Developing an avian collision risk model to incorporate variability and uncertainty

Lead Research Organisation: University of the Highlands and Islands
Department Name: The North Highland College UHI


The wind energy sector is an industry of strategic national importance, which can help secure our energy supplies, reduce our emissions and dependence on imported fossil fuels, and protect our environment. It is an industry on which our clean energy future rests. Despite the positive benefits of wind farms however, there is concern and uncertainty over the possible negative effects wind turbines may have on the environment, particularly on birds. For example, uncertainty remains over collision mortality i.e. the number of birds killed directly through collision with wind turbines. These uncertainties are far from trivial for the industry and have real consequences, potentially delaying wind farm projects and inhibiting the ability of the UK to meet its binding 2020 targets. Three projects in Round 2 of wind farm developments in the UK were delayed by over three years due in part to uncertainties over the assessment of impacts. Therefore better quantification of the uncertainty and variability associated with the estimation of impacts is required.

During Environmental Impact Assessments of wind farm developments, bird collision mortality is estimated using a mathematical model which describes the interaction of a bird with a wind turbine and predicts the risks of bird collisions with turbines. There are a limited number of collision risk models in use, not only in the UK but globally. However, it is recognised by many, including industry, statutory nature conservation agencies and academics that there is much room for improvement of these models. For example, collision risk models are deterministic and rarely include variation in the input parameters such as bird density, or bird biometrics which are inherently variable, but instead use average values. Additionally, any uncertainty in these values is not expressed. Adopting a single best value for parameters may reduce complexity and increase the accessibility of results for decision-makers however it can be misleading because it ignores the range of consequences that are plausible.

This project aims to i) review current models that are used to predict bird collision mortality caused by wind farm developments, ii) determine statistical methods suited to address any shortcomings of current models and then, using this information, iii) develop an updated model which incorporates variability and uncertainty. Reviewing current models and highlighting their strengths and weaknesses, as well as reviewing methods to incorporate variability and uncertainty will aid the development of a product, a collision risk model, which is fit for purpose. Development of the understanding of uncertainty in the outputs of collision risk models will be a key part of this project, and will be of direct benefit to industry, government advisors and regulators in the assessment and licensing processes for wind farm projects. The involvement of these parties will be vital in steering this project because any revision of a collision risk model has to function to better inform planning decisions for wind farm developments. To ensure that all relevant parties are involved, contribute and ultimately buy-in to the development of a new, updated model, there will be a workshop to discuss issues surrounding current practices to which developers, licensing authorities, statutory nature conservation bodies, academics and others will be invited. Also, to ensure the outputs of this project have impact and are used by the industry, the model and any documents produced will be made freely available and accessible through a dedicated webpage.

Wind energy has an important role in meeting energy targets, so there is a clear need to ensure that decisions made through the planning processes use the best available information, data and models. Improved understanding of the risks of collision to birds - a key effect considered in ornithological impact assessments of wind farms - is thus vital.


10 25 50
publication icon
Masden E (2016) Avian collision risk models for wind energy impact assessments in Environmental Impact Assessment Review

Description We have developed an update to the Band avian collision risk model (that is used in relation to offshore wind energy developments) that includes uncertainty and variability. The simulation model update randomly samples from distributions for each of the model parameters and the simulations can then be used to derive average collision estimates, with associated confidence intervals. The model update will therefore allow for a better understanding of the uncertainty associated with the predicted collision impact of a wind farm development and provide confidence limits, something which has previously been absent. In addition, the incorporation of uncertainty would reduce the possibility that a collision estimate was driven by the choice of a single input parameter value. Ultimately, the update will aid streamlining of the planning/consenting stages of a development by providing information not only on the magnitude of collisions i.e. the number of collision events, but also the likelihood of that number of collisions occurring.

In developing the model update we collated opinions of stakeholders with experience of offshore wind and ornithology, regarding uncertainty and variability in collision risk models. The responses from the stakeholder interviews helped to direct the model development. As part of the model development we also reviewed avian collision risk models which are currently available for use. It is the intention that we shall publish this review in the scientific peer-reviewed literature and we currently have a draft manuscript.

In addition to the above, we also conducted sensitivity analyses for the different options with the Band model. This research proved to be very informative and highlighted where to focus future data collection in order to improve collision risk estimates. In particular it highlighted bird flight speed as an area where little is still known, though this may improve with the increased use of biologging technologies such as GPS tracking.
Sectors Energy,Environment

Description The project has been referenced in the Joint Statuatory Nature Conservation Bodies response to the British Trust for Ornithology Avoidance Rate Review as an important upcoming report, and was subject to a special meeting of the collision sub-group of the Marine Renewables Ornithologists Group (MROG), which comprises Natural England, Scottish Natural Heritage, Natural Resources Wales, Northern Ireland Environment Agency, RSPB and Marine Scotland Science. For the RSPB it has provided valuable insight into the collision risk process. The model is already being used by RSPB as part of analysis of recent development applications, and the surrounding review, sensitivity analysis and interviews have been used to inform the policy underpinning the response of RSPB to these developments. As such it will provide part of the evidence presented in precognition for an upcoming public enquiry and has helped informed the Petition to a Judicial Review. By greatly assisting the comprehension of, and defining the means to deal with, uncertainty and variability in collision risk modelling, it has resolved a number of issues for RSPB. In the future it is expected that the model update will have economic benefit, in two main ways. Firstly, when preparing environmental statements, developers and their environmental consultants will be required to devote fewer person-hours to undertaking collision calculations as the updated model uses semi-automated scripts. Secondly, it is expected that by including variability and uncertainty in the estimation of collision risk estimates, that the risk will be more readily assessed and therefore the consenting process will be streamlined.
First Year Of Impact 2015
Sector Energy,Environment
Impact Types Economic

Description Annex IV review
Geographic Reach North America 
Policy Influence Type Citation in systematic reviews
Title Band model simulation 
Description A Monte Carlo simulation version of the Band collision risk model produced to include uncertainty and variability. 
Type Of Technology Webtool/Application 
Year Produced 2015 
Impact Key impacts: • Standardisation of the methodology and automation of the calculation process reduces the processing and analysis cost of running an collision risk calculation by approx. 90%, from an estimated £336k for a single wind farm development to approx. £36k, according to Marcus Cross of Scottish Renewable Power, plus increases regulator and developer productivity by reducing the discussion time required in the consenting process. • Will provide more realistic and scientifically credible estimates of collision numbers due to the inclusion of uncertainty and variability in the output figures. Currently only a point estimate is produced. The most significant impact is likely to be realised when considering cumulative impacts of multiple wind farms, such as those in the North Sea.