---

Stress evolution following the 1999 Chi-Chi, Taiwan, earthquake: Consequences for afterslip, relaxation, aftershocks, and departures from Omori decay
Geophys. J. Int., (2009) doi: 10.1111/j.1365-246X.2008.04069.x
Chung-Han Chan and Ross S. Stein

Chinese Summary [Contact us for printable article: han@gfz-potsdam.de rstein@usgs.gov]

Non-technical summary: We explore how stress transfer controls the fault afterslip, post-earthquake relaxation and the evolution of aftershocks in a continental thrust fault system. The 21 September 1999 Mw=7.6 Chi-Chi shock, with a surface-cutting 30°-dipping fault merging into a near-horizontal décollement, is typical of continental thrusts throughout the world, and so inferences drawn from this uniquely well-recorded event may be widely applicable. In addition, both the coseismic and postseismic deformation were recorded by a dense GPS network spanning the northern half of the rupture. First, we find that the spatial and depth distribution of aftershocks and their focal mechanisms are consistent with the calculated Coulomb stress changes imparted by the fault rupture. Some 86% of the M≥3 aftershocks lie in regions for which the Coulomb stress increased by ≥0.1 bars. In addition, there is a 11-12% gain the percentage of focal mechanisms on which the shear stress has increased over the pre-Chi Chi period. Second, we find that the afterslip generally occurred where the calculated coseismic stress increased on the ramp, décollement and its down-dip extension. Inclusion of post-earthquake relaxation enhances the fit to the GPS data and aftershock distribution. We find that the seismicity rate increases beginning about 15 months after Chi-Chi in an annulus extending east of the main rupture. The spatial extent of the annulus resembles the calculated 0.05-bar Coulomb stress increase caused by afterslip on the decollement (see accompanying figure). Further, there is a 9-12% increase in the percentage of focal mechanisms with shear stress increases >0.01 bars imparted by the postseismic afterslip and relaxation. Thus we argue that postseismic stress changes for the first time can be shown to alter the production of aftershocks, as judged by their rate, spatial distribution, and focal mechanisms.