Bird collisions are a well-known problem in onshore wind farms. So far, less research has been done into how often seabirds fly into offshore wind turbines and which species could be particularly endangered. Now, the results of a two-year study at Vattenfall’s European Offshore Wind Deployment Center in Aberdeen Bay suggest that seabirds can fly around the obstacles well, at least during the day and when the weather is not too bad.
From 2020 to 2021, the DHI Group from Denmark commissioned by Vattenfall – a software development and engineering consultancy company specializing in water – observed the flight behavior of gannets and various species of gulls using 3D tracking technology from April to October. This included the breeding season and some time before and after. As the DHI researchers write in their report, their monitoring system did not record any collisions during the study periods.
In order to avoid the wind turbines, the birds initially flew more slowly and then mostly avoided horizontally. Kittiwakes braked 200 meters before the rotor tips and then swerved 140 to 160 away, mostly to the side. Herring gulls and unidentified great gulls slowed at 120 meters and changed direction at 90 to 110 meters. Gannets reacted most quickly, slowing down 100 meters in front of the wind turbines and only turning away 40 to 50 meters in front of them.
Wind turbines in the way of seabirds?
“We expected the collision risk to be low, but for these seabird species it actually seems very low. What was surprising, however, was that there were no collisions at all,” says Henrik Skov, who led the project for DHI Group Team expected to see more collisions with the advanced equipment.
The birds are most active from April to October, when thousands of them forage around the edge of the wind turbine and occasionally between the wind turbines during and after the breeding season, Skov says.
Unlike on land, however, collisions at sea cannot be reconstructed by counting dead bird bodies around the wind turbines. In order to observe the flight behavior of the birds, the DHI researchers placed a camera with a 360-degree all-round view on two of the eleven wind turbines – in the middle and on the edge of the wind farm – and also a radar device on the outer system. The system first detected the approaching animals by radar and automatically transferred them to the cameras for further tracking.
Create flight patterns of seabirds
Software supplemented the “flat” video images, so to speak, with the depth detection of the radar and triangulated altitude information to form three-dimensional flight patterns. The radar data also helped to determine the speed and direction of the birds. The system could track animals across both cameras, which was supported by an artificial intelligence system in the second year and increased the tracking accuracy.
The overall view was restricted by a little more than 90 degrees in the lower right corner of the wind farm. Here the turbine mast on which the radar was mounted covered the scanning field. In addition, the radar system was “blind” between each turbine’s blades, Skov said, because the towers and blades created too much “static interference.” It’s a kind of unwanted radar echo. Here it was checked purely by camera whether there were any collisions.
Video sequences with examples of micro avoidance behavior. The top sequence shows a Lesser Black-backed Gull flying along the plane of the rotor. The sequence below shows a Lesser Black-backed Gull flying vertically over a stationary rotor without any adjustments.
(Picture: DHI Group)
Nevertheless, “this is a very exciting study because impressive technology was used to take a closer look at the flight paths,” says Wolfgang Fiedler from the Max Planck Institute for Animal Behavior in Radolfzell. With 3,000 recorded flight sequences in two years, a lot has come together. “Perhaps the system overlooks one or two birds. This is probably not a big problem at the time of year and with the species, because there are no species that appear particularly frequently in the flock,” the scientist continues.
At the same time, however, we know that collisions are much more frequent at night and when visibility is poor than during the day and when the weather is nice. From April to mid-May and from September, migratory birds are mostly on the move at night over the sea. “The risk of collision could be higher in the times when we weren’t monitoring: at night, in very thick fog or very heavy rain,” Skov also confirms.
issue of habitat loss
For Fiedler, in addition to nocturnal and bad weather investigations, another question would be important: “Collisions are not the only problem, but there can also be a loss of habitat if the animals evade large areas and no longer use actually good feeding areas. This damage is then not so easy to quantify,” explains the researcher. “The avoidance observations only begin a few hundred meters in front of the wind turbines. This does not record whether they are avoided even more extensively, i.e. the birds do not even fly into this area.”
Another study with gannets using GPS telemetry data, in which one is not dependent on the range of optical systems, shows that the birds sometimes start to evade as little as one kilometer. “In the North Sea you can see relatively well that there are individuals who do not fly into these areas at all,” Fiedler continues.
The DHI technique would also be very beneficial on land, says Fiedler. “A lot is happening in this area. For example, there are approaches that observe birds in the vicinity of wind turbines either only with radar or with radar and intelligent cameras. There are also mechanisms that are supposed to switch off the wind turbine in an emergency if it notices that a vulture or an eagle flies straight towards it and doesn’t turn away.” Although “certainly more is promised than is really possible”, the systems are still very close to being ready for use.
(vs.)