GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 2.1 GNGTS 2024 Revealing anomalies in the Molise 2018 earthquake sequence S. Gentili 1 , P. Brondi 1 , G. Rossi 1 , M. Sugan 1 , G. Petrillo 2,3 , J. Zhuang 3 , S. Campanella 1 1 National Institute of Oceanography and Applied Geophysics - OGS, Italy 2 The Institute of Statistical Mathematics – ISM, Japan; 3 Scuola Superiore Meridionale – SSM, Italy Understanding the seismic clustering patterns in a region is crucial for statistical testing and forecasting. The NESTORE (NExt STrOng Related Earthquake – Gentili et al. 2023) algorithm has been shown to be a successful example of such applications of cluster analysis. It can be used for strong aftershock forecasting during an ongoing cluster. However, its forecasting performance can be compromised, if clusters are not properly detected. With various approaches, including traditional window-based methods, complex network-based techniques, stochastic declustering methods rooted in the Epidemic Type Aftershock Sequence (ETAS) model, and Principal Component Analysis (PCA), this study investigates the seismic sequence in Molise (southern Italy) in 2018. Ambiguous results were obtained when applying the NESTORE method. We utilize an enhanced template matching catalog and four different methods to identify earthquakes belonging to the cluster. While two methods indicate the presence of two distinct clusters in the same area and time period (from April to November 2018), Principal Component Analysis and Nearest Neighbor suggest the presence of a single cluster. Inconsistencies are attributed to the seismicity potentially being part of anomalous sequences. This research emphasizes the need for refined cluster identification methods and calls for further studies on how to characterize the specific seismic anomaly as in Molise in 2018. Acknowledgements Funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation