Quaternary climate change and Podocarpus elatus (Podocarpaceae).

Abstract

Understanding the effect of Quaternary climate change on the distribution, diversity and divergence of Podocarpus elatus (R.Br. ex Endl.) will contribute to the conservation of the east Australian rainforests in light of increasing ecological damage, rapid human population growth and anthropogenic-induced global warming. To know how intraspecific diversification in a wide-ranging long-lived species is driven by climate change will improve our understanding of how climatic-drivers are involved in the evolutionary process. This thesis examined the genetic consequences of Quaternary climate change in Podocarpus elatus, a long-lived rainforest conifer endemic to Australia. Firstly, eight polymorphic nuclear microsatellite and one chloroplast loci were isolated/characterised in P. elatus. I demonstrated the microsatellite primers could be applied to other podocarps (i.e. P. grayi, P. lawrencei, and P. smithii.). The markers were used to investigate the genetic diversity and structure of P. elatus throughout the broad-distributional range. Populations throughout the east Australian rainforests were screened and two divergent regions separated by the dry Clarence River valley (New South Wales) were discovered (i.e. Clarence River Corridor). Niche modelling techniques were employed to verify the incidence of climatic/habitat divergence between the two regions. Phylogeographic analysis and environmental niche modelling were combined to determine: (1) if post-glacial distributional dynamics could be described; (2) if the range-contractions suggested by southern fossil records are uniform across the entire distribution; and (3) if there is agreement between environmental niche modelling and molecular-based regional dynamics. Niche-modelling indicated that at the Last Glacial Maximum (21 Ka), the habitats suiting the two genetically differentiated regions of P. elatus were geographically disjunct. The northern distributional region persisted through the LGM in a small refugial area, which during postglacial periods has expanded. Conversely the southern range followed the opposite trend and has contracted since the LGM, but overall had greater genetic diversity. Coalescence-based analyses support these differential dynamics across the distribution of the species. The future climate-induced range shift of the two genetically differentiated regions of P. elatus were modelled and compared to coalescence-based inference of regional gene flow, genetic boundaries and expansion/contraction dynamics to provide information with regard to community response to climate cycles. A total of 405 occurrence records were obtained to model species’ distribution (21, 6, and 0 Ka) based on the current environment using MAXENT and forecasting future distribution (2050 A2) using an ensemble of thirteen atmospheric-oceanic global climate models. The analysis suggests the geographic shift in genetic diversity of P. elatus according to future climate change scenarios. Together these data sources provide a means to predict future distribution of genetic diversity, and infer rainforest areas at increased risk of localised extinction. It was found that P. elatus is considerably more threatened than shown by its current distribution, and I suggest the use, and extension, of habitat corridors to accommodate future climate-induced range shift of fragmented rainforest habitats along the east coast of Australia.