We used global sensitivity analysis to study how magnetic resonance sounding signals respond to water flow parameters during a snowmelt event in a mountain catchment. This geophysical sounding technique is directly sensitive to subsurface water content. We show that the signals have a strong, dynamic sensitivity to only some parameters, while others remain uninfluential. This deeper understanding clarifies the potential and limitations of using such data to calibrate hydrological models.

Land Surface Model outputs (evaporation, transpiration, groundwater recharge) are influenced by uncertain parameters. Global sensitivity metrics provide a ranking of the importance of uncertain factors. Evaporation is directly influenced by the net radiation and by the parameters associated with the top litter layer. Transpiration appears as mainly influenced by the vegetation characteristics and by albedo. Groundwater recharge is influenced mainly by soil-related parameters.

Vertical maps of seismic velocity reflect variations of subsurface porosity. We use such images to design the geometry of subsurface compartments delimited by velocity thresholds. The obtained patterns are inserted into a hydrogeological model to test the influence of random geometries, velocity thresholds, and hydraulic parameters on data estimated from the model: the depth of the groundwater and magnetic resonance sounding is a geophysical method sensitive to subsurface water content.

The reproduction of flows and contaminations underground requires a good estimation of the parameters of the geological environment (mainly permeability and porosity), in three dimensions. While most researchers rely on geophysical methods, which are costly and difficult to implement in the field, this study proposes an alternative using data that are already widely available: piezometric records (monitoring of the water table) and the lithological description of the piezometric wells.