Phosphorus (P) limits productivity across broad swathes of Earth. Soil microbes are the key players in controlling P availability: first because they are responsible for the conversion of organic to inorganic forms, and second because the microbial biomass accounts for a substantial fraction of organic P. To date little is known about the microbial traits that enable microbial communities to flourish across broad P availability gradients. There is some evidence that differences in elemental stoichiometry (e.g. C:N:P) play a role in adaptation to P availability, but as yet we do not know how microbial communities alter C:N:P.

Little progress has been made in understanding how soil microbial communities adapt to low P because we have lacked the tools to move beyond simplistic view of microbes as blackboxes of C, N & P. One way of moving beyond the restrictive lens of elemental stoichiometry is by harnessing the power of mass spectrometry. The aim of this work was to explore how soil microbes adapt to P by developing a mass spectrometry toolkit for ID & quantification of soil microbial cellular composition.

We show for the first time that one of the ways soil microbial communities from P-poor soils minimize cellular P requirements is by substituting phospholipids with P-free betaine lipids. In P-poor soils betaine lipids accounted for almost half of membrane lipids, thereby sparing about 15% of total cellular P. Lipid substitution is likely a key trait underpinning adaptation of soil microbes to P deficiency.

Recent work has shown that the network architecture of species interactions affects the diversity and resilience of plant communities over time, but the effect on invasibility over time has not yet been investigated. To explore this, we designed and conducted a series of virtual experiments using a spatially-explicit stochastic cellular automata simulation model of plant community dynamics. Interactions among plants were assumed to be mediated via plant–soil feedback, with the invader species having positive conspecific interactions.
We simulated the progress of invasion over 2000 years in communities with 88 different theoretical species interaction networks with a range of network architectures. We also considered the effect of whether invasion followed or didn’t follow a disturbance event and whether plant interactions occurred between neighbours or between future and previous occupants.
We found that the network architecture of plant species interactions influenced both the probability that the invader persisted and the rate at which its abundance (percentage biomass) increased. Surprisingly, disturbance did not always increase these measures of invasibility, and in some systems reduced them. Lastly, we found that there was greater abundance of the invader in communities where plant interactions occurred between the previous and future occupant compared to communities where interactions occurred between neighbours.
Our findings highlight the potential importance of plant interaction network architecture in determining the invasibility of plant communities, which could have useful applications for the management of invasive species.

Drought has the potential to affect the structure of root-associated fungal communities. Pathogenic, mutualistic, and saprotrophic fungi in the soil are expected to respond in several ways to drought, while these responses, in turn, would impact plant mechanisms to face drought. To test how different fungal groups respond to drought and whether their response is linked with root morphological adjustments to drought, we grew 24 different grasslands species under drought (30 % WHC) and non-drought (70 % WHC) conditions. The plants used included all major functional types (grasses, legumes, and forbs). At the end of the experiment, root traits, abundance, and richness of different fungal guilds were measured.
Our results show that different fungal guilds respond differently to drought. Soil pathogen abundance and richness were not affected by drought but differed among plant species. Physiological acclimation, synthesis of osmolytes to maintain hydration, could explain pathogens persistence under drought. Saprotroph abundance and richness increased with drought. Previous studies have seen that different enzymes which break down cellulose and degrades chitin increases their activity under drought. In contrast, AMF abundance and richness decreased with drought, which was linked with a higher specific root surface (i.e., fine roots) for grass species, suggesting that grasses rely more on root morphological adjustments than on mycotrophy in responding to drought. Our results evidence that drought has a strong effect on the assembly of fungal communities, but such responses depend on interactions between plant functional groups and fungal guilds.

Soil microbes play a pivotal role in the functioning of terrestrial ecosystems and are increasingly recognized as important drivers of plant diversity. eDNA technology has revolutionised our capacity to assess soil microbial communities. A powerful application of this technology is the assessment of responses and trajectories of microbial communities in restoration soils post-disturbance. This study assessed changes in bacterial and fungal communities in soils at 36 sites through a 30-year-old chronosequence of highly diverse forest-community rehabilitation after bauxite mining in south-west Australia. Bacterial and fungal communities were analysed using Illumina MISEQ sequencing of the 16S rRNA and ITS1 gene regions respectively. These were assessed against changing restoration practises, an equivalent number of adjacent intact remnant communities, and physical/chemical properties of the soil. Increasing microbial diversity and a strong trajectory towards the composition of reference soils was found. However, key compositional differences between reference and 30yo restoration soils were identified, the functional significance of which is to be explored. Measuring below-ground richness using eDNA considerably alters perceptions of biodiversity in restored and natural landscapes, and helps to optimise reconstructed substrates for aboveground and belowground biodiversity.

One of the few predicted benefits under a future climate is a potential increase in plant productivity associated with elevated concentrations of CO2. This benefit is physiologically constrained to C3 species, which classically grow in temperate biomes. Under a warmer and drier climate many regions currently growing C3 species for pasture may consider planting tropical grasses (C4), which do not experience the same benefit to productivity under elevated CO2 conditions. However, there still may be benefits from elevated CO2 in growing these C4 grasses when they are mixed with legumes. As part of a nutrient facilitation experiment in a glasshouse setting we compared pasture production among two pairs of tropical grasses and legumes. The grasses and legumes were grown both singly and together under ambient and elevated CO2 atmospheric conditions. As expected, grasses grown in monoculture did not benefit from elevated CO2 conditions while legumes increased biomass by 30-40-%. In general, grasses grown in a mixed sward had greater biomass than those grown in monoculture, however, the legumes grown with grasses declined in biomass as compared with legumes grown singly. Furthermore, the benefit to legumes from elevated CO2 was reduced (Burgundy bean) or removed (Greenleaf Desmodium) when grown in combination with grasses. Importantly, grasses grown with legumes under elevated CO2 conditions increased production thereby benefitted from the altered climate, suggesting that mixed swards may provide an added benefit nutritionally under future conditions for tropical species.

Urban land contamination is a persistent environmental problem in Australia, contributing to chronic human health issues and overall ecosystem degradation. Phytoremediation is an emerging plant-based biotechnology which uses select species to decontaminate polluted soils over time. In 2018 to 2019, the first large-scale, in-situ Australian phytoremediation garden was installed by a cross-institutional research team on the grounds of White Bay Power Station, Sydney; a former coal-fired power station of national heritage significance. The aim of this project was to plant specially designed fields of heavy metal-absorbing phytoremediator species to decontaminate the site's soils, and comparatively measure species effectiveness. Fields of 25 different annual species, native and non-native, were planted on the grounds of the power station in 2018. Species were only selected for use if they had been shown to work effectively in previous scientific trials. Exceptionally long and intense Summer heatwaves, combined with reduced seasonal rainfall, negatively influenced the survival and establishment of several plant species. ICP-MS analysis of harvested tissue samples revealed that plant species from different taxonomic groups, including Asteraceae and Brassicaceae, effectively absorbed and compartmentalised high quantities of heavy metals over time. This project demonstrates that phytoremediation is a viable, cost-effective and biodiversity-restoring solution to large-scale land contamination in Australian industrial zones.

Most invasive trees depend closely on mycorrhizal symbionts to provide required resources, and thus their invasive success depends on the dispersal of these mycorrhizae as well as their own dispersal. Previous empirical and theoretical work has shown that the dispersal and environmental tolerance characteristics of organisms can undergo selection pressure and evolution during the course of an invasion or colonization of new areas, but the evolutionary dynamics of these traits during co-invasion has not been considered. In this work, we use spatially-explicit eco-evolutionary simulation modelling to investigate how the dispersal and environmental tolerance characteristics of both trees and their mycorrhizal symbionts evolve over the course of a tree invasion. We find that the selection pressures manifesting during the invasion cause the dispersal and environmental tolerance characteristics of the different organisms to vary across time and space in complex and interdependent ways that can be influenced by human management. For example, dispersal ability of the two organisms increases over time and is higher at the fronts of an invasion, but managing trees through targeted removal of outliers decreases the rate at which increased dispersal ability evolves, which in turn greatly slows the rate of invasion. Our results show that eco-evolutionary dynamics should be considered when formulating management strategies for invasive plants.

The relationship between legumes (Fabaceae) and their bacterial symbionts evolved more than 60 million years ago. In particular, legumes associate with nitrogen-fixing bacteria (rhizobia), especially when soil nitrogen levels are low. Legumes therefore play a central role in nutrient cycling and productivity of many natural and managed ecosystems. The interaction takes place in the soil, which is a complex chemical and ecological environment. Silicon is the most abundant metalloid element in soil, some of which (bioavailable silicon) can be taken up by plants, especially the grasses. Little, however, is known about how silicon affects legumes or their relationship with rhizobia. Our research aims to redress this. We present novel results showing how silicon affects legume-rhizobium interactions using three different genotypes of the model legume _Medicago truncatula_. Plants were grown in soil with low bioavailable silicon and either supplemented or non-supplemented with silicon. All genotypes were inoculated with a model rhizobial strain _Ensifer meliloti_ SM1021. Three-month old plants were harvested and key symbiotic traits, such root nodulation and nitrogen fixation were analysed. We found that silicon promoted nitrogen fixation, by more than 40% and 50% in two of the genotypes. We used 15N natural abundance to quantify the exact amount of nitrogen being fixed by rhizobia. These results will be discussed in the context of plant silicon and its potential role in nutrient uptake. Finally, understanding these tripartite interactions between silicon, legumes and rhizobia may contribute to a more environmentally sustainable strategy for ensuring ecosystem health and global food security.

Soil carbon sequestration is a key mechanism through which greenhouse gas emissions can be reduced via the removal of carbon dioxide from the atmosphere by soil microbes (bacteria, fungi) and plants, and storing it in a solid form (soil organic carbon) within the soil. This project is testing the effectiveness of recycled organic compost and timed (or rotational) grazing as strategies to increase soil carbon sequestration and improve soil biology and structure in dryland grazing systems of northern Victoria. On each of five trial farms, compost (from household green waste) has been added to half of a 7 ha plot at a rate of 5 tonnes/ha in autumn and spring 2019, and crossed with three grazing treatments (timed grazing, ungrazed and business-as-usual). Soil moisture probes have been installed for continuous monitoring and soil biology, soil moisture, soil chemistry, pasture quality and quantity, and ground cover will be measured before, during and after the initiation of treatments. We predict that the treatments will stimulate biological activity within the soil leading to increased respiration and biological content, which in turn, will increase carbon sequestration in the soil. If successful, this has the potential to help mitigate climate change through increasing soil carbon sequestration and reducing emissions. We also predict an improvement in soil structure and condition, leading to better water infiltration and retention, and increased pasture cover. This will increase the reliability and quality of pasture, particularly in dry times, increasing resilience of farming enterprises to climate change.

Plants and herbivorous insects are locked in an evolutionary arms race in which plants deploy an arsenal of defences to which herbivores evolve counter-adaptations. How do plants perceive herbivore attack and respond accordingly? During herbivory, plants experience numerous stimuli including chemical signals either produced by the insect or microbes found in the insect’s gut and saliva. Considering insect-specific signals (e.g. mechanical damage) and microbial signals (e.g. microbe-derived chemicals) activate divergent signalling pathways, gut microbes may serve as insect symbionts by suppressing defences against herbivores. However, many grasses have evolved the ability to accumulate silicon (Si) from the soil to use for defence against insects and microbes. Moreover, Si-uptake is an inducible defence suggesting it is linked to plant perception of herbivore-associated signals.

By exposing plants to herbivore-associated stimuli, we have identified how different components of herbivory trigger specific responses in Brachypodium distachyon with and without Si. Upon wounding, leaves treated with oral secretions (OS) containing both microbe- and insect-derived elicitors from Helicoverpa armigera larvae increased the area of necrosis around wound openings compared to untreated wounds. Necrosis is characteristic of microbial pathogens in the OS that originate from the insect gut. Subsequent removal of microbes from the OS using antibiotic herbivore diets provided support for this. Moreover, Si-assisted wound healing by initially increasing wound shrinkage, after which necrosis area did not change, potentially indicating wound ‘sealing’. In conclusion, B. distachyon may overcome the herbivore symbionts by effecting a faster wound-healing response facilitated by Si, potentially decreasing susceptibility to herbivores.