COS 113-7
Relationships between building façade characteristics and bird-window collision mortality during fall migration in Stillwater, Oklahoma
Recent estimates indicate that hundreds of millions of birds die annually in North America from collisions with building windows. Some bird species appear disproportionately prone to collisions, and building and landscape features, such as window area and proximity to vegetation, may escalate this risk. Previous studies have assessed building- or neighborhood-level correlates of avian collision rates. However, we hypothesize that smaller-scale characteristics of individual building facades are also correlated with bird-window collision rates.
We surveyed for window-killed birds in Stillwater, OK during fall migration in 2013-2014. We conducted surveys for 21 consecutive days at six buildings in 2013 and ten buildings in 2014. We continued surveys for an additional 25 days at one building and 30 days at another building in 2014. The façade-scale factors we evaluated include (1) proportion of the façade covered by glass surfaces, (2) façade type as defined by length, shape, and position relative to adjacent façades, (3) façade height as number of floors, and (4) amount of vegetation at the façade base.
Results/Conclusions
We found 29 carcasses of 14 species and 7 additional carcasses that we could not identify to species. Twenty-seven carcasses were for species (n=11) considered more vulnerable to collisions than average at a national scale.
We found 28 (81%) carcasses at facades defined as alcoves or porticos. These façade types ostensibly produce a tunnel-like effect that may funnel birds towards windows or mask the presence of windows, especially under low ambient light levels. We also found 28 (81%) carcasses at façades that lacked vegetation at the base and instead had artificial surfaces (e.g., metal grates, concrete walkways, brick patios). Thus, vegetation in the immediate vicinity of building facades may be less important than façade shape for causing fatal collisions. Finally, most collisions (n=27, 77%) occurred at façades where glass covered ≥80% of the surface. This result is consistent with earlier findings documenting a relationship between window area and bird collision rates at the building scale. Our results suggest that a more complete understanding of the drivers of bird-building collisions—and therefore, development of approaches to effectively reduce collisions—requires the study of a hierarchy of factors that includes fine-scale features of individual building facades.