Phylogeography, genetic stocks, and conservation implications for an Australian endemic marine turtle

Abstract

Identification of the geographic extent of population boundaries, the distribution of genetic lineages, and the amount of genetic exchange among breeding groups is needed for effective conservation of vulnerable marine migratory species. This is particularly true of the flatback turtle (Natator depressus), which only breeds in Australia but has extensive migrations that can include international waters. This study investigated the phylogeography and genetic structure among 17 flatback turtle rookeries across their range by sequencing an 810 bp portion of the mitochondrial DNA in 889 samples and genotyping 10 microsatellite loci in 598 samples. There was low phylogenetic divergence among haplotypes and evidence of recent population expansion, likely in the late Pleistocene. A predominant haplotype was found across all rookeries, but other haplotype groups were regionally specific. In general, there was agreement in patterns of genetic differentiation in the mitochondrial DNA and microsatellite data, and in some pairwise comparisons a higher mutation rate of microsatellites provided stronger evidence of differentiation. These results suggest natal philopatry operates in the choice of breeding locations for males as well as females. Evidence of genetic connectivity among neighbouring rookeries led to the identification of seven genetic stocks. Geographic boundaries of rookeries used by genetic stocks varied widely (160–1,300 km), highlighting a need for field studies to better understand movement patterns. Hierarchical analysis of molecular variance identified significant genetic differentiation based upon genetic stock, nesting phenology (summer vs. winter nesters), and a west–east discontinuity across Torres Strait. A pattern of isolation by distance was identified, which was most strongly observed in the microsatellite data. In combination with tagging and telemetry studies, these results will allow better quantification of stock‐specific threats along migratory routes and in foraging habitats. Implications of climate change will be stock specific and may depend upon the extent of genetic connectivity between neighbouring stocks.

Variation in thermal traits describing sex determination and development in Western Australian sea turtle populations

“…These results support previous suggestions that some N. depressus populations may be resilient to the over‐production of females as a consequence of climate change through local adaptation (Howard et al., 2015), as the warmest rookery at Eighty Mile Beach (Bentley, Kearney, et al., 2020) had the highest T PIV . This large and genetically distinct N. depressus population (FitzSimmons et al., 2020) has presumably been under stabilising selection for TSD parameters that generate optimal operational sex ratios at warmer incubation temperatures (sensu Doody et al., 2006).…”

“…Chelonia mydas nest throughout the year in the region studied, with nesting peaks occurring during summer in Western Australia, and so for this species our focal rookeries represented geographically separated populations nesting on sub‐tropical and tropical beaches. The three N. depressus rookeries represented distinct genetic stocks, while the two C. mydas rookeries were also genetically distinct, based on 2,074 single nucleotide polymorphisms (SNPs; FitzSimmons et al., 2020). Microclimate models show substantial thermal variation in sand at 50‐cm depth among these rookeries (Bentley, Kearney, et al., 2020), with average temperatures during the nesting season ranging from 27.4 to 32.0°C (Figure 1).…”

“…Populations of N. depressus nesting around the Austral summer (November–February) and Austral winter (July–October) occur across this geographic range, with the phenological shift occurring around the King Sound region (Tucker et al., 2018; see Figure 1). This phenological divergence contributes to strong genetic structure in N. depressus (FitzSimmons et al., 2020). Microclimate models of beach temperature and empirical data point to substantial thermal variation across Western Australian rookeries, which is partly offset by the differences in peak nesting times (Bentley, Kearney, Whiting, & Mitchell, 2020).…”

Mitochondrial DNA Profiling to Combat the Illegal Trade in Tortoiseshell Products

“…Turtles are poached from both nesting and foraging areas and products may be seized at the poaching location, en-route to or at the destination of the illegal trade, hence it can be challenging to know which breeding populations the items came from. Genetic analysis has been used extensively in marine turtles to assess phylogeography (Vargas et al, 2016;Jensen et al, 2019a;FitzSimmons et al, 2020) and stock structure (Dutton et al, 2013;Matsuzawa et al, 2016). Once the stock structure for a species has been determined, that information has been used to estimate the stock origin of turtles sampled at foraging areas (Maffucci et al, 2006;Gaos et al, 2018a;Dutton et al, 2019), caught as bycatch in fisheries (LaCasella et al, 2013;Clusa et al, 2016;Stewart et al, 2016) and stranded turtles (Rankin-Baransky et al, 2001).…”

Optimising sample sizes for animal distribution analysis using tracking data

“…The new method uses the probabilistic property of utilisation distributions rather than a user‐defined volume contour and allows distribution analyses to more realistically account for variation in space use. We demonstrate this technique using a dataset consisting of 69 deployments of accurate Fastloc‐GPS tags on a single genetic stock of flatback turtles Natator depressus (FitzSimmons et al., 2020). This species provides an ideal case study since they have been the subject of a number of satellite tracking programmes, driven by the need for accurate and robust data on the distribution of these threatened marine reptiles in order to manage anthropogenic impacts, such as climate change, boat strikes, predation, and pollution (Critchell et al., 2019; Hamann et al., 2013; Lewison et al., 2013; Shimada et al., 2017).…”