Ecology and control of feral cats on a large island ecosystem

Invasive alien predators are a major cause of global biodiversity loss, and this is particularly true of feral cats (Felis catus) on islands. Invasive predator control has traditionally focused on lethal methods, such as trapping and poisoning, but these are problematic to implement at large scale, requiring ongoing resourcing and high effort. This has led to new focus on testing ecological and evolutionary levers to efficiently reduce invasive predator impact. Ecological mechanisms include restoring apex predators which suppress invasive mesopredators, controlling invasive prey to limit invasive predators, and restoring structural complexity in vegetation to increase refuges for prey. Lethal control of feral cats is most effective on islands, where reinvasion is restricted.

Eradication of a single pest species can have unpredicted and detrimental consequences where ecosystems are more complex, however. In many situations, both on islands and mainlands, there are multiple invasive and native species, both predators and prey. We need to base our approaches to eradicating or reducing the impacts of invasive predators in prior understanding of the mechanisms of interaction amongst the invasive and native predators and prey in the food web. This will enable the trajectory of changes to other species in the food web to be anticipated. This holistic approach to invasive predator management is needed both in cases of lethal invasive predator removal, more applicable to islands, and where leverage points are identified and tested to manipulate the ecological drivers of the invasive-native species relationships and abundances, more applicable to mainlands.

Using a large island case study, I investigate the potential ecological drivers of abundance and behaviour of an apex invasive predator, the feral cat, and a native mesopredator, the eastern quoll (Dasyurus viverrinus), and the effects of cat control on these and other native and invasive predators and prey. The purpose of the study is two-fold: 1) to understand the ecological dynamics on a large island with an intermediate complexity of the mammalian food web, with multiple native and invasive predator and prey species; and 2) to guide the lethal control of feral cats supported by ecological methods to reduce unintended and unpredicted consequences of predator removal. The case study is cat management on lunawanna-alonah/Bruny Island, a 362 km² island off the southeast coast of Tasmania, Australia. The island represents a case study in invasive predator management that is intermediate between the simple food webs on small islands and complex food webs of mainlands. Being an island, cat ecology and control can be studied at large-scale while edge effects and external influences are easier to control than would be the case on mainland sites. The Bruny Island Cat Management Program (BICMP) conducted foundational work from 2016 - 2022 to assess cat impacts and the effectiveness of control, as part of a feasibility study for cat eradication on the island. Substantial cat control was undertaken from 2018 to today in the vicinity of "The Neck" seabird colony, where migratory, colonially nesting short-tailed shearwaters (Puffinus tenuirostris) breed over the summer months.

The first aim of the thesis was to estimate the abundance and density of both the native and the invasive predators across the entire island and to determine the bottom-up or top-down factor(s) influencing their abundance, to infer predatory or competitive interactions between these predators and via their prey. The second aim was to test the effect of the two-year cat reduction operation on the density of feral cats and on the abundance of other mammalian predators and prey, both native and invasive. The third aim was to assess variation in spatial and movement behaviour of feral cats in contrasting environments.

First, by estimating both densities of feral cat and eastern quoll across an entire large island in relation to environmental variables, we provide evidence of spatial partitioning between high densities of an invasive and a native predator. Abundance of the native mesopredator was not limited by the apex invasive predator, despite cats being implicated in eastern quoll decline on mainland Tasmania. The spatial partitioning suggests competition as an underlying mechanism, mediated by bottom-up environmental factors. Our findings may represent a novel trophic cascade, based on exploitation competition, whereby biocontrol suppresses a highly fecund invasive prey species, the European rabbit (Oryctolagus cuniculus), which can then be kept at low density by a native mesopredator, the eastern quoll, which then prevents cats from reaching high densities and further protects quolls from cat predation. Bioclimatic suitability enables quolls to reach high density in the drier northern part of the island. In the wetter southern part of the island, cats may reach high density in the presence of an abundant sizeable prey, in this case the native pademelon (Thylogale billardierri), and lower bioclimatic suitability for quolls. Cats may be dependent on abundant populations of sizeable prey to reach high densities. Importantly, this system state may be recent, triggered by releases of calicivirus that cause rabbit haemorrhagic disease. Our findings highlight the fragility of invaded food webs and the importance of grounding conservation actions involving invasive predator control in a holistic understanding of ecological mechanisms.

Secondly, we show that colonially nesting, seasonally breeding seabirds create complex, subtle cascades of ecological effects among mesopredator and prey species. In this situation, cat control has limited top-down effects on other mesopredators and prey. Instead, there is a lagged response to the bottom-up productivity pulse provided by the breeding seabirds, with increased recruitment of native and invasive rodents. Our study demonstrates that cat density on islands, particularly around seabird colonies, can be very high, much higher than on mainlands, and that control using trapping is effective reducing cat density in this focussed landscape context. Our results highlight the complexity of the effects of control of an invasive predator on the broader ecosystem.

Finally, we provide the first insight into the spatial ecology and habitat selection of feral cats in Australian wet forests. Highly productive environments, even seasonally, correlate with smaller home ranges. Cats in low productivity environments travel further and intensify their use of anthropogenic features such as tracks and roads, features avoided by cats in productive environments. Our findings highlight the individual variation and behavioural plasticity of cats which have important implications for food web dynamics and wildlife management.

Overall, the results of this large island case study add to the growing body of evidence that demonstrates the potential for ecological methods such as control of rabbits to reduce the impacts of invasive predators. Ecological approaches when combined with lethal control, is especially useful on large islands with complex ecosystems and food webs, but has broader utility for large, unfenced landscapes.