A new multi-resolution based method for estimating local surface roughness from point clouds

From some empirical and theoretical research on the digital elevation model (DEM) accuracy obtained for different source data densities, it can be observed that when the same degree of data reduction is applied to a whole area, the rate of change in the DEM error is statistically greater in local areas where the surface is rougher. Based on this observation, it is possible to characterize surface roughness or complexity from the differences between two digital elevation models (DEMs) built using point clouds that represent the same terrain surface but are of different spatial resolutions (or data spacings). Following this logic, a new approach for estimating surface roughness is proposed in this article. Numerical experiments are used to test the effectiveness of the approach. The study datasets considered in this article consist of four elevation point clouds obtained from terrestrial laser scanning (TLS) and airborne light detection and ranging (LiDAR). These types of topographical data are now used widely in Earth science and related disciplines. The method proposed was found to be an effective means of quantifying local terrain surface roughness.