Abstract: In this paper, the analysis units were upgraded to the minor drainage basin level, and the subcatchments of the micro-scale oriented to the “cell” level were proposed as the classification idea of analysis object. The GIS spatial analysis was used to construct the geographic attribute database for subcatchments based on the available geomorphic data of the study area. Four key classification indicators including the elevation, relief degree of land surface, average slope and slope skewness were selected. The hierarchical classification method was used to set the thresholds of the corresponding topographic factors, and 8 types of landforms, i.e. the flood plains, residual loess tableland, loess platform, steep-slope gully, girder-like hilly gully, small rocky undulating temperate mid-mountain, small stone undulating steep-mountain and waters in Yaozhou District of Tongchuan City, Shaanxi Province were divided. The relative integrity of the landforms was maintained in the classification as much as possible. Compared with the field sampling observations performed by GPS, the classification accuracy of the study was as high as 88.19%, which is expected to be a digital geomorphological classification based on subcatchment.

Key words:  subcatchment, landform classification, drainage basin, grid DEM, Tongchuan, Shaanxi