GNGTS 2015 - Atti del 34° Convegno Nazionale

GNGTS 2015 S essione 3.2 115 2) areas indicated with “B”, with resistivity values between 50 and 100 Ω m; these values indicate the probable presence of areas where a phenomenon of instability is present too. The low resistivity values indicate that these anomalies are due to the potential presence of water saturated incoherent materials; 3) areas with resistivity values between 150 and 280 Ωm; these values indicate the probable presence of subsoil zones free of in-homogeneities. From the right hand panel of Fig. 2, instead, we can notice a quite uneven distribution pattern of the spontaneous potentials. In particular, it is possible to note two points, indicated with “R”, where there is a negative concentration of the spontaneous potential (-10 mV). It is quite probable that, in these points, a flow of materials in the directions indicated by the arrows (that is, towards positive spontaneous potentials values) was occurring at the moment of the measure (March 2014). GPR measurements. A large GPR survey was performed too. In order to carry out it, we made use of a pulsed RIS Hi Mod GPR system, manufactured by the Ingegneria dei Sistemi IDS-Corporation and belonging to the Laboratory of applied geophysics to archaeological and monumental heritage of IBAM-CNR. The system is equipped with a dual antenna with central frequencies at 200 and 600 MHz, plus an antenna at 900 MHz and plus an antenna at 2 GHz. For the case history at hand, in dependence of the several met situations, we have exploited the dual antenna or the antenna at 900 MHz. In particular, the main targets looked for were subservices and cavities, because possibly related to the present or to future subsidence events. So, the targets looked for were expected to have a size between 0.5 and 2 m, and their depth was expected variable from tens centimeters to a few meters. We set the time bottom scale to 60 ns for the 900 MHz antenna, 80 ns for the 600 MHz and 160 for the 200 MHz antenna. In the first phase, the GPR surveys were performed in the areas outside the dwellings, in the second phase, instead, inside a number of dwelling, especially those where the most meaningful cracks along the walls were present and those were some the floor had collapsed in some points. Comprehensively, we gathered 174 Bscans, covering an overall surface of about 964 m 2 . Afterwards, we performed a standard processing sequentially based on zero timing, background removal (Persico and Soldovieri, 2008), declipping, gain variable vs. the depth, one dimensional filtering when needed and finally migration in time domain (Persico, 2014). The processing of the collected data sets was performed making use of the commercial code GPRSlice (Goodman and Piro, 2013), which also was exploited in order to achieve depth slices when possible. A quite good penetration of the signal, that reached (on average) reliably the time level of about 60ns (corresponding on average to a depth of about 2.1m) was recorded. An average propagation velocity of the electromagnetic waves equal to about 0.07m/ns was evaluated thanks to the presence of the diffraction hyperbolas presumably ascribable to electrically small targets (Mertens et al., 2015; Persico et al., 2015a, 2015b; Soldovieri et al. , 2008, 2009). The profiles obtained with the antenna at 200 MHz showed the same anomalies recorded in the profiles obtained with the antenna at 600 MHz, but did not put into evidence significant deeper targets. Therefore, we will limit to consider the data gathered with the antennas at central frequency 600 MHz because of the obvious better available resolution with respect to the lower frequency antennas. Depending on the zones and on the exploited antennas, the interline spacing of the Bscan profiles was equal to 50 cm or less (we used a transect of 20 cm for the antennas at 900 MHz). This allowed us to join together the results and work out time depth slices. The time slices were built using a time intervals of 5 ns, about corresponding to 0.17 m. The slides reported in Fig. 3 are not picked up at a unique depth level but rather they are a combination of several slides at different estimated depths level into the different probed areas. This