Fault sources and kinematics of the 1897 Assam (MW8.1) and the 1934 Nepal (MS8.2) earthquakes retrieved by KF-NGA inversion and their seismotectonic implications

We performed automatic nonlinear geophysical source inversions of the 1897 Mw 8.1 Assam, India, and the 1934 Ms 8.2 Nepal earthquakes. Both events are key to understanding the seismotectonics and the hazard of the Himalaya–Assam region in the suture between the Indian subcontinent and Eurasia. Only macroseismic data were used to retrieve geometric and kinematic information on the two sources; the data were treated using the kinematic function model of radiation of S body waves in the 10–120 km distance range from the source. We used a genetic algorithm with niching in the inversion and found the minimum variance model to be a north-dipping fault of Himalayan type (strike, 289° 11°; dip, 36° 4°; rake, 87° 16°). This fault source produced a striking synthetic match of the empirical reference; that is, the macroseismic field intensities. The result gives credit to the interpretation of the regional fault bordering the southern edge of the Shillong plateau (Assam) as an example of thrust kinematics in the context of the subducting Indian subcontinent. The Himalayan source of the 1934 earthquake retrieved by the intensity-based inversion is fully compatible with that obtained from instrumental data, new field observations, and paleoseismological evidence. Thus the inversion method was again verified. Furthermore, the epicenter and the fault-plane solution of the 1934 earthquakewas demonstrated to be more reliable than the epicenter (26.5° N, 86.5° E, Bihar–Nepal border region) determined in 1980 from the instrumental data of 1934.