Flying high: The ethics of monitoring birds with drones

Unmanned aerial vehicles, more commonly known as “drones,” are quickly become useful tools for ecologists and biologists alike. They’re relatively cheap, ranging from a few hundred to a few thousand US dollars, and are fairly easy to learn how to operate. They can allow researchers access to landscapes that are otherwise inaccessible and to approach animals that are otherwise unapproachable.

Regulatory agencies are increasingly legislating the use of drones, but as Universite de Montpellier researcher Elisabeth Vas and colleagues argued recently in the journal Biology Letters, “no ethical guidelines exist [yet] with respect to their potential impacts on animal welfare.” They reason that the lack of guidelines regarding the use of drones to study animals is perhaps because there’s a lack of research regarding how animals react when drones approach.

As a first step towards addressing that issue, Vas’s team assessed the reactions of three species of waterbirds to a small quad-copter drone, one of the most affordable and widely used varieties. They focused on waterbirds because drones are already being used extensively to survey wetlands and coastal ecosystems. They decided to test mallard ducks in a semi-captive setting; these were birds living at the Zoo du Lunaret in Montpellier, France, but were capable of flying in and out of the zoo at their whim. In addition, they targeted populations of wild flamingos and common greenshanks. They limited their flights to non-breeding seasons, and only approached birds that were resting or feeding.

The specific drone they used was fairly quiet and was classified as non-impacting, so the researchers focused on the visual characteristics of the drone (color) as well as the flight trajectory. Drones were launched either 50 meters or 100 meters away from the target birds, and were colored either white, black, or blue. From the launch spot, the drones ascended vertically to thirty meters, and then approached the birds at one of four speeds (2, 4, 6, or 8 meters per second) and descended from one of four angles relative to the ground (20, 30, 60, or 90 degrees). Put together, this resulted in 36 different conditions for each of the three different species.

While one researcher operated the drone, a second monitored the birds’ reactions through binoculars. Reactions were classified as one of three types: no reaction, brief head or tail movements followed by moving away from the drone, or flying away. The approaches ended either when the bird reacted or when the drone was just four meters away from the bird. When at least one bird from within a group retreated from the drone – whether by walking, swimming, or flying – the group was considered “stressed” by the drone’s approach.

They found that in 80% of the 204 test trials, drones could approach waterbirds to a distance of four meters without any visible behavioral response. Approach speed, drone color, and number repeated approaches did not have any bearing on the birds’ responses. Instead, the main impact on the birds was the drones’ approach angle. Drones approaching at a more oblique angle were typically least stressful to the birds. As the angle of approach moved closer to vertical (90 degrees), the birds reacted more strongly. The researchers suspect this may because the waterbirds’ predators tend to approach from directly above, and a drone descended from directly above may have simulated a predator.

To minimize stress on the birds, they recommend launching drones no closer than 100 meters from the birds, approaching them at an angle rather than from directly above, and adjusting the approach distance according to species. Some species are more tolerant of novel stimuli than others, after all.

They recommend that future studies address the question of noise and drone size, two variables that the researchers in this case did not include. They also note that the study focused only on three types of waterbirds that feed mainly on plants and small invertebrates. Additional studies must be conducted with a wider range of taxa, with a wider range of feeding and social ecologies. A quick YouTube search, for example, reveals that raptors tend to attack drones, and the researchers surmise that the same may be true for corvids and larids. Finally, they recommend that future studies address the physiological responses that birds have to drones. Just because the researchers observed no overt behavioral responses to drones doesn’t mean that they didn’t have increased heart rates or respiration, for example.

Vas concludes that “drones may be used in ornithology for a wide range of population censuses, measurements of biotic and abiotic variables, and recordings of bird behaviour,” but cautions that the study is but a “first step towards a code of best practices in the use of drones for ecological research.” – Jason G. Goldman | 18 February 2015

Source: Vas, E., Lescroel, A., Duriez, O., Boguszewski, G., & Gremillet, D. (2015). Approaching birds with drones: first experiments and ethical guidelines. Biology Letters, 11(2), 20140754–20140754. doi:10.1098/rsbl.2014.0754

Header image: Photosebia /