Use of thermal and vegetation index data from earth observing satellites to evaluate the risk of schistosomiasis in Bahia, Brazil

A geographic information system (GIS) was constructed using maps of regiona l agroclimatic features. vegetation indices and earth surface temperature d ata from environmental satellites, together with Schistosoma mansoni preval ence records from 270 municipalities including snail host distributions in Bahia, Brazil to study the spatial and temporal dynamics of infection and t o identify environmental factors that influence the distribution of schisto somiasis. In an initial analysis, population density and duration (months) of the annual dry period were shown to be important determinants of disease . In cooperation with the National Institute of Spatial Research in Brazil (INPE), day and night imagery data covering the state of Bahia were selecte d at approximately bimonthly intervals in 1994 (six day-night pairs) from t he data archives of the advanced very high resolution radiometer (AVHRR) se nsor of the National Oceanic and Atmospheric Administration (NOAA)-11 satel lite. A composite mosaic of these images was created to produce maps of: (1 ) average values between 0 and + 1 of the normalized difference vegetation index (NDVI); and (2) average diurnal temperature differences (dT) on a sca le of values between 0 and 15 degreesC. For each municipality. NDVI and dr were calculated for a 3 x 3 pixel (9 km(2) area) grid and analyzed for rela tionships to prevalence of schistosomiasis. Results showed a statistically significant relationship of prevalence to dT(rho = - 0.218) and NDVI (rho = 0.384) at the 95% level of confidence by the Spearman rank correlation coe fficient. Results support use of NDVI, dT; dry period climatic stress facto rs and human population density for development of a GIS environmental risk assessment model for schistosomiasis in Brazil. (C) 2001 Elsevier Science B.V. All rights reserved.