A Semblance-based Microseismic Event Detector for DAS Data

Distributed Acoustic Sensing (DAS) is becoming increasingly popular in microseismic monitoring operations. This data acquisition technology converts fiber-optic cables into dense arrays of seismic sensors that can sample the seismic wavefield produced by active or passive sources with a high spatial density, over distances ranging from a few hundred meters to tens of kilometers. However, standard microseismic data analysis procedures have several limitations when dealing with the high spatial (inter-sensor spacing up to sub-meter scale) sampling rates of DAS systems. Here we propose a semblance-based seismic event detection method that fully exploits the high spatial sampling of the DAS data. The detector identifies seismic events by computing waveform coherence of the seismic wavefield along geometrical hyperbolic trajectories for different curvatures and positions of the vertex, which are completely independent from external information (i.e. velocity models). The method detects a seismic event when the coherence values overcome a given threshold and satisfies our clustering criteria. We first validate our method on synthetic data and then apply it to real data from the FORGE geothermal experiment in Utah, USA. Our method detects about two times the number of events obtained with a standard method when applied to 24h of data.