Interpretation of observed ground level changes due to the 1905 Kangra earthquake, Northwest Himalaya

Abstract

The Kangra earthquake of 1905 in the Northwest Himalaya was the first Indian earthquake for which coseismic ground elevation changes were measured through repeat precision levelling. The levelling Une from Saharanpur to Mussoorie via Dehra Dun was situated near the southeastern extremity of the NW-SE trending meizoseismal area. Maximum uplift of 14.3 cm was measured at a bench mark in Dehra Dun. Some details of the levelling data having been published recently, data for eight bench marks were considered for interpretation in this study. In principle, these are sufficient for us to estimate eight parameters of a rectangular rupture buried in a homogeneous Poisson solid. But because of some uncertainties regarding the bench mark positions and also the relatively small amounts of uplift measured at the bench marks, we have preferred to use the data as a means to check the reasonableness of rupture models proposed on other grounds rather than to propose a new model independently. The observed elevation changes are not in accord with the view that the Kangra earthquake was caused by a single 100 km long rupture along the Shahpur-Mandi sector of the Northwest Himalaya. However, the data can be explained if, in addition, a second smaller rupture extending on either side of the levelling Une near Dehra Dun is assumed. The data can also be explained if the rupture during the earthquake had (1) a length of 280 km parallel to the general strike of the Himalaya from around Kangra up to a short distance (about 10–15 km) northwest of the levelling Une near Dehra Dun, (2) a breadth of 80 km perpendicular to the strike of the Himalaya, (3) a northeasterly dip of about 5°, and (4) a depth of 10–15 km along its SW edge. Finally, the elevation data rule out any role for the Main Boundary Thrust in the occurrence of the Kangra earthquake. This major tectonic feature demarcating the geological boundary between the outer and the lesser Himalaya thus appears to be unimportant in the current seismotectonics of the region and the subduction of the Indian lithospheric plate under the mountains.