project legs 56 and 57, japan trench: organic petrography and

62. ORGANIC GEOCHEMISTRY OF SEDIMENTS CORED DURING DEEP SEA DRILLINGPROJECT LEGS 56 AND 57, JAPAN TRENCH: ORGANIC PETROGRAPHY AND

EXTRACTABLE HYDROCARBONS

J. Rullkötter, C. Cornford,1 P. Flekken, and D. H. Welte, Institute for Petroleum and Organic Geochemistry,Kernforschungsanlage Jülich, Jülich, West Germany

ABSTRACTThe quantity, type, and maturity of the organic matter of Quater-

nary and Tertiary sediments from the Japan Trench (DSDP Leg 56,Sites 434 and 436; and Leg 57, Sites 438, 439 and 440) were deter-mined. The hydrocarbons in lipid extracts were analyzed by capil-lary-column gas chromatography and combined gas chromatog-raphy/mass spectrometry. Kerogen concentrates were investigatedby microscopy, and vitrinite-reflectance values were determined.

Measured organic-carbon values were in the range of 0.13 to 1.00per cent. Extract yields, however, were extremely low. Normalized toorganic carbon, total extracts ranged from 4.1 to 15.7 mg/g Corg. Gaschromatography of non-aromatic hydrocarbons showed that all sedi-ments, except one Oligocene sample, contained very immature,mainly terrigenous organic material. This was indicated by π-alkanemaxima at C29 and C3) and high odd-carbon-number predominances.Unsaturated steroid hydrocarbons were found to be major cycliccompounds in lower- and middle-Miocene samples from the upperinner trench slope (Sites 438 and 439). Perylene was the dominatingaromatic hydrocarbon in all but the Oligocene sample.

Microscopy showed kerogens rich in terrigenous organic particles,with a major portion of recycled vitrinite. Nevertheless, almost allthe liptinite particles appeared to be primary. This is a paradox, asthe bulk of the samples were composed of hemipelagic mineral mat-ter with a major siliceous biogenic (planktonic) component. A trendof reduced size and increased roundness can be seen for the vitri-nite/inertinite particles from west to east (from upper inner trenchslope to outer trench slope). All sediments but one are relatively im-mature, with mean huminite-reflectance values (i?o)in the range of0.30 to 0.45 per cent. The oldest and deepest sediment investigated,an Oligocene sandstone from Site 439, yielded a mean vitrinite-reflectance value of 0.74 per cent and a mature n-alkane distribution.This sample may indicate a geothermal event in late Oligocene time.It failed to affect the overlying lower Miocene and may have beencaused by an intrusion. Boulders of acidic igneous rocks in theOligocene can be interpreted as witnesses of nearby volcanic activityaccompanied by intrusions.

INTRODUCTION

A transect of the Japan Trench east of Honshu Islandwas drilled on DSDP Legs 56 and 57. Drill-site locations(Figure 1) embraced the top of the trench inner wall(Sites 438 and 439, water depth -1600 m), the upperslope (Site 435, water depth 3400 m), a mid-slope terrace(Site 440, water depth 4500 m), the zone of accretion atthe base of the slope (Site 434, water depth 5985 m), andthe outer-trench slope (Site 436, water depth 5240 m). Amajor objective of this transect was the investigation of

'Present address: The British National Oil Corporation, Glasgow,Scotland.

the sedimentary organic matter in the active-marginzone of the Japan Trench. It was expected that analysisof drill cores would reveal any differences between theorganic matter in the deep-sea sediments deposited onthe Pacific plate, which is being subducted under thecontinental margin of Japan, and the organic matter ofthe continental-slope sediments. Another subject ofspecial interest was the so-called "zone of accretion" atthe foot of the slope, which was assumed to originatefrom sediments scraped off the oceanic crust duringsubduction. The Japan Trench had been considered anexception among the trenches of the circum-Pacific beltby being nearly devoid of terrigenous sediments (Scholland Mariow, 1974). Investigation of the organic matter

1291

J. RULLKOTTER, C. CORNFORD, P. FLEKKEN, D. H. WELTE

HONSHU IS

PACIFIC OCEAN

tions (approximate)

LEG 56

LEG 57

CONTOURS IN METERS

DEEP SEA TERRACE 435 440 441 434

INNER TRENCHMID. SLOPETERRACE

OUTER TRENCH .SLOPE

SCHEMATIC CROSS SECTION SHOWING DRILL SITE

LOCATIONS FOR DSDP - LEGS 56 + 57

Figure 1. Location of sites drilled on the Japan Trench transect and schematic sec-tion showing drill sites. Section based on Ishiwada and Ogawa (1976) and JapanNational Oil Corporation multi-channel data.

in the Legs 56 and 57 samples should indicate whetherthis statement is valid for the organic component of thesediments.

SAMPLESTwo core samples each from Sites 434 and 436, seven

samples from Site 438, three samples from Site 439, andfive samples from Site 440 were studied. Information ondepth, stratigraphy, and lithology is given in Table 1.

EXPERIMENTAL PROCEDURESSamples were received frozen. They were dried,

ground, and extracted with dichloromethane, using aflow-through extraction cell (Radke et al., 1978). Totalextracts were separated into non-aromatic hydrocar-bons, aromatic hydrocarbons, and heterocompounds bymedium-pressure micro-column liquid chromatogra-phy. Contents of total organic carbon were determined

by combustion of HCl-treated rock samples in an oxy-gen flow at 1300°C in a Leco carbon analyzer.

Capillary-column gas chromatography of hydrocar-bon fractions was performed on a Siemens L 402 gaschromatograph equipped with a 25 m × 0.3 mm i.d.glass capillary column coated with SE 54. Hydrogen wasused as carrier gas, and the oven temperature was pro-grammed as follows: 80 °C (2 min), 80 to 250 °C (3 °C/min), 250 °C (10 min).

Mass spectrometric analyses were carried out using aVarian MAT 112 S mass spectrometer linked directly toa Varian 3700 gas chromatograph. Samples were in-jected splitless onto a 25 m × 0.3 mm i.d. SE 54 glasscapillary column. The oven was programmed from100°C to 250 °C at 4°C/min (carrier gas: helium). Themass spectrometer was operated at 70 eV, and thesource temperature was 220° C. The magnet was scannedcontinuously at a rate of 1 s/dec. All data were storedand processed with a Kratos DS 50 S data system.

1292

ORGANIC PETROGRAPHY AND EXTRACTABLE HYDROCARBONS

TABLE 1Description of Samples (depth, age, lithology, and present structural situation), Organic/Carbon Values, and External Data

SiSample

(interval in cm)

434-1-3, 115-125434B-IS-I, 107-117

436-7-4, 115-125436-36-2, 115-125

438A^M, 110-120438-10-4, 115-125438A-I8-2, 1 15-125438A-24-4, 110-120438A-46-3, 110-120

438A-70-4, 110-120438A-84-1, 100-110

439-9-4, 100-110439-21-2, 115-125439-31-5,0-5

440B-3-5, 110-120440B-17-5, 125-135440B-39-3, 115-125440B-51-3, 125-135440B-68-2, 1 10-120

"Calculated.bValues questionable bee

lb-BottomDepth

(m)

4.2419.1

61.2333.7

47.784.7

223.2283.2492.7

722.2850.5

893.2994.7

-1105

165.6298.7504.7618.7778.6

Age

PleistoceneLower Pliocene

PleistoceneMiddle/Late Miocene

Pliocene/PleistoceneUpper PlioceneLower PlioceneLower PlioceneUpper Miocene

Middle MioceneMiddle Miocene

Lower MioceneLower MioceneUpper Oligocene (?)

PleistocenePleistoceneLower PlioceneLower PlioceneUpper Miocene

ause of extremely low weight of fr

Lithology

Diatomaceous oozeVitric-diatomaceous mudstone

Diatomaceous muddy oozeRadiolarian-diatomaceous

mudstone

Diatomaceous silty clayDiatomaceous silty clayDiatomaceous clayDiatomaceous clayMottled diatomaceous

claystoneDiatomaceous claystoneDiatomaceous claystone

Tuffaceous siltstoneClayey siltstoneSandstone

Diatomaceous clayClaystoneDiatomaceous claystoneClaystoneSilty claystone

actions.

Present StructuralPosition

Zone of accretion (lowerinner trench slope)

Outer trench slope

Upper innertrench slope

Upper innertrench slope

Mid-slope terrace

( org(%)

0.870.70

0.260.13

0.4 10.600.900.820.61

0.5 10.5 5

0.530.330.33

0.980.590.600.371.00

ppm

13690

489

415«

1246064

4873

592046

1 1243754541

Mg/g

Q>rg

15.712.8

18.56.8

10.19.7

13.87.3

10.4

9.513.3

11.18.5

14.0

11.47.2

12.512.14.1

Extrac

Non-AromaHC(%)

83

25,50 b

2726

82922

1319

319

49

1218111516

t Yields

tic AromaticHC(%)

1012

2125 b

2720111419

1823

141429

1618233 019

Hetero-Compoundsa

(%)

8285

5 4 i2 5 b

4654815759

6958

557722

7264665565

Details of the experimental procedures for micro-scopic investigations have been described in a previouspaper (Cornford et al., 1979). In brief, the extracted sed-iments were treated with HC1, and a kerogen concen-trate was floated off, using a solution of 1.9 g/cm3 den-sity. The kerogen concentrate was trapped on a filterand dried, and a smear slide was prepared for observa-tion in transmitted light. Where sufficient material wasobtained, the kerogen was embedded in "Araldite" andpolished for observation in reflected light. Reflectancewas measured with a Zeiss MPM 01 photometer inter-faced with a Digital Equipment PDP 11/10 computer,using non-polarized light and oil immersion. The humi-nite/vitrinite reflectance, where quoted, is the arith-metic mean of the values in the lowest histogram peak.Each reflectance value is fed into the computer with amaceral identifier (L = Liptinite, V = vitrinite/humi-nite, I = inertinite, X = unknown). Histograms arethen generated both for the reflectance values and forthe maceral distribution, using 0.05 per cent intervals.

RESULTS

Organic-Carbon Values and Extractable Organic Matter

The organic carbon content (Table 1) was found tovary in the range of 0.13 to 1.00 per cent: the lowestvalue comes from the oxidized pelagic Miocene of Sec-tion 436-36-2, while the majority of values are above 0.5per cent Corg. In contrast to the fairly high values oftotal organic carbon of the sediments, the yields of ex-tractable organic matter are low (Table 1). Normalizedto organic carbon, extract yields between 4.1 and 18.5mg/g Corg were obtained. Compared, for example, withthe results reported for Leg 47 A samples from the con-tinental rise of the eastern North Atlantic (Cornford etal., in press), the Leg 56/57 values are lower by a factorof five to ten.

Liquid chromatographic separation of the total ex-tracts yielded varying amounts of hydrocarbons (Table

1); typical values are between 25 and 45 per cent totalhydrocarbons. No clear depth or age trend can be estab-lished. All values, however, have to be interpreted withgreat care, because of the very low absolute weight ofthe chromatographic fractions.

Non-Aromatic Hydrocarbons

The capillary-column gas chromatograms of the non-aromatic hydrocarbon fractions show similar /r-alkanedistributions for all samples, with the exception of theoldest (Oligocene) section from Site 439 (439-31-5),which will be described separately. From the normalized«-alkane distributions (Figures 2-4), maxima at «-C29orn-C3l are evident. The high odd-carbon-number pre-dominance in the range of «-C25 to «-C31 indicates lowmaturities for all samples. Below «-C23 there is a moreuniform /7-alkane distribution without any pronouncedpredominance (Table 2).

Pristane and phytane are minor components in mostof the samples. Where calculation of a pristane/phytaneratio was possible (Table 2), this ratio turns out to bebelow 1 except for two Pleistocene sections (440B-3-5and 440B-17-5) and an upper-Miocene section (440B-68-2) from Site 440.

The general pattern of non-aromatic hydrocarbondistributions is represented by Section 440B-3-5 inFigure 5 (top). The Ai-alkanes are dominant, and thehigh-molecular-weight cyclic hydrocarbons mainly con-sist of saturated triterpanes, probably of the hopaneseries, although unambiguous identifications were notpossible, because of the low extract yields.

A significantly higher concentration of C27+ cyclichydrocarbons — relative to all other sediments in-vestigated — was observed in three lower- and middle-Miocene sections from Sites 438 and 439 (438A-70-4,438A-84-1, and 439-9-4). Figure 5 (bottom) shows thecapillar-column gas chromatogram of the non-aromatichydrocarbon fraction of Section 438A-84-1. Compoundstentatively identified by interpretation of their massspectra and by comparison with previous results from

1293


15

10 -

5 -

0

15

1 01

5 -

r 15 -

10 -

- 5

15 -

10 -

5 h

0

15

10 -

5 -

- 438/

- 438-10

. i

: 438>Q

4 3 8 A

1 .1

-1

-1

\-

-A

ε

4

- 438A-46

\

•

<•

-A

-3

>

1

1

1

1

4-

1

i i i i i

15

10h

5 -

0

15

10h

5 -

~ 15 -

3 10-

CC

0

15

10 -

5 -

0

15

10 h

1 1

1 1

1 I

1 1

1 1

1 1

1 1

I 1

1

- 4

CO

—

35

SA-70

3A-84

- 439-9-4

I

: 43E

4:

I T

3S

1-21-2

L

-31-5

-4

-1

I

I

I1

.1

l lI

i

i i i i i15 20 25 30 35

Carbon number15 20 25 30 35

Carbon number

Figure 2. Normalized n-alkane distributions for samples from DSDP Sites 438 and 439, upper innertrench slope.

1294


15

10 I-

5 :

0

15

10 -

5 -

15 -

5 10-Q

CC

15

10 -

5 -

10 -

5 h

- 44(

- 44(

" . 1

44(

Z- . ,

44(

~- 44(

)E

)E

_'J-5

-17-E

I I)B-3S

IIDE

)E

\-

-(

31

3£ »I

,l

.1

. 1

1 "Mi l15 20 25 30 35

Carbon number

C

CC

CC

1 5 r

10 -

5 -

• ^ 1 0 -

5 -

: 434-

- 434

I I

B

I-

-1

3

5-1

I i i i i i15 20 25 30

Carbon number35

Figure 3. Normalized n-alkane distributions for samplesfrom DSDP Site 440, mid-slope terrace.

25 30Carbon number

Figure 4. Normalized n-alkane distributions for samplesfrom DSDP Site 434, lower inner trench slope (zoneof accretion), and DSDP Site 436, outer trench slope.

DSDP Leg 47A (Cornford et al., 1979) and Leg 50 (Rull-kötter, unpublished) are listed in Table 3. Besides twoC30 pentacyclic triterpenes, a variety of homologous un-saturated steroid hydrocarbons in the range of C2η to C29were detected. Two double-bond isomers were present ineach case. Sterenes were found to be Δ4- and Δ5-isomers,steradienes could be identified as Δ4>22- and Δ5-22-com-pounds, respectively (Figure 6). Cyclic compounds not

1295


TABLE 2Carbon Preference Indexes (CPI) and Pristane/Phytane

Ratios for Non-aromatic Hydrocarbon Fractions

Sample(interval in cm)

434-1-3, 115-125434B-15-1, 107-117

436-7-4,115-125436-36-2,115-125

438A-4-4, 110-120438-10-4, 115-125438A-18-2, 115-125438A-24-4, 110-120438A46-3, 110-120438A-70-4, 110-120438 A-84-1, 100-110

439-9-4,100-110439-21-2, 115-125439-31-5,0-5440B-3-5, 110-120440B-17-5, 125-135440B-39-3, 115-125440B-51-3, 125-135440B-68-2, 110-120

CPI 19-23

1.111.19

0.980.92

1.031.051.071.011.071.140.98

1.071.201.041.171.281.301.121.11

CPI25-31

3.083.09

2.792.32

3.613.584.274.313.093.062.55

2.513.591.124.154.344.954.072.72

CPI29

4.304.464.003.24

4.735.025.626.264.344.363.61

3.634.631.115.165.726.445.403.65

Pristane/Phytane

0.60a

0.400.77a

-

0.670.25a

0.620.620.690.910.380.99a0.38a

1.9

1.11.40.770.911.3

aVery low pristane and phytane concentrations.

marked in Figure 5 (bottom) are saturated triterpanes ofthe hopane/moretane series (C27 to C31).

Section 439-31-5, probably upper Oligocene, shows acompletely different «-alkane distribution, with a max-imum at n-C22 (Figure 2) and a carbon-preference indexnear unity (Table 2), a distribution very similar to thatof a mature crude oil. The pristane/phytane ratio of 1.9is the highest value observed in the Leg 56/57 series ofsamples. Although the possibility of indigenous matu-ration or of migrated hydrocarbons present in this Oli-gocene sandstone sample will be discussed in the nextsection, contamination cannot be fully excluded.

Aromatic HydrocarbonsPerylene was detected as the most abundant aromatic

hydrocarbon in all immature sediment samples. TheOligocene Section (439-31-5) showed a distribution ofaromatic hydrocarbons typical for higher-maturitysediments.

MicroscopyThe microscopy of all samples revealed kerogens

dominated by terrestrial-higher-plant debris. A briefdescription of each sample is given in Table 4, the lip-tinite content being estimated from transmitted-lightpreparations, while the distinction between huminite/vitrinite and inertinite was made using polished blocks(Stach et al., 1975).

The majority of observed particles were huminite/vitrinite and inertinite. The reflected-light preparationsalso revealed a significant portion of cell-wall debris,consisting of amber-colored, relatively thin humoteli-nitic wall fragments. In Pliocene diatom oozes (43 8 A-18-2 and 438A-24-4) they were obviously derived fromxylem cells with bordered pits, probably indicating coni-

fers as the source. Pollen was common throughout, withthe exception of Sections 439-21-2 and 439-31-5: theformer contained a dominance of larger humic (lignite?)particles, while the latter, a silty sandstone, containedalmost exclusively finer-grained vitrinite and inertinite.Also of significance was a large (60-80 µm), subcircular,thin-walled pollen or algal body which appeared only inthe Plio-Pleistocene, but at all investigated sites.

Reflected-light preparations showed the kerogenfrom the Miocene claystone of 439-21-2 to containmany lignite-type fragments which were multimacerites(particles containing two or three maceral groups). Inthis case, huminite was mixed with sporinite (mega- andmicrospores). Multimacerites had slightly higher reflec-tance levels (0.43-0.49%) than monomacerites (0.41-0.44%), although such a small difference should not betaken as significant. Sclerotinite particles (fungal sporebodies) in bimacerites showed reflectance levels of 0.48and 0.43 per cent. Sclerotinites were also seen in the Pli-ocene diatom mudstone of 434B-15-1, the Pliocene dia-tom ooze of 438A-24-4, the Pliocene diatom clay of438B-3-5, and the Pleistocene diatom ooze of 440B-39-3, with reflectance values of 0.37, 0.49, 0.56, and0.39 per cent, respectively.

Two types of reflectance data were recorded, randomreflectance histograms of all measurable particles (Fig-ures 7-9) and selected measurements on the lowest re-flectance huminite/vitrinite particles found (Figure 10and Table 5). Reflectance histograms of all measurableparticles were generated in two ways (Figures 7a and7b), showing both the maceral distribution and the ac-tual reflectance values. Figure 7, representing a Pliocenediatom mudstone (434B-15-1), is generally typical of allsamples, except the lignite-type materials of 439-21-2. Itcontains multimodal huminite/vitrinite peaks, little lip-tinite, and a significant amount of inertinite. These his-tograms are biased, since often small liptinite or inerti-nite particles (e.g., <5 µm) could not be measured.

Figure 8 contrasts the maceral distributions in diatomoozes from the upper inner trench slope (438-10-4) andthe outer trench slope (436-7-4), the two samples beingof Pliocene and Pleistocene age, respectively. Figure 9shows the maceral distributions in the deepest sampleanalyzed, the Oligocene silty sandstone of 431-39-5, andthe Pliocene diatom claystone of 440B-39-3 from themid-slope terrace. Types of reflectance distribution,where measured, are noted in Table 5.

Table 5 also contains the huminite/vitrinite-reflec-tance data that come from selected particles of thelowest-reflectance huminite/vitrinite. The mean reflec-tance values place limits on the maturity of the samples,but for reasons given in the next section these data arenot very reliable.

SUMMARY AND INTERPRETATIONOn the basis of the low extract yields and the high

odd-carbon-number predominance of the C25 to C33«-alkanes in the extractable lipid hydrocarbon fractions,the organic matter of all sediments drilled during theDSDP Legs 56 and 57 transect of the Japan Trench —

1296

ORGANIC PETROGRAPHY AND EXTRACT ABLE HYDROCARBONS

440B-3-5, 110-120

165.6 m. Pleistocene

438A-84-1, 100-110

850.5 m, M. Miocene

60 40Retention time (min)

20

Figure 5. Capillary-column gas chromatograms of Sample 440B-3-5, 110-120 cmrepresenting the general pattern of non-aromatic hydrocarbon distributions inDSDP Legs 56 and 57 sediments, and of Sample 438A-84-1, 100-110 cm contain-ing large amounts of unsaturated polycyclic hydrocarbons. Numbers indicaten-alkanes, letters mark compounds tentatively identified by mass spectrometry(see Table 3).

except the Oligocene sandstone of 439-31-5 — can bedescribed as immature and predominantly terrigenous.This is true for the sequences of the continental-slopesediments as well as for the sediments deposited on thecrust of the Pacific plate, at least after the middleMiocene. This is also confirmed by the microscopic in-vestigations. Huminite-reflectance values (Table 5;Figure 10) are all below 0.52 per cent with the exceptionof the silty sandstone of 439-31-5 at about 1105 mdepth, which had a mean vitrinite reflectance of 0.74 percent.

However, because the total-reflectance histograms(Figures 7-9; Table 5) show such an abundance ofhigher-reflectance particles, there is no certainty thatany of the measured huminite/vitrinite is primary andaffected only by diagenesis/catagenesis during its lastdepositional cycle. Indeed, with the exception of the

lowest two samples at Site 439, the transmitted-lightpreparations all show highly immature yellow pollen,spores, and cuticles; only in Section 440B-68-2 werereworked (orange and granular) spores apparent.

As was discussed in studies of Cretaceous argil-laceous sediments from DSDP Site 398 (Cornford, 1979)the higher-reflectance huminite/vitrinite has been attri-buted to eroded coals or associated organic-matter-richsediments in the source area. Distal transport throughoxidizing waters could also increase the reflectance of theparticles. This explanation seems more reasonable for thebroad reflectance "humps" (e.g., Figure 9b) seen in anumber of samples (Table 5) than for the multimodal dis-tributions shown in Figures 7, 8a, or 9a. In addition, suchoxidation should produce some brighter-reflecting oxida-tion rims with darker cores in the particles: no such rimswere observed. In favor of a recycled-coal or -kerogen

1297


TABLE 3Unsaturated Polycyclic Hydrocarbons in the

Non-aromatic Hydrocarbon Fraction ofSample 438A-84-1, 100-110 cma

GC Peak

abcde

fghjk

1mno

Compound

Cholest-4,22-diene (?)Cholest-5,22-diene(?)Cholest-4-eneCholest-5-eneErgosta-4,22-diene

Ergosta-5,22-dieneErgost-4-ene (?)Ergost-5-ene (?)Sitosta-4,22-diene (?)Sitosta-5,22-diene(?)

Sitost-4-eneSitost-5-eneC30 pentacyclic triterpeneC30 pentacyclic triterpene

Molecular Ion (M+)

368368370370382

382384384396396

398398410410

aSee Figure 3, bottom, for GC peak identification.

STERENES

Cholestenes

Ergostenes

Sitostenes

STERADIENES

R = H :

CH3 :

CoHr :

Cholestadienes

Ergostadienes

Sitostadienes

Figure 6. Molecular structures of unsaturated steroidhydrocarbons tentatively identified in Section 438A-3-5.

origin is the presence of discrete peaks of higher-rankmatter, multimacerites in Section 439-21-2, and anoma-lously dark and granular spores in Section 440B-68-2.Spores generally lighten in color with oxidation (Stach etal., 1975). Further evidence of recycled matter is the poorcorrelation of the values of the lowest-huminite/vitri-nite-reflectance population with depth (Figure 10). Thissuggests that the particles reached their present maturitylevels before incorporation into the sediments.

The recycled-organic-matter origin requires a massiveand long-term input of particles from the Oligocene tothe Pleistocene, on both sides of the trench (Figures 7aand 7b). This homogeneity is surprising, but coals of ap-propriate maturities (brown coals to medium-volatilebituminous coals) are currently being eroded on theJapanese Islands (Matveef, 1972).

The huminite-reflectance trend, tentatively sketchedon Figure 10, is at variance with the logged down-hole-temperature data, which indicate gradients of about30°C/km at Site 438 and 10°C/km at Site 440. Nosignificant difference is observed in the organic-mattermaturity trends of the two sites. This anomaly can be

explained if most of the reflectance measurements ofFigure 10 were made on recycled particles, the trendwith depth then being largely fortuitous. This seems alsoto be indicated by consideration of the time/tempera-ture relationship to huminite reflectance. Section 439-21-2, at 994 m depth, has a reliable reflectance level of0.45 per cent. If the geothermal gradient at Sites 438 and439 is taken at 30°C/km, this corresponds to about36°C (estimate of 6°C at sea floor). The significantburial at Sites 438 and 439 is Plio-Pleistocene. A tem-perature of 36°C, however, is too low to give 0.45 percent with Plio-Pleistocene burial (Dow, 1978). This in-dicates that the particles arrived at the sites with anelevated reflectance level.

Estimates of the size of the huminite-vitrinite/iner-tinite particles seen in reflected light (Table 4) show atrend of coarser and more-angular grains in the inner-slope samples from Sites 438 and 439, while the more-distal, outer-slope Site 436 and the lower-inner-slopeSite 434 show finer and more-rounded grains. The mid-slope terrace Site 440, with more-rounded and finer-grained particles, shows more similarities with the distalsites (Table 4). This clear correlation of huminite-vitri-nite/inertinite particle size and shape with transportdistance must stem from the fact that most of the mater-ial is recycled and started its final depositional cycle inthe form of partially consolidated peats or coals. It hasthus been fractionated mainly on the basis of size, sincedensity and shape will have been relatively constant. Afurther trend, one of reduction in diversity of the lipti-nites away from the continent, appears to occur, but thesmall number of samples and the qualitative nature ofthe analysis means that confirmatory data should besought. Both these trends — reduction in size and angu-larity, and the impoverishment of the mainly higher-plant-derived liptinite populations — indicate a deriva-tion of the organic components from the west, the mid-slope terrace trapping the material destined for the outertrench slope.

The multimacerite particles of the Miocene clay stoneSection 439-21-2 may well be attributable to a source onthe postulated nearby Oyashio landmass (Figure 11).The existence of the Oyashio landmass near Sites 438and 439 also explains the higher abundance of unsatu-rated cyclic hydrocarbons in lower- to middle-Miocenesediments drilled at these sites (Figure 5, bottom; Table3). Preservation of large amounts of chemically ratherunstable compounds like sterenes and steradienes indeep-sea sediments has been observed during investiga-tion of core samples from DSDP Leg 47A (Cornford etal., in press) and Leg 50 (Rullkötter, unpublished). Inall cases, these compounds were found in slumped andturbiditic sediment sequences originating from down-slope mass flows at a continental margin. It can be as-sumed that the organic matter in these sediments origi-nally was produced in an oxygen-minimum layer withhigh bioproductivity at the shelf edge (Welte et al.,1979). Early down-slope transport after primary sedi-mentation can account for the preservation of unsatu-rated steroid hydrocarbons, because early diagenetic

1298


TABLE 4Kerogen Compositional Data

Sample(interval in cm) Maceralsa

Sizeb

(µm)Preser-vation0 Features Apparent in Transmitted Light

434-1-3, 115-125434B-15-1, 107-117

436-7-4, 115-125436-36-2, 115-125

438A-4-4, 110-120438-10-4, 115-125438A-18-2, 115-125438A-24-4, 110-120438A-46-3, 110-120438A-70-4, 110-120438A-84-1, 100-110

439-9-4, 100-110439-21-2, 115-125439-31-5,0-5

440B-3-5, 110-120440B-17-5, 125-135440B-39-3, 115-125440B-51-3, 125-135440B-68-2, 110-120

30:40:3070:10:20

65:20:1548:50:2

30:40:3045:25:3030:30:4050:40:1060:20:2040:30:2060:20:20

50:20:3095:5:185:15:1

65:15:2045:50:550:30:2055:30:1555:35:10

10-2015

1015

803025050302030

20500-

3015101520

MF

FMGFGFFFMM_M

FMFFF

Pollen, cuticle, woody cell debris.Much fine-grained vitrinite and inertinite; pollen and

woody debris.

As above.Poor recovery; rounded, well-sorted inertinite.

Inertinite subangular;a few reworked spores.Subangular to rounded inertinite; degraded liptinite.Inertinite subangular to angular; pollen, cuticle.Inertinite rounded to subangular; cuticle, pollen.Inertinite rounded and subrounded; degraded liptinite.Few larger vitrinite and inertinite fragments.Fine-grained, subangular inertinite.

As above; pollen.Lignite (?) fragments;multimacerites.As above, but much finer grained.

Cuticle, pollen, xylem tracheids with bordered pits.Inertinite rounded to subangular; fine liptinite debris.Much fine, rounded inertinite.Rounded inertinite, flaky liptinite.As above; some reworked spores.

aHuminite/vitrinite: inertinite: liptinite.t>Size of vitrinite/inertinite particles in transmitted light.cPoor, moderate, fair, good.

transformation of sterols assisted by micro-organismsstarts in the uppermost sediment layers (e.g., Nishi-mura, 1978). Rapid burial of the slumped and turbiditicsediments may have protected their organic matteragainst further attack by oxygen-enriched deep watersand micro-organisms. The «-alkane distributions ofthese sediments from Sites 438 and 439 (Figures 2 and 5,bottom) show, however, that they contain considerableamounts of terrigenous organic matter. Together withthe organic-carbon content of only about 0.5 per cent,this means that even in these sediments, which show bet-ter preservation of organic matter than any other sedi-ments drilled during Legs 56 and 57, the potential forpetroleum-hydrocarbon generation has to be rated low.

Typical non-aromatic-hydrocarbon distributions ofLegs 56 and 57 sediments (Figure 5, top) show a domi-nance of terrigenous wax alkanes. Distal transport fromthe Japanese landmass prior to final deposition — per-haps comprising periods of non-reducing environments— favors preservation of fully saturated alkanes with-out functional groups. Terrigenous organic materialgenerally seems to be more resistant to degradation thanmarine organic matter (cf. Cornford et al., in press).This might explain the terrigenous character of the solu-ble organic matter even in the oxidized pelagic Miocenesediment of 436-36-2, recovered at Site 436 on the outertrench wall. This gives rise to difficulties in differentia-tion between the organic matter of the deep-sea sedi-ments and the organic matter in the continental-slopedeposits, especially in consideration of the origin of thesedimentary organic matter of the accretionary wedge.

The Oligocene silty sandstone of 439-31-5 showshydrocarbon distributions that are typical for mature

organic matter. Although contamination should not befully excluded, an explanation for this observationmight be given by boulders of acidic igneous rocksdiscovered in Cores 439-32 to 439-36, which point tonearby volcanic activity. Thus, it is not completelyunlikely that locally high heat flow produced high-ma-turity organic matter, which led to migration of hydro-carbons into the sandstone. However, since microscopyshows Section 439-31-5 to contain vitrinite with anom-alously high reflectance (Table 5; Figure 9a), the sedi-ment itself may either be of higher maturity or containallochthonous, recycled particles of higher rank. In anycase, the good agreement between extractable-hydro-carbon analysis and kerogen microscopy seems to ruleout contamination. Recycled particles normally do notcontain much extractable organic matter, while thissample has a high extract yield (14 mg/g Corg) comparedto most others (Table 1). In the absence of an Oligo-cene/Miocene unconformity, the evidence thus seems topoint to a late-Oligocene source of geothermal heatwhich produced a vitrinite-reflectance level of 0.74 percent in Section 439-31-5, but the heat source had cooledby the early Miocene, leaving Section 439-21-2, some110 m higher, with a reliable huminite-reflectance levelof not more than 0.45 per cent.

ACKNOWLEDGMENTS

We would like to thank Dr. M. Radke and Dr. R. G.Schaefer for carrying out extraction and gas chromatography,respectively, and Miss I. Jacobs, Miss M. Neumann, Mr. W.Benders, Mr. U. Disko, Mr. F. J. Keller, Mr. K. Otterberg,Mr. J. Schnitzler, Mr. H. G. Sittardt, and Mr. H. Willsch fortechnical assistance. Our thanks go to Dr. M. Bjor0y (Con-tinental Shelf Institute, Trondheim, Norway), Dr.-Ing. H. W.

1299


fc . 0 .

********** X

*LLL*LVL

G. NUMBER. K4617C

XVUw

xvvvv uvw w w vv wuyvuv vwwu iVVVVVVXWVVVVVV III1

vyyuvwuuyyywuvuiu

.JO A

i:

I I11VI

IVIV V

HATE:

Ii

I i1111 i

24-AU6-78

I 11 1 I— *' 3 . 5

MACERAL COMPOSITION ( # )

V t l l L T × : 31 : 5 : 4

==== = NUMBER OF REFLECTANCE VALUES MEASURED^. — ^ — — — ^ f. — Φ " •πr * * * '

0

101520

30

3540

45

50

5560

657075

80

85

90

95

10 0105

110

1 15

120125130135

140145150

1 5 5

160

165

170

175180

185

190

195

L>OO

4

9

- 14

24- 29- 34

3V44

49

- 5459

- 64

697479

- 84- 89- 94

99

- 104- 109

- 114- 1.1.9

- 124- 129- 134- 139

- 144- 149- 154

- 159 .

- 164 :169

- 1.74 :

- 179 :- 184

189 :- 194 :~ 199- 204 '

I 6

: ii16

' 2 7: 3i

3540

> 4551

5861

66

7075

8086

90

96

104106

111115

1211251 30

1.35

140

15716216?

172

185

196

9

14

17

2 73135

40

46

5 158

63

68

70/' J

81

87

90

96

104107

112

11 5

126

134139142

L58

1 68

19

2831

3641

46

5259

63

69

7 276

8189

93

9 7

108

1 1 4116

1.28

139

169

3 2

3 7

41

46

!j J

7 A

77

8?

89

94

98

116

129

169

33

37414 6

53

77

82

9498

34

3742

4754

78

83

38 39 39

43 43 4347 4954

84

43 44 44

Figure 7. Reflectance histograms of kerogen particles from Section434B-15-1. a. Macerals, using codes V = huminite/vitrinite, I = inerti-nite, L = liptinite, and X = unknown, b. Reflectance values (×IOO)plotted at 0.05 per cent intervals; mean reflectance of lowest peak is 0.41per cent.

Hagemann (Rheinisch-Westfálische Technische Hochschule,Aachen), and Dr. D. Leythaeuser (KernforschungsanlageJülich) for carefully reading and reviewing the manuscript.

Samples were made available by participation of the Deut-sche Forschungsgemeinschaft (DFG) in the DSDP/LPOD pro-gram. This is gratefully acknowledged, as is financial sup-port by the German Ministry for Research and Technology(BMFT), grant No. ET 3070 B.

REFERENCESCornford, C , 1979. Organic petrography of Lower Cre-

taceous shales at DSDP Leg 47B, Site 398, Vigo Seamount,eastern North Atlantic. In Ryan, W. B. F., Sibuet, J.-C,et al., Init. Repts. DSDP, 47, Part 2: Washington (U.S.Govt. Printing Office), 523-528.

Cornford, C , Rullkötter, J., and Welte, D. H., 1979. Theorganic geochemistry of DSDP Leg 47A, Site 397, eastern

1300


I.1..O.G. NUMBER: K4619C.M.JA DATE: 25--AUG-78

* X* V V* XVWV VVVI* vvvvu vvvvvv v i :i i* vvvvvvuvvvuvwuw w i i i i II II

. 5 1 1 .5

I I I

3 . S

MACERAI... C O M P O S I T I O N ( * ) :

V : I : L { × 4i I 20 i o J 2

E . O . G . NUMBER} K2623C.M.JA DATE: 25-AUG-78

* X V K>* V V V* yy w y

* W W X V W III .* vy yy w w yyy VII i* X WVUWWWWVVV VII I II I I* L. WWWWWWWVI VI.TIT I I I I I I I I 1* L L X V V V V V V V V V V V V V V V V V V V V I V V I . I V I I I I . i l I I I I I I I I I I I I I I I :

^ -...-..—. — ̂ ― -– — _.. - __ .._ ._. ̂ — .„. ._ _ ^ __ ._ _ ^ — ..„ _- - -.. .„. ^_ Φ_

.5 1 1.5 2 2.5 3 >3.=== ZR-OIL == >>

MACERAI... C O M P O S I T I O N ( # ) I

V : I : L . { X VO : 56 : 3 : 4

KEY:V V I T R I N I T E » I INERTINITE»L LIPTINITE!.X UNKNOWN»>==VALUE :: 3.5ZR

#BASED ON POLISHED PARTICLES. IE.NO AMORPHOUS MATTER.

Figure 8. Maceral-reflectance histograms of kerogen particles, a. Section 436-7-4,mean reflectance of lowest peak is 0.32 per cent. b. Section 438-10-4.

North Atlantic: organic petrography and extractable hydro-carbons. In Ryan, W. B. F., von Rad, U., et al., Init.Repts. DSDP, 47, Part 1: Washington (U.S. Govt. Print-ing Office), 511-522.

Dow, W. G., 1978. Petroleum source beds on continentalslopes and rises. Bull. Am. Assoc. Petrol. Geol., 62, 1584-1606.

Ishiwada, Y., and Ogawa, K., 1976. Petroleum geology ofoffshore waters around the Japanese Islands. United Na-tions ESCAP, CCOP Tech. Bull., 10, 23-24.

Matveef, A. K., 1972. Map of the Coal Fields of the World.Moscow (Nedra).

Nishimura, M., 1978. Geochemical characteristics of the highreduction zone of stenols in Suwa sediments and the en-

vironmental factors controlling the conversion of stenols tostanols. Geochim. Cosmochim. Ada, 42, 349-57.

Radke, M., Sittardt, H. G., and Welte, D. H., 1978. Removalof soluble organic matter from rock samples with a flow-through extraction cell. Anal. Chem., 50, 663-5.

Scholl, D. W., and Marlow, M. S., 1974. Sedimentary se-quence in modern Pacific trenches and the deformedcircum-Pacific eugeosyncline. Soc. Econ. Paleont. Miner-al., Spec. Publ., 19, 193-211.

Stach, E., Mackowski, M. Th., Teichmüller, M. Taylor,G. H., Chandra, D., and Teichmüller, R., 1975. Stach'sTextbook of Coal Petrology (2nd Edition): Berlin (Ge-brüder Borntraeger).

Welte, D. H., Cornford, C , and Rullkötter, J., 1979.Hydrocarbon source rocks in deep sea sediments. Proc.11th Annual Offshore Tech. Conf, 1, 457-464.

1301


E . O . G . NUMBER: K 4 6 3 6 J . J O A DATE. 29-AUG-78

W XVVV IV X

UVVVV V W X

I Ii

II

1 . 5%R- ü IL

3 . 5

MACERAL C O M P O S I T I O N ( # )

V:I :L:× 42

E . O . G . NUMBER: K4641J.JOA DATE: 29-AUG-78

* V* X V* V V V V VV* w wyv v yv v i i* x xvy uuvuvv w v v i i* xxxwvuwyvwvvyvy vxxn n^ _ _ Φ _ * . . - * * . * . . . . ~ — * - - - *

.5 1. 1.5 2 2.5 3 >3.5= := = = :=.-== %R - O I L -- = y;;

MACERAL COMPOSITION ( # ) J

V : I J L : X • - 46 : e : o : e

K E Y : V -" VI: T R I N I T E » I = I N E R T I N I T E > I.. L I P T I N I T E X U N K N 0 W N r > = V A L U E > 3 . 5 Z R

*BASED ON POLISHED PARTICLES. IE.NO AMORPHOUS MATTER.

Figure 9. Maceral-reflectance histograms of kerogen particles, a. Section 439-31-5,mean reflectance of lowest peak is 0.74 per cent. b. Section 440B-39-3, mean re-flectance of lowest peak is 0.43 per cent.

1302


αa>a

2 0 0

4 0 0

6 0 0

8 0 0

1000

1200

4 3 9

4 3 9

Localheating?

0.2 0.3 0.4 0.5 0.6 0.7

Vitrinite reflectance {% Ro)

Figure 10. Mean huminite/vitrinite-reflectance valuesfor samples from DSDP Legs 56 and 57, JapanTrench. The arithmetic mean and a standard devia-tion are plotted as a point and bar, respectively. Thetrend line may not be realistic, as argued in the text.

TABLE 5Huminite/Vitrinite Reflectance Data

Sample(interval in cm)

434B-15-1, 107-1 17

436-7-4, 1 15-125

438-10-4, 115-125438A-24-4. 110-120438A-46-3, 110-120438A-70-4, 110-120438A-84-1, 100-110

439-21-2, 1 15-125439-31-5,0-5

440B-3-5, 110-120440B-39-3, 115-125440B-51-3, 125-135440B-68-2, 110-120

Sub-BottomDepth

( i n )

419.1

61.2

84.7283.2492.7722.2850.5

994.7-1105

165.6504.7618.7778.6

* o('"')

0.36

0.32

0.350.370.380.350.38

0.450.74

0.300.430.520.36

Lowest-ReflectanceH uminite/Vitrinitea

N

8

17

165

1343

3023

812163

Std. Dev.

(0.03)

(0.07)

(0.03)(0.08)(0.02)

_

-

(0.04)(0.07)

(0.04)(0.05)(0.05)

Range

0.31-0.40

0.20-0.43

0.30-0.400.31-0.490.35-0.420.29-0.400.37-0.39

0.37-0.540.60-0.84

0.23-0.340.34-0.560.42-0.58

Comments on Total-Reflectance Histogram

Multimodal (Figure 5).

Multimodal (Figure 6).

Multimodal (Figure 6).Broad peak between 0.27 and 1.44%.Broad peak between 0.17 and 1.02%.Not measured.Scattered values between 0.38 and 1.49%.

Not measured.See Figure 7.

Values between 0.29 and 1.48%.See Figure 7.Multimodal between 0.25 and 1.38%.Not measured.

ΛRQ is the arithmetic mean of selected huminite/vitrinite reflectance values; .Vis the number of measurements

1303


WEST

HONSHU IS.

KitakaMassif

HOLOCENE

CONTINENTALSHELF

CONTINENTALSLOPE DEEP SEA TERRACE

438 439INNER TRENCH SLOPE

EAST

JAPANTRENCH

438 439

_L_L

LATE MIOCENE

EARLY MIOCENE

438 439

_LJ_ p^^^^?

LATEOLIGOCENE

OYASHIOANCIENT LANDMASS

438 439

congl.

Figure 11. Diagrammatic history of subsidence of the Oyashio landmass, based on DSDP Leg 57 sampling andJNOCmultichannel records. K = Cretaceous, ECz = early Cenozoic (Paleogene), LCz = late Cenozoic (Neogene). Zoneof accretion and oceanic igneous crust shown diagrammatically in the Holocene section. (Figure taken from Leg 57shipboard summary.)

1304

project legs 56 and 57, japan trench: organic petrography and

Documents