Towards a kala azar risk map for Sudan: Mapping the potential distributionof Phlebotomus orientalis using digital data of environmental variables

The need to define the geographical distribution of Phlebotomus orientalis results from its importance as the dominant vector of kala azar (visceral l eishmaniasis) in Sudan. Recent epidemics of this disease in southern and ea stern Sudan caused an estimated 100000 deaths and have renewed the impetus for defining the ecological boundaries of the vector. This information is a n essential prerequisite to the production of a risk map for kala azar. Thi s study uses data on the presence and absence of P. orientalis from 44 Coll ecting sites across the central belt of Sudan. A logistic regression model was used to estimate the probability of the presence of P. orientalis at ea ch collecting site as a function of climatic and environmental variables (r ainfall; temperature; altitude; soil type and the satellite-derived environ mental proxies - Normalized Difference Vegetation Index and Land Surface Te mperature). The logistic regression model indicates mean annual maximum dai ly temperature and soil type as the most important ecological determinants of P. orientalis distribution. An initial risk map was created in a raster- based geographical information system which delineates the area where P. or ientalis may occur. This map was then refined using a mask layer indicating the known rainfall-based boundaries of the distribution of Acacia-Balanite s woodland - a woodland type known to be associated with the distribution o f this vector. The predictive performance of the risk map is discussed.