Rapid inference of a seismic wavefield is essential for disaster assessment and emergency rescue.Interpolation methods based on observed strong motion recordings,supported by dense seismic networks,provide a computati...Rapid inference of a seismic wavefield is essential for disaster assessment and emergency rescue.Interpolation methods based on observed strong motion recordings,supported by dense seismic networks,provide a computationally efficient alternative to traditional numerical simulations.However,these purely mathematical approaches fail to account for the physical processes of source rupture and wave propagation,as well as the spatial correlation and nonstationarity of frequency content in ground motion,thereby lacking physical interpretability.This study presents a rapid seismic wavefield inference method that incorporates the spatial correlation and nonstationarity of ground motion.We introduce a spatial correlation factor into the inverse-proportional-weighted interpolation method to infer the ground-motion Fourier amplitude spectra(FAS)at un-instrumented sites,using the FAS of nearby observed recordings.Furthermore,we apply an equivalent group velocity model to derive the ground-motion phase spectra,which are combined with the interpolated FAS to generate nonstationary ground-motion time histories.Verification results from an actual earthquake case indicate that the inferred FAS within the 0.1-25.0 Hz frequency band and ground-motion intensity measures are generally consistent with observed values.This study modestly improves the accuracy of seismic wavefield interpolation and offers a mechanistic basis for conventional mathematical interpolation methods from the perspective of engineering seismology.展开更多
基金Key Science and Technology Project of Ministry of Emergency Management of People’s Republic of China under Grant No.2024EMST040401Natural Science Foundation of Heilongjiang Province under Grant No.LH2022E119+2 种基金National Key R&D Program of China under Grant No.2023YFF0725004Fund of Institute of Engineering Mechanics,China Earthquake Administration under Grant No.2024B11National Natural Science Foundation of China under Grant No.42404078。
文摘Rapid inference of a seismic wavefield is essential for disaster assessment and emergency rescue.Interpolation methods based on observed strong motion recordings,supported by dense seismic networks,provide a computationally efficient alternative to traditional numerical simulations.However,these purely mathematical approaches fail to account for the physical processes of source rupture and wave propagation,as well as the spatial correlation and nonstationarity of frequency content in ground motion,thereby lacking physical interpretability.This study presents a rapid seismic wavefield inference method that incorporates the spatial correlation and nonstationarity of ground motion.We introduce a spatial correlation factor into the inverse-proportional-weighted interpolation method to infer the ground-motion Fourier amplitude spectra(FAS)at un-instrumented sites,using the FAS of nearby observed recordings.Furthermore,we apply an equivalent group velocity model to derive the ground-motion phase spectra,which are combined with the interpolated FAS to generate nonstationary ground-motion time histories.Verification results from an actual earthquake case indicate that the inferred FAS within the 0.1-25.0 Hz frequency band and ground-motion intensity measures are generally consistent with observed values.This study modestly improves the accuracy of seismic wavefield interpolation and offers a mechanistic basis for conventional mathematical interpolation methods from the perspective of engineering seismology.
