Drivers and distribution of soil arsenic in China’s yellow river irrigation area by machine learning
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Summary Arsenic contamination in irrigated agricultural soils poses global health and ecosystem risks. Using China’s Yellow River Irrigation District as a case study, we developed an interpretable machine learning framework to predict soil arsenic distribution and identify its driving mechanisms. Among five models (XGBoost, RF, SVM, MLP, and MLR), XGBoost achieved the highest accuracy (R= 0.83, RMSE = 0.55). SHAP analysis revealed that cation exchange capacity, population density, and soil pH are the dominant factors controlling arsenic accumulation. Spatial autocorrelation further identified arsenic enrichment hotspots in the central and northern study regions. By integrating XGBoost with ordinary kriging, we produced a high-resolution (1 km × 1 km) arsenic map that overcomes the “bull’s-eye effect” of traditional interpolation. This framework offers a transparent, predictive tool for targeted pollution management in data-limited irrigated regions. abs0010 Graphical abstract http://www.w3.org/1999/xlink float portrait ga1.jpg undfig1 anchor portrait graphical abs0015 Highlights • An interpretable ML framework for high-precision prediction of soil As u0010 • SHAP identifies CEC, population density, and pH as the top three drivers u0015 • High-resolution As map eliminates the “bull’s-eye effect” of traditional methods u0020 • Spatial analysis reveals As hotspots co-aggregated with key drivers u0025 simple ulist0010 author-highlights abs0020 Environmental health; Environmental
Abstract
Summary Arsenic contamination in irrigated agricultural soils poses global health and ecosystem risks. Using China’s Yellow River Irrigation District as a case study, we developed an interpretable machine learning framework to predict soil arsenic distribution and identify its driving mechanisms. Among five models (XGBoost, RF, SVM, MLP, and MLR), XGBoost achieved the highest accuracy (R= 0.83, RMSE = 0.55). SHAP analysis revealed that cation exchange capacity, population density, and soil pH are the dominant factors controlling arsenic accumulation. Spatial autocorrelation further identified arsenic enrichment hotspots in the central and northern study regions. By integrating XGBoost with ordinary kriging, we produced a high-resolution (1 km × 1 km) arsenic map that overcomes the “bull’s-eye effect” of traditional interpolation. This framework offers a transparent, predictive tool for targeted pollution management in data-limited irrigated regions. abs0010 Graphical abstract http://www.w3.org/1999/xlink float portrait ga1.jpg undfig1 anchor portrait graphical abs0015 Highlights • An interpretable ML framework for high-precision prediction of soil As u0010 • SHAP identifies CEC, population density, and pH as the top three drivers u0015 • High-resolution As map eliminates the “bull’s-eye effect” of traditional methods u0020 • Spatial analysis reveals As hotspots co-aggregated with key drivers u0025 simple ulist0010 author-highlights abs0020 Environmental health; Environmental assessment; Machine learning teaser abs0025