Channel flow and ductile extrusion of the high Himalayan slab-the Kangchenjunga-Darjeeling profile, Sikkim Himalaya

The geology of the Kangchenjunga-Darjeeling profile in west Sikkim and north Bengal is consistent with the interpretation of the High Himalaya metamorphic sequence as a ductile channel approximately 15-20 km thick, extruding southwards, bounded by major ductile shear zones above (South Tibetan Detachment) and below (Main Central Thrust zone). In the Yoksam to Kangchenjunga section the entire slab is composed of migmatites or leucogranite sheets. Massive sills, or sub-horizontal to gently north-dipping sheets, of leucogranite make up the entire Kangchenjunga massif and extend west at least as far as Jannu. Shear sense indicators show both pure shear flattening fabrics and non-coaxial south-directed simple shear fabrics. The carapace of the melt-filled channel is seen around the southern margin of the Singallila ridge and the Darjeeling klippe, where inverted metamorphic isograds from sillimanite+K-feldspar grade down through kyanite and staurolite to garnet-biotite grade have been well documented. We present an interpretation of the crustal structure of the Sikkim Himalaya, based on field structural observations and utilizing the INDEPTH seismic profile to constrain sub-surface structure to the north. We suggest that widespread melting within sillimanite grade gneisses, triggered sudden and rapid ductile extrusion during the Miocene, by critically affecting the rheology of the middle crust under south Tibet. Leucogranite melts lubricated shear zones, facilitating rapid transport of heat towards the surface. When the leucogranite melts began to cool, ductile extrusion ended, as thrusting propagated down-section into the Lesser Himalaya. Since leucogranite expulsion was largely horizontal, rather than vertical, and since the STD normal sense shear zone is very low-angle, dipping to the north, decompression melting cannot be invoked as an origin for the leucogranites. The heat source for such widespread melting at such shallow levels in the crust can only be explained by having a highly radioactive protolith of Indian plate Proterozoic sediments. The overall map distribution and geometry of leucogranites, migmatites and metamorphic isograds is compatible with both the Everest-Makalu section to the west, and the Bhutan section, east of the Yadong-Gulu rift. We conclude that the channel flow model is a viable explanation for this section of the Himalaya.