Most developing countries require information on agricultural areas and their monitoring. For cotton production in Senegal, the national company uses farmers' claims and ground enquiries, but there is a very high error rate in the yield forecast and it is impossible to obtain an accurate estimation of production. In this context, space-derived data from satellites can provide useful information.

The present project studies the possibility of crop area monitoring using ERS-1 data, on Senegal. The main objective is to determine the possibility of crop detection and mapping especially for cotton, with a multitemporal data set of radar images and determine appropriate mapping methods.

Throughout the crop cycle, 8 SAR images of the same Track/Frame were acquired, one on each ERS-1 orbit, from May 95 to January 96. During the crop season, a set of ground measurements was made on selected test fields (about 160). The measurement consists of a monthly monitoring of the development for different types of crop (cotton, peanut, cereal). This is completed by geographical coordinates measured by GPS.

All the images are superimposed and this multi-temporal data set is used to study the backscatter. The backscattering coefficients of different crop types and their temporal variation been analyzed as a function of crop development stage and the recorded rainfall. It was found that, for most crop types in this representative agricultural region in Africa, the vegetation covers appear, for the C-band SAR, as sparse medium and the radar backscatter result from the soil backscatter. The difference in the radar backscatter of different crop types relies mainly on the attenuation of the soil backscatter by the vegetation. Based on the results of this analysis, a method of mapping different crop types is developed. In particular, some grouping of crops (cotton and most cereal, peanuts and mil) have been necessary. Also, the filtering and mapping methods have been selected to take into account the small range of the radar backscatter and their temporal variation. The results present a map of different crop types and their acreage.

In collaboration with the national company, the result of this estimation is then compared with complementary information (general knowledge, statistical information). This final evaluation assesses the possibility of deriving important information from SAR data during the crop cycle, whilst the cloud cover over this period does not enable optical data to be acquired.

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry