GNGTS 2022 - Atti del 40° Convegno Nazionale

GNGTS 2022 Sessione 2.1 195 (Lolli et al. , 2020). As a surveillance region R we consider a regular tessellation of the Italian territory made by squares with side L = 30 √− 2 km from 7°E to 19°E in longitude and from 36°N to 47°N in latitude. Because the completeness of the seismic catalogue is poor in offshore areas, according to Gasperini et al. (2021), we consider in our analysis only the squares within which at least one earthquake with Mw≥4.0 occurred inland from 1600 to 2021 according to the CPTI15 catalog excluding then all cells that are completely on marine areas. The cells that make up the surveillance region R are 177 (Fig. 1). We chose as target shocks threshold m T = 5.0 because in Italy such earthquakes potentially cause damage to buildings and threat the health and the life of inhabitants. This choice is also consistent with most of the applications of EEPAS model to other regions of the World (Evison and Rhoades 2005, Rhoades 2011; Rhoades and Evison 2004; Rhoades 2007).We also chose the time interval from 1990 to 2021 for fitting and retrospective forecasting evaluation of the EEPAS model as the accuracy and completeness of the Italian catalog improved significantly since 1990 (Gasperini et al. , 2013). Fig. 1 - Tessellation of the Italian territory region used for the fitting of parameters and for the retrospective experiment. The thick black line delimits the analysis region R. The cells that compose are only those within which at least one earthquake with M ≥4.0 from 1600 to 2021 have occured according to CPTI15 catalogue (Rovida et al ., 2020) and have 30 √− 2km of side.