Role of the repartition of wetland breeding sites on the spatial distribution of Anopheles and Culex, human disease vectors in Southern France

Using a logistic regression model, we first evaluated which land cover variables determined the presence of Culex and Anopheles larva. The resulting probability map of larval presence then was used to project the average probability of finding adults in a buffer area. This was compared to the actual number of adults collected, providing a quantitative assessment of adult dispersal ability for each species.The distribution of Cx. modestus and An. melanoon is mainly driven by the repartition of irrigated farm fields and reed beds, their specific breeding habitats. The presence of breeding sites explained the distribution of adults of both species. The buffer size, reflecting the adult dispersal ability, was 700 m for Cx. modestus and 1000 m for An. melanoon. The comparatively stronger correlation observed for Cx. modestus suggested that other factors may affect the distribution of adult An. melanoon. We did not find any association between Cx. pipiens larval presence and the biotope due to the species' ubiquist character.By applying the same method to different species, we highlighted different strengths of association between land cover (irrigated farm fields and reed beds), larval presence and adult population distribution.This paper demonstrates the power of geomatic tools to quantify the spatial organization of mosquito populations, and allows a better understanding of links between landcover, breeding habitats, presence of immature mosquito populations and adult distributions for different species.Environmental conditions may determine the presence of species and their population dynamics, especially for insects which are highly dependent on climatic conditions and landscape organization [1-4]. Geographical Information Systems (GIS) and spatial analyses are used widely in ecology [4-8] to understand the relation between habitats and the presence or abundance of species. When applied to pathogen vectors, these tools render it possible to develop control strategies