14:15 〜 14:30
[SSS07-09] Joint Transdimensional Bayesian Inversion of Rayleigh Wave Ellipticity and Dispersion for Near-Surface Shear-Wave Imaging
キーワード：seismic ambient noise, shear-wave velocity, joint inversion, transdimensional Bayesian
Understanding near-surface geology is crucial for assessing seismic hazards, especially in densely populated urban areas. Consequently, seismic ambient noise analysis has emerged as an effective tool for modeling near-surface shear-wave velocity (VS) structures. This method uses background vibrations from natural sources such as ocean waves or traffic to extract surface wave properties, such as the Rayleigh wave dispersion and ellipticity. Nowadays, the most common practices involve conducting single-station measurements to obtain ellipticity and array measurements to obtain dispersion. However, these data are rarely integrated during the inversion step, potentially leading to nonunique solutions. This study explores the feasibility of joint inversion of Rayleigh wave dispersion and ellipticity curves using a transdimensional Bayesian framework. By parameterizing the model space with 2D Voronoi cells and considering the lateral spatial correlation of the inversion result, this approach aims to provide a statistical analysis of the solution and reduce nonuniqueness. We demonstrated the proposed methodology using real data from the seismic array in Greater Jakarta. Our results present a significantly refined solution of the VS model of the Jakarta Basin. It is important to note that more than the ellipticity data alone would be required to invert for VS structures. However, we can highlight the complementary of Rayleigh wave ellipticity data to dispersion data for inferring 1-D VS depth profile in a joint inversion, especially in resolving the shallowest layer. This study suggests that integrating both Rayleigh wave dispersion and ellipticity data not only improves the model's accuracy but also adds valuable information to the depth resolution of near-surface imaging.