petrophysics of metamorphic and igneous rocks

Effect of compositional variations on log responses of igneous andmetamorphic rocks. II: acid and intermediate rocks

R. PECHNIGI, H. ogLtus' & A. BARTETZKOIlAngewandte Geophysik, RWTH Aachen University of Technology,

Lochnerstrasse 4-20, 52056 Aachen, Germany

( e - mail : r. p e c hni g @ g e ophy s ik. nuth- aac he n. de )2Department of Geotogy, University of Leicester, Leicester, LEI 7RH, UK

Abstract: An extensive data-set of petrophysical down-hole measurements exists forboreholes drilled into continental crystalline crust. We selected boreholes covering a

range of different types of plutonic rocks and gneisses in amphibolite or high-grade meta-morphic rocks. According to Serra's concept of electrofacies, a specific set of log responsesshould characterize one rock type. Here, we concentrate on the detection of compositionalvariations between rock types. Bulk composition of the protoliths influences the mineralo-gical composition of the metamorphic rock, and we demonstrate how this impacts on thedown-hole measurements. Integration of logging data with geochemical core data andmineralogical descriptions allows the calibration of the log responses to rock types. Therelationship of the log responses with core data shows a remarkably good correlation, anddiagnostic trends are detected. From the logs, potassium and neutron porosity are particu-larly helpful in distinguishing different types of gneisses and igneous rocks with respectto their protoliths. The propoftions of amphibole/pyroxene, mica * K-feldspar andfeldspar + quarlz in the rocks seem to control the direction of correlation in a cross-plot,i.e. positive or negative, depending on increasing or decreasing mineral proporlions. Thisis true for al1 boreholes, and a generalized classification scheme could be developed forthese crystalline rocks.

IntroductionDuring the last few decades, several boreholeswere drilled into igneous and metamorphicbasement rocks. Exploration drilling has beenperfomed for the purposes of mineral mining,tapping geothermal energy or disposal ofwastes as well as in the framework of researchdrilling in continental or oceanic crust. In con-trast to well logging in the traditional hydro-carbon environment, information about logresponses in igneous and metamorphic rocks isscarce. This is especially true for the acid andintermediate rocks of the continental basement.Besides some early compilations of 1og responsesfrom the principal igneous and metamorphicrocks (Keys 1979; Desbrandes 1982), moststudies performed hitherto have focused onsingle boreholes drilled into continental base-ment (Daniels et al. 1983; Sattel 1986; Paillet1991; Pratson et al.1992; Traineau et al.1992:'Nelson & Johnston 1994; Pechnig et al. 1997).These studies showed that the complex geologi-cal conditions of crystalline rocks, particularlymetamorphic rocks, are not easily determined

from the logs, because of the superyosition oflog responses produced by the varying compo-sitional and stmctural variations of crystallinerocks. Hitherto. no systematic inteqpretation orclassif,cation charts are available for crystallinerocks. Therefore, this study is focused on a

comparison of igneous and metamorphic rocktypes from drill-holes in continental basement.It comprises, besides foregoing electrofacies ana-lyses individually performed for the differentholes, a core log integration to develop classifi-cation charts. which include all availableinfomation on rock chemistry and mineralogy.

Log and lithological data compilationThis study is based on a comparison of wirelinedata with mineralogical and geochemical datafrom core and cuttings samples. Log, core, andcuttings data from the following boreholes werecompiled and analysed: KTB pilot hole,Leuggern, Schafisheim, Böttstein, Soultz-sous-Foreß GPKI, Moodus and Cajon Pass (Fig. 1).These boreholes cover a wide range of different

Frorr: Henvnv, P. K., BREwER, T. S., PEz.rno, P. A. & Pnrnov, V. A. (eds) 2005. Petrophysical Properties ofCrystalline Rocfts. Geological Society, London, Special Publications, 240, 219-300.0305-8719/05/$15.00 The Geological Society of London 2005.

280 R. PECHNIG ETAL.

Fig. 1. Location of the boreholes compared in this study.

types of acid to intermediate plutonics, orthog-neisses and paragneisses. From each borehole,one or several intervals were chosen to be repre-sentative for the lithology. Intervals wereselected using the following criteria:

(1) general availability and quality of the

logging data;(2) availability of core/cuttings information;(3) intervals with good borehole conditions to

avoid logging tool failure sources fromborehole enlargements, and

(4) homogeneous sequences and large bedthickness, to minimize log integrationeffects on bed boundaries. Since the focusof this study is concerned with the effectof the rock composition on the toolresponses, fractured, brecciated andstrongly altered intervals were genelallyexcluded.

Geological setting and lithology of the

selected boreholes

Metamorphic rocks were drilled in the boreholesKTB (Emmermann & l,auterjung 1991:.

Hirschmann et al. 1991), Moodus (Naumhoff1988), Leuggern (Peters e/ al. 19894) and

Cajon Pass (Sllver et al. 1988). The rocks

drilled in these boretroles mainly compriseparagneisses, orthogneisses and metabasitesoverprinted by amphibolite-facies up to granu-lite-facies metamorphism. Plutonic rocks weredrilled in the boreholes Soultz-sous-Förets(Traineau et al. 1992), Böttstein (Peters et al.1986) and Schafisheim (Matter et al. 7988'1, andin the deeper parts of the Leuggern borehole(Peters et al. 1989a). The drilled rocks includegranites to granodiorites, syenites and monzo-nites. An overview of the boreholes, their rockcontent and petrogenetic evolution is given inTable 1 and Figure 2, and is summarized in thefollowing section.

The German Deep Continental Drilling ProjectKTB drilled two deep holes at the western marginof the Bohemian Massif. The drill site itself lieswithin the Zone of Erbendorf-Vohenstrauss(ZEV), a small crustal segment of the Variscanorogenic belt. Two boreholes, separated by onlya short distance of 200 m from each other, weredrilled. The pilot hole (KTB-VB) reached a finaldepth of 4001 m, and was almost completelycored. The drilled crustal segment consists of analternating sequence of three main lithologicalunits: paragneisses, metabasites and'variegated'units of paragneiss-metabasite alternations(Emmermann & Lauterjung 1997). All theseunits have suffered pervasive metamorphism to

281LOG RESPONSES: ACID/INTERMEDIATE ROCKS

=--lEE=.=e. , p Ebtz + i:EE.=-iE==iEErI;. E - .'t.: .2 .== ; .=..tr4 .=.. ' ab g ä Es ä :eE H SeZni ?i-EE !;,' A ä:2 ä ö2 ) j'4 Z ==EiZ== ZEzä sF.! =€ "rS; EreA Ei" = :Egäi |EEA Ej + i_.> A-a: ,)b- .=tr

= tr q=- =),t- Ei e ü,E

-='l v- ä-=cci *-ici 9."^€ z.==, --N

=--N a= E >== .-L .- El?t - +ü ;F; E;Fu =€ E ;5ä.a; ,-*Eu F=,iE ;F

t E !E= öo> q>EiE >iF=E FE=E; ?i>.=e.ElE >aI eE? äRs-Es F; äF^'* a* z lEe El : re F> F: I s;s Täx[?1äi[sFäs[-:[€x[** =i[s!-:s Yäo öcO OOösO OO O OO OOd O!O O.Od!O O I O{C) n{= ö - öö - o ü 0 ü € o\ r-f, 6 d oo N n o 6 öl o c] & o 6

aq9o4-., o

o Y!anlEo =o!

* dläz u=u? a ^hI.idie qoq,= üa urikYlo* oQoo^tAta,: g5 9 ö.: r<öxFoI .: .:

äo.!ogi)-i3Oc

=a.-O

-: bO

;d

\a9 3,

ö-o bo

c0

-onh ö.1 \ocl*N9ttoc€inO r);NOor)c$o öncÖ+o\o\ |ö llon Or)C*\O O c.l a- <'

N N€NO

O

öoO

öca

<.

o

boO

E

&I

o

o

0

aN

aEo

ü-.o.=

H

@\o\o o\tt

\o t--

\oovt l-ttnO

dtrr

N6vo\&6

n c.lf\*I

c] @t-o,Q

On*ne Itooonoa-*

oöoöo

OJ

o

:Öcq

n\ooh$Oo+

lmor9<CN

(Jzt4

c]

v0.oN

(u

\o

al

o

do(a

0)o

P.

o

E6

bo

o

ooo0o

o

o

O

Ooü

c)O;]

Og

o

=! _

-E

o

I

o

o

b

öOI

\

ao

j

po

s!

ö4

ö'

{

3

282

KTB-VB

FFfj--l sranites

7aV7 paragneisses

R. PECHNIG EZAT.

ffi syenites, monzonites

F7TA orthogneisses

ll:Tl heterogenousseries

Moodus

E Metabasites

I selected intervals

Depth (m) Cajon

Ir,:'..1

Il' 'l.ll :,1L':l)[,'.]Frr l

Iit]

li'fl';, '1

t,--."1t-:-l

Soultz GPK

IItiiiitr,

tii.il

tiiiil

l1lt1l

lililll-t--,1t-.-.,l

llll

Böttstein

Hliiril

liiil

HN

Schafisheim

uIIHä

Leuggern

nrr-illtrlf'llr- F| "'I

t-+ffi

lrjriN

0

500-

1 000-

1 500-

2000

2500-

3000-

3500,

4000-

Fig. 2. Schematic lithological overview of the boreholes and the intervals selected for this study.

upper amphibolite facies, dated at between 405and 315 Ma ago (O'Brien et al. 1997). Thewhole section is characterized by steeplydipping foliation (50-70') and abundant faultzones resulting from multi-stage brittle defor-mation (Hirschmann et al. 1997).

As part of an extensive Swiss researchprogramme focused on the evaluation of possibleradioactive waste repositories, seven deep bore-holes were drilled in northern Switzerland.Three of these boreholes were selected for thisstudy: Leuggern, Schafisheim and Böttstein.The boreholes reached final depths between1300 and 2500 m. In total. about 3000 m weredrilled through crystalline rocks, most of it com-pletely cored and thus extremely well documen-ted. The boreholes revealed the crystallinebasement as a complex mixture of metamorphicand igneous rocks. The metamorphic units aremainly composed of metasediments with minorintercalations of mafic rocks (Peters et al. 1986,1989a, 1989b). These rocks underwent high-grade amphibolite-facies to granulite-facies

conditions about 500 Ma ago, followed by differ-ent stages of ductile deformation (Thwy et al.1994). During the final stage of the Variscanorogeny (330-315 Ma) large volumes of graniticmelt were intruded into the gneiss series. Theintrusions were accompanied by strong cataclas-tic deformation and hydrothermal overprintingof the metamorphic series. A later tectonohy-drothermal event during the Lower Permianstrongly altered the granites (Thury et al. 1994).

The borehole GPKI is located on the westernside of the Rhine Graben near Soultz-sous-For6ts (France), in a region with an unusuallyhigh geothermal gradient. The well was drilledas part of the Franco-German 'Hot Dry Rock'project, in order to test the feasibility of estab-lishing a deep geothermal heat exchanger.GPK1 was drilled in 1988, reaching a depth of2000m, through 1376m of sedimentary coverinto the underlying granite. Only a few cores,with a total length of 43 m, were taken. Thegranites crystallized about 316 * 7 Ma ago(Chevremont & Genter 1988). The massif is

mainly composed of porphyroid granite withintercalated flne-grained leucogranitic intru-sions, both affected by pervasive alteration.Hydrothermally altered zones of several metresin thickness are frequently observed alongmajor fracture zones, where the granite is signifi-cantly altered (Genter 1989; Genter et al. 1989;Traineau et al. 1992').

The Moodus borehole extends to a depth of1460 m near the town of Moodus in central Con-necticut. USA. In 1987 the well was drilled intothe crystalline rocks of the eastem Appalachians,with the objective of evaluating seismic hazardrisks for the installation of power plants. Ninecores, roughly 2.5 m rn length, were recovered at150 m intervals, and cuttings were collected at6 m intervals (Naumhoff 1988). The drilled lithol-ogy consists of paragneisses and orthogneisses ofacid to intermediate composition, with minorintercalations of mafic metamorphic rocks. Twodifferent terranes of the Appalachian mountainbelt were encountered. In the upper palt, the holepenetrates rocks of the Early Palaeozoic Merri-mack terrane, and passes at about 820 m into theunderlying Avalon terrane (Naumhoff 1988).Based on radiometric age determinations, theserocks are Late Precambrian in age (620 Ma;Wintsch & Aleinikoff 1987). The Merrimack andthe Avalon teffanes are separated by a majorductile thrust fault system called the Honey Hillfault zone. Mineral assemblages associated withthe fault zone indicate that the dominant ductilefabric developed under amphibolite-grade meta-morphic conditions about 390 Ma ago (Naumhoff1988). In contrast to the other boreholes described,the rocks drilled in the Moodus borehole onlylocally show a greenschist-facies ovelprint, andbrittle deformation is rare.

The Cajon Pass scientific borehole, 4 km northof the San Andreas Fault in California. wasdrilled to measure the heat flow and state-of-stress at seismogenic depth to explain the 'heat-flow-stress paradox' observed in the SanAndreas fault zone (Lachenbruch and Sass

1988). The borehole was drilled to 3.55 km intoa series of plutonic rocks and orthogneisses.Information on lithology is mostly based on cut-tings, since only 37o of the borehole was cored.The rocks are orthogneisses ofgabbroic to grani-tic composition, and were overprinted underamphibolite-facies conditions 75-81 Ma agoalmost sychronously with dioritic intrusions(Silver et a/. 1988). A Late Miocene tectonicphase resulted in the formation of 'low anglefractures', which, during the Pliocene, acted aspathways for hydrothermal fluid circulation andcorresponding zeolite-facies alteration (Vincentand Ehlig 1988).

283

Mineralogical and geochemical

composition of the rocks

Geochemical data-sets and information on rockmodal composition were collected from the lit-erature (Matter et al. 1988; Peters et al. 1989a,ä; Peters et al. 1986; Genter et al. 1989; Sllveret al. 1988; Ambers 1989; KTB-Data CD). Therock modal data comprise results from pointcounting and X-ray diffraction analyses. Theaverage geochemical and mineralogical compo-sitions for 25 rock types are displayed inTables 2 & 3. The bulk composition of therocks can be summarized as follows.

The biotite granites and aplites of the bore-holes: Schafisheim, Leuggern, Böttstein andSoultz-sous ForCts GPKl show comparablegranitic chemical signatures, with tendenciestowards alkaline composition for the Schafisheimgranite and aplite (Fig. 3). The main chemicaldifferences concern the K2O content, whichcorresponds to variations in the averagepotassium-feldspar volume between l77o and4lVo. The variations in K-feldspar contents arecaused by the substitution of K-feldspar by plagi-oclase, while qtartz and biotite proportionsremain almost constant in the granites. Com-pared to the granites, the aplites have lowerbiotite contents.

The monzonites and syenites of the Schafisheimborehole were classified after Streckeisen withregard to their modal composition (Matter et al.1988). Low quartz contents and high amphibolecontents separate them from the other plutonicrocks compiled for this study. Although themonzonites and syenites show considerabledifferences in their modal composition, theiraverage chemical composition is quite similar,both plotting into the syeno-diorite field(Fig. 3). Larger differences between syenitesand monzonites concelx the potassium andsodium concentrations controlling the varyingpotassium-feldspar/plagioclase ratio of theserocks (Table 3).

The rocks of the orthogneiss group span a suitefrom almost gabbroic composition to leucocraticgneisses of granitic composition. Their chemicalsignature is widely controlled by the stage ofmagmatic differentiation of the plutonic precur-sor rocks as visible for the Cajon Pass orthog-neisses (Fig. 3). The average element contentsand modal composition of the Cajon graniticgneiss are in the range of the biotite granitesfrom the other boreholes. Higher quartz con-tents are documented for the Moodus graniticgneiss. This observation is in agreement withthe higher SiO2 values measured in this rocktype.

LOG RESPONSES : ACID/INTERMEDIATE ROCKS

oO

a

Va5;_qRüxLoo

-.10

<.;95o

-=

q9.iö,^ : +..-sd o

s.:= >si--:i?

v ö'Ä

+o9>

ö al =.!i9:::2A;?! SO o %

&Eiled-: oi/::'--to:€ 4: @9€;x - ! o

E;F€?!r O Ca ! n! : ! o 9!!r-i 9 - i 9

<: o E ,, ,4o,^t=,r

=9P::EEai d A X Ao o* > i uC d 6 d r!

F ! >>5,&,3 t!,.7 2

R. PECHNIG ETAL

30033330 0000999 000s93s9seI I I lStsl | 3"1 | I I I I O:="

do -o<-oNrll ljrr 2222tt t r r:::.N*

l== lll I llll

,** r r r ({({ r rlvv I r r zzzzt I

|| rr l lllll

rllt i| 2222ttt

NhÄ r r- o roll I

-Onahn ' ' r N rvv;v I lJ v I

r,ä,äl r rr onl+-ll

OOO r @*rjji r rj jj

oo€+,lI o zzzz-t --l-=19 o

VV\\

:1 rh6 16o ? annonhono@Oi<;;; a\ö= ! n \t$6oNNölrNiiX r r r r 4dddd I r f L I I I{Nnn3--Nqil2777- | .!6nOOCOOnOO

: ;-- \ t---.. i-d N-S€rS - * -o n r =t - A -- ooNnv Xnn€\o\oooo i' cli-N-N+N --!+nnd--- ----N---++h

onnrrohn N n n n rh++NonoÖ Ö o o oFttrt

RKglgRg- zzzz?2Ä flol SolNo'vv

llll ll'I ll'l -:1:^l- n

lltl

A-@€-\OhN {Orro INN N I

@oh\oo60*hoN6_ NI

-h@6€6rr I i\O-+h+ S l€ i nh iN-

+oo-d-N- | :-+f,-,f. l-; rN

oor€od€@ lo*6<'Oc] Oo€o*@+\oNhdNNd No lNNclooo $ciN*oc.lc]o*o

Nodo€*n6 x * * * o*o 6* -Ä=9SN*i d

ä r

oc)oh+ooo o h hhhn6o€doQ-<++--<.h 444 j- - **d-***nn\O

I r I r I ;!>+> I I I I r I I I j

OOOvr€O€n -ZLZ-WA hhC)6Nhg€ONiööo+ Nol ooo dd:-odNoN$

NOo€h OO n 0 Öoonoono*c----J6-+ 411j- n-+h**o+N+, I I I I ;'>+) I r I I I I I I I I l

nnhO_On<. L4zzaae OhnnOhnOoO.j* r-;N N6 Oo0 O-:ClOiio o

J ..-!JJ = = r -:.--1 ::: ; ;

E Y=YEEE ry s,? iiä E E eYet ! ee EEet-:3ötAt 3 üüji=== .:jjt=.:VtUt!

!.Er s *is .: + -u -\ .SI P . \ :*. -: E so.Y Ss" i:\! ä* F t T*.: lJFr siIö öF* tlFs ü! s s R:G Sä== üööö vd a t vö

o

791.9;it2=*l+ts I

^o^

oo!coc3Evö

o

o

oo

o

H

elsl.9 Iboldl

ao

N

O

ä0oo

J

281

x

:ö4p

\o

o

ö

o

OOap

o

ato

Fi

LOG RESPONSES: ACID/INTERMEDIATE ROCKS

Table 3. Mean major-element values of the selected rocks

Lithology Hole SiO2 Al2Or KzO Na2O CaO Fe2O3 FeO MgO TiOz Ref

Biotite grdnite

MonzoniteSyeniteAplite

Plutonic rocks13.9 5.1 3.4 1.614.1 5.2 2.9 1.514.3 4.9 2.1 1.414.2 4.1 3.9 2.214.3 5.1 2.7 s.712.5 6.2 1.5 6.114.3 4.6 4.5 1.513.6 1.3 3.1 0.9Orthogneisses14.8 t.7 2.9 '7.7

t6.9 2.6 4.1 5.315.7 3.4 3.5 5.114.0 4.3 3.4 2.013.8 5.4 NA 2.3t2.9 4.5 NA 3.7t2.1 4.8 NA 2.0Paragneisses12.0 2.4 2.6 1.119.5 4.2 0.9 2.515.8 2.6 2.7 1.3t2.3 3.4 NA 4.8r1.7 2.1 3.5 3.51.5.0 2.o 2.9 5.214.0 2.'7 NA 5.215.5 3.3 3.7 5.2r 6.0 1.7 4.9 2.r

Schaf.Leug.Bött.GPKlSchaf.Schaf.Schaf.Leug.

9.2 0.80.1 2.10.5 t.64.3i.8 5.91.8 5.90.1 1.1

0.5 0.6

9.55.35.2r.4

1.2 0.51.3 0.50.9 0.41.5 0.71.1 1.510.1 1.50.1 0.30.4 0.1

8

10I2

3

8

t9'7

13

I5

61

.298* 577*586

69.069.17 t.166.850.149.369.774.1

53.860.361 .5

71.171 .6'73.5'74.4

(1)(2)(3)(4)(l)(1)(1)(2)

Gabbroic CajonDioritic CajonGranodioritic CajonGranitic Cajon

Mood.K-feldspar Mood.Quartz-feldspar Mood.

( Sillimanite)-biotite Leug. a

Leug. bKTBMood.

Hornblende Leug.KTBMood.

K-feLdpar KTBQuartz-plag. KTB

t0 74.218 60.332 64.4

*2541 70.83 60.37 56.0

-2707 68.38 52.42 66.8

0.53.01.6

0.81.66.04.73.28.33.58.63.9

2.44;7

4.8

5.1 1.0 (5)2.6 0.8 (s)3.0 0.1 (s)0.s 0.2 (5)0.1 0.11.6 0.54.6 0.3

t.2 0.5 (2)t.9 0.8 (2)2.4 0.9 (6,7)2.7 0.83.s t.4 (1)3.8 1.3 (6,8,9)5.6 0.63.1 1.1 (6,8)2.7 0.s (6,10)

References (.t) Matter e/ al. (1988); (2) Peters €/ al. (1989ö): (3) Peteß et al. (1986); (4) Tratneat et ql. (1992)t (5) Silver er a/. (1988);

(6) KTB-Data CD; (7) Heinschild et ql. (1988\; (8) Stroh & Tapfer (1988)l (9) Tapfer et al. (1989); Wittenbecher et al. (1989).

Leug. a: plagioclase-biotite gneiss (metagreywacke).

Leug. b: cordierite-bearing sillimanite biotite gneiss (metapelite).*Mean values calculated from geochemical logging data.

NA: Infomation is not available-

Compared to the selected plutonic rocks andorlhogneisses, the paragneisses show a morecomplex mineralogical composition. Besidesthe main components (quartz, feldspar, biotiteand amphibole in varying amounts), white mica,sillimanite, cordierite and garnet also occur. Thepresent mineralogical composition is mainlycaused by the original sediment bulk compo-sition, but is also affected by the degree of meta-morphism. The biotite gneisses show strongvariations in quartz content (15-55Va) and theamount of biotite, white mica and sillimanite(Hirschmann el a/. 1997; Naumhoff 1988;Peters e/ al. 1989a). Besides the migmatiticversion of the Leuggern paragneisses, all othertypes contain almost no K-feldspar. Pyroxene isobserved only in the Moodus biotite gneisses.

The chemical composition of the biotite gneisses

of the different boreholes varies significantly,colresponding with changes in the modalcomposition.

Homblende paragneisses were drilled in theboreholes Leuggern, KTB and Moodus. The

precursor rocks are siliciclastic sedimentsmixed with mafic rock components of volcanicorigin, such as ashes or volcaniclastics (Harmset al. l99l; Hirschmann et al. l99l; Naumhoff1988; Peters et al. 1989a). The relative pro-portions of siliciclastics and volcanic com-ponents determine the overall chemistry andmineralogy of the hornblende gneisses. Thus,they show considerable differences in theirmean mineralogical composition, particularlyregarding the quarlz, amphibole and biotitecontents.

Pelrophysical characteristics of the rocks

A wide spectmm of logging operations werecarried out in the selected boreholes, includingthe recording of acoustic, electric and nucleardata. The logs that were evaluated for thisstudy (Table 4) were performed by the followingSchlumberger tools: the dual laterolog (DLL);litho density tool (LDT); natural gamma-rayspectroscopy tool (NGT); compensated neutron

286 R. PECHNIG EZÄT.

40

+ biotite-granitex monzonitex syenite< apliteo gabbroic gneiss

^ dioritic gneissv qranodioritic gneiss

SiO2 (wt %)

granitic gneisspotassium feldspar gneissquartz plagioclase gneissplagioclase-biotite gneiss(cordierite)-sillimanite-biotite gneisshornblende gneiss

BcGKLS

*

oao

BöttsteinCajonGPK I

KTB-VBLeuggernSchafisheim

Fig. 3. Total alkali versus silica diagram for the chemical analyses of rock types selected for this study (modified afterWilson 1989).

tool (CNT); and different sonic tools, such as thedigital sonic array tool (DST) and the compen-sated sonic tool (CST). General information onlogging tools and their measurement principlescan be found in Ellis (1987), Rider (1996) andHearst & Nelson (2000).

Log analysis was carried out by applying theelectrofacies method, originally introduced bySerra (1984) for log interpretation in sedimentaryrocks. Like a lithofacies description, whichcomprises several petrographic and chemical

Table 4. Overview of tools and logs used for this study

characteristics, the electrofacies method com-bines the petrophysical value ranges recordedby different logging tools. An electrofaciescharacterizes a rock type by a specific set oflog responses, and thus allows it to be distin-guished from others. The concept is not restrictedto the use of electrical measurements. Any 1og

property suitable for the characterization of a

rock type may be used. This study incorporatesthe log curves for total gamma-ray, potassium,bulk density, neutron porosity, electrical resis-

Tool Log Principle

DLL (dual laterolog)SDT (Sonic digital tool)

DSI (Dipole shear sonic imager)CNT (compensated neutron tool)LDT (litho density tool)

NGT (natural gammaspectrometry tool)

LLD tohm m) ,

DTCO (u,s m ')

NPHI (7o)

RHOB (gcm ')

SGR (API-)

POTA (.Vo)

THOR (ppm)URAN (ppm)

Laterolog deepCompressional-wave

travel time

Neutron porosityBulk density

Spectral (total)gamma-ray

Potassium contentThorium contentUranium content

Electrical resistivityTravel time of sound

Absorption of neutronsAb sorption/ scattering

of gamma-raysNatural gamma-ray

emlsstons

*API: American Petroleum Institute

tivity and P-wave velocity as being standardlogging parameters.

Log analysis was carried out separately foreach borehole. All available information oncore and cuttings descriptions was collectedand, using core-log-correlation, the in situlogging data were assigned to the different mag-matic and metamorphic rock types. A total of 25electrofacies were defined. Tables 5-7 displaythe log value statistics for the rock types sep-arated within the different boreholes. Theminimum, maximum, mean values and standarddeviation are given for each electrofacie. Thedata are calculated from the borehole sectionsmarked in Figure 2, taking into account the selec-tion criteria defined on p. 280. Figure 4 displaysthese values graphically. The similaritiesand differences in the borehole geophysicalcharacteristics can be summarized as follows.

The gamma-ray values cover a large valuerange (40 550 APD within the plutonic rocks

28'7

and gneisses. The average garnma-ray values ofthe plutonic rocks are about 250 API for thegranites/aplites and about 350 API for thesyenites/monzonites, and are thus signiflcantlyhigher than the gamma-ray mean values of allthe gneisses. Highest values occur in theSchafisheim monzonites, with about 550 API.Besides the granitic gneiss and the K-feldspargneiss of the Moodus borehole, all of the othergneisses have lower gamma-ray values (<210API) than the igneous rocks. The potassium logresponses in general correlate with the gamma-ray ones, showing on average higher potassiumcontents for the plutonic rocks than for the gneisses.

The investigated rocks have bulk densityvalues from 2.30 g cm ' to 2.95 g cm ' withstrong variation within the three groups of plu-tonics, ortho- and paragneisses. The averagebulk density of granites and aplites rangesbetween 4.59 - 0.02 g cm ' and 2.65 -0.02 gcm-'. which is in excellent agreement


Table 5. Electrofacies statistics displayed for the different rock types. Minimum, marimum mean,

and standard deviation of gamma-ray and potdssium log responses

Lithology Hole Total gamma ray (API I Potassium (7o)

Min Max Mean SD Min Max Mean SD

BiotiteGranite

LeucograniteMonz.oniteSl,eniteAplite

GabbroicGranodioriticDioriticGranitic

K-feldspttrQuartz-.feldsp.

(Sillimanite)Biotite gneiss

Hornblende gneiss

K-feLdspar

Quartz-plag.

Plutonic rocks187 356 259 32163 312 257 1'.7

181 304 230 i9t96 308 260 18

130 280 225 35r21 548 333 64215 510 332 58168 329 244 33151 333 252 32

Orthogneisses44 100 73 11

66 2r0 91 20s9 r98 91 1680 165 106 15

156 433 202 31

111 363 t't6 4564 t34 i09 12

Paragneisses82 t92 l4l 1690 211 151 19'72 136 99 11

98 t67 t21 1075 169 133 22539977835 135 81 t619 185 138 2666 100 76 1

462 3.55 5.82984 4.31 6.92662 4.31 6.59

1005 3.13 7.93320 3.87 6.36456 4.31 8.91532 4.41 8.44242 4.06 6.62406 3.58 6.79

61 1.13 2.51720 2.36 4.0'1

1028 2.03 3.51343 3.35 4.s2298 3.43 7.99418 2.57 5.'7 5

264 t.7'7 5.3 r

903 2.33 5.761546 1.99 6.t418.56 1.44 3.34'755 2.06 4.22130 1.50 4.04664 1.33 2.57

1116 1.11 3.91101 1.83 5.62I 09 1.35 2.46

4.19 0.40 4625.85 0.38 9845.35 0.40 6624.96 0.50 10055.30 0.50 3206.30 0.90 4s66.84 0.'74 5325.29 0.39 2425.57 0.59 406

I .95 0. 18 613.21 0.28 7202.68 0.21 10284.03 0.28 3434.57 0.65 2983.62 0.55 4484.05 0.78 264

3.90 0.54 9034.32 0.61 t5462.39 0.38 18562.91 0.33 7552.53 0.63 130r.88 0.23 664234 0.46 11163.85 t.02 1011.8 0.05 109

SchafisheimLeuggerrrBöttsteinGPKlGPKlSchafisheimSchafisheimSchafisheimLeuggem

CajonCajonCajonCajonMoodusMoodusMoodus

Leuggern*Leuggern'KTB-VBMoodusLeuggernKTB-VBMoodusKTB-VBKTB-VB

*Plagioclase biotite gneiss (metagreywacke).'Cordierite

bearing sillimanite biotite gneiss (metapelite)

SD: standtrd deviation.

288 R. PECHNIGEZAI.

Table 6, Electrofacies statistics displaved for the dilferent rock tlpes. Minimum, maximum mean, andstandard deviation of density and neutron log responses.

Density (g cm 3) Neutron porosity (7o)

Min Max Mean sl-) Max Mean SD

BiotiteGranite

LeucograniteMonzoniteSyeniteAplite


K-feldsparQuartz-feldsp.


Hornblende gneiss

K-feldsparQuartz-plag.

Schafisheim 2.55Leuggern 2.52Böttstein 2.45GPK1 2.58GPK1 2.54Schafisheim 2.58Schafisheim 2.61Schafisheim 2.55Leuggern 2.51

Cajon 2.58Cajon 2.31Cajon 2.30Cajon 2.38Moodus 2.59Moodus 2.51Moodus 2.53

Leuggern* 2.49Leuggern' 2.50KTB-VB 2.40Moodus 2.68Leuggern 2.47KTB-VB 2.68Moodus 2.54KTB-VB 2.58KTB-VB 2.63

Plutonic rocks2.64 2.61 0.022.66 2.60 0.022.64 2.59 0.022.73 2.65 0.022.70 2.60 0.022.90 2.79 0.052.94 2.80 0.072.66 2.61 0.022.71 2.61 0.03

Orthogneisses2.92 2.73 0.092.19 2.58 0.092.9t 2.68 0.102.71 2.55 0.072.10 2.62 0.012.8r 2.71 0.052.85 2.68 0.07

Paragneisses2.81 2.68 0.042.8t 2.7 0.042.92 2.'74 0.042.82 2.'76 0.022.85 2.'72 0.063.01 2.86 0.052.95 2.73 0.062.87 2.69 0.06274 2.68 0.02

462 0.6 4.0984 0.8 1.5662 t.4 13.0

1005 0.7 5.9320 0.3 2.9456 6.5 16.1

532 3.2 205242 1.2 2.8406 0.8 8.9

6l - 0.1 9.6120 - 0.1 4.91028 -0.2 8.3343 - 1.3 2.4298 - 0.6 0.9448 0.6 4.3264 -0.9 2.0

903 3.6 19.61546 4.6 24.41856 2.8 14.6

7 55 t.4 5.8130 0.9 t9.7664 6.1 15.9r116 -0.6 5.6101 1.1 15.6109 3.8 9.1

t.'73.43.71.80.99.8

10.41.82.9

2.10.21.60.50.12.t0.1

0.6 4621.3 984t.6 6620.8 10050.3 3201.8 4562.9 5320.4 2421.8 406

2.2 61

0.9 1201.4 10280.5 3430.2 2981.0 4480.6 264

8.0 2.5 90312.5 3.9 1,5468.7 2.3 18563.2 0.8 7556.0 4.5 130

11.1 1.5 6641.1 0.9 r1165.9 3.1 101

6.3 1.2 109

*Plagioclase-biotite gneiss (metagreywacke).-Cordierite-bearing sillimanite-biotite gneiss (metapelite)

with granite density data from " laboratorymeasuremenls (2.60 I 0.07 g cm 'l Landolt-Bömstein 1982). Monzonites and syenites aredistinctly separated from lhe acid plutonics byhigher-density values of 2.79 + 0.05 g cm 'and 2.80 r 0.07 g cm '. Most of the gneissicrock types show broad value ranges, frequentlyskewed by very low-density values. This is par-ticularly the case lor the Cajon orthogneisseswith density records lower than 2.45 g cm '.Although we restricted the data selection to inter-vals of good borehole conditions (caliper devi-ations {107o of bit size) and a general lowtectonic overprint, low-density values wereregistered. They are explained as being relatedto small joints and fissures or borehole wall irre-gularities not detected prior to data analysis.

The P-wave velocity values of the selectedrocks are between 3.ikms 1 and 6.9 km s l.

Mean values of the different rocks do notscatter signiflcantly, and are between 5.1 *

0.03kms-r and 6.1 +0.02kms1. Suryris-ingly, P-wave velocity values are lower in themonzonites/syenites than in the granites/aplites, although the monzonites and syeniteshave higher density values. Values of the orthog-neisses are hish. with mean values of more than5.8 + 0.02 kÄ, ' and minimum values notbelo 5.0 km s-1. Besides the homblendegneisses. mean values of the paragneiss groupare below 5.7 km s-' and in a similar range tothe plutonic rocks. Very low values (<4 km s 'tare only observed for the Leuggern biotitegneisses and the KTB potassium-feldspar gneiss.

Beside the gamma-ray and the potassium logthe neutron porosity exhibits the strongest differ-ences between the rocks selected for this study.Most paragneisses are signiflcantly separatedfrom the orthogneisses, granites and aplites bytheir high neutron porosity values of up toalmost 25Vo. Only the monzonites/syenitesreach comparable value ranges. Granites,


Table 7. Electrofacies 'ttatistics displal'ed for the different rock types' Minimum, maximum mean, and

standard devintion of P-wave vektcity and electrical resistivity data

I-itholosv Electrical resistivity (1og ohm m) P-wave velocity (km s ')

Min Max Mean SD

289

Hole

SDMeanMaxMin

Biotite granite

LeucograniteMonzoniteSyeniteAplite


K-feldsparQuartz-plctg.


HornblendeGneis.s

K-feldsparQuartT-plag.

Schafisheim 1.6Leuggern 2.4Böttstein 2.6GPK1 1.8GPK1 2,3Schafisheim 1.3Schafisheim 1.8

Schafisheim 2.6Leuggern 2.6

Cajon 2.4Cajon 2.2Cajon 2.0Cajon 2.8Moodus 3.0Moodus 3.3Moodus 2.9

Leuggem* 1.2Leuggem' 1.3KTB-VB 1.5Moodus 3.5Leuggern 2.5KTB-VB 1.7Moodus 2.8KTB-VB 2.3KTB-VB 2.3

Plutonic rocks2.6 2.0 0.i 462 4.74.8 3.9 0.1 984 4.94.4 3.5 0.4 662 5.34.7 3.3 0.6 1005 5.34.6 3.3 0.5 320 5.33 2.1 0.4 156 4.9t.4 2.9 0.2 532 1.43.9 3.3 0.2 242 5.24.1 3.5 0.5 406 5.0

Orthogneisses3.8 3.2 0.3 61 5.54.6 3.9 0.6 720 5.04.6 3.5 0.6 1028 5.04.6 4.5 0.3 313 5.24.6 4.2 0.2 298 5.64.0 3.8 0.1 448 5.24.8 3.8 0.4 264 5.1

Paragneisse s

4.6 3.1 0.5 903 3.84.1 3.0 0.5 1546 3.24.6 3.4 0.1 1856 4.84.5 4.0 0.2 755 5.04.1 3.8 0.6 130 5.04.9 3.6 0.5 664 5.65.0 4.0 0.4 1116 5.33.4 2.9 0.4 r01 3.93.8 3.4 0.3 109 5.5

5.3 0.2 4625.6 0.2 9845.6 0.2 6625.8 0.2 1005s.8 0.2 320s.5 0.2 4565.4 0.2 5325.'.7 0.1 2425.5 0.2 406

5.'7

5.9596.36.15.95.96.05.9

6.56.56.96.46.46.06.4

6.36.06.26.05.86.66.65.85.8

6.r5.85.85.95.95.66.r

5.35.15.65.55.66.16.15.45.1

0.2 6l0.2 7200.2 10280.2 3430.2 2980.1 4480.2 264

0.2 9030.3 t5460.2 18560.1 '7 550.2 1300.2 6640.2 11160.3 101

0.1 109

*Plagioclase-biotite gneiss (metagreywacke).

Cordierite-bearing sillinanite biotite gneiss (metapelite).

aplites and orthogneisses generally do not exceed10Vo. It is noticeable that the neutron porosityvalues of the Moodus borehole also show lowvalues of less than 60/o for the paragneisses.

Electrical resistivity values .vary by sereralorders of magnitude. lrom l0' to l0) ohm m.

They do not show any significant differencesbelween the different rock types. Tn most cases

mean resistivity values are above l0'ohm m.

which is characteristic for massive igneous andcrystalline rocks known from laboratorymeasurements (Landolt-Börnstein 1982). Sig-nificantly lower value ranges are observed forthe Schafisheim granite, monzonite and syenitesthat point to borehole-specific influences, such

as an overall higher microcrack density orstronger alteration.

Integrated analysis of log and rock data

The geochemical, petrographic and boreholegeophysical data were compared in order to

extract significant relationships between rockcomposition and log responses. This was per-formed using corelation analysis. Pearson corre-lation coefficients were calculated using theaverages of logging, geochemical and mineralo-gical values of the different rock types. Besidesthe single mineral phases, corelation coefficientswere also calculated for the mineral groups:(1) quartz, plagioclase and K-feldspar; (2) thetotal amount of biotite, muscovite, chloriteand amphibole; and (3) biotite, muscovite andK-feldspar. The results are given in Table 8.

Results of correlation analysis between the logdata and mineralogical and geochemical rockcomposition will be discussed and interpretedin the following sections.

Effects of rock mineralogy on log responses

The correlation between mineralogical rock dataand wireline data reveals the following systema-tic observations.

290 R. PRCHNTG ETAl,.

Fig. 4. Electrofacies of the different rock types. Displayed are rninimum, mean and ma-rimum values for thestandard logs.

IOo.oCofE

rock type

gamma ray (APl)

0 200 400

density (g/cm3)

2.4 2.6 2.8 3

p-wave velocity (km/s

456

biotite granik

leucogranit(monzonit(

syenit€aplik

a

L

B

GSJa

L

-----r-'---

----T--

--

--1--_

-'...--t_

-T--T-

---'--T--

--_--------------T

-

------

----

---- ----

uoaIoc

ctro

gabbroi(granodioriti(

dioriticgraniti(

K-feldsparre rtz-falr'lcna r

U

(-M

MI\il

TT--l---T--

----

--

_-l-

--'----

-_T---

--

--'--'r-'--

-0)@60)coN(Eo

(sillimanitebiotite

hornblende

K-feldsparouarlz-olao

L

L

KM

L

KMKK

-'--T--

T--T

-T-

-'--r

-.T_-=------ ---- --'---

--

__T

----

-- -T-

---l_

@Yoo_oco=o

neutron porosity (%)

0 10 20

resistivity (log ohmm)

12345potassium (%)

2468

granik

leucogranitemonzonit(

syenit(aplit€

biotiteL

B

GöSa

L

T-

----T_T-

--

--

T'-.r.-

-T--

--.--..-r.-----

-

_-----

------

----

--------T-----

--

____-'-T---

--gabbroirgranodioriti<

dioritirgraniti(

K-feldsparquartz-feldspar

c

MMM

--T_-r-

T-7'-'r-

-

-_T_

---T-.---

-T---T-"-----

T-----T----

-- -T---r---

---.-r_

oa0)c(g(uo

(sililmanrrebiotitt

horn blend<

K-feldspaquartz-plag

LiL --l

K -.1

YJK --l

M]K --l

6-J

--T----

-

.-

-- -T_

--'T--

-_---------.------.

---.T----

-_-._--

--T- ------

--T_

-----_T_---r--

------T-

S: Schafisheim L: LeuggernC: Cajon M: Moodus

B: BöttsteinK: KTB-VB

G: GPK 1

LOG RESPONSES: ACID/INTERMEDIATE ROCKS 291

Table 8. Pearson correlation coefrcients calculated between mean values of geochemical, mineralogicaland borehole geophysical data-sets

Oxide/mineral Gammaray

Density P-wavevelocity

Neutron Potassium Resistivityporosity

sio2Al:o:KzoCaOMgoTi02QuartzK-feidsparPlagioclaseBiotiteMuscoviteChloriteSillimaniteGametAmphiboleMica*Chl.fAmp.Mica * K-feldsparQuartz * K-feidspar f Plag

- 0.0570.282

'0.807- 0.201

0.1 250.057

- 0.304

'0.634* 0.5910.201

-0.t]2-0.342

o.044

- 0.1 17

- 0.056

-0.042'''0.746

0.019

r -0.5'79-0.012-0.287*0.597

'o.iog''0.162

- 0.3000.369

-0.1110.2160.1170.3300.0490.109

*0.546*0.619

-0.227' 0.6i4

0.094

- 0.1 39

- 0.3 15

0.2680.102

-0.t440.0590.0920.311* 0.588

- 0.418

- 0.012

-0.137-0.412

0.2860.212

* -0.500

0.153

- * 0.5820.307

-0.1460.1550.,110

'0.665-0.265

0.296

- 0.315*0.500*0.510

0.4060.445

.0.610

0.254''0.704

0.065| 0.6'71

0.032 0.488

- 0.259 - 0.105r"0.874 - 0.213-0.218 0.1160.060 0.332

- 0.099 -0.417-0.26'7 -0.496io.6gl o.ogz

* - 0.568 0.2390.239 -0.2'75

-0.144 - 0.1 19

0.391 -0.1080.053 -0.159-0.026 0.152- 0.t14 0.2450.108 -0.401*0.825 0.2'720.120 0.410

*Conelation is different from zero at a significänce level of 0.05.Corelation is different from zero at a significance level of 0.01.

The most sensitive log for rock matrix vari-ations for all gneisses and plutonic rocks is thegamma-ray log and, in particular, the potassium1og, showing the highest corelation coefflcients(r > 0.8) between rock and log data. Bothparameters show comparable correlation coef0-cients with rock chemistry and mineralogy. Sig-niflcant positive correlations are calculatedbetween the K-feldspar content and the contentof the mineral group K-feldspar plus biotite andmuscovite. The relation to rock mineralogy isobvious, since K-feldspar, biotite and muscovitecontain a considerable amount of potassium intheir crystal lattice. Figure 5b shows the corre-lation between the potassium log and the totalamount of K-feldspar plus biotite and muscovite.With regard to rock type and mineralogical com-position (Table 2) the amount of K-bearing min-erals controls the potassium and gamma-ray datato different degrees. In the acid orthogneisses,the rocks' potassium content is mainly relatedto the K-feldspars, while potassium content andgamma activity in the intermediate to basicgneisses is dependent on the amount of biotite.

The only significant negative correlation iscalculated for the potassium, respectively thegamma ray versus the plagioclase content(Table 8). This can be explained by thesubstitution of K-feldspar by plagioclase inigneous rocks during magmatic differentiation,bearing in mind the rock classification afterStreckeisen (196'7).

Table 8 also reveals significant correlationsbetween neutron porosity and rock composition.The highest positive correlation (r : 0.7) wascalculated for the neutron log versus the amphi-bole plus biotite and muscovite content.Neutron porosity increases with the increasingamounts of these minerals (Fig. 5d). Rockswith high mica and amphibole contents, such as

hornblende gneisses, biotite paragneisses, sye-nites and monzonites (see Table 2), plot in theupper right corner of Figure 5d, while granites,aplites and granitic rocks plot in the lower left-hand corner. In contrast, the neutron log isnegatively correlated with the quartz plusK-feldspar and plagioclase content (Fig. 5c).Neutron porosity decreases with increasingamounts of quartz plus feldspar. In consequence,most granites, aplites and gneisses of graniticcomposition plot into the lower right edge ofthe cross-plot, while syenites/monzonites andmost paragneisses have higher neutron andlower quartz/feldspar values. The high neutron-porosity values of some rock types and theobserved correlation to rock mineralogy canbe explained by the physical principles of theneutron-log measurement. Here, the hydrogencontent is the most important factor controllingthe neutron-log responses of a formation, sincefaslemitted neutrons are slowed down predomi-nantly by hydrogen nuclei occurring in the for-mation. Hydrogen can be kept in a formation inthree different ways (Rider 1996):

a)

r=0.746 ox j

,G .1,< + ,cL S *B"

I\4x

Mu

LK''tN'4" e -a[/

^K*U*"t*K' j.L

"cK

292 R. PECHNIG ETA'

* granitic gneiss* K- feldspar gneiss* quartz plagioclase gneisso plagioclase-biotite gneissr (cordierite)-sillimanite-biotite gneissO hornblende gneiss

b)

r=0.825 ^S5X

L+B+

+ö

c*

L<S

+WL.MK *-,

IV

M.., + L

'uÄ K.c s

.KvK

350

E.=^a(g

o

o- zsosE zoo(E

Ech tsoo)

50

10

=a^oooo,cof-c)to

-2

^ 10

=a96ooc'o

=zc)c0

020406080K-feldspar + biotite + muscovite (vol.%)

40 60 B0 100

quartz + plagioclase + K{eldspar (vol.%)

+ biotite-graniteX monzonitex syenite{ apliteö gabbroic gneissa dioritic gneissv granodioritic gneiss

020406080K-feldspar + biotite + muscovite (vol.%)

0204060biotite + muscovite + amphibole (vol.%)

B BöttsteinC CajonG GPKIK KTB-VBL LeuggernM MoodusS Schafisheim

Fig. 5, Cross-plots of log responses of garnma ray, potassium, and neutron porosity, versus mineralogical compositionof the rocks. Displayed are the average data from Tables 2 and 4. (a) & (b) Show positive trends between the gamma-ray/potassium logs with the total amount of micas and potassium feldspar within the rocks. (c) Shows the strongnegative correlation between the neutron-porosity 1og and the quartz plus leldspar content. (d) Reveals the positiveconelation between the neutron log and the total amount ol OH-group bearing minerals. The conesponding Pearsoncorrelation coefücients are given in Table 8.

(1) as free water in pores, fissures andfractures:

(2) as bound water incolporated within min-erals; and

(3) as part of the crystal lattice in the form ofOH-groups.

Since data selection was restricted to fresh. tecto-nically almost undisturbed crystalline rocks witheffective porosities usually <37o, the effect onthe neutron 1og of free water in pores and f,ssuresbecomes negligible. This makes it evident thatthe recorded high neutron porosities in the crys-

r=-0.671 L.

s.. ^+\5K.oL

Le

oC

r=0.704Ks

*K

B'!+ Ma

< +i* lVsc 0cV

c

talline rocks are not real in the sense of indicatingrock porosity, but they are the result of rockmatrix effects. Minerals with bound water areclay minerals such as montmorillonite and min-erals of the zeolite group; however, neitheroccur in the investigated rocks. Therefore, onlyOH-bearing minerals can contribute significantlyto the total hydrogen content in crystalline rocks.This fits with the results of Table 8, since amphi-bole and micas have OH-groups incorporated intheir lattice and their concentration is positivelycorrelated with the neutron log. On the contrary,quartz and feldspars are free ofhydrogen and theocculrence of this mineral group is negativelycorrelated with neutron porosity.

There might be an argumeni that, besides thehydrogen in the formation, the atomic weightof the chemical elements also influences the(thermal) neutron log response (Schlumberger1994), as there is a trend that elements withhigh atomic weights capture neutrons moreeffectively. Furlhermore, the presence ofcertain rare-earth and trace elements withparticularly large capture cross-sections (e.g.boron, lithium and cadmium) can have a signifi-cant effect on the neutron-log readings (Harveyet al. 1996'1. Rare-earth elements occur in acid/intermediate rocks, sometimes making quite asignificant contribution. It can be assumed thatthe high neutron-log readings in crystallinerocks are produced by superposition of the differ-ent effects of hydrogen content, atomic weightand rare-earth element concentrations. Furtherinvestigations would be needed to separate theneutron-porosity readings for the differenteffects.

Correlation analysis has also revealed cone-lations between the density and rock mineralogy.A positive correlation (r - 0.546) is calculatedfor the relation between density and the amphi-bole content. A more significant negative corre-lation was calculated between density and thetotal amount of quartz and feldspar. The calcu-lated results are explained by the differentmineral densilies. increasing from potassiumfeldspar 12.56 g cm ') to quarrz (2.65 g cm ').to plagioclases .(2.61 2.76 g cm-') and micas(2.7-3.1 g cm 't. up to the mafic minerals ofthe amphibole group (3.15 1.25 gcp 3) andthe pyroxene group (2.9-3.5 g cm-') groups(Wohlenberg 1982; Schön 1996). Therefore,quartz/feldspar-rich rocks show low bulkdensity values, and maflc rocks with consider-able amounts of amphibole/pyroxene have highrock densities.

Correlation analysis shows no significant cor-relations between P-wave velocity and mineralcomposition. A weak correlation was calculatedonly for the relationship between P-wave vel-

293

ocity and biotite and mica contents, respectively.These correlations are negative, indicating thataverage P-wave velocity decreases with micacontent. At the moment there is no plausibleexplanation for this. In this study, corelationanalysis reveals that the P-wave velocity log isnot indicative of rock composition. However, itis known from laboratory measurements thatrock-building minerals have characteristicP-wave velocities (Schön 1996) and that arelationship with rock modal composition doesexist. Missing this relationship in our studycould be explained by structural effects such asmicro-cracks and foliation. Metamorphic rocksare characterized by pronounced foliation,which is known to influence the P-wave aniso-tropy considerably. In the KTB borehole, up to157o P-wave anisotropy was measured withinstrongly foliated biotite gneisses, even undersimulated in situ conditions (Berckhemer et al.1997). The highest P-wave velocity values aremeasured parallel to foliation, while the lowestvalues are measured perpendicular to the foli-ation. Therefore, rock anisotropy could be anexplanation for the strong scatter of in situ Vpdata for the different rocks.

The resistivity log does not show any signifi-cant comelation with any of the main mineralogi-cal components. This was expected, as the porespace of the selected crystalline rocks is verylow and rock-forming minerals usually act aselectrical isolators.

EJfects of rock chemistry on log respctnses

The highest conelation coefficient (r - 0.87)between log data and rock chemical data is calcu-lated between the potassium log and the K2Ocontent of the rocks (Table 8 & Fig. 6a). Smalldifferences between the two types of data-setare caused by the principal differences in inte-gration volume and sampling method betweenthe continuous log data and the discrete core ot'cutting analyses. Taking into account the strongdifference in the amount of data between coreand log data, the high conelation between thetwo data sources is remarkable. Nevertheless,high potassium values in the log data may bebiased, due to the calculation of potassiumcontent from the total energy spectra ofcounted gamma-rays or by accumulation ofpotassium in the borehole mud. The latter wasobserved in mica-rich gneisses in the KTB-VBhole (Pechni g et al. 1997). Mica crystalsbecame detached from the borehole wall andwere kept in suspension in the mud. Thiscaused gamma-ray 1og responses to increase byrp to l0%.


294 R. PECHNIG ETAL

A slightly smaller correlation coefficient(r:0.81) was calculated for the gamma-rayand the K2O content of the rocks. While therelation potassium log versus K2O from coresrepresents variations in only one chemicalelement, the relation between gamma ray andversus K2O from cores is also influenced by

thorium and uranium. These two elements con-tribute to the total gamma-ray responses andthus change the gamma-ray/potassium ratio.

Significant correlations exist between thedensity log and the TiO2 and MgO content.Titanium and magnesium have higher atomicweights than the other rock-forming elements

2.8

Eoo)> )7'6c0)E

E-raa(g

oo3

34K2O (o/")

46Mgo (%)

+ biotite-graniteX monzonitex syenite< apliteO gabbroic gneissa dioritrc gneissv granodioritic gneiss

* granitic gneiss* K-feldspar gneiss* quartz plagioclase gneisso plagioclase-biotite gneisso (cordierite)-sillimanite-biotite gneissO hornblende gneiss

60 70

sio2 (%)

B BöttsteinC CajonG GPKIK KTB-VBL LeuggernlV MoodusS Schafisheim

14

12

2.9

tr ^^6 2.0

o)

!a ta0)!

10

:o^oboo.co

0)to

-2

Fig. 6. Cross-plots of log responses of potassium, density, and neutron porosity versus the chemical composition of therocks. Displayed are the average data liom Tables 3 & 4. The corresponding Pearson correlation coefficients are givenin Table 8.

a)X

öX

L+

f.*B.Gs

++S10 Y M

; cM

M.c

r=0.874

Lo

c

I

o;MC^ 6''YoK

r=0.762

M.N4 K. .cMc

Nr1 o,

cL. I* oL +K

+ct\4* s -A<+L s+1 cB- vc*

c)r=0.709

Ke

siX

.M...K r ! lvltvt 4

L"oL ( Mo*S*KC+

,4 SG

311 c_

n-

L.

K' oL

KL0 t

,B', +Ta

GS.IV*"c-*c f,,l**"t M,

t**

fiv

r=-0.582

sKo*Xs

K

oc


silicon, aluminum, potassium and calcium, andthey are preferentially incorporated into maficminerals. High TiO2 contents and high densityvalues are therefore measured for the mostmafic rocks: the Schafisheim monzonites/sye-nites, followed by the hornblende gneisses ofLeuggern and KTB (Fig. 6b). The correlationbetween density and TiO2 content is strongerthan the one between density and MgO content.The latter is mainly caused by the very highMgO values of the monzonites and syenites,thus plotting separately from the general trend(Fig. 6c). A weak negative correlation(r: -0.58) occurs between density and SiO2content. A possible reason is a coupling of a

decrease in mafic constituents with an increasein both quartz and feldspar.

The neutron porosity shows conelations com-parable to those calculated between the densityand the chemical elements. The highest coeffi-cient is calculated for the relation betweenneutron porosity and the TiO2 content. Thismight be due to different effects of:

(1) OH-groups incorporated in some maficTiO2-rich constituents;

(2) higher atomic weights of elements in maficrocks; and

(3) the occurence of trace elements.

The negative trend between the neutron porosityand the silicon content is also visible, althougha strong scatter occurs between the data points.

The P-wave velocity and electrical resistivitylogs show no significant cor:relation with any ofthe chemical elements. This is a further indi-cation that these logging parameters are notsignificantly influenced by the chemical compo-sition of the selected rocks.

Log correlation trends as result ofmine ral o g i c al v a ri ab i I i ry*

Although significant correlations between logreadings and rock composition exist, correlationsbetween the log parameters are not detectable on

EOo

=ac0)E 2.6

potassium (wt.%)

paragneisses

-i. hornblende gneiss (M)

# hornblende qneiss (K) |

G hornblende gneiss (L) |

sillimanite/biotite gneiss 1L1 | I

' (sillimanite)-biotitegneisslK)l

biotite gneiss (M) I

8

orthogneissesa gabbroic gneiss (C)

Ö dioritic Aneiss (C)

4potassium (wt.%)

igneous rocksX biotite granite (S)

El biotite granite (L)

* biotite granite (B)

leucogranite (GP)

aplite (S)

]"

]'@ granodioritic gneiss (C) -'l

O qranitic gneiss (C) |

+ granitic gneiss (M) ltttK-feldspar gneiss (M) |

quartz-feldspar qneiss 1V1 JX hornblende-biotite monzonite (S)

] VM hornblende-biotite syenite (S)

Fig, 7. Cross-plots of (a) potassium versus neutron porosity and (b) potassium log versus density, both showing the

selected rock types. The potassium versus neutron porosity cross-plot makes it possible to span the classilication fieldsto distinguish between the main rock groups: granites/aplites, monzonites/syenites, paragneisses, and acid tointennediate orthogneisses. Similar relationships between are also visible lbr the potassium-density cross-plot, but the

separation ofthe groups is not as distinct as for the neutron potassium p1ot. L, Leuggern; B, Böttstein; S, Schafisheim;GP, GPK1; C, Cajon Pass; M, Moodus; K, KTB.

296 R. PECHNIG EZAL,

the f,rst viewing. When displayed as a cross-plot,the logging data show a large scatter (Fig. 7a &b). However, the cross-plot neutron porosityversus the potassium log allows the separationof the main rock types, since comparable rocksfrom different boreholes plot into the same plotareas (Fig. 7a). This is especially the case forthe granites and aplites. More than 90Vo ofthese data plot between 0 and 7 Vo neutron poros-ity and 4 and 7 wtTo potassium. Also, the plotarea of the orthogneisses is quite limited. Thelargest data scatter is produced by the para-gneisses, especially for the neutron-porosityvalues, which range between 27o and >157o.The monzonites and syenites also exhibit a

large data scatter. These rocks are clearly separ-ated from the granite field and the paragneissfield. Signiflcant overlaps between the mainrock types exist between the paragneiss and theorthogneiss fields in the neutron porosity versuspotassium plot. Rocks plotting in this areacannot be distinguished. Data clustering is alsovisible in the cross-plot density versus potass-ium, and the separation of the main rock typesis not as sharp as in Figure 7a. This is especiallythe case for the orthogneisses, biotite para-gneisses and hornblende gneisses, whichshow strong overlaps in the density field2.6-2.8 g cm-'. Cranites and aplites are betterseparated by their high potassium values. Thisis also the case for the monzonites/syenites.

By separating the rocks into paragneisses andplutonic rocks, significant trends between logparameters become visible. The most prominenttrends exist for the paragneisses, where two cor-relation trends occur. Within the large group ofbiotite gneisses, the gamma ray is positively cor-related with the neutron porosity (Fig. 7) and thedensity log (Fig. 8). This correlation is clearlyobserved for all the biotite gneisses of the differ-ent boreholes, although offsets exist between theLeuggern and the Moodus/KTB data. The trendscan be related to the rock mineralogical variabil-ity and will be explained by examples from theKTB biotite paragneiss. The mineralogical andchemical composition of this gneiss type variesconsiderably. with quar{z concentralionsranging between 257o and 55Va; biotite contentsbetween 5Va and 407a: and muscovite contentsbetween 37o and 38Vo (Table 2). The varyingquartz-mica ratios are reflected in the chemicalcomposition: i.e. SiO2 values between 45Vo andl57o and K2O contents between 2.2Vo and3.67a(Table 3). Petrographic studies revealed twoend-members of the KTB biotite gneisses:a quartz-rich type with high SiO2 values, and a

mica-rich rich variety with increased potassiumcontents (Müller et al. 1989). By comparing

oEOC')

=aCoE

3.'1

3.0

2.9

2.8

2.7

2.6

2.550 100 150 200

gamma ray (APl)

50 100 150 200gamma ray (APl)

Fig. 8. Diagram of the mineralogical end-memberscontrolling the relationship between the density andgamma array in paragneisses. For signatures, see

Figure 7.

mineralogical and chemical data with log-ging data it was demonstrated (Haverkamp &Wohlenberg 1991; Pechnig et al. 1997) thatthe mica-rich paragneisses have much highergamma-ray, potassium, density and neutron-porosity values than the quartz-rich types. Thiscan be explained by mineral physics and chem-istry. Micas are the most important potassium-bearing minerals in the biotite paragneisses, andthus they contribute signiflcantly to the measuredtotal gamma activity. Micas have higher densitiesthan quaflz (muscovite: 2.83 g cm '. biotite:3.0 I g cm '. quartz: 2.65 g cm 't. and they

3.1

3.0

E 2.9()O)I z.a=oc ^-Q z.l!

2.6

2.5


have incorporated OH-groups in their lattice.After Serra (1986) a rock consisting of l00%o ofmica would produce 20-25Vo apparent porosityin the neutronJog readings, while pure quartz

would produce a neutron-log response of -2Vo.The relationship between the quattzfmica

ratio and the logging data exists also for theLeuggern and Moodus gneisses. In all boreholes,the end-members of the observed trends are, as

marked by the arrows in Figure 8, quartz-richgneisses with low density and low gamma-rayvalues on the one hand, and mica-rich gneisses

with high density and high gamma-ray valueson the other hand. In particular, the Leuggernbiotite gneisses show a pronounced modal varia-bility. Quartz contents vary between L5Vo and50Vo, and total mica plus sillimanite content canbe more than 60Vo (Table 2). This produces totalgamma-ray values varying from 80 to more than200 API (Tabte 5a), and log density rangingbetween 2.49 and,2.81 g cm-3.

The hornblende gneisses of the Moodus andthe KTB borehole show divergent trends. In con-trast to the biotite gneisses, density and gammaray are negatively cotrelated (Fig. 8). The maincomponents of hornblende gneisses arc qüartz,feldspar, biotite and amphibole, in variableamounts (Table 2). Gneisses with low amphibolecontents have high quartz contents of, and viceversa. Enrichment in quartz corresponds withthe change to a more acid geochemical compo-sition of the rocks, which is also accompaniedby a potassium increase. Since amphiboleshave a significant higher density than quartz,the compositional changes produce a trend withthe following end-members: amphibole-richhornblende gneisses with low gamma-rayvalues and high density values on the one hand,and quartz- and plagioclase-rich hornblendegneisses with low densities on the other (Fig. 8).

Plotting data from the orthogneisses and pluto-nic rocks in the cross-plot density versus potassiumdoes not reveal signiflcant trends (Fig. 9) as

observed for the paragneisses. A negative corre-lation between density and potassium is visiblefor the orthogneisses, which can be explained bymineralogical variations similar to those of thehomblende gneisses. Mineralogical end-membersof this trend are biotite-amphibole-rich orthog-neisses of gabbroic/dioritic composition with thehighest densities, and quartz-plagioclase-richorthogneisses of granodioritic/granitic compo-sition with the lowest densities. Gamma activityand potassium values increase significantly in the

field of granites and aplites. Here, only a weaktrend is visible. This trend is caused by the enrich-ment of K-feldspar in these rocks. Due to the sub-

stitution of K-feldspar and plagioclase in granites,

3.0

EoDz.s.=aco)c

2.6

48potassium (wt.%)

+ amphibole+ K-feldspar

48potassium (wt.%)

Fig. 9. Diagram of the mineralogical end-members

controlling the relation between the density and thepotassium log in orthogneisses and plutonic rocks. Forsignatures, see Figure 7.

potassium values vary over a considerable range.Since the density of K-feldspar, plagioclase andquartz (the main components of the granites and

aplites, Table 2), is more or less the same, rockdensity is not affected by the mineralogicalvariability.

The monzonites and syenites are separatedfrom the granites and orthogneisses, and showa data scatter without any trends visible. Theserocks exhibit high variability in the potassiumand density values, which is related to the chan-ging amounts of their components. High densityand high potassium values in syenites/

3.0

E€ 2.8(')

=QCo)E

2.6

298 R. PECHNIG ETAZ

monzonites are related to the enrichment of bothminerals, homblende and potassium feldspar.

In summary, for the orthogneisses and pluto-nic rocks, the relation between the density andpotassium log is controlled by the relativeamount of the mineral assemblages: amphiboleplus biotite, quartz plus plagioclase, and potass-ium feldspar.

Conclusions

Mineralogical, geochemical and in situ petrophy-sical data-sets were compiled for this study inorder to explain log responses in relation torock composition for acid to intermediategneisses and plutonic rock. Correlation analysisbetween the different data types reveals signifi-cant correlations between some of the standardlogs and rock mineralogy. The most prominentare the gamma-ray and potassium logs as indi-cators for the potassium feldspar and micacontent. In addition, the density and neutron-porosity log are indicative of the rock compo-sition. They are strongly controlled by therelative amount of amphibole plus mica versusquafiz plus feldspar. P-wave velocity and electri-cal resistivity are essentially independent of rockcomposition.

These results show that in the investigatedrock types the standard logging parameters areof a different importance for lithology prediction.Density-log and neutron-porosity devices, whichare used in the sedimenlary environmenl as por-osity predictors, are indicative of mineralogicalchanges in the clystalline environment. Themost powerful discrimination between differentrock types is given by a combination of thegamma-ray and the potassium logs on the onehand versus the neutron and density logs on theother. The combination of neutron versus potass-ium logs is suitable for discriminating betweenthe main rock types.

Strong correlations between logging par-ameters are also observed within the majorrock groups. Integration of petrophysical logdata with information on rock compositionreveals three major mineralogical backgroundparameters which are controlling the logresponses in gneisses and acid to intermediateplutonic rocks. In paragneisses, log correlationsfollow the ratios of the amphibole, quartz andmica. In the orthogneisses and granites/aplitesthis is biotite plus amphibole versus K-feldsparplus quartz, following the magmatic differen-tiation from mafic to acid rocks. Syenitesand monzonites strongly separate from thistrend, since potassium feldspar occurs togetherwith amphibole and biotite.

This study used logging data and core data from variousboreholes, collected and composed from differentsources. The German Continental Drilling Project (KTB)was funded by the German Ministry of Research andTechnology (BMFI). KTB data were made available bythe GeoForschungszenftum (GFZ) Potsdam, Germany.Logging data from the Leuggem, Böttstein andSchafisheim boreholes were kindly made available byNAGRA (National Co-operative for the Disposal ofRadioactive Waste). Data from the borehole GPK1Soultz-sous-Foröts were made available by the Bureaude recherches g6ologiques et miniöres, and the digitallogging data-sets of the Moodus and Cajon Pass boreholewere kindly provided by the Borehole Research Group ofthe Lamont-Doheny Earth Observatory, NY, USA. Partsof this study were funded by the German Science Foun-dation (DFG, Wo-159/11;Wo-159/14). We would alsolike to thank Prof. J. Wohlenberg for initiating thesestudies and for making them possible.

ReferencesAorns, J. A. S., OsvoNo, J. K. & Rocsns, J. J. W.

1959. The geochemistry of thorium and uranium.In: AunnNs, L. H., PRESS, F., ReNrer,r,r, K. &RuNcoRN, S. K. (eds) Physics and Chemisny ofthe Earth,3,298-348.

Ar,teens, C. P. 1989. Core and Cuttings Descrilttionsof the 1987 Moodus 4764 Foot Deep Well, Connec-ticut. Connecticut Geological and Natural HistorySulvey, Open File Report 89-1, 207 pp.

BERCKHEMER, H., ReusN, A. et aI. 1997. Petrophysi-cal properlies of the 9-km-deep crustal section atKTB. Journal of Geophltsical Research, 102,18 337- 18 361.

CssvnEr,rorur. P. & GEurEn. A. 1989. Etude Pötolo-gique des Ecltantillons cle Granite de Soultz-sous-For€ts, Forage GPK1. Note interne 89/IRG, 20.

DeNrEr-s, J. J., Scorr, J. H., & Or-HoEpr, G. R. 1993.Interpretation of core and well log physical prop-efty data from drill hole UPH-3, StephensonCounty, Illinois. Journal of Geophysical Research,88.'7346-7354.

DEsennnoss, R. 1982. Encyclopedia of Well Logging,Edition Technips, Paris, 548 pp.

Et-rIs, D. V. 1987. Well Logging for Earth Scientists.Elsevier, Amsterdam, 532 pp.

Ervnrnnr'rlrlN, R. & LeurlnruNc, J. 1997. TheGerman Continental Deep Drilling Program KTB:overview and major rcsults. Journal of Geophysi-cal Research,102, 18 179 I8 201.

GrNrnn, A. 1989. G4othermie Roches ChaudesSöches: le granite de Soultz-sous-F6rets (Bas-Rhine, F ranc e ). Fracturation naturelle, ahö rationhltdrothermale et interaction eau-roche. Thösede doctorat de 1'Universit6 d'Orl6ans, France, 201pp.

GENrsn, A., Bs.ruponr, A., MtuNrpn, A. &.CHEvnpuoNr, P. 1 989. Etude des alt6rations hydlo-thermales du granite de Soultz-sous-Fdrets-ForageGPKI. Rapport BRGM B9, SGN 295 EEE/IRG.

Henvey, P. K., Lovgr-r-, M. A., BnpwEn, T. S.,Locxp, J. & MeNslEy, E. 1996. Measurementsof thermal neutron absorption cross section in

selected geothermal reference material. Geostan-

dard Newsletter, 20, 19 -85.HAVERKAMe, S. & WoulENnERG, J. 1991. EFA-Log-

Rekonstruktion kristalliner Lithologie anhand vonbohrlochgeophysikalischen Messungen für dieBohrungen URACH 3 und KTB-Oberpfalz VB.KTB Report, 9l-4, 213 pp.

Hsensr, J. R., NElsoN, P. H. & PrtlLst, F. L. 2000.

WelI Logging for Physical Properties. A Handbook

for Geophysicists, Geologists and Engineers, 2ndedn, Wiley, Chichester, UK.

HETNSCHTLD. H. J., Hour.NN, K.. Srnou, A. &TeppEn, M. 1988. Tiefbohrung KTB OberpfalzVB, Ergebnisse der geowissenschaftlichenBohrungsbearbeitung im KTB Feldlabor,Teufenbereich 480 bis 992 m: C. Geochemie. In:Er,tr'rEnr,teNN, R., DrErntcu, H. G., HBINrscs,M. & WöHnr, T. (eds) KTB-Report, 88-2,Cl C107, Hannover, Germany.

HrRscuueNN, G, Duvsrsn, J., HARMs, U., KorNv,4.,Lepp, M., oE WeLl, H., Zut-r.up, G. 1997. TheKTB superdeep borehole: petrography and stlrlc-ture of a 9-km-deep crustal section. GeologischeRundschau,86,3-14.

Krvs, W. S. 197 9. Borehole geophysics in igneous and

metamorphic rocks. Transactions, SPWI'A, 20thAnnual Logging Symposium, Tulsa, Oklahoma,paper OO.

LACHENBRUCH. A. H. & S,rss, J. H. 1988. The sffess

heat-flow paradox and thermal results from CajonP ass. Ge o ph,r sic al Re s e arch Lett e rs, 15, 98 1 - 984.

LANDoLT-BöRNsrplN 1982. Zahlenwerte und Funk-tionen aus Naturwissenschaften und Technik.Neue Serie Vol. I: Phltsical Properties of Rocks,Subvol. b. - 604 S., Springer, Beriin, Heidelbergand New York.

MATTER, A., PBrrns, T., BLAst, H., ScnnNrpn, F. &Wprss, H. P. 1988. Sondierbohrung Schafisheim -Geologie; Text- und Beilagenband. Nagra Tech-nische Berichte NTB 86-03, Nagra, Wettingen,Switzerland.

NAUMHoFF, P. 1988. Lithology and structure identifiedin a 1.5 km borehole near Moodus, Connecticut.EOS,69, p. 49l

Nrr-soN, P. H. & JoHNsroN, D. 1994. Geophysical andgeochemical logs from a copper oxide deposit,

Santa Cruz Proiect, Casa Grande, Aizona. Geo-physics, 59 (12), 1827- 1838.

O'BnrsN, P. J., DuvsrEn, J., Gn,quEnr, B.,ScHnEvEn, W., SröcxuEnr, B. & WeeEn, K.1997. Crustal evolution of the KTB drill site: fromoldest relics to the late Hercynian granites. JoumolGeophl,sical Research, I02, 18 203-I8 220.

PECHNTc, R., Hevsnr,qup, S., Wost-ENBERG, J.,

Zrr,rMpnNrlNN. G. & Buncxir,qnor, H. 7997.lnle-grated log interpretation in the German ContinentalDeep Drilling Program: lithology, porosity, and

fracture zones. Journal Geophysical Research,102. 18 363-18 390.

Perens, T. J., MerrEa, A., BlAsl-t, H.-R. &G,qutscnt, A. 1986. Sondierbohr-ung Böttstein -Geologie; Text- und Beilagenband, Nagra Techn.Ber. NTB B5-02, Nagra, Wettingen, Switzerland.

299

PprEns, T. J., Merrst, A., Br-Ast-t, H.-R.,IsrNscur,tto, Cu., KI-Enoru, P., MevEn, Cu. &Mevrn, J. 1989a. Sondierbohrung Leuggern -Geologie; Text- und Beilagenband, Nagra Techn.

Ber. NTB 86-05, Nagra, Wettingen, Switzerland.Persns, T. J., Merrsn, Mp,vr,n, J., IssNscutrltn, Cs.

& Zrecr-En, H.-J. 1989ö. Sondierbohrung Kaisten

- Geologie; Text- und Beila genband, Nagra Techn.B er. NTB 86-01, Nagra, Wettingen, Switzerland.

Pnnrsou, E. L., ANoEnsoN, R. N., Dovp, R. E., Lvle,M., Srr-vEn, L. T., Jeurs, E. J. & Csepppll, B. W.1992. Geochemical logging in the Cajon Pass drillhole and a new, oxide, igneous rock classificationscheme. Journal Geophysical Research, 97,

5167-5180.Rrosn, M. 1996. The Geological Interpretation of WeIl

logs. Whittles Publishing, Caithness, Scotland,UK, 280 pp.

Rocpns, J. J. W. & Aoevs, J. A. S. 1969a. Thorium.In: Wpopponl, K. H. (eds) llandbook o;f Geochem-istry, 2 (4), 908-90O, Springer, Berlin.

RocERs. J. J. W. & Aoerrls, J. A. S. 1969&. Uranium.In: WEoEponL, K. H. (eds). Handbook of Geo-chemistry, 2 (.4), 928 -92O, Springer, Berlin.

S,crrel, G. 1986. FACIOLOG - Anwendung eines

Programmsystems zur Charakterisierung der Kris-tallin-Lithologie. Nagra Infonniert, l, 18-24.

ScHr-ur,reencrp. 1994. Neutron porosity loggingrevisited. Oilfield Review, October, 4 8.

ScHöN, J. H. 1996. Physical properties of rocks -fundamentals and principles of petrophysics.HpLetc, K. & TnErrEr-, S. (eds) Harulbook o;f Geo-physical Exploration, Section I, Seismic Explora'tion,Yol. 18. Pergamon, Oxford, 583 pp.

Sltne, O. 1984. Fundamentals of Well-Log Interpret'ation. YoL 1, Elsevier, Amsterdam.

Senne, O. 1986. Fundamenta.ls of Well-Log Intetpret-ation. Y o1. 2. Elsevier, Amsterdam.

STLVER, L. T., Jeues, E. W. & Cueppst-l, B. W. 1988.Petrological and geochemical investigations at theCajon Pass Deep Drillhole. Geophysical ResearchLetters,15,961 964.

SrnecrErssN, A. 1967 . Classification and nomencla-ture of igneous rocks. Neues Jahrbuch Minerakt-gische Abhandluagen, 107, 144 214.

Srnon, A. & TeppEn, M. 1988. Tietbohlung KTBOberpfalz VB, Ergebnisse der geowissenschaftli-chen Bohrungsbearbeitung im KTB Feldlabor,Teufenbereich 0-480 m: C. Geochemie. 1n:

El,nrnnlteuN, R., DIsrntcs, H. G., HrrNIsctl,M. & Wösnr, T. (eds) KTB-Report,88-1, Cl-C73. Hannover.

Teppnn, M., HprNscuIr-o. H. J., Srnou, 4.,WrrrnNescHsn. M. &. Ztr,tr,tEn. M. 1989.Tietbohrung KTB Oberpfalz VB, Ergebnisseder geowissenschaftiichen Bohrungsbearbeitungim KTB Feldlabor, Teufenbereich 2500 3009,7m: C. Geochemie. In: Er'n'tnnueNn, R.,

Drprr.rcu, H. G., HsrNrscu, M. & WÖHnl-,T. (eds) KTB-Report,89-4, C1-C50, Hannover.

Tuunv, M., Geurscut, A. et al. 1994. Geologyand hydrology of the crystalline basement ofnofthem Switzerland. Geologische Berichte Nr.


300 R. PECHNIG ETA'.

L8, Bundesamt für Umwelt, Wald und Landschaft,Bern. Switzerland.

TnerNeeu, H. A., GsNrEn, J. P., Ceurnu, F.,FesR.roL, H. & CnevnnrrloNr, P. 1992. Petrogra-phy of the granite massif from drill cutting analysisand well log interpretation in the geothemal HDRborehole GPKI (Soultz, Alsace, France). In:BnBsBp, J.C. (eds). Geothermal Energy inEurope,1-29, Gordon and Breach Science Pub-lishers, Switzerland.

VrNcENr, M. W. & Esr-tc, P. L. 1988. Laumontitemineralization in rocks exposed of San AndreasFault at Cajon Pass, Southem Califomia. Geophysi-cal Research Letters, 15,977 -980.

WILsoN, M. 1989. Igneous Petrogenesis. UntvinHyman, London.

WrNrscs. R. P. & Ar-sINrropr, J. N. 1987. U-Pb Iso-topic and geological evidence for Late Paleozoic

anatexis, deformation, and accretion of the LateProterozoic Avalon Terrane. American Joumal ofScience, 287, 107 -126.

WTTTENBECHER, M., HErNscuILD, H. J., Srnou, A. &TAIFER, M. 1989. Tiefbohrung KTB OberpfalzVB, Ergebnisse der geowissenschaftlichen Boh-rungsbearbeitung im KTB Feldlabor, Teufenber-eich 3009, 7-3500 m: C. Geochemie. 1n;

EulrnnrraeNN, R., Dlernlcu, H. G., HEtNtscs,M. & Wöunr-, T. (eds) KTB'Report, 89-5,Cl - C62, Hannover, Getmany.

WoHLENBERG, J.1982. Dichte der Minerale und Ges-teine. In'. ANceNsusrsn, G. (eds) Inndolt Böm-stein - Physikalische Eigenschaften der Gesteine-Subvolume A, 66-119, Springer, Berlin, Heidel-berg and New York.

petrophysics of metamorphic and igneous rocks

Documents