Housing development erodes avian community structure in U.S. protected areas.

Protected areas are a cornerstone for biodiversity conservation, but they also provide amenities that attract housing development on inholdings and adjacent private lands. We explored how this development affects biodiversity within and near protected areas among six ecological regions throughout the United States. We quantified the effect of housing density within, at the boundary, and outside protected areas, and natural land cover within protected areas, on the proportional abundance and proportional richness of three avian guilds within protected areas. We developed three guilds from the North American Breeding Bird Survey, which included Species of Greatest Conservation Need, land cover affiliates (e.g., forest breeders), and synanthropic species associated with urban environments. We gathered housing density data for the year 2000 from the U.S. Census Bureau, and centered the bird data on this year. We obtained land cover data from the 2001 National Land Cover Database, and we used single- and multiple-variable analyses to address our research question. In all regions, housing density within protected areas was positively associated with the proportional abundance or proportional richness of synanthropes, and negatively associated with the proportional abundance or proportional richness of Species of Greatest Conservation Need. These relationships were strongest in the eastern forested regions and the central grasslands, where more than 70% and 45%, respectively, of the variation in the proportional abundance of synanthropes and Species of Greatest Conservation Need were explained by housing within protected areas. Furthermore, in most regions, housing density outside protected areas was positively associated with the proportional abundance or proportional richness of synanthropes and negatively associated with the proportional abundance of land cover affiliates and Species of Greatest Conservation Need within protected areas. However, these effects were weaker than housing within protected areas. Natural land cover was high with little variability within protected areas, and consequently, was less influential than housing density within or outside protected areas explaining the proportional abundance or proportional richness of the avian guilds. Our results indicate that housing development within, at the boundary, and outside protected areas impacts avian community structure within protected areas throughout the United States.