Phylogeny, Traits, and Biodiversity of a Neotropical Bat Assemblage: Close Relatives Show Similar Responses to Local Deforestation

Abstract

If species' evolutionary pasts predetermine their responses to evolutionarily novel stressors, then phylogeny could predict species survival in an increasingly human-dominated world. To understand the role of phylogenetic relatedness in structuring responses to rapid environmental change, we focused on assemblages of Neotropical bats, an ecologically diverse and functionally important group. We examined how taxonomic and phylogenetic diversity shift between tropical forest and farmland. We then explored the importance of evolutionary history by ascertaining whether close relatives share similar responses to environmental change and which species traits might mediate these trends. We analyzed a 5-year data set (5,011 captures) from 18 sites in a countryside landscape in southern Costa Rica using statistical models that account and correct for imperfect detection of species across sites, spatial autocorrelation, and consideration of spatial scale. Taxonomic and phylogenetic diversity decreased with deforestation, and assemblages became more phylogenetically clustered. Species' responses to deforestation were strongly phylogenetically correlated. Body mass and absolute wing loading explained a substantial portion of species variation in species' habitat preferences, likely related to these traits' influence on maneuverability in cluttered forest environments. Our findings highlight the role that evolutionary history plays in determining which species will survive human impacts and the need to consider diversity metrics, evolutionary history, and traits together when making predictions about species persistence for conservation or ecosystem functioning.

Precipitation and tree cover gradients structure avian alpha diversity in North‐western Costa Rica

“…In the tropics, conversion of forests to agriculture and climate drying are expected to cause drastic ecosystem‐level changes (e.g., Malhi et al, ). Indeed, the conversion of forests to agricultural lands leads to non‐random extirpation patterns across multiple taxa (Flynn et al, ; Newbold et al, ), including birds (Karp et al, ; Frishkoff et al., ), bats (Cisneros, Fagan, & Willig, ; Frank, Frishkoff, Mendenhall, Daily, & Hadly, ) and amphibians (Nowakowski, Frishkoff, Thompson, Smith, & Todd, ). Likewise, changes in climate threaten and favour certain species over others (Thomas et al, ; Thuiller et al, ; Urban, ).…”

Evolutionarily distinct amphibians are disproportionately lost from human‐modified ecosystems

“…), which in turn may be conserved across evolutionary history (Frank et al . ; Nowakowski et al . ).…”

“…Susceptibility to human land conversion has been shown to vary considerably across species and taxonomic groups (Gibson et al 2011;Keinath et al 2017;Pfeifer et al 2017), with some species not only persisting but also thriving in anthropogenic habitats (McKinney & Lockwood 1999;Thomas 2017). The propensity to perish or persist in human-modified ecosystems may be due to aspects of species' ecological and life history traits (Newbold et al 2013;Frishkoff et al 2015), which in turn may be conserved across evolutionary history (Frank et al 2017;Nowakowski et al 2018a). Evolutionary patterns in species' ability to persist in the face of humaninduced environmental change have also emerged for certain taxa: across several systems birds that are evolutionarily distinct tend to be disproportionately lost from human-modified agricultural (Frishkoff et al 2014;Edwards et al 2015) and urbanised sites (Morelli et al 2016;Ib añez-Alamo et al 2017;Sol et al 2017).…”

Integrating over uncertainty in spatial scale of response within multispecies occupancy models yields more accurate assessments of community composition

“…mean detectability of 0.4 for a study of bats (Frank et al 2017) and 0.26 for a study of birds (Carrillo-Rubio et al 2014)). 4) Low detection with detection affected by local environment: average detectability approximately 0.25, but rising to approximately 0.5 under high local values of E and dropping to approximately 0.1 under low local values of E (denoted 'env' below).…”

“…For most species detectability will be far below perfect (e.g. mean detectability of 0.4 for a study of bats (Frank et al 2017) and 0.26 for a study of birds (Carrillo-Rubio et al 2014)). However, detection may approach perfect for some sessile organisms.…”