Abstract
An immediate report of the source focal mechanism with full automation after a destructive earthquake is crucial for timely characterizing the faulting geometry, evaluating the stress perturbation, and assessing the aftershock patterns. Advanced technologies such as Artificial Intelligence (AI) has been introduced to solve various problems in real-time seismology, but the real-time source focal mechanism is still a challenge. Here we propose a novel deep learning method namely Focal Mechanism Network (FMNet) to address this problem. The FMNet trained with 787,320 synthetic samples successfully estimates the focal mechanisms of four 2019 Ridgecrest earthquakes with magnitude larger than Mw 5.4. The network learns the global waveform characteristics from theoretical data, thereby allowing the extensive applications of the proposed method to regions of potential seismic hazards with or without historical earthquake data. After receiving data, the network takes less than two hundred milliseconds for predicting the source focal mechanism reliably on a single CPU.
Highlights
An immediate report of the source focal mechanism with full automation after a destructive earthquake is crucial for timely characterizing the faulting geometry, evaluating the stress perturbation, and assessing the aftershock patterns
Recent efforts have been refined towards applying artificial intelligence (AI) technologies to estimate the source parameters because of its full automation, high efficiency, and human-like capability[4,5,6], which has been remarkably demonstrated in numerous seismic processing tasks such as earthquake detection[7,8], seismic phase picking[9,10,11], magnitude estimation[12], and others[13,14,15,16,17,18]
We train the Focal Mechanism Network (FMNet) model with the synthetic dataset and apply it to predict the focal mechanisms of four real earthquakes with magnitudes larger than 5.4 of the Ridgecrest sequence which occurred in July 2019 in southern California
Summary
An immediate report of the source focal mechanism with full automation after a destructive earthquake is crucial for timely characterizing the faulting geometry, evaluating the stress perturbation, and assessing the aftershock patterns. Obtaining the focal mechanism of an event from waveform data with processing effort as little as possible is more appealing Another recent progress that took advantage of an advanced search engine was performed[44] to estimate earthquake source parameters in less than 1 s. FMNet learns the universal characteristics of waveforms concerning the source focal mechanisms from the synthetic training data This considers the scenarios without enough historical source focal mechanisms for training the neural network model, especially for those regions with limited seismicity but having the potential seismic hazards. We produce a by-product of the encoder, which is a sparse representation of the input waveforms, to analyze the working mechanism and robustness of the FMNet
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
