TECTONICS/Mountain Building and Orogeny 423 plate to flex down and create the Siwalik molasse basin This basin accumulated all the debris eroded from the rising Himalaya, transported south by rivers The rivers converge into the Indus River in the west and the Ganges in the east Sediments eroded from the Himalaya have been transported by these rivers to the Indus Fan in the Arabian Sea and the Bengal Fan in the Bay of Bengal Tibetan Plateau The Tibetan plateau is the largest area of uplifted crust on Earth The plateau extends for over 3000 km east– west and 1500 km north–south It lies at an average elevation of just over km above sea-level The interior of the plateau is very flat with internal drainage, low precipitation, and low erosion rates The margins of the plateau are ringed by mountain ranges including the Himalaya along the south, the Karakoram and Pamirs to the south-west, the Tien Shan and Kun Lun to the north, and the Long Men Shan along the east Earthquakes reveal that the high plateau is currently undergoing east–west extension, whilst the margins of the plateau show compression or strike-slip faulting The geology of Tibet shows that the plateau region includes several different continental plates that were progressively accreted to the southern continental margin of Asia throughout the Phanerozoic (Figure 4) The most recent and probably the largest of these was the final plate collision, that of India with Asia The crust beneath the Tibetan plateau is between 65 and 70 km thick, double that of normal continental crust Several different models have been proposed to account for the thick crust and approximately 1000 km of crustal shortening required The extreme end-member models include underthrusting of India beneath the entire Tibetan plateau, a model proposed initially by Emile Argand in 1924, and homogeneous thickening of the plateau with very little underthrusting of Indian material at depth Recent deep crustal seismic reflection and refraction profiling of Tibet has revealed that the Indian plate lower crust probably underthrusts southern Tibet only as far as the Bangong suture zone approximately 450 km north of the Indus–Yarlung Tsangpo suture zone The equivalent amount of shortening in the Indian plate upper crust has been taken up by intense folding, thrusting, and crustal thickening in the Tethyan zone and Greater Himalaya The seismic profiling has also managed to trace the prolongation of the surface faults in the Himalaya, northwards beneath the southern part of the plateau Seismic and structural data has revealed that the Main Central Thrust and the South Tibetan detachment normal fault bound a mid-crustal layer of hot, partially molten rock that extends southwards to the high Himalaya Magnetotelluric studies have revealed the presence of ‘bright spots’ indicative of pockets of fluid or magma at relatively shallow depths beneath southern Tibet today These have been interpreted as pockets of granitic magma forming today in a structural position similar to those of the 20- to 17-My-old leucogranites cropping out in the high Himalaya This layer of partial-melt migmatites, high-grade gneisses, and leucogranites was extruded out from beneath the southern part of the Tibetan plateau as a channel of ductile-deforming rock bounded by a rheologically stronger upper crust (Tethyan zone) and lower crust (underplated Indian shield Precambrian and Early Palaeozoic rocks) Whereas the Tibetan plateau shows little relief despite being km above mean sea-level, the neighbouring Karakoram range of northern Pakistan and Ladakh shows the highest relief of all, with many 7000- to 8000-m-high mountains and deeply incised river valleys The Karakoram crust has also been tilted, revealing the deep crustal geology not exposed in Tibet (Figure 3) The Karakoram shows multiple episodes of crustal thickening and regional metamorphism spanning the past 65 My and multiple episodes of crustal melting, resulting in granite magmatism A series of pre-50-My-old granite– granodiorite intrusions indicate that the southern margin of Asia was probably a subduction-related Andean-type continental margin prior to the Indian plate collision 50 My ago The climax of mountain building in the Karakoram occurred between 24 an – 15 My ago with the emplacement of the huge Baltoro granite batholith, a series of intrusions of biotite monzogranite to biotite–muscovite–garnet leucogranite A suite of 24- to 22-My-old lamprophyre dykes intruding around the Baltoro granites indicates that parts of the upper mantle were melting at the same time as the lower crust The ages of the Baltoro granites are similar to the age of the Greater Himalayan leucogranites to the south, which span 24–12 My ago, with the majority between 21 and 17 My ago This suggests that following the India– Asia collision, crustal thickening, metamorphism, and magmatism spread both across the south Asian margin in the Karakoram and across the north Indian plate margin along the Himalaya Earthquake distribution across Tibet shows that the entire plateau region is deforming internally today, and not acting as a rigid plate Earthquakes in the high plateau show that the crust is undergoing east–west extension, whereas earthquakes in the mountain ranges bordering the high plateau are