Testing & Research

We test birds’ reactions to patterned glass using two glass testing tunnels — the only ones in the U.S. ABC started testing glass in 2010 at Carnegie Museum of Natural History’s Powdermill Avian Research Center (PARC), a permanent bird-banding station in Rector, Pennsylvania that bands more than 11,000 birds of more than 110 species in a typical year. In 2022, we expanded our glass testing capacity, joining forces with Washington College to establish a second glass testing tunnel at the college’s Foreman’s Branch Bird Observatory in Chestertown, Maryland.

On the outside, the testing tunnels — 24-foot-long, narrow constructions of plywood — look unusual, but the tunnels are scientific tools. By watching how birds fly through them, we can identify which glass materials birds can see, and which ones they can’t.

Birds are safely caught in mist nets (fine mesh netting used to gently capture birds) and retrieved by trained professionals. All personnel handling birds for the glass testing tunnels are experienced bird handlers under the supervision of a Master Bander operating under a federal permit from the U.S. Geological Survey’s Bird Banding Lab.

Birds mist-netted on trails near the banding station are removed from the nets by banding technicians, placed in cloth bags, and taken to the station for banding and measuring. After banding, birds chosen to be tested in the tunnel are put back into a cloth bag and brought to the tunnel.

When a bird is brought to the tunnel for testing, the tunnel-testing technician removes it from the bag and evaluates it. Any bird that appears stressed is safely released without being put through a tunnel trial. For birds in good condition, the band number is recorded and the bird is released into the tunnel by hand through a sleeve set into the tunnel’s end panel. Any bird that does not fly after 30 seconds in the tunnel is withdrawn and released.

Birds are placed in a 24-foot-long tunnel, constructed of plywood and particle board over a steel frame. The tunnel is open and lighted at one end with two panes of glass mounted 18 inches from the opening. This presents birds with a choice for how they will exit: via an “invisible” panel of plain glass that serves as a control or a panel of the patterned material being tested. A taut mist net stops the birds before they reach the glass, and each bird is released back into the wild after a single test flight.

Birds are observed and recorded as they fly through the tunnel. A bird’s flight type, as well as where it goes — towards the “invisible” control glass, the treated glass, the walls of the tunnel — gives us insight into how each bird perceives its surroundings.

Consistency matters, and the tunnels have some special features to ensure the efficacy of the trials. Each tunnel is mounted on a pivot and is rotated every five minutes to keep a constant orientation, with the sun always directly behind the operator. Mirrors at the sides of the tunnel reflect ambient light onto the front surfaces, the “bird approach side” of the glass. Greatly reduced natural light falls on the back surface of the glass, mimicking the lighting differential between indoors and outdoors in buildings.

Test materials (the treated glass) are presented in random order and in equal frequency on the left and right sides. At regular intervals, proofing trials using two clear panes or no glass are run as a control. During these trials, equal numbers of flights to left and right indicate that the tunnel itself is not influencing the choice made by the birds.

ABC’s rating system assigns a Threat Factor^TM score that indicates which materials birds are most likely to avoid during our tunnel testing process. We score at least 80 trials per tested material to account for environmental variation and the fact that we are running trials with a variety of bird species. The observations of birds in flight are analyzed, and from that data, we can assign a Material Threat Factor (TF) based on how many birds fly toward the glass. If birds avoid it, that means they can see it. ABC defines the Material Threat Factor for a pattern or material as the percentage of birds tested that flew toward that material. So, if 20 of 80 birds fly toward the tested glass, 20/80=25% and the TF=25.

The Material Threat Factor is a key component of our Products and Solutions Database. It’s a reflection of how well birds recognized the glass during trials. The lower the TF, the more effective the glass will be in reducing collisions. ABC defines “bird-friendly” materials as having a threat factor ≤ 30, corresponding to a conservative estimated reduction of collisions of at least 50% under real-world conditions.

To request glass testing and a Threat Factor^TM or to learn more, please contact Christine Sheppard, Senior Director, ABC Glass Collisions Program.