Xiong et al. (2025) Evaluating Terrestrial Water Storage, Fluxes, and Drivers in the Pearl River Basin from Downscaled GRACE/GFO and Hydrometeorological Data

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

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-11-25
• Authors: Y. Xiong, Jincheng Liang, Wei Feng
• DOI: 10.3390/rs17233816

Research Groups

Information not available from the provided text.

Short Summary

This study develops and validates a downscaled terrestrial water storage anomaly (TWSA) product for the Pearl River Basin (PRB) by integrating GRACE/GRACE Follow-On observations with the WaterGap Global Hydrological Model (WGHM) via joint inversion, demonstrating enhanced spatiotemporal fidelity and providing actionable information for water management.

Objective

• To enhance the spatiotemporal resolution of GRACE/GRACE Follow-On terrestrial water storage anomaly (TWSA) estimates in the Pearl River Basin (PRB) using a joint inversion downscaling approach.
• To validate the downscaled product against GRACE/GRACE Follow-On and WGHM, assessing its performance at basin and pixel scales.
• To characterize the seasonality and long-term variability of TWSA in the PRB and identify its primary drivers, including the influence of large-scale climate modes.
• To demonstrate the utility of the enhanced TWSA product for drought and flood diagnostics and water resources management.

Study Configuration

• Spatial Scale: Pearl River Basin (PRB), basin-scale, regional scales, pixel scales.
• Temporal Scale: 2002 to 2022 (21 years), with analysis of annual, multi-year, and seasonal (e.g., April to July) variations.

Methodology and Data

• Models used:
  • WaterGap Global Hydrological Model (WGHM) for high-resolution spatial patterns.
  • Joint inversion technique to integrate GRACE/GFO with WGHM.
  • Extreme gradient boosting (XGBoost) for driver attribution.
  • Shapley additive explanations (SHAP) for interpreting XGBoost model outputs.
• Data sources:
  • GRACE and GRACE Follow-On (GFO) satellite observations for terrestrial water storage anomalies (TWSA).
  • Precipitation (P), evapotranspiration (E), and runoff (R) for water balance calculations and driver analysis.
  • Vegetation and radiation variables for driver analysis.
  • Large-scale climate modes (El Niño, Pacific Decadal Oscillation - PDO) for modulating TWSA relationships.
  • Drought severity index (DSI) and standardized flood potential index (FPI) for diagnostics.

Main Results

• The downscaled TWSA product significantly outperforms WGHM at both basin and pixel scales in the PRB, exhibiting consistently lower errors and higher skill when validated against GRACE/GFO.
• Terrestrial water flux (TWF) estimates derived from the downscaled product show improved agreement in both magnitude and phase with water balance calculations.
• TWSA in the PRB displays strong seasonality, with precipitation exceeding evapotranspiration and runoff from April to July, leading to a storage peak in July.
• From 2002 to 2022, the PRB experienced alternating multi-year periods of terrestrial water storage decline and recovery.
• On an annual scale, TWSA covaries with precipitation and runoff, with these relationships modulated by large-scale climate modes: El Niño and a warm Pacific Decadal Oscillation (PDO) favor wetter conditions, while La Niña and a cold PDO favor drier conditions.
• XGBoost with SHAP attribution identifies precipitation as the primary driver of TWSA variability, followed by runoff and evapotranspiration, with vegetation and radiation variables playing secondary roles.
• Drought and flood diagnostics using the drought severity index (DSI) and a standardized flood potential index (FPI) successfully captured the severe 2021 drought and major wet-season floods in the PRB.

Contributions

• Develops and validates a novel downscaled terrestrial water storage anomaly (TWSA) product for the Pearl River Basin, significantly enhancing the spatiotemporal fidelity of satellite-informed storage estimates.
• Provides a robust methodology for integrating coarse-resolution satellite gravimetry with high-resolution hydrological models, applicable to other humid subtropical systems.
• Quantifies the primary drivers of TWSA variability in the PRB, including the influence of large-scale climate modes, using advanced machine learning techniques (XGBoost with SHAP).
• Demonstrates the practical utility of the enhanced TWSA product for improved risk assessment and water resources management, particularly for drought and flood monitoring.

Funding

Information not available from the provided text.

Citation

@article{Xiong2025Evaluating,
  author = {Xiong, Y. and Liang, Jincheng and Feng, Wei},
  title = {Evaluating Terrestrial Water Storage, Fluxes, and Drivers in the Pearl River Basin from Downscaled GRACE/GFO and Hydrometeorological Data},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs17233816},
  url = {https://doi.org/10.3390/rs17233816}
}