Résumé
The timing of E–W extension of the Tibetan plateau provides a test of mechanical models of the geodynamic evolution of the India–Asia convergence zone. In this work we focus on the Kung Co half graben (Southern Tibet, China), bounded by an active N–S normal fault with a minimum vertical offset of 1600 m. To estimate the onset of normal faulting we combined high and medium temperature (U–Pb, Ar/Ar) and low temperature ((U–Th)/He) thermochronometry of the Kung Co pluton, a two-mica granite of the northern Himalayan granitic belt that outcrop in the footwall of the fault. Biotite and muscovite Ar/Ar ages , are close from each other [not, vert, similar 16 Ma ± 0.2 (Ms) and not, vert, similar 15 ± 0.4 Ma (Bt)], which is typical of fast cooling. The zircon and apatite (U–Th)/He ages range from 11.3 to 9.6 Ma and 9.9 to 3.7 Ma respectively. These He ages are indicative of (1) fast initial cooling, from 11.3 to not, vert, similar 9 Ma, gradually decreasing with time and (2) a high geothermal gradient (not, vert, similar 400 °C/km), close to the surface at not, vert, similar 10 Ma. The Kung Co pluton was emplaced at about 22 Ma (U–Pb on zircon) at less than 10 km depth and 520–545 °C. Subsequent to its shallow emplacement, the pluton underwent fast thermal re-equilibration ending around 7.5 Ma, followed by a period of slow cooling caused either by the end of the thermal re-equilibration or by very slow exhumation (0.02–0.03 mm/yr) from not, vert, similar 7.5 Ma to at least 4 Ma. In either case the data suggest that the exhumation rate increased after 4 Ma. We infer this increase to be related to the initiation of the Kung Co normal fault. A critical examination of previously published data show that most not, vert, similar N–S Tibetan normal faults may have formed less than 5 Ma ago rather than in the Miocene as assumed by several authors. Such a young age implies that E–W extension is not related to the Neogene South Tibetan magmatism (25 to 8 Ma). Consequently, models relating E–W extension to magmatism, such as convective removal of the lower lithosphere, may be inappropriate. We rather think that this extension is related with local accommodation of boundary forces and displacements.