Mortality limits used in wind energy impact assessment underestimate impacts of wind farms on bird populations

Abstract

1. The consequences of bird mortality caused by collisions with wind turbines are increasingly receiving attention. So-called acceptable mortality limits of populations, that is, those that assume that 1%–5% of additional mortality and the potential biological removal (PBR), provide seemingly clear-cut methods for establishing the reduction in population viability. 

2. We examine how the application of these commonly used mortality limits could affect populations of the Common Starling, Black-tailed Godwit, Marsh Harrier, Eurasian Spoonbill, White Stork, Common Tern, and White-tailed Eagle using stochastic density-independent and density-dependent Leslie matrix models.

3. Results show that population viability can be very sensitive to proportionally small increases in mortality. Rather than having a negligible effect, we found that a 1% additional mortality in post-fledging cohorts of our studied populations resulted in a 2%–24% decrease in the population level after 10 years. Allowing a 5% mortality increase to existing mortality resulted in a 9%–77% reduction in the populations after 10 years.

4. When the PBR method is used in the density-dependent simulations, the proportional change in the resulting growth rate and carrying capacity was species-independent and largely determined by the recovery factor (Fr). When Fr = 1, a value typically used for robust populations, additional mortality resulted in a 50%–55% reduction in the equilibrium density and the resulting growth rate. When Fr = 0.1, used for threatened populations, the reduction in the equilibrium density and growth rate was about 5%.

5. Synthesis and applications. Our results show that by allowing a mortality increase from wind farm collisions according to both criteria, the population impacts of these collisions can still be severe. We propose a simple new method as an alternative that was able to estimate mortality impacts of age-structured stochastic density-dependent matrix models.

CONCLUSIONS

We show that the use of the ORNIS 1%, the 5% mortality criterion, and potential biological removal criteria are inadequate for providing safe thresholds with respect to the impact of wind turbine collisions on populations. The responses of the population to a mortality increase are generally much higher than the mortality increase itself, whereas the PBR method is determined by the recovery factor Fr. We propose a new method presented by Equation (4) as a viable alternative, provided that r0 can be estimated, is positive, and the population growth is density-dependent. This method directly relates population properties summarized in r0 and the allowed reduction in the equilibrium density of a population to the allowed mortality fraction. Also, it appears robust for all the used matrix approaches. 

When a population growth is not density-dependent or the population is declining, we propose the use of population viability analysis for more in-depth studies of such impacts. Any additional mortality reduces the population's buffer capacity to recover from any stochastic or structural hazard, and thereby increases the risk of extinction. Particularly, those populations with low-density dependence (low r0) are very sensitive to even a small increase in mortality. In declining populations, in which the recruitment rate is lower than the mortality rate, no “acceptable” additional mortality levels exist, as even a small increase in mortality leads to faster extinction. Finally, we should account for cumulative effects of multiple wind farms, as bird populations encounter increasing numbers of wind farms.