OOS 51-4
Evaluating the drivers of bird-window collisions in North America
Bird-window collisions (BWCs) are an important human-related threat to bird survival in developed landscapes. BWCs are thought to be affected by building structural features and land use at local and landscape scales, but we know little about whether the drivers of BWCs are consistent among urban areas. In 2013, the EREN network enabled collaboration among 13 sites throughout the eastern United States and central Mexico to assess the drivers of BWCs. Our short-term goals were to complete a pilot field season in fall 2013 to (1) test project methods and protocols and (2) conduct a preliminary analysis of the data collected.
Selection of study buildings, measuring environmental and structural factors, and carcass survey methods were standardized to make data comparable among sites. Carcass surveys were completed at all buildings (N = 87) for 21 consecutive days. We used generalized linear mixed models to assess the relationship between the number of carcasses resulting from window collisions and four factors: total window area for each building, window treatment (clear, tinted, or tinted-reflective), local digitized vegetation cover within 50 m of each building, and proportion of developed land within 10 km buffer surrounding each site.
Results/Conclusions
We found 91 bird carcasses (N = 39 species) resulting from window collisions. The most supported model explaining the number of carcasses included window area, local vegetation, and broad-scale development. Buildings with the most carcasses had high window area. Number of carcasses had weak positive associations with green space in the immediate vicinity of buildings and developed land in the broader landscape.
Project methods and protocols performed well, but will be modified for the fall 2014 field season to include, for example, a more effective carcass survey protocol that improves detection of carcasses by field workers. The project would benefit from additional collaborators in the western United States and throughout Mexico and Canada.
Results from the 2013 pilot field season are preliminary, but suggest strong potential to assess the drivers of BWCs at the continent scale in future field seasons. This information is crucial for predicting local and regional mortality, which would focus future conservation efforts aimed at reducing collision-related impacts. To date, the project has provided inquiry-based educational opportunities for 189 undergraduate students and 21 faculty/professional researchers at collaborator sites. Students at three campuses participated in professional development activities, e.g., poster presentations at local scientific meetings, which stemmed from their research on bird-window collisions.