Exploring the relationship between malaria, rainfall intermittency, and spatial variation in rainfall seasonality

Abstract:

Malaria is a major public health problem throughout tropical regions of the world. Successful prevention and treatment of malaria requires an understanding of the environmental factors that affect the life cycle of both the malaria pathogens, protozoan parasites, and its vectors, anopheline mosquitos. Because the egg, larval, and pupal stages of mosquito development occur in aquatic habitats, information about the spatial and temporal distribution of rainfall is critical for modeling malaria risk. Potential sources of hydrological data include satellite-derived rainfall estimates (TRMM and GPM), evapotranspiration derived from a simplified surface energy balance, and estimates of soil moisture and fractional water cover from passive microwave imagery. Previous studies have found links between malaria cases and total monthly or weekly rainfall in areas where both are highly seasonal. However it is far from clear that monthly or weekly summaries are the best metrics to use to explain malaria outbreaks. It is possible that particular temporal or spatial patterns of rainfall result in better mosquito habitat and thus higher malaria risk. We used malaria case data from the Amhara region of Ethiopia and satellite-derived rainfall estimates to explore the relationship between malaria outbreaks and rainfall with the goal of identifying the most useful rainfall metrics for modeling malaria occurrence. First, we explored spatial variation in the seasonal patterns of both rainfall and malaria cases in Amhara. Second, we assessed the relative importance of different metrics of rainfall intermittency, including alternation of wet and dry spells, the strength of intensity fluctuations, and spatial variability in these measures, in determining the length and severity of malaria outbreaks. We also explored the sensitivity of our results to the choice of method for describing rainfall intermittency and the spatial and temporal scale at which metrics were calculated. Results demonstrate that information about the seasonality and intermittency of rainfall has the potential to improve our understanding of malaria epidemiology and improve our ability to forecast malaria outbreaks.