T. M. Harrison

Did the Indo-Asian collision alone create the Tibetan plateau?

It is widely believed that the Tibetan plateau is a late Cenozoic feature produced by the Indo-Asian collision. However, because Tibet was the locus of continental accretion and subduction throughout the Mesozoic, crustal thickening during that time may also have contributed to growth of the plateau. This portion of the geologic history was investigated in a traverse through the central Lhasa block, southern Tibet. Together with earlier studies, our mapping and geochronological results show that the Lhasa block underwent little north-south shortening during the Cenozoic. Rather, our mapping shows that ∼60% crustal shortening, perhaps due to the collision between the Lhasa and Qiangtang blocks, occurred during the Early Cretaceous. This observation implies that a significant portion of southern Tibet was raised to perhaps 3–4 km elevation prior to the Indo-Asian collision.

Tectonic history of the Altyn Tagh fault system in northern Tibet inferred from Cenozoic sedimentation

The active left-slip Altyn Tagh fault defines the northern edge of the Tibetan plateau. To determine its deformation history we conducted integrated research on Cenozoic stratigraphic sections in the southern part of the Tarim Basin. Fission-track ages of detrital apatites, existing biostratigraphic data, and magnetostratigraphic analysis were used to establish chronostratigraphy, whereas composition of sandstone and coarse clastic sedimentary rocks was used to determine the unroofing history of the source region. Much of the detrital grains in our measured sections can be correlated with uplifted sides of major thrusts or transpressional faults, implying a temporal link between sedimentation and deformation. The results of our studies, together with existing stratigraphic data from the Qaidam Basin and the Hexi Corridor, suggest that crustal thickening in northern Tibet began prior to 46 Ma for the western Kunlun Shan thrust belt, at ca. 49 Ma for the Qimen Tagh and North Qaidam thrust systems bounding the north and south margins of the Qaidam Basin, and prior to ca. 33 Ma for the Nan Shan thrust belt. These ages suggest that deformation front reached northern Tibet only ∼10 ± 5 m.y. after the initial collision of India with Asia at 65–55 Ma. Because the aforementioned thrust systems are either termination structures or branching faults of the Altyn Tagh left-slip system, the Altyn Tagh fault must have been active since ca. 49 Ma. The Altyn Tagh Range between the Tarim Basin and the Altyn Tagh fault has been a long-lived topographic high since at least the early Oligocene or possibly late Eocene. This range has shed sediments into both the Tarim and Qaidam Basins while being offset by the Altyn Tagh fault. Its continuous motion has made the range act as a sliding door, which eventually closed the outlets of westward-flowing drainages in the Qaidam Basin. This process has caused large amounts of Oligocene–Miocene sediments to be trapped in the Qaidam Basin. The estimated total slip of 470 ± 70 km and the initiation age of 49 Ma yield an average slip rate along the Altyn Tagh fault of 9 ± 2 mm/yr, remarkably similar to the rates determined by GPS (Global Positioning System) surveys. This result implies that geologic deformation rates are steady state over millions of years during continental collision.

Late Cenozoic tectonic evolution of the southern Chinese Tian Shan

Structural, sedimentological, magnetostratigraphic, and 40Ar/39Ar thermochronological investigations were conducted in the southern Chinese Tian Shan. On the basis of our own mapping and earlier investigations in the area, the Late Cenozoic southern Tian Shan thrust belt may be divided into four segments based on their style of deformation. From west to east, they are (1) Kashi-Aksu imbricate thrust system, (2) the Baicheng-Kuche fold and thrust system, (3) the Korla right-slip transfer system, and (4) the Lop-Nor thrust system. The westernmost Kashi-Aksu system is characterized by the occurrence of evenly spaced (12–15 km) imbricate thrusts. The Baicheng-Kuche and Korla systems are expressed by a major north dipping thrust (the Kuche thrust) that changes its strike eastward to become a NW striking oblique thrust ramp (the Korla transfer zone). The Lop Nor system in the eastern-most part of the southern Chinese Tian Shan consists of widely spaced thrusts, all involved with basement rocks. Geologic mapping and cross-section construction suggest that at least 20–40 km of crustal shortening with a horizontal shortening strain of 20–30% has occurred in the southern Chinese Tian Shan during the late Cenozoic. These estimates are minimum because of both conservative extrapolation of the thrust geometries and partial coverage of the thrust belt by the cross sections. The timing of initial thrusting is best constrained in the Kuche basin where crustal shortening may have occurred at 21–24 Ma, the time of a major facies transition between lacustrine and braided-fluvial sequences constrained in general by biostratigraphy and in detail by magnetostratigraphy. This estimate represents only a minimum age, as development of thrusts in the southern Chinese Tian Shan may have propagated southward toward the foreland. Thus the sedimentary record only represents the southernmost and therefore youngest phase of thrusting. If our estimate of timing for the thrust initiation (21–24 Ma) is correct, using the estimated magnitude of shortening (20–40 km) and shortening strain (20–30%), the averaged rates of late Cenozoic horizontal slip and shortening strain are 1–1.9 mm yr−1 and 2.9–4.5 × 10−16 s−1, respectively. Our reconnaissance 40Ar/39Ar thermochronological analysis in conjunction with earlier published results of apatite fission track analysis by other workers in the Chinese Tian Shan suggests that the magnitude of Cenozoic denudation is no more than 10 km, most likely less than 5 km. We demonstrate via a simple Airy-isostasy model that when the thermal effect on changes in surface elevation is negligible, determination of the spatial distribution and temporal variation of both horizontal shortening strain and denudation becomes a key to reconstructing the elevation history of the Tian Shan. Using this simple model, the loosely constrained magnitude of crustal-shortening strain and denudation in the southern Chinese Tian Shan implies that it may have been elevated 1.0–2.0 km since the onset of Cenozoic thrusting.

When did the roof collapse? Late Miocene north-south extension in the high Himalaya revealed by Th-Pb monazite dating of the Khula Kangri granite

Th-Pb ion microprobe measurements made on 12 monazite grains from the Khula Kangri granite, Tibet-Bhutan frontier, are interpreted to indicate that crystallization occurred at 12.5 ± 0.4 Ma. The leucogranite is cut by the Gonto La detachment, part of the Southern Tibet detachment system that has allowed upper-level, north-directed extension of the Himalayan orogen. Significant orogen-normal extension in southern Tibet appears to have continued 8–10 m.y. later than previously recognized. This is the first reported crystallization age for a leucogranite east of the Yadong cross structure, an apparent 70 km offset of the high Himalaya and Southern Tibet detachment. West of the Yadong cross structure, reliable ages for high Himalaya events (major Main Central thrust slip, granite generation and emplacement, attainment of critical topography, and major detachment extension) group between ca. 24 and 19 Ma. We interpret the west-to-east change across the Yadong cross structure to be due to either (1) an abrupt, ∼10 m.y. younging of principal high Himalayan events or (2) a deeper (thus younger) exposed part of the footwall of the southern Tibet detachment. Near Khula Kangri, the Southern Tibet detachment is cut by the highly oblique Yadong-Gulu rift; a manifestation of Tibet plateau east-west extension. Integrated estimates of magnitude, and rate, of detachment displacement suggest that the observed postcrystallization north-directed extension lasted for 1–3 m.y., after which time the Yadong-Gulu rift formed. This interpretation is consistent with initiation of east-west extension of Tibet at ca. 8 Ma.

Thermal structure and exhumation history of the Lesser Himalaya in central Nepal

The Lesser Himalaya (LH) consists of metasedimentary rocks that have been scrapped off from the underthrusting Indian crust and accreted to the mountain range over the last ~20 Myr. It now forms a significant fraction of the Himalayan collisional orogen. We document the kinematics and thermal metamorphism associated with the deformation and exhumation of the LH, combining thermometric and thermochronological methods with structural geology. Peak metamorphic temperatures estimated from Raman spectroscopy of carbonaceous material decrease gradually from 520°–550°C below the Main Central Thrust zone down to less than 330°C. These temperatures describe structurally a 20°–50°C/km inverted apparent gradient. The Ar muscovite ages from LH samples and from the overlying crystalline thrust sheets all indicate the same regular trend; i.e., an increase from about 3–4 Ma near the front of the high range to about 20 Ma near the leading edge of the thrust sheets, about 80 km to the south. This suggests that the LH has been exhumed jointly with the overlying nappes as a result of overthrusting by about 5 mm/yr. For a convergence rate of about 20 mm/yr, this implies underthrusting of the Indian basement below the Himalaya by about 15 mm/yr. The structure, metamorphic grade and exhumation history of the LH supports the view that, since the mid-Miocene, the Himalayan orogen has essentially grown by underplating, rather than by frontal accretion. This process has resulted from duplexing at a depth close to the brittle-ductile transition zone, by southward migration of a midcrustal ramp along the Main Himalayan Thrust fault, and is estimated to have resulted in a net flux of up to 150 m^2/yr of LH rocks into the Himalayan orogenic wedge. The steep inverse thermal gradient across the LH is interpreted to have resulted from a combination of underplating and post metamorphic shearing of the underplated units.

Southward propagation of the Karakoram fault system, southwest Tibet: Timing and magnitude of slip

The net slip on the southern portion of the Karakoram fault system in southwest Tibet is estimated by restoring a piercing line defined by two key surfaces in the South Kailas thrust system, a regional counter thrust along the Indus-Yalu suture. Assuming that the thrust system is planar across the Karakoram fault, we calculate 66 ± 5.5 km of normal right slip. Documentation of the South Kailas thrust active at 13 Ma implies that the Karakoram fault in southwest Tibet did not initiate until after the cessation of motion on the thrust. However, field investigations of the central portion of the Karakoram fault system document the fault to have been active at 17 Ma and to have accumulated a maximum of 150 km of right slip. We suggest that these along-strike variations in the magnitude of slip and timing constraints are best explained by southward propagation of the Karakoram fault system. This is inconsistent with major right-lateral slip on the fault system, which was used in support of extrusion models for Tibet.

Structural evolution of the Gurla Mandhata detachment system, southwest Tibet: Implications for the eastward extent of the Karakoram fault system

Field mapping and geochronologic and thermobarometric analyses of the Gurla Mandhata area, in southwest Tibet, reveal major middle to late Miocene, east-west extension along a normal-fault system, termed the Gurla Mandhata detachment system. The maximum fault slip occurs along a pair of low-angle normal faults that have caused significant tectonic denudation of the Tethyan Sedimentary Sequence, resulting in juxtaposition of weakly metamorphosed Paleozoic rocks and Tertiary sedimentary rocks in the hanging wall over amphibolite-facies mylonitic schist, marble, gneisses, and variably deformed leucogranite bodies in the footwall. The footwall of the detachment fault system records a late Miocene intrusive event, in part contemporaneous with top-to-the-west ductile normal shearing. The consistency of the mean shear direction within the mylonitic footwall rocks and its correlation with structurally higher brittle normal faults suggest that they represent an evolving low-angle normal-fault system. 4 0 Ar/ 3 9 Ar data from muscovite and biotite from the footwall rocks indicate that it cooled below 400 °C by ca. 9 Ma. Consideration of the original depth and dip angle of the detachment fault prior to exhumation of the footwall yields total slip estimates between 66 and 35 km across the Gurla Mandhata detachment system. The slip estimates and timing constraints on the Gurla Mandhata detachment system are comparable to those estimated on the right-slip Karakoram fault system, to which it is interpreted to be kinematically linked. Moreover, the mean shear-sense direction on both the Karakoram fault and the Gurla Mandhata detachment system overlap along the intersection line between the mean orientations of the faults, which further supports a kinematic association. If valid, this interpretation extends previous results that the Karakoram fault extends to mid-crustal depths.

Geologic correlation of the Himalayan orogen and Indian craton: Part 2. Structural geology, geochronology, and tectonic evolution of the eastern Himalaya

Despite being the largest active collisional orogen on Earth, the growth mechanism of the Himalaya remains uncertain. Current debate has focused on the role of dynamic interaction between tectonics and climate and mass exchanges between the Himalayan and Tibetan crust during Cenozoic India-Asia collision. A major uncertainty in the debate comes from the lack of geologic information on the eastern segment of the Himalayas from 91°E to 97°E, which makes up about one-quarter of the mountain belt. To address this issue, we conducted detailed field mapping, U-Pb zircon age dating, and 40Ar/39Ar thermochronology along two geologic traverses at longitudes of 92°E and 94°E across the eastern Himalaya. Our dating indicates the region experienced magmatic events at 1745–1760 Ma, 825–878 Ma, 480–520 Ma, and 28–20 Ma. The first three events also occurred in the northeastern Indian craton, while the last is unique to the Himalaya. Correlation of magmatic events and age-equivalent lithologic units suggests that the eastern segment of the Himalaya was constructed in situ by basement-involved thrusting, which is inconsistent with the hypothesis of high-grade Himalaya rocks derived from Tibet via channel flow. The Main Central thrust in the eastern Himalaya forms the roof of a major thrust duplex; its northern part was initiated at ca. 13 Ma, while the southern part was initiated at ca. 10 Ma, as indicated by 40Ar/39Ar thermochronometry. Crustal thickening of the Main Central thrust hanging wall was expressed by discrete ductile thrusting between 12 Ma and 7 Ma, overlapping in time with motion on the Main Central thrust below. Restoration of two possible geologic cross sections from one of our geologic traverses, where one assumes the existence of pre-Cenozoic deformation below the Himalaya and the other assumes flat-lying strata prior to the India-Asia collision, leads to estimated shortening of 775 km (∼76% strain) and 515 km (∼70% strain), respectively. We favor the presence of significant basement topography below the eastern Himalaya based on projections of early Paleozoic structures from the Shillong Plateau (i.e., the Central Shillong thrust) located ∼50 km south of our study area. Since northeastern India and possibly the eastern Himalaya both experienced early Paleozoic contraction, the estimated shortening from this study may have resulted from a combined effect of early Paleozoic and Cenozoic deformation.

Evaluating the role of preexisting weaknesses and topographic distributions in the Indo-Asian collision by use of a thin-shell numerical model

Thin-viscous-sheet models have proved to be very useful in exploring the interaction between plate boundary and gravitational forces during continental collision. However, simplifications of these models (e.g., absence of faults, planar geometry of the collision zone) limit their use in making specific predictions regarding tectonic evolution, such as the role of eastward extrusion along major strike-slip faults during the Indo-Asian collision. This deficiency is overcome by a thin-shell finite-element model with faults that can assess the effects of preexisting fault configurations and topographic distributions on velocity fields. Numerical simulations of a palinspastically restored Asia at ca. 50 Ma suggest that preexisting lithospheric weaknesses favor north-south shortening during initial collision, whereas preexisting high topography in southern Asia promotes eastward extrusion. These results underscore the first-order importance of initial topography and the distribution of preexisting faults in the outcome of geodynamic modeling.

Did the Indo-Asian collision alone create theTibetan plateau?: Reply

doi: 10.1130/0091-7613(1999)027<0285:DTIACA>2.3.CO;2 1999;27;285-286 Geology M. A. Murphy, An Yin and T. M. Harrison Did the Indo-Asian collision alone create theTibetan plateau?: Reply Email alerting services articles cite this article to receive free e-mail alerts when new www.gsapubs.org/cgi/alerts click Subscribe to subscribe to Geology www.gsapubs.org/subscriptions/ click Permission request to contact GSA http://www.geosociety.org/pubs/copyrt.htm#gsa click