Evolution and biogeography of Australian tropical freshwater fishes

Abstract

Australia’s freshwater fish fauna is the most depauperate of any continent (256 formally recognised species), although endemism is exceptionally high (74%), largely due to its arid climate and history of isolation from other land masses. The Australian Monsoonal Tropics (AMT) biome in the tropical north is an exception. The AMT encompasses 33% of the Australia landmass, but contains 65% of the Australian fish fauna and, in a global context, the biome and many of its catchments contain moderate to high species richness relative to their size. However, the biodiversity, evolution, and biogeography of the AMT’s fish fauna remain poorly studied relative to the rest of the continent.

In this thesis I utilise samples from the most comprehensive region-wide collection of freshwater fish molecular and distributional data in the AMT to help answer three fundamental questions regarding the regions freshwater fish fauna: (1) what is the true biodiversity of the AMT; (2) what are the key evolutionary processes driving and maintaining freshwater fish diversity across the region, in particular the highly endemic fauna of the Kimberley bioregion; and (3) what are the key patterns in diversity and distributions across the landscape and how can they be arranged into a cohesive biogeographic framework?

First, I conducted a multigene molecular assessment of species boundaries in the AMTs most speciose freshwater family (Terapontidae) in order to assess the phylogenetic relatedness of terapontids in northwestern Australia (including the Kimberley) to the level of population, and to identify any unique genetic lineages that likely represent undescribed ‘candidate species’. I demonstrated the presence of 13 new candidate species within the Kimberley, more than doubling previous estimates of terapontid diversity in the region.

Second, I conducted an assessment of morphological (morphometric and meristic) data from seven of the genetically defined candidate taxa, and the four previously described species within the genus Syncomistes to see if the seven candidates can be discriminated morphologically and to determine which characters best delimit taxa. I found an impressive array of meristic and morphometric character differences between species within Syncomistes and determined that the head, particularly feeding structures such as the jaw and dentition, were the most important morphological features in discriminating between taxa.

Third, I looked for congruence between phylogenetic patterns in Kimberley terapontids and both past (low sea-level)/present (high sea-level) geological barriers and pathways as identified by GIS analysis, and tested the general hypothesis that geographic isolation of terapontid lineages during Pliocene and Pleistocene high sea-levels triggered the onset of reproductive isolation between taxa thus driving rapid speciation in the region. I found that most Kimberley terapontid species arose during the Plio-Pleistocene glacial cycles and are at different stages of allopatric divergence and speciation caused by the same vicariant processes. The results support the hypothesis that changing sea levels during late Pliocene and Pleistocene glacial cycles are a key driver of speciation and distributional patterns in the Kimberley.

Fourth, I combined phylogenetic, biogeographical and diversification analyses to examine the nature of the Kimberley as a mesic refugium. Specifically, I investigate the tempo and timing of endemic diversification to see if the Kimberley has been a ‘museum’ or a ‘cradle’ of diversification. My combined molecular clock estimates and likelihood-based historical biogeographic reconstructions suggest that terapontids recently transitioned into the Kimberley from the east during the late-Miocene. Outstandingly, ~80% of Kimberley terapontids diversified within the region in the last 3 Ma. Further diversification analyses identified a single significant shift in diversification rates ~1.4 Ma that corresponds with a significant change in global climate midway through the Pleistocene. Given these finding my findings suggest that the Kimberley has been acting as a cradle of Neoendemism.

Fifth, I generate a bioregionalisation of the freshwater fish in the AMT using the Simpson’s beta dissimilarity metric, and then assess the relationships of the biogeographic regions to their current environment using generalised dissimilarity modeling (GDM). I also estimate true species richness across catchments using the Chao 2 index in order to identify major sampling gaps. I propose three major freshwater fish bioregions and 14 subregions that differ substantially from the current bioregionalisation scheme. I found that species turnover was most strongly influenced by environmental variables that reflect changes in terrain (catchment relief and confinement) and productivity (NPP and forest cover). Current river orientation and historic connectivity between rivers during low sea-level events also appear to be influential. Three focal points of species richness and two of endemism were identified in the AMT, considerably expanding upon the spatial understanding of these patterns. Finally, a number of key sampling gaps are identified that need to be filled in order to fully refine the proposed regionalisation.

Overall the results of this thesis add considerably to biodiversity estimates and the taxonomic knowledge of freshwater fish communities in the AMT. It also helps determine the major drivers of speciation in the Kimberley, the mode of diversification, and provides insight into the regions function as an evolutionarily important mesic refugium. Finally, it provides a modern freshwater bioregionalisation of the AMT and helps to determine the environmental variables driving community change across the landscape. These findings have important ramifications for the conservation of Australia’s tropical freshwater fishes. The Kimberley in particular is highlighted as not only an important evolutionary refugium, but also as a catalyst for narrow range endemic speciation. As a result the regions contains some of the most threatened freshwater fish communities in Australia.