Stratigraphy of eastern

13. STUDIES ON SAPROPELS

13.1 STRATIGRAPHY OF EASTERN MEDITERRANEAN SAPROPEL SEQUENCESRECOVERED DURING DSDP LEG 42A AND THEIR PALEOENVIRONMENTAL

SIGNIFICANCE

Robert B. Kidd, Institute of Oceanographic Sciences, Surrey, United KingdomMaria Bianca Cita, Istituto di Geologia, University of Milan, Italy

andWilliam B. F. Ryan, Lamont-Doherty Geological Observatory, Palisades, New York

ABSTRACTDuring Leg 42A of the Deep Sea Drilling Project, we recovered

over 150 discrete layers of dark colored organic-rich sediment atfive sites in the eastern Mediterranean. We consider about two-thirds of these to be sapropels, on the basis of a definition thatrecognizes a sapropel as "a discrete layer, greater than 1 cm inthickness, set in open marine pelagic sediment and containinggreater than 2.0% organic carbon."

This paper includes a catalogue of all the sapropels andsapropelic layers recovered, including analysis of their sedimentarystructures, general composition, fossil content, and stratigraphicposition.

Sapropels are identified, dating back to the middle Miocene,and a single occurrence is recorded in the Pleistocene of thewestern Mediterranean. We recognize that the record of sapropelstratigraphy provided by Leg 42A and Leg 13 drilling is incom-plete in both time and space. Despite this, we can make tentativebasin-wide correlations between individual layers at least back tothe early Pliocene, proving that these stagnation events cannotmerely be linked to glacial phenomena. Their widespread distribu-tion and the wide range of depths argue against the presence of an"oxygen-minimum layer."

Stagnation of the Mediterranean basins resulted from a numberof interacting factors which caused stratification within the watercolumn. This was a transient situation that developed repeatedlysince the mid-Miocene when the Mediterranean took on it presentconfiguration. Glacial expansions were primary factors bringingabout stratification in the Pleistocene; whereas, similar factors areas yet unidentified in the Pliocene and Miocene sediments.

INTRODUCTION

During Leg 42A of the Deep Sea Drilling Projectwe recovered more than 150 discrete layers of darkcolored organic-rich sediment (sapropels and sapro-pelic layers) within the sediment columns at five sitesdrilled in the eastern Mediterranean. We believe theseare the sedimentary expression of periods of stagnationof the Mediterranean bottom waters.

Previous studies of similar layers in piston coreshave resulted in a correlated sapropel stratigraphy forthe Quaternary (McCoy, 1974; Ryan, 1972; Van Strat-ten, 1972). During Leg 13 some of these Pleistocenelayers and one upper Pliocene layer was recovered(Ryan, Hsü, et al, 1973). Leg 42A drilling (Figure 1)proved the existence of sapropels and sapropelic hori-zons in sediments back to the middle Miocene. How-

ever, the sapropels occur at relatively few DSDP sitesand none is thought to contain a complete sequence ofthese layers. Consequently our pre-Quateraary recordis incomplete in both space and time and we cannotdevelop a complete Pliocene and Miocene sequence ofcorrelations, as has been done for the Quaternarypiston cores. Nevertheless, projected additional analy-sis of materials from both cruises may result in thecorrelation of certain pre-Pleistocene horizons withinthe separate eastern Mediterranean basins. Indeed, thisalready appears possible for some Leg 42A layerswhich were cored more than once.

To this end, we present here an inventory of ourLeg 42A findings which includes all the observationaldata in terms of location, structure, general composi-tion, fossil content, and stratigraphic position availableto date for each of the sapropels and sapropelic

421

R. B. KIDD, M. B. CITA, W. B. F. RYAN

))/

-40°

f NORTH

•30°N

ó°

(

0°

E U

j / BALEARIC

10°

R O P E

\ \

i/\TYRRHENIAN.

BASIN I O 4

371

AFRICA

10°

> # 3 7 3 ^

SEA ,

•

20°

Λ

IONIAN SEA

.374

125°

20°

\i127

128 <5>

• 3 7 7126

\

/

\

° L

4% <S^AEGEANV

••SEA. ‰378 i*1 2 9 O

o130

^ 131

30°

B L A C K S E A

s ^ V ^^ - ^

^ ' A N T A L Y A - ^ )B A S I N , ^ ^ <

LEVANTINE /o BASIN /

30°

40°

r\ )

40°•

4 0 ° E 3 ( f •

Figure 1. Index map showing the location of Leg 42A and Leg 13 sites.

horizons. A discussion follows, which focuses mainly onstratigraphic aspects but also introduces comments onthe paleoenvironmental significance of the sequence oflayers. This discussion is preliminary in that the obser-vational data are still incomplete and further investiga-tions are foreseen especially with regard to isotopic,organic, and trace element geochemistry. The inventoryshould nevertheless provide useful data for otherscientists interested in oceanic stagnation processes.

TERMINOLOGY OF SAPROPELS

Dark colored layers rich in organic carbon set inmarine sequences were first described from easternMediterranean piston cores by Kullenberg (1952) andsimilar horizons were later named "sapropelitic layers"by Olausson (1960). Since that time beds of this typehave become known simply as sapropels. Loose defini-tion of this term has resulted in some inconsistenciesappearing in the literature. Strong variations in organiccarbon content exist between layers and, as an aid inenvironmental interpretation, we need to distinguishbetween "true sapropels" and the less organic carbon-rich "sapropelic layers." Our Leg 42A terminology,first proposed at a meeting of the shipboard sedimen-tologists at Lamont-Doherty Geological Observatory inSeptember 1975, defines, a sapropel as: a discrete layer,greater than 1 cm in thickness, set in open marinepelagic sediments and containing greater than 2.0%organic carbon by weight. A sapropelic layer is similarlydefined, but differs in that it contains between 0.5%and 2.0% organic carbon. The above definitions meanthat, for example, whole sequences as in the Burdiga-lian of Site 372 with just over 0.5% organic carbon, donot qualify as sapropelic layers and similarly, diatoma-ceous layers rich in organic carbon within the eva-

poritic sediments of Site 374 would not qualify foreither term. Of the 74 dark layers encountered on Leg42A and for which the organic carbon content wasdetermined by the DSDP (La Jolla), Technische Uni-versitat (Munich); and IFP (Paris) shorebases, 54 weredesignated sapropels.

RECORD OF SAPROPELS AND SAPROPELICLAYERS RECOVERED DURING LEG 42A

Sapropels and sapropelic layers, ranging in colorfrom black to very dark green and brown, were en-countered in all of the eastern Mediterranean sitesdrilled during Leg 42A. With one notable exception,they were entirely absent from the western Mediterra-nean sequences (Sites 371, 372, and 373A). Thisexception is the disturbed lump of greenish black marlrecovered in the topmost core of the Tyrrhenian Sea(37A-1-2, 0-5 cm). It contained 2.1% organic carbonand a mixed Pleistocene faunal assemblage.

Tables 1 to 5 catalogue the basic observational datafor both the sapropels and sapropelic layers found atSites 374 through 378, respectively. By definition, allare individual layers at least 1 cm thick, with theexception of some which we term "multiples." Theseare beds composed of a number of individual layers;some of which are laminae less than 1 cm thick."Multiples" composed of numerous laminae, all lessthan 1 cm thick, would still be grouped as a singlesapropel or sapropelic layer if the layers are obviouslyinterrelated, and if they contain greater than 0.5%organic carbon (Figure 2).

STRUCTURE

Because most of the layers reported here occur inunlithified Pliocene-Quaternary sediments in the upper

422

TABLE 1

Sapiopels and Sapiopelic Layeis at Site 374

Sec-Core tion

2 12223

CC

Location

Top(cm)

165225195

123

Base(cm)

166235296

127

CoreHole

Depth(m)

157.0

161.5

Multi-ples

(No oflayers)

Structure

Contacts

Sharp above & belowSharp above & belowSharp above & belowSharp above & belowSharp above & below

"Ooze"Layer

Above/Below

Above

O G

rad

ing

-αn>ot-tM

3

(B)

<L>

£< L

amin

a

Thick-ness(cm)

111149

Composition

Sand Fractions

Abun-dance Mineralogy

High Mica

OrganicCarbon

(max. %)

9

0.6

9

2.961.62

Classification

Sapro-pels(S)

S

Sapro-peüc

Layers(SL)

SL

SL

Org.CUn-

known

9

9

NannoZone

NN20

Age

ForamZone

N22

Designation

Pleistocene

1 103 106 208

1 121 1221 125 1291 140 143

CC 209

Sharp disturbed

Both sharpBoth sharpBoth broken

High

Above

Above

Low

LowLow

Vole, glass,quartz

Quartz

Mica, qz.pyrite

4.21

2.624.968.720.67

NN20 N22 Pleistocene

SL

22

CC

5662

60 251.564

Both sharpBoth sharp

Low

High

Mica qz.Mica qz.,

pyrite

2.48

2.03NN20 N22 Pleistocene CΛ

HtI

m

aHCD

S2Zm>ZCΛ

>

tarCΛ

CDOcwzninCΛ

22223

322233335

333848

12110

49194297

75063

124110

354551

12212

51204398

95466

126111

256.5297

9

304330.5

Both sharp>) Sharp curved

DisturbedBoth sharpBoth Thin lamina

1 cm belowBoth faulted topBoth sharpBoth sharpBoth sharpBoth faultedBoth sharpBoth sharpBoth sharp

>) Sharp

AboveMed.

High

High

MicaQuartz

-

_

MicaMicaMica

_

_

1.663.208.991.439.87

16.71i

9.561.823.85

2.882.772.13

SS

ss

s

ssss

SL

SL

SL


NN18

MPl-S LateNN17/ M n : > PüoceneNN16

11

Above

V. lowHighV. low

HighHighV. low

M P 1 " 5LatePliocene

340.0NN15

g

9

1025

3

23

36

014

39 349.5to 359

7 35916 to

368.5

? Disturbed

Sharp disturbedBoth sharp

Part of core catcher SapropelicSide wall core -- partly Sapropelic

3

72

V. low

V. lowV. low

Mica, qz.

-

9

1.932.31 S

SL

? NN13

NN12

MP1-3

MP1-2

EarlyPüocene

EarlyPUocene

EarlyPUocene

111222

91100110

57112137

93101111

611.21150

378.0

381.5

Sharp faultedBoth sharpBoth sharpBrokenSharpBase sharpTop gradn l

LayersinDolomiticL. St.

G

B

BB

XXX

21149

13

MP1-19

0.72

EarlyPUocene

SL

No Sapropels below but dolomitic mudstones (often diatomaceous) have organic carbon values ranging up to 5.3%.

to

ço

Core

4

5

6

7

8

9

Sec-tion

4

1223

35

CC

335

CC

2451

122334555

Location

Top(cm)

40

1464

8221

109139

455848

105215141

100488839

13316

713

113

Base(cm)

62

1476

8823

111141

535550

150345244

101499856

13418

814

114

CoreHole

Depth(m)

245.5to 252360

369.5

461

470565to574.5622 to631.5650.5

660

Multiples(No oflayers)

4 gradedunits

MultipleMultipleMultiple

MultipleMultiple

Structure

Contacts

Sharp baseGrad1 topsBoth sharpBoth sharpBoth sharpBoth sharp


Both sharpBoth sharpBoth sharp


Both sharp

"Ooze"Layer

Above/Below

2ü

(G)

G

TABLE 2Sapropeis and Sapropelic Layers at Site 375

T3

£Ot-*t-t3

«

(B) (

B

Numer'sLaminae

Numer'sLaminae

•αu

imin

a

j

:×)X

Thick-ness(cm)

22

1262

229

1322?

45?13

13

11

1017

12111

Composition

Sand Fraction

Abun-dance Mineralogy

Rich indetrital mins.

Mostly biogenicsome pyrite

Rich indetrital mins.

OrganicCarbon(max. %)

7.02

9

9

2.39

3.729

4.312.180.540.51

2.540.710.512.0

4.109

9

3.825.063.152.80

9

9

Sapro-peis(S)

S

S

S

ss

s

ss

s

s

Classification

Sapro-pelic

Layers(SL)

SLSL

SLSL

Org.CUn-

known(?)

9

9

9

?

9

?9

99

?9

Nan noZone

NN11

NN11

NN10

NN10

NN7

NN6

NN5

Age

ForamZone

N17

N16

N16

N16

N14/N i l

N12/N10

N9/N10

Designation

Late Miocene(Tortonian)



Late Miocene(Tortonian)Middle Miocene(Serravallian)

Middle Miocene(Serravallian)

DD

, M. B

. CIT

>

ço

X

TABLE 3Sapropels and Sapropelic Layers at Site 376

Location Structure

CoreHole

Sec- Top Base Depth MultiplesCore tion (cm) (cm) (m) (No of layers) Contacts

"Ooze" .1 | gLayer *g § E

Above/ o 2Below (G) (B) (X)

Composition Classification Age

Thick- Sand Fraction ^ ^ Sap ro_ness Abun- Miner- Carbon pels(cm) dance alogy (max. %) (S)

Sapro- Org. Cpelic Un-

Layers known(SL) (?)

Nan noZone

ForamZone Designation

334

13140

3

35150

5

1 1 52 80 0 Disturbed Disturbedprobx gradedunits (3)

2 40 146 Disturbed Intenselymin. (2) deformedPoss. (3) DisturbedPoss. (2) Disturbed

Disturbed

10 11.5 Disturbed

64 71 Disturbed

76 77 Disturbed

80 89 7.5 Disturbed

94

1311

1201

3 33 863 1404 04 70

96

15030

1452

301241484075

7.5

Poss. 4separate layers

(Poss. 3)(Poss. 2)(3)

17.0

DeformedintenselyDeformedDeformedDisturbedBoth sharpBoth sharpInts. defmd.Bounds sharpBounds sharpBounds sharpBounds sharpHighlydeformed

Thickn<1

Crest (2)Thicknd

Crest (1.5)Thick nd

Crest (7)ThickndCrest (1)ThickndCrest (8)

2.78

2.21

5.55

2.64


Vole.glass

9

5.281.04

2.707

1.902.32

S

S

S

SL

SL


30 67 17.0 to (3)26.5

Intenselydeformed

3.44 NN20 N23/ PleistoceneN22

112120

1456470

105

11512114070

1466572

111

36.0

45.5

Both sharpBoth sharpBoth sharpBoth sharpBoth sharpBoth sharpBoth sharpBoth sharp

31

1-521126

4.56

0.537

8.92


SL

NN18

NN17/NN16

MP1-5

MP1-5

LatePliocene

6

8

3442

945595

0

97759871

45.5to55.064.5 to74.0

2 thin layers4 thin layers

Both sharpFaultedSharp, faultedSharp

AboveAbove G

B XXX

320

3Up to 2

1.233.75

7

7

SSL NN19

, NN12

? NN11

N22

MP1-1

N17

PleistoceneEarlyPliocene

Late Miocene(Messinian)

Organic carbon rich mudstones and graded beds in Cores 9 to 12 — all late Miocene (Messinian)

wOmZom00

00TABLE 4

Sec-Core tion

1 22222

Location

Top(cm)

14010122897

Base(cm)

15011133298

CoreHole

Depth(m)

190.5

193.5

Multi-ples

(No oflayers)

Structure

Contacts

Gradation'lBoth sharpBoth sharpBoth sharpSharp-broken intoelongatedmud pebbles

"Ooze"Layer

Above/Below

Below?

radi

ng

O(G)

GG?

urro

w

oa(B)

ted

imin

a

(X)

X

Thick-ness(cm)

101141

Composition

Sand Fraction

Abun-dance

HighHigh

Miner-alogy

OrganicCarbon(max. %)

2.914.06

?3.473.47

Sapro-pels(S)

ss

ss

Classification

Sapro-pelic

Layers(SL)

Org.CUn-

known(?)

9

NannoZone

MixedPleist.

Age

ForamZone

N22

Designation

Pleistocene

öö

DO

Q

CO

(Marlstone at base of Core 4, Section 3 — organic carbon rich)

TABLE 5Sapropels and Sapropelic Layers at Site 378/378A

Location Structure

Sec- Top BaseCore tion (cm) (cm)

CoreHole Multiples

Depth (No of(m) layers)

"Ooze"Layer

Above/Contacts Below

v 1 1O m J

(G) (B) (X)

Composition Classification Age

Thick-ness(cm)

Sapro- Org. COrganic Sapro- pelic Un-

Abun- Miner- Carbon pels Layers known Nanno

Sand Fraction

dance alogy (max.' (SL) (?) ZoneForamZone Designation

1A 1 61 74 46.0 Disturbed3 10 15 Disturbed4 22 30 Top sharp, B

base grad11*4 105 116 At least 3 Bothgradnl B4 131 147 Poss. 3 Bothgradnl B5 70 89 At least 4 Both gradol Below G? B

_____ 55.5 or base sharp

13?5?8

9?16?19

7

1.72

7

2.39

SL


131 13315 23

84.0

65 72 93.5

Both? grad«iSharp base, Gtop grad1"Top gradnl Above & G

below

7.15

6.98NN19 N22 Pleistocene

112 122 103.0 to112.5

Both grad"! Above 10 4.08 NN19 N22 Pleistocene

6 12

3

3

14415

81

92

15021

83

95

141.0 Disturbed AboveBase sharp, Abovetop grad1"Base sharp, Above &top grad1" belowBase sharp,top grad1"

7

5.16

4.63

5.50

S

S

s NN18 MP1-6

3

3

99

121

103

125 150.5

Base grad"1,top sharpBase sharp,top gradn*

G

B

B

X

X

4

4

69 85 169.5 to179.0

Both gradn l16 3.22 NN16 MP1-5 Late

Pliocene1 117 126 217.0

to2 3 46 226.5

Top gradn l,base sharpTop gradn l,base sharp

9

43 Vole.glass

3.76

5.17NN16 MP1-5 Late

Pliocene

3A

CΛ "

01 *

st> pQCΛ

34 43 293.0

56 68

0 16

15 2052 6095 100

16 25

30 4370 83103 107

72 85

102 111

130 142

40 55126 137

302.5

Top grad1"

base distrbd.Top?,base gradn l

Both grad111,both gradn l

Both grad111,distrubedTop grad"1,base sharpBoth gradn l

Both gradn l

Both grad"1,disturbedTop sharp,base grad°lBoth sharpTop sharp,base?Both sharpBoth sharp,disturbed

G?G?

BB

B

9?12

16?Burrows below to 48 cm

G?

G

BBB

B

BBB

B

B

B

BB

B X

585

1313

4

13

912

159

21411

Burrows below to 106 en

G?G?

BBBBB

B

B

B

BB

X

XXX

14433625

24

3294

NN13 MP1-3EarlyPliocene

70

Im>GO

>

JβO

arCΛtnOGtnnmCΛ

11

303o

8.

la p

ro

4

4

44

144 150 302.5

1 318 3295 106

13018466369919

3746

144254966751124

39

50

63 6683 85123 131133 137 2?

Disturbed,sharpBoth sharpBoth grad"1

Both sharp

Both grad n l

Both grad"1

Both grad n l

Both sharp?Both sharp?Both sharpBase sharpTop grad11',both sharpTop gradnlBase sharpBoth grad n l

Both grad n l

Both grad n l

Both grad n l

NN14

4.161.695.01

S

S

1.51

SL

SL MP1-3

NN13

EarlyPliocene

MP1-2

NN12

312.0


Ov

Figure 2. "Multiple" sapropel horizon at 38 to45 cm in Core 5, Section 2 of Site 374(3.2% org.c). Note the relationship of thethin black lamina to the overlying main body

of the sapropel. Neither the thin lamina at28 cm nor the isolated lump of sedimentrich in organic carbon at 48 cm was, bydefinition a sapropel or sapropelic layer. Asapropelic layer is present at 33 to 35 cm.(1.66% org. C). (scale in cm)

parts of the holes, considerable drilling disturbance isoften evident (Figure 3). In these cases, detailed struc-tural analysis is impossible. Although we can recognizediscrete layers, flowage upcore caused by the drillingprocess may result in our recording a somewhat exag-gerated thickness.

Although the thicknesses of individual sapropels andsapropelic layers encountered on Leg 42A range up to45 cm, only a small number—46 out of 150 or morerecovered—are greater than 5 cm thick. The thickestlayers found at each site were respectively: 13 cm atSite 374, about 45 cm at Site 375, 20 cm at Site 376,10 cm at Site 377, and 43 cm at Site 378. At Sites 374,376, and 377, the layers noted above were the onlyones greater than 10 cm thick whereas at Sites 375 and378 beds of this thickness make up over a quarter ofthose recovered. Consequently we cannot demonstratea relationship between the occurrence of thick layersand the present environments of the sites; Site 375 waslocated on the Florence Rise and Site 378 was drilledin the Cretan Basin. Nor can thickness of the individ-ual Leg 42A sapropels be related to sedimentationrate; the thickest layers at Site 374 are recorded in theearly Pliocene where calculated sedimentation rates areconsiderably lower than in the rest of the hole.

Nesteroff (1973), in his study of the Leg 13 cores,identified two main sapropel types, primarily on thebasis of structure: "pelagic sapropels," similar to thosewhich had been described in piston cores up to thattime, and "sapropelitic turbidites" for those bedsencountered during Leg 13 drilling, which showedstrong evidence of current deposition. Pelagic sapropelswere described as always having sharp contacts withtheir host sediments, nannofossil oozes. They weremade up of a distinct layer of gray marly ooze overlainby a layer of black sapropel and this topped by graymarly ooze or by pure nannofossil ooze. The secondtype was considered by Nesteroff to be a sapropel-stained current deposit since it had structural charac-teristics interpreted as cross-bedding and grading, aswell as sand-silt laminae and winnowed concentrationsof foraminifers. On the basis of these criteria, such bedswere considered by Nesteroff to be either sapropel-stained turbidites or contourites. In Tables 1 to 5,structural features were recorded with this generaldistinction in mind. This proved useful, but also re-vealed a much greater structural complexity for sapro-pels than had hitherto been recognized. In addition,the importance of burrows within some sapropels,especially those of the Cretan Basin Site 378, wasrevealed. Their presence indicates either that a bottom-living fauna existed during the formation of some typesof sapropel, or that the post-sapropel populationsought out the organic carbon-rich substrate as particu-larly desirable grazing territory.

428

STRATIGRAPHY OF EASTERN MEDITERRANEAN SAPROPEL SEQUENCES

Figure 3. Sapropel layers showing flowage causedby drilling disturbance of soft sediments (376-2-4, 0-25 cm). The location and thicknesses of

such layers reported here are those of theircrests, (scale in cm)

We assign most of the Leg 42A sapropels andsapropelic layers to NesterorTs pelagic group. Thesehave sharp contacts with the host marls and have nointernal structure. At Site 374 all but 6 of the 38 layerswere of this type. Corresponding figures for the othersites are: Site 375, all except 5 of the 25 recorded andat Site 376 all but 5 of the 32 recorded are of pelagictype. On the other hand, the layers at Site 378 werealmost all burrowed throughout and those at Site 377were difficult to evaluate with at least half seemingly ofturbidity current origin. However, the Leg 42A recordclearly indicates that there are variations within thispelagic group. Occasionally only one of the contacts issharp and gradational boundaries can occur at eitherthe top or the base. Also, a thin sapropelic lamina issometimes present a few centimeters above or belowthe sapropel bed, separated from it by nannofossilmarl (Figure 4). We looked for the association ofrelatively pure nannofossil ooze layers with the sapro-pels, as reported by Nesteroff from the Leg 13 sedi-ments, in the Leg 42A cores. We found the associationonly in a small number of the layers and its occurrenceseems entirely random. In most cases a layer of purenannofossil ooze occurs above the sapropel or sapro-pelic horizon, but it can also be found below. Also itspresence is not confined to layers of the pelagic type; itis also found with clearly current-deposited beds. Theseobservations suggest that the onset and termination ofstagnation for deposition of the pelagic-type sapropelwas usually a relatively sudden event, but also, thatexceptions occurred when unstable conditions prevailedover a more prolonged period, resulting in gradual oroscillatory onsets and terminations.

Sapropels and sapropelic layers which, because oftheir cross and oblique lamination grading, and/orconcentrations of coarse material (Figure 5), we be-lieve to be of current-deposited origin, were encoun-tered at all the Leg 42A eastern Mediterranean sites.This is a total of 41 layers, but half of these are in theSite 378 cores. Site 377, drilled as a continuation ofLeg 13, Site 126, contained two sapropels of this type,and confirmed NesterofTs earlier observations.

Twenty-three of the total 41 current-deposited sap-ropels recovered were graded, but the exceptionsconfirm the notion that not all sapropels of this typecan be assigned to a turbidity current origin (Figure 5).No clear picture emerges from consideration of theircontacts with the host sediment. Almost all in Sites 374to 377 have sharp basal and upper contacts. But at Site378 a complete range from "both sharp" through "oneor other gradational" to "both gradational" occursand again the presence of an associated pure oozelayer appears to be random.

The burrow structures that were associated withmany of the Leg 42A sapropels and sapropelic layersprovide evidence of bottom conditions during, orimmediately after, their formation. Leaving aside Site378 for the moment, burrowing associated with sapro-pels and sapropelic layers occurs almost exclusively in

429


the current-laminated type. The only exception forSites 374 through 377 is the bed at Sample 374-11-1,91-93 cm which contains Chondrites burrows through-out and is not laminated. One might expect that theonset and termination of stagnant conditions would berecorded by burrow structures, either below the sapro-pelic layers, recording the last benthic fauna, or abovethem, recording the recolonization of the area of seafloor after the stagnation event. Evidence of suchoccurrences is present in the cores, and Figure 5 showssome examples. On the other hand, there are sapropelsand sapropelic layers in which the burrowing occurswithin the layer itself. Some, if not all, of this bioturba-tion could result from organisms recolonizing theseabed after stagnation had ceased and burrowingdown into the organic carbon-rich substratum. In somesapropels, the color of the burrow fillings is characteris-tic of the overlying sediment.

Site 378 represents an extreme case where all but 7of the 51 layers found there contained burrows actuallywithin the main part of the bed. Indeed, throughoutthe entire Site 378 sediment column burrow traceswere usually most prevalent in association with thesapropels and sapropelic layers (Figure 6). Chondritesis the dominant form but Planolites is common and afew Zoophycos are also present. Possibly Site 378, withits higher levels of burrowing associated with the sap-ropels, differs from the other eastern Mediterraneansites only because burrowing was restricted to sapro-pelic beds of current-deposited origin and conditionsconducive to this type of deposition were more preva-lent in the Cretan Basin. The lack of lamination inmany beds may be because primary lamination wasdisrupted by later activity by benthic fauna. This couldalso account for the lack of lamination in the bed atSite 374 reported above.

COMPOSITIONSmear-slide analysis was used aboard ship for pre-

liminary determinations of the composition of thesapropels and sapropelic layers. They were generallynannofossil marls which were rich in organic carbon.This richness in organic debris made identification ofother components difficult and estimations of theirabundances are unreliable. As examples, however, thelayers at Site 376 were estimated to contain: 10%-25%fine-grained organic matter, 20%-30% nannofossils,30%-40% clay minerals, 5%-10% pyrite, and 5%-15%foraminifers, while at Site 378 the estimates were thus:25%-40% fine-grained organic matter, 15%-5O% nanno-fossils, 0%-30% clay minerals, plus <1% to 5% fora-minifers and quartz. The apparent absence of pyrite inthe latter may not be accurate inasmuch as darkminerals were frequently undetected in these slides.

Sand fractions, resulting from sieving for microfau-nal analyses, were examined for most of the sapropelsand sapropelic layers at one of the sites; Site 374 (seeTable 1). Unexpectedly, we cannot use these qualita-tive estimates of sand fraction abundance as criteriawith which to make a distinction between the sapropelsbelieved to be of current origin in Core 5 Section 2,

Figure 4. Three types of relationships of the pelagic sapro-pel to the host sediment. (Scale in cm.) (A) Sharp con-tacts above and below (374-3-1, 125-131 cm).

430


Figure 4. (Continued).(B) Sharp contacts with sapropelic laminae a few centi-meters above (374-3-1, 118-124 cm).

Figure 4. (Continued).(C) Sharp contacts with sapropelic laminae 1 cm below(374-5-3, 6-14 cm).

431


1

1 /1

1

i8B*̂

... 1

I

« 1

7

Hi

HIBUUBUB

p

—

i

E-

Pf—hr

y

Figure 5. Three types of current-deposited sapropelic layer.(Scale in cm.) (A) Graded sapropelic layer with lamina-tion and sapropelic burrows above (374-11-2, 125-150cm).

Figure 5. (Continued).(B) Non-graded sapropelic layer with lamination and sap-ropelic burrows above and below (374-11-2, 100-125cm).

432


Figure 5. (Continued).(C) Disturbed, but complexly laminated non-graded sap-ropelic layer. Fine sapropel/marl lamination may be theresult of current reworking of the main sapropel.

and those that were of pelagic type. Also, the only sandsize minerals recorded were mica, quartz, and pyriteand their distribution is insufficient to suggest varia-tions from layer to layer. Shards of fibrous volcanicglass may have some correlative value, however. Theseappear in the layer at Sample 374-3-1, 103-106 cm.Others were found at Sample 376-2-2, 1-30 cm (Figure7), and also as elliptical pockets at Site 378, in its 43-cm-thick layer (Sample 8-2-3, 46 cm). The first two areof similar age (Pleistocene), and the latter is earlyPliocene (MP1-3 of Cita, 1975). After some months ofstorage at the DSDP East Coast Repository, the Leg42A cores were re-examined for additional ash mate-rial. A hitherto, undescribed, non-sapropelic layer,made up almost entirely of ash material was identifiedin Core 1 of Site 376. This had become more visibleafter slight drying on the sediment surface of the splitcore. No additional ash layers were found in the Leg42A cores, but this finding has allowed tentativecorrelation with the Leg 13 Site 125 (see later sec-tions).

For detailed petrography and clay-mineralogy ofindividual sapropel layers, the reader is referred to thecontribution by Sigl et al. (this volume) which makesup the second part of this chapter.

The organic carbon analyses made by the shore-based laboratories show clearly the wide variationsthat exist in organic content between individual sapro-pel layers. Of those analyzed, that at Sample 374-5-3,40-51 cm had the highest percentage: 16.71% organiccarbon by weight. This is the highest value recordedfrom any Mediterranean sapropel to date. It is the onlyone from the Leg 42A cores to exceed 10%, but, on theother hand, a further 12 exceed 5%. One point which isclear from the analyses completed to date is that thepelagic-type layers do not differ from the current-deposited type in percentage organic carbon content.The current-deposited type are not merely "sapropel-stained." All except three of the current-laminatedand/or graded layers are true sapropels (2.0%), andinclude five of those with greater than 5% organiccarbon.

FOSSIL CONTENTAll the sapropels investigated are fossiliferous.

Planktonic foraminifers encountered in the "normal"sediments are found in similar or greater abundance inthe associated sapropels. Giant Orbulinas, easily seenon the surface of a split core, are present in thesapropel at Sample 376-6-4, 55-75 cm, of earliestPliocene age from the Messina Abyssal Plain (Figure8). Sapropels containing a very high organic carboncontent usually do not contain many planktonic fora-minifers. This is the case in the Pleistocene sapropel atSample 374-3-1, 140-143 cm with 8.72% organic car-bon and in the late Pliocene sapropel at Samples 374-5-3, 49-52 cm, with its record value of 16.71% organiccarbon.

Faunal diversities, in contrast, are usually lower inthe sapropels and sapropelic layers than in the hostsediments. Discoasters are enriched in the sapropels ofSite 376, and similar enrichment is coincident with

433


«

HHfei

i

•

- •

378-6-3 378-11-2 378-11-4 378A-3-4 378A-3-5

Figure 6. Several 1.5-meter core sections showing the burrowed sapropels andsapropelic layers at Site 378. (scale in cms)

434


| ,;f•

' /

r.m•

Figure 7. Scanning electron microscope picture showing ashard of fibrous volcanic glass found in the sapropel at376-2-2, 1-20 cm. (× 500).

Figure 8. Scanning electron microscope picture showinggiant Orbulinids found in the sapropel at 376-6-4, 55-75cm. (× 46).

dissolution of the coccoliths in many of the layers ofSite 378 (Müller, this volume). There are also layersobviously enriched in nannofossils with no sign ofdissolution and some with large abundances of singlespecies of coccoliths.

Several sapropels yield only species of planktonicforaminifers known to live in the upper layers of thewater column (epipelagic). This is the case for the

sapropels recorded in the Antalya Basin (376-1-1, 50-80 cm; 376-1-3, 13-16 cm) and in the Cretan Basin(378-6-3, 92-95 cm). Others, however, also containspecies of globorotalids which lived between 100 and500 meters (mesopelagic). This suggests that stagna-tion here did not affect the structure of the upper watercolumn usually associated with the permanent thermo-cline.

Deformed specimens of large Globorotalia truncatu-linoides showing abnormally inflated chambers in theexternal part of the last formed whorl, were found inthe early Pleistocene sapropel at Sample 374-4-2, 56-64 cm of the Messina Abyssal Plain and in the AntalyaBasin sapropels at Samples 376-1-1, 50-80 cm, 376-1-2, 116 cm, and 376-2-3, 4-17 cm (Figures 9 and 10).Similar aberrant, deformed specimens of Globorotaliatruncatulinoides occur in the piston-cored sapropel S7of the late Pleistocene from the Mediterranean ridge(Cita et al., in press). Since Globorotalia truncatuli-noides is a well-known mesopelagic species, the devel-opment of the laterally inflated forms is interpreted asa response to stress conditions which did occur in theupper water mass.

Highly specialized foraminifer assemblages domi-nated by populations of the euryhaline species Globige-rina eggeri and Globigerina bulloides (Parker, 1958;Ryan, 1972), were found in the Messina Abyssal Plainsapropels 374-2-3, 123-127 cm; 374-3-1, 103-106 cm;376-2-2, 20-28 cm; 376-2-3, 4-17 cm; and 376-2-3,121-146 cm (Figure 11). Similar faunal assemblageswere reported in the "Brunhes-Matuyama" sapropelsof DSDP Site 125, by Cita et al.' (1973), and in thepiston-cored sapropels S6, S9, SI 1, and S12, by Cita etal. (in press).

Some sapropels, on the other hand, contain foramin-ifer assemblages indicative of redeposition by currents.For example, one sapropel analyzed from the Torto-nian of Site 375 (Sample 5-3, 21-23 cm) was domi-nated by detrital minerals, but also contained a fairlyrich although poorly diversified assemblage of plank-tonic foraminifers which are consistently small in size.The assemblage is devoid of mesopelagic speciesbut it also contains numerous large spores. Anotherfrom the Messinian of Site 376 (Sample 8-2, 7-9 cm),also contained numerous detrital minerals, togetherwith reworked benthic foraminifers, including the in-nershelf genus Discorbis, and sparse, obviously size-sorted, planktonic foraminifers. Intervals, in which dia-toms completely dominate the assemblage, were foundonly in the Cretan Basin sapropels, 378-7-4, 70-76 cmand 378-8-2, 20-40 cm (Figure 12). Both sapropels arelate Pliocene (MP1-5, foraminifer zone; NN 16 nanno-fossil zone) and contain poor foraminifer assemblageswhich also include species thought to be mesopelagic,such as Globorotalia puncticulata.

Palynological investigations were carried out onthree sapropels of early Pleistocene, late Pliocene, andearly Pliocene age, respectively from the MessinaAbyssal Plain Site 374 (Bertolani Marchetti and Ac-corsi, this volume).

Sapropels are usually very rich in palynomorphs.Strick-Rossignol (1973) demonstrated from the Leg 13

435


100 µm

Figure 9. Scanning electron microscope picture showing three left-coiling specimens of Globorotalia truncatulinoides withdeformed, laterally inflated chambers in the last whorl. Specimens A and B are from 376-1-1 (52 cm). Specimen C is from374-4-2 (57 cm), (scale in microns)

material that two discrete pollen assemblages arerepresented in the eastern Mediterranean sapropels:one of northern origin dominated by arboreous plantsand by Artemisia, the other of Nilotic origin domi-nated by Graminaceae and Cyperaceae. She suggestedthree different processes for the transport, dispersal,and sedimentation of the pollen grains:

1) The formation of the resedimented sapropels asfound at Sites 130 and 131A is associated with the firstprocess. These sapropels yielded the Nilotic pollenassemblage: Graminaceae, Cyperaceae, and Pterido-phytes. These pollen, derived from tropical trees, arebelieved to enter the sea in suspension with the sedi-mentary discharge and behave as a benthic element.

2) The formation of the finely laminated sapro-pels containing predominantly biogenic components isassociated with the second mechanism in which thesapropel results from wind-transport of pollen over ashort distance from the area of vegetation, followed bydispersal by surface currents over a very long distance(pelagic behavior). Note that only northern pollenwere found associated with these finely laminatedsapropels.

3) The third mechanism of transport is invoked toexplain the widespread distribution of certain small-sized non-bladed pollens. This mechanism involveslong-distance wind transport of the pollen in the upperatmosphere.

The investigations by Accorsi (Bertolani Marchetti,and Accorsi, this volume) on the three layers from Leg42A, Site 374, which are of early Pliocene (sapropellayer in 374-9-2), late Pliocene (sapropelic layer in374-5-2), and Pleistocene (sapropelic layer in 374-4CC) age, respectively, recorded the northern pollensonly. The authors suggest that the remarkable differ-

ences found in the percentages of the climaticallysignificant genera Pinus, Cedrus, Quercus, of the familyChenopodiaceae, and of Artemisia, are the result ofclimatic changes through time.

Investigations by Stradner and Bachmann (thisvolume), on the silicoflagellates found in the CretanBasin Site 378, suggest a paleotemperature of 15°-20°C for the sapropel Sample 8-2, 3-46 cm. Theirestimate is based on a Dictyocha/Distephanus ratio of1:1.7 recorded at 21 cm. The ratio between the coldwater indicator Dictyocha fibula and the warm waterindicator Distephanus speculum is considered climati-cally significant by Mandra and Mandra (1972). The tempera-ture estimated for this late Pliocene (MP1-5 Zone) sapropelis lower by approximately 5°C than that estimated for theoverlying "normal" sediment from Section 1.

Pteropods are common in the Pleistocene sapropels(see, also, Cita et al., 1973). Here, deposition in anoxic,stagnant bottom waters may have enhanced preserva-tion by preventing dissolution (Berger and Soutar,1970).

Tests of bottom living faunas are generally absentfrom the sapropels of Sites 374 through 377. Benthicforaminifers, ostracodes, and echinoid spines, whichare consistently present in the intervening "normal"host sediments, are rare or absent altogether in thesapropels and sapropelic layers. An exception is thesapropels at Site 378 in the Cretan Basin which containan epifauna of benthic foraminifers and ostracodeswith densities and diversities comparable to thosefound in the "normal" sediments. These sapropels arethe beds that also contain marked structural evidenceof burrowing. We are uncertain at present if thebenthic tests are from organisms living at the time ofsapropel deposition, or whether these tests are alloch-

436


Figure 10. Comparison between a deformed Globorotalia truncatulinoides (A) from376-1-1 (52 cm) and a normal specimen (B) from 132-3-4 (71 cm), both showing theapertural face, (scale in microns)

thonous and were carried into the sapropels in the gutsof the epifauna which repopulated the seabed with therecommencement of "normal" sedimentation.

STRATIGRAPHIC POSITIONPrecise correlations between a large number of

Mediterranean cores on the basis of sapropels which

were interpreted as "isochronous lithologies" togetherwith others using tephra horizons, have been con-structed by Olausson (1961), Ryan (1972), McCoy(1974), and Sigl and Müller (in press).

Nesteroff (1973) used the term "Upper Sapropel"for the DSDP Leg 13 sapropels referred to the Gephy-rocapsa oceanica Zone (NN 20), and "Lower Sapro-

437

00 CO

ço

Figure 11. Specimens of the euryhaline species Globigerina eggeri from 374-3-1 (103 cm), whose tests are very thin and fragile. The taxon is alsoknown as Neogloboquadrina dutertrei. A = spiral view. B-F = umbilical views, (scale in microns)


a thick bed of coarse ash of shard texture (non-fibrousand without appreciable inclusions) and index ofrefraction similar to a bed detected in Section 4 ofCore 1, at Site 125 (see Chapter 7 of Ryan, Hsü, et al.,1973). The lithologic correlation of these two ash beds,as shown in Figure 13, is further supported by thepresence of the Globigerina eggeri-Globigerina bulloidesassemblage within these levels at the two sites. Bothlines of evidence suggest that the uppermost sapropelsin Cores 1 and 2, of Site 376, are the Brunhes/Matuyama sapropels of Cita et al. (1973), and thebiostratigraphic assignment to the E. huxleyi nannofos-sil zone in the Site Chapter of 376 is questioned.

Of the eight sapropels recovered from the Pleisto-cene sequence at Site 374 in the Messina Abyssal Plain,only the highest two from Sample 2-3, 123 cm (160 msubbottom) and Sample 3-1, 103 cm (210 m subbot-tom) contain the Globigerina eggeri-Globigerina bul-loides assemblage, perhaps placing these layers alsowithin the group of "Brunhes/Matuyama" sapropels.

The sapropels from the upper part of Core 5 of Site376 and from Core 4 of 374 are older. From lithologic

Figure 12. Scanning electron microscope picture showing afaunal assemblage dominated by diatoms found in thesapropel at 378-8-2, 3-46 cm. (x 1000)

pel" for those referred to the Pseudoemiliania lacunosaZone (NN 9). Cita and Ryan (1973) proposed agrouping of the same Leg 13 sapropels arrangedaccording to their position with reference to the paleo-magnetic time-scale (Ryan, 1972; Cita et al., 1973) asfollows: "Brunhes sapropels"; "Brunhes/Matuyamasapropels"; "Matuyama/pre-Jaramillo sapropels";"Matuyama/pre-Olduvai sapropels".

Sapropels, referable to their "Brunhes" group, haveonly been recovered in piston cores. They are 12 in alland have been recently numbered independently byseveral authors (McCoy, 1974; Hieke, in press; Cita etal., in press). The youngest sapropel, S 1, is post-glacial, and has been dated several times by the C14

method; its absolute age is approximately 8000 yr. B.P.Sapropels referable to this group have not been recov-ered in any of the DSDP sites.

Details of the stratigraphic record provided by theLeg 42A coring, with reference to that of Leg 13 andto any available piston cores, are summarized below.

Quaternary

The sequence of sapropels and sapropelic layerswidely recorded from piston cores (e.g., horizons S 1,through S 12 of Cita et al., in press) were not positivelyidentified in any of the Glomar Challenger cores fromeither Leg 13, Site 125, or Leg 42A, Site 376. Theseyoung sapropels extend from isotope stage 1 of Emili-ani (1955, 1966), to stage 13 and are dated as rangingback to approximately 0.44 m.y. according to therevised time scale of Emiliani and Shackleton (1974).These layers have been referred to as the "Brunhessapropels" by Cita et al. (1973).

The most recent stratigraphic horizons in the DSDPcores are those from Core 1 of Site 376 which contain

100 m— X

X

X

X

X

X

X

i?f°ca

Brunhes/MatuyameSapropels

Matuyama/Pre-JaramillcSapropels

Matuyama/Pre-OlduvaiSapropels

\

Figure 13. Tentative correlations of Pleistocene sapropels,between Site 374 (Messina Abyssal Plain) and Sites 125and 376 (Mediterranean Ridge).

439


considerations of textures and colors of the intercalatednannofossil oozes along with bed thickness, this secondand older group of sapropels has the greatest physicalresemblance to the "Matuyama/pre-Jarmillo" sapro-pels of Site 125. However, a problem with this pro-posed correlation is that the Pliocene/Pleistocenebiostratigraphic boundary has been identified in Core 5at both Site 375 and Site 374, where it lies considera-bly below the sapropels of Core 3 at Site 125. There-fore, the sapropels in the lower part of Core 5 at Site376 and in Core 5 of Site 374, might be even older andcould belong to a "Matuyama/pre-Olduvai" group.

Pliocene

Late PlioceneLate Pliocene sapropels were first discovered during

Leg 13 of the Deep Sea Drilling Project in the crestalarea of the Mediterranean Ridge (Site 125A, Core 2-2,approximately 42 m subbottom): they are from astratigraphic interval referred to the Globigerinoidesobliquus extremus Interval-Zone (now MP1-5 Zone ofCita, 1975), and to the Discoaster surculus Zone (NN16).

Sapropels from the same biozone were recoveredduring Leg 42A in the Messina Abyssal Plain, in theAntalya Basin, and in the Cretan Basin.

We propose a correlation which places the jet-blacksapropel found in Sample 5-3, 49-51 cm at Site 374from about 300 meters subbottom as equivalent withthe similarly black sapropel found in Sample 5-4, 105-111 cm, from about 45 meters subbottom at Site 376.In addition to having the same biostratigraphic posi-tion, these sapropels have the highest organic carboncontent of the entire set of eastern Mediterraneansapropels discussed here: the former has a record valueof 16.7% and the latter has a value in excess of 8%.

No sapropels were found from either the latestPliocene (MP1-6 Zone), or from the earliest part of thelate Pliocene (MP1-4 Zone) at Sites 374 and 376,although sapropels at this biostratigraphic positionwere recovered in the Cretan Basin at Site 378.

Early PlioceneSapropels of early Pliocene age were first recovered

during Leg 42A. They are present in all three foramini-fer zones recognized in this stratigraphic interval.

MP1-3 Zone: One sapropel occurs in this biozone atSite 374 (Core 8-3, approximately 360 m subbottom);no sapropels occurred in this biozone at Site 376(condensed section with omission surfaces). Numeroussapropels were found at Site 378 (up to 15 in Core 3A,Zero Section to Section 6, and up to 10 in Core 11,Sections 1 to 4) all lie between 293 and 307 msubbottom.

MP1-1 Zone: Up to six sapropels occur in thisbiozone at Site 374 (Core 11, Sections 1 and 2 atapproximately 380 m subbottom). Note that Core 11 isdiagenetically altered by extensive dolomitization. Onethick sapropel occurs at Site 376 (Core 6, Section 4).This biozone is not represented at Site 378.

Miocene

Late MioceneDiatomaceous sediments rich in organic carbon

encountered within the Messinian evaporitic sequenceat Sites 374, 375, and 376 are, by definition, excludedfrom this account of sapropel stratigraphy. However,four thin sapropelic layers occur as a "multiple" inCore 8, Section 2, of the "hemipelagic" ("lago mare"facies) sequence at Site 376.

Fifteen sapropels and sapropelic layers of Tortonianage were recognized in the spot-cored hemipelagicsequence at Site 375; some of these were of thecurrent-deposited type. Seven layers were analyzed fororganic carbon content and five of these were desig-nated sapropels (organic carbon ranges 2.18% to4.31%). Two layers are rich in detrital minerals andcontain only a minor biogenic contribution whichshows strong evidence of redeposition (375-5-3, 21-23cm and 375-6, CC). A third however, from 375-6, CC,is predominantly biogenic and framboidal pyrite is itsonly obvious mineral.

Middle MioceneSerravallian sapropels, so far the oldest found be-

neath the Mediterranean deep sea floor, were identi-fied at Site 375 on the Florence Rise. They occur inCores 8 and 9, and to date all of those analyzed aresapropels containing from 2.8% to 5.06% organic car-bon.

DISCUSSIONWe consider, at this time, only the Pliocene to

Quaternary eastern Mediterranean sapropel stratigra-phy and its paleoenvironmental significance becauseour record becomes distinctly fragmentary even insediments older than Pleistocene. The interval of timesince the end of the Messinian "salinity crisis" is dividedinto four segments whose sediments were depositedunder different conditions of ocean circulation as infer-red from their sedimentary and fossil peculiarities.During three of these four intervals we believe that theeastern Mediterranean underwent periods of intermit-tent stagnation, whereas during the fourth interval itwas permanently ventilated.

Late Pleistocene IntervalThe "Brunhes sapropels" and "Brunhes/Matuyama

sapropels" generally have the following features:1) They are closely linked to climatic changes

(climatically modulated). Many of them occur duringglobal warming trends as deduced from analyses of themicrofaunal assemblages and from the studies of thevariation in <518/O of foraminifers from the Caribbeanand Pacific Ocean (Cita et al., in press).

2) They occur in the Mediterranean at levelscharacterized by a significant lightening of the oxygenisotopic composition (up to 4.8%o) relative to theintervening periods of effective ventilation and/orenhanced precipitation.

440


3) They bear (although, not always) a euryhalinefaunal assemblage dominated by Globigerina eggeriand Globigerina bulloides indicating the existence of anupper surface layer of seawater diluted by the run-offof melting glaciers.

4) They are distributed on both structurally ele-vated regions, such as the Mediterranean Ridge(DSDP Sites 125, 130, and 376), and in the deepestbasins (Sites 127, 128, and 374); consequently, the"oxygen minimum" model (Arthur, 1976) cannot beapplied. Some of the sapropels in piston cores werefound in water depths as shallow as 700 meters(McCoy, 1974) along the upper continental slope.The "Brunhes" and "Brunhes/Matuyama sapropels"accumulated during the "glacial Pleistocene" (sensuSelli, 1967; Cita el al., 1973) and corresponds to thewell-known cyclical glacial expansions in the Alpine-Mediterranean area (Würm, Riss, Mindel, etc.).

Early Pleistocene and Late Pliocene Interval

The so-called "Matuyama/pre-Jaramillop" and "Ma-tuyama/pre-Olduvai" sapropels are only known fromthe cores of Legs 13 and 42A. Their main characteris-tics are as follows:

1) They also are climatically modulated and occurduring, or near warming trends as recognized in thepopulation of planktonic foraminifers (Cita et al.,1973; Ciaranfi and Cita, 1973). The climatic fluctua-tions are distinct but not as large as in the "glacialPleistocene." Warm peaks are recorded, but coldpeaks are not comparable in magnitude to thoserecorded in the later part of the Pleistocene.

2) They do not yield specialized euryhaline as-semblages, suggesting that no dilution of the superficialwater masses occurred in times before the onset of theAlpine glaciations. If glacio-eustatic changes in the sealevel occurred (as postulated by Cita and Ryan, 1973),they were controlled by the Arctic glaciation at highlatitudes and did not cause a local dilution of the watermasses.

3) They are very rich in organic carbon. Thehighest values ever recorded from sapropels of any ageare to our knowledge from the late Pliocene (MP1-5foraminifers zone).

4) They are present both on rises and in basinsand the "oxygen minimum" model (Arthur, 1976) asis the case for the younger sapropels, cannot be appliedto them.

5) High abundances of siliceous plankton occa-sionally occur in some of these sapropels, as in Cores 6and 8 of Site 378 in the Cretan Basin.

Early Pliocene Interval

Sapropels of early Pliocene age (the Gilbert sapro-pels) are different from those previously discussed inseveral aspects:

1) They are not climatically modulated.2) They are found only in basins and are absent

on ridges.3) Their content of organic carbon is consistently

low (at the limit of the definition of a sapropel).

4) Stagnation is accompanied by a temporarydestruction of the epifauna and infauna, as is indicatedby the early Pliocene sapropels of Sites 374 and 376,but the benthic fauna is detected throughout the Plio-Quaternary section in the Cretan Basin, Site 378.

Mid-Pliocene Interval Without Sapropels

A time of relatively consistent ventilation withoutthe formation of strata rich in organic carbon occursthroughout the eastern Mediterranean, except in thebasins north of Crete, from the early/late Plioceneboundary (ca. 3.3 m.y.B.P.) to the extinction level ofthe genus Sphaeroidinellopsis (ca. 3.0 m.y.B.P.). Thisinterval is entirely within the Gauss Epoch, where thebiostratigraphic control is good and includes both thecontinuously cored sequences at Sites 125 and 376 andthe spot-cored sequence at Site 374. Sapropels did,however, form at Site 378.

The Gauss Epoch coincides with the initiation ofNorthern Hemisphere Arctic glaciation (Laughton,Berggren, et al., 1972) as recorded in sediments of thenorthern Atlantic by the deposition of ice-rafted mate-rial directly on top of deep-sea sediments yieldingsubtropical faunal assemblages. A cool episode("brown climatic eposide" of Ciaranfi and Cita, 1973)was recorded both in the eastern and in the westernMediterranean. This episode was correlated (Cita andRyan, 1973) with the Aquatraversan erosional phaseidentified on the Tyrrhenian coast of central Italy(Ambrosetti and Bonadonna, 1967).

The maintenance of well-ventilated conditionsthroughout most of the Mediterranean during theGauss Epoch, is attributed principally to the vigor thatcooling brings to bottom-water generation and circula-tion.

Comments on the Causes and Durations of StagnantEpisodes

We believe the principal cause of stagnant condi-tions was the development of a marked density strati-fication which inhibited sinking of the surface waterinto deeper levels. This prevented the renewal ofdissolved oxygen which was being consumed by thedecay and utilization of organic matter. The densitystratification may have been in the form of a strongtemporary thermocline or simply a lower salinity sur-face-water layer floating on higher salinity intermediateand deep-water layers as was suggested by Olausson(1961). The development of stratification could havebeen caused by preferentially greater (and more rapid)heating of the surface-ocean layer than the deep watermasses during an episode of regional (or global)climatic warming. In addition, stratification could havebeen initiated by an influx of fresh water to semi-isolated basins, either as excess rainfall (pluvials),glacial melting, or a combination of the two.

The duration of deposition of the individual sapro-pels in Mediterranean cores ranged from less than1000 years to generally no longer than 10,000 years.Ten thousand years is about the time it would take toheat a 1000-meter-thick bottom-water layer by 3°C

441


solely as a result of the geothermal flux through theocean floor. Large amounts of additional heat wouldalso be supplied by downward advection and conduc-tion from the ocean surface layers. Hence, stratificationwould be a non-equilibrium transient situation thatcould only be maintained for long periods of time by acontinuous applied perturbation (i.e., the steady inflowand outflow of less dense surface water such as occursin the Black Sea).

SUMMARYOf the 150 or more dark colored organic-carbon-

rich layers recovered during Leg 42A, about two-thirdsare sapropels as defined in this paper. The majority ofthese are pelagic but a significant number were prob-ably current-deposited on the basis of analysis of theirsedimentary structures, detrital mineralogy, and faunalcontent. These may represent brief incursions of turbid-ity currents into anaerobic basins analogous to those ofthe present Californian borderland (Berger and Soutar,1970).

All the sapropels are fossiliferous, but their faunaldiversities are usually lower than in the host sediments.Mesopelagic and epipelagic faunas are present but, asexpected, benthic faunas are absent in most of thelayers. Aberrant specimens of planktonic foraminifersand large concentrations of diatoms and coccolithstestify to conditions which were adverse to somespecies but were very favorable (presumably partlydue to lack of predators or competition) to others.

We plan to further investigate whether burrowtraces and the remains of benthic faunas found withinsome sapropels resulted from the activity of burrowerswho recolonized the sediment surface after a period ofstagnation or, whether a rich epifauna and infauna did,in fact, live on the sea floor during the time of stagna-tion. These burrow traces and faunas are characteristicof early Pliocene sapropels but occur throughout thePliocene and Quaternary at Site 378 in the CretanBasin. We may need to consider a different mechanismfor sapropel formation during the early Pliocene, onewhich has more recent analogies in the sedimentsequences in the basins north of Crete.

Whereas previous studies of piston cores have al-lowed the correlation of Quaternary sapropels overwide areas of the eastern Mediterranean, the drillingby Glomar Challenger on Legs 13 and 42A, althoughproving the existence of sapropels back to the middleMiocene, provides an incomplete pre-Quaternaryrecord. Despite this, we are able to tentatively correlatethe Pliocene and Quaternary sapropel sequences atthree of the eastern Mediterranean sites (Figure 13).Such correlations confirm that periods of widespreadstagnation took place at least as far back as thelowermost Pliocene, proving that their causes were notmerely linked to glacial phenomena (Olausson, 1961).Also, the identification of sapropels in the Serravail-lian, at one site (375) suggests that the events haveoccurred repeatedly over the past 11 million years ormore. Furthermore, a sapropel was encountered in thePleistocene of the Tyrrhenian Sea site (373A) which

shows that such stagnation was not entirely confined tothe eastern Mediterranean.

Sapropels, considered to be time-synchronous, arefound to have been deposited in a wide range of waterdepths. Consequently, they could not have beenformed by the presence of an "oxygen minimum"layer as a function of water depth.

We grouped the Pliocene and Pleistocene sapropelsinto three categories on the basis of time intervals. Weinfer that each type of sapropel was developed underdifferent conditions of ocean circulation. Their salientfeatures are summarized in Table 6.

We conclude that basin-wide stagnation within theMediterranean results from density stratification of thewater masses. This is a transient condition that hasdeveloped repeatedly since the middle Miocene whenthe Mediterranean took on its present form with onlyone outlet to the world ocean (Hsü et al, this volume),and is the result of a number of interacting factors. Themain influence on ocean circulation in the late Pleisto-cene that caused stratification was the repeated AlpineMediterranean glacial expansions. High latitude Arcticglacial expansions were possibly a prime cause of theearly Pleistocene/late Pliocene density stratification.However we are presently unable to identify anyoutstanding factors which could have caused earlierperiods of stratification.

TABLE 6Salient Features of the Pliocene and Quaternary

Mediterranean Sapropels

TimeInterval Distribution

FaunalAssemblages

Relationshipto Climate

Trends

OceanicCirculation

Influenced by

LatePleistocene

EarlyPleistocene/late Pliocene

EarlyPliocene

Both onridges andin basins

Both onridges andin basins

In basinsonly

Euryhahneplanktonicfaunas; ben-thonic faunasabsent

Planktonic faunasnot euryhaline;benthic faunasabsent

Planktonic faunasnot euryhaline;Benthic epifaunasand infaunas mayhave survived

On warmingtrends; coldpeaks distinct

On or nearwarmingtrends; coldpeaks lessdistinct

Notclimaticallymodulated

Alpine-Mediterraneanglacialexpansions

Arctic highlatitude glacialexpansions?

1

ACKNOWLEDGMENTSWe thank Dr. Steve Calvert of the Institute of Oceano-

graphic Sciences, Wormley, for his critical review of thiscontribution and for useful suggestions for improvements inthe manuscript. Financial support for WBFR was providedby the Division of Ocean Sciences of the U.S. NationalScience Foundation Grant NSF-OCE-76-0237.

REFERENCESAmbrosetti, P. and Bonadona, F. P., 1967. Revisiorie dei

data sul Plio-Pleistocene di Roma: Atti Soc. Gioenia,Catania, v. 18, p. 33.

Arthur, M. A., 1976. The Oxygen Minimum: Expansion,Intensification and Relation to Climate (Abstract): Joint

442


International Oceanographic Assembly, Edinburgh, Scot-land, September 1976.

Berger, W. H. and Soutar, A., 1970. Preservation of planktonshells in an anaerobic basin off California: Geol. Soc. Am.Bull., v. 81, p. 275-282.

Ciaranfi, N. and Cita, M. B., 1973. Paleontological evidenceof changes in the Pliocene climates In Ryan, W. B. F.,Hsü, K. J., et al., Initial Reports of the Deep Sea DrillingProject, Volume 13: Washington; (U.S. GovernmentPrinting Office), p. 1387-1399.

Cita, M. B., 1975. Planktonic foraminiferal zonation of theMediterranean Pliocene deep sea record; a revision; Riv.Ital. Paleont. Strat, v. 81, p. 527-544.

Cita, M. B. and Ryan, W. B. F., 1973. Time scale and generalsynthesis In Ryan, W. B. F., Hsü, K. J., et al., InitialReports of the Deep Sea Drilling Project, Volume 13:Washington; (U.S. Government Printing Office), p. 1405-1415.

Cita, M. B., Chierichi, M. A., Cliampo, G., Moncharmont Z.,M., D'Onofrio, S., Ryan, W. B. F., and Scorziello, R.,1973. The Quaternary Record of the Tyrrhenian andIonian Basins of the Mediterranean Sea In Ryan, W. B.F., Hsü, K. J., et al., Initial Reports of the Deep SeaDrilling Project, Volume 13; Washington; (U.S. Govern-ment Printing Office), p. 1263-1339.

Cita, M. B., Vergnaud-Grazzini, C, Robert, C, Chamley, H.,Ciaranfi, N. and D'Onofrio, S., in press. Paleoclimaticrecord of a long deep sea core from the Eastern Mediter-ranean: Quaternary Res.

Emiliani, C, 1955. Pleistocene temperature variations in theMediterranean; Quaternaria, v. 3, p. 109.

Emiliani, C, 1966. Paleotemperature analysis of Caribbeancores P. 6304-8 and P. 6304-9, and a generalized temper-ature curve for the last 425,000 years: J. Geol., v. 74, p.233.

Emiliani, C. and Shackleton, S., 1974. The Brunhes epochpaleotemperature and geochronology; Science, v. 183, p.511-514.

Hieke, W., 1976. Problems of Eastern Mediterranean lateQuaternary stratigraphy—a critical evaluation of literature:"Meteor" Forsch. Ergneb., v. 24, p. 68-88.

Kullenberg, B., 1952. On the salinity of water contained inmarine sediments; Medd. Oceanographiska Inst. Gote-borg, 21.

Laughton, A. S., Berggren, W. A., et al., 1972. Initial Reportsof the Deep Sea Drilling Project, Volume 12: Washington(U.S. Government Printing Office).

Mandra, Y. T. and Mandra, H., 1972. Paleoecology andTaxonomy of Silicoflagellates from an Upper MioceneDiatomite near San Felipe, Baja California, Mexico.Occos. Pep. Calif. Acad. Sci., v. 99, p. 2-35.

McCoy, F. W, Jr., 1974. Late Quaternary sedimentation inthe Eastern Mediterranean Sea: Unpublished Ph.D. thesis,Harvard University, p. 1-132.

Nesteroff, W. D., 1973. Petrography and Mineralogy ofsapropels. In Ryan, W. B. F., Hsü, K. J., et al., InitialReports of the Deep Sea Drilling Project; Volume 13:Washington; (U.S. Government Printing Office), p. 713-720.

Olausson, E., 1961. Description of sediment from the Medi-terranean and Red Sea; Rept. Swedish Deep-Sea Exped.1947-1948, v. 8, p. 337-391.

Parker, F. L., 1958. Eastern Mediterranean foraminifers;Rept. Swedish Deep-Sea Exped. 1947-1948, v. 8, p. 219-283.

Ryan, W. B. F., 1972. Stratigraphy of Late Quaternarysediments in the Eastern Mediterranean. In Stanley, D. J.,(Ed.), The Mediterranean Sea Strasbourg, Va. (Dowden,Hutchinson and Ross), p. 765.

Ryan, W. B. F., Hsü, K. J., et al., 1973. Initial Reports of theDeep Sea Drilling Project, Volume 13; Washington (U.S.Government Printing Office).

Selli, R., 1967. The Pliocene-Pleistocene boundary in Italianmarine sections and its relationship to continental stratig-raphies. In Progress in Oceanography; vol. 4, New York(Pergamon Press), p. 67.

Sigl, W. and Müller, J., 1976. Identification and correlationof stagnation layers in cores from the Eastern Mediterra-nean: Proc. Verb. Reun., Monaco.

Strick-Rossignol, M., 1973. Pollen analysis of some sapropellayers from the deep sea floor of the Eastern Mediterra-nean. In Ryan, W. B. F., Hsü, K. J., et al., Initial Reportsof the Deep Sea Drilling Project; Volume 13: Washing-ton, (U.S. Government Printing Office), p. 631-643.

Van Stratten, L.M.J.V., 1972. Holocene Stage of OxygenDepletion in Deep Waters of the Adriatic Sea. In Stanley,D.J., (Ed.), The Mediterranean Sea Strasbourg, (Dowden,Hutchinson and Ross), p. 631.

443

Stratigraphy of eastern

Documents

Transcript of Stratigraphy of eastern