This report presents the technical background to the ornithological environmental impact assessment for the construction of an offshore windpark at Horns Rev, 14 km west-south-west of Blåvandshuk, Denmark. Construction of the park is planned to commence in 2001. The park will consist of c. 80 wind turbines, each of at least 1.8 MW, and cover an area of 27.5 km2 (including the 200 m exclusion zone around the park).
The inner Danish waters and the eastern part of the North Sea constitute major staging and wintering grounds for huge numbers of water- and seabirds. The present gross estimate, to be regarded as a minimum estimate, is that 5-7 million birds of more than 30 species winter in these areas, and that even larger numbers stage or pass through on migration. These numbers, which for several species constitute more than half of the northwest Palearctic breeding or flyway populations, are of international importance for the conservation of waterfowl populations. As a consequence, Denmark has obligations under both the Ramsar and Bonn Conventions and the EU-Birds Directive to protect and maintain these populations. For this reason, investigations of the impacts of offshore wind parks on water- and seabirds have been made a requirement of the project.
Detailed knowledge exists concerning the numbers of birds roosting at and migrating past the westernmost point of Jutland, Blåvandshuk. In relation to the Ramsar Convention, an area is of international importance for a flyway population of a species, if 1% or more of the individuals are present in the area at some time during an annual cycle. According to this criterion, the area is of international importance to divers, Eider, Common Scoter, Common Tern, and Sandwich Tern. A number of other species, e.g. Guillemots and Razorbills, are present in the area in significant numbers as well, though these numbers do not make up 1% of the population.
Previous counts have been carried out almost exclusively from the coast. Thus the distribution of birds in the area around and at Horns Rev remained virtually unknown. Therefore, transect counts from aircraft and ship were initiated in April 1999 and have been continued up to May 2000. So far, nine aerial and three shipbased counts have been carried out, covering an area of 1,700 km2 centred on the Horns Rev project area.
From the aerial counts, precise information about the distributions of birds in the area has been obtained. In all counts, the two species feeding on the bottom fauna (Eider and Common Scoter) were found almost exclusively along the coast (within the 6 m depth contour), and only in a few cases more than 10 km offshore. Very few individuals of either species have been observed within 2-4 km of the projected wind park area. The only species recorded offshore in significant numbers were fish-eaters, divers, Gannets, auks and terns, plus large numbers of gulls, often concentrating around fishing vessels. The distribution of fish-eating species was highly variable, probably because distributions and densities of fish were variable in both space and time.
Assuming a random distribution of birds in the surveyed area, c. 1% of the total numbers recorded would be predicted to be present in the area of the projected wind park. Only divers (8 of 554 recorded individuals) and Kittiwake (11 of 1,118 recorded individuals) approached this threshold, though they did not exceed it significantly. All other species abundant in the general area were present in the projected park zone in lower than expected numbers. From this it is concluded that the specific construction area of the projected wind park is of very limited significance for water- and seabirds judged by their overall distribution in the waters around Horns Rev.
Analysis of both aerial and ship counts showed that the probability of detecting flocks of birds decreases rapidly with distance from the observer. For divers, detection probability from ship was reduced to 50% at a distance of 150 m. Estimates of densities and total numbers depend on the precision of being able to correct for birds overlooked by the observers. It is concluded that aerial counts provide objective data on bird distribution and of the minimum number in the area, but that further development of statistical methodology - considered outside the scope of the present study - is needed in order to calculate definite densities and total estimates.
As a result of the comparisons between the results of the aerial and ship surveys, preference is given to the former. Both methods have statistical problems associated with the conversion of count data into density and abundance measures. However, because a total aerial survey can be carried out in one day, whereas a ship survey (under ideal circumstances) takes four days to cover the same area, there is a much higher probability that birds will move inside the count area during the survey, and thus potentially may be counted twice or missed altogether.
Impacts on birds resulting from construction work are expected to be temporary and limited. The effect of laying the cable to land is also considered to be temporary and minimal in two proposed scenarios.
Potential permanent impacts on seabirds resulting from the long-term operation of the wind park are identified under three main headings:
- Physical changes of the habitat
- Disturbance effects resulting from avoidance of the turbines, equivalent to the loss of potential to exploit the otherwise available habitat
- Collision risk
Physical changes of the habitat include 1) the loss of the bottom area which supports the turbine foundations, 2) the provision of new underwater substrate for the settlement of larvae of marine invertebrates, and 3) the provision of platforms for birds to sit or perch on.
Loss of bottom substrate comprises less than 0.3% of the wind park area and is not expected to lead to measurable impacts. Settlement of fauna will mainly involve Balanus and some Polychaete colonisation that do not represent significant food items for birds, hence it is not expected that these modifications of the habitat will lead to any significant changes in bird numbers and distributions in the area. The provision of platforms for sitting/perching may attract some gulls and possibly Cormorant to the wind park area.
The degree of avoidance of wind turbines structures in the open sea have not been investigated for the species present in the area. Previous investigations of Eiders and geese show that avoidance of wind turbines is limited to distances within a range of 100-500 m. Even in a ‘worst possible case’ scenario (assuming birds completely avoid the wind park area up to a distance of 4 km) habitat loss within the total investigated area will not exceed 10% for most of the species involved.
A number of case studies carried out in terrestrial environments demonstrate that the collision risk of wind turbines is limited. Even at night, studies suggest that flights of Eiders and Common Scoters in inshore waters tend to avoid wind turbines. Most studies concerning collision risk cover situations with migrating land birds. Only a single study dealing with foraging birds of prey documents a relatively high number of collisions indicating that some raptors are less able to avoid turbines when foraging. For Gannets, terns and skuas, it is not possible to completely rule out the potential risk of collisions by these birds when foraging in the area. Similarly, the mounting of light on the turbines for ship navigation may attract nocturnal migrants during conditions of poor visibility. With the existing knowledge of the birds species in the area and of their reactions towards turbines, the possibility of some collision risk between birds and turbines cannot be ruled out.