Stable isotope studies of fossiliferous Palaeogene sequence of Kutch, Western India: palaeoenvironmental implications

E L S E V I E R Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77

PAI O

Stable isotope studies of fossiliferous Palaeogene sequence of Kutch, Western India: palaeoenvironmental implications

A. Sarkar a, A.K. Ray b, S.K. Bhattacharya ° " Department of Applied Geology, Indian School o f Mines, Dhanbad826 004, India

b Department o f Geology, Presidency College, Calcutta 700 073, India c Physical Research Laboratory, Ahmedabad 380 009, India

Received 25 January 1994; revised and accepted 3 May 1995

Abstract

Stable isotope analyses have been carried out on rocks and fossils of the Palaeogene carbonate-clastic sequence of the Kutch area in Western India. Isotopic, lithological and palaeontological data indicate that Tertiary sedimentation started in this region in the early middle Eocene (Lutetian) with the development of semi-locked basins (lagoons) in the depressions of pre-existing Deccan Trap basements. The larger benthic foraminifera Assilina was the dominant microbiotic species capable of surviving in this highly stressed environment. The basins were later cut off by the oscillatory withdrawal of the sea, which trapped large amounts of organic matter. Degradation of this organic matter exhausted the dissolved oxygen and biogenic methane was produced in anaerobic conditions. This methane, decomposed by sulphate reducing bacteria, produced some bioclastic limestones with extremely depleted carbon isotope values. The oxygen isotope values of the limestones suggest a significant contribution of freshwater into these lagoons. As a result, terrigenous clastics were formed; associated lignite beds developed in a warm, humid climate. Probably the area was part of a global warm humid belt which extended through Africa and Europe as far as 40°N during the Lutetian. A renewed marine transgression is recorded by the formation of aerated tidal lagoons having a more seaward extensions. The process culminated in a relatively open shallow marine condition when late middle Eocene and Oligocene bioclastic limestones (with a vast array of tropical chlorozoan assemblages) were deposited.

1. Introduction

The Kutch onshore basin in Western India encloses one of the few late Mesozoic-Cenozoic sequences of India. The basin occurs in an oil-rich belt which includes the Bombay offshore, Cambay and Saurashtra basins in the south and extends through Rajasthan in India and the lower Indus to the Salt Range and Baluchistan in Pakistan, in the north and northwest respectively (Eames, 1952; Nagappa, 1959; Bhatia, 1985). However, there is no consensus about the basic stratigraphic frame-

0031-0182/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0031-0182(94)00047-X

work of the Kutch region, especially that of the Tertiary. Neither have the palaeoenvironmental conditions under which these sediments were deposited been fully understood (Wynne, 1872; Biswas and Raju, 1973; Ray et al., 1984; Biswas, 1986). Such understanding, particularly of a tropi- cal region, would provide an important input to the reconstruction of global palaeoclimatic models of the early Palaeogene time.

Here we report analyses of oxygen and carbon isotopes in the Palaeogene carbonates of the Kutch Tertiary sequence with the aim of understanding

66 A. Sarkar et aL /Palaeogeography, Pa[aeoclimatology, Palaeoecology 121 (1996) 65 77

the depositional setting of these post-Deccan Trap sediments. In addition, some aspects of faunal assemblages are compared vis-a-vis isotope data and the stratigraphy is refined.

2 . G e o l o g i c a l s e t t i n g

Tertiary sediments of the Kutch onshore basin are exposed in a westerly convex belt covering ~400 km 2 parallel to the present-day coastline

(Fig. 1). The various lithological units, comprising mainly fossiliferous limestones, marls, clastics and evaporites, unconformably overlie either the Deccan Traps (65 m.y. old, Pande et al., 1988) or the older Jurassic sediments. The Pa|aeogene is classified here into three formations (Fig. 2a,b): the clastic evaporite marl (CEM), the white lime- stone marl (WLM) and the coral limestone marl elastic (CLMC). The CEM formation varies litho- logically and the constituent units, equivalent to member-rank, are discontinuous and lenticular.

68 ° 35' 68 ° 40' 6B ° 45'

L T U V O r . - ' ) ,¢%f'+ + ~ o

I ! ." 4 - / " -t=. ~. TM + ' " " ~ - ' - - - - ~ _ . ~ . . ; . ~ . _ + ' * t t _ , . * / ,v v ! N D E X

~_~~~~2t2,~, , , ~,Q%.,V~BARA_ND A ~ CLASTIC EVAPORITE MARL ( C E M ,

~ ~ - I - -t- ~-///zX;+.,." V VI-I V.V ~ U P P E R ASSILINA MARL(UAM){CEM)

' i ' "t'- " = 4" 1 ' f - i - _ L - ~ ~ G Y P S I F E R O U S SHALE (}NCEM)

, I I . ' ' - " .+ + , r~-~DECCAN TRAP BASALT 0

2? . . . . . . . . ; + " ' ' -23 30 '~ - 30'

68 ° 35 ~ 68 ° 40 68 ° 45 '

Fig. 1. Geological map of the Kutch showing the early Palaeogene outcrops.

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77 67

, I , , I I I I I I - 5 0

I I I I I I w I i I ] I i I i ] I I i

z I i I i I i I 11 i1 40 i z < t~ / / / / / / I

W -- i / / / / / / /

w I I I I l,,J-30

I ~ - - . t - - . . - . . 1 I I

',1 ~ : ' ~ I I l I I ' 6 0 " , , , , I ~ . . . . 16 ~ . ' ~ . - % - . ~ r - r - . ° . - o . o r - . - L , ,

• G C C C C " 17

o , / / / / / ~-! / / / / I i . ~

ILl ~ j ~ / s ~ _ _ ~ . f ~ + + + + ~ a r ° ,

! f V

v -10 w

\ :EF- " v -~ -Tv o m

0chreous marl i, ~ Green mar l l ~:~] Nummulitic Nummutitic

marl u ~ marl (NOM) ~ White coral . ~ U p . A s s i l i n a

l imestone ~ marl (UAM) Conglomerate _~' ~ M u d s t o n e Whi te /buf f 7 boulder bed

F771 Gypseous ~r l imestone ~r ~ s h a l e / Bu f f / b rown _ s i l ts tone w mar l Shale- (-' Conglomerate Jl ~ l ign i te

L r. Assl lina mar [ (LAM)

- - ~ Basal r conglomerate

Z n,~

W r Y a ~ - r

123

i 15

I I

1. Assi l ina regularia sijuensis 2. Ass i l ina regular ia 3. Nummul i tes obtusus 4. Nummuli tes beaumonti 5. Nummul i tes pengaronensis 6. Nummutites macuta tus 7. Nummulites skamineus 8. Nummul i tes t i t ch te l i 9. Discocyctina d ispansa

10. Discocycl ina sowerby i 11. Discocycl ina augustae 12. Discocycl ina se l la 13. Asterocycl ina a l l i cos ta ta 14. Alveol ina e l l i p t i ca 15. Ass i l ina exponens 16. Eulepidlna d i l a t a t a 17. Heterostegina koh t i i (~)

Fig. 2. (a) Generalised lithostratigraphic column of Kutch Palaeogene and (b) Range chart of different larger benthic foraminiferal species.

68 A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65- 77

Among them, the upper Assilina marl (UAM) , occurring at the top of the CEM formation, has more extensive outcrops than the lower Assilina marl (LAM), which is exposed only in the river- cutting sections. Both the units have abundant tests of benthic foraminifera Assilina as framework grains; the grain to matrix ratio, however, varies widely, from matrix-supported floatstone to occa- sionally mud-supported wackestone to grain- supported packstone. A variety of gypsiferous and non-gypsiferous clastics and carbonaceous beds (lignitic) exists between the LAM and UAM. While the WLM is rather monotonous in lithology (unlike CEM), the CLMC formation has a number of varieties e.g. bioclastic limestone, coral lime- stone, chalky limestone, ochreous marl with a few fossiliferous conglomerate, sandstone and silt- stone beds.

3. Fossils and stratigraphy

The entire sequence of the Kutch Palaeogene is richly fossiliferous. Larger calcareous benthic fora- minifera make the most important group, genera and species of which play significant roles in the bio- and/or chronostratigraphy. By virtue of their rock forming abundance many of these genera and species are also important in characterising litho- stratigraphic units. Thus Assilina regularia and Assilina regularia var. sijuensis characterise the LAM and UAM members, Nummulites obtusus the NoM member in the CEM formation; typical late Lutetian species of Discocyclina, NummuIites, Assilina, Alveolina and Dictyoconoides characterise the WLM formation and Nummulites intermedius- J:ichteli and/or Lepidocyclina (Eulepidina) dilatata characterise different parts of the CLMC forma- tion. Their bio- and chronostratigraphic signifi- cance has been pointed out by different authors (Raju, 1969; Samanta, 1969, 1981). Fig. 2a,b pre- sent the Kutch Palaeogene lithostratigraphic column (vide Ray, 1987) and range chart of different major larger benthic foraminiferal species.

Another interesting point about this group of microfossils is relevant in the present context. Both the LAM and UAM members of the CEM forma- tion are characterised by Assilina regularia and

Assilina regularia var. so'uensis. These were earlier identified as some Lower Eocene species of the genus Assilina, viz. granulosa, spinosa etc. (Mohan and Gupta, 1968; Raju, 1971). Subsequent work (Ray et al., 1984; Ray, 1987) showed that the present morphology of these Assilina tests is largely the results of differential micritisation during dia- genesis. This process acting on the outermost surface of the tests wears down the involute spiral lamina or granules; at different stages of this breakdown, the spiral laminae tend to show a semi-involute to even evolute nature of overlapping and granules may or may not be preserved at all (Fig. 3a-c). The morphology of these diageneti- cally affected tests conforms with the "Lower Eocene" species earlier identified and the real morphology is obtained only when traced back to the diagenetically unaffected or least affected tests. This helps to identify the species as A. regularia and A. regularia var. sijuensis of the Middle Eocene. This observation on the taphonomic changes affecting the systematics of microfossils has its bearing on the stratigraphy too. A. regularia and A. regularia var. sijuensis are the earliest fossils in the sequence immediately overlying the Deccan Trap basements; they thus help to establish the age of initiation of Palaeogene sedimentation as being in the Middle Eocene.

It is to be noted that both the LAM and UAM are characterised by a virtually monospecific assemblage of Assilina regularia (and var. sijuensis) in which the species and its variety are represented by the proliferation of individuals. Macrobiota are completely absent. This attests to the successful adaptation of these forms to this particular envi- ronment, without any significant competition from other species or genera of larger benthic foramini- fera. The environment was, thus, stressed because it was restricted, which weeded out other forms but Assilina regularia and its variety sijuensis, which could overcome the stress, flourished numer- ically without much competition. To some extent stress left its imprint on their growth as well. Size measurements (both the thickness and diameter) of the Assilina (particularly A. regularia and A. subassamica) tests show that they are relatively smaller than the same species (and variety) from the Siju limestones of the middle Eocene tethyan

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77 69

(a)

(b)

sequence of Assam which was deposited in an open marine condition with enormous generic and specific diversity. Table 1 shows the comparison of the maximum diameter and thickness of the Assilina species and its variety from two diverse middle Eocene geographical localities like Kutch and Assam.

Other microfossils include less common smaller benthic and planktonic foraminifera and rarer calcareous algae and pollen spores. Macrobiota like bivalves and gastropods are mostly found in the W L M and C L M C formations and are repre- sented by genera like Strombus and Conus in the W L M formation and Pecten and Ostrea in both the W L M and C L M C formations. Echinoids are represented by Echinolampas and Clypeaster in the W L M and C L M C formations and Schizaster and Breynia in the C L M C formation. Corals (Anthozoa) form another significant group, partic- ularly in the C L M C formation; Actinastrea, Astrocoenia etc. are the more common genera. Palaeocarpilius macrochelus, a well known crab, shark teeth and bone fragments are among the other fossils found in one or the other formation. The macro- and microbiota together broadly sug- gest a shallow marine, upper neritic to littoral environment for the Kutch Palaeogene.

(c)

Fig. 3. Taphonomic changes in Kutch Assilina, x 20. (a) Least altered test from LAM. (b) Tests with partly worn out granules, LAM; note the central port ion rendering biconcavity to an originally parallel sided test. (c) Completely altered test; black colour is due to ferrugination (pyrite).

Table 1 Compar ison of max imum thickness and diameter of benthic foraminifera Assilina and its variety from middle Eocene of Kutch and Assam. Note relatively smaller sizes of Kutch specimens

Assilina Assilina Assilina Assilina regularia regularia subassamica subassamica

var. var. sijuensis sijuensis

Diameter a 17 15 5.8 5.6 (mm) Diameter b 19.6 14.9 10 5.1 (mm) Thickness a 1.8 1.6 1.2 0.8 (mm) Thickness b 2.3 2.2 1.4 0.9 (mm)

aKutch Assilina, from Ray (1987). bAssam Assilina, f rom Samanta (1962).

70 A. Sarkar et al . /Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77

Table 2 O x y g e n a n d c a r b o n i so tope va lues o f K u t c h l imes tones

Sample no. R o c k type Loca l i t y S t r a t i g r a p h i c

un i t

~18 O

P D B

(%0)

/i~3 C

P D B (%,,)

A-82 O c h r e o u s n u m m u l i t i c J a d v a C L M C

m a r l P -24 J a m a n w a l a

2 - 5 b - T C Ber Mot i

27-91 Whi t e l imes tone W a i o r

W-91 K h a r i

7 -TC N u m m u l i t i c m a r l W a i o r A - 1 4 4 a W h i t e Assil ina L i m e s t o n e B a r a n d a W L M

C - 1 5 A " DS-25 " G o d h a t a d

5 - A R Sehe

A-47 J a d v a

A-70 3 2 - T C Buff l imes tone W a g a p a d h a r

A - 7 1 - P G W h i t e l imes tone J a d v a

A- 105 Buf f l imes tone "

A - 3 2 ( 2 ) . . . . A - 3 2 ( 2 ) D i s c o c y c l i n a test " K-1 O c h r e o u s Assilina K a k d i r iver sec t ion U A M

l imes tone

K-2 . . . . K-2 Smal l b iva lve "'

K-3 O c h r e o u s Assil ina ~'

l imes tone

K-4 . . . .

K -12 " ""

K-13 . . . . DS-61 " N a r e d o

D S - 6 / 1 A "'

DS_6/1B . . . .

DS-18 . . . .

D S - 4 5 A . . . . . .

D S - 7 0 A . . . . . .

D S - 7 0 B . . . . . .

A- 192A " N a r e d i "

C - 5 7 A . . . . ""

C-57B . . . . . .

P -20 /45 . . . . . .

P U - 9 . . . . . . U 1 /84A " K a k d i r iver sec t ion "

U 3 / 8 4 . . . . . . U 1 / ' 8 4 B . . . . . . K - 5 G r e y to " L A M

green i sh m a r l y l imes tone wi th Assil ina regularia

K - 6 " " K-7 . . . .

K-8 " K - 9 "

K - 1 0 . . . .

- - 3 . 6

2.68

- - 2 . 1 2

2.38

- - 3 . 0 0

- 2 . 9 9

3.7

4.3

- 1 . 8

- 2 . 2 3

- -3 .03

3.6

- 3.69

3.48

- 3 . 6 5

- 3 . 7 - 3 . 2

4.8

4.7

4.6 5.4

- 6 . 2

- -5 .3

- -4 .8

5.7

- -3 .1

- 2 . 3

- 3 . 2

- 3 . 1

- 1 . 5

- 1 . 3

- -3 .5

- -4 .5

- 5 . 7

3.9

1.8

- 3 . 1

+ 0 . 4

3.8 - - 4 . 6

5.8 - - 5 . 9

- -4 .3 5.9

6.3

0.8

- 1 . 2 8

+ 0 . 6 6

- -1 .31

0.59

0.38

- -0 .5

+ 1 . 5

+ 1 . 2

+ 0 . 4 8

+ 1 . 1 3

- 0 . 7

2.95

+O.83

+ 0 . 0 1

+ 0 . 4

+ 1.6

- 5 . 8

- 1 1 . 7

- 4 . 0

- 3 . 7

1.9

+ / / . 8

+ 0 . 6

- 4 . 7

- 7 . 9 - 7 . 8

- 7 . 0

13.7

1.4

1.7

5.3

2.6

1.3

- 7 . 7

- 0 . 9

- 0 . 2

- 1.0

- 0.7

+ 1 . 6

0.5

0.7 - 2 7 . 8

+ 3 . 1

2.8

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77

Table 2 (continued)

71

Sample no. Rock type Locality Stratigraphic 6~80 unit PDB

(%0)

~13 C

PDB (%o)

K-11 " -5.0 K - 1 4 . . . . . . -5.3 K - 1 5 . . . . -6.9 K - 1 6 . . . . . . -6.2 LDS-28A/1 " Godhatad " -4.4 LDS-28A/2 . . . . -4.6 LDS-28A/3 . . . . . . -4.5 LDS-28B . . . . --4.9 LDS-28C/1 . . . . . . -4.4 I ~DS-28C/2 . . . . - 4.4 LDS-28C/3 . . . . -4.7

-4.5 -3.0 -2.6 -3.0

--15.8 -29.9 --19.8 -21.9 -16.8

22.1 14.7

4. Stable isotopes

Most of the stable isotope analyses have been carried out on bulk samples from the LAM and UAM units of the CEM formation as well as a few measurements of the rocks of the WLM and CLMC formation. Due to the hard and compact nature of these carbonates, it was not possible to analyse the various components separately. In two samples from the CEM ( U A M ) and WLM forma- tions, we separated a bivalve and a benthic fora- minifera Discocyc l ina respectively and analysed them.

4.1. E x p e r i m e n t a l procedures

For the bulk analyses, the sample was cleaned under deionised water in an ultrasonic bath, to remove extraneous fine grained material, and then dried in an oven. Subsequently the sample was powdered, homogenised and again dried under vacuum at 150°C. For bivalve and foraminifera, the outer layer was scraped off and powdered samples were collected by microdrills. About 0.5 mg of sample was reacted with 100% ortho- phosphoric acid (H3PO4) under vacuum at 50°C in an on-line CO2 extraction system. The CO2 gas thus released was cleansed of water vapour through a - 1 0 0 ° C alcohol trap and taken into a VG 602D Micromass mass spectrometer to mea- sure the oxygen and carbon isotope ratios. Results are expressed in conventional 6-notation with

respect to the PDB standard. The experimental precision for both 6~80 and ~3C, based on repli- cate measurements of internal laboratory stan- dards Z-Carrara (courtesy Dr. N.J. Shackleton) and NBS-20 is +0.1%o (Sarkar et al., 1990).

5. Results

6180 and 613C values of the Kutch Palaeogene carbonates are given in Table 2 along with the rock types analysed, their corresponding strati- graphic units and the localities from which they have been collected. The data are plotted in a •180-613C scatter diagram in Fig. 4. The total spreads of c5~80 and c513C in these samples are -6.9%0 to +0.4%0 and -29.4%0 to +3.1%o, respectively. The corresponding spreads in the LAM unit are - 6 . 9 to - 4 . 3 and --29.4 to +3.1%o (field C, Fig. 4); - 6 . 2 to --1.3 and - 1 3 . 7 to +0.8%0 in the UAM unit (field B) and - 4 . 3 to - 1.8 and - 2 . 9 5 to + 1.6%o in WLM and CLMC suites, respectively (field A). Also shown in Fig. 4 are the average isotopic compositions of the Tertiary marine and freshwater limestones (Keith and Weber, 1964).

6. Diagenesis

Before we use the isotopic data for palaeo- environmental reconstruction, it is important to

72 A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77

m E3 r,

.-.z

+ 5

-5 _E)

- 15

-25

- 3 5 I

I

t -8

I I 'T" I I I I [ [ I

- ~ e '~ -..x CA ~ / k+O O + A / --~ /El \ + +o J • /

\ \ I o o • / / \ \ \ [ -/" \ \ /

C\ \~..I / / \\ + T" ~ -0~ /

\ .,~! • CLMC OLIGOCENE \ I X W L M LATE MID. E O C E N E - L U T E T I A N

\ + [ ~ D I S C O C Y C L I N A FROM W L M

\ ÷ + [ • U A M E A R L Y M I D , E O C E N E - L U T E T I A \ I 6 B I V A L V E F R O M U A M

\ [ • L A M E A R L Y M I D . E O C E N E

\ \ + [ 1:71 A V E R A G E T E R T I A R Y M A R I N E LIMESTOI

\__i'/ (~ AVERAGE T E R T I A R Y FRESH WATER L I M E S T O N E

I I I I I I L I I I - 6 -4 - 2 0 +2

818 0 ( ' / o . ) PDEi

Fig. 4. Plot of 5180 vs. (~13C values of Kutch carbonates. Average Tertiary marine and freshwater limestone values are taken from Keith and Weber (1964). Fields A, B, C represent total spread of isotopic values in WLM, U A M and LAM respectively (for details see the text).

examine the extent of diagenesis on the original composition. The essentially unaltered fresh fora- miniferal tests of a shallow marine environment and high Mg-calcite composition found in almost all the units suggest that the bulk composition of the dominant framework materials is largely unaffected. In many cases, foraminifera tests exhibit profuse micritisation suggesting an early, almost syndepositional, marine phreatic diagenesis under the influence of algal action (Bathurst, 1971). Compaction effects such as pressure weld- ing, squashing and cracking of foraminifera tests are rare, which implies an early lithification. Spar cements are rare; so is secondary moldic or vuggy porosity. In the matrix, too, micritic mud, present in varying amounts in different rocks, does not show any major evidence of neomorphism or other alterations. This micrite mud is also high in Mg-calcite as found in the framework test materi- als or the internal sediments in primary intragranu- lar voids; staining by Alizarin Red-S and ferric chromate was used to confirm the composition of the different carbonate components. Ferrugination locally affects the composition of tests in the LAM and UAM rocks. Ferrugination is also present in the CLMC rocks where only test surfaces show a coating of ferric limonites, which give an ochre•us

appearance. Micritisation and ferrugination are not found in the WLM rocks. Diagenesis in the WLM and CLMC rocks broadly includes a greater amount of compaction and occasional sparitisa- tion or development of spar cement. However, for isotope analyses those samples showing a lesser amount of spar, or none at all, were considered.

A critical look at the isotope data (Table 2) shows that compared to the total spread of 5180 (maximum 7%0 in the LAM and UAM), the spread of 513C is very high (14.5%o in the UAM, 32.5%o in the LAM), except in the WLM suite, and no concomitant change between 6180 and 513C is observed. Any pervasive post-depositional diagen- • sis, on the other hand, would produce a sympa- thetic variation between 5180 and 613C (Veizer and Hoefs, 1976). Based on these observations, we tend to believe that the isotopic compositions of the Kutch Palaeogene carbonates have not been seriously altered after their formation and that the diagenesis was early marine in nature.

7. Discussion

The 5180 of calcium carbonate primarily depends on two factors: temperature and the is•-

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77 73

topic composition of the ambient water (Shackleton, 1974). The range of values of 6180 of the Kutch carbonates is ~7%0 which is too large to be explained by the temperature change alone, since the temperature coefficient of oxygen isotopic fractionation is only 0.2%o/°C. The palaeo- geographic position of the Kutch area during the Tertiary period was at about latitude 8°S (Dietz and Holden, 1970). Hence a large change in the temperature (i.e. about 35°C), even on the seasonal scale, could not have been possible for this tropical region. Moreover, the biota, consisting of abun- dant calcareous foraminifera and associated mol- lusks, echinoids, corals and calcareous algae, represent a chlorozoan assemblage of well aerated, warm, shallow seas (Lees, 1975); as discussed earlier, certain parts of the sequence, such as those represented by the LAM and UAM rocks, were formed in more or less restricted conditions. In general, therefore, there was not much difference between the surface and bottom water temper- atures in this basin: a maximum 10-15°C vertical temperature gradient is estimated during the Eocene and Oligocene even in the open ocean (Shackleton and Boersma, 1981). Therefore, even if the majority of the carbonates were formed in the depth of the basin, the temperature was unlikely to introduce 7%0 spread in the 3180. We, therefore, conclude that this spread was caused mainly by the natural variability in the 6180 com- position of the ambient water.

The 6180 of ocean water at a particular place can be altered by various physical processes. The mixing of 180 depleted fresh water with the ocean water, as found in marginal environments like river mouths, tidal flats, lagoons etc., can shift the ~180 towards more negative values. Similarly, evaporation can enrich the 6180 of water, as found in the arid Red Sea region. For example, the 6~80 depletion in the Tertiary freshwater carbonate (Fig. 4) compared to the marine carbonate, is due to the fact that they were deposited in a lSO deficient environment. The 3180 and ~13C of the various units of the Kutch carbonates, when com- pared with the average Tertiary marine and fresh- water carbonates (Fig. 4, Keith and Weber, 1964), clearly show that the LAM unit was formed in an intermediate environment. In contrast, the rocks

of the WLM/CLMC suites were deposited in rela- tively more open marine conditions. The 6180 of the UAM unit indicates overlapping of marginal and open marine situations. The influx of fresh water in both the LAM and UAM was enough to affect both the salinity and 3180 values. The total spread of 613C, on the other hand, is about 32%o and much larger than that of 6~80 (Fig. 4). The most noticeable features in Fig. 4 are the presence of highly negative 613C values (up to -29.4%o) in some samples of the LAM and UAM. Such depleted 613C values have been reported earlier but mainly from the unfossiliferous inorganic car- bonates (Hathaway and Degens, 1969; Nelson and Lawrence, 1984).

In general, 613C of carbonate is controlled by the 613C of dissolved total carbonate in water or ZCO2 (Emrich et al., 1970) which is close to 0%o (in PDB scale) for sea water. The freshwater carbonates are depleted in 13C due to the contribu- tion from microbial degradation of terrestrial organic matter (613C~-26%o). Such process is found to be typically operative in the organic-rich sulphate-reducing zone. Anomalously depleted carbonates are also known to be formed through the oxidation of biogenic methane (CH4, 613C~ - 2 0 to --100%0) originated by the anaero- bic bacterial fermentation below the zone of sul- phate reduction. The methane production in the fermentation zone is, of course, associated with complementary CO2 production with an enriched 613C value of about +20%0 (Irwin et al., 1977). However, it is also possible to convert the methane by sulphate reducers under anaerobic conditions without any free oxygen (Schoell and Wellmar, 1981).

For the Kutch carbonates, various sedimento- logical evidence suggests a shallow depth of burial and hence the fermentation zone may not have been reached at all. Furthermore, the ~13C value here do not go beyond -29%o, indicating that any large-scale contribution from methane-derived CO2 was absent. We, therefore, attribute the depleted 613C of the Kutch carbonates to the combined effect of both anaerobic microbial degra- dation of organic matter and the decomposition of biogenic methanes by sulphate reducing bacte- ria. We are, however, unable to identify the exact

74 A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77

component of the carbonates responsible for such anomalous values. In one sample from UAM, we had separated a bivalve and analysed it along with the bulk. Results show (Table 2) that while the bivalve has a 613C value of -4%0, the bulk has a value of -11.7%0 and is distinctly more negative. 61sO, on the other hand, has identical values in both the components. Mollusks generally grow in oxygen isotopic equilibrium (Lloyd, 1964). The metabolic effect in carbon isotopes (if present) is minor. Therefore, the above data indicate that the anomalous 613C values were perhaps contributed by the micrite matrix and/or cements and not by the framework fossil grains and that both were precipitated from water having similar oxygen isotopic compositions. Probably the matrix/cement component was formed a little later than the test materials, when there was a significant change in the ambient carbon reservoir but not in the oxygen. Detailed geological investigation revealed that both the LAM and UAM occur in small pockets, with the latter having more extensive outcrops than the former. However, the laterally discontinu- ous nature of the beds and the varying thickness of both the units suggest that they were deposited within depressions of the pre-existing Deccan Trap basements (Ray et al., 1984). It is plausible that both the units were deposited over these undula- tory trap basements during the mid-Eocene trans- gression and under a fluctuating near-shore environment. We suggest that with the onset of the transgression, the LAM rocks and their fossils were deposited in these depressions, forming semi- locked basins. Subsequent (and immediate) fluc- tuation (regression) of sea level cut off these basins, trapping large amounts of organic debris. The SO4 ions (usually present in excess in the sea water) aided production of 13C depleted CO2 through degradation of organic matter. The situa- tion was conducive to the production of methane which was anaerobically oxidised at the sediment/water interface. The bicarbonates thus produced ended up as precipitating micrite matrix or cement carbonates along with the formation of pyrites. Perhaps these carbonates were formed at a stage when the residual methane concentration was very low because the LAM exhibits 613C which is less negative (only upto 29%0), than

compared to the 613C of the shelf carbonates (upto -60%0) off the NE United states and the Fraser delta which are suspected to be methane-derived (Hathaway and Degens, 1969; Nelson and Lawrence, 1984).

The depositional environment of the UAM was more or less similar to that of the LAM since they have similar negative (~13C values (up to about

- 13.7%0). However, the data indicate a significant contribution from bicarbonates produced in aero- bic tidal lagoons with relatively more seaward extensions, as opposed to the partially locked lagoons of the LAM. Contributions from metha- nogenic carbonates were perhaps negligible in the UAM. It should be noted, in this regard, that the decomposition of methane and/or organic matter in both these units were in all likelihood, operative in a few restricted pockets because the anomalous carbon isotope values are not uniformly found throughout these units.

The large scale presence of organic matter, in close proximity to a terrestrial environment, is evident from the occurrence of impressions of plant leaves, thick lignite beds almost at the same stratigraphic levels (Fig. 2a,b) and abundant opaque minerals perhaps derived from the basaltic trap basement. The occurrence of gypsiferous clas- tics (occasionally with 1 m thick gypsum layers) interbedded with the lignites or clastics indicates short intervening arid conditions during the post- LAM regressive phase. A reducing milieu in the LAM is attested by the presence of pyrites inside the spiral laminae of benthic foraminifera (replac- ing the high Mg-calcite; Fig. 3c). The UAM, of course, is characterised by limonitic iron oxides (and no pyrites), indicating a more aerated condition.

In contrast, we infer an open marine deposi- tional environment for the WLM and CLMC suites based on the clustering of both 3180 and 313C around the average Tertiary marine carbonate values (Fig. 4). Often the larger benthic foramini- fera grow in oxygen isotopic equilibrium (Luz et al., 1983) and hence we also analysed a separated D i s c o c y c l i n a test and the bulk in a sample of the WLM (Sample A-32(2), Table 2). c5180 and 613C in the test are -3.2%o and +1.6%o respectively and are not much different from those of the bulk

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77 75

(-3.7%0, + 0.4%0), indicating that both the matrix and framework had essentially marine sources of oxygen and carbon. The relatively more open marine condition for the WLM and CLMC is also attested by the typical chlorozoan assemblage found in the tropical belt. Both the WLM and CLMC are composed of pure carbonate bodies, distinctly different from the early middle Eocene CEM formation having prominent clastic deposits (terrigenous inputs), probably accumulated by intense land-derived run-off.

8. Conclusions

Studies of stable isotope and faunal assemblage, carried out on the early Palaeogene carbonate- clastic sequence of the Kutch, Western India show that the Tertiary sedimentation in this region started only in the early middle Eocene. With the onset of post-Deccan Trap transgression, bioclastic limestones were deposited in the semi-locked lagoons formed over the depressions of basaltic trap basements. The biota which could develop in this highly stressed environment consisted of the virtually monospecific assemblage of Assilina. Oxygen isotope analyses on these limestones indi- cate that apart from the sea water, these lagoons were receiving large amounts of fresh water too. These restricted basins were soon cut off due to the fluctuation of sea level trapping a large amount of organic matter (growing in and around these basins), evidenced by the presence of thick lignite beds above the limestones. The decay of this organic matter and the oxidation of biogenic meth- ane gas, generated in some pockets, produced bicarbonates extremely depleted in 13C as revealed by the carbon isotope values of these limestones. During this short post-LAM arid phase (when few intervening gypsum beds were deposited) conver- sion of organic matter by sea water SO4- ions and methane oxidation took place in anaerobic condi- tions. The reducing milieu was also conducive to the formation of pyrites.

The close proximity of these basins to land is also shown by the considerable volume of clastic deposits. A condition of high run-off, indicating abundant precipitation (with minor arid pulse in

between) may be envisaged which was conducive to luxuriant growth of plant matter (lignite). The global compilation of palaeoclimatic data by Barron (1989) shows that precipitation had increased during the Eocene, causing intense weathering products like laterites, bauxites, fer- ricretes, etc. often associated with lignites. In the arid and semiarid regions of western India, such early Tertiary laterites and ferricretes are abun- dantly found (Kumar, 1986). Based on proxy climatic data, a dominantly humid climatic zone with intermittent semihumid conditions in north- west and western India (Tethyan sequenses) during the early Tertiary has also been suggested (Oberhansli, 1992). In southern India too, early Tertiary red soils have been reported which indi- cates a humid climate with rainfall exceeding 2000mm yearly (Bruhn, 1989). A warm, humid climatic condition during Eocene has been inferred by various workers from wide geographic locations namely Afganistan, western Africa, central and northern Europe which indicate that the Eocene climatic optimum was global in nature and extended as far as 40°N (Parrish et al., 1982; Bardossy and Aleva, 1990; Murray and Wright, 1974; Valeton, 1983). The middle Eocene lagoons of the Kutch was perhaps also receiving large amounts of fresh water due to this dominantly wet climatic condition. Isotope and sedimentological evidence on the upper part of these bioclastic limestones indicate an aerated tidal lagoonal situa- tion with a more seaward extension. This indicates a renewed transgression of the sea which culmi- nated in the deposition of open marine (late middle Eocene and Oligocene) limestones. Apart from a great variety of benthic foraminifera, corals, bivalves, bryozoans, red algae etc., these rocks are also labelled by typical marine oxygen and carbon isotope values.

Acknowledgements

We thank Profs. A.K. Saha and T. Roychowdhury for encouragement. Comments from the reviewers H. Oberhansli and V. Grazzini were extremely helpful in improving the quality of the paper. Financial assistance for the field work

76 A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecolog) 121 (1996) 65 77

was provided to AS and AKR by the Indian National Science Academy and the Council of Scientific and Industrial Research, New Delhi respectively.

References

Bardossy, G. and Aleva, G.J.J., 1990. Lateritic bauxites. Dev. Econ. Geol., 27:1 624.

Barron, E.J., 1989. Climate variations and the Appalachians from the Late Palaeozoic to the present: Results from model simulations. Geomorphology, 2:99 118.

Bathhurst, R.G.C., 1971. Carbonate Sediments and their Diagenesis (Devel. Sedimentol., 12). Elsevier, New York, 822 pp.

Bhatia, S.B., 1985. Some aspects of early Tertiary biostratigra- phy and palaeogeography of the Indian subcontinent. Q. J. Geol. Min. Met. Soc. Ind., 57(2): 63-98.

Biswas, S.K. and Raju., D.S.N., 1973. The rock stratigraphic classification of the Tertiary sediments of Kutch. Bull. Oil Nat. Gas. Comm., 10(1 2): 37 45.

Biswas, S.K., 1986. Palaeogene of Kutch- a rejoinder. Ind. J. Earth. Sci., 13(4): 343 360.

Bruhn, N., 1989. Substrate Geneses Rumpfltachendynamic. Kieler Geogr. Schr., 74:173 179.

Dietz, R.S. and Holden, J.C., 1970. Reconstruction of Pangaea: Break-up and dispersion of continents, Permian to present. J. Geophys. Res., 75:4939 4956.

Eames, F.E., 1952. A contribution to the study of Eocene in western Pakistan and western India: Discussion of the faunas of certain standard section and their bearing on the classification and correlation of the Eocene in western Pakistan and western India. J. Geol. Soc. Lond., 107: 173 300.

Emrich, K., Ehhalt, D.H. and Vogel, J.C., 1970. Carbon isotope fractionation during the precipitation of calcium carbonate. Earth Planet. Sci. Lett., 8:363 370.

Hathaway, J.C. and Degens, E.T., 1969. Methane-derived marine carbonate of Pleistocene age. Science, 165:690 692.

Irwin, H., Curtis, C. and Coleman, M., 1977. Isotopic evidence for source of diagenetic carbonates formed during burial of organic rich sediments. Nature, 269: 209-213.

Keith, M.L. and Weber, J.N., 1964. Carbon and oxygen isotopic compositions of selected limestones and fossils. Geochim. Cosmochim. Acta, 28: 1787-1816.

Kumar, A., 1986. Palaeolatitudes and the age of Indian laterites. Palaeogeogr. Palaeoclimatol. Palaeoecol., 53: 231 237.

Lees, A., 1975. Possible influence of salinity and temperature on modern shelf carbonate sedimentation. Mar. Geol., 19:159 198.

Lloyd, M.R., 1964. Variations in the oxygen and carbon isotope ratios of Florida bay mollusks and their environmen- tal significance. J. Geol., 72: 84-92.

Luz, B., Reiss, J. and Zohary-Segev, T., 1983. Seasonal variation of stable isotopes in Amphisorous from Gulf of Aqaba (Elat), Red sea. In: J.E. Meulenkamp (Editor), Reconstruction of Marine Palaeoenvironments. Utrecht Micropalaeontol. Bull., 30:229 237.

Mohan, M. and Gupta, S.C., 1968. The microfauna and age of gypseous shales western Kutch, India. Contrib. Cushman Found. Foraminiferal. Res., 18:21 26.

Murray, J.W. and Wright, C.A., 1974. Palaeogene foraminifera and palaeoecology, Hampshire and Paris basins and the English channel. Spec. Pap. Palaeontol., 14:1 129.

Nagappa, Y., 1959. Foraminiferal biostratigraphy of Cretaceous Eocene succession in the India Pakistan Burma region. Micropalaeontology, 5:145 192.

Nelson, C.S. and Lawrence, M.F., 1984. Methane-derived high-Mg calcite submarine cement in Holocene nodules from the Fraser delta, British Columbia, Canada. Sedimentology, 31:645 654.

Oberhansli, H., 1992. The influence of the Tethys on the bottom waters of the early Tertiary ocean. In: The Antarctic Palaeoenvironment: a Perspective on Global Change (Antarct. Res. Ser., 56). Am. Geophys. Union, Washington, DC, pp. 167 184.

Pande, K., Venkateshan, T.R., Gopalan, K., Krishnamurthy, P. and Mcdougall, J.D., 1988. 4°Ar 3gAr ages of alkali basalts from Kutch, Deccan volcanic province, India. In: K.V. Subbarao (Editor), Deccan Flood Basalts. Geol. Soc. Ind., pp. 145 150.

Parrish, J.T., Ziegler, A.M. and Scotese, C.R., 1982. Rainfall patterns and the distribution of coals and evaporites in the Mesozoic and Cenozoic. Palaeogeogr. Palaeoclimatol. Palaeoecol., 40:67 101.

Raju, D.S.N., 1969. Eocene Oligocene boundary in Ramania, Kutch. Bull. Oil Nat. Gas Comm., 6(2): 71 73.

Raju, D.S.N., 1971. Observations on the Eocene, Oligocene and Miocene foraizqiniferal biostratigraphy of Kutch, Western India. Publ. Cent. Adv. Stud. Geol., Punjab Univ., Chandigarh, 10:136 155.

Ray, A.K., Ghosh, C.C., Sen, P., Ghosh, A., Nag, S., Saha, D. and Chowdhury, A., 1984. Palaeogene of Kutch an alternate view on the initial part of the sequence. Ind. J. Earth. Sci., 11:207 224.

Ray, A.K., 1987. Palaeogene stratigraphy of Kutch: interacting development of rocks and fossils. Thesis. Univ. Calcutta, 1 and 2, 269 pp.

Samanta, B.K., 1962. Assilina from Eocene beds of Garo hills, Assam. Bull. Geol. Min. Met. Soc. Ind., 27:1 25.

Samanta, B.K., 1969. Taxonomy and stratigraphy of the Indian species of Discocyclina (Foraminifera). Geol. Mag., 106(2): 115 127.

Samanta, B.K., 1981. Two stratigraphically important Nummulites species from the middle Eocene of India and Europe. Paleontology, 24:803 826.

Sarkar, A., Ramesh, R., Bhattacharya, S.K. and Rajagopalan, G., 1990. Oxygen isotope evidence for a stronger winter monsoon current during the last glaciation. Nature, 343: 549 551.

A. Sarkar et al./Palaeogeography, Palaeoclimatology, Palaeoecology 121 (1996) 65 77 77

Schoell, M. and Wellmer, F.W., 1981. Anomalous 13C depletion in early Precambrian graphites from Superior Province, Canada. Nature, 290: 696-699.

Shackleton, N.J., 1974. Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina: isotopic changes in the ocean during the last glacial. Colloq. Int. C.N.R.S., 219: 203-209.

Shackleton, N.J. and Boersma, A., 1981. The climate of the Eocene ocean. J. Geol. Soc. Lond., 138: 153-158.

Valeton, I., 1983. Klimaperioden lateritischer Verwitterung und

ihr Abbild in den synchronen sedimentationsr~iumen. Z. Dtsch. Geol. Ges., 134: 413-452.

Veizer, J. and Hoefs, J., 1976. The nature of ~80/160 and 13C/aZC secular trends in sedimentary carbonate rocks. Geochim. Cosmochim. Acta., 40: 1387-1395.

Wynne, A.B., 1872. Memoir on the Geology of Kutch to accompany the map compiled by A.B. Wynne and F. Fedden during the seasons of 1867-1868 and 1868 1869. Mere. Geol. Surv. Ind., 9, 289 pp.