Comparing habitat connectivity of 2D versus 3D-landscape representations for brown bear (Ursus arctos) and European bison (Bison bonasus L.) in the Northern Carpathians

The loss of ecological connectivity in increasingly fragmented and human-dominated landscapes threatens the long-term persistence of many species, especially large mammals. Habitat connectivity modelling typically relies on 2D representations of landscape and vegetation structure (Ziółkowska et al., 2014). While, the inclusion of 3D landscape and vegetation structure representations has been shown to improve the predictive power of such models (Dymytrova et al., 2016; Huber et al., 2016; Zellweger et al., 2016), the value of 3D representations have rarely been tested across larger regions.

The Northern Carpathians, particularly the border zone between Slovakia and Poland (a 40 km band into each country along the country border), are key for many populations of large mammals threatened elsewhere in Europe. Connectivity assessments in this region have typically been conducted for one of the two countries only, and using coarse, 2D land-cover data.

In this study therefore, we evaluate 3D versus 2D landscape representations and vegetation structure and their role in connectivity assessments for the Northern Carpathians of Slovakia and Poland, for two species: the brown bear (Ursus arctos) and European bison (Bison bonasus L.). In particular, we derive a 3D representation of vegetation structure from LIDAR (Light Detection and Ranging) data to examine its influence and importance for the delineation of individual habitat patches and movement corridors. Our connectivity assessment relied on graph- and circuit- theoretical approaches, calibrated with either the 2D or 3D landscape and vegetation structure representation.

Comparisons showed that shifts towards the 3D vegetation structure representation in habitat connectivity models have the potential to greatly affect composition and configuration of corridors (lengths, effective distance, and spatial location), but not the importance of patch level habitat quality rankings for both mammals. Moreover, our 3D connectivity assessment for the brown bear (Ursus arctos) and European bison (Bison bonasus L.) should substantially contribute to solving urgent conservation issues related to these species.

Acknowledgments

We gratefully acknowledge support by the Visegrad Fund: project “Transboundary ecological connectivity – modelling landscapes and ecological flows, CON@SK.PL, No. 21640051.