Community genetics of eucalypts : provenance effects on canopy communities, potential drivers and underlying QTL

Genetic variation in foundation trees can have an extended effect on their dependent communities. Extended genetic effects have been shown to influence the abundance, richness and composition of canopy communities in many forest tree systems. However, few community genetic studies have been undertaken on southern hemisphere trees. This thesis focuses on the important genus Eucalyptus. Trees of this genus dominate much of Australia's native forests and woodlands but are grown globally for forestry and other uses. Eucalypt species are genetically diverse and it is important to understand the influence of this genetic diversity on dependent biodiversity. This is particularly so for large-scale forestry and restoration plantings where there is increasing interest in the use of non-local provenances as a response to global climate change, but is also an issue for native forest management and conservation. Three well-studied eucalypt systems ‚Äö- E. globulus, E. morrisbyi and E. pauciflora ‚Äö- were used to study the relative importance of provenance variation in shaping the canopy community of dependent arthropod and fungi, potential mechanisms driving genetic-based differences and underlying regions of the genome impacting on the dependent community. The thesis specifically aimed to determine: 1) the relative importance of extended genetic effects in comparison to known factors influencing biotic communities, including site, yearly and ontogenetic effects; 2) the stability of extended genetic effects across environments and years to better understand the potential extended consequences of provenance translocation in forestry, restoration and conservation; 3) the relative importance of foliar chemicals as potential drivers of genetic-based variation in dependent community responses in E. globulus; and 4) the regions of the genome impacting on dependent community responses. Significant variation in the dependent canopy community among provenances was detected following natural colonization of common garden field trials of all three eucalypt systems. The relative influence of genetic-based effects on dependent community composition was comparable to that of ontogenetic effects (E. morrisbyi), while site (E. globulus and E. pauciflora) and yearly effects (E. pauciflora) far outweighed that of genetic-based effects. Nevertheless, broad trends in genetic-based community variation among provenances were detected across sites and years and associated with reported patterns of adaptive differentiation within the focal tree species associated with latitude (E. globulus) and altitude (E. pauciflora). The association of genetic-based variation in physiochemical properties of leaves on the dependent community of E. morrisbyi and E. globulus provided insights into potential foliar chemical mechanisms driving these genetic-based community differences. Investigation of the cuticular waxes of E. globulus showed significant genetic control and differences among provenances, with signals of divergent selection in four of the thirteen quantified wax compounds. These four cuticular wax compounds along with seven foliar terpenes explained a combined 34% of the provenance differences in canopy community composition. Overall the influence of foliar wax compounds was comparable to that of the well-studied foliar terpenes. However, while quantitative trait loci (QTL) for community traits were detected in the mapping family studied, none of the chemical compounds associated with provenance differences in the canopy community were associated with these genomic regions. Furthermore, while many QTL were detected for individual community members, and some of these co-located with QTL for foliar chemicals, none co-located with QTL for community traits. The work presented in this thesis advances the field of community genetics in several ways. Extended genetic effects have been detected in numerous plants systems, however, the partitioning of variation to determine the relative importance of genetic effects and potential mechanisms is rare. While this thesis shows that extended genetic effects in eucalypts may be small in comparison to effects such as site and year, it provides evidence that adaptive variation in phenotypic traits can have extended effects on communities and ecosystem processes, and this influence appears consistent across sites and years. Additionally, it highlights the complexity underlying both phenotypic and genetic-based variation in dependent canopy communities with QTL providing evidence of community-level emergent genetic effects.