Document Type: 
Document Number (FOIA) /ESDN (CREST): 
Release Decision: 
Original Classification: 
Document Page Count: 
Document Creation Date: 
November 1, 2016
Sequence Number: 
Case Number: 
Content Type: 
PDF icon CIA-RDP82-00850R000200090007-3.pdf9.45 MB
APPROVEE3 FOR RELEASE: 2007/02/08: CIA-ROP82-00850R000200090007-3 3 JUNE 1980 ME NO. 3f MAi4Ci-i 1980 I OF 2 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000200094407-3 FOR OFFIC[AL USE ONI.Y JPRS L/9121 3 June 1980 USSR Report METEOROLOGY AND HYDRUIOGY No. 3, March 1980 ~BIS FOREICN BROADCAST INFORMATION SERVICE FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 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. Headlines, editorial reports, and material enclosed in brackets IL are supplied by JPRS. Processiag indicators such as [Text] _ or [ExcerptJ 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- 4 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 within 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 attitudes of the U.S. Government. For farther information on report content ca11 (703) 351-2938 (economic); 3468 (political, sociological, military); 2726 (life sciences); 2725 (physical sciences). 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: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY I USSR REPORT METEOROLOGY AND HYDROLOGY No. 3, March 19$0 JPRS L/9121 3 June 1980 Selected articles from the Russian-language journal METEOROLOGIYA I GIDROLOGIYA, Moscow. F CONTENTS Dynamic-Statistical Parameterization of the Ocean's Thermal Memory ~ (Sh. A. Musayelyan) 1 Zonal Dietribution of Cloud Cover Over the Earth (T. G. Berlyand, et al.) 13 Determinatione of the Mass Concentration of Aerosols in the Plumes of Industrial Plants Using a Lidar (I. M. Nazarov, et al.) 25 Change in Microstructure of Stratiform Clouds Under Influence of Condensation Nuclei (K. B. Yudin) 37 - Probabiliatic Approach to the Objective Classificatiou Problem (A. A. Burtsev) 45 = Structure of Tropical Cyclone "Carmen" Determined from Aerological Sounding Measurements Over the Ocean - (V. V. Galuahko, et al.) 53 Accuracy in Determining Atmospheric Ozone Content Using Data from . Measurements of Outgoing Radiation (Yu. M. Timofeyev, et a1.) 61 Model for Computing Ttnickness of the Quasihomogeneous Layer in the Ocean C (T. R. Kil'matov and S. N. Protasov) 7) - a - [III - USSR - 33 5& T FOUO) FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 , Computation of Characteristics of the Quasi-Isothermal Layer in the ' Equatorial Zone of the Ocean (A. B. Polonskiy) 80 Opposition in Ice Re3istribution in the Waters of the Foreign Arctic (V. I. Smirnov) .............e................................, 91 Methods for Regime and Operational Determination of River Runoff (I. F. Karasev) 98 Objective Method for Evaluating the Probable Success af an Individual Hydrological Forecast (G. N. Ugreninov) 113 Distribution of Current Velocity and Turbulent Friction Near the Water- Air Interface (N. K. Shelkovnikov, et al.) 118 . Method for Measuring the Thickness of a Petroleum Film on the Surface of a liater Basin (T. Yu. Sheveleva, et al.).................................... 124 Computation of Mean Oblast Yield of Winter Wheat (A. R. Konstantinov and D. N. Peradze) 130 Computation of Atmospheric Counterradiation (N. I. Rudnzv) 137 Nephelometric Method for Dctermining Light Attenuation in the UV and Visib].e Spectral Regions (T. P. Toropova) 141 Transmission of Intpgral Solar Radiation by a Grass Cover (T. K. Tammets) 146 Optical Properties of Crystalline Clouds . (0. A. Volkovitskiy) 150 - Sixtieth Birthday of Solomon Moiseyevich Shmeter 164 _ At the USSR State Committee on Hydrometeorology and Environmental Monitoring (V. N. D�rozdov and V. M. Voloshchuk) 166 Conferences, Meetings and Seminars � (L. S. Speranskiy, et al.) 169 _ News from. Abroad (B. I. Silkin) .............o................................. 178 - b - FOR OFFICIAL USE ONLY I APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY c PUBLICATION BATA English title : METEOROLOGY AND HYDROLOGY - Russian title ; METEOROLOGIYA I GIDROLOGIY A Author (s) . - Editor (s) : Ye. I. Tolstikov Publishing House : Gidrometeoizdat Place of Publication ; Moscow Date of Publication : March 1980 Signed to press 18 Feb 80 Copies ; 3780 COPYRIGHT . "Me'teorologiya i gidrologiya", 1980 � c - FOR OFFICIAL USE ONLY � i APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 F'i; : OFFICIAL USE ONLY UDC 551.509.33 DYNAMIC-STATISTICAL PARAMETERIZATION OF THE OCEAN'S THERMAL MEMORY Mb scow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 80 pp 5-14 (Article by Doctor of Phyaical and Mathamatical Sciences Sh. A. Musayelyan, USSR Hydrometeorological Scientific Research Center, submitted for publica- tion 5 July 1979] Abstract: The problems of predictability of atmospheric processes and long-range weather forecasting are discusaed. Particular atten- tion is devoted to the role of external ener- gy sources. The author emphasizes the role - of cloud cover over the ocean as a regulator of the heat influx in the process of absorp- tion of solar radiant energy by the ocean. A dynamic-atatistical method for parameter- - ization of the ocean's thermal memory is out- 1 ined . [Text] In the short-range forecasting of ineteorological fields the atmo- sphere in the first approximation can be regarded as an isolated me33um and - the processes transpiring in it can be regarded as adiabatic. In other words, in the first approximation the short-range forecasting of ineteor- ological fields is the Cauchy problem a problem with initial data. Ac- cordingly, in numerical shcrt-range weather forecasting the principal fac- tor is the quality of analysis of the fields of initial data. [T11e pro'..lem of bnundary conditions will not be discussed here (Author's note).] In ac- tuality, however, the processes transpiring in the atmosphere are strictly nonadiabatic, and when one speaks of study of atmospheric behavior over ~ long and very long time periods the external energp conditions must be re- $arded as determining factors. And this means that the problem of modeling of short-period climatic varigtions or long-range predictiofl of ineteorolog- ical fields is the Cauchy problem, but with a source. In other words, in . this case it is insufficient to have only qualitatively analyzed initial - fields; it is also neceasary to take into account, and rhis is partic-lnr_- ly important, the principal macroscale nonadiabatic factors. _ 1 tvin /1L'0TnrA1 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 L.,A.. - The contributions of initial data and external energy sources to numerical weather forecasting for different times in advance have been examined by a number of authors (for example, see [1, 4]). These contributions can alsu be inveatigated directly on the basis of the firat law of thermodynamica. Qualitative Analysis of Contributions of Initial Data and Nonadiabatic - Factors As is well known, the mathematical expression for the first law in thermo- dynamica, applicable to atmospheric conditions, is the energy equation, - which can be written in the following form: _ Or dT di k = E. - UF t u d H + a Sin N d A a'= A Here t is time, /k is geographic longitude, e= 7T12 - so, Pis latitude, ve ( 9, 21 , t) and va (B, A , t) are the velocity components the ~ B and A axes respectively, T( e, X , t) is temperature, a is the earth's ~ mean radius, k is the coefficient of horizontal macroturbulent exchange, A is the Laplace operator in spherical coordinates and the function ~ E( B, i1 , t) integrally describes all the heat influxes. We will assume that it is necessary to obtain a solution of equation (1), periodic relative to A , symmetric relative to the equator and limited at ' the pole, satisfying the initial condition T(d, X, t) jr=o= To (g,. i.). (2) For convenience in the subsequent presentation we introduce the notation - vp dT vX dT - a d 9 a sin d d?. - I(9, i., t) and equation (1) is rewritten in the following form: ar L A T~E-}-I. (3) dt a= Aesuming that the functions E(8, a, tj, I( e,'1., t) and Tp( e,~) can be repreaented in the �orm of series in spherical functions E-_ Re r~ EA ~t~ E-'m a Pn (COS 6), ~.r n m ~4 ~ I:= Re ,~,~J: I� (t) C- 'm I Pn (COS e)r nm T, = Re J:G, Totn e' r"' a Pn (COS fj), /t fi we will seek a solution ot equation (3) in the form T= IZe 11 Tm (t) C- iin a E7- (COS H)� (5) nm 2 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY In formulas (4) and (5) E1�(t), In(t) and Tn(t) are complex functions, TO n are complex numbers, and ~(cos e) is the adjoint Legendre p,~lynom~al. It ie easy to show that the considered problem is reduced to integration of the equation dTn"l k _L dl ~ tt= 11 ~J1 -j- ] ) 7''" _ ~ni -t j~ n (6) with the stipulated condition jm (t) I t-0 = Tomn� (7) _ The solution of problem (6)-(7) can be written in the form � /.n (t) - akn c Xp I a, 1l (lt 1) ll t ~ J) ~8) tl Ton ex p In Q, (I[ 1) t~. where ETn and In are values of the functions En(t) and In(t), averaged in . the considered time interval. Formula (8) is prognostic. The second term on the right-hand side of this formula shows that in the predictlon of temperature the role of initial data actually decreases.with time, and this decrease occurs exponentially. However, the first term on the right-hand side of (8) shows that the con- tribution of the energy sources increases with time in conformity to the law 1- eXp n, !t (lt -t- 1) t~. L It is entirely obvious that both these processes (decrease in the role of initial data with time and increase in the contribution of energy sources) occura differently for different wave numbers n. In addition, the intensity of both these processes ia essentially dependent on the value of the caef- ficient of horizontal macroturbulent exchange k, which, as is well known, variea in a rather broad range. For waves of the scale of middle-latitude cyclones nZ 5 and for the value k= 106 m2/sec the contributions of the above-mentioned processes are rep- resented in Fig. 1. In this figure we have plotted time in days along the horizontal axis, and along the vertical axis some conditional rp value characterizing the contribution of some factor to the precomputed meteorological f ield (in easence, this is the parameter characterizing the success of the predic- - tion, for example, the correlation coefficient). 3 F(1R (1FFTf;TAT. ttCF (luT v APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 o,e' \ - 2 0.61 - 04 - - A' ZC JO 40 JD t r - Fig. 1. Contributions of initial data (1) and hest influxes (2). The following cor.clusiona can be drawn on the basis of an analysis of Fig. 1: 1) In the time interval 0< tG 5 days the success of the forecast is deter- mined pr-Imarily by the quality of analysis of the initial field and the contribution of the energy sources is secondary. In this case the fore- casting problem in the first approu;.mation can be regarded as the classic- al problem with initial data. According to the terminology which is now - in use forecasts for up to five days are called short-range. 2) In the time interval approximately from one to two weeks the role of the above-mentioned factors is approximately identically important, that is, the success of the forecast is determined to an approximately equal degree both by the quality of analysis of the initial field and the con- trib ution of energy sources. In this case the forecasting problem can be cons idered as the Cauchy problem with a source; both these factors are identically important. This is a case of intermPdiate forecasts. 3) In the time interval uing decrease in the ro tion of energy sourcea. the Cauchy problem with by energy sources. This and a season. from two weeka to one-two months there is a contin- Le of initial data and an increase in the contribu- In this case the forecasting problem is evidently a source, but the main role is nevertheless played is the case of long-range forecasts for a month 4) Finally, in the time interval frnm one-two months to infinity the role - of initial data becomes negligib].e and everything (or almost everything) is determined by the external energy conditions. Th{s is the case of a superlong-range or climatic forecast. Othe r classificstions also exiat. We will return to formula (8). It is easy to show that the time 20, at whoae end the role of initial data becomes equal to the contribution of the energy sources, is determined using the formula 0,7 a2 - To kn (n + 1) ' 4 . FOR OFFICIAL USE ONLY ~ APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Below we give the 0 valuea for different wave nt;mbera with k- 106 m2/ _ sec. n 1 2 3 4 5 6 7 8 9 10 -6 o days 162 54 27 16 11 8 6 5 4 3 The cited data show that the ultralong components of the initial fields _ (n - 1, 2, 3) are conserved for a very long time in the atmospheric mem- ory and only after one or more mpnths their role becomes less than the contribution of energy so+srces. Waves of a synoptic acale (n s,*-5) in the atmoepheric memory, however, are conserved only for approximately two weeks, after which the role of nonadiabatic factors increases. Finally, for the high-frequency components the role of initial data de- creases very rapidly and literally after several days the external energy sources become the determining factor. Another conclusion which follows from formula (8) relates to the predict- ability problem. As is well known, the traditional interpretation of the problem of predic- tability of atmospheric processes is based on the conclusion that the err- ors contained in the initial fields are doubled each two to four days, as a reault of which predictability is a definite limit equal to five-aeven _ days (for example, see [6]). This, to a certain degree, agrees with the just mentioned formula. Moreover, curve 1 in Fig. 1 shows that the con- tribution of initial data decreases with time independently of their ac- curacy such is the structure of the equations in hydrothermodynamics. But, according to this same formula, with a decrease in the role of ini- tial data the contribution of energy sources increases. Therefore, to apeak of atmoapheric predictability, taking into account only the role of initial data and laying aside the contribution of energy sources,means - from the very beginning to set a limit to predictability of 5-7 days. ' Z'his follows from formula (8) and can be seen clearly in Fig. I, The na- ture of atmospheric processes is evidently such that allowance for only one factor the contribution of initial data leads to a sharp limita- tion on the predictability limit, whereas the addition of another factor energy sources makes this limit longer. In any case, such is the nature of the equationa of hydrothermodynamics, and this rigorously fol- lows from formula (8) and can be seem clearly in Fig. 1. The structure of the prognostic formula (8) is such that when having even precise initial data, atrictly speaking, it is impossible to precompute the predicted field precisely even for one day. However, if in addition precise data are available concerning the sources, uaing this formula it is possible to precompute the temperature ffeld for any with the ac- curacy with which equation (3) is integrated. However, the initial fie1d ia known very approximately and only for a part of the earth constituting 5 F(1R (1FFTr7AT. iTCx Our v APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 about 20% of its entire surface. With respect to the energy sources, ,judg- ing from everything, our ideas concerning them are extremely vague. There- fore, when reference is to long-range weather forecasting the principal = problem is that of discriminating and parameterizing the most important ~ large-scale energy sources controlling the behavior of the global atmo- , sphere over long time intervals. Everything set forth above indicates that when refereace is to short-range weather forecasting the term "atmo- ~ spheric predictability" in general is an apt expression. However, if ref- erence is to long-range forecasting of ineteorological fields the tradi- tional problem of predictability must evidently be dealt with in a new way and the *_erm "atmospheric predictabil ity" itself must be replaced by some other, more suitable term, such as the term "long-range predictabil- , ity of the eai�*_h-atmosphere system." In equation (3) and formula (8) the temperature T( e,X , t) and its ex- panaion coefficients n(t) are functions highly dependent~on time. Now - we introduce into consideration the climatic temperature T( e,,A ) and we will ass ume that it is slightly dependent on timp (as it actually is), so that a T/ a tN 0. Then, if^we denote the expansion coefficienta T( e ,~k) in spherical functions by 1n it is easy to show that 2 FE" + IR ~ (g) n kn (n + 1) . dn t�z other har.d, on the basis of (8) it is easy to find (10) lim Tn (t) r-oo kit (+t t I) ' - Comparing (9) and (10), we have - , T~'= lim Trt (t). (-U) Thus, according to formula (8), modern climate is some asymptomatic state of the real atmosphere with real boundary conditions and external energy - eources. 'Lt should be noted that the conclusions drawn above do not have a categor- ical nature due to a number of simplifications of the considered problem (in particular, no allowance ia made for the process of energy exc?-�--:ge between waves of different scales). However, it appears that these C-on.- - clugiQns, at least qualitativelq, agree with modern concepts. The enumerated classifications are based on a very important fact: the irr itial data on the state of the atmosphere sooner or later "are �orgotten" and the factors controllino atmospheric behavior over long time intervals will be the external energy sources. 6 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Thus, for developing a long-range weather forecasting method it is neces- - sary that we first discriminate and parameterize those factors which con- trol the processes transpiring in the "earth-atmosphere" system over the course of Iong time intervals. ! L7 9 P Y! FIr Pd! LT d dI dIl - S,1 S.2 S,? 5,1 3,7 5,1 5,2 5,1 �,7 S,? 5,2 5,2 t :Jw oa� oi�� c;~ oa. or� -oa�,c res7~o 'erti;e;ro Qz. ,e,m or~ q:ur~,se.~y 1 _ ;GU' J== � ;Ji 0~ D :OJ~~�~~ j.:w ~ ~0� %-0' ' " Ob 0~=~' :i~ ia 1�`D ' - - - - ~ xti Jb0 S;U 5[U n di ~v 700 6 lA,~ 1s 0 I DmG'FPIIPIDILddIdII1 Fig. 2. Mean monthly vertical water temperature profiles and Qi values. av-t c 012 -012 -0,2 0 D -42 _ 02 0,30 -0,4 'T -H -0,2 -0,41 IX -.T -0, 2 i7D-!X 01f 0.2 p OZ �OES '02 0,,'tlf Y YI ~J � 0, 6 -0, J4 02 0 p jv-P - os -~z D, 56 o,~ L149 �D,? ~ 59 \\0,6 1 1 1 , ~nz 1 ~ 4 6 B 10 z Fig. 3. Cross-correlation matrix of moving averages of the S'(tm) and T'(tnl - +t) values for two months. ~ Comments on Thermal Memory of Ocean In an investigation of large-scale atmospheric processes it can be assuined that the sun is the only energy source. 7 FOR OFFT(:TAT, TTSF nNr v 0 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/48: CIA-RDP82-00850R000200094407-3 rvn vrVi%,.LnL .,..L. The nature of atmospheric-oceanic processes is such that the ocean in the warm half of the year absorbs solar radiant energy, transforms it into thermal energy and "remembers" it, that is, accumulates it in its depths. In the cold half of the year, using different heat exchange processes, the ocean transmits this heat to the atmosphere. In this case the cloud _ cover is the principal regulator with a feedback [3-5, 7]. In order to trace this process we will turn to Fig. 2, where on the basis of data for weather ahip M(66�00'N, 02�00'E) we have shown the mean month- ly vertical water temperature profiles for all the months of 1953. Along the vertical axie we have plotted depth in meters, and along the horizontal axie time t(the months are denoted by Roman numerals) and temperature T in degrees Celsius. "Qp" denotes the heat content of an ocean layer with a thickness 150 m with a temperature conatant with height, equal to 5.2�C. "Qi" (i = 1, 2,.... 12) denotes deviations of the heat content of the men- tioned layer from Qo (heavy shading). An analysis of this figure indicates that an increase in the heat content of the active layer of the ocean occur_a in the warm half of the year, where- _ as in the winter the accumulated heat ia transmitted to the atmoaphere. This can be seen particularly well in the inset to Fig. 2, where the solid curve represents the approxi-mate depth of the homogeneous layer of the ocean, whereas the dashed curve represents the ratio Qi/Q4� For greater clarity we will cite the following well-known example [4]. If the temperature of the upper 100-m layer of the world ocean is reduced by 0.10C ' and it is assumed that all the released heat is expended on heating of the atmospheric air, the temperature of the entire atmosphere to an altitude 30 km ie raised by approximately 5-6�C. - Thus, in developing a method for long-range weather forecasts an acute need arisea for parameterization oE the thermal memory of the ocean. Role of the Cloud Cover of the Oceans The process of absorption of solar radiant energy by the ocean, its trans- - formation into internal energy stored in the warm half of the year and the release of this accumulated heat to the atmosphere in the cold half-year cannot for the time being be parameterized in general form with the accur- acy necessary for the development of long-range weather forecasting schemes - on thie basis. The author of [5, 7] gave a certain phenomenological interpretation of this problem. There, instead of the temperature of sea water, he proposes the use of information on the cloud cover of the oceans received from meteorological satellites. In particular, it was demonstrated that there is a quite close asynchronous correl3tion between anomaliea of summer cloud _ cover over the ocean, averaged for three-month time intervals, and devia- tiona of winter air temperature on the continent from the norm. 8 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL U5E Oi1LL Further investigations indicated that such a correlation also exists with ' two-month averaging of anomalies of the above-mentioned meteorological el- ements [2]. As an example, we will examine Fig. 3, where by means of isc,- carre.lates we have represented the correlation matrix for Leningrad. in computing this matrix on the basis of data from meteorological satellites for the 10 years from 1965 through 1974 we formed the mean monthly values for total cloud cover at the points of a geographic grid 5� x 5� for the part of the North Atlantic bounded by latitudes 40 and 65�N. Using these data we determined the norms for each month in the year and then computed the corresponding anomalies and averaged them for the considered part of the ocean. The monthly temperature anomalies used were those employed in the operational work of the USSR Hydrometeorological Center. iJsing these data we computed the moving mean two-month anomalies of the considered meteorological elements. Along the horizontal axis in Fig. 3.we have plotted the time shift 'C , and along the vertical axis the months of the year. The nature of the field of isocorrelates in this figure is approximately the s.ame as for the case of three-month averaging [S, 71, to wit: the field of isocorrelates is rep- resented by alternating positive and negative correlation regions extend- ing along the mairi diagonal of the matrix. It is easy to discriminate a region of negative correlation between the characteristics of summer cloud cover and winter temperature. Such cross-correZation matrices were computed for each of 31 stations uni- formly distributed over the European USSR. An lnalysis of these matrices made it possible to discriminate the most informative indicators for for- mulating a statistical scheme for predic.ting air temperature anomalies for the European USSR on the basis of cloud data registeYed over the North At- lantic. These prediEtors are listed in the table. One of the important peculiarities of this table is that in the prediction of the temperature anomaly for the cold half of the year the number of predictors invariably includes the cloud cover anomaly for the warm half-year. A prognostic re- gression equation was written on the basis.of this table. The regression coefficients were determined using data from a teaching sample for 10-12 years by the ieast squares method. Then, using data from rin examination sample we computed a series of forecasts of two-month air temperature anomalies for the European USSR. These predictions are usually Punrly suc- cessful with respect to sign, especially in those cases when tllere is a predominance of zonal circulation in the lower half of the troposphere. At the present time the prediction method is in the stage of operatio:al tests. The preliminary results of these tests are encouraging. s Thus, like the earlier published studies [5, 71, the results presented in this article convincingly demonstrate that there is a quite close negative asynchronous correlation between summer cloud cover anomalies over thc, North Atlantic and deviations of winter air temperature from khe norm ii 9 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 the European USSR. And this means that the heat absorbed and rlccumulaled in the active layer of the ocean during the warm half of the year iti rv- tained for a very long time in its depths and is released to the atmo- sphere during the winter months. Predictor Months Using Cloud Cover Anomaly for Temperature Anomalies Prediction months Predictor months .Tanuary-February' April-May ' February-March August-September March-April September-October . April-May September-October rtay-June February-March June-July February-March Ju].y-August January-February August-September June-July September-October . April-May October-November June-July November-Decemter May-June December-January July-August Dynamic-Statistical Approach to the Problem of Parameterization of the Oceanic Thermal Memory On the basis of the results obtained in the preceding sections we will at- tempt to formulate some phenomenolog ical approach to the problem of dynam- ic-statistical parameterization of t he process of the thermal effect of the ocean on the.atmosphere [5]. Bearing in mind the. above-r^ntioned conclusions as a first approximation, we will assume that the quantity of heat Qr( 0,2l, t) accumulated in the active layer of the ocean during some summer month is proportional to the function - N, (P, t) = 1-0,1 S, (H, t), ~ where r is the numbering of the months during the warm half-yeaX, Sr(0, P1, t) is the corresponding total quantity of clouds. It then can be written that C' Qr (H, ~'r t) _ ~r ~vr L) fY, (0, Xe t), (11) are some still unknown functions to be determined and which are where , r assumed to be quasiuniversal. It is,obvious that y., when Sr = 0/10 Q~ - ^ ' i o when Sr = 10/10 , , 10 FOR OFFICIAL USE ONLY \ APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY It evidently can be postulated that the quEntity of heat Qq released by the ocean into the atmosphere in the q-th winter month consists of the quantity of heat accumulated in the active layer of the hydrosphere dur- ing each of all the summer months. The contribution of each month of the warm half-year must enter with its weight. Now we will assume that as a teaching sample we have an archivea containing observational data for yeara. For each pair of succeseive warm and cold halves of the year belong- ing to this sample we have, on the basis of (11) Q�%C/ N' (12) Accordingly, the heat influx equation for predicting air temperature (with stipulated initial and boundary conditions and known horizontal velocity components) of the q-th winter month, with (12) taken into account, can be written in the form dT vd dT uk dT k dt + a d b + n sin 8 d?~ - nT ~ T= X, N,. Q (13) r For convenience in exposition, we will call the functions Y r( e, the asynchronous influence functions. ~ Our objective is a determination of elements of the matrix 1f xr(~, A)II functions for each pair of numbers r c y~ and q c yl possible for real at- mospheric-oceanic conditions. Equation (13) describes the change in air temperature caused by the thermal effect of the ocean on the atmosphere, advection and horizontal macrotur- bulent exchange. A distinguishing characteristic of this equation is that its right-hand side was written in a form with a"lagging" argument and its computation requires the availability of data on cloud cover only for the times which have elapsed (that is, cloud cover need not be predicted). - Since our objective is a determination of the asynchronous influence func- - tion )Cr, therefore, assuming that in equation (13) the functions T, vg , vIN _ and Nr are known, and x r are to be found, we denote its right-hand side by Fq( Q, ,X) and rewrite it in the following form: ~ N, F`i. r Hence the asynchronous influence functions )tr can be determined by differ- ~ ent methods. However, it seems that in this case the spectral method is the most natural. In this case the problem essentially involves solution of a system of linear algebraic equations for the expansion coefficients of the sought-for functions [5]. The degree of practical applicability of the asynchronous influence func- tions determined in this way can be Judged only after carrying out exten- sive diagnostic and prognostic experiments during which there should also be a atudy of the nature of the temporal variability of these functions. 11 RnR nl?FTrTeT rTeV rnvrv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 CUK Vrrl%..LnL u.._ Uaing the proposed method for parameterization of the oceanic thermal mem- ory it ia evidently impossihle to study atmospheric-oceanic proceBSes with any detail. However, we hope that on the basis of such parameterization it will be possible, in a rough approximation, to describe the principal large-acale procesaes of the thermal effect of the ocean on the atmosphere during the cold half of the ,year. - - BIBLIOGRAPHY 1. Blinova, Ye. N., "Hydrodynamic Prediction of Mean Monthly Temperature Anomalies for the Earth's Northern Hemiaphere Using IGY Data, DOKLADY AN SSSR (Repo-ts of the USSR Academy of Sciences), Vol 131, No 2, 1960. 2. Zadorozhnaya, T. N., "Evaluations of the Correlations Between Cloud Anomaly Fields Over the Ocean and Temperature Anomalies Over the Con- tinent With a Two-Month Averaging Interval," TRUDY GIDROMETTSENTRA SSSR (Transactiona of the USSR Hydrometeorological Center), No 192, 1977. 3. Marchuk, G. I., Musayelyan, Sh. A., "Methods for Computing Variations of the Total Flux of Radiant Energy for the Purposes of Long-Range Prediction of Large-Scale Meteorological Fields," METEOROLOGIYA I GIDROLOGIYA (Meteorology and Hydrology), No 8, 1974. - 4. Monin, A. S., "Physical Mechanism of Weather Changes," METEOROLOGIYA I GIDROLOGIYA, No 8, 1963. 5. Musayelyan, Sh. A., 0 PRIRODE NEKOTORYKH,SVERKFIDLITEL'NYKH ATMOSFER- NYKH PROTSESSOV (Nature of Some Super-Prolonged Atmospheric Processes), - Leningrad, Gidrometeoizdat, 1978. 6. Lorenz, E. N., "The Predictability of Hydrodynamic Flow," TRANS. NEW YORK ACAD. SCI., Ser. 2, Vol 25, 1963. 7. Muasaelyan, Sh. A., "The Use of Cloud Cover Satellite Information for Quantitative Long-Term Forecasting," PROCEEDINGS OF THE SYMPOSIUM ON i ME'rEOROLOGICAL OBSERVATIONS FROM SPACE, Philadelphia, Pennsylvania, ` USA, 8-10 June 1976 (19th COSPAR). ; 12 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 s FOR OFFICIAL USE ONLY UDC 551.(58+576.2)(100) ZONAL DISTRIBUTION OF CLOUD COVER OVER THE EARTH Moscow METEOROLOGIYA I GIL.^.^LOGIYA in Russian No 3, Mar 80 pp 15-23 [Article by Doctor of Geographical Sciences T. G. Berlyand, Candidate of Geographical Sciences L. A. Strokina and L. Ye. Greshnikova, Main Geophys- ical Observatory, aubmitted for publication 11 September 19791 Abstract: A atudy was made of the latitudinal diatribution.of the quantity of clouds on the basis of new global monthly maps of cloud cover compiled on the basfs of data from the world network of ineteorological stations. It is shown that in both hemispheres the mean quantity of clouds over the oceans is greater than over the land (by approximately one-tenth). The aonthly quantity of clouds for the earth as a whole is virtually constant, regardless of season of the yeaf. The results are compared with similar data from earlier studies made for climatological generalization of cloud cover distribution. [Text] Status of problem and data used. The study of characteristics of _ the global cloud cover regime is now attracting considerable attention in connection with work on estimating the earth's energy balance, model- ing of climate and its changes and also solution of many other timely _ problems of a scientific and practical nature. - One of the first climatological generalizations of cloud cover data was made by Brooka, who in 1927, on the basis of a relatively limited volume of obaervational data obtained the zonal distribution of the quantity of clouds over the oceans, land and the earth's surface as a whole [9]. Af- ter this study the zonal characteristics of cloud cover were virtually not investigated for a lonq time. Only after three decades did K. Tele- gadas and J. London [10] compute the zonal values of the quantity of clouds over the northern hemiaphere for two months (April, October) and two aeasons (winter, summer). In characterizing cloud cover over the oceans they employed data from the atlas of climatic maps of the oceans 13 Fnu nFFTrrer TrcL' n*rrv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 L'VL\ VL'aiva..u+ prepared by McDonald [12]. The results of these studies have found use in a number of climatological investigations, in particular, devoted to the modeling of climate. During the last 10-15 Veare the volume of cloud cover information has in- creased considerably as a result of the increasing series of observations, the implementation of international geophysical programs, determining the expansion of the ground network of stations in poorly studied regions, and also the launching of ineteorological earth satellites. This facilitated ~ the further atudy of the patterns of spatial-temporal distribution o� cloud cover. Using data from two years of observations from a satellite (1965-1966), J. Sadler [14] obtained the zonal distribution of the quantity of clouds for the tropical zone (30�N-30�S). - H. van Loon [18] used both ground and satellite information on cloud cover. On the basis of a generalization of [9, 14 and others] he investigated the characterietica of the latituclinal distribution of the quantity of cloada in the southern hemiephere for Jaizuary and July. In the fundamental publication ATLAS OF THE OCEANS [2], recently published in the USSR, there is a series of monthly cloud cover maps of the Pacific, Atlantic and Indian Oceans. They were compiled using data from regular and expeditionary observationa on ships over a long period of time (from the end of the last century through 1965-1970). Z. M. Makhover and L. A. Nudel'man [1] prepared an atlas containing a con- siderable number of different climatic characteristica of cloud cover for the southern hemisphere. In the course of preparation of the atlas maps they generalized for all montha of the year an extenaive volume of daily data from ground observations for a five-year period (1967-1972). Data on the distributior, of the quantity of clouds over the earth's surface, except for the equatorial zone (13�N-5�S), are cited in studiea by C. Shutz and W. Gates [15, 161. They were prepared for the purpose of obtaining init- ial material for investigations for the modeling of climate. Aa the charac- teristics of the cloud cover regime in the northern hemisphere the authors used the results of surface and satellite observations. The latter were made only at two times (midday and midnight). For the southern hemisphere the data were taken from the mentioned study by van Loon [18j. They also drew upon the reaults of generalization of a four-year series of observa- tiona (1967-1970) from a satellite, taken from the atlas published by D. Miller [13], relating only to the interval 1400-1600 hours local solar time and characterized the cloud cover conditions for a limited part of the day. The use of inadequately uniform observational data on cloud cover from the point of view of inethods for collecting information concerning it, the ob- servations made once or twice during the day and the relatively short (not 14 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL IrSE ONLY - alway$ corresponding to climatological requirements) obaervation periods all this could not but exert an influence on the accuracy of the valuea and completeness of the characteristica in the above-mentioned atudies. World monthly maps of the dietribution of the total quantity of clouda based on more uniform inaterial, covering a long period of observationa, were com- piled by T. G. Berlyand and L. A. Strokina. Some of these maps were publish- ed in 1975 [3]. The initial data used in their compilation was the informa- tion included in numerous handbooks, monographs, atlases, monthly summaries and yearbooks published by the meteorological aervices o.f different coun- tries. The results of modern investigations of the cloud cover regime in individual regions of the earth were also taken into account [4, 5, 7, 8, 17, 191 in this investigation. In the course of analysis of the various - initial data, in addition to surface data, use was made of data rrom sat- ellite obaervations of cloud cover [6, 13, 14]. In compiling monthly maps - of the quantity of clouds on the contine*ts the authors of [3] used data for about 3,500 points, of which 950 were for Europe, 1,150 for Asia, 450 for Africa, 300 for North America, 460 for Central and South America, 140 for Australia and Oceania and 30 for Antarctica (we note that in the men- _ tioned study by Brooks data for a total of 1,000 stations were uaed for _ compiling such maps [9]). At most stations the observation perioda exceed- _ ed 30 yeara, whereas for etations situated in the low latitudes from 15 to 20 yeare. Z'he distribution of cloud cover over the oceans was determin- ed for the most part uaing data from the MARINE CLIMP,TIC ATLAS [17], which contains data on the probability of the total quantity of clouda for indi- vidual regions characterizing the principal climatic regions of the oceans. For the area of the Atlantic Ocean the initial values were given for 93 regions, for the Pacific Ocean for 111 regions, for the Indian Ocean for 48 regions. This article is devoted to a further development of [3]. In that paper, on the basis of data taken from new world monthly maps, the authors computed the zonal values of the quantity of clouds for the continents, oceans and the earth's surface as a whole. Computation of Zonal Characteristics of Cloud Cover The mean latitudinal cloud quantities were determined by averaging the val- ues for 35 circles of latitude 30i = i-0, 5, 10, 15,...,85� with a longitude interval 5�. A total of more than 30,000 values were taken from all the monthly maps; these characterize the quantity of clouds at the intersec- tions in a regular 5� grid. - The mean value for a 10� i-th latitude zone, taking the oceans and contin- ents separately into account, was determined using the formula Ni _ 5 Nt ,vl+5 0.5 cos Tt_ 5~ ny_S) I + cos yr ~ n jj + 0, 5 cos v;~; ~ n0+s> ! > > I n! 0,.5 Ni-cos x_ N cos Y' 0 ~ s t s i i + ,5 Ni+5 co~ ft+5 15 FOR OFFICIAL USE ONLX APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 v..,.. l P Ult ur r i%. icil. U.._ where nij is the quantity of clouds at the j-th point of intersection of a regular grid at the circle of latitude ~Pi; Ni is the number of points of grid intersection at the circle 5pi, the subscript i= i0, 5, 10,..., - 85. The zonal value of the quantity of clouds over the earth's surface was com- puted as the mean weighted value of the corresponding data for the oceans and continenza, taking into account the areas of the latter. The results are given in tables accompanying the paper. The data in the table make it poeaible, more completely than up to this time, to ascertain the peculiar- ities of the latitudinal and annual variation of the quantity of clouds. over the earth's : urface. Tables 1 and 2 give the ar,nual variation of the zonal values of cloud cover in tenths over the land, oceans and for the earth as a whole respectively. Table 3 contains data on the mean monthly quantity of clouds for individ- ual continents and individual oceans. Analytical Reaults Table 1 indicates that during the course of the year both for the land and for the oceans in both hemispheres an identical latitudinal variation in the quantity of clouds with maxima in the temperate and equatorial lati- tudes and minima in the tropics is characteristic. The difference between the extremal values sometimes exceeds 3/10. The position of the maximum and minimum values does not remain constant during the course of the year. In each hemisphere during the warm half-year they shift in direction toward the higher latitudes; during the cold season the shift is toward the low latitudes. Thie ehift ia quite great, being 10-15� in latitude. The great- Pst withdrawal of the maxima of the quantity of clouds in the direction from the equator to the north is observed in the period from May through September; in the opposite direction from November through March. Ztaice a year (April, October) the extrema are situated in both hemispheres al- most symmetrically relative to the equator. The results of this analysis of the mean monthly cloud quantities for each of the 10� zones for the most part show that during the course of the year there is one maximum and one minimum. A maximum in the warm season and a minimum in the cold seaeon is the most commonly observed type of annual variation. It is characteristic of latitude zones situated above 50� in both hemispheres; with advance toward the polea there is an increase in the annual amplitude. The amplitude attaina its maximum values (4/10) over the north pole; this amplitude is twice as great as ovEr the sour', pole. The noted type of annual variation ia also characteriatic of the equatorial zone (20�N-20�S), where the annual amplitude is about 3/10 over the land and 1/10 over the oceans. 16 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/48: CIA-RDP82-44850R000200094407-3 FOR OFFICIAL USE ONLY 5ap~�~ Tenths . ~ � 8 a) . , y\~ . . . , . 7 / v r r 6 `J r \ I /x.11 ~l 1 / , S 1 ~ ~ w 9- ~ ~ ~ 6) B - ^ . ~ ' : Y/ : i�''.' . i / �~r , . S - ,�f } i ~i � 1 - 2 r j e-a9 ---5 Y .~6 7 ~ ~ ~ ~ ~ ~ ~_-1-_----�_�~.-=~ ~ ~ ' 90i 70 SO 747 0 10 70 g Fig. 1. Zonal distribution of cloud quantities according to data from dif- ferentauthora, a) January; b) July; 1) C. Brooks; 2) J. London; 3) J. Sad- ler; 4) H. ve.: Loon; 5) C. Shutz and W. Gates; 6) Z. M. Makhover and L. A. Nudel'man; 7) T. G. Berlyand, L. A. Strokina and L. Ye. Greshnikova 17 _ FOR OFFICIAL iTSR ONi.Y APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/48: CIA-RDP82-44850R000200094407-3 r Vn vr r i%,, w.,.. Table 1 Annual Variation of Cloud Quantities LUtipotal ( [ I II I !II I !V I V I VI I V11 I Vllll i`C I 1 I XI I XII I fo;t G 7 Han cyureA 80-70�c2 5,1; 6 3 ~i,fi 6 1 5,1 5 9 I 5,7 6 2 7,0 7 1 7,2 7 1 7,3 7 1 7,2 7 3 7,7 7 8 7,~4 7 8 6,3 9 6 5,3 6 5 G,;~ 8 6 70-60 . , , , , , , , , , , , , 60- b0 6,3 G,l 6,1 6,5 6,7 6,7 6,6 6,6 6,8 7,1 7,0 6,7 6, 6 " .50-40 5,7 5,8 5,9 6,1 5,9 5,5 5,0 4,7 4,7 5,1 5,8 6,1 5.5 40-30 5.1 5.0 5,0 4,9 4,5 3,9 3.3 3.7 3,4 3,5 4,0 4,7 4,3 30-20 3,7 3,6 3,7 3,6 3,6 3,7 4,0 4,2 3,7 3,0 3,4 3,6 3,7 20-10 3, z-, 3,6 3,8 4,0 4,6 5,2 6.0 6,2 5,6 4,1 3,9 3,6 4,1, .10-0 5,4 5,6 6,0 6,6 6,7 h,% 4,9 7,1 6,7 6,0 5,8 5,1 6,1 0-10� 103 7,2 7,3 7,3 6,8 6,4 5,6 .5,0 5,4 fi,0 f;,2 6,8 7,0 6.4 10-20 6,4 6 ,3 6,0 5,4 4,0 3,~ 3,1 3,0 4,2 :~.2 5,7 5,3 4,~+ 20-30 4,5 -a,i, 4,3 3,3 3,5 3,3 3,0 2.7 3,2 3,9 4,1 4,2 3,8 30-40 11 3 3 4 7 3 3 8 4 7 4,8 4.7 3,9 3,9 4,2 3,8 3,8 4,1 40-50 , 4,!) , 3,3 , 4,9 , 5,4 , 6,6 6,1 5,8 5,8 5,9 5,3 5,9 5,8 5,6 50-GO 60--70 7,1 6,7 6,7 6,5 5,8 5.7 5,9 6,4 6,4 6,4 6,2 6,3 6,3 70--80 6,2 6,3 5,4 5,7 5,0 4,9 �`t 0 5,7 5,8 6.2 5,8 6,2 5,7 60-90 5,4 5,7 5,9 4,5 4,2 4,1 3,9 5,0 5,0 5.U 5,4 5,7 5,0 'Ceecpnor 4 nonywapac 5,2 .5 .:i 5,2 5,4 5,6 5,5 5,6 3,6 5,5 5,3 5,3 5,2 5,4 .IOHaioc 5 no7yiuapF+e 5,9 5,9 5,8 5,3 4,8 4,4 4,2 4,4 4,9 5,3 5,4 5,6 5,2 Cywa n .ucnom 5,4 5,4 5,4 5,4 5,3 5,2 5,1 5,2 5,3 5,3 5,4 5,3 5,3 6 g Ha1c oxeauaMjs 90-80�C 5,3 5,2 5,3 5,7 7,7 8,5 3.7 9,0 I US ~ 8 5 8,0 1 8 6,2 5,4 ' 6 8 6 0 7,1I 1 7 80--70 6,0 5,9 5,7 6,3 7,7 8,0 8,1 8,4 , , , , , 70-60 7,0 6,3 6,3 7,0 7,6 7,7 7,7 7,6 3,U 7,8 7,4 7,1 7.4 -60-50 7,6 7.5 7,5 7,7 8,1 8.2 8,3 8,1 7,8 7,6 7,7 7,7 7,8 50-90 7,5 7,4 7,3 7,5 7,7 7,9 7.7 7,4 7,1 7,0 7,2 7,4 7,4 40-30 6,5 fi.r, 6,5 6,5 6,6 6,6 6.1 5,7 5,0 ~ G,I 6,2 6,4 6,3 30-20 J, r) l1,3 5,3 5,3 5,3 f,,[; 5,7 5,3 :i,'Z ~ 5,4 5,3 5,5 20-10 .5,2 5.0 :,U 5,3 5,4 6,1 6,3 6,1 (i,Q I 5,8 5,4 5,4 :),G 10�-0 5,8 Fi,6 53 5,9 6,1) 6,3 G,? 6.1 6,0 (>,O 5,$ ri.9 5,51 0-10� w 5,9 5,7 5.6 n,6 5,3 5,2 5,3 ~i,3 5,6 F,,fi 5,7 5,8 ' 5,& 10-20 5,7 i,5 5.6 5,3 M 5,0 ;5,3 :>,4 5,5 I 5,7 5,8 .`i,i ;r,.`, 20-30 5,G 5,5 5,7 5.6 :i, 5 5,4 5,6 5,6 5,8 6,0 .5,> (P, ~t.) = 3 K~; i:~ j)Z (1 2 n'� nt I''R ri P 1 ' (4) K (r, !t.) - � G Pn Kis (P, ~t) = j nn~K -}-2(1 - f ~n)~-~- (Yt - 1) g ta-n~ l where K(P) is the Van de Hulst function. ' > nrt' h~P~ 2 sin c sin-' ~ 1 - P ' ~ o= 4 r is a dimensionleas parameter) 30 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY (n)~ 1,05; n 1,5: C~ Thie function is correct ior any real values of the refractive index of perticlee. The mean appruximation error is 4% for a refractive index 1.5. Failure to take abeorption into account in K(p , n) gives errors oF a few percent for polydieperse media. The particle size function was atipulated in the form of a three-parameter gamma function f (r) = Rr'` e-bi (;L b (5) r which with a broad change in the b and i-L parameters describes well the diverae apectra of particles observed in the atmosphere. With (4) and (5) taken into account, the expression of the scattering function for the adopted model has the form [6] Nt~�~= 47tN + 1)(FL -F2) y�.ta b2 2 4' Y" (Et + ~ + y=) r) (6) -f-1/1 t y` cos (;.L + 1) ~1 1 + 2 b2N (IL+ 1)(lL -f- 2) X r n l~n - i X L 1 f(n) c�ty 1' ~ b n ~ p Y = 2=; C=1- 8nb x; x= r An analytical model of the volume scattering coefficient (6) can be used in determining the mass concentration and parameters of an aerosol in the interpretation of data from laser sounding. It makes it possible to reduce the problem of determining the micrnstructure of an aerosol to the optimization problem in parameter space by means of minimizing the functional k F r, N) _ ~ [~i Z) - P; (~�)1', (7) constituting the sum of the squares of nonclosures of the differences in the measured values of the scattering coefficients and the corresponding valuea in the model. In principle, for determining the four parameters of the diatributiion it is necessary to know the measured scattering co- efficients at four wavelengths. However, taking into account that the 31 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 variations of thP refractive index in the range from 1.3 to 1.7 lead to a change in the scattering coeff icients A) by not more than 10-15Y [10], with an accuracy in measuring )8(/A, z) of 20% the value of the refractive index can be selected taking into account the specific condi- tions of the investigated object, additional measurements of compoai- tion or taken as a mean. In other words, with a f ixed value of the re- fractive index in determining the aerosol parameters it is possible to limit ouraelves to three wavelengths. Having apecific experimental data on fi(,;1, z), uaing an electronic computer we obtain the numerical values of the selected P (a) model. The end result of the calculations is evaluations of the model parameters with a atipulated accuracy, deter- mined by the sum af the squarea of the difference in the values of the parameters i:, two successive iterations. With a fixed value of the re- fractive index n, the accuracy of the method gives an error in determin- ing the parameters of 3-5X [lO]. An assumption concerning the particle- size diatribution function does not limit the possibilities of the model employed because using the variations of r and � it is possible to obtain the most diverse particle distribution spectra. In order to obtain multi- modal spectra it is necessary to have measurements at a great number of wavelengths. Analyais of Inversion Results Tiue to the fact that two wavelengths were used in our experiment in sound- ing the plume, our method makes it possible to determine two parameters of the model. In this case the most interesting parameters of the plume should be the mean particle size and the numerical concentration. Precise- ly these parameters were determined in this experiment. Two other para- meters distribution width � and the refractive index of particles ft were selected from the accompanying samplings. The results of the samplings cited in Table 1, on the basis of the dispersion value, made it poseible to assume that the investigated aerosols have a quite broad size spectrum and therefore the � value should be selected equal to 0 or 2. The value of the real part of thp refractive index for particles of zinc, iron and copper, according to the Ivlev model [4] and cited in the book by Van de Hulst, can vary in the range 1.40-1.47. As mentioned above, such changes in n cannot introduce significant errors into the determination of mass concentration. - In additiony for work with the optimum parameterization method using an electronic computer it ie necessary to select two initial points contain- ing the surmised sought-for parameters n, the mean particle size r, � numerical N or mass M= N b 4/3110 concentration, where b is aerosol den- aity. In our experiment these data were also taken from microphysical _ measurementa. We note that in principle we can select any reasonable val- - uea of the parameters as the initial points. Their choice influencea only the convergence of the problem when seeking the minimum of the F function and the expenditure of computer time. In order to stipulate these points it is not at all mandatory to carry out simultaneous microphysical meas- urements, but with the availability of the latter the choice of initial points is simplified. 32 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 ; FOR OFFICIAL USE ONLY Thus, as a result of inversion of equation (3) by the optimum parameteriza- - tion method we obtained evaluations of model parameters as indicated in Table 2. Table 2 Evaluation of Aerosol Parameters by Optimum Parameterization Method � n r.41 KM _ I N c.u-3 I I eixa/M~~ ~,if % �m 2,0 1,40 0,20 5,0� 10' 3,2 S~0 0 1,47 0,50 3,7�107 74 40 2,0 1,44 0,58 5,7�107 985 50 Table 3 _ Values of Volume Scattering Coefficienta and Aerosol Parameters for Pure Atmosphere . - [IapaMer- I Afn.+t P(X. z) ceH Pb. �pac- i _ rtpe2e.7e- 1 HHA 2 Z , ~ � 0,.532 4�10-6 7-0,22 790 10 ' 0,694 3�10-6 �=2.0 1,064 1,7 � 10-6 n=1,33 0,532 2,1 � 10-e i=0.20 530 20 0,694 1,5�10-6 li-2,0 - 1,064 7�10-7 rr=1,33 KEY: - 1. �-m 2. Distribution parameters The magnitude of the error in determining the mass concentration !S M(with a density S= 2 g/cm3) was determined from L�he minimum nonclosure L1F, which characterizes the minimum deviation of our model p T(a) from the set of experimental data. - The mean radius of the particles falls in the region 0.3-0.6 � m and varies in dependence on the spectral variation of the scattering coefficient [2]. The concentration was equal to 5�107 cm 3. The mass concentration with a 50% error was 75 � g/m3, which corresponds to data from the sampling. TiLle 2 shows that the parametera most sensitive to variations of experimentai 33 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 data are N and the mean size of particles, and hence, to a high degree, also the mass concentration. Such high errors in determining M are at - tributable to this circumstance. For example, the inversion of the volume acattering coeff icients (Fig. 3) for a case when there were ac- companying,aircraft sa2plings with stipulated values 0, n= 1.47 made it possible to obtain r= 0.5~xm and a value of the mass concentration M = 74 � g/m3. The minimum value of the mass concentration, corresponding to the background value A (;1 , z) = 5�10-4 cm-1, is 2.6 �10'2 j.L. g/m3, whereas the concentration N is 1.3�104 cm 3. The maximum M value is 400 � g/m3/m3 and was observed with A(z) = 9.0 cm 1. The computed value of the scattering coefficient in the model for the assumed particle distrib- ution with N= 1 cm 3 is P(9l) = 3.77�10-8 cm 1f1.5�10-8 cm 1. The accur- acy in determininq the mass concentration in this case is 40%. ,a "r -",).o.u2 Ji't0 JIBO J6: U 3Bo: [ M Fig. 4. Dependence of the volume scattering coefficient on depth of pene- tration. Similar computations were made for other series of laser sounding of the plume; these are shown in Fig. 4. The mean values of the distribution and mass concentration parameters for these series are given in Table 2. As an illustration of the method described above we wi1Z give an example of inversion of the results of three-frequency laser sounding of the frAe atmosphere at Leningrad carried out in 1978. In this experiment we used an Al-Na garnet laser ( T= 1.06�.m and 0.532�-m) and a ruby laser (A _ 0.69 �.m) with a common receiving antenna designed in a Cassegrainian sys- tem with a diameter of the large mirror 0.26 m. Table 3 gives data on the composition of aerosol obtained with values of the meteorological range of visibility 10 and 20 km. 34 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY - 1'he refraction coefficient for the pure atmosphere was assumed equai to 1.33. The use of three sounding wavelengths made it possible to evaluate three aerosol parameters. The n value was stipulated, taking the experi- mental conditions into account. The determined values of tlie concentration N are in agreement with data in the literature for the free atmosphere [10]. Conclusion The results of this investigation of aerosol in smoke plumes using two-fre- quency laser sounding and the employed method fcr inversion of the volume scattering coefficients indicated that the mass concentration of particles obtained by this method agrees well with the results of sampZing. This confirms the considerable possibilities of remote sensing and monitoring of amoke plumes using laser techniques. In the case of three-frequency sounding the mass concentration can be obtained in the absence of ac- companying microphysical measurements. T~ao-frequency sounding also can be ~ used without accompanying microphysical measiirements if the element com- - position of the effluent is known. With an increase in the number of sounding frequencies the method describ- ed here makes it possible to obtain multimodal aerosol distribution spec- tra. Taking into account that enterprises of the same class give close spectra ' of aerosol particles in effluent, for solution of ~ypical problems in mon- itoring effluent it is sufficient to obtain statistically representative characteristics of effluent for individual branches of industry. This will make it possible ta organize remote routine monitoring of effluent by means of laser systems. BIBLIOGRAPHY 1. Badayev, V. V., Georgiyevskiy, Yu. S., Pirogov, S. M., "Aerosol At- tenuation in the Spectral Region 0.25-2.2 � m," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA (News of the USSR Academy of Sciences, Physics of the Atmosphere and Ocean), Vol II, No 5, 1975. 2. Zhitkov, L. V., Kaul', B. V., et al., "Use of a Lidar for Investi.gat- ing the Dynamics of a Smoke Plume,"TRUDY 5-go VSESOYUZNOGO SIMPOZIUMA PO LAZERNOMU T AKIJSTICHESKOMU ZONDIROVANIYU ATMOSFERY (Transactions of the Fifth All-Union Symposium on Laser and Acoustic oi = the Atmosphere), Tnmsk, 1978. - 3. Zuyev, V. Ye., et al., "Optical Experiment and Results of ::nversion of Data from Multifrequency Laser Sounding of the M4 crostructur2 of 'Aerosol in the Surface Layer," PROBLEMY DISTANTSIG`NNOGO 7.ONDIROVANIYA _ ATMOSFERY (Problema in Remote Sensing of the Atmoapheie), Tomsk, SO AN SSSR, In-t Optiki Azmosfery, 1976. 35 Vnu nVVTrrnT rrcV nlknv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 rvn vrri%.iti1j u.... ....L~ 4. Ivlev, L. S., Popova, S. I., "Complex Refractive Index of Matter of - the Disperse Phase of Atmospheric Aerosol," IZV. AN SSSR, FIZIKA ATMO- SFERY I OKEANA, Vol 9, No 10, 1973. 5. Kovalev, V. A., "One Method for Pracessing a Laser Signal," TRUDY GGO ('I'ransactions of the Main Geophyeical Observatory), No 312, 1973. 6. Martynova, V. I., "Polydisperse Scattering Coefficient for the Mie Approximation Function," TRUDY IPG (Transactions of the Institute of Applfed Geophysics), No 17, 1973. 7. Nazarov, I. M., Nikolayev, A. N., Fridman, Sh. D., DISTANTSIONNYYE I EKSPRESSIVNYYF METODY OPREDELENIYA ZAGRYAZNENIYA OKRUZHAYUSHCHEY SREDY (Remote and Express Methods for Determining Environmental Con- tamination), Mnscow, Gidrometeoizdat, 1977. 8. Pakhomov, P. V., Martynova, V. I., "Mie Approximation Function for the Factor of Effectiveness of Scattering," TRUDY IPG, No 17, 1973. 9. Rozhdestvenskaya, V. I., "Determinatiun of the Concentration and Para- meters of Aerosols from the Results oi Tailight Measurements," TRUDY IPG, No 32, 1977. 10. Rozhdestvenskaya, V. I., RASEYANIYE SVETA NA POLI7I:iPERSNYKH SISTEMAKH _ AEROZOL'NYKH CHASTITS (Light Scattaring on Polydfsperse Systems of Aero- sol Particles), Moscow, IPG, 1977. 1 I I ~ ' I ~ 36 ~ FOR OFFICIAL USE ONLY ~ ~ APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY UDC 551.509.616 CHANGE IN MICROSTRUCTURE OF STRATIFORM CLOUDS UNDER INFLUENCE OF CONDENSATION NUCLEI Moscow METEOROLOGIYA I GIDROLOGIYA in Ruasian No 3, Mar 80 pp 33-38 (Article by K. B. Yudin, Institute of Experimental Meteorology, submitted for publication 19 Suly 19791 - Abstract: The article gives the results of f ield experiments for the modification of warm strati- - form cloud cover by artificial hygroacopic eon- densation nuclei. The author gives some charac- teristics of a modified cloud medium and describes the nature of restructuring of the spectrum of - cloud element sizes ae a result of modification. [Text] The problem of artificial modification of warm clouds and fogs is of great scientific and practical interest. One of the objectivea of - artificial modification is an improvement in visibility in clouds and foga. Their total dissipation ie not always necessary. Zt follows from the theory of light attenuation of dispersive media [10] that it is pos- aible to improve viaibility by a restructuring of the microstructure of a cloud or fog. After Yu. S. Sedunov [6] proposed a method for the mod- ification of microstructure by the introduction of additional finely dis- - , perse condensation nuclei directly prior to the onset of cloud formation, aseries of theoretical studiea [1, 2, 5] was carried out in which a study was made of the effect of nuclei with different characteristics. The , experimental studies carried out in an aerosol chamber [3, 8] indicated that the considered method affords a possibility for modification of the _ process of forming of the droplet spectrum. A merit of the method is a high degree of use of the reagent, attributable to the fact that in the - proceas of formation of the microstructure the total mass, which can be extremely small, participates, and to a certain degree determines its - large number of condensation nuclei. The ob,jective of this study was a field investigation of the regularities in changes in the microstructure of the cloud medium with introductian of a reagent. Artificial hygroacopic particlea were introduced into the cloud near the condensation level. They must be entrained into a cloud by ascending flowa, and competing with natural nuclei, change the course 37 LV1D A'VrTl,T AT T7Vn n*rr V APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 L'VIl VL'l lVlau~ v.. . v.�..~~ r-1 d ~ ~ E~ Of N Q~ ar tD u 1+ G! ~ cd rl r-I a w 44 V o~o ~ ~ 7 G! U Gl O tl1 -rl y p 0 ~0 ,~C G b J M Gl ,W U P 0 i4 MH N y H ~O h o0 ON O r"i ~ ~ ~ ~ a~ ~d ro ~ n a o w -W Q4 ~5 q ~n O ~ Ir ri � cNd -W O p ~z � � � rlNM~t V1 ~ APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY the modification zone can be represented as an elliptical cylinder with a horizontal axis of 7-8 km. The lengtha of the axes of the ellipse, lying at the base of the cylinder, 5 minutes after modification were 600 and 200 m, and after 10 minutes 1,200 and 400 m. The mean concentration of reagent after 5 and 10 minutes was 20 � g/m3 and 5 N.g/m3 reapectively. However, the reagent, diffuaing in the cloud under the influence of dynamic turbulence, doee not fill the entire volume uniformly and at individual points of ita concentration can differ coneiderably from the mean values. For the mon- itoring and regietry of changes in microstructure the aircraft repeatedly returned to the modification track at an altitude exceeding by 20-30 m the altitude at which the reagent is introduced. The flight trajectory, the moments of entry into the modification zone and emergence from it were regiatered by a ground radar. The time of presence of the aircraft in the modification zone was about two minutes. The principal instruments for investigating microstructure and its changes were a photoelectric instrument for measuring the sizes and concentration of cloud droplets in the range from 0.5 to 401km [4] and an instrument for - measuring aerosol scattering in clouds. During presence in the modification zone 100-120 microstructure samples were taken for analysis. The number of droplets in the sample varied from 500 to 6,000, depending on their concen- - tration in the cloud. The background data on the state of clouds were col- lected in regions adjaCent to the modification zone during aircraft man- euvers for return to the modification track. - Table 1 gives data on the synoptic conditions under which the experiments were made and some integral characteristics of microstructure computed from the measured apectra. A preliminary analyaie indicated that in the modification zone there is an increase in the concentration of hydro- meteora and aerosol scattering. However, there is no direct proportional- ity between theae changes, which indicates a restructuring of the spectra. In order to explain the nature of this restructuring each of the measured spectra was divided into two parts water-enveloped nuclei and dropleta [7]. The part of the spectrum containing water-enveloped nuclei was de- scribed by a power-law diatribution in the form Y'' f (r)=A ,~where A is a normalizing factor numerically equal to the value of the above power function with r= rp = 1 � m. The A and -V parameters were computed by the least squares method. The-parameters of the distribution function, de- - scribing the droplet part of the spectrum, were computed from the modal values of the radius and distribution function and the liquid water content � value, computed from the measured spectrum. Table 2gives the parameters of the distribution function (and not the parameters of the mean spectrum), averaged from individual spectra, and their variability in background meas- urements and in the modif ication zone 5-7 minutes after introduction of ttie reagent. The last four lines in the table generalize data on change in the parameters as a result of modification and on the significance levels f,)r theae changes, evaluated using the Bartlett and Fisher tests. 41 - FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 An analysis of the table indicates that as a result of introduction of the reagent there was an increase in the concentration n of water-encased nuclei, that is, most of the additional nuclei were encased in water, but could not form droplete. The droplet concentration m and the aerosol acat- tering coefficient o4increase somewhat. The changes of the exponent in the inverse power-law diatribution -V, the mean radius r of drop.lets and the relative dispersion of the droplet part of the apectrum p,/r are in- significant at the level of the natural inhomogeneity of clouds. A sub- stantial increase in fluctuatione of some parameters ia alao observed in the wake of the effect. f!r) Nxrf I� cn"i J000 - a) 1000 � \ J00 ~ 100 � JO � 10 ~ 45 1 2 S 10 0,5 b) Z \ - - J \ \ \ l . ~ J 1 r M,Y,y t ~ Fig. 1. Spectra of size of hydrometeors before modification and 5-7 minutes after cloud modification. a) background, b) wake of modification; 2) in etratocumulus cloud, 3) in haze under clouds, 5) in stratiform cloud The additional information on restructuring of the spectra is shown in the figure. The numbers on the curves in this figure correspond to the numbers of the experiments. The figure shows that in the modification zone the value of the distribution fu:,::tion in the region 0.5-0. 7 � m increasea by a factor of 2-3, there is a deepening of the "dip" between the parts of the spectrum, and droplets which are larger than those observed in background measurements appear. However, droplets capable.of causing coagulation (r ) 20pL,a) have not been regiatered. Our work demonstrated that the introduction of additional artif icial con- densation nuclei somewhat modifies the microstructure of clouds. Under natural conditions the process is traced for 13-15 minutes after intro- duction of the reagent. At the end of this period the concentrations of 42 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY the diffueing reagent and modified droplets become so small that the mod- ification effect ia insignificant and all the measured parameters of the cloud medium approach the backgrouad levels. On the other hand, thia circumstance indicates that repeated flights of an a3rcraft through the _ modification zone do not result in eubstantial changes in etratiform cloude and exert no influence on "purity" of the eaperiment. It ie evi- dently poasible to trace the further course of the process only with a - _ marked increase in the seeding area. A significant effect from modifica- tion which could be used for practical purposes has not yet been estab- lished. The author expresses appreciation to A. V. Lachikhin, V. P. Bedrenko and B. S. Yurchak, apecialiats at the Institute of Experimental Meteorology, for assiatance in carrying out field experiments and measurements. BIBLIOGRAPHY 1. Aleksandrov, E. L., Klepikova, N. V., "Influence of Artificial Conden- sation Nuclei on the Kinetics of Formation of Cloud Droplets," TRUDY IEM (Transactions of the Institute of Experimental Meteorology), No 8(46), 1974. 2. Aleksandrov, E. L., Klepikova, N. V., "Effect of Artificial Condensa- tion 2Juclei on Development of the Cloud Spectrum," TRUDY IEM, No 9(52), 1975. 3. Aleksandrov, E. L., Yasevich, N. P., "Results of Preliminary Experi- tnents for Fog Modification in a Condensation Nuclei Chamber," TRUDY IEM, No 20, 1971. 4. Aleksandrov, E. L., Lachikhin, A. V., Posadskiy, V. I., Yudin, K. B., "Aircraft Photoelectric Instrument for Measuring Cloud Droplets," TRUDY IIIM, No 19(72), 1978. 5. Klepikova, N. V., Sedunov, Yu. S., "Kinetics of the Droplet Spectrum in the Initial Stage of Condensation," TRUDY IEM, No 3(37), 1973. 6. Sedunov, Yu. S., "Numerical Experiment for Modifying the Kinetics of Formation of the Cloud Spectrum by the Introduction of Additional Con- densation Nuclei," TRUDY IEM, No 6, 1969. - 7. Sedunov, Yu. S., FIZIKA OBRAZOVANZYA ZHIDKOKAPEL'NOY FAZY V A7.`MOSFERE (Physics of Formation of the Liquid Droplet Phase in the Atmosphere), Leningrad, Gidrometeoizdat, 1972. 8. Smirnov, V. V., "Restructuring the Microstructure of Fogs Under the Iri- fluence of Hygroscopic Particles," TRUDY IEM, No 25(93), in press. 43 FOR OFFICIAT iTSF f1NT.Y APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 - 9. Skhirtladze, G. I., Yurchak, B. S., "Measurement of the Coefficient of Horizontal Diffusion in Cumulus Clouda by the Radar Method," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA (News of the USSR Acadeary of Sciences, Phyaics of the Atmoaphere and Ocean), Vol 15, No 2, 1979. _ 10. Shifrin, K. S., RASSEYANIYE SVETA V MUTNOY SREDE (Light Scattering in a Turbid Medium), Moscow, Goatekhizdat, 1951. 44 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY UDC 551.509.314 PROBABILISTIC APPROACH TO THE OBJECTIVE CLASSIFICATION PROBLEM Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 80 pp 39-44 [Article by A. A. Burtsev, USSR Hydrometeorological Scientific Research Center, submitted for publication 13 July 1979] Abstract: In the probabilistic approach to objective classification the meteorological fields are not characterized as belonging un- ambiguously to some class, but by degrees of reliability of belonging to all classes. It ia demonstrated ttiat the proposed probabilistic approach ia more universal and has a number of advantaaes in comparison with the traditional determir.istic approach. In order to ascertain the degrees of reliability of such a determin- ation the author proposes the synthesis of a Golovkin algorithm with a potential functions method algorithm and its realization using ac- tual data. The results indicate an objective character of the restored degrees of.reliability. [Text] Recently in meteorology ever-increasing attention is being devoted to the problem of objective classif ication of the fields of ineteorolog- _ - ical elements. The timeliness of this problem is attributable not only - to theoretical and cognitive considerations, but to the fact that the results in one form or another can be used in weather forecasting. Many authors, concerned with classification, indicate the difficulty [2], and some [7] even suggest the impossibility of creating a universal ob- jective classification in meteorology. The author of [6] makes an attempt _ to create such a classification, but by virtue of the factors which will be considered below, this-approach can be considered universal only tc a limited degree. In this article we propose a new approach making it pos-- sible to come close to solution of this problem. _ Among the objectiva classification algorithms special attention shotll.c' be given to the B. A. Golovkin algorithm [5]. For the first time in meteor- ology it was applied in [4] and then with some modifications in [3]. 45 L'/1D !NL'VT/~TAT Ti[+T ALTrtI APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 This algorithm, constituting an approximate solution of the problem of linear programming with Boolean variables, makes it possible to discrim- inate in the set of objects X the centers of clustering of objects xi E X, and specifically, obtain a set of the most typical, or as they are uaual- ly called, "standard" objecta. Then the entire ensemble of objecta xi C X, i= 1,...,N (N is the number of objects) is broken down into classes by as- signing each xi to that class the distance to whose standard object is minimum. A merit of this algorithm is that it is completely free of any a priori assumptions concerning the spatial structure of the set of ob- jects to be classified and in the best way possible evaluates the number of classes and unambiguously distributes all the objects by classes. In this algorithm there is no need to stipulate any threshold distances; the only input data are the matrix (measuring N x N) of distances between each pair of objects. If the spatial structure of the set X is such that all xi can be broken - down into classes which are adequately compact and distant from one anoth- er, the Golovkin algorithm rather qualitatively solves the classification problem. The situation is far more complex in a case when the character- istic regions of the classes are situated close to one another and partic- ularly when the characteristic regions intersect. In this case objects _ which are extremely close to one another, situated in a zone of intersec- tion of the characteristic regions, can be assigned to different classes - and in th is case the probability of a classification error will be very - high. And precisely such a pecu7.iarity is characteristic for some large- scale fields of ineteorological elements. For example, it was established - in [3] that the fields of summer mean inonthly anomalies of H-SOQ (regis- - tered in the latitude zone 75-40�N during the period 1948-1977) have the property of intersection of characteriatic regions. This circumstance followed from the fact that after constructing a classification on the _ basis of the Golovkin algo:-ithm the zero hypothesis of equality of intra- - class means could not be refuted on the ba.sis of the Fisher test with a 5% significance level. The classification became significant orly after exclusion of a number of objects falling in the zone of intersection of the characteristic regions. But such an artificial procedure evident- ly cannot be considered justified because we obtain a number of objects i_ ' unrelated to any one of the classes. Returning to [6], it should be noted that the theory of a universal objec- tive classification constructed there in the considered case is inapplic- able since at its basis is the hypothesis of nonintersection of the char- acteristic regfons of classes. Thus, the classification in [6J can be re- garded as universal only in a limited class of fields of ineteorological elements. - In this article the author does not attempt to dispute the position that - in the atmoaphere there are some stable states around which the entire ensemble of temporal realizations of atmoapheric processes tends to be 46 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY grouped. However, the objectively established fact of the existence of some individual meteorological f ields not similar to one of the standard fields of determined classes forces us to seek a more flexible approach ta the objective classification problem than used now and before. Now we caill examine still another aspect of objective classification theory. In developing different meteorological classifications it is cus- - tomary to limit ourselve to the construction of "standard" fields of classes, assuming in this case that these standards are an exhaustive char- acteristic for all fields combined with them into a single class. Now we will consider the problem of the informational properties of the stan- dards, for clarity in description examining the case of two classes. We will denote these classes by A and B, and their standarda by xA and xB. The quantity of information I(y) concerning some field y EX consists of two ~arameters, to wit: the quantity of information abou* y included in xA and the quantity of information on y included in xB. I (y) _ / (v/xA) + f (.vlxR). The Iky/x) parameter is determined in the following way: 1 (J/�r) = H (x) +J/ (J) ---fl (X, Y), where c, H(x) f p(x) IogP (x) ttx is the statistical entropy of the x field and p(x) is probability density. = If y belongs to the region where p(x*A , y) is close to P(x*B, y), it is ob- vious that I(y/x ~)/I(y/.x*B) do not differ so much thar one of them can be neglected. Should we formally assign y to that class for which the p(x, y) value is maximum, we lose a considerable part of the information, which is scarcely justifiable. Thus, in the considered case of inter- sections of the characteristic regions the informational properties of the standards do not enable ue to solve the classification problem suc- cessfully if we employ traditional methods for breaking down the objects by classes. Another problem which must be dealt with in solving classification problems is the availability of a teaching sequence of sufficient length. V. N. Gli- venko has demonstrated the theorem of uniform convergence of the empirical distribution curve to the distribution function with an increase in the volume of the sample. It follows from this theorem that the length of the teaching sequence I ad adequate so that after ending the teaching process it will be possible to recognize the newly appearing objects is proportion- al to the dimensionality of criterial space and inversely proportional to the probability of classification error. Applicable to our data, in partic- ular, for the above-mentioned mean monthly 6H500 fields, it was establish- ed in [3] that lad N 6000, which is an order of magnitude greater than the archives of field data at our disposal. In order to reduce lad it is 47 FnR nFFTrTer TiCF nrrrv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 ~ . _ - _ _ _ necessary to reduce considerably the dimensionality ot criterial space. This can be achieved by the approach of expansion in natural orthogonal functions. However, since in order to obtain lad in our data we were forced to limit ourselves to a very small number of the first expansion terms, some perceiltage of the variabili[y (dispersion) of the fields re- mains not taken into account. In addition to what has been stated above, this circumstance indicates that salution of L�he classification problem, which in essence is a com- pression of the useful information, in the considered case a transforma- tion of this lnformation into a loss. It seems that this difficulty also cannot be overcome by any effective method if we remain within the framework of the traditional, that is, de- terministic theory of breakdown by classes. Accor3ingly, we will proceed to fo:mulation of a fundamentally new, probabilistic classification prob- lem: we will characterize each field xi E X not by an unambiguous assign- ment to the class A or B, but instead we will introduce the concept of the - degree of reliab ility of belonging DA(ai) and DB(xi), that is, we will say - that the xi field belongs to the class A with the degree of reliability DA(xi) and belongs to the class B with the degree of reliability Dg(xi). In the case of M classes each xi will be characterized by the vector D(xi) Dil,..., DiM 3, i= 1,..., M. Such an approach is far more universal a than the deterministic approach, and in addition, with such a representa- tion there is no informa.tion loss. In actuality, if we f ind all the degrees of reliability, the determined vector D(xi) will unambiguously characterize xi, that is, rigorously fix its position in the space X, not leavfng room for uncertainty. It is con- venient to uae a probabilistic characteristic as the degree of reliabil- ity. Thus, we will assume that in all X there are functions DA(a) and DE(x) which are the conditional probabilities that x belongs to class A or class B respectively. We note that such a probabilistic formulation of the prob- lem is a special case uf the deterministic approach, characterized by the fact that DA(x) and DB(x) only assume values equal to or close to 0 or 1 in objects of A and B, as occurs in the case of adequate compactness and a considerable distance between A and B. Now our problem is, using the objects xi E X appearing in the teaching pro- cess, to restore DAW and Dg(x) = 1- DA(x}. In order to approximate DA(x) we will use the adaptive algorithms approach, and specifically the poten- - tial functions method [1]. This method is attractive, first of all, because it is free of any a priori assumptions concerning the nature of the distrib- ution of objects, Lhat is, we need only assume that there is some non-zero probability of the :ppearance of objects xiE X; second, the method is at- Cractive due to its simplicity and the economy of computer time; third, the potential functions method allows successful synthesis with a Golovkin al- gorithm. Applying the Golovkin algorithm, we will seek the optimum set of standard objects and then we will "fix" them as the most typ3.ca1 represent- ativea of the classes, and after this, applying the potential functions 48 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY approach, we will ascertain the degree of reliability of whether each xi belongs to these standar.d objects. . TFius, aseuming that DA(x) is representable by a series in the form DA (x) cj?I (t), . ~ _ where (Pi(x) is some camplQte system of functions, we will use the recur- _ rent procedure fA 1 (x) =Jq (x) -f ~11 K ('CA+1, X), where fA(x) is the n-th approximation to DAx), K(xn+l, x) is a potential function. In general form K has the following form: _ C6 K )�i (_X) 'c(Y), (2) = where the followi..E, conditions are imposed on T i ~ -n : k,T--- L i t.ti'S; -10 n FE~;?'�-:~^,~�:~ �cnI 6 Fig. 2. Hydrodynamic and thermodynamic characteristics of atmosphere in _ zone of influence of typhoon "Carmen." 1) cloudless Trades zone, 2) north- ern periphery, 3) central cloud mass, 4) tail convective zone. KEY: 1. mb 2. m/sec 3. cal/g A11 the experimental data obtained cordance with the synoptic situatii sible to discriminate seveYul such 4. cm/sec 5. sec 6. cal/(cm2�sec) during this period were grouped in ac- 3ns in the polygon region. It was pos- situations: 1) Northern periphery of typhoon "Carmen"; 2) CentrRl cloud mass; 3) Tail convective zone. It ahould be noted that the proposed classification of synoptic situations reflects quite well the peculiarities of the state of the troposphere in the region of the aerological polygon. The distribution of the thermohydro- dynamic parameters of the atmosphere is qualitatively similar within each group, which makes it possible to consider the reaulta of the computations as objective characteristics of the corresponding synoptic eituations. The results of computations of the thermo- and hydrodynamic characteristics of the troposphere for the above-mentioned situations are presented in Fig. 2. The horizontal components of wind velocity u and v are determined in a Cartesian coordinate system with a reading origin situated at the center of Che polygon. The coordinate syatem is oriented in such a way that the OX- 56 FOR OFFICIAL USE ONLY f' . APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY axis is directed from the center of the typhoon. Here u and v are the radial and tangential components of wind velocity, directed along the OR and OY axes respectively. The positive direction of the tangential velocity com- ponent is counterclockwise. The positive sign for the vertical velocity component w corresponda to an upward direction, whereas relative vorticity ROT corresponds to counterclockwise rotation. Figure 2 also shows the thermodynamic characteristics of the atmosphere: latent heat LQ, static energy EL, and also their influxes (FLQ and FEL) per unit time in an air column with a baee 1 cm2 and an altitude 50 mb. _ Now we will examine the most significant characteristics of the troposphere atructure over the waters traver.sed by typhoon "Carmen" in its different zones. The aerological polygon of 10-11 August was to all intents and pur- poses aituated outside the zone of active influence of the typhoon (Table 1, Figure 1), on its northern pexiphery, at a distance of about 800 km from the center in a cloudless zone. As a result of typhoon movement on 12 Auguet the aerological polygon was within the cloud mass at a distance of about 400 km from the center. Then the typhoon moved and the aerological polygon occupied a position in its tail cloud system at distances from the _ � center of 800 km on 13 August and from 1,200 to 1,400 km on 14 August. - The greatest changes in the characteristics of the troposphere occurred _ with movement of the polygon from the northern periphery of the typhoon into its central cloud masa. The northern periphery of the typhoon, in all probability, was a slightly disturbed zone, subjected only to a limited in- fluence from the typhoon. This can be seen most clearly from the profile of the radial component of wind velocity, which does not have traits char- acteristic for zones of strong influence of typhoons (small influx in low- er troposphere, no upper tropospheric outflow). Other distinguishing char- acteristics of the considered zone are posi*_ive relative vorticity, the = presence of weak but noticeable ascending air flows, small heat influx ` and virtually zero moisture influx, on the average, in the entire thick- _ neae of the atmosphere. = ' The central cloud mass differs appreciably in its characteristics from the ~ considered situation. This difference was manifested most clearly in the following facts: there is a considerable change in the profile of wind radial velocity there is a great influx in the lower troposphere and an outflow in the upper troposphere; . there is a change in the stage of vorticity from positive on the north- ern periphery to negative (anticyclonic) within the cloud mass, especially at great altitudes; the descending flows in the entire thickness of the troposphere in- crease by a factor of 4-5; in comparison with the preceding situation the total tropospheric mois- ture content increases bq a factor of 1.4; 57 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 1'Vl\ there was a considerable warming of the troposphere a minimum in the profile of static energy (in the case of the northern periphery sit- uated in the layer 600-700 mb) was smoothed and the atatic energy in the entire air column increased. There was a subatantial change in the energy flux in a vertical column of . the tropoaphere. There was an increase in the influx of static energy FEL. It was distributed for thz most part in two layers� 950-700 and 700-300 mb. In the first layer the influx averaged about 2�10-~ cal/(cm2�sec), in the second approximately 5�10'3 cal/(cm2�sec), However, the influx of latent heat FLQ became negative and in the middle layers of the troposphere at- tained values 8�10-3 cal/(cm2�sec) (the integral moieture outflow from a vertical unit air column in this case attained 0.85 g/day). It should be noted that tt�e energy influx of 10'3 cal/(cm2�sec) in an air column with a height of 50 mb and with a base of 1 cm2 can heat (under the condition that energy is expended only on heating) air by 7�C per day. Thus, in the central cloud zone of the typhoan the tropesphere was intensively heated. Tha tail convective zone of the typhoon, in which the aerological polygon was present on 13 August, differed little with respect to hydrodynamic characteristics from the central cloud mass. As on the preceding day, there were descending vertical flows, but their velocitiea decreased by a factor of.4-5. Vorticity remained negative. The profile of radial wind velocity retains features characteristic of the central cloud mass, but the outflow in the upper troposphere substantially decreased. There wae a considerable change iz Lhe energy balance of the tropoephere. . Although the vertical distribution of EL and LQ virtually did not differ from the central cloud zone, the energy balance (that is, the influxea of heat and moisture into an air column) decrea.sed by a factor of 6-7. The total influx of all types of heat (static energy) was close to zero. - Table 2 Cobrdinates of Polygons and Observation Times (Cloudless Situation) - - ~ - KUQpitilnlbi $ C\1A Koor~,u~~r~ ~ - BPC\IA i10811C01(2, U ' 4itcno Ttuc.iol n ~rnHi� �ontiroiia, zpa~ 3 - ceFiTH6p� ~rpuneFi, zpao 3 CC}ITAUPA ( P I choe), ir cKoc), v - ~1 Z c. ut. 1 B. :1� 5 _ c. In. B. a. ~ 9 12 15,1 143.5 29 12 12,2 146,0 9 00 1511 143.5 29 18 12,2 146,7 pp 13,3 138,8 30 00 12,2 146,9 12 00 13,2 141,6 30 OG 12,1 146,?, 12 pp 12,8 141,3 30 12 11,5 141,4 12 12 15,8 143.5 2 12 12,3 145,5 29 OG 12,1 196,8 12 00 12,G 145.5 KEY: 1. Day in September 3. Polygon coordinates 5. E _ _ 2. Time (GMT), houra 4. N 58 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Thus, an analysis of the thermodynamic characteristir_s of typhoon "Car- men" makea it possible to draw the following conciusiona: 1. The tail convective zone of a typhoon ie characterized by a coneiderable heat content, differing little from the heat content of the central cloud mass, but differing appreciably from the Iittle-diaturbed northern peri- phery of the typhoon. 2. The tail convective zone is relatively conservative in the sense that the influx of latent heat FLQ and eepecially the total energy balance of the influxes FEL into a vertical air column are not great. 3. Both in the central cloud mass and in the tail convective zone of the ty- phoon there is a considerable anticyclonic vorticity in the entire thick- ness of the troposphere. It is particularly great in the layer 400-200 mb. 4. The vertical velocities both on the northern periphery and in the tail convective zone of the typhoon its eastern periphery are great. In this way the periphery differs appreciably from the central cloud mass of the typhoon, in which the vertical velocities are considerable. Some thermodynamic characteristics are given in Figures 2 and 3 for a com- parison of the thernadynamic characteristics of typhoon "Carmen" with the characteristics of the troposphere under undisturbed conditions. These thermodynamic characteristics are for the cloudless Trades zone, whose frequenc,y of recurrence during the time of the expedition was rather great. In order to compute these characteristics we averaged the experimental data obtained at different times during the course of the expedition and _ characterizing a relatively extensive cloudless zone. This fact was es- tabliehed ueing satellite photographs, nephanalysis maps and meteorolog- ical evaluationa of cloud cover over the polygon. Table 2 gives the coor- dinates of the polygons and the observation times for such cases. It should be noted that in this case (in the abaence of typhoons near the polygon) the coordinate system is oriented in such a way that the OX-axis is directed to the east and the OY-axia is directed to the north, that is, u is the ~ zonal component of wind velocity, v is the meridtonal component. We note the moat characteristic features of the thermohydrodyriamic parameters rep- resented on the graphs. These include: 1) Weak positive vorticity in the boundary layer of the troposphere and a considerable negative vorticity at great altitudes. 2) Weak, virtually zero ascending flows in the entire thickness of the tro- posphere. 3) A virtually complete absence (on the average) of exchange of heat and ua isture. The influx of static energy into an air column is virtually equal to zero. Thus, as follows from the computations, the cloudless Trades zone is char- acterized by a considerable conaervatiam, that is, a constancy of the ver- tical profiles of the principal thermodynamic characteristics and small (on the average) heat influxes into an air column. 59 F(1R l1FFTCTAT TTCF l1TTT V APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 BIBLIOGRAPHY 1. Minina, L. S., Arabey, Ye. N., "Structure of the Troposphere Under Tropical Cyclogeneais Conditions," METEOROLOGIYA I GIDROLOGIYA (Meteor- ology and Hydrology), No 12, 1979. 2. Petrova, L. i., Nesterova, A. V., "Dynamic and Energy Characteristics ~ of the Tropoephere on the Periphery of Typhoons and in Zones of Tteir Maximum Frequency of Recurrence," TAYFUN-75 (Typhoon-75), Vol I, Len- - ingrad, Gidrometeoizdat, 1977. 60 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY UDC 551.521.3 ACCYJRACY IN DETERMINING ATMOSPHERIC OZONE CONTENT USING DATA FROM - MEASUREMENTS OF OUTGOING RADIATION = Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 80 pp 51-58 - [Article by Candidates of Physical and Mathematical Sciences Yu. M. Timo- = feyev and M. S. Biryulina and V. P. Kozlov, Leningrad State University, _ aubmitted for publication 15 January 19791 - Abstract: Computations of the informativeness and - = optimum conditions for measurements of outgoing thermal radiation in the 03 absorption band 9.6 m are given in the problem of indirect determination _ of the vertical profile and total content of ozone - in the earth's atmosphere. The article gives an analysis of the physical reasons for the low in- formativeness and accuracy of the indirect method relative to the q(p) profile. Different ways are - proposed for increasing the accuracy in determin- ing the characteristics of ozone content using - measurements from meteornlogical artificial earth - _ satellites. It is shown that heterodyne measurements of outgoing radiation (with a resolution 10'2-10-3 = cm 1) make possible a substantial increase in the accuracy of the indirect method in the upper sttato- _ sphere. - [TextJ Measurements of the'total content of ozone U03 and the vertical pro- ` file of its concentration in the atmosphere, already carried out for a - quite long time, have become particularly timely during recent years, in = particular in connection with the problem which is arising of possible changes in the thickness of the ozone layer as a result of change in the chemical composition of the atmosphere [10, 18]. Present-day requirements for obtaining global information on atmospreric characteriatics stimulated the development of special satellite methods for determining ozone content based on interpretation of data on outgo-ing _ radiation in different spectral ranges. In one of these methods use is made 61 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY of ineasurements of outgoing thermal radiation in the region of the 03 ab- sorption band 9.6 � m. Sources [7, 12, 16] give the results of a theoret- ical analysis of the accuracy of this indirect method. The meteorological artificial earth sate113.tes "Nimbus-3" and "Nimbus-4" were used in measur- ing outgoing radiation, making it possible to evaluate its real accuracy, and also to obtain a great volume of valuable information on the global field of ozane content [13]. After analyzing the principal results of these studies it is possible to come to the frllowing conclusions: _ a) Theoretical investigations and satellite experiments indicated that the - informativeness of the considered indirect method when using an average epectral resolutian (Q y= 2-5 cm 1) is small. In particular, with the pres- ent-day accuracy of ineasurements of outgoing radiation it is possible to determine only 1-2 independent parameters of the vertical structure of the ozone content profile (for example, 1-2 coefficients in the expansion of the profile in the r_haracteristic empirical base). As a result, it is not possible to obtain sufficiently reliable data on the vertical structure - of the 03 content profile in the atmosphere. b) The indirect method makes it possible to determine the total ozone con- tent with a relatively hi;h accuracy. In particular, both theoretical in- vestigationa and the itice,:pretation of satellite data indicate the pos- sibility of determining U03 with an accuracy of 6-10% relative to the true or mean U03 value. c) In not a aingle one of the studies with which we are acquainted was any attempt made to analyze the physical reasores for a low informativeness of the indirect method or to find wa7s to increase it. In particuiar, until - now there have been no computations of the optimum canditions for measure- ments of outgoing radiation in the absorption band 9.6 �-m. Experience has ' indicated that a correct solution of this problem in a number of cases will considerably inerease tiie accuracy of indirect methods [3]. In this study an attempt is made to fill the existing gaps. The mathematical basis for solution of inverse problems in the theory of transfer of thermal radiation is the well-known integral form of the trans- - fer equation [4]. After linearization of the equation the considered prob- lem is reduced to solution of the following operator equation of the first kind: 8l=AXZq, (1) _ [It is assumed that the atmospheric temperature profile is known on the basis of solution of the problem of theroal sounding of the atmosphere [4l]� ~ _ where b I is the vector of deviations of the measured (with different fre- quenciea yi, i= 1, 2,...,m) radiation Z( v i) from the radiation I( V i), correaponding to the mean profile of of the mixture ratio 03 - q(p), 62 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Sq is the corresponding vector of deviation of the mixture ratio (at dif- ferent atmospheric levels p1,j - 1, 2,...,n), A is a matrix m x n, whose elements have a simple physical sense they characterize the sensitiv- ity of outgoing radiation in the i-th spectral range to variationa of the q(p) pref ile at the level with the preseure pi. n uF Fig. 1. Vertical temperature profile T and mean ozone profile q for which - Computationa were made (a), the kernels A(p) for the absolute variations of ozone concentration in three registry channels (b) and the kernels A'(p) for relative variations of the ozone concentration (c). 1) y= 1055 cm 1, ' 2) V =995cm1, 3)~) =1125cm1. KEY: 1) mb 2) 8/8 ' 3) erg/(cm2�sr�cm 1)/g/g Figure lb shows examples of computation of Ai(p) for three spectral ranges at the center and in the wings of the absorption band with a spectral reso- lution of the measurements Q,y = 5 cm'l. The computations were based on use of a atatiatical absorption ;nodel [9] and for the mean profiles q(p) and T(p) corresponding to the period May-September [17]. An analysis of the behavior of the Ai(p) curves shows that for all the con- sidered apectral intervals (we examined 20 intervals in the range from 965 to 1155 cm 1) the ma.ximum influence on outgoing radiation is exerted by absolute variations of ozone content at approximately one and the same level, situated in the troposphere, that is, svbstantially below thc, level where the maximum 03 content is situated (see Fig. la, which gives mean q(p) profile. A change in the optical density of the spectral :~lte_r- vals (with movement from the center of the band to its edges) leads only 63 - FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 ~ v y�iu c~c 2> 0 -A�~d~ e 3pt/(CM1CeKGO�Gti'l �AO0'30tAcHt�ctKtP-f4 41 ?OD ' ?SO T K t/t 3 ) APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-00850R040240090007-3 I to changes in the absolute values Ai(p), but not to a displacement of the curvea in the vertical coordinate. This leads to the conclusion that for the considered indirect method the information available for layers above and near the ozone content maximum is very low. A similar behavior of the _ Ai(p) curvea is also obaerved for other aeaeons. The Ai(p) curves shown in Fig. 1 characterize the sensitivity of I(v) to absolute variations of q(p). However, due to the strong variability of q(p) within the limits of the ozone laper (q(p) varies by two orders of magnitude) it is useful to analyze the behavioi. the Ai(p) curves cor- responding to an ~erational equation relative to the ~q/q values (it is easy to see that Ai(p) = q�Ai(p). Figure lc shows Ai(p) curves for these same spectrai intervals. It can be seen that although the Ai(p) curves are situated higher in the atmosphere, they are characterized by the above- mentioned peculiarities. Under conditions of a low information yield of the indirect method it is _ desirable to make computations of the optimum conditions for measuring outgoing radiation for the purpose of maximum use of the information con- tained in the I(v) spectrum. In this study we used the optimization meth- od proposed by one of the authors [13 and involving the optimum combining of the radiation c+f different spectral intervals in different measurement - channels. The choice of the optimum number of ineasurements (N) is accom- _ plished on the basis of the known criterion [2] ~k~/(y I~ (2) where /'\k is.the k-th eigenvalue of the covariation matrix of radiation KII = AKqqAT, O'I is the dispersion of random error, AT is a transposed matrix. Table 1 Eigenvalues ~k(erg/(cm2�sec�sr�cm 1))2 for Different Seasons k V-IX X-XII I-IV 1 0.4135�102 0.4375�102 0.1558�102 2 0.3694 0.2382 0.1565 3 0.3167�10'2 0.1480�10-2 0.1700�10-2 Table 1 gives the first three eigenvalues for three seasons when using co- . variation matrices of ozone content ICq.q from [17]. - It follows from the data in Table 1 that with the preseat-day accuracy in measurements ( O" =0.1-1.0 erg/(cm2�sec x sr�cm-1) with a spectral resolu- tion 6v= 5 cm"l) the crfter.ion (2) is satisfied by 1-2 eigenvaLues. 64 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY of, .�1 , 71 /65 1000 � IOSS 1010 11J5 o 961 1010 lPi010SJ 1,0 6) 0" , 96S 1D10 1055 1010 1135 1 ^ ~ S6S ~ 1010 f0~y0 fOSS n. . 96' 1010 1055 1070 1135 Fig. 2. Configuration of optimum plan for combining spectral intervals. a) for two-channel instrument (N = 2); b) for three-channel instr.ument (N = 3) (with a value of the ordinate equal to unity a channel is cut in t_o }he combining operation). Table 2 KEY: o, a0,1 3pe1(c.x..cen�cp.c.u-1) 2 c, =i,u~p2l (c.,,=.cen - �cp.c.s--1) . - . n ~ (1-7Qj7Q)�1M% - n (1--:Q17q)�100ti p ,04.~ 9, o ! 1009~ I nQ q�I00% ~ N M Nc3 ~ M N=3 N N II II M II t `1 N N II M II M II i 21 3 1a 1 .9 s 7 1 s 9 1 io 0.045 o 0 o 16,3 77 o 0 o 11,7 77 0,425 u u o 37,4 27 o u u 21i,6 27 s 7,0 a,c> 13,7 �10,5 16,5 3,8 ~1,0 4,3 17,4 18,7 10 10,2 17 5 6,3 28 2 19,8 4 28 61;,0 :31 5 10,1 9 1,8 J S 5,1 } 2 2 .5,9 25 6 25,8 25 6 11,9 i 9 20 , , , , ,1 , : , : , , 100 33,2 13,3 33,3 33.8 21,0 29.4 31,9 33,0 33,�1 21,1 200 Sfi,:i -A.3 57,2 59,2 26,2 ;52.5 53,9 :,i ,:3 27,0 400 24,5 18,5 31,6 40,6 19,4 12,1 25,5 2G,7 :3,1.7 21,7 800 4,7 2.6 9,1 17,6 32,3 l,'l IO,i 11,4 11.�S 31,6 A n A 1 -aL o 92,0 94,90 Jf,,16 9G,Eil =0,47% U 63,3 9~1,75 9:,,d~l or~.i,:! =0,55% G 7 U I p:~ I I t 0 1. mb 2. erg/(cm2�sec�sr�cm 1) It can be aeen in Fig. 2, which shows examples of the optimum plans with N= 2 and 3 for the season May-September~that the optimum measurement channels occupy rather broad and overlapping spectral regions and also 65 FOR OFFICIAL USE ONT.Y APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 include sectors in different parts of the band. Computations for other seasons indicated an absence of significant variations in the optimum plans. Now we will proceed to an analysis of the results of evaluations of errors in the indirect method for different measurement schemes. It should be _ noted that in a large number of studies (for example, see [12]) as the parameters characterizing the error in the method use is made of the relative errors in restoration, determined relative to the true or mean q(p) or U03 values. However, taking into account that the relative natur- al variations of the characteristics of ozone content are relatively small (12-60%) f17j, such an approach can create the impression of a rather high uccuracy of the indirect method. For example, in evaluating accuracy it is more correct to use the parameter ~ (v) Yo ip1= �y (p) characterizing the relative decrease in a priori uncertainty in knowledge _ of the profile q(p) after carrying out a satellite experiment ( O'q and ~ -ri ~ P4 04 D ~ b 0 ~ ~ P4 ~ 1. i r-~I ctl Cl O' .~G G a1 'z+ H D 00 O rI > a, ~ L ~ Cr' �rl "Cy 1J r+ 7 co ~ ~.I ~ ~ ~H r l C O U N p ~ ~ 0 ~ N ~ U T1 l.r Q cO i.+ cU . ~ ~ U u G1 �r l - p ~ ri ~ U w ~ co ~ G l + J ~ b0 Gl > ~f c0 H A; O V p rn 107 _ FOR (1FFT('. TAT. TTSF. nur.v APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 . _ 0 o C) ~ N N r--I ~O C w u o .,4 Ai ~ ~ ~ Q~N I I, I ,,..r U O as p O~7 r-i ~ O O~ Ctl r-i Q1 F-+ U i-+ O U O U G M 0 ~ O M u O r~ co II ~D ~ a0 ~1 N "O O 2~ ~ rn v ~ 0 00 CT% r- M O G �n O 0 SG ' \.O r-I ~ I C1. H C > r-~ c~~1 ri f~ rl O~ ON ~ C) 41 I r- 00 > rl MH r- 0 w ~4 O m r-i M O ~ 41 D, 00 II cn * � -rl r-I o Q) M � ~ --i ~ o~~ 41 co o c s o~~ ~ a H r-I oo � ~ i i ro ' N N v " ~ ~ > o 0 i c O u a a a~ y z o ~ b i i b 0 . > ~ > 9 " c~ � ri Cd 3 CY) ~ u 0 0 O F H O) d O U ' D+ d ~ O' m ~ O -H N 4.) cd cti 0 ~ U cC ~ ~ y O cd ~ ~ co V V 0 b0 41 H ~ ~ l U ~ bo W r w ~ 44 p P~G O A 10 8 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY extrapalation interval 'teX, so that Pj-),0 with 7 km-1 for Sc, As relate to clouds of the mixed phase. Measurements of spectral transmission in [18], made in the atmosphere in the presence of crystals In the form of needles, indicated that attenuation in the region of wavelengths X = 0.5-12 � m is close to neutral, although in some spectra there is a weak minimum at /^k= 9.2-10 � m. It is reported in - [38] that the attenuation of IR radiation with A = 10.6 � m in crystalline clouds is somewhat greater than in the case of visible radiation. On the basis of these data it can be concluded that in contrast to droplet clouds, attenuation of radiation in the range A = 0.5-12�.m by crystalline clouds is approximately identical. Therefore, Table 2 gives some idea concerning , the attenuation indices of crystalline clouds in the IR spectral region as well. 151 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY In order to compute the attenuation of radiation by crystalline clouds it is important to know whether it is possible to approximate crystals of dif- ferent configuration by ice spheres. At present there is no univerQal agree- _ ment on thia subject. Computations [15, 161 of the factora involved in the effectiveness of attenuation ICO(10.6) for radiation with a= by ice spheres, cylindera and platelets ahow that the influence of config- uration is manifeated for particlea with aizes ~o - 27Tr/,A < 6(where r is the radius of a aphere or cylinder and the half-thickness of the plate- let). The authors of [3, 36] compare the measured Kp(10.6) values with the . computed values for ice spheres and come to the conclusion that regardless of the configuration of the crystals, in computations of attenuation at 10.6 � m it ia possible to use a model of ice spheres of an equivalent sec- tion. In [15], on *he basis of a comparison of the results of their meas- urements of spectral transparency in crystalline fogs and computations for large water droplets, the conclusion is drawn that a model of spherical ; particles cannot be used in computing spectral attenuation. However, the authors of this study neglected the difference in the optical constants of water and ice. Accordingly, further investigation is required for an- swering the formulated question. ; Radiation Scattering by Ice Crystals One of the distinguishing characteristics of the scattering properties af j- _ crystalline clouds is manifested most clearly in observation of a number ' of optical phenomena accompanying the transmission of solar radiation _ through a veil of cirrostratus clouds (less frequently, cirrus, cirrocum- ulus and altocumulus). These include large and small halos, false suns, tangential and circumzenith ares, light pillars, etc. These phenomena have long attracted the attention of researchers and have been described in detail in [11, 30 and in a number of other studies]. All thesE pheno- mena are caused by the reflection of light rays from plane surfaces or by light refraction in a crystal, which occurs the same as in a prism [11]. An ice crystal, having the configuration of a regular hexagonal platelet or pillar, for the rays constitutes a prism with refracting angles 60 and 90�. A ray, entering a prism, emerges from it after double refraction, forming small and large halos in the direction of angles 22 and 46�. Due to light dispersinn during the transmission of rays in the prism the halos must be brightly colored (inner edge red, outer edge violet)., ` in contrast to the phenomena caused by reflection (for example, a hori- zontal circle). However, the color of a halo is frequently not pure or is totally absent. In this case there is a"white" halo. The reasons for this may be both oscillations or rotation of the crystals about their axes and the superposing of reflected, refracted and diffracted rays [11, 30]. All the enumerated phenomena arise only when clouds contain crystals and for observers serve as one of the methods for determining the phase state of cloud cover. Moreover, the appearance of most of them indicates that in clouds conditions sometimes arise for the predominant orientation of crystal axes [11]. The process of radiation scattering by particles of 152 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY any configuration is described by a transformation matrix (scattering matrix), by means of which it is possible to express the Stokes parameters - of the scattered wave through the Stokes parameters of an incident wave [19]. The Stokes parameters characterize the intensity and polarization of radiation. In a general case the scattering matrix has 16 components. Spherical particles of an optically inactive aubstance have a scattering matrix of the aimpleat type (the tranaformation matris containa only three independent parametera). The asphericity of particles exerts an influence on all the light scattering matrices. The predominant orientation of aspherical particles can cause linear and circular refraction effects in combination with dichroism, even if the particles are not optically active [19]. Due to the great mathematical difficulties no expression has been - derived for computing the components of the scattering matrix, but in [33] a study was made of the theoretical principles for their experimental de- termination. %7' iU I , P% ~ IN` ~ fA 20 0 II -20 -yo 0 60 9C 120 8� Fig. 2. Polarization of light scat- tered by crystals cylinders (1) and platelets (2) according to data in [35]. I1 F~t ~ 0 Fig. 1. Scattering indicatrices for radiation with a= 0.63 � m. 1) com- puted for prism and horizontal scat- tering plane, 2) same for vertical scattering plane; 3) measured in a crystalline medium (data from [1]). 153 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Table 1 Phase Structure of Clouds in Different Regions of the Earth Cloud form Eastern zone of Temperate Arctic tropical Atlantic latitudea Cu droplet - St droplet, mix2d, mixed droplet Sc droplet droplet, mixed, mixed dropl et Ns mixed mixed Ac droplet droplet, mixed, mixed droplet As mixed, mixed, crystalline, droplet crystalline mixed Cs mixed, crystalline crystalline crystalline C1 crystalline crystalline crystalline Table 2 - Values of Attenuation Index in Clouds of Different Forms, km71 Cloud form Eastern zone of Temperate Arctic trop{cal Atlantic latitudes Sc 25 As 7-20 25 4 Cs, Ci sp 8.3 2.5 2.5 Crystalline fogs 0.5-1.2 2.3-32 Until recently the attention of researchers has been concentrated only on measurement of the components fll and f12, which characterize the intensity - and degree of polarization of scattered light. Tte angular dependence of fll (scattering indicatrix) in crystalline fogs [27], measured for the firat time, did not exhibit intensity maxima in the range of halo angles, which in [28] is attributed to an inadequate angular resolution of the in- etrument. Later maxima of different intensity were noted at scattering angles of about 22� and 46� [4, 8, 9, 34, 35], corresponding to the amall and large halo circles observed in nature, and also with e= 142� [35] 154 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Bouguer halos [ll]. A halo with 6a;46�, also under natural conditions, was obaerved considerably less frequently. In additton, in these investi- gations greater scattering in the region of angles close to 90� was observed in comparieon with droplet clouds (for the first time in [12, 27]). However, the scattering indicatricea meaeured by different authors reveal a substantial scatter of values, especially in the region of lateral scatter- ing angles. An analysis of the data indicated that thia scatter cannot be attributed to a difference in the shape and size of particles. The reseon for the discrepancy in the eaperimental data was expleined in [13, 141, where it was demonetrated that the presence of droplets or ice spheres in the medium, together with crystals, exerts an influence on the angular dis- tribution of scattered radiation. With a clear identification of the phase state and discrimination of cases of a purely crystalline medium the scatter of scattering indicatrices for crystals of different shape and size in the range of angles 10-180� was 1-30X. In [1, 14] the authors also demonstrated that an increase in scattering by crystals in the region of lateral angles occurs due to a decrease in the region of angles 2< 6< 40�. These data were obtained when measuring scattering in a single plane. However, as dem- _ onetrated in [2, 8, 9, 36, 44], the predominant orientation of crystals - leads to an asymmetric distributi4n of the scattered radiation in different scattering planes, although in the range of angles 10-180� the difference _ in fll( B) for different planea does not exceed the discrepancy in the in- tenaity values measured in a single plane. Figure 1 shows the averaged - scattering indicatrix from [1, 13], and Figure 2 shows the degree of plane polarization of scattered light [35]. Xn) Hpucmonnu11r.cKUe _ . - - - - - ~ " 1 u6nann � V, o z d,o ~ ~q~ o s - O's ` 0 7 _ . ~ NcKyccmDerrNOn � 2� cp~da croetuaNyou , ~ ` ~'anena~reie p ~~-~-r3 06nara -ZO -10 0 ~ 10 t 'C Fig. 3. D('rf) values for clouds of different phase state according to data from [41]. 1) snowflakes, 2) snowflakes covered with hoarfrost, 3) fall- ing anow, 4) graupel, 5)-6) thawing particles, 7) freezing droplets of aspherical configuration. KEY: 1. Crystalline clouda 3. Drnplet clouds _ 2. Artificial medium of mixed phase 155 L�!1D AL'L`TrT AT ttOV nwn v APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 Measurements of the angular variation of other matrix components have been made only in [7], whose basic conclusion is that the matrix of light scat- tering by crystals is characteristic for asymmetric particles with a ran- dom orientation. In addition, it is noted that measurement of ite compon- ents can become a means for the remote atudy o� clouda, for example, study of the depolarization of acattered light. The conclusion that it is possible to identify the phase state of clouds on the basis of the depolarization ratio D(31) was drawn earlier in [20]. D(31) is determined as the ratio of the perpendicular and parallel compon- ents in the backscattering signal in the casp of irradiation of particles by plane-polaiized radiation. 1'he experiments revealed that for water droplets with a diameter up to 5.5 mm the depolarization ratio D(J7)< 0.03 [40], whereas in crystalline clouds and snowfalls from 0.5 to 1.0 [20, 41]. However, it is difficult to characterize clouds of differ- ent composition by any definite D(71) values. It is only possible to es- tablish some limits of these values in dependence on the phase state of cloucis at different temperatures, as is demonstrated by Fig. 3 from [41]. Thus, the singularity of the light scattering matrix for a crystalline me- - dium is caused for the most part by the configuration of the crystalline particles. Therefore, the use of the scattering indicatrices in computa- _ tions as the model of a crystalline cloud, as models of ice spheres [24, _ 39] or circular cylinders [31] does not give a satisfactory result. In [1] there was a study of light scattering by hexagonal ice prisms of finite length in the approximation of geometrical optics and diffraction and analytical expressions were derived for computing the scattering in- dicatrix. Figure 1 showa the computed scattering indicatrices in the hor- - izontal and verticaZ planes for prisms oriented in the horizontal plane - (the radiation beam is propagated in this same plane) [1]. The consider- ed case ma.kes iti possible to evaluate the maximum possible asymmetry of light scattering becauGe all other cases of orientation should lead to its decYeas2. The a.greement with the experimental data in [13] can be consfdered satisfactory. The discrepancy in the computed and experimental ~ [1, 421 indicatrices in the region of backscattering angles can be attrib- uted to the fact that in the computations no allowance was made for a aumber of effects, such as surface waves [19]. Thus, the scattering indicatrices shown in Fig. 1 can evidently character- ize the angular distribution of apparent radiation scattered by a crys- talline cloud with e,> 10�. Interrelationship Between Backscattering and Optical Thickness In interpreting the results of lidar sounding of clouds at different wave- lengths it is of interest to study the interrelationship between backscat- tering Crand the optical thickness 'G~ . Experimental investigations [3, 156 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY 13, 14] have shown that this interrelationship for ~k= 0.63~t.m, inde- pendently of the phase composition of the medium, is described by a re- gression equation with identical coefficients. For Z 0.63 < 0.4 the in- terrelatioriship between 0'it and optical thickneas can be considered linear. The deviation from linearity in the case of large 2 63is attributable to the increasing contribution of multiple scatterin$: The equality of the regrebsion coefficients means that the value of the backscattering in- dicatrix for ^A = 0.63 in a cryatalline medium is the same as in a droplet medium: fll(51 ='m 0.05 (f30Y). Measurementa [6, 42] for wave- lengths 0.57, 0.63 and 10.6 � m, the same as in [3, 141, did not re- veal C~rdifferences in droplet and crystalline fogs with identical 2~. Investigations were not made at other waveleagths. Thus, from the backscattering signal for A = 0.63, 0.57 and 10.6 � m it is evidently impossible to determine the phase composition of cluud cover. This requires that measurements of polarization, such as D(Tt), be made. Transmission and Scattering of Radiation by a Crystalline Cloud Crystalline clouds, the same as droplet clouds, constitute an optical medi- um which is quite dense in which single scattering is accompanied by mul- tiple scattering processea. The problem of the propagation of radiation in cloud media in which multiple ecattering occura and with allowance for ab- sorption by atmoapheric gases and aerosol has not been solved at all. In theoretical investigations it has been special methods for solving this problem which for the most part have been developed and the results of _ computations of the fluxes of solar radiation reflected, absorbed and - transmitted by layers of droplet clouds are given. Individual character- - istics of radiation transfer were also examined for crystalline clouds, a layer of ice spheres or circular cylinders being used as the model [23, 24, 31, 32, 391. It was noted earlier that these models cannot be re- garded as satisfactory for computations of the indicatrix of light scat- - tering by ice crystals. However, the results of some theoretical and ex- perimental studies give basis for assuming that a model of ice spheres can be employed in computing the characteristics of radiation transfer. For example, it was demonstrated in [32] that the computed values for radiation and transmission in the window 8.3-12.5 � m for layers of different thickness containing ice cylinders and spheres of equivalent cross section differ insignificantly. Measurements of the reflectivity of artificial crystalline fogs in the near-IR spectral region were made in [45]. The author of [24] demonstrated that the results of these measure- ments agree well with the camputed reflectivity values for spherical ice particles. The transmission and reflection of solar radiation by clouds is dependent both on the optical properties of droplets and crystals and on the char- acteristics of radiation absorption in and outside a cloud [21, 22, 261. The influence of absorption by atmospheric gases in the range a= 2.4-3.0 �m is illustrated by Fig. 4, which shows the solar radiation spectra 157 Vnv nvVTrrAT ttcV rnnv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 for radiation transmitted by crystalline clouds Ci [21]. The solar spec- trum was measured in a clear sky at the same altitude (6.2 km) at which in another flight the base of Ci clouds with a thickness from 0.5 to 3 - - km was si[uated. 1 lDO U 1, 4 v /r ~ ~ i i3 2, 9 J 1,4 MKM ~A m Fig. 4. Transmission of solar radiation by the pure atmosphere at an aYti- tude of 6.2 km (1) and by Ci clouds thin (about 0.5 km), (2) intermedi- ate thickness (3) and dense (about 3 km) (4). Ie11.a'yDm/(CNr� c,) � rrn.y} 1 M 1000I1 100 � 1CG ',`l ~ ip . f I ~ i,0 i,S 1,4 ?,.1,?AMIr+I 2 Fig. 5. Reflectivity of solar radiation by dense Ci (1) and dense Cu (2). KEY: 1. Irefl , 1A W/ (cm2 � sr �tkm) 2. m The inf luence of absorption by atmospheric gases is also manifested in the spectra of reflectivity of solar radiation. There are reflectivity minima corresponding to the absorption bands of o.rygen A = 0.76 and 1.26~i.m, the strong bands of water vapor ~1, = 0.95, 1.13, 1.38, 1.47, 1.86, 1.9 and 2.58 �,m, as well as -the H20 and C02 bands with = 2.01, 2.06 and 2.68~t,m. In addition, there are minima corresponding to the water or icE absorption bands [21, 22]. - 158 FOR OFFICIAL USE ONLY LA APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY The latter is attributable to the fact that the absorption of radiation by particles decreasea scattering. Figure 5 ehows the averaged reflectiv- ity spec tra for cryatalline Ci and droplet Cu clouds from [21]. In aome intervals of wavPlengths the reflectivity of these clouds is different. The difference is maximum where the complex refractive indices of water and ice are most different. Thus, the measurement of reflectivity in the IR spectral region can be used in identifying the phase state of clouds. This conclusion was later confirmed by computations for plane-parallel layers of the medium con- taining spherical particles of water and ice [24, 29, 39]. _ Summary - Thus, the optical properties of crystalline clouds at present have not been = fully studied. Some components of the scattering matrix of visible radi- - ation, especially the scattering indicatrix with 0> 10� and the polariz- = ation of scattered radiation for angles 15-150; can be considered known. - An analyais of published investigations shows that the singularity of the - scattering properties of crystalline clouds, distinguishing them from droplet clouds, includea: ~ 1) manifestation of such optical phenomena as halos, light pillars, false suns, different ares, etc., caused by refraction, reflection or diffrac- - tion of Iight on ice crystals in the case of a random or predaminant ori- - entation; _ 2) an increase in the fraction of radiation scattered in lateral direc- tions in compar3son with drops; 3) the partial depolatization of backscattered radiation. ' All these effects are caused by a differeace iri the shape of the crystals from spherical and therefore cannot be precisely modeled by means of poly- disperse ice spheres ar cylinders. In computations of the scattering indi- - catrix for visible radiation it is possible to use the analytical expres- - sions proposed in [1]. These were derived for a model medium consisting of ice hexagonal prisms. In addition, the spectral crystall ine and droplet c able for the most gart to and ice. Aecordingly, for ity of crystalline clouds corresponding size. variation of reflectivity of IR radiation by louds is different. This is evidently attribut- the infZuence of the optical constants of water computing the spectral variation of reflectiv- it is possi'ble to use a uodel of ice spheres of The characterist3cs of the optical progerties of crystalline clouds given above give basis for remote determi.nation of the presence of the crystall- ine phase in elouds and the relative cnntent of cryst.als and droplets. 153 rnID AVL�T/�7A� rrOa n*nv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000200094407-3 rvn vrri...Lru., v.... BIBLIOGRAPHY 1. Volkovitskiy, 0. A., Pavlova, L. N., Petrushin, A. G., "Light Scat- tering by Ice Crystals," IZV. AN SSSR, FIZIKA ATN:OSFERY I OKEANA (News of the USSR Academy of Sciencea, Physica of the Atmosphere and Oceun), Vol 16, No 2, 1980. 2. Volkovitskiy, 0. A., Pavlova, L. N., Snykov, V. P., "Asymmetry of the Scattering Properties of a Crystall ine Cloud Medium," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA, Vol 11, No 7, 1975. 3. Volkovitskiv, 0. A., Nikiforova, N. K., Pavlova, L. N., Petrushin, - A. G., Snykov, V. P., "Model Investigations of the Optical and Aero- dynamic Properties of Ice Crystals," VOPROSY FIZIKI OBLAKOV (Problems in Cloud Physics), Gidrometeoizdat, Leningrad, 1978. 4. Dugin, V. P., Golubitskiy, B. M., Mirumyants, S. 0., Paramonov, P. I., Tantashev, M. V., "Experimental Investigat ions of the Optical Charac- - teristics of Artificial Ice Clouds," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA, Vol 7, No 8, 1911. ~ - 5. Dugin, V. P., Volkovitskiy, 0. A., Maksimyuk, V. S., Mirumyants, S. - 0., Snykov, V. P., "Spectral Transmission of Artificial Crystalline Cloud Formations," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA, Vo1 12, No 4, 1976. ' 6. Dugin, V. P., Mirumyants, S. 0., Pavlova, L. N., "Experimental Inves- tigations of Backscattering at Wavelengths 10.6 and 0.57 �.m by Artif- icial Cloud Formations," TRUDY IEM (Transactions of the Institute of Experimental Meteorology), No 13(58), 1976. 7. Dugin, V. P., Mirumyants, S. 0., "Scattering Matrices of the Light of Artificial Crystalline Clouds," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA, Vol 12, No 9, 1976. 8. Dugin, V. P., Volkovitskiy, 0. A., Mirumyants, S. 0., Nikiforova, N. K., "Anisotropy of Light Scattering by Artificial Crystalline Cloud Formations." IZV. AN SSSR, FIZIKA ATMdSFERY I OKEANA, Vol 13, rro i, 1977. 9. Dugin, V. P., Maksimyuk, V. S., Mirumyants, S. 0., Nikiforova, N. K., "Anisotropy of Light Scattering by Artificial Cryatalline Cloud For- mations (Vertical Illumination of Medium)," TRUDY IEM, No 13(58), 1976. 10. Kosarev, A. D., Mazin, I. P., Nevzorov, A. N., Potemkin, V. G., Shug- ayev, V. F., "Comparison of Some Microphysical Characteristics of Clouds in Different Geographical Regions," VOPROSY FIZIKI OBLAKOV, Leningrad, 1978. 160 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2047102108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY 11. Mindart, M., SVET I TSVET V PRIRODE (Light and Color in Nature), Mos- cow, Nauka, 1969. 12. Nikiforova, N. K., "Experimental Study of Light Scattering by Droplet and Crystalline Fogs," TRUDY IEM, No 10, 1970. 13. Pavlova, L. N., "Investigation of the Scattering of Visible Radiation in Cloud Media Containing Crystals," IV VSESOYUZNYY SIMPOZIUM PO RAS- PROSTRANENIYU LAZERNOGO IZLUCHENIYA V ATMOSFERE (Fourth All-Union Symposium on the Propagation of Laser Radiation in the Atmosphere), Summaries, Tomsk, 1977. 14. Pavlova, L. N., "Measurement of the Coefficients of Directed Light - Scattering in Droplet and Crystalline Cloud Media," IZV. AN SSSR, FIZIKA ATMOSFERY I OKEANA, Vol 14, No 9, 1978. 15. Petrushin, A. G., "Attenua.tion, Scattering and Absorption of Radia- tion at-10.6 m by a Model Cloud Medium Containing Ice Circular Cyl- inders of Infinite Length," TRUDY IEM, No 11(54), 1975. 16. Petrushin, A. G., "Attenuation of IR Radiation in Crystalline Cloud Media," TEZISY DOKLADOV NA IV VSESOYUZNOM SIMPOZIUME PO RASPROSTRAN- ENIYU LAZERNOGO IZLUCHENIYA V ATMOSFERE (8ummaries of Reports at 4th All-Union Symposium on the Propagation of Laser Radiation in the At- mosphere), Tomsk, 1977. 17. FIZIKA OBLAKOV (Cloud Physics), edited by A. Kh. Khrgian, Leningrad, Gidrometeoizdat, 1961. 18. Filippov, V. L., Ivanov, V. P., Makarov, A. S., "Variations of Aero- sol Attenuation of Radiation Under Conditions of an Ice Fog," TEZISY DOKLADOV NA IV VSESOYUZNOM SIMPOZIUME PO RASPROSTRANENIYU LAZERNOGO IZI,UCHENIYA V ATMOSFERE, Tomsk, 1977. 19. Van de Khyulst, G., RASSEYANIYE SVETA MALYMI CHASTITSAMI (Light Scat- tering by Small Particles), Moscow, IL, 1961. 20. Shupyatskiy, A. B., Shlyakhov, V. I., Kravets, V. V., Tyabotov, A. Ye., "Use of Laser Technology in Polarization Investigations of Meteorolog- ical Formations," METEOROi.OGIYA I GIDROLOGIYA (Meteorology and Hydre,l- ogy), No 2, 1967. _ 21. Blau, H. H., Espinola, R. P., Reifenstein, E., "Near-IR Scattering by Sunlit Terrestrial Clouds," APPL. OPTICS, Vol 5, No 4, 1966. 22. Blau, H. H., Espinola, R. P., "Spectral Property of Cloude from 2.5 to 3.5 Microns," APPL. OPTICS, Vol 7, No 10, 1968. 161 xnu nFFTrTer TrcV n7%nv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-00850R040240090007-3 23. Dave, J. V., "Intensity and Polarization of the Radiation Emerging from a Plane-Parallel Atmosphere Containjng Monodispersed Aerosol.y," APPL. OPTTCS, VoJ 9, Na 12, 1970. 24. Hansen, J. E., Cheynay, H., "Theoretical Spectral Scattering of Ice Clouds in the Near Infrared," JGR, Vol 74, No 13, 1969. _ 25. Heymsf ield, A. J., Knollenberg, R., "Properties of Cirrus Generating Cell," J. ATMOS. SCI., Vol 29, No 7, 1972. 26. Hovis, W. A., Blaine, L. R., Forman, M. L., "Infrared Reflectance of High-Altitude Clouds," J. APPL. OPTICS, Vol 9, No 3, 1970. 27. Huffman, P. J., Thursby, W. R., "Light Scattering by Ice Crystals," J. ATMOS. SCi., Vol 26, No 5, 1969. 28. Huffman, P. J., "Polarization of Light Scattering by Ice Crystals," J. ATMOS. SCI., Vol 27, No 8, 1970. 29. Irvine, W. W., Follack, J. B., "Infrared Optical Properties of Water and Ice Spheres," ICA1tUS, Vol 8, 1968. - 30. Lenggenhager, K., "Eine Meteorologische Erscheinung. Z. Erklarung der - "unchten" Spektralfarben des Sonnenhalos von 22� Radius," UNIVERSUM, " Vol 30, No 3, 1975. 31. Liou, K. N., "Transfer of Solar Irradiance Through Cirrus Cloud Lay- ers," JGR, Vol 78, No 9, 1973. 32. Liou, K. N., "On the Radiative Properties of Cirrus in the Window Re- . , gion and Their Influence on Remote Sensing of the Atmosphere," J. ATMOS. SCI., Vol. 31, No 3, 1974. 33. Liou, K. N., "lfieory of the Scattering Phase-Matrix Determination ; for Ice Crystals," JOSA, Vol 65, No 2, 1975. 34. Liou, K. N., Baldwin, R., Kaser, T., "Preliminary Exgeriments on the Scattering of Polarized Laser Light by Ice Crystals," J. ATMOS. SCI., - Vol 33, No 3, 1976. 35. Morita, Y., "Scattering Cross Section of Freely Suspended Ice Crys- ; tals for Visible Light," TENKI, Vol 20, No 3, 1973. - 36. Nikiforova, N. K., Pavlova, L. N., Petrushin, A. G., Snykov, V. P., Volkovitskiy, 0. A., "Aerodynamic and Optical Properties of Ice Crystals," J. AEROSOL. SCI., Vol 8, No 3, 1977. 37. Ohtake, T., Huffman, P. J., "Visual Range in 7ce Fog," J. APPL. METEOROL., Vol 8, 1969. 162 FOF OFFICIAL USE OtdLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY 38. Pembrook, J. D., Gryvnak, D. A., Burch, D. B., "Attenuation of Solar Radiation by Natural Clouds at Several Wavelengths in the Visible and Infrared," JOSA, Vol 61, No 4, 1971. 39. Plass, G. N., Kattawar, G. W., "Radiative Transfer in Water and Ice Clouds in the Visible and Infrared Region," J. APPL. OPT., Vol 10, No 4, 1971. 40. Sassen, K., "Depolarization of Laser Light Backscattered by Artif- icial Clouds," J. APPL. METEOROL., Vol 13, No 8, 1974. 41. Sassen, K., "Laser Depolarization 'Bright Hand' from Melting Snow- flakes," NATURE, Vol 255, 1975. 42. Sassen, K., "Backscattering Cross Sections for Hydrometeors: Measure- ments at 6328 A," J. APPL. OPT., Vol 17, No 5, 1978. 43. Schaaf, I. W., Williams, D., "Optical Constants of Ice in the Infra- red," JOSA, Vol 63, I?o 6, 1973. 44. Thucnan, W. C., Brown, A. G., "Preliminary Studies of Intensity af Light Scattering by Water Fogs and Ice Fogs," SCIENCE, Vol 120, No 3128, 1954. - 45. Zander, R., "Additional Details on the Near-Infrared Reflectivity of Laboratory Ice Clouds," JGR, Vol 73, No 20, 1968. 163 Fnu nFFTrrer TiCV l1TTT V APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 SzXTIETH BIRTHDAY OF SOLOMON MOISEYEVICH SHMETER Moacow METEOROi,OGIYA I GIDROLOGIYA in Russian No 3, Mar 80 p 122 [Article by personnel of the Central Aerological Observatory] [Text] Solomon Moiseyevich Shmeter, a well-known scientist in the field of atmospheric physics, a Doctor of Physical and Mathematical Sciences, _ head of the Division of Cloud Physics and Atmospheric Dynamics of the Central Aerological Observatory, marked his 60th birthday on 10 March _ 1980. Solomon Moiseyevich came to meteorology af�er graduating from the Mechanics- Mathematics Faculty of Moscow State University and then Khar'kov Hydro- ~ meteorological Institute. During the period 1944-1947 he worked at the Yak- utsk Geophysical Observatory, first as an aerological engineer and then as head of the aerology division. In 1947 he undertook graduate studies at the 164 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY ' Central Aerological Observatory. S. M. Shmeter was the first graduate stu- dent at the obaervatory and after then all his scientific activity over a period of 32 yeare hae been closely aseociated with the Central Aerological - Observatory. S. M. Shmeter dealt with a very broad range of problems during these years. Original investigations of the chemical composition of clouds were the sub- ject of his Candidate's dissertation, which he successfully defended in 1951. During 1951-1957 he carried out a series of important theoretical and experimental studiea devoted to an improvement in aerological measure- mente. Since 1957 the scientific activity of Solomon Moiaeyevich for the most part hae been concentrated on experimental investigations of atmo- ~ spheric turbulence and cloud physics. His !nany years of investigations of the dynamics and mesostructure of convective clouds later served as a ba- sis for his doctoral disaertation, successfully defended in 1968. The book by S. M. Shmeter, entitled FIZIKA KONVERTIVNYYE OBLAKOV (Physics of Con- - vective Clouds), received.broad international recognition. S. M. Shmeter repeatedly headed flight expeditions and participated in flights carried out under highly complex meteorological conditions. Dur- ing theae flight expeditions he created a new method for aircraft inves- tigationa of the atmosphere and obtained unique materials substantially enriching our concepts concerning turbulence, convection, and especially - the physics of convective clouda. The results of the studies of S. M. Shmeter have been published in more than 60 scientific articles. He is the author and coauthor of six mono- graphs on cloud physics and atmospheric turbulence. Three of his mono- graphs have been translated and published abroad. The rich experience of thia working aerologiat and At the same time hia excellent masL�ery of - theory assisted him in creating a text for uae in the aerology course (1965), he being the coauthor. Solomon Moiseyevich devotes great attention to teaching work. Under his , direction several Candidate's dissertations were prepared on different aepects of atmospheric physics. S. M. Shmeter is engaged in much public and political work. A member of the CPSU aince 1945, he devotes much time and energy to the indoctrina- tion of young scientists. Now S. M. Shmeter is at the height of his creative forces and we wish him good health and retention of the same vigorous activity and further creative successes. 165 FOR OFFTrrAT, T1GF nNr.v APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR ur'r'lulAL uoL unLY A'1' THE USSR STATE COP4fITTEE ON HYDROMETEOROLOGY AND ENVIRONMENTAL MONITORING Moscow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 80 pp 122-123 [Article by V. N. Drozdov and V. M. Voloshchuk] _ [Text] A regular session of the Scientific Council of the State Committee on Hydrometeorology and Environmental Monitoring on the problem "Study of - the Oceans and Seas" was held on 28 November 1979. A report entitled "Preliminary Results of an Interdepartmental Expedition ~ - _ in the Baltic Sea and Prospects for its Hydrometeorological Investigation" was presented by I. N. Davidan. In his report he noted that the Leningrad Division of the State Oceanographic Institute and the Northwestern Admin- istration of the Hydrometeorological Service, within the framework of the Commission "Study of Hydrology and Water Contamination of the Baltic Sea," - during 1978-1979 carried out much work for bringing together the efforts of different departments in the field of organization and implementation of comprehensive investigations of the Baltic Sea, a generalization of the reaults of two interdepartmental eapeditions. The expeditions were carried out to study the most timely problems of the Baltic Sea (processes on a synoptic scale, variability and spatial nonuniformity of contamination fields) in accordance with programs developed at the Leningrad Division _ of the State Oceanographic Institute. ; Due to the close contacts between the Leningrad Division of the State Oceanographic Institute, the Northwestern Administration of the Hydrometeor- ological Service and the Arctic and Antarctic Scientific Research Institute, the following organizations became involved in the expeditions: Baltic Sci- entific Research Institute of Fisheries, Central Scientific Research Inst- tute of Geodesy, Aerial Mapping and Cartography, All-Union Scientific Re- search Institute o� Ma.rine Geology and the TEF and GS DKBF [expansions un- - known]. The expeditions were carried out successfully and made it possible to obtain new, original results, important for study of the Baltic Sea. Then the speaker discussed in detail the results obtained from the observa- tions and those preliminary conclusions which have been drawn. In conclu- sion he analyzed the principal directions in hydrometeorological investiga- tions in the Baltic Sea. 166 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY The Council noted that the results obtained from the interdepartmental ex- peditions in the Baltic Sea and also the results of intradepartmental eaped- itions in Tallinn Gulf, in which several shipa and aircraft participated eimultaneously, make it poseible to coneider such an organization of field investigationa to be the most rational in light of new tasks for preserva- tion of the environment. In its resolution the Council approved the organization and implementation of auch work and recommended that the Leningrad viviaion of the State Ocean- ographic Institute prepare materials on the organization and implementation of such expeditions for extending the experience to other basins. The coun- cil also decided to consider an interdepartmental program for investigations _ of the Baltic Sea as a unified geographical feature over the extended period - of 15-20 yeara (Project "Baltika"). A representative of the State Oceanographic Institute, A. N. Ovsyannikov and a representative of the Arctic and Antarctic Scientific Research,Institute, A. T. Bozhkov reported on the second problem "Status and Prospects of De- - velopment of the Sea Coastal Network." After hearing and discussing the reports, the Council noted that the marine administrations of the Hydrometeorological Service are devoting great at- - tention to supporting the normal work of the marine hydrometeorological net- work. The quality of observations in the marine network on the whole is good. The State Oceanographic Institute and the Arctic and Ar.tarctic Scientific Reaearch Institute are carrying out a systehiatic checking of operation of the marine network and are rendering the necessary assistance. During 1980 the State Oceanographic Institute, in collaboration with the Arctic and Antarctic Scientific Research Institute, Far Eastern Scientific Research Hydrometeorological Institute and hydrometeorological observator- ies of marine administrations of the Hydrometeorological Service will carry out scientific research work on the theme "Development of a Scheme for the Distribution of the Marine Coastal Hydrumeteorological Network in the Seas of the USSR," which will make possible a scientific validation of the prospects for development of the marine network. V. N. Dro zdov An unofficial conference of experts on statistical planning of the Precip- itation Enhancement Project was held at the State Committee on Hydrometeor- ology during the period 29 October-2 November 1979. It was devoted to the International Precipitation Enhancement Project being carried out in the basin of the Douro River (Spain). The scientific objectives of this project were defined by the 28th Seseion of the Executive Committee WMO in 1976. There have already been a series of unofficial conferences of experts at 167 T/~T I~TTTnT JT n~~ t� APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 which there was a discussian of different aspects of the Precipitation En- hancement Project. The materials of these conferences were given in dif- ferent WMO bulletins and reparts (Weather Modification Programme. Precip- itation Gnhancement Project). - At the conference of experts on statistical planning of the Precipitation ' Enhancement Project there was discussion of the present status of work on the project, the specialists analyzed the aspects of statistical planning and evaluation of fielci experiments with artificial modification for the purpose of obtaining additional precipitation and formulated recommenda- tions for further preliminary investigations which must be made for the purposea of the Precipitation Enhancement Project in the Douro basin. The Soviet delegation familiarized the foreign specialists with the prob- lems involved in the planning of field experiments for enhancing precipi- tation in the steppe region of the Ukraine and in the basin of Lake Sevan, with the results af experiments already carried out, and also with plans for further improvement of statistical evaluations of the results of field experiments being carried out at the present time in the USSR under the direction of Academician A. N. Knlmogorov. Foreign specialists presented concise information on studies discussed at the Sixth Conference on the Use of Probabilistic and Statistical Methods in the Atmospheric Sciences held by the American Meteorological Society during the period 9-12 October 1979 in Canada. The conference materials will be published in the next WMO bu'Lletin (Weather Modification Programme. Precipitation Enhancement Pro- ject). V. M. Voloshchuk 168 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY CONFERENCES, MEETINGS AND SEMINARS Moacow METEOROLOGIYA I GIDROLOGIYA in Russian No 3, Mar 80 pp 124-128 [Article by L. S. Speranskiy, K. Ya. Vinnikov, N. F. Dement'yev and V. D. - Komarovj _ [Text] An All-Union Conference on "Hydrodynamic and Statistical Methods for Local Weather Forecasting and Problems in Mesometeorology" was held in Novosibirsk during t%e period 15-19 October 1979. '1'he conference was attended by representatives of 15 acientific institutes and colleges in the country. A total of 36 reports were presented. A large group of repoxts was devoted to hydrodynamic methods for local weather forecaeting. Here, in particular, we should mention a report by Academician G. I. Marchuk, V. V. Penenko, A. Ye. Aloyan and G. L. Lazri- yev (Computation Center Siberian Department USSR Academy of Sciences) on nwnerical modeling of the microclimate of cities, with anthropogenic fac- turs taken into account. Individual aspecte of thie problem were examined in reports by A. Ye. Aloyan, D. L. Iordanov and V. V. Penenko, entitled - "Numerical Modeling of Tranaport of Passive Admixtures," A. Ye. Aloyan, G. L. Lazriyev and V. V. Abramenko, "Methods for Taking the Turbulent En- ergy Equation into Account," and A. Ye. Aloyan and G. V. Isayev, "Propaga- tion of an Admixture in the Surface Layer from Fixed and Moving Sources." A series of reports presented by specialists of the USSR Hydrometeorolog- ical Center was devoted to studies for creating a local forecasting model on the basis of deep convection equations. A numerical algorithm for in- tegrating a system of deep convection equations and investigating the properties of this system was diacuased by V. Z. Kisel'nikov, Ye. M. Pekelis and D. Ya. PreBSman. A report by M. Alautdinov and N. P. Vel'tishch- ev presented a method for computing radiation heat influxes and gave the = results of modeling of the diurnal variation cf ineteorological elements under moist deep convection conditions. A. A. Zhelnin and A. A. Bregman told of an algorithm for the parameterization of microphysical processes related to the formation of clouds and precipitation. V. M. Losev reported on computation of stationary mesoscale temperature and wind fields on the baeis of synoptic information using linearized deep coavection equations. - 169 FnD nL`L`rnTAT ifc`c nIkrrv APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 . ~ . _ _ P. Yu. Pushistov, L. S. Speranskiy, T. A. Toloknova and Ye. G. Sementsova (West Siberian Regional Scientific Research Hydrometeorological Institute) reported on a model of local forecxsting for a lowland territory and pre- sented some results of forecasts made on the basis of this model. V. K. Arguchintsev (Irkutsk State University) told about a two-dimensional model `l for investigating breeze effects, constructed without the simplification of quasistatics. B. G. Vager (Leningrad Institute of Construction Engin- eers) reported on allowance for horizontal diffusion in a model of the eurface layer intended for calculation of the heat and moisture transfer reglme in the preaence of inhomogeneities of a small ].inefir scale (canal in the desert, flight atrip). His second report was devoted to experience in the use of cplines in boundary layer problems. M. S. Akhmetov (Pskov State University) told of some aspects of interaction between open pit mines and the atmosphere. A number of reports were devoted to an investigation of convection and tur- bulence. R. S. Pastushkov and S. A. Vladimirov told about numerical model- ing of the convective cloud cover accompanying a squall. G. G. Goral' and 0. I. Chapovskaya (High-Mountain Geophysical Institute) reported on exper- imental investigations of convective pracesses accompanied by squalls. The synoptic and thermodynamic conditions determining the localization and intensity of convection in the Northern Caucasus were eaamined by L. M. Fedchenko and V. A. Belentsova (High-Mountain Geophysical Institute). The report of A. N. Koval`chuk and T. N. Terskovoy (High-Mountain Geophysical Institute) was devoted to an investigation of the dynamics of cumulonimbus clouds and the conditions for the formation of well-developed hail cells. An investigation of a model of a cloud ensemble on the basis of GATE data was preaented by A. I. ral'kovich (USSR Hydrometeorological Center). The subject of two reports by P. Yu. Pushistov, V. M. Mal'bakhov and S. M. Kononenko was the numerical modeling of penetrating convectton in the boundary layer dnd an inves~igation of two-level convective ensembles and the mechanism of formation of cumulus clouds. A repart by A. S. Gavrilov (Leningrad Hydrometeorological Institute) was devoted to an investigation of the convective boundary layer using a dif- ferent tool: systems of Fridman-Keller moment equations. In this report, as in the communication of the above-mentioned authors, great attention was devoted to the problem of antigradient heat transfer. A. G. Tarnopol'- skiy (State Oceanographic Institute) and V. A. Shnaydman (Odessa Hydro- meteorological Institute) made a comparative investigation of the charac- teristics of turbulence obtained employing two closing methods: on the basis of an equation for the rate of dissipation of turbulent energy and using the generalized Karman hypothesis. An investigation of the struc- ture of vertical movements in the boundary layer on the basis of experi- mental data was the subject of a report by Yu. P. Perevedentsev and Yu. G. Khabutdinov (Kiev State University). The reaults of experimental in- vestigationa and methods for computing turbulence in the layer 50-500 m were reported by V. N. Barakhtin and E. A. Morozov (West Siberian Regional Scientific Research Hydrometeorological Institute). 170 FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200090007-3 FGR OFFICIAL USE ONLY A report by B. N. Sesgeyev (Central Aerological Observatory), tell ing about a two-dimensional model of an atmospheric front, devoted great at- tention to the parameterization of phase transitions and the formation of precipitation. A model of formation of a frontal zone, taking into ac- count nonadiabatic factors, was examined by B. Ya. Kutsenko (Central Aero- logical Observatory). A report by Z. N. Kogan and N. Z. Pinus (Central Aerological Observatory) was devoted to an evaluation of the budget of the principal types of energy in a middle-latitude cyclone. The subject of a aecond communication by Kogan was a atudy of the kinetic character- istics of a cyclone on the basis of radiosonde data. Inveatigations of the interaction between the boundary layer and the free atmosphere for processes on a subsynoptic scale were reported by V. A. Shlychkov and P. Yu. Pushistov. T~ao reports were devoted to problems involved in flow around barriers: "Lee Wave Resistance Exerted by a Real Mnuntain System" V. N. Kozhevaikov, N. N. Zidlev and N. N. Pertsev (Moscow State University), "Numerical Mod- eling of the Problem of Flaw Around Obstacles by an Incompressible Fluid" I. G. Granberg (Institute of Physics of the Atmosphere). A number of reports examined models of prediction of local phenomena on the basis of statistical methods. The methods for ordering and sampling of parameters in local alternative models were investigated in a report by T. A. Anikina, G. M. Vinogradova and L. N. Romanov (West Siberian Regional Scientific Research Hydrometeor- ological Institute). One atatistical forecasting method, based on the ex- prinHlon of the FieldH of ineteoroloAical elemente into serieA in u syAtem of linezirly independent Eunctiona, was discusaed by V. V. Kostyukov (West Siberian Regional Scientific Research Hydrometeorological Institute). A _ model of prediction of thP frequency of recurrence of breezes was examin- _ ed in a report by E. A. Burman and F. Ya. Stupina (Odessa Hydrometeorolog- _ ical Institute). The following reports of specialists of the West Siberian Regional Scientific Research Hydrometeorological Institute were also de- voted to the prediction of individual weather phenomena: M. Ya. Kogan and L. N. Romanov "Quasilinear Models for Prediction of Diurnal Temperature Variation," G. G. Polyakov and Z. V. Torbina "Prediction of Precipita- tion Using a Piecewise-Linear Model," D. I. Zenkevich "Prediction of Visibility in," I. P. Prokop'yev "Prediction of Signif icant Snowfalls in the Southeastern Part of Western Siberia." " L. S. Speranskiy A bilateral Soviet-American symposium on the modeling of climate, climatic changes and the statistical processing of climatic data was held at Tbil- isi during the period 15-22 Octaber 1979. - It was the sixth of a series of symposia held during 1976-1979 within the framework of project 02.08.11 "Effect of Changes in the Heat Balance ,-)n Climate" of Working Group VIII ("Effect of Environmental Change on Climate"; 171 nnn nnnrnrAT y.r.,. n+.... APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000200094407-3 rVn vrri%..Lnu - of the Soviet-American Commiasion on Scientific Cooperation in the Field - of Preservation of the Environment. The project is headed on the Soviet side by Corresponding Member USSR Acad- emy of Sciences M. I. Budyko, and on the American side Professor W. L. Gates. Well-known American scientists participated in the symposium: W. L. Gates, E. N. Lorenz, R. D. Sess, R. M. Chervin, S. Esbensen, M. Halem, I. M. Held, R. D. Jenne, R. Madden, G. R. North, A. Robock, M. J. Suarez, J. M. Wal- lace, E. R. ;.eiter and E. W. Bierley. On the Soviet side there were about 30 specialists, for the most part from the inatitutes of the State Com- mittee on Hydrometeorology and the USSR Academy of Sciences. The symposium program included 37 reports, including 20 reports of Soviet - - scientists and 16 reports of scientists of the United States delegation. = One report was presented jointly by the scientists of tne USSR and the United States. The most general problems were considered in the reports of M. I. Budyko and E. N. Lorenz. The report of M. I. Budyko contained a synthesis of modern concepts con- cerning impending changes in global climate and the biosphere under the _ influence of anthropogenic change in the content of carbon dioxide in the _ atmosphere. The report expressed the assurance that already at the pres- ent time it is possible to prepare forecasts of climatic change necessary for working up long-range economic development plans. The problem of the predictability of behavior of a climatic system was discussed in a report by E N. Lorenz. Touching upon the problems involv- ed in prediction of climatic changes in relation to the increase in con- tent of atmospheric C02, E. N. Lorenz surmises that this is precisely the type of climatic problem whose solution will be most feasible by ap- plication of the numerical modeling method. Empirical evaluations of recent changes in the thermal and ice regimes of the northern hemisphere were examined in ajoint report of specialists of four institutes of the StaCe Committee on Hydrometeorology: K. Ya. Vin- nikov, G.V. Gruza, V. F. Zakharov, N. P. IGovyneva, K. M. Lugina and E. Ya. Ran'kova. It was demonstrated in the report that beginning from the middle 1960's there has been a warming of the northern hemisphere with a mean intensity 0.1-0.2� C/10 years. This warming is accompanied by a sub- stantial reduction in the total ice content in the Arctic Ocean. In a report by A. Robock, on the basis of an analysis af empirical data characterizing the change in the mean temperature of the northern hemi- - sphere during the last 400 years, carried out using a numerical zonal model of a climatic system with parameterized dynamics, the author - 172 - FOR OFFICIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2047102108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY refuted the hypothesis that "solar activity" is one of the imporctant fac- tors in change in global climate. Empirical data on recent changea in climate and the results of a statis- tical analysis and generalization were also presented in the reporta of E. R. Reiter, K. Ya. Vinnikov and P. Ya. Groysman, G. V. Gruza and E. Ya.Ran'kova, R. D. Madden, 0. A. Drozdov, J. M. Wallace and D. S. Gutsler, Ye. M. Dobryshman, M. I. Fortus and Ya. M. Kheyfets, N. V. Koby- sheva and L. P. Naumova, L. S. Petrov, et al., V. N. Adamenko, Yu. L. Rauner. The climate modeling problem was discussed in more than ten reports. Eval- uations of the influence of an increase in the content o� C02 in the at- mosphere on the climatic regime, obtained using the Oregon University model, were presented in a report by W. L. Gates. Although the model in- cludes a realistic description of the topography of the earth's surface and the seasonal variation cf solar radiation, these evaluations cannot be used for predicting c'limatic changes, since in this investigation the temperature of the ocean surface is considered stipulated and does not change with an effect on the climatic system. 'Che necessity for allowance for the influence of thermal inertia of the ocean on the change in air temperature accompanying an anthropogenic in- crease in the atmospheric content of C02 was demonstrated in a report by R. D. Sess. In another report by this author it was stated that the pro- cess of global warming can be accelerated as a result of the increasing anthropogenic discharge of the compounds C0, NOx, CH4, forming from fuel combustion. In a report read by V. P. Dymnikov, Academician G. I. Marchuk and his col- leagues gave detailed information cuncerning the status of studies on con- struction of models in the theory of climate at the Computation Center Siberian Department USSR Academy of Sciences. I. V. Trosnikov told about a model of general circulation of the atmosphere developed at the USSR Hydrometeo rological Center. It was demonstrated in a report by V. P. Meleshko and R. T. Vezerold that allowance for the real geographical distribution of cloud cover exerts a substantial influence on the local characteristics of the climatic and circulatory regime of the atmosphere. Important problems in study of the laws of behavior and sensitivity of a global climatic system by means of models were also examined in reports by Corresponding Member USSR Academy of Sciences G. S. Golitsyn, G. North, M. Suarez, I. L. ICarol' and I. M. Held. The reports of V. A. Aleksandrov and V. Ya. Sergin, R. M. Chervin and M. Halem were devoted to a study of the statistical characteristica of the meteorological regime, reproducible using models of general circulation 173 'VAD /1VL�TIIT AT iinn /%%n lr APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000200094407-3 of the atmosphere. The studies of these authors give additional informa- tion on how realistic climatic models are and make possible a more cor- rect formulation of numerical experiments for evaluating the sensitivity of cl.imate to eaternAl effects. Without dwelling on the other reports (the materials contained in them have been partially published), we note a paper by I. M. Held, which was discussed from the floor, telling about a new study by Manabe and Weth- erhold which gave an evaluation of the influence of doubling and quadrupl- ing of atmospheric C02 content on the mean annual climatic regime over the continents and oceans separa*_ely. According to these evaluations, the warm- ing of climate can be accompanied by a substantial change of moistening conditions i^ agriculturally important regions of the continents. The participants in the symposium expressed the opinion that it is neces- sary to summarize the results of the series of Soviet-American symposia held during the period 1976-1979 at a conference of Soviet and American experts on the problem of the influence of C02 on climate. Upon completion of the symposium this proposal was adopted at a session of the Eighth Working Group of the Soviet-American Commission. R. Ya. Vinnikov The Tenth Conference of the Danubian Countries on Hydrological Forecasts was held in Vienna during the period 11-14 September 1979. It was attend- - ed by delegations from Austria, Bulgaria, Hungary, USSR, West Germany, Czechoslovakia and Yugoslavia, as weZl as two hydrologists from Switzerland and one each from East Germany and the Netherlands. Three international or- ganizations were represented at the conference: Danubian Commission, World Meteorol.ogical Organizatio.L and International Institute of Systems Analy- sis. The total number of participants was 134. The working languages were Russian and German. - In evaluating the organization of the conference work we should note the high level of its preparation and implementation. Before the conference began two volumes of mater.ials were published which contained national _ reports on the methods used in each country for predicting the water - and ice regimes of the Danube (one volume) and the reports presented at the conference (second volume). Also included were author's simmaries of reports and detailed programs of the scientific sessions and field trips. Each conferee received these materials. We should also note the exceptionally high quality of simultaneous translation. All this favored the complete success of the conference. Thirty-four reports were presented and discussed at six scientific ses- sions. Most of these were devoted to the following problems and aspecte of hydrological forecasts for the Danube and its tributaries: short-range 174 FOR OFFIGIAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2047102108: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY forecasting of levels and discharges, long-range forecasting of levels and discharges, forecasting of ice phenomena, forecasting of channel processea and runoff of sediments, evaluation of the accuracy and econom- ic effectiveneas of forecasts. The greatest number of reports (15) dealt with methods for the short-range forecasting of water discharges and levels in the Danube and its tributar- y ies. The basis for the methods is mathematical models of the channel phase of runoff, diffe:ing in complexity, and models known under the name "pre- cipitation-runoff." Among the first of these models attention was centered on an adaptive lin- ear model of channel runoff whose parameters vary continuously in de- pendence on the observed error in forecasts on the preceding days. Pre- computations of river discharges and levels using an electronic computer on the basis of such a"self-adjusting" forecasting method are probably of considerable practical interest. The model was developed by the Hungar- ian hydrologist Sh. Ambrush for the reach of the Danube between Dunafoldvar and Baja, but it naturally has quite general significance. In the fore- casting methods which were discussed in the reports of Kresser, Gutknecht and Dreher (Austria), Engel and Mendel, Schiller and Tissa (West Germany) and Stefanova (Bulgaria) the emphasis was on computations of the travel- time curve of water on the basis of models of a cascade of linear and nonlinear capacitances. In general, these models are quite well known. Nevertheless, the reports were of considerable interest because they fam- iliarized the conferees with some modifications of these models and pro- - cedures for determining their parameters. It is also important that river reaches with a cascade of reservoirs were among the investigated objects. - "Precipitation-runoff" models were discussed in reports by Lutz, Koch and ~ Rosemann (West Germany), Lukachova (Czechosluvakia), Georgevic (Yugoslav- ia) and Nachtnebel (Austria). In scientific respects the most interesting - reports were an investigation of the influence exerted by the intensity of precipitation, runoff losses and some other factors on the form of the - unit hydrograph (report by Lutz) and a nonlinear model of the "runoff- precipitation" type (reports by Lukachova and Nachtnebel). In the reports presented at the preceding conference in Budapest it could be seen that in the "precipitation-runoff" model there is still a part ~ which has not been adequately developed; reference is to computations of runoff losses with time. Unfortunately, this situation still persists, as follows from the conference reports in Vienna. The reason for this is the _ absence of extensive investigations af the problem of infiltration of rain water, especially investigations on the basis of experimental and representative basins. The use of a mathematical approach alone does not solve this complex problem. Attention was given to this circumstance _ at the final session of the conference. The reports on the considered problem also contained important informa- tion relating to remote measurements of hydrometeorological elements and the automated collec*ion of the results of these measurements at the 175 Ff1R !1VFTf�TAT TTCF r1NTV APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102108: CIA-RDP82-00850R000240090007-3 rviN. ul i forecast preparation center. Since the maximum possible advance time for predicting rain-induced high waters in many cases is less than 24 and even 12 hours, an acceleration of the collectior of initial data for preparation of a forecast is of very great practical importance. A pre- diction of precipitation, to be sure, makes it possible to increase the advance time for predicting high water. In this connection informational reports on precipitation forecasts were presented at the conference. - Ir should be noted that in connection with the construction of a large nur~lber of reservoirs on the Danube tributaries the problem of their con- Lrol ts becoming increasingli, ucute. For this reason, and Also in rela- tlon to the prob~.ems involved in monitoring water quality (two reporta from the USSR and Bulgaria were devoted to water quality problems), short-range Forecasts are acquiring ever-greater importance. Naturally, the prediction of exceptionally high waters, constituting a danger for the population and enterprises along the shores, also retains its im- portance in the presence o'L reservoirs. At this conference there was not a single report on the prohlem of long- range forecasting of low levels of the Danube, having great importance for navigation. For that reason particular interest was shown in a re- port by V. M. Mukhin (USSR Hydrometeorological Center) on a method for predictinR the 1 evels of the Danube for 10-20 days in advance. The basis - of the method is a mathematical model which takes into acc:,unt the role of water distribution along the length of the river and the role of the dtstribution of precipitation over the area of the basin in the formation of c�iter discharges at the lowest-lying station on the river. Such fore- casts make it possible to know the depth in the main channel during the - mc,vemeiit of a ship in the Danube reach from Bezdan to Tulcea, with an ex- tent oC about 1,400 km. The report caused intereat as well due to the - fact that for solving the nroblem V. M. Mukhin used a modern approach (component analysis, theory of solution of incorrectly formulated prob- lems). - Four reports were devoted to a superlong-range forecasting of the water voltune in the Danube and an analysis of its changes over a period of many years. 'i'hey did not contain new approaches to solution of the prob- lem oF superlong-range forecasting of river runoff. We note that among the hydrologists of the Soviet Union the opinion prevails that there is _ no firm scientif ic basis for the superlong-range forecasting at this - time and that investigat'ons of this problem have a strictly explora- tory nature. The Soviet delegation presented no reports on the super- long-range forecasting of river volutae, although at the preceding con- ference at Budapest we did present reports on this problem. _ Six reports were presented on the problem of channel processes and the runoff of sediments. However, only the reports from the Soviet Union = were devoted to a quantitative pred iction of river channel deformations (g. F. crishcherko, State Hydrological Institute and V. N. Mikhaylov, 176 FOR OFFIr,IAL USE ONLY APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200090007-3 APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200090007-3 FOR OFFICIAL USE ONLY Moscow State University). The reports on sediments, in essence, contain- ed no methods for predicting the solid runoff of the Danube and its trib- utaries. Neverthelesa, they were interesting since they gave the resulta of investigations which without question will be used in developing meth- ods for predicting the runoff of sediments. For example, in a report pre- _ sented by Bulgaria there were data characterizing the distribution of the granuiarity of bottom deposits in a reach of the Danube with an ex- - tent of about 2,000 kil.ometers and casting light on the evolution of the c eompoHitton oC depoAits alonR the lenRth of thiR river. The acouRtLe _ method for meayuring bottom sedimenta is quite novel; it was deecribed in a report by Sehlagge (West Germany). It is well known that the groblems involved in predicting the ice regime of rivers in the Danube basin are being studied primarily in the Soviet - Union. It is therefore not surprising that only our reports were present- ed at the conference: one about thP hydrological conditions of winter navigation and another about the method for long-range forecasting of the duration of ice phenomena and the times for clearing of the Danube _ from ice. There were only two special reports on the problem of evaluating the econ- - omic effectiveness of forecasts. The matter of economic effectiveness was _ touched upon in a number of other reports in which data were cited quanti- - _ tatively characterizing the advantage gained from different specific fore- casts of the water volumes in the Danube and its tributaries. Upon complet.ion c>f the scientific program of the conference an excursion w;iq orKiinlzed for its piirtic-iPunts to the Danuhe Me]k and ihba-PersenhoyK llydroelectric I'uwer S