The Far Reach of Megathrust Earthquakes: Evolution of Stress, Deformation and Seismicity Since the Sumatra-Andaman Rupture

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

The December 26, 2004 Sumatra-Andaman earthquake was the largest event since the 1960s and the first magnitude > 9 earthquake to occur in the era of modern space geodesy and broadband seismology. This earthquake produced measurable static surface displacements at distances as large as 4500 km. Given the tremendous size of the 2004 earthquake, a ?second wave? of accelerated asthenospheric flow has developed that is going to perturb the regional deformation field for many decades. In addition, other postseismic deformation processes such as localized afterslip and poroelastic rebound, will also affect the regional deformation field. In fact, deformation measurements of the first few months following the Sumatra-Andaman earthquake can be explained quite well by either kinematic models of localized afterslip along and mostly below the rupture, or by viscoelastic relaxation. The 2004 earthquake was followed three months later by a great magnitude 8.7 event on an adjoining segment of the subduction zone, and again in 2007 by a magnitude 8.4 earthquake. This earthquake sequence provide the rare opportunity to fundamentally improve our understanding of the earthquake cycle of megathrust ruptures and the constitutive properties of the adjoining oceanic and continental crust and upper mantle. The knowledge the research team is gaining from this work will be applied towards improved stress models of how earlier ruptures influenced the timing and nature of subsequent events. This study includes a comprehensive analysis of GPS-measured deformation, seismicity changes and other geophysical data in order to develop well-constrained models of regional deformation and stress transients. The main goals of the project are to (1) differentiate the contributions from the various postseismic deformation processes, (2) to quantify the reach and distribution of transient deformation and stress across the region, especially from deep-seated relaxation in the upper mantle, and (3) to determine the extent, magnitude and pattern of time dependent stress perturbations affecting active seismicity throughout the region. In their modeling, the investigators rely on analytical viscoelastic normal mode solutions for post-earthquake relaxation on a spherically symmetric, compressible, self-gravitating Earth and on numerical finite element techniques. The finite element techniques allow for consideration of lateral heterogeneity across the subduction plate boundary zone and for the role of the subducting slab. The most important impact of this work is to contribute to earthquake hazard assessments for the Burma-Andaman-Sumatra-Java plate boundary zone by improving the accuracy of earthquake scenarios and probabilities.