[J05-P] Poster

[J05-P-08] Crustal stress and strain inversion of the Taiwan orogen using a mixed linear-nonlinear Bayesian approach

^○Ray Y. Chuang (National Taiwan University, Taipei, Taiwan)

The state of stress and strain in the crust is fundamental information for understanding bonging tectonic status and processes. It is common to use seismic data for inverting stress state at depths and geodetic data for estimating strain rates on the surface. However, it is arbitrary to determine spatial smoothing in regular damped least-squares for stress and strain inversions. Therefore, we extend the common stress and strain inversions into a mixed linear-nonlinear Bayesian framework, which provides objective way to estimate regularization parameters. In this method, one can estimate principal stress orientations with 3D smoothing by using focal mechanism solutions. The smoothing and fault plane selections can be determined by the Markov Chain Monte Carlo (MCMC) approach with the Metropolis algorithm. We show examples of stress and strain inversions for the Taiwan orogen and demonstrate how crustal stress changes after a large seismic event. The results show that the stress state of Taiwan is mainly dominated by strike-slip and reverse faulting with normal faulting in the northern and northeastern Taiwan, which is consistent with other studies. The stress state changes after the 1999 Chi-Chi earthquake is consistent with postseismic movement.