GNGTS 2019 - Atti del 38° Convegno Nazionale

130 GNGTS 2019 S essione 1.1 Recently, through the integration of shallow electrical resistivity profiles, electromagnetic soundings, ambient noise recordings and seismic anisotropy data, Villani et al. , (2019) proposed an interpretation of two cross-sections within the PGC basin. However, high-resolution information on location, geometry and throw of buried fault splays, as well as fine details of the continental infill architecture, are still lacking. Active-source seismic profiling is crucial to fill this gap of knowledge. To improve the subsurface basin imaging, on September-October 2017 we acquired three high-resolution seismic profiles within the PGC with a multi-fold wide-aperture acquisition geometry that allows collecting simultaneously reflection and refraction dense data in a wide offset range. The first profile (labelled Vettore ) is 2635 m-long, trends WNW, and crosses nearly orthogonal the 2016 surface ruptures as well as an important splay of the VBFS. The second profile (labelled Maneggio ) is 2035 m-long, trends WNW, and intercepts one of the basin-bounding faults to the south. The third profile (labelled Castelluccio ) is 3355 m-long, trends NNE, and connects the first two profiles. All profiles were collected by deploying a 1195 m-long linear array of 240 vertical geophones 5-m spaced that was shifted using a roll-along technique to cover the overall profiles length. The geophones were connected to ten 24-bit seismometers. We used a 6-tons vibroseis (Minivib of IVI©) as a powerful yet safe energy source, with average 5-m spacing. A total number of 1212 shot- gathers (consisting of 290880 seismic traces) were collected. We calibrated our geophysical data with the nearby available shallow boreholes (Ge.Mi.Na., 1963), some of which reach the pre-Quaternary limestone bedrock, and geological data (by updating the map of Pierantoni et al. , 2013, in particular regarding the PGC deposits). Our imaging strategy combines reflection data processing and non-linear refraction tomography. For each profile, we handpicked first-arrival traveltimes in order to perform a non-linear tomographic inversion of a dense dataset (94320 readings for profile Vettore , 47520 for line Maneggio and 37920 for line Castelluccio ). We use a multi-scale imaging strategy, successfully applied for imaging complex thrust-structures, fault-controlled basins and shallow fault zones (Improta et al. , 2002, 2003; Improta and Bruno, 2007; Villani et al. , 2015, 2017). Seismic tomography is fundamental for inferring the large-scale geometry of the basin. The obtained P-wave velocity models illuminate the PGC structure down to 350-500 m depth along the three key sections. The limestone bedrock is imaged as a high-velocity region (Vp ~4000-5500 m/s) showing a complex geometry due to the presence of a large number of subsurface faults defined by abrupt lateral Vp changes. We apply an advanced processing flow to the reflection data taking advantage of the high fold and large offset range of the Common-Mid-Point (CMP) gathers that show reflections with an overall good signal-to-noise ratio and coherency (see methodological details in Bruno et al. , 2019). The obtained migrated and depth-converted stack sections show details of the depositional architecture down to about 700-800 m depth. Within the continental infill, we observe clear and continuous high-amplitude reflections, corresponding to low-Vp regions in the tomographic models (Vp < 2500 m/s), define alternating fine and coarse sediments (silty sands and sandy gravels), which may be related to distal alluvial fan or to lacustrine deposits Fig. 2 - The Minivib used for the acquisition of seismic data in the Pian Grande di Castelluccio basin.