GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 1.1 GNGTS 2024 To constrain the depth geometry of Serre and Citanova faults, a set of 10-km spaced seismological sectons (fve sectons for each fault) with a bufer projecton of 5 km was created (Fig. 2). the geometry at depth of the considered faults (red lines) was traced following earthquake clustering startng from the intersecton of the fault on the surface (black crosses in Figure 2). Clusters are visible, especially in S1 and S6 corresponding to the southern tp areas of the CF and SRF, respectvely. Despite the other sectons that do not exhibit clear clusters useful to infer the attude of the studied faults at depth, we traced their geometry considering the same trend of S1 and S6 and the earthquakes with the highest magnitude for each secton. Subsequently, using a trial-and- error approach, we geometrically tested the modelled planes with the known empirical scaling in order to fnd a reliable soluton for fault planes capable of generatng events with a magnitude of 7. Fault response modelling of the Citanova and Serre faults We combined the feld structural data, literature data, and kinematcs observed at surface with the seismic dataset in order to develop a reliable 3D model of the fault planes. According to the proposed fault model (Figures 3 a), the CF is an almost 44-km-long fault, roughly N40E-oriented with a plane dipping toward NW, whereas the SRF is a N30E-striking, 40-km long fault with a plane dipping toward NW. The average dip of the CF is 57°, while the SRF exhibits an average dip of approximately 60°. All geometric parameters are summarized in Table 1. The parameters derived from the 3D model were used to estmate the expected magnitude for each plane, assuming an actvaton of the faults for their entre length and using empirical relatonships. We used the surface rupture length (SRL) and the rupture area (RA) vs. magnitude empirical scaling both for Wells and Coppersmith (1994) and Leonard (2010). The resultng magnitudes for the given faults are comparable (see Table 1). Moreover, the Fault Response Modelling (FRM) module was applied to kinematcally test the model and verify the maximum vertcal displacement and its spatal distributon associated with the actvaton of the fault planes for their entre length (which is consistent with a maximum expected magnitude of approximately 6.8 – 7). The frst simulaton (Figure 3c) shows the displacement feld for the actvaton of both faults. The simulated vertcal displacement ranges from 0.5 to 1.7 m (with a maximum value equal to 2.2 m). The correspondence of the mesoseismal areas of 5 and 7 February and for the 1 March shocks, macroseismic data and simulated coseismic displacement (Figure 3c) confrms the choice of the CF and SRF faults, respectvely, as the most likely causatve sources for the considered events. The second simulaton (Figure 3d) shows the cumulatve displacements ranging from −354 m to 112 m. In this case, the displacement values exhibit an abrupt change across the fault traces; the vertcal cumulatve displacement of CF and SRF reaches almost 450 m (see also the vertcal displacement dz profles shown in Figure 3f), which is consistent with the minimum vertcal ofset estmated for these faults (see also Jacques et al., 2001).