Exhumation of the ultrahigh-pressure Tso Morari unit in eastern

Ladakh (NW Himalaya): A case study

Julia de Sigoyer

Laboratoire de Geologie, CNRS-UMR 8538, Ecole Normale Superieure de Paris, Paris, France

Stephane Guillot

Laboratoire de Sciences de la Terre, CNRS-UMR 5570, Universite Lyon I et Ecole Normale Superieure de Lyon,Villeurbanne, France

Pierre Dick

Institut de Geologie, Universite de Neuchatel, Neuchatel, Switzerland

Received 12 December 2002; revised 6 February 2004; accepted 18 February 2004; published 13 May 2004.

[1] Exhumation processes of the ultrahigh-pressure(UHP) Tso Morari dome (NW Himalaya) areinvestigated using structural, petrological, andgeochronological data. The UHP Tso Morari unit isbounded by the low-grade metamorphic Indus SutureZone to the NE and Mata unit to the SW. Threedeformation phases (D1, D2, and D3) are observed.Only D3 is common to the UHP unit andsurrounding units. In the UHP unit, the firstdeformation phase (D1) produced upright folds,under eclogitic conditions (>20 kbar; 580 ± 60�C).D1 is overprinted by D2 structures related to aNW-SE trending open anticline. D2 is characterizedby blueschist mineral associations, and correspondsto the quasi-isothermal decompression from a depthof 90 km up to 30–40 km. The final exhumationphase of the Tso Morari unit is dominated by tectonicdenudation and erosion (D3), associated with a slighttemperature increase. Radiochronological analysesindicate that the UHP exhumation process beganduring the Eocene. Exhumation was fast duringD1-D2 and slowed down through D3 during theOligocene. The change in the deformation style fromD1-D2 to D3 in the Tso Morari unit coincides withchanges in the exhumation rates and in themetamorphic conditions. These changes may reflectthe transition from an exhumation along thesubduction plane in a serpentinized wedge, to thevertical uplift of the Tso Morari unit across the uppercrust. INDEX TERMS: 8110 Tectonophysics: Continental

tectonics—general (0905); 3660 Mineralogy and Petrology:

Metamorphic petrology; 8015 Structural Geology: Local crustal

structure; 9320 Information Related to Geographic Region:

Asia; KEYWORDS: exhumation processes, ultrahigh-pressure

metamorphism, Himalaya, horizontal shortening, folding,

India-Asia convergence. Citation: de Sigoyer, J., S. Guillot,

and P. Dick (2004), Exhumation of the ultrahigh-pressure Tso

Morari unit in eastern Ladakh (NW Himalaya): A case study,

Tectonics, 23, TC3003, doi:10.1029/2002TC001492.

1. Introduction

[2] High to ultrahigh-pressure metamorphic rocks ofcontinental or oceanic origins are always found in conver-gent zones [Ernst and Liou, 1999]. Some of these eclogiticmetamorphic rocks formed under high-pressure and low-temperature conditions indicate that they have been buriedin a subduction zone context [Platt, 1993]. The subsequentreturn of these well-preserved eclogitic rocks to the Earth’ssurface often implies a rapid exhumation [e.g., Duchene etal., 1997]. However, as they display different tectono-metamorphic evolutions [Cloos, 1982; Spalla et al.,1996], the exhumation processes still remain a matter ofdebate where various partly contradictory models exist:coaxial extension, associated with a detachment fault[Ruppel et al., 1988; Jolivet et al., 1996]; extensionalcollapse [Dewey et al., 1993]; thrusting toward the foreland[Argand, 1916; Steck et al., 1998]; buoyancy forces assistedby erosion and tectonic processes [Chemenda et al., 1996];corner flow [Platt, 1993; Allemand and Lardeaux, 1997], orchannel flow [Cloos, 1982; Guillot et al., 2000, 2001];exhumation by extrusion within a soft zone of deformationcompressed between two rigid blocks [Thompson et al.,1997a].[3] Each exhumation model predicts the nature of the

contact between the high-pressure (HP) or ultrahigh-pres-sure (UHP) unit and surrounding lower-grade rocks, and thekinematics of penetrative structures. Detailed structural,petrological, and geochronological analyses of HP to UHProcks thus appear necessary to precise the exhumationprocesses of such rocks.[4] The outstanding preservation of petrologic and

structural features in and around the coesite bearingeclogitic Tso Morari massif [Sachan et al., 2001], makesthis area an ideal zone to study the exhumation processesof UHP rocks.

TECTONICS, VOL. 23, TC3003, doi:10.1029/2002TC001492, 2004

Copyright 2004 by the American Geophysical Union.0278-7407/04/2002TC001492$12.00

TC3003 1 of 18

[5] The petrological and geochronological results havepreviously been published on the Tso Morari massif [deSigoyer et al., 1997, 2000; Guillot et al., 1997; O’Brien etal., 2001; Sachan et al., 2001]. In this paper we will focuson the structural analysis of the UHP unit and its surround-ing areas.[6] The synthesis of structural, petrological and geochro-

nological studies will lead to a discussion on the exhuma-tion processes of UHP rocks, which will be used as a basisfor future modeling.

2. Geological Transect

[7] The investigated area is located in eastern Ladakh(northwest India), in the internal part of the Himalayan belt(Figure 1). It spreads from the Ladakh batholith to the northto the Indian continental margin to the south, between78�250E and 77�500E of longitude and 34�N and 32�50Nof latitude (Figure 2).[8] In this study we highlight a new northeast-southwest

geological transect from the Ladakh batholith to the Tethyansedimentary cover of the Indian margin (Figure 3). Alongthis cross section three domains are distinguished: (1) theLadakh Batholith and Indus Suture Zone, (2) the Tso Morariunit, and (3) the Mata-Karzog unit.

2.1. Ladakh Batholith and Indus Suture Zone

2.1.1. Ladakh Batholith[9] The Ladakh Batholith or Trans-Himalayan Batholith

is located between the Shyok Suture Zone to the north andthe Indus Suture Zone to the south. According to Bassoulletet al. [1983]; Reuber et al. [1987]; Maheo et al. [2000];Rolland et al. [2002] this calc-alkaline batholith corre-sponds to a volcanic arc, formed along the Asian marginduring the northward subduction of the Neo-Tethys oceanfrom Lower Cretaceous to early Eocene [Debon et al., 1986;Le Fort, 1989; Weinberg and Dunlap, 2000]. Before 75 Ma,the Ladakh arc was accreted onto the Asian margin creatinga thick crustal zone [Rolland et al., 2002].2.1.2. Indus Suture Zone[10] The Indus Suture Zone is composed of six units,

from north to south: (1) the Indus Sequence, (2) theNindam Flysch, (3) the Nidar Ophiolite, (4) the ShergolConglomerate, (5) the Drakkarpo unit, and (6) the Ribilunit.2.1.2.1. Indus Sequence[11] The Indus sedimentary sequence was deposited in an

episutural basin that evolved from a marine to a continentalenvironment and recorded the closure of marine domainbetween the two converging continents. The sequencebegins with Late Cretaceous detritic formations eroded fromthe Ladakh batholith [Van Haver, 1984; Mascle et al., 1986;Robertson, 2000]. They are followed by the Gongmaru-Laformation, composed of red deltaic sediments and nummu-litic lower Eocene limestones accumulated on the Asianmargin [Blondeau et al., 1986; Garzanti et al., 1987]. Themiddle to upper Eocene Choksti conglomerate withstretched granodioritic and andesitic pebbles overlies the

Gongmaru-La formation [Mascle et al., 1986]. Southward,red continental pelites alternating with paleosoils and greenconglomeratic sandstones lie in channels. They are typicalof the Oligocene Nurla formation related to the erosion ofthe Indian continental margin [Baud et al., 1982].2.1.2.2. Nindam Flysch[12] The thick (hectometric to kilometric) Cenomano-

Maastrichian Nindam Flysch consists of sandstones andpelitic rocks that derive from the volcanoclastic productsof the Ladakh magmatic arc [Bassoullet et al., 1983;Robertson, 2000].2.1.2.3. Nidar Ophiolite[13] The ophiolite is well preserved and mainly unmeta-

morphosed (only local occurrence of chlorite suggestshydrothermal metamorphism). The base of the Nidar Ophio-lite consists of 1000 m of well-preserved pillow lava andbasalt. A large dyke complex overlays the pillow lava. It ismade of microgabbros infilled by doleritic dykes. South-ward, the dyke complex is replaced by a 1-km wide level ofgabbros. Above the gabbros lies serpentinites, within whichone can distinguish cumulates of lower oceanic crust andmantle residue rocks [Guillot et al., 2000]. Geochemistry onthe metabasalts [Thakur and Bhat, 1983; de Sigoyer, 1998;Maheo et al., 2000] suggests an intraoceanic arc setting thatcorresponds to the eastward equivalent of the SpontangOphiolite (W Ladakh) (Figure 1).2.1.2.4. Shergol Conglomerates[14] The Shergol conglomerates thrusts over the Nidar

Ophiolite (Figures 2 and 3); they lie in between the NidarOphiolite and the Drakkarpo unit [Mascle et al., 1986]. Thesequence consists of coarse continental conglomerates withpebbles coming from the Nidar ophiolite, Drakkarpo, Ribiland Tso Morari unit. This formation corresponds to apostcollisional product of erosion from the internal Hima-layan orogen. It is not directly dated in Ladakh, but similarOligocene conglomerates located in an identical structuralposition were observed in south Tibet [Van Haver, 1984;Colchen et al., 1987].2.1.2.5. Drakkarpo Unit[15] The Drakkarpo unit (Figure 2) is 5 km wide. Its

base consists of a 2000-m thick polygenic conglomerate,its matrix is composed of schists, green sandstones, orcalcareous slates in which lenses of tuffs, basalts, serpen-tinites, quartzites, micaschists and radiolarites are ob-served. The holes and fractures of the tuffs are filledwith carbonates, chlorites and oxides suggesting hydro-thermal metamorphism. These volcanic rocks have alka-line affinities; their geochemical features suggest anoceanic island (OIB) origin [Fuchs and Linner, 1997;de Sigoyer, 1998]. Thick white limestone typical ofplatform facies environment, probably Permian in age[Colchen et al., 1987; Corfield et al., 1999] is embeddedin Upper Albian to Mid Cenomanien red sandstones[Fuchs and Linner, 1996]. The location of this unit,between the Nidar Ophiolite and the Indian margin,the succession of the different lithologies, and the geo-chemical data suggest that the Drakkarpo unit representsa remnant of former seamounts such as the Photangunit observed below the Spontang Ophiolite (Figure 1)

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

2 of 18

TC3003

[Colchen et al., 1987; Reuber et al., 1987; Corfield et al.,1999].2.1.2.6. Ribil Unit[16] Southward, the Drakkarpo unit thrusts over the Ribil

unit metamorphosed under greenschist facies conditions(Figures 2 and 3). The Ribil unit consists of agglomeraticslates, overlaid by reddish brown dolomite marbles, pyrox-enes bearing basalts, and vesicular basalts. Brachiopodfragments, from the Upper Paleozoic (Upper Carboniferousto Permian) where found in the marbles by Fuchs andLinner [1996]. A similar succession of rocks is described inthe basement of the Lamayuru formation [Colchen et al.,1994] suggesting that the Ribil unit may represent the distalpart of the Indian continental margin. However, geochem-

ical analyses carried out on the basalts [de Sigoyer, 1998]show an alkaline OIB (oceanic island basalt) origin, similarto those observed in the Drakkarpo unit. We consider thatthese two units are remnants of seamounts accreted on theIndian margin.

2.2. UHP Tso Morari Unit

[17] The Tso Morari unit is one of the westernmostNorth Himalayan gneissic domes, but unlike the otherdomes, it is metamorphosed under UHP conditions. TheTso Morari unit (100 km � 50 km) outcrops south of theRibil unit (Figure 4). It has an elongated shape strikingnorthwest-southeast (Figures 2 and 3) [Thakur, 1983].According to stratigraphic constraints, the Tso Morari unit

Figure 1. Geological map of the NW Himalaya, modified after Steck et al. [1998]. TM is the TsoMorari unit. Inset shows location of the study area.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

3 of 18

TC3003

represents a remnant-tilted block of the distal Indiancontinental margin [Colchen et al., 1994; Steck et al.,1998]. In its central part, orthogneiss are intrusive intomore or less deformed Cambro-Ordovician sediments[Trivedi et al., 1986; de Sigoyer, 1998, Girard and Bussy,1999]. Some metabasic rocks are infilled in the orthog-neiss and may be associated with the Ordovician magmaticevent, as they present continental affinities [de Sigoyer,1998]. Toward the rims of the dome, the orthogneiss isoverlaid by an Upper Carboniferous to Permian metasedi-mentary cover [Colchen et al., 1994; Fuchs and Linner,1996], which consists of metapelites, metagraywackes,metacarbonates, reddish Permian metadolomites andquartzites. Hectometric lenses of metabasic rocks areassociated with the Permian dolomitic limestones, theyshow continental tholeiitic affinities such as Panjal traps[de Sigoyer, 1998], and are related to the Carboniferous-Permian rifting of the Neo-Tethys [Bassoullet et al., 1983;Honegger et al., 1982; Spencer and Gebauer, 1996].[18] An undeformed granite outcrops at the Polokongka

La pass. According to geochemical data this granite formspart of the Cambro-Ordovician orthogneiss; Sm/Nd (onapatite, garnet and whole rock [de Sigoyer, 1998]), Rb/Sr

[Trivedi et al., 1986] and U-Pb (on zircon [Girard andBussy, 1999]) ages are respectively 458 ± 14 Ma, 487 ±25 Ma and 479 ± 2 Ma. Similar Ordovician granite andorthogneiss are observed in the Nyimaling area [Stutz andSteck, 1986], they present a Gondwana affinity as many ofthem developed along the Indian continental margin [LeFort et al., 1986].[19] Petrological observations and thermobarometrical

estimations carried out on the Tso Morari metapelitic andmetabasaltic rocks are detailed by Guillot et al. [1995], deSigoyer et al. [1997], Guillot et al. [1997], de Sigoyer[1998], O’Brien et al. [2001], and Sachan et al. [2001](Figures 5 and 6). Evidence of ultrahigh-pressure andrelatively low-temperature metamorphism (20–25 kbarand 580� ± 60�C) are deduced in the Tso Morari unit bythe occurrence of coesite, garnet, omphacite, phengite,glaucophane, and zoisite in the basic rocks [de Sigoyer etal., 1997; Sachan et al., 2001] and by the association ofjadeite, garnet, chloritoide, phengite in the metapelites[Guillot et al., 1997]. In the orthogneiss the UHP metamor-phic conditions are much more difficult to characterize asthe mineral assemblage presents a high variance. Thereforechanges in pressure and temperature conditions only modify

Figure 2. Geological map of the Tso Morari area based on satellite Spot images combined with our fieldobservations and previous studies [Berthelsen, 1953; Thakur, 1983; Fuchs and Linner, 1996; Steck et al.,1998]. AB is trace of profile shown in Figure 3.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

4 of 18

TC3003

the composition of the coexisting mineral phases. However,in the Tso Morari orthogneiss, the magmatic plagioclase andbiotite have reacted at the expense of Ca rich garnet(Grossular 52%), kyanite, phengite (with a Si4+ = 3.36)and zoisite, associated with rutile. This mineral assemblageis often described in other UHP orthogneiss as in the GranParadiso or Monte Rosa in the Alps [Le Goff and Ballevre,1990; Dal Piaz and Lombardo, 1986], and suggests that theorthogneiss underwent the same metamorphic conditionsthan the metasedimentary and metabasic rocks. Thesemetamorphic conditions reveal the subduction of the TsoMorari unit, and consequently of the Indian margin down toa minimum depth of 90 km (Figure 5). During its exhuma-tion up to 40–30 km depth the Tso Morari unit underwentisothermal decompression under blueschist facies condi-tions (11 ± 3 kbar; 580� ± 50�C) as shown by thecrystallization of secondary glaucophane, in metapeliteson the eastern part of the unit [de Sigoyer et al., 1997;Guillot et al., 1997]. In the western part of the unit, calcicamphibole (and not glaucophane) and garnet crystallized inbetween omphacite and garnet in the metabasic rocks [deSigoyer et al., 1997; O’Brien et al., 2001; Girard, 2001].Staurolite associated with chlorite and phengite in the S2plane reacted to give kyanite and biotite, in the metagray-wackes, such a reaction implies a temperature increases(Figures 5 and 7) [Guillot et al., 1997]. Thermobarometricalstudies carried out on these western rocks show a temper-ature increase up to 630� ± 50�C under amphibolitic facies

conditions at 30 km (9 ± 3 kbar) depth. In all the rockschlorite and white micas continued to crystallize in C3 shearbands under greenschist facies conditions during the end ofthe Tso Morari exhumation. The recrystallization of theeclogitic rocks into garnet bearing amphibolite and thenunder greenschist conditions is mainly observed in thesouthern and western part of the Tso Morari unit, while

Figure 3. NS cross section from the Ladakh batholith (A) to the Tethyan sediments (sedimentary coverof the Indian margin) (B); trace is given in Figure 2. Deformation phase D1 corresponds to obduction ofthe Nidar ophiolite obduction, and to the eclogitization of Tso Morari unit at the Paleocene-Eocenetransition. D2 corresponds to the early part of the exhumation of the Tso Morari unit, D3 to the later part.

Figure 4. Photograph of the Zildat normal shear zonewhich separates the slightly metamorphosed Ribil unit fromthe Tso Morari eclogitic dome. Serpentinite lenses underlinethis shear zone. This photograph also shows the thrustingcontact between the Drakkarpo unit and the Ribil unit.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

5 of 18

TC3003

Figure 5. Map of the studied area showing the metamorphic facies or mineral occurrences in themetabasalts and metapelites in the different units [de Sigoyer et al., 1997; Guillot et al., 1997]. The fresheclogitic rocks are mainly preserved in the central part of the Tso Morari unit. In the northeast border ofthe Tso Morari unit, blueschists mineralogical assemblages overprint the eclogitic paragenesis, whereaseclogites are overprinted by garnet bearing amphibolite paragenesis in the western part of the Tso Morariunit. Note also the metamorphic contrast between the eclogitic Tso Morari unit and the surrounding units,metamorphosed under lower grade (epidote amphibolite to greenschist metamorphic facies).

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

6 of 18

TC3003

they are retrogressed under blueschist and greenschist faciesconditions in the eastern part (Figures 5 and 6).

2.3. Mata-Karzog Unit

[20] South of the Tso Morari unit, some chromitic podsassociated with serpentinites and basaltic lava crop out inthe core of a synclinorium (Figure 2). As first suggested byBerthelsen [1953] and according to geochemical data theserocks represent a relict of an intraoceanic arc ophiolite[Maheo et al., 2000]. The Karzog Ophiolite may representthe southward continuity of the Nidar Ophiolite [Maheo etal., 2000]; (Figure 3). Under the Karzog Ophiolite lies theCarboniferous to Permian sedimentary sequence of Mata

[Berthelsen, 1953; Virdi et al., 1978]. South of the Karzogvillage, the Mata-Rupshu granite is observed in the core of arecumbent north verging anticline. Trivedi et al. [1986]proposed a Rb-Sr age of 487 ± 14 Ma for the Mata-Rupshugranite, confirmed by a U-Pb zircon age of 482.5 ± 1 Ma[Girard and Bussy, 1999]. In the upper part of the Matacrest (6275 m), doleritic sills intrude the granite. South ofthe Mata-Rupshu granite, a thick normal metasedimentarycover is developed starting with reddish brown dolomiticslates, overlaid by dark slates, which show alternation ofquartzitic and calcareous levels. In these latter levels, frag-ments of ammonites were found (G. Mascle, personalcommunication), and were described by Virdi et al. [1978]as Permian fossils.[21] In the Mata-Karzog unit no relics of eclogites were

found. Magmatic pyroxene relics are still observed in thebasic rocks located in the northern part of the unit, else-where these metabasic rocks have recrystallized into actin-olite, biotite, plagioclase, zoisite, chlorite, magnetite andcarbonate. Such a mineral association is typical for uppergreenschist to epidote amphibolite metamorphic facies(Figure 5) [de Sigoyer, 1998]. The lack of eclogitic rocksand the occurrence of magmatic relics in the basic lensesindicate that the Mata-Karzog unit has never undergone HPmetamorphism.[22] The major feature observed in the different units is

the combination of strongly contrasted metamorphic andstratigraphic domains. The unmetamorphosed Indus SutureZone (with Asian and oceanic affinities) is bounded by theUHP Tso Morari unit (with Indian affinities) which isbounded to the south with the weakly metamorphosed(upper greenschist conditions) Mata-Karzog unit (Figures 2,3, 5, and 6).

3. Deformation Pattern

[23] In order to understand the exhumation mechanismsof the UHP Tso Morari unit and the nature of the contactsbetween the three main domains, we compared the tectonicevolution of the Tso Morari unit with contiguous units. Thebulk finite strain pattern and the principal directions of finitedeformation in the three main domains are deduced from thestructural observations. The numbering of deformationalphases D1, D2 and D3 on the different units is proppingon the Tso Morari unit evolution. The geological descriptionof the studied area shows a general northwest-southeasttrending of the different units (Figure 8).

3.1. Structural Evolution of the Indus Suture Zone

3.1.1. Observations[24] The Indus Suture Zone is characterized by a fan-

shape geometry with northeast and southwest verging struc-tures, respectively north and south of this zone (Figures 3and 8).[25] 1. The Ladakh batholith is locally reworked by a

post magmatic N120�/40�S foliation plane, that bears aN165�/35� mineral lineation underlined by amphiboles. Tothe south the batholith is in stratigraphic contact with theIndus sequences, which are deformed by F1 south verging

Figure 6. P-T-t path showing the tectometamorphicevolution of the Tso Morari unit deduced from samplescoming from all over the unit [de Sigoyer et al., 2000;Sachan et al., 2001]. Boxes represent the differentmetamorphic stages. The ages of the Tso Morari evolutionwere obtained using different radiochronological systems onspecific paragenesis. The eclogitization is dated at about55 ± 7 Ma by Lu/Hf, Sm/Nd and U/Pb geochronology ongarnet-omphacite-whole rock, garnet-glaucophane-wholerock, and allanite, respectively. Minerals related to amphi-bolite metamorphic recrystallization have been dated atabout 47 ± 6 Ma) by Sm/Nd, Rb/Sr and 40Ar/39Argeochronology on garnet-Ca-amphibole-whole rock (in thebasic lens), phengite-apatite-whole rock, and phengite (onmetapelites), respectively. The end of the Tso Morariexhumation is dated by 40Ar/39Ar age at 29 ± 0.4 Ma onbiotites and muscovites which crystallized under greenschistconditions. Ve is estimated vertical exhumation ratesdeduced from ages and depth.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

7 of 18

TC3003

Figure

7

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

8 of 18

TC3003

folds. These F1 folds are strongly overprinted by hectometerto kilometer northeast verging F2 folds. The F2 folds aresynfoliation, they have a N120�/40�S axial plane and aN135�/35� axis.[26] To the south, the Indus Sequence is separated from

the Nindam Flysch (Figure 3) by a steep south vergingextensional fault (N070�/80�S) with an oblique N240�/50�striae suggesting a strike-slip component. This brittle faultreactivates an earlier northeastern verging thrust asdescribed by Van Haver [1984] in western Ladakh.[27] Locally, evidences of recent northwest-southeast

dextral strike-slip faults, parallel to the Karakorum faultare also observed in the Indus valley. They locally crosscutthe Quaternary alluvial terraces.[28] 2. The Nindam Flysch was first thrust over the Nidar

Ophiolite toward the southwest D1. Later, north verging F2folds, similar to those observed in the Indus Sequence,redeformed the Nindam Flysch, and produced a penetrativefoliation S2.[29] 3. The Nidar Ophiolite has been affected by a

penetrative planar fabric D2 oriented N150�/70�S, associ-ated to the formation of the large northeast verging foldswith N140� fold axis and by top to the northeast thrust. ThisD2 deformation partly overturned the southern part of theophiolite (Figure 3). At a regional scale, the Nidar Ophioliteshows a sigmoid shape, striking N120�, suggesting that ithas also recorded dextral strike slip movement compatiblewith the movement observed on the Karakorum fault to thenorth (Figure 1).[30] 4. The Oligocene Shergol conglomerates overthrust

the Nidar Ophiolite toward the northeast.[31] 5. The Drakkarpo unit (Figure 2) extends along a

N120�–140� direction and shows a double verging struc-ture (Figure 3). It overthrusts the Nidar Ophiolite toward thenortheast, and the Ribil unit toward the southwest. Somereverse faults striking N135 40�N are observed close to thesouthern boundary of the Drakkarpo unit. They havedeformed limestones and basalts and are overprinted byS2 foliation, which were probably active during D1.[32] In the Drakkarpo unit the alignment of white lime-

stones can be observed in the landscape and in Spot images.This alignment is parallel to the N120�–140� trendingdirection of the Drakkarpo unit, and to the F2 folds axis(La2) in the core of the unit (N317�/20�) (Figure 8). The F2

folds and the associated S2 foliation overprint a previous S1foliation. The L2 mineral lineation is oriented N120�/6� inthe southern part of the unit and N080�/36� in the northernpart of the unit. S2 dips to the south in the northern part ofthe unit and to the north in the southern part, leading to thefan shape of this unit. On the southwestern border of theunit a L3 stretching lineation oriented N030�/45� is locallypresent on the main foliation plane. The northern contact ofthe Drakkarpo unit involves the Shergol conglomeratessuggesting that thrusting toward the north was probablyactive after the Oligocene and could be related to D3(Figure 3).[33] 6. In the Ribil unit the main structures are hecto-

metric south verging F1 folds with fold axis orientedN322�/13�, associated with a northeast dipping foliation(N135�40�N) and L1 lineation (N060�/30�). These foldsare associated with shear bands showing top to thesouthwest thrusting movement (Figure 3). A few hundredmeters before the Tso Morari unit, extensional structuresD2–3 appear. D2–3 structures are characterized bynortheast dipping S/C structures, drag folds with axialplane oriented N130�/25�N, and late kink bands. Thestretching lineation L3 is perpendicular to the boundariesof the units, and is oriented N030�. These structures arecarried by chlorite and quartz assemblages suggestinggreenschist facies metamorphic conditions during D2-3.Finally, a late brittle extensional fault dipping at about60� toward the northeast marks the boundary between theRibil and Tso Morari units. This brittle fault belongs tothe Zildat zone, which separates the weakly metamor-phosed Indus Suture Zone from the UHP Tso Morari unit(Figures 4, 5, and 6). The D2-3 normal ductile structuresare mainly observed in the footwall of the Zildat zone, inthe Tso Morari unit.3.1.2. Interpretation[34] Regionally, the first deformation phase (D1) ob-

served in the Indus Suture Zone corresponds to the south-ward thrusting of this zone over the Indian continentalmargin. Relicts of the Nidar ophiolite are observed in thecore of the Mata-Karzog synclinorium, suggesting theobduction of the Nidar ophiolite onto the Indian marginduring D1 (Figure 9).[35] The D2 phase is related to back thrusting of the

Indus Sequence and Drakkarpo unit toward the northeast,

Figure 7. Photographs ofD1-D2-D3 phases of deformation in the TsoMorari unit. (a) F1 upright foldwith axis (N040�/10�),close to the Pologonka La. F2 recumbent folds overprint F1 fold, La2 (N175�/14�), orthogneiss (o), metabasalts (mb),metapelites (mp), (ap) trace of the axial plane. (b) Photomicrograph under crossed nikols of the eclogitic foliation inthe core of a metabasic lens, underlined by the dynamic recrystallization of omphacite (omph) and garnet (grt).(c) Photomicrograph under crossed nikols of a Mg-rich metapelite showing the S1 foliation defined by kyanite and phengite(with a Si4+ content of 3.58). F2 folds folded S1 foliation during the retrogression under blueschist metamorphic conditionsas shown by the crystallization of Mg-chlorite and phengite with a Si4+ content of 3.45 in the hinge of the F2 fold.(d) Deformed potassium feldspar porphyroclast in the northern limb of Tso Morari orthogneiss showing top to the SSW C2movement (shear). (e) Metabasic lens in the southern limb of the Tso Morari Unit showing a (C2) top to the north shearing.(f ) C3 normal shears in metabasic lenses in the northern limb of the dome showing top to the north movement. (g) C3normal shears in the orthogneiss showing top to the north extensional movement in the northern limb of the Tso MorariUnit. (h) C3 normal shears in the southern part of the Tso Morari orthogneiss. These C3 shear planes show top to the southnormal movement. (i) Photomicrograph of the C3 shears in a potassium rich metapelite sampled in the western part of theTso Morari unit. Kyanite + biotite are developed in the C3 shear plane at the expense of staurolite + chlorite S2 foliation.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

9 of 18

TC3003

while the Ribil unit underwent ductile extension toward thenortheast. The S2 pattern of the Indus Suture Zone ischaracterized by a S-shape curvature compatible with adextral strike-slip movement. This S-shape of the foliation

combined with the fan-shape geometry of the Indus SutureZone, of the Drakkarpo unit and with the asymmetric shapeof the Nidar Ophiolite suggests a dextral transpressiveregime in the Indus Suture Zone during D2.

Figure 8. Structural map and stereoplots of the studied area showing the structures related to the D2deformation phase in the Drakkarpo, the Ribil, the Tso Morari, and the Mata units. Some D3 structuresare also represented on this map in the Mata and Tso Morari units.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

10 of 18

TC3003

[36] D3 is mainly observed closed to the Tso Morari unitand is related to normal movements top toward the northeastunder ductile and brittle conditions.[37] In the studied area the global D1–D2 strain pattern is

partitioned between a N30� direction of shortening and aN120� direction of dextral wrenching. This shorteningdirection is close to the global N020� convergence directionbetween Asia and India [Patriat and Achache, 1984]. Theobliquity between the convergence direction and the short-ening direction in the Indus Suture Zone may lead to strainpartitioning, and explain the occurrence of dextral strike slipmovement on the Indus Suture Zone. A hundred kilometersnorth of the Indus suture zone, the dextral, Karakorum fault,is still active (Figure 1).

3.2. Structural Evolution of the Tso Morari Uhp Unit

3.2.1. Observations[38] The Tso Morari unit has an elongated dome geom-

etry, corresponding to a doubly plunging anticline towardthe northwest and southeast (Figures 2 and 8) [Thakur,1983]. The principal axis of this elongated dome has aNW-SE orientation and dips 10� to the NW (Figure 8).According to this geometry the Tso Morari dome has amaximum thickness of 7 km. The dome is characterized bya flat foliation S2 in its central part, which becomes steeperwith opposite dipping directions on its borders (Figure 3).Two normal ductile shear zones bound the dome, the Zildatshear zone to the northeastern limb and the Karzog shearzone to the southwestern limb (Figures 3 and 4).[39] Three main phases of ductile deformation are ob-

served in this crystalline massif (Figure 9). The first phase(D1) is only preserved in the central part of the unit. Thesecond one (D2) is well developed all over the Tso Morariunit except on its border, where D3 has strongly overprintedit. The structures related to each phase of deformation were

observed either in the orthogneiss, in metabasic or meta-pelitic rocks. Our structural study concerns mainly D2 andD3 events, which are clearly related to the exhumation ofthe UHP Tso Morari unit.3.2.1.1. D1 Deformation[40] There is very few evidence of D1 structures

(Figures 7a, 7b, and 7c), which are only observed in thenorthwestern part of the dome close to the Polokongka La(Figure 8). D1 is characterized by steep tight to isoclinalfolds (F1) of centimeter- to hectometer-scale associated witha subvertical axial plane cleavage (S1), oriented N050�/70�NW (Figure 7a). These folds deform the orthogneiss aswell as the metabasic levels and the metasediments. The S1foliation in the metabasic rocks is borne by eclogiticminerals, garnet and omphacite (Figure 7b). The omphaciteshave recrystallized dynamically and define the foliation andthe mineral lineation, showing that D1 was recorded undereclogitic facies conditions (Figure 6) [de Sigoyer et al.,1997]. In the metapelitic rocks D1 defines a kyanite,phengite, quartz foliation [Guillot et al., 1997].3.2.1.2. D2 Deformation (Figures 7a, 7c, 7d, 7e, 8, and 9)[41] Recumbent isoclinal to open metric folds (F2)

deforms the S1 foliation (Figures 7a and 7c). D2 ispervasive and characterizes the main deformation phase ofthe Tso Morari unit (Figures 3 and 8); D1 structures aretransposed to shallow dipping S2-L2 structures. In thecentral part of the unit, the S2 schistosity is flat, the foldaxis La2 strike of about N130�, they are subhorizontaland parallel to the mineral and stretching lineation (L2)(Figure 8). Toward the edges of the dome, the S2 foliation issteeper. The general trend of the S2 foliation defines thedome shape of the Tso Morari. On the border of the domethe L2 stretching lineation is perpendicular to the La2fold axis along a N040� direction. In the northern limb ofthe dome centimeter to hectometer Z-shape F2 folds are

Figure 9. Summary of the structural features related to (D1-D2-D3) phases of deformation in thedifferent units.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

11 of 18

TC3003

observed, whereas on the southern limb, the F2 folds have aS shape. As the polarity of the series remains normal allover the Tso Morari dome, the F2 folds are south verging inthe northern limb of the unit and north verging in itssouthern limb (Figure 3). These antagonistic fold vergingbetween the limbs of the dome are confirmed by 60%(38 over 64) of unambiguous shear criteria (C/S structures)which shows top to the southwest sense of shear in thenorthern limb of the dome (Figure 7d), and top to thenortheast sense of shear in the southern limb (Figure 7e).The C/S structures are associated with pressure shadows onthe K-feldspar clasts in the orthogneiss which is a stronglylamined porphyric granite-mylonite, the shear criteria are ofopposite sense in both side of the dome (Figure 7d) as wellas in the metapelites. Asymmetric basic lenses also showtop to the SW shear criteria in the northeastern limb and topto the NW shear criteria in the southern limb of the dome(Figure 7e). As we never observed evidence of superposi-tion of these structures interpreted as D2 structures, wepropose that the opposite sense of shear and oppositeverging folds on both limbs of the Tso Morari domecorrespond to a single deformation event.[42] In the core of unaltered metabasic lenses the S1

eclogitic foliation (defined by garnet and dynamic crystal-lization of omphacite) (Figure 7b) is partly replaced byblueschist mineral association while on the border of thelenses eclogitic minerals are mainly replaced by amphib-olitic mineral assemblages. This suggests that basic rockswere stretched and sheared during the blueschist D2 phaseof deformation (Figure 7e). In the Fe-rich metapelitic rocksphengite and glaucophane define the L2 mineral lineation.The S2 foliation, which represents the axial surface of theF2 folds, is underlined by blueschist mineral associations. Inthe Mg-rich metapelites, Mg-chlorite crystallized in thehinge of F2 folds at the expense of kyanite and phengite,Mg-chlorite is frequently described in HP rocks, its crystal-lization at the hinge of the folds shows that the F2 folds weredeveloped at the beginning of decompression (Figure 7c).These petrological observations suggest that D2 structureswere recorded during the first part of the Tso Morari rockexhumation from eclogitic conditions down to blueschistmetamorphic conditions (Figure 6).3.2.1.3. D3 Deformation[43] In the central part of the dome, D2 structures are

slightly overprinted by D3 shear zones (C3). The D3structures are mainly localized on the borders of the domein two kilometric wide zones defined as the Zildat normalshear zone to the north and the Karzog normal shear zone tothe south (Figures 3, 4, and 8). The Zildat normal shear zonerepresents the ductile northern boundary of the Tso Morariunit; it separates the UHP unit from the low-grade metamor-phic Indus Suture Zone (Figures 3 and 4). The Zildat normalshear zone follows the Zildat valley west of Sumdo village,and continues in the southern side of the Ribil valley, eastof Sumdo (Figure 8). The footwall of this shear zone islocated within the Tso Morari unit, it corresponds to a largekilometric deformed band characterized by northeast dip-ping S3/C3 structures. The shear planes trend N120/30� NE,S3 foliation plane bears a stretching lineation L3 oriented

(N065�/30�). F3 drag folds with fold axis La3 orientedN345�/15� are usually associated with the S3/C3 shearbands. Shear criteria indicate top to the northeast movement(Figures 7f and 7g). Within the Zildat normal shear zone,occur hectometric lenses of serpentinites (Figure 4). Geo-chemical analyses indicate a depleted mantle origin for theserpentines. The serpentinites were removed from thehydrated mantle wedge by the Tso Morari massif duringits exhumation, suggesting that the Zildat normal shear zonewas active from the start of the Tso Morari exhumation[Guillot et al., 2000, 2001]. Close to the contact with theRibil unit, the C3 shear bands are underlined by phengites,chlorites, secondary chloritoid, albite and rare biotites in themetapelites, relics of glaucophanes are totally transformedinto chlorite and albite. These indicate that D3 continue todevelop under lower amphibolite to greenschist metamor-phic conditions [Guillot et al., 1997]. This ductile shear zoneis later crosscut by a steep normal brittle fault, strikingnorthwest and dipping 60�N (Figure 4). This steep normalfault crosscut the gently NE dipping cleavage. In the hangingwall (Ribil unit) of the Zildat normal shear zone (Figure 3)few shear indicators are observed. The contrasted record ofD3 deformation between the footwall and the hanging wallof the Zildat shear zone is interpreted as a consequence of astrong vertical motion of the UHP Tso Morari unit relative tothe weakly metamorphosed Ribil unit. The southern limb ofthe Tso Morari unit is affected by a wide south dippingextensional ductile shear zone, more than five kilometerswide, the Karzog normal shear zone. This zone separates theTso Morari unit from the less metamorphosed south Mata-Karzog unit. In this zone the S3 schistosity (N070�/20�SE)bears a stretching lineation L3 (N170�) that is parallel to theL2 stretching lineation, and perpendicular to the C3 shearplanes. The S3/C3 structures observed in the orthogneisssuggest top to the southwest movements (Figure 7h). TheKarzog normal shear zone is cross cut by the late Peldonormal brittle fault (Figure 8). No serpentinites were foundwithin the Karzog normal shear zone.[44] In the rest of the Tso Morari unit the C3 shear planes

are mainly developed in the metagranites and metapelites(Figures 7f, 7g, 7h, and 7i). In the orthogneiss, the C3 shearplanes are observed on the border of the K-feldspar por-phyroclasts (Figures 7g and 7h). In the metagraywackesfresh kyanite and biotite developed in C3 shear planes at theexpense of staurolite, phengite and chlorite in the S2foliation plane (Figure 7i), suggesting that these shear bandswere developed under amphibolitic conditions during theexhumation of Tso Morari unit [Guillot et al., 1997].Metabasic lenses were also sheared and fractured duringD3 (Figure 7f ). Along the fractures, the normal shear zoneand on the borders of the basic lenses, amphibolitic mineralassemblages (Ca amphibole and biotite) have crystallized atthe expense of eclogitic and blueschist minerals [de Sigoyeret al., 1997]. All the steps from fresh eclogites to deeplyrecrystallized amphibolites are observed from the core to theborder of the basic lenses or toward the normal C3 shearplanes. The thermobarometrical estimates carried out on theamphibolitic assemblage of metasediments and metabasicrocks suggest a temperature of 630� ± 30�C for a pressure of

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

12 of 18

TC3003

9 ± 1 kbar [de Sigoyer et al., 1997; Guillot et al., 1997](Figure 6).[45] Tilted kilometric blocks bounded by normal faults

are also observed in the Tso Morari dome, they are northverging in the northern flank and south verging in thesouthern flank. Thus the normal phase of deformation D3was first ductile under amphibolitic (Figures 7f, 7g, 7h,and 7i) and greenschist facies conditions, then brittle.[46] Finally, Steck et al. [1998] described Quaternary

dome and basin structure in this area with a wavelengthof 10 to 150 km for an amplitude of 3 to 5 km. ManyQuaternary faults are also observed, some of them arereported on the structural map (Figure 8). The Tso Morariand Kiagar Tso lakes are located on conjugated faults[Berthelsen, 1953]. At Puga, close to Zildat normal zone,hot sulphur spring is situated on an active fault. Steck et al.[1998] attributed these Quaternary structures to an activedextral transpressional regime, which created N-S strikingnormal faults, NW-SE striking dome and basin structures ina main shear zone, parallel to NW-SE striking Indus Suturezone. These quaternary structures may partly overturnedprevious structures.3.2.2. Interpretation[47] The different stages of deformation described in the

Tso Morari unit can be interpreted as a consequence ofprogressive deformation during the exhumation of this UHPunit (Figure 9). The rare F1 upright folds observed suggest astage of horizontal shortening under eclogitic conditions.During D2 (the most penetrative deformation phase), thedevelopment of the flat S2 foliation in the central part of theunit implies a component of vertical shortening. Thisflattening is associated with a NW-SE lineation L2, parallelto the La2 folds axis and to the long axe of the Tso Moraridome. These structures suggest a component of horizontalstretching during D2 along a NW-SE direction. On theborders of the dome the opposite vergence of the D2structures toward the core of the unit can be interpreted assecondary structures of an N120–130� huge anticlinecompatible with a N020�–30� direction of shortening[e.g., Burg, 1987; Burg and Podladchikov, 1999]. Contraryto Thakur [1983] and Steck et al. [1998], who haveinterpreted the Tso Morari dome as the result of a latecompressional event, we propose that the dome geometry ofthe Tso Morari is an earlier compressional structure mainlydeveloped during D2 in continuity with D1 horizontalshortening structures. The combination of previous petro-logical studies with our interpretation of the D2 structuresallows proposing a model for the first part of the Tso Morariexhumation. We propose that D2 structures reflect theexhumation of the Tso Morari unit, realized by a combina-tion of vertical and horizontal component of displacementsprobably along the subduction plane up to 40–30 km. Theend of the exhumation is controlled by D3 structures, whichwere developed from ductile amphibolitic-greenschist facies(40–30 km depth) to the brittle conditions close to thesurface. D3 structures are mainly localized on the borders ofthe dome and characterized by normal shear zones thatsuggest a local noncoaxial strain regime (Figure 9). Wepropose that the gently NE dipping cleavage in the footwall

of the Zildat normal shear zone, and SW dipping cleavagein the footwall of the Karzog normal shear zone wereformed in a zone of subvertical shortening below thesetwo major low angle shear zones. At regional scale, thefinite D3 strain pattern suggests a subvertical uplift of theTso Morari dome across the upper crust associated withtectonic denudation and erosion (Figure 10). This strainevolution is very similar to the strain evolution recorded bythe Kangmar dome in southern Tibet [Lee et al., 2000]. TheD3 shear structures could have been developed deeperwithin the Zildat zone in order to accommodate the relativemotion of the Tso Morari with its surroundings. Theevolution from D1 to D3 during the Tso Morari exhumationmay reflect the transition from an exhumation along thesubduction plane, in a serpentinized channel toward avertical uplift across the upper crust (or the accretionarywedge) (Figure 10).

3.3. Structural Evolution of the Mata-Karzog Unit

3.3.1. Observations[48] As discussed previously, the occurrence of chromitic

pods and basalts at Karzog suggest the obduction of theNidar Ophiolite on the Mata sequence (Figures 2 and 9).The contact of the Karzog ophiolitic complex with the Matasediments is concordant and strongly deformed, it is inter-preted as a tectonic contact. This obduction is considered asthe first phase of deformation D1 in the Mata unit. D2structures can be observed within the entire Mata unit. TheS2 foliation trend in the Mata unit is slightly discordant withthe S2 foliation in the Tso Morari unit. S2 foliation dipssouthward and has been developed as an axial surfacestructure for the ductile recumbent F2 folds (Figure 8).Asymmetric (F2) folds deformed the sedimentary sequenceof Mata and the Karzog ophiolite, as the Karzog ophiolitelies in the core of a synclinorium. The sedimentary se-quence is normal above the Mata granite (G. Mascle,personal communication, 1997) suggesting that the F2 foldsare NE verging. The fold axes strike of about N120� in thenorthern part of the unit, and are along a N045� direction inthe southern part (Figure 8). The stretching lineation L2trends N160� in the northern part of the unit and is orientedN020� in the southern part. Mylonitic zones are observed onthe borders of the Mata-Rupshu granite, showing top tothe north thrusting shear criteria on the lower contact. Themylonitic zones localize the deformation between rocks ofdifferent competence (granite and sediments). In basicrocks metamorphic minerals, such as zoisite, actinoliteand biotite, bear the S2 cleavage, and suggest uppergreenschist to epidote-actinolite amphibolite conditionsduring D2 (Figures 5 and 6) [de Sigoyer, 1998]. Later,S3/C3 shear bands, showing top to the SW extensionalmovements overprint D2 structures. These extensional shearbands are mostly developed in the southern part of the unit,close to the Phirse valley, where F3 SW verging folds areobserved. They correspond to top to the SW extensionalmotion of the Spiti sedimentary cover relative to the Mataunit (Figure 9). Chlorite and micas underline D3 structures,showing that D3 began under greenschist facies conditionsand continued under brittle conditions.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

13 of 18

TC3003

[49] All the structures described in the Mata-Karzog unitwere disturbed and rotated by a late fault network still activetoday. According to satellite images, this fault network isrelated to the formation of N-S pull-apart basins. The N080�dextral strike slip fault that cross cut the Tso Morari lake,could be directly related to the present day activity of thedextral Karakorum fault (Figures 1, 2, and 8).3.3.2. Interpretation[50] The Mata-Karzog unit is separated from the Tso

Morari unit by the south dipping Karzog normal shearzone, this unit lies structurally over the Tso Morari unit.This observation is consistent with the metamorphic con-ditions recorded by these two domains (UHP in the TsoMorari unit, upper greenschist conditions in the Mata unit).As in the Indus suture zone, the first stage of deformation inthe Mata-Karzog unit is related to the obduction of theKarzog ophiolite onto the Mata unit. Contrary to theSpontang ophiolite that thrust over the Eocene sedimentsmore to the west (Figure 1), the Karzog ophiolite lies abovePermo-Carboniferous sediments. No Jurassic and Creta-ceous rocks are observed in the Mata unit nor in the TsoMorari unit. These rocks were either not deposited or eitherwere eroded or scrapped off during the subduction of theIndian margin [Guillot et al., 2000].[51] After the ophiolite obduction, the Mata-Karzog

unit was deformed by the D2–D3 deformation phases(Figure 9). The north verging F2 folds observed in thisdomain, can be compared to those observed in the south-ern limb of the Tso Morari unit. They are all compatiblewith a N030� direction of shortening, as observed else-where in the study area (Figure 8). However, in the TsoMorari unit D2 structures were developed under blueschistconditions, whereas greenschist metamorphic conditionswere recorded during D2 in the Mata-Karzog unit. D2structures were not recorded at the same structural level inMata and Tso Morari units. The late D3 phase that over-printed the F2 folds in the Mata unit may be partlycontemporaneous to the extensional D3 deformation phaserecorded by the Tso Morari unit. The bulk D2 deformationrecorded in the Mata unit after the ophiolite obduction canbe explained by the pinching of this unit between theIndian convergent plate and the Tso Morari unit when thislater was coming up through the surface. At the end of theTso Morari exhumation, the Mata unit slides southwardduring D3. Note that all the structures in the Mata-Karzogunit have opposite vergence by comparison to the struc-tures observed in the Ribil unit, north of the Tso Morariunit (Figures 3, 8, and 9). Thus, at the scale of the studyarea, the bulk strain pattern is compatible with a NE-SWdirection of horizontal shortening from the ophioliteobduction to the D2 phase. It was followed by extensionaltectonics during D3 corresponding to the tectonic denuda-tion and erosion of the whole area, which is associatedlocally (on the Zildat and Karzog normal shear zones) withvertical shortening (Figures 3, 8, and 9).

4. Timing of the Deformations

[52] The bulk strain pattern emphasizes that the TsoMorari unit and the surrounding units have undergone a

Figure 10. Simplified geodynamic evolution of the TsoMorari exhumation. The geometrical observations done onthe different units of the area suggest that the exhumationof the UHP Tso Morari unit was mainly ruled by horizontalshortening between the Indian and Asian convergentlandmasses. The Indian-Asian convergent rates are takenfrom Patriat and Achache [1984]. (HHC is HigherHimalayan Crystallines.)

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

14 of 18

TC3003

synchronous evolution only since D3 deformation phase.Previous deformations were independent in the differentunits. The first phase of deformation in the weakly meta-morphic units (Indus Suture Zone and Mata unit) is relatedto the obduction of the Nidar Ophiolite onto the Drakkarpoand Ribil oceanic islands and on the Indian Margin (Mataunit). This obduction may have occurred during Eocenetimes as suggested by the 40Ar/39Ar age at 50 Ma obtainedon amphibole from the Karzog ophiolite (G. Maheo andI. M. Villa, personal communication), as the Spontangophiolite [Colchen et al., 1987; Reuber et al., 1987]. Thetiming of D2–D3 in the Indus sequence is estimated inthe western part of Ladakh at about 40–35 Ma by VanHaver et al. [1986], on the basis of K/Ar cooling ages. Theoccurrence of the Oligocene Shergol conglomerates inthe tectonic contact between the Nidar Ophiolite and theDrakkarpo unit shows that D3 was still active during and/orafter Oligocene times.[53] Schlup et al. [2001] report fission track ages in the

Mata unit at about 45 ± 2 Ma on zircon and 40 ± 3 Ma onapatite. The Mata unit was then at shallow crustal level atthe end of Eocene times (Figure 10).[54] In the Tso Morari unit, D1 recorded the eclogitiza-

tion and is dated at about 55 ± 6 Ma (Paleocene/Eoceneboundary) by Lu/Hf, Sm/Nd and U/Pb geochronology oneclogitic mineral assemblages (Figures 6 and 10) [deSigoyer et al., 2000]. The transition from D2 to D3occurred during the transition from blueschist to amphib-olitic metamorphic conditions. The amphibolitic mineralshave been dated at about 47 ± 2 Ma (lower Eocene) bySm/Nd, Rb/Sr and Ar/Ar geochronology (Figure 6) [deSigoyer et al., 2000]. Thus D2 occurred between 55 andbefore 47 Ma. Finally, 40Ar/39Ar age of 30 ± 1 Ma(Oligocene) have been obtained on newly crystallizedbiotites and muscovites sampled in metagraywackes ofthe western part of the massif [de Sigoyer et al., 2000].In the rest of the unit, fission track ages on zircon,between 40 ± 2 and 34 ± 2 Ma, and on apatite, between24 ± 2 and 8 ± 2 Ma are reported by Schlup et al.[2001]. The youngest apatite fission track ages wereobtained in the Zildat normal shear zone, indicating thatthe normal fault was recently active [Schlup et al.,2001]. The 40Ar/39Ar age and fission track ages confirmthat the Tso Morari unit was already exhumed up to uppercrustal levels by upper Eocene to lower Oligocene times(Figures 6 and 10).

5. Exhumation Processes in the Himalayan

Context

[55] Combining the ages obtained on the Tso Morarimassif and converting the pressures to depth, exhumationrates turn out to be greater than 7 mm/yr for the first part ofexhumation (between 90 to 30 km) (Figures 6 and 10) [deSigoyer et al., 2000]. This first part of exhumation tookplace during the lower Eocene (between 55 and 47 Ma). Itwas mainly controlled by the D2 deformation. Petrologicaldata suggests that this part of exhumation was quasiisothermal [de Sigoyer et al., 1997; Guillot et al., 1997]

(Figures 6 and 10), and occurred within a partially serpenti-nized mantle wedge [Guillot et al., 2000, 2001]. Paleomag-netic data on the Indian ocean and oceanic sediments showthat during the lower Eocene the convergence rate betweenIndia and Asia was greater than 10 cm/yr, suggesting thatcontinental subduction was still active and rapid at this time[Patriat and Achache, 1984; Klootwijk et al., 1992; Guillotet al., 2003]. The beginning of continental erosion is datedat 52 Ma by Garzanti et al. [1987] in the northwestHimalaya. Therefore the first part of the Tso Morari exhu-mation associated with rapid convergence rate took placewithout significant erosion [de Sigoyer et al., 2000; Guillotet al., 2003].[56] From upper Eocene to Oligocene times, the Tso

Morari unit was exhumed from 30 to 10 km depth with aslower exhumation rate (1.2 mm/yr). This second part ofexhumation is associated with the D3 normal structures. Atthe Himalayan scale, this period corresponds to a decreaseof the convergence rate between India and Asia (from10 to 5 cm/yr after 50 Ma) [Patriat and Achache, 1984]. Thisdecrease is due to the thickening of the Himalayan wedgeby the underthrusting of the Higher Himalayan Crystallineunit under the internal zone (Figure 10) [Treloar et al.,1989; de Sigoyer et al., 2000]. According to numericalmodeling [Jamieson et al. 1996], a simple decrease of theconvergence rate associated with a progressive thickeningof a collision zone can explain the temperature increaserecorded in the internal part of the belt at this time, as itwas recorded by the Tso Morari unit. Goffe et al. [2003]show that the type of rocks accreted to a wedge hasimportant effects on the thermal regime of orogenicwedge. Prior to the continental subduction of the Indianmargin only oceanic crust and sediments were subducted,which explains the low temperature metamorphismrecorded by the Tso Morari UHP unit. Later, during theexhumation of the Tso Morari unit, granitic material withhigher radioactive heat production was accreted to thewedge, which led to a temperature increase. In contrast,Chemenda et al. [2000] and Guillot et al. [2003] explainthe heating of the internal zone by the break-off of theIndian continental slab and the attached subducted oceaniclithosphere. Whatever the cause of the slight temperatureincrease, the last part of the Tso Morari exhumationoccurred by vertical extrusion associated with tectonicdenudation and substantial erosion.

6. What is the Motor of the Exhumation?

[57] The different phases of deformation recorded in theTso Morari unit imply several exhumation processes, acti-vated at successive structural levels. The underthrusting ofthe buoyant Higher Himalayan Crystalline unit below theTso Morari unit from middle Eocene to Oligocene times,can easily explain the end of the Tso Morari unit exhuma-tion by underplating and consequently tectonic denudationas observed in the internal part of accretionary prism[Platt, 1986; Goncalvez et al., 2000; Rolland et al., 2000](Figure 10). However problems remain in understanding thebeginning of the Tso Morari unit exhumation from 90 to

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

15 of 18

TC3003

30 km depth. In the following, we will confront our data tothe classical models of exhumation.[58] The high rate of exhumation of the Tso Morari

eclogitic rocks (>7 mm/yr) argues against an exhumationruled by that erosion or isostatic reequilibration [Duchene etal., 1997]. The relatively low temperature (<630 ± 50�C)recorded during the decompression of the Tso Morari unitand the geodynamical context is inconsistent with anexhumation ruled by extensional collapse [Dewey et al.,1993].[59] Buoyancy forces have been classically admitted as

the major motor to exhume HP to UHP rocks [Englandand Richardson, 1977; Chemenda et al., 1996, 2000; Ernstand Liou, 1999]. However, Henry et al. [1997] show thatthe density of an eclogitized upper or intermediate conti-nental crust (3.06–3.31g cm�3) is close to the density ofthe mantle (3.3 g cm�3). In the Tso Morari unit most ofthe rocks were recrystallized under eclogitic conditionsapart some orthogneiss and granites and therefore we canthen assume a density of 2.9–3 g cm�3 for the whole HPmassif. Cloos [1993] shows that the subduction of alithospheric slab is possible only if the difference ofdensity between the mantle and the subducted slab isgreater than 10%. If the exhumation of an eclogitic massifacross the mantle is only driven by buoyancy forces, asimilar density difference of 10% should be expectedbetween the exhumed massif and the mantle to initiateexhumation. The analogue modeling of Chemenda et al.[1996] suggests that the exhumation of eclogitized terrainsby buoyancy forces is possible only if a thick piston ofbuoyant (unmetamorphosed) continental crust push theUHP rocks toward the surface. As the Tso Morari unitrepresents a small portion of the upper crust (100 � 50 �7 km) the buoyancy forces were probably not the onlymotor for the exhumation of this UHP massif.[60] The symmetry of the structures observed in the

Tso Morari dome and in the surroundings and thedirection of L2 lineation parallel to the long axe ofthe Tso Morari dome argue against an exhumation bythrusting of nappes toward the foreland as suggested bySteck et al. [1998].[61] The model of exhumation in an accretionary sedi-

mentary prism by corner flow is very efficient to explain therapid synconvergent exhumation of HP rocks located at therear of the bulk accretionary prism [Platt, 1986; Allemandand Lardeaux, 1997]. The location of the Tso Morari unitclose to the paleosubduction plane, between the suture zoneand the Indian shelf at the rear of orogenic prism, argues forthis model. The dynamics of corner flow model is governedby underplating. Nobody knows what is beneath the TsoMorari massif, neither if there was some underplating belowthe Tso Morari at the beginning of its exhumation, i.e.,before upper Eocene. On the other hand, many observationssupport the channel flow model.[62] The serpentinized rocks observed around the Tso

Morari dome, within the Zildat ductile extensional faulthave a lower density (2.5 g cm�3) than the eclogitic rocksand than the mantle wedge, a low coefficient of viscosityof about 1019 Pa s�1 at 550�C [Carter and Tsenn, 1987],

and low coefficient of friction (mi � 0.15 � 0.3) [Escartinet al., 2001]. The presence of serpentinites even in smallamounts, allow the localization of deformation. As pro-posed by Guillot et al. [2000], the serpentinites haveprobably played a leading role in the exhumation of theUHP Tso Morari dome acting as a lubricant in the mantlewedge. In this model, the exhumation within a low-viscosity channel is effectively controlled by a hydrody-namic return flow along the subducting plate [Cloos, 1982,1986]. The Tso Morari dome could be exhumed as a rigidUHP unit embedded in light and soft serpentinite sea-mounts as observed today in the Mariana zone [Yamamoto,1995]. Moreover, Guillot et al. [2001] show that the lownormal deviatoric stresses (few tens of MPa) developed atthe interface between a rigid subducting plate and a low-viscosity serpentinized wedge can produce a hydrodynamicreturn flow necessary to move an eclogitic unit back up.These normal stresses could also be responsible for thedevelopment of the D1 and D2 structures observed withinthe Tso Morari unit, parallel to the slab surface. The onsetof the Tso Morari exhumation was likely rapid andoccurred along the subduction plane allowing a componentof lateral movement (associated with the vertical compo-nent) as deduced form this study. In contrast the exten-sional structures developed closer to the surface were mostprobably associated with a decrease of the exhumationrate. The changes in the Tso Morari exhumation rate andin its metamorphic conditions coincide with changes of thedeformation phase and with the variation of the India-Asiaconvergence rate. This observation also suggests thatboundary forces which developed at the interface betweenthe subducting plate and the mantle wedge, exert effec-tively a strong control on the exhumation of high- toultrahigh-pressure rocks.[63] It is noticeable that in the Himalayas, so far eclogitic

rocks have only been described around the NW syntaxis inPakistan and Ladakh [Pognante and Spencer, 1991; Le Fortet al., 1997; O’Brien et al., 2001]. The scarcity of eclogiticrocks in the Himalayas could be either due to a lack ofknowledge of the internal zone or related to the obliquity ofthe convergence between Indian and Asian plates along thewestern syntaxis [Seeber and Pecher, 1998]. Strain parti-tioning caused by this obliquity and emphasized in thewestern syntaxis by the active Karakorum fault, and favoredthe earlier dextral motion deduced from this study in theIndus suture zone, has probably helped the exhumation ofthe Tso Morari. As suggested by numerical modeling[Fossen and Tikoff, 1998; Thompson et al., 1997a, 1997b]oblique convergence, and the ensuing strain partitioningplay a major role in the vertical motion of HP to UHProcks.

7. Concluding Remarks

[64] As suggested 10 years ago by Platt [1993], there isno single mechanism that can explain the exhumation of allhigh-pressure terrains. Today two main hypotheses aredeveloped concerning the motor of the exhumation; forsome, modeling proposes that volume forces act as the

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

16 of 18

TC3003

motor of exhumation, whereas for others exhumation isdriven by the forces acting on the boundaries of thedeforming system. Our geological study on a naturalexample shows that boundary forces and velocity changeat the lithospheric plate scale controlled the exhumation ofthe UHP massif and the associated tectonic denudation.However, most of the high-pressure rocks involved in ourcase have a sedimentary or granitic protolith, suggesting

that buoyancy forces played a role in the exhumationprocesses.

[65] Acknowledgments. We kindly acknowledge the financial sup-port of the CNRS-INSU by the IDYL-HIMALAYA program. Discussionson the structural evolution of the Himalayan belt and on exhumationprocesses with A. Pecher and G. Mascle in the field, J. P. Burg, PascalAllemand, J. M. Lardeaux, J. Platt were very stimulating. A. Pfiffnerprovided careful reviews that greatly improved the manuscript.

References

Allemand, P., and J. M. Lardeaux (1997), Strain parti-tioning and metamorphism in a deformable oro-genic wedge: Application to the Alpine belt,Tectonophysics, 280, 157–169.

Argand, E. (1916), Sur l’arc des Alpes Occidentales,Eclogae Geol. Helv., 14, 145–204.

Bassoullet, J. P., M. Colchen, T. Juteau, J. Marcoux,G. Mascle, and G. Reibel (1983), Geological studiesin the Indus suture zone of Ladakh (Himalayas), inContribution to Himalayan Geology, vol. 2, editedby Gupta, pp. 96–124, Indust. Publ. Corp., Dehli.

Baud, A., R. Arn, P. Bugnon, A. Crisinel, E. Dolivo,J. G. Esther, J. G. Hammerschlang, M. Marthaler,H. Masson, and A. Steck (1982), Le contact Gond-wana-peri-Gondwana dans le Zanskar oriental(Ladakh, Himalaya), Bull. Soc. Geol. Fr., 24,341–361.

Berthelsen, A. (1953), On the geology of the Rupshudistrict, NW Himalaya, Medd. Dansk. Geol. Foren.Kobenkhavn, 12, 350–415.

Blondeau, A., J. P. Bassoullet, M. Colchen, T. L. Han,J. Marcoux, G. Mascle, and T. Van Haver (1986),Disparition des formations marines a l’Eocene in-ferieur en Himalaya, Sci. Terre Mem., 47, 103–111.

Burg, J. P. (1987), Regional shear variation in relationto diapirism and folding, J. Struct. Geol., 9(8),925–934.

Burg, J. P., and Y. Podladchikov (1999), Lithosphericscale folding: Numerical modelling and applicationto the Himalayan syntaxes, Int. J. Earth Sci., 88,190–200.

Carter, N., and M. Tsenn (1987), Flow properties of con-tinental lithosphere, Tectonophysics, 136, 27 –63.

Chemenda, A. I., M. Mattauer, and A. Bokun (1996),Continental subduction and a mechanism for exhu-mation of high-pressure metamorphic rocks: Newmodeling and filed data from Oman, Earth Planet.Sci. Lett., 143, 173–185.

Chemenda, A., J. P. Burg, and M. Mattauer (2000),Evolutionary model of the Himalaya-Tibet system:Geopoem based on new modelling, geological andgeophysical data, Earth Planet. Sci. Lett., 174,397–409.

Cloos, M. (1982), Flow melanges: Numerical model-ling and geological constraints on their origin in theFranciscan subduction complex, Geol. Soc. Am.Bull., 93, 330–345.

Cloos, M. (1986), Blueschists in the Franciscan com-plex of California: Petrotectonic constraints on up-lift mechanisms, Geol. Soc. Am. Bull., 164, 77 –93.

Cloos, M. (1993), Lithospheric buoyancy and collisionalorogenesis: Subduction of oceanic plateaus, conti-nental margins, island arc, spreading ridges, andseamounts, Geol. Soc. Am. Bull., 105, 715–737.

Colchen, M., I. Reuber, J. P. Bassoullet, J. P. Belier,A. Blondeau, M. Lys, and P. De Wever (1987),Donnees biostratigraphiques sur les melangesophiolitiques du Zanskar, Himalaya du Ladakh,C. R. Acad. Sci., 305, 403–406.

Colchen, M., G. Mascle, and G. Delaygue (1994),Lithostratigraphy and age of the formations in theTso Morari dome (abstract), J. Nepal Geol. Soc.,10, 23.

Corfield, R. I., M. Searle, and R. G. Owen (1999),Photang thrustsheet: An accretionary complexstructurally below the Spontang ophiolite constrain-

ing timing and tectonic environment of ophioliteobduction, Ladakh, Himalaya, J. Geol. Soc. Lon-don, 156, 1031–1044.

Dal Piaz, G. V., and B. Lombardo (1986), Early-alpineeclogite metamorphism in the Penninic MonteRosa-Gran Paradiso basement nappes of the north-western Alps, in Blueschists and Eclogites, Mem.Geol. Soc. Am., 164, 249–265.

Debon, F., P. Le Fort, S. M. F. Sheppard, and J. Sonet(1986), The four plutonic belts of the Transhima-laya-Himalaya: A chemical, mineralogical, isotopicand chronological synthesis along a Tibet-Nepalsection, J. Petrol., 27, 219–250.

de Sigoyer, J. (1998), Mecanismes d’exhumation desroches de haute pression basse temperature, en con-texte de convergence continentale (Tso Morari, NOHimalaya), Ph.D. thesis, 236 pp., Univ. Claude Ber-nard, Lyon, France.

de Sigoyer, J., S. Guillot, J. M. Lardeaux, and G. Mascle(1997), Glaucophane-bearing eclogites in the TsoMorari dome (eastern Ladakh, NW Himalaya),Eur. J. Mineral., 9, 1073–1083.

de Sigoyer, J., V. Chavagnac, J. Blichert-Toft, I. M.Villa, B. Luais, S. Guillot, M. Cosca, and G. Mascle(2000), Dating the Indian continental subductionand collisional thickening in the northwest Hima-laya: Multichronology of the Tso Morari eclogites,Geology, 28, 487–490.

Dewey, J. F., P. D. Ryan, and T. B. Andersen (1993),Orogenic uplift and collapse, crustal thickness, fab-rics and metamorphic phases changes: The roleof eclogites, in Magmatic Processes and Plate

Tectonics, edited by H. M. Prochard et al., Geol.Soc. Spec. Publ., 76, 325–343.

Duchene, S., J. M. Lardeaux, and F. Albarede (1997),Exhumation of eclogites: Insights from retrogradedepth-time path analysis, Tectonophysics, 280,125–140.

England, P. C., and S. W. Richardson (1977), The in-fluence of erosion upon mineral facies of rocksfrom different metamorphic environments, J. Geol.Soc. London, 134, 201 –213.

Ernst, W. G., and J. G. Liou (1999), Overview of UHPmetamorphism and tectonics in well-studied colli-sional orogens, Int. Geol. Rev., 41, 477–493.

Escartin, J., G. Hirth, and B. Evans (2001), Strength ofslightly serpentinized peridotites: Implications forthe tectonics of oceanic lithosphere, Geology, 29,1023–1026.

Fossen, H., and B. Tikoff (1998), Extended models oftranspression and transtension, and applications totectonic settings, in Continental Transpressionaland Transtensional Tectonics, edited by R. E.Holdsworth, R. A. Strachan, and J. R. Dewey, Geol.Soc. Spec. Publ., 135, 15– 33.

Fuchs, G., and M. Linner (1996), On the geology of theSuture Zone and Tso Morari Dome in EasternLadakh (Himalaya), Jahrb. Geol. Bundesanst.,139, 191–207.

Fuchs, G., and M. Linner (1997), Multiphase tectonicsin the Indus suture zone of eastern Ladakh, paperpresented at 12th Himalayan Karakorum Tibet In-ternational Workshop, Acad. Naz. dei Lincei andSoc. Geogr. Ital., Rome, Italy.

Garzanti, E., A. Baud, and G. Mascle (1987), Sedimen-tary record of the northward flight of India and its

collision with Eurasia (Ladakh Himalaya, India),Geodin. Acta, 1, 297–312.

Girard, M. (2001), Metamorphism and tectonics of thetransition between non metamorphic TethyanHimalaya sediments and the North HimalayanCrystalline Zone (Rupshu area, NW India), Ph.D.thesis, 100 pp., Univ. Lausanne, Switzerland.

Girard, M., and F. Bussy (1999), Late Pan-Africanmagmatism in NW Himalaya: New geochronologi-cal and geochemical data from the Ordovician TsoMorari metagranites (Ladakh, NW Himalaya),Schweiz. Mineral. Petrogr. Mitt., 79, 399 –417.

Goffe, B., R. Bousquet, P. Henry, and X. Le Pichon(2003), Effect of the chemical composition of thecrust on the metamorphic evolution of orogenicwedges, J. Metamorph. Geol., 21, 123–141.

Goncalvez, P., S. Guillot, C. Nicollet, and J.M. Lardeaux(2000), Thrusting and sinistral wrenching in apre-Eocene Caribbean accretionary wedge (Samanapeninsula -Dominican Republic), Geodin. Acta, 13,119 –132.

Guillot, S., J. M. Lardeaux, G. Mascle, and M. Colchen(1995), Un nouveau temoin du metamorphisme dehaute-pression dans la Chaıne Himalayenne: Leseclogites retromorphosees du Dome du Tso Morari,(Est Ladakh, Himalaya), C. R. Acad. Sci., Ser. II,320, 931–936.

Guillot, S., J. de Sigoyer, J. M. Lardeaux, and G. Mascle(1997), Eclogitic metasediments from the TsoMorari area (Ladakh, Himalaya): Evidence for con-tinental subduction during India-Asia convergence,Contrib. Mineral. Petrol., 128, 197–212.

Guillot, S., K. Hattori, and J. de Sigoyer (2000), Mantlewedge serpentinization and exhumation of eclo-gites: Insights from eastern Ladakh, northwestHimalaya, Geology, 28, 199–202.

Guillot, S., K. H. Hattori, J. de Sigoyer, T. Nagler,and A. L. Auzende (2001), Evidence of hydrationof the mantle wedge and its role in the exhuma-tion of eclogites, Earth Planet. Sci. Lett., 193,115 –127.

Guillot, S., G. Garzanti, D. Baratoux, D. Marquer,G. Maheo, and J. de Sigoyer (2003), Reconstructingthe total shortening history of the NW Himalaya,Geochem. Geophys. Geosyst . , 4(7), 1064,doi:10.1029/2002GC000484.

Henry, P., X. Le Pichon, and B. Goffe (1997), Kine-matic, thermal and petrological model of the Hima-layas: Constraints related to metamorphism withinthe underthrust Indian crust and topographic eleva-tion, Tectonophysics, 273, 31– 56.

Honegger, K., V. Dietrich, W. Frank, A. Gansser,M. Thoni, and V. Trommsdorff (1982), Magmatismand metamorphism in the Ladakh Himalayas (theIndus-Tsangpo suture zone), Earth Planet. Sci.Lett., 60, 253–292.

Jamieson, R. A., C. Beaumont, J. Hamilton, andP. Fullsack (1996), Tectonic assembly of invertedmetamorphic sequences, Geology, 24, 839–842.

Jolivet, L., B. Goffe, P. Monie, C. Truffert-Luxey,M. Patriat, and M. Bonneau (1996), Miocenedetachment in Crete and exhumation P-T-t pathsof High-pressure metamorphic rocks, Tectonics,15, 1129–1153.

Klootwijk, C. T., J. S. Gee, J. W. Peirce, G. M. Smith,and P. L. McFadden (1992), An early India-Asia

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

17 of 18

TC3003

contact: Paleomagnetic constraints from NinetyeastRidge, ODP Leg 121, Geology, 20, 395–398.

Lee, J., B. Hacker, W. S. Dinklage, Y. Wank, P. Gans,A. Calvert, J. Wan, W. Chen, A. E. Blythe, andW. Mc Clelland (2000), Evolution of the Kangmardome, southern Tibet: Structural, petrologic andthermochronologic constraints, Tectonics, 19,872–895.

Le Fort, P. (1989), The Himalayan orogenic segment, inTectonic Evolution of the Tethyan Region, NATOASI Ser., Ser. C, vol. 259, edited by A. M. C. Sengor,pp. 289–386, Kluwer Acad., Norwell, Mass.

Le Fort, P., F. Debon, A. Pecher, and J. V. P. Sonet(1986), The 500 Ma magmatic event in alpinesouthern Asia, a thermal episode at Gondwanascale, in Evolution des Domaines Orogeniquesd’Asie Meridionale (de la Turquie a l’Indonesie),edited by P. Le Fort, M. Colchen, and C. Montenat,Sci. Terre, 47, 191–209.

Le Fort, P., S. Guillot, and A. Pecher (1997), HPmetamorphic belt along the Indus suture zone ofNW Himalaya: New discoveries and significance,C. R. Acad. Sci., 325, 773–778.

Le Goff, E., and M. Ballevre (1990), Geothermobarom-etry in the albite garnet orthogneisses: A case studyfrom the Grand Paradiso nappe, Western Alps,Lithos., 25, 261–280.

Maheo, G., H. Bertrand, S. Guillot, G. Mascle,A. Pecher, C. Picard, and J. de Sigoyer (2000),Les ophiolites Cretace du sud Ladakh (NWHimalaya, Inde): Temoins d’un arc immature intra-oceanique, C. R. Acad. Sci., 330, 289–295.

Mascle, G., G. Herail, T. Van Haver, and B. Delcaillau(1986), Structure et evolution des bassins d’episu-ture et de perisuture lies a la Chaıne Himalay-enne, Bull. Cent. Rech. Elf Explor. Prod., 10,181–203.

O’Brien, P. J., N. Zotov, R. Law, M. A. Khan, and M. Q.Jan (2001), Coesite in Himalayan eclogite andimplications for models of India-Asia collision,Geology, 29, 435–438.

Patriat, P., and J. Achache (1984), India-Eurasia colli-sion chronology has implications for crustal short-ening and driving mechanisms of plates, Nature,311, 615–621.

Platt, J. P. (1986), Dynamic of orogenic wedges and theuplift of high-pressure metamorphic rocks, Geol.Soc. Am. Bull., 97, 1037–1053.

Platt, J. P. (1993), Exhumation of high-pressure meta-morphic rocks: A review of concepts and processes,Terra Nova, 5, 119–133.

Pognante, U., and D. A. Spencer (1991), First record ofeclogites from the High Himalayan belt, Kaghanvalley (northern Pakistan), Eur. J. Mineral., 3,613–618.

Reuber, I., M. Colchen, and C. Mevel (1987), The geo-dynamic evolution of the South-Tethyan margin inZanskar, NW Himalaya, as revealed by the Spon-tang ophiolitic melange, Geodin. Acta, 1, 283–296.

Robertson, A. H. F. (2000), Formation of melanges inthe Indus suture zone, Ladakh Himalaya by succes-

sive subduction-related, collisional and post-colli-sional processes during Late Mesozoic-LateTertiary time, in Tectonics of the Nanga ParbatSyntaxis and the Western Himalaya, edited byM. A. Khan et al., Geol. Soc. Sped. Publ., 170,333–374.

Rolland, Y., J. M. Lardeaux, S. Guillot, and C. Nicollet(2000), Extension syn-convergence, poinconne-ment vertical et unites metamorphiques contrasteesen bordure Ouest du Grand Paradis (Alpes Franco-Italiennes), Geodin. Acta, 13, 133–148.

Rolland, Y., C. Picard, A. Pecher, H. Lapierre,D. Bosch, and F. Keller (2002), The CretaceousLadakh arc of NW Himalaya: Slab melting andmelt-mantle interaction during fast northward driftof Indian Plate, Chem. Geol., 182, 139–178.

Ruppel, C., L. Royden, and K. Hodges (1988), Thermalmodeling of extensional tectonics: Application topressure-temperature-time history of metamorphicrocks, Tectonics, 7, 945–957.

Sachan, H. K., B. K. Mukherjee, Y. Ogasawara,S. Mayurama, A. A. K. M. Pandey, N. Yoshioka,and H. Ishida (2001), Discovery of coesite fromIndian Himalaya: Consequences on Himalayantectonics, paper presented at Ultrahigh-PressureMinerals (UHPM) Workshop, Waseda Univ.,Tokyo, Japan.

Schlup, M., A. Carter, and A. Steck (2001), Exhuma-tion history revealed by fission track cooling ages inthe North Himalayan Crystalline Zone of easternLadakh, NW Himalaya, J. Asian Earth Sci., 19,56– 57.

Seeber, L., and A. Pecher (1998), Strain partitioningalong the Himalayan arc and the Nanga Parbat anti-form, Geology, 26, 791–794.

Spalla, I., J. M. Lardeaux, G. V. Dal Piaz, G. Gosso,and B. Messiga (1996), Tectonic significance ofalpine eclogites, J. Geodyn., 21/3, 257–285.

Spencer, D. A., and D. Gebauer (1996), Shrimp evi-dence for a Permian protolith age and a 44 Mametamorphic age for the Himalayan eclogites(Upper Kaghan, Pakistan): Implications for the sub-duction of Tethys and the subdivision terminologyof the NW Himalaya (abstract), in 11th Himalaya-Karakorum-Tibet Workshop, Flagstaff, Ariz., 28April – 1 May.

Steck, A., J. L. Epard, J. C. Vannay, J. Hunziker,M. Girard, A. Moraro, and M. Robyr (1998), Geo-logical transect across the Tso Morari and Spitiareas: The nappe structures of the Tethys Himalaya,Eclogae Geol. Helv., 91, 103–121.

Stutz, E., and A. Steck (1986), La terminaison occi-dentale du Cristallin de Tso Morari (Haut-Hima-laya; Ladakh meridional, Inde): Subdivision ettectonique de nappe, Eclogae Geol. Helv., 79,253–269.

Thakur, V. C. (1983), Deformation and metamorphismof the Tso Morari crystalline complex, in Geologyof Indus Suture Zone of Ladakh, edited by V. C.Thakur and K. K. Sharma, pp. 1– 8, Wadia Inst.of Himalayan Geol., Dehra Dun, India.

Thakur, V. C., and M. I. Bhat (1983), Interpretation oftectonic environment of Nidar ophiolite: A geo-chemical approach, Geology of Indus Suture Zoneof Ladakh, edited by V. C. Thakur and K. K.Sharma, pp. 33 – 40, Wadia Inst. of HimalayanGeol., Dehra Dun, India.

Thompson, A., K. Schulmann, and J. Jezek (1997a),Extrusion tectonics and elevation of lower crustalmetamorphic rocks on convergent orogens, Geol-ogy, 25, 491 –494.

Thompson, A. B., K. Schulmann, and J. Jezek (1997b),Thermal evolution and exhumation in obliquelyconvergent (transpressive) orogens, Tectonophysics,280, 171–184.

Treloar, P. J., R. D. Broughton, M. P. Williams, M. P.Coward, and B. F. Windley (1989), Deformation,metamorphism and imbrication of the Indianplate, south of the Main Mantle Thrust, northPakistan, J. Metamorph. Geol., 7, 111–126.

Trivedi, J. R., K. Kewal, Sharma, and K. Gopalan(1986), Widespread Caledonian magmatism inHimalaya and its tectonic significance, Terra Cog-nita, 6, 144.

Van Haver, T. (1984), Etude stratigraphique sedimento-logique et structurale d’un bassin d’avant arc: Ex-emple du Bassin de l’Indus, Ladakh, Himalaya,Ph.D. thesis, Univ. Joseph Fourier, Grenoble,France.

Van Haver, T., M. Bonhomme, G. Mascle, andJ. Aprahamian (1986), Analyses K/Ar de phyllitesfines des formations detritiques de l’Indus auLadakh (Inde): Mise en evidence de l’age eocenesuperieur du metamorphisme, C. R. Acad. Sci.,Ser. II, 302, 325–330.

Virdi, N. S., V. C. Thakur, and R. J. Azmi (1978),Discovery and significance of Permian microfossilsin the Tso Morari crystallines of Ladakh, India,Himalayan Geol., 8, 993–1000.

Weinberg, R. F., and W. J. Dunlap (2000), Growth anddeformation of the Ladakh batholith, northwestHimalayas: Implications for timing of continentalcollision and origin of calc-alkaline batholiths,J. Geol., 108, 303–320.

Yamamoto, K. (1995), Origin of blueschists-faciesclasts in the Mariana forearc, western pacific, Geo-chem. J., 29, 259–275.

���������J. de Sigoyer, Laboratoire de Geologie, CNRS-

UMR 8538, Ecole Normale Superieure de Paris, 24 rueLhomond, F-75231 Paris cedex 05, France (sigoyer@geologie.ens.fr)

P. Dick, Institut de Geologie, 11 E. Argand,Universite de Neuchatel, 2007 Neuchatel, Switzerland.

S. Guillot, Laboratoire de Sciences de la Terre,CNRS-UMR 5570, Universite Lyon I et Ecole NormaleSuperieure de Lyon, 2 rue Dubois, Batiment Geode,F-69622 Villeurbanne cedex, France.

TC3003 DE SIGOYER ET AL.: HP EXHUMATION PROCESSES IN THE HIMALAYAS

18 of 18

TC3003