Résumé
Changes in groundwater levels associated with hydroclimatic variations and anthropogenic water extraction can deform the solid Earth, both elastically and inelastically. Satellite-based geodetic techniques which measure the Earth's surface displacements can thus be used to track changing conditions in aquifer systems. However, accurately extracting groundwater-induced deformation signals still poses a challenge as geodetic techniques like GNSS and InSAR also record noise, systematic errors and other sources of deformation. In this study, we take advantage of the relatively dense in situ groundwater monitoring network of the Sacramento Valley aquifer in California to constrain its deformation and hydromechanical properties. We start by characterizing the main seasonal and multiannual fluctuations in groundwater levels with an Independent Component Analysis (ICA) and exploit the resulting temporal signature to extract the associated deformation field from GNSS and InSAR time series. We then develop a poroelastic model of the aquifer to invert for its elastic storage capacity and estimate the respective contributions of elastic and inelastic processes to long-term subsidence. Our modeling also suggests that depth-dependent elastic properties are necessary to explain the spatial distribution of horizontal poroelastic displacements measured by GNSS. This work has important implications for the sustainable management of heavily-stressed Californian aquifers but also serves as a calibration between in situ and remote sensing techniques, which is essential for the successful deployment of satellite-based groundwater monitoring in areas with sparse field-based instrumentation.