Scottish Marine and Freshwater Science Volume 5 Number 16:The Avoidance Rates of Collision Between Birds and Offshore Turbines

1. Introduction

The Scottish Government has a target for 100% of Scottish demand for electricity to be met from renewables by 2020 by creating a portfolio of both onshore and offshore technologies (Marine Scotland 2011). However, concern over the environmental impacts of these developments in the UK, and in particular the risk of birds colliding with wind turbines, has contributed to the delay and cancellation of some projects. In order to quantify the risk of birds colliding with wind turbines, a number of collision risk models have been developed (Band 2012, Smales et al. 2013). These include an update to the Scottish Natural Heritage ( SNH) collision risk model, originally developed for onshore windfarms (Band 2000, Band et al. 2007), redeveloped to better reflect data collected in relation to impact assessments for offshore windfarms (Band 2012). This work was undertaken as part of one of the projects undertaken through the Strategic Ornithological Support Service ( SOSS) programme, a joint initiative involving industry, statutory nature conservation bodies ( SNCBs) and the RSPB. These models combine a series of parameters describing the turbine design and operation with estimates of a bird's size and behaviour in order to predict the number of birds that would be expected to collide with a turbine over a given time period. Of these parameters, detailed analysis has suggested that these models are highly sensitive to variation in the avoidance rate, the proportion of birds which take action to avoid colliding with a turbine (Chamberlain et al. 2005, 2006). Despite this, there has been relatively little empirical evidence put forward to support avoidance rates for offshore windfarms, which are likely to vary according to species and weather conditions, in particular visibility.

Whilst avoidance rates can be determined from observed mortality rates or actual observations of birds' behaviour, defining robust values for use in collision risk modelling can be extremely challenging. However, there are concerns that avoidance rates derived from observed mortality rates may act as a 'fudge-factor', incorporating observer biases and model error, as opposed to the actual behaviour of the birds (May et al. 2010, Douglas et al. 2012). Current guidance from SNH (2010) is that, in the absence of species-specific empirical data, a default avoidance rate of 0.98 should be used for most species in onshore windfarm assessments and this value has been widely used in the offshore environment as well. However, in light of recent evidence ( e.g. Everaert & Stienen 2007, Krijgsveld et al. 2011) the validity of this approach has been questioned and concerns have been raised by developers that it will lead to an over-estimate of the likely number of collisions (Moray Offshore Renewables Limited 2012, Trinder 2012, Smartwind/Forewind 2013) and, as a consequence, potentially contribute to the delay and cancellation of key projects. In a policy environment where there is limited evidence on which to base decisions it is important to reflect uncertainty, but not to apply unrealistic levels of precaution which will make it difficult to reach informed decisions about where and where not to build windfarms.

There is a strong need for a consensus on the appropriateness of recommended avoidance rate values given the influence they have on collision estimates and, therefore, consenting decisions. However, at present, there is a lack of clarity over the interpretation of studies of avoidance behaviour and the applicability of the resultant avoidance rates to different collision risk models, study sites and species. As a result, details presented in reviews of avoidance behaviour of birds in the marine environment ( e.g. Maclean et al. 2009, Cook et al. 2012) have been subject to confusion. A key reason for this is the lack of consistency in the terminology applied to different spatial scales of avoidance, and the widely varying interpretation of the types of avoidance behaviour occurring. There is therefore, an urgent need for a review of avoidance behaviour in offshore windfarms in order to provide a clear appraisal of the existing evidence base, provide a robust critique of the data available with which to refine recommendations on avoidance rates and offer clear guidance as to how they should be used in future collision risk modelling scenarios. Whilst the focus of this review will be on collision risk modelling and species relevant to the UK context, it will draw on evidence from Europe and beyond.

This work aims to reduce the current level of uncertainty around appropriate avoidance rates for seabird species within collision risk modelling by providing a thorough review of the existing evidence base. The scope of this review is broader than those previously undertaken ( e.g. Cook et al. 2012) and includes quantitative and qualitative analyses of the data identified with a view to identifying representative avoidance rates for five priority species - northern gannet, black-legged kittiwake, lesser black-backed gull, herring gull and great black-backed gull. The review identifies current knowledge gaps and aims to ensure that future strategic work is targeted at addressing the most appropriate issues. Due to the sensitivity of the work and the importance of its conclusions, the work has been overseen by a steering group of key stakeholders and experts, with a view to gaining widespread acceptance of its conclusions.