An optimized online version of NESTORE software package for the forecasting of strong aftershocks: an application to Italian clusters

NESTORE (Next STrOng Related Earthquake) is a recently developed algorithm (Gentili & Di Giovambattista 2017, 2020) to recognize clusters in which a strong mainshock is followed by an aftershock of similar magnitude. Specifically, NESTORE labels clusters as type A if the magnitude difference between the mainshock and its strongest aftershock is less than or equal to 1, otherwise as type B. After an intense earthquake, the prediction of strong following events is strategic for civil protection purposes. In fact, already weakened structures may suffer further damage, increasing the risk of collapse and casualties. The goal of NESTORE is a near real-time estimation of the probability that the ongoing cluster is type A. The software is based on a set of parameters (features) of seismic clusters calculated at increasing time intervals after the mainshock. In particular, the algorithm exploits a training procedure with a feature-based machine learning approach. The features are related to the evolution of the number of events and their space-magnitude distribution over time. To make NESTORE a suitable software for online sharing, we optimized its structure. Specifically, some functions have been improved, further ones have been added, and a new name structure has been introduced to better characterize the three independent modules of NESTORE (cluster identification, training, and testing). This software renovation has been developed in the frame of project “Analysis of seismic sequences for strong aftershock forecasting” funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation within the collaboration in science and technology between Italy and Japan. We applied this new version of NESTORE to Italian seismicity and in particular to North-Eastern Italy, and obtained information on the features with best performances in terms of type A and B cluster discrimination.

 

Funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation