Abstract
To predict the spatial distribution of anopheles in the Dielmo village (located in the southeastern part of Senegal), we used residual fauna collected from 104 different rooms during four separate trips conducted in 1994 and 1995. Thanks Generalized Estimating Equations, we were able to identify factors influencing the distribution of Anopheles in the village. Several variables, such as the number of persons sleeping in the room, population density around the hut, construction materials, presence of mosquito nets, were found to be significant, while many spatial variables relevant to the scale of a region (vegetation index, distance to larval sites...) were not found to be significant on the village level. As a result, it became clear that it is difficult to correctly predict the anopheline density for each house even with precise spatial data created with Satellite imagery and Geographical Information Systems (GIS). This work highlights the complexity of the geographical study of anopheline density and its limits on a small space.