Next-Generation Seismic Monitoring with Neural Operators

Seismic phase picking is the task of annotating seismograms with seismic wave arrival times and underpins earthquake monitoring operations globally. State-of-the-art approaches for phase picking use deep neural networks to annotate seismograms at each station independently; this is in stark contrast to the way that human experts annotate seismic data, in which waveforms from the whole network are examined simultaneously. With the performance gains of single-station algorithms approaching saturation, it is clear that meaningful future advances will require algorithms that can naturally examine data for entire networks at once. Here, we introduce a general-purpose network-wide phase picking algorithm, PhaseNO, that is based on a recently developed machine learning paradigm called Neural Operator. PhaseNO can use data from any number of stations arranged in any arbitrary geometry to pick phases across the entire seismic network simultaneously. By leveraging the natural spatial and temporal contextual information, PhaseNO achieves superior performance over leading baseline algorithms by detecting many more earthquakes, picking many more phase arrivals, yet also greatly improving measurement accuracy. Following similar trends being seen across the domains of artificial intelligence, our approach provides but a glimpse of the potential gains from fully-utilizing the massive seismic datasets being collected around the world.