Erosion of phylogenetic diversity in Neotropical bat assemblages: findings from a whole-ecosystem fragmentation experiment

The traditional focus on taxonomic diversity metrics for investigating species responses to habitat loss and fragmentation has limited our understanding of how biodiversity is impacted by habitat modification. This is particularly true for taxonomic groups such as bats which exhibit species-specific responses. Here, we investigate phylogenetic alpha and beta diversity of Neotropical bat assemblages across two environmental gradients, one in habitat quality and one in habitat amount. We surveyed bats in 39 sites located across a whole-ecosystem fragmentation experiment in the Brazilian Amazon, representing a gradient of habitat quality (interior-edge-matrix, hereafter IEM) in both continuous forest and forest fragments of different sizes (1, 10, and 100 ha; forest size gradient). For each habitat category, we quantified alpha and beta phylogenetic diversity, then used linear mixed-effects models and cluster analysis to explore how forest area and IEM gradient affect phylogenetic diversity. We found that the secondary forest matrix harboured significantly lower total evolutionary history compared to the fragment interiors, especially the matrix near the 1 ha fragments, containing bat assemblages with more closely related species. Forest fragments ≥ 10 ha had levels of phylogenetic richness similar to continuous forest, suggesting that large fragments retain considerable levels of evolutionary history. The edge and matrix adjacent to large fragments tend to have closely related lineages nonetheless, suggesting phylogenetic homogenization in these IEM gradient categories. Thus, despite the high mobility of bats, fragmentation still induces considerable levels of erosion of phylogenetic diversity, suggesting that the full amount of evolutionary history might not be able to persist in present-day human-modified landscapes.