GNGTS 2013 - Atti del 32° Convegno Nazionale

A shallow coastal aquifer flows out along the Casalabate coast. The groundwater circulation is influenced by the tectonic fracture systems and is manifested by some submarine springs. Underwater investigations of the sinking of a portion of submerged beach in 1997 revealed the presence of a karst spring (Delle Rose and Federico, 2002) probably joined with the cave that induced the 1993 sinkholes. Such springs are not perennial and can occur suddenly. The last case of new spring formation occurred below an house on the beginning of September 2013 and caused a new warning of the residents. 3D models. Using the software Groundwater Modeling System (GMS) a 3D geological model and a 3D groundwater model are in progress. The scope of building the models is to simplify the field problem and organize the field data so that the system can be analyzed more readily (Anderson andWoessner, 2002). The conceptualization of the models includes synthesis and framing up of data pertaining to geology, hydrogeology, hydrology, and meteorology. The lithologic data were collected from 23 core lithologs data (kindly provided by the technical office of Lecce municipality). Such a borehole survey was performed on 2008 and was related to the project “actions to protect the soil in the hydrogeological risk areas of the municipality of Lecce” funded by the Ministry of Environment and Protection of Land. The geological model was created using a block-centric finite-difference grid. The gridding was performed through the eight nearest-neighbor methodology with 3-D interpolation by average minimum distance. The solid model was developed to a depth of about 10 m below the ground level. The sensitivity of the model was tested by varying the horizontal and vertical spacing of the nodes, and an optimized final model, least sensitive to changes in spatial resolution (i.e., smaller grid sizes), was built. The resolution of the final model was 100m×100m×2m. The resulting discretization consisted of 250nodes×250nodes×100nodes, obtaining 6,250,000 solid model nodes, each with a voxel volume of 20,000 m 3 . It should be noticed that the results of the modeling are not free from uncertainties, which could be reduced by increasing datapoints, but they illustrate one of the most probable scenarios. The smallest scale of variation that the model is able to depict is equal to the resolution of one voxel. The 3D geological model (Fig. 2a) with the cross-sections in Fig. 2b provides a depictions of the bedrock of the studied area. Such model was vertically discretized in 8 layers (paving; artificial landfills; sands of beach; organic soils; residual soils; marsh deposits; cemented thinly laminated and cross-stratified calcarenites; moderately cemented massive calcarenites). The layers were allowed to have seepage from the top and leakage through the base, making them hydraulically connected. The top-most layer and the second layer were defined as unconfined. The rest of the layers were defined as confined, as the water table was not expected to fall > 10 m. Such a model has reached a first level of advancement and needs several new data to be fully efficient, to start from the boundary conditions as well as the real physical characters of the higher layers. Given the heterogeneity and complexity associated with the study area (e.g. local-scale variations of aquifer properties), some generalizations, simplifications and assumptions will be made to construct the groundwater flow model. The modeling area is limited in the east by the Adriatic sea, and it accounts for about 1,5 km in length and 0,5 km in width. The 3-D constant- density groundwater flow will be simulated by a block-centric, finite-difference grid model, using MODFLOW code inside GSM software. Furthermore MODAEM code, which allows the analysis to the analytical elements (polygons, lines and points), will be used. New geophysical measures. In our previous work, we have proposed a method to integrate geophysical survey and geological knowledge to characterize the sinkhole hazard at Casalabate. As regards geophysics job, the performed 3D GPR and 3D ERT allowed us to reveal a complex hypogean cavity system inside the urban area affected by sinkhole since the 1993 (Delle Rose and Leucci, 2010). Here we propose an ERT survey carried out in an area (Fig. 3a) next to the 1993 sinkhole collapsed that caused the collapsing of three buildings and irreparable damage to various others, which were demolished later. A 48-channel Syscal-Pro Resistivity-meter in a multielectrode configuration using a Wenner–Schlumberger array (Loke 2009) was used. This 108 GNGTS 2013 S essione 3.2