JPRS ID: 10564 USSR REPORT METEORLOGY AND HYDROLOGY NO. 3, MARCH 1982

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000500070010-8
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 53
• Document Creation Date: November 1, 2016
• Sequence Number: 10
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFFIC[AL USE ONLY
JPRS L/ 10564
4 June 1982
USSR Report
. METEOROLOGY AND HYDROLOGY
No. 3, March 1982
.
_ Fg~$ FOREIGN BROADCAST INFORMATION SERVICE
. ~
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/49: CIA-RDP82-00850R040500070010-8
NOTE
JPRS publications contain information primarily from foreign
newspapers, periodicals and books, but also from news agency
transmissions and broadcasts. Materials from foreign-language
sources are translated; those from English-language sources
are transcribed or reprinted, with the original phrasing and
other characteristics retained.
Headlir.es, 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
processedo 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 preceded 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-
ciPS, views or at.titudes of the U.S. Government.
COPYRIGHT LAWS AND REGULATIONS 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-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2447102/09: CIA-RDP82-44850R444544474414-8
JPRS'L/10564
4 June 1982
USSR REPORT
METEOROLQGY AND HYDROLOGY
No. 3, Marcr 1982
Translation of the Russian-'_anguage monthly journal METEOROLOGIYA I
- GIDROLOGIYA published in Moscow by Gidrometeoizdat.
CONTENTS
~Variability of Temperature and Circulation Regimes in Northern Hemisphere
Atmosptiere 1
kInfluence of Temperature Stratification on Adiabatic Fluctuations.in Polytropic
Atmosphere.................................................................... 2
Major Tempe'cature and Precipitation A;nomalies Over European USIOR, Western
Siberia a;nd Kazakhstan 3
*Interaction Between SuUclo:::?.,and Cloud Layers in Tropical Cyclones and'
_ Intertropical Convergence Zone.......................................... 15
*Investigation of Cloud Cover Dynamics in Tropical Zone Using Z`wo-Dimensional
Spectral Analysis 16
*Models of One-Dimensional and Joint Distributions of Non-N egative Random Jalues. 17
_ 'tissipation of Kinetic Energy in Cyclone Over Ocean 18
- *Prediction of Water Salinity in Gulf of Riga Bottom Horizon 19
Status and Prospec~s of Numerical Hydrodynamic Investigations of Fluctuations of
LevEls in Arctic 5eas......................................................... 20
*Description of River Runoff at Several Points by Multivariate Canonical
T:xpansion Afethod 30
_ *Use of Surfa,ce Generaters of Ice-Forming Aerosols in Work for Artificially
Augmenting Precipitation in Mountainous Regions 31
~
_ Denotes items which have been abstracted. ,
~
- a- [III - USSR - 33 S&T FOUO]
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00854R000540070014-8
FOR OFFICIAL USE UN1,Y
Lvaluation of State of Agricultural Crops From Satellite Data 32
~Evaluation of Spectral Resolution of Optical Apparatus for Measuring Minor
Gas Components in Atmosphere 42
*Possibility of Determining Atmospheric Temperature Profile by Acoustic
Sounding 43
*Inadvertent Influence of Ma3or Cities and Industrial Centers on Precipitation.. 44
*Review of Monograph 'Gidrologicheskiye protsessy i ikh rol' v formirovanii
k-ictiestva vody' ('Hydrological Processes and Their Role in Forming Water
c)ual.it,y' by V. A. Znamenskiy, Le, ingrad, Gidrometeoizdat, 1981, 248 Pages.. 45
*Conferences, Meetings, Seminars................................................ 46
*Notes From Abroad....~ 47
*Obituary of Yekaterina Nikitichna Blinova (1506-1981) 48
- *Obituary of Sergey Ivanovich Sivkov (1901-1981) 49
* Denotes items which have been abstracted.
- b -
FOIt OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000500074410-8
FOR OFFICIAL USE ORLY
'1TLC 551.524.3+551.513(215-17)
VARIABILITY OF TEMPERATURE AND CIRCULATION REGIMES IN NORTHERN HEMISPHERE
ATMOSPHERE
Moscow METEOROLOGIYA I GIDROLOGIYA in Russiazi No 3, Mar 82 (manuscript received
1,7 Jun 81) pp 8-20
_ [Article by G. V. Gruza, professor, L. K. Kleshchenko, candidate of geographical
sciences, and T. P. Timofeyeva, All-Union Scientific Research Institute of Hydro-
mereorological Information-World Data Center]
[Abstract] A joint analysis of the long-term changes in a number of key character-
istics of state of the atmosphere was r,ade. The spatial and temporal.changes in
_ global and local parameters of the tem?erature ancl circulation regimes of the
J northern hemisphere were studied using data for the period 1949-1979 on the basis
of the mean monthly fields of temperature and pressure at sea level and the geo-
- potential field at the 500 gPa surface as interpolated at the points of inter-
section of. a geographic grid with a 10� interval along the parallels an3 a 5� in-
terval along the meridians. The investigation was limited to the latitude zone
- 30-80�N. In applying the empirical-statistical method it was not only the initial
~ fields which were analyzed, but also the fields of their anomalies, t},.e f~elds
of zonal and meridional components of the horizontal gradient anc the modulus of
the vector gradient. This analysis indicated ttrat there is 4 comglete noncorres-
pondence between regional and global climatic changes, revealing an exceedingly
complex pattern of climatic changes resulting from interaction of factors of
global and local scales acting in unison over extensive regions of the earth only
undei, exceptional conditions. This in-depth study clearly suggests that the long-
- term changes in the temperature at.d circulation regimes are characterized by a
- great diversity, have a complex and unstable character. During the 30-year per.iod
studied there are no linear trends in these changes. However, in the 1970's it
is possible to see a transition from cooling to warming in mean hemisphere tem-
perature and some increase in anomalousness of processes at the earth's surface
manifested in an increase in the gradients of temperature and surface pressure
anomalies. Figures 4, tables 3�, references 22: 14 Russian, 8 Western.
1
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFFICIAL USE ONLY
UDC 551.509.313+551.524
INFLU1:NCE OF TEMPERATURE STRATIFICATION ON ADIABATIC FLUCTUATIONS IN POLYTROPIC
ATMOSPHFRE
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, yiar 82 (manuscript received
12 Jun 81) pp 21-29
jArticla by 0. K. Gorbunova and V. M. Kadyshnikov, candidate of physical and mathQ-
matical sciences, USSR Hydrometeorological Scientific Researc'n Center]
[Abstract] This is essentially a continuation of earlier work by one of the auth-
ors (V. M. Kadyshnikov, "Small Oscillations of a Polytropic Atmosphere and Filter-
� ing Role o.f the Hydrostatic Approximation," METEOROLOGIYA I GIDROLOGIYA, No 11,
1979). Now, using the solution obtained in that article, a study is made of the
dependence of the spectrum or oscillations in its different parts on Y, which is
' the principal characteristic of a polytropic atmosphere Cy is the temperature
~ gradient af the main state). The dispersion expression determines two families
of frequencies: S+ gives acoustic waves and S_ gives gravitational waves. The
~ dispersion expression contains two key variable parameters, a and )s , and their
behavior is examined in detail. In three cases, A9b, A,~,oc., OC :~4 0, 2)
(G < 3) 06 > 1, oC9j b, aCusing asymptotic formulas for confluent hypergeo-
- metric functions, a solution of the dispersion expression can be obtained in ex-
p].icit form. Case 1, corresponding to long waves, gives a solution for both acous-
zic and gravitational cases; case 2, also corresponding to long waves, has a so-
-luticsn anly for gravitational waves; case 3, corresponding to short waves, has a
solution for both types of waves. With this taken into accoun'c, it was then pos-
sible to study the frequency sgectrum of the equations of hydrothermodynamics lin-
earized relative to a state of rest with a polytropic temperature distributic..
After investigating the dependence of frequencies and group velocities on the ver-
tical temperature gradient of this state and on the vertical variability of dis-
- turbances it was posslble to demonstrate that with an increase in stability and
- a decrease in variability there is a decrease of acoustic frequencies and an in-
; crease in gravitational frequencies. It -lso becomes clear how the velocity of
- propagation of both types of atmospheyic disturbances changes. Figures 2; refer-
ences: 4 Russian. 2
FOR OFF'ICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00854R000540070014-8
_ FOR OFFIC[AI. USE ONLY
LTDC 551.509.314+551.524:551.577(47)(571.1)(574)
MAJOR TEMPERATURE AND PRECIPITATION ANOMALIES OVER EUROPEAN USSR, WESTERN
SIBERIA AND KAZAKHSTAN
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 (manuscript received
18 Jun 81) pp 30-38
[Article by 0. V. Batyreva, candidate of physical and mathematical sciences, .
- and L. Ye. Lukiyanova, candidate of geographical sciences, USSR Hydrometeoro-
- logical Scientific Research Center]
. [Text] Abstract: The article gives a classification
- of g.rowing seasons on the basis of the ratio
betweet;-large positive and negative anomalies.
The value 1.2 CS was adopted as the criterion
for a major anomaly. It was possible to define
four classes of growing seasons and an attempt
. is made to predict these classes by means of
linear discriminant analysis. Evaluations of
- the probable success of the classification
surpass the evaluations of a random classifica-
' � tion. The mean probable success (guaranteed
probability) is 0.70 (for a random classifica-
tion 0.56), whereas the probability of a pre-
cise prediction of the class is 0.48 (for a
random classification 0.25). The hypothesis
of nondependence of the actual and prognostic
classes is refuted using the x2 test with a 5%
significance level.
- The problem of long-range forecasting of major temperature and precipitation
anomalies is one of the most important and difficult problems in modern meteor-
ology. It has been dealt with in a great number of studies [1, 2, 13] in which
the authors develop primarily synoptic methods for the long-range forecasting
of large anomalies. As criter ia for discriminating major anomalies use is made
either of.the excess of the cY value over a large part of the territory or the
excess of some absolute value of the anomaly (for example, for temperature,
etc.). Zn other studies for defining the types of anomalies use is made of the
anoma].ousness coefficient K proposed by N. A. Bagrov; K+ and K_ are considered
separately for positive and negative anomalies. The anomalousness coefficient
incl.udes all the values of the anomalies in the considered territory, both
3
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000500070010-8
- FOR OFEICIAL USE ONLY
large and small. However, in studying large anomalies it is possible that anom-
alies less than some critical value will not be taken into account at all. In
a study by Aristova [3] the anomalousness criterion used was the fraction of
the considered area for which the anomaly exceeds a stipulated level. Unfor-
tunately, only whole values multiples of d( 0, 2d , 3 (Y, -or, -2d ,-3 d etc. )
are considered as such levels and intermediate levels are not considered in
the corresponding intervals.
In this article as a criterion for aefining a ma3or anomaly we used the value
1.2c7. In the normal distribution oE a meteorological element the probability
that this value will be exceeded is 0.12 for an anomaly.of each siEn (a total
of 0.24), that is, approximately about 1/4 of all the cases. Then for a char-
acteristic of anomalousness of the territory it is possible to take the index
v, that is, the relative number of stations.,for which the anomaly exceeds the ,
tevel l:20*. In this way it is also possible to characterize the anomalousness
of a stipulated period of time (number of months for which the anomaly A 3
1.2cr). Like IC+ and K_ it is necessary to compute v+ and v separately for pos-
itive and negative anomalies. The v value in part corresponds to the mean num-
her oE "surges" beyond a stipulated-level. However, in the theory of surges
a"surge" is an event in which the random value for tiie first time intersects
a stipulatad level and therefore the frequency and duration of the surge dif-
fer. The application of the theory of sEirges in meteorology is discussed, for
examPl.e, in [8, 9] aud elsewhere. The proposed v value includes all cases of
exceeding of the level 1.2 d, regardless of the preceding and adjacent values
of the meteorological element, that is, both the frequsncy of the surge and
its duration are taken into account.
In the analysis of the anom.3lousness of precipitation and temperature anomal-
ies we used mean monthly data frnm 1901 through 1978 for temperature for 20
stations located in the territory of the main agricultural regions of the USSR
(European USSR, Western Siberia, Kazakhstan), for precipitation [10] the av-
eraged values for 20 economic regions. The computed temperature and precipita-
tion anomalies were normalized for the Cf value and maps of major anomalies ex-
_ ceed'ng 1.2 d were constructed. The maps were analyzed from month to month in
their natural sequence, which made it possible to prepare a catalogue of the
most signlficant anomalies.
It should be noted that in most studies [13] an analysis was made of large anom-
~ alies for individual months with a discreteness interval of the data of one
. year.
An analysis of monthly maps of anomalies in their natural sequence made it pos-
sible to detect the characteristics of individual anomalies, couomon for both
precipitation and temperature. As a rule, on each map somewhere there is an
anoma'ly with >,1.2a. Among the 936 maps of anomalies (78 years x 12 months)
there were onl,y 116 on'which not at a single station did the temperature anom-
;ily not exceed 1.2 d and only 51 such maps of precipitation anotnalies where
there was not an anomaly with > 1.2U and this was true of all economic regions.
However, anomalies with > 1.20' rarely persist at one and the same point or in
one and the same region for m6re than two months. The greatest duration was
five months for precipitation and six far temperature. As a rule, the centers
4
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00854R000540070014-8
- oE the anomalies appearing initially at one point or tn one region during the
subsequent months move into adjacent regions, frequently increasing in area
and withdrawing considerable distances from the initial center. The total dura-
tion of the large anomalies of the same sign, with movement taken into account,
can attain 7-8 months and even a whole year. Usually the larger the anomaly i.s
in area, the greater is the area occupied by centers of the same sign and the
- greater is the duration of this anomaly.
It is asserted in [1, 21 that a large value of the anomalousness index corres-
Ponds to a predominance of an anomaly of one sign and large K.F and K_ values
nre r:irely noted simultaneously. We feel that this is not entirely so. There
were 50 such precipitation anomaly maps on which there were not less than
three centers of different signs simultaneously. A typical example is the sum-
mer of 1972 when there were very large precipitation anomalies of different
signs simultaneously in different parts of the considered territory: a strong
drought in the northern and central regions of the European USSR and excess
moistening in the western regions. During the warm months there is aci inverse
r.el.ationship between the temperature and precipitation anomalies, whereas in
the cold months there is a direct relationship. The number of major positive
precipitation anomalies in all months of the year is greater than the number
of negative anoxnalies, which corresponds to data on the asymmetry of precipit-
ation [4]. In summer positive temperature anomalies are encountered more fre-
quentLy than negative anomalies, and in winter, vice versa, which also corres-
ponds to the asymmetry values [5]. The mean frequency of recurrence of major
temperature anomalies is 20-22%.; for precipitation it is 18-20%. An increase
in the total number of major anomalies has been clearly noted during the last
decade (since 1970).
The maximum frequency of recurrence of major temperature anomalies, 25%, was
observed from 1931 through 1940 and from 1971 through 1979. The maximum fre=
quency of recurrence of major precipitation anomalies is 23% and falls in the
1960's and 1970's. This conclusion agrees well with the results in [6, 7, 14,
]_S], where the authors noted an appreciable increase in anomalousness in the
_ 1970's both for the territory of the USSR and for the entire northern hemi-
spher.e. The spatial differences in the frequency of recurrence of major anom-
alies are not very great (from 18 to 24%); the anomalousness of the northern
regions is somewhat greater than in the southern regions. There is a very small
number of major negative precipitation anomalies for the 19th region Trans-
caucasia (11%). The frequency of recurrence of values > 1.2cY on the average
Is somewhat less than for an independent normal sample (24%), which is attrib-
utahle to coherence in time and deviations from normality. The frequency of re-
currence of major anomalies of different duration corresponds well to the
evaluations of the probability of a surge beyond the level 1.2 d for a two-
di.mensional normal sequence, which, according to [8, 9], can be evaluated as
_ ().08-0.09. Tlie differences in the character of the anomalousness between indi-
vidual. months of the-year are also small (from 18 to 23%) and correspond well
- to the asymmetry and excess values. Since an analysis oi individual anomalies
iti diff.icult due to the great differences in their duration and geographic po-
sition, firsC we analyzed the generalized characteristics of major anomalies as
a who]e for the entire territory and for a long period of time (year and grow-
ing season).
5
FOIt OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFNICIAL USE UNLY
52

~
~
f6 � i
~
r
. ~
w
~
~ ; ;
~ ~i ~ i' r i i
VAJ
i~
{B� +
.w*f' .
4Q
f2
'
281
2y
10
; !2
r 9
:I !B /1 i
: �~P ~/f
y ' : � r s
~~I7
.
70
/ ty. ~
6J 71 ~1
� ~4� 'S 75
~!~X ~4.JB � .J�~
� ~i Jb
/ ~ : �J4 ~'f
1t! L�-.._1..._._~__~_.i~..a.-� .
f90.s f91.f l.r: f iSJS ~5~~ f9Ss M5 07J ,
9p ~o ?p rf�)
Y
Fig. 1. Ten-year� moving averages v+, Fig. 2. Classification of growing sea-
v_, v'q of precipitation, sons on basis of character of anotaalous-
ness of precipitation.
In order to define the climatic characteristics the values v+, v_ and also
their sums vy for temperature and precipitation were summed both for the en-
' tire year and for the growing season (from April through September) and we
constructed graphs of the temporal variation of these indices and also their
five- and ten-year moving averages. Year-to-year variations are 15-20% of the
� annual sums v+, v_ and vp . There is a general increase in the total anomal-
ousness of precipitation in the 1930's up to 1936, gfter which there is a
decrease to a minimum in 1942. A generai increase in the anomalousness of
temperature and precipitation in the 1970�s with sharp variations for in.di-
vidual years was characteristic. The moving averages graphs make it possible
to define quasiperiodic variutions with periods of 15-20 years. As a rule,
the v+, v_ curves vary in antiphase, although there are periods of a simultan-
eous increase in anomalousness of both signs. For example, all the mouing five-
and ten-year periods, including the 1970's, give an appreciable increase in
the annual and semiannual sums v+, v_ and v% simultaneously (the total anom-
alousness of precipitation during recent years excee.ds by approximately 20%
the values encountered earlier). For temperature this characteristic is ex-
pressed to a lesser degree. Examples of the graphs are shown in Fig. 1.
Since major temperature and precipitation anomalies are most important for
agriculture when they occur in the growing season, the breakdown of all the
years into classes on the basis of`the character of the anomalousness was
_ carried out using the sums v.}., v_ and vS for the entire growing season. A
classification of all the growing seasons from 1901 through 1978 was carried
out on the basis of the ratio of the total number of ma3or positive and nega-
tive anomalies. The values of the sums v+ and v_ for the growing season for
all years from 1901 through 1978 were regarded as the coordinates of a point
on a plane in a Cartesian coordinate system and were plotted on the graph.
6
FOR OFFICIAL USE ONLY
J~
�
~
' n ,
J~ � ~
b �
.
ll 7: ~y lY J6 ~ b
� � � � 7f /
.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R040500070010-8
FOR OFFICIAL USE ONLY
7'tie number of the year in the 20th century to which the corresponding values
related was plctted near each point.
~ T}le d istribution of points for precipitation is shown in Fig. 2. The separa-
tion of growing seasons on the basis of temperature was carried out in a sim-
ilar way, with the single difference that the E v+ and S v_ axes ehanged
p1y v+ entered into the first class for precipitation, whereas for temper-
ature the anomalously hot years when Tv+ > 2 v_. In the fourth class the
moist periods with 'Z v+ I v_ enter for precipitation (and accordingly, for
temperature, the cold years when T v_ >I v+). The "middle" classes (second and
third) are characterized by approximately equal values of the sums v+ and v_,
wiiose difference, as a rule,does not exceed the standard deviation of the
corresponding value. The second class includes years with small 2 v+ and ? v_
values (less than the mean values), whereas the third class includes years
with large values (gre6ter than the means). In the first stage the separation
into classes was carried out separately for temperature and precipitation.
The lists of the corresponding classes are given in Table 1.
The cited lists show that the defined classes for precipitation and temperature
partially correspond to one another, that is, about half the growing seasons
fall in one and the same classes. '.~he best correspondence is attained for the
first class (dry and hot) and a worse correspondence for the fourth class.
In actuality, as is well known, hot and dry years are characterized by a pre-
dominance of anticyclonic weather, whereas major positive anomalies of pre-
_ cipitation in summer are associated with an intensification of cyclonic activ-
ity and the transport of moist air masses from the Atlantic; najor negative
temperature anomalies are associated with the intrusion of cold arctic air
Crom the north. tC the moistening conditions during the growing season are considered, it is
necessary to take into account temperatur.e and moisture anomalies simultaneous-
1y. For example, D. A. Ped' [12] proposed an aridity index which represents the
dtfference between the normalized temperature anomaly and precipitation. All
the droughts known in history were accompanied by major anomalies of both pre-
- cip.itation and temperature (droughts of 1921, 1936, 1972, 1975, etc.). Thus,
in order ta characteriz.e the moistening conditions during the course of the
- growing season it is possible to consider the values w_ = v_ (precipitation)
+ v+ (temperature), w+ (precipitation) + v_ (temperature). These values in
turn can be divided into classes on the basis of the same principles as
7
- FOPt OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000500074414-8
FOR OFFI('IA1, IiSN: ()NI.Y
temperature and precipitation separately. The cited list of classes to a
high degree corresponds to the classes of temperature and precipitation.
Table 1
List of Classes of Anomalousness of Temperature, Precipitation and the
Ceneralized Index w(the Table Columns Give the Years of the 20th Century. For
Example, 01 Should be Read 1901, 45 194.5, etc.)
' 1 I
II I
lll I
IV I
I I
ll I
1I1 I
IV I
9 I
ll I
lll
I IV
Precipitation
Temnerature
General
ized
wind index
01
02
0.3
15
01
03
06
02
01
05
03
.02
04
(),i
p(i
22
20
U.`'~
09
04
OG
07
13
04
07
US
12
21i
11
10
15
07
C18
lU
23
09
09
11
13
78
24
22
17
OS
20
18
29
11
- IU
19
14
31
31
Yti
23
11
21
22
31
12
- 17
2:3
16
32
32
27
29
12
24
25
34
14
- 18
1�1
27
:i i
36
30
34
13
36
27
46
15
20
2: i
3.1
4 3
37
33
39
14
37
30
49
16
?I
3()
:tti
53
38
35
49
16
38
32 '
50
17
29
35
46
54
40
4:.;
50
13
31)
35
52
19
36
40
�17
.56
46
�41
52
19
40
43
68
26
37
41
49
58
43
-59
54
26
48
94
73
23
.
;39
42
53
GU
51
61
62
28
51
53
77
33
51
43
52
Gl
53
64,
fi:i
41
55
54
4l
55
.14
57
V:)
55
VS
7a3
42
57
159
42
63
43
72
70
57
70
76
4.i
(ifi
GI
45
li3
59
73
77
63
71
77
47
67
62
47
71
li(1
74
(iG
74
56
71
64
56
75
62
76
67
J$
72
63
58
611
78
72
(1o
7)
sU
� s;>
75
69
74
69
uu
7s
76
70
67
Table 2
Mean dik
and Maximum dik(maX) Distances Between Classes
Init-.al montr
\ I xI
I :CII l I I
]I
I III
Precipitation
- Jik
0,77 0,62
0,84 0,85
1,04
1,03
I
1 ,fi5 I 1,13
I 1,24 1,60 (
1,73
1,35
-
Te:nperature
I1;R I
1,33 0,92
1,23 0,62
0,87
0,83
r!lkt_11111%1
1,59 I. 1,32
I 1,56 0,82 I
1,39
! 1,36
8
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500074414-8
FOR OFFICIAL USE ONLY
CY)
. ~
.--I
.n
cd
H
H
H
H
I
H
H
DC
1
~
~
N
r{
H
U
H
O
I
~
~
H
bp
.
H
~
H
r-I
I
O
~
41
O
~
~
~

G
W
ro
H
t11
H
N
N
G'+
~
H
O
F"{
r-I
ro
O
H
~
d
W
H
O
~
N
U1
W
~
~
DC
U
4a
O
~
~
~
N
U
U
~
~
N
r--I
~
~

O
p
W
f'^
. 1~
~o
op
O
O
M 00
~ un
O O
~--I I~
n ~
O O
N
u'1
O O
N
H
~
O
~t
cd
I
o
O
N
a
~
v
o
O
O
O
0
OG
O~
O
O
O
~
%O
n
O
M
~
O
O
O
M
~-I
~
-:T
n
O
O
O
00
00
0
0
0
O~
~G
r-I
O
O
O
I O
A
cn
a
~
N
44 b(1
4I
U
O 0
U
U

fn
1~ r~
r1 cti
N
e--I 4-~
4)
r-i
-rl N
,-i
,a
,.a N N
,O
m
cv ~n ro
4
n ,a ~
a
O
O D U
H

~
a
a
~
a
0
~
I
~
c+1
O
v1
~
O
N
~
O
~
~
O
~
~
O
~
u'1
O
I O
N
p. �rl
4-4 CO
O r
U
~ ' r-I
fn 1.~ r-I
-rl R3
rl w
�ri tA
,.o N N
~ co ~n ro
.n ~ ~
O O D U
~4
w a
9
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFFICIAL USE ONLY
Tt should be noted that the defined classes make it possible to describe only
the general character of anomalousness during the entire growing season and
for the region as a whole (region of the European USSR, southern part of
[destern Siberia, Kazakhstan). The classification does not take into account
the location of the centers of positLve and negative anomalies within the re-
- Rion and their change during the course of the growing season. Allowance for
- all these differences would lead to a far greater number of classes, which
would make difficult their prediction. Nevertheless, the classification cor-
responds fairly well to the separation of years obtained by other methods.
All the best known droughts fell in the first class. The only exception is
- the summer of 1972, which with respect tu'precipitation fell in the third
class, not the first, since there were no major negative anoma.lies over the
entire territory: a drought occurred primarily in the northern and central
regions of the European USSR, whereas in the western regions at the same time
- there was a considerable positive precj.pitation anomaly. However, with respect
- to temperature and the w index 1972 fell in the first class.
It is of interest to give a prediction of the defined classes on the basis of
zin analysis of the preceding conditions in the atmosphere. The longest series
are for the temperature field. It was selected as a set of criteria for making
t}ie classification. Since the years entering into each class were known in ad-
vance, we carried out the procedure of "learning with a teacher," that is,
breakdown of criterion space into regions best corresponding to stipulated
- classes. The teaching sample represented the temperature field expanded in
natural orthogonal components:' The first eight normalized coefficients of the
expansion were taken as criteria in the discriminant analysis. Linear dis-
criminant analysis was used on the assumption of a normality of the distribu-
tion of criteria and the equality of the covariation matrices of the classes.
Although these assumptions, strictly spea;cing, are not satisfied, nevertheless,
- as was demonstrated in [10], the advantage of use of quadratic discrimi.nant an-
alysis is not maintained with a changeover to independent material. In this
case, since the breakdown was made into four classes, the samples for evaluat-
ing the covariation inatrices of the four classes were too small. Accordingly,
we used one covariation matrix for all classes. A linear parametric algorithm
was used; the vectors of the mean criteria in space were found for each class
and the corresponding linear discriminant functions were constructed.
' As indicated in [11], the linear discriminant �unetions for normal objects with
equal covariation matrices eorrespond to the classificator for the minimum dis-
tance From the class "standard" (the standards are the mean fields for each
- class). Since the expansion coefficients are statistically independent criter-
= ia, the distances d�k are determined as the sum of the squares of the normal-
ized expansion coef~icients.
Table 2 gives the mean and maximum distances d-k between the class standards.
Tiiis value (the Mahalanobis distance) serves as an indirect evaluation of clas-
sification quality. Table 2 shows that the minimum distances were obtained for
tlte prediction for January. The classes of temperature anomalousness are less
- different in the space of the selected criteria than the precipitation criteria.
'1'fie clistanc.e between the f.irst and fourth classes on the average exceeds the
distance between the other classes (by approximately 0.4).
10
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R040500070010-8
- FOR OFFICIAL iJSE ONLY
TL is of interest ro check the significance of the determined distances be-
tween classes. For this we will assume that the entire criterion space belongs
to one and the same multidimensional normal set with a zero vector of the means
zind a unit covariation matrix. These conditions are well satisfied for the nor-
mal coefficients of expansion in natural components. We will find the critical
distances between the means of the classes which would be obtained if the class-
es were formed randomly. The Mahalanobis distance between the j-th and k-th
classes is
- ) = nl~
i d; h = - (1, , . u,)'$ ~f t"
!_1 a ) ~ cl>
where Lj, � k are the vectors of the mean classes, S is the covariation ma-
trix, Aji, Ak are the mean values for the j-th, k-th classes of coefficients
of thE expansion, djk has the distribution
2 G2 ~t
_ wliere x 2 is the xZ distribution with 8 degrees of freedom. The parameters in
the paregtheses have the dispersion
`';'1 �1 I
n '
where n is the number of the class. Assuming the number of years in the class
to be 20, we find that the dispersion of each term 2/20�1 is approximately
equal to 0.1. Since the dispersion DXk = 2k = 2.8, then the dispersion dJ-k
=0.1�2�8 = 0.16, and-the standard deviation o'(djk) = 0.4. The mean value djk
is approximately equal to
- 202 ( n)i:k 0,1�s-0,8.
(2)
Tabl.e 2 shows that the maximum distance between the classes in many cases ex-
ceeds 10% and even 5% of the critical distribution value x 8(~), equal to
1.33 aiid 1.55 respectively.
- We evaluated the quality of the classification for all the teaching sample
fr.om 1901 through 1978. The evaluation was made in the following way: with
precise falling into a class the evaluation of the forecast was equal to 1;
. when f.alling into the adjacent class 1/2, and in the opposite case 0. A
purely random forecast of the equiprobable classes with such an evaluation
- mzttrix }ias a probable success'(guaranteed probability) of 0.56, whereas the
probaUility of precise falling into a class is 0.25 (each matrix "box" corres-
ponds to a probability of entry equal to 1:16). The classes which we defined
are nonequally probable and therefore the probable success of a random fore-
- cast is dependent both on the natural frequency of recurrence of classes and
on t11e frequency of forecast of each class. We computed evaluations of random
Eorecasts with the actual probabilities taken into account; they do not ex-
ceed 0.58-0.59, but the probability of precise entry into a class for a random
Corecast does not exceed 0.27-0.28.
~ 11
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-40850R000500470010-8
FOR OFFIC'IAL USF. OVI,Y
. Table 3 gives evaluations of the probable success of the discriminant analysis
and also the frequency of precise falling into a class. Table 3 shows rhat the
evaluations everywhere exceed the probable success of the random classification.
- 'PIie mean rrobable success p is equal to 0.70. The probablc success af the cla5s-
es for temperature for January is-somewhaE lower. Falling into a class in
most cases exceeds by almost twice the corresponding number for a random class-
ification and is about half of all the cases. The mean probability of a precise
falling into a class is 0.78. With an increase in the advance time of the fore-
cast the quality of the classification does not change significantly.
The next step was a classificatfon on the basis of the minimum distance from
the standard for the sum of the squares of.the difference in the normalized
expansion coefficients for two, three or more months. Although the expansion
coefficients f:,r the successive months have some coherence, it is not great
and it can be neglected. The expansion coefficients for the different months
can serve as new criteria for the classification. The combining of two, three
or more months as prognostic criteria will make it possible to improve the
evaluations of the classification up to 83%.
Table 4
Fact
Conjugation of Criteria
Forecast
I II III'. IV
� I 6 5 5 3
II 4 12 2 4
III 6 2 10 2
IV 3 2 4 8
Ttie hypothesis of a nondependence of the actual a.nd prognostic classes was
ctiecked using the k0 test. The matrix of conjugation of the criteria for the
for.ecast for October is represented in Table 4. The computed values of the cri-
terion were highly significant (24.96), whereas the 5% critical value was
)C 9(0.05) = 16.56. In this case the matrix of conjugation of criteria was con-
structed for predicting the classes of anomalousness of precipitation for Octo-
ber, having the evaluations p= 0.69 and pii = 0.46. As indicated by the conjug-
- ation matrix, the frequency of recurrence of prognostic classes approx�imately
cor.responds to the actual frequency of recurrence. The best probable success
is for the "middle" classes (second and third) and the minimum saccess is _`or
ttie first class. The falling of a forecast into a class opposite the actual
cl 44
a~ w
a)
o o a)

ao o

- m
,H ~4+

m
.-1 G
o
Cd
o
a)

ova) o~
bo
m
4+p 4
o
a) pG
O
~-H
o r.
H~n
r+
cn F-+
i
�d
rv a-a
G
o
0
o
a
~
r
p 0
w�H
0
3P
a~
N 4)
u0 q
(L)~
~v~
~ �
rov
u i
0o
0
0 o
�a
vo ~

q ~
.-I

r
10
o 3

o
a)
cu o p o+J
o
,-I
-H o b
o
~ o+J
o
-H
v4+
~ o v
a
H
o cv
4 +1
o~
G
,H u
-W
,H
P. -W
$4
4J ,H rI cd ,H :J
ri ,H
cd
N
-H
rl o
~-I u
,
D
E
1
p rl
41
rl
LJ
41 G+
O
cd " ld JJ r-I
N4J
tA 0 U
1J
0 0
+3
.0 0
td " O
tA
3
u-+ 3
cd 0
3
m
rv a
uW
10 ca 0 0 ed o
a m
0
ca 0 0
m
m o
0
0 P
0 cd 0
44
w
b0 ti-1

0
C." O O qrn
O b0
-rl
Gl -H "d
A
G+ -rl 44
b0 P cd
'J+ G
O G
O'Ll
Ul w
rl
p
�rl
44 M
UI
:J rl rl -r-1 rl Gl
*ri *rl
rD 11 r.
rl
N4.1 O
~-I
Gl .0
'b rl 44
td
L'
+1 O
~ N
O
~
O O
'd
41 N+J g W
W 4-J
"
D,
ed 'H
'
'
~
19 t'd
m

cA
'd U
l
-
O
3
~ 41
U) F~
~ 0
U
t!~
f~
I
..1
O0
r
4
~ q
N~
ri
Gl
~ N

tn 41 N N N ri
N N
tn
N ri 'd
N
C'+ ~--I G
N
m 41
~1 N N
rl
td ~
~ O
'J . V1
S-i
1.
O�>
1J
+J r-I 1J cS1
f3 P
~
N Rl ~
~
+1
N R1 O

Z
>
~
~ p

11 11 41
a~ cd
co
}-i 00 p H
a~

a cd
cd
a~
~ a)
a)
a a) ~ a)
0
3 y
a)
y
J
o,
) *H
N
�
$4 4J
H,n
a
A
41A
q
-Hfa'tiM A +1
10 H
v
+1 w 3
A
aW +
H
u bD
3A 14
U
u
41 ctl
a~
k ~
w o
w
~
~
*
^
o
~
o
~
-
~
~ c n
~
~
~
-
- ~
~
i
c
~
v
i ~
o
10
id +J

=
N
P.
0
G
~
-

- in

c
d
v
, p �,-I
.
.

.
z
~
~
~
~
cq
U
~
~
~

CO
W
N
4J
i.
~
tn 1J
i.
~ tn
-rl
N'b
r-I
41
r1

N.C q
b0
U1
rl 0
'
U
~
-rl w
U rl
cti
~ N
Gl
GJ lY >1 N rl
F"+
44 4-I
rl
ri N
4-1
-ri
-W Ul
R1 :j
P.
44
'J, 0 `1-4 cd r-I
~
cd
4J 0)
r4
O �rl
44
-ri Fi
U1 �rl
44 t7'
-.i
o
r-4 ~

- i-4
-H 41 ,.4 y, P.
~
.c
r+
o
ti-+ : 'o
~
o
s m
-H ,J
14 IH
u
co
2
~
rl N
GW bip 0
o
v
0
u
x
-r-1
w
-rl
p
P b cd
o 0 P
14
o
r4
~
N0 4J N
~>1 co "
:1 a
En ~
a
a)
4
(1)
4J
-H a o
v
f-4

0
Cd u
p
U) a)
p w.0 o cj
'
ro
w
w c~
=
Q
c~ $40
7 =
,J
cn
A'
u EO
cn =
'K
i.+
td
x=
F
O>
H= N V3
O
=
1:4
~4
C
a
22
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFRiCIAL USE ONLY
_ I.nvestigation of the regime characteristics of wind-induced level fluctuations
tn regions not supplied with observational data.
Two- and three-dimensional models of storm surges were used in solving these
I)robl.ems [11, 12]. The realization of models for arctic seas was described in
[6-8]. Table 1 gives a list of the principal numerical experiments carried out
as a result of investigation of the features of level fluctuations observed
in arctic seas. In the first approximation a study was made of the most signif-
icant level fluctuations during the navigation season when the influence of the
i.ce cover can be neglected. In order to satisfy the requirement of adequacy
oE the results of computations and data from in situ observations the modeling
- of wind-induced level fluctuations was accomplished simultaneously for the en-
tire shelf zone of arctic seas. The criteria of agreement between computed and
= observed level values were the statistical evaluations used in numerical hy-
clrociynamic weather forecasts: mean computation error, mean absolute error in
c.omputations, mean square error, computation quality parameter and correlation
coefFicient between the computed and actual level variation. A comparison of
tlie results of modeling with observational data was carried out at 35 points
located on the coast and on islands of the studied seas.
Table 2 gives the mean statistical quality characteristics obtained in r.he mod-
eling of 10 synoptic situations causing the most significant level fluctuations.
'Ihe presented evaluations are evidence of the accuracy which at the present
_ time meets the requirements imposed on comgutations oriprediction of wind-in- ~
d tic ed phenomena [9].
'C}ie developed methods for computing level fluctuations in ar.ctic seas can be
used f.or prognostic purposes when there is a reliable prediction of the surface
field of atmospheric pressure or wind. Exper ience shows that for computing a
situation with a duration of 36 hours it is necessary to expend 40 minutes
witll a"Minsk-32" electronic computer. The collection and processing of initial
data require 40 minutes and the preparation of the collected information in a
for.m convenient for analysis requires 20 minutes. It is desirable that the
making of forecasts by hydrodynamic methods be carried out in combination with
numerical forecasts of ineteorological and other hydrological elements. This
wi11 make it possible to reduce the random errors in the transmission of data
::ind reduce time consumption on preparatory work.
'I'ile advance time and accuracy in predicting wind-induced level fluctuations with
: use of the considered methods is dependent for the most part on the correspond-
i.�g characteristics of prediction of the fields of atmospheric pressure and
wi.nd. In actuality, using as a point of departure an analysis of the equations
of motion employed in computing wind-induced processes, it can be concluded
chat tfie level errors must be dependent not only on the errors in meteorolog-
ical parameters, but also on the absolute velocity, direction and duration of
- the wind effect. An analysis made it possible to explain why in the computations
_ oE wind-induced phenomena for real situations the best results are obtained
wfien modeling the most significant level fluctuations. First, considerable lev-
el ('luctuations are caused by pressure formations with large horizontal* gradi-
encs of atmospheric pressure, and in this case the errors in computing wind
vc:loclty and direction are usually minimum. Second, the wind in the case of
23
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/49: CIA-RDP82-00850R040500070010-8
Qreat level fluctuations operates along the effective direction of wind-in-
dticeci Eluctuations and an error in wind direction in this case of even 30�
r.auses an error in level of not more than 20 cm. Third, the duration of opera-
tion oE strong winds usually does not exceed 6-12 hours, which also does not
1 e:id to signif icant level distortions.
- In the modeling of insignificant level fluctuations caused by low-gradient pres-
sure fields the relative errors in the computed levels sharply increase, for the
most part due to an increase in the errors in determining the direction of the
prevailing wind as a result of the nonrepresentativeness of blurred pressure
Elelds resulting from the thin network of ineteorological stations in the Arctic.
Thus, an evaluation of the accuracy of numerical hydrodynamic methods for com-
Puting, wind-induced level fluctuations with the existing accuracy in determin-
ed and computed meteorological f ields must be carried out using data on the
mztximtim level fluctuations. Accordingly, the statistical evaluations of the
nuality of modeling were ranked using the magnitude of level fluctuations. It
is desirable to formulate the requirements on the accuracy of ineteorological
for.ecasts in different regions, upon attaining which a numerical hydrodynamic
Prediction of wind-induced level fluctuations is applicable.
Wittiin the framework of the developed methods an attempt was made to obtain re-
p,ime characteristics of wind-induced phenomena of practical importance. One
of these is the extreme level fluctuations which are possible once a year or
once in 5, 10, 20 and 50 years. Maps with such characteristics were preparel for
all the studied arctic seas and were included in a scient'Lfic-practical manual
prepared for publication. A comparison of the computed extremal level values
_ determined by hydrodynamic modeling and the results of computations of the ex-
_ tremal levels at shore stations, carried out using observational data and using
ttle method described in [1], gave satisfactory results. For a number of points
- along the stlore the data computed using the model were too low by 20-40 cm.
7'his must be expected because in the developed method the computation of extrem-
al levels did not take into account the movement of pressure systems. Within the
- Eramework of the created method it is evidently necessary to develop procedures
- for taking the movement of pressure formations into account in order to make
allowance for possible resonance phenomena and also to take into account the
interaction of wind-induced phenomena with tidal level fluctuations.
A definite step has been made in this direction. In particular, with the partic-
ipation of Ye. N. Uranov there was a classification of atmospheric processes
evidently causing considerable level fluctuations in the Kara Sea. This,was
clone on the basis of a joint analysis of three interrelated synoptic fields
(stirE>n fin rsn 1717 r

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000500070010-8
The described preliminary stages in establishing relationships between the state
of agricultural fields and their brightnesses make it possible to proceed to a
solution of the problem of evaluating the state of agricultural fields using
data from space images.
A distinguishing characteristic of satellite images with an average resolution
(about 250 m) is the small linear dimensions of agricultural fields, which
makes difficult the tie-in of individual fields to data from field measurements.
- Accordingly, in the processing of satellite images with an average xesolution
it evidently is necessary to take into account not individual agricultural
fields but a group of fields within the limits of individual districts, regions,
- etc.
The algorithm for the processing of a satellite image involved the following.
The entire data bank of optical densities of space images was quantized into
classes corresponding to different agricultural crops (feature recognition).*
The next stage in the processing is the discrimination of optical density class-
es corresponding to a different state of the grass stand of a particular crop.
A determination of the boundaries (threshold levels) between optical density,
classes was carried out using characteristic breaks in the histogram (Fig. 4).
The histogram represents the ratio of the number of elements with a particular
optical density to the total number of optical density elements of the sector
- of the image being processed (in our case the territory of Khersonskaya Ob-
last).
The presence of well-expressed peaks (modes) or so-called breaks (change in his-
_ to;ram slope) indicates the presence of natural formations with different re-
flective properties. Intuitively it can be postulated that over the coiisidered
territory the natural features represented on this histogram (Fig. 4) include
a water surface, plowed soil and fields with different projective coverage of
the soil. On the date of the survey, 28 May 1980 (Fig. 5a), the following
groups of agricultural crops differed considerably with respect to the height
of the grass stand: winter, early spring and late spring. A knowledge of these
facts and the differences in the spectral variation of agricultural crops con-
sidered above made it possible, using the breaks on the histogram in Fig. 4,
- to def ine 5 classes of objects differing significantly in brightness: winter,
early and late spring crops, plowed soils and water bodies.
The threshold levels obtained on tliebasis of the histograms in Fig. 4 were
_ used in quantizing the optical density of the analyzed image into individual
classes (Fig. 5b), that is, the entire image, in accordance with the histogram,
was represented in the form of 5 optical density classes.
A comparison of the actual data on the structure of sown areas in the Kherson-
~ , skiv polvgon, obtained by an expedition of the Geography Facult.y at Moscow
ide will assume that there is a mutually unsmbiguous correspondence between
the optical density of the processed photograph and the registered brightness.
In the absence of such a correspondence it is necessary to make a brightness
correction of the images. We will not examine this problem here.
37
FOR OFFIC[AL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-40850R000500470010-8
- FOR OFFICIAL LSE ONLY
M
G!
H
~
D
u u
a~
�n N
~
-1

4)
b
0
a
cd
00
~
~
a
o
M
Cn
O
u
~
W
~
~
0
~
a
o
a
N
~

d
O
'd
U
~
m

a
~
$4
u u
a~
~
y
o
0
a ~
bo
$4
c
d
0
w
N
o
~
3

~
U
~
W
41
O
co
cd
a
0
-M
N
~
a
i
ca
~
.
~
b
a~
~
A
I
~
o
c~
.o
rn
~
0
0
0
0
0
U
v
a~
~
~d
3

14
~
o
a
a oo
c~

a
cd
I f
w
~
O
O
~
>1
~
N
v
~
b0
0 10
�

~
[

.
-i
~
.
+1
H
W
f0
e
d
~
h-1
Gl
00
td
~
~
ul)
~
i
a
P.
.
0
0
0
o
u
rn
~
o
~
41
u
w
g
�n
a
bo

.
~
,..7 0o
ch
A oo
0
~.1 .
R~ 00
U rl
W
0
'd
~
60
oo
p
r-i
~ u
~
3
~
"
1+
o
~
~ ~
v
,
r~
w
a(d
b
~
~
~ o
�
oD -W
~
M
w
a
N
N
hi
N
1C
38
FOR OFF[C[AL USE OIdLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

FOR OFFICIAL USE ONLY
State University and the results of quantization of image optical densities in-
to classes indicates that there is a correspondence between them (Table 4).
Table 4 ;
Structure of Agricultural Fields for Fragment of Khersonskaya Oblast on
28 May 1980 .
Land use areas Area, %
results of quantization of according to data of
optical densities Moscow State Univer-
sity expedition
Water bodies
8
8 ,
Plowed soil
12
15
Late spring crops
16
20
ERrly spring crops
20
11
Winter crops
44
46
Table 5
Evaluation (units)
Above average (4)
Average (3)
Below average (2)
Area, %
results of quantization of
optical densities
22
42
36
according to data of
Moscow State Univer-
sity expedition
24
51
25
The determination of areas with different fields from space images is dependent
to a considerable degree on the correctness of discrimination of the threshold
levels of optical densities from the histograms. These should be selected on
the basis of a priori information, standard measurements, etc. As already stat-
ed above, for this purpose we used histograms of the distribution of optical
density and a priori information on the phenophases of developmant of agricul-
tural crops.
The sectors with groups of agricultural crops similar with respect to reflect-
ivity, defined on the images, are also nonuniform with respect to optical den-
sity. In this case the optical density of the image is dependent on variations
in state of the crop: the greater the degree to which the vegetation cover
conceals the soil, the less is the optical density on the positive images. This
Eact can be used in discriminating areas having a different state of agricul-
tural crops on the images. A further interpretation involves a determination of
the quantitative relationships between the image optical density and the state
� 39
FOit OFF[C[AL USE OWLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2047/02/09: CIA-RDP82-00850R000504070010-8
FOR OFFICIAL USE ONLY
of a narticular crop (the quantity of vegetation mass or some other parameter).
i1 -
_ o
a
B) .
C3 '
B
Eg ~ .
1 .,..r'

Fig. 5. Fragment of space image with average resolution on 28 May 1980 (a) and
results of image processing. a) structure of agricultural fields (1 water
bodies, 2) plowed soil, 3) spring crops, 4) winter crops); b) state of winter
crops (1 above average, 2) average, 3) above average, 4) other land use
areas.
40
- FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
FOR OFFICIAL USE ONLY
Figure 5c gives an analysis of the state of winter wheat for the selected frag-
ment of Khersonskaya Oblast_ The iuiage optical density of the selected sector,
fed out to a color display, was uniformly divided into three parts correspond-
ing to the mean state of winter wheat and states above and below the means,
wh4ch approximately corresponds to a projective coverage of less than 60%,
80% and more than 80%. The selected image density gradations corresponded to
those threshold levels which were mentioned above. Then by program all the
analyzed fragment was represented-in gradations of optical density in accord-
ance with the threshold levels. Thus, on the basis of a specialized complex
for the digital processing of images it was possible to compute areas with a
different state of winter wheat. The results of an analysis of the space image
are given in Table 5 together with the actual data obtained by an expedition
of the Geography Faculty of Moscow State University on the state of winter
wheat during this period. Table 5 shaws that there is a definite correspondence
between the results of an image and actual data, which indicates the possibil-
ity of using the described semiautomatic processing scheme for evaluating the
state of agricultural crops over great areas.
To be sure, there is�definite arbitrariness in the choice of threshold levels
- of optical density when using the described method for evaluating state. These
levels, and also the interrelationships between density gradations and the
- quantitative indices of state must be found on the basis of simulated models
of interaction between radiation and the soil-vegetation syste.m with the use
of a priori information. There are considerable difficulties in making a bright-
ness correction of space images. However, the preliminary results of analysis
of natural phenomena cited above and the results of processing of satellite
images make it possible to conclude that there are real possibilities for eval-
uating the state of agricultural fields using space survey data with apparatus
having an average resolution.
BIBLIOGRAPHY
l. Belyayeva, I. P., Rachkulik, V. I. and Sitnikova, M. V., "Correlation Be-
tween the Brightness Coefficient of the Soil-Vegetation System and the
_ Quantity of Vegetation Mass," METEOROLOGIYA I GIDROLOGIYA (Meteorology
and Hydrology), No 8, 1965.
- 2. ISSLEDOVANIYE PRIRODNOY SREDY S PILOTIRUYEMYKH ORBITAL'NYKH STANTSIY (In-
vestigation of the Environment Froffi Manned Orbital Stations), edited by
K. Ya. Kon3rat'yev, Leningrad, Gidrometeoizdat, 1973.
3. Kondrat'yev, K. Ya.'and Fedchenko, P. P., "Possibilities of Using Reflec-
tion Spectra of Soils for Studying Their Properties," ISSLEDOVANIYE ZEMLI
IZ KOSMOSA (Investigation of the Earth From Space), No 1, 1980.
4. Rachkulik, V. I. and Sitnikova, M. V., "Some Problems in Determining the
Biomass o.f Desert Pastures and Fields of Agricultural Crops METEOROLOG-
IYA I GIDROLOGIYA, No 6, 1976.
41
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2407/42/09: CIA-RDP82-40850R000500470010-8
FOR OFFiCIAL USE ONLY
UDC 551.508.953
EVALUATION OF SPECTRAL RESOLUTION OF OPTICAL APPARATUS FOR MEASURING MINOR GAS
COMPONENTS IN ATMOSPHERE
-Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 (manuscript received
1 Jun 81) pp 102-106
[Article by A. V. Polyakov and Yu. M. Timofeyev, candidate of physical and mathe-
matical sciences, Leningrad State University]
[Abstract] Great difficulties are encountered in studying-the spatial-temporal
- variations in the content of minor gas components in the atmosphere. This dic-
- tates the creation of special optical instruments with a high or ultrahigh spec-
tral resolution. In order to ascertain the required resolution of the needed in-
struments; as well as for evaluating already existing instruments, the authors
examine a simplified model of transfer of solar radiation on oblique tra3ector-
ies (measurements from satellite, aircraft or balloon) and have constructed nomo-
grams (and expressions supplementing them) which make it possible to solve this
problem in a broad spectral r.egion (1-100 � m) for a large number of minor gas
components. The rules for using the nomograms are given. The procedures for de-
termining 4det and Ayo t are detined (d y det is the spectral resolution neces-
sary for detecting a spectral line with a known half-width and intensity for a
stipulated altitude corresponding to a minor gas component with a known concen-
tration; QVopt is the optimum spectral resolution in line measurements at a
stipulated altitude). By way of illustration, a table gives the AYdet and 'A yopt
values for different altitudes applicable to the conditions of a satellite exper-
iment. Figures 2, tables l; references 9: 7 Russian, 2 Western.
42
F'Olt OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2047/02109: CIA-RDP82-00850R000504070010-8
UDC 551.501.796+551.501.724
FOSSIBILITY OF DETERMINING ATMOSPHERIC TIIHPERATURE PROFILE BY ACOUSTIC SOUNDING
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 (manuscript received
1 Jun 81) pp 106-110
[Article by A. Ya. Bogushevich and N. P. Krasnenko, candidate of technical sci-
ences, Institute of Atmospheric Optics, Siberian Department, USSR Academy of Sci-
ences]
[Abstract] At present it is possible to determine the temperature profile only by
the combined radioacoustic method, based on the scattering of electromagnetic
waves on periodic inhomogeneities of air pressure created by an acoustic packet.
A new method for measuring the temperature profile is proposed here. Data on the
- temperature profile can be obtained by using a bistatic sounding geometry and by
measuring the time of propagation of an acoustic pulse along its trajectory. In
this method, assuming a stationarity and horizontal homogeneity of ineteorological
fields at the measurement scales (0.3-6 sec and 30-300 m), the entire boundary
layer is broken down into n layers (altitudes Hl-Hn) and a piecewise-linear ap-
proximation of the temperature profile is used. It is shown that by successively
� measuring the sounding altitude Hl,it is possible to determine the temperature
profile. The most significant-factors exerting an influence on the accuracy in
measuring temperature are: wind, acoustic noise, sound refraction, variation of
the Q coefficient, f3nite character of the dimensions of the scattering volume
and turbulent fluctuations of ineteorological fields. The principal systematic and
random errors associated with these influencing factors are individually analyzed.
The influence of the wind, for example, can 'be taken into account by employing
compensating geometries; wind refraction corrections can be made'by the same pro-
cedure. A correction must be introduced for real humidity at the ground level. In
general, it is easy to compensate for the systematic errors. Fluctuation errors
are more important and the most important of the fluctuations is noise, resulting
in fluctuations of signal arrival time. A formula is derived to take this into ac-
count. The article examines a model with an atmosphere with standard profiles of
meteorological parameters and a Kolmogorov turbulence spectrum. There are opti-
mum frequencies for acoustic sounding and optimwn spacings between the antennas
of the bistatic sounder with which the random error in determining temperature
is minimum. In a particular example it is shown that in a suburban area the mean
temperature profile could be measured by the bistatic acoustic sounder to alti-
tudes 300-500 m with a good-accuracy. Figures 2; references 7: 4 Russian, 3 West-
ern.
43
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102109: CIA-RDP82-00850R000500070014-8
FOR OFFICIAL USE ONLY
UDC 551.577.53
' INADVERTENT INFLUENCE OF MAJOR CITIES AND INDUSTRIAL CENTERS ON PRECIPITATION
' Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 (manuscript received
_ 14 .Tul 81) pp 111-119
[Article by I. V. Litvinov, candidate of physical and mathematical sciences, Insti-
tute of Experimental Meteorology]
- [Abstract] Precipitation zones are known to develop more frequently over a city
than in the suburbs. There is also an increase in the number of thunderstorms
over cities and an increased frequency of recurrence of days with precipitation,
as well as the frequency of extremal precipitation and the duration of precipit-
ation. However, this pattern is notialways observed: Leningrad, Moscow and De-
troit, for example, do not entirely adhere to this generalization. The review of
the literature presented in this article makes it clear that cities and industrial
areas do exert an influence on precipitation sums both directly over built-up re-
- gions and nearby. It is the heat release of the cities and the structure of the
_ built-up area which exert the greatest influence. It is not entirely clear how
ice-forming nuclei of anthropogeriic origin exert an influence on precipitation,
but they do not alter precipitation~sums by more than a few percent. Nowever, ur-
banization does not exert an influence on the quantity of precipitation falling
over extensive territories because city areas account for but a tiny percentage
of the total area of a country, even in Europe. Over large cities and industrial
areas this precipitation increase may be 30%, but the overall effect for the land
is no more than a few tenths of a percent.During the last 100 years in Europe
there has been only an insignificant increase in precipitation in the northwest
and a decrease in the southeast. Nevertheless, there is an anthropogenic effect
on precipitation and this must be taken into account in climatic studies and in
investigating the results of artificial modification In the neighborhood of
large cities and industrial centers. Figures 2; references 89: 38 Russian, 51
Western. 44
FOR OFF[C[tiL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8
Ff
,
REVIEW OF MONOGRAPH 'GIDROLOGICIiESKIYE PROTSESSY I IRH ROL' V FORMIROVANII KACHESTVA
VODY' ('HYDROLOGICAL PROCESSES AND THEIR ROLE IN FORMING WATER QUALITY'), BY
V. A. ZNAMENSKIY, LENINGRAD, GIDROMETEOIZDAT, 1981, 248 PAGES
kNioscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 pp 120-122
[Review by A. V. Karaushev, professor]
[Abstract] This new monograph is devoted to an investigation of the processes of
transport of water masses and soluble substances in large water bodies situated
in different zones. The author has developed a new multisided hydrological and
hydrodynamic approach for evaluating the processes of formation of water quality.
The book definitely has a practical character. Methods are proposed for the model-
ing and laboratory investigation of water exchange and the transport of substances
in water bodies. The proposed methods can be used in planning systems for inven-
tarying and monitoring water quality. Mathematical models have been proposed which
can be used in solving ecological problems and in developing schemes for the multi-
' sided use and conservation of water resources. The book has an introduction, five
chapters and summary. The introduction enumerates the principal problems examined
by the author. Each chapter gives an evaluation of the present status of. the con-
sidered matters. The first chapter is a detailed examination of water exchange
processes in water bodies. The second chapter examines the hydrodynamic structure
of flows in water bodies. The third chapter is devoted to the influence of water
exchange and dynamic structures on the transport and distribution of dissolved
substances. The fourth chapter gives. the author's own methods for the hydraulic
modeling of the processes of transport of water masses and dissolved substances,
including the results of numerous experimental investigations. The fifth chapter
gives proposals on use of such methods in solving problems in the rational use and conservation of waters, monitoring water quality and establishing norms for
the admissible discharge of waste waters into water bodies, in the development
of mathematical models of water quality and in planning the multisided use of
water resources. The book wi11 be highly useful in solving many problems in the
national economy.
45..
FOR AFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000500070010-8
FOR OFF7CIAL USE ONLY
CONFERENCES, MEETINGS, SEMINARS
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 pp 123-125
[Article by P. Yu. Pushistov, A. A. Zhelnin and V. F. GridasovJ
[Abstract] An international symposi,4m "Current Problems in LJeather Forecasting"
was held in Vienna during the period 23-26 June 1981. It was organized by the
Austrian and American Meteorological Societies. Representatives of 14 countries
discussed theoretical and statistical aspects of short-range weather forecasting,
intermediate- and long-range weather forecasting. This brief summary of the sym-
posium gives the content of reports by American, British, Austrian, West German
and other specialists, with a few lines or short paragraph being devoted to each
report.
A national conference on "Problems and Ways to Develop Methods for Observing Mois-
ture Supplies for Crops" was held at Cherkassy during the period 8-12 September
- 1981. It was sponsored by the State Committee on Hydrometeorology and Environmen-
= tal Monitoring and the Scientific and Technical Society of Agriculture. The con-
ference revealed that during the last decade great advances have been made in
determining soil moisture content and in ascertaining the parameters of moisture
supply for agricultural crops, including the use of remote methods for measuring
- moisture, such as SHF radiometry and aerial gamma surveys. Portable multiparameter
hygrometers are being readied for production (these are based on measurement of
the electric parameters of the soil); some stations naw have neutron hygrometers;
more attention is being given Co computation methods; the drying of soil samples
- in a thermostat is being replaced by high-frequency drying. There are still no
adequate methods or instruments for replacing the thermostat-weight method for
deCermining soil moisture content: neutron hygrometers, aerial gamma surveys, SHF
radiometry and computational methods all have their shortcomings. Different organ-
izations are striving to upgrade these methods or combine them for achieving be'tter
results. The reports and communications presented at this conference are to be pub-
lished in the Transactions of the All-Union Scientific Research Institute of Agri-
cultural Meteorology in 1982. The next such conference will be held in 1984.
' 46
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2047/02109: CIA-RDP82-00850R000504070010-8
FOR OFFTCiAL USE ONLY
NOTES FROM ABROAD
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 pp 125-127
[Article by B. I. Silkin]
[Abstract] In the journal NEW SCIENTIST, Vol 91, No 1267, 1981, it is reported that
Doctor Stanley Shannon and his colleagues at the Illinois Hydrological Administra-
tion have completed a cycle of studies of the interrelationship between aircraft
flights and weather conditions on the approaches to 0'Hare Field. Since the 1960's
these flights have caused a 10% increase in cloud cover over an extensive area
along these routes. This has resulted in a definite cooling in Illinois and an in-
crease in precipitation. On the other hand, at nighttime the temperatures remain
higher and the winters have become warmer.
*~r*
NEW SCIENTIST, Vol 91, No 1274, 1981, reports that by making use of satellite data
R. Orville and B. Vonnegut have compiled a map of the global distribution of light-
ning discharges. This phenomenon is virtually absent in the high latitudes:and
quite rare in the middle latitudes, being most common in the equatorial regions of
the land. It still remains a mystery why lightning is so rare over the oceans.
In another issue of NEW SCIENTIST'(Vol 91, No 1267, 1981) it is reported that a
specialist at the Atlantic Richfield Company, R. Currie,.has proposed a method for
superlong-range forecasting of ineteorological conditions on the basis of cycles
of the lunar tidal effect, making use�of maximum entropy spectrum analysis. It ap=
pears that there is a correlation between the 18.5-year cycle of the maximum of
the lunar-tidal effect and arid years in the.Western United States (although in individual cases the drought coincides with the 22-year double cycle of the maxi-
- um in spot-forming solar activity).
m
~
NEW SCIENTIST, Vol 91, No 1267, 1981, reports that World Data Center A on Solar-
Terrestrial Physics at Boulder was threatened with complete shutdown as an econ-
omy measure, much to the dismay of the world scientific community. Now it has been
decided that part of the work will-be contract.ed out, which will result in higher
user cost.
47
FOR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2047/02109: CIA-RDP82-00850R000504070010-8
FOR OFFIC(AL USE ONLY
OBITUARY OF YEKATERINA NIKITICHNA BLINOVA (1906-1981)
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 pp 127-128
[Article by staff of the Board of the USSR State Committee on Hydrometeorology and
Environmental MQnitoring and personnel of the USSR Hydrometeorological Scientif ic
-Research Center]
[Abstract] Yekaterina Nikitichna Blinova, corresponding member, USSR Academy of Sci-
ences, died on 15 December 1981. She was head of the Division of Planetary Atmo-
spheric Dynamics and Hydrodynamic Long-Range Weather Forecasting at the USSR Hy-
drometeorological Scientific Research Center. First working (1934) at the Main Geo-
physical Observatory, she later (1943) moved to the Central Institute of Forecasts.
Her entire career was devoted to the field of long-range weather forecasting. As
early at 1943 she published the.major monograph GIDRODINAMICHESKAYA TEORIYA VOLN
DAVLFNI'YA, TEMPERATURNYKH VOLN I TSENTROV DEYSTVIYA ATtZOSFERY (Hydrodynamic Theory
of Pressure Waves, Temperature Waves and Atmospheric Centers of Action), which out-
lined the first theory of numerical,long-range weather forecasting. In her doctor-
al dissertation of 1946'she described how theoretically it was possible to repro-
duce the really observed distribution of ineteorological elements. She inspired oth-
ers to follow her in advancing the development of the hydrodynamic theory of cli- mate and long-range forecasting. The basis for these investigations was the spec-
tral approach, applied widely thereafter in many linear models of climate, general
circulation of the atmosphere and long-range forecasting. She supervised the f irst
operational preparation of several variants of long-range forecasts. Ye. N. Blinova
then proceeded from linear to nonlinear schemes and application of primitive equa-
tions of dynamic meteorology. During recent years she developed a method for long-
range forecasting on the basis of discrimination of the "main oscillations" and pub-
lished an algorithm for solving prognostic spectral equations. A model was proposed
in which the changes in meteorological elements with time are computed simultaneous-
ly with the climatic background. All this represents but a fraction of her contrib-
utions to this field. She was active in the GARP program, being a member of the .
Soviet committee on this program and chairman of the subgroup on numerical experi-
mentation. Figures: 1. 48
F'OR OFF[CIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8

APPROVED FOR RELEASE: 2447102/09: CIA-RDP82-44850R444544474414-8
FOR OFFICIAI, USE ONLY
ORITUARY OF SERGEY IVANOVICH SIVKOV (1901-1981)
Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 82 p 128
[Article by a group of comrades]
[Abstract] Sergey Ivanovich Sivkov, doc-,tot of geographical scii-inces, an outstand-
ing Soviet specialist in the field of actinometry, died at age 80 on 28 May 1981.
He began his work at Kursk Observatory, advanced through the ranks and served as.
its head during the years 1928-1930. Thereafter he was director of the Tien Shan
High-MountainObservatory and from 1935 through 1950 was head of the Karadag Af-
filiate of the Actinometry Institute. On two occasions he wintered at polar sta-
tions in the Arctic. Subsequently he laid aside administrative assignments and ded-
3cated himself exclusively to scientif ic research. His publications, 54 in all,
dealt with such matters as methods for computing actinu:netric'fluxes and inves-
tigating instrument performance. Mueh of his work is still of practical importance
in the operation of Soviet and foreign actinometric services. During his work at
the Main GeopHysical Observatory he wrote (1968) a monograph entitled METODY RAS-
CHETA KHARAKTERISTIK SOLNECHNOY RADIh.TSII (Methods for Computing Characteristics
of Solar Radiation) and thereby earned the academic degree of doctor of sciences.
His final area of research and publication was the investigation of cloud cover
and radiation. . COPYRIGHT: "Meteorologiya i gidrologiya", 1982
5303
CSO: 1864/9
- END -
49
FOR OFFICtAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000500070010-8