Analysis of Local Variation of Soil Surface Parameters With TerraSAR-X Radar Data Over Bare Agricultural Fields

The objective of this paper is to analyze the sensitivity of very high resolution TerraSAR-X radar data taken over bare soils to surface soil parameters and to study the spatial variability of these parameters at a fine scale (within a field plot). The relationship between the backscattering coefficient and the soil's parameters (moisture, surface roughness, texture, and local topography) was examined by means of four satellite images, as well as ground truth measurements, of each of the three agricultural plots, recorded during several field campaigns in the winter and spring of 2008. TerraSAR images demonstrate high potential for the identification of local variations of roughness and texture. An approach for the estimation of local moisture is proposed using an empirical method adapted to the scale of an individual field. The results show that, by using TerraSAR-X data to study bare agricultural fields, local variations in soil moisture can be retrieved with a root-mean-square error of 0.05 $\hbox{cm}^{3} \cdot \hbox{cm}^{-3}$.