NucL Tracks Radiat. Meas., Vol. 17, No. 3, pp. 301-307, 1990 Int. J. Radiat. Appl. Instrum., Part D Printed in Great Britain

0735-245X/90 $3.00 + .00 Pergamon Press plc

THERMAL HISTORY OF THE KUNLUN BATHOLITH, N. TIBET, AND IMPLICATIONS

FOR UPLIFT OF THE TIBETAN PLATEAU

CHERRY L. E. LEwis University of London Fission Track Research Group, Department of Geological Sciences,

University College London, Gower St, London WCIE 6BT, U.K.

(Received December 1988; in revised form 2 August 1989)

Abstract--At an average elevation of 5 km, the Tibetan plateau is the highest in the world. The timing of uplift and the mechanism by which the crust was thickened to 70 km have long been a subject for debate. Preliminary Rb-Sr and 4°Ar/39Ar biotite ages combined with zircon and apatite fission track dates from the Kunlun batholith in northern Tibet allow for comparison with the previously established uplift history of southern Tibet, and inferences to be made about uplift of the plateau as a whole.

A major divide in the Kunlun Terrane is identified at the Golmud Fault, across which apatite ages decrease from 100 Ma in the north to 20 Ma in the south. This is believed to be as a result of a thrusting event at 120 Ma which uplifted the Northern block and overthrust the Southern block to the south. Subsequent to this event, surface approach rates (SAR) of less than 70 m Ma -t from biotite closure at 120 Ma to apatite cooling at 20 Ma in the Southern block, indicate a long uninterrupted period of thermal equilibration in an essentially static continental block.

Mean confined track lengths of 12.5-13.5/~m in apatites suggest that samples exposed at the surface today resided in the fission track partial annealing zone prior to the onset of uplift which must have occurred subsequent to 20 Ma. This indicates that at least 3-4 km have been uplifted since this time and that the onset of this phase could have been as recent as 8 Ma.

1. INTRODUCTION

THE KUNLUN batholith defines the northern edge of the Tibetan plateau (Fig. 1), which comprises four microplates successively accreted to the Eurasian continent since the Devonian. Northward subduction of the Songpan-Ganzi Terrane along the Kun- lun-Qilian suture line led to emplacement of the main Kunlun batholith during the Permian at 255 Ma (Harris et al., 1988), while southward subduction resulted in collision with the Qiangtang Terrane around 200 Ma along the poorly defined Jinsha suture zone (Dewey et al., 1988). At 120 Ma collision of the Lhasa and Qiangtang Terranes reactivated the Kunlun-Qilian suture along the Xidatan Fault caus- ing the Kunlun Terrane to be overthrust to the south. This event is believed to mark the initiation of movement on the Xidatan Fault which today is a major left-lateral strike-slip fault. The subduction of India beneath Tibet resulted in emplacement of the Gangdese batholith, which now defines the southern limit of the plateau, and eventual collision at 45-40 Ma to form the Himalayas (Searle et aL, 1987).

From palaeo-samples of flora and fauna collected in Tibet, Li et al. (1981) and Xu (1981) argued that the elevation of Tibet was less than 1 km until the end of the late Pliocene, and that consequently the uplift to its present altitude of some 5km must have occurred within the last two million years. Other

workers however, believe that the uplift process was a more gradual one and that it has been occurring since the collision with India some 45 Ma ago (Dewey and Burke, 1973; Him et al., 1984a,b; Chang et al., 1986; Molnar et al., 1987; Copeland et al., 1987).

All previous uplift studies have concentrated on the southern edge of the plateau in Tibet and NW Pakistan (Zhao and Morgan, 1985; Copeland et al., 1987; Zeitler 1985, Zeitler et al., 1982). In particular, the work by Copeland et al. (1987) showed that the Quxu pluton in the Gangdese batholith (Fig. 1) uplifted in discrete pulses with the most rapid phase occurring between 20 and 17 Ma when uplift rates approached a peak of 4.4 mm yr - ' . Unfortunately there is little exposure of suitable granitoid material for dating in the centre of Tibet so a complete examination of the whole plateau was not possible. Nevertheless, a study of granites from the Kunlun Terrane allowed uplift comparisons to be made with data from the south and inferences to be drawn regarding uplift of the plateau as a whole. Full details of the fission track data, which were calibrated using the zeta method described in Hurford and Green 0983), can be found in Lewis (1988).

2. THE KUNLUN TERRANE

Immediately north of the Xidatan Fault (Fig. 2), which marks the southern boundary of the Kunlun

301

302 C. L. E. LEWIS

microplate, a group of post-tectonic plutons give Rb-Sr whole rock ages around 190 Ma (Harris et al., 1988; Lewis, 1988), However, Rb-Sr and 4°Ar/39Ar ages from biotites in the Xidatan granite, which is now cut by the Xidatan Fault, have been re-set to 120 Ma by partial melting of the granite as a result of a thrusting event at this time. Ages of the other post-tectonic plutons increase northwards away from the fault having only been partially re-set. Two exposures of the Yie Nin Gou pluton show biotites that have been partially reset in the more southerly outcrop (Yie Nin Gou X) while those from the more northerly exposure still retain an age of emplacement. Similarly the biotites from the 390 Ma Wanbaogou pluton also record intrusion ages, but ~°Ar/39Ar data show a partial re-setting of the feldspars by intrusion

of the Permian batholith around 255 Ma. North of the Golmud Fault (Fig. 2), plutons record a re-setting of biotites at 228 Ma which may be associated with movement on the Golmud Fault at this time.

3. FISSION TRACK RESULTS

3.1 Post-tectonic plutons

Figure 3 plots apparent ages against assumed closure temperatures for the post-tectonic granites north of the Xidatan Fault, and curves can be drawn through these points to give an average cooling rate for each pluton. However, in this type of plot infor- mation is only available at the point of mineral closure, with relatively few constraints being placed

Kunlun batlloilth ~ KUNIr UN T E R R A N E ~ . . . .

s o . - - "LTF' Z-Q%3 G,, d ,

r ; ~ QIANTANG TERRANE

C:=

Eb

Banggon~uture 1120 Ma)

~ 2 J 1 ~ 0 / " " ~ Granite

~ ~ LHASA TERRANE

~ ~ . T s a n g p o Su tu re - ¢2-P-s -M =--~-) -

HIMALAYAS

FIG. 1. Gereralized map of the 1985 Tibet Geotraverse route showing the outcrop of plutonic rocks. The names of the four microplates are also shown along with the position of suture lines and approximate

times of collision.

UPLIFT OF THE TIBETAN PLATEAU 303

km 20 !

Golmud F a u l t _ . . . . . . . . . - - - -

/

M i d - K u n l u n F a u l t / . . . . . . . . . . . . . . . . . . . . . . ...f_

W a n b a o g o u ~

D u o

Golmud East

Hydro

. . . . . S O U T H E R N B L O C K - -

.- Xidatan Fault

Fro. 2. Simplified map of the Kunlun Terrane showing the plutons sampled for this study.

on the shape of the curve between points. Neverthe- less, following the Cretaceous thrusting event at 120 Ma with the associated re-setting of the biotite ages, the cooling curve for the Xidatan granite is well constrained in having both a zircon age (81 + 3 Ma) and a calculated low temperature 40Ar/39Ar feldspar age (34 Ma at 155 + 20°C; Harrison, personal com- munication) between the biotite and apatite ages. Assuming a constant rate of cooling between biotite closure at 120Ma and feldspar closure at 34Ma allows for interpolation of the zircon closure tem- perature between these two points which, at 234 + 10°C, agrees well with 240 + 25°C obtained by Hurford (1986). The very slow cooling rates from

biotite re-setting to apatite closure at 13 + 2 Ma are calculated to be in the region of 2°C M a - l .

3.2. Wanbaogou and Golmud Hydro plutons

At 255 Ma the feldspars in the Wanbaogou pluton were re-set at 250°C (Harrison, personal communica- tion), by circulating hydrothermal fluids resulting from intrusion of the Permian batholith (Fig. 4). Zircons give a cooling age of 83 + 3 Ma, but if a closure temperature for this age is interpolated in the same way as for the Xidatan pluton between the high and low temperatures calculated from the 4°Ar/39Ar age spectra, a very low closure temperature of

T*C

700

600

500

4OO

300

200

100

0

Emplacement

• Rb/Sr biotite Z i r m n FY

Low T At /At

Al~t i te F r

@* XidaUm

m. Yie Nia Gou

.0o Yie Nin Gou X

N,Ij T~

- - - ; . . . . ; . . . . ; . . . . ; . . . . _- . . . . ; . . . . ; . . . .

5 0 I00 150 200 250 300 3 5 0 4 0 0

Apparent Ages ( M a )

FiG, 3. Cooling curves for the Southern Region post-tectonic granites, Note the increase in biotite ages away from the Xidatan Fault.

304 C . L . E . LEWIS

Toc

700 :Emplacement

600

50O

4O0

_~_ Al~te Fr

0 .50 100 150 200 250 300 350 400 Apparent Ages (Ma)

FIG. 4. Cooling curves for the Wanbaogou and the Golmud Hydro plutons.

4~ Wimbmgou

o- Oolmd Hydro

136 + 10°C is inferred for the zircons. Alternatively, if temperatures were raised sufficiently at 120 Ma to fully or partially anneal the zircons, although clearly they were insufficient to re-set the biotites, then a more realistic closure temperature of 165°C could be inferred for the zircons. Similar slow cooling results are obtained from the Golmud Hydro Group and both plutons give apatite ages within error of each other at 20 Ma, which is also within error of plutons from the Southern Region.

3.3. Northern Region

North of the Goimud Fault there is a pronounced increase in both the zircon and particularly the apatite ages, which are some 100 Ma older than those in the south (Fig. 5). The Goimud Fault is also identified by geochemical data (Lewis, 1988; Harris et al., 1988) as being a significant boundary in the region and it allows the terrane to be divided up into the Northern and Southern blocks. The age difference across the Golmud Fault suggests that there must

have been differential movement between the two blocks which, from the age of the apatites in the Northern block is considered to have occurred during the Cretaceous thrusting event at 120 Ma, uplifting the apatites in this block into the zone of relative track stability.

4. DISCUSSION

The different isotopic analytical techniques applied to the Kunlun granitoids provide a thermal history for the region stretching back 400 Ma, and the data are presented graphically against distance from the Xidatan Fault in Fig. 6. From this it can be seen that significant events occurred around 390, 255, 120 and 20 Ma and that thermal equilibration of the Southern block occurred between 120 and 20 Ma subsequent to the disturbance caused by the Cretaceous collision between the Lhasa and Qiangtang terranes.

If the data are related to thermal equilibration in this way, it becomes apparent that the reduction in mineral ages with time does not necessarily reflect a

Toc

700

600

500

400

30O I

200

100

o. o

d RblSr biotite . / 1

5O 100 150 200 Apparent Ages

f 230 300 350 400

I: - Golmud East

- Doo Ya He

FIG. 5. Cooling curves for the Northern Region plutons.

UPLIFT OF THE TIBETAN PLATEAU 305

400

350

30O

250

Ale (Ma) 200

150

100

50

I F ~

SoWl~m Naril~m

I / \

S i i w

l0 2O 30 4O 50 Distance north of Xidatan Fault (kin)

0 - - ~ - • • l

-10 60 70

2a~T

O-AFr

FIG. 6. Summary plot of all age data against distance north of the Xidatan Fault. Rb/Sr = data from biotites. ZFT = zircon fission track data. AFT --- apatite fission track data. A single zircon fission track cooling age from a quartz porphyry south of the Xidatan Fault is also shown on the diagram. On its own this age is not very informative, except that at 120 Ma it is further evidence for a significant event

at this time.

rise of the rock towards the surface as a result of uplift, but a relaxation of the geothermal gradient in an essentially static continental block, coupled with normal erosion. It then becomes much more mean- ingful to interpret the data in terms of defining periods of crustal stability, which are occasionally punctuated by orogens, rather than looking for periods of uplift that probably represent only a small percentage of the time scale involved. Discussions of "uplift rates" in this type of environment then be- come meaningless and the term "surface approach rate" (SAR) is preferred to avoid ambiguity.

Conventionally it is believed that fission track ages decrease with depth as tracks enter the zone of partial annealing and become shortened. Classic examples are to be found in the fission track analysis of deep boreholes (Gleadow and Duddy, 1981; Gleadow et al., 1983; Hammerschmidt et aL, 1984). As a result of partial annealing the track density becomes reduced which in turn reduces the measured age. However, in regions such as the Kunlun Terrane where slow cooling over a prolonged period of time has been followed by a phase of uplift, then it is possible for the zone of partial annealing to become uplifted and exposed. The measured ages will then be a mixture of shortened tracks accumulated prior to uplift while sitting in the annealing zone, and new longer tracks formed since uplift into the stability zone. This will result in a mixed age that is older than the actual onset of uplift. Ideally these will result in a bimodal distribution of track lengths which can then be identified but in reality the two distributions often merge and overlap.

Confined tracks from the Kunlun apatites were hard to find, due to low uranium concentrations and young ages, but typical distributions from the Southern and Northern blocks show a fairly broad range of tracks with a mean length of

13.52_+ 1.35#m and 12.62 + 1.21/am respectively. Compared to the average track lengths of 15 tam from age standards, the Southern block shows a reduction in the mean track length of only 10%, suggesting they are from the upper region of the partial annealing zone. Those from the Northern block have been reduced by about 15% and are therefore from a slightly lower level in the annealing zone, but since these are a 100 Ma older than those in the Southern block, they are believed to be associated with the earlier Cretaceous phase of uplift in the region at 120 Ma.

Examination of SAR in Fig. 7 confirms a long period of slow cooling in the Kunlun Terrane fol- lowed by a period of uplift in the Southern block subsequent to an apparent age of 20 Ma. On this diagram an "averaged" uplift rate from the Quxu pluton in southern Tibet over the last 20 Ma is included for comparison and to demonstrate how these curves can hide peaks of uplift such as the period between 20 and 17 Ma, from which it is known that rates accelerated to as much as 4ram yr -I. Clearly, without further constraints it is not possible to be more specific about when periods of accelera- tion occurred in the Kunlun terrane. Nevertheless, since the onset of uplift, 3--4 krn must have been removed to expose at the surface today samples from the top of the partial annealing zone. Had the onset of uplift occurred as long ago as 20 Ma, then it is to be expected that even moderate erosion rates would have penetrated to the base of the annealing zone to reveal true uplift ages. For example, assuming ero- sion rates equal uplift rates, even at a moderate 0.5 m m y r -m only 8 Ma would be required to pene- trate to the base of the partial annealing zone to reveal true uplift rates. Since this does not appear to be revealed it must be concluded either that uplift has been relatively recent or that erosion rates are

NT 17/~J

306 C . L . E . LEWIS

Gangdese bathollth Kunlun batholith

i 0°0i r

I°°I /. . A.,,. 5O ~ y ~ " - ~

,'4'

~ - S o u t h

- Cent ra l

• A- N o r t h

x - Quxu

0 20 40 60 g0 100 120 140

Apparent Ages (Ma)

FIG. 7. Surface approach rates tSAR) for plutons in the Kunlun Terrane compared with average uplift rates from the Quxu pluton in southern Tibet, assuming a geothermal gradient of 30°C km-~.

unusually slow. That there is no evidence for a major escarpment at the Golmud Fault suggests that the latter is not the case. Thus from these data it appears that the maximum age for the onset of uplift in the Southern block can be constrained to being less than 20 Ma, and that the min imum age could be as recent as 8 Ma.

Recent uplift in the Kun lun is in accordance with a model which suggests that thickening and uplift of the Tibetan plateau would propagate northwards subsequent to the Himalayan collision (Dewey et al., 1988, and references therein), rather than uplifting the whole plateau simultaneously. From their work in the Gangdese batholith, southern Tibet, Copeland et al. (1987) show that a major pulse of uplift occurred in that region between 20 and 17 Ma. It is therefore geologically realistic to infer that a peak of uplift in the Kunlun occurred subsequent to this between 17 and 8 Ma.

Acknowledgements--I am particularly grateful to Dr Anthony J. Hurford for introducing me to the wonders of fission track dating, and to Professor Emili6 J~.ger for use of the fission track dating facilities at the University of Berne, Switzerland. I should also like to thank Professor Hawkesworth and Dr Hurford for perusing this manuscript and making helpful suggestions. This project was financially supported by the Royal Society, London, and the Open University, Milton Keynes.

REFERENCES

Chang C. et al. (1986) Preliminary conclusions of the Royal Society and Academia Sinica 1985 geotraverse of Tibet. Nature 323, 501-507.

Copeland P., Harrison T. M., Kidd W. S. F., Xu R. and Zhang Y. (1987) Rapid early Miocene acceleration of uplift in the Gangdese Belt Xizang (southern Tibet), and its bearing on accommodation mechanisms of the India-Asia collision. Earth planet. Sci. Lett. 86, 240-252.

Dewey J. F. and Burke K. C. A. (1973) Tibetan, Variscan and Precambrian basement reactivation: products of continental collision. J. Geol. 81, 683--692.

Dewey J. F., Shackleton R. M., Chengfa C. and Yiyin S. (1988) Tectonic evolution of the Tibetan Plateau. Phil. Trans. R. Soc. Lond. A327, 379--413.

Glcadow A. J. W. and Duddy I. R. (1981) A natural long term annealing experiment for apatite. Nucl. Tracks Radiat. Meas. 5, 169-174.

Gleadow A. J. W., Duddy I. R. and Lovering J. F. (1983) Fission track analysis: a new tool for the evaluation of thermal histories and hydrocarbon potential. Aust. Petrol. Explor. Ass. J. 23, 93-102.

Hammerschmidt K., Wagner G. A. and Wagner M. (1984) Radiometric dating on research drill core Urach I l i - a contribution to its geothermal history. J. geophys. Res. 54, 97-105.

Harris N. B. W., Xu R., Lewis C. L., Hawkesworth C. J. and Zhang Y. (1988) Isotope geochemistry of the Tibetan Plateau. Phil. Trans. R. Soc. Lond. A327, 263-285.

Hirn A. et al. (1984a) Crustal structure and variability of the Himalayan border of Tibet. Nature 307, 23-25.

Hirn A. et al. (1984b) Lhasa block and bordering sutures--a continuation of a 500kin Moho traverse through Tibet. Nature 307, 25-27.

Hurford A. J. (1986) Cooling and uplift patterns in the Lepontine Alps, South Central Switzerland and an age of vertical movement on the Insubric fault line. Contr. Miner. Petrol. 92, 413--427.

Hurford A. J. and Green P. F. (1983) The zeta age calibration of fission track dating. Isotope Geosci. 1, 285-317.

Lewis C. L. E. (1988) Petrogenesis and thermal history of the Kunlun batholith, Northern Tibet. Ph.D. thesis, Open University, Milton Keynes.

Li J., Li B., Wang F., Zhang Q., Wen S. and Zhang B. (1981) The process of the uplift of the Qingai-Xizang plateau. In Proc. Syrup. Qinghai-Xizhang (Tibet) Plateau. Geology, Geological History and Origin of the Qinghai-Xizhang Plateau, Vol. 1, pp. I 11-118.

Molnar P., Burchfield B. C., Liang K. and Zhao A. (1987) Geomorphic evidence for active faulting in the Altyn Tagh and northern Tibet and qualitative estimates of its contribution to the convergence of India and Eurasia. Geology 15, 249-253.

U P L I F T O F T H E T I B E T A N P L A T E A U 307

Searle M. P., Windley B. F., Coward M. P., Cooper D. J. W., Rex A. J., Rex D., Tingdong L., Xuchang X., Jan M. Q., Thakur V. X. and Kumar S. (1987) The closing of Tethys and the tectonics of the Himalaya. Bull. Geol. Soc. Am. 98, 678-701.

Xu R. ( 198 !) Vegetational changes in the past and the uplift of Qinghai-Xizhang plateau. In Proc. Symp. Qinghai- Xizhang (Tibet) Plateau. Geology, Geological History and Origin of the Qinghai-Xizhang Plateau, Vol. 1, pp. 139-141.

Zeitler P. K. (1985) Cooling history of the NW Himalaya, Pakistan. Tectonics 4, 127-151.

Zeitler P. K., Tahirkeli R. A. K., Naeser C. W. and Johnson N. M. (1982) Unroofing history of a suture zone in the Himalaya of Pakistan by means of fission track annealing ages. Earth planet. Sci. Lett. 57, 227-240.

Zhao W. and Morgan W. J. (1985) Uplift of the Ti~tan Plateau. Tectonics 4, 359-369.