作者: Natalia Poiata , Claudio Satriano , Jean-Pierre Vilotte , Pascal Bernard , Kazushige Obara
DOI: 10.1093/GJI/GGW071
关键词: Signal 、 Pixel 、 Geology 、 Aftershock 、 Vertical seismic profile 、 Remote sensing 、 Coherence (signal processing) 、 Energy (signal processing) 、 Signal processing 、 Tectonics
摘要: We present a new methodology for detection and space–time location of seismic sources based on multiscale, frequency-selective coherence the wave field recorded by dense large-scale networks local antennas. The method is designed to enhance signal statistical features across array sensors consists three steps: processing, imaging, location. first step provides, each station, simplified representation extracting multiscale non-stationary characteristics, through multiband higher-order statistics or envelopes. This processing scheme account priori unknown transients, potentially associated with variety (e.g. earthquakes, tremors), prepare data better performance in posterior steps. Following imaging carried 3-D spatial mapping summation station-pair time-delay estimate functions. produces time-series images representing likelihood that pixel makes part source. Detection performed final maxima from images. demonstrate efficiency detecting locating low signal-to-noise ratio an example aftershock earthquake records stations International Maule Aftershock Deployment Central Chile. potential detect, locate characterize energy release possibly mixed radiation earthquakes low-frequency tectonic tremors further tested continuous southwestern Japan.