Microseismic Event Classification With Time-, Frequency-, and Wavelet-Domain Convolutional Neural Networks.

Passive seismics help us understand subsurface processes, for example, landslides, mining, geothermal systems, and so on and help predict and mitigate their effects. Continuous monitoring results in long seismic records that may contain various sources, which need to be classified. Manual detection and labeling of recorded seismic events are not only time-consuming, but can also be inconsistent when done manually, even in the case where it is done by the same expert. Therefore, an automated approach for the classification of continuous microseismic recordings based on a convolutional neural network (CNN) is proposed, with a multiclassifier architecture that classifies earthquakes, rockfalls, and low signal-to-noise ratio quakes. Furthermore, we propose three CNN architectures that take as input time-series data, short-time Fourier transform (STFT), and continuous wavelet transform (CWT) maps. The suitability of these three networks is rigorously assessed over five months of continuous seismometer recordings from the active Super-Sauze landslide in France. We observe that all three architectures have excellent and very similar performance. Furthermore, we evaluate transferability to a geographically distinct seismically active site in Larissa, Greece. We demonstrate that the proposed network can detect all 86 cataloged earthquake events, having only been trained on the Super-Sauze dataset and shows good agreement with manually detected events. This is promising as it could replace the painstaking manual labeling of events in large recordings.