Modeling the relationship between landscape characteristics and water quality in a typical highly intensive agricultural small watershed, Dongting lake basin, south central China

Understanding the relationship between landscape characteristics and water quality is critically important for estimating pollution potential and reducing pollution risk. Therefore, this study examines the relationship between landscape characteristics and water quality at both spatial and temporal scales. The study took place in the Jinjing River watershed in 2010; seven landscape types and four water quality pollutions were chosen as analysis parameters. Three different buffer areas along the river were drawn to analyze the relationship as a function of spatial scale. The results of a Pearson's correlation coefficient analysis suggest that "source" landscape, namely, tea gardens, residential areas, and paddy lands, have positive effects on water quality parameters, while forests exhibit a negative influence on water quality parameters because they represent a "sink" landscape and the sub-watershed level is identified as a suitable scale. Using the principal component analysis, tea gardens, residential areas, paddy lands, and forests were identified as the main landscape index. A stepwise multiple regression analysis was employed to model the relationship between landscape characteristics and water quality for each season. The results demonstrate that both landscape composition and configuration affect water quality. In summer and winter, the landscape metrics explained approximately 80.7 % of the variance in the water quality variables, which was higher than that for spring and fall (60.3 %). This study can help environmental managers to understand the relationships between landscapes and water quality and provide landscape ecological approaches for water quality control and land use management.