Overlap and Segregation among Multiple 3D Home Ranges: A Non-Pairwise Metric with Demonstrative Application to the Lesser Kestrel Falco naumanni

SIMPLE SUMMARY: Modeling animal space use in 3D is more realistic than confining research to 2D methods and can greatly increase our understanding of interspecific and intraspecific competition, predator–prey relationships, habitat selection and use. In particular, home range overlap/segregation is a fundamental property of animal interactions with deep implications for biodiversity conservation and management. In order to solve the issue of measuring the degree of overlap/segregation among an arbitrarily large number of 3D volumetric home ranges, we introduced the novel non-pairwise index MVOI (Multiple Volumetric Overlap Index) and its complement to 100 MVSI (Multiple Volumetric Segregation Index). Results show that traditional 2D spatial analyses can significantly overestimate the overlap between the individuals, population and species that occupy a habitat with a strong vertical component. Both the MVOI and MVSI can also be used to quantify how the 3D home ranges change over time (i.e., 4D home ranges) and the robustness of 3D home range assessment through the degree of overlap among different 3D estimators. We applied the MVOI and MVSI to birds, but they can be readily applied to any animal species, in particular those with a significant vertical component to their space use. ABSTRACT: In this study we solved the issue of measuring the degree of overlap/segregation among an arbitrarily large number (n ≥ 2) of 3D volumetric home ranges (i.e., x, y, and h(g); where h(g) is height above ground level) for the first time. For this purpose, we introduced the novel non-pairwise index MVOI (Multiple Volumetric Overlap Index) and its complement to 100 MVSI (Multiple Volumetric Segregation Index). Regardless of the number of 3D volumetric home ranges, the MVOI and MVSI generate a single score of overlap/segregation between 0 and 100, making ecological interpretation much easier and more meaningful when compared to n × n pairwise overlap indices. As a case study, we applied the MVOI and MVSI to 12,081 GPS points of five lesser kestrels (Falco naumanni) during the nesting period at Santeramo in Colle (Apulia region; Italy) in an area with the most elevated density of lesser kestrels in urban colonies worldwide. The 3D volumetric home ranges ranged between 1.79 km(3) and 8.19 km(3). We found that the tracked birds had different vertical profiles, possibly to limit intraspecific competition, resulting in a 3D home range overlap that was only 61.1% of the 2D overlap and 52.8% of the probabilistic one.