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

This study reviewed data that have been collected from offshore windfarms and considers how they can be used to derive appropriate avoidance rates for use in the offshore environment.

Executive Summary

1. The selection of appropriate avoidance rates for use in collision risk models at offshore windfarms is often a key part of the Environmental Impact Assessment process. Ideally, these avoidance rates should reflect the behavioural responses of birds to turbines. However, they are often used as a 'fudge-factor' to incorporate aspects of model error. The situation is further complicated by a lack of data for marine birds and offshore windfarms. As a consequence, present guidance is based on values that have been derived for terrestrial species at onshore windfarms. This study reviewed data that have been collected from offshore windfarms and consider how they can be used to derive appropriate avoidance rates for use in the offshore environment. Aims of the study were five-fold:

  • To produce definitions for the types and scales of avoidance;
  • To review current use of avoidance rates;
  • To review and critique existing avoidance behaviour studies and any derived rates;
  • To provide summary avoidance rates and a total avoidance rate for each priority species/species group based on the evidence available at present;
  • To undertake an assessment of the sensitivity of the conclusions reached to inputs and conditions under which they were collected.

The study focussed on five priority species - northern gannet, black-legged kittiwake, lesser black-backed gull, herring gull and great black-backed gull - whose behaviour and distribution make them particularly prone to collision with offshore turbines.

Definitions ( section 3)

2. A key hurdle to defining appropriate avoidance rates for use in the offshore environment has been a lack of clear, agreed definitions of avoidance behaviour. Therefore, the first step of this review was to define the different scales at which avoidance behaviour may occur. Three categories of behaviour were initially defined - macro-, meso- and micro. Micro-avoidance refers to 'last-second action taken to avoid collision, which is considered to occur within 10 m of the turbine rotor blades. Meso-responses reflect all responses to individual turbines occurring between the base of each turbine and the windfarm perimeter (defined as 500 m from the base of the outermost turbines). Macro-responses reflect all behavioural responses to the presence of the windfarm that occur at distances greater than 500 m from the base of the outermost turbines. Avoidance rates are typically derived by comparing observed collision rates to the number of collisions that would be expected in the absence of avoidance behaviour, considering all bird movements within the perimeter of the windfarm. Consequently, calculations do not usually consider whether any avoidance action takes place at the meso- or micro-scale. It was thus also necessary to consider a fourth category, within-windfarm avoidance, which combines micro-avoidance and meso-responses.

Current use of avoidance rates ( section 4)

3. The avoidance rates used with collision risk models have shown substantial variation over time. Initially, very high values, often based on incorrect interpretations of data, were used. Since the earliest environmental impact assessments, there has been a broad tendency to follow standard guidance with avoidance rates of 0.95 and more recently, 0.98 used. However, in light of recent evidence from both on- and offshore windfarms these values are coming under increasing scrutiny from developers and their consultants.

Macro-responses ( section 5.1)

4. As with micro-avoidance and meso-responses, the evidence for macro-responses to the presence of a windfarm was typically inconsistent for gulls. Studies designed to look at potential displacement effects reported both evidence for attraction and for displacement and others no significant response at the limited number of sites which were available for consideration. Thus, for gulls, the balance of evidence suggests a macro-response of 0 ( i.e. no attraction to or avoidance of the windfarm). However, the response of northern gannet to the presence of windfarms appeared to be more consistent, with strong avoidance evident at several sites, although again it was not always clear whether the macro-response was a result of barrier effects or displacement. Based on the evidence currently available, it is suggested that a macro-response rate of 0.64 is a suitable precautionary value for northern gannet.

Micro-avoidance ( section 5.2) and meso-responses ( section 5.3)

5. Data for micro-avoidance and meso-responses were extremely limited. No clear and consistent patterns were evident for any of our five priority species. For this reason, it was not possible to derive micro-avoidance or meso-response rates for these species.

Within-windfarm avoidance ( section 5.4)

6. A total of 20 sites were identified as having sufficient data to derive within-windfarm avoidance rates by comparing observed collision rates to those expected in the absence of avoidance behaviour. Of these, nine were considered to have data of sufficient quality to estimate robust within-windfarm avoidance rates to be calculated using the Band (2012) collision risk model. Within-windfarm avoidance rates were derived for use with both the basic Band model (Options 1 and 2), that assumes that birds are distributed evenly within the rotor-swept area of a turbine, and with the extended Band Model (option 3) that uses a continuous flight height distribution to estimate collision risk at different points within the turbines rotor-swept area. Based on these data within-windfarm avoidance rates of 0.9959 (± 0.0006 SD) and 0.9908 (± 0.0012 SD) were derived for herring gull for use with the basic Band model and extended Band model respectively. Similarly, within-windfarm avoidance rates of 0.9956 (± 0.0004 SD) and 0.9898 (± 0.0009 SD) were derived for large gulls for use with the basic Band model and extended Band model respectively, and rates of 0.9921 (± 0.0015 SD) and 0.9027 (± 0.0068 SD) derived for small gulls also for use with the basic Band model and extended Band model respectively. Within-windfarm avoidance rates of 0.9893 (± 0.0007 SD) for the basic Band model and 0.9672 (± 0.0040 SD) for the extended Band model were derived for all gulls. Insufficient data were available to calculate a within-windfarm avoidance rate for northern gannet. (Note, where we report the standard deviation around the derived within windfarm avoidance rates, this relates variability between sites and not to uncertainty in the model input parameters. Estimating the contribution of the model input parameters to the uncertainty associated with the derived avoidance rates requires a more detailed understanding of the real range of values associated with each parameter than is available currently.)

Sensitivity of derived within-windfarm avoidance rates ( section 6)

7. The sensitivity of within-windfarm avoidance rate values to model input parameters was also assessed and it was found that the final derived values were most sensitive to assumptions about the proportion of birds at collision risk height. However, it was also found that sensitivity to input parameters declined as the number of flights through a windfarm increased.

Recommended total avoidance rates ( section 7)

8. Whilst we have estimated within-windfarm avoidance rates to four decimal places, current guidance from SNH is that expressing avoidance rates to more than three decimal places is unwarranted ( SNH 2013). Given the inherent uncertainty in the data we feel that this is a sensible approach to apply to total avoidance rates. For this reason, we round within-windfarm avoidance rates down to three decimal places when deriving recommended total avoidance rates. For gulls the balance of evidence suggests a macro-response of 0 ( i.e. no consistent attraction to or avoidance of the windfarm). Consequently, the recommended total avoidance rates for these species are equal to the within-windfarm avoidance rates. Therefore, avoidance rates of 0.995 for herring gull, lesser black-backed gull and great black-backed gull and 0.992 for black-legged kittiwake are recommended for use with the basic Band model. Based on the evidence available, it is suggested that the total avoidance rate for northern gannet is unlikely to be lower than that for all gulls. Assuming a macro-avoidance rate of 0.64, this would reflect a within windfarm avoidance rate of 0.9703. We acknowledge that this is precautionary, but in the absence of more species-specific data, we feel it is appropriate. Hence, an avoidance rate of 0.989 for northern gannet is recommended when using the basic Band model. For the extended Band model, avoidance rates of 0.990 for herring gull and 0.989 for lesser black-backed gull and great black-backed gull were recommended. Based on the evidence available at present, it was not possible to recommend an avoidance rate for use with the extended model for either black-legged kittiwake or northern gannet.

Transferability of avoidance rates between models ( section 8)

9. Whilst the basic and extended Band models are the most widely used collision risk models at present, there are a number of alternatives. Based on our assessment of the alternative models which we were able to obtain descriptions of, the definitions and values we present in this report are likely to be broadly applicable to other models.


Back to top