Figure 1.
Using pairwise distances to rank species or populations.
Consider a hypothetical group of taxa (A)—a set of closely-related species or populations of a single species—that is distributed across several islands in an archipelago (B). Differences among the taxa, labeled x1 through x6, can be organized into a pairwise distance matrix (C). We can represent this matrix either as a phylogenetic tree or as a phylogenetic network (D), where a set of weighted splits describes the relationships among the taxa (E). Altogether, these splits represent the group's phylogenetic diversity (PD). By selecting subsets of splits that exclude a given taxon, we can calculate each taxon's contribution to the total PD of the tree or network (F). The Shapley metric (SH) and expected PD complementarity (PD_c) are different approaches for ranking taxa based on split data. Note that the highest-scoring taxa (highlighted values) can differ considerably depending on the type of metric used and whether the splits come from a tree or network. We discuss the reasons for these differences and methods for ranking taxa in Section (ii) of the main text.
Figure 2.
Conservation prioritization of spotted owl (Strix occidentalis) populations.
(a) Distribution of spotted owls in the United States and the populations sampled by Barrowclough et al. [48], [51]. Shaded areas denote suitable habitat based on forest cover data [73]. Colors denote the subspecies S. o. caurina (blue), S. o. occidentalis (green), and S. o. lucida (orange). Populations 31 and 32 represent the S. o. juanaphillipsae subspecies in Mexico (range not shown). (b) NeighborNet of sampled populations based on mtDNA differentiation (pairwise ΦST values). (c) Histogram of SH values, highlighting the populations with the highest scores. See Table 1 for an explanation of abbreviations used.
Table 1.
Spotted owl populations sampled by Barrowclough et al. [48], [51] and ranked by Shapley value (SH) and heightened evolutionary distinctiveness (HED).
Figure 3.
Conservation prioritization of mountain pygmy-possum (Burramys parvus) populations.
(a) Distribution of mountain pygmy-possums in Australia (gray inset), showing populations sampled by Mitrovski et al. [58]. Shaded areas denote suitable habitat above 1,400 m. (b) NeighborNet of sampled populations based on microsatellite differentiation (pairwise FST values). (c) Histograms of SH and HED values, highlighting the populations with the highest scores. See Table 2 for an explanation of abbreviations used.
Table 2.
Mountain pygmy-possum populations sampled by Mitrovski et al. [58] and ranked by Shapley value (SH) and heightened evolutionary distinctiveness (HED).