The aim of this paper is to show the ability of multitemporal short-wave satellite data with high spatial resolution (SPOT data) to characterize crop systems in two ways: (1) description of crop-rotation management by identification of the existing crops and optimization of data acquisition for that purpose; (2) analysis of radiometric time evolution of the crops—evolution of mean response (for radiances and vegetation index) and variability (within and between fields), related to their phenological development, cropping practices and weather conditions. The need for precise corrections of atmospheric effects for multitemporal data series is shown.