Item

Abstract

Metropolises such as Los Angeles, San Francisco, Tokyo, etc are faced with high earthquake risk, because their location within the actively deforming plate boundaries. The large population and density of infrastructure increase the earthquake risk exposure, and also make the earthquake monitoring difficult, due to various types of anthropogenic noise generated in cities and the logistical difficulties of instrument deployment.We have developed a deep-learning-based denoising algorithm, UrbanDenoiser, to filter out the strong urban seismological noise, and improve the capacity for earthquake monitoring in urban settings. UrbanDenoiser strongly suppresses noise relative to the signals, because it was trained on waveform datasets containing rich noise sources from the urban Long Beach dense array and high signal-to-noise ratio earthquake signals from the rural San Jacinto dense array. Application to urban seismic data shows that UrbanDenoiser can effectively suppress the high noise level at daytime, allowing us to work on the entire day’s data, not just during night when anthropogenic noise is lower, which doubles the utility of existing data. We apply UrbanDenoiser to a regional seismic network for the La Habra earthquake sequence in the urban area, and leads to an increased detection rate amounting to more than 4.5 times the number of detections in the Southern California Seismic Network catalog. Earthquake location using the denoised Long Beach data does not support the previous report of mantle seismicity beneath Los Angeles, but suggests a fault model featuring shallow creep, intermediate locking, and localized stress concentration at the base of the seismogenic zone.