 This study investigated the potential of SAR images to improve early crop type mapping compared to optical images alone. It was found that SAR images can provide significant improvements in terms of land cover classification accuracy, both at the end of the season and for early crop identification. The most relevant SAR image features were identified as heralic textures, entropy, inertia, the polarization ratio and the local mean together with the VV imagery. Working at 10M resolution and using speckle filtering yielded the best results. Furthermore, it was demonstrated that the use of SAR imagery allows to use optical data without gap filling, thus avoiding the need for gap filling in the case of imperfect cloud screening. When combined with imperfect cloud screening, the use of SAR imagery yields results which are equivalent or even better than the use of gap filling. This article was authored by Jordi Anglotta, Arthur Vincent, Marcella Arias and others.