Supporting Information - 1

Supporting Information - 1 Projecting Global Land Use Driven Evolutionary History Loss Abhishek Chaudhary1*, Vahab Pourfaraj2, Arne O. Mooers2 1

Institute of Food, Nutrition and Health, ETH Zurich, Schmelzbergstrasse 9, Zurich 8092, Switzerland (*Corresponding author: e-mail: [email protected]; phone: +41 79 170 1766; fax: +41 44 632 1155) 2

Department of Biological Sciences and IRMACS, Simon Fraser University, Burnaby, British Columbia, Canada V5A1S6

Calculation of evolutionary distinctiveness (ED) scores of each species To compare PD loss with cumulative ED loss from random pruning on the three clades used here (mammals, birds, and amphibians), we needed taxonomically complete trees. For all three taxa, these complete trees are actually tree distributions that account for possible placement of taxa that can only be placed approximately in a dated phylogeny based on taxonomy (Martyn et al., 2012; Jetz et al., 2014 and Isaac et al., 2012). For mammals, we took the distribution of 1000 trees used to generate the evolutionary indices in Martyn et al. (2012). We re-calculated ED scores (see below) for these 5139 species across the 1000 trees. For birds, we calculated ED scores using an updated distribution of phylogenies (K. Magnuson-Ford, pers. comm.), incorporating a total of 10,284 extant bird species as recognized by the IUCN, based on the distribution of "Hackett backbone" trees used by Jetz et al. (2014). Updates primarily involved the splitting of one named species into two or more. The bird phylogenies used by Jetz et al. (2014) were modified to be consistent with the 2016 IUCN Red List taxonomy (BirdLife International 2015). Thus, (1) the scientific names of 561 species on the phylogeny were changed to match those of the IUCN Red List, (2) 136 species from the phylogeny that were not recognized by the IUCN Red List were removed, and (3) 427 species (details below) that are currently recognized by the IUCN Red List (BirdLife International 2015 (checklist version 8)) as extant but are not on the Jetz et al. (2014) tree were added. These changes were applied to the complete distribution of 10,000 trees, each which now include 10,284 species, fully consistent with the IUCN Red List taxonomy. Of the 427 species added, 405 were species that had recently been elevated from subspecies to species status, i.e. splitting from a known sister species. In these cases, the new species was inserted into the phylogeny halfway down the branch of its sister. In cases where one species was split into more than two species, the topology of the new clade (including the sister already existing in the tree) was generated randomly (rtree in the R package ape) and all branch lengths were set as equal. This clade was then inserted into the tree, replacing the original sister branch and the branch lengths of the pendant edges were increased such that the tree remained ultrametric.

The remaining 22 species added to the tree included 11 newly described species, and 11 species not recognized by Jetz et al. (2014). These 22 species were added to the tree as described above, where sister species were identified based on BirdLife species factsheets or recent scientific literature. The only exception was Heliangelus zusii, which was inserted halfway down the branch leading to the clade containing the Taphrolesbia and Aglaiocercus genera: BirdLife states in its fact sheet “Heliangelus spp. typically occur in cloud-forest and shrubbery at elevations of 1,200-3,400 m, mostly at 1,400-2,200 m. This species is probably more closely allied to Aglaiocercus and Taphrolesbia, and should be sought in humid or semi-arid habitat as high as 3,200 m from northwestern Venezuela to northern Peru (Kirchman et al., 2010).” Finally, for the amphibians, we used the polytomy rich amphibian tree used by Isaac et al. (2012), which incorporated 5713 species. We applied the method described by Kuhn et al. (2011) to resolve the polytomies in the original tree to produce a distribution of 10,000 trees. We used BEAST version 1.7.5. (Drummond et al., 2012), sampling every 1000 trees and setting the burn-in rate at 10%. Calculation of ED scores: For each of the three clades, we calculated ED scores for each tree in these distributions. For each species on a tree, ED is the weighted sum from the root to the tip of phylogeny (Redding, 2003) using caper (Orme et al. 2013) package of R: 𝑟

𝐸𝐷 (𝑇, 𝑖) =

∑ 𝑒 ∈(𝑇,𝑖,𝑟)

λ𝒆 𝑐𝑒

Where 𝑖 is the species of interest, λe is the edge length divided by number of species (ce ) it subtends. We compared PD loss to cumulative ED loss under random extinction separately for the three taxa using the following protocol. For each tree from our posterior distribution, we calculated the ED score for each species. We then randomly pruned species iteratively from that tree using the drop.tip function in the R package ape (Paradis et al., 2004), from one up to 15% (771, 1543 and 857 species for the mammals, birds and amphibians respectively), recording the summed ED loss and the attendant PD loss from that tree. We then used the lm function in the stats package of R (R Core Team, 2016) across the set of pruned trees to generate a single estimate of the slope describing the relationship between the two measures (Figure S1 below). The mean ED scores for each species across the posterior distributions for all three taxon are provided in Table S1 of Supporting Information-2 excel file.

Figure S1. Relationship between cumulative ED and PD loss for random loss of species, up to 15% species loss for a) mammals, b) birds and c) amphibians. In all panels the main line indicates the slope from a linear model that considers pruning across each of 1000 trees, with shading around the line corresponding to density of individual measures across these pruned trees Standard deviation (SD), coefficient of determination (r2) and pvalues are also shown for the fitted linear regression line.

References: BirdLife International (2015) Available at: http://www.birdlife.org/datazone/userfiles /file/Species/Taxonomy/BirdLife_Checklist_Version_80.zip (.xls zipped 1 MB). Drummond A.J., Suchard M.A., Xie D. & Rambaut A. (2012) Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29, 1969– 1973. Isaac N.J.B., Redding D.W., Meredith H.M. & Safi K. (2012) Phylogenetically-Informed Priorities for Amphibian Conservation. PLoS One, 7, 1–8. Jetz W., Thomas G.H., Joy J.B., Redding D.W., Hartmann K. & Mooers A.O. (2014) Global Distribution and Conservation of Evolutionary Distinctness in Birds. Current Biology, 24, 919–930. Kirchman J.J., Witt C.C., McGuire J.A. & Graves G.R. (2010) DNA from a 100-year-old holotype confirms the validity of a potentially extinct hummingbird species. Biology letters, 6, 112–5. Kuhn, T.S., Mooers, A.Ø. & Thomas, G.H. (2011) A simple polytomy resolver for dated phylogenies. Methods in Ecology and Evolution, 2(5), 427-436. Martyn I., Kuhn T.S., Mooers A.O., Moulton V., & Spillner A. (2012) Computing evolutionary distinctiveness indices in large scale analysis. Algorithms for molecular biology, 7, 6. Orme D., Freckleton R., Thomas G., Petzoldt T., Fritz S., Isaac N., & Pearse W. (2013) caper: Comparative Analyses of Phylogenetics and Evolution in R. R package version 0.5.2. . Paradis E., Claude J. & Strimmer K. (2004) APE: Analyses of Phylogenetics and Evolution in R language. Bioinformatics, 20, 289–290. R Core Team (2016) R: A Language and Environment for Statistical Computing. . Redding D. (2003) Incorporating Genetic Distinctness and Reserve Occupancy into a Conservation Prioritization Approach. Master Thesis. University Of East Anglia, Norwich UK.