SEISMICITY, STRUCTURE AND RHEOLOGY OF THE LITHOSPHERE IN THE LAKE BAIKAL REGION

Abstract

This paper combines observations of seismicity, gravity, topography and thermal and velocity structures to investigate the rheological properties of the lithosphere in the Lake Baikal region. We examine the seismogenic thickness (Ts) using 25 earthquakes of Mw 5.1-7.1, whose full source parameters have been determined by inversion of teleseismic waveforms, 13 of which are presented here for the first time. These 25 events, plus six others (Mw 5.0-5.8) whose depths are well constrained, show that moderate earthquakes occur at depths up to ~30 km in the northeast Baikal rift. Based on the teleseismic waveform modelling results and published relocations of microearthquakes using regional networks, we conclude that the mantle is not a significant source of seismicity in the Baikal region. Using the admittance between free-air gravity and topography, we estimate the effective elastic thickness (Te) in the region to be between 5 and 20 km. Nowhere do the data require that Te > Ts, consistent with the simple interpretation that the long-term strength of the lithosphere resides in its seismogenic layer. A weak mantle in the Baikal region can be explained by its high temperature, which we estimate by combining local geotherm estimates with the regional upper mantle velocity structure, obtained from fundamental and higher-mode surface waves. Geotherms are fitted to pressure and temperature estimates from mantle nodules at four sites, both within and outside the Siberian shield. In order to constrain the temperatures at the Moho, we estimated crustal thicknesses using teleseismic receiver functions. Moho temperatures are estimated to exceed ~550°C beneath the Siberian shield and are higher in the more recently deformed mountain belts to the south. Based on a reassessment of oceanic geotherms and seismicity, it seems likely, therefore, that the mantle in the Baikal region is too hot to be a source of long-term strength. This is consistent with the recent suggestion that the distribution of mantle seismicity in both the oceans and the continents is dependent on temperature alone. Finally, we note that results from S-wave tomography studies, combined with the observed locations of rift-related earthquakes, lead us to suspect that the frequently published position of the edge to the Siberian shield at the surface provides a poor description of that same boundary at depth. © 2006 The Authors Journal compilation © 2006 RAS.