A decision classifier to classify rivers for river management based on their structure in China: an example from the Yongding river.

River classification is a very useful tool for river management yet still a difficult task. This paper proposed a new decision classifier (DCF) to classify rivers for Chinese river management based on existing classification systems. Aimed at river function management, the DCF with the five-layers frame was developed on reach level in a spatially nested pattern that from top to bottom are natural province, basin, valley, reach, habitat and microhabitat. Five indexes (artificial degree, closeness, sinuosity, bed material texture, geomorphic units (GUs)) were selected and organized into the DCF according to the importance of the influence on river structure from macro to micro, large to small and top to bottom, because they represent main aspects of river structures and are easy to obtain. In addition, the closeness index is another good connector between valley level and reach level, and the GUs index links reach level to habitat level. The overall procedure to use DCF includes primary indoor classification and field validation. Remote sensing, geographical information system and global positioning system technologies were adopted in the process to dramatically reduce workload, especially fieldwork. Finally, the approach was applied to the Yongding river as a good example, and 17 river styles were identified.