Foam injection is a promising option for soil remediation applications. However, predicting how it will propagate in highly permeable aquifers under groundwater flow is challenging. Here, we have studied pressure and saturation variations during foam propagation. A 2D tank packed with 1 mm glass beads was used to study foam injection in highly permeable porous media under lateral flow. Specifically, we evaluated the efficiency of pressure and time-domain reflectometer (TDR) sensors to predict foam propagation using an imaging technique. A numerical model coupling two-phase flow and surfactant transport was developed to simulate the experimental results. This model takes into account the effect of non-Newtonian behavior of foam, surfactant concentration, and critical capillary pressure through the definition of the mobility reduction factor (MRF). The experimental results show that the foam injection pressure first increases with a logarithmic law and then stabilizes. This pressure stabilization can be related to the state of pseudo-equilibrium between foam generation and destruction. We observed an asymmetrical foam propagation due to water lateral flow. Comparisons of the liquid saturation fields calculated by analysis of TDR probes and estimated by imaging show that the TDR sensors monitor foam propagation well in saturated porous media. They can predict the shape of the injected foam. Contrary to pressure sensors, it is possible to capture weak foam behavior using TDR sensors. Finally, the numerical model we have developed correctly captures the shape of foam propagation and its ability to divert water flows. This model produces the propagation of the strong foam well and predicts the saturation and pressure fields with good precision.< Réduire