Phylogeography of the Australian freshwater turtleChelodina expansareveals complex relationships among inland and coastal bioregions

Abstract

We examined range‐wide mitochondrial phylogeographical structure in the riverine freshwater turtle Chelodina expansa to determine whether this species exhibits deep genetic divergence between coastal and inland hydrological provinces, as seen in co‐distributed freshwater taxa. We sequenced two mitochondrial loci, genealogical relationships were assessed using a network approach, and relationships among biogeographical regions were tested using analyses of molecular variance. Population history was evaluated using neutrality tests, indices of demographic expansion, and mismatch analyses. Twenty‐one haplotypes were recovered across two mitochondrial haplogroups separated by approximately 4% nucleotide divergence. The haplogroups have discrete geographical boundaries but only partially support a hypothesis of deep divergence between coastal and inland bioregions. The first haplogroup comprises populations from the inland Murray‐Darling Basin and from coastal catchments south of the Mary River in south‐east Queensland. The second haplogroup comprises populations from coastal catchments north of the Mary River. Cryptic phylogeographical barriers separating adjacent coastal populations are congruent with those demonstrated for other freshwater taxa and may result from the combined influences of the Conondale Range and alluvial deposits at the mouth of the Mary River. The findings of the present study demonstrate that freshwater taxa commonly display genetic differentiation within a biogeographical region where no boundaries have been recognized, highlighting the need to uncover cryptic microbiogeographical regions to aid conservation of freshwater biota. © 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, 111, 789–805.

Significant genetic structure despite high vagility revealed through mitochondrial phylogeography of an Australian freshwater turtle (Chelodina longicollis)

“…It is particularly subdued at the drainage boundary between the MDB and the Bass Province and at the boundary between the MDB the Fitzroy-Dawson and Burnett drainages. Exchange of freshwater taxa across the Great Dividing Range has been long recognised Keenan 1992, 1996;McGlashan and Hughes 2001;Unmack 2001;Murphy and Austin 2004;Cook et al 2006;Hammer et al 2007;Thacker et al 2007;Jerry 2008;Faulks et al 2008Faulks et al , 2010Unmack and Dowling 2010;Hodges et al 2014) and we expect C. longicollis to easily transverse regions with low to moderate elevational gradients.…”

“…Sequences were aligned using ClustalX 1.81 (Thompson et al 1997) to yield final edited alignments of 1042 bp, comprising 595 bp of ND4, 68 bp of tRNA Pro and 379 bp of control region (GenBank accession numbers for ND4 haplotypes are KM581393-KM581420; GenBank accession numbers for CR haplotypes are KM581421-KM581448). Four methods, described in greater detail in Hodges et al (2014), were used to confirm the genuine mitochondrial origin of the sequences and minimise the chance of undetected inclusion of nuclear paralogues in our analyses. Specimens examined in the Supplementary material details the two samples used in the present study for the mitochondrial enrichment procedure.…”

“…We suggest that the barrier presented by the ancient and topographically complex Great Dividing Range drove independent evolution of the two mitochondrial lineages and has also maintained separate distributions of the two contemporary haplogroups at least since the early Pleistocene. A range of sympatric freshwater taxa, including fish, crustaceans and a turtle, also display intraspecific phylogeographic structure in varying extent and age as a result of the Great Dividing Range (Rowland 1993;Unmack 2001;Murphy and Austin 2004;Hammer et al 2007;Faulks et al 2008Faulks et al , 2010Unmack and Dowling 2010;Hodges et al 2014). Despite its often subdued character, this landscape feature is an important driver of evolutionary diversity in freshwater taxa, regardless of life history.…”

“…The McPherson Range forms the high-elevation (,1500 m above sea level) northern boundary of the Richmond drainage and acted as a significant barrier to gene flow in some species throughout the Miocene and Pliocene (McGuigan et al 1998;Keogh et al 2003;Chapple et al 2011;Smissen et al 2013). The Conondale Range delineates the Mary and Brisbane drainages and is an influential biogeographic barrier in many freshwater species, including a turtle (Page and Hughes 2014;Hodges et al 2014). Future sampling is required in C. longicollis to determine the exact location of the phylogeographic break, and whether haplogroups overlap or have hard boundaries in this region.…”

“…Our study builds on the phylogeographic dataset for sympatric freshwater taxa in eastern Australia (Hughes et al 2013;Hodges et al 2014) to test the extent of influence that freshwater bioregions and hydrological connectivity have on genetic structure. C. longicollis represents the maximum dispersal capacity of an obligate Australian freshwater vertebrate and can potentially highlight the upper limit beyond which freshwater bioregions have no effect on phylogeographic partitioning.…”