Exploring the Impact of Land Use Scales on Water Quality Based on the Random Forest Model: A Case Study of the Shaying River Basin, China

Abstract

Optimizing the land use structure is one of the most effective means of improving the surface water aquatic environment. The relationship between land use patterns and water quality is complex due to the influence of dams and sluices. To further investigate the impact of land use patterns on water quality in different basins, we use the Shaying River as an example, which is a typical tributary of the Huai River Basin. Utilizing 2020 land use data and surface water quality monitoring data from two periods, this study employs GIS spatial analysis, the Random Forest Model, redundancy analysis, and Partial Least-Squares Regression to quantitatively explore how different-scale buffer zone land use patterns impact surface water quality. The key findings include: (1) notable seasonal differences in water quality indicators within the basin. The Water Quality Index (WQI) is significantly better in the non-flood season compared to the flood season, with water quality deteriorating towards the lower reaches. Key indicators affecting water quality include dissolved oxygen (DO), ammonia nitrogen (NH3-N), total phosphorus (TP), and turbidity (Tur) in the flood season and NH3-N, permanganate index (CODMn), and electrical conductivity (EC) in the non-flood season. (2) Cultivated land and construction land are the main land uses in the basin. The sub-basin buffer zone was identified as the most effective scale for land use impact on water quality indicators in the Shaying River. (3) Partial Least-Squares Regression (PLSR) analysis revealed that cultivated land, construction land, and grass are the primary land use types influencing surface water quality changes, and the PLSR model is better during the non-flood season. Cultivated and construction lands show a positive correlation with most water quality indicators, while forest land, water bodies, and grasslands correlate positively with DO and negatively with other indicators. The study underscores that rational land use planning in the sub-basin is crucial for enhancing the quality of the surface water environment.