| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 41 Number 7 Pages 1099-1120 2000
© Oxford University Press 2000
Geochemistry of Flood Basalts of the Toranmal Section, Northern Deccan Traps, India: Implications for Regional Deccan Stratigraphy
1SCHOOL OF OCEAN AND EARTH SCIENCE AND TECHNOLOGY, UNIVERSITY OF HAWAII, HONOLULU, HI 96822, USA
2DEPARTMENT OF EARTH SCIENCES, INDIAN INSTITUTE OF TECHNOLOGY, POWAI, BOMBAY 400 076, INDIA
Received August 20, 1999; Revised typescript accepted February 18, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Tholeiitic lavas forming a flood basalt sequence of 870 m thickness at Toranmal in the northern Deccan Traps have a large range in isotopic ratios [
Nd(t) = +2·1 to –15·7, (87Sr/86Sr)t = 0·70467–0·71416, 206Pb/204Pb = 16·699–20·246], similar to that of lavas in the well-studied southwestern part of the province. The basalts with the lowest Nd(t) values display distinctive lows at Nb, Ta, P and Ti, and large positive Pb spikes in their primitive-mantle-normalized element patterns, indicative of significant continental lithospheric influence in their petrogenesis. As in much of the southwestern Deccan, Nd(t) exhibits a rough negative correlation with Mg/Fe and SiO2 and a positive correlation with Fe, consistent with temperature-controlled assimilation. Overall, the Toranmal section appears distinct from sections in the northwestern sector of the province; however, some Toranmal basalts are isotopically and chemically similar to flows in the northeastern Deccan, and a thick pile of lavas resembling the Poladpur Fm of the southwestern Deccan, the closest type-sections of which lie 
380 km to the south, is present. If it indeed represents a northern remnant of this formation, the Poladpur Fm, which also extends far into the central and southeastern parts of the province, is one of the most widespread of Deccan formations, with a possible original extent 
3 x 105 km2. The Ambenali Fm, which forms a thick sequence lying above the Poladpur in the southwestern Deccan, is not present at Toranmal. Several flows have broad geochemical affinities with the southwestern Bushe and Mahabaleshwar formations, which respectively lie below the Poladpur and above the Ambenali; however, these flows are not in the southwestern stratigraphic order and are probably of relatively local origin as dikes compositionally very similar to these flows are present at Toranmal and elsewhere in the vicinity. KEY WORDS: Deccan Traps; geochemical stratigraphy; flood basalts; large igneous provinces
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
An understanding of the nature and magnitude of magma generation, storage, and transport in flood basalt provinces requires knowledge of the stratigraphic and geochemical relationships between lava sequences in different regions of any given province. For most provinces, including the Deccan Traps of India (Fig. 1), such knowledge is still fragmentary. Although much reduced by erosion and, in the west, subsidence below sea level, the on-land extent of the Deccan remains vast, at 500 000 km2. Recent 40Ar–39Ar ages and Re–Os isotopic data for basalts from different parts of the province indicate that most of it formed in a geologically brief period around 67–64 Ma (e.g. Courtillot et al., 1988
; Duncan & Pyle, 1988; Venkatesan et al., 1993; Baksi, 1994; Baksi et al., 1994; Bhattacharji et al., 1996; Sheth et al., 1997; Allègre et al., 1999). Combined field and geochemical work in the 1980s produced a stratigraphic framework for the well-exposed southwestern sector (boxed area in Fig. 1) that divides the lava sequence in this region into three subgroups and 11 formations with a maximum stratigraphic thickness of 3·4 km (Table 1; e.g. Cox & Hawkesworth, 1985; Beane et al., 1986; Khadri et al., 1988; Subbarao et al., 1988; Lightfoot et al., 1990). More recent work has shown that several of these formations extend into the central Deccan (particularly the Khandala and Poladpur formations; Peng, 1998; K. V. Subbarao et al., unpublished data, 1998) and the southeastern Deccan (particularly the Poladpur and Ambenali formations; Mitchell & Widdowson, 1991; Bilgrami, 1999) for as much as 350 km from their southwestern type-sections. In addition, lavas isotopically and chemically equivalent to those of the Ambenali Formation (Fm) are present in the northeastern sector of the province as far as 900 km from their southwesternmost counterparts (Mahoney, 1988; Deshmukh et al., 1996; Peng et al., 1998).
|
|
However, many of the flows in the northeastern region (Mhow–Chikaldara–Jabalpur area), although similar to the southwestern Poladpur Fm in their chemical and Nd–Sr isotopic characteristics, possess Pb isotopic ratios higher than those of Poladpur lavas and thus must have taken somewhat different pathways through the crust on their way to the surface (Peng et al., 1998). Lavas in the northwestern sector of the province (from approximately Rajpipla westward) differ markedly as a group from those in the other regions studied. In particular, whereas the latter are almost exclusively tholeiitic, many northwestern lavas are relatively alkalic (e.g. Krishnamurthy & Cox, 1977; Mahoney et al., 1985; Melluso et al., 1995; Peng & Mahoney, 1995).
The relationships between the northwestern, northeastern and southern lava sequences—the extent to which they are interdigitated, or stratigraphically or structurally isolated from one another—are unknown. In part, this uncertainty reflects a lack of combined Nd–Sr–Pb isotopic and chemical studies in the area between these three major regions of the province; that is, roughly in the area between Rajpipla, Igatpuri, Buldana and Mhow. Here, we discuss the elemental and isotopic geochemistry and stratigraphy of the lava pile at Toranmal, the thickest in this area (Fig. 1).
|
FIELD GEOLOGY, PETROGRAPHY AND ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Toranmal (1152 m; 21°53’N, 74°28’E) forms a prominent peak of the Satpura range, which constitutes a horst separating the Narmada and Tapi grabens (along which flow rivers of the same names; Fig. 1). The Tapi graben is downfaulted along the Satpura Foothill Fault (Guha, 1995
; Sheth, 1998), which has produced a steep scarp in the vicinity of Toranmal; however, we have found no evidence of faulting within the Toranmal section itself. The section exposes a basalt sequence of 870 m thickness, intruded by several approximately east–west-trending basaltic and doleritic dikes, none of which are obvious feeders to the flows. To our knowledge, the only previous geochemical work on this section was carried out by Nair et al. (1996), who analyzed major elements and a suite of trace elements by X-ray fluorescence spectrometry (XRFS). We have identified, conservatively, 26 flows and two interspersed bole horizons on the basis of exposed flow boundaries, the presence of major vesicular–amygdular zones interpreted as flow tops, simple or compound nature of outcrops, and macroscopic and microscopic petrography (see Fig. 2). The flows have northerly dips of 2–3° (see Nair et al., 1996) and thicknesses of 10–120 m. Except for two compound flows at about 450 m and 710 m, the lava pile consists of simple flows, most of which exhibit large cooling columns (lower, and sometimes also upper, colonnades with middle entablatures). The thickest flows identified in our field section are thicker than normal for simple flood basalt flows, possibly indicating the presence of one or more unexposed flow boundaries. Nair et al. (1996) reported 41 flows in this section (see fig. 1 of their paper) but relied on criteria that we consider unreliable; specifically, they inferred flow boundaries between amygdaloidal zones and underlying massive zones, and between outcrops having different weathering patterns, including the presence or absence of spheroidal weathering (K. K. K. Nair, personal communication, 1999).
|
Spheroidal weathering is common, and soil development and extensive weathering significantly restrict the number of outcrops suitable for sampling. From examination of samples with the naked eye and hand lens, we classified the basalts into aphyric, plagioclase-phyric and giant plagioclase (GPB) types, the last having plagioclase phenocrysts >2 cm in length. Nair et al. (1996) reported no GPBs, but we encountered one at 700 m elevation (represented by sample SH105). A roughly equal number of macroscopically aphyric and plagioclase-phyric flows are present (Fig. 2), although most of the ‘aphyric’ flows are seen to be microphyric in thin section. The phenocryst assemblage of the Toranmal lavas is not notably different from that in the southwestern and northeastern Deccan (see Nair et al., 1996); however, weathering alteration tends to be more extensive than we have seen in most samples that we have collected from those regions (see also below), although primary textures remain easily recognizable. Plagioclase is the most abundant microphenocryst and is often accompanied by augite and altered olivine; the phenocrysts commonly form clots, producing a cumulophyric texture. Groundmasses are fine grained with intersertal or intergranular textures, and consist of plagioclase and augite with minor iron oxides. The dikes are texturally similar to the flows, although slightly coarser grained, and the boles are composed of altered tuffaceous material.
After rejecting the most altered samples, we chose 27 for geochemical work (23 flows, three dikes and a sample from a boulder that probably fell from a nearby flow outcrop; see Fig. 2). Major elements were analyzed by XRFS and trace elements by inductively coupled plasma–mass spectrometry (ICP-MS) on agate- and alumina-ground powders at the University of Hawaii (Table 2) following the methods of Norrish & Chapell (1977) and Jain & Neal (1996), respectively. In addition, some Sr abundances were measured by inductively coupled plasma–atomic emission spectrometry at the Indian Institute of Technology, Bombay. Two dike samples and 17 samples from flows were analyzed for Nd, Sr and Pb isotopic ratios and abundances of Nd, Sm, Sr, Rb and Pb by isotope dilution at the University of Hawaii [Table 3; see Mahoney et al. (1991) for methods]; as in our previous studies, isotopic work was carried out on small, hand-picked rock chips cleaned briefly in weak acid to avoid possible Pb contamination (e.g. Peng & Mahoney, 1995). These small, picked chips may not always be representative of the bulk-rock mineralogy, particularly in coarser-grained or patchily altered samples; thus, although the isotope-dilution and ICP-MS values are generally close, we use the isotope-dilution abundance data here only for age-correcting Sr and Nd isotopic ratios. For direct comparison with the Pb isotope data in the southwestern Deccan dataset, very few of which are age corrected, Pb isotope ratios are present-day values (note, however, that age corrections would generally be very small relative to the variation seen within individual Deccan formations; e.g. 238U/204Pb estimated from whole-rock ICP-MS data ranges between six and 20, corresponding to a 66 Ma age correction of only –0·06 to –0·2 in 206Pb/204Pb).
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Isotopic data
For correlation of physically separated lava piles, elemental, isotopic, stratigraphic and absolute-age relationships must all be evaluated. In general, assuming flow ages allow a correlation, Nd–Sr–Pb isotopic ratios can be critical because, unlike elemental abundances, they are not modified significantly by post-contamination differentiation during flowage or in magma chambers, local-scale crystal accumulation or moderate amounts of subaerial weathering alteration (see below), and tend to be more sensitive to differences in amount of contamination and end-member composition. Combined Nd–Pb–Sr isotopic data are a powerful tool for regional stratigraphic correlation in the southwestern, southeastern and central Deccan, because the data fields of the different southwestern formations display only limited overlap with each other in one or more of the isotope diagrams of Fig. 3; the Khandala, Bushe, Poladpur, Ambenali and Mahabaleshwar formations are particularly well characterized isotopically. The Toranmal samples encompass a wide range of isotopic values: e.g.
Nd(t) = +2·1 to –15·7 (t = 66 Ma), (87Sr/86Sr)t = 0·70467–0·71416, and present-day 206Pb/204Pb = 16·699–20·246. However, most of the data lie in or rather close to the Poladpur, Bushe or Mahabaleshwar Fm fields in Fig. 3, indicating the operation of similar processes in the petrogenesis of the Toranmal basalts and these southwestern formations.
|
, dikes). Fields are shown for southwestern Deccan formations, lavas of the Jabalpur area, and normal MORB (mid-ocean ridge basalts) of the Central Indian Ridge (CIR) (from Peng et al., 1998
In particular, data for many of the lavas fall within the field of the Poladpur Fm in all isotopic diagrams (samples SH93, -94, -99, -100 and SH109–112); these flows encompass a large portion of the Toranmal section. Flows with Poladpur-type Nd–Pb–Sr isotopic ratios also have been found recently in the Shahada–Shirpur area (Fig. 1) some 50–80 km to the south of Toranmal (Chandrasekharam et al., 1999). The GPB sample, SH105, is Poladpur-like in its Nd and Pb isotopic ratios (Fig. 3a), but its relatively low (87Sr/86Sr)t value puts it outside the Poladpur field in Fig. 3b and c. Likewise, data for samples SH102 and SH104, from two thick flows in the middle of the section, lie within the Poladpur field in the Nd–Sr isotope diagram (Fig. 3c) but fall well to the high-206Pb/204Pb side of this field in Fig. 3a and b; significantly, this same combination of features is seen in many lavas in the northeastern part of the Deccan (Jabalpur–Chikaldara–Mhow area), where truly southwestern-Poladpur-type Pb isotopic compositions appear to be rare (Peng et al., 1998).
In contrast to the Poladpur-like basalts, samples SH113, -114 and -116 from flows in the upper part of the sequence, SH90 from a flow near the base, and dike SH89 have distinctive, low-Nd signatures equivalent to those of the southwestern Bushe Fm in Fig. 3a. However, the (87Sr/86Sr)t ratios of all of these samples are substantially lower, relative to their Nd(t) values, than those of any Bushe Fm lavas yet analyzed (Fig. 3b, c; note that this is true whether or not the Sr isotope ratios are age corrected). Several flows with broadly similar isotopic compositions have been documented within the northeastern sector of the Deccan in the Mhow and Chikaldara sections (Peng et al., 1998), whereas a dike in the Shahada–Shirpur area south of Toranmal has similar Pb isotope compositions and even more extreme Nd and Sr isotopic values [Nd(t) = –20·2, (87Sr/86Sr)t = 0·72315; Chandrasekharam et al., 1999].
Other dikes in the Shahada–Shirpur area are known to have Mahabaleshwar-Fm-type isotopic signatures (Sheth et al., 1997; Chandrasekharam et al., 1999). At Toranmal, data for sample SH115, from a flow near the top of the section, plot within the Mahabaleshwar Fm field in all isotopic diagrams. Sample SH95, from a flow much lower in the section, is also Mahabaleshwar-like in Fig. 3a, but has (87Sr/86Sr)t slightly higher than found in Mahabaleshwar Fm lavas (Fig. 3b, c). A dike cutting this flow, SH96, has isotopic characteristics very similar to those of the flow.
Elemental data
All of the Toranmal basalts are differentiated, with MgO between 2·98 and 7·11 wt %, and mg-number between 31·6 and 60·3; where mg-number = 100 x [atomic Mg2+/(Mg2+ + Fe2+)], assuming 85% of total Fe is Fe2+. Nair et al. (1996) showed that the major-element relationships are compatible with fractionation of a gabbroic assemblage of plagioclase, clinopyroxene and olivine, consistent with the presence of these phases as phenocrysts in the lavas and similar to most southwestern and northeastern Deccan basalts. They also observed that many of the Toranmal lavas have elemental affinities with formations in the upper part of the southwestern Deccan sequence. As with virtually all southwestern and northeastern Deccan basalts, and consistent with petrographic observations, the Toranmal data lie in the field for tholeiitic basalts in a total alkalis vs silica diagram (not shown). The only exception is sample SH90, which has unusually high Na2O (4·31 wt %) and K2O (2·12 wt %) contents; in contrast, several of the other samples have total alkali-element abundances lower than those of most Deccan lavas. These differences at least partly reflect the comparatively intense subaerial weathering alteration that has affected much of the Toranmal section. This alteration is also evident in the variable but generally high weight-loss-on-ignition values (LOI) in Table 2, which reach as high as 6·68 wt % in SH99.
Previous work on Deccan lavas has shown that among the commonly analyzed elements, K and Rb are redistributed (locally decreased or increased) most strongly by such alteration; Ba and Na can also be redistributed at higher levels of alteration, whereas isotopic ratios do not appear to be affected significantly (e.g. Cox & Hawkesworth, 1985; Mahoney et al., 1985; Beane et al., 1986; Beane, 1988; Lightfoot & Hawkesworth, 1988; Peng, 1998). Of course, with advanced subaerial alteration the abundances of many elements, including Si, Ca, P, Ni, Sr, Y and the lanthanide rare earths can be modified relative to elements such as Al, Nb, Zr, Fe, Cr and Ti (e.g. Price et al., 1991; Fodor et al., 1992; Wilkins et al., 1994; Widdowson & Cox, 1996; Moore, 1997). One ratio that appears to change little, even during extreme alteration, is Nb/Zr; in the southern Deccan, this ratio is among the most useful, along with isotopic ratios, in distinguishing highly altered Mahabaleshwar Fm lavas (Nb/Zr > 0·10) from those of the Ambenali and Poladpur formations (which both have Nb/Zr in the 0·05–0·10 range; e.g. Widdowson & Cox, 1996). For mildly altered rocks in the south, Sr, Ba, Ba/Y and even Rb have been used to help distinguish among these formations (e.g. Cox & Hawkesworth, 1985; Devey & Lightfoot, 1986; Mitchell & Widdowson, 1991).
Figure 4 summarizes several aspects of the geochemical stratigraphy at Toranmal, showing the variation with altitude in mg-number, several interelement ratios and Sr isotopes. Nair et al. (1996) presented a somewhat similar figure of elemental variations. Unfortunately, detailed comparison of elements analyzed in common in the two datasets is seriously handicapped because the two suites of samples cannot be correlated; our field stratigraphic section (Fig. 2) is not comparable with that of Nair et al. (their fig. 1) as a result of the different criteria used for demarcating flows coupled with what appear to be significant disagreements between their altimeter readings and ours. On x–y plots, both sets of data define generally similar trends for most of the alteration-resistant elements. Differences in some elements, such as La/Ce, may indicate significant interlaboratory biases, whereas the very low K, Rb and Ba contents reported by Nair et al. for a number of samples may reflect advanced alteration.
|
All of the Toranmal basalts are enriched in the light rare earth elements (LREE; e.g. La/Sm = 1·8–4·7), all have peaks at Pb in their primitive-mantle-normalized incompatible element diagrams, and most have normalized Nb/La and Ta/La <1. Much of the section is composed of flows with a limited range of Th/Nb (0·1–0·2), Pb/Nb (0·1–0·3), La/Nb (0·9–1·5) and Nb/Zr (0·06–0·10) ratios (Fig. 4b, c); those that we analyzed isotopically have Poladpur-type Nd–Sr isotopic compositions or broadly Mahabaleshwar-like isotopic values. A selection of incompatible element patterns for this group is presented in Fig. 5. Alteration effects are most evident in highly variable contents of Rb, K and, to a lesser extent, Ba among lavas with otherwise rather similar incompatible element characteristics (e.g. compare the patterns for SH88 and -93, and for SH100 and -102 in Fig. 5a, b). Alteration-related modification of incompatible elements appears most severe in isotopically Poladpur-like sample SH99 (Fig. 5c). The much fresher SH97 has similar abundances of Th, U, Nb, Ta, Pb, Zr and Ti, but comparison of the two patterns indicates a nearly total loss of Rb, K and Ba, and significant loss of Sr and P in SH99. In addition, partial loss of the lanthanide rare earths in SH99 is suggested by their low abundances relative to Ti, Zr, Nb and Ta. The mg-number is essentially the same (45·6 and 45·0), but SiO2 and Ni are both much higher in SH99 (53·22 vs 48·22 wt % and 110 ppm vs 83 ppm, respectively), whereas CaO/Al2O3 is much lower (0·60 vs 0·76). Importantly, despite the variable alteration in this group of samples, many have patterns whose shapes strongly resemble those of southwestern Poladpur Fm lavas, particularly in the more alteration-resistant elements; Figure 5d–f provides several comparisons with Poladpur patterns.
|
Four of the most differentiated flows (i.e. with the lowest mg-numbers, 31·6–43·7), SH92, -94, -106 and -109, possess Th/Nb, Pb/Nb, La/Nb and Sm/Yb values similar to those of the Poladpur-like samples, but have significantly greater overall abundances of incompatible elements as well as higher La/Yb and/or Nb/Zr ratios; these flows appear in both the lower and middle levels of the section (e.g. Fig. 4e). The two we analyzed isotopically (SH94 and -109) have nearly identical isotopic ratios falling within the Poladpur Fm field in Fig. 3. The incompatible element pattern for SH106 is characterized by higher La/Nb and lower Nb/Zr than the other three samples; it is more enriched in the highly incompatible elements than any known Poladpur Fm lavas but matches rather well with the pattern of the Monkey Hill member of the Khandala Fm (Fig. 6a). The incompatible element patterns of SH92, -94 and -109, which are closely similar to each other, are not matched by either Poladpur or Khandala Fm patterns, although some Khandala Fm patterns have a similar overall slope from Rb to Lu (e.g. the Giravli member; Fig. 6b).
|
The five low-
Nd samples (dike SH89, flows SH90, -113, -114 and -116) display significantly elevated Pb/Nb and Th/Nb values (Fig. 4b, c), relatively high SiO2 (51·18–52·80 wt %) and low total iron (10·91–11·43 wt % for three of the samples), and four of these samples have the highest mg-numbers (53·1–60·3) in our collection. Distinctive features of their incompatible element patterns (Fig. 6c) are a very prominent Pb peak, lows at Nb, Ta, P and Ti, and relatively flat right-hand portions (i.e. from Nd to Lu, also evident in their low Sm/Yb values in Fig. 4d). In all of these respects, these samples resemble Bushe Fm lavas (a pattern for a lava of the Pingalvadi member of the Bushe is shown for comparison in Fig. 6d). Sample SH90 differs from the others in that it has lower abundances of the REE relative to Nb, Ta, Zr and Ti, such that a peak at Zr is evident in its incompatible element pattern (Fig. 6d). As with the Poladpur-like sample SH99, this feature is likely to be a result of the comparatively advanced alteration in SH90; however, whereas SH99 has lost most of its Rb, Ba and K, SH90 has high Rb, Ba and K (and, as noted above, high Na2O). Finally, a pattern intermediate between the Bushe-like and Poladpur-like patterns is exhibited by isotopically Poladpur-like sample SH112; in particular, the portion of its pattern from Nd to Lu is rather flat (Fig. 6e). Sample SH115, which is Mahabaleshwar-like in all isotope diagrams, lacks the degree of relative LREE and Th enrichment seen in the Mahabaleshwar Fm; instead, except for its markedly higher Pb content, its pattern is more similar to those of Ambenali Fm lavas (Fig. 6f). The patterns of the other two isotopically rather Mahabaleshwar-like samples, SH95 and -96, are similar to each other in most elements just as they are isotopically; their patterns are not Mahabaleshwar-like but resemble those of some Poladpur Fm lavas (Fig. 5f).
To include the major elements in our evaluation of the Toranmal basalts, we followed the rationale of previous workers (e.g. Beane et al., 1986; Beane, 1988; Khadri et al., 1988; Peng et al., 1998) and performed a discriminant-function analysis comparing the combined data for the major elements and several commonly analyzed trace elements with those of the southwestern Deccan formations, for which XRFS data on some 1400 samples are available. The analysis was performed as described by Peng et al. (1998), using the same software (SPSS 7.5 for Windows, professional version) and standard set of southwestern Deccan basalt data; the trace elements used were Ni, Y, Zr, Nb, Ba and Sr, for which abundant good-quality data exist in the southwestern Deccan dataset (data for many of the other trace elements in Table 2 are limited to relatively few southwestern samples or are of highly variable quality in the southwestern dataset). Table 4 summarizes the results, listing the closest southwestern-formation match and corresponding Mahalanobis distance for each Toranmal sample. Samples having no match or one at a Mahalanobis distance greater than 20 (corresponding to a conditional probability near zero) are shown with question marks. Figure 7 compares scores of the first two discriminant functions, which together account for 73% of the total variation in the southwestern dataset, with the fields defined by the relevant southwestern formations.
|
|
Despite the variable alteration in the Toranmal section, many of the basalts identified as Poladpur-like on the basis of isotopic and incompatible element signatures are matched with either the Poladpur (SH93, -101, -103, -104, -110) or Ambenali (SH100, -102, -107) formations in Table 4 at Mahalanobis distances similar to those pertaining among the southwestern basalts. The close similarity in major elements and the commonly analyzed trace elements between the Ambenali chemical type and many positive-Nd Poladpur lavas has been emphasized by previous workers (in relatively fresh samples, Ba, K and Rb are higher in Poladpur than in Ambenali lavas, but alteration can obscure the difference); a clear distinction is provided only by isotopic measurements (and seldom-analyzed elements such as Pb) (e.g. Cox & Hawkesworth, 1985; Devey & Lightfoot, 1986; Beane, 1988). Therefore, these results support an overall Poladpur affinity. The correspondence to the Poladpur is not perfect, however: low-mg-number lavas with incompatible element enrichment similar to SH92, -94, -106 and -109 have not been reported from the Poladpur Fm, and these four samples are not matched with any southwestern formation in Table 4 (see also Fig. 7). SH88, another low-mg-number lava, is matched with the Bhimashankar Fm, despite having incompatible-element characteristics similar to those of SH93 for most elements (Fig. 5a). Likewise, SH111, -112 (with the flat Nd-to-Lu pattern), and the highly altered SH99, all isotopically Poladpur-like, have no close southwestern counterparts in Table 4.
All of the low-Nd samples with Bushe-Fm-like incompatible element patterns, SH89, -113, -114 and -116, are matched with the Bushe Fm (SH90, strongly modified by alteration, is also matched with the Bushe Fm, but at a very large Mahalanobis distance of 84·2). The isotopically Mahabaleshwar-like sample SH115 is matched with the Ambenali Fm, consistent with the overall shape of its incompatible element pattern (see above). So too are the other two isotopically rather Mahabaleshwar-like samples, SH95 and SH96, which, however, have Poladpur-like incompatible element patterns (Fig. 5f).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Contamination
Peng et al. (1994)
concluded that two stages of mixing (indicated schematically by arrows in Fig. 3a, b) were required to explain the subparallel fields defined by many of the southwestern Deccan formations in Nd–Pb and Sr–Pb isotope diagrams. They postulated the first stage to be between an isotopically Ambenali-type end-member (representing a major, transitional-MORB-like source, presumably in the convecting mantle) and high-206Pb/204Pb, high-87Sr/86Sr, low-Nd lithospheric mantle material or amphibolite-grade lower crust, a crustal role being supported by elevated oxygen isotope ratios in phenocrysts of lavas that appeared to have been affected mainly by this stage of contamination. The second stage was proposed to have occurred between the products of variable amounts of first-stage mixing and several different low-206Pb/204Pb materials, some probably granulitic, giving rise to the elongated isotopic fields in Fig. 3a and b. Although most data available for the northwestern Deccan do not fit this two-stage model (Peng & Mahoney, 1995; J. Mahoney & L. Melluso, unpublished data, 1998), recent results for the northeastern Deccan support it: lavas in the Jabalpur area define an array paralleling part of the stage 1 mixing trend, and many Chikaldara and Mhow basalts have similarly high 206Pb/204Pb for their Nd and Sr isotopic values (i.e. like chemically similar Jabalpur-area lavas, their data points lie well to the right of the Poladpur Fm field in Fig. 3a, b, similar to SH102 and -104 at Toranmal). Thus, many of the northeastern basalts appear to have been affected less by the second stage of mixing than most of the southwestern lavas (Peng et al., 1998). In addition, Ambenali-type basalts are present in the Chikaldara and Jabalpur areas.
The Toranmal data are also consistent with the two-stage model. Isotopic data for many of the flows form an array, corresponding to a variable stage 2 influence, which considerably overlaps the Poladpur Fm field and, with SH102 and -104, extends to the high-206Pb/204Pb side of it. Values for the five low-Nd(t) basalts trend subparallel to the Bushe Fm field, whereas those for SH115, -95, and -96 are in or close to the Mahabaleshwar Fm field, which presumably reflects little or no contamination by the stage 1 end-member.
Values of Pb/Nb, Th/Nb and La/Nb vs Nd(t) or (87Sr/86Sr)t show good overall correlations (e.g. Fig. 8a, b) consistent with variable amounts of contamination involving high-Pb, -Th, -La, low-Nb continental lithospheric material. As expected, correlations between such elemental indicators of contamination and Pb isotopes are much poorer because, in contrast to Nd and 87Sr/86Sr, continental end-members with Pb isotope ratios both lower and higher than those of the isotopically Ambenali-like mantle source were presumably involved. Also evident are positive correlations of Nd(t) with TiO2 (Fig. 8c) and Fe2O3* (total iron as Fe2O3; Fig. 8d), and a rough negative correlation of Nd(t) with SiO2 (excepting two low-MgO samples and the highly altered SH99; Fig. 8e). Despite variable amounts of fractionation, Fe2O3* and MgO have virtually no correlation with each other but, as Fig. 8f shows, a negative overall correlation is present between Nd(t) and MgO/Fe2O3* or mg-number. Similarly, using an alteration-resistant, moderately incompatible element such as Zr as an index of progressive fractionation (e.g. Hooper, 1994), negative correlations with elemental ratios such as Pb/Nb and Th/Nb are seen (not shown). Thus, the most contaminated samples tend to be the least differentiated and the least contaminated samples are all rather highly evolved.
|
A very similar situation pertains in the southwestern Deccan, particularly in the Bushe–Poladpur–Ambenali part of the sequence, where it has been interpreted by most researchers to reflect temperature-controlled assimilation (e.g. Cox & Hawkesworth, 1985
; Devey & Cox, 1987; Mahoney, 1988). In this process, hotter, more primitive magmas experience greater amounts of contamination than do cooler, more evolved ones; more primitive magmas also have lower abundances of incompatible elements than their more evolved counterparts, so that a given mass fraction of contamination has a larger effect on Nd, Sr and Pb isotope ratios and incompatible trace elements. In principle, contaminants can be either crustal or suitable lithospheric mantle materials, although the high magmatic 
18O values (up to +7·6
vs +5·9 for the Ambenali) recorded in pyroxene and plagioclase phenocrysts of the Bushe Fm favor crustal involvement (Peng et al., 1994). The similar major element, trace element and isotopic characteristics of the Bushe Fm and low-Nd Toranmal basalts imply similar sources and a similar contamination process. Also, in each case there was apparently nothing to prevent some post-contamination differentiation (e.g. SH116 has an mg-number of only 47·8, and SH89, -90 and -113, with identical (within errors) Nd(t) of –12·6 to –12·8, have mg-numbers of 53·1–59·1).
Nair et al. (1996) took Ba/Ti as an index of contamination; noting a rough positive correlation between Ba/Ti and mg-number, they argued that it might reflect progressively greater insulation of crustal wall rocks by solidification of basaltic magma on feeder-conduit walls as the magma evolved. Although possible in principle, this explanation implies a systematic decrease up-section in the level of contamination (Mahoney et al., 1982; Cox & Hawkesworth, 1985; Devey & Cox, 1987), which is not seen at Toranmal (e.g. Fig. 4a–c).
Effects arising from variable pre- and post-contamination differentiation, the specific nature of the contamination process (e.g. partial melting of wall-rock during ascent or in a magma chamber, selective contamination, etc.), and the two stages of contamination by different low-Nd end-members as required in the two-stage model are difficult to quantify. The same is true of alteration effects for many elements at Toranmal. Moreover, although Archean or early Proterozoic end-members are indicated as probable contaminants by the terrains around and beneath the Deccan Traps, the specific contaminants involved are unknown, and in some cases even their crustal vs lithospheric–mantle nature is unclear (e.g. Mahoney, 1988; Lightfoot et al., 1990; Peng et al., 1994; Turner & Hawkesworth, 1995; Lassiter & DePaolo, 1997; Allègre et al., 1999). Detailed modeling of contamination is thus precluded. However, at the most basic level, we note that incompatible element patterns broadly similar to those of the elementally Poladpur-like Toranmal lavas can be produced for most alteration-resistant elements just by assuming 3–13% bulk contamination of chemically Ambenali-like magmas with globally averaged Archean crust of intermediate to felsic composition, and allowing for variable fractionation (e.g. Fig. 9a, which shows an example assuming a purely felsic crustal end-member). Similar overall levels of contamination were inferred for the Poladpur Fm by Lightfoot (1985).
|
The low-
Nd Toranmal basalts are arguably more problematic in terms of contamination, yet reasonably good incompatible-element fits for most elements (including Nb and Ta; see Arndt & Christensen, 1992) also can be achieved for these basalts (excluding the highly altered SH90) by assuming 20–25% bulk contamination of suitable magma compositions, similar to estimates for the Bushe Fm (e.g. Lightfoot, 1985; Mahoney, 1988). Figure 9b shows an example for SH114 (the lava with the highest mg-number), assuming a purely felsic average Archean crustal end-member and no post-contamination fractionation. Of course, given the wide range of rock compositions present in Archean and early Proterozoic terrains, such averages represent at best only crude approximations to actual contaminant compositions; also, bulk assimilation is unlikely to be the main process operating at large scales (e.g. Sinigoi et al., 1996). Nevertheless, in Fig. 8, schematic mixing curves are included to illustrate that key features of the low-Nd basalts are consistent with contamination by broadly similar material, followed by variable fractionation. For these curves, the high-Nd mantle end-member is assumed, as in the southwestern and northeastern Deccan (e.g. Peng et al., 1994, 1998, and references therein), to be isotopically equivalent to the least-contaminated Ambenali Fm basalts. However, because the Ambenali basalts themselves are differentiated (most are ferrobasalts) and also preserve a record of progressively decreasing degree of partial melting in the source (e.g. Mahoney et al., 1982; Cox & Hawkesworth, 1985; Devey, 1986; Devey & Cox, 1987; Lightfoot et al., 1990), they are not a suitable chemical end-member for the magmas inferred in the genesis of the low-Nd Toranmal (or Bushe Fm) basalts. For the illustrative mixes in Figs 8 and 9b we assumed an unfractionated, high-MgO basaltic composition and an average transitional-MORB (mid-ocean ridge basalt) incompatible element pattern, flatter than the Ambenali pattern (i.e. corresponding to a higher degree of partial melting) yet consistent with the required Ambenali source characteristics (e.g. Lightfoot & Hawkesworth, 1988; Mahoney, 1988). Evidence for the existence of magmas with relatively flat patterns before contamination is provided at Toranmal by the isotopically Poladpur-like sample SH112 [Nd(t) = +0·4], which reflects a much smaller amount of contamination than the low-Nd basalts yet, like them, has a relatively flat pattern from Nd to Lu (Fig. 6e); the SH112 flow lies just beneath the upper group of low-Nd lavas (Figs 2 and 4d). Also, little-contaminated lavas with rather flat incompatible element patterns and even higher Nd(t) (to +2·5, only slightly lower than the Ambenali Fm range) are present west of Toranmal in the northwestern sector of the Deccan (see next section).
Regional geochemical comparisons
Comparison with the northwestern Deccan
Geochemical stratigraphic work in the northwestern Deccan is severely hindered by a lack of good exposures. Relatively few multielement isotopic and comprehensive trace element data are available, and fewer 40Ar–39Ar ages and paleomagnetic measurements; nevertheless, the existing data permit some first-order conclusions to be made. Geochemical studies of this region have focused on cores from drillholes (penetrating lava sequences 150–420 m thick) near Dhandhuka, Wadhwan and Botad (Fig. 1), on outcrops in the Rajpipla–Navagam area 60–100 km west of Toranmal in the Narmada graben, and on relatively thin lava sections and intrusions in several other areas (Alexander & Gibson, 1977; Krishnamurthy & Cox, 1977, 1980; Mahoney et al., 1985; Melluso et al., 1995; Peng & Mahoney, 1995). As noted in the Introduction, many of the northwestern lavas are rather alkalic, unlike the Toranmal samples (except SH90, and its high alkali element content appears likely to be a result of alteration). Also, a number of the lavas in the drillholes (West, 1958; Krishnamurthy & Cox, 1977) and elsewhere (Melluso et al., 1995) are picritic or high-MgO basalts, which are not present at Toranmal. Further, the drillhole lavas define very different Nd–Pb and Sr–Pb isotopic trends from any of the southwestern or Toranmal basalts, such that in Fig. 3a and b their data fields (not shown) would be nearly vertical (Peng & Mahoney, 1995).
In addition to relatively alkalic lavas similar to those in the drillholes, Melluso et al. (1995) described several areas where tholeiitic flows with fairly flat primitive-mantle-normalized Nd-to-Lu patterns and variable enrichment in the highly incompatible elements are exposed. Unlike the low-Nd Toranmal basalts, these lavas range from picritic basalts to rather low-MgO basalts. Isotopic data for these lavas (J. Mahoney & L. Melluso, unpublished data, 1998) reveal a range of values [e.g. Nd(t) = +2·5 to –6·0] very different from those of the low-Nd Toranmal basalts. The only hints of a possible petrogenetic link between these flows and the low-Nd basalts are provided by (1) SH112 [with Nd(t) = +0·4], which lies immediately beneath the upper three low-Nd flows, and (2) the relatively high-MgO pre-contamination magmas inferred for these Toranmal flows (see the previous section); it may be that similar high-MgO magmas with relatively flat incompatible element patterns were involved in the generation of both the low-Nd Toranmal basalts and these northwestern lavas. If so, the contamination processes and/or pathways involved appear to have led to different end-products, in general. A possible exception is a flow with broadly Bushe-like Sr and Nd isotopic ratios reported from the extreme northwestern corner of the Deccan (Krishnamurthy et al., 1988).
Comparatively close to Toranmal, the flows in the Rajpipla–Navagam area consist of interbedded, isotopically similar alkalic and tholeiitic basalts, which define a Nd–Sr isotopic array that coincides with the Poladpur Fm field at positive values of Nd(t) but, unlike the Poladpur-type Toranmal lavas, diverges from it at lower values toward lower (87Sr/86Sr)t (Mahoney et al., 1985). Pb isotopes have not been measured for these lavas, and trace element data are limited. However, all of the lavas analyzed for Nb, for example, have high Nb contents; among the tholeiites, Nb varies from 27 to 52 ppm. Only the low-MgO (4·02–4·85 wt %) Toranmal flows SH92, -94 and -109 have Nb abundances in this range, but the Navagam tholeiites all have higher MgO, mostly in the 6·7–8·6 wt % range. The Navagam tholeiites also have Sr contents between 343 and 592 ppm, whereas the Toranmal flows all have values <307 ppm. Thus, at present, we cannot correlate the Navagam lavas with the Toranmal basalts.
Overall then, the currently available data show little correspondence between the Toranmal section and areas farther to the west. Major structures in this region that existed during the Deccan episode, such as the Cambay graben (which runs approximately north–south just east of Dhandhuka), may have impeded movement of flows between Toranmal and some of the more westerly areas. In addition, at least some of the northwestern lavas may lie at different levels in the Deccan volcanic stratigraphy than most of the Toranmal basalts. The depth to the base of the volcanic succession beneath Toranmal is unknown, but many of the northwestern basalts are near at least the local base: two of the drillholes bottomed in pre-Deccan sedimentary rocks, which also lie just beneath the Navagam section. However, the 40Ar–39Ar ages available for two Dhandhuka drillhole lavas are in the same range, within errors, as those for other parts of the Deccan Traps (R. A. Duncan, personal communication, 1997), including dikes that cut Poladpur-type lavas in the Shahada–Shirpur area (Sheth et al., 1997); thus any stratigraphic difference must correspond to a smaller difference in age than can at present be measured radiometrically.
Comparison with the southwestern and northeastern Deccan
The fact that a number of the Toranmal samples have both Poladpur-type isotopic and incompatible element signatures suggests that the Poladpur Fm may extend as far north as Toranmal. The results of our discriminant-function analysis using major elements and a subset of trace elements are more mixed but also indicate that some of the lavas strongly resemble the Poladpur basalts in the southwest. If the Poladpur Fm is present at Toranmal, it would mean that this formation is one of the most extensive in the Deccan Traps: the northernmost type-sections in the southwestern region are 380 km from Toranmal, somewhat south of the latitude of Bombay (e.g. Beane, 1988), and the southernmost known occurrences in the southeastern Deccan (Mitchell & Widdowson, 1991; Bilgrami, 1999) are 580 km from Toranmal. The only other formation that may have a comparable extent is the Ambenali (Peng et al., 1998).
The existence of sections containing Poladpur-type basalts in the corridor between Toranmal and the southwestern type-sections would effectively settle the issue. As noted earlier, broadly Poladpur-like flows are exposed in the Shahada–Shirpur area east of Kondaibar in thin (<100 m thick) sections that crop out above the alluvium filling the Tapi graben (Sheth, 1998; Chandrasekharam et al., 1999). Flows identified isotopically and chemically as Poladpur Fm are also present in several areas farther to the southeast, near Buldana (Peng, 1998; K. V. Subbarao et al., unpublished data, 1998). However, no geochemical data have been published for the crucial region between Igatpuri and Kondaibar–Shirpur (Fig. 1), although Subbarao et al. (1994) presented an interpretive north–south cross-section showing only formations below the Khandala Fm in this area.
In lieu of evidence for the Poladpur Fm in the Igatpuri–Kondaibar region, the presence of consistent, southwestern-type interformational stratigraphic relationships at Toranmal, such as Ambenali-type lavas overlying and/or Bushe-type ones underlying the Poladpur-type flows (see Table 1), would strongly support a direct link with the southwestern Deccan formations. Deshmukh et al. (1996) suggested that the Toranmal flows with low Ba/Ti might belong to an ‘eastern’ Ambenali Fm, which they postulated to be present across the northern and northeastern Deccan. The 690 m section south of Mhow (Fig. 1) contains no Ambenali-like lavas, but the chemically and isotopically Ambenali-type flows documented in the Chikaldara and Jabalpur areas indeed may represent a distal portion of the Ambenali Fm, and they lie above lavas with Poladpur-like Nd–Sr isotopic and elemental characteristics (Peng et al., 1998). However, none of the Toranmal samples possess the distinctive Ambenali-type isotopic signature or primitive-mantle-normalized incompatible element pattern. Those with the lowest Ba/Ti values mostly appear to record Ba loss during alteration (see above). Sample SH115 has an incompatible element pattern rather similar to an Ambenali Fm pattern but has a significant peak at Pb (Fig. 6f) and Mahabaleshwar-Fm-type isotopic ratios (Fig. 3). Therefore, we conclude that no truly Ambenali-type basalts are present in the Toranmal section.
Likewise, the Mahabaleshwar Fm, which overlies the Ambenali Fm, is not evident at Toranmal. Samples SH115, -96 and -95 all differ from Mahabaleshwar Fm basalts in ways not attributable simply to alteration, phenocryst accumulation or post-eruptive differentiation—and, importantly, only SH115 lies stratigraphically above the Poladpur-like basalts. The fact that SH96 is a dike, together with the presence of other isotopically Mahabaleshwar-like dikes in the Shahada–Shirpur area (Chandrasekharam et al., 1999), strongly suggests that the SH95 and -115 flows have a relatively local origin in the northern Deccan. Similar dikes and sills appear to exist farther east in the northeastern Deccan and beyond (Sen & Cohen, 1994; Kent et al., 1997; Peng et al., 1998).
Nor is the Bushe Fm obviously present at Toranmal. The low-Nd basalts have incompatible element patterns resembling those of Bushe lavas but their isotopic ratios fall outside the known Bushe Fm range in Fig. 3b and c. Thus, these basalts are unlikely to be simply northern extensions of the same Bushe flows seen in the southwest (i.e. there is no obvious reason why the northern portions of flows would consistently have lower Sr isotopic ratios than the southern portions), although they clearly reflect overall similarities in petrogenesis. Most importantly for the present discussion, the Bushe Fm lies below the Poladpur Fm (Table 1) but only SH90 is situated near the base of the Toranmal section; the other three low-Nd flows lie above the sequence of Poladpur-type lavas. Because (1) the SH89 dike closely resembles these flows both chemically and isotopically (see SH113 and -89 in Figs 3 and 6c), (2) dikes with rather similar elemental characteristics are present in the Nandurbar area near the Tapi River south of Toranmal (Melluso et al., 1999), and (3) a chemically similar dike with even more extreme isotopic ratios is known in the Shahada–Shirpur area (Chandrasekharam et al., 1999), a relatively local origin in the northern Deccan appears highly likely for these low-Nd lavas.
Several chemically and isotopically similar low-Nd flows are also present in the upper part of the Chikaldara section (within a group of Ambenali-type lavas) and in the upper and middle levels of the Mhow section (Peng et al., 1998). On the basis of the then-available XRFS elemental data, Deshmukh et al. (1996) speculated that such lavas might be correlated with the Panhala Fm at the top of the southwestern Deccan stratigraphic sequence (see Table 1). However, any direct connection with this formation is ruled out by the isotopic data and more comprehensive trace element data now available [this paper and Peng et al. (1998)], because the Panhala basalts do not have appropriate incompatible element patterns or isotopic signatures [e.g. the Desur unit of the Panhala has (87Sr/86Sr)t = 0·7073–0·7081 and Nd(t) = –6 to –10·6, and the isotopic ratios of other Panhala lavas closely resemble those of the Mahabaleshwar Fm; e.g. Lightfoot & Hawkesworth (1988)].
In short, the stratigraphic context of the Poladpur-like flows at Toranmal cannot be ascertained from the non-Poladpur-like lavas. It may be significant that, unlike the broadly Mahabaleshwar- and Bushe-like dikes, no obviously Poladpur-type dikes have yet been found in this region or in the northeastern Deccan. The apparent lack of such dikes may suggest a relatively distant eruptive origin for the Poladpur-type flows at Toranmal. However, we emphasize that in contrast to, for example, the Columbia River province (e.g. Hooper, 1997), understanding of the relationship between Deccan dikes and flows remains at a very preliminary level (e.g. Beane, 1988; Hooper, 1990; Bhattacharji et al., 1996). Indeed, very few combined Nd–Pb–Sr isotopic and comprehensive elemental data are yet available for dikes in any part of the Deccan.
In the southwestern part of the province, the Poladpur Fm comprises about 20 flows and has a maximum thickness of 375 m (e.g. Devey, 1986; Beane, 1988; Lightfoot et al., 1990). In the Toranmal section, the thickness of lavas with Poladpur-type Nd–Sr isotopic ratios and incompatible element patterns is 450–470 m; inclusion of the isotopically Poladpur-like low-MgO flows (SH92, -94, -106 and -109) increases this value to 550–570 m in at least 16 flows. If the Toranmal lavas indeed represent the Poladpur Fm, this formation thus would be of comparable thickness or even thicker far to the north than to the south of Bombay. If so, only a fraction of the geochemical diversity in the formation as a whole is seen in either the northern or the southern Deccan. For example, several lavas in the southwest have Nd(t) < –4 (see Poladpur field in Fig. 3c); values this low have so far not been found among the northern or northeastern Poladpur-type basalts. Likewise, as noted earlier, the abundant Poladpur-like flows in the northeastern Deccan (with a thickness of 340 m at Chikaldara) possess similar elemental and Nd–Sr isotopic characteristics to Poladpur flows in the south but generally have higher 206Pb/204Pb (Peng et al., 1998). At Toranmal, lavas with both southern-type and northeastern-type (i.e. SH102 and -104) isotopic characteristics are intercalated, providing an important link to the northeastern sections.
Interestingly, however, despite the relative proximity of the Toranmal and Mhow sections (Fig. 1), the detailed stratigraphic–geochemical correspondence between them is poor. The main similarity is that several low-Nd, broadly Bushe-like flows are present in both areas. No isotopically Mahabaleshwar-Fm-like basalts (like SH95 and -115) have been found in the Mhow area, and all of the lavas with Poladpur-like Nd–Sr isotopic signatures have high 206Pb/204Pb values, placing them to the right of the Poladpur Fm field in Fig. 3a and b, like only SH102 and -104 (of the samples analyzed) at Toranmal. Moreover, Toranmal lacks the chemically Khandala-Fm-like lavas that are abundant in the Mhow section (Peng et al., 1998). Thus, most flows reaching Toranmal apparently did not reach the Mhow area, and vice versa, or the two sections expose somewhat different stratigraphic levels (see below). Presumably, most flows sampled in the two areas were fed by different feeder dikes. The Mhow section is on the north side of the Narmada graben, which was tectonically active before, during and after the Deccan event (e.g. Sheth & Chandrasekharam, 1997, and references therein), and structures associated with this feature may have effectively hindered most flows traveling in a direction between Toranmal and Mhow.
|
CONCLUDING REMARKS |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Our results, together with those of Peng et al. (1998)
, Chandrasekharam et al. (1999) and K. V. Subbarao et al. (unpublished data, 1998), indicate the extent of broadly Poladpur-type basalts to be considerably greater than recognized previously. Consideration of all the documented occurrences suggests that such basalts originally may have covered an area of 300 000 km2 or more; in comparison, the largest units in the Columbia River province have an areal extent roughly a third this size (Tolan et al., 1989). However, determining whether the northern and northeastern Poladpur-type flows and the southern Poladpur lavas belong to a single volcanostratigraphic formation, in the normal sense of a packet of flows emplaced everywhere in the same relatively brief interval of time, remains a critical problem.
The thick southwestern Deccan lava pile contains only one magnetic polarity reversal, R (reversed) to N (normal), with formations below the Mahabaleshwar Fm all having R polarity (e.g. Vandamme et al., 1991, and references therein). In contrast, Venkata Rao et al. (1996) reported a lower R interval in the Toranmal section between 295 and 445 m [elevations are those in the stratigraphic section of Nair et al. (1996); see above], a very thin N polarity zone consisting of two inferred flows between 445 and 490 m, an upper R interval from 490 to 930 m, and discordant directions above 930 m, which they suggested recorded a transition to the N epoch seen in the southwest. In the Narmada graben to the southwest of Mhow, Sreenivasa Rao et al. (1985) and Dhandapani & Subbarao (1992) reported an N–R–N sequence and concluded that the lower N lavas at the bases of their sections were older than the entire southwestern Deccan stratigraphy. If the thin lower N interval at Toranmal represents a true N polarity epoch, then the R sequence below it would be even older. Yet four of the six flows below 500 m that we analyzed isotopically are Poladpur-type (SH100, -99, -94 and -93).
It is difficult to evaluate the conclusions of Venkata Rao et al. (1996) because their data are presented only in the form of a brief summary. However, because the thin N interval is recorded by only two inferred flows and is not bounded on either side by a bole or other evidence of a protracted period between eruptions, we question whether this interval represents a true N polarity epoch. It may record a short-lived ‘cryptochron’ (such as those documented recently on the sea floor; e.g. Cande & Kent, 1992) within chron 29R, the polarity epoch to which most workers assign the main Deccan eruptions (e.g. Vandamme et al., 1991). Alternatively, it might be an artifact of sampling in physically disturbed zones; for example, in the entablature zone, where blocks of rock are often tilted or overturned, which may not be evident in areas of limited exposure like Toranmal. A third possibility, that the normally magnetized interval consists of a sill or sills, seems less likely as neither we nor Nair et al. (1996) observed any evidence of sills.
In view of these uncertainties, two very different interpretations of the geochemical data are possible. If a true, early N polarity epoch is recorded at Toranmal, a petrogenetic connection with the southwestern Poladpur basalts remains plausible, but Poladpur-type lavas in the Deccan would not constitute a volcanostratigraphic formation in the usual sense. Rather, Poladpur-type magmas must be distinctly time-transgressive, having been produced in one or more places starting before the eruption of the stratigraphically lowest exposed lavas of the Jawhar Fm in the southwest until at least the emplacement of the uppermost southwestern Poladpur lavas (Fig. 10, right). Poladpur-type flows may have been confined to a relatively small part of the northern Deccan during the lower R and N polarity intervals, for example, becoming widespread to the south and east only during the upper R polarity epoch (note that the SH102 and -104 flows at Toranmal, which provide a link with the northeastern high-206Pb/204Pb Poladpur-type sequences, are at elevations above the purported N interval). In this case, future work is likely to discover Poladpur-type lavas interfingered with those of the lower southwestern formations in places in the region between Toranmal–Shirpur and Igatpuri. A somewhat analogous situation may be found in the Paraná province of South America, where several spatially extensive lava types appear to span a nearly 10 my interval (e.g. Peate, 1997).
|
Alternatively, if the thin lower N interval at Toranmal does not represent an epoch of N polarity, then the entire Poladpur-type sequence was formed during an epoch of R polarity, consistent with the magnetostratigraphy of the southern and central Deccan. The Poladpur-type lavas at Toranmal are then most likely to represent a northern remnant of a vast Poladpur Fm (Fig. 10, left), which, however, is geochemically more diverse than indicated by the flows exposed in any one region. We note that the Poladpur-like lavas with high 206Pb/204Pb in the Chikaldara and Jabalpur areas are also in R polarity sequences, as are the Ambenali-like basalts above them [see Peng et al. (1998) and references therein]. Most of the lavas in the sections south of Mhow studied by Sreenivasa Rao et al. (1985) also have R polarity. Beyond confirmation and clarification of the reported N interval at Toranmal, systematic field, geochemical and paleomagnetic study of additional flow sections in the northern and central Deccan and of dikes in the western, northern and northeastern areas of the province should reveal how the various Poladpur-type lavas are related. The answer is vital for understanding the large-scale volcanic structure of the Deccan Traps, whether volcanism migrated with time (e.g. Beane et al., 1986; Devey & Lightfoot, 1986; Widdowson & Cox, 1996) and geochemical evolution on a province-wide, rather than just regional, basis.
|
ACKNOWLEDGEMENTS |
|---|
Keith Cox’s keen and abiding interest in Deccan geochemical stratigraphy was long a source of inspiration for our own efforts. We thank Peter Hooper for kindly making his ICP-MS data for the southwestern Deccan available, and Denys VonderHaar, Khal Spencer and Nancy Hulbirt for help with various aspects of the project. Ray Kent, Gautam Sen and Mike Widdowson provided constructive critical reviews. This work was supported by NSF Grant EAR-9418168 to J.M. and a Department of Science and Technology, Government of India, grant to D.C.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: +1-808-956-8705; e-mail: jmahoney{at}soest.hawaii.edu
Present address: Centro de Investigacion en Energia, Universidad Nacional Autonoma de Mexico, Apartado Postal 34, Temixco, Morelos 62580, Mexico.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD GEOLOGY, PETROGRAPHY AND...RESULTSDISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
Alexander, P. O. & Gibson, I. L. (1977). Rare earth abundances in Deccan Trap basalts. Lithos 10, 143–147.[Web of Science]
Allègre, C. J., Birck, J. L., Capmas, F. & Courtillot, V. (1999). Age of the Deccan Traps using 187Re–187Os systematics. Earth and Planetary Science Letters 170, 197–204.
Arndt, N. T. & Christensen, U. (1992). Role of lithospheric mantle in continental volcanism: thermal and geochemical constraints. Journal of Geophysical Research 97, 10967–10981.
Baksi, A. J. (1994). Geochronological studies on whole-rock basalts, Deccan Traps, India: evaluation of the timing of volcanism relative to the K–T boundary. Earth and Planetary Science Letters 121, 43–56.[Web of Science]
Baksi, A. J., Byerly, G. R., Chan, L.-H. & Farrar, E. (1994). Intracanyon flows in the Deccan province, India? Case history of the Rajahmundry Traps. Geology 22, 605–608.
Beane, J. E. (1988). Flow stratigraphy, chemical variation and petrogenesis of Deccan flood basalts from the Western Ghats, India. Ph.D. Dissertation, Washington State University, Pullman.
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. Bulletin of Volcanology 48, 61–83.
Bhattacharji, S., Chatterjee, N., Wampler, J. M., Nayak, P. N. & Deshmukh, S. S. (1996). Indian intraplate and continental margin rifting, lithospheric extension, and mantle upwelling in Deccan flood basalt volcanism near the K/T boundary: evidence from mafic dike swarms. Journal of Geology 104, 379–398.
Bilgrami, S. Z. (1999). A reconnaissance geological map of the eastern part of the Deccan Traps (Bidar–Nagpur). In: Subbarao, K. V. (ed.) Deccan Volcanic Province. Geological Society of India Memoir 43(1), 219–232.
Cande, S. C. & Kent, D. V. (1992). A geomagnetic polarity time scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research 97, 13917–13951.
Chandrasekharam, D., Mahoney, J. J., Sheth, H. C. & Duncan, R. A. (1999). Elemental and Nd–Sr–Pb geochemistry of flows and dikes from the Tapi rift, Deccan flood basalt province, India. In: Verma, S. P. (ed.) Rift-Related Volcanism: Geology, Geochemistry and Geophysics. Journal of Volcanology and Geothermal Research 93, 111–123.
Courtillot, V., Féraud, H., Maluski, D., Moreau, M. G. & Besse, J. (1988). Deccan flood basalts and the Cretaceous/Tertiary boundary. Nature 333, 843–846.
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. Journal of Petrology 26, 355–377.
Deshmukh, S. S., Sano, T., Fujii, T., Nair, K. K. K. & Yedekar, D. B. (1996). Chemical stratigraphy and geochemistry of basalt flows from the central and eastern parts of the Deccan volcanic province of India. In: Deshmukh, S. S. & Nair, K. K. K. (eds) Deccan Basalts. Gondwana Geological Magazine, Special Volume 2, 145–170.
Devey, C. W. (1986). Stratigraphy and geochemistry of the Deccan Trap lavas, Western India. Ph.D. Dissertation, University of Oxford.
Devey, C. W. & Cox, K. G. (1987). Relationship between crustal contamination and crystallisation in continental flood basalt magmas with special reference to the Deccan Traps of the Western Ghats, India. Earth and Planetary Science Letters 84, 59–68.
Devey, C. W. & Lightfoot, P. C. (1986). Volcanological and tectonic control of stratigraphy and structure in the western Deccan Traps. Bulletin of Volcanology 48, 195–207.
Dhandapani, R. & Subbarao, K. V. (1992). Magnetostratigraphy of the Deccan lavas, south of the Narmada River. Journal of the Geological Society of India 24, 63–79.
Duncan, R. A. & Pyle, D. G. (1988). Rapid eruption of the Deccan flood basalts at the Cretaceous/Tertiary boundary. Nature 333, 841–843.
Fodor, R. V., Frey, F. A., Bauer, G. R. & Clague, D. A. (1992). Ages, rare-earth element enrichment, and petrogenesis of tholeiitic and alkalic basalts from Kahoolawe Island, Hawaii. Contributions to Mineralogy and Petrology 110, 442–462.
Govindaraju, K. (1989). 1989 compilation of working values and sample descriptions for 272 geostandards. Geostandards Newsletter 13, 1–113.
Guha, S. (1995). Tectonic framework and evolution of the Tapi basin—an intracratonic half-graben in west–central India. Indian Minerals 49, 61–78.
Hooper, P. R. (1990). The timing of crustal extension and the eruption of continental flood basalts. Nature 349, 246–249.
Hooper, P. R. (1994). Sources of continental flood basalts: the lithospheric component. In: Subbarao, K. V. (ed.) Volcanism. New Delhi: Wiley Eastern, pp. 29–53.
Hooper, P. R. (1997). The Columbia River province: current status. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism. Geophysical Monograph, American Geophysical Union 100, 1–27.
Jain, J. C. & Neal, C. R. (1996). Report of the ICP-MS facility, 1993–1996. Notre Dame University Open File Report, 30 pp.
Kent, R. W., Saunders, A. D., Kempton, P. D. & Ghose, N. C. (1997). Rajmahal basalts, eastern India: mantle sources and melt distribution at a volcanic rifted margin. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism. Geophysical Monograph, American Geophysical Union 100, 145–182.
Khadri, S. F. R., Subbarao, K. V., Hooper, P. R. & Walsh, J. N. (1988). Stratigraphy of Thakurvadi Formation, western Deccan basalt province, India. In: Subbarao, K. V. (ed.) Deccan Flood Basalts. Geological Society of India Memoir 10, 281–304.
Knaack, C., Cornelius, S. & Hooper, P. R. (1991). Trace element analysis of rocks and minerals. Washington State University 1991 Open File Report, 36 pp.
Krishnamurthy, P. & Cox, K. G. (1977). Picrite basalts and related lavas from the Deccan Traps of western India. Contributions to Mineralogy and Petrology 62, 53–75.
Krishnamurthy, P. & Cox, K. G. (1980). A potassium-rich alkalic suite from the Deccan Traps, Rajpipla, India. Contributions to Mineralogy and Petrology 73, 179–189.
Krishnamurthy, P., Pande, K., Gopalan, K. & Macdougall, J. D. (1988). Upper mantle xenoliths in alkalic basalts related to Deccan Trap volcanism. In: Subbarao, K. V. (ed.) Deccan Flood Basalts. Geological Society of India Memoir 10, 53–69.
Lassiter, J. C. & DePaolo, D. J. (1997). Plume–lithosphere interaction in the generation of continental and oceanic flood basalts: chemical and isotopic constraints. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism. Geophysical Monograph, American Geophysical Union 100, 335–355.
Lightfoot, P. C. (1985). Isotope and trace element geochemistry of the south Deccan lavas, India. Ph.D. Dissertation, The Open University, Milton Keynes, UK.
Lightfoot, P. C. & Hawkesworth, C. J. (1988). Origin of Deccan Trap lavas: evidence from combined trace element and Sr-, Nd- and Pb-isotopic studies. Earth and Planetary Science Letters 91, 89–104.
Lightfoot, P. C., Hawkesworth, C. J., Devey, C. W., Rogers, N. W. & van Calsteren, P. W. C. (1990). Source and differentiation of Deccan Trap lavas: implications of geochemical and mineral chemical variations. Journal of Petrology 31, 1165–1200.
Mahoney, J. J. (1988). Deccan Traps. In: Macdougall, J. D. (ed.) Continental Flood Basalts. Dordrecht: Kluwer Academic, pp. 151–194.
Mahoney, J. J., Macdougall, J. D., Lugmair, G. W., Murali, A. V. & Gopalan, K. (1982). Origin of the Deccan Trap flows at Mahabaleshwar inferred from Nd and Sr isotopic and chemical evidence. Earth and Planetary Science Letters 60, 47–60.
Mahoney, J. J., Macdougall, J. D., Lugmair, G. W., Gopalan, K. & Krishnamurthy, P. (1985). Origin of contemporaneous tholeiitic and K-rich alkalic lavas: a case study from the northern Deccan Plateau, India. Earth and Planetary Science Letters 73, 39–53.
Mahoney, J. J., Nicollet, C. & Dupuy, C. (1991). Madagascar basalts: tracking oceanic and continental sources. Earth and Planetary Science Letters 104, 350–363.
Melluso, L., Beccaluva, L., Brotzu, R., Gregnanin, A., Gupta, A. K., Morbidelli, L. & Traversa, G. (1995). Constraints on the mantle sources of the Deccan Traps from the petrology and geochemistry of the basalts of Gujarat state (western India). Journal of Petrology 36, 1393–1432.
Melluso, L., Sethna, S. F., Morra, V., Khateeb, A. & Javeri, P. (1999). Petrology of the mafic dyke swarm of the Tapti River in the Nandurbar area (Deccan Volcanic Province). In: Subbarao, K. V. (ed.) Deccan Volcanic Province. Geological Society of India Memoir 43(2), 735–755.
Mitchell, C. H. & Widdowson, M. (1991). A geological map of the southern Deccan Traps, India and its structural implications. Journal of the Geological Society, London 148, 495–505.
Moore, C. L. (1997). Evaluation of regolith development and element mobility during weathering using the isocon technique. In: Eggleton, R. A. (ed.) The State of the Regolith. Geological Society of Australia, Special Publication 20, 141–147.
Nair, K. K. K., Chatterjee, A. K. & Sano, T. (1996). Stratigraphy and geochemistry of the Deccan basalts along Toranmal section, western Satpura region. In: Deshmukh, S. S. & Nair, K. K. K. (eds) Deccan Basalts. Gondwana Geological Magazine, Special Volume 2, 23–48.
Norrish, K. & Chapell, B. W. (1977). X-ray fluorescence spectrometry. In: Zussman, J. (ed.) Physical Methods in Determinative Mineralogy, 2nd edn. New York: Academic Press, pp. 201–272.
Peate, D. (1997). The Paraná–Etendeka province. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism. Geophysical Monograph, American Geophysical Union 100, 217–246.
Peng, Z. X. (1998). Geochemical studies of the northwestern, central and northeastern Deccan Traps, India: sources, petrogenesis, and relations to southwestern Deccan stratigraphy. Ph.D. Dissertation, University of Hawaii, Honolulu.
Peng, Z. X. & Mahoney, J. J. (1995). Drillhole lavas from the northwestern Deccan Traps, and the evolution of Réunion hotspot mantle. Earth and Planetary Science Letters 134, 169–185.[Web of Science]
Peng, Z. X., Mahoney, J. J., Hooper, P. R., Harris, C. & Beane, J. E. (1994). A role for lower continental crust in flood basalt genesis? Isotopic and incompatible element study of the lower six formations of the western Deccan Traps. Geochimica et Cosmochimica Acta 58, 267–288.[Web of Science]
Peng, Z. X., Mahoney, J. J., Hooper, P. R., Macdougall, J. D. & Krishnamurthy, P. (1998). Basalts of the northeastern Deccan Traps, India: isotopic and elemental geochemistry and relation to southwestern Deccan stratigraphy. Journal of Geophysical Research 103, 29843–29865.
Price, R. C., Gray, C. M., Wilson, R. E., Frey, F. A. & Taylor, S. R. (1991). The effects of weathering on rare earth elements, Y, and Ba in Tertiary basalts from southeastern Australia. Chemical Geology 93, 245–265.
Rudnick, R. & Fountain, D. M. (1995). Nature and composition of the continental crust: a lower crustal perspective. Reviews of Geophysics 33, 267–309.[Web of Science]
Sen, G. & Cohen, T. H. (1994). Deccan intrusion, crustal extension, doming and the size of the Deccan–Réunion plume head. In: Subbarao, K. V. (ed.) Volcanism. New Delhi: Wiley Eastern, pp. 201–216.
Sheth, H. C. (1998). Geochemistry, petrogenesis, stratigraphy and structure of Deccan flood basalts of the western Satpura–Tapi region, India. Ph.D. Dissertation, Indian Institute of Technology, Bombay.
Sheth, H. C. & Chandrasekharam, D. (1997). Early alkaline magmatism in the Deccan Traps: implications for plume incubation and lithospheric rifting. Physics of the Earth and Planetary Interiors 104, 371–376.[Web of Science]
Sheth, H. C., Duncan, R. A., Chandrasekharam, D. & Mahoney, J. J. (1997). Deccan Trap dioritic gabbros from the western Satpura–Tapi region. Current Science 72, 755–757.[Web of Science]
Sinigoi, S., Quick, J. E., Mayer, A. & Budahn, J. (1996). Influence of stretching and density contrasts on the chemical evolution of continental magmas: an example from the Ivrea–Verbano Zone. Contributions to Mineralogy and Petrology 123, 238–250.
Sreenivasa Rao, M., Ramasubba Reddy, N., Subbarao, K. V., Prasad, C. V. R. K. & Radhakrishnamurty, C. (1985). Chemical and magnetic stratigraphy of parts of Narmada region, Deccan basalt province. Journal of the Geological Society of India 26, 617–639.[Web of Science]
Subbarao, K. V., Bodas, M. S., Hooper, P. R. & Walsh, J. N. (1988). Petrogenesis of Jawhar and Igatpuri formations, western Deccan basalt province, India. In: Subbarao, K. V. (ed.) Deccan Flood Basalts. Geological Society of India Memoir 10, 253–280.
Subbarao, K. V., Chandrasekharam, D., Navaneethakrishnan, P. & Hooper, P. R. (1994). Stratigraphy and structure of parts of the central Deccan basalt province: eruptive models. In: Subbarao, K. V. (ed.) Volcanism. New Delhi: Wiley Eastern, pp. 321–332.
Sun, S.-s. & McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publication 42, 313–345.
Todt, W., Cliff, R. A., Hanser, A. & Hofmann, A. W. (1996). Evaluation of a 202Pb + 205Pb double spike for high-precision lead isotopic analyses. In: Basu, A. & Hart, S. (eds) Earth Processes: Reading the Isotopic Code. Geophysical Monograph, American Geophysical Union 95, 429–437.
Tolan, T. L., Reidel, S. P., Beeson, M. H., Anderson, J. L., Fecht, K. R. & Swanson, D. A. (1989). Revisions to the estimates of the areal extent and volume of the Columbia River Basalt Group. In: Reidel, S. P. & Hooper, P. R. (eds) Volcanism and Tectonism in the Columbia River Flood-Basalt Province. Geological Society of America, Special Paper 239, 1–20.
Turner, S. & Hawkesworth, C. J. (1995). The nature of the sub-continental mantle: constraints from the major-element composition of continental flood basalts. In: McDonough, W. F., Arndt, N. & Shirey, S. (eds) Chemical Evolution of the Mantle. Chemical Geology 120, 295–314.
Vandamme, D., Courtillot, V., Besse, J. & Montigny, R. (1991). Paleomagnetism and age determinations of the Deccan Traps (India): results of a Nagpur–Bombay traverse and review of earlier work. Reviews of Geophysics 29, 159–190.[Web of Science]
Venkata Rao, K., Nair, K. K. K. & Padhi, R. N. (1996). Magnetostratigraphy of the Deccan basalts of the western Satpura region. In: Deshmukh, S. S. & Nair, K. K. K. (eds) Deccan Basalts. Gondwana Geological Magazine, Special Volume 2, 431–438.
Venkatesan, T. R., Pande, K. & Gopalan, G. (1993). Did Deccan volcanism pre-date the Cretaceous/Tertiary transition? Earth and Planetary Science Letters 119, 181–189.[Web of Science]
West, W. D. (1958). The petrography and petrogenesis of forty-eight flows of Deccan Traps penetrated by borings in western India. Transactions of the Indian National Institute of Science 4, 1–56.
Widdowson, M. & Cox, K. G. (1996). Uplift and erosional history of the Deccan Traps, India: evidence from laterites and drainage pattern of the Western Ghats and Konkan coast. Earth and Planetary Science Letters 137, 57–69.
Wilkins, A., Subbarao, K. V., Ingram, G. & Walsh, J. N. (1994). Weathering regimes within the Deccan basalts. In: Subbarao, K. V. (ed.) Volcanism. New Delhi: Wiley Eastern, pp. 217–231.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. E. JAY and M. WIDDOWSON Stratigraphy, structure and volcanology of the SE Deccan continental flood basalt province: implications for eruptive extent and volumes Journal of the Geological Society, January 1, 2008; 165(1): 177 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Kumar, R. Menon, and G. K. Reddy Crustal geotherm in southern Deccan basalt province, India: The Moho is as cold as adjoining cratons Geological Society of America Special Papers, January 1, 2007; 430(0): 275 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.C. Sheth Plume-related regional prevolcanic uplift in the Deccan Traps: Absence of evidence, evidence of absence Geological Society of America Special Papers, January 1, 2007; 430(0): 785 - 813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. SANO, T. FUJII, S. S. DESHMUKH, T. FUKUOKA, and S. ARAMAKI Differentiation Processes of Deccan Trap Basalts: Contribution from Geochemistry and Experimental Petrology J. Petrology, December 1, 2001; 42(12): 2175 - 2195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. WIDDOWSON, M. S. PRINGLE, and O. A. FERNANDEZ A Post K-T Boundary (Early Palaeocene) Age for Deccan-type Feeder Dykes, Goa, India J. Petrology, July 1, 2000; 41(7): 1177 - 1194. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||