The roles of mosquito and bird communities on the prevalence of West Nile virus in urban wetland and residential habitats.

This study investigated the impacts of urban wetlands and their adjacent residential environments on the transmission dynamics of West Nile virus (WNV) within the state of New Jersey (USA). A working hypothesis was that urban wetlands decrease the local prevalence of WNV through the dilution effect from increased bird diversity, and through relative reductions in the numbers of competent avian host and mosquito species commonly associated with WNV. Surveys of mosquito and bird communities were undertaken at six urban wetlands and their adjacent residential environments over two seasons (2009, 2010). The community compositions of both avian and mosquito species differed significantly across habitats, and over relatively short geographical distances. Residential areas contained significantly higher proportions of WNV-competent mosquito species (31.25±5.3 %; e.g. and , and WNV-competent avian host species (62.8±2.3 %, e.g. House Sparrow and American Robin) when compared to adjacent urban wetlands (13.5±2.1 %; 35.4±2.1 % respectively). Correspondingly, WNV infection rates within local populations indicate that WNV was more prevalent within residential areas (28.53/1000) compared to wetlands (16.77/1000). Large urban wetlands (>100 ha) produced significantly lower weekly WNV infection rates in local (6.67±2.84/1000) compared to small (<15 ha) wet-lands (22.57±6.23/1000). Avian species richness was also influenced by patch size. Large urban wetlands contained significantly more species than small wetland patches. These results confirm that the community compositions of mosquito and avian hosts are important drivers in WNV infections, and that the ecological conditions that favor transmission are more strongly associated with urban residential environments than with adjacent urban wetlands.