Basin formation in the High Himalaya by arc-parallel extension …easd.geosc.uh.edu/murphy/pdf/Saylor et al. 2010.pdf · 2012-02-21 · Zhada basin are the South Tibetan Detachment

Basin formation in the High Himalaya by arc-parallel extension

and tectonic damming: Zhada basin, southwestern Tibet

Joel Saylor,1,2 Peter DeCelles,1 George Gehrels,1 Michael Murphy,3

Ran Zhang,3 and Paul Kapp1

Received 3 September 2008; revised 25 August 2009; accepted 23 September 2009; published 27 February 2010.

[1] The late Miocene–Pleistocene Zhada basin insouthwestern Tibet provides a record of subsidenceand basin formation within an active collisional thrustbelt. The >800 m thick basin fill is undeformed andwas deposited along an angular unconformity on topof Tethyan strata that were previously shortened inthe Himalayan fold-thrust belt. Modal sandstonepetrographic data, conglomerate clast count data, anddetrital zircon U-Pb age spectra indicate a transitionfrom detritus dominated by a distal, northern source to alocal, southern source. This transition was accompaniedby a change in paleocurrent directions from uniformlynorthwestward to basin-centric. At the same time thedepositional environment in the Zhada basin changedfrom a large, braided river to a closed-basin lake.Sedimentation in the Zhada basin was synchronouswith displacement on the Qusum and Gurla Mandhatadetachment faults, which root beneath the basin andexhume midcrustal rocks along the northwestern andsoutheastern flanks of the basin, respectively. Theseobservations indicate that accommodation for Zhadabasin fill was produced by a combination of tectonicsubsidence and damming, as midcrustal rocks wereevacuated from beneath the Zhada basin in responseto arc-parallel slip on crustal-scale detachment faults.Citation: Saylor, J., P. DeCelles, G. Gehrels, M. Murphy, R. Zhang,

and P. Kapp (2010), Basin formation in the High Himalaya by arc-

parallel extension and tectonic damming: Zhada basin, southwestern

Tibet, Tectonics, 29, TC1004, doi:10.1029/2008TC002390.

1. Introduction

[2] Since they were first recognized [Molnar andTapponnier, 1978; Ni and York, 1978], Miocene normalfault-bounded basins in the High Himalaya and TibetanPlateau have fueled speculation on their origins and havebeen used as fundamental arguments in various tectonicmodels. These include basin formation and arc-parallel

extension resulting from oblique convergence between Indiaand Asia [McCaffrey and Nabelek, 1998; Seeber and Pecher,1998], eastward extrusion of Asian lithosphere [Armijo et al.,1986], attainment of a limiting crustal thickness and sub-sequent gravitational collapse of the upper crust [Molnarand Tapponnier, 1978; Molnar and Lyon-Caen, 1988; Kappand Guynn, 2004], removal of dense mantle lithosphere[England and Houseman, 1989], oroclinal bending of theHimalayan arc [Ratschbacher et al., 1994], and formationas a result of arc-normal extension [Hurtado et al., 2001].Moreover, these basins provide a record of tectonic andclimatic processes essential for discriminating among com-peting hypotheses about the evolution of the Himalayan-Tibetan orogenic system.[3] The largest (>150 km long and >60 km wide, with an

outcrop extent of >9,000 km2) of these enigmatic basinsis theMiocene-Pleistocene Zhada basin in the western portionof the Himalayan arc (Figure 1) [Zhang et al., 1981; Zhu etal., 2004, 2007;Meng et al., 2008;Wang et al., 2008a, 2008b;Saylor et al., 2009]. The Zhada basin trends NW–SE and islocated just north of the High Himalayan ridge crest in thewestern part of the orogen (�32�N, 82�E). The basin occu-pies a region that is more than 1000 m lower in elevation thananywhere else along strike in the High Himalaya and itdwarfs other late Cenozoic sedimentary basins in southernTibet and the northern Himalaya [Ganser, 1964]. Stableisotope data indicate that the Zhada region may have lostup to 1.5 km of elevation since the late Miocene [Saylor etal., 2009], suggesting a direct link between basin-formingprocesses and regional elevation.[4] The published literature on Zhada basin presents con-

flicting information and interpretations [Ni and Barazangi,1985; Zhang et al., 2000; Murphy et al., 2002; Thiede et al.,2006; Valli et al., 2007; Wang et al., 2008a]. For example,the basin-filling Zhada Formation has been reported asboth an upward fining fluviolacustrine sequence [Zhang etal., 1981; Zhou et al., 2000; Li and Zhou, 2001] and aboulder conglomerate-capped succession [Zhu et al., 2004,2007]. The basin has been interpreted to be the result ofmovement on the South Tibetan Detachment system due toarc-perpendicular extension [Wang et al., 2004], upliftcaused by arc-perpendicular compression [Zhou et al.,2000; Zhu et al., 2004], and arc-perpendicular rifting [Wanget al., 2008a].[5] In this paper we examine the development of the

Zhada basin and its relationship to basin-bounding faults asinformed by new sedimentological and provenance data. Ourresults show that the basin evolved from a throughgoingfluvial system to an internally drained depocenter concom-

TECTONICS, VOL. 29, TC1004, doi:10.1029/2008TC002390, 2010ClickHere

for

FullArticle

1Department of Geosciences, University of Arizona, Tucson, Arizona,USA.

2Now at Department of Geological Sciences, Jackson School ofGeosciences, University of Texas at Austin, Austin, Texas, USA.

3Department of Earth and Atmospheric Sciences, University ofHouston, Houston, Texas, USA.

Copyright 2010 by the American Geophysical Union.0278-7407/10/2008TC002390$12.00

TC1004 1 of 24

http://dx.doi.org/10.1029/2008TC002390

Figure 1. (a) Elevation, shaded relief, and generalized tectonic map of the Himalayan-Tibetan orogenicsystem showing the location of the Zhada basin relative to several other normal fault bounded basins.Image from the UNAVCO Jules Verne Voyager and the Generic Mapping Tools (GMT). (b) Generalizedgeologic map of the Zhada region [Chen and Xu, 1987;Murphy et al., 2000, 2002;M.Murphy, unpublishedmapping, 2005, 2006, 2007]. Locations of measured sections presented in Figure 6 are indicated by solidblack lines. Location of Figure 2 is outlined.

TC1004 SAYLOR ET AL.: ZHADA EVOLUTION ARC-PARALLEL EXTENSION

2 of 24

TC1004

itant with exhumation of midcrustal rocks to the northwestand southeast in the footwalls of major arc-parallel exten-sional systems. By combining our results with regionalstructural studies, we present a tectonic model that explainsZhada basin development through a combination of sill upliftand basin subsidence due to arc-parallel extension on crustal-scale detachment faults.

2. Geologic Setting

[6] Major structural features near the margins of theZhada basin are the South Tibetan Detachment system(STDS) to the southwest, the Great Counter thrust and rightslip Karakoram fault to the northeast, and the Leo Pargil andGurla Mandhata gneiss domes to the northwest and south-east, respectively (Figure 1b). The STDS is a series of northdipping, low-angle, top-to-the-north normal faults that placePaleozoic-Mesozoic low-grade metasedimentary rocks ofthe Tethyan sequence on high-grade gneisses and granitoidsof the Greater Himalayan sequence [Hodges et al., 1992].No ages for movement on the STDS in this area have beenpublished, but the timing of displacement along strike isbracketed between 21 and 12 Ma [Hodges et al., 1992,1996; Searle et al., 1997; Murphy and Harrison, 1999;Hodges, 2000; Murphy and Yin, 2003; Searle et al., 2003;Yin, 2006; Cottle et al., 2007]. In the Zhada region the southdipping, top-to-the-north Oligocene-Miocene Great Counterthrust system cuts the Indus suture [e.g., Ganser, 1964; Yinet al., 1999; Murphy and Yin, 2003]. The region north of theIndus suture is dominated by Mesozoic and Cenozoicigneous rocks of the Gangdese magmatic arc [Yin et al.,1999; Aitchison et al., 2002; Murphy et al., 2002; Murphyand Yin, 2003]. Exhumation of migmatites and paragneissesin the Leo Pargil gneiss dome by normal faulting initiated ateither �15 Ma or �9 Ma, as indicated by 40Ar/39Ar andAFT cooling ages, respectively [Zhang et al., 2000; Thiedeet al., 2006]. Exhumation of Greater Himalayan and LesserHimalayan rocks in the Gurla Mandhata gneiss dome beganat �9 Ma [Murphy et al., 2002; Murphy, 2007]. FaultedQuaternary basin fill indicates that gneiss dome develop-ment continues today [Murphy et al., 2002]. Althoughprevious studies have linked development of the Zhadabasin with movement on faults to the northeast and south-west of the basin [Zhou et al., 2000; Wang et al., 2004; Zhuet al., 2004; Wang et al., 2008a], our mapping (Figure 2)reveals that only in its northwest corner is the basin fill cutby a basin-bounding fault. Elsewhere in the basin, theundeformed Zhada Formation onlaps paleotopography inthe hanging walls and footwalls of the Great Counter thrustand the STDS, indicating that it entirely postdates slip onthese major fault systems (Figure 2).[7] The Zhada basin is filled by the Zhada Formation,

which consists of >800 m of fluvial, lacustrine, eolian, andalluvial fan deposits (Figures 2 and 3). The basin fill is largelystructurally undisturbed and sits above shortened Tethyansequence strata along an angular or buttress unconformity(Figures 4a–4c). The Zhada Formation is capped by ageomorphic surface that is correlative across the basin(Figures 4b and 4d). The geomorphic surface is interpreted

as a paleodepositional plain that marks the maximum extentof sedimentation prior to integration of the modern SutlejRiver drainage network. After deposition, the basin wasincised to basement by the Sutlej River, exposing the entirebasin fill in spectacular canyons and cliffs. The best estimatefor the onset of sedimentation in the Zhada basin is 9.2 Mabased on magnetostratigraphy (Figure 5) [Wang et al.,2008b; Saylor et al., 2009]. Correlations to the geomagneticpolarity timescale [Lourens et al., 2004] were constrained byMiocene-Pliocene mammal fossils and by an observed shiftfrom pure C3 to mixed C3 and C4 vegetation in the Zhadabasin. The correlation presented in Figure 5 accounts for allof our normal and reversed intervals; places the C3-C4

transition at �7 Ma, which is consistent with its age else-where in the Himalaya [France-Lanord and Derry, 1994;Quade et al., 1995; Garzione et al., 2000; Ojha et al., 2000;Wang et al., 2006]; and yields relatively constant andreasonable sedimentation rates. For an in-depth discussionof alternative correlations see Saylor et al. [2009].

3. Sedimentology of the Zhada Formation

[8] We measured 14 stratigraphic sections spanning thebasin from the Zhada county seat in the southeast to the LeoPargil/Qusum Range front in the northwest (Figures 1b, 2and 3). Here, we define the Zhada Formation as the entirebasin fill extending from the basal unconformity to the paleo-depositional surface. Sections were measured at centimeterscale and correlated by tracing major stratigraphic unitsand the paleodepositional surface that caps most sections(Figures 3, 4b, and 4d). The 14 lithofacies associations inTable 1 are defined on the basis of lithology, texture andsedimentary structures, and are grouped into five deposi-tional environment associations. Only abbreviated descrip-tions and interpretations are presented here; for details seeSaylor [2008]. Lithofacies codes following the lithofaciesassociations in Table 1 are based on the work byMiall [1978]andDeCelles et al. [1991] and described in Table S1.1 Unlessotherwise indicated, all deposits are laterally continuousfor hundreds of meters to several kilometers.

3.1. Zhada Formation Members

[9] We identified five members of the Zhada Formationbased on lithofacies assemblages (Figures 2 and 6). Thelower conglomerate member (Nlc) consists of lithofaciesassociations F1, F2, and rare F3 (Figures 7a, 7b and 7e).Where present, it extends from the base of the section to thetop of the last F1 interval >1 m thick. Above this, the lowerfluvial member (Nlf) consists of lithofacies associations F2,F3, S2 and S3 and extends from the top of the lowerconglomerate member to the base of the lacustrine clay-stone. The lacustrine member (Nl) includes primarily lith-ofacies associations P1, P2, L1, L2, S1 and minor S2, S3and F2 (Figures 7d and 7f). The lacustrine member extendsfrom the first occurrence of lacustrine claystone to the firstthick (>1 m) interval displaying soft-sediment deformation.

1Auxiliary material data sets are available at ftp://ftp.agu.org/apend/tc/2008tc002390. Other auxiliary material files are in the HTML.


3 of 24

TC1004

Figure 2. Geologic map of the Zhada region showing the traces of measured sections and therelationship between the Zhada Formation and major faults in the region.


4 of 24

TC1004

Figure

3.

Measuredsections,paleocurrentmeasurements,clastcompositions,andlithologic

correlationsfrom

theZhada

basin.See

Figure

2forlocations.


5 of 24

TC1004

Figure

3.

(continued)


6 of 24

TC1004

The upper fluvial member (Nuf) is similar to the lowerfluvial member with the exception that it is coarser andcontains more abundant soft-sediment deformation. Thismember is composed primarily of sand- to boulder-sizedmaterial in contrast to the lower fluvial member which isdominantly silt- to sand-sized material. The upper fluvialmember extends from the lacustrine member to the firstsignificant (>10 m thick) conglomeratic interval. The upperalluvial fan member (Nal) includes lithofacies associationsA1–A4, F3, P1 and P2 (Figure 7c). It extends from the firstsignificant conglomeratic interval above the lacustrine mem-ber to the top of the Zhada Formation.

3.2. Paleocurrent Measurements

[10] Paleocurrent data consist of measurements of 298limbs of trough cross strata (method I of DeCelles et al.[1983]) at 29 sites and measurements of 568 long-axistransverse imbricated clasts at 47 sites in measured sectionsthroughout Zhada basin (Figure 8).[11] Paleocurrent measurements fall into two groups: those

that show a northwestward paleoflow direction and thosedirected toward the basin center. The first group comes fromthe lower two members of the Zhada Formation (Nlc and

Nlf). The average paleocurrent azimuth from 8 measuredsections in the lower members ranges between 270� and 349�(Figure 8a and Table S2). The second group of measurementscomes from the two upper members (Nuf and Nal) oranomalously coarse-grained tongues (cobble-boulder) withinlower members. These paleocurrent indicators are less uni-form but generally indicate flow toward the center of thebasin in seven of the nine measured sections (Figure 8b andTable S2). In the northwest end of the basin there is a trendfrom uniformly northwestward directed paleoflow (Qusumsection; Figure 8a) to a more complex, but average north-westward directed paleoflow (1NWZ and 2NWZ sections)and finally a complete reversal to an eastward averagepaleoflow direction (3NWZ section; Figure 8b).

4. Provenance Analysis

4.1. Methods

4.1.1. Sandstone Petrography and ConglomerateClast Counts[12] Modal sandstone composition data were collected

from 35 standard petrographic thin sections from 9 measured

Figure 4. The Zhada Formation is undeformed and lies in buttress unconformity upon deformedTethyan sequence strata. (a) The unconformity relationship in the basin center. (b) The unconformityrelationship at the basin margins. (c) Deformation in the central Zhada basin is limited to small-offsetfaults. The minor offset, lack of basement involvement, and the fact that deformation is limited to a thinstratigraphic interval indicates that this deformation is related to syndepositional slumping rather thansignificant tectonic activity. (d) The geomorphic surface which caps the Zhada Formation


7 of 24

TC1004

sections across the basin. Thin sections were stained forpotassium and calcium feldspar and point counted (onlygrains larger than silt sized, 403–744 counts per slide) usingthe Gazzi-Dickinson method [Ingersoll et al., 1984]. Thepetrographic counting parameters are listed in Table S3, andrecalculated modal data are given in Data Set S1.[13] Conglomerate composition data consist of 3447

identified clasts from 32 sites throughout Zhada basin,and bed loads of several modern streams and rivers.Conglomerate clast counts involved identification of clastlithology of �100 pebble-sized or larger clasts at regular

intervals from within individual beds. Bed load countsinvolved identification of �100 pebble-sized or larger claststhat were dredged from the streambed.4.1.2. U-Pb Detrital Zircon Geochronology[14] U-Pb geochronology was conducted on zircons

separated from nine sandstone samples and one paragneisssample from the Zhada basin and the footwall of the QusumDetachment, respectively, and were processed for detritalzircon analysis using procedures described by Gehrels[2000] and Gehrels et al. [2008]. Analyses were conductedusing the laser ablation multicollector inductively coupledplasma mass spectrometer (LA-MC-ICPMS) at the ArizonaLaserChron Center. Approximately 100 individual zircongrains were analyzed from each sample (with the exceptionof samples 2EZ88 and 2EZ60.7A). These were selectedrandomly from all sizes and shapes, although grains withobvious cracks or inclusions were avoided. In-run analysis offragments of a large zircon crystal (generally every sixthmeasurement) with known age of 564 ± 4 Ma (2-sigmaerror) was used to correct for interelement and intraelementfractionation. The uncertainty resulting from the calibrationcorrection is generally 1%–2% (2 sigma) for both 206Pb/207Pband 206Pb/238U ages. Common Pb was corrected by using themeasured 204Pb and assuming an initial Pb composition fromStacey and Kramers [1975] (with uncertainties of 1.0 for206Pb/204Pb and 0.3 for 207Pb/204Pb).The analytical data arereported in Data Set S2. Details of the operating conditionsand analytical procedures can be found in the work byGehrelset al. [2008].[15] For each analysis, the errors in determining 206Pb/238U

and 206Pb/204Pb result in a measurement error of �1%–2%(at 2-sigma level) in the 206Pb/238U age. The errors in

measurement of 206Pb/207Pb and 206Pb/204Pb also result in�1%–2% (at 2-sigma level) uncertainty in age for grains thatare >1.0 Ga, but are substantially larger for younger grains

due to low intensity of the 207Pb signal. Formost analyses, the

crossover in precision of 206Pb/238U and 206Pb/207Pb ages

occurs at 0.8–1.0Ga. Interpreted ages are based on 206Pb/238U

for <�300 Ma grains and on 206Pb/207Pb for >�300 Ma

grains. Though 206Pb/238U ages are more precise below 1 Ga,the occurrence of numerous zircon grains with lead lossdue to Cenozoic metamorphism necessitated a youngercutoff. Analyses with >20% uncertainty, that are >30%discordant (by comparison of 206Pb/238U and 206Pb/207Pbages) or >5% reverse discordant are omitted from interpreta-tion (Figure S1).

4.2. Results

4.2.1. Sandstone and Conglomerate Compositions[16] Quartz is present in monocrystalline, polycrystalline

(commonly in granitoid lithic grains), and foliated poly-crystalline (commonly in mylonitic grains) grains, and alsomonocrystalline grains within sandstone or quartzite grains(Figure S2). Other lithic grains include phyllite, schist,carbonate, chert, andmudstone. Volcanic grains are common,particularly in the lowest part of the Zhada Formation. Themost abundant are felsic volcanic grains. This group includesporphyritic, highly altered quartzofeldspathic grains, quartz-rich grains with disseminated plagioclase, and volcanic

Figure 5. Correlation of the composite magnetostrati-graphic section with the GPTS of Lourens et al. [2004]. Thecomposite magnetostratigraphic section was constructed fromtwo magnetostratigraphic sections, each of which spannedthe entire thickness of the Zhada Formation [Saylor et al.,2009]. The stratigraphic location of the switch from plantsutilizing exclusively the C3 carbon fixation pathway to a mixof C3 and C4 plants is indicated by the gray box (between 170and 250 m in the South Zhada measured section). Thelocation of the first Hipparion fossils is at �240 m in theSouth Zhada measured section.


8 of 24

TC1004

Table

1.Lithofacies

AssociationCodes,Descriptions,andInterpretationsa

Lithofacies

Association

Description

Interpretation

F1:Gcm

i,Gch,Gt,Gcf

Amalgam

ated

channel

form

sandtabularunitsofmoderatelysorted,

clast-supported

pebble

conglomeratefeaturingabundanttraction

sedim

entary

structures;depositsare2–10m

thick;erosivebases.

Depositsofmigratingchannels,bars,bed

load

sheets,andgravel

dunes

ofa

braided

river

within

alow-gradientfluvialsetting[BristowandBest,1993;

CantandWalker,1978;Collinson,1996;HeinzandAigner,2003;

Luntet

al.,2004]

F2:St,Sp,Sh,Sc

Tabularunits0.1–5m

thickofhorizontallyorcross-stratifiedsandstone;

abundantload

castsandball-and-pillow

structuresandrare

channel

form

sSheetflooddepositsonmedialto

distalfluvialfansin

amarshyfloodplain

ormarginal

lacustrinesetting[H

amptonandHorton,2007;

Saez

etal.,2007]

F3:Sm,Sc,

Mm,Mc,

Ml,Mh

Massive,

organic-rich,strongly

calcareoussiltstones

andsandstones

0.1–4m

thickfeaturingsoft-sedim

entdeform

ationbutno

pedogenesis;containsostracods,roottraces,andsm

all

planorbid

gastropods(G

yraulussp?)

Marshyfloodplain

orlakemargin

wetlanddeposits[Allen

andCollinson,1986]

S1:Gct,Gcm

i,Sp,St,Sh,Sm,Sr

Tabular,upwardcoarseningsandstoneto

pebble

conglomerate0.25to

6m

thick;erosional

bases

andabrupttops;horizontalstratification,

clim

bing,trough,andplanar

cross

stratification;channelized

depositsupto

0.5

mthickarelocallypresent;abundant,

well-preserved,robustgastropodshells

Product

ofsedim

entationwithin

adelta

complex[Allen

andCollinson,1986;

Saez

etal.,2007]

S2:Sh

Low-angle

laminae

upto

3cm

thickwithoutinternal

structure;developed

inwell-sorted,fine-grained

sand;roottraces

andsoft-sedim

ent

deform

ationarecommon

Vegetated

eolian

sandflat

[Hunter,1977]

S3:Sf

Stacked

units(0.5–1.5

mthick)ofsteeply

inclined,upwardcoarsening

laminations;developed

inwell-sorted,fine-grained

sandstone;

roottraces

andsoft-sedim

entdeform

ationarecommon

Lacustrinecoastalplain

dunedeposits

L1:Ml

Papery,

laminated

siltstone;

abundantfossilgrasses,fragmentary

shellmaterial

andoccasional

fish

fossils

Depositsofagrass-rich,shallowly

submerged,low-gradient,lakemargin

environment[Allen

andCollinson,1986;TalbotandAllen,1996]

L2:Mr,Sr,Srw,Sp,Sh,Sm

Upwardcoarseningsandstonedeposits0.5–3m

thick;gradational

bases

anderosional

tops;clim

bingripple

andwaveripple

cross

stratification

andmuddrapes

Depositsofaterm

inal

lobeonaprogradingdelta

front

[Allen

andCollinson,1986;Dam

andSurlyk,1993;Jopling,

1965;Saez

etal.,2007;TalbotandAllen,1996]

P1:Mh,Mm

Upwardcoarseningmassiveorhorizontallylaminated

siltyclaystone;

upto

10m

thick,includes

dispersed,plant,ostracod,andfish

fossils

Profundal

lacustrinedeposition(below

fair-w

eather

wavebase)

P2:Sh,Sm,Mh

2m

thickstacksofupwardfininglaminae

upto

3cm

thick;basal

surfaces

ofindividual

laminae

areabruptandtheupwardtransitionto

siltstone

isgradational,lower

portionismassiveandgrades

upward

tohorizontallylaminated

Stacked

depositsofturbiditycurrents(primarilyBoumaA

andB)

[Bouma,1962;Giovanoli,1990;Lowe,

1982;Muttiet

al.,2003]

A1:Gcm

,Gch,Gcm

iPoorlysorted,veryangularto

subrounded,pebble

toboulder

conglomerate;

0.5–20m

thick;

erosivebases,horizontalstratificationandlong-axistransverse

imbrication

High-concentration,unconfined

sheetflooddeposits[BlairandMcP

herson,

1994b;Blair,2000;DeC

elleset

al.,1991;Nem

ecandSteel,1984;

Pierson,1980,1981]

A2:Gmm,Gcm

Massive,

unorganized,matrix-supported

boulder

conglomerate0.5–3m

thick;

noerosivebases

Debrisflow

deposits[BlairandMcP

herson,1994a;

Pierson,1980]

A3:Gcm

i,Gch,Gcf,Gt

Imbricated,lenticular,clast-supported,pebble-boulder

conglomeratebodies;

erosivebases

andhorizontalandcross

stratification

Depositedbystream

sin

alluvialfanchannels[D

eCelleset

al.,1991]

A4:Gx

Chaoticmatrix-andclast-supported

boulder

conglomeratedeposits;

abundantload

castsandballandpillow

structures(upto

2m

ofrelief

onsomestructures)

High-concentrationfloodordebrisflow

depositsinto

awater-saturated

environment[Blair,2000;Miall,2000;Nem

ecandSteel,1984;

Pivnik,1990;Postma,1983]

aWeidentify

14lithofacies

associationsand5depositionalenvironments.Depositionalenvironmentsareas

follows:F,fluvial;S,supralittoral;L,littoral;P,profundal;andA,alluvialfan.Modernanalogsto

theseenvironmentsoccurboth

atKungyuCo(N

30�360 ,E82�110 )andto

thewestofZhongba(N

26�,E84�)

insouthernTibet.


9 of 24

TC1004

grains that include exotic lithic fragments. Lath work volcanicgrains contain lath-shaped plagioclase crystals and are typi-cally felsic in composition. Vitric volcanic grains are thosewith pseudoisotropic textures. These commonly resemblechert, but, where stained, have taken a pink plagioclase stain.In most thin sections plagioclase is slightly more abundantthan potassium feldspar. Both biotite and muscovite arepresent in most samples with muscovite typically being morecommon. Accessory minerals include zircon, tourmaline,garnet, olivine, serpentine, chlorite, and cordierite. The mostabundant cement is carbonate with secondary occurrences ofgypsum cementation.[17] Sandstones from the lower conglomeratic and fluvial

members of the Zhada Formation consist of poorly lithifiedcarbonate- or gypsum-cemented litharenites. Subroundedto subangular lithic grains, including polycrystalline quartz,are the dominant constituents with lesser amounts of mono-crystalline quartz grains (Figures 9b, 9e, and 9h). The lithicfraction is dominated by felsic and vitric volcanic grains(Figures 9c and 9i). Feldspar constitutes �10%–20% of thetotal compositions (by comparison with either Qt and L orQm and Lt) with plagioclase being slightly more abundant.Whereas volcanic grains are the most abundant lithic grainsin lower member sandstones, in some samples up to 66%of the lithic portion is composed of either metamorphic orsedimentary lithic grains (Figure 9c samples 0SZ1 and0.2SZ30).[18] As in the lower members, sandstones from the upper

members consist of poorly lithified carbonate- or gypsum-cemented litharenites. However, starting in the middle of thelower fluvial member the composition of sandstones changesabruptly. This is particularly evident in the lithic fraction,which changes from being dominated by volcanic grains tosubequal amounts of sedimentary and metamorphic grains(Figures 9c and 9i) or entirely metamorphic grains (Figure 9f).Feldspar constitutes a larger component of upper membersandstones (10%–55% of the total), with the proportion of

feldspar increasing up section. Plagioclase dominates overalkali feldspars. In contrast to lower member sandstones, vol-canic lithic grains are a minor constituent of upper membersandstones.[19] Conglomerate clast count data are consistent with

modal sandstone analyses. Sedimentary and volcanic clastsdominate the lower members of the Zhada Formation;metamorphic clasts become more abundant up section,and volcanic clasts less so (Figure 3).4.2.2. U-Pb Geochronologic Analyses of Detrital Zircons[20] A total of 840 new zircon ages from 10 samples are

reported here. The preferred ages are shown on relative age-probability diagrams [from Ludwig, 2003]. These diagramsshow each age and its uncertainty (for measurement erroronly) as a normal distribution and sum all ages from asample into a single curve (Figure 10).[21] Detrital zircon samples fall broadly into two groups:

those with significant peaks >100 Ma and those with almostexclusively <100 Ma detrital zircon ages (Figure 10d). Nocorrelation was found between the degree of zircon grainshape and age. The youngest zircon age population is 23 Ma(n = 8, LwrCgm58.15) though a small cluster (n = 2) are asyoung as 17 Ma (2EZ66.5; Figure 10 and Data Set S2). The>100 Ma age population has prominent peaks at 450–550,800–1200, and 2500 Ma and minor peaks at 550–650,1300–1450 and 2100 Ma. The <100 Ma population hasmajor peaks at�35, 45–50, and�55 Ma and minor peaks at17–23, 60–65 and 90–100 Ma.

4.3. Interpretation

4.3.1. Sandstone and Conglomerate Provenance[22] Detrital modes of sandstones show an up-section

shift from an arc-orogen source to a recycled orogen source.The lowermost samples are from sandstones within thefluvial Nlc and Nlf members. Their lithic fractions fall largelywithin the arc-orogen source field in Qp-Lv-Ls ternarydiagrams and straddle the border between the arc-orogen

Figure 6. Stratigraphic cross section across the Zhada basin showing the relationship between ZhadaFormation members. Cross-section line is indicated in Figure 3. Vertical exaggeration is 10X.


10 of 24

TC1004

and recycled orogen fields in Qm-F-Lt and Qt-F-L diagrams(Figure 9) [Dickinson, 1985]. Sandstone compositions fromthe top of the Nlf member are more scattered but trend towardthe quartzose, recycled orogen fields in all ternary diagrams(Figure 9). Samples from the upper members of the ZhadaFormation (Nl, Nuf, Nal) continue the trend toward morecompositionally mature, recycled orogen sources, evenextending into the continental block field in Qm-F-Lt andQt-F-L diagrams (Figures 9g and 9h). These trends are alsopresent in samples from the northwestern part of Zhadabasin (Figures 9e and 9f).

[23] The primary source of arc-orogen-derived volcanicdetritus for samples from the fluvial interval of the lowermember of the Zhada Formation is interpreted to be theMount Kailash region to the northeast of Zhada basin (northof the Karakoram fault; Figures 1 and 8). The region southof the Zhada basin is dominated by high-grade metamorphicrocks and the only local sources of volcanic detritus arerelatively small zones of generally mafic ophiolitic rocks inthe Tethyan Himalaya. In contrast, the Mount Kailashregion is dominated by both volcanic and plutonic rocksassociated with the Cretaceous-Tertiary Gangdese magmatic

Figure 7. (a) Gravel foresets of lithofacies F1. Foresets are 3–4 m tall. (b) Large-scale trough crossstratification in F2 sandstones. (c) Coarse progradational sequences at the top of the Zhada Formation.Visible in this image is an organic-poor example of lithofacies P1 and lithofacies A4 and A1. TS,transgressive surface; P, upward coarsening trends inferred to represent progradation. (d) Horizontallystratified and planar and low-angle climbing ripple cross stratification of lithofacies S1. (e) F2 (nowcontorted) overlying marshy wetland deposits (F3). Staff is �50 cm long. (f) Upward fining laminae oflithofacies P2 that are interpreted as turbidites.


11 of 24

TC1004

arc and younger clastic rocks derived from these igneousrocks [Yin et al., 1999; Aitchison et al., 2002; Murphy et al.,2002; Murphy and Yin, 2003]. This area is also the head-waters of the modern Sutlej River (Figure 8b). Probablesources of the recycled orogen detritus which dominatesthe upper members of the Zhada Formation include rocksin the Ayi ShanMountains northeast of the basin, the Tethyansequence strata which underlie the basin, and the GreaterHimalayan and Tethyan sequence rocks southwest of thebasin. Greater Himalayan rocks are also exposed in the GurlaMandhata dome to the southeast of Zhada basin [Murphy,2007].[24] The modal petrography points to an up-section shift

from a northeasterly source to a source located southwest ofthe Karakoram fault. The enrichment in recycled orogendetritus is also evident in the northwestern Zhada basin.However, being farther from the inferred source of volcanicdetritus, the initial volcanic fraction is reduced and so theevolution from an arc-orogen dominated source to a more

quartzose source is not as dramatic as in central Zhadabasin.[25] Like the petrographic data, conglomerate clast count

data show an up-section decrease in the abundance ofvolcanic detritus and an increase in the relative abundanceof high-grade metamorphic and plutonic clasts (Figure 3).Particularly in the southern margin of the basin, wherepaleocurrent indicators from the upper members of the ZhadaFormation show northward paleoflow directions, the onlysource of high-grade metamorphic clasts is the GreaterHimalayan sequence to the south of Zhada basin. However,low-grade metasedimentary clasts dominate the clast countsthroughout the basin, indicating that the local Tethyanbasement was a significant source of detritus. In the northernand northwestern parts of the basin, paleocurrent data indi-cate that high-grade metamorphic and plutonic rocks werederived from the footwall of the Qusum detachment. Con-sistent with greater distance from the source of volcanicrocks (the Kailash Range), the ratio of volcanic clasts to

Figure 8. Paleocurrent data for (a) the lower members of the Zhada Formation and (b) the uppermembers of the Zhada Formation superimposed on topography. Paleocurrent data show an up-sectionevolution from uniformly northwestward directed to basin-centric paleocurrent indicators. Numbers inrose diagrams identify measured sections in Table S2. The outline of Zhada basin sediments is shown inFigure 8a. For comparison, the modern drainage network is shown in Figure 8b. The modern Sutlej Riverrises in the region of Rakkas tal and Manasarovar (near Mount Kailash) and exits the Zhada basin nearthe Leo Pargil Dome.


12 of 24

TC1004

Figure 9. Ternary diagrams showing modal petrographic data from this study. Samples show anup-section trend from a northern Gangdese Arc dominated source to a southern Himalayan fold-thrust beltdominated source. The change in source terrane coincides with the reorientation of paleocurrent indicators(Figure 8). Fields are from Dickinson [1985].


13 of 24

TC1004

other clasts is lower in the northwestern Zhada basin thanin the central Zhada basin. The conglomerate clast count dataindicate that the source of sediments in the Zhada basinchanged from a distal, northeastern source, with local input,to a predominantly proximal southwestern source.4.3.2. U-Pb Detrital Zircon Geochronology[26] Samples with significant detrital zircon ages <100Ma

come from the bottom of the Zhada Formation and thosewith almost exclusively >100 Ma ages come from the top(Figure 10d). Possible sources for detrital zircons includethe Cretaceous-Tertiary igneous rocks related to Gangdese

(Kailash) arc magmatism, Paleozoic-Mesozoic metasedi-mentary rocks of the Tethyan Sedimentary sequence andlate Proterozoic–early Paleozoic rocks of the GreaterHimalayan sequence.[27] Good correlation exists between most of the <100 Ma

age peaks with zircon age spectra from regions north ofZhada basin (compare the left plots of Figures 10a, 10b and10c). Specifically, age peaks at 17–23 Ma, �55 Ma and78–80 Ma correlate with igneous or detrital zircon agespectra from the Mount Kailash area (Figure 10a). The onlydocumented source for�55Ma zircons is from the Gangdesearc (Figure 10a). However, as there are no good correlationsfrom this region for peaks from 35 to 50 Ma, these zirconsmust have been derived from the Ayi Shan. Ayi Shan zirconsalso may contribute to the 78–80Ma peak in the Zhada basin(Figure 10a).[28] Ages of detrital zircons from the top of the Zhada

Formation (e.g., 4SZ14, 3NWZ45.9 and 3NWZ77) aredominantly Paleozoic-Proterozoic; these cannot have beenderived from the Mount Kailash region. The NeogeneKailash Formation has a greater proportion of <100 Mazircons than do the samples mentioned above (Figure 10a,Kailash). Hence, the Kailash Formation may be a minorsource, but the major source of detritus must lie elsewhere.Likely sources include inherited zircons from Ayi Shanparagneisses or leucogranites which have Tethyan sequenceprotoliths, the Tethyan sequence basement, or GreaterHimalayan sequence rocks (compare the right plots ofFigures 10a, 10b and 10c) [DeCelles et al., 2000, 2004;Martin et al., 2005; Gehrels et al., 2006a, 2006b, 2008].Consistent with the petrographic data, the detrital zircon datashow a significant up-section shift from a source in theMountKailash area to a local basement or southwestern source.[29] There is an apparent disconnect between the relative

abundance of detritus derived from northeast of Zhada basin

Figure 10. U-Pb relative age-probability density diagramsfor (a) possible source terranes for Zhada basin sedimentand (b, c, and d) detrital zircon samples from this study.Figure 10b shows detrital zircon samples from the northwestof Zhada basin with age populations split into <100 Magrains in the left plots and >100 Ma grains in the right plots.Figure 10c shows detrital zircon samples from central Zhadabasin with age populations split into <100 Ma grains in theleft plots and >100 Ma grains in the right plots. Figure 10dshows detrital zircon samples from this study plotted withoutsplitting young and old age populations. Within Figures 10b,10c, and 10d, samples are plotted stratigraphically up sectionfrom bottom to top. Sample names correlate to measuredsections and heights within thosemeasured sections. Samplesshow an up-section change from a Kailash or Ayi Shansource to a local (Tethyan sequence) or southerly (GreaterHimalayan sequence) source that is coincident with thereorientation of paleocurrent indicators (Figure 8). Stars inFigure 10a indicate igneous ages, and colors correlate withtext color. Bin sizes for histograms associated with relativeage-probability diagrams are 5 Myr for ages between 0 and100 Ma and 50 Myr for ages between 100 Ma and 3500 Ma.Source region data from Gehrels et al. [2008].


14 of 24

TC1004

indicated by the detrital zircon data and the conglomerateclast count data. The conglomerate clast counts indicate adominance of low-grade metamorphic rocks (i.e., TethyanSequence) over volcanic rocks (i.e., Gangdese source) in thelower members of the Zhada Formation while the detritalzircon data show the opposite. One possible reason for thisis downstream grain-size sorting. Unstable volcanic andigneous clasts were trapped or disaggregated close to theirsource north of the Zhada basin while sand-sized igneouszircon grains were transported to and deposited together withlocally derived low-grade metasedimentary clasts in theZhada basin. This explanation is consistent with a moredistal northeastern source and a more proximal local orsouthwestern source.

5. Subsidence Curve

[30] A tectonic subsidence curve for the Zhada basin wasproduced from the South Zhada section by decompactingthe sediments using standard backstripping techniques(Figure 11a, decompacted subsidence curve) [Vanhinte,1978; Dickinson et al., 1987; Allen and Allen, 1990] andthen removing the load of the sediment assuming Airyisostacy (Figure 11a, tectonic subsidence curve). Age controlis based on magnetostratigraphic tie points (Figure 5) [Sayloret al., 2009]. Bathymetry was not a major contributor tosediment accommodation because the lake was never morethan a few tens of meters deep based on the thickest observedupward coarsening parasequence within the Zhada Forma-tion. The essentially linear Zhada basin curve best matches

typical subsidence curves from strike-slip or rift basins(Figure 11b), though the subsidence rate is low: 0.09 mm/yrand 0.06 mm/yr for the decompacted and tectonic subsidencecurves, respectively. In addition, Zhada basin fill is thinnerthan classic rift and strike-slip basin fills, probably as a resultof both low subsidence rates and the brief duration of thebasin. Though exceptions are admitted rift basins are typi-cally longer lived than supradetachment basins [Friedmannand Burbank, 1995].

6. Discussion

6.1. Zhada Basin Evolution

[31] Sedimentology and paleocurrent data indicate that alarge, northwestward flowing, low-gradient river depositedthe lower members of the Zhada Formation. Whereas thismay seem to contradict the argument for a northern sourcefor the volcanic detritus based on petrography, clast com-position, and detrital zircon analyses, we note that themodernSutlej River has its headwaters in the Mount Kailash region,flows southward into, and then northwestward along theZhada basin and currently exits the basin to the southwest(Figure 8b). The presence of volcanic detritus in the lowermember of the Zhada Formation indicates that the paleo-Sutlej River followed a similar path into the Zhada basin andflowed northwestward along the basin axis. Paleocurrentindicators at the base of the Zhada Formation are orientednorthwestward even within 10 km of the modern Qusumrange front, indicating that, at the time that the sediments

Figure 11. (a) Decompacted and tectonic subsidence curves for the Zhada basin. (b) Tectonicsubsidence curve for the Zhada basin compared to tectonic subsidence curves for a variety of tectonicsettings [from Allen and Allen, 1990; Angevine et al., 1990].


15 of 24

TC1004

were deposited, the Leo Pargil/Qusum Range presented noobstacle to northwestward flow.[32] Up section, paleocurrent indicators, provenance data,

and sedimentology indicate that the Zhada basin becamefully or partially closed, and that detritus was derived fromlocal highlands surrounding the basin. Northwestwardpaleoflow suggests that the cause of ponding within Zhadabasin is to be found to the northwest. Deformation andgrowth structures (Figure 12) in the Zhada Formation in thenorthwestern part of the basin, near the Leo Pargil/QusumRange, indicate that movement on the Qusum detachmentand exhumation of the range were synchronous with depo-sition of the Zhada Formation. It is unlikely that the Zhadabasin was entirely hydrologically closed because a largeriver, such as the Sutlej, would have quickly filled andoverflowed a closed basin. Moreover, the Zhada Formationcontains no significant thicknesses of evaporites, whichwould be expected in a closed basin. Some outflow probablycontinued either via a small outlet in the southwest of thebasin or groundwater flow which would not be detected bypaleocurrent data [Duhnforth et al., 2006]. However, paleo-current indicators and provenance data show that majorsurface flow during the Pliocene-Pleistocene was basin-centric and stable isotope data indicate that the lake under-went significant evaporation [Saylor et al., 2009].[33] The presence of Zhada basin sediments in buttress

unconformity on top of deformed Tethyan sequence strataindicates that there was some paleotopography on the

basement prior to deposition of the Zhada Formation. Inmeasured sections located near Tethyan basement rocks,alluvial fan facies interfinger with fluvial facies. Paleocurrentindicators from alluvial fan facies are commonly at highangles to those from axial fluvial facies. This was observedparticularly in the northwestern end of the basin and near theGuga section and is the cause of the anomalous northwardoriented petals on the 2NWZ, 3NWZ and 1NZ rose diagramsand east or southward oriented petals on the Guga rosediagrams (Figure 8). These data add detail to environmentalreconstructions, but do not change the overall basin evolutionpicture.

6.2. Tectonic Origins of the Zhada Basin

[34] Any model for Zhada basin must explain (1) thetransition from nondeposition to deposition at �9.2 Ma,(2) the relationship of the Zhada Formation to preexistingtopography in the Tethyan fold-thrust belt, (3) the varietyand stacking patterns of depositional environments in theZhada Formation, (4) the lack of widespread deformationin the Zhada Formation, (5) the lack of cyclic incision andinfilling or significant hiatuses in sedimentation in theZhada Formation, and (6) the transition from deposition tonondeposition and incision at <1 Ma.[35] All evidence indicates that closure of the Zhada

basin was the result of uplift of the Leo Pargil and possiblyGurla Mandhata domes, which created tectonic dams or sills

Figure 12. Growth structure in Zhada Formation in the proximal hanging wall of the Qusumdetachment fault. This is the only significant deformation observed in the Zhada basin. Telephone polesare 8 m tall. View is toward the northeast. The normal fault shown is antithetic to the top-to-the-east masterQusum detachment fault (not visible, to the west). See text for additional discussion of the growth-stratageometry.


16 of 24

TC1004

and caused basin subsidence. Exhumation of both the LeoPargil/Qusum Range and the Gurla Mandhata dome wassynchronous with sedimentation in the Zhada basin. Theonly basement-involved deformation observed in the ZhadaFormation is a growth structure at the northwestern end of thebasin,�10 km from the Qusum detachment fault (Figure 12).The observed structure is adjacent to the Qusum measuredsection (Figures 2 and 3) which we correlate to the base of theZhada Formation (�9–4.5Ma). The growth-stratal geometryis best explained by progressive sedimentation and rotationof basin sediments in the proximal hanging wall above ablind, upward propagating normal fault [Gupta et al., 1999;Sharp et al., 2000; Gawthorpe and Hardy, 2002]. Paleocur-rent indicators from the base of the Zhada Formation areuniformly northwestward, even in the Qusum measuredsection (Figure 2 and 8a), indicating that prior to uplift ofthe Leo Pargil/Qusum Range, the paleo-Sutlej River flowedunobstructed toward the northwest. The linkage betweenbasin formation and exhumation is less direct for the GurlaMandhata detachment than for the Qusum detachment.However, several lines of evidence suggest that exhumationon the Gurla Mandhata detachment also drove basin sub-sidence. (1) The temporal coincidence between exhumationof the Gurla Mandhata dome and basin development allowsa causal link. (2) The Gurla Mandhata detachment accom-modates between 35 and 66 km of displacement andexhumation [Murphy et al., 2002] but the basin in itsimmediate hanging wall (Pulan basin) is less than 15 kmwide. (3) Rollover structures in the Pulan basin fill showthat the Gurla Mandhata detachment flattens at an extremelyshallow depth. The implication is that exhumation-relatedsubsidence was not focused in the proximal hanging wall andbut rather was accommodated in the distal hanging wall.[36] Basin subsidence in the hanging wall of the Qusum

(and possibly Gurla Mandhata) detachment was spatiallyvariable. Correlation of magnetostratigraphy between mea-sured sections shows that there has been more subsidencein the basin center (the South Zhada and Guga sections)than toward the basin margins (East Zhada or SoutheastZhada sections) (Figure 3) [Saylor et al., 2009]. Significantdifferential subsidence and sediment accumulation are evi-dent between the South Zhada and Southeast Zhada sec-tions between 9.2 Ma and 7 Ma (Figure 3a). Differentialsubsidence between the South Zhada and East Zhada sec-tions provides evidence for ongoing subsidence post-7 Ma(Figure 3b). Additional evidence for basin subsidence isprovided by the change in paleoslope indicated by Figure 3a.Paleocurrent analysis indicates that the paleoslope at the timeof deposition of the lower Zhada Formation was northwest-ward (to the right in Figure 3a). This would have beenimpossible if the basin was in its current configurationbecause both the southeast (Southeast Zhada section) andnorthwest portions of the basin (1Northwest Zhada andQusum sections) are higher than the center of the basin. Thiscould result from subsidence of the basin center or fromuplift of the northwestern margin of the basin. Regardless,it provides evidence for removal of footwall rocks frombeneath the center of the basin toward the northwest, andpossibly southeast, of the basin.

[37] Migmatites and paragneisses were exhumed frommidcrustal levels in both the Leo Pargil/Qusum Range andthe Gurla Mandhata dome [Zhang et al., 2000; Murphy etal., 2002; Thiede et al., 2006]. Kinematic analysis indicatestop-to-the-east sense of shear on the Qusum detachment[Zhang et al., 2000] and top-to-the-west sense of shear onthe Gurla Mandhata detachment [Murphy et al., 2002]. Inboth cases, the midcrustal footwall rocks have been exhumedfrom beneath the Zhada basin synchronously with sedimen-tation in the hanging wall. The thin basin fill, short-livedextension and low subsidence rate are characteristic of supra-detachment basins [Friedmann and Burbank, 1995], which isgenerally consistent with the regional tectonic setting ofZhada basin in the hanging walls of the Qusum and GurlaMandhata detachments.[38] The elongate shape of the Zhada basin reflects the

spatial extent of subsidence caused by exhumation. TheGurla Mandhata detachment is kinematically linked to andspatially limited on the north side by the Karakoram fault[Murphy et al., 2000, 2002; Murphy and Burgess, 2006].Although a similar relationship has not been documentedfor the Leo Pargil/Qusum system, the abrupt northwardtermination of arc-parallel extension as both detachmentfaults approach the Karakoram fault, the strain compatibilitybetween the Leo Pargil and Qusum detachments and theKarakoram fault, and the synchronicity in deformationbetween the Leo Pargil/Qusum Range and the Karakoramfault indicate that all three systems, Karakoram, Leo Pargil/Qusum, and Gurla Mandhata, may be kinematically linked[Zhang et al., 2000; Phillips et al., 2004; Thiede et al.,2006; Valli et al., 2007], and that the Zhada basin istherefore a hybrid supradetachment/strike-slip basin. Oneattractive aspect of this interpretation is that these structurescan accommodate arc-parallel extension and be the primarycontrols on Zhada basin evolution in the absence of a largestrike-slip faulting on the southwestern margin of the basin(see section 7).[39] The onset of sedimentation in Zhada basin at�9.2Ma

is within error of the oldest AFT ages determined by Thiedeet al. [2006] from the west side of the Leo Pargil Dome. Theincrease in exhumation rate indicated by the AFT data wasalso synchronous with an increase in tectonic subsidencerate at 5.23–2.58 Ma and with the onset of lacustrinesedimentation in the Zhada basin [Thiede et al., 2006]. Thisfurther implicates exhumation of the Qusum/Leo PargilRange as a controlling factor in Zhada basin development.The synchronicity of movement on the Karakoram fault inthe Zhada region (10 Ma [Yin et al., 1999; Murphy et al.,2000]) and exhumation of Gurla Mandhata (9 Ma [Murphyet al., 2002]) with the 9–10 Ma AFT ages from Leo Pargiland the 9.2 Ma onset of sedimentation in Zhada basinpoints to the late Miocene as the time of major structuralreorganization in the area.[40] The presence of �15 Ma 40Ar/39Ar cooling ages,

from both the west side of the Leo Pargil Dome and the eastside of the Leo Pargil/Qusum Range, has been used to arguefor a mid-Miocene onset of exhumation [Zhang et al., 2000;Thiede et al., 2006]. If true, the fact that sedimentation inZhada basin did not begin until the onset of exhumation of


17 of 24

TC1004

the Gurla Mandhata dome, despite apparent ongoing exhu-mation of the Leo Pargil/Qusum Range, indicates that thecombination of sill uplift and basin subsidence was necessaryfor basin formation in this region. However, in an alternatescenario, mid-Miocene 40Ar/39Ar cooling ages may reflectcooling during slip on the Main Central Thrust [Robinson etal., 2003, 2006]. This is suggested by the abundance of earlyto mid-Miocene 40Ar/39Ar cooling ages from the GreaterHimalayan and Tethyan sequences on the south flank of theHimalaya [Metcalfe, 1993; Vannay et al., 2004; Thiede et al.,2005; Robinson et al., 2006]. Thus, the mid-Miocene coolingages obtained from the Leo Pargil/Qusum Range may beinherited from tectonic events related to arc-normal thrustingrather than arc-parallel extension. Determining which sce-nario is accurate, and hence whether exhumation of Gurla

Mandhata played a role in subsidence and sedimentation inZhada basin, must await additional research.[41] Finally we point to an anomalous valley, possibly a

wind gap, striking orthogonally across the Leo Pargil/QusumRange (32.33�N, 79�E; Figure 13) as the possible outlet ofthe paleo-Sutlej River. This is the only significant valley inan otherwise unbroken mountain range. The valley extendsacross the entire width of the range and yet is currentlyoccupied by an underfit stream <1 m deep and <5 m wide.This stream originates at a glacier midway through the valleyand thus cannot account for a valley transecting the entirerange. The valley is mantled by terrace deposits that haverecently been incised (Figure 13b). We tentatively interpretthis valley as the course of the paleo-Sutlej River prior toexhumation of the Leo Pargil/Qusum Range. Whereas this

Figure 13. (a) LandSat image of the northwestern corner of the Zhada basin showing a wind gap,possibly representing the course of the paleo-Sutlej River. (b) A view to the southeast from within thevalley in Figure 13a. Inset box in Figure 13a is enlarged in Figure 13b. Note the relict surface that hasbeen abandoned and incised by the modern river which occupies this valley. The cliff face indicated by‘‘a’’ is �20 m tall.


18 of 24

TC1004

interpretation remains speculative pending further investiga-tion, if true, the paleo-Sutlej may have been the headwaters ofeither the Indus or the Chenab Rivers as suggested byBrookfield [1998].[42] Several explanations for the existence of the Zhada

basin can be ruled out based on the variety and stackingpatterns of lithofacies assemblages in the Zhada Formation.The basin exhibits none of the lithologic, stratigraphic orstructural features of flexural or rift basins. Specifically,there are no thrust or high-angle normal faults adjacent to thebasin that were active during the mid to late Miocene. Thetectonic subsidence curve does not match the classic upwardconvex flexural basin subsidence curve, and the uniformlyupward coarsening textural trend and strong asymmetrytypical of flexural basins are absent (Figures 3 and 6). Unlikerift basins, where subsidence is maximized adjacent to themaster fault [Gawthorpe and Leeder, 2000], the greatestsubsidence took place in the middle of the basin; far removedfrom any structures. The basin lacks the wedge-shaped orsteep-sided cross-sectional geometry which might indicatethe presence of a half-graben or graben. Whereas the sub-sidence curve superficially matches that of other rift basins,the Zhada Formation is much thinner than typical rift basinfill [Friedmann and Burbank, 1995] and mapping of thenortheastern and southwestern margins of the basin revealedno geometric relationships other than onlap of basin sedi-ments onto the basin bounding highlands. In no place alongthese margins is the basin fill faulted.[43] The duration of the basin and the lack of incision or

hiatuses in sedimentation indicate that climate change wasnot the direct cause of sedimentation. It is difficult to envisiona scenario in which climate change, acting alone, wouldinitiate sediment accumulation and basin development, pro-duce the variety of sedimentary environments archived inthe Zhada Formation, drive continuous sedimentation for�9 Myr, and then abruptly terminate deposition and returnthe region to an erosional, incising state. The basin could alsobe the result of damming by landslides, glaciers or tectonicactivity. However, the duration of the basin and the lack ofglacial deposits argue against the first two causes, leavingonly structural damming as a viable possibility.[44] The South Tibetan detachment system (STDS) had

little to no effect on the Zhada basin. The Zhada Formationon the southwestern margin of the basin is undeformed andonlaps preexisting topography on the southwestern marginof the basin as far southward as the trace of the STDS.Where documented elsewhere in the Himalaya, the youn-gest slip on the STDS predates the onset of sedimentation inZhada basin [Hodges et al., 1992, 1996; Searle et al., 1997;Murphy and Harrison, 1999; Hodges, 2000; Murphy andYin, 2003; Searle et al., 2003; Yin, 2006; Cottle et al., 2007].In contrast to quiescence on the STDS, kinematic analysesof the Gurla Mandhata and Qusum detachments indicatethat these faults accommodate arc-parallel extension and areactively exhuming midcrustal rocks from beneath the Zhadabasin [Zhang et al., 2000; Murphy et al., 2002]. Structuraland isotopic analysis also indicates that the Gurla Mandhatadetachment cuts the STDS, indicating that slip on the STDS

had ceased prior to 9 Ma [Murphy et al., 2002; Murphy andCopeland, 2005; Murphy, 2007].

7. Synthesis

[45] We divide the evolution of the Zhada basin into5 stages (Figure 14). Before 9.2 Ma the Zhada basin wasoccupied by a large, northwestward flowing, braided riverwhich had its headwaters in the Mount Kailash region(Figure 14a). Exhumation of the Leo Pargil/Qusum Rangeand the Gurla Mandhata gneiss dome along a system ofnortheast striking detachment faults beginning at �10–9.2Ma decreased the fluvial gradient and caused ponding andupstream sediment accumulation. Before being completelydefeated, the paleo-Sutlej River cut and flowed through thevalley identified above as a wind gap (Figure 14b). Thecombination of sill uplift and basin subsidence (owing tosimultaneous footwall uplift and hanging wall subsidenceduring slip on the detachments) resulted in basin closureand lacustrine and alluvial fan sedimentation upstream ofthe dam between �5.6 Ma and <1 Ma. The Zhada basinbecame a holding pond for detritus derived from local high-lands (Figure 14c). This situation continued until the sill wasbreached at the current outlet in the southwestern corner ofthe basin at <1 Ma (Figure 14d), probably by either head-ward erosion or overtopping. Following breach of the sill themodern Sutlej River system began to reequilibrate to the newbase level resulting in upstream migration of a knickpointand incision of the Zhada Formation (Figure 14e). A similargeomorphological evolution characterized the Thakkholagraben in central Nepal [Garzione et al., 2003]. Integrationof the modern Sutlej drainage network and enhanced erosionassociated with upstream knickpoint migration also providea mechanism to explain the post-5 Ma reorientation of trans-Himalayan rivers and increase in Himalayan, as opposed tosuture zone, detritus delivered to the Indus fan [Clift andBlusztajn, 2005].[46] Arc-parallel extension is clearly accommodated by

extension across the Leo Pargil/Qusum Range and GurlaMandhata, and by dextral slip on the Karakoram fault [Zhanget al., 2000;Murphy et al., 2002; Phillips et al., 2004; Thiedeet al., 2006]. It is unclear which structures accommodateextension on the southwestern side of Zhada basin. No large-scale structure comparable to the Karakoram fault is presenton this side of the basin. It is possible that extension istaken up by distributed faulting directly south of the basin(Figure 15b). Another possibility is that slip on the Leo Pargiland Qusum detachments is transferred westward or north-westward, a process similar to that described by Murphyand Copeland [2005] for the Gurla Mandhata detachment(Figure 15b).[47] The arc-parallel extension process provides a mech-

anism to explain the loss of elevation suggested by Saylor etal. [2009]. They inferred a 0.8–1.5 km loss of elevationbased on oxygen isotope paleoaltimetry. Assuming Airyisostacy, that loss of elevation would require between 4.5and 8 km of crustal thinning. This could be achieved by twointeracting detachment fault systems below the Zhada basin


19 of 24

TC1004

(Figure 15d). The Gurla Mandhata detachment fault exca-vates structurally deeper rocks [Murphy, 2007] and thus isinferred to cut the Qusum detachment fault. The zone ofinteraction of these two fault zones is inferred to be below

the zone of maximum subsidence (the depocenter) of theZhada Basin. A flat in the Gurla Mandhata detachment isinferred based on a zone of minimal subsidence (and non-deposition) between the Pulan and Zhada basins. Flattening

Figure 14. Perspective diagram showing the interpreted basin evolution of the Zhada basin from athroughgoing river to a closed-basin lake before final breaching of a new sill resulting in integration ofthe modern Sutlej River.


20 of 24

TC1004

of the Gurla Mandhata detachment is also indicated byrollover structures in the Pulan basin fill. In Figure 15d,the preextension depth of the STDS is inferred fromDeCelles et al. [2001]. Crustal thinning and the associatedelevation loss also explain the anomalously low elevations inthe Zhada region.

8. Conclusions

[48] The sedimentary facies in the Zhada formation showthat depositional environments in the Zhada basin included

braided fluvial, marshy wetland, lacustrine, dune field, sandflat, lacustrine fan delta, and alluvial fan systems. Through-out the basin’s history the climate was sufficiently arid toproduce dune fields, evaporative marshes, evaporative sandflats, and bedded gypsum. These environmental conclusionsare consistent with high d18O values from lacustrine gastro-pods [Saylor et al., 2009].[49] Paleocurrent data indicate that the Zhada basin

evolved from a northwestward draining river valley to abroad depositional plain and finally to a largely closed basin.Closure of the basin resulted in lacustrine and lake margin

Figure 15. Paleogeographic maps showing the influence of arc-parallel extension on the Sutlej Riverand Zhada basin. (a) Prior to 9.2 Ma the paleo-Sutlej River drained regions north of the Indus suture andflowed to the northwest. (b) With the onset of arc-parallel extension and exhumation of the GurlaMandhata and Leo Pargil/Qusum core complexes the Sutlej River begins to pond, eventually forming thelarge Zhada paleolake. Arc-parallel extension is accommodated on the southwestern side of the basineither by distributed shortening, shown schematically by circle a, or by transferring extension to the westor northwest (circle b). (c) In the Pleistocene a new sill is breached in the west of the basin, resulting inthe integration of the modern Sutlej River system. (d) NW–SE cross section showing the relationshipbetween extension on the Qusum and Gurla Mandhata detachment faults and sedimentation in the Zhadaand Pulan basins. The cross-section line is indicated in Figure 15c.


21 of 24

TC1004

alluvial fan sedimentation until the drainage sill wasbreached in the southwest corner of the basin. The changein sedimentation style was accompanied by a change insediment provenance, from a volcanic-rich source in theMount Kailash region, to more local metamorphic/clasticsources, mainly in the Tethyan and Greater Himalayansequence rocks evidenced by modal petrography and detritalzircon U-Pb frequency spectra.[50] Sedimentation in the basin was due to a combination

of topographic damming by sill uplift and tectonic subsi-dence. Integrating this basin history with existing regionalstructural models suggests that the Zhada basin is the resultof arc-parallel extension which was accommodated oncrustal-scale detachment faults at the northwest and possibly

southeast extremities of the basin (Figure 15). Subsidencelikely took place in response to vertical thinning andexhumation of the midcrust during late Miocene detachmentfaulting.

[51] Acknowledgments. The authors would like to thank RichardHeermance for his contribution to this paper. Thanks also for fieldassistance from Jay Quade, Cai Fulong, Jeannette Saylor, and ScottMcBride.We are thankful for thorough reviews by two anonymous reviewersand Tectonics Associate Editor Eric Kirby which helped to improve thismanuscript. U-Pb geochronological analyses were conducted with supportfrom the National Science Foundation grants EAR-0443387 and 0732436with lab help from Victor Valencia. Additional support was provided by theNational Science Foundation Tectonics Program, ExxonMobil, Chevron-Texaco, and the Galileo Circle of the University of Arizona.

ReferencesAitchison, J. C., A.M. Davis, B. Badengzhu, and H. Luo

(2002), New constraints on the India-Asia collision:The lower Miocene Gangrinboche conglomerates,Yarlung Tsangpo suture zone, SE Tibet, J. AsianEarth Sci., 21(3), 251 – 263, doi:10.1016/S1367-9120(02)00037-8.

Allen, P., and J. Allen (1990), Basin Analysis: Principlesand Applications, 451 pp.,Blackwell Sci., Oxford,U. K.

Allen, P., and J. Collinson (1986), Lakes, in SedimentaryEnvironments and Facies, edited by H. G. Reading,pp. 63–94, Blackwell Sci., Oxford, U. K.

Angevine, C. L., P. L. Heller, and C. Paola (1990),Quantitative Sedimentary Basin Modeling, Am.Assoc. of Pet. Geol., Tulsa, Okla.

Armijo, R., P. Tapponnier, J. L. Mercier, and H. Tong-Lin (1986), Quaternary extension in southern Tibet:Field observations and tectonic implications, J. Geo-phys. Res., 91(B14), 13,803–13,872, doi:10.1029/JB091iB14p13803.

Blair, T. C. (2000), Sedimentology and progressive tec-tonic unconformities of the sheetflood-dominatedHell’s Gate alluvial fan, Death Valley, California,Sediment. Geol., 132(3 –4), 233–262, doi:10.1016/S0037-0738(00)00010-5.

Blair, T. C., and J. G. McPherson (1994a), Alluvial fansand their natural distinction from rivers based onmorphology, hydraulic processes, sedimentary pro-cesses, and facies assemblages, J. Sediment. Res.,Sect. A, 64(3), 450–489.

Blair, T. C., and J. G. McPherson (1994b), Alluvial fanprocesses and forms, in Geomorphology of DesertEnvironments, edited by A. D. Abrahams and A. J.Parsons, pp. 354–402, Chapman and Hall, London.

Bouma, A. (1962), Sedimentology of Some FlyschDeposits: A Graphic Approach to Facies Interpreta-

tion, Elsevier, Amsterdam.Bristow, C. S., and J. L. Best (1993), Braided rivers:

Perspectives and problems, in Braided Rivers, editedby J. L. Best and C. S. Bristow, Geol. Soc. Spec.Publ., 75, 1 – 11.

Brookfield, M. (1998), The evolution of the greatriver systems of southern Asia during the CenozoicIndia-Asia collision: Rivers draining southwards,Geomorphology, 22(3 – 4), 285 –312, doi:10.1016/S0169-555X(97)00082-2.

Cant, D., and R. Walker (1978), Fluvial processes andfacies sequences in sandy braided South SaskatchewanRiver, Canada, Sedimentology, 25(5), 625 – 648,doi:10.1111/j.1365-3091.1978.tb00323.x.

Chen, J., andG. Xu (1987), Geologicmap of theGerdakeregion at a scale of 1:1,000,000 and geologic report,Xizang Bur. of Geol. and Miner. Resour., Lhasa,Tibet.

Clift, P. D., and J. Blusztajn (2005), Reorganization ofthe western Himalayan river system after five mil-lion years ago, Nature, 438(7070), 1001 – 1003,doi:10.1038/nature04379.

Collinson, J. (1996), Alluvial sediments, in SedimentaryEnvironments: Processes, Facies and Stratigraphy,edited by H. G. Reading, pp. 37–82, Blackwell Sci.,Oxford, U. K.

Cottle, J. M., M. J. Jessup, D. L. Newell, M. P. Searle,R. D. Law, and M. S. A. Horstwood (2007), Struc-tural insights into the early stages of exhumation alongan orogen-scale detachment: The South TibetanDetachment system, Dzakaa Chu section, easternHimalaya, J. Struct. Geol., 29, 1781 – 1797,doi:10.1016/j.jsg.2007.08.007.

Dam, G., and F. Surlyk (1993), Cyclic sedimentation ina large wave- and storm-dominated anoxic lake:Kapp Stewart Formation (Rhaetian-Sinemurian),Jameson Land, East Greenland, in Sequence Strati-

graphy and Facies Associations, edited by H. W.Posamentier et al., Spec. Publ. Int. Assoc. Sedimentol.,18, 419–448.

DeCelles, P. G., R. P. Langford, and R. K. Schwartz(1983), Two new methods of paleocurrent determi-nation from trough cross-stratification, J. Sediment.Petrol., 53(2), 629–642.

DeCelles, P. G., M. B. Gray, K. D. Ridgway, R. B. Cole,D. A. Pivnik, N. Pequera, and P. Srivastava (1991),Controls on synorogenic alluvial-fan architecture,Beartooth Conglomerate (Paleocene), Wyomingand Montana, Sedimentology, 38(4), 567 – 590,doi:10.1111/j.1365-3091.1991.tb01009.x.

DeCelles, P. G., G. E. Gehrels, J. Quade, B. LaReau,and M. Spurlin (2000), Tectonic implications ofU-PL zircon ages of the Himalayan orogenic beltin Nepal , Sc ience , 288 (5465) , 497 – 499,doi:10.1126/science.288.5465.497.

DeCelles, P. G., D. M. Robinson, J. Quade, T. P. Ojha,C. N. Garzione, P. Copeland, and B. N. Upreti(2001), Stratigraphy, structure, and tectonic evolu-tion of the Himalayan fold-thrust belt in westernNepal, Tectonics, 20(4), 487–509.

DeCelles, P. G., G. E. Gehrels, Y. Najman, A. J. Martin,A. Carter, and E. Garzanti (2004), Detrital geochro-nology and geochemistry of Cretaceous – earlyMiocene strata of Nepal: Implications for timingand diachroneity of initial Himalayan orogenesis,Ear th Plane t . Sc i . Le t t . , 227 , 313 – 330 ,doi:10.1016/j.epsl.2004.08.019.

Dickinson, W. (1985), Provenance relations from detritalmodes of sandstones, in Provenance of Arenites, edi-ted by G. Zuffa, pp. 333–361, D. Reidel, Dordrecht,Netherlands.

Dickinson, W., R. Armin, N. Beckvar, T. Goodlin,S. Janecke, R. Mark, R. Norris, G. Radel, and

A. Wortman (1987), Geohistory analysis of ratesof sediment accumulation and subsidence for se-lected California basins, inCenozoic Basin Develop-ment of Coastal California, edited by R. Ingersolland W. Ernst, pp. 1 –23, Prentice Hall, EnglewoodCliffs, N. J.

Duhnforth, M., A. G. N. Bergner, and M. H. Trauth(2006), Early Holocene water budget of the Nakuru-Elmenteita basin, central Kenya Rift, J. Paleolimnol.,36(3), 281–294, doi:10.1007/s10933-006-9003-z.

England, P., and G. Houseman (1989), Extensionduring continental convergence, with application tothe Tibetan Plateau, J. Geophys. Res., 94(B12),17,561–17,579, doi:10.1029/JB094iB12p17561.

France-Lanord, C., and L. A. Derry (1994), d13C oforganic carbon in the Bengal Fan: Source evolutionand transport of C3 and C4 plant carbon to marinesediments, Geochim. Cosmochim. Acta, 58(21),4809–4814, doi:10.1016/0016-7037(94)90210-0.

Friedmann, S. J., and D. W. Burbank (1995), Rift basinsand supradetachment basins: Intracontinental exten-sional end-members, Basin Res., 7(2), 109 – 127,doi:10.1111/j.1365-2117.1995.tb00099.x.

Ganser, A. (1964), Geology of the Himalayas, 289 pp.Wiley-Interscience, London.

Garzione, C. N., D. L. Dettman, J. Quade, P. G.DeCelles,and R. F. Butler (2000), High times on the TibetanPlateau: Paleoelevation of the Thakkhola graben,Nepal, Geology, 28(4), 339 – 342, doi:10.1130/0091-7613(2000)28<339:HTOTTP>2.0.CO;2.

Garzione, C. N., P. G. DeCelles, D. G. Hodkinson,T. P. Ojha, and B. N. Upreti (2003), East-west exten-sion and Miocene environmental change in thesouthern Tibetan Plateau: Thakkhola graben, centralNepal, Geol. Soc. Am. Bull., 115(1), 3 – 20,doi:10.1130/0016-7606(2003)115<0003:EWEAME>2.0.CO;2.

Gawthorpe, R., and S. Hardy (2002), Extensional fault-propagation folding and base-level change as con-trols on growth-strata geometries, Sediment. Geol.,146(1 – 2), 47 – 56, doi:10.1016/S0037-0738(01)00165-8.

Gawthorpe, R. L., and M. R. Leeder (2000), Tectono-sedimentary evolution of active extensional basins,Basin Res., 12(3 – 4), 195 – 218, doi:10.1046/j.1365-2117.2000.00121.x.

Gehrels, G. E. (2000), Introduction to detrital zirconsstudies of Paleozoic and Triassic strata in westernNevada and northern California, in Paleozoic andTriassic Paleogeography and Tectonics of Western

Nevada and Northern California, edited by M. J.Soreghan and G. E. Gehrels, Spec. Pap. Geol. Soc.Am., 347, 1 – 17.

Gehrels, G. E., P. G. DeCelles, T. P. Ojha, and B. N.Upreti (2006a), Geologic and U-Pb geochronologic


22 of 24

TC1004

evidence for early Paleozoic tectonism in theDadeldhura thrust sheet, far-west Nepal Himalaya,J . As ian Earth Sc i . , 28 (4 – 6) , 385 – 408 ,doi:10.1016/j.jseaes.2005.09.012.

Gehrels, G. E., P. G. DeCelles, T. P. Ojha, and B. N.Upreti (2006b), Geologic and U-Th-Pb geochrono-logic evidence for early Paleozoic tectonism in theKathmandu thrust sheet, central Nepal Himalaya,Geol. Soc. Am. Bull., 118(1 – 2), 185 – 198,doi:10.1130/B25753.1.

Gehrels, G. E., V. A. Valencia, and J. Ruiz (2008),Enhanced precision, accuracy, efficiency, and spa-tial resolution of U-Pb ages by laser ablation –multicollector – inductively coupled plasma–massspectrometry, Geochem. Geophys. Geosyst., 9,Q03017, doi:10.1029/2007GC001805.

Giovanoli, F. (1990), Horizontal transport and sedimen-tation by interflows and turbidity currents in LakeGeneva, in Large Lakes: Ecological Structure andFunction, edited by M. M. Tilzer and C. Serruya,pp. 175–195, Springer, Berlin.

Gupta, S., J. R. Underhill, I. R. Sharp, and R. L.Gawthorpe (1999), Role of fault interactions in con-trolling synrift sediment dispersal patterns: Miocene,Abu Alaqa Group, Suez Rift, Sinai, Egypt, BasinRes., 11(2), 167 – 189, doi:10.1046/j.1365-2117.1999.00300.x.

Hampton, B. A., and B. K. Horton (2007), Sheetflowfluvial processes in a rapidly subsiding basin, AltiplanoPlateau, Bolivia, Sedimentology, 54(5), 1121–1148,doi:10.1111/j.1365-3091.2007.00875.x.

Heinz, J., and T. Aigner (2003), Hierarchical dynamicstratigraphy in various Quaternary gravel deposits,Rhine glacier area (SW Germany): Implications forhydrostratigraphy, Int. J. Earth Sci., 92(6), 923–938,doi:10.1007/s00531-003-0359-2.

Hodges, K. V. (2000), Tectonics of the Himalaya andsouthern Tibet from two perspectives,Geol. Soc. Am.Bull., 112(3), 324 – 350, doi:10.1130/0016-7606(2000)112<0324:TOTHAS>2.3.CO;2.

Hodges, K. V., R. R. Parrish, T. B. Housh, D. R. Lux,B. C. Burchfiel, L. H. Royden, and Z. Chen (1992),Simultaneous Miocene extension and shorteningin the Himalayan orogen, Science, 258(5087),1466–1470, doi:10.1126/science.258.5087.1466.

Hodges, K. V., R. R. Parrish, and M. P. Searle (1996),Tectonic evolution of the central Annapurna Range,Nepalese Himalayas, Tectonics, 15(6), 1264–1291,doi:10.1029/96TC01791.

Hunter, R. E. (1977), Basic types of stratification insmall eolian dunes, Sedimentology, 24(3), 361–387,doi:10.1111/j.1365-3091.1977.tb00128.x.

Hurtado, J. M., Jr., K. V. Hodges, and K. X. Whipple(2001), Neotectonics of the Thakkhola graben andimplications for recent activity on the South Tibetanfault system in the central Nepal Himalaya, Geol.Soc. Am. Bull., 113(2), 222 – 240, doi:10.1130/0016-7606(2001)113<0222:NOTTGA>2.0.CO;2.

Ingersoll, R. V., T. F. Bullard, R. L. Ford, J. P. Grimm,J. D. Pickle, and S. W. Sares (1984), The effect ofgrain size on detrital modes: A test of the Gazzi-Dickinson point-counting method, J. Sediment.Petrol., 54(1), 103–116.

Jopling, A. V. (1965), Hydraulic factors controlling theshape of laminae in laboratory deltas, J. Sediment.Petrol., 35(4), 777–791.

Kapp, P., and J. H. Guynn (2004), Indian punch riftsTibet, Geology, 32(11), 993 – 996, doi:10.1130/G20689.1.

Li, J., and Y. Zhou (2001), Pliocene palynoflora from theZanda basin West Xizang (Tibet), and the palaeoen-vironment, Acta Micropalaeontol. Sin., 18, 89– 96.

Lourens, L., F. Hilgen, N. J. Shackleton, J. Laskar,and D. Wilson (2004), The Neogene period, in A

Geologic Time Scale 2004, edited by F. M. Gradstein,J. G. Ogg, and A. G. Smith, pp. 409–440, CambridgeUniv. Press, Cambridge, U. K.

Lowe, D. (1982), Sediment gravity flows. 2. Deposi-tional models with special reference to the depositsof high-density turbidity currents, J. Sediment. Petrol.,52(1), 279–298.

Ludwig, K. (2003), Isoplot 3.0, Spec. Publ. 4, 70 pp.Berkeley Geochronol. Cent., Berkeley, Calif.

Lunt, I. A., J. S. Bridge, and R. S. Tye (2004), Aquantitative, three-dimensional depositional modelof gravelly braided rivers, Sedimentology, 51(3),377–414, doi:10.1111/j.1365-3091.2004.00627.x.

Martin, A. J., P. G. DeCelles, G. E. Gehrels, P. J. Patchett,and C. Isachsen (2005), Isotopic and structural con-straints on the location of the Main Central Thrust inthe Annapurna Range, central Nepal Himalaya,Geol. Soc. Am. Bull. , 117(7 – 8), 926 – 944,doi:10.1130/B25646.1.

McCaffrey, R., and J. Nabelek (1998), Role of obliqueconvergence in the active deformation of the Hima-layas and southern Tibet Plateau, Geology, 26(8),691 – 694, doi:10.1130/0091-7613(1998)026<0691:ROOCIT>2.3.CO;2.

Meng, X. G., D. G. Zhu, Z. G. Shao, C. B. Yang, J. N.Han, J. Yu, Q. W. Meng, and R. P. Lu (2008), LateCenozoic stratigraphy and paleomagnetic chronol-ogy of the Zanda basin, Tibet, and records of theuplift of the Qinghai-Tibet Plateau, Acta Geol. Sin.(Engl. Ed.), 82(1), 63 –72.

Metcalfe, R. (1993), Pressure, temperature, and timeconstraints on metamorphism across the Main CentralThrust zone and High Himalayan slab in the GarhwalHimalaya, in Himalayan Tectonics, edited by P. J.Treloar and M. P. Searle, Geol. Soc. Spec. Publ.,74, 485–509.

Miall, A. D. (1978), Lithofacies types and vertical pro-file models in braided river deposits: A summary, inFluvial Sedimentology, edited by A. D. Miall, Mem.

Can. Soc. Pet. Geol., 5, 597–604.Miall, A. D. (2000), Principles of Sedimentary Basin

Analysis, 616 pp.Springer, Berlin.Molnar, P., and H. Lyon-Caen (1988), Some simple

physical aspects of the support, structure, and evolu-tion of mountain belts, in Processes in ContinentalLithospheric Deformation, edited by S. J. Clark Jr.,B. C. Burchfiel, and J. Suppe, Spec. Pap. Geol. Soc.Am., 218, 179–207.

Molnar, P., and P. Tapponnier (1978), Active tectonicsof Tibet, J. Geophys. Res., 83(B11), 5361–5375,doi:10.1029/JB083iB11p05361.

Murphy, M. A. (2007), Isotopic characteristics of theGurla Mandhata metamorphic core complex: Impli-cations for the architecture of the Himalayan orogen,Geology, 35(11), 983–986, doi:10.1130/G23774A.1.

Murphy, M. A., and W. P. Burgess (2006), Geometry,kinematics, and landscape characteristics of an activetranstension zone, Karakoram fault system, southwestTibet, J. Struct. Geol., 28, 268–283, doi:10.1016/j.jsg.2005.10.009.

Murphy, M. A., and P. Copeland (2005), Transtensionaldeformation in the central Himalaya and its role inaccommodating growth of the Himalayan orogen,Tectonics, 24, TC4012, doi:10.1029/2004TC001659.

Murphy, M. A., and T. M. Harrison (1999), Relation-ship between leucogranites and the Qomolangmadetachment in the Rongbuk Valley, South Tibet,Geology, 27(9), 831–834, doi:10.1130/0091-7613(1999)027<0831:RBLATQ>2.3.CO;2.

Murphy, M. A., and A. Yin (2003), Structural evolutionand sequence of thrusting in the Tethyan fold-thrustbelt and Indus-Yalu suture zone, southwest Tibet,Geol. Soc. Am. Bull., 115(1), 21– 34, doi:10.1130/0016-7606(2003)115<0021:SEASOT>2.0.CO;2.

Murphy, M. A., A. Yin, P. Kapp, T. M. Harrison, D. Lin,and G. Jinghui (2000), Southward propagation of theKarakoram fault system, southwest Tibet: Timingand magnitude of slip, Geology, 28(5), 451 –454,doi:10.1130/0091-7613(2000)28<451:SPOTKF>2.0.CO;2.

Murphy, M. A., A. Yin, P. Kapp, T. M. Harrison, C. E.Manning, F. J. Ryerson, D. Lin, and G. Jinghui(2002), Structural evolution of the Gurla Mandhatadetachment system, southwest Tibet: Implications forthe eastward extent of the Karakoram fault system,Geol. Soc. Am. Bull., 114(4), 428–447, doi:10.1130/0016-7606(2002)114<0428:SEOTGM>2.0.CO;2.

Mutti, E., R. Tinterri, G. Benevelli, D. di Biase, andG. Cavanna (2003), Deltaic, mixed and turbidite

sedimentation of ancient foreland basins, Mar. Pet.Geol., 20(6 –8), 733–755, doi:10.1016/j.marpetgeo.2003.09.001.

Nemec, W., and R. Steel (1984), Alluvial and coastalconglomerates: Their significant features and somecomments on gravelly mass-flow deposits, in Sedi-mentology of Gravels and Conglomerates, editedby E. H. Kloster and R. J. Steel, Mem. Can. Soc.Pet. Geol., 10, 1 –31.

Ni, J., and M. Barazangi (1985), Active tectonics of thewestern Tethyan Himalaya above the underthrust-ing Indian plate: The Upper Sutlej River basin as apull-apart structure, Tectonophysics, 112(1 – 4),277 –295, doi:10.1016/0040-1951(85)90183-0.

Ni, J., and J. E. York (1978), Late Cenozoic tectonics ofthe Tibetan Plateau, J. Geophys. Res., 83(B11),5377–5384, doi:10.1029/JB083iB11p05377.

Ojha, T. P., R. F. Butler, J. Quade, P. G. DeCelles,D. Richards, and B. N. Upreti (2000), Magneticpolarity stratigraphy of the Neogene Siwalik Groupat Khutia Khola, far western Nepal, Geol. Soc. Am.Bull., 112(3), 424 – 434, doi:10.1130/0016-7606(2000)112<0424:MPSOTN>2.3.CO;2.

Phillips, R. J., R. R. Parrish, and M. P. Searle (2004),Age constraints on ductile deformation and long-termslip rates along the Karakoram fault zone, Ladakh,Earth Planet. Sci. Lett., 226, 305–319, doi:10.1016/j.epsl.2004.07.037.

Pierson, T. C. (1980), Erosion and deposition by debrisflows at Mt. Thomas, North Canterbury, NewZealand, Earth Surf. Processes, 5(3), 227 –247.

Pierson, T. C. (1981), Dominant particle supportmechanisms in debris flows at Mt Thomas, NewZealand, and implications for flowmobility, Sedimen-tology, 28(1), 49 – 60, doi:10.1111/j.1365-3091.1981.tb01662.x.

Pivnik, D. A. (1990), Thrust-generated fan-deltadeposition: Little Muddy Creek Conglomerate, SWWyoming, J. Sediment. Petrol., 60(4), 489–503.

Postma, G. (1983), Water escape structures in the con-text of a depositional model of a mass flow domi-nated conglomeratic fan-delta (Abrioja Formation,Pliocene, Almeria basin, SE Spain), Sedimentology,30(1), 91 – 103, doi:10.1111/j.1365-3091.1983.tb00652.x.

Quade, J., J. Cater, T. Ojha, J. Adam, and T. Harrison(1995), Late Miocene environmental change and inNepal and the northern Indian subcontinent: Stableisotopic evidence from paleosols, Geol. Soc. Am.Bull., 107(12), 1381–1397, doi:10.1130/0016-7606(1995)107<1381:LMECIN>2.3.CO;2.

Ratschbacher, L., W. Frisch, G. Liu, and C. Chen(1994), Distributed deformation in southern andwestern Tibet during and after the India-Asian colli-sion, J. Geophys. Res., 99(B10), 19,917 –19,945,doi:10.1029/94JB00932.

Robinson, D. M., P. G. DeCelles, C. N. Garzione, O. N.Pearson, T. M. Harrison, and E. J. Catlos (2003),Kinematic model for the Main Central Thrust inNepal, Geology, 31(4), 359 – 362, doi:10.1130/0091-7613(2003)031<0359:KMFTMC>2.0.CO;2.

Robinson, D. M., P. G. DeCelles, and P. Copeland(2006), Tectonic evolution of the Himalayan thrustbelt in western Nepal: Implications for channel flowmodels, Geol. Soc. Am. Bull., 118(7 – 8), 865–885,doi:10.1130/B25911.1.

Saez, A., P. Anadon, M. J. Herrero, and A. Moscariello(2007), Variable style of transition between Palaeo-gene fluvial fan and lacustrine systems, southern Pyr-enean foreland, NE Spain, Sedimentology, 54(2),367–390, doi:10.1111/j.1365-3091.2006.00840.x.

Saylor, J. E. (2008), The late Miocene through modernevolution of the Zhada basin, south-western Tibet,Ph.D. thesis, 306 pp.Univ. of Ariz., Tucson.

Saylor, J. E., J. Quade, D. L. Dellman, P. G. DeCelles,P. A. Kapp, and L. Ding (2009), The late Miocenethrough present paleoelevation history of southwes-tern Tibet, Am. J. Sci., 309(1), 1 –42, doi:10.2475/01.2009.01.

Searle, M. P., R. R. Parrish, K. V. Hodges, A. Hurford,M. W. Ayres, and M. J. Whitehouse (1997), ShishaPangma leucogranite, South Tibetan Himalaya:


23 of 24

TC1004

Field relations, geochemistry, age, origin, andemplacement, J. Geol., 105(3), 295 – 318,doi:10.1086/515924.

Searle, M. P., R. L. Simpson, R. D. Law, R. R. Parrish,and D. J. Waters (2003), The structural geometry,metamorphic and magmatic evolution of the Everestmassif, High Himalaya of Nepal –South Tibet,J. Geol. Soc. London, 160, 345–366, doi:10.1144/0016-764902-126.

Seeber, L., and A. Pecher (1998), Strain partitioningalong the Himalayan arc and the Nanga Parbat anti-form, Geology, 26(9), 791 –794, doi:10.1130/0091-7613(1998)026<0791:SPATHA>2.3.CO;2.

Sharp, I. R., R. L. Gawthorpe, J. R. Underhill, andS. Gupta (2000), Fault-propagation folding in exten-sional settings: Examples of structural style and syn-rift sedimentary response from the Suez rift, Sinai,Egypt, Geol. Soc. Am. Bull., 112(12), 1877–1899,doi:10.1130/0016-7606(2000)112<1877:FPFIES>2.0.CO;2.

Stacey, J. S., and J. D. Kramers (1975), Approximationof terrestrial lead isotope evolution by a two-stagemodel, Earth Planet. Sci. Lett., 26, 207 – 221,doi:10.1016/0012-821X(75)90088-6.

Talbot, M., and P. Allen (1996), Lakes, in Sedimentary

Environments: Processes, Facies and Stratigraphy,edited by H. G. Reading, pp. 83–124, BlackwellSci., Oxford, U. K.

Thiede, R. C., J. R. Arrowsmith, B. Bookhagen, M. O.McWilliams, E. R. Sobel, and M. R. Strecker(2005), From tectonically to erosionally controlleddevelopment of the Himalayan orogen, Geology,33(8), 689 –692.

Thiede, R. C., J. R. Arrowsmith, B. Bookhagen,M. McWilliams, E. R. Sobel, and M. R. Strecker(2006), Dome formation and extension in the TethyanHimalaya, Leo Pargil, northwest India, Geol. Soc.Am. Bull., 118(5 – 6), 635 – 650, doi:10.1130/B25872.1.

Valli, F., N. Arnaud, P. H. Leloup, E. R. Sobel, G.Maheo,R. Lacassin, S. Guillot, H. Li, P. Tapponnier, andZ. Xu (2007), Twenty million years of continuousdeformation along the Karakorum fault, westernTibet: A thermochronological analysis, Tectonics,26, TC4004, doi:10.1029/2005TC001913.

Vanhinte, J. E. (1978), Geo-history analysis: Applicationof micro-paleontology in exploration geology,AAPG Bull., 62(2), 201–222.

Vannay, J.-C., B. Grasemann, M. Rahn, W. Frank,A. Carter, V. Baudraz, and M. Cosca (2004),Miocene to Holocene exhumation of metamorphiccrustal wedges in the NW Himalaya: Evidence fortectonic extrusion coupled to fluvial erosion, Tec-tonics, 23, TC1014, doi:10.1029/2002TC001429.

Wang, S. F., P. Blisniuk, O. Kempf, X.M. Fang, F. Chun,and E. Wang (2008a), The basin-range systemalong the south segment of the Karakorum fault zone,Tibet, Int. Geol. Rev., 50, 121 –134, doi:10.2747/0020-6814.50.2.121.

Wang, S. F., W. L. Zhang, X. M. Fang, S. Dai, andO. Kempf (2008b), Magnetostratigraphy of theZanda basin in southwest Tibet Plateau and itstectonic implications, Chin. Sci. Bull., 53(9),1393–1400, doi:10.1007/s11434-008-0132-9.

Wang, W., J. Zhang, and B. Zhang (2004), Structuraland sedimentary features in Zanda basin of Tibet,Acta Sci. Nat., 40, 872–878.

Wang, Y., T. Deng, and D. Biasatti (2006), Ancientdiets indicate significant uplift of southern Tibetafter ca. 7 Ma, Geology, 34(4), 309 – 312,doi:10.1130/G22254.1.

Yin, A. (2006), Cenozoic tectonic evolution of theHimalayan orogen as constrained by along-strikevariation of structural geometry, exhumation history,and foreland sedimentation, Earth Sci. Rev., 76(1–2),1–131, doi:10.1016/j.earscirev.2005.05.004.

Yin, A., T. Harrison, M. Murphy, M. Grove, S. Nie,F. Ryerson, W. Feng, and C. Le (1999), Tertiarydeformation history of southeastern and southwes-tern Tibet during the Indo-Asian collision,Geol. Soc.Am. Bull., 111(11), 1644–1664, doi:10.1130/0016-7606(1999)111<1644:TDHOSA>2.3.CO;2.

Zhang, J., L. Ding, D. Zhong, and Y. Zhou (2000),Orogen-parallel extension in Himalaya: Is it theindicator of collapse or the product in process ofcompressive uplift?, Chin. Sci. Bull., 45(2), 114–120,doi:10.1007/BF02884653.

Zhang, Q., F. Wang, H. Ji, andW. Huang (1981), Pliocenesediments of the Zanda basin, Tibet, J. Stratigr., 5,216–220.

Zhou, Y., L. Ding, W. Deng, and J. Zhang (2000),Tectonic cyclothems in the Zanda basin and its sig-nificance, Sci. Geol. Sin., 35, 305–315.

Zhu, D., et al. (2004), Features of Pliocene – lowerPleistocene sedimentary facies and tectonic evolutionin the Zanda basin, Ngari area, Tibet, J. Geomech.,110, 245–252.

Zhu, D., X. Meng, Z. Shoa, C. Yang, J. Han, J. Yu,Q. Meng, and R. Lu (2007), Evolution of the paleo-vegetation, paleoenvironment and paleoclimate dur-ing Pliocene – early Pleistocene in Zhada basin, Ali,Tibet, Dizhi Xuebao, 81, 295–306.

��P. DeCelles, G. Gehrels, and P. Kapp, Department

of Geosciences, University of Arizona, Tucson, AZ85721, USA.

M. Murphy and R. Zhang, Department of Earth andAtmospheric Sciences, University of Houston, Houston,TX 77204, USA.

J. Saylor, Department of Geological Sciences, JacksonSchool of Geosciences, University of Texas at Austin,Austin, TX 78712, USA. ([email protected])


24 of 24

TC1004

Documents

Basin formation in the High Himalaya by arc-parallel extension …easd.geosc.uh.edu/murphy/pdf/Saylor et al. 2010.pdf · 2012-02-21 · Zhada basin are the South Tibetan Detachment