Identifying excavation damaged zones using2D electrical resistivity tomography modeling

Underground excavations provoke in their vicinity a region where the rock is disturbed, i.e., loosened due to micro as well as macro fractures. The shapes, dimensions, and properties of such so-called excavation damaged zones (EDZ) are of increasing importance for the planning and construction of geotechnical barriers in underground repositories for toxic and problematic wastes.
Dynamic stability assessment of rocks after underground excavation is important. Mechanical changes related to an excavation damage zone (EDZs) leads to changes the physical properties of rocks. In fractured and unsaturated materials, resistivity is sensitive not only to the matrix electrical properties but also to the saturation of the water phase and to the density and orientation of cracks. Recent studies show that electrical resistivity tomography (ERT) and induced polarization (IP) methods are capable of monitoring the mechanical behaviour of EDZs.
ERT and IP methods are performed in the galleries which are excavated in clay-rocks of the Tournemire test site located in the south of France. The aim of these geophysical investigations is the characterization of EDZ zones and hydro-mechanical behaviour. ERT is performed for an arc profile on the walls of the gallery, with 43 electrodes arranged on the floor and wall with a distance of 20 cm between them. Non-polarizing electrodes of Cu/CuSO4 were used. Interpretation of the ERT section on the straight line profile, carried out on the floor of the tunnel, confirmed the existence of a high electrical resistivity zone near the surface (a fractured and partially saturated zone with a depth of 50 cm).
A 2D electrical resistivity model was developed to perform a tomography survey for the arc profiles on the walls of the underground excavations (horseshoe section). All the Wenner, dipole-dipole, and Schlumberger electrode arrays can be used for ERT surveying.
Current and potential electrodes can be arranged on the floor, walls and ceiling of the tunnel, with equal intervals. A finite element method was performed in order to solve the Poisson equation for all points of the model space with respect to boundary conditions. The finite element approach involves solving a discretized form of the weak formulation of the Poison equation. For each quadripole (two current and two measured electrodes), the code is run once and the electrical resistance can be calculated. The geometric factor of electrical array can be calculated when the code is run for a homogeneous electrical conductivity earth around the tunnel.
The space model surrounding the gallery (up to 10 times the distance to the tunnel diameter) is divided into quadrilaterals whose conductivity can be changed. The Neumann boundary condition is considered for the inner and outer surface of the tunnel wall. The outer surface is far from the walls of the gallery.
The code (Forward model) is programmed in COMSOL Script software using Matlab language. An input file was used that determines the location of each quadripole for each electrode array according data acquisition. Eventually, the apparent resistivity pseudo-section is calculated from the code.
An interpreted cross-section, obtained from manual inverse modelling using the code, shows good conformity with the results of the tomography obtained from the straight profile taken on the floor of the tunnel. The results also show that ERT is capable of investigating the resistivity changes near the surface of tunnel walls. The depth of the investigation is up to the tunnel radius.