Towards the Use of Passive Seismic for Hydrogeological Characterization
Résumé
Hazards related to groundwater (GW), such as floods, landslides, and sinkholes, present significant risks to infrastructure integrity and public safety. A deeper understanding of groundwater table (GWT) geometry and dynamics, along with GW infiltration rates, closely linked to soil permeability, is crucial for anticipating and managing this risks effectively. By analyzing dispersion data acquired from train-induced seismic noise, we estimate daily GWT level maps. This is achieved through the utilization of a neural network, which processes the dispersion data to infer GWT levels across the study area. This approach provides a comprehensive and spatially explicit representation of GWT dynamics. Furthermore, we develop a method for estimating GW infiltration rates, which can be regarded as an apparent permeability, across different depths. This method involves analyzing the temporal evolution of VR (Rayleigh-wave phase velocity) over various frequencies. The observed phase-shift in VR over time between frequencies are indicative of changes in GW infiltration rates, with higher frequencies being particularly sensitive to this fluctuations. We estimate infiltration rates at different depths, providing valuable insights into subsurface hydrogeological processes. This two interdisciplinary approaches bridge the gap between geophysics and hydrogeology, enabling a comprehensive characterization of both saturated and unsaturated zones.