GNGTS 2015 - Atti del 34° Convegno Nazionale

GNGTS 2015 S essione 3.2 113 to provide some insight with regard to the entity and extension of the damages within a part of the old town. The results of these investigation were particularly important because the municipality had to establish whether some tens of people could come back to their houses, or rather they should look for alternative solutions. In particular, the municipality had provided provisional accommodations to several people, which was of course a meaningful cost for them. The situation was made somehow even more complicated by the fact that, indeed, several artificial cavities are present under the houses of the historical center of Mesagne, due to a number of cisterns (whose dating is quite uncertain and variable from case to case, but the most ancient ones are thought to be many centuries old) reused in several ways during subsequent periods (included the exploitation as waste containers) and then definitely abandoned in the 20 th century and in most cases currently corked. Indeed, in many cases also the memory of the presence of these structures is lost, because the houses changed ownership several times. In particular, the old town of Mesagne was inhabited by the Messapic people since the 6 th century B.C.. Then, the town passed through the Roman, Greek-Byzantine, Norman, Svevian, French and Spanish dominations that left traces of great historical and artistic value on which, however, the inevitable effects of anthropogenic modifications are evident. In particular, it is probable that important remains are still hidden under the soil, because still in 1999 Messapic tombs dating back to the 5 th century B.C. were found. The old town was also affected by significant morphological changes because of occurred leveling, excavations and subsequent debris flows, as it is put into evidence by several documents in the archives of the technical office of the Municipality of the town. The geology of the historic centre and more in general of the town of Mesagne is characterized in the first meters by the presence of clayey weakly cemented sands. The water table is located at an average depth of about 4 m from the current planking level. In particular, the geological substrate, formed by sands, is affected by some degradation processes in its superficial parts whose most significant effect is the gradual removal of the cemented part of the subsoil, due to the rainfall and to the meaningful amount of dropped water from the supply networks and from the drainage systems. Indeed, these degradation processes caused in the old town centre of Mesagne several documented subsidence events also in the past. In order to probe the current event, an integrated ERT (Loke and Barker, 1996), SP (Fedi and Abbas, 2013) and GPR prospecting was performed, showing one more time the capabilities of the noninvasive techniques in identifying, within the first meters of depth, a number of anomalies, partially ascribable to collapses and of cavities. The paper is organised as follows: In the next section the ERT and SP are presented. In Section 3, the GPR results are presented. Conclusions follow in section 4. ERT and SP results. ERT and SP results have been achieved making use of the same cable displacements. The instrumentation has been supplied to the Laboratory of Applied Geophysics to the Archaeological and Monumental Heritage (IBAM) of Lecce. In particular, the ERT measurements was a SYSCAL kid-switch suitably modified in order to work both as a georesistivimeter and as an equipment for the measure of the spontaneous potentials. An axial dipole (dipole-dipole) configuration was exploited, that allowed (thanks to a suitable procedure set on purpose) to achieve both vertical (TGV) and horizontal (TGO) tomography images, that allowed to achieve a quite clear picture of the subsurface resistivity distribution. The maximum length of the profiles was chosen based on the required resolution and on the maximum depth of interest, namely the first few meters from the planking level. We made use of 24 electrodes with an electrode distance of 2 m. The inversion of the data has been carried out through an iterative least square minimization process that decreases progressively the difference between the measured apparent resistivity and the model apparent resistivity calculated according to the updated current model of the subsoil. Both the distributions of the electrical resistivity and of the spontaneous potential parameters in the subsurface have been determined in this way. Let us stress, in particular, that a non-standard acquisition geometry