Remote sensing of anophelines in rice-cropping villages in Mali: Patterns of vector abundance and malaria transmission

The explosive population growth and widespread urbanization in Africa requires a significant increase in food production. Crop irrigation is therefore expected to increase in the future, although it is often blamed for aggravating the health risk of local communities—by providing habitats suitable for mosquitoes vectors of malaria (Anopheles gambiae s.l. and An. funestus in our study area) and other diseases. An epidemiological paradox sometimes occurs, however, when an increase in vector numbers is accompanied by a reduction of the risk of infection, due to a reduction in mosquito longevity and of their tendency to bite human (vs. animals). The objective of this dissertation was to determine how agricultural patterns mapped using satellite data affected vector densities and malaria transmission parameters in 18 rice-cropping villages in Mali. I used a combination of optical (Landsat ETM+) and synthetic aperture radar (ERS-2 SAR). Using Landsat data, rice was distinguished from other land uses with 98% accuracy and rice cohorts were discriminated with 84% accuracy (three classes) or 94% (two classes). ERS-2 SAR backscatter was correlated with the height and biomass of rice plants and was therefore useful to distinguish among rice growth stages. As in previous studies, the early vegetative stage was associated with higher larval production. SAR was further able to distinguish between agronomic practices linked to high and low-production within those early stages. The landcover maps were integrated with archived data on adult and larval anopheline densities and malaria transmission parameters. The area of several landcovers explained up to 89% of the variability in mosquito numbers. The maximum correlation was obtained when landcover was measured in a 1-km buffer area. Vector density was negatively associated to parity and anthropophilic rates. An. gambiae showed higher vectorial capacity (VC) than An. funestus , with seasonal variations. Peak VC for both species was in April 2000, coincident with an irrigation shortage that limited double cropping. The resulting ‘intermediate’ An. gambiae density was coupled with peak parity rates and VC. ‘Intermediate’ An. gambiae densities may therefore result in the highest level of malaria transmission.