Photo credit: Pexels
In an effort to mitigate anthropogenic carbon emissions related to energy production, wind energy is promoted as a plausible alternative. However, installing wind power facilities is directly related to land take and habitat fragmentation, through roads and artificial surfaces needed for installing wind turbines. In addition, the operation of wind turbines is directly linked to biodiversity loss via avian species fatalities. Biodiversity through various ecosystem services moderates extreme climatic events, and soil erosion among others. We are thus facing a paradox of impacting biodiversity to modulate potential anthropogenic climatic impacts, as there are strong interaction effects between biodiversity, land use and climate.
Wind energy facilities are increasing worldwide. Bats are long-lived mammals with low reproductive potential and thus high levels of adult individuals survivorship is important for maintaining a viable population. All European bat species are strictly protected under the Habitats Directive and the national legislation of many countries. Nevertheless, in the absence of specific, mandatory regulations, the effect of wind farm installation and operation on bat populations is rarely examined to a sufficient extent in environmental impact assessments. Risk assessment is of particular importance for the estimation of bats killed under wind turbines which might be often underestimated due to the fact that carcasses are of small size, not easily detected, quickly consumed by scavengers or destroyed from collisions. Death counts are improved by the use of detection dogs, cameras and digital technologies, but these methods are costly and the scale and extent of their applications limited. Ultimately for detecting how ‘clean’ wind energy is there is a need for land use and wildlife deaths assessment.
Bat deaths depend upon characteristics of the environment around wind turbines such as cover of natural, agricultural, or artificial surface areas. It is often assumed that the interaction scale of wind turbines and the surrounding environment is a few hundred meters, but in general it rarely is quantified and thus not well known. Deaths depend on wind turbine characteristics such as rotor diameter, speed, tower height which are correlated. To that end there are arguments regarding several small or fewer large wind turbines that may be less harmful. However, a study in the USA indicated deaths are best explained by the energy produced by wind turbines.
Methods: The effects of wind turbine features and environmental variables at different spatial scales associated to bat deaths in a mountainous and forested area in Thrace, Greece were investigated. Initially, the study sought to quantify the most lethal wind turbine characteristic between tower height, rotor diameter and power. The scale of interaction distance between bat deaths and the land cover characteristics surrounding the wind turbine was quantified. A statistical model was trained and validated against bat deaths and wind turbine, land cover and topography data (dataset 1). Variance partitioning between bat deaths and the explanatory covariates was conducted. The trained model was used to predict bat deaths attributed to existing and future wind farm development in the region (datasets 2 -5).
The study is published in a peer-reviewed journal:
Moustakas, A., Georgiakakis, P., Kret, E. & Kapsalis, E. (2023). Wind turbine power and land cover effects on cumulative bat deaths. Science of The Total Environment 892: 164536.
https://www.sciencedirect.com/science/article/abs/pii/S0048969723031571?via%3Dihub
What is the most ‘lethal’ wind turbine feature?
A study in Greece 1 explicitly investigating the most lethal wind turbine feature in bats among tower height, rotor diameter, and wind turbine power (nameplate capacity) indicated that all three are highly correlated and that it is wind turbine installed power that best explains bat deaths.
Figure 1: fewer large or several small wind turbines? What is the most ‘lethal’ wind turbine feature to wildlife?
What is the interaction scale between wind turbines and the surrounding environment?
The same study 1 indicated that among 3 scales examines, 250, 1000, and 5000 meter radius around each wind turbine, it was 5 km, the larger distance than the ones examined, that best explained the interaction between the surrounding environment and deaths.
Figure 2: Wind turbines (black dots) examined with buffers of 250, 1000 and 5000 meters around them. What is the interaction scale between wind turbines and the surrounding environment? Photo credit: P. Georgiakakis
How many bat deaths pass under the radar?
In terms of percentages in comparison with recorded deaths (dataset 1; Figure 3), there is a higher probability of deaths of 377.8% in operating, but not sufficiently surveyed wind turbines (dataset 2 & 3). Wind turbines not operating yet, but bound to (data set 4) or likely to operate soon (data set 5) together will contribute to 210.2% excess deaths than the ones recorded. Under the scenario that the wind turbines under production and installation licence proceed to operation, the total excess deaths will be 588% higher than the ones recorded1. The number of bat deaths recorded in the field (dataset 1) is an underestimate of the actual one due to difficulties in detecting them (they are small, there is vegetation around), quickly consumed by scavengers, or destroyed from collisions.
Figure 3. Bat deaths per year predicted in unmonitored neighbouring wind energy facilities (datasets 2-5) compared with the monitored one (dataset 1). Circles indicate the mean while whiskers indicate a 95% confidence interval of the mean. The horizontal dotted red line indicates the actual recorded number of deaths in dataset 1 in a year. Photo credit: A. Moustakas
Where should wind turbines be avoided?
Bat deaths were best explained by wind turbine power1. Natural land cover was the second best predictor of bat deaths in terms of variance explained. It also had the steepest coefficient slope across all variables, deaths steeply increased for every unit (1%) of natural land cover increase. Thus, the ‘perfect storm’ for bat deaths needs a combination of high total power and wind turbines installed in natural land cover areas. To that end results from this study confirm conclusions from Canada indicating that ‘the larger a facility is, the more important specific spatial and environmental context becomes in determining bat mortality’ 2. Results from Greece, in addition to previous studies suggest that wind turbines should not be licensed in areas where natural land cover at a radius of 5 km exceeds 50%.
Outside these high-risk areas, environmental impacts assessment studies should extend the search range for important foraging, drinking and roosting sites in a sufficiently large area, typically no smaller than 5 km. The spatial cumulative effect of at least 5 km should be assessed as a combined effect of wind energy developments comprised of multiple wind turbines from different wind energy facilities, taken together, rather just for a single wind turbine.
Figure 4: a. Effects of each covariate of the statistical model on bat deaths. Solid blue line indicates model fit, while shaded light blue envelopes indicate a 95% confidence interval. Vertical axis indicates probability of death for each variable, while horizontal axis the data range of each predictor in the final model. Thickness of black bars in the horizontal axis indicates data density. b. Variance partitioning of the explanatory covariates on bat deaths. Variance partitioning is quantifying the percentage of variance explained by each variable, and ranking their relative importance. Results add to 100%. Photo credit: A. Moustakas
Wind energy or power and deaths
Results of the study in Greece indicate deaths are significantly associated to wind turbine power. Power alone explained a third of the variance of bat deaths and it is more lethal than tower height or rotor diameter. Energy, the product of power and time, produced by wind turbines is determined by the size of the turbine (tower height, rotor diameter), and rotor speed 3. To that end decreasing rotor speed would need to be compensated by increasing wind turbine size and vice versa for producing the same amount of energy. Thus, altering wind turbine features will not minimize impacts on bats, as long as the energy produced remains the same or even increases. In addition, having several small or fewer large wind turbines is unlikely to minimize deaths as long as total produced energy remains constant. Avian and bat mortality was reported to be constant per energy unit produced across wind turbine power capacities in a study in the USA 4, confirming that not simply the size of turbines but power is the factor better explaining deaths.
Conclusions:
Among all wind turbine features and land cover characteristics, wind turbine power is the most significant factor associated to bat deaths. Wind turbines located within 5 km buffer comprised of natural land cover have substantial higher deaths and should not be licensed. More wind turbine power will result in more deaths. Unrecorded deaths are > 500% higher than the recorded ones.
Wind turbines are directly related with land loss and degradation by artificial base surfaces and road-induced fragmentation. Ranking of biodiversity threats in Europe and North America indicates that historically threats deriving from land use changes are several folds higher than the ones from climatic changes 5, with comparable results in other studies 6. In the most recent assessments, land-use change is still the biggest current threat to nature, destroying or fragmenting the natural habitats of many plant and animal species on land, in freshwater and in the sea 7.
A paradox emerges of impacting biodiversity to mitigate the climate: biodiversity, through the ecosystem services it supports, makes an important contribution to both climate mitigation and adaptation 8. It thus needs to be prioritized what is the relative gain and loss of wind farm installation and operation in the climate-land-biodiversity-energy nexus.
References:
1 Moustakas, A., Georgiakakis, P., Kret, E. & Kapsalis, E. Wind turbine power and land cover effects on cumulative bat deaths. Science of The Total Environment 892, 164536, doi: https://doi.org/10.1016/j.scitotenv.2023.164536 (2023).
2 MacGregor, K. A. & Lemaître, J. The management utility of large-scale environmental drivers of bat mortality at wind energy facilities: The effects of facility size, elevation and geographic location. Global Ecology and Conservation 21, e00871, doi: https://doi.org/10.1016/j.gecco.2019.e00871 (2020).
3 Dixon, S. & Hall, C. Wind Turbines. Fluid Mechanics and Thermodynamics of Turbomachinery, 419-485 (2014).
4 Huso, M. et al. Relative energy production determines effect of repowering on wildlife mortality at wind energy facilities. Journal of Applied Ecology 58, 1284-1290 (2021).
5 Almond, R. E. A., Grooten, M. & Peterson, T. Living Planet Report 2020-Bending the curve of biodiversity loss. WWF, Gland, Switzerland. (World Wildlife Fund, 2020).
6 IPBES. Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. (Bonn, Germany, 2019).
7 Almond, R.E.A., Grooten, M., Juffe Bignoli, D., Petersen, T., 2022. Living Planet Report 2022 –
Building a Naturepositive Society. WWF, Gland, Switzerland
8 EuropeanCommission. WHITE PAPER: Adapting to climate change: Towards a European framework for action. (Commission of the European Communities, 2009).