Patterns of bird-window collisions inform mitigation on a university campus.
ABSTRACT: Bird-window collisions cause an estimated one billion bird deaths annually in the United States. Building characteristics and surrounding habitat affect collision frequency. Given the importance of collisions as an anthropogenic threat to birds, mitigation is essential. Patterned glass and UV-reflective films have been proven to prevent collisions. At Duke University's West campus in Durham, North Carolina, we set out to identify the buildings and building characteristics associated with the highest frequencies of collisions in order to propose a mitigation strategy. We surveyed six buildings, stratified by size, and measured architectural characteristics and surrounding area variables. During 21 consecutive days in spring and fall 2014, and spring 2015, we conducted carcass surveys to document collisions. In addition, we also collected ad hoc collision data year-round and recorded the data using the app iNaturalist. Consistent with previous studies, we found a positive relationship between glass area and collisions. Fitzpatrick, the building with the most window area, caused the most collisions. Schwartz and the Perk, the two small buildings with small window areas, had the lowest collision frequencies. Penn, the only building with bird deterrent pattern, caused just two collisions, despite being almost completely made out of glass. Unlike many research projects, our data collection led to mitigation action. A resolution supported by the student government, including news stories in the local media, resulted in the application of a bird deterrent film to the building with the most collisions: Fitzpatrick. We present our collision data and mitigation result to inspire other researchers and organizations to prevent bird-window collisions.
Project description:Millions of birds die in bird-window collisions in the United States each year. In specialized test settings, researchers have developed methods to alter window designs to mitigate collisions. However, few published studies provide pretest and posttest evaluations of mitigation treatment areas and untreated control areas on existing buildings. We initially monitored bird-window collisions at a single building on the University of Utah campus in Salt Lake City, Utah, USA, during winter 1 (November 9, 2017-January 2, 2018). We found 15 bird-window collisions, most under a portion of the building with a mirrored façade. To test a mitigation treatment, we installed Feather Friendly® bird deterrent film on part of the mirrored façade after winter 1. The unmitigated areas of the same building served as a control area. We continued monitoring during the following winter 2 (November 15, 2018-January 12, 2019). The treated area collisions declined from seven before mitigation to two after mitigation, a 71% reduction. The control area had eight collisions at both times. Results of a generalized estimating equation yielded a significant area by season interaction effect (p = 0.03) and fewer collisions in the mitigated area than the control area at winter 2 (p = 0.03), supporting efficacy of the mitigation. In winter 2 we also expanded monitoring to eight total buildings to evaluate the risks of mirrored windows and proximity to fruiting pear trees (Prunus calleryana) and the benefits of bird-friendly glass. Bird-friendly glass, found on two buildings, included windows with permanent fritted dots or embedded ultraviolet patterns. We counted 22 collisions across the eight buildings. Mirrored windows and proximity to fruiting pear trees related to higher odds of bird-window collisions, based on separate generalized estimating equations. The best fit model included mirrored windows and pear trees. The two buildings with bird-friendly glass had only one collision, suggesting that these designs deter collisions, although the difference was not statistically significant. To publicize the study and to receive reports of additional bird collisions or fatalities on campus, we created a citizen science project on iNaturalist and engaged in additional outreach efforts that yielded 22 ad hoc reports. Many previous studies have documented Cedar Waxwing (Bombycilla cedrorum) collisions, but at relatively low numbers. Cedar Waxwings accounted for 31 of 34 identifiable collisions from the monitoring study and 4 of 21 identifiable collisions or fatalities from ad hoc reports.
Project description:Bird-building collisions are the largest source of avian collision mortality in North America. Despite a growing literature on bird-building collisions, little research has been conducted in downtown areas of major cities, and no studies have included stadiums, which can be extremely large, often have extensive glass surfaces and lighting, and therefore may cause many bird collisions. Further, few studies have assessed the role of nighttime lighting in increasing collisions, despite the often-cited importance of this factor, or considered collision correlates for different seasons and bird species. We conducted bird collision monitoring over four migration seasons at 21 buildings, including a large multi-use stadium, in downtown Minneapolis, Minnesota, USA. We used a rigorous survey methodology to quantify among-building variation in collisions and assess how building features (e.g., glass area, lighting, vegetation) influence total collision fatalities, fatalities for separate seasons and species, and numbers of species colliding. Four buildings, including the stadium, caused a high proportion of all collisions and drove positive effects of glass area and amount of surrounding vegetation on most collision variables. Excluding these buildings from analyses resulted in slightly different collision predictors, suggesting that factors leading some buildings to cause high numbers of collisions are not the exact same factors causing variation among more typical buildings. We also found variation in collision correlates between spring and fall migration and among bird species, that factors influencing collision fatalities also influence numbers of species colliding, and that the proportion, and potentially area, of glass lighted at night are associated with collisions. Thus, reducing bird collisions at large buildings, including stadiums, should be achievable by reducing glass area (or treating existing glass), reducing light emission at night, and prioritizing mitigation efforts for glass surfaces near vegetated areas and/or avoiding use of vegetation near glass.
Project description:Background:To reduce bird fatalities from millions of window collisions each year in North America, it is important to understand how design and landscape elements relate to collision risk. The current study extends prior research that found that buildings near ornamental pear trees (Prunus calleryana) and buildings with mirrored windows significantly increased odds of collisions among eight buildings on the University of Utah campus in winter. The previous study found bird-friendly glass was not related to collision risk, although only one fatality occurred at two buildings with ORNILUX® ultraviolet (UV) or fritted windows. We reasoned that extending data collection to include fall might provide a better test of efficacy. We tested the following three hypotheses: (1) Buildings with mirrored windows would experience more collisions, replicating the original study; (2) the addition of fall migration data would reveal fewer collisions at the buildings with bird-friendly windows; (3) the danger of pear tree proximity would be heightened in winter, when fruit is ripe enough to appeal to frugivores, especially the Cedar Waxwings (Bombycilla cedrorum) that frequent these trees. Methods:Trained observers monitored buildings three times per week in Fall (September 12 to October 27, 2019) and Winter (October 29, 2019 to January 24, 2020). Collisions were photographed and documented in the iNaturalist University of Utah Bird Window Collision Project. Results:There were 39 total collisions, from 0 to 14 per building.Using generalized estimating equations, buildings near pear trees had 3.33-fold increased odds, mirrored windows had 5.92-fold increased odds, and bird-friendly windows had an 84% lower odds (Odds ratio = 0.16) of bird window collisions when analyzed separately; all were statistically significant (p < 0.01). A test of all possible combinations of risk and protective factors revealed that the best fit model included pear trees (odds = 2.31) and mirrored windows (odds = 2.33). A separate analysis tested the pear tree by season interaction model; it yielded the deadliest combination, with 40-fold increased odds for buildings near pear trees in winter season. Discussion:This research provides the first peer-reviewed evidence found for the efficacy of bird-friendly fritted windows and ORNILUX ® UV windows in buildings. In addition, it replicated a study that established the dangers of mirrored windows and fruiting pear trees near buildings. These risks were especially dangerous to Cedar Waxwings, who constituted 62.2% of the identifiable window collision victims. This research highlights how building risks depend on window design, landscape choices, species, and season. If replicated, analyses of risk factors can help identify buildings that require mitigation to make existing windows less deadly. Results also support the installation of bird-friendly glass in new or renovated buildings to reduce fatalities.
Project description:Collisions with glass are a serious threat to avian life and are estimated to kill hundreds of millions of birds per year in the United States. We monitored 22 buildings at the Virginia Tech Corporate Research Center (VTCRC) in Blacksburg, Virginia, for collision fatalities from October 2013 through May 2015 and explored possible effects exerted by glass area and surrounding land cover on avian mortality. We documented 240 individuals representing 55 identifiable species that died due to collisions with windows at the VTCRC. The relative risk of fatal collisions at all buildings over the study period were estimated using a Bayesian hierarchical zero-inflated Poisson model adjusting for percentage of tree and lawn cover within 50 m of buildings, as well as for glass area. We found significant relationships between fatalities and surrounding lawn area (relative risk: 0.96, 95% credible interval: 0.93, 0.98) as well as glass area on buildings (RR: 1.30, 95% CI [1.05-1.65]). The model also found a moderately significant relationship between fatal collisions and the percent land cover of ornamental trees surrounding buildings (RR = 1.02, 95% CI [1.00-1.05]). Every building surveyed had at least one recorded collision death. Our findings indicate that birds collide with VTCRC windows during the summer breeding season in addition to spring and fall migration. The Ruby-throated Hummingbird (Archilochus colubris) was the most common window collision species and accounted for 10% of deaths. Though research has identified various correlates with fatal bird-window collisions, such studies rarely culminate in mitigation. We hope our study brings attention, and ultimately action, to address this significant threat to birds at the VTCRC and elsewhere.
Project description:Bird-window collisions are a major and poorly-understood generator of bird mortality. In North America, studies of this topic tend to be focused east of the Mississippi River, resulting in a paucity of data from the Western flyways. Additionally, few available data can critically evaluate factors such as time of day, sex and age bias, and effect of window pane size on collisions. We collected and analyzed 5 years of window strike data from a 3-story building in a large urban park in San Francisco, California. To evaluate our window collision data in context, we collected weekly data on local bird abundance in the adjacent parkland. Our study asks two overarching questions: first-what aspects of a bird's biology might make them more likely to fatally strike windows; and second, what characteristics of a building's design contribute to bird-window collisions. We used a dataset of 308 fatal bird strikes to examine the relationships of strikes relative to age, sex, time of day, time of year, and a variety of other factors, including mitigation efforts. We found that actively migrating birds may not be major contributors to collisions as has been found elsewhere. We found that males and young birds were both significantly overrepresented relative to their abundance in the habitat surrounding the building. We also analyzed the effect of external window shades as mitigation, finding that an overall reduction in large panes, whether covered or in some way broken up with mullions, effectively reduced window collisions. We conclude that effective mitigation or design will be required in all seasons, but that breeding seasons and migratory seasons are most critical, especially for low-rise buildings and other sites away from urban migrant traps. Finally, strikes occur throughout the day, but mitigation may be most effective in the morning and midday.
Project description:Bird-window collisions are one of the main causes of avian mortality worldwide, with estimations reaching up to almost one billion of dead individuals annually due to this cause in Canada and the USA alone. Although this is a growing conservation problem, most of the studies come from North America, evidencing the lack of knowledge and concern in countries with high biodiversity and growing population development. Our objectives were: (1) to estimate the current situation of bird-window collisions in Argentina, a country with around 10% of the world's avian biodiversity, and, (2) to identify drivers of bird-window collisions at a national and local scale, focusing on a city surrounded by a protected area. We used a citizen science project called "Bird-Window Collisions in Argentina" that consisted of an online survey that collected data on collision metrics and risk factors. We found that more than half of participants reported at least one collision during the last year, suggesting this issue is common and widespread. In addition, our data show that the number of windows and the presence of vegetation reflected in windows are factors that strongly influence the risk of collision at national scale. On the other hand, the environment surrounding buildings affects the rate of bird-window collisions at local scale, being greater in buildings surrounded by tall vegetation than in buildings surrounded by a greater proportion of urbanization (human-made structures). We call for attention on a topic that has been poorly evaluated in South America. We also encourage future scientific studies to evaluate additional risk factors and mitigation strategies accordingly, to provide a better understanding of bird-window collisions particularly in a highly biodiverse region as South America.
Project description:Bird collisions with buildings are the second largest anthropogenic source of direct mortality for birds (365-988 million birds killed annually in the United States). Recent research suggests that this mortality occurs disproportionately across species. However, previous work had relied on regional and annual measures of relative species abundance. Our research identifies which species experience higher or lower collision rates than expected from local abundances using two sets of citizen science data: Minnesota Project BirdSafe and the Mississippi River Twin Cities Important Bird Area Landbird Monitoring Program. Our analysis used a measure of relative species abundance that spatially overlaps the area monitored for building collisions and was measured weekly, allowing for a temporally and spatially more specific analysis than most previous analyses. Abundance and collision data were used to model phylogenetic and behavioral traits associated with increased collision risk. Behavioral traits included diurnal/nocturnal migration timing, length of migration, and foraging strategies. Our analysis shows that birds that predominately migrate during the day have a decreased risk of building collisions despite peak collision numbers occurring during early morning; this result suggests that more nuanced behavioral or physiological differences between diurnal and nocturnal migrants could contribute to bird-building collision risk. Additionally, for many species, local abundance is the predominant determining factor for collision risk. However, for ~20% of species studied, the family, genus, and/or species of a bird may affect the collision risk.
Project description:Collisions with windows are an important human-related threat to birds in urban landscapes. However, the proximate drivers of collisions are not well understood, and no study has examined spatial variation in mortality in an urban setting. We hypothesized that the number of fatalities at buildings varies with window area and habitat features that influence avian community structure. In 2010 we documented bird-window collisions (BWCs) and characterized avian community structure at 20 buildings in an urban landscape in northwestern Illinois, USA. For each building and season, we conducted 21 daily surveys for carcasses and nine point count surveys to estimate relative abundance, richness, and diversity. Our sampling design was informed by experimentally estimated carcass persistence times and detection probabilities. We used linear and generalized linear mixed models to evaluate how habitat features influenced community structure and how mortality was affected by window area and factors that correlated with community structure. The most-supported model was consistent for all community indices and included effects of season, development, and distance to vegetated lots. BWCs were related positively to window area and negatively to development. We documented mortalities for 16/72 (22%) species (34 total carcasses) recorded at buildings, and BWCs were greater for juveniles than adults. Based on the most-supported model of BWCs, the median number of annual predicted fatalities at study buildings was 3 (range?=?0-52). These results suggest that patchily distributed environmental resources and levels of window area in buildings create spatial variation in BWCs within and among urban areas. Current mortality estimates place little emphasis on spatial variation, which precludes a fundamental understanding of the issue. To focus conservation efforts, we illustrate how knowledge of the structural and environmental factors that influence bird-window collisions can be used to predict fatalities in the broader landscape.
Project description:Perhaps a billion birds die annually from colliding with residential and commercial windows. Therefore, there is a societal need to develop technologies that reduce window collisions by birds. Many current window films that are applied to the external surface of windows have human-visible patterns that are not esthetically preferable. BirdShades have developed a short wavelength (ultraviolet) reflective film that appears as a slight tint to the human eye but should be highly visible to many bird species that see in this spectral range. We performed flight tunnel tests of whether the BirdShades external window film reduced the likelihood that two species of song bird (zebra finch, Taeniopygia guttata and brown-headed cowbird, Molothrus ater) collide with windows during daylight. We paid particular attention to simulate the lighting conditions that birds will experience while flying during the day. Our results indicate a 75-90% reduction in the likelihood of collision with BirdShades-treated compared with control windows, in forced choice trials. In more ecologically relevant comparison between trials where all windows were either treated or control windows, the estimated reduction in probability of collision was 30-50%. Further, both bird species slow their flight by approximately 25% when approaching windows treated with the BirdShades film, thereby reducing the force of collisions if they were to happen. Therefore, we conclude that the BirdShades external window film will be effective in reducing the risk of and damage caused to populations and property by birds' collision with windows. As this ultraviolet-reflective film has no human-visible patterning to it, the product might be an esthetically more acceptable low cost solution to reducing bird-window collisions. Further, we call for testing of other mitigation technologies in lighting and ecological conditions that are more similar to what birds experience in real human-built environments and make suggestions for testing standards to assess collision-reducing technologies.
Project description:Bird-window collisions are the second leading cause of human-related avian mortality for songbirds in Canada. Our ability to accurately estimate the number of fatalities caused by window collisions is affected by several biases, including the removal of carcasses by scavengers prior to those carcasses being detected during surveys. We investigated the role of scavenger behavior in modifying perceived carcass removal rate while describing habitat-specific differences for the scavengers present in a relatively scavenger-depauperate island ecosystem. We used motion activated cameras to monitor the fate of hatchling chicken carcasses placed at building (under both windows and windowless walls) and forest (open and closed canopy) sites in western Newfoundland, Canada. We recorded the identity of scavengers, timing of events, and frequency of repeat scavenging at sites. Using 2 treatments, we also assessed how scavenging varied with 2 levels of carcass availability (daily versus every third day). Scavenger activities differed substantially between forest and building sites. Only common ravens (Corvus corax) removed carcasses at building sites, with 25 of 26 removals occurring under windows. Burying beetles (Nicrophorus spp.) dominated scavenging at forest sites (14 of 18 removals), completely removing carcasses from sight in under 24 hours. Availability had no effect on removal rate. These findings suggest that ravens look for carcasses near building windows, where bird-window collision fatalities create predictable food sources, but that this learning preceded the study. Such behavior resulted in highly heterogeneous scavenging rates at fine spatial scales indicating the need for careful consideration of carcass and camera placement when monitoring scavenger activity. Our observations of burying beetle activity indicate that future studies investigating bird collision mortality near forested habitats and with infrequent surveys, should consider local invertebrate community composition during survey design. The high incidence of invertebrate scavenging may compensate for the reduced vertebrate scavenger community of insular Newfoundland.