The role of breeding phenology and aggregation of waterfowl on avian influenza dynamics in southern Africa

Recent outbreaks of highly pathogenic avian influenza virus (AIV) in birds, humans and other mammalian species calls for a better understanding of virus dynamics in wild bird species and populations that act as maintenance hosts. Host ecology influences the transmission of pathogens and can be used to explore and infer pathogen dynamics. Most of the ecological processes proposed to explain AIV transmission in wild birds have been derived from studies conducted in the temperate and boreal regions of the northern hemisphere. We evaluate the role of two key drivers of AIV dynamics in a waterfowl community in Zimbabwe (southern Africa): (1) the recruitment of young birds and (2) the seasonal aggregation of birds. We analyse the seasonal variation of AIV prevalence in waterfowl and overlay these data with the phenology of reproduction and the seasonal variation in the local abundance of these species. We find that the breeding period of southern Afrotropical waterfowl species is more extended and somewhat less synchronized among species in the community than is the case in temperate and boreal waterfowl communities. Young birds are recorded at most times of the year, and these immunologically naïve individuals can therefore act as new hosts for AIV throughout the year within the wild bird population. Although host aggregation peaks in the cold-dry to hot-dry season, birds still aggregate throughout the year and this potentially spreads the opportunities for first infection of juveniles and other naïve birds temporally. We did not find a relationship between season, AIV prevalence in waterfowl, the influx of juveniles or the gradual aggregation of birds during the dry season. Therefore, the main drivers of AIV dynamics (juvenile influx and host abundance/aggregation), although present in Afrotropical regions, could not explain the AIV seasonal patterns in our study in contrast to results reported from temperate and boreal regions. These differences imply variation in the risk of AIV circulation in waterfowl and in the risk of spread to poultry, other animals or humans. (Résumé d'auteur