Geometry of the Main Himalayan Thrust and location of the locking line in the Garhwal Himalaya from receiver function CCP stacking

We use common conversion point (CCP) stacking of Ps receiver functions to image the crust and upper mantle beneath the western (Garhwal) Himalaya. Our seismic array of 21 broadband seismometers located at 79°-80°E longitude spans the Himalayan thrust wedge between the MFT and STD, and was operated for 21 months in 2005-2006 by India's National Geophysical Research Institute (NGRI). The Garhwal Himalaya has seen relatively little geophysical investigation compared to central Nepal, so this study and ones like it are important for constraining along-strike variations in crustal thickness and thrust wedge geometry. We calculate receiver functions using iterative time-domain deconvolution on a catalog of 450 events. Our CCP stacks show a nearly horizontal Moho at 35-40 km depth beneath the Lower Himalaya which deepens to 45-50 km beneath the Higher Himalaya. This Moho depth is ~5 km shallower than that observed using receiver function imaging in central Nepal and 10 km shallower than in the NW Indian Himalaya, implying an along-strike variation in either the thickness of pre-subduction Indian crust or in the strain pattern in the down-going plate. The observed thickness of subducted Indian crust is 25 km (the thickness between the ~15 km deep MHT and the ~40 km deep Moho). We image the MHT as a positive conversion (velocity increasing with depth) at 12-17 km depth beneath the Lesser Himalaya. This depth is consistent with that inferred from receiver function studies several hundred kilometers east in central Nepal, although in Nepal the conversion is negative (velocity decreasing with depth). A velocity increase with depth is consistent with the inferred (from geologic mapping) juxtaposition of rock types across the MHT in Garhwal. The MHT in our CCP stack takes on a negative amplitude ~10 km north of the surface location of the MCT-I (Munsiari Thrust in Garhwal), where the thrust may form a geometric or structural ramp that has, in central Nepal, been linked to the abrupt rise in topography at the base of the High Himalaya. Our image can be fit to within uncertainty by MHT geometries both with and without a ramp, but our image does place an upper bound of 5 km on the height of the ramp. The zone of negative amplitude on the MHT, at the likely ramp, coincides with a peak in electrical conductivity (>0.3 S/m) and is down-dip of a peak in microseismicity, suggesting the presence of fluids trapped below the out-of-sequence Munsiari Thrust and an associated transition from locked to creeping behavior on the MHT. Also suggestive of free fluids in this region are tomographic inversions of microseismicity which indicate low Vp and high Vp/Vs. Because fluids facilitate ductile deformation, this location is the likely location of the locking line on the MHT, consistent with the observation of concentrated microseismicity (which is known to focus at the down-dip edge of locked faults).