Remote sensing based identification of environmental risk factors associated with West Nile disease in horses in Camargue, France.

Geographic information system and remote sensing technologies were used to identify landscape features associated with risk of West Nile virus transmission as defined by the presence of confirmed horse cases. SPOT-4 images of Camargue area were used to generate a map of landscape categories of epidemic foci and the geographic information system was employed to determine the proportion of landscape components surrounding 10 horse case sites and 17 control sites. The spatio-temporal analysis of the cases outbreak gave the best results for a spatial window of 9 km and a temporal window of 18 days. Two clusters were identified (relative risk=3.35), both in the wet area of Camargue, near the town "Les Saintes Maries de la Mer". The first one was 4 km radius, the second 9 km. The relationships between the presence of cases and proportions of landscape categories were analyzed using generalized linear model. The best model indicated that rice fields and dry bushes, wet "sansouire" and open water were the major components of the landscape that were associated with the presence of West Nile virus cases.