Abstract
DONG Energy has commissioned a consortium of Orbicon and DHI in association with BIOLA to undertake a study of bird migration as part of the monitoring program for the planned Horns Rev 2 offshore wind farm. This report contains the results of the baseline phase running from October 2007 to April 2008. With respect to bird migration the monitoring program is focused on documenting the long-distance migration across and along Horns Rev during autumn and spring. This information is needed in order to describe any impacts of the construction of Horns Rev 2 offshore wind farm on migratory birds and to describe any cumulative effects, here especially potential barrier effects as a consequence of both the Horns Rev 1 and Horns Rev 2 wind farms. The potential barrier effects on feeding movements of birds staging in the area is given less attention. The statistical design of the monitoring program is based on the BACI approach applied to radar recordings (direction and density of echoes), and it incorporates data from the PSO monitoring programme for the Horns Rev 1 wind farm (HR1).
The establishment of the Horns Rev 2 offshore wind farm (HR2 OWF) was granted by the Department of Energy on the 19th March 2007 on the basis of DONG Energy’s application of 13th October 2006. The location of the HR2 OWF is planned for the outer part of Horns Rev, and it consists of a total of 91 turbines, each 2.3 MW which are placed with 13 east-west oriented rows of 7 turbines. The monitoring of bird migration was based on combined visual and radar studies. Due to the late start of construction works at the HR2 OWF only the spring period was covered, and no platforms were available for mounting the radar at the site. Hence, the study had to be made from an anchored vessel at HR1 and HR2 in combination with visual observations and a fixed installation at Blåvands Huk. Two X-band ship surveillance radars with a power output of 12 kW and 25 kW, respectively, were used for the radar observations. They were both operated with the antenna rotating horizontally. The two radar devices in use were a Decca “BridgeMaster E” at Blåvands Huk and a Furuno “FR 2105” on the vessel.
Post-processing was made with the purpose to compare bird movements observed at the two anchoring positions at the HR1 and HR2 OWF’s as well as to compare the results of bird migration/movements at Horns Rev with those at Blåvands Huk in terms of timing, intensity, direction, flight altitude and species composition. Radar images of the horizontal radar were captured from the screen signal via video splitter by a framegrabber card on a PC using custom-made software called SWARM (by TriOS Mess- und Datentechnik, Oldenburg). One screenshot was stored every minute and a cumulative picture every 5 minutes. All radar recordings were transferred to a GIS database allowing the analyses of flight directions and patterns of tracked trajectories. At each location, flight intensities were estimated both as effort-corrected totals for all birds recorded and for selected species groups (visual observations). In addition, to demonstrate the apparent concentration of bird movements and migration near the coast during the study period flight intensities were estimated in relation to distance from the coast at Blåvands Huk. An analysis of flight intensities with distance was made using radar data from Blåvands Huk and summarising flight intensities in 15 north-souths oriented transects, each with a width of 1 km, extending from the coast to the Horns Rev 1 offshore wind farm.
At both offshore sites movements of birds were dominated by ducks, terns and gulls with Common Scoter, Herring Gull, Lesser Black-backed Gull, Common Gull and Sandwich Tern being numerically the most important species. As these species groups usually were stationary for longer periods and certain individuals were likely to be present in the area for the duration of several hours or even days, more than 80 % of the observations in both study areas might represent foraging flights or flights of resting birds. Divers were observed more frequently at HR2, and Common Scoters less frequently as compared to HR1. The vast majority of the observations at Blåvands Huk were of typical coastal water birds, whereas no songbirds were seen. The three most abundant species at Blåvands Huk were Common Scoter (57.6 %), Divers (15.6 %) and Knots (4.7 %).
As a result of the dominance of scoters, terns and gulls the altitudinal distribution of flights displayed a concentration at altitudes lower than 20 m. At Blåvands Huk, due to the migration of geese, a moderate proportion of birds were observed at altitudes between 40 and 200 m.
The study revealed low flight intensity over Horns Rev and hardly any migration activity during daylight hours – most birds being local displaying feeding movements in all directions. Between HR1 OWF and Blåvands Huk profile analyses showed a steep increase in flight intensities approaching the coast at Blåvand. In general, flight intensities were much higher both for migration and feeding movements near the coast than offshore.The movements of birds at Blåvands Huk were dominated by southward movements of waterbirds compensating for northward drift by the current during the night.
The bird migration study indicated that movements driven by local and in most cases moderate feeding concentrations are more important on Horns Rev in spring, while close to Blåvand compensating movements of Common Scoters are more important. This is corroborated by the baseline monitoring of resting birds in the area showing a concentration of Common Scoters west of Blåvands Huk and only few birds feeding over Horns Rev. No response behaviour to HR1 OWF could be established due to the low flight intensity at the site and lack of feeding concentrations of Common Scoter. In addition, the use of ship-based radar shortened the effective detection range limiting the scope for tracking behavioural changes of birds passing the wind farm.
The low intensity of bird movements and migration during spring indicate that potential barrier effects of the two wind farms on long-distance migrants most likely are of limited magnitude during this season. However, the profile analyses of the radar data from Blåvand indicate that larger movements of Common Scoters occur irregularly offshore, and the aerial surveys over the past five years have shown that a larger proportion of the scoters in the area may concentrate over the western part of the reef, especially during late winter. It should be noted that the potential barrier effects on long-distance migrants may be larger during autumn migration due to the much higher volume of birds passing through the area during autumn as compared to during spring.
As the data on bird migration collected during the PSO programme did not cover the central and western areas of Horns Rev the cumulative database on bird migration across Horns Rev is judged insufficient for the baseline of the HR2 monitoring programme. For this reason the baseline has been extended into the autumn season 2008 using the established foundations in the HR2 OWF as platforms for continuous recordings using 25 kW radars concurrently with recordings from the HR1 transformer station and from the coast at Blåvands Huk.