l?b,_

ELSEVIER Tcctont~ph~ sic,, 288 ( I t~gX) I - 16

TECTONOPHYSICS

Seismic studies around the Kola Superdeep Borehole, Russia

Y,V. G a n c h i n ~'::, S .B. S m i t h s o n d, I .B. M o r o z o v ", D . K . S m y t h e h, V.Z. G a r i p o v ~

N . A . K a r a e v 'J, Y. K r i s t o f f e r s o n "

" I)cpartment o f (h'olo. W and Geophv.~ic~. Unit er~it 3 0 l !,Vw,nin.e,. l.oramw. WY X2071-.'Ot)6. I 'S.t

l, (;e,,h).e~ mid ,41~ldied (.;e~,h*g3: Unirur~itv ,~1 (Ha.won: (;la.~,,,ow (;12 ,~QQ. ~ :K

' Run,tun State ( 'ontmtttcc ,m (;uo/og.~: 4-0 B (iru;m.~ka]a. 123242 .$h,.~ctm; R u ~ i , ,i VIR(.;-'Rtulgeo[i-tka'. 21) l.~!ian.~owqa. /9 . .¢#/9 St. Putm'.sl~uce. Russia

' In,/tittle o / S o l i d I:arth Ph~i¢'.v l'nn'(v:sit~ O/Bcrtwn...tll('.~l..11. 5007 Ih'/~en. :\:,nL~ o~

Abstract

The Kola Superdeep Borehole (SG-3) proxided an ideal optx~rtunity to lest hypotheses on the origins ot crustvl retlections and on the presence and seismic expression of fluids in the upper crusl. The ahemalive sources of crustal reflections include compositional changes, shear zones, fluids, and metamorphic facies changes, all of which arc represented at the well. Both the 3g-kin-long CDP section and the borehole VSPs in the range 2.2 6.0 km demonstrate the presence of retlections from dipping comlx)sitional layering, shear zones, and tluid-filled zones. Subhori/ontal rellectivity zones are interpreted as horizontal fluid-lilled fracture-t}pe reservoir rocks. Results suggest the presence of fluids dm~.n to a depth of at least 12 km in the upper crust: the presence of these tluids lowering seismic ~elocity causes estimates of upper crustal composition to he too felsic..C, 1998 Elsevier Science B.V. All rights reser,,ed.

Kevword~: horehole, CDP" VSP: retlections: Kola

1. Introduct ion

The geologic interpretation of crustal seismic data is still speculative and raises many questions concerning the nature of crustal reflectors (e.g. Fuchs, 1969: Smithson et al., 1977: Christensen, 1989: Valasek et al.. 1989; Kremenetsky, 1990; Pavlenkova, 1991: Mooney and Meissner, 1992; l,evander cl al., 1994: Rudnick and Fountain, 1995). Mooncy and Meissner (1992) suggest a large number of proposed models for the origin of reflections from the deep crust, including: ( 1 ) differing physical prop-

• Ct~rrc~ponding author, l 'ax : ~ I (307 ) 766 -6679 : [ : .-mai l :

.~ u ri (a u~ ~ txcdu

erties of rocks: (2) lithologic variations: (3) seismic anisotropy: (4)juxtaposit ion of difl~'rcnt rock types caused by faulting: (5) ductile shear zone,,: (6) zones containing lluids: (7) ductile llow: (8) molten and partially molten bodies m the crust.

The Kola Superdeep Borehole (SG-3) in the Kola Peninsula, northwest Russia. is one of the l'cv, sci- entilic boreholes [e.g. KTB deep drilling project in Germany (Hohrath el al., 1992). Gravberg-Siljan (Juhlin. 1990), Cajon Pass (Rector, 1988)l where hy- potheses on the nature of crustal rellectivily can he tested against in situ geological control. Thc SG-3 borehole has been cored down to the depth of 12.25 km. The comprehensive investigation of its core was carried out together with different kinds of geophys-

tit)40- I ~5119SIS 19(X) ." 19t,~X Else,. icr Science B.V. All right, rc~,cr', cd. PII S ( ) ( ) 4 { ) I q 5 1 ( t ~ 7 1 0 0 2 , 4 0 - 1

2 Y.V. ( h m ( h m ('t al. / li'( lonol>hvsi(~ 2,'~,'~ ~ Ioo,'~'j I 16

ical logging experiments in the borehole (Kt)zlovsky. 1987). Single-fold retlection shooting and deep seis- mic sounding (DSS) were conducted in the SG-3 borehole region, but until now it has lacked deep- crustal, common-depth-point (CI)P) seismic prolil- ing. To fill this major gap in the knowledge of the region, a multinational seismic experiment was per- formed during winter and spring of 1992. The data acquisition was carried out in cooperation with the Ministry of Geology of thc Russian Republic, thc Institute of Physics of the Earth of the Academy of Sciences, Moscow, and the Universitics of Wyoming, Bcrgcn, Glasgow and Edinburgh. In addition to a -,38-kin-long CDP profile, two deep VSPs (up to 6 km deep) and 4 shalh)w VSPs (0.5 km deep) were acquired during this experiment.

The purpose of this publication is to o,)rrelate seis- mic retlections (present on both VSP and CDP data) with the known geological interfaces ahmg tile SG-3 section. This analysis also involves the effect of lluid- lilled voids in association with subhorizontal rellec- tivity detected in VSP and CI)P data in tile SG-3 vicinity.

2. Main geniogical and geophysical features of the SG-3 borehole region

The Kola Supcrdeep Borehole is located in the northeastern part of the Baltic Shield inside the Early Pmterozoic Pechenga-hnandra---Varzuga Greenstone Belt, which is surrounded by l.atc Archean terrains. The main metamorphism, deformation and thrust- ing in the Belt have occurred ca. 1.95-1.85 Ga (Melezhik and Sturt, 1994), and its evolution is compatible with interpretations in terms of mc)dern plate tectonics (e.g. Berthe[sen and Market, 1986: Melezhik tuld Sturt. 1994). The early Proterozoic Pechenga structure (Fig. It, one of thc largest zones comprising the Greenstone Belt, can be character- ized as a synform consisting of two distinct units: the North and South Zones separated by the Poritash system of faults. The North Zone is an isometric SW dipping (20 60 °) half-graben separated into sev- eral blocks by a widespread, transverse fault systcm (Melezhik and Stuff, 1994). The South Zotle con- sists of steeply dipping (40-90°), strongly folded and imbricated rock assemblages.

Stratigraphy of the whole Pechenga structure is

characterized by the cyclical build-up of sedimen- tary-volcanic formations. Each cycle begins with thin sedimentary unit and ends with thick volcanic deposi- tion (Mclezhik and Sturt, 1994). Approximately 7()+~i. by area. of the structure is occupied by volcanic rock,,. The metamorphic grade in the SG-3 section gradu- ally increases with depth from prehnile-pumpcllyile to amphibolite facies (Fig. 2), though most commonly it is greenschist in the Proterozoic.

Pillowcd and naassi,,'c tholeiitic basahs and fcr- ropicritic lavas dominate in the upper f'rotcrozoic part of the SG-3 section (0.0-4.5 km deepl and are characteri/ed by P-wave velocity of 6.7--6.8 km/s. Subalkaline basalts with foliated and brecciated tex- ture (!./), = 6.1-6.3 kin/s) are more widely spread from 4.5 to 6.8 km in the SG-3 section. The most conllnon metasedimentary rocks phyllites, tufts, and sandstones are characterized by lower sonic P-'aa~,e velocities ranging from 5.6 to 6.0 km/s. The rest of tile section (6.g 12.2 knl) consists mainly of the I+ate Archean gneiss miglnatite rocks (V t̀ -- 6.0 kin/s) hosting abundant arnphibolite bodies (~,q>-. 6.8 ktn/s) up to 30 nt in thickness. The major shear zone at a depth of 4.5 km where velocity drops from 6.7 to 5.4 km/s is a boundary separating rocks v+ith massive from toliated textures. No delinite trend in P-wave xelocitv variations can be observed in the depth interval 4.5-11.5 kin. Below 4.5 kin. \'elocit+~ neither increases nor decreases with depth, thot,gh h)cal velocit 5 contrasts reach 1.0 klll/s and nlore. Gneissosity, foliation, and contact surfaces in thc whole SG-3 section are commonly parallel to each other and normally dip southwards at angles front 20 to 60 °.

Among other gases, helium is the onl'v one that was not affected by technogenic lactors during SG-3 gas-logging because it was not present in the drilling mud. in the Proterozoic complex (0-6.8 kin)helium accumulations have a local character. Starting from a depth of 7.7 km, the frequency of an anomah)us amount of helium increases reaching highest concen- trations in between 10.1 and 10.3 km (Fig. 1.66. p. 244 in Kozhwsky, 1987). High gas concentration in the Archean rock complex correlates with the zones of fractured rocks with highly mineralized waters (Fig. 1.77, p. 273 in Kozh)vsky, 1987).

Previous seismic studies in the SG-3 region (e.g. Kaius el al.. 1987: Mints et al., 1<-.187: Kremenetsky.

}i ~ ( ; { m ( h m (.! (11 / l i . ( l ,)m)phv~l(~ 2,~,~ (/oo:%' / /6 ~

~ n f f h ~ r n

LEGEND ARCHEAN ROCKS

a b Gneisses, amphibolites, gramto-gnmsses a) northern unit: b) southern unit

EARLY PROTEROZOIC ROCKS N O R T H P E C H E N G A G R O U P

I '.~.~'kt' ::'

kol

a b

Ahmalahti Formation (polymict conglomerates, sandstones, basalts, basallJc andesJtes )

I Kuetsjarvi Formation ( red-coloured sandstones, dotos[ones, alkaline picntes, basalts, andesitic da~tes )

I Kolasjoki Formation (volcanodastic polymict conglomerates, sandstones, tholeitic basalts )

Pilgujarvi Formation (conglomerates, sandstones, ferrooicntic laves and tufts, acidic lavas and tufts )

a) Ni-Cu beanng 'Prnduc~ve' FonnatJon, and b) acidic !avas and tufts inside Pilgujarvi Formation

S O U T H P E C H E N G A G R O U P

~ Bragino Formation (bimndal picrite-andesite volcanic association, andesitic volcanoclas~ conglomerates and sandstones )

I kp I Kaplja Formation ('black shales', arcos=c sandstones, cherts, thole~itLc basalts )

I N T R U S I V E R O C K S a b

a) Orogenic granites: b) subvolcamc andes~les

G E O L O G I C A L A N D O T H E R S Y M B O L S a b

a) Fau!'4 i0) stnke and dip ol bedgin,q

a b al The Kola-92 CDP line location b) The Kola Superdeep 8oreho!e ',ocatloq

Fig. I. (;ct4o~ic~d ",ketch m~q~ (~1 tile PechenL:d qruclurc. M~,dilicd h'om Mclczhik ~md ?';tur[ (I ()t,~q.

4 Y. U ( ; . n ( h m el . I . / 7~.('1~m~phv.~i( ~ 2,~,'~' ( / 99,'.;i I - I h

SW

8S0

E E . = = > ,

S G - 3 . c ; ~ ~= ¢o E o ~ -~

i '~ r- o 2 . 5 k m 900 950 ._E = ,,, . ¢ . . . . u 0 u, " - -

@

b>,

L ~

° ~

e -

t,-

k,,.

o

..Q

e-

e-)

E

<

4

Fig. 2. Northern part of the CDI" line (Migrated scaled lime scctkm) hI comparison ',,,ill) the I)-wavc acoustic log data lacotl',ljc impedance, reflection coel'licient, and ~,ynthetic scisrnogranl) lind geological mt'orn)~ition h'orn the S(;-++ horchol¢. S.vnthetic seismogram position on lhC section corresponds to the boreholc location. I+itholo~.~ ¢xplatlalion: / = domh)antly massi',c basalt,,,: 2 = I'oliutcd basalts and schists with thin intcrheds of metasedimcntary rocks: .? -- dottlinantl+v Illct~L~,cdimcnt~Lry rocks: 4 = gncinse> with nuhist>, and pla~io-~ranitcs, ttorizonlal line sC~IIICIIIs in th¢ Archcan dcliotc Itlc Ihickcnt amPhibt>litc bodies. RI-R,~ dcl)olc rcllcctitmn di,,cussed in the text.

}': ~,: Gum hin el al. I ' l ) '+' t ,mWfln~ics 2,'~,'~ ; IOq,'fl I 16 5

1990: Pavlenkova, 1991: Digranes el al.. 1996: Carr el al., 1996) testify to the presence of numerous dipping reflections and subhorizontal reflectors con- centrated below 7 km deep (e.g. Pavlenkova. 1991: Mints el al., 1987).

3. 1992 year data acquisition and processing

Approximately 20-fold CI)P profiling provided increased signal/noise, lateral resolution and obser- vation density relative to previously conducted seN- mic investigations in this region. The 3g-kin-long line was shot t'rom south-west to north-east, along a straight track and ended just north of the borehole. Four seismic vibrators were used as a source, and three recording systems were used for data recording (Table 1). Besides the major CI)P line, two deep VSPs were recorded in the Kola Superdeep Borehole from a depth of 2.2 to 6.0 km at 25-m intervals. The source offset was 0.2 km south of the well for VSP-I and 2.0g km north of the w, ell R~r VSP-2.

The CDP line was processed to 15 s using an in-

"lahlc I

"fhc K~fla-92 data acquisition paramete rs

Paramcler Dc~,crJplion

Ree ¢.~link' InMrtllllClll

Ntl l l lhcr o l t'hanllcln ( ',.vvcragc

Ro,. ,,trd i n,.2' t ime ( 'orrc la lcd record length

NillllpIc I'HIC

St+tlrcc lyp¢ Peak ftlrcc

NUiilhci ol \ ihr..d¢+r~,

NUillht.'l ill s"ACCpx

1 :p.,'.\ ucp Ircquenctcx

S~\ ccp lenglh

SOllrt. u' mler\'al ( "',Ullig U f;.It ion

( ic~ph,mc t).pc

Rc~ ci\,.'r illtCl\ al (;cophLmc alr;.l\ Nulilhu'r o f COlilplH1ent~, Nominal o['f,;cl Val'iillJolls

MDS- 10 recording s.',slcm

96 20-fold mmfinal

4(I ,,

20 ~,

( ' I ) R 4 m,: VSP. 2 In,,

Failing .. ihralorx

27.0()0 pound~, 3 4

('1)1! I f l - f~) I-Iz: VSP. 15-0( I I-tz 2() ,,

I(X) rn m'mlh+,al

( )ff-cn,.l

I() I I / ( ] c o , source M D R I ( vertical i lO H / I .RS (hc, riz.ontall 50 m I0 ph,.mex spaced 2.5 rn 3 (I vertical and 2 horizontal) 10(X) 5450 m

Table 2

The Kola-02 ( 'DP prc~.'c.;,.ing paranlclcrs

P,~llillllClcr I)c,,cription

P R ' - ( I ) I ' pro~ c~m, . ,

(_'o r r,,." h.ltJon

Interactive editing Noisc-adapt i ; c l ihcr ing

k,2"rtical I rate qacking

S,.nlhclic I0 6(I I I / ups,.\ccp

"I race kil li n.,.z ~lalion~.ll\ "lllLIChillL'r\" llOIxC

q $0 I I z :.uld if,, halnl,,mic,, LIII~.I

IS I Ix from punlps a! the v, el l~

I)i,. crxit,. ,tack

("1)1' I,r~'-~lm'a ImU es.~in,. , f l~m (al l I ' roMA k" t,,,d.~,

2-D Imc b'comclrv

Rclract i (m stalic~,

I':ir,q-brcnk picks Intcracte .c

Retraction statics I ntcracli\u t)Lll Ulll 320 nl

Spectral balancing Pas~, bund 1(l f i ( t H /

KI.T m~i-,c can,,.'clcr Noixe-adapli , .e Iihcr based on

Ihc cigen-dccoml ' ,os i t ion ol the

d~ll;l L'O\;l['l~lllt'C 1114[[JX

Veltx.'il) unahsi, Repealed three IlIIICN S()RI" tu cummon offset

dIllII~li[l

NM()

I ) M O

lmcrsc N M ( )

Veloci ty anah.~i~ Semhlancc and ( 'VS S()RT m ( ' D P domain

N M ( ) Final \ elocilics

( ' D P stack

Po~t-~tack i.oce~.~in,~t 'all I ' m M A X tool~;

FX-dccon~olu t ion R a n d o m noise climinalion FK and l in i lc-di f fercncc

M ig rat ion

teractive and iterative processing sequence (Table 2). The 1992 Kola VSP data processing w'as reported b\ Carr el al. (1996). Two additional processing steps were added in the present study: (I) interval veloc- ities were obtained using least-squares inversion of travel times (Stewart, 1984); (2) deep reflections on multicomponent VSP records were analyzed using 2-I) ray-tracing for dipping interfaces.

4. Seismic wavefield description

4.1. The .scismi( ,section ahm~ the whole ( 'I)P line

The seismic section ahmg tile whole CI)P line shows numerous rellections, mostly dipping SW (Fig. 3al. The section can be subdivided into Ihrec

6 ) :~ Gan( hin ('1 ul./7~'(l~)m,/~/~wic~ 2A'(k' t/9(,',',;'; / I(>

(a) SOUTH PECHENGA NORTH PECHENGA SG-3 , i

S t a l i o n # 300 4 0 0 _~oo ~,oo " /00 ~,00 ~ O O l 0 . . . . . . . . . . . . . - ~ r 0

• -~ ." ". '- .- .-~ i: • - . -:;. + -. ~ ~ : . , . ~- . - . ~.~,:;.-+~ . - ~ + . ~ : - ' , ~ " ". ~t~-, . ~ : ~ - -: >.- ;'::;'-.?~->..".e" ". . . . . "-'-- : . . ~ :."~.~... +_'.'~>:-" +"-+.~" ,/ - " -.-.-. . . . . ~

. . . . . r...~:=++...:.--.-_... ..-.~.:-:- ... -+ + ..r . . . . .&z -- - ~ e ---+.+ ~ = " + . - . . ..+..~++ ~ . . + ~ e - . . . - - ~ - . - - . ~<m g _ - - - ~ - - ~ ~ ! : , : . . , - - . ' - } : - ~ ; 3 . : : . . ' ~ - ~ , - V ~ , : t ' < - . . . . . ~ . . . : " . ~ ; , ~ . , ; '+,+ .- ~ . . . . . ~..:;-+'K?_,j:¢ ~ ' - + - . ~ . . . g . - - . ~ ? ' . ~ - ~ " , , , ~

- - --.: < ' - - - ] ' - ~ - : : ~ ' - ' - < - ' - . : " • . . . . . . . . + ~ ~ - : * ~ + - - ' : "~ : ~ a ' > ' - ~ * + , ' ~ ~ : i

~ ! 7;- "~'< ~" ~=~--~ '" :~Z;2C"~' -~":~ 'T:~ ~ - ~ . " - - ' " . "

. . . . _ . . . . . . . . . ~ . . . . . . . -_-.: .-,+:. . . . . . . <~...= . . . . , ~- _ =: .+~g_-~., ~ ~ . . : . , = . . . . . . . . . . . . . .

: :~-: .~ : .= - . ~-.-=.-..->--~_-. ; +>~.-::.,-.:~-.:~:.-~ + . - . 4 ~ 7 , ~ g , - ~ :a,-- --:,--..~.-"_=-r--.-:.~ m~:---~-&_~-+-'-'~.~..: -" ~-~--.:--,,~. .-.-.-~-+'.:.

k m .,,+o S W 0 5 10 km ~ O" ,,,,P" N E

' . ,++ ^ ^ . 2

0

( b ) E

5 -

c:i.

1:3

10

15

F--- l l I 2 r-.-q3 1 + m s F--q6 F-Tq8 Fig. 3. Sob, nil,,. '..ccti,.',n u l (mg tile Ku lu ( ' D P lir, c (aL Po,,i-,.ta,,:k l iNi lc d i f fe rence nl igr~l l i tm. N,a ',crlic41 cxag,_'cr~lti,.m, lntcrr , rctulr,.c c ; . t r l o o n I b l s h o , . , . i n g m a i o r l a t i h ] o n L " , a n d t i t h o h + g i c a l t lI1it ' , . / 2 • %i ) t l th l ) u ' d l c n g a ( i l ' t l t i p . I = m c t a b u > , a l t c s a n d m c t a s c d i i n c n t , , . 2

- l}lelaalldi_+~,itu'~+,; . ? - / = N ( u ' t h Pe, ._ 'hu 'nga ( ; r o u p . . ? - lllCt;tl1~l',~t]tL", ;.111(.I 111el;.t~.L'dmlt.'lltn t+l l > i l g u . l a l \ i l 'tH111~ltiu11. -/ = 11"ict;.II'ILP.a]Ic', a n d

nicta ' ,udi incnt~ i l l Kti la>,joki. Kuot , t iars i and .&khnia laht i l :ornlut ion>,; 5 = I ' a r l \ Proterozt l i+ rhcon lo rph ic ,L'ranit(+idn: 0 = ,,\rchcun rock>: 7 ~ I ] l t l l l > ; ~ i t h ~ l l r ( i x t ~, J l l d i c a t i l l 7 r o l a t i x o I I I O \ L ' I I 1 L ' I l l , d o t , d i e d \~. h c l t . " h l l ' c r l c d : ,"g - - , . : o n e s o f i l i l c n s e , , c J l i s l l ) , , i l , , . t ] 4 i a t i o n i l l 3 d i11 \ h H l i l i Z i l l l O l l

, ,ho\v jng ;ippr(+xiniatL' clip direl : lJon: 0 : /,.'llll_,,~ in II1o C r t l ' q t . ' ( ir lckpiHldi l l~ Io ,,Lil+~ht+rJ/tirllal ~,L.i,,nlic rel lccti~ i l l

l a t e r a l l y d i s t i n g u i s h a b l e p a r i s : t h e n o r t h e r l l p a r t w i t h m o n o c l i n a l ( - - 3 0 ° S W ) c l ip o f r e f l e c t o r s v , . ' h ich Ctll

g r a d u a l l y l l a t t e n i n g s o u t h d i p p i n g r e l l e c t o r s ( f r o m 3 0 tilT t h e n o r t h e r n p a r t a t a b o u t s u r f a c e s t a t i o n l o c a t i o n

to 10<'): the central part, which is charactcrizcd h) 8()0: ;.illd the SOtllherll part which interferes with thc

E I'; ( ;am 'h in c! al. I li'( wm<,pln.Wc~ 2,~,~ i Ig~),k'; I 16 7

central part in the vicinity of station number 50(). Reflections v, ith the maximum SW dip (50-60 °) can be found souttl of station 500.

Thc most coherent and continuous SW dipping rc- llections on the section are interpreted as the fault zones mapped on the surface (Fig. 3b). The /.u- oln;.i fault (station 740) manifests transition from rel- atixelv undisturbed layering in the northern part to strongly faulted and fl~lded rocks occurring in the cen- tral part of the section. The Poritash fault zone (sta- tion 520) is ecologically characterized by a system of closely spaced imbricate thrust faults along which the South Pechcnga sequence was thrust onto the North Pcchcnga units. The southernmost part of the pro- lile crosses a rcgJon o[ granitoid intrusion which nlay be allochthonous. The corresponding rcflections from tcctoni/cd contact between granitoids and surround- ing rocks can be found south of station 400.

Numerous subhorizonlal reflections with different degrees of coherency can be observed throughout the ,.vholc section (Fig. 3). Several mechanisms can bc suggested for their generation: (I) subhorizontal dikes: ~21 subhorizontal faults..juxtaposing different rock types: (3) fluid-filled fracture zones: (4) side- swipc. We shall discuss below those of subhorizontal reflections, w.hich appear in the vicinity of the SG-3 borcholc.

4.2. Kola 1992 VSP Irar~lield patteml.~

The thst arrivals observed in the Kola VSPs cot- respond to an inc'ident P-wave of complicated shape due to revcrbciaiJons and nlode conversions ou nu- Illerotlk iilterlaces (Figs. 4 and 5). Next arrives an incident shear wave with velocity about 1.72 times slov,cr than the direct P-wave. It also comprises sev- eral trains of reverberations and mode conversions.

Rel]ected waxes have shorter continuity and less amplitude than the direct arrivals, but still can be easil', recognized like those marked A, B. and C in Fig,,. 4 and 5. As noted by' Karus ct al. (1987), short coherent line-ups of the up-coming waves and the complex structure of the geological medium do not pern|it unambiguous determination of the ref lect ion w a \ e type. But in SOlllC eases, due to

a\aih.iblc geological and geophysical Jnforinalion. tlllalllbJguotlS deternlJnatJon of wave type is possible. .~c\ 0ral represcnlali\e COillinuous reflections 'acre

picked fl~r ray-tracing arialysis in the time range 0.5 4.5 s on the vertical (Z) and two horizontal IX and Y) components of VSP-I and VSP-2 (Figs. 4 and 5). Some possible solutions for the picked reflections are presented in Tables 3 and 4 after comparison of calculated travel times with the observed ones.

The major conclusions from analysis of the VSP reflected wavefield arc: (1) SS and PS reflections dominate over PP- and SP-type: (2) moderately dip- ping (15-40 °) inteH'aces can be found mostly in the Protcrozoic part of the section (up to 6.8 krn dcepl: and 13) subhorizontal reflections dominate in thc Archean part (at least for the northcrnnlost VSP-2L Points 2 and 3 repeat previously reported results (e.g. ( 'arr el al.. 1996: Mints el al.. It)g7: Pavlcnko\a, Igg l ). but point 1 needs to be explained.

"Phe direct shear wave fronl VSP-2 was subjected to polarization analysis by l)igranes et al. (1996L They have observed two cleat orthogonal pc, h.uJza- tions in the iriterval from 4.4 to 6.0 kin. The theoreti- cal polarizations were calculated for several nlodcls. [)igrancs c ta l . (1996) report that only the moclel containing aligned, vertical cracks can explain their observations.

Gibson and I~en-Mellahenl (1c)91) presented Ihc P, .~V and Stl radiation pallerns for lWO case,,: ( I ) isotropic with randomly oriented fractlircs; I'~l anisoiropic with aligned cracks. They have estab- lished some distinct differences in the scattered cli,,- ph.iccment liclds. In particular, considering the in- cident P-~.~,'ave, the backscallered P-v,a',e llclcl I]~i aligned fractures is less than lw:Jcc thai for the isotropic obstacle (figs. 5 and 6 in Gibson and Bell-Menahenl. 1791). ttowever, the shapes of Ihe radiation patterns for SV incidence arc the same. The radiation patterns (Gibson and Ben-Menahenl. leg I) irl case of the anisotropic voJtinle explairl rela- tively strong PS alld S,'~ rellcctions on the Koh.i \:SI "k, (Figs. 4 and 5 t.

4..'~. Wal'~' i~am,rn.~ on tlw CI)I ~ ~~,ction in the 5"(;-.?

ricinitv

Tile seismic v, avelield m the vicinity of tile S(;-3 borehole (Fig. 2) can be subdivided into pat)parts+

• , 3 "3 the SW dipping reflectors ill Ihc time range ()-_._ s and neark, horizontal reflectors belov~ 2.2 ,,. The donlJnant dip of reflectors cross-cutting thc S(I-3

8 Y U ( ; a . ( ' h i , ( ' I a/./7~'('t(,,,/Hh~i('~ 2,'~,'~t/997~i /.. /6

Vertical Z-component

, ) i re<,

~ # ' <f:~}t'(', It+,>. +W/({O+ .f')',,';/,+';+,~;,~+'+ ,' ' L~ ,' t ' " !3;2#,~tl9 . . . . . . . . . . . . . . .

- , ' r l ' = " . . " tl~,, ": I~%l'l" 7 + ' I ' , l . l l l ! t l ' l i l l i { t . l J ~ l ) t l [ ; t I ¢'1" 7 #

: , i l l ~ i l f l U ~ / i ~ ' t~\ ' ' P.IX ' l i ' : ' , ) : /Lt~/ j&l I l l .~ ' l { i ! ' I ! {11 ~' :.:,,~ t.,,...~., ,., ,,,ILL, ;?lf~X~'7,~/!~i~..~,.h,.~,..,,,: ~f.+, fJ{ , , ~ } l ~ . l J ~ ; ~ " l / : . ' l / l l t##//H/ I ! I I

Horizontal X-component x ; . . . ~xx " , ' ~ : ~ ";~ , . . 't ~ t 1 : " ,'(.V"I! f: ~t'f '(.t~"l~ # % ' ~?, ' ,"q; t'r~'~.,'l,% ,,:,.~¢~r{/- It' . - / .t~.".

• t~;'..).¢l(~.~"~.fi LLIl~,, ,Yd'l~ v ~ • I ( ' t . ( , , '~: ' , t l . , , ~ti.'~,',,~.,, '~'~'V~.;,~'r. ,~.'!i ,"~z, rt, ;~ r--':.:./"

~':.;!".~`~i.~.'.~¢¢'nw`~```'~a:~.u~t}*5~%`¢.~`':tr..~x:`~`.U` ~,;~W~X~.~` ;~)~v. ' -7~) i~nh~ t.]:~ A I % ,4. .....

',', ', , " ' , v " l ' i ' l l l ' v,~.v,~ • I l i t ( . r ' • i) ' .~t '~l ' ,( l , " m l , ' : ' g ,a .' ,./ ... " ': ,. 7]li 'i ' iil - . . \ , , . . . . - -

.%%.{/ - .V,"~l l ) , ; : - . : t~ ' , , , " , I , i ' : . , , q i . : l i / ' , f i , " i . , - ( t # . ( ~ h : ~ 1 l ; o ' 7 ~ : I ~ ~ / , ; ]77 ] " % ' u , , , . ~ . ,

.,.,,~4, ~,..,,,:,7~, ~ ,,,~...~,, , S . . - , , , , ~ : , b ,,,..., ..... ~t~.., . . . . . . . . . . . ,,-., ,, ":!,.->'.r,-$,~,'",, ' .,|: +, ; : , ,':.:,' ..,.,~,~ ..,,.,,.,,i <~. ' ~ , , ~ . . - , , . . . . . ,- ,

I I I I

Horizontal Y-component

I i h ' " : I~ I ' l I l l | I l l . ¢ ' : ' i~ l l i l l } '~" ' n i ( ~ l . I l i t i i¢ l i l l I l l . t is . ." " ¢111 I ' "~ ! . ' I : .," , ~ / 1 '

I; i ,V"l ' ?',<; ' (/ti~ "" ;,,.It {.,¢<<,(.tll A ' : , iv , '¢ ' : ;~.~; t , , ,'.'.-If,,r'l):,',Ig'~.~:r, tK!ll lAlif(~'¢l~:~',,t ' l , w . l / , l l . ~ I ~ ] - 4 ~ / i " " . . . . . . I ! I I ! i i , I ~ l l G I ' ' i i I l l I I l l " l } , ' l * l . i l ' t l l l i l I 1 " . ' l i I I l l I l l I i ' l l i " l i l * l ) r ' I . i l # " ", ' < , , '

r:, ' ,. i l l , '( l i i~;,t,!. '~\%\': '¢,tlt;-7i¢":~.~'{ ;W , " i t l l l , ' ) : , l i , , ; ' ;~lv,",, ' , : ' l ' l ( 'dt l l~)l"; f f l , ' t ( l l l l i 'b~'~lt. .~P;.~I~i l7 I I IF~.~,:iI,.././ ' ' " i l l l l ~ i ? " f - . : l T l l ~ l ~ " I I l i l . . '1 I " ' l " i / " l l / ~ I " *i l / ; - V ' t , ~ ' ' I ) I f I . , ' ( . ' . ..,; . ""

I;'ri$7 b ) t d \ % iiJ, ,m)i,~'~,';':",'!.~,~".,'i;:l!lv,)~fflfk"';,"!(;':': :~)k.k ~ i ' ,T t { i ' ,7 ;~ '~ i~ i l l~ ' . l (~ ! i$ : )~ l l I I v ~ j , / ~ U , t,~l ;,~,~7(6t:~lTi'llfi c d;, :'.,'!),]I! ,,~¢;,'.,':]:ir,:;:.gY,,l,-~7i~('?,., ; ~ l . ~ i ' M ) ) ~ ( ( ~ / I ~ / ! . ' L , ; . ~. 4

I i . , i )',!r4 %Q V~)li f ,~) :: "/~;1 ', I?.~}~+:.', ill,~'(,\', ',1? t:, : , , i l / ; , ~11.1 t+~ l l~ , )~ : i~ [ I , ( E ~ I - ! ] g;. ' , , . .. % _ .

I l i,:.; iO ': "i)i~%(~.~'li~'t~ip.~l(i!,v.i,, '.,,., : / r , 0 , lHi / .~}/ I , :~,D~7, l ¢ ~ l t [ ~ ~ q ; ~ . l ~ i ~ J l V / ) ! / Z , k .'.'. ' ' , i t " $1 )W, ' l l l l f r r f ' : ~ l l ~ . Y l f l i . ' ~ : l , w ( ~ # . l l | ' ' , ' l~it i , ' ."~'.~'! f .~, . ' l ' , - : ' ~ l l l t L ' . l t I I i ~ q ~ ( l ( - ~ i l " ~ l i , ' l / , # . . ' t # / I . # " ' : : } "JE'' i 5 I ' [ ! . . . ' r " ¢ 1 . ; i * ' " " * ' ~ • I ° l " l l l~ " l l l ~ l ~ t ~ " ~ ' ; ' ~ " " IX I I l , ,1 . . . . . . T ! r . . . . . . . . I . . . . ~'"~ ....... . ~ t ; ,.,, -~ $~., . i ...... , "., . . . . . . 7#.. r ~. ,~,~ ., . . . . . . . ' . . . . . . ,t'

t t Q t , ~ , , . ~ , f l F / r ~ ," +' ~ i | l : - ? .~ . . , ; L . , , , t l i l - , r / i ~ , ' , . "~v L I I r e ¢ 1 3

I I I I

4 3 2 1 0 T i m e , s e c

Fig . 4. V S P - I t l l r cc -c 'Oml~) l len l d i ~ p l a ) . Sou rce . 0 .2 k i n ~ou th t ' roi l l the h o r c h o l e : hand-pa...~, t i l ter . 22-514 I I z : A ( ; C . I(X)O ins. Arl 'o ' ,~r i n d i c a t e r e f l e c t i o n n d i ~ . c u s n e d i l l t h e t e x t a n t i in te t j '~r , . : lcd irl T a b l e 3.

Y V. ( ; w w l f i , c~ al . /~ 'ct<mul~ln.w~.~ 2,";,'~' (199<~>) I I h ~)

Vert ical Z -componen t

~;!~!v~! "ii i~t ~t!:'::'!l.~,~i~tt ~ , ~ i ,~t~l' !!:,~ i ~

I I I

. . . . Hor izonta l X -componen t

'N ,' ,~ I',, la~2i~D, " , : , ' ~)~ ~1.7 1~"

~t, it "~i: "~7t"~' ,t_,~yl~;~'.,!j~ (f.#).,,~! 7,,

I I I

. ,{i7,; : . . 4

U:{., SP-conversi°n

ttiT/tlllllllli I

,' ~ f f / i l : . / r : l ,

;t- ~' II).')'/L" 17Dl': 'l/

,171Ill/ D i r e c t S

O , r < c , ,

ii;{ht.;$~'.TD~L¢ f:/

SP-conversion , / ( ~ , Y l l # ' i , " ~

i

E

.0.1

I Z

. . . . . Hor izonta l Y -componen t

~(~Lv" , ~')~'pF.i~ii,~. # ~_~(~I li~li~t.! ].ll ~ .;,Directs

l| ' '..II ~ri.i ,, ,'.:~ , - ;.,.,:~:~ ,SR,:~.~" -,%.".~ r(, i ! ~il ,, D:-..g-,',:~.,):.,.- Direct P

~('7!,q~ ' i '<i;i ,,'@~i,~'.l,~',~'. ~" .~ ,: ' "P;"~'~II/.

4 3 2 1 0

T i m e , s e c I-i~. 5 \ 'SI>-- ~ IhF¢c-c'llllll'..~llCl]l dinphi). Source. 2.{)N kni norlh lr'Olll lht" bllrehole: baild-r)tlSn liher. 22 5,~ I I / . (;;lll'l cllrrcclc'd l'l~r ,_,conlc i r ica] ~prcad inT . Arl'tl~,t, ,, illtii~.'tilc' r c l | c ' c l i on~ dikc'u,,xcd hi thc I,..'~,1 ; l l ld i n t e rp re ted in " l 'ablc J+

I 0 E U ( ;anch in el al. 17"e~ tOnol,h.vsics 2b:b; t 199~%' 1 /-- / 0

Table 3 Ray-tracing results for reflectors from VSP- I seismic waxel ie ld"

Componen! Rel lecmr Apparent vehm'ily Retlected wave Reflectors Angle of

(kin/s) type Dip (degrees) Deplh (kin) Incidence (degrees) Approach (deg,ec',,I

Z A I 4.5(X) PS 25 4.9 5 - 2 0 28- 37

A2 4.5(X) SS 25 4.9 4-2(I 30-45

B 8.21X) I'l' 15 ~ I I. 3 -6 3 - ~

SP 30 ~8 .0 5 • 20 35-55 (" 5. I(R) SS 35 8 .0-11.0 5--15 38-55

X AI 4,5(X1 PS 25 4.9 5-211 28-37 A2 4.500 SS 25 4.9 4 20 30-45

l?, 4.5(X) SS 25 ~6.0 4 - 18 3 0 - 4 0

('1 3.71X) SS 0 "-7.0 0 2 0-2 C2 3.7(X) SS 0 --9.0 0 2 o -2 I) 5.300 SS 35 ~9.0 5 15 38-55

Y A I 4.5(10 I)S 25 43) 5 - 2 0 28 37

A2 4.5(X) SS 25 4.9 4 20 3(I-45

B 4.5(X) SS 25 --6.0 4 18 3 0 - 4 0

(" 3,8(X) SS 0 ~8.5 0 2 0 - 2

l) 5,200 SS 35 "~ I 1.0 4- 14 36-52

a Source ol]'sel ().2 km soulh of Ihe borehole.

axis is 25-30 ° (above 2.2 s). Rettections R I-R5 (Fig. 2) arc associated with acoustic contrasts in- side the Proterozoic Pechenga complex. A group of

"lablc 4

Ray-tracing resuhs for rellectors Irom VSP-2 seismic v, avelield :'

coherent reflections R I inside the "Productive" For- mation ( I. I-2.8 km deep) correspond to contacts of metasedimentary layers with ultramalic intrusions.

( 'omponenl Relleclor Apparent ~elocil) Reflec(ed wa,,c Relleclors

(kin/s) lypc Dip (deglecM l)eplh (kml

Z A 8.500 SS 25 4 .5 .5 .2 P, 5,8()) Pl' 0 8.0, 9.0

SS 30 5 .5 .6 .5 PS 4(I 8.0, 9.11

(" 6.5(R) SP 0 7.(I- I 1.11 SS 35 8.0.-12.1) Pl ' 0 12(I 14.(1

A 7,500 SS 2o 4.5

B 4.6(X1 SS 25 5 .5-7 .0 PS 30 6 .0- IO.O

( ' 3.6(X) PS 0 9 .0-12.0 SS 0 7.0- I 0.0

A 9.50t) SS 30 4.7. 5.2 I'l II .1100 PP 35 "- 11.0

SP 35 -8.5 ( ' 3.600 PS (1 6.0.- I 1.0

SS (1 (LO I 1.0

" Source ofl'sel 2.(18 km north o l the borchole.

Anglc of

Incklcncc {dcgreesi , \ppmadl Idcg,'cc~l

25 45 50 -70

8 - 1 0 6 12 3(1 40 60-70

20 4(1 50-70

6 - 8 4 I0

12-25 45 -.6O 6 -8 4--8

20 -40 4t)-60

18-42 51)- 7O

15 35 55-75 5 -7 4 - 8 6 8 4 - 8

30-50 60 80

12-25 45 (}I) 5-3O 4O 5O 5 -8 4 8 5 8 4 8

Y v Gain'bin el al./'l~'ctOnOldO~ic~ 2&'; I I9~.~,~j I - 16 I I

Thcse boundaries arc characterized by the largest velocity contrasts (more than 1.0 knfs), and the corresponding reflections are among the strongest m thc CDP section. The reflectivity character of the rocks constituting the 'Productive' strata is en- hanced by several shear zones associated mostly with metasedimentary layers. Several relatively thin ( 10-20 m) horizons of sedimentary-tuffogenic rocks ( Vp = 5.7-6.0 kin/s) intercalated with sheets of mas- sive and globular lavas (Vp = 6.7 kin/s) at a depth of 3.65-3.85 km produce visible coherent reflec- tions R2. Reflection R3 manifests a transition in the SG-3 section between rocks with massive texture and foliated ones (4.5 km deep). This depth also corresponds to the top of the major l,utchlompol Fauh Zone (LFZ, 4.5-4.7 km deep), where P-wave ,,elocitv drops by more than 20~7~. In spite of such velocity contrast, the retlection coefficients do not exceed (LOg m the LFZ (Fig. 2) because of the gradual character of this zone.

The muhicyclic reflection R4 can be associated v, ith a metasedimentary layers (recta- limestones. sandstones, and carbonate schists in the depth range 5.65- 5.75 kin) inside foliated basalts and plagioclas¢ schisls. Though the velocity contrast in this inter- val is relatively small (5% and less), we observed increased amplitudes on both the synthetic seismo- gram and the CDP section (Fig. 2). This results from laminated geometry of reflectors. The laminated con- liguration below the LFZ (4.5 kin) is produced not only by sedimentary layers alternated with basalts, but also by zones with strong cataclastic fabrics ~vithin layers of unmylonitized rocks. Two impor- tant characteristics of myhmitic rocks - preferred orientation of minerals and retrograde mineralogy (l:ountain ctal. . 1984) - sugges t that velocities for wa,,es traveling normal to mylonitic layering can be substantially reduced with respect to umnylonitized rocks. Preferred orientation of minerals, particularly biotite or other phyllosilicates, can be expected to cause anisotropy below the I,FZ (reflections R3-R5 in Fig. 2).

The ProtcrozoJc-Archean (Pr-Ar) contact (6.X4 knl deep in the SG-3 section, reflection R5 in the seismic section).juxtalx~ses Proterozoic diabase schists and i\rchean gneisses. The P-wave velocity drops to 5.7 km/s in the corresponding depth inter- \al. ()ullv i11 situ increased fracturing or anisotropy

can explain such ,,elocity drop. Reflection R5. corre- sponding to the Pr-Ar contact is, most likely, caused by a fault zone dipping in the same general direc- tion as the whole Pechenga structure. This fault zone changes physical characteristics of the rocks com- prising it, making them less competent and more fractured thal decreases acoustic properties. Con> pared to the dipping reflection R5. the reflection R6 is nearly horizontal, as well as the deeper ones R7 and R8 {Fig. 2). The possible nature ~)I" these sub- horizontal reflections will be discussed belov< after a correlation of reltections found on the CI)P section with lhose from the near-offset VSP- I.

4.4. C o r r e h l l i o n o I ( ' D P a n d V S I ' re.ltc¢'lirily

p~lllcvll,~'

The ton'elation of CI)P and VSP rellectivhy pat- terns is displayed in Fig. 6. Due to predonmlantly dipping interfaces in tile upper Proterozoic part of the SG-3 section, it is unlikely to calibrate the CDP sec- tion by means of conventional corridor stack based on a simple flattening of first-break limes (Fig. 6a). Instead, better results are obtained by a corridor stack performed after flattening of reflections h 3 pr¢calcu- lated dip moveout corrections (Fig. 6b). Such I)M() con'ections were obtained by the lllcans of ray-trac- ing for the known velocit}, model (based on inte- grated AI, data and inversion of VSP lit'st arri,,als) and variable dip angles of interfaces. A dip angle of 25 ° w.as found to produce the host P-v,..a\.e llat- tenmg results for VSPs at least m lhe I,FZ \ icinitv. This value is in general agreement with geological observations and with the results of VSP ray-tracing tk~r the SS reflections from the I,I:Z ITables 3 and 4). The resulting corridor slack IFig. 6hi positions properly the reflector R I and makes the rcllector R2 much more pronounced compared to convelltionaJ stack (Fig. 6a). Reflections R3 and R5 corrcspond- ing to IJ:Z and Pr-Ar coulacl, respectively, as well as the multicyclic reflections between tlletn are in accordance with surface CMP reflections (Fig. 6hi. There is a long train of reflcctions below Ihe Pr Ar contact which is difficuh Io correlate with thc ('1.)1 ~ data. To understand better the rel]cclivity patterns on both VSP and CDP data. another corridor stack using I)MO corrections for SS rellections was perlk~rmed. After the stacking, the resultant trace v,a,, rescalcd

SW

0-

SU

RF

AC

E

SE

ISM

IC

2

NE

C

OR

RID

OR

S

TA

CK

S

VS

P-1

20

0 m

so

urc

e of

fset

, ve

rtic

al c

om

po

nen

t

I I

i I

3 4

5 6

Dep

th,

km

rave

s

i

C~

3 T W T T

i se

c se

c

Fi~.

6.

(.',:

.rrc

lafio

n oL

t11

¢ ul

'~'n

in ~

'~'

.:.~'

,'¢lic

l,,I

hC~m

the

\"S

P

whh

Itl

~ nu

rl':.l~

.'c n

cism

j~'s

. Th

e co

rrid

or

',,I:.L

c],,n

~.trc

l~,rt

~,..l

uccd

h,.

,~un

mfin

~ V

SP

n al

ter

llallc

njng

(a

) b'

, I[I

c (,I

jrc,..

'l ]~

-'..~.

~.l'.c

aFF

j'.L~I

s. (

h)

b'.'

2~ ~

dil',

m

,~Y'

.',..'O

tlt c

,:,rr

cctio

ns

for

PP

-rel

lecl

cd

w~v

..c,

and

(c)

hv

25 °

dip

nlo'

..'oo

l]l

C('~

l'rCclj

(.)n~

, t('

~r S

S-r

clle

ctcd

v,

':.r..c

(r

esca

lcd

in

tin|c

{o

F)

P-',',

~.I\c

~t

rrj,.~

.dn)

. Th

e ni

IlgjC

Ira

ccn

I'r,{)

lll c,

,~rrj

,..]o

r sl

:.lck

s LI

I'L' re

pel.L

Ied

t't~r

"..is

ual

cfl'c

...'t.

VS

P-I

i,,

plo

tted

al

lot

(.'()r

rC,,:

lj()n

x I'(

~1 g

ct~n

lctr

jcal

xI

',rc~

ldin

~,

I:K

-lih

crin

g ((.

1(~,

.,. tlg

oirlg

,.,

.:.r, ¢

licld

IC

jCLl

C~.])

. Ll

lld t

lLfll

u'llj

Il~

I'~\

-~5 ~

dip

III

()\L

'()L]

I C

OI'F

CC

IiOIIx

I'O

f |h

C l

)])-r

cl'tC

,,:lc

d '..~

...Y'.C

n.

Y ~: ( ;am'hi t t et al. 17~'ctomq~h.l~i~ ~ 2,'~'~ ¢1~9~f I 16 13

(T, /7 i, = Vp/V, = 1.72) and plotted (Fig. 6c). In such a stack, the reflections R3-R6 became much more pronounced than on PP wave corridor stack. This observation is in accordance with the radiation patterns by Gibson and Ben-Menahem (1991) for an anisotropic scattering volume. The deeper retlec- tions on an SS wave stack (Fig. 6c) are concentrated around 3 s (9.5 km deep), and according to our interpretation (Tables 3 and 4) they originate from horizontal interfaces.

5. Discussion: causes of velocity variations in the crust sampled by the Kola Well

()he way to correlate litholo,,ic cross-section of a &..

crustal segment with the corresponding data derived from sonic logs and VSPs is direct measurement of elastic wave velocities in core samples at high confining pressures when most of microcracks are closed Vernik et al. (1994) performed such measure- ments for selected rock samples from the SG-3 bore- hole. A comparison of limited (12 samples, 15 cm total length) high-pressure velocity' measurements with sonic log and VSP data. showed some agree- ment (Vernik et al., 1994). Based on this comparison, Vernik el al. (1994) explain all depth intervals with relatively low velocities by the combined effects of variations in lithohwv~, and rock fabric only. All other possibilities, as free-water-tilled in situ microcracks {e.g. Kremenetsky, 1990) or tectonically induced di- latanc 3 (e.g. Mints ct al., 1987) arc neglected. The presence of subhorizontal seismic reflections in the Kola area at a depth of 7.5-8.5 km lbund on wide angle rellection and deep seismic sounding data (e.g. Mints et al.. 1987; Pavlenkova. 1991) is explained by Vcrnik el al. (1994) b 5, the variations in amphibolite bodies" concentration.

Another approach Io correlate lithologies with scismic velocities was introduced by Babeyko et al. (19941. They have used a petrophysical modeling technique Io calculate intrinsic (crack-free) elastic properties in the lower part of the Kola borehole (6.84--10.53 kin) from bulk chemical compositions of core samples. Calculations were performed for a total of 896 core samples. Babeyko el. al. (1994) used the comprehensive method of vehx:ity mod- eling. First they compiled the detailed lithologic cross-section of the Archean pa.rt o[" the Kola bore-

E

e-

a

5.5

7.0

Vp, kmls 5.7 5.9 6,1 6.3 6.5 6.7

8.0

9.0

10.0

.= ............ I I

Fig. 7. Zonc~, m the Archcan section of TilL' S(;-. ~' c,hadcd arcas l

where i11cilMir,2d velocities arc lm~cr Th;.ln Ihc inlrin:,ic icrack-

Ir¢cl onc~,. ( 'onlpar ison is ba~,cd on Ihe re~,uhs ~fl' i~em~l~hy,,ical

modeling (Habc,,ko c¿ al.. Ig94l ~.;ith The interval P-v, axc ' ,clot-

ilion, froln the observed VSP and acoustic log data. All ~clocit,, '~il]tlC'~ Tire aD, cr:Agcd dl 200-Ill Jntcr \a ls .

hole including 554 individual layers, and only after that, the3. estimated crack-free isotropic properties of all the layers using calculations for core samples and lithologic infornlation. The results of such calcula- tions averaged tit 200-nl intervals are presented in Fig. 7 in comparison with velocities obtained from the acoustic log and the deepest Russian VSP (least- squares inversion of travel times performed in the present study).

The shaded areas A, B, C. and D on the veloc- it3,' plot in Fig. 7 correspond to the depth intervals where both sonic and VSP velocities are lesser than the intrinsic isotropic "crack-free" velocities. Espe- cially' large velocity contrasts arc observed for zones

A (7.0 km deep) and I) (10.0 kin): 0.3 and 0.4 kin/s, respectively. Zone A corresponds Io the fault zone jusl below Pr-Ar contact and correlates wilh a group of reflections R5 {Fig. 2) dipping in Ihe same general direction as the layers of the whole Pechenga complex. All other coherent events below the group of rellections R5 are st, bhorizonl,al with reflection R8 being the strongest and mosl laterally continuous <more than 5 kin). The reflections R6

14 )( V. G a m h m et al. / 7+'+ton+ q~hv,sic ~ 2,~8 ( I t)9,'~' J I--16

and R7 (Fig. 2) correspond t¢3 the velocity contrasts between zones B-E and E-C (Fig. 7), respectively. As reported by, Lanev et al. (1987), amphibolites are mainly in conformable occurrence with gneisses and contacts form an angle of about 40-60 ° with the borehole axis. Gneisses of the upper parts of the Archean complex (down to a depth of 9.45 km in the borehole section) are widely developed at the surface north from the SG-3. These facts lk~rce us I¢5 look for another explanation of the origin of subhorizontal reflections R6-R8, rather than excess of amphibolite bodies.

The effect of fluid-filled voids and anisotropy due to preferred minerals orientation (PMO) can be a reasonable explanation in the case of subhorizontal reflections. The Archean biotite-plagioclase gncisses in the borehole are foliated, and the foliation plane has an inclination of 40-60 ° to the borehole axis (Ko- zlovsky, 1987). Taking this into account, t:~abeyko et al. (1994) performed calculations of velocity devi- ations from isotmpic values in Kola gneisses with average biotite content. They found that anisotropy due to PMO should not lower P-wave velocities more than 5c~. Comparison of isotropic crack-free vehx:- ity calculations with VSP P-wave velocities shows that anisotropy due to PMO is not enough to ac- count for a velocity reduction (7 -8~) in zones A and D (Fig. 7). That is why, we propose fluid-filled voids as major mechanism generating subhorizontal reflections.

The structure of fractures and Ix)re space of the Kola core samples have been studied by many au- thors (e.g. Kozlovsky, 1987: Morrow ct al., 1994: Abdrakhimov et al., 1996). All the investigators agree that many of the observed nlicrocracks re- suited fronl drilling-induced damage, but some of the fractures have characteristics clearly suggesting that they have fomled naturally (e.g. Morrow et al., 1994. fig. 5; Abdrakhirnov et al., 1996). To clarify the nature of fluid and its effect on rocks, Abdrakhi- nlov et al. (1996) performed special experiments in a "thermal pressure chamber', which allowed variation of temperature up to 300°(7, hydrostatic pressure up to 200 MPa, and deviatoric uniaxial compression up to sample destruction. The experiments were carried out with samples (gneisses and amphibolites) taken from the surface near the SG-3 wellhead. It was es- tablished (Abdrakhimov et al., 1996) that subcritical

opening of intergranular space in a crystalline rock becomes possible only with the presence of water. Neither pressure, nor temperature alone do not cause subcritical rock damage. The occurrence ¢/1" subcrit- ical crustal rock destruction due to stress corrosion has already been documented (e.g. Kerrich et al., 1981: Etheridge, 1983). The particular importance of the work by Abdrakhimov et al. (1996) is that they have estimated the 9-12 km depth interval in the Kola SG-3 section t¢5 he the most favorable tor stress corrosion cracking. At greater depth, with tempera- tures increasing to 2(X)"C, this process is changed by crystallization and crack healing. The CDP seisnlic section incorporates the nlost prominent subhorizon- tal reflection R8 at a depth of 10 km (Fig. 2). Ap- proximately the same depth generates subhorizontal reflections recorded on 1992 Kola VSPs (Figs. 4 and 5, retlections C). The maximum helium concentra- tion at a depth of 10 km and increased water inflow (Kozlovsky. 1987) together make the rock-destruc- tion process at 9-12 km, caused by stress corrosio,i, the most probable explanation of the observed sub- horizontal rellectivity. The subhorizontal character of seismic retlectivity also finds its explanation in stress corrosion effect, because of approximately linear P- T distribution in the upper crust.

6. Summary

The Kola ClIP section is highly rellective; dip- ping rellections are generated from lithologic con- tacts within the supracrustal series and from shear zones. Reflectivity of several zones including a ma- jor shear zone at 4.5 km is enhanced by fluid-tilled fractures indicated by VSP interval velocities that are too low for non-porous crystalline rocks. Shear zone retlectivity is also enhanced by anisom)py. Re- flections may be caused by compositional bound- aries, shear zones, anisotropy and fluid-tilled ",oids or some combination thereof. Horizontal rellections are most ct)nlmon in the Archean below 6.8 kin, but are also found crossing dipping reflections through- out the CDP profile. No horizontal dips or layers are observed in the core material so no evidence exists for the presence of horizontal dikes to explain tile reflections as has been suggested earlier. Inter- val velocities determined from VSPs are lower than intrinsic rock velocities in the Archean section and

E t.: (;am'hi . ctal. / 7~.clo.oph.wic~ 2,~,~' ~/~.'9,~t I /6 15

also Ior the LFZ. The lowered seismic velocity is at- tributed to fluid-filled microcracks and the horizontal rellections to lluid-tilled fractures. Results suggest the presence of tluids down to a depth of at least 12 km in the upper crust: the presence of these fluids lowering seismic velocity causes estimates of upper crustal composition to he too felsic. This rnay be a widespread phenomenon which means that global estimates of crustal composition based on seismic velocity are biased toward compositions that are too felsic.

Acknowledgements

Thanks are due to both National Science Foun- dation (Grant EAR-91186(X)) and RossComNedra lk~r linancial support. The authors would like to thank 1). Guberman for access to the Kola Superdeep Borchote, Dr. Yu. Kuznetsov for providing us with original log information, and Dr. M. I,izinsky for providing access to the additional Russian deep VSP data. We also wish to thank C. Humphreys and the other tnembers of international seismic team who made the acquisition possible under difticult lield conditions. We thank (?. Juhlin, H. Harjes, S. Klem- perer, and W, Mooney, for many helpful comments and suggestions.

References

Abdrakh imo\ . M.Z.. Kuznet~.o,.. Yu.l.. Zonn. M.S., 199b. t"orL" ~,pa,++,+. ' struc'tur,+." o f deep cruslM rt~_+ks frtm~ the Kok, OvL'rdccp drfll ir,# data. Phy>,. SolM I']arth 32 (5). 426-436. Translated h't ,n tgz ika / .eml i (Russia).

BabL'\ k~. ,\.Yu.. Sobol¢\. S.V.. Sinelniko',, E.l).. Sndrno~,. Yu.P.. l)L'ru\,,chiko\a. N.A.. It,U-J4. "l'hL' Kola Borehole from hulk chL'lni,.:al cOnll)t~nJtJon ~tl L'ol'c ,ample',. Surv. (ie~ph',.~,. 15. 545 -573.

I'lL'rthcl,,cn. A.. Marker. M.. 1986. "l'cctonics of the Kola coll ision

~,UltlrL" and adjacent :\rcllL'an and Earl,, Proterozoic tcrrahls In the nt+rthL'a,,tL'rn rL'gion of the Balt ik shield. II1: (;alstm+ I ) ,A.. MudlL'r. St. reds.). The European (;cotra\'ersL'. Part I. "Fccttm~q'~h,.,,ics ] 2~3. 31 55.

('err. I'l.J . Snlith~,on. SB. . Karae~. N.A.. Ronin. A.. ( iar ipo\ . V.. Kre, t~llL'rsun. Y.. l.)igranes. P.. Smythe. D.. Gil len. C.. I t~t~n. ~,crtical sei~,mic pmtilL' results l 'rom the Kola Superdeep I'lt~rchulL'. Russia. TL'ct(mophysic~, 2(M.. 295 307.

('hristcnsL'n. N.[.. 1989. Rel lcct iv i t ) and scisnlic r>ropcrtics ol Ihc dcL'p continL'ntal crust..I. ( icophys. RL's. 94 ( B 12 L 17793-

178()I. l)igrunu,,. P. Kri~,t~ffl_'rscn. Y.. Kara t \ , N.. 1996. An unalysi'~ oF

shear waves obser\ed in VSP data from the nupL'rdcep v, ell al

Kola. Run~iu. (';coph~.,. J. Int. 126. 545-554. [ ' )hcrMg¢. M . A . . 1983. Dif lercnt ial ,,tress, nlagniludc,, during rc

gional deformation and nmtamorphism: upper hound ilnpt~,,ed h) tensile Iracturing. C;eolog~, I I . 27,1 234.

I-oUillairJ. I ) .M.. Hurich. (.'.A.. Smith~,un. S.B.. 1984. Sci~,miu rcfleL'ti',it+v o t ' m \ h m i t e /ones in the crusl. ( ;cology 12. Ig5 198.

I'uch,,. K.. It;'(~9. On the pn~perti¢,, o f deep cru,,lal rcfleclinn~,..I ( ieophw, . ,'15. 133 149.

(;ih,,on. RI , . . Ben-Mcnahcm, A.. 1991. Ela,,tic v.a\e ,,catl¢l'in:2 b)" allJX+i~tl't~pic obntacles: i.lppJJcatJnll to lraclt l lL 'd \~IUlllC~. I (;cnphy,,. Ren. 06. Ig905 Ig924.

Hohrath. A.+ L~ram, K.. Hanitzsch. C.. Hubrul. P, K,.i,,tncr. t . . I.u,,chen. I!.. Riihl. "I.. Schrulh. P.K.. S011nct. \~... 10~)2 l{~al uutiun end interpretation of \:,%P-nlcanure.ncnl, rill l i l t KFB. ( )hcrp la l ] pilot horcholc. Sci. l)t'fl l. 3. St). 9t:.

JuhlilL ('.. 1990. ll+terpretatJon ol rcllcclor,, in the Sillan Ring ~.IIL'~I b~lse,,l on icstllls fr{'~lll tilL' ( l l ' ; l \ hL'r~-I }',{u¢holc, 'I"L'L'hHI(~ ph).,,io, 173. 345 3bO.

Karus. I-.W.. Kuznel,,o\. ().L., Kuzncl~,o\. Y.I.. Rudenku. (;.1{.. I,izm',k>. M.I) . . l , i hmcnko . 1.'¢.. Kud/m,,k~. [ .I... IghX7 SL'i , IlliL" jlI\L'MI~;IIIOI]n ill 1he Borcholc. In: Ku/ lu \ , ,k \ . Y.A. i l :d.I. "l'hL' SupL'rdccp Well of IhL" Koltl Pcnm',ula. Sl~rm~2L'r-Vcrlag. BL'rlin, pp. 351 35,~.

KL'rrich. R.. l.u rour . T.E.. BurnL't. RI , . . I~;~1. MinL'ral rL'actlons participating in intrugranular IracturL' prt~pagution: implica- tJOll'~ for ,,tre,,s cnrro,don ',.+racking. J. Stru,._'l. (;e,,d. ~,. 77 S7.

Kozh+\nk}. Yo.A. i l !d.). 1987. The SupL'rdeep W,.'II o f tlIL" K~da PL'ninsula. Springcr-VerhLg. Berlin. 558 pp.

Kren]cnetnk ' , . . . \ .A. . 1990. The geological natmL" o l nJ..'i,.Fm,.. b~mndaric, in the ,contincnl;d ',.TIIM. Ill: l'u',:h',. K.. Ku l lo \ , k , , . "tL'...\.. Kn \ t , ,o \ . A.I. . Zohuck. M. I ) . qlid,,.L SupL'r-I)eL'p ,:un- tinL'ntul dri l l ing and dcL'p gco1"~h}.sical >,~mndin.t.,. ,~pllng¢l-\ '~'r- lak2. l'lerlin, pp+ 353 -3t33.

l.anc',. V.S+. Nal i \k ina. t{.B.. \:akim~shr\a. V .V. (hdL'nkina. li+,\+. Ru>,unu\. M+S+. Smirnt+\. YR . Su>.h+\a. S+N++ I)uk. (i+(i.. Koltstt\u. "I'.V. Ma, lcnn iko; . V.A., "l'im~lcc'.. H.V.. /,a~,la\nk\. \".(i.. Ig87. (;c~lugicul :,cctitm of the \~ctl. In: Ko/h~\~,k\, "~.A.

(t-M.). ThL" ,',;upL'rdcep WL'II ol + the Kt+la PL'nnl,,ula+ ~,pringel.

\.'erlag. Berlin. pp+ 4(')-73. [+c',ander...\.. lh+hh>+. R.V~'.. Snnth. S.K+, l++n,L']and. R+W. Svnd.._'r.

I )B. . t lolliger. K+. I tA(-}4+ The cru,,t w, a hL'tL'rtL,2'¢nL'ou,. "Optl-

cul" lllcdh~+'n, tu "crt~'udilL's m lh,..' nli,,t'. "l'cchmoph~sic., 232.

2,'-; I - 297. Mc lc /h ik . V.A.. %furl. ILA. . 1994. (iL'llL'ral L'L'~dt~gb und c ~ -

I t l t iOllar\ hlMt)l '\ t+t thL" carl.', Plot~.'l't~/oic P o h l l a k - P a s \ i k lh.tchcng.:t hnandra/Varzuga-L'>,t'l)ono_ ,, (irv.ennlonL' I,k..h m tIlL' nortllca>.tern IL0.1tJc ShiL'l,,.l. ~allh-Su'i . RL'\ .t, fl. 2()5- 245.

Mint,,. M V.. Kolpakm. N.I.. l.anL'\. V.S.. Rtnsanm. M.S.. Ig~7. ['he charitctcr ol the >,ubht+riztmtal ,.einnuc l-,~mndaric,, \ \ i lhm

the Upl+~cr r~;ll.l tel the 1"arth',, L'rllnl tacct'~r,.lhlg to data lltVn the Kola t J ltradL'L'p Well ). (+cotectt>nl£:., 21 t 5 I. 444- 451.

Motme,.. W.D.. ML'b, snL'r. R.. 1992. Multi-~2cnctic uriL.m ~1 crustal rcl]cctP, ity: a r¢,.iev, t>l" sci,.m~c rcl lcctitm pml i l ing t+l the ctmlinel}tM crust and Mol+~t). In: l:t)uI1t;.un. I).M.. ,,\rcultts. R.. Ka',. R.W. II 'ds.). ( 'ont incntal I,t~\~cr ('ru,,I - I)c\clo1"~

16 E ~{ (;am'hin et M. / Te('t~m~q~h~.~i('.~ 288 (/99,~; I 16

merits in (;eotectonics. Vol. 23. Elsevier. Amsterdam. pp. 45- gO.

Morrow. ('.. l,ockner. D., ttickman, S.. Ru~,ano',. M.. Rockel. T.. 1994. Effects o f l i thology and depth on the permeability of core sanlples from the Kola and K'I'B drill hole,,,.. J. Geoph.,,s. Res. 99. 7263-7274.

Pavlenkova. N.I.. 1991. The Kola nuperdeep drillhole and the nalure of seismic Ixmndariex. Terra Nova 4. I 17.. 123.

Rector, J.W.. 1988. Acquisition and I~relinlinar,, • anal,csb, of ori- ented nluhicolllponent multi-offset VSP data: DOSE('(" ('ajol! Pass Deep Scientilic Drillhole. Geophys. Ren. Left. 15. 1061-. I o64.

Rudnick. R.I,.. Fountain, D.M.. 1995. Nature and composition of the continental crust: a lower cruMal perspecti,,e. Re,.'.

Geophys. 33 13), 267-307. Smifllson, S.B., Shire, P.N., I'~rown, S.K., 1977. Sei~,mic rellec-

lions from tkecamhrian crust. Earth Planet. S,,+'i. [.eli. 37. 333 33g.

Ste~.art. R.R.. 1984+ VSP interval ,.elocities froln traxeltilnc inversion. (ieophys. Prosl~Ct. 32.60g 62g.

Valasek, PA.. Snoke. A.W.. Hurich. ('.A.. Smithson. S.B.. 1989 Nature and origin of seismic r¢llection l;ahric, Ruh',-East Ilum- boldl metamorphic core complex, Nevada. "12"ctonics 8. 391- 415.

Vernik, I,.. Hickman. S.. I,tu:kner. D., Rusano'.', M., 1994. t lhra- sonic velocities, m cores from Ihe Kola Superdeep '*ell and Ihc nature of suhhorizontal ~,cisnlic reflections. J. (]eoph~. Res 99. 2420t.~ 24219