Cenozoic evolution of the eastern Pamir: Implications for
strain-accommodation mechanisms at the western end of
the Himalayan-Tibetan orogen
Alexander C. Robinson†
An Yin‡
Craig E. Manning§
T. Mark Harrison#
Department of Earth and Space Sciences, University of California, Los Angeles, California 90095-1567, USA
Shuan-Hong Zhang††
Xiao-Feng Wang‡‡
Institute of Geomechanics, Chinese Academy of Geological Sciences, Beijing 100081, People’s Republic of China
ABSTRACT
Detailed field mapping, geochronologic
and thermochronologic analyses, and pet-
rologic investigations conducted along the
southern segment of the late Cenozoic Kon-
gur Shan extensional system provide new
information on the Cenozoic tectonic evolu-
tion of the eastern Pamir at the western end
of the Himalayan-Tibetan orogen. Field rela-
tions and cooling-age patterns in the hanging
wall and footwall of the active faults show a
southward decrease in the magnitude of east-
west extension along the southern Kongur
Shan extensional system, from 20 km or less
along the Muztaghata massif in the north,
to <3 km along the Tashkorgan fault in the
south. These results, in conjunction with
previously published work on the northern
segment of the Kongur Shan extensional sys-
tem, show a general southward decrease in
east-west extension along the entire length of
the extensional system, consistent with mod-
els of extension primarily driven by oroclinal
bending or radial thrusting of the Pamir. Pet-
rologic data, 40Ar/39Ar cooling ages, and mon-
azite Th-Pb ages from schists and gneisses
in the footwall of the southern Kongur Shan
normal fault along the Muztaghata massif
†
Present address: Department of Geosciences,
University of Houston, Houston, Texas 77204, USA;
acrobinson@uh.edu.
‡
E-mail: yin@ess.ucla.edu.
§
E-mail: manning@ess.ucla.edu.
#
E-mail: tmh@argon.ess.ucla.edu.
††
E-mail: zhangshuanhong@sina.com.
‡‡
E-mail: wxf@cags.net.cn.
GSA Bulletin; July/August 2007; v. 119; no. 7/8; p. 882–896; doi: 10.1130/B25981.1; 12 figures; 1 insert; Data Repository item 2007134.
882
record two tectonic events that immediately
preceded late Miocene initiation of east-west
extension: (1) high-grade schists and gneisses
experienced upper amphibolite facies meta-
morphic conditions (9–10 kbar, 700–750 °C)
dated as late Oligocene to middle Miocene
by in situ ion-microprobe analyses of mona-
zite inclusions in garnet; and (2) high-grade
schists and gneisses were subsequently rap-
idly exhumed to shallow crustal levels in the
late Miocene with 40Ar/ 39Ar biotite cooling
ages of 7.5–9 Ma. Rapid exhumation was
accommodated in part by the east-west–
striking, south-dipping, Shen-ti normal fault.
Field relations and regional geologic corre-
lations indicate that this exhumation event
was related to the formation of the Central
Pamir gneiss domes, and the antiformal
Muztaghata massif is the eastward continu-
ation of the Sares dome of the Central Pamir.
These observations suggest that the antifor-
mal gneiss domes of the Central Pamir have
not been offset across the Karakorum right-
slip fault from the Qiangtang anticlinorium
in Tibet. Instead, we propose that the devel-
opment of the Central Pamir gneiss domes
may have been related to Oligocene-Miocene
northward underthrusting and thickening of
crust beneath the Pamir.
Keywords: Pamir, extension, gneiss dome, geo-
chronology, metamorphism, Karakorum.
INTRODUCTION
While knowledge of the tectonic history of
the Tibetan Plateau has substantially improved
For permission to copy, contact editing@geosociety.org
© 2007 Geological Society of America
in recent years (e.g., Matte et al., 1996; Mur-
phy et al., 1997; Horton et al., 2002; Haines
et al., 2003; Kapp et al., 2003a, 2003b; Tay-
lor et al., 2003; Chung et al., 2005; Spurlin et
al., 2005; Leier et al., 2007), the evolution of
the northwestern portion of the Himalayan-
Tibetan orogen, especially within the Pamir
(Fig. 1), is relatively poorly understood. One of
the more difficult problems has been determin-
ing the correct correlation of tectonic terranes
between the Tibetan Plateau to the east and the
Pamir–western Himalayan syntaxis region to
the west (e.g., Schwab et al., 2004) due in part
to political borders, difficult access over high-
altitude terrane, and a lack of clear offset geo-
logic markers. This problem has hindered our
ability to determine (1) the magnitude of lateral
offset along Cenozoic right-lateral Karakorum
fault and its role in lateral extrusion tectonics
(e.g., Peltzer and Tapponnier, 1988; Searle,
1996; Searle et al., 1998; Lacassin et al., 2004;
Schwab et al., 2004), and (2) whether the
asymmetry of the India-Asian collision zone is
primarily the result of eastward lateral extru-
sion, along-strike variations in the distribution
of shortening, or thrust system geometry (e.g.,
Tapponnier et al., 1982; Peltzer and Tappon-
nier, 1988; Houseman and England, 1996; Neil
and Houseman, 1997; Cowgill et al., 2003).
Our field and analytical results along the south-
ern end of the Kongur Shan extensional system
suggest that structural features of the Central
Pamir terrane, the Muskol-Sares domes, con-
tinue across the north-striking Cenozoic Kon-
gur Shan extensional system and have not been
offset from the Qiangtang anticlinorium along
the Karakorum fault.

REGIONAL GEOLOGY
Tectonic Divisions
Understanding of the Cenozoic deformation
history of the Pamir requires knowledge of both
the original distribution of markers prior to the
India-Asia collision and the mode of deforma-
tion during the collision. The western Hima-
layan-Tibetan orogen is composed of several
distinctive terranes that evolved along, or were
accreted to, the southern margin of Asia dur-
ing the late Paleozoic–Mesozoic closure of the
Tethys Ocean (e.g., Tapponnier et al., 1981;
Şengör and Okurogullari, 1991; Burtman and
Molnar, 1993; Hsu et al., 1995; Pan and Bian,
1996; Şengör and Natal’in, 1996; Yin and Harri-
son, 2000; Xiao et al., 2002; Schwab et al., 2004)
(Figs. 1 and 2). While these terranes are gener-
ally interpreted as along-strike equivalents of ter-
ranes in central and eastern Tibet to the east and
Afghanistan to the west (Tapponnier et al., 1981;
Burtman and Molnar, 1993; Yin and Nie, 1996;
Gaetani, 1997; Schwab et al., 2004), the correla-
tion of these terranes is a subject of considerable
debate. One interpretation is the South Pamir–
Karakorum–Hindu Kush terrane correlates to the
Qiangtang terrane of Tibet (Fig. 1) (e.g., Yin and
Harrison, 2000), and this is supported by inter-
preted low magnitudes of right-lateral strike-slip
motion along the southern Karakorum fault of
60–150 km (Searle, 1996; Searle et al., 1998;
Murphy et al., 2000). A competing interpreta-
tion is that the Qiangtang and Lhasa terranes
are equivalent to the Central Pamir and South
Pamir–Karakorum–Hindu Kush, respectively,
implying hundreds of kilometers of right slip
along the Karakorum fault (Lacassin et al., 2004;
Schwab et al., 2004). The latter is supported in
part by correlation of similar-age magmatic belts
and an anticlinorium in the Central Pamir, which
includes the Muskol and Sares domes, that is
interpreted to be equivalent to the Qiangtang
anticlinorium (Schwab et al., 2004).
Cenozoic Tectonic Framework
The western Himalayan-Tibetan orogen is
bounded in the north by the south-dipping Main
Pamir thrust and thrusts of the Western Kunlun
Shan (Figs. 1 and 2). The Main Pamir thrust
has been interpreted to have accommodated
~300 km of southward subduction of crust of the
Tarim-Tajik Basins during the Cenozoic (Burt-
man and Molnar, 1993; Thomas et al., 1994).
Initiation age of slip along the Main Pamir
thrust is considered to be either late Oligocene
(Thomas et al., 1994, 1996; Sobel and Dumi-
tru, 1997) or middle Eocene (Yin et al., 2002).
In addition to overthrusting of the Pamir along
Geological Society of America Bulletin, July/August 2007
Cenozoic evolution of the eastern Pamir
70°E 80°E
40°N n
Sha
Tian
MPT
1
Figure 2
Tajik 2 Tarim Basin
Basin Pamir
3
4 AKMS
Ka
rak
or
Tibetan Plateau
um
Fa
30°N
ult
Hi
m al
M
ain
Fro
n tal
India
70°E 80°E
Figure 1. Simplified tectonic map of the Indo-Asian collision zone showing major active
structures and suture zones (after Burtman and Molnar, 1993; Yin and Harrison, 2000).
MPT—Main Pamir thrust; IYS—Indus-Yalu suture; BNS—Bangong-Nujiang suture; JS—
Jinsha suture; AKMS—Ayimaqin-Kunlun-Mutztagh suture. Terranes of the western Indo-
Asian collision zone are: 1—Northern Pamir; 2—Central Pamir; 3—South Pamir–Karako-
rum–Hindu Kush; 4—Kohistan arc.
the Main Pamir thrust, the region has also been
internally shortened by 340 km along thrusts
including the Tanymas and Pshart fault zones
(Fig. 2) (Burtman and Molnar, 1993; Waldhör et
al., 2001). However, the timing of this shorten-
ing is not well determined, and much of it could
be pre-Cenozoic in age.
Active deformation in the Pamir is domi-
nated by east-west extension along the 250-km-
long Kongur Shan extensional system in the
east (Arnaud et al., 1993; Brunel et al., 1994;
Robinson et al., 2004) and the Karakul rift in
the west (Strecker et al., 1995; Blisniuk and
Strecker, 1996) (Figs. 2 and 3). Age constraints
from the northern Kongur Shan system indicate
initiation of extension at 7–8 Ma (Robinson et
al., 2004). Three kinematic models have been
proposed to explain the active extension in the
Pamir: (1) topographic collapse of the eastern
Pamir during growth of the Kongur Shan and
Muztaghata gneiss domes over a crustal-scale
thrust ramp (Brunel et al., 1994), (2) radial
thrusting along the Main Pamir thrust and/or
oroclinal bending of the Pamir salient (Strecker
et al., 1995; Yin et al., 2001), and (3) a trans-
tensional bend in the right-slip Karakorum
fault, linking late Cenozoic extension in the
Pamir with extensional structures in southwest
Tibet (Ratschbacher et al., 1994; Murphy et al.,
2000). However, our previously published work
along the Kongur Shan extensional system (and
observations presented here) argues against any
90°E 100°E
40°N
Altyn Tagh Fault
Son
JS
gpan
- Gan
Qiangtang zi
BNS
Lhasa 30°N
IYS
ayas
Thrust
90°E 100°E
kinematic link with the Karakorum fault to the
south (Robinson et al., 2004).
Active internal deformation within the Pamir
also occurs along a series of northwest-striking
right-slip faults in the southeastern Pamir region
(e.g., the Aksu-Murgab and East Pamir faults;
Fig. 2), which have been interpreted to link
with the Karakorum fault to the south (Strecker
et al., 1995). Slip rates on these structures are
estimated to be <1 mm/yr, indicating that, aside
from the extensional structures, active internal
deformation of the Pamir is limited.
Central Pamir Gneiss Domes
Within the Central Pamir terrane, fault-
bounded high-grade schists and gneisses with
Miocene cooling ages are exposed in the cores
of the east-west–trending Muskol and Sares anti-
forms (Pashkov and Dmitriyev, 1982; Peykre et
al., 1982; Schwab et al., 2004) (Fig. 2), which
we refer to as the Central Pamir gneiss domes.
The metamorphic cores experienced greenschist
to upper amphibolite facies metamorphic con-
ditions and locally preserve migmatitic textures
inferred to be Cenozoic in age (Pashkov and
Dmitriyev, 1982; Schwab et al., 2004). Schwab
et al. (2004) interpreted the eastern margin of
the Sares dome to be truncated by the Karako-
rum fault, where the antiformal structures rep-
resent the offset westward continuation of the
Qiangtang anticlinorium (Kapp et al., 2000; Yin
883
 Robinson et al.

73° E-O 74° 76° Geologic units:

Q

nsu fa
u lt
N
Main Pa Q Quaternary
Marka mi Neogene
D r N
E-O Eocene/Oligocene

Th

Sedimentary and Metamorphic units

P
Muji
Karakul Q
fault
Tarim Basin K Cretaceous

rus
Tr Trgr D
Trgr gr
J Jurassic
39° graben
Trgr P
Q
Kon Tr-J Triassic-Jurassic
K

t
Pzm
J

gu
Trgr Trgr Pzgr
N Tr Triassic
Trgr E-O
Trgr

rS
C-P K P-Tr Permian-Triassic
Trgr D D
Ta C C P Permian

ha n
K Pzm
ny Ghez

Kumt
C
m as K Trgr J C-P Carboniferous-Permian
K K fault
E-O
MuskolJ Antiform fau Carboniferous
K C

norma
Tr-J
lt Q E-O S
Pzgr J

agh f
Czgr Pz2
Czgr Sares
C-P Antiform Pzgr
D Devonian
P-Tr Pz2
C
Tr
-J Czgr Tr-J
P N S Silurian
C
t Pshart fault P
rez faul
P-Tr
Lyangar-Sa
Ordovician

ault
Pz2 O
Tr E-O C Tr-J

l fa
Tr Czgr O P E-O
E-O Czgr J
Rushan-Pshart fault K Czgr
Trgr Pzgr K Pzm Paleozoic Metamorphics
P-Tr gr
J
P-Tr J C

ult
Kgr
Czgr
Kgr C-P J
P-Tr
Trgr Pz2 Paleozoic basement (2)
Tr
Ea

C-P Czgr
J Tr J Q
38° P-Tr Q Pz1 Paleozoic basement (1)

KS
st

C-P P
Kgr
Pa

Pz2 E-O C
So C-P Pz2 C Z Proterozoic basement
m

Kgr
ut D
ir

Kgr Kgr
Alichur fault h P-Tr Pzgr Pzgr O
NTra ern
fa

J Czgr Cenozoic plutons

Igneous units
Tash
Czgr S
ul

Pz2 pp Pa J J
t

Kgr Kgr Jgr

Q e m C-P Trgr Kgr Cretaceous plutons
irJ Czgr
korg
Pz1
O
Pz2 Q Jgr Jurassic plutons
Kgr P
P-Tr
Aks P-Tr Q Pzgr
an
Kgr
u-M Trgr Triassic plutons
ult
P
urg
fault

E- O J J
a ab S Pz1 Jgr P
Paleozoic Plutons
rf
Pzgr
i Q
K fa ult E-O
amPz
Tr
JS

2
Pzgr
hP
K
J
S
C Map Symbols:
ut
Z Kgr Q J
P Pzgr
D Thrust fault
So Kgr
Figure 3 K
K

lt Pz1 Pzgr Normal fault

n fau
C-P J
Z
a Kgr
Vakh P Pz1
Strike-slip fault
Ka fa

AK

Pzgr
37°
C-P Ka Pz 1
37°
C-P Buried or poorly
ra ult

r
MS

Fa aka
Kgr J

fault constrained fault

ko

Tr-J Jgr

Reshun
Kgr
Kilik ul x
ru

faul t Suture zone

73° P 74° 75° 76° 77°
m

P D Kgr t Pzm S Pz1

Figure 2. Simplified geologic map of the eastern Pamir, northern Karakoram, and Western Kunlun Shan showing distribution of lithologic
ages and Cenozoic faults (after Yin and Bian, 1992; Strecker et al., 1995; Sobel and Dumitru, 1997; Robinson et al., 2004; Schwab et al.,
2004). KS—Kudi suture; AKMS—Ayimaqin-Kunlun-Mutztagh suture; JS—Jinsha suture.

and Harrison, 2000), yielding ~200 km of offset a northwest-trending syncline (Figs. 3 and 4) Early Jurassic granites and granodiorites that
along the northern Karakorum fault. (Liu, 1988; Yin and Bian, 1992). intrude high-grade metasedimentary rocks and
gneisses in the hanging wall and footwall of the
LITHOLOGIES OF THE SOUTHERN Quaternary Units Kongur Shan normal fault, and (2) Cenozoic
KONGUR SHAN REGION Quaternary units along the Kongur Shan plutons of the Tashkorgan igneous complex
extensional system consist dominantly of alluvial (Zhang et al., 1996) exposed along the western
Sedimentary Rocks and fluvial deposits (Q1 and Q2; Figs. 3 and 4) side of the central Tashkorgan Valley.
and glacial moraine deposits (Qm). Fine-grained
Paleozoic and Mesozoic Units lacustrine deposits of inferred Quaternary age Mesozoic Units
The southern end of the Tashkorgan Valley are exposed sporadically along the northern end Triassic to Early Jurassic granites and grano-
exposes a sequence of Permian limestones and of the Tashkorgan Valley, possibly deposited diorites (Mzgr) consist of medium-grained bod-
slates (Liu, 1988; Yin and Bian, 1992) (Figs. 3 while the outlet of the valley was dammed dur- ies ranging in size from sills several meters in
and 41). The sequence consists of structurally ing a glacial maximum. Fine-grained deposits thickness to regionally extensive plutons. Plu-
higher limestones interlayered with minor slate drape the western side of the Tashkorgan Valley tons within the hanging wall of the Kongur Shan
and sandstone, (Pl), which overlie a thick slate at the southern end of the Tashkorgan fault (Ql), normal fault intrude upper amphibolite schists
unit (Psl) that dips moderately to the northeast. and they are interpreted to be loess deposits. (Pzsch1) and are unfoliated to weakly foliated
Along the western side of the valley, Permian While these are inferred to be Quaternary, no (Robinson et al., 2004). Footwall plutons are
units are juxtaposed across a north-striking fault age constraints are available. exposed primarily within the core of the Muz-
against a thick sequence of Triassic-Jurassic taghata massif and intrude quartzofeldspathic
limestone (Jl1 and Jl2) unconformably overlain Igneous Rocks gneisses (Pzgn1) (Liu, 1988; Yin and Bian, 1992).
by Cretaceous red beds (Kss) that are folded into Granitic rocks are weakly foliated along the
Igneous units along the southern end of the southeastern side of the Muztaghata massif,
1
Figures 4 and 5 are on a separate insert accompa- Kongur Shan extensional system consist of but are strongly foliated with mylonitic fab-
nying this issue. two primary groups (Fig. 3): (1) Triassic to rics to the northeast of the massif. U-Pb zircon

884 Geological Society of America Bulletin, July/August 2007

 Cenozoic evolution of the eastern Pamir

dating of plutons in the hanging wall (Rob-

inson et al., 2004) and the footwall (Xu et al.,
1996; Robinson et al., 2004) has yielded ages
75° of 230–199 Ma, consistent with previous age
Tarim Basin
Muji
Fault
Qm assignments to the late Paleozoic–early Meso-
Pzsch2
Q2
zoic South Kunlun plutonic belt (Pan, 1992;
Qm Youngun and Hsu, 1994; Matte et al., 1996; Xu
Q1
Kon Shala Tala Fault et al., 1996; Cowgill et al., 2003).

Kalagile Fault Cenozoic Units

Qm Qcgl 39°
g

Dgw
QcglJss
ur

Mzgr
Q2 Cenozoic plutons of the Tashkorgan alka-
Sh

Pzsch2 Pzsch2 Ov
an Mzgr
Mzgr Q2 Fa
Kss1
line complex are exposed for ~60 km along the
Jss
ult Dgw western side of the Tashkorgan Valley (Esy and
Qi
ak Q2 Pzgn3
N Pzsch1 lak
Fa Ov Mgr; Figs. 3, 4, and 5 [see footnote 1]) (Zhang
ult Pzgn1 et al., 1996). They intrude high-grade meta-
Pzsch1
Q1 Pzgn2 Mzgr Qm
morphic rocks of inferred Permian age to the
Dgw
Mzgr
Ghez Fault
west (Zhang et al., 1996) and are covered by
Pz
Pzgw
Mzgr gn1 Quaternary deposits or cut by the Tashkorgan
Baoz

0 20 km Q2 Kongur Shan normal fault to the east. The Tashkorgan alka-

Ko 7530
iya Fa

line complex consists of three rock types with

ng

Geologic Units:
progressively younger ages (Xu et al., 1996;
ur
ult

Q1 Mzgr
Young Quaternary deposits
Sh

Figure 6
Pzgw Qm Zhang et al., 1996): a medium-grained diopside
an
Quaternary Units

Qm Quaternary glacial deposits AR 6/19/00-11

D''' syenite (Esy), (K-feldspar 40Ar/39Ar age: 52.0
Q2 Old Quaternary deposits
Fault

Pzsch1
± 0.7 Ma), a coarse-grained granite that intrudes
Qcgl Pliocene-Quaternary conglomerates Pzgn1
Mzgr
the syenite (K-feldspar 40Ar/39Ar age: 18.2 Ma),
QL Quaternary loess A

D''
and a biotite granite (Mgr) exposed along the
Qss Quaternary sandstones Q1
Mgr
northern end of the complex (biotite 40Ar/39Ar
Kss1 Cretaceous sandstone Pzsch3 7509 Qm age: 11.45 ± 0.30 Ma; K-feldspar 40Ar/39Ar age:
Sedimentary and Volcanic Units

Muztaghata B'
Kss2 Cretaceous red beds Q2
A'
11.58 ± 0.76 Ma). Geochemical analyses from
B
Jss Jurassic sandstone
Qm all members of the Tashkorgan igneous com-
JL Jurassic limestone plex suggest a high contribution of crustal mate-
Mzgr
JL Jurassic limestone rial with Sr concentrations of 1000–3500 ppm
PL Permian limestone (Zhang et al., 1996; Table DR12), and 87Sr/86Sr
Psl Permian slates ratios of 0.7094–0.7100 (Zhang et al., 1996).
t
aul

Ov Ordovician volcanics Mgr

Q1 38°
nF

C'
Pzgn1
ma

Dgw Devonian metagraywacke Metamorphic Rocks

Tah

Pzsch1 Pzsch3 Pzsch3S

Paleozoic high grade schists hen
Metasedimentary Units

-ti F
Pzgw Paleozoic metagraywacke Pzsch5 ault Hanging Wall of Kongur Shan Normal Fault
D' Torba
Pzsch2 Paleozoic schists shi Fa
ult
The hanging wall of the Kongur Shan fault
Pzgn1 Paleozoic quartzofeldspathic gneisses
C
consists of two metasedimentary units separated
Q1
Pzsch5
Pzgn2 Paleozoic calcareous biotite gneisses
D
by the Baoziya thrust (Figs. 3 and 5) (Robinson
Tash

T as

et al., 2004): the structurally higher Pzsch1, and

hk

Pzgn3 Paleozoic mafic gneisses and schists Pzsch4 Qss

korg

or

Esy
the structurally lower Pzgw. Pzsch1 is composed of
ga

Wu

Pzsch3 Paleozoic schists and marbles Qm Q2

nV
an fa

qa

pelitic schist, quartzite, and minor amphibolite.

all

Pzsch4 Paleozoic schists

Va
ey

Q2
lle
ult

Pzsch5 Paleozoic schist (low grade) Typical mineral assemblages in the pelitic units
y

Qm Mzgr
QL
Plutonic Units

Mgr Miocene granite To

yield upper amphibolite facies assemblages of
rb quartz + plagioclase + biotite + garnet ± musco-
Esy
a sh
G'
Eocene syenite
iF
Mzgr Triassic-Early Jurassic granites
au
lt
vite ± sillimanite ± K-feldspar. Previous studies
QL
along the northern portion of the hanging wall
Map Symbols: PL
have shown that the unit is part of the Late Tri-
Kongur Shan normal fault Approximate or inferred

Normal faults of the Kongur

fault contact
Psl assic medium- to high-grade, high-temperature
Buried fault contact JL1
Shan extensional system
? ?
and low-pressure metamorphic terrane, which
Normal faults not associated with Lakes and rivers Qm
the Kongur Shan extensional system Glaciers Kss2 developed during Triassic-Jurassic regional
Strike-slip fault JL1 JL2
Sample locality and Qm Psl igneous activity (Robinson et al., 2004). While
Thrust fault AR 6/19/00-11
station number Figure 6 the northern portion of the hanging-wall

Figure 3. Geologic map of the Kongur Shan extensional system. For cross sections, see Fig-
2
ure 6. GSA Data Repository item 2007134, Tables DR1–
DR5 and Figures DR1–DR5, is available on the Web
at http://www.geosociety.org/pubs/ft2007.htm. Re-
quests may also be sent to editing@geosociety.org.

Geological Society of America Bulletin, July/August 2007 885

 Robinson et al.

-10

-10
-5
metamorphic sequence exhibits a distinct east-

-5
8

0
B'

D'''
southeastward increase in metamorphic grade,

Pzgw
the southern portion is uniformly amphibolite

Q2
facies. Pzgw is composed of light-green quartz-

Kongur Shan Fault

Q1
rich chloritic and calcareous schists, with minor

N 40.1° W
layers of marble and dark-gray schist. Typical

Pzgn1
Mzgr

?
schist assemblages include quartz + white mica
+ calcite ± chlorite, indicating greenschist facies

Mzgr
metamorphic conditions.

D''
?
Footwall of the Kongur Shan and Tahman
Normal Faults

Figure 6. Geologic cross sections from Figure 3. (A) Cross section along A–A′. (B) Cross section along B–B′. (C) Cross section along C–C′.
High-grade metamorphic rocks in the foot-

?
wall of the Kongur Shan and Tahman normal

Pzsch3
faults are composed of two main units: a struc-

?
turally lower unit of quartzofeldspathic gneisses
Kongur Shan Fault

Mzgr
Q2
(Pzgn1) and a structurally higher metasedimen-
tary unit (Pzsch3) (Figs. 3, 5, and 6). The gneiss
unit exhibits well-developed compositional

N 5.6° E

Mzgr
banding (Fig. 7A) with assemblages of quartz

?
Pz gn
+ plagioclase + biotite ± hornblende ± garnet
Mgr

± K-feldspar (Fig. 8A). The overlying Pzsch3 is

?
tens of meters thick along the western flank of
N 83.9° E

the Muztaghata massif, but it is at least 10 km

?

thick south of Muztaghata (Figs. 5 and 6C).

Pzsch3

Pzsch3 is composed of interlayered pelitic schist,

Q2

mafic schist, marble, and calc-silicate. Pelitic

Pzgn1
B

schists in the footwall of the Kongur Shan and

8

-5

-10

?
B

Elevation (km)
Tahman normal faults preserve primary meta-
morphic assemblages of quartz + biotite + kya-
-10
-5
8

nite + garnet + plagioclase + K-feldspar within

A'

the structurally deepest levels (Fig. 8B). Trace

fibrolitic sillimanite is often observed in kyanite-
Shen-ti Fault

bearing samples, but it becomes more abundant

Pzsch3
at higher structural levels along with the appear-

?
?

D'

Torbashi Fault

ance of muscovite. Migmatitic textures are

Mzgr

common in the basal part of the pelitic schists,

N 3.6° E

indicating widespread partial melting through

4
ch
Pzsch5

Pz s
muscovite dehydration (Fig. 7B). Thermobaro- ?
?

metric calculations from electron microprobe

D

analyses of a garnet-kyanite-K-feldspar–bear-
8

-5

-10
D

ing schist yield peak metamorphic conditions of Elevation (km)

9–10 kbar, 700–750 °C (Table DR2, Fig. DR1,

?

see footnote 1) using GRAIL (Garnet-Rutile-

-10
-5
Pzsch3

Illmenite-Al2SiO5-Quartz) and GASP (Garnet-

C'

Plagioclase-Quartz-Al2SiO5) equilibria (Bohlen

Kongur Shan Fault

and Liotta, 1986; Koziol and Newton, 1988),

Kongur Shan Fault
Q2

(D) Cross section along D–D′–D″–D′′′.

Mzgr
z gn1

garnet-biotite Fe-Mg exchange, and THERMO-

P

Q1

CALC 2.7 (Powell et al., 1998). The calculated

Pzgn1
Pzsch3

peak metamorphic conditions are consistent

?
Eastern Pamir Detachment

with the presence of metamorphic K-feldspar,

?

kyanite, and migmatitic textures. As kyanite

Tahman Fault
Pzsch1

is the aluminosilicate phase coexisting with

K-feldspar in many samples, partial melting
8 Tashkorgan Fault
Q2
N 74.9° E

appears to have begun during peak metamor-

?
Pzsch3

N 42° E

phic conditions rather than during exhumation.

?

?
Mgr

Hanging Wall of Tashkorgan Normal Fault

C
A

Along the east-flowing portion of the

5

-5

-10
8

-5

-10

C
A

Elevation (km) Elevation (km)

Tashkorgan River, upper amphibolite facies

886 Geological Society of America Bulletin, July/August 2007

 Cenozoic evolution of the eastern Pamir

Southeast
A B

South North
C D

Figure 7. Gneisses of (A) Pzgn1 showing typical layering between the quartzofeldspathic gneisses and biotite-hornblende–rich gneisses in
the core of the Muztaghata massif. (B) Pzsch3 showing migmatitic textures in the high-grade pelitic schists. Lens cap for scale. (C) Looking
east at south-dipping Pzsch3 at the southern tip of the Kongur Shan normal fault along the Tashkorgan River. (D) Looking west at Pzsch3
immediately north of the Shen-ti fault. Asymmetric boudinage of a quartz vein shows top-south sense of shear. Rock hammer for scale.
Qtz—quartz; Chl—chlorite; Act—actinolite; Plg—plagioclase; Ky—kyanite; Grt—garnet.

A C East
B
S Ky

Plg
Figure 8. Photomicrographs of
Ky
metamorphic units along the south-
ern Kongur Shan extensional sys-
tem. (A) Mylonitic schist of Pzsch3
Grt
from the western flank of Muz-
taghata with top-to-the-west sense
1 mm
Qtz 1 mm of shear. Outcrop is ~200 m high.
(B) Garnet-kyanite-K-feldspar
C South
D assemblage of Pzsch3 from sample
AR7/7/00-7a. (C) Mylonitic S-C
fabric in quartz-rich schist from
Chl
C
Chl Pzsch4 showing top-to-the-south
S Act sense of shear. Relief is ~800 m.
(D) Mafic schist from Pzsch4 with
actinolite crosscutting foliation.
Act
Qtz

Qtz
1 mm 1mm 1 mm

Geological Society of America Bulletin, July/August 2007 887


schists and quartzofeldspathic gneisses from the
footwall of the Kongur Shan normal fault are
juxtaposed against structurally higher medium-
grade schists of Pzsch4 (Figs. 3 and 5). Pzsch4 is
composed of interlayered quartz-rich, calcare-
ous, and mafic schists. Mafic schists consist of
chlorite + quartz + epidote ± actinolite ± calcite
(Figs. 8C and 8D), indicating greenschist facies
metamorphic conditions. Actinolite porphyrob-
lasts crosscut the foliation in several samples,
suggesting postdeformation peak metamorphic
conditions (Fig. 8D). Pzsch4 continues to the
southeast, where it is exposed along the eastern
side of the Wuqa Valley (Figs. 3 and 4).
Structurally above the greenschist facies schists
of Pzsch4 and exposed between the Tashkorgan
Valley and Wuqa Valley are high-grade schists of
Pzsch5 (Figs. 5–6). Pzsch5 consists of pelitic schists
and quartzofeldspathic gneisses intruded by
small bodies of undeformed to mylonitic gran-
ite. Metamorphic grade in the unit decreases to
the south from upper amphibolite facies (garnet
± sillimanite ± kyanite ± K-feldspar) to green-
schist (garnet + biotite) facies.
STRUCTURAL GEOLOGY OF THE
SOUTHERN KONGUR SHAN
Kongur Shan Extensional System
From north to south, the 250-km-long Kongur
Shan extensional system consists of the Muji
fault, Kongur Shan normal fault, Tahman nor-
mal fault, and Tashkorgan normal fault (Figs. 2
and 3) (Robinson et al., 2004). The geology of
the northern portion of the extensional system
including the Muji fault and the northern seg-
ment of the Kongur normal fault has been previ-
ously described by Arnaud et al. (1993), Brunel
et al. (1994), and Robinson et al. (2004). Here,
we describe the structures along the central and
southern segments of the Kongur Shan exten-
sional system. Together with previously pub-
lished work, our new results provide a coherent
view of the overall structural framework of the
Kongur Shan extensional system.
Southern Kongur Shan Normal Fault
The Kongur Shan normal fault is generally
north-striking south of the Kongur Shan massif,
but it becomes highly convex to the west along
the western margin of the Muztaghata massif
(Figs. 3 and 5). While the location of the fault
is easy to determine in the field within a few
meters and by satellite images, the fault trace
itself is commonly buried under glacial deposits
and other Quaternary debris and was not directly
observed. Along the western flank of the Muz-
taghata massif, a 5–10-m-thick zone of chloritic
breccia is present, immediately beneath which
888 Geological Society of America Bulletin, July/August 2007
Robinson et al.
are mylonitic schists and high-grade gneisses
with west-plunging stretching mineral linea-
tions. Kinematic indicators such as S-C fabrics
and rotated quartz pods show top-to-the-west
sense of shear (Figs. 7A). Mylonitic fabrics die
out ~200 m below the fault surface at deeper
structural levels and are also absent along the
southern flank of Muztaghata. Foliations in the
schists and gneisses and compositional band-
ing in the gneisses are generally parallel to the
slope of the western mountain flank of the Muz-
taghata massif, dipping ~30° west and increas-
ing to ~60° southwest along the southern flank
of the massif (Figs. 5, 6A, and 6B). The 30° dip
of the mylonitic foliations along the western
Muztaghata flank is interpreted to be subparallel
to the dip of the Kongur Shan normal fault.
Foliations within the Muztaghata massif
define a west-northwest–trending antiform, with
a gently dipping northern limb and a moderately
dipping southern limb (Figs. 5 and 6D). Field
observations in the eastern core of the massif
indicate the foliation rotates to the east from
west-dipping to subhorizontal (Figs. 6A and
6B). This is notably different from the structure
of the Kongur Shan gneiss dome to the north
(Brunel et al., 1994; Robinson et al., 2004),
where foliations define a north-northwest–trend-
ing, north-plunging dome, with the axis subpar-
allel to the strike of the Kongur Shan normal
fault and steeply (>70°) east-dipping foliations
along the eastern limb of the dome.
The Kongur Shan normal fault continues for
~20 km south of the Tagamansu Valley before
terminating at the Tashkorgan River (Fig. 5).
Along this segment of the fault, the internal
structure of the footwall changes significantly:
Footwall foliations and compositional layering
change from orientations subparallel to the strike
of the Kongur Shan fault to highly oblique. At the
southern tip of the Kongur Shan fault, foliations
are perpendicular to the fault, striking east-west
and dipping moderately to the south (Figs. 5 and
7C). Mylonitic fabrics and stretching lineations
are abundant in the southern exposures of Pzsch3,
with generally south-plunging orientations and
top-to-the-south sense of shear (Fig. 7D). The
contact between footwall units Pzgn and Pzsch3
is cut at a high angle by the Kongur Shan nor-
mal fault and continues across the fault into the
footwall of the Tahman normal fault with little
or no observable offset (Fig. 5). This is the only
place along the trace of the Kongur Shan normal
fault where rock units can be mapped across the
structure, supporting the observation that the
fault terminates in this area.
Tahman Normal Fault
The 20-km-long Tahman normal fault
cuts obliquely across the northern end of the
Tashkorgan Valley, strikes N25°E along the
northern two-thirds of the fault and N5°E along
the southern third of the fault, and dips steeply
to the west-northwest (Fig. 5). Quaternary
activity along the fault is demonstrated by well-
preserved 1–2-m-high fault scarps that cut Qua-
ternary alluvial deposits. Maximum local relief
in the footwall of ~500 m suggests low mag-
nitudes of offset across the fault. The Tahman
fault is interpreted to be a transfer structure,
linking the west-dipping Kongur Shan normal
fault and the east-dipping Tashkorgan normal
fault. As with the southern end of the Kongur
Shan normal fault, footwall structures and fab-
rics in the high-grade schists and gneisses are
truncated at a high angle by the Tahman fault
and show no relation to the active structure.
Mineral lineations and shear sense indicators in
the metamorphic rocks trend generally north-
south and are top-to-the-south, respectively.
Muztaghata Dome
North-south cross sections through the foot-
wall of the Kongur Shan and the Tahman nor-
mal faults (Figs. 8C and 8D) show that the high-
grade schists and gneisses of Pzsch3 and Pzgn1
define a large upright to slightly overturned
south-vergent antiform with the west-plung-
ing domal structure of the Muztaghata massif
described above as defining the hinge of the
structure. Since the active Kongur Shan and Tah-
man normal faults clearly crosscut the southern
limb of this antiform, the domal structure of the
Muztaghata massif is interpreted to largely pre-
date the initiation of east-west extension along
the Kongur Shan extensional system.
Tashkorgan Normal Fault
The 75-km-long Tashkorgan normal fault is
the southernmost segment of the Kongur Shan
extensional system. The fault strikes north to
north-northwest, dips steeply to the east, and
bounds the western side of the Tashkorgan
Valley (Figs. 3 and 4). Expression of the fault
is variable along strike: North of the junction
with the Tahman normal fault along the gently
sloping western side of the valley, interpretation
of satellite images shows that the Tashkorgan
normal fault consists of several discontinuous
faults 5–10 km in length (Fig. 3). South of the
juncture with the Tahman fault, the Tashkorgan
normal fault becomes a prominent morphologic
feature, bounding the western side of the valley
(Figs. 3 and 4), with well-developed triangular
facets along the eastern flank of the mountain
range. The Tashkorgan fault is well defined for
another 30 km to the south, while topographic
relief of the footwall decreases. At its southern
tip, the fault branches into two smaller faults,
which form scarps with ~100 m of offset within

a loess-covered region (Ql) on the western side
of the Tashkorgan Valley (Fig. 4). Offset of the
loess dies out to the south over a distance of
~10 km, and the Tashkorgan fault is interpreted
to terminate without linking to structures further
south. An 40Ar/39Ar K-feldspar cooling age from
the syenite in the northern footwall of the Tash-
korgan normal fault of 54 Ma (Xu et al., 1996)
indicates temperatures below 150 °C since the
early Eocene, indicating ≤5 km of exhuma-
tion due to late Cenozoic east-west extension
(assuming a geothermal gradient of 30 °C/km).
Kalaqigu Fault
Directly along strike south of the southern
termination of the Tashkorgan fault, a north-
south–striking subvertical gouge zone, 40–50 m
thick, called the Kalaqigu fault, separates beige
to gray Jurassic marbles to the west from Per-
mian slates to the east (Fig. 4). No clear kine-
matic indicators were found within the gouge
zone, and no offset features were identified to
determine a sense of shear. While the Kalaqigu
fault is in the correct orientation to link the active
Kongur Shan extensional system to the north
and the Karakorum fault to the south, as has
been previously proposed (Arnaud et al., 1993;
Brunel et al., 1994; Ratschbacher et al., 1994;
Strecker et al., 1995; Murphy et al., 2000), our
field observations indicate that the fault has not
been recently active (Figs. 4A and 4D). Specifi-
cally, the fault trace cuts perpendicularly across
several narrow, deeply incised valleys without
offsetting them, and ridges above the valleys
where the gouge zone is exposed are capped by
Quaternary loess deposits that show no evidence
of tectonic disturbance.
Shen-ti Normal Fault
At the southern tip of the Kongur Shan nor-
mal fault, upper amphibolite facies schists and
gneisses in the footwall of the Kongur Shan
and Tashkorgan faults (Pzsch3 and Pzgn) are sepa-
rated from greenschist facies schists to the south
(Pzsch4) along the east-west–trending Shen-ti
fault (Fig. 5). The Shen-ti fault cuts obliquely
through the high-grade schists and gneisses in its
footwall juxtaposing the low-grade Pzsch4 against
high-grade pelitic schists (Pzsch3) to the west and
quartzofeldspathic gneisses (Pzgn) to the east.
While exposures of fault gouge were observed
along the trace of the fault, the fault surface that
bounds the hanging wall and the gouge zone was
never directly observed in the field. Subparallel
south-dipping foliations in the hanging wall and
footwall along most of the trace of the Shen-ti
fault are interpreted to indicate a southerly dip,
although to the east, footwall fabric orientations
Geological Society of America Bulletin, July/August 2007
Cenozoic evolution of the eastern Pamir
within the quartzofeldspathic gneisses are mod-
erately to highly oblique to the orientation of the
Shen-ti fault (Fig. 5). Kinematic indicators in
the southern exposures of the high-grade schists
show top-to-the-south sense of ductile shear
adjacent to the Shen-ti fault (Fig. 7D), while
mylonitic S-C fabrics within the calcareous and
mafic schists in the hanging wall of the Shen-ti
fault also show top-to-the-south sense of shear
(Fig. 8C). Top-to-the-south shear sense indica-
tors and the juxtaposition of high-grade rocks
in the footwall against lower-grade rocks in the
hanging wall all suggest that the Shen-ti fault is
a top-to-the-south normal fault. The east-west
change in footwall units and fabric orienta-
tions suggests that the brittle Shen-ti fault cuts
obliquely through the footwall, exposing struc-
turally deeper (and non-mylonitic) portions of
the footwall to the east (Fig. 5).
Torbashi Thrust
Immediately south of the Shen-ti fault, green-
schist facies calcareous and mafic schists of
Pzsch4 are juxtaposed against upper amphibolite
facies schists of Pzsch5, indicating the presence of
a fault we call the Torbashi fault (Figs. 3–4). The
contact between Pzsch4 and Pzsch5 is strikes east
subparallel to the Shen-ti fault before bending
to the southeast along the Wuqa Valley (Fig. 4).
Foliations in the low-grade footwall schists
and high-grade hanging-wall schists generally
follow the orientation of the contact, changing
from east-striking and south-dipping along the
Tashkorgan River to northwest-striking, south-
west-dipping along the Wuqa Valley. Although
we never directly observed the Torbashi fault in
the field, south- to southwest-dipping foliations
on either side of the inferred fault trace imply a
south- to southwest-dipping fault, with mylonitic
fabrics in the hanging wall showing top-to-the-
southwest shear sense (Fig. 4). Based on the
juxtaposition of high-grade schists structurally
overlying low-grade schists, we interpret the
Torbashi fault to be an originally northwest-dip-
ping, southwest-directed thrust, which has been
subsequently rotated to a southerly dip. Along
the eastern side of the Tashkorgan Valley, fab-
rics in the high-grade schists in the hanging wall
of the Torbashi thrust change to the south from
south-southwest–dipping to northeast-dipping,
with east-west– to southwest-northeast–trend-
ing mineral lineations (Fig. 4). Further south,
northeast-dipping medium-grade schists (Pzsch5)
lie structurally above northeast-dipping Permian
limestones and slates. As to the north, the juxta-
position of higher-grade rocks over lower-grade
rocks suggests that the contact is a thrust fault
that dips to the northeast. We interpret this con-
tact to be the southern exposure of the Torbashi
thrust, where the high-grade rocks of Pzsch5 form
a large synformal klippe.
GEOCHRONOLOGY AND
THERMOCHRONOLOGY
In order to better understand the tectonic evo-
lution of the southern Kongur Shan region, we
used several techniques to determine the timing
of igneous activity, high-grade metamorphism,
and exhumation of the various units. These
included U-Pb isotopic analyses of individual
zircon grains and in situ Th-Pb isotopic analyses
of individual monazite grains using the UCLA
CAMECA IMS 1270 ion microprobe, and
40
Ar/39Ar step-heating analyses of biotite and
muscovite separates from numerous samples.
Analytical procedures for the various techniques
are the same as those described in Robinson et
al. (2004) and references therein.
U-Pb Zircon Geochronology
Cenozoic Biotite Granite
Twenty-seven zircon grains from three
samples (AY8/21/99–8, AY8/23/99–3, and
AR4/27/00–10) were analyzed from the large
biotite granite in the hanging wall of the Kongur
Shan normal fault west of the Muztaghata mas-
sif (Mgr; Fig. 5; Table DR3 [see footnote 2]).
Zircon analyses yielded a spread in ages from
Late Proterozoic to late Miocene, with a strong
cluster of late Miocene ages (Figs. 9A and B).
Older ages generally came from the cores of zir-
con grains and are interpreted as inherited zircon
from the protolith. The seventeen youngest anal-
yses from the biotite granite yielded a weighted
mean 206Pb/238U age of 11.46 ± 0.71 Ma (mean
square of weighted deviates [MSWD] = 5.5),
while a strong cluster of 12 analyses yielded an
indistinguishable age of 11.68 ± 0.45 (MSWD
= 1.5) (Fig. 9B), which we interpret as the intru-
sion age of the granite.
Footwall Orthogneiss
Sample AR6/15/00–3 was collected from a
deformed biotite granite located in the footwall
of the Kongur Shan normal fault immediately
north of the Muztaghata massif (Fig. 5). Four-
teen analyses were made from 11 zircon grains
(Fig. 9C). One analysis (11–2) yielded a nega-
tive U-Pb 206Pb*/238U age with large errors and is
not discussed further. The other analyses yielded
Late Cretaceous to early Miocene 206Pb*/238U
ages (105–22 Ma), with nine analyses yielding
late Eocene to early Miocene 206Pb*/238U ages
of 22–44 Ma. The nine youngest 204Pb corrected
ages were slightly reversely discordant, and two
analyses yielded negative 207Pb*/235U ages (anal-
yses 8–1 and 1–1; Table DR3).
889
 Robinson et al.

0.150 0.004 0.025

25 Ma

0.125
A 800 Ma B C 140 Ma

0.020
0.003 120 Ma

0.100
600 Ma 100 Ma
Pb/238U

Pb/238U

Pb/ U
15 Ma 0.015

238
500 Ma
0.075 0.002 80 Ma
206

206

206
400 Ma
0.010 60 Ma
0.050 300 Ma
0.001 40 Ma
200 Ma 5 Ma 0.005
0.025
20 Ma

0.000 0.0 0.000

0.00 0.25 0.50 0.75 1.00 1.25 1.50 -0.02 0.00 0.02 0.04 0.06 0.08 0.00
0. 0.05 0.10 10.5 0.01 .0.10 1.00
207 235 207 235 207 235
Pb/ U Pb/ U Pb/ U Th/U

Figure 9. Concordia plots of U-Pb zircon analyses. (A) All zircon analyses from biotite granite Mgr (samples AR4/27/00–10, AY8/21/99–8, and
AY8/23/99–3). (B) Miocene analyses from biotite granite Mgr. Analyses in black yield a weighted mean age of 11.46 ± 0.71 Ma (mean square
of weighted deviates [MSWD] = 5.5). Analyses in bold yield a weighted mean age of 11.68 ± 0.45 Ma (MSWD = 1.5). (C) Zircon analyses from
footwall mylonitic granite (sample AR 6/15/00–3). The Th/U ratios for each analysis are plotted to the right of the concordia plot.

Zircon grains from AR6/15/00–3 have a
mottled internal zoning pattern in cathodolu-
minescence images, and several grains contain
small patches of truncated oscillatory zoning
(Fig. DR4, see footnote 2). Analyses show a
direct correlation of age with internal zoning:
Spot analyses of overlapping portions of oscil-
latory zoned zircon yield Late Cretaceous ages
while analyses exclusively from mottled por-
tions of zircon yield Cenozoic ages. The mottled
internal zoning is interpreted to be zircon grown
or recrystallized during high-grade metamor-
phism (Hoskin and Black, 2000), while oscil-
latory zoned patches represent preserved por-
tions of premetamorphic igneous zircon. This
interpretation is supported by low Th/U ratios
of the analyses (Th/U < 0.10; Fig. 9C) typical
of metamorphic zircon (e.g., Hoskin and Black,
2000; Ding et al., 2001; Mojzsis and Harrison,
2002). The timing of metamorphism is tenta-
tively placed as Oligocene–early Miocene based
on the six youngest analyses (33–22 Ma).
Th-Pb Monazite Geochronology
in garnet yielded ages from Triassic (228
± 10 Ma) to late Oligocene (24.1 ± 1.7 Ma).
Matrix analyses all yielded Miocene ages from
15.9 ± 0.3 Ma to 7.5 ± 0.2 Ma. Sixty-eight mon-
azite grains were analyzed from four samples of
garnet–kyanite–K-feldspar–bearing rocks at the
southern end of the Kongur Shan normal fault
footwall and the footwall of the Tahman normal
fault (AR 9/6/03–3, AR 9/6/03–5, AR 7/7/00–
7a, and AR 7/8/00–5b) (Fig. 5); thirty-seven of
these occurred as inclusions in garnet, and thirty-
one were from the matrix of the sample. Analy-
ses from monazite inclusions in garnet (almost
entirely from samples AR 7/7/00/7a and AR
7/8/00–5b) yielded ages from Jurassic to middle
Miocene. With one exception, matrix monazite
analyses yielded Oligocene to Miocene ages,
with most falling between 10 and 16 Ma.
The cumulative distribution of all analyses
from monazite inclusions in garnet from Pzsch3
yielded a wide spread in ages from Mesozoic to
Miocene (Fig. 10). Mesozoic ages ranged from
Middle Triassic to Late Cretaceous, and most
ages fell between the Middle Jurassic and Early
Cretaceous. While the Mesozoic ages indicate
ages, (2) a narrow middle Miocene peak at
ca. 14 Ma, which overlaps with the second peak
in inclusion ages, and (3) a broad peak in the late
Miocene from ca. 11 to 8 Ma that is not present
in the distribution of garnet inclusion ages.
Based on the distribution of the inclusion
and matrix ages, together with the high grade of
metamorphism and migmatitic textures, we inter-
pret the monazite analyses as follows: Cenozoic
burial and prograde metamorphism of the schists
and gneisses to mid-crustal depths (30–35 km
based on thermobarometry) occurred during
the Oligocene–early Miocene, resulting in the
broad peak in ages in both the matrix and inclu-
sion analyses (as well as the metamorphic zir-
con ages from AR6/15/00–3). Continued crustal
thickening and/or thermal relaxation during the
Miocene led to partial melting of the schists at
ca. 14 Ma, resulting in the narrow peak in ages in
40
Inclusion ages (n = 22)
35 Matrix ages (n = 63)
Relative Probability

30
Seven samples from the high-grade pelitic a polymetamorphic history for Pzsch3, the wide
schists of Pzsch3 were dated by in situ ion-micro- spread makes it difficult to determine the tim- 25

probe Th-Pb analyses of monazite from two ing of this earlier metamorphic event (or events). 20
different regions: the western flank of Muz- Cenozoic ages from monazite inclusions in gar- 15
taghata, and the southern footwall of the Kongur net ranged from Eocene to middle Miocene with 10
Shan and footwall of the Tahman fault (Fig. 5) two main peaks: (1) a broad Oligocene to early
5 0 10 20 30 40 50
(Table DR4, see footnote 2). Miocene peak from 30 to 20 Ma, and (2) a nar-
0
Thirty-five grains were analyzed from three row middle Miocene peak at ca. 14 Ma. Matrix 0 25 50 75 100 125 150 175 200 225 250
samples of mylonitic high-grade schist from the monazite analyses yielded only one pre-Oligo- Age (Ma)
western flank of Muztaghata (AR 8/31/03–3, cene age of 119.2 ± 5.7 Ma (Fig. 10). The cumu-
AR 8/31/03–5, and AR 9/1/03–1a) (Fig. 5), lative distribution of Cenozoic matrix ages show Figure 10. Cumulative probability plots
seven of which occurred as inclusions in garnet three main peaks: (1) a small broad late Oligo- of all monazite Th-Pb ages from Pzsch3
and twenty-eight of which were from the matrix cene to early Miocene peak (ca. 20–25 Ma), grouped into inclusion ages (black) and
of the sample. Analyses of monazite inclusions which overlaps with the first peak in inclusion matrix ages (gray).

890 Geological Society of America Bulletin, July/August 2007


both matrix and inclusion analyses. Since mona-
zite in the matrix of the samples would have
been susceptible to dissolution/reprecipitation
during anatexis, much of the monazite grown
during the Oligocene–early Miocene (as well
as surviving Mesozoic grains) would have been
reset or destroyed. Recrystallization processes
during exhumation-related deformation resulted
in continued monazite growth, producing late
Miocene matrix ages from 8 to 10 Ma.
40
Ar/39Ar Thermochronology
Hanging Wall of Kongur Shan Normal Fault
A sample from Pzsch1 in the hanging wall of
the Kongur Shan normal fault (AR6/19/00–11a)
northwest of the Muztaghata massif (Fig. 3)
yielded a Late Jurassic biotite age of 159 ± 4 Ma
(Table DR5, Fig. DR5, see footnote 2), which
overlaps with previously published Late Juras-
sic ages from the northern portion of the hang-
ing wall (Robinson et al., 2004). As discussed in
Robinson et al. (2004), we interpret these ages
to reflect postmetamorphic cooling through the
biotite closure temperature.
Sample AR4/27/00–10 is from the biotite
granite at the northern end of the Tashkorgan
alkaline complex (Mgr), southwest of the Muz-
taghata massif (Fig. 5). This sample yielded a
biotite 40Ar/39Ar weighted mean spectra age of
11.2 ± 0.2 Ma, which is in agreement with the
biotite and K-feldspar cooling ages from Xu et
al. (1996) and overlaps the U-Pb zircon age of
11.68 ± 0.45 Ma. The overlap in age between
the zircon U-Pb analyses and the biotite and
K-feldspar 40Ar/39Ar analyses shows that the
granite was emplaced at shallow depths and
cooled rapidly in the middle Miocene.
Footwall of Kongur Shan Normal Fault
Twenty-two biotite analyses and five musco-
vite analyses were obtained from samples taken
from the footwall of the Kongur Shan normal
fault (Fig. 4). Most analyses yielded relatively
flat age spectra with agreement within error
between total gas, weighted mean age spectra,
and inverse isochron ages. Several analyses
yielded slightly disturbed, but interpretable age
spectra (e.g., AR7/4/00-1; Fig. DR5Z), while one
analysis yielded a highly disturbed uninterpreta-
ble age spectra (AR6/30/00-10b; Fig. DR5W).
North of the Muztaghata massif, muscovite
and biotite analyses from footwall gneisses
yielded ages of 6.2–2.7 Ma (Fig. 5). The data
from this area show two trends: (1) ages increase
slightly to the east with distance from the fault,
and (2) ages increase to the south along the trace
of the fault, where muscovite ages increase from
2.9 Ma to 6.2 Ma over a distance of 10.5 km
along the northern flank of Muztaghata.
Geological Society of America Bulletin, July/August 2007
Cenozoic evolution of the eastern Pamir
South of the Muztaghata massif, biotite and
muscovite analyses from the footwall of the
Kongur Shan normal fault are characterized by
remarkably consistent ages of 7.5–8.5 Ma over
a broad area (Fig. 5). One biotite age of 10.1
± 0.2 Ma at the mouth of Tugenmansu Valley is
anomalously old, which we interpret to be due
to excess 40Ar. While there was only one musco-
vite analysis along the southern footwall of the
Kongur Shan normal fault (AR7/4/00–5), the
age overlaps with nearby biotite ages, indicating
rapid cooling in the late Miocene. These ages are
in general agreement with two phlogopite K-Ar
ages and a biotite K-Ar age reported by Xu et al.
(1996) of 9.0 Ma, 6.7 Ma, and 6.7 Ma, respec-
tively, from the Tugenmansu Valley. However,
since no uncertainties were reported, it is difficult
to assess the discrepancy of their younger ages.
The consistency of the ages from the footwall is
notable because there are no obvious relation-
ships between the biotite ages and proximity to
the Kongur Shan fault, structural trends, or rock
type. This is strikingly different from the pattern
of increasing age with distance from the Kongur
Shan fault observed in the footwall immediately
north of the Muztaghata massif and along the
northern flank of the Kongur Shan massif (Arn-
aud et al., 1993; Robinson et al., 2004).
Footwall of Tahman Normal Fault
Four biotite and two muscovite analyses were
obtained from schists of Pzsch3 within the foot-
wall of the Tahman fault (Fig. 5). Analyses from
the Tahman fault footwall generally yielded ages
of 8–10 Ma, which overlap with footwall ages
from the Kongur Shan fault directly to the east.
One analyses from the footwall of the Tahman
normal fault yielded a biotite age of 13.7 Ma,
which we interpret to be the result of excess
40
Ar because it overlaps with the timing of peak
metamorphic conditions for the footwall deter-
mined from monazite Th-Pb geochronology.
Hanging Wall of Torbashi Fault
Three biotite analyses and one muscovite
analysis were obtained from high-grade schists
of Pzsch5 at the northern end of the Torbashi fault
hanging wall (Fig. 5; Figs. DR5AJ–DR5AM
[see footnote 2]). Biotite analyses yielded
late Early Cretaceous total gas ages between
107 ± 2 Ma and 124 ± 4 Ma, with somewhat
disturbed age spectra. The muscovite analy-
sis yielded an erratic age spectra with a total
gas age of 105 Ma. Since this is significantly
younger than the biotite age of 120 Ma from the
same sample, it suggests that either the biotite
is contaminated with excess 40Ar, or the mus-
covite has been subjected to alteration. Despite
the relatively poor quality of the data, these
results indicate that the high-grade schists in the
hanging wall of the Torbashi thrust fault were
cooled below the muscovite and biotite closure
temperatures by the Late Cretaceous.
DISCUSSION
Field investigations of the southern Kongur
Shan extensional system and related geochrono-
logic and thermochronologic analyses provide
insight into the Cenozoic tectonic evolution and
pre-Cenozoic tectonic configuration of the Pamir
region. Next, we discuss the geologic implica-
tions of our observations and analytical results.
Kongur Shan Extensional System
Cooling-age patterns and geologic relations
across the active faults at the southern end of the
Kongur Shan extensional system provide impor-
tant information on the magnitude and timing of
extension along the southern end of the Kongur
Shan extensional system. While cooling ages in
the footwall of the southern Kongur Shan fault
and Tahman fault are uniformly late Miocene,
several observations demonstrate that this cool-
ing event was unrelated to the active east-west
extension: (1) late Miocene cooling ages show
no variation relative to the active structures, with
little or no offset in ages across the Kongur Shan
fault and no variation in cooling ages up to at
least 15 km east of the active Kongur Shan fault
(Fig. 5). (2) Kinematic indicators in the southern
exposures of Pzsch3, which were at ~35 km in the
middle Miocene show top-to-the-south sense
of shear, perpendicular to the active east-west
extension. (3) The sharp break in cooling ages
from Miocene to Cretaceous and between upper
amphibolite and greenschist facies schists occurs
across the south-dipping Shen-ti fault. (4) Geo-
logic mapping shows no detectable offset of the
contact between the high-grade schists (Pzsch3)
and quartzofeldspathic gneisses (Pzgn1) across the
Kongur Shan fault and no observable continua-
tion of the fault south of the Tashkorgan River.
Cooling ages along the northern flank of
the Muztaghata massif are more difficult to
uniquely interpret. The decrease in cooling ages
across the massif from ca. 7.5 Ma to 6.2 Ma and
younger could represent a break between pre–
east-west extension ages to the south and syn-
extensional ages to the north. This would date
initiation of extension along the southern end of
the Kongur Shan fault to between 6.2 and.5 Ma,
similar to the documented initiation of extension
at the northern end of the fault (Robinson et al.,
2004). Alternatively, the 6.5 Ma muscovite age
could represent cooling due to re-equilibration
of high thermal gradients from rapid middle
to late Miocene exhumation of the schists and
gneisses immediately to the south. If so, ages
891

farther north would suggest initiation of exten-
sion since ca. 5 Ma, indicating southward prop-
agation of east-west extension.
Regardless of these two interpretations, cool-
ing ages indicate that the southernmost change
from pre– to syn–east-west extension ages occurs
along the western flank of the Muztaghata mas-
sif, where we interpreted the fault to dip ~30° to
the west. Rapid middle to late Miocene exhu-
mation documented immediately to the south
would have resulted in advection of isotherms
and a high local geothermal gradient along the
Muztaghata massif. Using a reasonable elevated
geothermal gradient of 40 °C/km, rocks along
the western flank of the massif at or near the bio-
tite and muscovite closure temperatures would
record ~9 km of exhumation, yielding ~20 km
of east-west extension, with higher geothermal
gradients (or synextensional ages located farther
north) yielding less extension.
Eocene K-feldspar cooling ages demonstrate
<5 km of exhumation along the steeply east-
dipping Tashkorgan, which results in <3 km of
extension for a fault dipping 60°. This, along
with the apparent termination of the fault to
the south within thick Quaternary(?) deposits,
shows low magnitudes of extension along the
Tashkorgan fault and a southward decrease in
the magnitude of extension along the southern
half of the Kongur Shan extensional system.
To address along-strike variation of east-west
extension along the entire Kongur Shan exten-
sional system, it is necessary to reexamine one of
our calculations from Robinson et al. (2004). One
of the two points where magnitudes of extension
were calculated along the Kongur Shan normal
fault was ~35 km north of the Kongur Shan mas-
sif, where K-feldspar Multiple Diffusion Domain
modeling showed rapid cooling from ~250 °C
at 7–8 Ma. The segment of the fault where the
sample was located dips ~30° to the southwest,
and the calculated 17 km of extension was based
on ~10 km of exhumation along a 30°-dipping
plane. However, since extension is east-west
rather than northeast-southwest, exhumation
would be along a path plunging ~20° west,
yielding ~30 km of east-west extension. While
we did not obtain detailed kinematic data in the
field, analysis of satellite images indicates right-
lateral deflection of geomorphic features along
this fault segment, which is consistent with the
above interpretation. It is important to note that
the estimated magnitude of extension is still less
than the ~35 km of extension calculated along
the Kongur Shan massif farther to the south.
East-West Extension in the Pamir
Calculated magnitudes of extension along the
Kongur Shan fault system from north to south
are as follows: (1) ~30 km at the northern end
892 Geological Society of America Bulletin, July/August 2007
Robinson et al.
of the Kongur Shan fault, (2) ~35 km along the
Kongur Shan massif, (3) ~20 km (or less) along
the Muztaghata massif, and (4) <3 km along
the Tashkorgan fault. While uncertainties in
the individual calculations are significant, due
primarily to uncertainties in subsurface fault
geometry (e.g., if the fault shallows at depth)
and geothermal gradients, the overall pattern of
decreasing magnitude of extension from north
to south is robust. These results are consistent
with east-west extension in the Pamir that was
primarily driven by oroclinal bending–radial
spreading of the Pamir salient during northward
overthrusting along the Main Pamir thrust.
One key observation is that our calculations
show the largest magnitude of extension located
along the Kongur Shan massif; this is incon-
sistent with a simple oroclinal bending–radial
spreading model, which would predict a con-
stant southward decrease in the magnitude of
extension. This suggests that other factors play a
role in the evolution of the Kongur Shan exten-
sional system, such as synorogenic uplift asso-
ciated with the exhumation of the Kongur Shan
massif (e.g., Brunel et al., 1994).
Oligocene-Miocene Crustal Thickening and
Gneiss Dome Formation
Oligocene-Miocene Crustal Thickening
Monazite and zircon ages from high-grade
schists and gneisses along the southern foot-
wall of the Kongur Shan fault document crustal
thickening in the eastern Pamir during the Oli-
gocene to middle Miocene. The timing of crustal
thickening overlaps with the interpreted late Oli-
gocene initiation of slip along the Main Pamir
thrust as determined from sedimentologic, ther-
mochronologic, and paleomagnetic evidence in
the Tarim and Tajik Basins (e.g., Thomas et al.,
1994, 1996; Sobel and Dumitru, 1997; Burt-
man, 2000), as well as metamorphic ages from
near-ultrahigh-pressure crustal xenoliths from
the southeast Pamir (Ducea et al., 2003; Hacker
et al., 2005). Together, these studies point to a
major change in the dynamics of the Pamir in
the late Oligocene with the initiation of major
crustal thickening and shortening. Crustal thick-
ening also appears to have led to a period of
extensive Miocene middle- and lower-crustal
melting based on the presence of migmatitic
textures in the high-grade schists, lower-crustal
xenoliths that underwent partial melting (Hacker
et al., 2005), and numerous Miocene granitic
bodies in the Central Pamir (Fig. 11) (Xu et al.,
1996; Ducea et al., 2003; Schwab et al., 2004).
Miocene Gneiss Dome Formation
As discussed already, rapid late Miocene
exhumation of high-grade schists and gneisses
at the southern end of the Kongur Shan normal
fault predated the active east-west extension
and was instead accommodated in part by the
south-dipping Shen-ti fault. Geochronologic
and thermobarometric constraints indicate that
the schists were brought from 30 to 35 km depth
to within 10 km of Earth’s surface in ~6 m.y.,
which requires exhumation rates of 3–4 mm/
yr. We interpret this event to have resulted in
the formation of the west-northwest–trending
domal structure of the Muztaghata massif and
southern Kongur Shan fault footwall. Further,
several lines of evidence indicate that the Muz-
taghata antiform is the eastward continuation of
the Muskol and Sares antiforms of the Central
Pamir (Figs. 2 and 9), and that these features
have not been offset from the Qiangtang anti-
clinorium along the Karakorum fault. These
include: (1) Lithologies in the fault-bounded
core of Muskol and Sares domes are described
as granitic gneisses, kyanite-garnet–bearing
micaceous schists, and interlayered marble and
amphibolite, with local migmatization of the
high-grade schists (Pashkov and Dmitriyev,
1982; Peykre et al., 1982), identical to the high-
grade schists and gneisses in the footwall of the
Shen-ti fault (Pzsch3 and Pzgn1). While there are
differences compared to some of the lithologies
described by Schwab et al. (2004) (i.e., lack of
gabbros and preserved pillow basalts), these
appear to be primarily confined to the upper units
(and lower metamorphic grades) of their sche-
matic stratigraphic sections. This is consistent
with our observation that the Shen-ti fault cuts
“down-section” to the east, excising the shal-
lower levels of the domes. (2) Regional geologic
maps and interpretation of satellite images show
the contact of the southern margin of the Sares
antiform intersecting the Tashkorgan Valley
~10 km south of the westward projection of the
Shen-ti fault (Figs. 2 and 11) (Pashkov and Dmi-
triyev, 1982; Yin and Bian, 1992). (3) The axes
of the Sares and Muztaghata domes lie directly
along strike of each other (Fig. 11) (Pashkov and
Dmitriyev, 1982), and restoration of east-west
extension places the Muztaghata massif directly
beneath the Sares dome. (4) Map relations and
interpretation of satellite images suggest that,
like the Muztaghata dome, the Sares dome is
asymmetric with a shallower-dipping northern
limb, defined by a sinuous contact that interacts
with local topography, and a steeper-dipping
southern limb, which is straighter and does not
appear to interact with topography. (5) The Sares
and Muskol gneiss domes record early Miocene
40
Ar/39Ar hornblende and biotite cooling ages,
with Cretaceous 40Ar/39Ar cooling ages in the
Central Pamir south of the domes (101 Ma;
Schwab et al., 2004), similar to the age juxtapo-
sition across the Shen-ti and Torbashi faults.
 Cenozoic evolution of the eastern Pamir

A 74° 76°
Geologic units:
Quaternary deposits
Kalagile Cenozoic midcrustal rocks
fault Cenozoic plutons (from
Schwab et al., 2004,
Ghez and Hubbard et al., 1999)
Map Symbols:
fault Gneiss dome axes

Boundary of Miocene
detachments
A Thrust fault
Muskol dome
Normal fault

Strike-slip fault
B
Buried fault

C
Sares dome
38° 38°

Shen-ti
fault

B Kongur Shan
A
C
5 5
B
t
ul u lt
fa
Elevation (km)

Elevation (km)

0 al Ghez fault 0 l fa
rm ma
no or
-5 EPSZ n -5 EPSZ nn
ha Sha EPSZ
-10 urS -10 ur
ng ng
Ko Ko
-15 -15

EPSZ
Muztaghata
C
5
Elevation (km)

ult
al fa
0 orm
nn
D -5 r Sh
a
n gu
Ko
-10

Figure 11. (A) Simplified tectonic map of the Pamir emphasizing Cenozoic high-grade rocks exposed in the cores of the Muskol and Sares
domes and the footwall of the Kongur Shan normal fault, and the distribution of the Cenozoic Sares-Murgab belt of plutons in the Pamir.
(B–D) Schematic cross sections A, B, and C across the Kongur Shan normal fault and Eastern Pamir shear zone showing the interpreted
relation between the two structures. The line denoting the Eastern Pamir shear zone (EPSZ) represents the schematic location of a shear
zone several kilometers thick.

Geological Society of America Bulletin, July/August 2007 893

 Robinson et al.

Although cooling-age data from Schwab et
al. (2004) for the Sares and Muskol domes are
limited, they suggest a complicated history of
synchronous gneiss dome formation and crustal
thickening during the Miocene. Early Miocene
cooling ages for the Sares and Muskol domes
overlap with prograde metamorphism docu-
mented within the Muztaghata gneiss dome and
are older than the late Miocene cooling ages in
the footwall of the Shen-ti fault. The eastward
decrease in cooling ages along the gneiss domes
suggests that either gneiss dome formation was
diachronous, migrating from east to west, or
that exhumation lasted longer in the east than in
the west, eventually exhuming rocks that were
still undergoing prograde metamorphism in the
middle Miocene. We tentatively prefer the lat-
ter, based on the apparent eastward increase in
north-south width of the Central Pamir gneiss
domes (Fig. 11).
Tectonic Model for Crustal Thickening and
Gneiss Dome Formation
While the northern margin of the high-grade
core of the Sares antiform is fault-bounded, our
field observations indicate that the gneisses of
the Muztaghata antiform are structurally con-
indicating that the right-slip Ghez shear zone
has been rotated from an originally shallowly
dipping to subhorizontal position (Fig. 11B),
with a top-to-the-south sense of shear. While the
southward continuation of the eastern bound-
ary of the high-grade metamorphic rocks has
not been documented in detail, all regional geo-
logic maps place it in roughly the same location
(Figs. 2 and 11A), and several depict it as a fault
(e.g., Yin and Bian, 1992; Xinjiang Bureau of
Geology and Mineral Resources, 1993). Fur-
thermore, this eastern contact of the terrane is
depicted as connecting with the east-striking
Shen-ti fault in the south.
We propose that the eastern contact of the
Cenozoic high-grade terrane is a continuous
shear zone that links with the Shen-ti fault to the
south, which we refer to as the Eastern Pamir
shear zone (Fig. 11). In our model, the Eastern
Pamir shear zone is a regional ductile décolle-
ment that accommodated northward relative
displacement of the middle and lower crust rela-
A Oligocene Eastern Pamir shear zone
tive to the upper crust beneath the northeastern
Pamir, which allowed Oligocene to Miocene
crustal thickening to occur with limited or
no shortening in the upper crust of the North-
ern Pamir (Fig. 12A). Based on metamorphic
assemblages within the Ghez shear zone (garnet
+ biotite) and pervasive ductile shear distributed
over a zone several kilometers wide, the East-
ern Pamir shear zone was located below the
brittle-ductile transition at mid-crustal depths of
~15 km. This is similar to the depth to detach-
ment for low-angle normal fault systems (e.g.,
Davis and Lister, 1988; Wernicke and Axen,
1988; Lister and Davis, 1989), including those
in Tibet (e.g., Kapp et al., 2005). Crustal thick-
ening in the footwall of the ductile décollement
led to melting of the middle and lower crust
beneath the Central Pamir during the Miocene
(Fig. 12B) and the formation of the Sares-Mur-
gab granitoid belt (Fig. 11A). Continued north-
south shortening in the footwall of the Eastern
Pamir shear zone, facilitated by a decrease in
Main Pamir Thrust

tinuous with those of the Kongur Shan massif to S N

the north, and they form a coherent high-grade
metamorphic terrane in the footwall of the Kon- Tarim

gur Shan normal fault. This indicates that the

tectonic evolution of this metamorphic terrane
involved synchronous middle to late Miocene
exhumation at its southern end (this study) B Early Miocene
and continued prograde metamorphism at its S N
northern end (Robinson et al., 2004). Here, we
discuss several other regional observations and Tarim
propose a tectonic model for the evolution of the
Cenozoic metamorphic terrane.
One observation from our field studies in the
region and previously published information
(Burtman and Molnar, 1993) is the lack of Ceno-
zoic internal shortening of the metamorphic
C Middle Miocene
rocks that make up the Northern Pamir terrane S N
south of the Oligocene-Holocene fold-and-thrust
belt in the Alai Valley. This requires the Miocene Tarim
crustal thickening and prograde metamorphism
documented in the Kongur Shan massif (Robin-
son et al., 2004) to have been driven by under-
thrusting of material beneath the Northern Pamir
from either the north or the south.
Where the eastern boundary of the Cenozoic
high-grade metamorphic terrane has been well
documented east of the Kongur Shan massif,
it coincides with the right-lateral Ghez shear Figure 12. Cenozoic tectonic evolution of the middle and lower crust in the Pamir: (A) Oli-
zone (Fig. 3) (Brunel et al., 1994; Robinson et gocene underthrusting of the middle and lower crust along the Eastern Pamir shear zone
al., 2004). The 40Ar/39Ar cooling ages and ther- with crustal thickening and prograde metamorphism. (B) Continued thickening and heat-
mokinematic modeling from the Kongur Shan ing of the middle and lower crust leads to dehydration melting in the early to middle Mio-
massif require significant footwall rollover cene, generating granitic melts. (C) Weakened middle and lower crust is extruded upward
(>40°) during extension (Robinson et al., 2005), during continued north-south compression, forming the Central Pamir gneiss domes.

894 Geological Society of America Bulletin, July/August 2007


both the density and strength of the footwall
of the shear zone due to melting, resulted in
extrusion of the Central Pamir gneiss domes
synchronous with continued crustal thickening
and metamorphism (Fig. 12C) (e.g., Teyssier
and Whitney, 2002; Burg et al., 2004; Mahéo et
al., 2004). Synchronous gneiss dome formation
and continued northward relative motion of the
footwall of the Eastern Pamir shear zone during
gneiss dome formation accounts for both late
Miocene (ca. 9 Ma) prograde metamorphism
along the Kongur Shan massif (Robinson et al.,
2004) and the asymmetric, south-vergent shape
of the Muztaghata gneiss dome (e.g., Whitney
et al., 2004).
A possible mechanical drive for our model of
deformation, crustal thickening, and gneiss dome
formation is the proposed Cenozoic northward
subduction of the Karakorum terrane beneath
the Central Pamir (Ducea et al., 2003; Hacker et
al., 2005). Thus, structures accommodating top-
to-the-south Oligocene-Miocene displacement
along the Eastern Pamir shear zone footwall
would surface where intercontinental subduction
was being accommodated, in the southern Kara-
korum terrane (Fraser et al., 2001), or along one
of the suture zones in the Southern Pamir and
farther south (Hacker et al., 2005).
CONCLUSIONS
Field and analytical investigations along the
southern end of the Kongur Shan normal fault
integrated with previously published results
from the northern end of the fault system show a
generally southward decrease in the magnitude
of east-west extension along the Kongur Shan
extensional system. These results are consis-
tent with oroclinal bending–radial thrusting of
the Pamir salient as the primary driver of east-
west extension in the Pamir. However, other
processes, such as synorogenic extension, may
have played a role in the exhumation of the Kon-
gur Shan massif.
High-grade schists at the southern end of
the Kongur Shan extensional system along the
Muztaghata massif record upper amphibolite
facies metamorphic conditions (~9 kbar, 750
°C), which locally resulted in muscovite dehy-
dration melting. Ion-microprobe Th-Pb mona-
zite and U-Pb zircon ages yield a distribution
of ages from the Oligocene to late Miocene,
which we interpret to document prograde meta-
morphism during the late Oligocene to middle
Miocene, culminating in partial melting in the
middle Miocene. Our interpreted timing of
anatexis overlaps with the age distribution of
Miocene granitic plutons in the Central Pamir,
indicating widespread melting of the middle and
lower crust at this time.
Geological Society of America Bulletin, July/August 2007
Cenozoic evolution of the eastern Pamir
Exhumation of high-grade schists and
gneisses along the southern Kongur Shan nor-
mal fault occurred during the late Miocene prior
to initiation of east-west extension, and exhu-
mation of the southern end of the metamorphic
terrane was accommodated along the south-dip-
ping, top-to-the-south Shen-ti fault. We interpret
this rapid exhumation event, and formation of
the Muztaghata antiform, to be related to the
development of the Miocene Central Pamir
gneiss domes, where the Muztaghata antiform is
the eastward continuation of the Sares antiform
of Schwab et al. (2004).
Our interpreted eastward continuation of the
Sares antiform into the footwall of the Kongur
Shan normal fault, and lack of evidence for
strike-slip deformation along the Tashkorgan
Valley, suggests that the Central Pamir antiforms
have not been offset across the Karakorum fault
from the Qiangtang anticlinorium as suggested
by Schwab et al. (2004).
ACKNOWLEDGMENTS
Discussions with Eric Cowgill, Mike Taylor, and
Alex Webb helped clarify many of the ideas presented
here. We thank Marty Grove for assistance with the
Ar analyses, Frank Kyte for assistance with the elec-
tron microprobe analyses, and Marty Grove and Axel
Schmidt for support with the ion microprobe analyses
and interpretation of data. Careful and critical reviews
by Michael Edwards, Rasmus Thiede, and especially
Lothar Ratschbacher substantially improved the initial
manuscript. Cumulative probability plots were gener-
ated using Isoplot/Ex v 3.0 (Ludwig, 2003). This work
was supported by National Science Foundation grant
EAR-0126122 and a Grants-in-Aid of Research from
the Geological Society of America. The ion micro-
probe facility at University of California–Los Ange-
les (UCLA) is partially supported by a grant from the
Instrumentation and Facilities Program, Division of
Earth Sciences, National Science Foundation.
REFERENCES CITED
Arnaud, N.O., Brunel, M., Cantagrel, J.M., and Tapponnier,
P., 1993, High cooling and denudation rates at Kon-
gur Shan, eastern Pamir (Xinjiang, China) revealed by
40
Ar/39Ar alkali feldspar thermochronology: Tectonics,
v. 12, no. 6, p. 1335–1346.
Blisniuk, P.M., and Strecker, M.R., 1996, Kinematics of
Holocene normal faulting in the Northern Pamir: Eos
(Transactions, American Geophysical Union), v. 77,
no. 46, p. F693.
Bohlen, S.R., and Liotta, J.J., 1986, A barometer for garnet
amphibolites and garnet granulites: Journal of Petrol-
ogy, v. 27, p. 1025–1034.
Brunel, M., Arnaud, N., Tapponnier, P., Pan, Y., and
Wang, Y., 1994, Kongur Shan normal fault: Type
example of mountain building assisted by extension
(Karakoram fault, eastern Pamir): Geology, v. 22,
p. 707–710, doi: 10.1130/0091-7613(1994)022<0707:
KSNFTE>2.3.CO;2.
Burg, J.-P., Kaus, B.J.P., and Podladchikov, Y.Y., 2004,
Dome structures in collision orogens: Mechanical
investigation of the gravity/compression interplay, in
Whitney, D.L., Teyssier, C., and Siddoway, C.S., eds.,
Gneiss Domes in Orogeny: Geological Society of
America Special Paper 380, p. 47–66.
Burtman, V.S., 2000, Cenozoic crustal shortening between
the Pamir and Tien Shan and a reconstruction of the
Pamir–Tien Shan transition zone for the Cretaceous
and Palaeogene: Tectonophysics, v. 319, p. 69–92, doi:
10.1016/S0040-1951(00)00022-6.
Burtman, V.S., and Molnar, P., 1993, Geological and geo-
physical evidence for deep subduction of continental
crust beneath the Pamir: Geological Society of Amer-
ica Special Paper 281, 76 p.
Chung, S.-L., Chu, M.-F., Zhang, Y., Xie, Y., Lo, C.-H.,
Lee, T.-Y., Lan, C.-Y., Li, X., Zhang, Q., and Wang, Y.,
2005, Tibetan tectonic evolution inferred from spatial
and temporal variations in post-collisional magma-
tism: Earth-Science Reviews, v. 68, p. 173–196, doi:
10.1016/j.earscirev.2004.05.001.
Cowgill, E., Yin, A., Harrison, T.M., and Xiao-Feng, W.,
2003, Reconstruction of the Altyn Tagh fault based
on U-Pb geochronology: Role of back thrusts, mantle
sutures, and heterogeneous crustal strength in forming
the Tibetan Plateau: Journal of Geophysical Research,
v. 108, no. B7, p. 2346, doi: 10.1029/2002JB002080.
Davis, G.A., and Lister, G.S., 1988, Detachment faulting
in continental extension: Perspectives from the south-
western U.S. Cordillera, in Clark, S.P., Jr., Burchfiel,
B.C., Suppe, J., eds., Processes in continental litho-
spheric deformation: Geological Society of America
Special Paper 218, p. 133–159.
Ding, L., Zhong, D., Yin, A., Kapp, P., and Harrison, T.M.,
2001, Cenozoic structural and metamorphic evolution
of the eastern Himalayan syntaxis (Namche Barwa):
Earth and Planetary Science Letters, v. 192, p. 423–
438, doi: 10.1016/S0012-821X(01)00463-0.
Ducea, M.N., Lutkov, V., Minaev, V.T., Hacker, B., Ratsch-
bacher, L., Luffi, P., Schwab, M., Gehrels, G.E., McWil-
liams, M., Vervoort, J., and Metcalf, J., 2003, Building
the Pamirs: The view from the underside: Geology,
v. 31, no. 10, p. 849–852, doi: 10.1130/G19707.1.
Fraser, J.E., Searle, M.P., Parrish, R.R., and Nobel, S.R.,
2001, Chronology of deformation, metamorphism,
and magmatism in the southern Karakoram Moun-
tains: Geological Society of America Bulletin,
v. 113, no. 11, p. 1443–1455, doi: 10.1130/0016-
7606(2001)113<1443:CODMAM>2.0.CO;2.
Gaetani, M., 1997, The Karakorum block in Central Asia, from
Ordovician to Cretaceous: Sedimentary Geology, v. 109,
p. 339–359, doi: 10.1016/S0037-0738(96)00068-1.
Hacker, B., Luffi, P., Lutkov, V., Minaev, V., Ratschbacher,
L., Plank, T., Ducea, M.N., Patiño-Douce, A.E., McWil-
liams, M., and Metcalf, J., 2005, Near-ultrahigh pres-
sure processing of continental crust: Miocene crustal
xenoliths from the Pamir: Journal of Petrology, v. 46,
no. 8, p. 1661–1687, doi: 10.1093/petrology/egi030.
Haines, S.S., Klemperer, S.L., Brown, L., Jingru, G., Mechie,
J., Meissner, R., Ross, A., and Wenjin, Z., 2003,
INDEPTH III seismic data: From surface observations
to deep crustal processes in Tibet: Tectonics, v. 22,
no. 1, p. 1001, doi: 10.1029/2001TC001305,2003.
Horton, B.K., Yin, A., Spurlin, M.S., Zhou, J., and Wang,
J., 2002, Paleocene-Eocene syncontractional sedi-
mentation in narrow, lacustrine-dominated basins of
east-central Tibet: Geological Society of America
Bulletin, v. 114, no. 7, p. 771–786, doi: 10.1130/0016-
7606(2002)114<0771:PESSIN>2.0.CO;2.
Hoskin, P.W.O., and Black, L.P., 2000, Metamorphic zircon
formation by solid-state recrystallization of protolith
igneous zircon: Journal of Metamorphic Geology, v. 18,
p. 423–439, doi: 10.1046/j.1525-1314.2000.00266.x.
Houseman, G., and England, P., 1996, A lithospheric-thick-
ening model for the Indo-Asian collision, in Yin, A.,
and Harrison, T.M., eds., The Tectonic Evolution of
Asia: New York, Cambridge University Press, p. 3–17.
Hsu, K.J., Pan, G., and engör, A.M.C., 1995, Tectonic evo-
lution of the Tibetan Plateau: A working hypothesis
based on the archipelago model of orogenesis: Interna-
tional Geology Review, v. 37, p. 473–508.
Kapp, J.L.D.A., Harrison, T.M., Kapp, P., Grove, M., Lov-
era, O.M., and Lin, D., 2005, Nyainqentanglha Shan:
A window into the tectonic, thermal, and geochemical
evolution of the Lhasa block, southern Tibet: Journal
of Geophysical Research, v. 110, no. B08413, doi:
10.1029/2004JB003330.
Kapp, P., Yin, A., Manning, C.E., Murphy, M., Harrison,
T.M., and Spurlin, M., 2000, Blueschist-bearing meta-
morphic core complexes in the Qiangtang block reveal
deep crustal structure of northern Tibet: Geology, v. 28,
895

no. 1, p. 19–22, doi: 10.1130/0091-7613(2000)28<19:
BMCCIT>2.0.CO;2.
Kapp, P., Murphy, M.A., Yin, A., Harrison, T.M., Lin, D.,
and Jinghui, G., 2003a, Mesozoic and Cenozoic tec-
tonic evolution of the Shiquanhe area of western Tibet:
Tectonics, v. 22, 1029, doi: 10.1029/2001TC001332.
Kapp, P., Yin, A., Manning, C.E., Harrison, T.M., and
Taylor, M., 2003b, Tectonic evolution of the early
Mesozoic blueschist-bearing Qiangtang metamorphic
belt, central Tibet: Tectonics, v. 22, no. 4, 1043, doi:
10.1029/2002TC001383.
Koziol, R., and Newton, R.C., 1988, Redetermination of the
anorthite breakdown reaction and improvement of the
plagioclase-garnet Al2SiO5-quartz geobarometer: The
American Mineralogist, v. 73, p. 216–223.
Lacassin, R., Valli, F., Arnaud, N., Leloup, P.H., Papuette, J.L.,
Haibing, L., Tapponnier, P., Chevalier, M.-L., Guillot,
S., Mahéo, G., and Zhiqin, X., 2004, Large-scale geom-
etry, offset and kinematic evolution of the Karakorum
fault, Tibet: Earth and Planetary Science Letters, v. 219,
p. 255–269, doi: 10.1016/S0012-821X(04)00006-8.
Leier, A.L., DeCelles, P.G., Kapp, P., and Ding, L., 2007,
The Takena Formation of the Lhasa terrane, southern
Tibet: The record of a Late Cretaceous retroarc fore-
land basin: Geological Society of America Bulletin,
v. 119, no. 1/2, p. 31–48, doi: 10.1130/B25974.1.
Lister, G.S., and Davis, G.A., 1989, The origin of metamor-
phic core complexes and detachment faults formed
during Tertiary continental extension in the northern
Colorado River region, U.S.A.: Journal of Struc-
tural Geology, v. 11, p. 65–94, doi: 10.1016/0191-
8141(89)90036-9.
Liu, Z.Q., 1988, Geologic Map of the Qinghai-Xizang Pla-
teau and its Neighboring Regions: Beijing, Chengdu
Institute of Geology and Mineral Resources, Geologi-
cal Publishing House, scale 1:5,000,000.
Ludwig, K.R., 2003, Users’ Manual for Isoplot/Ex Version
3.0: A Geochronological Toolkit for Microsoft Excel:
Berkeley, Berkeley Geochronology Center Special
Publication 4, 70 p.
Mahéo, G., Pêcher, A., Guillot, S., Rolland, Y., and Dela-
court, C., 2004, Exhumation of Neogene domes
between oblique crustal boundaries in south Karako-
rum (northwest Himalaya, Pakistan), in Whitney, D.L.,
Teyssier, C., and Siddoway, C.S., eds., Gneiss Domes
in Orogeny: Geological Society of America Special
Paper 380, p. 141–154.
Matte, P., Tapponnier, P., Arnaud, N., Bourjot, L., Avouac,
J.P., Vidal, P., Liu, Q., Pan, Y., and Wang, Y., 1996,
Tectonics of Western Tibet, between the Tarim and
the Indus: Earth and Planetary Science Letters, v. 142,
p. 311–330, doi: 10.1016/0012-821X(96)00086-6.
Mojzsis, S.J., and Harrison, T.M., 2002, Establishment
of a 3.83-Ga magmatic age for the Akilia tonalite
(southern West Greenland): Earth and Planetary Sci-
ence Letters, v. 202, p. 563–576, doi: 10.1016/S0012-
821X(02)00825-7.
Murphy, M.A., Yin, A., Harrison, T.M., Durr, S.B., Chen, Z.,
Ryerson, F.J., Kidd, W.S.F., Wang, X., and Zhou, X., 1997,
Did the Indo-Asian collision alone create the Tibetan Pla-
teau?: Geology, v. 25, p. 719–722, doi: 10.1130/0091-
7613(1997)025<0719:DTIACA>2.3.CO;2.
Murphy, M.A., Yin, A., Kapp, P., Harrison, T.M., Ling, D.,
and Jinghui, G., 2000, Southward propagation of the
Karakoram fault system, southwest Tibet: Timing and
magnitude of slip: Geology, v. 28, no. 5, p. 451–454, doi:
10.1130/0091-7613(2000)28<451:SPOTKF>2.0.CO;2.
Neil, E.A., and Houseman, G.A., 1997, Geodynamics of the
Tarim Basin and the Tian Shan in central Asia: Tecton-
ics, v. 16, no. 4, p. 571–584, doi: 10.1029/97TC01413.
Pan, Y., 1992, Introduction to Integrated Scientific Investi-
gation on Karakoram and Kunlun Mountains: Beijing,
China Meteorological Press, 92 p.
Pan, Y., and Bian, Q., 1996, Tectonics, in Pan, Y., ed., Geo-
logical Evolution of the Karakoram and Kunlun Moun-
tains: Beijing, Seismological Press, p. 187–229.
Pashkov, B.R., and Dmitriyev, E.A., 1982, Muzkol’ crys-
talline massif (Central Pamir): International Geology
Review, v. 24, no. 3, p. 285–296.
Peltzer, G., and Tapponnier, P., 1988, Formation and evo-
lution of strike-slip faults, rifts, and basins during the
India-Asia collision: An experimental approach: Jour-
nal of Geophysical Research, v. 93, p. 15,085–15,117.
896 Geological Society of America Bulletin, July/August 2007
Robinson et al.
Peykre, Y.B., Ye, A.P., and Zotov, I.A., 1982, The Muzkol’
metamorphic complex of the Central Pamir: Interna-
tional Geology Review, v. 24, no. 3, p. 297–303.
Powell, R., Holland, T., and Worley, B., 1998, Calculating
phase diagrams involving solid solutions via non-linear
equations, with examples using THERMOCALC: Jour-
nal of Metamorphic Geology, v. 116, no. 4, p. 557–588.
Ratschbacher, L., Frisch, W., Liu, G., and Chen, C.C., 1994,
Distributed deformation in southern and western Tibet
during and after the India-Asian collision: Journal of
Geophysical Research, v. 99, p. 19,917–19,945, doi:
10.1029/94JB00932.
Robinson, A.C., Yin, A., Manning, C.E., Harrison, T.M.,
Zhang, S.-H., and Wang, X.-F., 2004, Tectonic evolu-
tion of the northeastern Pamir: Constraints from the
northern portion of the Cenozoic Kongur Shan exten-
sional system: Geological Society of America Bulletin,
v. 116, p. 953–974, doi: 10.1130/B25375.1.
Robinson, A.C., Yin, A., and Lovera, O.M., 2005, Thermo-
kinematic modeling of the effect of footwall erosion on
cooling age patterns in detachment fault settings: Eos
(Transactions, American Geophysical Union), v. 86,
no. 52, p. 243.
Schwab, M., Ratschbacher, L., Siebel, W., McWilliams, M.,
Minaev, V., Lutkov, V., Chen, F., Stanek, K., Nelson,
B., Frisch, W., and Wooden, J.L., 2004, Assembly of
the Pamirs: Age and origin of magmatic belts from
the southern Tien Shan to the southern Pamirs and
their relation to Tibet: Tectonics, v. 23, TC4002, doi:
10.1029/2003TC001583.
Searle, M.P., 1996, Geological evidence against large-scale
pre-Holocene offsets along the Karakoram fault: Implica-
tions for the limited extrusion of the Tibetan Plateau: Tec-
tonics, v. 15, no. 1, p. 171–186, doi: 10.1029/95TC01693.
Searle, M.P., Weinberg, R.F., and Dunlap, W.J., 1998,
Transpressional tectonics along the Karakoram fault
zone, northern Ladakh: Constraints on Tibetan extru-
sion, in Holdsworth, R.E., Strachan, R.A., and Dewey,
J.F., eds., Continental Transpressional and Transten-
sional Tectonics: Geological Society of London Spe-
cial Publication 135, p. 307–326.
Şengör, A.M.C., and Natal’in, B.A., 1996, Paleotectonics of
Asia: Fragments of a synthesis, in Yin, A., and Har-
rison, T.M., eds., The Tectonic Evolution of Asia: New
York, Cambridge University Press, p. 486–640.
Şengör, A.M.C., and Okurogullari, A.H., 1991, The role of
accretionary wedges in the growth of continents: Asi-
atic examples from Argand to plate tectonics: Eclogae
Geologicae Helvetiae, v. 84, no. 3, p. 535–597.
Sobel, E.R., and Dumitru, T.A., 1997, Thrusting and exhu-
mation around the margins of the western Tarim
Basin during the India-Asia collision: Journal of Geo-
physical Research, v. 102, no. B3, p. 5043–5063, doi:
10.1029/96JB03267.
Spurlin, M.S., Yin, A., Horton, B.K., Zhou, J., and Wang, J.,
2005, Structural evolution of the Yushu-Nanqian region
and its relationship to syncollisional igneous activity, east-
central Tibet: Geological Society of America Bulletin,
v. 117, no. 9/10, p. 1293–1317, doi: 10.1130/B25572.1.
Strecker, M.R., Frisch, W., Hamburger, M.W., Ratschbacher,
L., Semiletkin, S., Zamoruyev, A., and Sturchio, N.,
1995, Quaternary deformation in the Eastern Pamirs,
Tadzhikistan and Kyrgyzstan: Tectonics, v. 14, no. 5,
p. 1061–1079, doi: 10.1029/95TC00927.
Tapponnier, P., Mattauer, M., Proust, F., and Cassaigneau, C.,
1981, Mesozoic ophiolites, sutures, and large-sale tec-
tonic movements in Afghanistan: Earth and Planetary
Science Letters, v. 52, p. 355–371, doi: 10.1016/0012-
821X(81)90189-8.
Tapponnier, P., Peltzer, G., Le Dain, A.Y., Armijo, R., and
Cobbold, P., 1982, Propagating extrusion tectonics in
Asia: New insights from simple experiments with plas-
ticine: Geology, v. 10, p. 611–616, doi: 10.1130/0091-
7613(1982)10<611:PETIAN>2.0.CO;2.
Taylor, M., Yin, A., Ryerson, F.J., Kapp, P., and Ding, L.,
2003, Conjugate strike-slip faulting along the Ban-
gong-Nujiang suture zone accommodates coeval
east-west extension and north-south shortening in the
interior of the Tibetan Plateau: Tectonics, v. 22, no. 4,
p. 1044, doi: 10.1029/2002TC001361.
Teyssier, C., and Whitney, D.L., 2002, Gneiss domes and
orogeny: Geology, v. 30, no. 12, p. 1139–1142, doi:
10.1130/0091-7613(2002)030<1139:GDAO>2.0.CO;2.
Thomas, J.-C., Chauvin, A., Gapais, D., Bazhenov, M.L.,
Perroud, H., Cobbold, P.R., and Burtman, V.S., 1994,
Paleomagnetic evidence for Cenozoic block rotations
in the Tadjik depression (Central Asia): Journal of Geo-
physical Research, v. 99, no. B8, p. 15,141–15,160,
doi: 10.1029/94JB00901.
Thomas, J.-C., Cobbold, P.R., Wright, A., and Gapais, D.,
1996, Cenozoic tectonics of the Tadzhik depression,
Central Asia, in Yin, A., and Harrison, T.M., eds., The
Tectonic Evolution of Asia: New York, Cambridge
University Press, p. 191–207.
Waldhör, M., Appel, E., Frisch, W., and Patzelt, A., 2001, Pal-
aeomagnetic investigation in the Pamirs and its tectonic
implications: Journal of Asian Earth Sciences, v. 19,
p. 429–451, doi: 10.1016/S1367-9120(00)00030-4.
Wernicke, B., and Axen, G.J., 1988, On the role of isostasy in
the evolution of normal fault systems: Geology, v. 16,
p. 848–851, doi: 10.1130/0091-7613(1988)016<0848:
OTROII>2.3.CO;2.
Whitney, D.L., Teyssier, C., and Vanderhaeghe, O., 2004,
Gneiss domes and crustal flow, in Whitney, D.L., Teys-
sier, C., and Siddoway, C.S., eds., Gneiss Domes in
Orogeny: Geological Society of America Special Paper
380, p. 15–33.
Xiao, W.J., Windley, B.F., Chen, H.L., Zhang, G.C., and
Li, J.L., 2002, Carboniferous-Triassic subduction
and accretion in the western Kunlun, China: Implica-
tions for the collisional and accretionary tectonics of
the northern Tibetan Plateau: Geology, v. 30, no. 4,
p. 295–298, doi: 10.1130/0091-7613(2002)030<0295:
CTSAAI>2.0.CO;2.
Xinjiang Bureau of Geology and Mineral Resources, 1993,
Regional Geology of Xinjiang Uygur Autonomous
Region: Beijing, Geological Publishing House, 841
p. (in Chinese with English abstract).
Xu, R., Zhang, Y., Xie, Y., Vidal, P., Arnaud, N., Zhang, Q.,
and Zhao, D., 1996, Isotopic geochemistry of plutonic
rocks, in Pan, Y., ed., Geological Evolution of the Kara-
koram and Kunlun Mountains: Beijing, Seismological
Press, p. 137–186.
Yin, A., and Harrison, T.M., 2000, Geologic Evolution of
the Himalayan-Tibetan orogen: Annual Review of
Earth and Planetary Sciences, v. 28, p. 211–280, doi:
10.1146/annurev.earth.28.1.211.
Yin, A., and Nie, S., 1996, A Phanerozoic palinspastic
reconstruction of China and its neighboring regions, in
Yin, A., and Harrison, T.M., eds., The Tectonic Evolu-
tion of Asia: New York, Cambridge University Press,
p. 442–485.
Yin, A., Robinson, A., and Manning, C.E., 2001, Oroclinal
bending and slab break-off causing coeval east-west
extension and east-west contraction in the Pamir–
Nanga Parbat syntaxis in the past 10 m.y.: Eos (Trans-
actions, American Geophysical Union), v. 82, no. 47,
p. F1124.
Yin, A., Rumelhart, P.E., Butler, R., Cowgill, E., Harrison,
T.M., Foster, D.A., Ingersoll, R.V., Qing, Z., Xian-
Qiang, Z., Xiao-Feng, W., Hanson, A., and Raza, A.,
2002, Tectonic history of the Altyn Tagh fault sys-
tem in northern Tibet inferred from Cenozoic sedi-
mentation: Geological Society of America Bulletin,
v. 114, no. 10, p. 1257–1295, doi: 10.1130/0016-
7606(2002)114<1257:THOTAT>2.0.CO;2.
Yin, J., and Bian, Q., 1992, Geologic Map of the Kara-
koram–Western Kunlun and Adjacent Regions: Bei-
jing, Science Press, scale 1:2,000,000.
Youngun, Y., and Hsu, K.J., 1994, Origin of the Kunlun
Mountains by arc-arc and arc-continent collisions: The
Island Arc, v. 3, no. 2, p. 75–89, doi: 10.1111/j.1440-
1738.1994.tb00096.x.
Zhang, Y., Xie, Y., Xu, R., Vidal, P., and Arnaud, N., 1996,
Geochemistry of granitoid rocks, in Pan, Y., ed., Geo-
logical Evolution of the Karakoram and Kunlun Moun-
tains: Beijing, Seismological Press, p. 94–136.
MANUSCRIPT RECEIVED 27 JANUARY 2006
REVISED MANUSCRIPT RECEIVED 16 JANUARY 2007
MANUSCRIPT ACCEPTED 19 JANUARY 2007
Printed in the USA
 Cenozoic evolution of the eastern Pamir: Implications for strain-accommodation mechanisms at the western end of the Himalayan-Tibetan orogen
Alexander C. Robinson, An Yin, Craig E. Manning, T. Mark Harrison, Shuan-Hong Zhang, Xiao-Feng Wang
Figures 4 and 5
Supplement to: Geological Society of America Bulletin, v. 119, no. 7/8, doi: 10.1130/B25981.S1.
© Copyright 2007 Geological Society of America
A 75° 00′ 75° 30′
52
44 79

Kongur Shan
19 15 31 49 84
54
38
Pzgn1 45 84

lt
Q2 65

fau
87 40
52 66 65 67
40 51
61 4000 23

n
62 69
Q1 51 46

ma
75 44 50
28
77 Q1 47 44
Pzgn1

Tah
55

C
76 23
00
73 76 45
50

fault
20 42
60 87 75 61

44
46
75 72 Pzsch4 400
40 0
n = 37
Ta

48 35
17
65 53
66 75 area unmapped
s

46 38
72
hk

37 45 74
61 68

0
29 33 86 77

400
48 40
o rg

10 28 79 85 49 71
Mgr 41 67 63 Pzsch3
24 48 55
Pzsch3 20
an

55 66 28
18 71 52
0 10 84
72 82
fau

17 56 69 74
28 54 60
58
Q2 80 59 32 32
77
Shen-ti
lt

72
45 25 37 69 87 34 34
12
38 65 65
Shen-ti fault
38 72 30 47 62 15
13 61
27 37 33
42
18 Pzsch4 28
32
36 38 4000
33 23 37

Q2
18
31 Torbashi fault 33
83

64

84 54 Q1
Data from the Tashkorgan River

30
34
?

00
17 82 Data from the Wuqa Valley

30
2 53

00
40

40
Q2
Pzsch5

00
Mgr
43
4
Tashkorgan
62
Shen-ti fault 44 area unmapped

40
00
Pzsch5
5000
Pzsch4
4000

Q1 52
62 83
12 81 2 50
64
44 27 00
38
?? 4

To
Qss
Ta

56

r
64

Wu

ba
sh

sh
ko

75

q
rga

i fa
a fa
Esy

Wu
nR

ult
qa

ult
ive

31 65

R
64

ive
0 53
r

500 Qss
49
Pzsch4
34

r
37º 40'

37º 40'
Pzsch5 48
4000

Torbashi Q2
Qm 6

46
40 ?
00
?
Tash

area unmapped
Q1
k orga

B Q1 area unmapped Q2
Pzsch5
n fau

n = 22
Gl

Qm Qm
ac

Esy
lt

47
ier

14
5000

Glacie QL
r
Mzgr
Geologic Units:
Qm
Q1 Quaternary alluvial/fluvial deposits

40
Qm Quaternary glacial moraines
00
Q2
Q2 Old Quaternary alluvial/fluvial deposits
41 Pzsch5
0 QL Quaternary (?) loess
400
9
Sedimentary Units

D 5000 Qss Quaternary (?) medium-grained deposits

Kss Cretaceous red beds

48
26

area unmapped JL2 Jurassic limestone

37 31
? JL1 Jurassic limestone

PL Permian limestone

Ps Permian slate
Q1
QL
To
rb
Metamorphic Units

Pzgn1 Paleozoic quartzofeldspathic gneisses.

as
00
40

hi
fau Paleozoic upper amphibolite facies metasediments.
lt Pzsch3
Structurally lower portions dominated by pelitic
schists. Structurally higher portions dominated by
marble, calc-silicates, and mafic schist.
0
500
N Pzsch4 Paleozoic upper greenschist facies schists and
4000
mafic schists.
QL PL Pzsch5 Paleozoic upper greenschist to upper amphibolite
?
22 facies schists and gneisses.
PL
57 65
Plutonic Units

81 86 Mgr Miocene biotite granite

43
2 km
39 87
Figure Esy Eocene diopside-bearing syenite
5D
37º 20'

Mzgr Triassic-Early Jurassic granites and granodiorites

46
69 Q2
4000 53
76

77 75
Ps 48
34
44 Map Symbols:
area unmapped Kongur Shan normal fault
JL1 ? ?
Normal faults associated with the Kongur Shan
Kal

N
5000
extensional system
Normal faults not associated with the Kongur Shan
aqig

47
extensional system
Thrust fault
u faul

37
47
Anticline
Ps
t

Syncline
JL2 ? ? Qm

Approximate or inferred fault contact

48
?
Fault with unknown kinematics
59
?
41
JL1 Kss 20 Buried structure
33 85
68 Strike and dip of foliation and
53
85 trend and plunge of lineations
47
Q1
70
35 47 Strike and dip of bedding
68

0 5 10 Qm
Ps Lakes and rivers
Contour Interval: 200 m
4000

4000
Q2 Glaciers

75° 00′ JL1

44

Figure 4. (A) Geologic map of the Tashkorgan normal fault, NE Pamir, China. Lower-hemisphere equal-area stereonets show foliation planes and lineations from (B) unit
Pzsch4, and (C) unit Pzsch5. (D) ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) visible bands (red—3; green—2; blue—1) of the Kalaqigu
fault (black arrows) south of the Tashkorgan fault. Undisturbed Quaternary deposits capping ridges (white arrows) indicate no recent motion along the fault.
 A 75° 00′ 75° 30′
Pzgw Qm

Mzgr
Qm 6525
B
Muztaghata
Pzsch1 Q2 massif
Pzgw n = 111

Kongur Shan
4000
Glacier
30
30 Pzgw
57 60 40 54 27
70 54
36
65
47 62
56
area unmapped

6220

fault
32
35
Pzgw 38 38
32 Q1
10 32 Glacier
25 00
50 Lineations
Bao
29 AR 6/12/00-4
30 40
AR 6/12/00-2
23 00 2.92 ± 0.17(b)
Pzsch1
ziya 40 2.96 ± 0.03 (m)
Poles to Planes
fau
35
32 32
11 35

4000 lt 11
25

17 34 27
2
30 Pzgn1
63
Geologic Units:
18
AR 6/15/00-3
10 2.72 ± 0.07(b)
6

25 AR 6/15/00-5
Q2 AR 6/18/00-2
17
3.12 ± 0.07(b) Q1 Quaternary alluvial/fluvial deposits

Quaternary Units
21 14
9
31 Pzsch1 4.71 ± 0.10(m)
38 14 20

21
24
22
Qm Quaternary glacial moraines
6 61 5 Mzgr

Qm 4
Q2 Quaternary alluvial/fluvial deposits
38° 20′

Qm 10
8
Q2
10 AR 6/18/00-7 Qss Quaternary sandstones
38°
27
4.83 ± 0.09(m)
32
55 Pzgn1
Pzsch3
Q1
AR 6/13/00-1b
5000 Pzsch1 Paleozoic upper-amphibolite facies schists.
6.20 ± 0.13(m)
00

Paleozoic metagraywacke. Lower greenschist

40

AR 8/31/03-5 area unmapped

8
AR 8/31/03-3 500 Pzgw facies schists and calcareous schists, with minor
4 0 0 0
60 marble and graphitic schist.

Metasedimentary Units
27
10
25 14
7184 6088 Q2
Pzgn1 Paleozoic quartzofeldspathic gneisses.
Pzgn1
Pzsch3
Qm
Paleozoic upper-amphibolite to granulite facies
metasediments. Structurally lower portions
Pzsch3

Pzsch3
7509
Muztaghata dominated by pelitic schists. Structurally higher
portions dominated by marble, calc-silicates, and
7277 mafic schist.
Qm
Pzsch4 Paleozoic upper greenschist facies schists and
AR 9/1/03-1 mafic schists.
Glacie
7
25 Pzsch3 r Paleozoic upper greenschist to upper amphibolite
Pzsch5
37 facies schists and gneisses.
6 19 7028
Q2

Plutonic Units
Mgr Miocene biotite granite
24 6000
11 71
Esy Eocene diopside bearing syenite
26
Pzgn1 6841
5000
Mzgr Triassic-Early Jurassic granites and granodiorites
18
21 19
38
8
17
36 Glacier 6355
50
00
5144
Map Symbols:
Qm
Kongur Shan normal fault

Ko 6108 5000
Normal faults associated with
Qm n gu AR 6/29/00-3 AR 6/29/00-4 Qm the Kongur Shan extensional system

rS Qm
8.03 ± 0.21(b) 7.50 ± 0.13(b)
Normal faults not associated with
AR 4/27/00-10
11.2 ± 0.2(b) ha 65
the Kongur Shan extensional system
n
fa
AY 8/21/99-8 Pzgn1 56

u
58
5358 50 Mzgr
lt
50
57 AR 6/22/00-7
Thrust fault
78
Qm 60 8.36 ± 0.18(b)
66
AR 6/25/00-3 y 40
lle
AR 6/29/00-1
0 ± 0.13(b) 55 8.07 ± 0.14(b) AR 6/22/00-6 Approximate or inferred fault contact
40
07.87 56
va 7.42 ± 0.26(b)
Q2 54
su 58

AR 6/25/00-1 an area unmapped Buried fault contact

8.60 ± 0.25(b)
e nm 60
g 52 85
T55u 5384 Strike and dip of foliation and
63 85 trend and plunge of lineations
AR 6/22/00-6
8.05 ± 0.19(b) Pzgn1
36
AR 6/22/00-6
AR 6/30/00-10 Sample locailty, station number, and weighted
9.42 ± 0.94(b)
Mgr 10.1 ± 0.2(b) mean 40Ar/39Ar age (b = biotite, m = muscovite)

47 72 Lakes and rivers

15
62 82
AR 6/23/00-2 86 78
8.02 ± 0.17(b)
66 Glaciers

AY 8/23/99-3 33 84
6
74 5148
0
80
38° 00′

AR 6/24/00-7
38° 00′

Q1 54
7.74 ± 0.30(b)
5000

75
AR 6/24/00-3
76 8.38 ± 0.23(b)

Tahman 31
10 25
area unmapped
Kongur Shan

46
48
32 58
52 79
18
44 84
19 15 31 49 AR 6/30/00-10
ult

54 AR 6/28/00-9 9.42 ± 0.94(b)

45 8.32 ± 0.17(b) 84
38 Pzgn1
fa

87 65
65 40
66 67
AR 7/8/00-7
an

51
40 61 8.51 ± 0.14(m) 52 4000 23
51 69
m

8.55 ± 0.14(b) 46
75 44
AR 9/6/03-3
Tah

AR 7/6/00-10
4000
77 50
fault

28
Q
62
47 44 AR 9/6/03-5 Pzgn1 8.08 ± 0.16(b)
1 55
AR 7/8/00-5
76 23 00
Mgr
9.20 ± 0.19(b) 76
42
45 50
Ta

20
AR 7/1/00-2 60 73 87 75 61
Pzsch3
46 8.47 ± 0.17(b) AR 7/1/00-6
sh

44 72
AR 7/8/00-9 40 8.27 ± 0.13(b) AR 7/7/00-7 75
ko

9.18 ± 0.17(m) AR 7/2/00-8 35

48 65 53 7.93 ± 0.18(b)
17 10.3 ± 0.2(b) 66 75
r

46
ga

38 74 72
45 61 68 77 AR 7/4/00-5
29 33 37 86
0
nf

AR 7/8/00-12 48 40 8.08 ± 0.14(m) 71

400

28 79 49
13.7 ± 0.2(b) 10 85
67 63 AR 7/6/00-1
au

41 Pzsch3 48
24
20
Q2 7.46 ± 0.30(b) 55
55 66
28
lt

18 10 71 52 AR 7/4/00-1
0 84
Pzsch3 56 72 82 8.35 ± 0.23(b) 74
17 69
28 54 60
58
Q2
45 25
80
72
37
59 32
87
32
77
Shen-ti
69 34 12
38 65 34

Shen-ti fault
65

N
38 72 30 47 62 15
13 61
37 33
27 42
18 Pzsch4 28
32
Q2 36 38 33 4000
23 37
area unmapped

18
31 Torbashi fault 33
83

64 Q1
AR 4/28/00-8
0
500 102 ± 2(m)
84 Pzsch5 120 ± 2(b) 54
34
?
Ta
sh

17 82
ko

30

2
r

53 AR 4/28/00-9
ga

00

107 ± 2(b)
n

40

Q2
Ri

40
00

Pzsch4
ve

43 W
uq
r

4 62

Shen-ti fault Tashkorgan 44 a

Ri
ve
r
Q1
Tash

5000
0 5 10
4000

Contour Interval: 200 m

korg

Esy 62 Pzsch5
Qss
12

Q1
a

Mzgr
n fau

AR 4/2700-11 Q
ss
56
123 ± 3(b)
64

75° 00′ 75° 30′

lt

75

Figure 5. (A) Geologic map of the southern Kongur Shan normal fault. (B) Lower-hemisphere equal-area stereonet showing foliation planes (black lines), poles to planes (blue
crosses), and lineations (red squares) from the Muztaghata massif.