Prédiction de la distribution des glossines dans la zone des Niayes par l'utilisation de télédétection

Mémoire ingénieur

Master

English. African animal trypanosomiasis (AAT) are a major pathological constraints to livestock in the Niayes (Senegal). In the Niayes area, the main vector of this disease, Glossina palpalis gambiensis (Diptera : Glossinidae) is present and this species is subject to an eradication campaign launched in 2005. The aim of the study is to establish a spatial distribution of the species in order to define an infested area. This infested area will be used in a targeted control. An approach based on ecological knowledge of the species has been developed to fight against this vector borne-disease. We used several species distribution models to understand the impact of environmental variable on the species. But we also used those models to map the potential distribution of G. p. gambiensis through our area of study. Theoretical knowledge on the species were operated through the theory of ecological nichesfrom which distribution models are based. The data required for modeling were obtained using satellite image (MODIS). This time series of images were used to derive climatic and environmental data which could be linked to tsetse fly distribution. The analysis of the species responses to these variables have been made through exploratory analysis based on the concept of ecological niche that allowed to describe and select variables that influence the most the distribution of the vector of the AAT. The Resultsshowed a high sensitivity to variables related to vegetation. These variables were then used by three predictive models to delineate the targeted area. Among those models, two have a good predictive power (AUC = 0.8), and the best of them (MaxEnt) were used to obtain results that are consistent with the ecology of the species. Hopefully, the former control area are not going to change a lot. However, these models showed some suitable area wherethere were no sampling.