Assessment of three mixed arrays dataset for subsurface cavities detection using resistivity tomography as inferred from numerical modelling

Abstract

Abstract The present study deals with the evaluation of a three-mixed array dataset for the detection of subsurface cavities using conceptual air-filled cavity model sets at different depths. Cavity models were simulated using the forward modelling technique to generate synthetic apparent resistivity data for three common individual arrays. These arrays are dipole–dipole (DD), pole–dipole (PD), and Wenner–Schlumberger (WS). The synthetically apparent resistivity data obtained from two different individual arrays were merged to form a high-resolution single model. Consequently, three possible mixed arrays datasets can be obtained: the dipole–dipole-Wenner–Schlumberger (DD+WS), pole–dipole, and Wenner–Schlumberger (PD+WS), and dipole–dipole and pole–dipole (DD+PD). The synthetically apparent resistivity data for both the individual and mixed arrays were inverted using Res2dinv software based on the robust constrain inversion technique to obtain a 2D resistivity model section. The inverted resistivity sections were evaluated in terms of their recovering ability of the model’s parameters (e.g. resistivity, and geometry). The results show that the individual arrays can resolve the location and dimensions of the cavity within reasonable accuracy only at a depth not exceeding 6 m below the surface. On the other hand, a significant resolution enhancement in model resistivity with increasing depth was observed when the mixed arrays were used. The (DD+WS) mixed arrays dataset brings up better model resistivity and shows closer parameters to the true actual model among the other mixed arrays. So it is strongly recommended for cavity detection studies.