The species-specificity of energy landscapes for soaring birds, and its consequences for transferring suitability models across species

Context Soaring birds depend on atmospheric uplifts and are sensitive to wind energy development. Predictive modelling is instrumental to forecast conflicts between human infrastructures and single species of concern. However, as multiple species often coexist in the same area, we need to overcome the limitations of single species approaches. Objectives We investigate whether predictive models of flight behaviour can be transferred across species boundaries. Methods We analysed movement data from 57 white storks, Ciconia ciconia, and 27 griffon vultures, Gyps fulvus. We quantified the accuracy of topographic features, correlates of collision risk in soaring birds, in predicting their soaring behaviour, and tested the transferability of the resulting suitability models across species. Results 59.9% of the total area was predicted to be suitable to vultures only, and 1.2% exclusively to storks. Only 20.5% of the study area was suitable to both species to soar, suggesting the existence of species-specific requirements in the use of the landscape for soaring. Topography alone could accurately predict 75% of the soaring opportunities available to storks across Europe, but was less efficient for vultures (63%). While storks relied on uplift occurrence, vultures relied on uplift quality, needing stronger uplifts to support their higher body mass and wing loading. Conclusions Energy landscapes are species-specific and more knowledge is required to accurately predict the behaviour of highly specialised soaring species, such as vultures. Our models provide a base to explore the effects of landscape changes on the flight behaviour of different soaring species. Our results suggest that there is no reliable and responsible way to shortcut risk assessment in areas where multiple species might be at risk by anthropogenic structures.

Automated summary

This study examines the transferability of predictive models of flight behavior across species boundaries. Using movement data from white storks and griffon vultures, the study finds that topographic features can accurately predict soaring behavior for storks, but less efficiently for vultures. The study emphasizes the need for more research to accurately predict the behavior of highly specialized soaring species.