HR: 0800h
AN: S31A-0214    [Abstracts]
TI: 3-D Seismogenic Stress Fields in and around Japan, Inferred from the CMT Data Inversion
AU: * Terakawa, T
EM: terakawa@eps.s.u-tokyo.ac.jp
AF: Department Earth and Planetary Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan
AU: * Terakawa, T
EM: terakawa@eps.s.u-tokyo.ac.jp
AF: CREST, Japan Science and Technology Agency, Kawaguchi Center Building, 1-8 Honcho, Kawaguchi-shi, Saitama, 332-0012, Japan
AU: Matsu'ura, M
EM: matsuura@eps.s.u-tokyo.ac.jp
AF: Department Earth and Planetary Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan
AU: Matsu'ura, M
EM: matsuura@eps.s.u-tokyo.ac.jp
AF: CREST, Japan Science and Technology Agency, Kawaguchi Center Building, 1-8 Honcho, Kawaguchi-shi, Saitama, 332-0012, Japan
AB: We developed a robust inversion method to estimate the pattern of the stress fields related to earthquake generation (seismogenic stress fields) from the centroid moment tensors (CMT) of seismic events by using Akaikefs Bayesian information criterion (ABIC). The CMT solution of a seismic event is conventionally defined by the surface integral of 2-D moment tensor density over a rupture area. Applying Gauss' divergence theorem, we can transform the surface integral of 2-D moment tensor density into the volume integral of stress release over a finite elastic region surrounding the dynamic rupture area. The volume integral representation of CMT is more essential than the surface integral representation, because dynamic rupture growth is controlled by energy flow into the rupture zone from the surrounding region storing elastic strain energy. Since the occurrence of an earthquake releases some part of the seismogenic stress field around its hypocenter, we can relate CMT data with the seismogenic stress field with the volume integral representation. We represent the CMT of a seismic event by a weighted volume integral of the true but unknown seismogenic stress field. The weighting function is taken to be a 3-D Gaussian-type distribution with its peak at the hypocenter and variance proportional to the two- thirds power of the seismic moment. Representing each component of the seismogenic stress field by the superposition of a finite number of 3-D basis functions, we obtain a set of linear observation equations to be solved for the expansion coefficients (model parameters). We introduce prior constraint on the roughness of the seismogenetic stress field and combine it with observed data to construct a Bayesian model with a hierarchic flexible structure controlled by hyper-parameters. The optimum values of the hyper-parameters are objectively determined from observed data by minimizing ABIC. Given the optimum values of the hyper-parameters, we can obtain the best estimates of model parameters by using a maximum likelihood algorithm. We applied the inversion method to observed CMT data in and around Japan (the NIED Moment Tensor Catalogue, 1997.1.31-2007.1.31) to reveal the 3-D patterns of seismogenic stress fields associated with plate subduction. We compared the stress patterns in the northeast Japan arc and the Ryukyu arc. In both regions the stress patterns in the shallow part of the oceanic plates and the descending slabs are characterized by normal and reverse faulting with strikes parallel to the trench axes, respectively. However, the stress patterns in the back- arc regions are quite different from each other: reverse faulting in the northeast Japan arc and normal faulting in the Ryukyu arc. The difference in the stress pattern results from the difference in the tectonic loading mechanism.
DE: 3260 Inverse theory
DE: 7230 Seismicity and tectonics (1207, 1217, 1240, 1242)
DE: 7240 Subduction zones (1207, 1219, 1240)
DE: 8123 Dynamics: seismotectonics
DE: 8164 Stresses: crust and lithosphere
SC: Seismology [S]
MN: 2007 Fall Meeting