managing the impacts of dingoes and other wild dogs...managing the impacts of dingoes and other wild...

Managing the Impacts of Dingoes and Other WildDogs is the first book to provide a comprehensivereview of the history and biology of wild dogs in Australia, the damage they cause, andcommunity attitudes to their management.

Australia's wild dogs include dingoes, introducedaround 4000 years ago, feral domestic dogs andhybrids between the two. They are widelydistributed throughout Australia. Predation andharassment of stock by wild dogs causes millionsof dollars worth of losses to Australian sheep,cattle and goat producers each year. There arealso opportunity costs in areas where sheep arenot grazed because of the high risk of wild dogpredation. For this reason, wild dog control is asignificant expense for many pastoralists andgovernment agencies. Yet conservation of puredingoes is also important and is threatened bytheir hybridisation with feral domestic dogs on themainland.

Key strategies for successful wild dog managementare recommended by the authors, who are scientificexperts on wild dog management. The strategiesare illustrated by case studies.

Managing the Impacts of Dingoes and Other WildDogs is an essential guide for policy makers,pastoralists, conservation reserve managers andall those interested in wild dog management.

Managing the Impacts of

Managing the Im

pacts of Dingoes and O

ther Wild D

ogs

A G R I C U L T U R E , F I S H E R I E S A N D F O R E S T R Y - A U S T R A L I A

Dingoes and Other Wild Dogs

Managing the Impacts of Dingoes

and Other Wild Dogs

Peter Fleming, Laurie Corbett,

Robert Harden and Peter Thomson

Scientific editing by Mary Bomford

Published by

Bureau of Rural Sciences

© Commonwealth of Australia 2001

ISBN 0 644 29240 7 (set)

ISBN 0 642 70494 5 (this publication)

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be

reproduced by any process without prior written permission from the Bureau of Rural Sciences. Requests

and inquiries concerning reproduction and rights should be addressed to the Executive Director, Bureau of

Rural Sciences, PO Box E11, Kingston ACT 2604.

The Bureau of Rural Sciences is a professionally independent scientific bureau within the Department of

Agriculture, Fisheries and Forestry — Australia. Its mission is to provide first-class scientific research and

advice to enable the department to achieve its vision — rising national prosperity and quality of life through

competitive and sustainable agricultural, fisheries and forestry industries.

The Commonwealth and all persons acting for the Commonwealth in preparing the booklet disclaim all

responsibility and liability to any person arising or indirectly from any person taking or not taking action

based upon the information in this booklet.

Credits for cover photographs: Main: Laurie Corbett. Inset: NSW Agriculture.

Affiliations:

Authors: Peter Fleming, NSW Agriculture – Vertebrate Pest Research Unit

Laurie Corbett, EWL Sciences Pty Ltd

Robert Harden, NSW National Parks and Wildlife Service – Biodiversity Research Group

Peter Thomson, Agriculture Western Australia – Vertebrate Pest Research Section

Editor: Mary Bomford, Bureau of Rural Sciences, Canberra.

Typeset by Lisa Curtin

Printed by Pirie Printers Pty Limited

Preferred way to cite this publication:

Fleming, P., Corbett, L., Harden, R. and Thomson, P. (2001) Managing the Impacts of Dingoes and Other

Wild Dogs. Bureau of Rural Sciences, Canberra.


PO Box E11

Kingston ACT 2604

Ph: 02 6272 4282

Fax: 02 6272 4747

Internet: http://www.affa.gov.au/outputs/ruralscience.html

AFFA Shopfront

GPO Box 858

Canberra ACT 2601

Ph: 02 6272 5550

Fax: 02 6272 5771

Email: [email protected]

Copies available from:Published by:

Wild dogs, which include feral domesticdogs, dingoes and their hybrids, are a prob-lem in Australia because their predation andharassment of stock causes millions of dol-lars worth of losses to sheep, cattle and goatproducers each year. There are also opportu-nity costs in areas where sheep are notgrazed because of the high risk of wild dogpredation. Yet dingoes are also valued as anative species and their conservation isimportant to many people. The survival ofpure dingoes on mainland Australia is threat-ened by hybridisation with feral domesticdogs.

There is little reliable information about thecost of wild dog predation or the benefits ofwild dog control. The relationship betweendog abundance and livestock predation isoften complex and variable and sometimesstock losses can be high even when wild dognumbers are low. Although spending onpest control should be justified in terms ofeconomic returns on such investments, thisis clearly difficult when changes to livestockproductivity in response to dog control areoften poorly quantified. This can be furthercomplicated where pastoral properties abutgovernment lands where dingo conservationis a management objective and dogs movebetween these areas.

This book is one in a series produced by theBureau of Rural Sciences as part of theNational Feral Animal Control Program — aNatural Heritage Trust initiative. Others in theseries include guidelines for managing feralhorses, rabbits, foxes, feral goats, feral pigs,rodents and carp. The principles underlyingthe strategic management of vertebrate pestshave been described in Managing VertebratePests: Principles and Strategies (Braysher1993) and in Australia’s Pest Animals: NewSolutions to Old Problems (Olsen 1998). Theemphasis is on the management of pest dam-age rather than on simply reducing pest den-sity. The guidelines recommend that, wher-ever practical, management should concen-trate on achieving clearly defined economicor conservation benefits.

To ensure the guidelines are accepted as abasis for wild dog management, commenthas been sought from State, Territory andCommonwealth Government agencies andfrom land managers and community andresearch organisations. The StandingCommittee on Agriculture and ResourceManagement has endorsed the publicationof these guidelines.

These guidelines provide natural resourceusers, managers, advisers and funding agen-cies with ‘best practice’ national guidelinesfor managing the economic and environ-mental damage caused by wild dogs. Theywill help land managers reduce harm to live-stock caused by wild dogs and assist in theconservation of pure dingoes through theuse of scientifically based management thatis humane, cost-effective and integrated withecologically sustainable land management.

Peter O’Brien

Executive Director


Managing the Impacts of Dingoes and Other Wild Dogs iii

Foreword

dsgsdf

FOREWORD iii

ACKNOWLEDGMENTS ix

SUMMARY 1

INTRODUCTION 5

1. NOMENCLATURE, HISTORY, DISTRIBUTION AND ABUNDANCE 11

Summary 11

1.1 Nomenclature 11

1.2 Origin, spread and distribution of dingoes throughout the world 12

1.3 Introduction, spread and distribution of dingoes and other wild dogs in Australia 13

2. BIOLOGY AND ECOLOGY 17

Summary 17

2.1 General description 17

2.2 Habitats 20

2.3 Diet and hunting strategies 20

2.4 Home range and movements 27

2.5 Social organisation and behaviour 29

2.6 Reproduction 32

2.7 Mortality and disease 33

2.8 Population dynamics and changes in abundance 35

2.9 Hybridisation 39

2.10 Co-occurence with other predators 41

3. ECONOMIC AND ENVIRONMENTAL IMPACTS AND VALUES 43

Summary 43

3.1 Economic impact 43

3.2 Environmental impact 49

3.3 Resource and conservation value 50

3.4 Diseases and parasites 51

3.5 Interactions between wild dogs, marsupial carnivores and introduced predators 53

3.6 Predator–prey relationships 54

3.7 Interactions between humans and wild dogs 60

Managing the Impacts of Dingoes and Other Wild Dogs v

Contents

4. COMMUNITY ATTITUDES AFFECTING MANAGEMENT 63

Summary 63

4.1 Community perceptions and attitudes 63

4.2 Animal welfare issues 65

4.3 Public health issues 68

4.4 Conservation issues 69

5. PAST AND CURRENT MANAGEMENT 71

Summary 71

5.1 Past legal status and management 72

5.2 Current legal status 75

5.3 Current management strategies 77

6. TECHNIQUES TO MEASURE AND MANAGE IMPACT AND ABUNDANCE 83

Summary 83

6.1 Introduction 83

6.2 Estimating abundance 84

6.3 Estimating agricultural and environmental impacts 87

6.4 Control techniques 93

6.5 Costs of control 107

6.6 Environmental and non-target issues associated with 1080 baiting 109

7. STRATEGIC APPROACH TO MANAGEMENT 111

Summary 111

7.1 Strategic approach 112

7.2 Defining the problem 112

7.3 Developing a management plan 113

7.4 Economic frameworks 124

7.5 Implementation 129

7.6 Monitoring and evaluation 129

7.7 Case studies 130

8. DEFICIENCIES IN KNOWLEDGE AND PRACTICE 135

Summary 135

8.1 Assess relationship between wild dog abundance and predation of cattle 135

8.2 Assess relative effectiveness and efficacy of baiting strategies 135

Bureau of Rural Sciencesvi

8.3 Assess effect of Rabbit Calicivirus Disease on dingo predation of livestock 136

8.4 Investigate feasibility of compensation schemes for wild dog predation 136

8.5 Train vertebrate pest control operators and managers 136

8.6 Improve public awareness of agricultural production, conservation and animal welfare issues for wild dog control 136

8.7 Develop species-specific and more humane control techniques for wild dogs 137

8.8 Assess economic importance of hydatids in wild dogs 137

8.9 Assess the role of disease induced mortality in wild dogs 138

8.10 Assess the role of wild dogs if rabies were introduced 138

8.11 Assess risks to non-target species of 1080 poisoning 138

8.12 Assess the ecological effects of wild dog control on feral cat and fox populations 138

8.13 Assess the interactions of wild dogs and native carnivore populations 139

8.14 Assess effects of wild dog abundance on macropods 139

8.15 Assess the value of dingo conservation 139

8.16 Develop a method to identify genetically pure dingoes 140

8.17 Improve knowledge about genetics of wild dogs 140

8.18 Assess the ecological role of dingo hybrids 140

REFERENCES 141

APPENDIX A Parasites and pathogens recorded from wild dogs in Australia 157

APPENDIX B Getting the best out of extension 161

APPENDIX C Authors’ biographies 165

ABBREVIATIONS AND ACRONYMS 167

GLOSSARY 169

INDEX 175

FIGURES

Figure 1 Strategic approach to managing the impacts of wild dogs. 9

Figure 2 Distribution of wild dogs and livestock. 15

Figure 3 The pelts of wild dogs, showing variety of colours. 19

Figure 4 The breeding cycle of adult (more than one-year-old) and young (less than one-year-old) female dingoes in central Australia. 33

Figure 5 Fluctuations in dingo density in the Fortescue River region 1976–84. 37

Figure 6 A conceptual model of the dynamics of a population of wild dogsin an area exposed to annual baiting programs. 37

Figure 7 The process of hybridisation between dingoes and domestic dogs. 40

Managing the Impacts of Dingoes and Other Wild Dogs vii

Figure 8 Cattle numbers and rainfall in central Australia from 1874 to 1985. 55

Figure 9 A model of predation by wild dogs in a pristine coastal ecosystemin tropical Australia. 56

Figure 10 Models of predation by wild dogs in disturbed ecosystems inarid Australia. 59

Figure 11 Dingo predation on sheep. 91

Figure 12 A wire-netting wild dog-proof fence in north-eastern New South Wales. 94

Figure 13 New wire-netting fences. 94

Figure 14 A decision-making framework for devising a plan of management for reducing predation of livestock by dingoes and other wild dogs in easternAustralia. 122

Figure 15 Some hypothetical relationships between dog density and damage. 125

Figure 16 The relationship between density of wild dogs and the damage causedby wild dogs to cattle enterprises in north-eastern New South Wales. 125

Figure 17 A marginal analysis of wild dog control. 126

Figure 18 An example of benefit–cost ratio analyses. 132

TABLES

Table 1 The occurrence of major food groups (% of samples) in the dietof dingoes and other wild dogs in Australia. 21

Table 2 Sheep losses caused by dingoes over an 18-day period. 45

Table 3 Predation of livestock by wild dogs in north-eastern New South Wales. 46

Table 4 Australian legislation and policies for dingoes and other wild dogs. 81

Table 5 The effort expended for the control of wild dogs by landholdersin north-eastern New South Wales. 109

Table 6 A decision table of strategic and reactive control measures for wild dogs in New South Wales. 117

Table 7 Hypothetical benefit–cost comparison of two wild dog control strategies using two sets of sheep productivity data. 133

BOXES

Box 1 Recognising wild dog predation of sheep 89

Box 2 A decision-making framework for wild dog control 119

Box 3 Economic framework for wild dog management 124

Bureau of Rural Sciencesviii

Special thanks are due to the following peoplewho provided detailed comments whichenhanced the accuracy and usefulness of thispublication: David Adams, Lee Allen, PeterBird, John Burley, Peter Catling, Brian Coman,Carole de Fraga, Chris Dickman, GlennEdwards, Penny Fisher, Hugh Gent, CliveMarks, Clyde McGaw, Sean Moran, AlanNewsome, Barry Oakman, Syd Shea andChristopher Short. We also thank all the partici-pants of the Wild Dog Management Workshop,held in Canberra in December 1996, whichwas the precursor of this publication.

Several individuals from the Bureau of RuralSciences deserve mention. Quentin Hart man-aged the overall publication production pro-cess and assisted with editing, design andprint management. Lisa Curtin incorporatedthe numerous modifications to the final drafts,assisted with copy editing, compiled theindex, typeset the document and had majorresponsibility for final production. DanaBradford helped collate earlier drafts of themanuscript. Kim Tatnell redrew the figures.

The draft manuscript was circulated to thefollowing organisations for comment:

• Animals Australia

• Australian Conservation Foundation

• Australian Veterinary Association

• Central Land Council

• Commonwealth Department ofAgriculture, Fisheries and Forestry

• CSIRO

• Land and Water Resources Research andDevelopment Corporation

• National Consultative Committee onAnimal Welfare

• National Farmers’ Federation

• Rural Industries Research andDevelopment Corporation

• Standing Committee on Agriculture andResource Management

• Standing Committee on Conservation

• Vertebrate Pests Committee

We thank these groups and hope that thisdocument will facilitate their involvement inmore strategic management of wild dogimpacts.

Managing the Impacts of Dingoes and Other Wild Dogs ix

Acknowledgments

Wild dogs are widely distributed throughoutAustralia and are pests in agricultural areas,particularly in areas dominated by sheepenterprises. Predation of sheep and cattlethreatens the economic viability of someproperties and the costs of wild dog controlcan be substantial. At the same time, in unoc-cupied lands and areas of extensive cattlegrazing wild dogs are often tolerated anddingoes are actively conserved in parts oftheir range.

These guidelines are a comprehensive reviewof the origins of dingoes and other wild dogs inAustralia, their biology and ecology, the dam-age they cause, and past and current manage-ment. The attitudes of various communitygroups to wild dogs and the damage theycause through predation of livestock, and tothe conservation of dingoes were sought dur-ing the production of these guidelines. A strate-gic approach to management is recommendedto reduce predation on livestock by wild dogsand to allow conservation of dingoes. Thisapproach is illustrated by case studies.Deficiencies in knowledge, management andlegislation are identified.

These guidelines have been prepared pri-marily for State and Territory managementagencies as a basis on which to consult withland managers and relevant interest groupsand to prepare state, regional and localstrategies for managing wild dogs and reduc-ing the damage they cause to livestockindustries. Their purpose is to assist in devel-oping the most cost-effective strategies toreduce wild dog damage to production.Ideally, such strategies are based on reliablequantitative information about the damagecaused by dogs, the cost of control measuresand the effect that implementing control hason reducing damage. In developing theseguidelines the authors have used all suchavailable information. In some instances,however, where reliable information is notyet available, land managers responsible forwild dog management will still have to makeassumptions about impacts and the efficacyand cost-effectiveness of control techniques.

Biology, ecology and taxonomy

The wild dog population comprises two sub-species of canid, dingoes (recommendednomenclature, Canis lupus dingo) and feraldogs (recommended nomenclature, C. l.familiaris) and hybrids of the two. Dingoeswere first introduced to Australia some 4000years ago and domestic dogs have been pre-sent since first European settlement in 1788.Dingoes and other wild dogs are widely dis-tributed throughout the country and are pre-sent in most environments. However, din-goes and other wild dogs have beenremoved from much of the agricultural zoneover the past 200 years and hybridisationbetween the subspecies over that time hasresulted in a lesser proportion of pure din-goes, especially in south-eastern Australia.

The average adult dingo in Australia weighs16 kilograms and, although feral dogs andhybrids may weigh up to 60 kilograms, mostare less than 20 kilograms. Pure dingoes aredistinct from similar-looking domestic dogsand hybrids because they breed once a yearand have some different skull characteristics.The present distribution of dingoes andother wild dogs covers most of the mainland,except for the sheep and cereal growingareas of south-eastern Australia. Wild dogslive in small groups or packs in territorieswhere the home ranges of individuals varybetween 10 and 300 square kilometres.Packs are usually stable but under certainconditions some wild dogs, usually youngmales, disperse.

Although wild dogs eat a diverse range offoods, from insects to buffalo (Bubalusbubalis), they focus on medium and largevertebrates. Hunting group size and huntingstrategies differ according to prey type tomaximise hunting success. Larger groups ofwild dogs are more successful when huntinglarge kangaroos (Macropus spp.) and cattleand solitary animals are more successfulwhen hunting rabbits and small macropods.

Managing the Impacts of Dingoes and Other Wild Dogs 1

Summary

Female dingoes become sexually mature bytwo years and have only one oestrus periodeach year, although some do not breed indroughts. Female feral dogs of a similar sizeto dingoes have the potential to have two lit-ters each year but this is rarely achievedbecause of the high nutritional demands ofraising young. Litters average five pups andare usually whelped during winter.

Agricultural impacts

Wild dogs prey on livestock and predationon sheep and cattle can threaten the eco-nomic viability of properties in some areas.Sheep are the most commonly attacked live-stock, followed by cattle and goats.

Some individual wild dogs cause far moredamage than others, although many individ-uals will attack or harass sheep, sometimesmaiming without killing. Wild dogs some-times chase sheep without attacking them.Even when wild dogs kill sheep, they oftenleave carcasses uneaten. Wild dogs that fre-quently kill or maim sheep often eat otherprey, indicating that predation of livestockmay be independent of the abundance ofother prey. Surplus killing, where moresheep are killed than are needed for food,means that stock losses can be high evenwhen wild dogs are at low densities.

Wild dogs are implicated in the spread ofhydatids, a risk to human health and the causeof losses of production associated with hydati-dosis (causal agent Echinococcus granulosus)in cattle and sheep. They also provide a reser-voir for heartworm (Dirofilaria immitis) infec-tion and dog diseases such as parvovirus(causal agent Parvovirus). Wild dogs pose thegreatest potential risk of maintaining andspreading dog rabies (Rhabdoviridae) if itwere to be introduced to Australia.

Conservation of dingoes

The dingo is usually considered a nativeAustralian mammal. Dingoes are an intrinsicpart of natural ecosystems and they alsohave aesthetic value. There is some publicexpectation that dingoes should be con-served and dingoes are legally protected insome States and Territories. In Australianwildlife communities, wild dogs are top

order predators, and as such probably have amajor influence on the abundance of thespecies they compete with or prey on. Theinteractions between wild dogs and foxes(Vulpes vulpes) are not well understood. It isunknown whether the presence wild dogsreduces fox abundance and hence whetherwild dogs reduce the impact of foxes onnative animal prey.

The greatest threat to the survival of dingoesas a protected sub-species is hybridisationwith other dogs. In the more settled coastalareas of Australia and increasingly in out-back Australia, the barriers to matingbetween domestic dogs (feral and owned)and dingoes are rapidly being removed.Hence hybridisation is becoming more com-mon and the pure dingo gene pool is beingswamped. In south-eastern Australia, morethan half the wild dogs are hybrids. Changesto policies on wild dog management andpeople’s attitudes would be needed to pre-vent the extinction of pure dingoes on themainland. The main hope for conservation isto educate people about the plight of din-goes and to manage pure dingoes on largeislands such as Fraser Island and MelvilleIsland.

Community attitudes affecting management

Opinions vary as to the pest status of din-goes and other wild dogs. People in the agri-cultural sector often view wild dogs as a pestto be removed from the environment. Incontrast, Aboriginal peoples, urban peopleand conservationists often view dingoes asnative wildlife that should be conserved.Public opinion influences not only the typeof management strategies that are developedbut also the type of control methods that areused. Wider public attitudes rightly demandthat the techniques used for wild dog controlbe as humane as possible and minimise risksto non-target animals and other environmen-tal values. Management strategies that do notaddress or acknowledge broad communityattitudes are susceptible to disruption orinterference.

Bureau of Rural Sciences2

Past and current management

In the past, legislation for the managementof wild dogs has included punitive Acts andActs dealing with the conservation ofwildlife. Management of wild dogs reliedheavily on labour-intensive techniques, suchas trapping, shooting, and ground baiting,with bounty payments being offered as anincentive to kill dogs. Much of the controlwork was reactive, dealing with problems asthey arose. Nevertheless, some strategic, pre-ventative control was carried out includingthe construction of district-wide exclusionfences.

The dingo is extinct in much of the sheepand cereal production zones of eastern andsouthern Australia because of habitat modifi-cation and the success of early poisoningcampaigns. The areas that are largely with-out wild dogs are separated from areaswhere they are still present by dog-prooffences that were erected around the turn ofthe century and are still maintained.

In most States and Territories, there is a legalrequirement to destroy wild dogs in sheepand cattle grazing zones. Poisoning pro-grams form the basis of lethal control effortsalthough trapping and shooting are alsoimportant.

Current management strategies focus on theobjective of minimising the impact of wilddog predation on livestock, not just onkilling wild dogs. Aerial baiting with 1080(sodium fluoroacetate) baits forms a majorpart of most management programs and isprimarily targeted at limited zones adjacentto livestock grazing areas. Large coordinatedcampaigns have generally been adopted,being more efficient and effective than smalllocalised efforts. Bounty payments have notbeen successful in reducing predation bywild dogs and are subject to abuse.

Policy and legislation to encourage the con-servation of pure dingoes is required insome States and Territories and a concertednation-wide effort is needed to ensure thatdingo conservation is not thwarted by con-flicting legislation. Simultaneously, the con-trol of wild dogs, including dingoes, must bepermitted where predation of livestockoccurs.

Techniques to measure and manageimpact and abundance

To formulate wild dog management plans, itis necessary to measure the level of preda-tion inflicted by wild dogs and to measurechanges in wild dog abundance. These twomeasures enable an assessment of whenwild dog control is required and how effec-tive it is.

The principal techniques to control wilddogs are exclusion fencing, shooting, trap-ping and poisoning. Poisoning using 1080 isthe most cost-effective means of reducingpopulations of wild dogs over large areas ofremote or inaccessible country. Various baittypes are used and methods of placementrange from burying individual baits to drop-ping baits from aircraft. Trapping is still usedfor wild dog control and will probablyalways be needed to target particular dogsthat cannot be removed by other means.New techniques such as the use of livestock-guarding dogs, poison ejecting devices andtoxic collars have been suggested as alterna-tives to current methods.

Strategic approach to management

The strategic approach to wild dog manage-ment allows improvements at both the localand regional scale. The strategic approachhas four components: defining the problem;developing a management plan; implement-ing the plan; and monitoring and evaluatingprogress and outcomes.

Defining the problem involves the identifica-tion of who has a wild dog problem, whatharm the dogs cause, where, when and whydamage occurs and how much it costs.

The development of a management planrequires setting management objectives thatshould include interim and long-term goals,a time frame for achieving them and indica-tors for measuring performance. Options forwild dog control include local eradication,strategic management, reactive managementor no dog control.

Economic frameworks are needed for assess-ing the value of alternative strategies to man-age wild dogs. In some situations, manage-ment plans that include conservation strategies


for dingoes are required so that potentiallyconflicting goals can be encompassed.Consultation between stakeholders and clearidentification of the goals is critical for avoidingpotential conflicts between stakeholder groupswith different legal obligations and objectives.

Wild dogs have large home ranges and oftentraverse boundaries between lands managedby different stakeholders. Action by groups,including government agencies, is thereforean essential element of planning and imple-mentation. By pooling resources, wild dogcontrol groups and boards have been betterable to manage wild dog problems.Management programs must be flexibleenough to account for the different objec-tives of stakeholders.

Monitoring and evaluation occur at differentlevels throughout the implementation andon completion of actions. Operational moni-toring records and reviews the costs ofactions during the program and ensures thatthe management plan is executed in themost cost-effective manner. Performancemonitoring assesses the effectiveness of themanagement plan in meeting the agriculturalproduction or conservation objectives thatwere established initially. Evaluation of datafrom both forms of monitoring enables thecontinuing refinement of the managementplan. Strategic management of wild dogs isbased on the concept of adaptive manage-ment, in which the management plan is flex-ible, responding to measured changes in

economic, environmental and pest circum-stances. By adopting the strategic approach,predation by wild dogs should be minimisedwhile the conservation of the dingo propor-tion of the wild dog population will beenhanced. Under such an approach, limitedresources will be better allocated and thescale of management will be more appropri-ate for wild dog problems.

Deficiencies in knowledge and practice

Although there is much knowledge aboutthe ecology, behaviour and effects of preda-tion by dingoes and other wild dogs, sometopics require further research to enable bestpractice management to be implemented.These include better definition of the agricul-tural impacts of wild dogs and control pro-grams for different enterprises in differentregions, study of the interactions betweenthe control of rabbits and wild dog predationof livestock and the effects of wild dog con-trol on the abundance of kangaroos and wal-labies (Macropus spp.), and the effects ofthis on agriculture and forestry. There arealso knowledge deficits relating to the con-servation of dingoes, the effects of wild dogcontrol programs on persistence of pure din-goes, the interactions between predation bywild dogs and the conservation status ofnon-target animals, and the interactions ofwild dogs with feral cats, foxes and nativecarnivores.


These guidelines for managing the impactsof dingoes (Canis lupus dingo) and otherwild dogs (C.l. familiaris) are the eighth inthe Managing Vertebrate Pests series beingpublished by the Bureau of Rural Sciences(BRS) in cooperation with the VertebratePests Committee of the Standing Committeeon Agriculture and Resource Management(SCARM). These guidelines were fundedunder the agricultural component of theNational Feral Animal Control Program(NFACP) of the Natural Heritage Trust(NHT). A fundamental difference betweenthese guidelines and the preceding publica-tions exists because dingoes hold a legalposition unique amongst Australian mam-mals. Unlike most of the other speciesaddressed by the series, dingoes are simulta-neously a protected native species anddeclared vermin. The dingo and some nativebirds and rodents are both protected anddeclared according to their occurrence andsituation.

Other guidelines in the series include thosefor managing feral horses (Dobbie et al.1993), rabbits (Williams et al. 1995), foxes(Saunders et al. 1995), feral goats (Parkes etal. 1996), feral pigs (Choquenot et al. 1996),rodents (Caughley et al. 1998) and carp(Koehn et al. 2000). A companion volume,Managing Vertebrate Pests: Principles andStrategies (Braysher 1993), which explainsthe principles on which best practice pestmanagement is based, can be read in con-junction with all of these guidelines. There isalso an overarching publication (Olsen1998), designed for general reading, whichreviews past management of pest animals inAustralia and promotes a more strategicapproach for future management. The bene-fits of focusing on the damage caused by apest and not the pest itself are explained.Olsen (1998) also explains the need to takeinto account the links between different feralanimal species and other aspects of landmanagement, consistent with the holisticapproach advocated under the EcologicallySustainable Development (ESD) Strategy andLandcare.

A single publication considering the mainvertebrate pests would be desirable and con-sistent with the holistic approach to landmanagement advocated under the ESDStrategy and Landcare objectives. Such apublication would take into account linksbetween pests and links between pests andother aspects of land management.However, the complexities posed by such anapproach and current limited knowledge ofinterspecific interactions has made thisimpractical. All the guidelines, includingthese, consider interactions between speciesand the relationships with other aspects ofland management.

These guidelines are principally for State andTerritory land management agencies, toassist them to better coordinate, plan andimplement regional and local programs thatcan more effectively manage adverseimpacts of wild dogs. The CommonwealthGovernment has an interest in improvingstrategies, techniques and approaches tomanage the damage caused by wild dogs,both through its responsibilities as a manag-er of Commonwealth lands and resources,and through programs such as NFACP andthe National Landcare Program (NLP) of theNHT, and the National Strategy for theConservation of Australia’s Biological Diversity(Commonwealth of Australia 1992).

Vertebrate pests in Australia: speciesor situations?

The definition of pest status can be con-tentious. Some species are regarded as pestsall the time in all situations because of theircurrent detrimental impacts or their potentialadverse impacts, given their biology,behaviour and historical performance aspests in similar or other habitats. Other ani-mals are generally regarded as having eitherbeneficial or neutral net impact in most situa-tions. Some species, such as the dingo andthe feral goat, may be both a significant pestand a significant conservation or economicresource. Perhaps the most useful criterionfor evaluating the status of an animal is to


Introduction

evaluate it in terms of its value in a particularsituation. For example, cats are valued highlyas pets by many people and some pedigreecats have a high market price. Conversely,predation by feral cats is regarded as a pro-cess threatening some endangered nativevertebrates with extinction (Dickman 1996).

The National Feral Animal ControlProgram

NFACP is working with State, Territory, andlocal governments to reduce damage by pestanimals to agriculture and the environment.The agricultural component of NFACP isadministered by BRS; the environmentalcomponent by Environment Australia.

Under its component of NFACP, BRS is pro-ducing these national management guide-lines for the main pest species of agriculturalproduction and supporting projects toaddress the information, management andextension deficiencies they identify and todemonstrate the strategic managementapproaches they advocate.

Applying a strategic approach to the man-agement of the impacts of wild dogs involvesthe establishment of four essential compo-nents (Figure 1) This approach has beenadopted in the development of each set ofnational guidelines.

The strategic management approach

Problem definition and planning ofmanagement strategies

There are two problems requiring manage-ment. The first problem is predation of live-stock by dingoes and other wild dogs.Although there are no estimates of theAustralia-wide losses to livestock productioncaused by wild dogs (including dingoes), theestimated annual expenditure on controlactivities of $7 million is second only to thatfor rabbits. The historical threat of predationby wild dogs has largely determined the distri-bution of sheep and cattle in Australia. A barri-er fence stretching from the Great AustralianBight through South Australia andQueensland and ending in north-eastern NewSouth Wales has been built and maintained by

government agents and graziers to excludewild dogs from sheep grazing lands.

Secondly, the dingo has been in Australialong enough to be regarded as part of thenative fauna. The existing dynamics ofAustralia’s fauna have evolved with thedingo and the conservation of dingoes withinnon-agricultural environments is a legitimateaim. Since European settlement, the increas-ing presence in the population of genes fromferal and domestic dogs has reduced thedingo population’s integrity. If this trendcontinues it is predicted that the increasingoccurrence of domestic dog genes in wildpopulations will effectively lead to theextinction on the mainland of the dingo as asubspecies by 2100 (Corbett 1995a).

Strategies to conserve pure dingoes can con-flict with strategies to control wild dogs toreduce their impacts on livestock enterpris-es. Developing approaches to satisfactorilyaddress both problems requires clarificationof issues and knowledge of the biology andstatus of dingoes and other wild dogs. Thus,Chapter 1 discusses the taxonomy of dingoesand other wild dogs, and details their origins,distribution and abundance, and Chapter 2reviews their biology and ecology. InChapter 3, the impacts of wild dogs onhuman activity and environments are dis-cussed. Public attitudes can strongly influ-ence the perceived nature of dingoes andother wild dogs as a resource or as a prob-lem, and these issues are addressed inChapter 4. The legal status of dingoes andother wild dogs, and past and current man-agement practices, are reviewed in Chapter5.

The objective of the national guidelines is tostimulate a widespread change in approachto the management of dingoes and otherwild dogs from ad hoc measures to a strate-gic management approach based on cooper-ative action and the most recent knowledge.An integrated approach on a regional or totalcatchment scale is advocated because theproblems associated with dingoes and wilddogs usually extend past the boundaries ofindividual land holdings.


The primary aim of a land manager is tomeet their desired conservation and/or agri-cultural production goals using practical andcost-effective means. This must be done ashumanely as possible and without degradingother natural resources on which the long-term sustainability of agriculture and biodi-versity depend. There is great variabilitywithin and between the environments inwhich wild dogs occur and this influencesmanagement activities. The factors that affectthe desired outcomes include fluctuatingcommodity prices, legal constraints, climaticvariability including drought, interactions ofwild dogs with prey, grazing pressure, live-stock genetics, conservation objectives, ani-mal welfare considerations and social fac-tors.

Legislative constraints and the extensivenature of wild dog predation problems haveresulted in a strategic approach being manifestin many areas. These guidelines will haveachieved their purpose if the advocated strate-gic approach is widely accepted and imple-mented. Strategic management requires themeasurement of the impacts and abundanceof wild dogs, and this can be achieved in anumber of ways which are detailed in Chapter6. Many people and agencies, including gov-ernments and community groups, jointly ownwild dog problems and need to work coopera-tively to find strategic solutions (Chapter 4 andChapter 7). In some cases, inadequacies inavailable knowledge may prevent identifica-tion of the best strategy. A flexible approach,where the implementation of managementactions are continually monitored and evaluat-ed and modified if necessary (‘learning bydoing’ or ‘adaptive management’) is often thebest approach. Strategic approaches to themanagement of dingoes and other wild dogsare described in Chapter 7.

Implementation, monitoring andevaluation of strategic programs

A group approach to the implementation ofmanagement programs reflecting coopera-tion between individuals and agencies at thelocal and regional level is encouragedthroughout the guidelines and Chapter 7outlines features to aid the implementationof management plans. A group approach

involves all affected landholders and otherswith a significant interest in the managementand conservation issues associated with din-goes and other wild dogs from early plan-ning stages through to implementation,monitoring and evaluation.

At a national level, such an approachrequires that the various roles and responsi-bilities of government agencies, individualsand interest groups are taken into accountand integrated. State and Territory govern-ments provide the legislative and regulatoryinfrastructure, and conservation and pestcontrol agencies administer the appropriateActs and regulations. Responsibility for localmanagement of wild dogs rests with theowners and occupiers or administrators ofland. The active participation of theVertebrate Pests Committee (VPC) and allassociated government agencies in develop-ing these guidelines is thus important inobtaining their acceptance and support forimplementation by both agricultural andconservation interests.

For a strategic management program to besuccessful, it must be continually monitoredand evaluated so that modifications andimprovements can be incorporated. Suchmonitoring, evaluation and re-evaluation isan ongoing process and techniques forassessing impacts and monitoring manage-ment practices and programs are detailed inChapter 6.

Strategic management at the localand regional level

The management of wild dogs is a complexissue because the pest status and conservationstatus of the species must be balanced. Thisdocument presents the best practices for themanagement of dingoes and other wild dogs.Management must attempt to reduce theadverse impacts of wild dogs while maintain-ing viable populations of genetically pure din-goes and these guidelines amalgamate thebest available information on effectiveapproaches. These guidelines consider theconservation values of dingoes and the influ-ence of hybridisation on their genetic integrity.Conservation priorities affect managementdecisions for wild dogs, particularly at the


interface of developed agricultural lands andland managed for conservation. The emphasisin these guidelines is therefore to concentrateon managing the impacts of wild dogs on agri-cultural and environmental resources whileconserving the dingo as a sub-species. At thelocal and regional level, land managers needto use the information in the book to developand apply their own strategies. Examples ofsuccessful strategic approaches, both hypo-thetical and real, involving private and govern-ment land managers are given in Chapter 7.

These guidelines outline best practices basedon present knowledge. A number of defi-ciencies in that knowledge are identified inChapter 8. It is expected that best practices

will evolve through adaptive managementand that community-based groups willbecome more involved in the strategic man-agement of wild dogs. These guidelinesallow local groups to own the pest or conser-vation problem as well as managementstrategies derived from the guidelines. It isintended that these guidelines will also assistState and Territory governments in their roleof providing legislative, technical and policysupport for the management of dingoes andother wild dogs.

All dollars have been converted to 1999–2000 Australian dollars unless otherwise stated in the text.



Pro

ble

m d

efin

itio

n

(Cha

pter

s 3

and

4 an

d Se

ctio

ns 6

.2,

6.3

and

7.2)

•w

hat i

s th

e pr

oble

m–

pred

atio

n–

cons

erva

tion

•w

ho h

as th

e pr

oble

m•

who

else

is in

volv

ed•

defin

e im

pact

–ec

onom

ic–

envi

ronm

enta

l•

mea

sure

impa

ct•

map

ping

Managem

ent

Pla

n

(Cha

pter

s 5

and

6 an

d Se

ctio

n 7.

3)

•de

fine

obje

ctiv

es•

partn

ersh

ips

•go

vern

men

t rol

e•

sele

ct m

anag

emen

t opt

ions

–lo

cal e

radi

catio

n–

cons

erva

tion

–

strat

egic

man

agem

ent

–re

activ

e m

anag

emen

t–

no d

og c

ontro

l•

deve

lop

a m

anag

emen

tstr

ateg

y•

defin

e pe

rform

ance

crite

ria

Imple

men

tation

(Sec

tions

6.4

and

7.

4)

•gr

oup

actio

n–

owne

rshi

p•

who

le fa

rm

catc

hmen

t/di

stric

t•

gove

rnm

ent r

ole

•m

onito

ring

regi

me

Monitori

ng a

nd

Evalu

ation

(Sec

tions

6.2

, 6.3

and

7.

5)

•as

sess

con

trol a

ndco

nser

vatio

n•

com

pare

ove

r tim

e•

com

pare

tech

niqu

es•

eval

uatio

n of

ou

tcom

es

Fig

ure

1:S

trat

egic

appro

ach to m

anag

ing

the

impac

ts o

f w

ild d

ogs

(af

ter

Bra

ysher

199

3).

Summary

There is currently debate about the correcttaxonomy of dingoes and other wild dogs.Both are derived from wolves (Canis lupusssp.). In this book, the scientific names Canislupus dingo and Canis lupus familiaris arerecommended for the subspecies dingo anddomestic dog respectively.

Dingoes were brought to Australia from Asiaabout 4000 years ago. They were present inAsia possibly 10 000 to 14 000 years agoand were derived from wolves. The dispersalof dingoes throughout Australia was aidedby Aboriginal people who used dingoes forfood, companions, hunting-aids and bed-warmers. The dingo never reachedTasmania. Domestic dogs were brought intoAustralia by Europeans as early as 1788and their release into the wild has contin-ued since then.

The distribution of dingoes in Australia hasdecreased since European settlement, althoughthe abundance of wild dogs in some areasmay have increased due to the provision ofpermanent water. Food, water and cover areprobably the most important factors affectingthe distribution of wild dogs in areas withoutintensive control. Dog-proof fences that protectsheep from predation also limit the distributionof wild dogs.

1.1 Nomenclature

The scientific names applied to animals thathave been selected by domestication are atpresent the subject of much debate in taxo-nomic circles (Gentry et al. 1996; Brisbin1998). Wild-living dogs of Australia are mem-bers of the family Canidae, belonging to theorder Carnivora. The dog, C. familiaris, isthe type species for the genus Canis,although it is not considered a naturalspecies but rather one developed by humansfrom grey wolves (C. lupus) (Stains 1975).Analyses of chromosomes have shown thatthe karyotypes of C. lupus and C. familiariscannot be differentiated (Chiarelli 1975).

While this does not assist in the differentia-tion of Canis species, it demonstrates recentor close common evolutionary lineage.

The name ‘dingo’ is probably a European cor-ruption of the word ‘tingo’, used by Aboriginalpeople at Port Jackson to describe camp din-goes (Thomson 1870). Other Aboriginal namesfor dingoes include ‘warrigal’, ‘maliki’ (campdogs) and ‘wantibirri’ (wild dingoes)(Breckwoldt 1988). The scientific name of thedingo has undergone much synonymy but in1982, the specific designation Canis lupusdingo was recommended (Honacki et al.1982). This name was proposed to reflect theirwolf ancestry, the uniformity of dingo popula-tions throughout their huge distribution in Asiaand Australia and ‘universal usage’; however,C. l. dingo is not in common usage. Of 109documents in a search for the keyword ‘dingo’in the Wildlife Worldwide database, 1935 toJune 1995 (National Information ServicesCorporation), none used this nomenclature.Canis familiaris dingo was the most common-ly used name for dingoes. Domestic and feraldogs were usually grouped as C. f. familiaris.The name C. f. dingo has not been suppressedand is still in common usage (Dr J. Clutton-Brock, British Museum, London, pers. comm.1997).

The names C. f. dingo for the dingo proportionof the wild dog population and C. f. familiarisfor both wild-living and commensal domesticdogs have the greatest use in scientific litera-ture. It logically follows from the derivation byhuman selection of both dingoes and dogsfrom wolves and the genetic similarities that allthree animals should be C. lupus. Wolves aremorphologically separable from both dingoesand domestic breeds of dog (Lawrence andBossert 1967) and have such morphologicaland behavioural dissimilarities (Newsome et al.1980) for all three to be considered separatesubspecies allocated sub-specific names,lupus, dingo and familiaris respectively. Thisdelineation of the classification of dingoes andother wild dogs has the potential to affect man-agement strategies through acts of law (Brisbin1998).


1. Nomenclature, history, distribution and abundance

Corbett (1995a) concludes that wild-livingdogs in Australia are subspecies of C. lupus,that is C. l. dingo and C. l. familiaris, butthese designations are yet to be formallyaccepted (Corbett 1995b). However, becauseof genetic similarities that indicate commonlineage with wolves and sufficient differ-ences that they justifiably be regarded as dis-tinct sub-species, the recommended nomen-clature is: dingoes, C. l. dingo; and domesticdogs, C. l. familiaris. Recent evidence, basedon skull morphology, size, coat colour andreproduction, indicates the existence ofregionally distinct populations of dingoesbetween Australia and Thailand (Corbett1985; 1995a) but not within Australia(Corbett in press).

‘Delineation of the classification of dingoes and

other wild dogs has the potentialto affect management strategies

through acts of law’.

The terms wild dog, feral dog, dingo andhybrids mean different things. We define thevarious meanings as follows:

Dingoes: native dogs of Asia. Dingoes werepresent in Australia before European settle-ment and still occur in the wild here. Puredingoes are populations or individuals thathave not hybridised with domestic dogs orhybrids.

Domestic dogs: dog breeds (other than din-goes) selectively bred by humans, initiallyfrom wolves and/or dingoes, that usuallylive in association with humans. Introducedto Australia by European settlers.

Hybrids: dogs resulting from crossbreedingof a dingo and a domestic dog and thedescendants of crossbred progeny.

Wild dogs: all wild-living dogs (includingdingoes and hybrids).

Feral dogs: wild-living domestic dogs.

Free-roaming dogs: dogs that are ownedby humans but not restrained so they are freeto travel away from their owner’s residence.

Commensal dogs: wild dogs (includingdingoes and free-roaming domestic dogs)

living in close association with but indepen-dently of humans.

Where we were unable to ascertain the sta-tus of animals in the publications consulted,we have followed the original authors’usage. Whether this usage is in accordancewith the above definitions is unknown.

1.2 Origin, spread and distribution of dingoesthroughout the world

Recent work using molecular techniques(DNA fingerprinting) indicates that a dingo-like canid existed perhaps 100 000 before pre-sent (BP) and that it was distinctly dingo-likeabout 10 000–14 000 BP (Gentry et al. 1996).However, the earliest known dingo-like fos-sils are from Thailand (dated at 5500 BP) andVietnam (5000 BP) (Higham et al. 1980).Based on skull morphology, these earlycanids evolved from the pale-footed (alsoknown as the Indian) wolf (C. l. pallipes)and/or the Arabian wolf (C. l. arabs) between6000 and 10 000 years ago (Corbett 1995a).

‘The dingo’s general morphology has remained

virtually unchanged for the past5500 years, although this situation is now rapidly

changing through hybridisationwith domestic dogs.’

This phase of rapid evolution coincided withthe time when people in southern Asiachanged their nomadic hunter-gathererlifestyle to a settled agricultural subsistencethat allowed commensal relationships todevelop between wild animals and people(Clutton-Brock 1989; 1992). Effectively thiswas the start of the evolution of the earlydingo-like canids into dingoes and other dogs,but subsequent evolution proceeded alongdifferent pathways in western and easternAsia, and at different rates.

In western Asia and southern Europe, peopleselectively bred these primitive canids toimprove the characteristics of dogs for hunting,herding, hauling, guarding, scavenging and


fighting, as well as for therapeutic, companion,symbolic and novelty values (that is, domesti-cation). The outcome is the immense range ofsize, shape, colour and temperament found inthe 600 or so modern domestic breeds of dogs.

In eastern Asia, people used the early canidsfor food, hunting, alerting and perhaps forcultural reasons, but they were not selective-ly bred. Morphological comparisons of skullsof the early Asian fossils show a close simi-larity with modern dingoes from Thailandand Australia, but a clear difference to mod-ern domestic dogs (Corbett 1985; 1995a).There are also close similarities in bodyshape, breeding pattern, coat colour andsocial behaviours between dingoes in south-east Asia and Australia. This indicates thatthe dingo’s general morphology hasremained virtually unchanged for the past5500 years, although this situation is nowrapidly changing through hybridisation withdomestic dogs (Section 2.9).

The early dingo became cosmopolitan throughits association with the movements of earlyhumans as their populations expanded(Bellwood 1978, 1984; Thorne and Raymond1989). It was during this expansion that din-goes were transported to Australia where themost recent introductions were by theMacassan trepangers (90–350 years ago)(Macknight 1976), and the ‘boat people’ fromVietnam and Indonesia.

Dingoes probably accompanied the Asianseafarers mainly as a source of fresh foodduring long sea voyages, or as guard dogsduring stopovers. There were also culturalreasons for transporting dingoes (Clutton-Brock 1977; Medway 1977;Corbett 1995a).

The primitive dogs of most Pacific islands(Titcombe 1969), the ancient kirri dog ofNew Zealand (Colenso 1877; Bay-Petersen1979), the basenji in the African Congo (Coe1997), and the New Guinea singing dog(Troughton 1957) are descended from south-east Asian dingoes. According to fossil evi-dence, the primitive dogs of the Americaswere also morphologically very similar todingoes and they probably arrived theretogether with people via the Bering Strait(Olsen and Olsen 1977). The Carolina dog isthe remaining descendent of the earlyAmerindian canids (Brisbin 1989).

1.3 Introduction, spread and distribution of dingoes andother wild dogs in Australia

1.3.1 Introduction to Australia

The oldest reliably dated dingo remains inAustralia are from 3450 ± 95 years BP(Milham and Thompson 1976) and fossils of


Thai dingo, which has similar morphology to the Australian dingo (Source: L. Corbett).

about this age have been found throughoutmainland Australia (White and O’Connell1982). This suggests that, having reached thiscontinent, dingoes colonised the mainlandand many offshore islands quickly and com-pletely, although they never inhabitedTasmania.

This dispersal was probably assisted byAboriginal people who had arrived on thecontinent at least 40 thousand years earlier.Some Aboriginal tribes used dingoes to huntgame, especially kangaroos and wallabies(Macropus spp.), possums (Phalangeroidea)and echidnas (Tachyglossus aculeatus). SomeAboriginal people suckled pups and slept withdingoes for warmth (Lumholtz 1889; Finlayson1935; White 1972; Dixon and Huxley 1985;Pickering 1992). Dingoes are well-representedin Aboriginal mythology and rock art(Breckwoldt 1988).

Domestic dogs were first introduced intoAustralia in 1788 (Australian GeographicSociety 1996) and dispersal into the wild(both deliberate and accidental) has beencontinuing since then. The assumed sourceof feral dogs in Australia is the abandon-ment, neglect, loss or deliberate release intothe wild of domestic dogs by humans since1788 (Gould 1863; Corbett 1995a). Althoughthere are few records of such releases, theiroccurrence is supported by reports of free-living dogs of specific breeds being seen orcaptured in remote areas (Newsome andCorbett 1985; Jones 1990; Corbett 1995a).Some of the larger feral dogs may have beenbred to hunt feral pigs and other game andbecome lost on hunting expeditions. Theincidence of C. l. familiaris and hybrids inwild dog populations is higher in south-east-ern Australia than in inland and north-west-ern Australia (Newsome and Corbett 1982;1985; Jones 1990; Thomson 1992a).

1.3.2 Distribution in Australia

The distribution of dingoes in Australia (Figure2) has been reduced since European settle-ment in 1788, when dingoes occurred through-out the mainland. For example, in SouthAustralia they now occupy about 60% of theirformer range (P. Bird, Primary Industries andResources, South Australia, pers. comm. 1999).However, the abundance of dingoes is likely to

have increased over much of their remainingrange (Corbett 1995a) as a result of increases inwatering points and food supplies. Dingoesand other wild dogs are widely distributedthrough mainland Australia to the north andwest of the barrier fence (Breckwoldt 1988;Thomson and Marsack 1992). Most popula-tions throughout Australia comprise pure din-goes, although in south-eastern Australia themajority are hybrids (Figure 2) (Newsome andCorbett 1985; Jones 1990).

‘Control by humans has had significant impact on the distribution and abundance of

wild dogs since European settlement.’

In Queensland, most wild dogs occur out-side the Dog Fence (Figure 2 and Chapter 6)which surrounds the sheep grazing areas ofcentral and southern Queensland althoughthere is no information about relative dogdensities either side of the fence (Fleming etal. 1992). Although wild dogs sporadicallyoccur in the Western Division inside the DogFence that runs along the north-western bor-ders of New South Wales, wild dogs are mostabundant and most commonly occur intableland and coastal environments in theeast of the State. The highest densities andgreatest impact of wild dogs are in theNorthern and Southern Tablelands, the latterbeing contiguous with the Eastern Highlandsof Victoria which is the location of most wilddogs in that State (Mitchell 1986). The activi-ties of free-roaming dogs are most commonlycentred near towns and cities (Coman andRobinson 1989; Meek 1999).

The absence of wild dogs from most of themore closely settled and agriculturally devel-oped areas, and areas within the exclusionfence, indicates that control by humans hashad a significant impact on the distributionand abundance of wild dogs since Europeansettlement.



Common

Uncommon

Generally common, but high levels ofcontrol within parts of this zone mean thatdingoes may be absent in certain areas.

Naturally sparse

Absent

Predominantly cattle

Absent

Predominantly sheep

Sheep and Cattle

Dog-prooffence

Dog-prooffence

A

A

B

C

Dingoes, hybrids and feral dogsA

Known hybridisation with feral dogs BArea within recently rehabilitatedQueensland Barrier Fence.Dingo numbers likely to decline.

C

Mostly pure dingoes - above dashed lineD

DD

Distribution of Wild Dogs

Distribution of Livestock

Perth

AliceSprings

AdelaideMelbourne

Canberra Sydney

Darwin

Hobart

Darwin

Hobart

Brisbane

Fraser Island

Fraser Island

Townsville

Perth

AliceSprings

AdelaideMelbourne

Canberra Sydney

Brisbane

Townsville

Section ofrabbit-proof fencewithin dog-proof fence

Section ofrabbit-proof fencewithin dog-proof fence

Figure 2: Distribution of wild dogs and livestock (after Breckwoldt 1988; Corbett 1995a; Fleming 1996b)

Summary

The average adult dingo in Australia stands57 centimetres at the shoulder, is 123 cen-timetres long from nose to tail tip andweighs 16 kilograms. Dingoes are smaller inAsia. Feral dogs of up to 60 kilograms havebeen recorded but most are less than 20kilograms and their height and length aresimilarly variable.

The coat of dingoes is typically ginger butvaries from sandy-yellow to red-ginger andis occasionally black-and-tan, white orblack. Most dingoes have white markings onthe feet, tail tip and chest, some have blackmuzzles and all have upright ears andbushy tails. Coats with a dark dorsal strip ordappling in the white areas usually indicatehybrids. Both hybrids and feral dogs havehighly variable coat colours and patterns.

Pure dingoes have only one oestrus cycle ina year whereas hybrids and domestic dogsmay have two cycles. Dingoes also differfrom hybrids and domestic dogs in someskull characteristics and dingoes usually donot bark whereas most hybrids and domesticdogs do. Hybridisation between dingoes andother wild dogs is becoming more common.

The present distribution of dingoes andother wild dogs covers most of the mainlandexcept for the sheep and cereal growingareas of south-eastern Australia. They preferhabitats that have adequate free water andcover for concealment. Wild dogs live insmall groups or packs in territories where thehome ranges of individuals vary between 10and 300 square kilometres. However, thereis considerable overlap in the home rangesof pack members. Wild dogs eat a diverserange of foods, from insects to buffalo.Hunting group size and hunting strategiesdiffer according to prey type to maximisehunting success. Larger groups of wild dogsare more successful when hunting largekangaroos and cattle and solitary dogs aremore successful when hunting rabbits andsmall macropods. The main prey are: mag-pie geese, rodents and agile wallabies in the

Top End (Kakadu National Park); rabbits,rodents, lizards and red kangaroos in cen-tral Australia; euros and red kangaroos innorth-west Australia; rabbits in theNullarbor region; and wallabies, possumsand wombats in eastern Australia. In Asiamost dingoes have a commensal relation-ship with humans and mainly eat rice, veg-etables and other table scraps.

Female dingoes become sexually mature bytwo years, although some do not breed indroughts. Males in arid Australia also havea seasonal breeding cycle of about sixmonths. Gestation takes about 63 days andlitters of 1–10 pups (the average is 5) arewhelped in winter, usually in an under-ground den. Dingo pups usually forageindependently of their parents at 3–4months or, if in a pack, when the nextbreeding season begins. In eastern Australiapups may become independent at 6 monthsor stay in the family group for up to 12months.

2.1 General description

2.1.1 Size and coat colour

The average measurements of a maturedingo in Australia are: total length, 123 cen-timetres; shoulder height, 57 centimetres;head length, 22 centimetres; ear length, 10centimetres; hindfoot length, 19 centimetres;tail length, 31 centimetres; and weight, 16kilograms (Thomson 1992a; Corbett 1995a;).Males are universally larger and heavier thanfemales. Dingoes from northern and north-western Australia are larger than dingoes incentral and southern regions. All Australiandingoes are heavier than their Asian counter-parts (Corbett 1995a). Feral dogs may weighup to 60 kilograms (Korn and Fleming 1989)but are usually 11–24 kilograms (males), and10–22 kilograms (females) (Jones 1990).Free-roaming domestic dogs in Meek’s(1998) study near Jervis Bay ranged from15–31 kilograms.


2. Biology and ecology


Hybridisation may be obvious as in this case, but other hybrids may be hard to distinguish from pure dingoes, posingmanagement problems in conservation areas (Source: L. Corbett).

Feral dogs may weigh up to 60 kilograms (Source: P. Fleming).

In order of frequency of occurrence, the coatcolours of pure dingoes are: ginger (red tosandy), black and tan, white and black(Newsome and Corbett 1985). Most dingoeshave white points (white toes, white feet orwhite socks and a white tail tip) (Newsomeand Corbett 1985; Thomson 1992a). Accordingto early explorers’ reports, solid black dingoes,widespread throughout Asia, may once havebeen widespread in Australia (‘Collin’sVoyage’, undated c. 1790 and ‘Mr Gilbert’snote’, undated, cited in Gould 1863) but theyhave been rarely recorded in Australia inrecent times. However, animals from a popu-lation in the Victorian and South Australianmallee are consistently black with white points(P. Bird, Primary Industries and Resources,South Australia, pers. comm. 1999).

All other colours including sable (sandy withblack shoulders and back, commonly seen inGerman Shepherd breeds), brindle, patchyginger and white and patchy black and whiteindicate hybrids or domestic dogs. Ginger-coloured hybrids, usually resulting from hybri-dising with heelers, kelpies or collies, appearvery similar to pure dingoes and are oftenimpossible to distinguish on external features(Figure 3).

2.1.2 Longevity and methods toassess age of wild dogs

In the wild, dingoes can live for 10 years andferal dogs for at least 12 years, but most dieat about 5–7 years. Methods to estimate theage of wild dogs use: head length and eye-lens weight (Catling et al. 1991); the weightor length of bacula (penis bones — juvenile,immature and mature adult age classes); theeruption pattern of adult teeth (useful up to6 months); tooth wear (6–12 months), andthe annular cementum bands in the root tis-sue of teeth (dogs older than 12 months)(Jones 1990). Alternative methods use clo-sure of the foramen at the root tip of canineteeth to distinguish juveniles from older ani-mals; and for dingoes with closed root tipson all canines, the width of the pulp cavity ofcanine teeth distinguishes yearlings fromadults (Thomson and Rose 1992). Directobservation is also used to distinguish juve-niles from older animals.

2.1.3 Water needs

Wild dogs generally drink water every day,about one litre in summer and half a litre inwinter (Newsome et al. 1973). In winter in


Figure 3: The pelts of wild dogs, showing a variety of colours (Source: NSW Agriculture).

arid areas, when prey such as long-hairedrats (Rattus villosissimus) are plentiful, din-goes may live solely by absorbing waterfrom prey (Green 1973). Also, in arid centralAustralia, many weaned pups obtain most oftheir water from food. Females have some-times been observed to carry water in theirbellies to dens and regurgitate it for theirpups (Corbett and Newsome 1975).

During lactation, captive females havealmost no increase in their water intakebecause they ingest the faeces and urine oftheir small pups, thus recycling the waterthey contain as well as keeping the denclean (Green 1973). Wild-living dogs proba-bly have higher rates of water turnover thandogs fed by people because they need topursue and catch their own food.

2.2 Habitats

Prior to the arrival of Europeans, dingoesoccurred across all of mainland Australia butwere absent from Tasmania. They probablyoccupied all habitats, although at variousdensities. Today, the overall distribution ofdingoes has been reduced by the long histo-ry of control and exclusion fencing, particu-larly in the sheep grazing areas of the conti-nent (Figure 2). They are absent from themajority of New South Wales and Victoriaexcept for the eastern highlands and coast ofboth States, from the south-east third ofSouth Australia, and from most of the south-west tip of Western Australia (Figure 2).

‘Dingo numbers may have increased greatly in some

arid areas since European settlement as a result of the

pastoral industry, more watering points and the introduction of rabbits.’

Habitat use by wild dogs has not been stud-ied in detail. Their present distribution cov-ers the majority of mainland habitat types,and they are considered common across thisrange except for the arid eastern half ofWestern Australia and adjoining parts ofSouth Australia and the Northern Territorywhere they are thought to be naturally

sparse (Figure 2). Corbett (1995a) has sug-gested that dingo numbers may haveincreased greatly in some arid areas sinceEuropean settlement as a result of the pas-toral industry, more watering points and theintroduction of rabbits (Section 2.8.4).

On a smaller scale, wild dogs favour somehabitats more than others. These preferencesappear to be related to habitat features suchas prey distribution, presence of cover orother shelter and presence of water. In thehot, semi-arid Fortescue River region ofWestern Australia, for example, packs of din-goes spend proportionately more time inriverine areas than in other parts of theirrange (Thomson 1992d). This is likely to beassociated with the presence of water, thickercover and greater prey abundance in theriverine areas.

2.3 Diet and hunting strategies

2.3.1 Diet of wild dogs in Australia

The diet of wild dogs has been studied moreextensively than any other aspect of theirbiology. Over the past 30 years or so, almost13 000 stomach and faecal samples from sixmajor climatic regions have been analysed.Of 177 prey species identified, most weremammals (72.3% by occurrence, 71 species);others were birds (18.8%, 53 species), vege-tation (3.3%, mainly seeds), reptiles (1.8%,23 species) and an assortment of insects,fish, crabs and frogs (3.8%, 28 species)(Table 1; Corbett 1995a).

‘Dietary information may be misleading because wild dogs willkill stock without eating them andconversely because canids often

take livestock carrion.’

Despite the large range of prey eaten by wilddogs throughout Australia, almost 80% of thediet comprised only ten species. In order ofgreatest frequency these were: red kangaroo(Macropus rufus), rabbit, swamp wallaby(Wallabia bicolor), cattle, dusky rat (Rattuscolletti), magpie goose (Anseranas semipal-mata), brushtail possum (Trichosurus


vulpecula), long-haired rat, agile wallaby(Macropus agilis) and common wombat(Vombatus ursinus). Of these, only cattle(mostly as carrion) were eaten in each of thesix regions reported in Corbett (1995a). Thisnarrow range of major prey indicates thatwild dogs are specialists and not the oppor-tunistic generalists they are often assumed tobe. However, wild dogs do use a broadrange of hunting tactics involving solitaryand cooperative hunting in groups (Section2.3.4).

Fresh sheep and cattle rarely occur in thediets of wild dogs (Robertshaw and Harden

1985a; Lunney et al. 1990; Corbett 1995a).However, this may reflect the relatively lowabundance of livestock in many of the areaswhere samples were taken, sheep in particu-lar being few or at low density in most stud-ies. For arid and semi-arid cattle country, theabsence of cattle in the diets of wild dogsmay reflect the abundance and ease of pro-curement of native prey rather than the den-sity of cattle. Thomson (1992c) found thatsheep were an important component of thediet of dingoes on a pastoral lease in north-west Western Australia where merino sheepwere the main livestock enterprise. In con-trast, livestock were a minor component of


Table 1: The occurrence of major food groups (% of samples) in the diet of dingoes and other wild dogs in Australia(adapted from Corbett 1995a). f = faecal samples, s = stomach samples. Size categories from Brown and Triggs (1990).

Samples 6722 (f) 1480 (s) 131 (f/s) 413 (f/s) 2063 (f/s) 1993 (f)

Others 0.4 0.5 0.8 0.2 16.3 1.4

Largemammalsa

12.5 36.4 39.7 100 22.9 0.4

Mediummammalsb

26.6 41.7 69.5 4.8 72.6 85.5

Smallmammalsc

34.3 28.0 0 0.2 13.2 20.6

Reptiles 0.1 14.1 1.5 3.4 1.0 1.3

Birds 33.8 11.9 2.3 5.6 21.7 2.7

Insects 1.3 4.1 0.8 2.9 2.2 <0.1

Plants 7.3 0.1 0 0 1.2 0.2

Wet–drytropics

CentralAustralia

NullarborPlain

Fortescueregion(WA)

South-eastAustralia

North-eastNew South

Wales

aAnimals > 10 kilograms mean adult body weight.bAnimals 750 grams to 10 kilograms mean adult body weight.cAnimals < 750 grams mean adult body weight.

the diet of dingoes at two other sites wherethe land was vacant government land orunimproved pastoral leases running cattle(Thomson 1992c). Both Corbett (1995a) andThomson (1984a, 1992c) caution that dietaryinformation may be misleading, in relation toattacks on livestock, because wild dogs willkill stock without eating them (Section3.6.1). Conversely, the occurrence of theremains of livestock in the digesta and faecesof predators does not necessarily implicatethem as serious pests (Ginsberg andMacdonald 1990) because livestock kills canbe revisited and conversely because dogsoften take livestock carrion (Corbett 1995a).

2.3.2 Regional differences in diet

In tropical coastal regions of the NorthernTerritory, dusky rats, magpie geese and agilewallabies together form over 81% occur-rence in the diet (Corbett 1989) and there is aseasonal pattern of predation on them. Mostgeese are eaten as fledged young during thedry season. Dusky rats are eaten when inhigh abundance, about every fourth year.Agile wallabies are eaten throughout theyear but mainly in the wet season.

In arid central Australia, rabbits, small mam-mals and lizards are the main prey duringsequences of good rainfall years whereasduring droughts, dead cattle provide most ofthe diet (Corbett and Newsome 1987). Onthe Barkly Tableland, Northern Territory,where rabbits do not occur, no single nativespecies predominates in the diet except forlong-haired rats when they form hugeplagues, about once every nine years(Newsome and Corbett 1975; Corbett 1995a).

In north-west Australia, the large native mam-mals, red kangaroos and euros (Macropusrobustus), predominate in the diet, probablybecause of the paucity of small and medium-sized mammals, particularly rabbits(Thomson 1992c). However, on the NullarborPlains where both red kangaroos and rabbitsare abundant, rabbit occurs twice as frequent-ly as kangaroos (Macropus spp.) in the diet(Marsack and Campbell 1990).

In the Gulf region of Queensland, feral pigsand agile wallabies are important diet items.As in tropical Northern Territory, rabbits are

absent from this region. Macropods, includ-ing eastern grey kangaroos (Macropusgiganteus), red-necked wallabies (M. rufo-griseus), rat kangaroos (Potoroidae) andbrushtail possums were found commonly insamples collected in the Maranoa region ofcentral Queensland (L. Allen, Department ofNatural Resources, Queensland, pers.Comm. 1997; Allen et al. 1997).

‘In the Gulf region of Queensland, feral pigs andagile wallabies are important

diet items.’

In the cool temperate mountains of south-eastern Australia, the medium-sized walla-bies (swamp wallaby and red-necked walla-by) predominate in the diet of wild dogs inthe lower slopes, and the common wombatpredominates at higher altitudes. Brushtailpossums and ringtail possums(Pseudocheirus peregrinus) are also com-monly eaten (Corbett 1995a).

2.3.3 Dingo diet in south-east Asia

Very few dingoes in Asia live totally indepen-dently of humans and the main food of Asiandingoes is carbohydrate (rice, fruit and otherfood scraps) supplied by people or scav-enged. In rural areas of Thailand and northSulawesi, dingoes have been observed hunt-ing insects, rats, lizards and other live preyalong roadsides, rice paddies and in forests(Corbett 1988a).

2.3.4 Hunting strategies, huntingsuccess and anti-predatorbehaviour of prey

In Australia, as elsewhere, predation of live-stock by wild dogs conflicts with humanagricultural activity. The foraging behaviourand feeding ecology of wild dogs are ofinterest because they may affect the patternsof predation on livestock and the susceptibil-ity of wild dogs to control practices. Forexample, an abundance of alternative foodhas been cited as a possible reason for thefailure of strychnine baiting programs(Newsome et al. 1972).


The foraging behaviour of dingoes has beencharacterised differently by various authors(Whitehouse 1977; Robertshaw and Harden1986; Corbett and Newsome 1987; Newsomeand Coman 1989; Thomson 1992c). Theavailability of different prey species in differ-ent localities determines the dingo’s huntingstrategies. For example, the dingo may be:

• an opportunistic feeder (Whitehouse 1977)

• a selective predator of medium-sizedmacropods, such as swamp wallabies(Robertshaw and Harden 1986; 1989)

• a predator of large macropods (Thomson1992c), or it may

• combine a staple diet of medium-sizedmammals with a supplementary diet oflarger mammals and an opportunisticdiet of small mammals and carrion(Newsome et al. 1983a).

‘Alternative food has been cited as a possible reason for the

failure of strychnine baiting programs.’

When dingoes forage, the size of huntinggroups is determined by the type and abun-dance of prey (Thomson 1992c; 1992d). Iflarge prey such as kangaroos and cattle arebeing targeted, then dogs form large groupsto hunt (Corbett and Newsome 1987;Thomson 1992d). Schoener (1971) postulat-ed that hunting efficiency for large prey isincreased through group formation and thishypothesis is supported by Kruuk’s (1972a)study of spotted hyenas (Crocuta crocuta).Kruuk (1972a) found that individual hyenasbenefited from hunting in groups becausefor every doubling in the weight of preytaken, the number of hyenas that were fedincreased threefold. In addition, whencanids hunt in groups they are able to cap-ture larger prey (Van Valkenburgh andKoepfli 1993).

Hunting kangaroos and wallabies

Throughout Australia various species ofmacropods (mean adult weight 17–66 kilo-grams) are the most commonly killed prey(Corbett 1995a) and the tactics dingoes useto catch them are generally the same: theysight them, bail them up, then kill them.Groups of dingoes are more than three timesas successful at bailing up kangaroos andmore than twice as successful at killing themthan dingoes hunting by themselves. Forexample, in the Fortescue River region innorth-west Australia, the success rate for sin-gle dingoes bailing up and killing was 5.5%for red kangaroos and 33.3% for euros. Theequivalent success rate for groups of three ormore dingoes hunting together was 18.9%and 74.2% respectively (Thomson 1992c).

In contrast, Marsack and Campbell (1990)reported that all attacks on red kangaroosobserved in the Nullarbor region of WesternAustralia (sample size = 10) were by solitarydingoes. However, this may have beenunusual because the primary prey of dingoesin the Nullarbor region are rabbits and soli-tary wild dogs are more successful thangroups when hunting rabbits. Therefore,dingoes were probably hunting alone forrabbits when they saw a red kangaroo andtook the opportunity to hunt it.

Jarman and Wright (1993) reported observa-tions of attacks on eastern grey kangaroosby wild dogs at Wallaby Creek in north-east-ern New South Wales. The wild dogs theysaw with grey kangaroos were more often ingroups than the dogs seen without kanga-roos. However, this does not mean that wilddogs usually formed groups to hunt largemacropods in the area because wild dogswere seen in groups in only 21.3% of allobservations of wild dogs hunting kangaroos(Jarman and Wright 1993). Solitary wild dogswere also observed hunting red-necked wal-labies (Jarman and Wright 1993), the mostabundant medium-sized macropod atWallaby Creek (Southwell 1987).

The advantage of numbers when pursuinglarge kangaroos is that the leading dingooften makes the kangaroo change directioninto the path of other dingoes involved inthe chase, who in turn are skilled at cutting


corners; the combined effort soon exhauststhe quarry and it is easily bailed up(Thomson 1992c). This technique is similarto that used by wolves (Canis lupus) (Mech1970), hyenas (Kruuk 1972a) and Africanhunting dogs (Lycaon pictus) (Bekoff 1975).Another variation that African hunting dogsand dingoes sometimes use is the ‘relayhunt’, in which exhausted leaders arereplaced by following pack members.

The advantage of hunting in groups whenkilling large kangaroos is that wild dogs canmaintain relentless pressure on the quarryafter a successful chase. Since the leadingwild dogs become exhausted during thechase, the killing is done by other membersof the group coming along behind them. Thequarry can also be attacked simultaneouslyfrom several directions. Autopsies of kanga-roos killed by wild dogs (Thomson 1992c)suggest two patterns of attack: (1) nipping orhamstringing the hind leg to slow the kanga-roo sufficiently to attack its throat (on adultand juvenile kangaroos), and (2) runningalongside and biting the dorsal thorax andneck regions (on juveniles and small adultfemales).

Wild dogs are probably more successful athunting large kangaroos in more open areas,particularly arid areas where kangaroos tendto concentrate around permanent sources ofwater, especially during droughts (Corbett1995a). This can lead to surplus killing(Shepherd 1981). In contrast, hunting suc-cess is lower where kangaroos use the ter-rain to escape pursuing dingoes or hampertheir attack. In the dissected ironstoneranges of the Fortescue River region innorth-west Australia, 62.5% of unsuccessfulattacks by groups of dingoes on adult kanga-roos failed because the kangaroos backedup against natural barriers, preventing attackfrom the rear and reducing the attack to onlyone dingo at a time (Thomson 1992c).

Presumably, the success rate of dingoes andother wild dogs hunting grey kangaroos in thedensely forested regions of eastern Australia isalso low because of similar obstacles, but littleinformation is available. However, the appar-ent differences between hunting success inopen and closed habitats may be reflected inthe alert and flight distances of kangaroos in

these habitats. Eastern grey kangaroos(Jarman and Wright 1993) alert to dingoessooner than red kangaroos (Shepherd 1981)(means: 121 metres and 150 metres respective-ly) but both flee at similar distances (means: 98metres and 105 metres respectively).

‘Hunting tactics developed formacropods pre-adapted wild dogs

for hunting sheep and cattle.’

There are many anecdotal records of adultkangaroos and wallabies (Macropus spp.)taking to water to escape attack by dingoes(Bacon 1955). This ploy is unlikely to be suc-cessful if there is no escape route (as at asmall dam or waterhole) because the din-goes often wait for the quarry to emerge andthen continue the attack. Most reports ofkangaroos drowning dogs that enter thewater to attack them relate to domestic dogsrather than dingoes.

The tactics dingoes use to chase and killsmall and medium-sized macropods (meanadult weight 3–15 kilograms) are similar tothose they use for large kangaroos, but somedifferences are apparent. Dingoes huntingalone rely more on scent than sight to trailthe quarry, so they are often several hundredmetres behind them and the chase may lastfor several hours. The success rate of lonedingoes hunting medium-sized macropods isprobably greater than it is for lone dingoeshunting large macropods, but there are nodata to confirm this.

In north-eastern New South Wales,Robertshaw and Harden (1986) recordedthat female swamp wallabies pursued bydingoes eject their pouch young (greaterthan approximately 800 grams), apparentlybecause they are a burden when escapingpredators. However, there are no data onwhether or not more females escape by thisploy; it is only known that the populationage structure of wallabies is altered.

Most studies indicate that wild dogs killedthe young of large (mean adult weight >15kilograms) and medium-sized macropodsmore frequently than they killed adults. Forexample, in arid north-western New SouthWales, Shepherd (1981) recorded that almost


all of the 83 red kangaroos killed by a groupof five dingoes around a dam were juvenilesless than 18 kilograms (96.4%) and mostwere females (81.3%). In the moist forestedmountains of north-eastern New SouthWales dingoes also indicated a strong prefer-ence for dependent juvenile swamp walla-bies (32%), but not for gender (Robertshawand Harden 1985b, 1986). Several studieshave concluded that large male kangaroosare not routine prey for wild dogs (Shepherd1981; Robertshaw and Harden 1989; Jarmanand Wright 1993; Wright 1993).

Hunting tactics developed for macropodspre-adapted wild dogs for hunting sheep,cattle and other large animals. These tacticsmay also result in surplus killing of sheepand cattle.

Hunting rabbits and small prey

In central Australia, dingoes usually huntalone to catch rabbits, often by sighting andrunning them down, or by scenting andpouncing on a rabbit in a grass hutch. Theirgreatest success apparently comes fromknowing where to find rabbit warrens, espe-cially warrens containing rabbit kittens; anddingoes, in groups or alone, include suchwarrens in their regular hunting circuit(Corbett 1995a). Hunting success is also like-ly to be related to rabbit density and agestructure but there are few data; Corbett(1995a) observed a dingo hunting alone forone hour in a rabbit-infested area, with 80holes investigated, and two successful killsfrom four attempts.

Dingoes mainly rely on their hearing andsmell to find rodents, grasshoppers andother small prey moving about in grass andother vegetation. They capture these prey bypouncing on them.

Hunting birds

Many adult birds are captured when they aremoulting and unable to fly. Dingoes also eatyoung nestlings and newly fledged birds thatare easily captured. For example, at a coastalsalt lake in south-east Australia, a dingo ‘cor-nered’ six moulting black swans (Cygnusatratus) in a small shallow inlet. When theswans made a break for open water, a second

dingo rushed out of the fringing scrub andthey swiftly dispatched all six swans withbites across the back behind the wings; butonly four of the swans were eaten (Newsomeet al. 1983a). In the coastal wetlands of north-ern Australia, fledgling magpie geese form amajor part of the dingoes’ diet (Corbett 1989)and are captured in much the same way asswans are. However, healthy adult geese(mean weight 2.4 kilograms) are alsoobtained by stealing the kills of large preda-tory birds such as white-breasted sea eagles(Haliaaetus leucogaster) (Corbett 1995a).

Hunting cattle, buffalo, horses andsheep

The formation of hunting groups is usuallyessential when wild dogs hunt large ungu-lates such as cattle (Corbett and Newsome1987; Thomson 1992c), feral horses(Newsome and Coman 1989) and buffalo(Bubalus bubalis). Nearly all attacks on cat-tle and buffalo are aimed at young animals,from the newborn to subadults; healthy full-grown adults are rarely attacked. There arethree basic attack tactics (Corbett 1995a):

1. Constantly harass a mother with adependent calf. Solitary wild dogs orgroups keep the prey constantly on thealert, so it eventually tires or relaxes itsguard enough for a dingo to deliver acrippling bite.

2. Actively spook a mob of cattle to sepa-rate the calves from the adults, then focuson the calves.

3. Wait-and-watch tactic, observing and test-ing the most vulnerable members of aherd.

In the third tactic, a group of wild dogs, usu-ally sits, sometimes for hours, and watches aherd from a distance. They watch for cattleor buffalo that exhibit unusual behaviourthat might indicate vulnerability, such as dis-ease or injury. Common examples of suchbehaviour are: cows giving birth; individualsthat are stationary or not moving with themain body of the herd when grazing ordrinking; and calves running and frequently


bawling. Wild dogs will either attack orharass the targeted animal directly or waituntil it dies. If the targeted individual turnsout to be healthy, the wild dogs promptlyretreat and may resume waiting and watch-ing for other potential quarry.

Although wild dogs can kill cattle, there arefew data on how successful they are. In theFortescue River region of north-westAustralia, Thomson (1992c) recorded only 73interactions between cattle and dingoes overseven years. Most interactions (81%)involved cattle reacting defensively to din-goes and 26 (36%) involved attacks oncalves. However, only four calves werekilled, so the dingoes’ success rate was atbest 15% (4/26), or only 5% (4/73) if all inter-actions are counted.

‘The formation of huntinggroups is usually essential when

wild dogs hunt cattle.’

Cattle have a greater tendency to defendthemselves than do sheep, and trained cattlehave been used to protect sheep from preda-tors in New Mexico (Anon. 1987). Thegrouping exhibited by cattle acts as an anti-predator behaviour that reduces the risk ofan individual being taken and allows for ani-mals on the outside of a group to aggressive-ly defend the whole group (Alcock 1989). Insome areas of Australia, predation of calvesby wild dogs is believed to be minimised ifhorned cows are run instead of polled cows(E. Maskie, grazier, New South Wales, pers.comm. 1984).

Wild dogs evolved for running and they canoutrun both sheep (Thomson 1992c) andcattle (Alexander 1993). There are manyanecdotes about the dingo’s prowess athunting and killing sheep but the onlydetailed evidence is from Thomson’s (1992c)study in the Fortescue River region in north-west Australia. Thomson (1984a, 1992c)found that both groups and solitary dogskilled sheep, that dingoes often attackedmore than one sheep in a mob, and that inmost attacks the sheep were not killed out-right. From a total of 61 attacks, 26 sheep(43%) were seriously injured, but only eight

(13%) were killed outright. In 69% of attacks,dingoes broke off the attack before thesheep was killed, although in 31% of attacksthey then attacked another sheep. Sheepshowed no aggressive defensive behaviourand in 11 (18%) of the attacks they struggledso little that dingoes began eating thembefore inflicting a killing bite.

The reaction of sheep to the appearance of adog in their vicinity is to congregate in a mob(Kilgour 1985). Sheep chased by dogs oftenrun in a circle, the leading animals beingturned back to the mob by a pursuing dog.These behaviours make sheep more suscep-tible to surplus killing.

Scavenging and caching

Dingoes readily scavenge food, particularlycattle and kangaroo carcasses that becomeplentiful during drought in arid regions ofAustralia. For example, Corbett andNewsome (1987) studied dingo diet overseven years at Erldunda in central Australiaand found overall that cattle carrionoccurred in 6.3% of 285 stomachs comparedto remains of live cattle in only 2.1%. Morecarrion was eaten in droughts (mean 10%occurrence) than flush periods (3%).Dingoes usually gorge themselves on thecarcasses of large prey either in one feed orintermittently over several days until it is alleaten (P. Thomson, unpublished data1976–1984). Sometimes rival dingo groupsdispute over a carcass (Corbett 1995a).

‘Cattle carrion occurred in 6.3% of stomachs compared to

remains of live cattle in only 2.1%’.

Dingoes living on the coast regularly patrolthe beaches and scavenge fish, seals, pen-guins and other birds that are washed up(Newsome et al. 1983a). In Kakadu NationalPark, dingoes regularly scavenge theremains of prey underneath the nests andfeeding platforms of white-breasted seaeagles (Corbett 1995a).


In the Fortescue study, caching by dingoesof a small euro was observed (P. Thomson,unpublished data 1983). Captive dingoes atAlice Springs have also been observedcaching excess food, especially dog biscuits.In central Australia, wild dogs frequentlyreturned to carcasses, although these werenot buried or covered. Food remains at denswere often abundant and always conspicu-ously placed (L. Corbett, unpublished obs.).

2.4 Home range and movements

2.4.1 Home range

In common with most terrestrial mammals,individual wild dogs spend the majority oftheir lives within discrete areas or homeranges. Data on the boundaries and size ofhome ranges are usually derived from radio-tracking studies of collared animals. Withindiscrete social groups or packs, the homeranges of individual pack members overlapconsiderably (Section 2.5). The sizes of homeranges appear to be determined to a large

degree by the availability of resources. Whereresources are plentiful, individuals do nothave to roam far for food or water, and homeranges tend to be smaller. For example, radio-tracked adult wild dogs in the heavily forestedescarpment areas of the northern tablelandsof New South Wales had a mean home rangesize of 27 square kilometres, with consider-able overlap in the ranges of different individ-uals (Harden 1985). McIlroy et al. (1986a)found similar sized home ranges (21.5 squarekilometres) for wild dogs in a mountainousforest area of south-eastern Australia. Theseareas have a high rainfall, a temperate climate,and a large diversity of potential prey species(Newsome et al. 1983a; Robertshaw andHarden 1985a). In contrast, in the less produc-tive arid environment of the Fortescue Riverin north-west Western Australia and theSimpson Desert in Central Australia, dingoeshad larger home ranges, averaging 95.8square kilometres (Thomson and Marsack1992) and 67 square kilometres (Corbett1995a) respectively. In the Nullarbor region ofsouth-east Western Australia, where preypopulations fluctuated widely and water was


The social organisation of dingoes has implications for predation dynamics (Source: L. Corbett).

sparse, home ranges were larger still, rangingfrom 90 to 300 square kilometres (Thomsonand Marsack 1992).

‘The sizes of home rangesappear to be determined to a

large degree by the availabilityof resources.’

Home ranges are generally stable over time,although they may shift in response tochanges in food availability or social organi-sation. Individuals that begin to separatefrom social groups or packs, prior to dispers-ing from their natal home range tend to roamover larger areas (Thomson 1992d). In SouthAustralia, Bird (1994) estimated that 300–400dingoes were sharing a single watering pointin a particularly arid region. Eighty individualdingoes were simultaneously visible nearanother water source. These appeared to bemerely aggregations of individuals ratherthan social groupings, and probably reflect-ed the large area serviced by widely spacedwaters, and that dingo numbers were highbecause of a recent rabbit plague.

2.4.2 Movements

Harden (1985) carried out an intensive studyof short-term movements of dingoes in theescarpments of north-eastern New SouthWales and showed that dingoes exhibitedtwo distinctly different patterns of move-ment. The first was characterised by theintense use of a small area with manychanges in direction and was believed to beassociated with hunting. The second was amore directed movement that frequently tra-versed a large part of the home range. Oneof the functions of the latter might have beento maintain communication between individ-uals by regular visits to scent posts (Section2.5.2).

In Harden’s (1985) study, the wild dogs wereactive throughout the day with peaks ofactivity at dawn and dusk. Hourly rates ofmovement were equal for diurnal and noc-turnal periods and the average distancemoved in 24 hours was around 15 kilome-tres. Wild dogs spent 65% of the day activeand 35% resting. Periods of activity were

short (65% less than 60 minutes) and inter-spersed with shorter rest periods (70% lessthan 30 minutes). In contrast, in the hotter,less vegetated environment of north-westWestern Australia, dingoes tended to beinactive during the middle of the day, withmajor peaks of activity and movement occur-ring around dawn and dusk (Thomson1992b). Similar patterns of activity have beenrecorded for dingoes in the Simpson Desert(Corbett 1995a) and coyotes (Canis latrans)living in hot environments in North America(Andelt 1985).

‘Dispersal appears to be related to food supply and mediated by social factors.’

In north-west Western Australia, dingoes didnot regularly travel large distances acrosstheir ranges (mean maximum range widthwas 10.5 kilometres) (Thomson 1992d).Movements were localised, and packs tend-ed to focus their activities in a particular sec-tion of their range for a time, with occasionalforays into other parts of their range. Thispattern was consistent through most of theyear. The major seasonal factor influencingmovement patterns was related to the raisingof pups. This activity dramatically affectednursing individuals, restricting their move-ments, but other pack members were affect-ed as well.

Dingoes displayed strong site fidelity, sel-dom engaging in forays far beyond theirhome ranges or territories (only 1.2% of allradio-locations represented foray move-ments) (Thomson et al. 1992a). Most foraymovements were within four kilometres ofthe territory boundaries and rarely exceededsix kilometres. There was no seasonal pat-tern to these movements. Males were morelikely than females to engage in forays andjuvenile dingoes only engaged in forays inthe company of older dingoes. Some foraysappeared to be associated with subsequentdispersal movements.

Dispersal represents a move that places ananimal permanently beyond the area that itnormally occupied, and usually involvesmovement out of the natal home range.


Dispersal appears to be related to food sup-ply and mediated by social factors. In north-west Western Australia, individuals as well asgroups dispersed (Thomson et al. 1992a).Males tended to disperse further thanfemales and had a higher incidence of dis-persal. One male dingo in arid centralAustralia travelled 250 kilometres (Newsomeet al. 1973). Dispersal distances in the studyby Thomson et al. (1992a) ranged from1–184 kilometres, with a mean of 20 kilome-tres. The incidence of dispersal was highestwhen the density of the dingo populationwas high and food supply low, and wasfacilitated by the availability of nearby vacantareas. On the Nullarbor Plain, dingoes tend-ed to move further than those in north-westWestern Australia particularly as dams driedup. Less than 10% of dingoes moved morethan 50 kilometres and the longest move-ment was 250 kilometres (Thomson andMarsack 1992). In South Australia, the twolargest males of 59 radio-collared individualsdispersed 150 and 225 kilometres respective-ly after a dam they were watering at dried up(Bird 1994). Dingo numbers were very highthroughout the region at the time and foodsupply low.

Meek (1999) reported that radio-collaredfree-roaming dogs travel 8–30 kilometres tohunt macropods in bushland adjacent to thehomes of their owners. Coman andRobinson (1989) reported that commensaldogs from the Victorian city of Bendigo trav-elled smaller distances, usually less than 6kilometres. Most of these forays were intoadjacent agricultural land where livestockwere attacked in 51 of 84 forays. The man-agement implications of wild dog move-ments are discussed in detail in Chapter 6.

2.5 Social organisation andbehaviour

2.5.1 Wild dog society

Dingoes and other wild dogs, like their wolf(Canis lupus ssp.) forebears, are social ani-mals. Where conditions are favourable, theyform stable packs that maintain distinct terri-tories that overlap little with neighbouringpacks (Thomson 1992d). However, regional

variations are seen, reflecting the flexiblenature of dingo social structure (Green andCatling 1977; Newsome et al. 1983a; Corbettand Newsome 1987; Bird 1994). This flexibil-ity is not surprising in view of the wide vari-ety of habitats, prey species, climatic condi-tions and levels of human exploitationencountered across Australia. Studies byRobertshaw and Harden (1986), Corbett andNewsome (1987) and Thomson (1992d) sup-port the notion that increasing specialisationon larger prey such as wallabies and kanga-roos favours increasing sociality and the for-mation of larger groups. Corbett (1995a)concluded that the primary function of dingopacks is to defend hunting areas and otheressential resources.

The social organisation of dingoes in theFortescue River area of north-west WesternAustralia provides an example of dingo soci-ety in a relatively undisturbed ‘wilderness’area, where euros and kangaroos were themain prey and natural water was widely dis-tributed (Thomson 1992d). Here, dingoeswere organised into stable packs occupyingdiscrete territories that overlapped little withthose of neighbouring packs. Packs com-prised a dominant male and female and theiroffspring of various ages. Territory bound-aries were stable over time and between-pack encounters were extremely rare. Packsvaried in size (mean monthly pack sizes forfive packs ranged from 3–12 individuals),with smaller packs tending to be found inthe poorer areas and occupying largerranges. Pack members cooperated to huntprey and took part in communal activitiessuch as feeding (Section 2.3.4), resting andraising pups. Pack members were not alwaystogether at one time; they usually operatedin sub-groups that were flexible in size andcomposition. Lone dingoes were sometimesidentified; these individuals displayed nopack affiliations, occupied large ranges thatoverlapped the mosaic of pack territoriesand avoided encounters with packs. Theywere probably individuals in varying stagesof dispersal, seeking a mate and a vacantarea in which to settle.

Corbett (1995a) reported a more fluid situa-tion at Kapalga, in the northern tropicalNorthern Territory. Here, stable packs occu-


pied territories, but they altered where andwhat they hunted according to season andprey availability. In the Simpson Desert,another wilderness area, the necessity ofsharing sparsely distributed watering pointsmeant that distinct spatial separation ofpacks was not possible. Howling and otherforms of communication (Section 2.5.2) weremore pronounced in these situations, allow-ing a temporal separation of packs (Corbett1995a). When prey (mostly small mammals)were more abundant and widespread aftergood rains, dingoes tended to spread out toexploit these resources. Thus, packs werenot as constantly stable or confined to suchdefined areas as in the more predictablenorthern environments.

‘The influence of domestic dogs on the social structure ofwild-living dogs, dingoes and

hybrids in Australia is not documented.’

Research in the arid pastoral regions of cen-tral Australia revealed that most dingoeswere seen alone, although they were looselybonded in small amicable groups sharing acommon living area (Corbett 1995a). Rabbitswere a major dietary item and individualstended to hunt alone (Section 2.3.4). Theranges of different groups tended to overlapconsiderably, sharing common resourcessuch as the relatively sparse watering points.A similar pattern was observed in a relativelyundisturbed pastoral region on the NullarborPlain in Western Australia (Thomson andMarsack 1992, and unpublished data1982–1987). Here, dingoes preyed principal-ly on rabbits. Water was sparse and sharedby adjacent social groups and there was noevidence of strong territoriality. Groupingsof dingoes were most commonly seen dur-ing the months leading up to mating, duringthe raising of pups, and when dingoes werefeeding on large prey (kangaroos) (Section2.3.4).

The large degree of overlap in the homeranges of wild dogs in north-eastern NewSouth Wales (Harden 1985) suggests that thewild dogs studied were members of a singlepack. Free-roaming domestic dogs in south-

eastern Australia have also been reportedhunting or foraging in groups (Meek 1999).Feral dogs in Baltimore, North America alsoformed groups of 2–17 animals althoughapproximately 51% of dogs were seen alone(Beck 1973). Free-roaming dogs in centralVictoria were usually seen in pairs (54% ofsightings) or singly (34%), with packs ofthree to seven dogs being seen in 12% ofsightings (Coman and Robinson 1989). Theinfluence of domestic dogs on the socialstructure of wild-living dogs, dingoes andhybrids in Australia is not documented.However, it is likely that the factors thatinfluence the social organisation of dingoesin different areas would also influence thesocial behaviour of feral dogs or hybrids.

2.5.2 Communication

Howling

Vocal communication is important for din-goes and other wild dogs because they areoften spatially separated. Dingoes do notbark in the wild as domestic dogs andhybrids do (Corbett 1995a) but howling iscommon to all wild dogs.

There are three basic types of howl used,moans, bark-howls and snuffs, and thesehave at least 10 variations (Corbett 1995a).Howls travel over large distances and havethe purposes of locating other wild dogs,attracting pack members and repellingintruders from the pack home range. The fre-quency of howling varies throughout eachday and throughout the year, and is affectedby breeding, dispersal, lactation, social sta-bility and dispersion. Wild dogs often huntalone and in times of food shortage, packmembers may become more widely dis-tributed within their home range. Howling ismore pronounced at these times (Corbett1995a).

Scent Marking

Dogs have a highly developed sense of smellthat they employ in social communication.They use chemical signals originating inurine, faeces and scent glands (Ralls 1971).Scent marking in canids is oriented to specificobjects, elicited by familiar conspicuous land-


marks and novel objects or odours, andrepeated frequently on the same object(Kleiman 1966). Scent marking probably orig-inated as a means to familiarise and reassureanimals when they entered strange and fright-ening situations. Bringing together of thesexes and pack members and maintaining ter-ritory are probably secondary functions thatare important for survival (Kleiman 1966).

‘Preliminary work has been conducted to find attractants thatcould be used to lure wild dogs to

traps or poison baits.’

There is much evidence that dingoes and wilddogs defecate and urinate on objects at partic-ular sites to communicate with members oftheir own and rival packs (Thomson 1992b;Corbett 1995a). The most frequently recordedobjects that wild dogs mark are grass tussocks,small bushes, logs, fence-posts, rocks and thefaeces of other animals. Most scent posts arelocated at shared resources such as water inarid areas, hunting grounds, trails and roads,particularly intersections. Males scent-mark,

use raised-leg urination and rake the groundafter marking more frequently than females(Corbett 1995a). In north-west and centralAustralia, raised-leg urination and ground-rak-ing peaked in the mating season (Thomson1992b; Corbett 1995a). One of the functions ofscent-marking may be to synchronise repro-duction between pairs, as has been suggestedfor wolves and coyotes (Rothman and Mech1979; Wells and Bekoff 1981).

Another form of scent marking is scent rub-bing whereby an animal rolls on its neck,shoulders or back. Scent sources that elicitscent rubbing are associated with food,chemicals, catmint, urine, faeces and carcas-es, and scent markings of conspecifics(Reiger 1979). The function of scent rubbingis unknown but one hypothesis is that it mayincrease the social attractiveness of a particu-lar animal (Fox 1971). Dingoes and otherwild dogs are known to scent rub this way(Corbett 1995a).


In central Australia, dingo dens are commonly developed from enlarged rabbit burrows (Source: L. Corbett).

Scent-marking may be used by managers toassess the numbers of canids (Section 6.2.2).For example, the United States Fish andWildlife Service have developed a scent-sta-tion index method to assess trends in carni-vore populations (Roughton and Sweeny1982). This method has not been used inAustralia for assessing dog numbers but pre-liminary work has been conducted to findattractants that could be used to lure wilddogs to traps or poison baits (Jolly and Jolly1992a,1992b; Mitchell and Kelly 1992). Forexample, Jolly and Jolly (1992b) found thatabout half of ten captive wild dogs respond-ed to six chemical attractants in pen and fieldtrials which suggests that less expensive andbetter controlled pen trials might find aneffective attractant to lure wild dogs to trapsand poisons. Those trials were conducted inNovember and better responses might beobtained from trials in the breeding season.

2.6 Reproduction

2.6.1 Breeding: dingoes

The pattern of breeding has implications formanaging dingo predation on cattle (Section3.6.5).

Dingoes produce only one litter of pupseach year, and except for tropical habitats,litters are usually whelped in winter. Thisbreeding pattern is determined by thefemale’s annual oestrus cycle as males arecontinuously fertile in most regions (Catling1979). The precise onset and extent ofbreeding varies with age, social status, geo-graphic latitude, seasonal conditions andwhether animals are pure dingoes or hybrid(Jones and Stevens 1988; Corbett 1995a).

In arid central Australia, most wild femalescommence breeding when they are twoyears old and usually mate in April–May,about one month earlier than most of theircounterparts in southern temperateAustralia. In stable packs, the most dominant(alpha) female, usually the oldest, tends tocome into oestrus before subordinatefemales. Some of the subordinate femalesundergo pseudopregnancy (Corbett 1988b).During droughts, all young females less than

one-year-old and some older females do notbreed at all, and for those adults that do, theonset of breeding is delayed by about sixweeks (Figure 4).

Pro-oestrus and oestrus periods for captivedingoes in central Australia and domesticdogs, as determined from vaginal smears,last about 10–12 days (Corbett 1995a).However, in the Fortescue River region,behavioural data suggest that pro-oestrusmay last about 30–60 days (Thomson1992b).

Males reach full sexual maturity at 1–3 years.Although continuously fertile, males in hotarid regions appear to be true seasonalbreeders with maximum testis weightsoccurring about April, the peak of the matingseason (Catling et al. 1992). In contrast,males in the cooler temperate highlands ofsouth-east Australia can breed throughoutthe year and many successfully sire pupswith domestic bitches outside of the dingobreeding season (Catling et al. 1992).

‘Seasonality in breeding increases with latitude.’

Gestation lasts 61–69 days in captive din-goes. The average litter size for dingoes isfive (range 1–10) throughout Australia andThailand, and usually more males are bornthan females (Corbett 1995a).

Seasonality in breeding increases with lati-tude. In temperate regions of Australia, mostdingo litters are born in winter (June–August).Litters born in summer and autumn(November–April) in south-east Australia,are almost certainly from hybrid femaleswhereas in northern Australia, litters of puredingoes have been recorded in most monthswith a peak in July (Corbett 1995a).

2.6.2 Breeding: feral dogs andhybrids

Female feral dogs and hybrids of similar sizeto dingoes may have two oestrus cycles eachyear, although it is unlikely that they success-fully breed twice every year (Jones andStevens 1988). Evidence from south-eastAustralian populations of wild dogs indicates


that the regular birth pulse of dingoes is dis-rupted where there is a large proportion offeral dogs and hybrids present (Jones andStevens 1988). Male domestic dogs andhybrids do not show any seasonal changesin testicular activity and are fertile through-out the year (Jones and Stevens 1988).Gestation is 58–65 days for hybrids anddomestic dogs in the same climate (Corbett1995a). The average litter size for hybridsand domestic dogs is similar (about 5 pups)but varies with breed (Spira 1988).

2.6.3 Characteristics of dens

In central Australia most dens have beenrecorded in enlarged rabbit-holes, caves inrocky hills, under debris in dry creek beds,under large tussocks of spinifex, among pro-truding tree roots and under rock ledgesalong water courses (Corbett 1995a).

In the Fortescue River region, most denswere recorded in cave complexes in hilly ter-rain; others were in hollow logs, underspinifex or in enlarged goanna (Varanusspp.) holes (Thomson 1992b). Similarly, intropical northern Australia, most dens werefound in enlarged burrows of large goannas,with others under fallen trees and other

debris (Corbett 1995a). Most dens in theFortescue River region faced the south-east,possibly to minimise excessive heating fromthe sun (Thomson 1992b).

In contrast, in the cooler eastern highlands,most dens have been reported in hollowlogs, old wombat burrows and occasionallyin caves under rock ledges (Corbett 1974;Jones and Stevens 1988) and commonlyfaced north (B. Harden, unpublished data1970–84).

2.7 Mortality and disease

In south-eastern Australia, pups becomeindependent at 6–12 months (Harden 1981).Those that become independent and dis-perse at six months are less likely to survivethan those that stay with the parents for 12months (B. Harden, unpublished data1970–84). The higher survival may be due tothe increased hunting success that dogs ingroups may have with larger prey (Thomson1992c) or through the adults coaching thepups in hunting (Corbett and Newsome1975). Coaching of pups has been observedin captive-bred hybrids in north-east NewSouth Wales (T. Kempton, grazier, New


Adult, flush seasonYoung, flush seasonAdult, droughtYoung, drought

Mea

n ut

erin

e w

eigh

t (g)

D

Month

0

20

40

60

80

100

120

140

J F M A M J J A S O N

Figure 4: The breeding cycle of adult (more than one-year-old) and young (less than one-year-old) female dingoes incentral Australia (adapted from Catling et al. 1992). In flush seasons, most adult females are heavily pregnant by May–July,but not all young females become pregnant, as indicated by their low uterine weights. In droughts, young females do notbreed at all, and breeding is delayed by about six weeks for older females that do breed.

South Wales, pers. comm. 1979). In the sta-ble packs of the Fortescue River region, mostjuvenile dingoes do not become indepen-dent and disperse in the first year, but stay inthe social group, sometimes permanently.

2.7.1 Dispersal sinks

One of the main, consistent causes of dingomortality is a cycle involving dingo popula-tion density, food supply and human con-trol. When food becomes scarce for a largepopulation of wild dogs in a ‘safe’ area(source), they disperse, singly or in groups,to resource-rich, pastoral and agriculturalareas where there are fewer wild dogs.These are ‘danger’ areas, where human con-trol measures are intense. Wild dogs are poi-soned, trapped or shot here, creating vacantareas (sinks) and perpetuating the dispersal-mortality cycle.

For example, over nine years in theFortescue River region, 25 lone dingoes andsix groups (comprising 24 individuals fromfractured packs) dispersed from non-pastoral‘safe’ areas to sheep paddocks where controlwas intensive. All except one of these 49 din-goes (98%) were eventually killed — eithertrapped, poisoned or shot. This was twicethe average mortality for dingoes dispersingto other areas (Thomson et al. 1992a).

2.7.2 Human-induced mortality

Humans have had a major impact on wilddog populations through habitat manipula-tion, the introduction of domestic dogs andcontrol practices. For example, habitatchanges associated with agriculture in south-ern Australia caused the demise of dingoesin these areas. Effective control programs intargeted areas of the Fortescue River regionremoved most dingoes (Thomson et al.1992b). Human-induced mortality may alsobe high for wild dogs in areas of southernAustralia where coordinated control pro-grams operate annually. Although the effec-tiveness of wild dog control programs varies,severe mortality can be inflicted on wild dogpopulations (Fleming et al. 1996). The effectof trapping, poisoning and shooting on wilddog populations is discussed in detail inChapter 6.

2.7.3 Parasites, diseases and associated causes of death

Wild dogs are susceptible to all the diseasesthat affect domestic dogs. Thirty-eightspecies of parasites and pathogens have sofar been recorded in wild dogs. In Australia,at least another 50 infectious organisms indomestic dogs have been recorded (Seddon1965a,1965b; 1966; 1967; 1968; Seddon andAlbiston 1967; Kelly 1977; Appendix A), allof which could potentially become estab-lished in wild dog populations. The condi-tions under which establishment might hap-pen and the characteristics of disease organ-isms that would allow establishment andpersistence of these diseases require furtherevaluation. Appendix A is a detailed list ofdiseases affecting wild dogs.

‘Wild dogs are susceptible to all the diseases that affect

domestic dogs.’

In most cases, diseases have little effect onthe survival of adult wild dogs. Exceptionsinclude: canine distemper (Paramyxovirus),hookworms (Unicinaria stenocephala andAncylostoma caninum) and heartworm(Dirofilaria immitis) in northern Australiaand south-eastern Queensland (Coman1972a; Corbett 1995a, Meek 1998; L. Allen,Department of Natural Resources, Queensland,pers. comm. 1989).

The effects of most diseases and parasites onwild dog mortality rates and morbidity rateshave not been measured. The proportion ofa population dying during a given time inter-val is the mortality rate. Morbidity rate is theproportion of a population affected by dis-ease in a given time interval. The debilitatingeffects on wild dogs are unknown for 45% ofdingo parasites and pathogens, but 29% (11of 38) are known to be fatal, especially inpups (7 of 11 fatal infections). Lungworm(Oslerus osleri), whipworm (Trichurusvulpis), hepatitis (Adenovirus), coccidiosis(Isospora rivolta and Eimeria canis), lice(Trichodectes canis and unidentifiedspecies) and ticks (Ixodes holocyclus,Rhipicephalus sanguineus and Amblyommatriguttatum) can kill pups and thereby


decrease recruitment to dingo populations.Heartworm virtually eliminated dingo popu-lations at Kapalga (Kakadu National Park) inthe 1980s and distemper decimated BarklyTableland populations in the early 1960s andagain in the 1970s (Corbett 1995a).

Sarcoptic mange (causal agent Sarcoptesscabiei) is a widespread parasitic disease indingo populations throughout Australia butit is seldom debilitating. The highest preva-lence (20%) was recorded in the FortescueRiver region where 21% of mange-affectedadult dingoes were in poor condition com-pared to 5% for mange-free dingoes; howev-er, only one dingo was suspected to havedied from mange (Thomson 1992b). Maleswere twice as likely to be affected by mangeas females (Thomson 1992b) and it is likelythat mange is associated with particular preyspecies (Corbett 1995a).

Hydatidosis (caused by the cestodeEchinococcus granulosus) results in seriousillness in infected humans and in the devalu-ation of infected livestock carcases at slaugh-ter (Section 3.4). E. granulosus infectiondoes not cause serious illness in dogs.

In central Australia, wild dogs mostly eat rab-bits and consequently have relatively highinfestations of rabbit stickfast fleas(Echidnophaga myrmecobii) and tapeworm(Taenia pisiformis). By comparison, the mostcommon parasites of wild dogs in the nearbyBarkly Tableland are lice and an unidentifiedtapeworm species which are both probablyderived from eating long-haired rats, a majorprey species in this region. Also, sarcopticmange is usually associated with plagues ofdusky rats in the wet–dry tropics. In south-eastern Australia, kangaroo flies (Hippoboscidspp.) were often found on free-roamingdomestic dogs that frequently hunted macrop-ods (Meek 1998). These insects are most likelyto cause irritation rather than death.

2.7.4 Other causes of death

Where irruptions of rabbits are common, forexample, in arid South Australia and theNullarbor region of Western Australia, manywild dogs die from starvation when theirruption is over. Severe drought and concur-rent declines in water availability and prey,

particularly rabbits, are also significant caus-es of dingo mortality.

Other less common causes of mortality fordingoes and other wild dogs include: beingrun over by vehicles or horses, being chasedand killed by people on horseback (Hardenand Robertshaw 1987); buffalo and cattlegoring and kicking; snake bite; and preda-tion on pups by wedge-tailed eagles (Aquilaaudax). Unlike in north-east Thailand,where about 200 dingoes are butchered eachweek (Corbett 1985), human predation ofdingoes and other wild dogs for food nolonger occurs in Australia.

2.8 Population dynamics andchanges in abundance

2.8.1 Overview of predatorpopulation dynamics

In contrast to the situation with dingoes, theregulation and dynamics of populations ofother social canids such as wolves have beenextensively studied (Mech 1986; Petersonand Page 1988). Food supply is thought tobe the principal common natural factoraffecting canid social organisation, groupstability, dispersal strategies, mortality andreproductive success (all in turn influencingpopulation dynamics). The supply of prey,predominantly herbivorous animals, isaffected by environmental conditions (espe-cially the extent and pattern of rainfall), aswell as being influenced by the predatoritself (Section 2.3). When the effects of peo-ple on the environment are added (alteredlandscape, altered fire regimes, introducedexotic animals, increased watering points,imposed control on predators and some-times prey species), it follows that thedynamics of predator populations can bevery complex.

2.8.2 Dynamics of wild dog populations

The most intensive study of the populationdynamics of wild dingoes was carried out inthe Fortescue River region of WesternAustralia (Thomson 1992a). Dingoes were


radio-tracked over a nine-year period in anarea with minimal human disturbance. At thestart of the study, dingo numbers were rela-tively low because of control work (trappingand baiting) that had been undertaken in thearea. As the study progressed, dingo num-bers increased to high levels (Figure 5).Dingoes at that time had access to a moder-ate and apparently adequate supply of theirprincipal prey, kangaroos. The area was fullyoccupied by territorial packs. The popula-tion rose as a result of reproductive successand consequent increases in the size of thepacks. At the high population level, dingoesbegan to disperse, resulting in a subsequentfall in population size.

In the second period (1979–80, Figure 5),population density was still high, but fellmarkedly when dingoes were deliberatelyexposed to aerial baiting during research tri-als (Thomson 1986). At the time of baiting,kangaroo numbers were low and dingoeshad begun to increase their use of otherfood, including cattle carrion and smaller

prey species. The earlier signs of social per-turbation, indicated by an increase in disper-sals, were followed by other more dramaticchanges prior to baiting. Large packs beganto dissociate, dispersal continued, territoryshifts were recorded, and signs of changes inactivity patterns emerged. These featureshave also been recorded in populations ofwolves facing food shortages (Messier 1985).

‘Monitoring to detect and act onsuch circumstances is clearly asensible option for managers.’

After baiting removed almost all wild dogs,re-colonisation from adjacent unbaited areaswas rapid (Figure 5). The population rose,though not to the extent that occurred duringthe first period of the study. The new immi-grants occupied large areas, and in somecases, their reproductive potential wasunknown. When breeding increased, youngdingoes were able to establish new territories


Dingoes may be attracted to urban settlements in remote areas (for example, Moomba Tip in northern South Australiapictured above) by the prospect of easy food. Such scenarios are likely to increase the rate of hybridisation with domesticbreeds and may also create a public nuisance (Source: P. Bird, Animal and Plant Control Commission, SouthAustralia).

at the edge of their natal range. This meantthat pack size did not necessarily increase asit did during the first period of the study.There were now vacant areas available inwhich wild dogs could settle, unlike the earli-er period, when the area was fully occupied.

There are several important implications forwild dog management in these findings(Section 6.4.4). One of the most obvious isthat populations of wild dogs within closeproximity to livestock pose the greatest threatwhen their numbers are high and their foodsupply is limited. Movements into grazingareas are likely at these times. Monitoring todetect and act on such circumstances is clear-ly a sensible option for managers (Chapters 6and 7). The research findings from theFortescue River study also support the strate-gy of maintaining a buffer zone adjacent tostocked areas, where control work is carriedout to remove resident wild dogs and create a

‘dispersal sink’ (Section 6.4.4). This reducesthe risk of wild dogs moving from the adja-cent area into the paddocks, and also pro-vides an area for wild dogs dispersing fromfurther out to settle before moving intostocked areas (Chapter 6).

A study by Fleming et al. (1996) in north-eastern New South Wales suggested that thedynamics of wild dog populations underregular high levels of control by aerial bait-ing were greatly influenced by the annualnature and timing of the baiting. A conceptu-al model of population dynamics underannual aerial baiting (Figure 6) shows thatthe sink caused by annual population reduc-tion is filled by the time of the next annualbaiting. There was evidence that the newdogs in the area were both young dogs bornlocally and immigrants. The time at whichrepopulation occurred was not determined.Repopulation and its timing obviously affectdecisions about the timing of control pro-grams (Section 6.4.4).


Den

sity

(din

goes

/100

km

2 )

Cen

sus

area

(km

2 )

20

15

10

1976

1977

1978

1979

1980

1981

1982

1983

1984

5

25

0

350

300

250

200

400

150

Figure 5: Fluctuations in dingo density (•) in theFortescue River region 1976–84. Census data excludepups but all other dingoes known to have been presentat any time during a particular quarter were counted.Dissociation of dingoes from packs and a paucity ofground survey work precluded an accurate tally in somequarters (••). Population size was probably underesti-mated in these periods. The histogram depicts the sizeof the area censused in each quarter. Arrows indicatewhen baiting took place (after Thomson et al. 1992b).

Abu

ndan

ce

3

Months

21 4 5 6 7 8 9 10 11 12 13 14

Births

Aerial baiting

Aerialbaiting

Immigration

Figure 6: A conceptual model of the dynamics of apopulation of wild dogs in an area exposed to annualbaiting programs (Month 1 is March) (after Fleming1996a).

2.8.3 Social factors limiting population growth

A number of factors influence dingo num-bers and affect their population density(Section 2.7). Breeding and immigrationboost numbers and mortality, and emigra-tion depletes numbers. Mortality factors mayinclude disease, starvation and the directeffects of control actions. The type of foodavailable, its abundance, and its dispersionall affect the social dynamics of dingo popu-lations, and this in turn affects populationdynamics (Section 2.3). Some of these socialfactors are outlined in this section.

In most stable packs, only one litter of pupsis generally raised (Thomson et al. 1992b;Corbett 1995a), although several females inthe pack may be capable of breeding andmay even whelp. Social factors, includingdominant female infanticide (Corbett 1988b),appear to limit the production or raising ofadditional litters. In a three-year study of acaptive dingo pack, Corbett (1988b) foundthat less than half the potential increase indingo numbers was realised, due to inhibi-tion of breeding in subordinates, matingpreferences by dominants, infanticide andthe killing of subordinate females. In captivewolves, dominant males and females aggres-sively prevent subordinates from sexualsoliciting and mating activities (Zimen 1976).

An important consequence of infanticide bydominant females is that subordinatefemales help to rear and even suckle the off-spring of the dominant female. This repro-ductive strategy ensures the dominantfemale’s offspring not only receive addition-al care from subordinate helpers, but alsohave no competition for resources from theoffspring of other females. Infanticide inAustralian dingoes may have evolved inresponse to widespread drought and firewhen packs are forced to split into smallerunits to survive on the patchy and scarceresources available. The more pups born,the greater the chance that some would sur-vive adverse periods. Since most breedingdingoes would have been closely related, atleast some of the dominant pair’s geneswould survive to the next generation if allpregnancies went to term and if some of the

smaller social groups were able to survive.Once groups reformed, the dominant din-goes’ genes would be better favoured byinfanticide.

‘In most stable packs, only one litter of pups is

generally raised.’

Breeding success can be affected in a lessdirect way by food supply and the socialcontext. In the Fortescue study (Thomson etal. 1992b; referred to in Section 2.3.4), din-goes appeared to have a lower chance ofsuccessfully raising litters when acting alone.The absence of helpers to provide additionalfood for mother and pups probably con-tributes to poor pup survival, as does theinability of solitary dingoes or pairs to suc-cessfully hunt kangaroos, the primary preyin the area. Larger groups are more success-ful at hunting kangaroos (Section 2.3.4), anadvantage in providing food for the packand its offspring. When sheep, an easy preyfor solitary dingoes, became available to twobitches who were each caring for pups large-ly on their own, both litters were successfullyraised (Thomson et al. 1992b).

2.8.4 Changes in abundance sinceEuropean settlement

After dingoes were transported to Australiaby Asian seafarers about 4000 years ago,their numbers would have been kept downby food supply, social factors and diseasessystematically and periodically reducingpopulations. However, following Europeansettlement and the development of the pas-toral industry in inland Australia about 100years ago (Bauer 1983), dingo numbers rosedramatically as food such as rabbits, stockand some macropods became more plentifuland dams and artesian bores increased watersupply during droughts. In many of the moreclosely settled areas of Australia, however,changes to the environment and intensivecontrol of dingoes by shooting, trapping andpoisoning largely eradicated dingoes.Dingoes are now absent from large tracts ofagricultural and grazing land in southernAustralia. Dingo control over the extensive


pastoral areas of inland Australia was proba-bly not as effective, although the erection ofbarrier fencing probably tipped the balancein favour of graziers in some parts ofAustralia (Section 5.1.1).

‘The dingo population probablypeaked during the 1930–50s and

since then has remained highbut the proportion of hybrids

has increased greatly.’

In some instances, haphazard control workprobably was largely inefficient, in manycases only fracturing dingo packs into smallerunits, each with a breeding female. In thisway, the dingo’s natural population suppres-sion method, dominant female infanticide(Corbett 1988b), was discouraged, so thatdingo numbers could increase abnormallyduring flush periods. Further, most of thevictims of dingo control programs wereprobably young and very old dingoes. Thismeant that surviving populations tended toconsist mainly of middle-aged dingoes, theage group most likely to mate and raise lit-ters successfully.

The total dingo population in Australia prob-ably peaked during the 1930–50s, and sincethen numbers have remained high but theproportion of hybrids in the overall popula-tion has increased greatly (Corbett 1995a)(Section 2.9). Dingo samples collected in the1960–70s indicated that about half the wilddog populations in southern Australia werehybrids (Newsome and Corbett 1982), andthe most recent surveys in the early 1980s(Jones 1990) confirmed the trend of increas-ing hybridisation. Even in the presumed safebastion for pure dingoes on the mainland,Kakadu National Park in northern Australia,hybrids now occur (Corbett in press). If thecurrent rate of hybridisation continues, puredingoes may well be extinct by 2100 withonly hybrids and feral dogs remaining on theAustralian mainland.

Radio-tracking and mark–recapture studieshave indicated dingo densities of 0.17 persquare kilometre in Kosciusko National Parkin south-eastern Australia (McIlroy et al.1986a) and 0.14 per square kilometre at

Kapalga in Kakadu National Park in northernAustralia (Corbett 1995a). In the Guy FawkesRiver region in north-eastern New SouthWales crude estimates (using theindex–manipulation–index technique) ofwild dog density in unpoisoned areas werebetween 0.19 and 0.3 wild dogs per squarekilometre (Fleming 1996b). An adjacent areain the Guy Fawkes River region that wasannually baited with 1080 (sodium fluoroac-etate) had pre-baiting densities of between0.1 and 0.17 wild dogs per square kilometre.In the Fortescue River region the density ofdingoes in pack territories varied between0.03 and 0.25 per square kilometre (mean0.08 per square kilometre). In SouthAustralia, at the height of a rabbit plague,dingo numbers were crudely estimated at 0.3per square kilometre based on the maximumnumber of dingoes observed visiting an iso-lated water and the area serviced by thatwater (Bird 1994).

2.9 Hybridisation

2.9.1 The extent of hybridisation

The expansion of farming and grazing activi-ties in the nineteenth century led to thespread of domestic and feral dogs and theirconsequent hybridisation with dingoes.More than half of the wild dog population insouthern and eastern Australia are hybrids(Section 2.8.4; Figure 2). Although wild dogpopulations in northern Australia and otherremote areas remain essentially pure din-goes, hybrids are now being recorded there.Even on Fraser Island, five of a sample of 35wild dogs recently culled there were hybridanimals (Woodall et al. 1996).

The genetic make-up of wild hybrid dogshas not been evaluated except at the broadlevel of presence or absence of domestic doggenes in the wild dog population (Newsomeet al. 1980). Genetic evaluations have beendifficult because hybrids often look like puredingoes. New techniques are being devel-oped to better assess the genetics of wilddog populations (Wilton et al. 1999).


2.9.2 The main processes ofhybridisation

Hybridisation mostly occurs when domesticdogs go bush and dingoes come to town.However, since interactions between din-goes and feral dogs in the bush differ greatlyfrom those in urban places (Figure 7), so toodo the rates of hybridisation (Corbett 1995a).

‘Once hybrids become more prevalent, the resultant

lessening of behavioural differences will expedite the

hybridisation process.’

It is well known that the dogs of cattlemen,recreational fishers, bushwalkers, holiday-makers and Aboriginal people are occasionallylost in the bush (Corbett 1995a). However, thebehavioural differences between dingoes anddomestic dogs seem great enough to make itdifficult for dogs to infiltrate dingo society andbreed, particularly in remote areas where thereare more dingoes. Once hybrids become moreprevalent, the resultant lessening ofbehavioural differences will expedite thehybridisation process. This could partlyexplain the greater proportion of hybrids insouth-eastern Australia.

There is now a trend for people to acquire(often illegally) dingo pups as pets. The pupsmay be easily handled, but adults are usuallynot good pets, simply because they are wildanimals that have not been selectively bred forthe behavioural characteristics that make agood pet (RSPCA 1997). A pet dingo is likely touse its owner’s home as a base from which toroam and do as it pleases, or else it is aban-doned when it becomes an adult. The upshotof this pet trend is that dingo–domestic dogcontact is increased; because pet dingoes havegrown up without learning the socialbehaviours that curb crossbreeding withdomestic dogs, they are more likely tohybridise with domestic dogs than wild breddingoes. Many such hybrids are rejected byowners or stray to the bush where they mayinfiltrate wild dingo society and breed withpure dingoes. This process occurs more fre-quently in semi-rural areas near large urbancentres. If dingo breed societies promote andsell hybrids as pure dingoes, the rate ofhybridisation will increase (Section 5.3.3).

Some of the dogs that contribute to the genepool of wild-living dogs are not truly feral butare strays or free-roaming pet domestic dogs(Meek 1998). There have been reports ofdingo-like wild dogs mating with restrainedfemale domestic dogs (A. Melling, grazier, NewSouth Wales, pers. comm. 1984).


URBAN ANDRURAL CENTRES

Dingoes

BUSH AND OUTBACK STATIONS

Domestic dogs

Stray

Lost/abandoned

(this link is inevitable)

Rejected/stray

(this link is inevitable)

Pets/reared in centres

(this link can be broken if publicattitude changes)

Hybrids

PURE DINGOESDECREASE

MANY HYBRIDS

* Not shy* Easy to infiltrate domestic dog society

* Difficult to infiltrate dingo society

* Easy to infiltrate dingo society

FEW HYBRIDS

Figure 7: The process of hybridisation between dingoes and domestic dogs (after Corbett 1995a).

2.10 Co-occurrence with other predators

2.10.1 European red fox

The spatial relationship between wild dogsand the introduced European red fox (Vulpesvulpes) in Australia is poorly understood.Observations of an inverse relationshipbetween wild dog and fox abundance(Jarman 1986) have been popularly inter-preted to mean that wild dogs may limit thedistribution and abundance of foxes.However, this speculation is based largely onthe correlation of relative abundances andneeds to be confirmed by a demonstration ofcausality.

‘Although wild dogs may notinfluence the distribution of

foxes, they do influence theiraccess to resources.’

In areas of eastern Australia where wild dogshave not been excluded by human manage-ment, there is little evidence that wild dogsexclude foxes because the two species com-monly co-exist (Newsome et al. 1983a;Triggs et al. 1984; Robertshaw and Harden1985a and unpublished data 1977–84;Catling and Burt 1995; Fleming 1996a).Catling and Burt (1995) found no evidence inforested areas that foxes were excluded by,or avoided, wild dogs. They argue that popu-lations of red foxes in south-eastern Australiaare more likely to be limited by factors otherthan the presence of wild dogs. In manyareas, the marked differences between thehabitats and management regimes of agricul-tural (higher fox abundance) and non-agri-cultural land (higher wild dog abundance)confounds the comparison of the relativeabundances of the two species, and providesa number of alternative hypotheses thatcould explain observed differences in therelative abundances of the two species.

In central Australia foxes appear to avoidwild dogs, especially at widely spaced andshared watering points. Wild dogs alsoexclude foxes from kangaroo and cattle car-casses during droughts (Corbett 1995a), so

that, although wild dogs may not influencethe distribution of foxes, they do influencetheir access to resources. In northern SouthAustralia, foxes were also seen to avoid din-goes at waters and were frequently drivenoff by dingoes (P. Bird, Primary Industriesand Resources, South Australia, pers. comm.1999).

Similar evidence exists for Western Australia.In the Fortescue River region, the presenceof foxes was noted (sightings, captures intraps, tracks and other signs) during a nine-year intensive radio-tracking study of din-goes (Thomson 1992a; unpublished data1976–1984). In this area, foxes were relative-ly common on the productive coastal plains,where sheep were grazed and dingoes weresubject to intensive control. In the adjacentrugged unstocked areas, dingoes were ini-tially at low levels and foxes were present,albeit in low numbers. When dingo numbersincreased, foxes were only ever recorded inthe unstocked areas at the edge of the sheeppaddocks. When dingo numbers fell, foxactivity began to appear again deeper intothe unstocked areas. Although circumstan-tial, it seems likely that foxes were affectedby the abundance of dingoes. When dingonumbers were low in the unstocked areas,foxes appeared to invade from the adjacentcoastal plains.

In contrast, in another study on the NullarborPlain in Western Australia (Thomson and P.Marsack, unpublished data 1982–1987), din-goes, foxes and feral cats commonly co-occurred in the same local areas. Althoughthis seems to contradict the evidence fromthe Fortescue region, there are features ofboth environments that may explain the dif-ferences. In the Nullarbor area, rabbits werenumerous and distributed relatively uniform-ly across the landscape. Foxes would havebeen able to hunt rabbits within dingo homeranges, while avoiding resident dingoes. Incontrast, resources in the Fortescue areawere patchy and rabbits were absent.Dingoes favoured the most productive areas(Section 2.2), so it would have been difficultfor foxes to avoid dingoes, unless theyremained in the poorer areas. In addition,the presence of rabbit warrens in theNullarbor area gave foxes an opportunity for


escape from dingoes should conflict arise.Dingoes kill foxes (Marsack and Campbell1990) and avoiding dingo attack would havebeen more difficult for foxes in the Fortescuearea.

2.10.2 Cats

Feral cats co-occur with wild dogs through-out most of Australia but almost nothing isknown of their interrelationships. Fleming(1996b) found that wild dogs co-occurredwith feral cats (and red foxes and spotted-tailed quolls (Dasyurus maculatus)) on aregional scale in north-eastern New SouthWales but suggested that there might besome spatial segregation at a more localisedlevel. Corbett (1995a) suggested that wilddogs exclude feral cats from kangaroo and

cattle carcasses during droughts. Wild dogsoccasionally eat feral cats (Robertshaw andHarden 1985a; Thomson 1992c, Corbett1995a) although the impact of this on catpopulations is unknown.

2.10.3 Quolls

Prior to European settlement, dingoes almostcertainly co-occurred with the spotted-tailedquoll, eastern quoll (D. viverrinus), westernquoll (D. geoffroii) and northern quoll (D.hallucatus). Today, the distribution of wilddogs and quolls has been much reduced byhabitat changes associated with Europeansettlement. However, dingoes and other wilddogs still co-exist in many areas with allquoll species except the eastern quoll, whichis believed to be extinct on mainlandAustralia.


Summary

Wild dogs have impacts on agriculturethrough predation of livestock. Measuringstock losses alone usually underestimates thetrue economic cost of wild dogs becausethese estimates do not include costs of dogcontrol, the opportunity costs and other costsassociated with wild dogs (for example, notgrazing sheep on otherwise suitable land).However, most economic assessments havebeen limited to calculations of stock lossesbecause these are the easiest and most com-prehensible measures for landholders to use.Studies in Western Australia, Victoria, NewSouth Wales and Queensland have shownthat predation of sheep and cattle canthreaten the economic viability of propertiesin some areas. Sheep are the most commonlyattacked livestock, followed by cattle andgoats. There is evidence of seasonal peaks inpredation on livestock, possibly related to theseasonal breeding activity of wild dogs, aswell as the timing of lambing and calving.

Although many wild dogs will attack orharass sheep, some individual wild dogscause far more damage than others, some-times maiming without killing outright. Thepresence of wild dogs can adversely affectthe distribution and behaviour of sheep,even when dogs do not actively harass them.Wild dogs sometimes chase sheep withoutfollowing through with an attack. This maylead to stress-associated behaviour such asmismothering of lambs and loss of produc-tion. Even when wild dogs kill sheep, theyoften leave carcasses uneaten. Individualwild dogs that kill sheep often eat naturalprey such as kangaroos, indicating thatkilling of livestock is independent of the pres-ence of natural prey.

Wild dogs have been in Australia longenough to become a functional part of themammalian predator–prey relationshipsand fulfil an important role in the function-ing of natural ecosystems. Dingoes areimplicated in the extinction of Tasmaniandevils and thylacines from mainlandAustralia. Dingoes are regarded as a native

species and are subject to legislative protec-tion in some States and Territories.

Wild dogs are implicated in the spread ofhydatids which is a risk to human healthand the cause of losses of production associ-ated with hydatidosis in cattle and sheep.Wild dogs also provide a reservoir for heart-worm infection and diseases such as par-vovirus. Dog rabies is presently exotic toAustralia, but of all Australian wildlife, wilddogs pose the greatest risk of maintainingand spreading dog rabies if it was intro-duced.

The dingo, often considered a nativeAustralian mammal, has an intrinsic andaesthetic value and there is a public expec-tation that it will be conserved. Other wilddogs, as top order predators, may have animportant, but as yet unclear, influence onthe biodiversity of animal communities.They might also have an inverse densityrelationship with foxes and therefore beimportant in limiting the impact of foxesand cats on populations of small and medi-um-sized mammal prey.

In areas where dingoes are a major touristattraction, they occasionally show aggres-sive behaviour towards people, particularlyif they are often fed to encourage closerviewing.

3.1 Economic impact

The threat of predation of livestock by wilddogs has largely determined the distribution ofsheep and cattle in Australia (Figure 2)(Newsome and Coman 1989). Rapid expan-sion of the sheep industry after the successfulearly 1800s ventures of the MacArthurs andMarsden into merino breeding brought theproblem of predation by wild dogs to theattention of early legislators (Section 5.1.3).Sheep were under the constant supervision ofshepherds (Gould 1863; Rolls 1984) who hadresponsibility for preventing sheep strayingand preventing predation by wild dogs. Theamount of fencing for the enclosure of live-stock increased after the Nicholson Land Act


3. Economic and environmental impacts and values

Victoria 1860 and the Robertson Free SelectionAct NSW 1861, and the shortage of labour thataccompanied the 1860s goldrushes.Consequently, the use of shepherds became lesscommon although the threat of wild dog preda-tion ensured that considerable investment oflabour continued to be placed on wild dogcontrol. Predation has continued until the pre-sent although there have been few attempts toquantify losses at more than the property scale.

A Western Australian example of losses causedby dingoes over an 18-day period is given in(Thomson 1984a) (Table 2). These loss figuresare considered conservative as some eventscould have been missed and sheep were beingmustered from the paddocks so not all wereavailable to dingoes for the full period.Extrapolation of the data in Table 2 suggests anannual loss of 33% in area A and 16% in AreaB. Had dingo activity continued, direct lossesof this magnitude, along with those due toharassment, would have seriously threatenedthe viability of the enterprise (Thomson1984a).

Predation of livestock by wild dogs continuesto threaten the livelihood of some livestockproducers in tableland environments of east-ern New South Wales, the Australian CapitalTerritory and Victoria. In parts of the easterntablelands of New South Wales, wild dogs areregarded as the major limitation to sheep pro-duction (Fennessy 1966; New England RuralDevelopment Association (NERDA) undatedc.1966; Hone et al. 1981; Schaefer 1981;Fleming and Robinson 1986; Fleming and Korn1989). Holdings where wild dogs are a prob-lem are mostly situated along the Northern andSouthern Tablelands of New South Wales(Fleming and Korn 1989). The study byNERDA (c. 1966) also estimated the opportuni-ty cost of sheep not being run in areas thatwere suitable but for the presence of wilddogs. The timing of the survey correspondedwith a regional increase in pasture improve-ment with superphosphate. It was believedthat pasture improvement combined withdingo control could increase sheep numbersby 93%.


Even when livestock are not killed outright by wild dogs, economic losses may arise due to veterinary costs, decline inlivestock condition and downgraded sale prices (Source: L. Corbett).

‘There were seasonal peaks in predation, possibly

related to the seasonal breedingof wild dogs and control activity,as well as the timing of lambing

and calving.’

The surveys of dingo damage (Table 3) con-ducted by NERDA (c. 1966) and Schaefer(1981) were inherently biased because theirsamples were obtained from graziers attend-ing meetings to discuss the issue of wild dogpredation. Fleming and Korn (1989) report amonthly survey of authorised wild dog con-trol officers (of Rural Lands ProtectionBoards) from eastern New South Wales overthe four years from 1982 to 1985 (Table 3).During this period, 25 644 livestock wereattacked by wild dogs. Sheep were the mostcommonly attacked animal, followed by cat-tle and goats. Regional differences wereapparent in the livestock species attackedand these reflected the ratio of sheep to cat-tle grazed in each region. Not surprisinglytherefore, the major losses occurred in areaswhere sheep were present. Fleming andKorn (1989) found seasonal peaks in preda-tion, possibly related to the seasonal breed-ing of wild dogs and control activity, as wellas the timing of lambing and calving.

In 1984 and 1985, another survey of 111 ran-domly selected livestock producers withland close to or within terrain inhabited bywild dogs was conducted in north-easternNew South Wales (Table 3; Fleming 1987).Thirty-six per cent of producers reported thatwild dogs had attacked their livestock duringthe survey and 1194 sheep and 127 cattlewere killed in those attacks. The mean lossesduring the survey were 7.17 sheep per prop-erty per year and 0.76 cattle per property peryear. The mean value of losses suffered bythose participants that had livestock preyedon by wild dogs was $1900 (Fleming 1987).The mean value was largely dependent onthe values of livestock that prevailed duringthe survey. The cost to the sheep industry ofpredation by wild dogs in eastern New SouthWales for 1988 was estimated at around $4million (Saunders and Fleming 1988). Of thesurveys undertaken in north-eastern NewSouth Wales, only one survey (Fleming 1987)was an unbiased sample from landholderswithin or adjacent to terrain inhabited bywild dogs.

In 1985, Backholer (1986) mailed a question-naire to 809 properties in 23 shires ofVictoria that had a history of livestock preda-tion by wild dogs. The 508 respondentsreported mean losses per property of


Radio-collared dingoes involved

Potential sheep available

Harassments, no injuries

Minor injuries

Kills/mortal injuries

Other verified deaths

Total identifiable losses

Area A Area B

Table 2: Sheep losses caused by dingoes over an 18-day period, detected during radio-tracking in two areas on MardieStation, Western Australia where dingoes were being controlled (after Thomson 1984a).

3

800 4200

5 3

1

13

33

0 7

17*

26

3

6

*Total includes four sheep killed on a neighbouring property by one of the radio-tagged dingoes.

between $700 and $7400 per annum, 0.1 to24.9% of the total value of the enterprises.The total loss was $835 000. As well as this,opportunity costs (including time, movingstock, repairing fences) totalled $662 500and government agencies spent $1 444 500.Therefore the total annual loss was around$3 million. During the preceding eight years,the annual average of the total losses thatwere reported in the surveyed area was 2400sheep which would be equivalent to an aver-age annual loss of $1650 per property(Backholer 1986).

The most recent survey of livestock lossesinflicted by wild dogs was in the NorthernTerritory in 1995. The results of a question-naire of approximately 67% of pastoralistsled to a rough estimate that annual calf lossesattributable to predation by wild dogs werebetween 1.6% and 7.1% (Eldridge and Bryan1995). The estimated annual value of all cat-tle killed by wild dogs in the NorthernTerritory (assuming an average value of $540per head and 100% calving — such calvingrates are optimistic) was $13.5 million withcontrol costs of $300 000. This representsaverage losses of $89 000 per property andcosts of $2000 per property. Despite theextreme losses and comparatively small

investment in wild dog control, 96% ofrespondents rated wild dog control as beingworthwhile.

One of the difficulties in providing clear dataon the impact of wild dogs on livestock pro-duction is that most wild dog control pro-grams manage to achieve, at least to someextent, their aim of reducing predation onlivestock. Thus, figures on losses or damagecan be misleading, rarely reflecting thepotential impact that could arise in theabsence of wild dog control. The economicimpact of wild dog predation of livestockcannot simply be measured in terms of live-stock killed by wild dogs. Losses other thandirect maimings and killings of livestockcaused by wild dogs are especially difficultto quantify. The costs of control activities,losses of genetic material, capitalisation ofthe risk of wild dog predation into land val-ues (Schaefer 1981) and opportunity costs(such as the allocation of labour and capitalto predation mitigation and planning insteadof other on-farm activities) need to beaccounted for in economic assessments.Other opportunity costs are the impositionof sub-optimal enterprise mixes for a particu-lar agro-environment and sub-optimal pas-ture management caused by the presence of


Table 3: Predation of livestock by wild dogs in north-eastern New South Wales. Mean losses were calculated as the totalnumber of sheep killed by wild dogs divided by the total number of sheep run by the survey participants and expressedas a percentage (from Fleming 1996b).

Year(s) Sheep killed(Number/property/year)

Mean lossesof sheep (%)

Source

1961–62

1980–81

1982–85

1984–85

19.5a

19.4

N/A

14.5

1.33

0.9

0.7b

0.8

NERDA (undated c. 1966)

Schaefer (1981)

Fleming and Korn (1989)

Fleming (1987 and unpublished data1985)

aTo equate the data with previous biased surveys (NERDA undated, c. 1966; Schaefer 1981), only thelosses of Wild Dog Control Association members in Fleming (1987) are presented in this column.bLosses reported to local Rural Lands Protection Boards multiplied by three to account for two-thirds under-reporting recorded in Fleming and Korn (1989).


(a) Calves are most vulnerable to wild dog attack when they are newborn and left while the cow feeds; (b) ‘nursery groups’ ofcalves protected by older cows are common in areas where cattle are familiar with wild dog attacks (Source: L. Allen, Department of Natural Resources, Queensland).

a

b

wild dogs. Although the mix of sheep andcattle may be partially offset by capitalisationof the costs of wild dogs into the purchaseprice of land, enterprise mixes that arebelow the optimum are likely to negativelyaffect the cash flows and long-term prof-itability of grazing enterprises in wild doginhabited areas.

‘Figures on losses or damage can be misleading,

rarely reflecting the potentialimpact that could arise in theabsence of wild dog control.’

Despite potential shortcomings, surveys ofthe type outlined above substantiate the gen-erally accepted view that the major lossescaused by wild dogs occur amongst sheepflocks. An attempt to specifically documentand quantify the types of losses caused tosheep flocks was made on Mardie Station inthe Pilbara region of Western Australia(Thomson 1984a). In this study, 26 radio-col-lared dingoes in sheep paddocks wereobserved from the air. Interactions withsheep were followed up by investigations onthe ground. Additional information wasobtained when sheep were mustered andinspected at shearing time. The major find-ings were:

• Some dingoes caused far more damagethan others did, although most attackedsheep, sometimes maiming without killingoutright.

• The presence of a dingo could adverselyaffect the distribution and behaviour ofthe sheep, even if the dingo did not active-ly harass them.

• Dingoes sometimes chased sheep with-out following through with an attack.This could still lead to harm such as mis-mothering of lambs.

• When dingoes killed sheep they often leftcarcasses uneaten.

• Individual dingoes that frequently killedsheep often ate natural prey such as kan-garoos (Macropus spp.).

Dingoes easily outpaced sheep, although itwas clear that many of the witnessed chaseswere not motivated by hunger; chases weresometimes seen after wild dogs had beenobserved feeding. At times wild dogs simplybroke off chases, or moved on to chase othersheep in the mob, then left with no actualphysical contact with the sheep. Sheep weresometimes chased through fences (this hasalso been reported in north-eastern NewSouth Wales), on occasions into waterlessareas. At times, clear changes in the distribu-tion of sheep within large (up to 150 squarekilometres) paddocks were observed; sheepabandoned favoured grazing areas whendingoes were present. When ewes andlambs were chased, ewes ran off wildly,leaving lambs to keep up as best they couldand potentially causing their death throughmismothering. All these events could causeproduction losses, although they would notnecessarily be reflected in a ‘carcass count’.

The survival of most bitten sheep was poor.Far fewer bitten sheep were tallied at shearingthan would have been expected based onobservations in the paddocks. Nevertheless,rams seemed to survive severe scrotal injuries,with some being fully castrated by wild dogsattacking from behind (Figure 11).

In some areas, producers have elected to runcattle instead of sheep because of the effects ofwild dog predation. For example, Backholer’s(1986) survey showed that, to minimise wilddog predation, 12% of eastern Victorianrespondents reduced their sheep numbers ordid not run sheep. Although wild dogs cancause losses to cattle herds, as discussed next,the impacts are usually less severe and moreeasily overcome. Nevertheless, some grazinglands are more suited to running sheep thancattle, and the move to other enterprises suchas cattle production can be viewed as anopportunity cost of wild dog predation.

Damage by wild dogs is likely whenever theirranges overlap those of sheep. Damage occurslargely independent of age and condition ofsheep, age and density of the dogs, seasonalconditions and availability of alternative foodfor dogs. However, most of these factors influ-ence wild dog predation on cattle (Rankineand Donaldson 1968; Corbett 1995a; Allen andGonzalez 1998). In contrast to the situation


with sheep, the impact of wild dogs on cattleproduction is more variable. Generally, attackson young calves are the major cause of cattlelosses to wild dogs (Corbett 1995a; Flemingand Korn 1989). However, the cost of wilddog predation to the beef industry in Australiahas not been estimated. The profitability ofnorthern Queensland cattle enterprises is sub-stantially affected by branding percentage(Sullivan et al. 1992). Predation of calves bywild dogs is the main cause of neonatal lossesin northern Australian cattle (Rankine andDonaldson 1968) and hence has major bearingon branding precentage. Branding percent-ages also vary considerably with seasonal con-ditions, pasture improvement and level of ani-mal husbandry. Predation is greater whenalternative food is scarce (Thomson 1992c;Corbett 1995a; Allen and Gonzalez 1998).Improved weaner management and nutritionhave been shown to potentially increasebranding percentages from 40% in 1968–69 to1970–71 (Anderson and McLennan 1986) to80% twenty years later (Fordyce and Entwistle1992).

‘Predation may be higher when control operations cause

invading young wild dogs tocome into contact with

cattle herds.’

Estimates of predation losses of calves andweaners in normal conditions in rangelandgrazing areas are in the range of 0–29.4% peryear (Rankine and Donaldson 1968) andstudies of reproductive failure in cattle herdsin Queensland have suggested up to 30%loss of calves caused by predation by wilddogs (Allen and Gonzalez 1998). Such losseswould negate the potential gains due toimproved livestock and pasture managementin northern Australia.

There is evidence that the age and socialorganisation of a wild dog population canaffect the extent of predation on calves.Calves are most vulnerable when newborn,though the protective behaviour of the cowcan be sufficient to deter wild dog attacks(Thomson 1992c, and unpublished observa-tions). Experienced wild dogs, operating as ahunting unit, are more likely to be successful

at killing calves. On the other hand, there isevidence from north Queensland (Allen andGonzalez 1998), that risks of predation maybe higher when control operations cause apreponderance of invading young wild dogsto come into contact with cattle herds.Seasonal effects (Corbett 1995a) and thescale of control may modify the damageresponse; a larger baited area would reducethe rate of repopulation and the number oftransient dogs, as Thomson (1984b) found inWestern Australia.

It is possible that increased predation couldbe caused by the greater mobility of immi-grant dingoes, increasing the probability ofan encounter with a newborn calf. Thismight be exacerbated by the lone status andrelative inexperience of immigrant wilddogs, resulting in their poor success at hunt-ing larger and faster native prey such as kan-garoos.

Appropriate strategies to manage the impactsof wild dogs differ markedly depending onthe type of livestock and the conditions pre-vailing in an area (Chapters 5, 6 and 7).

3.2 Environmental impact

Predation by wild dogs may have an impacton the survival of remnant populations ofendangered fauna. For example, predation bythe dingo has been implicated in the extinc-tion of the Tasmanian native-hen (Gallinulamortierii) from mainland Australia (Baird1991). Endangered populations of marsupialsmay require management of their predators tobecome re-established or to survive (Johnsonet al. 1989). Predation by wild dogs is less like-ly to threaten populations of more abundantmarsupials (Robertshaw and Harden 1989).

The other environmental impacts of wild dogsrelate to their management, rather than to thepresence of the animals themselves. First,control measures may have a direct impact onnon-target species (Section 6.6.2). Second,reducing wild dog density may result in anincrease in other predators with overlappingdiets. (This process of substitution of preda-tors is called ‘mesopredator release’ (Soulé etal. 1988)). There is a commonly held opinion(Denny 1992; Smith et al. 1992) that removing


wild dogs from a system where foxes (Vulpesvulpes) also occur will result in an increase infox numbers with consequent increased pre-dation on critical body weight range (CWR)mammals (35–5500 grams) (Burbidge andMcKenzie 1989). However, neither anincrease in fox numbers resulting from areduction in wild dog numbers nor a resultingincrease in predation of CWR mammals havebeen demonstrated. Schlinder (1974) has alsoargued that, by the same mechanism, feralpigs could replace dingoes in some situations.

‘Predation by wild dogs may have an impact on the

survival of remnant populationsof endangered fauna.’

Predation on environmentally damaging feralungulates has been suggested as a positiveimpact of dingoes and other wild dogs. Forexample, Parkes et al. (1996) suggested thatthe limited distribution of feral goats in north-ern Australia was attributable to dingo preda-tion. Twenty sterilised dingoes were releasedon Townshend Island, off Queensland, in anattempt to reduce the population of 2000–3000feral goats. Within one-and-a-half years 99% ofthe goats had been removed and feral goatswere eradicated from the island after three-and-a-half years (L. Allen, QueenslandDepartment of Natural Resources, pers. comm.1998). Given that the dingoes survived for twoyears on very few goats, this raises the interest-ing question of the impact of dingoes on alter-native prey (Section 3.6.5) that must have beenkilled during the latter part of the exercise.Caution should be exercised in translatingthese results to mainland ecosystems. Pavlov(1991) provided circumstantial evidence that,while dingoes undoubtedly kill feral pigs, theydo not significantly affect feral pig abundance.

3.3 Resource and conservationvalue

Although feral dogs and hybrids are seldomregarded as a wildlife resource, dingoes havepotential value from five resource perspec-tives: (1) harvest for food or skins; (2) conser-vation of native species and natural communi-ties (Section 4.4); (3) in Aboriginal mythology;(4) tourism; and (5) as a specialised dog breed.

1. Unlike in some Asian countries where din-goes are eaten (Corbett 1995a), there is nocurrent harvest of dingoes for food inAustralia. There is evidence that Aboriginalpeople ate dingoes and other wild dogs(Manwell and Baker 1984). Dingoes couldbe harvested for their pelts, but these aregenerally of low value. It is illegal underthe Commonwealth Wildlife Protection(Regulation of Exports and Imports) Act1982 and subsequent amendments, andassociated Amendment Acts of 1986,1991 and 1995, to export native wildlifeproducts, including pelts, without appro-priate licenses. Trade in skins, except forthe recovery of bounties, is not permittedin those States and Territories where thedingo is afforded some conservation sta-tus.

2. The dingo has been in Australia longenough to have colonised most suitablehabitats and, in many instances, is still afunctional part of predator–prey relation-ships. The implication from this is that thedingo fulfils an important role in the func-tioning of ecosystems. Legislation in someStates acknowledges the ecological signifi-cance of dingoes and provides protectionin some situations (Section 5.2). Whetherdingoes have conservation value becausethey have a functional role in the conser-vation of natural ecosystems is unclear(Sections 3.4, 3.5 and 3.6). Regulation ofmacropod and emu (Dromaius novaehol-landiae) populations by wild dogs hasbeen inferred, but not proven, in a num-ber of studies (Caughley et al. 1980;Shepherd 1981; Robertshaw and Harden1986; Thomson 1992c; Fleming 1996b;Pople et al. 2000). Prey regulation by wilddogs has been inferred for rabbit popula-tions after droughts have reduced rabbitnumbers in arid areas (Corbett 1995a).Conversely, in an experiment in thewet–dry tropics, Corbett (1995c) demon-strated that dingoes do not regulate feralpig populations (Section 3.6.5).

In contrast, there is evidence that dingonumbers in many areas of central andnorthern Australia have increased since


European colonisation. Predation bydingoes may threaten the survival ofsome CWR mammals. Where this occurs,dingoes may be considered a biologicalliability rather than a biological resource.Where feral dogs and hybrids have sup-planted dingoes, these too may performa similar role in the function of theecosystem. Predators affect the distribu-tion and abundance of their prey(Huffacker 1970) and so the substitutionof one canid subspecies with a similarsubspecies is likely to maintain thedynamics of the community in whichthey occur.

3. The dingo is an important animal inAboriginal mythology. For some Aboriginalpeoples (for example, the Pitjantjatjara)dingoes are associated with sacred sitesand are considered the physical embodi-ment of a Dreamtime character orTjukurpa. The local disappearance ofnative species significant to Aboriginalpeople has caused them distress in thepast and similar distress may be caused ifdingoes become extinct.

4. Dingoes provide an indirect resourcevalue from tourism. They are present inmany zoos and private wildlife parks.Fraser Island is an example where free-living dingoes are a tourist attractionproviding income to island businesses.Tourism by recreational hunters is anoth-er source of value of dingoes (Allisonand Coombes 1969). The ‘howling-up’ ofa dingo may be regarded as a test ofhunting skills and the presentation ofscalps for bounties (Section 5.1.1) as asmall remuneration to help cover thecosts of the hunt (Allison and Coombes1969).

5. The dingo is recognised as an officialbreed of dog by the Australian NationalKennel Council and was adopted asAustralia’s national breed in November1993. In New South Wales, this status hasbeen recently recognised in legislation(Companion Animals Act 1998) whichallows people to keep dingoes under thesame restrictions as other breeds. Animals

kept by breed societies are traded as petsor specimen animals. This is potentiallydamaging for the conservation of puredingoes because many animals held bybreed societies are likely to be hybridsand there are no valid checks in place todetect them (Corbett 1995a; Corbett inpress; Section 2.9).

The presence of dingoes has an ‘unpriced’value (Sinden and Worrell 1979) that is diffi-cult to quantify and comprises the non-mon-etary value that many people place on din-goes. The dingo is an icon which many peo-ple value knowing is present in the wild, thatis, its ‘existence value’.

3.4 Diseases and parasites

Endemic diseases and parasites

Hydatidosis (causal agent Echinococcusgranulosus) causes fatalities and morbidityin humans. The prevalence of hydatidosis inhumans is often linked to sylvatic cycles inwild dogs and wildlife (Coman 1972a;Thompson et al. 1988). The prevalence inhumans is relatively low but is more com-mon in south-eastern Australia (Jenkins andPower 1996).

Hydatidosis associated with a sylvatic cyclewithin wild canids and macropods (Durie andRiek 1952; Coman 1972b) leads to the condem-nation of offal from up to 90% of slaughteredcattle from endemic areas (D. Jenkins,Australian Hydatids Control and EpidemiologyProgram, unpublished data 1999). An abattoirsurvey of sheep from the Southern Tablelandsof New South Wales from 1970–72 foundhydatidosis in up to 40% of carcasses (Hunt1978). Bovine hydatidosis (causal agentEchinococcus granulosus) prevalences of2.2–55.7% have been reported in south-easternQueensland (Baldock et al. 1985) and of0.5–7% in north-eastern Victoria (D. Jenkins,Australian Hydatids Control and EpidemiologyProgram, Australian Capital Territory, pers.comm. 1998). The latter prevalences were inspite of an extensive hydatid control programaimed at domestic and farm dogs. Hydatidosisis an occupational risk for wild dog trappersand researchers.


Red foxes have also been identified as defini-tive hosts and macropods as intermediatehosts for hydatid transmission in south-east-ern Australia (Obendorf et al. 1989; Reichel etal. 1994). Where feral dogs, dingoes and free-roaming domestic dogs co-occur with foxes(for example, in coastal south-easternAustralia, Meek 1998) the control of humanhydatidosis becomes more difficult.

‘The prevalence of hydatidosis inhumans is often linked to sylvaticcycles in wild dogs and wildlife.’

Wild dogs also transmit the viruses thatcause canine distemper (Paramyxovirus),canine hepatitis (Adenovirus) and parvovirusdisease (causal agent Parvovirus). Althoughthese diseases adversely affect domesticdogs, their transmission by wild dogs isunlikely to pose a significant threat becausethese diseases can be controlled.

The presence of heartworm (Dirofilariaimmitis) in dog populations is linked to thepresence of mosquitoes in endemic areas(Russell 1990). Heartworm infections areuncommon in the tablelands of south-easternAustralia, with most cases of infection indomestic dogs being in animals that havebeen moved by their owners to and fromendemic areas (Carlisle and Atwell 1984).Heartworm infection has not been recorded infoxes from the Northern, Central and SouthernTablelands of New South Wales (P. Flemingand B. Kay, NSW Agriculture, and D. Jenkins,Australian Hydatids Control and EpidemiologyProgram, Australian Capital Territory, unpub-lished data 1999). It is therefore unlikely thatwild dogs and dingoes from tablelands ofsouth-eastern Australia will experience heart-worm infection and that, more commonly,infection will be in coastal areas and northernAustralia. In urban Melbourne, where heart-worm infection has been recorded in foxes, asylvatic cycle of canine heartworm has beenpostulated (Marks and Bloomfield 1998). Thisprocess may also apply in endemic areaswhere foxes and wild dogs co-occur.

Seddon and Albiston (1967) suggest that wilddogs may act as hosts for the parasite thatcauses sheep measles (Taenia ovis) and theconsequent condemnation of sheep carcasses.

However, Coman (1972a) failed to find evi-dence of T. ovis infection in a sample of 204dingoes and other wild dogs in north-easternVictoria and linked this absence to the infre-quent occurence of sheep and cattle in the dietof dogs in his sample.

Parasites have also been instrumental in theidentification of dingo origins. The occur-rence of the biting lice (Heterodoxusspiniger) on dingoes and macropods inAustralia and on Asian dingoes implies thatdingoes were transported to and from Asia(Corbett 1995a) (Section 1.2).

Exotic diseases and parasites

Canids are regarded as the most importantsource of rabies (Rhabdoviridae) in humans(Garner 1992), with dogs causing an estimat-ed 75 000 cases annually throughout theworld (Fenner et al. 1987). As dog rabies ispresently exotic to Australia (there has beenone reported outbreak since European settle-ment in Tasmania in 1867; O’Brien 1992), thepotential role of wild dogs can only be specu-lated. Newsome and Catling (1992) suggestthat of all Australian wildlife, wild dogs andfoxes pose the greatest risk of maintainingand spreading dog rabies after introduction,and Thomson and Marsack (1992) proposeaerial baiting of buffer zones as the primaryweapon against the spread of rabies amongdingoes in rangelands. Forman (1993) indicat-ed that the establishment of a sylvatic dograbies cycle in Australia was remote but possi-ble. The dog strain of rabies remains the mainfocus for quarantine barrier prevention of thisdisease in Australia.

If dog rabies were to become endemic inAustralia, interaction between free-roamingdogs and feral dogs and dingoes would be themost likely avenue for dog rabies transmissionto humans. Free-roaming dogs have beenrecorded making linear movements of up toeight kilometres into bushland where wilddogs and foxes co-occur (Meek 1998). Theseanimals entered bushland with high macrop-od density to hunt and then returned to theirowners. The mean duration of hunting forayswas 23 hours, and it is probable that interac-tions between wild and free-roaming dogsoccurred on these trips. Newsome and Catling(1992) consider that, at the high densities of


wild dogs and dingoes found in northernAustralia and south-eastern Australia, dograbies would persist were it introduced.Interactions of unrestrained and unvaccinateddomestic dogs with wild dogs would con-tribute to human infection if rabies were tobecome established in wild dogs. Rabies is notsolely a problem of human health but alsoaffects livestock production; for example, batrabies from vampire bats (Desmodus rotun-dus) is a serious cause of mortality in cattle inCentral and South America (Garner 1992).Should a sylvatic cycle of dog rabies becomeestablished in wild dogs in Australia, it mightaffect sheep and cattle production and makethe treatment of animals injured by wild dogsmore risky.

‘Wild dogs and foxes pose the greatest risk of maintainingand spreading dog rabies after

introduction.’

A number of other diseases are importantpathogens of dogs in other parts of the worldand their introduction to Australia wouldadversely affect domestic dogs, particularlyin breeding kennels, in much the same wayas canine distemper does. Wild dogs at highdensities may also be affected and their pop-ulations limited by infection levels. Amongthese diseases are: canine brucellosis (infec-tive agent Brucella canis), which causesabortion and infertility; Chagas’ disease(Trypanosoma cruzi), which may causemyocardial and central nervous systemdegeneration; and tropical canine pancy-topaenia (Ehrlichia canis), an often fatalparasite of the blood associated with thecommon brown dog tick (Rhipicephalussanguineas) (Geering and Forman 1987).Canine brucellosis and Chagas’ disease areboth zoonoses when endemic but humaninfections are few. Geering and Forman(1987) suspect that tropical canine pancy-topaenia may be present in northernAustralian wild dogs although no positivediagnoses have been made.

Dogs are also susceptible to infection bythree other exotic disease organisms: theviruses causing Aujeszky’s disease(Herpesviridae, Alphaherpesvirinae) and

transmissible gastroenteritis (Coronaviridae),and screw-worm fly (Chrysomya bezziana)(Saunders et al. 1999). Wild dogs mightspread transmissible gastroenteritis, a dis-ease affecting young pigs, but dogs areunlikely to be important in spreading theother two diseases if they were to enterAustralia.

3.5 Interactions between wilddogs, marsupial carnivoresand introduced predators

The thylacine (Thylacinus cynocephalus), amarsupial carnivore about the size of thedingo, was once distributed throughoutAustralia, but ‘suddenly’ disappeared fromthe mainland about 3000 years ago (Archer1974; Dixon 1989; Rounsevell and Mooney1995). The Tasmanian devil (Sarcophilusharrisii), a marsupial carnivore about halfthe size of a dingo, was also widespreadthroughout Australia about 4000 years ago,but its population declined and it becameextinct on the mainland about 450 years ago(Jones 1995). Their demise can be attributedto competition with dingoes according to the‘superior adaptability’ hypothesis (Corbett1995a). This hypothesis hinges on the supe-rior social organisation of dingoes duringcritical periods when food supplies werescarce, widely dispersed or clumped, whichusually occurs during drought or after exten-sive wildfire. Only dingoes form large inte-grated packs and cooperate to catch largeprey and to defend carcasses, water andother crucial resources. On the other hand,thylacines hunted alone or in pairs and devilswere essentially solitary so that neither couldsuccessfully compete against the weight ofdingo numbers during those critical periods.

This contention is supported by early recordsof thylacines as having a stiff gait; they proba-bly could not run after their prey (mainlymacropods) as fast as dingoes could. Theyapparently located prey by scent and tired itby dogged pursuit, usually alone, as there areno records or anecdotes of thylacines huntingcooperatively. This apparent lack of packhunting is supported by bushmen’s observa-tions that the thylacine was normally muteexcept for a coughing bark (Rounsevell and


Mooney 1995). Social hunters, such as din-goes, have a large vocal repertoire for com-municating over distances (Corbett 1995a).Similarly, observations of devils in Tasmania,where they are still common, confirm thatthey hunt alone, and that although they cancatch a variety of live prey, they subsist main-ly on carrion such as macropods and sheep(Pemberton and Renouf 1993). It seems thattheir aggregations around carcasses are notcohesive social units, so, as for thylacines, it isquite likely that devils could not successfullycompete with dingoes when food was scarceduring drought and after fires.

‘Dingoes form large integrated packs and cooperate

to catch large prey and to defendcarcasses, water and other

crucial resources.’

Wild dogs may now present foxes and feralcats with a similar kind of competition. Incentral Australia, the most common preyspecies of wild dogs, foxes and feral cats arerabbits and small rodents. During a droughtbetween 1969 and 1972 these prey becamescarce and wild dogs changed their diet tored kangaroos (Macropus rufus) and cattlecarcasses (Corbett and Newsome 1987). Wilddogs were more successful at catching kan-garoos by hunting cooperatively than alone(Section 2.3.1), and stable packs of wild dogsdefended carcasses and waters more success-fully than less cohesive groups (Corbett1995a). During the first year of this drought,cats and foxes were also seen scavenging cat-tle and kangaroo carcasses, but sightings ofthem ceased and their tracks disappearedabout midway through the drought. Theymost likely starved because many emaciatedcats suddenly appeared around homesteadsand many were easily killed by park rangers(Hooper et al. 1973). Wild dogs undoubtedlycontributed significantly to the demise of catsand foxes by their increased monopoly ofcarcasses as the drought persisted. Foxeswere observed avoiding wild dogs at sharedwaterholes and an increase of cat in the wilddogs’ diet was recorded. In any event, therewere no signs of cats or foxes until thedrought broke and after the rabbit and rodentpopulations had resurged (Corbett 1995a).

3.6 Predator–prey relationships

Dingoes and other wild dogs have changedin abundance and status since European set-tlement led to modified ecosystems (Section2.8.4), and so has the nature of their socialityand predation.

3.6.1 Dingo behaviour and predation on cattle

There are potential problems if peak calvingcoincides with the dingo mating season. Forexample, in the Barkly Tableland of theNorthern Territory, most dingoes operate inde-pendently but during the dingo mating season(about four months peaking in March–April),dingoes form temporary breeding groupswhich often comprise one oestrous female andseveral males (Corbett 1995a). The dingo mat-ing season coincides with the peak in calvingand this coincidence contributes to the deathsof many calves (Corbett 1995a). Calves anddingoes are often together at water, which iswhere many attacks occur. In many cases,attacks on calves are probably more of a dis-placement activity than a hunger drive, per-haps because dingoes become frustrated fromcompeting over oestrous (in heat) females andfighting with rival males. Within dingo groups,there are many aggressive interactionsbetween males but actual fighting is uncom-mon because of complex behaviours associat-ed with dingo dominance hierarchies.Aggressive behaviour can be appeased ordiverted by submissive behaviour to avert seri-ous wounding and death (Corbett 1988b).However, a calf cannot appease or divert theaggression as a submissive dingo would, so thedingo, irrespective of social rank, continues toattack, often joined by other dingoes, until thecalf becomes wounded or dies. Calves killedthis way are rarely eaten. Even if the calf sur-vives well enough to be sent to market, themeat is often classified as second class becauseof scars from dog bites. Such loss to the cattleindustry is probably substantial but it is difficultto quantify (Section 3.1). As wild dogs seldomeat calves, examining stomachs or faeces ofwild dogs would be misleading.


3.6.2 Extinctions of native fauna incentral Australia

In central Australia, before the 1930s, 14species of bandicoots (Peramelidae), macrop-ods and rat-kangaroos (Potoroidae) were com-mon in the areas where cattle now graze, butonly five species survive today, and of these,two are rare and endangered (Newsome andCorbett 1977; Morton 1990).

It is probable that a combination of factorsoperated to cause those extinctions anddeclines including:

• habitat fragmentation and modificationfrom heavy grazing by rabbits and live-stock

• increased competition from rabbits andlivestock

• altered fire regimes

• predation by feral cats, foxes and dingoes.

With respect to predation, Corbett (1995a)indicates that cat and fox numbers were lowat the time and that dingoes played a major

role in the demise of those medium-sizedmammals that mostly sheltered on the sur-face amongst grass and shrubs.

The expansion of grazing enterprises from the1930s was due to the establishment of supple-mentary water from artesian bores (Bauer1983), which allowed cattle to graze furtherfrom natural water sources and modify andfragment habitats. This expansion coincidedwith the most severe droughts (Foley 1957)on record and widespread severe grassfires(Friedel et al. 1990) (Figure 8).

During drought the native fauna declined. Incontrast, dingo populations remained highdue to cattle carrion and water provided forstock. Dingo predation on macropod andbandicoots would have become increasinglysevere as dingo populations grew and as theprotective shelters were removed by cattleand rabbits. It is probably no coincidencethat the native mammals became extinct orrare.


Cat

tle n

umbe

rs ('

000s

)

Rain

(mm

)

Year

0 0

200

400

600

800

1000

500

400

300

200

100

600

1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980

Rain

Cattle

Years of extensive wildfires

Years ofdrought

Figure 8: Cattle numbers and rainfall in central Australia from 1874 to 1985. The increase in cattle numbers from the1930s corresponded with the availability of sub-artesian bore water. Droughts are unpredictable and common in aridareas, yet cattle numbers quadrupled between the worst droughts (1924–1930 and 1958–1965). Wildfires became morefrequent in the 20th century. The combination of extensive fierce fires and cattle grazing exposed many medium-sizemammals to predation by wild dogs. This predation was probably a cause of some native species’ extinctions (afterCorbett 1995a).

3.6.3 Factors affecting wilddog–prey interactions withnative prey

In the tropical coastal wetlands of theNorthern Territory, most dingoes live inpacks of three to eight and defend more orless fixed territories. They mostly eat duskyrats (Rattus colletti), magpie geese(Anseranas semipalmata) and agile walla-bies (Macropus agilis). When these prey areunavailable dingoes can switch to a range ofat least 33 species of substitute prey althoughthey usually concentrate on only one or twospecies at a time (Corbett 1989; 1995a).

Wallabies are available all year round where-as the supply of rats and geese varies withwet and dry seasons. Most geese are eaten asfledglings in the dry season. Rats irrupt intohuge plagues on the floodplains about everythree or four years, but they are only avail-able to dingoes during the dry months.Floodplain fauna (rats and geese) are mostlyeaten during the dry months and more forestfauna (wallabies (Macropus spp.) and pos-sums (Phalangeroidea)) are eaten during thewet months (Figure 9).

Climatic conditions influence both when andwhere dingoes hunt particular prey species.This alternation of predation between habi-tats, illustrated in Figure 9, is a well-defined,predictable cycle in which dingoes do notappear to influence the abundance anddiversity of any particular prey.

In the temperate coastal mountains of south-east Australia wild dog–prey interactions aredetermined more by wildfire than by rainfall.Most fires are low to moderate intensity sothat the environment remains fairly stable andfood supplies for wild dogs are usually high.Fires of high intensity, although infrequent,devastate entire forests and change the preyavailable to wild dogs, but the total food sup-ply usually remains high. It is therefore notsurprising that the predatory cycle is lessdefined than in the tropical wetlands. Lowerdensities of wild dogs living in smaller packs(averaging three members) and in smaller ter-ritories are also a consequence of this envi-ronmental stability (Newsome et al. 1983a).

This predatory cycle still alternates betweenconsistently available prey and seasonal

prey. At Nadgee Nature Reserve in south-eastern New South Wales, for example, themain prey are medium-sized mammals (wal-labies, rabbits, possums) and waterbirdssuch as black swans (Cygnus atratus) andcoots (Fulica atra). Both prey types areeaten seasonally and are supplemented bylarge macropods (eastern grey kangaroo)


% occurrence in diet

General model of dingo-prey interactions

% Forest fauna (wallaby + possum)%

Flo

odpl

ain

faun

a (ra

ts +

gees

e)

80

70

60

50

40

3010 20 30 40 50

w

w

w

wd

d d

d

w, wet seasond, dry season

dd

d

w

w

60

90

Seasonally predictible prey(Magpie geese)Eruptive prey(Dusky rats)

Consistently available prey(Agile wallabies)

DRYSEASON

WETSEASON

Figure 9: A model of predation by wild dogs in a pristinecoastal ecosystem in tropical Australia. Predation by wilddogs alternates between habitats, switching from forestduring the wet season to floodplain in the dry after thefloodplains dry out. The main prey are dusky rats and magpie geese in the dry and agile wallabies in the wet(after Corbett 1995a).

(Macropus giganteus) and small mammals(Newsome et al. 1983a). In this ecosystemwild dog predation sometimes affects preydiversity, abundance and population struc-ture, probably because severe wildfireschange habitats and thus alter the composi-tion of the prey base. This encourages wilddogs to either concentrate on a vulnerable,relatively uncommon species, or to prey onan abundant species, thereby relieving thepressure on another species and allowing itspopulation to recover. At Nadgee, wild dogsusually concentrate on macropods immedi-ately after severe fires and sometimes elimi-nate local populations of eastern grey kanga-roos (Newsome et al. 1983a).

‘The predatory cycle alternatesbetween consistently available

prey and seasonal prey.’

The effects of such intense predation is alle-viated when waterbirds are in great abun-dance, as sometimes occurs when severestorms replenish the coastal lakes with waterand food for the birds. When there are nosubstitute prey available after severe fires,wild dog numbers decline soon aftermacropods decline (Newsome et al. 1983a).

In the mountains, where waterbirds are lessavailable, there is no clear cycle. In KosciuskoNational Park, wild dogs mainly hunt wom-bats (Vombatus ursinus), wallabies and rab-bits, and these prey are supplemented by avariety of other species (Newsome et al.1983a). Similarly, in the mountains nearArmidale in north-east New South Wales, themain prey of wild dogs are macropods, espe-cially swamp wallabies (Wallabia bicolor),red-necked wallabies (Macropus rufogriseus)and eastern grey kangaroos (Robertshaw andHarden 1985a; 1985b). Wild dogs concentrateon juveniles (pouch young and young at foot)and reduce macropod recruitment rates soconsiderably that the populations of thesespecies may decline. In some areas, small iso-lated populations of eastern grey kangaroosand red-neck wallabies have been completelyeliminated (Robertshaw and Harden 1986).This happened because enough substituteprey was available to support the wild dogpopulation; otherwise the wild dogs wouldhave moved away or starved.

Another outcome of the wild dogs’ concentra-tion on swamp wallabies was a disruption ofthe usual seasonal pattern of wallaby births(Robertshaw and Harden 1986). Manyfemales ejected their pouch young when pur-sued by wild dogs, but as most of these off-spring were soon replaced (96% of sexuallymature females carried a blastocyst), therewas a continuous output of young instead ofthe usual spring–summer peak. Besides thischange in breeding pattern, the number ofovulations per female increased as predationpressure of wild dogs increased, and so didmale swamp wallabies’ testicle and epi-didymis weights.

3.6.4 Factors affecting wilddog–prey interactions on pastoral lands

Two major environmental disturbance fac-tors in much of Australia have been the intro-duction of exotic animals, especially rabbitsand livestock, and pastoral industry infras-tructure, such as artesian bores and dams. Intemperate zones, land clearing has alsochanged the structure of the landscape. Thisalteration in habitats and introduction ofexotic animals has made some native preyspecies increase and others decrease. Wilddogs have mainly benefited from extra sup-plies of food and water, which have helpedthem to survive drought and increase theirnumbers, but these extra resources have alsochanged the natural pattern of predation.

‘Dingo predation is greatest on small and medium-sized

mammals during flush periods and greatest on

large mammals in drought.’

The interplay between seasons (drought andflush years), native prey and introduced prey(pests and cattle) and predation by dingoes iswell illustrated by a study at Erldunda in centralAustralia (Corbett and Newsome 1987). Whenrains broke the longest drought on record(1958–65), rodents irrupted over widespreadareas and dingoes concentrated on them forabout a year. Then rabbits predominated in thedingo’s diet for the next three years. When


another drought reduced the rabbit popula-tions, predation on red kangaroos increased,even though they became uncommon. Then,as this drought lengthened, cattle began to dieand carrion became more frequent in the diet.This sequential emphasis upon vertebrate preyof increasing body size as aridity increased canbe summarised as a general model (Figure10)which indicates that dingo predation is greateston small and medium-sized mammals duringflush periods and greatest on large mammals indrought. This seasonal variability provides abasis to understand the impact of dingo preda-tion on prey, critical periods when dingoes killcattle, and whether or not predation can regu-late or limit prey populations.

3.6.5 Does predation by wild dogsregulate prey populations?

Predation by wild dogs has often beenassumed to be a cause of fluctuation or lackof fluctuation in some prey populations, asindicated by the following examples:

• Wild dog predation has been assumed toaccount for the contrast in density of redkangaroos and emus on the two sides ofthe Dog Fence between the borders ofQueensland, New South Wales and SouthAustralia. Outside the fence, where wilddogs abound, kangaroos and emus arerare; but the opposite applies inside thefence (Caughley et al. 1980; Pople et al.2000). There is evidence that, in this situa-tion, the predation rate by dingoes on redkangaroos and emus regulates their popu-lations at low densities (Pople et al. 2000).

• Feral goat populations persist only inareas where dingoes are absent or aresubjected to high levels of control(Parkes et al. 1996).

• At Petroi, in the mountains of north-east-ern New South Wales, the occurrence ofswamp wallabies in the diet of wild dogswas proportionally higher than expectedfrom the number of observed wallabies.When wild dog numbers increased, so toodid their consumption of wallabies whichwas soon followed by a marked decline inthe wallaby population (Robertshaw andHarden 1986).

• At Nadgee Nature Reserve, post-fire pre-dation by wild dogs on macropods heldtheir numbers in check for 2–3 years,probably because the fire opened uphabitats and made these prey more vul-nerable (Newsome et al. 1983a).

• In arid central Australia, red kangaroosbecame more vulnerable to predationduring drought, partly because theywere clumped around waterholes andremaining feed. At one site on the plains,kangaroo populations declined during adrought when dingo predation becameprogressively greater, and kangaroosremained low after drought — at about15% of their pre-drought numbers(Corbett and Newsome 1987).

• In the Harts Ranges, near Alice Springs,Northern Territory, red kangaroosdeclined from being common to rareafter a 7.5-year drought, and populationsdid not recover in the subsequent 10years even though pasture was generallybetter than average. There is evidencethat dingoes, whose numbers hadremained stable throughout the drought,mediated competition between rabbits,cattle and kangaroos to the detriment ofkangaroos (Corbett and Newsome 1987).

• In the Fortescue River region of north-west Western Australia, euro (Macropusrobustus erubescens) populations werefairly low in an area where dingo popu-lations were allowed to remain high,because they preyed selectively on par-ticular age classes of euros. When din-goes were greatly culled by a baitingprogram, euro populations immediatelyand dramatically increased (Thomson1992c).

• In the Guy Fawkes River region of north-ern New South Wales there is circumstan-tial evidence that dingoes limited theabundance of macropods (Fleming 1996b;Fleming and Thompson unpublished data1993). It is possible that annual removalsof wild dogs and foxes by aerial baitingincreased the abundance of macropods inbaited areas.



Rabbits

Kangaroos

Cattle

Rodents

Flush years Drought years

% d

ingo

es e

atin

g pr

ey

50

0

100

Smallprey (e.g.rabbits, rodents)

Large prey (e.g. cattle)

Economic calf loss

Flush Drought Flush

% d

ingo

es e

atin

g pr

ey

50

0

100

Beneficial calf loss

Curb increase in rabbits

(a)

(b)

Figure 10: Models of predation by wild dogs in disturbed ecosystems in arid Australia showing: (a) sequential predationon prey of increasing size; and (b) dingo predation on small and large prey. Note, the time scale on the x-axis is longer ingraph (b) (after Corbett 1995a).

These examples provide evidence that dingopredation can affect prey populations, espe-cially macropods, but the long-term effect ofthis predation, and whether or not the effectis regulatory, cannot be assessed withoutexperimental studies (Sinclair 1989). Forsome of the listed examples, alternativeexplanations could be feasible for theobserved changes in prey abundance.

‘Cattle carrion enabled dingoes to survive droughts andsubsequently to concentrate on

red kangaroos and perhaps regulate their population

after a drought.’

In Queensland, analyses of bounties paidannually on wild dogs and feral pigs over 24years indicated that pig mortalities increasedmore than threefold with every doubling ofwild dog numbers, and this inverse relation-ship suggested that predation on pigs bywild dogs was a limiting and potentially reg-ulating factor for populations of feral pigs(Woodall 1983). However, reanalysis ofWoodall’s data indicates that wild dog boun-ties were negatively related to rainfall overthe previous year, suggesting that fewer wilddogs were killed in the 12 months followinghigh rainfall (Choquenot et al. 1996).Correcting for the effects of rainfall removesthe apparent relationship between pig mor-tality rate and wild dog density. Other prob-lems with the analysis of scalp returns areidentified in Section 6.2.1.

There has only been one experimental studyto assess the impact of dingo predation onferal pig populations. This was at Kapalga innorthern Australia and clearly demonstratedthat predation alone did not regulate (seeGlossary) feral pig populations. Instead, pre-dation by dingoes was one factor acting inconcert with interference competition frombuffalo (Bubalus bubalis) to limit (SeeGlossary) feral pig numbers (Corbett 1995c).That is, feral pig numbers would have beenhigher if competition with buffalo and pre-dation by dingoes were absent. Dingoes didnot cause higher pig mortality as pig abun-dance increased.

Evidence from other regions of Australia(Newsome and Coman 1989, Newsome1990, Pech et al. 1992) indicates that preda-tion alone can significantly curb prey popu-lations only when prey numbers are initiallydepressed by a widespread environmentalevent such as drought in arid rangelands orintense wildfire in temperate forests. Thetheory is that prey are trapped in a ‘predatorpit’ (Walker and Noy-Meir 1982) where thereare too few animals successfully breeding forbirths to exceed off-take by predation. At thesame time, the extra food that such circum-stances usually provide allows the predatorsto survive. For example, in arid centralAustralia, cattle carrion enabled dingoes tosurvive droughts and subsequently to con-centrate on red kangaroos and perhaps regu-late their population after a drought.However, in tropical regions of Australia,with the absence of prolonged droughts orother circumstances to simultaneouslyreduce populations of main prey, it is unlike-ly that dingoes could ever limit feral pig pop-ulations to low levels.

3.7 Interactions between humansand wild dogs

There are few records of dingoes attackingor killing Aboriginal people, either in campsor in the wild but such incidents are not like-ly to be reported. There are more reports ofdingo attacks on non-Aboriginal people butmost are anecdotal (Savant 1969). During theRoyal Commission into the Disappearance ofAzaria Chamberlain, evidence was presentedon the deaths of five children caused bydingo attacks and several other dingo attackson children and adults throughout Australiaover the past 50 years or so (Morling 1987).

‘Aggressive behaviour is apparently most common during

the dingo’s breeding season.’

In recent years, dingoes have become a majortourist attraction at sites in outback Australiaand Fraser Island in particular. Consequently,many visitors and residents have deliberatelyor inadvertently fed dingoes to encourage con-tact for close viewing and photographs. This


has led to many dingoes and other wild dogslosing their fear of people and occasionally dis-playing aggression towards people, especiallyat commonly used areas such as campinggrounds and picnic areas. Aggressivebehaviour is apparently most common duringthe dingo’s breeding season. This provision offood for wild dogs by people and the resultingchange in wild dog behaviour has been docu-mented on Fraser Island by Marsterson (1994);Moussalli (1994); Price (1994) and Twyford(1994a,1994b). Current management of thisproblem is addressed in Section 5.3.2.

Wild dog territories that are centred on areasof high human activity, such as townships,town refuge dumps, camping grounds, pic-nic areas and resorts, appear to be smaller insize but have relatively higher numbers ofwild dogs per pack compared to wild dogsthat rely on natural prey in bush areas(Corbett 1998). The nature, frequency andintensity of interactions with people are like-ly to vary depending on the age and sex ofwild dogs, pack size and composition, timeof year, supplementary natural food suppliesand human reactions to wild dogs. Types ofwild dog–human interactions include wilddogs stealing and soliciting food, wild dogsstalking and harassing (nipping, ‘playful’ bit-ing) humans and outright attacks.

Similar human–predator interactions (involv-ing coyotes (Canis latrans) or bears (Ursusspp.)) have been recorded in North America(Howell 1982; Carbyn 1989). There appearto be many parallels in the case of the coyote(the ecological equivalent of the dingo inNorth America) where many attacks aredirected towards young human females,with a baby being killed in one instance(Carbyn 1989).

Interactions between humans and wild dogsmay occur for other reasons. These include:

• Wild dogs regarding humans as competi-tors or intruders into wild dog domainsand thus defending oestrous females,pups and ‘hunting’ areas (garbage sites,camp sites, barbecue areas, beaches).

• Wild dogs (mainly adults) regardinghumans (mainly children) as prey.

• Juvenile and subadult wild dogs ‘play-ing’ with humans.

In relation to the latter example, it is likelythat in places such as Fraser Island andUluru, many generations of dingoes havebeen reared in the close presence of humans(imprinting) so many young dingoes engagein what appears to be ‘playful’ behaviour.These are normally directed towards otherdingoes. In reality, the dingoes are practisingbehaviours that will be vital for their survivalin later life such as nipping and biting toassess the vulnerability of prey (to avoid seri-ous injury to themselves) or to achieve dom-inance amongst litter mates. Humans, espe-cially children, naturally do not understandthis and are most likely to turn and flee(often shrieking) which generally will stimu-late chasing and further aggression from din-goes.


Summary

The management of dingoes and other wilddogs is affected by community attitudes andperceptions. Opinions vary as to the pest sta-tus of dingoes and other wild dogs. Some pri-mary producers view them simply as anunwanted pest to be removed from the envi-ronment while other sections of society viewthem as wildlife or icons to be conserved asfar as possible. Public opinion influences notonly the type of management strategies thatare developed but also the type of controlmethods that may be deployed. Wider publicattitudes rightly demand that the techniquesused for wild dog control must be as humaneas possible and expose non-target animals tominimal hazard. Management strategies thatdo not address or acknowledge broad com-munity attitudes are susceptible to disruptionor interference.

4.1 Community perceptions andattitudes

Community attitudes towards dingoes andother wild dogs are diverse and greatly affectmanagement decisions. Dingoes and otherwild dogs were condemned from the earliesttime of European settlement in Australia asvicious killers of livestock and a threat to thedomestic animals raised by struggling set-tlers. The dingo is also regarded by somepeople as a desirable species for recreationalhunting (Allison and Coombes 1969) andwas hunted by early setters as a substitute forfoxes (Vulpes vulpes) (Rolls 1984). The term‘dingo’ came into usage as an insult whenapplied to a person. Later, the high profile ofthe dingo saw it being used as an advertisingimage (for example, Dingo Flour and DingoBitter beer).

With the increased urbanisation of Australiaand a growing awareness of environmentalissues, other attitudes towards dingoes haveemerged. A recent demographically represen-tative survey of 2000 Victorians of voting ageshowed that 79% of respondents classified

wild dogs as pests regardless of the respon-dents’ background (Johnston and Marks1997). The survey did not ask respondents toclassify ‘dingoes’ as pests or otherwise. Giventhat the majority of respondents regardedother native species (including possums(Phalangeroidea), kangaroos (Macropus spp.)and wombats (Vombatus ursinus)) as ‘non-pests’, the responses may have been differentif dingoes had been separately identified.

‘Nor is there a common view about dingoes and other wild

dogs amongst livestock producers.’

The dingo is viewed by many as an animalwhich has an important place in Australianecosystems, and one which should be pre-served as far as possible (Section 5.3.3). Suchattitudes have been enshrined in legislationin some States and Territories (Sections 3.3and 5.2). Strength of opinion can be influ-enced by the perception of purity of strain;conservationists wish to preserve pure din-goes but see hybrids and feral dogs as athreat to the dingo (Section 2.9). Othersbelieve that because a dingo-sized canid hasbeen present in Australia long enough tohave affected the biodiversity of the commu-nities in which they occur, the removal ofwild dogs from these environments mighthave unforeseen impacts on biodiversity,and that some wild dogs should therefore beretained irrespective of their genetic status.

Nor is there a common view about dingoesand other wild dogs amongst livestock pro-ducers. Sheep graziers would be unanimousin condemning the presence of a single wilddog near their flocks, whereas cattle produc-ers tend to have a variety of opinions. Somecattlemen are ambivalent towards dogs.Others reflect previous or current bad experi-ences of calf predation by wild dogs and reg-ularly control them; for example, 71% ofNorthern Territory pastoralists surveyed in1995 indicated that dingoes were a major orserious pest on their properties (Eldridge and


4. Community attitudes affecting management

Bryan 1995). Another group recognise thepotential role of dingoes in controllingmacropod abundance and are prepared toexperience some losses of calves in the beliefthat overall enterprise productivity is betterwhen there is less potential competition forforage between cattle and macropods(Section 3.6.5). Still others tolerate dingoesand other wild dogs until predation of calvesbecomes apparent and then institute a con-trol program (K. Watters, grazier, New SouthWales, pers. comm. 1984; D. Wurst, Parks andWildlife Commission (Northern Territory),Northern Territory, unpublished data, 1995).Some are strongly opposed to baitingbecause of the perceived risk to their owndogs. There is also the perception amongsome landholders living near national parksand other conservation reserves that wilddogs belong to the government and thereforeresponsibility for wild dog control on adja-cent grazing lands lies with State agencies (J.Burley, Department of Natural Resources andEnvironment, Victoria, pers. comm. 1999). Aswith other groups, trappers hold differentviews on the relative merits of different con-trol methods and the role of wild dogs in ani-mal communities (Ward 1986). The views oftrappers must be considered because thesepeople are often responsible for much of thecontrol effort spent on wild dogs.

In 1981, evidence from a series of publicmeetings and submissions from landholderswithin the original barrier fence inQueensland (Holden 1991) showed that atti-tudes toward wild dogs were explained inpart by the location of their holding in relationto the fence. Similar trends are evident inother States. Those individuals who have first-hand and ongoing experience with livestockpredation by wild dogs generally express thestrongest sentiments. These people often ratewild dogs as their biggest productivity inhibi-tion. Those landholders that are far removedfrom any threat of predation tend to be lessconcerned about wild dogs and may be moresubject to influences from non-agriculturalsectors of the community. Fear of predationby wild dogs has meant that in some areasgreater control effort has been spent on wilddogs than other pests that may be more dam-aging in a less obvious way.

Traditional Aboriginal groups generally per-ceive dingoes and other wild dogs as a com-ponent of the natural landscape with a right-ful place therein. Dingoes often feature inAboriginal mythology and are therefore seenas part of the cultural heritage of Australia.These interests must be kept in mind whendevising management programs for dingoes.In general, conflicts with the views ofAboriginal people rarely occur because ofthe alienation of traditional Aboriginal peo-ple from most of the grazing areas wherecontrol work is undertaken. However, whereAboriginal people maintain an ongoingaffinity to these areas their concerns must beconsidered and included in the planningprocess.

‘In some areas greater control effort has been spent onwild dogs than other pests thatmay be more damaging in a less

obvious way.’

Public perception of wild dogs as objects offear will also influence attitudes towardslocal management. Without doubt, the mostpublicised case of dingoes as predators ofhumans involved the alleged taking of ababy, Azaria Chamberlain, by a dingo atUluru in central Australia in 1980. Reaction tothe allegation led to an immediate controlprogram around the camping area where theChamberlains were staying. Local controland education programs have been instigat-ed in other national parks and naturereserves in response to attacks on people(for example, Fraser Island; AustralianAssociated Press 1998).

As a result of these various views in society,as well as the results of scientific research,the approach to wild dog control has alteredconsiderably in the past 20–30 years(Chapter 5). Management strategies nowfocus on problem areas, that is, in livestockpaddocks and on neighbouring land. Forexample, dogging or baiting forays are nolonger made into the vast areas beyond pas-toral leases in Western Australia. Indeed,such control work would be condemned asbeing economically and environmentallyundesirable.


Nevertheless, scientific research and analysisdoes not always alter management practices.For example, the usefulness of bounties hasbeen discredited in a number of reviews(Section 5.1.1) yet the recent proposed removalof bounties from Queensland legislation result-ed in heated debate (C. McGaw, Department ofNatural Resources, Queensland, pers. comm.1999). A compromise allowing the payment ofbounties by local governments resulted frompressure by some community groups.

Attitudes about controlling wild dogs areinfluenced by the control methods used. Forexample, pressure from animal welfaregroups has resulted in changes to legislationand policy for the use of leghold traps inVictoria, the Australian Capital Territory andNew South Wales. Some sections of the com-munity do not accept the use of poisons, andwould oppose any control campaigns basedon baiting (Section 4.2.4). Indeed, some peo-ple regard the killing of an animal, even if itis considered to be a pest, as wrong.

4.2 Animal welfare issues

4.2.1 General

Animal welfare groups aim to ensure that allanimals are treated humanely and thatactions that cause stress and suffering shouldbe minimised as far as is feasible. Groupssuch as the Royal Society for the Protectionof Cruelty to Animals (RSPCA) and theAustralian and New Zealand Federation ofAnimal Societies (ANZFAS) accept that con-trol of wild dogs may be required in certaincircumstances, but advocate restrictions onthe type of control techniques that are used.For example, the RSPCA believes that the useof 1080 (sodium fluoroacetate) poison onnative species is unacceptable, and that itsuse against introduced species should bestrictly controlled by legislation. Non-lethalmeans such as exclusion fencing are encour-aged.

Conversely, owners of livestock have legaland moral obligations under the various StatePrevention of Cruelty to Animals Acts to pro-vide the best husbandry possible for their live-stock and there are model codes of practice to

ensure this (Animal Health Committee, of theStanding Committee for Agriculture 1990).Predation by dogs is not humane and mostgraziers are concerned by the suffering anddistress imposed on their animals. This is par-ticularly so when surplus killing and injuryoccurs. The adverse effects of predation onthe welfare of livestock is likely to influencelandholders’ attitudes to management of wilddogs, including the control and conservationof dingoes. Counter to this is the argument thatsheep should not be grazed in areas wherepredation by wild dogs is likely and poses thequestion whether governments should sub-sidise wild dog control in such areas.

‘Counter to this is the argumentthat sheep should not be grazed

in areas where predation bywild dogs is likely.’

Clearly, welfare concerns must be consid-ered in all control programs involving lethaltechniques. These should be as target-specif-ic as possible and take into account maxi-mum welfare for the target species as well aswelfare issues relating to the accidental cap-ture or killing of other species.

4.2.2 Shooting

Shooting by skilled marksmen is probablythe most humane method of controlling wilddogs and for this reason is the favoured con-trol technique of the RSPCA. The objectiveshould be to kill the animal as quickly andcleanly as possible with a shot to the head.In some States and Territories, a Code ofPractice or government agency policy speci-fies the minimum specifications (calibres,projectile weight, range) for the shooting offeral or pest animals (Section 6.4.2).Shooting can usually only be viewed as anopportunistic method of wild dog control,although it can sometimes be useful to targetindividual animals inside sheep zones. It isnot a cost-effective option for reducing pop-ulations of wild dogs.


4.2.3 Trapping

Steel-jawed leg-hold traps have been tradi-tionally used in Australia for wild dog control.These traps are generally viewed as inhu-mane, with the bare steel jaws causing tissuedamage, serious cuts, broken bones, dislocat-ed joints and sometimes amputation of feet ofthe captured animal (Fleming et al. 1998). Thedegree of injury and suffering of a trappedanimal is also related to the length of time thatthe animal spends in the trap (that is, howoften traps are checked) and whether anal-gesic drugs (C. Marks, Victorian Institute ofAnimal Sciences, Victoria, pers. comm. 1998)or poisons have been applied to the trap toreduce stress or hasten death (Fleming et al.1998). As well, the sites where traps are setcan greatly affect the likelihood of catchingnon-target animals (Newsome et al. 1983b).Different groups of non-target animals sufferdifferent levels of injury (Fleming et al. 1998).For example, possums mostly had minorinjuries in Soft Catch‚ traps whereas 73% ofvaranids experienced major trauma.

There is an increasing awareness of the inhu-maneness of steel-jawed traps. Modificationsto existing and new traps and capture devicesthat are more humane continue to be devel-oped and are progressively being adopted.The Victorian treadle-snare and the VictorSoft Catch‚ padded trap and modified Lanestraps have been shown to result in fewerinjuries to trapped animals (Meek et al. 1995;Fleming et al. 1998) and their use is preferredover steel-jawed traps by animal welfaregroups. In Victoria, the treadle snare hasreplaced the steel-jawed trap for use in wilddog control, except in special circumstances.In New South Wales, steel-jawed traps areoutlawed, though the use of padded trapsand treadle snares is allowed. These must beused in accordance with a code of practicethat specifies, amongst other things, that trapsshould be checked daily. In South Australia,steel-jawed traps are outlawed in the 60% ofthe State outside the Dog Fence. Trapping isstill the preferred technique to target dingoeskilling sheep inside the fence and where poi-soning has proved unsuccessful.


Padding steel-jawed traps should be promoted to improve the humaneness of trapping (Source: P. Thomson).

‘There is an increasing awarenessof the inhumaneness of

steel-jawed traps.’

In some areas, the practice of applyingstrychnine poison to the jaws of traps isadvocated. The strychnine is bound into acloth and the trapped animal bites on thesoft material and ingests the poison. Thisresults in a rapid death, and is seen by someto be preferable to the animal remaining inthe trap until dying of exhaustion or expo-sure, or being discovered many hours laterby the trapper. However, strychnine isclassed as an inhumane poison (Section6.4.4) and its use for other purposes is notsanctioned. Strychnine cloths are generallyfavoured in the more remote pastoralregions, where trappers cover large areasand cannot return to their traps within a rea-sonable time. Additionally, many doggersbelieve that traps set for rogue animalsshould not be visited too often as the pres-ence of human scent near the trap may deterthe target animal (B. Morris, dogger, YassRural Lands Protection Board (New SouthWales), pers. comm. 1998). Daily checkingof traps is simply not a practical option in allcircumstances. In Western Australia andSouth Australia, steel-jawed traps can beused for wild dog control only if strychnineis applied to the traps. In New South Wales,strychnine cloths must be applied to trapsthat are not checked daily. A fast-acting,more-humane replacement poison forstrychnine on trap jaws would be preferableand requires investigation. (Section 8.7). Theuse of 1080 poison on traps is not practicalbecause it is too slow acting.

4.2.4 Poisoning

The RSPCA is generally opposed to the useof poisons (RSPCA 1997) but may accept tar-get-specific baits containing humane toxins.It is widely recognised that poison baitingremains one of the few viable options avail-able to control wild dogs, particularly in themore remote rangeland areas (Section 6.4.4).

Strychnine baits are viewed as inhumanebecause the affected animals remain con-scious and appear to suffer pain and anxiety

from the onset of clinical signs through todeath from asphyxia and exhaustion. Theclinical signs of strychnine poisoning in dogsare: deep reflexes and cramping of muscles,particularly in the legs; muscular spasms thatincrease in severity and extent through todeath from respiratory failure; and vomitingand diarrhoea.

1080 is now the preferred poison for use inwild dog control throughout Australia. It is afar more selective poison than strychnineand poses minimal risk to non-target ani-mals. Canids are particularly susceptible to1080. Many other mammals are less sensitiveto the poison, particularly many nativespecies in Western Australia that haveevolved an enhanced tolerance due to expo-sure to plants containing 1080 (King 1984).Birds and reptiles are in turn less susceptibleto 1080 than mammals.

‘Research is underway to develop an orally-active analgesic

for incorporation into baits tocounter debate about the

humaneness of 1080.’

1080 causes the blocking of the Krebs cycle,the major cellular biochemical mechanismfor releasing energy from food. In dogs, theprimary action of 1080 is on the central ner-vous system. Symptoms appear after a latentperiod of up to several hours, the periodvarying according to the amount of poisoningested. Dogs become excited, maybecome nauseated, frequently howl, andexhibit running fits (McIlroy, 1981). The finalphase of poisoning involves continuousmuscular contraction and death through lackof oxygen supply to the respiratory centre.During convulsions, affected dogs areunconscious and appear to be unaware oftheir surroundings, suggesting that they arenot suffering pain or distress (Gregory 1991).Cases of sub-lethal human poisoning sup-port the view that 1080 is a relativelyhumane poison; victims convulsed but laterreported no recollection of pain or physicaldistress (Gregory 1991). In Victoria, researchis underway to develop an orally-active anal-gesic for incorporation into fox baits tocounter debate about the humaneness of


1080 and to reduce regurgitation and multi-ple bait take by foxes (C. Marks, VictorianInstitute of Animal Sciences, Victoria, pers.comm. 1997). Success in this work wouldalso have application to wild dog baiting.

4.2.5 Control at breeding dens

In the past, doggers searched for dens eachbreeding season, in order to fumigate or cap-ture the pups, bitch, and any other associat-ed adults. This activity was aided by the useof sniffer dogs to seek out the dens. As forother control techniques, the associated wel-fare issues need to be considered. The prac-tice of blocking off the den entrances if thepups cannot be retrieved is clearly unaccept-able from a welfare aspect.

4.2.6 Exclusion fencing

The use of exclusion fencing is generallyregarded as causing fewer animal welfareimpacts than lethal control means. However,fences alter the movements of larger ground-based animals and care must be taken withdesign and placement of fences so as not toadversely affect their survival. Electric dingofences can trap and kill macropods, emus(Dromaius novaehollandiae) and echidnas(Tachyglossus aculeatus).

4.2.7 Biological control

Wild dogs are susceptible to many diseases(Section 2.7). It is unlikely that the release ofdiseases already present in the wild wouldbe effective for biological control. Given thefact that lethal diseases generally cause suf-fering and emaciation, and also affectdomestic dogs, this type of control is unlikelyto be supported by welfare groups or thegeneral public.

Other forms of biological control may be pos-sible in the future. The RSPCA supports theconcept of hormonal control to limit repro-duction (RSPCA 1997). The RSPCA and ANZ-FAS strongly support the development of fer-tility control measures, such as immunocon-traception (Tyndale-Biscoe 1994), as humanetechniques for controlling pest animals.Immunocontraception research is still in itsearly stages and the technique has not yet

been used for the control of any pest species.Domestic dogs and desirable populations ofdingoes would be susceptible to any form ofbiological control targeting problem wilddogs. Transmissible forms of fertility controlsuch as immunocontraception are thereforeunlikely to gain favour with the general com-munity.

4.3 Public health issues

4.3.1 Diseases and parasites

There have been a number of publicity cam-paigns aimed at reducing the prevalence ofhydatidosis (causal agent Echinococcus gran-ulosus) in humans (D. Jenkins, AustralianHydatids Control and Epidemiology Program,pers. comm. 1998). Although occurrence israre, the disease can be fatal. In areas wherehydatid infection of domestic dogs and live-stock is endemic, awareness of the disease ishighest. However, few people are aware ofthe sylvatic hydatids cycles between wild dogsand wildlife or between wild dogs and live-stock (Section 3.4). Local awareness is likely toincrease participation in programs to controlwild dogs and increase the pressure on gov-ernment instrumentalities to control wild dogsat the interface of public and grazing lands.

If rabies (Rhabdoviridae) ever became estab-lished in Australia, there would be a muchgreater public impetus to control all free-roaming dogs. Trade-offs between the desireto conserve dingoes and concerns abouthuman health would require much discus-sion, as the range of stakeholders involvedwould increase substantially.

4.3.2 Interactions with humans

Aboriginal people sometimes used dingoes forhunting, as food, decoration (fur), mobile blan-kets or currency (scalps) (Meggitt 1965;Hamilton 1972) but preferred Europeandomestic dogs when they became available.These dogs remained in the camps and thusobviated the need to seek new dingo pupseach year or to retain adult animals by break-ing one of their legs. Domestic breeds alsobarked and thus were better sentinels(Hamilton 1972; White 1972).


Free-roaming dogs are still common inremote Aboriginal settlements and the pres-ence of these dogs has human health impli-cations for the communities (Section 3.4).Human health programs that involve con-trolling wild dogs in and around Aboriginalcommunities need to address the potentialconflicts that arise from traditional Aboriginalviews of dingoes and other wild dogs(Sections 3.3 and 3.7).

4.4 Conservation issues

Community attitudes to dingoes havechanged considerably since the early days ofEuropean settlement when they were seenonly as a threat to the fledgling pastoralindustries. Research in the last 30 years hasgreatly increased the understanding of din-goes and led to more rational and effectivemanagement techniques that significantlyreduce the risk that dingoes pose to the pas-toral industries. Concurrently, there has beenan increase in public awareness of theimportance of conserving biodiversity.Dingoes are now commonly perceived asnative fauna and are formally recognised assuch in some States (for example, New SouthWales). Consequently, there is now a publicexpectation that dingoes will be conservedas part of Australian biodiversity.

‘There is now a public expectationthat dingoes will be conserved aspart of Australian biodiversity.’

During the past 30 years, there has been con-siderable conflict between conservationistsand managers of wild dogs. Three issueshave been central to this conflict. The first isthat the public perception of dingoes as anative species creates an expectation thatthey will be conserved. Thus, some sectionsof the public are opposed to all wild dogcontrol, particularly when it is conducted inconservation reserves.

Resolution of this conflict has been achievedin some States and Territories by attributing adifferent status to wild dogs dependent ontheir location. For example, in New SouthWales, the National Parks and WildlifeService effectively recognises dingoes as

native fauna, and while they are ‘unprotect-ed’ in conservation reserves, they are not adeclared noxious species until they move offthe reserve (Chapter 5). Control within con-servation reserves is only permitted whendingoes emanating from there are implicatedin stock losses and measures conducted out-side the reserve have failed to solve theproblem.

‘Resolution of this conflict hasbeen achieved in some States and Territories by attributing a different status to wild dogsdependent on their location.’

The second issue arising from this is thedilemma of how to differentiate dingoesfrom other wild dogs. This is compoundedby the exhibition of dingo coloration inmany hybrids. While DNA fingerprintingmight determine the frequency of dingo anddomestic breed genes, this technique doesnot address public perceptions. The geneticsof a wild dog is largely irrelevant to manypeople. These people are more concernedwith the appearance of a particular wild dogand compare it with their concept of a dingo.For conservation directions to be set, publicand expert inputs must be collected. A deci-sion is needed on what proportion of a wilddog population is required to be pure dingobefore conservation efforts are instituted.The feasibility of conservation strategies onthe mainland also needs investigation. Afterthese decisions are made, conservation poli-cy can be formulated.

The third issue is the risk to non-targetspecies during wild dog control programsand particularly during aerial baiting pro-grams with 1080. Many concerns about non-target risks have been allayed by the seminalworks on the toxicity of 1080 to Australianfauna by John Mcllroy and Dennis King andtheir collaborators (McIlroy 1981). In someStates, reductions in loadings of 1080 inbaits, better bait placement and a reductionin the number of baits used have resulted in atheoretically negligible risk to non-targetspecies (Fleming 1996a). However, becauseof public misconceptions or lack of educa-tion about 1080, there remains some concern


about non-target risk during wild dog baiting(Section 8.11). This issue needs to beaddressed by a public education programand by field-based studies to assess non-tar-get risk and confirm that wild dog baiting istarget-specific and effective.

The commonly held but unproven belief of auniversal, inverse relationship between wilddog and fox abundance (Section 2.10.1) hasbeen popularly interpreted to mean that con-serving dingo populations may limit the dis-tribution and abundance of foxes and thusreduce their impact on small and medium-sized native animals. If this proves true, con-serving dingoes and other wild dogs couldenhance the conservation of native speciesthreatened by foxes.


Summary

In the past, legislation for the managementof wild dogs has included punitive Acts andActs dealing with the conservation ofwildlife. Dingoes have been included withother dogs in early colonial legislationdesigned to remove troublesome dogs and toreduce the threat of predation of livestock.Management of wild dogs relied heavily onlabour-intensive techniques of trapping,shooting and ground baiting, with bountypayments being offered as an incentive tokill dogs. With cheap labour and materials,and a profitable grazing industry, the erec-tion of exclusion fencing was also feasible insome parts of Australia, and became anessential element of wild dog control in thoseareas. Much of the control work was reac-tive, dealing with problems as they arose.Nevertheless, some strategic, preventativecontrol was carried out, and in extreme butmisguided cases, efforts were made to con-trol wild dogs hundreds of kilometres fromthe nearest sheep grazing enterprises.

The dingo is extinct in much of the sheepand cereal production zones of eastern andsouthern Australia because of habitat modi-fication and the success of early poisoningcampaigns. The areas largely without wilddogs are mostly separated from areas wherethey are present by dog-proof fences thatwere erected around the turn of the century.

In most States and Territories, obligations orprovisions are made for the destruction ofwild dogs in sheep and cattle grazing zones.The Dog Fence, which runs for about 5600kilometres from Fowlers Bay in SouthAustralia to south of Dalby in south-easternQueensland, is maintained or subsidised bygovernment agencies. Poisoning programsform the basis of most lethal control efforts,although trapping and shooting are impor-tant in some contexts. Groups and boardshave provided finances that allow for bettermanagement.

The scientific information on the biologyand movements of dingoes and other wilddogs did not begin to accrue until the late

1960s. Since then, research has served tocorrect or refute much of the folklore andmythology surrounding dingoes andenabled management programs to be moresoundly based. In addition, research hasallowed for the objective evaluation of con-trol techniques and strategies. For example,the demonstration of the effectiveness ofaerial baiting with 1080 in pastoral regionsof Western Australia led to this techniquebeing adopted far more widely than previ-ously. A further influence on wild dog man-agement has been a growing public interestin conservation and animal welfare; con-trol programs had to be not only effectivebut to show due regard to welfare and risksto non-target species.

Current management strategies focus on theobjective of minimising the impact of preda-tion on livestock, not on killing wild dogs.Aerial baiting with 1080 baits forms amajor part of most management programsand is primarily targeted at limited zonesand buffers adjacent to livestock grazingareas. Larger coordinated campaigns havegenerally been adopted, being more efficientand effective than localised, ad hoc efforts.Ground-based baiting and trapping are stillcarried out, although to a lesser extent thanearlier times. Far fewer professional doggersare now employed, although they still playan important role in targeting specific ani-mals and in monitoring buffer areas.Bounty payments have not been successfulin reducing predation by wild dogs and aresubject to abuse. The use of government-sub-sidised bounties should cease.

The greatest threat to the survival of dingoesis hybridisation with domestic dogs. In themore settled coastal areas of Australia, andincreasingly in outback Australia, the barri-ers between domestic dogs (feral andowned) and dingoes are rapidly beingremoved. Hence hybridisation is becomingmore common and the pure dingo gene poolis being swamped. Already in the south-east-ern highlands, more than half of wild dogsare hybrids. The extinction of pure dingoeson the mainland is probably inevitable


5. Past and current management

unless there are changes to community atti-tudes and government policies on wild dogs.In particular the keeping of ‘dingoes’, whichare often hybrids that later ‘go bush’, canincrease the rate of hybridisation of wilddingo populations.

Conservation of dingoes was indirect untilthe 1970s and 1980s when dingoes werelisted on conservation schedules in someStates and Territories. Policy and legislationto encourage the conservation of pure din-goes is required in some States and a con-certed nation-wide effort is needed to ensurethat dingo conservation is not thwarted byconflicting legislation. Simultaneously, thecontrol of wild dogs, including dingoes,must be permitted where there is a recenthistory of livestock predation.

The main hope of conservation is to educatepeople about the plight of dingoes and tomanage pure dingoes on large islands suchas Fraser Island and Melville Island.

5.1 Past legal status and management

5.1.1 Control measures and legislation

During the 1800s, the combination of clear-ing for farming, exclusion fencing, poisoningand trapping resulted in the dingo becomingextinct over much of its previous range insouthern Queensland, New South Wales,Victoria and South Australia. By 1889, allmainland States and Territories had enactedlegislation to facilitate and administer thecontrol of wild dogs. Control was organisedby government agencies or regional semi-government organisations, run by boards oflocal landholders, and funded by govern-ment and by rates levied on landholders.There were four elements to wild dog con-trol.

The bounty system

Bounties were paid on presentation of a wilddog scalp to the appropriate authority. Thefirst record of a bounty system is from 1836in the Melbourne district (Breckwoldt 1988).A bounty system was soon introduced in all

mainland States and Territories and persisteduntil recently. The aim was to create anincentive for the control of wild dogs. Thisencouraged the perception that every wilddog was a sheep killer regardless of itsaccess to sheep, the hunting and trapping ofwild dogs by individuals rather than groups,and fraudulent claims (Tomlinson 1958a).

‘Despite the millions of dollarspaid in bounties, there is little

evidence that the bounty systemis or was an effective tool for

managing wild dogs.’

The control value of bounties has long beendebated. In 1930, the Royal Commission intothe Dingo and Stock Route Administration inQueensland concluded that bounties shouldnot be paid as they were subject to fraud(Holden 1991). Tomlinson (1958a) outlinedarguments for and against the system, con-cluding that the evidence was overwhelm-ingly against bounties. A resolution recom-mending that all bounties in Australia bestopped was passed by the Vertebrate PestsCommittee in 1975 (Smith, 1990). Bountieswere accordingly reduced to $2 (around $10in current values) in South Australia in 1975and were phased out in 1990. Anotherinquiry, held in Queensland in 1975, recom-mended that the bounty system be abolishedthere (Smith 1990). Saunders et al. (1985)proposed significant changes to wild dogmanagement in New South Wales includingthe abolition of bounties. Because of thatreview, bounties are no longer paid for wilddog scalps in the Eastern Division of NewSouth Wales. A wide-ranging review ofbounty payments by Smith (1990) concludedthat all bounty systems were subject to fraud,were ineffective in creating incentive, notcost-effective and were not related to wilddog abundance. More recently, Hassall andAssociates (1998) concluded that, world-wide, most bounty schemes have failed todeliver effective vertebrate pest control. Inresponse to the recommendations of Smith(1990), new legislation in Queensland willremove the provision for bounties althoughlocal governments can still pay them. (C.McGaw, Department of Natural Resources,Queensland, pers. comm, 1999).


Despite the millions of dollars paid in boun-ties, there is little evidence that the bountysystem is or was an effective tool for manag-ing wild dogs. In north-eastern New SouthWales, Harden and Robertshaw (1987) foundthat between 1958 and 1983, two-thirds ofbounties were paid to graziers, the group atrisk from wild dog predation. The incentiveof the bounty was questionable because ofthere was no relationship between the valueof a bounty and the number of bountiesclaimed.

Professional doggers

The employment of professional doggers bygovernment agencies, wild dog controlorganisations, and sometimes by groups oflandholders, was an important part of theimplementation of wild dog control.Doggers were responsible for both strategicand reactive control of wild dogs by trap-ping, shooting and poisoning. They weresometimes offered substantial additionalbounties by landholders to kill particularlytroublesome dogs.

‘As fencing materials became more sophisticated and

more readily available, the use ofexclusion fences became

feasible.’

In some areas (for example, south-east NewSouth Wales, Victoria, the Australian CapitalTerritory and some parts of WesternAustralia) doggers are still an integral part ofongoing wild dog control. In much of range-land Australia, however, increasing labourcosts and the introduction of mandatorywages for Aboriginal workers in the 1960s,plus the declining profitability of the pastoralindustry, resulted in most stations ceasing toemploy specialist doggers. For these rea-sons, and because of the greater adoption oflarge-scale aerial baiting, the number of gov-ernment-funded doggers has also fallen dra-matically.

Shepherding and exclusion fencing

Before the extensive fencing of pastoral runsto manage the movements of sheep, the first

method used to reduce predation by wilddogs was shepherding of flocks by paidshepherds. Shepherds were often sent intoisolated areas where they had to protect theirstock from human and wild dog predation.Clearing and fencing of pastoral lands andextensive strychnine baiting programspushed wild dog populations towards thefringes of the ‘improved’ country as graziersradiated from the central settlements.Shepherding to prevent predation by wilddogs is now only practised as a last resort (C.Young, grazier, New South Wales, pers.comm. 1984) because of the expense andtime constraints.

As fencing materials became more sophisti-cated and more readily available, the use ofexclusion fences as barriers to wild dogmovements into sheep country became fea-sible. Exclusion fencing for wild dogs beganat least 100 years ago with the erection of 13-wire, 1.8 metre high fences laced with verti-cal wires at about 15 cm intervals (Harden,unpublished data 1991), and becamewidespread after the introduction of wirenetting. Often a continuous fence resultingfrom adjoining landholders fencing aroundtheir own properties, protected groups ofproperties. As examples: in Western Australiafences were erected around holdings tokeep emus (Dromaius novaehollandiae) outand were maintained dog-proof (Holden1991); and at least 1000 kilometres of non-continuous barrier fencing was erected bylandholders on the New England tablelandsin New South Wales in the 1920s and 1930s(NERDA undated c. 1966). Similarly, in SouthAustralia, about 3800 kilometres of privatedog fences were erected by 1908. Between1896 and 1908, an additional 5000 kilometresof fencing was erected to maintain the StateVermin Fenced Districts of South Australia(Holden 1991).

Governments and control organisations whol-ly funded or subsidised the erection of otherbarrier fencing. The best known such fence(known as the Dog Fence or barrier fence)extends 5614 kilometres from near Dalby insouth-eastern Queensland to Fowlers Bay onthe Great Australian Bight in South Australia(Figure 2). Prior to shortening of theQueensland section of this fence in 1989, itwas 8614 kilometres long (Breckwoldt 1988).


The Queensland–New South Wales fence(359 kilometres) was originally built as a rab-bit-exclusion fence. This failed and it wasconverted to a dog-proof fence in 1914.However, the agreement between the Statesrequires that the fence be maintained rabbit-proof. The fence between South Australia andNew South Wales was converted from a rab-bit-proof fence to a dog-proof fence in 1917.The New South Wales Wild Dog DestructionAct 1921 placed the responsibility for theexclusion fence under the Western LandsCommission and an amendment to the Act in1957 established the Wild Dog DestructionBoard (WDDB) which retains responsibilityfor the fence and the payment of bounties forscalps. The South Australian dog fencereplaced a series of separately fenced vermindistrict fences in 1947 following introductionof the Dog Fence Act 1946. Responsibility forthe fence is with the Dog Fence Board, mem-bership of which is mostly landholders.

Poisoning

From the early 1800s, when strychnine wasfound to be useful for poisoning wild dogs,control programs were instigated at theproperty level or cooperatively. Cooperationbetween landholders was necessary becausestrychnine was expensive and could only beimported in quantities too large for individu-al landholders. Stockmen carried strychninethat they inserted into carcasses they found,and in some areas bait stations were estab-lished and maintained. In 1946, a manufac-tured brisket fat and strychnine bait wrappedin paper (the ‘Minty’ bait) was developed inQueensland and was subsequently used inQueensland, New South Wales, WesternAustralia and the Northern Territory.

Aerial baiting began with experimentaldrops of the Minty bait in Western Australiaand Queensland in 1946 (Tomlinson 1954),and continued for a number of years there,in the Northern Territory and in SouthAustralia. It was also used on the coast andtablelands of New South Wales from 1957.

Since the mid-1960s, 1080 (sodium fluoroac-etate) has largely replaced strychnine inbaits. However, strychnine can still be usedin baits in South Australia, Queensland and

parts of Western Australia. Because 1080 isclosely regulated, baiting programs areunder much tighter control than previously.Both fresh meat and manufactured baits areused.

‘Aerial baiting was generallyregarded as successful and

many dog fences were allowed tofall into disrepair.’

1080 meat baits were first aerially distributedin the Northern Tablelands of New SouthWales in 1964 and had replaced strychninebaits in aerial baiting programs in most areasby the late 1960s. Aerial baiting was general-ly regarded as successful and many dogfences were allowed to fall into disrepair.The reduction in the area of Queenslandprotected by the barrier fence could beattributed to the perceived success of 1080baiting, especially aerial baiting (Holden1991). Fixed-wing aircraft were used until1986, when helicopters became mandatoryfor aerial baiting in the east of New SouthWales because baits could be placed withmore accuracy (Thompson et al. 1990).Aerial baiting with 1080 has evolved consid-erably to increase its efficacy against wilddogs while reducing potential non-targeteffects. It is now generally accepted as acost-effective, safe method for the extensivestrategic management of wild dogs(Thomson 1986; Thompson and Fleming1991), and is used in Queensland, NewSouth Wales, Western Australia and theNorthern Territory.

Aerial baiting is the major control method inWestern Australia where, in 1996–97, 823 900baits were dropped during 505 flying hours;the average number of baits used annuallyhas varied little over the past decade.Similarly, in north-eastern New South Wales,aerial baiting using helicopters is the primaryform of wild dog control.

Fixed-wing aircraft are used for baiting inQueensland, western New South Wales andWestern Australia but not in the other Statesand Territories.


5.1.2 Past management strategies

When labour was cheaper, and before scientif-ic information about dingo movements andbehaviour began to emerge, many controloperations took place that would now beviewed as inefficient and inappropriate. InWestern Australia, for example, there was astrong belief in dingo migration routes from fardistant ranges and deserts to the grazing leases(Tomlinson 1958b). The attitude that any dingokilled would be ‘one less to attack a sheep’ pre-vailed. This led to major expeditions beingmounted to seek out dingo ‘breeding areas’(Tomlinson and Blair 1952), and to consider-able effort being expended in attempts to con-trol dingoes in these remote areas, hundreds ofkilometres from the closest livestock.

Despite much wasted effort in earlier times,there was still the recognition that a coordi-nated, community approach to wild dogcontrol was the most effective strategy. Thisgenerally took place in the form of ‘baitingdrives’, including aerial baiting (Tomlinson1954). Tomlinson (1958b) wrote:

‘Wild dog baiting drives, organised on a dis-trict-wide basis and combining all availablemanpower and aids such as aerial baiting,are without doubt the most effective destruc-tion method…. Careful planning andorganisation to ensure the work is properlycoordinated, is carried out at the best possi-ble time, and gives the most effective cover-age, is essential. Possibly, the most importantrequirement is to secure the participation ofthe landholders in these drives and the con-tinuation of control work afterwards.’

Over the years, this approach has beenrefined, promoted and increasingly adopted,as outlined in Section 5.1.3.

5.1.3 Historical /past organisationof control

Since European settlement, two organisa-tional levels of wild dog control have exist-ed:

• By 1889, all mainland States had enactedlegislation to facilitate the control of wilddogs (Breckwoldt 1988). Semi-govern-ment administrative structures, usuallygoverned by boards of local landholders,

were created and empowered to levy rateson landholders to fund the bounty systemand other control measures. These fundswere commonly supplemented by govern-ment subsidies. For example, until themid-1980s, most of the control of wilddogs in Victoria was by Government-employed doggers. Groups to facilitate thecontrol of wild dogs were common in east-ern Australia where the resources of thelocal groups were supplemented by gov-ernment contributions and rates collectedby the boards of management. Until recenttimes, wild dog control groups often heldimportant social significance. Dog drivesand poisoning programs served as meet-ings where neighbours could get togetherto socialise and discuss other issues affect-ing their holdings.

• Many properties, leases and runsemployed doggers and boundary riderswho had responsibility for trapping ‘ver-min’ and maintaining fences in dog-proofcondition. Privately employed doggerswere more common in the extensive pas-toral leases in northern central Queens-land, South Australia and WesternAustralia.

5.2 Current legal status (aroundAustralia)

5.2.1 Legislation

The legal status of dingoes and other wilddogs varies between States and Territories(Table 4). This status affects the control mea-sures that are applied and the level of coop-eration between individuals and groups.

Western Australia

Dingoes and hybrids are ‘declared animals’under the Agriculture and Related ResourcesProtection Act 1976 and are placed intoCategories A4, A5 and A6. These categorisa-tions are determined by the AgricultureProtection Board (APB) and are administeredby Agriculture Western Australia (AGWEST).Populations must be controlled and animals


cannot be introduced or kept in captivityexcept in approved institutions or under apermit which carries specific conditions.Although category A5 requires that popula-tions should be controlled throughout theState, it is recognised that dingoes pose nothreat in vast areas beyond the limits of pas-toral or agricultural land. APB policy restrictscontrol activities to stocked land and itsimmediate environs. Domestic dogs (runwild, feral, or being at large) are classified inCategory A5, meaning that they must be con-trolled. In municipal areas, domestic dogs arecovered by the Dog Act 1976.

Dingoes are covered by the Western AustralianWildlife Conservation Act 1950, administeredby the Department of Conservation and LandManagement (CALMWA). Under this Act, theyhave been listed as ‘unprotected fauna’,although they are not subject to control in faunareserves and National Parks without appropri-ate consultation between CALMWA, landhold-ers and AGWEST. Despite the declared pest sta-tus of dingoes outside conservation estates,they are not controlled over most of their range.

Part of the funding for wild dog control inpastoral areas of Western Australia is derivedfrom rates levied on pastoral leases matchedby government funds.

Northern Territory

Dingoes are undeclared in all areas of theNorthern Territory (Table 4). They areunprotected in all areas of the NorthernTerritory outside parks, reserves, sanctuaries,wilderness zones and the Arnhem LandAboriginal Reserve (Parks and WildlifeConservation Act 1993). Although there isno obligation on landholders to controlthem, the Parks and Wildlife Commissionorganises aerial and ground control pro-grams if requested by graziers. The dingohas been protected within national parksand nature reserves since 1984 when din-goes were removed from the declared ver-min list. The Aboriginal Land Rights(Northern Territory) Act 1976 requires thattraditional owners be consulted before anywild dog management programs are under-taken on their lands. Dingoes are unprotect-ed in Arnhem Land Aboriginal Reserve.

South Australia

Dingoes and hybrids are ‘proclaimed’ pestsunder the Animal and Plant Control Board(Agricultural Protection and OtherPurposes) Act 1986 in the sheep zone southof the Dog Fence (Figure 2). Dingoes mustbe controlled and can only be kept there inauthorised zoos and wildlife parks. Moniesfor the control of dingoes (Dingo ControlFund) is levied from all landholders withmore than ten square kilometres and ismatched by the government.

North of the dog fence the dingo is regardedas a legitimate wildlife species and althoughunprotected, is afforded a level of protectionby the South Australian Dingo Policy(Animal and Plant Control Commission1993). This policy was formulated in 1977with input from the Vertebrate Pests ControlAuthority (now Animal and Plant ControlCommission), pastoralists, the Dog FenceBoard, the Pastoral Board and the AustralianConservation Foundation. Beyond a 35 kilo-metre-wide baited buffer zone, conservationof dingoes is enhanced by restrictions toground baiting, prohibition of aerial baitingand phasing out of bounties.

The Dog Fence Board, under the Dog FenceAct 1946, administers maintenance of the2178 kilometre dog fence. Funding is sharedequally between the Government and land-holders receiving protection of the fence.

Queensland

Dingoes and other wild dogs are declaredpests under the Rural Lands Protection Act1985. The responsibility for wild dog controlin Queensland lies with landholders and isadministered by the Land Protection Branchof the Department of Natural Resources(QDNR) and local governments. Dingoesand dingo hybrids can only be kept withministerial approval and this is restricted tozoos and wildlife parks. The Queenslandbarrier fence (about 2500 kilometres long)(Figure 2) and funding for its maintenancehas averaged around 60% from State fundsand 40% from ‘precepts’ (levees) chargedwithin the ‘benefited area’ over the longterm. Local governments also maintain anumber of smaller ‘check’ fences. A team of


people employed by QDNR is responsiblefor the continued maintenance of the fence.

New South Wales

The Act to Consolidate the Acts for theProtection of Pastures and Live Stock fromthe Depredations of Noxious Animals 1898in New South Wales declared certain animalsincluding ‘native dogs’ noxious and obligedland owners to control them. The PasturesProtection Act 1939 reiterated this position asdid the Rural Lands Protection Act 1989 andthe Rural Lands Protection (Amendment)Acts 1994 and 1997. The wild dog policy ofthe National Parks and Wildlife Service(National Parks and Wildlife Act 1974)effectively protects dingoes within nationalparks and nature reserves, and the dingo isrecognised as a native species under theThreatened Species Conservation Act 1995.The Rural Lands Protection Act 1998 allowswild dogs to be declared as pest animals andrequires that the government be responsiblefor their management on government lands.Dingoes can be kept as pets under therestrictions of the Companion Animals Act1998.

In the Western Division of New South Wales,the WDDB administers the control of wilddogs and is responsible for the maintenanceof the 584 kilometres of the Dog Fence(Figure 2). The Board was established underthe Wild Dog Destruction Act 1921 and isfunded by rates on Western Division land-holdings and State Government subsidies.

Australian Capital Territory

The Nature Conservation Act 1980 in theAustralian Capital Territory defines protectedspecies in the Territory which includes din-goes. Control of wild dogs, including din-goes, on private lands is allowed subject to apermit authorising the killing of a protectedspecies being issued by Environment ACT.

Victoria

In Victoria, dingoes are offered some protec-tion within the lands administered under theNational Parks Act 1975 through the Wild DogPolicy of the National Parks and ConservationReserves Guidelines and Procedures Manual.

Elsewhere they were ‘declared vermin’ as wereall wild dogs under the Vermin and NoxiousWeeds Act 1958 and, since its repeal, under theCatchment and Land Protection Act 1994.Landholders have a legal obligation to controldeclared animals on land they own or occupy.

Tasmania

There have never been dingoes in Tasmaniaand the import of dingoes is banned alongwith a schedule of other exotic animals(National Parks and Wildlife Act 1970).Punitive action against feral and commensaldogs preying upon livestock is coveredunder the Dog Control Act 1987.

5.3 Current management strategies

5.3.1 Threats to livestock

It is now widely accepted that the threat tolivestock from wild dogs comes from withinthe stocked areas and immediately adjacent‘refuge’ areas. This has been supported byconsiderable scientific research (Section 2.4).With the increasing need to achieve the opti-mum cost-effectiveness of control work, themessage to managers about confining workto high-risk areas becomes even more attrac-tive. This coincides with a greater publicinterest in the preservation of dingoes(Chapter 3), and a greater public scrutiny oflethal control methods (Chapter 4).

One of the major changes in managementstrategies has been the abandonment of thebounty system in some States and Territories(Section 5.1.1). Many of the arguments sur-rounding bounties are well recognised,although it is worthwhile emphasising severalthat have undoubtedly hampered effectivewild dog control in the past, and may well stilldo so in some areas. Apart from the encour-agement of fraudulent practices, the paymentof bonuses encourages a ‘scalp count’ mental-ity. This can result in the targeting of areaswhere dogs are easy to catch, rather thanareas where dogs are posing the greatest risksto livestock. As well, bounties clearly encour-age the use of techniques that yield easily-found carcasses (trapping, shooting, and the


use of inappropriate poisons). There is strongevidence in Western Australia that the use ofhighly efficient 1080 baiting is not undertakenby some operators because carcasses are sel-dom found (Chapter 6). Success of controloperations should be measured by a reduc-tion or elimination of livestock losses, not bya scalp count.

‘Success of control operations should be measured by

a reduction or elimination of livestock losses, not by

a scalp count.’

The one exception to the general failure ofbounties in preventing livestock losses is thecase of affected landholders paying largebounties for the destruction of identifiabledogs responsible for extensive predation onsheep (that is a ‘rogue animal’ bounty or‘smart’ bounty, Hassall and Associates 1998).Because ‘rogue’ animals have often thwartedconsiderable attempts to kill them, they maytake a long time to catch, and although thesums offered may appear large ($100–$1000,E. Lackey, Rural Lands Protection Board(New South Wales), Inverell, pers. comm.1984), the effort required to catch the offend-ing dog(s) and the cost of their impact mayalso be large. This happens on a case-by-case basis and is different from the standardbounty system.

All of these factors have driven controlauthorities to adopt cost-effective, target-specific and humane control techniques andstrategies. The major features of current wilddog management in Australia are sum-marised below:

• A strategic approach to management, withan emphasis on identifying and concen-trating control in areas where stock are atrisk.

• A streamlined control effort to reflect thedegree of risk (the objective in sheepgrazing areas being to keep paddocksfree of wild dogs, with less stringentrequirements in the case of cattle enter-prises).

• Conducting control work in buffer areasimmediately adjacent to stocked pad-docks, to provide a sink for dispersingwild dogs to settle before they reach thepaddocks.

• The application of control on a largerscale and integrating the efforts of differ-ent groups.

• An increased use of aerial and groundbaiting with 1080 as cost-effective, strate-gic control methods.

• Less reliance on professional doggers andthe bounty system than previously(although bounties are still paid inQueensland, the Western Division of NewSouth Wales, and parts of WesternAustralia).

• Acts and policies to protect dingoes withinnational parks and similar fauna reservesin some States and Territories providing alegislative framework by which dingoescan be conserved.

5.3.2 Attacks on people

Wild dogs which come into close contactwith people may become aggressive (Section3.7). Authorities have attempted to managesuch problems through culling and/or legis-lation to control the feeding of wild dogs andthrough a public education program.Warnings about the dangers (to humans) offeeding wild dogs and seeking close contactare sometimes provided in signs andbrochures. People, especially children, wholive in areas where wild dogs (including din-goes) are present, need to be taught safe andappropriate behaviour towards dogs.

In addition to culling, methods to minimisewild dog–human interactions in tourist areasmay include aversive conditioning techniquesthat educate dingoes to avoid specific areas(Tauchmann 1998). For example, food scrapsspiked with lithium chloride, which if eatenwill cause dingoes to vomit, can be placedaround camp grounds. Similarly, deviceswhich emit high pitched sounds and impingeon the dingo’s sensitive hearing might be adeterrent; as are electric shocks administeredby modified cattle prodders (dingo prodders)or electrified fences (Bird 1994). Other deter-rents include weapons that fire kinetic energy


rounds (sting balls), bean bags, or sponge car-tridges (containing an irritant) (English andTaske 1998). The effectiveness of all theseapproaches for deterring dingoes is largelyunevaluated (Section 8.7).

5.3.3 Conservation of dingoes

(i) Public and governmenteducation

Education programs are required to helppeople recognise the dingo as a nativeAustralian species, understand its ecologicalrole and its plight, and push for policies toretain it as part of Australia’s national her-itage.

‘Governments can use improvedknowledge of dingo ecology to

instigate better control methods.’

To prevent hybridisation, people who wishto keep pure dingoes or hybrids as petsshould require permits and these animalsshould be neutered. Hybridisation wouldalso be reduced if domestic dogs kept in out-back mining towns, outback cattle stationsand Aboriginal settlements in wildernessareas such as Jabiru in Kakadu National Park,were neutered unless they are specificallykept for breeding. However, introducingsuch changes would require extensive con-sultation with dog owners in these places.

State and Territory governments can playtheir part here by recognising the overall roledingoes and other wild dogs play in wilder-ness and pastoral areas, and by legislatingaccordingly. The Northern Territoryremoved dingoes from the pest list in 1976but does not protect them, except in parks,reserves and in Arnhem Land. In SouthAustralia in 1977, dingoes north of the DogFence (60% of the State) were declassified aspests, but not protected. Survival of thedingo is ensured by restricting the availabilityof organised baiting campaigns to certainareas and to times when dingoes are presentin excessively high numbers and causinghardship for cattle producers. No bountiesare paid on dingo scalps and trapping andthe aerial laying of baits is prohibited outside

the Dog Fence. Dingoes remain proclaimedpests in the 40% of the State inside the dingofence. New South Wales protects dingoesonly in national parks and the conservationestate, as does the Australian CapitalTerritory. Dingoes remain ‘declared pests’subject to various levels of control through-out Queensland, Victoria and parts ofWestern Australia. Any new laws on thekeeping of dingoes by the general publicshould take into account the fact that suchgeneral ownership will increase the rate ofhybridisation (Corbett in press).

Governments can also use improved knowl-edge of dingo ecology to instigate better con-trol methods. For example, the NorthernTerritory Government was the first to stopannual broadscale aerial baiting and theWestern Australian Government drew on newknowledge about dingo movements to set upbuffer zones (nominally two dingo territorieswide) between pastoral and wilderness areas.

(ii) Dingo preservation societies

The Australian National Kennel Council(ANKC) is a co-ordinating body for State andTerritory canine controlling organisations. TheANKC recognises the dingo as an official dogbreed and has adopted it as Australia's nationalbreed with the proviso that exhibition, breed-ing or ownership of dingoes is not allowed inStates where these activities are prohibited. Insome States, preservation societies (such asthe Australian Native Dog ConservationSociety Ltd at Bargo, the Australian DingoConservation Association Incorporated atErindale in New South Wales and the DingoFarm at Chewton in Victoria) legally obtaindingoes to preserve and enjoy them. The phi-losophy and attitude of such societies isadmirable and their aims can be achieved ifthey take a united and scientifically validapproach. This is being done by the Bargo andErindale groups that are collaborating withresearchers at the University of New SouthWales (Barry Oakman, Australian DingoConservation Association Incorporated, NewSouth Wales, pers. comm. 1999). Otherwisedingoes may become inbred or the artificialenvironment and selective breeding may dis-courage the natural selection of wild charac-teristics. The best scientific knowledge must


be used to ascertain the dingo's general andspecific characteristics and this knowledgeshould be derived from samples collectedover most of the dingo's huge geographicrange in Australia and Asia.

With agreement from ANKC affiliated Stateand Territory canine councils or associationsand the dingo preservation societies, a nation-al register of dingoes and breeders could beconstituted. The following recommendationswould ensure that only pure dingoes wereregistered:

• Registered dingo breeders ensure thattheir stock comprises only pure dingoes.At present, this can only be done usingskull measurements from dead animalssupported by coat colours and breedingpatterns. In future, dingoes may beassessed for purity by DNA fingerprintingtechniques (Wilton et al. 1999) or possiblyby skull measurements from x-rays of livedingoes.

• Only verified pure dingoes be included onthe national register; that is, animalswhose parents are both from a pure dingobreeding line confirmed by the skull mea-surements or DNA analysis of the found-ing parents of that line. During the initialyears when stocks of proven pure dingoesare low, care must be taken to minimiseinbreeding and other genetic problems. Itshould be a registration requirement thatthe purity of every third generation ofeach breeding line is confirmed by refer-ence to skull measurements.

• The overseeing canine controlling bodies,dingo preservation societies and registeredbreeders be encouraged to educate thegeneral public about the plight of dingoesand the measures being taken to preservethem. This education should include theresponsibilities and problems of keepingdingoes in captivity and the reasons whymost people would not qualify to keep adingo. In addition, close collaboration withthe Royal Society for the Prevention ofCruelty to Animals (RSPCA) and other dog'shelters' should be established to ensurethat suspected dingoes and hybrids areculled rather than 'recycled' to members ofthe public.

• Most importantly, considerable effort bemade to win over members and support-ers of the pastoral industries, particularlythose in the sheep industry. Accordingly,the interested bodies should assist govern-ments to pass legislation that not onlyensures the preservation of pure dingoes,but safeguards the credibility of preserva-tion societies and combats the chicanerythat unscrupulous 'dingo breeders' mightemploy.

(iii) Island refugia

It is unlikely that most Australian mainlandhabitats will stay or become free of hybrids,so that large offshore islands and other refu-gia offer the best hope of preserving puredingoes in their natural habitat (Corbett1995a). There are many islands around theAustralian coastline, representing many cli-mates and habitats, excepting hot deserts.Some, such as Fraser Island, Melville Islandand Groote Eylandt, are large enough fordingoes to live and breed in natural condi-tions.

Hybrids would need to be eliminated. Dingopopulations would need to be managed so thatthey did not over-exploit their natural foodsupply and crash. Also, local regulations ban-ning the hand-feeding of dingoes must beenacted and strictly adhered to; otherwisemany dingoes will not only become depen-dent on food handouts but also become accus-tomed to the close presence of humans. Thisincreases the risk of dingoes annoying or bitingpeople. Management plans, such as the onedeveloped for Fraser Island (Queensland Parksand Wildlife Service 1999), are essential for themanagement of dingo populations on islandrefugia.

Islands do not necessarily need to be off-shore. They could be islands of well-protect-ed and maintained sanctuaries on the main-land, which would have the advantage ofadditional habitats, such as deserts, not avail-able on offshore islands. Such sanctuariesalready exist on the mainland for conserva-tion of endangered mammals and birds(Wamsley 1998).



Tab

le 4

: Aust

ralia

n leg

isla

tion a

nd p

olic

ies

for

din

goes

and o

ther

wild

dogs

. D

= D

ecla

red a

nim

al w

hic

h lan

d o

wner

s an

d o

ccupie

rs a

re o

blig

ed t

o c

ontrol;

U= U

ndec

lare

d a

nim

alw

hic

h lan

d o

wner

s an

d o

ccupie

rs h

ave

no o

blig

atio

n to c

ontrol

or pro

tect

; P= P

rote

cted

anim

al w

hic

h it is a

n o

ffen

ce to k

ill; N

= N

ot dec

lare

d n

oxi

ous

with

in lan

ds

man

aged

by

som

eSt

ate

agen

cies

; so

me

pro

tect

ion a

fford

ed to d

ingo

es e

ither

thro

ugh

polic

y or

in p

ract

ice.

QLD

Dw

hole

Sta

te

DD

Dep

artm

ent o

f

Nat

ural

Res

ourc

es;

loca

l gov

ernm

ent

NSW

Agr

icul

ture

;

Rura

l Lan

ds

Prot

ectio

n Bo

ards

Agr

icul

ture

Wes

tern

Aus

tralia

Park

s an

d

Wild

life

Com

mis

sion

Ani

mal

and

Pla

nt

Con

trol

Com

mis

sion

Dep

artm

ent o

f

Nat

ural

Res

ourc

es

and

Envi

ronm

ent

AC

TPa

rks

and

Con

serv

atio

n

Serv

ice

Dep

artm

ent o

f

Prim

ary

Indu

strie

s,

Wat

er a

nd

Envi

ronm

ent

Envi

ronm

ent

Aus

tralia

Obl

iged

to c

ontro

l

in p

asto

ral a

reas

Obl

iged

to

redu

ce/c

ontro

l

num

bers

whe

re

caus

ing

dam

age

No

oblig

atio

n to

cont

rol o

r pro

tect

Obl

iged

to c

ontro

l

in s

heep

pas

tora

l

area

s

Obl

iged

to c

ontro

l

in p

asto

ral a

reas

It is

an

offe

nce

to

kill

thes

e an

imal

s

Impo

rt pr

ohib

ited

Din

goes

pro

tect

ed

on C

omm

onw

ealth

land

Land

hold

ers

requ

ired

to c

ontro

l

in a

ll ar

eas

Rura

l Lan

ds

Prot

ectio

n A

ct

(198

5); N

atur

e

Con

serv

atio

n A

ct

1996

Wild

dog

Des

truct

ion

Act

1921

; Nat

iona

l

Park

s an

d W

ildlif

e

Act

197

4;

Thre

aten

ed

Spec

ies

Con

serv

atio

n A

ct

1995

; Rur

al L

ands

Prot

ectio

n

(Am

endm

ent)

Act

1998

;

Com

pani

on

Ani

mal

s Act

199

8

Agr

icul

ture

and

Rela

ted

Reso

urce

s

Prot

ectio

n A

ct

1976

; Wild

life

Con

serv

atio

n A

ct

1950

Terr

itory

Par

ks

and

Wild

life

Con

serv

atio

n A

ct

1993

Ani

mal

and

Pla

nt

Con

trol

(Agr

icul

tura

l

Prot

ectio

n an

d

Oth

er P

urpo

ses)

Act

198

6

Cat

chm

ent a

nd

Land

Pro

tect

ion

Act

199

4;

Nat

iona

l Par

ks A

ct

1975

Nat

ure

Con

serv

atio

n A

ct

1980

; Dog

Con

trol A

ct 1

975

Nat

iona

l Par

ks a

nd

Wild

life

Act

197

0;

Dog

Con

trol A

ct

1987

Wild

life

Prot

ectio

n

(Reg

ulat

ion

of

Expo

rts a

nd Im

ports

)

(Am

endm

ent)

Act

1995

Statu

s of

din

goes

Stat

us o

f wild

dog

s

Age

ncie

s

resp

onsi

ble

for

mana

gem

ent

Land

owne

r or

occu

pie

r

resp

onsi

bilit

y

Rel

evant

legis

lation

NSW

D p

asto

ral a

reas

N N

PWS

land

WA

NT

SAV

ICA

CT

TAS

Com

monw

ealth

D p

asto

ral a

reas

N N

atio

nal P

arks

U1

D in

side

barri

erfe

nce

N o

ther

are

as

D p

asto

ral a

reas

N N

atio

nal P

arks

PN

ot p

rese

nt

Impo

rt pr

ohib

ited

Expo

rt no

t per

mitte

d

P N

atio

nal P

arks

DU

1D

DD

ange

rous

dog

s

may

be

destr

oyed

Subj

ect t

o co

ntro

lC

ontro

lled

whe

re

impa

ct d

emon

strat

ed

contin

ued

ove

r pag

e


Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

No

No

Yes

Yes

No

No

Yes

No

No

Yes

No

No

No

No

No

No

Yes

Bonuse

s paid

2

Gov

ernm

ent

financ

e3

Coord

inate

d

contr

ol4

Wes

tern

Div

isio

n

only

Yes

(Pilb

ara

and

Gas

coyn

e/

Mur

chis

on)

1 Din

goes

are

und

ecla

red

in a

ll ar

eas

of th

e N

orth

ern

Terr

itory

and

are

unp

rote

cted

out

side

par

ks, r

eser

ves,

san

ctua

ries,

wild

erne

ss z

ones

and

the

Arn

hem

Lan

d A

borig

inal

Res

erve

.2 B

onus

es p

aid

on p

rese

ntat

ion

of p

roof

of d

estru

ctio

n of

a w

ild d

og; u

sual

ly a

sca

lp. B

onus

es a

re n

ot p

aid

unde

r Con

serv

atio

n or

Wild

life

Act

s.3 G

over

nmen

t ass

ista

nce

to c

ontro

l wild

dog

s pr

ovid

ed th

roug

h ba

rrie

r fen

ce m

aint

enan

ce, t

he e

mpl

oym

ent o

f dog

gers

and

sub

sidi

satio

n of

bai

ting

costs

. Gov

ernm

ent f

inan

ce is

not

ava

ilabl

e un

der

Con

serv

atio

n or

Wild

life

Act

s.4 C

oord

inat

ed c

ontro

l gro

ups

are

enco

urag

ed o

r req

uire

d.

Tab

le 4

contin

ued

QLD

NSW

WA

NT

SAV

ICA

CT

TAS

Com

monw

ealth

Summary

To formulate management plans, somemeasure of trends in dog population densi-ties is required, although an assessment oftotal numbers is not necessary. Trends innumbers indicate potential predationthreats to livestock and allow advanceaction to be taken. Threats vary accordingto the type of livestock at risk. What may be a‘safe’ rise in the abundance of wild dogs ona cattle station might seriously threaten theviability of a sheep station.

To assess the effectiveness of programs tomanage wild dogs, it is essential to monitorthe impacts of wild dogs on livestock produc-tion. Although knowledge of trends in wilddog numbers may assist in determiningwhether a new management approach iseffective, the most important measure of suc-cess is a reduction in livestock losses due towild dogs.

The principal techniques to control wilddogs are exclusion fencing, shooting, trap-ping and poisoning. Exclusion fences rangefrom the famous barrier ‘dog fences’ ofAustralia, which aim to keep wild dogs frommoving into huge sections of various States,to shorter fences maintained to protect indi-vidual properties. Traditionally, fences havebeen constructed from wire netting, butelectric fences are now used in some areas.Poisoning with 1080 is the most cost-effec-tive lethal means of reducing populations ofwild dogs over large areas of remote or inac-cessible country. Various types of bait areused, with methods of placement rangingfrom burying individual baits to droppingbaits from aircraft. Trapping is still used forwild dog control and will probably alwaysbe needed to target particular dogs that can-not be removed by other means.

At present, there are no suitable agents forthe biological control of wild dogs. Althoughin many ways a theoretically attractive con-trol option, biological control would havepotential drawbacks, including the risk to

domestic dogs, and the threat to the conser-vation goal of maintaining populations ofdingoes in reserve areas.

6.1 Introduction

Programs to manage wild dogs need to bestrategically based, focusing the effort inareas where the highest risks to livestockoccur. Not only is this economically sound, italso takes account of the conservation statusof dingoes in areas where they pose littlehazard to livestock. Vertebrate pest controlprograms are coming under increasingscrutiny by the public and the need forimplementing each program must be soundlyjustified. There is also a need to justify thecontrol methods used, not only in economicterms, but also on the grounds of animalwelfare (Section 4.2) and hazards to non-tar-get species.

‘Managers should avoid the pitfall of continuing with actionthat is based on tradition ratherthan an objective assessment ofthe best options for addressingtheir current wild dog problem.’

This chapter outlines the various techniquesthat managers can use to effectively plan,implement and evaluate the effectiveness ofmanagement programs. The range of controltechniques that are available are described,and suggestions are given on the circum-stances that warrant their use. Strategies willvary from region to region across Australia,but the fundamental aim of most manage-ment programs, to protect livestock fromwild dog attacks, will remain the same.Managers need to be aware of all options.They should avoid the pitfall of continuingwith a course of action that is based on tradi-tion rather than an objective assessment andevaluation of the best options for addressingtheir current wild dog problem.


6. Techniques to measure and manage impact and abundance

6.2 Estimating abundance

6.2.1 Background

Wild dogs are elusive, mostly active at dawn,dusk and during the night, have seasonallyvariable patterns of activity and usuallyoccur at relatively low densities (Chapter 2).These factors all contribute to the difficultyof obtaining accurate measures of wild dogabundance. However, measures of absolutedensity are usually not necessary and mostmanagers only need a measure of changes inabundance that may result from manage-ment programs or environmental conditions(for example, changes in the availability offood). Such changes may indicate, forinstance, that an increasing threat is emerg-ing, warranting a reallocation of resources orindicate the success or otherwise of a controlprogram.

Scalp returns are not a good measure of theabundance of wild dogs because they arestrongly influenced by the level and type ofcontrol effort. For example, scalps are rarelyrecovered during baiting programs (Section6.4.4). Scalp returns may also reflect theexpertise of a particular operator, rather thanreflecting the number of wild dogs in anarea. Scalp returns would only be a guide tonumbers if the same control methods wereused throughout, and the effort remainedconstant. These criteria are seldom met, andin any event, capture success can declineover time as surviving animals becomeincreasingly wary.

In many instances, estimates of the numberof wild dogs in an area, based on simplecounts of footprints, or sightings of wilddogs, provide sufficient information onwhich to base management decisions. Insheep paddocks, for example, the presenceof any wild dogs warrants some controleffort because sightings of wild dogs arehighly correlated with predation (Flemingand Korn 1989) and because once predationcommences it will continue until the preda-tor or the sheep are removed (Thomson1984a). In cattle grazing areas, assessmentsof potential risk may be related to the size ofgroups or packs of wild dogs seen in thearea. In this case, the regular sighting of large

packs throughout the area could indicate anincreased risk of predation on calves, partic-ularly if coinciding with a decline in theavailability of the wild dogs’ usual prey.

On a broader scale, managers may require ameasure of overall trends in wild dog densi-ties, particularly in the ‘refuge’ areas andbuffer areas adjacent to livestock paddocks.In these situations, the option of carrying outpreventative control is more efficient thanundertaking reactive control after damagehas already started. The most practicalmeans of measuring changes in populationdensity is to derive indices of abundance.The level of precision of the indices is mostimportant. If the inherent variability of theindices is too large, then real changes mayremain undetected or changes may be indi-cated when none really occurred.

‘In many instances, estimates based on simple

counts of footprints, or sightingsof wild dogs, provide sufficientinformation on which to base

management decisions.’

In most circumstances, indices of abundanceare adequate to assess the effectiveness of acontrol program. Indices of abundance arealso useful for making comparisons betweensites of the effectiveness of a particular strate-gy for reducing abundance. If conservationof dingoes is the aim, the long-term numbersof dingoes must be assessed. Regularmark–recapture assessments of areas withpure dingoes, coupled with DNA samplingwould be required to assess the success ofsuch a conservation program.

Detailed discussions on techniques and anal-yses for measuring population abundanceare given in Otis et al. (1978), Caughley(1980), and Caughley and Sinclair (1994).Some of the most useful techniques forderiving counts or indices of wild dog num-bers for management purposes are describedin the following sections.


6.2.2 Sign counts

Sign counts are often the most suitable meth-ods for assessing relative abundance. Faecaldeposits detected by trained ‘sniffer’ dogs canbe used to index abundance (B. Morris, dog-ger, Yass Rural Lands Protection Board (NewSouth Wales), pers. comm. 1998), as can foot-prints. Sign counts can be ‘passive’ (that is,the animals are detected in their normalmovements without detection altering theirbehaviour, for example, faecal deposits perunit time), or ‘active’ (that is, the animals areencouraged by lures to leave their sign at sta-tions, for example, scent stations and bait sta-tions, see below). Sign counts can be madealong roads (Allen et al. 1996; Catling et al.1997; Edwards et al. 2000), animal tracks andcreek beds (Fleming et al. 1996) or at ran-domly-placed or stratified stations (Mahon etal. 1998). The main problem with ‘passive’indices is that it is often difficult to generatesample sizes large enough to adequatelyquantify low-density populations of wilddogs. On the other hand, ‘active’ indices mayinterfere with normal behaviour patterns and

this can influence counts. In both cases, therelationship between actual abundance andthe index is assumed rather than known.Further work to determine the relationshipbetween indices and abundance is required.A study is currently being undertaken in theEastern Highlands of Victoria which will shedlight on the relationship for that environment(F. Gigliotti, Department of Natural Resourcesand Environment, Victoria, pers. comm.1999).

Many studies of carnivores in North Americahave relied on scent stations to obtainindices of abundance. These ‘stations’ con-sist of an area of cleared ground on whichtracks can be readily seen, with a scent lureused to attract animals to the site. The princi-ple underlying this method is that the visita-tion rates are proportional to animal density.However, visitation of scent stations hasbeen shown to be independent of density forsome species (Smith et al. 1994). Bait stationsare essentially scent stations modified byplacing a bait in the centre. For canids, it isrecommended that the baits be covered with


Cleared transects allow track counts to be used as an index of wild dog acitivity (Source: Queensland Rural LandsProtection Board).

soil to avoid removal by birds (Allen et al.1989; Thompson and Fleming 1994).

‘Further work to determine therelationship between indices and

abundance is required.’

Frequency of visitation to both scent stationsand bait stations would be affected by fac-tors including spacing of individual stations,presentation of baits and attractants, habitatdifferences between sites, frequency ofoperations, length of index period and quali-ty of tracking surfaces. These factors shouldbe standardised to facilitate within andbetween site comparisons. Bait stations canalso be used to assess cumulative baituptake. This can then be used as an abun-dance estimator. Thompson and Fleming(1994) were able to demonstrate that cumu-lative bait uptake data produced reliableindices of fox (Vulpes vulpes) abundance,with low variability.

6.2.3 More detailed counts or estimates for research

In most circumstances, indices of abundanceare adequate to assess the effectiveness of acontrol program as changes in abundanceare reflected by changes in the indices.However, for more detailed estimates ofnumbers for specific research studies, othermethods are also used. These includemark–recapture studies, where animals arelive-trapped, marked (tagged and collaredwith plain or radio-collars), then released.Estimates are made of the overall populationdensity when marked animals are eitherrecaptured or re-sighted during specific trap-ping or monitoring phases. Estimates areaffected by the altered behaviour of markedanimals. Increased wariness after initial cap-ture, for example, can produce overesti-mates of the overall population (Caughley1980).

A novel assessment of population assess-ment that may have potential for monitoringwild dogs has been developed for grizzlybears (Ursus arctos) in Canada (Mowat andStrobeck 2000). DNA profiles are establishedfrom the hairs of bears that visit a bait site

and mark–recapture analysis is used todetermine population size. Likewise inCalifornia, DNA residual in faeces has beenused to identify individual mountain lions(Puma concolor) (Ernest et al. in press).DNA recovered from mountain lion salivaleft in wounds on bighorn sheep (Oviscanadensis) can then be matched to individ-ual lions identified from faecal DNA (H.Ernest, University of California, Davis, pers.comm. 2000). Because mountain lions aremostly loners, control can then be targeted atthe individuals responsible for killingbighorn sheep. In future, problem individualwild dogs might be better targeted throughDNA analysis of faeces, saliva and tissuesfrom captured wild dogs. That is, salivaryDNA from wounds on sheep identifies theindividual wild dog; faecal DNA identifieswhere the wild dog lives and tissue DNAconfirms that the wild dog killed by con-trollers was indeed responsible for killing thesheep.

Cyanide baiting is a technique that can beused by researchers to generate ‘catch-per-unit-effort’ (CPUE) indices. Algar and Kinnear(1992) demonstrated the usefulness of cyanidebaiting to generate CPUE indices of fox abun-dance. The technique has not yet been fullyassessed for use with wild dogs. Using theknown number of removals obtained by sucha program, cyanide baiting also has the poten-tial for estimation of absolute abundance. Thisis achieved with index–removal–index calcu-lations (Caughley 1980) using the knownnumber removed in conjunction with beforeand after manipulation indices.

Probably the most detailed information onthe abundance of a wild dog populationswas obtained during intensive research stud-ies in the Pilbara region of Western Australia(Thomson et al. 1992b). There, dingoes wereradio-collared and observed from aircraft.Because the dingoes were living in packs,information was obtained on the entire packeven when only one pack member was col-lared. Uncollared packs or individuals wereidentified by comparing aerial observationsagainst data on dingo tracks counted duringroutine ground surveys. In this way, a reli-able tally of total population size was possi-ble over a number of years. Such data aredifficult and expensive to obtain.


6.3 Estimating agricultural andenvironmental impacts

6.3.1 Agricultural impacts

Producers need to assess the type and extentof livestock losses from different causes as afirst step in deciding how to deal with them.Measuring the impact of wild dogs on live-stock production has always been difficult.Part of the problem is that, in rangelandareas in particular, there is little intensivehusbandry of livestock. Musters are infre-quent, stock are regularly seen only atwaters, and the remains of animals killed bydogs are generally undetected. Often the firstindication of dog attacks is finding injuredanimals at waters or during musters. In addi-tion, livestock losses from other natural caus-es, including reproductive failure, can rarelybe quantified, making it difficult to deducelosses due to dogs based purely on mustertallies. Further, information is usuallyobtained from areas where some form ofwild dog control is being carried out, makingit difficult to deduce what the impact mayhave been in the absence of any controlwork (Section 3.1).

In earlier scientific studies, there was a ten-dency to make deductions from dietary anal-ysis about the likely impact that wild dogshad on livestock (Whitehouse 1977).However, such interpretations are generallynot reliable, because wild dogs do notalways eat the animals they kill. Neither isthe relative abundance of predators and preyknown so that a small occurrence of live-stock in the faeces may represent a substan-tial livestock loss if there are many wild dogspresent. In addition, wild dogs cause lossessuch as wounding and mismothering whichare not reflected in stomach or scat samples(Section 3.1; Thomson 1984a; 1992c).

Various surveys have been undertaken in anattempt to quantify predation losses, generallyover broad regions (Section 3.1; Fleming andKorn 1989). These have the limitations of anysurvey (for example, variable response rate,inaccurate estimations), although they cangive a largely objective indication of whereproblems occur and trends over time. It is rec-ommended that these data be collected as part

of property inspection reports by staff fromthe relevant land management authority. Mapsof such data provide managers with a power-ful tool with which to assess resource alloca-tion to reduce predation impacts. In NewSouth Wales, it is a statutory requirement thatrecords of stock losses be provided by a land-holder when seeking 1080 (sodium fluoroac-etate) baits for emergency baiting. Suchreports can be included in regional maps.

‘Where there are areas with similar environments and

production systems where wilddogs are controlled and

uncontrolled the effectiveness of control can be assessed.’

To obtain scientific estimates of reductions inpredation losses under a particular controlstrategy, comparisons must be madebetween at least two similar areas over anumber of years. One area must have nocontrol and the other uses the control strate-gy (Hone 1994). This accounts for seasonalvariation in wild dog activity and avoids con-trol strategies being erroneously deemedsuccessful or unsuccessful when the cause ofchanges in predation is seasonal. Obviously,it is unlikely that sheep producers wouldallow wild dogs to go uncontrolled and sothere will always be an element of doubt inassessing the success of a program. In theextensive cattle areas of northernQueensland, there are areas with similar cli-mates, environments and production sys-tems where wild dogs are controlled anduncontrolled. In these cases, the effective-ness of control can be assessed (Allen andGonzalez 1998).

Recognising predation

On an individual property scale, landholdersneed to be aware of how to determine lossesdue to predation. A number of factors tolook for are summarised in Box 1. This infor-mation is derived from direct observations ofradio-collared dingoes in sheep paddocks inWestern Australia (Thomson 1984a), thoughmuch of it is also applicable to predation oncattle and other livestock.



Skinning of carcasses reveals the extent of wild dog attack and, particularly for woolly sheep, may be necessary to confirmwhether a wild dog fed on the carcass post-mortem or was the cause of death (Source: (a) P. Thomson, (b) L. Corbett).


often found caked on the hind legs. Thepattern made by blood flowing down thelegs while the sheep was still upright canbe clearly distinguished from blood or flu-ids which may flow as a result of feeding,decomposition, or animals feeding on acarcass (Figure 11b). Saliva, even whendry, can sometimes be seen on the wool ofattacked sheep.

Skinning the throat and hind legs is oftensufficient to reveal hidden damage. Toothpunctures in the hide, subcutaneousbleeding, bruising and tissue damage willindicate wild dog attack (Figure 11c).External signs of tissue damage due tofeeding or scavenging activities can bereadily confirmed in a fresh carcass bysuch simple dissection. Tissue damagecaused after death will show no bruising orbleeding as this only occurs if the animal isbitten while still alive.

Decomposition and/or scavenger damagecan mask the cause of death in older car-casses. Often, however, blood-stainedwool still remains (especially on the lowerlegs which often dry out intact). Again,care must be taken to ensure that decom-position fluids are not mistaken for blood.Depending on the carcass age, dissectionis often warranted, as bruised tissue canstill be distinguished from undamaged tis-sue in a decomposing carcass. Providedthat the skin is still intact, simply pullingout the wool can often reveal damage.Puncture wounds in the throat will beuncovered in this manner (Figure 11d).

Carcasses found (sheep aged lessthan six months)

There are no predators in Australia apartfrom wild dogs that are large enough toinflict the sort of damage to adult sheepdescribed above. In the case of sheepyounger than about six months of age,however, the situation is often more com-plex. Many young lambs die from causesother than predation, and predators areoften suspected as a result of scavenging

Carcasses found (sheep agedmore than six months)

Signs — Depending on the type of groundand the amount of time elapsed since theattack, tracks may be found indicating astruggle. Both the wild dog and its preyoften leave deep prints with toes spreadout. Freshly broken vegetation, often hold-ing tufts of wool, is sometimes found at thesite of the attack. Pieces of wool withpatches of torn skin attached, as well asblood trails, are good indicators of preda-tion by wild dogs. Often, however, scav-engers obscure the tracks and other signs.The presence of dog tracks at a carcassdoes not necessarily mean predation wasthe cause of death, particularly if the tracksare more recent than the carcass.

Position of carcass — In the hotter, aridregions, sheep dying from natural causesusually die in shady places, sometimesnear water. A carcass found out in theopen, away from available shade can indi-cate predation. In the same way, the ageand condition of the sheep can give a clueto predation — if young and apparentlyhealthy sheep are found dead, this oftenindicates predator attacks.

Carcass examination — Although wilddogs usually attack from the rear as theypursue their prey, they generally kill bybites to the throat, damaging the tracheaand the major blood vessels of the neck.Blood and puncture marks on the throatare therefore good evidence of wild dogpredation. Blood is often found at themouth and nose, although care should betaken to distinguish this from other bodyfluids that drain from a decomposing car-cass (Figure 11a).

Attacks from behind result in injury to thesheep’s hind legs. Inexperienced wilddogs, or those attacking without anyapparent motivation to feed, frequentlyinflict considerable damage to the hind endof the sheep. This often leads to its deathafter the event. In these cases, blood is

Box 1: Recognising wild dog predation of sheep


Sheep showing signs of injury should beexamined for bite marks and blood, partic-ularly on the hind parts. As noted above,wild dogs often bite the hindquarters,causing easily identified injuries. Rams usetheir horns to deter wild dogs from frontalattacks and tend to suffer more rear attacksthan other sheep. A torn scrotum or some-times even complete castration may befound (Figure 11e).

Damage to the head or neck area of sheepusually results in a quick death, so almostinvariably animals found with theseinjuries are already dead. Any doubtsabout the cause of substantial injuries canbe easily solved by post-mortem examina-tion.

Occasionally sheep bearing scars or freshbite marks from wild dog attacks are discov-ered, particularly at shearing (Figure 11f).Care should be taken to distinguish bitemarks from shearing or wire scars. Fewsheep recover from severe wild dog-inflictedinjuries; blood loss, shock, infection andinhibited movement are probably the mostimportant factors in later death (Figure 11g).

Injured lambs are rarely found in range-land areas, as they seldom survive a wilddog attack. In areas of prime lamb produc-tion in eastern Australia, a number oflambs may be injured or killed by wilddogs while the accompanying ewes are leftunscathed.

on already dead or moribund lambs. Toverify predation, dissection must showhaemorrhage and bruising as described foradult sheep (other methods can be used todetermine whether the lamb was healthyand likely to have survived in the absenceof predation; Saunders et al. 1995).

A further complicating factor in relation tolambs is that predators other than wilddogs can be involved. It can be difficult todistinguish between fox and wild dog pre-dation on lambs. The size of bites andpuncture marks probably provide the mostreliable guide, dogs having much largerand more widely spaced canine teeth.Although foxes often tend to feed from theheads of lambs, there are variations in thebehaviour of both individual foxes andwild dogs that sometimes make definitepredator identification difficult.

Injured sheep found

Injured sheep are sometimes found inareas where wild dogs are active. Sheepthat have been bitten often drift towardsthe rear of a mob, and can frequently beseen hobbling behind. They sometimesremain close to water, although if theinjuries are severe they may be unable tomove at all.


Figure 11: Dingo predation on sheep showing: (a) ewe killed by dingoes, note blood on ground but little external damageapparent; (b) sheep attacked at rear end, note pattern of blood stain which took place when sheep was standing; (c) skinrevealing tooth punctures, bleeding and tissue damage, clearly distinguishable from undamaged portions of the carcass; (d)tooth punctures and bruising revealed in a decomposing carcass by pulling out neck wool; (e) ram with damaged scrotumfollowing dingo attack; (f) scrotum of ram showing scars from a previous dingo attack; (g) sheep severely injured by dingoattack that would not have survived (Source: P. Thomson).

a

d

c

b

g

f

e


(a) Traditional form of dog fence which is prone to breaching and expensive to construct and maintain (Source: P. Fleming);(b) Electric fence replacing old netting fence behind (Source: P. Bird, Animal and Plant Control Commission, SouthAustralia); (c) Solar panel providing charge for electric fence (Source: P. Bird, Animal and Plant Control Commission, SouthAustralia); (d) An alternative to a full electric fence is the placement of electrified outriggers (Source: Queensland RuralLands Protection Board).

a b

c d

Indirect losses

Estimating losses caused by general harass-ment and mismothering is difficult.Mismothering is the most significant cause ofneonatal lamb loss in merino sheep but it isdifficult to distinguish from that caused bywild dog harassment. Producers sometimesbecome aware of the presence of wild dogsin sheep paddocks by observing unusualbehaviour of sheep, such as sheep avoidingusually favoured grazing areas, or exhibitingpanicky behaviour. Wild dogs sometimeschase sheep through fences into waterlessareas, causing further indirect losses. Suchinstances can sometimes be positively identi-fied by tracking back to where the chasetook place.

‘Long-term branding figures with coincident data on wild dogcontrol can reveal a predation

effect not apparent from carcass counts.’

Evidence of wild dog predation of cattle maybe carcasses (rarely), lower-than-expectednumbers of calves at branding, or numbersof injured or scarred calves or weaners atmuster. The latter may be indicative of indi-rect production losses that are not evidentfrom livestock audits. For example, in thecoastal hinterland of eastern Australia, calvesinjured by wild dogs seldom recover frombites that invariably become infected and flyblown (P. Fleming pers. obs., 1989–91). Ifcalves recover, they usually remainunsaleable even after a considerable conva-lescence. Long-term branding figures withcoincident data on wild dog control cansometimes reveal a pattern suggesting a pre-dation effect not apparent from carcasscounts (Allen and Gonzalez, 1998).

6.3.2 Environmental impacts

The exact nature of past and present impactsof wild dogs on native fauna is unclear(Sections 3.2 and 3.6). In areas where there isa conservation goal to maintain dingoes, itmust be accepted that there will be predationon native species. Depending on the size ofthe area, and the degree of disturbance that

the environment has been subjected to, din-goes and prey would be expected to remainin an ecological balance. Only in situationswhere a remnant population of a particularlyvulnerable prey species was at risk of extinc-tion might a deliberate reduction of dingonumbers be considered (Section 3.6).Predation impact would be estimated byobserving and monitoring populations of theprey species. In these situations, the sec-ondary effects of any predator control cam-paign must be considered. There is some evi-dence, for example, that the removal of din-goes has resulted in an increase in the num-ber of foxes in some areas (Sections 2.10.1and 3.6). If this occurs, these smaller preda-tors may pose a greater threat than dingoes tosome native mammals.

6.4 Control techniques

A variety of control techniques are availableto manage the impact of wild dogs on live-stock production. These include the famous‘dog fences’ of eastern Australia which aimto physically exclude dogs from the livestockgrazing areas, the laying of poison baits, aswell as specific techniques such as trappingand shooting to remove individual animals.

In many cases, a combination of methodsmay be used. Choice of control techniquedepends on a number of variables, such asthe type and value of livestock being protect-ed, the presence of, and the potential risksto, non-target animals posed by each controltechnique, the size of the area being protect-ed, the accessibility of the terrain and thehumaneness of the technique. The manage-ment strategies developed for a particularregion (Chapters 3 and 5) will take these fac-tors into account, along with an assessmentof the cost-effectiveness of the variousoptions.

6.4.1 Exclusion fencing

Exclusion fencing remains a popular methodto prevent the incursions of wild dogs intosheep-grazing lands. Fencing ranges fromthe famous ‘barrier’ or ‘dog fences’ (Section5.1), which aim to keep wild dogs from mov-ing into huge sections of various States, to



Figure 13: New wire-netting fences like this one are rarely constructed because of the cost. The fence is 1.8 metres high withthe bottom half constructed with rabbit-proof netting and the top with lighter, marsupial-proof netting. A 25 centimetre flapof rabbit-proof netting is either buried or laid flat on the ground to prevent wild dogs, rabbits and macropods from diggingunder the fence (Source: P. Fleming).

Figure 12: A wire-netting wild dog-proof fence in north-eastern New South Wales. The fence has a single electrified outrigger wire and logs placed at the bottom to prevent damage by macropods (Source: P. Fleming).

shorter fences maintained to protect individ-ual or groups of properties. Traditionally,exclusion fences have been constructedfrom wire netting, but electric fences arenow used in some areas. Netting barrierfences are usually about 1.8 metres high,constructed either entirely from marsupialnetting or with the bottom half rabbit-proofnetting (Figure 12). To prevent wild dogsgetting under the fence, approximately 30centimetres of netting is turned out (andsometimes buried) at the bottom on the wilddog side of the fence. These fences areexpensive to construct ($8500 per kilometre;Bird et al. 1997). Marsupial netting is nolonger manufactured and new fence designsusing 1.8 metre high prefabricated deer fenc-ing have been developed (Figure 13) at acost of $5000 per kilometre for materials only(B. Harden, unpublished data 1999).

‘The electrified outrigger proved to be a relativelyinexpensive method of increasing

the security of older fences.’

The security of 35 kilometres of netting barrierfencing under landholder management innorth-eastern New South Wales was exam-ined monthly for two years (B. Harden,unpublished data 1991–93). Fence mainte-nance was inadequate on four of the six prop-erties (there was a hole a wild dog could useon average every 60 metres) and on someproperties maintenance tended to be left untilstock were killed. On average, a new holewas made in each kilometre of fence every 18days. Approximately one-third of holes weremade below ground level and another thirdjust above ground level. Almost all (93%)were caused by animals, predominantlymacropods, echidnas (Tachyglossus aculea-tus), wild dogs and foxes. Wild dogs regularlymoved through the fence on four properties,although not all incursions resulted in stocklosses. Sheep were killed on three properties,including one with a well-maintained fence.The wild dog responsible for losses on the lat-ter property crossed the fence on an adjoiningproperty then moved between propertiesinside the fence, illustrating the need forcooperation between neighbours in fencemaintenance.

Recently, effective all-electric barrier fenceshave been developed in South Australia as acheaper alternative to traditional nettingfences (Bird et al. 1997). A number ofdesigns were tested and the researchersfound that simple, upright seven or eight-wire, 975 millimetre-high electrified fencesexcluded most dingoes. The most effectivedesign was a seven-wire upright fence, with900 millimetre-wide wire netting placed to aheight of 400 millimetres, with the remaining500 millimetres turned out on the ground toform a skirt on the dingo side of the fence.Two of the four wires above the netting wereelectrified. This fence cost $3700 per kilome-tre and excluded dingoes for the 69 monthsof the trial. A cheaper alternative ($2700 perkilometre) that prevented dingo incursionsfor 46 months was a 10-wire electric fencewith the bottom 4 wires sloping to the dingoside of the fence at an angle of 45o to the ver-tical. Bird et al. (1997) concluded that themaintenance cost of this fence was likely tobe less than that for the fence with the net-ting, particularly in areas where corrosionrates are high.

The increased security provided by addingan electrified outrigger near ground level onthe wild dog side of existing old netting bar-rier fencing has also been evaluated (B.Harden, unpublished data 1992–94). Anelectrified wire was added to alternate onekilometre lengths of six kilometres of barrier


Snare trap being dug in beside a fence along whichwild dog movement is likely (Source: Queensland RuralLands Protection Board).

fence in north-eastern New South Wales, andthe fence was inspected and repairedmonthly for two years. The electrified wirewas run on Insultimber™ posts 300 millime-tres from the fence and 210 millimetresabove the ground. There was an earth returnwire 50 millimetres above the ground andthe barrier fence netting was also connectedto the energiser earth. This wire spacing waswide enough to prevent trapping (andkilling) echidnas under the electrified wirewhile preventing dingoes from reaching thebottom of the barrier fence without contact-ing the live wire. The rate of hole formationwas lower in almost all months in the sec-tions of electrified fence (average one holeper ten days per kilometre) than the non-electrified sections (one hole per six daysper kilometre). Macropods, echidnas andfoxes caused the majority of holes. In thetwo years of study, no dingoes or foxesbreached the electrified sections from theoutside, whereas four holes were made bydingoes, and five by foxes, in the non-elec-trified sections.

The total cost of the electrified outrigger was$900 per kilometre. This proved to be a rela-tively inexpensive method of increasing thesecurity of older fences. Although the elec-tric fencing components required little main-tenance, the annual cost of maintaining thebarrier fence itself was high. Most of this wasthe labour cost of inspecting the fence, sothe annual cost was largely a function of thenumber of inspections. The annual cost ofweekly inspections was $230 per kilometreof fence while the annual cost of monthlyinspections was only $80 per kilometre. The‘best’ inspection interval is a trade-offbetween security and cost.

One of the most important but often neglect-ed aspects of electric fencing is proper earth-ing. Australian soils are usually very dry andare poor conductors, so electric fences willnot work effectively unless they are properlyearthed. For the control of wild dogs, it isrecommended that one or more earth returnwires be incorporated in the design to over-come poor soil conductivity; dogs contactingboth the live and earth return wires simulta-neously then receive the full shock(Gallagher 1991).

Well-designed netting and electric barrierfences are expensive to erect and to maintainbut offer relatively continual protection fromwild dog predation if well maintained.Adequate maintenance is the crucial issue;without this, the fence may do little morethan create a false sense of security.

6.4.2 Shooting

Wild dogs are seldom seen during the dayand are especially wary where they are sub-ject to persecution by humans. Shooting istherefore generally regarded as an oppor-tunistic method of wild dog control (Hardenand Robertshaw 1987). Exceptions to this aresituations where hunters ‘howl up’ and thenshoot wild dogs or where a particular dogestablishes a regular pattern of visiting a par-ticular paddock or site and can be specifical-ly targeted. Organised wild dog ‘drives’,where wild dogs are flushed into a line ofshooters, were often unsuccessful and arenow rarely undertaken. However, excep-tions may occur. For example, a respondentto a survey by Fleming in 1985 (P. Fleming,unpublished data 1985) told of 40 men, inresponse to sheep losses over a period ofmonths and unsuccessful trapping and poi-soning campaigns, undertaking a dog drivein which only one wild dog was shot. No fur-ther sheep were killed for several months,indicating that the control was successfuland the benefits made the effort worthwhile.Generally, shooting is not an appropriatetechnique to reduce populations of wilddogs over extensive areas.

‘Generally, shooting is not anappropriate technique to reduce

populations of wild dogs overextensive areas.’

A .22 calibre rifle is suitable for the humanedestruction of wild dogs caught in traps. Forfree-ranging wild dogs, a high-powered cen-tre-fire rifle should be used (minimum .222calibre). The head or chest should be targeted,so that death is rapid. Less powerful weaponssuch as the rimfire .22 calibre should not beused for free-ranging wild dogs because of thegreater risk of non-lethal wounding.


In the Fortescue River region of WesternAustralia, the ‘Judas’ technique was used toremove specific dingoes. Radio-collared din-goes were regularly followed up by trackingand observation from an aircraft (Thomson1992a). These dingoes were generally loneindividuals that often formed associations withother lone, uncollared dingoes in the area.When uncollared dingoes were seen with theJudas individuals, the animals were followedto a resting site, and a marksman on theground was guided into the site by an observerin the aircraft. The method was expensive butextremely effective, with over 30 uncollareddingoes being selectively shot. When the col-lared dingoes were eventually shot as well, thesheep paddocks had been completely clearedof dingoes. Although effective control wasachieved, it is unlikely that this techniquewould ever be economically justifiable for anindividual grazier. In arid South Australia, air-craft are occasionally used for spotting din-goes that have breached the dingo fence.Ground shooters were guided to and shotabout 30 dingoes in two days on one property

following a major breach by dingoes in 1991(P. Bird, Primary Industries and Resources,South Australia, pers. comm. 1998).

6.4.3 Trapping

Trapping is still widely used against wild dogsin Australia, although in most areas, theemphasis has changed towards more specifictargeting of individual problem wild dogs,rather than general population control. InWestern Australia for example, trapping ismostly carried out by doggers employed bygroups of stations. Individual doggers maycover large areas (areas of one million hectaresare not uncommon) and rely on ground bait-ing and trapping to target wild dogs that havenot been killed during large-scale aerial baitingoperations (Section 5.1.1). Trapping tends tobe used where baiting is less effective, forexample, within sheep paddocks, where dogsare less likely to take baits because of theabundant and easily obtained food (P.Thomson, unpublished data 1976–1984).


Drying the surface of meat baits prior to injecting with 1080. In some areas, complete (rather than just surface) drying ofbaits may reduce non-target risks (Source: L. Allen, Department of Natural Resources, Queensland).

Trapping has the disadvantage that it islabour-intensive and requires training andexperience to be effective. Incorrect settingof traps can be inhumane and also result intarget animals becoming trap-shy. Carefulselection of trap sites is necessary to reducethe chance of trapping non-target animals(Newsome et al. 1983b). The proportion ofnon-target animals caught in traps variesaccording to the suite of animals present inan area and their relative abundance, theexperience of the trapper and the type oftrap used (Fleming et al. 1998).

‘Trapping is usually regarded as an inefficient method of

population control and is moreeffective against targeted

individuals.’

In much of Australia, the steel-jawed leg-hold traps used in the past are beingreplaced by more humane, padded-jawedtraps or snares (Fleming et al. 1998; Section4.2.3). Traps are buried and usually set withsome form of decoy (odours or carcasses) toattract the wild dog onto the trigger-plate.

McIlroy et al. (1986a) reported that trappingwas more successful than the use of groundbaits. Traps captured 56% of known wilddogs whereas baits removed 44%. However,because the control effort was not reportedand the methods indicate that it was unlikelyto be equivalent for trapping and poisoning,comparisons are impossible. It could beargued that neither technique was successfulbecause a large proportion of the knownwild dog population remained after bothtreatments. The true degree of success of aparticular method can only be determinedfrom a measure of the reduction of damageto livestock.

Trapping is usually regarded as an inefficientmethod of population control and is moreeffective against targeted individuals.However, a study in the Australian CapitalTerritory by Don Fletcher and Ian Faulkner(Environment ACT, unpublished data 1998)recorded trapping effectiveness in terms ofcatch per unit of trapping effort. They foundthat prior to 1993, trapping effort was largely

responsive to sheep losses by neighbouringlandowners. In 1995, in response toincreased sheep predation, trapping effortwas increased ten-fold from 1994. The num-ber of dogs caught per 1000 trap nightsdeclined linearly from 1994 to 1997 and in1998, approximately 1.5 dogs were caughtper 1000 trap nights compared to nearly 4per 1000 trap nights in 1993.

In terms of cost-effectiveness, trapping is anexpensive control option, the majority of thecosts being for time and labour. The returnfor effort of trapping is generally low, oftenin the order of five wild dogs caught per1000 trap nights (Fleming et al. 1998).However, this does not necessarily indicate apoor ‘success’ rate. If the five wild dogstrapped were the last particularly damagingindividuals encroaching on a sheep grazingarea, the effort could be deemed worthwhilebecause of the resultant long-term protectionfrom predation. Because trapping is some-times the only practical option for removingtroublesome individual wild dogs, the costsmust be weighed against the value of thedamage caused.

6.4.4 Poisoning

Poisoning is the most cost-effective lethaltechnique available to control wild dogs andit is often the only practical means forachieving population control in remote andinaccessible areas.

1080 (sodium fluoroacetate)

1080 has generally replaced strychnine asthe poison used in baits for wild dogs inAustralia (Section 4.2.4 and Chapter 5). 1080is also widely used in Australia for the con-trol of other vertebrate pests, particularlyrabbits and foxes. It is absorbed through thegastrointestinal and respiratory tracts andaffects the central nervous system, causingconvulsions. Its use is strictly controlled bygovernment regulations, which restrict theuse of the concentrated active ingredient totrained staff of government or semi-govern-ment agencies (for example, Rural LandsProtection Boards in New South Wales).Landholders are only permitted to use pre-pared products containing 1080 and thengenerally only after appropriate training.


Relative to native species, dogs and foxes areparticularly sensitive to 1080, so this makes itan appropriate toxin for these species interms of target specificity. The LD

50(the dose

of toxin required to kill 50% of test animals)for dingoes is approximately 0.1 milligramsper kilogram of body weight (McIlroy 1981).An estimate of 0.3 milligrams per kilogram foran LD

100(McIlroy 1981) means that a 20 kilo-

gram dingo would have to ingest six mil-ligrams of 1080 to ensure a lethal dose. Inreality, the LD

100for wild dogs is probably

somewhat less than this, based on records oflarge farm dogs that have been accidentallypoisoned and killed by eating single fox baitscontaining 4.5 milligrams or sometimes threemilligrams of 1080 (P. Thomson, unpublisheddata 1999).

‘Relative to native species, dogs and foxes are particularly

sensitive to 1080.’

1080 is broken down by micro-organisms inthe bait, soil and water (Wong et al. 1991).This detoxification is likely to vary consider-ably between different types of bait, meth-ods of preparation, and temperature andother environmental conditions (Wong et al.1991). These losses of toxin can be seen asan advantage over the longer term, reducingthe extended hazard to domestic dogs andother non-target species that may arise frombaits remaining toxic indefinitely. The lossesare not, however, justification for raising the1080 content of baits above the recommend-ed six milligrams per bait. Fleming andParker (1991) found that baits laid on thesurface in north-eastern New South Walesduring winter retained enough 1080 to deliv-er a theoretical LD

99for 42 days after injec-

tion. Baits collected after 90 days stillretained low levels of 1080 although on eachsampling occasion, 10% of baits still con-tained more than 2.0 milligrams of 1080, thetheoretical LD

99for wild dogs in eastern

Australia (Fleming and Parker 1991). Meatbaits buried during hot humid weather maybe completely decomposed and incorporat-ed into soil within four days (P. Fleming,unpublished data 1993). In drier areas, 1080may persist in baits for long periods. Twigg

et al. (1999) found most surface-laid meatbaits (67%) retained 2.5 milligrams or moreof 1080 eight months after placement.

‘The losses are not justificationfor raising the 1080 content abovethe recommended six milligrams

per bait.’

Concern is sometimes expressed that not allthe 1080 in a bait may be available to the tar-get animal. However, evidence from studieson foxes reveals that this is unlikely to be apractical issue (Saunders et al. 1995). In addi-tion, the target animals often eat more thanone bait (Thomson 1986), thereby theoreti-cally ingesting considerably more than theminimum amount of 1080 required to killthem.

A more serious concern about sub-lethalamounts of poison in a bait stems from thefact that 1080 is water-soluble and may beleached from fresh meat baits by rain fallingsoon after placement (McIlroy et al. 1988;Fleming and Parker 1991). Possible losses of1080 caused by rainfall immediately afterplacement must be taken into account. Thus,the suggested national standard 1080 contentfor a ‘dog bait’ in Australia of six milligrams(Thompson 1993) is soundly based.

Strychnine

Strychnine can still be used to poison thejaws of traps in Western Australia, SouthAustralia, Queensland and New South Wales.In all States and Territories exceptQueensland and some areas of SouthAustralia and Western Australia, strychninehas been phased out in baits for animal wel-fare and non-target susceptibility reasons. Itis absorbed through the gastrointestinal tractand acts on the central nervous system, caus-ing violent tetanic spasms (Seawright 1989).Strychnine is an odourless powder with a bit-ter taste and rapid action. Despite its rapidaction, strychnine is less humane in its actionthan 1080 because it causes deep reflexes,cramps and spasms in the muscles leading torespiratory failure and death (Hone andMulligan 1982). Strychnine is not a selectivepoison at the dosages required to kill wild


dogs (LD50

= 0.5–1.0 milligrams per kilogram(Hone and Mulligan 1982) and carnivorousbirds and mammals are susceptible to strych-nine meat baits. There is also the risk of sec-ondary poisoning (Hone and Mulligan 1982).

Cyanide

Sodium cyanide is a white powder that ishighly soluble in water; on contact withmoisture, hydrogen cyanide is released. Oninhaling hydrogen cyanide or absorbing itthrough the gastrointestinal tract, a poisonedanimal has difficulty in breathing, becomesunconscious, suffers convulsions and dies.This process can be very rapid but deathmay take up to four hours (Hone andMulligan 1982) and sub-lethal doses mayresult in bait shyness. In the USA, cyanide iscommonly deployed for the control ofcanids in a M-44® cyanide ejector and thesedevices have been tested for the control offoxes in Australia (Busana et al. 1998). Whendistributed in wax capsules for foxes,cyanide baiting caused many non-target poi-sonings in north-eastern New South Wales(Thompson 1994), but non-target deaths

were rarely recorded in Western Australiaduring extensive trials (P. Thomson, unpub-lished data 1992–1997). The main benefit ofcyanide over alternative poisons, especially1080, is its faster action. This is likely toresult in a reduced incidence of multiple baittake and hence caching or vomiting of sur-plus baits by wild dogs, which lessens therisk to farm dogs of accidental poisoning.Cyanide baiting for wild dogs is not currentlylegal in any State or Territory (Busana et al.1998).

Bait materials

Three types of bait are used in wild dog con-trol. The first two, moist meat and dried meat,are the most commonly used and are madeon-site from meat cut from culled kangaroos(Macropus spp.), cattle or horses. Baits are cutto size then injected with six milligrams of1080, except in the Australian Capital Territoryand Victoria, where 4.5 milligrams of 1080 isused, and parts of Queensland where up toten milligrams of 1080 is surface-applied tobaits. In Western Australia pastoralists are alsoable to make small quantities of baits by


The use of helicopters allows more precise aerial baiting in rugged terrain (Source: P. Fleming).

inserting an oat grain impregnated with sixmilligrams of 1080 into the meat; the 1080leaches from the oat into the meat. Moist meatbaits are used immediately but dried meatbaits are sun-dried before use to about 40% oftheir original weight.

The toxicity of moist meat baits declinesmuch more rapidly than dried meat baits(McIlroy et al. 1988; Fleming and Parker1991; Kirkpatrick 1999). The dry baits arequite hard and therefore difficult for non-tar-get species (birds and small carnivorousmarsupials) to eat (Calver et al. 1989).

‘The toxicity of moist meat baitsdeclines much more rapidly than

dried meat baits.’

The third type of bait is the manufacturedDoggone® bait. These baits contain six mil-ligrams of 1080 and are made from a softmeat-like substitute based on meat meal.They also contain fat, preservatives, bindingagents and flavour enhancers.

Moist meat baits are used in the NorthernTerritory (300–400 grams), Queensland (125grams in western areas and 250 grams incoastal areas), New South Wales (230 grams),the Australian Capital Territory (230 grams),South Australia (150 grams) and Victoria(approximately 125–230 grams). Dried baitsare used in South Australia and WesternAustralia (110 grams wet weight). Doggone®

baits (60 grams) are sometimes used in Victoriaand the Australian Capital Territory.

Occasionally, meat baits are dyed with tastelessgreen vegetable dye to reduce the chance ofdetection and removal by birds (McIlroy et al.1986b).

Ground baiting

Ground baiting is still an important controlmethod used throughout the country and maybe strategic or reactive (Section 7.3.4). Groundbaiting allows for baits to be strategicallyplaced to maximise the chances of wild dogsencountering them, at the same time reducingthe risks of them being taken by non-target ani-mals. The more strategic placement means thatfewer baits are used than in aerial baiting oper-ations.

Studies (Best et al. 1974; McIlroy et al. 1986a;Bird 1994; Fleming 1996a) of the effectivenessof ground baiting have revealed widely vary-ing results. Variable results can be attributed to:the availability of natural food (Best et al. 1974;Thomson 1986); the removal of multiple baitsby a single wild dog (Bird 1994; Fleming,1996a); removal of baits by non-target animals(particularly ants, foxes and birds; Newsome etal. 1972; Allen et al. 1989); the number of baitsavailable (or remaining) for the target animalsto find (Fleming 1996a); and the attractivenessof the baits (Allen et al. 1989). There is evi-dence that burying baits reduces the take bymost non-target animals (particularly birds)without reducing the uptake by wild dogs(Allen et al. 1989). Fleming (1996a) usedburied baits and reported a very low removalby native non-target animals. For this reason,bait mounds are used in the forests of south-eastern New South Wales, the AustralianCapital Territory and north-eastern Victoria.Bait mounds are bait stations comprising oneor more baits buried in a mound of earth andoften surrounded by an area of raked soil. Theraked soil allows the identification from foot-prints and other signs of animals visiting thestation. If a baiting program is conducted withan initial free-feeding period when unpoi-soned baits are placed in the mound, visits bynon-target animals can be identified and onlymounds visited solely by target species refilledwith poisoned baits.

‘Burying baits reduces the take by most non-target animalswithout reducing the uptake by

wild dogs.’

The use of bait stations is becoming morecommon. Replacement-baiting in bait sta-tions was suggested by Thompson andFleming (1994) as a method where baits arereplaced as they are removed by foxes overa period of ten days or so to reduce the bait-susceptible population to near zero. Fleming(1996a) showed replacement-baiting is asuccessful method for reducing wild dog andfox numbers in eastern New South Wales. Itis cheaper than aerial baiting and requiresless skill and experience than trapping.Ground baiting remains a primary control


tool in many accessible areas because theplacement of baits can be better controlled.

Some doggers are reluctant to use 1080 baitsbecause the carcasses of baited animals areseldom found. This means that it is more dif-ficult to tally a definite ‘kill’ and some seethis as a lost opportunity to retrieve scalpsfor bounty payments. These operators preferto use traps instead. This is a major practicaldrawback of bounty or bonus payments, andone that managers need to take into account.In many instances, baiting is a more cost-effective technique, with trapping only beingdesirable to target specific individual dogs.

Aerial baiting

Aerial baiting using 1080 baits is a cost-effec-tive method for the strategic management ofwild dogs over large areas (Thomson 1986;Thompson and Fleming 1991). It is also aneffective means of delivering baits in remoteand inaccessible areas where ground-basedoperations are impossible or impractical.Aerial baiting began in Western Australia andQueensland in 1946 (Tomlinson 1954).Subsequently, considerable improvements tothe materials and operations have been made.

‘Helicopters allow more accurate delivery of baits in

rugged mountain areas.’

Fixed-wing aircraft are the most suitable fordropping baits in flat terrain whereas themanoeuvrability of helicopters allows moreaccurate delivery of baits in rugged mountainareas (Thompson et al. 1990). Baits aredropped through a chute in the aircraft on thecommand of a navigator with local knowledgeof wild dog activity. Increasingly in New SouthWales and Western Australia, the aircrafts’flight-paths are automatically logged into anon-board global positioning system (GPS),providing accurate records of baiting opera-tions. A future development in WesternAustralia will be the incorporation of sensorsinto the bait chute so that the location of eachbait dropped will also be automatically logged.

Timing and frequency of baiting

The question of when and how often baitingshould be carried out is frequently raised butthere is no simple answer. Variables includeresources (usually dollars), value and vulner-ability of the livestock being protected, avail-ability of alternative prey for the wild dogsbeing targetted, and season (weather, avail-ability and distribution of water, stage of thebreeding cycle).

‘Aerial baiting is usually repeated on an annual cycle.’

Traditionally, baiting campaigns in WesternAustralia (Thomson 1986) and Queenslandwere carried out in autumn (late April–May)and spring (September–October). Autumncoincided with the breeding activity of din-goes when mating takes place and bitchesare in early pregnancy. Spring coincidedwith the time when pups begin to moveabout, removing the restrictions on move-ments associated with denning (Thomson1992d), and increasing the likelihood of ani-mals encountering baits. Food demands arealso likely to be high for groups of dingoes atthis time, and in northern areas surface waterbecomes more restricted, making it easier totarget the limited number of waterholes withbaits. This approach is still followed,although baiting is often now only done inspring in Western Australia. Baiting earlier inthe year is sometimes abandoned due to costand the possibility of rain reducing the life ofthe baits (due to leaching of 1080). In someareas, baiting even later in the year has beenconsidered, when the availability of water iseven more restricted.

In eastern Australia, aerial baiting is usuallyconducted in late autumn and winter. This isbecause:

• Baits distributed in autumn and winter lastlonger than those distributed in spring andsummer (Saunders and Fleming 1988;Fleming and Parker 1991), as baits maydecompose in a few days in the conditionsof high temperatures and humidity, or maybe quickly removed by ants (Fleming1996b).


• Bait take by non-target species that areless active in cooler weather, such asgoannas (Varanus spp.), is reduced.

• The population of wild dogs is potential-ly at its lowest prior to whelping inspring and so a given proportionalreduction in abundance in autumn andwinter equates to a smaller remnant pop-ulation than in spring and summer.

• In eastern New South Wales most of theaerial baiting occurs in areas of extensivecattle enterprises adjacent to sheep coun-try and mustering is usually completed bywinter reducing the chances of workingdogs being baited while spring mustering.

• Lambing in north-eastern New SouthWales is most common in spring andbaiting conducted in winter prior tolambing is sensible.

• Movements of wild dogs are traditionallybelieved to be greatest in autumn andwinter when young dogs are dispersingand mating is occurring.

Aerial baiting is usually repeated on anannual cycle. Fleming et al. (1996) found thatindices of wild dog abundance taken prior toannual aerial baiting in north-eastern NewSouth Wales were similar between yearsindicating that the wild dog population hadrecovered during the year following the pre-vious baiting. In some areas, there may beoccasions when baiting in buffer areas orbaited zones (Section 6.2.1) could be missedin some years without jeopardising livestock(Thomson 1986). However, this could be arisky undertaking unless detailed informa-tion was available on how many wild dogswere in an area, how many were breeding,and food supply. Otherwise, the safest pre-cautionary action is to bait the known prob-lem areas on an annual basis.

Baiting strategies

One of the concerns raised about the appli-cation of haphazard control efforts is that thedisruption caused by killing a few membersof a pack could lead to increased pup pro-duction (Section 2.8). This could lead torapid repopulation and possibly even an

increase in dispersal from such an area.Therefore, strategies should be developedfor baiting and other control efforts.

‘A buffer zone is more efficient than controlling

populations less intensively over much larger areas.’

The use of poison baiting is either strategicor reactive (Section 7.3.4) and this affectshow it is conducted. Reactive baiting is inresponse to predation incident(s) and is usu-ally conducted at the property level and con-tinues until the offending animal(s) are killedand predation stops. It is essential that thisfacility remains available to graziers particu-larly so that they can respond to the occa-sional failure of strategic programs.

It is clear from studies in various parts ofAustralia that the movements of wild dogsare usually limited. It is rare for individuals tomake long-distance forays beyond their nor-mal range, refuting the commonly-heldnotion that wild dogs regularly travel largedistances overnight to attack livestock. Inaddition, there is little evidence that dingoesor other wild dogs undertake any seasonalmigration. Under severe conditions, dispers-ing wild dogs may congregate around a limit-ed resource (for example, a waterhole), pos-sibly giving rise to the notion that some formof ‘migration’ had taken place. Wild dogs donot engage in the annual juvenile dispersalmore common among less social canids suchas red foxes (Saunders et al. 1995), however,individual wild dogs do occasionally dis-perse (Thomson et al. 1992a; P. Fleming andD. Jenkins unpublished data 1999). It followsthat wild dogs that reside within paddocks oradjacent areas pose the greatest danger tolivestock. Control activities should thereforebe focused in these areas.

The introduction of aerial baiting enabled wilddog control to be undertaken over large areasadjacent to areas where livestock were grazed.Thomson (1984b) investigated the value ofbaiting within these adjacent areas (bufferzones) to protect livestock from predation bydingoes in Western Australia and found that itwas reduced or eliminated for two or moreyears after buffer zones were established.


The strategy of creating a buffer zone(Thomson 1984b), approximately one or twowild dog home ranges wide, adjacent tostocked paddocks is based on the rationalethat by removing most resident wild dogs a‘dispersal sink’ results. This effectivelyencourages dispersing wild dogs to settlerather than keep moving to reach the pad-docks and is more efficient than to attemptto control populations less intensively overmuch larger areas. With a well-establishedbuffer zone, constant control work is unnec-essary. However, monitoring and periodicmaintenance control work is needed toavoid any substantial build-up of numbers inthe buffer zone. The buffer zone conceptforms the basis of most aerial control pro-grams in Western Australia (Thomson 1986)and northern New South Wales (Fleming etal. 1996). The environments and the size ofholdings in north-eastern New South Walesare vastly different from Western Australiaand buffer zones of two home range widthsare not practicable. Baiting is focussed at theinterface between sheep country and wilddog country (Fleming 1996b).

In South Australia, a buffer zone of up to 35kilometres has been recently established out-side the Dog Fence (Animal and PlantControl Commission 1993). The aim of thebuffer is to reduce the incursions of wilddogs through the Dog Fence and decreasethe vigilance required to maintain the DogFence in dog-proof condition.

6.4.5 Control at breeding dens

The practice of locating breeding dens(Section 4.2.5) has diminished as a routineseasonal activity and tends to be undertakenopportunistically. In areas where bountiesare paid, there is sometimes a tendency fordoggers to delay targeting pregnant bitchesuntil pups are born and ‘collectable’, increas-ing the tally of scalps and therefore pay-ments received. This is a further example ofhow bounty payments can influence controlwork in an undesirable way.

6.4.6 Sheep-guarding dogs

In North America and some areas of CentralEurope and the Mediterranean, special dogshave been bred to guard sheep from predators

including coyotes (Canis latrans) and wolves(Canis lupus ssp.) (Copinger and Copinger1978). These large dogs live in the paddockswith sheep as part of the flock and chase off orkill other canids that approach their flock.

In Australia, two breeds, Anatolian karabashand maremma, have been used to protectsheep and goat flocks from predation by wilddogs and foxes. Sheep-guarding dogs have notbeen experimentally tested in Australian condi-tions but studies in the United States ofAmerica (USA) have given uncertain results.Although sheep producers with sheep-guard-ing dogs suffered fewer predation losses tocoyotes than those without (Andelt 1992),results are inconsistent (Green et al. 1994). InVictoria, maremmas have been used to reducestock losses from 10% to 3% in eight yearssince their first use (Balderstone 1992) andhave been used successfully for approximatelyten years in preventing predation by free-roaming dogs at Livingstone Farm, Moree innorthern-central New South Wales (G. Esdaile,University of Sydney, pers. comm. 1986;Balderstone 1992).

6.4.7 Aversive conditioning andtoxic collars

Aversive conditioning is the process of traininga predator to avoid an activity or place by asso-ciating the activity or place with unpleasantexperiences. Aversive conditioning of preda-tors with lithium chloride has been the subjectof much research (Burns 1983; Gustavson et al.1983; Tauchmann 1998), although the resultshave been mixed. Cornell and Cornely (1979),for example, successfully used lithium chlorideto deter coyotes from scavenging and beggingat camp grounds in a National Park in Americabut other studies produced inconsistent results.Currently, the use of aversive conditioning isbeing trialed at Fraser Island to deter dingoesfrom camping areas.

Toxic collars have been suggested and triedfor the control of coyote predation of sheepin the USA (Connolly 1980). The collars con-tain pouches of toxic liquid such as a solu-tion of 1080. The technique relies on thecoyote biting the sheep at the neck, punctur-ing the collar, and ingesting the toxin.Scrivner and Wade (1986) in a review of field


trials, concluded that the collars were foundto be an effective method when used in con-junction with other control techniques. Thetechnique was most effective when a groupof collared ‘sacrificial’ animals were isolatedfrom other livestock. Problems experiencedincluded the fact that coyotes sometimesattacked elsewhere than at the throat, collarswere damaged or lost (caught on thorns,wire), or collars were pulled out of idealposition on the animal.

The toxic collar technique is labour intensive,and relies on intensive husbandry of livestockand isolation of animals in small areas. It istherefore unlikely to be useful against wilddogs in rangeland areas of Australia, but itmight be feasible for use in the more intensivegrazing areas. However, the same problems asthose identified in the USA are likely to occurin Australia. Like coyotes, wild dogs do notalways attack the throat of sheep; for example,in a sample of 133 sheep killed by dingoes inWestern Australia, 56% had received mortalwounds to the throat (Thomson 1984a).Welfare concerns may well be raised about thenotion of sacrificing sheep to wild dogs, partic-ularly if other methods are available to preventattacks in the first place. In addition, the use of1080 in this way may breach current State orTerritory regulations or policy on 1080 use. Atpresent, there are no other obvious candidatetoxins available. Interest in toxic collars hasbeen shown in Queensland (L. Allen,Queensland Department of Natural Resources,pers. comm. 2000).

6.4.8 Livestock management strategies to reduce or prevent predation

The most obvious management strategy toavoid the predation of sheep by wild dogs isenterprise substitution, where sheep arecompletely or partly replaced by cattle. Thisstrategy has been employed by some pro-ducers in the sheep zone of northernQueensland outside the current Dog Fence(L. Allen, Queensland Department of NaturalResources, pers. comm. 1997). This was inpart caused by changes in profitability ofsheep and cattle production and seasonalconditions. Similar changes have occurred ineastern New South Wales. When the margin

from sheep production becomes too low,relatively low levels of predation may limitprofitability and the high cost of enterprisesubstitution may become beneficial.

‘The most obvious managementstrategy to avoid predation is

enterprise substitution.’

Enterprise substitution is not always possiblebecause the land may be better suited forsheep production and the cost of replacingsheep with cattle can be high. There is atime-lag between setting up the cattle enter-prise and the receipt of income, particularlyin breeding enterprises and the lack of cash-flow may be limiting in the first phases.Change of enterprise also raises the level ofrisk because of the delayed returns exposethe producer to greater likelihood ofunfavourable seasonal conditions. Lean cat-tle are severely discounted or unsaleablewhereas lean sheep still produce wool. Theknowledge base required to become a suc-cessful cattle producer is not the same as thatrequired for successful wool or sheep meatproduction. There are important emotionalconsequences suffered by producers whoregard themselves as sheep graziers andtherefore find the change to cattle produc-tion difficult. These factors must be consid-ered before substituting cattle for sheep.

In eastern Australia, in dog-inhabited areaswhere both sheep and cattle are grazed, pre-dation of calves is rare (Fleming and Korn1989). Graziers put cattle in paddocks closeto timbered wild dog refugia and sheep inpaddocks further away so that the cattle canprovide an internal buffer to predation.Similarly in South Australia, sheep graziersoften run cattle in paddocks adjacent to theDog Fence where risks of predation aregreatest. Although this strategy has not beenformally tested, logic and circumstantial evi-dence suggest that it reduces predation ofsheep. Other graziers run wethers in pad-docks closer to wild dog refugia and grazetheir breeding sheep further away. While thisstrategy may not reduce the incidence ofpredation, the economic consequences ofpredation are lessened because profitabilityof breeding ewe enterprises is most strongly


linked to lambing and weaning rates andbecause genetic material is less likely to belost.

‘Most calves killed by dingoes during these periods do

not result in economic lossesbecause calves cannot survive

long droughts.’

On the Barkly Tableland of the NorthernTerritory, the dingo mating season coincideswith the peak in calving and calves. At thistime dingoes move around a lot and areoften seen together with calves at wateringpoints (Corbett 1995a). This contributes tothe deaths of many calves. Therefore, anappropriate management option would beto manipulate cattle herds so that fewercalves are born in the dingo mating season.Strategic mating of cattle has also beenshown to be more economically productivein much of tropical Australia (Sullivan et al.1992). This management strategy is practisedin the Alice Springs region and calf losses arerarely recorded, probably because the peakof the calf drop (September and October)does not coincide with the dingo mating sea-son. Although droughts in the Alice Springsregion are unpredictable in frequency andseverity, the effects of wild dog predation arepredictable and pastoralists can benefit fromthis factor in the following three ways(Corbett 1995a):

• In the Alice Springs region, high numbersof dingoes are maintained because humancontrol methods effectively increase thenumbers of breeding dingoes and becauseabundant natural food ensures the survivalof most pups. During runs of flush years,dingoes eat mostly rabbits and rodents, butin the middle and latter years of drought,dingoes survive well on cattle (Figure 10a),usually carcasses. Most calves killed bydingoes during these periods do not resultin economic losses because calves cannotsurvive long droughts. In fact, such calflosses are probably an advantage to pas-toralists because the mother’s chance ofsurviving the drought is increased if shehas no calf to feed. Thus, in situations like

this, dingoes benefit pastoralists by help-ing breeders survive droughts and themanagement strategy is easy: relax dingocontrol in droughts.

• Management decisions during the initialperiods of droughts are complex. After afew dry months, the populations of smallnative mammals and rabbits declinebefore cattle begin to die. Dingoes there-fore have no choice but to tackle largekangaroos or kill calves, so dingo controlis necessary. If the drought breaks after afew months, such calf losses are indeedeconomic losses (Figure 10b). Since thelosses are beneficial if the drought con-tinues, when should a pastoralist height-en or relax dingo control? One pastoral-ist in central Australia is reported to haveshot calves, not wild dogs, after the firstseven months of a drought, and also tohave swapped truckloads of his calvesfor hay from farmers in southernAustralia. It was also reported that thisproperty recovered from the droughtmuch earlier than neighbouring proper-ties, which suggests that predation bydingoes and destocking both work well.A strategic approach for drought man-agement in northern and centralAustralia is provided in RANGEPACK(Stafford-Smith and Foran 1990).

• The presence of dingoes may also benefitpastoralists immediately after drought.Rabbits and small native mammals quick-ly resurge when rain comes, and wilddogs concentrate on them (Figure 10b). Ifdingoes numbers are high due to relaxedcontrol effort during the drought, theheavy predation will hold down rabbitnumbers so that they take longer (per-haps two years) to reach their former lev-els. This gives a pastoralist more time toestablish other forms of rabbit control(Williams et al. 1995) before they becomea serious problem again. In effect, wilddogs and pastoralists work in tandem thisway to control rabbits and other grass-eating competitors of cattle.


6.4.9 Biological control

At present, there are no biological controlagents suitable for use against wild dogs(Section 4.2.7). With the goal to protect din-goes in conservation areas in Australia, it isunlikely that any self-disseminating means offertility or lethal biological control of wilddogs would ever be sanctioned. Domesticdogs would also require protection againstsuch an agent.

It is possible that with advances in molecularbiology, some form of agent could be devel-oped that is specific to dogs, targeting hor-mones or proteins important in the repro-ductive process (Tyndale-Biscoe 1994).However, this would need to be bait-deliv-ered to avoid the problems already outlined.There may be little practical benefit of such abait, unless it was target-specific and couldbe used in specific areas where toxic baitsare perceived as posing too great a risk toother fauna such as quolls (Dasyurus spp.).

6.4.10 Livestock insurance and compensation schemes

In USA, damage compensation schemeshave been instigated for coyote damage tosheep flocks, ungulate damage, and bear(Ursus spp.), wolf and mountain lion (Pumaconcolor) predation (Wagner et al. 1997).Such schemes recognise the national bene-fits of conservation of native wild animals(such as the dingo) without farmers bearingthe costs associated with the presence ofthose animals.

An alternative to spending money on controlof wild dog populations and barrier fences isthe introduction of compensation schemes.These may be self-funding, funded by gov-ernment agencies or by a combination ofrates and subsidies. There are risks associat-ed with compensation schemes, as they arereliant on the financial reserves of thescheme and the honesty of claimants. Self-funding schemes, with in-built checks suchas validation of claims by independentagents, potentially overcome the problem ofdishonesty but do not address funding limits.A self-funding compensation scheme mightbest be operated like an insurance programwhere annual rates (equivalent to insurance

premiums) are levied on all holdings includ-ing government lands, and distributed toaffected landholders in response to lossesverified by an authorised person. If no claimsare made, the funds might be invested andthe rates recalculated. Farmers should beencouraged to minimise damage by ratereductions for predation mitigation and com-pensation should not be open ended(Wagner et al. 1997). The feasibility andoperation of a compensation or insurancescheme requires investigation under the dif-ferent agricultural systems where wild dogpredation occurs before such a scheme isimplemented.

6.5 Costs of control

There are a number of techniques for mea-suring the costs of control programs. Themost basic in which all the costs of the con-trol program are added together at the end ofthe program is adequate for planning inmany cases. For example, the helicoptertime used for the annual aerial baiting pro-gram in north-eastern New South Wales iscollated each year so that costs can be pro-jected during planning and price risesaccounted for (A. Barnes, NSW Agriculture,pers. comm. 1998). Helicopter time accountsfor approximately 50% of the total cost ofbaiting (Thompson and Fleming 1991) andso is a useful index of costs. Thompson andFleming (1991) also found a strong linearrelationship between the amount of meatused and the total cost of aerial baiting in1988. Similarly, the costs of fence mainte-nance and erection could be estimated usingthe cost of materials per kilometre, the timerequired, the cost of labour and the distanceto be repaired, built or modified.

More complex evaluations of control programsare not commonly attempted. Benefit–costratios are simple estimators of the effectivenessor the efficiency of a program (Hone 1994;Case study 2 in Section 7.7.2). These analyseshave usually been applied to government-sub-sidised control programs. For example, inVictoria during the 1980s, much of the controlof wild dogs was undertaken by 21 doggerswho were employed by the State (Mitchell1986). A benefit–cost analysis showed that the


agricultural benefits were hard to estimatebecause of the externalities including theabsence of livestock production estimates fromcomparable areas with unmanaged wild dogpopulations. However, the control programwas considered beneficial when productivitygains needed to break even were comparedwith the productivity gains that were experi-enced in areas where control occurred. A num-ber of problems associated with analysis ofcost-effectiveness were raised including thatthe indicator of cost-effectiveness should becompared over years; and the measurement ofeffectiveness should not be the simple calcula-tion of the cost per wild dog killed becausetrapping return for effort decreases exponen-tially as the population declines towards zero(Mitchell 1986).

‘Graziers commonly regard theircontrol activities as an insurance

policy against stock losses.’

Marginal analysis (Leftwich and Eckert 1982;Hone 1994; Box 3 in Section 7.4) aims toallow a producer to estimate the increasedmargin (profit) from an enterprise that can bemade from different levels of input.However, marginal analysis appears to be oflimited use for evaluating the economics ofcontrol methods for wild dogs because of thehigh variability in the nature and extent ofpredation. For example, the flexible nature ofwild dog society, prey switching, and surpluskilling all affect the relationship between wilddog abundance and livestock predation, thatis, the damage function. Fleming and Nicol(unpublished data 1999) have evidence fromnorthern New South Wales that, while thedamage function is significantly linear, thefunction is a poor predictor of damage at dif-ferent densities of wild dogs. Thomson’s(1984a) data indicate that, in WesternAustralia, where sheep and wild dogs co-occur, predation of sheep is inevitable. This isthe underlying assumption of control strate-gies aimed at reducing wild dog abundanceand buffer zone strategies (Thomson 1984b;Fleming 1996a). The variability of damagemakes marginal analyses of control methodsdifficult (Box 3 in Section 7.4).

Another method of evaluation is to comparethe cost of control with the potential savings.This is a comparison of costs with potentialbenefits rather than achieved benefits. Forexample, a survey of participants in the 1988aerial baiting program on the NorthernTablelands of New South Wales costed thatsingle operation at $152 750 (Thompson andFleming 1991). The cost per participant was$540 of which $350 was subsidised by theNew South Wales Government in recogni-tion of the control that private landownersconducted in government lands. The costs ofthe aerial control program in 1988 wereequivalent to the loss of 7.5 wethers perlandholder (Thompson and Fleming 1991)and this was the potential benefit.

The cost per unit effort of control is anothermethod of estimating the cost of control.Table 5 shows the effort expended on con-trol of wild dogs in a number of surveys innorthern New South Wales. The Table showsthat the amount of labour expended on wilddog control decreased after the advent andwidespread use of aerial baiting in north-eastern New South Wales. Greater value canbe gained from such comparisons if thepotential and real benefits are calculatedsimultaneously.

Most commonly, graziers regard their controlactivities as an insurance policy against stocklosses. That is, if control is not exerted, dam-age will occur and continue to occur until theoffending dog is removed, so that moneyspent now is money saved later. The secondassumption underlying control programs isthat there is a positive relationship betweenthe abundance of wild dogs and the probabil-ity of damage. Where buffer zones are used,the assumption is that reducing the popula-tion of wild dogs in the buffer will reduce theimpetus for emigration of wild dogs to theadjacent grazing land where predation thenbecomes inevitable (Thomson 1984a). If theseare the underlying assumptions, then thestrategy must be to achieve the greatestreduction in the probability of predation forthe lowest possible cost. The cost is not open-ended; funds for wild dog control in this caseare usually allocated on the basis of howmuch a particular program costs and availablecash flows.


6.6 Environmental and non-target issues associated with1080 baiting

6.6.1 Environmental risks

The environmental fate of 1080 has beenstudied more extensively than any other ver-tebrate pesticide, particularly in NewZealand, where possum (Phalangeroidea)and rabbit control accounts for approximately70% of the world-wide use of the toxin(Eason et al. 1998). These authors report thatsodium fluoroacetate does not bind to soilconstituents and is detoxified quickly by soilorganisms, the rate depending on soil tem-perature and moisture. Only very small quan-tities are absorbed by plants so there is a neg-ligible risk of poisoning herbivores. None of857 surface water samples collected immedi-ately after aerial baiting programs for rabbitsor possums exceeded the acceptable (NewZealand) concentration for drinking water offive parts per billion. Because of very lowconcentration of 1080 applied to the environ-ment, and the rapid biodegradation of thetoxin, wild dog baiting is very unlikely tocause environmental hazards.

6.6.2 Risks to non-target species

McIlroy (1981) tested a number of nativespecies potentially at risk of poisoning during1080 baiting campaigns. On a weight-for-weight basis, native mammals are more toler-ant of 1080 than dogs. Birds and reptiles areeven more tolerant. In Western Australia, King(1989) assessed the northern quoll (Dasyurushallucatas) as being the species most likely tobe at risk during baiting campaigns for wilddogs. King radio-tracked a sample of northernquolls during a wild dog baiting campaignand found that all survived, despite condi-tions of apparent food shortage and potential-ly enhanced vulnerability. He concluded thatpopulations of northern quolls faced little riskfrom wild dog baiting campaigns. This alsoappears to be the case in parts of north-east-ern New South Wales where spotted-tailedquolls (Dasyurus maculatus) are still abun-dant after three decades of baiting (B. Harden,unpublished data 1999).

In eastern Australia, the spotted-tailed quoll isconsidered potentially at risk during baitingcampaigns against wild dogs and foxes (Smithet al. 1992; Belcher 1998). Adult female spot-ted-tailed quolls weigh up to four kilogramsand males to seven kilograms, and reach maxi-mum weight in their third year (Belcher 1998).


Table 5: The effort expended for the control of wild dogs by landholders in north-eastern New South Wales. Data werecollected from members of Wild Dog Control Associations or equivalent organisations.

Year

1962a

1981

1985

1988

Major method of control

Dog-proof fencing, hunting, trapping

Aerial baiting (fixed-wing aircraft),fencing

Aerial baiting (fixed-wing aircraft),fencing

Aerial baiting (helicopter), fencing

Control effort(hours/property/year)

39.3

15.9

12.5

11.6b

Source

New England RuralDevelopmentAssociation (undated c. 1966)

Schaefer (1981)

Saunders and Fleming(1988)

Thompson and Fleming(1991)

aFirst aerial baiting conducted in 1962 but widespread use did not occur until after 1965.bControl effort for aerial baiting alone.

McIlroy (1981) estimated their 1080 LD50

at 1.85milligrams per kilogram (95% confidence limitsof 1.28–2.68) and suggested an LD

100at 2.56

milligrams per kilogram. On that basis, a singlefresh wild dog bait (6 milligrams of 1080) con-tains an LD

50for spotted-tailed quolls less than

3.2 kilograms, or an LD100

if less than 2.3 kilo-grams in weight. The complete ingestion ofone fresh wild dog bait then has the potentialto kill a spotted-tailed quoll juvenile of eithersex or a small adult female. However, the toxic-ity of fresh meat wild dog baits declines rapidlyafter injection of the 1080, reducing the risk tospotted-tailed quolls. In the Australian CapitalTerritory, McIlroy et al. (1988) estimated that awild dog bait contained an LD50 for an aver-age-sized (2.8 kilograms) spotted-tailed quollfor 4–15 days after placement in winter butonly 2–4 days in summer. In a similar experi-ment during winter in forest in north-easternNew South Wales, Fleming and Parker (1991)estimated the same reduction in toxicityoccurred within two days of bait placement.

1080 ingested by the mother can be trans-ferred in the milk to suckling young. McIlroy(1981) found that more pouch young brushtailpossums (Trichosurus vulpecula) died thantheir mothers at each dose level of 1080 heevaluated. Eight pouch young of one northernquoll also died after their mother received asub-lethal dose, although the five young of aspotted-tailed quoll survived in similar circum-stances.

Captive spotted-tailed quolls can locate, exca-vate and consume fresh meat and non-poi-soned commercial fox bait buried to depths ofless than about ten centimetres (Belcher 1998;Murray 1998). However, the extent of baituptake by spotted-tailed quolls during wildcanid baiting programs is unclear. In north-eastern New South Wales, Fawcett (1994) andFleming (1996a) both reported that few buriedmeat baits were taken by spotted-tailed quollsduring baiting programs for the control of wilddogs. Fleming (1996a) concluded spotted-tailed quolls were not as readily attracted tobait stations as wild canids and appeared reluc-tant to remove buried bait from bait stations.In two extensive Victorian trials with buriednon-poisoned commercial fox bait, Murray(1998) also reported low visitation rates byspotted-tailed quolls, and was uncertain thatbait was taken by the quolls. While these

results indicate that spotted-tailed quolls maytake some baits during wild dog baiting pro-grams, they provide little insight into the extentand severity of any impact of baiting programson spotted-tailed quoll populations. That ques-tion can only be answered by field-basedresearch that directly examines bait uptake andthe effect on populations.

‘It may be possible to target wild dogs and foxes

preferentially in areas wherethey co-occur with quolls.’

Smith et al. (1992) also suggested that poi-soning programs to control wild dogs mightadversely affect the viability of critical bodyweight range (CWR) mammals (35–5500grams) by allowing mesopredator release offox populations (Sections 3.2 and 3.6).However, Fleming (1996a), at sites in north-eastern New South Wales, showed that foxpopulations experienced greater impactfrom a ground-based replacement-baiting(Section 6.4.4) than did populations of wilddogs, and that the remaining populations ofwild dogs were more abundant than those offoxes.

The management of wild canids, in areaswhere spotted-tailed quolls occur, requiresthought. Spotted-tailed quolls are likely to havedifferent foraging and movement patterns towild dogs and foxes, and may be more inde-pendent of roads and cattle pads than wildcanids (Fleming et al. 1996; Fleming 1996b).Consequently, it may be possible to target wilddogs and foxes preferentially in areas wherethey co-occur with quolls by using bait buriedat bait stations along roads and tracks, as theseare commonly used by wild dogs and foxes(Harden 1985; Catling and Burt 1995).

Marks et al. (1999) are investigating the pos-sible benefits of using the M-44® device todistribute 1080 in a more target-specific way.Less 1080 is required and the positioning ofthe distribution device can be modified toencourage wild dogs to trigger the deviceand obtain a full dose. Both these factorswould reduce the potential for quolls to befatally poisoned during wild dog control.These devices can also be used to applycyanide.


Summary

There are four stages in a strategic manage-ment program for wild dogs at the local andregional level: (1) problem definition; (2)developing a management plan; (3) imple-menting the plan; and (4) monitoring andevaluating progress.

Defining the problem is the first stage ofstrategic management planning. This involvesidentifying who has the problem with wilddogs, what the problem is, when, where andwhy it happens and how much it costs.

The costs and benefits of reducing the agri-cultural impact of wild dog predation canbe measured in dollars. In some situations,sufficient information is available to esti-mate the point where the costs of undertak-ing dog control equal the benefits. However,each situation must be reviewed individual-ly because interactions between wild dogsocial behaviour and predation may makeconventional economic analyses inappro-priate.

The second stage in strategic managementplanning is the development of a manage-ment plan. This requires setting manage-ment objectives that should include interimand long-term goals, a time frame for achiev-ing them and indicators for measuring per-formance. Developing a management planalso requires the selection of appropriatemanagement options. Options for pest controlinclude local eradication, strategic manage-ment, reactive management or no dog con-trol. A management strategy is then devel-oped that prescribes what will be done andwho will do it. The management strategy alsodescribes how the selected control techniqueswill be integrated and implemented. Strategicmanagement of wild dogs is based on theconcept of adaptive management, in whichthe management plan is flexible, respondingto measured changes in economic, environ-mental and pest circumstances. Economicframeworks are needed to assess the value ofalternative strategies to manage wild dogs.

In some situations, management plans thatinclude conservation strategies for dingoes arerequired so that potentially conflicting goalscan be encompassed. A management planmay need to integrate control and conserva-tion techniques into a systematic program.Consultation between stakeholders and clearidentification of the goals of management pro-grams is critical for avoiding conflicts betweenstakeholder groups with different legal obliga-tions and objectives. Consultation and part-nership encourages mutual ownership of aproblem and results in ongoing participationand cooperation, group reinforcement,improved communication and enhanced effi-ciency and allocation of resources.

Wild dogs have large home ranges and oftentraverse boundaries between lands managedby different stakeholders. Group action is anessential element of the implementation ofmanagement plans. Management requirespartnerships between stakeholders if it is to beeffective. Programs must be flexible enoughto account for the different legal obligationsand different ecological, social and econom-ic imperatives of stakeholder groups.

Monitoring and evaluation of outcomesoccurs at different levels throughout theimplementation and on completion ofactions. Operational monitoring, where thecosts of actions are recorded and reviewedduring the program, ensures that the man-agement plan is executed in the most cost-effective manner. Performance monitoringassesses the effectiveness of the managementplan in meeting the agricultural productionor conservation objectives that were estab-lished in the management plan. Evaluationof data from both forms of monitoringenables the continuing refinement of themanagement plan, where necessary. Realand hypothetical examples of the strategicmanagement of wild dogs in agriculturalproduction scenarios are presented.

It is the responsibility of government agen-cies to encourage best practice managementand evaluation and monitoring of existing


7. Strategic approach to management

management and new technologies.Extension services can play an importantrole in the coordination of diverse stake-holder groups when planning and imple-menting management programs. For effec-tive, goal-orientated management to occur,knowledge of current best practice and newtechnologies has to be available to manage-ment groups, with extension officers havinga central role in knowledge transfer.

7.1 Strategic approach

The four steps that constitute a strategicapproach to the management of wild dogsare: defining the problem; developing a man-agement plan; implementing the plan; andmonitoring and evaluating progress (Figure1). The strategic approach to pest manage-ment incorporates adaptive managementprinciples recommended by Walters (1986).In passive-adaptive management (Waltersand Holling 1990) a single strategy is select-ed, implemented, monitored and evaluated,and adapted according to the success or oth-erwise of the strategy. The active-adaptiveapproach puts up a number of alternativestrategies which are all implemented, moni-tored and evaluated, and adapted accordingto which strategies work best (Walters andHolling 1990). The latter technique is moreexperimental and requires standardisation ofmonitoring and effort across strategies, repli-cation of strategies and, ideally, nil-treatmentareas where no control strategy is imposed.The challenge for managers is to use theinformation in this book, and consideringhow the land is being used, to develop astrategic management plan to address thedamage caused by wild dogs. This chapterexplains how this might be achieved, anddescribes its special features for wild dogs inagricultural and conservation areas. The flowchart in Figure 1 outlines the steps requiredto follow the strategic approach and casestudies from a cattle breeding enterprise innorth Queensland and from sheep grazingareas in north-eastern New South Walesdemonstrate how the strategic approach canbe implemented.

7.2 Defining the problem

The first stage of the strategic approach tomanaging the impacts of dingoes and otherwild dogs is to define the problem at hand.This is more complex than simply measuringthe impact in terms of predation. There aresix components of problem definition(Figure 1) that must be identified for a strate-gic approach. The questions asked are ‘who,what, when, where, why and how much?’Problem definition may be complex wherepredation of livestock occurs in land adjoin-ing conservation areas. Here, a balance isrequired, so it is crucial that the problem isadequately defined before control activitiesbegin.

7.2.1 Agricultural impacts

Wild dogs adversely affect agricultural enter-prises by reducing profits through decreasedyields or increased costs (Section 3.1). Thelosses associated with predation and harass-ment of livestock by wild dogs have beenmeasured on occasions, as have the costsassociated with control activities (NewEngland Rural Development Associationc.1966; Schaefer 1981; Thomson 1984a, 1986;Fleming and Korn 1989; Thompson andFleming 1991). To make wise decisionsabout management it is necessary to estimatethe point at which the costs of wild dog con-trol are equal to or are less than the benefitsof the control (that is, the break-even point).

7.2.2 Human and animal healthimpacts

Wild dogs are known to be associated withdiseases of people and livestock (Section3.3). Reducing the levels of hydatid infectionin endemic zones may be the objective ofwild dog management and the benefit–costratio for this needs quantification. The pres-ence of free-roaming dogs may place thesuccess of a campaign to eradicate hydatidsfrom domestic dogs and livestock in jeop-ardy. In such cases, the presence of wilddogs will need to be incorporated in thehydatid control campaign. Similarly, eradica-tion of rabies or other exotic diseases ofdomestic dogs will rely to some extent onthe management of wild dogs.


Attacks on humans by wild dogs can reducethe number of visitors coming to popularconservation reserves (for example, FraserIsland). The occurrence of attacks needsmeasuring so that performance objectivescan be set. Needless-to-say, zero attacks isthe usual objective.

7.2.3 Conservation

The presence of dingoes has unpriced value(Sinden and Worrell 1979). Various tech-niques can be used to estimate a monetaryvalue for unpriced values. These monetaryequivalents have not been calculated fordingo conservation, but, where the controlof dingoes and other wild dogs in areas ofthe interface of government and grazinglands is deemed necessary, the intrinsic(Sinden and Worrell 1979) and contingentvalues (Wilks 1990) of dingoes should beincluded in the cost–benefit analysis.

Before economic frameworks can be used toassist meeting conservation goals, the valuethe community places on the conservation ofthe dingo, and other native species vulnera-ble to predation by dingoes, should be esti-mated. This would require research. The costand effectiveness of implementing wild dogcontrol techniques to protect conservationvalues also needs to be assessed so that themost cost-effective management strategiesfor meeting community conservation valuescan be determined. This would only be thecase if scientific data verified that controllingwild dogs actually protected conservationvalues, and that the costs of such controlequated with the contingent conservationbenefits.

7.3 Developing a managementplan

7.3.1 Identify management objectives

The implementation of a strategic approachto managing wild dogs requires the clearidentification of the different goals of stake-holders. These include the general public,local communities, welfare groups, land-holders and government agencies. Legal

obligations differ between stakeholdergroups and so do management objectives.The essential element in identifying objec-tives for wild dog management is to involveall the relevant stakeholders so that differentgoals can be raised and debated and com-promises reached.

The final management objectives are a state-ment of planned achievements which areusually agricultural or conservation benefits.It is desirable that all stakeholders agree onthe final objectives. Once the objectives havebeen set, they need to be defined in terms ofoutcomes which can be measured by perfor-mance criteria (Section 7.3.6).

7.3.2 Partnerships

After identifying agreed objectives for wilddog management, the next step is to create agenuine partnership and cooperative actionbetween stakeholders. The impacts of wilddogs must be seen as a community problemto be solved by and for the community; theyare not someone else’s problem to be solvedby the government or the next-door neigh-bour. They are also problems of all land man-agers upon whose land the wild dogs live.Integrated approaches to wild dog manage-ment are only possible through partnershipsbetween government and private partnerswhere mutual respect and inclusion prevail.

Predation of livestock is the problem of gra-ziers and the conservation of dingoes is theresponsibility of government agencies asproxy for the wider community. The involve-ment of the appropriate government agen-cies as stakeholders is crucial when control isproposed near or on public lands.

The involvement of other interest groups suchas animal welfare groups and dingo preserva-tion societies should be encouraged. Thisinvolvement ensures that management prac-tices are responsive to community attitudesand in turn, that the wider community under-stands the limitations that are often inherent inthe management of vertebrate wildlife.Although complete agreement may not bereached between groups, conflict can beavoided by both sides being aware of theother’s position and acknowledging their rightto hold that opinion (Pretty 1994).


The need for cooperative partnerships forthe control of wild dog predation has longbeen recognised and this is reflected throughthe long history of group control schemes.Many have been formalised in legislationand some control activities are not permittedwithout evidence that control is to be under-taken by recognised groups. Cooperativeaction between land managers is an essentialstrategy for effective wild dog managementbecause wild dogs have large home rangesand management needs to be applied syn-chronously to a relatively large area if it is tobe effective (Section 7.3.5). This requires thecollective action by groups of adjoiningproperty owners. The crucial first step is tocreate a genuine partnership and coopera-tive action by land managers, governmentsand others who will benefit from the man-agement or have some other stake in the out-come.

‘Management needs to be applied synchronously to a relatively large area if it is

to be effective.’

These stakeholders come from diverse back-grounds, and may have different responsibil-ities and requirements that may cause con-flict. There is more potential for conflictbetween stakeholders in wild dog manage-ment than for any other vertebrate pestspecies. In many areas, wild dog–livestockproblems occur at the interface of agricultur-al and government lands, and property own-ers may wish to control wild dogs on thegovernment land. However, that may be aconservation reserve where the conservationauthority has a responsibility to conservedingoes, or it may have populations ofthreatened species that are at risk from wilddog control methods. Private property own-ers may also oppose baiting on their proper-ties, limiting the potential for large-scalestrategic baiting programs.

In such cases, management may fail unlessacceptable, alternative control methods can beintegrated into the management plan anddeployed in these areas. Compromises areeasiest to reach when there is a clear and well-articulated statement of each stakeholder’s

position from the beginning of the planningprocess. In the case of government agencies,policy on wild dog management should beformulated through extensive public consulta-tion and be articulated through clear opera-tional guidelines to staff. In this way, eachstakeholder will understand the constraintsunder which others are operating and theactions that may be possible in any given situ-ation. The inability of agency representativesto meet these criteria is one contributor to seri-ous conflict over wild dog management.

One of the major advantages of coordinatedgroup control is that it provides a forumwhere these different interests and responsi-bilities can be aired and compromisesreached. Other advantages of group controlare that it can:

• better integrate a range of control methods into the management plan

• respond more effectively to wild dog predation

• facilitate awareness and peer pressure onthose unconvinced of the need for wilddog management

• make more effective use of resources suchas fencing materials, baits and aircraft

• provide the basis for more effective,long-term management of wild dog pre-dation.

All stakeholders need to be involved at allstages from the beginning of the planningprocess through to the implementation andevaluation of the plan if they are to haveownership of both the problem and its solu-tion. The group may need to collect and pre-sent information necessary to meet the leg-islative and policy requirements of govern-ment agencies. For example, in New SouthWales, aerial baiting can only be carried outby a recognised group such as a Wild DogControl Association (WDCA). Aerial baitingon conservation reserves requires specificapproval subject to the conditions of theThreatened Species Conservation Act 1995,and wild dogs cannot be controlled on con-servation reserves unless National Parks and


Wildlife Service policy is met. In such cir-cumstances there are advantages in estab-lishing the group on some formal basis tofacilitate participation by both governmentand non-government participants.

‘Groups need to develop both strategic and reactive

management plans.’

An example of where a formal approach toplanning and implementing broadscale aeri-al baiting has generally worked well is innorth-eastern New South Wales (Thompsonand Fleming 1991). In a sense, the strategicapproach is facilitated by the regulatoryrequirements of the approval process. Theseforce the various parties to acknowledgetheir different objectives and responsibilities,resulting in the main in an acceptance ofthese differences from which grow respect,compromise and cooperation. The process isalso assisted by the fact that most WDCAs aredecades old and have well established groupdynamics, and by increasing cooperationbetween National Parks and Wildlife Serviceand its neighbouring landholders on bothwild dog control and a range of other issues.

Groups need to develop both strategic andreactive management plans (Section 7.3.4),and to clearly define the roles and responsi-bilities of stakeholders in the implementationof these plans. This is particularly importantif there is to be a rapid and effective manage-ment response to wild dog predation.

7.3.3 Government as stakeholder

Government agencies have a legitimateinterest in managing vertebrate pests, includ-ing the impacts of wild dogs. Predation oflivestock is often associated with landhold-ings adjacent to government lands includingState forests and national parks. In thoseStates and Territories where dingoes are pro-tected, governments are involved throughtheir legislative responsibilities. However,Williams (1993) cautioned that managementby government agencies should not be per-ceived as a subsidy by landholders. Such aperception has been shown to reduce stake-holder participation in rabbit control. In a

small study conducted in the New Englandarea in New South Wales, Schaefer (1981)showed that the cost of predation by wilddogs was capitalised into the value of theland. Therefore, additional subsidies in theform of government funded pest control onprivate land would be difficult to justify.

‘The State should be seen as agood neighbour in its response toimpacts of wild dogs, yet ensure

the conservation of dingoeswhere it must.’

Different government agencies within a Stateor Territory may have potentially conflictingresponsibilities. This emphasises the needfor cooperation between all stakeholders.Management of wild dogs on governmentlands needs to be integrated with manage-ment on adjoining private lands. The Stateshould be seen as a good neighbour in itsresponse to impacts of wild dogs, yet ensurethe conservation of dingoes where it must.Government support for group managementcan increase the probability of achievingsuccessful outcomes. Government-fundedfacilitators and extension specialists can pro-vide expertise wanted by coordinatedgroups (for example, Landcare groups)(Chamala and Mortiss 1990). A program todevelop an integrated approach to the man-agement of wild dogs in the south-east ofNew South Wales, the Australian CapitalTerritory and north-eastern Victoria, whichincorporates the efforts of 14 landholder andgovernment bodies, was only made possiblethrough Commonwealth Government sup-port which allowed the employment of aproject coordinator and trainees (P. Fleming,D. Jenkins Australian Hydatids Control andEpidemiology Program and H. Cathles, YassRural Lands Protection Board, unpublisheddata 1998).

Much of the responsibility for the regulationof pesticides used for vertebrate pest controlis vested in government agencies. Hence,governments should also be responsible forensuring appropriate training of field staffusing the pesticides.


The role of extension services

The strategic approach to managing theimpacts of wild dogs incorporates the adop-tion of a variety of control and evaluationand monitoring techniques. Traditionally,State and local government extension ser-vices have been the conduit for informationtransfer to landholders. Extension officersstill play an important role in assisting theadoption of new and more effective tech-niques by land managers.

In addition to maintaining the flow of newinformation to managers, extension officersoften act as facilitators and coordinators foractivities within and between groups. Giventhe central role of extension people in themanagement of vertebrate pests, their con-tinued training and education is important.Many State government agencies (for exam-ple, Agriculture Western Australia and theQueensland Department of NaturalResources) place emphasis on developingthe extension skills of their staff. Theseorganisations ensure that all field staff withan advisory role receive training, eitherthrough dedicated Technical and FurtherEducation Courses or regular in-servicecourses. Extension training should focus onthe theory and practice of extension as wellas technical knowledge (Appendix B).Officers should be trained in individual andgroup communication skills, including prob-lem and goal identification, conflict resolu-tion, and negotiation. Once the objectives ofthe individual or group being addressed areidentified by the advisory officer, transfer ofknowledge and the implementation of coor-dinated management programs are morelikely.

7.3.4 Select management options

The five options for the management of wilddogs are eradication, strategic control, reac-tive control, conservation or no wild dogcontrol (Figure 1) (Braysher 1993).

There are three criteria that must be metbefore eradication can be considered: (1) atall densities, wild dogs can be killed morequickly than they can breed; (2) there is noimmigration into the controlled area; and (3)all wild dogs are at risk from the control

methods used (Bomford and O’Brien 1995).Eradication of wild dogs is unlikely to be fea-sible except at the local level or where habi-tats have been grossly changed by agricultur-al practices. In much of the wild dog-affect-ed area of north-eastern New South Wales(for example, Wongwibinda Wild DogControl Association area described byFleming (1996a), dog-proof fencing providesa barrier against immigration. Usually, fewwild dogs breach the fence. Any wild dogsthat encroach upon the area enclosed by thefence are pursued until they are killed.Because wild dogs are birth pulse animals, itis usually possible to kill the dogs inside thefence before they breed and thus achievelocal eradication. Wild dogs have been erad-icated from the sheep/wheat belt of south-eastern Australia through widespread controlcampaigns in the 1800s combined withbroad-scale changes to the landscape byclearing for crops. Eradication is undesirablewhere dingo conservation is desired.

Strategic control implies that sufficientknowledge is available to take precautionaryactions to prevent livestock predation or thespread of wild dogs from source areas intointensive grazing areas. Reactive control isconducted in response to predation, obser-vations of increased abundance of wild dogsor observations of wild dogs in sheep areas.Strategic approaches may include reactivecontrol as part of the overall plan.Management for the conservation of dingoesis complex.

Management options

Most management of wild dogs falls into twomain categories: strategic and reactive (Box 2;Table 6). Reactive management can also becategorised into: one-off, where control mea-sures are conducted in response to a specificproblem until the problem is solved; and con-stant, where control options are conducted ona continual basis usually because no strategyto prevent predation has been planned. Thereare a number of management options avail-able to landholders and groups (Chapter 6). Amechanism is required in the planning pro-cess to help decide the best course of action.The simplest approach to this is a decision-making framework comprising a progressive



Table 6: A decision table of strategic (S) and reactive (R) control measures for wild dogs in New South Wales. Similar constructscan be formulated for different States and Territories incorporating appropriate policies and laws.

Control method

Controltype

Appropriate situations Constraints

Baiting Ground Bait lines S

R

R

S

S

R

Bait station

Bait linesAerial(not availablein all states)

Alternative to aerial placement during strategicbaiting

In response to stock losses

Alternative to aerial placement during strategicbaiting. In response tostock lossesHigh risk areas which warrant more continuouscontrolFor agreed strategic controlin areas with a history oflosses

In response to stock losses

Generally less than two kilometres (and nevergreater than five kilometres) inside publicestate area; baits average 100 metres apart.

Generally less than two kilometres (and nevergreater than five kilometres) inside publicestate area; baits average 100 metres apart;(for example, NSW limit of 50 baits per prop-erty per day unless otherwise approved).

Must be near boundary of public estate; 1–2 baits per station

Must be near boundary of public estate; 1–2 baits per station

Helicopter only in eastern New South Wales,fixed-wing aircraft in Western Division.Placement less than two kilometres (and nevergreater than four kilometres) from public estateboundary; baits average 100 metres apart onpublic estate areas, 25 metres elsewhere;Ministerial approval required in NSWHelicopter only in eastern New South Wales,fixed-wing aircraft in Western Division. Placementless than two kilometres (and never greater thanfour kilometres) from public estate boundary; baits average 100 metres apart on public estate areas,25 metres elsewhere; Ministerial approvalrequired in NSW – rarely possible because oftime required to obtain approval

R

S

In response to stock losseswhere a small number ofdogs and their routes areknownLong-term populationreduction in small areas

Requires an experienced trapper; licencerequired in some areas

Requires an experienced trapper; licencerequired in some areas, labour intensive

Trapping

S At edges of sheep country Expensive to construct; continued maintenancerequired

Barrierfencing

R In response to stock losseswhen other methods havefailed; may also be usefulfor individual problemdogs

Most successful when an experienced operatoris used

Shooting

series of decision points (Box 2). Normallysuch an approach requires that there be onlyone criterion (question) for each decision.The criteria which apply at each decisionpoint are listed below. When a managementaction results, the form of that action is shownin the diagram, and the information needed toimplement the action can be found by refer-ring back to the control methods (Section6.4).

An alternative aid to making decisions aboutthe most appropriate control techniques toexert in a particular situation is a decisionmatrix (Caughley et al. 1998; Norton 1988).To construct a decision matrix, the knownalternative control options are listed in thefirst column and questions relating to techni-cal feasibility, likely success, economic desir-ability, environmental safety, and politicaland social acceptance are listed across thefirst row. Each question is answered for eachcontrol method with a ‘yes’, ‘no’ or questionmark. The most desirable option has themost ‘yes’ answers, provided that themethod works or is expected to work.

7.3.5 Develop management strategy

Once the management option has been decid-ed, a strategy is needed to achieve it. Strategiesfor the management of dingoes and other wilddogs prescribe what is to be done, when it is tobe done, in what order and how often it isdone and who does what. The strategy alsoprescribes how selected control techniqueswill be integrated. Table 6 and Box 2 can beused to assist in deciding what actions arerequired to achieve the selected managementoption (Section 7.3.4). For example, if theselected option is strategic control with reac-tive control when necessary, a buffer zone orbaited zone strategy may be chosen. Thebuffer might be achieved primarily by an exist-ing exclusion fencing, with seasonal groundbaiting programs to reduce abundance outsidethe fence and to establish the buffer. The strate-gy will therefore include identification of whomaintains and checks the fence to ensure that itis dog proof, who conducts the ground baitingprograms, the frequency at which the fence ischecked and the baitings occur, and the moni-toring regime. Reactive control might be target-ed trapping when an incursion by wild dogs

into the area protected by the fence is detectedin the regular monitoring program. Whoundertakes the reactive trapping and who paysis also decided beforehand and the owners ofprivate and public lands or their employees orthird parties such as contractors and controlassociations are selected and allocated theirtasks and responsibilities.

The management strategy also defines thesize of the treatment area. Plans to managewild dogs must be for defined areas of landand can be at any scale — national, State,regional, district or property. Managementplanning requires determining the right sizefor a management unit. Selection of an appro-priate size management unit will often be atrade-off between several factors, such as:

• Risk of reinvasion if the management unitis small — dingo home ranges may crossproperty boundaries and dispersing din-goes can travel long distances (Section2.4) so it is often necessary to conductwild dog control over larger areas such asseveral neighbouring properties or onproperties plus adjacent governmentland.

• Economies of scale for management oflarger units — for example, some expen-sive dog control techniques such asexclusion fencing (Section 6.4.1) and aeri-al baiting (Section 6.4.4) are best under-taken over large ares because the costs toindividual properties are lessened.

• Commonality of topographic area andcommunity of interest — the advantages ofsmall community groups working togetherto solve a problem in a local area (Section7.3.2) may be lost if the chosen manage-ment area is too large or extends beyond amanageable topographic unit or commoncommunity of interest. Stakeholders loseinterest if a project appears too large tosolve or is outside their area of familiarity(that is, it is ‘someone else’s problem’).

The management strategy also defines thetimeframe for management. The timeframe isset according to the management option(s)chosen. For instance, reactive control canoccur immediately if the problem is current



A decision-making framework comprising a progressive series of unambiguous decision pointscan help landholders and groups to decide what course of action to take in response to historicaland present problems with wild dog predation. This framework may assist managers who areable to monitor livestock regularly and may not apply in all situations. Normally this approachrequires that there be only one criterion (question) for each decision. To simplify the representa-tion of the framework, in some cases criteria have been grouped at appropriate decision points.To prevent ambiguity in these cases, the groupings are such that a ‘YES’ decision requires thatALL criteria have a ‘YES’ answer. Thus a single ‘NO’ results in a ‘NO’ decision. The result at eachdecision point leads to either a further decision point or to a management action. The flowchartin Figure 14 will help users work through the framework.

Using the decision-making framework

This framework works like a botanical key. Start at 1 and work through until a ‘YES’ decisionis reached. A ‘YES’ decision requires ALL criteria within the question to be answered YES.

1. Present losses?

Criteria Stock losses are presently occurring

These are caused by wild dogs from within the property and/or adjacentlands. (This is relevant to determine whether other agencies and neigh-bours should be involved in management. Losses are unlikely to becaused by wild dogs more than one home range width from such adjacentlands [about ten kilometres in south-eastern Australia])

If NO Go to 2 (Strategic management)

If YES Go to 1a

1a. Barrier fence?

Criteria Barrier fence separates wild dogs and stock

If NO Take actions to stop stock losses (go to 3)

If YES Go to 1b

1b. Fence secure?

Criteria The fence is wild-dog proof

If NO Repair the fence THEN take action to stop losses (go to 3)

If YES Take action to stop losses (go to 3)

2. Past losses? (Strategic management)

Criteria Stock have been lost in the past two years

These are caused by wild dogs from within the property and/or adjacentgovernment lands. (Losses are unlikely to be caused by wild dogs morethan one home range width from such adjacent lands (say 10 kilometres insouth-eastern Australia))

If NO No further actions or decisions are necessary

If YES Go to 3

Box 2: A decision-making framework for wild dog control


3. Strategic management?

Criteria Losses are of sufficient size and frequency to justify management

There is a high risk of future stock losses

Losses will be reduced by strategic management on the property and/oradjacent lands

If NO No further actions or decisions are necessary

If YES Go to 4

4. Constant management?

Criteria Losses occur at any time throughout the year

Intermittent forms of strategic control have failed

The severity of losses justifies the high cost of continuous strategic management

If NO Go to 5

If YES Go to 4a

4a. Erect barrier fence?

Criteria It is practical to erect a barrier fence (consider electrification)

The fence can be built on or very near to relevant boundary

The location of the fence on the Service boundary will prevent the wilddogs responsible for losses from reaching stock

Capital funds are available to erect the fence

If NO Go to 4b

If YES Erect the fence

4b. Employ dogger?

Criteria Trapping is consistent with agency policy

Employing doggers is consistent with agency policy

A skilled and reliable dogger is available

Recurrent funds are available

If NO The problem should be rethought from 4

If YES Employ dogger

5. Wild dog movements known?

Criteria The movements of wild dogs in relation to stock losses is sufficientlyunderstood to effectively and efficiently deploy intermittent strategic orreactive management

If NO Collect movement information THEN go to 6

If YES Go to 6


6. Baiting allowed?

Criteria Baiting is possible and consistent with government regulations

Baiting is consistent with agency policy

If NO Go to 8

If YES Go to 6a

6a. Ground baiting possible?

Criteria The area is accessible by a ground vehicle

Only a relatively small amount of bait is required in a small area

If NO Go to 7

If YES Spread baits from a ground vehicle

7. Aerial baiting

Criteria The area is in rugged country with steep ridges and gullies or the area is interrain that is difficult to access because of distance

A large quantity of bait is required over a large area

If NO Go to 8

If YES Spread baits by helicopter in mountainous terrain (for example, easternNew South Wales). Spread baits by fixed-wing aircraft in extensive, flatterterrain. Adhere to ministerial requirements if needed

8. Trapping allowed?

Criteria Trapping is allowed under government and agency policy

There is a skilled and reliable dogger available when required

If NO The problem should be rethought from 6

If YES Strategic trapping should be used when it is necessary, employ a dogger


1Presentlosses ?

2Past

losses ?

3Strategicmgmt. ?

4bEmploydogger?

4Constantmgmt. ?

4a Erectbarrierfence?

6Baiting

allowed?

8Trappingallowed?

6a Hand

baitingpossible?

1aBarrierfence ?

1bFence

secure ?

No action

No action

RETHINK Strategictrapping

Aerial bait

Erect barrierfence

Strategic trapping,bait stations

RETHINK

Move stock. Hand bait, trap,

shoot

Repair

Hand bait

7Aerial baiting

possible?

START

STRATEGIC

REACTIVE

YES YES

YES

YES

YES

YES

YES

YESYES

YES

YES

YES

YES

YES

NO

NO

NONO

NO

NO

NO

NO

NO

NO

NO

NO

NO 5 Dogmovements

known?

Determinemovements

Figure 14: A decision-making framework for devising a plan of management for reducing predation of livestock by dingoes and other wild dogs in eastern Australia.

or can be budgeted for against the contin-gency of a future increase in predation.Strategic management can be ongoing, regu-lar, or occasional and may be implementedimmediately or according to a plannedschedule. Short and long-term timeframesand achievement dates are defined in themanagement strategy.

7.3.6 Define performance criteria

Performance criteria are a list of measurablefactors which will be used to determine if themanagement objectives are met once the man-agement strategy is implemented. Becauserelationships between wild dog density andthe damage they cause are often complex andvariable (Box 3, Step 3), it is important that themanagement plan includes a monitoring com-ponent. Monitoring will allow the results ofstrategic action to be compared with the situa-tion preceding the action and with historicaldata, or with sites with no management orwith different management strategies.

Agricultural performance criteria

Agricultural performance criteria are usuallydefined in terms of reduced losses or increasesin production. For example, where calving per-centages are consistently 85% in a region andthe dog-affected properties in question yield80%, a performance criterion might be toincrease calving percentages to the regionalaverage over the next five years. In this case,the time frame is set long enough to encompassa range of expected seasonal conditions andtheir likely effect on fecundity in the herd.However, setting quantifiable and realistic per-formance criteria for management is complex.For example, the scale of the problem and thecontrol effort that is to be applied need consid-eration. The large home range of wild dogs inrelation to the stocking rate and size of sheepproperties in south-eastern Australia means thatmanagement groups rather than individualsmust decide on objectives. Conversely, largecattle stations in central and northernQueensland and the Northern Territory maycover many wild dog home ranges so the scaleof an effective control program might be oneproperty – although immigration of dogs into a‘sink’ property may actually result in increasedimpact (Section 3.1). The performance criterion


here might be to reduce calf losses to a prede-termined break-even level, which is set by thearea of the property, the value and cashflow ofthe enterprise, and the expected level of controlper unit of control effort. Economies of scaleare important and the economics of group con-trol are more complex than for individuals.

If conventional economic analyses cannot beused to account for intangible values andother complexities it may be best to provide amanagement group with as much informationas possible and let them decide what changein yield or loss should be set. The managersthen apply a level of control that experience orintuition predicts will lead to the productiongoal. Monitoring and evaluation (Section 7.6)will allow the adaptive management processto set new levels of control to achieve thegoals. This process allows the unpredictabilityof predation and the landholders’ attitude torisk to be included in the decision-makingprocess and for the plan to progress.

Conservation performance criteria

The setting of conservation performance cri-teria is even more complex than those ofagricultural production. Unpriced valuationsare required to take into consideration thetourism potential and inherent/contingentvalue of dingoes to Australians. It is impor-tant to assess whether a level of hybridisa-tion is acceptable to the general public. It isunlikely that the majority of people wouldknowingly pay to visit national parks to see aferal dog. Once such decisions are reached,performance criteria can be set. An exampleof a performance criterion for dingo conser-vation might be that the proportion of puredingoes in a population in a defined arearemains above a pre-defined percentage forthe next ten years.


Simple stepwise approach

Land managers who wish to determine theoptimal economic strategy for managing aproblem caused by wild dogs could usethe following stepwise approach (modi-fied from Bomford et al. 1995).

Step 1. Desired outcomes

Identify desired outcomes and estimate adollar value for each of these. Where out-comes are the protection of commodities(Section 3.1), such as reduced wild dogattacks on livestock, this should be reason-ably easy. Where outcomes are difficult tomeasure or intangible, such as the protec-tion of biodiversity or threatened nativespecies (Section 3.6), land managers maybe obliged to estimate how much theyconsider is an acceptable amount to spendto achieve that outcome.

Step 2. Control options

List all control options and how much theywould cost to implement (Section 7.5).Control options can be different techniquesor combinations of techniques, or differentlevels or frequencies of application of tech-niques. It is important that the options forcontrol are expressed as activities that amanager can select either to do or not to do.

Step 3. Density–damage relationships

Estimate the relationship between wild dogdensity and damage for each resource dam-aged by dogs (Figure 15). Such relationshipsallow managers to estimate the likely benefitfor a given control effort. For example, if a

control program reduces the density of wilddogs by 50%, how much will this reduce theincidence of attacks on calves? These rela-tionships may be complex or there may bemultiple relationships. For example, if wilddogs switch to different prey and/or changefrom hunting singly to hunting in packs, theslope or position of the line relating densityand damage may shift at the density wherethis occurs. Also, young inexperienced dogsmay inflict a different level of damage thanan equivalent density of experienced olderdogs. There is anecdotal evidence thatyoung dogs may be more likely to attacklivestock than older dogs which have theskills to prey successfully on kangaroos(Macropus spp.) (Section 3.6; Allen andGonzalez 1998). Surplus killing and changesin hunting strategy will also affect the rela-tionship.

In Figure 16 there is a significant linear rela-tionship between wild dog density and dam-age to cattle but there is high level of scatterof the data points around the fitted line. Thismeans that, although the relationship is lin-ear, the line is a poor predictor of the likelydamage caused by a given density of wilddogs. The poor fit of the line is partlybecause some individual dogs cause farmore damage than others (Section 3.1) andjust a few dogs may inflict high levels ofdamage on some properties. This is particu-larly true for sheep properties where surpluskilling by wild dogs occurs (Section 2.3.4). Inother areas, the density of older dogs in sta-ble packs might be quite high and yet attackson calves may be minimal (Allen andGonzalez 2000).

7.4 Economic frameworks

Economic frameworks are needed to assistmanagers in assessing the relative value ofalternative control strategies and the relativebenefits compared with other risks that mustbe managed (Hone 1994). Such frameworksrequire: definition of the economic problem;data on the relative costs and benefits of differ-ent management strategies; an understanding

of why the actions of individual land man-agers may not lead to optimal levels of control;and assessment of the means by which gov-ernments might intervene to overcome identi-fied market failures. Land managers can usesuch economic frameworks to select the mostappropriate management strategy for their cir-cumstances.

Box 3: Economic framework for wild dog management


Hence managers may need to collect andassess their own data on dog abundance(Section 6.2) and establish which dogs(mature dogs with established territories, ornewly arrived young dogs) do the damage.This information, together with records ofdamage levels, will enable managers todetermine density–damage relationships fortheir own circumstances.

Step 4. Control efficiency

Estimate the control efficiency of each con-trol option. That is, how much will a giveneffort using a particular control optionreduce wild dog damage? Most current con-trol programs inherently assume that thedensity–damage relationship for wild dogpredation of livestock is a simple linear func-tion (Line C in Fig 15) and that a given con-trol effort will result in a commensuratereduction in wild dog abundance and dam-age. Clearly, this is not always the case.

Step 5. Benefit–cost relationships

Use the information from Steps 1-4 to esti-mate costs and benefits of implementingeach control option, including optionswhich combine more than one technique,or the option of implementing no dog con-trol (Table 7; Section 7.6.2). Costs will bethe cost of implementing each controloption, and should include costs of moni-toring pests and planning. Benefits will bethe value of the reduction in damage to thevalued resource caused by implementingcontrol (that is the desired outcomes listedunder Step 1 above), plus any profits (forexample, profits made from tourists whopay to look at pure dingoes if a decision ismade to conserve them).

Different pest management options willgenerate a variety of benefit–cost relation-ships. Estimates of benefits and costs canbe discounted back to net present values(usually using a discount rate equivalent tothe interest rate the landholder pays on

Dam

age

Wild dog density

ABC

Figure 15: Some hypothetical relationships betweendog density and damage. Density–damage relationshipsare highly variable for wild dogs and there may be mul-tiple relationships. Just a few dogs might inflict high lev-els of damage on some sheep properties but as the den-sity of wild dogs increases, damage levels off, resultingin line A. On some cattle properties, damage levelsmight be low or nil when dogs are in low densities andare individually hunting for small prey, but if dog num-bers build up and they start to hunt in packs for calves,damage might increase sharply (line B). Line C wouldoccur when an incremental increase in the density ofwild dogs resulted in a proportional increase in preda-tion losses (Figure 16).

Mon

thly

loss

es ($

per

pro

perty

)

300

350

400

250

200

150

100

50

0.50 1.0 1.5 2.0 2.5 3.0

Monthly density index (sign)

Cattle-only enterprises

Figure 16: The relationship between density of wilddogs and the damage caused by wild dogs to cattleenterprises in north-eastern New South Wales. Theindex of wild dog density relied on observations oftheir sign and hence losses were sometimes sustainedat ‘zero’ density. Data from Fleming and Nicol(unpublished data 1999).


financing the control operation). This willreduce the value of costs and benefitsaccruing in the distant future relative tothose accruing in the near future.

Step 6. Marginal analysis

Plot both the incremental change in thecost of wild dog control (marginal costs ofcontrol) and the incremental change in thecost of damage (marginal costs of damage)caused by wild dogs against the level ofcontrol activity contemplated (Figure 17).For example, marginal costs might bechanges in the cost of finding and remov-ing an extra wild dog that occur as wilddog density is reduced. An example ofmarginal benefits could be changes to live-stock losses that occur as wild dog densityis reduced. Where the two lines cross istheoretically the optimal level of pest con-trol (Hone 1994). Further increases in con-trol activity will not cause commensuratereductions in damage, so at higher levelsof control beyond this point, costs willexceed savings in reduced damage.

The problem for managers is that theyoften do not have good information on thedensity–damage relationship, or this rela-tionship is highly variable. This makes ithard to estimate the position of themarginal benefits line which in turn meansthe optimal control point is hard to estab-lish. Even if managers can make a goodguess on an optimal wild dog density toaim for, it is not usually practical with mostcontrol techniques to simply cut off controlefforts at some pre-determined wild dogdensity. It is preferable to have a range ofcontrol strategies ranked along the x-axis,For example, different frequencies of trap-ping could be put along the x-axis. Theassociated cost and benefit values forimplementation would be plotted for eachstrategy, so a manager could then selectwhich strategy (level of control) is optimal,for example, the optimum trapping fre-quency.

Step 7. Pay-off matrices

Construct a table listing all the controloptions and their associated costs and ben-efits (economists call this a pay-off matrix).For example, the costs and benefits of nodog control, ground baiting at waterholes,and widescale aerial baiting with 1080(sodium fluroacetate) could be compared.Managers may wish to construct differentmatrices for different conditions, such asdifferent stocking densities, seasonal con-ditions, or commodity values for wool,lambs or calves. Managers will also need toconsider timescales when constructingthese matrices: what time span is coveredand how will this affect costs and benefits?

These matrices can then be used to selectthe option(s) which best meet the man-agers’ goals. If the manager is risk averse,the best options will be those that bring inreasonable returns (benefits in relation tocosts) under the widest range of conditions(that is, in most seasons and with a widerange of commodity prices). If the manag-er’s priority is to maximise profit, the pre-ferred options will be those that are likely

Control effortHighdogdensity

$

Lowdogdensity

Marginal costof control

Marginal benefitsof control

Figure 17: A marginal analysis of wild dog control.The marginal costs are the costs of incrementallyincreasing the level of control effort. The marginalbenefits are incremental changes in the level of dam-age caused by wild dogs at different levels of controleffort. The optimal level of control is where the twolines cross. Units on the x-axis are level of controleffort (for example, trapping frequency) not wild dogdensity.


to give the highest returns on investment,even though there may be some risk ofhaving no returns or even a loss if the sea-sons and prices go badly.

Pay-off matrices can also be used by a landmanager to compare returns on investmentin pest control with returns on using themoney for some other purpose, such ascapital investments, increasing stockingrates and pasture improvement.

Steps 1–7 complete the basic model. Oneway of improving accuracy may be toreplace single estimates with a range ofpossible values, and give associated proba-bilities for each value in the range.

The model can also be made more accu-rate by adding additional features.Incorporation of such additional featureswill make the model more complex, butincluding at least some of them may benecessary to make it accurate enough to beuseful:

• Social benefits could be included inStep 1, such as:

- off-site effects and good neighbourrelations — (Sections 7.3.2 and 7.3.3).

- conservation of pure dingoes(Sections 3.3, 4.1 and 4.4).

- animal welfare management — animalwelfare organisations would like the suf-fering caused by harvesting or controltechniques considered as a cost that istaken into account in vertebrate pestmanagement decisions (Choquenot etal. 1996: Section 4.2). Alternatively, wilddog control can alleviate the sufferingcaused to prey, reducing the welfarecosts to the community.

• Risk management for spread of diseaseby wild dogs could also be included inStep 1.

• Effects of government interventioncould affect value of benefits (in Step1) or costs (in Step 2).

• Indirect effects of pest control (for exam-ple, controlling wild dogs may lead to anincrease in kangaroo numbers) could beincluded as interaction effects in Step 3.

• The form in which benefits come maybe significant to a manager (Step 5).For example, cash ‘bonuses’ for dingoscalps may be more attractive as imme-diate cash for spending, than futuremoney from increased lambing per-centages, which may be committed inadvance to servicing debts or meetingfarm running costs.

Ideally, land managers could use this step-wise approach to optimise the control effort,but often capital is constrained by compet-ing demands and sub-optimal amounts areavailable. The processes outlined are sensi-ble and relatively simple ways of balancingcompeting demands. We recognise thatmanagers will have incomplete knowledgeof the information necessary to fully com-plete many of these steps. Some projectsfunded by the Vertebrate Pest Program andthe National Feral Animal Control Programin the Bureau of Rural Sciences aim to col-lect some of these data. Even where infor-mation is incomplete, the exercise ofattempting to complete the process, andrecording the assumptions and best guessestimates that are made, may prove a usefulaid to decision making for wild dog anddingo management.

Even where sufficient information exists toenable simple marginal or other economicanalyses of different dog control options,valuing the reductions in yield caused bywild dog predation is not straightforward.For example, the economic cost of preda-tion of a merino ewe can be simply esti-mated by the potential value of its wooland its replacement cost, yet valuing thecontingent loss of its cumulative geneticvalue is not simple. This is particularly sowhen the sheep involved are from a studor commercial breeding enterprise. Morecomplex economic models are needed toinclude the intangibles of predation onsheep production.

By contrasting the marginal ratios of differentapproaches on a common measure of bene-fit, for example, decreasing loss of biodiver-sity, or increasing farm income, break evenpoints can be established and the allocationof resources can be optimised. In such cases,


managers have to prioritise where controlwill be conducted. Accurate information tosupport many of the decisions needed forthis process will almost always be absent andmanagers will often have to make ‘bestguess’ estimates. However, the process willgive defensible decisions especially if theyare empirically tested by monitoring the out-comes. For example, sheep graziers mightneed to estimate the losses caused by wilddogs, both immediately, through predationon livestock, and longer term, throughreducing reproductive success and geneticadvancement of the flock or herd. In thiscase, future losses would need to be dis-counted at some appropriate rate (commonly5% per year). The costs of control would alsoneed to be assessed, examining differentcontrol strategies to see which are cheapestand most effective. Alternative options andopportunity costs would also need to beexamined.

More complex models

A more complex analysis of the economicsof wild dog management is possiblethrough linear programming. In linear pro-gramming, the underlying function (in ourcase the relationship between damage andpest density) and its constraints areassumed or known to be linear(Luenberger 1984). By fitting known valuesto the function and the constraints, andsolving the equations, control is optimisedin terms of cost. For example, in a regionallinear programming model of serrated tus-sock (Nassella trichotoma) control in NewSouth Wales, Jones and Vere (1998) max-imised the regional gross margin of controlof this weed within the constraints of soilfertility and rainfall. Their model incorpo-rated a linear relationship between the den-sity of serrated tussock and the reduction inwool production and carrying capacity,and subject to an array of rainfall regimesand soil fertilities. Linear programming isnot applicable if the functions are not linearand is limited to short-term evaluations.This process is probably most useful at thepolicy-making and program-funding level.

A problem with marginal analysis and withlinear programming is that they are simplis-

tic and do not include pest populationdynamics. The inclusion of populationdynamics in weed control models hasshown that control should be instigated atlower thresholds than predicted by the sim-ple threshold models often used in marginalanalyses (Bauer and Mortensen 1992).Stochastic dynamic programming (SDP)models attempt to link biological and eco-nomic data in such a way that accounts forthe variability of the biological components.One advantage of these models is that thepopulation dynamics of the subject pest canbe incorporated. Without these compo-nents, models underestimate the long-termcosts of not taking control action. Ignoringthe rate of increase of a population resultsin a lower level of instantaneous controlthan is optimal. This decreases the long-term economic benefits of control andresults in greater long-term costs of control(R. Jones, senior research economist, NSWAgriculture, pers. comm., 1999). As with lin-ear programming, SDP is complex and bestused by experts to evaluate managementstrategies and to aid policy and fundingdecisions. An example of SDP is a CD-ROMrabbit control simulation model which con-siders interactions between rabbit numbers,pasture production and sheep flock perfor-mance accounting for climatic stochasticity(Bureau of Rural Sciences/Centre forAgricultural and Regional Economics 1999).The model is based on a New South WalesTablelands grazing system and takes intoaccount uncertainty in the wool productionsystem, rabbit population dynamics and theeffectiveness of rabbit control options.Numerous simulations can be run to showthe expected or average improvement infarm gross margin resulting from rabbit con-trol across a range of environmental condi-tions. Rabbit control will not always beprofitable because competition with sheepfor pasture will not always be a problem. ASDP model for wild dog managementwould have greater uncertainty than that forrabbit control because of the much highervariability in density–damage relationshipsfor wild dog impact on sheep production.


7.5 Implementation

When the management plan is completed,implementation can start. It is usually desirableto involve as many stakeholders as possible inthe implementation stage. The value of thegroup approach in the implementation of apest management plan (Section 7.3.2) hasbeen discussed in detail in the earlier guide-lines for managing rabbits (Williams et al.1995) and feral pigs (Choquenot et al. 1996).

Maintaining commitment and enthusiasmamong all stakeholders is essential. Good com-munication between all participants through-out the implementation of the managementplan will ensure that difficulties are identifiedand addressed early, and if necessary supportor peer group pressure can be provided ifsome people have problems implementingtheir part of the plan (Section 7.3.2). Rapidcommunication of the results of the monitoringof performance indicators is also important sopeople see the rewards of their efforts whichwill motivate them to continue. If monitoringshows that expected benefits are not occurringthis also needs to be communicated rapidly sothe stakeholders can decide if the objectives orthe management plan need to be modified(Figure 1).

7.6 Monitoring and evaluation

Monitoring enables the continuing refinementof a control strategy in relation to the set objec-tives. It is important to distinguish betweenefficiency (operational objectives) and effec-tiveness (performance objectives) as manage-ment can be efficient but ineffectual. For exam-ple, 75% of wild dogs might be killed efficient-ly for little cost, but this strategy would fail ifthere was no concurrent reduction in preda-tion or increase in production.

A management plan must have operationalobjectives (for example, ‘was the plannedmanagement action carried out efficiently?’)and performance objectives (‘did the manage-ment action achieve the resource protectiongoals used to justify management action?’; thatis, ‘were the performance criteria met?’).

7.6.1 Operational monitoring

Records need to be maintained describingwhat was done, how many wild dogs were

killed or what proportional reduction of abun-dance index was achieved (Section 6.2),where, for how long, by whom, and at whatcost. Measurements need to be taken andreported routinely. Reports should includedetails of number of properties treated peryear, money allocated, total number of wilddogs killed, number of wild dogs killed perunit effort, number of wild dogs remaining, thecost per unit reduction and mapping of data.

The results of operational monitoring will beused to assess the efficiency of the manage-ment strategy. They will be useful for answer-ing such questions as:

• Were the best people chosen to under-take the fieldwork?

• Were the selected techniques applied atthe best times and did they reduce wilddog densities to desired levels?

• What were the costs? Were there costoverruns? Can budgets be reallocated toreduce costs?

7.6.2 Performance monitoring

Performance monitoring measures the effect ofmanagement of the resources to be protected,by comparing the outcome of managementagainst performance criteria. For example,‘how well have we achieved our goals?; or,‘have we increased our calving percentage tothe regional average?’. To answer these typesof question, production records need to bekept and compared with the values precedingimplementation of the management plan. Mostperformance measures can be assessed bylandowners or pest managers. Measures ofmore complex ecological relationships mayrequire interpretation by experiencedresearchers. Performance monitoring usuallyrequires a long-term perspective, and someexperimental and scientific rigour for results tobe interpreted and transferable.

The results of the performance monitoringwill be used to assess the effectiveness of themanagement strategy against the objectivesdefined in the management plan. They willbe useful for answering such questions as:

• Were some or all of the objectives met?Why or why not?


• Were the objectives met within the time-frame? Should the timeframe be changed?

• Should the problem definition or manage-ment objectives be changed (Figure 1)?

7.7 Case studies of the strategicplanning process

These case studies are to demonstrate strate-gic planning methods. The first case study isfrom an extensive cattle enterprise in northQueensland. The authors were unable tofind suitable data to provide a case study for

This case study draws on real data from anextensive cattle enterprise in northQueensland (from Allen et al. (1997); Allenand Gonzalez (1998)). Although the enter-prise did not undertake a formal planningprocess, the data are used here to demon-strate the strategic approach to wild dogmanagement. The specific objectives andassessments against those objectives werehypothetical but the responses to controlare real.

The site

The site is a 52 000 hectare property in thewet–dry tropics near Cape York in northernQueensland which has an annual rainfall ofabout 900 millimetres. The property is anextensive breeding cattle enterprise. Feralhorses, kangaroos and wallabies(Macropus spp.) have been controlled onthe property.

Defining the problem

From 1968 to 1988, branding percentageson the property were between 42% and70% and of these, the percentage of calvesthat had been bitten by wild dogs rangedfrom 8% to 19%. The property was baited atleast annually with strychnine baits laidaround waterholes from 1968 to 1988. Overthat time, because of pasture improvementand better management, there was a gradu-al increase in the number of cattle and in

the number of calves branded. The brand-ing percentage and the percentage ofcalves bitten by wild dogs both remainedfairly constant but the potential brandingpercentage for the region was not beingreached.

Objectives

Although specific production goals werenot set, the general objectives over the 20years prior to 1988 had been to maximisebranding percentage and income. For thepurpose of this example, the specific goalsfrom 1988 were to:

• increase the average branding percent-age to the potential levels for theregion (80%)

• reduce the number of calves bitten bywild dogs to below 2%.

Management options

There were three options in 1988:

• The ‘no dog control’ (Section 7.3.4)option was not appealing because bit-ten calves were strong evidence thatpredation by wild dogs was significant.

• Although there had been a steadyincrease in the number of calves brand-ed during the existing strychninebaiting program based at waterholes,

7.7.1 Case study 1 — Extensive cattle enterprise in north Queensland

the sheep industry. Instead, a hypotheticalcase study of a benefit–cost analysis hasbeen provided.

In the cattle case study, there has been anongoing problem of predation by wild dogsand below average livestock production. Amajor change from individual-based controlprograms to coordinated group activitiesoccurred, and records of production andlosses before and after the change enablesmonitoring of the progress towards objec-tives. The program has run over more thanten years.


improved general management hadincreased the number of breedersbeing run. This confounded the mea-surement of the effectiveness of thebaiting program. However, as therehad been a steady increase in calfturnoff, the baiting program mighthave been at least partly responsible.

• In 1988, cooperative aerial baitingusing 1080 baits was being encour-aged by neighbours and governmentinspectors in the region and else-where. This alternative was attractivebecause inaccessible areas of the prop-erty could be baited and beinginvolved in a joint community programwas regarded as a responsibility to thecommunity. Although aerial baitingwas expensive, the spectacularincreases in returns reported on thebush telegraph would more than coverthe expense.

Implementation

In cooperation with neighbouring cattlestations, approximately 50 000 square kilo-metres were baited in 1988 using a fixed-wing aircraft and 1080 meat baits. Theproblem and its solution were ‘owned’ bythe landholders involved and the govern-ment’s contribution was through theRegional Inspector (from the then RuralLands Protection Board) who was respon-sible for adding poison to the baits andadvising and coordinating the group.

Monitoring, evaluation and outcomes

Records of rainfall, calving percentages (atbranding), bites to calves by wild dogs andwild dog control activities have been keptfrom 1968 to the present. At the 1989branding, the number of bitten calvesdropped to zero and the highest-everbranding percentage to date (approxi-mately 75%) was achieved. In the eightyears following the first aerial baiting,branding percentages have averaged75.3% (s.d. = 1.2%) and have not been

below 70%. The percentage of bitten cattlewas less than 1.2%. If the goals had been toreach the potential branding percentagefor the region, the program has fallenshort; however, large increases have beenachieved and 320 additional calves havebeen produced per year. The baiting pro-gram did achieve the objective of reducingthe percentage of bitten calves to below2%.

The stocking rate of the property and theeffect of greater calf survival will need tobe assessed over the next few years todetermine if the extra calves are drawingon the pasture capital of the station.Monitoring and evaluation should contin-ue and alternative strategies be planned incase there is a trend towards decreasingeffectiveness.

Authors’ comment

Although there was no equivalent areawhere wild dogs were not controlledagainst which to compare these results, thecircumstantial evidence for the success ofthe aerial baiting strategy was strong andcertainly strong enough to convince mostlandholders of its value. Conversely, studiesundertaken in other smaller areas of centralQueensland (800 to 9000 square kilometresbaited) have shown that, although wild dognumbers were reduced, there was no con-current reduction in calf losses and in somecases calf losses increased (Allen andGonzalez 1998). In response to those find-ings, the manager of one of the studiedproperties in central Queensland has notbaited since 1997 (Allen and Gonzalez2000). Scale differences, behavioural andage structure changes in baited populationsof wild dogs, differences in repopulation,the level of population reduction or combi-nations of these (Allen and Gonzalez 1998)may explain the contrasting results betweenthe northern property and those wheresmaller areas were baited. Allen andGonzalez (2000) recommended that for fullvalue to accrue from aerial baiting, coopera-tive projects covering large contiguousareas are preferable to smaller, single orpart-property efforts.


To undertake a benefit–cost analysis, data onthe value of production, cost of wild dog con-trol and control effectiveness are required. Anexample of a benefit–cost analysis for wild dogcontrol on sheep properties is shown in Table7. In the table, hypothetical values are givenfor production parameters, wild dog controlcosts and sheep losses for two wild dog con-trol strategies:

• Wild Dog Control Strategy 1: trapping,reactive ground baiting, fence mainte-nance and opportunistic shooting; nobaited zone.

• Wild Dog Control Strategy 2: baitedzone (applied by aerial baiting) aswell as fence maintenance, oppor-tunistic shooting and reactive groundbaiting.

The benefit of these two strategies inreducing wild dog impact is determined bycomparing them against two hypotheticalpercentage stock losses (16% and 33%)when no wild dog control was applied.

Process

In this hypothetical case, it was assumed thatthe relationship between damage and thedensity of wild dogs is linear and hence therelationship between benefit–cost ratios andsheep losses is also linear (Figure 15, Line C).

To estimate the break-even points (thepoints at which the benefits equalled thecosts), linear functions were drawn for bene-fit–cost ratios at different percentage lossesfor the two strategies (Figure 18). Using theequations of the lines, the break-even pointswere estimated by determining when thebenefit–cost ratio (y-axis) equalled one.

In general, it is better to use long-termaverage losses than single year figuresbecause of the large variability in preda-tion that is experienced between years.

Analysis

The hypothetical benefit–cost ratios werebetter for Strategy 2 than Strategy 1 (Table 7,Figure 18) because the percentage predationexperienced and the relative costs werelower in Strategy 2. The control effort in bothstrategies was beneficial when compared

with the hypothetical percentage losseswhere no control effort was expended. Thebenefit was greater when the imposition ofthe baiting zone was the primary controlstrategy. For Strategy 1 (no baited zone),break-even would have been achieved ifexpected losses were 9.6%, whereas Strategy2 (with baited zone) was beneficial if expect-ed losses were above 4.0%. Aerial baiting isalso likely to have a greater impact on wilddog populations, but targeted baiting/trap-ping may be more effective in removing indi-vidual dogs which are causing problems inlocalised areas.

The benefit–cost analysis is clearly depen-dent on the per head value of the livestockand the potential activity — that is, the fre-quency of attacks and number of sheepmaimed or killed during attacks — of indi-vidual dogs that may be killed as a result ofcontrol effort. The latter is difficult to predictbecause relationships between wild dogdensities and livestock predation levels arehighly variable (Box 2, Step 3) and depen-dent on a wide range of factors includingavailability of alternative food, social struc-ture of dogs and activity of individual dogs.

7.7.2 Case study 2 — Hypothetical case study of benefit–cost analysis for sheepproperties

Bene

fit–c

ost r

atio

9

8

7

6

5

4

3

2

5 10 15

Sheep losses (%)

20 25 30 35

1

0

-1

10

Strategy 2

Strategy 1

Figure 18: An example of benefit–cost ratio analysesshowing the benefit–cost of effort to control wilddogs including a baited zone (Strategy 2 in Table 7)and control effort without a baited zone (Strategy 1 inTable 7) for sheep graziers in north-eastern NewSouth Wales. Benefit–cost ratios below 1(dotted/dashed line) are uneconomic.


Table 7: Hypothetical benefit–cost comparison of two wild dog control strategies using two sets of sheep productivitydata. (Hypothetical annual percentage losses experienced in the absence of control extrapolated from Thomson (1984a);sheep production and control effort figures for Strategy 1 extrapolated from NERDA (undated c. 1966) and those forStrategy 2 extrapolated from Thompson and Fleming (1991) and Fleming (1996b); average hourly wages are fromConsumer Price Index figures, Australian Tax Office (unpublished data 1996).

Parameter

Mean sheep numbers(head per property)

Greasy fleece weight (kilograms per head)

Value of sheep ($ per head)

Annual mean sheep killedby wild dogs (head per property)

Annual control effort (hours per property)

Cost of predation ($ per property)

Cost of control ($ per property)

Wild dogcontrol

Strategy 11

1030

3.7

42.4

25.6

185.6

1085

2975

Wild dog control

Strategy 22

1030

3.7

42.4

11.2

100

475

1790

No control(16% losses)

1030

3.7

42.4

165

0

6996

0

No control(33% losses)

1030

3.7

42.4

340.4

0

14432

0

Benefit-cost ratios3

Control

Strategy 1

Strategy 2

@16% annual losses

1.46:1

2.87:1

@33% annual losses

3.29:1

6.15:1

1Strategy 1: Control with no buffer-zone2Strategy 2: Control with a buffer-zone3Benefit–cost ratio is calculated by dividing the benefits (cost of predation when there is no control minuscost of predation under a particular control strategy) by the total costs of the control strategy. The totalcosts of the control strategy are the cost of control effort plus any predation costs that are still incurreddespite the control effort.

Summary

Although there is much knowledge about theecology, behaviour and effects of predationby dingoes and other wild dogs, some topicsrequire further research to enable best prac-tice management to be implemented. Thereare also knowledge deficits relating to theconservation of dingoes, the effects of con-trol programs on populations of non-targetanimals and the interactions between wilddogs and feral cats, foxes and native carni-vores. These knowledge deficiencies are notlisted in priority order.

8.1 Assess relationship betweenwild dog abundance and predation of cattle

Deficiency

A poor understanding of the relationshipbetween wild dog density and the predationof cattle in extensive cattle areas.

Developments required

Measurement of predation of cattle and wilddog density is required in different regions.This can be done during normal wild dogcontrol programs. As for sheep enterprises inthe eastern highlands, the costs of the ‘nodog control’ option requires evaluation(Section 7.3.4).

Consequences

If consistent relationships exist between wilddog density and predation of cattle, marginalanalyses have more credibility and becomevery useful in best practice management.Negative relationships between cattle preda-tion levels and wild dog density would indi-cate that control of wild dog populations wasdisadvantageous and should be discontin-ued and alternative strategies used, or thatcontrol should be coordinated over largerareas.

8.2 Assess relative effectivenessand efficacy of baiting strategies

Deficiency

The relationship between the cost of control(particularly aerial and ground baiting) andthe impact of wild dog predation on theprofitability of livestock (particularly sheep)in the eastern highlands is unknown. Thisknowledge is essential for decisions aboutthe continuing expenditure on wild dog con-trol, dingo conservation and the suitability ofsome lands for grazing enterprises. If possi-ble, areas where no wild dog control isexerted should be incorporated in theassessments so that the costs of the ‘no dogcontrol’ (Section 7.3.4) option can be evalu-ated and benefit–cost ratios of different con-trol programs can be compared against noaction.


Experiments to evaluate the relative costsand benefits of the various methods of wilddog control.

Consequences

Improved management decisions relating towild dog control, conservation and the prof-itability of grazing enterprises. If the relation-ships between wild dog density and preda-tion are tenuous or highly variable, the con-tinued use of all control strategies thatreduce dog abundance in sheep areas arejustified provided they fall within budgetallocations.


8. Deficiencies in knowledge and practice

8.3 Assess effect of RabbitCalicivirus Disease on dingo predation of livestock

Deficiency

The introduction of rabbit calicivirus has hada varied impact on rabbit numbers with thegreatest reductions being in arid and semi-arid areas (rainfall less than 300 millimetres).The effect of reduced abundance of rabbitson wild dog predation of cattle and nativeanimals and on wild dog population abun-dance is poorly known.


The interaction of rabbit calicivirus on rabbitpopulations and the subsequent effects onwild dog predation of cattle and native ani-mals requires research.

Consequences

Prey abundance may be an important indica-tor of the probability of wild dog predationof livestock. If so, understanding of preda-tor–prey relationships and the effects ofRabbit Calicivirus Disease on rabbit popula-tions is essential.

8.4 Investigate feasibility of compensation schemes forwild dog predation

Deficiency

A self-funding insurance or compensationscheme has been suggested to reimburselandholders for livestock losses. Suchschemes could be a substitute for control oran adjunct to reduced levels of control. Thefeasibility and function of such schemes hasnot been investigated.


The feasibility and operation of self-fundedcompensation schemes require investigationunder different agricultural systems wherewild dogs occur.

Consequences

Determination of the likely success of com-pensation or insurance schemes as an alter-native management strategy for wild dogs.

8.5 Train vertebrate pest controloperators and managers

Deficiencies

There is a need for training packages toinform field operatives and managers of thestrategic approach to wild dog managementand the conservation issues relating tohybridisation.


The information contained in this book pro-vides a starting point for the development oftraining packages for field operatives andmanagers. These should be continuouslyupdated to account for changes in legislationand new research findings.

Consequences

Soundly-based management will be imple-mented and the results of new research willbe incorporated into management plans.

8.6 Improve public awareness ofagricultural production, conservation and animalwelfare issues for wild dogcontrol

Deficiency

Public perceptions of vertebrate pest controlprograms are often based on incorrectpremises or worst-case scenarios. Many peo-ple are not fully aware of the range of agri-cultural and conservation reasons for wilddog control. There is also misinformationabout the use and safety of poisons, and inparticular the characteristics of 1080 (sodiumfluoroacetate) (Sections 4.2.4 and 6.4.4) thatmake it the poison of choice for controllingwild dogs. Although considered a humane


poison by many researchers and managers,there is a perception by many in the widercommunity that 1080 causes painful deaths.There is increasing community interest in theneed for, and humaneness of, control tech-niques used on pest animals. These issuesrequire extension to the broader community.


Education and media programs to extendinformation on the welfare issues of wilddog management. Include animal welfareissues associated with wild dog control inschool curricula.

Consequences

More techniques for management of wilddogs that include non-lethal strategies willbe available.

Better informed public debate about theconservation, livestock production and ani-mal welfare issues relating to the manage-ment of wild dogs.

8.7 Develop species-specific andmore humane control techniques for wild dogs

Deficiency

The increasing interest of the wider commu-nity in animal welfare requires that morehumane methods of control for pest animalsbe continually investigated. For example, theuse of strychnine on trap jaws is the presentlyaccepted and recommended method for min-imising the suffering of dogs held in traps.Strychnine causes a quick yet painful death.


Development and monitoring of new andimproved techniques to minimise or eliminatethe potentially adverse effects of control mea-sures on the welfare of wild dogs and non-tar-get animals is needed. For example, a fast-act-ing more humane poison for use as a substitutefor strychnine on trap jaws requires investiga-tion. The economic viability of new techniqueswill need to be assessed.

The use of deterents to minimise interactionsbetween wild dogs and people is untested.Alternative management tools such as live-stock guarding dogs and toxic collars havebeen tried in other countries but their applica-tion and benefit–cost analyses for Australianconditions are unknown.

Consequences

Ongoing assessment and improvement of ani-mal welfare aspects of wild dog control. Themost humane methods of control will be usedfor managing wild dogs and the welfare ofnon-target animals will not be compromisedduring control programs.

8.8 Assess economic importanceof hydatids in wild dogs

Deficiency

The relationship between wild dog popula-tions and the prevalence of hydatidosis(causal agent Echinococcus granulosus) inlivestock has not been fully investigated. Atlocal levels, it is known that hydatidosis incattle is sometimes associated with grazinglands adjacent to or within country inhabitedby wild dogs; however full epidemiologicalstudies have not been completed. The eco-nomic importance of hydatid infection inlivestock remains unclear and strategies toprevent or reduce their occurrence are yet tobe formulated.


Research is needed to estimate the economicimportance of hydatid infection in wild dogsand associated livestock.

Consequences

Development of strategies to prevent orreduce the occurrence of hydatid infection inwild dogs and livestock.


8.9 Assess the role of diseaseinduced mortality in wilddogs

Deficiency

A poor understanding of disease and para-site-induced mortality in wild dogs.


The role of various parasites and diseases onthe mortality of wild dogs and the impact ofthis on population dynamics in differentenvironments requires research.

Consequences

A better understanding of disease and para-site-induced mortality in wild dogs.

8.10 Assess the role of wilddogs if rabies were introduced

Deficiency

A question that is raised by the fact that din-goes came to Australia from Asia is, ‘Why isthere no rabies (Rhabdoviridae) in Australia?What is different here compared with similarhabitats and fauna types in Asia?’


Modelling using demographics and estimatesof rabies transmission coefficients will pro-vide useful indicators of the likelihood ofrabies becoming established in wild dogpopulations. Further research is needed onthe demographics and interactions of com-mensal dogs and wild dogs in the more set-tled areas of eastern Australia and in north-ern Australia.

Studying the ecology and demographics ofdingoes in Asia will assist in understandingwhy rabies is not endemic to Australia orwhat to do if it is introduced to Australia.

Consequences

Understanding the ecology and demograph-ics of dingoes in Asia will assist in rabiescontingency planning.

8.11 Assess risks to non-targetspecies of 1080 poisoning

Deficiency

Although there is evidence indicating thatthe poisoning of wild dogs with 1080 is likelyto have little impact on non-target popula-tions, this requires confirmation in easternAustralia. The potential impact of 1080 bait-ing for wild dog control on populations ofnon-target carnivorous species (includingphascogales (Phascogale spp.) and spotted-tailed quolls, (Dasyurus maculatus)) is themain factor potentially limiting the use of1080 (and particularly aerial baiting) to con-trol wild dogs in eastern Australia.


Scientific assessment of 1080 baiting pro-grams on populations of non-target carnivo-rous native animals in a variety of environ-ments. Investigation of the potential of alter-native baiting strategies (bait substrate,placement of baits and timing of baiting) toreduce non-target risks.

Consequences

A scientific basis for improved risk manage-ment for non-target species in areas wherewild dogs are baited.

8.12 Assess the ecologicaleffects of wild dog controlon feral cat and fox populations

Deficiency

There have been few investigations into theecological relationships between feral cats,wild dogs and foxes (Vulpes vulpes). The effectof controlling wild dogs on the abundance ofother predators requires further study.



Research to assess the factors influencing thedynamics of feral cat and fox populationsand the interplay of these factors with wilddog control. Relationships between wilddogs and these carnivores require researchto enable better management for conserva-tion.

Consequences

Best practice management of wild dogs,foxes, and feral cats can be based on scientificdata rather than unsubstantiated assumptions.

8.13 Assess the interactions ofwild dogs and native carnivore populations

Deficiency

Inadequate understanding of the interplay ofwild dog control with the factors affectingpopulations of native carnivores, particularlyquolls (Dasyurus spp.).


Research to assess the factors influencing thedynamics of native carnivore populationsand wild dog populations, and the interplayof these factors with wild dog control.

Consequences

Improved management for conservation ofnative carnivores in areas of wild dog con-trol.

8.14 Assess effects of wild dogabundance on macropods

Deficiency

The control of wild dogs in forested areas ofsouth-eastern Australia may have causedpopulations of kangaroos and wallabies(Macropus spp.) to increase with concomi-tant grazing impacts on agricultural andforestry enterprises. On some holdings, a

balance is attempted between calf losses dueto predation by wild dogs and the benefitsthat predation on macropods may have inreducing grazing pressure.


Predator–prey relationships between wilddogs and their macropod prey require inves-tigation in south-eastern Australia.

Consequences

Knowledge of predator–prey relationships incattle country of south-eastern Australia willallow for management strategies that bal-ance an acceptable abundance of macrop-ods with acceptable level of predation ofcalves.

8.15 Assess the values of dingoconservation

Deficiency

Economic frameworks are needed to assistthe community in meeting dingo conserva-tion goals. Better management of budgetsrelating to the conservation of dingoes canonly be achieved if the unpriced values ofdingoes are estimated enabling marginalanalyses and cost–benefit analyses of con-servation strategies. The contingent andinherent values the community places on theconservation of dingoes have not beenestablished.


Establish the contingent and inherent valuesof dingoes to the wider community.

Consequences

Estimation of unpriced values of dingoes willenable marginal analyses and cost–benefitanalyses of conservation strategies to deter-mine the most cost-effective managementstrategies for meeting community conserva-tion objectives.


8.16 Develop a method to identify genetically pure dingoes

Deficiency

If governments enact legislation to conservedingoes (Section 5.3.3) while controllingother wild dogs, non-destructive methods todistinguish between dingoes and other wilddogs are required. For example, conserva-tion programs on Fraser Island wouldrequire the capture of all wild dogs over timeto enable their genetic purity to be assessed,with subsequent release or destructiondepending on their genetic status. At pre-sent, there is no consensus applicable to thewhole country on what constitutes a ‘pure’dingo.


Techniques to differentiate between dingoesand other wild dogs are being developedand the continuation of this work should beencouraged. A national decision must be sci-entifically made on what genotype and/orphenotype constitutes a pure dingo.

Consequences

Policy decisions and management strategiesfor conservation of dingoes depend on theability to differentiate between subspecies ofwild dogs. Without a method of differentiat-ing that can be applied to live animals, con-servation strategies are impossible to imple-ment. A national policy on the genotyperequired for genetic purity will enable con-servation to advance; without such a policy,dingo conservation is a lost cause.

8.17 Improve knowledge aboutgenetics of wild dogs

Deficiency

Information on the taxonomic status of wilddogs throughout Australia is required, espe-cially in climatically different regions whereraces of pure dingoes may occur.


Public awareness of the issue of hybridisa-tion, a rapid method of field assessment ofpure dingoes, and strategies and techniquesfor the removal of hybrids and free-roamingdogs from areas of pure dingoes arerequired. The adequacy of refugia on islandsand the mainland to allow populations ofpure dingoes to be maintained naturallyneeds investigation.

Consequences

Genetic assessment of regional variations indingo populations and self-sustaining popu-lations of pure dingoes.

8.18 Assess the ecological roleof dingo hybrids

Deficiency

Although the role of dingoes in central andnorthern Australia is understood, the role ofwild dogs in eastern Australian environmentsis less well known. Also, it is unknownwhether the increased proportion of hybridswill change the ecological role currently heldby dingoes.


Scientific investigations are required tounderstand the similarities and differencesbetween dingoes and hybrids and whetherthese differences will alter predation rates onnative fauna and livestock.

Consequences

If the roles of dingoes and hybrids are differ-ent, a new suite of management decisionswill be required. Balancing the requirementsfor control and conservation in managementplans requires knowledge of potentially dif-ferent ecological roles of dingoes and otherwild dogs.


Alcock, J. (1989) Animal Behaviour: anEvolutionary Approach. SinauerAssociates Incorporated, Sunderland.

Alexander, R.M. (1993) Legs and locomotionin Carnivora. N. Dunstan and M.L.Gorman (eds) Mammals as Predators.Symposium of the Zoological Society ofLondon 65: 1–13.

Algar, D. and Kinnear, J.E. (1992) Cyanidebaiting to sample fox populations andmeasure changes in relative abundance.Pp. 135–138 in: P. O'Brien and G. Berry(eds) Wildlife Rabies ContingencyPlanning in Australia. Bureau of RuralResources Proceedings Number 11,Australian Government PublishingService, Canberra.

Allen, L., Engerman, R. and Krupa, H. (1996)Evaluation of three relative abundanceindices for assessing dingo populations.Wildlife Research 23: 197–206.

Allen, L. and Gonzalez, T. (1998) Baitingreduces dingo numbers, changes agestructures yet often increases calf losses.Australian Vertebrate Pest ControlConference 11: 421–428.

Allen, L. and Gonzalez, T. (2000) Results ofthe dingo impact study. Beefy and theBeast, Department of Natural Resources(Queensland) newsletter, Sherwood.Issue 6: 1–4.

Allen, L., Gonzalez, T. and Vitelli, S. (1997)What do dingoes eat? Beefy and the Beast,Department of Natural Resources(Queensland) newsletter, Sherwood.Issue 3: 4.

Allen, L.R., Fleming, P.J.S., Thompson, J.A.and Strong, K. (1989) Effect of presenta-tion on the attractiveness and palatabilityto wild dogs and other wildlife of twounpoisoned wild-dog bait types.Australian Wildlife Research 16: 593–598.

Allison, C. and Coombes, I. (1969) TheAustralian Hunter: A Comprehensive Guideto Game, Equipment, Hunting andPhotography. Cassell Australia, NorthMelbourne.

Andelt, W.F. (1985) Behavioural ecology ofcoyotes in south Texas. WildlifeMonographs 94: 1034–1036.

Andelt, W.F. (1992) Effectiveness of livestockguarding dogs for reducing predation ondomestic sheep. Wildlife Society Bulletin20: 55–62.

Anderson, E.R. and McLennan, S.R. (1986)Better management of Queensland nativepastures. J.R. Wythes and T.H. Rudder(eds) New ways to maximise returns fromQueensland beef. Proceedings of theAustralian Society of Animal Production16: 105–117.

Andrew, J. (undated c. 1999) Integratingeducation and extension. Bureau ofResource Sciences, Canberra.

Animal and Plant Control Commission(1993) Present policy on management ofdingo populations in South Australia.Unpublished policy document, SouthAustralian Animal and Plant ControlCommission, Adelaide.

Animal Health Committee (1990) Model codeof practice for the welfare of animals — thesheep. No. 34. Standing Committee onAgriculture Technical Report Series,Australian Government Publishing Service,Canberra.

Anon (1987) Cattle guard lambs. The Furrow92: 5.

Archer, M. (1974) New information about theQuaternary distribution of the thylacine(Marsupialia, Thylacinidae) in Australia.Journal of the Proceedings of the RoyalSociety of Western Australia 57: 43–50.


References

Australian Associated Press (1998) ‘Childfourth dingo victim’. Sydney MorningHerald, April 9., p6.

Australian Geographic Society (1996) FirstFleet. Pp. 1342–1345 in: The AustralianEncyclopaedia. Australian Geographic PtyLtd, Terry Hills.

Backholer, J.R. (1986) A survey of landholderson the wild dog problem in eastern Victoria:a preliminary analysis. Unpublished reportto Land Protection Service, Department ofConservation, Forests and Lands, Victoria.

Bacon, J.S. (1955) The Australian Dingo: TheKing of the Bush. McCarron Bird,Melbourne.

Baird, R.F. (1991) The dingo as a possiblefactor in the disappearance of Gallinulamortierii from the Australian mainland.Emu 91: 121–122.

Balderstone, E. (1992) Maremma guard dogsbeat lamb predators. Australian FarmJournal (February): 61.

Baldock, F.C., Arthur, R.J. and Lawrence, A.R.(1985) A meatworks survey of bovinehydatidosis in southern Queensland.Australian Veterinary Journal 62:238–243.

Bauer, F.H. (1983) The coming of Europeanman. Pp. 26–45 in: G. Crook (ed) Man inthe Centre. CSIRO, Melbourne.

Bauer, T.A. and Mortensen, D.A. (1992) Acomparison of economics and economicoptimum thresholds for two annual weedsin soybeans. Weed Technology 6: 228–235.

Bay-Petersen, J.A. (1979) The role of the dogin the economy of the New ZealandMaori. Pp. 165–181 in: A. Anderson (ed.)Osteological and Archaeological Papersfrom the South Pacific in Honour of R. J.Scarlett. New Zealand ArchaeologicalMonograph II. BAR International Series62.

Beck, A.M. (1973) The Ecology of Stray Dogs:A Study of Free-ranging Urban Dogs.York Press, Baltimore.

Bekoff, M. (1975) Social behaviour and ecolo-gy of the African Canidae: a review. Pp.120–142 in: M.W. Fox (ed.) The WildCanids: Their Systematics, BehaviouralEcology and Evolution. Nostrand ReinholdCompany, New York.

Belcher, C.A. (1998) Susceptibility of thetiger quoll, Dasyurus maculatus, and theeastern quoll, D. viverrinus, to 1080-poi-soned bait in control programmes for ver-tebrate pests in eastern Australia. WildlifeResearch 25: 33–40.

Bellwood, P. (1978) Man's Conquest of thePacific: The Prehistory of Southeast Asiaand Oceania. Collins, Auckland.

Bellwood, P. (1984) The great Pacific migra-tion. Pp. 80–93 in: Yearbook of Scienceand the Future 1984. EncyclopaediaBritannica, Washington.

Best, L.W., Corbett, L.K., Stephens, D.R. andNewsome, A.E. (1974) Baiting trials fordingoes in Central Australia, with poison'1080', encapsulated strychnine, andstrychnine suspended in methyl cellulose.CSIRO Division of Wildlife ResearchTechnical Paper No. 30.

Bird, P. (1994) Improved electric fences andbaiting techniques — a behaviouralapproach to integrated dingo control.Unpublished Final Report on Project DAS39 to the Wool Research andDevelopment Corporation.

Bird, P., Lock, B. and Cook, J. (1997) TheMuloorina Cell: a long-term trial to assesselectric dingo fence designs. Unpublishedreport to the South Australian Dog FenceBoard, Adelaide.

Bomford, M. and O’Brien, P. (1995) Eradicationor control for vertebrate pests? WildlifeSociety Bulletin 23: 249–255.

Bomford, M. and workshop participants(1995) Integrating the science and eco-nomics of vertebrate pest management.Unpublished workshop papers. Bureau ofResource Sciences, Canberra.


Braysher, M. (1993) Managing VertebratePests: Principles and Strategies. Bureau ofResource Sciences, Australian GovernmentPublishing Service, Canberra.

Breckwoldt, R. (1988) A Very ElegantAnimal: the Dingo. Angus and Robertson,Australia.

Brisbin, I.L. (1989) The Carolina dog. Off-lead (Jan): 18.

Brisbin, I.L. (1998) Comment on the pro-posed conservation of usage of 15 mam-mal specific names based on wild specieswhich are antedated by or contemporarywith those based on domestic animals.Bulletin of Zoological Nomenclature 53:28–37.

Brown, G.W. and Triggs, B.E. (1990) Diets ofwild canids and foxes in east Gippsland1983–1987, using predator scat analysis.Australian Mammalogy 13: 209–13.

Bureau of Rural Sciences/Centre forAgricultural and Regional Economics(1999). Vertebrate Pest Control SimulationModel (CD-ROM), v. 1.1.20.

Burbidge, A.A. and McKenzie, N.L. (1989)Patterns in the modern decline of WesternAustralia's vertebrate fauna: causes andconservation implications. BiologicalConservation 50: 143–198.

Burns, R.J. (1983) Coyote predation aversionwith lithium chloride: management impli-cations and comments. Wildlife SocietyBulletin 11: 128–132.

Busana, F., Gigliotti, F. and Marks, C.A.(1998) Modified M-44 cyanide ejector forthe baiting of red foxes (Vulpes vulpes).Wildlife Research 25: 209–215.

Calver, M.C., King, D.R., Bradley, J.S. andMartin, G.R. (1989) An assessment of thepotential target specificity of 1080 preda-tor baiting in Western Australia.Australian Wildlife Research 16: 625–638.

Carbyn, L.N. (1989) Coyote attacks on chil-dren in western North America. WildlifeSociety Bulletin 17: 444–446.

Carlisle, C.H. and Atwell, R.B. (1984) A sur-vey of heartworms in dogs. AustralianVeterinary Journal 61: 356–360.

Catling, P.C. (1979) Seasonal variation inplasma testosterone and the testis in cap-tive male dingoes, Canis familiaris dingo.Australian Journal of Wildlife 27:939–944.

Catling, P.C. and Burt, R.J. (1995) Why arered foxes absent from some eucalyptforests in eastern New South Wales?Wildlife Research 22: 535–546.

Catling, P.C., Burt, R.J. and Kooyman, R.(1997) A comparison of techniques usedin a survey of the ground-dwelling andarboreal mammals in forests in north-east-ern New South Wales. Wildlife Research24: 417–432.

Catling, P.C., Corbett, L.K. and Newsome,A.E. (1992) Reproduction in captive andwild dingoes (Canis familiaris dingo) intemperate and arid environments ofAustralia. Wildlife Research 19: 195–205.

Catling, P.C., Corbett, L.K. and Westcott, M.(1991) Age determination in dingoes andcross breeds. Wildlife Research 18: 75–83.

Caughley, G. (1980) Analysis of VertebratePopulations. John Wiley and Sons Ltd,Chichester.

Caughley, G., Grigg, G.C., Caughley, J. andHill, G.J.E. (1980) Does dingo predationcontrol the densities of kangaroos andemus? Australian Wildlife Research 7: 1–12.

Caughley, G. and Sinclair, A.R.E. (1994)Wildlife Ecology and Management.Blackwell Scientific Publications, Boston.

Caughley, J., Bomford, M., Parker, B.,Sinclair, R., Griffiths, J. and Kelly, D.(1998) Managing Vertebrate Pests:Rodents. Bureau of Resource Sciences,Australian Government PublishingService, Canberra.

Chamala, S. and Mortiss, P.D. (1990)Working together for Land Care: GroupManagement Skills and Strategies.Australian Academic Press, Brisbane.


Chiarelli, A.B. (1975) The chromosomes ofthe Canidae. Pp. 40–53 in: M.W. Fox (ed.)The Wild Canids: Their Systematics,Behavioural Ecology and Evolution. VanNostrand Reinhold Company, New York.

Choquenot, D., McIlroy, J. and Korn, T. (1996)Managing Vertebrate Pests: Feral Pigs.Bureau of Resource Sciences, AustralianGovernment Publishing Service, Canberra.

Clutton-Brock, J. (1977) Man-made dogs.Science 197: 1340–1342..

Clutton-Brock, J. (ed.) (1989) The WalkingLarder: Patterns of Domestication,Pastoralism and Predation. UnwinHyman, London.

Clutton-Brock, J. (1992) The process ofdomestication. Mammal Review 22:79–85.

Coe, S. (1997) The Basenji. Doral Publishing,Wilsonville.

Colenso, W. (1877) Notes on the ancient dogof the New Zealanders. Transactions andProceedings of the New Zealand Institute10: 135–155.

Coman, B.J. (1972a) Helminth parasites ofthe dingo and feral dog in Victoria withsome notes on the diet of the host.Australian Veterinary Journal 48:456–461.

Coman, B.J. (1972b) A sylvatic cycle for thehydatid tapeworm (Echinococcus granu-losus) in remote areas of eastern Victoria.Australian Veterinary Journal 48:552–553.

Coman, B.J. and Robinson, J.L. (1989) Someaspects of stray dog behaviour in an urbanfringe area. Australian VeterinaryJournal 66: 30–32.

Commonwealth of Australia (1992) NationalStrategy for the Conservation of Australia’sBiological Diversity. Australian GovernmentPublishing Service, Canberra.

Connolly, G.E. (1980) Use of compound1080 in livestock neck collars to killdepredating coyotes: a report on field andlaboratory research. U.S. Department ofthe Interior., Denver, Colorado.

Copinger, R. and Copinger, L. (1978)Livestock Guarding Dogs for U.S.Agriculture. Livestock Dog Project,Montague.

Corbett, L.K. (1974) Contributions to thebiology of dingoes (Carnivora: Canidae)in Victoria. Unpublished M.Sc. thesis,Monash University, Melbourne.

Corbett, L.K. (1985) Morphological compar-isons of Australian and Thai dingoes: areappraisal of dingo status, distribution andancestry. Proceedings of the EcologicalSociety of Australia 13: 277–291.

Corbett, L.K. (1988a) The identity and biolo-gy of Thai dingoes. Unpublished FinalReport to the National Research Councilof Thailand, Bangkok.

Corbett, L.K. (1988b) Social dynamics of acaptive dingo pack: population regulationby dominant female infanticide. Ethology78: 177–98.

Corbett, L.K. (1989) Assessing the diet of din-goes from feces: a comparison of 3 meth-ods. Journal of Wildlife Management 53:343–346.

Corbett, L. (1995a) The Dingo in Australiaand Asia. University of New South WalesPress Ltd, Sydney.

Corbett, L. (1995b) Dingoes: Expatriatewolves or native dogs? Nature Australia25: 46–55.

Corbett, L.K. (1995c) Does dingo predationor buffalo competition regulate feral pigpopulations in the Australian wet–drytropics? An experimental study. WildlifeResearch 22: 65–74.

Corbett, L. (1998) Management of dingoes onFraser Island. Unpublished report to theQueensland Department of Environment,Brisbane.


Corbett, L. (In press) The conservation statusof dingoes (Canis lupus dingo) inAustralia, with particular reference to NewSouth Wales: threats to pure dingoes andpotential solutions. Australian Zoologist.

Corbett, L. and Newsome, A. (1975) DingoSociety and its Maintenance: a PreliminaryAnalysis. Pp. 369–379 in: M.W. Fox (ed.)The Wild Canids: Their Systematics,Behavioural Ecology and Evolution.Nostrand Reinhold Co., New York.

Corbett, L.K. and Newsome, A.E. (1987) Thefeeding ecology of the dingo. III. Dietaryrelationships with widely fluctuating preypopulations in arid Australia: an hypothe-sis of alternation of predation. Oecologia74: 215–227.

Cornell, D. and Cornely, J.E. (1979) Aversiveconditioning of campground coyotes inJoshua Tree National Monument. WildlifeSociety Bulletin 7: 129–131.

Denny, M. (1992) Dorrigo ManagementArea Fauna Impact Statement ofProposed Forestry Operations. ForestryCommission of New South Wales, Sydney.

Dickman, C.R. (1996) Overview of theimpacts of feral cats on Australian fauna.Australian Nature Conservation Agency,Canberra.

Dixon, J.M. (1989) Thylacinidae. Pp. 549–559in: D.W. Dalton and B.J. Richardson (eds)Fauna of Australia. Mammalia. AustralianGovernment Publishing Service, Canberra.

Dixon, J.M. and Huxley, L. (eds) (1985)Donald Thomson's Mammals and Fishesof Northern Australia. Nelson,Melbourne.

Dobbie, W.R., Berman, D.M. and Braysher,M.L. (1993) Managing Vertebrate Pests:Feral Horses. Bureau of Resource Sciences,Australian Government Publishing Service,Canberra.

Durie, P.H. and Riek, R.F. (1952) The role of thedingo and wallaby in the infestation of cattlewith hydatids (Echinococcus granulosus)(Batsch 1786; Rudolph 1805) inQueensland. Australian Veterinary Journal28: 249–254.

Eason, C.T., Wickstrom, M.L. and Spurrs, E.B.(1998) Review of impacts of large-scalesodium monofluroacetate (1080) use inNew Zealand. Australian Vertebrate PestControl Conference 11: 105–110.

Edwards, G.P, de Preu, N.D, Shakeshaft, B.Jand Crealy, I.V. (2000) An evaluation oftwo methods of assessing feral cat anddingo abundance in central Australia.Wildlife Research 27: 143–149.

Eldridge, S.R. and Bryan, R. (1995) Dingoquestionnaire survey June–November1995. Unpublished report to Parks andWildlife Commission, Northern Territory,Darwin.

English, A.W. and Taske, J.E. (1998) A possi-ble role for the application of veterinarycapture techniques to the restraint ofhuman offenders. P 48 in: UnpublishedHandbook of the Australian VeterinaryAssociation Annual Conference, Sydney.

Ernest, H.B., Penedo, M.C.T, May, B.P.,Syvanen, M. and Boyce, W.M (2000).Molecular tracking of mountain lions inthe Yosemite Valley region in California:genetic analysis using microsatellites andfaecal DNA. Molecular Ecology 9:433–441.

Fawcett, A.M. (1994) Assessment of bait sta-tions for the control of wild dogs on theNew England Tablelands. UnpublishedB.Sc. (Hons) thesis, University of NewEngland, Armidale.

Fenner, F., Bachmann, P.A., Gibbs, E.P.J.,Murphy, F.A., Studdert, M.J. and White,D.O. (1987) Rhabdoviridae. Pp. 531–541in: Veterinary Virology. Academic Press,London.

Fennessy, B.V. (1966) The impact of wildlifespecies on sheep production in Australia.Proceedings of the Australian Society forAnimal Production 6: 148–156.

Finlayson, H.H. (1935) The Red Centre: Manand Beast in the Heart of Australia.Angus and Robertson, Sydney.


Fleming, P.J.S. (1987) Wild dog predation innorth-eastern New South Wales. AustralianVertebrate Pest Control Conference 8:103–106.

Fleming, P.J.S. (1996a) Ground-placed baitsfor the control of wild dogs: evaluation ofa replacement-baiting strategy in north-eastern New South Wales. WildlifeResearch 23: 729–740.

Fleming, P.J.S. (1996b) Aspects of the man-agement of wild dogs (Canis familiaris)in north-eastern New South Wales.Unpublished M.Res.Sc. thesis, Universityof New England, Armidale.

Fleming, P.J.S., Allen, L.R., Berghout, M.J.,Meek, P.D., Pavlov, P.M., Stevens, P.,Strong, K., Thompson, J.A. and Thomson,P.C. (1998) The performance of wild-canid traps in Australia: efficiency, selec-tivity and trap-related injuries. WildlifeResearch 25: 327–338.

Fleming, P.J.S., Allen, L.R. and Mitchell, J.L.(1992) The control of wild canids in east-ern Australia in relation to wildlife rabies.Pp. 139–146 in: P. O'Brien and G. Berry(eds) Wildlife Rabies ContingencyPlanning in Australia. Bureau of RuralResources Proceedings Number 11,Australian Government PublishingService, Canberra.

Fleming, P.J.S. and Korn, T.J. (1989)Predation of livestock by wild dogs ineastern New South Wales. AustralianRangelands Journal 11: 61–66.

Fleming, P.J.S. and Parker, R.W. (1991)Temporal decline of 1080 within meatbaits used for wild dog control in NewSouth Wales. Wildlife Research 18:729–740.

Fleming, P.J.S. and Robinson, D. (1986) Theimpact of wild dogs on livestock produc-tion. Proceedings of the AustralianSociety for Animal Production 16: 84–87.

Fleming, P.J.S., Thompson, J.A. and Nicol, H.I.(1996) Indices for measuring the efficacy ofaerial baiting for wild dog control in north-eastern New South Wales. Wildlife Research23: 665–674.

Foley, J.C. (1957) Droughts in Australia.Review of records from earliest years of set-tlement to 1955. Bureau of MeteorologyBulletin 43: 1–281.

Fordyce, G. and Entwistle, K.W. (1992)Improving breeder herd efficiency innorth Queensland. M.T. Sullivan (ed.)Changing beef markets — should northQueensland producers change manage-ment to meet them? Proceedings of theAustralian Society of Animal Production19: 55–56.

Forman, A.J. (1993) The threat of rabiesintroduction and establishment inAustralia. Australian Veterinary Journal70: 81–83.

Fox, M.W. (1971) Behaviour of Wolves, Dogsand Related Canids. Harper and Row,New York.

Friedel, M.H., Foran, B.D. and StaffordSmith, D.M. (1990) Where the creeks rundry or ten feet high: pastoral managementin arid Australia. Proceedings of theEcological Society of Australia 16:185–194.

Gallagher, B. (1991) 8th International PowerFence Manual. Gallagher Electronics Ltd,Hamilton, New Zealand.

Garner, M.G. (1992) World rabies picture:implications for Australia. Pp. 23–37 in: P.O'Brien and G. Berry (eds) Wildlife RabiesContingency Planning in Australia.Bureau of Rural Resources ProceedingsNumber 11, Australian GovernmentPublishing Service, Canberra.

Geering, W.A. and Forman, A.J. (1987)Animal Health in Australia. Volume 9.Exotic Diseases. Australian GovernmentPublishing Service, Canberra.

Gentry, A., Clutton-Brock, J. and Groves, C.P.(1996) Case 3010: Proposed conservationof usage of 15 mammal specific namesbased on wild species which are antedat-ed by or contemporary with those basedon domestic animals. Bulletin ofZoological Nomenclature 53: 28–37.


Ginsberg, J.R. and Macdonald, D.W. (1990)Foxes, Wolves, Jackals and Dogs. An ActionPlan for the Conservation of Canids.IUCN/SSC Canid and Wolf Specialist Groups,IUCN Publications Services Unit, Cambridge.

Gould, J. (1863) The Mammals of Australia.Volume III. Republished (1976) by J.MDixon as The Placental Mammals ofAustralia, with modern commentaries.Macmillan Company, South Melbourne.

Green, B. (1973) The water requirements ofdingoes. Australian Vertebrate Pest ControlConference 5: 76–77.

Green, B. and Catling, P. (1977) The biologyof the dingo. Pp. 51–60 in: H. Messel andS.T. Butler (eds) Australian Animals andtheir Environment. Shakespeare HeadPress, Sydney.

Green, J.S., Woodruff, R.A. and Andelt, W.F.(1994) Do livestock guarding dogs losetheir effectiveness over time? VertebratePest Conference (California) 16: 41–44.

Gregory, G. (1991) Perception of pain asso-ciated with 1080 poisoning. AustralianVertebrate Pest Control Conference 9:300–302.

Gustavson, C.R., Gustavson, J.C. and Holzer,G.A. (1983) Thiabendazole-based tasteaversions in dingoes (Canis familiarisdingo) and New Guinea wild dogs (Canisfamiliaris hallstromi). Applied AnimalEthology 10: 385–388.

Hamilton, A. (1972) Aboriginal man's bestfriend. Mankind 8: 287–295.

Harden, R. (1981) A look at the dingo.Australian Natural History 20: 191–194.

Harden, R.H. (1985) The ecology of thedingo in north-eastern New South Wales.I. Movements and home range.Australian Wildlife Research 12: 25–38.

Harden, R.H. and Robertshaw, J.D. (1987)The ecology of the dingo in north-easternNew South Wales. V. Human predation onthe dingo. Australian Zoologist 24: 65–72.

Hassall and Associates Pty Ltd (1998)Economic evaluation of the role of boun-ties in vertebrate pest management.Unpublished report to the Vertebrate PestProgram, Bureau of Resource Sciences,Canberra.

Higham, G.H.E., Kijngam, A. and Manly,B.F.J. (1980) An analysis of prehistoriccanid remains from Thailand. Journal ofArchaeological Science 7: 149–165.

Holden, P. (1991) Along the Dingo Fence.Hodder and Stoughton, Sydney.

Honacki, J.H., Kinman, K.E. and Koeppl,J.W. (eds) (1982) Mammal Species of theWorld: A Taxonomic and GeographicReference. Allen Press and Association ofSystematic Collections, Lawrence, Kansas.

Hone, J. (1994) Analysis of Vertebrate PestControl. Cambridge University Press,Cambridge.

Hone, J. and Mulligan, H. (1982) VertebratePesticides. Science Bulletin 89. Departmentof Agriculture New South Wales.

Hone, J., Waithman, J., Robards, G.E. andSaunders, G.R. (1981) Impact of wildmammals and birds on agriculture in NewSouth Wales. Journal of the AustralianInstitute of Agricultural Science 47:191–199.

Hooper, P.T., Sallaway, M.M., Latz, P.K.,Maconochie, J.R.,Hyde, K.W., and Corbett,L.K. (1973). Ayers Rock – Mt. OlgaNational Park environmental study, 1972.Land Conservation Series 2: 1–52.

Howell, R.G. (1982) The urban coyote prob-lem in Los Angeles county. Vertebrate PestConference (California) 9: 21–23.

Huffacker, C.B. (1970) The phenomenon ofpredation and its role in nature. Pp 327–343in: P.J Den Boer and G.R Gradwell (eds)Dynamics of populations: proceedings ofthe Advanced Study Institute on Dynamicsof numbers in populations. Oesterback, theNetherlands, 7–18 September 1970.


Hunt, E.R. (1978) Hydatid disease — dangerto man's health. Agricultural Gazette ofNew South Wales. 85: 1–4.

Jarman, P. (1986) The red fox — an exotic,large predator. Pp. 43–61 in: R.L. Kitching(ed.) The Ecology of Exotic Animals andPlants: Some Australian Case Histories.Wiley, Brisbane.

Jarman, P.J. and Wright, S.M. (1993)Macropod studies at Wallaby Creek. IX.Exposure and responses of eastern greykangaroos to dingoes. Wildlife Research20: 833–843.

Jenkins, D.J. and Power, K. (1996) Humanhydatidosis in New South Wales and theAustralian Capital Territory. MedicalJournal of Australia 164: 18–21.

Johnson, K.A., Burbidge, A.A. andMcKenzie, N.L. (1989) AustralianMacropodidoidea: status, causes ofdecline and future research and manage-ment. Pp. 641–657 in: G. Grigg, P. Jarmanand I. Hume, eds, Kangaroos, Wallabiesand Rat-kangaroos. Surrey Beatty andSons Pty Limited, Sydney.

Johnston, M.J. and Marks, C.A. (1997)Attitudinal Survey on Vertebrate PestManagement in Victoria. Report SeriesNumber 3. Agriculture Victoria andDepartment of Natural Resources andEnvironment, Frankston.

Jolly, S.E. and Jolly, L.M. (1992a)Environmental influences of the ability ofcaptive dingoes to find meat baits.Journal of Wildlife Management 56:448–452.

Jolly, S.E. and Jolly, L.M. (1992b) Pen andfield tests of odour attractants for thedingo. Journal of Wildlife Management56: 452–456.

Jones, E. (1990) Physical characteristics andtaxonomic status of wild canids, Canisfamiliaris, from the eastern highlands ofVictoria. Australian Wildlife Research 17:69–81.

Jones, M. (1995) Tasmanian Devil Sarcophilusharrissii. Pp. 82–84 in: R. Strahan (ed.) TheMammals of Australia. AustralianMuseum/Reed Books, Chatswood.

Jones, E. and Stevens, P.L. (1988) Reproductionin wild canids, Canis familiaris, from theeastern highlands of Victoria. AustralianWildlife Research 15: 385–394.

Jones, R.E. and Vere D. T. (1998) The eco-nomics of serrated tussock in New SouthWales. Plant Protection Quarterly13:70–76.

Kelly, J.D. (1977) Canine Parasitology.University of Sydney Post-graduateFoundation in Veterinary Science, Sydney.

Kilgour, R. (1985) Management in behaviour.Pp. 445–458 in: A.F. Fraser (ed.) Ethologyof Farm Animals: A Comprehensive Studyof the Behavioural Features of CommonFarm Animals. Elsevier, Amsterdam.

King, D.R. (1984) 1080 — A selective poisonfor pests. Journal of Agriculture WesternAustralia 25: 12–14.

King, D.R. (1989) An assessment of the haz-ard posed to northern quolls (Dasyurushallucatus) by aerial baiting with 1080 tocontrol dingoes. Australian WildlifeResearch 16: 569–574.

Kirkpatrick, W.E. (1999) Assessment of sodi-um monofluroacetate (1080) in baits andits biodegradation by microorganisms.Unpublished M.Sc. thesis, CurtinUniversity of Technology, Perth.

Kleiman, D. (1966) Scent marking in theCanidae. Symposium of the ZoologicalSociety of London 18: 167–177.

Koehn, J., Brumley, A. and Gehrke, P. (2000)Managing the Impacts of Carp. Bureau ofRural Sciences (Department ofAgriculture, Fisheries and Forestry —Australia), Canberra.

Korn, T. and Fleming, P. (1989) ControllingWild Dogs. Agfact A9.0.14. New SouthWales Agriculture and Fisheries, Sydney.


Kruuk, H. (1972a) The Spotted Hyaena: AStudy of Predation and Social Behaviour.The University of Chicago Press, Chicagoand London.

Kruuk, H. (1972b) Surplus killing by carni-vores. Journal of Zoology, London 166:233–244.

Lawrence, B. and Bossert, W.H. (1967)Multiple character analysis of Canis lupus,latrans, and familiaris, with a discussionof the relationships of Canis niger.American Zoologist 7: 1659–1673.

Leftwich, R.L. and Eckert, R. (1982) The PriceSystem and Resource Allocation. Eighthedition. The Dryden Press, Chicago.

Luenberger, D.G. (1984) Linear and Non-linear Programming. Addison-WesleyPublishing Company, Reading,Massachusetts.

Lumholtz, C. (1889) Amongst Cannibals. J.Murray, London.

Lunney, D., Triggs, B., Eby, P. and Ashby, E.(1990) Analysis of scats of dogs Canisfamiliaris and foxes Vulpes vulpes(Canidae: Carnivora) in coastal forestsnear Bega, New South Wales. AustralianWildlife Research 17: 61–68.

Macknight, C.C. (1976) The Voyage toMarege: Macassan Trepangers inNorthern Australia. Melbourne UniversityPress, Melbourne.

Mahon, P.S., Banks, P.B. and Dickman, C.R.(1998) Population indices for wild carni-vores: a critical study in sand-dune habi-tat, south-western Queensland. WildlifeResearch 25: 11–22.

Manwell, C. and Baker, C.M.A. (1984)Domestication of the dog: hunter, food,bed-warmer, or emotional object. Zeitschriftfür Tierzüchtung und Züchtungsbiologie101: 241–256.

Marks, C.A. and Bloomfield, T.E. (1998)Canine heartworm (Dirofilaria immitis)detected in red foxes (Vulpes vulpes) inurban Melbourne. Veterinary Parasitology78: 147–154.

Marks, C.A., Busana, F. and Gigliotti, F.(1999) Assessment of the M-44 ejector forthe delivery of 1080 for red fox (Vulpesvulpes) control. Wildlife Research 26:101–109.

Marsack, P. and Campbell, G. (1990) Feedingbehaviour and diet of dingoes in theNullarbor region, Western Australia.Australian Wildlife Research 17: 349–357.

Marsterson, S. (1994) Monitoring and man-agement of dingoes in the Central Stationarea, Fraser Island. Unpublished report tothe Queensland Department ofEnvironment, Brisbane.

McIlroy, J.C. (1981) The sensitivity ofAustralian animals to 1080 poison II.Marsupial and eutherian carnivores.Australian Wildlife Research 8: 385–399.

McIlroy, J.C., Cooper, R.J., Gifford, E.J.,Green, B.F. and Newgrain, K.W. (1986a)The effect on wild dogs, Canis f. famil-iaris, of 1080 poisoning campaigns inKosciusko National Park, New SouthWales. Australian Wildlife Research 13:535–544.

McIlroy, J.C., Gifford, E.J. and Carpenter,S.M. (1988) The effect of rainfall andblowfly larvae on the toxicity of 1080-treated meat baits used in poisoning cam-paigns against wild dogs. AustralianWildlife Research 15: 473–483.

McIlroy, J.C., Gifford, E.J. and Cooper, R.J.(1986b) Effects on non-target animal pop-ulations of wild dog trail-baiting cam-paigns with 1080 poison. AustralianWildlife Research 13: 447–453.

Mech, L.D. (1970) The Wolf: Ecology andBehaviour of an Endangered Species.Natural History Press, New York.

Mech, L.D. (1986) Wolf population in theCentral Superior National Forest,1967–1985. NC-270. USDA Forest ServiceResearch Paper.

Medway, L. (1977) The ancient domesticdogs of Malaysia. Journal of the RoyalAsia Society (Malaysia) 50: 14–27.


Meek, P.D. (1998) The biology of theEuropean red fox and the free roamingdog on Bherwerre Peninsula, Jervis Bay.Unpublished M.Appl.Sc. thesis, Universityof Canberra, Canberra.

Meek, P.D. (1999) The movements, roamingbehaviour and home range of free-roam-ing domestic dogs, Canis lupus familiaris,in coastal New South Wales. WildlifeResearch 26: 847–855.

Meek, P., Jenkins, D.J., Morris, B., Ardler, A.J.and Hawksby, R.J. (1995) Use of twohumane leg-holding traps for catchingpest species. Wildlife Research 22:733–793.

Meggit, M.J. (1965) The association betweenAustralian Aborigines and dingoes. Pp7–26 in: Man, Culture and Animals: TheRole of Animals in Human EcologicalAdjustments. Publication 78, AmericanAssociation for the Advancement ofScience, Washington, D.C.

Messier, F. (1985) Social organisation, spatialdistribution and population density ofwolves in relation to moose density.Canadian Journal of Zoology 63:1068–1077.

Milham, P. and Thompson, P. (1976) Relativeantiquity of human occupation andextinct fauna at Madura Cave, south-east-ern Western Australia. Mankind 10:175–180.

Mitchell, J. and Kelly, A. (1992) Evaluatingodour attractants for control of wild dogs.Wildlife Research 19: 211–219.

Mitchell, R.J. (1986) Some perspectives of thecosts and benefits of the Department'swild dog control program. Unpublishedreport to the Department of Conservation,Forests and Lands, Melbourne.

Morling, T.R. (1987) Report of theCommissioner the Hon. Mr Justice T.RMorling. Royal Commission of Inquiry intoChamberlain Convictions. Governmentprinter, Darwin.

Morton, S.R. (1990) The impact of Europeansettlement on the vertebrate animals ofarid Australia: a conceptual model.Proceedings of the Ecological Society ofAustralia 16: 201–213.

Moussalli, A. (1994) A preliminary field studyof dingoes inhabiting the Waddy pointregion, northeast Fraser Island: diet, socialorganisation and behaviour. UnpublishedB.Env.Sc. thesis, Griffith University,Brisbane.

Mowat, G. and Strobeck, C. (2000)Estimating population size of grizzly bearsusing hair capture, DNA profiling, andmark–recapture analysis. Journal ofWildlife Management 64:183–193.

Murray, A. (1998) Tigers and 1080.Unpublished report to the NaturalHeritage Working Group of the AustralianAlps Liaison Committee.

New England Rural DevelopmentAssociation (undated c. 1966) Dingoes inNew England. Community DevelopmentPamphlet No. 7. University of NewEngland, Armidale.

Newsome, A. (1990) The control of verte-brate pests by vertebrate predators.Trends in Ecology and Evolution 5:187–191.

Newsome, A. and Catling, P. (1992) Hostrange and its implications for wildliferabies in Australia. Pp. 97–107 in: P.O'Brien and G. Berry (eds) Wildlife RabiesContingency Planning in Australia.Bureau of Rural Resources ProceedingsNumber 11, Australian GovernmentPublishing Service, Canberra.

Newsome, A.E., Catling, P.C. and Corbett,L.K. (1983a) The feeding ecology of thedingo II. Dietary and numerical relation-ships with fluctuating prey populations insouth-eastern Australia. AustralianJournal of Ecology 8: 345–366.

Newsome, A.E. and Coman, B.J. (1989)Canidae. Pp. 993–1005 in: D.W. Walton andB.J. Richardson (eds) Fauna of Australia.Mammalia. Australian GovernmentPublishing Service, Canberra.


Newsome, A.E. and Corbett, L.K. (1975) VI.Outbreaks of rodents in semi-arid and aridAustralia: Causes, preventions, and evolu-tionary considerations. Pp. 117–153 in: I.Prakash and P.K. Ghosh (eds) Rodents inDesert Environments. Dr. W. Junk, TheHague.

Newsome, A.E. and Corbett, L.K. (1977) Theeffects of native, feral and domestic ani-mals on the productivity of Australianrangelands. Pp. 332–356 in: The Impact ofHerbivores on Arid and Semi-aridRangelands. Australian RangelandSociety, Perth.

Newsome, A.E. and Corbett, L.K. (1982) Theidentity of the dingo II. Hybridisation withdomestic dogs in captivity and in the wild.Australian Journal of Zoology 30:365–374.

Newsome, A.E. and Corbett, L.K. (1985) Theidentity of the dingo III. The incidence ofdingoes, dogs and hybrids and their coatcolours in remote and settled regions ofAustralia. Australian Journal of Zoology33: 363–375.

Newsome, A.E., Corbett, L.K., Best, L.W. andGreen, B. (1973) The dingo. AustralianMeat Research Committee Review 14:1–11.

Newsome, A.E., Corbett, L.K. and Carpenter,S.M. (1980) The identity of the dingo. I.Morphological discriminants of dog anddingo skulls. Australian Journal ofZoology 28: 615–626.

Newsome, A.E., Corbett, L.K., Catling, P.C.and Burt, R.J. (1983b) The feeding ecolo-gy of the dingo. I. Stomach contents fromtrapping in south-eastern Australia, andnon-target wildlife also caught in dingotraps. Australian Wildlife Research 10:477–486.

Newsome, A.E., Corbett, L.K. and Stevens,D.R. (1972) Assessment of an aerial bait-ing campaign against dingoes in centralAustralia. CSIRO Division of WildlifeResearch Technical Paper No. 24.

Norton, G.A. (1988). Philosophy, conceptsand techniques. Pp. 1–17 in: G.A. Nortonand R.P. Pech (eds) Vertebrate PestManagement in Australia: A DecisionAnalysis/ Systems Analysis Approach.Project Report Number 5, CSIRO, Divisionof Wildlife and Ecology, Canberra.

O'Brien, P. (1992) Introduction: rabies inwildlife. Pp. 3–5 in: P. O'Brien and G.Berry (eds) Wildlife Rabies ContingencyPlanning in Australia. Bureau of RuralResources Proceedings Number 11,Australian Government PublishingService, Canberra.

Obendorf, D.L., Matheson, M.J. andThompson, R.C.A. (1989) Echinococcusgranulosus infection of foxes in south-eastern New South Wales. AustralianVeterinary Journal 66: 123–124.

Olsen, P. (1998) Australia's Pest Animals:New Solutions to Old Problems. Bureau ofResource Sciences, Canberra andKangaroo Press, Sydney.

Olsen, S.J. and Olsen, J.W. (1977) TheChinese wolf: ancestor of the New Worlddogs. Science 197: 533–535.

Otis, D.L., Burnham, K.P., White, G.C. andAnderson, D.R. (1978) Statistical inferencefrom capture data on closed animal popu-lations. Wildlife Monographs 62: 1–135.

Parkes, J., Henzell, R. and Pickles, G. (1996)Managing Vertebrate Pests: Feral Goats.Bureau of Resource Sciences, AustralianGovernment Publishing Service,Canberra.

Pavlov, P.M. (1991) Aspects of the ecology ofthe feral pig (Sus scrofa) in semi-arid andtropical areas of eastern Australia.Unpublished Ph.D. thesis, MonashUniversity, Melbourne.

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


Pemberton, D. and Renouf, D. (1993) A fieldstudy of communication and socialbehaviour of the Tasmanian Devil at feed-ing sites. Australian Journal of Zoology41: 507–526.

Peterson, R.O. and Page, R.E. (1988) The riseand fall of Isle Royale wolves, 1975–1986.Journal of Mammology 69: 89–99.

Pickering, M. (1992) Garawa methods ofgame hunting, preparation and cooking.Records of the South Australian Museum26: 9–23.

Pople, A.R., Grigg, G.C., Cairns, S.C., Beard,L.A. and Alexander, P. (2000) Trends inthe numbers of red kangaroos and emuson either side of the South Australiandingo fence: evidence of predator regula-tion? Wildlife Research 27: 269–276.

Pretty, J.N. (1994) Alternative systems ofenquiry for sustainable agriculture.University of Sussex, United Kingdom.Institute for Developmental StudiesBulletin 25: 37–46.

Price, S. (1994) Monitoring and managementof dingoes in the Waddy Point area, FraserIsland. Unpublished report to theQueensland Department of Environment,Brisbane.

Queensland Parks and Wildlife Service(1999) Draft Fraser Island dingo manage-ment strategy. The State of Queensland,Environmental Protection Agency,Brisbane.

Ralls, K. (1971) Mammalian scent marking.Science 171: 443–449.

Rankine, G. and Donaldson, L.E. (1968)Animal behaviour and calf mortalities in anorth Queensland breeding herd.Proceedings of the Australian Society ofAnimal Production 4: 138–143.

Reichel, M.P., Lyford, R.A. and Gasser, R.B.(1994) Hyperendemic focus of echinococ-cosis in north-eastern Victoria. The MedicalJournal of Australia 160: 499–501.

Reiger, I. (1979) Scent rubbing in carnivores.Carnivore 2: 17–25.

Robertshaw, J.D. and Harden, R.H. (1985a)Ecology of the dingo in north-easternNew South Wales. II. Diet. AustralianWildlife Research 12: 39–50.

Robertshaw, J.D. and Harden, R.H. (1985b)Ecology of the dingo in north-easternNew South Wales. III. Analysis of macrop-od bone fragment in dingo scats.Australian Wildlife Research 12: 163–171.

Robertshaw, J.D. and Harden, R.H. (1986)Ecology of the dingo in north-easternNew South Wales. IV. Prey selection bydingoes and its effect on the major preyspecies, the swamp wallaby (Wallabiabicolor Desmarest). Australian WildlifeResearch 13: 141–163.

Robertshaw, J.D. and Harden, R.H. (1989)Predation in Macropodoidea: a review.Pp. 735–753 in: G. Grigg, P. Jarman and I.Hume (eds) Kangaroos, Wallabies andRat-kangaroos. Surrey Beatty and SonsPty Limited, Sydney.

Rolls, E.C. (1984) They All Ran Wild: theAnimals and Plants that PlagueAustralia. Angus and Robertson, Sydney.

Rothman, R.J. and Mech, L.D. (1979) Scent-marking in lone wolves and newly formedpairs. Animal Behaviour 27: 750–760.

Roughton, R.D. and Sweeney, M.W. (1982)Refinements in scent-station methodologyfor assessing trends in carnivore popula-tions. Journal of Wildlife Management 46:217–229.

Rounsevell, D.E. and Mooney, N. (1995)Thylacine. Pp. 164–165 in: R. Strahan (ed.)The Mammals of Australia. Reed Books,Chatswood.

RSPCA (Royal Society for the Prevention ofCruelty to Animals) [Online] (1997).Dingoes: Position Paper. Available:www.rspca.org.au/.

Russell, R.C. (1990) The relative importanceof various mosquitoes for the transmissionand control of dog heartworm in south-eastern Australia. Australian VeterinaryJournal 67: 191–192.


Saunders, G., Bunn, C., Eggleston, G., Garner,G. and Henzell, R. (1999). Ausvetplan WildAnimal Management Manual. 2nd Edition.Agriculture and Resource ManagementCouncil of Australia. Agriculture, Fisheriesand Forestry – Australia, Canberra.

Saunders, G., Coman, B., Kinnear, J. andBraysher, M. (1995) Managing VertebratePests: Foxes. Bureau of Resource Sciences,Australian Government Publishing Service,Canberra.

Saunders, G. and Fleming, P. (1988)Interaction between feral pigs, wild dogsand agriculture on the tablelands andcoast of N.S.W. Unpublished Final Reporton Project DAN 15P to the Wool Researchand Development Fund.

Saunders, G., Korn, T. and Giles, J. (1985)Review into wild dog control in easternN.S.W. Unpublished report to the DirectorGeneral of New South Wales Agricultureand Fisheries, Sydney.

Savant, P. (1969) Dingoes will attackhumans. People Magazine (May): 7.

Schaefer, M.M. (1981) The feral dog:Economic implications for graziers operat-ing within the Armidale PasturesProtection Board. UnpublishedB.Ag.Econ. thesis, University of NewEngland, Armidale.

Schlinder, C. (1974) Wild pigs and wildlife.Pastoral Review 84: 605–606.

Schoener, T.W. (1971) Theory of feedingstrategies. Annual Review of Ecology andSystematics 2: 369–404.

Scrivner, J.H. and Wade, D.A. (1986) The1080 livestock protection collar for preda-tor control. Rangelands 8: 103–106.

Seawright, A.A. (1989) Animal Health inAustralia. Volume 2. Chemical and PlantPoisons. Australian GovernmentPublishing Service, Canberra.

Seddon, H.R. (1965a) Diseases of DomesticAnimals in Australia. Part 5. BacterialDiseases. Volume I. Department of HealthPublication No. 9, Commonwealth ofAustralia, Canberra.

Seddon, H.R. (1965b) Diseases of DomesticAnimals in Australia. Part 5. BacterialDiseases. Volume II. Department of HealthPublication No. 10, Commonwealth ofAustralia, Canberra.

Seddon, H.R. (1966) Diseases of DomesticAnimals in Australia. Part 4. Protozoanand Viral Diseases. Department of HealthPublication No. 8, Commonwealth ofAustralia, Canberra.

Seddon, H.R. (1967) Diseases of DomesticAnimals in Australia. Part 2. ArthropodInfestations (Flies, Lice and Fleas).Department of Health Publication No. 6,Commonwealth of Australia, Canberra.

Seddon, H.R. (1968) Diseases of DomesticAnimals in Australia. Part 3. ArthropodInfestations (Ticks and Mites. Animals,Insects and other Agents harmful toStock). Department of Health PublicationNo. 7, Commonwealth of Australia,Canberra.

Seddon, H.R. and Albiston, H.E. (1967)Diseases of Domestic Animals inAustralia. Part 1. Helminth infestations.Department of Health Publication No. 5,Commonwealth of Australia, Canberra.

Shepherd, N.C. (1981) Predation of red kan-garoos, Macropus rufus, by the dingo,Canis familiaris dingo (Blumenbach), innorth-western New South Wales.Australian Wildlife Research 8: 255–262.

Sinclair, A.R.E. (1989) Population regulationin animals. J.M. Cherrett (ed.) EcologicalConcepts. Symposium of the BritishEcological Society of London (1989):197–241.

Sinden, J.A. and Worrell, A.C. (1979)Unpriced Values: Decisions WithoutMarket Places. Wiley-Interscience, NewYork.

Smith, A.P., Moore, D.M. and Andrews, S.P.(1992) Proposed Forestry Operations inthe Glen Innes Forest Management Area:Fauna Impact Statement. ForestryCommission of New South Wales, Sydney.


Smith, M.J. (1990) The role of bounties inpest management with special referenceto state dingo control programs.Unpublished Ass.Dip.Appl.Sc. thesis,Charles Sturt University – Riverina, WaggaWagga.

Smith, W.P., Borden, D.L. and Endres, K.M.(1994) Scent-station visits as an index toabundance of raccoons: an experimentalmanipulation. Journal of Mammalogy 75:637–647.

Soulé, M.E., Bolger, D.T., Alberts, A.C.,Wright, J., M., S. and Hill, S. (1988)Reconstructed dynamics of rapid extinc-tions of chaparral-requiring birds in urbanhabitat islands. Conservation Biology 2:75–92.

Southwell, C.N. (1987) Macropod studies atWallaby Creek. II. Density and distributionof macropod species in relation to envi-ronmental variables. Australian WildlifeResearch 14: 15–33.

Spira, H.R (1988) ‘Small companion animalreproduction behaviour and breeding’ in:Reproduction small companion animals:Proceedings of the J.D Stewart MemorialRefresher Course for Veterinarians,Sydney, No. 108, 8–12 August 1988,University of Sydney.

Stafford-Smith, D.M. and Foran, B.D. (1990)Rangepack: the philosophy underlyingthe development of a micro-computer-based decision support system for pas-toral land management. Journal ofBiogeography 17: 541–546.

Stains, H.J. (1975) Distribution and taxono-my of the Canidae. Pp. 3–26 in: M.W. Fox(ed.) The Wild Canids: Their Systematics,Behavioural Ecology and Evolution.Nostrand Reinhold Company, New York.

Sullivan, N.T., Rudder, T.H. and Holmes,W.E. (1992) Impact of younger turnoff onthe profitability and structure of cattleherds. M.T. Sullivan (ed.) Changing beefmarkets — should north Queensland pro-ducers change management to meetthem? Proceedings of the AustralianSociety of Animal Production 19: 53–54.

Tauchmann, A. (1998) Aversive conditioningtrial on dingoes using lithium chloride andthiabendazole. Unpublished report to theQueensland Department of Environment,Gatton College, University of Queensland.

Thompson, J. (1993) Report to the VertebratePests Committee (VPC) on the nationalstandardisation of ‘1080’ dose rates forwild dog and fox baits. Department ofLands, Brisbane.

Thompson, J.A. (1994) The management ofred foxes (Vulpes vulpes) in the northerntablelands of New South Wales.Unpublished M.Res.Sc. thesis, Universityof New England, Armidale.

Thompson, J.A. and Fleming, P.J.S. (1991)The cost of aerial baiting for wild dogmanagement in north-eastern New SouthWales. Australian Rangelands Journal 13:47–56.

Thompson, J.A. and Fleming, P.J.S. (1994)Evaluation of the efficacy of 1080 poison-ing of red foxes using visitation to non-toxic baits as an index of fox abundance.Wildlife Research 21: 27–39.

Thompson, J.A., Fleming, P.J.S. and Heap,E.W. (1990) The accuracy of aerial baitingfor wild dog control in New South Wales.Australian Wildlife Research 17: 209–217.

Thompson, R.C.A., Lymberry, A.J., Hobbs,R.B. and Elliot, A.D. (1988) Hydatid dis-ease in urban areas of Western Australia:an unusual cycle involving western greykangaroos (Macropus fuliginosus), feralpigs and domestic dogs. AustralianVeterinary Journal 65: 188–190.

Thomson, A.R. (1870) Guide to AustralianRemains Exhibited by the Trustees of theAustralian Museum. Australian Museum,Sydney.

Thomson, P.C. (1984a) Dingoes and sheep inpastoral areas. Journal of AgricultureWestern Australia 25: 27–31.

Thomson, P.C. (1984b) The use of bufferzones in dingo control. Journal ofAgriculture Western Australia 25: 32–33.


Thomson, P.C. (1986) The effectiveness ofaerial baiting for the control of dingoes innorth-western Australia. AustralianWildlife Research 13: 165–176.

Thomson, P.C. (1992a) The behavioural ecol-ogy of dingoes in north-western Australia.I. The Fortescue River study area anddetails of captured dingoes. WildlifeResearch 19: 509–518.

Thomson, P.C. (1992b) The behaviouralecology of dingoes in north-westernAustralia. II. Activity patterns, breedingseason and pup rearing. Wildlife Research19: 519–530.

Thomson, P.C. (1992c) The behavioural ecol-ogy of dingoes in north-western Australia.III. Hunting and feeding behaviour, anddiet. Wildlife Research 19: 531–542.

Thomson, P.C. (1992d) The behavioural ecolo-gy of dingoes in north-western Australia. IV.Social and spatial organisation, and move-ments. Wildlife Research 19: 543–564.

Thomson, P.C. and Marsack, P.R. (1992) Aerialbaiting of dingoes in arid pastoral areas withreference to rabies control. Pp. 125–134 in:P. O'Brien and G. Berry (eds) WildlifeRabies Contingency Planning in Australia.Bureau of Rural Resources ProceedingsNumber 11, Australian GovernmentPublishing Service, Canberra.

Thomson, P.C. and Rose, K. (1992) Agedetermination of dingoes from character-istics of canine teeth. Wildlife Research 19:597–600.

Thomson, P.C., Rose, K. and Kok, N.E.(1992a) The behavioural ecology of din-goes in north-western Australia. VI.Temporary extra-territorial movementsand dispersal. Wildlife Research 19:585–596.

Thomson, P.C., Rose, K. and Kok, N.E.(1992b) The behavioural ecology of din-goes in north-western Australia. V.Population dynamics and variation in thesocial system. Wildlife Research 19:565–584.

Thorne, A. and Raymond, R. (1989) Man onthe Rim: The Peopling of the Pacific.Angus and Robertson, Sydney.

Titcombe, M. (1969) Dog and Man in theAncient Pacific. Bernice P. Bishop MuseumSpecial Publication 59, Honolulu.

Tomlinson, A.R. (1954) Aerial baiting againstwild dogs and foxes in Western Australia.Journal of Agriculture Western Australia3: 37–49.

Tomlinson, A.R. (1958a) Bonuses for vermincontrol. Journal of Agriculture WesternAustralia 4: 465–469.

Tomlinson, A.R. (1958b) Wild dog control.Vermin Control Conference Perth 1–11.

Tomlinson, A.R. and Blair, C.K. (1952) Wilddog investigations. The Warburton Rangesarea—22 July to 29 August 1952. Journalof Agriculture Western Australia 1:883–893.

Triggs, B., Brunner, H. and Cullen, J.M.(1984) The food of the fox, dog and cat inCroajingalong National Park, south-east-ern Victoria. Australian Wildlife Research11: 491–499.

Troughton, E. (1957) A new native dog fromthe Papuan highlands. Proceedings of theRoyal Society of New South Wales(1955–56): 93–94.

Twigg, L., Eldridge, S., Shakeshaft, B.,Edwards, G., de Preu, N. and Adams, N.(1999) Longevity of 1080 meat baits incentral Australia. Unpublished final reportto National Feral Animal Control Program,Bureau of Rural Sciences.

Twyford, K. (1994a) Fraser Island dingomanagement plan: Community EducationProgram. Unpublished Draft Report to theDepartment of Environment, Queensland.

Twyford, K. (1994b) Investigations into thediet of dingoes on Fraser Island.Unpublished Second Interim Report to theDepartment of Environment, Queensland.


Tyndale-Biscoe, H. (1994) The CRC forBiological Control of Vertebrate PestPopulations: fertility control of wildlife forconservation. Pacific Conservation Biology1: 160–162.

Van Valkenburgh, B. and Koepfli, K.P. (1993)Cranial adaptations to predation in canids.N. Dunstan and M.L. Gorman (eds)Mammals as Predators. Symposium of theZoological Society of London 65: 15–37.

Wagner, K.K., Schmidt, R.H. and Conover,M.R. (1997) Compensation programs forwildlife damage in North America.Wildlife Society Bulletin 25: 312–319.

Walker, B.H. and Noy-Meir, I. (1982) Aspectsof the stability and resilience of savannaecosystems. Pp. 556–590 in: B.J. Huntlyand B.H. Walker (eds) Ecology of TropicalSavannas. Springer-Verlag, New York.

Walters, C.J. (1986) Adaptive managementof renewable resources. McGraw-Hill,New York.

Walters, C.J. and Holling, C.S. (1990) Large-scale management experiments and learn-ing by doing. Ecology 71: 2060–2068.

Wamsley, J. (1998) Rehabilitate for what?Workshop on Fauna HabitatReconstruction After Mining, Adelaide,10–11 October 1997:166–144.

Ward, F.G. (1986) The Vanishing Dogman.James Yeates Printing, Australia.

Wells, M.C. and Bekoff, M. (1981) An obser-vational study of scent-marking in coy-otes, Canis latrans. Animal Behaviour 29:332–350.

White, I.M. (1972) Hunting dogs at Yalata.Mankind 8: 201–205.

White, J.P. and O'Connell, J.F. (1982) APrehistory of Australia, New Guinea andSahul. Academic Press, Sydney.

Whitehouse, S. (1977) The diet of the dingoin Western Australia. Australian WildlifeResearch 4: 145–150.

Wilks, L. (1990) A Survey of the ContingentValuation Method. Resource AssessmentCommission Research Paper 2, AustralianGovernment Publishing Service, Canberra.

Williams, J.M. (1993) The factors and forcesshaping rabbit control policies and prac-tices in New Zealand: what lessons for thefuture? Pp. 42–50 in: B.D. Cooke (ed.)Australian Rabbit Control Conference.Anti-rabbit Research Foundation ofAustralia, Adelaide.

Williams, K., Parer, I., Coman, B., Burley, J. andBraysher, M. (1995) Managing VertebratePests: Rabbits. Bureau of Resource Sciences,Australian Government Publishing Service,Canberra.

Wilton, A. N., Steward, D. J. and Zafiris, K.(1999). Microsatellite variation in theAustralian dingo. Journal of Heredity 90:108–111.

Wong, D.H., Kirkpatrick, W.E., Kinnear, J.E.and King, D.R. (1991) Defluorination ofsodium fluoroacetate (1080) by microor-ganisms found in bait material. WildlifeResearch 18: 539–545.

Woodall, P.F. (1983) Distribution and popula-tion dynamics of dingoes (Canis famil-iaris) and feral pigs (Sus scrofa) inQueensland, 1945–1976. Journal ofApplied Ecology 20: 85–95.

Woodall, P.F., Pavlov, P. and Twyford, K.L.(1996) Dingoes in Queensland: skulldimensions and the identity of wildcanids. Wildlife Research 23: 581–587.

Wright, S.M. (1993) Observations of thebehaviour of male eastern grey kangarooswhen attacked by dingoes. WildlifeResearch 20: 845–849.

Zimen, E. (1976) On the regulation of pack sizein wolves. Zeitschrift fur Tierpsychologie 40:300–341.



Appendix A: Parasites and pathogens recorded from wild dogs inAustralia (after Corbett 1995a) and their potential effects onwildlife, humans and livestock (after Kelly 1977).

Organism

Cestodes

Echinococcus granulosus

Taenia ovis

Taenia hydatigena

Taenia pisiformis

Taenia serialis

Dipylidium caninum

Spirometra erinacei

Undeterminedspecies

Locality recorded

QueenslandNE NSWSE Highlands

Probably widespreadat very low levels

SE Highlands

Central AustraliaSE Highlands

SE Highlands

QueenslandSE Highlands

QueenslandCentral AustraliaSE Coastal

North AustraliaBarkly Tableland

Site of infection

Affectshumans?

yes

no

no

no

yes

yes

no

no

Affectswildlife?

no

yes

yes

no

yes

no

?

yes

Affectslivestock?

no

no

no

no

no

yes

yes

yes

Tapeworms

Small intestine

Small intestine

Small intestine

Small intestine

Small intestine

Small intestine

Small intestine

Small intestine

Nematodes

Uncinaria stenocephala

Ancylostoma caninum

Undeterminedspecies

SE Highlands

SE HighlandsQueenslandSE Coastal

North Australia

yes

yes

no no

no

no

no

no

no

Hookworms

Small intestine

Small intestine

Stomach/intestine


Organism

Toxacara canis

Undeterminedspecies

Dirofilaria immitis

Oslerus osleri

Trichurus vulpis

Locality recorded

SE Highlands

Central AustraliaNorth Australia

North AustraliaBarkly TablelandSE Highlands

SE Highlands

SE Highlands

Site of infection

Affectshumans?

yes

no

yes

no

no

Affectswildlife?

no

no

no

no

no

Affectslivestock?

no

no

no

no

no

Roundworms

Heartworm

Lungworm

Whipworm

Small intestine

Small intestine

Right ventricleand pulmonaryartery

Trachea

Caecum andlarge intestine

Acanthacephala sp. Barkly Tableland no no no

Thorn-headedworm

Stomach

Cyathospiruradasyuridis

SE Highlands no yes no

Spiruroid

Alimentary tract

Viruses

Protozoa

Paramyxovirus

Canine Hepatitis

Adenovirus

Isospora rivolta

Eimeria canis

Central AustraliaBarkly Tableland


SE Highlands

SE Highlands

no

no

no

no

no

no

no

no

no

no

no

no

Canine distemper

Coccidia

Respiratorytract

Liver

Alimentary tract

Nematodes


Organism

Protozoa

Sarocystis sp.

Giardia sp.

Locality recorded

QueenslandSE Coastal

SE Coastal

Site of infection

Affectshumans?

no

yes

Affectswildlife?

?

?

Affectslivestock?

?

?

Sarcosporida

Giardia

Insects

Striated muscleand heart muscle

Gastrointestinalsystem

Undetermined speices

Barkly TablelandCentral Australia

no no noSkin

Trichodectes canis SE Australia no no no

Biting lice

Skin

Ctenocephalidescanis

Probablywidespread

yes no no

Fleas

Skin

Ctenocephalidesfelis

Echidnophagamyrmecobii

Undetermined species

Probablywidespread


North Australia

yes

no

no

yes

no

no

no

no

noSkin

Skin

Skin

(Tabanidae)

Kangaroo flies

Widespread yes yes yes

Marchflies

Skin

(Hippoboscidae) SE Coastal no no noSkin

Mosquitoes

(Culicidae) Widespread yes yes yesSkin

Blowflies

(Calliphoridae) Widespread no no yesWounds


Organism Locality recorded

Site of infection

Affectshumans?

Affectswildlife?

Affectslivestock?

Ticks

Ixodes holocyclus SE Coastal high-lands

yes yes yesSkin

Rhipicephalus sanguineus

Central AustraliaNorth Australia

no no ?Skin

Amblyomma triguttatum

Queensland no yes ?Skin

Mites

Sarcoptes scabieri Widespread yes yes no

Sarcoptic mange

PentastomeArthropod

Skin

Linguatula serrata SE AustraliaCentral Qld

no ? ?

Tongueworm

Nasal cavities

Fungus

Microsporum canis Central Australia yes yes yes

Ringworm

Skin

Leech

Hirudo sp. SE Highlands yes yes yesSkin

Demodex folliculorum

Central Australia yes no no

Demodectic mange

Sebaceousglands and skinfollicles aroundeyes and nose

(after Andrew 1999)

The following recommendations, combinedwith available extension materials, will helpwild dog management coordinators incorpo-rate effective education and extension intotheir programs. The recommendations aredrawn from pest animal projects that were sup-ported under the Bureau of Resource Sciences’Vertebrate Pest Program (1993–1996) and areaimed at government officers, researchers,extension officers and community membersinvolved in group management. The recom-mendations are not in priority order althoughthey have been grouped so that similar issuesare listed together.

1. Wild dog management is essentially asocial issue.

While it is true that effective management ofwild dogs requires a certain baseline of sci-entific and technical information, the actualprocess of management is a social oneinvolving the ideas, activities and relation-ships of landholders and agency representa-tives. This means that there is often no pre-dictable and generalisable right or wronganswer to questions to do with management.The values and views of people should beconsidered and debated as part of the pro-cess of management.

2. Agency representatives who deal withlandholders need to have personal quali-ties, commitment and experience that arerespected by the landholders they areworking with. It is also important thatagency representatives see their interac-tions with landholders as long-term rela-tionships. Many landholders have seen a‘turnover’ of government staff and exten-sion officers and they sometimes have torepeat the same messages to consecutivegovernment representatives. Agency repre-sentatives also need to respect the personalqualities and experience of landholders.

This is perhaps one of the most crucialaspects of a project which involves differentinterest groups. Relationships between peo-ple are complex and interpersonal relation-ships are central to the success of projects.Projects and working arrangements can stallor fail because of personal clashes.Persistence and patience, conflict resolutionabilities and a sense of humour are importantattributes of good extension officers andother government officers. Extension andother government officers involved in landmanagement issues need to develop andsustain a genuine level of familiarity andtrust with the landholders that they seek tosupport. This genuine familiarity and trusttakes time to develop. Landholders live with-in social settings with often long-establishedcommunication patterns. It is unrealistic toexpect to be able to gain worthwhile insightsinto a social setting with a single, short visit.

3. Most projects and programs need to be conducted over long periods.Consequently, project managers need tobe responsive to change.

Most farmers and graziers view their enter-prise as a long-term business. Many are third,fourth or even fifth generation propertyowners and consider the significance of theirdecisions and practices on their children andgrandchildren. Some see governments asonly being interested in the short-term andwith little commitment to the long-term via-bility of the agricultural sector. Projects needto reflect a commitment to long-term out-comes and must consider changes that occurwithin government sectors and farming andgrazing communities.


Appendix B: Getting the best out of extension: 20 recommendations to help integrate education andextension into wild dog management

4. Landholders are ultimately responsiblefor ‘on the ground’ decisions about landmanagement involving their land. Inthe long run, landholders will determinewhether programs and actions succeedor fail.

Wild dog management within the agriculturalsector is mainly the responsibility of land-holders. Therefore they must have a majorrole in decision making if strategies are to besuccessful. Nevertheless, they do need toactively consult other groups and individualswith a major interest in the management ofwild dogs.

5. Management options for wild dogs mustbe seen to be practical by landholders.

Wild dog management options must be realis-tic with regard to equipment, labour require-ments and cost-effectiveness, and should fitwithin routine farm operations. These are thepractical considerations that landholders faceon a daily basis but are often overlooked byresearchers and/or agency representativeswho have different timetables.

6. Encourage the involvement of all stake-holders.

It is important to attempt to involve anyonewho is interested in land management issuesand also to attempt to seek opinions from awide range of community and other groups.Individuals within a community have diversetalents, interests and experience. Managers ofdifferent enterprises bring different perspec-tives to resource management issues. Womenhave contributed greatly to changes withinrural communities and it is now acknowledgedthat they are equal partners in many ruralenterprises. Elder members of communitiesshould not be forgotten, as they often knowmore about a particular district than anyoneelse. Different cultural groups and individualsoften have different traditional ways of know-ing and different ways of doing things thatbring a greater range of possible solutions toland management problems. Children also canbecome involved in projects through clubs,schools or through local events and activities. Iffunds are available, consider using an indepen-dent facilitator to manage diverse views or tobring diverse views together.

7. Identify and work with key players inthe community.

Within communities there are manylandowners who are respected for their landmanagement expertise and innovation.These people are often leaders in their com-munity. Working with these people will helpgain an insider’s view of local land manage-ment issues.

8. Land management issues attract arange of views: expect and respect diver-sity of opinion.

It is not reasonable to expect all landholdersand agency representatives to agree on anygiven issue. Aim for a give and take arrange-ment and value argument: it is a good way offinding out what people want and feel aboutan issue. It is reasonable for some people toconcede on some points but not on others.

9. The scientific community, supported bygovernment, should provide accessibleinformation. They should also involvelandholders and agency representativesin research and data collection.

Pest animal management requires supportinformation from research agencies and fromlandholders. If research is to have practicaloutcomes then it should include landholderinformation as part of the research process.Researchers should not regard landholdersas just data collectors. They have an essentialrole in ensuring that the research is testingpractical management options. Landholdersobservations may also provide useful direc-tions for research.

10. The rights of individual land ownershipshould be respected and data collectedfrom properties should be treated confi-dentially.

Landholders have the right to privacy and torespect of both their land and ‘intellectual’property rights. Release of data identifyingindividual landowners should be negotiatedwith those landowners in the first instance.Usually, it is the pooled information that is ofmost value to researchers and agency repre-sentatives so that individual landholders arenot identified.


11. Landholders and others often responddifferently to written information andquestionnaires than to conversations.

This may simply be because landholders’usual medium of communication is throughconversation rather than writing. This pointis important to remember when collectingdata about peoples’ opinions on land man-agement issues.

12. Sustained direct communication is themost valued and effective form of con-veying and receiving perspectives on themanagement of wild dogs.

Talking one-to-one with people is the mosteffective way of getting people involved andmaintaining their involvement in a program.Newsletters, field days, local shed meetings,media promotion and talks at local meetingsare also encouraged.

13. Access to information should be efficientand easy.

All people in a community should beinformed about and able to access informa-tion associated with land management pro-jects that is relevant to their district.Information could be presented via a num-ber of different means such as: local media,newsletters, mail-outs, presentations at meet-ings and informal gatherings. Individualsand groups should be provided with a con-tact list of people with various types ofexpertise and knowledge that they can callupon if required (for example, indigenous,historical, scientific, financial etc).

14. Consider ways of getting information tothe wider community and other interest-ed groups.

Many cheap and easily accessible informationchannels exist to inform other people aboutprojects. Local media will often respond to astory involving local landholders. Displays andnotices on bulletin boards and notice boards atlocal stores, halls and at special events willhelp spread the news. Write stories forLandcare and other rural newsletters and writeto or give talks at meetings of other groups inthe district. Word-of-mouth is one of the mostimportant ways of spreading the word aboutlocal projects.

15. Frequent, organised meetings thatinvolve committed representatives fromall stakeholder groups are necessary if‘on the ground’ change is expected asan outcome.

Information about the issues to be discussedshould be circulated to all stakeholders wellahead of meetings so that representativesmay consult with and receive advice fromtheir constituents. Representatives should bepeople within an organisation or district whohave a commitment to consultation and whowill give feedback to the people whoseviews they represent.

The management group size should bedetermined by the ability of those taking themain responsibility for pest management(those representing stakeholders) to feed-back information on a regular basis andreceive responses from others and to meetany queries from those whom they repre-sent. Mechanisms for informing all peopleshould be clearly thought out so that repre-sentatives do not merely ‘rubber-stamp’ pro-posals and actions that their constituentsmay not know about.

16. Meetings should always have a clearpurpose that relates to the needs of land-holders. They should be held at timesthat suit landholder schedules.

The best meetings are those that reflect wilddog management developments that haveoccurred on the ground between meetings,rather than those that can be perceived aspurely paper exercises. Meetings should alsobe held at times that are negotiated withlandholders. This includes consideration ofseasonal work, day of the week, time of day,and length of the meeting. Try to tie wild dogmanagement meetings into existing meet-ings such as landcare and bushfire meetings.Often a social event such as barbecue is auseful way of bringing people together. Thisalso gives people time to discuss issues in arelaxed atmosphere.


17. Wild dog management practices need tobe viewed within a broader legal andpolicy context.

Governments at all levels should providebroader framework considerations (for exam-ple Ecologically Sustainable Development,biological diversity, legislative requirements)within which on the ground pest managementpractices can be located and to which they cancontribute. Plain language versions of thesepolicy statements should be made readilyavailable to all participants.

18. All costs and benefits need to be considered.

Whatever technique is used to calculate costsand benefits it should be remembered that dif-ferent managers have different concerns andpriorities. Many aspects of land managementand broader lifestyle interests have values thatare difficult to quantify in monetary terms andcannot necessarily be generalised. A commondifficulty with attempts at cost–benefit analysisis the time it takes and the complexity of someformulas that seem irrelevant to landholders.This is so, particularly for those who considerthat the impact of any pest is worth managingor, at the other extreme, those that are com-pletely uninterested in pest managementregardless of the demonstrated economic ben-efit.

19. The pain and suffering of animalsshould be considered in decision making.

Ethical considerations are an aspect of pestanimal management that cannot be ignored.Many methods of pest managementinevitably inflict pain and suffering on ani-mals. Discussion about this issue and rele-vant information about various managementmethods must be considered as part of anypest animal management project if thesemethods are to be regarded as best practice.The most humane methods available shouldbe preferred.

20. Cultural and heritage issues should beconsidered in decision making.

Many land management decisions affect cul-tural and heritage sites. Consideration mustbe given to the relationship between culturaland economic interests as part of any pestanimal project.

Some suggested extension materials

a. National pest management guidelinesincluding this publication. Other guide-lines are available for rabbits, foxes, feralgoats, feral pigs, rodents, feral horsesand carp.

Australia’s Pest Animals: New Solutions toOld Problems (Olsen 1998)

New Approaches to Managing Pest Animals(an extension folder)

These are available from the Bureau of RuralSciences in Canberra (internet:http://www.affa.gov.au/outputs/ruralscience.html).

b. State and Territory agriculture and con-servation departments provide factsheetson various pest animal issues.

c. Specific local information can be obtainedfrom regional agricultural protection offi-cer and/or park service.


Peter Fleming

Peter Fleming is a senior research officerwith the Vertebrate Pest Research Unit ofNSW Agriculture. Since 1983, he has studiedthe impacts and management of flying foxes,bird pests, European red foxes, feral pigsand feral goats. Most of Peter’s research hasbeen in the highlands of eastern New SouthWales where he is primarily interested inassessing the impacts of wild dogs on live-stock production and the improvement ofmanagement strategies for wild canids. Peterhas published over 60 research and exten-sion articles about vertebrate pests. He iscurrently researching the interactions of feralgoats and domestic livestock for exotic dis-ease contingency planning, and is alsoresponsible for an integrated program forwild dog management in south-eastern NewSouth Wales and the Australian CapitalTerritory.

Laurie Corbett

Dr Laurie Corbett is a specialist in ecosystemassessment and management with over 30years field experience studying the ecologyof vertebrate predators and their prey, fireecology and the management of feral ani-mals. He has extensive knowledge of canids,felids, raptors and reptiles in Australia,Scotland and south-east Asia and is recog-nised as a world authority on dingoes. He isthe author or co-author of more than 140publications, of which over 50 specificallyrelate to wild dogs. Formerly with the CSIRODivision of Wildlife & Ecology, Dr Corbett iscurrently a principal ecologist with EWLSciences Pty Ltd, based in Darwin.

Robert Harden

Robert Harden is the Vertebrate Pests TeamLeader in the Biodiversity Research Group ofNSW National Parks and Wildlife Service. Hehas more than 30 years of research experi-ence of the ecology and management of wilddogs in the eastern highlands of NSW andhas published many research and extensionarticles on vertebrate pest management. Hisrecent research includes the development ofmanagement strategies for introducedrodents on Lord Howe Island, the rehabilita-tion of the Lord Howe Island woodhen, themanagement of feral goat impact in the east-ern highlands of NSW and the impact of foxbaiting on spotted-tailed quolls.

Peter Thomson

Peter Thomson is a research scientist withover 20 years field experience working onthe ecology and management of canids inAustralia. Much of his career was devoted toa long-term study of dingoes in the north-west and Nullarbor areas of WA, and in morerecent years, to various studies relating to thecontrol of foxes. He has published numerousarticles and scientific papers on dingoes andfoxes. He leads the Vertebrate Pest Researchgroup of Agriculture Western Australia and iscurrently involved in studies on a number ofdifferent pests.


Appendix C: Authors’ biographies

ANKC Australian National KennelCouncil

ANZFAS Australian and New ZealandFederation of AnimalSocieties

APB Agriculture Protection Board(Western Australia)

APCC Animal and Plant ControlCommission, South Australia

AGWEST Agriculture WesternAustralia

BRS Bureau of Rural Sciences(formerly Bureau ofResource Sciences)

CALMWA Department of Conservationand Land Management(Western Australia)

CPUE Catch-per-unit-effort

CWR Critical Weight Range (mammals weighing35–5500 grams)

DNA Deoxyribonucleic acid

ESD Ecologically SustainableDevelopment

GPS Global positioning system

NERDA New England RuralDevelopment Association

NFACP National Feral AnimalControl Program (a NHTprogram)

NHT Natural Heritage Trust

NLP National Landcare Program

NPWS National Parks and WildlifeService (New South Wales)

PWC Parks and WildlifeCommission (NorthernTerritory)

QDNR Department of NaturalResources (Queensland)

RLPB Rural Lands ProtectionBoard (New South Wales)

RSPCA Royal Society for thePrevention of Cruelty toAnimals (Australia)

SCARM Standing Committee onAgriculture and ResourceManagement

s.d. Standard deviation

SDP Stochastic dynamic pro-gramming

VPC Vertebrate Pests Committee

WDCA Wild Dog ControlAssociation (New SouthWales)

WDDB Wild Dog Destruction Board(western New South Wales)


Abbreviations and Acronyms

1080: Sodium fluoroacetate. An acutemetabolic poison without antidote; particularly toxic to canids.

Bait: Attractive substance fed to pest animalsthat can be used to carry a poison or contraceptive or to lure animals into traps.

Bait mound: Specialised bait station wherebait is buried in the centre of a mound offriable soil to minimise bait removal bynon-target animals.

Bait station: A specific site used for therepeated placement of bait.

Biocontrol (biological control): Control ofpest populations using a specific biological agent such as a virus, bacteriumor predator (for example, myxomatosis).

Biodiversity: Biological diversity. The natural diversity of living things, usuallydefined at three levels: genetic, speciesand ecosystem.

Blastocyst: Early, multicellular stage ofembryo development in marsupials atwhich development is postponed. Delayedimplantation of the blastocyst in the uterusallows birth to correspond with seasonalflushes of food or the loss of an embryo orpouch young.

Bounty (bonus): Predetermined reward paidon presentation of evidence (for example,scalp) of the destruction of an animal (forexample, wild dog).

Canid: A member of the family Canidaecomprising 13 species of wolves, jackals,dogs and foxes.

Carnivore: A flesh-eating animal or a member of the Order Carnivora.

Carnivora: Order of mammals includingwolves, dogs, foxes, otters, cats, weasels,bears, raccoons, civets and hyenas.

Carrying capacity: Density of an uncon-trolled population of animals that is in equi-librium with their natural resources andcompetitors.

Chromosome: Threadlike structure in thenucleus of a cell which carries the geneticmaterial (genes) of heredity.

Commensal dogs: Wild dogs (includingdingoes, domestic breeds and hybrids)living in close association with but inde-pendently of humans (for example, dingoes foraging in rubbish bins at camping grounds on Fraser Island).

Competition: A number of organisms of thesame or different species using commonresources that are in short supply (exploita-tion competition), or organisms seeking acommon and abundant resource harmingeach other in the process (interferencecompetition).

Conservation values: Values attributed tomaintaining biodiversity, including thepreservation of viable populations of nativespecies and natural communities over theirnatural range, preservation of wildernessand prevention of land degradation.

Conspecific: A member of the same species.

Contingent value: The unpriced value thatpeople place on maintaining things suchas open space, clean air and endangeredspecies. The social value of these things isoften poorly represented by the marketvalue of the land where they occur.Contingent value can be estimated by ask-ing consumers of their willingness to payto maintain a resource.

Correlation: Statistical relationship betweentwo or more variables where a change inone variable is reflected in a proportionalchange in the other.


Glossary

Critical weight range (CWR) mammals:Australian mammals with live weightsbetween 35 grams and 5500 grams thatare believed to be more vulnerable toextinction by predation.

Curvilinear relationship: Curved line rela-tionship between two or more variables.

Demographics: Statistics relating to popu-lation dynamics, including birth rates,mortality rates, age and sex ratios.

Density dependence: Regulation of thesize of a population by mechanisms thattend to retard population growth as densi-ty increases and enhance populationgrowth as density declines.

Dingoes: Native dogs of Asia selected byhumans from wolves. Present in Australiabefore domestic dogs. Pure dingoes arepopulations or individuals that have nothybridised with domestic dogs or hybrids.

Discount rate: The rate used to calculatethe present value of future benefits orcosts. It is calculated using the reverseequation to that used to calculate interestrates on invested money.

Dispersal: Movement of an animal awayfrom its birth or breeding site.

Dispersion: The spatial pattern of a popula-tion of organisms relative to one another.

Distribution: The geographical area (range)in which a group of organisms occurs.

Diurnal: Active during the day.

DNA: (deoxyribonucleic acid) The geneticmaterial in the cells of most living organ-isms, which is a major constituent of thechromosomes in the cell nucleus.

DNA fingerprinting (or mapping): Atechnique in which an individual's DNA isanalysed to reveal the pattern of geneticmaterial within particular segments. Thispattern is claimed to be unique to individ-uals, and closely related individuals havesimilar patterns.

Dogger, dogman: A pest controller whospecialises in the removal of wild dogs,usually by trapping or shooting.

Domestic dogs: Dog breeds (other thandingoes) selected by humans, initiallyfrom wolves, that usually live in associa-tion with humans. This selection processis ongoing.

Ecosystem: Ecological system formed byinteraction of living things and their envi-ronment.

Efficacy: Ability to produce desired effects.

Efficiency: The accomplishment of desiredeffects in relation to the effort (or cost)expended to produce those effects (oftenexpressed as a rate).

Endangered species: Species in danger ofextinction and whose survival is unlikelyif causal threatening processes continue tooperate.

Endemic: Limited to a certain region, coun-try or group.

Endemic disease: Disease that occurs in aregion or country.

Eradication: Permanent removal of all indi-viduals of a species from a defined area.

Exotic: Introduced from another country(for example, exotic species).

Exotic disease: Disease that does not occurin a region or country.

Extant: Still existing; not destroyed or extinct.

Extrapolation: Interpreting data beyond thedimensions within which it is collected (forexample, assuming conclusions drawn fromdata collected in one region will be relevantelsewhere).

Fecundity: The number of live births overan interval of time.

Feral: Domesticated species that has estab-lished a wild population.

Feral dogs: Wild-living dogs of domesticbreeds.


Fertility: The ability to produce young.

Free-roaming dogs: Dogs that are owned byhumans but not always restrained and soare free to travel away from their owner’sresidence (includes commensal dogs).

Friable: [Soil that is] easily crumbled.

Functional response: Relationshipbetween per capita food intake rate andfood availability.

Genotype: Genetic constitution of an organism.

Gestation: Pregnancy.

Global positioning system (GPS): Smalldevice that uses satellite signals to accu-rately locate the user’s position (latitude,longitude and altitude).

Gregarious: Living in groups.

Home range: Area that an animal (or group ofanimals) ranges over during normal dailyactivities. The boundaries of the home rangemay be marked (for example, wild dogs usescent marks) and may (see territory) or maynot be defended, depending on species.

Howl up: Wild dogs are lured to a hunterimitating the howling of a dingo.

Hybrids: Progeny resulting from the cross-breeding of two different species or sub-species and the descendants of crossbredprogeny (for example, dogs resultingfrom crossbreeding of a dingo and adomestic dog).

Hydatidosis: Disease caused by hydatidworm (Echinococcus granulosus) infec-tion.

Immunocontraception: A form of fertilitycontrol where a substance that triggers animmune reaction causes sterility orreduced fertility in affected animals.

Indices of abundance: Field signs that cangive a relative measure of dog abundance(for example, howls, fresh droppings,tracks, bait acceptance).

Ingested: Taken orally.

Intangibles: Values that cannot be numeri-cally quantified (for example, that forwhich it is difficult to estimate a moneyvalue).

Interference competition: see Competition.

Isopleth: A line drawn on a map throughpoints having the same numerical valuefor any element (for example, an isobarjoins points with the same barometricpressure).

Karyotype: Number and structure (genesequence) of the chromosomes in thenucleus of a cell. All the cells in an indi-vidual have the same karyotype (exceptfor sperm and egg cells).

Latent period: The time lag between anaction and a response.

LD50

: Dose (per kilogram of body weight)that will, on average, kill 50% of treatedanimals.

LD100

: Dose (per kilogram of body weight)that will, on average, kill 100% of treatedanimals.

Linear programming: A mathematicalmodelling approach that uses simultane-ous linear equations for optimising deci-sions under resource limitations. A linearprogramming problem has a linear objec-tive function (for example, to maximisewhole-farm gross margins from livestockproduction) and a set of linear constraints(for example, the density of wild dogs,enterprise type, labour and capitalresources) arranged in an array.

Linear relationship: Straight-line relation-ship between two or more variables.

Lure: Attractant (usually an odour-producingsubstance) which is used to enhance theeffectiveness of baiting programs or toattract an animal to a trap site.


M-44 ejector: mechanical ejector for deliveringencapsulated toxin to canids. The device istriggered by the canid pulling on bait materi-al connected to the ejector. This releases acompressed spring within the ejector whichdrives a rod through a toxin capsule, pro-pelling the contents of the capsule into theanimal's mouth. The advantages of this sys-tem are that devices can be left loaded assentinel stations, without degradation of thetoxin over time. The pull pressure can alsobe adjusted so that it is only likely to be trig-gered by canids and the device ensures thatthe full complement of toxin is ingested toreduce sub-lethal poisoning.

Marginal benefits: The shift in benefit val-ues that occur as incremental changes aremade in the factor(s) which affect level ofbenefits (for example, changes to live-stock losses that occur as wild dog densityis reduced).

Marginal costs: The shift in cost values thatoccur as incremental changes are made inthe factor(s) which affect level of costs(for example, changes in the cost of find-ing and removing an extra wild dog thatoccur as wild dog density is reduced).

Mark–recapture: Technique of live catch-ing, tagging, releasing and then recaptur-ing animals, and using a formula to esti-mate population size from the proportionof recaptured animals that are tagged.

Market failure: Occurs when resources arenot allocated efficiently through the use ofthe market, that is, when the costs andbenefits to society are not equated by thenatural market forces of supply anddemand (for example, unsustainable useof natural resources or development ofsocial inequities).

Mesopredator release: The process where-by the removal or loss of higher orderpredators results in the increase in abun-dance of and substitution by lower orderpredators. The negative impacts of meso-predators on small mammal populationsmay be greater than the higher orderpredators.

Monoestrus: Having a single oestrus periodin one sexual season or year.

Morbidity rate: Proportion of a populationaffected by disease for a given time inter-val. Usually expressed as a per capita rate.

Mortality rate: Proportion of a populationdying during a given time interval. Usuallyexpressed as an instantaneous per capitarate. In seasonal breeders such as dingoesmortality may be an annual death rate.

Nocturnal: Active at night.

Non-target (animal/species): Animal orspecies that is accidentally killed orinjured by a control measure (for exam-ple, domestic dogs or native wildlifecaught in wild dog traps).

Oestrus: The phase of the female reproduc-tive cycle when they are fertile and ovula-tion occurs, sometime referred to as ‘sexu-al heat’.

Opportunistic feeding: Non-selective feed-ing occurring when the opportunity arises.

Pack: A social grouping of canids, usuallygenetically related. Wild dogs may hunt as awhole pack, as a sub-group or individually.

Per capita: Per head of population (forexample, food consumption per sheep isper capita food consumption).

Pest: Harmful, destructive or troublesomeorganism.

Phenotype: The characters of an organismdue to the interaction of genotype andenvironment.

Population: Groups of animals of a particu-lar species occupying an area where theyare subject to the same broad set of envi-ronmental or management conditions.

Population dynamics: The process ofnumerical and structural change withinpopulations resulting from births, deathsand movements.


Population limitation: A factor is limitingif a change in the factor produces achange in average or equilibrium densityof a population. For example, predationby wild dogs may limit the density of aprey population if abundance of the preyis higher when wild dogs are absent.

Population regulation: A factor is regulat-ing if the percentage mortality that it caus-es increases with population density(sometimes called density-dependent reg-ulation). For example, a disease may regu-late wild dog abundance if it causes higherpercentage mortality as wild dog densityincreases.

Predator: An animal that kills other animalsfor food.

Prevalence: The number of instances of dis-ease, or related attributes (e.g. infection orpresence of antibodies) in a known popu-lation at a designated time, without dis-tinction between old and new cases.

Prey: An animal hunted or seized as food bya flesh-eating animal.

RANGEPACK: A computer software package(produced by CSIRO) with modules that aidmanagement decisions for livestock enter-prises, particularly for those in the arid zone.Modules include Herd-Econ, which modelsherd dynamics and property economics,Climate that uses past rainfall data to modelthe probability of rainfall events andPaddock, which predicts grazing patternswithin paddocks.

Pro-oestrus: Preparatory phase of the oestruscycle when the female reproductive systemis active but preceding ovulation (eggrelease).

Reactive control: Control activities inresponse to the presence of or damage byvertebrate pests.

Regression equation: An equation whichdescribes the relationship between two ormore variables.

Species: Group of interbreeding individualsnot breeding with another such group andwhich has characteristics that distinguishit from other groups.

Species-specific: Affecting only the target-ed species.

Standard deviation (s.d.): The standarddeviation of a sample is an estimate of itsvariability around a mean value, and iscalculated from the square root of thevariance (s2):

Strategic control: Using historical evidenceand current knowledge to devise strate-gies that prevent damage caused by verte-brate pests before damage commences.

Stochastic: Incorporating some degree ofnatural variation which has a mathemati-cally calculable probability.

Stochastic dynamic programming: Amathematical approach to modelling theeffects of problems (such as weeds, pestsor harvest rates) on production systems,which incorporates measured uncertainty(stochasticity) and the dynamics of thepopulation (of weeds, pests or resource).The objective function is often specifiedas the present value of the expectedreturns, which may include the decisionmakers' risk preference.

Subspecies: Group of individuals within aspecies, having certain characteristicswhich distinguish them from other mem-bers of the species, and forming a breed-ing group.

Surplus killing: Predatory activity where preyare attacked and killed in excess of theimmediate and short-term food require-ments of the predator (Kruuk 1972b).Surplus killing behaviour may result in anumber of surviving prey showing injuries.


Sustainability: Continuing in present formand at current level in the longer term.

Sustained control: Continued control inthe longer term.

Territory: The area occupied by an animal,or by a pair or group of animals, which itor they will defend against intruders.‘Territory’ and ‘home range’ are synony-mous for some canids.

Transect: A line (linear plot) through a studyarea on which data collection occurs.

Trap night: One trap set for one night (forexample, if three traps were set for twonights each, this would be six trap nights).

Top order predator: Animal at the top ofthe food chain. These animals are onlypreyed upon by other top order predatorsincluding humans.

Ungulate: Hoofed herbivore such as thehorse, goat, sheep, pig and antelope.

Unpriced value: Values for things that arenot exchanged in regular markets and assuch do not have a monetary price, forexample, scenery.

Varanid: The family Varanidae comprises agroup of about 30 species, generally knownin Australia as goannas.

Vertebrate: Animal with a backbone.

Wild dogs: All wild-living dogs, includingferal dogs, dingoes and hybrids.

Wilderness: Land that has been essentiallyunmodified since European settlement.

x-axis: The horizontal axis on a graph.

y-axis: The vertical axis on a graph.

Zoonoses: Diseases that are transmittedbetween animals and humans.


1

1080, 39, 65, 74, 87, 97, 98, 131, 136

clinical signs, 67

detoxification by soil organisms, 109

environmental fate, 109

human poisoning, 67

LD100,

99, 110

LD50,

99, 100, 110

LD99,

99

see also poisons

A

Aboriginal

mythology, 14, 50, 51, 64

people, 11, 14, 40, 50, 51, 60, 64, 68, 69,73, 76, 79

abundance, 125

absolute

estimation, 86

animal tracks, 85

estimator, 86

footprints, 84, 85, 101

index, 85

relative, 41, 85, 87, 98

Acts

Agriculture and Related ResourcesProtection Act 1976, 75, 81

Animal and Plant Control (AgriculturalProtection and other purposes) Act 1986,76, 81

Catchment and Land Protection Act1994, 77, 81

Companion Animals Act 1998, 51, 77, 81

Dog Control Act 1975, 81

Dog Control Act 1987, 77, 81

National Parks Act 1975, 77, 81

National Parks and Wildlife Act 1970,77, 81

National Parks and Wildlife Act 1974,77, 81

Nature Conservation Act 1980, 77, 81

Nature Conservation Act 1996, 81

Nicholson Land Act Victoria 1860, 43

Parks and Wildlife Conservation Act1993, 76

Pastures Protection Act 1939, 77

Robertson Free Selection Act NSW 1861,44

Rural Lands Protection Act 1985, 76, 81

Rural Lands Protection Act 1989, 77

Rural Lands Protection (Amendment)Act 1994, 77

Rural Lands Protection (Amendment)Act 1997, 77

Rural Lands Protection (Amendment)Act 1998, 77, 81

Territory Parks and WildlifeConservation Act 1993, 81

Threatened Species Conservation Act1995, 77, 81, 114

Wild Dog Destruction Act 1921, 74, 77,81

Wildlife Conservation Act 1950, 76, 81

Wildlife Protection (Regulation ofExports and Imports) (Amendment) Act1995, 50, 81

Adenovirus

see hepatitis

aerial observations, 86

African hunting dog, 24

agile wallaby, 21, 22, 56

agricultural impacts, 7, 8, 34, 87, 107, 108,112, 116, 123, 136, 139

Alice Springs, 15, 27, 58, 106

Amblyomma triguttatum

see ticks

Ancylostoma caninum

see hookworm

animal welfare, 7, 65, 66, 68, 83, 99, 113, 127,136, 137

Anseranas semipalmata

see magpie goose

Arabian wolf, 12


Index

arid, 21, 22, 26, 27, 28, 29, 30, 32, 35, 58, 59,60, 97, 136

areas, 20, 24, 31, 50

Arnhem Land Aboriginal Reserve, 76, 82

Asia, 11, 12, 13, 17, 19, 22, 38, 50, 52, 80, 138

Asian seafarers, 13, 38

Aujeszky’s disease, 53

Australian Capital Territory, 44, 51, 52, 65,98, 100, 101, 110, 115

distribution of wild dogs, 15

legislation, 77, 81

management of wild dogs, 73, 77, 79

Australian Hydatids Control andEpidemiology Program, 51, 52, 68, 115

Australian National Kennel Council, 51, 79,80

aversive conditioning, 78, 104

B

bait stations, 74, 85, 86, 101, 110

baiting, 22, 64, 65, 67, 68, 70, 84, 86, 87, 104,110

aerial, 36, 37, 52, 58, 69, 73, 75, 76, 79, 97,100, 101, 103, 107, 108, 109, 114, 115,117, 118, 121, 122, 126, 131, 132, 138

accuracy, 74

efficacy, 74

fixed-wing aircraft, 74, 102, 109, 117,121

helicopters, 74, 100, 102, 109, 117, 121

bait materials, 100

Doggone bait, 101

dried meat baits, 97, 101

moist meat baits, 101

decomposing baits, 102

ground baiting, 76, 78, 97, 117, 118, 126,132, 135

bait mounds, 101

burying baits, 101

free-feeding period, 101

replacement baiting, 101

strategies, 103, 138

timing and frequency, 102

Barkly Tableland (Northern Territory), 22,35, 54, 106

basenji, 13

bears, 61, 86, 107

benefit–cost

analysis, 132

comparison, 133

ratios, 132, 133

relationship, 125

bighorn sheep, 86

biting lice, 52

bounties, 50, 51, 60, 65, 73, 74, 75, 76, 77, 78,79, 102, 104

control value, 72

bovine hydatidosis, 51

branding percentages, 49, 130, 131

breeding

cycle, 33

pattern, 13, 32, 57, 80

Brucella canis (infective agent)

see canine brucellosis

brushtail possum, 20, 22, 110

Bubalus bubalis

see buffalo

buffalo, 25, 35, 60

buffer

areas, 78, 84, 103

zones, 52, 79, 103, 104, 108, 133

Bureau of Rural Sciences (BRS), 5, 6, 127,164

C

caching, 27, 100

Canidae, 11

canine brucellosis, 53

canine distemper, 34, 52, 53

canine hepatitis, 52

Canis latrans

see coyote

Canis lupus

see wolf

Canis lupus arabs

see Arabian wolf, 12

Canis lupus dingo (sub-species dingo), 5,11, 12

Canis lupus familiaris

see domestic dog


Canis lupus pallipes

see pale-footed wolf

carbohydrate, 22

Carnivora, 11

Carolina dog, 13

carrion, 21, 22, 23, 26, 36, 54, 55, 58, 60

catch-per-unit-effort (CPUE), 86

cattle, 6, 20, 21, 22, 23, 25, 26, 32, 35, 36, 41,42, 43, 45, 46, 48, 51, 52, 53, 54, 55, 57,58, 59, 60, 63, 64, 78, 79, 84, 87, 93, 100,103, 105, 106, 110, 112, 123, 124, 125,130, 131, 135, 136, 137, 139

‘nursery groups’, 47

protective behaviour, 49

strategic mating, 106

see also livestock

Chagas’ disease, 53

chromosomes, 11

Chrysomya bezziana

see screw-worm fly

climatic variability, 7

coat colour, 12, 13, 17, 19, 80

coccidiosis, 34

commensal, 11, 29, 77, 138

relationships, 12

commodity prices, 7

common brown dog tick, 53

Commonwealth Government, 5, 81, 115

community, 127

groups, 7, 65, 118

perceptions, 63

contingent

loss, 127

value, 113, 123, 139

control, 5, 7, 14, 20, 22, 34, 35, 36, 37, 38, 39,41, 44, 45, 49, 51, 52, 58, 63, 64, 65, 66,67, 68, 69, 72, 73, 74, 75, 77, 78, 79, 81,83, 84, 86, 87, 93, 96, 97, 98, 100, 101,104, 105, 106, 109, 110, 115, 116, 118,123, 129, 130, 136, 137, 138, 139, 140

activities, 6, 46, 76, 103, 108, 112, 114

cost of control, 46, 107, 108, 113, 124,126, 135

see also management and control

coot, 56

Coronaviridae

see transmissable gastroenteritis

techniques for measuring, 107

coyote, 28, 31, 61, 104, 105

critical body weight range (CWR) mammals,50, 51, 110

Crocuta crocuta

see spotted hyena

cumulative bait uptake, 86

cyanide, 86, 100, 110

see also poisons

Cygnus atratus

see swan (black)

D

Dasyurus maculatus

see spotted-tailed quoll

decision matrix, 118

decision-making framework, 116, 119, 122

dens, 14, 20, 27, 31, 39, 52, 53, 56, 58, 64, 84,104, 108, 116, 129

blocking of entrances, 68

characteristics, 33

location, 104

densities, 21, 25, 29, 34, 36, 38, 48, 58, 85, 86,135

dingoes, 20, 37, 39, 52

wild dogs, 14, 39, 52, 53, 56, 84, 108, 116,124, 125, 126, 129, 132

density–damage relationships, 124, 125, 126

Desmodus rotundus

see vampire bat

diet, 20, 21, 22, 23, 25, 26, 49, 52, 54, 57, 58

alternative food, 48, 132

dingoes

abundance, 14, 41

as a sub-species, 8

as an advertising image, 63

as an icon, 51

as an official breed, 51

as human food, 35, 50

as native fauna, 69

as pets, 40

attacks on humans, 60

Azaria Chamberlain case, 60, 64, 72


breeders, 80

communication, 54

howling, 30, 67

scent marking, 30, 31

conservation, 6, 8, 65, 70, 76, 79, 84, 115,116, 123, 127, 135, 140

issues, 7, 69, 136

objectives, 7, 139

priorities, 7

status, 7, 50, 83

values, 7, 50, 113

diet, 21, 22, 26

overlapping, 49

distribution, 6, 11, 12, 13, 14, 20, 41, 42,43, 48, 50, 51, 70, 102, 110

dominance hierarchies, 54

extinction, 39

female infanticide, 38, 39

foraging behaviour, 23

genetically pure, 6, 7, 12, 14, 15, 18, 19,32, 39, 40, 51, 63, 79, 80, 84, 123, 140

immigrant, 49

interactions with humans, 68

introduction to Australia, 13,14

legal status, 6, 75

movements, 37, 103

dispersal movements, 28

site fidelity, 28

numbers, 20, 28, 29, 36, 38, 39, 41, 50, 53,93

pelts, 50

population

assessment, 86

emigration, 38

immigration, 38

overestimates, 86

preservation societies, 79

repopulation, 37, 49, 103

reproduction

annual oestrus cycle, 32

average litter size, 32, 33

birth pulse, 33

gestation, 32, 33

pseudopregnancy, 32

seasonality, 32

sexual maturity, 32

testis weights, 32

role in functioning of ecosystems, 50

rival dingo groups, 26

size

average measurements, 17

ear length, 17

head length, 17, 19

hindfoot length, 17

tail length, 17

total length, 17

size of hunting groups, 23

taxonomy, 6

temporary breeding groups, 54

unpriced value, 113

weight, 17, 109

Dirofilaria immitis

see heartworm

disease, 25, 33, 34, 35, 38, 51, 52, 53, 68, 112,138

dispersal sink, 34

DNA, 86

fingerprinting, 12, 69, 80

sampling, 84

Dog Fence Board, 76

doggers, 67, 68, 73, 75, 78, 82, 85, 97, 102,104, 107, 120

dog-proof fence, 15, 74, 92, 94, 109, 116, 119

domestic dog, 6, 11, 12, 13, 14, 17, 24, 30, 32,33, 34, 35, 40, 52, 53, 68, 69, 76, 79, 99,107, 112

breeds

collies, 19

heelers, 19

kelpies, 19

commensal, 11

domestication, 11, 13

drought, 7, 22, 24, 26, 32, 33, 35, 38, 41, 42,50, 53, 54, 55, 57, 58, 60, 106

dusky rat, 20, 22, 35, 56

E

echidna, 14, 68, 95, 96


Echinococcus granulosus (causal agent)

see hydatidosis

ecological, 50, 61, 79, 129, 138, 140

balance, 93

Ecologically Sustainable Development (ESD)Strategy, 5

economic

assessments, 46

frameworks, 113, 124, 139

impacts, 9, 43, 44, 45, 46, 48, 49

loss to cattle industry due to wild dog predation, 54, 59

Ehrlichia canis

see tropical canine pancytopaenia

Eimeria canis

see coccidiosis

enterprise substitution, 105

Environment Australia (EA), 6

Erldunda (central Australia), 26, 57

European settlement, 6, 12, 14, 20, 38, 42, 52,54, 63, 69, 75

euro, 22, 23, 27, 29, 58

evolution, 11, 12

evolutionary lineage, 11

extension

officers, 116, 161

services, 6, 115, 116, 137, 161, 164

training, 116

F

fencing

barrier, 6, 14, 39, 64, 73, 74, 76, 82, 95, 96,107, 117, 119, 120, 122

electric

outrigger, 92, 94, 95, 96

exclusion, 20, 65, 68, 72, 93, 118

Dog Fence, 14, 58, 66, 73, 74, 76, 77, 79,104, 105

wire netting, 73, 92, 94, 95

see also dog-proof fence

maintenance, 95

marsupial netting, 95

rabbit-proof, 15, 74, 94

netting, 95

feral, 11, 12, 17, 19, 32, 33, 40, 51, 52, 63, 65,76, 77, 123

cats, 6, 41, 42, 54, 55, 138, 139

goats, 5, 50, 58, 164

horses, 5, 25, 30, 164

pigs, 5, 14, 22, 50, 60, 129, 164

rabbits, 5, 6, 20, 22, 23, 25, 28, 30, 35,38, 39, 41, 50, 54, 55, 56, 57, 58, 98, 106,109, 115, 129, 136, 164

fire, 38, 54, 56, 57, 58, 78

altered regimes, 35, 55

flush periods, 26, 33, 39, 58, 59

food supplies, 14, 29, 34, 35, 37, 38, 53, 56,61, 80, 103

Fortescue

River, 20, 23, 24, 26, 27, 29, 32, 33, 34, 35,37, 39, 41, 58, 97

study, 20, 23, 24, 26, 27, 29, 32, 33, 34, 35,37, 38, 39, 41, 42, 58, 97

fossils, 12, 13

foxes (European red), 5, 31, 42, 52, 58, 63,67, 68, 86, 95, 96, 98, 99, 100, 101, 103,109, 110, 139, 164

affect on abundance by wild dogs, 41, 50,54, 55, 70, 93, 138

as predators, 41, 55, 104

competition with wild dogs, 41, 54

Fraser Island, 15, 39, 51, 60, 61, 64, 80, 104,113, 140

free-roaming dogs, 12, 14, 29, 30, 52, 68, 69,104, 112, 140

Fulica atra

see coot

G

Gallinula mortierii

see Tasmanian native-hen

genetics

purity, 63, 80, 140

DNA testing, 12, 69, 80, 84, 86

see also dingoes (genetically pure)

government, 5, 6, 8, 22, 64, 65, 68, 75, 76, 77,79, 80, 98, 108, 118, 140, 161, 162, 164

agencies, 46, 72, 73, 81, 107, 113, 114,115, 116

roles and responsibilites, 7, 9


grazing pressure, 7, 139

grizzly bear, 86

group

approach, 7, 9, 129

formation, 23, 119

guard dogs, 13

see also sheep-guarding dogs

Gulf region, 22

Guy Fawkes River (northern NSW), 39, 58

H

habitat, 20, 34, 42, 55, 80, 86

Harts Ranges (near Alice Springs), 58

hauling, 12

heartworm, 34, 35, 52

hepatitis, 34, 52

herding, 12, 24, 25, 48, 49, 106, 123

Herpesviridae

see Aujeszky’s disease

Heterodoxus spiniger

see biting lice

Hippoboscid spp.

see kangaroo flies

home range, 30, 41, 104, 114, 118, 123

availability of resources, 27

boundaries, 27

natal, 28, 29

overlap, 27

size, 27, 28

hookworm, 34

hunting by dogs, 12, 13, 14, 26, 28, 29, 30, 31,33, 35, 38, 41, 51, 52, 53, 56, 57, 63, 68,72, 109

alone, 23, 24, 25, 124

birds, 21

nestlings, 25

newly fledged birds, 25

efficiency, 23

groups, 21, 23, 24, 25, 124

large kangaroos, 24

large ungulates, 25

peak in calving, 54, 106

strategies, 20, 22, 124

bailing up kangaroos, 23

pouncing, 25

success, 22, 24, 25, 33

tactics, 21, 25

unit, 49

hybridisation, 7, 12, 13, 14, 15, 18, 19, 30, 32,33, 36, 39, 40, 50, 51, 63, 69, 75, 76, 80,123, 136, 140

main processes, 40

prevention, 79

hydatid control program, 51

hydatidosis, 68, 137

in humans, 35, 51, 52

hydatids, 51, 68, 112, 115, 137

transmission, 52

foxes as definitive hosts, 52

macropods as intermediate hosts, 52

hyenas, 23, 24

I

implementation, 5, 7, 9, 73, 83, 113, 114, 115,116, 118, 129, 131, 140

indices

of abundance, 84, 85, 86, 103

relationship with abundance, 85

Indonesia, 13

insects, 20, 22, 35

island refugia, 80

Isospora rivolta

see coccidiosis

Ixodes holocyclus

see ticks

J

Jervis Bay, 17

K

kangaroo flies, 35

kangaroos, 14, 20, 26, 29, 30, 35, 36, 38, 41,42, 48, 49, 59, 63, 100, 106, 130, 139

autopsies, 24

drowning dogs, 24

eastern grey, 22, 23, 24, 56, 57

red, 20, 22, 23, 24, 25, 54, 58, 60

Kapalga (northern Australia), 29, 35, 39, 60

karyotypes, 11

kirri dog, 13

Kosciusko National Park, 39, 57


L

lactation, 20, 30

land

management, 87, 161, 162, 163, 164

agencies, 5

authority, 87

managers, 7, 8, 113, 114, 116, 124, 125,126, 127, 131

Landcare, 5, 115, 163

landholder survey/questionnaire, 45

legal, 5, 6, 65, 72, 75, 77, 100, 113, 164

constraints, 7

legislation, 7, 8, 50, 51, 63, 65, 72, 78, 80, 81,114, 115, 136, 140, 164

current, 75

lice, 34, 35, 52

lithium chloride, 78, 104

livestock, 6, 15, 21, 29, 35, 37, 43, 45, 46, 49,53, 55, 57, 63, 64, 65, 68, 75, 77, 78, 83,84, 87, 98, 102, 103, 105, 107, 108, 113,115, 116, 130, 132, 135, 136, 137, 140, 157

attacks on, 22

genetics, 7

harassment, 25, 26, 44, 48, 61, 93, 112

injuries, 88

castration, 48, 91

scrotal injuries, 48, 91

insurance and compensation schemes,108, 136

mismothering (cause of death), 48, 87, 93

production losses, 6

see also sheep, cattle

lizards, 22

longevity, 19

long-haired rat, 20, 21, 22, 35

lungworm, 34

Lycaon pictus

see African hunting dog

M

Macassan trepanger, 13

macropods, 22, 23, 24, 25, 29, 35, 38, 50, 51,52, 53, 54, 55, 56, 57, 58, 60, 64, 68, 94,95, 96, 139

Macropus agilis

see agile wallaby

Macropus giganteus

see kangaroos (eastern grey)

Macropus robustus

see euro

Macropus rufogriseus

see red-necked wallaby

Macropus rufus

see kangaroos (red)

magpie goose, 20, 22, 25, 56

maiming, 48

management and control

adaptive management

active, 112

passive, 112

biological control, 68, 107

molecular biology, 107

choice of control technique, 93

defining the problem, 112, 124, 130

developing a management plan, 113

flexible approach, 7

hormonal control, 68

humaneness, 66, 93

immunocontraception, 68

implementation, 7, 73, 113, 114, 115, 116,129

location of breeding dens, 104

management options, 9, 106, 116, 118,162

conservation, 9, 116

local eradication, 116

no control, 9, 87, 112, 130, 133

reactive management/control, 9, 73, 84,115, 116, 117, 118

strategic management/control, 5, 6, 7,8, 9, 74, 78, 102, 106, 111, 112, 113, 115,116, 117, 123, 136

management plan, 7, 9, 80, 112, 113, 114,115, 118, 123, 129, 136, 140

management strategies, 6, 8, 11, 64, 75,77, 93, 105, 106, 113, 118, 123, 129, 136,139, 140

operational monitoring, 129

performance criteria, 9, 113, 123, 129

performance monitoring, 129


preventative control, 84

scale of control, 49

maps, 9

regional, 87

Maranoa region, 22

marginal analysis, 108, 126, 128, 135, 139

mesopredator release, 110

monitoring and evaluation, 6, 7, 9, 34, 37, 39,83, 86, 93, 95, 104, 108, 110, 112, 114,116, 118, 123, 129, 130, 131, 135, 137

morbidity, 51

rate, 34

mortality, 33, 34, 35, 38, 53, 60, 138

mountain lion, 86, 107

musters, 87

N

Nadgee Nature Reserve, 56, 58

National Feral Animal Control Program(NFACP), 5, 6, 127

National Landcare Program (NLP), 5

native, 5, 12, 21, 50, 51, 56, 57, 63, 65, 67, 70,77, 79, 99, 101, 107, 113, 124, 136, 138,139

endangered vertebrates, 6

fauna, 6, 55, 69, 93, 140

remnant endangered populations, 49

mammals, 22, 55, 93, 106, 109

Natural Heritage Trust (NHT), 5

nature reserves, 56, 58, 64, 76, 77

neighbours, 75, 127, 131

cooperation, 95

New Guinea singing dog, 13

New South Wales, 6, 20, 23, 24, 25, 26, 27, 28,30, 33, 37, 39, 40, 42, 44, 45, 48, 51, 52,56, 57, 58, 64, 65, 66, 67, 69, 85, 87, 95,96, 98, 99, 100, 101, 102, 103, 104, 105,107, 108, 109, 110, 112, 114, 115, 116,117, 128, 132

distribution of wild dogs, 14, 15

legislation, 77, 81

management of wild dogs, 72, 73, 74, 77,78, 79

nomenclature, 11, 12

non-target

deaths, 100

poisonings, 97, 100

species, 49, 69, 83, 99, 101, 103, 109, 138

impact of control measures, 49

hazards to, 83

risks to, 109

trapping, 98

Northern Territory, 20, 22, 30, 46, 54, 56, 58,63, 64, 101, 106, 123


legislation, 76, 81

management of wild dogs, 74, 76, 79

Nullarbor Plains, 22, 23, 27, 29, 30, 35, 41

O

offal, 51

opportunistic feeder, 23

Oslerus osleri

see lungworm

Ovis canadensis

see bighorn sheep

P

Pacific islands, 13

packs, 20, 24, 27, 28, 29, 30, 31, 32, 34, 36,37, 38, 39, 53, 54, 56, 61, 84, 86, 103

pale-footed wolf, 12

Paramyxovirus

see canine distemper

parasites, 35, 35, 51, 52, 53, 68, 138, 157, 158,159, 160

partnerships, 9, 113

group control, 114, 123

parvovirus disease, 52

pastoral, 81

industry, 20, 38, 57, 73

lease, 21, 22, 64, 75, 76

pathogens, 34, 53, 157, 158, 159, 160

pay-off matrices, 126, 127

peak in calving, 54, 106

pest control agencies, 7

Petroi (north-eastern NSW), 58

Pilbara region, 48, 86

plagues, 22, 28, 35, 39, 56


poisons, 32, 34, 38, 65, 66, 67, 72, 73, 74, 75,93, 96, 98, 99, 100, 101, 103, 109, 110,136, 137, 138

baiting, 86, 100

bait-shyness, 100

cooperation between landholders, 74

detoxification, 99

humaneness, 67

M-44 cyanide ejector, 100, 110

Minty bait, 74

sub-lethal doses, 67, 99, 100, 110

population dynamics, 35, 37, 38, 128, 138

Potoroidae, 22, 55

predation, 7, 9, 26, 32, 35, 48, 49, 50, 51, 54,55, 56, 57, 60, 63, 64, 65, 73, 77, 78, 84,87, 89, 93, 96, 98, 103, 104, 105, 106, 107,108, 114, 116, 122, 123, 125, 129, 130,135, 136, 139, 140

on livestock, 6, 22, 43, 44, 46, 112, 113,115, 136

measuring the impact, 87

regulation or limiting of prey populations, 58

seasonal peaks, 45

predators, 22, 23, 24, 26, 35, 41, 49, 51, 53,60, 64, 84, 87, 93, 104, 138

selective, 23

predatory cycle, 56

prey, 7, 38, 48, 51, 58, 61, 84, 87, 102, 139

abundance, 20, 60, 136

anti-predator behaviour, 22

availability, 23

large, 23, 26, 29, 30, 33, 53

live, 22, 54

native, 21, 49, 56, 57

regulation, 50

small, 25

species, 20, 23, 27, 29, 35, 36, 54, 56, 57,93

switching, 108, 124

see also predation (on livestock)

problem definition, 6, 9, 112, 130

problem dogs/individuals, 98, 132

property inspection reports, 87

Pseudocheirus peregrinus

see ringtail possum

public

attitudes, 6

education, 70

health issues, 68

Puma concolor

see mountain lion

Q

Queensland, 6, 22, 34, 49, 50, 51, 58, 60, 64,65, 73, 80, 87, 99, 100, 101, 102, 105, 112,116, 123, 130, 131


legislation, 76, 77, 81


R

Rabbit Calicivirus Disease (RCD), 136

rabbit stickfast fleas, 35

rabbit-proof fence

see fencing (rabbit-proof)

rabbits, 5, 6, 20, 22, 23, 25, 28, 30, 35, 38, 39,41, 50, 54, 55, 56, 57, 58, 59, 94, 98, 106,109, 115, 128, 129, 136, 164

rabies, 53, 68, 112

transmission, 52, 138

radio-collar, 29, 45, 48, 86, 87, 97

rangelands, 52, 60

RANGEPACK, 106

rat kangaroo, 22

rats, 20, 21, 22, 35, 56

Rattus colletti

see dusky rat

Rattus villosissimus

see long-haired rat

recreational hunters, 51

red-necked wallaby, 22, 23, 57

regulatory infrastructure, 7

Rhabdoviridae

see rabies

Rhipicephalus sanguineas

see common brown dog tick

ringtail possum, 22

riverine areas, 20


Royal Society for the Prevention of Cruelty toAnimals (RSPCA), 40, 65, 67, 68, 80

S

sanctuaries, 76, 80

Sarcophilus harrisii

see Tasmanian devil

Sarcoptes scabiei (causal agent)

see sarcoptic mange

sarcoptic mange, 35

scat samples, 87

scavenging, 12, 22, 26, 54, 104

scent stations, 85, 86

scent-station index, 32

screw-worm fly, 53

seasonal conditions, 32, 48, 49, 105, 123

seasonal variation, 87

sheep, 25, 26, 34, 38, 41, 43, 44, 45, 46, 48,49, 51, 53, 54, 63, 65, 66, 72, 73, 76, 80,84, 86, 87, 95, 96, 97, 103, 104, 105, 107,108, 116, 123, 125, 127, 128, 130, 132,133, 135

grazing, 6, 14, 20, 78, 98, 112

measles, 52

merino, 21, 93, 127

recognising predation by wild dogs, 89,90, 91

see also livestock

sheep-guarding dogs

Anatolian karabash, 104

maremma, 104

shepherding, 73

shepherds, 43, 44, 73

shooting, 34, 38, 65, 73, 77, 93, 96, 117, 122,132

sign counts, 85

skull

morphology, 12

comparisons, 13

social

behaviour, 13, 30, 40

factors, 7, 29, 38, 127

organisation, 27, 28, 29, 30, 35, 49, 53,132

discrete territories, 29

distinct territories, 29

neighbouring packs, 29

spatial separation, 30

stable packs, 29, 30, 32, 34, 38, 54

temporal separation, 30

sodium fluoroacetate

see 1080

solitary dogs, 26

South Australia, 6, 20, 28, 29, 35, 39, 41, 58,66, 67, 81, 95, 97, 99, 101, 104, 105


legislation, 76

mallee, 19


spotted hyena, 23

spotted-tailed quoll, 42, 109, 138

bait uptake, 110

stakeholders, 68, 113, 114, 115, 118, 129,162, 163

Standing Committee on Agriculture andResource Management (SCARM), 5

stock losses, 69, 95, 104, 108

records, 87

strategic approach

see management and control (strategicmanagement/control)

strategic management

see management and control (strategicmanagementj/control)

strychnine, 22, 73, 74, 98, 99, 100, 130, 137

clinical signs, 67

success rate, 23, 24, 26

Sulawesi, 22

surplus killing, 24, 25, 26, 65, 108, 124

surveys, 39, 44, 45, 46, 48, 51, 63, 86, 87, 96,108

swamp wallaby, 20, 22, 23, 57, 58

juvenile, 25

pouch young, 24

swan (black), 25, 56

sylvatic cycles, 51, 52, 53

T

Tachyglossus aculeatus

see echidna


Taenia ovis

see sheep (measles)

Taenia pisiformis

see tapeworm

tapeworm, 35

Tasmania, 14, 20, 49, 52, 53, 54


legislation, 77, 81

management of wild dogs, 77

Tasmanian devil, 53

Tasmanian native-hen, 49

Thailand, 12, 13, 22, 32, 35

thylacine, 53, 54

Thylacinus cynocephalus

see thylacine

ticks, 34, 53

tourism, 50, 51, 123

Townshend Island, 50

toxic collars, 104, 105, 137

transmissable gastroenteritis, 53

trapping, 34, 36, 38, 66, 72, 73, 75, 77, 79, 86,93, 96, 97, 101, 108, 109, 118, 120, 121,122, 132

decoy (odours or carcasses), 98

padded-jawed traps or snares, 66, 98, 95

targeting specific individual dogs, 102

trap-shy, 98

Trichodectes canis

see lice

Trichosurus vulpecula

see brushtail possum

Trichurus vulpis

see whipworm

tropical canine pancytopaenia, 53

Trypanosoma cruzi

see Chagas’ disease

U

United States of America, 100, 104, 105, 107

Ursus arctos

see grizzly bear

Ursus spp.

see bears

V

vampire bat, 53

Vertebrate Pests Committee (VPC), 5, 7, 72

Victoria, 19, 20, 29, 30, 44, 45, 48, 51, 52, 64,65, 66, 67, 68, 85, 100, 101, 104, 107, 110,115


legislation, 77, 81

management of wild dogs, 72, 73, 75, 77,79

Vietnam, 12, 13

Vombatus ursinus

see wombat (common)

W

Wallabia bicolor

see swamp wallaby

wallabies, 14, 20, 21, 22, 23, 24, 29, 56, 57,58, 130, 139

Wallaby Creek, 23

water

artesian bores, 38, 55, 57

intake, 20

needs, 19

permanent sources, 24

supplementary, 55

turnover, 20

watering points, 14, 20, 30, 35, 41, 106

Western Australia, 20, 21, 23, 27, 28, 29, 30,35, 41, 44, 48, 49, 58, 64, 67, 86, 87, 97,99, 100, 101, 102, 103, 104, 105, 108, 109,116


legislation, 75, 76, 81


Western Division, 14, 77, 78

whipworm, 34

wild dogs

agricultural impacts, 112

as human food, 35

assessment of age, 19

bacula, 19

canine teeth (dingoes), 19

cementum bands (teeth), 19


closure of foramen, 19

eruption pattern of adult teeth, 19

tooth wear, 19

attacks on humans, 78, 113

culling, 78

diet, 20, 21, 22, 56, 58

analysis, 87

stomach and faecal samples, 20

emigration, 108

environmental impacts, 9, 49, 50

feeding ecology, 22

genetic evaluations, 39

human and animal health impacts, 112

interactions with native prey, 56

population dynamics, 37

refugia, 105

troublesome individuals, 98, 132

wildlife parks, 51, 76

wolf, 5, 11, 12, 24, 29, 31, 35, 36, 38, 104, 107

wombat (common), 21, 22, 57, 63

Z

zoonoses, 53

zoos, 51, 76


Managing the Impacts of Dingoes and Other WildDogs is the first book to provide a comprehensivereview of the history and biology of wild dogs in Australia, the damage they cause, andcommunity attitudes to their management.

Australia's wild dogs include dingoes, introducedaround 4000 years ago, feral domestic dogs andhybrids between the two. They are widelydistributed throughout Australia. Predation andharassment of stock by wild dogs causes millionsof dollars worth of losses to Australian sheep,cattle and goat producers each year. There arealso opportunity costs in areas where sheep arenot grazed because of the high risk of wild dogpredation. For this reason, wild dog control is asignificant expense for many pastoralists andgovernment agencies. Yet conservation of puredingoes is also important and is threatened bytheir hybridisation with feral domestic dogs on themainland.

Key strategies for successful wild dog managementare recommended by the authors, who are scientificexperts on wild dog management. The strategiesare illustrated by case studies.

Managing the Impacts of Dingoes and Other WildDogs is an essential guide for policy makers,pastoralists, conservation reserve managers andall those interested in wild dog management.

Managing the Impacts of

Managing the Im

pacts of Dingoes and O

ther Wild D

ogs

A G R I C U L T U R E , F I S H E R I E S A N D F O R E S T R Y - A U S T R A L I A

Dingoes and Other Wild Dogs

managing the impacts of dingoes and other wild dogs...managing the impacts of dingoes and other wild...

Documents