Read doi:10.1016/j.palaeo.2004.11.007 text version

Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303 ­ 332

Coastal ecosystem responses to late stage Deccan Trap volcanism: the post K­T boundary (Danian) palynofacies of Mumbai (Bombay), west India

J.A. Crippsa,*, M. Widdowsonb, R.A. Spicerb, D.W. Jolleyc


School of Earth Sciences and Geography, Kingston University, Kingston-upon-Thames, KT1 2EE, United Kingdom b Department of Earth Sciences, The Open University, Milton Keynes, MK7 6AA, United Kingdom c Centre for Palynology, University of Sheffield, Sheffield, S3 7HF, United Kingdom Received 24 March 2004; received in revised form 23 August 2004; accepted 12 November 2004

Abstract The Deccan Trap continental flood basalt eruptions of India occurred c. 67­63 Ma, thus spanning the Cretaceous­Tertiary boundary (65 Ma). Deccan eruptions were coeval with an interval of profound global environmental and climatic changes and widespread extinctions, and this timing has sparked controversy regarding the relative influence of Deccan volcanism upon endCretaceous catastrophic events. If Deccan Trap activity was capable of affecting global ecosystems, evidence should be present in proximal Indian sedimentary facies and their palaeontological contents. The impact of late stage Deccan volcanism upon biota inhabiting Mumbai (Bombay) Island's post K­T boundary lagoonal systems is documented here. Sediments (or bintertrappeansQ) which accumulated within these lagoons are preserved between Trap lavas that characterise the closing stages of this flood basalt episode. Mumbai Island Formation intertrappean faunal and floral communities are conspicuously distinct from those common to many pre K­T boundary, late Maastrichtian intertrappeans across the Deccan province. The latter sedimentary intercalations mostly developed in cognate semiarid, palustrine ecosystems; by contrast, those around Mumbai evolved in sheltered, peripheral marine settings, within subsiding continental margin basins unique to this late Deccan stage, and under an increasingly humid Danian climate. Geochemical analyses reveal that Mumbai sedimentation and diagenesis were intimately related to local explosive volcanic and regional intrusive activity at c. 65­63 Ma. Although tectonic and igneous events imprinted their signatures throughout these sedimentary formations, organisms usually sensitive to environmental perturbations, including frogs and turtles, thrived. Critically, palynofacies data demonstrate that, whilst plant material deposition was responsive to environmental shifts, there were no palpable declines in floral productivity following Mumbai pyroclastic discharges. Therefore, it is implausible that this late stage explosive volcanism influenced major ecosystem collapses globally. D 2004 Elsevier B.V. All rights reserved.

Keywords: K­T boundary; Deccan Traps (India); Flood basalt; Mass extinction; Palaeoecology; Palynofacies

* Corresponding author. Fax: +44 20 8547 7497. E-mail address: [email protected] (J.A. Cripps). 0031-0182/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2004.11.007


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

1. Introduction Continental flood basalt provinces are laterally extensive lava accumulations of substantial thickness and low topographic relief (Rampino and Stothers, 1988). India's dominantly tholeiitic Deccan Trap flood basalt province presently extends across approximately one sixth of the subcontinent, encompassing up to 106 km2 of its western portion (Deshmukh, 1982; Fig. 1). The basalts include Traps downfaulted into the Arabian Sea west of Mumbai (Bombay) and forming part of the Seychelles microcontinent (Tandon, 2002; Devey and Stephens, 1991), and possibly originally occupied a volume of up to 106 km3 prior to their erosion (Courtillot et al., 1986). The duration of the whole Deccan volcanic episode remains a polemic issue, and advocates exist for both a brief (b1 m.yrs., e.g., Duncan and Pyle, 1988; Hofmann et al., 2000) and extended (e.g., Widdowson et al., 2000; Sheth et al., 2001a) period of activity. This theme is particularly pertinent when assessing the effects of flood basalt volcanism upon local, regional and even global ecosystems. A rapid

emplacement of an entire flood basalt province would theoretically prove more detrimental than a series of events separated by protracted dormant intermissions. Proof of quiescent phases exists in the form of sedimentary sequences that accrued between the Traps. Subsequent extrusives often preserved these bintertrappeansQ, and evidence can be sought within them regarding the influence of volcanism upon sedimentary systems, microclimates and biota. Because substances released during mafic eruptions are less likely to reach potentially damaging stratospheric levels than those expelled by felsic volcanism, the effects of late stage, increasingly felsic, explosive Mumbai volcanism are of interest. Controversially, a study of massive, well-constrained pyroclastic events (Erwin and Vogel, 1992) found that these did not reduce the ecological diversities of land and marine ecosystems on regional or global scales, and hence were unlikely to be responsible for mass extinctions. A bolide impacting Mexico's Chicxulub platform (Hildebrand et al., 1991) is broadly accepted to have exacerbated, if not singularly forced, end Maastrichtian extinctions across the

Fig. 1. Present-day Deccan Trap outcrop extent. Major tectonic structures redrawn from Biswas (1991).

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


planet (e.g., Pope et al., 1994; Sweet et al., 1999; Vajda et al., 2001). The literature review we offer draws together c. 100 years of disparate observations, with the benefit of a much improved chronostratigraphic framework, and represents the most comprehensive overview yet produced on Mumbai sequences. Data presented here are placed within this context, to illustrate the ecology of a Deccan volcanic region towards the close of this flood basalt episode. This is one of the

first attempts to evaluate ecosystems within a flood basalt succession using an integrated palaeobotanical, geochemical, geochronological and sedimentological approach. A similar study was conducted for central India's Jabalpur region, near the Narmada­Tapti rift zone, and the Nagpur area to the south, by Tandon (2002; Fig. 1). Tandon's article described the environmental changes leading up to the onset of local Trap emplacement that are recorded in central Indian

Fig. 2. Mumbai District, including localities visited, adapted from Subbarao and Sukheswala (1979).


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 Table 1 (continued) Section Worli tunnel Worli tunnel Worli tunnel Worli tunnel Sample Wo 2736 Wo 2850 Wo 3128 Wo 3408 Description 2736 m west from organic-rich shale 2850 m west from organic-rich shale 3128 m west from organic-rich shale 3408 m west from organic-rich shale shaft, shaft, shaft, shaft,

Table 1 Intertrappean sample lithologies chosen for palynomorph analyses and additional techniques Section Bandra tunnel Bandra tunnel Bandra tunnel Bandra tunnel Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Jogeshwari Worli tunnel Worli tunnel Worli tunnel Worli tunnel Worli tunnel Worli tunnel Worli tunnel Sample B 2800 B 3000 B 3130 B 3510 Bom Bom Bom Bom Bom Bom Bom Bom Bom Bom Bom Bom 1/98 2/98 3/98 4/98 5/98 6/98 7/98 8/98 9/98 10/98 11/98 12/98 Description 2800 m From entrance: coaly layer 3000 m From entrance: dark, flat-laminated shale 3130 m From entrance: compact, flat-laminated shale 3510 m From entrance: dark shale with pyrite cubes Fairly coarse, carbonate-rich Dark, carbon-rich, laminated Coarse, pale and dark laminations Thick, carbon-rich, burrows, pyrite Rippled silt Predominantly coarse Tuff Fissile, laminated Tuff/calcareous mix Dark, carbon-rich Coarse, plainly bedded Dark shale and pale, coarser sediment interlaminated Dark shale From bdoggerQ layer with calcite veins Ash containing small white flecks Light olive-grey silt Trap basalt (top of section) Rippled, dark grey silt Finely laminated very dark grey silt Float crustacean claw Fragments from coarse bed, possible tuff Phlogopite-rich, ?rhyolitic tuff Slatey layers, flat-bedded, v.dark, ?organic-rich Volcanic bombs 2001 m west from shaft, organic-rich shale 2100 m west from shaft, organic-rich shale 2210 m west from shaft, organic-rich shale 2210 m west from shaft, organic-rich shale 2600 m west from shaft, organic-rich shale 2610 m west from shaft, organic-rich shale 2735 m west from shaft, organic-rich shale

Bom 13/98 Bom 15/98 Bom Bom Bom Bom Bom 16/98 17/98 18 /98 19/98 20/98

Bom 22/98 Bom 23/98 Bom 1/99 Bom 2/99 Bom 3/99 Wo 2001 Wo 2100 Wo 2210a Wo 2210b Wo 2600 Wo 2610 Wo 2735

Lameta Formation sediments. Here, topographic adjustments caused fluvial currents to redirect, and periodically submerged terrain to became increasingly subaerial. Although this dynamic landscape was influenced by regional volcanic activity, it was exploited by sauropod dinosaurs prior to the first local lava incursion (Tandon, 2002). The Mumbai peninsula is investigated by the present authors. Originally a series of islands (e.g., Bombay Island, Salsette Island), the landmass projects southwards into the Arabian Sea at c. 198 north (Fig. 2). Three intertrappean sections on the western side of the peninsula were investigated: an outcrop at Amboli quarry in Jogeshwari, and two tunnel cuttings excavated seawards from the coast, just south of Worli and near Bandra (Fig. 2). Both tunnels extend westward into the Arabian Sea, and samples were extracted along them between 2001 and 3408 m in the Worli tunnel, and 1890 and 3740 m in the Bandra tunnel (Table 1). Since completing fieldwork, the Amboli section has been demolished for housing construction. This work provides a graphic log and field summary of the lost section. A brief description of Amboli, Worli and Bandra lithologies is given in Table 1.

2. Geological setting 2.1. Stratigraphy and field relationships The Mumbai and Salsette Islands landmass comprises a linear depression bounded by easterly and westerly ridges (Sukheswala, 1956). Muddy sediments deposited in the central lowland dip 12­158 west, and lavas up to 258 west (Sheth et al., 2001a). A separate classification to the Deccan chemostratigraphy, established in the Western Ghats and now

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


covering much of the main Deccan province (MDP), exists for the distinct geochemistries of Mumbai intrusives and extrusives (Sethna, 1999; Table 2). The Amboli (Bom), Worli (Wo) and Bandra (B) intertrappean shale sections detailed here occur within the Mumbai Island Formation, the lowermost of the Salsette Subgroup (Table 3). Sethna (1999) placed this above the highest of the MDP, the Wai Subgroup. According to him, Worli intertrappeans occur stratigraphically above the Malabar Hill flow (Fig. 3). Sethna (1999) estimated this shale's thickness at c. 150 m, interrupted only by a 10-m tuffaceous breccia (hyaloclastite) horizon, and a 5-m basaltic layer. The nearby Bandra tunnel also runs through this sedimentary unit, and the onshore Amboli section possibly represents a lateral equivalent. Pandey and Agrawal (2000) detected several sedimentary basins offshore of Mumbai and in adjacent western Indian offshore areas, retaining India's largest hydrocarbon reserves (Gombos et al., 1995). Stratified intertrappeans in quarries around Jogeshwari (Fig. 2) have been intruded by a columnar jointed tholeiitic lopolith (Subbarao and Sukheswala, 1979) and are conformably overlain by a basaltic lava flow. The position of Jogeshwari exposures within the regional stratigraphy, and possible provincial north­ south correlations, are given in Fig. 4.

Table 2 Deccan chemostratigraphy from Mitchell and Widdowson (1991) Subgroup Salsette (4) Formation Manori (4) Madh­Utan Mumbai Island (4) Desur Panhala Mahabaleshwar (1) Ambenali (1) Poladpur (1) Bushe (2) Khandala (3) Bhimashankar (3) Thakurvadi (3) Neral (3) Igatpuri (3) Jawhar (3)

Table 3 Stratigraphical position of present samples within the Salsette Subgroup, after Sethna (1999) Subgroup Formation Geology Samples ­ Salsette Manori Formation Trachyte and Subgroup basalt intrusions Mahd­Utan Formation Mumbai Island Formation Rhyolite lava flows


Hyaloclastites, spilites, basalts and shales

Bom, Wo, B

Magnetostratigraphical correlations between Mumbai flows and the MDP volcanic pile have been attempted. Vandamme et al. (1991) and Vandamme and Courtillot (1992) detected a reversed-normal boundary obscured by a secondary palaeomagnetic component in some localities. These authors established that the changeover occurred at much lower altitudes than the typical 600-m elevation observed elsewhere in the Deccan (e.g., Western Ghats), and interpreted the Mumbai boundary to possibly represent a later, younger magnetic reversal. 2.2. Age An early Tertiary age was first assigned to uppermost Mumbai intertrappeans by Blanford (1867), and an inferred close affinity of Mumbai intertrappean biota with modern forms led Sukheswala (1956) to support this. However, Singh and Sahni (1996) found that several Mumbai taxa additionally occurred in intertrappeans as divergent as Kutch (Gujarat), Jabalpur (Madhya Pradesh), Nagpur (Maharashtra), Gurmatkal and Marepalli (Andhra Pradesh; Fig. 1), indicating correlations between all these sections. Mumbai ostracod assemblages were observed to have affinities with late Cretaceous and Palaeocene forms. The authors ultimately ascribed a Maastrichtian date, attributing contrasts between Mumbai and other Deccan facies to environment rather than age differences. Highly accurate radiometric dates of Mumbai extrusives recently obtained (e.g., Table 4) are closely comparable with those received for late stage MDP feeder dykes (Widdowson et al., 2000). Sheth et al. (2001a) argued that Mumbai volcanism continued for

Wai (3)

Lonavala (3) Kalsubai (3)

Data compiled from: (1) Cox and Hawkesworth (1984), (2) Cox and Hawkesworth (1985), (3) Beane et al. (1986) and (4) Sethna (1999). Initial Salsette Subgroup eruptions were coeval with Mahabaleshwar-Desur Formations of the Wai Subgroup.


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 3. Schematic section across Mumbai Island identifying the major lava flows, separated by intertrappeans (marked as bIQ), encountered in boreholes and detected outcropping at Sewri and Malabar Hill, from Sethna (1999).

z1 m.yrs. Hence, it strongly appears that Salsette Subgroup igneous activity was coeval with terminal Wai Subgroup eruptions along the Western Ghats, although the flow-types are not geochemically related. By this closing stage, the most intense and voluminous MDP lava formations had already erupted (Table 2). Locally restricted Mumbai Island magmatism directly proceeded the major K­T boundary global

extinctions, and represents the final throes of the Deccan flood basalt episode. 2.3. Tectonic setting Sukheswala (1956) determined that a narrow basin and common volcanic centres occurred along subsurface fracture zones, trending north­south across the

Fig. 4. Possible correlation of Mumbai province stratigraphy.

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


Table 4 Published ages of a variety of Deccan igneous rocks occurring around the Mumbai and Salsette Islands, in reverse chronological order (dates acquired from Amboli samples by M. Widdowson); wr=whole rock, pl=plagioclase Rock Basalt (tholeiite) Rhyolitic tuff Trachyte Trachyte Basalt (tholeiite) Intermediate rock Not specified Not specified Trachyte Rhyolite Basalt (tholeiite) Olivine nephelinite Basalt (tholeiite) Mugearite


Method Ar/39Ar (wr) 40 Ar/39Ar (wr) 40 Ar/39Ar (wr) 40 Ar/39Ar (wr) 40 Ar/39Ar (wr) 40 Ar/39Ar (wr) K­Ar (pl) K­Ar (pl) Rb­Sr (wr) Rb­Sr (wr) K­Ar (pl) 40 Ar/39Ar (wr) 40 Ar/39Ar (wr) K­Ar (wr)


Date (Ma) 64.55F0.59 64.64F0.39 60.4F0.3 61.8F0.3 60.5F1.2 62.4F1.0 60.2F2.5 62.8F3.0 ~60 61.5F1.9 88.8F4.0 72.0F6.9 74.1F3.3 38.7F0.9a

Confidence 2r 2r 2r 2r 2r Unspecified 1r 1r ­ Unspecified 1r ­ ­ 1r

Comments Sample Bom18/98 Sample Bom1/99 Unaltered sample Unaltered sample Unaltered sample From Salsette Island Unaltered sample Unaltered sample No clear isochron High initial 87Sr/86Sr Argon excess No plateau ages No plateau ages Altered sample

Reference Widdowson et al. (2000) Sheth et al. (2001b) Sheth et al. (2001a) Kaneoka et al. (1997) Vandamme et al. (1991) Lightfoot et al. (1987) Balasubrahmanyan and Snelling (1981) Kaneoka (1980) Kaneoka and Haramura (1973)

Age corrected with new decay constants by Vandamme et al. (1991).

Mumbai and Salsette Islands. A regional, ovalshaped, 12 km height by 35 km base diameter positive gravity anomaly, with its focus along the west coast of Salsette Island, coincides with an area of high heat flow (Negi et al., 1992, 1993; Fig. 1). Hooper (1999) and Sen (2001) inferred that mildly alkaline and tholeiitic dykes bearing mantle xenoliths, again trending roughly north­south, created this gravity high, and Sethna (2003) associated the Mumbai anomaly with intermediate and felsic igneous rocks underplated by gabbroic intrusive complexes. Vertical movements played a key role in shaping Mumbai Trap palaeoenvironments. Blanford (1872) proposed a mechanism which instigated alternating rising and sinking events across Mumbai Island, and structures across the district have recently been attributed to tectonic deformation (Widdowson, 1997; Sheth and Ray, 2002). North­south trending fractures through, and the block tilting of, offshore Mumbai basement rock have been related to the western margin of India rifting from Madagascar, then the Seychelles bank, respectively, before or during the Deccan volcanic episode (e.g., Devey and Lightfoot, 1986; Singh and Sahni, 1996). Inferring a different sequential order from flowmapping, Hooper (1990) concluded that the lithospheric thinning, shearing and rotation which produced the present regional westward dips only ensued after Reunion mantle plume emplacement, litho´ spheric doming and MDP eruptions. This crustal

extension arguably promoted the mantle upwarping that resulted in the Mumbai gravity anomaly (Dessai and Bertrand, 1995). Lightfoot et al. (1987) considered this to have triggered partial melting of lower crust gabbroic complexes and an associated production of trachytic magmas, whilst contamination from assimilated crust was debated to have generated the more acidic suites present. Negi et al. (1992) interpreted the Salsette Island gravity anomaly as a magma conduit, discrete from the main Deccan plume, which breached the continental margin fracture zone offshore of Mumbai. This fracture, and the Seychelles block detachment, were stated to be related to a bolide collision. Chatterjee and Rudra (1996) submitted the Mumbai High (Fig. 1) oilfield and Deccan intrusives as evidence of an impact (the bShiva craterQ), embroiling a putative offshore Mumbai meteorite strike with K­T boundary extinctions. Shale organic maturation was allegedly instigated by impact-induced lithospheric heating, and the offshore region, uplifted by earlier Deccan magma accumulation, sank in response (Pandey and Agrawal, 2000). Mumbai regional tectonic characteristics are more widely implied to be entirely products of terrestrial processes (e.g., Sethna, 2003; Table 5). Gombos et al. (1995) suggested that India's west coast hydrocarbon reserves resulted from a Mesozoic collapse of Proterozoic mobile belts into passive margin basins, during and following the rifting that produced the


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Table 5 Chronology of tectonic events influencing Mumbai Island Formation pyroclastic and sedimentary facies Stage Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Events Lithospheric doming above Reunion plume, ´ flood basalt activity across main Deccan province Rifting begins along previously existing N­S crustal fractures, crustal blocks tilted westward Development of shallow gulf as rifting and subsidence propagate, water invades depressions Magma upwells beneath thinned crust and intrudes into tensive crustal fractures Mumbai Island Formation explosive volcanism; shale and ash deposition into basin systems Intertrappeans buried as subsidence continues and thermally metamorphosed by intrusions Tertiary erosion onshore and deposition offshore isostatically enhances westward dips

Stage 6 Stage 7

Mumbai High fault block. Sedimentation into Mumbai High rifts was dominated by organic-rich shales, with continued subsidence promoting their thermal heating and maturation (Gombos et al., 1995). Widdowson (1997) attributed the current western Indian margin geomorphology to simultaneous onshore erosion and offshore deposition operating throughout the Tertiary. 2.4. Facies The crustal subsidence that accompanied Mumbai Island Formation activity represents the waning phase of Deccan activity (Singh and Sahni, 1996). Consequentially, Mumbai intertrappeans are generally much thicker than MDP sequences. An exceptionally thick shale overlying the Malabar Hill flow reflects a prolonged volcanic hiatus (Sethna, 1999; Fig. 3), and The Worli and Bandra tunnels cut into extensive, carbonaceous shales (Sethna, 1999). Sukheswala (1956) described the western ridge at Malabar and Worli as composed of a repetitive series of green and black ashes, and similar facies occur further north, around Jogeshwari (Sukheswala and Awate, 1957; Fig. 2). Volcanic and pyroclastic units were substantially reworked during repose phases, becoming increasingly clay and organic-rich, as reflected in a transition from greenish ashes and rhyolites to dark, fossiliferous shales in the Malabar and Worli hills of the western ridge (Sukheswala, 1956).

Structures including ripple marks prompted Sukheswala (1956) to advocate shallow lakes as likely depositional environments for the lowermost sediments. Oblong concretions of V10 cm diameter in a prominent ash bed exposed along Mumbai Island's western ridge were interpreted by this author to represent ash bombs which coalesced during pyroclastic eruptions, and a recurring subaqueous influence was deduced from the widespread occurrence of laminated beds. Deshmukh (1984) recognised that breccias had evolved from explosive volcanic activity, such volatility being enhanced by the invasion of water following subsidence. Sethna (1999) described most Mumbai district flow facies as at least partially subaqueous. Extrusive breccias in the Amboli section, Jogeshwari, are composed of basaltic and altered vesicular glass clasts in a fine- to medium-grained clay, carbonate and quartz-rich matrix. Their petrography indicated a spilitic origin to Tolia and Sethna (1990), the hyaloclastites having consolidated during phreatomagmatic basalt effusions. These authors did not detect volcanic bombs, finding infrequent subcircular objects possessing chilled margins to be pillow structures. The angular shapes of most volcanic fragments suggested to Singh (2000) that these underwent minimal aqueous transportation; consequentially, the eruptive centres themselves are likely to have occupied low-grounds. Sharma and Pandit (1998) assigned ignimbrite facies to cycles of felsic tuffs overlying intermediate to mafic pyroclastic flows in the Sasunavghar­ Juchundra area, c. 5 km north of Salsette Island. The greater pyroclastic content of such sequences around Mumbai than other Deccan fringe regions was regarded by Singh and Sahni (1996) to reflect a closer proximity to their volcanic source, their evolved chemistries pointing to the termination of Deccan events. Igneous, tectonic and hydrological activity greatly influenced Mumbai shale as well as ash facies. Amboli intertrappeans display a hardened, baked margin where they contact the tholeiitic lopolith, and elsewhere exhibit plastic deformation (Tolia and Sethna, 1990). Singh (2000) attributed shale baking to heat conducted from overlying lavas. Mumbai shales are indicative of sedimentation under waters with low oxygen concentrations (Singh and Sahni, 1996), as

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


reflected in pyrite precipitation along many carbonaceous laminations (Singh, 2000). However, occasional subaerial exposure led to desiccation and swamp formation under semiarid climes, as evidenced by calcite-filled rain prints and mud cracks (Singh, 2000). 2.5. Geochemistry Sukheswala (1956) identified pyroxenes and feldspars flanking calcite crystals in a Worli ash, and thus inferred a mafic chemistry. Partially decomposed feldspars, pyroxenes and biotite also occur in Jogeshwari tuffs, with calcite and quartz forming the major minerals here. Amboli hyaloclastites contain higher H2O and Na2O proportions than the local tholeiites, these enrichments in hydrous and alkali phases having been influenced by magma contacting water during its crystallisation according to Tolia and Sethna (1990). These authors recognised Amboli plagioclase to be a sodium-rich variety, and found that much of the calcite and quartz occurred as secondary minerals filling veins alongside zeolites. Metasomatism related to tectonism and intrusions is likely to have instigated zeolite precipitation across the Mumbai district (Sabale and Vishwakarma, 1996). Evidence of pyroclastic activity associated with terminal Deccan tensional regimes is preserved in the clay fractions of Mumbai shales. An X-ray diffraction (XRD) study of Amboli, Worli and Malabar intertrappean mineralogies (Singh, 2000) revealed matching mineral suites that indicated a mafic ash provenance for the shales' clastic components. Pyroxenes degraded, glass devitrified and smectitic clays evolved during reworking, the smectites producing few reflection angle peaks due to their weak crystal structure development (Singh, 2000). Smectites and chlorite constitute the most important Mumbai clays, and combine to form a mixed-layer superlattice. 2.6. Palaeontology Owen (1847) assigned frog remains within shales underlying the Malabar Hill Trap at Worli Hill the species Rana pusilla, although the fossil evidence for Maastrichtian Indian ranids has since been queried (Bossuyt and Milinkovitch, 2001). Turtles and molluscs from this section were detailed by Blanford

(1867), and additional species of frogs, the most abundant faunal element, by Chiplonkar (1940). Sukheswala (1956) extracted two Carteremys leithii freshwater Pelomedusidae turtle specimens, and a tooth later diagnosed as crocodilian (Singh and Sahni, 1996). According to Singh and Sahni (1996), preservation within the Mumbai shales is unique to the Deccan, being superior to that within most MDP intertrappeans. These authors examined the faunal component of sections at Worli Hill, Amboli and Malabar, unearthing Shweboemys (Carteremys) leithii skull and carapace fragments within the latter. This genus was further documented in MDP sediments at Nagpur, Marepalli and Kutch (Fig. 1). Similarly, the Mumbai ostracod genera Mongolianella, Altanicypris, Cypridea (Pseudocypridina), Timiriasevia and Cyprois were associated with those from MDP intertrappeans (e.g., Bhatia et al., 1990; Whatley et al., 2003). A new pelomedusoid turtle species, Sankuchemys sethnai, has recently been extracted at Amboli (Gaffney et al., 2003). Genera common to inland and marginal marine ecosystems signify that either lagoon waters were occasionally virtually freshwater, or that central Indian lakes tended towards brackish. However, Singh and Sahni (1996) emphasised that dinosaur and fish taxa, important in several widely distributed MDP localities, are entirely absent in the Mumbai shales (Table 6). The lack of fish was attributed to water turbidity or contamination, conditions frogs were capable of tolerating (Singh and Sahni, 1996), although turbid waters of modern coastlines are often colonised by fish. Even the Mumbai Leptodactylidae frog Indobatrachus was distinguished from MDP Pelobatidae and Discoglossidae forms (see also Khosla and Sahni, 2003, and references therein). The absence of some important MDP taxa around Mumbai, despite favourable preservation conditions, led Blanford (1867) to speculate that the cumulative effects of previous Deccan volcanism suppressed rainfall and damaged Mumbai environments to the extent that most MDP organisms lapsed into extinction. He interpreted poorly fossiliferous volcaniclastics low in the Malabar and Worli sequences to signify originally barren ecosystems, and suggested that Mumbai communities were a replacement biota to MDP fauna. Sukheswala (1956) reasoned that con-


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 Table 6 (continued) Organism Algae Acritarcha Botryococcus Dinoflagellate Zygnemataceae Poladpur Y Y Y Y Ambenali Y Y Y Y Mumbai Island ­ Y ­ ­

Table 6 Important organism groups in the Poladpur, Ambenali and Mumbai Island Formations (based upon a collation of results presented in Cripps, 2002 and references therein) Organism Dinosaur Crocodile Fish Poladpur Y ­ Y Y Y Y Y Y ­ Y Y Y Y Y ­ Y ­ Y ­ Y ­ Y Y Y Y Y Y Y Y ­ ­ ­ Y Y Y ­ Y Y Y Y Y Y Y Y Y ­ Y ­ ­ Y Ambenali Y Y Y Y ­ Y Y ­ Y Y Y Y Y ­ Y Y Y Y Y ­ ­ Y ­ ­ ­ Y ­ ­ ­ Y Y Y ­ Y Y Y Y Y Y Y Y Y Y Y Y ­ Y Y Y Y Mumbai Island ­ Y ­ ­ ­ ­ ­ ­ Y Y ­ ­ ­ ­ ­ Y ­ ­ Y ­ Y Y ­ ­ ­ Y ­ ­ ­ ­ Y ­ ­ ­ ­ ­ ­ Y ­ ­ ­ ­ ­ ­ ­ Y ­ ­ ­ Y

Apateodus Lepisosteus Phaerodus Pycnodus Ray Stephanodus

Turtle Frog Gastropod

Bivalve Ostracod





Lymnaea Paludina Physa Planorbis Unio Altanicypris Bisulocypris Candona Cypridea Cyprinotus Cypris Cyprois Drawinula Eucandona Metacyprois Mongolianella Mongolocypris Paracypretta Paraconadona Talicypridea Timiriasevia Harrisichara Microchara Peckichara Platychara Stephanochara Aquilapollenites Arecaceae ?Betulaceae ?Caprifoliaceae ?Mimosaceae ?Araucariaceae Bennettitaceae Ginkgoaceae Pinaceae Podocarpaceae Gleicheniaceae Osmundaceae Polypodiaceae Salviniaceae

temporaneous local, rather than preceding regional, volcanic activity generated a terrain inhospitable for Mumbai life. A thick basal greenish ash was thought to indicate an extended extrusive episode prior to a period of diminishing volcanism and community regeneration, represented by upper dark, fossiliferous shales. According to Mumbai Trap radiometric dates (Table 4), the diverse shale communities survived regional and global K­T boundary events. Bossuyt and Milinkovitch (2001) detailed archaeobatrachan frog lineages enduring the Deccan volcanic episode along the Indian island's peripheries, and thriving during the early Tertiary, notwithstanding their probable confinement along the western fringe by volcanism to the east and an ocean to the west. Although many frog groups are environmentally sensitive, some Leptodactylidae species have broad physiological tolerances, and today populate habitats undergoing ecological or climatic disturbances (Kaiser, 1997). 2.7. Palaeobotany Mumbai intertrappean plant megafossils are uncommon and distinct from those of the MDP (Blanford, 1867), but have similarly originated from land plants (Sukheswala, 1956). Bande et al. (1988) and Bande (1992) found limited Bambusaceae and Podocarpaceae wood, leaflets of possible Acacia (Leguminosae) affinity, and seeds similar to Artabotrys (Annonaceae). Megafloral remains are allochthonous within Mumbai basin facies, and the buoyancy of such organs as bamboo cane probably assisted their transportation. Leptodactylidae frog taxa that currently inhabit marine supra- to intertidal zones and consume saline marine food must regulate their osmotic balance (Abe and Bicudo, 1991). It thus seems plausible that Indobatrachus consumed terrigenous plant detritus washed down from vegetated

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


areas, a diet that ostensibly safeguarded the frogs from any effects of temporary productivity declines driven by volcanic disturbances (cf. Sheehan and Fastovsky, 1992). Palynofacies analyses are useful in combination with sedimentological investigations, potentially distinguishing environmental transitions before macroscopic change is visible (Tyson, 1985). An amalgamation of the ecology of organic matter (OM) producers, palynodebris transportation, decomposition prior to burial and alterations during diagenesis generates a sediment's palynofacies characteristics. According to Cross and Taggart (1982), the principal factors determining plant fossilisation are tissue durability, transportation distance, the existence and persistence of viable preservation sites, and sedimentation rates and consistency. No palynofacies analyses have previously been performed upon Deccan intertrappean floral material.

3. Data collection 3.1. Field observations (1) Amboli quarry, 19808V03WN; 072850 V30W 10 m E, a.s.l. exposes an intertrappean of z10 m thickness, dipping westward c. 88, terminating in a junction with basalt above (Fig. 5). Its base is obscured by the quarry floor (the underlying flow, occurring c. 3­4 m beneath ground level here, outcrops to the northeast). Sediments range from dark grey, flat-laminated shales, through silts, to pale grey, cross-rippled sands (the latter occurring exclusively around Jogeshwari). Coarse grains, rarely present along certain laminations, include well-rounded c. 0.4 mm diameter carbonate clasts and rounded quartz sands (e.g., Fig. 6b). Dark, laminated horizons (e.g., Bom 4/98 and Bom 12/98; Table 1) contain pyrite framboids. The majority of units are planar-bedded, although one chaotically deposited, coarser layer contains btabletsQ of flat-laminated sediment. Ripples of 0.1 cm amplitude by 2 cm wavelength traverse another upper bedding plane, and some ripple crests have been transformed into flame structures (e.g., Fig. 7c). Undulose upper bedding planes


frequently exhibit fine, laterally continuous organic drapes. A 1.22-m ash, Bom 1/99, forms a salient, continuous bed through the section's centre. This resistant unit yields virtually unaltered crystals the potassic mica phlogopite and quartz. Beneath, the uppermost fraction of Bom 8/98 consists of a series of fining-upwards beds. Fining-upwards cycles throughout this section tend to be continuous but thin, containing neither body nor trace fossils. However, small (1­2 cm) internal moulds of bivalves and gastropods occur sporadically elsewhere. An upper bedding plane exposed upon the quarry floor is pitted by common burrows (cf. Thalassinoides), these being virtually absent in higher beds. These are subhorizontal, smooth-walled, pellet back-filled, c. 1.5 cm diameter and 6 cm length, connecting at triple-junctions. Slightly oblate features of 1­1.5 cm diameter in Bom 6/ 98, viewed in cross section in the quarry face, initially appear to be higher, slightly compressed, vertical expressions of these horizontal traces. However, laminations cup underneath them and, when excavated, their true subspherical rather than cylindrical shape becomes apparent. Several ash beds are indistinctly stratified, either coarsening or fining-upwards. Some layers are dominated by grains, commonly feldspars, of up to 1 cm, horizontally aligned in thin, parallel bands. A coarse carbonate cement envelops the Bom 1/98 and Bom 2/98 matrices. Crystalline cement is particularly evident towards the uppermost basalt. Slickensides both follow and cross bedding planes. Sediments contacting the columnar lopolith exposed in the quarry face exhibit polygonal cracks. In sharp contrast with many MDP sections, no reddened ashes are present. The Worli and Bandra Tunnels are inaccessible, hence their overall sedimentological context is impossible to gauge. However, cuttings reveal that the tunnels pass through similar lithologies to those present in the Amboli section, except that the sediments generally lack cemented layers, being dominated by shale and OM (Table 1).


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 5. Amboli sedimentary summary log (for detailed log, refer to Cripps, 2002).

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


Two Amboli and two Worli tunnel samples were chosen for stable carbon isotope composition determination. Kerogen palynological residues (outlined next) were prepared for stable carbon isotope analyses by repeatedly centrifuging dry samples in 9:1 dichloromethane:methanol solvent. Stable isotope ratios were measured on an elemental analyser-isotope ratio-mass spectrometer. 3.3. Palynofacies A sediment's palynofacies is its content remaining after maceration in hydrochloric and hydrofluoric acids (Combaz, 1964). The desired end products of palynofacies maceration processes are slides clearly displaying an optimum number of phytoclasts (clasts of plant origin), with as little accompanying extraneous material as possible. Ideally, techniques employed should not alter the proportions of phytoclasts as they occur in their host sediment by biasing particular grain sizes or types. Standard palynological processing techniques to produce kerogen slides (Moore et al., 1991) were followed for the present study.

Fig. 6. Thin-section micrographs (plane-polarised light). (a) Finegrained clay and OM laminae undulating and bifurcating around coarser ash clasts and cement in silt sample Bom 3/98; (b) laminations compressed and distorted about a coarse, weathered pyroclast in ash sample Bom 16.

4. Results 4.1. Geochemistry The Amboli spherical clast Bom 3/99 is dominated by calcite, quartz, smectite and feldspar (Fig. 8a). A minor peak at 3.6 2 in the clay separate diffraction profile (Fig. 8b) denotes the presence of an ordered super-lattice, produced by two different minerals alternating regularly, constituting the mixedlayer clay corrensite. Peak positions confirm the super-lattice to be chlorite interleaved with a saponitic smectite (approximately 80:20 chlorite:smectite; Clayton, personal communication). A trace of kaolinite is evident in the whole-rock profile, although, interestingly, this clay is unusual in MDP intertrappeans (Cripps, 2002). XRF results reveal that, although Amboli ash and tuff chemistries vary considerably, all the samples possess elevated Na2O levels (Table 7). The two Amboli ashes analysed for stable carbon isotopic composition exhibit marginally lighter d 13C values

3.2. Geochemistry A thorough account of Mumbai clay mineralogy is given in Singh (2000). To provide comparison, the mineralogy of Bom 3/99, a spherical clast from Amboli quarry, was assayed by X-ray diffraction (XRD) for this work, after preparation using standard whole-rock and clay-separate methods (Hardy and Tucker, 1988). The clay separate was subjected to glycolation and heating, to distinguish between smectites, chlorites and kaolinites. Element concentrations were established using X-ray fluorescence spectroscopy (XRF). Analyses of major elements were performed on glass discs, and powder pellets were used for trace element analyses. Losses on ignition (LOI) were recorded to account for volatile contents.


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 7. Amboli section photographs. (a) Entire section, (b) typical organic-rich shale to marly sandstone bedding cycles, (c) flattened ripples on upper bedding plane of siltstone.

than the Worli shales (Table 8). The significance of these findings is discussed in Section 5. 4.2. Palynofacies Although significant volumes of organic residue remained after macerating Mumbai intertrappeans, palynomorphs supplied a negligible contribution. Spinizonocolpites palm pollen, Azolla water-fern massulae, Botryococcus algal colonies and various fungal spores were exceptionally logged in some shales. While this paucity means that a comprehensive palynological interpretation is unfeasible, similar lithologies through the Amboli, Worli and Bandra sections permit comparisons of their palynodebris

characteristics. Mumbai shales and silty sands are suited to palynofacies investigations due to their high concentrations of well-preserved, structured organic clasts. Seventeen Amboli (Bom), 11 Worli (Wo) and 4 Bandra (B) specimens were examined; samples were selected to typify the range of sediment types present (Table 1). Two hundred phytoclasts were logged for each sample, and grains allocated 1 of 16 designated microfloral categories (Table 9; Fig. 9). Palynodebris percentages are displayed at their stratigraphical positions through the Amboli sequence in Fig. 10. Six Amboli ashes proved unproductive (Table 1), and only one ash horizon macerated trapped significant quantities of organic clasts (Bom 16/98). By contrast,

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


Fig. 8. Amboli XRD profiles. (a) Bom 3/99 whole-rock profile, (b) Bom 3/99 clay separate profile. sme=smectite, cal=calcite, qtz=quartz, feld=feldspar, latt and csme=chlorite:smectite superlattice (corrensite), kao=kaolinite.

all 11 Worli and 4 Bandra samples contained abundant palynodebris. Changes in absolute palynodebris abundances occur with lithology transitions through these beds, the changes being accompanied by variations in the relative percentages of some phytoclast categories to others. For example, taking into account that drops in angular black clast numbers will force rises in other category percentages, decreases in small and large angular black clasts in Worli samples are accompanied

by marked increases in fragments displaying tracheids (Fig. 9). Following a different trend, low amounts of angular black clasts in Amboli samples generally accompany augmented amorphous organic matter (AOM) and branching leaf-like fragment percentages (Fig. 10). Small angular black clasts are consistently present in high percentages; the largest concentration occurs in Bom 4/98, a laminated, pyrite-rich bed (Table 9). Large angular black clasts are less concentrated, but


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Table 7 Infratrappean and intertrappean major (wt.%) and trace (ppm) element compositions received from XRF analyses (for lithologies, refer to Table 1) Sample SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Rb Sr Y Zr Nb Ba Pb Th U Sc V Cr Co Ni Cu Zn Ga Mo As S Bom 1/98 60.62 0.729 14.25 5.25 0.085 2.43 3.82 5.52 1.72 0.154 4.33 56.8 159 35.1 472 110.6 500 12 23 3 13 217 228 21 81 51 57 14 0 4 232 Bom 9/98 52.13 0.544 11.9 4.5 0.103 4.4 8.43 5.66 1.01 0.077 11.18 26.4 148 24.8 420 97.9 220 10 19 5 10 76 222 7 24 42 32 11 0 4 415 Bom 16/98 37.15 0.557 3.78 5.52 0.161 11.53 16.16 0.28 0.05 0.147 24.34 2 155 18.4 74 10.8 30 1 2 0 17 156 41 16 24 39 38 6 7 4 2571 Bom 23/98 71.46 0.635 12.09 3.57 0.048 1.5 1.04 3.21 3.21 0.19 2.51 113 166 33.3 360 79.1 919 9 17 4 12 103 272 32 309 56 74 11 1 8 125 Bom 1/99 64.59 0.751 15.93 4.24 0.077 1.74 2.3 6.89 1.92 0.107 1.88 55.5 143.6 35.3 582.8 142.6 537 14.7 29.6 4.9 9.8 67 212.8 9.8 3737 22.7 49.8 15.2 0.4 6.7 424 Bom 3/99 36.76 0.893 9.06 9.74 0.19 6.83 15.49 1.76 0.41 0.103 15.38 12.8 87.1 26 69.1 9.8 80.6 2.1 0 2.2 39.8 273.4 115 35.1 58.6 58.7 51.9 15.1 0 6.9 705 Other Deccan 42.46 1.655 11.01 11.66 0.18 3.78 13.07 0.17 1 0.09 15.16 34.87 106.5 22.71 108.7 10.58 131.4 5.35 3.93 1.372 30.53 243.9 111.5 25.92 47.15 136.7 46.12 14.9 0.564 3.32 258.8

Other Deccan=mean result obtained from a variety of ash intertrappeans from the Western Ghats, the Krishna­Godavari basin and the Mandla Lobe (Fig. 1).

follow a similar pattern up the samples. Branching leaf-like clasts are important in Amboli and Worli sediments, and Bandra cuttings are dominated by AOM. Amorphous matter and parenchymatous tissues are more abundant in Amboli than Worli samples, while fragments displaying tracheids are only important in Worli sediments. As with the small and large black clasts, branching leaf-like fragments and black laths typically exhibit angular edges. Phytoclast colours are recorded in Table 10, following the thermal maturity scheme of Batten (1996). Derived plant material is dominantly blackened, creating high thermal maturity estimations

Table 8 Results of stable carbon isotope analyses of kerogen samples (PDB=Peedee belemnite standard) Sample Bom 5/98 Bom 16/98 Wo 2001 Wo 2850 d 13Cx PDB À26.39 À26.97 À25.4 À25.77 À24.78 À24.99 À24.86 À25.03 Mean d 13Cx PDB À26.68 À25.58 À24.89 À24.94 Standard deviation 0.409 0.261 0.148 0.12

Table 9 Relative percentages of palynofacies categories for productive B, Bom and Wo samples Sample AOM Black Black, Branching Brown, Brown, Fungal Large, Palynomorph Parenchyma Small, Small, Subspherical Tracheid Cuticle? Noncellular lath (?wood) porous (?leaf) angular porous black, (non fungal) black, translucent black membrane angular angular 51.5 68.5 44.5 35 2 60.5 28 17.5 0.5 72 2 42 0 8.5 0.5 47.5 0.5 2.5 5 1.5 2 0 0 1.5 0 2 0.5 2.0 3.5 3.5 0 3 6 3 2 0.5 3.5 6.5 2.5 0 1 4.5 1.5 0 2 1.5 1.5 1.5 4 2.5 5 5 4.5 6 7 6 30.2 1.5 0 0 0 0 1 0 0 0 0.5 0 1 0 19.5 0 0 8.5 0 0 7.5 0 2.5 0 5 2.5 0 0 2.5 0 0 0 0 0 0 0 0 31 10.5 2.5 0 49 0 1 2 0 6.5 0 1.5 8.5 0 19.5 0 1 15 0 0 4.5 1.2 10.5 0 0 0 0 0.5 0 0.5 0 0 0 0 0 0.5 0 0 0 1 0 4.5 24.5 6.5 2.5 1 2.5 4 3.5 6.5 0 1.5 0 0 0 0 0 0 1.5 0 0 0 0 0 1 0 0 0 0 2 9 0 5 9 0 6.5 11.5 0 2 0 2.5 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1.5 0.5 7 0 0 0 0 0 0 0.8 0 8 2 11 17.5 23 6.5 8 14.5 26.5 3.5 10.5 11 28 12.5 30 13 19 12 20 14 9 11 24.5 8 24 5 18.5 12.7 16 0 0 0.5 0 0 0 0 0 0 0 0.5 0 0 0.5 0 0 0.5 0.5 0 0.5 0 0 0 0 0 1.5 0 0 0 0 0 0 0 70.5 0 1.5 3.5 0 0 13 0 0 0 0 0 0 0 12 0 5 0 0 0.5 0 0 0 0 11.5 33.5 28.5 38 41.5 0 31 20.5 36 60 21.5 9.5 45.5 44.5 74 66.5 21 72 64.5 17 25.5 24.5 30.5 60.5 37 29.5 53 56.5 23.7 34.5 0 0 0 0 0 0 0 8.5 1.5 0.5 0 0 0 0.5 1 0 2.5 11.5 3.5 0 1.5 11.5 0.5 6 0 16 1.5 0 0.5 3.5 1 0.5 0 0 0 0.5 4 0.5 0 0 0 0 0.5 2 1.5 0.5 2 0.5 1.5 3 3 2.5 5.5 1 3 1 15.9 5.5 0 0 0 0 0 0 0.5 0 0 0 0 0 1 0 0 0 2.5 2 3 22 4 27 0 1.5 22 7.5 0.5 6.1 1.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 1.5 0.5 0 6.9 0 0 0 2.5 0 0 0 7.5 2 0.5 0 14.5 0.5 0 0 0 0 0 0 6.5 3 8 0.5 0 9 0.5 1 0 0.4 11 J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

B 3510 B 3130 B 3000 B 2800 Bom 20/98 Bom 19/98 Bom 16/98 Bom 15/98 Bom 13/98 Bom 12/98 Bom 10/98 Bom 8/98 Bom 5/98 Bom 4/98 Bom 3/98 Bom 2/99 Bom 2/98 Bom 1/98 Wo 3408 Wo 3128 Wo 2850 Wo 2736 Wo 2735 Wo 2610 Wo 2600 Wo 2210b Wo 2210a Wo 2100 Wo 2001



J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 9. Relative percentages of palynofacies categories for Bandra (B) and Worli (Wo) samples.

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


Fig. 10. Distribution of palynofacies types with height through the Amboli section (details given in Table 9). Grey bands mark the positions of unproductive ashes.


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 10 (continued).

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 ~30% ~30% ~40% 5­6 Wo 3408


~45% ~30% ~25% 5

Wo 2610

(Amboli mean 6.3; Worli mean 5.6). E:L ratios (equant to lath-shaped clasts; Table 10) were received from counts of 50 black wood grains. Mean E:L ratios (38.4:11.6 for Amboli, and 35.8:14.2 for Worli) are similar, equant-shaped grains dominating over lathshaped in both sequences. Fig. 11 compares thermal maturity with black wood shape and size ratios through the Amboli section. Overall, b40 Am grains marginally form the greatest black wood size component, although there is a relatively even distribution of b40 Am, 40­80 Am and N80 Am clasts.

~40% ~45% ~15% 5 ~30% ~40% ~30% 5 ~30% ~35% ~35% 5­6 ~35% ~40% ~25% 6­7 ~30% ~40% ~30% 5­6

Wo 3128

Wo 2850

Wo 2736

Wo 2735

Table 10 Black wood phytoclast size, colour and shape statistics for productive Bom and Wo samples (thermal maturation after Batten, 1996)

Wo Wo Wo 2210a 2210b 2600

42 8

41 9

29 21

32 18

33 17

37 13

5. Interpretation 5.1. Facies The conspicuous absence of archetypal MDP boles and calcretes in Mumbai Island intertrappeans highlights a general lack of sediment subaerial exposure. Tectonic adjustments controlled the subaqueous nature of Mumbai sediments and Traps, allowing water to flood into the developing shallow basins as rifting and foundering of the margin progressed. Slickensides that both follow and cross bedding planes probably developed during this period of tectonism. Substantial intertrappean thicknesses are partly due to the extent of contemporaneous regional subsidence. Shale laminations indicate a lack of bioturbation, suggesting that infauna were unable to exploit these sediments, possibly due to inadequate interstitial oxygen levels. The combination of swamp facies and anoxic laminated sediments implies that water levels were generally very shallow, yet liable to stagnation. This was perhaps a consequence of restricted water mixing through a low-energy column, the aqueous body being isolated from a fully open marine influence. A stratified water column with a high potential towards basal anoxia may have resulted from a subtly more dense, brackish layer separating surficial, aerated freshwater from the sediments, such circumstances being liable to occur in partly enclosed, sheltered lagoons fed by rivers. Shale carbon concentrations appear to have been optimised by low clastic sediment input combined with high terrigenous organic productivity, and OM decomposition would

~70% ~25% ~5% 5­6 ~40% ~25% ~35% 6 ~20% ~50% ~30% 6 ~40% ~25% ~35% 5­6 ~45% ~30% ~25% 6 ~45% ~25% ~30% 6 ~35% ~35% ~30% 7 ~40% ~35% ~25% 5 ~40% ~25% ~35% 7 ~40% ~35% ~25% 6­7 ~40% ~30% ~30% 6­7 ~40% ~30% ~30% 6­7 ~40% ~40% ~20% 6

Wo 2100

Bom Bom Bom Bom Wo 10/98 13/98 15/98 16/98 2001

Bom 5/98

Bom 4/98

Bom 3/98

Bom 2/98

Bom 1/98

Phytoclast size: Small (b40 `m) i Medium (40­80 `m) i Large (N80 `m) i Thermal maturation Pytoclast shape: Equant (out of 50) Lath (out of 50)


41 9

43 7

23 27

43 7

45 5

28 22

42 8

36 14

45 5

34 16

40 10

44 6

34 16

28 22


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Fig. 11. Log of Amboli section palynofacies characteristics.

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


in turn have depleted oxygen resources. Clastic sediments are dominated by volcanic material, signifying that sedimentation rates diminished during nonvolcanic periods. Although OM was largely introduced, the dearth of eroded clastic material points to hinterland gradients having been negligible. Water energy infrequently increased, and undulating or rippled horizons became deposited above flatlaminated sediments. Paler, ash-rich units typify these faintly higher energy facies, the water movement perhaps initiated by ash introductions that triggered minor density currents. Tablets of flatlaminated shale in one sandy ash appear to have been ripped up and reworked after their compaction but before lithification. Horizons bearing asymmetric ripples indicate directed flow, potentially having resulted from such ash-bearing currents progressing across lagoon floors. Ripple tops were sometimes preserved flattened or altered into flame structures during their rapid deposition, dehydration and collapse (e.g., Fig. 7c). The Mumbai lagoons were stable environments that were disrupted by ash eruptions. Rare fine, laterally continuous organic drapes settled above rippled layers, as the water reverted to its calm state. Repetitive pyroclastic influxes established the series of fining-upwards, ash-rich rhythms through the Amboli section. The transition from Bom 8/98 to Bom 1/99 appears to equate to a gradual increase in pyroclastic activity, culminating in a major local event. Many ash beds are indurated, their matrix, having been welded. Spherical to ovoid objects, constituting Bom 3/99, lack internal structure, more closely resembling the coalesced ash bombs described by Sukheswala (1956) than the spilitic fragments or pillows detailed by Tolia and Sethna (1990), occurring in an ash rather than a flow breccia. Laminations cup underneath these bombs, as though the pyroclasts dropped upon and depressed unconsolidated sediments. These accretionary lapilli strongly suggest that ejecta cones were in close proximity to the Amboli lagoon. The lamination deficit through most ashes probably resulted from their accelerated, chaotic deposition styles. Air-fallen and fluvially deposited loose pyroclastics were possibly aerated enough to support burrowing organisms that obscured original bedding features.

When present, bivalve and gastropod internal moulds are of small (1­2 cm) sizes. This might be consequential to oxygen deficiency having stunted growth and/or caused large proportions of the mollusc populations to die prior to reaching maturity. The sizes of feeding traces upon a quarry floor bedding plane point to excavation by small crustaceans, and float crustacean claw sample Bom 22/98 (Table 1) may have originated from this horizon. Subhorizontal burrowing activity suggests sedimentation rates were low when organisms exploited the sediments. Their near absence in higher beds might be consequential to subsequent ash injections. The prevalence of shales through the extensive Worli and Bandra sequences points to continually low sedimentation rates here, and therefore substantial sedimentation durations. Discrepancies in ash and Trap frequencies between Amboli and the Worli and Bandra tunnels indicate that either volcanic centres were closer to Amboli, or activity was more intense at the time of Amboli deposition. Worli and Bandra shales are not as well-cemented as the Amboli sediments, suggesting cement migrated from ash horizons. Diagenetic events have altered the Amboli section, and recrystallisation during lithification is particularly evident towards the uppermost basalt. Polygonal cracks in sediments contacting the columnar lopolith are likely to have evolved simultaneously with the intrusion's contraction upon cooling. 5.2. Geochemistry The XRD profile of a volcanic bomb (Bom 3/99) exhibited numerous, clearly defined reflection peaks at positions signifying well-developed corrensite crystals (Fig. 8). Relatively fresh feldspars produce peaks; thus, it seems unlikely that sedimentary processes occurred over an extended enough period to permit the development of regularly alternating chlorite:smectite lattices. Rather, increasing diagenesis temperatures and durations transformed smectites into this mixed-layer, chloritic clay, by means of repeated dissolution and precipitation events. The ratio of chlorite to smectite (c. 80:20) indicates a heating event of z100 8C during lithification, possibly accompanied by a degree of saline fluid flow (Beaufort et al., 1997; Murakami et al., 1999).


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

Kaolinite forms a minor contribution to the volcanic bomb. As weathering continues, smectite can alter to kaolinite through a succession of smectite­kaolinite mixed-layer transitions. Its near absence in weathered Deccan volcanics suggests these fossilised at early stages of modification. Kaolinite crystals can, however, grow within substrates subjected to prolonged waterlogging, and while MDP boles were largely too well-drained to promote its precipitation, the Mumbai lagoonal basins provided more favourable precipitation sites. Kaolinite is a common alteration product of felsic igneous rocks, and phlogopite micas present through tuff sample Bom 1/99 may be indicative of a transformation to more felsic late stage volcanism as the region rifted and subsided. The varied chemistries of Amboli ashes are a reflection of their occasional explosive genesis in aqueous facies, clastic contamination, element mobilisation prior to lithification and hydrothermal alteration resulting from nearby intrusions. High sodium levels through these relative to MDP ashes (Table 7) may be consequential to their deposition in saline lagoons, although sodium from albites present would have augmented these concentrations. Amboli kerogen possesses marginally lower carbon isotopic signatures than those of Worli (Table 8). Thermal maturation, induced by local intrusion emplacement, is one means by which original Amboli OM d 13C could have been lowered. Dykes cross-cut intertrappeans offshore Mumbai (Sethna, personal communication); if these imparted a greater influence on Amboli than Worli sediments, they might additionally have been responsible for the darker Amboli phytoclast colours (Table 10). Relative depletions in Mumbai shale 13C through heating was possibly influenced by a selective preservation of organic fractions with augmented 12 C comparative to the total OM. Lipids, the most stable of plant constituents, are enriched in 12C by up to 8x compared with other biogenic compounds (Faure, 1986), and their hydrocarbon composition closely resembles that of petroleum. Smectitic clays catalyse lipid transformations to hydrocarbons virtually identical to petroleum (Faure, 1986), and the offshore Mumbai region is rich in source rocks.

5.3. Palaeontology Although molluscs are sporadically distributed through Amboli shales, no typical MDP genera (e.g., Physa gastropods, Unio bivalves) were identified during the present study. Since shale faunal material possessed high preservation potentials, the absence of ubiquitous MDP forms almost certainly reflects their intolerance to marginal marine environments. Investigations are required to ascertain whether these genera continued to occupy contemporaneous MDP Danian, ?Desur Formation palaeoenvironments (e.g., Singh and Kar, 2002), and thus survived the full effect of the Deccan episode proximal to the principal focus of flood basalt activity. Invertebrates which did inhabit Mumbai lagoons were periodically capable of exploiting oxygenated surface sediments, as demonstrated by the pellet back-filled feeding traces. No macroflora was recovered from the Amboli section by the present authors, although this sequence is extremely rich in disseminated plant matter. Parent plants possibly colonised firm terrain tens of metres from the low-angled, muddy lagoon shores and, consequentially, intact plant organs were seldom fluvially transported into the lagoons. 5.4. Palynofacies analyses Of the 16 palynofacies categories selected to represent the Mumbai phytoclasts (Table 9), 14 symbolise land-derived plant fragments which received their shapes, colours and sizes from their parent plant and organ varieties and taphonomic (including sedimentological) effects. (AOM is of unknown derivation, and fungal remains are virtually ubiquitous.) To classify the OM according to kerogen type (Tyson, 1985), these palynofacies are rich in humic kerogens (higher plant wood and parenchymatous tissues), much of this having altered to inertinite (carbonised black wood). The sapropelic kerogen component (structureless matter, largely planktonderived) is negligible and fusinite (fossil charcoal) is rare. Any volcanogenic charcoal potentially entered the open sea due to its slow waterlogging rate (cf. Nichols et al., 2000). All phytogenic clasts of known origin are terrigenous, reflecting deposition proximal to land, sheltered from a strong marine influence. The lack of

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


macroflora, and fragmented nature of wood and leaf material, infers these are allochthonous phytoclasts, allowing their characteristics to be applied sedimentologically to interpret their transport histories and depositional environments. Because the palynodebris were from palynomorphreleasing plants, the dearth of pollen and spores (Table 9) is unlikely to reflect low productivity, and sporopollenin-walled grains doubtfully degraded prior to wood and leaves. Whereas larger, more massive phytoclasts settled upon lagoon floors near river mouths, palynomorphs, particularly those trapping air, probably floated into the marine realm. This would necessitate slight horizontal water movements within the lagoon. Nonetheless, shales trap rare palm pollen, water-fern massulae, algal colonies and fungal spores, these grains also occurring in MDP intertrappeans (Cripps, 2002). Well preserved, seemingly autochthonous Botryococcus algae conceivably bloomed following seasons of high rainfall, within negligible salinity surficial layers of density- and salinity-stratified lagoons. Through the Amboli sequence, depletions in angular clasts correspond to rises in AOM and/or branching leaf-like fragments (Fig. 10), these changes normally accompanying transitions into shale facies (Table 1). Large angular clasts (e.g., Bom 13/98) accumulated when high runoff volumes rapidly transported palynodebris to the basins. Bom 4/98 has the most small angular black clasts. A lower velocity waterway would permit palynodebris darkening through oxidation during protracted transit times, selectively entraining finer grains, without promoting significant rounding. High AOM percentages in the Bandra samples and Bom 12/98, as well as a relatively large concentration of parenchyma in Bom 20/98, signify periods of diminishing river currents, during which only the lightest material reached the lagoons. Large black angular clast abundances mimic the trend of their small counterparts, implying that black clast percentages are predominantly associated with preservation effects. At Worli, negative correlations of palynodebris exhibiting tracheids to angular black clasts (Fig. 9) are likely to have resulted from runoff fluctuations vertically displacing the oxygen minimum zone. Palynodebris displaying tracheids increased in importance when this zone rose, while

falls increased biodegradation, deteriorating ultrastructural details and blackening palynodebris. Branching leaf-like material forms the second most common structured palynodebris type after angular black clasts. Worli sediments appear to have accumulated in a more distal setting than the Amboli intertrappeans, typically beneath the oxygen-minimum zone, with many particles exhibiting tracheids or leaf ultrastructures. The presence of leafy material in many Amboli and Worli sediments indicates that the decay of vast quantities of introduced leaves exhausted oxygen supplies, leaving much of the litter to become buried beneath muds (e.g., Bom 10/98). Bom 16/98, the only productive ash bed, contains a disproportionately high percentage of leaf-like material, and this is mostly pale. It is plausible that pyroclastic debris fell into a river, charging the water with sediment until it breached its banks, overwhelming and incorporating leafy floodplain plants. Upon entering the lagoon, the prompt and chaotic dumping of this material prohibited organic biodegradation. Bom 16/98 phytoclast distributions are very different to those of adjacent samples (Figs. 10 and 11), due to its distinct deposition style. Diluted palynodebris concentrations in other ash horizons have resulted from extremely rapid ash accumulation rates. The Amboli logs do not exhibit a constant palynofacies evolution timeline (Figs. 10 and 11) because of substantial disparities in shale and ash deposition rates and compaction extents. Although palynofacies data cannot delineate successional seres following pyroclastic events, low OM concentrations would be anticipated after catastrophic eruptions, and these are not apparent above Amboli extrusives. Ash Bom 16/98 terminates in black shales, and Bom 10/98 is a very productive bed occurring above ash Bom 9/98. It appears that Amboli vicinity ash-falls were localised phenomena, imparting minimal disturbance upon plant-life in the surrounding watershed. The manifest lack of a consistent palynoclast distribution pattern following ash emplacements is perhaps due to the erratic natures of the fluvial and aerial transport mechanisms. Minimal sorting has led to black wood exhibiting angular shapes and notable size ranges through the three sequences. It is thus surmised that much of the parent vegetation grew behind the lagoon shores, separated by stretches of muddy coastline. The


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332

dominance of equant- over lath-shaped phytoclasts is a reflection of debris buoyancy. Laths remain in suspension longer than equant grains, their high surface area to weight ratios retarding settling through the water column (Tyson and Follows, 2000). Palynofacies in which laths become increasingly significant (e.g., Bom 3/98, Wo 2600) delineate times of low lagoon energy, when sediment barriers to the sea developed. The marginally lower mean E:L ratio for Worli than Amboli suggests that Worli sedimentation occurred slightly further from river mouths. Most phytoclasts have high thermal maturities (6.3 Amboli mean, 5.6 Worli mean; Fig. 11; Table 10). While colour differences between palynodebris are influenced by variations in OM type and pre-burial oxidation (McArthur et al., 1992), diagenetic heating by intrusions was important around Mumbai.

6. Discussion Mumbai palynofacies are the products of tectonic and igneous activity, the proximity of plant communities, runoff volumes and velocities, airborne particle

fluxes and lagoon oxygen levels. The latter fluctuated consequential to depth changes, salinity stratification, turbulence and OM additions. Palynodebris sedimentological and preservational responses to environmental transformations produced the palynofacies patterns present. Mumbai intertrappean phytoclasts were deposited in extensive lagoons which experienced mild horizontal currents but insignificant vertical mixing. Forests persistently occupied river watersheds draining into the lagoons, on the solid, gently sloping hinterland beyond their muddy shores, regardless of sporadic regional pyroclastic volcanism. Terrigenous OM was supplied by rivers following precipitation. Although the Amboli, Worli and Bandra sections may not be contemporaneous, their depositional facies are likely to have coexisted concurrently in adjacent areas (Fig. 12). Worli and Bandra sediments accumulated further from the mouths of palynodebris-bearing rivers and ash cones than those at Amboli. Megaflora is scarce, although large land plant organs have been identified (e.g., Podocarpaceae wood; Bande, 1992). Sizeable fragments, such as logs, would be most effectively transported by an

Fig. 12. Palaeogeographical reconstruction of Amboli, Worli and Bandra depositional environments (aerial view, not to scale). Fauna based on Singh and Sahni (1996); flora based upon current work, Bande (1992) and Bande et al. (1988).

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332


aggressive, erosive drainage system. At Mumbai, however, megafloral remains were deconstructed through biological activity into smaller fragments, light enough to be suspended in distributaries meandering down the low-gradient back-shores to the lagoons. Charcoal is uncommon in Mumbai palynofacies, although it might be anticipated in volcanic regions (e.g., Uhl et al., 2004). Molten lava could promote the vaporisation and/or aerobic ashing of flora it advanced upon, rather than anaerobic charcoalification. Sedimentological and geochemical evidence points to Mumbai activity being more explosive than previous MDP tholeiitic eruptions; Mumbai surges presumably had lower expulsion temperatures than those of mafic effusions, and potentially cooled before contacting plants, as documented at Merapi, Java by Kelfoun et al. (2000). OM was infrequently scorched either within airborne plumes or by volcanic bombs. Eruptions doubtless uprooted or smothered plants, killing those worst affected, and forcing others to defoliate. However, the palynofacies reveal that few burned and, more importantly, a significant percentage survived to produce both OM and later generations. Paroxysmal volcanic explosions repeatedly showered Mumbai coasts, and organisms at bground zeroQ must have been subjected to serious environmental trauma. However, palaeontological evidence confirms the establishment of enduring communities of remarkable diversity and sensitivities. As well as the magnitudes of volcanic impacts decreasing with radial distance from ash vents, effects would be dependent upon the direction of pyroclastic flow and plume movements. Refugia clear of gravity flow drainage routes and upwind of plumes could have remained comparatively unscathed. Pyroclastic releases can decouple into flows and clouds (cf. Kelfoun et al., 2000) that generate at least two deposits, and fluvial reworking into further accumulations is possible. In such ways, one Mumbai expulsion might be represented by several lagoon layers, giving the semblance of more frequent eruptions than actually prevailed. If ashes did not reach Worli and Bandra because pyroclastics were transported in negligible distances, eruptions would have been of an inadequate magnitude to have injected materials into the stratosphere.

Hence, it is rational to conclude that no long-term climatic ramifications resulted. Negi et al. (1993) proposed that an offshore Mumbai extraterrestrial impact triggered Deccan volcanism, and argued for a bimodal origin for the K­T boundary extinctions. The palaeontology of intertrappeans near the focus of this alleged bolide cataclysm demonstrably contradicts this. The establishment of a varied biota shortly after the MDP eruptions calls into question their efficacy regarding environmental devastation. These data add credence to models implicating the Chicxulub impact as the main cause of organism turnover at the K­T boundary.

7. Conclusions ! This work highlights the imperative for future investigators to identify irrefutable K­T boundaries within the main Deccan tholeiitic succession, and thereby divide intertrappeans into pre- and post-boundary environments. Tertiary Deccan intertrappeans are extremely scarce (e.g., Singh and Kar, 2002), and the Danian Mumbai Island Formation possesses facies and organisms that are highly distinctive from all known MDP sediments. ! Mumbai facies are dominantly subaqueous, the sediments being significantly more organic-rich than those of the MDP. This is due to their coastal palaeoenvironments having undergone syndepositional subsidence, although it may additionally reflect the influence of an increasingly humid Danian climate. Crustal extension supplied pathways for intrusions which heated basin sediments, and created the Mumbai gravity high. ! Mumbai ecosystems represent the legacy of global K­T boundary phenomena combined with local and preceding intense regional Deccan flood basalt activity. Numerous pyroclastic eruptions influenced lagoon intertrappean accumulation; however, no evidence exists for extensive wildfires ensuing or for floral mass mortality events. Rather, abundant plant material entered the lagoons throughout this active period. ! Based upon the current findings, the authors stress a need to reassess the palaeoenvironments of other continental flood basalt provinces that are temporally correlated with ecological crises.


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 India. Tectonic Context. Gyanodaya Prakashan, Nainital, India, pp. 74 ­ 103. Blanford, W.T., 1867. On the Traps and intertrappean beds of western and central India. Mem. Geol. Surv. India 6, 137 ­ 162. Blanford, W.T., 1872. Sketch of the geology of the Bombay presidency. Rec. Geol. Surv. India 5, 82 ­ 102. Bossuyt, F., Milinkovitch, M.C., 2001. Amphibians as indicators of Early Tertiary bout of IndiaQ dispersal of vertebrates. Science 292, 93 ­ 95. Chatterjee, S., Rudra, D.K., 1996. KT events in India: impact, rifting, volcanism and dinosaur extinction. Mem. Queensl. Mus. 39, 489 ­ 532. Chiplonkar, G.W., 1940. A new species of fossil frog from the InterTrappean beds of Worli Hill, Bombay. J. Bombay Nat. Hist. Soc. J. 40, 799 ­ 804. Combaz, A., 1964. Les palynofacies. Rev. Micropaleontol. 7, ` ´ 205 ­ 218. Courtillot, V., Besse, J., Vandamme, D., Montigny, R., Jaeger, J.-J., Cappetta, H., 1986. Deccan flood basalts at the Cretaceous/ Tertiary boundary? Earth Planet. Sci. Lett. 80, 361 ­ 374. Cox, K.G., Hawkesworth, C.J., 1984. Relative contribution of crust and mantle to flood basalt magmatism, Mahabaleshwar area, Deccan Traps. Philos. Trans. R. Soc. Lond., A 310, 627 ­ 641. Cox, K.G., Hawkesworth, C.J., 1985. Geochemical stratigraphy of the Deccan Traps at Mahabaleshwar, Western Ghats, India, with implications for open system magmatic processes. J. Petrol. 26, 355 ­ 377. Cripps, J.A., 2002. Environmental impact of Deccan Trap flood basalt volcanism: assessment of regional floral responses to late Cretaceous­early Tertiary activity. PhD thesis, The Open University, Milton Keynes, UK, 502 pp. Cross, A.T., Taggart, R.E., 1982. Causes of short-term sequential changes in fossil plant assemblages; some considerations based on a Miocene flora of the Northwest United States. Ann. Mo. Bot. Gard. 69, 676 ­ 734. Deshmukh, S.S., 1982. Volcanological and petrological appraisal on Deccan basalts. Science Lecture Series, vol. 1. Gondwana Geological Society, India, pp. 30 ­ 56. Deshmukh, S.S., 1984. Geological and petrographic studies of the Deccan basalt flows and intercalated volcanoclastic beds encountered in drill holes in Bombay city and harbour areas. Rec. Geol. Surv. India 113, 33 ­ 51. Dessai, A.G., Bertrand, H., 1995. The bPanvel flexureQ along the western Indian continental margin: an extensional structure related to Deccan magmatism. Tectonophysics 241, 165 ­ 178. Devey, C.W., Lightfoot, P.C., 1986. Volcanological and tectonic control of stratigraphy and structure in the western Deccan Traps. Bull. Volcanol. 48, 195 ­ 207. Devey, C.W., Stephens, W.E., 1991. Tholeiitic dykes in the Seychelles and the original spatial extent of the Deccan. J. Geol. Soc. Lond. 148, 979 ­ 983. Duncan, R.A., Pyle, D.B., 1988. Rapid eruption of the Deccan flood basalts at the Cretaceous/Tertiary boundary. Nature 333, 841 ­ 843. Erwin, D.H., Vogel, T.A., 1992. Testing for the causal relationships between large pyroclastic volcanic eruptions and mass extinctions. Geophys. Res. Lett. 19, 893 ­ 896.

Acknowledgements The authors would like to thank the reviewers for their constructive criticisms, and K.V. Subbarao (IIT Mumbai) and A. Sahni (Panjab University) for their invaluable comments and field support. Cuttings from the Worli and Bandra tunnels were kindly donated to this investigation by S.F. Sethna (St. Xavier's College). We are indebted to V. Pearson, J. Watson, M. Sephton (Open University), R. Williams and T. Clayton (University of Southampton) for their advice and assistance with various analyses. This work was conducted during the tenure of a project funded by the Natural Environment Research Council. References

Abe, A.A., Bicudo, J.E.P.W., 1991. Adaptations to salinity and osmoregulation in the frog Thoropa miliaris (Amphibia, Leptodactylidae). Zool. Anz. 227, 313 ­ 318. Balasubrahmanyan, M.N., Snelling, N.J., 1981. Extraneous argon in lavas and dykes of the Deccan volcanic province. In: Subbarao, K.V., Sukheswala, R.N. (Eds.), Deccan Volcanism and Related Flood Basalt Provinces in Other Parts of the World, Geological Society of India Memoir, vol. 3, pp. 259 ­ 264. Bande, M.B., 1992. The Palaeogene vegetation of peninsular India (Megafossil evidences). Palaeobotanist 40, 275 ­ 284. Bande, M.B., Chandra, A., Venkatachala, B.S., Mehrotra, R.C., 1988. Deccan Intertrappean floristics and its stratigraphic implications. In: Maheshwari, H.K. (Ed.), Palaeocene of India. Indian Association of Palynostratigraphers, Lucknow, India, pp. 83 ­ 123. Batten, D.J., 1996. Palynofacies and palaeoenvironmental interpretation. In: Jansonius, J., McGregor, D.C. (Eds.), Palynology: Principles and Applications. American Association of Stratigraphic Palynologists Foundation, Salt Lake City, Utah, pp. 1011 ­ 1064. Beane, J.E., Turner, C.A., Hooper, P.R., Subbarao, K.V., Walsh, J.N., 1986. Stratigraphy, composition and form of the Deccan basalts, Western Ghats, India. Bull. Volcanol. 48, 61 ­ 83. Beaufort, D., Baronnet, A., Lanson, B., Meunier, A., 1997. Corrensite: a single phase or a mixed-layer phyllosilicate in the saponite-to-chlorite conversion series? A case study of Sancerre­Couy deep drill hole (France). Am. Mineral. 82, 109 ­ 124. Bhatia, S.B., Prasad, G.V.R., Rana, R.S., 1990. Deccan volcanism: a Late Cretaceous event: conclusive events of ostracodes. In: Sahni, A., Jolly, A. (Eds.), Cretaceous Event Stratigraphy and Correlation of the Indian Non-marine Strata. International Geological Correlation Project, Chandigarh, pp. 47 ­ 49. Biswas, S.K., 1991. Stratigraphy and sedimentary evolution of the Mesozoic Basin of Kutch, western India. In: Tandon, S.K., Pant, C.C., Casshyap, S.M. (Eds.), Sedimentary Basins of

J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 Faure, G., 1986. Principles of Isotope Geology. Wiley-Europe, Chichester, UK. 589 pp. Gaffney, E.S., Sahni, A., Schleich, H., Singh, S.D., Srivastava, R., 2003. Sankuchemys, a new side-necked turtle (Pelomedusoides : Bothremydidae) from the Late Cretaceous of India. Am. Mus. Novit. 3405, 1 ­ 10. Gombos, A.M., Powell, W.G., Norton, I.O., 1995. The tectonic evolution of western India and its impact on hydrocarbon occurrences--an overview. Sediment. Geol. 96, 119 ­ 129. Hardy, R., Tucker, M.E., 1988. X-ray powder diffraction of sediments. In: Tucker, M.E. (Ed.), Techniques in Sedimentology. Blackwell Science, Oxford, UK, pp. 191 ­ 228. Hildebrand, A.R., Penfield, G.T., Kring, D.A., Pilkington, M., Camargo, A., Jacobsen, S.B., Boynton, W.V., 1991. Chicxulub crater--a possible Cretaceous Tertiary boundary impact crater on the Yucatan Peninsular, Mexico. Geology 19, 867 ­ 871. Hofmann, C., Feraud, G., Courtillot, V., 2000. 40Ar/39Ar dating of mineral separates and whole rocks from the Western Ghats lava pile: further constraints on duration and age of the Deccan traps. Earth Planet. Sci. Lett. 180, 13 ­ 27. Hooper, P.R., 1990. The timing of crustal extension and the eruption of continental flood basalts. Nature 345, 246 ­ 249. Hooper, P.R., 1999. The winds of change: the Deccan traps, a personal perspective. In: Subbarao, K.V. (Ed.), Deccan Volcanic Province, Memoir of the Geological Society of India, vol. 43, pp. 153 ­ 165. Kaiser, H., 1997. Origins and introductions of the Caribbean frog, Eleutherodactylus johnstonei (Leptodactlidae): management and conservation concerns. Biodivers. Conserv. 6, 1391 ­ 1407. Kaneoka, I., 1980. 40Ar­39Ar dating on volcanic rocks of the Deccan Traps, India. Earth Planet. Sci. Lett. 46, 233 ­ 343. Kaneoka, I., Haramura, H., 1973. K­Ar ages of successive lava flows from the Deccan Traps, India. Earth Planet. Sci. Lett. 18, 229 ­ 236. Kaneoka, I., Iwata, N., Takigami, Y., 1997. 40Ar­39Ar dating: investigation of some technical problems and its application to the Deccan Traps rocks and a unique meteorite from Antarctica. Sci. Rep. Res. Inst., Tohoku Univ., Ser. A--Phys. Chem. Metall. 45, 47 ­ 51. Kelfoun, K., Legros, F., Gourgaud, A., 2000. A statistical study of trees damaged by the 22 November 1994 eruption of Merapi volcano (Java, Indonesia): relationships between ash-cloud surges and block-and-ash flows. J. Volcanol. Geotherm. Res. 100, 379 ­ 393. Khosla, A., Sahni, A., 2003. Biodiversity during the Deccan volcanic eruptive episode. J. Asian Earth Sci. 21, 895 ­ 908. Lightfoot, P.C., Hawkesworth, C.J., Sethna, S.F., 1987. Petrogenesis of rhyolites and trachytes from the Deccan Trap: Sr, Nd and Pb isotope and trace element evidence. Contrib. Mineral. Petrol. 95, 44 ­ 54. McArthur, J.M., Tyson, R.V., Thomson, J., Mattey, D., 1992. Early diagenesis of marine organic-matter--alteration of the carbon isotopic composition. Marine Geol. 105, 51 ­ 61. Mitchell, C., Widdowson, M., 1991. A geological map of the southern Deccan Traps, India and its structural implications. J. Geophys. Soc. Lond. 148, 495 ­ 505.


Moore, P.D., Webb, J.A., Collinson, M.E., 1991. Pollen Analysis. Blackwell Science, Oxford, UK. 216 pp. Murakami, T., Sato, T., Inoue, A., 1999. HRTEM evidence for the process and mechanism of saponite-to-chlorite conversion through corrensite. Am. Mineral. 84, 1080 ­ 1087. Negi, J.G., Agrawal, P.K., Singh, A.P., Pandey, O.P., 1992. Bombay gravity high and eruption of Deccan flood basalts (India) from a shallow secondary plume. Tectonophysics 206, 341 ­ 350. Negi, J.G., Agrawal, P.K., Pandey, O.P., Singh, A.P., 1993. A possible K­T boundary bolide impact site offshore Bombay and triggering of rapid Deccan volcanism. Phys. Earth Planet. Inter. 76, 189 ­ 197. Nichols, G.J., Cripps, J.A., Collinson, M.E., Scott, A.C., 2000. Experiments in waterlogging and sedimentology of charcoal: results and implications. Palaeogeogr. Palaeoclimatol. Palaeoecol. 164, 43 ­ 56. Owen, R., 1847. On the Batracholites, indicative of a small species of frog (from Bombay). Q. J. Geol. Soc. Lond. 3, 224 ­ 225. Pandey, O.P., Agrawal, P.K., 2000. Thermal regime, hydrocarbon maturation and geodynamic events along the western margin of India since late Cretaceous. J. Geodyn. 30, 439 ­ 459. Pope, K.O., Baines, K.H., Ocampo, A.C., Ivanov, B.A., 1994. Impact winter and the Cretaceous­Tertiary extinctions--results of a chicxulub asteroid impact model. Earth Planet. Sci. Lett. 128, 719 ­ 725. Rampino, M.R., Stothers, R.B., 1988. Flood basalt volcanism during the past 250 million years. Science 241, 663 ­ 668. Sabale, A.B., Vishwakarma, L.L., 1996. Zeolites and associated secondary minerals in Deccan volcanics: study of their distribution, genesis and economic importance. Gondwana Geol. Mag. Spec., Vol. 2, 511 ­ 518. Sen, G., 2001. Generation of Deccan Trap magmas. Proc. Indian Acad. Sci., Earth Planet. Sci. 110, 409 ­ 431. Sethna, S.F., 1999. Geology of Mumbai and surrounding areas and its position in the Deccan volcanic stratigraphy, India. J. Geol. Soc. India 53, 359 ­ 365. Sethna, S.F., 2003. The occurrence of acid and intermediate rocks in the Deccan volcanic province with associated high positive gravity anomalies and their probable significance. J. Geol. Soc. India 61, 220 ­ 222. Sharma, R.K., Pandit, M.K., 1998. Ignimbrite deposits from North of Mumbai in western part of Deccan flood basalt province, India. J. Geol. Soc. India 51, 813 ­ 815. Sheehan, P.M., Fastovsky, D.E., 1992. Major extinctions of landdwelling vertebrates at the Cretaceous­tertiary boundary, eastern Montana. Geology 20, 556 ­ 560. Sheth, H.C., Ray, J.S., 2002. Rb/Sr­87Sr/86Sr variations in Bombay trachytes and rhyolites (Deccan Traps): Rb­Sr isochron, or AFC process? Int. Geol. Rev. 44, 624 ­ 638. Sheth, H.C., Pande, K., Bhutani, R., 2001a. 40Ar­39Ar ages of Bombay trachytes: evidence for a palaeocene phase of Deccan volcanism. Geophys. Res. Lett. 28, 3513 ­ 3516. Sheth, H.C., Pande, K., Bhutani, R., 2001b. 40Ar­39Ar age of a national geological monument: the Gilbert Hill basalt, Deccan traps, Bombay. Curr. Sci. 80, 1437 ­ 1440. Singh, S.D., 2000. Petrography and clay mineralogy of intertrappean beds of Mumbai, India. J. Geol. Soc. India 55, 275 ­ 288.


J.A. Cripps et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 216 (2005) 303­332 Tyson, R.V., Follows, B., 2000. Palynofacies prediction of distance from sediment source: a case study from the Upper Cretaceous of the Pyrenees. Geology 28, 569 ­ 571. Uhl, D., Lausberg, S., Noll, R., Stapf, K.R.G., 2004. Wildfires in the Late Palaeozoic of Central Europe--an overview of the Rotliegend (Upper Carboniferous­Lower Permian) of the Saar­Nahe Basin (SW Germany). Palaeogeogr. Palaeoclimatol. Palaeoecol. 207, 23 ­ 35. Vajda, V., Raine, J.I., Hollis, C.J., 2001. Indication of global deforestation at the Cretaceous­Tertiary boundary by New Zealand fern spike. Science 294, 1700 ­ 1702. Vandamme, D., Courtillot, V., 1992. Paleomagnetic constraints on the structure of the Deccan Traps. Phys. Earth Planet. Inter. 74, 241 ­ 261. Vandamme, D., Courtillot, V., Besse, J., Montigny, R., 1991. Palaeomagnetism and age determinations of the Deccan Traps, India--results of a Nagpur­Bombay traverse and review of earlier work. Rev. Geophys. 29, 159 ­ 190. Whatley, R.C., Bajpai, S., Whittaker, J.E., 2003. The identity of the non-marine ostracod Cypris subglobosa Sowerby from the intertrappean deposits of peninsular India. Palaeontology 46, 1281 ­ 1296. Widdowson, M., 1997. Tertiary palaeosurfaces of the SW Deccan, western India: implications for passive margin uplift. In: Widdowson, M. (Ed.), Palaeosurfaces: Recognition, Reconstruction and Palaeoenvironmental Interpretation, Geological Society of London Special Publication 120, pp. 221 ­ 248. Widdowson, M., Pringle, M.S., Fernandez, O.A., 2000. A post K­T boundary (Early Palaeocene) age for Deccan-type feeder dykes, Goa, India. J. Petrol. 41, 1177 ­ 1194.

Singh, R.S., Kar, R.K., 2002. Palaeocene palynofossils from the Lalitpur Intertrappean Beds, Uttar Pradesh, India. J. Geol. Soc. India 60, 213 ­ 216. Singh, S.D., Sahni, A., 1996. Bombay inter-trappeans: new data on age and faunal affinities. Contributions to the XV Indian Colloquium on Micropalaeontology and Stratigraphy, Dehra Dun, India, pp. 465 ­ 469. Subbarao, K.V., Sukheswala, R.N., 1979. Deccan volcanism and related basalt provinces in other parts of the world: field guide for excursions--Bombay and Khandala. International Group Symposium. 29 pp. Sukheswala, R.N., 1956. Notes on the field occurrence and petrography of the rocks of the Bombay Island, Bombay. Trans. Min., Geol. Metall. Inst. India 50, 101 ­ 126. Sukheswala, R.N., Awate, G.S., 1957. Correlation of the ash beds occurring in the western parts of Salsette Island, Bombay. J. Univ. Bombay, 48 ­ 52. Sweet, A.R., Braman, D.R., Lerbekmo, J.F., 1999. Sequential palynological changes across the composite Cretaceous­Tertiary (K­T) boundary claystone and contiguous strata, western Canada and Montana, USA. Can. J. Earth Sci. 36, 743 ­ 768. Tandon, S.K., 2002. Records of the influence of Deccan volcanism on contemporary sedimentary environments in central India. Sediment. Geol. 147, 177 ­ 192. Tolia, N., Sethna, S.F., 1990. Lopolithic intrusion of basalt in the intertrappeans at Amboli Hill, Jogeshwari, Bombay. J. Geol. Soc. India 35, 524 ­ 528. Tyson, R.V., 1985. Palynofacies and sedimentology of some late Jurassic sediments from the British Isles and northern North Sea. PhD thesis, The Open University, Milton Keynes, UK.



30 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


Notice: fwrite(): send of 199 bytes failed with errno=104 Connection reset by peer in /home/ on line 531