Predicting bird-window collisions with weather radar

Up to 1 billion birds die annually in the U.S. from window collisions; most of these casualties represent migratory native species. Because this major mortality source likely contributes to the decline of the North American avifauna, mitigation tools are needed that accurately predict real-time collision risk, allowing hazards to be minimized before fatalities occur. We assessed the potential use of weather surveillance radar, an emerging tool increasingly used to study and to predict bird migration, as an early warning system to reduce numbers of bird-window collisions. Based on bird-window collision monitoring in Oklahoma, USA, we show that radar-derived migration variables are associated with nightly numbers of collisions. Across the entire night, numbers of collisions increased with higher migration traffic rate (i.e. numbers of birds crossing a fixed line perpendicular to migration direction), and migration variables for specific periods within the night were also related to nightly collisions. Synthesis and applications. Our study suggests that radar can be an invaluable tool to predict bird-window collisions and help refine mitigation efforts that reduce collisions such as reducing nighttime lighting emitted from and near buildings.