The genetic and demographic impacts of contemporary disturbance regimes in mountain ash forests

Abstract

Timber harvesting, frequent wildfires and a changing climate are influencing ecosystem composition, structure and function globally, with resulting losses to biodiversity and economic indicators. In south-eastern Australia, these factors are causing the rapid ecosystem collapse of montane forests. Here, I characterise and quantify the demographic and genetic impacts that changing environments are having on mountain ash (Eucalyptus regnans), a foundation species and one of the world's tallest trees. To test whether mountain ash populations exhibit variation in susceptibility to increasing fire frequency, I investigated the response of growth rates and seed production to stand age under different environmental conditions. My results show that environmental factors determine the age of maturation, in turn affecting the time taken for populations to develop reproductively viable amounts of seed. This suggests that reduced fire return intervals may result in niche contractions of obligate seeders such as mountain ash. Next, I conducted a range-wide analysis of mountain ash population genetic structure to determine the extent of hybridisation with messmate stringybark (Eucalyptus obliqua), and investigate how genetic diversity parameters are influenced by hybridisation. I found that hybrid occurrence was not distributed evenly across environmental gradients or populations, and after accounting for admixture, mountain ash showed very little population genetic structure, with a small effect of isolation-by-distance and low global FST (0.03). This suggests that decisions around provenancing for restoration may depend on knowledge of how admixture influences population genetic structure, and that for some species there may be little benefit in planning conservation strategies around environmental adaptation of seed sources. Fire and silvicultural practices may be modifying patterns of within- and among-population genetic diversity and fine-scale spatial genetic structure, across a mountain ash-dominated landscape. As chloroplast DNA and nuclear DNA are dispersed via different mechanisms, manual sowing of logged sites using non-local seed is likely to have differing effects on these two genomes. To test this, I utilised chloroplast microsatellites and genome-wide single-nucleotide polymorphisms to compare genetic parameters between undisturbed, burnt, and logged stands. The patterns and extent of genetic diversity and genetic differentiation among stands at nuclear loci were similar among disturbance histories, but chloroplast microsatellites revealed significantly higher levels of genetic diversity in logged stands. This suggests that logging is having minor impacts on the nuclear genome but large impacts on the chloroplast genome, with haplotypes entering the system via the use of non-local seed in the regeneration process. Understanding patterns and drivers of local adaptation is important for conservation management of foundation species. I investigated the genome of mountain ash for signatures of local adaptation at a regional and range-wide spatial scale, using three methods at each scale to identify outlier loci. I found 40 loci that were significantly associated with environmental variables, and demonstrate that investigation of multiple spatial scales provides a greater understanding of local adaptation. This thesis provides novel insights into impacts that modified disturbance regimes have on mountain ash populations. I found geographic variation in vital rates; genetic patterns suggestive of low recolonisation ability; high levels potential adaptive capacity; high potential for adaptive introgression; and identified loci showing signs of local adaptation. Importantly, this knowledge will assist with the conservation management of foundation tree species and forest ecosystems, contributing to the maintenance and/or maximisation of adaptive capacity and allowing forests to persist into a changing and uncertain future.