ln the framework of the European project CoForChange (predicting the effects of global change on forest biodiversity in the Congo Basin region) we propose a new method of satellite imagery analysis to characterize the spatial patterns of tropical forest in Central Africa. We used 10-year time series remotely sensed data from TERRAMDDI5 sensor to identify the variations of forest photosynthetic activity intensity within a synthetic year. The processing was done using the 16-days enhanced vegetation index (EVI) data. We crossed the resulting information with a data base of field inventories provided by forest companies (Central African Republic, Republic of Congo, and Cameroon). We found that the vegetation types detected with satellite imagery match with a gradient of evergreen dominance. These results are of crucial importance to predict spatial variations of carbon stocks and fluxes, because evergreen/deciduous forests (i) have contrasted annual dynamics of photosynthetic activity and foliar water content; (ii) differ in their community dynamics; and (iii) differ in ecosystem processes. This detailed spatial and temporal information should provide a better basis to monitor carbon in this poorly known region.