{"title":"基于Optuna超参数优化的SWAT与SWAT耦合XGBoost模型在泰国南河上游流域养分模拟中的对比分析","authors":"Chayut Pinichka , Srilert Chotpantarat , Kyung Hwa Cho , Wattasit Siriwong","doi":"10.1016/j.jenvman.2025.126053","DOIUrl":null,"url":null,"abstract":"<div><div>Agricultural runoff leading to nitrate (NO<sub>3</sub>-N) and orthophosphate (PO<sub>4</sub>-P) contamination poses significant environmental and public health risks. This study integrates the Soil and Water Assessment Tool (SWAT) with eXtreme Gradient Boosting (XGBoost), optimized using Optuna hyperparameter tuning, to enhance predictions of nutrient concentrations in water bodies. The hybrid SWAT-XGBoost model combines physical hydrological processes with machine learning for improved accuracy.</div><div>The optimized model demonstrated significant improvements over SWAT alone, with R<sup>2</sup> values increasing by up to 50 % and RMSE decreasing by up to 75 % across datasets. These results highlight the enhanced predictive capabilities of the hybrid approach in capturing nutrient transport dynamics. SHAP analysis further identified key factors, such as sediment dynamics and nutrient mineralization, as dominant drivers of contamination, providing actionable insights for effective watershed management.</div><div>By integrating SHAP analysis, the study identified key processes influencing nutrient transport, such as sediment dynamics and nutrient mineralization, offering deeper insights into pollution pathways. This approach provides a scalable and adaptable framework for improving watershed management, supporting sustainable agricultural practices, and mitigating contamination risks. The findings highlight an approach that combines physical modeling with machine learning to effectively address complex environmental challenges.</div></div>","PeriodicalId":356,"journal":{"name":"Journal of Environmental Management","volume":"388 ","pages":"Article 126053"},"PeriodicalIF":8.4000,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Comparative analysis of SWAT and SWAT coupled with XGBoost model using Optuna hyperparameter optimization for nutrient simulation: A case study in the Upper Nan River basin, Thailand\",\"authors\":\"Chayut Pinichka , Srilert Chotpantarat , Kyung Hwa Cho , Wattasit Siriwong\",\"doi\":\"10.1016/j.jenvman.2025.126053\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Agricultural runoff leading to nitrate (NO<sub>3</sub>-N) and orthophosphate (PO<sub>4</sub>-P) contamination poses significant environmental and public health risks. This study integrates the Soil and Water Assessment Tool (SWAT) with eXtreme Gradient Boosting (XGBoost), optimized using Optuna hyperparameter tuning, to enhance predictions of nutrient concentrations in water bodies. The hybrid SWAT-XGBoost model combines physical hydrological processes with machine learning for improved accuracy.</div><div>The optimized model demonstrated significant improvements over SWAT alone, with R<sup>2</sup> values increasing by up to 50 % and RMSE decreasing by up to 75 % across datasets. These results highlight the enhanced predictive capabilities of the hybrid approach in capturing nutrient transport dynamics. SHAP analysis further identified key factors, such as sediment dynamics and nutrient mineralization, as dominant drivers of contamination, providing actionable insights for effective watershed management.</div><div>By integrating SHAP analysis, the study identified key processes influencing nutrient transport, such as sediment dynamics and nutrient mineralization, offering deeper insights into pollution pathways. This approach provides a scalable and adaptable framework for improving watershed management, supporting sustainable agricultural practices, and mitigating contamination risks. The findings highlight an approach that combines physical modeling with machine learning to effectively address complex environmental challenges.</div></div>\",\"PeriodicalId\":356,\"journal\":{\"name\":\"Journal of Environmental Management\",\"volume\":\"388 \",\"pages\":\"Article 126053\"},\"PeriodicalIF\":8.4000,\"publicationDate\":\"2025-06-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Environmental Management\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0301479725020298\",\"RegionNum\":2,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Environmental Management","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0301479725020298","RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
Comparative analysis of SWAT and SWAT coupled with XGBoost model using Optuna hyperparameter optimization for nutrient simulation: A case study in the Upper Nan River basin, Thailand
Agricultural runoff leading to nitrate (NO3-N) and orthophosphate (PO4-P) contamination poses significant environmental and public health risks. This study integrates the Soil and Water Assessment Tool (SWAT) with eXtreme Gradient Boosting (XGBoost), optimized using Optuna hyperparameter tuning, to enhance predictions of nutrient concentrations in water bodies. The hybrid SWAT-XGBoost model combines physical hydrological processes with machine learning for improved accuracy.
The optimized model demonstrated significant improvements over SWAT alone, with R2 values increasing by up to 50 % and RMSE decreasing by up to 75 % across datasets. These results highlight the enhanced predictive capabilities of the hybrid approach in capturing nutrient transport dynamics. SHAP analysis further identified key factors, such as sediment dynamics and nutrient mineralization, as dominant drivers of contamination, providing actionable insights for effective watershed management.
By integrating SHAP analysis, the study identified key processes influencing nutrient transport, such as sediment dynamics and nutrient mineralization, offering deeper insights into pollution pathways. This approach provides a scalable and adaptable framework for improving watershed management, supporting sustainable agricultural practices, and mitigating contamination risks. The findings highlight an approach that combines physical modeling with machine learning to effectively address complex environmental challenges.
期刊介绍:
The Journal of Environmental Management is a journal for the publication of peer reviewed, original research for all aspects of management and the managed use of the environment, both natural and man-made.Critical review articles are also welcome; submission of these is strongly encouraged.