Research On The Relationship Between Landform Types And Distribution Of Slope

Slope is an important indicator of morphology, watershed hydrology and evaluation of soil erosion. Most of the previous studies focused on algorithms, uncertainty, and transformation of slope, which are usually limited to small scale such as hillslope and watershed scale. The studies about slope statistical distribution, as well as its relationship between statistical metrics and landform types at large scale are relatively weak. In this study, slope distribution indicators (such as relief, hypsometric integral, elevation gradients, slope steepness, distributed slope length, USLE-LS factor, etc.) that are based on higher resolution DEMs have been calculated within a 5km*5km extent, and then raster maps have been generated. The relationship between metrics and landform types has been analyzed primarily. At the same time, detailed spatial pattern and statistical distribution has been analyzed in 17 sampling sites. The results suggest that elevation increases from valley, loess hills and sandy hill, earth hills, to eastern part of Qinghai-Tibet Plateau. They also show that statistical parameters, such as relief, elevation gradients, slope steepness, and USLE-LS factor) in mountainous areas are lager, while in valley and sandy hills are smaller, and in loess hills with severe erosion are moderate. Slope in loess hills exhibits a unimodal distribution and near normal distribution, while LSfactor obeys a bi-modal distribution possibly due to the interaction of the slope and slope length.