Eucalypt decline and ectomycorrhizal fungal community ecology of Eucalyptus delegatensis forest, Tasmania, Australia

Ectomycorrhizal fungi underpin critical ecosystem processes which affect tree health. Eucalypt decline is widespread throughout Australia, and its cause has been attributed to a variety of factors, including forest management. In Tasmania, dieback in Eucalyptus delegatensis R.T. Baker has been linked to altered fire regimes and associated changes in mycorrhizal communities. This thesis presents a study that explores ectomycorrhizal species richness and community composition in relation to eucalypt health, understorey vegetation and soil chemistry in the context of fire history. The results further our understanding of ectomycorrhizal ecology and elucidate factors important to the maintenance of a healthy forest ecosystem. Study sites were established in E. delegatensis forest with either sclerophyll understorey (six plots) or rainforest understorey (six plots). Eight of the plots, located in north-east Tasmania, had been established for a study of fire ecology and had known fire histories ranging from 42 years since the last fire to long unburnt (>120 years). Four plots located in north-western Tasmania were long unburnt but had been disturbed by logging 22-25 years previously. Ectomycorrhizal fungal sporocarps, root tips and soil samples were collected during a three-year period from all 12 plots. Samples from soil, root tips and sporocarps gave rise to different but complementary information about ectomycorrhizal communities. Fungal operational taxonomic units were identified through DNA sequencing and phylogenetic analysis. At each site, understorey vegetation was characterised, soil and eucalypt foliage chemistry was analysed, and eucalypt crown condition was assessed. Primary crown dieback was identified as the most effective method for the measurement of eucalypt health. Multivariate statistical analyses were used to explore the relationships among ectomycorrhizal communities, eucalypt health, vegetation and abiotic variables. E. delegatensis forest with rainforest understorey was more than likely to be affected by severe eucalypt decline, had higher concentrations of soil inorganic nitrogen (nitrate and ammonium) and eucalypt foliar nitrogen, and had lower concentrations of soil and eucalypt foliar phosphorus, than forest with sclerophyll understorey. As forest declined in health the ecosystem moved from being nitrogen limiting to phosphorus limiting due to reduced phosphorus availability and plant uptake, potentially due to altered mycorrhizal activity. Ectomycorrhizal communities differed between moderately and severely declining forest and were correlated to crown health and altered soil chemistry associated with the two levels of decline. The Cortinariaceae had high species richness in healthiest sites while the Russulaceae and Thelephoraceae were rich in forest affected by severe decline. In northwestern and north-eastern Tasmania unique and distinctly different ectomycorrhizal fungal communities were found to occur in E. delegatensis forest with rainforest understorey versus those with sclerophyll understorey. Irrespective of understorey type and health status, the Cortinariaceae were highly diverse, and were the most species-rich family within the ectomycorrhizal community of E. delegatensis forest. The Cortinariaceae also was the most abundant family in the root tip community. The Helotiales, Russulaceae and Thelephoraceae also were important components. The importance of the Cortinariaceae in E. delegatensis forest is similar to other Australian eucalypt forests but distinctly different from northern hemisphere forests, which tend to be dominated by the Russulaceae, Thelephoraceae and Corticeaceae. Distance-based multiple linear regression models using only significant predictor variables based on soil and foliage nutrient concentrations and crown health were able to explain 52% of the variation in fungal community composition, and 44% of the variation in ectomycorrhizal community proportional composition at the family level. Soil pH, total soil nitrate, soil organic carbon and soil phosphorus were significant in predicting ectomycorrhizal species composition and proportional composition in the final models. A multiple linear regression model showed that available soil nitrate and phosphorus were significant in predicting ectomycorrhizal community richness. High richness was associated with low available soil nitrate or phosphorus. Northern hemisphere studies which show that changes in soil chemistry, especially mineral nitrogen, can strongly influence mycorrhizal species richness, species composition and community structure corroborate the likely influence of soil nitrogen on the ectomycorrhizal communities of E. delegatensis forest. This is the first study to find a strong correlation between ectomycorrhizal fungal communities and the status of eucalypt forest health. The results support the currently proposed model that, in the absence of fire, premature decline of temperate Australian eucalypt forests is closely linked to changes in soil chemistry, understorey vegetation and mycorrhizal communities.