GNGTS 2014 - Atti del 33° Convegno Nazionale

The contamination phenomenon was simulated using a PVC pipe characterized by a hole in the middle of the diameter of 1 cm. Ten litres of saline solution (with a salt concentration of 100 g/l) were injected from the hole in ten days and a well, located on the opposite side, has generated a steady stream into the vat pumping 35 ml/minute of water for the entire experiment. The pumping well was inserted in a drainage system like that used to recharge the aquifer. The piezometric head was kept constant at a depth of about 30 cm from the surface with a hydraulic circuit consisting of two tanks connected to the column. Furthermore an integrated monitoring system was used to identify the contaminant flow consisting of geo-electrical cables (for ERT cross-borehole), piezometer (for groundwater measurements) and multi-parameter probes (for evaluation of water properties). To perform ERT cross-borehole in the sand box, four piezometers instrumented with 48 steel plate electrodes and two cables with 24 steel electrodes are made in laboratory. With piezometer systems it was possible monitor the aquifer and analyse the water properties. All the cables were connected to the electrical resistivity meter Syscal Pro that provides electrical measurements with 96 electrodes and has ten channels to carry out up to 10 readings at the same time. The Syscal Pro, connected to a PC, allowed to acquire electrical resistivity data and to control them in real time. The 72 steel electrodes were used in order to perform the 2D cross-borehole resistivity measurements within 0.05÷0.65 m in depth. The cables were placed vertically in boreholes at a reciprocal distance of 0.35-0.37cm. A cross-borehole azimuthal dipole–dipole array was adopted using a current and potential dipole separation (D) of 0.15 m up to 0.30 m according seven different sequences. In order to define the electrical behaviour in uncontaminated and contaminated conditions seven ERT cross-borehole according to the sequences -AB-CD-EF-AD-BC-CF-ED- were performed and inverted each day where A-B-C- D-E-F are the borehole (see Fig.1) . For each one the reciprocal error was evaluated. Collection of measurements in the reciprocal configuration, which took approximately 180 minutes, permitted assessment of data errors (Binley et al. , 2002a). To investigate the effects caused by salt contamination several GPR acquisition were performed with SIR-3000 radar coupled to 400, 900 e 2000 MHz antennas also using a survey wheel. In this way the analysis are characterized by different resolution and depth. Three different acquisitions were made, two in the transverse direction with respect to the flow and one in longitudinal direction. To facilitate the estimation of the physical parameters of interest were placed anomalous bodies in the sand (steel rod, pvc pipe and steel plates already described). The experiment was monitored for thirty-one days with two different multi-parameter probes, the Multi 340i WHW handheld meter and the HI 9828 multi-parameter system (Hannah Instruments). The first probe, thanks to the small size, was used to measurements into the piezometers; the second one was located in a tank with overflow placed at the water outlet. The Multi 340i WHW was used to carry out pH measurements and conductivity measurements in the instrumented piezometer during the experiment usually at time intervals of four hours. The HI 9828 probe allowed to analyse and record on internal memory measurement of dissolved oxygen, pH, ORP, conductivity and related parameters, temperature, atmospheric pressure, related on the outlet water. Therefore, two different scenarios were analyzed: a. Sand-box uncontaminated saturated for 60 cm from the bottom with tap water (Sc-1); b. Sand-box contaminated with 10 litres of salt water (the concentration of NaCl was 100 g/l) and groundwater at 60 cm from bottom (Sc-2). In uncontaminated conditions (Sc-1) the resistivitymaps are performed in order to investigate the electrical starting behaviour of the sand and provide the data necessary to the assessment of the formation factor and the TDS concentrations. The test started on June 24 2014 at 08:00 in uncontaminated scenario with the water table fixed to a depth of about 32 cm from the ground level. The tracer test injection started on June 24 2014 at 11:00 am (Sc-2). Ten litres of salt water solution, with a concentration of 100 g/l 218 GNGTS 2014 S essione 3.3