Limited influence of landscape on the genetic structure of three small mammals in a heterogeneous arid environment

Abstract

Aim:The complexity and biologically challenging nature of arid landscapes can generate high inter-and intra-species diversity, although these biomes remain poorly studied. We investigated whether prominent geomorphic features in an Australian arid landscape had similar influences on patterns of intra-specific genetic diversity of three small mammals with different distribution and life history traits. Specifically, we tested (a) whether mountain ranges harbour high diversity and act as refugia, (b) the extent to which ephemeral river systems pose barriers to gene flow or (c) whether environmental differences explain genetic structure in these species. Location: The topographically complex Pilbara bioregion of the Australian arid zone. Taxa: Two rodents, the Western pebble-mound mouse, Pseudomys chapmani and the Sandy inland mouse, Pseudomys hermannsburgensis and a dasyurid, Ningaui timealeyi. Methods: We used maximum parsimony networks of mitochondrial haplotypes and spatially-explicit Bayesian clustering analysis of microsatellite genotypes to identify historical and contemporary genetic structure in each species. Based on measures of individual diversity and genetic distance, we distinguished 'hotspots' and 'coldspots' (>1.5 standard deviations from the mean) of genetic diversity and genetic connectivity and assessed their concordance with landscape features. The influence of environmental dissimilarity on observed genetic differentiation was evaluated using generalized dissimilarity modelling.Results: Patterns of genetic diversity and connectivity were complex, with few consistent patterns across species. Both microsatellite and mitochondrial diversity were high across species, with limited spatial genetic structure. Rivers had the most influence on genetic patterns but were not consistently identified as barriers. Environmental dissimilarity amongst sampling sites had low explanatory power. Main Conclusions:Limited genetic structure in each of the three species suggests that the structurally complex Pilbara landscape remains relatively permeable to species with high dispersal capacity. Additionally, population responses to boom-bust cycles in arid environments are likely to obscure spatial genetic patterns. K E Y W O R D Sarid zone, genetic structure, landscape, mammal, microsatellites, phylogeography, Pilbara

Multi-Species Phylogeography of Arid-Zone Sminthopsinae (Marsupialia: Dasyuridae) Reveals Evidence of Refugia and Population Expansion in Response to Quaternary Change

“…Our results add support to a growing number of studies that have found low population differentiation and potential panmixia within mammal species occurring in the Pilbara and other arid regions of Australia [ 66 , 68 , 69 ]. A recent study on three small mammals occurring in the Pilbara, two widespread habitat generalists (a murid, Pseudomys hermansburgensis and dasyurid, Ningaui timealeyi ) and a pebble surface specialist mouse ( Pseudomys chapmani ) found little evidence for population structure using mitochondrial and microsatellite sequence data [ 68 ]. Levy et al [ 68 ] proposed this resulted from the high fecundity and dispersal potential of these species, resulting in panmixia despite the complex geological landscape of the Pilbara presenting numerous potential barriers to gene flow.…”

“…A recent study on three small mammals occurring in the Pilbara, two widespread habitat generalists (a murid, Pseudomys hermansburgensis and dasyurid, Ningaui timealeyi ) and a pebble surface specialist mouse ( Pseudomys chapmani ) found little evidence for population structure using mitochondrial and microsatellite sequence data [ 68 ]. Levy et al [ 68 ] proposed this resulted from the high fecundity and dispersal potential of these species, resulting in panmixia despite the complex geological landscape of the Pilbara presenting numerous potential barriers to gene flow. Panmixia has also been suggested for the mainland Pilbara population of the northern quoll [ 70 ], the largest Pilbara dasyurid (<1 kg), which is a species that has been documented to have excellent dispersal abilities [ 71 ].…”

Empirical landscape genetic comparison of SNPs and microsatellites in three arid-zone mammals with high dispersal capacity

“…A previous study found limited evidence that the Pilbara landscape influenced patterns of genetic structure in these species (Levy et al, 2019). With new, high-resolution genetic and spatial data, we investigate the relative performance of microsatellites and SNPs to resolve functional connectivity in the Pilbara by (1) identifying population genetic structure and potential physical barriers to dispersal (isolation-by-barrier; IBB); and (2) determining the role of dispersal capacity (isolation-by-distance; IBD) and specific landscape attributes (isolation-by-resistance; IBR: aridity, soil moisture, substrate, topography, distance to water, vegetation, and/or fire) in facilitating or restricting realised dispersal.…”

“…Situated in the Australian arid biome, the topographically complex Pilbara region is a biodiversity hotspot that supports rich faunal diversity including both endemic and widespread mammals (McKenzie, van Leeuwen, & Pinder, 2009). Despite this, functional connectivity in Pilbara mammals is poorly resolved and the few genetic studies in the region are based on microsatellite and mitochondrial markers and reveal low genetic structure (Hohnen et al, 2016;Levy et al, 2019;Umbrello, Didham, How, & Huey, 2020). Threats including resource extraction, grazing pressure, and inappropriate fire regimes all impact habitat connectivity in the Pilbara (Cramer et al, 2016), highlighting the need to understand functional connectivity in the region.…”

Range‐wide genetic structure of a cooperative mouse in a semi‐arid zone: Evidence for panmixia

“…Markers G6 and C1 were also detected as having null alleles when the sampling sites were analysed separately, but this trend was not pervasive, so these loci were retained for further analyses. mtDNA sequences and haplotype information were obtained from (Levy et al.,) for a 383‐bp segment of the highly variable, noncoding mitochondrial control region (D‐loop). DNA sequences were available for 81 of our western pebble‐mound mouse samples.…”