GNGTS 2023 - Atti del 41° Convegno Nazionale

Session 1.1 GNGTS 2023 Fault structures of the Lake Pertusillo reservoir induced seismicity (Southern Italy) highlighted by a enhanced template-matching catalogue L. Valoroso 1 , D. Piccinini 2 , L. Improta 1 , S. Gaviano 2 1 Istituto Nazionale di Geofisica e Vulcanologia (INGV), Osservatorio Nazionale Terremoti (Roma) 2 Istituto Nazionale di Geofisica e Vulcanologia (INGV), Sezione di Pisa The Val d’Agri basin in the southern Apennines seismic belt is one of the central Mediterranean regions with highest seismic hazard, testified by the strong 1857 M7.1 Basilicata earthquake ( Mallet, 1862). The Val d’Agri is also a natural laboratory for studying anthropogenic seismicity since it hosts two well-documented cases of induced seismicity: the first is a wastewater-injection induced seismicity case observed at the Costa Molina injection well (Improta et al., 2015); the second case, which is investigated in this study, is the protracted reservoir induced seismicity (RIS) related to the seasonal water level oscillation in the medium-size Pertusillo water reservoir (PWR) (Valoroso et al., 2009; Stabile et al., 2014). Protracted RIS has been associated with destructive earthquakes, as testified by the 1967 M6.3 earthquake in the Konya region (India), followed by persistent moderate-magnitude M5+ events. In recent years, many other M4+ RIS earthquakes have been observed worldwide (Gupta, 2022 and references therein), fostering the seismological community to better understand the physical mechanism behind this process and its repercussions on the seismic hazard. The PWR seismicity is characterised by low energy with largest events around M L 3. However, a deeper understanding of this phenomenon is compelling due to the presence of active faults, potential source of M6+ earthquakes close to the lake (Improta et al., 2010), in an area where the source of the large M7.1 1857 Basilicata earthquake is still debated (DISS Working Group, 2021; Bello et al., 2022). We study the PWR induced seismicity through a very-high resolution earthquake catalogue computed by applying template-matching detection techniques to seismic data recorded during a 1-year-long dense passive survey. Template-matching (TM) uses cross-correlation between continuous seismic data and known earthquake recordings ( templates ) to detect previously unidentified events in continuous data. This method is very useful compared to traditional techniques (e.g., STA/LTA), especially in areas characterised by dense clusters of small-magnitude seismic events (e.g., Shelly et al., 2016; Ross et al., 2019 among many others). Enhanced TM catalogues have low completeness magnitude (M C ) allowing to image the geometry of faults and fractures and to furnish an extreme detail in describing the spatio-temporal seismicity evolution.