a= 0.748, b= 0.586 with a correlation coef-
ficient equal to 0.68.
Substituting the I and t values, computed using formulas (11)-(13), into formula
(17), we obtain the quantity of precipitation for 12-hour time intervals for an
advance time up to 36-48 hours.
It should be noted that computations of precipitation on the basis of formulas (11)-
_ (13) can be made for any forecasting regions, whereas formula (17) must be derived
statistically specifically for each forecasting point.
BIBLIOGRAPHY
1. Yeviyevich, T. V., "Empirical Formulas for Betermining the Total Moisture Con-
tent in the Atmosphere Over Moscow," RADIATSIONNYY REZHIM I OSADKI V MOSKVE
(Radiation Regime and Precipitation in Mnscow), Moscow, Izd-vo MGU, 1967.
2. Marchuk, G. I., "Theoretical Mode1 of Wea.ther Forecasting for a Short Time in
Advance," IZV. AN SSSR: SERIYA GEOFIZ. (News of the USSR Academy of Sciences:
Geophysical Series), No 5, 1964.
3. Orlova, Ye. M., KRATKOSROCHNYY PROGNOZ ATMOSFERIdYKH OSADKOV (Short-Range Fore-
casting of Precipitation), Leningrad, Gidrometeoizdat, 1979.
4. Petrichenko, I. A., "Prediction of Precipitation Using Data on the Liquid-
Water Content of Clouds," METEOROLOGIYA I GIDROLOGIYA (Meteorology and Hydrol-
ogy), No 10, 1971.
5. Uspenskiy, B. D., Mertsalov, A. N., Orlova, Ye. M., Belousov, S. L., Petri-
chenko, I. A. and Veselova, G. K., "Synoptic-Hydrodynamic Scheme for the Quanti-
tative Prediction of Continuous and Shower Precipitation," TRUDY GIDROMET-
TSENTRA SSSR (Transactions of the USSR Hydrometeorological Center), No 157,
1976.
6. Uspenskiy, B. D., Mertsalov, A. N., Orlova, Ye. M. and Petrichenkb, I. A.,
"Peculiarities of Prediction of Precipitation in Numerical Operational Synop-
tic-Hydrodynamic Schemes With an Advance Time to 36 Hours," TRUDY GIDROMET-
TSEIITRA SSSR, No 176, 1977.
7. Yudin, M. I., NOVYYE METODY I PROBLEMY KRATKOSROCHNOGO PROGNOZA POGODY (New
Methods and Problems in Short-Range Weather Forecasting), Leningrad, Gidro-
meteoizdat, 1963.
124
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 551.594.221
POSSIBILITY OF PREDICTION OF LIGHTNING DISCHARGES
Moscow METEOROLOGIYA I GIDROLOGIYA in Rus.sian No 3, Mar 81 pp 107-108
[Article by V. I. Arabadzhi, professor, Tula Polytechnic Institute, manuscript re-
ceived 23 Jul 80]
[Text] Abstract: The possibility of monitoring the
electric field in a well-developed cumulus
cloud up to the formation of a lightning
discharge is examined. On this basis the
- author proposes prediction of development
of lightning discharges in clouds.
In order to clarify the possibility of predicting lightning discharges in the
weather service it is necessary to have a more complete knowledge of the condi-
tions for the development of cumulus clouds into thunderstorm clouds. We will at-
tempt to give some rough evaluations of this process.
We will examine a well-developed cumulus cloud, in the first approximation regard-
ing it as a heat engine. The work performed by this engine in the separation of
electric charges and in creating an electric field can be expressed by the for-
mula

.4 - ~ Qi ( T, T, T2
where 41 is the energy expended on charge separation in the cloud, T1 is the tem-
perature of the heater (the mean temperature of the lower part of the cloud), T2
is the temperature of the cooler (the mean temperature of its upper part), y is
a factor less than unity taking into account the efficiency of operation of a real
heat engine in comparison with an ideal engine (in our case it characterizes the
energy losses in the formation of the electric field in the cloud).
We will compare the work according to (1) with the work which is performed in a
lightning discharge. Assume that the electric energy of the cloud before and af-
ter the discharge is equal to CV~/2 and CV2/2, where C is cloud capacity and Vp
and V are its potential before and after the discharge. Then the work of the dis-
charge is
CV2 CV2 z C~'~~ Y).
A (2)
Equating (1) and (2), we obtain
125
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Equating (1) and (2), we obtain
CV V- VT, - T
o ( o )=7Qi( ~ (3)
T, r
We will ass wne further that the kinetic energy of a cloud unit volume, as a re-
sult of charge separation in it, is transformed into the electric field energy
in conformity to the equation
?vI I vo E'
Y Z --2-,
(4)
where,P is the density of the charged component of the aerosol medium (cloud), v
is the rate of separation of charges of opposite signs, � 0 is the electric con-
stant, E is the dielectric constant, E is electric field strength, Y is the already
mentioned coefficient of energy losses associated with formation of the electric
field in the cloud. Expressing'Y f rom (4) and substituting the result into (3),
we obtain
~c�Ei-
CVl. ( ti n- Y) u: = Qi T T ( T~
(S)
We rewrite Q1(T1 - T2)/T1) in the form S(T1 - TZ), where S is the entropy lost by
the heater and we will denote the mean value of lightning discharge energy, which
we will use henceforth as an approximate parameter, by W. Then for the electric
field developing in the cloud we will have .
~ W
(6)
Since (6) was derived with allowance for the entropic exchauge in the cloud, it
can be regarded as quite important for evaluating the possihility of development
of a discharge. By stipulating the mean energy of the lightning discharge W and
experimentally determining the /4 , v, T1, T2, E and S values, using formula (6)
it is possible to evaluate the degree of approximation of the electric value to
that critical value which will ensure the development of a lightning discharge
- (we note that with allowance for d der.sity ot charged aerosols ana thL-'r curona
discharge typical for thunderstorm clouds the field strength responsible for a dis-
charge in a cloud has a value of about 106 V/m).
For practical purposes it is important to foresee the region of the beginning of
the discharge and the direction of its branching. In most cases the discharge be-
gins in a droplet or droplet-ice part of a thunderstorm cloud where there are
favorable conditions for this (additional electric fields as a result of fragment-
ation of the charged droplets and the falling of precipitation, some increase in
the length of the free path of electrons in comparison with surface conditions,
which ensiire tite development of tlie discliarRe channel, and a decrease in the ion-
ization energy of molecular associations and complexes with individual molecules).
The discharges, beginning and ending in the cloud or in the space directly adja-
cent to it, require a lesser energy for their development and therefore are observed
more frequently than discharges at the ground. However, often there are dischar.ges
excited by the electric field of a thunderstorm cloud, but not beginning in the
cloud, but at the ground surface discharges into high atructures. For example,
it is. known that discharges into the Ostankinskaya television tower above 500 m
are ascending, i.e., branch from the tower to a cloud. Discharges into the Empire
State Building in the United States become ascending beginning at a height of 390
126
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034465-3
m. This effect is attributable
upper parts of high buildings.
conformity to the law
FOR OFF(C1AL USE ONLY
to the high strength of the electric field at the
In the first approximation this field increases in
h
E==F.~, R,
(7)
where Eo is the field of the thunderstorm cloud at a plane earth's surface, h is
height of the structure, R is the effective radius of curvature of the discharge
apparatus at the top of the structure.
Assuming for the Ostankinskaya television tower h= 536 m, R-1-0.7 m and at the
earth�s surface Ep N 4�103 V/m, for the electric field strength at the top of the
tower ~ze obtain E N 106 V/m. With such a field strength in dry and pure air at the
earth's surface there is an electric charging of the air. At the top of the tower,
for the most part due to coronal discharge suitable conditions will prevail with a
lesser field s*_rength. Thus, by knowing the height of the structure it is possible
to foresee the direction of development of the lightning discharge over it.
We have briefly examined some points which can find application in work on the pre-
diction of lightning discharges. Further research work is required for explaining
the possibility of introducing these proposals into weather service practice.
BIBLIOGRAPHY
1. Arabadzhi, V. I., GROZA I GROZOVYYE PROTSESSY (Thunderstorms and Thunderstorm
- Processes), Minsk, 1960.
2. Gorin, V. P., Sakharova, G. S., Tikhomirov, V. V. and Shkilev, A. V., "Results
of Observations of Lightning Damage From the Ostankinskaya Television Tower,"
SBORNIK TRUDOV GOSUDARSTVENNOGO NAUCHNO-ISSLEDOVATEL'SKOGO INSTIW[JTA IM. G. M.
KRZHIZHANOVSKOGO (Collection of Papers of the State Scientific Research Insti-
tute imeni G. M. Krzhizhanovskiy), No 43, 1976.
127
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 556.34
DETERMINATION OF FILTRATION COEFFICIENTS OF COHESIVE SOILS IN A FROZEN STATE
THROUGH THEIR KINETIC SPECIFIC SURFACE
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 109-112
[Article by V. I. Shtykov, candidate of technical sciences, Northern Scientif ic Re-
search Institute of Hydraulic Engineering and Melioration, manuscript received 15
J ul 801
[Text] Abstract: The article describes the derivation
of a dependence for determining the filtration
coefficients for cohesive soils in thawed and
frozen states through their kinetic specific
surface. The author gives recommendations on
the determination of individual characteristics
of the ground entering into the cited dependences
- and also the results of a comparison of the fil-
tration coefficients computed using the proposed
dependences with their values obtained from ex-
periments under laboratory conditions.
A determination of the filtration coefficients for cohesive soils in a frozen state
involves great difficulties under both field and laboratory conditions. It is there-
fore of considerable practical interest to examine the possibility of using other
methods, such as computations, when only individual characteristics of the ground
in a thawed state are to be determined under laboratory conditions.
As in the case of friable soils, it is desirable to use as a point of departure a
filtration scheme for the ground in which it consists of a mass of filtration pas-
sages. For friable soils we have the following dependences for dztermining the rate
of movement and position of the percolation front with time [4]:
dy 7 a~t r 4 A: ros H h: t (1)
. ,
d c `J r= y~, du 'i Y T` y ~
S !a 4 :1 o rus H-}- Q, 7 h �1 .4 r cuti A a- d, h 1
t i ali ( d. I - In { A: cus H t du ~ h t ifn T Y ~(2)
where y is the distance from the ground water level to the percolation front in a
vertical direction (with percolation from above, from the surface), cm; 'G is time,
in seconds; Y is water density, g/cm3; du is the computed diameter of the filtra-
tion passage, computed from the dependence (6), cm; f,e. is the dynamic viscosity
coefficient, g(cm�sec);
128
FOR OFFICIAL USL ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
,
(n I. --2.ti(i- [i)
.4 = ) l l ti ~ i ~ H
(n - i B) I 1 - n - i~'� - 1.8 (i - BI ~ .
~ t --U,h ~ i T B
A is a parameter taking into account the content of unfrozen water in the friable
- soil, which with Wtm = 0 is equal to 1; d is the coefficient of water surface ten-
sion, n/cm; B is the wetting angle; h is the water layer at the ground surface with
percolation fr.om above (with movement from below h= 0), cm; n is porosity in frac-
tions; i is volumetric ice content in fractions; B is the content of entrapped air
in fractions.
In the case of cohesive soils the computation diameter of the filtration passage
is expressed through the kinetic apecific s,irface of the ground. We use the known
fact [1] that the 2n/E0 ratio is equal to the mean radius of the pores. We will as-
sume that the above-mentioned expression is equal to the computed radius of the
filtration passage when � p is understood as the kinetic specific surface of the
ground, that is, "dead-end" pores are excluded from consideration. Taking into ac-
count what has been said, dependence (1) for coherent soils in a dry state will
have the form
'2 z n': rus H j h
,t : . L � y Y y ~ ( 3)
where g is the acceleration of free falling, cm/sec2; is the kinematic viscosity
coefficient, cm2/sec.
With y-�oo, and also joining of the percolation front with the ground water level,
the infiltration process passes into a filtration process and by virtue of the
known expression dy KI
d'G n
from (3) we obtain a dependence for determining the filtration coefficient for co-
liesive soils.
z sn` - (4)
where I is the pressure head gradient, in this case equal.to l.
In the case of clayey ground the term in parentheses in expression (3) can be
neglected, and with dependence (4) taken into account we obtain the B. V. Deryagin
zormula [1]
Y2 2 Ktocos9
S = n: Y � (5)
In the derivation of formula (5) the author used as a point of departure the equal-
ity of work of wetting forces and frictional forces, that is, the work of gravita-
tional forces is neglected. Accordingly, it is rigorously applicable only to the
case of horizontal percolation, and in the case of vertical percolation to fine-
grained soils, and in particular, to clayey soils, for which the capillary potential
is much greater than the gravitational potential.
129
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034465-3
FOR OFFICIAL USE ONLY
Dependence (4) gives the value of the filtration coefficient at the beginning of
the process of infiltration into dry ground. Upon the elapsing of some time with
the saturation of cohesive soils with water the latter "looaen up." A film of
bound water is formed around each of the particles and this film, with an accuracy
adequate for practical purposes, can be considered immobile. With a moistening of
the cohesive ground there is no substantial increase in the initial ground volume,
but the total surface of the particles increases since for this we use the surface
of particles with allowance for the immobile film of bound water, which leads to a
decrease in the activo porosity of the ground and an increase in the kinetic spe-
cific surface. An increase in the specific kinetic surface is confirmed experiment-
ally (see Table 2, columns 3 and 4).
We will assume that in the case of cohesive ground as well the structure of the for-
mula for determining the computacion diameter of the filtration passage is the same
as in the case of friable soils [S]. Then for cohesive soils in a moist state we
write
n13 - 7,. It~ ~
n T li + t~~ lf', (6)
where Yo is the volumetric mass of ground in a dry state, g/cm3; Wc is the content
of firmly bound moisture in fractions of the mass of dry ground; C is a parameter
dependent on the kinetic specific surface of the ground.
For cohesive ground in a dry state dependence (6) has the form
n
[lu = C i _n �
(7)
Using the relationship between the mean radius of the filtration passage and the
kinetic specific surface of the ground, we express the C parameter through E p:
Csm 4 (1 --n)
10

The value of the computation diameter of the filtration passage, computed using
the dependence (6), corresponds to the kinetic specific surface of the cohesive
ground in a water-saturated state W). Using the dependences (6), (8), and also
the expression .
. 4 (n
dn
we obtain a formula reflecting the interrelationship between the kinetic specific
surface of the cohesive ground in dry and moist states,
(I-n-R+Y, W,) co
`w= (1 --n)
(9)
By analogy with moist cohesive ground, in the case of its freezing the dependences
for determining the computation diameter of the filtration passage and the kinetic
specific surface will have the form
n-i.. B-~1~c
d� C - 2,8 (i + BI ' (10)
1 n a '(n Wr + 1 -0,6 V! B
130
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
4 (n-B--l--Yn�'c)
d.
,
(11)
As a result the kinetic specific surface of cohesive ground in a frozen state is
expressed through the kinetic specific surface of this same ground in a dry state
in the following way :
II-t+7nWc+l -�O,fivl7fj~~O
-n ~
(12)
Thus, for determining the filtration coefficients for cohesive soils in a water-
saturated or in a frozen state we have the dependences
2Bln-B-7uWc1'(1--a1~ (13)
Kw r.2 0 - n+ B-F- 70 Wc)l e0 ~
2g(n-B- ioWc-1)3 (I - njl
K,t~ = r 2,8 (i tt) 13 s' (14)
r2II-n;7nWcT . 1 ap�,
I. t-o,6j/l.rtt
It should be noted that dependence (14) is valid only for cohesive ground in which
heaving is absent. The volumetric ice content of the ground i consists of the ini-
tial io ice content and the increment to ice.content as a result of the partial
freezing of water filtering into the ground i'. It follows from an analysis of de-
pendence (14) that frozen friable soil can become impermeable with an i value sub-
stantially less than the porosity due to the presence of entrapped air and tiound
water in it.
Infiltration into frozen cohesive soils and the conditions under which it ceases
were examined in a study by I. L. Kalyuzhnyy and others [3].
Table 1
Content of Unfrozen (Wun) and Firmly
Bound.(Wc) [Jater in Frozen Ground
Name of soil by Wun with t= 0�C Wc in fractions
mechanical compo- in fractions
sition
Sandy loam
0.07
0,02
Clayey loam
0.10
0.04
14edium clay
0.20
0.08
iieavy clay
0.35
0.20
Now we will examine the problem of determining the individual parameters entering
into the dependences cited above. In the absence of data on the content of en-
trapped air its value must be assumed equal to 0.03. Having data on the content
of unfrozen water in the ground (Table 1), the initial value of the volumetric
ice content can be determined using the N. A. Tsytovich formula
131
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
io = Yp(W - Wun)/ rice, (15)
where W is the total moisture content of the ground by mass in fractions; wun
is the content of unfrozen water by mass in fractions in the frozen ground; -Y ice
is ice density, g/cm3.
Table 2
Results of Determination of Filtration Coefficient of Plowed Layer in Thawed State
Kinetic specific surf ace Filtration coefficient
n B of ground, cm 1 10-4 cm/sec
dry moist experi- using (14)
mental
0.484
0.020
1410
1586
5.5
5.2
0.501
0.024
1370
1550
4.9
6.0
0.498
0.031
1376
1576
5.0
5.4
0.493
0.029
1390
1589
5.3
5.2
0.489
0.034
1400
1610
4.7
4.7
0.478
0.039
1430
1660
3.9
3.9
0.473
0.032
1446
1660
3.5
3.9
0.464
0.033
1454
1670
3.8
3.6
Table 3
Results of Determination of Filtration Coefficient of Plowed Layer in Frozen State
~ Kinetic spe- I cif3:c surface Filtration coefficient,
B of ground cm L cm/sec
0,468
0.000
0.021
1460
1810
1,9� 10-1
2,1 � 10-'
0,491
0,059
0,022
1400
2270
1,3-10-4
1,1 � 10-'
0,500
' 0.084
0,028
1365
2560
0,8�10-4
0,7�10-4
0.506
0,111
0.000
1353
2520
8,0� 10-s
7,8. 1p_a
0,547
0,121
0,020
1240
3020
5,2 � 10-6
5,9 0-6
0.513
0.161
0,029
1330
3430
2,4� 10-5
2,0� 10-5
0,525
0,177
0.032
1300
3630
1,6�10-6
1,7�10-5
0,540
0,212
0,037
1260
4100
0,8� 10-5
1,0.10-5
0,528
0.266
0,039
1290
4900
2.4�10-"
2,7�10-6
In conclusion we wi11 cite experimental data for determining the filtration coef-
f.icients in cohesive soils in thawed and frozen states. The kinetic specific surface
of the ground in a thawed state was determined by the B. V. Deryagin method [2] and
in frozen ground was computed using the expression (12).
132
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Tables 2 and 3 give the results of computation of the filtration coefficients for
two different soil types of the plowed layer in thawed and frozen states and also
data from their experimental checking.
Tables 2 and 3 show that the coincidence of the results of computations using the
proposed expressions and the experimental determination of the values of the fil.-
tration coefficients for cohesive soils is entirely satisfactory.
BIBLIOGRAPHY
1. Deryagin, B. V., "Determination of the Specific Surface of Porous Bodies From
the Rate of Capillary Saturation," KOLLOIDNYY ZHURNAL (Colloidal Journal),
Vol VIII, Nos 1-2, 1946.
2. Kachinskiy, N. A., FIZIKA POCHVY (Soil Physics), Part I, Moscow, Vysshaya
Shkola, 1965.
3. Kalyuzhina, I. L., Morozova, N. S., Pavlova, K. K. and Romanov, V. V., "Ther-
mophysical Method for Computing the Losses of Melt Water on Infiltration Into
Frozen Soil," METEOROLOGIYA I GIDROLOGIYA (Meteorology and Hydrology), No 1,
19 72 .
4. Shtykov, V. I., "Permeability of Frozen Friable Soils and Analysis of the Dy-
namics of its Change During the Period of Snow Melting and Thaws," METEOROLOG-
IYA I GIDROLOGIYA, No 12, 1979.
5. Shtykov, V. I., "Rate of Advance of the Percolation Front in the Ground,"
SBORNIK DOKLADOV PO GIDROTEKHNIKE (Collection of Reports on Hydroengineering),
No 13, Leningrad, Vsesoyuznyy NII Gidrotekhniki im. B. Ye. Vedeneyeva, per
I I3-409, Informenergo, 1977.
133
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
UDC 556.132(571.1)
EVAI,UATION OF A}.'PLICABILITY OF DIFFERENT METHODS FOR DETERMINING EVAPORATION FROM
A WATER SURFACE IN A ZONE OF HUMMOCKED SWAMPS
Moscow METEOROLOGTYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 112-115
[Article by Yu. P. Moskvin, State Hydrological Institute, manuscript received
8 Aug 80 ]
[Text] Abstract: A study was made of the problems
involved in applicability of different meth-
ods for determining evaporation from a water
surface in the zone of hummocked swamps in West-
ern Siberia. As a standard use was made of the
method for determining evaporation on the basis
of observational data in the GGI-3000 floating
_ evaporator with the introduction of correspond-
ing corrections. The accuracy in determining
evaporation from a water surface is evaluated
by computation methods with and without observ-
ational data over the surface of a lake.
E}dsting methods for computing evaporation from a water surface for the most part
are based on experimental materials obtained on the basis of observations over the
European USSR, in Central Asia, Kazakhstan and in the southern part of Siberia. Due
to the intensive economic exploitation of regions in the Far North the need arises
for developing an observation method and choice of an optimiun method for computing
evaporation from a water surface for these regions. The West Siberian Expedition
of the State Iiydrological Institute over a period of years carried out work for in-
vestigating the water balance components of hummocked swamps and in particular, for
estimating evaporation from the water surface of small lakes within swamps employ-
ing a floating apparatus with a standard GGI-3000 evaporator.
The basis for this study was observational data obtained during the warm periods
1978-1979. In making the observations we selected a lake with an area of 0.178 km2
situated within a swamp which was the most characteristic for the investigated re-
gion. This lake has an average depth of about 1 m and is situated in an unwooded
complex with flat-topped hummocks. Trees with a height of 2-3 m were situated 30-80
m from the lake shore on the northern and northwestern sides and 150-200 m from
the western part of the lake. There were no trees on the southern and eastern
sides.
134
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
Eobs
$Q
YO
j
nx
_ XECO!np
10 f0 E. nn ~
Fig. i. Correlation graphs of 10-day sums
of evaporation from water surface deter-
mined from data from observations with
floating evaporator apparatus (Eo S) and
computed using formula (2) (E mp� 1978-
1979. 1) observations 4 timesca day; 2)
observations 8 times a day; 3) correla-
tion for observations faur times a day;
4) correlation for observations eigrt
times a day.
:"ObS HM
16
:)mp
Fig. 2. Dependence of monthly sums of
evaporation from water surface deter-
mined from observational data from
floating evaporator apparatus and com-
puted using data from continental sta-
tion in absence of observational data
for 1978-1979. 1) using formula (5),
2) using formula (2).
A full complex of observations with the floating evaporation apparatus, including
observations of the meteorological elements over the water surface, were made in
accordance with the INSTRUCTIONS [2]. In addition, in 1979 observations were made
with continuous registry of the principal meteorological elements (wind speed,
relative humidity and air temperature),.which made it possible to obtain the mean
daily characteristics of the mentioned elements of the meteorological regime over
the water surface for eight observation times. The elasticity of water vapor over
a water body was determined using psychrometric tables on the basis of data regis-
tered eight times a day on relative air humidity taken f rom hygrograph tapes and
reduced to the readings of an aspiration psychrometer.
As a standard we employed the method for determining evaporation from a water body
on the basis of observational data obtained with the GGI-3000 floating evaporator.
In accordance with jl], evaporation from the lake at the site of the floating evap-
orator was computed using the formula
E - 0.8 E' ep e'"" 4 - e:bn , (1)
where E' is evaporation according to the GGI-3000 floating evaporator, in mm; 0.8
is a factor taking into account the instrument correction for the GGI-3000 evapor-
ator; eo, e'o is the maximum water vapor elasticity, computed on the basis of tem-
perature of the water surface in the water body and in the evaporator respective-
ly, in mb; e200 is absolute air humidity at a height of 200 cm from the water sur-
face on the floating evaporation apparatus, mb.
135
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
0 40 BO f, nn

APPROVED FOR RELEASE: 2047102109: CIA-RDP82-00850R400404030065-3
FOR OFFiCIAL USE ONLY
Table 1
Determination of Evaporation (mm) From Water Surface by Different Methods
Using . Usingr State HXrirol o~i cal Inst _ formulas
(1) with in absence of obser-
Ten-day with observationa
Month GGI-3000 vational data
period evapora-
tor data g times 4 times uRintt (2) usintz (5)
1978 r.
July
August
September
June
July
August
September
October
2
26(8)
29(8)
3
33
35
1
I
18
21
2
20(9)
22(9)
' 3
15
16
, Month
53(30)
59(30)
62
72
1
14
15
2
7(9)
10(9)
I 3
4(7)
5(7)
Month
25(26)
30(26)
27
35
1979 r.
19-30
22
24
27
3
13
1
32
41
41
2
34
41
42
3
46
49
52
Month
112
131
135
113
123
1
24
27
27
2
29
34
33
3
20
22
23
Month
73
83
83
91
91'
1
18
18
21
2
9
9
10
3
11
10
15
Month
38
37
46
50
50
1-3
3
3
4
248
278
295
257
284
Note: The number of days of observations is given in parentheses.
The evaporation quantity determined using formula (1), as a result of the insignif-
icant area of the lake, was used as the mean layer of evaporation from the entire
lake. For evaluating the applicability of the computation method when observational
data are available we computed the 10-day evaporation sums using the generalized
formula of the State Hydrological Institute
E-0,14 n (eo-em) (1+0,72 u2m), . (2)
where n is the number of days in the computation time interval, u200 is the mean
caind velocity over the water body at a height of 200 cm, m/sec.
136
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
In 1978 the computations were made using four observation times each day and in
1979 using eight observation times each day.
The results of the computations are given in Table 1 and are illustrated in a graph
(Fig. 1). An analysis of the data in the table and graph shows that computations of
evaporation using the formula of the State Hydrological Institute with the use of
data from observations four and eight times each day leads to a systematic exagger-
ation of the results. The magnitude of the indicated exaggeration in the case of ob-
servations eight times each day averages 10-13%, and in the case of four times each
day 17-20%. The correlation coefficients for these dependences are equal to 0.9,
which makes it possible to recommend as a computation method the determination of
evaporation from intraswamp lakes in the considered region using the dependences
E = 0.84Efour timesp (3)
E = 0.90Eeight times� (4)
For evaluating the applicability of computation methods for the conditions of hum-
mocked swamps in the absence of observations at the lake we made computations of
the monthly sums of evaporation from the water surface using data from the near-ly-
ing meteorological station Tarko-Sale. In computing evaporation we used formula (2)
for the conditions of absence of ohservations at the lake and the formula
E = EarbKul~ ean'
(5)
where Earb is the evaporation from an arbitrary water body, Ku and KLmean are co-
efficients taking into account the actual wind velocity and the length of the fetch
over the water body.
The computations of the parameters entering into the mentioned expressions were made
using data from a continental station with allowance for the transformation of
meteorological elements by methods recommended in [3]. The'results of the computa-
tions are given in Table 1 and in a graph (Fig. 2). The correlation coefficients
ior the derived expressions (Fig. 2) are equal to 0.81. The collected data give
Uasis for assuming that the computation of the monthly sums of evaporation on the
basis of ineteorological data for continental stations can also be accomplished for
a zone of hummocked swamps. However, final conclusions concerning the accuracy of
computation of these parameters can be drawn after obtaining the necessary addition-
al observational data.
BIBLIOGRAPHY
1. Vuglinskiy, V. S., Starovoytova, V. K. and Cher.skaya, Ye. N., "Method for Eval-
uating Evaporation From the Surface of a Water Body Using Data From the GGI-
3000 Continental Evaporator," TRUDY GGI (Transactions of the State Hydrological
institute), No 274, 1980.
2. NASTAVLENIYE GIDROMETEOROLOGICHESKIM STANTSIYAM I POSTAM (Instructions for Hy-
drometeorological Stations and Posts), Issue 2, Part 2, Leningrad, Gidrometeo-
izdat, 1961.
3. UKA7.ANIYA PO RASCHETU ISPARENIYA S POVERKHNOSTI VODOYEMOV (Instructions on Com-
puting Evaporation From the Surface of Water Bodies), Leningrad, Gidrometeoiz-
dat, 1969.
137
FOR rr FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400034065-3
FOR OFFICIAL USE ONLY
UDC 551.(524.73+507.362)
RECONCILING OF MESOSPHERIC TEMPERATURE VALUES MEASURED BY DIFFERENT ROCKET
SOUNDING SYSTEMS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 116-120
[Article by Yu. P. Koshel'kov, candidate of geographical sciences, Central Aerolog-
ical Observatory, manuscript received 21 May 80]
[Text] Abstract: The long-term atmospheric temper-
ature values obtained at different rocket
sounding stations using different measure-
ment methods are compared. The results of
these comparisons make possible the reconcil-
ing of inean data on the temperature of the
mesosphere obtained at a number of stations.
Introduction. In order to construct reference models of the atmosphere, in the an-
alysis of the temperature and pressure fields in the middle atmosphere at the scale
of a hemisphere and for other purposes it is necessary to reconcile the data ob-
tained using different rocket sounding systems, that is, obtained by different
methods. At Soviet stations sounding is now accomplished using the M-100B meteor-
olo gical rocket with a"361M" nosecone (prior to 1978 with a"361" nosecone) and
using a resistance thermometer. In addition, in such work use is made of the AQiR-06
rocket in which the sensing element for measuring temperature is a thermistor. In
the United States it is customary to employ Super-Loci-Datasonde rockets (prior
to 1970 the Arcasonde rocket), also employing thermistors. It is less common to
use the acoustic rocket-grenade method, the method of f alling gas-filled spheres,
manometric methods (Pitot tubes, etc.).
It is known that data from measurements made by different methods differ above 50
km. This was confirmed in direct comparisons of rocket systems [8, 10]. However,
such comparisons of necessity were based on a limited number of synchronous meas-
urements and involved only some sounding systems. Therefore, it is desirable that
their results be supplemented by "indirect" comparisons in which there is a compar-
ison of the mean temperature values obtained at different points (or in different
years) under similar atmospheric conditions, for example, in summer when the longi-
tude temperature differences in the stratomesosphere are small (the influence of
variation with latitude can be excluded by the interpolation of temperature at
the necessary latitude). The temperature differences found in such comparisons are
attributable both to the difference in the sounding equipment used and the resid-
ual difference in atmospheric conditions in the compared series of ineasurements.
138
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
The results of this sort of comparison are given below.
M-100B rockets and Ameri^an network rocket systems. Table 1 gives the differences
between the mean temperatures obtained using the M-100B ("361") system and Amer-
ican systems. For the northern hemisphere we have compared M-100B sounding data
(stations at Kheys Island, Volgograd, Tumlia) and Datasonde data (North American sta-
tions) for the summer of 1973-1978, in the southern hemisphere shipboard sound-
ing data with the 14-100B (Indian Ocean) for 1961-1977 and Arcasonde and Datasonde
data (Mar Chiquita station in Argentina) for the siunmer of 1968-1977 at latitude
38�S and mean annual data for 1964-1975 at latitude 8�S (Easter Island).
Some of the discrepancies (L 1) in Table 1 can be attributed to the peculiarities
of the aerodynamics of ineasurements of temperature by an M-100 ("361") rocket.
(The 0 1 values for winter conditions should be greater than the summer values
cited in Table 1). In addition, during recent years there has been an improvement
in the calibration method (the A2 effect in Table 1) and there has been some re-
finement of the temperature restoration coefficients used in computations of the
velocity correction for the M-100B rocket (,63 effect). Another part of the discrep-
ancies (44) arises as a result of the difference in the time of day when the M-100B
("361") and Datasonde rockets were launched. There are evidently also other reasons
for the discrepancy in data, including those associated with errors in American
measurements.
As indicated in Table 2, the data from an indirect comparison in general are con-
firmed by the results of the direct comparison of 1977, involving the M-100B
("361M") and Datasonde systems. Due to the approximate nature of the quantitative
evaluations of the differences in the conditions for the indirect and direct com-
parisons (4 1, A2, Q 3 and Q q), as well as the assumptions made in the indirect
comparison, one should not expect a total coincidence of the results of these com-
parisons.
Grenades, Pitot tubes and gas-filled spheres. A study was made of the differences
between the mean temperatures determined on the basis of data from 20 rocket-gren-
ade measurements (in 1960-1970) and 5 measurements using Pitot tubes (in 1966-1970)
at Wallops Island during summer [15, 16] (see figure): The measurements made using
Pitot tubes give somewhat lower temperatures than the grenade measurements in the
lower mesosphere and higher temperatures in the upper mesosphere. However, the sig-
nificance of these differences is great only at an altitude of 55 km. The differ-
ence decreases still more if only the grenade measurements for August (when all
the measurements with the Pitot tube were made) are used. Although the volume of
data is not great, it can be concluded that it is possible to make joint use of data
obtained with grenades and Pitot tubes.
Measurement data obtained using spheres in the summer of 1970-1974 [12] at Woomer
station (31�S), supplemented by data from a so-called "dropsonde" (in which the
systematic discrepancy with spheres was eliminated 14]) and reduced to a latitude
38�, are 1-3�C warmer than grenade data for 38�N (see figure). However, the differ-
ences are not significant.
139
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
~
Gl
.-1
Cd 4-1
H 3
~ 4
41
~3
~ v
A -W
d .~L
~ V
~a
1.1 PQ
~-l 0
b0'j ~
~ z $4
N P a1
D O 9a
44 ~ c,ud ~
~ A w
00
O ~ ~
o
v r--{ U)
u (j~
:3 0 ~
H 4-4 00
Rf 41
y }Q)j ~.coi
4.1 U N
cd
N ~
a cn
y v pp
U tco.i iH-~
q A
Q)

~ N
tU N
d U m
co�z�
~ v -
p c''1
N r� Z
G1 ^
~
~ u
CA
~

U ~
H H
L+4 O
44 U
~
A
a
~
u:kAtq,LqLqLq
~
00 riun d'un
N M N ^
�oo r~ti�
M
00
~I Lq ,I Ln Lq
o 9M1-~1l ^ i
Q~ M CV
~
I N N
~ NMN0
w
zl
~ MMN~~~
J
~

t~ x
N
N
~
~
H
co
~4 H
b
U ~
'
b ~ ~ ~b
Q
~
~
~
o ~
i c
a
i a
d
cd 'b q q ~
~
41 O N ~
~t
A N E. H.fl
a
~
~
L'' 3
O .
O
-F
1-1
~ N
9:
cd
r-I
'C O -ri
-I-
H U~
O (k) rl ~T ~
~N
U G 4-~ Q O
)"U)
G

p
~ r-1 U r--H
r-I
-T
co
o 44
Q
-i
41 r
N
U
10
O ~
O
J
L'+
j
~
~
~ O
,
(L) c
d
N 0
P
~
A

,
-I
N
)IJ a
1J ~--I cd -H %O
M
>
~
F+ ? O
N
a
N W ri
41
~
'b O
i~
~
~
q
-
H N A ~a bO
rl
U �rl ~--I A
M
U
Q tA
U'C
H c
d
Q
C
,
g
id ~ U
C
.
pn ~
~
QI
~..J ~1
N
l.i U ~ fA ~
V
d a~ r-i D, ~
rn
z
v
pq
r.
.0
0 ~ q
Q
�
! t
G
a1 0
d
10 '1
a, ~n
+
~ o cv
w +
a
A
U
4N
'd
9: O ~
41
a
4
~ ~4
~
4
d
~ t rn O
ri W
~'1
O CJ
N
O
U ~ 4~ -r-I U
~
p v 41
GJ
~
i
PQ ~ N
~
44 O C+' 1J 44
rl
W r-1 cd tA 44
1J
140
FOR OFF[CIAL USE ONLY
r-i -It 10 U1 N
r-i r-i
in
,O N I, M r-i
r-i r-I
u1 Ln u1 u1
r-I rl O O r-I
Ln Ln
p .-t M r-I O
Ln
N N Cn N r-I
c7% cn N I- N
r-i '--I
Lr'1 u1 Ln u'1 u1
. . . . .
00 M tf1 ~t u'1
N C'rl N rl
O u1 O U1 O
I- %O 10 ul U1
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000400030065-3
~
a~
~
H 'd 1 cd
cC30 ~ F+ ~
y,l w
~ ~
~ ~ 0 0.

g
G A
�rl rl 01
u q.~ ~r
1.1 .{~i � ~ M
0 Ocn�a)
a! �rl o
C~1 U ~ p
F+ G1 O
fn ia 4-i ~
a) 000 ~
O~
41 1.~ r-1 ~I
I O ~
i'Cd.' `;nwo
a) 0 rnoo ~o
CW7b4rn 3~
0 p r-I O
7+ O O I 4-+
~ cd ~ ~O 00
'C 41 (A ON %.D
Q) A ~ ~
O cU Cl v1
41 +1 D 0 %o
.a rn
~ G c~d cn ~
O
o H qrlW
~ 0 Q
U ~ rj)
g ~ a
~ a) r-4
> i ~ o ~
$4 N -rl r-I
v o w ,-1 0
~ c ~d WC N ~
~d N ru1~ f-1 ~
p U
Rz
a p +j
Hv~~
~m U)
~ G~! O ~
Gqzo
L)

ia
z w 0
i
rn
~ornoo
G 41 r- G
1+� ~�d
4-I U
A d
FOR OFFICIAL USE ONLY
~
w
~
~
N
r'I ~"I 'da
H J 4"+
-I
4. 0 ticd
r~
41
tn i-.
Gl t-i O
.A cd 00
Z Q) o.%
H D+ ~-I
~ F+ O~
0 0 r-
4H O%
v j
0
41 p ~
~ C cd
~ W A
cd c~o q ~
$4 Ctl V
'U 0 0 P4
$4 cn w ~
C7 ~ Rt I
o a). w
41 a~ F+ M
N U'
aua
~
y
=

a0~N00
U"~
viuo uo
1 a
n
~o
N
cD I GV
I n 1 O
N
z
( ( ~ ~
o ,n ir~

co IM ~ O,~ cV
C,
~
1
~

141
30
o..,j r-4

rn y
fA C ~
N U
0 ca
f~1 N
5 O Gl
N ji 4w-i
14
1~.~ 3+ A
coo~
P "4 Cd
00 ~
H C14.~C
ri l1
Gv
3 ro ~
a A ai
b ~
co .z
u 4+
~ r. r.
w a~i a
~ aa
A v] rl
FOR OFFICIAL USE ONLY
G I
~
~
O
~~r
ve0 `t~ i0
~
NCVCV^
z
'O
o
�0~8n`�
0
00
z
Lo'a 'q
cn
~
N
~
�
C
V
N
o~
~
F+
O
~
~ to u')
a~aocvo--
5
CV N
Q'
w
~
Lq
oMorno
~
cv
M
b
~
CD
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
N,rM o 4 a \ _ `
i ;
'ti 3 ~
4 t 2~ ~
70 o a
i
~
~
.

so~o -s" " o eF�c
Fig. 1. Deviations of inean temperatures obtained using Pitot tubes and gas-filled
splieres from grenade data during sumtner months, 1) data from Pitot tubes for Aug-
ust and grenade data for June-August; 2) data from Pitot tubes for August and gren-
ades for August; 3), data from Australian spheres for December-February and grenades
for June-August; 4) data for American spheres and grenades for June-August.
The mean long-term temperatures measured with American spheres in the subtropics in
the upper mesosphere were lower than the temperature measurea using grenades (see
figure). Some of the differences, however, must be attributed to the diurnal varia-
tion of temperature because the spheres were launched for the most part at 0500-
1200 hours (LT) whereas the grenades were launched more uniformly in the course of
the day. An influence of the 11-year cyclicity of temperature in the stratomeso-
sphere is also possible [3, S, 13].
Grenades (spheres) and Datasondes (Arcasondes). The differences between summer tem-
peratures according to grenade measurements and data from Datasonde and Arcasonde
probes for North America were important at the level 60 km and above (Table 3). Dur-
ing the winter and transitional seasons the "grenade-Datasonde" differences for 38
and 59�N were somewhat greater: -1 --3�C at the level 55 km, -3 --6�C at 60 km
and -8 --9�C at 65 km. The influence of the diurnal variation and the possible 11-
year cycle on the accuracy of the comparisons in this case should not be great.
According to data from about 200 paired launchings of Datasondes and spheres in the
United States during 1971-1978 j18J at latitudes 22, 28 and 34�Pd the spheres give
lower temperature values than the Datasondes (with the correction [11]): on the
average by 4�C at the level 45 km, 5.5�C at 50 km, 3�C at 55 km, 0.5�C at 60 km and
(with a reduction of the number of patrs to 35) 4�C at 65 km. These results agree
with the "grenade-sphere" differences.
142
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000400034465-3
FOR OFFICIAL USE ONLY
Grenades (Pitot tubes, spheres) and M-100B rockets. The differences between the
temperature values es:timated on tfie basia of grenade measurements (also using
Pitot tubes and Australian spheres) [12, 15, 16] for the latitudes of the Soviet
stations and determined using the M-100B ("316") rocket are maximum for 80�N
(Table 4). The reason for this may be an inaccuracy in the extrapolation of gren-
ade data to 80�N, tfie small volume of data, etc. At the 80-km level the values of
the differences are varia5le.
When using the modernized "316M" nosecone with the M-100B rocket (instead of the
"361") the discrepancies with grenade data are reduced, but their values are pre-
liminary.
M-lOQB and MMIt-06 rockets. Table 5 shows that the temperature values according to
rIIKR-06 data in the lower mesosphere are somewhat higher than according to M-100B
("3619") data by approximately 1�C at altitudes 45-50 km and 6�C at 55 km. With
the results given in Table 4 we find that the temperatures measured with the MMR-06
rocket and with use of grenades are in agreement with one another at altitudes 50-55
km.
Summary. Our analysis systematizes the discrepancies in long-term data from rocket
measurements of temperature in the mesosphere carried out at a number of stations
using different methods and apparatus. This makes possible the reconciled use of
data from different sources. Hocaever, there is no unanimous opinion concerninc, the
choice of a reference method because investigations of the errors of individual
methods have not been completed. In the formulation of standard and reference
models of the mesosphere it has become the practice to use data on temperature ob-
tained by the grenade method as the fundamental data [2, 6, 17]. Accordingly, in
the formulation of new mQdels a continuity is best ensured by continuing to employ
data obtained by the grenade method as the reference data for the mesosphere (Table
6). Data obrained with the MMR-06 rocket, Australian spheres (in the 1970's) and
Pitot tubes on the average agree with grenade data. In the case of the 14-100 rocket
reconciling corrections in Table 6 are considerably smaller for data obtained at the
present time using the "361M" nosecone than for archival data (prior to 1978) ac-
cumulated using the "361" nosecone.
We note that grenade observations in general are complex and expensive and for all
practical purposes are not being made at the present time. In choosing a method as
a reference (for example, M-100 ("361-M") data) the values of the reconciling cor-
rections vary in accordance with the differences between the temperatures obtained
using this and the grenade method.

143
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R440400030065-3
FOR OFF[CIAL USE ONLY
Table 5
Dif�erences Between Atmospheric Temperature Values (�C) Measured Using MMR-06 Rocket
and M-100B ("361M") Rocket
Period
Region
Altitude, km
40
45 50
55
Summer 1979-1980
Middle latitudes
0.5
1.5 2.5
5
May 1979
Equator
-3.5
0 -0.5
8
February 1980
Equator
1.5
2.5 1.5
7
Table 6
Reconciling Empirical Increments (�C) to Long-Term Temperature Values Obtained
in Former Years by Various Rocket Measurement Methods (Using Grenade
Measurements as Reference Value$)
Alt
km
M-1006
Datasond
Spheri
Ltude,
"361"
vith cor;
th (US A)
without
(sum-
rection
correc- rrec
r
)
mer
[11]
tion on
tF
1
75
I
-
6
0
70
9
21
5
8
-
-7
-
-
-
0
6,5
60
24
21
10
?
-5
-3
-1 b
-3
-2
-2
55
50
12
g
6
4
-1
I ~
-5
-1
5
45
2
1
0
-3
-1
4
.
BIBLIOGRAPHY
l. Avdeyev, V. N., Lysenko, Ye. V. and Chernova, G. G., "Aerodynamic Error in Tem-
perature Measurements in the Atmosphere With a Rocket Thermometer," TP.UDY TsAO
(Transactions of the Central Aerological Observatory), No 144, 1980.
2. GOST 4401-73. STANDARTNAYA ATMOSFERA (State Standard 4401-73. Standard Atmo-
sphere), Moscow, Izd-vo Standartov, 1974.
3. Kokin, G. A., Bugayeva, I. V., Ryazanova, L. A. and Speranskiy, K. Ye., "Cor-
relation Between Stratospheric Processes and Solar Activity," METEOROLOGIYA I
GIDROLOGIYA (Meteorology and Hydrology), No 7, 1977.
- 4. Koshel'kov, Yu. P., "Mean Monthly Temperature in the Southern Hemisphere Meso-
sphere," TRUDY TsAO, No 140, 1978.
5. Angell, J. K. and Korshover, J., "Recent Rocketsonde-Derived Temperature Vari-
ations in the [destern tiemisphere," J. ATMOS. SCI., Vol 35, 1978.
144
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
6. COSPAR INTERNATIONAL REFERENCE ATMOSPHERE 1972 CIRA 1972, Berlin, Akademie-
Verlag, 1972.
7. Exemenary, F. R. C., "On the Magnitude and Uncertainties of Correctiona to Arc-
asonde 1 A Temperatures," J. APPL. METEOROL., Vol 11, No 4, 1972.
8. Finger, F. G., et al., "Compatability of Meteorological Rocketsonde Data As
Indicated by International Comparison Tests," J. ATMOS. SCI., Vol 32, No 9,
1975.
9. Hoxit, L. R. and Henry, R. 24., "Diurnal and Annual Temperature Variations in
the 30-60 km Region as Indicated by Statistical Analysis of Rocketsonde Tem-
perature Data," J. ATMOS. SCI., Vol 30, No 5, 1973.
10. Ivanovsky, A. I., et al., "Preliminary Results of the Intercomparison Test
of US and USSR Meteorological Systems at the Wallops Island in August 1977,"
Preprint to XXI COSPAR, Innsbruck, 1978.
11. Krumins, M. V. and Lyons, W. C., "Corrections for the Upper Stratosphere
Temperatures Ustng a Thin Film Loop Mount," MOLTR 72-152. Naval Ordnance Lab-
oratory, Md., 1972. .
12. Pearson, P. H., TECHNICAL NOTES, Weapons Res. Establish., Salisbary, Austral-
ia, 1970-1975.
13. Quiroz, R. S., "Stratospheric Temperatures During Solar Cycle 20," JGR, Vol
84, 1979.
14. Schmidlin, F. ,T., "Diurnal Tidal Analysis in tha Equatorial Stratosphere and
Mesosphere," Preprint to COSPAR XIX, Philadelphia, 1976.
15. Smith, W. S., et al., "Temperature, Pressure, Density and Wind Measurements
in the Upper Stratosphere and Mesosphere," NASA TECHN. REPORTS R-211, R-245,
R-263, R-288, R-316, R-340, R-360, R-391, 1964-1972, Wash. D. C.
16. Theon, J. S., et al., "Tfie Mean Observed Meteorological Structure and Circul-
ation of the Stratosphere," NASA TR R-375, Wash. D.C., 1972.
17. US STANDARD ATMOSPHERE, 1976, NOAA, NASA, USAF, Wash., D.C., 1976.
18. World Data Center A, METEOROLOGY. HIGH ALTITUDE METEOROLOGICAL DATA, Asheville,
N. C., 1965-1977.
145
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-40850R040400034065-3
FOR OFFiCIAL USE ONLY
UDC 551.508.29
RADIO DEVICE OF A SYSTEM FOR THERMAL SOLTPdDING OF THE ATMOSPHERE BY THE
RADIOACOUSTIC SOUNDING METHOD
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 81 pp 120-123
[Article by V. M. Bovsheverov, M. A. Kallistratova and L. V. Knyazev, candidates of
physical and mathematical sciences, A. G. Gorelik, professor, and M. Yu. Yegorov]
[Text] Abstract: Radioacoustic sounding (RAS) is a
promising method for prolonged measurement
of the temperature profile in the atmospheric
houndary layer. The authors examine the various
aspects of design of the receiving and trans-
mitting radio elements of a radioacoustic sound-
ing system. The optimum specifications for these
devices are set forth. An experimentally deter-
mined temperature profile is given and this is
compared with radiosonde results.
Radioacoustic sounding (RAS) is an effective method for long-term monitoring of
the temperature profile T in the atmospheric boundary layer at a real time scale.
The method is based on measurement of the speed of sound vs in the atmosphere,
vs = 20.05 ,
which is determined from the Doppler frequency shift fD of an electromagnetic sig-
nal reflected from an acoustic wave. A distinguishing characteristic of the RAS
system is that it effectively functions near the synchronism 2~ 5 ='1 [5, 6], when
the electromagnetic energy reflected from each of the acoustic wave fronts arrives
in the receiver antenna in phase. In this case
fD = 2vs/~, _ vs/ 'Xs = fS,
where ~ and a S are the wavelengths of the electromagnetic signal and sound.
A highly important prohlem in developing the radio device for the RAS system is the
choice of a working wavelength. The systems described in the literature have used
either very short electromagnetic waves (about 3 cm, 10 cm [2, 3]) or very long
waves (0.7 m, 8 m[4, S]). In the first case it is impossible to obtain a reflec-
tion with great ranges since the acoustic wave rapidly attenuates; in the second
146
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3
FOR OFFICIAL USE ONLY
r~Y}
the acoustic wave almost does not attenuate during propagation, hut the realiza-
tion of such a system requires the construction of unwieldy narrow-beam radio and
acoustic antennas.
r--------------~
Transmitter (
t ~
L--~------ --J
~
FI iF h Receiver ~
! e ~p > BF i
~ I
Al. Da EFA P' M IF : i ~
8
I ~
i I

PA Sw 0~--~
I ~
I G,AP ~
_ Bc,Qitar,i.t__PYS1ejL
Fig. 1. Block diagram of RAS system radio complex. CC) capacitive coupling; PA and
PA3) power amplifiers for transmitter and acoustic system; MO and M03) master os-
cillators; PI) phase inverter; HFA) high-frequency amplifier; DC) directional coup-
ling; P) preselector; BMsig and BMPh) balance mixers of signal and phase channels
of receiver; Het) heterodyne; IrAsi , IFA ph) intermediate frequency amplifiers; PD)
phase detector; BF) band filter; SFI synchronous filter; M) multiplier; LFF) 1ow-
frequency filter; Sw) switch; GAP) generator of acoustic pulses.
The developed RAS system operates at a wavelength of 30 cm. This is a compromise
in the effort to obtain a significant effective range of the RAS system with rela-
tively small dimensions of the antenna system.
According to [5], for the reception of a signal from a kilometer range with an acous-
tic power of about 5 W and with an antenna directional diagram of about 10� the ratio
of the power of the electromagnetic signal Prec reflected from the acoustic pulse to
the transmitter power Pt (radar potential) should be not less than 140 db. This does
not take into accc;:nt the attenuation of the signal caused by turbulence and the
shift of the acoustic pulse relative to the radio antenna ray by the horizontal wind
component. The attenuation of the reflected signal caused by the enumerated factors
147
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000400030065-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000400030065-3
cOR OFFICIAL USE ONLY
is 50 db or more. The required potential is therefore -190 db.
Fig. 2. Output voltages of circuit for synchronous shift of Doppler frequency and
uand filter. The scale 0.2 sec/cm corresponds to a range scale 70 m/cm.
,y Mr
~ ~ev 0800 hours
7y yOM!/N 0740 hours
\ t 0