High-resolution (0.1°) Groundwater Storage Anomalies for France (2003-2023) Derived from the Downscaling of the GFZ G3P Product

Downscaled High-Resolution (0.1°) G3P-derived Groundwater Storage Anomaly (GWSA) for France (2003-2023) This repository contains the associated datasets generated as a result of the study “Improving the Spatial Resolution of GRACE-derived GFZ G3P Groundwater Storage Anomaly Product Through Unsupervised Deep Learning Downscaling”, by C. A. Farías and G. Goracci. In this work, an unsupervised U-Net-based convolutional neural network framework was applied to downscale groundwater storage anomalies (GWSA) from the German Research Centre for Geosciences in Postdam (GFZ) G3P product from its original 0.5° spatial resolution to 0.1°. The processing has been tested over the Paris basin area, covering the period from January 2003 to September 2023. Groundwater monitoring wells from the IGRAC Global Groundwater Monitoring Network (GGMN) were employed for the validation of the product. Dataset content The folder in this repository contains: The High-Resolution (HR) GWSA products resulting from the application of the downscaling framework (HR_GWSA_0.1deg_DE_2003-2023.nc) The downsampled Low-Resolution (LR) version of the HR GWSA (Downsampled_HR_GWSA_0.5deg_DE_2003-2023.nc) The original GFZ G3P GWSA product, clipped to the boundaries of the test area (France) and study time period (2003-2023) (Original_G3P_GWSA_France_2003-2023.nc) A README text file, with information about the datasets A LICENCE text file, containing information on licenses and attributions