TY - JOUR
T1 - A critical appraisal of 40 years of avian collision risk modelling
T2 - How have we got here and where do we go next?
AU - Cook, Aonghais s.c.p.
AU - Salkanovic, Eldina
AU - Masden, Elizabeth
AU - Lee, Hsiu eik
AU - Kiilerich, Alexander holm
N1 - © 2024 The Authors. Published by Elsevier B.V.
PY - 2024/11/12
Y1 - 2024/11/12
N2 - Collision risk models (CRMs) form a key part of the Environmental Impact Assessment (EIA) process for wind farms. It is 40 years since the first CRM was published. We take a critical approach to reviewing the CRMs available in the peer-reviewed and grey literature, with a view to better understanding the current challenges, and how these may be overcome. In total, we identified 52 models, which include 75 input parameters amongst them. There has been a tendency for models to become more complex over time, incorporating more input parameters with a view to improving biological realism. However, many of these models have not been applied within the EIA process. Consequently, despite advances in modelling and the types of data available to inform models, few of these advances have been implemented in regulatory processes. As few models have been validated, it would be beneficial to make use of data from any existing or planned post construction monitoring campaigns to do so, where possible. Additionally, we recommend for sensitivity and benchmarking analysis to compare results from established models for any new CRM proposed. As the industry continues to expand, it is important that stakeholders have confidence in any approach used to assess collision risk. This will likely require close collaboration and careful discussion between all stakeholders.
AB - Collision risk models (CRMs) form a key part of the Environmental Impact Assessment (EIA) process for wind farms. It is 40 years since the first CRM was published. We take a critical approach to reviewing the CRMs available in the peer-reviewed and grey literature, with a view to better understanding the current challenges, and how these may be overcome. In total, we identified 52 models, which include 75 input parameters amongst them. There has been a tendency for models to become more complex over time, incorporating more input parameters with a view to improving biological realism. However, many of these models have not been applied within the EIA process. Consequently, despite advances in modelling and the types of data available to inform models, few of these advances have been implemented in regulatory processes. As few models have been validated, it would be beneficial to make use of data from any existing or planned post construction monitoring campaigns to do so, where possible. Additionally, we recommend for sensitivity and benchmarking analysis to compare results from established models for any new CRM proposed. As the industry continues to expand, it is important that stakeholders have confidence in any approach used to assess collision risk. This will likely require close collaboration and careful discussion between all stakeholders.
U2 - 10.1016/j.eiar.2024.107717
DO - 10.1016/j.eiar.2024.107717
M3 - Article
SN - 0195-9255
VL - 110
SP - 107717
JO - Environmental Impact Assessment Review
JF - Environmental Impact Assessment Review
ER -