Farías et al. (2026) Improving the Spatial Resolution of GRACE-Derived GFZ G3P Groundwater Storage Anomaly Product Through Unsupervised Deep Learning Downscaling

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-09-11
• Authors: Celina Anael Farías, Gilberto Goracci
• DOI: 10.3390/rs18183119

Research Groups

• Hydrology and Water Resources Group, University of [Institution]
• Climate and Environmental Physics Group, University of [Institution]

Short Summary

This study presents an unsupervised deep learning approach to downscale groundwater storage anomalies from GRACE-derived data with a spatial resolution of 0.1°. The method outperforms the original product when compared with in situ measurements.

Objective

• Investigate the feasibility of using machine learning to improve the spatial resolution of GRACE-derived groundwater storage anomalies.

Study Configuration

• Spatial Scale: Regional scale (France) and local scale (Paris Basin)
• Temporal Scale: 2003–2023

Methodology and Data

• Models used: Unsupervised deep learning approach with a neural network architecture
• Data sources:
  • GRACE/GRACE–FO missions (GFZ G3P product) for groundwater storage anomalies
  • ERA5 climatic variables (13)
  • Digital elevation model

Main Results

• The downscaled product shows good agreement with the original G3P product (average temporal and spatial correlations of r = 0.99).
• At the basin level, the downscaled product outperforms the original G3P data when compared with in situ measurements (r = 0.67 vs. r = 0.62).
• The accuracy is maintained at the local well scale (r = 0.39 vs. r = 0.38).

Contributions

This study contributes to the development of machine learning techniques for improving the spatial resolution of GRACE-derived groundwater storage anomalies, which can help monitor this critical resource more effectively.

Funding

• This research was funded by the [Project Name] (Grant Code: [Reference Code]) and the [Program Name] (Grant Code: [Reference Code]).

Citation

@article{Farías2026Improving,
  author = {Farías, Celina Anael and Goracci, Gilberto},
  title = {Improving the Spatial Resolution of GRACE-Derived GFZ G3P Groundwater Storage Anomaly Product Through Unsupervised Deep Learning Downscaling},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18183119},
  url = {https://doi.org/10.3390/rs18183119}
}