Progression from South-Directed Extrusion to Orogen ...web.utk.edu/~mjessup/Jessup/Ama_Drime_Massif_files... · Cottle et al. 2009a), Leo Pargil dome (Thiede et al. 2006), and Gurla

[The Journal of Geology, 2010, volume 118, p. 467–486] � 2010 by The University of Chicago.All rights reserved. 0022-1376/2010/11805-0002$15.00. DOI: 10.1086/655011

467

Progression from South-Directed Extrusion to Orogen-ParallelExtension in the Southern Margin of the Tibetan Plateau,

Mount Everest Region, Tibet

Micah J. Jessup and John M. Cottle1

Department of Earth and Planetary Sciences, University of Tennessee, Knoxville, Tennessee 37996, U.S.A.(e-mail: [email protected])

A B S T R A C T

In the South Tibetan Himalaya, major detachment systems exhumed midcrustal rocks from different structural levelsthat are exposed in the Ama Drime and Mount Everest massifs. The South Tibetan detachment system (STDS)accommodated exhumation of the Greater Himalayan Series (GHS) until the Middle Miocene. Field- and laboratory-based structural data presented here indicate that early melt-present deformation in the footwall of the STDS ac-commodated large-scale flow of a low-viscosity middle crust. Decompression-related anatexis and emplacement ofleucogranites in the structurally highest positions of the GHS (∼16–18 Ma) mark the final stages of south-directedextrusion within a relatively narrow solid-state mylonite zone that progressed into a brittle detachment. Estimatesof mean kinematic vorticity and deformation temperatures record a progression in deformation conditions, from ultra-mylonitic leucogranite (∼400�–500�C; 51%–67% pure shear) to mylonitic marble (1300�C; 41%–55% pure shear) tomoderately foliated marble ∼200�–300�C (46%–59% pure shear). Previous investigations demonstrated that the ces-sation of movement on the STDS at ∼16–13 Ma was followed by anatexis in the lower portion of the middle crust(∼12–13 Ma; Ama Drime Massif). The Ama Drime and Nyonno Ri detachments (∼8–12 Ma; Dinggye graben) exhumedrocks from the deepest structural position in the central Himalaya during orogen-parallel extension. Brittle faultingdissected the upper crust (!4 Ma; Tingri–Kung Co graben) in a setting that was kinematically linked to extension inthe interior of the Tibetan plateau.

Introduction

Midcrustal rocks exposed along the southern mar-gin of the Tibetan plateau in the Mount Everestregion were exhumed by older, orogen-parallel de-tachments (e.g., South Tibetan detachment system[STDS]) and younger, orogen-perpendicular detach-ment systems (e.g., Ama Drime and Nyonno Ridetachment; fig. 1). The STDS is a system of low-angle shear zones and detachments that extends for12500 km along orogenic strike and consistentlyrecords top-down-to-the-north displacement (Burget al. 1984; Burchfiel et al. 1992). It juxtaposes theanatectic core of the Himalaya (Greater HimalayanSeries [GHS]) with the Tibetan Sedimentary Se-quence (TSS) in the hanging wall (Burg et al. 1984;Burchfiel et al. 1992). The STDS is generally inter-

Manuscript received June 11, 2009; accepted March 31, 2010.1 Department of Earth Science, University of California,

Santa Barbara, California 93106, U.S.A.

preted as either a low-angle normal fault (Burchfielet al. 1992) or a passive roof fault that forms theupper boundary to south-directed extrusion of theGHS (Beaumont et al. 2001; Searle et al. 2003; Lawet al. 2004; Jessup et al. 2006; Cottle et al. 2007).Alternatively, tectonic wedging models interpretthe STDS as a passive roof thrust with severalphases of reactivation (Webb et al. 2007).

Migmatite-cored domal structures, such as theAma Drime Massif (ADM; Jessup et al. 2008b;Cottle et al. 2009a), Leo Pargil dome (Thiede et al.2006), and Gurla Mandhata core complex (Murphyet al. 2002), offset and/or reactivate the STDS (fig.1, inset). The ADM reached granulite facies meta-morphic conditions (∼750�C and 0.7–0.8 GPa) at∼12–15 Ma before being exhumed during orogen-parallel extension (fig. 2; Jessup et al. 2008b; Cottleet al. 2009a). Relatively narrow fault zones con-taining solid-state fabric development overprinted

468 M . J . J E S S U P A N D J . M . C O T T L E

Figure 1. Simplified interpretive block diagram of the Mount Everest region, Tibet-Nepal, adapted from Jessup etal. (2008b). Inset map shows location of Namche Barwa, Ama Drime, Gurla Mandhata, Leo Pargil, Gurla Mandhata,and Nanga Parbat. ADD, Ama Drime detachment; NRD, Nyonno Ri detachment; STDS, South Tibetan detachmentsystem; MCTZ, Main Central Thrust zone.

early stage, melt-present deformation fabrics dur-ing exhumation (Langille et al. 2010). The easternlimb of the ADM is bound by ductile shear zonesand faults that also define the western margin ofthe Dinggye graben (Zhang and Guo 2007).

Two (Tingri–Kung Co and Dinggye) graben thatextend from the Himalaya into the interior of theTibetan plateau (Taylor et al. 2003) are locatedwithin the Mount Everest region (Armijo et al.1986; Maheo et al. 2007). The graben are kinemat-ically linked to extension in the interior of the Ti-betan plateau (Taylor et al. 2003; Kapp and Guynn2004). The interaction between the STDS, domes,and graben provides critical information regardingthe progression from a kinematic setting domi-nated by south-directed extrusion of midcrustalrocks along the STDS to orogen-parallel extensionthat is accommodated by orogen-perpendicular nor-mal faults and detachments. To explore this aspectof the evolution of the Himalayan-Tibetan plateauorogenic system, we integrate field-based structuraldata with microstructural and kinematic analysisto characterize two sections of the STDS that areoffset by the Tingri–Kung Co graben (Maheo et al.2007; fig. 2). New results are combined with exist-

ing data from the GHS and ADM to provide insightinto the role of crustal melting and strain parti-tioning during the progression from south-directedextrusion to orogen-parallel extension.

Background

South Tibetan Detachment System. In the MountEverest area, the STDS includes two low-angle,high-strain zones: the structurally lower ductileLhotse detachment (LD) and structurally higherbrittle Qomolangma detachment (QD; Carosi et al.1998, 1999; Searle 1999b; Law et al. 2004; Sakai etal. 2005; Cottle et al. 2007; Jessup et al. 2008a). Thesouthernmost exposure of the QD occurs on thesummits of Mount Everest (8850 m), Changtse(7583 m), and Cheng Zheng (6977 m). From thesummit of Cheng Zheng, the QD dips gently to thenorth and is exposed along the sides of, and drain-ages into, Rongbuk Valley (Burchfiel et al. 1992;Carosi et al. 1998; Murphy and Harrison 1999;Searle et al. 2003; Law et al. 2004; Jessup et al. 2006;fig. 1). Between these structures lies a wedge ofEverest Series, which includes greenschist to am-phibolite facies schist and calc-silicate (Lombardo

Journal of Geology P R O G R E S S I O N T O O R O G E N - P A R A L L E L E X T E N S I O N 469

Figure 2. Faded Landsat 7 image with overlay of major fault systems in the Mount Everest region (Armijo et al.1986; Burchfiel et al. 1992; Maheo et al. 2007; Zhang and Guo 2007; Jessup et al. 2008b). South Tibetan detachment(red line), extensional shear zones and normal faults (black line), normal faults (yellow line). Ra Chu and Gondasampatransects highlighted by open white circles.

et al. 1993; Carosi et al. 1998, 1999; Searle 1999a,1999b; Jessup et al. 2008a; Cottle et al. 2009b).

These two detachments merge near the northernlimit of exposure in Rongbuk Valley (Carosi et al.1999; Searle et al. 2003). Mylonite developmentwithin the upper ∼300 m of the footwall gneiss,leucogranite, calc-silicate, and marble occurred at∼17 Ma and progressed into a brittle detachmentat ≤16 Ma (Hodges et al. 1998; Murphy and Har-rison 1999; Searle et al. 2003). The 300-m-thickductile shear zone records telescoping of isotherms(Law et al. 2009) and a progression in deformationconditions during exhumation (Law et al. 2004;Jessup et al. 2006). Detailed kinematic and vorticityanalysis indicates that during the early stages(higher temperature) exhumation was accommo-dated within the upper 300 m of the GHS (∼62%–35% pure shear). As exhumation continued, defor-mation was partitioned into the overlying marble/limestone (45%–28% pure shear), followed by dis-placement on the brittle detachment (Law et al.2004; Jessup et al. 2006).

The trace of the STDS toward the northeast, de-fined by field mapping and Landsat7 image inter-pretation, extends from Rongbuk Valley acrossDoya La and into the Dzakaa Chu Valley (fig. 2).Here, the STDS is significantly thicker (∼1 km)than exposures in Rongbuk Valley (Cottle et al.2007). U(-Th)-Pb monazite geochronology on apostkinematic leucogranite dike that crosscuts themylonitic foliation yields a crystallization age of∼20 Ma and provides a minimum age for high-temperature fabric development in the deepest por-tion of the shear zone (Cottle et al. 2007). Unlikeexposures in Rongbuk and Dinggye, no discretelayer-parallel brittle detachment was observed. In-stead, a right-way-up apparent metamorphic fieldgradient of at least 260�C/km—1 across the entireshear zone is interpreted to represent excision ofmaterial and telescoping of isotherms within a pas-sive roof fault during south-directed extrusion(Cottle et al., forthcoming).

The STDS is deflected around the ADM and ex-posed in Dinggye Valley (fig. 2; Burchfiel et al. 1992;


Cottle et al. 2007). The Dinggye detachment is par-allel to the pervasive fabric in the amphibolite fa-cies GHS of the footwall as well as the greenschistfacies calc-silicate rocks in the hanging wall(Burchfiel et al. 1992). The Sa’er detachment is a1- to 110-m-thick breccia zone that juxtaposes am-phibolite facies marble and thin-bedded quartzitein the footwall with unmetamorphosed carbonaterocks that lack a penetrative fabric in the hangingwall. The Dinggye and Sa’er detachments were ac-tive at deeper and shallower structural levels, re-spectively (Burchfiel et al. 1992). 40Ar/39Ar ther-mochronometry in Dinggye Valley records rapidexhumation along the STDS between ∼13 and 16Ma (Hodges et al. 1994).

Ama Drime and Nyonno Detachments. The ADMis an antiformal structure located ∼50 km northeastof Mount Everest (figs. 1, 2) that is bounded to thewest and east by the Ama Drime and Nyonno Ridetachments, respectively (Jessup et al. 2008b). Thepredominant composition of the ADM is migma-titic orthogneiss with lenses of garnet amphiboliteand granulitized eclogite that are boudinaged andfolded (Groppo et al. 2007; Jessup et al. 2008b;Cottle et al. 2009a). Granulite facies metamor-phism (750�C and 0.7–0.8 GPa) overprinted earlyeclogite-facies metamorphism and was followed byin situ partial melting at ! Ma (Groppo13.2 � 1.4et al. 2007; Cottle et al. 2009a). Postkinematic leu-cogranite dikes that crosscut the pervasive mig-matitic foliation and boudinaged mafic lenses inthe central portion of the dome provide a maximumage for fabric development and melt-present defor-mation at Ma during orogen-parallel ex-11.6 � 0.4tension (Cottle et al. 2009a). Low-temperature (U-Th)/He apatite thermochronometry conducted onsamples collected along two transects across therange yield a minimum exhumation rate of ∼1 mm/yr between 1.5 and 3.0 Ma (Jessup et al. 2008b).

Structural, petrological, and isotopic evidencefrom the Ama Drime detachment (ADD) indicatesthat the range-bounding shear zone is potentiallya branch of the Main Central Thrust zone (Lom-bardo et al. 1998, 2000; Lombardo and Rolfo 2000;Groppo et al. 2007; Liu et al. 2007; Langille et al.2010). This implies that (1) some portion of theADM may be part of the lowermost GHS or perhapsLesser Himalayan Sequence (Lombardo et al. 1998;Visona et al. 2000; Visona and Lombardo 2002;Groppo et al. 2007) and (2) the ADD potentiallyexperienced an early history as a thrust fault thatwas subsequently reactivated during orogen-paral-lel extension (Jessup et al. 2008b; Langille et al.2010).

The ADD juxtaposes upper amphibolite facies

GHS rocks in the hanging wall from granulite faciesAma Drime orthogneiss in the footwall. Polyphasefolding is overprinted by the main shear zone fabricthat strikes NNE–SSW, dips 4�–66�W, and containsa well-developed, approximately down-dip stretch-ing lineation (Jessup et al. 2008b). The ADD is com-posed of interlayered leucogranite, quartzite, calc-silicate, and marble that consistently recordtop-down-to-the-west sense of shear (Jessup et al.2008b). Estimates of deformation temperature ob-tained using quartz microstructures, strain-inducedasymmetric myrmekite, crystallographic slip sys-tems, and quartz c-axis opening angle indicate thattop-down-to-the-west shear sense dominated be-tween 400� and 650�C and suggest a minimumdetachment-parallel displacement of ∼21–42 km(Langille et al. 2010).

The Nyonno Ri detachment (NRD) is a 100–300-m-thick zone of pervasively foliated orthogneissand leucogranite that strike NNE–SSW, dip ∼45�E,and contain a down-dip stretching lineation (Burch-fiel et al. 1992; Jessup et al. 2008b). Asymmetrictails on feldspar porphyroclasts, S-C fabric, andshear bands within the footwall record top-down-to-the-east sense of shear (Jessup et al. 2008b). TheNRD offsets the STDS by 20 km of right-lateralseparation (Burchfiel et al. 1992; Hodges et al. 1994;Liu et al. 2007; Zhang and Guo 2007; Jessup et al.2008b). Quaternary deposits are offset by normalfaults located along the base of the triangular facetsand record top-down-to-the-east displacement,which is younger than extensive moraine depositson the eastern limb.

Dinggye Graben. While major shear zone and de-tachment systems (i.e., ADD and NRD) accom-modated orogen-parallel extension largely withinthe middle crust, brittle normal faults were formingin the upper crust. One such example is the Pumqu-Xianza-Dinggye graben system (Armijo et al. 1986;Burchfiel et al. 1992; Zhang and Guo 2007) thatextends from the margins of the ADM into the in-terior of the Tibetan plateau, where it intersectsthe right-lateral Gyaring Co fault (Taylor et al.2003). Near Sa’er, the western margin of the Ding-gye graben is defined by the NRD and normal faults(fig. 2; Burchfiel et al. 1992; Zhang and Guo 2007;Jessup et al. 2008b). The hanging wall is composedof GHS that is juxtaposed with the TSS along theSTDS. This sequence is rotated and dissected by aset of southeast- and northwest-dipping normalfaults (Burchfiel et al. 1992). 40Ar/39Ar thermo-chronometry from the Xainza-Dinggye rift (∼8–13Ma) are younger than movement on the STDS (∼24–13 Ma; Hodges et al. 1994) and provide further evi-dence for a transition from motion on the STDS to


the NRD at ∼13 Ma (Hodges et al. 1994; Zhang andGuo 2007).

Tingri–Kung Co Graben. The Tingri–Kung Cograben is the southernmost portion of the Tangra–Yum Co graben that extends from near the northface of Cho Oyo (8201 m) northward into the in-terior of the Tibetan plateau (Armijo et al. 1986).The Kung Co half-graben crosscuts the western por-tion of the Burta Range (Armijo et al. 1986; Maheoet al. 2007). The footwall experienced shallow ther-mal relaxation and cooling (16 and 7.5 Ma) that wasfollowed by increasing exhumation rates from 7.5Ma to the present. Because the north-south-strikingfault offsets the Gyirong-Kangmar thrust and lacksductile fabric development, it is interpreted as abrittle upper-crustal normal fault that formed at !4Ma (Maheo et al. 2007). The southern terminationof the Kung Co fault occurs near the Tsamda hotspring fissure ridge, which formed in response totectonically induced hydrofracturing during exten-sion (Armijo et al. 1986; Hoke et al. 2000; Newellet al. 2008). A dextral en-echelon array of travertineridges (trend N ) records oriented10� � 11� j � 3perpendicular to the fracture plane (Armijo et al.1986). Water issuing from these faults was shal-lowly circulated in carbonate-rich country rock(Hoke et al. 2000; Newell et al. 2008).

Gondasampa Transect on the South TibetanDetachment System

The Gondasampa Transect is located on the west-ern side of a north-south-striking ridge that extendsinto the central portion of the Tingri graben (fig.2). The ridge rises toward the southwest, wherefootwall rocks of the STDS are exposed along aprominent ridgeline that extends into the interiorof the Lapchi range (fig. 2).

Mesoscale Observations. The structurally highestposition reached during this transect (∼5890 m) iscomposed of mylonitic marble (Cal � Ca-Plg � Dol� Di � Qtz � phyllosilicates). It contains a per-vasive foliation (representative; 100�, 11� NE) andmoderate lineation (representative; ) that10� r 022�are defined by aligned calcite grains. Single grainsand aggregates of diopside within a matrix of calcitehave long axes that are oriented at various anglesto the main foliation. The foliation is deflectedaround angular blocks of mylonitic leucogranitethat are suspended in the marble.

Structurally below the marble, ribbons of dynam-ically recrystallized quartz and white mica definea pervasive S-C fabric in mylonitic leucogranite(representative; 085�, 29� NW). Elongate quartz andfeldspar create a well-developed stretching linea-

tion (representative; ). Together these25� r 031�record top-down-to-the-northeast sense of shear.Narrow (∼4 cm) ultramylonite zones contain rigidfeldspar porphyroclasts that are set within a matrixof dynamically recrystallized quartz and whitemica (Jessup et al. 2007). Quartz and mica are de-flected around rigid feldspar porphyroclasts and de-fine a prominent S-C fabric (fig. 3B). These are pos-sible equivalents to mylonitic leucogranitesexposed in Rongbuk Valley (Burchfiel et al. 1992;Law et al. 2004; Jessup et al. 2006, 2007). Less abun-dant quartz-rich layers with a pervasive foliation(representative; 081�, 31� NW) and stretching lin-eation (representative; ) that are defined26� r 030�by slightly anastomosing ribbons of quartz withrare isolated feldspar porphyroclasts are alsopresent.

Amphibolite facies (Bt � Kfs � Hbl � Grt) Rong-buk Formation gneiss, located structurally belowthe leucogranite, preserves a range of deformationconditions. Near the base of the leucogranite,gneiss contains a pervasive foliation that is definedby aligned biotite and quartz (representative; 069�,31� NW) and a weak lineation. Biotite-rich layersalternate with quartz- and feldspar-rich layers.Feldspar in these layers lack evidence of internaldeformation, are clustered and, in places, separatedby ribbons of dynamically recrystallized quartz thatdefine shear bands. Toward structurally deeper po-sitions, rocks contain semirigid feldspar and an in-creasing percentage of leucosomes. Large (14 cm #4 cm) feldspar porphyroclasts are rigid with minorsymmetrical tail development, while smaller feld-spar porphyroclasts (!1 # 2 cm) are elongate andaligned with the main fabric (fig. 3C). The deepeststructural position of this transect contains mig-matitic gneiss that records melt-present deforma-tion. Leucogranite bodies are oriented parallel tofoliated host rock (fig. 3D). Shear bands are com-mon (fig. 3E). Garnet porphyroblasts occur in themigmatitic gneiss (fig. 3F). Garnet porphyroblastsare also found in relatively high concentrationswithin a single pelitic layer (fig. 3G).

Microscale Observations. Marble at the structur-ally highest position reached during this investi-gation (G05-05) contains aligned, elongate calcitegrains that define the main foliation (table 1). Thefoliation wraps around diopside and feldspar por-phyroclasts. Mica fish record top-down-to-the-northeast sense of shear. Some calcite contains thin(!1 mm) Type I twins (Burkhard 1993; Ferrill et al.2004) that indicate temperatures !170�–200�C (fig.4A). Thick (11 mm) Type II twins are also presentand indicate temperatures between 200� and 300�C(fig. 4B; Weber et al. 2001; Ferrill et al. 2004). These

Figure 3. Images from the Gondasampa Transect, Tibet. A, South Tibetan detachment exposed on the summit ridgeof transect viewed toward the southeast. B, Hand sample of ultramylonitic leucogranite with white feldspar porphy-roclasts in a matrix of dynamically recrystallized quartz and mica. C, Vertical outcrop of semirigid feldspar in dy-namically recrystallized matrix. D, Migmatitic gneiss in foreground with Tingri graben in background. E, Represen-tative migmatitic gneiss with shear band. F, Garnet porphyroblast in gneiss. G, High-resolution scan of a thin sectionof G05-10.


Table 1. Ra Chu and Gondasampa Transects

Sample Rock type Elevation (m) Wm

Deformationtemperature

(�C)

Ra Chu transect (lat 28�21.318N, long 086�40.342E at highest sample):T07-01 Marble 4860 ... 200–300T07-02 Marble 4855 .68–.75 200–300T07-03 Marble 4854 .60–.65 200–300T07-04 Marble 4853 .65–.73 1300T07-05 Marble 4851 .70–.80 1300T07-06 Marble 4849 .66–.80 1300T07-07 Marble 4847 ... 1300T07-08 Calc-silicate 4846 ... 280–400T07-09 Calc-silicate 4845 ... 280–400T07-10 Calc-silicate 4843 ... 280–400T07-12 Calc-silicate 4841 ... 400–500T07-13 Calc-silicate 4841 ... 400–500T07-14 Calc-silicate 4839 ... 400–500T07-15 Calc-silicate 4836 ... 400–500T07-16 Leucogranite 4834 ... 1500T07-17 Calc-silicate 4832 ... 1500T07-18 Calc-silicate 4830 ... 1500

Gondasampa transect (lat 28�24.521N, long 086�30.052E at highest sample):G05-05 Marble 5869 ... 200–300G05-04 Qtz vein 5864 ... 400–500G05-03 Leucogranite 5765 .60–.70 400–500G05-02 Leucogranite 5695 ... 400–500G05-01 Qtz vein 5695 .50–.60 400–500G05-06 Bt-schist 5885 ... 1500G05-07 Grt-Bt schist 5865 ... 1500G05-08 Grt-Bt schist 5865 ... 1500G05-09 Augen gneiss 5828 ... 1500G05-10 Grt-Bt-Sill schist 5806 ... 1500G05-11 Grt-Bt schist 5816 ... 1500

Note. Samples are in order of decreasing structural position.

samples define a range of deformation temperatures(200�–300�C) at the structurally highest position ofthis transect.

Layers (∼30 cm) of ultramylonite (G05-01 andG05-03) are oriented parallel to the main foliationin the leucogranite and contain multiple kinematicindicators. Rigid feldspar porphyroclasts with j-and d-type asymmetric tails are suspended in a ma-trix of fine-grained recrystallized quartz and whitemica (fig. 4C). Group 4 (Grotenhuis et al. 2002)white mica fish were formed by removal of materialalong C′ shear bands. Aligned white mica or do-mains of dynamically recrystallized quartz defineS-C and C′ shear bands. Quartz-rich layers containa well-developed oblique grain shape fabric definedby aligned recrystallized quartz grains (fig. 4C) andare locally disturbed by asymmetric folds. Theseindependent shear sense indicators consistentlyrecord top-down-to-the-northeast sense of shear.Quartz-rich domains are interlayered with veryfine-grained biotite and white mica. Quartz grainboundaries record subgrain rotation recrystalliza-tion (SGR; 400�–500�C; Stipp et al. 2002).

At intermediate structural positions (G05-06),feldspar porphyroclasts are rigid and set within amatrix of quartz that records grain boundary mi-gration recrystallization (GBM; fig. 4D). Asymmet-ric biotite tails on rigid feldspar record top-to-the-northeast sense of shear. Biotite also definesextensional shear bands that confirm this sense ofshear. Myrmekite is present in some K-feldspargrains. In one example, blades of plagioclase withvermicular quartz are oriented ∼90� to the mainfoliation (fig. 4D). Quartz grain boundaries recordGBM recrystallization (1500�C; Stipp et al. 2002).

Large K-feldspar porphyroclasts are semirigid(G05-10/11) in the lowest structural position of thetransect. Polygonal intergrowths of plagioclase andquartz record grain boundary area reduction duringhigh-temperature annealing. Biotite laths arewrapped around rigid feldspar porphyroclasts andare interlayered with sillimanite. Fibrolite andelongate blades of sillimanite are present (fig. 4E).Kyanite inclusions (∼465 mm # 90 mm) are presentin K-feldspar grains (fig. 4F). Garnet porphyroblastsare elongate (4 mm # 2 mm), and the outer ∼50

Figure 4. Photomicrographs of samples from the Gondasampa transect. A, Type I deformation twin in calcite (cal).B, Type II deformation twins in calcite with phlogopite (phl) and diopside (di) porphyroblast. C, Mylonitic leucogranitewith rigid feldspar (fld), muscovite (ms) fish, and dynamically recrystallized quartz (qtz) matrix. Main foliation (Sa)and oblique grain shape fabric (Sb) are present. D, Semirigid K-feldspar porphyroclast in matrix of recrystallized quartzand plagioclase. Myrmekite is marked by intergrowths of vermicular quartz and plagioclase. E, Prismatic sillimanitein garnet � biotite � kyanite schist (G05-10). F, Kyanite inclusion in K-feldspar grain (G05-10).


mm of the porphyroblast is inclusion free. Towardmore internal sections of the grain, an ∼300-mm-thick zone of opaque inclusions is followed by rel-atively inclusion-free garnet. Rare, subhedral kya-nite is intergrown with sillimanite and is the firstdocumentation of kyanite from the footwall of theSTDS in the central Himalaya.

Ra Chu Transect on the South TibetanDetachment System

The Ra Chu section of the STDS is exposed alonga prominent ridgeline that rises above the Ra Chu(river) on the road from Tingri to Rongbuk Valley(fig. 5; Pertusati et al. 2003). Toward the south, theSTDS can be projected to Cho Oyo (8201 m). At itswesternmost exposure, the STDS is abruptly trun-cated by normal faults that define the eastern mar-gin of the Tingri–Kung Co graben (Armijo et al.1986).

Mesoscale Observations. Two ridges provide ex-posure of the uppermost section of the STDS (fig.5). An ∼15-m-thick zone of cataclasitic Lower Pa-leozoic TSS juxtaposes Upper Carboniferous terrig-enous sediments above from marble below (Per-tusati et al. 2003). An ∼5-m-thick zone ofmoderately foliated marble (representative; 129�,24� NE) defines the base of the cataclasite zone (fig.5B). Toward deeper structural positions below thecataclasite (15 m), the marble (Cal � Ca-Plg � Dol� Di � Qtz � phyllosilicates) becomes penetra-tively deformed and preserves a well-developed fo-liation (representative; 150�, 24� NE) and lineation(representative; ; fig. 5B). When viewed20� r 030�parallel to the foliation and perpendicular to thelineation (XZ section), quartz, feldspar, and diop-side grains are suspended in the calcite matrix. Leu-cogranite and calc-silicate mark the base of thissection, which creates the major break in slope be-tween light gray sheared marble above and deeporange calc-silicate and leucogranite below (fig.5C).

Calc-silicate (Cal � Ca-Plg � Dol � Di � Qtz �phyllosilicates) with a pervasive foliation (repre-sentative; 175�, 16� NE) and lineation (representa-tive; ) marks the base of the mylonitic04� r 014�marble unit. Where exposed on the top of foliationsurfaces, the foliation within the calc-silicate wrapsaround leucogranite pods (fig. 5C). Although Per-tusati et al. (2003) document sheath folds in thecalc-silicate, we were unable to confirm their pres-ence and instead prefer to interpret these as isocli-nal folds with strongly attenuated limbs (fig. 5D).Pods and lenses of mylonitic leucogranite with awell-developed foliation (representative; 144�, 6�

NE) and lineation (representative; ) are02� r 004�folded and boudinaged. Elongate tourmaline withinmylonitic leucogranite is exposed in orthogonalsections. Tourmaline is aligned so that it appearsas elongate, boudinaged clusters (fig. 5E), whereasin sections that are oriented perpendicular to thestretching lineation, the terminations of elongatetourmaline are exposed (fig. 5F).

Pseudotachylite crosscuts the pervasive foliationwithin mylonitic leucogranite at a high angle (fig.5G). The contact between the pseudotachylite andthe leucogranite is sharp. Angular to rounded leu-cogranite fragments are suspended in the thickerportions of the vein. An intensification of the fo-liation defines localized high-strain regions withinthe leucogranite (fig. 5H). Strain gradients occur onvarious scales and record the interplay betweenstrain rate, lithology, and fluids.

Microscale Observations. Samples from the high-est structural position (T07-01-03) are predomi-nantly composed of calcite that defines a moder-ately developed foliation with a moderate lineation(table 1). Quartz grains are set within a calcite ma-trix (fig. 6A). The lack of undulose extinction inquartz indicates that they are undeformed. Quartzgrains are generally dispersed throughout the sam-ple and only occasionally form clusters of severalgrains. Calcite grains preserve thin Type I twins (fig.6A) and tabular Type II thick twins (fig. 6B). Rarecurved and tapered Type III twins are also present.Types II and III indicate temperatures 1200�C (Burk-hard 1993; Ferrill et al. 2004). Since there is noevidence for dynamic recrystallization (1300�C;Weber et al. 2001), these bracket the temperatureof deformation between 200� and 300�C (fig. 7).White mica fish consistently record top-down-to-the-northeast sense of shear.

A decrease in the grain size of calcite occurs atan intermediate structural position between sam-ples T07-03 and T07-04. Beneath this threshold,calcite twins are absent (fig. 6C–6E). The foliation,defined by elongate calcite, is wrapped around rigiddiopside and feldspar porphyroclasts. Diopside por-phyroblasts presumably grew during an earlier,higher-temperature event (Cottle et al., forthcom-ing) and were subsequently rotated during shearingat lower deformation temperatures. Quartz grainscontain undulose extinction, and some are lens-oidal and aligned parallel to the main foliation (fig.6D). Lack of twins indicates complete dynamic re-crystallization of calcite at temperatures 1300�C(fig. 7; Weber et al. 2001; Ferrill et al. 2004). Feldsparand diopside porphyroclasts have d and j tails andlong axes that are oriented at a range of angles fromthe foliation (fig. 6C, 6D).

476

Figure 5. Field images of the Ra Chu transect. A, Ra Chu is located at the base of the image and flows from thenorthern slopes of Everest into the Tingri graben. Total structural thickness of the transect is ∼30 m. Lower-hemisphereequal area stereonet of foliation and stretching lineation from transect ( ). B, Mylonitic marble located at then p 27top of the transect. C, Well-foliated calc-silicate warped around pod of leucogranite. View is oblique across the foliation.D, Folded calc-silicate at the base of transect. E, F, Mylonitic leucogranite with boudinaged tourmaline viewed inorthogonal exposures. Black line work highlights the orientation of the foliation and stretching lineation. G, Pseu-dotachylite crosscutting mylonitic leucogranite viewed in vertical outcrop. H, Strain gradient in mylonitic leucograniteviewed in vertical outcrop.

477

Figure 6. Representative photomicrographs of samples from the Ra Chu section. A, Type I deformation twins incalcite (cal) from structurally highest sample of the mylonitic marble section (T07-01). B, Type II twins in calcitefrom intermediate structural position in mylonitic marble section (T07-02). C, Dynamically recrystallized calcite inthe lower portion of the mylonitic marble section (T07-05). Diopside (di) and feldspar (fld) porphyroclasts are rigidobjects in a fine-grained calcite matrix with a pervasive foliation (Sa). D, Rigid diopside porphyroclast with d-typetails in the dynamically recrystallized calcite matrix (T07-07). E, Diopside grains in the structurally highest sectionof the calc-silicate portion of the transect (T07-08). F, Dynamically recrystallized quartz (qtz) and muscovite (ms)grains from a mylonitic leucogranite within the calc-silicate (T07-16).


Figure 7. Mean kinematic vorticity estimate (Wm) andestimates of deformation temperatures for the upper ∼20m of the Ra Chu transect. Arrows indicate minimumtemperature estimates for dynamically recrystallizedcalcite.

The top of the calc-silicate section is near thelocation of sample T07-08. This sample contains agreater concentration of diopside and plagioclaseporphyroclasts that are set within a very fine-grained calcite matrix. Porphyroclasts are closelypacked and lack tails that are common in struc-turally higher positions. Calcite grains in the ma-trix lack twins and indicate dynamic recrystalli-zation at (Ferrill et al. 2004). GrainT 1 300�Cboundaries in quartz-rich layers are bulging intoeach other, suggesting that bulging recrystalliza-tion (BLG) was the dominant mechanism at 280�–400�C (Stipp et al. 2002).

Calc-silicate extends to deeper structural posi-tions (T07-10) where bulging quartz grain bound-aries are still dominant with minor developmentof SGR. As structural depth increases (T07-12),SGR becomes dominant (400�–500�C; Stipp et al.2002). A lens of leucogranite (T07-16) is transitionalbetween SGR and GBM (1500�C; Stipp et al. 2002).Mica fish, S-C fabric, and sigma-type asymmetrictails on feldspar porphyroclasts define top-to-the-northeast sense of shear. The deepest structural po-sition exposed in the Ra Chu transect (T07-18) con-tains amoeboidal quartz grain boundaries thatrecord GBM (1500�C).

Mean Kinematic Vorticity

Vorticity analysis on samples from within theSTDS and MCTZ attests to its applicability to shearzones in the Himalaya (Grasemann et al. 1999; Law

et al. 2004; Carosi et al. 2006; Jessup et al. 2006,2007; Larson and Godin 2009). Estimating the rel-ative contributions of pure and simple shear (vor-ticity) during exhumation of the GHS is an impor-tant aspect of this investigation because asignificant contribution of pure shear predicts anincrease in extrusion and exhumation rates relativeto simple shear dominated flow (Law et al. 2004).

Kinematic vorticity number (Wk) is a direct es-timate of pure ( ) and simple ( ) shearW p 0 W p 1k k

during plane strain deformation (Means et al. 1980).Equal contributions of pure and simple shear occursat (Law et al. 2004). Mean kinematicW p 0.71k

vorticity (Wm) uses a time-averaged Wk to accom-modate for variability of vorticity during non–steady state deformation (Fossen and Tikoff 1997,1998; Jiang 1998). Plane strain deformation hasbeen demonstrated for the STDS exposed in Rong-buk Valley (Law et al. 2004). Vorticity estimates(two dimensions) on rocks that deviate from planestrain yield values that may be overestimated by!0.05 (Tikoff and Fossen 1995). This is minimalcompared with other errors associated with Wm

measurement used during this investigation. Wm

was estimated by applying the rigid grain technique(Passchier 1987; Wallis et al. 1993; Jessup et al.2007) to samples of mylonitic marble and leucogra-nite from the immediate footwall of the STDS.

Models demonstrate that during simple shear( ), rigid objects with a range in aspect ratiosW p 1m

(R) will rotate infinitely in a ductile matrix. Witha higher contribution of pure shear ( ),0 ! W ! 1m

porphyroclasts will rotate until they reach a stable-sink (Rc) position that is defined by a unique com-bination of the aspect ratio (R), the acute angle fromthe foliation (v), and Wm (Jeffrey 1922; Ghosh andRamberg 1976; Passchier 1987; Simpson and DePaor 1993):

2R � 1cW p .m 2R � 1c (1)

The rigid grain net (RGN) plots the shape factor(B∗) versus v on a theoretical net for combinationsof B∗ and v at a range in Wm values:

2 2M � Mx n∗B p ,2 2M � Mx n (2)

where Mx and Mn are the long and short axes of theellipse, respectively (Passchier 1987). The RGN un-ifies existing methods that use rigid porphyroclastsin a flowing matrix (Passchier 1987; Wallis 1992;Simpson and De Paor 1993, 1997; Wallis et al.1993), simplifies plotting of data, and limits am-


biguity in estimating Wm by enabling the user tocompare complex natural data sets with the theo-retical net (Jessup et al. 2007).

Samples analyzed in this study using the rigidgrain method were deemed appropriate if they metthe following criteria (Passchier 1987): (1) deformedhomogeneously, (2) rigid porphyroclasts signifi-cantly larger than the matrix grains, (3) high finitestrains to rotate objects toward stable-sink posi-tions, (4) porphyroclast shape that is close to or-thorhombic, (5) significant number of porphyro-clasts with a range in B∗, (6) porphyroclast thatpredates the deformation fabric, and (7) no me-chanical interaction between grains. For an over-view of systematic source error estimation for therigid grain method, see Iacopini et al. (2008).

Only two samples from the Gondasampa transectpassed the criteria for rigid grain vorticity analysis(G05-01 and G05-03). These are ultramylonitesfrom within the mylonitic leucogranite near theupper portion of the STDS. G05-01 and G05-03were the basis for creating the RGN, with originalinterpretations defining a range of Wm estimates of0.57–0.60 and 0.65–0.70, respectively (Jessup et al.2007). In order to maintain consistency with theapproach for determining the critical thresholdused by this investigation, we prefer to use abroader range of values (G05-01, Wm p 0.50–0.60;G05-03, Wm p 0.60–0.70) for the critical thresholdvalues using data from both positive and negativesides of the RGN. Together these samples define arange of Wm estimates (0.50–0.70; 51%–67% pureshear) for high-strain ultramylonite in leucogranitethat occurred at deformation temperatures of∼400�–500�C.

Mylonitic marble from the upper ∼10 m of theRa Chu transect proved to be well suited for rigidgrain vorticity analysis using the RGN. These sam-ples contain rigid diopside, feldspar, and quartz ina matrix of calcite. Johnson et al. (2009a, 2009b)proposed that an envelope of phyllosilicates aroundporphyroclasts can potentially decouple it from thematrix and yield vorticity estimates that underes-timate the simple shear component. The lack ofphyllosilicates around the porphyroclasts in thesemylonitic marbles indicates that the potential fordecoupling at the clast/matrix interface was min-imal. The structurally highest of these samples(T07-02; ∼5 m below detachment) yields a Wm es-timate of 0.68–0.75 (figs. 7, 8A; 46%–52% pureshear). The next structurally lower sample (T07-03;∼6 m below the top of the section) yields a Wm

estimate of 0.60–0.65 (55%–59% pure shear) andrepresents the upper portion of penetratively de-

formed marble (figs. 7, 8D). These two samples rec-ord the lowest deformation temperatures estimatedduring this investigation (200�–300�C). SamplesT07-04, T07-05, and T07-06 were collected frombetween 7.5 and 11 m below the detachment in azone of sheared marble that contains a pervasivefoliation and lineation and dynamically recrystal-lized matrix calcite (1300�C; fig. 7). These samplesyield a range in Wm estimates of 0.65–0.80 (41%–55% pure shear) at deformation temperatures1300�C (fig. 8C–8E). Structurally below T07-07,calc-silicate becomes the dominant rock type, andbecause of grain interaction, these samples are un-suitable for rigid grain vorticity analysis.

Discussion

Kinematic Evolution of the South Tibetan Detach-ment System. Data from two transects (Gonda-sampa and Ra Chu) across the STDS exposed oneither side of the Tingri–Kung Co graben provideimportant new constraints on the south-directedextrusion and exhumation of the GHS (figs. 2, 3).Melt-present flow dominates the deepest structuralpositions in the Gondasampa transect and recordsthe early stages of south-directed flow at high de-formation temperatures. At decreasing structurallevels, melt-present deformation is overprinted bysolid-state fabric development that is recorded bymylonitic and ultramylonitic marble, calc-silicate,and leucogranite. We interpret this as a transitionfrom early, high-temperature melt-present flow tolater, lower-temperature deformation that was ac-tivated during progressive south-directed exhu-mation along the STDS.

Microstructures from the Gondasampa and RaChu transects provide important qualitative esti-mates of deformation temperatures. Types I and IItwins in calcite are preserved in the uppermost sec-tion of the transects and indicate that deformationtemperatures reached ∼200�–300�C at this struc-tural level (fig. 7). Beneath this, calcite was dynam-ically recrystallized at higher temperatures(1300�C) in the Ra Chu transect. Quartz-rich do-mains in calc-silicate and leucogranite record atransition from BLG through SGR to GBM. Theserecord an apparent increase of 200�–300�C over an∼30-m-thick section of the Ra Chu transect. The200�–300�C temperature distribution recordedwithin these rocks is likely a combination of tel-escoping and/or flattening of isotherms beneath thedetachment and the presence of a range of defor-mation mechanisms that accommodated exhu-mation at progressively lower temperature inter-

Figure 8. Rigid porphyroclasts plotted on the rigid grain net (Jessup et al. 2007). v, angle between the long axis of agrain and the foliation. , critical threshold values used to estimate mean kinematic vorticity (Wm).∗B


vals (Law et al. 2004; Jessup et al. 2006; Cottle etal. 2007, forthcoming).

Mean kinematic vorticity estimates and defor-mation temperatures on samples of mylonitic le-ucogranite and marble quantify three general stagesof this exhumation history. Ultramylonite leu-cogranite records 51%–67% pure shear at ∼400�–500�C. During progressive exhumation, strain waspartitioned into the structurally higher marble at1300�C (41%–55% pure shear) and finally into themoderately foliated marble at ∼200�–300�C (46%–59% pure shear). Mylonitic marble samples col-lected immediately below the detachment fromthree locations in Rongbuk Valley (Jessup et al.2006) yield Wm estimates that range between 0.74and 0.91 (28%–45% pure shear), a significantlylower contribution of pure shear than our resultsfrom the equivalent rock type and structural po-sition in the Ra Chu transect. Wm estimates frommylonitic and ultramylonitic leucogranite in theGondasampa transect (51%–67% pure shear) recorda higher contribution of pure shear than the equiv-alent rock type and structural position in RongbukValley (39%–61% pure shear; Jessup et al. 2006).The structurally highest samples (lower deforma-tion temperature) in both areas record the highestcontribution of simple shear, whereas deeper struc-tural positions (higher deformation temperature)yield higher contributions of pure shear.

Jessup et al. (2006) interpreted the relationshipbetween Wm estimates and deformation tempera-tures as the result of spatial and temporal parti-tioning of flow during exhumation. In this model,the structurally lowest samples record early,higher-temperature deformation conditions with agreater contribution of pure shear. These were jux-taposed with later, lower-temperature deformationconditions that are dominated by simple shear thatoccurred during the later stages of exhumation. An-other possibility is that decoupling between theporphyroclasts and matrix is more prominent in thephyllosilicate-rich leucogranites. Higher concen-trations of phyllosilicates, particularly around theporphyroclast matrix interface, are predicted to de-couple the porphyroclast from the matrix, resultingin data that overestimate the pure shear component(Johnson et al. 2009a, 2009b). Leucogranite samplesthat record the highest contribution of pure shearalso contain higher concentrations of phyllosili-cates, and for these samples, decoupling could offeran alternative explanation.

Strain partitioning within the footwall of theSTDS is common. Blocks of mylonitic leucogranitethat are suspended in mylonitic marble are evi-

dence for brittle fracturing and rotation of the gran-ite after it developed a mylonitic fabric and beforethe partitioning of strain into the weaker marble/calc-silicate section at temperatures of ∼200�–300�C. Similarly, foliation in calc-silicate iswrapped around rigid leucogranite pods in the RaChu section. Pseudotachylite that crosscuts leu-cogranite records high-strain events associatedwith seismic slip during exhumation along theSTDS. In general, marbles are the weakest rocktype, and strain is partitioned into these over arange of deformation temperatures. These obser-vations indicate that melting, rheologic contrastsbetween rock types, strain rate, and the onset ofdifferent deformation mechanisms are the majorvariables that control the extent to which strain ispartitioned and/or localized during exhumationalong the STDS.

Melt-Present Deformation. Melt-present defor-mation is widespread throughout the GHS andADM. Melting within the GHS exposed in theMount Everest region spans ∼8 m.yr. (22–16 Ma).The high-melt fraction present during this time in-terval had the potential to decrease the effectiveviscosity of the crust, resulting in dramatic rheo-logical weakening and midcrustal flow. Graniteemplacement mechanisms within the GHS arecharacterized by forceful injection along and acrossthe main foliation (Weinberg and Searle 1999) andprovide evidence for high fluid pressure gradientswith the GHS. Within the 30-km-thick GHS, themajority of mylonites are limited to the uppermost(Dzaka Chu, 1000 m; Rongbuk and Gondasampa,300 m) section of the GHS-STDS system (Law etal. 2004; Jessup et al. 2006, 2008a; Cottle et al.2007). Rare occurrences of mylonites have been re-ported in the interior of the GHS near Mount Ev-erest (Carosi et al. 1999) as well as more extensiveareas in western Bhutan (Carosi et al. 2006) andwestern Nepal (Carosi et al. 2007). The lack of my-lonite development and the widespread occurrenceof migmatite and leucogranite in the core of theGHS indicates that high-temperature melt-presentdeformation was the dominant mechanism bywhich the GHS was extruded southward.

While the final stages of leucogranite injectionand crystallization in the upper 1–2 km of the GHSoccurred during south-directed extrusion and ex-humation, the lower portion of the middle crust(130 km below the STDS), as exposed in the centralportion of the ADM, reached granulite facies (∼15Ma) and by ∼11–13 Ma was experiencing decom-pression and anatexis during orogen-parallel exten-sion (Cottle et al. 2009a; fig. 9). Partial melting at


Figure 9. Synthesis of thermochronologic data from the Mount Everest region (Hodges et al. 1994, 1998; Murphyand Harrison 1999; Simpson et al. 2002; Searle et al. 2003; Zhang and Guo 2007; Jessup et al. 2008b; Cottle et al.2009a) using published closure temperatures (Hodges et al. 1994; Farley 2000; Hacker et al. 2009). General range fordeformation temperatures and percent pure shear recorded by the South Tibetan detachment system exposed inGondasampa and Ra Chu sections.

this deeper structural position weakened the crustin a kinematic setting that was dominated byorogen-parallel extension.

These data indicate the following: (1) Midcrustalflow on the southern margin of the Tibetan plateauis decoupled from the upper crust via a network ofrelatively narrow shear zones and fault systems. (2)Between ∼24 and 16 Ma, south-directed extrusionand exhumation along the STDS is recorded by the∼30-km-thick GHS. (3) Anatexis at ∼11–13 Ma waslimited to the lower portion of the midcrust (asexposed in the ADM), where it flowed into areasof localized orogen-parallel extension and was ex-humed by the ADD and NRD. (4) The transitionfrom partial melting within the upper (GHS) tolower (ADM) portions of the midcrust resulted inlarge changes in crustal rheology (i.e., relativestrength) and played a substantial role in initiatingexhumation of both of these two distinct high-temperature structural domains during the middleand Late Miocene, respectively.

Implications for the Onset of East-West Exten-sion. Two major graben begin in the Mount Ev-erest and Ama Drime areas and extend from thecrest of the Himalaya into the interior of the Ti-betan plateau. The Tingri–Kung Co graben marksthe southern end of the more extensive Tangra–Yum Co graben that extends northward, where itintersects a northwest-striking, right-lateral strike-slip fault (Armijo et al. 1986; Taylor et al. 2003;Kapp and Guynn 2004). Thermochronometry on aportion of the Tingri–Kung Co graben constrainsthe onset of footwall exhumation and orogen-parallel extension to !4 Ma (fig. 9; Maheo et al.2007). The Dinggye graben, located ∼140 km eastof Tingri–Kung Co, initiated at ∼8–13 Ma (Zhangand Guo 2007). It extends from the eastern side ofthe ADM northward between Mabja and LhagoiKangri South Tibetan gneiss domes and into theXianza graben system that terminates at the north-west-striking, right-lateral Gyaring Co strike-slipfault. The Gyaring Co strike-slip fault forms a con-


jugate fault with northeast-striking, left-lateralfaults to the north of the Bangong-Nujiang suture(Taylor et al. 2003). These are kinematically linkedin a system of faults where conjugate sets of strike-slip faults and graben accommodate east-west ex-tension (Armijo et al. 1986; Taylor et al. 2003).The Dinggye-Pumqu-Xianza and Tingri–Kung Co–Tangra Yum Co grabens as well as some of the mid-crustal rocks exhumed in their footwalls (i.e.,ADM) are kinematically linked to active extensionin the interior of the Tibetan plateau.

The onset of orogen-parallel extension in theMount Everest region occurred between ∼11 and13 Ma, when the lower portion of the middle crust(i.e., ADM) experienced muscovite dehydrationmelting and anatexis (fig. 9; Cottle et al. 2009a).Melting within the overlying middle crustal rocks(GHS) had ceased by this time, and they as well asthe overlying STDS were dissected by normalfaults. The penetration depth of Tingri graben wasapparently limited to the upper crust (!4 Ma),whereas the Dinggye graben (∼8–13 Ma) appears torecord some early component of deformation be-low the brittle ductile transition zone (Zhang andGou 2007). We propose that early (11–13 Ma) melt-present low-viscosity flow weakened the crust ina kinematic setting that was dominated by crustal-scale extension. As a result of lateral gradients inlithostatic pressure, melt-weakened rocks were ex-humed from mid- to lower crustal depths withinareas of localized extension (ADM), while brittlenormal faulting occurred at shallower crustal levels(fig. 9). The ADD and NRD exhumed rocks fromone of the deepest structural positions exposed inthe central portion of the Himalayas. Footwall my-lonites in the ADD record top-down-to-the-westduring a minimum of 22–42 km of down dip dis-placement (Langille et al. 2010). These mylonitesprogressed into a detachment that, along with thePumqu fault systems, accommodated the final low-temperature ductile-brittle stages of exhumation.

Conclusions

Rocks exposed in the Mount Everest region providewindows into midcrustal processes of deformation,strain partitioning, and melting beneath the south-ern margin of the Tibetan plateau. Data from twonew transects across the STDS record a progressionin deformation mechanisms. During exhumation,strain was partitioned into an ∼150-m-thick shearzone that records solid-state fabric development.Ultramylonitic leucogranite records 51%–67%pure shear at ∼400�–500�C that was followed bypartitioning of strain into the structurally highermarble at 1300�C (41%–55% pure shear) and themoderately foliated marble at ∼200�–300�C (46%–59% pure shear). By 12–16 Ma, the STDS was in-active and no longer capable of accommodatingsouth-directed extrusion/exhumation. The locus ofmelting migrated into a structural position in thelower part of the midcrust by 11–13 Ma and wassubsequently exhumed within a kinematic settingthat was linked to extension.

A C K N O W L E D G M E N T S

R. D. Law and M. P. Searle provided the initial mo-tivation and funding for this study. D. Newell, L.Duncan, and D. Breecker provided valuable assis-tance in the field. S. Wangdu coordinated our tripsand brought us to Gondasampa. B. Dunne providedadvice on calcite twin morphology. J. Lee, R. Ca-rosi, D. Grujic, and J. M. Langille provided detailedreviews that helped strengthen a previous versionof the manuscript. N. Costello helped with initialpetrography as part of his undergraduate researchproject. Funding was provided by grants from theCollege of Art and Sciences at the University ofTennessee and National Science Foundation grantEAR-207524 to R. D. Law and M. P. Searle.

R E F E R E N C E S C I T E D

Armijo, R.; Tapponnier, P.; Mercier, J. L.; and Han, T.-L.1986. Quaternary extension in southern Tibet: fieldobservations and tectonic implications. J. Geophys.Res. 91:13803–13872.

Beaumont, C.; Jamieson, R. A.; Nguyen, M. H.; and Lee,B. 2001. Himalayan tectonics explained by extrusionof a low-viscosity crustal channel coupled to focusedsurface denudation. Nature 414:738–742.

Burchfiel, B. C.; Chen, Z.; Hodges, K. V.; Liu, Y.; Royden,L. H.; Deng, C.; and Xu, J. 1992. The South Tibetan

detachment system, Himalayan Orogen: extensioncontemporaneous with and parallel to shortening ina collisional mountain belt. Boulder, CO, GeologicalSociety of America, 41 p.

Burg, J. P.; Guiraud, M.; Chen, G. M.; and Li, G. C. 1984.Himalayan metamorphism and deformations in thenorth Himalayan Belt (southern Tibet, China). EarthPlanet. Sci. Lett. 69:391–400.

Burkhard, M. 1993. Calcite twins, their geometry, ap-pearance and significance as stress-strain markers and


indicators of tectonics regime: a review. J. Struct.Geol. 15:351–368.

Carosi, R.; Lombardo, B.; Molli, G.; Musumeci, G.; andPertusati, P. C. 1998. The South Tibetan detachmentsystem in the Rongbuk Valley, Everest region: defor-mation features and geological implications. J. AsianEarth Sci. 16:299–311.

Carosi, R.; Lombardo, B.; Musumeci, G.; and Pertusati,P. C. 1999. Geology of the Higher Himalayan Crys-tallines in Khumbu Himal (eastern Nepal). J. AsianEarth Sci. 17:785–803.

Carosi, R.; Montomoli, C.; Rubatto, D.; and Visona, D.2006. Normal-sense shear zones in the core of theHigher Himalayan Crystallines (Bhutan Himalaya):evidence for extrusion? In Law, R. D.; Searle, M. P.;and Godin, L., eds. Channel flow, ductile extrusionand exhumation in continental collision zones. Geol.Soc. Lond., p. 425–444.

Carosi, R.; Montomoli, D.; and Visona, D. 2007. A struc-tural transect in the lower Dolpo: insights on the tec-tonic evolution of western Nepal. J. Asian Earth Sci.29:407–423.

Cottle, J. M.; Jessup, M. J.; Newell, D. L.; Horstwood, M.S. A.; Noble, S. R.; Parrish, R. R.; Waters, D. J.; andSearle, M. P. 2009a. Geochronology of granulitized ec-logites from the Ama Drime Massif, implications forthe tectonic evolution of the South Tibetan Himalaya.Tectonics 28:TC1002.

Cottle, J. M.; Jessup, M. J.; Newell, D. L.; Searle, M. P.;Law, R. D.; and Horstwood, M. S. A. 2007. Structuralinsights into the early stages of exhumation along anorogen-scale detachment: the South Tibetan detach-ment system, Dzakaa Chu section, eastern Himalaya.J. Struct. Geol. 29:1781–1797.

Cottle, J. M.; Riley, N.; Waters, D. J.; Beyssac, J.; andJessup, M. J. Forthcoming. Thermal structure of theSouth Tibetan detachment system revealed by phaseequilibria modeling and Raman spectroscopy of car-bonaceous material. J. Metamorph. Geol.

Cottle, J. M.; Searle, M. P.; Horstwood, M. S. A.; andWaters, D. 2009b. Timing of midcrustal metamor-phism, melting, and deformation in the Mount Ev-erest region of southern Tibet revealed by U(-Th)-Pbgeochronology. J. Geol. 117:643–664.

Farley, K. A. 2000. Helium diffusion from apatite: generalbehavior as illustrated by Durango fluorapatite. J. Geo-phys. Res. 105:2903-2914.

Ferrill, D. A.; Morris, A. P.; Evans, M. A.; Burkhard, M.;Groshong, R. H.; and Onasch, C. M. 2004. Calcite twinmorphology: a low-temperature geothermometer. J.Struct. Geol. 26:1521–1530.

Fossen, H., and Tikoff, B. 1997. Forward modeling of non-steady-state deformations and the “minimum strainpath.” J. Struct. Geol. 19:987–996.

———. 1998. Forward modeling of non-steady-state de-formations and the “minimum strain path”: reply. J.Struct. Geol. 20:979–981.

Ghosh, S. K., and Ramberg, H. 1976. Reorientation ofinclusions by combination of pure and simple shear.Tectonophysics 34:1–70.

Grasemann, B.; Fritz, H.; and Vannay, J. C. 1999. Quan-titative kinematic flow analysis from the Main Cen-tral Thrust zone (NW-Himalaya, India): implicationsfor a decelerating strain path and the extrusion of oro-genic wedges. J. Struct. Geol. 21:837–853.

Groppo, C.; Lombardo, B.; Rolfo, F.; and Pertusati, P.2007. Clockwise exhumation path of granulitized ec-logites from the Ama Drime range (eastern Himala-yas). J. Metamorph. Geol. 25:51–75.

Grotenhuis, S. M.; Passchier, C. W.; and Bons, P. D. 2002.The influence of strain localization on the rotationbehaviour of rigid objects in experimental shear zones.J. Struct. Geol. 24:485–499.

Hacker, B. R.; Wallis, S. R.; McWilliams, M. O.; and Gans,P. B. 2009. 40Ar/39Ar constraints on the tectonic historyand architecture of the ultrahigh-pressure Sulu oro-gen. J. Metamorph. Geol. 27:827–844.

Hodges, K. V.; Bowring, S. A.; Davidek, K. L.; Hawkins,D. P.; and Krol, M. A. 1998. Evidence for rapid dis-placement on Himalayan normal faults and the im-portance of tectonic denudation in the evolution ofmountain ranges. Geology 26:483–486.

Hodges, K. V.; Hames, W. E.; Olszewski, W.; Burchfiel,B. C.; Royden, L. H.; and Chen, Z. 1994. Thermo-barometric and Ar-40 Ar-39 geochronological con-straints on Eohimalayan metamorphism in the Ding-gye area, southern Tibet. Contrib. Mineral. Petrol.117:151–163.

Hoke, L.; Lamb, S.; Hilton, D. R.; and Podera, R. J. 2000.Southern limit of mantle-derived geothermal heliumemissions in Tibet: implications for lithosphericstructure. Earth Planet. Sci. Lett. 180:297–208.

Iacopini, D.; Carosi, R.; Montomoli, C.; and Passchier,C. W. 2008. Strain analysis and vorticity of flow inthe northern Sardinian Variscan belt: recognition of apartitioned oblique deformation event. Tectonophy-sics 446:77–96.

Jeffrey, G. B. 1922. The motion of ellipsoidal particlesimmersed in a viscous fluid. Proc. R. Soc. 102:161–179.

Jessup, M. J.; Cottle, J. M.; Searle, M. P.; Law, R. D.;Newell, D. L.; Tracy, R. J.; and Waters, D. J. 2008a. P-T-t-D paths of Everest Series schist, Nepal. J. Meta-morph. Geol. 26:717–739.

Jessup, M. J.; Law, R. D.; and Frassi, C. 2007. The rigidgrain net (RGN): an alternative method for estimatingmean kinematic vorticity number (W-m). J. Struct.Geol. 29:411–421.

Jessup, M. J.; Law, R. D.; Searle, M. P.; and Hubbard, M.S. 2006. Structural evolution and vorticity of flow dur-ing extrusion and exhumation of the Greater Hima-layan Slab, Mount Everest Massif, Tibet/Nepal: im-plications for orogen-scale flow partitioning. In Law,R. D.; Searle, M. P.; and Godin, L., eds. Channel flow,ductile extrusion and exhumation in continental col-lision zones. Geol. Soc. Lond., p. 379–413.

Jessup, M. J.; Newell, D. L.; Cottle, J. M.; Berger, A. L.;and Spotila, J. A. 2008b. Orogen-parallel extension andexhumation enhanced by focused denudation in Arun


River gorge, Ama Drime Massif, Tibet-Nepal. Geology36:587–590.

Jiang, D. 1998. Forward modeling of non-steady-state de-formations and the “minimum strain path”: discus-sion. J. Struct. Geol. 20:975–977.

Johnson, S. E.; Lenferink, H. J.; Marsh, J. H.; Price, N. A.;Koons, P. O.; and West, D. P. 2009a. Kinematic vor-ticity analysis and evolving strength of myloniticshear zones: new data and numerical results. Geology37:1075–1078.

Johnson, S. E.; Lenferink, H. J.; Price, N. A.; Marsh, J. H.;Koons, P. O.; West, D. P.; and Beane, R. 2009b. Clast-based kinematic vorticity gauges: the effects of slip atmatrix/clast interfaces. J. Struct. Geol. 31:1322–1339.

Kapp, P., and Guynn, J. H. 2004. Indian punch rifts Tibet.Geology 32:993–996.

Langille, J. M.; Jessup, M. J.; Cottle, J. M.; Newell, D. L.;and Seward, G. 2010. Kinematic evolution of the AmaDrime detachment: insights into orogen-parallel ex-tension and exhumation of the Ama Drime Massif,Tibet-Nepal. J. Struct. Geol., doi:10.1016/j.jsg.2010.04.005.

Larson, K. P., and Godin, L. 2009. Kinematics of theGreater Himalayan Sequence, Dhaulagiri Himal: im-plications for the structural framework of central Ne-pal. J. Geol. Soc. Lond. 166:25–43.

Law, R. D.; Jessup, M. J.; Searle, M. P.; Cottle, J. M.; andWaters, D. 2009. Telescoping of isotherms beneath theSouth Tibetan detachment, Mount Everest Massif:implications for magnitude of internal flow during ex-trusion of the Greater Himalayan Slab. Himalayan J.Sci. 5:86–87.

Law, R. D.; Searle, M. P.; and Simpson, R. L. 2004. Strain,deformation temperatures and vorticity of flow at thetop of the Greater Himalayan Slab, Everest Massif,Tibet. J. Geol. Soc. 161:305–320.

Liu, Y.; Siebel, W.; Massonne, H. J.; and Xiao, X. 2007.Geochronological and petrological constraints for tec-tonic evolution of the central Greater Himalayan Se-quence in the Kharta area, southern Tibet. J. Geol.115:215–242.

Lombardo, B.; Pertusati, P. C.; and Borghi, A. 1993. Ge-ology and tectonomagmatic evolution of the easternHimalaya along the Chomolungma-Makalu transect.Himalayan Tectonics 74:341–355.

Lombardo, B.; Pertusati, P. C.; Rolfo, F.; and Visona, D.1998. First report of eclogites from the E Himalaya:implications for the Himalayan Orogeny. Mem. Sci.Geol. Univ. Padova 50:67–68.

Lombardo, B.; Pertusati, P. C.; and Visona, D. 2000. Ge-ology of the northern Arun tectonic window. 15thHimalayan-Karakoram-Tibetan International Work-shop, Earth Science Frontiers. Beijing, China Univer-sity of Geosciences, p. 37–38.

Lombardo, B., and Rolfo, F. 2000. Two contrasting eclo-gite types in the Himalayas: implications for the Him-alayan orogeny. J. Geodyn. 30:37–60.

Maheo, G.; Leloup, P. H.; Valli, F.; Lacassin, R.; Arnaud,N.; Paquette, J. L.; Fernandez, A.; Haibing, L.; Farley,K. A.; and Tapponnier, P. 2007. Post 4 Ma initiation

of normal faulting in southern Tibet: constraints fromthe Kung Co half graben. Earth Planet. Sci. Lett. 256:233–243.

Means, W. D.; Hobbs, B. E.; Lister, G. S.; and Williams,P. F. 1980. Vorticity and non-coaxiality in progressivedeformations. J. Struct. Geol. 2:371–378.

Murphy, M. A., and Harrison, T. M. 1999. Relationshipbetween leucogranites and the Qomolangma detach-ment in the Rongbuk Valley, South Tibet. Geology 27:831–834.

Murphy, M. A.; Yin, A.; Kapp, P.; Harrison, T. M.; Man-ning, C. E.; Ryerson, F. J.; Lin, D.; and Jinghui, G. 2002.Structural evolution of the Gurla Mandhata detach-ment system, southwest Tibet: implications for theeastward extent of the Karakoram fault system. Geol.Soc. Am. Bull. 114:428–447.

Newell, D. L.; Jessup, M. J.; Cottle, J. M.; Hilton, D. R.;Sharp, Z.; and Fischer, T. 2008. Aqueous and isotopegeochemistry of mineral springs along the southernmargin of the Tibetan plateau: implications for fluidsources and regional degassing of CO2. Geochem. Geo-phys. Geosyst 9:Q08014.

Passchier, C. W. 1987. Stable positions of rigid objects innon-coaxial flow: a study in vorticity analysis. J.Struct. Geol. 9:679–690.

Pertusati, P. C.; Musumeci, G.; Lombardo, B.; Peruzzo,L.; and Rolfo, F. 2003. The South Tibetan detachmentbetween Doya La and Tingri (southern Tibet). 18thHimalaya-Karakoram-Tibet International Workshop.Zurich, ETH, p. 92.

Sakai, H.; Sawada, M.; Takigami, Y.; Orihashi, Y.; Dan-hara, T.; Iwano, H.; Kuwahara, Y.; Dong, Q.; Cai, H.;and Li, J. 2005. Geology of the summit limestone ofMount Qomolangma (Everest) and cooling history ofthe Yellow Band under the Qomolangma detachment.Island Arc 14:297–310.

Searle, M. P. 1999a. Emplacement of Himalayan leu-cogranites by magma injection along giant sill com-plexes: examples from the Cho Oyu, Gyachung Kangand Everest leucogranites (Nepal Himalaya). J. AsianEarth Sci. 17:773–783.

———. 1999b. Extensional and compressional faults inthe Everest-Lhotse Massif, Khumbu Himalaya, Nepal.J. Geol. Soc. 156:227–240.

Searle, M. P.; Simpson, R. L.; Law, R. D.; Parrish, R. R.;and Waters, D. J. 2003. The structural geometry, meta-morphic and magmatic evolution of the Everest Mas-sif, High Himalaya of Nepal–South Tibet. J. Geol. Soc.160:345–366.

Simpson, C., and De Paor, D. G. 1993. Strain and kine-matic analysis in general shear zones. J. Struct. Geol.15:1–20.

———. 1997. Practical analysis of general shear zonesusing the porphyroclast hyperbolic distributionmethod: an example from the Scandinavian Caledon-ides. In Sengupta, S., ed. Evolution of geological struc-tures in micro- to macro-scales. London, Chapman &Hall, p. 169–184.

Simpson, R. L.; Parrish, R. R.; Searle, M. P.; and Waters,D. J. 2000. Two episodes of monazite crystallization


during metamorphism and crustal melting in the Ev-erest region of the Nepalese Himalaya. Geology 28:403–406.

Stipp, M.; Stunitz, H.; Heilbronner, R.; and Schmid, S.M. 2002. The eastern Tonale fault zone: a “naturallaboratory” for crystal plastic deformation of quartzover a temperature range from 250 to 700�C. J. Struct.Geol. 24:1861–1884.

Taylor, M.; Yin, A.; Ryerson, F.; Kapp, P.; and Ding, L.2003. Conjugate strike slip faulting along the Bang-ong–Nujiang suture zone accommodates coeval east–west extension and north-south shortening in the in-terior of the Tibetan plateau. Tectonics 22:2044.

Thiede, R. C.; Arrowsmith, J. R.; Bookhagen, B.; Mc-Williams, M.; Sobel, E. R.; and Strecker, M. R. 2006.Dome formation and extension in the Tethyan Him-alaya, Leo Pargil, northwest India. Geol. Soc. Am.Bull. 118:635–650.

Tikoff, B., and Fossen, H. 1995. The limitations of three-dimensional kinematic vorticity analysis. J. Struct.Geol. 17:1771–1784.

Visona, D., and Lombardo, B. 2002. Two-mica and tour-maline leucogranites from the Everest-Makalu region(Nepal-Tibet): Himalayan leucogranite genesis by iso-baric heating? Lithos 62:125–150.

Visona, D.; Villa, I. M.; Pertusati, P.; and Lombardo, B.2000. Different varieties of Miocene leucogranite in

the Arun Valley–Everest–Makalu area: field relations,petrology and isotope geochemistry. 15th Himalayan-Karakoram-Tibetan International Workshop, EarthScience Frontiers. Beijing, China University of Geo-sciences, p. 100–101.

Wallis, S. R. 1992. Vorticity analysis in a metachert fromthe Sanbagawa Belt, SW Japan. J. Struct. Geol. 14:271–280.

Wallis, S. R.; Platt, J. P.; and Knott, S. D. 1993. Recog-nition of syn-convergence extension in accretionarywedges with examples from the Calabrian Arc and theEastern Alps. Am. J. Sci. 293:463–495.

Webb, A.; Yin, A.; Harrison, T.; Celerier, J.; and Burgess,P. W. 2007. The leading edge of the Greater Himalaya:implications for the evolution of the Himalayan oro-gen. Geology 35:955–958.

Weber, J. C.; Ferrill, D. A.; and Roden-Tice, M. K. 2001.Calcite and quartz microstructural geothermometryof low-grade metasedimentary rocks, Northern Range,Trinidad. J. Struct. Geol. 23:93–112.

Weinberg, R. F., and Searle, M. P. 1999. Volatile-assistedintrusion and autometasomatism of leucogranites inthe Khumbu Himalaya, Nepal. J. Geol. 107:27–48.

Zhang, J., and Guo, L. 2007. Structure and geochronologyof the southern Xainza-Dinggye rift and its relation-ship to the South Tibetan detachment system. J. AsianEarth Sci. 29:722–736.

Documents

Progression from South-Directed Extrusion to Orogen ...web.utk.edu/~mjessup/Jessup/Ama_Drime_Massif_files... · Cottle et al. 2009a), Leo Pargil dome (Thiede et al. 2006), and Gurla