JPRS ID: 8237 EXPERIMENTAL SEISMIC STUDIES OF THE EARTH'S INTERIOR

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000100010044-1
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 179
• Document Creation Date: November 1, 2016
• Sequence Number: 44
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
23 'JAlillARY 1979
;FOVO
I

i OF 2
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
t'UH Ut't'11,1HL UJt UIVI.Y
JPIt5 L/8237
23 January 1979
EXPERIMENTAL SEISMIC STUDIES
OF THE EARTH'S INTERIOR
U. S. JOINT PUBi.ICATIONS RESEARCH SERVICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
NO'1'L
JpEt5 publicntiong conCuin information primrily from foreign
newspapcrs, periodiCals and books, but also from news agency
transmisgions und brottdcasCs. Materials from foreign-lnngunge
sources are translrated; Choye fram Cnglish -languuge sources
arc transcribed or reprinted, wieh ttie original phrgsing nnd
other characteristics retnined.
}leadlincs, editorial reports, and maCerial enclosed in brackeCs
(I are supplicd by JPEt5. Processing indicators such as (TexC)
or (ExcerptJ in Che firsr line of each iCem, or following the
lxse line of a brief, indicute how the originnl information was
processed. Where no processing indicator is given, rhe infar-
mation was summarized or exCracted.
Unfamiliar names rendered phonetically or transliterated are
enclosed in parenClieses. Words or names preceded by a ques-
tion mark and enclosed in parentheses were not clear in the
original but have been supplied as appropriaCe iii context.
OCher unattributed parenthetical notes within Che body of an
item originate with the source. Times within iCems are as
given by sourcc. lk
The contents of this publication in no way represent the poli-
cies, views or atCitudes of the U.S. Government.
PROCUREMENT OF PUBLICATIONS
JPRS publications may be ordered from the National Technical
Information Service, Springfield, Virginia 22151. In order-
ing, it is recommended that the JPRS number, title, date and
author, if applicable, of publication be cited.
Current JI'RS publicAtions are announced in Government Reports
Announcements issued semi-monthly by the National Technical
Information Service, and are listed in the Monthly Catalog of
U.S. Government Publications issued by the SuperinCendent of
Documents, U.S. Government Printing Office, Washington, D.C.
20402.
Indexes to this report (by keyword, author, personal names,
title and series) are available through Bell & Howell, Old
Mansfield Road, Wooster, Ohio, 44691.
Correspondence pertaining to matters other than procurement
may be addressed to Joint Publications Research Service,
1000 North Glebe Road, Arlitigton, Virginia 22201.
~
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

EPORT DOCUMENTATION
~S.pEPOpf N0,
3, Rae1plenCf Ateoftion No.
PAGE
.y~
F
JPRS L/8237
~
1111� .
nd tuODUe
S. IIepoA 0~1~
CXPE;R [MI;N''Af, 5l; I SM [C S'I'UI)I I.S 0I.' 'I'NL I:Att'CH' 5 INTCItInR
2 3 Janua 19 79
IL
1. Auphor(U
L. V. AnConovtt, F. P. Antikayev, et al.
P~rfarmint Or~~nltatlon q~pl, P1o,
. Pefermine Ottanltaflon Name end Addnf�
10. Pro1601T.6M/Work Unll No,
Joint PublicgCions Research Service
1000 N. Glebe Rd.,
u. conir. - q X' ++1 p> =o+, xhere o2 i s the stundard deviation, we have
C (t) a 210' - K (tll.
As t-+m, K(t) -*0, whence CH = 2a2, and the normalized correlstion rstio is
K CtW - 1-- C(t)12o9.
Fbrmuln establishes gsimple relution between the gtructure function
:ind thc rorrelation functton: uhen t= 0, C(t) 0, and K' ( t) = 1; when t
C(r)=2o2 und Y.'(t)-+0.
M'or t.lit! a:riltr nP convenience rre nrbitrari ly denate the structure funrtion of
fluctuntionr. d 1g A, bL, n4 by , , . Fig. 6 shows the
atrur.t�re t"unctionn for fluctuations or Lrsvel time of the P aave, end
21
FOR OFFICIhL U5E ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOtt OFFtCtAL U5ti ONLY
ddtW ydN11,0141
~
\
~
;
~.....,n ~e
d
t/ !D /i0 IYA I/0 A.~~
Fig. 6. StruGt,urt funetidn of fluctuatidng in trtLve1 time and varintiono
3.n thicknonr, of the earth'a erust for the territnry of the LASA group in
nccordance With datu of H. Iyer and J. i{eti1y (1972). 1--varigtiong in
r,hi(iknenn of the earth's cruat; 2--fluctuations of travel tim; 3--gmoothed
s;tructure functioa n!' varigtiofls in crunt thickness
ulso thc structure function tddN2> for vnrigtiong in thieknegs of the
et1I'th'r crust bii fAccdrding tn dgtn of the LASA grdup nnd 20 portable
t15GS stntinns (tyer, Hea1y, 1972). An r_nn be geen, thp etructure, f'unction has an abaolute maximum at a flistence
- c,t' 300 km und n nerfen oP relatively 1du extrema. IP we consider the
:;moothed funetibn ea6l17y we etLn see that the first maximum eames at a distan
ce
equal to 100 km, and the firgt minimum is c1ose to 180-190 km. Thie indfcates
that in the change of crustsl thieknegs in the territory of the LABA group
n chararterintic rhythm cen be observed thtLt has a period of the ordpr of
1$0 km nnd accordingly g haif-perind of tihe order of 90 km. In eddition,
or,her charncterigtic rhythms are obgerved ag we11.9 but leas pronounced, e. g.
of the ordrr of 35, 110 km, etc.
I� We look at the structure functian of fluctuationg of travel time of the
P Wave, We can see that it is p ractically identical in shape to the func-
tion . Zhis indicates a direct genetic relatioi, betWeen fluctuations
o: trnvel time of the P Wave and variations in the thickness of the earth's
crust, ttnd gppnrently the upper mantle.
1000 and A < 1000 km. Tn fact, if expression is used to
upproximnte the amplitude curve of the Lg wave, assuming that it is related
to higher modes of surface wn.ves propagating with minimum group velocity,
then thr be3t agreement of' the theoretical curve with experimental data is
observed for Y= 0.20 when A < 1000, and Y= 0.13 when A > 1000 km. Taking into
con,ideration that T is approximately 3 s, V= 3.57 km/s for A 1000 km, we get for tFie f.-rst case a Q of approxi.mately
160, iind for the yecond case a Q of approxirnately 250. Such a simplified
cstimate nliows us to assume that the Lg wave recorded at distances A> 1000 km
prnpagntes in a relatively high-Q medium.
TABLE 2.
Va1ues of crslibration functions QM(0) for longitudinal Pg and PV waves from
from shallow (H ;
2) evnlunting the relation betWeen the given scale and knoti+n scales thgt
have been plotted With respect to other types of Wavea.
7'he time interv$ls in Which the umplitudeg of seismic aaves Were measured
aere chocen in accordgnce with the plotted hodograph.
On the section up to 15,000 km the maximn of the F'KP uaves lie no further
ttian 10 s beyond the first entry. bn the 15,000-16,000 km section where
waves of the Clt brgnch cnn sometimeu be seen in the first entries, to avoid
errors thc intervxl of inea:surementu Was increased to 15 g after the first
entry rep,qrdless of which branch these entries belong to. Beyond the zone of
intcrference, the intervsl of mcanurements in the first group of Waves
(thc bt' and CH branches) ir bounded by the instant of entry of the PKP2 (AA)
branch.
For the PKP2 brosnch, the time interval in Which amplitude maxima is observed
i:; somer+hut greater than for waves of the first group, and averages 20 s
nfter Lhe first entry of the PKP2 group.
Datu processing uez done separately for the two differentiated groups. 7'he
A/T vnluer, were rcduced to the level of rrtb = 5.5, and were then averaged in
35
- FOR OfFICIl,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fblt UN'FIGIAL U5E ONGY
hurnfl-F3d-ki ].r,mr_ter i rt,ervatc; Ot' rp1 r.entrti1 dtr,tEtnCe. 7'he regulttant curve
arui t;}in bFU;ir; t'ar n c1t1 it1rnti0n Gurvh in tinftn ni' the U. S. magnitu~e tiC�L1e.
'!'o r.heck qatitit'riCt,iori or t.hc bricair frincipter, or chnntructinn of tna,nitude
;,c~te:;, r eomt,rzri;;c,t1 wtw, madr or the cleviati.ohc or the reduced AIT values
rrom t,he avernrtv? r.urvr on the une hrind, nnd the magnitudes of the corre=
~~pondtrig shvr.kr: on tte ot,her hnn(l. Sur_h cnmpurighns wrre mdde in dif't'erent
ratiges or r_Y,tcentral dintattces. Nv reltition wnn observecl betWeen A/T 8evia=
r,ioris and msatrnitude or epieentral (1istance,
'Rie mensurement:; uere mude ori seismogramn obttLined on standard channeln.
'ilif, materinls Were rroCessed in nuch ft uay thnt the 1eve1 or the amplitude
curves corregpnnded td u mugnftude evn.luntion df 5.5 in mb gGn1e unitg.
f) reduce the valibrntion curves plotted frnm type pKP Wgve$ iaccor8ance
ait.h ttle m~~`t ::cu1e, an empirie~. r.urve rel~.ting the trtb and mpy~ sealeg Wan
u;od (Fi vl. 15). Let us notc that close tn the va1ue mb = 5. 5iacbliflg bf
t.~
ae - i ~ - - - ~
- . I - _t-
QM - - - :
~
o s ` . M
14
F'ig. 15. Relution betWeen the nb (United
5tates) and mW (Gamprehensive Seigmologfca1
Expedition) mggnitude gGaleg
+,he rr,agni tudes td crnvert from one geale tn the other takes place with
mitflimiun errorr, since the formulas obtuined by different authors close ta
t>he value mti =5.5 yield prnCtically identical regultg.
I3,~;;ides, for enrthqualtes with nnrmal focus depth ae hgd magnitude estimates
in unita or the MLN scale (gbout 50 csses). Thege estimates Were also
converted nccording to knoun formulas (Antonova et al., 1968; "Ma itude
and Energy C1$ssificution of Earthquakes" 1974) to units of the M
scale.
FT_
hor some earthquakes, aur etation netuork simultaneousl}r recorded waveg of
+he P rund YKP types (about 40 cases . This enabled a direet comparison of
ttie A/m values for PKp wnves With m magnitude estimstes.
'Tt;e difCerences in the ealibration curve tie-in level found by the three
rV!t.hods did not exceed 0.1 mugnitude unit. Estimatea ccording to the
c~~ri5tructed mnpnttude ocales thst We Will designate m~~, snd m~02 eoineide
~~n t.hr, uveraf;e xith tlie mPV sca1e.
;lie n^curticy or sepsirste determinstion or magnitude by scales for PKP Wgves
i:, approximntcly the same as for F' Wave3. Increased regional variability
or mni;nitude deviation:s is observed only close to the minimum of the cali-
bt'rtLinn func:Lion. 710s crror cnn be eliminated by introducing regional
mrLf;nitude corrections. The c:ulibrntion funetion is given in Tables 4 and 5
ntifi nrl 6'iF;. tG, uhere dntn from reference sources are shown for comparison.
36
FOR OFFICIt,L USE OYLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOIt tlFF'ICIAL U5E dNLY
w
u , tiyhUs� k111
Fig. 16. Calibration curves rdr pKp Wuves: 1--gecording to datg oi' the
Comprehensive Seigmalogirsl Expedition; thp upper branch corregponds to
the pKpI group, the lrn+er td pKP2; 2--aGcording te datg of I. V. aor-
bunovn snd N. V. Shatorngyrg (1976); 3--$ccarding to data of S. Miyemura (1974)
'PAHI,E li
Mngnitude calibrntion curve for F'fCpI Waves
*
ne. e~
ON
1. ~re. ar
Qiy
A, ~ra
Oiy
12,0
~~~0
fl~0
7,20
IS,A
1
00
1l.1
7,80
14.0
7,70
15.0
.
6
80
. 1:,i
1,80
14,1
1e,0
,
6
73
1:,3
t,b
14,2
16,1
,
6
70
1s.4
'
7,e0
14.3
7.15
11,2
,
6
40
12,6
1.10
l4,4
7,10
I6,3
,
6
60
12,6
7,10
14,6
7,10
1e,4
.
4
e0
i:,>
7,73
14,s
1.10
1e,3
,
6
00
+:,e
7,73
f4.7
7,10
Ie.$
,
e6
6
11.9
7.7$
14.8 .
7,10
16,7
,
6
70
11,0
1,70
14,0
7,15
1{,t
,
6
80
is,1
1,70
Is,o
1,20
1e,9
.
�
+a
u,:
1,43
13,1
7,20
17,0
,
1
0
u.s
t,b
is,:
7,23
17,1
,
7
13
11,4
1,36
15,3
7,35
11,1
,
7
25
13,6
7,4$
tS.4
1,73
17,3
,
7
35
1s.$
7.76
fS'S
7,73
17.4
,
l
Irr
I3,1
1.30
I6.4
7,20
17,5
.
1
45
1l,s
7,35
fi.?
7,10
17.6
,
7,4S
*A, Lhour.. km
37
FOR OFFICI6.;. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fntt hFFtCIAL U5E dNLY
mnut,E 5
Cntibrri.tton Cunctinn qM for i1Kti2 waveri
s, rw. w I Ow I e, w*e..r l dM I e, t+~ie. .r I nM
Ie,$
e,4
6.0
17,5
7,0+
1e,e
I,e
e,00
17,e
1,06
ie,t
e's
6,06
Itj
1,0e
Ie,e
A,"
it,a
t,w
it,e
t,06
is,o
7,00
17,0
v,ue
18,0
7114
*A, thous. km
Uincus ,ion of tftie Fiegult5
inveutigttion nf the trgvel times and amplitude$ df body aaves, phase veloci- ~
ttes of surfnre Waves gnd free agcillations di the earth has given an idea of
the layered structure of the earth and pnabled congtruction of a radia].ly ~
inhomof;cnenus und spherically symmetric model. At the same time, marv
knorrri factg cnnnot be explnined Within the franeaork of g radially inhomo- geneoti:s madel and require expunaion of the class of models of the earth '
thnt gccnunt for bnth r$dial and lateral inhomogeneity of the atructure of the egrth.
Fxperimental investigation of the horizantal inhomogeneity of the earth
renuireg the develnpment df detgiled cross-aectionxl gnd aresl seismologic
sy,tems that in nccordance with the degree of detgil of the investigation ,
3hould bc oriented LoWard some characteristic scale of inhomogeneity of '
structure. 'fo do this, it is necessary to establish a hierarchy of natural
acules of inhomogeneities of the structure of the earth's interior and the
relation betWeen these scales gnd the scales of variations in parameters of
the wave field.
mo detect and study the scgle effects that determine the regional variability
of elements of the wave field, it is necessary to introduce a number of neW
morEiholoqical characteristics of the field and to develop relstively simple
nnd effectivr techniques for data anal}sis.
In t;hi- t'irst part of the monograph on Lhe basis of experimental flata about
the pnttern:: of propup,atfon of seismic Waves in an inhomngeneous earth it is
::t:oan hos the cnn,yor elements of the wave field are related to the nature of
dintr�il,tition and the scale of variation of velocities of propagation of
::ei::mi c r+rive:: and the absorbing properties of the medium.
'Mc mitin procedural technique used to analyze regional variability of the
rhnrnrteriz:tic:; ot' hod} and surface rraves is to sepgrate the observed wave
t'ir.1d:, into txi) mnin r.omponcnts: the average field (background) and the
devtiitions from thc aversge fluctuations. Simple and comparatively
38
FOR OFFICII,:. U5E UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FdK dFFICtAL U5E ONLY
et'f'ectivv t,r,rhniqucr, nrr rrofldned for dnmlyzinf; the gtntinticnl strur.ture of
t'tucturitionEi, Charncterintit, exramplen are given td ahnw the actual pnsgi-
bilitien bt' methodr, thut have been developed i'dr gtudying the gce.l.e effects
in manii`edtatibn of inhbmngeneitien of the ntrueture in the pattern of fluea
tuatinns nf r+uve fie1d patterna.
It io ohoun that to analyxe the gtructure oi' rluctutLtiang of thq Wave fie18
parameterg, guch atatistieal eharacterintien an functians of coherence,
Gorrelntion ahd struGture funGtidng hgve great pngsibil3tieg. A gummary
Curve rdr the cdrreltttidn rutid as u function of epicentrgl distgnce in
thr range of 10 km to 125000 km ig plotted from the dtLtn fnr P wgve pulse
ghapt rdrrelatidn fdund in diff'erent partg of the Wor1d. Analy$is of these
data ghawn that by uning nspatial Cdrrelation function in eggentitLlly
dii`�erent genlogiCU1-teCtnnic regions of' the earth dne Gan diatingutsh the
acalen of fluctuationa in pnrtuneterg of the wtkve fiel{d thgt are typical af
these regions, tLnd put them into correspnndence with the seal.eg of inhomo-
geneitien di the medium.
3pntial gtruCLure functinns sre egpecially cdnvenient for evaluating the
spatial gcalen of inhomogencities in gtructure from the nature of fluctu-
utions in wnve rield patternn. mhey Were cglculaLed for fluctuations in
the travel time of the p Wave for the LASA group gnd 20 USdS stationa located
in $%-,he continentul regiah, gnd also for n group of seismic atations in
Californin in the zone of tr$nsitinn from cnntinent to�oeean, and for a
system of geinmic gtatidna in other territnrips of the United Stgtes nnd
Cenada. Anpatial structare i'unetion of fluctuations in the logarithm of
the cLmplitude of a longitudinal w$ve is plotted from ststion data on the
territory of the United 5tates. The resulLgnt data are compared rtith the
structure function of thiekness of the earth's crust in the territory of the
LASA group und With the system of velocity pmfiles of the crust and upper
mantle of six regions of the United States and Canada. A direct relation
is estsblished betr+een the'structure function of wave field elements and
horizontal inhomop,eneities of the structure of the crust and mantle.
Astructure function of fluctuations in the times of arrival of the PeP Wave
in rlotted for a system of World-r+ide seismic stations fram data of regis-
trntion o1' thc LONG SHbT underground nuclear explosion. It is shoWn that
by us-Ing gtrurture t'unctions, horir.ontal inhomogeneities cen be traced to
grent depthr,, ineluding in the region of transition from the lower mantle
to the outer core.
It i:: also shown thut u:~e of the system of regionally differentiated curves
of I'n, 11g, P, Sn, S, Lg and R asves can appreCigbly extend the dynamic range
rind the ronp,e of epicentral distances in solving the problem of estimating
tiie magnitudes of seismic sources.
39
FOR OFFICII,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FdR dFFICtAL U5E hNLY
ll. ItE(S1UNA1, F'ECULIMI'I'IF.S nF SEISMtC WA1/ES ANU STtiUCmUIiE bF 7'NE UF'PER
MAN'CLE CN GEW'I'IZAL ASIA
7'hc c;econd pnrt of thc mandgruph present r, the renulta d� yegrs of renearch
donc in the centi�nl part of the AsiutiC Continent. A mttjer place is given
to Irivrzit,iNtion ot' t1it, titructure of thn upper mgntle in thh pamir-BtLyknl
rr(;Lon, where f;einmnlogIC profile dbserviitions Were mnde in 1961-1963 (Lukk,
Ner:;euov, 1965; Neraenov, hitutian, 1964). Tn gubnequent years, additiongl
r1utu huve been nbtained thut have enab1ed more fletuiled analygis of the
int,ernril strurture of the upper mantle Within the limitg of this region.
'Irtc ter�r-itory oi' Ccntrnl Asiu betaeen Pfunir gnd BtLykal includes a serfes of
tectoriie structures that comprige the eastern branch of the Ura1-Mongolian
geonyncl Lnul fdlded region: 'Cyan' -Shgn' , Knzakhstnn, Mongnlia, Altay and
('ri.bnyknl'ye.
Yc. N. Altukhov et a?. (1974) f'eel that these relatively "widely aged folded
�r.ones r.an be unified into three graups that have similarity of dpvelopment
und structure." In thi3 Cnnnection it is hypothetically gssumed that the
reE;ionn.l atructures r.un be sy3tematized uccording to the t,ypes of the earth's
cru:st thut mnde up the folded bgse of Archegn age for the "primary gnd
regcnerfsted gea3ynclines" th$t have developed since the Lower Proterozoic
Firid luter. F'ig. 17 gives u diggram of the ma,jnr structural elements of the
_ eu3tern pnrt of the Ural-Mongolian folded zone (Altukhov et al., 1974).
.
.

.
�
. . .
~
. . ~ ~
.
~
*
+
~
+
~
� . � . � . �

� � � ~ � ~ � �
+
t
+
G; ' . . .

. `
~
~
+
   ~
~
� � �


~

:
t: . � �

.
~
~
.r~~>, r;. ; , ~ - � ~
r. ~  ~ �
: { ` f . ~ i ~  t*~**~~*~*

   f 444 f    t f~1'f'~*1*~*1'f~1~4
 t f  ~         *        f 
  1 =J EMI Ml C~I ME  Ma1 200
F'ig. 17. Majar structural Plements cf the eastern part of thP Ural~Mangolian
x(,nf, ( A1 tukhov et al., 1974 1-� ancienf� platforms and massi fs; 2-4-r
f�oi(led ;;y; tr.ms LT1F1L Ilrlve devrloped on a crust of continental type (Kazakhatan��
'(ti�rin' :;lian'. Sayan-Ba,ykal. Khinqgn�-Gobi); 2--primary Keasynclines; 3, 4--
rc�t;rtwruted gceosyncli nes (3--early Caledonian, 4-.�Hercynian) ; 5. 6---folded
.:y::t,pm Ltiat has developed on a crust of transition type (Mongol�-Altay),
lwi msir;y lfro::ynr.lines; 5-� enrl,y Caledanian. F-�-Hercynten; 7-��folded system
ruit, Iin;: oicveloped on n crust of oceanic type (Zaysan Gobi, Hercynian);
:f--:.hcnth of the West Si herian Ylate; 9��- zones of fractures; a-along
t+aundnrie:: of folded s,ystem::; b--alonq houndaries of geos,ynclinal zones of
di f'ferr_nt, nt;es. 'I'he double line shoWS the Pamir-Baykal profile.
40
FOR OFFICIt.:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR OFFICtAL USE ONLY
It i:,, r.onnidr�rrd r.hnrnrt.rrivt,iv faature r,.f t.hp t'nlded LAyntemn thnt hplong
t,o Hir. I'Irrit grnuh. inrludlnw thr rli,yllfl=f1n,ykMl, Kny,khnr.rin-'Pynn' Shnn' and
KhInNun-Obbl uyrythmt, tliaL thh t'o1ded zanen of this grdup werp fdrmed on a
bguement with a continental type of cruet "with g thick granite-Cneins lnyer
of Archenn age." 7'he primary geonynclintil gysterns of Proterozoie�reefogenic
age comprise the franework of the Siberian and Chinesp platforms and mtiinly
repronent Atructures of the Sayan-gaykal folded gysteM. '1'he regenerated
gedsynGlines ghnwed up mainly Within the limits of the Kazalchstan-Tyan' Shan'
aygtem: the Kokctietav-North Tyan' Shan' gensyncline of the Caledonian Era,
and the zhunggr-ga.lkhanh and South 7'yan' Shan' geosynrlineg of the Nercynian.
Heldnping to the third type of folded systems th$t rrere formed an a basement
aith ocpanic type of crust (rrithout g siglic 1syer) in the Znyann-(3obi falde8
region ttiat "extendg fdr more than 2000 km in g continuous strip 150 200 km
Wide from Semipalatingk dn the West through the Zaysgn-Ulengur esuldran,
along the gouthwentern slapes of Mongolian Altay, acrogs Cobi A1ttLy to the
1oWlunds of the Eastcrn Gobi."
mtius Ye. N. Altukhnv et al. assume that during the evolution of the earth's
crust in thc Ural-Mongolian xone, the Archean granite-metamorphic layer of
crust from the LoWer proterozoic and later period Was converted by successive
geosynrlinal cycles to a new type of crust. The Bqyka1-5ayan and the
Kaxakhaten-'Pyan'Shan' folded gystems With continental type of crust were
formed in the procens of the primary geosynclinal cycle in the Proterozoic
utid the regenerative ayrle in the Caledonian and Nercynian epochs. A crust
of intermediate type was formed Within the limits of the Mongolian-Altqy
t'olded system in the LoWer-Upper Pgleozoic epoch. It is assumed that the
grouth of the crust took place from the edges of the system toWard the center.
In the Ordovician-Silurign, as a consequence of disintegration and "stretching
of the cru3t nearly from the axial part of the Ural-Mongo]ian belt the
Zaysun-Gobi system arose aith a crust of oceanic type."
'Phe given pattern of evoliation of the crust of the Ural-Mongolian belt is to
a certain extent hypothetical. The extent to Which such a treatment is
reulistic may deppnd directly on the possibility of explaining a number of
facts discovered in recent yesrs by geologyr, geophysics and geochemistry.
1t ,ee= thnt some of the assumptions ntated by geologists on the basis of
genornlixation of various data relative to the origin and tectonic structure
- of ttie ec:;f.ern part of the iJral�-Mongolian folded region (Altukhov et al.,
19711; Gayt3ev et al., 1974; 'Lonenshtayn, 1974, Shul'ts 1974) may be confirmed
to some extent by the results of seismologic studies presented in our
monograph.
C1ittPter 1. Spoetrnl Chnrncteristics of P, Pg and Lg Waves
Invc,tigation of the 3tructure of the upper mantle is based mainly on the
tr�vel Limea of body waves, the phase velocities of surface waves, and on
dutu ubout free oscillations of the earth. The use of dynamic characteristics
of seismic wavPS for these purposes opens up neW possibilities (Azbel' et al.,
1966; Yanovskeya et al., 1964). Of greatest promise is the consideration of
41
FOR OFFICIh;. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
k~;
FUR UFFICtAL U5E dNLY
rprctrbt timplitude r.urvcn (Vdiknv, Ygnavska,ya, 1974). Noaever, so fer this
quention hnn nnt t,een gtven ndequute ntitentiian. We knorr nf on1y an extremely
- mmnl] numbrr nC papern in Which antiefactory dntft uould ha,ve been found on
thc apeetrdl cunplitude curveg of different phmen of bddy and gurface waveg
(Mnlinnvnkayn, 1971; Arehambegu et n1,, 1969). 9o much the more deserving
of interecit are the data on amplitude curvpg of 8ifferent frequeneieg found
by the Cbmpretiertgive Seigmologica1 Expedition ag a restil.t of mare than a
decade of obnervationg uging the Ch255 frequpncy-selective seismiC stationg
(Zapol'okiy, 1971),
1, Amplitudp curves of differenr frequenrieg
The Matprials Procpsged. Recdrdings of earthqualteg by gseven-channel
ChI5S ntation located in Talgsr were uged fnr experimental invegtigation
of the genernt gtructure of neismic Waves in the range of epicentral dis-
tances frnm 500 to 3500 km, F'or the sake of brevity, this diatance range
is nometimeg called the intermediste zone. The avergge frequencies f of the
regigtrgtion channels of the station Were 0.35, 0.7, 1.4, 2.8, 5.6, 11 and
22 }iz. butn of gbout 800 earthquakeg Were processed in all. t
lnformation on egrthquakes (time at the focus, coordinates, depth, magnitude
and clasa) Were talcen from the seismir bulletins of the U53R and the United
States, and also from data of regional netWOrks (Baykal, Altay and North
'Iyan' Shnn'). Moat of the earthquakes in the xone up to 1000 km hgve a
value of K from rlass 9 to 11, and in the zone beyond 1000 km, the magnitude
M is i'rom 4 to 5.5.
'I'he entire uQgregate of epicentera of the earthquakes Was divided into four
directions. In accordance srith the location of the epicenters relative to
ttie malp,ar stetion these directions were termed: Northeast, East, South
nnd West, designated from here on for brevity by NE, E, S and W. 2'he epi-
centrel diat$nceg within each direction With varying detail covered the
range from 200-300 to 3000-3500 km. The distribution of the number of epi-
centers by directions is shoan in Table 6.
, TABLE 6
Data on the number of earthquakes as a
function of di^tance interval and direction
e, xm
~ rrE
I E
I s I
w
200-1000
59
153
60
1000-3000
108
127
130 132
'Phe Northeast direction is formed by the epfcenters of Ze~ysan, Altay, Sqyan,
F'ribaykul'ye and Trans-Baykal. The East by earthquakes of Dzhungaria,
NorthWest China, Mongolia. Comprising the South direction are the epicenters
42
FOR OFFICII,:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOtt dFFICIAL USE dNLY
oi' Kunhp,nria, C'akintnn, Nepa1, tndin and the 8gyr nf Bengu].. In the Went
direction are Snuth 'I'yan' Shan', pamir, the mgdzhik Deprension, Kopet-nag,
tr+.n and the CEiur.unus.
Meggurpmpnt and procegsing Methods. In each di the four directionn, gpectra1
rsmplitude Curveg were plotted for the mujor Wave groupg reliably fliffer-
entiated in the intermpdiate zone: pn, p, pg, 5and Lg. 'I'he hndograph
canpiled by T. L. Nergegov and T. (S. Pgutiun (1964) rrag uged in flgtn analygis,
Msximum amplitudeg were menaurpfl in intervalg lagting 10-15 s fnr longi-
tudinal anveg, and 15-20 n for S and Lg waveg.
Uue td atrong abnorption, high frequencieg could ttot be followed throughout
the distxnce rgnge. For instnnce the recdrdings dn the geventh channel
(i'= 22 Hx) were mgde at digtnnceg up to 300 km, nn the sixth (f = 11 Hz)
up to 500-700 km, on the fi.fth (f =5�6 Nz) up to $00-1000 km, 'I'he ampli-
tude Curves for the fdur lnw-frequency chunnela aould be folloWed throughout
the intervel of investigated digtgnces,
'I'he single-statinn method was used in data procegsing. mhe emplitude curves
rrere plotted fram recordings ni' many earthquakes with epicentral distances
coverinq the range of interest. Inten3ity estimatea were knorn for ell these
enrthqunkes either the energy class K(st digtances up to 1000 km) from
observations of regional netKOrks, or the magnitude M determined from surface
waves (at distances beyond 1000 km). It is important that these estimates
do not depend on the behavior of the body waves in the intermediate zone.
If we have an estimate of earthquake intensity Mi or Ki and We know how the
spectral amplitudes A(f) depend on M or K(Nurmagambetov et al., 1975), We
can normalize the measured amplitudes, reducing them to a fixed value of
Mp or Kp. It was such normalized amplitudes that were used to plot the
umplitude curves.
All measured spectral amplftudes were reduced to the reference class Kp = 10
or to the reference magnitude Mp = 5. Normalization Was done by the formulas
AU, Q - A U. K& fUgK v I"
ar
A .4f.) -A (1, 4)� iam M VY.w,,,
Values of functions R(f) for average frequencies of a ChISS station obtained
experimentally (Nurmagambetov et al., 1976) are summarized in Tab1e 7.
I?ie function a(f) aas taken as the same for different waves and regions.
Ttie vnlues of RK and RM are related by the formula
aX - o.w pM.
implied by the energyr relation between the M and K scales that is valid
Within the limits of the dynamic range of the earthquakes used (Zapol'skiy
et nl., 1974).
43
FOR OFFICIkL llSE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fdtt dFF'ICIAL U5E tlNLY
'I'AHI~C 7
Values oi' p(P) for channels df a ChI55 station
4u:~ 6
, O,ls 0,10 I140 i,AO b,AO 11t00
aK o,M o,U 0.50 o,M 0.50 0.46
a,N t,ao o,a 0101, o,w 0,7e 0,72
KEY : 1--Parnmeter
2-- f, Hz
Si nce the amplituden were normalized to Kp = ld in the near xone (up to 1000
km) and td Mp =5 in the far zone, in compnring the amplitude curves gwnia-
rizing the dgta for the entire distance interval (from 200 td 3000 km), the
curves oP the near zone were shifed by an amount proportional to BK until
they coincided with the curve of the far zone on the section of coverage.
Since 6 depends on frequency, this ghift was different for different recording
chunnels.
'I'tius in plotting the spectral amplitude curves we used normalized values of
the maximum amplituden of wave groupg Pn, P, Pg, Sn, S and Lg.
Craphs for the dependence of maximum amplitudes on distance Were plotted
from six frequency channels for each wave group. The experimental values
(from 60 to 250 points) were avereged, and a smoothed curve was drawn through
the centers of gravity of the individual distance intervals. The method is
described in detail in another work (Antonova et al., 1968). As a result,
familieu of spectral amplitude curves were obtained for each direction by
wave group3.
Shown on graphs plotted from recordings of the ChISS station at Talgar are
amplitude curves For approximately the seme azimuths according to data of
aide-band instruments of the general type SK for six stations located in
Northern Tyan' Shan'. Graphs for the ratios of amplitudes of Pg and Pn Waves
ss u function of epicentral distance are plotted for each direction.
Let u3 go on to a detuiled exemfnation of the amplitude curves.
Spectral amplitude curves of Pn and P waves for the four directions are shoWn
in Fil;. 18. Characteristic peculiarities of all curves are: uniform drop
in the interval from 250 to 350-400 km, increase in intensity at about
350-400 km, and a second spike at 500-600 km that shows up more reliably in
all directions, und then a section of strong decline from 600 to 800-1000 lan.
'Me structure of the amplitude curve of the Pn Wave on the section up to
1000 km is the most unstable in comparison with all the Wave groups con-
:;idered. The amplitude curves are characterized by strong oscillations that
44
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR nFFICtAL U5E nNLY
41
~qa
A/T, um/u
R4
A/m,um/g
l 1 ! 1 I 1 1 1 1 1 I 1
~0~rog rN aoo 700 laoon rooe d~Aae� oo ni nr iooa rari rAr
FtE;. 18. Spectrnl umplitude curves of P and Pn Waves for Northeast (a),
Eust (b), South (c) and West (d) directions: 1--average channel frequency
0.35 }Iz; 2--0.7; 3--1.4; 4--2.8; 5--5.6; 6--11 Hz; 7--from recordings of
Wide-bund ,^,K equipment
nrc di fficiiiL ta distinguin}i With the methods used for the observations and
(intn proce::.ing. A:: nn cxanple, we pive a version of possible approximation
ot' experimentrsl dnta l'or the spectral cvnplitudes of Pn waves for the South
tlirectlon (F'Ig. 19).
-
-

.
~n
1
I
-

J
~
-
~
i
~
r
I
b jf
\
~
;
~
~


- -

.

'

~
~
,
;
v
~
\
~
45
FOR OFFICIk:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOk nFFICIAL USE dNLY
A,NM
Fii;. lq. Att exrunple of possible approximation of experimental values of
the srectral amplitude:; of a Pn wave on the 300-1000 km section for each
directinn: channelr are separated; see F'ig. 18 for designation of curves
Unc: can clegrly distinguish three sharp spikes of intensity at 400, 600 and
800-900 km, wtiere the funplitude differentials reach one and n half orders of
magnitude. An example of another technique for averaging experimental data
is shown by the curves for the West direction (see Fig. 18d) that are plotted
with strong smoothing, resulting in revelation of the most "long-period"
romponPnt of the attenuation curve and total filtering of the "short-period" -
(oscillating) component. Ttie given two versions of averaging show the inhomogeneity of structure of
oY the spectral amplitude curves in the near zone.
Let us continue ttie examination of curves for all directions (Fig. 18). The
frequency-selective nature of attenuation on the section up to 800 km shows
up weakly. The curves of the first four channels (frequencies from 0.35 to
' 3 E{z) are nearly parallel, the attenuation curve of the fifth channel (5�5 Hz)
}iitis u somewhat greater slope, and it is only the curve of the sixth channel
(]1 }iz) ttiat is appreciably steeper. The most pronounced differences in the
behnvlor of ,pectral curves of different frequencies are observed on the
8)0-1000 km section. At a distance of 600-800 km the amplitudes on the
Ci ft.li nnd r.ixth channels (5-10 Hz) are approximately equal to the amplitudes
cxi the fi rst and .econd r.hannels (0.35-0.7 Hz) ; at distances of 1300-1500 km
tlicy becomf- 20-30 times lower.
t'r�am ii comparison of curves of different directions on the section up to
- ]UUO km it is clear that as tLere is a chaxige from NE to E, and then toward
, W und S there is a reductian in the slope of the curves and displacement of
� ttie position of extrema toward greater distances.
46
FOR OFFICIti:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
Om, um/n

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR UrFICIAL U5E ONLY
On curves for directinnv NE and E near 1000 km, oscillation ia orserved with
a minimum at 800-850 km and a maximum at 950-1050 km. This .feature shows
up vFry wealcly on curves Por directions S and W.
At cllstances greater than 1000-1500 km the general attenuation of amplitudes
decreases sharply, ttie curves flatten out, reaching a minimum at distances
nf about 1500 km, which is clearly distingu3shed on curves of directions
NE, E ancl S. Apparently for the West direction the minimum is shifted toward
2000-2200 km.
Further on a maximum is observed that is most pronounced on curves of direc-
tions NE and W, where its "amplitude" reaches 0.7-0.9 log unit. For directions
E and S, the "amplitude" of the spike is lower: 0.3-0.4 log unit. The
position of the maximum of the amplitude curve gradually shifts from 1800 km
to 2400 km with a change from directions NE and E to S and W. This maximum
has a complex structure and can be resolved into two maxima located at
distances of about 2000 and 2500 km.
Unfortunately, it has been nearly impossible to determine the spectral
structure of this spike, i. e. to trace stab].e differences in the behavior
of different frequencies. The curves of frequencies from 0.35 to 2.8 Hz
betiave about the same, and the higher frequencies disappear at distances
greater than 1500 km.
For a quantitative description of regional differences we introduce several
parameters that characterize the ma,jor features of'the amplitude curves:
Zimin. or Zimax. the position (in km) of the extrema of the curves;
rli the logarithmic steepness of the slope (negative exponent of the
approximating power function) within the limits of a certain distance
interval; di the "amplitude" of the maximum of the curve (in log units);
and the characteristics of level the value of the ratio A/T (in um/s) at
a certain distance. We will determine the values of these parameters for
three frequency intervals: 0.35-0.7, 1.4-2.8 and 5.6-11 Ha, that we will
designate for brevity by the values of the average frequencies 0.5, 2 and
8 Hz. A11 values of the parameters for ttie amplitude curves of P and Pn waves
as well as Pg waves are summarized in Table 8. The unreliable determinations
of paremeters are enclosed in parentheses.
The amplitude curves of Pg waves are shown in Fig. 20. These curves also
show a characteristic oscillatory structure; however it shows up much more
weakly than in Pn waves. Two sections of monotonic fall-off in ampli:tude
are observed: the first from 250 to 350-500 km, and the second from 500-600
to 800-1000 km. These two sections are separated by an oscillation that
shows up clearly on curves of directions NE and E, and to a much lesser extent
on curves for S and W.
On curves for directions NE, E and S one can distinguish a minimum located
at distances of 3401 420 and 450 km respectively, and a maximum at dis-
tances of 400, 500 and 520 km. The average values of logarithmic steepness
47
FOR OFFICIIsL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOK UP'FICIAG USL ONUY
Viilueis of' prirametcrn nC npcctriA nmplitude curven af Pn, i' unven nnd Pg waven
t'oc� t'our direr.tlan;> vt' rrdpr.tf;sLtion atcl t}iree e.verage f'reyuenries: 0,5, 2, g 11z
n N
~
A
cr =NH~
s=L
'~S

r
�
.
$,1
� s
1 1
4,6 1 - I -
6,6
t

1
i'n
uncl E'
Wnvrg
~h
aso-aoo

4,5
a
1,7
2.2
3,5
(a)
3,0
i,s (o,e)
1,e
2.0
li rrM.` ?
300-450
-
Sb
NO
/SO
490
400
420
420
- -
_

r, 3
400-e00
-
4e0
eso
un
aio
470
4e0
- -
-
m
aoo-eoo
4,0
e,s
8.0
4,e
5.0
e,s
3,5
3,0
n,e
1.6
2.0
Air
eoo
zo
23
to
33
eo
is
70
00
- Iso
lw
s
Al1'
2
IS00
IJ
S
1
35
11
3
70
10
- IW
SO
~
It YNN.
1~1~
1 W
I~
-
1~
1~
-
1~

- o
r

it 3
17oo-woo
iuoo
1e70
-
1eso
1700

aaou
ioso
- -
-
-
a1 w,p
i
17uo_2000
0,7
o,s
-
0.3
d,i
-
o,~
0133
~
t, ws~e
2400-1700
23w
2100
-
1500
1700
-
3700
ZtlUO
1W
14SU
-
e, wa. 3
2400-2700
0,25
0.15
-
0,13
0.20
-
0,33
o,b
0,40
0153
-
Pg wnve
q,
7S0-3W ~
3,5-5
5
A
(1)
3,7
3,7
3-,1
1,0
6,0 13,3f
1,
~
3,0
rye
So0--800
7,6
0
10
5.3
4,5
S
(4,2)
8
- 3,3
7,6
(3,0)
KF;Y: 1--ParomeLer; 2--mtn;
3--mfut
A/m,
A
c
e
~V
\
.
`

i
ia
�I
1 -i
g)
l I I I I I I` I
4 JOO J00 700 /040 JO/ 7I01 /M
J// ADY JAI 7N /A'/ J/0 J/9 700 /00/ d�"K
d,Nh
Fig. 20. Spectral amplitude curves of Pg waves of the Northeast (a), East
(b), South (c) and West (d) directions; see Fig. 18 for designation of curves
48
FOR OFFICItiL USE ONLY
d

~
. ~

,
~
.
\
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOtt UFFtCtAL U5E hNLY
ti1 nn(l ry2. 01' Lhrne tuo sircLioun or t1ic, nmplltuclr, r.urvrr, are itivr.n im mable t3.
'I1~~~ ('a Wuvrit dtuni, (jut v,ery i~t,ri~tir;ty ~riN~E~roxtmcit~!ly utt A'3-A�~' wit}tin the
.ltmitti or the rlrgt srction, nnd ntlll mdre strdngly within the 13tnits ai'
thc aecond nection),'1'he extent; or attenuation wenkenn uith n transitidn frdm
dtrer,t iori Nr Ln E and then Crtim ro' ta W. 'I'he speetrally gelective nature or
rittcnuution in uenlcly exrronnod, nn thut p,enerul dnmping Canndt be nttributed
to the ef Cect of ubsnrptibh, A compurtndn or the Pg gnd pn uaves rrith
respect to level shows un thut nt a di titanr.e of 300 km they differ by a �
t'sctor or 5-15 on f'requencied nf abnut 1 Hz, find by a factnr ot' 3-4 0n
Crequcnrirs nP 3-6 NZ,
From a cnmpgrinon of the umplitude turven of Pg and pn Waves Nith regpect
to nhape, ue gee thnt the exponent of the apprdximeting pnWer functidtt i'dr
IlN Wuven in nbout two unito highei� thun fbr I'n Waven. purtieulgrly deserving
or nttentian tn the CninCidence or ponitiona (wii;h regpect to the digtarce
,txl., ) rff the extremu dC thr wmplitude curveg of beth typeg of Wgveg. For
innt,nr.e their minfma cnincide rsnd eome nt 34n-420 km for the direetions NE
and E reapeCtively. Similarly coineident nre the positiong nf the nert
maximum: nt 400-450 gnd 500-520 km for the given direetians. F'vr con-
venience in cnmpciring, thr npectral emplitudeg nf Pg and pn r+gves, graphs werp
r � a c
J f f I~ _
J~ .
~rii aov ,ra M na

i , b
f t i
I /
/
1a A. Nh
d
Fig. 21. Amplitude ratio of Pg and Pn waves as a function of epicentral
distunce for the Northeast (a), East (b), South (c) and West (d) directions;
ttie curve numbers, correspond to the conventional designations in Fig. 18a
l+9
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Ftlk 0MC1 AL 1181: t1NI,Y
p.lor,ted Cor thc ch,nge aith dir,tnnCe in the ratio of theit' amplitudes w > 1 J J 7 ia m
11
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
A/P91~i9/Pn/
20

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOk ON'FiC1:Al, USl: ()NLY
d i nt1Lnr_V (300, 500 untl 800 krn) uncl Cour di rections nf nrnpfi_p,ntidn. 'I'tte
vrLluF;a ot' the ratios were culculated Crom direct, tneasuremcritg of the
amplitudes of i�,hene wuven rnther than with renpect to Qmplitude curveg or
C,ec:~rrs.
I,e1, iiu con:;ider Lhe (;raph i'or 300 km in more detuil (nee Fig. 28b). At thin
cli:;t,unce one begins to track the f'g wave. Crom recordings di' the Ta.lgnr
ctatinn thi:ti wrive in distingultlhed f'rom distnnces nf 250-280 km. gy 300 km
the k'N wnve is usunll.y dcl.ayed reluttve to the first entiry by no more than
it cun be ur,sumed thnt at thi3 distnnce the conditiong of propap,ation
hrLve not ,yet had any nppreciable influence on the npectrum of the Pg wave.
Uetui led studies of the apectra nY loCal eurthqucilces undertaken by us
prevlously on muterials of the 'I'algar ChIS5 station showed that the spectrum
nt' the 1'n wave inherlts the spectral pectLlio.rfties nf the direct p wave
propagrsting in the eurth's crust. This gives us a bgsis for assuming that
the dtf�erence spectrn shown in F'ig. 30a can be interpreted as the frequency
response of the mechanism of fbrmn,tion of the Pg wave.
'I'tic rimi>litude diffcrence^ of this response are quite considerable. For
in:,t,rince the p,reutest relative intensity of the Pg wave is noted for earth-
qurikes o� the Northea3t and West directions Por which the ratio of amplitudes
of Pg to k'n is equal to 10-15 on a frequency lower thrin 1 Hz. The ratio of
these wuves begins to decrease from a frequency of 1.4 Hz in the NE direction,
1111cl fz'om 0.7 fiz in the West, and is inversely proportiongl to frequency with
cxponcnt 0.7-0.8.
For di rect,ion:, S nnd 1s the frequency dependence is much less pronounced.
In thc frequency runge of 0.3-3 liz the wave ratio is approximately constant
rsnci ecjual to 4-6. For higher frequencies the ratio decreases weakly: in
i nverse pi�oportion to frequency to a power of 0.3-0.4.
Witti increasing distance the azimuthal differences damp out, and at 800 km
(Fig. 30c) they become insignificant.
'I'hE spectra of S waves are shown on Fig. 25 for three directions of propa-
F;at,tor. Of all the investigated types of waves, the data on transverse
waves are the least re].iable and representative. Therefore let us deal with
them briefly.
.Jivot tt; for longitudinal waves, the hfghest-frequency spectra are those of
the E;ust direct:ion. The spectra of the West and South directions are
appreciably lower-frequency ones. F'or directions E, S and W the ratio of
runplitudes of ttie frequencies 0.7 and 2.8 Hz at A distance of 1000 km is
0.22, 0.70 and 0.68 log unit respectively. By a distance of 1500-2000 km
the difference: in apectra of different directions abate.
FoT� t.1ie npectra of transverse waves of the East direction the effect of the
iricrr.use in t,mplitude on a frequency of 0.35 Hz fs obse.rved to a still
greater extent than for longitudinal waves.
62
FOR OFFICII,~ L'SE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
rOk nNFrr,rnt, usE' nNt,Y
'I1ie uprctru ot' U wrxveu nhow n more ntable chungc with dlntanee than longi-
tudinnl wuves. An uppreri.able change in the gpectra talces place in a more
extended intervEil.: f'rom 800-1000 to 2000-2500 km. The prinCipal ogeil-
14tibtl:3 are conCined to 2000 and 2500 km.
'Che gpectrn of Lg wavee an a funCtion of distance are shnwn in Fig. 26 and 27
for the 'I'alp,ar and Temporary qtations,
7'he umplitude differences of the spectru of Lg wgves that show up at ghort
distuncer3 repeat the peculiarities of the spectra of longitudinal waves.
' 'I'he spectra of the NE direction are the highest-frequency. With an increese
- in uzimuth one notes a congiderable increass,e in low-frequency components of
the npectrwn. For inntuncc: 'Puble 13 shows that the maximum differences in
the rutin of umPlitudes- or rrequencien of 0.7 and 2.8 Hz rench 0.7-0.8 ing u,
'1'he maximum oC the apectrum oF directinns NE and E is at a frequency of
1.4 lfz, and fbr the West direct3on at f'requettcies lower thgn 0.35 Hz.
'i'hcse ertimaten apply to epicentrnl distunces of 350 1vr.
1lnomalou:,ly high amplitudes of the 0.35 Hz �requency can be seen on the
spectra at ohort distances for all directions except the South. -
Now let un exumine the change with distance in the spectrum of Lg waves. ~

I'ig. 29 shows n grnph of the distance dependence of spectral ratios as
plotted from data of the 'Pemporary ChIS5 station. The plot of the graph
repcats the main peculiarities noted previou3ly for other types af'waves. =
In the interval from 200-250 to 500-700 km the spectrum of Lg waves ctiangea -
more weakly than at greater distances. The strongest changes t ake place =
on thc section between 600-800 and 1300-1400 km. At distances of 1300-1'(00 km -
the oscillation showy up cleo,rly, after which the regular change in spectral
shape with di stance abates noticeably.
In conclusion let us repeat the main peculiaritiea of spectra of different
waves and their chnnge with distance.
In cornparing the spectra of different azimuths of propagation for all waves,
one trend shows up: the sPectra become lower-frequency as the azimuth
incrcn::es, i. e. with u trrsnsition from the NF, direction to S. For Pg and
Lg wnves these differences show up even on the initial tracking stage, and
con::equently they are due to u grenter extent to the mechanism of formation
Lhun to the conditions of propapation.
The patterns of chnnge in spectrrs with distance are most graphically destribed
hy curves for the dependence of spectral ratios on distance. These graphs
s}iow several di::tance intervals within which the changes in spectral shape
of all Kaves are approximutely uniform. In the initial part (from 200-300
to 500-700 km) sPectral changes are insignificant. In the next distance
interval (from 500-800 to 1300-1700 km) the strongest change in spectra
occurs with principal damping on higher frequencies. Between 1300 and 2000 km
63
FOR OFFICI!L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
>
FOk UFA'tC[AL U51. UNLY
~
~ir~ tn~c~rvrll of t,110 oPpo~ite chtinge in spectra can be distinguished, i. C.
r'
:t.rungcr drsmpinK of' lowor t'requencies. I'ttin shdwu up ag oncillatinri nn
gruplm of the dtutunce df-rendence of npectral ratias. I'he next osril.lation, '
uith iL lowcr "runp]itiide," ir, noted ut distnnces of 2000-2500 km.
Af, d:i:~tnnce3 greater Lhnn 1500-2000 ktti there ig practicglly no systematic
compo�rit in the ctiange with distance in the spectra of longitudinal wave:;
coi- frequencics lnwer than 3 Hz. At thene distances the spectral ghape of
I-1 waves on the averagc is the seme a3 in the teleseismic zone. The spectra
oP S waves tnd especiully Lg waves continue to change, but noticeably nore
wcakly than at ohorter distances.
At distances up to 800-1000 km the amplitude3 on a frequency of 0.35 Hz are
unomalousl.,y high fbr the spectra of al.l types of waves (especially S and Lg
waves). '1'his ef'f'ect is characteristic to g grester extent of recordings in
the NL and E directions.
' T1ie spectra of P and Lg waves at Temporary station are appreciably higher-
fre(luency spectra than on Talgar station. However, the general patterns of
change in the spectra with distance and the distance intervals with charac-
teri.stic chariges of spectra for both stations are the same.
Cliaptrr 2. t)ifferentiation of Large Hnrizontal Inhomogeneities with Respect
to the Criaracteristics of P, Lg and Rg Waves
Iri t}ii:: ctiapter an investigation is made of the spatial structure of the
a�r,imuthal differences of emplitude curves and the spectra of seismic waves
described in the preceding chapter. It is shown that these differences are
farmed chiefly on certain sections of the paths of propagation of seismic
rays when they cross certain geologicsl boundaries or structures.
The principal form of representation of experimental data is by spatial
constructions: comparison of amplitude and spectral characteristics on
different paths, mapping of these characteristics, localization of sections
of abrupt changes in the kinematic and ci,ynamic parameters of seismic waves.
Ic, section 1 an investigation is made of the spatial distribution of the
spectral characteristfes ef longitudinal waves from remote earthquakes.
tt i;; shown that regional differences in the value of this parameter are
stati::tically significant. A description is given of the spectral charac-
toristics of P waves as distributed for earthquakes of the ma,jor seismically active zones of the earth, and in more detail for Central Asia. The region
af the most minimal values of spectral parameters is confined to the Tibetan
rnassif, and a region of somewhat higher values to the Persian Plateau.
`Ifiey are interpreted as regions of elevated absorption of seismic waves in
the upper mentle. _
Lg and Rg interference waves are registered only in the case where the entire
path of the seismic ray traverses a crust of continental type. Consequently
d
64
FOR OFFICI6+:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FdEt OFFICIAL USC ONLY
by compuring the intep3ity of these waves on different propagation routeg
one can loculite tfie blocks in which there is no "grunite" lqyer, or other
appreciable r.hnnNea rire noted in the sCructure of the earth's crust.
Iri seCtions 2 und 3, ari invesL�igution i:; mndE of the peculigrities of propa-
gation of these waven on different paths that cross Central Asia and the
ndjoining territorien. Frnm observations of the system of stations, the
_ bounduries are locnlized at which a change in the intensity of Lg wavea
occurs. These waves disgppear completely in those cases where the paths of
the seismic rays even partly intersect the Tibetan massif. On the boundaries
of ttiis massif, sharp che.nges are also noted in the parameters of s train of
Rg surface waves.
1. Mapping of spectral characteristics of P waves
In thi:; section an enaly3is is made of the results of mapping of the spectral
ctlaracteristics of individual earthquakes done with respect to observations
in the intermediate and teleseismic zones. The regional differences f'ound
in the spectral characteristics of P waves are interpreted as a manifestation
of horizontal inhomogeneities of the absorbing properties of the upper mantle
in the vicinity of the focus.
5pectral CharacterisCics of Remote Earthquakea and Absorbing ProperCies of
the Upper Mantle. The experimenta.l data presented below and the results
obtained by other authors (Tsu,jiura, 1969) show the stability of regfonal
differences in the spectral composition of longitudinal waves according to
observations at remnte stations. The spntial degree of ordering of the
spectral differences can be attributed either to the particulars of focal
radiation or to horizontal inhomogexeities of absorbing properties of the
medium along the paths of the seismic rays.
We assume that the particulars of the focus, which are definitely important
in formation of the spectral composition of an individual earthquake, may not
be predominant in a set of earthquakes over an extensive terr3.tory. Experi-
ence in stucLying the spectral particulars of focal radiation of local earth
tremors of a number of regions (Molnar et al., 1976; Hanks, Wyss, 1972:
`IfiatcYier, Hanks, 1973; Tucker, Brune, 1973) has shown that within the limits
of euch region there is a considerable variety of spectral characteristics,
primarily of the angular frequencies, that is due to the difference in the
mechanisms of foci and to variations in the stress field. Spatial ordering
of sE,ectrnl parameters is noted only for quite small epicentral zones.
Ii'or extecisive global seismically active regions, no syGtematic deviations of
tiic spectral composition are noted in observations at small distances.
`Ptierefore we assume that the spectral differences observed in the teleseismic
xone nre formed mainly on the path of the seismic ray.
it, has been estnb].ished that the greatest distortions are introduced into the
nal :;pec l.rum when a seismic ray crosses the low-R layer associated with the
65
FOR OFFICIl,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
i:ott OFFtrrnt, usi. oKLv
aul;hcnuci phcre. Miln,y rv,,crlrr.hor:; Ivavr, deLer.tud con:siderable rep,ional dif-
t'ererice:, in thc absorbing prnpcrtirs n� tMa luyer. ApptLrently it is the
mauaic nature of the abaorbing properties of thE upper mantl.e thQt determineg
t.he grent varieLy of steepnens of the high-f'requency nlnpe of the spectru -
of remoi.e earthqutLkes,
I'he lic-erence of the u3thenospheric layer has an effeet both in ttie region of
the epicenter and in the vicinity of the station. For a tectonically homn-
geneou3 region, the influence of the anthenospheric layer in the vicinity o; t;he extt of t}ie ruys ct.n be taken as identical in registratinn of earth- ;
(i�r-tke;; at usingle atution from different azimuths. Use of a system of ob-
sec�vutidn3 from qeverul stations almost completely eliminates this influence.
'I'hus we u:jsume that the observed differences in the high-frequency sl.ope of
the spectra of remote earthquake3 reflect mainly horizontal inhomogeneity
of the absorbing properties of the upper mantle in the vicinity of the focus
(witti consideration of drift of the seismic ray). Spatial mapping of the
E,eculiurities of the high-frequeney part of the spectrum (with reference of
these peculiarities to the epicenter) will reveal the regional differences
_ of' nbsorbing properties of the upper mantle in seismieally aetive regions.
'I'he Maeerials and Technique Used. About a thousand recordings of earthquakes
of i;he territory of Central Asia and ad,jacent territories, recorded at ChI55
stations (Garm, Talgar nnd Temporary) were processed. Besides, about 700
c�ccordings of earthquakes from different epicentral regions of the earth
wei�e processed at these same stations. The magnitudes of the earthquakes
were from 4.5 to 5�5.
'llie opectral rutio of cunplitudes for channels with average frequencfes of
2�5 and 0,7 Hz or 2.5 and 0.35 Hz was determined from the seismograms. The
vr.Llue of this parameter calculated from each recording was mapped, i. e. it
was assigned to the epicenter of an eaxthquake. Maps were then constructed
for un intermediate smoothing stage in which the average values of the
parameter were calculated for a group of 5-10 closely spaced earthquakes.
On ttie next smoothing stage the values of the parameters (the averages for
large regions) were calculated, or maps of isolines of the parameter Y were
constructed.
'Loning of the Earth by Spectral Peculiarities of Remote Earthquakes. On the
f'irst strige to determine the stability and scale of regional differences of
1.1ie ^pectral composition of P waves an examination was made of the materials
oi' tlic teleseismic zone, where the conditions of observations are more
f'avoruble for detecting the investigated effect than in the intermediate zone.
- As is known (Antonova et al., 1968; Carpenter et al., 1967) in the teleseismic
�r.one the distance factor can be disregarded, the seismic signal is simple in
_ shupc, rsrid consequently the measurements of the maximum amplitudes are more
uriartibi Ei"uous .
66
FOR OFFICII01. USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
1
FUk UN F[C t AI. U51; f1NLY
7?ic cxprrimental rlritit nn remoLe enrthqunkey thet were uspd were obtained at
; Temporary atatian, Materi.uls of the (7arm and 'i'algar atations were utilixed
nnly in part,
Mdre than 700 recordingg of remote earthqualces from e11 ma,jor seismically
active znnes of the earth were procesged. The 1ogarithm of the amplituae
ratio on frequencieg nf 2,5 gnd 0.5 tiz wag taken as the characteristic of
, the apectrum.
i
! Z'he average valuea of thi3 parwneter fbr thirty epicentral territories are
~ ohown on the map in i'ig. 31. It can be seen that the spectrum does not
~ flepend on epicentral distance. Eaxthquakes in relatively near regions
; 'ribet, Iran, the Tyan' Shan'-Pamir chain are enmparatively 1ow-frequency.
� 6 a . ir
� ~ .

�o.8
.o.!
i
V
�0.~ �o.
~ �o,e 30
~o.
. ~
o.
~ - - e
�o.~
�a.~ � .
N; �u.~ 1�

�
M
IM M M ~ M M Ii~ IM Ip If~
Fig . 31., Map of the spectral ratio Y(2.5 Nz/0.5 Hz) for the major seismically _
uctive zones of the earth (from data of the Temporary ChISS station)
In examining the map, we can see certain patterns in the spatiel distribution
of values of the paxameter. All seismically active zones of the earth were
classified into four groups differing in the parameter Y by 0.20-0.25 (i. e.
by 50-80%).
The first, highest-frequency group, included earthquakes of the Aleutians,
the Kuril-Krunchatka arc and Japan. The average value of the parameter Y for
the3e territories is 0.16 log unit.
67
FOR OFFICLAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOEt nFFICIAL liSL ONLY
'I'te earthqunlcea of Soviet Midd1e Agia and the part df Centrnl Asig situatefl
t;o tihc north of' 'I'ibet ure ulno high-frequency. Nere the uverage value of the
parcuneter in 0.20. t3ut these regione are lncated rather close to the
rer.ording qtutioti (dtstance 1.000-2000 km), which may explain the high-
3'recluency nature o f the eurthqutilces.
"hie earthquakes of tndonesia und the I'hilippines, for whicti the gverage
pnrtmeter is 0.33, make up the next group.
Sttll lower-frequency are the recordings of earthqualces of the Alpine belt
ttiu Mediterrnnean, 'Curkey, the C$ucasus and Tran where the paremeter
uverages 0.55, Abdut ttie same values oP the parameter are typical of earth-
quuke3 of South America and the continuation of the Yndonesian belt to the
eaet New Guinea, the Solomon Isleaids, the Fiji territory,
'.I9ir 1oaest-frequency group is comprised by earthquakes of mibet (Y =-0.70),
Nortti and Central America ~(Y =-0.'j2), South and East Africa (Y =-0.82) .
`I'tic clifference:; of the paranieter between ad,jacent groups are significant
sitice there were from 20 to 100 earthquakes in each regional sample, the
avei�age value of ttie standard deviation of an individual measurement within
a 3smp].e being 0.25 log unit.
Alialogous zoning, but witti respect to a smaller number of epicentral regiotis,
was also done from recordings of remote earthquakes at Garm and Talgar
:;tftitions. Joint examination of the materials of the three stations showed
a 3y3tematic influence, comon to all zones, introduced into the spectrum
by station conditions. In comparisan with the Garm station, the spectra of
ttie Temporary station were on the average more high-frequency, while those
of Talgar station were more low-frequency.
The relative spectral differences of the epicentral regions were retained
with respect to the recordings of the three stations as well. For instance
the tiIghest-frequency 3pectra were those of Northern Japan, the Aleutians and
Indonesia. The spectra of the Mediterranean, South America and New Guinea
occupy an intermediate position. The spectra of North America, Tibet, the
Arabinn Peninsula and South Africa are definitely low-frequency.
Ot' course one should not overestimate the generality of these data. The
:,tations that were used are situated comparatively close together (average
distance about 1000 km). It is possible that observations on completely
different azimuths of propagation of reys, especially for seismically active
zones, that extend along oceanic archipelagoes, will show other tendencies.
Zfie spatial structure of the spectral parameter was studied in more detail
for the northern and northwestern part of the Pacific are of seismicity.
'I'hc results of mapping of this region with respect to the parameter Y on the
intermediate averaging stage are shown in Fig. 32. Here the average value of
the parameter is shown for each group of epicenters made up of 3-6 earthquakes.
. 68
FOR OFFICIi,L USE UNLY
~

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
P'Uk UF!'CCiAL U;;P; nNLY
Fig. 32. Map of values of the npectral ratio Y(2.5 Hz/0.9 Hz) for the
northern and northwestern partg of the Pacific Ocean seismically active
belt (from data of the Temporary Ch255 station)
KEY: 1--Sea of Japan 3--Pacific Ocean
2--Sea of Okhotsk 4--Aleutian Islands
On this map, one can clearly see the pattern of spatial distribution of the
parameter Y. For instance for the Aleutian arc as one advances from east
to west, the vulues of the parameter first increase from -0.7 to +0.15, i, e.
the spectra become higher-frequency, and then they become lower-frequency
once more, the parameter decreasing to -0.25. -
For t!ie Kuril-Kamchatka arc one notes a tencenc,y for the parameter to in-
c:rease with movement from the northeast to the southwest from -0.4 to +0.25�
For Northern and Central Japan the patterns show up more weakly. Nevertheless
two tendenaies can be noted: concentration of high-frequency foci near the
island of Hokkaido and intensifi.cation of the high-frequency character of
earthquake spectra in the dj.rection perpendicular to the axis of the trend
of seismicity from a deep-water trench toward the continent.
To evaluate the statistical significunee, we present t2ie following data.
The total number of epicenters with respect to which the values of the
parameter were determined for the investigated part af the Pacific Ocean
arc was 247, the average value oF the parameter for the entire region was
-0.17 log unit, the standard deviation of an individual measurement with
respect to i,he entire region o.f investigation was -0.27 log unit, the standard
deviation of ari individual measurement wi*,hin each group was -0.16 log unit.
The difference of average values of the parameter for groups of earthquakes
appreciably exceeds the standard deviation of an individual value.
The statistical significance of the results is also confirmed by the pattern
of change in the value of the parameter Y along the Pacific Ocean arc.
Mapping of Spectral Parameters for the Central Part of the Asiatic ConCinent.
Observations within the limits of this territory covered an interval of
epicentral distances from 1000 to 4000 km. Here, in cnrtrast to the
69
,
FOR OFFICII,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR OFFICIAL U5E dNLY
te:l.esetsmic zorie, ttin infliaence of dintance ori the apectra of earthquakes
beeamen appreciable and thc, reuult3 of mapping should in principle reflect
the patterns of ctiangc: in the spectra for different directions. Such
mapE,lng wi11 enable detecti.on of reg3ona:l differences in the value3 of the
pni�ameter Y not wiLh respnct to directions the,t are aelecte8 a priori, but
wl:l.:l reveal their :3patial structure.
'i'he atrong dependence of the spectra on distance impedes quantitative com-
parinons and estimates, makirig them possible only for regions that are
located rit different distances from the station. However, the actual s3tu-
ati.on is more fnvorable for quantitative comparisons of spectral parameters
of dit'ferent regions, since the spectra of longitudinal waves at dist,ances
greater than 1500 km change insignificantly, at least on frequencies lower
than 2500 Hz. Most experimental data used in this division apply to distances
in excess of 1500 km, while estimates of the steepness of the spectrum apply
ta fi-equericies lower than 2.5 Hz. Al.l this has enabled us to apply to the
irit;ermediatc zone as well the method of mapping spectral parameters already
used n.bove in the processing of observations in the teleseismic zone.
Lstimn,tes of the spectra o.f the same epicentral regions from data of different
statlons diverge more strongly in the intermediate zone than in the tele-
seismic zone, and it is difficult to make a quantitative compaxison of maps
plotted for the three stations. Nevertheless, even for these distances the
results of the studies show coincidence of the position of regions of
reduced and elevated values of the paraemter.with respect to all threP
stations.
k'ig� 33. Maps of isolines of values of the spectral ratio y(2.5 Hz/0.5 Hz)
for ttie central part of the Asiatic continent (from data of the ChISS stations
of Ternporary (a) and Garm (b))
Stiown in Fig. 33 are maps of isolines of the spectral ratio y for the central
- part of the Asiatic continent plotted from data of the Temporary and Garm
stations. The region of minimum values of the parameter'is confined to the
70
FOR OFFICIt,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FO[t OFFIC1AL US(i ONLY
TtbEtun plateau. Accordittg to the Garm data it is yituated between 85-1000
G long and 30-36� Nlat., and according to data of Temporary station
between 81-97� E long and 29-35� N lat. According to the data of Talgar
statian, tihe region of minimum values is less pronounced, and is aituated
between 84-100� E long and 29-35� N lat.
If consideration is taken of drift with respect to the directions of propa-
gation of rqys :from the region oF the minimum to the recording stations
(fbr Garm towar-d the west-northwest, for Temporary toward the north, and for
'I'algar toward the northwest), the coincidence of the position of the dif-
ferentiatcd minimum va].ues w3th respect to the data of all three stations
r.en be considerefl good.
,Judging from the map of isolines, the maximum d3fference 3n values of the
parumeter within the limits of the region from its center toward the
periphery is 0,6 10g unit for Temporar,y station, and 1.0 1og unit for Garm.
7'he number of determinations of the spectral ratio for territories inside
Tibet and its framework (conventially taken as a strip 300 km wide) is
21 and 24 respectively for Temporary station, and 11 and 22 respectively
for Carm station. The average values of the parameter inside and outside
of Tibet are 0.75 and 0.35 log unit respectively according to Temporary,
and 1.2 and 0.3 log unit respectively according to Garm. The value of the
st andard deviation of an individual determination from the average in all
cases is equal to 0.2 log unit. The simplest analysis of these estimates
confirms the statistical significance of the values found for Y in Tibet.
7'he few del:erminations of spectral ratios for Central Asia that have been
made From recordings of ChISS stations at Novosibirsk and Bodon also show
n minimum of the values of the parameter in the territory of Tibet.
Another region of minimum values of the parameter is located in the territory
of the i'ersian plateau, approximately between 50 and 63� E long.
In more detailed zoning of the territory of Soviet Middle Asia from the data
_ of Temporary station a strip of low-frequency earthquakes is distinguished
tha,t runs along the Tyan' Shan'-Pamir zone between 67 and 79� E long. Within
the strip the average value of the parameter is minus 0.6, and along its
_ periphery minus 0.2.
According to data of Garm and Talgar stations, a minimum is noted in the
territory of Mongolia with a differential in values of the parameter between
the center and the periphery that is equal to 0.4 log unit.
A region of high-frequency earthquakes is situated to the north and to the
west of Tibet. These are the territories of Northwest China, Dzhungaria,
Altay and Zaysan, Karakorum, the Hindu Kush and Kok-Shaa1. However, the
regions of maxima are not as pronounced as the regions of minima. The differ-
ential in values of the parameter Y between the center of the maximum and
its periphery averages 0.4 log unit.
71
FOR OFFICIAL U5E UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
b'Ok UFFICIAL USL ()NLY
F'1.g. 311. Map diagram of the location of regions of elevated (light shading)
and reduced (heavy nhading) absorption for the central part of the Asiatic
contirient. Repiona are shown that are repeated with respect to the data of
the ttiree stationa Carm, Talgar and Temporary
Ttie most general peculiariti.es th at are steadily repeated in observatfons on ~
the three st ations are shown on the map diagram in Fig. 34� Regions of ~
negative values of Y, I. e. elevated absorption, and positive values, I. e.
relatively weak absorption, axe distinguished.
'Ifi e results, in particular for the regions of Tibet and Mongolia, agree well '
with independent determinations of the absorbing properties of the upper '
mantle within the limits of the investigated territory made by L. P. Vinnik ,
and A. A. Godzikovskaya (1975)� _
2. Particulars of propagation of Lg wavea on paths that croas Central Asia
The intensity of interference groups of waves of the Lg or Rg type depends
considerably on the type of structure of the earth's crust on the path of
propagation. If the path crosses even a small section with crust of oceanic
type or transition type, this leads to sharp damping or even total disap-
[iearance of these waves. It is possible that other peculiarities in the
:;tructure of the earth's crust (disruption of the granite layer and the like)
may also lead to effects of t,his kind. This is what makes it possible to
use Lg und Rg waves as a simple and promising tool in studying the earth's
crust.
In this section an examination is made of particulars of propagation of Lg
waves on different routes that cross the central part of Asia. It is shown
that Lg waves propagate over an appreciable part of the continent situated
approximatel,y to the north of the 35� parallel of north latitude. On routes
that cross the Tibetan plateau these waves completely disappear. The study
tec}lnique that is used reliably localizes large horizontal inhomogeneities
in the structure of the crust, and possibly the upper mantle as well.
72
FOR OFFICIl,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOEt OFF'ICIAL USC ONLY
Region of Seudiee and the Mnterialp U9P.d. 'f'}he lntpnqit,y oi' Lg waven waq
studicd on pathn crossing, Centra1 Asia und the r.nntip,uous territories of
Chinrt, Mdngolla, the Boviet, Unian (Midclle Ahta, Kgzrikhatan, Siberia), India
and in part F'akiytan, Afganistan and Iran. 7'he patha were wibhin the terri-
tory bounded approximately by 75 and 125 E long and 20 end 55 N iat.
'I'he particularg of propagatidn of Lg wavea on routeg crossing the mibetan
plateau were studied in more detail. 7'his territory has received very
little geophysciel study, even ttiough infbrmation on its structure is
extremely importunt for tectonic constructiona and for an understanding of
the mechanism of convergence and motion of tectonic plates. Therefore the
dynnmic pec uliaritien of propagation of Lg waveq and the dispersion properties
of surface waves are so far almost the only sourcen of inforrnation on the
3tructure of the earth's crust within the borders of this mountain region.
Recordings of earthquake3 on two groups of stations situated to the north
and south of Central Asia were examined. Included in the northern group
were the 5oviet stations of Talgar (Tl11"), Novosibirsk (HC6), Bodon (6,QH)
situated 50 km to the east oP Ltilce Baykal, and in part Garm (f'PM). Included
in the southern group were statfons of the world-wide seismic network
(WWSSN) siturzted in India and Southeast Asia: New Delhi (NDI), Chiangmai
(cftc), Shillong (sHr), etc.
The four 5oviet stations used ChISS seismograms in the frequency band from
It to 0.05 Hz. The stations of the world-wide network used recordings of
standard short-period and medium-period equipment. An analysis was made of
several hundred recordings of earthquakes from different focal zones located
both within the investigated region and on its outer border. The magnitudes
of the earthquakes lay mainly in the ran ge of 4.5-5�5. Epicentral distances
were from 300 km to 3000-3500 km.
Regional Differences in the Intensity of Lg Waves. In most cases Lg waves
have the macimum amplitude on earthquake recordings. This applies to the
seismograms of short-period instruments or ChISS channels in the range from
0.5 to 39. The wave group as a rule has clear entries.
F'urticularly clear and intense entries of Lg waves are observed on recordings
of the north group of stations for earthquakes of Altay, Sayan, Pribaykal'ye,
Mongolia and Eastern Siberia. One of the most characteristic features of
the dynamics of Lg waves in the investigated region is that this group of
waves ulways predominates on the recording of all earthquakes of the eastern
und northeastern a2imuths of propagation (relative to the northern group of
:.tntions). T'or earthquukes of the western and southwestern azimuths of
propagation, Lg waves are re.liably observed and predominate on the recording
only within the limits of the first 2000 km of the epicentral distance.
After that their intensity gradually declines, and the spectrum shifts
townrd longer periods.
The Lg wave group shows up on al.l earthquakes situated to the easi, of the
Caspian, to the north and northeast of the Persian plateau and to the north
73
FOR OFFICII,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
rc)i: ort~~rc; i nr, utii: nrvi.,Y
oi' Che '1'.Ibetnm ma:3ni f, With.i n the l imits of' t;he extensive t;erritory con-
veritionally bnundcd by the enumeruted regions, Lg waves propage.te without
iaridr.c�l;c>i.ng rsrky ripprer.inble chnnges. I'tiese estimates app'ly to routes of
nurtheust, nortti und south (to the limits of 7'ibet) directions of propagatiori
re].at;ive to stations located iri the USSR ( north group)..
An exceptionally abrupt change in the arnplitude of Lg waves is noted when
pnths of propagation cross the north border of the Tibetan massif. As long
ctic; thc epicenters of eurthRuakes are to the north of this boundary, running
eiPprnximntely al.orig 35� N 1at, Lg waves are clearly appaxent on the recordings
of ct1.1 northern stutions and predominate in umplitude. On the recordings of
eric-thquakes with epicenters located close to the northern border of the
Tibetun massif, the Lg waves are noticeably attenuated, and are totally
ahsetit if the epicen-ters are 100 km oi� more to the south of this border.
- h'or iristance, for two earthquakes with approximately equal amplitude of
I' wavcs, but citunted on diffei�ent sides of the boundary with a distance of
the order of 150-200 km between epicenters, the differences in intensity of
the Lg wctves may reach a factor of 100 or more. In this connection, on
seismograms of the focus further to the south in the pair being compared,
no wtive grouPs at all can be distinguished at the place where one would `
expect the entry of Lg waves.
I;verything that has been said here is illustrated by the set of recordings
of eurthquakes of two*azimuths of propagation shown in Fig. 35 (Baykal and
2'ibet) obtained at Temporary station. The coordinates of the epicenters of
these earthquakes are given in Table 14. It is apparent that on the
recordings of the Baykal direction (Fig. 35a) the Lg waves can be confidently
identified and predominate at a11 distances right up to 3000 km.
TABLE 14
Coordinates of epicenters of earthquakes with recordings
shown in Fig. 35
B~~Y~tlO~ M~IIp~MMMM
MOlitltOl' M/IIp~~MMM!

J~ 11l11

1 .A.
~ Ilhl
.W.

1
52.9
e9,5
~t
40,0
77,4
1
55,1
03,f
f2
37,4
77,8 , i
3
46,1
88,8
13
37,8
86,9
(
44,9
101,0
14
35,3
86,4 '
S
53,1
f07,8
15
35,8
87,4
6
56,6
114,0
11i
34,0
87,6
7
58,2
120,7
17
3U,6
84,4 ;
'8
56,8
f23,8
f8
32,5
03,7
p
58,9
f27,7
f9
213,2
03,3
10
44,7
82,8
KLY: 1--Ba,yka]. direction 3--~� N lat
2--Tibet direction 4--X0 E long
74
FOR OFFICIi,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
10, 41
~S ~~rG+A+I~+r.+w ~+~~t~'~ ~'~'^"k~'"
L r .
44'
6yt~
iy
s~ 4/
6 .,.AwA...�.......t+.v.~..�w~.N.M,.y�M"~8h'K�~~;~
.
~
'tJ
6g ~ ~
1 minute
' ~
/2 ._._._......_........~:��rM.�~...r~l f~`"~~~-'M~1+~.-h
r
if I
rok c~Hrr.rrni, Ilsi, ntvi.,v
a
~ :..........`..r,,,~,
1 minute
1
 . _ _ _ _�.....,rl,y~~,~,
~9t
f r..~..... ...~-.�........w....... ~ . .....~M.w.,o.~~,Mw,w~..ti.
, w.... . . ~~,+MM,~'~n~t~.�u~.,~,~....
e �~ti.-.�~.~a.....M,.w~..-,4..+~-*.*-�~h,~:. e.M�,wy,*.ry.,~,n/v~! ~ ' '~'(~1~~~;''~;~*~iy'~~
Ly/~~ ~ I f
t �
ow~-TMM�~M......... il ~ i~ 
'9 ~+tMM~w.t~MNAd+I'Ar;;�y!~^�M'+'y1~,~~.~^(~ ~ ~I ~ r~ I ~ ~I' ~I, ~i ~
Ly1
P+
As
19
~
c9l
r-vM'.Vl.~~jr"'"""'w""..y"+"' �
s i9, ~l
75
FOR OFFICII,L USE ONLY
Fig. 35. Examples of recording
of Lg waves at the ChTSS Temporary
station for earthquakes in the
northeast (Baykal) and south
( Tibet) directions : passband
of the channel 1.4-2.3 s. The
numbers correspond to the
earthquakes shown in Table 14APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fit;. 36. Mn,p:; of epicenters
t.lic relative intenr,ity of Lfr,
(b) stations: 1--Lg waves p
::ci::mogram; 2--Lg waves have
ma,for fructures are shown by
N'dtt UFFI(:1 A1, 115I.: c)NLY
.
of earthquakes showing qualitative evaluation of
waves from recordings of Talgar (a) and Temporary
redominate with respect to amplitude on the ,
low amplit ude; 3--Lg waves entirely absent;
solid and broken lines.
76
FOR OFFICIt'iL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
1~uk ii;;E ONI,v
An i.ppruxlmaLrly i3imf lar puttern i3 nbservcd nn recordingn of the 7'ibet
cltrect,ton (Fig. 35b) un, 1dnN nn the epir.enters are ta thE north of the Tibetari
masni f( No 10-13 Cpi ccntc rc: rs i tuuted di rec tl,y nn the boundary of the
rnassni f' (recording No 14 ) rire eharneteri zed b,y u noticeably attenuated Lg
wuvr., A rtill nhrirpr.r cnntrnit, i:s upparent between the recordings of eurth-
cluuicen tJo 15 cxnd No 16, locnt.ed directly on the boundary nf the massif and
90 km to the south respectivc:ly. On the recording of the south earthquake
- the Lg wuves are totally absent. I'hey cannot be seen at all either on
enrthquukes stil]. further to the sout;h (No 17-19
Siach strong difi'erenr;e in the intrnsity of' Lg wave3 enabled us to use a
simplified method of classiC,ytrig recordings. All seismograms considered
were hroken down into thr,ec group:; wlth respect to amplitude level. 'I'he
f'1 r:it qrotap included recorclings with FLn 3ntense Lg wave (of the type of
No 1-1.2, Fig. 35a, b), in the secnnd group the amplii;ude of the Lg waves
ic upproximately equal td that of p waves or somewhat lower (type 13-15 on
35b), und the third group included no Lg wavea at all (type No 16-19
ori Fi g. 35b).
`l'he resu].ts of' such : orting of thc recordings obtained at the Talgar and
`I'emporury stationy are shown in Fig. 36 in the form of maps of the epicenters
of the g.iven earthqutikes. Tbe type of recording is arbitrarily assigned to
t,he epi centcr.
Unfortunately, such maps werc con:;tructed only for northern stations since
wc had little available experimental duta on southern stations.
A compiirison of the recordinp,s of the sr3me earthquake at different stations
z;hows thnt the intensity of the Lg waves is practically independent of the
source, but is determined by the location of the path. F'or some earthquakes
foc� which we had recordings made at several stations, the positions of the
epicenters are indicated together with the paths of the seismic rays corre-
sponding to the system of observations (rig. 37). Here we selected only
those epicenters such that theii� position relative to the recording stations
enabled more precise determination of the boundary where the Lg waves
disuppear. The prevailing system of observations does not permit an answer
to Ghe cluestion of whether the disappearance of the Lg waves is connected
w.i t:h the in fluence of thi , geol.ogi ca]. boundary or these waves do not propagate
withi.n ttic limits of +,he tei�ritory of the Tibetrsn plateau.
FiW. 38 shows the amplitude graph, of Lg waves for two directions: Baykal
_ uncl 7'i bct. On these g,raphs, plotted from data of the Talgar and Temporary
:;tfitions, the shading shows the interval of epicentral distances that corre-
opond:; to the north boundary of Tibet. It can be seen that for the Tibetan
direction in this region the umplitudes decrease by at least an order.
;;uch an cotimate is the lower limit of the change in amplitudes since on the
eart}iquake recordings where the Lg waves were absent the phase amplitudes of
scuttered oscillations were measured on the section of existence of Lg waves
thrst was isolated in accordance with the average hodograph.
77
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
i:
Fnk aFrtcinL usE ONLY
~
11 W
M irr &AN - '
`'-y...
~ -
0 :
p tlll 1 Il ~
r '~h i. ' ~
M ~  . .
~ ' 10 r .
~ ~ m ~ ~ �n _S i~= r ~+i
~ ~ ~ ~ ~1�~ ~ -'~i..':.? s: t 11
M~ 1 Q 1 `i. a'.:. ;N
~
e�v~ 1 _
. ..La
w w u~ i�
U M M W IN 1)~ If1
M
w
1
0.
1
\
y '
L s_\IU
,
O
% % A

.:E�f
A ~
~ z~
~
Y
,
/ �j
. .
~ ~
p
..7
r 11~ . 171i:
Fig. 37, Maps of epicenters and paths of propagation of Lg waves toward the
tinrthern (a) and southern (b) groups of stations: 1--selsmic stations; 2--
ep:icenters; 3--"sha'rp" boundaries that totally screen Lg waves with position
f'air].y reliably determined; 4--the same boundaries drawn hypothetically; 5--"weak" boundaries where Lg waves are appreciably attenuated when they ,
cross; 6--puths corresponding to recordings with intense entries of Lg waves;
7--paths with recordings of intermediate type; 8--paths corresponding to '
recordings with total absence of Lg waves; 9--deep-level breaks that outline
tte Tibetan maJJlf according to the tectonic map of Eurasia ,
E'rom the graphs of the Baykal direction we can see that the intensity of Lg
waves chan6es weakly when they cross such a large tectonic structure as
the Aayknl rift zone.
The position of the boundaries that cause disappearance of Lg waves or that
separate regions (blocks) that are "favorable" or "unfavorable" for the
78
FOR OFFICIl,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOlt nb'FICIAL USL; UNLY
t
!
a

1N A/T, um/s
, . u
 �
.
A. i.~..
. f-r�-
.
� � � �
� � \ ~ �
. ~ ~ �r y.,~ � .
� \ _
^ ~ ' � .
~ .
N~ ~ � ~
�
J � � �
� ~ �
~ � ~ ~
�
~ � ~
. � I ///%%I ~ � J
.r
100 J00 J000 lJ00 1000 d,MN
lg A/T, um/s
� ~
� aAl+ �
/
0

.
� ~ e �
oa
_ `q,\�~ e e �e
- e,
a,"
F'ig. 38. Gruphs of attemiation of the amplitudes of Lg waves according to
the data of Talgar (a) and Temporary (b) stations for the Baykal (a) and
Tibet (b) directions. The shading shows the distance interval that corre-
sporids to the north boundary of the Tfbetan plateau
propagation of Lg waves can be more precisely determined from aaialysis of
the maps shown in Fig. 36 and 37� First let us consider the north boundary.
This boundary passes upproximately between points with coordinates 34� N lat
nnd 780 E long, and 35�5� N lat and 90� E long. Its position with respect
to latitude is most reliably determined with an error of less than 100 km.
This is o.pparent from the maps (see Fig. 36) when a comparison is made of the
intensity of the Lg waves from recordings of two closely spaced epicenters
(No 8 and 9 on Fig. 37a). Analysis of all data confirms that the north
boundary is the sharpest and its position is determined by the accuracy of
calculation of the coordinates of the epicenters, which is apparently not
high here, and accordinQ to our estimates amounts to 30-50 km, which is due
to the unilateral location of the st ations.
We cannot speak so confidently of the western and eastern ends of this
boundary. F'or instance the west end is apparently at 780 E long, which is
vcrified by the data of the maps (see Fig. 36) and the sharp difference in
the intensity of Lg waves on paths going from epicenter No 2 toward the
northern stations (see Fig. 37a).
The pasition of the eastern terminus of the boundary, according to data of
different stations, is ambiguously determined. With respect to the recordings
of Talgar station (see Fig. 36a) the boundary that sepaxates the two types
of patr.s goes along the parallel of 36� N lat and terminates at about
97-980 E long. The few determinations from recordings of the Novosibirsk
. station indicate that beyond the point with coordinates 360 N lat.and
79
FOR OFFICIl,L USE UNLY
~
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fnk UFF'ICIAL U;ils UN1,Y _
90" E long t;}e hnuttdury flnes in the northeusterly direction ta approximately
961 L lonp,, cainCidintf wit:h the deep-level brenlc thgt bouncle ttie 'Pibetan
mnsaif on the northefLet. 'I'hie hypothesis is also confirmed by data on the
int,enuity of I,g wo,veu on diFf'ereni; rautes nccording tn the recordings of
fiodon station.
'I'tie southwest und south boundurieo of the block, which do not transmit
Lg waves, are localized by recordings nf the southern group of stations
(see Fig. 37b) with lesa surety. On i;he section between 78 and 901 E long
they apparEntly coincide with the system of deep-1eve1 breaks that separates
ttie mibetan ma: sif from ttie iiimalnyas (the Karaltorwn break on the southwest
and the "Indian suture" on the south). Purther on this boundary goes off
toward the soutlieast, apparently coinciding with the general trend of geo-
logical structures, but in view of the dearth of experimental 3ata on the
intensity.af Lg waves, it can be only hypothetically drawn.
'I'hc effect of disappearance nf Lg waves is observed, in addition to Tibet,
on paths ttiat cross the Zagros break zone separating the Iranian plateau
from ttie 7,ugros Mountains. However, we have not examined this region in
detail.
Along with the "sharp" boundaries, one can distinguish "weak" boundaries at
which a redtiction in amplitudes of Lg waves takes plaee. Such a boundary
(Fig. 37) passes along the Hindu Kush-Karakorum arc, and its continuat3on
to the southeast� Among the "weak" boundaries is the strip that coincides
with the Kopetdag Foothill downwarp, and further to the east with the Herat
brefilc; n,nd also the northwest border of the Indian shield (Beluchistan, the
Sulaiman Range). Thus an investigation of the relative intensity of Lg
waves on routes crossing Central Asia and the contiguous territory has
shown ttie foll.owing.
- Lg waves are reliably registered over the entire extensive territory of the
Asiatic continent that is situated to the north of the Kopet-Dag and its t
continuation, }iindu Kush and the Tibetan plateau, and also on the terri-tories
of the Indian shield.
Lg waves are attenuated as they cross Pamir-Hindu Kush, the north border of
the Indian shield, in propagation along Tyan' Shan'. `
Lg waves arc completely absent on recordings if the epicenter is located
wittiin the Tibetan massif, or the path of the ray even partly crosses this ,
territory. Even 100-150 km of travel through Tibet is sufficient for total
disappearance of these waves on recordings.
3. Regional differences in the characteristics of Rg waves
The Rg wave groug in most instances is observed on seismograms simultaneously
with Lg waves. The properties of both wave groups are to a grtat extent
analogous. They propa.gate only on continental paths, have comparatively
clear entries, their velocity is independent of distance.
80
~
- FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
Fok aFFrr,rnL usi; ONLv
In the intermedictte diytanr.e zane under t,he conditions of tihe southeastern
USBIi, kig wztive3 utiual l,y prec3ominate t n nmpl i. t;ude throuph the enti re trai n of
csurt'ucc waven, oLnd hava a p,rour velnc.ity ot' aboiat 3 km/si (Ru2aykin, Khalturin,
1)'lli), For eartliquakes with rnagnltude of less thun 5.5, thia group is
frequcntly the only one to be distinguished in a train of surface waves,
especially at distances of up tn 1500-2000 km.
7'he Rg wave group usually tates the form of a short train on the geismogram,
consisting of several extrema wittiout pronounced dispersion. 7'he apparent
period of the oecillations in the group is 8-12 s. The low-frequency charac-
ter us well as the fuct ttiat ottier poorly identi fiable surface waves follow
ttiis group make it difficult to diatinguish Rg waves on seismograms. There-
fore, parumeters that ctiaracterize the entire train of surface waves were
picked out ns the object of rreasuremecits. In the great majority of cases
they applied directly to the Hg group.
~ 7:n this sectivn we give the results of investigation of the parameters of
n train of Rg surface waves its duration, shape of the envelope, time of
onset of maximum amplitude observed on different paths crossing the
Central part of the Eurasian continent. It is shown that the changes in
these parameters with transition to more remote earthquakes take place in
rnfiny instances abruptly rather than Qradually. This made it possible to localize the territories responsible for such changes, and then to interpret
them a,s large horizontal inhomogeneities in the structure of the earth's -
r.rust. The Tibetan plateau is the largest such inhomogeneity in the investi-
gated region.
Moreover, to ctieck the methods developed, the characteristics of the train
' of' surface waves were studied on dif.ferent paths corssing the Black Sea.
llccording to data of several stations, we have reliably localized the
deep-water part of the Black Sea Basin for which a crust of suboceanic or
transitional type is typical.
Materials Used. Method of Measurements and Processing. The recordings of .
a channel with passband of 10-20 s of the Temporary ChISS station were used
for studying the parameters of Rg waves. The data of a total of 300 earth-
quakes of Soviet Middle Asia, Siberia, Mongolia, China, India, Pakistan, Iran
and the Caucasus were processed in all. The range of epicentral distances
was from 600 km to 400-5000 km. The use of band filtration with magnifi-
cation of the order of 15,000-20,000 enabled confident differentiation of
trains'of surface waveseven for cvmparatively weak eaxthquakes *aith magnitude
of 4.5-5�
- The upparent periods of maximum amplitudes ranged from 7-9 s at the beginning
of the tracking interval to 12-14 s at the end.
At distances up to 1500 km in all cases, and at greater distances in nearly
all cases, the train of surface waves was simple in form and consisted of
3-4 extrema with a comparatively clear entry and very weakly expressed
81
FOR OFFICie,:. USE U1VLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
i
6"dK UFFCCIAL USIs ONLY
d:1 3peraton. 7'he grotap vel.oeity way equnl to approximately 3.0 km/s, varying
over u range of 7-8% i n cit f'f'erent di reetions .
_ `l'o dc:scribe kinetnutic peculiarities the parametcr t'max = tmaX -(0/3.0) was
u;leci, where t,maX Is tho time of onset of the maximum of funplitudes in the
ti,fiin of :iui�Face wuves ut epicentral distance of A km.
`1'he form of the train was characteriLed by duration T-- the time of existence
on the setsmo6ram af omp]itudes of at least 1/3 of the max3mum.
'I'tie amplLtude chnnge3 on the route were characterized by the magnitude
correction c1M = Mst - M, where Mgt is calculnted for the given station, and
the average value of the magnitude M is determined with respect to the _
network of stations.
'1'hc Cttrther technique for representation of primary data was identical to
ttint used in processing the materials of Lg waves. Curves were plotted for
the parameters t'maX and T as funetions of distance for several directions
oP propagation, these parameters were mapped by assigning their values to
eptcenters, and the values of the parameters were compared on different
ptitlis ci-ossing the investigated territory.
Ctiarges in parameters with distance were considered for four directions of
propagation relative to Temporary station: Northeast Altay, the Sayans,
1'ribaykal'ye, East Siberia, Kamchatka, the Aleutians; East Northwest
Cii ina, Moril;olia, Northeast China, Sakhalin, Southern Japan; South Northwest
China, Tibet, South China, India, the Indian OcFan; Southwest Soviet
tdiddle Asia, Afganistan, Pakistan, Iran, the Caucasus, Turkey.
The time of onaet of the maximum amplitude t'maX reduced to a group velocity
oF 3.0 km/s is shown on Fig. 39 as a function of epicentral distance for the
four directions of propagation relative to Temporary station.
It can be seen that on the graph of the Northeast direction (Fig. 39a) the
velocities lie in a range of 3.0-3.2 km/s for the entire continental segment
of the path within which the Rg group is distingizishable. At distances of
more than 4000 km (beginning of oceanic routes) the times t'maX and ac-
cordingly the average group velocities increase sharply.
Typical of the East direction (Fig. 39b) is appz�oximately constant time t'max,
group velocities are 2.9-3.1 km/s. The most remote earthquakes (Sakhalin,
Southern Japan) are situated on routes with a continental type of crust.
For tlie South direction (Fig. 39c) the Rg group is reliably distinguishable
anly up to 1500 km, beynnd which it disappears on all recordings for paths
ttint cross the Tibetan platenu. On other routes in the same direction the
Rg wave group gradually stretches out, the number of extrema increases.
On Tibetan routes, recordings show up with large values of t'max, wk-iile for
other routes this quantity increases gradually. The avPrage group velocities
82
FOR OFFICIl,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
roR orrxcr.nc, usE' ONLY
.
c
,
~
90

.

,
,
�
s
�
. �
'
~o
�
'
t,/
!so
�o
O
~
. .
_
d
,
.

'
.
_ . ~90o zoa,~ rnoo ~coa a.NM
Fig. 39� `r{mP o.f onset of .maxi.mum amplitude t'm in a train of sw�face
waves xeduced to a groizp velocity of 3 km/s as aahnction of epicentral
distance from data of the Temporary ChISS station (.passband of the channel
10-20 s) for the Northeast (a), East (b), South (_c) and 8outhwest (,d)
directions
lie in a range of 2.7-3.0 km/s, and axe much lower for recordings from foci
locuted inside Tihet (distanceG from 1600 to 220 km).
In l:fie Southwe;t direction (Fig. 39d) up to distances of 1500-2000 km the
trroup velocities are 2. 8-3.0 km/s, and then they gradually decrease to
2.6-2.8 km/s with increasing distance. Crossing the Caspian leads to total
transformation of the train of Rg waves; it is considerably stretched out,
and the times of onset of maximum amplitude increase shaxply to 150-300 s.
Duration of Recording of a Train of Suriace Waves. Graphs of the change in
duration with distance for three directions are shown in Fig. 40.
Typical of the East direction (Fig. 40a) is a weak change in duration: within
the limits of the continental section of the path (up to 3700 km) it rema.ins
83
FOR OFFICIl,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
v b .
d/ � '
:
. ~
. .

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
r01t OFrcc;rnL usE nrvi.v
Tg ;
tuu a
JO - D ' '
. .
10 " , �
/d
6 ~
b .
~ i ' �
JO� ~ ~.ti ~ ~
)0 . . . .
!0 ~ ~ip1
B C ' � '
~ �
/00 - ' + � ,
. , ~
d0 , M. .
' � .
10 . � 4.40
~ A~~Y SG'9 /0T4 I000 000 y000 JOOOd,MN i
Fig,. 40. Distance dependence of duration of a train of surface waves for
- the Northeast and East (a), South (b) and Southwest (c) directions
approximately equal to 15-30 s. For earthquakes from the territory of the
- Prlcific Ocean the duration increases sharply to values of 100-300 s. The
Ni, K.roup follows and retains its simple shape over the entire continental
path. It is lacking in more remote earthquakes; a complex distended pattern
of surface waves is observed. .
~
- In the South direction (Fig. 40b) the investigation of duration with distance
follows another pattern. Here the train keeps its original for.m, and accord- ,
ingly a short duration of the order of 12-30 s up to distances of about
1300 km. A bend then talces place in the t.rain that is especially noticeable on recordings of earthquakes located within the Tibetan plateau. On these
recordings the compact Rg group disappears, and a distended group of surface ,
waves shows up at later times. The ovei-all duration increases sharply and
then changes but little, remaining within limits of 40-150 s.
In the Southwest direction (Fig. 40c) the duration increases approximately '
in proportion to the square of the distance. A sharp increase in duration ~
uT) to 150-200 s--- takes place on the recordings of earthquakes on routes
ttiat have crossed the Caspian Sea.
Tl1e parameters o� the train ef surface waves tlmaX and T were mapped by thE
met}iod described above an applied to the spectral parameters of longitudinal
84
FOR OFFICIr%L USE UNLY
u
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FoR oFrr.c; CAL USC ()NI,Y
wuvea. '1'ie val.ue of � pnrr.unctrr was nom! nu1:1y assigned to nn epicenter and
lntermediut;e uvec-aging wa:3 then done with respect to 3-6 closely spaced
epicentera. Maps of' thls stuge of averaging of the parameters t'mgX and T
Kre ahown on T'if,. 41. They illustrate ttie ma,jor trends in variation of
parnunetcrrti ln different c3i rectlnns, rznd tndi r.ate the spatial confinement of
pnintu of greatest criarige;s in ttie pareaneter:; (maximum values of the gradient).
Fig. 41. Maps of the parameters t'max (a) and T(b) from data of the Tempo-
ra:r,y ChISS station
Within the limits of +he transition zone from the Asiatic continent to the
Yacific Ocean, abrupt changes in the parameters occur only where the rou:;es -
cover a section of crust of oceanic type.
Among those territories with a maximum gradient is the Caspian Sea. On
roiates that cross or even graze the southern part of the Caspian Sea Basin
(with crust of transitional type) the train is completely broken, resulting
ici nn nbrupt rise in the measured parameters.
Somewhat less of an increa.se in the val.ues of the parameters also takes
P1ace wheci the 1'ersian plateau is crossed.
85
FOR OrFICIl,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
N'nR ONFIC CAG USI.: UNLY
A:;r:ct,ion of the :3trongest chunges in the pnrameters of a trn,in is situated
on t;hc terrltar,y of Centrul Aaia, confined to the northern boundar,y of 'I'ibet.
For in:itance on recordinNq of eurthquake.; with epicenter;, to the north of
'1'tt,ot wtthln iL f3t,rlt) fippr-oxlmatel.,y 500 km wic3e t,tie vMlues of the parnmeters
ti nu1x and r ure equal. to 35 nnd 28 s; on the recnrdings of earthquFiltes
s.i.tuated i.nside 'I'ihet, the values of the parctmeters increase right away
t:n EiU rznd 125 s re:;pectivrly. A comparison of the parameters of earthquakes
c::lo:,e to the boundary of TibEt but on both sidea of that boundary showed that
Lhe pnro.neters change sharrl,y rather than gradually.
W
Q,MN
Fif;. 42. Specimens of recordings of trains of surface waves for directions
ttiat pass to the west of Tihet(a), cross the Tibetan plateau (b) and pass to
ttic east of Tibet (c). For the direction that crosses the Tibetan massif (b),
we d.i:;tinguish recordings from earthquakes located to the north of Tibet (bl)
n.nd within Tibet (b2). The vertical line corresponds to a group velocity of
3.0 kmjs. The arrows indicate the times on the seismograms that correspond
to group velocities of 3.1, 2.9, 2.8, 2.7 and 2.6 km/s. Temporary ChISS
stFZtion, channel with passband of 10-20 s
`I4ie influence that the boundary of .:ibet has on the shape of the recording of
a tr-ain of surface waves is illustrated by the set of seismograms shown in
Fig,, 42, Shown here are specimens of the recording of a train of the Rg
f;rour for three comparatively narrow directions of propagation: a--for
paths t,o the west of Tibet; b--for paths that cross Tibet, and c--for those
goinE; to the east of Tibet. A distinction is made between the recordings
oP carthquakes from epicenters located to the north of Tibet (bl) and directly
in t;he Tibetan ma^sif (b2).
ilE, to distances of 1500 km the sh ape of the train is approximately the same
rinci cornparKtively simple in all di rections. After crossing the northern
86
FOR OFFICIAL USE U[VLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FQFt nFFICTAL US}; (1Nl,Y
boundary of 'I'Ibet; or it:3 cont.tnuiLtion (1600 km fnr ttie western sector, 1700 km
fbc- the 7'ibetan ncetor slnd 2000 km for the eastern), the shape changes
noticeably, enpecially aharp.l,y on the recording
,s of earthquakes located
innide ttie massi f, 7'he t;irne nP onset of' the maximum amplitude of the train
iricre:ased by 100 s relative to the velocity of 3.0 km/s, and the shape of
ttie train wa,s noticeably stretched out.
Crndual].y, with increasing epiceni;ra:l d.tstance in the South direction, the
train is somewhat cotisol.iclated, and its shape becotnes closer to that of
the neighboring directions at the so.me d.istances.
7'hu.~~ when the northern boundary of the 'Pibetan plateau is crossed there is
a loss of the Rg phase and assoctated strong changes in both parameters of
ttie train. These changes }inve ttie same scale as with a transition from
paths of propagation alnng an oceanic type of crust to routes with a conti-
nental type.
Propagation of Surface Waves on Paths that Cross the Black Sea. As has
ttlreudy been stated, to evaluate the possibilities of the method and compare
ttie scales of distori;ions that a.rise in a train of surface waves when hori-
zonta]. inhomogeneities are crossed in the earth's crust, anal ogous work was
dorie f'or the territory of the Black Sea Basin.
It is known (Balavadze, Mindeli, 1966; Neprochnov, 1962) that the deep-water
part of the Black Sea is characterized by a crust of intermediate type in
- which there is no granii;e la,yer. The boundaries of this section and its
structure have been independently studied by several geophysical methods
wi.th the maximum possible detail.
,;eisrnologists have established (Savarenskiy, Val'dner, 1960; Sikharulidze,
1963) that Lg and Rg waves are totally screened by this part of the Black Sea
Basin. The belt of active seismicity that passes to the south and west of
the Black Sea, and the presence of a network of recording st ations provide
good conditions for "x-raying" the entire Black Sea area and ad,jacent terri-
tories b,y using Rg waves from seismic sources on different paths.
A11 ttiis makes the region of the Black Sea an ideal testing ground for
evaluating the possibilities of the method and quantitative comparisons of
thc scales of ineasurements of paT�ameters here and in Tibet.
To study cha.nges in the parameters of a train of surface waves on routes that
cross the deep-water part of the Black Sea and pass outside of it, more than
300 recordings of 120 earthqurskes on stations of Obninsk, Simferopol' and
Ba.kuriani were processed. Zfi e parameters t'max and T were measured on
seismogrtims of SK and SKD instruments. Besides, the magnitude correction AM
was calcul.ated fo.r each recording.
~
Ttie results of ineasurements in the form of maps of the intermediate stage of
avcrAging are shown in Fig. 43. Given here for each of the three stations
87
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR OFFICIAL USIs ONLY
Fi. g. 43. Maps of values of the parameters OM (a), t' max (b) and T( c) [next
� pn.ge] according to data of stations at Obninsk (top), Simferopol' (center)
and Ba.kuriani (bottom)
considered are maps of the magnitude deviation AM (Fig. 43a), the parameter
t'maX (Fig. 43b) and the parameter T(Fig. 43c). .
Une can readily see the effect of an abrupt iracrease in parameters t'max
FLna T and a reduction i.n AM on paths that cross the deep-water part of the sea. The average values of all three parameters with respect to each st ation
88
FOR OFFICIitiL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
a b

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000100010044-1
FOR OFrICIAI. USL" ONLY
c
~
Fig. 43 (continued)
separately for routes that cross the deep-water basin and pass by it are
summarized in Table 15 along with the standard deviations of the individual
determinations of +.he parameters.
89
FOR OFFICIt,;, USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
23 JANUARY 1979 FOUO
2 OF 2
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOk UEFtC:IAL USIs UN1.Y
'I'Alli,r I',;
{'arumetor: of u train oi' uurCur.c Wrivc:, f'or routcn t.hat Croar,
t;he rtvep-wrt,~r bcic3in (1) ttn(l pa:;:-, by it (2)
~
~M
,
~MU ~ t n
t
CMr
.-4,7
--O,Ib
IMI 50
9.i1:
110
Caropono,ft
-0,7
-0,10
1711 5c)
1N11
DU
6nrPrwr
-U,A
0
1U) 15
250
YU
; 5 cva.N
-o,et
--o,oe
iw w
xa
iw
0
0.12
0.16
a 23
50
as
P;'r:Y: 1--Station 4--REikurifini
2--Ohninrk S--Average
3--;~ i m!'erc,pnl'
An:1;/:,is of thc dntu in Lhe Tahle :hnW: that crocGinK of the dpep-watpr basin
b; the path of a i,eiamic ru,y rihOW3 up in the fo1laW,ina changPS of parameters;
rcr3uotion of the maxi.mum nmplitude of the train of :urface wevea t;y a factor
af' 4, inCrenr,e in duration by g fuctor of 2.5, end in the paraineter t'max
,t fnctor o^ 4.
'1lif. ;;t.at,i st i rKl nipni fi rnnce of tbi K ei'fect ca.nnct hP dovtted 4ince thP
,ii ff'(�r(-nre ot' t.he nvPrap,e3 excePdn n the stFSndnrd deviAtton of an indi-
v:duai determini,.tion of the parameter by a factor of 4_5, t3ssuminq thnt
ttve uverage value was determined from samples containing 20-25 individual
cieterminations.
Compartson of the Influence that Large Inhomogeneities have on the Parameters
of u Train of Surface Waves. An examination uas made of the largest inhomo-
s'eneit:es of the earth's crust of the investigated territory whose influence
- nr, t.he parametcrs of the trair, wa,s detected in anal,ysis of seismologic
rr,a+.eri ri13. Surf ace Nnven experience strong distortions when they cross the
rtm mar:-df, the Caspian snd black ueas, the Persian plateau and the
1.r�fin::itlcin zoric from the continent to the ocean in the territory of the
Kuri 1-Kuncnritka arc.
';1ic riveruge value , of the parameters of the train t'max end T were determined
Tor ttif- cnumerated structures from the recordings of earthqualtes with foci
toc:nted on both sides of the tttructure Within a 300-500 km strip. The
fir:iL group (1) includer, eurthquakes whose routes cross the investigated
utrucLure; the paths of the second group (2) do not eross (Table 16).
'I'iie muunitude devfation AM was also determined for the Black Sea territory.
Tt,e value of this parameter t'ox the transition zone betWeen the continent
t
90
FOR OFFICInL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'Olt UNF[C1AL U5L ONLY
lru1 1,10 ocern i n the v i (l. i n I t,,y oi' t,he Kur{ 1-Knmr.hat,ka are it; ttilten f'rom the
work of I,. ;olov'yc!v rlnd V. ii, N,h~;tn (1959).
`PAE3LF; 1(
F'arcuretcrg of a train of r,urface wavcA o1' groups of earthqunkes with paths
that cross (1) und do not crors (2) 1arge utructural elements
~mu
~
AM
(1) cnr.,r.ms .~...ti
2 KacnMleaa "
100
43
230
45 - -
~ 4ePra rope
I54I
Ml
260
IOU --0,7 -0,1
(tePeMneIIa nAno
DO
50
I10
40
r) Tn6eteul rseeas
Eu
95
tJU
30 -
b Kerutu-- KyOneY
lw
u
20U
3U -O,S u
KEY: 1--Structural elements 4--Percian Plateau
2--Ca3pian Sen 5-=Pibetun Massif
3--Black Seu 6--Kamchatka-Kurils
Malysis of the duta in the muble uhowc that all th:vi structures where
routes pa.,a over a crust of ocennic or trnnsitional types the Black and
Caspicsn seas, and also the transition zone between the continent and the
coeun di3tort the train of surface waves to an approximately equal extent.
Someahat leaser distortions ta}te place with croasing of the Tibetan ma3sif,
and still t;maller when the Persinn plateuu is croased.
7'tiu r, "x-rqyinp," of extensive and varied territories by Lg and Rg Waves
has demonstrated the feasibility of distinguishing and localizing the
large.t inhomogeneities in the structure of the earth's crust, and possibly
the upper mantle. The developed technique can be used primarily for
st.udying the position of buundaries of continental plates in little investi-
gated regions or localizing sections of the earth's crust characterized by
the absence of a"grnnite" lqyer.
Ch3ptcr 3. Particulars of the Structure of the !lPper Mantle in Centi�ul Asin
1. Structure of .*.he upper mantle along the Baykal��Pamir profile
'I'lie ::1,ructure of Lhe ijprr-r mrsntle of the territory of Central Asia was
ci on ttie bosis of sei:mop,raphic materissls obtained on the profile of
:ztu(ite
f'r3mir-I3aykal seismic :tations (Nersesov, Rautian, 1964). Observations on
thir, profile uere bep_,un in 1961 and completed in mid 1963. In 1962 the
number of stations reached 54.
Earlier, on the basis of a detaileri study of the deep-focus aone of Pamir-
Flindu Kush carthqurskes, ttie structure of the upper mantle in this territory
had bcen ctudied to depttis of the orcier of 800 km (Lukk, Nersesov, 1965).
91
FOR OFFICII,L UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'OR nFFICIAL USL UNLY
f,ut.er on, thF~ matr_rin1,3 oC obucrvution:; on the Ptinir-BaykFSl profile wcre ,
pro(_-vn;~r,d on t.hr_ t,uE~in ot' meLhodr (levelohed nt the cnmputing center of the
,",(t,urirun l~~,~~~rt,rnun~. oI' t,hr, ;;oV1rt ACilclF:ttky ot' "cicncrs t'or numericnl solution
r) f' i rivuresr_ t;hrri~-t U mrriej 1 rinni k i nanut i r. p roblem:s ( Alekseyev et al 1969 ,
~ 19Y1) ,'Che revultr guve ngenernl ideu of thr. ntructurp of the upper mtintle
in the centra.l pnrt of the proi'ile.
'Chtn oection giveu recsultr of ii more complete analysia of the structure oi' '
t,he upper muntle based on the use of proft lr_ (ltsta on the kinematic Chargeter-
i st.ica oC first and subsequent entries us well as spectrnl dynrimic character- ,
i.:,t,i.r.n of longitudinal wgvefi from the materinls of the Talgar ChI55 gtatinn.
in the procers of interpretation, exNrimentnl hodogrgphr, were compared with
r,hcoretical hodographtt calculsted by M. V. Alekseyeva for a number of eross
:,rrtions of the upper mantLe. 't'he nuthors are sincerely grateful to her
f'or furnt nhinp these muteri als.
TAI3LE 17
List of the earthquakes used in constructing graphs
for the Pamir-Bqykal profile
1
1
~ K�rA..Aw
( 2) . (
M~M~fp~

r 7. ~
'~M~K~ '
K~ AIt!
/A-H
~

e 7,.r.
B,.A.
~Y
K UIK1l~1
Soviet Middle Asia and Kazakhstsn
23.v111.19e1
a
u
3e,5 3e133' ee�ar
2o
13,0
21.Ix,19E1
pS
p
08.0 40 2t/ YO 13
fS-ZO
12.3
011I1.1901
pY
32
SS.O 41 JS 7508
b-fS
10.0
31.1.10e2
oo
aa
s3,0 3839 7007
5-10
ia,s
29.111.1063
13
,6
23,0 . 008 82 32
5-10
12,6
71.IV.1962
01.
11
0,6 {4 13 7843
S
12.4
01.V.f062
11
12
70,0 38 36 68 30
S
U.0
Altay
and the Seyana
21.V.1911
01
06
57,0 is'21 SS*30
5-20
10.4 �
20.Xt.1i11
01
W
47,0 5040 03 45
b-.2p
17.0
13.IV.1163
le
36
0.1,0 49 77 A1 JD
5--20
12.7
22.1,19bl
01
26
41.0 S2 n 10023
5-90
13.2
Pribayksl'ye
4t.X3961
22
13
08.0 S3�35 IOd'SO
0--3J
11,0
. 33. X(.1061
01
11
0.0 SS S3 10053
0-31
13.0
11.I.1Ob!
14
fb
37.2 $430 fll 00
0--3i
11.4
KE,Y : 1--Date
6--Coo; linates
of
epi centers
2--mime zt focus
7--N lat
3--hours
$--E long
4--minutes
9--Depth, km
5--second3
10--Energy clas
s K
(lOK J) `
'I?ic sei:;mograms of earthquukes with data summarized in Tabl.e 17 Were used
~ for ttie kinemntic constructions. The earthquake materials Were supplemented
92
FOR OFFICIAL L'tiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Pnk nFFIC1Af. USN: dNLY
Yz
roo
!00
/00
/00
JOB
JOD
700
Nr
d
t
`
B
d 0
Q
0
43
~
~
~
�
P'4
~
km/s
7
/O
i
I
e y,u ii r e.r iu n
~r
r
~
-
.
/
1J

i~I iql
e,r
42f -
i
f
i
Fig. 44. Diap,ram of hodogrxphs of longitudinal waves on the Pamir-Baykal
profile: solid lines--first entries; dashed lines--Pg Waves; dotted lines--
reflected waves for the loop section. Shown belox are the velocity profiles
of the upper mantle for the corresponding sections of the profile, plotted
nn the ba,is of interpretation of hodo;~raphs and spectral amplitude curves.
SeP Fig. 45 for symbols. The .hading :.hows the regions of ambiguous determi-
nation of depth and velocity
93
FOR OFFICI149. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOkt OFFICIAL USE ONLY
by selc:tnograrns ai' three r.ompnrgtively 1arge pxplosians (600-700 metric tons)
1 n the Eastern Sa,Yann set aff In 1 ate 1961 Find eqrl,y 1962 In construction of
I,hr Abaknn-TNyrhet Ftat 1 ro}id.
'Che hodographs of the earthquaker; and explnsinns were plotted in a scheme of
cuiitrury ud overtalting systems along a nrortie with overall extent of
ribout 3500 km as shnwn i n FiE;. 44 .'I'he enti re system of hodographs was
ttecl in wtth respect to first entries at mutuol points with reduction of ttle
fuci of eurthquakeg to the earth'3 surfac:e. 'I'he accurxcy of this coordination
wa-, :t1.5 s, and r.an be taken as satisfactory considering the great extent
of thc profile, the di:splacement of the epicenters with relation to the
prof'ile and inexact knowledge of the depths of the earthquake foci.
'['he tiodop,raphs of two earthquakes 31.I.1962 In the Vakhsh Mountains on the
souttiwest of the profile, and 28.X.1961 In the northeast direr.tion in the
victriity of Eastern pribaykal'ye were the reference base for the entire
cy:item of observations. A comparison of these two hodographs with the
st!sndard of E. Herrin's commission (1968) shows that in the northeast
direction (from Parnir to Baykal) at distances up to 3000 km the experimental
hodograph is more high-velocity, and in the opposite direction contrariwise,
teus high-velocity.
'Iilus the general tendency of the structure of the upper man-tle in the east ~
is characterized by lower velocities, anci on the west by hip,her velocities.
Fnr epicentral distances that correspond to the position of the interface at
a depth of about 700 km, both hodographs have about the same velocities.
At the srune tirre, waves that are tracked in the second entries and are
associated with this interface are better distinguishable on the eastern
brunch of the hodograph and more weakly distinguishable on the Western
branch, which may indicate a difference in the sharpness of this interface
in the territories of Soviet Middle Asia and the Sayans.
Subsequent entries associated with a 400 km interface are also identically
expressed on both hodographs. For the eastern direction the reflected waves
(or loop sections of the hodographs) are observed in fragments, and for the
western direction they are clearly tracked at a great distance. A com-
, purison of ttie experiment al hodographs of second entries with theory for
ttie interfaces at depths of 400 and 700 km shows disagreement in the times
of c3elrky of the second entries relative to the first. According to the
cxperimental data these delays are shorter.
In the wc::tern direction the velocity of longitudinal waves of the initi,q?
- purt of tlte hodol;raph (earthquake of 28.X.1961) in the first entries has a
vnlue of 7.9 km/s, and does not increase to 8.0 km/s until a distance of the
ordc:r of 1000 km. In the vicinity of 2000 km the velocity increases abruptly
from 8.7 to 10.0 km/s, and then at a distance of 2800 km to 12.6 km/s.
'ne contrury tiodograph (earthquake of 31.I.1972) has considerably different
velocittes. Up to approximately 600 km the velocity of longitudinal waves
94
FOR OFFICIti:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOft ONFICIAL USL c1NLY
Su ubout, 8.0 km/cj. 'I?i(in it lncreases to n.baut 8.3 km/s and gradually risea
ta valuea of 8.4-8, 5 km/a. At about 2200 km the velocity of longitudina].
wuveg increages abrupt7.y frorn 9.6 to 11,0 km/a and at 2800 km reaches values
of 12.6 km/n, -
'I'he contrury-nvertaking hodograph from ttie ee,rthquake of 26.IV.1962 in the
eafitern directidn has a velocity of longitudinal waves of 8.1 km/s in the
first entrien in the intial pnrt, which gradually increases to 8.5 km/s with
u,jump to 9.7 km/s at about 2000 km. 'I'he hodographs of the central part of
the profile (earthquakes 28.III.1962 and 13.IV.1962) are higher-velocity in
the initial part; here the velocity lies in a range of 8.2-8.3 km/s. 7'he
hodograph of road blasting in the territory of the Eastern S qyans is a lower-
velocity hndograph, the velocity in the first entries being 8.0 km/s.
Tn the western direction the velocities of the first entries for the expln-
sions in the initial part of the tiodograph have a value of 8.0 km/s. For the
earttiquake from the territory of the Sayans (13.TV.1962) the velocity is
tiigher, and comes to 8.3 km/s. The more ettsterly earthquake (28.III.1962)
shows n change in velocity of the first entries in the western direction
from $.0-$.1 to $.2-8.3 km/s.
Oi' definite interest is the behavior o1' longitudinal Pg interference waves,
since they give some idca of the average velocity of longitudinal waves in
thc rrust. On the eastern end of the profile their average velocity is close
to 6.0 km/s, and they are triicked to e p,reater extent than on the west. In
the vicinity of the Western 5ayans and Altay the velocity of these waves
increases to 6.2-6.3 km/s. Such a velocity is typical of all of Kazakhstan
as we11. Within the limits of Soviet Middle Asia the velocity of Pg waves
ranges from 6.0 to 6.3 km/s. In comparison with the western direction a
general tendency is observed toward a reduction in the intervals over which
longitudinal interference waves can be tracked.
A general examination of the wave pattern done with the use of data on the
cLynamics of longitudinal waves obtained with the use of the ChISS station
at Talgar gives us a basis for some remarks on trends in the structure of
the crust and mantle in the territories that are crossed by the Pamir-Bqykal
profile. In the preceding chapter a detailed examination was made of the
3pcctral. umplitude curve; of Pn, Pg, Sn and Lg waves (Fig. 18, 20, 22, 23).
The averun,e velocity in the earth's crust in Pribaykal'ye and the Eastern
;;,lyans i:c close to 6.0 km/s. The upper part of the mantle from the underside
ot' ttie crust rutd down to 100-130 km has a weak gradient and is characterized
by velorities from 7.9 to 8.0 km/s. In this connection, difficulties arise '
in evaluutinN the exi;,tence of the low-velocity layer that is usually asso-
ciated with the a,thenosphere. In any event, it can be assumed that if it
occurs, it is very faint.
The interface nt 400 km is evidently upheaved, and should not be deeper than
360 km. This interface is clear.ty expressed and is characterized by a sharp
95
FOR OFFICIhL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'qEt OEFICIAL USE ON1,Y
griidicnt. BPlow thiy interfnce, the velor.ity approaches 10 km/q. A deeper
boundary (700 km division) is located in the 640-660 km intervsl and is not
very clearly expresned.' It probably has a weak grad3ent.
1:n the vicinity of the transiton from the WestPrn Snyans to Altay the crust
ha:i un uvernge velocity of 6, 2-6. 3 km/s. Beneath the undersPe of the crust
tte muntle becomes more high-velocity. An abrupt decllne in amplitudes of
sei:;mic oscilla�tions (Fig. 18a) according to data of the Talgar ChI55 station
in the dtctance ranWe f'rom 700 to 1700 km, and the presetice of a small
3ncrea:ie in cunplitude3 at a distance of about 1000 km show the existence of
ari n:.thenospheric layrr ef reduced velocity, and beyond it an interface.
A cl.aar boundary is dintinguished at a depth of the order of 360 km. Asso-
r,tated wtt}i this boundary is asharp emplitude sp3ke observed at a distance
of 1$00 km. 'rhe deeper 700 km interface is much more weakly expressed.
[n ttiis territory it is at a depth of 650-700 km.
A comparison of ma.ntle velocity data for depths ranging from 350 to 630 km
in the territories of Pribaykal'ye, Kazakhstan and A].tay shows that the
velocities in this interval are hiBher in the east than in the west: 10 and
9.6 k:n/s respectively. Thus in the ;^astern section of the proffle the lower
veloc:ity in the upper part of the cross section is compensated by higher-
velocity low levels.
The presence of an amplitude spike on the amplitude curves of the Northeast
direction (Fig. 18a, b) at distances of the order of 450 km, and u branch
that can be tracked on the hodograph at the same distances of a weak reflected
wave with chort delay with respect to the first entries indicate an inhomo-
genelty in the topmost levels of the mantle at depths of the order of
80-90 km. It can be assumed that there is a boundary of low rigidity at
this deptti. Territori.ally, this inhomogeneity is confiend to the section of
the profile between Lake Balkhash and the boundary of the Northern Tyan'-Shan'.
In the environs of Northern Tyan'-Shan' the earth's crust has an average
velocity close to that observed in Kazakhstan 6.3 km/s. Further toward
the southwest the average velocity in the crust decreases to 6.0 km/s.
In the first entries, deep longitudinal waves have a velocity of 8.1-8.3 km/s.
Ttiese velocities are close to those observed for the territory of Kazakhstan;
however, in contrast to the northeast direction (recalling that the azimuth
fs read from Talgar station), the amplitude curves are situated considerably
higher with respect to level, and they have no characteristic spikes. This
circumstance, as we11. as the absenr.e of pronounced waves or loops in the
hodagraph show that the low-velocity channel in the Middle-Asiatic direction
is poorly expressed.
'Ctie poor 3liow of the 20-degree spike on the curves ZPACIB us to assume that
r,he deep-level boundary close to 400 km is a1Ro less p.ronounced. The 700 km
boundary in this rf:gion can be much better tracked on the amplitude and
kinematic curves of refleated waves (Fig. 18e, d).
96
FOR OFFICII,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F()It (1FFI(; I AI. II!;I? (1NI.Y
m
A much tigher-velncity cros:3 section is tyr,ical of the southern part of the
profile the region of the Pfunir-Nindu Kush earthquakes (Lukk, Nersesov,
1965). flere thr_ main difference betwcen the deep-level cross section and
those examined previousl.y is observed in the upper lev31s of the mantle
to a depth of 400 km. 'I'he high-ve7.ncity cross section of the mantle to the
south of Pnmir is confirmed by research of Indian aeismologists as Weii.
In evalusttng the peculiaritie3 of ttie structure oi' the mantle of Soviet
Middle Asia as a whole we cn.n note that it is apparently transitional between
the high-velocity upper mantle of Pamir (with poorly expressed deep-level
interfaces) an8 the muntle of Kaze.khstan. It is quite significant that the
asthenospheric channel in Soviet Middle Asia is much more poorly expressed
than in the territory of I:ustern Kazakh9tan. It is a triking fact that the
presence of the 1ow-velocity ctiannel in the asthennsphere of Kazakhstan is
ar..companied by somewhat of an increase in velocity at a relative],y greater
dept}i (down to 400 km), w}ii1e in Soviet Middle Asia the absence and weak
cxpre:ssion of thin c}iannel is associated with somewhat of a reduction in
vclocity at these depth:,.
Unfortunately, the available material does not permit detection of clear
boundaries of division of the upper mantle among the individual noted blocks.
We can only present a few considerations. The boundary between the mountai.n-
ous Pcunir block and the central part of Soviet Midsi].e Asia runs in the zone
of the Darvaz-Karakul' break r.one (Vinnik, Lukk, 1975)� 7cie transition from
the Middle Asia type of mantle to the Kazakhstan type is noted fn the region
of t}ie northern spurs of the Trans-Ili Mountains on the border of the Ili
Ilusin. Ttie boundary betwepn the Sayans proper and Alta}r is quite pronounced.
in deep-level structure, these two regions differ considerably. Besides
the cnateriuls that we have used, confirmation of the existence of this
boundary is to be found in the results of study of Lg waves (Nersesov,
Rautian, 1964). The pattern of ctiange in velocities in the upper mantle for
the territories crossed by the profile is shown in Fig. 44.
The principles governing the relation between the spectral emplitude curves
and kinematic curves lead to a number of considerations of a more general
nature for thc eastern and southern directions (see Fig. 18b, c). In the
eastern direction (western territories of China) the spectral curves have a
poorly expressed amplitude spike associated with a depth of 360-400 km,
Apparently this boundary is fuzzy, which is also confirmed by the relative
spectral width of this spike: the low-frequency components of the ChISS
spectrum come close together. The spike at 25� is located at a comparatively
great distance, and is more pronounced on the high-frequency components,
which indicates that it is associated with a rather sharp boundary at depths
oF uround 700 km.
A comParison of the eastern and northeastern direcLions shows that con-
sidernble changer; take place in the structure oF the upper mantle to the
south of the proPile: the low-velocity asthenospheric channel dies out;
ttie 1100 km boundary loses clarity and rigidity, and its depth apparently
97
FOR OFFICI/,L USE ONLY
IL
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOEt dFCICIAL U5E ONLY
:4
i.ncremucs; the s;hurpners of t,he boundary of' the upper mantle at a depth of
700 Scrn increasei3 nomewhat, the boundary itself also becoming somewhat deeper
cu; comnurecl with the northeast. New data ap,ree with the particu].ars of the
nLruct,ure af the eustei�n purt of the given territory found previously as a
result; ot' procescing of the mnterimls of observat'ions on the pamir-Baykal
profi.Le ( tiugqycvrkiy et a,l., 1971)�
It can be e,ssumei that the enumerated considerable changes in the structure
of the mantle occur between Altey and dort;hern Dzhungaria both within the
borders of the USSN nnd in China.
Tn the southern direction (see Fig. 18c) oriented from Talgar mainly toward
'Pibet und the northern provinces of India, the deep-1eve1 cros5 section
rilso changes considerably. The low-veloc3ty channel again begins to be
trncked, although wecilcly, and the presence of a small spike in amplitudes at
distances of about 1500 km indicates either that its depth increases to anproximately 200 km, nr that the nature of the boundary itself changes.
7'he I00 km boundary in this region is weakly expressed. Tt is characterized t~y narrowing of the spectrum of the reflected wave, which is evidence of a
strong gradient. In addition, the lower boundary of the upper mantle is
much more clearly expressed than for previously examined directions. The
two-hiunped nature of the amplitude curve is an indication of the complexity
o.C this t,oundary, and the configuration of the first maximum confirms a
tendency to increase. In comparing the southern direction with the southwest
direction considered above, one can note an increase in expression of the
lower boundary of the mantle, as we11 as less rigidity, which shows up in
narrowing of the spectrum of waves reflected from it.
The use of the spectral curves of lotigitudinal waves obtained at Garm station
und at Temporary station situated in Northeastern Kazakhstan to study
genera]. patterns of structure provides additional information on the structure
of the mantle within the limits of the southern and southwestern sections,
and also on the eastern continuation of the profile. All available data
are shown in the diagram on Fig. 45.
7'he shading on the diagram shows regions of development of mountain struc-
tures. A number of interesting conclusions derived from the given data.
The prairie regions of Kazakhstan are characterized by a higher-velocity
cru:;t and the presence of a low-velocity channel in the upper part of the
mnntle. /11so striking is the circumstance that an asthenospheric channel can
bc noteci both in the vicinity of the Tarim plate and in the western parts of ~
.4 a1
cd u~ cd t0. cdtn u
ROO2fo-H M
~M g R R G" iG~i
w-,md-+Ma
113
FOR OFFICII,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Fnk oN�rtccnl, usr, ONLY
Let; usi exuminc the dependencen nf the second dif'i'erences ddm on the distance
betwttn stations. mn do thin, we refer to the praph ghown in F3g. 50. Tt is
cleEir from thig graph the,t the necond differencee ddm increase w3th digtanee
between gtntions, reachinp, a mttximum in thp 600-1000 km interval. On the
bnnin nf the analnp,y between t,he funCtion of gecond differenc:es and the
utructure functidn, the pohition of this maximiun on the A3 axis can be
intErpreted an half the 3patial dimension of the character~gtic inhomo-
gcncitieg, which ngrees with the neact minimum of the function in the
1.500-2000 km interva].. 7'he seCOnd maximum of the given graph is obgerved
in ttie vtctnity of Aij = 4000-5000 km.
3. Amplitude fluctuations on the NorCh Tyan'-Shan' station group
Fot� :;tatioris of the North 7'yan'-5han' network (Fig. 51) the second dSfferences
ot' logarithms of the P wave were considered. 7'he processing technique here
iA

A6 A~
�f
�r
� r
v~~ rsrM
F'ip,. 51. Diagram
of location of
station3 of the
IJorth Tyan'-Shan'
group
wsu somewhat different in connection with the fact that
in the case of short distanr.es between stations there
wag no need to introduce a correction into the epi-
central diatance and average values of the logarithm of
the amplitude for each earthquake were determined as the
arithmetical means, whil.e the fluctuations were determined
as the deviations from these means.
In this we4Y the errors associated with individual
peculiarities of separate earthquakes were sharply
reduced.
7'he mean valuea of fluctuations in amplitudes are
swnmarized in Table 21. The experimental data were
selected with respect to the qua].ity and nature of the
P wave recording; obligatory requirements Were clarity
of the first entry, pulse shape of the wave and a con-
siderable, at least tenfold, excess in amplitude over the
microseismic background. Altogether, 89 earthquakes
were selected with M 2 5�75.
`I'tie individual error of ineasurements of the maximwn amplitude of oscillations
in our procedure is about 0.04 log unit. Each value in the table is the
result of averaging of the data of 5-15 earthquakes. Thanks to this, the
relative error is reduced by a factor of rn (n is the number of ineasurements),
und hence the error of each tabulated value becomes no greater than 0.02.
Get us note that this is considerably less than the variations e.
Thc iiverage station values �dlg A>e>g lie in the interval from -0.06 to 0.04,
and the average values wi.tli respect to epicentral regions �d lg A>e>a lie in
ttue interval from -0.13 to 0.06.
in rtcr.ordnnce with the previously introduced definition, the quantities
,:>, should be equal to zero since for each individual earthquake
g The difference of this quantity from zero that is observed in
114
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
COIt UFF1CrA1. USL ONLY
TABLE 21
Average vnlueg of the fluctuations of ~unplitudeg d 1g A of a P wave
' on Nnrth 'I'yun'-Shatt' stationa
, t1,
311MYlMf~1~MN14
CwAtY~MtM1~ ltIMYM
(9),
N
^
p~~ONY
~
11
11 >
7 T
71
~
1
~
V
2 a KsrliiN -p,il{-u,IN 0,15-0,09
3 MyTY-NUMANAU~1N~.1
I11~-U
UU 0
I7 U
Q~
0.01
I9
0
0,01--0,43-0,01-0,02 0,00
I2
3
,
,
,
i
.
4
~AM, UaunNne p~~M._U,11-U,ZO 0,01
.
0,01
0,
-0.0
0,06 0,03
0
11 0,I2-0
03 -0
09
0.I3
0
13
~
,
,
,
.
ntlNCKNe nctpnu .~11,113 II,U7 O,tS~--tl,Uh...0,01
0.10 0.04 0.08 0,64
O,fI
NAONS INM -.0,IS-U,Zi~-11,U1-(1,03
7~II~IOII~NIR4l~1N111~~
U
~
U

0,16-0,03 0,18 0,17 0
0,14
~
,
j~
~
~-11~113 0.14
p
0,14
~~20-00 0,03-0,03
0,13
~
U lAlINAMII, rpkNAAIb ll, 11 ~-II,1U (1,17 U,1B~,~
r~ADNxnu
3
0,I2 -0,03 -0,10 0.~
0.14
ewe~
(tAItA>,>, -0,118 -0,13 n,uS 0,04 00 0,03 0,01 0
11,111 11.10 11,10 U,W 0,10 0,14 0,00 0,09
KEY: 1--Epicentral regions 6--:[ndonesia
2--Kodi ak I aland 7--Etirnpe and North Afri ca
3--A].eutians, Komandors B--Iceland, Greenland, Baffin Island
4--Kurils, Sea of Okhotsk 9--Seiymic stations
5--Japanese Islands
reality is probably due to the fact that individual averaging operations were
done both with respect to insufficient data and with respect to appreciably
different stations. The confidence interval for each average is equal to o/Fn
where n i3 the number of terms. Thus the values of averages with respect to
epicentral regions do not exceed the confidence interval, and their values are
not ntatistically significant. At the arune time, the difference of inean
square deviations is significant. In fact a exceeds e>gs as implied by our method of determining fluctu-
ations, should be cloge to zero, and therefore slight corrections have been
made in the eolwnng of Tnble 26 go that �600g =0.
Taat.E 26
Averuge valupa of deviation (s) from the straight-line hodograph e
of the p wave on stationa of the North Tyan'�-Shan' group
o~~..... h.....
s.n..+~.~
~
K+1m
-0~1~ 0,01-0,11-1~~1 o,st O,OQ--O,tl
2,00
I.W
~l~~ Kawyq~
0~3l-0~~-0~86 O~~Y f.1~ 0~1~-i~ri
f~de
1~3{
xMAw o,.hM W"
1,46. o.m 0,42-1.e6
o.b
1,13
p~o~esM ~~fer
-0,13-O'Y-1.28-4,16 0,U-0,16 0,01
f.77
0.00
-O4tl-01l1-l,f0 I,11-0.4l-0.01-2,35
=.46
I.91
M~ � Ca~~ A*
0,11 O,p-f,61 f~1l-0~70-0,1l-0,l0-0.1f
I.W

t9 e. r.""Wa�,
-1,74 1,0 0,07-0,28
o.u
1,31
400a.
o,a 0.07-0.78 0,61 0,07 0,28-1,02
1,3e
te�)s
0,26 0.0 t1b I,36 0.97 0.59 1,18
1.18
nEY: 1--Seismic stgtions 6--Japanese Islands
?.--Epicentral regions 7--Indonesia
3--Kodiak Island $--Europe and North Africa
Is--Aleutians, Komandors 9--Iceland, Greenland, Baffin Island
S--Kurils, 5ea of Okhotsk
122
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOR C)FN�ICiAL 1I51s t)NLY
Attriking peeulinrity di the fiuctuationo iu the Wide range of variations
in , from -2.72 to *3.11, eonsiderable differences being noted in ico-
lated caaps on the enme station fdr adjacent regions (for inetance for
station 7and regions iV and V the dit'f'erenee in 3�21 a) The differences
nf average sbation flustuationa e>a are congiderable; from -1.02 tio
+1.38. TheOe valueg characterize the inhamogeneitiy of the upper parti of the
rronn nection benrntih thc qeinmic otatinnn. If eongideratir,n is talten of
the fQCt that ail stationa are located on cryntal.line roeks, and inhomdgeneity
oceurs only rrithin the limits of the earth'o crust, ahich is about 45 km thirk,
bhe contraot in veloeity fluetugtions should be about 10%, A prreciable
difi'erences ot' the mear square deviatiions oea (from 0.25 to 1.55) eharactierize
the strong contrast between inhomogeneities of the crust and thoae of the
mantle, the letter not being part of the upper croee secb3on. The presence
of thege inhomogeneibies shoWg up aith greatest contragt in the flata of
station 4. If it in asgwned that the characteriatic vertical and horizontal
gize of thege inhomogeneitiea in the same, and equal to approximately 100 km,
then the fluctuetiong of velocity mugt average 10-12x, and the maximum devi-
ationa up Lo 15-20%. dn the other hand, 3f such anomalies are impossible,
we mugt assume that their vertical dimenaion is appreciably greater than tihe
horizontal. .
Shown in Fig. Sg are the seeond differences o2' the travel time ddtij.
~
I
~ I
I
JW /N tN /M
I ~ I
1 ~
I
I
tIN aIY AIMI
ey,PA
Fig. 59. Average values of second differences of travel times ddt for
seisntic atations (a) and epicentral distances (b) from observations on the
North Tyan'-Shan' group. See Fig. 52 for explanation
The structure of the grapha and the behgvior of the average ddt are very
gimilar to What wss notefl in Fig. 52 for the second differences of the
loKarithrro of amplitudeg. The elements of the graph of ddt have been shifted
to the let't with respect to the data for dblg A. The interval aF correlation
of fluctuationa of travel time is less than the fluctuations of emplitudes,
nnd Lfierefore the inflection a.nd decline of the graph of ddt thet correspond
tn the p,raph of ddlg A have been pushed off to the left out of the figure.
Let us note that in wave propagation in media with random inhomogeneities
the intervals of correlAtion of fluctuations of amplitudes must be greater
thun the fluctuations in travel times.
123
FOR OFFICIl,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOR OFFICIAL U5E ONLY

t
�
~

Fig. 60. Correlation ot' fluctuatidns in
P

~ ~
~

the amplitudes nnd trave1 times of P Waves
' '
�
�
.
from data of the North Tyan'-Bhanl station

~
group

� : .
. .
Simtlgrity in the behavior of grapts of dd 1g A end ddt ig aleo a rather
convincing gign of statistical significance of the results.
'I'he correletion of fluctuationg in amplitudes gnd travel times is 111ustrated
by Fig. 60 plotted !'rom data of the North Tynn'-Shan' group of stationa.
Wp ran readily aee that there ia practically no correlation. Th1s reaulti is
important for judging the nature ofAnhomogeneities of tihe upper part of the
cross sectian Lhat determine the average atation anomalies of ampl3tude and
travel time: vElocity inhomogeneities of the upper part of the medium that
cover pointg of observation should generate weakly correlated funetions
d 1g A and dt thut are characterized by a correlation coefficient of 0.4 in
the case of 1oa-contrast anoma].ies of velocity (Chernov, 1975). Thus the
lack of correlation can be interpreted as compnrative homogeneity of the
very top of the cross section, about a 10 km 1ayer. 7'hen the average station
anomalieg ahould be considered as the result of inhomogeneities that are
aitugted more deeply, right down to the underside of the earth's crust.
Anomalies of amplitudee may be generated both by velocity inhomogeneities
and by the inhomogeneities of absorbing properties. Such inhomogeneities
r+ith dimensions of 50-100 km and average Q of about 300 should be charac-
terized by variutions of Q rat,ging from 150 to 400.
Chapter 2. Investigation of Fluctuations in the Shape of a P Wave Recording
1. Study of the ehape of the recording of a P wave on stations situated in
different territoriea of the Soviet Union
This study is based on calculating the coefficients of correlation between
seismic nip,nals registered by different seismic stations. The initial
sections of the seismogrems that contained the recording of the P wave Were
quantized. 7'he seismograms that were used to study the spatial structure of
the P wave Were obtained on 55 seismic stations situated in different regions
of Ehe US5H, and in part included data used in stuc'~ying the fluctuations of
amplitudes and travel time described in the preceding chapter.
In selectinR the material, consideration was given to the folloWing peculiari-
tics of seinmlc recordings: the pulse shape of the initial part of the
recording, i. e. those sources that had short-term action and were
124
FOR OFFICIAL USE ONLY
t
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FtJR OFFIGIAL U5E ONLY
Gh~ra~tr,rizrc9 by tiimlllir.ity 01' LIiO nhuls rit' t.lie t'ircit aive., nnd ttn ramluLrn-
i,tvuly Wlclc-bnttd upcctrun; 11ul't'leirnu Intrnsity nP the recording en
three ntationn nn a minimum; mnximum peak-td-peak amplitude of tihe f irgt
pulse on the saismngram of nt 1eant 10 mm; ciear entries.
7'hree-component recordingg of yl earthquakes with magnitude M >5 aere uspci
with epicentere removed by a di8tance of 30-801 from the otationa. mhe foCi
of the earthquakes aere situated in different regions of the earth; in the
western part of the PaCific dcean be1t, in Indonesia, in the Mediterranean
5ea, in the North At].nntic and ih ttie Aratic ncegn.
7'he X-COmponent cdrresponding to the direcbidn touard the source Was pre-
determinefl from the hnrixnntc;l cnmpanents ef each recording N-S and W-E.
In thia way tihe vertical and rgdigl X-cnmpdnent n!' the P wave recording
Were nubjected to nubgequent corre]..ation analysis.
The ghgpe of Che reeording is analyzed by cdmparing recordinge of the P Wave
fram nne nource on different gtations. A quantitative measure of the
degree of reaemblance of shape in the Correlation caefffcient Kij of the
recordingu of the i-th and J-th seigmic stationg deterrnined fram the forniula
p,Ii) P pI dr
Kry ~Jp,2(1Jd1J p/JtJd1 u8 �
Since Kij = Kji the entire eet of correlution coeff3cients for n earthquakes
can be representefl as a symmetric matrix containing n -1 roWS and columns.
7'he va1ue of the correlation cnefficient fbr a pair of similar functions
sucfi us the recardings of remote earthquglceg is n maximum with the proper
coincidence of the firat entries. In reality, there is no firm certainty
thut the beginning of coding of the recording coincides with the first entry.
Yossible errors reach a quarter of the predominant period.
In order to keep the errors associated with uncertain determination of the
beginning of readout from "spoiling" the maximum value of Kij, the signals
Were "synchronized" by calculating the correlation coefficients for several
relutive locations. The maximum shift of one signal relative to another was
cquo1 to t0.4 s, which enabled calculation of nine values of the correlation
coefficient for esch pair of recordings in the case of a 0.1 s displaeement
ctel, cqual to the quantization step.
iri nttier aords, in the "toynchronization" process the section of the corre-
lation of tWO signals in the vicinity of their first arrivals is calculated.
After thnt, only the maximum value of the flanetion on this section Was used.
Cince there were comparatively few data of ,joint observations on individual
puir:s of otatfons, they Were processed all together rather than being
sc:parrited by epicentral reqiona. It s+as established as a result of such an
unulysis that ctribtlity of the shape of the vertical component is higher on
125
FOR OFFICIML USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOK UFFIGIAL USE tlNLY
t,he averfsgr thnn the rt,dinl romponent. 'I'his ghdwn up in the fact thtLt high
vcLluen of the correlation coefPieient K s d.7 appear for thp vertical com-
ponent compartLtively more often than far the horizontal component.
of inteveat id an exgmination dr thh correlgtinnn both within groups of
scismic ntgtiona nnd betueen neparate groups. tt hfati been established that
the t;rdupa et' atntiang oC the BuropecLn ptLrt of the US9R and Soviet Middl,e
Asiu are ehgrgeterized by a relatively high 1eve1 a� interr~al sorrelatian.
mhc:ne groupg arp unequnlly related wi th each other : the Middle Asin stationti
ntiow gdnd correlation aith Northern CauCagus attLtiana, while thoge of the
E;uropean part of the nation shoa good correlation With both of these groups.
`I'he Ya1cut stationn correlate weii with ali regional groups of stat3ons. It is
nnt pdgsiblp to trace the nature oi' the correlation for other groups of
stutinnn due to insufficient data.
7'hus relatively hip,h correlation of the vertical component of the P Wave
rer.ording is noted both uithin separate groupa (European, Mi8d1e Asian)
And between remotely spaced stiationa. At the same time, in mar~y instances
Weak correlation ig observed between eomparatively cloee stations. With
regard to the gpatinl correlation of the X-component it can be gtated that
the nnture of the correlation dependenceg in basica].ly tbe same gs for the
vertical component. At the same time, the vertical components correl.ate
with each other more strongly on the Who1e than do the radiat components.
7'he distortion of signal $hape that ehous up in impairment of itg correlation
Wtth nnothe.r, reference aigngl, mqy be attributed to several causeg. In
pnrticular to scattering of seiemic energy When the r+ave passes through a
regton that contains inhonageneities that are contrasty and small compared
with a WavelenEth, as well as to the distorting effect of the contrasty
relie!' of buried and surface boundaries.
ttigti correlation of the ghape of a p Wave mey show up et distances of several
thousand kilometres betWeen stations if the deep interior of the earth intro-
duces xeak distortions, i. e. if it is comparativel,y uniform; moreover, it
is important that the signals radiated by the source in different directions
have n similer shape, even though they mqy differ appreciably in amplitude.
A reduction of correlation is usually caused by the screening action of
sections of the medium in Which minor inhomogeneities are strongly developed.
Ime dimensions of these sections amount to several hundred kilometres, While
the average distances between them are considerably greater thousands of
kilometres.
Huvlng made a detailed analysis of the spatial structure of the P wave, it
i:; of interest to evaluate the principal scale eFfects of fluctuations of
wiiveshape and to relate them to the inhomogeneity of the structure of the
c,rrth's interior.
126
FOR OFFICII,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
li
i
,
FOR OFFICtAL U9L ONLY
2. inveatigation of detailed seruceure nE the shape of the P wave
'i?� detailed ntruCture of the lonp,itudinnl wave wan etudieA from the datia of
nbgervatiiong on dense nygtemg of neigmic atationa; North Tyan'-Shan', North
Kntalchatan, European, Centra1 'i'yan'-5han', Qarm and Zeyok. R'heee groups of
ntntions operated at different times, the number of stations and their
relative iocation in the group alon changed, attd therefore the study was
baneri nn initial dgta differing in d.,~tvi1. 'I'he dnta of only the vertical Gompdnent nr the recording are considered here.
mhe mcthod of data preparatinn and cmlculations of the correlation coef-
t'icienta in degcribed in the preceding section.
`Che North Tyan'-Shan' Group of Seatian~. The recordings of 89 earthqualtes
were uged, the same oneg that Were uged tio study the structure of fluctuationa
of tunplitudeg and trave1 time. A matrix of correlabion coefficients KiA Was
calculated for each earthqualce; the vg1ues of the matrix aere rather high
nn the average, exeeeding a level of 0.5. 7'he matrices of average valueg of
the correlaLidn coefficientg ,>Si are presented in Tab1e 28. mnr3LE 27
Averaped mgtricen of the coefficients of correlation of Lhe shape
of the inittal ghnpe of the recording from materials
of ntations of North 7'yan'-Shsn'
i s 1
1
L$
f
A I
I
9
f I
1
f 1
t
1I
7 1 I 1 I 1
a
1
1
I
_
U. ;1 11,61 o.6S
11,88 O.LI
f1,!~
0,65
0,53
11.56
0.14
O.M
o.lt2
U,149
1
4
n,54 n,A:
O.CS
o,dS A,6l1 0.60
11,4i 0,40 0.60
O,:iA
11,74
O,iO
0.65
f
3
0,S7 0.32 b,S,1 0.00
0.:~3 0,41 0,11
0,56
0,311
n.3i
0.410
n,+T
41.11
4
0,11
0,70
U.76
0,96
0.70
O.OfI
0,32
0.69
11,71
3
S
0 ,S4 0,62 0,75
0,65 O.S7
0.83
0.{3
0.68
0.33
9
4
0,54 O,S3
0,141
A,:is
0,62
~1.3
O5S6
n.ti1
A.:~
S
6
t 1~~ Ruw
0.
v,3b
0.64
n,77
0,71
S
6
0,67
~ 2)Awnr. Kwa~
0.67
0~61
O.tu
0-75
3
A~ 3) KMw.
11,67
O,AI
0,3~
11.31
1~.82
7
O,bS
7
0,63
7
020eew nM
u,Sd
~ 1 1
i
1
t
1
3 f
1 1 ~
t
1
! ~
~ !
~
t
~
1
2
0,44 0,~ 0,63
u,N 0,S1
11,61
9,64
0.00
9.0
0,51
O,t8
O,b:i
0.51
1
I
O,:r4 O,S3
0,~
O,SO 0151 O,S4
0,50 0,43 0,37
O,:rB
0,61
0.11
0,38
1
3
0,46 0,33 A.N O,S11
0.3i 0,3! 0.45
0.48
10.46
O,tt
0.56
0.'!9
0,45
3

0,13
0.79
0,67
0.76
0.U
0.78
0.48
U,66
0,62
3
4
0,6$ U,Si 0,67
0.53 O,SY
O,QO
O,i~
0,63
0.31
3
t
U,37 O,33
0.58
u,76
n.37
0.0
0.510
0.39
A,;~6
S
7
~~t ~ IMwn.~ ~nrw~
0,70
0,07
0.71
0.71
0;~
S
5~
O.AS
0 ,57
0,67
O.Sa
SS
O,
9
6(6) dwN~
tr,sQ
n.9t
0~31
u.il
11,36
~
O,
t2
7
0.61
I
! ~ 1
1
1
t
1
I
I
i
n'~ u:sq ~;s �6 i
o:~
o:si
ii:p
u;ii
i{EY:
1--Kodialt Islan
4--Ja
panese
Islands
3
4
u,eu
o,n
u,a
0.00
0.67
o,a
0.44
o.ee
o,U
2--Aleutains
,
S--Indonesia
~ ~ 7
s
�'+t
�o:u
�o'%o
Komandors
6-��Europe, North Africe
7
0:74
3--Kurils, Sea
7--Iceland.
Greenland,
of Okhotsk Baffin Island
127
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOR OFFtCtAL U3E ONLY
'CA1ILL PfI
Avernge vniuen dt' the coefficientg nf Gorrelation of the shape
ai' a P Wave reaording far gtatiOt19 of Northern Tyan'-Shan' _
~
1
!
1
1
/
1
f
1
1
o.. ttw�.
n,ee
o,eo
o,e~
o,e~
0,60
n,ea
o,du
u,ee
0,e1
Amrri. Kom,sm"
o,w
o,ao
0,e4
0,e1
u,M
o,es
0,42
a,U
0,02
K1NAm. Oexea reN
O,M
0,60
0,41
O,SB
0,6A
0,S1
O,63
0,41
tl,u
A~e~e~r~ ect~ew
O,bs
n,6A
0,61
0,17
0,70
O,bd
U,M
O,IIA
0,44
~
o,U
o,U
0,61
0,53
0,57
0,e1
0,63
o,et
0,e7
EMMA. eMo�+N Mo��& o,u
0,50
o,es
0,53
o,an
o,ee
oX
o,M
uX
h
9~
m" a1,
o
'
0,e7
o,eo
o,es
01e1
o,eo
n,eo
n,ae
0.61
o,su
~
a
n
"
( ldCp;Am
O,bS
O,bl
0,lU
0167
0,63
O,tll
U,60
d,AO
KEY: 1--Epicentral regions 6--Japanese ialands
2--Seismic atations 7--Indonesia
3--Kodinlc igland B--Europe, North Afr3ca
4--AleuEiana, Komandors 9--lceland, dreenland, Baffin Island
S--Kurila, Sea of Okhotsk 10--Average =
7'he average va.7.ues of the coefficients calculated from the roWS of the table
are the station characteristics, ahile those calculated from the columns are
the averaged cheracteriatics of the given earthquake sample for a specif3c
epicentral region. The average values of the coefficientis of correlation of
the shape of the recording of a P Wsve for stations of Northern Tyan'-Shan'
vary 2'rom 0.43 to 0.70. 3ince each val.ue is found by averaging at least 50
tndividual patimates, differences greater than 0.02 should be tiaken es sig-
ni t'i cant .
For eaeh seismic station the values of the coefficients are varied as s
funetion of the azimuth to the epicentral region, the amount of the varigtion
reaching 0.2. For each of the seven regions the values of the coefficients
change little as a function of the position of the recording atation, end
vsriations remain within limita of 0.1. This meana that the change in
spatial stability of the ahgpP of the P wave ia related primarily to the
position of the epicentral region, and depends comparetively little on the
individual peculiarities of the seismic atationa. Let us note that the
differenceu in frequency responses of the recording chsnnels should shoW up
in variations of the average values of the coefficienta With respect to
different pairs of stations for eepara.te epicentreS regions. The compara-
tively slight valuea of these variations show that the effect of the differ-
ence in spatial atability of the P wave from diPferent epicentral regions
is reliably detected on the background of varietions thst mey be possibly
eagociated with some non-identity of equipment characteristics.
The appreciable influence that the region of location oP the source has on
the gpatial atructure of the ahape of the longitudinal wave shoWS up in
comparatively vesk similarity of the correlation matrices for diPferent '
regions.
i 128 ~
FOR OFFICItiL U3E ONLY
~ .
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOk UFFtCIAL USL dNLY
TABLE 29
Correlatinn of mntricen of the choffieiehtg of correlation
of ttie nhape of the reCarding of a Pwave tar a11 epicentral reginns
~ M~'
1
11
111
IY
V
T VI
I VII
t !
O,fA 3-0109 tl,81
0130
0167
O,Ond
It
1 -11,9U 0,61
0,40
0,14
0,21
111
IV
f 0,002
0,01
0,02
-0,1
v
, 1
tl,/7
0,36
0,48
VI
1
0,44
0,26
Vit
1
0,11
I
9ftHg@HTpAJ1bHNe paAoHbtn Epicentral regions
Note: Roman numerals denote i--Kodialc inl.and
II--A1eutians, Komandors
III--Kuri1e, 5ea of Okhotak
IV--Japanege Islandg
V--Indonegi a
Vi--Europe, North Africa
VII--Iceland, Oreenland, Baffin Is1and
To find a quantitative measure of similarity of matrfces of e, ae corre-
lgted them With each other, getting the geventh-order matrix ahoWn in
mable 29. In correlating the matricea, they xere a1.1 converted to numerical
series, the valuea of each seriea being referred to a zero average. 71wenty-
one correlation coefi'icients Were calculated, of Which three are negative, but
have low nbsolute val.uea. The positive vglUes of the coefficients range in
absolute vaJ.ue from 0.02 to 0.57.
mtius each value in Table 29 Was calculated from two series containing 21
terms apiece; the mean square error of the result close to zero ia t0.3,
and on the 0.4 level from 0.1 to 0.5. Therefore values exceeding +0.4
and -0.3 have a considerable difference from the background. A comparatively
htp,h correlation is observed in the initiel matrices for reuate territories _
of Alaska and the Mediterranean. No dependence of the degree of correlation
on distance between epicentral regions is observed.
Let us consider the wQyr that the rorrelatian coefficient e depends on the
distance Aii between pairn of seismic stations i and j and on their position
relative to the source. In doing this we separate a11 data into tao parts
corresponding to longitudinal and transverse correlations. In the firat case
the source is in line with the stations, and in the second in the direction
perpendiculgr to the line joining Lhe gtations.
Shorn in Fig. 61 $re data obtained in our observations on North Tyan'-Shan'
stutions and also in observations by G. M. Tsibul'chik in the Sayans (1968)
nnd by V. Din in North America (1965). All these data were obtained by
129
FOR OF'FICIAL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
N'Ott 0FFIC iAL U5l; ONLY
..,z . ~ .w
. . . J
,t,` . ~ . d
~ . ~..-E a.~~1~ . ~1.1 L...:z..
/ t J v,r ~ /0 t ~ v r f!00 t J M00
dt/~NN
TFig, 61. Coefficienbg of correlation of the ghape di the recording as a
func:tinn of the distanCe between stationg: 1--correlations with regpect to '
c11 directions; 2--longitudinal correlation; 3--transverse correl.ation;
4--our dstu from stations of the North Tyan'-3han' group; 5--data of V. Din
4965); 6--data of G. M. Tsibul'chik (1968)
ahort-period vertical inatruments. 7'hp observationg 3n the Sayans are
ctiaracterized by short diatances between stations from 1 to 30 km. The
duta on correlation are not differentiated with respect to the position of
stations relative to the direction to the aource. The lPVe1 of correlation
t'a11.s off uniformly with increasing Aij from 0.85 to 0.70.
V. bin's dsta relate to a wide range of distances from 2 to 400 km. The
giaphs of K(Dij) have a characteristic shape: as A3j increases one first
obnerves a drop, then a riae and once again a drop to an asymptotic value
oi' Ki~1 �0.5 When Aij= 400 km. Din's curve is situated considerably lower
than Tsibul'chik's as a consequence of the difference in the length of the
uignuls being correlated: 20 s for V. Din, and 3 s for G. M. Tsibul'chik.
It is important to note that the graph of longitudinal correlation is as 3t
Were the graph of trgnsverse correlation stretched out in the Aij scale.
7'he same characteriatic features can be seen as well on our graphs in the
interval of Aij = 70-400 km. At Aij = 120-190 km a maximum K is observed
that can be identified with the corresponding maximum of the riin curve:
the minimum of transverse correlation is at Ai~i _70 km. Corresponding
characteristics af an element of the curve of Iongitudinal correlation are
noted at Ai~i ? 340 km (maximum) and at Ai~i = 125 km (minimum). Thus our data
stiow general similarity to those of V. Din as expressed in the identical
gtructure of the graphs of K(Ai~) and their close levels at extremum val.ues.
This peculiarity of the spati al change of shape of the longitudinal wave is
probably a consequence of similarity in the structure of inhomogeneities of
the earth in different regions. Let us examine this problem in more detail.
7'tie differences in the intervals of longitudinal and transverse correlation .
Indicate that the horizontal inhomogeneities that distort the waveshape lie
nt nucti depttis that their upper edges are below the open surface. In this
connection, the "shadows" of inhomogeneities that are projected by seismic
rey3 onto the surface of the earth are stretched out in the source-
130
FOR OFFICIAL U.SE ONL'Y
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
coti orFtctnL us: oNt,Y
inhdmoNeneity direction. Let uz; note that in the given ca.ge when remote
eurthquakes are uued we sec longitudinal correlai;ion with respect to snme
intermediate direction cldser to the wnve front rather than in pure form,
i. e. along the raya. Therefore the true longitudinal correlation of the
shupe must extend to sti11 greater distances.
'I'hc interval of traneverse cnrrelatinn of anplitude and phase of a plarie wave
iu clour to thc chur�aeterlstic: dimension of anomalies of the field of veloci-
t,lcn of r.luiitic wtLVe:s (Chernnv, 1958). 'I'h:ls :;tatement can be applied to
wuvC:itiape a:c weii nlnr.e ttc lutt;er is aetermined by the ratio of amplitudes
of' nepurate extretna and by their position in time from the Initial part of
Lhe seis-mic recarding.
Let us define the intervul of correlation of shape as the distance at which
the first minimum in observed on the curve of K(Aij). Taking into consider-
ation thut the wave front in the repion of exit to seismic stations has a
low 31ope, wr conClude that the characteristic sca.1.e of a horizontal inhomo-
geneity in the vicinity of North 'I',yan'-Shan' ic no more than 70 km; extrapo-
- lation of the graph of K(Ai~) tcward low values of A~j to a level of Kii = 0.5
(the some us thr. 1eve1 of t e minimum on Din's curve gives a value of about
50 km for the imtervtal of correlation of inhomogeneities. In North America
(region of t}ie Tonto I'orest group) the corresponding dimension is 30 km.
The North Kazakhstan Group of Stations. In the given case a more compli-
c:ated tectinique was used for studying the shape of the P wave from remote
eurthqualces.'Phe :;imilarity of shape was studied in three frequency ranges :
ut) to 1 Ilz, frorn 1 t;o 1.5 Hz rsnd from 1.5 to 3 Hz; to do this, three numerical
filLcr:: were Irtroduced into ttie program for calculating the correlation
coel'ficientr, urici a:s a re::ult of filtration, three signals P1(t), P2(t)
and ['g(t) charucterizing clearly bounded spectra, were derived from
each signal P(t).
The correlation coefficients were calculated f'or two groups of earthquakes
in intervals of epicentral distances of 3000 km < A < 10,000 km and A>10,000 km;
the latter intervEl corresponds to registration of the PKP wave.
F'or the first group of earthquakes the coefficients of correlation of shape
were calculated as well from wide-band recordings from 0.5 to 3 Hz (Fig.
62).

~
~
Of . ~/0
/00 " tOOdV,sh
Fig. 62. Correlation coefficient for
shape of recording as a function of the
distance between stations of the North
Kazakhstan group from wide-band
(0.5-3 Hz) recordings. See Fig. 52
for explanation
131
FOR OFFICIii:. USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Ft11t 01WIC111L Il;il; t1NI,Y
'1111' ftl, 1!1'illl1t Wvf'1~ ill1d,1-t'11 !'f~~~ Lhr'v1, I'vr.ijiu,tif'y runit,l1t1
iri i.lit! t'orrn ul' t,tic! riw~rrai!~~ vnluou ol' Lhe r.orrCltitidn cnrt't'irient uti d
I'urivt.lUn nP ttic dinttuice hetWacn ntut.lnnts A, I csnd nre nhoWn in Fia. 63.
'11w inllivifiwil vnl1ns.c wrri, tivrrti!t!d ovrr- I,hr dint,r,.nee irit.rrvrLin 11~ thnt
,~r(~ rili+wri hy 1ai,~~ t~ori .uril.rtl i in~~;;, 'I'in, ttinl i~l VCt'f,iCtl.l I Lr1C ~hoWU ~hc 7U%
voril't.clc.`flCt: {t1tCCVCLL f'or r_vnl.untlon ni' thr -LVr~rnga, an8 the broken 1inC
t'cjr t,he tndtviduu.l vtducu.
f.
4s t
o . b i , j
t t
~
' ~
aJ !
A j .-L,
C t_..._r fr..Y
.
T +
1 1 1 ~
t~
I
~
e d
T
t
.
,
" /D , !0' #0 � 70 !0f 10 / J
f + j f 4 .
vii;. 63. Coefficient of correlation of the shape of a recording as a function
of distunce betwecn stati.ons of the North Kazakhstan group: a, b, c--for the
intcrval of epicentral distances of 3000 < A 10,000 km, PKP Wave; a, d--frequency
.intvrvrLl of 1.5-3 tlz; L, e--1-1.5 Hz; c, f--1e3s than 1 Hz. See Fig. 52 for
(!xplatuLion
r,nrt,li(junkr,:: from the t.eleaeismic rep,ion 3000 < A< 10,000 km Kood r-patial
liL;/ of ..hupe i:; obrerved in all the investip,nted frequenc,Y ranges.
(-'or t'requency bundr, of 1.5-3.0 fiz and 1� 1.5 Hz the graphs have identical
rivhrLvior, but, on n hlgher level in the latter cewe. The decline in the
cor�rclation coeCficient witli nn increase in Dij up to 30-60 km should be
- ntLrthuted to tte chrirfLcteristic inhomopeneity of structure in the locality
nt' Ltic :stntion, of the Nortli Kazakhstan group. There is a striking feature
in tiehnvior of ttic corrclntlon coeffiMent in the t'requency interval beloa
1 ifr Wtthin ttic li.mitr oi' thls uame range of epicentral distances: compara-
ti vcly tiigh correlfition of lonp,-period components of the recording at
di::t.nnces of morc than 30 km between stations.
132
FOR OFFICIAu UtiB UNLY
~
i
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
I4111 tlFl-At>IAI. 1INI.Y
t'ompnri hg thr rr.uial.t,so WI th datri crl' othrr hut,hnt'i; (r;he h'tg, 5 nhd 61), Wp nrc-
cunvtrir.rf thrst thc gpnernl naturr. nt' th, ohntiql gtructurr nP tht, P wavp
culttClrleti I'nr vnnenL{nl l,y di f t'err.nt rertlann or t.hr r_nrth. ACtuglly, na uo
r.nn rter. Crcym Ftg. Gs, Lhr cor�reltsticiri or ciiwnnlt; rst pointtl df' regintrntion
c]rc:reuaes with tnrreafltlg dIflttitlCt' heLWrrn them. At thr t4hme timq, fo11oving,
n rrlnt,ive minimum in the cerrelnLion hC signals nt didtanccg of the order or
`,;U-'rU km ( t'nr frequrncien nC 1-1,5 tlx nnd diattinGen or 3000 -c A-c 10,000 km) onc
observen acompnrrstively ntrong increane in cdrrelation dt digLaflCEn about
tNicr nEl grent.
Attention uhnuld n1no br given th thc chanr;e in pt,t;ittans or the minimum and
mrix.imum ot' the dral, and np{ ke i n rsignal correlation er, dependent on f'requenCy
uria WrvC tyre,
We rnn nttcmpt, to cxpltsin the 1ndiCn~ed peculinritien in the behavior or
K(Ai,I ) r~n Lhe buuiu r~f the f'ollouin~; Cc~nCeptr. When u h Wnve propggateg in
r, m~c~t,am wit,h inherroi;enelLtcn thriL urr riindomly dintributed, fluGtuationg are
ct,:.~~rvr.~l cf tmul.t,nncr,ussly i n thc umpli tudr. nncl nhase nf the Wnve. 7'his fact
may be dur_ to f,hr, i;uperpouittun or Wnvec; Pormed nn inhomdgeneities or the
medium on the primary 11 Wuve. A:s u result ef such superposition, fluctuationn
arinc in thc t'orm or n trnin dt' nrcillattons bf' the P Wnve that arp rep_,istered
rkt thr br.i!Inntni; or triC recordtnp. Crnduul nttenugtion or rarrelgLednesa af
7ij;nalc: with un tncrcune tn thc relative tli.:;tance t,etWeen utations charac-
terizcu inrreasing dispnrity or the parumeterr. or inhomogeneitiea in the
strurturc or the mcdium directly bcneuth the stations. The observed minims
and muxima af correlation or uignalu nre nzssociated with the effeet of the
mo::t pronounced inhomnE;eneiLies, which ure probably t'airly extended and
nbrupt riner, and dropr, in cnntrqnty seinmic bnundsrirn.
'I1ke grrntcrt contribution tv the t'ormntion di the wave registered on the
open surruce i3 made by a certain rep,ion of the interface beneath the station
uitti ef'fective dimensionn that cnn be reprcsented by an ellipse uith samimajor
nxin equnl to
~ ~/`D17T'
R, x1~n  1' iin
whf,rt! V t:: t,hc vr.lori ty or the wnve, m ts the period of the wave, H i3 the
nw-r:u;i, dcEt.h or th(. IntCri'ncc, 0 t:, Lhe rLngle at' incidence of the wave on
ttie intcrPucc or the ungle or inrlinntion of the interface in the case of
normnl inciderirc of the Wnve (F'ip,. 64).
A/', / X 417
!l h ~
Fig. 614. Diagram or the influence that the
relief or boundrsry M has on propag$tion of a
P wave: n, b, c, d--sections of boundary M;
~--angle of incidence of the Wave on the
boundary
133
FpK Ob'F'ICIAL UtiB UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Fdit 0FFICtAL U5F: tlNLY
oI,vic�un.ly thr nnture nC thc ncttnn or boundnry M nn the ptLrnmeterd nf a p Wnve
Wil.l. ba Cnnri3ciernbly difi'rrent Cor Ct1nCE1 ahere the e11ipno With nemimt'Jnr
axlu ttI cdvcrn nectiona a, b, e gnd d respeetively. 'i'hin Wi11 ahoW up in
t.hr behrsvlor or t,hc r.orrrlritidn caefi'iGient t'or the correoponding ptiiria af
rr.t;intered ntgnulr;, nCrardinly, td avnilnblr informatfdn (Vn1'kdvskiy, 1973)
Hie nhapc or the rel iet' or boundary M in Northern Kaznkhntnn uhpre the group
o1' rcel.timte ntntlentt Wnu ldcated corregpondn tn the diagrnm shavm in Fig. 64.
l,et ut; enrry out nimplL, rnleultttinng that Will give qu�kntitntive eatimateB
c,t' thc ponniblr i'luetuntfenn in Wnvenhnpe. We t+i11 aanump thftt the gverage
ticr,Lh or or.CUrrence or bnundnry M is N=5n km, tand the veldcity or a t+ave
tn V-a 8 km/n, 'C -t 1 n, ~ = 501). For nuch Nunditions tlie effective dimenginng
ot' the mr_dium thnL pluyn an ennentia1 pgrt in fdrmation or the p utive trgin
ure determined by the qunntity 2tt~ u6O km.
'I'tiiis mr.nnn thnt, !'or nvernge Chkraeteriatic dimeniiidng of undulations ef
t,oundary M or Y,he drder ot' 60 km the coeft'iGtenc of Carrelatian for a P uave
uith pcriod or 1 u mny hnve nnhsrp drop (minimum) in the interval of
dtr,tianr.c:i betwecn uttttitinc3 of 50��'r0 km, und a rise (maximwn) at distanceg
or the order nt' 120 km.
AnEiol,ou:; connLructtons can be done f'or evgluating the influence of the
t,u:srmcnt, boundnry, ahi ch occurs nt an nverage depth of 5 km with velocity
or hnrixnntal Wave prepngatidn or gbout 6 km/g, 2R1 a 12 km, which also
nr,recn With the hEhrivtdr at' the function of lsptial correlation of wgveshape.
' !t, t.hi! :,ame Limr, ur.eordinf; to the eonsiderntions presented above, the
po:;it,ions: or the relative maxlmwn and minimwn or Lhe funetion of spatial
corrclatidn ::hould ahift With increasing period of the Wave tos+ard greuter
va,lues or relativr dintances betr+een ratations of registration, Which ia
probubly Just :rhnt happens in reality.
A clusc relution betWeen the correlation of shgpe and frequency is also
observed for PKE' waves recorded st flistances of more than 10,000 km from
the source. It should be noted that the spatial stability of the recording
inr.ren:;er. r+ith u reduction in frequency: for the loW-frequency range one
oboerve:: the hip,heat vglues of the correlgtion coefficienLg. 2'his is
'ippnrcntLy dur_ tn tWO circumstances. In the first place the PKP wave
approache.: Lhc group nf netamic atationn practicslly from below, and therefore
tho ::er.wion uf it:; pnth through the upper part of the mediure that is charac-
ter�i �r.vd b,y inrreru;ed inhomoGeneity is shorter than for the P wave. In the
:-wcornl Eirrr_, the predominuting periods of the P Wave are longer than for
t,ht- tIKlI w,ivc, and theret'ore the major part of the energy of the signal is
ornnrv!nt,ratcd [n the frcqucncy hnnd below 1 H2; in the processing system used,
thw :0.1t,nai-to-crror rittio of the mersurements of experimental data was com-
prrnt.i vcly lox.
An exrunirintion of the nveraf;e level of spatial correlation within identical
tand:; (::vc 11,,. 63) ::hoW:: ttiut some reduction of correlation s+ith inereasing
134
FOft OFFICII,L UtiE UNi.Y
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'nlt 011tf;tAL 11S1; ONI,Y
1j1iir~~!nl,ri 111--11,lrwl- i;, ubwrvud t'nr, the intervrd ut' 1.15-3 Iln.; !'or the
lnt.r.rv+d of' 1-1 ilz Ltic lr-vrl c,i' r.nrrelntion its in a rangr ef 0.5-0,8,
brin.V. appreGiably htE;hcr Por thr. t,rlr:;rinmic region thnn fnr the inter-
rnc,cfintr. yhnr, nnd thr_ dtntrsnt, >onc (Mh 41FVr.). tn tMe t'renuency interval
IorloW 1 IIv, n rr.i;u1nr irir.rrn.r tc; r,t,nr.rvrtl iri t,lw ntntiK1 edrrelqtion
til' t.lw u,tvr. Wil,h rui in+-rum,c in r,p1ount.rr1 diutunec�.
A Common typiCU1 f'cnturt! nC t.11 graphn to a1sn a werslcly L-xpreaged rise in
tht, region or et I is ln km, Whlch in a11 prdbability in aagoeiate8 with the
utpearaneC nf inhorropeneitien u.ltugted in thc very toprnost igypr or the
earth'r; crunt.
Thp CCntral 'Cyan'-5han' Grdup df Stntiong. 'J'he tranaverae gpatial eorre-
latidn or the sihnpc nf tihe h Wave Wns estimated Cram recordings of 37
ebCthqutikeu frem dif'f'errnt tricentrai regidnn. For each earthquake, in
nGcnrdunre With the method dencribed above, the mgtrix of the correlation
CuefCicient wnu Culculatr_d ibr the ::hnpe o� thc nignnl recorded on the
vcrtir.nl chmponcnt. 'Phe intervsl or relntive digtnncen between recording
rtution:: vuricd f'rom 2 te 30 km. '1'he epicentral difltanCeg rangcd from 3000
to 9000 km.
Shown in F'ig. 65a in thc dependenre ef the cerrelatiofl coefficipnt on the
dtstanre bctween seicmic gtationr,, unc] in Fig. 65b on the distgnce beLwpen
thr eri Crnterr, nP egrthnuukes.
- _ - - ~
Htf'~t+
w
-f/ ' t F �J.7.�/0 lI
. ~
.
1 ~ 1 f
~
~III /N/ 1W ,Alr pM
dy,Pw
b'ig. 65. Uependence or the cocfficient of eorrelation of shape on the
di::tuncr betWeen statfona (n) and between epicenters (b) from observations
nri i.hc Centrul Tjrun' :Shnn' Crout. See Fig. 52 far explnnation
For thc function or r.jiatinl correlntion ns related to the distance between
epi crntern or ersrthquukc:;, spikea and drops in correlation of the shape of
t.hr I' wavc urr noted at dir,tnnces of 1500, 3500 end 5000 km. The general
l'iorm nf 6iii- relntion :nd the locntion or the characteristic peculiarities
:grve r+cl i wit.h Lhe dnLn on other regton:. or the t+oz'ld (see Fig. 5), as Well
rt;s wi t,h Llic dnt,n ort t'LuctursLionr, of the anplitudea ttrtd travel times of the
I' u:wc Ltint ure given in Lhe preceding chapter.
Tlie corrclutiun function fnr the shape or the P wave as dependent on the
rlt::t,nnce heLween seiHmic .tations, just un in the preceding cases, shoWS a
135
FOR nFFIC1nL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'dk t)h'FtCIAL U51. CINLY
i,uWard r;raduiil no,i,onurttoti ut th 1nCretuing rt-lnttve digtance betr+een
;;I.rittf-,riEt, Moreuver, i Cnmhbrutively c1enr minimum in correlatian in dbserved
in the vicinit,y ni' 8-11 km. tt can bt, antiumed thgt this minimum on Fig. 65g
arh.un under tlir! Int'luenvr nt' uridultitidncs vf ct nhnrp geinmic intprfqep in the
upicrmoat luirt or' thr_ rnrth'cti erutst, nnd the naturt, nf the overta1 deG1ine in
oor�r�rl.attun to tL mnnl t'clitation af the prLttern:s di' distributidn of inhamoge-
nrit,te:; in the crur.t nnd mnntle like uhat in nated in other station groupg.
11ie Garm 5tgtion Grnup. 'I't� dependence dt' the cnrrelation eoefficient an the
rnltttivr diAtnnce bGtueen ceismie ntntidns (F'ig. 66) in founfl from the flata
oP 18 enrthqunkett With epicentern nituated chiefly in the seigmically active
r�"ginnes nf the PacifiC dcenn be1t. 'Itie digtunce betueen rerording stations
did nnt exceed 25 km.
Ct+�. 66. nependence of the coefficient of
correlgtion of shnpe an distance between
ntations from observgtidns on the (Iarm group.
:ee Fig. 52 for explgnation
c'ornpur t ni; the dutFi o C the t,urm group Wi th thc datu of stations of Central
'I;y:Ln'-:;hun;, we are convinced of their identity. 7'he decline in correlation
ot' t,he :shtpe ot' the P r+ave is aomewhat Weaker in the 12 km region, which
mi,y be due to n less shgrply prnnounced boudary that is tracked in the upper
E.nrt of the earth'u erust.
'11ie Europegn Station Group. To evaluate the spatial correlation of the
:shape of the P wuve, recordings of 33 earthquakes from di fferent epicentral
regionn were used. 'I'he matrices of the correlation coefficients were calcu-
latcd for eucti earthquate, as in the other cases. The distances between the
recot�cling stutiono varied from 3 to 100 km. 7'he epicentral distances were
3500-10,000 km. 'I'he resultant data are shown in Fig. 67 in the form oi'
r!!lntion:: for the rorrelation coefficient as a funetion of the relative
di:;tnnre betWeen stations (Fig. 67a) and the relative distance between the
opirentcrt oP carthyuakes (Fig. 67b).
v
0

~
~IV
~
++i,,++f+f
frr ~.v a~n
dplin
Pir,. G'l. Dependenr.c: of the coefficient of correlation of shape on distance
twtwcrn stutionz (u) and between epicenters (b) from observations of the
LuroEn-un Group. See Fig. 52 for explanation.
136
FOR OFFICIi,L UtiE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1
. , . . .
ey,.~

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Mk UFF[ttIAL USL t)NLY
llrinl,yn.tt, (W thc rerui trLnt cititrt nhnWri (Fit!, (i`tu) that in the lora].ity di'
iit,il, ionu ui' tlir Muropoui I!roup juu;t, rics i n Lhe vt ainity ot' the Ndrth Kaxtilthgtnn
iLrltion I;rnup Lha der.ltne in corrrlation oi' ntiape of tihe p wave is mogt
prennunced .tn thr rtgion hC 50-60 km, un in the rine notied in the region
cloue to 100 km. '1'tl e drap in tha funCtinn of spatial correlation in the
vicinity of 10-12 km thut in noted in nther regions is practical].y unexpressed
in this Caue.
In thin connectidn we may unnume that in the lncality of the European grnup
of utatinns tiy analnp,yr with the reginn of North Kaxa}chstn,n there sre prob$bly
vnriations In the dpeth ot' occurrence ot' the MnhnrovMc diacontinuity, which
uccnrd+.ng ;o the duta in the litergture (Belyayevskiy, 1974; Pavlenkova, 1973)
i u fiitunted here nt an average depth of 40-45 km.
7'1ic gruph ohnwn in F'ig. 67b far the cnet'ficient of correlation of shape of
t,tiLI I' anve uu u functinn af' tlie distunce:; between epicentera demonstrates the
:uunc nLablc Cnordination of drdps, and spikes in correlation thst is observed
in other regions. Ap,nlnst the bnckground of cu general decline in the function
of :ipatiral correlation that nsymptotically npproaches a 1eve1 of 0.5, one
obsrrve; relntive minittia of correlation for relative distances of 1500, 3500,
yUUU and 9000 km und correlution spikes at 2000, 4000 and 8000 km.
The Zeysk 5tation Croup. Datn on the Zeysk group of stations were found as
u result of processing recordings of 16 earthqualces of the Pacific Ocean belt.
Epicentrnl distuncec are 3000-6000 km. 7'he relative d3stances between
utntinns vnried from 2 to 20 km. The results of calculations of the
cnrrelation matrices are shown in Fig. 68 in the form of the dependence
on dintunce between sei,mic stutions.
N0
1 ~
Fif;. 68. Dependence of the coefficient of
correlution of shape on the distance between
stations in the Zeysk group. See Fig. 52
for explanation
dU, iY1
Aruit,y::i:: of the>e datn ;;how, that in the locality of the Zeysk station
jr�uup one ob;:erves two relat.ive minima of correlation confined to relative
di^.t:inccs oP 4 nnd 7 km.
Ttic uvernll decline in the relation K(A1j) toward 20 km takes place more
intercn-ively thun in the regions considered previously. It ean be assumed
ttiuL ttie rcliitive minimum in the funetion of spatial correlation that is
di:',tingui .hed in the regions of the North Kazakhstan and European groups
of ::Cation:: i.n thc vicinity of 50-60 km wi11 shoW up here at somewhat
clo:,cr rclztivc di:;tunces.
137
FOR OFFICI/,L UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Ft1R OFFIl;lAl, II;;I; 11NI.Y
1)Lir_iiustlon oi' Lhr [Ze:3ultu
~ 7'11r thi rd part nt' the mnnograp}i hns nresented experimental data on the
- :slifit,iul ntrurture oC t'luCtuationg in the charncteristica of a longitudinal
wttvr, 'I'tie nuthnry net themselves the problem of evuluating t'ne characterfgtic
sr.ates dC ma,jor otructural elementig of ttie wave f'ield and shdwing tiheir
rclation to cnrreipnnding inhomogeneitien in tihe inner structure of the
dcpthn, of the earth.
'i'o t;et. c1aLu on thc rntto of typicul 3ca.l.es of elements of the wgve field,
irif'urmtLLion wan u:sccl on drvintinnn t'rom the avera;e (standard) val.ues of
;unp]. i tudc3, truvel time;y und thr ahnpe of the longitudinal wave as provided
by dceh aeiomic sounding, on t;emporary expeditionary stations in different
t.ectonlc province:3 oC the Snviet Uninn.
`ChV mcLiri procedurul technique on which the study of the fine structure of
ttic 11 wave to based, consists in evaluating the stability of deviations from
averat;e fluctuations of kinematic and dynamic parameters of the wave field
~s, a Cunction of the relatfve distances between recording stations or between
the epicentral regions of earthquakes.
Convcntional methods of statistical processing of experimental data were
ti::ed for cluuntitatfve evaluation of the stability of deviations. In par-
ticulur, extensive use is made of correlation and structure functions.
A:; a re::ult of the studies it was found that the spatial structure of fluc-
l:uuLlons of such elements of the wave fiel'd as amplitude, travel time and
c of t,lie lnngitudinul wave huve distinct characteristic scales. The
conccpt ot' ocalc in the t;iven instance is arbitrary, and has the following
concc:ptual signil'icance. Tfie depths of the earth are characterized by such
z::tructure that [n the :~election of a pair of recording stations or a pair
cil' ::ei::mic soilrce:: on ttie riverage one will observe an increase or reduction
in the deviation o!' wave parameters from certain standard values if the
c�eltttive distance:: between corresponding pairs satisfy the observed spatial
cnlc.
s. Table 30 summarizes data on the spatial scales of elevated arid
rc>cliiceci value:; on the average in fluctuations in the parameters of a longi-
tud[nal wave. Fig. 69 summarizes experimental data on the spatial corre-
lution of the shrspe of the longitudinal wave.
It is rnther difficu.lt to give un unambiguous quantitative interpretation
ot' the scrsles differentiated for the fluctuations in paxameters of the P wave
and to druw :;pecific conclusions on the structure of the depths of the earth.
At ttie ;.aunc time, the � resultant data give such clear evidence that the
110iticidence of the characteristic scales of fluctuations of elements of
thc wave ('ield ttiat ure distinguished fn different tectonic provinces is
riot, acciricntal, t1iat we must accept at least their qualitative relation to
inhomogeneitie:: of' thc internal structure of the earth. .
138
FOIt OHFiCIt'%L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
IeUIt uFFIc; I nl. UyL c)Nt,Y
'I'/1151h; 10
Chrxrrieteriistic tahntiu] ssea1rr, o[' elevuted (max) and depregsed (min)
valucn oi' deviutiorui f'rom cLVerivr,e (ntandgrd) fl11CtUgEiOfl9 of th2
amplitude dd 1g A nnd truvel timen ddt, und aldn in the va1ues of
thc GnQrricient nf correlation Kij ot' longitudintil waveshape
r
~
rtpee,e~ee�u�w we.r~eM (~r1
ne A.~T~ ~
rp ~ln
ltaMMalI KC~

u Ir
r
me 1 roln
Mae I nNn mIt I
leln
\ J hIU @NNNNYm P1Ctttl11NNI1M MlNAr ttiN11NilYN
14 61 teppMtopNO CCCp
5
3
a..ixp iwp-.12111 Wu-.Ia0015d0-2WO
51-
(2U11-'tl~q 111N1--1UIUAUNI--5U1N1 `.~AtlW -

T
11~nneAeM~~ M~tt
CCCP
,ti'~
'~'i
6Ln
A K
so-so
pNY
u~ttTiN
' 50-.80 I20-I50 I00-I40
a
Q
7 6*PMYA ToN4WsN
JtXI--SUO 80-1UU 75-tW 53--75 Ix-190
50-70
>e:tic t'ield nnomulies with respect to the scale of inhomogeneity is uncertain
(more quulitntlve thun quantitative); nevertheless it is useful to account
t'or Lhts relution for decisinn making in controversial situations.
F()r roj,ianvs oC Soviet Middle Asia and Kazakhstan we had data on magnetic field
nnomu.l ies A7'a averup,ed over agrid of 10' x151, I. e, corresponding to a
:3caAe on f;eophyoical mnps of ~.;2500000. Just as for the gravitational f3eld,
no rclrition was ob3ervcd between the seismic and magnetic field.
An exr.epl;ion is the grapll fbr Am, DTQ, where a linear relation is observed
t'ur points corresponding to North Tyan'-Shan' stations; the correlation
roe�ficient is equul ta -0.7. However, this relation is based on the data of
unly eig}it seismic stations, and therefore the confidence interval of the
rectuced estimate at the 95% level of significance is from -0,10 to -0.95,
frocn wtiich it can be assumed that the peculiarity under discussion came up
t'ortuitously.
One of t;}ic rearons for u lack of connection between Am and the investigated
1,,eop}iysical fields, (both gruvitational and magnetic) is that in the case of
compnrrttively short epicentral distances, up to 2000 km, seismic waves go
t'roni the source to the receivers through.the upper mantle. The latter is
char+scterized by consideruble inhomogeneities of the velocitfes of seismic
waves and ubsorbing properties, and therefore the observed magnitude correc-
Cions reflect the influence of all inhomogeneities that lie on the path of
hropagation of the wave, whereas the geophysical anomalies characterize only
:Lova.L iriliomoirerieitien close to observation points. The fnfluence of these
intomogeneitie: on the nmplitude peculiarities of seismic waves that traverse
thc., grenter pnrt of the path in the very inhomogeneous top of the upper
mant.le is appreciably weaker than for waves that go through the appreciably
rnor�r homogeneous lower mantle. Thus the geological�-geophysical character-
i:;tics of the medium permit a more certain prediction of the effect of an
anumalouo- change in the amplitudes of seismic waves only from remote sources
(A - 7,000-10,000 km), I. e. for cases where a considerable part of the path
oF j,r�opugation falls to the lower mantle.
*llt Wmpcr:titure:a in excess of 700�C rocks are completely nonmagnetic.
/1ric>malier. of the magnetic field can be related only to the topmost layer
o 140 ktn.
152
FUR OFFICiAL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FUk UFFtCtAL U5L ONLY
Chapter 2, Inveutitpution of the Relution E1etween Ainplitude Anomalieg aiid
Oeolot;icul Cnnditions
1. Influencp that genernl fentures of geological geruceure have on the
level of amplltudes of spismic wavey
Ceological condibinns in the vicinity of' u ntation may have an appreciable
influence ori the size of the magnetic c:orrection Am. Apparently it is the
inCluence oC 1oca.l geologtca1 conditionn that is regponsible fbr the f'act
that in aorne ca3eo the em differ by 0.4-0,5 fbr stntiinns separated by e 2'ew
kllometres.
A whole complex ni' detailecl geologir.al studies in the vicinity of each
station is required to determine the way that the amplitude of a recording of
selsmic signals depends on the pecul.l.aT�ities of geological conditions. It
wnulci be necessary to study the wuy that the value of Am 3s influenced by
the thickness and makeup of loose deposits, the ground water level, ani-
sdtropy of the physical properties of rocks of the basement and the sedi-
mentary sheath, relief of the basement, the thickness of the earth's cruet
nnd the inclinations of its boundaries, the petrographic composit3on of the
- busement rock, the presence of deep breaks and tectonic fractures, the type
of geological structures that the rocks belong to and so forth.
Let us try to evaluate the influence of the most general geological and
tectonic conditions on the values of Am. Most stations were located in
territories where fairly ancient crystalline i�ocks (Paleozoic or older)
crop out on the surfuce. Such ure the stations located on the south of
ttc: Luropeun Part, within the borders of Kaza.kfista.n, in the Urals, in
'I'rnns-Bayknl, on t;he Kolri Penin sula and in the vicinity of Noril'sk. It
o;hould be noted ttiat all these stations are characterized by values of Am
from -0.1 tn -0.6, i. e. they are quite fuvorable for registration of seismic
yignals. M exception is observed in all the stations on the south of the
European part, where the values of em are about +0.20.
A second group of stations is situated within the limits of ancient platforms
with a thick cover of sedimentary rocks. Such are the stations in the
vicinity of Moscow, Kirov, Riga, on the south of the Ukraine, in the
vicinity of Mirnyy and Yakutsk. The vFil.ues of the magnitude corrections
on tllese statlons vmry over u wide range, and in some instances there are
stiurp differences between stutions locuted only a few kilometres apart.
In this connection it is quite clifficult to extrapolate the values of Am
to ottier terri tories.
LcL un di:,tinr;ui;;ti the stntions thrst fall within the limits of so-called
"tecLonic nctivrttior." TEiese are stations that are situated in Tyan'-Shan',
1111,n,y, Ku�r.ba;;u, thc Wc:;tern iind Fastern Sayuns, and in Pribaykal'ye. The
vulur. , oi' /1m herc hnve n range from -U.hO to +0.30. An increase is observed
in Fhc v.Lluc:v, of Am in the easterly direction. The region of tectonic
at:tivzti.on is charlcterized b,y a differentiated relief, alternation of
153
FOR OFFICIAL L'SE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
P'OR OFFIC tAL U5l: dNLY
;,nrl',icrct uiit,or~~pplrigu or rockEi af' thc I'o.ldotl t'ouridution ttincl nedimentar,Y btiuina.
N(j isy,;tcnu.ttlC pr.CU11ur1Ltcn r.nuld be Cdund in the relatinn between the
vatuc: nC Am tlfld ({ebIC1giCti1 cdnditions i.n the vicinity at' the stationg.
PInrill,y, fs nurnber ot' stntionn are nituuted within the limits or an the
boundnry ot' the zone oC the MeAO-Cenozdic p,eooynclinal regfon. 5ueh are the
;;1,n1,1.on,n in thc Cartathinnn and 'I'rnnn-CnrrtLt}ittt, in mranayCauCasia. 5outh
'I'urkrncnirl, nn Chukntka, within the limita of the 'I'ttdzhik nepresgion and on
the l1umirs. 'I'heac gtutionn fiLso dif'f'er otrnngly in values of Am. For
lnotance in the Cuucnsus, in murkmenig and on Chukdtkg the vglueg of Am
nc�e iet;utive, while in the mudxhik tiepresginn und the Parnirs they have
po:;itive valueg,
'i'tus the genernl feature:i of geological atructure in the vicinity of a
:3trrtian do not determine the particularg of the nature of registration of
ou'iomic uignals in explicit fdrm. Apparently to determine the influence df
genlogicut conditions we need to do incnmparubly more thorough, detail.ed and
comprehensivr research on the geology and genphysics of the station locality.
2. [nfluence tliat breaks heve on the parameters of seiemic waves
More obvious was the inf'luence of breaks on the shape and amplitude level of
:,ci3mic signal:s recorded on stations situated in direct proximity to these
breaks. For seismic wnves the fault zone may be a shield on the one hand
thnt reClerts purt of' the wave energy, and on the other hand may be an
.inhcimogcneity wit}i anomalously high absorbing properties.
A:; cxperimental material, we used recordings of 40 remote earthquakes from
Lhree ericentral zones: Aleutian, Japanese and Indonesian. The recording
wa, done by standard 3eismic ch$nnels of the SK-?M seismic receiver. Signal
amplitudes were measured from recordings of the vertical component.
'Ptie amplitude of the maximum phase was selected as the parameter character-
izing the dynamics of the longitudinal wave of a remote earthquake. 2'his
chotce is dictated by the fact thet the maximum phase of the P wave eorre-
::ponds i'airly uell to the maximum of the spectrum of oscillations and is
ttic most stable in comparison with other phases of the given wave group when
tho frerluericy respon:;e of the recording channel is used (Antonova et al.,
1908; I1u:,clicnik, 1970).
'I'ir�ee major factors influence the amplitudes of remote earthquakes: con-
s i t. i un:; at, ttie focua (energy, depth, directionality of radiation, etc.),
tlW c,l~rtic purumetera of the medium through which the waves pass, and the
environment close to the recording station.
Con:, icferiition of the influences of the first two factors on the amplitude of
t.hr ::t~i:;mic sirnril wns trsken care of by calibrating the recordings of closely
Evaceci tut i ons wi. t.h re:: pect to n single stution taken as the reference; the
raLio (xi = Ai/Aut ts calcutRted, wtiere Ai fs the amplitude on the i-th station,
urid A;;L is the amplitude on the reference station.
154
FOE2 OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
PUlt UFh' i C 1 AL U51, c1NLY
/ t J
~w~~- rW1Y~- ..w~~ on~~
- ^.W W*f ~~,~~I~W ~-.+,-+.,,,r,.,~,, em v
"M~VwM1A~- ~~1~1~11~- ~^w~M�.,r~. em!
-~+rw~` ,M'-'A~ - em r
AYI~
te io 0
it , I , I
PIg. 74. Recordingr of remote earthquE,kess on seismic stations of the aarm
group: 1--Aleutians, 21.I7.1968, to =1gho8"'; 2--Japan, O1.V.1968, tp = 08h43m;
3--Indonesiu, 23.II.1968, tp =16h14m; cm = stntion number
'I'akeci an the reference was station 1(see Fig. 76) which is located in
rclatively quiet geological conditions and is fairly remote from tectonic
t'rnctures. Hesides, on ttiis ste.tion one observes a comparatively stable
nimpte pu13e shape of the first wave group for the given recordings of
Japane3e, Aleutian and Indone3iMn earthquokes (Fig. 74).
The nature of the influence of fault zones on the amplitude of a registered
remote earthquake signal was studied from pairs of earthquakes; Aleutian
earthcluakes were taken us the reference. This is explained by the fact that
in compnri:;on with Japanese und Indonesian zones, the Aleutian zone has
higher seismicity, i. e. prACtically every month there is a tremor there that
is clearly recorded on the Carm station group. In the second place, this
zone i:; very stablc with renpect to the amplitude parameter ai.
Ttic ::tAbillty of the parameter ai was studied as follows. Spatiel constancy
uf' ttie qunntities ai rrri:i verified. To do this, pairs of earthquakes were
::c.lec:ted with close aPpnrent periods of the recording of oscillations, and
tttr ri1Lio of nmplitude3 of like phases ai =ai/ai was examined. The paired
cnc�tlitlunkes were :;elected within the limits of R strirtly bounded time
:;ci;ment not exceedlnM ttiree duys. 'I'his limitation is explained first of all
by poe,:;ible ctluIpment in:;tubility iri time, which had to be excluded as much
tuc po:;:;ible.
11:: a rr:;ulL of un c:xctimi.nation of about fifty independent earthquakes ft is
uppnrent thut Lhe vulue of u; for ever,y i-th station is fairly stable, and
155
FOR OPFICI/,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'dk UFFiC1AL USN, ONLY
0, I
lri t'aet t,l,c uverat;e vrilur. u( for Lluch stn.tion iE; r.inge tn unity, whi1e the
rivcrrlNa tic:ilt;tcr oC inciividuul va1ueti ddcn not exCeed 15%.
Chnuge tn Pulse Shnpe dC u ltemnCe Carthquake as a FuncCinn of the Aximueh Cn
the Lpic:euter. Prel.{rninriry vistutl annl,ysis of recdrdingn of remote earth-
ilunken allowr, tira Ld drrkw the qualitat;ive cdnCl.usinn thut Cdr some statiidns
Lhe pu1:3e qhEipe ut' a lnngitudinal wuve is diHtorted with a change in the
fLLI muth to Lhe c:pi center, f ig. 'tli nhnwt; nn examp1e di' u recording of tihr~e
vurt,hrlunken Crnm diPf'c:rent epicentrn.l ronr:s. It is nppgrent thnt the nhape
ot' t,he rer.ordi_ng oC the N.eutian etLrthqunke hrIy the gnme clegr pulse fdi-m
on rtl l:;tutions, and di Cferu dnly in mrLxinium nmplitudes. For the Jnpanese
cnc�t,h(juuke this CitLrity oi' the Pirst wave grnup is disrupted on stations
4 rind 5. F'dr the Aleutian earthquake, the rgtins of amplitudeg of subsequent
o:;c111ittlons ta thc mnximum tirnplitudeq AmaX were less than unity; for the
Japutneno curthquuke theae ratios becfune greater thgn unity.
wi th rr.::pr.ct to the rcr.nrdingn ot' the Indonesian earthqualce the clarity of
i.ic t'( ruL wave tratn lu disrupted on stations 6 and 13. For the Aleutian
eart,hquake on ttiese stations the measure of distinetness of the pulse A/AmaX
wri.s loc:s than 1.0, bei.ng in n rgnge of 0.35-0.80; for the Indoneaian egrth-
quuke the value of A/Arnax is in a r$nge of 0.60-1.60.
'I'hc, pul:.e shape on stution 1 does not undergn nny appreciable changes.
'['he value of A/At~ax hei�e for mil earthquakes is lesa than 1.0, and is in a
r:itit*,e o C 0. 30-0 .60 .
ln thr inve3til;nted time reriod (F'ebruary, - May, 1968) the frequency responses
ori the ot;itions wrre rigidly stnndardized. 7'hi3 enabled comparison of the
rcr.ording:; 3hawn in F'ig. 74 not only with respect to shape, but also with
rr^,pr.cL to nmplitudes for the three azimuths (Fig. 75). As we can see,
t,he cii t'f'erent azimuth3 are characterized by different values of the reference
rccordings normalized with respect to amplitudes.
~
y
10 II
tw6
a6
tm
~s
QI
~ C~J ~
,
r,s
�
n
co�
~e
~ a
J
!,0
10 _ E J
,i a1 t
VO 60 s0 !00 !IO VO i0 !00 yr�
F'ig. 75. Graphs of the dependence of rela-
tive amplitude al on the azimuth to the
epicentral zone. 2'he shading indicates
azfmuth intervals of about 400 (Aleutians),
70� (Japan) and 120� (Indonesia).
cm = stntion number
156
FUlt OFFICIIw UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FUk nFFtC tAL U5L qNLY
Piir,, 76, merrhhiC dinp,ram of the tlarm
rr.r,irn. 'Lonen ot' hrogkn, tt--Karalcul' ;
6--t'r, troviik; rr--;;urkhnhcik; e--Crat,mhntn
ot' Quaternary breakn; 0--geismir. otat3onn
with values of azimubh4l sensitiivity
Quantitative Evaluation of Change in the Amplieude ParameCer as a Function
of AzimuCh. Crnphs were plotted fer each ntation for bhe average 61 as a
t'unction of the uverage azimuth to the epicentral zone. 7'hege relations
ure nhown in Fig. 75. C}iven for cach nverage value of al on the graph is
t,lic scatter of ttie individunl vuluen of the data in the form of "whiskers,"
nnd t;fir number of pairs of eurthquakes from which the relative amplitudes
ul were determined.
7'tie vnlue of aj(~) is a measure of the azimuthal amplitude sensitivity of
tfie i-th stgtion with respect to the direction of the average azimuth 0�.
11t nel:imic ntatinns 3, 11 and 13 thia partimeter varies in the following way:
It clecreasr.s by u i'actor of 1.7-2.4 for earthquakes of Japan, and by a
fnctnr ot' 1.7-3.0 fnr the enrthyuakes from Indonesia.
Otuttoro 2, 5 und 6"feel" thc Japanese earthquakes about the same as the
Aleutian ones; their 3enuitivity in the Indonesian direction is 1.7 times
lcss. Stations 11 and 12 show exceptionally high sensitivity to Japanese
earthquakes, exceeding their sen3itivity to Aleutian earthquakes by a factor
of 2.1 und 1.4. The sensitivity of the 11-th station to Indonesia is the
same as for the Aleutians, while for the 12-th station it is less by a
factor of 1.8.
Influence of Breaks on the AmpliCude Level of Signals. Before examining the
relati4n between the position of breaks relative to the station and the direc-
tion of nrrivsl of waves with the amplitude sensitivity of each station,
let us turn to the tectonic scheme of the Garm region. P;ai rly cnmplete duta on the geological structure and tectonics of the region
rire to be tiud in the work of M. V. (;zovskiy et al. (1960). We have taken
;i:: our i,aoin t}ie tectonic diagram presented in this work (Fig. 76).
'I'1ir Garm rcl!ion ir. un alpine structure comprised of two structural stages
iuul broken up into individun.l blocku (structural steps) by deep tectonic
brectk., Lh:LL .inLCrsect the earth's crust.
157
FOR OFFICI/,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Fntt nCCLCIAL USL ONLY
'I'hrec ;ti1;rr.p1y I'n.1.1 Lng, zonvit n(' rlerii-lrvr.l rrhnka nxe digbinguished t Karakul',
I'r. t rov:k nn(l �iar-kliob;tit , '
t~I add [ t tori tc) the cleep-1eve:1 breaks, fIlUlt5 of i;he Ruaternn.ry Era are
cl f:: t! ni;u 1 nhed thnt arr: rihown i n t'ragmenrs nn the diagrem acco.rding to data
tnken ['rnm the miip dC recent movementg nE' the earth's crust in the Carm
rer;ton compiled by V. K. Kuchn.,y,
t'tg. 'tE shows anly thone faulLs that are situated no more than 5 km from the
11pc (omi c :ttatiatin .
- On thr tectonic diagrum the map,nitudes of a.zimuthal sensitivity 514) for,
eac}i stution are shown Uy vectors. It can be seen f'rom the size of the
vcctoc�s t}iat the earthquakes with rays crosaing nearby fault zones are
"t'e7t" more poorly hy the stations. An exception is station 3 for which
the (iircetionul senait,ivity was lower than expected. 7'his, can be explained
by thc fuct thnt the lo'urktiobsk fault znne in the v3cinity of station 3 is
muc:ti wtcier than stlown dn the tectonic diagram (see F'ig. 76), and the boundary
of' t11e zone pnsme3 to the south of the station.
'1't,iu tiypothesis is confirmed by data on recent movements of the earth's
crust Fram the results of geodetic observations to the south of station 3.
t,}ifit :;how vertical displacements of the earth's crust, although such dis-
plrxcements have not been observed to the north (Enman, Sambireva, 1971).
`I'tc iriCluence of Quaternary breeks is also felt by stations 12 (Petrovsk
zone), and 2(Karakul' zone): the amplitudes of Indonesian earthquakes
tit tlie k;ive:n stations nre 1.8 times lower.
'I'lie wuy ttiat the reduction in amplitude of the recordings of remote earth-
cjuzilces is related to the position of the station relative to tectonic breaks
can be explained in the following way. The boundaxies of the fault zones
zii�e 2'ormed by surface breaks; therefore within the limits of the zone itself
that is t'illed with crushed rocks the amplitude of the signal must decrease,
nnd reflectioris from the :;urfaces of faults may inerease the amplitude,
i'oc�::ing it in the recording area. The ratio of these opposed effects is
ii:;uociated with ttie seismic wavelength and its path within the limits of the
Ihul.L 'I.OfIC (width of the znne).
l1n un:.t;il,le vrilue of nmplitudes is also noted on stations situated directly
on cZu:LLernnry breaks. For instance on station 13 the signal of Aleutian
r.�Lrttiqut,lce s i:, 1.5-2.5 times as strong as at the other stations. Japanese
cUrt.hcjurilce:, give recordi.ng:s here that are 1.5-2.0 times weaker than Aleutian
cairth(juakos. `I'he ::nme instability of behuvior is noted at station 12
i tuaCed on ri cZunternrir,y fault.
An :inuly: io of ttie amplitudes of P waves on stations of the Garm group
stowed Lhat ttie fault xones have a considerable effect on the spatial
:tructure of the wave field. This influence can be seen in distortion of
158
FOR OHFICIAL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOR OFFTCI'AL U5L ONLY
I.lic., pulne :lIlI1j1C rlnd tn t;he con3iderrLblc Cluctuut;Inns of the anpl3tudes of
:sciarruc wtLveu.'I7ie 1atter rnay have two pns sible causes. 2'he first is the
reflection of seinnic waves from the boundaries of fault zones and the
ubsorption of their energy due to propagation through the fault zones.
Aa a result ot' the action of these f'actors the amplitudes of the first
phu:;c;, of r,nc l' wave u,re reduced, while thosc of the 3ubsequent phases are
distorted by interference. 'I'he second possible cause is the waveguide
properties of the i'ault zones, r['hese zones are comparatively thin steeply
f'a11int; 1a,yer~ with reduced velocity. The difference of velocity from the
values typicul of the 3urrounding meflium decreases with depth, and at depths
oC 10-12 km disappears cnmpletely, i;he fault is "hea7.ed." The amplitudes of
:zeismic waves incident on ttiis waveguide increase as the surface is
approuched ur, a consequence of a, gradual reduction in elastic moduli.
In cases where the seismic rays are parallel to the fault plane, favorable
conditions arise for propagation of waves in the waveguide fault zone, and
one notes an increase in amplit ude 1eve1 on the surface as compared with
c1osely spaced stations. This mechanism is well explained by the increase
in umplitudes of Japanese earthquaces on the 11-th station.
'I'hus the given analysis shows that in selecting the points of location of
;ensitive stations in regions with complicated tectonic conditions it is
necessary to do detailed research that will establish the spatial block
structure of the amplitudes of seismic waves. In doing this it should be
tacen into consideration that the strongest deviations from average values
AT�e obser�ved close to ftzult zones.
'.I'}ie formation of the seismic wave field is determined mainly by the elastic
anci dissipative properties of the earth's interior. At present these
propcrties huve been inadequately studied, and therefore it is difficult _
to curry out n, detailed calculation of the kinematic and dynamic character-
isticr, of t}ie wave field. At the same time, a practically important problem
localization of the horizontal inhomogeneity of acoustic properties in
the depths of the earth is completely solvable by using supplemental
geophysicul.information.
As has been shown, there are certain correlations between the physical
parameters of matter in the depths of the earth; therefore, the magnetic
arid t;ravitational fields for instance may be used to get information on the
elastic properties of the medium, and consequently to predict the dynamic
peculiarities of seismic waves.
'!'he most characteristic peculiarities of seismic waves, that are associated
with inhornogeneities of the structure of the lower part of the crust and the
upper mantle, show up wit}i fairly good contrast in isost atic anomalies of
ttie gravita.tional field and in the nature of the block structure of the
ariomulous rnagnetic field. At the same time, sma11-scale changes in the
dynrLrnic characteristics of the seismic waves are intimately related to the
inhomogerieity of the ver,y top part of the cross section, which is expressed
159
FOR OFFICIrsL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Mlt ttFi,'It;lAL II;;I{ t)NI,Y
i ii It 1-1ru1,1,1' i ri t,I' 1~~!d1nn:n1,ii.r,y Li., tnnr,; r~t' t;~~Ctc~ntC brr.nku
llhtI rll,fl~!1' 1q!C111111N~~,111!1{ (-)r 0IC t1t;N11(!t.UCf?.
( ~t>11,' 1IIi; 1 I 1!I
'f',,n y,,rir;. lirivr, ;.lner_ publlcrtL{oh or Lhe Cbtnprelic:ti,tivh Sei.ftmolngica1
rnonogruph "Furidumentul. Expertmentfil F'attcrnr3 or the Uynamicn
(>t' ;;cili(ILIc, W(f.VL't," (MLonovu et ttl 1968), puring thltt time the expedition
hnn nI,r,ainefl rir.w ext,enn.ive experlmentri.l mtLteriul on whic:h the prenent
rn~7n~~~;�r~E~h i;; 1,risucl. A cornpuric.on or two pc.~rtod:l ;zenuruLed by a decade giver,
nji t(l(,,u oC t,tie cilrer,tton o(' develonment or c:xpcrtmentn,l ;3eismological rescureh
t'rom iunnl,vof:: nf' genernl (uvcrage) pMttern:; or uc:iumic wuven to thc st;uciy
()t' ~~~,rnF>>r�c~Ltv~~l,y firie dcLnil;z oi' kl,netnnttc and dynnmtc charncteriatica and
I-vrINal,ic,n ot' irilionogenett;f.cu of' the r.zirth':; s;tructure.
'17io, in-ain purpooe or thi:, work hiu, beeti thc investitr,atton oi' the mutual
r-lat, ion t,ctwecri i ntornol;enc [ t t eL of the enrth' ns.tructure and fluctuations
ut' 1,1w W11VQ ficlcl. 'Me riol.ution of thtn nroblem han beem made pos3ible by
t.h~, dr!vc.lopmr_nC ol' deriow o,ystemfti or aciumir abaervations.
'I'he ::I.u(l,y o!' Whc r.ompl i ecit;ed opncc-1;1me strur,ture or the wave ficld ha:;
r-yui rr_d otronN cciiemritixution: rl rough hierarchy or spntial ncales has
t,-cri i rt,roduced, the fielcl itse1P hn:s been represented uo the sum of com-
of cii ffcrent scales in a background-fluctuntion relation with each
'l'h(2 c�rnj)oncnts oi' the Ciel(I aro separnted by geolop,ieal eriteria and methoda =
r)t' :,~~u~�c~-t'r~~r~ucncy t'iltrution.
'i'1w irnorphology uf t.he fluctun.tions i.s constructed on t;he basis of the
0111Iriu:t,eri:,ttc:. of rs rilndom fiel(i. DesPtt:e the fact that such an approach
hfui h~'01) :;u~~~~ertc~cf t'nirl,y long fu!o, Lhere has been no unified system of
1l1.u4nr, i r�lt,ive (lecrri ption of the ueismic wave field up until now. 'Phe system
d,,,:sCr�ihcd hcr-c i, formulated on the ba?,is of the traditfons of research of
t'()rn;cr Yc11.1%, (Antonuva cr, u1., 1968; Nersesov, Ruutian, 1964; Nersesov et al.,
19(,'; Nikolnyev, 1972); it is fuirly complete and universal: any detail of
t.ho wrtve rJuld in, 3ub,yect to achematization and division into parts, each
parl, rnakc:: tt:, own c(.)ntrihution to the qunntitative description of the
::IuLtirt.l. ::enle. 7?iiu technique i.;; applicable both to analysis
()I' i.hw :;irnpio r,hrlrUc:.eri:ctic:; of 3eismic wnve:; amplitudes, travel times,
:M11 f,c, ,LrlrLly:,i:z or fsalrly complicated purameters such as those of polari-
z.,O.ion, cnergir;; in certain frequency bands a.nd so forth.
rc:,ult:, rire: in mfiny instsnce:s the fii�st of their kind. From our stand-
~loint, tho! mnin siir,nit'icUnre of these results is that they provided new geo-
iog,ic"11. anci geoph,y:'icnl inf'ormrition on thc medium and in this way proved the
oompct.enre of thw re::e,irch technique uNed. This gives us the impetus for
rocummen~l i n~~ ttutt rc i;;mo.logir,ts use our method extensively under conditfons
ot' i,hc.- widest, ;Lnd :,y,ternc of ob:;ervntions.
i6o
FOk ONFICIr+L UtiE UtJi,Y
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'Oit UFFtC I Af, USI: l7NLY
Natural.ly In the Caurse of rcscureh thc method will undergo certain changes;
nevcrthelena the renultfs found in vurioufi regiona of the glnbe can be com-
pared, uniCied and pr.rt'cr,trrl,
'Phin work haa entabliohed the fundrunentnl rough npatial characterigticg of
ttiC atructure oC the aeiemir wave field und horizontal inhomogeneities; the
t'I rnt, afl yet apprnximgtc, t'eatures ot' dintribution of hnr3zontal inhomo-
Keneitieg of various ncalen in the eurth hnve been eatabllahed, 7'he re-
Cinement af theae feuturen, the invecttgution of unstudied regiann of the
, carth, irnpravement of the relinbility and detail of infbrmntion all this
lu n rnutter Cnr the future. 'I'he nuthors tinpe thnt future experimentsl
;SCiumic ,t,udiec nF the ini,grtoc- of' the earth will take the path of development
oC the methndtl pre:ientad in i,hia book, '
[iCCEiiLNCT:S
l. I. Yu. Azbel', V. I. Kcyltn-t)orok, T. B. Yanovnkayg, "A Method of Joint
Interpretation of Hodogruhhs o.nd Anplitude Curves in the Study of the
Upper Mantle" in the book: "Muahinnqya interpretatsiya seygmicheskikh
voln" (Machlne tnterpretution of Seismic Waves), No 2, Moacow, Nauka, IU966.
K. Nci, I. L. Nersesov, A. V. Nikolayev et al. ,"Time Changes in the
V.luctuatinnn of Amp1i tuden ancl 'I'ravel Times of a Teleseismic P Wave
on the Curm, Cnlifornia, LE:S/1 und NORSNi Croups" in the book: "Sovetsko-
:zmerikanskiye isslednvaniya po prognozu zemletrvaseniy" (Soviet-U. S.
fiESeurch on Earthquake Prediction), Vol 1, vushanbe, Donish, 1976.
- 3. A. S. Alekseyev, M. M. Luvrent'yev, R. G. Mukhometov, V. G. Romanov,
"A Numerical Method of Solving the Three-Dimensional Inverse Kinematic
f'roblem" in the book: "Matematicheskiye problemy geofiziki" [Mathematical
Flroblems of Ceophysics], No 1, Novosibirsk, Computing Center, Siberian
Depurtment of the AcadertW of Sciences USSR, 1969�
J. A. S. Alekseyev, M. M. Lavrent'yev, R. G. Mukhometov et al., "A Numerical
Method of Determining ttie Structure of the Upper Mantle of the Earth" in
ttie book: "Mntematicfieskiye problemy geofiziki," No II, Novosibirsk,
Computinl; Cr-ntcr, : iberian tiepartment of the Academ}r of Sciences USSR,
19'tl .
5. Ye. N. A.Ltukhov, K. f.. VolochkovicTi, B. N. Krasil'nikov, A. D. Smirnov,
"h;xpcrLcnce in Stnndiirdizing the H'olded S,ystems of the Ural-Mongolian
13eLt wltti Conr,Ideration of the Structure of Their Basement" in the book:
"`1'ektonikn (Jrnlo-Mongol' skoflp skladchatogo poyasa" [Tectonics of the Runl-Mongolinn Foldcd Belt], Moscow, Nauka, 1974�
6. L. V. Antonova, F. F. Aptikayev, R. I. Kurochkina et al., "Osnovnyye
ek::pr.rimentul'nyye zakonomernosti dinamiki seysmieheskikh voln" [Funda-
mental F:xperirtientnl Patterns of the Dynamics of Seismic Waves], MoscoW,
NrLuku, 1968.
161
FOR OFFICIAL UtiE UNLY
s
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
r014 nN r rr. i nt, utii: nNi,Y
'f � Y,. t. Ar:Lnuv[c:li, (!cl, ,"llpprlruturri t rnetodikri uc,yntnometrichesk.ikh nablyu-
(It:nty v0,0SIi" [Methudr3 rin(l E'quiprnent; Cor Seiamntechnical Observatidnn],
Moi; r.oW , Nuuktx, 19'th,
11. 'l,, I. Arrinovteh, N. A. Vvedenqkuyn, t, Ye. Gubtn et ul., "Inntruktniya
c, pucyudke proixvndutvu i dbrnbntki nablyudeniy na neysmirheskikh
utantniyalch Yedinoy titaterry neyamirhegkikh ngblyudeniy SSSTi" [Instructions
oii tht! Orcler t'or Cnrrying out und t'rocenning Observations on Seismic
113tations of the UniCied Syntem of Spinmic nbservationn of the USSR),
Moucow, Institute of Ph,ynics of the Earth, Arnderty ni' Scienceg USSR, 1966,
9. M. Ye. Artem'yev, "i'lunetnry and 'Lona]. Intiomogeneities of the Upper
Muntle and R'heir�ftelution to peculiarities of Regional Tectonica" in the
book: "SvynL' pwverklinontnykh struktur zemnoy kory s glubinnymi" [The
Itr_.latioti Between Surfnce Strurturee of the Earth's Crust and Deep-Leve1
Ctruc:tures], Kiev, "Nuukovn dumka," 1971.
10. Ye. V. Artyuslikov, "Layer of Reduced Vincoaity in the Upper Mantle of the
E.firth and Phenomena that ure Nelated to it," BYULLETEN' MOSKOVSKOCO
U"CE(k;STVA tSPYmA`I'ELEY PRIRbDY, OTDELENIYE GEOLOGII, 1970, No 2.
ll , li. K. Balnvadze, E1. Sh. Mindeli, "Principal Results of Geophysical
Uudies of the Structure of the Eartti'3 Crust in the Black Sea Basin"
in t}ie book: "Stroyeniye Chernomorskoy vpadiny" (Structure of the Black
Sea. Busin Moscow, Nauka, 1966.
- 12. N. A. Fielyayevskiy, "Zemnqya kora v predelakh territorii SSSR" [The
1,4lrtt';, Cru:st Within the Borders of the USSR], Moscow, Nauka, 1974.
l:i. G. N. Hugayevskiy, I. L. Nersesov, V. A. Rogozhina, "Horizontal Inhomo-
j;eneities of the Upper Mantle in Central Asia," IZVESTIYA AiCADDiSI NAUK
10-0:3fi, SERIYA GEOFIZICHESKAYA, 1971, No 6.
14. V. I'. Valyus, "A1Gorithm for Rapid Calculation of Refracted Waves" in
the book: "Algoritmy interpretatsii seysmicheskikh dann,ykh" [Algorithms
i'ar Interpretation of Seismic Data], No 5, Moscow, Nauka, 1971.
J.S. Vrinek, A. zutopek, V. Karnik et al., "Standardization of the Scale
of Mar;nitudes," I7.~t/r,TIYA AICADEMII NAUK SSSR, SERIYA GEOFIZICHESKAYA,
1962, No
lG. A. Itit:;crnii, eci., "Verkhn,yuya munti,ya" [T1ie Upper Mantle], Moscow, Mir,
1975.
17. L. P. Vinnik, "Isuledovaniya mantii Zemli seysmicheskiMi metodami"
[S,tuclie: of the Mrintle of the Earth by Seismic Methods], Moscow,
Nuuku, 1976�
18. L. 1'. Vinnik, A. A. Godzikovskaya, "Lateral Variations of Absorption in
Ltie UE)Exr Mantle 13enersth Asia," IZVESTIYA AKADEMII NAUK SSSR, FIZIKA
'l,EML[, 1975, No 1.
162
FUk OFFrCIAI. UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
A'Uk OFF'fCIAL Utii; (1NGY
11), L. I'. Vlntilk, A. A. Liilck, "'i'ecl,urilc lnl.crprutration nC the Ueep-Level
Structurc of the Pamirn," I'LVt5mIYA AKAUEMII NAUK SS3R, CE07'EKTnNIKA,
1975, No 5.
20. L. 11. Vinnik, A. V. Niknla,yev, "Velocity Crnsg Section of the Lower
Mantle from I7irect Measurementg of dt/dA," IZVESTIYA AKADEMII NAUK SSSR,
FIZIKA ZEMLI, 1970, Ho 11.
21. I, Ye. Volkov, T. B. Yanovskn.ya, "On Uetermining the Velocity and Quril.ity
Fnctor in the Upper Muntle f'rom Amplitude Curvea" in the book: "Mashinnyy
unalix teifrnvykh seyamicheskikh dannykh. Vychiolitel'naya seysmologiya"
(Mnchine Malygis of Digital Seismic Dgta. Computational Seismology),
Ho 7, Moscow, Nauka, 1974.
22, I, S. Vo1'vovskiy, "Seyamicheskiye inaledovaniya zemnoy kory v SSSR"
(Seismic Studies of the Earth's Crust in the USSR), Moscow, Nedra, 1973.
23. T, N. Calkin, A. V. Nikolayev, Ye. A. Starshinova, "Fluctuations of Wave
Characteristics and Minor Inhomogeneity of the Earth's Crust," IZVESTIYA
NCADEMII NAUK SSSR, F'IZIKA 7.EMLI, 1970, No 11.
24. M. V. Gxovskiy, V. N. Krestnfkov, N. N. Leonov, et al., "Map of Most
Recent Tectonic Movement.n of Soviet Middle Asia," IZVESTIYA AKADEMII
NAUK SSSR, SERIYA GEOFIZICHESKAYA, 1960, No 8.
25, t. V. Gorbunova, N. V. Shutorneya, "Calibration Curve for Determining
the tdngnitude of Earthquake3 from PKPIKP Waves," IZVESTIYA AKADEMII NAUK
SSSR, H'I ZI KA ZEMLI, 1976, No 7. .
26. B. Cutenberg, "Fizika zemnykh nedr" [Physics of the Depths of the EarthJ,
Moscow, IL, 1963.
27. A. A. Dergachev, "Evaluating the Q of the Earth's Crust at Tuva" in the
book: "Voprosy seysmichnosti Sibiri" [Problems of Seismicity of Sfberia],
part 1, Novosibirsk, Institute of Geology and Geophysics, Siberian
Department of the Acadetty of Sciences USSR, 1972.
28. V. Din, "Correction of P Wrsves and Group Adjustment," "Trudy Instituta
inzhenerov elektroniki i radiotekhniki" [Proceedings of the Institute
of Engineerc of Electronics and Radio Engineering], No 12, 1965.
29. V. V. Zhadik, A. A. DerEachev, "Measurements of the Qual.ity Factor of
the Earth's Crust from Recordings of Microtremors," IZVE3TIYA AKADEMII
NAUK SSSR, FIZIKA ZEMLI, 1973, No 2.
30. V. N. Zharkov, L. N. Dorofeyeva, V. M. Dorofeyev, V. M. Lyubimov, "Zone
of Reduced Values of the Dissipative Function in the Shell on the
Boundary With the Core," DOKLADY AKADEMII NAUK SSSR, 1974, Vol 214, No 4.
163
FOR OFFICIlt:. USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'dk UF'FtCtAL USE nNLY
;il. N. l,
nytaev, L, 1-1. Ldnr-.nuhtuyn, N, G, Markovg et a1,, "mectonics of
Monga.tiu" in Lhe bonk: "mektdnika Ura].n-Mnngol'skngn skladchatogo
poynsoct,' Mdscow, Nauka, 197b.
:32, K. K. Zapol'skiy, "ChI55 F'requency-5elective 5eismic Stations" in i;he
book ; "Lknperimentnl' nqyn seyemologiya" [Experimental 3eismology] ,
Moncow, Naukn, ly'tl.
33. K. K. 7,n,po1'skiy, I, L. Nersesov, T. C. Rgutian, V. 7. Khalturin, "physi-
cril ilrinciplec nf' Msgni.tude C1agsificution of Earthquakes" in the book:
"Mugnituda i energeticheakaya klassifikatsiya zemletryaseniy" (Magnitude
und Energy Clagsification of Earthquakes), Vol 1, MS53S, Acadeny of
5ciences USSR, 1974.
:ih. L. P. 'Lonenshtuyn, "Model of IDevelopment of the Geogynclinal Frocess" in
Liie book: "7'ektonikg I1ralo-Mongol'skogo skladchatogo poyasa," Mosenw,
Nntalca, 197h.
35. "Local Structure of Turbul.ence in an Tncompressible Liquid at Very High
Rrynolds Numbers," bOKLADY AKADEMII NAUK SSSR, 1941, No 30.
36. A. A. Lukk, I. L. Nersesov, "Structure of the Upper Part of the Earth's
Mantle from Observntions on Earthquakes with Intermediate Depth of the
F'ocus," DOKLADY AKADEMII NAUK SSSR, 1965, Vol 162, No 3.
37. Ye. PJ. Lyustikh, "I:ostasy and Isostgtic Nypotheses," "Trudy Geofizi-
che3kogo instituta Akademii Nauk SSSR" [Proceedings of the Geophysics
Institute, Acadecry of 5ciences USSR], 1957, No 38.
38. "Magnituda i energeticheskqya klassifikatsiya zemletryasneniy," collection
of articles, Vol 1, 2, Moscow, MSSSS, Acaderry of Sciences USSR, 1974.
39� L. N. Malinovskaya, "Spectral Amplitude Curves of P Waves" in the book: -
"Algoritmy interpretatsii seysmicheskikh dannykh. Vychislitel'nqya
seysmologiya," No 5, Moscow, Nauka, 1971. .
40. 11. Molnar, T. G. Rautian, V. I. Khalturin, "Experie:+ce in Studying the
Spectra of Local Earthquakes of the Garm Region" in ti:e book: "Sovetsko-
amerikanskiye issledovaniya po prognozu zemletryasneniy," Vol l,
Dushanbe, Donish, 1976.
41. Y�. P. Neprochnov, "Results of Deep Seismic Sounding on the Black Sea"
in the book: "GSZ v SSSR" [Deep Seismic Sounding in the USSR], Moscow,
(lostekhizdat, 1962.
+2. .I. L. Nersesov, T. G. Rautian, "Kinematics and Dynamics of Seismic Waves
zt Distances ur to 3500 km from the Epicenter" in the book: "Eksperi-
metital'nqya seysmika" [Experimental Seismics], Moscow, Nauka, 1964.
164
FOR OFFICIIdL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
Fok OrFtc;tnL uSN: OrrLv
10. I, L. NEruennv, A. V. Nikoluyev, Ye. N. lu'edovu, "mhe Ngture of Norizontal
fnhomo,?,(,nrait;y al' the Mii.nl,l.c, of' rfie f;rlrth Crom lo'einmie bata," DOKLAbY
11Kl11*;MI f N/111K 19'f;1, V()I ;'U'(, Nt, h.
44. A. V. Nikdlayev, "Seyumiku neodnorodnykh i mutnykh sred" (Seismics of
Inhomogeneaun and 7'urbid Media], Mdscow, Nedra, 1972.
115. A. V. Niknlayev, F. S. 'I'regub, "i2egults nf 7nvestigation of a Sta-
tisticnl Model nf the Earth's Cruat," DnKLADY AKADEMIT NAUK SSSR, 1969,
vol 189, No 6,
46. A. Nurmagambetov, "nomping of Seismic Wavee and Energy Classification of
Egrthquakes f'rom Ubservations by ChI55 Equipment" in the book: "Magnituda
i energetichcsk qya klaanifikatsiya zemlptr,yagneniy," Vol 2, Moscow,
MJJSJ, Acaderr~y of Sciences USSFt, 1974.
IVf. A. hurmugrunbetov, T. G. ituutian, V. i, Khalturin et al., "Dependpnce of
the Spectra of .ri@i51TIfC Oscillations on the Energy of Earthquakes" in the
book: "Vopro:sy kolichestvennoy otsenki seysmicheskoy opasnosti" [Problems
of Quuntitntive Evaluation of Seismic banger), Moscow, Naukg, 1975.
48. N. I. Pnvlenkovu,"Voluuv,y,yc polya i mddel' zemnoy kory" [Wave Fields and
a Model of ttie Earth's Cru3t], Kiev, "Naukova dumka," 1973.
49. I, p. F'usechnik, "Kharakteristika seysmicheskikh voln pri yaderr~ykh
vzryvnkh izemletryasneniyakh" [Characteristics of Seismic Waves in
the Cuse of Nuclear Explosions and Earthquakes], Moscow, Nevuka, 1970.
50. K. S. Ponamorevu, "On Propagation of Diffusion Sound," UCHENYYE ZAPISKI
KURSKOGO PEDACOGICHESKOGO INSTITUTA, 1969, No 54.
51. F. Pre3s, "Internal Structure of the Earth from Data of Theoretical
Models" in the book: "Priroda tverdoy Zemli" [Nature of the Solid Earth],
Moscow, Mir, 1975.
52. "Priroda tverdo,y Zemli," Moscow, Mir, 1975.
53. N. N. i'uzyrev, ij'. V. Krylov, B. P. Mashen'kin, "Metodika rekognos-
tnirovochnykh F;lubinnykh seysmicheskikh issledovaniy" [Methods of
Reconnni.ssance Deep-Levcl Scismic Studies], Novosibirsk, Nauka, 1975.
54. A. I. Nuzaykin, V. I. Ktinlturin, "Hodograph of the Maximum Phase of a
Rnyleigti Wave at Distances up to 3500 km" in the book: "Magnituda i
energeticheskaya klassifikatsiya zemletryasneniy," Vol 2, Moscow,
M.^,SO;, Acuderty of Sciences USSR, 1974.
55� Ye. F. Savarenskiy, N. G. Val'dner, "Lp, and Rg Waves of Earthquakes of
the Black Sea Basin, and Some Considerations on Their Nature" in the book:
"Seysmicheskiye issledovaniya" [Seismic Research], No 4, Moscow, Acadeiry
of Sciences USSR, 1960.
165
FOR OFFICII,L USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
F'Olt UNF1(;IAL IISL c1NLY
1)f1. Yc. N. c;rflova, ";1linLi L "Lrucl,ilrc
i;enoll,y o f' thc! Crtuib iin(l Mslnt le,"
ZLMLI, ly'f h, No 12,
oC t 11 Wri.ve rind I101.1rnnh,nl lnhomo-
1~'l,Vi~;;~''1 YA I~KAbi!~fiT NAUK "~~Sit, A"IZIKA
57. D. I. :;ikharuliclze, "'I'he Nature or Lg, and Hg Waves and Ittvestiggtion of
the 5tructure o1' the Earth'n Crunt," "mrudy Tnotituta genfixiki AN OSSR"
[i'roceedings dt' the Institute of Cenphysics, Acgderty of Sciences of the
Genrgian SSfi], 1963, Vol 21.
5fi. S. Smlt, "Nonidenl Elasticity" in the book; "Verkhny$ya mantiya" [The
Upper Murt1.e j, Moscow, Mi r, 1975.
'59. S. L. "olov'yev, V. B. Shein, "Intensity of an Egrthquake from Data
oC Far En3tern and Continentnl "tutions of the U5SR" IZVESTIYA AKADEMII
NAitK SLHIYA GEOF'IZIC}IESKAYA, 1959, Nd 9.
60. V. T. Khalturin, N. B. Urusova, "Evaluation of the Absorption of Longi-
tudlnal and '1'ransverse Waves from Observations on Local Earthquakes,"
"'I'rudy Tnstituta fizfki Zemli Akademii Nauk SSSR" (Proceedings of the
tn:;titute of 1'hysics of the Earth, Acadetty of Sciences USSR], 1962,
No 25.
61. V. .L, Khalturtn, "Absorption of Seismic Waves in the Earth's Crust of
Norttiern Tynn'-Shan" in the book: "Eksperimental'naya seysmologiya,"
Moucow, Nauka, 1971.
G;,. M. 'I'; ibul'chik, "IYekotnryye chislennyye metody analiza struktury
::e,ystnogramm i opredeleniya srednikh kharakteristik zemnoy kory" [Some
Numec�lcul Method3 of Mul,yzing the Structure of Seismograms and Deter-
mining the Average C}iaracteristics of the Earth's Crust], Candidate's
Dissertation, Novosibirsk, Institute of Geology and Geophysics, Siberian
Department, Academy of Sciences USSR, 1968.
63. L. A. Chernov, "Ra:;pi�ostraneniye voln so sluchaynymi neodnorodnostyami"
[I'ropugution of Waves with Random Inhomogeneities], Moscow, Acadetty of
Sc iences l1 SSR, 1958.
611. I,. A. Ciiernov, "Volny v sluchayno-neodnorodr~ykh sredakh" [Waves in
Iintuirnnly icihomogeneou:. Media], Moscow, Nauka, 1975�
6,. :1. :'hul'tn, "Formation of the Continental Crust of Paleozoic Folded
Br, t t:> >Lnd 'I'he i r Pre:;ent Structure" in the book : "Tektonika Uralo-
Mongo.l'::kn1to :,klaclchatogo poyasa," Moscow, Nauka, 1974.
06. V. U. Enrnan, T. G. Simbireva, "Vertical Shifts on the Nimich Proving
Graundo, and ttie Mechanism of Foci of Nearby Earthquakes," "Tezisy
dnkladov V:;eso,yuznogo soveshchaniya po sovremennym dvizheniyam zemnoy
kary" [Abstracts of Reports to thc All-Union Conference on Recent Move-
ments of the Earth's Crust], Alrt:rt-Ata, 1971.
166
FOR OFFICIAL USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FnR bFFICtAL U5C ONLY
67. T. h. Ynnuvnkayu, 0, V. Unllkovn, Yu, A. "urkov, 'Amplitude Curves nf
Wuven" in l,ho book; "Vdrrony dinrunir.henkoy teorii raaprostraneniya
r,cynmiChenktkli vdln" (Problcron of the Uynrunic '1'henty of 5eigmic Wave
Propagntion), No 7, i,eningrad, Leningrad State University, 1964,
68. T. B. Yanovaka,ya, "Some Data dn Seiamic Waves in the Earth's Core" in
the book; "Algoritny interpretataii seyamicheskikh clannykh," No 5,
- MUBCOW, Nauka, 1971.
Gv. R. D. Adcunn, E. R. Lngdahl, "P-Wave Velocities in the Earth's Core: an
Interim Eteport," PHYS. EAn'I'H AND PLANET. INTERxORS, 1974, Vol q, No 1.
70. K. Aki, "Scattering of P-Waves Under ttie Montana LASA," J. GEOPNYS. RES.,
1973, vol 78, No g.
71. U. L. Anderson, A. Ben-Menahem, C. B. ArchammUeau, "Attenuation of Seismtc
Energy in the Upper Muntle," J. CEOPHYS. RES., 1965, Vo1 70, No 6.
72. E. V. Artyushkov, "Iiheological Properties of the Crust and Upper Mantle
According to Data on Inostatic Movements," J. GEOPHYS. RES., 1971, Vol 77,
Na
73. C. B. Arcliambeau, E. A. F'linn, D. C. Lambert, "Fine Structure of the
Upper Mantle," J. GEOPNYS. RES., 1969, Vol 74, No 25.
' 74. R. G. Bacer, "F'reliminary Study for betermining Magnitude from Lg,"
A;ARTNQUNCE NOTES SEISMOL. SOC. AMERICA, 1967, Vol 38, March-June.
- � 'r'~. B. A. l3olt, "Entimntion of PKP Travel Times," BULL, SEISMOL. SOC.
I1MrRICA, 1968, vol 58, No 4.
76. I. Capon, R. J. Greenfield, R. J. Kolker, R. T. Lacoss, "Short-Period
Signal Processing Results for the Large-Aperture Seismic Array," GEO-
PHYSICS, 1968, Vol 33, No 3.
77� U. S. Carder, D. W. Cordon, J. N. Jordan, "Analysis of Surface-Foci
Travcl Times," BULL. SEISMOL. SOC. AMERICA, 1966, Vol 56, No 4.
'ff;. E. K. Crtirpenter, Y. P. Marshall, A. Douglas, "The Amplitude-Distance
Curve Cor :ihort-Period Teleseismic I'-Waves," GEOPHYS. J., 1967, Vol 13,
No 1-3.
'f'). M. A. Cliinnery, M. N. `I'okstSz, "P-Wave Velocities in the Mantle Below
700 km," DULL. SEISMOL. SOC. AMERICA, 1967, Vol 57, No 2.
8U. M. A. Chinnery, "Velocity Anomalies in the Lower Mantle," PHYS. EARTH
AND YLANET. INTERIORS, 1969, Vol 2, No 1.
81. J. Cleary, "Analysis of the Amplitudes of Short-Period P-Waves Recorded
by Long-Range Seismic Measurement Stations in the Distance Range of 30�
to 1200," J. GEOPHYS. RES., 1967, Vol 72, No 18.
167
FOR OFFICIe,L USE U[VLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
I~01( 010Fl(, I AI, USI: c1NLY
tfP. I). ,1. Cc,rbiuhlcy, "Mu1.t;lple-Array Meagurementg of tihe p-Wave Travel-mime
Uerlvntive," ciL01'FfY5. J. ttnY, ASmfiON, 50C., 1970, vnti 17, No 1,
83. A. M. Uainty, M, N. 'Ibknft, K. R. Andernon et a1., "5eismic Scattering
rsnd 5hulLow Structure oC the Monn in Oceanus prncellgrum," MOON, 1574,
Vol y, No 1/2.
fllt, S. K. bey-Sarknr, H. A. Wigginn, "ilpper Mgntl.e 5tructure in Western
Canxda," J.GEOt'lIY" . fiLS 1976, Vo1 $1, Nn 1-2. .
85. A. 61. Uziewonski, H. A. Nsddon, "'1'he Rad3us of the Core-Mantle $oundary
Cni'erred f'ron 7'ravel 7'ime and r'ree 05Ci11ation Data: a Critical fieview,"
I1IIY;.;, CAR'I'H ANb PLANE7', INmERIOHS, 1974, Vol 9, No 1.
86. J. P. I;vernden, "Mngnitude beterminations at Regional and Near-Fiegional
bistances in the United States," BULL. 5EISMOL. SOC. AMERrca, 1967,
Vol 57, No 4.
67� B. rutenberg, C. F. Richter, "Magnitude and Energy of Earthquakes,"
ANN. CEOL., Rome, 1956, Vol 9, No 1.
88, A. L. flttles, J. Cleary, J. Roberts, "Velocity Distribution in the Lower ~
Mant.Le," BULL. SEISMOL. SOC. AMERICA, 1968, Vol 58, No 6. i
89� 1I'.:1.f{unks, M. Wyss, "Z'he Use of Body-Wave Psectra in the Determination
of Seismic-Source Parameters," BULL. SEISMOL. SOC. AMERICA, 1972, vol 62, No 2.
yU. H. S. Hasegawa, "Short-Period P-Code Characteristics in the Eastern '
Canudian Shield," BULL. SEISMOL. SOC. AMERICA, 1970, Vol 60, No 3.
91. E. flerrin, "Introduction to 1968 Seismological Tables for P-Phases,"
BULL. ;EISMOL. SOC. AMERICA, 1968, Vol 58, No 4. !
92� 1[. M. I,yer, I. }i. Nealy, "Teleseismic Residuals at the LASA-USGS Extended
Array and Their Interpretation in Terms of Crust and Upper Mantle
Structure," J. GEOP11Y S. RES., 1972, Vol 77, No 8.
93. B. Jttns:;on, E. S. Ifusebye, "Application of Array Data Processing Tech-
niques to a Network of Ordinar,y Seismograph Stations," PURE AND APPL.
croruY;, 1968, Vol 69, rto ii . '
)lt. L. H. lohnson, "Arra,}r Measurement of P-Velocities in the Upper Mantle,"
J. Ch:OPEiYS. RES,, 1967, Vol 72, No 24.
95. L. H. Johnson, "Array Measurements of P-Velocities fn the Lower Mantle,"
13ULL. :;EI'o'MOL. SOC. AMERICA, 1969, Vol 59, No 2.
yG. K. L. Knila, D. Sarkar, "P-Wave Amplitude Variation with Epicentral .
Uiotance and the Mngnitude Relacions," BULL. SEISMOL. SOC. AMERICA,
1975, vol 65, No 4.
168
H'UR OFFICIAL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
FOk UCFCC f AL USN: t)NLY
97. Ii. Kunnmory, "ALtentani:ion oC p-WuvEC in the tlpper nnd Lower Mgntle,"
H111,I� HAli'i'11QU/1KN. 1' Nil7', 'i'OKYn UN1V,, 1967a, Vol 115, pt, 2,
yU, II . Kuncuuory, "UniRr Mantle "t,ructure from Appurent Velncitie$ or p-Wave s
Ilrcorded t1: Wukaywnfs Micro-Earthqunkv hb;;ervatnt;y," BULL, EAIimNQUAKE R&".
t NST, 'I'OKYU UNIV., 19671), Vo1 h5 , pt . 3,
99. H. Kfinrunory, "Spectrum di' 1' iind PcP in Cielation to the Mantle-Core
ISoundnry und Attenuntion in the Muntle," J. CEbPNYS, RES., 1967C, Vol 72,
_ No 2. 100. II. Kunamnr,y, "14)rr.Lrurn nf' ;;liort-Period Core Phttseq in Relation to the
At,trnuutinn irt the Mnntlr," '1, GEOi'F(YS, EiES., 1967d, Vo1 72, Ho 8.
101, lI, Kanumory, "Vcl.ocit,y and Q oC Mantle Waves," pHYS, EARTH AND pLANE'.C.
I Nmh;(ZI0(7 19'(0, Vn1 2, No b,
LUP, B. I,. N. KenneLL, "7'tic lO.f'1'rr,ts of Attenuation on Seismogrems," BULL.
;;LI:;MO1,. SOC,/1MCRI C11, 1975, Vo1 65, No 6.
103, R. L. Kovuch, L. Mderaon, nAttenuation of Shear Waves in the Upper
and Lower Mantle," bULL. SEISMOt,. SOC. AMERICA, 1964, Vol 54, No 6.
104. G. A. McMechan, "/lmplitude Constraints and the Inversion of Canadian
Shield Early fiise Explosion Datu," nULL. SEISMOL. SOC. AMERICA, 1975,
Vol 6$, No 5.
10>. S). M.iy;unura, "beterminat;ion of Hod,y-Wgve Magnitudes for Shallow Earth--
quailce:; In ttie New Zeulund and Macquurie Loop Region Using PKP Data,"
I'liY;t. EAIi'I'il ANU I'LAKI?T. INTI:ItIURS, 1974, Vol 8, No 2.
1U6. 0. W. Nuttli, "Ttie IUnplitudes of Tele.^,eismic P-Waves," BULL. SEISMOL.
Snc. nrERr Cn, 1972. vol 68, rto 1.
107. M. Nir.izl, L. D. Anderson, "Upper Mantle Strticture of Western North
11mor�ir.a Crom Apprirent Velocitiey of P-Waves," J. GEOPHYS. RES., 1965,
vot 70, Nn la.
108. R. )ic,l,in:.on, I(. M. l,yAr, "'1'emporal and Spatial Variations of Travel-Time
Itc:;icluml:; in Centrnl Ca1i['ornin for Novayn. Zemlya Events," BULL. SEISh10L.
00C. M1E1iICA, 1976, Vol 66, No 5.
lU). R. : rito, A. F. F;spinozn, "Rua.lit,y Far.tor Inversion Determination from
the Mnl,ysi r, of fiod,y Wuve Data," Yt. 1, PURE AND APPL. GEOPHYS., 1967,
Vol 67, f)p 1-26.
lUl. M. K. R. R. Julii;n, "Amplitude-Distance Calihration Curve ,
I'rurn 1)cr.prPocu^ Narthquake: Ri1LL. ,^,EISMOL. SOC. AMERICA, 1976, Vol 66,
No 5. �
169
FOR OrFIM:. UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-44850R000100014444-1
rnlt UI~P't(;fAL USL UNLI'
I Il. I). W. H.
P. Mrroii, 11,
W. KiIiIS,
"An Arrn,y 11,tiiciy
af' I'-Wnve
Vclut!iLl(?u tu tlm
Miuil,lL,
'1'r-unnitio
n 'Lnne becieath
Northeaetern
n�strai.in," IiULL.
SCtSMOL, SnC,
AMERTCA,
1974, voi 64,
No 6.
114. H. L. 5trcet, "Sculing Northeaatern United Stateg-Southeastern Canadian
l-:rsrthqunken l~y 'I'heiv L[; Wnve.3," bULL, SEISMOL, SdC, AMERICA, 1976, Vol
66 , No 5.
113. T. L. 'I'eng, "Atterrtaution of Bocly Wttven and 7'he Q. Structure of the
Muntl.e," J. (ItL.UhiIYS. FiLS,, 1978, Vol 73, No 6.
101. W. I4iutctiei�, T. S. Hunks, "Source Parametei-s of Southern California
Earthquakes," J. CCOPtIYS. RES., 1973, Vol 78, No 35.
115. M. '1'sujiura, "Ref;ional Vurintion df p-Wave Gpectrum (1)," BULL. EAfiTii-
(JUAKL NES. I NS'I'. 'I'OKYO UAIV., 1969, voi 47, No 4.
.116, f). L. 'I~ZCker, I. N. Brune, "Seismograms, 5-Wave Spectra and Source
Pnrarnetero for Aftershocks of the San Fernando Earthquake" in: "San
i'ernando, Californig Earthqualce of February 9, 1971," Vol III, Washing-
t,on, 1973.
1.17. K. C. Veitti, C. E. Clawson, "Magnitude from Short-Period P-Wave Data,"
IIULL. SEI0"h10I,. ; OC. AMERICA, 1972, Vol 62, No 2.
118. J. Wesley, "Diffu:;ton of Beismic Energy in the Near Range," J. GEO-
I'kIYS. RES., 1965, Vol 70, No 20.
11.9. H. A. WiLt;in:i, D. V. Fielmberger, "Upper Mantle Structure of the Western
United States," J. GCOP}[YS. RES., 1973, Vol 78, No 11.
120, A. D. Zoltun, R. P. Masse, J. P. Gurski, "Regional Attenuation of
;liort-Period P- and S-Waves in the United States," Geophys. res.,
1975, vol 40, No 1.
COPY[t1GIf'I': Izdztel'stvo "Nauka", 1978
66 lp
CSO: 81)01/0481 - END -
170
FUR OFFTCIAL UtiE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000100010044-1