Multiscale surface roughness of bare soil

Up to very recently the remote sensing community has used for vertical and horizontal surface roughness a description based on single scale rms height and correlation length, but they do not characterise all the properties of natural surfaces that are relevant for scattering. These two parameters are not intrinsic properties of the surface, but they depend on the measurement length (outer scale) and the resolution (inner scale). The capability of studying pixel size surface statistics is thus important, since not a single surface can be proved to be stationary by measurements, because the results are affected by the inner and outer scales of the measurements. The objective of the present work has been to develop a realistic surface roughness description for natural surfaces like agricultural fields. The surface is described with the inclination angle obtained by linear regression of the surface and multiscale surface roughness on top of this inclination. The main idea of the multiscale surface roughness description is that the surface roughness is based on a superposition of various wavelengths.