Many important geological questions are related to the understanding of fluid flow through the pores of geological materials (rocks, sediments and soils) on the field scale. Often, these questions involve the flow of two (or more) fluids which do not mix, in which case their behavior is driven by interfaces between the fluids at the microscopic scale of the pores. The current lack of understanding of the dynamics of these interfaces inside the pore system impedes the derivation of accurate models at the field scale, significantly impacting e.g. the access to clean drinking water, the protection and remediation of drinking water towards pollution, enhanced geothermal energy generation, production and storage of hydrocarbons, and various other processes related to the industrial use of porous materials. These limitations plead for a rational mathematical approach, based on upscaling of models developed at the pore scale , which incorporate various characteristics that can be observed and measured there. The objective of this research proposal is thus to develop reliable, experimentally-driven macroscale mathematical models of two-phase flow through geological media, by combining pore scale visualization experiments with mathematical modeling and upscaling techniques. The key challenges in this proposal will be the characterization, modeling and upscaling of the behavior of the fluid interfaces (notably wettability) throughout geological materials.