Thickness of underthrust Indian crust in the Garhwal Himalaya

Using common conversion point (CCP) stacking of teleseismic receiver functions (RFs), we image the Moho and Main Himalayan Thrust (MHT) beneath the Garhwal Himalaya (Alaknanda Valley, Uttaranchal, India). Beneath the Main Frontal Thrust (southern margin of the Himalaya), we image the Moho at a depth of 40 km, increasing to a depth of 55 km beneath the South Tibet Detachment (STD). These depths are shallower than the 50-55 km found c. 300 km west by the HIMPROBE team (Rai et al., GRL 2006), but are comparable to those found c. 600 km east by the HiCLIMB team (Nabelek et al., Science 2009). However, the HiCLIMB team imaged a Moho with a nearly constant dip, whereas in our region we image an apparent steepening of the Moho beneath the Greater Himalaya which accommodates most of the observed 15 km Moho depth increase. This steepening is spatially coincident with both the rise in elevation of the Tibetan plateau and with the Munsiari Thrust (MT), which we also image. The Munsiari Thrust is a brittle thrust fault that is the southern-most, structurally-lowest, and currently-active component of the Main Central Thrust (MCT) deformation zone. Beneath the Lesser Himalaya we image the MHT at a depth of 15 km and the Moho at a depth of 40 km, suggesting that the thickness of Indian continental basement subducting beneath the Himalaya in this region is as thin as 25 km. These results also suggest that ~15 km of thrust sheets lie beneath the Lesser Himalaya. Beneath the STD, this increases to ~25 km of thrust sheets. Our data are from a broadband seismic array consisting of ~20 stations with ~10 km spacing arranged linearly from SW to NE across the Himalayan thrust belt at ~80°E, from the Main Frontal Thrust (MFT) to the South Tibet Detachment (STD). The array was operated by India's National Geophysical Research Institute in 2005-2006. We generate crustal images using stacking of P-S receiver functions. We calculate receiver functions using an iterative time-domain method and depth-convert them by back propagation in an assumed velocity model, then bin and stack them to obtain two-dimensional images. Our model has a bin size of 1 km in depth and 10 km horizontally.