Genome-wide data delimits multiple climate-determined species ranges in a widespread Australian fish, the golden perch (Macquaria ambigua)

Abstract

Species range limits often fluctuate in space and time in response to variation in environmental factors and to gradual niche evolution due to changes in adaptive traits. We used genome-wide data to investigate evolutionary divergence and species range limits in a generalist and highly dispersive fish species that shows an unusually wide distribution across arid and semi-arid regions of Australia. We generated ddRAD data (18,979 filtered SNPs and 1.725million bp of sequences) for samples from 27 localities spanning the native range of golden perch, Macquaria ambigua (Teleostei; Percichthyidae). Our analytical framework uses population genomics to assess connectivity and population structure using model-based and model-free approaches, phylogenetics to clarify evolutionary relationships, and a coalescent-based Bayesian species delimitation method to assess statistical support of inferred species boundaries. Addressing uncertainties regarding range limits and taxonomy is particularly relevant for this iconic Australian species because of the intensive stocking activities undertaken to support its recreational fishery and its predicted range shifts associated with ongoing climate change. Strong population genomic, phylogenetic, and coalescent species delimitation support was obtained for three separately evolving metapopulation lineages, each lineage should be considered a distinct cryptic species of golden perch. Their range limits match the climate-determined boundaries of main river basins, despite the ability of golden perch to cross drainage divides. We also identified cases suggestive of anthropogenic hybridization between lineages due to stocking of this recreationally important fish, as well as a potential hybrid zone with a temporally stable pattern of admixture. Our work informs on the consequences of aridification in the evolution of aquatic organisms, a topic poorly represented in the literature. It also shows that genome-scale data can substantially improve and rectify inferences about taxonomy, hybridization and conservation management previously proposed by detailed genetic studies.