Abstract
Understanding how to prioritize among the most deserving imperilled species has been a focus of biodiversity science for the past three decades. Though global metrics that integrate evolutionary history and likelihood of loss have been successfully implemented, conservation is typically carried out at sub-global scales on communities of species rather than among members of complete taxonomic assemblages. Whether and how global measures map to a local scale has received little scrutiny. At a local scale, conservation-relevant assemblages of species are likely to be made up of relatively few species spread across a large phylogenetic tree, and as a consequence there are potentially relatively large amounts of evolutionary history at stake. We ask to what extent global metrics of evolutionary history are useful for conservation priority setting at the community level by evaluating the extent to which three global measures of evolutionary isolation (evolutionary distinctiveness (ED), average pairwise distance (APD) and the pendant edge or unique phylogenetic diversity (PD) contribution) capture community-level phylogenetic and trait diversity for a large sample of Neotropical and Nearctic bird communities. We find that prioritizing the most ED species globally safeguards more than twice the total PD of local communities on average, but that this does not translate into increased local trait diversity. By contrast, global APD is strongly related to the APD of those same species at the community level, and prioritizing these species also safeguards local PD and trait diversity. The next step for biologists is to understand the variation in the concordance of global and local level scores and what this means for conservation priorities: we need more directed research on the use of different measures of evolutionary isolation to determine which might best capture desirable aspects of biodiversity.
Highlights
Making informed decisions about the appropriate focus of conservation investment has become a central theme of both academic research and conservation action
Removing the top 500 high scoring global average pairwise distance (APD) species from all communities resulted in reduction in an average mean trait pairwise distance of 8%, meaning that the remaining species tended to be more similar in terms of their functional traits compared with the unperturbed community
There appears to be potential for close agreement between different metrics used to create conservation priorities at APD ED PE APD ED PE
Summary
Making informed decisions about the appropriate focus of conservation investment has become a central theme of both academic research and conservation action. We calculated the median of each of three measures of evolutionary isolation across the 10 000 phylogenies, each at three scales: the length of the terminal branch [20,21] linking a species to the tree (unique PD contribution, called phylogenetic endemism or pendant edge (PE) cophenetic, R package ape [22]); the APD, i.e. the mean APD to all other species in the tree (cophenetic, R package ape [22]) and the fair proportion measure of ED (ed.calc, R package caper [23]) These metrics include the two ends of an axis of evolutionary isolation measures that weights information nearer the root (APD) or nearer the tip (PE [24]), alongside the only currently used measure in active conservation prioritization (ED: [6,11]). We selected the top scoring 500 species for each global metric and determined how many of these top-ranking species were in each community We removed these species from each community they were found in and measured biodiversity change in terms of mean reduction in local PD and change to mean average trait pairwise difference (measured using TAPD). We removed the same number of species randomly from each community (replicated 500 times), to create an expectation of PD lost and TAPD change given random loss of species from communities
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
