FINAL REPORT FOR THE AUSTRALIAN GOVERNMENT DEPARTMENT OF THE

ENVIRONMENT AND HERITAGE

Interactions between feral cats, foxes, native carnivores, and rabbits in

Australia.

Published September 2004

Prepared by: Robley, A1., Reddiex, B1., Arthur T2., Pech R2., and Forsyth, D1.,

¹ Arthur Rylah Institute for Environmental Research

Department of Sustainability and Environment
PO Box 134
Heidelberg
Victoria 3084

² CSIRO Sustainable Ecosystems

Gungahlin Homestead
GPO Box 284
Canberra
ACT 2601

© Commonwealth of Australia (2004).

Information contained in this publication may be copied or reproduced for study, research, information or
educational purposes, subject to inclusion of an acknowledgment of the source.

This report should be cited as: Robley, A., Reddiex, B., Arthur T., Pech R., and Forsyth, D., (2004).
Interactions between feral cats, foxes, native carnivores, and rabbits in Australia. Arthur Rylah Institute for
Environmental Research, Department of Sustainability and Environment, Melbourne.

The views and opinions expressed in this publication are those of the authors and do not necessarily reflect
those of the Commonwealth Government or the Minister for the Environment and Heritage.

This project (ID number: 40593) was funded by the Australian Government Department of the Environment
and Heritage through the national threat abatement component of the Natural Heritage Trust.
Table of contents

EXECUTIVE SUMMARY AND RECOMMENDATIONS ......................................................................................... 1

1 BACKGROUND ...................................................................................................................................................... 4

2 OBJECTIVES .......................................................................................................................................................... 5

3 INTRODUCTION.................................................................................................................................................... 6
3.1 INFORMATION USED ............................................................................................................................................... 7
3.2 BACKGROUND TO PREDATOR AND PREDATOR–PREY INTERACTIONS..................................................................... 7
4 LITERATURE REVIEW...................................................................................................................................... 10
4.1 CHANGE IN ABUNDANCE OF PREDATORS ............................................................................................................. 10
4.1.1 Control of both feral cats and foxes (implications for primary and alternative prey)................................ 10
4.1.2 Control of foxes only (implications for primary and alternative prey) ....................................................... 13
4.1.3 Control of feral cats only (implications for primary and alternative prey) ................................................ 15
Summary .................................................................................................................................................................. 17
4.2 INTERACTIONS BETWEEN FERAL CATS AND FOXES ............................................................................................. 18
Summary .................................................................................................................................................................. 22
4.3 CHANGE IN ABUNDANCE OF PRIMARY PREY (RABBITS) ...................................................................................... 23
4.3.1 Effects of changes in abundance of primary prey on feral cat abundance and impacts on native prey ..... 23
4.3.2 Effects of changes in abundance of primary prey on feral cat diet and impacts on native prey................. 25
4.3.3 Effects of changes in abundance of primary prey on fox abundance and impacts on native species ......... 26
4.3.4 Effects of changes in abundance of primary prey on fox diet and impacts on native prey ......................... 27
Summary .................................................................................................................................................................. 28
4.4 INTERACTIONS BETWEEN NATIVE AND INTRODUCED PREDATORS, AND RABBITS ............................................... 32
4.4.1 Canids ......................................................................................................................................................... 32
4.4.2 Dasyurids.................................................................................................................................................... 32
4.4.3 Raptors........................................................................................................................................................ 33
4.4.4 Varanids...................................................................................................................................................... 34
Summary .................................................................................................................................................................. 34
5 INTERACTIVE MODELS OF PEST POPULATION DYNAMICS ............................................................... 35
Summary .................................................................................................................................................................. 49
6 IMPLICATIONS FOR INTEGRATED CONTROL ......................................................................................... 52

7 GAPS IN KNOWLEDGE ..................................................................................................................................... 53

7.1 PRIORITIES IN CURRENT GAPS IN OUR UNDERSTANDING ....................................................................................... 53
7.1.1 Further information requirements .............................................................................................................. 54
8 FILLING THE GAPS............................................................................................................................................ 55

9 REFERENCES....................................................................................................................................................... 57

10 ACKNOWLEDGMENTS ..................................................................................................................................... 68

APPENDIX............................................................................................................................................................. 69

Interactions between feral cats, foxes, rabbits and native carnivores

Figures and Tables
Figure 1. Some key processes that may affect interactions between predators and their prey........................7
Figure 2. Total response curves for (a) type II and (b) type III responses (from Sinclair and Krebs 2003). .....9
Figure 3. Distribution of a) fox, b) feral cat and c) European rabbits in Australia............................................20
Figure 4. Location and extent of a) feral cat (n =96) and b) fox control (n = 777) operations in Australia......21
Figure 5. Interactions in a simplified system. ..................................................................................................36
Figure 6. Prey dependent (dotted line) and ratio dependent (solid lines) functional responses. ....................37
Figure 7. Rabbit vegetation models.................................................................................................................38
Figure 8. Fox and rabbit spotlight counts from the Flinders Ranges...............................................................40
Figure 9. Rate of increase of foxes (r) during (a) the recruitment phase (spring – late summer) and (b)
the winter decline (late summer to spring), plotted against rabbit index of abundance...................41
Figure 10. Fox rate of increase from (a) peak to peak and from (b) trough to trough plotted against the
rabbit index mid-way between the peaks or troughs........................................................................41
Figure 11. Rabbit, cat and fox indices of abundance at Roxby Downs...........................................................42
Figure 12. Cat (open diamonds) and fox (closed squares) functional responses to rabbits in the Flinders
Ranges. ............................................................................................................................................43
Figure 13. Simulated population trajectories without additional density dependence in fox population
dynamics (i.e. the original Pech and Hood 1998) model (a & b), with density dependence
added (c & d, g = 0.0015).................................................................................................................46
Figure 14. Effects of rabbit control on fox population dynamics......................................................................47
Figure 15. Simulated population dynamics using the Pech and Hood (1998) model with feral cats
added. ..............................................................................................................................................48
Figure 16. Simulated population dynamics using the Pech and Hood (1998) model with feral cats
added. ..............................................................................................................................................49

Table 1. Examples of studies in Australia that have experimentally assessed the impact of predation on
rabbit population densities through manipulations of predator densities. T = Treatment, NT =
Non-treatment sites. .........................................................................................................................16
Table 2. Examples of comparative feral cat and fox diet studies in areas where rabbits are present. ...........22
Table 3. Examples of studies that have assessed the impact of changes in rabbit population densities
on predators and alternative prey. ...................................................................................................30

Interactions between feral cats, foxes, rabbits and native carnivores

Executive summary and recommendations

Through the Natural Heritage Trust, the fox control is inconsistent between studies and
Department of the Environment and Heritage may be confounded by inadequate monitoring
(DEH) is working to develop and implement techniques and behavioural changes.
coordinated actions to reduce damage to the
A potential cost of predator control is an increase
natural environment and primary production
in rabbit abundance, which may cause increased
caused by feral animals.
competition for food and other resources with
Predation by foxes (Vulpes vulpes) and feral cats native herbivores. Several studies suggest that
(Felis catus) have been identified as known or predators can exert prolonged regulating pressure
perceived threats to 34 and 38 native species, on rabbits at low densities and can impede
respectively, in threat abatement plans provided recovery of rabbit populations. Particularly when
for under the Environment Protection and populations have already been significantly
Biodiversity Conservation Act 1999 (EPBC Act). reduced through external factors such as disease,
Land degradation and competition with native drought, high or low rainfall, floods or warren
species by European rabbits (Oryctolagus ripping. However, predator manipulation studies
cuniculus) is also listed as a key threatening over a wide range of habitats have provided
process under the EPBC Act. inconsistent evidence of predator regulation of
rabbits. Predation appears to play an important
The aim of this report is to review the evidence of
role in regulating rabbit populations in arid and
the interactions between these three pest species,
semi-arid systems under certain conditions (e.g.
their control and the impact they have on
after drought has reduced rabbit populations), but
Australian native species. The objectives of this
has weaker effects in more temperate
report are:
environments or when environmental conditions
1. To determine the nature of interactions improve and rabbits escape regulation. It is
between feral cats and foxes (competition important to note that many of the studies that
and/or predation), especially in relation to have shaped our understanding of population
control of either or both species, and the regulation of rabbits in Australia were undertaken
associated impacts on native species and prior to the escape of Rabbit Haemorrhagic
ecological communities (especially those disease (RHD) in Australia. The potential
listed as threatened under the EPBC Act), regulatory effect of RHD on rabbit populations and
and feral rabbit populations within Australian the effect this could have on rabbit–predator
habitats/regions. interactions is largely unknown. The impact of
rabbits on flora and soils is well documented, but
2. To determine the implications of feral rabbit the impact on native mammal species is poorly
control to feral cat, fox and native prey understood.
populations, and the importance of rabbits for
maintaining high feral cat and fox numbers The impact of changes in predators and their
within Australian habitats/regions. primary prey on native mammal species has been
the focus of few experimental studies. Studies
3. To determine the interactions between feral
that have discussed the role of foxes and feral
cats, foxes and native carnivores and relative cats in regulating rabbit populations have largely
significance of competition and predation by not investigated the benefits or costs of predator
feral cats and foxes to these native species.
control to native species. Other studies that have
Based on the degree of overlap in distribution and investigated the impact of fox and cat control on
diet of feral cats and foxes, there is a potential for native mammal species have reported benefits
competitive interactions. There is circumstantial from pest control; however, there are many
evidence of foxes excluding feral cats from food acknowledged limitations of these studies. While
resources, and of foxes killing feral cats. No several studies have reported that fox removal
studies have experimentally demonstrated an has benefited a range of native species, many
increase in the rate of predation by feral cats on have not assessed pre-control population
native species following a reduction in fox parameters, do not have control sites, are not
abundance in Australia. Several studies have replicated, and have not attempted to test
described increases in cat abundance following alternative hypotheses to predation, such as
reductions in fox numbers resulting from control competition by herbivores. Also there are several
operations. However, the evidence for an notable exceptions to a general response to fox
increase in abundance in cat abundance following control (e.g. mixed responses of small mammal

Interactions between feral cats, foxes, rabbits and native carnivores 1

abundance from Operation FoxGlove WA, Project impact of foxes and feral cats on both rabbits and
Eden, WA and Project Deliverance, Vic). While native prey requires kill rates of these prey to be
the limitations cited above might have resulted assessed in relation to the availability of all prey
from limited budgets and logistical constraints types. This is particularly important for native prey.
associated with large-scale operations, the It is also important to understand the population
inferences that can be drawn are limited dynamics of native Australian prey and the
nevertheless. population dynamics of rabbits following the
arrival of RHD, in the absence of predation from
From the studies reviewed it is unclear what the
introduced predators.
impact of a decline in rabbits is on native species.
In the studies reviewed in this report, both feral The limited data available for temperate systems
cats and foxes shift consumption to the next most suggest fox population dynamics may not be
abundant prey item (e.g. invertebrates, reptiles, or linked as strongly to rabbit dynamics as they
birds) in the absence or decline of rabbits. There appear to be in semi-arid systems. Alternative
is no evidence that as a result of a decrease in models are thus required for temperate systems.
rabbits there is an increase in predation rates on These models will almost certainly require data on
populations of rare or endangered species. the interactions of predators and a wide variety of
foods. Feral cats are rarely seen in spotlight
The interactions between rabbits and predators in
counts in temperate systems and no quantitative
arid and semi-arid environments have been
numerical relationships can be established from
relatively well studied in comparison to more
the available data.
temperate parts of Australia. Our level of
understanding of these interactions and the Several studies have reported that integrated
impact on native species in arid and semi-arid and control (ripping, RHD or both poison baiting and
temperate environments is less well understood. RHD) has enhanced the decline of predator
In temperate environments the relationship species, but to our knowledge no studies have
between changes in rabbit abundance and investigated the costs and benefits of integrated
declines in either feral cats or foxes has not been feral animal control. A risk-averse approach would
clearly demonstrated. Also, no studies showed be to undertake integrated control wherever feral
that a decline in rabbit abundance leads to an cats, foxes and rabbits co-occur. However, this
increased rate of predation on native species. It may not be practical or possible due to limitations
appears that in systems where rabbits are not the on resources. At present we have no clear
staple prey item, changes in rabbit abundance understanding of the costs and benefits
have little impact on populations of feral cats or associated with integrated control programs.
foxes.
Despite a number of studies that have provided
Little quantitative information is available on the valuable insights into the impacts that changes in
interactions between introduced predators and prey abundance can have on populations of
native carnivores. Available data suggests that introduced predators, and how predators can
dingoes (Canis lupus dingo), may be capable of influence the abundance of prey species, there
suppressing fox populations, but that this is likely are many gaps in our understanding of predator-
to be mediated by specific environmental prey interactions.
conditions such as drought. There is some
The four main areas where further information
evidence to suggest that foxes spatially and
would improve our understanding of the
temporally avoid wild dogs and that only during
interactions between feral cats, foxes, rabbits,
times of limited resources do the two come into
their control and the impacts on native species
direct conflict. Similarly, there is a lack of
are:
knowledge on the impacts of feral cats and foxes
on native predators. 1. How to effectively monitor changes in
We used simulation models to explore the abundance of introduced predators,
potential interactions between rabbits, foxes and particularly feral cats. At this point in time we
feral cats. The sensitivity of the model to small are limited in our ability to control feral cats
changes in rainfall suggests a more detailed over large areas, although this is an area of
understanding of the relationships is required. current research.
More specifically, there is a need to quantify the 2. The impact of predator control operations on
relationship between rabbits and foxes and feral the population dynamics and social
cats. Numerical responses for the two predators organisation of sympatric predators and the
should be determined in relation to both the impacts on native species and communities.
abundance of rabbits (or juvenile rabbits) and
simultaneously the abundance of alternative food
sources. To properly quantify and model the

Interactions between feral cats, foxes, rabbits and native carnivores 2

3. The role of rabbits in temperate systems in conducted over appropriate temporal and spatial
supporting elevated numbers of foxes and scales is likely to produce important advances in
feral cats. our understanding of the interactions between
feral cats, foxes, rabbits, their control and native
4. The effects of disease (RHD and species. It is recommended that at the completion
myxomatosis), particularly in temperate of such studies the information gained is used to
environments, on the interactions between update the models of the systems as presented in
predators and their prey this review, that the results be peer reviewed and
A combination of focused research programs on made widely available, and the outcomes from the
the more tractable parameters of the above models should be used to direct management
identified gaps, and larger scale experiments strategies for these pest species.

Interactions between feral cats, foxes, rabbits and native carnivores 3

1 Background
The Department of the Environment and Heritage
(DEH) is the Australian Government’s major
environmental agency and is responsible for
achieving the Government’s environmental
objectives. Through the Natural Heritage Trust,
DEH is working to develop and implement
coordinated actions to reduce damage to the
natural environment caused by pest animals.
Since their arrival in Australia over a century ago,
introduced herbivores such as the European rabbit
Feral Cat (Felis catus) Photo: Department of Natural
(Oryctolagus cuniculus) and introduced predators Resources, Mines and Energy, Queensland.
like the feral cat (Felis cattus) and red fox (Vulpes
vulpes) are thought to be responsible for the
extinction or decline of a wide range of native
species. Foxes and feral cats have been
identified as known or perceived threats to 34 and
38 native species, respectively, in threat
abatement plans provided for under the
Environmental Protection and Biodiversity
Conservation Act 1999 (EPBC Act). Competition
and land degradation by rabbits is also listed as a
key threatening process under the EPBC Act.
Both State and Federal governments annually
commit significant funds to manage the impact that
these pest animals have on our environment.
Between 1992 and 1999 the Federal government
committed $4.7, $1.2 and $2.1 million to fox, feral
cat and rabbit research and control programs,
respectively.
Understanding the mechanisms that influence the Red Fox (Vulpes vulpes) Photo: P.Menkhorst.
abundance of these pest species, and the nature
of the interactions between pest species and
native species is critical to increasing our capacity
to manage the threats they pose, and to optimise
expenditure on pest animals management.

Dingo (Canis lupus lupus) Photo: DSE

European Rabbits (Oryctolagus cuniculus) in plague

numbers, South Australia. Photo: P. Bird.

Spotted-tailed Quoll (Dasyurus maculatus) Photo: DSE

Interactions between feral cats, foxes, rabbits and native carnivores 4

2 Objectives
Predation by the feral cat and the red fox, and
competition and habitat modification by the
European rabbit have been listed as threatening
processes by the Commonwealth under the EPBC
Act 1999. State legislation also recognises these
species as threats to biodiversity while control to
mitigate their impacts is conducted throughout
Australia. The role the rabbit plays in supporting
populations of feral cats and foxes, and the effect
of control of one or more of these species has in
altering their impact on native species, is poorly
understood. The aim of this report is to review the
evidence of interactions between these pest
species, their control and the impact on Australian
native species. The objectives of this report are:
• To determine the nature of interactions
between feral cats and foxes (competition
and/or predation), especially in relation to
control of either or both species, and the
associated impacts on native species and
ecological communities (especially those listed
as threatened under the EPBC Act), and feral
rabbit populations within Australian habitats
and regions.
• To determine the implications of feral rabbit
control on feral cat, fox and native prey
populations, and the importance of rabbits for
maintaining high feral cat and fox numbers
within Australian habitats and regions.
• To determine the interactions between feral
cats, foxes and native carnivores and relative
significance of competition and predation by
feral cats and foxes to these native species.

Interactions between feral cats, foxes, rabbits and native carnivores 5

 Foxes: The first reliable record of a successful fox
3 Introduction release was near Geelong, in 1871, where rabbits
had been released a few years earlier. The
Since the European settlement of Australia in
subsequent spread of foxes across Australia is
1788, 59 species (24%) of the mammalian fauna
closely linked to the spread of rabbits. Australian
have become rare, vulnerable or extinct (Short
studies on the food habits of foxes (Coman and
and Smith 1994). These extinctions and declines
Brunner 1972; Myers and Parker 1975a,b,
have not been spread evenly across the
Brooker 1977; Jones and Coman 1981; Catling
continent, with a greater number of extinctions
1988; Paltridge et al. 1997; Risbey et al. 1999)
being recorded in the semi-arid and arid parts of
highlighted the importance of rabbit in their diet,
Australia than in the more temperate areas
and data on the early spread of foxes suggested
(Woinarski and Braithwaite 1990).
the spread was more rapid where rabbits were
There has been considerable debate as to the present (Saunders et al. 1995).
cause of these extinctions. As early as 1856–57,
Feral cats: The timing of the arrival of domestic
observations were made of the decline in a range
cats in Australia is less clear. Baldwin (1980)
of native species. Finlayson (1961) writes on
suggested that cats could have been introduced
observations he made in Central Australia
to north-western Australia by Indonesian trading
between 1931–35 and 1950–56. He notes the role
vessels as early as the sixteenth century.
of the “three major scourges, the rabbit, the fox
However, Abbott (2002) reviewed historical
and the feral house cat”, and he describes their
sources and found no evidence that the cat was
impact as catastrophic, the rabbit by “competition
present on mainland Australia prior to settlement
for food plants and the latter two by direct
by Europeans. He reported that cats spread from
predation”. Evidence that predation by introduced
multiple coastal introductions in the period 1824–
predators is the primary cause of extinction and
86 and by 1890 nearly the entire continent had
decline in populations of native species has
been colonised. Abbott (2002) concluded that the
gathered momentum in the past decade (Dickman
evidence for early impacts of feral cats causing
et al. 1993; Short and Smith 1994; Smith and
major and widespread declines in native fauna is
Quinn 1996; Short 1998). This is mainly the result
considered tenuous and unconvincing.
of the experimental demonstration of the impacts
of predators on remnant populations of mammals Populations of feral cats were increased in the
(Kinnear et al. 1988, 2002; Friend 1990) and their nineteenth century by the planned release of
impact on reintroduced mammals (Friend 1990; thousands of feral cats (Rolls 1969) in an attempt
Short et al. 1992). to control mice (Dickman 1996), rabbits (Rolls
1969; Fuller 1970) and native rats (Bennett 1879
Rabbits: European rabbits first established in
in Dickman 1996). The impact of feral cats on
Australia near Geelong in Victoria, in 1859. The
native fauna has not been critically investigated,
rapid spread of rabbits across mainland Australia
but numerous historical and circumstantial
was probably aided by the presence of burrows of
accounts suggested that feral cats may have
native species, the lack of predators, changes
deleterious effects on native fauna (e.g. Dickman
made by the development of land for agriculture,
et al.1993; Dickman 1996).
and in some cases by their deliberate
transportation. Rabbits have caused significant Interactions: Understanding the dynamics of
damage to the environment directly, preventing predator–prey systems is fundamental to
the regeneration of some plant species, and effectively managing the threats and benefits that
indirectly by impacting on bird and mammal introduced predators pose to Australia’s
populations through altering vegetation conservation and agricultural values. The
community structure, damage to soils (erosion, interactions between feral cats and the red fox,
loss of fertility, and increased run-off). The and their reliance on introduced mammals,
impacts of rabbits have been most significant in especially European rabbits as their primary prey
the rangelands of central Australia, where are of key importance. Abundance of primary
numerous plant species and the animals that are prey can influence the extent to which these
dependent on them are threatened with extinction predators impact on secondary prey, many of
or are suffering range reductions (Williams et al. which are indigenous species.
1995). Rabbits can compete with sheep for
pasture, particularly when biomass falls below a Predator–prey interactions also have significant
threshold (estimated at below 259 kg ha-1, Short implications for Australia’s agricultural industries.
1985). The estimated impact of rabbits in lost For example, foxes can kill lambs (Lugton 1991;
production for the wool industry in 1989 was $115 Saunders et al. 1995: Greentree et al. 2000) but
million per annum (Williams et al. 1995). predation by foxes and feral cats can also
regulate rabbit numbers and help reduce their
impact on farm production by reducing

Interactions between feral cats, foxes, rabbits and native carnivores 6

 Dingo
competition with livestock and reducing the costs Contr
Contr
of rabbit control (Williams et al. 1995). Foxes Cats
RHD

3.1 Information Used

This review draws on many studies from a wide Contr Rabbits Native prey
range of environments across Australia. We have
included research programs or control operations
that were useful in addressing the objectives of
the study. Although we were not given approval Myxomato Vegetation Non-
to include several unpublished studies by some native
government organisations, a number of these are Fir
reputedly close to publication and should provide
Climat e
additional information soon.
The majority of information we review is from Figure 1. Some key processes that may affect
experiments or control operations that have either interactions between predators and their prey.
reduced or removed predators or their primary
prey. We do not review studies of the diet of feral The arrows represent the direction, but not the
cats or foxes per se as this does not provide direct strength of the interaction. Circular arrows indicate
evidence of the nature of the interactions between density-dependant regulation. Factors that disrupt
these species and prey species. However, we any one of these interactions can result in flow on
include studies that describe the diets of multiple effects to other parts of the system. For example,
predator species, or those that include shifts in reducing the abundance of foxes may influence
diet in response to changes in the abundance of the interactions between: feral cats, rabbits and
primary prey and predators, either via deliberate native prey; native prey and rabbits; rabbits,
experimental manipulation or as a result of native prey and vegetation; vegetation and native
management actions. These studies are included prey; and so on.
as they may provide information on the potential The interactions between predators and their
for competition between predator species. prey, and the implications for conserving species
We use models of trophic interactions to: a) that are threatened by predation have been
formally identify the various interactions and their extensively written about (Holt 1977; Sinclair et al.
inter-relationships, b) identify where there are 1990; Pech et al. 1992, 1995; Sinclair and Pech
gaps in our current understanding, and c) 1996; Pech and Hood 1998).
investigate the outcome of various management Predation can either limit a prey population’s
scenarios of integrated control and the potential growth, or it can act to regulate a population’s
benefits to native species. abundance. This review is not concerned with the
Finally, we outline areas for future investigations mechanisms that might limit a population, as all
including experimental designs to address the forms of mortality and reproductive loss set a limit
identified critical gaps in knowledge for various about which populations fluctuate (Sinclair and
geographical locations around Australia. Pech 1996). Understanding if predation regulates
prey does provide insights into the risks faced by
3.2 Background to Predator and populations of native species (i.e. small
populations are potentially at greater risk of
Predator–Prey Interactions extinction), and it also provides information on
This section provides a brief overview of how a population might respond to the removal of
predator–prey interactions, as these concepts predation pressure. The role that primary prey (i.e.
provide the background upon which the rest of the rabbits) plays in population increase of predators
report is based. and the flow on effect to alternative prey is also
important.
Untangling the complex nature of the population
dynamics of predators and their prey is difficult Understanding the interaction between feral cats,
(Figure 1). foxes and their prey relies on knowing how
changes in predator abundance (numerical
response) and the rate at which they depredate
prey (functional response) relate to changes in
prey density (Solomon 1949; Sinclair et al. 1990;
Pech et al. 1992). It also requires an
understanding of the role that primary prey
(especially rabbits) has in maintaining the

Interactions between feral cats, foxes, rabbits and native carnivores 7

abundance of both foxes and feral cats, and which prey is regulated by food. At higher levels of
whether this alters predation rates on alternative predation, prey rate of increase is positive
(native) prey species. between points B and C, if prey numbers fall
below point B prey species can be driven to
Two different types of functional response are
extinction. At very high levels of predation, prey
commonly used to describe the way in which the
will go extinct. These curves represent the
number of prey consumed per predator changes
situation where prey species are secondary prey
as prey density changes (Holling 1959; 1965). A
items and predators are reliant on some other
Type II functional response predicts that predators
staple prey(e.g. rabbits).
will have a progressively decreasing effect on
prey as prey abundance increases (i.e. inversely For the Type III total response (Figure 2b), there
density-dependent, Sinclair et al. 1990; Pech et al. are a number of outcomes from different predation
1992). A Type III response can be represented by rates. At low levels of predation the outcome is
a threshold S-shaped curve. At low-prey densities the same as in a Type II response and prey are
the proportion of prey consumed increases as not regulated by predation (point C). As predation
prey density increases (i.e. density-dependent levels increase there can be two possible stable
response). This arises through prey-switching points: firstly (points A, C), where predation is
behaviour or decreased social and territorial regulating prey at point A and food or some other
constraints of predators when prey becomes more factor at point C; and secondly a single stable
abundant (Pech et al. 1992). At high-prey density point (A) where predation alone is regulating prey,
the proportion of prey consumed slows, with a or prey are not able to persist due to predation
similar response to a Type II curve (Pech et al. rates being too high (level 4). Point B is an
1992; Sinclair and Krebs 2003). unstable threshold (level 2) where populations
move either towards point A or C.
The numerical response follows the basic shape
of functional responses, whereby the number of
predators increases in response to increases in
prey density through mechanisms such as
reproduction and immigration. Predator numbers
can also reach an asymptote due to reproductive
and social constraints such as territoriality and
emigration (Sinclair et al. 1990; Thompson 1994).
In practice, it is difficult with empirical data to
distinguish between Type II and Type III
responses. The total response of predators is the
product of the functional and numerical response
and is shaped by the types of functional and
numerical response experienced by the predator.
If there is no density dependence in either
functional or numerical response, then the
proportional effect of the total response is
uniformly inversely density-dependent and is of
Type II form (Figure 2). If there is density
dependence then the shape of the total response
is of Type III form and shows density dependence
at low-prey densities while remaining depensatory
at high-prey densities (Sinclair 1989; Sinclair et al.
1998).
A prey species’ instantaneous rate of increase
(i.e. the difference between net recruitment and
predation) can be used to assess the impacts of a
Type II and Type III predator total response.
Figure 2a shows prey rates of increase relative to
population size (number of prey) for a range of
Type II total predator responses (level 1–3).
Figure 2b shows the equivalent curves (level 1–4)
for Type III total response (Sinclair and Krebs
2003). In Figure 2a, at high-prey density, low
levels of predation (level 1) do not regulate the
prey species, and there is one stable point (C) at

Interactions between feral cats, foxes, rabbits and native carnivores 8

 a) b)

Rate of increase (r)

Rate of increase (r)

C A B C
B
1 1
2
2 3
3
4

Number of prey Number of prey

Figure 2. Total response curves for (a) type II and (b) type III responses (from Sinclair and Krebs 2003).
The instantaneous rates of change of the prey population experiencing different levels of (a) Type II and (b) Type III
predation. Point A represents a stable point from regulation by predators, point C a stable point due to regulation from
food with predation not regulating, and point B is an unstable threshold. Curves 1-4 represent different intensities of
predation: 1, Iowest predation level; 4, highest predation level.
Recent theoretical developments have highlighted Non-manipulative studies have demonstrated that
some of the limitations of functional and numerical predation plays a role in limiting primary prey
responses in understanding predator–prey populations, but they cannot be used to
dynamics (Alonzo 2002). These developments unequivocally assess whether predation is a
suggest that prey vulnerability can play an regulating factor because of the potential
important role in how predators maximise their confounding effect of other factors. Sinclair (1989)
foraging efficiency by selecting prey based on suggested that predator regulation of prey can be
poor anti-predator behaviour (Quinn and tested by removing predators, and then, after the
Cresswell 2004). prey has increased permitting predators to
reinvade. If predators are regulating prey
Prey species weigh up the cost of an activity
numbers, the return of predators should result in
against the risks of predation. The impacts of
prey populations returning to pre-predator removal
these decisions manifest in a reduced amount of
densities (assuming that all other factors are
time spent foraging, reducing the food intake of a
equal). Pech et al. (1995) and Krebs et al. (2001)
prey species, which in turn can act to reduce
described the possible manipulations of prey,
health and fecundity. For example, rodents and
which include changes in prey density through
gerbils reduced foraging and shift foraging activity
reintroduction’s, altering food supply or the
when the risk of predation was high (Brown 1988;
abundance of alternative prey species and
Kolter et al. 1991; Hughes et al. 1994) and
predators.
Antechinus species in Australia displayed different
foraging effort under risk of predation (Stokes et In addition to interactions between predators and
al. 2004; Arthur 2001; Arthur and Pech 2003). prey, predators that share food resources can
Arthur et al. (2003) showed that populations of compete, either via intraspecific competition or
house mice reproduced earlier and reached intraguild predation. Changes to the composition
higher densities in locations where the risk of of the predator assemblage can result in altered
predation was low compared to areas where it rates of predation on prey species (see section 4
was high. for further details).
We need to understand the types of interactions In complex systems with multiple predators and a
that currently exist. This includes the types of range of prey species, community food web
response predators’ experience from changes in models may prove to be more insightful (Chase
prey abundance that might change the abundance 2003; Navarrete and Castilla 2003). These are
of predators or alter the rates of predation on areas of active research and development that
native species. Several authors have suggested may provide increased understanding of the
that the best approach to determine which interaction between predators, prey and their
response describes the interaction between management in the future; however, they are not
predators and their prey is through perturbation discussed further in this report.
experiments (Sinclair 1989; Pech et al. 1995;
Cappuccino and Harrison 1996; Korpimaki and
Krebs 1996; Sinclair 1996; Krebs et al. 2001).

Interactions between feral cats, foxes, rabbits and native carnivores 9

4 Literature Review
Although predators have been regarded as
4.1 Change in Abundance of contributing to population control of low-density
Predators rabbit populations worldwide (Newsome et al.
1989; Trout and Tittensor 1989; Gibb and
The control of introduced predator species,
Williams 1990; Rogers et al. 1994), there have
particularly foxes, is widespread throughout
been few experimental tests in Australia to assess
Australia for both agricultural and conservation
whether predation is indeed a regulating factor
reasons. Preliminary results of a review of feral
(Table 1). Trout and Tittensor’s (1989) review of
vertebrate pest control operations throughout
the worldwide rabbit–predator literature suggested
Australia, commissioned by the DEH, indicate the
that predators do not have a regulatory effect on
magnitude of this control. For example, fox
high-density rabbit populations but may regulate
control was undertaken on at least 9 million ha of
low-density populations, in particular those
predominantly public land during 2002 (Reddiex et
populations that have been reduced by extrinsic
al. 2004).
factors.
If predation is regulating rabbit abundance, a
Study 1: Yathong Nature Reserve
reduction in the abundance of foxes (and feral
cats) could result in an increase in the abundance Newsome et al. (1989) and Pech et al. (1992)
of rabbits. Rabbits are well suited to respond reported a predator-removal experiment
quickly to the removal of predation, as they have a conducted at Yathong Nature Reserve, New
high rate of increase and respond rapidly to South Wales between June 1981 and January
improvements in environmental conditions. This 1984. Newsome et al. (1989) reported that low-
has potential ecological consequences that may density populations of rabbits increased rapidly
indirectly lead to impacts on native fauna through where foxes and feral cats were continually shot.
loss of vegetation, soil structure, and changes in After only 14 months, densities of rabbits at the
nutrient levels (Banks et al. 1998) or competition predator-removal sites were 3.5–4 times greater
for food (Williams et al. 1995). Dawson and Ellis than those sites where predators were not
(1979) found considerable overlap in the diet of controlled, whereas untreated populations had
the rare yellow-footed rock wallaby (Petrogale remained low. Pech et al. (1992) subsequently
xanthopus) and rabbits in western New South showed that when predators were allowed back
Wales, and Dawson and Ellis (1994) found into the predator-removal areas, rabbit
evidence of competition for food between rabbits populations continued to increase and did not
and red kangaroos. decline to the density in the untreated area.
These studies demonstrated that predators could
Fox control operations aimed at restoring native
regulate rabbit populations at low densities in
species or communities have been undertaken
semi-arid and arid habitats (less than 9–15 per
across Australia over recent decades. These
spotlight km), but that populations could escape
operations rarely control all pests (i.e. feral cats,
predator regulation and result in a higher stable
foxes and rabbits) or report on the presence of
state. The mechanism whereby rabbit populations
non-target pest species. This hinders the
are reduced to enable regulation at low density
interpretation on the impact that changes in
may vary across: the range of rabbits (eg. drought
rabbits abundance arising from predator control
reducing food resources) in the above studies; or
might have on native species.
disease such as RHD (Mutze et al. 1998;
Saunders et al. 1998); or conventional rabbit
4.1.1 Control of both feral cats and foxes control operations. The importance of
(implications for primary and environmental stochasticity in enabling regulation
alternative prey) of rabbit populations resulted in Newsome et al.’s
(1989) concept of ‘environmentally modulated
This section focuses on the effects of changes in predation’. Newsome and Sinclair (1995)
the abundance of feral cats and foxes on rabbit suggested that predator–prey dynamics in
populations and on native species or Australia are influenced by El Nino Southern
communities. Studies that have experimentally Oscillations (ENSO) causing wide environmental
manipulated predator populations receive most fluctuations (e.g. prolonged periods of drought).
attention (Table 1). Interactions between
introduced predator species following control of Study 2: Peron Peninsula
only one of a suite of predator species are also Project Eden is an ‘operational experiment’ with a
reviewed. primary goal of reconstructing the pre-European
fauna on Peron Peninsula (100 000 ha), Shark
Bay, Western Australia (Thompson and Shepherd

Interactions between feral cats, foxes, rabbits and native carnivores 10

1995). This area can be classified as semi-arid, native animals at Heirisson Prong, Shark Bay,
with a mean annual rainfall of 220 mm. In Western Australia, from 1990 to1994. This area
addition to predator-removal operations over a can be classified as semi-arid and coastal, with a
1050 km2 area, herbivores (goats, sheep and mean annual rainfall of 280 mm. The experiment
rabbits) have also been controlled. The relative comprised three different ‘predator zones’: zone
abundance of predators and larger native fauna 1) an area with low cat and low fox abundance
such as euros, echidnas, emus and goannas have following eradication of foxes and intensive cat
been monitored using track transects, where control on the northern tip of Heirisson Prong that
animal tracks are monitored over a 80 km was isolated by an electrified barrier fence (c. 12
transect, and small mammals have been km2); zone 2) an area with low fox abundance,
monitored on six grids using pitfall and Elliott comprising 120–200 km2 immediately adjacent to
traps. the barrier fence; and zone 3) an area with no fox
or cat control over an unspecified area adjacent to
Aerial baiting to control foxes commenced in
the previous zone. Spotlight surveys targeting
1995, resulting in an approximately 95% reduction
foxes and feral cats were undertaken at three-
in fox abundance, and subsequently an electrified
monthly intervals to assess changes in relative
barrier fence was constructed to prevent/reduce
abundance over the three zones. Rabbits were
reinvasion by foxes. Feral cat control commenced
also counted during these surveys. Pitfall traps
in 1996 and has involved a number of trapping
were monitored one year before and three years
and baiting regimes. Feral cats have been
after predator control began and were used to
maintained at approximately 30–50% of pre-
determine the relative capture success of small
control levels, a level that is believed to be too
mammals and reptiles between the three zones.
high to permit the establishment of many species
of small mammal (K. Morris, pers. comm.). Rabbit Control of foxes and feral cats in zone 1 resulted
population did not increase significantly following in low densities of foxes throughout the study
fox-control. It is possible that rabbits had escaped period (<0.05 km-1), whereas feral cat numbers
predator regulation prior to the commencement of initially increased following the initial drop in
fox control operations, and thus a reduction in spotlight counts of foxes, but then declined when
predators would not have affected the rabbit intensive cat control began. In zone 2, fox control
population. Presence/absence monitoring over resulted in fox abundance remaining low, but
eight years following the initial fox control spotlight counts of feral cats showed a three-fold
operation indicated a seasonal fluctuation in increase from 0.06 km-1 to 0.18 km-1 over three
rabbits, supporting the idea that rabbits were years. Both foxes and feral cats showed no trend
being regulated by food resources and not in abundance in zone 3.
predation. The release of myxomatosis and more
Indices of rabbit abundance were low (<1 rabbit
recently RHD on Peron Peninsula in 1996 may
km-1) in all zones prior to predator control, but
confound the effects of fox control on rabbit
increased in both of the predator control zones
populations.
once control commenced (to a maximum of
Trap-catch indices of small native mammals did 7 km-1). There was an apparent difference in the
not appear to have increased significantly after average annual rate of increase between zone 1
intensive predator-control. In 1995, trap success and zone 2 (r = 0.27, and r = 0.53, respectively;
for all small mammals ranged from 8 to18%, and calculated from Risbey 2000), with rabbits
increased to 40% in 1996 (seasonal rainfall increasing more rapidly in zone 2. Rabbits were
events may have influenced this measurement), periodically poisoned in zone 1 from 1993
but has averaged 10–25% since 1996, except for onwards, with the objective of killing feral cats and
2001 when trap success fell to 8% (K. Morris, foxes by secondary poisoning. Risbey (2000)
pers. comm.). Trapping along road transects has suggested that this may have accounted for low
also showed variable results, with some species rabbit counts in autumn 1992 and winter 1993.
apparently increasing (e.g. hopping mouse Rabbit abundance remained low in the zone
[Notomys spp.], goanna, bobtail skink and blue where no predators were controlled.
tongue skink), and others declining (e.g. bilbies
Risbey (2000) suggested that rabbit populations
[Macrotis lagotis] and woylies [Bettongia
at Heirisson Prong were regulated by the onset of
pencillata]). The relative abundance of reptiles
seasonal rainfall and predation from foxes and
and small mammals on trapping grids might be
feral cats; and that the removal of foxes in
influenced more by rainfall than predator-control
particular may have allowed rabbits to escape
(K. Morris, pers. comm.).
predator-regulation. However, the observed
Study 3: Heirisson Prong increase of rabbits following removal of predators
was extremely low compared with other studies
Risbey (2000) described a predator-removal
where regulation has been inferred (Pech et al.
experiment undertaken to protect reintroduced
1992; Banks 2000).

Interactions between feral cats, foxes, rabbits and native carnivores 11

Risbey et al. (1999) stated that their study 1999 at two levels of rabbit density (medium 13.7
presents the first experimental evidence that feral ha-1 and high 38 ha-1) in Risbey’s zone 1. They
cats have a negative impact on populations of found no significant overlap in diet and that
small mammals on the mainland of Australia. In bettongs were capable of shifting their diet in
zone 1, where foxes and feral cats were response to changes in resource availability while
maintained at low abundance, captures of small rabbits were far less flexible, resulting in a
mammals increased over the duration of the study dramatic population decline. Robley et al. (2002)
(42 captures in June 1990 to 93 captures in July found no difference in burrowing bettong body
1994). However, in zone 2, where only foxes were condition, reproductive output or resource use
controlled, small mammals (Ash-grey mouse during the same period.
[Pseudomys albocinereus]; Sandy inland mouse
Between 1992 and 1997, foxes gained entry to
[Pseudomys hermannsburgensis]; Little long-
the reintroduction site (zone 1) at Heirisson Prong
tailed dunnart [Sminthopsis dolichura]; and house
on three separate occasions lasting between
mouse [Mus musculus]) declined by 80% (55
several days and several weeks. On each
captures in March 1990 to 7 in March 1994).
occasion foxes killed between 36 and 77% of the
Small mammals were variable over the study
re-established population of burrowing bettongs
period in the experimental control zone, where
(Short et al. 2002). Foxes engaged in surplus
foxes and feral cats were not controlled. Changes
killing of bettongs, rarely consuming any of the
in indices of predator abundance did not appear to
carcasses. Foxes killed bettongs despite an
influence capture success of reptiles.
abundant rabbit population that, at the time of
The acknowledged limitations of Risbey et al.’s incursions in 1996 and 997, outnumbered
(1999) study included no replication of both bettongs by 350 and 700 to one, respectively
experimental and control study sites; incomplete (Short et al. 2002). This provides some evidence
sampling before the manipulation at all sites that predator-naive prey are highly vulnerable to
(experimental control was not monitored pre- predation regardless of the form of the response
manipulation); and the manipulation of rabbit predators might have to changes in prey
numbers was not replicated across all study abundance, and that even well-established
zones. These limitations prevent extrapolation of reintroduction populations are still vulnerable to
the results outside Heirisson Prong, but provide predation.
valuable insights into predator–prey interactions in
Study 4: Arid Recovery Program
this system; however, the influence of increasing
rabbit abundances after the removal of foxes (and The Arid Recovery Program is a joint conservation
feral cats) remains a confounding factor. CSIRO initiative involving Western Mining Company
has continued to control foxes and feral cats in Resources, Department of Environment and
zones 1 and 2, and monitor the changes in Heritage (South Australia) and the University of
abundance of prey species, but at the time of Adelaide, and was established in June 1997
writing, the data were not available for inclusion. (Moseby 2002). An area of 60 km2 was
progressively fenced to exclude rabbits, feral cats
Robley (1999) undertook a study in Risbey’s zone
and foxes, with all these species removed from
1 between 1995 and 1998 that investigated the
inside the reserve between 1997 and 2000.
interactions between rabbits and burrowing
bettongs (Bettongia lesueur); the later being Small mammal and reptile pitfall trapping from
reintroduced in 1992. Periodic rabbit control using sites inside (n=12) and outside (n=11) the reserve
1080 poisoned with one-shot oats was has been undertaken annually since 1998. In the
discontinued in 1996 with the reintroduction of first three years there was no reported difference
western barred bandicoots (Perameles between the average number of small mammals
bougainville). Fox and feral cat control continued captured inside and outside the reserve. In 2001
through poisoned meat baiting and shooting and 2002 a significant difference in the average
(Richards and Short 2003). Rabbit populations number of small mammal captures was reported,
increased 1 per spotlight km in 1995 to a peak of with a four-fold increase inside the reserve
13 rabbits per spotlight km in the summer of between 2001 and 2002, and in 2002 a similar
1997–1998. The rabbit population collapsed in the difference between captures inside and outside
months following to approximately 3 per spotlight the reserve (Moseby 2002). The author also noted
km in July 1998. This decline was aided by the an increase in the number of observations of Barn
introduction of RHD and the rabbit population Owls (Tayo alba), Frogmouths (Podargus
continued to decline and remained at less than 1 strigoides) and Boobook Owls (Ninox
per spotlight km to October 1998 (Robley et al. novaeseelandiae) inside the reserve, and while
2002). acknowledging the limitations. Moseby (2002)
suggested this might be expected given the
Robley et al. (2001) investigated dietary overlap
increased number of small mammals. The results
between bettongs and rabbits between 1996 and

Interactions between feral cats, foxes, rabbits and native carnivores 12

for reptiles are more complex and little can be Several other studies have focused on the
drawn from it at this stage. Moseby (2002) noted a benefits to native species of controlling foxes.
general decline in captures in 2001–2002 and Saunders et al. (1995) and Kinnear et al. (2002)
suggested that this may be linked to changes in reviewed predator-removal studies that have
vegetation structure resulting from the removal of attempted to quantify the removal of fox predation
rabbits. on native species. Some of these studies, which
have primarily been undertaken in Western
4.1.2 Control of foxes only (implications Australia, are summarised below.
for primary and alternative prey) Study 6: Mt Carolyn
Study 5: Namadgi National Park Kinnear et al. (1988, 1998) reported on a fox-
removal study at Mt. Carolyn commencing in 1979
Banks et al. (1998) described a predator-removal
where foxes were controlled using 1080 baits at
experiment that examined the role of fox predation
two colonies of rock-wallabies (Petrogale
in suppressing rabbit population growth in
lateralis), and not controlled at three other
Namadgi National Park, Australian Capital
colonies. Populations receiving fox control
Territory, between 1993 and 1995. This area can
increased four- to five-fold over 8 years, while
be classified as sub-alpine forest with reclaimed
those not receiving fox control remained at similar
pasture. The experiment involved suppressing
levels to pre-control levels in 1990, having
fox abundance on two sites, using 1080 poisoned
remained low or increased and then declined over
FoxOff™ over an 18 month period and comparing
the duration of the experiment. No information is
rabbit responses on these and two other non-
provided on the presence or abundance of
treated sites. All sites were approximately 10 km2
rabbits. Hone (1994) statistically analysed the
in area. In the two removal sites, foxes declined
data from Kinnear et al. (1988) using three
in abundance, and rabbit populations grew to 6.5
different approaches (ANOVA, changes in
and 12.0 times their initial population size within
abundance, and changes in rate of increase), only
18 months. In the untreated sites, rabbit
one of which demonstrated a statistically
populations showed a very small population
significant effect of fox control on rock-wallaby
increases over the same period (10%). An
populations. Kinnear et al. (1998) rebutted Hones’
interesting observation made by these authors
analysis by highlighting differences in each of their
was that other non-target predators of rabbits (e.g.
underlying assumptions for each of the tests that
feral cats, dingos], wedge-tailed eagles [Aquila
are described by Hone, and suggested an
audax]) appeared unaffected by the fox baiting.
alternative modelling approach to analysing the
The authors noted that feral cats were seen
data. Hone (1999) pointed out that despite the
infrequently throughout the study (<1 animal every
various approaches to analysing the data there
2 months).
are still potential alternative hypothesis that could
Banks (2000) reported on a follow up experiment explain the observed patters, and that it is the
at this site where foxes were permitted to re- acquisition of reliable knowledge that will improve
invade the two sites where they had previously our capacity to understand and manager
been removed (at the time of reinvasion rabbit threatening processes. Sinclair and Krebs (2003)
population densities were estimated at 44.2 and reviewed the results of Kinnear's studies and
21.6 per spotlight km). On one predator-removal showed that the rate of increase for these rock-
site rabbit populations declined immediately after wallabies was higher when predators were absent
foxes reinvaded and remained low for 16 months, than when they were present. They suggest that
suggesting that fox predation was effective at fox predation was inversely density dependant
regulating numbers. Banks (2000) suggested that and that foxes were treating rock-wallabies as
part of the observed decline in rabbit numbers secondary prey to some more abundant and
might have been attributable to changes in habitat persistent primary prey.
use following the reinvasion of foxes. However, on
Study 7: Dolphin Island
the other predator-removal site rabbit densities
dropped slightly following the reinvasion of foxes, Kinnear et al. (2002) reported a nearly thirty-fold
but then continued to increase (23%) over the (1 in 1979 to 27 in 1990) increase in Rothschild’s
following 16 months, suggesting that rabbits were rock-wallabies (Petrogale rothschildi) following fox
not regulated by fox predation. control on Dolphin Island, Western Australia.
Over the same time period, rock-wallaby numbers
Spotlight counts, including indices of several
remained constant on fox-free Enderby Island.
native species, have been continued to the
Counts were undertaken over a single period of
current day at Banks’ study sites by Environment
several days, and no data on trends in fox
ACT. These unpublished data are not available for
abundance, changes in causes of mortality,
discussion in this report, but one is incorporated
alternative predators or underlying environmental
into the trophic interaction model (see section 5).

Interactions between feral cats, foxes, rabbits and native carnivores 13

conditions are presented. Hence it cannot be Operation Foxglove involved aerial baiting of
concluded that fox predation regulates the foxes over 400 000 ha of northern jarrah forest
abundance of these rock-wallabies. from 1993 to 1999 (de Torres 1999). The control
was largely designed to determine the optimal
Study 8: Tutanning Nature Reserve
frequency of poison baiting (unbaited versus 2, 4
At Tutanning Nature Reserve (2200 ha), brush- and 6 baitings per year), with fauna responses to
tailed bettongs (Bettongia penicillata) increased baiting monitored at each site. Monitoring
from 7 captures in 1984, prior to fox baiting, to 64 included sand plot indices for fox abundance,
in 1989 following 5 years of intensive fox control. radio-telemetry of translocated populations of
Numbers of common brushtail possum woylies, radio telemetry of common brushtail
(Trichosurus vulpecula) and tammar wallaby possums, and trapping and spotlight counting of a
(Macropus eugenii) also increased over this suite of native fauna. Preliminary results in
period, and it was suggested that fox control regards to native fauna response to fox baiting are
enabled the burrowing bettong to inhabit and inconsistent, as radio-telemetry monitoring has
reproduce successfully in a larger part of the shown significant differences in survivorship of
reserve (Kinnear et al. 2002). translocated woylies between treatments, while
trapping data showed no significant difference in
Study 9: Dryandra State Forest abundance between treatments for the three most
An intensive fox removal program in Dryandra frequently trapped native mammals.
State Forest, Western Australia, where control High-intensity (6 baitings per year at 5 baits per
was undertaken monthly over 5 years, resulted in km2) fox control in northern jarrah forest in
a significant increase in numbat (Mymecobius
Western Australia over 6 years has not resulted in
fasiatus) in the baited area, but not in the unbaited a population increase of quokkas (Setonix
area, and burrowing bettongs also appeared to brachyurus). Low recruitment has been
increase in the baited area (Friend 1990).
suggested as the cause for the lack of response in
Study 10: Jarrah Forest this species (Hayward et al. 2003). However,
habitat preferences are suggested as a potential
Morris et al. (1995) reported on an experiment limiting factor with no evidence presented to show
that investigated the impact of foxes on western that foxes have been killing quokka.
quoll (or chuditch; Dasyurus geoffroii) in a jarrah
forest of Western Australia. Fox baiting using Study 12: Red-tailed Phascogale
1080 baits was undertaken from 1991-1994 over
Friend and Scanlon (1996) reported on the effect
an initial area of 4 000 ha, which was later of fox control on populations of red-tailed
increased to 17 000 ha. Ten chuditch were phascogale (Phascogale calura) in the Western
monitored using radio-collars for the first 12 Australian wheatbelt. Trapping grids were
months and then by trapping only. In the first 12 established to monitor numbers of red-tailed
months following fox baiting, trap success rates
phascogale on nine reserves in 1993, two of
were stated as increasing significantly. Trap which received no fox control, three received fox
success in the unbaited area was low (0–1.2%) control since 1985, and four received fox control
compared to the baited area (1.2–8.6%), however,
since 1994. Trap success data from 1994 to1996
trap success was not assessed prior to control suggested that fox control benefited populations
commencing in the baited area and trap success of red-tailed phascogale; however, it was also
on the unbaited area was an order of magnitude
noted that rainfall and population abundance from
less than on the treated site. It is therefore the previous year are strongly related, which
possible that chuditch abundance may have obscures the effect of other factors.
naturally varied between baited and unbaited
areas and the observed differences may not Study 13: Western Shield
necessarily be a result of fox predation. Morris et
Project Western Shield, is a fox management
al. (1995) also reported an increase in trap
program that involves the application of poison
success rates of brush-tailed bettongs (or woylie,
baits (4 baitings per year at 5 baits km2) on 3.6
Bettongia pencillata), common brushtail possum
million ha of Western Australia. This project was
and southern brown bandicoots (or quenda,
formally launched in 1996 and is currently in
Isoodon obesulus) in the baited areas, and in
operation. Native species (particularly small
addition to changes in numbers these species
mammal) abundance is monitored at 40 sites,
broadened their distribution following fox control.
using a range of techniques including trapping
Study 11: Operation FoxGlove and spotlighting, and nest boxes to monitor fauna
not readily trapped. Mammal reintroductions have
During the 1990’s the impact of fox control on
also occurred at sites where foxes are being
native fauna in Western Australia has also been controlled, and these reintroductions are also
investigated in two large control operations. monitored. At the time of writing the results of this

Interactions between feral cats, foxes, rabbits and native carnivores 14

project were being prepared for publication by the the consequences for rabbit, and cat populations,
Department of Conservation and Land and their impacts on ‘at-risk’ species despite
Management, Western Australia (CALM) and decades of publicly funded research. Some of this
were not available for inclusion in this review. uncertainty may be addressed if funding providers
insisted on peer-reviewed publication of results.
Study 14: Project Deliverance
In 1995, Project Deliverance was established in 4.1.3 Control of feral cats only
Eastern Victoria. This project aimed to measure (implications for primary and
the response of medium-sized native mammals to
broad-scale fox control. Three locations, the West alternative prey)
Coast, East Coast and Stony Peak sites were Feral cats are thought to have had a destructive
established (A. Murray, pers. comm.). Each impact on a wide array of native vertebrate fauna
location comprised a poisoned ’treatment site‘ and (Atkinson 1985, Dickman 1996). The impact of
a paired unpoisoned ‘non-treatment’ site. 1080 feral cats have been most obvious on islands
FoxOff bait was buried in bait stations spaced at 1 (Nogales et al. 2004), in part, as native species
km intervals, with baits being replaced every 3-4 have evolved in the absence of such a predator,
weeks. Non-poisoned baits were laid at the same and due to a general lack of appropriate anti-
rate and intensity on the non-treatment sites (A. predator behaviour.
Murray, pers. comm.). The sites were between
7000 and 14000 ha and were paired to match Feral cats either acting alone or in association
dominant vegetation community and structure, with factors such as rabbits or rats and mice, has
topography and geographic location. been considered responsible for the local
extinction of a number of species on islands. For
Medium-sized mammals were cage-trapped on example, predation by feral cats, that were
both treated and non-treated sites by placing 60 supported by high numbers of introduced rabbits,
cage traps at 300 m intervals along a single was the mechanism that caused the extinction of
transect in each treatment and non-treatment site. the Macquarie parakeet, (Cyanoramphs
Traps were operated for several days four times novaezelandiae erythrotis) on Macquarie Island
per year from 1998 to 2003. (Taylor 1979). This pattern is considered to be
Preliminary results from this project indicate that widespread across Australian islands (Burbidge
long-nosed potoroos (Potorous tridactylus) and 2002).
southern brown bandicoots (Isoodon obesulus) The eradication of feral cats from islands has
may have responded positively to the fox control been achieved a number of times on islands
treatment. around Australia and the world (Algar et al.
Summary 2002Harper and Dobbins 2002; Wood et al. 2002;
Rauzon et al. 2002). The impact of feral cats on
The assumption underlying these fox control mainland Australia is poorly understood. They are
operations is that species thought to be at risk will cited as the primary cause for the failure of a
respond to a reduction in fox abundance. While reintroduction program of Rufous hare-wallabies
there have been increases in population estimates (Lagorchestes hirsutus) in the Tanami Desert in
for some species at some locations, the response Western Australia (Gibson et al. 1994;
has been variable. Many of the studies reviewed Christensen and Burrows 1995).
have not assessed pre-control population
parameters, lack control sites, have no replication, Despite several years of research (Algar and
do not attempt to assess changes in fox and/or Sinagra 1996; Algar et al. 1999), control programs
cat abundance, and plausible alternative aimed at reducing feral cat abundance are
hypothesis remain untested. Little is known on the hindered by a number of factors, including the
impact of changes in rabbit abundance following limited reliability of the available indices to
predator control on native species, and the accurately assess changes in abundance, and by
importance of differences in underling prey the lack of efficient control techniques.
vulnerability to predation. We know of no control operations or research that
The rock-wallaby studies in Western Australia has specifically investigated the relationship
serves to highlight the difficulty in interpreting the between feral cats and rabbits in isolation from
results from studies into predator–prey foxes and/or dingoes. It is unclear whether or not
interactions, and the need for well designed and feral cats are capable of regulating rabbits, or that
implemented programs that will enable robust feral cat populations show a numerical response
analysis resulting in reliable information. to a reduction in rabbit numbers. It is also unclear
what effect feral cat control alone has on
There still remains uncertainty of the effect of fox populations of native species.
control on ‘at-risk’ species and uncertainty about

Interactions between feral cats, foxes, rabbits and native carnivores 15

Table 1. Examples of studies in Australia that have experimentally assessed the impact of predation on rabbit population densities through
manipulations of predator densities. T = Treatment, NT = Non-treatment sites.

Study Location Study area Habitat Duration of Experimental Techniques T & NT Replication Key rabbit population density Factors other
study treatment used sites changes than predation
(years) considered

b
Newsome New South 3 sites, Semi-arid 2 Fox and cat Rabbit and Yes n=1–2 Increased 11.7 times on T sites Drought
2
et al. (1989) Wales 50–180 km removal predator compared with 2.8 times on NT
per site abundance sites Food supply

b
Pech et al. New South 3 sites, Semi-arid 5 Fox and cat Rabbit and Yes n=1–2 T site populations remained Drought
2 a
(1992) Wales 50–180 km removal , followed predator higher than the NT sites, despite Myxomatosis
per site by no predator abundance, reintroduction of predators
removal Predator diet

Banks et al. Canberra 4 sites, 10 Sub-alpine 2 Fox removal Rabbit and Yes n=2 T sites increased 6.5 and 12 None
2
(1998) km Forest predator times compared to 2 times and a
per site abundance decline on NT sites

Banks (2000) Canberra 4 sites, 10 Sub-alpine 1½ Allowed foxes to Rabbit and Yes n=2 One T site declined and None
2
km Forest re-invade sites predator remained low following predator
per site abundance reinvasion. Other T site
declined, then increased by 23%
Risbey (2000) Heirisson 3 sites, 120 - Semi-arid 5 Fox and cat Predator Yes No Two T sites increased while no Rainfall
2
Prong >200 km removal abundance change on NT site.
a b
Predator removal was carried out by Newsome et al. (1989) Level of replication changed during the study.

Interactions between feral cats, foxes, rabbits and native carnivores 16

 test alternative hypotheses, such as competition by
Summary herbivores.
Improving the level of reliable knowledge on the Several studies have reported that fox removal has
interactions between feral cats, foxes and rabbits benefited a range of native species. However,
will increase our capacity to manage the impact of there are several notable exceptions (e.g. mixed
predation on populations of native species and responses of small mammal abundance from
when to undertake integrated control. Operation FoxGlove, Project Eden, and Project
A potential cost of predator control is the release of Deliverance).
rabbits from regulation resulting in a numerical The only site we reviewed where foxes, feral cats
increase in rabbit abundance, which may cause and rabbits were controlled was the Arid Recovery
increased competition for food with native Program in South Australia (Moseby 2002). A
herbivores (Dawson and Ellis 1979; Dawson and positive response in small mammal species was
Ellis 1994). The damage by rabbits is well reported at this site; however, we have not seen
documented (Williams et al. 1995), but the impact detailed results from the study. An interesting
of rabbits on native species is poorly understood ( observation of the study was the reported increase
Robley et al. 2001, 2002). in avian predators and the shift in the reptile
Several studies suggest that predators can exert community structure, possibly in response to
prolonged regulating pressure on rabbits at low changes in vegetation brought on by the removal
densities and can impede recovery of rabbit of rabbits (Moseby 2002). This highlights the
populations. This is particularly so when those complex and interactive nature of ecosystem
populations have already been significantly management.
reduced through external perturbations of density- One of the major limitations in the manipulative
independent extrinsic factors such as disease, experiments described above is the short time
drought, high or low rainfall, floods, and warren frame of the studies. For example, Banks (2000)
ripping (Newsome et al. 1989). However, predator monitored rabbit and predator populations for 16
manipulation studies over a wide range of habitats months following re-colonisation of study sites by
have provided inconsistent evidence of predator foxes; however, this period only covered one
regulation of rabbits. Predation appears to play an breeding season for foxes. Ideally, manipulation
important role in regulating prey populations in studies should cover a time period sufficient to
some arid systems under certain conditions (e.g. allow variation in temporal trends to be accounted
after drought has reduced rabbit populations), but for in the analysis. Also, many studies lack either a
has weaker effects in more temperate treatment and non-treatment comparison or at
environments or when environmental conditions least before and after measures of abundance.
improve and prey escape regulation. In contrast, The latter is not ideal, as direct comparisons are
in New Zealand, Reddiex (2004) found that confounded by temporal changes (e.g. rainfall from
predation, at least when combined with RHD, is a the previous year, Friend and Scanlon 1996).
significantly stronger process in temperate than Another difficulty is that most studies reviewed did
semi-arid regions. It is important to note that many not quantify the abundance of alternative predators
of these studies were undertaken prior to the (e.g. feral cats or native predators) or whether they
escape of RHD in Australia. The potential were even present in the study areas. The
regulatory effect of RHD on rabbit populations and potential influence of these species may confound
the effect this could have on rabbit–predator the interpretation of the reported results.
interactions are largely unknown.
There is a need for more manipulative experiments
The impact of changes in predator and primary to examine the role of predation in regulating
prey abundance on native mammal species has rabbits, particularly in temperate environments,
been the focus of few experimental studies. and the impacts a range of rabbit abundances
Several studies that have discussed the role of have on native prey species. These studies need
predation (feral cats and foxes) in regulating rabbit to control for the effects of other regulating factors
populations have not investigated the benefits or if they are to test unequivocally the role of
costs of predator control to native species. Other predation in altering rabbit abundance and the
studies that have investigated the impact of fox impacts on native species and/or communities.
and cat control on native small mammal species These studies need to operate at a spatial and
reported benefits from pest control; however, there temporal scale that encompasses the full range of
are many acknowledged limitations of these environmental factors that are likely to influence
studies. Many of the studies did not assess pre- the dynamics of rabbit populations in a given area,
control population parameters, did not have control and those of their predators. However, it may be
sites, were not replicated, and did not attempt to difficult to temporally or spatially manipulate some

Interactions between feral cats, foxes, rabbits and native carnivores 17

potential regulating factors (e.g. diseases) on an The majority of studies reviewed reported that the
appropriate scale. diets of both feral cat and foxes included rabbits,
birds, reptiles, rodents, invertebrates, and plant
Alternatives to large-scale experimental
material. Most of these studies have been
manipulations include engaging management
undertaken in systems where rabbits were the
agencies to use control operations as large-scale
dominant dietary species (i.e. primary prey; but see
experiments (Walters 1986, NSW NPWS 2001,
Sandell 1999).
Robley and Wright. 2003). Another approach is the
development of predictive population models. Several studies observed that the relative
These models can be used to explore the importance of secondary prey species varied
sensitivity of the various parameters that are between predators (Bayly 1978; Catling 1988;
critical to predator–prey dynamics, and identify Risbey et al. 1999). For example, Risbey et al.
areas on which management and experiments (1999) found that sheep carrion and invertebrates
should focus. were more prevalent in the diet of foxes, and native
rodents, birds and reptiles were more prevalent in
4.2 Interactions Between Feral Cats the diet of feral cats. Molsher (1999) reported an
and Foxes overall dietary overlap between foxes and feral
cats of 75%, but also reported that foxes and feral
In Australia, little is known of the relationships cats used many of the same prey types in different
between feral cats and foxes. Mesopredator proportions. The removal of foxes from one of
release (sensu Soulé et al.1988) occurs when a Molsher’s (1999) sites resulted in a significant
dominant predator is reduced in abundance, thus increase in the frequency of carrion consumption
allowing a population increase in lower-order by feral cats compared with the sites where foxes
predators that results in an increase in predation were not controlled, but there were no other
on shared prey species. Mesopredator release significant differences for any of the other prey
has been documented in a range of studies. For types. It is believed that foxes and feral cats co-
example (in the following format: dominant exist in many areas due to specialisation of
predator, lower order predators, prey): Coyotes different age classes of rabbits, with feral cats
(Canis latrans), foxes (Vulpes spp.), skunks focusing on juvenile rabbits and foxes on adult
(Mephitis spp.), domestic cats and small birds rabbits (Catling 1988). Jones (1977) and Liberg
(Soulé et al. 1988; Estes 1996); Iberian lynx (Felis (1984) have also reported a preference of feral
pardina), mongoose (Herpestes ichneumon), and cats for juvenile rabbits.
rabbits (Palomares et al. 1995); coyote, red fox,
The potential for dietary competition between foxes
and duck (Sovada et al. 1995); coyote, badger,
and feral cats may be exacerbated in periods such
gray fox, bobcats, jack-rabbits and rodents (Henke
as droughts or following disease outbreaks (e.g.
and Bryant 1999); coyote, gray fox, cat, opossum,
myxomatosis or RHD) where food resources,
and scrub-feeding birds (Crooks and Soulé 1999).
particularly young rabbits, become limited. Few of
Several other studies have inferred potential
the studies listed in Table 2 attempted to compare
mesopredator release from changes in abundance
diet with prey availability, and those that have
of sympatric predator species (e.g. Dekker 1986;
focused on rabbit abundance (e.g. Molsher 1999;
Litvaitis and Harrison 1989; Sargeant et al. 1993;
Read and Bowen 2001; Holden and Mutze 2002).
Lindstrom et al. 1995). The key mechanism that
Molsher (1999) reported a positive correlation
may enable a dominant predator to have an effect
between the occurrence of rabbit in the diet of both
on mesopredator populations is interspecific
feral cats and foxes and rabbit abundance. Read
competition (e.g. exploitation or interference
and Bowen (2001) reported that diet between
competition) and/or intraguild predation (e.g. direct
foxes and feral cats were similar when rabbit
predation).
numbers were high, but varied substantially when
The potential for mesopredator release of feral cats rabbit numbers were low. Holden and Mutze
as a result of changes in fox abundance is of most (2002) investigated the impact of RHD on feral cat
concern, but poorly understood. These species and fox diet in South Australia. Fox diet changed
overlap in their distribution (Figure 3), there are to include substantially less rabbit, and more
many areas where fox control is undertaken invertebrates and carrion, while there was little
throughout Australia and where these species change in cat diet.
co-occur (Figure 4) and very few control operations
In general, diet studies showing a high degree of
directly target feral cats. The potential for
overlap support the concept of interspecific
interspecific competition between foxes and feral
competition between feral cats and foxes, but they
cats is supported by the numerous studies that
do not demonstrate such a relationship.
have described a strong dietary overlap between
the two species from the same sites (Table 2). Dietary studies aside, little is known about the
interaction between feral cats and canids (Dickman

Interactions between feral cats, foxes, rabbits and native carnivores 18

1996). Dingoes have been reported to directly 1994 and 1997. The extent of the perceived
predate feral cats (Corbett 1995) and foxes interspecific competition was investigated using a
(Coman 1973; Molsher 1999; Risbey et al. 1999), fox-removal experiment. The study area comprised
however, the incidence was extremely low. From a two treated areas, foxes were removed on one and
sample of 15 dietary studies covering Victoria, New reduced by 50-75% on the second, and two
South Wales, Northern Territory and Western untreated areas, where foxes were not controlled.
Australia, and comprising 2133 fox stomachs and
7718 fox scats, the remains of feral cats occurred
only four times. There is no evidence to determine
whether these were a result of direct predation or
scavenging dead individuals, though Risbey et al.
(1999) suggested that direct predation was likely in
their study.
Several studies have described increases in cat
abundance following reductions in fox numbers
resulting from control operations (Algar and Smith
1998; Catling and Reid 2003), and following local
declines in dingo abundance in Queensland
(Pettigrew 1993). Christensen and Burrows (1995)
observed a three-fold increase in the abundance of
feral cats in the Gibson Desert, and suggested this
increase was likely a result of exceptional rainfall
resulting in a large numerical response of rabbits.
Catling and Burt (1995) have also reported that the
abundance of feral cats is negatively correlated
with both foxes and dingoes at a site in New South
Wales. Read and Bowen (2001) did not manipulate
predators, but reported that cat abundance peaked
when fox numbers were low and when rabbit
numbers were relatively high.
Risbey et al. (1999) suggested that fox control at
Heirisson Prong could lead to a mesopredator-like
response resulting in an increase in cat
abundance. This prediction was based on the
observation of cat remains in fox diet studies, and
hence it was inferred that intraguild predation
occurred. This prediction was tested in a
subsequent study, (Risbey et al. 1999) where fox
and cat numbers were counted on spotlight
transects and small mammals were surveyed by
live capture methods in areas where fox and cat
populations were controlled (see section 4 study 3
for further detail on the experimental design of this
study). At the site where only foxes were
controlled, spotlight counts of feral cats increased
three-fold over three years, but small mammals
declined in numbers (indexed by captures/100 trap
nights).
Despite the limitations in the experimental design
of this project (Risbey et al. 1999), they believed
that the above results were sufficient to infer that
fox control may lead to increased abundance of
feral cats. Molsher et al. (1999) investigated the
potential for competition between feral cats and
foxes by concurrently comparing diet, home range
and habitat use, and using video observation and
simultaneous radio-tracking to study avoidance
and aggression between both feral cats and foxes
at Lake Burrendong, New South Wales, between

Interactions between feral cats, foxes, rabbits and native carnivores 19

 (a)

(b)

(c)

Figure 3. Distribution of a) fox, b) feral cat and c) European rabbits in Australia.

(Source: Environmental Resource Information Network, Department of the Environment and Heritage).

Interactions between feral cats, foxes, rabbits and native carnivores 20

 # 75000 - 300000
# 30000 - 75000
(a) # 10000 - 30000
#
1000 - 10000
#
1 - 1000

## #

# # # #
###
# #
#
#

##
## ##
# ## # # #
#

# # # #
#

# #
# # #
# # ##
#
###

(b)
# 300000 - 800000
# 75000 - 300000
# 30000 - 75000
# 10000 - 30000
#
1000 - 1000
#
0 0
- 1 000

#
#
#
#
#

# ## #
###
#
#

# # # ##
#
# #
#
# # ## ##
#
# ##
#
# # # # # # #

# #

# ## ## # #

###
#
#

# ## #
####### ## ## ###
#
#
#
# #
# #
## #
#
#
# #
#
### # ## #
## # #
#
#
# ## #
##
# #
### # ##
#
#
## # # # ## # #
##
## ### # # #
## # #
#

####
########## #
#
###
#
#
#
###
# #
# ###
#
##
# ##
# #
## # # # # # #
# # #######
############
# # # # # # #######
## # # #
#
#
#
## # ##
#
## #

# #
## ## # # #
# ###### # ##
# #

#
#
# #
#
#
# # ## # #
# # #### ## # ##
### # # # # ### # #
# #
#
#

# # ### ##### ## ## ##### #

# # # # ##
#
##
# # ## # #
#### ## ###### ##
# ### ## ######### ###
#
#

###
## # # # ## # # ##
#
# ##

##
## #
#

###

Figure 4. Location and extent of a) feral cat (n =96) and b) fox control (n = 777) operations in
Australia.
(Reddiex et al. 2004). Area of control operations (ha) not shown to scale.

Interactions between feral cats, foxes, rabbits and native carnivores 21

Table 2. Examples of comparative feral cat and fox diet studies in areas where rabbits are present.

Study Location Stomach Technique Number of Rabbit in fox Number of Rabbit in cat
or scat used to fox samples diet (% cat samples diet (%
assess diet occurrence) occurrence)

Bayly (1978) Mt Lyndhurst, Stomach Percent 29 44.5 21 68.2

SA occurrence
Catling (1988) Yathong Nature Stomach Percent 288 45.1 112 54.0
Reserve, NSW occurrence
Molsher Lake Scats Percent 343 52.2 499 81.6
(1999) Burrendong, occurrence
NSW & Volume
A
Risbey et al. Heirisson Prong, Stomach Percent 47 76.6 171 49.1
(1999) WA occurrence
Read and Roxby Downs, Stomach Percent 105 ~0-100 516 ~22-70
Bowen (2001) SA occurrence
Holden & Flinders Stomach Percent Pre-RHD 105 Pre-RHD 63.1 Pre-RHD 73 Pre-RHD 37.3-
Mutze (2002) Ranges, SA occurrence 46.2
Post-RHD 774 Post-RHD 6.9- Post-RHD
15.7 294 Post-RHD 9.8-
35.6
A
Feral cats (n=109) and Semi-feral cats (n=62) combine

A large overlap in resource use, home range and measure the actual number of individuals in a
diet between feral cats and foxes suggested a high population or the number within a given area, as
potential for competition. In both areas where these measurements can be labour intensive and
foxes were controlled there were significant expensive, and in the majority of ecological
behavioural changes, including increased use of investigations unnecessary (Krebs 1999). Instead,
carrion and increased use of grassland habitat. indices of density that are correlated with absolute
Molsher et al. (1999) suggested that these density are useful (Caughley 1977). Unfortunately,
behavioural changes indicated interspecific the current techniques available (bait take,
competition; however, as acknowledged by the spotlight counts, sand plot activity and scat counts)
author, there was no increase in cat abundance are generally imprecise, and/or have restrictions on
over the 2.5 years following the control of foxes, their application. The relationship between
therefore mesopredator release cannot be changes in the index and actual abundance
demonstrated statistically. Interference remains untested. There is a need for further
competition was also recorded, with three radio- development of more reliable techniques to
collared feral cats believed to be killed by foxes, accurately assess changes in the abundance of
and foxes were observed acting aggressively predator species in Australia.
towards feral cats. However, no cat remains were
found in any of 255 fox stomachs or 343 fox scats, Summary
suggesting that if intraguild predation did occur it
was relatively rare. Feral cats and foxes overlap in distribution and
diet, and there is circumstantial evidence of
A major limitation of many of the above mentioned interspecific competition, where foxes may
studies is that reported increases in cat abundance competitively exclude feral cats from food
following fox control may in fact be an artefact of resources, and of intraguild predation where foxes
the census methods rather than an actual increase may prey upon feral cats.
in cat abundance. Indices of cat abundance using
track counts may increase following a reduction in Foxes, but not feral cats have been controlled over
foxes; however, this may be related to changes in large areas, and there is a possibility that impacts
cat activity patterns not changes in abundance on shared prey can increase following fox control if
(Molsher 1999). While spotlighting is often feral cat numbers increase after fox control.
undertaken over an inappropriate transect length Several studies have described increases in cat
for predators and/or is assessed at an appropriate abundance following reductions in fox numbers
scale for rabbits, but not predators. resulting from control operations. However, the
Monitoring changes in abundance of introduced evidence for a numerical response in cat
predators can be expensive and problematic as abundance following fox control is inconsistent
these species are often cryptic, elusive and occur between studies and may be confounded by
in low densities. It is often not necessary to inadequate survey techniques and behavioural
changes that may influence cat activity.

Interactions between feral cats, foxes, rabbits and native carnivores 22

There remains a great deal of uncertainty on rates with poorly developed anti-predator
mesopredator release of feral cats following fox behaviour.
control, but it would not be surprising if adequately
In many parts of Australia, particularly semi-arid
tested, given the above circumstantial evidence.
and arid areas where drought is common, rabbit
populations fluctuate markedly (Williams et al.
4.3 Change in Abundance of Primary 1995), and a lagged increase in predator numbers
Prey (Rabbits) can result (Saunders et al. 1995; Read and Bowen
2001). However, when rabbits decline, either as a
The European rabbit, which occupies 90% of
result of drought or disease, or through rabbit
Australia (Figure 3), forms the major component of
control operations, it is possible that predators may
the diet of feral cats and foxes in many areas,
switch their consumption to the next most
particularly the pastoral zones of southern
abundant alternative prey (Pech and Hood 1998;
Australia (Coman and Brunner 1972; Myers and
Table 3).
Parker 1975a,b; Brooker 1977; Jones and Coman
1981; Jarman 1986; Catling 1988; Dickman 1996; Prey switching occurs when the proportional
Paltridge et al. 1997; Molsher et al. 1999; Risbey et contribution of a species to a predators diet does
al. 1999). not match its relative abundance (Murdoch 1969;
Murdoch and Oaten 1975). While prey switching
Rabbits may influence the persistence of native
has not been demonstrated for feral cats or foxes
species by increasing the population size of a
in Australia, some work has been undertaken on
shared predator. In their review of patterns of
their numerical and dietary response to changes in
decline and extinction of Australian rodents, Smith
rabbit abundance and in some instances the
and Quin (1996) concluded that high levels of
changes in abundance of alternative prey (see
rabbits and house mice may have supported foxes
section 4.3).
and feral cats leading to declines and extinctions of
native prey species; they term this process, In a review of the potential impacts on Australian
‘hyperpredation’. Jarman (1986) noted that where native fauna of RHD, Newsome et al. (1997) noted
rabbits supported numerous foxes, more that there was little rigorous evidence of impacts of
vulnerable species such as rat-kangaroos or predation on wildlife populations as primary prey
bandicoots may be subjected to unsupportable (rabbits) collapse. This was regardless of the
levels of predation. In north-eastern New South mechanism that caused the crash (i.e. drought,
Wales, Rufous rat-kangaroos (Aepyprymnus myxomatosis or conventional control). Their report
rufescens) persist only where foxes and rabbits was commissioned as RHD escaped Wardang
were scarce (Schlager 1981), and Christensen Island of the South Australian coast in 1995 and
(1980) argued that densities of brushtailed was spreading across the Flinders Ranges in
bettongs (or woylies Bettongia penicillata), foxes South Australia. Since then, there have been a
and rabbits are similarly related in south-western number of studies on the impacts of reduced rabbit
Western Australia. It was thought that predation by numbers (mainly) resulting from RHD on the
feral cats (supported by high numbers of abundance of feral cats and foxes. Far fewer
introduced rabbits) was the mechanism that studies have simultaneously investigated changes
caused the extinction of the Macquarie parakeet) in predator diet and the flow-on effects to
on Macquarie Island (Taylor 1979). Parakeets and alternative prey species, and many of these have
feral cats coexisted for 60 years before rabbits been conducted over short (<2 years) timeframes.
were introduced onto the island, with feral cats
being presumably controlled by density-dependant This section reviews the interactions between a
reduction in primary prey (rabbit) on feral cats,
mechanisms. Within 20 years of the introduction of
rabbits, parakeets and banded rails had become foxes and native carnivores and the effects of
extinct. It was suggested that rabbits provided a these predators on alternative prey. We review the
impact on each predator individually and the
year round food supply, supporting feral cats at
high densities, thus increasing predation on potential impact on alternative prey.
alternative prey.
4.3.1 Effects of changes in abundance of
Smith and Quin (1996) suggested that declines
primary prey on feral cat abundance
and extinctions in native species are more likely: to
occur in areas where rabbits, rats and mice are and impacts on native prey
abundant and the alternative prey species are There are a limited number of examples in the
terrestrial; fall within the prey-size class of dingoes published literature that illustrates the relationship
(which can predate almost any prey-size), foxes between a change in rabbit abundance and a
and feral cats (that are restricted to small and change in the abundance of feral cats. By far the
medium-sized prey); and have low reproductive majority of evidence comes from studies in semi-
arid and arid Australia.

Interactions between feral cats, foxes, rabbits and native carnivores 23

Study 1: Yathong Nature Reserve populations had declined by 85% due to RHD
(Holden and Mutze 2002). No information was
Newsome et al. (1989) and later Pech et al. (1992)
provided on changes in cat numbers on the broad-
reported on a predator-removal experiment at
scale sites. In the year following the arrival of RHD,
Yathong Nature Reserve, New South Wales,
the previously distinctive seasonal dispersal peaks
conducted between June 1981 and January 1984.
in cat abundance were absent. The lack of rabbits
This area can be classified as semi-arid, with a
to support recruitment into the population was cited
mean annual rainfall of 200–350 mm. The
as a major reason for the decline in feral cat
experimental design for the Newsome et al. (1989)
numbers across all areas of this study. However,
study was, 1) removal of predators over one area
the authors noted that cat numbers increased
of 70 km2 (Block A), 2) no predator control was
through 1996-97 in the National Park concurrent
undertaken on two areas of 180 and 50 km2
with an increase in house mice in the diet. No
(Blocks B & C, respectively). After one year of the
measure of changes in abundance were made for
study Block B was sub-divided and predator
house mice, thus this result needs to be interpreted
control commenced over a 90-km2 area (Block B1)
with caution.
of the Block to examine repeatability and site
specificity of the results. No predator control was Study 16 Roxby Downs
undertaken on the remaining area (Block B2).
Read and Bowen (2001) reported on the
Pech et al. (1992) advanced the above study by
population dynamics of feral cats and foxes in
analysing an additional experiment beginning in
relation to changes in rabbit abundance over a ten-
mid-1983. Predator-removal continued in Block A,
year period between 1989 and 1999. Outbreaks of
but only limited predator-removal occurred in Block
myxomatosis and RHD occurred in 1993 and 1996,
B1. Control of predators ceased in all sites in
respectively. Changes in feral cat, fox and rabbit
August 1983 and they were allowed to reinvade
abundance were monitored using spotlight counts
the experimental blocks. In both studies densities
along two 20 km transects.
of rabbit and predator populations were assessed
on all sites by spotlight counts. The effects of Feral cats reached a peak density of 3 km-2 prior to
predator-removal on native prey were not the release of RHD. Rabbit populations declined
assessed in either of these experiments. On sites following the release of RHD (Read and Bowen
where no predator control was undertaken 2001), and cat populations were reported to have
spotlight counts of feral cats were highest subsequently declined. However, no figures were
(1.5 km-1) in 1979, corresponding with the period of provided on the relative abundance of feral cats
highest rabbit numbers (310 km-1). Pech et al. post-RHD.
(1992) report that feral cat abundance was linked
to the maximum density of rabbits in the previous Study 17: Lake Burrendong
three months. At Lake Burrendong, in central New South Wales,
Molsher et al. (1999) reported no clear change in
Study 15: Flinders Rangers National Park
the abundance of feral cats to changes in rabbit
In the Flinders Ranges National Park, South abundance. Rabbits at Lake Burrendong declined
Australia, Holden and Mutze (2002) reported on a following the arrival of RHD from ~18 km-1 to
study investigating the impact of RHD on <5 km-1 six months later. Cat numbers were low six
introduced predators from 1994 to 2000. The months after high rabbit numbers. However,
study involved a combination of fox baiting across changes in feral cat numbers were also weakly
the entire park and rabbit control operations correlated with the abundance of carrion, small
(warren ripping) covering 10% of the park. An mammals (4-month lag), reptiles (1-month and 2-
experimental site was established in 1992, month lag), and grasshoppers (3-month lag).
containing eight treatment blocks (each 3–4 km2). These authors suggest that behavioural responses
Rabbits were controlled on half the plots. RHD altering the sightability of feral cats may account
reached the area in 1995. Spotlight transects, each for recorded changes in population size rather than
running 2 km long, were established in each actual numerical responses associated with
treatment block, concentrating on strips 30–60m changes in prey abundance.
wide. In addition to this study they conducted a
broad-scale study of introduced predators. This Study 18: Northern Territory
covered three areas, a fox baited area, an Edwards et al. (2002a) report on population trends
unbaited area adjacent to the park, and a remote in rabbits and other wildlife following the arrival of
unbaited area. No details of the size of these areas RHD in the Northern Territory. They used spotlight
were provided. counts conducted at 3-monthly intervals along
10 km transects over two successive nights. They
On the experimental plots, feral cat numbers
declined from ~15 feral cats sighted per 100 km to recorded an 85% reduction in rabbit numbers
an average of 3.2, six to ten months after rabbit across six locations in the Northern Territory, but
no detectable decline in feral cat numbers. The

Interactions between feral cats, foxes, rabbits and native carnivores 24

authors noted that a weakness in their study was However, several sites had pre-RHD data on rabbit
the lack of a control site where RHD was not numbers, in some cases predator numbers, and
present, and that they could not separate out only at Lake Burrendong, in parts of Hattah-
environmental effects. Kulkyne, Flinders Ranges and Nullarbor was there
any information on native fauna pre-RHD in
In a related study that covered 2.5 years, Edwards
relation to feral cat and rabbit abundance.
et al. (2002b) established paired ripped and
unripped warren plots on four of the above sites.
Rabbits were monitored via spotlight counts along 4.3.2 Effects of changes in abundance of
a fixed transect of 10 km, and populations of feral primary prey on feral cat diet and
cats, foxes, and dingoes were monitored using impacts on native prey
track counts along the same transect. Both before
and after the arrival of RHD, there was significantly Feral cats have a catholic diet, but prefer live prey.
less sign of feral cats on sites where rabbit warrens Younger rabbits appear to be a staple of their diet
were ripped compared to unripped sites. when abundant, but a number of diet studies
indicate that they are capable of eating a variety of
Study 19: RHD Monitoring Program items including small mammals, birds, reptiles,
The RHD Monitoring and Surveillance Program invertebrates and carrion (Dickman 1996).
was established to monitor the impacts of the Study 1: Yathong Nature Reserve
spread of RHD on biodiversity (Neave 1999). This
program covered 10 intensive (Table 3) and 54 Catling (1988) found that feral cats displayed a
broad-scale monitoring sites, collectively covering Type III functional response (see section 4) to
all the principal biomes occupied by rabbits in rabbits at Yathong Nature Reserve, New South
Australia. At each of the intensive monitoring sites, Wales but only during the rabbit-breeding season.
attempts were made to assess changes in the Cats ate fewer rabbits when the rabbit population
abundance of rabbits, small to medium-sized was low and when fewer young rabbits were in the
mammal fauna, other fauna (birds, macropods and population, but more when the rabbit population
wombats), predators and vegetation (pasture and was increasing. When rabbits were scarce (or
perennials). The RHD science sub-committee set declined after the breeding season), feral cats
standard methods for the collection of data at each changed their diet to include (in order of
of the ten sites (Neave 1999). However, on many importance) invertebrates, birds, reptiles and small
of the sites native species response monitoring mammals. This resulted in an annual prey cycle.
was stopped after a year or two due to a lack of Catling (1988) made no assessment of alternate
response or a shortfall in funding. prey species abundance so it is not possible to
infer what the impact of this change in diet would
Sandell and Start (1999) summarised the results have on populations of native species.
from the Australia-wide RHD monitoring program
and the implications for biodiversity. They reported Study 15: Flinders Rangers National Park
that a decline in feral cat numbers following the In the Flinders Ranges, (South Australia), Holden
arrival of RHD was recorded on 6 of 10 intensive and Mutze (2002) reported that despite a
monitoring sites. These were the Nullarbor Plain significant decline in rabbit abundance following
(anecdotal only), the Northern Territory (one of RHD, the remaining cat population continued to
which was included in Edwards et al. 2002b prefer rabbits in their diet (pre-RHD rabbits
above), the Flinders Ranges (one being reported occurred in 42% of cat stomachs n=73, versus
by Holden and Mutze 2002 above), on the central 24%, n=293 post-RHD). Reptiles and birds were
tablelands of New South Wales, and at Hattah- consumed at a similar rate pre- and post-RHD, but
Kulkyne National Park. At this latter site, sightings invertebrates became a more frequent item in feral
of feral cats during the course of rabbit spotlight cat diet post-RHD. No assessment was made of
transects were the only measure available. An changes in the abundance of alternative prey, so
average of 0.025 feral cats per km was recorded the potential impact cannot be discussed.
from ten spotlight counts prior to the arrival of RHD
(137 km of transect). An average of 0.01 feral cats Study 20: Roxby Downs
per km was recorded from six spotlight counts At Roxby Downs, South Australia, rabbits were the
post-RHD (Sandell and Start 1999). most important component of feral cat diet when
At two of the RHD monitoring sites there was no rabbit counts exceeded 10 km-2; below this figure
pre-RHD data available and reports of decline other vertebrates increased in importance (Read
were only anecdotal. It was not possible to and Bowen 2001). The authors suggested that the
establish control sites (i.e. RHD absent) at any of decline in cat abundance, following the decline in
the monitoring locations. rabbits, was buffered by the ability of feral cats to
shift their hunting to include a wide range of native
vertebrates. Generally, feral cats were reported to

Interactions between feral cats, foxes, rabbits and native carnivores 25

consume prey in proportion to its availability, thus Feral cats showed a marked increase in the
when rabbit populations crashed feral cats consumption of birds, prior to an increase being
consumed small sand-dwelling lizards, with house detected in the field. However, the author
mice and small passerines also contributing to their suggested that bird abundance had actually
diet (Read and Bowen 2001). The impact on increased but the survey technique failed to record
alternative prey cannot be discussed, as changes the increase.
in abundance of alternative prey were not
assessed during this study. 4.3.3 Effects of changes in abundance of
Study 17: Lake Burrendong primary prey on fox abundance and
At Lake Burrendong, New South Wales, rabbits impacts on native species
remained the dominant prey type of feral cats The fate of foxes and rabbits has been linked since
despite a 90% reduction in rabbit numbers their introduction to Australia some 130 years ago.
(Molsher et al. 1999). There was no evidence of It has been suggested that the spread of the fox
feral cat numbers proportionally increasing across Australia was in part facilitated by the
consumption of reptiles, invertebrates or small presence of rabbits, which had been introduced
native mammals after rabbit abundance had earlier (Saunders et al. 1995). However, the
declined. However, house mice were found to form interactions between rabbit abundance and foxes
a significant component (100% occurrence in scats and the impacts on native fauna have only been
in autumn and 43% in winter) of the cat diet 10 investigated more recently.
months post-RHD. Previously, house mice had not
occurred in more than 19% of cat diet. Study 19: RHD Monitoring Program

Study 21: Hattah-Kulkyne National Park Declines in spotlight counts of foxes were reported
following the arrival of RHD at four (Nullarbor
Cavanagh (1998) and Sandell (1999) assessed Plains, Muncoonie, Hattah-Kulkyne and
changes in feral cat diet pre- and post-RHD at Tablelands) of the nine national RHD monitoring
Hattah-Kulkyne National Park. Four cat stomachs sites (Sandell and Start 1999; an additional site
were collected pre-RHD and eleven were collected was located in Tasmania where foxes were
post-RHD. Feral cats consumed mammals, absent)(Table 3). At two sites (Nullarbor and
invertebrates, reptiles, and birds, with rabbits being Flinders Ranges subsite) the reports of decline
the staple prey item. The authors acknowledged were only anecdotal. Two sites reported no long-
that sample sizes were small but suggested that term change in fox abundance (Lake Burrendong
post-RHD feral cats shifted their diet from rabbits and Northern Territory aggregated sites) and one
to birds (50% by occurrence and 16% by volume site was not assessed due to low fox densities
pre-RHD to 67% and 88% post-RHD). (Coorong). As mentioned earlier, these findings
Study 22: Tanami Desert need to be interpreted with caution.

Paltridge (2002) investigated the diet of feral cats, Study 16: Roxby Downs
foxes and dingoes in relation to prey availability at At Roxby Downs, high fox numbers coincided with
two separate sites in the Tanami Desert, Northern peaks in rabbit abundance. Fox densities peaked
Territory, between 1995 and 1997. Rabbits were at >3 km-2 one year after rabbit densities peaked at
absent from this study area. Monitoring focused on ~375 km-2, but declined to <0.5 km-1 several
changes in abundance of invertebrates, reptiles, months after rabbit populations crashed following
and small mammals via pitfall and Elliott trapping 3 the arrival of RHD. Foxes were rarely seen for the
times per year. Bird species were monitored along two years of this study (Read and Bowen 2001).
1 km walked transects using distance sampling
methods. Macropods, goannas and bilbies were Study 15: Flinders Rangers National Park
monitored by track counts along 10 km track In the Flinders Ranges National Park, Holden and
transects. The diet of predators was assessed Mutze (2002) reported that fox numbers were
(frequency of occurrence) through analysis of scats reduced by 96% (54 per 100 spotlight km to 7.8
collected along the track transects and from active spotlight km) following a fox-baiting program. After
searches. the arrival of RHD fox abundance declined to 1.6
In the absence of rabbits, feral cats relied on per 100 spotlight km with a lag of about 6 months.
reptiles as a summer staple with an increased The authors suggested that the reduction in rabbit
reliance on birds during winter when reptiles where numbers was partially responsible for the
less active. In most cases the relative abundance additional decline. The authors noted that a critical
of prey items in the diet of feral cats followed that impact on fox numbers was the lack of rabbits
of their relative availability. The consumption of during the rabbit-breeding season; this resulted in
small mammals (both sites) and skinks (one site) no peak in rabbit numbers, which normally
was strongly correlated with their field abundance.

Interactions between feral cats, foxes, rabbits and native carnivores 26

supports recruitment of juvenile foxes (Holden and following the arrival of RHD. However, for both of
Mutze 2002). these observations no statistical tests or details of
variance in the data were presented. Several
Study 18: Northern Territory:
alternative hypotheses for changes in scat
In contrast to the above studies, Edwards et al. accumulation and den activity are possible. It is
(2002a) reported that following an 85% reduction possible that repeat visits to the same dens may
in rabbit numbers across six locations in the have resulted in the decline in occupancy by foxes
Northern Territory there was no detectable decline through time, and that changes in food resources,
in fox numbers (see section 4.3.1 for comments on environmental conditions or fox population
the limitations of this study). structure may have influenced defecation rates.
Sandell (1999) reports on the changes in rabbit,
fox and native species following the arrival of RHD 4.3.4 Effects of changes in abundance of
at Hattah-Kulkyne, Victoria. This site comprised six primary prey on fox diet and
sub-sites, at which rabbits were monitored via impacts on native prey
spotlight counts and active warren entrance
counts. Warren entrance counts did not commence At Yathong Nature Reserve, New South Wales,
until after the arrival of RHD, as did spotlight foxes appeared to display a Type III functional
counts at one of the six sites. Spotlight counts response (Pech et al. 1992) to increasing rabbit
were conducted over a total of 108 km per annum numbers during winter and spring, eating more
from 1991 to 1999, with sampling repeated at least rabbits during the rabbit breeding season than
once. during the non-breeding season. Once the rabbit
breeding season had finished, foxes relied more
Sandell (1999) reports that at two of the sites in the on supplementary prey items (in decreasing order
Hattah-Kulkyne National Park, pre-RHD spotlight of importance): invertebrates, reptiles, carrion and
counts for rabbits ranged between 2 and 8 km-1, birds (Catling 1988). However, no assessment of
but have remained below 0.5 km-1 since the arrival abundance was provided and it is therefore not
of RHD. At the site in the Murray Sunset National possible to properly quantify the functional
Park, pre-RHD counts ranged from 0.2 to 6 km-1, response of foxes to alternative prey.
while post-RHD counts did not exceed 0.3 km-1.
On the dryland agricultural site rabbit counts When rabbits were abundant (>10 km-2) at Roxby
declined from an average of 4 per km pre-RHD to Downs, South Australia, they formed the major
1.2 per km over a two year period following the dietary component for foxes, occurring in more
arrival of RHD. than 70% of fox stomachs. Post-RHD, when rabbit
abundance declined, foxes shifted from their pre-
Fox abundance was assessed during quarterly RHD reliance on rabbits to mainly invertebrates
spotlight counts for rabbits between 1990 and and slow moving fossorial reptiles (Read and
1999. Sandell (1999) reports that there was no Bowen 2001). Despite the presence of a range of
decline in fox abundance coincident with the small native mammals (three native mice, two
decline in rabbits following the arrival of RHD. The hopping-mice and two dunnart species) house
data are aggregated from a series of transects in mice was the only small mammal species
the Murray Sunset National Park totalling 137 km. consumed by foxes (Read and Bowen 2001).
Cavanagh (1998) reports that rabbits did not
appear to be the staple prey item of foxes pre- In the Flinders Ranges National Park, South
RHD, rather foxes relied on carrion and reptiles, Australia, rabbit was the most common prey item
which buffered the impact of rabbit population taken by foxes pre-RHD, occurring in 65% of
decline on the fox population. stomachs (n = 105). Post-RHD, where rabbit
populations were reduced by 85%, the occurrence
Scats collected on an annual basis between 1995 of rabbit in fox stomachs was only 16% (n = 774;
and 1999 from Mallee Fowl (Leipoa ocellata) nests Holden and Mutze 2002). The authors reported an
were also used to assess a change in relative apparent shift in diet towards more invertebrates,
abundance of foxes (n=568). Sandell (1999) reptiles and kangaroo (carrion from harvesting
concluded that there was a significant decline in operations). The authors acknowledged that the
the proportion of nests at which fox scats were limited availability of pre-RHD dietary data made it
collected and attributed this to a decline in fox difficult to determine clearly the change in fox diet
abundance. Sandell (1999) also assessed active (Holden and Mutze 2002).
fox dens quarterly on five Mallee Fowl monitoring
grids between 1996 and 1999 by repeat visits. The Sandell (1999) assessed changes in fox diet at the
average proportion of dens occupied decreased Hattah-Kulkyne RHD monitoring sites in Victoria,
through time. The author suggested that the by analysing stomach and scat contents. Eleven
combined information from scat counts and active stomachs were collected pre-RHD and 57 post-
dens indicate a steady decline in fox abundance RHD. Foxes consumed mammals, reptiles, birds,

Interactions between feral cats, foxes, rabbits and native carnivores 27

fish/crustaceans, invertebrates and vegetation. experimental non-treatment sites, means that
Sandell (1999) found that carrion was the most these results need to be interpreted cautiously.
important component of fox diet pre-RHD (55% by
In the Northern Territory, Edwards et al. (2002b)
occurrence and 36% by volume) and that this did
studied the effect of warren ripping on rabbits and
not change post-RHD (52% and 41% respectively).
other wildlife. They found that there was less sign
Cavanagh (1997) further analysed these data and
of foxes and feral cats on ripped plots than on
concluded that the risk of foxes shifting from
unripped plots, but could not detect a change in
rabbits to other prey post-RHD was minimal.
the abundance of red kangaroos, small mammals
Saunders et al. (2004) looked at changes in fox or raptors following a decline in rabbit numbers.
diet pre and post RHD. They collected fox
stomachs from undulating to hilly lowland country, Summary
around Orange, NSW. The area has an annual
rainfall of between 500 and 800 mm. RHD arrived Rabbits are common throughout 90% of Australia
at the site in 1996 and was widely established by and have been associated with the spread of the
the end of that year. The authors state that RHD fox since its arrival 130 years ago. The association
had an effect in the rabbit population but do not between the abundance of rabbits and feral cats is
provide data on the size of the effect. Foxes were less well understood.
shot at night throughout the year between 1995 Increased predator density may result from a
and 1998. Foxes were divided into pre (n=240) and reliance on abundant staple prey
post (n=269) RHD samples. (e.g. rabbits). This may result in a numerical
Dietary data was analysed as both percentage increase in predator species, which may have
occurrence (%O) and percentage by volume implications for predation rates on some native
(%BV). These authors found no dramatic RHD- species if the predator species that increases
induced differences in fox diet. Rabbit comprised specialises in certain prey types. Perturbation
20.8 %O and 16.2 %BV pre-RHD and 19.3 %O experiments looking at changes in staple prey
and 16.1 %BV post-RHD. There was an increase abundance and dietary responses of predators, in
in the %BV consumption of rodent. This was conjunction with population studies of predators
related to an eruption of house mice numbers. and prey would provide a test for this hypothesis.
There was no detectable shift to increase The use of predator manipulation studies or the
predation rates on other prey (sheep, macropod, monitoring of RHD outbreaks has provided insights
possum, bird, reptile, invertebrate or plant). into the interaction between changes in the
Saunders et al. (2004) suggested that the abundance of rabbits and the flow-on effects to
combination of drought (which preceded the arrival predators and alternative prey. The level of our
of RHD) and RHD had acted in concert to understanding of the interactions varies between
suppress rabbit abundance below a critical biogeographical regions in Australia.
threshold resulting in a lack of shift in dietary In semi-arid and arid areas of Australia, where
selection. rabbits are the primary prey of feral cats and foxes,
Paltridge (2002) monitored changes in the diet of the abundance of both predator species appears to
foxes in two areas of the Tanami Desert, where be strongly correlated with rabbit abundance. The
rabbits do not occur. The author found that in abundance of both predators is associated with
contrast to dietary studies elsewhere in Australia, peaks and troughs in rabbit abundance, and both
reptiles were an important component of the diets predator species show a lagged decline in
of foxes and should be classified as seasonal abundance of 6 to 12 months after rabbits are
staples. When reptiles were less active during substantially reduced. In temperate environments
winter, birds increased in importance in the diet of this relationship is less well understood, and in the
foxes. few studies in these habitats foxes and feral cats
have not shown the same marked response to
Of the dietary studies reviewed, only Edwards et changes in rabbit abundance. In areas where
al. (2002a) described changes in the abundance of rabbits are not the primary prey, or where
native carnivores and alternative prey with a environmental conditions and/or disease have
decline in primary prey (rabbit abundance). They suppressed rabbit populations below a critical
report that there were more dingoes and varanids threshold, the decline in rabbits has had no
post-RHD and less wedge-tailed eagles. The measurable effect on the abundance of these
authors noted that the data for wedge-tailed eagles predators.
was highly variable and should be interpreted with
caution. Similarly, they report no change in the In a few examples the use of integrated control
relative abundance of red kangaroos or small (ripping, RHD or poison baiting and RHD) has
mammals, but that these data are also highly enhanced the decline in predator species.
variable. This variability combined with a lack of

Interactions between feral cats, foxes, rabbits and native carnivores 28

A number of studies have assessed changes in the
diet of feral cats and foxes with changes in rabbit
abundance. In arid and semi-arid systems, where
rabbits were abundant, feral cats were able to kill
rabbits even at low densities or to shift to
alternative prey species, including lizards, house
mice and birds when rabbits were less abundant
providing some buffering against declining rabbit
populations. In temperate habitats, house mice
may play an important role in supporting feral cat
abundance, acting as staple prey. Foxes in these
areas appeared less capable of killing rabbits at
low densities and relied more on invertebrates and
reptiles.
In areas where rabbit populations had been
suppressed, either by environmental conditions,
disease, or in areas where rabbit populations are
‘naturally’ at low densities, it appears that changes
in rabbit abundance have little effect on the diet of
foxes. Little is known about feral cat diet in this
situation.
Feral cats and foxes also occur in areas where
rabbits are either absent or uncommon. In arid
areas where rabbits are absent, invertebrates and
reptiles comprise the bulk of the diet of foxes, with
rare or endangered small mammals comprising a
relatively small proportion of their diet (Paltridge
2002). Rabbits are uncommon in the higher
altitude areas of Australia where prey availability
varies seasonally (Osborne et al. 1978). In these
areas invertebrates are the major dietary item of
foxes in snow free months, with native small
mammals found in all months but reported as the
winter staple (Green and Osborne 1981).
From the studies reviewed it is unclear what the
impact of a decline in primary prey is on native
species. In the studies reviewed in this report, both
feral cats and foxes shift consumption to the next
most abundant prey item, (e.g. mice, invertebrates,
reptiles, or birds). There is no evidence that as a
result of a decrease in primary prey there is an
increase in predation rates on populations of rare
or endangered species. We are not discounting
that this is a real possibility. Rather, we were
unable to find or access studies that demonstrate
such an effect.
Our level of understanding of the interactions
between feral cats, foxes and rabbits in temperate
environments is less clear. The relationship
between changes in rabbit abundance and
declines in either feral cats or foxes has not been
clearly demonstrated and no information is
available that demonstrates that a change in rabbit
abundance leads to increased rates of predation
on native species.

Interactions between feral cats, foxes, rabbits and native carnivores 29

Table 3. Examples of studies that have assessed the impact of changes in rabbit population densities on predators and alternative prey.
T = Treatment and NT = Non-Treatment.
Study Location (# sites) Study Area Habitat Duration Treatment Technique T & NT Replication Key Changes in Changes in
(ha) (months) Used Sites population alternative prey
Yes / No

Pech et al. 1992 Yathong, NSW Semi-arid 30 Pre-control Spotlight counts Y Y Increase in # rabbits Not monitored

Holden and Mutze Flinders Ranges NP, SA 400 each Arid 36 RHD Spotlight counts P&P n = 10 Decline in feral cats Monitored
2002
Read and Bowen Roxby Downs, SA 20 km Arid 10 yrs RHD Spotlight counts N Decline in feral cats
2001 transects P&P
2
Molsher et al. 1999 Lake Burrendong, NSW 90 km Temperate 3 yrs RHD Spotlight counts N Decline in feral cats Monitored
P&P
30 km Small Mammal
transects Trapping
Active searches
Edwards et al. 2002a Multiple sites, NT 10 km Arid 2.5 yrs RHD Spotlight counts n=4 No Decline in feral
transects P&P cats
2
Edwards et al. 2002b Multiple sites, NT 20 – 140 km Arid 2.5 yrs RHD / Spotlight counts Y n=4 Decline in feral cats
Warren (Warren
ripping Track Counts ripping
Small Mammal only)
Trapping
Sandell and Start Nullarbor Plain, WA 25 km transect Arid 16 RHD Spotlight counts N N Decline in feral cats Monitored
1999 and foxes
21 km transect
Central Australia sites 400 each Arid 36 RHD Spotlight counts P&P n=4 Decline in foxes but Monitored
not in cats
Muncoonie Lake, QLD 1050 Arid 24 RHD Spotlight counts N N Decline in foxes, not Monitored
in feral cats
A
Balcanoona / Wertaloon, 400 each Arid 24 RHD Spotlight counts N n=4 Decline in foxes and Monitored
SA feral cats
Hattah, VIC ~700 each Semi-arid 24 - 84 RHD Spotlight counts P&P n=6 Decline in foxes, not Monitored
in feral cats

P&P = Pre and Post treatment monitoring, A = Anecdotal, NA = not assessed due to low numbers

Interactions between feral cats, foxes, rabbits and native carnivores 30

Table 3 cont/.

Study Location (# sites) Study Area Habitat Duration Treatment Technique T & NT Replication Key Changes in Changes in
(ha) (months) Used Sites population alternative prey
Yes / No

Coorong, SA 60–120 Temperate 24 RHD Spotlight counts Y N=2 NA Monitored

Lake Burrendong, NSW 800–1200 Temperate 24 RHD Spotlight counts N N No decline in foxes Monitored
Central Tablelands, 250 each Temperate 48 RHD Spotlight counts P&P n=3 Decline in foxes and Monitored
NSW cats
North Tasmania 1500 Temperate 24 RHD Spotlight counts P&P N No decline in feral Monitored
cats
Catling 1988
Yathong, NSW Semi-arid 30 Pre-control Spotlight counts Y Y Cat & Fox show Not monitored
Type III response to
Diet study Freq. Occur rabbits

Cavanagh 1998 & Hattah, VIC ~700 each Semi-arid 24–84 RHD Spotlight counts P&P n=6 Cat shift in diet Monitored
Sandell 1999 from rabbit to birds
Diet Study Freq. Occur/vol.

P&P = Pre and Post treatment monitoring, A = Anecdotal, NA = not assessed due to low numbers

Interactions between feral cats, foxes, rabbits and native carnivores 31

 evidence of exclusion or avoidance by foxes of wild
4.4 Interactions Between Native and dogs (Catling and Burt 1995). Thus, it is possible
Introduced Predators, and Rabbits that only when resources are limited that foxes and
wild dogs come into conflict. Foxes appear to avoid
4.4.1 Canids wild dogs in central Australia at sparsely separated
watering points (Fleming et al. 2001) and at
Dingoes, domestic dogs (Canis lupus familiaris) carcasses of kangaroos and cattle during drought
and their hybrids (collectively known as wild dogs) (Corbett 1995).
occur throughout most of mainland Australia, and
while dingoes can be considered native species, The interactions between feral cats and wild dogs
the functional role hybrids play in the ecosystem are poorly understood. The two species co-occur in
may be sufficiently similar for them to be many areas of Australia (Figure 3) and wild dogs
considered as acting like native dingoes. The are capable of consuming food items that are also
distribution wild dogs overlaps with both feral cats eaten by feral cats, and wild dogs have been
and foxes (Fleming et al. 2001) and rabbits (Figure recorded occasionally eating feral cats (Fleming et
3). al. 2001).

Over the past 30 years, the diet of wild dogs has The distribution of wild dogs and the spotted-tailed
been extensively studied. While over 170 species (Dasyurus maculatus), western (Dasyurus
have been identified (Corbett 1995), 80% of the geoffroii), and northern quoll (Dasyurus hallucatus)
diet of dingoes comprised only 10 species. These overlap, but the nature of any interactions between
were: red kangaroos (Macropus rufus), rabbit, wild dog and quoll species is not understood.
swamp wallaby (Wallabia bicolor), cattle, dusky rat
(Rattus colletti), magpie goose (Anseranas 4.4.2 Dasyurids
semipalmata), common brushtail possum long-
haired rat (Rattus villosismus), agile wallaby
(Macropus agilis) and common wombat (Vombatus 4.4.2.1 Quolls
ursinus) (Corbett 1995). No studies have investigated the interactions of
Mitchell (2003) studied the dietary and spatial any of the quoll species with foxes, feral cats or
overlap of wild dogs and foxes in the Greater Blue changes in primary prey abundance. This is
Mountains. He examined scats collected from 10 despite the fact that at least two of the four species
sites in autumn and winter 2002 (a minimum of 25 of quoll kill rabbits when available (Belcher 1995).
scats were collected for each species from each Quolls are smaller than both feral cats and foxes,
site). Mitchell (2003) also undertook a meta- with the spotted-tailed quoll, the largest quoll
analysis of 19 previous studies from eucalypt species ranging from 1.5 to 4 kg. The eastern quoll
woodland/forest areas that compared fox and wild (Dasyurus viverrinus) ranges from 0.8 to 1.3 kg,
dog diets. Mitchell concludes that the diets of foxes the northern quoll ranges from 0.5 to 0.8 kg and
and wild dogs showed a high degree of overlap, the western quoll (or chuditch) ranges form 0.8 to
and suggested that this was evidence for potential 1.3 kg. In comparison, adult red foxes weigh
competition. This author also found that at a fine between 4.5 and 8.3 kg (Coman 1983), and adult
scale there was some indication of temporal feral cats have been reported to weigh as much as
avoidance, but that at a landscape scale foxes and 6.2 kg (Jones 1983).
wild dogs co-existed.
Potential interactions between quoll species and
Given the potential for dietary overlap and the introduced predators could arise through
overlap in distribution of all three predators, there interspecific competition (e.g. exclusion via
is potential for wild dogs to suppress, either aggressive competition) and/or intraguild predation
through competition or direct predation, (e.g. direct predation). Given the overlap in diet
populations of feral cats and/or foxes (Jarman (see below) and size differences: competition or
1986; Robertshaw and Harden 1985; Thompson predation remains an untested possibility. It is also
1992; Corbett 1995; Fleming et al. 2001). possible that with a reduction in feral cats and/or
However, this has yet to be confirmed foxes, quoll species may increase in abundance
experimentally. (i.e. mesopredator release; see section 3.2).
On the Nullarbor Plain, Western Australia, foxes
and wild dogs were reported to be able to co-exist
because foxes were able to hunt rabbits inside wild Spotted-tailed Quolls
dog territories and possibly escape conflict by Belcher (1995) studied the diet of the spotted-
using rabbit warrens (Thomson and Marsack tailed quoll in East Gippsland, Victoria, and found it
unpubl. data in Fleming et al. 2001). In forested to be largely dependent on medium-sized
areas in south-eastern Australia, there was no mammals (0.5 to 5 kg). The most important prey

Interactions between feral cats, foxes, rabbits and native carnivores 32

species were the European rabbit, the common Oakwood (2000) studied reproduction and
brushtail possum and the common ringtail possum demography in northern quolls and reported that
(Pseudocheirus peregrinus). Other prey included the most common proximate cause of mortality
Antechinus species, bush rats (Rattus fuscipes), was predation, probably by dingoes Morris et al.
echidnas (Tachyglossus aculeatus), macropods, (2003) suggested that the western quoll (or
wombats (Vombatus ursinus), birds, invertebrates, chuditch) could be detrimentally impacted by foxes
and reptiles. A shift in diet between years was through direct predation of young quolls, and/or
attributed to the variation in rainfall and the effect competition for food as both species overlap in diet
this had on prey species abundance, suggesting (Coman 1973; Lunney et al. 1990; Soderquist and
that this predator, like foxes and feral cats (see Serena 1994).
above) selects prey items in relation to their
abundance. Significant differences in diet were 4.4.3 Raptors
found between adult and sub-adult quolls. Sub-
adult quolls consumed significantly more small Newsome et al. (1997) provides a comprehensive
mammals, ringtail possums, invertebrates and review of the potential impacts of a decline in
reptiles, and significantly fewer rabbits than adult rabbits on raptors in Australia. They identified that
quolls. Belcher (1995) found that medium-sized four of the 24 raptor species in Australia relied on
prey contributed more than 80% of the biomass of rabbits as a major dietary item and another five
prey consumed. utilise rabbits as alternative prey when abundant.
In order of significance the top four are the wedge-
Two unpublished studies provide additional tailed eagle (Aquila audax), little eagle (Hieraaetus
information on dietary overlap and interactions with morphnoides), brown falcon (Falco berigora) and
introduced predators. Alistair Glen at the Institute brown goshawk (Accipiter fasciatus).
of Wildlife Research, University of Sydney, New
South Wales, is currently investigating comparative Newsome et al. (1997) reported that, in general, in
diet and habitat use of foxes and spotted-tailed areas where wedge-tailed eagles rely on rabbits, a
quoll in coastal forest in New South Wales. decline in rabbits would result in a decline in either
Unpublished data from this study indicates that the number of adult birds, clutch size or young
quolls are killing rabbits near the margins of the produced. Evidence presented by Ridpath and
forests (A. Glen pers. comm.). The New South Brooker (1986) indicates that wedge-tailed eagles
Wales National Parks and Wildlife Service is would not breed if rabbit abundance fell below 60
investigating the impact of bush fires on quolls, in km-2. At the Western Mining Company Olympic
particular diet in the Byadbo region. Preliminary Dam site 4–5 wedge-tailed eagle nests were
results suggest that the abundance of rabbits and regularly observed to raise 2 chicks each prior to
possums decreased immediately post-fire, but the the arrival of RHD. Post-RHD no successful nests
proportions of rabbit in the quoll diet increased (J. were seen for 4 years (John Read pers. comm.).
Dawson, pers. comm.). Interestingly, the remains Little eagles are probably reptile specialists, but
of a cat were found in a single quoll scat in this take advantage of abundant rabbit populations.
study. Whether this was carrion or not is unknown. They can survive almost exclusively by feeding on
At Lake Burrendong in New South Wales, Molsher young rabbits in spring. It has been reported that
et al. (1999) compared the diet of feral cats and peaks in laying season coincide with the peak in
spotted-tailed quolls and suggested that there was rabbit breeding season (Baker-Gabb 1984;
enough overlap for potentially exploitative Mallinson et al. 1990; Olsen and Marples 1992 in
competition, although the author notes that the Newsome et al. 1997). The potential impact of a
sample size of quoll scats was small (n=12). The reduction in rabbit abundance on this species is
author found that rabbits were the main prey item poorly understood.
of both species with invertebrates second in Apart from utilising rabbits in both winter and
importance. spring little else is known about the interactions
between brown falcons and rabbits. Like brown
falcons, the brown goshawk consumes a
Other Quoll Species considerable amount of rabbit during winter and
Little information is available on the interactions spring, and where rabbits are not available birds,
between the remaining quoll species and feral cats reptiles and invertebrates form the major dietary
and foxes. The three smaller species are active components (Newsome et al. 1997).
hunters, preying on invertebrates, small mammals, Raptors were monitored at several of the RHD
birds, lizards, frogs and plant matter. Invertebrates, monitoring sites (see section 4.3), but no clear
particularly arthropods, form an important relationship between short-term changes in
component of their diet (Blackball 1980; Godsell abundance and breeding output and declines in
1982; Johnson and Roff 1982; Begg 1983; Serena rabbits was demonstrated (Sandell and Start
et al. 1991; Soderquist and Serena 1994).

Interactions between feral cats, foxes, rabbits and native carnivores 33

1999). This was partly due to the high degree of capable of suppressing fox populations, but that
variation in density of territories, breeding pairs and this is likely to be mediated by specific
the number of young produced per year, the size environmental conditions such as drought. There is
of monitoring sites, small samples sizes, seasonal perhaps stronger evidence to suggest that foxes
variations in climatic conditions, and no measures spatially and temporally avoid wild dogs and that
of alternative prey consumption. only during times of limited resources do the two
come into direct conflict.
Edwards et al. (2002a) reported that in the
Northern Territory, although populations of raptors Similarly, there is a lack of knowledge on the
(including wedge-tailed eagles, little eagle, brown impacts of feral cats and foxes on quoll species.
falcon and brown goshawk) fluctuated post-RHD, While there is some information to indicate that
no significant reduction in populations were there is potential for a negative effect through
detected. In a related study, Edwards et al. competition or direct predation, this is also likely to
(2002b) investigated the impacts of warren ripping be moderated by specific environmental conditions
on rabbits and other wildlife. They reported that (e.g. drought or fire altering prey composition or
rabbits were less abundant on ripped plots both abundance).
before and after the arrival of RHD, but that there
A reduction in the abundance of a shared primary
was no treatment effect on the abundance of
prey item (e.g. rabbits) may result in increased
raptors, including brown falcons and brown
competition, direct aggression, increased levels of
goshawks. A limitation on this study was that it only
predation on alternative prey (both from native and
ran for 2 to 3 years, and that ripped areas were
introduced carnivores), any of which has the
relatively small compared to the territories of large
potential to negatively impact on native carnivore
raptors. This may not have been long enough to
populations.
detect numerical changes in populations following
declines in rabbit abundance. The available evidence suggests that wedge-tailed
eagles are likely to experience a reduction in
4.4.4 Varanids numbers if a reduction in rabbit abundance is
significant and sustained. While other raptor
No studies have investigated the interactions of species (little eagle, brown falcon and brown
any of the varanids with foxes, feral cats or goshawk) utilise rabbit when abundant or at
changes in primary prey abundance. There are 26 particular times of the year, it is less clear what the
described species of goanna (the terms ‘monitor’ outcome would be for these species. Further
and ‘goanna’ are interchangeable) in Australia. All quantitative information is required for these
are carnivorous, consuming almost anything that species.
can be caught and eaten. Varanids are active
diurnal hunters which stalk, run down or dig out The current detail of information available for
their animal prey: smaller species take larger varanids is insufficient to be able to draw any
insects, spiders and small frogs, lizards and conclusions about the impact of changes in these
snakes; larger species hunt lizards, snakes, small predators or rabbit abundance. Studies on the role
birds and mammals but also feed on carrion. varanids play in the predator–prey dynamics of the
Species that are capable of or known to consume Australian ecosystem are required.
rabbits include the perentie (Varanus giganteus),
sand monitor (Varanus gouldii), lace monitor
(Varanus varius), and yellow-spotted monitor
(Varanus panoptes) (King and Green 1999).
Most species are active for only six months of the
year, but become most active during late spring
and summer, which coincide with the emergence
of young rabbits and increases in rabbit numbers.
Williams et al. (1995) made reference to an
increase in numbers of goannas with an increase
of rabbits. If rabbits comprise a significant
component of their diet then these varanids might
be affected by a decline in rabbit numbers.

Summary
Little quantitative information is available on the
interactions between introduced predators and
native carnivores. The information that is available
suggests that dingoes and wild dogs may be

Interactions between feral cats, foxes, rabbits and native carnivores 34

5 Interactive models of pest population dynamics

A major aim in ecology is to produce dynamic and it is likely the conclusions from them are
models that allow us to predict the effects of qualitative rather than quantitative.
changing parts of the system. To date this has
Here we set out to refine and extend the Pech and
proved largely unattainable, particularly in natural
Hood (1998) model. We aim to make the model
systems (Abrams 2001). Part of the problem lies
more predictive, but recognise this will be
in the complex nature of these systems, but also in
constrained by whether we can identify critical
a lack of focus on the components of many of the
interactions for these species, and by the data
relationships such as the functional response
available. The simplest models express rates of
(Abrams 2001).
increase of consumer species in terms of the
Interactive models abundance (or intake) of resources (‘prey’), so our
initial intention was to find relationships between
Interactive models attempt to model the
rates of increase for foxes and/or feral cats in
relationships advocated by Abrams (2001) and
relation to the abundance of rabbits.
have been used in Australian systems because of
Unfortunately, data were not available to do this.
the strong environmental variability characteristic
We focus initially on semi-arid systems, because
of these systems. They were pioneered in
the original model was developed for these
Australia by Caughley (1987) and colleagues
systems, but we also explore temperate systems.
(Bayliss 1987), who used them to model kangaroo
population dynamics. At the base of the model is One of the problems we have with developing
rainfall, (Figure 5) which drives pasture production models for these systems is that we don’t really
and pasture senescence. Herbivore offtake from know what the population dynamics of the different
pasture is determined by the functional response species are. For rabbits we have a reasonable
of the herbivore to pasture. The numerical idea, but for predators our understanding of their
response, or instantaneous rate of increase of the population dynamics is very uncertain. A ~20 year
herbivore, is expressed in terms of the biomass of data set of Brian Cooke’s (unpubl. data) indicates
grassland vegetation. For a more mechanistic that rabbit populations were generally low (<20 per
approach rate of increase should be expressed in spotlight km), but showed sharp increases,
terms of intake rate, rather than the density of the sometimes up to 400 per spotlight km, and sharp
resource. declines in density. This suggests rabbits respond
rapidly to good conditions and then crash just as
Pech and Hood (1998) developed a three trophic
rapidly when conditions deteriorate. Whether
level interactive model for a semi-arid system: with
rabbits are regulated by predators under certain
grassland vegetation at the bottom level, rabbits
conditions is not really known, but was suggested
and a model native Australian prey in the middle
by some of the results from Yathong (Newsome et
level, and foxes at the top level. Their model was
al. 1989; Pech et al. 1992). For predators, we
developed to explore whether reduced rabbit
don’t really know whether their populations
abundance due to RHD was likely to benefit or
fluctuate markedly or are reasonably stable despite
negatively affect native prey subject to fox
large fluctuations in rabbit density. We may expect
predation. Their model calculated rabbit
with predators to see recruitment peaks in late
population rate of increase as a function of pasture
summer/autumn, because of their seasonal
biomass, but adds a term to account for fox
breeding. This was observed in the Flinders
predation on rabbits.
Ranges, but spotlight counts there were over
They made numerous assumptions because of the extremely long distances (Holden and Mutze
lack of detailed information on many of the critical 2002). In many other areas these recruitment
parameters for a model of this type. The first major peaks are not apparent, possibly reflecting the
assumption was that there is a relationship short spotlight distances and the limitations of
between fox population rate of increase and rabbit spotlight counts for tracking changes in the true
density, which we explore below. density of predator populations.
Another key parameter was determined in their
model by trial and error, to produce population
A modelling framework
dynamics broadly consistent with those that occur
in the field. They set minimum densities on rabbit We begin with the premise that the important
and fox populations to stop them going extinct in interactions that drive the population dynamics of
the model. These assumptions make the the species in this system are those shown in
predictive power of these early models uncertain, Figure 5 and we discuss each of these interactions
in detail. This diagram is not exhaustive. For

Interactions between feral cats, foxes, rabbits and native carnivores 35

example, we have ignored native predators in the determined by the availability of prey (in this case
system. pasture) only. Other authors have used ‘ratio
dependent’ functional responses, which express
intake as a function of both prey and predator
density, and there has been some debate over
9 12 11
which form is the most appropriate starting point
Foxes Cats
RHD for model building (Abrams and Ginzburg 2000).
8 14 The original motivation for ratio-dependent
15 10 13 functional responses was that predator
Rabbits Native prey
interference or predator facilitation would affect the
4 7 intake of prey by predators and hence predator
6
abundance should be taken into account.
3 5 However, other mechanisms could lead to a ratio
dependent functional response being appropriate.
Vegetation
2
1 Food is unlikely to be distributed evenly in a
Climate landscape, but when we model animal populations
with non-spatial models we use spatially-averaged
Figure 5. Interactions in a simplified system. values. The functional response, for example,
Two-way arrows represent interactions that operate in would reflect the average intake rate of the
both directions e.g. vegetation affects the growth of
population. As food declines, we expect the
rabbit populations and rabbit populations affect the
growth of vegetation by consuming it. They do not imply average intake rate of consumers to decline, but
equal strength of the interaction in each direction. we also expect the consumer population to decline
in response to this. Those consumers that are left
Interactions 1 and 2 would likely be in areas where food is still relatively
The effect of climate on vegetation biomass and abundant and hence the average intake rate of
growth in Australia in semi-arid systems was those consumers would appear higher. Figure 5
quantified by Robertson (1987). Most published illustrates this.
interactive models for semi-arid systems in For example, for the prey-dependent functional
Australia (Caughley 1987; Choquenot 1998; Pech response illustrated in Figure 6, the average intake
and Hood 1998) have used Caughley’s rate of the population would be ~0.03 kg rabbit-1
modification of Robertson’s (1987) pasture growth day-1 at an average pasture biomass of 100 kg ha-
model (Caughley 1987; Appendix 1). The model 1
. If we only consider a prey dependent functional
accounts for the fact that pasture growth is response then as the rabbit population declines
determined by both rainfall and standing biomass (say from 50 ha-1 to 5 ha-1) intake would remain the
at the start of the growth period. The latter same. At 5 rabbits ha-1, however, it is likely these
component represents intra- and inter-specific rabbits are in areas where pasture biomass is
competition within the plant community (Interaction actually higher than 100 kg ha-1 even though the
2). The model ignores any changes in pasture overall average biomass is 100 kg ha-1. The
composition, and assumes an even spatial average intake per rabbit would hence be higher
distribution of pasture biomass. than would be expected. This type of relationship
has important consequences for the models,
because it provides a mechanism in the model
Interaction 3 where animals maintain sufficient intake at low
population densities so their populations do not go
Effects of rabbits on vegetation extinct . A ratio dependent functional response
The effect of rabbits on vegetation was measured could have the form provided in equation 3 in
by (Short 1987) using an intensive grazing trial in Appendix 1.
Kinchega National Park. The daily per capita
consumption of pasture by rabbits, adjusted for
body weight and expressed as kg animal-1 day-1 is
given by equation 2 in Appendix 1. This is an Ivlev
form of a type II functional response. No other
assessments of the rabbit functional response to
vegetation in semi-arid systems were available at
the time of writing.
This type of functional response is known as ‘prey
dependent’ because intake for a given body size is

Interactions between feral cats, foxes, rabbits and native carnivores 36

 0.08
A more mechanistic approach would be to model
rabbit rate of increase as a function of both pasture
0.07
growth (rabbits respond to growing pasture by
0.06 breeding), and standing biomass (standing
0.05 biomass may contribute to rabbit survival), but to
0.5
our knowledge this has not been attempted and
Intake

0.04 5
prey dependent
data are not available. (As an example of this
0.03
50 approach see equations 7 and 8 in Appendix 1).
0.02 The behaviour of this model is shown in Figures 7c
0.01 & 7d. Qualitatively this type of model appears to
0
better reflect the abrupt changes in rabbit
0 100 200 300 400 500 600 700 abundance evident in semi-arid systems (B. Cooke
Pasture biomass unpubl. data, Pech et al. 1992), but data are not
available to parameterise this model properly.
Figure 6. Prey dependent (dotted line) and
ratio dependent (solid lines) functional
responses.
The values 0.5, 5 and 50 indicate the abundance of the
consumer.
We include the functional response of rabbits in
our models to take account of their effect on their
food supply. However, other herbivores in the
system are often ignored (e.g. Pech and Hood
1998), which may cause serious errors if we want
to properly account for pasture biomass. Other
significant herbivores would be stock (in pastoral
areas), kangaroos (Caughley 1987), large feral
herbivores such as goats, and invertebrates.

Vegetation effect on rabbit rate of increase

In semi-arid systems rabbits begin breeding in
response to rainfall (Wheeler and King 1985;
Wood 1980). Pech and Hood (1998) found a
relationship between rainfall, lagged three months,
and the rabbit rate of increase at Yathong,
consistent with this observation. They then
modelled the rate of increase of rabbits as a
function of standing pasture biomass three months
prior, to capture this lag. The relationship used by
Pech and Hood (1998) for the quarterly rate of
increase of rabbits is provided in equation 4 in
Appendix 1. The maximum rate of increase (5.5 -
4.6 = 0.9 per quarter) and decrease (-4.6 per
quarter) were estimated by fitting the relationship
to data from Yathong during a period of predator
control, while the demographic efficiency (0.0045)
was estimated by Choquenot (1992). This
relationship could be altered to express r as a
function of intake rather than standing biomass
(see equation 5 and 6 in Appendix 1).
The resultant dynamics in a model including
rainfall, vegetation and rabbits, where the intake
rate is prey dependent and the numerical response
is intake dependent, is shown in Figure 7a.
Replacing the prey dependent intake rate with a
ratio dependent intake rate stops the population
from declining to very low levels and allows it to
respond more rapidly to improved conditions
(Figure 7b).

Interactions between feral cats, foxes, rabbits and native carnivores 37

 (a) (c)
80

80
60

60
Rabbits/ha

Rabbits/ha
40

40
20

20
0

0
0 5000 10000 15000 0 5000 10000 15000
Time (days) Time (days)

(b) (d)
80

80
60

60
Rabbits/ha

Rabbits/ha
40

40
20

20
0

0 5000 10000 15000

0

Time (days)
0 5000 10000 15000
Time (days)

Figure 7. Rabbit vegetation models.

(a) Rabbit-vegetation model with ‘prey dependent’ intake and the numerical response determined by intake. (b) Rabbit-
vegetation model with ‘ratio dependent’ intake where the numerical response is determined by intake. (c) Rabbit-
vegetation model with ‘prey dependent’ intake, rabbit populations increase in response to pasture growth and decline as
a function of intake when pasture is not growing. (d) Rabbit-vegetation model with ‘ratio dependent’ intake, rabbit
populations increase in response to pasture growth, and decline as a function of intake when pasture is not growing.

Interaction 4 Interaction 5
Density-dependence in rabbit populations Native prey effect on vegetation
Interaction 4 allows for unexplained density- Two general categories of native prey could be
dependence in rabbit populations. This could considered: Abundant native prey such as
come from social interactions, (see equation 9 in kangaroos, which are likely to have an impact on
Appendix 1 for the numerical response function). the vegetation (Caughley 1987); rare and
This approach was proposed by Caughley and threatened native prey. The latter may be
Krebs (1983) and has been used for possums by herbivores, or may feed on alternative foods such
Bayliss and Choquenot (2002) and for ferrets and as insects, but they are unlikely to have significant
feral cats by Arthur and Norbury (unpubl. data). effects on vegetation biomass at the low densities
This has not been considered for rabbits, but there at which they currently exist.
is no evidence to our knowledge that it is
necessary for rabbits.

Interactions between feral cats, foxes, rabbits and native carnivores 38

Effect of vegetation on native prey The daily consumption of rabbits in grams per fox
The effect of vegetation on kangaroo dynamics per day is given by equation 10 in Appendix 1.
was modelled by Caughley (1997) and Bayliss
To measure the true impact of foxes on rabbits, kill
(1987). No data were provided to model the
rates are required. However, these data are not
dynamics of threatened native prey in response to
available for semi-arid systems to our knowledge.
vegetation or rainfall in the absence of predators.
If the average size of rabbits in fox stomachs is
Pech and Hood (1998) modelled native prey using
782 g (Pech et al. 1992) this implies that foxes kill
the same numerical response to vegetation
a maximum of 1096/782 = 1.4 rabbits day-1. The
biomass as rabbits, because they were mainly
average field metabolic rate of foxes in the central
interested in the effect of a different functional
western tablelands of NSW in autumn was
response of foxes to the two types of prey.
estimated as 2 328 kJ day-1 for male foxes (av.
Marsupials have a lower maximum birth rate than
weight 5.6kg) and 1 681 kJ day-1 for female foxes
eutherian mammals of the same size (Sinclair
(av. weight 5.4kg) (Winstanley et al. 2003). To
1997), which may suggest they have a lower
yield this much energy requires ~ 435 g of
maximum rate of increase than eutherian
mammalian prey for males and ~ 314 g of
mammals of the same size. However, rate of
mammalian prey for females (Winstanley et al.
increase is determined by both reproduction and
2003). This is well below the satiating intake
survival (assuming closed populations), and
estimated by Pech et al. 1992, but may provide an
Sinclair (1997) suggested that marsupials may
estimate of the sustained daily intake rate below
have higher survival rates than eutherian mammals
which fox rate of increase is negative. However,
of the same size, which offsets lower reproduction.
applicability of these results from a temperate
Hence, it is uncertain how native mammals
system to semi-arid systems is uncertain.
respond to their food supply in the absence of
predation. These data are required to properly As with rabbits some form of ratio dependent
model the population dynamics of threatened functional response may be required to capture the
native Australian prey. likelihood that if fox populations decline in
response to lower rabbit densities the average
intake of rabbits by the remaining foxes may
Interaction 6 increase. Data were not available to explore this.
Other Regulating Factors
There are no data available to our knowledge on Effect of rabbits on fox rate of increase
other factors, such as social factors, which may The common occurrence of rabbit in the diet of
regulate the abundance of native prey. foxes at high rabbit density and the low occurrence
at low rabbit density, combined with the observed
Interaction 7
decline in fox populations following declines in
Competitive interactions between introduced and rabbit populations in times of drought (Myers and
native species Parker 1975a; Myers and Parker 1975b; Newsome
Interaction 7 represents competitive interactions et al. 1989) suggested the abundance of rabbits
between introduced and native species that are not strongly influences the abundance of foxes.
captured by competition for resources through the However, it is likely many other foods consumed
functional response (e.g. competition for shelter). by foxes are also reduced in abundance during
To our knowledge few studies have directly drought, and confirmation of the reliance of foxes
addressed competition between rabbits and native on rabbits to maintain high densities requires a
species. In one study that did, no evidence was reduction in rabbit density when environmental
found that competition with rabbits affected conditions are still good (i.e. times of average or
burrowing bettongs (Robley et al. 2002). above average rainfall). Before the arrival of RHD
in Australia, To our knowledge there is no data for
good conditions where rabbit density declined in a
Interaction 8 semi-arid system, before the arrival of RHD in
Australia. The introduction of myxoma virus in the
Effect of foxes on rabbits (semi-arid and arid 1950s might have resulted in these conditions but
systems) only anecdotal information is available on the
In semi-arid systems rabbits comprise a large consequences for predator populations at that time
percentage of the fox diet, particularly when at high (Newsome et al. 1997).
density (Pech and Hood 1998). Pech et al. (1992)
estimated the functional response of foxes to
We have obtained two sets of data from semi-arid
rabbits at Yathong based on the weight of rabbit
systems where fox density was monitored before
found in fox stomachs and an estimate of gut
and after the arrival of RHD. In one, Holden and
passage rates. They fitted a Holling Type III
Mutze (2002) described the response of fox
functional response to the data (Holling 1959).
populations to the large reduction in rabbit density

Interactions between feral cats, foxes, rabbits and native carnivores 39

that occurred due to RHD in the Flinders Ranges same conversion to rabbits ha-1 used by Pech and
National Park (FRNP). Rabbit populations Hood (1998) is applicable in this system (they
dropped from a long-term average of ~32 km-1 to assumed 40% sightability within a 150 metre wide
~5 km-1 after RHD was first detected at a site that transect) these rabbit densities (over 4 ha-1) should
comprised part of the area where fox density was have produced close to maximum rates of increase
assessed. In the year following the arrival of RHD for foxes every year, rather than the stable
(1996) the area received average rainfall, while in population densities observed. Under their model
1997 it received above average rainfall. These the rate of increase of foxes is zero when rabbit
data suggest that rabbit population density had a density is ~0.4 ha-1. This suggests either fox
significant effect on fox population density in the density was regulated by some other factor,
Flinders Ranges. In the other study, at Roxby possible social interactions (Interaction 9), or the
Downs (Read and Bowen 2001; WMC Olympic conversion to rabbits ha-1 used by Pech and Hood
Dam unpubl. data), rabbit and fox populations (1998) greatly overestimated the rabbit density in
were monitored by spotlight counts across two this system. Another possibility is that the model of
areas. The fox population declined prior to the Pech and Hood (1998) greatly underestimates the
arrival of RHD in early 1996, but this was during a rabbit density at which the fox rate of increase is
period of above average rainfall (1992) making the zero.
relationship between rabbit population density and
fox population density less certain. After the arrival No experimental control transects were monitored
of RHD the fox population remained low despite at the same time as the FRNP fox transect, but in
good rainfall in 1997. February 1996, after the arrival of RHD in
November 1995 on the sites used by Mutze et al.
We analysed the data from the Flinders Ranges in
(2002), two additional large areas were assessed
more detail by assuming rabbit density across the
for fox abundance: an adjacent unbaited area and
entire study area followed the pattern observed by
a distant unbaited area. Rabbit densities were low
Mutze et al. (2002) on unripped sites. The fox
in all of these areas after the arrival of RHD
counts shown in Figure 3 of Holden and Mutze
(Holden unpubl. data). In April 1996 these two
(2002) were taken from a number of different
areas had indices of 0.79 foxes km-1 and 1.07
areas. We combined data from all of the areas to
foxes km-1 respectively at the time of peak
construct one time series for fox population
dispersal, again consistent with the dispersal
density. The result is shown in Figure 8, and
peaks recorded by Mutze et al. (2002). In October
suggests that before RHD arrived in 1995, foxes
1996 the ‘troughs’ were 0.13 foxes km-1 and 0.22
had consistent dispersal peaks in late summer and
foxes km-1 respectively, also generally consistent
consistent troughs prior to the peaks. Dispersal
with the pattern pre-RHD. However, the following
peaks occur in late summer and early autumn due
year the large dispersal peak did not occur in either
to the seasonal nature of fox breeding, when
area. If we consider the distant unbaited area as
juveniles disperse. In other words, prior to RHD
the area where fox density was not manipulated
the spotlight data suggested stable fox population.
(i.e. foxes were not controlled) and combine it with
1.20 50
the data on foxes collected by Mutze et al. (2002),
45
we can compare rates of increase from the troughs
1.00
40
to the peaks (‘recruitment phase’) and from the
35
peaks to the troughs (‘decline phase’) in relation to
0.80
rabbit density from unripped transects (Mutze et al.
Rabbits/km
Foxes/km

30
0.60 25
2002). We can also compare the rates of increase
20
from trough to trough, and from peak to peak, (i.e.
0.40
15
different estimates of yearly rates of increase).
0.20
10 The rate of increase over the recruitment phase or
5 decline phase was not related to rabbit density
0.00 0 (Figure 9).
Dec-91 Sep-94 Jun-97 Mar-00

Figure 8. Fox and rabbit spotlight counts from

the Flinders Ranges.
Data from a number of sites within each area were
combined to generate single time series for each
species. (Black line, Holden and Mutze 2002; Grey line,
Mutze et al. 2002).
Over this three year period, the rabbit index
remained high, ranging from ~16 km-1 up to ~45
km-1. Most of that time it was over 25 km-1. If the

Interactions between feral cats, foxes, rabbits and native carnivores 40

 (a) (b)
1.6 0 10 20 30 40
1.4 0
1.2 -0.2
1 -0.4
R2 = 0.0002
r 0.8 -0.6
0.6 -0.8
r
-1
0.4 R2 = 0.0231
-1.2
0.2
-1.4
0 -1.6
0 10 20 30 40 -1.8
rabbits/km rabbits/km

Figure 9. Rate of increase of foxes (r) during (a) the recruitment phase (spring – late summer) and (b)
the winter decline (late summer to spring), plotted against rabbit index of abundance.
The recruitment rate of increase was calculated as ln(Nsummer/Nspring), where the spring estimate was taken as the lowest
fox index in Sep-Nov, and the summer index was taken as the highest estimate in Feb-May the following year. The
rabbit index was the corresponding spotlight count at the start of the recruitment phase in spring. The winter decline rate
of increase was calculated as ln(Nspring/ Nsummer), where the spring estimate was taken as the lowest fox index in Sep-
Nov, and the summer index was taken as the highest estimate in Feb-May prior to the spring. The rabbit index was the
corresponding spotlight count in summer.

The rate of increase from peak to peak (and trough to trough) was not related to rabbit density (Figure 10).

(a)
(b)
1
1
0.5
R2 = 0.0062 0.5 R2 = 0.012
0 0
r 0 10 20 30 40 50 r 0 10 20 30 40
-0.5 -0.5

-1 -1

-1.5 -1.5
rabbits/km rabbits/km

Figure 10. Fox rate of increase from (a) peak to peak and from (b) trough to trough plotted against
the rabbit index mid-way between the peaks or troughs.
Despite no apparent relationship between fox At Roxby Downs the fox population declined prior
population rate of increase and rabbit density, fox to the arrival of RHD in 1995 (Figure 11).
density appeared to decline following the arrival of Unfortunately we do not know whether rabbit
RHD if we consider either peak or trough indices density was high prior to April 1989, this could
of foxes (Figure 8). At the distant site the post- explain the initially high fox density. If the same
RHD peak fox index was ~36% of the pre-RHD conversion to rabbits/ha used by Pech and Hood
peak index, while the post-RHD trough fox index (1998) is applicable in this system rabbit densities
was ~50% of the pre-RHD trough index (Holden during the period when foxes declined should
and Mutze 2002). have produced positive rates of increase for foxes
every year, rather than the observed decline.
If shooting foxes had little impact on the fox index,
the FRNP data suggest fox density was limited by
some factor other than rabbit density when rabbit
density and fox density was high, but following the
decline in rabbit density post RHD, fox density
may have been limited by rabbit density.

Interactions between feral cats, foxes, rabbits and native carnivores 41

 0.2 20
where lagged rainfall was used as an index of
food availability (Pech et al. 1997), but data were
0.18 18
not available. An example of the approach is
0.16 16 described in equations 12 and 13 in Appendix 1.
0.14 fox 14
Interaction 9
Foxes and cats km -1

cat
0.12 12

Rabbits km -1
rabbit
Density-dependence in fox populations
0.1 10
Interaction 9 allows for unexplained density-
0.08 8 dependence in fox populations. There is some
0.06 6 evidence from the Flinders Ranges this may
0.04 4
occur, based on the observations that fox density
remained relatively stable prior to the release of
0.02 2
RHD despite high rabbit densities, although
0 0 tenuous (see above). Foxes are highly territorial,
Aug-87 May-90 Jan-93 Oct-95 Jul-98 Apr-01 Jan-04
and probably exist as a family group with one
male, one female and cubs prior to cub dispersal.
Figure 11. Rabbit, cat and fox indices of Family groups with more than one adult vixen
abundance at Roxby Downs. have been observed outside Australia, although it
Read and Bowen 2001 and Read and Bowen, unpubl. is highly unlikely in semi-arid and arid
data. To simplify the data, which were noisy, we fitted environments (Saunders et al.1995) and adult
smoothed splines.
females with overlapping home ranges have been
This highlights one of the problems we have in recorded in eastern Australia (Saunders et al.
developing quantitative models for these systems; 2002), but these tend to occur in areas with
the data are usually not collected in an abundant resources such as urban environments.
appropriate way. Rabbit abundance indices are Hence, fox density may be regulated by social
usually expressed as rabbits per spotlight km, and interactions.
it is unclear whether an observed number per
Density dependence could be included by adding
spotlight km in one study is equivalent to the
a term to the Pech and Hood (1998) model for fox
same number per spotlight km in another study.
numerical response, (equation 14 in Appendix 1),
In the Flinders Ranges study the peak rabbit
or by adding a density dependent term to either
density in April 1992 averaged ~40 km-1, while in
the decline phase or increase phase in the more
the Roxby Downs area it averaged ~5 km-1. Was
detailed model but is not warranted on the
there an eight-fold difference or were the apparent
available data.
different densities due to partly to measurement
protocols or site-dependent factors such as Interaction 10
sightability? It is also unclear whether the
The effect of feral cats on rabbits
temporal sequence of observations within sites
In semi-arid systems rabbits comprise a large
really reflects true densities because of the
percentage of the diet of feral cats (Holden and
influence of changing vegetation on spotlight
Mutze 2002; Read and Bowen 2001; Olympic
counts. The numerical response of foxes to
Dam unpubl. data; Risbey et al. 1999). Based on
rabbits is expressed in equation 11 in Appendix 1
the data presented in Holden and Mutze (2002)
and is from Pech and Hood (1998).
and rabbit spotlight counts from (Mutze et al.
The model was based on: the allometric estimate 2002), feral cats have a type II functional
of maximum rate of increase, an observed response to rabbits, and have higher predation
maximum rate of decline during the drought at rates on rabbits at lower densities compared to
Yathong, and a value for demographic efficiency foxes (Figure. 12). As with the Yathong data
obtained by trial and error, to produce an overall (Pech et al. 1992) foxes appear to have a type III
model that showed qualitatively reasonable functional response to rabbits. The fitted model
behaviour in terms of rabbit population dynamics. for feral cats is a Holling type II functional
response, (see equation 15 in Appendix 1).
A more mechanistic approach would express the
fox rate of increase in terms of food intake, and To establish a cat functional response to rabbits
would also break the year into recruitment and we have assumed that occurrence in stomachs
decline (non-recruitment) periods, but whether this reflects actual intake. To properly model the
is feasible requires future investigation. Our initial offtake rate of feral cats on rabbits we need to
intention was to at least express rate of increase express the equation in terms of biomass of rabbit
over the recruitment period and the decline period rather than percentage occurrence in the diet. To
in terms of rabbit availability, as has been done for measure the true impact of feral cats on rabbits,
predators in New Zealand (Arthur and Norbury, as for foxes, we actually require kill rates but
unpubl. data), and for foxes in temperate areas these data are not available for semi-arid

Interactions between feral cats, foxes, rabbits and native carnivores 42

systems. If we assume percentage occurrence in rabbit density could mean that feral cats are not
the diet reflects actual intake of rabbit by feral cats as capable of accessing alternative prey as foxes
then we require an estimate of satiating intake. are, and hence when rabbit numbers decline
As a first estimate of this we apply fox satiating foxes are still able to maintain their population
intake (see equation 10 in Appendix 1), scaled to density, while feral cats decline.
reflect average body size of feral cats and foxes in
The results from South Australia suggested that
semi-arid areas (based on data from Roxby
cat density was affected by rabbit density, but
Downs (Read and Bowen 2001). This gives an
whether feral cats continue to maintain higher
estimate of (3.4[average weight of feral
rates of increase than foxes at some rabbit
cats]0.75/4.6[average weight of foxes]0.75) x 1096 =
densities was equivocal. In the Flinders Ranges
874g. To express the functional response in
cat density declined and feral cats were in poor
terms of a rabbit density we also need to convert
condition following the reduction in rabbit density
rabbits per spotlight km to rabbit ha-1.
despite good environmental conditions (Holden
80 and Mutze 2002). At Roxby Downs, cat density
70
also declined following the arrival of RHD, but
when rabbit density was intermediate between the
60
initial high density and the low density that
% occurrence of rabbit

50
occurred after RHD arrived, there was some
40 evidence the cat population increased while the
30 fox population declined (Figure 11).
20
Establishing a numerical response for feral cats
10 It has been reported that feral cats preferentially
0 prey on juvenile rabbits (Catling 1988; Jones
0 5 10 15 20 25 30 35 40
1977) and this may explain the slow population
Rabbits/km
recovery observed in Victoria following a cull of
Figure 12. Cat (open diamonds) and fox feral cats when few young rabbits were available
(closed squares) functional responses to (Jones and Coman 1981). If feral cats respond to
rabbits in the Flinders Ranges. the availability of juvenile rabbits rather than the
Data on occurrence of items in stomachs are from total abundance of rabbits, then it may be
Holden and Mutze (2002). Rabbit abundance indices necessary to express the numerical response of
are from unripped transects of Mutze et al. (2002). feral cats in terms of availability of juvenile rabbits.
To our knowledge data are not available to
While not from a semi-arid system, data from parameterise a model of this form. One possibility
Burrendong Dam in NSW provides some may be to model the numerical response of feral
evidence that feral cats continue to prey on cats, as a function of the numerical response of
rabbits at lower densities than foxes (Davey et al. rabbits, assuming those juvenile rabbits are
In Prep; Molsher 1999). available when the rate of increase of rabbits is
The effect of rabbits on cat rate of increase positive.
The common occurrence of rabbit in the diet of If we use a functional form based on the
feral cats at high rabbit density, combined with the abundance of rabbits as for foxes (e.g. like eqn
observed decline in cat populations following 11, Appendix 1), the maximum rate of increase of
declines in rabbit populations in times of drought feral cats based on an average body size of 2.8
(Newsome et al. 1989), suggests cat rate of kg is 0.99 per year (0.25 per quarter). The
increase could depend on the abundance of maximum rate of decrease is unknown, and
rabbits. However, the availability of alternative difficult to estimate because of the difficulty in
food is also likely to decrease during droughts. If assessing the abundance of feral cats. The
we consider the functional response of feral cats demographic efficiency is unknown, but if feral
to rabbits relative to the functional response of cats can increase when rabbit densities are low
foxes to rabbits we could make contrasting (cf. foxes) their demographic efficiency would be
predictions about how we expect these predators larger than foxes. The effect of demographic
to respond to a reduction in the density of rabbits. efficiency on resultant dynamics is explored
One possibility is that feral cats continue to exploit below.
rabbits at lower densities than foxes because they
are better able to catch rabbits than foxes. Then Interaction 11
we would hence predict that when rabbit densities
Density-dependence in cat populations
drop foxes are disadvantaged and decline, while
Interaction 11 allows for unexplained density-
feral cats can continue to exploit rabbits and can
dependence in the dynamics of cat populations.
maintain their population density. Alternatively,
Whether some form of social regulation affects the
the continued presence of rabbit in the diet at low
abundance of feral cats is unknown.

Interactions between feral cats, foxes, rabbits and native carnivores 43

Interaction 12 Interaction 14
Interactions between foxes and cats not captured Effect of feral cats on native prey
by competition The functional response of feral cats to native
Interaction 12 represents interactions between prey is unknown, but feral cats have been
foxes and feral cats not captured by competition implicated in the decline and extinction of a
for resources through the functional response. number of native species (Dickman 1996; Risbey
This could be interference competition or direct et al. 2000). To properly quantify the impact of
predation. The effect of foxes on feral cats and feral cats on native prey we ideally need kill rates
vice versa is unknown. Possible increases in cat of native prey by feral cats as a function of both
abundance in response to a reduction in fox native prey abundance and rabbit abundance. As
populations through poisoning at Herrison Prong for foxes it is essential to know whether kill rates
(Risbey et al. 2000) and to a natural reduction in of native prey by feral cats are high because
the fox population at Roxbury Downs (Read and native prey are intrinsically vulnerable to feral
Bowen 2001) suggested that foxes may reduce cats.
cat abundance. However, large-scale control of
Effect of native prey on the rate of increase of
foxes in the Flinders Ranges undertaken while
feral cats
rabbit populations were still high did not result in
Threatened native species are unlikely to
an increase in the cat population (Holden and
contribute significantly to the rate of increase of
Mutze 2002).
feral cats, but the contribution of more abundant
One way to model these interactions is by native prey is unknown.
reducing the numerical response of one predator
Interaction 15
species in response to the presence of the other
(equation 16 in Appendix 1). Data are not Disease in rabbit populations
available to parameterise this relationship. RHD had a large impact on rabbit populations in
semi-arid systems when it was first introduced,
Interaction 13
and it appears to be still effectively limiting the
Effect of foxes on native prey abundance of rabbits at Roxby Downs (Read and
The functional response of foxes to native prey is Bowen 2001) and in the Flinders Ranges (Mutze
unknown, but foxes have been implicated in the et al. 2002) and elsewhere (Sandell and Start
decline and extinction of many native species in 1999). The longer-term effectiveness of RHD at
the critical weight range (Burbidge and McKenzie limiting rabbit populations is uncertain. The effect
1989). Pech and Hood (1998) modelled a generic of myxomatosis on rabbit population dynamics
type II functional response, which is capable of since the release of the arid-adapted rabbit flea is
driving prey to extinction (Sinclair et al. 1998). To also uncertain.
properly quantify the impact of foxes on native
prey we ideally need kill rates of native prey by A simple model for preliminary
foxes as a function of both native prey abundance exploration
and rabbit abundance. The latter factor is
The discussion above has identified many gaps in
required to deal with any preference foxes might
our understanding of the relationships identified in
have for different prey. In particular, it is essential
Figure 5. There are far too many permutations
to know whether kill rates of native prey are high
and combinations to simulate the effects of
because native prey are intrinsically vulnerable to
varying all the different parameters. Below we
foxes, because this changes the structure of the
present a simple model that builds on the Pech
model required for this part of the system.
and Hood (1998) model. We explore the
consequences of adding density dependence in
fox population dynamics and extend the model by
Effect of native prey on rate of increase of foxes adding feral cats and ignoring any effects of RHD
Threatened native species are unlikely to on rabbit populations. The model is simulated
contribute significantly to the rate of increase of using 120 years of actual rainfall from a semi-arid
foxes, but many other relatively abundant native area of Australia (annual mean 321 mm, sd 110
prey sources might be important. Foxes have a mm). For vegetation we use equation 1 with the
diverse diet, for example insects are often eaten V* term removed; for the rabbit functional
in summer, but data are not available to properly response we use equation 2; for the rabbit
quantify the contribution of each prey type to the numerical response we use equation 4; for the fox
rate of increase of foxes. functional response we use equation 10; for the
fox numerical response we use equation 14; for
the cat functional response we use a modified
form of equation 15 (see equation 17 in Appendix
1). For the cat numerical response we use

Interactions between feral cats, foxes, rabbits and native carnivores 44

equation 16, with a = 0.56 (the same as foxes),
and c-a = 0.25 (i.e. the intrinsic rate of increase rm
= c-a is estimated using the allometric relationship
of Sinclair (1996) and an average body weight of
2.8 kg for adult females). We vary the
demographic efficiency d and the effect of foxes
on feral cats h. In the absence of any other
information we set the minimum density of feral
cats the same as foxes (0.1 km-2).
The effect of density dependence on fox–rabbit
interactions
Figure 13 shows simulations from a vegetation-
rabbit-fox model with density dependence in the
numerical response for foxes (g > 0 in equation
14) and without (a & without density dependence
g = 0 in equation 14). Density dependence
restricts fox population growth and allows rabbit
populations to achieve higher densities.
As mentioned above, whether these dynamics
properly reflect fox population dynamics is
uncertain, and the sensitivity of the model to
changes in rainfall raises some concerns. The
rainfall data used for the simulation are actual
rainfall data from a semi-arid area in Australia,
and there is a slight upward trend in the yearly
rainfall. This slight upward trend had a significant
effect on fox population dynamics, with fox
populations at higher densities in the last 50 years
of the simulation compared with the first 50 years.
(See also Davis et al. 2003 for a discussion of the
effect of changing either the mean or the variance
of rainfall distribution in the model used by Pech
and Hood 1998).

Interactions between feral cats, foxes, rabbits and native carnivores 45

(a) (b)

2.5
40

2.0
Foxes per sq. km
30
Rabbits/ha

1.5
20

1.0
10

0.5
0.0
0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

(c) (d)
2.5
40

2.0
Foxes per sq. km
30
Rabbits/ha

1.5
20

1.0
10

0.5
0.0
0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

Figure 13. Simulated population trajectories without additional density dependence in fox population
dynamics (i.e. the original Pech and Hood 1998) model (a & b), with density dependence added (c &
d, g = 0.0015).
Figure 14 shows the effect of controlling rabbits the sensitivity of the model to an increase in
on fox population density. Controlling rabbits to rainfall. When rabbits are controlled below ~0.4
less than 1 ha-1 has little effect on fox population ha-1 the fox population must decline to its
dynamics. Controlling rabbits to 0.5 ha-1 has a minimum allowable density under this model (this
large effect on fox populations during the first 50 is the value below which the demographic
years of the simulation, but fox populations are efficiency of foxes results in fox rate of increase
relatively high during the last 50 years, showing being negative).

Interactions between feral cats, foxes, rabbits and native carnivores 46

(a) (b)
1.2

Foxes per sq. km

2.0
Rabbits/ha

0.8

1.0
0.4
0.0

0.0
0 20 40 60 80 100 0 20 40 60 80 100

Years Years

(c) (d)
1.2

Foxes per sq. km

2.0
Rabbits/ha

0.8

1.0
0.4
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

(e) (f)
1.2

Foxes per sq. km

2.0
Rabbits/ha

0.8

1.0
0.4
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

Figure 14. Effects of rabbit control on fox population dynamics.

The density dependent factor, g in equation 14, is zero for all the simulations. The figures on the left show the level to
which rabbit density is controlled (1 ha-1, 0.5 ha-1, 0.3 ha-1). The grey line in the figures on the right shows the fox
density when rabbit density is not controlled. The black line shows the response of fox density to rabbit control.
The effect of adding feral cats is required to properly model their dynamics. If a
Adding feral cats to the model (with parameter h = negative effect of foxes on feral cats is added, the
0 in equation 16) reduces the abundance of cat population decreases and the fox population
rabbits through predation, and reduces the increases (Figure 16). The problem remains that
number of foxes through competition for rabbits rabbits and foxes are often at their defined
(compare Figure 14 a & b with Figure 15). The minimum values.
strength of the effect is determined by the
demographic efficiency of feral cats (Figure 15). It
is evident from this formulation of the model that
all species are often at their defined minimum
values. This indicates that a much better
understanding of the relationships between the
species (and the importance of prey items other
than rabbits for maintaining predator populations)

Interactions between feral cats, foxes, rabbits and native carnivores 47

 10 15 20 25

10 15 20 25
Rabbits/ha

Rabbits/ha
5

5
0

0
0 20 40 60 80 100 0 20 40 60 80 100

Years Years
2.0

2.0
Foxes per sq. km

Foxes per sq. km

1.5

1.5
1.0

1.0
0.5

0.5
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years
3.0

3.0
Cats per sq. km

Cats per sq. km

2.0

2.0
1.0

1.0
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

Figure 15. Simulated population dynamics using the Pech and Hood (1998) model with feral cats
added.
A comparison between the figures on the left and those on the right shows the effect of changing the demographic
efficiency of feral cats from 2 to 5: feral cats become more competitive, their population increases, the rabbit population
is generally kept lower, and the fox population is reduced.

Interactions between feral cats, foxes, native carnivores and rabbits 48

 15

15
Rabbits/ha

Rabbits/ha
10

10
5

5
0

0
0 20 40 60 80 100 0 20 40 60 80 100

Years Years
1.5

1.5
Foxes per sq. km

Foxes per sq. km

1.0

1.0
0.5

0.5
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years
3.0

3.0
Cats per sq. km

Cats per sq. km

2.0

2.0
1.0

1.0
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100

Years Years

Figure 16. Simulated population dynamics using the Pech and Hood (1998) model with feral cats
added.
The demographic efficiency of feral cats was set at 5. A comparison between the figures on the left and those on the
right shows the effect of adding a negative effect of foxes on feral cats (h=0.003, eqn 16). Feral cats average lower
densities, foxes average higher densities and rabbit density increases also as the more efficient predator is suppressed.

Summary
address issues of stability at lower resource
The preliminary simulation models explore the densities. More specifically, there is a clear need
potential interactions between rabbits, foxes and to properly quantify the relationship between
feral cats, but they are based mainly on rabbits and the two predators. Numerical
hypothesised relationships. The sensitivity of the responses for the two predators should be
model to small changes in rainfall suggests a more determined in relation to both the abundance of
detailed understanding of the relationships is rabbits (or juvenile rabbits) and simultaneously the
required. The tendency of the model to reach the abundance of alternative food sources. Based on
lower defined minimum for all species when both diet studies both foxes and feral cats consume
predators are present also suggests a much better many prey species other than rabbits, but to our
understanding of the relationships is required. knowledge no quantitative information currently
Refinements to the model such as the utilisation of exists to build this into the predators’ numerical
ratio-dependent functional responses may partly responses. Ideally, numerical responses would be

Interactions between feral cats, foxes, native carnivores and rabbits 49

based on intake rates, but whether this is feasible 1.2
is currently unknown. Foxes and feral cats have
clearly defined breeding seasons, and future 1
investigation of the relationships should focus on
these characteristics. The pulsed dynamics 0.8

foxes/km
resulting from breeding and non-breeding seasons
0.6
will affect model behaviour and are likely to R 2 =0.0124

generate predictions that are substantially different 0.4

to the continuous models we have used above.
This is an important topic for future research. The 0.2
effect of foxes and feral cats on the rate of
increase of each other through interference 0
competition, competition for resources or predation 0 10 20 30 40 50
is essentially unknown. rabbits/km
To properly quantify and model the impact of foxes
and feral cats on both rabbits and native prey Figure 17. Rabbit and fox abundance
requires kill rates of these prey, assessed in relationship in temperate systems.
relation to the availability of all prey types. This is Composite plot using spotlight data from Burrendong (A.
particularly important for native prey and we Newsome unpubl. data; Davey et al. in prep.), the ACT
(Don Fletcher et al. unpubl. data) and the central
currently have no data on this. At the same time it
western slopes of NSW (Glen Saunders unpubl. data).
is essential to develop an understanding of the
population dynamics of native Australian prey and Because of likely differences in sightability, the
the population dynamics of rabbits using data continued use of spotlight counts makes it difficult
obtained since the arrival of RHD, and in the to produce quantitative models using data from
absence of predation from introduced predators. different areas. However, initial interpretation
suggests it may not be appropriate to link fox
Temperate systems
dynamics to rabbit dynamics in temperate systems.
There are no interactive models currently available
Alternatively, despite some reductions in rabbit
for temperate systems that include foxes. We
abundance due to RHD, rabbit populations may
have explored data sets from Burrendong Dam (A.
still not have reached the low densities required to
Newsome unpubl. Data; Davey et al. in prep.), the
reduce fox density below the current partially
ACT region (Don Fletcher et al. unpubl. data)
controlled or socially regulated level in agricultural
andfrom the central western slopes of NSW (G.
landscapes. This requires further exploration.
Saunders unpubl. data). The data sets are
Certainly, the functional response of foxes to
characterised by short spotlight transects (range
rabbits estimated before and after the arrival of
4.1km – 30 km, with most <10 km). Short transects
RHD at Burrendong suggests foxes have a diverse
result in highly variable counts of foxes. This is
diet (Davey et al. in prep) and may not be reliant
because foxes have large home ranges and are
on rabbits.
generally shy and cryptic in their behaviour. With
one exception, none of the data sets show the Feral cats
clear seasonal recruitment peaks that could be Feral cats are rarely seen in spotlight counts in
anticipated for foxes. Nonetheless, spotlight temperate systems and no quantitative numerical
transect lengths of 16.2 km, 6.2 km and 4.2 km relationships can be established from the available
were sufficient to detect the effects of poison data. Dietary data indicate rabbits are important
baiting on fox populations at Burrendong (A. prey for feral cats even at low rabbit population
Newsome unpubl. data; Davey et al. in prep.). densities (Davey et al. in prep). Molsher (1999)
suggested feral cats showed a behavioural
However, we found no evidence from any of the
response to the removal of foxes at Burrendong,
data from temperate systems that the reduction in
indicating potential competition between these
rabbit density due to RHD had any effect on fox
species, but no numerical response was evident.
abundance. Using a composite plot, which
However, appropriate data to assess whether cat
assumes that spotlight indices were comparable
populations increased in response to the reduction
between all temperate areas for which data were
in foxes were not available, and a way of
available, we found no relationship between rabbit
quantifying cat abundance is needed for future
abundance and fox abundance (Figure. 17).
studies.

Interactions between feral cats, foxes, native carnivores and rabbits 50

Summary – temperate systems
The few data available for temperate systems
suggested fox population dynamics may not be
linked as strongly to rabbit dynamics as they
appear to be in semi-arid systems. Alternative
models are likely to be required for temperate
systems (e.g. Pech et al. 1997). Almost certainly
these models will require data on the interactions
of predators and a wide variety of foods, including
foods associated with human activity.
At this stage there is no evidence that interactive
population models can be transferred directly
between ecosystems in different climatic zones.
Research to fill knowledge gaps will need to be
applied separately in temperate and arid areas of
Australia. In addition, no information is available to
test whether or not the models, or the existing
demographic data, are applicable to areas such as
northern Australia (where the range of foxes and
rabbits may be expanding); or to areas such as
Tasmania, where there are substantial differences
in the abundance of native predators and the
availability of native prey species for foxes (should
they ever establish).

Interactions between feral cats, foxes, native carnivores and rabbits 51

6 Implications for Integrated control

The general principals and strategies of integrated the resource it threatens, or the relative costs and
control are outlined in Braysher (1993). Williams et benefits of integrated control techniques for a
al. (1995) and Saunders et al. (1995) provided combination of predators and rabbits, and native
guidelines for the application of these principals species.
and strategies to rabbits and foxes, respectively.
Implementing an integrated control program is
The general principals set out by Braysher (1993) therefore limited by the above constraints.
included defining the problem, clearly stating the Monitoring and evaluating control programs is
objectives and setting out the criteria of success limited by the lack of reliable techniques to monitor
and failure, evaluating various management changes in the abundance of feral cats and many
options, implementing actions; monitoring and populations of native species, and to a lesser
evaluating the outcomes against the objectives. extent changes in fox abundance.
Being able to clearly state the objectives of an
integrated control program requires an
understanding of the impacts of feral cats, foxes
and rabbits, or a combination of these species, on
native fauna. This review highlights the general
lack of knowledge on the impacts of these species
and the interactions between species.
A risk adverse approach would be to undertake
integrated control wherever feral cats, foxes and
rabbits co-occur. However, this may not be
practical or possible due to limitations on
resources.
There appears to be a link between feral cats and
rabbit abundance and fox and rabbit abundance in
semi-arid and arid areas. A reduction in rabbits,
under the right circumstances, can lead to a
lagged reduction in both fox and feral cat
abundance. In situations were there are small
populations of native species that are at risk from
predation, and where rabbits are the primary prey
of foxes and/or feral cats, it may be beneficial to
undertake integrated control. In areas where
rabbits are not the primary prey of feral cats or
foxes, integrated control may not be necessary,
and targeted predator control may be a better
investment of limited resources.
Assuming that reliable information on the impacts
of feral cats, foxes and rabbits can be obtained for
a particular area, setting the criteria of success or
failure is currently hindered by the available
techniques to assess changes in abundance of
feral cats and in small populations of rare native
species.
Currently, there is an array of control techniques
and strategies used for rabbits and foxes, and only
limited strategic approaches and tactical tools for
the control of feral cats (Algar et al. 1999). To our
knowledge, few studies have investigated the
efficiencies and effectiveness of these control
strategies, in terms of the target prey species and

Interactions between feral cats, foxes, native carnivores and rabbits 52

7 Gaps in Knowledge
on from these are a series of subset areas that
also require investigation:
The aim of this review was to synthesise the
current state of knowledge on the interactions
between feral cats, foxes, rabbits (their control)
and their impacts on native species, including 1. How to effectively monitor changes in
native herbivore and predator species. abundance of introduced predators,
A number of studies have provided valuable particularly feral cats.
insights into the impacts that changes in prey Currently, there are no robust methods for
abundance (via control or natural events) can have assessing changes in predator abundance in
on populations of introduced predators, and how Australia, particularly for feral cats. Thus, there are
predators can influence the abundance of prey no means of assessing the effectiveness of control
species. The interactions between foxes (and to a operations at reducing predator abundance. This
lesser extent feral cats) and native species have is the highest priority, as the remaining knowledge
also been studied through both field gaps all require accurate assessment of changes
experimentation and theoretical modelling. in predator abundance as a result of pest control
Despite these important studies, there are many operations. A review of monitoring techniques for
aspects of predator–prey interactions that lack feral cats is currently being undertaken by the
reliable information. Therefore, managers charged Arthur Rylah Research Institute for Environmental
with controlling these species for agricultural and Research.
environmental benefits must make decisions based
on unreliable knowledge and are faced with the
uncertainty that their actions may not reliably result 2. The impact of predator control
in the desired outcome. operations on the population dynamics
In this section we highlight the key areas where and social organisation of sympatric
further investigation would improve our predators and the impacts on native
understanding of the impacts feral cats, foxes and species and communities.
rabbits, and the control of combinations of these
species have on native species. While there is some evidence that both fox and cat
abundance is related to rabbit abundance in arid
and semi-arid areas, we have little information on
7.1 Priorities in current gaps in our the potential for feral cats to increase in
understanding abundance or to alter their foraging behaviour
We believe that all of the knowledge gaps following the removal/reduction of a higher order
presented below are vital to improving our predator. We also know little about the impact of
understanding of the interactions between feral changes in predator communities on the
cats, foxes, rabbits and native species. However, persistence of native species. However, the
we have prioritised these gaps in order of limitations of techniques for estimating predator
importance in regard to the potential level of abundance, particularly feral cats, restrict our
benefit for managers. ability to identify if feral cats do increase in
abundance and/or impact on native wildlife.
We provide basic design scenarios and
suggestions for the general location where these
investigations might be conducted. This should not 3. The role of rabbits in temperate
be viewed as an exhaustive description of
experimental design, but rather a guide.
systems in supporting elevated
numbers of foxes and feral cats.
At present very few control operations concurrently
target feral cats, foxes and rabbits. There is no The little data available suggests fox population
information that we are aware of on the costs and dynamics may not be related to rabbit abundance
benefits of integrated control. Before this type of in temperate systems in Australia. There is no
analysis can be properly undertaken the identified evidence either way for feral cats. Investigations
gaps in our current level of understanding need to into these interactions and the effects on native
be addressed. species will require studies covering a wide range
of foods as rabbits are unlikely to be the primary
Set out below are the four key areas that we prey of these predators in these areas.
consider need the most urgent attention. Following

Interactions between feral cats, foxes, native carnivores and rabbits 53

4. The effects of disease (RHD and mortality as a function of changes in native prey
myxomatosis), particularly in temperate abundance. Quantification of kill rates for native
prey (and rabbits) in relation to the abundance of
environments, on the interactions between all food sources in an area and in relation to the
predators and their prey. abundance of predators.
Much of the information we have on the
interactions between feral cats, foxes and rabbits
has been gained either prior to or at the time of the • What are the potential costs of releasing
arrival of RHD. We have little information on the rabbits from predator induced regulation, both
effects of combinations of RHD and predation, and ecologically and economically?
the flow-on effects to native prey species. The damage rabbits cause to flora and soils is well
documented, but the flow-on effects to native
7.1.1 Further information requirements fauna species are less well understood. We
The following gaps in the knowledge are in some currently have little reliable information on the
instances a subset of the broader areas of potential impacts of increases in rabbit abundance
investigation highlighted above; many arise as a result of a release in regulation.
through the model development process described
in Section 5. We provide these as a guide to a
more targeted research approach that will help • What are the interactions between dingoes,
parameterise the models and reduce the gaps in foxes, feral cats and quoll species, and how do
our understanding. changes in rabbit abundance affect these?
We currently know very little about the impact of
introduced predators and predator control on the
• What factors regulate populations of native long-term persistence of populations of native
species? predators, in particular quoll species, or how
There is currently no information on how native changes in rabbit abundance might influence these
species respond to changes in food supply in the interactions. To fill this gap requires experimental
absence of predation, or how social factors may evidence of the changes in rates of increase (or
influence population dynamics. This information is relevant demographic parameters) of one predator
needed to properly develop models for the in relation to manipulation in the density of the
population dynamics of native prey in the absence other predator and / or their primary prey.
of predation. Assessing demographic parameters would allow a
This will require identification of key resources, and better understanding of the mechanism of
quantification of rates of increase in relation to the interaction.
availability of these resources. This type of study • Quantification of rates of increase of foxes and
should fit within broad scale predator removal feral cats (or relevant demographic parameters
undertaken for management as a more detailed such as reproduction and survival) in relation
exploration of the response of threatened prey to to availability of resources including rabbits (or
predator removal. juvenile rabbits), native prey and other foods.
• What factors regulate predator populations? Rates of increase should be divided into a
It is possible that there are density dependent recruitment phase and a decline phase. This
regulatory mechanisms that act to regulate research will require better ways of quantifying the
populations of both feral cats and foxes. There is abundance of the predators and their prey, and/or
some evidence for this for foxes in semi-arid techniques to assess survival and reproduction.
systems in South Australia, but nothing is known of The use of DNA technology to estimate population
this for feral cats. abundance is one example of a method that may
improve monitoring techniques.

• What is the functional response of foxes and

feral cats to changes in the abundance of
native species at risk from predation?
While these predators have been implicated in the
decline and extinction of a range of native species,
many still persist. To properly understand the
impact of feral cats and foxes on native species we
need to quantify the rate of predator-induced

Interactions between feral cats, foxes, native carnivores and rabbits 54

8 Filling the gaps

To gain reliable knowledge on the most effective However, we do provide some guidelines as to the
and efficient combination of pest management general design features for studies to fill the gaps
strategies that would bring about gains in in our knowledge.
biodiversity, we could apply the basic principals of
Community and Institutional Support
investigative science. That is, undertake large-
scale manipulative experiments that are replicated, It is likely that community engagement will be
randomised and controlled. These three tenets of essential, as control operations will probably be
the scientific method underpin the acquisition of undertaken across tenure due to the distribution of
reliable knowledge. However, large-scale the species and the scale at which experiments are
manipulative experiments are difficult to implement likely to be undertaken. Accordingly, the
in the field and require long-term support and implementation and long-term success of large-
investment from management agencies. Financial scale projects will in part rely on community
and logistical constraints have the effect of support. Institutional support is also vital, providing
reducing the temporal scale over which studies are the resources to implement projects.
able to operate. This is a major limitation as
It is important that in the development stage of
patterns and processes often take many years to
emerge. It is often the case that large these projects the expectations of the timing and
management-scale experiments must sacrifice one magnitude of results are kept realistic. Failure to
deliver on unrealistic expectations can lead to a
or more of the components of experimental design.
withdrawal of community and institutional support,
Our ability to impose reductions of specified levels resulting in premature cessation of projects.
on feral cats and foxes is limited. The techniques
available are essentially blunt instruments, able Legislative Requirements
only to impose changes at a coarse level. This A vital component to any experiment is the use of
limits our capacity to fine-tune our understanding of non-treatment sites. However, the control of pest
many of the interactions between management animals is often legislated and is obligatory for
actions (control), predators, their prey and native State Government and other land managers. Not
species. It is also difficult to plan for outbreaks of controlling pest species may contravene local
disease such as myxomatosis or RHD that may legislation.
confound experimental manipulations.
Scale
These limitations can have the effect of restricting
the generality of the outcomes or reducing the It is important that manipulative experiments are
strength of the inferences that can be drawn. conducted at the appropriate scale. For example,
However in some instances manipulative for an experiment investigating the impact of
experimentation, at the scale of management or at changes in rabbit abundance on fox population
least at the scale of the predators, is the only way dynamics, it would be necessary to have at least
to improve our knowledge. two experimental units and one control unit that
each encompasses the home range of several fox
A combination of management scale groups. Similarly, experiments on competition
experimentation and smaller scale research between rabbits and medium-sized native
targeted at specific questions, such as those listed herbivores would be at the scale of the herbivores.
in the previous section, will lead to increased levels
of reliable knowledge that can also be used to The scale of the monitoring program also needs to
optimise the modelling processes. Based on be appropriate for the species investigated. For
improvements in knowledge, management actions example, spotlight transects that monitor changes
may then be altered to optimise benefits of pest in rabbit abundance are typically shorter (< 5 km)
control. Continual updating of system models than those required to assess changes in fox
decreases the amount of time it takes to improve abundance (> 10 km).
the reliability of management decisions. Duration
Experimental Design One of the major limitations on many of the studies
This review does not provide detailed designs for to date has been the limited duration of the
each identified gap in current knowledge as this experiments. If the experiment was investigating
would require knowledge of site specific features, changes in survival rates, the study needs to be
history of control, infrastructure, limitations on able to account for the natural variation in these
access, community and State Government rates. This may take several years.
support, and legislative requirements.

Interactions between feral cats, foxes, native carnivores and rabbits 55

Experimental Control areas set aside as non-treatment sites
(6 x 20 000 ha). In the north-west of Victoria there
The use of non-treatment (experimental controls)
are several large (> 100 000 ha) parks that also
sites is often the best approach and a basic feature
have the potential to undertake management scale
of experimental design. These areas are often
research in a semi-arid setting and which all
difficult to incorporate at the scale of management,
contain feral cats, foxes and rabbits. Victoria also
or managers are very reluctant to ‘do nothing’. It is
has a number of fox free islands that have
important that pre-manipulation assessments of
established feral cat and rabbit populations.
both treatment and non-treatment sites are
undertaken to ensure that the sites are comparable The New South Wales fox threat abatement plan is
and that underlying differences are not influencing being implemented across a wide rage of habitat
species abundance prior to beginning the study. types specifically aimed at protecting a suite of
Pre-manipulation assessment may need to native species.
account for seasonal or yearly variations in
Also in NSW is Lake Burrendong, a temperate
population demography.
grazing area that has the advantage of having had
It may not be possible to find areas that have not research undertaken in the past on feral cats,
had some management intervention; however, it foxes and rabbits, although Lake Burrendong is no
may be possible to cease management in some longer an active research site.
areas and compare changes in the system to
In South Australia, project Bounce Back and the
areas where control is maintained.
Arid Recovery Program operate in semi-arid and
Randomisation arid environments, and have established
infrastructure.
In order to make statements about the generality of
the outcomes from experiments, treatments need In Western Australia there are a number of large-
to be allocated randomly. This is to avoid scale predator control projects that have been
underlying patterns and bias. In large-scale underway for a number of years. These projects
experiments this is not always possible. Without focus on the control of foxes, and to a lesser
randomisation results from large-scale experiments extent, feral cats, and cover a wide range of
are specific to that area in which they are biogeographic areas from temperate forest to
conducted. It is not possible to transfer the semi-arid coast.
knowledge gained in one area to another.
All these areas have a range of species that are
Location considered threatened by predation or potential
competition from rabbits.
There are a number of large-scale management
programs currently underway across a range of It is recommended that at the completion of each
biogeographic regions in Australia. In Victoria, investigation the information be used to update
Southern Ark is a large-scale fox control program models of the systems and its predictions.
that operates over 1 million hectares of south-
eastern Victoria. While this program is focused on
fox control it offers an opportunity to incorporate
integrated control options, including substantial

Interactions between feral cats, foxes, native carnivores and rabbits 56

9 References
Abbott, I. (2002). Origin and spread of the cat, Felis catus, on mainland Australia, with a discussion of the
magnitude of its early impact on native fauna. Wildlife Research 29, 51-74.

Abrams, P. A. (2001). Describing and quantifying interspecific interactions: a commentary on recent

approaches. Oikos 94, 209-218.

Abrams, P. A., and Ginzburg, L. R. (2000). The nature of predation: prey dependent, ratio dependent or
neither? Trends in Ecology & Evolution 15, 337-341.

Algar, D., and Smith, R. (1998). Approaching Eden. Landscape 13, 28-34.

Algar, D., and Sinagra, J. A. (1996). ‘Methods of broadscale control of feral cats in Western Australia. Feral
Pests Program, Project 11’. (Department of Conservation and Land Management: Western Australia).

Algar, D., Angus, G. J., and Sinagra, J. A. (1999). ‘Preliminary assessment of a trapping technique to
measure feral cat abundance. Final Report to Environment Australia’. (Department of Conservation
and Land management: Western Australia).

Algar, D. A., Burbidge, A. A., and Angus, G. J. (2002). Cat eradication on Hermite Island, Montebello Islands,
Western Australia. In ‘Turning the tide: the eradication of invasive specks’. (Eds C. R. Veitch and M. N.
Clout.). pp. 118 (World Conservation Union, Gland: Switzerland).

Alonzo, S. H. (2002). State-dependent habitat selection games between predators and prey: the importance
of behavioural interactions and expected lifetime reproductive success. Evolutionary Ecology
Research 4, 759-778.

Arthur, A. D. (2001). Effects of predation and habitat structure on the population dynamics of prey:
experiments in a model, field based system. Ph.D. Thesis, University of Sydney.

Arthur, A. D., and Pech, R. P. (2003). The non-lethal impacts of predation on mouse behaviour and
reproduction – implications for pest population dynamics. In ‘Rats, mice and people: rodent biology
and management’. (Eds G. R. Singleton, L. A. Hinds, C. M. Krebs, and Spratt, D. M.) pp 329-333
(ACIAR Monograph No.96).

Arthur, T., Pech, R., Drew, A., Gifford, E., Henry, S., and McKeown, A. (2003). The effect of increased
ground-level habitat complexity on mouse population dynamics. Wildlife Research 30, 565-572.

Atkinson, I.A.E. (1985). The spread of commensal species of Rattus to oceanic islands and their effect on
island avifaunas. In ‘Conservation of Island Birds’ (Ed. P. J. Moors). pp. 35-81. (ICBP Technical
Publication no. 3).

Baker-Gabb, D. J. (1984). The feeding ecology and behaviour of seven species of raptor over-wintering in
coastal Victoria. Australian Wildlife Research 11, 517–532.

Baldwin, J. A. (1980). The domestic cat, Felis catus L., in the Pacific Islands. Carnivore Genetics Newsletter
4, 57-66.

Banks, P. B. (2000). Can foxes regulate rabbit populations? Journal of Wildlife Management 64, 401-406.

Banks, P. B., Dickman, C. R., and Newsome, A. E. (1998). Ecological costs of feral predator control: foxes
and rabbits. Journal of Wildlife Management 62, 766–772.

Banks, P. B., Newsome, A. E., Dickman, C. R. (2000). Predation by red foxes limits recruitment in
populations of eastern grey kangaroos. Austral Ecology 25, 283-291.

Interactions between feral cats, foxes, native carnivores and rabbits 57

Bayliss, P. (1987). Kangaroo dynamics. in Kangaroos: their Ecology and Management in the Sheep
Rangelands of Australia. (Eds. G. Caughley, N. Shepherd and J. Short) pp119-134 (Cambridge
University Press: Cambridge).

Bayliss, P., and Choquenot, D. (2002). The numerical response: rate of increase and food limitation in
herbivores and predators. Philosophical Transactions of the Royal Society of London Series B-
Biological Sciences 357, 1233-1248.

Bayly, C. P. (1978). A comparison of the diets of the red fox and the feral cat in an arid environment. South
Australian Naturalist 53, 20-28.

Begg, R. J. (1983). Northern Quoll. In: ‘The Australian Museum Complete Book of Australian Mammals’. (Ed.
R. Strahan.) p. 23. (Angus and Robertson: Sydney).

Belcher, C. A. (1995). Diet of the Tiger QuoII (Dasyurus maculatus) in East Gippsland, Victoria. Wildlife
Research 22, 341-57.

Blackball, S. (1980). Diet of the eastern native-cat, Dasyurus viverrinus (Shaw), in southern Tasmania.
Australian Wildlife Research 7, 191-198.

Baysher, M. (1993). ‘Managing Vertebrate Pests: Principles and Strategies’. (Australian Government
Publishing Service: Canberra).

Brooker, M. G. (1977). Some notes on the mammalian fauna of the western Nullarbor plain, Western
Australia. Western Australian Naturalist 14, 2-15.

Brown, J. S. (1988). Patch use as an indicator of habitat preference, predation risk, and competition.
Behavioural Ecology and Sociobiology 22, 37-47.

Burbidge, A. A., and McKenzie, N. L. (1989). Patterns in the modern decline of Western Australia's
vertebrate fauna: causes and conservation implications. Biological Conservation 50, 143-198.

Burbidge A. A., and Manly, B. J. F. (2002). Mammal extinctions on Australian islands: causes and
conservation implications. Journal of biogeography. 29, 465-473.

Cappuccino, N. and Harrison, S. (1996). Density-perturbation experiments for understanding population

regulation. In ‘Frontiers of population ecology.’ (Eds R. B. Floyd A. W. Sheppard and P. J. De Barro)
pp 53–64 (CSIRO Publishing: Melbourne).

Catling, P. C. (1988). Similarities and contrasts in the diets of foxes, Vulpes vulpes, and cats. Felis catus,
relative to fluctuating prey populations and drought. Australian Wildlife Research 15, 307-317.

Catling, P. C., and Burt, R. J. (1995). Why are red foxes absent from some eucalypt forests in eastern New
South Wales? Wildlife Research 22, 535-546.

Catling, B., and Reid, A. M. (2003). ‘Predator and critical weight range species 5. Results of spring 2002 and
autumn 2003 surveys’. (CSIRO Sustainable Ecosystems: Canberra).

Caughley, G .(1977). ‘Analysis of Vertebrate Populations”. (John Wiley and Sons, Chichester: United
Kingdom).

Caughley, G. (1987). Ecological relationships. In ‘Kangaroos: their Ecology and Management in the Sheep
Rangelands of Australia’. (Eds. G. Caughley N. Shepherd and J. Short) pp 159-187 (Cambridge
University Press: Cambridge).

Caughley, G., and Krebs, C. J. (1983). Are big mammals simply little mammals writ large? Oecologia 59, 7-
17.

Interactions between feral cats, foxes, native carnivores and rabbits 58

Cavanagh, F. A. (1998). ‘The effects of Rabbit Calicivirus on the Diet of Feral Cats and Foxes in North-West
Victoria: A Progress Report’. (Department of Natural Resources and Environment: Melbourne).

Chase, J. M. (2003). Strong and weak trophic cascades along a productivity gradient. Oikos 101, 187-195.

Choquenot, D. (1998). Testing the relative influence of intrinsic and extrinsic variation in food availability on
feral pig populations in Australia's rangelands. Journal of Animal Ecology 67, 887-907.

Choquenot, D. (1992). The outsiders: competition between introduced herbivores and domestic stock in
rangeland grazing systems. Australian Rangelands in a Changing Environment, 7th Biennial
Conference of the Australian Rangeland Society. Australian Rangeland Society.

Christensen, P. E. S. (1980). The biology of Bettongia penicillata Gray, 1837, and Macropus eugenii
(Desmarest, 1817) in relation to fire. Forest Department of Western Australia Bulletin 91, 1-90.

Christensen, P. E. S. and Burrows, N. D. (1995). Project Desert Dreaming: the reintroduction of mammals to
the Gibson Desert. In ‘Reintroduction Biology of Australian and New Zealand Fauna’. (Ed. M. Serena)
pp. 199-208 (Surrey Beatty and Sons: Chipping Norton).

Coman, B. J. (1973). The diet of red foxes, Vulpes vulpes L, in Victoria. Australian Journal of Zoology 21,
391-401.

Coman, B. J. (1983). The fox. In ‘Complete Book of Australian Mammals’. (Ed. R. Strahan), pp 486-487.
(Angus and Robertson: Sydney).

Coman, B. J., and Brunner, H. (1972). Food habits of the feral house cat in Victoria. Journal of Wildlife
Management 36, 848-853.

Corbett, L. (1995). ‘The Dingo in Australia and Asia’. (University of New South Wale Press Ltd: Sydney).

Crooks, K. R., and Soule, M. E. (1999). Mesopredator release and avifauna extinctions in fragmented
systems. Nature 5, 563-566.

Davey, C., Sinclair, A. R. E., Pech, R. P., Arthur, A. D., Newsome, A. E., Hik, D., Molsher, R., and Allcock, K.
(In Prep). Do exotic vertebrates structure the biota of Australia? An experimental test in New South
Wales.

Davis, S. A., Pech, R. P. and Catchpole, E. A. (2003) Populations in variable environments: the effect of
variability in a species’ primary resource. In ‘Wildlife population growth rates’. (Eds. R.M. Sibly, J.
Hone and T.H. Clutton-Brock) pp. 180-197 (Cambridge University Press: Cambridge).

Dawson, T. J., and Ellis, B. A. (1979). Comparison of the diets of Yellow-footed Rock-wallabies and
sympatric herbivores in western New South Wales. Australian Wildlife Research 6, 245-254.

Dawson, T. J., and Ellis, B. A. (1994). Diets of mammalian herbivores in Australian arid shrublands: seasonal
effects on overlap between kangaroos, sheep and rabbits and on dietary niche breadths and
electivities. Journal of Arid Environments 26, 257-271.

de Torres, P. (1999). Control and Ecology of the Red Fox in Western Australia – prey response to 1080
baiting over larger areas. CALMScience, Wildlife Research Centre, Department of Conservation and
Land Management, Western Australia.

Dekker, D. (1986). Wolf (Canis lupus) numbers and colour phases in Jasper National Park, Alberta: 1965-
1984. Canadian Field Naturalist 100, 550-553.

Dickman, C. (1996). ‘Overview of the Impacts of Feral Cats on Australian Native Fauna”. (Australian Nature
Conservation Agency: Canberra).

Interactions between feral cats, foxes, native carnivores and rabbits 59

Dickman, C. R., Pressey, R. L., Lim, L. Parnaby, H. E. (1993). Mammals of particular conservation concern
in the Western Division of New South Wales. Biological Conservation 65, 219-48.

Edwards, G. P., Dobbie, W., and Berman, D. McK. (2002a). Population trends in European rabbits and other
wildlife of central Australia in the wake of rabbit haemorrhagic disease. Wildlife Research 29, 557-565

Edwards, G. P., Dobbie, W., and Berman, D. McK. (2002b). Warren ripping: its impacts on European rabbits
and other wildlife of central Australia amid the establishment of rabbit haemorrhagic disease. Wildlife
Research 29, 567-575.

Estes, J. A. (1996). Predators and ecosystem management. Wildlife Society Bulletin 24, 390-396.

Finlayson, H. H. (1961). On Central Australian mammals. Part IV. The distribution and status of central
Australian species. Records of the South Australian Museum 14,141-191.

Fleming, P., Corbett, L, Harden, R. and Thomson, P. (2001). ‘Managing the Impacts of Dingoes and Other
Wild Dogs’. (Bureau of Rural Sciences: Canberra).

Friend, J.A. (1990). The numbat Myrmecobius fasciatus (Myrmecobiidae): history of decline and potential for
recovery. Proceedings of the Ecological Society of Australia 16, 369-377.

Friend, J. A. and Scanlon, M. D. (1996). ‘Assessment of the effect of fox control on populations of the red-
tailed phascogale. Phase 4: final report’. ANCA Feral Pests Program Project 18. (Department of
Conservation and Land Management: Western Australia).

Fuller, B. (1970). ‘The Nullarbor story’. (Rigby: Adelaide).

Gibb, J. A., and Williams, J. M. (1990). ‘European rabbit’. In ‘The handbook of New Zealand mammals’. (Ed.
C. M. King) (Oxford University Press: Auckland).

Gibson, D. F., Lundie-Jenkins, G., Langford. D. G., Cole, J. R, Clarke. D. E., and Johnson, K. A. (1994).
Predation by feral cats, Felis catus, on the rufous hare-wallaby, Lagorchestes hirsutus, in the Tanami
Desert. Australian Mammalogy 17, 103-7.

Godsell, J. (1982). Aspects of the population ecology of the eastern quoll, Dasyurus viverrinus (Dasyuridae,
Marsupialia) in southern Tasmania. In ‘Carnivorous Marsupials’. (Ed. M. Archer) pp. 199-207 (Royal
Zoological Society of New South Wales: Sydney).

Green, K. and Osborne W. S. (1981). The Diet of Foxes, Vulpes vulpes (L.) in relation to abundance of prey
above the winter snowline in New South Wales. Australian Wildlife Research 8, 349-60.

Greentree, C., Saunders, G., Mcleod, L., and Hone, J. (2000). Lamb predation and fox control in south-
eastern Australia. Journal of Applied Ecology 37, 935 – 943.

Harper, G. A., and Dobbins, M. (2002). Control of cats on mountain “islands”, Stewart Island, New Zealand In
‘Turning the tide: the eradication of invasive specks’. (Eds. C. R. Veitch and M. N. Clout) pp. 408
(World Conservation Union, Gland: Switzerland).

Hayward, M. W., de Tores, P. J., Dillon, M. J., and Fox, B. (2003). Local population structure of a naturally
occurring metapopulation of the quokka (Setonix brachyurus Macropodidae: Marsupialia). Biological
Conservation 110, 343 – 355.

Henke, S.E. and Bryant. F.C. (1999). Effects of coyote removal on the faunal community in western Texas.
Journal of Wildlife Management 63,1066-1081.

Holden, C. and Mutze, G. (2002). Impact of rabbit haemorrhagic disease on introduced predators in the
Flinders Ranges, South Australia. Wildlife Research 29, 615-626.

Interactions between feral cats, foxes, native carnivores and rabbits 60

Holling, C. S. (1959). The components of predation as revealed by a study of small mammal predation of the
European pine sawfly. The Canadian Entomologist 91, 293–320.

Holling, C. S. (1965). The functional response of predators to prey density and its role in mimicry and
population regulation. Memoirs of the Entomological Society of Canada 45, 5–60.

Holt, R. D., (1977). Predation, apparent competition, and the structure of prey communities. Theoretical
Population Biology 12, 197-229.

Hone, J. (1994). ‘Analysis of Vertebrate Pest Control’. (Cambridge University Press: Cambridge).

Hone, J. (1999). Fox control and rock-wallaby population dynamics – assumptions and hypotheses. Wildlife
Research 26, 671-673.

Hughes, J. J., Ward, D., and Perrin, M. R., (1994). Predation risk and competition affect habitat selection and
activity of Namibia Desert gerbils. Ecology 75, 1397-1405.

Jarman, P. (1986). The red fox – an exotic, large predator. In ‘The Ecology of Exotic Animals and Plants,
Some Australian Case Studies’. (Ed. R. L. Kitching) pp. 45-61 (Wiley and Sons: Brisbane).

Johnson, K. A., and Roff. A. D. (1982). The western quoll, Dasyurus geoffroii (Dasyuridae, Marsupialia) in
the Northern Territory: historical records from venerable sources. In ‘Carnivorous Marsupials’. (Ed. M.
Archer). pp. 209-20. (Royal Zoological Society of New South Wales: Sydney).

Jones, E. (1977). Ecology of the feral cat, Felis catus (L.), (Carnivora: Felidae) on Macquarie Island.
Australian Wildlife Research 4, 249-262.

Jones, E. (1983). Feral Cat. In ‘Complete Book of Australian Mammals’. (Ed. R. Strahan). p. 489 (Angus and
Robertson: Sydney).

Jones, E., and Coman, B. J. (1981). Ecology of the feral cat, Felis catus (L.), in south-eastern Australia. I.
Diet. Australian Wildlife Research 8, 537-547.

King, D., and Green, B. (1999). ‘Goannas. The Biology of Varanid Lizards’. (University of NSW Press:
Sydney).

Kinnear, J. E., Onus, M.L., and Bromilow, R. N. (1988) Fox control and rock-wallaby population dynamics.
Australian Wildlife Research 15, 435-450.

Kinnear, J. E., Onus, M. L., and Sumner, N. R. (1998). Fox control and rock-wallaby population dynamics –
II. An update. Wildlife Research 25, 81-88.

Kinnear, J. E., Sumner, N.R., and Onus, M. L. (2002) The red fox in Australia-an exotic predator turned
biocontrol agent. Biological Conservation 108, 335-359.

Korpimaki, E., and Krebs, C. J. (1996). Predation and population cycles of small mammals. Bioscience 46,
754–764.

Krebs, C. J. (1999). ‘Ecological Methods’. (Addison-Wesley Educational Publishers: California).

Krebs, C. J., Boutin, S., and Boonstra, R. (2001). ‘Ecosystem dynamics of the Boreal Forest: the Kluane
Project’. (Oxford University Press: New York).

Liberg, O. (1984). Food habits and prey impact by feral and house-based domestic cats in a rural; area in
southern Sweden. Journal of Mammalogy 65, 424-432.

Lindstrom, E. R., Brainerd, S. M., Helldin, J. O. and Overskaug, K. (1995). Pine marten- red fox interactions:
a case of intraguild predation? Annales Zoologici Fennici 32, 123-130.

Interactions between feral cats, foxes, native carnivores and rabbits 61

Litvaitis, J. A. and Harrison, D. J. (1989). Bobcat-coyote niche relationships during a period of coyote
population increase. Canadian Journal of Zoology 67, 1180-1188.

Lugton, I. (1991). Dietary and feeding studies of Australian foxes and their relevance to lamb predation.
Australian Vertebrate Pest Conference, pp 181-186.

Lunney, D., Trigg, B., Eby, P., and Ashby, E. (1990). Analysis of scats of dogs, Canis familiars, and foxes,
Vulpes vulpes (Canidae Camivora), in coastal Bega, New South Wales. Australian Wildlife Research
17, 61-68.

Mallinson, D., Olsen, P., and Olsen, J. (1990). A note on the breeding of the Little Eagle (Hieraaetus
morphnoides) and other raptors in the Mt Mugga area, ACT. Corella 14, 88-91.

Mitchell, B. (2003). Interactions between wild dogs and foxes in the Greater Blue Mountains region. Honours
Thesis, University of NSW, NSW.

Mitchell, N., Haeffner, R., Veer, V., Fulford-Gardner, U., Clerveaux, W., Veitch, C. R., and Mitchell, G.
(2002). Cat eradication and the restoration of endangered iguanas (Cyclura carinata) on Long Cay,
Caicos Bank, Turks and Caicos Islands, British West Indies. In ‘Turning the tide: the eradication of
invasive specks’. (Eds. C. R. Veitch and M. N. Clout.) pp. 206 (World Conservation Union, Gland:
Switzerland).

Molsher, R. (1999). The ecology of feral cats, Felis catus, in open forest in New South Wales: Interactions
with food resources and foxes. Ph.D. Thesis, (University of Sydney: NSW).

Molsher, R., Newsome, A., and Dickman, C. (1999). Feeding ecology and population dynamics of the feral
cat (Felis catus) in relation to the availability of prey in central-eastern New South Wales. Wildlife
Research 26, 593 – 607.

Morris, K., Johnson, B., Orell, P., Gaikhonf, G., Wayne, A., and Mora, D. (2003). Recovery of the threatened
Chuditch (Dasyurus geoffroii): a case study. In ‘Predators with Pouches: The Biology of Carnivorous
Marsupials’. (Eds. M. Jones, C. Dickman and M. Archer). (CSIRO Publishing, Collingwood: Victoria).

Morris, K., Orell. P., Brazell, R. (1995). The Effect of Fox Control on Native Mammals in the Jarrah Forest,
Western Australia. 10th Australian Vertebrate Pest Control Conference, Hobart, Tasmania.

Moseby, K. (2002). ‘The Arid Recovery Project. Annual Report 2002’. (Department of Environment and
Heritage: Adelaide).

Murdoch, W. W. (1969). Switching in general predators: experiments of predator specificity and stability of
prey populations. Ecological Monographs 39, 335-354.

Murdoch, W., and Oaten, A. (1975). Predation and population control. Advances in Ecological Research 9,
2-130.

Mutze, H., Bird, P., Kovaliski, J., Peacock, D., Jennings, S. and Cooke, B. (2002). Emerging epidemiological
patterns in rabbit haemorrhagic disease, its interaction with myxomatosis, and their effects on rabbit
populations in South Australia. Wildlife Research 29, 577-590.

Myers, K., and Parker, B. S. (1975a). Effect of severe drought on rabbit numbers and distribution in a refuge
area in semi-arid north-western New South Wales. Australian Wildlife Research 2, 103-120.

Myers, K., and Parker, B. S. (1975b). A study of the biology of the wild rabbit in climatically different regions
in eastern Australia. VI. Changes in numbers and distribution related to climate and land systems in
semi-arid north-western New South Wales. Australian Wildlife Research 2, 11-32.

Navarrete, S. A., and Castilla, J. C. (2003). Experimental determination of predation intensity in an intertidal
predator guild: dominant versus subordinate prey. Oikos 100, 251-262.

Interactions between feral cats, foxes, native carnivores and rabbits 62

Neave, H. M. (1999). Rabbit Calicivirus Disease Program Report 1: Overview of Effects on Australian Wild
Rabbit Populations and Implications for Agriculture and Biodiversity. A report of research conducted by
participants of the Rabbit Calicivirus Disease Monitoring and Surveillance Program. (Bureau of Rural
Sciences: Canberra).

Newsome, A. E., Parer, I., and Catling, P. C. (1989). Prolonged prey suppression by carnivores-predator
removal experiments. Oecologia 78, 458-467.

Newsome, A., Pech, R., Smyth, R., Banks, P., Dickman, C., (1997). ‘Potential impacts on the Australian
native fauna of rabbit calicivirus disease’. (Environment Australia: Canberra).

Newsome, A. E., and Sinclair, A. R. E. (1995). Testing new theory in predator-prey relationships. Wildlife
Research 22, 1-3.

NSW National Parks and Wildlife Service (2001). ‘Threat Abatement Plan for Predation by the Red Fox
(Vulpes vulpes)’. (NSW National Parks and Wildlife Service: Hurstville).

Nogales, M., Martin, A., Tershy, B. R., Donlan, J. C., Veitch, D., Puerta, N., Wood, B., and Alonso, J. (2004).
A review of feral cat eradication on islands. Conservation Biology 18, 310-319.

Oakwood, M. (2000). Reproduction and demography of the northern quoll, Dasyurus hallucatus, in the
lowland savanna of northern Australia. Australian Journal of Zoology 48, 519-539.

Osborne, W., Preece, M., Green, K., and Green, M. (1978). Gungartan: A winter fauna survey above 1,500
m. Victorian Naturalist 95, 226-35.

Palomares, F., Gaona, P., Ferreras, P., and Delibes, M. (1995). Positive effects on game species of top
predators by controlling smaller predator populations: An example with Lynx, Mongoose, and Rabbits.
Conservation Biology 9, 295–305.

Paltridge, R., Gibson, D., and Edwards, G. (1997). Diet of the feral cat (Felis catus) in central Australia.
Wildlife Research 24, 67-76.

Paltridge, R. (2002). The diets of cats, foxes and dingoes in relation to prey availability in the Tanami Desert,
Northern Territory. Wildlife Research 29, 389-403.

Pech, R. P., Sinclair, A. R. E., Newsome, A. E., and Catling, P. C. (1992). Limits to predator regulation of
rabbits in Australia: evidence from predator-removal experiments. Oecologia 89, 102–112.

Pech, R. P., Sinclair, A. R. E., and Newsome, A. E. (1995). Predation models for primary and secondary prey
species. Wildlife Research 22, 55–64.

Pech, R., Hood, G. M., McIlroy, J., and Saunders, G. (1997). Can foxes be controlled by reducing their
fertility. Reproduction, Fertility and Development 9, 41–50.

Pech, R. P., and Hood, G. R. (1998). Foxes, rabbits, alternative prey and rabbit calicivirus disease -
consequences of a new biological control agent for an outbreaking species in Australia. Journal of
Applied Ecology 35, 434-453.

Pettigrew, J. D. (1993). A burst of feral cats in the Diamantina – a lesson for the management of pest
species? In: ‘Cat management workshop proceedings’ (Eds. G. Siepen and C. Owens.) (Queensland
Department of Environment and Heritage: Brisbane).

Quinn, J. L., and Cresswell, W. (2004). Predator hunting behaviour and prey vulnerability. Journal of Animal
Ecology 73,143-154.

Rauzon, M. J., Forsell, D. J., and Flint, E. N. (2002). Seabird re-colonisation after cat eradication on
equatorial Jarvis, Howland, and Baker Islands, USA, Central Pacific. Zealand. In ‘Turning the tide: the

Interactions between feral cats, foxes, native carnivores and rabbits 63

 eradication of invasive specks’. (Eds. C. R. Veitch and M. N. Clout) pp. 411 (World Conservation
Union, Gland: Switzerland).

Read, J., and Bowen, Z. (2001). Population dynamics, diet and aspects of the biology of feral cats and foxes
in arid South Australia Wildlife Research 28, 195-203.

Reddiex, B. (2004). ‘Effects of predation and rabbit haemorrhagic disease on rabbit population dynamics in
New Zealand’. Ph.D. Thesis, (Lincoln University: New Zealand).

Reddiex, B., Forsyth, D. M., McDonald-Madden, E., Einoder, L. D., Griffioen, P. A., Chick, R. R. and Robley,
A. J. (2004). ‘Review of existing feral goat, feral cat, feral rabbit, feral pig, fox, and wild dog control in
Australia’. (Arthur Rylah Institute for Environmental Research, Department of Sustainability and
Environment: Melbourne).

Richards, J. D. and Short, J. (2003). Reintroduction and establishment of the western barred bandicoot
Perameles bougainville (Marsupialia: Peramelidae) at Shark Bay, Western Australia. Biological
Conservation 109, 181-195.

Ridpath, M. G., and Brooker, M. G. (1986). The breeding of the Wedge-tailed Eagle, Aquila audax in relation
to its food supply. Ibis 128, 177-194.

Risbey, D. A., Calver, M. C, and Short, J. (1999). The impact of cats and foxes on the small vertebrate fauna
of Heirisson Prong, Western Australia. I. Exploring potential impact using diet analysis. Wildlife
Research 26, 621-630.

Risbey, D. (2000). ‘The impact of cats and foxes on small terrestrial vertebrates and the control of feral cats
at Heirisson Prong’. Ph.D. Thesis (Murdoch University: Western Australia).

Risbey, D. A., Calver, M. C., Short, J., Bradley, J. S., and Wright, I. W. (2000). The impact of cats and foxes
on the small vertebrate fauna of Heirisson Prong, Western Australia. II. A field experiment. Wildlife
Research 27, 223-235.

Robertshaw, J. D. and Harden, R. H. (1985). Ecology of the dingo in north-eastern New South Wales. II.
Diet. Australia Wildlife Research 12, 39-50.

Robertson, G. (1987). Plant dynamics. In ‘Kangaroos: their Ecology and Management in the Sheep
Rangelands of Australia’. (Eds. G. Caughley, N. Shepherd, and J. Short.) pp 50-67 (Cambridge
University Press: Cambridge).

Robley, A. (1999). ‘The comparative ecology of the burrowing bettong (Bettongia lesueur) and the European
rabbits (Oryctolagus cuniculus)’. Ph.D. Thesis (Murdoch University: Western Australia).

Robley, A., Short, J., and Bradley, S. (2001). Dietary overlap between the burrowing bettong (Bettongia
lesueur) and the European rabbits (Oryctolagus cuniculus) in semi-arid coastal Western Australia.
Wildlife Research 28, 341 – 349.

Robley, A. J., Short, J., and Bradley, S. (2002). Do European rabbits (Oryctolagus cuniculus) influence the
population ecology of the burrowing bettong (Bettongia lesueur)? Wildlife Research 29, 423-429.

Robley, A, and Wright, J (2003). ‘Adaptive Experimental Management of Foxes. Annual Report for Year 2
July 2002 – June 2003’. (Arthur Rylah Institute for Environmental Research, Department of
Sustainability and Environment: Melbourne).

Rogers, P. M., Arthur, C. P., and Soriguer, R. C. (1994). The rabbit in continental Europe. In ‘The European
rabbit: the history and biology of a successful colonizer’. (Eds. H. V. Thompson, and C. M. King)
(Oxford University Press: Oxford).

Rolls, E. G. (1969). ‘They all ran wild’. (Angus & Robertson: Sydney).

Interactions between feral cats, foxes, native carnivores and rabbits 64

Sandell, P. (1999). Assessing the Implications of RCD for the Management of Large Conservation Reserves,
Grazing leasehold and Dryland Agriculture in North-west Victoria. Final Report to Bureau of Rural
Science. National RCD Detailed Monitoring and Surveillance Program. (Parks Victoria and
Department of Natural Resources and Environment: Melbourne).

Sandell, P. R., and Start, A. N. (1999). Rabbit Calicivirus Disease Program Report 4: Implications for
Biodiversity in Australia. A report of research conducted by participants of the Rabbit Calicivirus
Disease Monitoring and Surveillance Program and Epidemiology Program, Prepared for the RCD
Management Group. (Bureau of Rural Sciences: Canberra).

Sargeant, A. B., Greenwood, R. J., Sovada, M. A. and Schaffer, T. L. (1993). Distribution and abundance of
predators that affect duck production: Prairie Pothole Region. U.S. Fish and Wildlife Service Resource
Publication, 194.

Saunders, G., Coman, B., Kinnear, J., and Braysher (1995). ‘Managing Vertebrate Pests: Foxes’. (Bureau of
Resource Sciences: Canberra).

Saunders, G., Choquenot, D., McIlroy, J., and Packwood, R. (1998). Initial effects of rabbit haemorrhagic
disease on free-living rabbit (Oryctolagus cuniculus) populations in central-western New South Wales.
Wildlife Research 26, 69–74.

Saunders, G.R., McIlroy, J.C., Berghout, M., Kay, B., Gifford, Perry, R., and van de Ven, R. 2002, The
effect of induced sterility on the territorial behaviour and survival of foxes. Journal of Applied Ecology
39, 56-66.

Saunders, G., Berghout, M., van de Ven, R., and Winstanley, R. (2004). The diet of foxes (Vulpes vulpes) in
south-eastern Australia and the potential effects of rabbit haemorrhagic disease Wildlife Research 31,
13-18.

Schlager, F. E. (1981). ‘The distribution, status and ecology of the rufous rat-kangaroo, Aepyprymnus
rufescens, in northern New South Wales’. Masters of Natural Resources, (University of New England:
Armidale).

Serena, M., Soderquist. T. R. and Morris, K. (1991). ‘The Chuditch’. Western Australian Wildlife Management
Program, No. 7. (Department of Conservation and Land Management: Perth)

Short, J. (1985) The functional response of kangaroos sheep and rabbits in an arid grazing system. Journal
of Applied Ecology 22, 435-47.

Short, J. (1987). Factors affecting food intake of rangelands herbivores. In ‘Kangaroos: their Ecology and
Management in the Sheep Rangelands of Australia’. (Eds. G. Caughley, N. Shepherd, and J. Short.)
pp 84-98. (Cambridge University Press: Cambridge).

Short, J. (1998). The extinction of rat-kangaroos (Marsupialia: Potoroidae) in New South Wales, Australia.
Biological Conservation 86, 365-377.

Short, J. and Smith, A. (1994). Mammal decline and recover in Australia. Journal of Mammalogy. 75, 288-
297.

Short, J., Bradshaw, S. D., Giles, J., Prince, R. I. T., and Wilson, G. (1992). The reintroduction of macropods
(Marsupialia: Macropodoidea) in Australia – a review. Biological Conservation 62, 189-204.

Short, J., Kinnear, J., and Robley, A. (2002). Surplus killing by introduced predators in Australia – evidence
for ineffective anti-predator adaptations in native prey species. Biological Conservation 103, 282-301.

Sinclair, A. R. E. (1989). Population regulation in animals. In: ‘Ecological Concepts: the Contribution of
Ecology to Understanding of the Natural World’. (Ed. J.M. Cherrett). pp. 197-241. (Blackwell Scientific
Publications: Oxford).

Interactions between feral cats, foxes, native carnivores and rabbits 65

Sinclair, A. R. E. (1996). Mammal populations: fluctuation, regulation, life history theory and their implications
for conservation. In ‘Frontiers of population ecology’. (Eds. R. B. Floyd, A. W. Sheppard, and P. J. De
Barro). pp. 127–154. (CSIRO Publishing: Melbourne).

Sinclair, A. R. E. (1997). Fertility control of mammal pests and the conservation of endangered marsupials.
Reproduction, Fertility and Development 9, 1-16.

Sinclair, A. R. E., and Krebs, C. J. (2003). Complex numerical responses to top-down and bottom-up
processes in vertebrate populations. Philosophical Transactions of The Royal Society London B 357,
1221–1232.

Sinclair, A. R. E., Olsen, P. D., and Redhead, T. D. (1990). Can predators regulate small mammal
populations? Evidence from house mouse outbreaks in Australia. Oikos 59, 382–392.

Sinclair, A. R. E., and Pech, R. P. (1996). Density dependence, stochasticity, compensation and predator
regulation. Oikos 75, 164–173.

Sinclair, A. R. E., Pech, R. P., Dickman, C. R., Hik, D., Mahon, P., and Newsome, A. E. (1998). Predicting
effects of predation on the conservation of endangered prey. Conservation Biology 12, 564-575.

Smith, A. P., and Quin, D. G., (1996). Patterns and causes of extinction and decline in Australian conilurine
rodents. Biological Conservation 77, 243-267.

Soderquist, T. R., and Serena, M. (1994). Dietary niche of the western quoll, Dasyurus geoffroii, in the jarrah
forest of Western Australia. Australian Mammalogy 17,133-136.

Solomon, M. E. (1949). The natural control of animal population. Journal of Applied Ecology 18, 1–35.

Soule, M.E., Bolger, D.T., Alberts, A.C., Wright, J., Sorice, M. and Hill, S. (1988). Reconstructed dynamics of
rapid extinctions of chaparral-requiring birds in urban habitat islands. Conservation Biology 2, 75-92.

Sovada, M. A., Sargeant, A. B., and Greir, J. W. (1995). Differential effects of coyotes and red foxes on duck
nest success. Journal of Wildlife Management 59, 1-9.

Stokes, V. L., Pech, R., Banks, P. B., and Arthur, A. (2004). Foraging behaviour and habitat use by
Antechinus flavipes and Sminthopsis murina (Marsupialia: Dasyuridae) in response to predation risk in
eucalypt woodland. Biological Conservation 117, 331-342.

Taylor, R. H., (1979). How the Macquarie Island parakeet became extinct. New Zealand Journal of Ecology
2, 42-45.

Thompson, P. C., and Shepherd, R. (1995). Return to Eden. Landscope 10, 21 – 25.

Thomson, P. C. (1992) The behavioural ecology of dingoes in north-western Australia. III. Hunting and
feeding behaviour, and diet. Wildlife Research 19, 531-542.

Thompson, H. V. (1994). The rabbit in Britain. In ‘The European rabbit: the history and biology of a
successful colonizer’. (Eds. H. V. Thompson, and C. M. King) pp. 64–107 (Oxford University Press:
Oxford).

Trout, R. C., and Tittensor, A. M. (1989). Can predators regulate wild rabbit Oryctolagus cuniculus population
density in England and Wales. Mammal Review 19, 153–173.

Walters, C. (1986). ‘Adaptive Management of Renewable Resources’. (MacMillan Publishing Company: New
York).

Wheeler, S. H. and King, D. R. (1985). The European rabbit in south-western Australia. II. Reproduction.
Australian Wildlife Research 12, 197-212.

Interactions between feral cats, foxes, native carnivores and rabbits 66

Williams, K., Parer, I., Coman, B., Burley, J., and Braysher, M. (1995). ‘Managing vertebrate pests: rabbits”.
(Australian Government Publishing Service: Canberra).

Winstanley, R. K., Buttemer, W. A., and Saunders, G. (2003). Field metabolic rate and body water turnover
of the red fox Vulpes vulpes in Australia. Mammal Review 33, 295-301.

Woinarski, J. C. Z. and Braithwaite, R. W. (1990). Conservation foci for Australian birds and mammals.
Search 21, 65-67.

Wood, D. H. (1980). The demography of a rabbit population in an arid region of New South Wales Australia.
Journal of Animal Ecology 49, 55-79.

Wood, B., Tershy, B. R., Hermosillo, M. A., Donlan, C. J., Sanchez, J. A., Keitt, B. S., Croll, D. A., Howald, G.
R., and Biavaschi, N. (2002). Removing cats from islands in north-west Mexico. In ‘Turning the tide:
the eradication of invasive specks’. (Eds. C. R. Veitch and M. N. Clout) pp. 206 (World Conservation
Union, Gland: Switzerland).

Interactions between feral cats, foxes, native carnivores and rabbits 67

10 Acknowledgments
This review would not have been possible without the collaboration of many researchers who allowed us
access to their experiences, knowledge and unpublished data. We are especially thankful to John Anderson
(CALM, WA); Peter Banks (University of NSW, NSW); James Dawson (NPWS, NSW); Paul de Torres
(CALM, WA); Glen Edwards (PWC, NT); Alistair Glen (University of Sydney, NSW); Chris Holden (NPWS,
SA); Paul Mahon (NPWS, NSW); Stephen McPhee (DSE, VIC); Keith Morris (CALM, WA); Greg Mutze
(Plant Animal Control Commission, SA); Jenny Nelson (DSE, VIC); Jacqui Richards (CSIRO-CSE, WA);
Daniel Risbey (CSIRO –CSE, WA); Peter Sandell (Parks Victoria, VIC); Jeff Short (CSIRO-CSE, WA); Peter
Thompson (Agriculture WA.
We would like to thank G. Saunders, J. Parkes, Susan Wright, Bruce Mitchell and D.McRae for reading
earlier drafts of this report.

Interactions between feral cats, foxes, native carnivores and rabbits 68

11 Appendices

Appendix 1. Predictive model of pest species interactions in Australia.

Interactions 1 and 2
The effect of climate on vegetation biomass and growth in Australia in semi-arid systems was quantified by
Robertson (1987). In Australia most published interactive models for semi-arid systems (Caughley 1987;
Choquenot 1998; Pech and Hood 1998) have used Caughley’s modification of Robertson’s (1987) pasture
growth model (Caughley 1987), where the quarterly change in pasture biomass is given by:

∆V = −55.12 − 0.0153V − 0.00056V 2 + 2.5R + V * eqn 1

Pasture biomass V is in units of kg ha-1, R is the quarterly total rainfall in mm and V* is drawn from a normal
distribution with the mean equal to the estimate from the regression equation and a standard deviation of 52
kg ha-1 (Caughley 1987).

Interaction 3
Rabbits effect on vegetation
The effect of rabbits on vegetation was measured by Short (1987) using an intensive grazing trial in
Kinchega National Park. The daily per capita consumption of pasture by rabbits, adjusted for body weight
and expressed as kg animal-1 day-1 was:

g R = 0.068(1 − e −V / 138 )( w0.75 ) eqn 2

where V is the pasture biomass in kg ha-1 and w is the weight of a rabbit in kg. The satiating intake is
0.068w 0.75. This is an Ivlev form of a type II prey dependent functional response.
A ratio dependent functional response could have the form:

xV

g R = 0.068(1 − e R )( w0.75 )
y
eqn 3

where the parameters x and y determine the shape of the relationship.

Vegetation effect on rabbit rate of increase

The relationship used by Pech and Hood (1998) for the quarterly rate of increase of rabbits was:

r = −4.6 + 5.5(1 − e −0.0045V ) eqn 4

where V was the pasture biomass in the previous quarter.

This relationship could be altered to express r as a function of intake rather than standing biomass. For
example:

r = − a + c(1 − e − dI ) eqn 5

Interactions between feral cats, foxes, native carnivores and rabbits 69

where a is the maximum rate of decrease, c-a is the maximum rate of increase, d is the demographic
efficiency, and I is the intake. d is related to the proportion p of satiating intake sw0.75 below which the
population declines by:

− ln(1 − a / c)
d= eqn 6
s. p.w0.75

A more mechanistic approach would be to model rabbit rate of increase as a function of both pasture growth
(rabbits respond to growing pasture by breeding), and standing biomass (standing biomass may contribute to
rabbit survival), but to our knowledge this has not been attempted and data are not available. As an
example of this approach rabbit dynamics could be divided into periods where rabbits breed and periods
where breeding ceases and populations decline. Breeding could be determined as a function of a threshold
biomass (i.e. breeding only occurs when a certain biomass of vegetation is present) combined with an Ivlev
numerical response of the form:

r = rm (1 − e − d∆V ) when V > threshold eqn 7

where rm is the maximum rate of increase, d determines the shape of the relationship between growth and
rate of increase and ∆V is the growth. When the vegetation is not growing rabbits do not breed and the
population declines depending on available biomass (or intake):

r = −a + c(1 − e − dV ) eqn 8

where a in this case reflects the maximum rate of decrease, c = ln(maximum finite survival rate in non-
breeding period), d is the demographic efficiency and V is the biomass.

Interaction 4
Interaction 4 allows for unexplained density-dependence in rabbit populations. This could come from social
interactions for example, and is modelled by a numerical response function of the form:

r = − a + c(1 − e − dV ) − jR eqn 9

where j defines the density dependence.

Interaction 8
Effect of foxes on rabbits
In semi-arid systems rabbits comprise a large percentage of the fox diet, particularly when at high density
(Pech and Hood 1998). Pech et al. (1992) estimated the functional response of foxes to rabbits at Yathong
based on the weight of rabbit found in fox stomachs and an estimate of gut passage rates. They fitted a
Holling type III functional response to the data (Holling 1959). The daily consumption of rabbits in grams per
fox per day was:

g F ( R) = 1096 R 2 /[(1.32) 2 + R 2 ] eqn 10

where R is the number of rabbits ha-1 (Pech and Hood 1998).

Effect of rabbits on fox rate of increase

Pech and Hood (1998) expressed the numerical response of foxes to rabbits as:

Interactions between feral cats, foxes, native carnivores and rabbits 70

 rF = −0.56 + 0.77(1 − e −3.2 R ) eqn 11

where R is the number of rabbits ha-1. For computer simulations they defined a minimum fox density of 0.1
km-2 to prevent fox populations from reaching unrealistically low densities and/or going extinct.
A more mechanistic approach would express fox rate of increase in terms of food intake, and would also
break the year into recruitment periods and decline (non-recruitment) periods, but whether this is feasible
requires future investigation. As an example of the approach, recruitment could be considered a point event
(say weaning) and modelled as a function of intake integrated over a preceding period (or average intake for
simplicity):

rF (recruits) = c(1 − e − d ( i −T ) ) for i ≥ T eqn 12a

rF (recruits) = 0 for i < T eqn 12b

where c is the maximum recruitment rate, d is the demographic efficiency, i is the averaged intake over a
defined period, and T is a threshold intake below which recruitment is not possible. The maximum intake
rate is defined by the maximum number of females produced per female per recruitment period c =
ln(1+max. females produced). After the recruitment peak the population declines until the next recruitment
phase at a rate according to the equation:

rF (decline) = − a + c(1 − e − d ' I ) eqn 13

where a is the maximum rate of decline, c-a is the minimum rate of decline, d’ is the demographic efficiency
based on intake and I is the intake. As an example of estimating c-a, if litters of four cubs, with two females
and two males is a reasonable approximation of the maximum recruitment of foxes, and an rmax of 0.84 per
year is a reasonable approximation of the maximum rate of increase of foxes, then the maximum finite yearly
survival rate (c-a) is about e0.84/3 = 77%.

Interaction 9
Interaction 9 allows for unexplained density-dependence in fox populations. Density dependence could be
added by adding a term to the Pech and Hood model for fox numerical response:

rF = −0.56 + 0.77(1 − e −3.2 R ) − jF eqn 14

or by adding a density dependent term to either the decline phase or increase phase in the more detailed
model.

Interaction 10
The effect of feral cats on rabbits
The fitted model for feral cats is a Holling type II functional response (Figure 12):

g C ( R ) = 52.6 R /(5.9 + R ) eqn 15

where R is the number of rabbits km-1, and the intake rate is expressed as a percentage occurrence in the
diet per unit time.

Interaction 12
Interaction 12 is added to represent interactions between foxes and feral cats not captured by competition for
resources through the functional response. This could be interference competition or direct predation.

Interactions between feral cats, foxes, native carnivores and rabbits 71

One way to model these interactions is by reducing the numerical response of one predator species in
response to the presence of the other. For example, the cat numerical response could be expressed as:

rC = −a + c(1 − e − dR ) − hF eqn 16

where R is the abundance of rabbits (or juvenile rabbits), and F is the abundance of foxes.

A simple model for preliminary exploration

For cat functional response we use a modified form of equation 15 (i.e. the satiating intake re-scaled from the
value for foxes, and rabbit abundance converted from a spotlight count to an estimate per unit area):

g C ( R) = 0.874 R /(0.5 + R) eqn 17

where R is the density of rabbits ha-1. For the cat numerical response we use equation 16, with a = 0.56 (the
same as foxes), and c-a = 0.25 (i.e. the intrinsic rate of increase rm = c-a is estimated using the allometric
relationship of Sinclair (1996) and an average body weight of 2.8 kg for adult females). We vary the
demographic efficiency d and the effect of foxes on feral cats h. In the absence of any other information we
set the minimum density of feral cats the same as foxes (0.1 km-2).

Interactions between feral cats, foxes, native carnivores and rabbits 72