Mohammed et al. (2025) Deep-Learning-Based Probabilistic Forecasting of Groundwater Storage Dynamics in Sudan Using Multisource Remote Sensing and Geophysical Data

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

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-09-12
• Authors: Musaab A. A. Mohammed, Norbert Péter Szabó, Joseph Omeiza Alao, Péter Szűcs
• DOI: 10.3390/rs17183172

Research Groups

Not explicitly mentioned in the provided text.

Short Summary

This study integrates GRACE satellite data with GLDAS land surface variables to assess and forecast groundwater storage (GWS) dynamics in Sudan, revealing a positive GWS recovery across all regions, particularly strong in the south and southwest.

Objective

• To assess and forecast groundwater storage (GWS) dynamics across eight major regions in Sudan by integrating satellite-derived water storage from GRACE with land surface variables from GLDAS.

Study Configuration

• Spatial Scale: Eight major regions in Sudan.
• Temporal Scale: Time series analysis and forecasting of groundwater storage dynamics.

Methodology and Data

• Models used: Random Forest (for GRACE TWS reconstruction), Bidirectional Long Short-Term Memory (BiLSTM) with bootstrapping (for GWS trend forecasting).
• Data sources: Gravity Recovery and Climate Experiment (GRACE) satellite observations (terrestrial water storage), Global Land Data Assimilation System (GLDAS) land surface variables (surface and root-zone components).

Main Results

• Probabilistic forecasts of GWS anomalies were generated with narrow confidence intervals consistent with historical ranges.
• Forecasted GWS anomalies indicate positive recovery across all eight regions of Sudan.
• Sen’s slope values for GWS recovery range from 0.014 to 0.051 per month.
• The strongest GWS recoveries are evident in the southern and southwestern regions.
• Northern and eastern areas display more modest GWS gains.

Contributions

• Represents one of the first applications of deep learning with uncertainty quantification for GRACE-based groundwater analysis in Sudan.
• Demonstrates the potential of an integrated GRACE-GLDAS-deep learning approach to support informed and sustainable groundwater management in data-limited environments.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Mohammed2025DeepLearningBased,
  author = {Mohammed, Musaab A. A. and Szabó, Norbert Péter and Alao, Joseph Omeiza and Szűcs, Péter},
  title = {Deep-Learning-Based Probabilistic Forecasting of Groundwater Storage Dynamics in Sudan Using Multisource Remote Sensing and Geophysical Data},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs17183172},
  url = {https://doi.org/10.3390/rs17183172}
}