Seasonality of agricultural exposure as an important predictor of seasonal yellow fever spillover in Brazil

Yellow fever virus (YFV) is a zoonotic arbovirus affecting humans and non-human primates (NHPs) in Africa and South America. Prior explanations of yellow fever (YF) seasonality have focused on climate, despite many cases occurring in people engaged in agriculture. Using random forest classification models, the authors predicted monthly occurrence of YF in humans and NHPs across Brazil by fitting four classes of covariates: host demography, agricultural seasonality (planting/harvesting), agricultural output, and climate/vegetation seasonality. Models captured seasonal YF reporting in humans and NHPs more accurately when including agricultural seasonality rather than climate alone, particularly for monthly aggregates. Results highlight seasonality of exposure through agricultural activities as a key component of zoonotic spillover and identify crop types and anthropogenic seasonality that could help target areas at highest spillover risk.