COSMO-SkyMed imagery for crops characterization

This paper shows some preliminary results on the exploitation of COSMO-SkyMed (CSK) data for the characterization of agricultural fields. In particular, we analyze the impact of the planting row direction on to the radar backscattering response. We considered two different crop types, namely carrots and potatoes, and we grouped them in two subsets, the first including fields with row direction roughly perpendicular to the Synthetic Aperture Radar (SAR) Look Direction (LD) and the second including fields with row direction parallel to the SAR LD. A multi-temporal analysis of the CSK backscattering coefficients for this two different sensor-target acquisition geometries, at three different incidence angles and at VV, HH, VH polarization are presented. Moreover, the influence of the row direction on the retrieval of the Normalized Difference Vegetation Index (NDVI) was investigated. The preliminary results show that CSK at VH polarization is sensitive to crop biomass independently of the sensor incidence angle and rows direction of the crops fields. On the contrary, the backscattering coefficients at HH and VV polarization are sensitive only when the sensor incidence angle is parallel to crops rows.