Influence of resolution on slope in areas with different topographic characteristics

This article studied slope scaling effects in six study areas in China. The areas had different terrain characteristics and were examples of distinct regions that are regarded in China as a distinct form of modeling and assessment of erosion and hydrology. The resolutions used in this paper including 10, 25, 50 and 100 m. Mean slope, standard deviation of slope and slope distribution were calculated as basic slope statistics in the six study areas at four resolutions to generate information on change with resolution and rate of change with resolution. A theoretical slope distribution model was developed and used to model the basic statistics to help to understand the changes with resolution. Geostatistical methods were also used to help to understand the nature of the changes. The results showed that: (1) The change and rate of change of slope statistics with resolution differed between the study areas; (2) a simple driving parameter in the theoretical model (sig_g) was useful in studying the changes of slope basic statistics with resolution since direct relationships exist between it and the other slope statistics; (3) the geostatistical analysis showed how the way terrain structure changed with resolution corresponded to the correlation distance of slope surface increasing and the total variance decreasing as resolution became coarser and (4) the change and rate of change of slope basic statistics were found to be related to the changing rate of range (or correlation length) for detailed structures (R_origin). This suggested that the change and rate of change of slope with resolution were decided by the amount of higher spatial frequency information (small valleys, gullies and hills) at which scale would be smoothed by the given resolution changing process and that rate of change analysis provided a useful way to analyze the composition of multi-resolution structures.

► Changes and rates of change of slope with resolution were studied in six sites.
► The slope statistics change was modeled and explained.
► Semi-variogram range structure and slope change with resolution were related.
► The variance of elevation gradient could describe most slope statistics.
► Slope change with resolution and differences in terrain structure are well related.