Bird-building collisions increase with weather conditions that favor nocturnal migration and with inclement and changing weather

Lay Summary center dot Collisions with windows are a top bird mortality source. Information about weather effects on bird-window collisions could help predict collisions. center dot In Minneapolis, Minnesota, we studied 21 buildings over 4 migrations to evaluate how weather affects collisions for birds that migrate at night. center dot Spring and fall collisions increased with weather favoring migration, like tailwinds. We also found evidence that collisions increase with weather changes that cause migrating birds to be grounded, like changes from tailwinds to headwinds during the night. center dot We also found more-nuanced weather effects, including possible effects of weather 2-3 nights before collisions occur, and of multi-day sequences of conditions. center dot Weather forecasts may help target temporary measures to reduce bird-window collisions, but relationships are complex, and recommendations to building managers could be difficult to implement on short notice. More-permanent practices, like turning off nighttime lights throughout migration and using bird-friendly building design and glass treatments, should therefore be prioritized. Collisions with building windows are a top bird mortality source, but few studies have evaluated how bird-window collisions are influenced by weather. By monitoring collisions daily at 21 buildings in Minneapolis, Minnesota, over 4 migration seasons, we show that weather influences numbers of window collisions of nocturnal migrants in spring and fall, indicating that collisions may be forecastable based on weather conditions. Collisions increased with weather favoring migration, such as consecutive nights of south wind in spring and even short periods of north wind in fall. We also found evidence that spring and fall collisions increase with weather changes that impede migration, such as changes from fair conditions and tailwinds early in the night to headwinds near sunrise. Our study suggests complex weather effects never before considered in the context of bird collisions, including possible time lag effects of conditions 2-3 nights before collisions occur, effects of multi-day sequences of conditions, and interactions between conditions at different times of night. More research is needed to determine if the accuracy of weather-based collision prediction systems improves by integrating such nuances and to clarify mechanisms through which these complex effects operate, such as influences of weather on migration intensity and collision avoidance behavior. Weather-based forecasts may allow refinement of collision mitigation approaches (e.g., reducing building lighting on certain nights or using temporary glass coverings or treatments). However, because challenges remain to communicating such temporally targeted actions and implementing them in a timely manner, other bird-friendly practices (e.g., season-long lighting reduction and permanent glass treatments) should continue to be prioritized.