Abstract
Operational offshore wind farms are known to have a number of potential impacts on birds and these include mortality from collision with turbine blades and ancillary structures (moving and stationary). Offshore windfarm developers routinely use collision risk models (CRMs) to assess this potential impact on birds when undertaking environmental impact assessments. In the UK, for offshore windfarms, the most frequently used avian collision risk model is the Band model (Band 2012). The Band (2012) model requires a number of input parameters, including information on the density of birds in the windfarm area, bird avoidance rates, flight speed, flight height and size information for the bird species involved and various turbine parameters like rotor diameter, pitch and operational time. All of these input parameters have variability and uncertainty associated with them and since the predicted collision risk from the Band model is sensitive to the input parameters, variability in the input parameters can have a significant effect on predicted collision risk. However, consideration of this variability in the key input parameters is not routinely included when collision risk modelling is undertaken as part of the Environmental Impact Assessment (EIA) process, and uncertainty/variability around the collision predictions is rarely presented in environmental statements from offshore windfarm (OWF) developers. For these reasons a project was undertaken to develop the Band (2012) model using a simulation approach to incorporate variability and uncertainty in the collision risk modelling process. The output of this project was the development of a stochastic version of the Band (2012) collision risk model (Masden 2015) which allows variability around input parameters to be entered in the model and used to calculate a distribution of collision risk estimates which reflects the variability in the input parameters. Natural England, as part of its statutory advice responsibilities in relation to Nationally Significant Infrastructure Projects (NSIPs) in the offshore environment, would like developers to take account of variability and uncertainty in their assessment of potential collision impacts, and the stochastic version of the Band model developed by Masden (2015) offers a means of doing that. However, there has been limited testing of the application of this stochastic version of the Band model to datasets typically used by developers for collision risk modelling. Therefore Natural England commissioned this project to review and test the stochastic version of the model to determine the best way to parameterise the model using data available from EIAs, and to compare outputs derived from the stochastic version of the model against those generated by the Band (2012) model. Natural England will use the results of this project to inform our advice to offshore windfarm developers and the Planning Inspectorate regarding the assessment and significance of potential collision impacts to birds as part of the Environmental Impact Assessment (EIA) and Habitats Regulations Assessment (HRA) processes. The results of this Natural England project will also be used in a project commissioned by Marine Scotland that is developing an updated version of the stochastic Band model that builds on the work undertaken to date and will address the gaps and issues identified in the current version by industry and statutory agencies.