Wang et al. (2025) Depth-Specific Prediction of Coastal Soil Salinization Using Multi-Source Environmental Data and an Optimized GWO–RF–XGBoost Ensemble Model

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

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-12-16
• Authors: Yuanbo Wang, Xiao Yang, Xingjun Lv, Wei He, Ming Shao, Hongmei Liu, Chao Jia
• DOI: 10.3390/rs17244043

Research Groups

Not explicitly mentioned in the provided text.

Short Summary

This study developed an integrated modeling framework combining ensemble machine learning and spatial statistics to investigate depth-specific soil salinity dynamics in the Yellow River Delta, revealing distinct vertical control mechanisms where surface salinity is dominated by vegetation and soil water, while deeper layers are influenced by total dissolved solids, pH, and groundwater depth.

Objective

• To investigate the depth-specific dynamics of soil salinity and resolve the vertical differentiation and spatial heterogeneity of salinity drivers in the Yellow River Delta, a vulnerable coastal agroecosystem.

Study Configuration

• Spatial Scale: Yellow River Delta, with harmonized samples at 30 m resolution.
• Temporal Scale: Not explicitly mentioned in the provided text.

Methodology and Data

• Models used: Hybrid Gray Wolf Optimizer–Random Forest–XGBoost model, SHapley Additive Explanations (SHAP), Spatial autocorrelation analysis (Global Moran’s I).
• Data sources: Multi-source environmental predictors, 220 field samples.

Main Results

• The hybrid Gray Wolf Optimizer–Random Forest–XGBoost model achieved high predictive accuracy for surface salinity (R² = 0.91, RMSE = 0.03 g/kg, MAE = 0.02 g/kg).
• Spatial autocorrelation analysis (Global Moran’s I = 0.25, p < 0.01) revealed pronounced clustering of high-salinity hotspots associated with seawater intrusion pathways and capillary rise.
• Distinct vertical control mechanisms were identified: vegetation indices and soil water content dominate surface salinity, while total dissolved solids (TDS), pH, and groundwater depth increasingly influence middle and deep layers.
• SHAP quantified nonlinear feature contributions and ranked key predictors across layers, offering mechanistic insights.

Contributions

• Presents an integrated modeling framework combining ensemble machine learning and spatial statistics to investigate depth-specific soil salinity dynamics.
• Provides mechanistic insights into vertical control mechanisms of soil salinity, moving beyond conventional correlation.
• Demonstrates how explainable machine learning (SHAP) can bridge the gap between black-box prediction and process understanding in environmental science.
• Highlights the importance of depth-specific monitoring and intervention strategies for soil salinization.
• Offers a generalizable framework adaptable to other coastal agroecosystems with similar hydro-environmental conditions.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Wang2025DepthSpecific,
  author = {Wang, Yuanbo and Yang, Xiao and Lv, Xingjun and He, Wei and Shao, Ming and Liu, Hongmei and Jia, Chao},
  title = {Depth-Specific Prediction of Coastal Soil Salinization Using Multi-Source Environmental Data and an Optimized GWO–RF–XGBoost Ensemble Model},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs17244043},
  url = {https://doi.org/10.3390/rs17244043}
}