Inferring inter‑colony movement within metapopulations of yellow‑footed rock‑wallabies using estimates of kinship

Abstract

<p>Understanding the exchange of individuals between wildlife populations, particularly those with naturally fragmented habitats, is important for the effective management of these species. This is of particular consequence when the species is of conservation concern, and isolated populations may be lost due to pressures from predation or competition, or catastrophic events such as wildfire. Here we demonstrate the use kinship and population structure analysis to show potential recent movement between colonies in metapopulations of yellow-footed rock-wallaby (<i>Petrogale xanthopus</i> Gray 1854) at two sites in the Grey Range of Queensland, and at four sites in the Gawler Ranges of South Australia. These colonies are also compared to a single colony from the Flinders Ranges, a connected landscape of rock-wallaby habitat. Using reduced representation next-generation sequencing, we acquired and filtered a set of similar to 17,000 single-nucleotide polymorphisms to examine population genetic variation, structure and relationships within populations, and also identify putative migrants. Initial STRUCTURE analysis re-confirmed each population should be considered separately. Tests of population genetic variation identify several colonies appearing to be experiencing genetic erosion, also with low calculated effective population sizes (N_e = 4.5-36.6). Pairwise comparisons of individual relatedness (relatedness coeffiecients; <i>r</i>) implied several contemporary movement events between colonies within both the Gawler and Grey Ranges (<i>r</i> > 0.125), which was then affirmed with tests for putative first generation migrants. These results are of particular note in South Australia, where threat abatement (management of key predators and competitors) may facilitate dispersion. Additionally, in Queensland, colonies are separated by anthropogenic barriers: predator exclusion fencing designed to exclude dingoes (<i>Canis familiaris</i>) from grazing land, which may hinder dispersal. This work highlights the usefulness of population genetics to inform management outcomes in wildlife, in this case, highlighting the need for threatened species management at the landscape level.</p>