[S-SS09] Seismic wave propagation: Theory and Application

[SSS09-10] Ambient Noise Differential Adjoint Tomography Applied to a Linear Array in Los Angeles Basin

*Xin Liu^1,2, Gregory C Beroza¹ (1.Stanford Earth, 2.JAMSTEC)

Keywords:ambient noise adjoint tomography, near surface shear velocity

The small time shifts that are used in ambient-field interferometric Green’s function analysis may arise from either structural anomalies of interest, or spatio-temporal behavior of the noise source, which is usually not of interest. In seismic noise interferometry, the differential time kernel has been shown to be much less sensitive to variations in the spatio-temporal distributions of the noise source and more sensitive to the structural anomalies of interest, thus using differential measurements is an effective approach for reducing bias in tomographic images (Liu 2020). In this study, we generalize this technique by deriving the shear-wave velocity directly from differential time kernels for fundamental mode Love waves in noise cross-correlations along a linear array spanning ~37 km in the Los Angeles basin without the 1D depth conversion step that assumes a layered medium. The initial results show strong sensitivity to near-surface shear velocity structure in the uppermost 500 m. We then apply iterative inversion for shear velocity, and will present an updated near-surface shear velocity structure that should help to constrain future urban earthquake ground motion.