Abstract
Large increases in the capacity for producing renewable energy are necessary if the UK and EU are to meet their greenhouse gas emission targets. Much of this increased capacity will come from offshore wind farms. The installation of these structures may have impacts on marine wildlife, including seabirds, requiring robust environmental impact assessments to identify and mitigate against potential risks. The overall aim of this thesis was to assess whether remote radar technology could be used to collect ornithological data required for EIA for offshore wind farms. An S-band marine surveillance radar was installed on the Beatrice Alpha oil platform, 22 km offshore in the Moray Firth, Scotland and operated from June 2006. This was adjacent to the site of the EU DOWNVInD demonstrator project, which installed two 5 MW wind turbines at the most offshore deep water site yet developed. Commercially available software linked to the radar was used to automatically detect and track birds, but significant amounts of radar clutter were also recorded in this offshore environment. Bespoke filters were then developed to eliminate non-avian tracks from the dataset. The filtered data showed temporal patterns over scales of days to seasons that could be linked to existing knowledge of the use of the site by seabirds. Patterns in flight directions during the breeding season indicated that birds using the site were also attending colonies at the East Caithness cliffs SPA to the north west. Radar data were also used to assess the flight speed parameter used in the Band collision model, which calculates collision probability between birds and wind turbines. Collision risk was shown to increase with decreasing ground speed of the bird. The flight speed values used in most calculations of collision risk are the published mean airspeeds, which do not account for the effect of wind on flight speed. Ground speeds of radar tracks were highly variable and on average were 0.707 ms-1 slower than airspeeds, indicating that in many cases it is not appropriate to use airspeed as a proxy for ground speed. Radar was shown to be a useful tool for answering some questions relating to offshore wind turbine impacts on seabirds, but should be viewed as one technique within a suite available and may not be the most appropriate to answer every question.
The second aim of this thesis was to use visual boat-based surveys in the spring of 2006 and 2007 to assess the impact of the installation of the two wind turbines on seabird abundance and distribution at the site. No impact of turbine installation was detected, despite large differences in abundance between the years. However, environmental variation had a highly significant effect on the density of birds, demonstrating the difficulties in assessing impacts against the background variability inherent in marine environments.