The impact of cats and foxes on small terrestrial vertebrates and the control of feral cats at Heirisson Prong

Settlement by Europeans brought with it many changes to the Australian environment including: altered fire regimes; land clearing; hunting native animals; and the introduction of exotic diseases, herbivores and predators. Many of these factors have been proposed as causes of declines in the diversity and ranges of native vertebrates, especially mammals, although few have been studied experimentally. This study at Heirisson Prong in Shark Bay, Western Australia, examined the role of exotic predators in the decline of native fauna by investigating the predatory impact of feral cats and foxes on the small (<50 g) mammal and reptile fauna using both dietary analysis and a controlled field experiment. The study arose from perceived shortcomings in the knowledge of the impacts of these predators upon Australian native vertebrates and apparent difficulties in the control of feral cats in particular. Heirisson Prong was an appropriate site for such a study, as populations of feral cats and foxes were already being controlled to protect threatened mammals reintroduced to mainland Australia from offshore islands. The methods of predator control used at Heirisson Prong created three broad zones differing in cat and fox activity. This provided the opportunity to investigate the inter-specific interactions between feral cats and foxes and their predatory impact on small mammals and reptiles. Options for the control of feral cats were also explored by testing a range of trapping and poisoning techniques. Populations of three native species of mammals, two introduced species of mammals and three native families of reptiles were monitored one year before and three years after predator control. Fencing, shooting, poisoning and trapping were used to eradicate feral cats and foxes from a 12 km^2 area on the northern end of Heirisson Prong peninsula. Such intensive control led to a rapid decline in spotlight counts of foxes (from 0.162 ± 0.052 foxes.km^-1 in December 1989 to <0.010 foxes.km^-1 after December 1991) and a rise, followed by a slow decline in spotlight counts of feral cats (from 0.063 ± 0.034 cats.km^-1 in December 1989 to 0.131 ± 0.013 cats.km^-1 in February 1992, declining to 0.000 - 0.050 cats.km^-1 after July 1992). A buffer zone of 200 km^2 immediately south of the fenced area was baited twice a year with dried meat baits containing sodium monofluoroacetate (1080) to reduce the chance of foxes reaching the barrier fence. This level of control was effective against foxes, but not against feral cats. Spotlight counts of foxes declined (0.015 ± 0.010 foxes.km^-1 in March 1990 to 0.000 foxes km^-1 in April 1991) while spotlight counts of feral cats rose three-fold (0.059 ± 0.012 cats.km^-1 in April 1991 to 0.184 ± 0.081 cats.km^-1 in March 1994). Spotlight surveys of a third area, in which no feral cat or fox control was conducted, showed only seasonal fluctuations (peaks in summer, lows in winter). The effect of different densities of feral cats and foxes upon the abundance of small mammals and reptiles (< 50 g) was assessed using pitfall traps to monitor the capture rates of these animals. Concurrently, the diet of feral cats and foxes was assessed by analysing gut contents. Populations of rabbits (~1.5 kg), an exotic prey item for feral cats and foxes, were monitored during spotlight surveys of feral cats and foxes. Mammals were the most important group of prey to both predators, but native rodents were more important to feral cats than to foxes. The capture rates of native mammals increased following the removal of feral cats and foxes from Heirisson Prong (42 mammals trapped in June 1990 vs. 93 mammals in July 1994). declined by 80% where spotlight counts of feral cats rose following fox control (55 mammals in March 1990 vs. 7 in March 1994) and remained low, but steady, where feral cats and foxes were not controlled. The strongest responses were shown by the ash-grey mouse and the sandy inland mouse, which are known to spend up to 61% of their time foraging in open habitat, and were therefore more likely to encounter predators. Reptiles were not as important as native mammals in the diet of both cats and foxes and did not appear to show the same trends in response to changes in predator density. Rabbits were important prey for both feral cats and foxes and their populations appear to be regulated by both the onset of seasonal rainfall and predation by feral cats and foxes. Research into trapping and poison baiting techniques targeting feral cats assessed four different types of traps, a range of lures and four types of poison baits. The most effective trap tested at Heirisson Prong was the Victor Soft Catch® trap (No. 1.5) (0.86 feral cats per 100 trap nights) and the lure with the best performance used at trap-sites at Heirisson Prong was the social scent lure. Pro’s Choice. Four methods of poison baiting were tested on a radio-collared population of feral cats, ranging from nine cats in the first trial, to eight cats in the second and third, to seven cats in the fourth. The baits tested were dried meat baits, baiting rabbits to kill cats through secondary poisoning, a fishmeal-based bait and a semi-dried meat bait coated in the flavour enhancer Digest. Only one radio-collared cat, whose home range included a rubbish tip, died after eating a fishmeal-based bait. Such a low kill rate in this series of bait trials suggests that none of the methods of poison baiting tested could be recommended for the control of feral cats in semi-arid Australia. This study at Heirisson Prong represents the first experimental field study which manipulated densities of feral cats and foxes to show that feral cats can have a negative impact on populations of small native mammals. It showed that selective control of foxes alone was deleterious to populations of small native rodents, because feral cats, which included more small mammals in their diet than foxes, increased in abundance when relieved from the possible competitive and/or predatory forces exerted by foxes. Techniques are already established for the control of fox.es in many parts of Australia, but in the absence of an effective poison bait to control feral cats, options to control populations of feral cats are limited to trapping using Victor Soft Catch® traps (No. 1.5) baited with a social scent lure such as Pro’s Choice in small areas such as fenced reserves or offshore islands. Finally, monitoring both predator and prey populations should be considered as an essential element in all predator control campaigns to ensure that the desired outcome is achieved