Abstract

We present an analysis of time-dependent precursory source processes and associated ground surface deformation prior to the great earthquake rupture of a long seismic gap zone along a transform plate boundary. This work is based on a theoretical model proposed by Li and Rice (1983) and reviewed briefly here. In the model, thickness-averaged stress transmission in the lithosphere is analyzed by a generalized Elsasser plane stress model which includes coupling to a viscoelastic asthenosphere. Upward progression of preseismic rupture at each section along strike is analyzed as quasi-static extension, in local antiplane strain, of an elastic-brittle crack; this provides the necessary boundary condition (in the context of the “line-spring procedure”) along strike for the Elsasser plate model. The approach to instability is simulated using what are thought to be typical material and tectonic parameters. It is found that prior to a large earthquake, the shear strain increases rapidly near the seismic gap zone but tends to diminish at moderate distances from the plate boundary. Such a strain-reversal phenomenon, although difficult to measure at present due to its small magnitude, may be helpful in the interpretation of geodetic measurements and in the planning of geodetic network locations for the purpose of earthquake prediction. The acceleration of slip, with consequent acceleration of stress increase in the upper brittle crust, predicted by the model may provide a plausible explanation for nonlinear features associated with seismicity preceding some major events (see, e.g., Raleigh et al., 1982). A “precursor time”, tprec, during which anomalous strain rate at the seismic gap exceeds twice its background level, is found to be in the range of a couple of months to 5 yr, depending on the tectonic loading rate; tprec varies approximately as R−1.3 (where R is a dimensionless loading rate parameter). This precursor time may be associated with physical anomalities related to rapid stress increase such as ground tilting, water table variation, and radon emission which precede some large earthquakes. Consistent with observations, the analysis predicts longer precursor times for larger earthquakes.

First Page Preview

First page PDF preview
You do not currently have access to this article.