Did you hear that we can predict earthquake occurrence in the laboratory using Machine Learning? You might have already seen some great and interesting papers related to this topic. However, nearly all studies focused so far on the smooth faults, which display repetitive patterns of quasi-periodic slips, and limited roughness evolution. Contrary, natural faults display structural complexity (fault networks) and enhanced roughness. Our new study published recently in the Earth and Planetary Science Letter attempts to verify our forecasting capabilities for rough faults using quite unorthodox feature pool derived from acoustic emission data. So if you wish to see how earthquake prediction goes for the most complex scenario, see our new contribution!
Reference:
Karimpouli, S., D. Caus, H. Grover, P. Martínez-Garzón, M. Bohnhoff, G. C. Beroza, G. Dresen, T. Goebel, T. Weigel, and G. Kwiatek (2023). Explainable machine learning for labquake prediction using catalog-driven features. Earth and Planetary Science Letters 622, 118383, DOI: 10.1016/j.epsl.2023.118383. [ Article Page ] [ Download open-access Article ]