GNGTS 2021 - Atti del 39° Convegno Nazionale

GNGTS 2021 S essione 2.1 208 DECISION TREE-BASED MACHINE LEARNING APPROACH FOR FOLLOWING STRONG EARTHQUAKE FORECASTING IN SEISMIC CLUSTERS S. Gentili 1 , R. Di Giovambattista 2 1 National Institute of Oceanography and Applied Geophysics - OGS, Italy 2 Istituto Nazionale di Geofisica e Vulcanologia, 00143 Roma, Italy NESTORE (Next STrOng Related Earthquake) is a machine-learning method for the next strong following earthquake forecasting in seismic clusters. We define an o-mainshock (operative mainshock) as the first event in a cluster with magnitude Mm greater than a given threshold, and we are interested in clusters with at least one subsequent earthquake with magnitudeMa≥Mm-1; we refer to these clusters as “type A,” while to the others as “type B.” Our study focuses on a decision tree-based approach, which uses seismological features extracted from the first hours/ days seismicity of the cluster, in order to forecast if the cluster following a strong earthquake will be a type A. Specifically, we analyzed features based on the spatial and temporal distribution and on the energy of the earthquakes in the cluster. The method has been successfully applied to Northeastern Italy - Western Slovenia (Gentili and Di Giovambattista, 2020), to Italy as a whole (Gentili and Di Giovambattista, 2017) and to California (Gentili and Di Giovambattista, 2021). Recently, the NESTORE development has been funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation within the framework between the Government of Italy and the Government of Japan on Cooperation in Science and Technology for a bilateral project entitled “Analysis of seismic sequences for strong aftershock forecasting”. In this work, we will provide a summary of previous work and of the methodology future objectives. Acknowledgements Funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation References Gentili S. and Di Giovambattista R.; 2017: Pattern recognition approach to the subsequent event of damaging earthquakes in Italy. Physics of the Earth and Planetary Interiors, 266, pp. 1-17. Gentili S. and Di Giovambattista R.; 2020: Forecasting strong aftershocks in earthquake clusters from northeastern Italy and western Slovenia. Physics of the Earth and Planetary Interiors, 303, 106483. Gentili, S. and Di Giovambattista, R.; 2021: Strong following earthquake forecasting by a pattern recognition approach in California , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-4381, https://doi. org/10.5194/egusphere-egu21-4381, 2021. Corresponding author: sgentili@inogs.it