The capability of COSMO-SkyMed (CSK) in monitoring vegetation has been investigated in this paper. SAR data from CSK were collected on two agricultural areas in Italy from 2010 to 2012, at different dates during the vegetation cycle. X-band data have been compared to accurate ground truth measurements of soil and vegetation parameters carried out simultaneously to satellite passes. Significant sensitivity of backscatter to vegetation water content of agricultural crops was observed. However, the backscattering showed an opposite trend as a function of biomass of wheat and sunflower, which belong to two very different vegetation types, namely narrow-leaf and broad-leaf crops. Similar trends emerged at lower frequencies (i.e., C and L bands) for the same crop types. In order to investigate the role of different parameters of soil and vegetation (e.g., surface roughness and moisture, plant density and height, dimensions of leaves, stem diameter, and water content) on the backscatter behavior for these two different crop types, model simulations were performed using a discrete element radiative transfer model for vegetation, whereas soil was modeled using the advanced integral equation model (AIEM). A sensitivity analysis of the model was carried out by varying the dimensions of vegetation components within the range of parameters directly measured on ground during the experimental campaigns. The model simulations were successively compared with experimental data of backscattering. The good agreement found between experimental and simulated data encouraged the follow-up of the research toward the implementation of inversion algorithms, which can be able to retrieve vegetation biomass from SAR data and from an operative point of view.