A central goal of restoration ecology is to reverse degradation, re-establishing structural, compositional and functional features of ecosystems to some prior state. Initially, restoration requires the causes of degradation to be abated and may subsequently require interventions to reconstruct or reintroduce features of the prior state. Measurement of restoration requires a frame of reference to quantify change against management goals. Reference sites, representing plausible restoration targets, are key elements of the Australian Standard for Ecological Restoration. We applied this framework to assess outcomes of livestock exclusion to restore degraded grassy ecosystems in central NSW. We assessed restoration trajectories in plant species composition and soil properties based on a 17 year chronosequence of livestock exclusion relative to unfenced controls and ungrazed reference sites. Convergence and divergence were modelled as a function of time since livestock exclusion, initial state and environmental covariables. Livestock fencing initiated divergence from unfenced controls in both vegetation and soils, which was explained partly by the degree of difference between the initial and reference states. There was limited convergence with reference states in species composition, but this was variable between sites and unrelated to time since fencing. There was no evidence of convergence in soil properties. Fencing initiated the development of novel communities comprising common native and exotic taxa. The findings suggest site selection is crucial and reinforce the need to consider supplementary management actions, such as planting, scalping, direct seeding and weed control, in addition to livestock grazing exclusion, to restore woodlands towards desired states.

Conservation fences are increasingly used globally to protect threatened species. In addition to the conservation benefits within fenced reserves, spillover effects are an expansion of conservation benefits beyond protected area boundaries and are a potential benefit of fenced reserves. We studied a range of wildlife gradients caused by conservation fences by measuring the activity of birds, reptiles and mammals at various distances inside and outside the Arid Recovery Reserve fence. We recorded a range of fence-induced effects including spillover, barrier, step and source sink effects which combined to create a zone of overlapping wildlife gradients both inside and outside the fence. We termed this zone a Sepesphere and found that it extended for over a kilometre from the fence. Spillover effects were only recorded in two species of rodent and not in any reintroduced mammal species despite them being able to permeate the fence, at least as juveniles. Several groups of reptile had higher activity outside the fence than inside. Barrier effects were recorded in one native and one feral species where higher activity was recorded close to the fence compared with more remote sites. The various activity trends were related to each animal's ability to permeate the fence, their susceptibility to the threats outside the fence and the trophic cascades triggered by the removal of cats, foxes and rabbits inside. Results suggest that conservation fences can have varying and far reaching effects on reintroduced and resident fauna and that not all species will benefit from conservation fencing.

The arrival of novel predators can trigger trophic cascades driven by shifts in prey numbers. Predators also elicit behavioural change in prey populations, via plasticity and rapid evolution, and such behavioural response by prey may also contribute to trophic cascades. Here we document the effects of a novel predator on the behaviour and demography of a native prey species. We reveal rapid behavioural responses in the prey species (grassland melomys, a granivorous rodent) following the introduction of a novel marsupial predator (northern quoll). Within months of quolls appearing, populations of melomys exhibited reduced survival and population declines relative to control populations. Quoll-exposed populations were also significantly shyer than nearby, predator-free populations of conspecifics. This rapid but generalised response to a novel threat was replaced over the following two years with more threat-specific antipredator behaviours (i.e. predator-scent aversion). Predator-exposed populations, however, remained more neophobic than predator-free populations throughout the study. These behavioural responses manifested rapidly in changed rates of seed predation by melomys. Quoll-exposed melomys populations exhibited lower per-capita seed take rates, and rapidly developed an avoidance of seeds associated with quoll scent, with discrimination playing out over a spatial scale of tens of metres. Presumably the significant and novel predation pressure induced by the introduction of quolls drove melomys to fine-tune their behavioural response through time as the nature of the threat became clearer. These behavioural shifts could reflect individual plasticity in behaviour or may be adaptive shifts from natural selection imposed by quoll predation. Our study provides a rare insight into the rapid ecological and behavioural shifts enacted by prey to mitigate the impacts of a novel predator and shows that trophic cascades can be strongly influenced by behavioural as well as numerical responses.

Climate change is having measurable impacts on our land, water and biodiversity. In line with climate-driven impacts on the natural environment, there is a pressing need to understand and plan for the climate impacts on restored habitats and develop guidelines to implement and monitor climate adapted plantings and seed production areas. These areas have the potential to improve habitat resilience of native species within the wider landscape and provide empirical data to develop effective adaptive management strategies in the face of uncertainty. By including a diversity of species and seed provenances in revegetation projects, we can maximise the resilience of plantings to climate change and other environmental stressors.
In Victoria guidelines have been developed for climate adapted plantings to help restore native, biodiverse habitats under a changing climate. This information will link to national guidelines. Information includes selecting a diversity of species for restoration and nursery development, seed provenance, genetic profiling and climate trajectories for different climatic regions. The establishment of climate future plots will (a) validate the ‘climate readiness’ of different provenances (identified through genetic and climate modelling), (b) act as seed production areas (helping overcome the current risks of seed shortages), (c) identify how diverse ecological plantings influence faunal communities, and (d) provide demonstration sites for the public and restoration industry. This presentation will outline the development of these climate plots and discuss how information from experts and practitioners is being used to guide their implementation.

The impacts of invasive predators can be amplified by high densities of invasive prey species. This is evidently the case in Australia, where hyperabundant rabbit populations lead to high densities of feral cats and correspondingly high impact of cats on native species, especially small mammals. This suggests that reducing rabbits could also reduce abundance of cats, and thereby alleviate predation on native small mammals. However, cats might respond to the loss of rabbits by, at least temporarily, prey-switching to native small mammals, resulting in increased predation on those species.
We investigated the short-term effects of an experimental reduction of rabbit abundance on feral cats and their small-mammal prey in arid South Australia. We reduced the rabbit population in a 37 km2 experimental enclosure by ~ 80% (2,215 rabbits removed from an estimated population of ~2,800), while monitoring an adjacent unmanipulated area as a control. Cats were present in both the experimental and control areas.
This rabbit reduction had a major impact on cat diet, condition and survival. There was no evidence for an increase in the consumption of native rodents, though the activity of small mammals declined.
Our results demonstrate that individual feral cats prey-switch in response to removal of their primary prey. However, we also show that survival and overall activity of cats decreased, which could result in net, long-term benefits for native prey threatened by cats.

Biodiversity loss has long-term economic and ecological impacts. Preventing further damage and restoring the original biodiversity is therefore essential and one of the great challenges of ecology and conservation of our time. In my project, I aim to experimentally test methods to facilitate increase in the extent of area occupied by reintroduced western quolls (Dasyurus geoffroyi) in the Ikara-Flinders Ranges National Park/ South Australia. This is the intended conservation outcome of the project. My research is focussing on providing better knowledge about which ecological levers can be manipulated to achieve this outcome. I will answer these questions:

1. How do western quolls and feral cats select habitat temporally and spatially across a landscape gradient? With GPS-collars I investigate how western quolls and feral cats use the habitat from the rocky ranges to open plains. This helps to better understand whether the presence and absence of feral cats influences the habitat selection of western quolls.

2. Which pest control treatments are most effective at reducing local cat activity? Using landscape-scale manipulations, I compare the effects of warren-ripping vs. shooting of rabbits on local and landscape-level activity of cats and other warren-using fauna.

3. How does rabbit removal affect habitat selection of western quolls and feral cats? In areas of shared and potential occupancy of quolls and cats, I will use GPS-collars to investigate changes in habitat use of cats and quolls before and after experimental rabbit removal.

Australia spends more than $16 million annually on red fox control to mitigate their impact on our native fauna and reduce stock predation events. Despite this considerable investment, the effect of fox control on fox densities remains poorly understood, largely due to the relationship between traditional index-based measures of fox density (e.g. bait take rates, occurrence on motion-sensing cameras) and true fox density being complex and non-linear. Therefore, in order to produce robust estimates of fox density we need to develop new monitoring tools. Non-invasive genetic sampling offers a powerful alternative to traditional approaches and can produce robust estimates of fox density from analysis of DNA from the surface of scat.
The “Otway Ark”, a landscape scale fox control program, was established by Parks Victoria in late 2017. To evaluate the Ark, we surveyed transects for fox scat before and after baiting commenced - one in a baited site, and the other in an unbaited site. Scats were genotyped using microsatellites to identify individual foxes and this data was used to generate fox density estimates at several key periods throughout the control program. Despite declines in the relative abundance of fox’s post baiting (scats/km2), our results suggest a more modest impact of baiting upon fox density thus far. Additionally, individual identification of scats allowed for insight into individual persistence post-baiting and movement of foxes, parameters of key interest to land managers undertaking fox control. Our study demonstrates the power of genetic sampling approaches when evaluating the efficacy of predator control.

Populations of feral horses and deer are rapidly increasing in Kosciuszko National Park. Impacts on vegetation and soil stability have become highly visible and widespread. We investigated these impacts in the White Cypress Pine – White Box woodlands of the lower Snowy River valley. This woodland is a component of the White Box‐Yellow Box‐Blakely's Red Gum Grassy Woodland and Derived Native Grasslands complex that is nationally listed as a critically endangered ecological community. Exclosures and grazed plots were established and surveyed in 1984 and 1987. We resurveyed the plots to investigate the severity of ungulate impacts in the valley in 2013, 2017 and 2018. We compared the response of 24 environmental variables to ascertain recent herbivore impacts. While there was no evidence of horses or deer from dung surveys in 1987, in 2018, 84% of the dung was from horses, 13% from deer, 1% from rabbits and 2% from macropods. Total herbivore dung density increased fourfold since the 1987 survey. On the understanding that all plots had the same starting condition in 1984 with respect to prior herbivory, we deduce that horses and deer are having significant ecological impacts. There was a far greater cover of understorey plants and the midstorey was denser and taller inside the exclosures. Outside the exclosures, the vegetation cover was much sparser and soil erosion was active and extensive. Invertebrate count was nearly double inside the exclosures. The dense even‐aged regrowth overstory stands of White Cypress Pine have changed little in 34 years.