Simultaneous inversion of about 5000 seismic phase data of 391 aftershocks of the 29 March 1999 Chamoli earthquake (mb 6.3), recorded by 28 closely spaced temporary and permanent seismic stations, sheds new light on understanding the seismogenic structure of the source area. The best-located 160 aftershocks (md 2.0-4.0) with root mean square error less than 0.2 sec and epicentral and depth error less than 1 km are clustered at a depth range of 8-16 km, above the plane of detachment, the interface between the Himalayan sedimentary wedge and the Indian shield. The mainshock nucleated by thrust faulting at the “fault end,” where the Alakananda fault meets the plane of detachment at a depth of 15 km, and the aftershocks occurred by shear adjustment at the Alakananda fault.
The 3D P- and S-wave velocity structures of the source area show an east-west-trending low-velocity zone (LVZ). The LVZ is comparable with the north-dipping or near-vertical Alakananda fault on the surface. It extends from the surface down to 15 km depth. It broadens in the north-south direction at shallower (0-10 km) depth and narrows at deeper depth (10-15 km). The P-wave tomography shows a high-velocity zone (HVZ) at the eastern margin of the LVZ. The mainshock and the aftershocks occurred at the contact of these two zones and within the HVZ. The S-wave velocity structure resembles that of the P wave, but it shows an LVZ at the mainshock source area at a depth of 15 km where the higher P-wave velocity structure (HVZ) is observed. The seismogenic source area is thus characterized by a higher VP/VS or higher Poisson's ratio at the fault end. This anomaly may be due to fluid-filled fractured rock matrix, which might have contributed to the nucleation of the mainshock.