TECTONICS, VOL. 20, NO. 6, PAGES 799-809, DECEMBER 2001
Southward extrusion of Tibetan crust and its effect on Himalayan
tectonics
K.V. Hodges,J.M. Hurtado,andK.X. Whipple
Dcpartmcntof Earth,Atmosphcric,
andPlanetarySciences,
Massachusctts
Instituteof TechnologyCambridge,
Massachusetts
Abstract.
USA
The
Tibetan
Plateau
is a storehouse
of
excess
gravitational potential energy accumulatedthrough crustal
thickening during India-Asia collision, and the contrast in
potential energy between the Plateau and its surroundings
strongly influencesthe modern tectonicsof south As•a. The
distribution of potential energy anomalies across the region,
derivedfrom geopotentialmodels,indicatesthat the H•malayan
front is the optimal locationfor focuseddissipationof excess
energystoredin the Plateau.The modempatternof deformation
and erosionin the Himalayaprovidesan efficientmechanismfor
such dissipation, and a review of the Neogene geological
evolution of southern Tibet and the Himalaya shows that this
mechanismhas been operationalfor at least the past 20 million
years. This persistenceof deformational and erosional style
suggeststo us that orogens,like other complex systems,can
evolve toward "steady state" configurationsmaintainedby the
continuousflow of energy.The capacityof orogenicsystemsto
self-organizeinto temporallypersistentstructuraland erosional
patternssuggeststhat the tectonichistoryof a mountainrange
may dependon local energeticsas much as it doeson far-field
plate interactions.
favors the localizationof high-efficiencydissipationprocesses
alongthe Himalayanorogenicfront. A closerlook at the geology
and physiographyof this margin of the Tibetan Plateaureveals
that large amountsof energyare dissipatedthroughcoordinated
deformational and erosional activity that accommodate
southward
extrusion of the middle crust from beneath the Tibetan
Plateau. These processesare not just modern phenomena;
evidence that they have been active in the Himalaya for at least
the past 20 million yearssuggeststhat the southernmargin of the
orogenicsystemmay have achieveda dynamical steady state
reflecting a rough balancebetween the accumulationof energy
and its dissipation.
2. Potential Energy Gradients Around the
Tibetan
Plateau
Of all the energyaccumulativeprocessesthat operateduring
orogeny the increaseof gravitational potential energy in the
systemdue to crustalthickeningis among the most important.
The gravitationalpotentialenergy(U) per unit area of a column
of isostatically compensatedlithosphere is equivalent to the
integral of the body force over the vertical distancebetweenthe
surfaceand the compensation
horizon:
1. Introduction
Fromtheperspective
of thermodynamics
the development
and
evolutionof orogenicsystemsinvolvesthree kinds of processes:
those that serve to accumulateenergy, those that transfer energy
from one part of the system to another, and those that help
dissipateexcessenergy.Crustalthickening and other forms of
energy accumulation in orogenic settings are predominantly
causedby lithosphericplate interactions;as a consequence,the
rates at which they occur and the geometriesof the structures
they produceare at leastbroadly relatedto platemotions.On the
otherhand,processes
responsiblefor the transferalor dissipation
of energyare more dependenton the abundanceand distribution
of matter within an orogenicsystem,making their spatial and
temporal evolution impossibleto predict from plate tectonic
principlesalone.If we aspireto a comprehensive
understanding
of orogenesis,
we must look more closelyat the distributionof
energyin orogenicsystemsbecause,at leastin theory,gradients
in that distributionshoulddrive the processesof energytransfer
anddissipation.
In this paper, we show that the distributionof gravitational
potential energy in the Himalayan-Tibetan orogenic system
Copyright2001 by thc AmericanGeophysical
Union.
Papernumber2001TC001281.
0278-7407/01/2001TC001281
$12.00
799
(•)
z=0
where the spatial coordinate (z) ranges from zero at the
compensation
horizonto S at the Earth'ssurface,p(z) is density
as a functionof depththroughthe column,andg is gravitational
acceleration[Molnar and Lyon-Caen, 1988]. Since the variation
in U from one columnof lithosphereto anotheris a measureof
the contributionmade by buoyancyforcesto the deformationof
the lithosphere,it is convenientto comparethe gravitational
potentialenergyfor a specificlithospheric
columnof interest(U,)
to that of a referencepotentialenergy,for example,the mean
value for the Earth's lithosphere(Ur) [Cobblentzet al., 1994],
throughthepotentialenergyanomaly(Eu):
œc=
(2)
The calculationof Eu requiressome means of estimating
density structuresfor both the column of interest and the
referencecolumn. Perhapsthe most effective approach is to
invert seismicvelocitydatathroughempiricalrelationships[e.g.,
Joneset al., 1996],but suchdataareunavailablefor mostof the
HimalayaandTibet.We insteadfollowtheapproach
of Coblentz
et al. [1994], who notedthat E• is approximately
relatedto the
morefamiliargeoidanomaly(Es) throughtherelationship:
800
HODGES ET AL.: SOUTHWARD
where G is the gravitationalconstant.Because we are interested
only in variationsin the potentialenergy field that are related to
lithosphericdensityand thicknessvariations,we have filtered out
long-wavelengthcontributionsto the geoid height, which might
be related to sublithospheric
heterogeneities,
by removingterms
belowdegreeandorderseven(cosine-tapered
to degreeandorder
11) from the geoid prior to applyingequation(3), as advocated
by Joneset al. [1996].
Plate la is a map of Eu for the Tibetan Plateauregion derived
in this way from the NASA-National Mapping and Imagery
Agencyjoint geopotentialmodel EGM96 [Lemoineet al., 1998].
In general, positive values of Eu imply that deviatoric stresses
arising from horizontal variations in the lithospheric density
structureare tensile,whereasnegativevaluesimply that they are
contractile. The Himalayan-Tibetan orogenic system is well
defined in Plate l a as a band of high gravitational potential
separatingthe low-Eu regionsof the Indian subcontinenton the
south and the Tarim and Qiadam Basins on the north. As has
been noted elsewhere [e.g., Molnar and Tapponnier, 1978'
Molnar, 1988; England and Houseman, 1989], Eu contrasts
betweenTibet and its surroundings
providestrongmotivationfor
lateral redistributionof the Tibetan crust in order to dissipate
excess gravitational potential energy. Indeed, the nature of
EXTRUSION
OF TIBETAN
CRUST
wherev is the flow velocity,r is a transportcoefficient,and•SEv
/6x (or AEu, for simplicity)is the spatialgradientin Eu. The
nature of r reflects the mechanismassumedfor crustal flow, and
it is likely to be dependent
on factorsthatvary in time andspace.
For the specialcaseof channelizedflow of a viscouslower crust
underthe influenceof crustalthicknessvariations[Kruse et al.,
1991;Clark and Royden,2000;McQuarrieand Chase,2000],
tc •
,
(6)
seismicactivityin Tibet over the 1977-1998intervalimpliesthat
the plateauregion is actively losingpotentialenergy[Tanimoto
and Okamoto,2000].
The distribution of Eu across the plateau provides some
insightsregardingthe probabledirection,magnitude,and rate of
crustalflow relatedto spatiallyvariablebuoyancyforces.(In this
paper, we use the term "crustal flow" to refer to the spatial
transfer of crustal material without regard to mechanism. For
example, the viscous transport of lower crustal material [Bird,
1991; Kruse et al., 1991] and the fluvial transportof eroded
where D is the channel thickness and g is the viscosity of
material flowing in the channel. (This relationship involves a
proportionality rather than an equality because A Eu is
proportional to, but not equivalent to, the lateral pressure
gradientsinvokedby Kruse et al. [1991] and subsequent
authors
as the impetus for channelized flow.) For the special case of
erosion and fluvial transport of eroded sedimentsout of the
orogenic system, r is strongly dependenton factors such as
lithologic resistance to erosion, erosional mechanism, stream
network geometry, climate, and the exact relationshipbetween
stream gradient and AEu [Howard et al., 1994; Tucker and
Slingerland,1997; Sklar and Dietrich, 1998; Whipple and
Tucker, 1999]. In reality, these and many other processes
combineto accommodateenergy transfer in orogenicsystems,
such that r is very complicated, but equation 5 nevertheless
capturesthe essentialphysicsof how potentialenergygradients
influencematerial fluxes in orogenic systems.It clearly doesnot
provide a comprehensivedepictionof material fluxes, however,
becauseexternally imposed, "far-field" tectonic forces, basal
tractions,andthe like alsoplay importantroles.
The negative sign is equation (5) implies that crustal flow
proceedsdown gradientsin gravitational potential energy. In
sediments
Platela, valuesof Eu are highest(>_5.9
x 10•2 N/m) in the
are both mechanisms
of flow
in the context
of our
working definition.) Imagine, for a moment, that the orogenic
systemis not subjectto lateral forcesrelatedto lithosphericplate
interactions.In that case, Plate l a is a map of a potential field
throughwhich gradientsin Eu drive the flow of material. This
situation is analogous to the influence of gradients in
temperature, hydraulic pressure,or chemical concentration on
heat flux, fluid transport, or chemical diffusion. Structural
homologies are notorious among equations governing such
transportprocesses.For example, Fourier's law for heat flow,
Darcy's law for groundwaterflow, and Fick's law for chemical
diffusion all have the same basic form in one dimension:
q,
o•x,
(4)
whereq is a flux per unitcross-sectional
area,•Sb/6x
is thechange
in potential over some spatial dimension x, and K is a
proportionalityterm [Carslaw and Jaeger, 1986; Furbish, 1997;
Crank,1975].For manyproblemsin diffusiveflow, K is assumed
to be constant,
but thisneednotbe thecasein general.
By analogywith thesemore familiar examplesa generallaw
for crustalflow drivenby gravitational
potentialenergygradients
is
Karakoram Range and along the Himalayan arc. North of the
Himalaya, on the physiographic Tibetan Plateau, Eu drops
smoothly eastward from the Karakoram to-95øE longitude
before rising again slightly toward the Longmenshanmountains,
immediately west of the Sichuan Basin. This "saddle" in the Eu
field is relatedto a region of relatively low gravitationalpotential
energy extending southward from the Qiadam Basin. High
positiveEu acrossthe plateauand the broadnegativegradientin
Ev from west to east is consistentwith geologicalevidenceof
NeogeneE-W surfaceextension[Molnar and Tapponnier,1978;
Armijo et al., 1986] and interpretivemodelsof eastwardlowercrustalflow [Roydenet al., 1997; Clark and Royden,2000].
Equation (5) also implies that the rate of crustal flow is
proportionalto spatial gradientsin Eu [see also England and
Molnar, 1997]. For the Himalayan-Tibetan system we can
evaluate,in a generalway, where the mostefficientdissipationof
excessgravitationalpotential energy might occurby studyinga
map of the spatial derivative of the potential energy anomaly
(AEu) over the region (Plate lb). This exerciseshowsthat the
steepest
gradients
(upto-3.3 x 107N/m2)areconcentrated
along
the southern plateau boundary, between -80øE and 95øE
longitudes.A few discontinuouszones of high AEu also exist
along the westernKunlun mountain front. Slightly lessdramatic
IIODGESET AL.' SOUTHWARDEXTRUSIONOF TIBETAN CRUST
801
50 ø 15'N
Potential
Energy
Anomaly (• u)
TA
Q
T
Fig. 2
Fig. I
-'
, I
>5.8(
!12
18 ø 15'N
be
Spatial
Gradient
in •u (A•u)
WK
EK
ws
ES
MOHGOLIA
TA
TU
0.0
> 1.0 (x 107
m 2)
k
PEOPLœS REPUBLIC
OF CHINA
AKISTAN
N
BH
IND•A
Plate 1. (a). Distributionof the gravitationalpotential energy anomaly(œu) in the Himalayan-Tibetansystemand
surroundingregionsof southAsia. Areasof specialinterestincludenorthernIndia (I), the KarakoramRange(K),
the central Tibetan Plateau(T), and the Sichuan(S), Tarim (TA), and Qiadam (Q) Basins.The rectangularbox
showsthe areacoveredby Figure2. Within that box the obliqueline indicatesan approximateprofile for whichthe
generalizedcrosssectionin Figure 1 is appropriate.(b) A map of the samearea as Plate la, showingthe spatial
derivative of the potentialenergyanomaly (Age). The highestgradientsoccur along the range front betweenthe
western(WS) and eastern(ES) syntaxesof the Himalaya. Gradientsare nearly as high along the westernKunlun
Mountains(WK) but are more subduedalong the easternKunlun Mountains(EK) and the Longmenshan(L). Lowgradientgapsto the north and southof the Longmenshan,designatedby blue arrows, have been postulatedas
pathwaysfor the flow of lower crustoutward from beneaththe centralplateauregion [Clark and Royden,2000].
Inset showsthe political geographyof the region of Plates la and lb: A, Afghanistan;BA, Bangladesh;BH,
Bhutan; K, Kyrgyzstan; L, Laos; N, Nepal; R, Russia; TA, Tajikistan; TU, Turkmenistan; TH, Thailand; U,
Uzbekistan;and V, Vietnam. Platesla and lb are in unprojectedgeographiccoordinatesand the insetis an Albers
conical projection.
802
HODGES ET AL.: SOUTHWARD EXTRUSION OF TIBETAN CRUST
gradients
(2 to 3 x 10• N/m•) characterize
theeastern
Kunlunand
the orogenicsystemthrough sedimentaryprocesses,is most
effectivealongthe marginsreceivingthe heaviestprecipitation.
Thus we might expect the Himalayan margin, which receives
extremelyhigh monsoonalprecipitation,to supporta higherrate
of El/dissipationthanthe westernKunlun marginwhichreceives
far less precipitation, even though the two margins have
comparableAœt/insomeareas.Howeverit alsomightbe argued
that the high-El/ gradient along the Himalayan margin is
dynamicallysupportedby India-Eurasiaconvergence,and that
this convergence effectively counteracts the horizontal
componentsof buoyancyforcesthat might drive the southward
Longmenshanfronts.
It is unlikelythatall high-Aœt/margins
in Plate lb are zonesof
very rapid energy dissipationby crustalflow for three reasons:
(1) the lithosphere is not mechanically isotropic, (2) the
mechanismsavailablefor transferof matterout of the system
vary from placeto placealongthe margins,and (3) the systemis
subjectto anisotropicforcesrelatedto plateconvergence.
Clark
and Royden[.2000]have describedhow rheologicalvariability
mightbe importantif, for example,the low-El/regionadjacentto
a high-Aœt/margin is strongenoughto act as a barrier to the
lateralflow of lower crust.In particular,they suggested
that the
Longmenshanfront may have developedbecausethe crustwas
anomalously
strongin the regionof the SichuanBasin,diverting
eastwardlower crustalflow throughweakerchannelsto the north
andsouth(Plate lb)
It might be arguedthat one of the mostimportantagentsof
energydissipation,the erosionand transportof materialout of
flow of material from the Tibetan Plateau toward India. In order
to evaluate the actual significance of dissipation along a
particularhigh-Aϥ;margin it is important to investigatethe
dissipativecapacityof deformationalanderos•onalprocesses
that
occur now along those margins and to ask whether or not the
geologicalrecordsuggests
a historyof efficientmaterialtransport
out of the system.In the next sectionwe review geomorphologic
Foreland
.-.
Zon
•
E 20- •;
.c:
•.
•
Zone HST
........LesserHimalayan
Zone
HST
................
•:•..........
•..
MOHO
b.
Little
Uplift,
Moderate
Uplift,High
Uplift, Little
Uplift,
Erm•on
Erosion
•
Erosio•
......
• •.-•.'•: • •;"'::•;;.::•/'-•••'½:.• •?.' .......' ' '"•'•"••'••
'•••-'•
'•• '•- -•-••••:•
.....
..::•?•
g::;:....•
.g•?:?f•;•
.•.•-v..:.,..
•.•
;;3.
•ler•ce.
-• :..•.• f::• .•
•..:•:•::•.x:g:-..•
;.• , e•-
Figure 1. (a) Generalizedcross section of the Himalayan margin of the Tibetan Plateau showingprincipal
tectonostratigraphic
zonesseparatedby the SouthTibetan fault (STF), Main Central thrust(MCT), Main Boundary
thrust(MBT), and Main Frontalthrust(MFT) systems.The subsurfacestructuralconfigurationshownhere, slightly
modifiedfromPandeyet al. [1995] and Cattinand Avouac[2000],presumes.
that all majorthrustsystems
in the
Himalaya root into a commonbasald•collement:the Himalayan Sole thrust(HST). The dashedline designatesa
now-inactivethrustfault that lies beneathklippen of metamorphicrockswithin the LesserHimalayanZone; it has
been interpretedalternativelyas an erosionalremnantof the MCT systemor as a separatethrustsystem[Hodges,
2000]. Vertical lines indicatethe principalphysiographictransitionsof the Himalaya (PTi-PT3) as discussedin the
text. The approximatepositionsof the axial planesof a major anticlinorium(A)-synclinorium(S)pair in the Lesser
Himalaya are labeled as well. (b) Interpretive model of the extrusion-erosionmechanism for excess energy
dissipationalongthe southernTibetanPlateaumargin.A channelof middle to lower crustalmaterial(designatedby
the random-dashpattern) extrudessouthwardfrom the central Tibetan Plateaubetweenthe STF and HST. Relief
and geodetically determineduplift rates are relatively high in the regions between the surface traces of these
structures,implying high erosion rates compared to the Gangeric Foreland and Tibetan Plateau. Much of the
shorteningbetweenIndia andTibet duringthe Holoceneepochhasbeenaccommodated
by slip on the MFT system
[Law3and Avouac,2000], and it may be that all of the regionbetweenPTi and PT3 representsthe free boundaryof
the extrudingwedgeover that timescale.However,the highestratesof erosionand uplift todayoccurnorthof the
MCT trace, which may indicateinsteadthat extrusionis presentlyconcentratedin a channelthat surfacesbetween
STF and MCT traces(the unshadedzone markedwith the random-dashpattern),and that relatively little extrusion
occursto the southtoday.
HODGES ET AL.' SOUTHWARD
andgeologicalevidencethat the activetectonicsof the Himalaya
are stronglyinfluencedby the southwardflow of Tibetan middle
and lower crustto the Himalayantopographicfront,whereit is
removed from the system by orographically focused erosion.
Theseprocessesact togetheras an efficient enginefor energy
dissipationalong the southernmargin of the Tibetan Plateau,
even as continuedIndia-Eurasiaconvergenceaddsenergyto the
systemas a whole.
EXTRUSION
OF TIBETAN
CRUST
803
3. Physiography and Neotectonicsof the
Himalaya
The southernmargin of the Tibetan Plateauis markedby a
series of sharp physiographictransitions (Figure l a). The
northernmost
of these(PTi) separates
the rugged,high peaksof
the mainHimalayanrangesandthe moresubduedtopography
of
the Tibetan Plateau.A second(PT2) occurssouthof the range
crest and representsthe main topographicfront of the central
Himalaya,wherethe southernslopesof the highpeaksgive way
to a foothills region with much lower relief and mean elevation.
A lesssignificantfront (PT3) separates
the foothillsregionfrom
33'N
the GangeticPlains.
PhysiographictransitionsPTi-PT3 are related to neotectonic
activityat or belowthe surface.The PT3 frontcoincidesroughly
w•th the trace of the north dipping Main Frontal thrust(MFT)
31'
system (Figure l a), a feature which was first established in
Figure 2. (a) Simplifiedtectonicmap (Albersconicalprojection)
of the centralregionof southernTibet and the Himalaya,after
Hodges [2000]. Half-arrows and ball-and-bar symbolsadorn
Quaternarytranscurrentand normal faults (respectively)in
southernTibet. The double-tick symbols indicate the dip
29ø•
I•! Tibetan Realm
27'
•
Greater HimalayanZone
•
LesserHimalayan
Zone
I• Subhimalayan
Zone
80'
direction
100km
I• Gangetic Foreland
82'
84'
86'
of the basal detachment
of the South Tibetan
fault
(STF) system,whereasthe barbsindicatethe dip directionsof
reverse faults of the Main Central (MCT), Main Boundary
88'E
(MBT), and Main Frontal (MFT) thrust systems. The MCT
33'N.i.•
.•...,::,..:
:.•.
•:,.w.:.:.
•.:X.............................................................................................
system is geometricallycomplex and includesnumeroussplays
with different movement histories; only the roof fault of the
system is shown here for the sake of clarity. Lithologic
,: ..... ...
descriptions of units in the Greater Himalayan, Lesser
Himalayan, and Subhimalayan Zones are given by Hodges
".... '* '"*'"'s "- :"•"
-.
[2000]. The hangingwall of the STF systemcorrespondsto the
.;;.....•,..,.,.. •'.
.......
..
.:
Tibetan Zone in that paper, but the field shown as "Tibetan
Realm" in Figure 2a also includes other tectonostratigraphic
packages,suchas the Indus-Tsangpoand TranshimalayanZones
of Hodges [2000]. Landmarksincludethe AnnapumaRange(A),
the Dangardzongfault (DF), the Dhaulagiri Range (D), Mount
Manaslu (M), the Thakkholagraben(TG), and the Yadong-Gulu
2,' ::
......... :..,.,:-.--,
' "•:.,{*.:''
...-;•;
':;::/v:.'.-----.rift (YG; just eastof the mapmarginat •89ø-900E longitude).(b)
0
Slope map of the same area derived from GTOPO 30 digital
elevation data. White-shaded regions indicate the steepest
80'
82'
84'
86'
88'E
33' • • :.•.:,•g••'•2• ............................. '....'........................topography.Note the regionof poor sloperesolution(designated
with "P") near the southeasterncorner due to poor GTOPO 30
data coverage.The highest relief in the Himalaya occurs in a
zoneboundby abruptphysiographic
transitionsreferredto as PT•
and PT2in the text. (c) Equivalentto Figure 2b with fault systems
from Figure 2a superimposed.The trace of the STF system
(correspondingwith PT•) separatesthe east-west extensional
..
.,
:
'..**.-
....
.
.
.
.
...
....
....
. ............................
... ...........
...
......
strain field of southern Tibet
29 o :
27'
•
•
•.. •
Nonreal and Transcurrent
Faults of Tibet
'
South
Tibetan
Fault
System
MaorThrust
S terns
oftheHimala
a
......
..................................................
80'
82'
84'
86'
88'E
from the north-south
contractional
strain field of the Himalayan realm. Divergence of the southern
margin of the high-relief zone (PT2) and the MCT trace often
occursin regionswhere the MCT is poorly mapped (designated
here with questionmarks). However, there are someareaswhere
the divergenceis well documented;note, for example,the linear
segmentof PT2 (markedwith an L in Figure 2c) that lies tensof
kilometers south of a segmentof the MCT zone mapped by
St6cklin et al. [1980], Macfarlane et al. [1992], and Upreti and
Le Fort [1999].
804
HODGES ET AL.: SOUTHWARD
Pliocene time but has experiencedepisodesof slip in the more
recentpast [Nakata, 1989; Yeatset al., 1992; Lav• and Avouac,
2000]. Marking the southern outcrop boundary of NeogeneQuaternary foreland basin sedimentary rocks of the
"SubhimalayanZone" (Figure 2a), the MFT system projects
northward into a shallowlydippingthrustsurfacethat marksthe
principalhorizonalongwhichthe Indianplateis subducted
[e.g.,
Seebetet al., 1981; Yeatsand Lillie, 1991]. This Himalayan Sole
thrust (HST; Figure l a) is typically drawn as extending
northward,with a dip of no more thana few degrees,beyondthe
surfacetrace of the Main Boundarythrust(MBT) system[e.g.,
Baranowskiet al., 1984;Ni and Barazangi, 1984]. Developingin
late Miocene-Pliocene time, the MBT system separates the
Subhimalayan Zone from overlying greenschist- to lower
amphibolite-facies
metamorphic
rocksof the "LesserHimalayan
Zone" (Figure 2a). In the central Himalaya of Nepal and
adjacentregionsof India andBhutanthe internalstructureof the
LesserHimalayan zone is characterizedby a long-wavelength
synclinorium-anticlinoriumpair (Figure l a). Stratal dips in
Lesser Himalayan rocks decreaseprogressivelyfrom southto
north, toward the core of the synclinorium,suggestingthat the
MBT systemrootssmoothlyinto the HST as shownin Figure la
[Gansser,1964;Le Fort, 1975;SchellingandArita, 1991].
In the eastern and central Himalaya the complementary
anticlinorium to the north has been interpretedas a fault bend
fold structuredeveloped above a ramp in the HST with a
structuralrelief of more than 5 km [Brunel, 1986; Schellingand
Arita, 1991; Schelling, 1992]. Although a somewhat more
complexgeometryhasbeenproposedfor the Himalayawestof
central Nepal [Srivastava and Mitra, 1994; DeCelles et al.,
1998], generalizedcrosssectionsdrawn acrossthe Himalayan
front almostalwaysshowsomeform of "crustal-scale"ramp in
the HST [e.g.,Gansser,1964;Le Fort, 1975;Haucket al., 1998;
Harrison et al., 1998]. While sucha featureis not requiredby the
surface geology, it nevertheless provides a convenient
explanationfor several geophysicalobservations,including
patternsof seismicitywithinthe orogenicwedge[Molnar, 1990;
Pandey et al., 1995, 1999], the depth to seismic reflectors
interpreted
astheHST beneathsouthern
Tibet [Zhaoet al., 1993;
Hauck et al., 1998], and evidence that much of the estimated
India-Tibet convergencerate has been accommodatedon the
MFT systemduringthe Holocene[Lav• andAvouac,2000].
Becausethepositionof the postulatedramp is generallydrawn
below PT2,manyauthorshaveattributedthe topographic
frontto
block uplift of the Greater Himalayan zone due to strain
accumulationover the ramp [Jacksonet al., 1992;Jacksonand
Bilham, 1994; Bilham et al., 1997; Biirgmann et al., 1999;
Jouanneet al., 1999; Larson et al., 1999; Pandey et al., 1995;
Cattin and Avouac,2000]. However,the mountainfront alsolies
near and sometimes coincident with the 1- to 5-km-thick
shear
zone that marks the trace of the MCT system, a third important
structurethat dividesthe LesserHimalayan rocks from middle to
upper amphibolite-faciesmetamorphicrocks of the Greater
Himalayanzone(Figuresla and 2a). Althoughslip on the MCT
commencedat 20-23 Ma (see review by Hodges [2000]), the
structurehasbeenactiveepisodicallysincethat time [Macfarlane
et al., 1992; Harrison et al., 1997], and some researchers[e.g.,
Seebet and Gornitz, 1983] have attributed PT2 to active
deformationon the system.A somewhatmodified interpretation
is that a newly emergent,out-of-sequencethrust system,partly
EXTRUSION OF TIBETAN CRUST
reactivatingstrandsof the MCT system,partly breakingalong
new fault traces,might be responsiblefor PT2. Regardlessof the
exact relationship between this physiographic transition and
specific structures, there seems little doubt that it is a
manifestationof recentfaultingin the Himalayanrealmand is not
a passive erosional front, as has been suggestedelsewhere
[Maseket al., 1994].
The topographic
breakat PT• marksa boundarybetweentwo
regionsof the lithosphere
thatareresponding
in distinctive
ways
to continental collision. South of PT•, deformation over the past
23 million yearshasproducedcontractional
structures
suchasthe
MCT, MBT, and MFT systems.Seismologicand geodeticstudies
showthat the currentrate of shorteningacrossthe rangeis 17-18
mm/yr [Bilhamet al., 1997].Farthernorth,on the plateauitself,
the modem strain field is much different. Landsatimagery, field
maps,andseismicdatasuggest
that activedeformation
in Tibet
includesroughlyE-W extensionon numerousnorthstrikingrift
systems,right-lateraldisplacements
on NW strikingtranscurrent
faults,and left-lateraldisplacements
on NE strikingtranscurrent
faults [Molnar and Tapponier, 1975; Armijo et al., 1986].
Collectively, these fault systems are compatible with an
extensional
regimein the southern
plateauandeastwardextrusion
of Tibetan crust as India convergeswith the rest of Eurasia
[Tapponierand Molnar, 1976]. Geodeticmeasurements
are not
yet adequateto documentmodernstrain rates for the plateau
region,but neotectonic
studiessuggest
HoloceneE-W extension
at an integratedrate of-10 mm/yr [Armijo et al., 1986], or
roughlyhalf the rate of N-S shorteningacrossthe Himalayan
rangesto the south.
The abrupt surface transition between "Himalayan" and
"Tibetan" strain fields shouldbe manifestedby a major, active
structuralsystem.Indeed,an importantfamily of faultshasbeen
mappedat thisposition,the SouthTibetanfault (STF) system
(Figuresl a and2a). The STF systemtypicallyseparates
thehighgrade Greater Himalayan Zone to the south from lower
amphibolitefacies to unmetamorphosed,
Cambrian-Eocene
sedimentary
rocksof the TibetanZone to the north.It consists
principally of regionallyextensive,low-anglenormal faults
(detachments)
thatdip shallowlynorthwardbeneaththe Tibetan
Zone, although some faults and ductile shear zones at the
boundaryalso have accommodated
N-S shorteningand rightlateraltranscurrent
displacement
[P•cher, 1991; Coleman,1996;
Hodgeset al., 1996].Whereit hasbeenextensively
studied
in the
AnnapurnaRangeof northcentralNepal (Figure2a), the STF
systemdisplaysevidence
of a complexsliphistorythatincludes
cyclingbetween
N-S extension
andN-S shortening
overa period
of severalmillion years [Hodgeset al., 1996; Fannayand
Hodges, 1996].
Structuraland geochronologicdata indicate that the STF
systemwas established
in early Miocenetime [Hodges,2000],
but the durationof STF activity is not well constrainedfor most
regionsin the Himalaya.In at leasttwo areas(Figure2a), near
the 8000 m peaks of Annapurna[Hodgeset al., 1996] and
Manaslu [Gulllot et al., 1994], major detachmentsof the STF
systemwere intrudedby middleto early Miocene(16-22 Ma)
granitesand cannothave moved subsequently.In most areas,
however,the systemincludesmultiple faultswith complicated
movementhistories,and the agesof many of thesestructuresare
constrainedonly to be youngerthan middle to early Miocene
[e.g.,Hodgeset al., 1996;Searleet al., 1997].
HODGES ET AL.: SOUTHWARD
EXTRUSION
OF TIBETAN
CRUST
80:5
Tibetan Plateaulies in a well-definedzone generallyboundto the
north and south by the surface traces of the STF and MCT
systems.This implies a direct relationshipbetween coordinated
displacementon extensionaland contractionalfault systemsand
high erosion rates. We interpret the neotectonic and
physiographiccharacteristicsof the southernplateau margin as
indicatingthe southwardextrusionof a crustalwedge which is
bound above by the STF system and below by the Himalayan
sole thrust (Figure lb). The current geometry of contractional
structuresin the Himalaya is such that the rate of extrusion is
Recent mapping at the southernterminus of the Thakkhola
graben,one of the mostconspicuous
N-S rift systemsin southern
Tibet (Figure 2a), provides important constraint on just how
young STF activity might be [Hurtado et al., 2001]. The main
growthstructureof this graben,the Dangardzongfault, has had a
history of episod•cdisplacementstretchingfrom circa 11 Ma to
as late as the Quaternary [Fort et al., 1982; Garzione et al.,
2000]. As the Dangardzongfault is traced southwardtoward the
Tibetan Zone - Greater Himalayan Zone contact,it can be shown
to offset an early strandof the STF systembut to be cut by the
latest strand of the system.This relationship,documentedand
describedin detail elsewhere [Hurtado et al., 2001], requiresthat
low-angle normal faulting along STF systempersistedinto the
Quaternary period at the longitude of the Thakkhola graben.
Here, at least, the surface strain discontinuity between the
Himalayan realm and the Tibetan Plateau has been
accommodated
by slip on the STF system.
Whether or not the STF system experienced Quaternary
displacementalong the entire length of the Himalayan chain
systems,producinga relatively narrow zone of rapid uplift that
correspondsto the high-relief band in Figures 2b and 2c. The
coincidenceof high relief and high rock uplift rate implies that
this is also a zone of unusually rapid erosion,an interpretation
that is consistentwith thermochronologicand isotopic data for
detrital mineral suites from modern Himalayan rivers and the
Bengal Fan, which indicatethat the GreaterHimalayanZone has
been the principle source for sediments removed from the
remains unknown, but several lines of evidence are consistent
with sucha scenariofor the central and easternHimalaya. Nakata
[1989] documented the existence of an important WNW-ESE
striking,NE dipping normal fault which offsetsglacial moraines
near the westernend of the Dhaulagiri Range (Figure 2a). Recent
field mapping and satellite image analysishave shown that this
structureis the probable westward extensionof the detachment
that truncatesthe Dangardzongfault, extendingthe strike length
of the STF structurewith documentedpost-Pleistoceneslip to at
least 100 km west of the Thakkhola graben (J.M. Hurtado and
K.V. Hodges, manuscript in preparation, 2001). Much farther
east, at roughly 89øE longitude, a reflection seismic profile
collected across the STF west of the Yadong-Gulu rift of
southern Tibet (Figure 2a) shows a Pliocene-Pleistocene
extensionalbasin developedin the hanging wall of the Zherger
La detachment,a relationshipwhich requires relatively recent
southern flank of the orogenic system since Miocene time
[France-Lanordet al., 1993; Brewer et al., 2000].
Numerical experiments such as those of Beaumont et al.
[1992] and Willett [1999] suggesta strongfeedbackrelationship
between deformational and erosional processesin orogenic
systems.In the case of the Himalaya the processesillustratedin
Figure lb togetherprovide a remarkablyefficient mechanismfor
the dissipation of excess potential energy. While southward
extrusionreducesthe crustalthicknessof the Himalayan-Tibetan
system(and thus its gravitationalpotential energyper unit area),
accelerated erosion at the margin of the system leads to the
removal of mass (and therefore energy) from the system
altogether. In our view, it is no accident that this "extrusionerosion" mechanismhas developed along the southernplateau
margin, where its dissipative efficiency is magnified by the
highest potential energy gradients in the Himalayan-Tibetan
slip on the basal STF detachmentin the region [Hauck et al.,
system.
1998].
Additional evidence for neotectonic activity on the STF
system comes from the nature of PT•. This transition is
manifestedin the field by a combinationof drainagedivides and
major knickpointson large, trans-Himalayanrivers, and it marks
a particularlyobviousdiscontinuityin slopemapsof the southern
Tibetan Plateauregion (Figure 2b). PT• probablyis not relatedto
a changein precipitationrate (becauseit lies entirely within the
rain shadowof the Himalaya) or the erosionalsusceptibilityof
the bedrock. It is also unlikely to be associated with a
One of the most interesting characteristicsof the extrusionerosion mechanism is that it accomplishesthe dissipation of
energy along a front of active N-S shorteningwithout interfering
with concurrent processesof energy accumulation related to
India-Eurasia convergence.Unlike the dissipative extensional
processesthat accommodatethe "collapse"of orogenicsystems
after convergence has ceased [Dewey, 1988], the processes
illustratedin Figure lb may persistover a significantpart of the
constructionalhistory of an orogen. Indeed, coupled thermalmechanicalmodels of large orogenswith orographicallyfocused
erosion,such as that taking place along the southernmargin of
the Tibetan Plateau, naturally evolve toward channeledflow of
the middle crusttoward the erosionfront [Jamiesonet al., 2001].
Studentsof the earlier history of the Himalayan orogen will
note that the extrusion-erosionmechanismillustratedin Figure lb
bearsstrongsimilarity to previouslypublishedinterpretationsof
the deformationalprocessesoperativein Miocene time, just after
the STF had been establishedand when the MCT probablyserved
as the Himalayan sole thrust. Burchfiel and Royden[1985] first
suggestedthat the STF may have developedduringgravitational
collapse of the Miocene topographic front between India and
Eurasia and may have accommodatedthe southwardextrusionof
a wedge of material between the MCT and STF systemsat that
time. Additional geological, petrologic, and geochronological
evidencefor a dynamicalcouplingof Miocene movementon the
fundamentalchangein erosionalmechanism,sincethe transition
is not spatiallyrelatedto the limit of glacialactivity in the region
[Duncan et al., 1998]. We interpret the sharpnessof PT• as an
indication of recent tectonic activity. Indeed, the discontinuity
itself correspondsalmost exactly to the youngeststrand of the
STF system in those regions where the fault system has been
mappedin detail and where digital elevationdata of sufficiently
high quality are availableto constrainthe positionof PT• and
knickpointson major drainages(Figure2c).
4. Effect of Coordinated
Deformation
and Erosion
on the SouthernPlateauMargin
Close scrutinyof the topographicslopemap in Figures2b and
2c showsthat the highestrelief along the southernmargin of the
accelerated
between
the surface
trace
of the
STF
and MCT
806
HODGES ET AL.: SOUTHWARD EXTRUSION OF TIBETAN CRUST
two fault systemswas publishedsubsequently[Searle, 1986;
SearleandRex, 1989;Burchfielet al., 1992;Hodgeset al., 1992,
1993, 1996],andMiocenewedgeextrusionalongtheHimalayan
frontwasarticulatedas a channelflow processby Grujic e! al.
[1996] and Grasemannand gannay.[1999]. A key question
regardingthe long-termefficiencyof thissetof processes
is how
energy is transferredfrom other parts of the systemto the
dissipativemargin. In our view this processrequiresthe
developmentof one or more channelsof middle or lower crustal
flow that feeds material to the southernmargin, a notion
suggested
earlierby Nelsone! al. [1996]andWue! al. [1998]on
the basisof geophysical
data interpretedas favoringa partially
moltenmiddlecrustbeneathTibet. The rockscurrentlyexposed
in the GreaterHimalayanZone,whichdisplayabundantevidence
of Miocenein situanatexisat middlecrustallevels,represent
the
modemleadingedgeof this feederchannelbut alsopreservea
record of the kinematic
evolution
of the channel in their
deformational
fabrics.Studiesof thesefeaturesby Grujic e! al.
[ 1996],Hodgeset al. [ 1996],and Grasemann
and Vannay[ 1999]
suggest that the extrusion-erosion
mechanism had been
established
alongthe southernplateaumarginby earlyMiocene
time,andtheneotectonic
evidencereviewedin thispaperargues
thatit is still activetoday.
5. Implications
From a thermodynamic
perspectivean orogenis an open
systemwhosebehavioris dictatedlargelyby how energyflows
through it. The relative rates of energy accumulationand
dissipationhelp definethree stagesof orogenesis.
In the first
stage newly establishedmountain ranges stockpile energy
becausethe rate of accumulationthroughcrustalthickening
greatlyexceeds
therateof dissipation.
In thesecondstagea fully
developedmountain range reachesits maximum elevation as
energy accumulativeand dissipative processescompete for
prominence.In the third and final stage,as plate convergence
wanes, energy dissipation mechanismspredominateand the
mountainrangeis eventuallydestroyed.
Between the time of initial India-Eurasia collision and the end
of the Oligoceneepochthe Himalayan-Tibetansystemwas a
first-stageorogen,with crustal shorteningand thickening
outpacingerosionand othermodesof energydissipation.
Near
the Oligocene-Miocene
boundary,therewas a dramaticincrease
in the rate of energydissipationas evidencedby an increased
erosionalinputto the HimalayanforelandandBengalFan [Galy
e! al., 1996], the earliestdocumentedcontinentalextrusionat the
southeastern
edgeof the Tibetanplateau[Chunget al., 1997],
andcoevaldevelopment
of the STF andMCT systems
[Hodgeset
al., 1996]. We interpretsuchchangesas evidenceof a transition
to the secondstageof orogenesis.
Along the southernmarginof
the Tibetan Plateau,linked displacementon the STF and MCT
systems permitted southward extrusion of the middle crust
(represented
by the GreaterHimalayanZone) towardthe Indian
forelandin early Miocenetime. However,this energydissipative
process has persisted episodically throughout much of the
Neogene, as has the energy accumulative process of crustal
accretionalongthe southernflank of the orogen,suggesting
the
spontaneousdevelopmentof a near steady state conditionthat
fluctuatesarounda balancebetweenenergyinflux andeffiux.
Exactlyhow the Himalayan-Tibetansystemhas responded
to
naturalperturbations
from the steadystateconditionover the past
20 million yearshasdependedon threeparameters
that change
over time and from place to place: the theology of the
lithosphere,the rate of crustalshortening,and the rate of erosion
alongthe Himalayanrangefront. A stronglithosphereis capable
of achievinga much greaterthicknessthan a weak lithosphere
beforecollapsingunderits own weight.At severaltimesduring
the early to middle Miocene, rocks of the Greater Himalayan
Zone underwentpartial melting, resultingin a dramaticdrop in
crustalstrength.We might expectsuchepisodesto coincidewith
major episodesof movement on the STF and MCT systems.
Indeed, a close relationshiphas been demonstratedbetween
crustalanatexisand STF and MCT displacement[Hodgese! al.,
1996]. The specific kinematics of movement on structures
boundingthe wedge during any phase of activity dependson
whether the orogenicsystemneeds to gain or lose energy in
responseto a perturbation from the steady state. If energy
accumulationis favored, the upper and lower bounding fault
systemsmay accommodateshortening. This may explain, for
example,evidencefor some episodesof early Miocene thrustsensedisplacement
on the STF systemin centralNepal [Hodges
et al., 1996; Vannayand Hodges, 1996]. If energydissipationis
favored,extrusionof the wedgeby normalslip on the upperfault
system and reverse slip on the lower bounding fault system
increasesstructuralrelief and enhancesthe orographiceffect on
precipitation,thus increasingthe efficiency of erosionalong the
Himalayanfront.
Somedeformationrelatedto the redistributionof energyalong
the southern margin of the Tibetan Plateau may be more
complicatedthan simple southwardextrusion.Figure l a reveals
significant variations in gravitational potential energy along
strike in the Himalaya. Evidence in the geological record for
transcurrentmovementon the STF and MCF systems,as well as
for range-parallel shortening and extension in the Greater
Himalayan zone [Burg et al., 1984; P•cher, 1991; Coleman,
1996], might indicate the lateral flow of material down such
gradientsin œu.
Many mechanismsbeganto work in concertto dissipatelarge
amountsof gravitationalpotential energy from the HimalayanTibetan systemin middle to late Miocene time. In addition to the
extrusion-erosionengine featured in this paper these include
processsetsthat accommodateE-W extensionof the uppercrust
in southernTibet [Armijo et al., 1986], eastwardextrusionof the
lower crust in easternTibet [Royden et al., 1997], and eastward
extrusion,clockwise rotation, and---N-S extensionof the upper
crustalongthe easternmarginof the TibetanPlateau[Burchfielet
al., 1995]. The vigorouscollaborationof theseprocessesin the
relatively recent past may simply be coincidental,but it might
also suggestthat someunusualand abruptenergyaccumulation
mechanismmay have driven the system far from its steadystate
configuration in middle Miocene time. One possibility is that
convectiveremoval of denselower lithospherebeneathTibet, and
its subsequent replacement by less dense asthenosphere,
catastrophicallyincreasedthe gravitationalpotential energyof
the region [England and Houseman, 1989], but this hypothesis
remainsuntestableby geologicmeans.
As reviewed by Molnar et al. [1993], there have been major
changesin climate in the Himalayan-Tibetanregionover the past
20 million years, most notably an abrupt intensificationof the
Asian monsoon in late Miocene time and an increasein glacial
ttODGES ET AL.' SOUTHWARD
activity in the Quaternary.The most likely impact of theseevents
would be to increaseerosionrates along the Himalayan front,
with possiblefeedbacksto other dissipativemechanismsas well
[Willerr, 1999; Beaumont et al., 1992], although there is no
unequivocalgeologicevidencefor a stronggeodynamicresponse
of the Himalayan-Tibetansystemto climatic forcing [Burbank et
al., 1993; Derry and France-Lanord, 1997]. The persistenceof
the extrusion-erosionenginedescribedhere, despitesuchevents,
implies that the effect of secularclimate changeis to increaseor
decreasethe efficiency of energy dissipation and to alter the
relative importanceof specificdissipativemechanisms,but not to
affect a major changein the overall processor to switch off the
engine altogether.
Evidence that the STF system is an active feature has
important implications for the modern geodynamicsof Tibet.
Termination of the principle growth fault of the Thakkhola
graben at the shallowly north dipping STF system [Hurtado et
al., 2001] implies that the rift systemsof southern Tibet are
rootless, kinematically decoupled from the remaining Tibetan
lithosphere. This observation,consistentwith geophysicaldata
from southernTibet [Nelsonet al., 1996], is at oddswith tectonic
models that require strong mechanical coupling between the
Tibetan and Himalayan realms [McCaffrey and Nabelek, 1998].
More generally,it suggestscautionin the use of surfacefaulting
patterns to infer a single style of deformation for the entire
lithosphere; for example, the lower crust of southern Tibet
appearsto extrude southwardover the downgoing Indian plate
while the upper crust extendseast-west.To make matterseven
more complicated,the fault plane solutionsof great earthquakes
beneath the southern plateau suggest that E-W extension,
kinematically similar to deformation in the upper crust but
fundamentallydissimilarto that in the lower crust, occursin the
mantle lithosphereof Tibet [Chen and Kao, 1997; Holt, 2000].
Thus the kinematic responseof the lithosphere to continental
References
Arm0o, R., P. Tappomer, J Mercier, and T. Han,
Quaternary extension in Southern Tibet: Field
observationsand tectonicunpllcanons,J. Geophys.
Res., 91, 13803-13872, 1986.
Baranowskl, J., J. Arlnbruster, L Seeber, and P.
Molnar, Focal depthsand fault plane solutionsof
earthquakesand active tectonicsof the Himalaya,
J. Geophys.Res.,89, 6918-6928, 1984.
Beaumont, C., P. Fullsack, and J. Halmlton, Eroslonal
control of active compressionalorogens,m Thrust
Tectontcs, edited by K R McClay, pp 1-18,
Chapmanand Hall, New York, 1992.
Bilham, R., K. Larson, J. Freymuiler, and P.I.
Members, GPS measurements of present-day
convergenceacrossthe Nepal HunMaya,Nature,
386, 61-64, 1997.
Bird, P., Lateral extrusion of lower crust from under
high topography,in the isostatichlmt, J Geophys
Res., 96, 10,275-10,286, 1991.
Brewer, I.D., D.W. Burbank, and K.V. Hodges, The
downstreamevolution of detrital mineral cooling
ages - Insights froin the Marsyandl Valley of
central Nepal, EOS Trans. AGU, 81 (48), Fall
Meet. Suppl.,AbstractT51C-04, 2000.
Brunei, M., Ductile thrusting in the Himalayas: Shear
sensecriteria and stretchingilneanons, Tectonics,
5, 247-265, 1986.
Burbank, D.W., L. Derry, and C. France-Lanord,
Lower Himalayan detrital sediment delivery
EXTRUSION OF TIBETAN CRUST
807
collisionis variablein threespatialdimensionsand the additional
dimension
of time.
If a steady state conditionsuch as the one describedhere can
developin orogenicsystemsandpersistover long timescales,it
followsthat the structuralhistoryof an orogendoesnot place
uniqueconstraints
on the historyof plateinteractions
responsible
for their development,limiting the value of simplekinematic
models. For example,evidencefor extensionaldeformationor
lateralextrusionin a convergent
orogenat a particulartime does
not necessarily
meanthatwholesalegravitationalcollapseof the
systemhad begunby that time. Instead,suchprocesses
may
simplyaccommodate
energydissipation
at a rate comparable
to
thatof energyaccumulation
throughotherprocesses
operatingat
the sametime elsewherein the system.A deeperunderstanding
of orogenic systemsrequires that we concentrateresearchon
more thanjust the temporalevolutionof geologicalstructuresand
landforms. We must also examine the geologic record for
evidence of persistentprocesssetsthat might imply dynamical
steady state conditions,and we must determine whether or not
the forcing factors responsiblefor the initial development of
major structuralfeaturescontinueto dictate their evolution over
periods that may last tens of millions of years. Expanding our
perspectiveto view the patternsof deformationand erosionin the
contextof energy flow may be an importantpart of rising to one
of the great challenges before continental tectonics today: to
define the processrelationshipsthat link the simple physics of
mountainbuildingto the rich complexityof mountainranges.
Acknowledgments.The authorswould like to thankSam Bowring,
Clark Burchfiel,Marin Clark, Kevin Furlong,Brad Hagcr, and Leigh
Roydcnfor valuablediscussions
rcgardingthe influenceof gravitational
potentialenergyon Himalayan-Tibetan
orogenesis.
CraigJonesprovided
codcto facilitate gered filtering. Both he and Clcm Chaserevicwcd thc
penultimateversion of the manuscript,and we appreciatetheir efforts.
This rcsearchwas sponsored
by grantsfrom theTectonicsProgramof thc
National
Scicnce Foundation.
Burchhei, B.C., and L.H. Royden, North-south
extensionwithin the convergentHimalayanregion,
Geology,13,679-682, 1985.
Burchfiel, B.C., Z. Chen, K.V Hodges,Y. L!u, L.H.
Royden, C. Deng, and J. Xu, The South Dbetan
Detachment System, litmalayan Orogen '
Extension ContemporaneousWtth and Parallel to
Shorteningin a ColhsionalMountatnBelt, 41 pp.,
Geol.Soc.Am., Spec Pap, 269, 1992.
Burchfiei, B,C., Z. Chen, Y. L!u, and L.H. Royden,
Tectonics of the Longmen Shan and adjacent
regions, central China, lnt Geol Rev, 37, 661735, 1995
Burg, J.P., M. Brunei, D. Gapals, G.M. Chen, and
G.H. L!u, Deformation of ieucogramtesof the
crystalline Main Central Sheet in southernTibet
(China), d Struct Geol, 6, 535-542, 1984.
Btirgmann, R., K.M. Larson, and R. Bllham, Model
inversion of GPS and leveling measurements
acrossthe Himalaya: Implicationsfor earthquake
hazardsand future geodeticnetworks,ttimalayan
Geol., 20, 59-72, 1999.
Carslaw,H.S., andJ.C. Jaeger,Conductionoj Heat in
Solids, 510 pp., Clarendon, Oxford, England,
1986.
Cattin, R., and J.P. Avouac, Modeling mountain
buildingand the seismiccyclein the Himalayaof
Nepal, d. Geophys.Res., 105, 13,389-13,407,
2000.
Chen, W.-P., and H. Kao, Seismotectomcsof Asia:
Chung,S.-L., T.-Y. Lee, C.-H. Lo, P.-L. Wang, C.-Y.
Chen,N.T. Yem, T.T. Hoa, andG. Wu, Intraplate
extensionprior to continentalextrusionalong the
Ailao Shan-RedRiver shear zone, Geology,25,
311-314, 1997.
Clark, M.K., and L.H. Royden, Topographicooze
Building the eastern margin of Tibet by lower
crustalflow, Geology,28, 703-706, 2000
Cobblentz, D.D., R.M. Richardson,and M. Sandiford,
On the gravitational potential of the Earth's
lithosphere,Tectonics,13,929-945, 1994.
Coleman, M.E., Orogen-parallei and orogenperpendicularextensionin the central Nepalese
Himalayas, Geol. Soc. Am. Bull., 108, 1594-1607,
1996.
Crank, J., The Mathematics oJ DtJJbsion,414 pp.,
Oxford Univ. Press,New York, 1975
DeCelles,P.G., G.E. Gehreis,J. Quade,andT.P. Ojha,
Eocene-earlyMioceneforelandbasindevelopment
and the history of Himalayan thrusting,western
and centralNepal, Tectonics,17, 741-765, 1998.
Derry, L.A., and C. France-Lanord, Himalayan
weathering and erosion fluxes: Climate and
tectonic controls, in Tectonic Uplift and Climate
Change, edited by W.F. Ruddunan,pp. 289-312,
Plenum,New York, 1997.
Dewey, J.F., Extensional collapse of orogens,
Tectonics, 7, 1123-1139, 1988.
Duncan. C., A. Klein, J. Masek, and B. Isacks,
despitean intensifiedmonsoonat 8 Ma, Nature,
somerecentprogress,
in The TectonicEvolutionof
Asia, editedby A. Yin and M. Harrison,pp. 37-62,
Comparisonof late Pleistocene
andmodernglacier
extentsin centralNepal basedon digital elevation
dataand satelliteimagery,Quar Res.,49, 241-254,
364, 48-50, 1993.
CambridgeUniv. Press,New York, 1997.
1998.
808
HODGES ET AL.: SOUTHWARD
England, P., and G. Houseman, Extension during
continentalconvergence,with application to the
Tibetan Plateau, J. Geophys Res, 94, 1756117579, 1989.
England,P., and P. Molnar, Active deformationof
Asia: From kinematicsto dynamics,Science,278,
647-650, 1997.
Fort, M., P. Freytet, and M. Colchen, Structuraland
sedlmentologlcalevolution of the ThakkholaMustang graben (Nepal Himalayas), Z
Geomorphol,42, 75-98, 1982.
France-Lanord, C., L. Derry, and A. Mlchard,
Evolution of the Himalaya since Miocene time:
Isotopic and sedimentologlcevidencefrom the
BengalFan,in ttimalayanTectomcs,
editedby P.J.
Treloar, and M. Searle, Geol Soc. Spec.Publ. 74,
605-621, 1993.
Furbish, D.J., Fluid Phystcs in Geology, 476 pp.,
Oxford Univ. Press,New York, 1997.
Galy, A., C. France-Lanord,
andL A. Derry, The late
Oligocene-early Miocene Himalayan belt:
Constraintsdeducedfrom isotopiccompositions
of
early Miocene turbidltes in the Bengal Fan,
Tectonophysics,
260, 109-118, 1996.
Gansser, A., Geology oJ the ttimalayas, 289 pp.,
Wfiey-Intersclence,
New York, 1964.
Garzlone, C.N., D.L. Dettman, J. Quade, P.G.
DeCelles, and R.F. Butler, High times on the
Tibetan Plateau: Paleoelevatlon
of the Thakkhola
Graben,Nepal, Geology.,
28, 339-342, 2000.
Grasemann, B., and J.-C. Vannay, Flow controlled
inverted metamorphismin shear zones, J. Struct.
Geol., 21,743-750, 1999.
Grujlc, D., M. Casey, C. Davidson, L.S. Hollister, R.
Kilndig, T. Pavlis, and S. Schmld, Ductile
extrusionof the Higher Hunalayan crystallinein
Bhutan: Evidence from quartz microfabrics,
Tectonophystcs,
260, 21-43, 1996.
Guillot, S., K.V. Hodges, P. Le Fort, and A. Pacher,
New constraints on the age of the Manaslu
!eucogranite: Evidence for eplsodic tectonic
denudationin the centralHimalayas,Geology.,22,
559-562, 1994.
Harrison, T.M., F.J. Ryerson, P. Le Fort, A. Yin, O.
Lovera, and E.J. Catlos, A late Mlocene-Phocene
origin for the central Himalayan inverted
metamorphism,Earth and Planetary Science
Letters, 146, El-E7, 1997.
Harrison, T.M., M. Grove, O.M. Lovera, and E.J.
EXTRUSION
OF TIBETAN
Tibetan Fault System in the central Nepalese
Himalaya, Geol Soc Am. Bull., 113, 222-240,
2001.
Jackson, M.E., and R. Bllham, 1991-1992 GPS
measurements across the Nepal Himalaya,
Geophys.Res.Lett., 21, 1169-1172, 1994.
Jackson,
M., S. Bamentos,R. Bfiham,D. Kyestha,and
B. Shrestha,
Uplift in theNepal Hunalayarevealed
by spirit leveling, Geophys.Res Lett., 19, 15391542, 1992.
Jamieson,R.A., C. Beaumont,M.H. Nguyen, and B.
Lee, Interaction of metamorphism,deformation,
and exhumationIn large convergentorogens,J.
Metamorph.Geol.,in press,2001.
Jones,C.H., J.R. Unruh, and L.J. Sonder,The role of
gravitational potential energy in active
deformationin the southwesternUnited States,
Nature, 381, 37-41, 1996.
Jouanne,F., J.L. Mugruer,M.R. Pandey,J.F. Gamond,
P. Le Fort, L. Serruner, C. Vigny, and J.P.
Avouac,Obliqueconvergence
in the Himalayasof
westernNepal deducedfroin preliminaryresultsof
GPS measurements,
GeophysRes.Lett., 26, 1933-
Kruse, S., M. McNutt, J. Phlpps-Morgan,and L.
Royden,Lithosphericextensionnear Lake Mead,
A model for ductile
flow
in the lower
crust,J. GeophysRes, 96, 4435-4456, 1991.
Larson, K., R. Burgmann, R. Bllham, and J.
Freymueller, Kinematics of the India-Eurasia
collision
zone froin
GPS
measurements,
J.
Geophys.Res, 104, 1077-1093, 1999.
Lav•, J., and J.P. Avouac, Active folding of fluvial
terraces across the Siwalik Hills, Himalayas of
central Nepal, J. Geophys.Res, 105, 5735-5770,
2000.
Le Fort, P, Himalayas: The collided range. Present
knowledgeof the continentalarc, Am. J. Sci., 275A, 1-44, 1975.
Lemmne, F.G., et al., The developmentof the joint
NASA GSFC and the National Imagery and
Mapping Agency (NIMA) geopotenttal model
EGM96, NASA Tech. Pap NASA/TP-1998206861, 600 pp., 1998.
Macfarlane,A., K.V. Hodges,and D. Lux, A structural
analysisof the Main Central thrustzone, Langtang
National Park, centralNepal Himalaya, Geol. Soc.
Am. Bull., 104, 1389-1402, 1992.
Masek, J.G., B.L. Isacks, T.L. Gubbels, and E.J.
Catlos, A model for the origin of Himalayan
anatexisand inverted metamorphlsm,J. Geophys.
Fielding, Erosion and tectonics at margins of
continentalplateaus,J. Geophys Res., 99, 13,941-
Res., 103, 27,017-27,032, 1998.
Hauck, M.L., K.D. Nelson, L.D. Brown, W. Zhao, and
A.R. Ross, Crustal structure of the Himalayan
orogen at-90 ø east longitude from ProJect
INDEPTH deep reflection profiles, Tectonics,17,
McCaffrey, R., and J. Nabelek, Role of oblique
convergence in the active deformation of the
Himalayasand southernTibetan plateau, Geology,
481-500, 1998.
112, 324-350, 2000.
Hodges, K.V., R. Parrish, T. Housh, D. Lux, B.C.
Burchfiel, L. Royden, and Z Chen, Simultaneous
Miocene extension and shortening in the
Himalayanorogen,Sctence,258, 1466-1470, 1992.
Hodges,K.V., B.C. Burchfiel, L.H. Royden, Z. Chen,
and Y. Llu, The metamorphic signature of
contemporaneousextension and shortening in the
central Himalayan orogen:Data from the Nyalam
transect,southernTibet, J. Metamorph. Geol, 11,
721-737, 1993.
Hodges, K.V., R R. Parrish, and M.P. Searle, Tectonic
evoluuon of the central Annapuma Range,
Nepalese Himalayas, Tectomcs, 15, 1264-1291,
1996.
26, 691-694, 1998.
Plateau,Geology,,
28, 91-94, 2000.
Molnar, P., A review of geophysicalconstraintson the
deep structureof the Tibetan Plateau, the Himalaya
and the Karakoram,and their tectonicimplications,
in The GeologicalEvolution of Tibet, vol. 326 A,
edited by C. Chang, et al, Philos Trans. R. Soc.
London,Set. A, 33-88,1988.
Molnar, P., A review of the seismicity and rates of
active underthrusting and deformation at the
Himalaya, d. Himalayan Geol, 1, 131-154, 1990.
Molnar, P., and H. Lyon-Caen, Some simple physical
aspectsof the support,structure,and evolution of
mountain belts, in Processes tn Conttnental
LithosphericDeformation,edited by S. Clark, B.C.
Burchfieland J. Suppe,Geol. Soc Am. Spec.Pap.,
218, 179-207, 1988.
Molnar, P., and P. Tappomer, Cenozoic tectonicsof
Holt, W.E., Correlated crust and mantle strain fields in
Tibet, Geology,,28, 67-70, 2000.
Howard, A.D., M.A. Smdl, and W.E. Dietrich,
Modelingfluvial erosionon regionalto continental
scales,J. Geophys.Res.,99, 13,971-13,986,1994.
Hurtado, J.M., K.V. Hodges, and K.X. Whipple,
Neotectonics
13,956, 1994.
McQuarne, N., and C.G. Chase, Raising the Colorado
Hodges,K.V., Tectonicsof the Himalaya and southern
Tibet from two perspectives,Geol Soc.Am Bull.,
of
the
Thakkhola
Graben
and
implications for recent activity on the South
Asia: Effects of a continental collision, Science,
189, 419-426, 1975.
Molnar, P., and P. Tapponmer, Active tectonics of
Tibet, J. Geophys.Res., 83, 5361-5375, 1978.
Molnar, P., P. England, and J. Martinod, Mantle
dynamics, uplift of the Tibetan Plateau, and the
Indian monsoon, Rev. Geophys., 31, 357-396,
1993.
Nakata, T., Active faults of the Hunalaya of India and
Nepal, in Tectontcs oJ the Western ttimala12as,
edited by L.L. Malincomco and R.J. Lllhe, Geol.
Soc Am Spec.Pap, 232,243-264, 1989.
Nelson, K.D., et al., Partially molten middle crust
beneath southern Tibet: Synthesis of Project
INDEPTH results,Science, 274, 1684-1688, 1996.
N1, J., and M. Barazangl, Seismotectonics of the
Himalayan collision zone: Geometry of the
underthrustingIndian Plate beneaththe Himalaya,
d. Geoph12s.
Res.,89, I 147-1163, 1984.
Pandey, M R., R.P. Tandukar, J.P. Avouac, J. Lav6,
and J.P. Massot, Interseismic strain accumulation
on the Hnnalayan crustalramp (Nepal), Geophys
Res Lett, 22, 751-754, 1995.
Pandey, M R., R.P. Tandukar, J.P. Avouac, J. Vergne,
and T. H6nuer, Seismotectomcs of the Nepal
Himalaya from a local seismic network, d. Astan
Earth Sct., 17, 703-712, 1999.
Pficher,A., The contactbetweenthe Higher Himalyan
crystallinesand the Tibetan sedimentary series:
Miocene large-scale dextral shearing, Tectomcs,
10, 587-599, 1991.
Royden,L.H., B.C. Burchild, R.W. King, E. Wang, Z.
1936, 1999.
Nevada:
CRUST
Chen, F. Shen, and Y. Llu, Surface deformation
and lower crustal flow in eastern Tibet, Sctence,
276, 788-790, 1997
Schelling,D., The tectonostratigraphy
and structureof
the eastern Nepal Hnnalaya, Tectomcs,11, 925943, 1992.
Schelling, D., and K. Arita, Thrust tectonics,crustal
shortening, and the structure of the far-eastern
Nepal Himalaya, Tectomcs,10, 851-862, 1991.
Searle, M.P., Structural evolution and sequence of
thrusting in the High Himalayan, Tlbetan-Tethys
and Indus suture zones of Zanskar and Ladakh,
Western Himalaya, d. Struct. Geol., 8, 923-936,
1986.
Searle, M.P., and A.J. Rex, Thermal model for the
Zanskar Himalaya, d. Metamorph. Geol., 7, 127134, 1989.
Searle, M.P., R.R. Parrish,K.V Hodges,A Hurford,
M.W. Ayres, and M.J. Whitehouse, Shisha
Pangma leucogranite, South Tibetan Himalaya:
Field relations, geochemistry,age, origin, and
emplacement,d. Geol., 105, 295-317, 1997.
Seebet, L., and V. Gornitz, River profiles along the
Himalayan arc as indicators of active tectonics,
Tectonophystcs,
92, 335-367, 1983.
Seeber, L, J.G. Armbruster, and R.C. Quittmeyer,
Seismicity and continental subduction in the
Himalayanarc, in Zagros, Hindu Kush, ttimalaya
GeodynamtcEvolution, Geodyn Ser., vol. 3,
edited by H.K. Gupta and F.M. Delany, pp. 215242, AGU, Washington,DC, 1981.
Sklar, L., and W.E. Dietrich, River longitudinal
profiles and bedrock incision models: Stream
power and the influence of sediment supply, in
Rivers Over Rock. Fluvial
Processes m Bedrock
Channels,Geophys.Monogr. Set., vol. 107, edited
by K.J. Tinkler and E.E. Wohl, pp. 237-260, AGU,
Washington,D. C., 1998.
Srivastava, P., and G. Mltra, Thrust geometries and
deep structureof the outer and lesserHimalaya,
Kumaun and Garhwal (India): Implications for
evolution of the Himalayan fold-and-thrustbelt,
Tectomcs, 13, 89-109, 1994.
St6cklin,J., K.D Bhattarai,V. Slngh-Chetri,and A.N.
Bhandan, Geologic map of Kathmanduarea and
CentralMaharabharatRange(1:250,000), Dep. of
Mines and Geol., Kathmandu,Nepal, 1980.
Tammoto, T., and T. Okamoto, Change of crustal
potentialenergyby earthquakes:
An indicatorfor
extensionaland compressional
tectonics,Geophys.
Res. Lett., 27, 2313-2316, 2000.
Tapponier, P., and P. Molnar, Slip-line field theory
and large-scalecontinentaltectonics,
Nature,264,
319-324, 1976.
Tucker, G.E., and R.L. Slingerland,Drainage basin
responseto climate change,Water Resour.Res.,
33,2031-2047, 1997.
Upretl, B.N., and P. Le Fort, Lesser Himalayan
crystalline nappes of Nepal: Problemsof their
HODGES ET AL.: SOUTHWARD
origin, in H•malaya and Tibet, in Mountatn Roots
to Mountatn Tops, edited by A. Macfarlane, R.B.
Sorkhabi and J Quade, Geol. Soc Am Spec. Pap
328, 225-238, 1999.
Vannay, J.-C., and K.V. Hodges,Tectonometamorphic
evolution of the Himalayan metamorphic core
betweenAnnapumaand Dhaulagin,centralNepal,
J Metamorph Geol, 14, 635-656, 1996.
Whipple, K.X, and G.E. Tucker, Dynamics of the
stream-power river incision model: Implications
for height lumts of mountain ranges, landscape
response timescales, and research needs, J
GeophysRes, 104, 17,66i - 17,674, 1999.
Wlllett, S.D., Orogeny and orography:The effects of
EXTRUSION
OF TIBETAN
erosion on the structure of mountain belts, J.
Geophys Res, ]04, 28957-28981, 1999.
Wu, C., K.D. Nelson, G Wortman, S.D. Samson, Y.
Yue, J. Ll, W.S.F. Kidd, and M.A. Edwards,
Yadong cross structure and South Tibetan
detachment in the east central Hnnalaya (89 ø90øE), Tectomcs,17, 28-45, 1998.
Yeats, R.S., and R.J Lfihe, Contemporarytectonicsof
the Hnnalayan Frontal Fault System. Folds, blind
thrustsand the 1905 Kangraearthquake,d. Struct
Geol, 13,215-225,
CRUST
Zhao, W., K.D Nelson, and P.I. Teain, Deep seismic
reflection evidence for continental underthrusting
beneath southern Tibet, Nature, 366, 557-559,
1993.
K.V. Hodges, J.M. Hurtado, and K.X. Whipple,
Department of Earth, Atmospheric,and Planetary
Sciences, Massachusetts Institute of Technology,
Cambridge,MA 02139, USA. (kvhodges•mit edu,
hurtado(•mit.edu,kxw(•mit edu).
1991.
Yeats,R.S., T. Nakata,A. Farah,M Fort,M.A. Mlrza,
M.R. Pandey, and R.S. Stein, The Himalayan
frontalfaultsystem,Ann Tecton.,6, 85-98, 1992.
809
(ReceivedJanuary12,2001;
revised June 14, 2001;
acceptedJune22, 2001 )