Abstract
Rijkswaterstaat, commissioned by the Ministry of Agriculture, Nature and Food Quality, requires a threshold value for an Acceptable Level of Impact (ALI) for a number of bird species, which can be used to assess the effects of offshore wind farms (OWF) on their population size. These thresholds are to be used in KEC (Kader Ecologie en Cumulatie) version 4.0.
Currently, effects of offshore wind farms are assessed using relatively simple criteria and thresholds like the ORNIS 1% criterion and the Potential Biological Removal (PBR). These methods are easy to apply and understand but come with limitations. The PBR approach has certain clear drawbacks, such as e.g. that it provides a fixed and very static figure that does not take any environmental variability into account. Moreover, it implicitly assumes a fixed level of undemonstrated density dependence in population development (O'Brien et al. 2017).
As a novel method of impact assessment, population models have been created for the species of interest in the southern North Sea (Potiek et al. 2019; van Kooten et al. 2019). This gives more insight into the current expected population trajectory, and the effect of additional mortality. This project describes a method to determine threshold for determining acceptable levels of impacts (ALIs), which can be assessed using population models. The required threshold will have to consist of two parts:
• A threshold population decline 30 years after the impact, as a percentage X of the projected population size without the impact, which is considered ‘acceptable’.
• A threshold probability Y that X is below this acceptable level after 30 years, which is considered an acceptable risk.
Together, X and Y lead to an ALI expressed as ‘The probability of a population decline of X% or more, 30 years after the impact, cannot exceed Y’. In order to help the Ministry of Agriculture, Nature and Food Quality to determine these threshold values for the relevant species, this report describes a newly developed method to derive X and Y, and presents four options for Y, corresponding to four different levels of ‘acceptable risk’, for each species. The method is consistently applicable to both mortality due to habitat loss and collisions with turbines. Additionally, the general framework of the method could be applied to assess any (combination of) impact(s). The requirement for application is a well-formulated matrix population model, for which there is a long and rich tradition in conservation biology.
We also provide factsheets for each species, outlining the most relevant species-specific data. This factsheet should aid in the choice of the most appropriate option for Y among the presented candidates. Finally, we provide a guideline on how this information may affect this choice. This choice has to be made by the Competent Autority and is therefore not addressed in this report. Acceptable levels of impact 8 The species for which we apply the method and derive potential X and Y values are:
• Brent goose
• Bewick's swan
• Common shelduck
• Eurasian curlew
• Red knot
• Black-legged kittiwake
• Great black-backed gull
• Herring gull
• Lesser black-backed gull
• Sandwich tern
• Common tern
• Black tern
• Great skua
• Arctic skua
• Guillemot
• Razorbill
• Atlantic puffin
• Red-throated diver (diver sp.)
• Northern fulmar
• Northern gannet
• Common starling