Re-Os isotope systematics of sediments of the Brahmaputra River system

Geochimica et Cosmochimica Acta, Vol. 67, No. 21, pp. 4101– 4111, 2003 Copyright © 2003 Elsevier Ltd Printed in the USA. All rights reserved 0016-7037/03 $30.00 ⫹ .00 Pergamon doi:10.1016/S0016-7037(00)00201-1 Re-Os isotope systematics of sediments of the Brahmaputra River system SUNIL K. SINGH,* LAURIE REISBERG, and CHRISTIAN FRANCE-LANORD Centre de Recherches Pétrographiques et Géochimiques (CRPG/CNRS), 15, rue Notre Dame des Pauvres, B.P. 20, 54501 Vandœuvre-lès-Nancy Cedex, France (Received July 15, 2002; accepted in revised form March 11, 2003) Abstract—Re-Os analyses were performed on suspended loads and coarser grained bank sediments of the Brahmaputra River system. Re and Os concentrations of these sediments vary from 7 to 1154 ppt and from 3 to 173 ppt, respectively. 187Os/188Os ratios range from 0.178 to 6.8, and thus vary from nearly mantle to very radiogenic crustal values. Nevertheless, most of the sediments have 187Os/188Os ratios less than 1.5, and nearly all of the samples of the Brahmaputra main channel have ratios less than 1.2. Thus, as previously suggested, the Brahmaputra is much less radiogenic than the Ganga. The Siang River, the northern extension of the Brahmaputra, is quite radiogenic in Os despite receiving sediments from the Tsangpo River, which flows along a suture zone with ultramafic outcrops. The Brahmaputra main channel has a fairly constant 187Os/188Os ratio even though its tributaries contribute sediments with very heterogeneous Os isotopic compositions. These data, along with the corresponding Nd isotopic compositions, suggest that about 60 –90% of the sediment in the Brahmaputra system is derived from Himalayan formations (Higher Himalaya and Lesser Himalaya) whereas 10 – 40% comes from ophiolite-bearing sequences, perhaps eastern equivalents of those of the Transhimalayan Plutonic Belt. Os data also confirm previously published Sr and Nd results, indicating that about half of the sediments delivered to the Brahmaputra are supplied by the Siang River, while the Himalayan and the eastern tributaries account for 40 and 10%, respectively. The lower 187Os/188Os of the Brahmaputra River compared to that of the Ganga is due to two factors. One is the more limited presence of the Lesser Himalaya and hence the lower black shale content of the eastern Himalaya. The other is the non-radiogenic Os supplied by the eastern and southern tributaries, reflecting the presence of mantle-derived lithologies in this region. Despite the lower sediment supply from these tributaries, they contribute greatly to the Os budget of the Brahmaputra River. This study indicates that the Brahmaputra River has little effect on the present-day seawater Os budget. However, reconsideration of this budget suggests that the Ganga, which provides the most radiogenic Os of major rivers studied to date, may have significant impact on the marine Os isotopic composition. The Indo-Asian collision cannot be excluded as an important cause of the increase in the marine 187Os/188Os over the past 16 million years until the contributions of all of the rivers draining the Himalayan Tibetan Plateau are known. Copyright © 2003 Elsevier Ltd this issue, more data from rivers draining the HTP region are needed. There now exists a significant amount of Os data for dissolved and particulate matter in the Ganga River (Pegram et al., 1994; Levasseur et al., 1999; Sharma et al., 1999; PiersonWickmann et al., 2000). However, the information available for the Brahmaputra is much more limited. The few data that exist suggest that while both rivers drain Himalayan lithologies, the Os composition of the dissolved (Sharma et al., 1999) and the particulate (Pierson-Wickmann et al., 2000) load of the Brahmaputra is much less radiogenic than that of the Ganga. This has been attributed to the presence of ophiolites in the IndusTsangpo suture zone, which is drained by the Tsangpo River, the upstream continuation of the Brahmaputra, though this assertion has not been substantiated. Such mantle rocks have unradiogenic Os isotopic ratios, but high Os concentrations, and thus are expected to contribute sediments with low 187Os/ 188 Os ratios. This investigation is one of a growing number of studies aimed at understanding the geochemical cycle of Os at the surface of the Earth, and the behavior of Re and Os during crustal weathering and fluvial transport. Such information is necessary before the Os isotopic record of seawater can be used to track possible variations of climate and tectonics in the past. 1. INTRODUCTION Chemical weathering in the Himalaya has influenced the chemical and isotopic budget of the ocean and atmosphere since the onset of the Himalayan orogeny. Following the analogy of marine Sr isotopic evolution, it has been suggested that Himalayan weathering has provided large amounts of radiogenic Os to seawater (Pegram et al., 1992; Peucker-Ehrenbrink et al., 1995), thus explaining the marked increased in the marine 187Os/188Os ratio over the past 16 Ma. The presence of old black shales in the Himalaya, rich in Re and hence in radiogenic Os (Singh et al., 1999; Pierson-Wickmann et al., 2000), further strengthens the above hypothesis. On the other hand, Levasseur et al. (1999) have suggested that the Himalayan Os flux, though quite radiogenic, is not large enough to have much affect on the oceanic composition. Thus, while it is generally accepted that weathering of the Himalayan-Tibetan Plateau (HTP) has largely controlled the isotopic evolution of Sr in late Cenozoic seawater (Palmer and Edmond, 1989; Krishnaswami et al., 1992; Galy et al., 1999), its impact on the marine Os isotopic record remains controversial. To resolve * Author to whom correspondence should be addressed, at Physical Research Laboratory, Ahmedabad, India (sunil@prl.ernet.in). 4101 4102 S. K. Singh, L. Reisberg, and C. France-Lanord Fig. 1. Map showing sample locations and 187Os/188Os ratios. Suspended load samples are denoted SL; all others are bank sediments. The Tsangpo River drains through the Indus Tsangpo Suture Zone in Tibet, takes a 180° turn around Namche Barwa and drains through various Himalayan formations between Namche Barwa and Pasighat, where it is known as the Siang River. Afterwards it merges with the Dibang and the Lohit (two eastern tributaries) and becomes the Brahmaputra until the confluence with the Ganga in Bangladesh. It is joined by six major tributaries from the Himalaya and three from its southern drainage system. In the present contribution, we report results from a Re-Os isotopic study of sediments from the Brahmaputra River and its major tributaries. Both suspended particles and coarser grained bank sediments were investigated, allowing us to examine the possible effects of granulometric sorting. Similarly, sediments collected during the monsoon and non-monsoon periods were analyzed, permitting possible seasonal variations to be detected. The results of this study allow us to establish a budget of the major sources controlling the Os isotopic composition of Brahmaputra River sediments, and to compare this budget with that obtained on the basis of Sr and Nd isotopes (Singh and France-Lanord, 2002). We also reevaluate the possible contribution of Himalayan weathering to the Os isotopic evolution of seawater. We conclude that while the Brahmaputra River probably had little influence on this evolution, a possible large role for other rivers draining the HTP cannot be excluded on the basis of current data. 2. GEOLOGIC SETTING AND SAMPLING STRATEGY The Tsangpo River, the Tibetan part of the Brahmaputra, originates in Kailash mountain and then flows along the IndusTsangpo Suture zone (Fig. 1). The main lithologies in the drainage basin are the Paleozoic sedimentary sequences of Tibet and the Trans Himalayan batholiths, which consist of ultramafic to felsic rocks, ranging from peridotite to granite. At the Eastern Syntaxis, near Namche Barwa, the river takes a 180° turn as it flows through highly metamorphic rocks of the Higher Himalayan Crystallines (HHC) (Burg et al., 1998). Between Namche Barwa and Pasighat, where it is referred to as the Siang, the river passes through the Abor volcanics and the Miri limestone, which are interbedded with shales and other sedimentary rocks including black shales of the Lesser Himalaya and the Gondwana Group (Thakur, 1986). After Pasighat the river flows through the alluvium of the Assam Plain, which was deposited on the Indian craton in response to the Himalayan induced subsidence of the region between the Himalaya and Naga Patkoi ranges (Kumar, 1997). The Northern tributaries drain first the various crystalline and metasedimentary lithologies of the Higher Himalaya, then the sedimentary rocks, including carbonates, shales, slates and quartzites, and some crystallines of the Lesser Himalaya and finally the Tertiary sediments of the Siwaliks. The Eastern tributaries, the Dibang and the Lohit, drain the Mishmi hills. The geology of the Mishmi hills is poorly understood. Earlier workers considered Re-Os systematics in the Brahmaputra River system this region to be the continuation of the eastern extension of the Ladakh ranges or of the Burmese mountains (Kumar, 1997). Geologic and tectonic information (Sharma, 1991; Kumar, 1997) now suggest that it represents the eastern continuation of the Transhimalayan Plutonic Belt (TPB). It seems plausible that the classical Himalayan Formations terminate against the Mishmi hills along the Tidding Suture, thought to be the continuation of the Indus-Tsangpo Suture. In the Mishmi hills both the Dibang and the Lohit flow through tholeiitic metavolcanics of island arc affinity and calc-alkaline diorite-granodiorite complexes similar to the Dras volcanics and the Kargil igneous complex of the Ladakh magmatic complex (Sharma, 1991). The rivers drain through the graphitic schists, grey slates and marble bands of the Yang Sang Chu formation, altered metavolcanics with chlorite and phyllite and limestone of the Tidding formation and diorite and tonalite of the Mishmi formations. In the Lohit valley, hornblende schists, marble with diorite and diorite-granodiorite intruded by mafic rocks and schist are also exposed (Kumar, 1997). The main southern tributaries of the Brahmaputra are the Nao Dihing (a tributary of the Lohit River), the Burhi Dihing, the Dhansiri and the Kopili Rivers, which flow through the Naga-Patkoi Ranges, including the Shillong plateau and Mikir hill. These ranges, which are composed of Tertiary sequences, are similar to those of the Assam basin resting on Precambrian basement (Kumar, 1997). The rivers of this region flow through the sedimentary rocks of the Arakan-Assam basin which resulted from the collision of the Indian plate with the Tibetan and the Central Burmese plate. The western part of the IndoBurman ranges, along the Arakan coast, consists of Cretaceous and Oligocene sedimentary shales hosting volcanic dykes and ophiolites. Some of the sediments in this region were deposited in the Paleocene to Oligocene and were derived from the Inner Volcanic Arc of Burma (Colin et al., 1999). Naga ophiolites are present in this section (Kumar, 1997). Grains of iridosmine along with native gold and platinum have been reported in the sand of the Nao Dihing River (Mallet, 1882). Both suspended load and bank sediment samples were collected during two field campaigns, the first in October 1999 and the second in July 2000, representing, respectively, the postmonsoon and monsoon seasons. Samples were collected along the course of the main Brahmaputra River from Pasighat in Arunachal Pradesh, India to Dhubri near Bangladesh, as well as from most of the important tributaries joining from north, east and south in the Assam Plain (Fig. 1). For suspended load samples, ⬃5 L of water were collected from the middle of the stream. During 5 to 7 d after collection, the suspended sediment was allowed to settle to the bottom of the plastic carboys. After this time, most of the water above the sediment column was siphoned out and the remaining sediment was stored in plastic bags along with some water. Bank sediments were collected mostly from the riverbanks but in some cases were collected from sand bars in the middle of the river. The sediments of the Brahmaputra River system (BRS) are analyzed and reported earlier for their major element compositions, Rb, Sr, Sm and Nd concentrations and Sr and Nd isotope compositions (Singh and France-Lanord, 2002). These compositions were used to trace sediment provenance of the BRS. 87Sr/86Sr of these sediments varies from 0.7053 to 0.8250 4103 where as the variations in ␧Nd is from ⫺20.5 to ⫺6.9. Large variations in Sr and Nd isotope compositions of the Siang, Eastern and the Northern tributaries enable us to quantify the sources of the sediments from the various drainage and from the various lithologies. Based on these data, it has been estimated that half of the sediments to the BRS is derived from the Eastern Syntaxis region, whereas the contributions from Himalayan and Eastern drainage are 40 and 10%, respectively. In terms of lithology, the Himalayan rocks are representing ⬃70% of the total sediment flux and remaining portion is being supplied by Transhimalayan Plutonic Belt (Singh and FranceLanord, 2002). 3. MATERIALS AND METHODS About 1 to 3 g of sample powder along with 190Os and 185Re spikes were weighed into 60 mL PFA bombs and heated with HF and HBr in an oven at 145°C for ⬃24 h. After cooling, the Os was oxidized with CrO3 in HNO3 and extracted into Br liquid following the technique of Birck et al. (1997). The details of the implementation of this technique at CRPG can be found in Pierson-Wickmann et al. (2000). The extracted Os was purified by micro-distillation (Roy Barman, 1993; Birck et al., 1997). Re was separated from the remaining liquid after Os extraction using isoamylol (Birck et al., 1997). Os and Re concentrations were determined by isotope dilution. Os isotopic compositions were measured by NTIMS (Creaser et al., 1991; Völkening et al., 1991) using a Finnigan MAT 262 mass spectrometer. Re isotopic compositions used for concentration calculation were measured with an Elan 6000 ICP-MS. In the latter part of this study Re compositions were measured with an Isoprobe MC-HEX-ICP-MS. The total procedural blank for Os was quite low (from 0.3 to 1.0) pg. As the isotopic composition of the blank was not well constrained, blank corrections were not performed. Considering an average Os blank of 0.5 pg and 187Os/188Os of 0.15, the blank contribution to the concentration and 187Os/188Os did not exceed ⬃1% for most of the samples (⬃5% in the worst case, sample BR 76 with 3.1 ppt Os). Thus the lack of blank correction does not effect the conclusions of this study. Re blanks were moderately high and mostly derived from the CrO3. Using two different batches of CrO3 the total Re blanks were 35 and 25 pg. The Re concentrations of the samples are corrected for this blank accordingly. During the course of the measurement, 187Os/188Os of the in-house standard was 0.17383⫾0.00064 (2␴), consistent with earlier measurements (Pierson-Wickmann et al., 2000). This value is also in agreement with previous measurements of this standard performed at LDEO and at WHOI. 4. RESULTS The Re and Os concentrations and 187Os/188Os of the sediments are given in Table 1. Os isotopic ratios are indicated with the corresponding sample numbers on the map in Figure 1. Os and Re concentrations and Os isotopic compositions vary widely among the sediments of the Brahmaputra River system. Os and Re concentrations range from 2 to 193 ppt and from 7 to 1154, respectively. 187Os/188Os ratios vary from 0.18 to 2.9, except for one sediment from a small stream with a ratio of 6.8 (Table 1). The samples richest in Os have the least radiogenic compositions, as seen by the rough correlation between 187Os/ 188 Os and 1/188Os (Fig. 2). Os concentration generally increases with that of Re, except for the few eastern and southern tributaries that have sediments with exceptionally high Os contents. Re and Os concentrations are controlled mostly by the lithology mineralogy of the sediment source regions, but other factors play a role as well. In particular, both Re and Os concentrations of the suspended load 4104 S. K. Singh, L. Reisberg, and C. France-Lanord Table 1. Re and Os concentrations and Sample Brahmaputra Main Channel BR 19 BR 29 BR 65SL BR 66 BR 3 BR 4 BR 6 BR 9 BR 52SL BR 53SL BR-55SL BR 56 BR 73SL BR 74 BGP 14c BGP 82c Eastern Tributaries BR 15 BR 17 Tsangpoc Nimuc Siang BR 59SL BR 60 Himalayan Tributaries BR 21 BR 61SL BR 62 BR 25 BR 58 BR 27 BR 64 BR 35 BR 70 BR 33 BR 71SL BR 72 BR 76 BGP 11c Southern Tributaries BR 31 BR 68 BR 11 BR 13 BR 13R BR78 Type 187 Os/188Os of sediments of the Brahmaputra River system. River Bank sed. Bank sed. Susp Load Bank sed. Susp load Susp load Susp load Bank sed. Susp load Susp load Susp load Bank sed. Susp load Bank sed. Bank sed. Bank sed. Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Brahmaputra Bank Bank Bank Bank 188 Os (fm/g) Re (ppt) 187 Os/188Os ␧Ndb 10.1 13.3 42.6 17.3 55.8 29.1 56.7 14.6 38 21.7 84.5 18.7 37.7 17 12 18 6.0 8.3 26.5 10.9 36.0 18.5 36.3 9.3 24.8 13.4 55.0 11.7 23.2 11.2 7.3 11.4 69.8 52.4 302 57 274.7 142.8 269.8 59.9 359 419 365 75 210 32 161 803 1.454 0.969 1.07 0.926 0.766 0.899 0.834 0.835 0.678 1.091 0.68 1.007 1.161 0.64 1.596 0.815 ⫺12.6 ⫺14.04 ⫺13.96 ⫺13.63 ⫺13.4 ⫺13.2 ⫺12.5 ⫺13.4 ⫺13.29 — ⫺12.79 ⫺13.44 ⫺14.01 ⫺14.36 ⫺16.9 ⫺13.6 Dibang Lohit Near Lhasa PM PM 92 40 44.15 2.01 62.9 24.5 29.4 1.2 294.9 1039.5 230 117 0.286 1.215 0.501 1.388 ⫺6.9 ⫺12.4 ⫺10 Susp load Bank sed. Siang Siang M M 32.1 17.5 19.1 9.5 331 397 1.443 2.294 ⫺14.63 ⫺12.03 Bank sed. Susp load Bank sed. Bank sed. Bank sed. Bank sed. Bank sed. Bank sed. Bank sed. Bank sed. Susp load Bank sed. Bank sed. Bank sed. Subansiri bank Subansiri Subansiri bank Ranga Nadi bank Ranga Nadi bank Jia Bhareli bank Jia Bhareli bank Puthimari bank Puthimari bank Manas bank Manas Manas bank Tipkai Tista PM M M PM M PM M PM M PM M M M 15.8 36.2 10.8 15.2 10.6 10.7 14.4 18.8 26.7 14 26.4 9.9 23.0 6.4 9.6 6.6 6.5 8.5 10.8 17.2 8.3 16.2 3.1 8 1.8 4.3 82.7 164 55 141.8 98 68.1 106 179.5 141 62.5 167 44 7 1154 1.048 0.9 1.431 0.924 1.07 1.377 1.486 1.8 0.774 1.528 1.151 1.21 1.571 2.859 ⫺15.6 ⫺12.74 ⫺14.09 ⫺12.76 ⫺12.27 ⫺16.31 ⫺16.36 ⫺19.9 ⫺17.55 ⫺15.95 ⫺17.22 ⫺16.35 ⫺20.2 ⫺21.20 Bank Bank Bank Bank Bank Bank Kopili bank Kopili bank Dhansiri Buri Dihing Buri Dihing Basistha Dhara at Guwahati PM M PM PM PM M 21.5 9.8 49 173 16.84 8.1 13.7 6.1 32.7 120.2 11.4 3.0 76.4 37 190.5 35.4 — 25 0.845 1.003 0.471 0.178 0.380 6.8 ⫺12.7 ⫺20.49 ⫺8.4 ⫺18.7 sed. sed. sed. sed. sed. sed. Dibrugarh Tezpur Tezpur Tezpur Guwahati Guwahati Guwahati Guwahati Guwahati Guwahati Guwahati Guwahati Dhubri Dhubri Chilmari Chilmari Os (ppt) PM PM M M PM PM PM PM M M M M M M M PrM sed. sed. sed. sed. at at at at at at at at at at at at at at at at Season ⫺12.63 a M ⫽ monsoon, PM ⫽ postmonsoon, PrM ⫽ premonsoon, R ⫽ replicate. From Singh and France-Lanord (2002). c From Pierson-Wickmann et al. (2000). b samples are higher than those of the bank sediments (Fig. 3). This systematic difference cannot be attributed to a source difference as both types of sediments collected in a given location have similar Os isotopic compositions. Part of this concentration difference can be explained by dilution with quartz which is more abundant in bank sediments. As shown in Figure 4, among Brahmaputra main channel sediments a negative correlation exists between Os and SiO2 content. However, this correlation does not intersect the x-axis at 100% SiO2, so quartz dilution does not completely explain the lower bank sediment Os concentrations. Dilution by other Os-poor phases, perhaps feldspars, may also play a role. The suspended sedi- ments are poorer in quartz and feldspar, but conversely, richer in micas and other phyllosilicates than the bank sediments. This may partly explain the higher Re and Os contents of the suspended load, since micas are enriched in Os and Re relative to most other phases, as shown by a previous study of river sediments from Central Nepal (Pierson-Wickmann et al., 2002b). Another factor may be that both Re and Os have an affinity for organic carbon, which is more abundant in the suspended load (⬃0.5%) than in the bank sediments (⬃0.2%) (Singh and France-Lanord, unpublished data). Identification of the main Os and Re bearing phases in crustal rocks and sediments clearly requires further investigation. Re-Os systematics in the Brahmaputra River system Fig. 2. Mixing diagram showing 187Os/188Os versus inverse of 188Os. In general 187Os/188Os increases with decreasing Os concentration. As both suspended load and bank sediments are pooled together, the scatter is increased because at a given site Os concentrations for suspended loads are higher than those of bank sediments. Nevertheless, their isotopic ratios are similar because they are derived from the same source regions. 4.1. Brahmaputra River In the northeastern portion of the main channel of the Brahmaputra River, Os isotopic compositions become progressively less radiogenic in the downstream direction. The 187Os/188Os ratios of the samples collected furthest upstream, at Pasighat in the Siang River (the Brahmaputra in the Arunachal Pradesh region), are 2.3 and 1.44 for bank sediment and suspended load, respectively. These rather radiogenic compositions are noteworthy, as the Tsangpo River is the dominant water source of the Brahmaputra upstream of Pasighat. The Tsangpo flows 4105 Fig. 4. Os concentration vs. SiO2 wt. percent for Brahmaputra main channel sediments. A negative correlation is observed, but this trend does not intersect the x-axis at 100% SiO2, indicating that dilution by quartz alone is not the only factor controlling the difference in Os concentration between suspended and bank sediment samples. along the Indus-Tsangpo suture in Tibet (Fig. 1) and has a bed load with a low 187Os/188Os ratio (⬃0.5; Pierson-Wickmann et al., 2000) reflecting the presence of ophiolites in its drainage basin. Downstream from the Siang, the main Brahmaputra channel has 187Os/188Os values between 0.64 and 1.57 while Os concentrations vary from 10.1 to 84.5 ppt. At Dibrugarh 187 Os/188Os is ⬃1.45. This ratio decreases to ⬃1.0 at Tezpur. Further downstream, at Guwahati, 187Os/188Os ranges from 0.64 to 1.0 among the 8 samples analyzed. No further significant decrease in 187Os/188Os is observed beyond this point. At Dhubri the Os isotopic ratios are 0.65 and 1.16 for bank sediment and suspended load, respectively. In the Bangladesh, further downstream and near to the river’s mouth two values of 0.8 to 1.6 have been measured (Pierson-Wickmann et al., 2000). Throughout the length of the river, there is no evidence that seasonal variations have a systematic effect on Os isotopic compositions. Monsoon and postmonsoon samples generally have the same isotopic ratios. The sole exception occurs in Bangladesh, where the ratio changed from 0.8 in the premonsoon to 1.6 in the monsoon. The singularity of this monsoon sample (BGP14) is also observed for Sr and Nd (Galy and France-Lanord, 2001). Given the constancy between seasons of samples collected elsewhere along the Brahmaputra, the reason for this difference is not obvious, but it may result from a sudden input from Himalayan tributaries during flash flooding. 4.2. Himalayan Tributaries Fig. 3. A plot of Re versus Os concentrations in the sediments. Re and Os concentrations in suspended loads are higher than those of bank sediments, partly due to quartz dilution in the bank sediments and partly probably because of their association with organic carbon or phyllosilicates which are more abundant in the suspended load. In the Himalayan tributaries of the Brahmaputra, Os concentrations range from 3.1 to 36.2 ppt. With the exception of one sample (the Tista, with a ratio of 2.86), 187Os/188Os ratios vary from 0.77 to 1.57. These isotopic compositions are similar to the ratios of those central Himalayan rivers (Pierson-Wickmann et al., 2000) that drain mainly the Tibetan Sedimentary Series (TSS) and the High Himalaya (HH). In contrast, they are 4106 S. K. Singh, L. Reisberg, and C. France-Lanord lower than the Os isotopic ratios of central Himalayan rivers that include a large Lesser Himalaya (LH) area in their drainage basins. This suggests that in the eastern section the LH contribution, which is highly radiogenic in Os due to the presence of ancient black shales, is less important than in the central and western Himalaya. This interpretation is consistent with conclusions based on Sr and Nd isotopic data (Singh and FranceLanord, 2002) and with the geology of this region (Thakur, 1986; Kumar, 1997), which all indicate that the areal extent of the LH formation decreases towards the east. This geographic trend also explains why the Tista River, the westernmost of the Himalayan tributaries to the Brahmaputra, has a much more radiogenic Os composition and a higher Re concentration than the other samples. The Tista is more radiogenic than the other Brahmaputra tributaries because its drainage basin contains a larger proportion of the ancient black shale bearing LH formation. 4.3. Eastern Tributaries Two eastern tributaries of the Brahmaputra were sampled. The Lohit bank sediment has a rather high Re concentration (1040 ppt) and an Os concentration (40 ppt) and 187Os/188Os ratio (1.2) very similar to that of average upper continental crust (Esser and Turekian, 1993; Peucker-Ehrenbrink and Jahn, 2001). In contrast, the Dibang sediment has a 187Os/188Os ratio of only 0.286 with 92 ppt Os content. This non-radiogenic Os signature is derived from mantle lithologies present in the drainage basin of the Dibang, and is consistent with the low Sr and high Nd isotopic composition of this river (Singh and France-Lanord, 2002). As mentioned in the geologic description, the bedrock of this drainage basin includes lithologies similar to those of the Transhimalayan region. Of particular importance for Os isotopes is the likely presence of the Tidding Suture, which may contain ophiolites and thus ultramafic rocks that are rich in Os of non-radiogenic composition. 4.4. Southern Tributaries Re and Os concentrations of sediments of the southern tributaries vary from 25 to 191 ppt and from 8.1 to 173 ppt, respectively. With the exception of one highly radiogenic sample, 187Os/188Os varies from 0.18 to 1.0. The least radiogenic sample is the bank sediment of the Burhi Dihing, which has a 187 Os/188Os ratio of 0.178 and an Os concentration of 173 ppt. A duplicate analysis of this sample yielded a 187Os/188Os ratio of 0.38 with an Os concentration of 16.8 ppt, suggesting that the Os is distributed quite heterogeneously. Iridosmine grains, along with gold grains, have been found in the sand of the Nao Dihing River (Mallet, 1882). In its upper reaches, the Burhi Dihing is located very close to the Nao Dihing and crosses a similar lithology, and thus it is possible that the high Os concentrations of the Burhi Dihing samples reflect the presence of rare iridosmine grains (a single grain of iridosmine 5 ␮m in diameter could account for the entire Os budget of the more Os-rich sample). Regardless of the exact identity of the Os-rich phase, it is quite likely that the high Os concentration and the low Os isotope composition of the Burhi Dihing is due to the presence of mantle derived minerals in the bank sediment. This may also be the case for the Dhansiri River. The Basistha Dhara is a very small stream south-east of the Guwahati that flows on the Indian basement. Its water is slightly acidic. The bank sediment of this stream has a low Os concentration (8 ppt) and an extremely radiogenic Os signature (187Os/188Os ⫽ 6.8). As no other samples in this study have Os ratios approaching this value, the highly radiogenic Os of this sediment appears to be highly localized. 5. DISCUSSION 5.1. Sources of Osmium to the Brahmaputra As shown above, the Os isotope compositions of the sediments of the Brahmaputra River tributaries are quite heterogeneous. Nevertheless, for most of the length of the Brahmaputra main channel the Os composition of the sediments is fairly uniform and somewhat less radiogenic (187Os/188Os ⫽ 0.93 ⫾0.24, simple average; 187Os/188Os ⫽ 0.87 weighted average, including all samples from Tezpur downstream) than that estimated for the average upper continental crust (187Os/188Os ⫽1.3, Esser and Turekian, 1993; 187Os/188Os ⫽1.1, PeuckerEhrenbrink and Jahn, 2001). To understand the origin of the Brahmaputra Os ratio, it is first necessary to explain the Os compositions of the main sources of sediments to the Brahmaputra. These comprise the Siang River, the Himalayan tributaries, and the eastern and southern tributaries. In the Siang River, the bank sediment sample has a surprisingly high 187Os/ 188 Os ratio (⬃2.3). The Siang is the southern continuation of the Tsangpo River. Tsangpo River sediment has a 187Os/188Os of only ⬃0.5 (Pierson-Wickmann et al., 2000), reflecting the fact that this river flows through the ophiolite of the IndusTsangpo suture zone in Tibet. For this reason, earlier workers (Sharma et al., 1999; Pierson-Wickmann et al., 2000) have thought that the Siang should be unradiogenic as well. To explain the Siang sediments with 187Os/188Os ratios as high as 2.3, the presence of a lithology with radiogenic Os, such as black shales, is required. Further evidence for the presence of black shales is provided by the high P2O5 content of the Siang bank sediment (Singh and France-Lanord, 2002). There are reports of black shales in the Siwalik and Gondwana group formations in the drainage basins of the Siang upstream of Pasighat. In particular, Thakur (1986) noted the existence of carbonaceous shale with phosphatic limestone in the Gondwana group in this area. A second factor explaining the high 187 Os/188Os ratio of the Siang River sediment is the presence of a knickpoint (Zeitler et al., 2001) before Namche-Barwa (Fig. 1) that may block part of the sediment coming downstream. Thus much of the unradiogenic Tsango sediment may never reach the Siang. The limited influence of the Tsangpo sediment is also consistent with the erosion budget based on Sr-Nd data (Singh and France-Lanord, 2002), which indicates that the Tsangpo contributes only ⬃5% of the total sediments of the Brahmaputra, with the rest of the Siang contribution coming from erosion in the syntaxis region, i.e., region surrounding Namche-Barwa. The Os isotopic signatures of the major Brahmaputra tributaries and the main Brahmaputra channel are plotted as a function of ␧Nd in Figure 5 (␧Nd from Singh and FranceLanord, 2002). To quantify the sediment fractions derived from each of the main Himalayan units (Lesser Himalaya, Higher Re-Os systematics in the Brahmaputra River system Fig. 5. Variation diagram relating ␧Nd and 187Os/188Os for the sediments of the Brahmaputra River system. Nd data for the sediments are taken from Singh and France-Lanord (2002). Fields of various end members based on data from France-Lanord et al. (1993), Turner et al. (1996), Singh et al. (1999), Ahmad et al. (2000) and Pierson-Wickmann et al. (2000) are also plotted. The 187Os/188Os ratios of the end members of the mixing curves are based on sediments of the Narayani River outflow for the LH and the average of Central Nepal stream sediment data for the HH (both from Pierson-Wickmann et al., 2000), while that of the TPB is taken from the Dibang River bank sediment (this study). The plot shows that 60 –90% of the Brahmaputra River sediments may be derived from Himalayan formations with Os characteristics similar to those of the Higher Himalaya. This result is consistent with those based on Sr-Nd systematics of these sediments (Singh and France-Lanord, 2002). Himalaya, and Transhimalayan Plutonic Belt) the Os isotopic compositions of these units are required. The Os systematics of these formations have not been studied in the eastern Himalaya, but the lithology (Thakur, 1986) and existing Sr and Nd data (Dietrich and Gansser, 1981; Bhalla and Bishui, 1982; Bhalla et al., 1982; Trivedi, 1990; Dikshitulu et al., 1995; Burg et al., 1998) strongly suggest that the different formations of the eastern section of the Himalaya are comparable to those of the central and western sections. Therefore 187Os/188Os ratios of the main lithologies in the eastern Himalaya are assumed based on values available from these units in the central and western Himalaya (Singh et al., 1999; Pierson-Wickmann et al., 2000). These potential end members are indicated in Figure 5. The LH have very unradiogenic Nd isotopic compositions (FranceLanord et al., 1993; Ahmad et al., 2000) coupled with radiogenic Os compositions reflecting the presence of 1 to 2% black shale (Singh et al., 1999; Pierson-Wickmann et al., 2000). The HH have higher Nd (France-Lanord et al., 1993; Ahmad et al., 2000) and lower Os isotopic ratios (Pierson-Wickmann et al., 2000). The TPB has high Nd isotopic ratios (Turner et al., 1996) suggesting recent derivation from the mantle, consistent with the presence of a suture zone in this region. While we have no Os isotopic data from rocks in the TPB, geologic, geochemical and tectonic information (Sharma, 1991; Kumar, 1997) suggest that this formation continues eastward into the Mishmi hills. As the Dibang River flows through the Mishmi hills and has a 143Nd/144Nd ratio within the TPB range, we use the Os isotopic ratio of this river to define the TPB field in Figure 5. 4107 Unfortunately, no isotopic data is available for the rock formations that feed the southern tributaries. When viewed in Figure 5, the suspended sediment (BR 59SL) from the Siang appears to fall in the HH field. The Siang bank sediment (BR 60), however, requires an additional radiogenic Os component, likely provided by black shales of the Gondwana group, as suggested above. Most of the sediments of the Himalayan tributaries are derived from the Higher Himalaya or from mixtures of HH with LH sources containing ⬃2% black shale. Nevertheless a few Himalayan tributaries, such as the Subansiri River, may include a minor contribution from the Transhimalayan belt. As noted earlier, the sediment of the northernmost of the eastern tributaries, the Dibang, was used to define the TPB field. The other Eastern tributary, the Lohit, has a Nd composition indicating that the TPB contribution was diluted with more ancient crustal material. This is confirmed by the 187Os/188Os ratio of this sample, which is much higher than that of the Dibang. The presence of a highly radiogenic component is required to explain this Os composition, given that the Os concentration of the TPB component, represented by the Dibang sediment, is quite high. Again, ancient black shales would be the best candidate, especially considering the high Re concentration of the Lohit sediment. The southern tributaries have highly variable isotopic compositions. The Burhi Dihing and the Dhansiri both have very low Os isotopic ratios, consistent with the presence of the Indo-Burmese suture zone in the region, but their Nd isotopic compositions are quite different. Despite the large heterogeneity observed among sediments from the various tributaries, the sediments of the Brahmaputra main channel are very tightly clustered in the Os-Nd variation diagram, with the exception of one sample from Bangladesh (BGP 14) and to a lesser extent, that from Dibrugarh (BR 19). This implies very uniform mixing proportions of sediments from the different source formations. The isotopic compositions of most of the Brahmaputra sediments can be explained by mixing of 10 – 40% sediments from the TPB with 60 –90% from the Himalaya, including a possible small (⬃1%) contribution from Lesser Himalayan type black shales (Singh et al., 1999; Pierson-Wickmann et al., 2000). The unusually radiogenic character of the LH unit results mainly from the presence of these ancient black shales; most other LH lithologies have Os compositions similar to those of HH rocks (Pierson-Wickmann et al., 2000). Thus it is impossible to distinguish the contributions from the LH (other than black shale) and the HH using Os data. The relative proportions of Himalayan and TPB material are in the range of those obtained using Sr-Nd data (Singh and France-Lanord, 2002). The Brahmaputra sources can also be examined in terms of the sediment proportions provided by each drainage system. On the basis of Sr and Nd isotopes, it has been found (Singh and France-Lanord, 2002) that the Siang represents about half of the sediment budget of the Brahmaputra system whereas the Himalayan tributaries and the eastern tributaries contribute ⬃40 and 10% to the overall budget. The Os systematics are consistent with these proportions. Average Os concentrations and weighted average 187Os/188Os ratios were determined for each drainage system. For the eastern system, the Burhi Dihing was also included as it has very high Os concentrations and may thus be important for the Os budget despite its low 4108 S. K. Singh, L. Reisberg, and C. France-Lanord Table 2. Sediment proportions from each drainage system. Drainage Siang Himalayan Eastern Calculated Dhubri average a Proportion of sedimenta 188 Os (fm/g) 50% 40% 10% 100% 14 10 70 18 17.2 187 Os/ Os 188 1.7 1.2 0.3 1.0 0.99 143 Nd/ ␧Nda 144 ⫺13.3 ⫺16.4 ⫺9.6 ⴚ14.17 ⫺14.2 From Singh and France-Lanord (2002). sediment discharge. The average values calculated for each drainage system are given in Table 2. The calculated composition of the sediment mixture is also given. This calculated value is in excellent agreement with the composition of sediments collected at Dhubri (weighted average of bank sediment and suspended load given in Table 2), near the outflow of the Brahmaputra. Note that this value is also consistent with the Os composition of sediments delivered to the ocean by the Brahmaputra, inferred on the basis of analyses of sediments from the Bay of Bengal and the Ganga (PiersonWickmann et al., 2001). In other words, the Os compositions of turbidites sampled near the active canyon and shelf off Bangladesh (187Os/188Os ⫽ 1.2–1.5) reflect a mixture of more radiogenic sediments delivered by the Ganga (187Os/188Os ⬃ 2.3–2.6, Pierson-Wickmann et al., (2000) with Brahmaputra sediments similar to those collected at Dhubri. Our study supports the earlier conclusion (Singh and FranceLanord, 2002) that the Higher Himalaya formation is the major source of sediments to the Brahmaputra River system with some contribution from the Transhimalayan formations. About half of the sediment is delivered by the Siang River. Nevertheless, even though the sediment supply from the eastern and southern tributaries is not very significant in terms of volume percent, these tributaries contribute considerably to the Os budget of the Brahmaputra as they contain high concentrations of Os. As the sediments of the Siang at Pasighat have fairly radiogenic Os, it is the rivers from the east and south that explain the relatively low Os isotopic ratios of the Brahmaputra main channel sediments. Thus the lower 187Os/188Os ratios of Brahmaputra River sediments relative to those of the Ganga are primarily due to the presence of mantle derived lithology in the drainage basins of the eastern and southern tributaries. The lower proportion of LH (the formation that includes radiogenic black shales) in the eastern Himalaya, compared to the LH proportion in the central and western Himalaya, also plays a role. 5.2. Relation between Dissolved and Particulate Os The 187Os/188Os ratio of the Brahmaputra sediment at Guwahati varies from 0.7 to 1.1 with a weighted average of 0.8. The isotopic ratio of the dissolved Os at Guwahati is 1.07 (Sharma et al., 1999) which is just slightly more radiogenic than the sediment isotopic composition. Similarly, there is a good correspondence between water (187Os/188Os ⫽ 2.9; Levasseur et al., 1999) and bank sediment (2.3 and 2.6; PiersonWickmann et al., 2000) collected at the same location in the Ganga River, with the water 187Os/188Os ratio being again slightly higher. In a study of runoff in Papua New Guinea, Martin et al. (2000) also found that the water sample was slightly more radiogenic than the corresponding sediment. Taken together these observations suggest that in general the isotopic composition of particulate Os is a little less radiogenic than that of dissolved Os collected in the same place. In fact, the average 187Os/188Os ratio of river water from throughout the world (⬃1.5; Levasseur et al., 1999) is slightly more radiogenic that of estuarine sediments thought to represent the upper continental crust (⬃1.3; Esser and Turekian, 1993), suggesting that this pattern may hold true on a global level. These observations support the suggestion that there is some preferential dissolution of 187Os during weathering (PeuckerEhrenbrink and Blum, 1998; Jaffe et al., 2002; Pierson-Wickmann et al., 2002a), or alternatively, that Re rich minerals (like black shales) are more easily weathered. Nevertheless, these data also suggest that this effect is small, that is, that fractionation of radiogenic Os from non-radiogenic Os during weathering and transport is limited. 5.3. Contribution to the Os Isotope Evolution of Seawater It is well established that the 187Os/188Os ratio of the ocean has increased markedly during the Cenozoic, and particularly over the past ⬃16 My (Pegram et al., 1992; Ravizza, 1993; Peucker-Ehrenbrink et al., 1995). In analogy with the interpretation of the marine Sr isotopic record, it was suggested that this increase was due largely to uplift and erosion of the Himalaya (Pegram et al., 1992; Peucker-Ehrenbrink et al., 1995). This possibility seemed especially promising as the Lesser Himalaya were known to contain ancient black shales, which have very radiogenic Os compositions (Singh et al., 1999; Pierson-Wickmann et al., 2000). However, recent workers (Levasseur et al., 1999; Sharma et al., 1999) have argued that Himalayan erosion did not have much effect on the Os composition of seawater, because the Os flux delivered by the Ganga, though quite radiogenic, is small. If the increase in seawater 187Os/188Os does not reflect erosion of highly radiogenic Himalayan lithologies, it may instead imply higher overall continental weathering rates over the last 16 Ma, which would in turn have implications for the CO2 content of the atmosphere. Thus it is important to consider whether Himalayan erosion can really be excluded as the primary cause of the recent increase of radiogenic Os in the oceans. We first consider the total Os flux from the Ganga-Brahmaputra (G-B) river system. We use the Os concentration (12 pg/kg) and isotopic ratio (187Os/188Os ⫽ 2.9) determined by Levasseur et al. (1999) for Ganga River water. (The lower ratio obtained by Sharma et al., 1999, is unlikely to be representative of Ganga water at the outflow, as it was determined from a sample collected upstream of the confluence with the Kosi River, which drains Lesser Himalayan terrain rich in black shales.) We use the water Os composition determined by Sharma et al. (1999) for the Brahmaputra (187Os/188Os ⬃ 1.07), which is supported by our measurements of sediments collected in the same location and elsewhere in the main Brahmaputra channel. The combined 187Os/188Os ratio of the G-B is ⬃1.8 with an Os concentration of 10.7 pg/kg, as the water fluxes for the Ganga and the Brahmaputra are 4.9 ⫻ 1011 m3/a and 6.3 ⫻ 1011 m3/a, respectively (Rao, 1979). Together these rivers Re-Os systematics in the Brahmaputra River system supply ⬃4% of the total global riverine Os delivered to the ocean, with an Os isotopic ratio significantly more radiogenic than that (⬃1.54) estimated for average river water (Levasseur et al., 1999). The G-B 187Os flux is ⬃4.6% of the global river flux. Thus, the proportion of the global riverine Os flux supplied by the G-B is greater than that of its water discharge (⬃2.8%) and almost double that of its Sr flux (⬃2%). Even though the Sr discharge from the G-B represents only 2% of the total riverine Sr flux, it plays a major role in controlling the Sr isotopic evolution of seawater (Krishnaswami et al., 1992; Richter et al., 1992; Galy et al., 1999). So it cannot be assumed that the contribution of the G-B to the seawater Os isotopic budget is negligible, solely on the grounds that it provides only 4% of the Os flux. To get a rough idea of the influence of the Ganga on seawater Os, we consider the effect of changing the highly radiogenic Os isotopic composition of Ganga water to a value more typical of most of the world’s rivers. As the residence time of Os in seawater is quite short (between 5000 and 50000 yr; see discussion in Oxburgh, 2001), the marine Os composition will most probably reflect the currently observed inputs. We assume that all of the radiogenic input is delivered as dissolved Os in river water, and use Levasseur et al.’s (1999) estimate for the mean 187Os/188Os of the world’s rivers (1.54). We also assume an average 187Os/188Os ratio of 0.126 for the non-radiogenic Os inputs (cosmic dust and alteration of oceanic crust). Mass balance then indicates that ⬃66% of the ocean’s Os is derived from river water. As the Ganga Os flux, with 187Os/188Os ⬃ 2.94 (Levasseur et al., 1999), represents ⬃2% of the global riverine total, the average 187Os/188Os ratio of the rest of the world’s rivers is ⬃1.51. If the Ganga River had an Os isotopic composition similar to this average value, seawater would have a 187Os/188Os ratio of 1.039, instead of the current value of ⬃1.06, assuming constant relative proportions of riverine and non-radiogenic Os sources. In other words, ⬃7% of the change in the seawater 187Os/188Os ratio over the past 16 Ma (from 0.75 to 1.06) could be explained by increasing the Os isotopic ratio of Ganga water to its present value. There are of course very large uncertainties on this figure, due to our limited knowledge of the required data, in particular the average Os isotopic composition of river water, and the Ganga Os flux. This calculation also assumes that the total Os flux delivered by the Ganga has not changed. If, as seems likely, Himalayan erosion has increased the Os flux from the Ganga, then this figure is an underestimate. This will be even more true if suspended particles and bed load carried by the Ganga eventually release some of their Os to seawater. Even more uncertainty concerns the possible contribution of other rivers that drain the Himalayan-Tibetan Plateau (HTP). Our sediment results, as well as the water results of Sharma et al. (1999), demonstrate that the Brahmaputra River provides Os with an isotopic composition similar to that of present-day seawater. Thus the Brahmaputra is likely to have very little effect on the present-day marine Os isotopic budget. Nevertheless, this does not mean that other HTP rivers will have little effect. In particular, the Indus may play an important role. Clift et al. (2002) demonstrated that the Nd isotopic composition of sediments in the lower part of the Indus are similar to those of the Ganga. Singh et al. (1999) showed that the Os compositions of black shales collected in the Indus drainage basin are among 4109 the most radiogenic of the Lesser Himalaya. The importance of black shales in this drainage basin is also shown by the high uranium content of the Indus River water (Pande et al., 1994), which is similar to that of the Ganga. (The Indus water sample [Sharma et al., 1999] and paleosols [Chesley et al., 2000] that yielded relatively unradiogenic Os compositions were collected upstream of the black shale exposures and thus are unlikely to reflect the composition of the Os delivered by the Indus to the ocean.) In addition, several other major Chinese and Indochinese rivers have their headwaters in the HTP. These include the Chang Jiang (Yangtze), which has a high water discharge (9.3 ⫻ 1011 m3/a), a high 187Os/188Os ratio (1.95) and a relatively high Os concentration (13.9 pg/kg) (Levasseur et al., 1999). Thus the possibility that uplift of the Himalayan-Tibetan Plateau has contributed to the late Cenozoic increase of the marine Os isotopic ratio cannot be rejected at this time. More complete data sets for all the HTP rivers are needed to test the importance of the Indo-Asian collision for the Os isotopic evolution of seawater. 6. SUMMARY AND CONCLUSIONS For most of the length of the Brahmaputra River, both suspended and bank sediments have a fairly uniform 187Os/ 188 Os ratio of ⬃1, which is much lower than the isotopic ratio (⬃2.6; Pierson-Wickmann et al., 2000) of sediments delivered by the Ganga River. While little difference exists between the Os isotopic compositions of bank sediments and suspended matter collected in the same location, suspended sediments nearly always have higher Re and Os concentrations than the corresponding bank sediments. In most cases, no systematic differences in either concentration or isotopic composition are observed between samples collected in a given locality during or outside of the monsoon period. Analysis of the Os compositions of the main tributaries suggest a sediment source budget for the Brahmaputra River that is consistent with the source proportions (50% Siang River, 40% Himalayan tributaries, 10% eastern and southern tributaries) previously determined on the basis of Nd isotopes (Singh and France-Lanord, 2002). While the volumetric fraction of sediments derived from the eastern and southern tributaries is small, these rivers play a disproportionately large role in controlling the Os isotopic compositions of the sediments of the main Brahmaputra channel. This is because they provide material rich in unradiogenic Os, reflecting the presence of mantle-derived ultramafic rocks in their drainage basins. On the other hand, the Siang River, the southern continuation of the Tsangpo River that flows along the Indus-Tsangpo suture in Tibet, does not provide unradiogenic Os, in contrast to what has been suggested in the past. This may reflect the presence of a knickpoint before Namche-Barwa that blocks much of the sediment carried by the Tsangpo. Thus the markedly lower 187Os/188Os ratio of the Brahmaputra River sediments compared to those of the Ganga results mainly from the contribution of the eastern and southern tributaries that carry mantle-derived phases from the suture zones in that region. The smaller proportion of Lesser Himalayan lithology, which includes radiogenic black shales, in the eastern part of the Himalaya relative to the western and central regions drained by the Ganga, probably also plays a role. This suggestion is supported by the low uranium content of Brahmaputra River 4110 S. K. Singh, L. Reisberg, and C. France-Lanord water (Sarin et al., 1990; Chabaux et al., 2001), which argues against an important component of black shale in the drainage basin. The relatively unradiogenic Os isotopic signature of the Brahmaputra sediments is consistent with that of a water sample collected in the same locality (Sharma et al., 1999). This result confirms the suggestion that the Brahmaputra outflow has had very limited effect on the Os isotopic composition of seawater. On the other hand, reconsideration of the Ganga River outflow suggests that its contribution may be significant. Although we have excluded an important role for the Brahmaputra River, the total effect of the Indo-Asian collision on the seawater Os composition cannot be evaluated until information is available for all of the rivers that drain or have headwaters in the Himalayan Tibetan Plateau. Acknowledgments—This work was supported by a MNESR fellowship to SKS and grants from the French “Programme National Sol Erosion.” We thank Amulya Narzary for help during sampling in Assam and Catherine Zimmermann for laboratory and analytical support. Constructive reviews of Dr. D. K. McDaniel and associate editor R. J. Walker helped improve the manuscript. Associate editor: R. Walker REFERENCES Ahmad T., Harris N., Bickle M., Chapman H., Bunbury J., and Prince C. (2000) Isotopic constraints on the structural relationship between the Lesser Himalaya Series and the high Himalayan Crystalline Series, Garhwal Himalaya. Geol. Doc. Am. Bull. 112, 467– 477. Bhalla J. K. and Bishui P. K. (1982) Geochronology and geochemistry of granite emplacement in north eastern Himalaya. Geol. Surv. India Calcutta 122, 18 –20. Bhalla J. K., Srimal N., and Bishui P. K. (1982) Isotopic study of Mishmi complex, Arunachal Pradesh, North Eastern Himalaya. Geol. Surv. India Calcutta 122, 20. Birck J.-L., Roy Barman M., and Capmas F. (1997) Re-Os isotopic measurements at the femtomole level in natural samples. Geostandard Newslet. 20, 19 –27. Burg J.-P., Nievergelt P., Oberly F., Seward D., Davy P., Maurin J.-C., Dio Z., and Meier M. (1998) The Namche Barwa syntaxis: Evidence for exhumation related to compressional crustal folding. J. Asian Earth Sci. 16, 239 –252. Chabaux F., Riotte J., Clauer N., and France-Lanord C. (2001) Isotopic tracing of the dissolved U fluxes of Himalayan rivers: Implications for present and past U budgets of the Ganges-Brahmaputra system. Geochim. Cosmochim. Acta 65, 3201–3217. Chesley J. T., Quade J., and Ruiz J. (2000) The Os and Sr isotopic record of Himalayan paleorivers: Himalayan tectonics and influence on ocean chemistry. Earth Planet. Sci. Lett. 179, 115–124. Clift P. D., Lee J. I., Hildebrand P., Shimizu N., Layne G. D., Blum J. D., Garzanti E., and Khan A. A. (2002) Nd and Pb isotope variability in the Indus River system: Implications for sediment provenance and crustal heterogeneity in the Western Himalaya. Earth Planet. Sci. Lett. 200, 91–106. Colin C., Turpin L., Bertaux J., Desprairies A., and Kissel C. (1999) Erosional history of the Himalayan and Burman ranges during the last two glacial-interglacial cycles. Earth Planet. Sci. Lett. 171, 647– 660. Creaser R. A., Papanastassiou D. A., and Wasserburg G. J. (1991) Negative thermal ion mass spectrometry of osmium, rhenium, and iridium. Geochim. Cosmochim. Acta 55, 397– 401. Dietrich V. and Gansser A. (1981) The leucogranites of the Bhutan Himalaya (crustal anatexis versus mantle melting). Bull. Suisse Mineral. Petrogr. 61, 177–202. Dikshitulu G. R. Pandey B. K., Krishna V., and Dhana Raju R. (1995) Rb-Sr systematics of granitoids of the Central Gneissic Complex, Arunachal Himalaya: Implications on tectonism, stratigraphy and source. J. Geol. Soc. India 45, 51–56. Esser B. K. and Turekian K. K. (1993) The osmium isotope composition of the continental crust. Geochim. Cosmochim. Acta 57, 3093– 3104. France-Lanord C., Derry L., and Michard A. (1993) Evolution of the Himalaya since Miocene time: Isotopic and sedimentological evidence from the Bengal Fan. In Himalayan Tectonics (eds. P. J. Trelor and M. Searle), pp. 603– 621. Special Publication 74. Geological Society of London. Galy A. (1999) Etude Géochimique de l’érosion Actuelle de la Chaı̂ne Himalayenne. Ph.D. Thesis. Institut National Polytechnique de Lorraine, France. Galy A., France-Lanord C., and Derry L. A. (1999) The strontium isotopic budget of Himalayan rivers in Nepal and Bangladesh. Geochim. Cosmochim. Acta 63, 1905–1925. Galy A. and France-Lanord C. (2001) Higher erosion rates in the Himalaya: Geochemical constraints on riverine fluxes. Geology 29, 23–26. Jaffe L. A., Peucker-Ehrenbrink B., and Petsch S. T. (2002) Mobility of rhenium, platinum group elements and organic carbon during black shale weathering. Earth Planet. Sci. Lett. 198, 339 –353. Krishnaswami S., Trivedi J. R., Sarin M. M., Ramesh R., and Sharma K. K. (1992) Strontium isotopes and rubidium in the Ganga-Brahmaputra River system: Weathering in the Himalaya, fluxes to the Bay of Bengal and contributions to the evolution of oceanic 87Sr/ 86 Sr. Earth Planet. Sci. Lett. 109, 243–253. Kumar G. (1997) Geology of the Arunachal Pradesh. Geological Society of India. Levasseur S., Birck J.-L., and Allegre C. J. (1999) The osmium riverine flux and the oceanic mass balance of osmium. Earth Planet. Sci. Lett. 174, 7–23. Mallet F. R. (1882) On iridosmine from the Nao-dihing River, Upper Assam, and on platinum from Chutia Nagpur. Rec. Geol. Survey India 15, 53–55. Martin C. E., Peucker-Ehrenbrink B., Brunskill G. J., and Szymczak R. (2000) Sources and sinks of unradiogenic osmium runoff from Papua New Guinea. Earth Planet. Sci. Lett. 183, 261–274. Oxburgh R. (2001) Residence time of osmium in the oceans. Geochem. Geophys. Geosystems 2. p. 2000GC000104. Palmer M. R. and Edmond J. M. (1989) The strontium isotope budget of the modern ocean. Earth Planet. Sci. Lett. 92, 11–26. Pande K., Sarin M. M., Trivedi J. R., Krishnaswami S., and Sharma K. K. (1994) The Indus river system (India-Pakistan): Major-ion chemistry, uranium and strontium isotopes. Chem. Geol. 116, 245– 259. Pegram W. J., Krishnaswami S., Ravizza G. E., and Turekian K. K. (1992) The record of seawater 187Os/186Os variation through the Cenozoic. Earth Planet. Sci. Lett. 113, 569 –576. Pegram W. J., Esser B. K., Krishnaswami S., and Turekian K. K. (1994) The isotopic composition of leachable osmium from river sediments. Earth Planet. Sci. Lett. 128, 591–599. Peucker-Ehrenbrink B., Ravizza G., and Hofmann A. W. (1995) The marine Os-187/Os-186 record of the past 80-million years. Earth Planet. Sci. Lett. 130, 155–167. Peucker-Ehrenbrink B. and Blum J. D. (1998) Re-Os isotope systematics and weathering of Precambrian crustal rocks: Implications for the marine osmium isotope record. Geochim. Cosmochim. Acta 62, 3193–3203. Peucker-Ehrenbrink B., Jahn B. (2001) Rhenium-osmium isotope systematics and platinum group element concentrations: Loess and upper continental crust. Geochem. Geophy. Geosystems 2. p. 2001GC000172. Pierson-Wickmann A.-C., Reisberg L., and France-Lanord C. (2000) The Os isotopic composition of Himalayan river bedloads and bedrocks: Importance of black shales. Earth Planet. Sci. Lett. 176, 203–218. Pierson-Wickmann A.-C., Reisberg L., France-Lanord C., and Kudrass H. (2001) Os-Sr-Nd results from sediments in the Bay of Bengal: Implications for sediment transport and the marine Os record. Paleoceanography 16, 435– 444. Re-Os systematics in the Brahmaputra River system Pierson-Wickmann A.-C., Reisberg L., and France-Lanord C. (2002a) Behavior of Re and Os during low-temperature alteration: Results from Himalayan soils and altered black shales. Geochim. Cosmochim. Acta 66, 1539 –1548. Pierson-Wickmann A.-C., Reisberg L., and France-Lanord C. (2002b) Hydrothermally altered marble: Another source of continental radiogenic osmium. Earth Planet. Sci. Lett. 204, 203–214. Rao K. L. (1979) India’s Water Wealth. Orient Longman. Ravizza G. (1993) Variations of the 187Os/186Os ratio of seawater over the past 27 million years as inferred from metalliferous carbonates. Earth Planet. Sci. Lett. 118, 335–348. Richter F. M., Rowley D. B., and DePaolo D. J. (1992) Sr isotope evolution of seawater: The role of tectonics. Earth Planet. Sci. Lett. 109, 11–23. Roy Barman M. (1993) Mesure du rapport 187Os/188Os dans les basaltes et les péridotites: Contribution à la systématique 187Re-187Os dans le manteau. Ph.D. Thesis. Université de Paris VII, France. Sarin M. M., Krishnaswami S., Somayajulu B. L. K., and Moore W. S. (1990) Chemistry of uranium, thorium and radium isotopes in the Ganga-Brahmaputra River-systems: Weathering processes and fluxes to the Bay of Bengal. Geochim. Cosmochim. Acta 54, 831– 845. Sharma K. K. (1991) Geological settings of the ophiolites and magmatic arc of the Lohit Himalaya (Arunachal Pradesh), India with special reference to their petrochemistry. In Geology and Geodynamics Evolution of the Himalayan Collision Zone, Physics and Chemistry of the Earth, Vol. 18(I–II), pp. 277–292 Pergamon. 4111 Sharma M., Wasserburg G. J., Hoffmann A. W., and Chakrapani G. J. (1999) Himalayan uplift and osmium isotopes in oceans and rivers. Geochim. Cosmochim. Acta 63, 4005– 4012. Singh S. K., Trivedi J. R., and Krishnaswami S. (1999) Re-Os isotope systematics in black shales from the Lesser Himalaya: Their chronology and role in the 187Os/188Os evolution of seawater. Geochim. Cosmochim. Acta 63, 2381–2392. Singh S. K. and France-Lanord C. (2002) Tracing the distribution of erosion in the Brahmaputra watershed from isotopic compositions of stream sediments. Earth Planet. Sci. Lett. 202, 645– 662. Thakur V. C. (1986) Tectonic zonation and regional framework of eastern Himalaya. Sci. Terre 47, 347–360. Trivedi J. R. (1990) Geochronological studies of Himalayan granitoids. Ph.D. thesis. Gujarat University, India. Turner S., Arnaud N. O., Liu J., Rogers N., Hawkesworth C. J., Harris N., Kelley S., Calsteren P., and Deng W. M. (1996) Post-collision, shoshonitic volcanism on the Tibetan Plateau: Implication for convective thinning of the lithosphere and the source of ocean island basalts. J. Petrol. 37, 45–71. Völkening J., Valczyk T., and Heumann K. G. (1991) Osmium isotope ratio determinations by negative thermal ionisation mass spectrometry. Mass Spectr. Ion Processes 105, 147–159. Zeitler P. K., Meltzer A. S., Koons P. O., Craw D., Hallet B., Chamberlain C. P., Kidd W. S. F., Park S. K., Seeber L., Bishop M., and Shroder J. (2001) Erosion, Himalayan geodynamics, and the geomorphology of metamorphism. GSA Today, 11, 4 –9.