Résumé : [en] Petroleum hydrocarbons (HC) represent the most widespread contaminants in the world and in-situ bioremediation remains a competitive treatment in terms of cost and environmental concerns. However, the efficiency of such a technique (by biostimulation or bioaugmentation) strongly depends on numerous environmental characteristics (heterogeneities of the subsurface structure, soil moisture, oxygen and pollutants bioavailability, microbial nichesdots) and is still difficult to predict a priori. In order to lower these uncertainties, Electrical Resistivity Tomography (ERT) appears as a valuable non-invasive tool to detect soil heterogeneities and to monitor biodegradation. The main objective of this study was thus to isolate with ERT an electrical signature corresponding to an enhanced biodegrading activity, in an aged HC-contaminated clay loam soil. To achieve it, a pilot tank with metric dimensions (3.6 times 0.9 times 0.6 m) and a recirculating system (which is quite unique for this type of purpose) was built to mimic field conditions and to control the evolution of the bio-physico-chemical parameters (microbial concentration in soil and groundwater, temperature, pH, pO2, redox potential, bulk and fluid conductivities, water flow, hydrocarbon content) through time and space. Five panels of electrodes were placed at different locations in the tank to detect lithological heterogeneities and to monitor the bulk resistivity variations with time-lapse ERT.Compared to a first insufficient biostimulation phase with H2O2 and KNO3, bioaugmentation with Rhodococcus erythropolis T902.1 led to a HC depletion of almost 80% (6900 to 1600 ppm) in 3 months in the center of the contaminated clay, where pollutants were less bioavailable. Furthermore, lithological heterogeneities (clay, sand, gravels) and microbial activities (growth, degradation and biosurfactant production) were successfully discriminated by ERT images obtained during both remediation phases. In the future, this cost-effective technique should be transferred to the field in order to either (i) detect and forecast biodegradation processes before choosing an appropriate remediation technique, or (ii) monitor the efficiency of this biodegradation during an in-situ bioremediation.