���Ĵ�ý

Hanyong Cai had barely begun his first year at Duke Kunshan University (DKU) in China when he witnessed a grim scene on campus: A spotted dove had collided with a glass pane and was lying on the ground. The bird, distinctive for speckled plumage along its nape, had died from the impact. “We also observed a feather stuck to the window,” Cai says. 

The encounter dug at Cai’s curiosity. Why did this bird crash?

Now, two years later, the DKU junior volunteers with a nationwide research program, established by DKU conservation biologist Binbin Li, to better understand the problem of bird collisions across the country’s varied urban landscapes. 

The work builds on a ���Ĵ�ý initiative, led by Nicki Cagle, a senior lecturer in ecology and environmental education at the Nicholas School of the Environment. In spring 2014, Cagle and former Nicholas School Ph.D. students Natalia Ocampo-Peñuela and Scott Winton launched the to monitor buildings on the Durham campus for birds that had flown into glass panes.

“The reason this is happening, at least physiologically, is that the birds don’t perceive the glass as a physical barrier,” explains Cagle, who is also the director of natural and cultural resources interpretation at Duke Forest. Birds either think they can whiz through the glass, or they mistake reflections of habitat for the real thing and veer in, only to knock themselves out. 

In fact, window collisions are a leading human-caused source of bird deaths, along with habitat loss and cats. Researchers estimate that in the United States alone, up to 1 billion birds die each year by crashing into glass.

This avian attrition can have far-reaching ecological implications.

“Birds are a crucial link in the chain of life,” Cagle says. They prey upon insects and mice that attack agricultural crops, for instance, and provide sustenance to other organisms. 

But to lose a bird is to also lose beauty, Cagle adds. “You see a bird flying, you can’t help but imagine what it’s like to be wild and free like that. They bring us joy and hope and inspiration.”

Frequent Flyers

Bird collisions with windows can happen year-round at buildings of all sizes. However, in the U.S., the rate increases during fall and spring migration season, when birds are moving through urban areas in large numbers, according to Cagle.

Through the Bird-Window Collision Project, Cagle and Duke students work to identify major collision hotspots at campus buildings. They rely on reports of dead or injured birds from passersby, as well as on data from their own monitoring.

Monitoring involves walking around the foundation of designated buildings, typically during morning hours in fall and spring. Participants record when and where they find fallen birds, as well as the species if possible. 

Using that data, members of the Bird-Window Collision Project team can determine how crash patterns vary across buildings over time, and which species appear most vulnerable to campus collisions.

Most avian fatalities at Duke involve cedar waxwings — sleek, grayish-brown songbirds with Zorro-mask markings. Ruby-throated hummingbirds, which flash iridescent green in the sun, are also common victims.

Upon finding deceased birds, building monitors carefully collect them for later preservation.

“We donate all of our carcasses to the North Carolina Museum of Natural Sciences in Raleigh, so [they’re] not going to waste,” says Madison Chudzik, a Ph.D. student in the biology department who has been leading the monitoring effort for the last couple years on a volunteer basis. “They put them in the collections for researchers or historians to use.”

Using their monitoring data, the team can inform practical changes on campus that make Duke safer for birds.

For example, back in 2015 the team Duke’s Fitzpatrick Center for Interdisciplinary Engineering, Medicine and Applied Sciences, or FCIEMAS, as a major threat to birds. The main offender was a windowed bridge connecting two parts of the center. (See banner image.) 

The team presented their findings, which they later , to university leadership. Within months, Duke Facilities Management had the FCIEMAS with a bird-friendly veneer consisting of small, evenly placed dots that birds can detect. In effect, birds read this pattern as a “DO NOT ENTER” sign. 

The update appeared to reduce observed collisions, the monitoring team , though the data set was too small to be considered statistically significant. 

According to Cagle, Duke’s attention to bird safety represents a larger national trend. 

“Since I've been working on this project, I think awareness about bird-widow collisions has increased dramatically,” she says. “Architects have become more willing to use and experiment with bird-friendly glass products in particular, and even just the average homeowner is now aware of the problem and wants to do something.”

Smoother Sailing

At the time of the FCIEMAS makeover, Binbin Li was a Ph.D. student at the Nicholas School. The bird collision research got her wondering whether windows pose similar problems in China, where she’s from. 

Li found her answer a couple years later after joining the faculty at DKU, where she is an associate professor of environmental sciences: “I got several calls from my colleagues saying, ‘Oh, we got birds colliding with the windows. It’s dead or it’s injured,’ and they asked for my help,” recalls Li, who has a secondary appointment at the Nicholas School. 

“Suddenly I realized, okay, the bird window collision actually is not just a unique problem in North America. It can happen in China as well.”

In fall 2018, Li launched DKU’s first campus-wide bird monitoring survey, adapting Nicholas School methodology. The participants, including undergraduate students, found that nearly half of collisions occurred at buildings certified as environmentally friendly. Spotted doves were among the commonly found species.

The results prompted DKU to reduce the amount of glass used in  and to incorporate fritted glass — glass etched with patterns that birds can discern — and other bird-friendly design, according to Li. 

Encouraged by DKU’s response to the research, Li expanded her study area. Working with several nonprofit groups, she initiated the first national survey of bird collisions in China in 2021. The following year, Li established the Anti-Bird Collision Action Alliance, a nationwide network designed to improve bird safety through community-engaged scholarship, or citizen science. 

Members of the alliance train volunteers to conduct periodic bird collision monitoring in their vicinity, then compile and crunch the data for annual public reports. Results from the alliance’s  indicated a “very clear correlation between bird collisions with the glass coverage in buildings,” Li says. 

The surveys have attracted wide participation. More than 3,500 volunteers from more than 200 cities and 32 provinces joined the 2026 effort, according to Li. 

“We think the citizen science approach is very effective in changing how people think about why we need conservation, and how development in human society may threaten other species,” Li says. “Especially when the volunteers see a dead bird and hold it in their hands, they feel the weight of it, and that changes a lot of them.”

Additionally, participating in bird collision monitoring “gives a lot of people, especially in urban areas, a very easy way to understand and participate in conservation,” Li says. “They collect data to provide evidence, and then they have the evidence to support their interventions and actions. And after that, we encourage the volunteers to continue to do monitoring so they can evaluate the effectiveness of the interventions.” 

Indeed, the alliance’s research, along with shifts in public perception about the issue, have inspired some Chinese cities, including Shenzhen and Shanghai, to incorporate bird-friendly design into urban planning, according to Li. 

New Heights

Growing interest in bird safety across China also inspired Li to develop a complementary national study to address a question that on-the-ground surveys can’t: When it comes to bird collisions, how do we account for fallen birds that escape public notice?

For example, when birds hit a window, they can drop into vegetation, out of public view. Small mammals or other birds might steal away with a carcass before detection. Additionally, cleaning crews may sweep up fallen birds during early-morning shifts. 

Li’s study aims to identify who — or what — removes birds before they’re detected, and how long it takes for those bodies to vanish. 

Determining the rate of bird disappearance at different locations can inform more robust estimates of avian attrition from window strikes. Decision-makers and conservationists can then use those estimates to push for policy changes and design approaches that benefit birds.

“It is really basic research for us to do, but it is also important,” says Hanyong Cai, the DKU junior who assists with bird-window collision research.  

In May, Cai and fellow DKU junior Zeheng Li shared the research concept with Cagle, who subsequently invited the students to help implement similar methodology at Duke. 

This past summer, the team set about installing unobtrusive camera traps, used to record wildlife, at seven locations across the Durham campus. At each site, the team placed a deceased bird that they had procured through approved, standardized protocol. They repeated those steps in Duke Forest as a point of comparison.

“In the forest and on campus, the environment is very different. So, by doing the comparison, we can study the relation between the environmental factors and the removal rate,” of the birds, Zeheng Li explains.

Although the team is still collecting data, early results suggest slight differences between bird snatchers at the two locations. In camera trap data from campus, the team has identified species “that we are describing as somewhat more urbanized,” Cai says. “These animals include cats and also some American crow,” Campus footage has also revealed gray foxes, according to Cai.  

Meanwhile, raccoons and opossums have unwittingly mugged for the cameras in both locales. 

Ultimately, the Duke and DKU teams plan to determine how their data stacks up.

“We thought, wow, wouldn’t it be cool to see how the two campuses compare and if we're seeing similar trends in China and the United States?” Cagle says. “They have a whole different slew of mesopredators in urban areas” in China, she says, referring to midsize animals in the middle of the food chain, “so it’s just a completely different system.” 

The Thing With Feathers

Several months ago, Nicholas School junior Madison Hobbs was walking across the Durham campus when she spied a small bird lying eerily still on the ground — the victim of a fatal window collision, she suspected. She and a friend buried the carcass amid native North Carolina vegetation near Grainger Hall, part of the Nicholas School.

The encounter struck a chord with Hobbs, who remembers feeling helpless as a kid when birds occasionally crashed into her family home. “I always just thought it was kind of inevitable that birds run into windows,” says Hobbs, who plans to volunteer as a building monitor with Cagle’s team this spring.  

“Doing this work really brings up mixed feelings,” Cagle says. There is grief but also hope. 

“We know how to fix this problem, and it really takes just that interdisciplinary approach between ecologists and architects and administrators to come together to make those solutions a reality.”

If you find a bird on Duke’s campus that may have collided with a window, whether dead or alive, visit the  to learn how to report it.