Abstract
DONG Energy has commissioned a consortium of Orbicon and DHI in association with Marine Observers to model the distribution of prey to Common Scoter Melanitta nigra by means of a habitat model as part of the monitoring program for the planned Horns Rev 2 offshore wind farm. 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.
Common Scoters feed almost exclusively on filter-feeding bivalves, and in the Danish part of the North Sea they are known to feed on both cut trough shells Spisula subtruncata and American razor clams Ensis americanus. The preference for razor clams seems to be of recent origin, and this preference has been suggested as a possible reason for the dispersal of scoters in the Horns Rev area experienced during the PSO-related monitoring programme. This dispersal pattern of razor clams possibly may affect the presence of scoters near the Horns Rev 2 site. As both primary prey species to Common scoter in the North Sea occur commonly in the region knowledge of the available supply of both razor clams and trough shells in the site and on Horns Rev in general is critical to the application of a BACI design for monitoring the impact of the wind farm on scoters. Lack of detailed hydrodynamic and geo-biochemical model to construct physiology-based growth models for standard knowledge about the distribution of prey can lead to ambiguous interpretations of monitoring results, and even bias conclusions on possible impacts.
The modelling work integrated dynamic model data on the potential carrying capacity of the two potential prey species in the area and field data on benthos and surface sediments and the relief of the sea floor. The carrying capacity models were developed using the output from a local combined hydrodynamic and geo-biochemical model to construct physiology-based growth models for standard individuals with an advection term that replenish the food ingested by filter-feeders.
Existing knowledge of food selection and prey dynamics by Common Scoter in the Horns Rev area could not be corroborated. Both additional dietary investigations in the main feeding area of the scoters, the linked benthos sampling programme and the model results unambiguously point at two rather well-defined prey communities of razor clams and trough shells, which are both utilised by the scoters. The community of through shells is related to the area of fine sediments and very high food supply in terms of nearseabed chlorophyll concentrations found in the Esperance Bugt and at the ground Cancer. Most of the areas over the reef are unsuitable to the trough shells, due to the sediment structure and the frequent turn-over of sediments as a result of periodic strong currents over the reef. The community of Razor clams, on the other hand, prefer the medium-sized sediments over the reef and to the northwest of the reef, including the planned wind farm site, and are due to their deep penetration into the sediment able to survive the frequent sediment re-bedding in these areas. Although this community is also found in the eastern part of the area the fine sediments seem to be unsuitable to the species, and prevent the razor clams to utilise the higher concentrations of food found just off the Danish coast. Rather than an ecosystem dominated by one prey species, American razor clams, our results indicate that both prey species are important at Horns Rev, and that most birds now and historically have been feeding on trough shells in the area.
The change in the historic distribution of Common Scoter recorded after 1999 showed a relocation of a proportion of 10-25 % of the birds along the reef into the offshore areas shallower than 15 m on the reef as well as on the plains northwest of the reef. These areas can now be identified as suitable habitat for razor clams and as mainly unsuitable for trough shells. Thus, it is highly likely that Common Scoters mainly feed on razor clams in the „new‟ area of their distribution. As the birds collected in the area in 2006 and in 2008 all were in excellent condition the most probable function of this dual prey system is that razor clams provide a supplementary source of food for scoters which are generally experiencing a rich supply of food in terms of trough shells. This is supported by the fact that scoters seem to use the razor clam habitats more during the later part of the season, when food availability in their primary feeding area may become scarce as a result of their own predation.
The distribution model for razor clam for the period 2000 to 2008 clearly shows that most of the footprint area of the HR2 OWF is prime habitat for this species, and hence that concentrations of Common Scoters may occur regularly in the wind farm site, as found during the surveys from 2005 to present. Based on the model results it is questionable whether scoters will use the existing HR1 wind farm site frequently, as the area does not seem to be highly suitable to razor clams.
Data on surface sediments and benthos are lacking in the areas to the west of Fanø and the whole area northeast of Horns Rev, - areas which are important to describe in terms of potential food supply to Common Scoter. Equally important, the available data on the Spisula and Ensis do only allow for a break-down of habitat suitability based on presence/absence. Quantitative data are needed to base suitability estimations on biomass levels rather than presence/absence patterns, as such estimations most likely will result in better spatial information on gradients and patchiness in the distribution of the two species on Horns Rev.