North China Plain Groundwater Downscaling from Gravity Satellites (~5 km)

Groundwater in the North China Plain (NCP) has been severely depleted by intensive irrigation and urban extraction, raising the need for high-resolution monitoring and analysis of groundwater trends and recovery potential. Traditional GRACE-based assessments offer valuable basin-scale insights but remain limited by their coarse spatial resolution and inability to resolve vertical aquifer responses or distinguish complex driving mechanisms. This study develops an integrated framework that combines blind super-resolution of GRACE terrestrial water storage, vertical separation of shallow and deep aquifer signals using well observations. The blind downscaling model increases GRACE resolution from 1° to 0.0625° (~5 km) and improves groundwater level retrieval accuracy to 89.5%. This improvement allows detailed quantification of a long-term decline (~1.74 m/yr, 2002–2021) followed by recent recovery (~1.12 m/yr, 2021–2023), alongside dynamic shrinkage and fragmentation of groundwater drawdown funnels. These findings provide new insights into the spatiotemporal evolution of groundwater in overexploited basins and offer a technical foundation for targeted, adaptive groundwater management strategies under intensified climate and anthropogenic pressures.