Abstract
Between February 2000 and April 2025, 62 observer-based aerial surveys of birds were conducted at Horns Rev, an area of the North Sea off central Jutland, using a Distance Sampling line transect survey design. The survey area covered the offshore wind farm (OWF) areas of Horns Rev I (HR I), Horns Rev II (HR II) and Horns Rev III (HR III). The surveys were classified into four main phases according to the construction and operation of the three wind farms developed within the area. Phase 0 included 15 surveys prior to any wind farm construction, Phase 1 included 25 surveys post-construction HR I and pre-construction HR II and HR III, Phase 2 included 10 surveys post-construction HR I and HR II, but pre-construction HR III, while Phase 3 included 12 surveys post-construction of all three OWF’s. Owing to changes in survey coverage and expansion in geographic ranges during Phase 1, which could bias the results, Phase 1* was created to represent the latter surveys from November 2005 to late spring 2007. The combined data set collected between 2000 and 2025 offers a unique opportunity to address the potential change in the displacement of birds over time, based on empirical data.
This report describes the changes in abundance and distribution of common scoter Melanitta nigra and divers (predominantly red-throated diver Gavia stellata but potentially including some black-throated divers Gavia arctica) over the period, based on the statistical analysis of visual aerial survey data in the Danish North Sea Horns Rev area gathered during the surveys described above. Species-specific Distance Sampling analyses were undertaken to correct for various aspects of avian detection probabilities, pooled across surveys. This was since followed by survey-specific spatial analyses with covariates including water depth (bathymetry), distance from the coast and/or a geographical covariate to model the distributions of common scoters and divers at a fine geographical scale for each survey.
The number of common scoters in the entire survey area increased markedly from Phase 0 to Phase 2 due to the distributional expansion west and northward from the coastal area and not necessarily an increase in overall abundance. The density of birds declined in Phase 3, which was dominated by the decline in the 2024/2025 winter surveys compared with previous survey years. Because of a general shift in the distribution of common scoter in the survey area over the first years of the survey period, it was difficult to assess the impact of the installation of the HR I. Common scoters showed significant displacement from the footprint of HR II within a distance of 3 km between Phase 1* and 2 and within a distance of 6 km between Phase 1* and 3. The displacement effect was also be analysed in direction-specific sectors around HR II. The longest displacement distances occur in the western sector in both phase differences (5 km and 8 km respectively). Within the HR II footprint (the area within the outer perimetry of the Horns Rev II offshore wind farm) common scoter densities decline by 50% between Phase 1* and 2 and a further 65% in Phase 3. Common scoter densities significantly increased in the HR III footprint between Phase 1* and 2, when HR III was not constructed. Between Phase 2 and 3, after the construction of HR III, significant declines of common scoters were observed. These declines were in keeping with a general density decline in much of the survey area and therefore not necessarily directly related to the construction of HR III.
While common scoters occurred in the Horns Rev survey area in high densities, diver densities were much lower. Diver densities in the HR I area were found to be fairly stable between Phases 0 - 2, but significantly decreased between Phase 2 and 3, a decline that was in line with declines across much of the study area. HR II showed strong evidence for divers being significantly displaced around its footprint within 5.5 km between Phase 1* and 2 (pre- and post-construction) and within 8.5 km between Phase 1* and 3. The largest displacement distances were observed in the northeast, Phase 1*-2, and the southeast, Phase 1*-3. Initially post construction, Diver densities declined by 75% within the footprint of the wind farm (100% grid cells significantly decreasing) and then by a further 24% between Phase 2 and 3 (30% cells significantly decreasing).
We evaluated the long-term distribution of common scoter and diver species, bird species classified as sensitive to human disturbances, in and around the HR I, II and III OWF’s at Horns Rev. Due to low and variable numbers of divers and common scoters in the HR I area this wind farm did not allow for definite conclusions of the displacement effect of that wind farm. We found that divers and common scoters decreased in and around the HR II wind farm after its construction. Both divers and common scoter continued to be found in decreased densities in the HR II area between Phase 2 and 3. Within the HR III area, with larger and more widely spaced turbines, declines in the density of common scoter or divers was observed in the post-construction phase, though not as pronounced as in the HR II area and at least in part owing to survey area wide declines observed in the 2024/2025 surveys.