Division Design of Water Eco‐Functioning of the River Basin

Abstract

Regionalization plays an important role in delineating landscape heterogeneity and providing spatial frameworks for environmental management. Historically, the most widely used method for regionalization is qualitative synthesis of various mixes of environmental or biotic data as the whole. However, such methods based on expert judgment and whole data set are less repeatable and defensible than their quantitative counterparts. In this study, a comprehensive regionalization approach that integrated multiple quantitative techniques was developed to differentiate two distinct types of landscape variables, i.e., response attributes and driving factors. This approach was applied to the regionalization in the Huai River Basin (HRB), China. In the regionalization scheme, the sub‐watersheds of the basin were adopted as the basic spatial unit; different response attributes were used to characterize different water eco‐functioning. The sub‐watersheds were classified into a certain number of groups using the k‐means cluster analysis based on the response attributes. A goodness‐of‐fit index, calculated as the ratio of between‐group/within‐group variation, was employed as a quantitative criterion to assess the statistical performance of the classification results. Driving forces were used to assess the strengths of the forces operating at different spatial and temporal scales on water eco‐functioning. A general, quantitative, regionalization framework was developed; this framework can reveal underlying mechanisms of basin divisions and landscape patterns through integrating response attributes and driving factors. This framework was implemented by using a basin‐scale regionalization of Huai River Basin as an example; however, the framework can be adapted to other regionalization practices within broad contexts.