The cascading ecological effects of changes in abundance of an apex predator, the Tasmanian devil

Large carnivores have declined across most of the earth. After centuries of decline, however, some carnivores have begun returning to their former ranges. These changes in carnivore abundance have in some cases triggered trophic cascades and mesopredator release. The effects of predators arise through their direct consumptive effects, as well as behavioural effects that can in turn affect a species' fitness, and ultimately shape food webs. The Tasmanian devil (Sarcophilus harrisii) is the apex predator of Australia's island state of Tasmania. It has suffered severe population declines caused by a clonal, transmissible cancer, devil facial tumour disease (DFTD). Since its discovery in 1996, DFTD has spread across 80% of the devil's range, causing 80% population declines on average. In response to fears the species could go extinct, an insurance population of devils was established on the formerly devil-free Maria Island. The progressive spread of DFTD and the introduction of devils to Maria Island has created two concurrent natural experiments of opposing direction, allowing us to assess the ecological effects of rising and falling abundance of this apex predator. Through this thesis, I used one or both of these natural experiments to test the ecological and behavioural effects of rapid changes in devil abundance. First, I assessed changes in carcass persistence and consumption by experimentally placing carcasses with remote cameras across the gradient of devil population decline, ranging from long-term diseased areas to those without DFTD. Second, I assessed changes in the risk-sensitive foraging behaviour of a key prey species, the brushtail possum, before and after the introduction of devils to Maria Island, using a giving-up density (GUD) foraging experiment (a measure of perceived risk). Third, I combined remote camera surveys of the two natural experiments into a single study to investigate how rising and falling devil abundance affects community-wide temporal activity. Fourth, I used remote cameras to assess the cascading effects of devil population declines on community composition across Tasmania. Declines in devil abundance reduced competition for carrion. This allowed other scavengers, including the invasive feral cat, to increase carrion consumption. Although mesoscavengers increased their consumption of carrion, they were unable to functionally replace the devil. Carcasses persisted ‚Äöv†¬¿2.6-fold longer in long-diseased areas, highlighting an underappreciated ecological function of larger, more specialised carnivores. Changes to devil abundance triggered rapid behavioural shifts in other species. Previous work showed that before devils were introduced to Maria Island, GUDs of the brushtail possum were indistinguishable between the long-diseased area, where devils were rare, and Maria island, where devils were absent. This equivalence suggests that devils in long-term diseased areas are functionally extinct with respect to their influence on possum foraging behaviour. Three years after the introduction of devils to Maria Island, I show that possum GUDs were 64% higher on Maria Island than the long-diseased control region, indicating the rapid reinstatement of risk-sensitive foraging behaviours within a single generation. The temporal activity of mesopredators and prey also showed rapid shifts. At sites where devils had declined severely, the devil's closest competitor, the spotted-tailed quoll increased its use of the devil's temporal niche, strongly suggesting they have been released from interference competition. On Maria Island, key prey species responded to rapidly increasing devil densities by reducing their use of the devil's temporal niche, indicating a re-instatement of subtle behaviours that reduce the risk of encountering a devil. The direct and indirect effects of declining devil abundance have culminated in a restructuring of the food web. Devil population declines were associated with increased abundance of feral cats, which in turn had a negative effect on the southern brown bandicoot, a small mammal with similar characteristics to the many species already driven extinct by invasive mesopredators on the Australian mainland. This highlights the important role that apex predators can play not only in regulating the behaviour and abundance of native species, but also their capacity to confer ecological resistance to the harmful effects of invasive predator species, which are a major global driver of vertebrate extinctions. Importantly, the effect of devils on the alien feral cat was larger than their effect on a native mesopredator, the spotted-tail quoll, suggesting the limiting effects of apex predators might be stronger on alien species than on coevolved species. In sum, this thesis provides new evidence on the direct and indirect mechanisms by which apex predators structure communities and highlights their powerful but currently underutilised capacity to mitigate biodiversity loss due to invasive mesopredators.