Detecting land use-water quality relationships from the viewpoint of ecological restoration in an urban area

Urbanization increases impervious area, generates pollution and transforms the configuration, composition and context of land covers and thus has direct or indirect impacts on aquatic systems. Detecting land use-water quality relationships is of significance for both urban sustainable development and environmental risk management. We first selected suitable landscape metrics at landscape and class levels as well as productivity as land use indicators; second, we integrated ten water quality indicators into a comprehensive pollution index (CPI) that serves as the primary water quality indicator for measuring the effects of diffusive pollution in an urban area; finally, taking water quality indicators as dependent variables and land use indicators as independent variables, we detected land use-water quality relationships using both ordinary least squares (OLS) and geographically weighted regression (GWR) models. The main findings are (1) the CPI is a good water quality indicator for detecting land use-water quality relationships and the performance of the GWR model is superior to that of the OLS model and (2) landscape metrics have potential for predicting water quality; however, improving the productivity of natural vegetation is a more tangible and reliable way to promote water resources restoration than rearranging the configuration of land use types.

► A new comprehensive index integrating multiple water quality parameters was designed. ► A methodology and standard for selecting suitable land use indicators were suggested. ► Vegetation productivity is a better indicator to water quality than pattern metrics. ► Our method can detect the interaction hotspots between land use and water quality. ► We focused on what can be done to mitigate the effect of land use on water quality.