Among the various soil properties, drainage is the most important one for land productivity as well as the environmental management. Classification and mapping of regional and local soil drainage conditions and hydrologic soil groups are often required as input in many hydrological models used for assessing soil degradation processes and soil vulnerability to environmental, losses under intensive agricultural production. The objective of this study was to evaluate the capability of high resolution SAR remote sensing data for mapping soil drainage. Backscattering coefficients (HH, VH, VV, RR, LL, ψ45° and ψ135°) extracted from CV-580 SAR were analyzed as function of drainage classes according to the land use (bare soil, annual crops). The multipolarization data showed a maximum of difference of 2 dB between drainage classes. Discriminant analysis performed after applying Principal Component Analysis (PCA) showed overall classification accuracy of 60% for drainage classes. Only the poorly drained class was clearly discriminated (73-97%) under agricultural fields. Introducing the Polarimetric parameters (mean a angle, entropy (H), and anisotropy (A)) in, the general disciminant classification, the soil drainage classification accuracy was improved particularly by the anisotropy parameter. The study demonstrated the potential, of C-band multipolarized and Polarimetrie SAR for soil drainage classification and mapping. However, the high spatial resolution of the CV-580 showed high variability of scattering mechanisms associated to the fall acquisition period, where many crop fields were harvested but some corn fields being not. Spring acquisition near seeding of RADARSAT-2 (launched on December 14, 2007) imagery, should give better classification results. This hypothesis will be tested next spring (May 2008).