GNGTS 2013 - Atti del 32° Convegno Nazionale

2D-3D GPR survey across the Castrovillari Fault. A wide number of 2D GPR profiles and a 3D GPR volume were acquired during the last field survey across the Castrovillari fault (Fig. 1). The survey site has been identified by the UR of Dr. Francesca Cinti (INGV), after an accurate geological survey on the area. The same was employed for the data recording. In the present paper only some main results on the preliminary 2D 300 MHz profiles are reported. The 2D profile in Fig. 3a was recorded close to an outcrop showing the studied fault and close to the area on which the 3D GPR acquisition grid was materialized. An inter-trace distance of 0.01 m (Δx) has been used for data recording with both the used antennas. The data have been first processed with a basic flow. The velocity estimation has been defined through an hyperbola diffraction analysis (“hyperbola fitting”): a resulting average velocity value of 0,095 m/ns has been estimated for the investigated subsoil, and then used for the final time to depth conversion. A static (“topographic”) correction and a f-k migration algorithm has been applied to the data employing the same velocity, in order to restore the true dips to the reflectors and collapse the hyperbolic diffractions. The fault zone has been better focused in a narrow line at about 22 m, providing reliable information on its geometric characteristics. Fig. 3a illustrates the processed un-migrated profile, more suitable to highlight the fault zone even using the diffractions. Then, a 3D GPR volume has been acquired employing a 300 MHz antenna, employing a distance of 0.1 m between profiles and an inter-trace distance of 0.01 m. The size of the acquisition grid was 5,5 x 20 m, respectively in NW-SE and NE-SW directions, consisting in 56 2D GPR profiles each of 2000 (Fig. 3). The flow used for the data processing was analogous to the one already described for the 2D lines. Tab. 1 – Table summarizing the main acquisition parameters used for the 2D/3D GPR survey done on the Castrovillary area. Acquisition parameter Time window (ns) Estimated RelativeDielectric Permittivity ( r ) Total number of samples for trace (n°) Inter-trace distance (m) Profile distance (3D data) (m) Antenna Central Frequency 300 MHz 200/300 14 (Pollino) 512/1024 0.05/0.01 0.10 (Castrovillari) 500 MHz 100/200 10 (Castrovillari) 512/1024 0.01 / Integrated interpretation of the 2D/3D data. The available trench data in Fig. 2b across the Mt. Pollino fault (Grotta Carbone site) have been extended in length as well as in depth by the GPR profiles, imaging the geological structures and the fault zone with high-resolution. The original stratigraphic information about the units described by Michetti et al. (1997) as alluvial fan deposits (Pleistocene) were highlighted in depth by a discontinuity dividing gentle inclined bedding on underlying dipper layers (Fig. 2a, 100 ns): these can be interpreted as an older unit belonging to the alluvial fan deposits, like a seismic “bedrock”, compatible with the high probing depth investigated. The attribute analysis helps the visual interpretation of the tectonic discontinuities and the sedimentary units, which can be accurately deduced and easily followed. The fault offset can be estimated from some steps separating the colluvial materials on the fault hanging wall, that looks like “transparent” to the radar energy due to the strong reflection of the continuous basal reflector, comparable with the one of the foot wall units (Fig. 2a,b). Some geophysical signatures already observed in literature by some authors (Liner and Liner, 1997; Bano et al. , 2002; Pauselli et al. , 2010; Ercoli et al. , 2013) were identified, like relative differences in signal amplitude, attenuation of the radar units, interruptions of the lateral continuity and dip of the reflectors, diffraction hyperbolas (in un-migrated data) and 172 GNGTS 2013 S essione 1.2