Groundwater storage trends in northern Italy as observed by GRACE, well measurements, and vertical land motion

Geodetic observations of the Earth’s gravitational and deformational response to changes in terrestrial water storage (∆TWS) have been essential measurements to identify regions experiencing long-term wetting and drying driven by a combination of climate and anthropogenic forces. The northern Italian Plains, home to a third of the country’s population and contributing more than half of the agricultural output, have experienced a dryer-than-normal two decades. Here, we investigate what impact these dry conditions have on the long-term groundwater storage (GWS) using observations of change in terrestrial water storage (∆TWS) from the Gravity Recovery and Climate Experiment (GRACE) and the second-generation follow-on (GRACE-FO) missions and in-situ groundwater level time series from 820 wells over the period of 2003-2022. We use a wavelet time-frequency analysis to deconstruct each signal into seasonal and long-term components and identify multi-year dry and wet epochs. We find two long periods of declining groundwater storage (2003-2007, 2015-2022), two short periods of groundwater recovery (2008-2009, 2012-2014), and one period of near-zero ∆GWS (2010-2011). We find a net volume loss of 12.0 km³ from 2003-2022. Further, we validate these ∆GWS trends and total volume loss estimates using a combination of in-situ groundwater level variations and vertical land motion observed at nearly 500 Global Navigation Satellite System (GNSS) stations. These stations show poroelastic deformation over aquifers related to groundwater storage changes and elastic loading deformation that is highly correlated with predicted elastic loading displacements from GRACE(-FO) ∆TWS outside of aquifer areas. To calculate groundwater storage from groundwater level, we estimate spatially- and depth-variable aquifer storage coefficients using a combination of lithologic information and co-located well and GNSS observations. By analyzing all three datasets in combination we can evaluate the impacts of multi-year dry- and wet- periods on groundwater resources, providing essential contextual information for future water management.