5:15 PM - 7:15 PM
[SSS09-P04] The frequency-Bessel transform method based on surface-integral for seismic records of earthquake event

Keywords:Surface wave, Multimodal dispersion curves, Multi-component dispersion curves, Leaking modes, F-J method
The frequency-Bessel transform (F-J) method is a well-established tool for multimodal surface wave tomography, particularly in ambient noise imaging. However, its application to seismic records of earthquake event encounters challenges due to the azimuthal anisotropy of Green's functions. The original F-J method is based on azimuthal isotropy assumption, which may probably confuse the dispersion curve extracted from earthquake events when seismic array spanning a wide azimuthal range. In previous studies, this challenge was typically addressed by carefully selecting stations within a specific azimuthal range relative to the seismic source to minimize anisotropic interference. To address these limitations theoretically, starting from the semi-analytical solutions of wave-fields, we introduce the idea of surface integral to the F-J transform by deriving the surface-integral-based expressions of expansion coefficients, whose extreme value can provide us the solutions of dispersion curves. The results of synthetic tests and real data cases demonstrate the feasibility of our method, confirming its capability to improve the quality from seismic records of earthquake event. Furthermore, our method extends the F-J framework to handle multi-component seismic event data. By using time-window-based mode separation, it successfully extracts not only Rayleigh and Love wave components but also leaking modes with more dispersion characteristics. This advancement may contribute to subsurface imaging by providing additional information and improved constraints on structural properties. The surface-integral F-J method offers a promising approach for seismic data processing, facilitating a more comprehensive utilization of multimode dispersion information from seismic records of earthquake event.
