ELSEVIER Marine Geology 136 (1997) 299-315

Palaeoceanographic conditions during the formation of a ferromanganese crust from the Afanasiy-Nikitin seamount,

North Central Indian Ocean: geochemical evidence

V.K. Banakar *, J.N. Pattan, A.V. Mudholkar

National Institute of Oceanography, Dona Paula, 403004 Goa, India

Received 5 October 1995; revision 26 August 1996; accepted 26 August 1996

Abstract

A ferromanganese crust dredged from the summit of the Afanasiy-Nikitin seamount in the North Central Indian Ocean (NCIO) has recorded Neogene oceanographic events. The substrate of the crust is composed of fresh-water phreatic calcite cement, Terebratulinae casts, rounded and ferruginised basalt clasts and weathered coralline algal fragments suggesting subaerial exposure during the Oligocene (N 30 Ma) global sealevel drop.

The mineralogy, major, trace and REE element geochemistry and Co-model age estimates suggest three distinct accretionary environments during the crust growth: (1) a period of contemporary precipitation of Fe-Mn oxide and carbonate fluorapatite (CFA) indicate an intensified intermediate water oxygen minimum zone (OMZ) in the late Miocene, (2) a pulse of very high CFA deposition and detrital input in addition to Fe-Mn oxide accretion at the close of Miocene reflect a more intense OMZ and the erosion of the Himalayas and, (3) more oxidizing conditions of the ambient seawater due to contraction of the late Miocene OMZ facilitated the accretion of almost pure Fe-Mn oxide since the Pliocene. These interpretations solely depend upon the chronology based on an empirical relationship and are subjected to the confirmation by radiometric dating.

As a consequence of significant deposition of CFA in addition to Fe-Mn oxide during the intensified OMZ, the normally coherent behaviour of trivalent rare earth elements (3 + REE) is not observed in the crust. La, Yb and Lu show a positive association with the CFA phase elements (Ca and P) and Pr, Nd and Sm with the Mn-oxide phase elements (Mn, Co and Ni). This leads to an abnormal, incoherent behaviour of the 3 + REE in multimineral authigenic system.

Keywords: Central Indian Ocean; ferromanganese crust; seamount; palaeoceanography; geochemistry

1. lutroduction

Ferromanganese encrustations (crusts) are largely hydrogenous colloidal precipitates (Hein

* Corresponding author.

et al., 1992a and references therein) which are known to record oceanographic conditions during their growth (Segl et al., 1984; Banakar and Borole, 1991; De Carlo, 1991; Eisenhauer et al., 1992; Hein et al., 1992b). Crusts accreting on seamounts have been shown to concentrate cobalt (Halbach et al., 1983) and the cobalt concentration has

0025-3227/97/$17.00 0 1997 Elsevier Science B.V. All rights reserved PII SOO25-3227(96)00065-5

300 V.K. Banakar et ul.JMarine Geology I36 (1997) 299-315

been used to derive an empirical relation [G (mm/m.y.) = 1.28/[Co(%) - 0.241 for estimating their growth rates (G) (Puteanus and Halbach, 1988). The occurrence of hydrothermal crusts has also been reported from several seamounts and can be differentiated from the hydrogenous crusts by their distinctly different composition (Hein et al., 1992a; Usui et al., 1993). The primary requirements for accretion of the hydrogenous crusts are (a) a sediment-free hard substrate, (b) fairly oxic ambient waters, (c) a large supply of oxidisable Mn. Such conditions are generally found on seamounts at shallow-water depth below the OMZ. The study of hydrogenous crusts from seamount areas therefore provides valuable evi- dence for variations in the palaeo-accretionary and palaeo-OMZ conditions. In addition, the nature and composition of substrate of the crust may reflect oceanographic conditions prior to the crust accretion.

The chronological resolution of crusts by several direct methods (isotope and biostratigraphy dating) as well as the empirical method (Co-model) is rather poor compared to that obtained for sediments as a consequence of their very slow accretion rate. Nevertheless, crusts offer reliable information regarding palaeoenviron- mental conditions during their growth.

The present study is aimed at understanding the oceanographic conditions during the crust accre- tion and substrate formation in a specimen dredged from the upper flank of the Afanasiy-Nikitin sea- mount (hereafter refered as seamount), in the NCIO. We believe that the studied specimen repre- sents accretionary environment in the NC10 based on the facts; (a) several studies using individual crusts of world oceans in the past have demon- strated that the seamount crusts are the condensed stratigraphic sections that record oceanographic and geologic conditions of the surrounding envi- ronment (see Hein et al., 1992b and references therein), and (b) the oceanographic changes in the equatorial Indian ocean are of regional scale. This study is of significance because, (a) the seamount is located at the center of the late Miocene intraplate deformation zone in the NC10 (Karner and Weissel, 1990) during which a major hydro- thermal discharge was recorded in the sediments

of the distal Bengal fan (Boulegue and Mariotti, 1990) and (b) Late Miocene/Pliocene area1 expan- sion and intensification of OMZ in the Indian ocean was proposed from the study of ODP sedi- ment cores of Broken ridge (Dickens and Owen, 1994). These oceanographic variations are expected to influence the composition and mineral- ogy of the crust.

In this paper, we present a detailed study of the major, trace and REE element behaviour in a crust from the seamount. To evaluate the relationship between the growth history of the crust and the palaeoenviromnent we have used ages based on the cobalt concentration accretion rate method (Puteanus and Halbach, 1988).

2. Tectonic setting of the seamount

The seamount is presently located at 3”s latitude and 83”E longitude in the NC10 about 1000 km southeast of Sri Lanka (Fig. 1). It extends about 300 km in N-S direction (Paul et al., 1990) and its summit is located at 1550 m water depth

Sediment thickness adjacent to the seamount is -1.5 km (Levchenko et al., 1993). Our single

(Rudenko, 1994). The emplacement of the sea-

beam echosounder bathymetric data on either side of the seamount yields an average seafloor depth

mount was

of 4800 m. The calculated basement depth below the sea level is therefore -6300 m and the relief

-75 Ma (Cande and Kent, 1992;

of the seamount is

Karner and Weissel, 1990).

-4750 m above basement (Fig. 2).

Using the age versus depth curve of Parson and Sclater ( 1977) the expected subsidence due to lithospheric cooling places the basement at a depth of - 5500 m. The difference of - 800 m between expected and actual depths of basement is therefore attributed to subsidence due to sediment loading. Fig. 2 presents a simplified subsidence diagram of the seamount which gives a cumulative subsi- dence rate of -50 m/m.y. Assuming uniform, uninterrupted subsidence since the formation of the seamount, a simplified subsidence history of the seamount can be summarised as follows

WC Banakar et al./Marine Geology 136 (1997) 299-315 301

03

OE 83’03’ 06’ 09*

Fig. 1. Schematic map showing the location of the Afanasiy-Nikitin seamount. The dredge track is represented by a thick line on the magnified portion of the seamount bathymetry (adopted from Rudenko, 1994).

using the plate reconstructions of Royer et al. (1992):

(a) from the original emplacement on the mid- ocean ridge at a height of N 2300 m above sea level around N 75 Ma at w 50”s latitude and 50”E longi- tude, the seamount subsided to sea level -30 Ma when it was at N 20”s latitude and 65”E longitude.

The seamount summit therefore experienced varying depositional conditions at various water depths over time.

3. Material and methods

(b) continued subsidence brought the summit of the seamount to the present water depth ( 1550 m).

Two large pieces of crusts were dredged along a precise track from 1700 to 1600 m on the upper-

302 V. K. Banakar et ai.IMarine Geology I36 (1997) 299-315

Presenr l000tion

3OS 6t83OE

Expected depth of basement (-55OOm) due to lithospheric cooling.

_ 7_ @ Actual depth of basemen? (-63OOm) due to combination of lithospheric cooling and sediment loading.

I 75 7’0

I 40

I I 60 40 30 2’0

I IO 6

Ma

Fig. 2. Schematic diagram of simplified subsidence history of the Afanasiy-Nikitin seamount. The three standings of the seamount for different time periods with respect to the sea-level is based on the estimated subsidence rate of - 50 m/m.y. : (a) using Royer et al. (1992) plate reconstructions, (b) Kashintsev et al. (1987), (c) Haq et al. (1987). Present-day OMZ stand was obtained by extrapolating the OMZ areal expansion in the Indian ocean given by Dickens and Owen ( 1994).

most flank of the seamount (Fig. 1) on board R/V AA. Sidorenko during December 1994. This is the first recovery of crusts from this area and these crusts comprised the entire dredge haul. The thick- ness of the crust layer (black portion in Fig. 3) varied from a few mm on one side (bottom of the figure) to 41 mm on the opposite side (top of the figure). The top of the crust which was in direct contact with the overlying water is compact and thick. The thin crust layer on the opposite side is highly granular and loosely held suggesting depos- ition of the Fe-Mn oxide through a crevice in the substrate outcrop. This side was therefore assumed to be the bottom of the sample and is not consid- ered in this study. The substrate is made up of conglomerate limestone cemented by calcite (white portion in Fig. 3). Conglomerate components

include several large brachiopod casts and gravel size basalt clasts.

A small portion of the crust having nearly uniform surface texture and maximum thickness (41 mm) was cut from the largest specimen (i.e., from left edge of the specimen shown in Fig. 3) for detailed analyses. The sample was carefully scraped using a surgical blade to obtain subsamples at uniform depth intervals perpendicular to the growth axis of the crust. Scraped layers were ground to - 230 mesh in an agate mortar. Aliquots dried overnight in an oven (105110°C) were leached with 20 ml of 3 N HCl in a CEMR MDS-2000 microdigest to extract the hydrolysate fraction of the crust. The residue was dried, weighed and stored. The filtrate was diluted to volume using BarnstedR 18 mohm deionised water.

FC K. Banakar et al.lMarine Geology 136 (1997) 299-315 303

Fig. 3. The studied Fe-Mn crust. Conglomerate substrate con- sists of Terebratulina casts and rounded basalt clasts cemented by microspar calcite (white portion). The Fe-Mn crust (black portion) is thick on the side facing top of the photograph. On the bottom side the substrate is convex and hence the crust appears to be thick but the actual crust is only 14 mm thick and is highly granular. For the present study, the sample was cut from the left top edge of the specimen displaying maximum crust thickness of 41 mm.

A 15 ml batch of the solutions was used for the analyses of major and trace elements. The 85 ml batch of the solutions and synthetic crust solution were subjected to cation chromatographic extrac- tion of REEs following standard procedures. The extract was analysed for ten REEs using the emis- sion lines recommended by De Carlo ( 1990). All the analyses were carried out on a Perkin-ElmerR ICP-AES, Plasma-400 calibrated using a near- crust composition multi-element calibration stan- dard prepared from single element stock solutions of AldrichR chemicals. The REE values obtained for synthetic crust solution indicated that the recovery of REEs was nearly complete (98- 102%). Duplicate analyses of U.S.G.S. A-l nodule stan- dard were used for the accuracy estimation of our analytical results. The analytical errors were within + 4% for Mn, Fe, Co, Cu, Ca, La, Ce, Ndj Sm, Eu, Tb, Yb and Lu; f5% for Ni, Ba, P and Pr; f 11% for Ti and Gd. Error in the SC values was not determined but considered to be acceptable based on its strong sympathetic association with detrital components in the crust and consistent results (N 7 ppm) for U.S.GS. A-l.

Powdered samples from zones with distinct col-

ours in the crust, the pmleaehable residue of the crust, and the substrate cement were subjected to mineralogical &ilyses on & PhilipsR X-Ray diffractometer nsing Cu( Ka) radiation. Thin sec- ti,ctns of the,,lsut+.rate cement and basalt clasts were s&died under a polarised light microscope.

Growth rates for the individual layers of the crust were estimated using the cobalt concentration empirical formula of Puteanus and Halbach (1988). These rates were used to assign ages to individual depth layers. The oceanic Co flux remains nearly constant at all depths in the water column (Halbach et al., 1983). The Co concen- tration empirical formula was derived after evalu- ating several chemical analyses and “Be based growth rates of the crusts accreting above the carbonate compensation ‘depth (CCD) (Puteanus and Halbach, 1988). Therefore, we assume that this empirical formula is also valid for the present crust accreted well above the CCD. However, we are aware of the limitations in such estimates, p&&larly for the apatite enriched layers where highly complex post-depositional chemical exchange of metal species is expected. The chronol- ogy is therefore approximate with about 20% absolute error. We give more emphasis to the dates obtained for pure Fe-Mn oxide portion than for the apatite enriched portion of the crust.

4. ResuIts and discwsion

4.1. The substrate

The conglomerate substrate is composed of Terebratulinae casts (Muir-Wood et al., 1965) (Fig. 4a), subrounded to rounded, ferruginised basalt clasts (Fig. 4b, c), reworked shallow-water coralline algal fragments (Fig. 4e) and several other unidentified skeletal fragments. Deep-water fauna such as cocoliths are not present either in the cement or in the assemblage of skeletal frag- ments. The calcite cement exhibits neomorphic features (Fig. 4d) with microspar texture (matrix material in Fig. 4e). The results of chemical analy- sis and mineralogy (Fig. 5a) suggest that the cement in the conglomerate substrate is a low

304 V. K Banakar et al./Marine Geology 136 (1997) 299-315

Fig. 4. Various components of conglomerate substrate. (a) photograph of two large Terebratulinae casts extracted from the substrate, (b) photograph of basalt clasts extracted from the substrate, (c) photomicrograph (50 x) of a basalt clast, (d) photomicrograph (250 x) showing the neomorphic features in calcite cement, and (e) photomicrograph (125 x) of algal clast. Note the roundess of basalt and algal clasts.

magnesium calcite with a composition

CaI.~M&.ozS (CO&. Low Mg-calcite has been shown to form under

fresh-water phreatic (Bathurst, 1975) as well as deep-water marine conditions (S&lager and James, 1978). Our results clearly indicate that subaerial exposure and magnesium leaching under fresh-water phreatic conditions have led to the formation of microspar calcite (Bathurst, 1975;

Rao and Nair, 1994). Microspar calcite has cemented various components of conglomerate which in turn acted as a suitable substrate for the crust.

From the subsidence diagram (Fig. 2), it appears that the summit of the seamount was at sea-level -30 Ma. A major global drop in sea- level of more than 100 m occurred during this period (Haq et al., 1987). The neomorphic calcite

V.K. Banakur et al./Marine Geology I36 (1997) 299-315 305

I

34 30 25 I

30 25 20

CFA

Fig. 5. X-ray diffractograms of (a) substrate cement: C= calcite, (b) nonleachable residue: Q= quartz and F= feldspar, (c) apatite- rich intermediate zone of the crust: CFA = carbonate fluorapatite.

cement and rounded basalt clasts in the substrate suggest that the Oligocene sea-level drop might have exposed the substrate to subaerial high energy conditions. The subaerial exposure of the ridges during this period has been documented in other sites of the Indian ocean (Rea et al., 1990). Based on micropalaentology and isotope dating of the calcite cement in a tutI breccia from a seamount located 200 km northeast of the Afanasiy-Nikitin seamount, Kashintsev et al. (1987) deduced contin- ued volcanic activity until the Eocene. Fresh-water phreatic calcite cement in the conglomerate sub- strate might therefore be the result of later subaer- ial exposure of the seamount.

4.2. The crust

Megascopically, the crust is composed of three zones, 41-28 mm is dark brown (older zone),

28-19 mm is light brown (intermediate zone) and 19 mm to surface is black (younger zone).

(a) Age estimates: Three distinct growth periods have been recorded in the three zones of the crust. The older zone displays a slow but rapidly fluctu- ating growth (2.3-5.1 mm/m.y.) which corres- ponds to the late Miocene. The intermediate zone displays a short-lived episode of very rapid growth (6.4-12.8 mm/m.y.) at the close of the Miocene. The younger zone displays a fairly constant slow growth (2.6-4.7 mm/m.y.) since the Pliocene (Table 1).

(b) Major and trace metal variation: All the analyzed major and trace elements exhibit distinct variations in the three different growth zones.

In the older zone, Fe-Mn oxide phase elements display nearly constant concentrations (e.g., Mn -2296, Fe -lO%, co -0.7%, Ni -0.496, cu -0.06%) with Mn/Fe ratios around 2 (Table 1,

306 V.K. Banakar et al./Marine Geology 136 (1997) 299-315

Fig. 6). This zone contains, &MnO, as the major mineral phase, significant amounts of CFA (Ca - 9%, P - 3%) and low detritus (residue - 2%, Ti - 0.6%, SC - 3 ppm). In the intermediate zone, an extreme dilution of the hydrogenous ferromanga- nese material (Mn < 14%, Co <0.4%, Ni ~0.3% and Mn/Fe<l) by CFA (Ca 12.8% and P 5.1%) and detritus (residue 18.5%, Ti 1.5% and SC 10 ppm) is evident (Table 1 and Fig. 5c and 6). The detritus in the crust is composed of quartz and feldspars (Fig. 5b). Intermediate and older zones contain about 25 and 10% CFA, respectively which is intimately intermixed with the Fe-Mn oxides.

The intimate intermixing of the CFA within the Fe-Mn oxide material suggests contemporaneous deposition of Fe-Mn oxide and CFA during the late Miocene. Suboxic conditions have been shown to favour the contemporaneous deposition of these minerals (Rao and Burnett, 1990). Alternatively, the microenvironments favourable to CFA precipi- tation (Slansky, 1986) might have fluctuated rapidly within the suboxic waters yielding an inti- mate mixing. The development of intensified suboxic OMZ due to enhanced global productivity during the late Miocene (see Peterson et al., 1992) may have taken place at the beginning of crust accretion. The Ca/P ratios in the crust are <3 which is generally found in CFA (Slansky, 1986) indicating the absence of hydrothermal calcite in the crust matrix. The Mn, Fe, Co, Ni and Cu concentrations and Mn/Fe ratios (Table 1) are in the range typical of hydrogenetic crusts. We there- fore rule out any effect of hydrothermal activity associated with the late Miocene intraplate defor- mation (Boulegue and Mariotti, 1990) in the prox- imity of the seamount. The sharp increase of CFA and residue contents in the intermediate zone of the crust occurred at N 5 Ma (Table 1) and might reflect the peak productivity in the equatorial Indian Ocean resulting in most intense OMZ at the Miocene/Pliocene boundary (Dickens and Owen, 1994). The record of high productivity during late Miocene (Peterson et al., 1992) is also shown by high concentration (up to 2400 ppm) of the proxy element for productivity, Ba (Dymond et al., 1992) in older and intermediate zones of the crust (Fig. 6). A high content of detrital elements

in the intermediate zone might reflect an intensified erosion of the Himalayas during the close of Miocene (Rea, 1992) resulting in an increased input of suspended detritus to the NCIO.

In the younger zone, hydrogenous ferromanga- nese material is maximum with minimum CFA and low lithogenous detritus (Mn -21%, Fe -17%, co -0.6%, Ni -0.3%, Mn/Fe -1; Ca - 2.3%, P -0.4%; residue -2%, Ti -0.7%, SC - 5 ppm; Table 1, Fig. 6). &MnO, is the only identifiable mineral phase in this zone with broad peaks at 2.4 A and 1.4 A. Sharp depletion in CFA and detrital content (Ca 2.62 from 6.28% and P 0.58 from 2.16%; residue 1.1 from 11.5%, Ti 0.74 from 1.06% and SC 4.4 from 7.7 ppm) in 17-19 mm interval (Table 1 and Fig. 6) corresponds to the beginning of the Pliocene and marks the com- mencement of almost pure Fe-Mn oxide accretion. The Ba content of the crust also drops from -2400 ppm (in the intermediate zone) to - 1400 ppm (Table 1, Fig. 6) indicating reduced productivity since the Pliocene. The drop in the Pliocene productivity level might have resulted in the contraction of late Miocene intense OMZ in NCIO. A combination of contracting OMZ due to reduced equatorial productivity and continued subsidence of the seamount have therefore grad- ually moved the crust ( - 1650 m water depth) into less oxygen depleted suboxic waters during the early Pliocene leading to the precipitation of almost pure hydrogenous Fe-Mn oxide.

A simple interelement correlation matrix for the crust (Table 2) yields strong sympathetic associa- tions among Mn, Co and Ni (Y= >0.8); Ca and P (r= >0.9) and residue, Ti and SC (Y= > 0.7) suggesting three distinct group of elements: hydrogenous Mn-oxide, CFA, and, lithogenous detritus respectively, in accordance with the different growth zones accreted in response to varying palaeo-environment in NCIO. Cu does not show specific association with any of the above groups. Ba is moderately sympathetic with the CFA group (r= -0.6, Table 2) indicating its rela- tionship to productivity. Fe exhibits a strong anti- thetic association with the CFA group and no association with the other two groups (Table 2) indicating its independant precipitation as an oxy- hydroxide. These relationships are in general

V.K. Banakar et aLlMarine Geology 136 (1997) 299-315 307

Mn %

I = a 5 r:

T 5

IO

14

15

Jz

22.5

27

51

33

41

Residue %

I u 0 u

5

IO

14

le

3’ E

22.5

27

0 m

33

4’ -0

-I P %

* I-

N-

-lW-

- 5

2

IO

3 I4

3 10 22.5

22

31

35

41 ti

Ce - anomaly b L u

I u u ”

5

IO

I4

18

22.5

27

31

35

41 ~

Fe %

Ti %

8 i? 3 %

E 3+REE pp m

e 8

co % L I ;r

--i

SC. ppm L ID Fi

E Ba x (IO31 ppm Ce x 1 IO31 ppm

: 7 N N

ET”1

Mn/ Fe

Fig. 6. Depth variation in major, trace and REE element content of the crust. For convenience, the mean depth of each sampling interval (Table 1) was used.

308 V.K. Banakar et al,/Marine Geology 136 (1997) 299-315

agreement with the earlier studies of hydrogenous strongly associated with Fe (Y= > +0.7) but nega- crusts (Bonatti et al., 1972; Halbach et al., 1983; tively correlated with Mn, Co and Ni (Y= > -0.7) De Carlo et al., 1987; Hein et al., 1992a,Hein confirming the role of Fe-oxyhydroxides as the et al., 1992b; Banakar and Borole, 1991). sole carrier phase for the 3 + REE.

(c) 3 + REE variation: In the following discus- sion, we analyse the REE behaviour in the three zones of the crust. As a consequence of redox sensitive nature of Ce (Glasby et al., 1987; Elderfield, 1988), we treat its behaviour separately from that of the other 3 + REEs.

The concentrations of all the analysed REEs (Table 1) are within the ranges observed in the hydrogenous Fe-Mn nodules and crusts (Glasby et al., 1987; De Carlo, 1991). The variation of X3+ REE is not systematic with depth in the crust and ranges from 434 to 750 ppm (Table 1, Fig. 6). Nontheless, an increasing content of X3 + REE in the younger zone is seen. Below this zone, the variation in X3 + REE increases (Fig. 6).

The associations of 3+ REE with any of the three major mineral phases in older to intermediate zone (41-19 mm) are not significant (see Table 3). La, Yb and Lu are more sympathetic with Ca and P (CFA phase); Eu and Gd more sympathetic with Fe and Ti (Fe-oxyhydroxide or Fe-Ti hydrate phase (Koschinsky and Halbach, 1995)), and Pr, Nd and Sm more sympathetic with Mn, Co and Ni (Mn-oxide phase). These associations clearly suggest the possibility of random fractionation of trivalent REE when several potential carriers are present in the water column.

The coherent behaviour of 3 + REEs in natural systems is well known (Elderfield et al., 1981; Calvert et al., 1987; Glasby et al., 1987; De Carlo, 1991; Murray et al., 1992; Nath et al., 1992; Sholkovitz et al., 1993; Bertram and Elderfield, 1993; Pattan and Banakar, 1993 and references therein). In the present case, La amongst light REEs (LREE) and Yb and Lu among heavy REEs (HREE) deviate from the above general rule (Table 2). La exhibits a moderate positive associa- tion (r= -0.5) with its neighbours and HREE but poorly associated with middle REEs (MREE). On the other hand, HREE exhibit poor association with all other REEs except La and Gd. Pr, Nd, Sm, Eu, Gd and Tb exhibit an expected strong coherence (r= > +0.8). The association of REEs with major mineral phases in the crust is also very complex (Table 2) and deviate from the generally expected association with Fe-hydroxides in hydrogenous manganese nodules (Glasby et al., 1987) and crusts (De Carlo, 1991). We therefore obtained separate correlation matrices for the CFA enriched intermediate and older zones (Table 3) and the Fe-Mn oxide enriched younger zone to resolve the complex associations of the REEs in the total crust.

From La,/Yb, ratios (Table 1 and Fig. 6) and shale-normalised REE patterns with depth in the crust (Fig. 7), the above interpretation becomes clear. LaJYb, ratios with depth exhibit two dis- tinct trends reversing at a depth of 26 mm corre- sponding approximately to the end of Miocene. La enrichment over Yb is evident in the older zone whereas the opposite occurs in the younger zone. In CFA enriched zone therefore La is preferentially incorporated into the crust via a phosphatic phase even though HREE endmembers also exhibit sym- pathetic association with this phase (Table 3). A strong positive association of La,/Yb, with Mn in CFA-enriched zones (Y = 0.79; Table 3) probably reflects the influence of Mn-oxide phase in the fractionation of trivalent REE. The above observa- tions suggest that the crust mineralogy and major element composition which have evolved in response to varying OMZ conditions have played a decisive role in the fractionation of the triva- lent REEs.

In Fe-Mn oxide enriched younger zone, the normally expected coherent behaviour of all 3 + REE is evident (I= > +0.8). These elements are

(d) Ce-anomaly: The Ce-anomalies were calcu- lated as 2(Ce,)/( La, + Pr,), where n indicates shale- normalized concentration. Despite large variation in chemical and mineralogical composition within the crust, the Ce-anomalies have remained positive (Fig. 7) and vary between 2.8 and 5.5 (Table 1 and Fig. 6). Ce exhibits strong positive association (> 0.7) with Mn, Co and Mn/Fe (Mn-oxide phase) and insignificant association with all other REE and CFA phase in multimineral older to intermedi- ate zone (Table 3). The behaviour of Ce confirms

V. K. Banakar et al./Marine Geology 136 (1997) 299-315

Table 1

Depth distribution of major, trace and REE elements in the crust from the Afanasiy-Nikitin seamount, Central Indian Basin

Depth G. Rate Age ResX Ti Mn Fe % Ca P Co Ni Cu Ba SC Mn/ La

0nm) (mm/Ma) (Ma) (%I (8) (8) (8) (%o) (8) (%) (%) (%) (ppm) (ppm) Fe (ppm)

00-02 2.6 0.77 3.8 0.55 19.78 18.69 2.52 0.54 0.70 0.27 0.016 1121 6.5 1.06 266 02-04 3.0 1.44 2.1 0.56 19.21 16.78 2.41 0.49 0.65 0.27 0.020 1199 6.3 1.14 255 04-06 3.1 2.08 1.3 0.63 20.38 17.85 2.34 0.45 0.64 0.30 0.025 1267 5.5 1.14 243 06-09 3.0 3.06 3.5 0.73 22.75 17.09 2.32 0.41 0.63 0.39 0.038 1416 4.9 1.33 203 09-11 3.3 3.67 1.4 0.75 21.74 15.80 2.26 0.35 0.63 0.43 0.047 1370 4.2 1.38 187 11-13 3.1 4.31 1.4 0.75 21.83 15.01 2.35 0.35 0.64 0.41 0.051 1379 4.1 1.45 190 13-15 3.7 4.86 1.5 0.79 21.72 16.06 2.31 0.36 0.58 0.38 0.052 1446 4.5 1.35 186 15-17 4.0 5.36 1.6 0.79 21.70 16.77 2.42 0.37 0.55 0.39 0.055 1477 3.9 1.29 174 17-19 4.7 5.78 1.1 0.74 20.74 16.12 2.62 0.58 0.51 0.36 0.053 1391 4.3 1.29 202 19-21 8.5 6.02 11.5 1.06 12.76 14.56 6.28 2.16 0.28 0.31 0.075 1620 7.7 0.88 214 21-24 12.8 6.25 18.5 1.49 9.65 13.68 7.16 2.93 0.19 0.24 0.073 2003 9.8 0.71 239 24-26 12.8 6.41 14.9 1.07 8.57 11.50 11.70 5.19 0.19 0.15 0.047 1675 7.7 0.74 240 26-28 6.4 6.72 5.1 0.79 13.29 10.64 12.81 4.77 0.36 0.22 0.051 1778 6.3 1.25 280

28-30 4.6 7.16 1.2 0.73 19.08 9.84 10.98 3.59 0.51 0.35 0.065 1851 7.3 1.94 398 30-32 3.7 7.70 1.0 0.62 22.25 9.45 8.54 2.72 0.59 0.43 0.082 2231 3.7 2.35 256 32-34 2.7 8.45 2. I 0.60 23.66 10.18 6.27 1.77 0.71 0.45 0.074 2323 2.6 2.32 215 34-36 2.3 9.33 2.4 0.50 21.87 10.14 7.12 2.09 0.78 0.37 0.052 2278 3.0 2.16 240 36-38 3.2 9.95 1.3 0.55 21.11 11.10 8.22 2.42 0.63 0.34 0.039 2332 2.6 1.90 217 38-41 5.1 10.54 1.5 0.67 20.30 11.14 9.04 2.92 0.49 0.30 0.035 2422 3.0 1.78 284

G. rate=Growth rate, Resn = Residue, T REE= Total REE, cc = shale normalised ratio, Ce* = Cerium-anomaly,. These value

on dry basis.

Table 2

Interelemental associations in the entire crust. (n = 19, level of significance at 99% is 0.55)

Ti Fe Mn Ca P Res” Co Ni Cu Ba SC La Ce

Ti

Fe

Mn

Ca

P

Resff

co

Ni

cu

Ba

SC

La

Ce

Pr

Nd

Sm

Eu

Gd

Tb

Yb

Lu

3+REE

LaJYb, Ce*

Mn/Fe

1.00

0.01

-0.79

0.21

0.34

0.88

-0.87

-0.43

0.44

- 0.02

0.73

-0.08

-0.60

-0.46

-0.35

-0.33

-0.16

-0.00

-0.28

0.22

0.28

-0.21

-0.37

-0.40

-0.64

1 .oo

0.14 1 .oo

-0.85 -0.52 1.00

-0.77 -0.66 0.98

-0.07 -0.90 0.34

0.17 0.91 -0.51

-0.00 0.83 -0.44

-0.55 -0.17 0.36

-0.87 0.00 0.66

0.17 -0.82 0.21

-0.45 -0.21 0.61

-0.64 0.51 0.40

0.30 0.28 -0.26

0.35 0.18 -0.26

0.57 0.27 -0.52

0.70 0.09 -0.55

0.57 -0.13 -0.32

0.3 I 0.08 -0.18

0.21 -0.32 0.07

0.18 -0.34 0.10

-0.01 -0.05 0.18

-0.73 0.13 0.58

-0.39 0.56 0.05

-0.62 0.67 0.21

1.00 0.49

-0.63

-0.56

0.34

0.57

0.34

0.56

0.24

-0.31

-0.30

-0.53

-0.52

-0.27

-0.20

0.12

0.15

0.13

0.46

-0.05

0.04

1.00

-0.83

-0.65

0.28

0.05

0.77

-0.00

-0.61

-0.30

-0.22

-0.23

-0.07

0.08

-0.12

0.15

0.17

-0.09

-0.17

-0.52

-0.63

1.00

0.65

-0.36

-0.06

-0.72

-0.13

0.39

0.46

0.35

0.44

0.27

0.05

0.25

-0.28

-0.32

0.09

0.18

0.36

0.60

1.00

0.32

0.12

-0.68

-0.34

0.40

-0.07

-0.14

-0.02

-0.17

-0.35

-0.25

-0.33

-0.32

-0.30

-0.01

0.65

0.66

1 .oo 0.48

0.07

-0.00

0.11

-0.53

-0.54

-0.59

-0.58

-0.50

-0.56

-0.09

-0.04

-0.32

0.05

0.26

0.29

1.00

-0.33

0.25

0.63

-0.18

-0.27

- 0.45

-0.62

-0.61

-0.18

-0.51

- 0.49

-0.07

0.86

0.44

0.65

1.00

0.34

-0.63

0.04

0.19

0.13

0.33

0.57

0.21

0.70

0.69

0.37

-0.42

-0.84

-0.73

1.00 0.43 1.00

0.44 0.09

0.49 0.01

0.18 -0.17

0.16 -0.38

0.37 -0.40

0.46 0.00

0.60 -0.22

0.57 -0.23

0.85 0.19

0.40 0.72

-0.34 0.64

0.20 0.86

Resn= Residue, Ce* = Cerium-anomaly, La/Yb = Shale normalised ratio

pp. 309-3 10

Ce Pr Nd Sm Eu Gd Tb Yb Lu T REE 3+REE La/ Ce*

@pm) @pm) (ppm) (wm) (ppm) (wm) (ppm) Qwm) (wm) @pm) (ppm) Yb”

1409 51.4 213 43.5 12.4 53.1 8.40 23.1 3.64 2089 680 0.9 2.8 1391 48.8 192 40.1 11.2 47.8 1.62 23.0 3.61 2020 629 0.95 3.0 1439 45.2 183 35.8 10.2 42.5 6.85 22.4 3.61 2032 593 0.93 3.3 1579 36.0 143 28.6 8.2 35.8 5.73 20.5 3.42 2063 484 0.85 4.5 1621 30.8 129 26.5 7.3 30.1 5.23 20.3 3.36 2063 1440 0.79 5.1 1541 31.5 125 26.8 7.3 30.8 4.95 19.6 3.24 1980 439 0.83 4.8 1444 36.8 136 27.8 7.9 33.0 5.42 19.2 3.19 1899 455 0.83 4.3 1387 34.8 131 28.1 7.5 31.1 5.25 18.8 3.20 1821 434 0.79 4.4 1592 33.7 140 21.3 7.9 30.3 5.46 19.3 3.21 2055 463 0.90 4.6 1218 35.0 144 21.7 8.3 38.0 5.83 20.6 3.25 1715 497 0.89 3.4 1046 34.3 144 26.7 8.2 37.2 6.14 20.9 3.53 1566 520 0.98 2.7 1313 27.7 111 18.8 6.1 29.5 4.86 22.3 3.80 1777 464 0.92 3.7 1748 28.2 119 19.0 5.6 30.1 4.93 21.2 3.59 2260 512 1.13 4.4 2057 45.0 182 30.0 8.5 42.6 6.68 31.4 5.42 2807 750 1.08 3.5 2053 36.8 134 24.3 6.3 26.6 5.83 18.5 2.98 2564 511 1.18 5.0 1977 35.2 138 26.2 6.6 26.1 5.29 14.5 2.24 2446 469 1.27 5.5 1782 41.3 148 25.6 7.2 29.0 5.58 15.0 2.30 2296 513 1.37 4.3 1873 35.8 138 24.8 6.3 24.9 6.13 13.0 2.03 2341 468 1.43 5.1 2271 43.1 163 29.2 7.3 30.5 6.85 16.4 2.52 2854 583 1.48 4.9

s are

Pr Nd Sm Eu Gd Tb Yb Lu 3+REE La,/Yb, Ce* Mn/Fe

1.00 0.96 0.89 0.83 0.77 0.93 0.28 0.18 0.80 0.17

-0.49 0.08

1.00 0.90 1.00 0.89 0.93

0.85 0.85 0.95 0.87 0.42 0.32 0.31 0.21 0.86 0.64 0.08 -0.13

-0.58 -0.54 -0.04 -0.14

1.00 0.93 0.84 0.45 0.36 0.63

-0.31 -0.71 -0.37

1 .oo 0.80 1 .oo

0.68 0.32 1.00 0.59 0.22 0.99 1.00 / 0.75 0.81 0.65 0.58 1 .oo

-0.34 0.16 -0.47 -0.50 0.20 1 .oo

-0.83 -0.58 -0.64 -0.58 -0.59 0.33 1.00

-0.46 -0.08 -0.39 -0.39 -0.03 0.66 0.66 1.00

pp. 311-312 V.K. Banakar et al.IMarine Geology 136 (1997) 299-315

Table 3

Interelemental associations in the phosphate enriched zone of the crust. (n = 10, level of significance at 99% is 0

Ti Fe Mn Ca P Res# Co Ni Cu Ba SC L

Ti 1 .oo Fe 0.78 Mn -0.88 Ca -0.00 P 0.31 Resn 0.95 co -0.89 Ni -0.64 CU 0.28 Ba -0.59 SC 0.90 La -0.11

Ce -0.85

Pr -0.42

Nd -0.18

Sm -0.08

Eu 0.39

Gd 0.54

Tb -0.06

Yb 0.38

Lu 0.40

3+ REE -0.08

LaJYb, -0.76

Ce* -0.83

Mn/Fe -0.88

1.00

-0.68

-0.34

-0.08

0.78

-0.69

-0.46

0.14

-0.47

0.62

-0.40

-0.80

-0.28

-0.09

0.12

0.47

0.44

0.05

0.04

0.02

0.25

-0.54

-0.64

-0.80

1.00

-0.34

-0.63

-0.90

0.95

0.89

-0.00

0.79 -0.88

0.07

0.83

0.62

0.41

0.42

-0.11

-0.43

0.31

- 0.43

-0.45

0.14

0.79

0.78

0.97

1.00 0.92

-0.02

-0.33

-0.58

-0.41

-0.42 0.25

0.57

0.14

-0.30

-0.22

-0.55

-0.41

0.08

-0.23

0.53

0.57

0.31

-0.16

-0.09

-0.21

1 .oo

0.32 1.00

-0.60 -0.85

-0.79 -0.70

-0.33 0.19

-0.59 -0.61

0.48 0.82

0.37 -0.31

-0.18 -0.94

-0.51 -0.55

-0.40 -0.38

-0.66 -0.26

-0.36 0.24

0.13 0.36

-0.37 -0.27

0.54 0.23

0.58 0.25

0.12 -0.29

-0.41 -0.76

-0.31 -0.77

-0.51 -0.91

1 .oo

0.82

-0.06

0.75

-0.87

-0.00

0.73 0.53

0.30

0.28

-0.17

-0.48

0.14

-0.48

-0.50

0.04

0.77

0.73

0.94

1.00

0.42

0.62

-0.65

-0.02

0.58

0.56

0.41

0.52

0.10

-0.24

0.27

-0.32

-0.36

0.10

0.47

0.58

0.87

1 .oo

-0.22 0.31

-0.05

-0.26

-0.02

0.04

0.21

0.33

0.30

-0.08

0.25

0.23

0.02

-0.52

-0.23

0.05

1.00

-0.82 1.00

-0.17 0.21

0.64 -0.77

0.49 -0.34

0.26 -0.06

0.34 -0.10

-0.17 0.44

-0.58 0.74

0.37 -0.07

-0.69 0.71

-0.69 0.72

-0.09 0.20

0.89 -0.88 -

0.70 -0.93 -

0.74 -0.83

Rest= Residue, Ce* = Cerium - anomaly, La,/Yb, = Shale normalised ratio.

71)

a Ce Pr Nd Sm Eu Gd Tb Yb LU 3 +REE LaJYb, Ce* Mn/Fe

1.00 0.42 1 .oo 0.51 0.57 1.00

0.63 0.41 0.93 1.00 0.31 0.28 0.86 0.91 1.00 0.32 -0.24 0.60 0.76 0.81 0.58 -0.34 0.30 0.57 0.48 0.45 0.36 0.82 0.87 0.85 0.78 -0.16 0.09 0.32 0.09 0.77 -0.17 0.05 0.28 0.04 0.95 0.38 0.69 0.82 0. 57 0.03 0.76 0.48 0.26 0.24 0.22 0.76 0.09 -0.13 -0.08

0.16 0.85 0.57 0.35 0.31

1.00

0.83 0.65

0.44 0.39 0.57

-0.27 - 0.63

-0.18

1.00 0.42 1.00

0.81 0.15 1.00 0.79 0.12 0.99 1.00 0.70 0.65 0.73 0.71 1.00

-0.58 0.34 -0.64 -0.64 -0.00 1.00 -0.81 -0.09 -0.65 -0.65 -0.28 0.75 1.00 -0.43 0.22 -0.33 -0.35 0.18 0.12 0.76 1.00

V. K. Banakar et aLlMarine Geology 136 (1997) 299-315 313

28-30mm ,” $

!&I L/J- * 20 \

w 10 30-32mm

-1 1A

: :1- ,w ; 20

IO 32-34mm

I IL/--

5 3

9-llmm L L II-13 mm

Fig. 7. NASC-normal&d REE pattern with depth in the crust. Note the positive Ce-anomalies throughout the crust. 0 corres- ponds to the surface of the crust.

that, Mn-oxide phase is the only carrier of oxidised Ce, probably analogous to the mechanism of Co incorporation into hydrogenous crust (Halbach et al., 1983). The positive Ce-anomalies indicate

fairly oxic water (suboxic) throughout the period represented by the crust. However, random varia- tion in the Ce-anomaly with depth in the crust (Figs, 6 and 7) does not allow us to draw any conclusion regarding the fluctuations in the inten- sity of palaeo-OMZ.

5. summary

The Fe-Mn crust and its substrate from the upper flank of the Afanasiy-Nikitin seamount have recorded Neogene oceanographic variations. On one hand, the calcite cement in the substrate exhibiting subaerial fresh-water phreatic neo- morphism reflects the Oligocene global sea-level drop. On the other hand, mineralogical and com- positional characteristics suggest, (a) late Miocene intensified OMZ due to high productivity coupled with enhanced supply of lithogenous detritus at the close of Miocene has resulted in multimineral crust accretion and, (b) subsequent contraction of OMZ coupled with drastic reduction in the input of lithogenous detritus in the early Pliocene has favoured the accretion of almost pure hydrogenous crust in a less oxygen depleted OMZ.

High residue contents in the intermediate zone of the crust indicates an increased input of litho- genous detritus to NC10 at the close of Miocene probably resulting from the increased erosion of the Himalayas.

The availability of several potential REE carrier phases (CFA, Mn-oxide and Fe-hydroxide) appears to be responsible for the observed incoher- ant behaviour of REEs in the crust. Contemporaneous accretion of CFA and Fe-Mn oxide during the intensified OMZ in the late Miocene has resulted in random fractionation among strictly trivalent REEL

“Be and Sr isotope dating of the studied crust would help in a better understanding of palaeocea- nographic conditions in the NCIO.

Acknowledgements

We thank E. Desa, Director, for permission to publish this manuscript and R.R. Nair for sugges-

314 K’.K. Banakar et al./Marine Geology 136 (1997) 299-315

tions. Valuable suggestions by G.R. Dickens, G.P. Glasby and anonymous referees were of great ben- efit. We also thank M.S. Mannikeri, K.S. Krishna, P.C. Rao and K.A.K. Raju for discussions. The assistance from Shirsat, Javali, Chavan, Uchil and Prabhu is acknowledged. Financial support was from the Department of Ocean Development, Government of India under the project Polymetallic Nodules. This is N.I.O. Contribution No. 2515.

References

Banakar, V.K. and Borole, D.V., 1991. Depth profiles of

Z3DThcXCC~., transition metals and mineralogy of ferromanga- nese crusts of the Central Indian Basin and implications for palaeoceanographic influence on crust genesis. Chem. Geol. (Isot. Geosci. Sect.), 94: 33-44.

Bathurst, R.G.C., 1975. Carbonate Sediments and their Diagen- esis. Elsevier, Amsterdam, 658 pp.

Bertram, C.J. and Elderfield, H., 1993. The geochemical balance of the rare earth elements and neodymium isotopes in the ocean. Geochim. Cosmochim. Acta, 57: 1957-1986.

Bonatti, E.T., Kramer, T. and Rydell, H., 1972. Classification and genesis of submarine manganese deposits. In: D.R.Hom (Editor), Ferromanganese Deposits on the Ocean Floor. NSF, Washington, DC, pp. 149-166.

Boulegue, J. and Mariotti. A., 1990. Carbonate cement and fluid circulation in intraplate deformation. In: J.R. Cochran and D.A.V. Stow et al. (Editors), Proc. ODP Sci. Results, 116: 1355140.

Calvert, SE., Piper, D.Z. and Baedecker, P.A., 1987. Geochem- istry of rare earth elements in ferromanganese nodules from DOMES site A, northern equatorial Pacific. Geochim. Cos- mochim. Acta, 51: 2331-2338.

Cande, S.C. and Kent, D.V., 1992. A new geomagnetic polarity time scale for the late Cretaceous and Cenozoic. J. Geophys. Res., 97: 13,917-13,951.

De Carlo, E.H., 1990. Separation of lanthanide series elements in marine Fe-Mn crusts by ion-exchange chromatography and determination by ICP-AES. Separ. Sci. Tech., 25: 781-795.

De Carlo, E.H., 1991. Palaeoceanographic implications of rare earth element variability within a Fe-Mn crust from the Central Pacific ocean. Mar. Geol., 98: 449-467.

De Carlo, E.H., McMurtry, G.M. and Kim, K.H., 1987. Geo- chemistry of ferromanganese crusts from the Hawaiian archipelago-I. Northern survey areas. Deep-Sea Res., 33: 44-467.

Dickens, G.R. and Owen, R.M., 1994. Late Miocene-Early Pliocene manganese redirection in the Central Indian Ocean: expansion of the intermediate water oxygen minimum zone. Paleoceanography, 9: 169-18 1.

Dymond, J., Suess, E. and Lyle, M., 1992. Barium in deep-sea sediment: A geochemical proxy for paleoproductivity. Paleoceanography, 7: 163-181.

Elderfield, H., 1988. The oceanic chemistry of rare earth ele- ments. Philos. Trans. R. Sot. London, 325: 105-126.

Eisenhauer, A., Gogen, K., Pemicka, E. and Mangini, A., 1992. Climatic influence on the growth rates of the Mn-crusts during the late Quatemary. Earth Planet. Sci. Lett., 109: 25-36.

Elderfield, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 1981. Rare earth element geochemistry of oceanic ferro- manganese nodules and associated sediments. Geochim. Cosmochim. Acta, 45: 513-538.

Glasby, G.P., Gwozdz, R., Kunzendorf, H., Friedrich, G. and Thijssen, T., 1987. The distribution of rare earth and minor elements in manganese nodules and sediment from equato- rial and SW Pacific. Lithos, 20: 97-113.

Halbach, P., Segl, M., Puteanus, D. and Mangini, A., 1983. Relationship between Co-fluxes and growth rates in ferro- manganese deposits from the central Pacific seamount areas. Nature, 304: 716-719.

Haq, B.U., Hardenbol, J. and Vail, P.R., 1987. Chronology of the fluctuating sea levels since the Triassic. Science, 235: 1156-l 167.

Hein, J.R., Schulz, MS. and Gein, L.M., 1992a. Central Pacific cobalt rich ferromanganese crusts: historical perspective and regional variability. In: B.H. Keating and B.R. Bolton (Edi- tors), Geology and Offshore Mineral Resources of the Central Pacific Basin, Circum-Pac. Comic. Energy Miner. Resour. (Earth. Sci. Ser., 15). Springer, New York, pp. 261-283.

Hein. J.R., Bohrson, W.A., Schulz, M.S., Noble, M. and Clague, D.A., 1992b. Variation in the fine scale composition of a central Pacific ferromanganese crust. Paleoceanography, 7: 63377.

Kamer, G.D. and Weissel, J.K., 1990. Compressional deforma- tion of oceanic lithosphere in the central Indian ocean: why it is where it is. In: J.R. Cochran and D.A.V. Stow et al. (Editors), Proc. ODP Sci. Results, 116: 279-290.

Kashintsev, G.L., Peyve, A.A. and Rudnik, G.B., 1987. Subal- kaline volcanism in the mid Indian ocean basin. Int. Geol. Rev., 73-80.

Koschinsky, A. and Halbach, P., 1995. Sequential leaching of marine ferromanganese precipitates: Genetic implications. Geochim. Cosmochim. Acta, 59: 5 113-5132.

Levchenko, O.V., Milanovskiy, V.Y. and Popov, A.A., 1993. A sediment thickness map and the tectonics of the distal Bengal fan. Oceanology, 33: 2322238.

Muir-Wood, H.M., Stehli, F.G., Elliot, G.F. and Hatai, K., 1965. Terebratulida. In: R.C.Moore (Editor), Treatise on Invertebrate Palaeontology, Brachiopoda, part H. Geol. Sot. Am., pp. 728-856.

Murray, R.W., Buchholtzten Brink, M.R., Gerlach, D.C., Ruess, G.P. and Jones, D.L., 1992. Interoceanic variation in the rare earth, major and trace element depositional chemis- try of chert: perspective gained from the DSDP and ODP records. Geochim. Cosmochii. Acta, 56: 1897-1913.

V.K. Banakar et aLlMarine Geology 136 (1997) 299-315 315

Nath, B.N., Balaram, V., Sudhakar, M. and Pluger, W.L., 1992. Rare earth element geochemistry of ferromanganese deposits from the Indian ocean. Mar. Chem., 38: 185-208.

Parson, B. and Sclater, J.R., 1977. An analyses of the variation of ocean floor bathymetry and heatflow with age. J. Geo- phys. Res., 82: 803-827.

Pattan, J.N. and Banakar, V.K., 1993. Rare earth element dis- tribution and behaviour in buried manganese nodules from the Central Indian Basin. Mar. Geol., 112: 303-312.

Paul, J., Singh, R.N., Subrahmanyam, C. and Drolia, R.K., 1990. Emplacement of Afanasiy-Nikitin seamount based on transfer function analyses of gravity and bathymetry data. Earth Planet. Sci. Lett., 96: 419426.

Peterson, L.C., Murray, D.W., Ehrmann, W.U. and Hempel, P., 1992. Cenozoic carbonate accumulation and compensa- tion depth changes in the Indian ocean. In: R.A. Duncan et al. (Editors), Synthesis of Results from Scientific Drilling in the Indian ocean. AGU Monogr., 70: 311-334.

Puteanus, D. and Halbach, P., 1988. Correlation of Co-concentration and growth rates-a method for age deter- mination of ferromanganese crusts. Chem. Geol., 69: 73-85.

Rao, V.P. and Burnett, W.C., 1990. Phosphatic rocks and man- ganese crusts from seamounts in the EEZ of Kirabati and Tuvulu, Central Pacific Ocean. In: B.H. Keating and B.R. Bolton (Editors), Geology and Offshore Mineral Resources of the Central Pacific Basin. Circum-Pac. Count. Energy Miner. Resour. (Earth Sci. Ser., 15). Springer, New York, pp. 285-296.

Rao, V.P. and Nair, R.R., 1994. A re-valuation of palaeocli- matic conditions during the Pleistocene and Holocene from the western continental shelf of India, evidences from pet- rology of the limestones. In: B.N. Desai (Editor), Oceanog- raphy of the Indian Ocean. Natl. Inst. Oceanogr. India, pp. 423-437.

Rea, D.K., Dehn, J., Driscoll, M.W., Farrel, J.W., Jeneck, T.R.,

Owen, R.M., Pospichal, J.J. and Resiwati, P., 1990. Palaeo- ceanography of the eastern Indian ocean from ODP leg 121, drilling at Broken ridge. Geol. Sot. Am. Bull., 102: 679-690.~~. 423-437.

Rea, D.K., 1992. Delivery of Himalayan sediment to the north- em Indian ocean and its relation to global climate, sea level, uplift and sea water strontium. In: R.A. Duncan et al. (Edi- tors), Synthesis of Results from Scientific Drilling in the Indian Ocean. AGU Monogr., 70: 387-402.

Royer, J.Y., Sclater, J.G., Sandwell, D.T., Cande, S.C., Schlich, R., Munschy, M., Dyment, J., Fischer, R.L., Muller, D., Coffin, M.F., Patriat, P. and Bergh, H.W., 1992. Indian ocean plate reconstruction since the late Jurassic. In: R.A. Duncan et al. (Editors), Synthesis of Results from Scientific Drilling in the Indian Ocean. AGU Monogr., 70: 471-475.

Rudenko, M.V., 1994. Morphostructure of the Afanasiy-Niki- tin seamount. Oceanology, 33: 525-530.

Schlager, W. and James, N.P., 1978. Low magnesium calcite limestones forming at the deep seafloor, tongue of the ocean, Bahamas. Sedimentology, 25: 675-702.

Segl, M., Mangini, A., Bonani, G., Hofman, H.J., Nessi, M., Suter, M., Woltli, W., Friedrich, G., Pluger, W.L., Wiechow- ski, A. and Beer, J., 1984. iaBe dating of a manganese crust from central north Pa&c and implication for ocean palaeo- circulation. Nature, 309: 540-543.

Sholkovitz, E.R., Church, T.M. and Arimoto, R., 1993. Rare earth element composition of precipitation, precipitation particles and aerosols. J. Geophys. Res., 98: 20,587-20,599.

Slansky, M., 1986. Geology of Sedimentary Phosphates. North Oxford Acad. Publ., London, p. 210.

Usui, A., Nashimura, A. and Iizasa, K., 1993. Submersible observations of manganese nodules and crust deposits on the Tenpo seamount, northwestern Pa&c. Mar. Geores. Geotech., 11: 262-291.