Tremor Waveform Extraction and Automatic Location With Neural Network Interpretation

Abstract

Active faults release tectonic stress imposed by plate motion through a spectrum of slip modes, from slow, aseismic slip, to dynamic, seismic events. Slow earthquakes are often associated with tectonic tremor, nonimpulsive signals that can easily be buried in seismic noise and go undetected. We present a new methodology aimed at improving the detection and location of tremors hidden within seismic noise. After identifying tremors with a classic convolutional neural network (CNN), we rely on neural network attribution to extract core tremor signatures. We observe that the signals resulting from the neural network attribution analysis correspond to a waveform traveling in the Earth's crust and mantle at wavespeeds consistent with seismological estimates. We then use these waveforms signatures to locate the source of tremors with standard array-based techniques. We apply this method to the Cascadia subduction zone, where we identify tremor patches consistent with existing catalogs. This approach allows us to extract small signals hidden within the noise, and to locate more tremors than in existing catalogs.