Predictive modeling for allopatric Strix (Strigiformes: Strigidae) owls in South America: determinants of their distributions and ecological niche-based processes

Abstract

Strix (Strigidae) is a worldwide genus of 17 owl species typical of forested habitats, including Rusty-barred Owls (S. hylophila), Chaco Owls (S. chacoensis), and Rufous-legged Owls (S. rufipes) in South America. These species are distributed allopatrically, but the ecological traits that determine their distributions remain largely unknown and their phylogenetic relationships are unclear. We used species distribution models (SDMs) to identify variables explaining their distribution patterns and test hypotheses about ecological divergence and conservatism based on niche overlap analysis. For Rusty-barred Owls and Chaco Owls, climatic factors related to temperature played a major role, whereas a rainfall variable was more important for Rufous-legged Owls. When niche overlaps were compared, accounting for regional similarities in the habitat available to each species, an ecological niche divergence process was supported for Chaco Owl-Rusty-barred Owl and Chaco Owl-Rufous-legged Owl, whereas a niche conservatism process was supported for Rusty-barred Owl-Rufous-legged Owl. Different ecological requirements support current species delimitation, but they are in disagreement with the two main hypotheses currently envisaged about their phylogenetic relationships (Chaco Owls as the sister taxa of either Rufous-legged Owls or Rusty-barred Owls) and support a new phylogenetic hypothesis (Rufous-legged Owls as sister taxa of Rusty-barred Owls). Our findings suggest that speciation of Rusty-barred Owls and Rufous-legged Owls was a vicariant event resulting from Atlantic marine transgressions in southern South America in the Miocene, but their niche was conserved because habitat changed little in their respective ranges. In contrast, Chaco Owls diverged ecologically from the other two species as a result of their adaptations to the habitat they currently occupy. Ecological and historical approaches in biogeography can be embedded to explain distribution patterns, and results provided by SDMs can be used to infer historical and ecological processes in an integrative way.