Miocene Exhumation of the Pamir Revealed by Detrital Geothermochronology of Tajik Rivers C

TECTONICS, VOL. 31, TC2014, doi:10.1029/2011TC003040, 2012

Miocene exhumation of the Pamir revealed by detrital geothermochronology of Tajik rivers C. E. Lukens,1 B. Carrapa,1,2 B. S. Singer,3 and G. Gehrels2 Received 4 October 2011; revised 6 February 2012; accepted 26 February 2012; published 18 April 2012.

[1] The Pamir mountains are the western continuation of the Tibetan-Himalayan system, the largest and highest orogenic system on Earth. Detrital geothermochronology applied to modern river sands from the western Pamir of Tajikistan records the history of sediment source crystallization, cooling, and exhumation. This provides important information on the timing of tectonic processes, relief formation, and erosion during orogenesis. U-Pb geochronology of detrital zircons and 40Ar/39Ar thermochronology of white micas from five rivers draining distinct tectonic terranes in the western Pamir document Paleozoic through Cenozoic crystallization ages and a Miocene (13–21 Ma) cooling signal. Detrital zircon U-Pb ages show Proterozoic through Cenozoic ages and affinity with Asian rocks in Tibet. The detrital 40Ar/39Ar data set documents deep and regional exhumation of the Pamir mountains >30 Myr after Indo-Asia collision, which is best explained with widespread erosion of metamorphic domes. This exhumation signal coincides with deposition of over 6 km of conglomerates in the adjacent foreland, documenting high subsidence, sedimentation, and regional exhumation in the region. Our data are consistent with a high relief landscape and orogen-wide exhumation at 13–21 Ma and correlate with the timing of exhumation of the Pamir gneiss domes. This exhumation is younger in the Pamir than that observed in neighboring Tibet and is consistent with higher magnitude Cenozoic deformation and shortening in this part of the orogenic system. Citation: Lukens, C. E., B. Carrapa, B. S. Singer, and G. Gehrels (2012), Miocene exhumation of the Pamir revealed by detrital geothermochronology of Tajik rivers, Tectonics, 31, TC2014, doi:10.1029/2011TC003040.

1. Introduction 2000; Schwab et al.,2004;Kapp et al.,2007;Robinson et al., 2007] and host Cenozoic magmatic rocks [e.g., Schwab et al., [2] The Pamir Mountains of Tajikistan and China are part 2004; Chu et al.,2006;Wang et al., 2008] and local exten- of the greater Himalayan-Tibetan orogenic system and com- sional features [e.g., Hubbard et al.,1999;Kapp and Guynn, prise some of the highest mountains on Earth (Figure 1a). 2004; Amidon and Hynek, 2010]. However, the Pamir, which The dramatic relief and altitude observed in the Himalayas currently lie well within the Asian continent, over 250 km and Tibet have been largely associated with processes related north of the western syntaxis (Figure 1a), may have a very to pre- and syn-collisional shortening combined with Ceno- different tectonic history than those of Tibet and the Hima- zoic erosion [Avouac and Tapponier,1993;Hodges,2000; laya. Several authors have attempted to correlate tectonic Yin and Harrison, 2000; DeCelles et al.,2002;Kapp et al., terranes in the Pamir with those in Tibet [Yin and Harrison, 2007; Ritts et al., 2008]. In contrast, very little is known 2000; Schwab et al., 2004; Robinson et al.,2007;Robinson, about the timing and processes responsible for deformation, 2009], based on the age and character of magmatic belts exhumation and uplift of the Pamir. In particular, it is unclear and offset of individual local units across strike-slip faults. if the Pamir result directly from Indo-Asia collision or if they Currently, a single correlation model does not exist and along- are the result of significantly different, younger geodynamic strike correlation remains controversial. and erosional processes. Parallels have been drawn between [3] The Pamir differ from the Himalaya and Tibet in the the Pamir and Tibet based on the fact that both comprise areal extent of metamorphic domes [Dmitriev, 1976; several Paleozoic-Mesozoic accreted terranes [Harrison et al., Pospelov and Sigachev, 1987; Burtman and Molnar, 1993]. The Pamir domes are larger than those documented in Tibet, 1Department of Geology and Geophysics, University of Wyoming, they are Cenozoic in age, and have yielded predominantly Laramie, Wyoming, USA. Miocene exhumation ages [Hubbard et al., 1999; Robinson 2Department of Geosciences, University of Arizona, Tucson, Arizona, et al., 2007]. Previous workers have suggested that relief USA. (and related high rates of exhumation) developed in the 3Department of Geoscience, University of Wisconsin-Madison, Madison, Wisconsin, USA. southern Pamir after 11 Ma (based on xenoliths and thermal modeling [Ducea et al., 2003]) and, conversely, during Copyright 2012 by the American Geophysical Union. early mid Miocene time (based on the sedimentary record in 0278-7407/12/2011TC003040 the Alai Valley [Hamburger et al., 1992; Coutand et al.,

TC2014 1of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 1. (a) Location map; white box is the approximate area in Figure 1b. (b) Map of the western Pamir, showing major tectonic boundaries, surface geology, and locations of gneiss domes [Burtman and Molnar, 1993; Schwab et al., 2004; Hacker, personal communication, 2011]. See section 2.1 for additional description of rock types within each tectonic terrane.

2002] (Figure 1a). A few thermochronologic studies have and differences in tectonic and exhumation history between suggested early mid Miocene exhumation on a local scale the Pamir and Tibet and to gain new insights on possible [Hubbard et al., 1999; Amidon and Hynek, 2010], but fundamental differences in the geodynamics of the two no regionally extensive sampling has previously been regions. To determine the timing and distribution of source executed to determine the exhumation history of the Pamir, area crystallization and exhumation, we here use the detrital and thus the timing of orogenic-scale exhumation is geochronological and thermochronological signal recorded by not constrained. detrital minerals (zircon U-Pb and white mica 40Ar/39Ar) [4] Resolving both the crystallization and exhumation sig- derived from five major rivers draining the western Pamir nal in the Pamir is important in order to resolve similarities (Figures 1b and 2). Our new data set also contributes to

2of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 2. Map of the western Pamir, showing the areal extent of watersheds above each sample point. Tectonic boundaries are also shown (as in Figure 1) for reference. further constraining along-strike affinity between terranes in [6] The rocks of the Central Pamir record the presence of a the Pamir and in Tibet. small ocean basin in Mesozoic time. Deep marine sedimentary rocks, pillow basalts, and tuff of late Paleozoic to 2. Regional Setting Jurassic age were deposited [Pashkov and Shvol’man, 1979; Burtman and Molnar, 1993], and the basin closed in late 2.1. Geologic History of the Pamir Jurassic to Cretaceous time [Burtman and Molnar, 1993]. [5] The Pamir are generally divided into three distinct A thick Jurassic-Cretaceous sequence is today preserved in tectonic terranes: the Northern, Central, and Southern Pamir the Tajik depression [Hamburger et al., 1992; Nikolaev,2002] [e.g., Burtman and Molnar, 1993]. In the Northern Pamir, and constitutes the erosional and depositional record of these the closure of a large ocean basin in late Carboniferous time processes. Arc-type granitoids intruded into the Triassic is documented by the presence of island arc igneous rocks Rushan-Pshart marine basin sequence at 190–160 Ma may be [Pospelov and Sigachev, 1987; Burtman and Molnar, 1993], associated with subduction along the Rushan-Pshart suture marine carbonates, and conglomerates [Pospelov, 1987; and closure of the basin [Schwab et al., 2004] (Figure 1b). The Burtman and Molnar, 1993]. The suture associated with this Mesozoic basin evidenced in the Central Pamir is similar to basin closure has been correlated to the Kunlun suture in small basins documented in Afghanistan and in northwest Tibet [Schwab et al., 2004], and represents the suturing of Tibet, but there is no evidence that these basins were the Northern Pamir onto the Asian continent [Burtman and connected, and it is likely that several separate small basins Molnar, 1993]. Volcanic rocks exposed in the northern existed at this time [Burtman and Molnar,1993]. – Pamir from the Kunlun arc are dated at 370 320 Ma [7] Farther south, the Indus-Tsangpo and Shyok sutures [Schwab et al., 2004]. (Figures 1b and 3) and associated ophiolite belts most likely

3of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 3. Regional tectonic map, after Burtman and Molnar [1993]. document the collision of India with Asia in early Cenozoic 350 km [Burtman and Molnar, 1993; Schmidt et al., 2011]. time [Burtman and Molnar, 1993]. Exposed bedrock geol- High rates of convergence continue today, and geodetic ogy in the Southern Pamir includes Precambrian and Paleo- surveys have suggested that convergence across the northern zoic metamorphic rocks, Triassic, Cretaceous, and Paleogene Pamir and Alai Valley is as much as 21 mm/yr [Bazhenov and granites and diorites, and remnants of Carboniferous through Burtman, 1990; Reigber et al., 2001]. Prior to about 29 Ma, Paleogene sedimentary rock [Pashkov and Budanov, 1990; the Tarim Basin and Tajik Depression were connected by a Burtman and Molnar, 1993]. marine basin, of which the sedimentary record is preserved in [8] Correlation between terranes across Tibet and the the Alai Valley (Figure 1a). This basin began to close around Pamir is complicated by the ongoing controversy over dis- the Oligocene-Miocene boundary [Leith,1982;Hamburger placement along the Karakorum fault and related structures, et al.,1992;Coutand et al., 2002; Nikolaev, 2002], when a which separate the two regions. The magnitude of slip wedge of syn-tectonic Neogene conglomerates and coarse along the Karakorum strike-slip fault ranges from 65 km sediment was deposited [Hendrix et al., 1994; Sobel and [Murphy et al., 2000] to as much as 1000 km [Peltzer and Dumitru, 1997; Yin et al., 1999]. Tapponnier, 1988]. Recent studies have supported more modest amounts of displacement. The correlation of anti- 2.2. Metamorphic Domes formal domes in the Pamir and the Qiantang terrane in Tibet [11] Across the central and southern Pamir, a series of by Schwab et al. [2004] suggest displacement of appro- Cenozoic metamorphic domes have been recognized, termed ximately 250 km, whereas the correlation of the Aghil the Pamir gneiss domes [e.g., Ratschbacher et al., 1997; formation by Robinson [2009] revises that number to 149– Hubbard et al., 1999; Schwab et al., 2004; Robinson et al., 167 km of offset since Late Jurassic time. 2007]. In the central Pamir, the domes are largely com- [9] Under the correlation of Robinson [2009], the South- posed of amphibolite grade rocks, but also contain leuco- ern Pamir correlate to the Qiangtang terrane in Tibet, and the granites [Schwab et al., 2004], and have yielded early Northern Pamir correlate with the Sangpan-Garze terrane in Miocene cooling ages from 40Ar/39Ar dating on biotite, central Tibet (Figures 1b and 3). Between the two correlated muscovite, and hornblende [Schwab et al., 2004]. In the terranes lies the Central Pamir terrane, which has no Tibetan eastern Pamir, Robinson et al. [2007] correlated high-grade equivalent [Burtman and Molnar, 1993; Robinson, 2009]. schists and gneisses to rocks found in the Central Pamir Alternatively, Schwab et al. [2004] correlate the Southern using field relations. Deformation in these rocks was dated Pamir with the Lhasa terrane, the Central Pamir with the as late Oligocene to early Miocene using monazite inclu- Qiangtang terrane, and the Northern Pamir with the Songpan- sions in garnet, and exhumation to shallow crustal depths Garze terrane. was constrained to 7.5–9 Ma using 40Ar/39Ar dating of [10] Some north-south striking extensional faults have biotite [Robinson et al., 2007]. In the southern Pamir (farther been mapped in the northeastern Pamir, but few extensional south than the area sampled in this study), Hubbard et al. structures have been documented in the western Pamir. [1999] described the “whiteschists” of the Goran series, The Cenozoic history of the region is dominated by large which are of upper amphibolite grade. 39Ar/40Ar dating of amounts of north-south shortening. Shortening has been biotite, phlogopite, and hornblende in these rocks shows accommodated by northward movement of the entire Pamir Miocene cooling (16.9 Æ 0.3 Ma for phlogopite, 18.3 Æ block by at least 300 km [Burtman and Molnar, 1993], by 0.2 Ma for biotite). Domes have been recognized in every strike-slip faulting along the eastern and western Pamir on watershed sampled in this study, and in some cases contribute the edge of the Tarim basin and Tajik Depression [Burtman large portions of exposed surface geology (Figure 1b). and Molnar, 1993; Sobel and Dumitru, 1997; Sobel et al., [12] Exhumation of deep crustal rocks in the region during 2006] (Figure 3), and by internal shortening of at least late Miocene time has been documented by several authors

4of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014 based on the presence of deeply metamorphosed rocks and [16] Structural data from the Alai Valley, which represents xenoliths in ultra-potassic igneous rocks in the southern the last remain of a basin once connecting the Tajik Pamir [Hubbard et al., 1999; Ducea et al., 2003; Hacker depression to the Tarim basin, indicate that deformation et al., 2005; Robinson et al., 2007]. Xenoliths that experi- started in the Oligocene [Coutand et al., 2002] and is still enced temperatures of 1050–1200C and pressures of 2.4– active today [Arrowsmith and Strecker, 1999]. To the west, 2.7 GPa yield well-constrained eruption ages of 11 Ma, where instead the Pamir and the Tien Shan are today in based on 40Ar/39Ar dating of K-feldspar, groundmass, contact, collision has been proposed to have occurred since and biotite [Ducea et al., 2003]. The presence of metamor- the Miocene-Pliocene [Pavlis et al., 1997]. phic domes in the Pamir has been used to constrain the [17] Intracontinental subduction is supported by seismic timing of high-pressure metamorphism and subsequent tomography, which images a cold south-dipping slab under exhumation of these rocks to the surface at 57–11 Ma the Pamir [Negredo et al., 2007]. Shallow- and intermediate- [Hubbard et al., 1999; Hacker et al., 2005]. Workers have depth seismicity also follow a south-dipping band, and generally interpreted Miocene cooling ages to reflect tec- suggest that a rigid body must be present at depth in which tonic exhumation due to local extension [Dmitriev, 1976; earthquakes occur [Burtman and Molnar, 1993]. The high Ratschbacher et al., 1997; Hubbard et al., 1999; Schwab rates of convergence and large amounts of shortening across et al., 2004; Robinson et al., 2007], although the larger the Alai Valley [Reigber et al., 2001] are consistent with geodynamic processes controlling dome formation and exhu- intracontinental subduction. mation are poorly understood. [18] Subduction of thinned continental crust and/or the removal of the upper crust can result in upper crustal short- 2.3. Tectonic Models for Exhumation in the Pamir ening accommodated by the accretion or underplating [13] The presence of metamorphic domes and xenoliths in of sedimentary cover (and/or of the underlying basement the Pamir [Ducea et al., 2003; Hacker et al., 2005] requires a rocks) onto adjacent continental crust, and by internal crustal mechanism able to exhume rocks from depth in excess of thickening and thrusting [Molnar and Gray, 1979; Capitanio 10km. Metamorphic domes, characterized by Miocene et al., 2010]. Miocene exhumation in the Tien Shan [e.g., thermochronological ages, are widespread within the Pamir Sobel et al., 2006] supports high magnitude deformation [Schwab et al., 2004; Hacker et al., 2005] but the processes driven exhumation in the Cenozoic. Apatite fission track responsible for their formation and exhumation remain ages from the eastern Pamir also support a Miocene age of largely unresolved. exhumation and suggest that deformation has been active [14] Several different models have been proposed for the since 20 Ma [Sobel and Dumitru, 1997]. Cenozoic tectonic evolution of the Pamir, including exhumation via local extensional structures directly resulting 3. A Geothermochronologic Approach to Resolve from Indo-Asia collision, shortening related to the collision Exhumation and Tectonics in the Pamir with the Tien Shan [Sobel and Dumitru, 1997], and following intracontinental subduction of thinned Asian crust [19] In order to address the timing and distribution of [Burtman and Molnar, 1993; Negredo et al., 2007]. In the exhumation in the Pamir on an orogen-wide scale, a detrital eastern Pamir, tectonic exhumation along extensional sys- method is here applied. The distribution of detrital cooling tems such as the Konghur Shan was responsible for signifi- and crystallization ages has been used successfully in several cant exhumation of schists and gneisses in the late Miocene studies for provenance discrimination and to determine [Robinson et al., 2004, 2007]. However, few extensional timing of major tectonics events [e.g., Copeland and systems have been mapped in the western and high (central) Harrison, 1990; Najman et al., 1997; White et al., 2002; Pamir that could accommodate similar exhumation. Carrapa et al., 2003; Gehrels et al., 2003; DeCelles et al., [15] The exhumation of the Tien Shan, which lie north of 2007; Ruhl and Hodges, 2005; Carrapa, 2010]. In particu- the Pamir and the Alai Valley (Figure 1), has been associated lar, detrital geothermochronology of river sands has proven with Indo-Asia collision. Exhumation from depths of 2– to well represent the regional signature of the source areas 5 km is recorded in the Tien Shan by apatite fission track [e.g., Carrapa et al., 2004]. ages at 13–26 Ma, generally youngs southward [Sobel and [20] Sand samples from five major modern rivers draining Dumitru, 1997; Sobel et al., 2006]. Several authors have the western side of the Pamir in Tajikistan (Figure 2) were 40 39 advocated the presence of an intracontinental subduction collected and analyzed using standard white mica Ar/ Ar zone along the Alai Valley, to the north of the Pamir, where thermochronologic and zircon U-Pb geochronologic meth- a south-dipping slab of Asian crust may currently be sub- ods [Smith et al., 2003, 2006; Hodges, 2004; Gehrels et al., ducted to depths of more than 600 km [Burtman and 2006, 2008] (see Text S1 in the auxiliary material for 1 Molnar, 1993; Negredo et al., 2007]. Intracontinental sub- details). In general, one hundred grains [Vermeesch, 2004] duction along this south-dipping subduction zone and sub- were targeted for analyses if available. However, due to poor sequent collision between the thickened Pamir and Tien apatite and white mica mineral yields, less than one hundred Shan may have accommodated 300 km of northward analyses per sample are here presented. Detrital zircon U-Pb indentation of the Pamir salient along the south dipping ages provide the timing of crystallization (or metamorphic Main Pamir Thrust and Pamir Frontal thrusts (Figure 1b) recrystallization) of different source rocks [e.g., Gehrels [Sobel and Dumitru, 1997; Hendrix et al., 1994; Coutand et al., 2003] and thus help to discriminate between et al., 2002]. Such indentation has been partly accommodated by strike-slip movements along western and eastern 1Auxiliary material data sets are available at ftp://ftp.agu.org/apend/tc/ structures (e.g., Karakoram and Kashgar-Yecheng transfer 2011tc003040/. Other auxiliary material files are in the HTML. doi:10.1029/ system to the east) [Sobel et al., 2011]. 2011tc003040.

5of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 4. Summary of geothermochronologic data juxtaposed against U-Pb (zircon) ages from India and Asia. Curves for 40Ar/39Ar and U-Pb in both the larger figure and inset are from this study. Curves for Asian and Indian Plate affinity utilize data from the compilation of Henderson et al. [2010]. Inset includes all data acquired in this study younger than 1 Ga; note logarithmic horizontal (time) scale. different sources, whereas detrital white mica 40Ar/39Ar and are calculated using a weighted mean. Calculations were thermochronology provides the timing of cooling (through performed using AgePick. – the 350 420 C isotherm; [McDougall and Harrison, 40 39 1999], assuming that source rocks are either igneous or 3.1. U-Pb and Ar/ Ar Age Distributions reached peak temperatures higher than the closure tempera- of the Vanch River ture during metamorphism) and exhumation of presumably [23] The Vanch River (TJK-08) (Figure 2) is the north- the same sources. The area sampled covers the western side ernmost river sampled and drains the western slope of the of the Pamir from north of the Kunlun suture to south of the central Pamir, with source rocks including Cenozoic mag- Rushan-Pshart suture (Figures 1b and 2) and include the matic and volcanic rocks [Schwab et al., 2004], part of the North (TJK08 and part of TJK04), Central (TJK07, and parts Yazgulom gneiss dome (B. Hacker, personal communica- of TJK04 and TJK06) and Southern Pamir (TJK05 and part tion, 2011), and older sedimentary rocks from the Northern of TJK06) (Figures 1b and 2). and Central Pamir [Aghanabati, 1986]. [21] Detrital U-Pb geochronologic analyses were per- [24] Ninety-eight grains from sample TJK-08 were ana- formed using LA-MC-ICMPS in the LaserChron center at lyzed for U-Pb geochronology. Ages range from range from the University of Arizona according to the methods of 34.9 Æ 1.8 Ma to 3178.7 Æ 24.9 Ma. Most ages fall into one Gehrels et al. [2008]. Between forty and ninety-eight grains large age peak (n = 68) at 40 Ma (Figure 4), and 5 grains per sample yielded sufficiently precise and concordant ages are older than 1 Ga. Sixty-nine grains were analyzed (Data Set S1 in the auxiliary material). U-Pb zircon ages for 40Ar/39Ar thermochronology. Ages range from 32.71 Æ range from 32.6 Æ 3.4 Ma to 3485.4 Æ 30.5 Ma, with dis- 1.42 Ma to 348.18 Æ 3.94 Ma. Peaks are at 33 Ma (n = 5), tinct peaks in age spectra at 36–42 Ma and 98–104 Ma. 40 Ma (n = 10), and several peaks of n = 7–10 between As a whole, U-Pb ages indicate Asian crust affinity 231 Ma and 280 Ma (Figure 5a). [Henderson et al., 2010] (Figure 4), and reflect sources in 40 39 3.2. U-Pb and Ar/ Ar Age Distributions both plutonic (Mesozoic and Cenozoic), metasedimentary of the Murghab River (Paleozoic and Proterozoic), and possibly volcanic or metavolcanic (Cenozoic) rocks (Figure 1b). The full U-Pb [25] The Murghab River (TJK-04) (Figure 2) drains an data set can be found in the supplementary material (Data area with >2 km of relief, deep in the high Pamir. Source Set S1). rocks include the Muskol Dome, Cenozoic magmatic (gra- [22] Between 54 and 85 white mica grains from each of nitic) rocks [Schwab et al., 2004], and marine sedimentary the five samples, for a total of 334 grains, were analyzed by rocks of Carboniferous-Jurassic age from both the North and single-grain total fusion 40Ar/39Ar analysis at the Rare Gas Central Pamir. Geochronology Laboratory at the University of Wisconsin, [26] Forty grains from sample TJK-04 were analyzed for Æ according to the method outlined by Smith et al. [2003, zircon U-Pb geochronology. Ages range from 32.6 Æ 2006]. Ages range from 13.57 Æ 0.55 Ma to 348 Æ 4 Ma, 3.2 Ma to 3118.7 107.1 Ma. Major age peaks are present with distinct age peaks at 15–21 Ma, 26–29 Ma, and at 37 Ma (n = 10) and 540 Ma (n = 6) (Figure 5b). Nine 162–185 Ma (Figure 4 and Data Set S1). Age spectra grains are older than 1 Ga. [27] Fifty-four grains from sample TJK-04 were anal- (peaks) for individual samples are reported as the maximum 40 39 age probability within a group of more than three analyses, yzed for Ar/ Ar thermochronology. Ages range from

6of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 5. (a–e) Summary of geochronologic and thermochronologic data for grains younger than 1 Ga from each watershed sampled. Note the youngest age peak in Figures 5b–5e, which all fall between 13 and 21 Ma.

15.59 Æ 1.55 Ma to 21.14 Æ 1.54 Ma. One age peak dom- 65 Ma (n = 3), and the largest peak is at 614 Ma (n = 5). inates at 17 Ma (n = 52) (Figure 5b). Nine grains are older than 1 Ga (Figure 5c). Eighty-four 40 39 40 39 grains were analyzed for Ar/ Ar thermochronology. Ages 3.3. U-Pb and Ar/ Ar Age Distributions range from 17.82 Æ 0.75 Ma to 36.57 Æ 0.43 Ma. Peaks are of the Yazgulom River at 19 Ma (n = 12), and 27 Ma (n = 53) (Figure 5c). 28 [ ] The Yazgulom River (TJK-07) drains an area of high 40 39 relief on the western slope of the central Pamir (Figure 2). 3.4. U-Pb and Ar/ Ar Age Distributions Source rocks include Cretaceous and Cenozoic magmatic of the Bartang River rocks, Paleozoic sedimentary rocks of the Central Pamir, and [30] The Bartang River (TJK-06) drains a very large area the Yazgulom Dome [Aghanabati, 1986; Schwab et al., 2004]. (Figure 2), which covers much of the high terrain in the [29] Forty-four grains from sample TJK-07 were analyzed central and southern Pamir as well as the western slope for U-Pb zircon geochronology; ages range from 39.8 Æ of the mountain range. The drainage area includes the 0.9 Ma to 3485.4 Æ 30.5 Ma. The youngest age peak is at Shakhdara, Muskol, and Sarez domes. Rocks outcropping

7of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014 within this watershed also include Mesozoic and Cenozoic 4.2. Exhumation History of the Pamir 40 39 magmatic rocks, Paleozoic sedimentary rocks, and Paleozoic [37] When comparing zircon U-Pb and mica Ar/ Ar basement rocks from parts of the Northern, Central, and ages it is possible to differentiate crystallization from cool- Southern Pamir [Aghanabati, 1986; Schwab et al., 2004; ing (exhumation) history. In the northernmost sample, from Robinson et al., 2007]. the Vanch River, the youngest U-Pb zircon ages (34.9 Æ [31] Eighty-five grains from sample TJK-06 were ana- 1.8 Ma to 42.4 Æ 2.4 Ma) and 40Ar/39Ar ages (32.7 Æ lyzed for zircon U-Pb geochronology. Ages range from 1.42 Ma to 42.0 Æ 1.2 Ma) are indistinguishable from each Æ Æ ≤ 34.2 0.2 Ma to 2795.0 56.5 Ma. Small age peaks (n 8) other (Figure 5a). These ages are best explained to represent are present at 39 Ma, 104 Ma, 549 Ma, 559 Ma, crystallization of Cenozoic plutons rather than exhumation. 645 Ma, 802 Ma, and 839 Ma. The largest age peak This interpretation is supported by the work of Schwab et al. is at 972 Ma (n = 12), and twenty-four grains are older [2004], who mapped plutonic gneisses and volcanic rocks than 1 Ga (Figure 5d). 40 39 of unspecified Cenozoic age within the same structural ter- [32] Sixty-nine grains were analyzed for Ar/ Ar ther- rane. Our new data thus put a new constraint on the age of Æ Æ mochronology. Ages range from 13.57 0.55 Ma to 204 these rocks. The remaining four samples (from the Murghab, 6.82 Ma, with five grains between 14 Ma and 23 Ma, Gunt, Bartang, and Yazgulom Rivers), however, have and 51 grains between 131 Ma and 199 Ma (Figure 5d). distinct lags of ≥15 Myr between the youngest zircon U-Pb Age peaks are numerous and contain less than five grains, age and the youngest 40Ar/39Ar age (Figures 5b–5e). This and thus are not reported individually here. The full data set suggests that the 13–21 Ma youngest 40Ar/39Ar age peak is available in the supplementary material (Data Set S1). (Figure 5) represent exhumation ages rather than a plutonic 40 39 or magmatic signal. Given that all the rivers sampled drain 3.5. UPb and Ar/ Ar Age Distributions 40 39 of the Gunt River metamorphic domes, the young Ar/ Ar age component is best explained to represent exhumation of these domes in the [33] The Gunt River (TJK-05) drains an area of high relief Miocene. These metamorphic domes have been described in southern Pamir (Figure 1b), and is sourced in Cretaceous as amphibolite grade and leucocratic [Schwab et al., 2004], and Cenozoic magmatic rocks and Paleozoic sedimentary and are a likely source of white micas. rocks [Aghanabati, 1986; Robinson et al., 2007]. The 40 39 [38] Overall, our new Ar/ Ar data indicate a remarkably drained area includes the Shakhdara dome. widespread early mid Miocene thermochronological signal [34] Eighty-seven grains from sample TJK-05 were ana- from four different major rivers. Cooling through the 350– lyzed for U-Pb zircon geochronology. Ages range from 40 39 Æ Æ 420 C isotherm ( Ar/ Ar closure temperature window) 98.4 3.8 Ma to 2123.3 28.0 Ma. One large age peak [McDougall and Harrison, 1999] for a linear geothermal (n = 33) is present at 102 Ma (Figure 5e), and two grains are gradient of ≥25C corresponds to >11 km of crust removal older than 1 Ga. 40 39 during the Miocene. This is supported by the presence of [35] Forty grains were analyzed for Ar/ Ar thermo- >6 km-thick coarse alluvial deposits (e.g., Tavildara con- chronology. Ages range from 14.19 Æ 1.25 Ma to 104.97 Æ glomerate, Figure 7) of Miocene age in the adjacent Tajik 4.03 Ma, with distinct age peaks at 16 Ma (n = 40) and Depression [Nikolaev, 2002] and Alai Valley [Coutand 27 Ma (n = 11). Only one grain is older than 39 Ma, Æ et al., 2002]. This coarse detritus shed from the Pamir at 104.97 4.03 (Figure 5e). represents a large volume of material removed by erosion from proximal, high-relief hinterlands at this time. Subsi- 4. Discussion dence analysis and isopach data of the Tajik Depression 4.1. Pamir-Tibet Terrane Correlation [Leith, 1985; Nikolaev, 2002] support thermochronologic data, documenting a sharp increase in sediment accommo- [36] Our new detrital zircon U-Pb ages generally resemble – dation in the Tajik Depression, and thus tectonic loading and those of Tibet, with a large percentage of Cenozoic (40 thickening of the orogen, during Miocene time. 55 Ma), Mesozoic (100–250 Ma) and ages that are generally 40 39 [39] The widespread 13–21 Ma Ar/ Ar signal indi- younger than 1 Ga [Murphy et al., 1997; Chu et al., 2006; cates deeper exhumation in the Pamir with respect to Tibet, Leier et al., 2007a, 2007b, Wang et al., 2008; Wen et al., where Cenozoic ages recorded by 40Ar/39Ar and fission 2008]. The Pamir data set contains an additional age spec- track thermochronology are limited [e.g., Guynn et al., 2006; tra at 36–42 Ma that is likely sourced in the Cenozoic vol- Rohrmann et al., 2011]. Very slow regional exhumation canic and plutonic rocks recognized by Schwab et al. [2004] characterizes Tibet since the Cretaceous [Kapp and Guynn, (Figure 1b). Comparison to Tibetan age distributions 2004], resulting in the widespread preservation of Meso- (Figure 6) shows a strong similarity to ages from Qiangtang zoic and Cenozoic sedimentary rocks and only localized terrane lavas (36–52 Ma) [Wang et al., 2008], and to exposures of deeper crustal rocks [Kapp et al., 2000, 2007]. Gandese batholith ages from the Lhasa terrane (11–71 Ma, Remarkable differences also appear when the new data set is with the largest age peak at 52 Ma) [Ji et al., 2009]. This 40 39 compared to detrital Ar/ Ar ages from foreland basin supports the terrane correlations of Robinson [2009], who deposits in the Himalaya of India and Nepal [e.g., Szulc correlated the Qiangtang and Songpan-Garze terranes into et al., 2006; White et al., 2002]. Such deposits record a the Pamir, but does not exclude those of Schwab et al. 40 39 significant Miocene detrital white mica Ar/ Ar age com- [2004], who included the Lhasa terrane in correlations with ponent (15–20 Ma) that has been associated with activa- the southern Pamir. A few grains older than 1 Ga are present, tion of the Main Central Thrust and South Tibetan which reflect an unknown older source. More geochrono- Detachment [Hodges, 2000, and references therein]. How- logical work of this kind is needed in order to better resolve ever, the ages recorded by Miocene through Quaternary terrane correlations between the Pamir and Tibet.

8of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Figure 6. Summary of U-Pb ages from this study and from the Lhasa and Qiangtang terranes in Tibet Detrital samples are from Cretaceous deposits, and thus should not contain any grains younger than 65Ma. Data was complied from Leier et al. [2007a] (detrital samples), Wang et al. [2008] (Qiangtang lavas), Chu et al. [2006] (Lhasa granites and Gangdese batholith), Ji et al. [2009] (Gangdese), and Wen et al. [2008] (Gangdese). foreland basin deposits in the frontal Himalaya are >1000 Ma by Pavlis et al. [1997]. However, this would require sta- to <12 Ma [Szulc et al., 2006; White et al., 2002]. A similarly bility of several rivers over very long time scales, and is not wide spread of ages is observed in other collisional systems likely the sole mechanism to explain the magnitude of such as in the Alps, where the exhumation record is a result exhumation observed. of erosion of different nappes that underwent multiple [41] Geodynamic models suggest that extension could be tectono-thermal events [e.g., Carrapa et al.,2004]. the direct result of collision, which may produce late-stage [40] The detrital signal in the Pamir is derived from a large normal faults and exhumation of deep-crustal rocks as a area draining multiple terranes bounded by Paleozoic and result of the changing balance between downward-directed Mesozoic sutures and more recent (Cenozoic) structures, shear traction and upward-directed buoyancy forces as col- and thus a much wider spectra of ages was expected in lision progresses [Beaumont et al., 2009]. Depending on the the detrital record. The dominance of a Miocene signal initial convergence velocity, ultra-high pressure rocks may instead suggests provenance preferentially from the gneiss reach depths within 10 km of the surface within 11–7 Myr domes, and is here interpreted to document the timing of after collision [Beaumont et al., 2009]. Our new data are exhumation of these domes, which cover more than half the consistent with this model, including the exhumation of overall sampled area. The dominance of sources in the xenolith-bearing ultra-high-pressure rocks observed in the domes may also be influenced by focused erosion by rivers, Southern Pamir [Ducea et al., 2003; Hacker et al., 2005]. which may be preferentially eroding high elevation sources However, if extensional structures are indeed responsible in the core of the range. If the rivers have been stable for much of the Miocene dome exhumation in the Pamir, over long periods of time, they may have influenced the then more detailed mapping of the region is needed, as few exhumation of gneiss domes at the range core as suggested extensional structures are currently recognized.

9of12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

et al., 2007]. In this scenario, the initiation of intracontinental subduction would have occurred some time prior to 21 Ma and thinned continental crust would underlie the marine basin, which closed as subduction progressed until collision occurred between the relatively thicker crust of the Pamir and the Tien Shan. [44] Both of these processes are expected to result in upper crustal shortening and thickening of the Pamir in the early Miocene, and thus may explain enhanced erosion. Intra- continental subduction of Asian crust beginning in the early Miocene (25 Ma, when the Alai Valley began to close [Coutand et al., 2002]) is consistent with Miocene exhumation of the Pamir gneiss domes and enhanced foreland basin subsidence and deposition. [45] Regardless of the preferred exhumation scenario, it should be noted that these interpretations are based on a relatively small sample size, and thus would greatly benefit from additional study. Detrital studies of this kind typically aim for >100 dated grains [Vermeesch, 2004], and previous workers have suggested that sampling large rivers requires a larger data set (100 grains) in order to effectively char- acterize the area sampled [Ruhl and Hodges, 2005]. Smaller data sets have successfully been employed in characterizing the timing of exhumation in other orogens [e.g., Copeland and Harrison, 1990; Carrapa et al., 2004; Najman et al., 2005], and although our samples are small, they are likely to reflect the dominant age peak present in the source. The data set presented here represents a first step in understand- ing the orogen-scale exhumation of the Pamir and future work in this area is necessary in order to narrow the broad range of possible interpretations presented here.

5. Conclusions

[46] Our new U-Pb detrital zircon ages show an affinity to Figure 7. Photographs of the Miocene Tavildara conglom- age distributions from Tibet, and support the terrane corre- erate in the Tajik Depression. (a) Thickness of the conglom- lation of Robinson [2009], in which the northern Pamir erate in this area varies up to over 5000 m. (b) A closer look correlate with the Sangpan-Garze terrane and the southern at this poorly sorted conglomerate. Subangular clasts up to Pamir correlate with the Qiangtang terrane. However, ages 25 cm in diameter are predominantly limestone. Smaller, that match those found in the Lhasa terrane [Ji et al., 2009] well-rounded clasts of sedimentary and crystalline rocks are also present in samples from the Gunt and Bartang are also present, though less abundant. The conglomerate Rivers, in agreement of the correlation of the southern Pamir is well-cemented, with a muddy, calcitic cement. with the Lhasa terrane [Schwab et al., 2004]. 40 39 [47] The fact that the bulk of detrital white mica Ar/ Ar ages in the Pamir are early mid Miocene (13–21 Ma), [42] Alternatively, the Miocene signal recorded by detrital minerals of Tajik rivers may represent erosion of duplex suggests younger and higher magnitude exhumation than structures cored by gneiss domes, which were deeply incised that observed in Tibet. Widespread exhumation recorded by as a result of contractional deformation and increased relief our new data set combined with the presence of proximal during collision with the Tien Shan. Crustal thickening and and coarse-grained alluvial Miocene deposits in the Tajik depression requires relatively high relief and rapid removal subsequent exhumation may have resulted from a contrac- – tional regime across the Pamir and Tien Shan, resulting of several kilometers of material from the Pamir at 13 directly from Indo-Asian collision. In the Alai valley, which 21 Ma. This timing correlates with closure of the Alai Valley lies between the Pamir and Tien Shan, cessation of marine [Coutand et al., 2002] and collision with the Tien Shan deposition and initiation of intramontane deposition began at [Sobel and Dumitru, 1997; Sobel et al., 2006]. 25 Ma [Coutand et al., 2002]. The transition to terrestrial [48] Deep, extensive exhumation is consistent with a deposition at this time is consistent with the closure of a higher magnitude of shortening in the Pamir with respect marine Tajik-Tarim basin [Burtman and Molnar, 1993] and to Tibet, which resulted in upper crustal deformation, collision of the Pamir with the Tien Shan [Coutand et al., the exhumation of gneiss domes, and the rapid removal of 2002; Sobel et al., 2006]. crustal material from the hinterland. We suggest that this high magnitude, widespread exhumation is the result of [43] The closure of the marine Tajik-Tarim basin at this time is also consistent with models that include intraconti- geodynamic processes that differ significantly from those nental subduction [Burtman and Molnar, 1993; Negredo responsible for the development of the Tibetan-Himalayan

10 of 12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014 system, and are related to a pre-Neogene geological history Gehrels, G., V. Valencia, and A. Pullen (2006), Detrital zircon geochronol- that is unique to this region. ogy by laser-ablation multicollector ICPMS at the Arizona Laserchron Center, in Geochronology: Emerging Opportunities, Paleontol. Soc. [49] Widespread exhumation may have resulted from Pap., vol. 12, edited by T. D. Olszewski and W. D. Huff, pp. 67–76, increased shortening and crustal thickening following Indo- Paleontol. Soc., Boulder, Colo. Asia collision, or following subduction of an intraconti- Gehrels, G. E., V. Valencia, and J. Ruiz (2008), Enhanced precision, accu- – nental slab. Both of these models, or a combination of the racy, efficiency, and spatial resolution of U-Pb ages by laser ablation multicollector– inductively coupled plasma–mass spectrometry, Geochem. two, may explain the observed exhumation as well as the Geophys. Geosyst., 9, Q03017, doi:10.1029/2007GC001805. presence of gneiss domes in the Pamir, and further study is Guynn, J., P. Kapp, J. Volkmer, B. Carrapa, and M. Heizler (2006), warranted to resolve the geodynamics of the area. Minimal Cenozoic denudation of the central Tibetan Plateau revealed by low-temperature thermochronology, Eos Trans. AGU, 87(52), Fall Meet. Suppl., Abstract T43E-06. [50] Acknowledgments. This work was funded by a National Geographic exploratory grant and by the University of Wyoming. Hacker, B., P. Luffi, V. Lutkov, V. Minaev, L. Ratschbacher, T. Plank, Geochronology at the Arizona LaserChron Center is supported by NSF M. Ducea, A. Patiño-Douce, M. McWilliams, and J. Metcalf (2005), Near-ultrahigh pressure processing of continental crust: Miocene crustal EAR 0732436. We thank Brian Jicha and John Trimble for help with – sample analysis and data reduction. This manuscript benefited from xenoliths from the Pamir, J. Petrol., 46, 1661 1687, doi:10.1093/petrology/ egi030. thoughtful comments from Peter Molnar and Peter DeCelles. Hamburger, M. W., D. R. Sarewitz, T. L. Pavlis, and G. A. Popandopulo (1992), Structural and seismic evidence for intracontinental subduction in the Peter the First Range, central Asia, Geol. Soc. Am. Bull., 104, References 397–408, doi:10.1130/0016-7606(1992)104<0397:SASEFI>2.3.CO;2. Aghanabati, A. (1986), Geological map of the Middle East, scale Harrison, T. M., A. Yin, M. Grove, O. M. Lovera, F. J. Ryerson, and 1:5,000,000, Geol. Surv. of Iran, Tehran. X. Zhou (2000), The Zedong Window: A record of superposed Tertiary Amidon, W. H., and S. A. Hynek (2010), Exhumational history of the north convergence in southeastern Tibet, J. Geophys. Res., 105, 19,211–19,230, central Pamir, Tectonics, 29, TC5017, doi:10.1029/2009TC002589. doi:10.1029/2000JB900078. Arrowsmith, J. R., and M. R. Strecker (1999), Seismotectonic range-front seg- Henderson, A. L., Y. Najman, R. Parrish, M. BouDagher-Fadel, D. Barford, mentation and mountain-belt growth in the Pamir-Alai region, Kyrgyzstan E. Garzanti, and S. Andò (2010), Geology of the Cenozoic Indus Basin (India-Eurasia collision zone), Geol. Soc. Am. Bull., 111, 1665–1683, sedimentary rocks: Paleoenvironmental interpretation of sedimentation doi:10.1130/0016-7606(1999)111<1665:SRFSAM>2.3.CO;2. from the western Himalaya during the early phases of India-Eurasia Avouac, J.-P., and P. Tapponier (1993), Kinematic model of active defor- collision, Tectonics, 29, TC6015, doi:10.1029/2009TC002651. mation in central Asia, Geophys. Res. Lett., 20, 895–898, doi:10.1029/ Hendrix, M. S., T. A. Dumitru, and S. A. Graham (1994), Late Oliogocene- 93GL00128. early Miocene unroofing in the Chinese Tian Shan: An early effect of Bazhenov, M. L., and V. S. Burtman (1990), Structural Arcs of the Alpine the India-Asia collision, Geology, 22, 487–490, doi:10.1130/0091- Belt: Carpathians-Caucasus-Pamirs [in Russian], Nauka, Moscow. 7613(1994)022<0487:LOEMUI>2.3.CO;2. Beaumont, C., R. A. Jamieson, J. P. Butler, and C. J. Warren (2009), Hodges, K. V. (2000), Tectonics of the Himalya and southern Tibet from Crustal structure: A key constraint on the mechanism of ultra-high- two perspectives, Geol. Soc. Am. Bull., 112, 324–350, doi:10.1130/ pressure rock exhumation, Earth Planet. Sci. Lett., 287, 116–129, 0016-7606(2000)112<324:TOTHAS>2.0.CO;2. doi:10.1016/j.epsl.2009.08.001. Hodges, K. V. (2004), Geochronology and thermochronology in orogenic Burtman, V. S., and P. H. Molnar (1993), Geological and Geophysical systems, in Treatise on Geochemistry, edited by H. D. Holland and Evidence for Deep Subduction of Continental Crust Beneath the Pamir, K. K. Turekian, pp. 263–292, Elsevier, Oxford, U. K. Spec. Pap. Geol. Soc. Am., 281, 76 pp. Hubbard, M. S., E. S. Grew, K. V. Hodges, M. G. Yates, and N. N. Pertsev Capitanio, F. A., G. Morra, S. Goes, R. F. Weinberg, and L. Moresi (2010), (1999), Neogene cooling and exhumation of upper-amphibolite-facies India-Asia convergence driven by the subduction of the greater Indian ‘whiteschists’ in the southwest Pamir Mountains, Tajikistan, Tectono- continent, Nat. Geosci., 3, 136–139, doi:10.1038/ngeo725. physics, 305, 325–337, doi:10.1016/S0040-1951(99)00012-8. Carrapa, B. (2010), Resolving tectonic problems by dating detrital minerals, Ji, W. Q., F. Y. Wu, S. L. Chung, J. X. Li, and C. Z. Liu (2009), Zircon U-Pb Geology, 38, 191–192, doi:10.1130/focus022010.1. geochronology and Hf isotopic constraints on petrogenesis of the Gangdese Carrapa, B., J. Wijbrans, and G. Bertotti (2003), Episodic exhumation in the batholith, southern Tibet, Chem. Geol., 262,229–245, doi:10.1016/j. Western Alps, Geology, 31, 601–604, doi:10.1130/0091-7613(2003) chemgeo.2009.01.020. 031<0601:EEITWA>2.0.CO;2. Kapp, P., and J. H. Guynn (2004), Indian punch rifts Tibet, Geology, 32, Carrapa, B., A. Di Giulio, and J. Wijbrans (2004), The early stages of 993–996, doi:10.1130/G20689.1. the Alpine collision: An image derived from the upper Eocene–lower Kapp, P., A. Yin, C. E. Manning, M. Murphy, T. M. Harrison, M. Spurlin, Oligocene record in the Alps–Apennines junction area, Sediment. Geol., D. Lin, D. Xi-Guang, and W. Cun-Ming (2000), Blueschist-bearing 171, 181–203, doi:10.1016/j.sedgeo.2004.05.015. metamorphic core complexes in the Qiangtang block reveal deep crustal Chu, M.-F., S.-L. Chung, B. Song, D. Liu, S. Y. O’Reilly, N. J. Pearson, structure of northern Tibet, Geology, 28,19–22, doi:10.1130/0091- J. Ji, and D.-J. Wen (2006), Zircon U-Pb and Hf isotope constraints on 7613(2000)28<19:BMCCIT>2.0.CO;2. the Mesozoic tectonics and crustal evolution of southern Tibet, Geology, Kapp, P., P. G. DeCelles, G. E. Gehrels, M. Heizler, and L. Ding (2007), 34, 745–748, doi:10.1130/G22725.1. Geological records of the Lhasa-Qiangtang and Indo-Asian collisions in Copeland, P., and T. M. Harrison (1990), Episodic rapid uplift in the the Nima area of central Tibet, Geol. Soc. Am. Bull., 119, 917–933, Himalaya revealed by 40Ar/39Ar analysis of detrital K-feldspar and mus- doi:10.1130/B26033.1. covite, Bengal fan, Geology, 18, 354–357, doi:10.1130/0091-7613(1990) Leier, A. L., P. G. DeCelles, P. Kapp, and G. E. Gehrels (2007a), Lower 018<0354:ERUITH>2.3.CO;2. Cretaceous strata in the Lhasa terrane, Tibet, with implications for under- Coutand, I., M. R. Strecker, J. R. Arrowsmith, G. Hilley, R. C. Thiede, standing the early tectonic history of the Tibetan Plateau, J. Sediment. J. Korjenkov, and M. Omuraliev (2002), Late Cenozoic tectonic develop- Res., 77, 809–825, doi:10.2110/jsr.2007.078. ment of the intramontane Alai Valley, (Pamir-Tien Shan region, central Leier, A. L., P. Kapp, G. E. Gehrels, and P. G. DeCelles (2007b), Detrital Asia): An example of intracontinental deformation due to the Indo- zircon geochronology of Carboniferous-Cretaceous strata in the Lhasa Eurasia collision, Tectonics, 21(6), 1053, doi:10.1029/2002TC001358. terrane, southern Tibet, Basin Res., 19, 361–378, doi:10.1111/j.1365- DeCelles, P. G., D. M. Robinson, and G. Zandt (2002), Implications 2117.2007.00330.x. of shortening in the Himalayan fold-thrust belt for uplift of the Tibetan Leith, W. (1982), Rock assemblages in central Asia and the evolution of the south- Plateau, Tectonics, 21(6), 1062, doi:10.1029/2001TC001322. ern Asian margin, Tectonics, 1, 303–318, doi:10.1029/TC001i003p00303. DeCelles, P. G., J. Quade, P. Kapp, M. Fan, D. L. Dettman, and L. Ding Leith, W. (1985), A mid-Mesozoic extension across central Asia?, Nature, (2007), High and dry in central Tibet during the late Oligocene, Earth 313, 567–570, doi:10.1038/313567a0. Planet. Sci. Lett., 253, 389–401, doi:10.1016/j.epsl.2006.11.001. McDougall, I., and T. M. Harrison (1999), Geochronology and Thermo- Dmitriev, E. A. (1976), Cenozoic Potassium Rocks of Eastern Pamir chronology by the 40Ar/39Ar Method, 2nd ed., Oxford Univ. Press, [in Russian], Akad. Nauk Tadzhikskoy SSR, Dushanbe, Russia. Oxford, U. K. Ducea, M. N., et al. (2003), Building the Pamirs: The view from the under- Molnar, P., and D. Gray (1979), Subduction of continental lithosphere: side, Geology, 31, 849–852, doi:10.1130/G19707.1. Some constraints and uncertainties, Geology, 7,58–62, doi:10.1130/ Gehrels, G. E., A. Yin, and X.-F. Wang (2003), Detrital-zircon geochro- 0091-7613(1979)7<58:SOCLSC>2.0.CO;2. nology of the northeastern Tibetan plateau, Geol. Soc. Am. Bull., 115, 881–896, doi:10.1130/0016-7606(2003)115<0881:DGOTNT>2.0.CO;2.

11 of 12 TC2014 LUKENS ET AL.: MIOCENE EXHUMATION OF THE TAJIK PAMIR TC2014

Murphy, M. A., A. Yin, T. M. Harrison, S. B. Dürr, Z. Chen, F. J. Ryerson, of the Himalayan-Tibetan orogen, Geol. Soc. Am. Bull., 119, 882–896, W. S. F. Kidd, X. Wang, and X. Zhou (1997), Did the Indo-Asian colli- doi:10.1130/B25981.1. sion alone create the Tibetan plateau?, Geology, 25, 719–722, Rohrmann, A., P. Kapp, B. Carrapa, P. W. Reiners, J. Guynn, L. Ding, and doi:10.1130/0091-7613(1997)025<0719:DTIACA>2.3.CO;2. M. Heizler (2011), Thermochronologic evidence for plateau formation in Murphy, M. A., A. Yin, P. Kapp, T. M. Harrison, D. Lin, and G. Jinghui central Tibet by 45 Ma, Geology, 40, 187–190, doi:10.1130/G32530.1. (2000), Southward propagation of the Karakoram fault system, southwest Ruhl, K. W., and K. V. Hodges (2005), The use of detrital mineral Tibet: Timing and magnitude of slip, Geology, 28, 451–454, doi:10.1130/ cooling ages to evaluate steady state assumptions in active orogens: 0091-7613(2000)28<451:SPOTKF>2.0.CO;2. An example from the central Nepalese Himalaya, Tectonics, 24, TC4015, Najman, Y. M. R., M. S. Pringle, M. R. W. Johnson, A. H. F. Robertson, doi:10.1029/2004TC001712. and J. R. Wijbrans (1997), Laser 40Ar/39Ar dating of single detrital Schmidt, J., et al. (2011), Cenozoic deep crust in the Pamir, Earth Planet. muscovite grains from early foreland-basin sedimentary deposits in India: Sci. Lett., 312, 411–421, doi:10.1016/j.epsl.2011.10.034. Implications for early Himalayan evolution, Geology, 25, 535–538, Schwab, M., et al. (2004), Assembly of the Pamirs: Age and origin of mag- doi:10.1130/0091-7613(1997)025<0535:LAADOS>2.3.CO;2. matic belts from the southern Tien Shan to the southern Pamirs and their Najman, Y., A. Carter, G. Oliver, and E. Garzanti (2005), Provenance of relation to Tibet, Tectonics, 23, TC4002, doi:10.1029/2003TC001583. Eocene foreland basin sediments, Nepal: Constraints to the timing and Smith, M. E., B. Singer, and A. Carroll (2003), 40Ar/39Ar geochronology of diachroneity of early Himalayan orogenesis, Geology, 33, 309–312, the Eocene Green River Formation, Wyoming, Geol. Soc. Am. Bull., 115, doi:10.1130/G21161.1. 549–565, doi:10.1130/0016-7606(2003)115<0549:AGOTEG>2.0.CO;2. Negredo, A. M., A. Replumaz, A. Villaseñor, and S. Guillot (2007), Smith, M. E., B. S. Singer, A. R. Carroll, and J. H. Fournelle (2006), High- Modeling the evolution of continental subduction processes in the resolution calibration of Eocene strata: 40Ar/39Ar geochronology of Pamir-Hindu Kush region, Earth Planet. Sci. Lett., 259, 212–225, biotite in the Green River Formation, Geology, 34,393–396, doi:10.1130/ doi:10.1016/j.epsl.2007.04.043. G22265.1. Nikolaev, V. G. (2002), Afghan-Tajik depression: Architecture of sedimen- Sobel, E. R., and T. R. Dumitru (1997), Thrusting and exhumation around tary cover and evolution, Russ. J. Earth Sci., 4, 399–421, doi:10.2205/ the margins of the western Tarim basin during the India-Asia collision, 2002ES000106. J. Geophys. Res., 102, 5043–5063, doi:10.1029/96JB03267. Pashkov, B. R., and V. I. Budanov (1990), The tectonics of the zone Sobel, E. R., M. Oskin, D. Burbank, and A. Mikolaichuk (2006), Exhuma- of intersection between the southeastern and southwestern Pamir tion of basement-cored uplifts: Example of the Kyrgyz Range quantified [in Russian], Geotektonika, 3,70–79. with apatite fission track thermochronology, Tectonics, 25, TC2008, Pashkov, B. R., and V. A. Shvol’man (1979), Riftogenic margins of Tethys doi:10.1029/2005TC001809. in the Pamir [in Russian], Geotektonika, 6,42–57. Sobel, E. R., L. M. Schoenbohm, J. Chen, R. Thiede, D. F. Stockli, Pavlis, T. L., M. W. Hamburger, and G. L. Pavlis (1997), Erosional M. Sudo, and M. R. Strecker (2011), Late Miocene–Pliocene decelera- processes as a control on the structural evolution of an actively deforming tion of dextral slip between Pamir and Tarim: Implications for Pamir fold and thrust belt: An example from the Pamir-Tien Shan region, central orogenesis, Earth Planet. Sci. Lett., 304, 369–378, doi:10.1016/j. Asia, Tectonics, 16, 810–822, doi:10.1029/97TC01414. epsl.2011.02.012. Peltzer, G., and P. Tapponnier (1988), Formation and evolution of strike- Szulc, A. G., et al. (2006), Tectonic evolution of the Himalaya constrained slip faults, rifts, and basins during the India-Asia collision: An experi- by detrital 40Ar-39Ar, Sm-Nd and petrographic data from the Siwalik mental approach, J. Geophys. Res., 93, 15,085–15,117, doi:10.1029/ foreland basin succession, SW Nepal, Basin Res., 18, 375–391, JB093iB12p15085. doi:10.1111/j.1365-2117.2006.00307.x. Pospelov, I. I. (1987), Formations and tectonic development of the late Vermeesch, P. (2004), How many grains are needed for a provenance Variscides of the south Tien Shan and northern Pamir, in Early Geosyn- study?, Earth Planet. Sci. Lett., 224, 441–451, doi:10.1016/j. clinal Formations and Structures [in Russian], edited by Y. M. epsl.2004.05.037. Pushcharov and I. V. Khvorova, pp. 149–178, Nauka, Moscow. Wang, Q., et al. (2008), Eocene melting of subducting continental crust and Pospelov, I. I., and S. P. Sigachev (1987), On retrocharriage in the Pamir early uplifting of central Tibet: Evidence from central-western Qiangtang [in Russian], Dokl. Akad. Nauk SSSR, 3, 678–683. high-K calc-alkaline andesites, dacites and ryholites, Earth Planet. Sci. Ratschbacher, L., M. Schwab, S. Semiletkin, B. Hacker, and U. Hermann Lett., 272, 158–171, doi:10.1016/j.epsl.2008.04.034. (1997), The NE-edge of Tibet: Intracontinental shortening and extension, Wen, D.-R., D. Liu, S.-L. Chung, M.-F. Chu, J. Ji, Q. Zhang, B. Song, T.-Y. basement doming, and exhumation in the Pamirs, Geol. Soc. Am. Abstr. Lee, M.-W. Yeh, and C.-H. Lo (2008), Zircon SHRIMP U–Pb ages of the Programs, 26, 144. Gangdese Batholith and implications for Neotethyan subduction in southern Reigber, C., G. W. Michel, R. Galas, D. Angermann, J. Klotz, J. Y. Chen, Tibet, Chem. Geol., 252, 191–201, doi:10.1016/j.chemgeo.2008.03.003. A. Papschev, R. Arslanov, V. E. Tzurkov, and M. C. Ishanov (2001), White, N. M., M. Pringle, E. Garzanti, M. Bickle, Y. Najman, H. Chapman, New space geodetic constraints on the distribution of deformation in and P. Friend (2002), Constraints on the exhumation and erosion of the central Asia, Earth Planet. Sci. Lett., 191, 157–165, doi:10.1016/S0012- High Himalayan Slab, NW India, from foreland basin deposits, Earth 821X(01)00414-9. Planet. Sci. Lett., 195,29–44, doi:10.1016/S0012-821X(01)00565-9. Ritts, B. D., Y. Yue, S. A. Graham, E. R. Sobel, O. A. Abbink, and Yin, A., and T. M. Harrison (2000), Geologic evolution of the Himalayan- D. Stockli (2008), From sea level to high elevation in 15 million years: Tibetan orogen, Annu. Rev. Earth Planet. Sci., 28, 211–280, doi:10.1146/ Uplift history of the northern Tibetan plateau margin in the Altun Shan, annurev.earth.28.1.211. Am. J. Sci., 308, 657–678, doi:10.2475/05.2008.01. Yin, A., T. M. Harrison, M. A. Murphy, M. Grove, S. Nie, F. J. Ryerson, Robinson, A. C. (2009), Geologic offsets across the northern Karakorum W. X. Feng, and C. Z. Le (1999), Tertiary deformation history of south- fault: Implications for its role and terrane correlations in the western eastern and southwestern Tibet during Indo-Asian collision, Geol. Soc. Himalayan-Tibetan orogen, Earth Planet. Sci. Lett., 279, 123–130, Am. Bull., 111, 1644–1664, doi:10.1130/0016-7606(1999)111<1644: doi:10.1016/j.epsl.2008.12.039. TDHOSA>2.3.CO;2. Robinson, A. C., A. Yin, C. E. Manning, T. M. Harrison, S.-H. Zhang, and X.-F. Wang (2004), Tectonic evolution of the northeastern Pamir: B. Carrapa and G. Gehrels, Department of Geosciences, University of Constraints from the northern portion of the Cenozoic Kongur Shan Arizona, 1040 E. 4th St., Tucson, AZ 85721, USA. extensional system, Geol. Soc. Am. Bull., 116, 953–974, doi:10.1130/ C. E. Lukens, Department of Geology and Geophysics, University of B25375.1. Wyoming, Laramie, WY 82071, USA. ([email protected]) Robinson, A. C., A. Yin, C. E. Manning, T. M. Harrison, S.-H. Zhang, B. S. Singer, Department of Geoscience, University of Wisconsin- and X.-F. Wang (2007), Cenozoic evolution of the eastern Pamir: Impli- Madison, 1215 W. Dayton St., Madison, WI 53706, USA. cations for the strain-accommodation mechanisms at the western end

12 of 12