Global conservation of phylogenetic diversity captures more than just functional diversity

Abstract

The biodiversity measure, phylogenetic diversity (PD), links evolutionary history to the conservation of feature-diversity (broadly, the different evolutionary features of species), and so to future options for humanity (option value)1. Mazel et al.2 claim that (1) PD must perform better than random in capturing functional diversity (FD) if it is to have any validity for conservation; (2) PD captures FD unreliably; and (3) we may need to abandon the use of PD in conservation, depending on the outcome of further FD randomisation tests. We argue that Mazel et al.2 misrepresent the PD/feature-diversity framework as restricted to functional traits, and we illustrate how a conservation focus on functional traits could lead to the global loss of PD, feature-diversity and option values. The core rationale for PD conservation initiatives, such as the EDGE of Existence programme, should continue to build on the link from PD to broad feature-diversity1,3,4, not a link to a few nominated functional traits. Mazel et al.2 explicitly point to Faith’s1 original broad featurediversity arguments in describing the idea that conserving PD will conserve a wide variety of forms and functions. However, Mazel et al.2 then adopt a narrower functional perspective, asserting that such diversity can be measured as FD, calculated using selected traits of assumed ecological relevance (e.g., four mammalian traits: diet, body mass, activity cycle, foraging height). Mazel et al.2 incorrectly synonymise FD with feature-diversity by misrepresenting Faith’s reference to feature-diversity1 as a reference to FD, and Faith’s reference to future options arising from feature-diversity1 as a reference to future options from FD. This misrepresentation underpins their false claim: “the fundamental phylogenetic gambit at the heart of all PD-based conservation strategies ...[is that]... maximizing PD captures more FD than randomly choosing species.” Because they incorrectly equate PD’s broad feature-diversity with their narrowly defined FD, Mazel et al.2 have no justification for this claim that failure to recover FD in their randomisation tests casts doubt on all PD conservation initiatives. One such PD conservation initiative, the EDGE of Existence programme5, is characterised by Mazel et al.2 as following the logic of their FD phylogenetic gambit. In reality, an original rationale for EDGE explicitly drew on Faith’s1 feature-diversity arguments: Isaac et al.5 noted that “phylogenetic diversity is clearly related to character diversity”, arguing that PD-based scores indicate unique features and potential future utilitarian value. This general feature-diversity/future options link has supported the prioritisation of EDGE species for conservation, including those that might otherwise be neglected. The EDGE rationale echoes the history of studies on how phylogeny is informative about both known and unknown features of organisms3,6. Early studies examining character/feature data and phylogenies3,4 also provided tests corroborating the specific rationale for PD—that shared ancestry can generally account for shared features among species. Mazel et al.’s2 test of their FD gambit does not assess whether PD captures a wide range of features; indeed, a perfect PD-feature relationship can produce a failure in their test7. The well-corroborated PD-features relationship3,4 has supported the use of EDGE information by the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), as one indicator for “maintenance of options” (one of “nature’s contributions to people”)8. IPBES reports global estimates of total imperilled PD, over six major taxonomic groups8. The IPBES Asia-Pacific assessment8 provides recent examples of surprising global benefits that have been discovered, illustrating the broad range of these often-unfamiliar evolutionary features. These include the discovery that funnel-web spider venom (Hadronyche infensa) is the unlikely source for medication to avoid brain damage caused by strokes9, and that a substance in Tasmanian Devil milk (Sarcophilus harrisii) fights antibioticresistant bacteria10. This unusual mammal feature illustrates the scope of feature-diversity, in contrast to Mazel et al.’s2 limited four ecological traits. Other examples highlight how the feature of interest originated in an ancestral lineage (e.g. medicinal features in plants11), further corroborating the PD rationale. The EDGE of Existence programme interprets PD and featurediversity as indicating multiple values of biodiversity. A species’ https://doi.org/10.1038/s41467-019-08600-8 OPEN