Sedimentology of the Indus Group, Ladakh, northern India ... · PDF fileSedimentology of the...

Journal of the Geological Society, London, Vol. 158, 2001, pp. 151–162. Printed in Great Britain.

Sedimentology of the Indus Group, Ladakh, northern India: implications for thetiming of initiation of the palaeo-Indus River

H. D. SINCLAIR1 & N. JAFFEY1

1Department of Geology and Geophysics, Grant Institute, University of Edinburgh, West Mains Road,Edinburgh, EH9 3JY UK (e-mail: [email protected])

Abstract: The upper reaches of the Indus River form a longitudinal network that drains a large portion ofthe western Himalaya. It plays an important role as a sediment routing system, feeding the Indus Deltaand submarine fan, and has played a controlling role in the denudational history of the western Himalaya.Given the rivers long-term significance, its timing of initiation remains poorly constrained. The facies,palaeocurrents and provenance of the post-early Eocene Indus Group preserved along the upper reachesof the modern Indus reveal a history of fluvial/deltaic and deep-water sedimentation in an intermontanebasin dominated by internal drainage. Hence, the Indus Group is not considered to represent the depositsof a palaeo-Indus River as previously thought. Illite crystallinity values and apatite fission track dating ofthe Indus Group suggest that the succession was buried to temperatures of 155–280�C, and that unroofingstarted in early Miocene times and proceeded at 0.1–0.4 mm a�1 to the present. The youngest sedimentaryrocks preserved along the Indus Basin are early Miocene in age from the tectonostratigraphicallyequivalent deposits around Kargil to the west of the study area. The transition from sedimentaccumulation to erosional unroofing in early Miocene times coincides with accelerated regional unroofingof the High Himalayas to the south, and the initiation of the Indus Delta/submarine fan to the west.Differential uplift between the northward thrusting of the Zanskar and Indus sedimentary successionagainst the undeformed Ladakh Batholith provides a mechanism for post-early Miocene initiation of amajor longitudinal river. Hence, early Miocene times is believed to represent the earliest possible age ofinitiation of the palaeo-Indus river following this course.

Keywords: Indus, Himalayas, Ladakh, molasse, intermontane basins.

The Indus Basin of Ladakh, northern India, records a historyof sedimentation along the Indus Suture Zone that spansaccumulation in a fore-arc basin prior to collision, and anintermontane basin post-collision (i.e., from Early Palaeoceneto Miocene times; Garzanti & Van Haver 1988; Searle et al.1990). The younger, fluvially dominated sedimentary succes-sion termed the Indus Group are preserved along the present-day course of the Indus River, and have been cited as evidencefor a long-lived, antecedent history of development of theIndus system along this route (Searle 1996; Qayyum et al.1997; Friend 1998; Clift et al. 2000). New facies and palaeo-current data from the Indus Group are presented in order toevaluate this proposal.

The modern Indus River represents one of the major riversof the world, with the seventh highest sediment discharge of291 Mt a�1 from its river mouth (Milliman & Meade 1983;Hovius 1998). It is sourced from the southern margin of Tibetin the region of Mount Kailas, from where it flows west-northwestward, for approximately 880 km along the strikeof the mountain belt (Fig. 1). It initially follows theIndus–Tsangpo continental suture zone, then traverses theLadakh batholith, finally heading south, cutting throughthe Himalayan mountain belt around the Nanga ParbatMassif (Fig. 1). This distance of longitudinal flow is secondonly to the Yarlung/Tsangpo river which follows a similartectono-geomorphic route to the eastern syntaxis of theHimalaya before heading southward as the Brahmaputra.

Large-scale longitudinal rivers form the first harmonic oftopography in large mountain belts (Koons 1994). Theirdevelopment significantly modifies sediment routing systemscompared to the transverse networks found in many smaller

mountain belts (Hovius 1996). The most spectacular exampleof this is south of the Himalaya where the Indus and Bengalfans are fed by river systems running longitudinally within themountain belt (Indus and Yarlung–Tsangpo), and along themountain front (Ganges and Brahmaputra; Fig. 1).

Longitudinal rivers may also play an important role in thedenudation of a mountain belt. The longitudinal component ofriver systems are typically long, low gradient alluvial channels(i.e., channels bound by their own alluvial sediment as opposedto bedrock), and where fluvial incision rates, and hencedenudation rates, are low compared to topographic uplift(Koons 1995). However, where these rivers transverse the

Fig. 1. Map of main Himalayan rivers feeding the Indus and Bengalfans. The study area is boxed, and is located along the longitudinalportion of the Indus River in Ladakh, northern India (Fig. 3).

151

mountain range, high discharge combined with steep slopesresults in high rates of fluvial incision. Modelling suggests thathigh fluvial incision rates will generate accelerated regionaldenudation through the maintenance of hillslopes by mass-wasting (Anderson 1994); these models are supported bycomparisons of incision rates with longer-term denudationrates where the Indus River flows around the Nanga Parbatmassif (Burbank et al. 1996). Additionally, it has been sug-gested that these sustained high rates of denudation may haveresulted in the focusing of increased mechanical strain inthis area as evidenced by folding and thrusting (Seeber et al.1998; Zeitler & Koons 1998). Given the sedimentological,denudational and possible mechanical significance of the IndusRiver as a major longitudinal system, it is important to havesome knowledge of the timing of its initiation, and hence to beable to compare it to the sedimentary record in the Indus fan,and the erosional record of the Nanga Parbat Massif.

The natural location to look for evidence of a palaeo-IndusRiver running longitudinally within the mountain belt is alongthe course of the present-day river. Three localities along theupper Indus preserve a sedimentary record (Fig. 1): (1) MountKailas in southern Tibet where thick conglomerates drape thesouthern slopes of the Gangdese Batholith (Gansser 1964;Harrison et al. 1993); (2) Ladakh, northern India (this study);and (3) Kargil, NE Kashmir (Brookfield & Andrews-Speed1984b). The Indus Group in Ladakh was chosen for this studydue to the presence of previous background work (Van Haver1984) and logistical considerations. In order to test thepresence or absence of a palaeo-Indus within the IndusGroup, testable predictions must be made. During periods ofaggradation of a palaeo-Indus River along the Indus suturezone, facies and palaeocurrent data should record evidence ofsustained fluvial discharge to the NNW. During aggradation,the denudation of the drainage basin, and the discharge ofsediment out of the system would have been relatively low. Incontrast, during periods of relatively high denudation andsediment discharge, the fluvial system would be degradationaland incising, and the only preserved evidence may be an

unconformity. During these periods, the thermochronology ofthe rocks beneath the unconformity may be used to evaluatethe timing of initial degradation. This record may be compli-cated by localized, structurally induced uplift and degrada-tion during sedimentation; this would be recorded by growthstrata and time equivalent unconformities. These predic-tions are tested using the sedimentology, stratigraphy andthermochronology of the Indus Group.

The Indus Group, LadakhThe Indus Group has been defined by Searle et al. (1990), andrepresents a dominantly continental succession, at least 2 kmthick, of post-early Eocene age. It crops out SW of the IndusRiver between Khalsi in the NW, and Upshi in the SE.Underlying the Indus Group are a succession of marine strataof Late Albian to Eocene age described by Van Haver (1984)and Brookfield & Andrews-Speed (1984a). Structurally, theIndus Group are located in a 15–20 km wide zone, bound tothe SSW by a suite of highly deformed rocks derived from theIndian passive margin (Zanskar units), the forearc (NindamUnit) and the volcanic arc rocks (Dras Unit), and ophioliticmelange units of the Indus Suture Zone (Bassoulet et al. 1983;Searle et al. 1990). The north-northeastern margin of theIndus Group basin onlaps onto the Ladakh arc batholith ofthe southern margin of Tibet (Van Haver 1984; Searle et al.1997; Fig. 2). The Indus Group have been interpreted as theremnants of an intermontane basin (Van Haver 1984; Garzanti& Van Haver 1988; Searle et al. 1990), conformably underlainby sediments which represent marine conditions in a fore-arcbasin (Searle et al. 1990). Intense structural deformation andlack of preserved microfauna have hampered the establishmentof a reliable biostratigraphy for the Indus Group (Fig. 2).However, at the base of the succession are foraminiferallimestones of the Sumda Formation which were dated asPalaeocene by Van Haver (1984) using gastropods andbivalves, and have been recently dated as latest Palaeocene

Fig. 2. Comparison of proposedstratigraphies for the Tertiary of theIndus Basin, Ladakh. Wakka ChuMolasse shown but not correlated.

152 H. D. SINCLAIR & N. JAFFEY

(54.9 Ma; O. Green, pers. comm.) using nummulitid bio-stratigraphy. Higher in the succession are remains of palmleaves and bivalves dated as Upper Eocene/Lower Oligocene(Van Haver 1984). The uppermost strata (Nimu Formation) ofthe Indus Group are undated. In the Kargil area, 80 km westof this study area, are along strike, tectonostratigraphicallysimilar successions to the Indus Group called the Wakka ChuMolasse. The youngest dates from these rocks are from theTaramsu Formation that yields mammal remains dated asearly Miocene (Brookfield & Andrews-Speed 1984b).

Garzanti & Van Haver (1988) studied the petrographyand geochemistry of the Indus Group and underlyingsuccessions, and demonstrated that the mid-Cretaceous toearly Eocene succession records submarine sedimentationderived from the progressive unroofing of the volcaniccover of the magmatic arc, now represented by the LadakhBatholith; the resultant sandstone composition is similar tomodern and ancient fore-arc basins. At latest Palaeocenetimes (beginning Indus Group) there was a transition topredominantly continental sedimentation, associated witha sharp increase in plutonic detritus from the LadakhBatholith, along with ophiolite-derived sediment. The finaltermination of marine sedimentation in early Eocene times isinterpreted to record the initiation of continental collision(Searle et al. 1987).

Within this tectono-stratigraphic framework, the deposi-tional history of the post-early Eocene succession is analysedherein (i.e., the time from which no evidence of marinesedimentation occurred). This means that the ChogdoFormation which underlies the Nummulitic Limestones(Searle et al. 1990; Fig. 2) is not documented. The develop-ment of a detailed lithostratigraphy within which the sedi-mentology can be studied requires interpretation of thestructural sections, with correlation of units between fold andthrust packages. The assumption in building a more detailedstratigraphic succession through such a deformed region isthat, for the most part, reverse faults and thrusts carry oldermaterial over younger.

Structural cross-sections

Three cross-sections through the Indus Basin were alignedSSW–NNE, perpendicular to the structural strike (Figs 3 & 4).Multiphase folding and thrusting have resulted in the over-printing and interference between structures with vergenceto the NNE and SSW (Searle et al. 1990). The south-southwestern boundary of the Indus Basin sediments is theshear zone of the Indus suture, with an ophiolitic melangeexposed near Chilling (Fig. 3a). To the south of this suture isthe Lamayuru Formation which records deep-water sedimen-tation on the outer margins of the Indian passive margin(Searle et al. 1990, 1997). The northern boundary of the IndusGroup is the Ladakh Batholith, which is locally onlapped byIndus Group sedimentary rocks of the Maastrichtian BasgoFormation; these have been interpreted as alluvial fan depositson the northern margin of the fore-arc basin (Garzanti & VanHaver 1988; Garzanti et al. 1996).

The Indus Group is located in the north-northeasternportion of these cross-sections. Each of the structural sectionscan be divided into a northern and southern unit, separated bythe Choksti fault; a high-angle, south-southwestward dippingreverse fault that transported the Chogdo Formation over theNurla Formation (Searle et al. 1990; Figs 3 and 5). In sectionc (Fig. 4), this fault has been back-steepened to the extent thatit is now dipping north-northeastward. SSW of this fault, theJurutze Formation (Searle et al. 1990) forms the core of abroad (5–15 km), complex anticlinorium, with the SumdaFormation (Sumda Gompa Formation; Van Haver 1984),Chogdo Formation and the Nummulitic Limestones (VanHaver 1984; Searle et al. 1990) folded and thrusted in its limbs.Immediately north of Chogdo, 100 m scale folds with opposingvergence exhibit complex interference structures. Axial planarcleavages can be seen dipping both to the NNE and SSW.

In the Zanskar Gorge (Fig. 4a), to the north of the Chokstifault (in its footwall), is another large (3 km wavelength),upright anticlinorium with vertical fold hinges developed in theNurla Formation. The northern limb again shows interference

Fig. 3. Simplified geological mapshowing the location of the Indus Grouprelative to the modern Indus River inLadakh. The sedimentary rocks of theIndus Group have been thrusted andfolded in a northward direction againstthe undeformed Ladakh Batholith. Thecross-sections in Fig. 4 are located.Location of samples for apatite fissiontrack age determination are starred, withtheir ages shown next to the locations.

INDUS GROUP SEDIMENTATION 153

between upright and SSW-vergent folds on a 50–100 m wave-length. This anticlinorium is again bound to the north byanother thrust on to the Nimu Formation. Sections b and c

also show north-northeastward vergent folding and thrust-ing to the north of the Choksti fault, accompanied by aprogressive decrease in deformation. The thick-bedded Hemis

Fig. 4. Structural cross-sections through the Indus and Tar groups showing some complex interference structures, but with a dominantlynorthward vergence in the northern areas. Cleavage orientation is shown by the dashed lines. Measured sedimentary sections illustrated inFigs 5, 6, 8 and 9 are located.


Conglomerates found in the east, are steeply tilted, but onlylocally folded into large wavelength (several hundred metres)folds.

Stratigraphy and sedimentologyThe stratigraphic subdivision used here is based on Searle et al.(1990; Fig. 2); which itself was based on Thakur (1981) andThakur & Misra (1984). The Jurutze, and Sumda formationsunderlie the Indus Group which were termed the Tar Group bySearle et al. (1990; Figs 2 & 5). In the southern succession ofthe Zanskar Gorge (Fig. 4a), the Nummulitic Limestones areoverlain by the Nurla Formation. North of the Choksti fault,the lower part of the Nurla Formation comprises greysandstone/mudstone alternations exposed at the ChokstiBridge, and hence named the Choksti Bridge Member. It ispossible that the Choksti Bridge Member is synchronous with

the Nummulitic Limestones, as it is present at the base of theNurla Formation in a detached thrust unit to the north. Thefollowing descriptions and interpretations commence with theNummulitic Limestones which are thought to record the lastevidence of a marine seaway between the Indian and Asianplates, and hence sedimentation along the suture immediatelyprior to collision (Searle et al. 1987).

Nummulitic Limestones. The Nummulitic Limestones are bestexposed along the Zanskar Gorge, 2–4 km south of Choksti(Fig. 4, section 1; Fig. 5). Benthic nummulitid foraminiferahave yielded an age of 54.9 Ma (Green et al. pers. comm). Theformation is approximately 180 m thick, and commences at itsbase with 2–3 m thick, coarse sandstones and mudstonesinterbedded with approximately 1 m thick nummulitid-richpackstones. Many of the limestone units are sharp-based withlocally developed pebble layers; pebbles comprise micriticlimestone, quartz vein, schists and green mudstones, with nogranodiorite. Intervals of siltstone and fine sandstone showclimbing ripple lamination and low-angle cross-bedding.Higher up the section, packstones dominate and contain amore varied fauna with small isolated corals, oysters and otherbivalves. Details of the foraminifera present in the NummuliticLimestones can be found in Van Haver (1984).

The Nummulitic Limestones abruptly overlie floodplainmudstones of the upper Chogdo Formation (Searle et al. 1990;Fig. 5), and are here interpreted to represent a transgressioninto an inner, carbonate ramp setting; i.e., they accumulated atapproximately mean fair-weather wave-base (Burchette &Wright 1992). The packstones suggest wave reworking, butinsufficient to completely remove the lime mud component.The sharp, pebbly bases to some units suggest episodic storminfluence, and the sandstone intervals may record sand-richshoals or even shoreface deposits similar to sandstone unitswhich intercalate with the time equivalent NummuliticLimestones of the Alpine periphery (Sinclair et al. 1998).

Nurla Formation. The Nurla Formation has been documentedin the Zanskar Gorge (Fig. 6a) and between Shang andMartselang (Fig. 6b). The succession in the Zanskar Gorge islocated immediately north of the Choksti Conglomerate, i.e.,stratigraphically beneath it (Fig. 4a). It is also found further tothe north in a separate thrust sheet.

Overall, the total thickness for the Nurla Formation is atleast 700 m, comprising alternations of coarse sandstone unitsfrom 10 to 20 m thick, with thin to medium-bedded siltstones,fine sandstones and mudstones forming units from 8 to 20 mthick; the overall sandstone/mudstone ratio is approximately60%. The sandstones are sharp-based, with pebbly to con-glomeratic basal deposits; the basal surfaces locally show a10–20� disconformable relationship with the underlyingsandstone/mudstone alternations. Most units show somedegree of upward fining, and commonly terminate abruptlyupwards into the siltstones and mudstones. Bedding geo-metries are variable, with tabular to lensoid beds showinglow-angle cross-cutting relationships. The sandstone units con-tain planar and trough cross-stratification with up to 1 m thickcross-sets. Locally, the sandstone units are interbedded with0.5–1 m thick conglomerates. The intervening mudstones andsandstones vary in colour from red to green to dark grey.Siltstone beds range from 2 to 5 cm, and are normally graded.Remains of leaves and small wood fragments are common.Some of the red mudstones are mottled, with the localizeddevelopment of horizons rich in small (<2 cm) irregular, cream

Fig. 5. Sedimentary section through the succession immediatelyunderlying the Indus Group, measured south of Sumda-Do, ZanskarGorge (Fig. 3). This succession records fore-arc sedimentation priorto closure of the Tethys Ocean. The Nummulitic Limestones(uppermost Palaeocene; Van Haver 1984) record the lastunambiguous evidence of marine conditions, and have been used asan age constraint on continental collision (Searle et al 1987).


coloured, calcareous nodules. The outcrops north ofSumda-Do (Fig. 6) comprise thinner sandstone units (1–3 m)with extensive planar cross-bedding (30 cm cosets). Con-glomerates are also better developed (up to 5 m thick), com-prising well-rounded, clast-supported conglomerates withclasts up to 15 cm across. Clasts are dominated by meta-sediments including pelites, grey micrites and chert pebbles,approximately 20–30% of the clasts are granodioritic. VanHaver (1984) also identified fragments of peridotite which mayhave been derived from the ophiolitic suture zone or theSpontang Ophiolite (Searle et al. 1997). The conglomeratesshow low-angle cross-bedding. Palaeocurrent measurementsfrom the Nurla Formation indicate a variable southwardto south-eastward directed palaeoflow (Fig. 7a), with anadditional north-directed component (Fig. 7b).

This succession is interpreted as the aggradational depositsof meandering and braided, sandy and pebbly fluvial channelsover floodplain muds and silts. Rapid channel avulsion pro-cesses led to the abrupt transitions from channel sandstonesand conglomerates to floodplain fines. The thinner beddedsandstones are considered overbank flood deposits. The low-angle discordancies along bedding surfaces record broaddowncutting into underlying floodplain mudstones.

In the lower parts of the Nurla Formation near the ChokstiBridge along the Zanskar Gorge (Fig. 4a), is a succession ofstrata that are significantly different from the bulk of the NurlaFormation. It is at least 100 m thick, and comprises thin tothick-bedded (<1.5 m), medium to coarse sandstones, inter-bedded with dark grey siltstones and mudstones. The sand-stones are tabular bedded, sharp-based with locally developedload structures. The beds are non-graded to normally graded,with ripple laminated tops. The siltstones and mudstones arefinely laminated to bedded on a 1–15 mm scale, with ripple andplanar laminae. These strata are interpreted as the result ofepisodic deposition by turbidity currents in a sub-wave baseenvironment.

Choksti Conglomerate Formation. The Choksti ConglomerateFormation is exposed immediately south of Choksti villagealong the Zanskar Gorge (Fig. 4a). It is at least 300 m thick,with the lower 100 m of the succession being dominated bymedium to thin-bedded laterally variable, fine to mediumgrained sandstones. Palaeocurrents were predominantly to theNE to east (Fig. 7c). The upper 200 m of the formation isdominated by amalgamated, thick-bedded (up to 8 m) con-glomerates (Fig. 8). The conglomerates are dominantly, clast-supported with a very poorly sorted, very coarse sandstonematrix. Sandstone beds within the conglomerates are locallyplanar cross-bedded. Larger clasts (up to 30 cm) are rounded,with smaller clasts being more angular. Clast types higher inthe succession are dominated by red and green metasedimentswith approximately 10% volcanics and 10% granodiorite. Thegranodiorite content can get up to 60% lower in the succession.The overall succession is divided by two red mudstone units,approximately 4 m thick.

Conglomerate deposition occurred during ephemeral, highbedload and suspended load transport resulting in a-axisimbrication of pebbles; a fabric most typically found indeep-water conglomerates (Harms et al. 1982). However, theintercalated red mudstones suggest deposition in a continentalenvironment; an ephemeral, sheetflood dominated alluvial fan(Blair & McPherson 1994), or a braided fluvial system arethought the most likely candidates for the environment ofdeposition.

Fig. 6. Sedimentary sections through the Nurla Formation: (a)exposed along Zanskar Gorge (Fig. 4a), and (b) exposed in valleybetween Shang and Martselang (Fig. 4c). These sections record theintercalation of thick sandstone units with red, green and greymudstones and siltstones. The successions are interpreted asthe deposits of aggrading fluvial systems with meandering andbraided channels cutting through fine-grained floodplain deposits.

Fig. 7. Palaeocurrent roses for data from the Indus Group. Theplots are binned at 10� intervals, and the number within eachinterval increase linearly from the origin. All data have beencorrected for structural dip. (a) Dip of foresets from planarcross-bedding in the Nurla Formation, south of Stok Palace.(b) Plan-view ripple forms and planar cross stratification data fromthe Nurla Formation, Zanskar Gorge. (c) a-axis pebble imbricationdata from the Choksti Conglomerate Formation, Zanskar Gorge(only pebbles with all axes well exposed were measured). (d) Planarcross-stratification and a-axis pebble imbrication data from theHemis Conglomerate Formation, Zanskar Gorge. (e) Flute andgroove casts from the Nimu Formation, Zanskar Gorge. (f) Fluteand groove casts from Nimu Formation, Saspol.


Hemis Conglomerate Formation (late Eocene/early Oligocene;Van Haver 1984). The Hemis Conglomerate Formation is bestdeveloped in the east of the study area (Van Haver 1984;Searle et al. 1990), and were recorded in this study north ofMartselang (Fig. 4c) and north of Hemis. Their maximumthickness is approximately 1500 m in the region of Hemis, withonly a thin remnant (150 m) preserved 50 km west in theZanskar Gorge. This formation is characterized by laterallycontinuous (over several kilometres), 20–50 m thick con-glomerate and sandstone units interbedded with red mud-stones. The conglomerates are well-bedded, with a clastsupported texture, commonly imbricated, with well roundedclasts up to 30 cm. Clast lithologies are dominated (up to 90%)by granodiorites, acidic lavas and tuffs which are thought tohave been derived from the Ladakh Batholith (Van Haver1984), along with a few metasediments. Locally interbeddedcoarse sandstones show 75 cm thick, trough cross-sets. Overall,the conglomeratic units show a slight fining and thinning up,and their bases range from abrupt to gradual. The interveningred mudstones can locally develop up to 40 m thick, andcontain thin (1–5 cm), laminated, fining-upward siltstone beds,with some burrowing and wood fragments scattered alongbedding planes. Locally developed are green, branching net-works believed to represent rootlets. No palaeocurrentmeasurements were obtained from this formation. This succes-sion is interpreted as the intercalation of high-energy (possiblyephemeral) fluvial channels and floodplain mudstones in theeast of the area, thinning westward.

In the westernmost exposures of these conglomerateslocated 1.5 km south of the confluence of the Indus andZanskar rivers (Figs 4a & 9), the conglomerates have adifferent sedimentary character, and have previously been

interpreted as the thrusted repetition of the Choksti Con-glomerate Formation (Searle et al. 1990). However, theirstratigraphic position at the base of the Nimu Formation, andtheir granodiorite-clast-rich nature links them more closelyto the Hemis Conglomerate Formation, and here, they areinterpreted as along strike equivalents.

Near the base of the conglomerates in the Zanskar Gorge isa 20 m thick, clast-supported conglomerate dominated bygranodioritic boulders and mudstone clasts (Fig. 9). The baseof this conglomerate cuts down at least 10 m into the under-lying coarse sandstones. The upper part of the conglomerateunit fines upward into very poorly sorted, coarse sandstoneswith isolated, scattered boulders. The upper part of the forma-tion comprises 1–10 m thick, amalgamated, medium to coarsesandstones rich in mudstone clasts up to 3 m across. Beds arenormal to non-graded, and are laterally variable in thickness,and are sharp based, with extensive downcutting. Cross bed-ding is rarely developed, with cross-sets up to 20 cm thick.Locally developed are mudstone beds 0.1–4 m thick. Palaeo-current data indicate a predominantly north to north-northeastwards directed flow, with a minor southwestwardcomponent (Fig. 7). This succession is conformably over-lain by dark grey mudstones and sandstones of the NimuFormation.

This part of the Hemis Conglomerate Formation is inter-preted as the deposits of amalgamated, very coarse, high-concentration turbidity currents with a conglomeratic bedload(Lowe 1982). The extensive downcutting, and abundance of

Fig. 8. Sedimentary section through the Choksti Conglomeratemeasured from the Zanskar Gorge, opposite Choksti village.

Fig. 9. (a) Sedimentary section through the Hemis Conglomerateand Nimu formations measured along the Zanskar Gorge.(b) Close-up of sandstone/mudstone alternations from the NimuFormation interpreted as a succession of turbidites. Palaeocurrentsmeasured from cross-bedding are shown by the arrows with arrowedbases, data from flutes shown with dots at base; these data areplotted in Fig. 7.


mudstone clasts and rafts suggests that these beds accumulatedin channels cutting into a mud-rich, northward dipping,sub-aqueous slope. It is thought that these deposits recordsedimentation on a northward facing, fan-delta slope shedfrom coarse fluvial systems similar to those recorded in theHemis Conglomerate Formation; examples of fan delta-fedturbidite systems are summarized by Massari & Colella (1988),classifying them as slope-type fan-deltas.

Nimu Formation. The Nimu Formation forms the firstoutcrops found on the southern bank of the modern IndusRiver between the village of Saspol in the west, eastward toMartselang (Fig. 3). In this study, the Nimu Formation hasbeen studied from east to west at Stok, Pic Nimu, the ZanskarGorge and Saspol (Fig. 3). The maximum observed thicknessof the Nimu Formation was 600 m at Pic Nimu, although it islikely that the total stratigraphic thickness is considerablygreater.

Near Stok Palace, Stok, the Nimu Formation comprises0.04–1.5 m thick, coarse to fine sandstones intercalated withred mudstones and thin siltstones. The sandstones show vari-able grading, with planar, rippled and undulatory lamination.Abrupt basal surfaces locally show bottom structures such asgrooves. The siltstones are intensely rippled, with beddingplanes showing original bedforms with straight crests, andsymmetrical and trochoidal cross-sectional geometries;features typical of wave generated ripples (Allen 1982).Imprints of small bivalves are common on bedding surfaces;Van Haver (1984) identified them as a freshwater bivalve of thegenus Unio. Burrowing is common in the finer deposits of thissuccession. At Pic Nimu, the facies are similar, but sandstonebeds are thicker (up to 2 m), and occur as sandstone domi-nated packages ranging from 100 to 200 m thick; similarly, theintervening mudstones are up to 100 m thick. The succession atStok and Nimu suggest episodic sand accumulation in ashallow, sub-aqueous environment; it has been suggested thatthe system was deltaic (Van Haver 1984).

In the Zanskar Gorge and at Saspol, the Nimu Formationcomprises a different suite of facies. However, whether they aretime equivalent to the deposits of Pic Nimu and Stok (Searleet al. 1990) is questionable. The structural cross-section fromthe Zanskar Gorge suggests that the Nimu Formation south ofthe confluence has been structurally overthrusted onto themore northerly Nimu Formation (Fig. 4a) which is traceable2 km east to Pic Nimu. It is therefore proposed that theexposures at Pic Nimu are younger than those in the ZanskarGorge. In the Zanskar Gorge, the facies comprise sandstone/mudstone alternations (Figs 9 & 10) with coarse sandstonebeds up to 2 m thick, most showing normal grading, sharpbases and more gradually fining tops. Flutes and grooves arecommon on the base of beds, indicating palaeo-flow to thenorth (Figs 7e & 9b). Mudstone clasts up to 1 m across arecommon in the thicker beds. Amalgamation of coarse sand-stone beds forms sandstone packages up to 12 m thick(Fig. 9b). Thinner-bedded medium sandstones are also sharp-based and normally graded, and contain extensive current-ripple lamination, again indicating northward palaeo-flow.The mudstones are dark grey, and finely laminated.

In the region of Saspol (Fig. 3), the facies are similar to thosein the Zanskar Gorge, but show an increase in sandstonethickness (up to 4 m), with more pebbly, and partially erosivebases. Planar cross-bedding is also more common. Palaeo-flowfrom grooves and flutes in the Saspol region was directedprimarily to the SE (Fig. 7f).

The Nimu Formation is deformed throughout the region.Their upper surface is either the present-day erosion surface, ora palaeo-erosion surface that is now blanketed by undeformedsands and gravels (Fig. 10) deposited by alluvial fans, andQuaternary lakes caused by the damming of the Indus valley(Burgisser et al. 1982).

Palaeo-environments of the Indus Group; evidence forthe palaeo-IndusThe intermontane basin (post-Nurla Formation) was infilledby sediments deposited in environments which alternatedbetween fluvial and (probably) lacustrine (Fig. 11). The NurlaFormation and Choksti Conglomerates record coarse-grained,braided and meandering channels interbedded with floodplain

Fig. 10. Field drawing of unconformity at top Nimu Formation,overlain by undeformed Quaternary alluvial sediments. Locationapproximately 500 m south of the confluence of the Zanskar andIndus rivers, southwest of Pic Nimu (Fig. 3).

Fig. 11. Summary of sedimentary succession exposed along theZanskar Gorge. Columns on right-hand side indicate formationnames and approximate palaeo-environmental interpretations. Thezig-zagged lines through the section indicate the location of majorthrust faults.


mudstones. There is no evidence of major fluvial incision, andthere is high preservation of floodplain muds, suggesting thatthe bulk of this succession is aggradational (Bridge & Leeder1979). The palaeo-currents during this period were highlyvariable, with a predominantly northeasterly flow in theChoksti Conglomerate; this is highly at variance with theWNW directed flow required to suggest these sediments recordthe palaeo-Indus.

The Nimu Formation and western outcrops of the HemisConglomerate record >300 m of sediment deposited domi-nantly by mass flows and turbidity currents, at water depthsbelow the influence of fluvial or wave activity, representing awater depth of >20 m, and possibly significantly more. It hasbeen suggested that these formations were deposited in a largelake (Van Haver 1984; Searle et al. 1990). Palaeocurrents inthese formations are again variable, with a northwardand south-eastward component (Fig. 7). The upper NimuFormation records a regression into shallow water environ-ments. As with the previous formations, the Nimu Formationdoes not appear to record the early development of theIndus River. The Hemis Conglomerate records high energyfluvial sedimentation. However, its local appearance andwestward termination preclude its interpretation as asustained, basin-parallel flowing fluvial system.

Overall, the Indus Group record sedimentation withinvariable palaeo-environments. Palaeocurrents are oblique orperpendicular to the modern structural strike, and to thelongitudinal axis of the palaeogeographically restored IndusBasin (Van Haver 1984; Garzanti & Van Haver 1988; Garzantiet al. 1996). The limited range in compositions found withinthe clasts of the conglomerates (Van Haver 1984), and theirdominantly local origin (from the Ladakh magmatic arc),favour sedimentation from internally drained, transversefluvial and fluvio-deltaic fan systems as suggested by thepalaeocurrent data.

None of the sedimentological data suggests that there wasany time period of sustained, fluvial discharge to the NNWthat might record sedimentation by a precursor fluvial systemto the modern Indus. Therefore, the Indus River is not thoughtto have initiated along the Indus suture zone until afterdeposition of the Indus Group.

Burial and exhumation of the Indus GroupMaximum burial temperatures for the Indus Group can beestimated from illite crystallinity values presented by VanHaver (1984); values measured range from an illite crystallinityindex of 4 in the Nummulitic Limestones to 9.8 in the NurlaFormation. These values are inversely proportional to depth(Zen & Thompson 1974), and hence the maximum burialdepth is recorded by the Nummulitic Limestones. Correlationsof illite crystallinity values with coal rank and vitrinitereflectance values (Zen & Thompson 1974; Cornford 1984)allow estimates of the maximum possible temperaturesreached by the Nummulitic Limestones of 280�C. Maximumpossible temperatures reached by the Nurla Formation (illitecrystallinity index=10) are 155�C. The difference between thesetwo values (125�C) translates to a differential burial of 2.5–5 km assuming a possible range of geothermal gradients from25–50�C km�1. This would allow for the estimated totalthickness for the succession of 3.5 km.

Apatite fission track thermochronology enables an approxi-mation of the time at which an apatite crystal passed through

the 110�C (�10�C) isotherm during cooling to the surface(Brown et al. 1994). By combining the age at which a unit wasat 110�10�C with its maximum burial temperature, andassuming a linear cooling path, it is possible to estimate thetime at which maximum burial took place. This is used toapproximate the timing of transition from sediment accumu-lation (burial) to erosionally driven exhumation within theIndus Basin.

Apatite fission track ages from the Indus Group of theLadakh area are currently limited to two samples collected byP. Clift, and processed at University College London by A.Carter (Clift, unpublished data). One sample from the NimuFormation near the village of Nimu which has a central age of14�3 Ma, and one from a boulder in the Choksti Con-glomerate from the Zanskar gorge with a central age of 12 Ma.Plotting these values on to a temperature/time plot (Fig. 12),and projecting the line of cooling up to the maximum burialtemperature for the Nurla Formation enables a broad timeinterval of 26–14 Ma in which cessation of sedimentation andinitiation of exhumation in the basin began (Fig. 12). Thiscorrelates well with the youngest preserved sediments found tothe west in the Kargil area which are early Miocene (16–25 Ma). Apatite fission track ages from the neighbouringLadakh Batholith immediately north of the Indus River(Fig. 3) record older dates than those found in the IndusGroup. Two central ages for the batholith at Upshi are 28�3and 29�2 Ma.

This information on the burial and exhumation of the IndusGroup suggests that sediment accumulation in the Indus Basinterminated in early Miocene times. The transition from fluvialaggradation to subsequent fluvial downcutting records base-level lowering most likely associated with structurally drivensurface uplift (Ouchi 1985). Long-term fluvial incision is thedriving force behind erosional denudation and the consequentexhumation of rock to the surface (Anderson 1994; Burbank

Fig. 12. Time temperature plot for the Indus Group and for theLadakh Batholith. Illite crystallinity data give estimates of maximumtemperatures experienced by the Indus Group. Since the values aregreater than the temperature of the apatite fission track partialannealing zone (120–60�C), the apatite fission track ages will havebeen reset from the source rock age during burial. The neighbouringLadakh Batholith has older apatite fission track ages indicatingslower cooling rates during exhumation. The more rapid cooling ofthe Indus Group is interpreted as a result of structurally induceduplift and erosion during Zanskar backthrusting.


et al. 1996). The average rate of cooling can be translated intomean incision rates assuming a geothermal gradient. EarlyMiocene geothermal gradients of 40�C km�1 have been calcu-lated in Kohistan to the west of this area using fission trackdata (Zeitler 1985); in this calculation a range of 30 to 50�Ckm�1 is used. The apatite fission track ages from the IndusGroup translate to cooling rates of 5 to 11�C Ma�1 whichequate to long-term incision/denudation rates of 0.1 to 0.4 mma�1. These rates are slow compared to values from the Indusvalley around the Nanga Parbat massif of up to 10 mm a�1

based on apatite fission track ages (Burbank et al. 1996).The cooling rates from the neighbouring Ladakh Batholith

were slower, ranging from 3.2 to 4.8�C Ma�1 which equate todenudation rates of 0.06–0.16 mm a�1. The contrast in denu-dation rates from the lower values in the Ladakh Batholith tothe higher values in the Indus Group is interpreted to resultfrom the northward thrusting of the Zanskar and Indussedimentary rocks against the undeformed buttress of thebatholith, i.e., structurally driven surface uplift and erosion.

Initiation of the palaeo-Indus and regional upliftThe data presented comprise a sedimentary succession whichrecords accumulation in an internally drained intermontanebasin (Fig. 13a) rather than evidence of a major westward-flowing river system from Eocene to early Miocene times, anda cooling history which reveals relatively slow denudation inthe area since then. Hence, it is proposed that the earliestpossible time for initiation of a palaeo-Indus river in this areawas early Miocene (i.e. <26 Ma) as opposed to 45–55 Masuggested by Searle (1996), Qayyum et al. (1997) and Cliftet al. (2000).

The termination of sedimentation along the Indus suturezone, and the onset of denudation in early Miocene timesrelates to the larger scale development of the Himalayas at thattime. The Indus valley is bordered to the SW by the Zanskarrange and High Himalayas. Maximum P–T conditions, indica-tive of maximum crustal thickening in these areas, occurredduring Late Oligocene to Early Miocene times (Searle et al.1992). Comparisons of detrital zircon ages with their strati-graphic ages from molasse sedimentary rocks indicate regionalaccelerated unroofing of the Himalayas commencing in earlyMiocene times (Harrison et al. 1993), associated with initiationof the Main Central Thrust (Harrison et al. 1998). Similarresults have been interpreted from sandstone provenancestudies on the Murree and Siwalik supergroups recording theinfill of the foreland basin (Parkash et al. 1980; Critelli &Garzanti 1994).

The combination of maximum burial ages, and acceleratedunroofing of the Himalayas during early Miocene times sug-gests that this was the time at which the main Himalayan chainwas initiated as a zone of high elevation and relief (Searle1995). This period of thickening and uplift is likely to havebeen the mechanical force necessary to initiate northward-directed (back-) thrusting of the Zanskar and Indus sedimen-tary rocks resulting in termination of sedimentation in theIndus intermontane basin, and its subsequent erosionalunroofing (Fig. 13b). The differential uplift field as recordedin the cooling histories between the backthrust Zanskarand Indus sediments and the relatively undeformed LadakhBatholith provides a mechanism for the development of amajor longitudinal river system such as the Indus (Koons1995).

Evolution of the Indus River systemThe earliest evidence for some form of westward-flowingdrainage network in the western Himalayas comes fromthe late Eocene to early Miocene Khojak Formation in the

Fig. 13. Approximate palaeogeographic reconstructions along theIndus Suture Zone during Oligocene and Miocene times. (a) DuringOligocene times, the elevation of the present Tibetan Plateau wasbetween 1 and 3 km, with the incipient Himalayan mountain beltremained relatively low (Harrison et al. 1998). Additionalpiggy-back basins were located over this area, but are not shown atthis scale (Critelli & Garzanti 1994). Sedimentation of the IndusGroup along the Indus Suture Zone fluctuated between fluvial,deltaic and deeper water environments, with variable palaeocurrentdirections, and a predominantly local source (Van Haver 1984);these sediments do not represent sedimentation by anorth-northwestwardly through-flowing river that could beinterpreted as a palaeo-Indus River. (b). Accelerated regional upliftin Early Miocene times resulted in the termination of sedimentationalong the Indus Suture Zone, and the onset of fluvial incisiondriving denudation. Differential uplift between the activelybackthrusting Zanskar and Indus sedimentary rocks and theundeformed Ladakh Batholith provided a mechanism for theinitiation of a longitudinal river system. At a similar time, the IndusDelta and submarine fan prograded rapidly into its present position.It is possible that this may have been linked, and that it records thetiming of initiation of the main Indus River as a longitudinaldrainage network, but this remains speculative.


Katawaz Basin of Pakistan (Qayyum et al. 1997, in press).These sedimentary rocks record a delta and submarine fancomplex that accumulated in a remnant ocean basin on thewestern margin of the evolving collisional belt. Qayyum et al.(in press) provide sandstone provenance data that indicate thatthe Khojak Formation was derived from a collisional orogen,and not a magmatic arc. These data make it highly unlikelythat the Indus Group, which is rich in material derived fromthe Ladakh Batholith, could represent a proximal feedersystem for the Khojak Formation.

The present-day Indus floodplain of eastern Pakistan was ashallow seaway called the Kirthar Gulf in Oligocene to MiddleMiocene times (Kazmi 1984). The northern margin of theKirthar Gulf is recorded by the Bugti bone beds (Pascoe 1964),which contain a transition from coastal to marine deltaicenvironments in Early Burdigalian times, recording the initia-tion of southward progradation of the Indus Delta (Kazmi1984). Accelerated sedimentation in the present position of theIndus submarine fan appears to have commenced in LateOligocene/Early Miocene times (Whitmarsh et al. 1974; Kolla& Coumes 1987; Davies et al. 1995), although much of thedeep-sea drilling information is on the periphery of the mainIndus Fan, which is likely to have been initiated later thanmore proximal parts.

The earliest evidence for a palaeo-Indus river of comparabletectono-geomorphic status to the present-day river comes fromforeland basin deposits of the Siwalik Group (Abassi & Friend2000). The Janak Conglomerate on the western margins of thePotwar Plateau was deposited from 9 to 1 Ma, and recordssedimentation from a large braided river sourced in theHigh Himalaya of northern Pakistan, and interpreted as thepalaeo-Indus (Abassi & Friend 2000).

It is evident that the southward progradation of the IndusDelta and initial development of the Indus submarine fanstarted by at least end Oligocene/beginning Miocene times.Whilst there is evidence of sedimentation in an intermontanebasin (Indus Group) and a remnant ocean basin (KhojakFormation) prior to this, it would be incorrect to describe theseas the deposits of a palaeo-Indus River. These palaeogeo-graphic considerations support the notion that acceleratedunroofing of the Himalayas at end Oligocene/beginningMiocene times was associated with the expansion of a paleo-Indus delta/submarine fan system. At the same time, differ-ential uplift between the Zanskar and Indus sedimentary rocksand the Ladakh Batholith provided a mechanism for theinitiation of a major longitudinal river system along the routeof the modern Indus River (Fig. 13b). Whether the embryonicIndus delta was initiated by a river following this route isdifficult to assess, however, the earliest sedimentological recordfor a palaeo-Indus river of comparable tectono-geomorphicstatus to the modern system starts at 9 Ma.

Conclusions(1) The Indus Group of Ladakh, northwestern India, recordssedimentation from early Eocene times in an intermontanebasin. Sedimentation occurred in fluvial, shallow and deep-water environments. Palaeocurrent data through the IndusGroup are variable, recording flow to the south, SE, northand NNE. The environmental reconstructions, sedimentprovenance and palaeocurrent data suggest that the Indusintermontane basin was internally drained, and does notrepresent sedimentation by a through-flowing palaeo-IndusRiver shedding material west-northwestward.

(2) Maximum burial temperatures for the Indus Groupusing illite crystallinity ranged from 155 to 280 �C. Apatitefission track ages for the Indus Group range from 12 to 14 Ma.Combining these data with stratigraphic information fromthe along strike equivalent sediments of the Kargil areasuggests that the Indus Basin accumulated sediment until earlyMiocene times, when erosional unroofing was initiated. Time-averaged rates of unroofing since then have ranged from 0.1 to0.4 mm a�1; this is relatively fast compared to the unroofingof the neighbouring Ladakh Batholith which ranged from0.06–0.16 mm a�1.

(3) The early Miocene was a time of accelerated upliftthroughout the western Himalaya, and is coincident with thetime of initiation of the Indus Delta/submarine fan system. It isproposed that this is the earliest time at which a palaeo-IndusRiver of similar tectono-geomorphic status to the presentsystem could have been initiated. Differential uplift betweenthe Zanskar and Indus sedimentary rocks and the LadakhBatholith provide a mechanism for the initiation of a longi-tudinal river at this time following the route of the modernIndus. Although, whether this was the location of the maintrunk stream feeding the embryonic Indus delta remainsspeculative.

M. P. Searle first introduced us to the area and helped enormouslywith logistics and discussions of the geological setting for which areextremely grateful. T. Crosby, V. Sinclair and F. Hussein are thankedfor assistance in the field. P. Clift is thanked for the provision ofunpublished data. The paper was written whilst on sabbatical atPennsylvania State University, where much logistical support andscientific stimulus was given by D. W. Burbank and colleagues. H.D.S.was partly funded by a Birmingham University Science Faculty pilotresearch scheme. Reviews by E. Garzanti and P. Friend, plus editorialcomments by D. Pirrie are much appreciated. M. Qayyum is thankedfor providing pre-prints of forthcoming publications. Acknowledge-ment of the above colleagues does not necessarily mean they all agreewith the interpretations presented.

References

A, I.A. & F, P.F. 2000. Exotic conglomerates of the Neogene.Siwalik succession and their implications for the tectonic and topographic.evolution of the Western Himalaya. In: K, M.A., T, P.J.,S, M.P. & J, M.Q. (eds) Tectonics of the Nanga Parbat Syntaxisand the Western Himalaya. Geological Society, London, SpecialPublications, 170, 455–466.

A, J.R.L. 1982. Sedimentary Structures: Their Character and Physical Basis.Developments in Sedimentology, Vol. 1. Elsevier, Amsterdam.

A, R.S. 1994. Evolution of the Santa Cruz mountains, California,through tectonic growth and geomorphic decay. Journal of GeophysicalResearch, 99, B10, 20161–20179.

B, J.P., C, M., G, J., L, M., M, J. & M, G.1983. Geological studies in the Indus Suture Zone of Ladakh (Himalayas).In: G, Q.J. (ed.) Contributions to Himalayan Geology, 2, 1–30.

B, T.C. & MP, J.G. 1994. Alluvial fans and their natural distinctionfrom rivers based on morphology, hydraulic processes, sedimentaryprocesses, and facies assemblage. Journal of Sedimentary Research, A64,450–489.

B, J.S. & L, M. 1979. A simulation model of alluvial stratigraphy.Sedimentology, 26, 617–644.

B, M.E. & A-S, C.P. 1984a. Sedimentology, petrographyand tectonic significance of the shelf, flysch and molasse clastic depositsacross the Indus suture zone. Sedimentary Geology, 40, 287–291.

B, M.E. & A-S, C.P. 1984b. Sedimentation in a highaltitude intermontane basin—the Wakka Chu Molasse (Mid-Tertiary,Northwestern India). Bulletin of the Indian Geological Association, 17,175–193.

B, R.W., S, M.A. & G, A.J.W. 1994. Apatite fissiontrack analysis: its potential for the estimation of denudation rates and its


implications for models of long-term landscape development. In: K,M.J. (ed.) Process Models and Theoretical Geomorphology. John Wiley andSons Ltd, New York, 23–53.

B, D.W., L, J., F, E., A, R.S., B, N.,R, M.R. & D, C. 1996. Bedrock incision, rock uplift andthreshold hillslopes in the northwestern Himalayas. Nature, 379, 505–510.

B, T.P. & W, V.P. 1992. Carbonate ramp depositional systems.Sedimentary Geology, 79, 3–57.

B, H.M., G, A. & P, J. 1982. Late Glacial lake sedimentsof the Indus valley area, northwestern Himalayas. Eclogae GeologicaeHelvetiae, 75, 51–63.

C, P., S, N., L, G., G, C., S, H.-U., C, M.& A, S. 2000. Fifty-five million years of Tibetan evolution recorded inthe Indus Fan. EOS, Transactions, American Geophysical Union, 81,277–281.

C, C. 1984. Source rocks and hydrocarbons of the North Sea. In:G, K. (ed.) Introduction to the Petroleum Geology of the North Sea,171–204.

C, S. & G, E. 1994. Provenance of the Lower Tertiary Murreeredbeds (Hazara-Kashmir Syntaxis, Pakistan) and initial rising of theHimalayas. Sedimentary Geology, 89, 265–284.

D, T.A., K, R.B. & R, A.T.S. 1995. A time-slice approach to thehistory of Cenozoic sedimentation in the Indian Ocean. SedimentaryGeology, 96, 157–179.

F, P.F. 1998. General form and age of the denudation system of theHimalaya. GFF, 120, 231–236.

G, A. 1964. Geology of the Himalayas. J. Wiley Interscience, New York.G, E. & V H, T. 1988. The Indus clastics: forearc basin

sedimentation in the Ladakh Himalaya (India). Sedimentary Geology, 59,237–249.

G, E., C, S. & I, R.V. 1996. Paleogeographic andpaleotectonic evolution of the Himalayan Range as reflected by detritalmodes of Tertiary sandstones and modern sands (Indus transect, India andPakistan). Geological Society of America Bulletin, 108, 631–642.

H, J.C., S, J.B. & W, R.G. 1982. Structures and sequences inclastic rocks. Society of Economic Paleontologists and Mineralogists ShortCourse lecture notes, no. 9.

H, T.M., C, P., H, S.A., B, S., O, T.P. &K, W.S.F. 1993. Isotopic preservation of Himalayan/Tibetan uplift,denudation, and climatic histories of two molasse deposits. Journal ofGeology, 101, 157–175.

H, T.M., Y, A. & R, F.J. 1998. Orographic evolution of theHimalaya and Tibetan Plateau. In: C, T.J. & B, K.C. (eds)Tectonic Boundary Conditions for Climate Reconstructions. OxfordMonographs on Geology and Geophysics, 39, 39–72.

H, N. 1996. Regular spacing of drainage outlets from linear mountainbelts. Basin Research, 8, 29–44.

H, N. 1998. Controls on sediment supply by large rivers. In: S, K.& MC, P. (eds) Relative Role of Eustasy, Climate, and Tectonism inContinental Rocks. Society of Economic Paleontologists and MineralogistsSpecial Publications, 59, 3–16.

K, A.H. 1984. Geology of the Indus Delta. In: H, B.U. & M, J.D.(eds) Marine Geology and Oceanography of the Arabian Sea and CoastalPakistan. Van Nostrand Reinhold, New York, 71–84.

K, V. & C, F. 1987. Morphology, internal structure, seismic strati-graphy, and sedimentation of the Indus Fan. American Association ofPetroleum Geologists Bulletin, 71, 650–677.

K, P.O. 1994. Three dimensional critical wedges: Tectonics and topographyin oblique collisional wedges. Journal of Geophysical Research, 99,12301–12315.

K, P.O. 1995. Modelling the topographic evolution of mountain belts.Annual Review of Earth and Planetary Sciences, 23, 375–408.

L, D.R. 1982. Sediment gravity flows: II. Depositional models withspecific reference to deposits of high-density turbidity currents. Journal ofSedimentary Petrology, 52, 279–297.

M, F. & C, A. 1988. Evolution and types of fan delta systems insome major tectonic settings. In: N, W. & S, R.J. (eds) Fan Deltas:Sedimentology and tectonic settings. Blackie and Son, London, 103–122.

M, J.D. & M, R.H. 1983. World-wide delivery of river sediment tothe oceans. Journal of Geology, 91, 1–21.

O, S. 1985. Response of alluvial rivers to slow active tectonic movement.Geological Society of America Bulletin, 96, 504–515.

P, B., S, R.P. & R, A.K. 1980. The Siwalik Group (Molasse) –sediments shed by collision of continental plates. Sedimentary Geology, 25,127–159.

P, E.H. 1964. A Manual of the Geology of India and Burma, Vol. III.Government of India Press, Calcutta.

Q, M., L, R.D. & N, A.R. 1997. Discovery of the palaeo-indus delta-fan complex. Journal of the Geological Society, London, 154,753–756.

Q, M., N, A.R. & L, R.D. 2000. Detrital modes andprovenance of the Palaeogene Khojak Formation in Pakistan: Implicationson early Himalayan orogeny and unroofing. Geological Society of AmericaBulletin, in press.

S, M.P. 1995. The timing of metamorphism, magmatism, and cooling inthe Zanskar, Garhwal, and Nepal Himalaya. Journal of Nepal GeologicalSociety, 11, 103–120.

S, M.P. 1996. Geological evidence against large-scale pre-Holocene offsetsalong the Karakorum fault: Implications for the limited extrusion of theTibetan Plateau. Tectonics, 15, 171–186.

S, M.P., C, R.I., S, B. & MC, J. 1997. Structureof the North Indian continental margin in the Ldakh-Zanskar Himalayas:implications for the timing of obduction of the Spontang ophiolite,India-Asia collision and deformation events in the Himalaya. GeologicalMagazine, 134, 297–316.

S, M.P., P, K.T. & C, D.J.W. 1990. Restoration andevolution of the intermontane Indus molasse basin, Ladakh Himalaya,India. Tectonophysics, 174, 301–314.

S, M.P., W, D.J., R, D.C. & W, R.N. 1992. Pressure,temperature and time constraints on Himalayan metamorphism fromeastern Kashmir and western Zanskar. Journal of the Geological Society,London, 149, 753–773.

S, M.P., W, B.F., C, M.P., C, D.J.W., R, A.D., R,D., T, L., X, X., J, M.Q., T, V. & K, S.1987. The closing of Tethys and the tectonics of the Himalayas. GeologicalSociety of America Bulletin, 98, 678–701.

S, L., A, J.G., M, A.S. & K, W.S.F. 1998. Feedbackbetween uplift, denudation, and tectonics at Nanga Parbat, westernHimalayas. EOS Transactions, American Geophysical Union supplement:1998 Fall Meeting, F 909.

S, H.D., S, Z.R. & T, M.E. 1998. Carbonate sedimentation atthe onset of foreland basin subsidence: The Eocene of the French Alps. In:W, P. & B, T. (eds) Carbonate Ramps. Geological Society,London, Special Publications, 149, 205–227.

T, V.C. 1981. Regional framework and geodynamic evolution of theIndus-Tsangpo suture zone in the Ladakh Himalayas. Transactions of theRoyal Society of Edinburgh: Earth Sciences, 72, 89–97.

T, V.C. & M, D.K. 1984. Tectonic framework of the Indus and Shyoksuture zones in eastern Ladakh, northwest Himalaya. Tectonophysics, 101,207–220.

V H, T. 1984. Etude stratigraphique sedimentologique et structural d’unbassin d’avant arc: exemple du bassin de l’Indus, Ladakh, Himalaya. PhDthesis, University of Grenoble, France.

W, R.B., W, O.E. & A, S. et al. 1974. Initial Reports of the DeepSea Drilling Project, Site 221, Vol 23. US Government Printing Office,Washington, 167–184.

Z, P.K. 1985. Cooling history of the NW Himalaya, Pakistan. Tectonics, 4,127–151.

Z, P.K. & K, P.O. 1998. Nanga Parbat as tectonic aneurysm: ametamorphic signature of indentor-corner dynamics. EOS Transactions,American Geophysical Union supplement: 1998 Fall Meeting, F 910.

Z, E-A. & T, A.B. 1974. Low grade regional metamorphism:Mineral equilibrium relations. Annual Review of Earth and PlanetarySciences, 2, 179–212.

Received 6 August 1999; revised typescript accepted 26 July 2000.Scientific editing by Duncan Pirrie.


Sedimentology of the Indus Group, Ladakh, northern India ... · PDF fileSedimentology of the...

Documents

Transcript of Sedimentology of the Indus Group, Ladakh, northern India ... · PDF fileSedimentology of the...