Range-wide fragmentation in a threatened fish associated with post-European settlement modification in the Murray-Darling Basin, Australia

Abstract

Distinguishing the relative influence of historic (i.e. natural) versus anthropogenic factors in metapopulation structure is an important but often overlooked step in management programs of threatened species. Biotas in freshwater wetlands and floodplains, such as those in the Murray–Darling Basin (MDB)—one of Australia’s most impacted ecosystems, are particularly susceptible to anthropogenic fragmentation. Here we present a comprehensive multilocus assessment of genetic variation in the threatened southern pygmy perch Nannoperca australis (578 individuals; 45 localities; microsatellite, allozyme and mitochondrial DNA datasets), an ecological specialist with low dispersal potential. We assess patterns of spatial structure and genetic diversity in populations spanning the highly fragmented MDB and test whether recent anthropogenic modification has disrupted range-wide connectivity. We detected strong and hierarchical population structure, very low genetic diversity and lack of contemporary gene flow across the MDB. In contrast, the apparent absence of pronounced or long-term phylogeographic structure suggests that observed population divergences generally do not reflect deeply historic natural fragmentation. Coalescent-based analyses supported this inference, revealing that divergence times between populations from the upper and lower MDB fall into the period of European settlement. It appears that the observed contemporary isolation of populations is partly explained by the severe modification of the MDB post-dating the onset of European settlement. Our integrated approach substantially improves the interpretation of how fragmentation impacts present-day biodiversity. It also provides novel contributions for risk-assessing management actions in the context of captive breeding and translocations of small freshwater fishes, a group of increasing global conservation concern.