Se of the data satellite weather and expert nowled e for pre ention and mana ement of the acridien ris

The Desert locust represents a significant threat for agriculture in many areas of the ground, particularly the African continent. The invasions of desert locust put in evidence the necessity for having a permanent system, for the fight against these insects. This study intended to propose a methodology for the modelling of displacements of the desert locust in gregarious phase, can be integrated within the framework of such a system. This modelling, passes to the precondition by the design of a semi automatic system by combining the weather, the satellite data, and the integral of the expert knowledge expressed in the form of rules of explicit inferences drawn from the dynamics of population of the desert locust. For the methodology implementation, we tried to model displacements of the swarms of the desert locust, in the steppe zone of Algeria, starting from the descriptions given by the Food and Agriculture rganization of the United Nations (FA ). The rules associated with our application relate to variables, in the form of pre treated and generated data, as well as intermediate data obtained starting from functions. The pre treated data used is that of the Digital Elevation Model resulting from world data SRTM. The generated data are those obtained by the oron s diagram starting from the weather data. nly one function was used for obtaining an intermediate data, by using the red and near infra red channels of optical data M DIS, this function relates to the Normalized Difference egetation Index.