Ullah et al. (2025) Drivers and Future Risks of Groundwater Projection in Tangshan, China: Integrating SHAP, Geographically Weighted Regression, and Climate–Land-Use Scenarios

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

Identification

• Journal: Hydrology
• Year: 2025
• Date: 2025-11-30
• Authors: Zahid Ullah, Yicheng Wang, Hejia Wang, Jia Liu
• DOI: 10.3390/hydrology12120317

Research Groups

[Information not provided in the paper text.]

Short Summary

This study developed an integrated framework combining machine learning and scenario-based forecasting to evaluate spatial drivers and patterns of groundwater stress in Tangshan city and project future risks under climate and land-use change. It found that evapotranspiration and population density are key drivers of depletion, with future projections under RCP 8.5 showing highly unstable recharge and intensified depletion risks compared to RCP 4.5.

Objective

• To evaluate the spatial drivers and patterns of groundwater stress and project potential future risks in semi-arid regions like Tangshan city using an integrated framework.

Study Configuration

• Spatial Scale: Tangshan city, a semi-arid urbanizing region.
• Temporal Scale: Current analysis based on 2022 data; future projections from 2023 to 2049.

Methodology and Data

• Models used: XGBoost (Extreme Gradient Boosting) regression model, SHAP (Shapley Additive exPlanations), GWR (Geographically Weighted Regression), LISA (Local Indicators of Spatial Association), scenario-based recharge modeling.
• Data sources: Spatial groundwater table data (2022), environmental and anthropogenic variables (evapotranspiration, population, temperature, precipitation, land use and land cover changes), Coupled Model Intercomparison Project Phase 6 (CMIP6) climate data, land-use data.

Main Results

• SHAP analysis identified evapotranspiration (ET) and population density as prominent contributors to groundwater depletion in agricultural and urban zones.
• GWR was used to estimate localized coefficients and construct a Vulnerability and Resilience Index (VRI).
• LISA validated vulnerability zones and revealed transitional stress regions.
• Future projections (2023-2049) under Representative Concentration Pathway (RCP) 8.5 demonstrate highly unstable recharge with frequent negative episodes (evapotranspiration > precipitation).
• RCP 4.5 projections show relatively stable patterns of groundwater table.
• Coupled with urban and agricultural expansion, RCP 8.5 intensifies groundwater depletion risks.

Contributions

• Provides an integrated framework combining understandable machine learning (SHAP, GWR, LISA) with scenario-based recharge forecasting for spatial driver analysis and risk assessment.
• Offers analytical understandings of spatial driver patterns and scenario-based risk assessments under climate and land-use change.
• Identifies priority zones for intervention and underlines the importance of adaptive, scenario-sensitive groundwater governance in semi-arid, urbanizing regions.

Funding

[Information not provided in the paper text.]

Citation

@article{Ullah2025Drivers,
  author = {Ullah, Zahid and Wang, Yicheng and Wang, Hejia and Liu, Jia},
  title = {Drivers and Future Risks of Groundwater Projection in Tangshan, China: Integrating SHAP, Geographically Weighted Regression, and Climate–Land-Use Scenarios},
  journal = {Hydrology},
  year = {2025},
  doi = {10.3390/hydrology12120317},
  url = {https://doi.org/10.3390/hydrology12120317}
}