Abstract
The aim of the project is to assess the collision risk between birds and wind turbines at the Horns Rev wind farm. In 2003 the studies focused on describing bird movements in relation to the wind farm and to identify the species-specific behavioural responses towards the wind turbines shown by migrating and staging species. The Horns Rev area lies in a region known to be important for substantial waterbird migration as well as holding internationally important numbers of several wintering and staging waterbird species.
Theoretically, birds approaching the wind farm may:
- pass through the wind farm
- increase flying altitude and pass above the wind farm
- change direction and pass around the wind farm
Only birds passing through the wind farm risk collision with turbines, hence determining the proportions of all migrating birds adopting the above three alternatives is crucial to our assessment of collision risk. Having entered into the wind farm, the risk is assumed to be highest for birds flying in the altitude of the turbine rotors. Consequently, flight altitude is another critical factor for those species entering the wind farm in the assessment of collision risks.
The present study is restricted to the period after the construction of the wind farm. For practical reasons, data from the pre-construction period was not collected. Consequently, no base-line studies of bird movements in the area prior to establishment of the wind farm are available to which the present data can be compared.
All observations of birds were undertaken from the transformer station situated north of the northeasternmost turbine in the wind farm. Mapping of flight movements routes was undertaken using radar surveillance day and night. Visual observations were performed during the daytime along four transects, two located north and east of the wind farm, one along the eastern row of turbines and the fourth crossing diagonally through the wind farm in a southwesterly direction. Combined use of radar and visual observations during the daytime provided speciesspecific information on bird movements and orientations as well as data on flight altitude. Visual observations were performed in August 2002 and April-May and August-November 2003. Radar observations commenced in August 2003 and continued until November. Due to a temporary cessation of the study, it was not possible to collect data during February-March 2003, the period of peak occurrence of staging divers in the area.
Radar tracks of flying bird were entered into a GIS-database, from which subsets of data were selected to describe bird movements. In this report, radar observations were used to describe:
- The flight direction of migrating birds approaching the wind farm in order to assess the degree of avoidance towards the wind turbines
- The probability of birds flying into the wind farm from the outer edge to measure the overall response of passing birds to the presence of the wind farm *In combination with visual observations, to describe the species-specific responses (flight direction and altitude) to the wind farm.
Bird movements generally followed a southwesterly orientation and the intensity was highest during night. Only a small percentage of bird tracks entered the wind farm (14-22%). The majority of tracks either changed their orientation and passed around the wind farm, most reacting 400 m from the wind farm (north side) or 1,000 m (east side), or disappeared from the radar screen. The disappearance of radar tracks is most likely the result of birds changing flight direction, resulting in a change in body orientation and hence reduced reflection area of the birds and thus lower detection probabilities by the radar. Loss of tracks may also reflect birds landing on the water. Whatever the precise nature of these disappearances, it is clear that loss of tracks on the radar screen reflects an avian behavioural response to the wind farm by approaching birds. Since most bird tracks disappeared c. 400 m from the outer turbines of the wind farm (north side) or 1,000 m (east side), these distances may represent the general extent to which flying birds avoid such structures.
In the area north of the wind farm, bird movements followed a general southwesterly orientation at distances greater than 400 m from the wind farm. The orientation did not differ between day and night, nor was it affected by different wind directions. Bird tracks within 400 m of the wind farm were predominantly of a southerly orientation and differed significantly from the general southwesterly orientation further away. This suggests that birds approaching the wind farm adjusted their flight direction and those that did pass through the wind farm did so along the open corridors between turbine rows, thereby further reducing the potential collision risk. Birds approaching the wind farm from the east in a southwesterly orientation started to adjust their flight to a more westerly direction within 1,000 m from the eastern border of the wind farm. Probably due to the fact that gulls and terns, which seem to be attracted by the wind farm, were almost exclusively recorded by the radar in the area east of the wind farm, a clear pattern of deflection was not found in this area.
Analyses showed that adjustment of the flight direction (in respect of the turbine rows) was more accurate during the day than at night, which may relate to a more precise recognition of individual turbines by the birds during the hours of daylight.