Cenozoic convergence between the Indian and Asian plates produced the archetypical continental collision zone comprising the Himalaya mountain belt and the Tibetan Plateau. How and where India-Asia convergence was accommodated after collision at or before 52 Ma remains a long-standing controversy. Since 52 Ma, the two plates have converged up to 3,600 AE 35 km, yet the upper crustal shortening documented from the geological record of Asia and the Himalaya is up to approximately 2,350-km less. Here we show that the discrepancy between the convergence and the shortening can be explained by subduction of highly extended continental and oceanic Indian lithosphere within the Himalaya between approximately 50 and 25 Ma. Paleomagnetic data show that this extended continental and oceanic "Greater India" promontory resulted from 2,675 AE 700 km of North-South extension between 120 and 70 Ma, accommodated between the Tibetan Himalaya and cratonic India. We suggest that the approximately 50 Ma "India"-Asia collision was a collision of a Tibetan-Himalayan microcontinent with Asia, followed by subduction of the largely oceanic Greater India Basin along a subduction zone at the location of the Greater Himalaya. The "hard" India-Asia collision with thicker and contiguous Indian continental lithosphere occurred around 25-20 Ma. This hard collision is coincident with far-field deformation in central Asia and rapid exhumation of Greater Himalaya crystalline rocks, and may be linked to intensification of the Asian monsoon system. This two-stage collision between India and Asia is also reflected in the deep mantle remnants of subduction imaged with seismic tomography.continent-continent collision | mantle tomography | plate reconstructions | Cretaceous T he present geological boundary between India and Asia is marked by the Indus-Yarlung suture zone, which contains deformed remnants of the ancient Neotethys Ocean (1, 2) (Fig. 1). North of the Indus-Yarlung suture is the southernmost continental fragment of Asia, the Lhasa block. South of the suture lies the Himalaya, composed of (meta)sedimentary rocks that were scraped off now-subducted Indian continental crust and mantle lithosphere and thrust southward over India during collision. The highest structural unit of the Himalaya is overlain by fragments of oceanic lithosphere (ophiolites).We apply the common term Greater India to refer to the part of the Indian plate that has been subducted underneath Tibet since the onset of Cenozoic continental collision. A 52 Ma minimum age of collision between northernmost Greater India and the Lhasa block is constrained by 52 Ma sedimentary rocks in the northern, "Tibetan" Himalaya that include detritus from the Lhasa block (3). This collision age is consistent with independent paleomagnetic evidence for overlapping paleolatitudes for the Tibetan Himalaya and the Lhasa blocks at 48.6 AE 6.2 Ma ( Fig. 2; SI Text) as well as with an abrupt decrease in India-Asia convergence rates beginning at 55-50 Ma, as demonstrated by India-Asia plate circuits ...

The surface uplift history of the Tibetan Plateau and Himalaya is among the most interesting topics in geosciences because of its effect on regional and global climate during Cenozoic time, its influence on monsoon intensity, and its reflection of the dynamics of continental plateaus. Models of plateau growth vary in time, from pre-India-Asia collision (e.g., Ϸ100 Ma ago) to gradual uplift after the India-Asia collision (e.g., Ϸ55 Ma ago) and to more recent abrupt uplift (<7 Ma ago), and vary in space, from northward stepwise growth of topography to simultaneous surface uplift across the plateau. Here, we improve that understanding by presenting geologic and geophysical data from north-central Tibet, including magnetostratigraphy, sedimentology, paleocurrent measurements, and 40 Ar/ 39 Ar and fission-track studies, to show that the central plateau was elevated by 40 Ma ago. Regions south and north of the central plateau gained elevation significantly later. During Eocene time, the northern boundary of the protoplateau was in the region of the Tanggula Shan. Elevation gain started in pre-Eocene time in the Lhasa and Qiangtang terranes and expanded throughout the Neogene toward its present southern and northern margins in the Himalaya and Qilian Shan.climate ͉ tectonics ͉ magnetostratigraphy ͉ Hoh Xil Basin ͉ Cenozoic T he Tibetan Plateau is the most extensive region of elevated topography in the world (Fig. 1). How such high topography, which should have an effect on climate, monsoon intensity, and ocean chemistry (1-5), has developed through geologic time remains disputed. Various lines of investigation, including evidence from the initiation of rift basins (6), potassium-rich (K-rich) volcanism (7), tectonogeomorphic studies of fluvial systems and drainage basins (8), thermochronologic studies (9), upper-crustal deformation histories (10, 11), stratigraphic and magnetostratigraphic studies of sediment accumulation rates (12), paleobotany (13), and oxygen isotope-based paleoaltimetry (14-22), have suggested different uplift histories. Authors of recent geologic studies (11) have proposed that significant crustal thickening (and by inference, surface uplift) in the Qiangtang terrane occurred in the Early Cretaceous [Ϸ145 mega-annum (Ma) age], followed by major crustal thickening within the Lhasa terrane between Ϸ100 and 50 Ma ago. This hypothesis remains disputed (23). Other models of plateau growth range from Oligocene (e.g., Ϸ30 Ma ago) gradual surface uplift (7) to more recent (Ͻ7 Ma ago) and abrupt surface uplift (24), with oblique stepwise growth of elevation northward and eastward after the India-Eurasia collision (7,20,25,26). With few exceptions (e.g., see refs. 11 and 27), most of these models focus on data from the Himalaya and southern Tibet and remain relatively unconstrained by geologic data from the interior of the Tibetan Plateau.The Hoh Xil Basin (HXB) of the north-central Tibetan Plateau (Figs. 1 and 2) is the most widespread exposure of Paleogene sediments on the high plateau and contains Ͼ5,000...

No abstract

S U M M A R YOngoing controversies on the timing and kinematics of the Indo-Asia collision can be solved by palaeomagnetically determined palaeolatitudes of terranes bounding the Indo-Asia suture zone. We show here, based on new palaeomagnetic data from the Linzizong volcanic rocks (54-47 Ma) near the city of Lhasa, that the latitude of the southern margin of Asia was 22.8 ± 4.2 • N when these rocks were deposited. This result, combined with revised palaeomagnetic results from the northernmost sedimentary units of Greater India and with apparent polar wander paths of India and Eurasia, palaeomagnetically constrain the collision to have occurred at 46 ± 8 Ma (95 per cent confidence interval). These palaeomagnetic results are consistent with tomographic anomalies at 15-25 • N that are interpreted to locate the Tethyan oceanic slab that detached following collision, and with independent 56-46 Ma collision age estimates inferred from the timing of slowing down of India, high pressure metamorphism, the end of marine sedimentation and the first occurrence of suture zone and arc detritus on the Greater Indian margin. When compared with apparent polar wander paths of India and Eurasia, the ∼46 Ma onset of collision at 22.8 ± 4.2 • N implies 2900 ± 600 km subsequent latitudinal convergence between India and Asia divided into 1100 ± 500 km within Asia and 1800 ± 700 km within India.