Order in the Pack - Western Sydney

Order in the Pack:

Ecology of Canis lupus dingo in the

Southern Greater Blue Mountains

World Heritage Area

A thesis submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Natural Sciences

College of Science, Technology and Environment

University of Western Sydney

Hawkesbury

© Brad Vincent Purcell, 2008

All photos © Brad Purcell unless stated otherwise

The story of a woman on the morning of a war

Remind me if you will exactly what we’re fighting for

Calling, calling for something in the air

Calling, calling I know you must be there

I don’t want to be your little research monkey boy

The creature that I am is only going to destroy

Throw me to the wolves because there’s order in the pack

Throw me to the sky because I know I’m coming back.

Red Hot Chili Peppers, ‘Easily’, Californication 1999

Acknowledgements

Dr. Robert Claude Mulley. Talk about philosophy. I was interviewed by PhD student

Susan Mowbray for her study on postgraduate experiences during doctoral studies and

a part of the interview was to somehow, draw my relationship with my principal

supervisor. So I did. I drew a bottle of Wynns Coonawarra Cabernet Shiraz Merlot, a

huge fireplace, stars and two stick figures sitting beside each other with the fire and a

cheese platter. Rob made everything easy and fun. He listened and most importantly,

he provided endless support and advice. It doesn’t get any better than that!

Acknowledgements

Co-supervisors Robert Close (UWS) and Peter Fleming from the Vertebrate Pest

Research Unit at Orange also imparted constructive criticism of the highest quality.

Their enthusiasm and support was endless and they have been an honour to work

with. The enlightened perspectives and enthusiasm of Andrew Glover from the Moss

Vale Rural Lands Protection Board, the man who conceived the idea for this project,

became the backbone of its stability. Duncan Scott-Lawson and Andrew Simson from

the Parks and Wildlife Division of the Department of Environment and Climate

Change and the Sydney Catchment Authority respectively, also cradled this study

through their system to a happy ending. Sue Cusbert from UWS had a similar effect,

working behind the scenes to ensure all equipment was ordered and functional…and

everything else.

Many other staff from partner organisations are deserving of many thanks for funding

and in-kind contributions including field accommodation, aerial services, boat

services on Lake Burragorang and the willingness to participate. These include Bob

Conroy, Chris Banffy, Kim de Govrik, Geoff Ross, Glenn Meade, Mick English,

Acknowledgements

Steven Mills, Joe Corby, Adrian Johnstone, Andrew Horton, Peter Stiff, Dave Young,

Dee Campbell, Kylie Madden and flight team staff from DECC. From SCA, Brian

Waldron, Glen Capararo, Tony Kondek, Peter Cox, Vicky Whiffin, Loretta Gallen,

Jane MacCormick, Dennis Ashton, Martin Krogh, Ian Wright, Martin Gilmore, Brian

Davies, Terry Keogh, Ken Bailey, Ugo Mana, Charles Starr, Keith Maylin, James Ray

and James O’Keefe. Rural Lands Protection Board trappers Bill Morris, Mick Davis

and Lee Parker, ranger Nev Collins and free lance trappers David Jenkins and Andrew

McDougall all deserve many thanks for their valiant efforts. I also wish to thank

veterinarian Rus Dickens for teaching me how to extract blood from a dog.

The scope of this study was so broad that much assistance was required for data

analysis. Barbara Triggs (Dead Finish) went through a lot of faecal matter for me and

I am very grateful. Justyna Paplinska (Genetic Technologies Ltd.), Alan Wilton

(University of New South Wales) and George Sofrinidis (Genetic Technologies Ltd.)

ran expert analyses on the genetic structure of the animals. Oleg Nicetic (UWS) was

always available and very patient with statistical tests and I am also very thankful to

Michael Dingley (UWS) for preparing skulls for analyses. To Ben Allen (South

Australian Arid Lands Natural Resource Management Board) I am also grateful for

his voluntary analysis of the home range data.

Every single volunteer that spent a week in the field collecting dingo scats or walked

for kilometres through dense bush to retrieve dingo skulls or GPS collars deserve so

much recognition. They all contributed to the project in many unique ways. From

first appearance until last: Glenn Purcell, Chad Weston, Andrew Murray, William

Littleton, Nina Jørgensen (Danish), Daniel Kusche (German), Dianne Mulley, Adam

Mulley, Aran Anderson, Joseph Horvath, Christine McCormack, Jon Reid, Angelo

Acknowledgements

Indorato, Jason Flesch, Jamie Cross, Lowell Steffan, Kerrie McGuigan, Loren

Howell, Lynn Bowden, Wendy Fairs, Mick Fairs, Jann Fathers, Mark Pont, Tom

McGuigan, Iris Schöberl (Austrian), Ngaire McCubben, Janice Reid, Martin Convey,

Chris Williams, Adam Bryce, Bryan Purcell, Stuart Sutton, Andrew Riley, Ibrahim

Imseis, Kingjoe Rahe, Paul Mulley, Kristy Anderson, Simon Purcell, Jack Pascoe,

Julie Langford, Lindy Duncan, Gene Ramirez, Josh Edwards and Jade Wittman.

I am also indebted for the constant support from family and friends. My parents,

Bryan and Vicki Purcell, have been more than devoted to supporting my chosen style

in life. They made the choice seem more instinctual because of the childhood

experiences from living near the bush and interacting with nature. My sister Kristy

Anderson, my brother Simon Purcell and my brother in law Aran Anderson were

always there for me. I also wish to thank Kerrie McGuigan for her support and

especially for continuously stocking my refrigerator with chocolate and ice cream;

and her parents Wendy and Anthony McGuigan.

My gratitude is also extended to Andrew Murray. He volunteered on five field trips

and provided many jokes which helped keep me less stressed. On behalf of most

volunteers, I have to thank him for recording “The Catchment Mix” audio tapes. All

volunteers sang or danced to the songs and the nostalgia I get when I hear a song from

catchment is the best. I am sure some of the volunteers would feel the same.

I wish to thank the University of Western Sydney for the thesis opportunity and

experiences, especially access to a vehicle and office space, research funds and funds

to attend conferences. Thanks to all of the friendly and supportive staff including

Lindy Duncan, Grace Brown, Sue Reed, Sandra Tuckwell, Helen Reid, Peter Ward,

Acknowledgements

Jack Wolfenden, Burhan Amiji, Michael Franklin, Gavin Beveridge, Karen

Stephenson, Kerrie Bradley, Zina O’Leary, Gilbert de Chalain and Bruce Simmons,

Chris Derry, Charles Morris, John Bartlett and Shelley Burgin. Finally, I extend my

gratitude to postgraduate colleagues including Larissa Borysko, Laura Parker, Sharon

Kemp, Dianne Cooke, Carmina Rositano and Greg McKean and Jack Pascoe.

It was such a pleasure to associate with everybody listed during this study and others

that I may have forgotten. Thank you very much and I wish you the best of luck in

your own adventures.

vii

Statement of Authentication

The work presented in this thesis is, to the best of my knowledge, original

except as acknowledged within the text. I hereby declare that I have not

submitted this material, either in full or part, for a degree at this or any

other institution.

(Signature) (Date)

12 November 2009

Table of contents

viii

Table of contents

Acknowledgements .....................................................................................................iii

Table of contents .......................................................................................................viii

List of figures ............................................................................................................... ix

List of tables ............................................................................................................... xiv

List of plates ............................................................................................................... xvi

Initialisms and abbreviations .................................................................................. xvii

Abstract ....................................................................................................................xxiii

Introduction .................................................................................................................. 1

Dingo ecology, study site and research

Chapter one:

framework ....... 4

Chapter two:

What is a “pure” dingo? ................................................. 63

Chapter three:

Interactions between sympatric competitors and their prey ................................................................................ 129

Chapter four:

Intraspecific variations in spatial organisation, movement and activity ................................................. 176

Chapter five:

Effects of cultural transmission and the proposal for management................................................................... 238

References ............................................................................................ 256

Appendix .............................................................................................. 279

List of figures

ix

Chapter one:

List of figures

Figure 1.1: Phylogenetic tree displaying the relatedness of the contemporary

canids ...................................................................................................................... 11

Figure 1.2: Potential subspecies of the Gray wolf ................................................... 12

Figure 1.3: Current known world distribution of dingoes ..................................... 13

Figure 1.4: Potential distribution of wild dogs, excluding foxes, in Australia ..... 15

Figure 1.5: Location of the SGBMWHA in relation to Sydney. ............................ 36

Figure 1.6: Location of Site 1 / transect 1 and Site 2 / transect 2 in the

SGBMWHA ........................................................................................................... 46

Figure 1.7: Site 1 road infrastructure, political boundaries and Lake

Burragorang .......................................................................................................... 47

Figure 1.8: Site 2 road infrastructure, political boundaries and Lake

Burragorang and Wollondilly River ................................................................... 49

Figure 1.9: Experimental approach used in the current study to comprehend

aspects of the ecology of dingoes in the SGBMWHA

......................................... 52

Chapter two:

Figure 2.1: Locations of dingoes showing “hybrid” canonical scores due to

selection pressure from being persecuted and locations of dingoes showing

“pure” canonical scores due to natural selection pressure ................................ 73

Figure 2.2: Comparison of neonate canid skull with an adult canid skull showing

the profound developmental alterations that occur from neonate dog to adult

dog ........................................................................................................................... 76

Figure 2.3: Expanded view of neighbour-joining tree of domestic dog haplotypes

within clade I (of IV) to illustrate the apparent position of the dingo amongst

domestic dog breeds. ............................................................................................. 78

Figure 2.4: Expanded view of clade I (of IV) of neighbour-joining tree of 8 wolf

and 15 dog genotypes within clade I (of IV) to illustrate the apparent position

of the dingo amongst domestic dog breeds.......................................................... 79

List of figures

x

Figure 2.5: Results of genetic analyses for calculated hybrid mixture proportions

with reference domestic dog genes and reference captive dingo genes ............ 89

Figure 2.6: Results of genetic analyses for 47 capture/release, one deceased and

two recapture/ release dingoes from 2004 – 2007 ............................................. 100

Figure 2.7: Determining probability of K and identification of K = 2 as the most

probable number of genetic clusters or K = 8 using the second order rate of

change ................................................................................................................... 105

Figure 2.8: Population one and population two identified using STRUCTURE at

K = 2. ..................................................................................................................... 108

Figure 2.9: Representation of capture location and relatedness data for

population one and population two within the SGBMWHA .......................... 109

Figure 2.10: Representation of subpopulations from population one and

population two using K = 8 in STRUCTURE. ................................................... 111

Figure 2.11: Representation of capture locations for subpopulations and

relatedness data for population one and population two within the

SGBMWHA using K = 8. .................................................................................... 112

Figure 2.12: Genetic samples collected outside of the SGBMWHA study site ... 114

Figure 2.13: Relationship between “purity” and relatedness of SGBMWHA

dingoes

.................................................................................................................. 115

Chapter three:

Figure 3.1: Frequency distribution of dingo and fox scats collected monthly for

26 consecutive months at Site 1 and Site 2 ........................................................ 138

Figure 3.2: Proportion of prey for dingoes and foxes in Sites 1 and 2 of the

SGBMWHA. ........................................................................................................ 141

Figure 3.3: Changes in the proportion of prey for dingoes and foxes during

breeding, whelping, rearing and exploratory biological seasons in Sites 1 and

Site of the SGBMWHA during 2005, 2006 and 2007

Figure 3.4: Significance of dietary items for dingoes during breeding, whelping,

rearing and exploratory biological seasons for 2005 and 2006 ....................... 144

....................................... 142

Figure 3.5: Significance of dietary items for foxes during breeding, whelping,

rearing and exploratory biological seasons for 2005 and 2006 ....................... 146

List of figures

xi

Figure 3.6: Comparison of mean occurrence of prey categories in dingo and fox

scats in the SGBMWHA during breeding, whelping, rearing and exploratory

biological seasons during 2005, 2006 and 2007 ................................................. 147

Figure 3.7: Use of the Predator Diet Index to illustrate annual changes in the

monthly occurrence of swamp wallaby and brushtail possum within dingo diet

at Site 1 and Site 2 and fox diet at Site 1 and Site 2 from analyses of hair and

bone fragments in faecal material collected over 26 months. ......................... 149

Figure 3.8: Comparison of monthly “abundance” using the Abundance Rating

Formula with monthly “activity” using the Passive Activity Index as

determined using the sand plot technique for Site 1 dingoes, Site 2 dingoes,

Site 1 foxes, Site 2 foxes, Site 1 cats and Site 2 cats between March 2005 and

April 2007 ............................................................................................................. 150

Figure 3.9: Monthly activity of dingoes, foxes and cats as determined using the

sand plot technique and the passive activity index in Site 1 and Site 2 from

March 2005 until April 2007 .............................................................................. 151

Figure 3.10: Comparison of changes in “passive activity” for Site 1 dingoes with

Site 1 foxes, Site 2 dingoes with Site 2 foxes, Site 1 dingoes with Site 1 cats, Site

2 dingoes with Site 2 cats, Site 1 foxes with Site 1 cats and Site 2 foxes with

Site 2 cats .............................................................................................................. 152

Figure 3.11: Relationship between eastern grey kangaroo PAI and dingo PAI for

Site 1 and Site 2.................................................................................................... 153

Figure 3.12: Relationship between common wombat PAI and dingo PAI for Site

1 and Site 2 ........................................................................................................... 153

Figure 3.13: Relationship between brushtail possum ARF and dingo ARF in Site

1 and Site 2 ........................................................................................................... 154

Figure 3.14: Comparison of scats collected per month and passive activity per

month for Site 1 and Site 2 dingoes and Site 1 and Site 2 foxes of the

SGBMWHA ......................................................................................................... 158

Figure 3.15: Relationship between eastern grey kangaroo PDI and dingo ARF

for Site 1 and Site 2 ............................................................................................. 159

Figure 3.16: Relationship between brushtail possum PDI and dingo PAI for Site

1 and Site 2 ........................................................................................................... 159

Figure 3.17: Relationship between dingo PAI, dingo biological seasons and

swamp wallaby PDI from dingo scats from Site 1 with eastern grey kangaroo

List of figures

xii

PDI from dingo scats, Site 2 eastern grey kangaroo PDI from dingo scats, Site

1 brushtail possum PDI from dingo scats and Site 2 brushtail possum PDI

from dingo scats ................................................................................................... 160

Figure 3.18: Relationship between fox PAI, fox biological seasons and swamp

wallaby PDI from fox scats sampled in Site 1 with eastern grey kangaroo PDI

from fox scats, Site 2 eastern grey kangaroo PDI from fox scats, Site 1

brushtail possum PDI from fox scats and Site 2 brushtail possum PDI from

fox scats ................................................................................................................ 161

Figure 3.19: Relating dingo and fox activity with prey activity and the

occurrence of swamp wallaby, eastern grey kangaroo, brushtail possum,

European rabbit and common wombat as dietary items.

............................... 163

Chapter four:

Figure 4.1: Encounter history with known dingoes from April 2005 until July

2008 ....................................................................................................................... 195

Figure 4.2: Separate maps of total data points from retrieved GPS collars for

each dingo and all dingoes on one map ............................................................. 197

Figure 4.3: Comparison between 100% MCP, 95% MCP, and 50% MCP with

90%-50% kernel contours and raw data points for male dingo 2.4 during

breeding season 2006 ........................................................................................... 199

Figure 4.4: Home range of dingo 2.25 including and excluding extraterritorial

movements............................................................................................................ 199

Figure 4.5: Comparison of 95% Minimum Convex Polygons with 90% kernel

contours for all dingoes ....................................................................................... 201

Figure 4.6: Spatial separation of 50% MCP core areas of activity and their

relationship with topography and Lake Burragorang .................................... 203

Figure 4.7: Comparison of 50% MCP home range estimates with 100% MCP

home range estimates for seasonal differences ................................................. 205

Figure 4.8: Monthly GPS locations around 50% MCP for female dingoes 1.4 and

2.10 and male dingo 2.4 which highlight movement patterns associated with

whelping, rearing, exploratory and breeding seasons ..................................... 207

List of figures

xiii

Figure 4.9: Representation of the potential movements of dingo 2.4 in

comparison with how movements are presented using GIS during his

extraterritorial foray from 15 th-18th March 2006 ............................................. 208

Figure 4.10: Extraterritorial movements of dingo 2.25 during breeding season

2007 ....................................................................................................................... 209

Figure 4.11: Movements of dingoes 2.21 and 2.20 around their 50% MCP core

area during an identified border patrol ............................................................ 210

Figure 4.12: Comparison of sand plot activity with activity determined GPS

locations for Site 1 and Site 2 annual collars .................................................... 211

Figure 4.13: Mean monthly activity plots for all dingoes collared with annual

GPS collars ........................................................................................................... 212

Figure 4.14: Mean speed travelled by four male dingoes and one female dingo

every hour and every 10 minutes ....................................................................... 213

Figure 4.15: Proportion of time spent resting, walking, trotting and running for

dingoes 2.20 and 2.21 .......................................................................................... 214

Figure 4.16: Activity patterns showing minimal activity increased activity for

dingoes 2.20 and 2.21 .......................................................................................... 215

Figure 4.17: Proportion of time spent resting, walking, trotting and running and

example of low activity exhibited by dingo 2.23 ............................................... 216

Figure 4.18: Percentage of time spent resting, in low activity, in crepuscular

activity and in increased activity for dingoes 2.21, 2.23 and 2.25 who lived in

territories adjacent to one another. ................................................................... 217

Figure 4.19: Observations of extraterritorial movements and mortality of female

dingoes 2.8, 2.15, 2.30, 2.31 and 2.34.

................................................................. 219

List of tables

xiv

Chapter one:

List of tables

Table 1.1: Dominant prey observed in dietary studies for dingoes in eastern

Australia ................................................................................................................. 27

Table 1.2: Minimum and maximum morphometric measurements of dingoes as

described by Corbett (2001) or Thomson (1992a) .............................................. 58

Table 1.3: Details of measurements collected .......................................................... 59

Table 1.4: Details of samples collected

..................................................................... 60

Chapter two:

Table 2.1: Origin of microsatellite loci reference samples for “purity” tests ....... 87

Table 2.2: Assigned scores of dingo “purity” or percentage of hybridisation ..... 90

Table 2.3: Microsatellites typed for relatedness tests and origin of microsatellite

reference samples .................................................................................................. 91

Table 2.4: Capture data for the five trapping programs ....................................... 92

Table 2.5: Morphometric data for dingoes of the SGBMWHA ............................ 93

Table 2.6: Morphometric data for comparison of dingoes from the SGBMWHA

with dingo populations around Australia ........................................................... 93

Table 2.7: Percentage coat colours recorded from dingoes of the SGBMWHA .. 94

Table 2.8: Percentage coat colour for comparison of dingoes from the

SGBMWHA with dingo populations around Australia in order of latitude ... 94

Table 2.9: Comparisons of average canonical measurements of dingoes from

around Australia with dingoes from the SGBMWHA ...................................... 95

Table 2.10: “Purity” of dingoes in the SGBMWHA according to skull scores .... 96

Table 2.11: Results for “purity” tests for all 50 samples collected from

capture/release animals, one deceased individual and two recapture/release

individuals. ............................................................................................................. 97

Table 2.12: Percentage genetic “purity” of dingoes per site and in total in the

SGBMWHA ......................................................................................................... 101

Table 2.13: Comparison of “purity” score with coat colour and morphometric

measurements of captured dingoes during 2004, 2005, 2006 and 2007 .......... 102

List of tables

xv

Table 2.14: Relationship of genetic “purity” scores with coat colour ................. 103

Table 2.15: Relationship of dingoes which do or do not match coat colour,

morphometric measurements and genetic “purity” criteria ........................... 103

Table 2.16: Numbers assigned to dingoes by STRUCTURE V. 2.2.

..................... 107

Chapter three:

Table 3.1: Total count and percentage of total prey groups and items observed in

dingo and fox scats over 26-months during 2005, 2006 and 2007 from the

SGBMWHA ......................................................................................................... 139

Table 3.2: Explanation of acronyms for Tables 3.4 and 3.5 ................................ 155

Table 3.3: Pearson correlation coefficients for comparative analyses of

activity/abundance records and occurrence of medium sized vertebrate prey

in............................................................................................................................ 156

Table 3.4: Pearson correlation coefficients for comparative analyses of

activity/abundance records and occurrence of medium sized vertebrate prey

in

............................................................................................................................ 157

Chapter four:

Table 4.1: Summary of GPS data precision from dingoes collared with annual

GPS collars and dingoes collared with short term GPS collars ...................... 189

Table 4.2: Comparison of 95% MCP with 90% kernel contours and 50% MCP

with 50% kernel contours for 12 dingoes in the SGBMWHA ........................ 200

Table 4.3: Comparison of 100% MCP area with 50% MCP area during

whelping, rearing, exploratory and breeding biological seasons from dingoes

outfitted with annual GPS collars. ..................................................................... 204

Table 4.4: Total observations of deaths, potential dispersal and extraterritorial

forays, groups and pregnant females ................................................................ 219

List of plates

xvi

Chapter one:

List of plates

Plate 1.1: Steep gullies surrounding Scotts Main Range road at Site 1. ............... 48

Plate 1.2: Looking south from foothills of Centre Ridge towards Joorilands at

Site 2. ...................................................................................................................... 49

Plate 1.3: An example of a Victor ® trap being set in front of a bush scented

with lure. ................................................................................................................ 55

Plate 1.4: A dingo caught in a trap with a trapper approaching the animal to

noose its neck. ........................................................................................................ 56

Plate 1.5: The dingo from plate 1.4 secured to the board at the conclusion of

operation ................................................................................................................ 57

Plate 1.6: Weighing a dingo. ..................................................................................... 59

Plate 1.7: Assessing dentition whilst collecting saliva samples for fracture, decay,

loss and the presence or absence of fleur-de-lis on the incisors. ....................... 60

Plate 1.8: Collecting an ear biopsy and securing an ear tag on the proximal side

of the left ear. ......................................................................................................... 61

Plate 1.9: Four drops of dingo blood on an FTA Card .......................................... 61

Plate 1.10: An emaciated dingo; dingo 2.1 rolling in lure; three juveniles

showing interest in the lure; and a pack of dingoes feeding on an eastern grey

kangaroo during an ad hoc sighting

.................................................................... 62

Chapter two:

Plate 2.1: A piebald domesticated silver fox, a curly tailed domesticated silver

fox, a mottled domesticated dog and a mottled domesticated silver fox and

comparison of non-domesticated silver fox farm silver fox skull with a

domesticated silver fox farm silver fox skull showing variations in width and

length ...................................................................................................................... 75

Initialisms and abbreviations

xvii

2D Two dimensional

Initia lisms and abbreviations

3D Three dimensional

°C Degrees Celsius

A Rarefacted allelic richness (a statistical method to compare the allelic

richness of samples having different sizes)

ABI Applied Biosystems

Ad hoc Created or done for a particular purpose only

Anon. Anonymous

Apr. April

ARF Abundance Rating Formula calculated by dividing the presence or

absence of tracks of a species on total sand plot by the number of sand

plot nights in a specified timeframe

Aug. August

AUG Augathella

B05 or B06 Breeding biological season (March-May) of dingoes in the year 2005

or 2006

BASRA Bargo SRA

BMNP Blue Mountains National Park

BMWHI Blue Mountains World Heritage Institute

bp Base Pairs

BPAF ARF calculated from brushtail possum spoor observed on sand plots

BPAP PAI calculated from brushtail possum spoor observed on sand plots

BPAR Raw counts of brushtail possum spoor observed on sand plots

BPDI Proportion of dingo or fox scats containing brushtail possum prey

remains

BPDR Raw counts of dingo or fox scats containing brushtail possum prey

remains

Br. Breeding season

BUSRA Burragorang State Recreation Area

© Copyright

CA Central Australia


xviii

CB Citizen Band

cf. Compare with

cm Centimetres

CWR Critical Weight Range (Australian fauna weighing 500g – 1500g)

DAF ARF calculated from dingo spoor observed on sand plots

DAP PAI calculated from dingo spoor observed on sand plots

DAR Raw counts of dingo spoor observed on sand plots

Dec. December

DECC Department of Environment and Climate Change

DNA Deoxyribose Nucleic Acid

DNMT DNA methyltransferases

DOP Dilution of Precision

DPI Department of Primary Industries

E05 or E06 Exploratory biological season (December-February) of dingoes in the

years 2005 or 2006

EDC Endocrine disrupting compounds

EMT Epithelial-Mesenchymal Transformation

ERAF ARF calculated from European rabbit spoor observed on sand plots

ERAP PAI calculated from European rabbit spoor observed on sand plots

ERAR Raw counts of European rabbit spoor observed on sand plots

ERDI Proportion of dingo or fox scats containing European rabbit prey

remains

ERDR Raw counts of dingo or fox scats containing European rabbit prey

remains

et al. And others

EHV Eastern Highlands of Victoria

Ex. Exploratory season

Ex situ Out of place

FAF ARF calculated from red fox spoor observed on sand plots

FAP PAI calculated from red fox spoor observed on sand plots

FAR Raw counts of red fox spoor observed on sand plots

f/d Fixes per day

Feb. February

FI Fraser Island


xix

FIS

FTA Patented FTA® Cards are chemically treated to allow for rapid

isolation of pure DNA

Inbreeding coefficient

GBMWHA Greater Blue Mountains World Heritage Area

GIS Geographic Information System

GPS Global Positioning System

ha Hectares

HE

HDOP Horizontal Dilution of Precision

expected heterozygosity

hr (or hrs) Hour(s)

In situ In place

IUCN International Union for the Conservation of Natural Resources

Jan. January

K Estimated size of a population when using software program Structure

KAF ARF calculated from eastern grey kangaroo spoor observed on sand

plots

KAP PAI calculated from eastern grey kangaroo spoor observed on sand

plots

KAR Raw counts of eastern grey kangaroo spoor observed on sand plots

KBNP Kanangra-Boyd National Park

KDI Proportion of dingo or fox scats containing eastern grey kangaroo prey

remains

KDR Raw counts of dingo or fox scats containing eastern grey kangaroo

prey remains

km Kilometre(s)

km/hr Kilometres per hour

LOD Log of Odds

Ltd. Limited

M Month

m Metres

Mar. March

MCMC Markov Chain Monte Carlo

MCP Minimum Convex Polygon

MDB Murray-Darling Basin


xx

m/hr Metres per hour

ml Millilitres

mm Millimetres

MMI Microsatellite Marker Investigation

mtDNA mitochondrial DNA

n Total number

N, S, E, W North, South, East, West and any combination of N or S with E or W

(such as NE = Northeast) provide direction of travel according to the

Arctic and Antarctic poles, and the rotation of the earth toward or away

from the sun.

NB Note well

NGBMWHA Northern Greater Blue Mountains World Heritage Area

NNP Nattai National Park

NPWS National Parks and Wildlife Service

Nov. November

NSRA Nattai SRA

NSW New South Wales

NT Northern Territory

Oct. October

P Probability

PAI Passive Activity Index calculated by dividing total tracks of a species

by the number of sand plot nights in a specified timeframe

PC Pearson Correlation

PCR Polymerase Chain Reaction

PDI Predator Diet Index calculated by dividing total occurrence of a species

as prey by the number of predator scat or stomach samples analysed

within a specified timeframe, such as one month

PDOP Positional Dilution of Precision

per se By or in it itself or themselves

Pop. Population

Pty. Propriety

PWD Parks and Wildlife Division

QLD Queensland

® Registered


xxi

R05 or R06 Rearing biological season (September-November) of dingoes in the

years 2005 or 2006

RDH Resource Dispersion Hypothesis

Re. Rearing season

RLPB Rural Lands Protection Board

SA South Australia

SCA Sydney Catchment Authority

se Standard error of the mean

Sept. September

SEQ South East Queensland

SGBMWHA Southern Greater Blue Mountains World Heritage Area

SL Skull Length

SRA State Recreation Area

TDS Total dingo scats

TFS Total fox scats

TIH Territory Inheritance Hypothesis

™ Trade Mark

US United States

USA United States of America

V. Version

VDOP Vertical Dilution of Precision

VHF Very High Frequency

VIC Victoria

VNTR Variable Nuclear Tandem Repeats

W05 or W06 Whelping biological season (June – August) of dingoes in the years

2005 or 2006

WA Western Australia

WAF ARF calculated from small macropod (swamp wallaby, red-necked

wallaby or wallaroo) spoor observed on sand plots

WAP PAI calculated from small macropod (swamp wallaby, red-necked

wallaby or wallaroo) spoor observed on sand plots

WAR Raw counts of small macropod (swamp wallaby, red-necked wallaby

or wallaroo) spoor observed on sand plots


xxii

WDI Proportion of dingo or fox scats containing swamp wallaby prey

remains

WDR Raw counts of dingo or fox scats containing swamp wallaby prey

remains

Wh. Whelping season

WOAF ARF calculated from common wombat spoor observed on sand plots

WOAP PAI calculated from common wombat spoor observed on sand plots

WOAR Raw counts of common wombat spoor observed on sand plots

WODI Proportion of dingo or fox scats containing common wombat prey

remains

WODR Raw counts of dingo or fox scats containing common wombat prey

remains

YSRA Yerranderie SRA

Abstract

xxiii

This study describes aspects of the descriptive, functional and social ecology of

dingoes Canis lupus dingo from the Southern Greater Blue Mountains World Heritage

Area, Australia. Amendments to the wild dog control order of the Rural Lands

Protection act 1998 recommended that resident dingoes in this protected area be

conserved in situ to maintain ecological function and to preserve wild “pure” dingo

populations.

Abstract

Field work was based on and around two 25 kilometre transects, each within a

separate site of varied rugged and forested terrain, within the core of this reserve. The

aim was to investigate similar biological and ecological descriptors to those used in

past studies of dingoes in Australia for comparison with this population. Objectives

included morphometric measurements and colouration, genetic purity, patterns of prey

consumption and changes in abundance, activity and spatial organisation to assess the

functional role of dingoes in this universally significant protected area.

Dingoes were trapped using padded soft jaw leg-hold traps by professional dingo

trappers. Captured animals were weighed, measured, had tissue/blood samples

collected for genetic tests, collared or tagged to observe patterns of movement, and

released at the location of capture. Dingoes were tested for rare microsatellites found

in captive dingoes and domestic dogs to estimate “purity” , and multilocus genotype

data were compared within the population to determine relatedness. Patterns of prey

consumption were assessed using scat analysis from dingo and fox scats sampled

along each transect. Monthly variations in general patterns of dingo activity were

Abstract

xxiv

tested using indices calculated from counting spoor on sand plots spaced

approximately every kilometre on both transects. Spatial organisation was

investigated using data logging global positioning system telemetry collars and

analysed using minimum convex polygons and kernel contours. Twelve collars were

outfitted to dingoes for 13 months and scheduled to log six to eight locations per 24

hours. Five additional collars were outfitted to dingoes for the 2007 breeding season

and scheduled to log one location every ten minutes for 50-52 days.

From 47 live captures, average morphometric measurements included: weight 16 ±

2.8kg; head length 234.2 ± 12.6mm; ear height 97.2 ± 5.9mm (n = 34); shoulder

height 575.6 ± 29.4mm; hind foot length 183.6 ± 9.2mm (n = 17); tail length 423.3 ±

69.3mm (n = 46); and total length 1326.1 ± 88.7mm. Black and tan was the most

common coat colour (38.3%) followed by sable (31.9%), tan (23.4%) and patchy

(6.4%). Canonical scores to estimate “purity” ranged between 0.46 and 4.34 (n = 10)

suggesting that 80% of the population was “pure” according to previously published

“purity” descriptors. Alternatively, comparative analyses of microsatellites from

captive dingo populations with the 47 live captures and one deceased individual

sampled during field work suggested 2.1% of the population were likely “pure”

dingoes, 16.7% was less than 75% dingo, 43.8% was less than 65% dingo and 37.5%

was less than 50% dingo (n = 48). However, conclusions on “purity” depend on

having a “pure” standard against which to compare current samples. Tests to

investigate genetic relatedness showed each dingo could be assigned to one of eight

closely related groups. Comparison of capture locations with relatedness data showed

related individuals were either trapped within topographically defined areas or during

Abstract

xxv

extraterritorial explorations. Genetic relatedness data also showed variation in

colouration between packs.

In total, 1489 dingo and 962 fox scats were analysed for diet. Most commonly

detected prey remains identified in dingo faeces was swamp wallaby (Site 1: 43.2%

Site 2: 35.1%), brushtail possum (Site 1: 13.8%; Site 2: 7.8%) and eastern grey

kangaroo in Site 2 (Site 1: 1.5%; Site 2: 12.6%). Twenty five species accounted for

the remainder of the diet and included mammals, birds, reptiles and insects. Of the

11768 spoor counted whilst monitoring sand plots, 759 dingo tracks were recorded in

Site 1 and 1048 were in Site 2 across all seasons. Dingo activity peaked and receded

every six months on both transects and related directly with the expected seasonal

activity of dingoes. Many strong positive and negative correlations between diet and

activity of dingoes and activity of medium-large sized mammalian prey items were

observed. Results implied that these interactions were synchronised at a landscape

scale and potentially showed the functional role of dingoes as a higher order predator

in this study area. Similar data on foxes showed no clear interactions with dingoes or

prey species, although fox diet and activity patterns overlapped with dingoes.

Dingoes maintained a mean home range of 34.2km2 (± 8.2 se) for 90% kernel

contours and 5.9km2 (± 1.4 se) for 50% kernel contours (n = 47347 locations from 12

dingoes). Movement patterns identified included territory maintenance behaviours,

extraterritorial forays by males and potential dispersal behaviour by females. Annual

patterns of activity peaked during breeding seasons which could be related with

activity peaks observed using data from sand plots. Daily activity, however, varied

Abstract

xxvi

per individual but tended to have crepuscular peaks. Data implied breeding and pup

rearing behaviours may have affected patterns of movement and activity the most.

From this study, dingoes are redefined as a primitive hypercarnivorous member of the

family Caninae from Australia, distinguished by one annual breeding cycle and

morphometric parameters that characterise the dingoes found within specific

geographic locations. Future management practices should focus on targeting

problem dingoes in areas near pastoral land opposed to adopting widespread dingo

control campaigns across the Greater Blue Mountains World Heritage Area. The

cultural transmission of behaviours from dingoes to dingoes is identified as an

essential natural phenomenon to maintain ecosystem stability and minimise livestock

losses in pastoral areas surrounding the study area. Similarly, environmentally

sustainable agricultural practices need to be adopted for holistic adaptive management

of the region surrounding the Greater Blue Mountains World Heritage Area.

Properties which are affected by dingo predation should be audited for assessment of

objects which may attract or deter dingoes to or from their estate. During the process

of an audit, landowners and the greater public need to be educated about dingo

ecology and adaptive landscape management practices. Environmentally sustainable

economic and land management policies may then be created and implemented for

holistic management of Australian landscapes.

Order in the Pack:

Ecology of Canis lupus dingo in the Southern Greater Blue Mountains

World Heritage Area

Dingo ecology in the Southern Greater Blue Mountains World Heritage Area

1

Australia is a land with many varied landscapes and ecosystems including temperate

rainforests, alpine zones, deserts, dry mesic schlerophyll ecosystems and tropical

rainforests. There is a general consensus that the dingo Canis lupus dingo (Meyer

1793) was most likely being transported by seafarers from Southeast Asia and arrived

in the north of Australia between 4000 and 5000 years ago (Corbett 2001). The effect

of dingoes upon arrival in Australia during the first 3,800-4,800 years before

European settlement is not well known. Based on its reproductive rate and agility,

Kohen (1995) suggested dingoes could have occupied the entire continent within 500

years. Occupation of territory and adaptation to the varied ecosystems and landscapes

may have been made easier with assistance from the Aboriginal population. Potential

interspecific competition between dingoes, thylacines Thylacinus cynocephalus and

Tasmanian devils Sarcophilus harrisii may have contributed to the extinction of the

thylacine and the devil on mainland Australia (Kohen 1995; Wroe et al. 2007).

Introduction

Earlier interactions between dingoes and Aborigines are not well understood.

Dingoes were revered in some Aboriginal cultures and some sources suggested they

were used for hunting (Corbett 2001). Montagu (1942) observed dingoes scavenging

food scraps at Aboriginal camps and suggested such interactions between hunter-

gatherer societies and wild canids may have resulted in the origin of domestic dogs.

Parker (2006), studying “the narratives within which Australians have ‘trapped’ their

dingoes”, showed that Aboriginal stories of dingoes were either ennobling or were

warnings to improve the human condition in contrast to colonial ideology which

resulted in the degradation of dingoes and the environment.


2

The contemporary dingo has a global reputation as an iconic Australian mammal.

Dingoes may be seen on postcards and stamps, on instant lottery scratchies, in

brochures and even on network television and in movies. In contrast, dingoes are

persecuted by many present day Australians for killing introduced domestic

herbivores “protected” by meagre wire fencing adjacent to dingo habitat. If dingoes

are not being shot at or trapped and poisoned, they are being caged or isolated by

fencing (Fleming et al. 2001). Dingoes were seen as such an immediate threat to

livestock production in Australia that the largest fence on earth was built to exclude

dingoes from preferred livestock grazing areas. In addition, a dingo was implicated in

a controversial murder trial in 1980 for apparently seizing a baby from a tent in

central Australia (Marcus 1989). In 2001, a young boy was also killed by a dingo on

Fraser Island, off the coast of Queensland (Edgar et al. 2007). Despite these events,

dingoes are still recognised as an iconic Australian mammal.

The status of the “pure” dingo for preservation was elevated on the International

Union for the Conservation of Natural Resources (IUCN) Red List of Threatened

Species from lower risk in 1996 to vulnerable in 2004 (Anon. 2008). As a result of

hybridisation with domestic dogs Canis lupus familiaris, “pure” dingoes are expected

to be extinct within 50 years. The problem, however, is not that the dingo may be

going extinct, nor that the dingo remains relatively unprotected by Australian

legislation: the problem is how to manage the hybridisation process. In March 2000,

a submission for a second wild dog control order under the Rural Lands Protection

Act 1998 was prepared by the New South Wales (NSW) National Parks and Wildlife

Service (NPWS), State Forests NSW, Department of Land and Water Conservation

and the Sydney Catchment Authority (SCA) to declare some public lands in NSW


3

important for the survival of dingoes. A total of 209 reserves, from more than 1269

available reserves, were selected for dingo conservation.

The Southern Greater Blue Mountains World Heritage Area (SGBMWHA) forms part

of a series of reserves recognised for outstanding universal value of a natural heritage

property. This series of reserves, known as the Greater Blue Mountains World

Heritage Area (GBMWHA) spanning 1,032,649ha, was identified as potential habitat

for maintenance of unregulated ecological function and wild “pure” dingo populations

under the second wild dog control order. Key reasons for their selection included the

rugged topography, the large extent of the areas, their relatively undisturbed nature

and their ability be managed as an ecosystem as part of their charter.

Questions arising from the wild dog control order, therefore, included the following:

1. Does the resident wild dog population in the SGBMWHA consist of “pure”

dingoes, dingo X domestic dog hybrids or feral dogs?

2. What is their functional role in the ecosystem?

3. How should the population be managed: for their function or for their

“purity”?

This study sought information on the ecology of the largest mammalian carnivore in

this ecosystem regardless of whether they were “pure” dingoes, dingo X domestic dog

hybrids or feral dogs. Outcomes from this study are expected to assist adaptive

landscape management practices for the preservation and conservation of the

GBMWHA.

Chapter One:

Dingo ecology, study site and

research framework

Dingoes attacking an eastern grey kangaroo Macropus giganteus in Lake Burragorang © Dennis Ashton, SCA

Dingo ecology, study site and research framework

5

Chapter One: Dingo ecology, study site

and research framework

1.1 Introduction .................................................................................... 71.1.1 The family Canidae ....................................................................................... 9

1.1.2 The dingo in Australia ................................................................................ 141.1.2.1 History and distribution ........................................................................ 151.1.2.2 Genetics ................................................................................................... 181.1.2.3 Biology ..................................................................................................... 201.1.2.4 Diet .......................................................................................................... 251.1.2.5 Activity and abundance ......................................................................... 281.1.2.6 Movement, home range and dispersal ................................................. 301.1.2.7 Management ........................................................................................... 321.1.2.8 The dingo “ex situ” ................................................................................. 341.1.2.9 The dingo “in situ” ................................................................................. 35

1.2 The Southern Blue Mountains World Heritage Area ................... 351.2.1 Climate and hydrology ................................................................................... 37

1.2.2 Geology and topography ................................................................................ 38

1.2.3 Vegetation and habitat ................................................................................... 39

1.2.4 Fauna diversity ............................................................................................... 40

1.2.5 Occupation and disturbance .......................................................................... 41

1.2.6 Management regimes ..................................................................................... 421.2.6.1 Parks and Wildlife Division .................................................................. 431.2.6.2 Sydney Catchment Authority ............................................................... 441.2.6.3 Rural Lands Protection Boards ............................................................ 441.2.6.4 Department of Primary Industries ....................................................... 45

1.2.7 Study sites ....................................................................................................... 461.2.7.1 Site 1 – Scotts Main Range .................................................................... 471.2.7.2 Site 2 – Tonalli to Joorilands ................................................................. 48

1.3 Project design .................................................................................... 501.3.1 Hypothesis ...................................................................................................... 50

1.3.2 Aim .................................................................................................................. 50

1.3.3 Objectives ........................................................................................................ 51

1.3.4 Experimental approach .................................................................................. 51


6

1.4 General methods ............................................................................... 521.4.1 Animal capture and handling ........................................................................ 53

1.4.1.1 Capture and restraint ............................................................................ 531.4.1.2 Morphometric measurements ............................................................... 571.4.1.3 Collection of samples ............................................................................. 60

1.4.2 Additional techniques .................................................................................... 61


7

1.1 Introduction

The dingo is a terrestrial predator. The role of this animal in the wild will vary

between south East Asia, its country of origin, and Australia due to environmental

differences. There is a dearth of ecological research on the dingo in Asia though there

is evidence that it only had a commensal relationship with humans (Corbett 2001;

Fleming et al. 2001). Available research from Asia is not strictly comparable with

Australian based research due to differences in factors such as habitat, climate and

diet. In Australia, the dingo currently assumes the role of a top order predator because

it is the largest mammalian terrestrial predator apart from humans. The current study

investigated aspects of the biology, behaviour and functional role of dingoes in the

SGBMWHA, 65km west of Sydney.

Chapter one reviews past research on the dingo in Australia and Asia. This includes

taxonomic information on the family Canidae and the evolution of the dingo from the

ancestral wolf line (Section 1.1.1). Section 1.1.2, the dingo in Australia, introduces

characteristics of the dingo identified within past research in Australia. Background

information on the arrival of dingoes in Australia and their history with Aboriginals

and Europeans is provided. Appraisal of previous studies on the genetics, biology,

diet, behaviour and patterns in movement and management of dingoes is also

reviewed. Current ex-situ and in-situ management practices are referred to in Section

1.1.3 to provide context for the management of dingoes in the SGBMWHA.


8

Section 1.2 introduces the study site. Location, climate, geology and topography,

vegetation and habitat are described. Resources needed to sustain a population of

dingoes, including prey items and water availability, are introduced to describe the

landscape in which these animals live. Site occupation, disturbance and management

regimes are also described. Historical and contemporary reports of the dingo in the

study site are also reviewed and discussed. This information will contextualize how

the history and management of the dingo, introduced in section 1.1, has affected the

population of animals in the SGBMWHA. Research questions and experimental

approach are explained in section 1.3. General materials and methods used to answer

the research questions are described in section 1.4. Specific techniques are described

within their relevant chapters.

Data collected are analysed and discussed in chapters two, three and four. Chapter

two addresses population genetics and biology. Dingo “purity” is critically appraised,

and compared with data on skull allometry, morphometrics, coat colour and genetic

biotechnology estimates previously reported. Tests of genetic “purity” and familial

relatedness of dingoes in the SGBMWHA are discussed in the context of changes that

may occur within species across a range of geographically isolated regions in

Australia. Chapter three provides analysis of annual changes in diet, activity and

abundance. Significance of observed relationships with extant prey and non-prey are

investigated and discussed in relation to biological seasons of dingoes. Patterns in

movement are analysed in chapter four from traditional VHF telemetry and novel

GPS telemetry data. Habits observed as a result of acquired telemetry data are related

to recorded changes in diet, activity and abundance as discussed in chapter three.


9

Observations of mortality and dispersal are included using data collected from novel

motion sensing passive camera traps and documented deaths.

Chapter five contextualises the dingo in modern day Australia, with reference to the

key findings in this study and prejudices of contemporary Australians. Review and

discussion of the implications of research findings for the research site and the

SGBMWHA dingoes are also made here. Conventional approaches to dingo

management are discussed in relation to findings of this study. A management

strategy, concordant with public, agricultural and environmental perspectives is

proposed for the SGBMWHA. This strategy has potential to be an adaptive, holistic

framework for long term sustainable management of dingoes in all Australian

ecosystems. The global relevance of this study is discussed in relation to the

importance of preserving and conserving the globally significant research site: the

GBMWHA.

1.1.1 The family Canidae

The family Canidae evolved 10-12 million years ago. They are known for the

possession of uniform and unspecialised dentition, with the carnassial teeth (upper

fourth premolar and lower first molar) arranged for shearing flesh (Wang et al. 2004;

Van Valkenburgh 2007). They are recognised as the oldest form of the Order

Carnivora. Within the Order Carnivora, canids fall into the superfamily Cynoidea

under the suborder Caniformia. Cynoidea are a sister family to Arctoidea (includes

bears Ursidae, the red panda Ailuridae, raccoons Procyonidae and weasels

Mustelidae) within the Caniformia. All modern day canids are found in the subfamily


10

Caninae of the family Canidae under tribes known as Vulpini, or fox-like canids, and

Canini, or wolf-like canids (Figure 1.1) based on 2001 base pairs of DNA sequence

from mitochondrial protein coding genes (Wayne et al. 1997). There is debate

between paleontological, molecular and morphological research regarding the

radiation of the Canini tribe around the world (cf. Vilà et al. 1997; Wang et al. 2004).

If the wolf-like canids began in North America, their arrival in Eurasia commenced an

extensive radiation and range expansion throughout Europe, Africa and Asia.

Combined with the suggestion that the tribe Canini belonged to a circum-arctic fauna

which undergoes expansions and contractions with the fluctuating global climate, the

process of evolution for multiple closely related wolf-like canids is explained. With

the arrival of the dingo in Australia during the late Holocene period, the family

Canidae have been unsurpassed in world distribution by any other families from the

Order Carnivora (Wang et al. 2004).


11

Figure 1.1: Phylogenetic tree displaying the relatedness of the contemporary canids based on 2001 base pairs of DNA sequence from mitochondrial protein coding genes (Wayne et al. 1997)

It is difficult to ascertain how many subspecies of Canis lupus are extant. There are

many sources of information which provide estimates though none of these provided

certainty. Figure 1.2 lists 18 possible subspecies, including the controversial Eastern

timber wolf and some extinct subspecies.


12

Figure 1.2: Potential subspecies of the Gray wolf.

Phylogenetic studies cannot be satisfactorily performed on canid forms from any

single continent because of their Holarctic distribution and faunal intermingling.

Instead, canids may be classed as hypo- (slightly), meso- (moderately) or

hypercarnivorous (highly) (Wang et al. 2004). It is difficult to class the dingo as

hypercarnivorous because it generally weighs less than 21kg, unlike other

hypercarnivorous canids such as the African wild dog Lycaon pictus and the Gray

wolf which are much larger. The dingo does fit the definition of hypercarnivory,

however, because it hunts in packs, has hypercarnivorous dentition and preys on

Gray wolf Canis lupus lupus

Arabian Wolf Canis lupus arabs

Domestic dog Canis lupus familiaris

Dingo Canis lupus dingo

Caspian Sea wolf Canis lupus cubanensis

Arctic wolf Canis lupus arctos

Eastern timber wolf Canis lupus lycaon Egyptian wolf Canis lupus lupaster Eurasian wolf Canis lupus lupus Great Plains wolf Canis lupus nubilus Hokkaido wolf (extinct) Canis lupus hattai Honshu wolf (extinct) Canis lupus hodophilax

Italian wolf Canis lupus italicus Mackenzie valley wolf Canis lupus occidentalis

Newfoundland wolf (extinct) Canis lupus beothucus

Russian wolf Canis lupus communis Southern East Asian wolf Canis lupus pallipes Tundra wolf Canis lupus albus

Mexican wolf Canis lupus baileyi


13

animals larger than its weight (Macdonald and Sillero-Zubiri 2004b) such as the red

kangaroo Macropus rufus and the eastern grey kangaroo Macropus giganteus.

Shepherd (1981) showed dingoes have higher predation rate on kangaroos (0.38kg

prey per kg predator per day) than wolves do on moose (0.13kg prey per kg predator

per day), and wolves are classed as hypercarnivorous canids. It can also be claimed

that the dingo developed hypercarnivorous dentition due to interspecific competition

with larger carnivores before its arrival in Australia. The dingo can still be considered

a “primitive” canid because it has retained the key characteristic of breeding

seasonally once per year, similar to other primitive relatives such as the wolf C. lupus,

the coyote C. latrans, and the jackal C. adustus (Newsome and Coman 1989). In the

current study, the dingo in Australia will be treated as hypercarnivorous because of

the position this animal holds as a top order terrestrial predator. The worldwide

distribution of the dingo is currently thought to be limited to South East Asia and

Australia (Figure 1.3).

Figure 1.3: Current known world distribution of dingoes (© 2003 Canid Specialist Group and Global Mammal Assessment cited in Macdonald and Sillero-Zubiri 2004b)


14

1.1.2 The dingo in Australia

Corbett (2001), Dickman and Lunney (2001) and Fleming et al. (2001) provided

comprehensive reviews on the biology, ecology and historical management of the

dingo. These authors included information on the impact of European colonisation on

its ecology and the impact of dingoes on European colonisation, the Australian people

and the Australian economy respectively. Intensive studies on dingoes reviewed in

these texts include those of Baverstock and Green (1975), Newsome et al. (1980),

Newsome and Corbett (1982; 1985), Jones (1990), Thomson (1992a, b, c, d), Woodall

(1996), Wilton et al. (1999) and Wilton (2001) and have provided insight into the

evolution of dingoes from a shared ancestor of the wolf and how the dingo became

established in Australia.

Contemporary research on dingo genetics demonstrated differences between captive

dingoes and domestic dogs (Wilton et al. 1999; Wilton 2001). Archetypical

characteristics of a top order predator are prevalent in the information reported on

dingo morphology, biology and behaviour. Dietary studies (Whitehouse 1977;

Newsome et al. 1983a, b; Robertshaw and Harden 1985; 1986; Corbett and Newsome

1987; Thomson 1992c; Paltridge 2002) showed selective and opportunistic patterns of

prey selection. Patterns of activity or changes in dingo abundance showed dingo

populations peaked during breeding seasons (Allen et al. 1996; Fleming et al. 1996;

Edwards et al. 2000; Burrows et al. 2003). Records prior to all studies mentioned

above were limited, anecdotal and potentially biased diary entries from European

settlers (Marcus 1989; Freeman 2005; Parker 2007).


15

1.1.2.1 History and distribution

Since European colonisation and settlement of Australia, available habitat for the

dingo has been severely fragmented. The largest predator-proof fence in the world,

aptly named the dingo or the dog-proof fence, was constructed between 1880 and

1885, separating the dingoes of central Australia from the dingoes of south east

Australia. This fence was constructed to assist the productivity of livestock farmers in

the fertile southeast, following persistent attacks on introduced domesticated

herbivores, particularly sheep. Intense control programs, rapid development of towns

and infrastructure and land clearing for agriculture created more isolation of

remaining dingo populations. Figure 1.4 presents the current distribution of wild dogs

in Australia including dingoes, feral or wild domestic dogs and their hybrids. They

can inhabit a diversity of ecosystems from tropical to temperate rainforests, deserts

and all habitat types in between (Fleming et al. 2001).

Figure 1.4: Potential distribution of wild dogs (dingoes, feral or wild domestic dogs and their hybrids), in Australia (from Fleming et al. 2006). ■ SGBMWHA , research site for this study; solid line is the dingo fence; thin lines are state political boundaries; area above dashed line represents area of mostly “pure” dingoes; distribution of wild dogs is common to naturally sparse in spotted areas; wild dogs are mostly absent in areas without shading.


16

Dingoes are not found in fossil records in Tasmania (Kiernan et al. 1983) so it may be

postulated with some certainty that dingoes arrived in Australia after Tasmania had

become an island due to rising temperatures and sea levels between 10,000 and

18,000 years ago (cf. Nakada and Lambeck 1989; Kohen 1995). Dingoes inhabited

all of mainland Australia either in commensal relationships (Fleming et al. 2001) or as

free roaming competitors with Aborigines (Kohen 1995) prior to European settlement

and the construction of the dingo fence. Studies have indicated that the dingo

contributed to the extinction of the two large marsupial carnivores, the thylacine and

the Tasmanian devil, from mainland Australia due to direct or indirect competition

(Kohen 1995; Johnson and Wroe 2003; Wroe et al. 2007). Kohen (1995) also

suggested the arrival of the dingo in Australia coincided with:

1. Expansion of the Aboriginal population;

2. Developments in the small tool tradition of Aborigines arising simultaneously

in India;

3. Environmental change induced from firestick landscape management

practices; and

4. Climate change.

The dingo may therefore assume a shared responsibility with Aborigines and climate

change for the mainland extinction of the two largest endemic marsupial carnivores.

Johnson and Wroe (2003) also provided evidence consistent with those propositions

by Kohen (1995). Expansion of European settlement, developments in agriculture

and landscape changes resulting from agriculture since European colonisation can,

similarly, assume responsibility for the persistence of dingoes despite eradication

efforts. Water availability in the form of troughs , bore drains and dams has increased


17

prey availability potentially assisting their survival through droughts (Newsome and

Coman 1989). Prevalence of domestic dogs with settlers has also provided new

opportunities for dingoes to mate, posing a threat to the survival of “pure” dingoes

due to hybridisation.

Approximately 95 years after the dingo fence was erected (or 190 years since

colonisation), the first major research on the dingo was published. Green and Catling

(1977) provided a general overview on the biology of the dingo including genetics,

growth and ageing, reproduction, behaviour and physiology. Soon afterwards more

details on social biology, movements and the effects of control efforts in Western

Australia, central Australia and coastal New South Wales and Victoria were reported

(Anon. 1978). Findings from research on dingo skull allometry as a descriptor of

“purity” were first published in the early 1980’s by Newsome et al. (1980) and

Newsome and Corbett (1982; 1985).

Under experimental conditions, “pure” dingoes and domestic dogs were crossed and

levels of hybridisation in captivity were calculated using comparative analysis of

canonical scores (Newsome et al. 1980; Newsome and Corbett 1985; Corbett 2001).

Dingoes from central Australia were selected for these tests under the assumption they

had “purity of line”. Eight measurements, including length of auditory bulla (X1),

maximum maxillary width (X2), mid crown width of the p4 tooth (X3), basal crown

length of c1 (X4), opisthion to inion (X5), width of both nasal bones (X6), cranial

height (X7) and distance between the posterior alveolar rims c1-p4 (X8

Y = 0.249x

) were recorded

and applied to a standard equation:

1 – 0.261x2 + 1.999x3 – 1.137x4 + 0.318x5 + 0.475x6 – 0.205x7 + 0.136x8 – 3.717


18

Resulting canonical scores where Y was ≥1.271 were classed as “pure”, 1.270 to -

1.393 were classed as hybrid and scores ≤-1.394 were classified as domestic dogs.

Jones (1990) and Woodall et al. (1996) also used skull allometry to identify the dingo

and “hybrids” before the development of new genetic biotechnology techniques

(Wilton et al. 1999; Wilton 2001; Savolainen et al. 2004).

1.1.2.2 Genetics

Clarke et al. (1975) and Shaughnessy et al. (1975) (cited in Green and Catling 1977)

reported no discernable difference between blood proteins of dingoes compared with

domestic dogs using protein electrophoresis. However, with development of genetic

biotechnology techniques, Wilton et al. (1999) compared microsatellite variation in

Australian captive dingoes with domestic dogs. Fourteen previously described canine

microsatellites were typed from sixteen captive dingoes and sixteen domestic dogs.

Results showed different distributions in alleles when compared, which overlapped in

13 of the 14 loci. The fourteenth locus, CXX30 had no alleles common in each group.

Other than this finding, the level of variation in the dingo samples used was distinctly

lower than the level of variation observed in domestic dog alleles.

Wilton (2001) supported earlier findings when testing DNA methods to assess dingo

purity. Using 77 captive dingoes, 55 mixed breed domestic dogs and 50 “wild dogs”

from one site in south east Australia, he found that the majority of microsatellites

tested showed similar distributions within their genetic group. Once again, variability

was lower in the captive dingoes than in the dogs although it is not specified whether

the “wild dogs” sampled were grouped with domestic dogs as “dogs”. A technique to

ascertain a calculated probability that a genetic sample was more likely to be from a


19

“pure” dingo, a “hybrid” or a domestic dog based on allele sizes, distributions and

frequencies was developed. Although no definitive answer on “purity” or

hybridisation could be attained, animals could be assigned to categories of dingo, ¾

dingo, hybrid or domestic dog. As the number of alleles tested increased the more

likely it became to detect past dilution events, and to determine differences between

populations of dingoes across Australia. Using this information, Wilton (2001)

implied that the conservation value of a population could be determined and screened

through genetic means.

To estimate the time of the first introduction of dingoes to Australia, mitochondrial

DNA (mtDNA) was sequenced from 211 “dingoes” , 22 dogs and 19 pre-European

archaeological dog samples (Savolainen et al. 2004). These were compared with 654

dog and 38 wolf samples from another study (cf. Savolainen et al. 2002). Sequence

variation of mtDNA in dingoes was very restricted compared with dogs and wolves.

Results indicated that all dingo mtDNA types originated from type A29 because it

was observed in 53% of samples. Based on an estimated wolf-coyote divergence of

two million years ago, the arrival of the dingo in Australia was calculated to be

between 4,600-5,400 years ago. A considerable difference for the mean distance to

A29 was also observed between dingoes in the north west of Western Australia and

other parts of Australia. This discrepancy was treated as local genetic drift.

Observations of A29 in samples from East Asia, New Guinea and arctic America

showed 47% of genetic samples unaccounted for and the discrepancy in Western

Australia implied the introduction of dingoes to Australia may have occurred more

than once. However, Savolainen et al. (2004) refuted this finding because a number

of other mtDNA types were observed in samples from islands surrounding Australia.


20

Another important aspect of dingo genetics involves phenotypic differences between

geographically isolated groups but these have not been addressed in great detail.

Thomson (1992a) showed localised gene pools existed where variations of black and

tan and white colours occurred at higher frequencies in some packs more than others.

Observations such as this on a range of geographically isolated populations across

Australia could become increasingly important for conservation efforts to preserve

this iconic species.

1.1.2.3 Biology

General descriptors used for dingoes include erect ears, a bushy tail and ginger/tan,

black and tan, black or white coloured coats (Corbett 2001; Fleming et al. 2001).

White points may be present on some of the paws and the tail tip, but not always

(Fleming et al. 2001). Dingo X domestic dog hybrids have been described as having

dingo-like appearance but coat colour may vary to brindle or patchy (Green and

Catling 1977). Newsome and Corbett (1985), however, reported that dingoes with

“pure” skull allometry can be ginger with white spots, patchy ginger and white or

brindle. Fleming et al. (2001) suggested that these colours along with sable and

patchy black and white are indicative of hybridisation although Newsome and Corbett

(1985) stated sable is no indication of dog ancestry and reported observations of black

and white dingoes from settlers in the mid 1800’s. Compiled average measurements

of dingoes from around Australia show that their total length is 123 centimetres,

shoulder height is 57 centimetres, head length is 22 centimetres, ear length is 10

centimetres, hindfoot length is 19 centimetres, tail length is 31 centimetres and they

weigh approximately 16 kilograms (Fleming et al. 2001). Newsome and Coman


21

(1989) reported that dingoes bark when alarmed, but otherwise communicate by

howling.

The Australian National Kennel Council classified the dingo, as a breed, in group four

with hounds (Anon. 2007). Dingoes were described as a medium sized and elegant

breed with great agility and stamina. Dingoes are alert, rangy and instantly reflexive

with minimal excess flesh (Anon. 2007). Coat colour was reported to range from red,

ginger, gold to palest cream, and black and tan. Males stand 52-60cm high and

females less whilst displaying more femininity, with adult weight for males and

females ranging between 13.5-19kg. The Australian National Kennel Council also

stated that male dingoes should have two normal, fully descended testicles and,

despite information reported by Newsome and Coman (1989), dingoes do not bark.

Socio-biological research on wild canids may provide insight into the biology of

dingoes. Large canids generally have a skewed sex ratio towards males, resulting

from female emigration and retention of male helpers (Macdonald and Sillero-Zubiri

2004b). Due to the period of pup dependency, competition between females for male

helpers to secure resources is likely to be intense and drive the society towards

polyandry (Macdonald and Sillero-Zubiri 2004b). Moehlman (1986; 1989) and

Moehlman and Hofer (1997) showed that the body mass of the mother influenced the

mass of the neonate, overall litter mass and gestation length. Increased pressure on

available resources and conflict between females to secure male helpers was reported

to induce reproductive suppression, increasing investment for reproduction. As a

result, alloparental behaviour and shared responsibilities force the hypercarnivorous

canids to move and hunt in packs. Hunting in packs increases the chance of success

when hunting larger prey, defending territories and resources from competitors and


22

maintains energy requirements for all members of the pack. MacDonald and Sillero-

Zubiri (2004b) provided a more detailed review of sociality in hypercarnivorous canid

populations and the counter argument by Geffen et al. (1996). Geffen et al. (1996)

stated that facets of canid socio-biology are adaptive to changes within their territory

and related to resource availability.

Dingoes are communicative canids. Scientists have hypothesised that postural, vocal

and olfactory communication can: a) maintain or increase distance between

individuals; b) allow individuals to assess fighting or resource holding potential of

competitors; and c) provide a means to identify the location or movements of pack

members and neighbouring groups (Mech 1970; Corbett 2001). This is important

when prey or water resources may be more abundant in the interstices of territories

and resources are shared. It is also an essential trait when hunting larger game.

During the current study, a dingo was disturbed from a hide near open grassland

where large mobs of eastern grey kangaroos were feeding and frequently fed. This

individual ran to more protective cover and began howling. Another four dingoes

were then observed approximately 200 metres behind the original location of the first

dingo and proceeded to flee at the sight of the vehicle. Photographic evidence from

an infra-red camera in the area showed all five animals travelling together before the

presumed hunt was interrupted.

Original research on social biology of the dingo stated dingoes can hunt or navigate

alone and within groups (Anon.1978). Dominance hierarchies, similar to that

observed in wolves with an “alpha” pair, subordinates, juveniles and omega animals

are prevalent in captivity and the wild (Mech 1970; Corbett 2001). Communal

nurture, where subordinate animals in wild groups do not breed but assist to rear the


23

offspring of the dominant pair, is apparent. Infanticide has been reported (Corbett

1988) and a dominant bitch, in captivity, was observed to move her pups into the

same den as the pups borne by the subordinate. The young borne by the subordinate

eventually died of starvation whilst both bitches suckled the young of the dominant

female (Anon. 1978). Hypothetically, subordinate females breed each season to

suckle and assist the dominant female during lactation. This would act to regulate

populations on one hand and increase survivorship of the dominant bitch and her

young, on the other.

Thomson et al. (1992a) reported perturbations in the social organisation of dingo

packs after an increase in population density due to an increase in pack size.

Coinciding with a reduced food supply after a period of little emigration, emigration

increased, some packs disintegrated, territories shifted and others contracted into

favourable areas. Infanticide or population regulation by dominant bitches was

alleged to occur. In other instances, multiple litters within packs were reported. This

was hypothesised to allow rapid population increase following high mortality from

events such as eradication programs. Multiple litters within a pack have also been

reported for wolves (Van Ballenberghe 1983).

Breeding season usually commences around March and concludes during May, but

this can vary between regions and the biology of individuals, such as age (Fleming et

al. 2001). Pro-oestrus/oestrus in captivity lasted 10-12 days (Corbett 2001) but

behavioural data from Western Australia suggested that it can last 30-60 days

(Thomson 1992b). Catling (1979) reported male dingoes in central Australia are

almost aspermous outside of the oestrous cycle. Captive males from central Australia

housed in a cool Canberra climate, however, produced spermatozoa all year though


24

this may have been from housing effects or proximity to domestic dogs in oestrus.

Gestation periods observed in captivity in Australia lasted 61-69 days (Corbett 2001).

At the London Zoo, pro-oestrus lasted two-three weeks, oestrus lasted three to four

days and four recorded gestation periods were 59, 62, 62 and 73 days (Kleiman 1968).

Copulatory ties were observed in the northern hemisphere in October and as late as

December, possibly as a result of crossing the equator. Litters in Australia are usually

born in winter (June-August) (Fleming et al. 2001) confirming dingoes are short day

length breeders. There are contradictions in reports, however, when “purity” is being

estimated for populations based on the time of year that whelping occurs. Litters from

“pure” dingoes have been recorded in “most months” in central Australia, but if litters

are born between November and April in south east Australia, the female is presumed

to be a hybrid (Corbett 2001). Dingo seasons observed by Thomson (1992d) included

pre-breeding (14 Feb.-22 Apr.); breeding (23 Apr.-25 June); nursing one (26 June-14

Aug.); nursing two (15 Aug.-6 Oct.); post-nursing (7 Oct.-4 Dec.); and non-breeding

(5 Dec.-13 Feb.). Corbett (2001) suggested dingo breeding season was around April

and May and Jones (1990) showed most births were between June and August,

following an oestrus peak in May, and September-December was the non-breeding

season.

Lactating bitches are thought to recycle water by coaxing pups to urinate and defecate

through licking of their genitalia (Baverstock and Green 1975). Ingestion of faeces is

also thought to be a mechanism of developing immunity to infectious agents that pups

may be in contact with, by lactating females developing antibodies and returning these

to the pups via the milk. Dingoes consume approximately 7% (≈1kg) of their body

weight in food per day and in desert regions about 70-100ml/kg per day in water,

though fluids may also be absorbed from prey (Anon. 1978; Hulst 2008).


25

1.1.2.4 Diet

There does not appear to be any data on daily energy requirements for dingoes. Major

prey constitutes mammals, birds, vegetation and reptiles but varies depending on prey

abundance and prey availability (Whitehouse 1977; Robertshaw and Harden 1985;

1986; Newsome et al. 1983; Corbett and Newsome 1987; Thomson 1992c; Paltridge

2002). Dingoes generally exhibit a combination of opportunistic and selective

predation (Fleming et al. 2001). It is safe to assume dingo diet is subject to changes

in pack structure and hunting strategies or pack status of the individual being sampled.

For instance, dominant animals may be first to feed, consuming gut contents of the

prey whilst low-ranking animals may be forced to scavenge carcasses or prey remains.

Dominant dingoes may therefore show higher vegetation content in stomach or faecal

samples, whilst low ranking animals may show a higher incidence of fur or bone

fragments from scavenging. These ideas may not be testable in field studies but are

within the realms of dominance hierarchies. Similarly, neighbouring packs may

display dietary differences based on prey availability within their territory.

Dietary studies of dingoes from eastern, central and Western Australia displayed

dietary differences across Australia based on prey availability within their regions.

Analysing stomach contents of dingoes in Western Australia, Whitehouse (1977)

reported 40% of prey consumed was the euro Macropus robustus, 26.9% was the red

kangaroo Macropus rufus and 5.5% was the introduced European rabbit Oryctolagus

cuniculus. Dingoes were documented as a prey item as often as introduced sheep

Ovis aries (4.1%) which were “probably non-feral” but evidence is not provided.

Ground matter was observed in 54.5% of stomachs, birds were observed in 15.3% and

remaining contents ranged from large, medium and small native and introduced

mammals (11.2%), reptiles (5.6%), insects (5.5%) and miscellaneous matter (1.4%).


26

Conclusively, the dingo was deemed to be an opportunistic predator despite the fact

that little introduced domestic livestock were represented in the stomachs and they

were as common in the site as the euro and red kangaroo. Marsack and Campbell

(1990) made similar observations: 96% of food items were mammalian, significantly

dominated by rabbit and red kangaroo with minor insignificant contribution of cattle

which was available as carrion. Thomson (1992c) also reported red kangaroo as the

staple prey, even when cattle carrion and sheep were as abundant.

Dingoes in central Australia displayed prey switching related to prey availability

(Corbett and Newsome 1987). On the whole, rabbit was most commonly eaten

(56%), followed by small mammals (27%), cattle (16.8%) as prey (2.1%) and carrion

(6.3%)1

1 Remaining percentage of cattle was indeterminate.

, red kangaroo (15%), lizards (12%), birds (4%) and insects (2%). Corbett

and Newsome (1987) reported ‘alternation of predation’ by dingoes in their study.

Dingoes were observed to predate overabundant rodent populations for one year at the

cessation of a severe drought. Rabbit populations were then maintained by dingoes

for three years during wetter periods when they increased as items of diet from 69% in

the first flush to 90% in the second flush. Following decline of the rabbit population

in the proceeding drought, red kangaroo increased as a prey item, as did cattle as

carrion when they began to die off due to dry conditions. Paltridge (2002) provided

alternative evidence to suggest reptiles were a staple prey item for dingoes at one

locale, approximately 600-800km north of the site used by Corbett and Newsome

(1987), between 1995 and 1997. Eldridge et al. (2002), studying the same region as

Corbett and Newsome (1987), reported rodents (42%), macropods (39%) and rabbits

(29%) as the most common prey items. The occurrence of rabbits as prey directly

correlated with their abundance in their study.


27

Dietary studies in eastern Australia are represented in Table 1.1. Research areas

ranged from south east and north east Victoria and south east, east and north east

NSW. In general, all studies were linked through the Great Dividing Range.

Table 1.1: Dominant prey observed in dietary studies for dingoes in eastern Australia. Site Date Major prey % Reference

From mountains of Gippsland, NE Victoria and SE NSW and the SE NSW

coast

1969-1975 Wallabies Common wombats

a

Possums

-

European rabbits

b

50.8 12.9 11.6 7.8

(Newsome et al. 1983b)

Nadgee (SE NSW coast), and Kosciusko (SE NSW

mountains)

1971-1980 Wallabies Possums

a

Water birds

b

European rabbits

c

41.1 17.4 36.9 10.2

(Newsome et al. 1983a)

Bega, coastal SE NSW 1981-1982 Wallabies a

European rabbits

Possums

46

b 17.5 8.2

(Lunney et al. 1990)

GBMWHA, E NSW 2002 Wallabies a

Possums

b

Antechinus sp.

d

Avifauna

37.1 21.3 11.1 9.5

(Mitchell and Banks 2005)

NE NSW 1969-1974 Wallabies a

Possums

Bush rat*

b, e 41.7 11.3 12.2

(Robertshaw and Harden 1985)

a Includes swamp wallaby Wallabia bicolor and red-necked wallaby Macropus rufrogriseus b Includes common brushtail possum Trichosurus vulpecula, common ringtail possum Pseudocheirus peregrinus and indeterminate possum c Includes little penguin Eudyptyla minor, mutton bird Puffinis sp., Swan Cygnus atratus, Eurasian coot Fulica atra and indeterminate large water birds d Includes brown antechinus Antechinus stuartii and dusky antechinus Antechinus swainsonii

e Does not separate common brushtail possum or mountain possums T. caninus -

* Rattus fuscipes Vombatus ursinus

Additional research in Gippsland by Brunner et al. (1976) and Friend (1978) showed

a high incidence of possum (brushtail/mountain and ringtail), swamp wallaby, bush

rat and rabbit but did not separate scat of dog/dingo or introduced red fox Vulpes

vulpes. Other prey items reported within dietary studies are significant contributions

to diet when combined but are relatively insignificant on their own.

Publications from western, central and eastern Australia reported predation of

livestock without stating whether predated livestock were feral or domesticated.


28

Livestock was a minor contribution to diet in all studies and predation or scavenging

could not be distinguished in most. Some of these studies were directly questioning

the impact of dingo predation on livestock enterprises and were funded by the

Australian Meat Research Committee (Australian Meat and Livestock Corporation),

Agriculture Protection Board of Western Australia and the Rural Credits

Development Fund. Recently developed views (Allen and Gonzalez 1998) hold that

disturbing dingo pack structure, a consequence of dingo control, forces individuals to

hunt alone and subsequently predate opportunistically on easy and abundant prey such

as livestock. Research on cattle stations in Queensland showed that the highest levels

of calf predation by dingoes occurred on baited sites opposed to non-baited sites

(Allen 2000). In contrast to the increased levels of predation observed, estimates of

dingo activity showed that activity was reduced as a result of the baiting campaign.

1.1.2.5 Activity and abundance

The most commonly used techniques to estimate activity and abundance of dingoes

involve systematic sampling of areas where their sign are most easily detected

(Fleming et al. 1996). Sign can include direct sightings, hair, scent posts and scats but

footprints are used most commonly. Research objectives for studies on activity or

abundance of dingoes in Australia have been to test the efficacy of techniques to

monitor the population to determine the success of control programs.

Allen et al. (1996) compared the fatty acid scent index with buried meat index and the

passive track index in south western Queensland. Results showed that each index was

sensitive to variation between individuals (age/pack status) and environmental

constraints. Fleming et al. (1996) measured the efficacy of aerial baiting of a wild


29

dog population in north eastern tablelands of NSW. Results indicated levels of

activity and abundance varied at each site, before and after baiting. A decline in

activity and abundance was also observed at the site without treatment. Sites were a

maximum of 12.7km apart, within the scope of one dingo home range, and shared a

common boundary in Guy Fawkes River National Park that may have been the link

between sites. These data indicated that “ targeted” aerial baiting campaigns may

cause landscape scale perturbations to ecosystem function.

Burrows et al. (2003) conducted a long term experiment to trial and develop effective

aerial baiting control measures in the Gibson Desert of Western Australia. A density

index for cat, fox and dingo tracks was calculated pre- and post-baiting. Variations in

dingo track densities between trials and baited and unbaited sites were observed pre-

baiting and post-baiting at unbaited sites. Distance between sites was only 40km for

trial one and 5km for trial two, within the home range of dingoes in Western Australia

(cf. section 1.1.2.6). Edwards et al. (2000) reported that track surveys at one site can

be one to three times higher than at other sites. Discussion of track based surveys

indicated that individual activity is a function of population density and population

reduction may increase activity of remaining individuals. This was also observed by

Allen et al. (1996). Site differences may be based on resource availability and

competition (Edwards et al. 2000), intraspecific competition, spatial organisation and

geography.

Basic biological data, such as patterns of dingo activity or changes in abundance for

an extended period of time are scarce. Harden (1985) calculated relative frequencies

of activity and rest using telemetry techniques in north eastern NSW. Activity

patterns were highest at dawn and dusk and lowest around midday. The activity


30

patterns of dingoes in that study were not crepuscular, nor were they nocturnal. In

Western Australia (Thomson 1992b), movement patterns were crepuscular with

periods of rest in the heat of the day. February – May marked the pre-breeding season

when raised-leg urination and howling increased (Thomson 1992b). Dingoes were

most active in March, least active in June and activity from August to December

fluctuated.

Fleming et al. (1996) reported that data for indices of activity or abundance should be

compressed into discrete periods of the annual biological cycle of the species of

interest. In the case of the dingo discrete periods could be breeding, whelping, rearing

or exploratory seasons. Using this method, most data on activity and abundance

showed increased activity during breeding (March – May) and rearing seasons

(September – November), which is information that affects control effort. These data

have on occasion been collected with data on movement patterns, home range and

dispersal.

1.1.2.6 Movement, home range and dispersal

Harden (1985), Thomson (1992b, d) and Thomson et al. (1992b) provided the most

comprehensive data on movement ecology of dingoes. Harden (1985) tracked five

adults and four juveniles in north eastern NSW from 1970-1974. Using triangulation

and signal strength techniques, 4058 fixes were obtained over 515 days. Tracking

was intensive, initially obtaining fixes twice daily and then every 15 minutes for 100-

hour intervals. Results suggested that dingoes moved with two different patterns.

The first, identified as searching movement, was characterised by high activity in a

small area. The second was characterised by purposeful movements over a substantial


31

area and dubbed exploratory movement. Home ranges varied from 4.3km2 (from

1063 locations) for a juvenile female to 54.8km2

(from 52 locations) for an adult

male.

Adult males were observed to occupy larger home ranges than adult females in the

Fortescue region of Western Australia (Thomson 1992d). Mean 95% home range

areas were 84.8km2 for adult males, 56km2 for adult females and 159.6km2

for lone

dingoes (n = 3, 8 and 3 respectively). Pack territories revealed a large degree of

spatial separation between different packs and considerable overlap of individuals

within packs. Territories were stable from year to year and encounters between packs

were rare. The number of locations ranged from 71-448 over a minimum period of

eight months, spanning four of six biological seasons identified for that study.

Riverine areas were preferred habitat and core areas of activity randomly shifted

between seasons. Pup rearing, however, appeared to be the most influential factor in

movement patterns. Similar to Harden (1985), Thomson et al. (1992b) reported

extensive short-term or exploratory movements. Data from Thomson (1992d)

suggested that it was uncommon for dingoes to travel large distances across

territories, though one individual travelled 19.2km in 7.5hrs and another 17.2km in

6.1hrs. Since techniques employed by Thomson (1992d) incorporated aerial

telemetry, compared with triangulation used by Harden (1985), observations in

movement patterns cannot be directly compared. Corbett (2001) reported dingoes that

shared water resources howled to inform neighbouring packs of their approach. This

behaviour is expected to minimise conflict between, and injuries to, pack members

and may explain why inter-pack encounters were rarely seen by Thomson (1992d).


32

Thomson et al. (1992b) also explored the incidence of dispersal. This was highest

when population densities were at their peak and food supply was low. In some

instances whole packs were observed to disperse and in other instances lone dingoes,

whose chance of survival was comparatively slim, were seen. Vacant areas were a

requirement for dispersing dingoes. These were usually more available in areas where

humans were managing dingo population numbers via control and their staple prey,

red kangaroos, had abundant feed. This created a “catch-22”: if dingoes were

controlled near farmland, predator-free modified landscapes with abundant resources

for herbivores were made available. Emigration to farmland by dingoes would

subsequently increase since water, food and space for territories were abundant.

Requirements for dingo control would consequently increase. If dingoes were not

controlled near farmland, livestock loss may occur, but according to Allen (2000),

overall livestock loss may increase due to control. Hence, land managers may create

an endless need for dingo control or management by “controlling” or “managing”

dingo populations.

1.1.2.7 Management

Based on movement patterns, Harden (1985) made two points clear:

1. Limited movements and small home ranges indicated control only needed

to occur near or adjacent to affected properties; and

2. Delivery of poison baits by fixed wing aircraft would be ineffective to

target dingo travel routes such as creeks, ridgetops and firetrails.

Thomson (1992d) suggested that a buffer zone as large as the home range of a dingo

with sufficient resources, adjacent to livestock properties, needed to be kept free of


33

dingoes. Dispersing dingoes would presumably settle, not migrate onto farmland and

be effectively controlled by managers. To maintain efficiency, rivers and creeks

needed to be targeted and it was recommended that campaigns coincided with

biological seasons when dingoes were known to be more active.

These methods for control indicated livestock losses will be minimised with

continuous effort at specific times of the year. Dingo management prior to

implementation of buffer zone management techniques (Harden 1985; Thomson

1992d) was aimed at eradication but did not account for recruitment of dispersing

dingoes from other areas. Fleming et al. (2001) provided more detail on the history

and current status of dingo management than is required in this study. Settlers using

exclusion fencing, land clearing, poisoning, trapping and shooting were responsible

for the dingo becoming extinct during the 1800’s in the majority of south eastern

Australia. Shepherding slowly became a last resort for farmers as professional dog

trappers were employed by the state to target troublesome dingoes. Bounty systems

to promote dingo control were effective from 1836 until the 1990’s when research

revealed they were subject to fraud and ineffective (Allen and Sparkes 2001).

Poisoning has included use of strychnine but is now generally confined to close

regulation of compound 1080 (sodium monofluoroacetate) applied to fresh meat and

manufactured baits. Apparent “improvements” in the technique of aerial baiting

validated this approach as an effective method of dingo control (Thomson 1986;

Thompson et al. 1990), with ground baiting also a frequently used method. Public

scrutiny of lethal control, public interest in preservation of dingoes, and concern

regarding the impact of poisonous baits on native animals have improved due

diligence over administration of poisonous baits in recent times.


34

Legislation has led to a paradox for historical and contemporary dingo management.

Davis (2001), Fleming (2001) and Fleming et al. (2001) outlined the major details but

the dilemma is that dingoes are protected under acts enabled to protect native species,

yet are declared pests under acts enabled to protect livestock enterprises. Legislation

differs between states and territories of Australia also. In the Australian Capital

Territory (ACT), dingo control requires a permit whilst in NSW, the state that

surrounds the ACT, land owners are obliged to control dingoes as pests. There is no

indication that discrepancies in legislation between states and territories will be

changed in the near future. The case of the dingo is not the only instance where a

wild canid requires control and conservation. Macdonald and Sillero-Zubiri (2004a)

and Sillero-Zubiri et al. (2004) discussed a number of topics ranging from cultural

prejudices, economics, development, politics, improvements in animal husbandry,

animal welfare and maintenance of biodiversity that need to be considered when

attempting to manage carnivore populations for conservation.

1.1.2.8 The dingo “ex situ”

Legislation within particular states and territories permits dingoes to be kept as pets

(cf. Fleming 2001; Oakman 2001). Aside from this, dingoes are found in most

zoological preserves and within kennels registered by dingo preservation societies.

There is considerable emphasis on maintenance of dingo “purity” in privately held

and zoological collections, though this concept remains difficult to define and

validate. Registration of dingo breeders and “pure” dingoes on a national register is

recommended, with education of the public and the agricultural sector on

conservation efforts involving dingoes also important. Since hybridisation in the wild

is seen as a key threatening process for dingoes, isolation of “pure” dingoes in large


35

areas is suggested to have merit (Fleming 2001). Large offshore islands or fenced

areas are one extreme conservation method that may maintain “pure” dingo stock in

wild conditions. Such extremes might not be necessary, however, if improvements in

management for contemporary wild dingo populations are made on mainland

Australia.

1.1.2.9 The dingo “in situ”

Effective management of wild dingo populations will require a re-evaluation of the

dingo in Australia and the industries this animal is reported to threaten. Holistic, best

practice management objectives which include but are not limited to the topic areas

highlighted by Sillero-Zubiri et al. (2004; cf. previous section 1.1.2.6 for the cyclic

conflict between land managers and dingoes) will need to be included. In NSW, the

Rural Lands Protection Acts have traditionally defined dingoes as pests that require

control. Amendments to the Rural Lands Protection Act (Wild Dogs) 1998 identified

key habitats where dingo control could be minimised for conservation of “pure” dingo

stock, pending further research (Anon. 2000). The SGBMWHA is one area identified

as having potential for conservation of wild dingo populations. This amendment was

the foundation for this research project and the research site is described section 1.2.

1.2 The Southern Greater Blue Mountains World Heritage Area

This study site is significant because it is part of the GBMWHA and is the largest

water storage facility for the population of Sydney. The SGBMWHA is located

approximately 65km west of Sydney central business district (Figure 1.5) between

Latitudes 33°43’00” and 34°18’00” and Longitudes 150°00’00” and 150°30’00”. The


36

eastern face is generally bordered by small hobby farms and the towns of Penrith,

Warragamba, Oakdale, Picton and Mittagong. South borders hobby farms on the site

boundary and livestock enterprises between Mittagong and Taralga. State forests

(pine plantations) provide a partial buffer zone between the study site and livestock

enterprises on the western border, in parts. The northern boundary abuts the Great

Western Highway, an arterial road from/to Sydney, the western railway and the city of

Katoomba.

Figure 1.5: Location of the SGBMWHA in relation to Sydney (b). = Site 1 (top) and Site 2 (bottom); ■ = Total SGBMWHA area; ■ = Schedule two water catchment protection zone; = Lake Burragorang; Broken lines represent road infrastructure and political boundaries of land management organisations. The whole of the SGBMWHA has eleven major areas dedicated for conservation.

These include:

1. Blue Mountains National Park (BMNP)

2. Kanangra-Boyd National Park (KBNP)

3. Nattai National Park (NNP)

4. Burragorang State Recreation Area (BUSRA)

5. Yerranderie SRA (YSRA)

6. Nattai SRA (NSRA)

Site 1

Site 2


37

7. Bargo SRA (BASRA)

8. Lake Burragorang (Largest water storage for Sydney)

9. Schedule one catchment protection zone

10. Schedule two catchment protection zone

11. SCA area of operations

Beside governance by the Department of Environment and Climate Change (DECC),

the Oberon, Katoomba and Picton offices of the Parks and Wildlife Division (PWD)

of the DECC, and the SCA, Wollondilly and Oberon Local Governments, Moss Vale,

Central Tablelands and Goulburn Rural Lands Protection Boards (RLPB), the Blue

Mountains World Heritage Institute (BMWHI) and working groups formed by the

Department of Primary Industries (DPI – Agriculture) cooperatively oversee and/or

manage the region. In addition to this are private land tenures within and bordering

the park, aboriginal groups, heritage groups, recreational and research groups that all

make contributions to land stewardship. This section will focus on aspects of the

natural landscape of the three major national parks and the history of occupation by

Aborigines and Europeans. Much of this information is readily available within plans

of management for the region which were the information source in the proceeding

discussion unless stated otherwise (cf. Anon. 2001a, b, c). Management objectives

for: a) the PWD of the DECC for the natural areas; b) the SCA for protection of water

assets; c) the RLPB; and d) the DPI for protection of livestock enterprises bordering

the park will be outlined in section 1.2.6.

1.2.1 Climate and hydrology

Five major river tributaries flow into and through the SGBMWHA forming the

catchment system for Lake Burragorang (Cox’s, Kowmung, Nattai, Wingecaribee and


38

Wollondilly Rivers). The Kowmung River and two of its largest tributaries, the

Jenolan and Kanangra Rivers, provide the largest source of high quality water to Lake

Burragorang. Since the construction of Warragamba Dam on the Warragamba River

in the north eastern corner of the region, the SGBMWHA has contained 70 per cent of

water storage needs for the population of Sydney (Anon. 1995). Prior to the

construction of Warragamba Dam in the 1960’s, waters flowed through to the

Hawkesbury-Nepean river system, sometimes inundating agricultural enterprises and

settlements of the Cumberland Plain in the outskirts of Sydney during flood (Rosen

1995).

The SGBMWHA currently experiences a temperate climate. Average temperatures

recorded since European records began range between 5°C and 16°C during cooler

months and 16°C and 29°C in warmer months. Lowest recorded minimum and

highest recorded maximum temperature range of the SGBMWHA region is -10.1 to

42.8°C respectively (Anon. 2001a, b, c). Similar variation is observed in average

annual rainfall, ranging between 812.6mm and 1410mm (Anon. 2001a, b, c). Large

variations in temperature and rainfall can be expected for the region due to variations

in altitude from 160m to 1158m above sea level from the eastern to the western

borders respectively.

1.2.2 Geology and topography

Devonian sediments deposited during the Permian and Triassic periods up to 350

million years ago formed the general geology of the greater Blue Mountains region.

Uplift and 90 million years of erosion has characterised the topography with steep

cliffs and deep valleys. Volcanic rocks are present in the areas of Yerranderie and


39

Bindook. These are associated with a major relic crater filled with volcanic breccia

and ashfall tuffs, covering approximately 40km2

. Outcrops of Silurian limestone in

several creek systems further enhanced the geological diversity of the area.

Kanangra-Boyd NP contains limestone deposits which have formed a karst system.

Yerranderie SRA was once mined for silver-lead sulphide ores and gold, which were

emplaced during volcanic activity. Parts of BMNP, NNP and BSRA were mined for

coal throughout much of the twentieth century. Soils found in the area have very low

fertility and are described within the Penrith and Katoomba 1:100 000 map sheets

(Bannerman 1990; King 1994).

1.2.3 Vegetation and habitat

The BMNP plan of management lists 40 distinct plant communities that are found

within the park. These may not all be within the SGBMWHA portion of the BMNP;

however, the statement reflects the geological diversity, variations in climate and

topography, fire history and variations of aspect in the area. Dry forests cover 45% of

the park, woodlands cover 38% and the remainder currently consists of heaths, low

woodlands, moist forests and rainforests. Nattai NP is dominated by dry schlerophyll

eucalypt-dominated associations. Kanangra-Boyd NP is comprised mainly of

heathlands, swamps, wet and dry schlerophyll forests, rainforests and rocky outcrops

which all contribute to habitat structure. Parts of the SGBMWHA are cleared of

native vegetation due to a history of farming.

Fire management in the protected areas includes fuel or hazard reduction burns where

excess understorey and leaf debris is depleted using low intensity controlled fires.

The intent of these is to manage the effects of fire on plant and animal communities


40

that may or may not require fire to complete or commence parts of their biological

cycle. Wildfires are more common and intense during summer months when dry

northwest and westerly winds prevail, but they may not occur every year. Lake

Burragorang is subject to accelerated effects of erosion, sedimentation and

eutrophication processes following fire that may decrease the quality of water in

storage. After fire some fauna species are believed more prone to predation due to

loss of protective cover, which is of concern to land managers.

1.2.4 Fauna diversity

Variations in topography, microclimates, vegetation and habitat across the park create

a patchy distribution of native and introduced fauna. Blue Mountains NP has records

for 46 mammals including marsupials and monotremes, more than 200 birds, 58

reptiles and 32 amphibians. Introduced mammalian species include pig Sus scrofa,

goat Capra hircus, fallow deer Dama dama, horse Equus caballus, cattle Bos taurus,

European rabbit, and occasionally domestic dogs lost by pig hunters. A number of

threatened species in the area include birds, invertebrates and mammalian species

such as bats and the koala Phascolarctus cinereus, tiger quoll Dasyurus maculatus,

brush-tailed rock wallaby Petrogale penicillata, long-nosed potoroo Potorous

tridactylus, yellow-bellied glider Petaurus australis and the squirrel glider Petaurus

norfolcensis. These animals are threatened due to the initial and ongoing invasive

effects of European colonisation, such as altered fire regimes and introduced game

and livestock animals.


41

1.2.5 Occupation and disturbance

There are numerous culturally significant sites arising from Aboriginal and European

occupation of the area. Aboriginal heritage dates back at least 14,000 years from

archaeological records for the Blue Mountains and 20,000 years for the greater region

including coastal areas. Records have indicated the SGBMWHA incorporates areas

inhabited by the Wiradjuri (south west area), Gundangarra (Kanangra and southern

area) and D’harawal (Nattai/south east area) Aboriginal people. Management of the

area by Aborigines is thought to generally consist of firestick farming practices to

clear undergrowth, hunt and increase abundance of certain types of plant food. A

rock art painting of a dingo can be found in NNP, potentially reflecting the cultural

significance of this animal to local Aborigines.

Europeans first entered the area when attempting to cross the Blue Mountains. Ensign

Francis Barrallier entered the Nattai valley and established a depot to conduct

explorations from 1802. Barrallier penetrated as far as the Kowmung River valley

that year and reported disputes between his greyhounds and the local dingoes

(Macqueen 1993). Occupation of the Burragorang valley dates back as early as 1824,

approximately 36 years after colonisation. There was an influx to the SGBMWHA of

part-time explorers and graziers with the discovery of precious metals at Yerranderie

in 1871. Use of the area by graziers and colonists in general meant domestic dogs

were introduced into the valley for herding and company. Extant dingo populations

were poisoned, shot and trapped to minimise loss of livestock to predation. Due to the

construction of Warragamba Dam in the 1950’s, many of the leases were inundated

by Lake Burragorang. “Joorilands” homestead was the last agricultural enterprise

vacated in 1993 and is still maintained as a relic of significant local heritage. The


42

cessation of agriculture at Joorilands reduced the need for dingo control in the core of

the SGBMWHA. Remaining populations of animals affected by the settlement of the

region in 1824 have since repopulated areas they had been excluded from. All

animals with the exception of introduced species specified in section 1.2.4 now have

legislative protection as native species under the NSW National Parks and Wildlife

Act 1974.

1.2.6 Management regimes

National parks in Australia are categorised under Category II: protected area

managed mainly for ecosystem protection and recreation, within the IUCN protected

area management categories. Over one million hectares of the GBMWHA was

inscribed on the IUCN World Heritage List in November 2000 for satisfying two

criteria applied for outstanding universal significance of natural values, including:

1. Criterion II - Outstanding examples of significant ongoing ecological and

biological processes in the evolution and development of ecosystems and

communities of plants and animals, particularly eucalypt-dominated

communities; and

2. Criterion IV – Important and significant natural habitats for in-situ

conservation of biological diversity, including the eucalypts and eucalypt-

dominated communities, primitive species with Gondwanan affinities such as

the Wollemi Pine, and a diversity of rare or threatened plants and animals of

conservation significance.

Management of the SGBMWHA will now be explained in context for the

organisations with governing jurisdiction. All organisations are controlled by the


43

state government of NSW, and the priority of each is to serve the public, to protect

food and water supplies, and to preserve cultural and natural heritage.

1.2.6.1 Parks and Wildlife Division

Park management plans (cf. Anon. 2001a, b, c) provided comprehensive information

on protected areas and objectives for their management. These have formed the basis

for much of the preceding discussion. General management objectives for national

parks are stated in the NSW National Parks and Wildlife Act 1974 and include:

• Protection and preservation of scenic and natural features;

• Conservation of wildlife and natural biodiversity;

• Maintenance of natural processes as far as is possible;

• Preservation of Aboriginal and historic sites, features and places;

• Provision of appropriate recreation opportunities; and

• Encouragement of scientific and educational inquiry into environmental

features and processes, prehistoric and historic features and park use patterns.

Parks and Wildlife Division have three major management plans for the region which

include KBNP (west), BMNP (central) and NNP (east) which incorporates BUSRA,

YSRA and NSRA. Policies and management objectives for native animals within

these parks apply to dingo conservation because maintaining dingo populations

maintains natural processes. Dingoes are classed as wild dogs and subjected to

specified objectives for control of feral species also. These contradictory policies

frame part of the objectives of this research (section 1.3), to identify whether the

dingo maintains ecological processes or disrupts them.


44

1.2.6.2 Sydney Catchment Authority

The role of the SCA is to capture, store and supply quality raw water from well-

managed catchments. The current study was mainly resourced by the operations

division of the SCA because they controlled access into the scheduled areas for water

protection, accommodation, Citizen Band (CB) radio provisions and road

infrastructure. To minimise effects of erosion, roads in the area were closed for half a

day after eight millimetres of rain had fallen and closed for one or more days if ten or

more millimetres had fallen. River crossings were closed if water levels were in

excess of 0.8m. From an SCA perspective, dingoes may carry zoonoses transmissible

by water and require control. Alternatively, they may control pest species such as

feral pigs that carry zoonoses that humans may be more susceptible to, and require

higher conservation value in the catchment areas. The SCA objectives aim to

promote healthy ecosystems, hence, implying dingoes need to be conserved.

1.2.6.3 Rural Lands Protection Boards

The RLPB are funded by ratepayers to address problems at a local level. There are 47

boards in NSW (currently under review) which consist of directors elected by

landholders every four years. The mission for each board is to be highly accessible to

the community and to respond to queries by ratepayers. In relation to dingoes and this

research, the RLPB system aims to: a) protect the community from invasion by pest

animals; b) protect animal production industries from diseases; c) take a leadership

role in liaison with other community organisations; and d) keep the NSW Government

fully informed of rural issues and of the benefits provided to the community. The

current study is based within the Moss Vale RLPB, which is taking an alternative

approach to dingo management by supporting this research, to identify best practice


45

holistic management and appropriate policies. Boards are committed to the provision

of quality service, inclusion of ratepayers within activities and collaboration with

local groups and agencies for the effective transmission of accurate information.

1.2.6.4 Department of Primary Industries

The vision statement for the NSW DPI is profitable, adaptive and sustainable

primary industries. The DPI acts in partnership with industry and public sector

organisations to foster development of this vision for primary industries in NSW. The

department adopts an integrated approach to development of policy and scientific

research, commercial services, regulation, education, advice and corporate services. It

is in a leading position to provide primary industries with the technology and

partnerships needed for profitable and sustainable production and influence

environmental and natural resource access debates.

All stakeholders mentioned above collaborate with landholders through the Oberon

Wild Dog Working Group to manage the local wild dogs2

. Working groups for wild

dog management were an initiative by the NSW DPI to achieve collaboration for

sustained control of wild dogs where they present a problem to industry. The Oberon

group manages all areas west of the Wollondilly River, including both transects

within this study area.

2 Wild dog is a generic descriptor of dingoes, feral domestic dogs and their hybrids.


46

1.2.7 Study sites

Two 25km transects were selected within land scheduled as priority habitat for dingo

conservation in the Amendments to the Rural Lands Protection Act (wild dogs) 1998.

Both transects were located behind locked gates within schedule one and schedule

two restricted access catchment protection zones which minimised human intrusion.

Sites 1 and 2 were approximately 12km apart by air and 16km by road. Logistical

considerations were important when selecting sites since the Wollondilly River is

subject to flooding when water is pumped from Wingecaribee Dam to Lake

Burragorang and cannot always be crossed by a vehicle. Sites were therefore

established on the western side of Lake Burragorang and the Wollondilly River

because access remained open via alternative routes. Figure 1.6 outlines Site 1, Site 2

and significant topographical features.

Figure 1.6: Location of Site 1 (top red oval) / transect 1 (top black line) and Site 2 (bottom red oval) / transect 2 (bottom black line) in the SGBMWHA . Blue line outlines Lake Burragorang, thin red lines outline schedule 1 and schedule 2 restricted access areas and thin orange lines are road infrastructure.


47

1.2.7.1 Site 1 – Scotts Main Range

Site 1 is a ridge named Scotts Main Range which extends north from Yerranderie,

heading directly in line with Katoomba. There is history of farming in the area and

the Catholic Bushwalkers Club has inholding property in Schedule 2 lands which they

used annually as a depot for recreational activities. The western aspect of the range is

bounded by the Kowmung River, which flows into the Cox’s River, and the eastern

aspect of the range is the Butchers Creek gully, from which water flows directly into

Lake Burragorang (Figure 1.7). Water is abundant and small dams occur

approximately every kilometre along Scotts Main Range. Annual maintenance of the

road as a firebreak promoted growth of herbaceous plants on the edges which may

have attracted herbivores to the road. Plate 1.1 displays a small part of Site 1.

Figure 1.7: Site 1 (Scotts Main Range, black line) road infrastructure (light orange lines), political boundaries (green and red lines) and Lake Burragorang (outlined by blue lines).

▲North


48

Plate 1.1: Aerial photograph of Scotts Main Range road at Site 1 showing the heavily dissected topography characteristic of the study area.

1.2.7.2 Site 2 – Tonalli to Joorilands

Site 2 generally consisted of regenerating ex-pastoral leases. The transect meandered

over ridges and through creek beds adjacent to the Tonalli River, which flows from

Yerranderie into Lake Burragorang, and continues along the foreshore of Lake

Burragorang past the mouth of Byrnes Creek, then over the foothills of Centre Ridge

to the Wollondilly River and Joorilands. This is illustrated in Figure 1.8. Water is

abundant along the entire transect due to Lake Burragorang, the Wollondilly River

and small dams found approximately every kilometre, though water levels in the dams

was more variable in Site 2 than in Site 1. Roads were maintained less frequently

along this transect. Eastern grey kangaroos were abundant and seen more frequently

in open grassy habitat. Plate 1.2 displays the southern portion of Site 2, looking south

from foothills of Centre Ridge towards Joorilands.


49

Figure 1.8: Site 2 (Tonalli ♦ to Joorilands ●, black line) road infrastructure (light orange lines), political boundaries (green and red lines) and Lake Burragorang and Wollondilly River (outlined by blue lines).

Plate 1.2: Looking south from foothills of Centre Ridge towards Joorilands at Site 2.

▲North


50

1.3 Project design

This project was designed to investigate as many aspects as practicable of the ecology

of dingoes in the SGBMWHA to better understand their functional role. Functional

role of dingoes is defined as the dynamic responses of dingoes to immediate factors of

the environment including changes in populations of sympatric (competing and prey)

species (Krebs 2001). As a result, some of the “deficiencies in knowledge and

practice” raised by Fleming et al. (2001; cf. Wilton et al. 1999; Dickman and Lunney

2001; Wilton 2001; Daniels and Corbett 2003; Elledge et al. 2006) were addressed

along with questions raised within the second wild dog control order under the Rural

Lands Protection Act 1998. Data were collected for comparison with past studies

introduced in section 1.1.2 on genetics, diet, activity, movements and spatial

organisation to identify the type of canid endemic to the SGBMWHA.

1.3.1 Hypothesis

If dingoes in the SGBMWHA are a top order predator in this ecosystem, then

measures of diet, abundance, and activity and movement patterns will demonstrate

their functional role in this landscape.

1.3.2 Aim

The aim of this study was to document aspects of the ecology of dingoes, dingo X

domestic dog hybrids or feral dogs in the SGBMWHA using techniques with

minimum invasiveness.


51

1.3.3 Objectives

The above hypothesis and aim were addressed using five objectives:

1. To assess the gene pool for previously identified “pure” dingo alleles and

compare purity data with data for genetic relatedness and to assess the

relatedness of sampled individuals;

2. To compare biological parameters including morphometrics and pelage

variation with previously reported biological parameters from other dingo

populations;

3. To identify diet and seasonal dietary shift through hair and bone analysis from

scats;

4. To assess changes in activity and abundance of medium ground dwelling

vertebrate fauna and their relationship with dingo activity, abundance and diet;

5. To investigate short and long term dingo movement patterns, including

dispersal, and intraspecific spatial organisation in the core of the SGBMWHA.

1.3.4 Experimental approach

From the outset, this study was positioned to develop a database on numerous aspects

of the ecology of “pure” dingoes, dingo X domestic dog hybrids, feral dogs or a

combination of these “wild dogs” from the SGBMWHA. Therefore, a cross

disciplinary approach was required and is summarised in Figure 1.9. As has been

standard with past studies (cf. section 1.1.2), the population had to be defined

according to genetic and morphometric descriptors. To gain an understanding of the

ecological role of the SGBMWHA “wild dogs”, dietary analyses needed to be

comparable with that of an introduced canid, the red fox, and past studies detailing the

diet of wild canids in Australia. Patterns of activity had to be comparable with dietary


52

analyses (recorded as frequently) and previous studies so differences between

biological seasons could be tested, if there were biological seasons. Data on

movement patterns and spatial organisation of canids in the study area also had to be

comparable with previous studies of diet and activity. Using a layered approach, this

study was arranged to describe the “wild dogs” of the SGBMWHA, their diet, patterns

of activity, changes in abundance, movement patterns and spatial organisation.

Figure 1.9: Experimental approach used in the current study to better understand aspects of the ecology of dingoes in the SGBMWHA . NB: Genetics affect biology, which affect patterns in prey consumption which affect abundance which affect movements which affect function and management of dingoes by dingoes and dingoes by humans.

1.4 General methods

The conduct of this research required ethics approval from the University of Western

Sydney Animal Care and Ethics Committee (ARP 04/010) and a scientific licence

from DECC (licence number S11482). A safe work plan for working in remote areas

was prepared and approved by collaborators upon completion of accreditation for: a)

Safe use of firearms (Pest Control Officers Course Level 2: no 398) for ethical

euthanasia of fauna injured as a consequence of experimental procedures; b) Certified

statement of attainment in chainsaw operations (level 1); and c) Documentation of

first aid certification.

Management Dingo function Dingo movements Dingo abundance Dingo diet Dingo biology Dingo genetics Bottom up

Approach


53

1.4.1 Animal capture and handling

Professional dingo trappers from the NSW RLPB were employed to maximise trap

effort and capture success within a short timeframe (maximum of 10 days per trapping

program). Trapping events occurred in the dingo breeding seasons of April 2005,

April 2006 and March 2007, June 2007 (whelping season) and September 2007

(rearing season).

1.4.1.1 Capture and restraint

Dingoes were captured using Oneida Victor ® ‘Soft Catch’ No. 3 (Woodstream

Corporation, Pennsylvania USA) off-set leg hold traps (here after called Victor traps)

due to observed capture success rates and low injury rates reported (Meek et al. 1995;

Fleming et al. 1998). During each trapping program trappers were requested to

extend the potential trap effort to 400 trap nights over ten days. The jaw spread on

Victor traps is 15cm and these traps are regularly used to trap foxes, coyotes and

dingoes (Fleming et al. 1998). Newly purchased traps were modified by filing the

plate and trip tongue so less pressure was required to set the trap off under the weight

of an animal.

Traps were first secured to a drag (a strong log), using chain and wire, to slow down

the retreat of trapped animals. Logs were usually 2-3m long and their strength was

tested prior to use by beating both ends on hard ground to see if they could withstand

breakage. The chain was securely attached in two locations to the log, usually before

and after a protuberance on the log, with sturdy wire (16-17Ga). Drags were placed

behind the trap and the chain was dug into the soil. Exposed chain and wire on the

log were covered with bark, leaves or both. Trap sites were selected by the trappers


54

using trained domestic dogs to detect3

the scent of dingoes in the area, the

interpretation of the landscape by the trapper or by using information collected during

the course of this research from sand plots (cf. section 1.4.2) and dingo sightings.

Kneeling on a canvas mat in front of the trap location, trappers would dig a hole with

a small mattock or shovel, slightly deeper and slightly larger than the Victor trap

when set. Soil from the hole was placed on the canvas mat and used to stabilise the

trap, which was tested by pushing on the four corners of the jaws (Plate 1.4) and

rectified if required. Depending on the trapper, loose grass or stringy bark rubbed into

a ball were placed under or around the plate or fly screen mesh or waxed baking paper

were placed over the plate to stop soil or other material affecting the trip mechanism

from being set off under the weight of an animal standing on the plate. Remaining

soil on the canvas mat was then sifted over the trap to fill in around the plate and to

hide the trap. Sifting was required to remove any sticks or rocks from the soil and

prevent such material from being caught between the jaws when traps were set off. A

sprinkling of home-made lure, urine or faeces collected from the dogs of the trapper

were placed approximately one hand span away from the plate. Placement of lure on

surrounding bushes or rocks was intended to raise the nose of the dingoes off the

ground and distract the animal from smelling human scent or the trap. Upon

achieving this, the dingo would tread on the plate of the Victor trap when leaning into

the bush to urinate or defecate and re-mark its territory.

3 Dingo trappers used male and female dogs, behaving and marking territory as a pack. Females in oestrus were sometimes preferred depending on the dingo sign detected. Trappers ignore the effects of male/female catch rates because a bitch on heat can draw attention from a wild bitch (competition) as well as a wild dog (dog meaning male dingo in search of a mate).


55

Plate 1.3: An example of a Victor®

trap being set in front of a bush scented with lure.

Traps were checked daily, from first trap set to last trap set to reduce time animals

spent in traps. Any sprung traps were reset. Non target captures were released except

for foxes and cats which were euthenased in accordance with legislation and PWD,

SCA, RLPB and DPI management objectives. Trapped dingoes were restrained for

handling by placing a dog-catching noose around the neck using an ACES Catch Pole

(Animal Care Equipment & Services Pty Ltd, Moorabbin, Victoria) to secure the

animal and minimise injury to the animal and animal handlers (Plate 1.4). Once the

head of the animal was secure, the back legs were held and the animal was lifted off

the ground and secured to a restraint board (Plate 1.5). The restraint board was made

by cutting a 1m X 0.8m square from an unused backyard plastic table. A dingo sized

domestic dog was used to measure where slits were required on the board for securing

the neck, shoulders and hip. Narrow slits (6cm x 2mm) were cut in the holding board

to secure neck, shoulder and hip sections of restrained animals. Suitable lengths of


56

approximately 5cm wide seat belt fabric and large sturdy plastic buckles were

threaded through the slits and used to secure dingoes to the board. The restraining

straps remained adjustable to accommodate larger and smaller dingoes and to ensure

dingoes were tightly secured when attached to the restraint board. Other items

attached to the board included a canvas blind and two ropes, one attached from the top

left to the bottom left corner, the other to the right hand corners, so the board could be

carried and the animal could be weighed using hand held scales. Once the captured

dingo was secure, a canvas blind, secured to the top left of the board, was laid over

the head to reduce stress and calm the animal. The trap was removed from the foot

and the foot was gently massaged to restore circulation. The trap was reset by the

trapper whilst samples were collected and morphometric measurements were

recorded.

Plate 1.4: A dingo caught in a trap with a trapper approaching to noose its neck. Note the immobile left paw and the drag attached to the trap.


57

Plate 1.5: The dingo from plate 1.4 secured to the board at the conclusion of operation. Note the canvas blind to minimise stress to the animal (top left), the location of straps around the neck, shoulders and hip to secure the animal to the board and the rope attached to the board to carry the dingo into a suitable area for operation and weighing the animal.

1.4.1.2 Morphometric measurements

Morphometric measurements were recorded on a data sheet (cf. appendix). For

comparison with past research by Newsome and Corbett (1982;1985), Jones (1990)

and Thomson (1992a), measurements included head length, ear length4, shoulder

height, hind foot width and length5

4 Some measurements were first performed incorrectly

, tail length and bushiness, total length, weight,

dentition (for fracture, decay or loss and age), approximate age, colour and distinct

markings, an evaluation of “purity” or hybridisation from phenotypic appearance, as

defined by Corbett (2001) and Fleming et al. (2001), and any apparent injuries or sites

of infection from mange or fighting with other animals. Table 1.2 outlines minimum

and maximum morphometric measurements for male and female dingoes whilst Table

5 Some measurements were first performed incorrectly


58

1.3 details how measurements were taken in this study. Dingoes were marked with a

collar (cf. chapter four) and an ear tag (placed on the proximal side of the left ear) for

resighting. A Trovan unique™

microchip transponder for animal identification was

inserted subcutaneously between the shoulder blades of each animal for subsequent

identification of recaptured animals that lost their ear tag or collar. Identification of

each dingo in this study was by site (1 or 2) and the number on the ear tag. For

instance, dingo 2.1 in plates 1.4 and 1.5 was the first dingo captured in Site 2. With

the exception of dingo 2.37 being tagged incorrectly, this method enabled easy

numeration of captures.

Table 1.2: Minimum and maximum morphometric measurements of dingoes as described by Corbett (2001) or Thomsonb

Measurement

(1992b).

Male criteria Female criteria

(mm) Min. Max. Min. Max.

Head length 215.4 234 202.9 226.9

Shoulder height 554.5 616.4 537.9 587.9

Total length 1168.5 1373.3 1143 1298.8

Weight (kg) 13 21.44 11.3 b 17.9b


59

Table 1.3: Details of measurements collected (c.f. appendix) Measurement Method Head length Head length was measured from the tip of the nose to the tip of the sagittal crest

(Corbett 2001). Ear length The length of the left ear was first measured using tape measure from the back of

the ear to the top of the ear. Once corrected, the left ear was measured from the node at the bottom on the inside front of the ear to the top of the ear with pes callipers.

Shoulder height

The front left leg was stiffened by pushing the elbow forward and pulling the foot down. The tape measure was held at the tip of the left shoulder blade and measurement was recorded from the edge of the straightened foot pad (toe nails were excluded).

Hind foot width

Callipers were held face up, level with the foot pad and across the foot so the middle digits were level with the main bar. Callipers were tightened to the relaxed foot and the measurement recorded.

Hind foot length

Measured from the foot pad to the bent heel.

Tail length Measured from the coccyx, located between the two hip bones, to the terminal part of the tail (hair tip was not included). Tail was held hard against the tape measure.

Total length Tape measure was held against the animal at points to maintain tension from the nose tip to the tail tip (as specified above) following the vertebral column.

Weight Ropes attached to the board were placed over the hook of the 50kg scale and the animal was lifted off the ground. Weight was recorded to the nearest 0.5kg (Plate 1.6). The board was weighed at 2kg in the same fashion

Dentition The head of the animal was held, lips were lifted and teeth were assessed when taking saliva samples (Plate 1.7)

Age Age was determined by combining weight: age relationships defined by Jones (1990) and dentition. Age via dentition was usually determined by the presence (0-2 years) or absence (>2 years) of fleur-de-lis (crown like shapes on the animals incisors; see Plate 1.7) and tooth eruption. Teeth, however, were always fully erupted and animals were too old to use head length measurements defined by Catling et al. (1991). The teeth x-ray technique defined by Thomson and Rose (1992) could not be used because the animals were not being destroyed.

Colouration Basic colour options were recorded and drawn on a schematic dingo diagram.

Plate 1.6: Weighing a dingo.


60

Plate 1.7: Assessing dentition whilst collecting saliva samples for fracture, decay, loss and the presence or absence of fleur-de-lis on the incisors.

1.4.1.3 Collection of samples

Samples collected included an ear biopsy (Plate 1.8), hair, saliva and drops of blood

from the ear biopsy on an FTA ® Card6

(Whatman International Ltd, Kent, England)

for DNA tests (Plate 1.9). Details of methods of extraction are specified below (Table

1.4) however, tests were conducted and results were produced at external laboratories.

Researchers at University of New South Wales, Sydney, Australia were consulted to

assess wild dingo “purity” whilst Genetic Technologies Limited (Melbourne,

Australia) was consulted to assess population structure. One quarter of each sample

collected was sent for analysis.

Table 1.4: Details of samples collected Sample Method Ear biopsy A sterilised livestock ear punch was used to collect ear biopsies (Plate 1.8). A thin

piece of cardboard was held behind the ear to promote a clean punch. Any punch that did not completely sever was held with artery forceps and cut free with surgical scissors or a scalpel blade. Biopsies were placed directly into 5ml sample jars half filled with ethanol. The ear punch and scissors were washed with ethanol before and after each sample was collected. Samples were stored in the freezer prior to analysis.

Hair Hair with follicles was removed from the back of each animal using artery forceps and placed directly into 5ml sample jars half filled with ethanol. Forceps were washed before and after each use with ethanol. Samples were stored in the freezer.

Saliva Buccal swabs were rubbed inside the mouth, over and between gums and lips for 20-30 seconds then placed in 5ml sample jars or the container they were provided with. Samples were stored dry.

FTA ® Card Five drops of blood were placed into both circles on the FTA card (Plate 1.9). FTA cards were stored dry prior to analysis.

NB: Great care was taken to not get any human DNA on any of the samples during collection.

6 FTA Cards contain chemicals that lyse cells, denature proteins and protect nucleic acids from nucleases, oxidative and UV damage, inactivate organisms and prevent the growth of bacteria and other micro-organisms.


61

Plate 1.8: Collecting an ear biopsy and securing an ear tag on the proximal side of the left ear.

Plate 1.9: Four drops of dingo blood on an FTA Card. Blood was extracted from the tube of the butterfly needle7

in the background.

1.4.2 Additional techniques

Most techniques used were specific to each chapter and are introduced when

necessary, such as diet analysis and the sand plot technique (chapter three). Based on

the seasons used in past studies (cf. section 1.1.2.3), each year in the current study was

separated into four general periods: whelping season (June - Aug.); rearing season

(Sept. – Nov.); exploratory season (Dec. – Feb.); and breeding season (Mar. – May).

One technique which was not part of major analyses was the use of passive camera

traps and ad hoc sightings. TrailMAC (A-Tec, Middletown, USA) motion-sensing

digital camera traps were wired to tree trunks or stumps in locations of known high

dingo activity for resighting data and general information on the dingo population.

These provided additional information which assisted the interpretation of genetic

relatedness data (chapter two) and movement data (chapter four). Cameras were

programmed to take one photograph within three seconds of the passive infra-red

7 Butterfly needles were used to extract blood from the saphenous vein for disease analyses which were not included in this study.


62

beam being broken and to reset within 30 seconds. A maximum of six cameras was

set for three days and nights per site per field trip. Most cameras were positioned to

photograph along the road, rather than across the road, to improve chances of

photographing animals travelling quickly. A sprinkling of home-made lure was

placed approximately three feet from the camera to attract any travelling dingoes into

view of the lens. Sighting data were used in a similar fashion. When a dingo or

dingoes were sighted they were generally filmed with a mini digital video recorder.

Plates 1.10 (a-c) are examples of the types of photos taken by the passive camera traps

and Plate 1.10d is a photograph taken whilst filming dingoes feeding on an eastern

grey kangaroo carcass during an ad hoc sighting.

a. b.

c. d.

Plate 1.10: An emaciated dingo (a); dingo 2.1 rolling in lure (b. Ear tag and collar are in view and she was also observed alone during all ad hoc sightings cf. chapter four Figure 4.1); three juveniles showing interest in the lure (c. The outer Perspex lenses were easily damaged by dust and hazy photographs were common); and a pack of dingoes feeding on an eastern grey kangaroo during an ad hoc sighting (d. NB: Three individuals in this pack had been captured during April 2006 and the remaining three were captured during March 2007).

Chapter Two:

What is a “pure” dingo?


64

Chapter Two: What is a “pure” dingo?

2.1 Introduction ....................................................................................... 65

2.1.1 Skull allometry, coat colour and epigenetic effects .............................. 66

2.1.2 Genetic research on the family Canidae ............................................... 76

2.2 Methods .............................................................................................. 84

2.2.1 Animal handling and operation ............................................................ 84

2.2.2 Skull measurements, genetic “purity” and population structure ......... 85

2.2.2.1 Skull measurements ............................................................... 85

2.2.2.2 Genetic “purity” ..................................................................... 86

2.2.2.3 Population structure .............................................................. 90

2.3 Results ................................................................................................ 91

2.3.1 Animal handling and operation ............................................................ 91

2.3.1.1 Capture rates .......................................................................... 92

2.3.1.2 Morphometric measurements ............................................... 92

2.3.2 Skull measurements, genetic “purity” and population structure ......... 95

2.3.2.1 Skull measurements ............................................................... 95

2.3.2.2 Genetic “purity” ..................................................................... 96

2.3.2.3 Population structure ............................................................ 104

2.4 Discussion ......................................................................................... 116

2.4.2 Population genetic structure of SGBMWHA dingoes ........................ 117

2.4.3 Skull allometry, coat colour and epigenetic effects ............................ 125


65

2.1 Introduction

Introgression of dingo genes with domestic dog genes has been identified as a

management dilemma for the conservation of “pure” dingoes (Daniels and Corbett

2003). Attempts to define “purity” of wild dingoes have been limited to comparison

of morphological measurements (Corbett 2001), skull allometry (Macintosh 1976;

Newsome et al. 1980; Corbett 2001), and genetic biotechnology (Wilton 2001).

Newsome and Corbett (1982) cross bred dingoes with domestic dogs in captivity over

six years to investigate hybridisation. Results from their study have since been used

to classify wild living dingoes as “hybrids” or “pure” (Jones 1990; Woodall et al.

1996) though Jones and Stevens (1988) and Jones (1990) concluded that the “hybrids”

they sampled were more dingo than domestic dog and used patterns of reproduction to

support their premise.

Contemporary studies on the identification, biology, behaviour and ecology of

Australian dingoes were published between 100 and 150 years after the dingo fence

divided south eastern Australia from northern and central Australia. Dingo population

numbers were severely reduced on the south eastern side of the fence at this time,

which limited the distribution of dingoes to mountainous, forested habitats. Chapter

two therefore describes the dingoes residing in the SGBMWHA that have been

geographically isolated from other dingo populations in south eastern Australia by

urbanisation and rural development. For comparison with previous studies, analyses

of data from skull allometry, morphometric measurements, pelage variation, genetic

“purity” and genetic population structure of dingoes in the SGBMWHA are shown.

Research on developmental biology, skull morphology and genetic biotechnology


66

techniques is critically appraised to demonstrate how genetic and phenotypic

descriptors could be developed for studies of the family Canidae. With these in mind,

characteristics of “pure” dingo populations using genetic biotechnology techniques

(Wilton et al. 1999; Wilton 2001) and skull allometry (Newsome et al. 1980) alone

are questioned. Dingoes of the SGBMWHA are defined in this study by cross-

referencing collected data with techniques used in previous studies, and application of

emerging techniques.

2.1.1 Skull allometry, coat colour and epigenetic effects

As stated in chapter one, dingoes were originally classified as “pure” or “hybrid”

according to skull characteristics, coat colour and morphometric measurements

(Newsome et al. 1980; Newsome and Corbett 1982; 1985). Newsome et al. (1980)

stated that dingo skulls have characteristics for functions linked with predation. When

compared with domestic dog skulls, dingoes have a longer muzzle, a flatter cranium

and a higher nuchal crest. Main teeth in dingos are larger and the canines are longer

and slightly more slender for mastication and shearing flesh (Newsome et al. 1980).

These features are supported by other research on skull allometry of wild and

domestic canids (Wayne 1986; Wayne and Vilà 2001) used when defining

characteristics of top order predators.

The taxonomic status of dingoes has previously been reviewed (Elledge et al. 2006).

These authors reached the same conclusions as other studies on dingo “purity”,

showing that reference specimens are unreliable to accurately describe dingoes and

new discriminators are required. Jones (1990) first questioned the validity of

classifying dingoes as “pure” or “hybrids” using calculations of skull allometry when


67

he studied dingoes in the north eastern highlands of Victoria. Jones (1990) stated that

the wild canids he studied could not be called domestic dogs but could not be called

dingoes either because measurements recorded were outside of those specified in past

research (Newsome et al. 1980; Newsome and Corbett 1982; 1985). Conclusions

made by Jones and Stevens (1988) and Jones (1990) are relevant to the current

chapter which also questions the validity of the techniques currently used to classify

wild dingoes as “pure” or “hybrids”. Conclusions from those studies included:

1. No reliable methods available to classify the present canids into categories of

“purity”;

2. Canids with coat colour, canonical score and physical conformation of “pure”

dingoes were present;

3. Less dingo-like canids could not be called hybrids because genetic structure

could not be accurately quantified;

4. Extreme, domestic dog like traits for head, body and limb shapes were not

present;

5. Uniform, dingo-like physique was common;

6. Reproductive patterns (one short monoestrous breeding cycle per annum) did

not follow patterns expected for feral dogs (biannual breeding cycles); and

7. Feral dogs and /or “hybrids” were conclusively few in number.

An increased range of variability in skull shape, coat colour and physique compared

with dingoes considered “pure” was observed by Jones (1990). “Purity” was,

however, assigned according to measures identified by Newsome et al. (1980) and

Newsome and Corbett (1982; 1985) for comparative analysis. Notes on hybridisation

based on skull scores suggested that introgression had occurred but Jones (1990) only


68

speculated an explanation for hybridisation. Similarly, Woodall et al. (1996),

studying dingo skull allometry in Queensland, were equivocal about why, where and

when “hybrid” or “pure” specimens were collected. Comparative analysis, by Jones

(1990), of dingo coat colours identified in the Victorian highlands before 1974 (n =

320) with his study (n = 414) showed ginger/tan animals had decreased by 9% and

black and tan animals had increased by 9%. Sable colouration was not recorded by

Jones (1990) although it was recorded before 1974 in the Victorian highlands and in

central Australia. Brindle colour decreased by 2%, black increased by 6%, straight

white, another “pure” dingo colour, decreased by 4% and patchy colouration was not

recorded (Jones 1990). Jones (2009) reviewed his previous results and agrees that the

canids examined form a single population of dingo-like canids, not three sub

populations.

Contemporary studies on inheritance of coat colour (Roemer et al. 1997; Blewitt et al.

2004), skull morphology (Herring 1993; Trut 2001; Trut et al. 2004) and other

phenotypic descriptors (Gilbert 2002) for vertebrates and invertebrates have shown

that use of these techniques alone to characterise populations may be limited. Herring

(1993) for instance, showed skull allometry can vary due to a range of environmental

circumstances such as mechanical loading. Mechanical loading was defined by

Hildebrand (1988), cited by Herring (1993), as the compressive, tensile or shearing

force applied to a solid object. Herring (1993) stated that due to “the logical

connection between loading and the function of the skeleton, many proposals have

invoked loading as a direct or indirect causative agent in skeletogenesis”. In terms of

dingo ecology, mechanical loading is the pressure of exertion used by dingo facial

muscles to shear flesh and masticate prey. Mechanical loading, or in other words


69

topographic isolation, prey selection and diet, can therefore alter the skull shape of

dingoes.

Variations such as these skull changes are classed as epigenetic in this study under a

broad definition which includes all phenotypic variations other than those caused by

DNA sequence change. The definition includes both heritable and non-heritable

traits. Natural selection can produce most of these changes provided genetic variation

on which to work, however, epigenetic processes can induce phenotypic variation

without requiring nuclear DNA variation. Some authors have used slightly different

interpretations of epigenesis but in general it is used to describe phenotypic variations

that are not explained by conventional Mendelian genetics (Monk 1990; Herring

1993; Lyon 1993; Henikoff and Matzke 1997; Roemer et al. 1997; Gilbert 2002; Van

Speybroeck et al. 2002; Rakyan and Whitelaw 2003; Ashe and Whitelaw 2006;

Rakyan and Beck 2006; Goldberg et al. 2007). Waddington (1942), cited by Henikoff

and Matzke (1997), coined the term “epigenetic” to describe changes in gene

expression during development. Such changes are heritable and caused by changes in

chromatin structure and/or DNA methylation. The activation or inactivation of genes

by DNA methylation includes gene silencing (Monk 1990), genetic or parental

imprinting (Lyon 1993), ‘developmental noise’ (Blewitt et al. 2004) and epigenetic or

transgenerational epigenetic inheritance (Roemer et al. 1997; Rakyan and Whitelaw

2003; Ashe and Whitelaw 2006). Goldberg et al. (2007) explained epigenetic effects

as the bridge between genotype and phenotype; the study of stable heritable change in

gene expression or cellular phenotype. Epigenetic effects have been demonstrated in

a range of vertebrate and invertebrate species (Gilbert 2002; Monk 1990; Roemer et

al. 1997; Blewitt et al. 2004) and biological differences can occur within species in


70

one generation. Ruvinsky (2001) described developmental genetics in the family

Canidae from gametogenesis through embryonic development, development of the

notochord, neural crest and during morphogenesis. His descriptions were consistent

with epigenetic manifestations during cranial development in vertebrates (Herring

1993).

Dingo embryo(s) develop a neural crest which produces mesenchymal cells after the

epithelial mesenchymal transformation phase. Following division, mesenchymal cells

become committed to a phenotypic pathway for fabrication of unique tissue types

(Caplan 1991) such as melanocytes for expression of coat colour (Sponenberg and

Rothschild 2001). Post birth, the skull develops the required mechanical loading

(Herring 1993) for the local prey and allometry is affected by the “biomechanical

programming” as originally suggested by Morey (1992). Epigenetic models predict

that genes exposed to different environmental influences can lead to variations in

phenotypic expression (Gilbert 2002; Van Speybroeck et al. 2002). Due to temporal

changes and the range and changing nature of anthropocentric influences across

ecosystems, epigenetic effects on dingoes are likely to vary across their range. Many

external factors may affect development of phenotype in dingo populations.

Differences between these populations in genetic variation, skull allometry and

colouration are perhaps, therefore, not related to hybridisation alone. Epigenetic

processes and natural selection may explain these differences.

Bioaccumulation of endocrine disrupting compounds (EDC) is known to be

particularly persistent in long-lived organisms feeding at the top end of the food chain

(Crews et al. 2000). EDC can be naturally occurring or synthetic and have been


71

shown to either activate or inhibit endocrine pathways, depending on the receptors

they bind to during cell signalling (Kelce et al. 1998; Crews et al. 2000). In addition,

EDC can contaminate water resources as anthropogenic chemicals, such as hormones,

and their effect on ecosystems is of concern (Colborn et al. 1993; Kelce et al. 1998;

Zacharewski 1998; Crews et al. 2000; Choi and Jeung 2003). Industrial processes

(such as wool scour, livestock husbandry and agriculture) coupled with large

(residential) and small scale sewage treatment plants which impinge on the five major

river systems of Lake Burragorang are potential sources of EDC in the SGBMWHA.

Studies of the impacts of EDC have shown that mammals can be affected through

decreased fertility, demasculinisation and feminisation, and alteration of immune

function (Colborn et al. 1993).

Endocrine disrupting compounds and other epigenetic activities are applicable to

dingo ecology for two reasons: 1) Dingoes are found at the top end of the food chain;

and 2) Research has shown lactating dingoes consume the urine and faeces excreted

by their young (Baverstock and Green 1975). Atchley et al. (1991) concluded that

postnatal maternal performance, including quality and quantity of milk and milk

growth factors, affected young through epigenetic means. Research to identify uterine

effects, epigenetic influences and postnatal skeletal development in the mouse

concluded that some authors confounded heritable and non-heritable components with

“environment” under the term “epigenetics” (Atchley et al. 1991). Atchley et al.

(1991) stated that genetic models therefore needed to consider direct genetic effects

from the genotype of the progeny, heritable epigenetic affects from the genotype of

the progeny, indirect epigenetic effects from interactions of two separate genotypes

and non-heritable environmental effects, such as EDC. It appeared that the genotype


72

of the mother and the child interacted with the uterine, nursing and post-weaning

environments which demonstrated local and systemic affects on the adult skeleton

(Atchley et al. 1991). These data are consistent with hypotheses by Ruvinsky (2001)

on epigenetic effects in the family Canidae.

In an attempt to answer questions regarding the evolution of the domestic dog, Trut

(2001) selectively outbred silver foxes Vulpes vulpes from fur-farm bred populations

in Russia. The general aim of the experiment was to observe adaptations in

behavioural traits from foxes that retained standard phenotype, biological function and

wildness to that of domestication, and conditioning of fox behaviour to the presence

of human beings (Trut 2001; Trut et al. 2004). From 50,000 offspring studied over 40

years, Trut et al. (2004) showed that adaptation responses to domestication had

occurred within six generations and phenotypic variations were commonly seen due to

experimental selection pressure.

Observations made by Trut (2001) and Trut et al. (2004) indicated that within isolated

populations of canids, major genetic and phenotypic changes can occur within very

few generations. Dingoes from southeast Australia may therefore differ

morphologically to dingoes from central and northern Australia due to the

construction of the dingo fence in 1885. In contrast, Corbett (2001) and Fleming et

al. (2001) speculated that the areas where dingoes were less controlled, in the north

western half of Australia, were mostly “pure” dingoes according to canonical analyses

of skull measurements (Figure 2.1). Their data are consistent with hypotheses of

epigenetic effects on skull allometry. If control programs were to change social and

feeding habits, then skull allometry may also change in response.


73

Newsome and Corbett (1982) reported that three out of eight parental dingoes during

research experiments in captivity returned “ intermediate” canonical scores. Trut

(2001) reported cranial changes within eight generations of foxes reared in

experimental conditions. Scott (1968) also reported cranial changes in captive bred

wolves. It is evident from these studies that life in captivity from birth can affect

cranial development in the family Canidae. Similarly, it could be postulated that food

supply and methods of hunting could create rapid adaptive morphological response to

environmental conditions (Herring 1993). Observations made during past

experiments of captive-bred, primitive canid species may, therefore, have been

affected by the food the subjects were fed.

Wayne (1986; 2001) explored relationships between cranial morphology of wild and

domestic dogs to understand consequences of domestication. The canid species and

breeds studied exhibited strong allometric scaling of cranial growth. Epigenetic

effects described in those studies confirmed why cranial morphology is so diverse in

Figure 2.1: Locations of dingoes showing “hybrid” canonical scores due to selection pressure from being persecuted and locations of dingoes showing “pure” canonical scores due to natural selection pressure (Adapted from Corbett 2001). Words in brackets are the original descriptors of the dingo populations used by Corbett (2001)


74

the family Canidae. Both studies indicated that future research needs to focus on the

uncertain role of developmental mechanisms (Wayne 1986) and on identifying genes

that influence growth rate and timing of skeletal development (Wayne 2001).

Phenotypic variations in captive bred foxes included colour changes such as

yellowing-brown mottling and piebaldness, floppy ears, curly and short tails,

abnormal tooth-bite (under-bite, similar to bulldog breeds) and other craniological

changes (Plate 2.1) (Trut 2001; Trut et al. 2004). They also showed that pigmentation

loci were related to genes associated with regulation of behaviour and development.

Intangible variation, similar to that observed by Blewitt et al. (2004), was also

recorded within litters. Trut (2001) adduced expression of coat colour genes for gene

silencing. Reference to the appearance of the once missing fifth digit on the hindleg

of canid species is another example used by Trut (2001) to explain how regulatory

embryonic actions and gene activity or inactivity can induce variation in the family

Canidae. During the fox domestication process a change in seasonal reproduction

patterns was observed, whereby oestrous cycled in some vixens during autumn and

spring, though extra-seasonal matings were rare.


75

Plate 2.1: A piebald domesticated silver fox (a), a curly tailed domesticated silver fox (b), a mottled domesticated dog (inset) and a mottled domesticated silver fox (c) and comparison of non-domesticated silver fox farm silver fox skull (left) with a domesticated silver fox farm silver fox skull (right) showing variations in width and length (d). All photos © Trut (2001)

Transformations of dimensions in the domesticated silver fox cranium were

associated with widening of the face skull and a decrease in the width and height of

the cerebral skull (Plate 2.1d) (Trut 2001). Changes were more prominent in males

selected for domestication, whose skulls were smaller in almost all proportions when

compared with male foxes bred for fur (Trut 2001). In comparison with cranial

morphology between wild and domestic dogs, Wayne (1986) stated that the skull

length within all dog breeds mimics exact allometric dwarfism due to a lack of

developmental variation when they are compared with wolf-like canids. One

exception was that Gray wolves tended to have longer teeth than similar sized

domestic dogs. Wayne (1986) suggested that these results were due to natural and

artificial selection respectively, a comment that directly related to epigenetic

inheritance of variation due to environmental adaptations. Figure 2.2 illustrates

susceptibility of the neonate canid skull to developmental processes. Further,

b.

c. d.

a.


76

alterations in the timing and rate of postnatal growth in the canid skull can assist the

evolution of new cranial proportions (Wayne and Vilà 2001).

As discussed in chapter one, the “pure” dingo is reported to have three general coat

colours: tan/ginger, black and tan and white (Corbett 2001). Preceding studies,

however, showed “pure” dingoes may also be sable, brindle, black or even black and

white (Newsome and Corbett 1985) depending on how historical reports were

interpreted. Trut (2001) demonstrated how coat colour can vary from a uniform silver

coat to a range of unexpected colours within a selectively bred fox population. Thus

the question arises: If a high density outbreeding population of wild dingoes were

living in optimal wild conditions, similar to farm-bred domesticated foxes, would

selection pressure cause variations in phenotype? Results from the fox research

indicated substantial variations could occur within six generations (Trut 2001) and

evidence suggests that domestication of the dog began by selecting traits exhibited by

wolves (Vilà et al. 1997; Wayne and Ostrander 1999)

2.1.2 Genetic research on the family Canidae

Research on genetic variation in the family Canidae has embraced colour (Sponenberg

and Rothschild 2001), population structure (Lehman et al. 1992; Girman et al. 1997),

origin (Vilà et al. 1997; Wayne and Ostrander 1999), and genetic diversity (Wayne

Figure 2.2: Comparison of neonate canid skull (a.) with an adult canid skull (b.) showing the profound developmental alterations that occur from neonate dog to adult dog (Wayne 1986; 2001; Wayne and Ostrander 1999)


77

and Ostrander 1999). In this section the importance of identifying population

structure in communal living wild canid populations is presented. Perceptions of

“purity” and the success or otherwise of microsatellites to define dingo “purity” are

addressed, as is the subsequent limitation of research on mtDNA to identify an origin

for dingoes. Research on the genetic relatedness of the dingo to other canid species is

used to illustrate the position of the dingo genotype in the phylogenetic tree of the

family Canidae.

The origins of the wolf were introduced in chapter one (Figure 1.1). Figure 2.3 is an

adaptation from Vilà et al. (1997) which detailed the divergence of a number of

domestic dog breeds from the wolf and coyote sequences. Vilà et al. (1997) also

proposed an alternative for divergence of domestic dog breeds (Figure 2.4) which

suggested that the dingo diverged as a separate breed earlier than domestic dog

breeds.


78

Figure 2.3: Expanded view of neighbour-joining tree of domestic dog mtDNA haplotypes within clade I (of IV) based on 261 base pairs (bp) of control region dog sequences (adapted from Vilà et al. 1997) to illustrate the apparent position of the dingo (black arrow) amongst domestic dog breeds and how genetic data can provide variations for interpretation (NB: Multiple haplotype lineages for the Leonberger, chow chow, border collie, Irish setter, Tibetan terrier, German shepherd, English setter, samoyede, golden retriever, Labrador retriever, Siberian husky and crossbreeds).


79

Figure 2.4: Expanded view of clade I (of IV) of neighbour-joining tree of 8 wolf and 15 dog genotypes within clade I (of IV) based on 1030 bp of control region dog sequences (adapted from Vil à et al 1997) to illustrate the apparent position of the dingo (black arrow) amongst domestic dog breeds (NB: Two haplotype lineages were found for the papillon breed). Labels for identification of similarities or dissimilarities between 261 bp and 1030 bp sequences and indications for bootstrap support have been excluded.

So what is a dingo? Ignoring “purity” altogether, Figure 2.4 suggests the dingo

separated off the evolutionary line from wolves and coyotes before most domestic dog

breeds. Longer mtDNA sequences were used by Vilà et al. (1997) to construct Figure

2.4 which provides a more reliable measure of relatedness. Figure 2.3 shows dingoes

arising after the separation of some domestic dog breeds. Leonard et al. (2002)

deduced that 80% of domestic dog mtDNA haplotypes, including those of ancient

breeds, fall on Clade I, in accord with Vilà et al. (1997). Originating in “the new

world” and Oceania, ancient dogs were indistinguishable from Eurasian domestic


80

dogs, a conclusion that is consistent with Figure 2.4. More importantly, Vilà et al.

(1997) and Leonard et al. (2002) highlighted that the dingo samples they tested placed

them within Clade I, “the most diverse clade of dog sequences”. Savolainen et al.

(2004) claimed that the dingoes of Australia formed as a result of inbreeding due to

low population numbers within the last trickle of domestic dogs entering Australia

from Asia. There is potential, as suggested by Corbett (2001), that it was not a trickle

of dingoes between Asia and Australia. Introgression from dingoes of Asian descent

may have been common for the 3000-4000 years before Europeans arrived and may

be considered as a form of hybridisation, casting doubt over the concept of genetic

“purity”. One has to assume Vilà et al. (1997) sampled captive dingoes but no

research on dingo “purity” was published until Wilton et al. (1999), two years later.

There is an imperfect record of the dingo “breed” in captivity and there are no records

to show when the captive dingoes for reference genetic sampling became isolated

from wild dingoes. The notion of dingo “purity” might therefore be a construct of

human thought rather than the end result of an evolutionary lineage.

Other studies which extended the phylogenetic tree to domestic dog breeds included

Wayne and Ostrander (1999), Leonard et al. (2002), Savolainen et al. (2002) and

Wayne and Ostrander (2007). Alternatively, Zrzavý and Řičánková (2004) provided a

supertree, various strict consensus trees and various parsimonious trees from

cytochrome subunits, a simplified stability tree and a preferred phylogeny of the

family Canidae tree. They used protein coding genes and morphological,

developmental, ecological, behavioural and cytogenetic characters to define the

phylogenic relationships of all canid species, and concluded that the various

phylogenies of the wolf-like canids show little agreement. Vilá et al. (1997) used


81

mtDNA from coyotes, wolves and subspecies of the wolf to identify the origin of the

domestic dog. Using different methods, both studies showed a spectrum of possible

canid clades (Zrzavý and Řičánková 2004) or multiple origins of domestic dogs (Vilá

et. al. 1997).

Vil à et al. (1997) also showed dingoes clustered with the New Guinea singing dog,

the African basenji and the greyhound without explanation. Close clustering of the

dingo with the New Guinea singing dog is understandable due to the close proximity

of Papua New Guinea to Australia. Using microsatellite markers from “purebred”

breeds, Parker et al. (2004) assigned the greyhound to a cluster of European herding

dogs that then branched into European hunting dog breeds. The African basenji in

Parker et al. (2004) clustered separately to the European greyhound, however, and

was in strong statistical consistency with the nine breeds most closely related to the

Gray wolf as classified by the techniques they used. The Siberian husky was included

within the nine breeds most closely related to the wolf, evolving after the basenji, and

was also characterised taxonomically on the dingo branch of like breeds produced by

Vila et al. (1997). Alternatively, the Australian shepherd, an American breed used in

the research by Parker et al. (2004), was one of four breeds which failed to

consistently cluster with others of the same breed and Parker et al. (2004) do not

mention using dingoes.

The most definitive studies of the dingo genome were by Wilton et al. (1999), Wilton

(2001) and Savolainen et al. (2004). Wilton et al. (1999) attempted to define dingo

“purity” and sampled for microsatellite variation in Australian dingoes. In their study,

levels of microsatellite variation in captive dingoes of unspecified origin and dingoes


82

found in kennels of dingo conservation societies and in Australian zoological parks

were typed for similarities within breed, and dissimilarities with domestic dog breeds.

One aspect of the methods used by Wilton et al. (1999) to characterise “purity” was

that the effects of comparing captive-bred tame dingoes with wild-born dingoes

(Newsome et al. 1980, Newsome and Corbett 1982, Wilton et al. 1999, Wilton 2001)

were not accounted for. The effects of environmental influences reviewed in section

2.1.1 showed that characteristics such as coat colour, skull allometry and morphology

can change the characteristics of animals in captivity that could lead to selection bias

in sampling. Wilton (pers. comm. 2006) stated that dingoes on Fraser Island, off the

coast of Queensland Australia, carried alleles that were not found in mainland

Australia dingo populations. Data are yet to be published however genetic variation

such as this suggests captive-bred dingoes may not represent the full spectrum of

dingo alleles. As Wilton (2001) suggests, examination of early museum specimens is

needed to provide a better baseline for assessment of “purity”.

Using archaeological records and live specimens, Savolainen et al. (2004) found

dingo mtDNA types formed a single cluster around A29, a central haplotype found in

dingoes and domestic dogs. Based on the assumption that haplotype A29 was the

only founder type, Savolainen et al. (2004) estimated the arrival of the dingo in

Australia to be between 4,600 and 5,400 years ago. This was consistent with

speculations made by Corbett (2001) for the arrival of the dingo in Australia. The

largest representation of wild and captive dingoes sampled (n = 141 of 211) were

from south eastern Australia (Savolainen et al. 2004). Dingo mtDNA data were

compared with pre-European archaeological dog samples from Polynesia, 654


83

domestic dog samples and 38 wolf samples with anonymous life histories to track the

trail of the dingo from southeast Asia to Australia.

Future genetic research on dingoes then is to sample nuclear DNA from widely

distributed dingo populations, including captive-bred groups, and also identify the

genetic relationships between individual dingoes, and spatial relationships of dingo

packs in the wild. Perhaps the original allele distributions defined by Wilton et al.

(1999) were observed in all captive-bred dingoes, and less often in wild dingoes, due

to relatedness within captive bred dingo populations. Lehman et al. (1992) studied

wolf packs and expected genetic results from analyses of VNTR and mtDNA to show

that the packs consisted of a dominant breeding pair and their offspring. Nine of 27

packs sampled showed that one pair and their offspring did not account for all the

individuals in a pack. Dispersal among packs of near relatives, or the packs that

shared a common boundary, was common and contributed to population structure,

although each pack comprised closely related individuals. Lehman et al. (1992) used

genetic relatedness data to identify packs that were derivatives from larger packs

without intensive radio telemetry or aerial surveys used by Thomson et al. (1992a) on

dingoes. Girman et al. (1997), studying African wild dogs, also identified an alpha

pair and adults within each pack to be unrelated using 14 microsatellite loci. The high

level of unrelated individuals within packs and dispersal behaviour apparently limited

opportunities for incestuous mating.

Wayne et al. (2004) summarised canid genetic research. Their criticism of the use of

mtDNA to assess genetic variability within a population was the prevalence of biased

genetic expression due to maternal inheritance of mtDNA. As in research by Lehman


84

et al. (1992) and Girman et al. (1997), microsatellite DNA are best used to identify

familial relationships because of its high mutation rate and maternal and paternal

inheritance. Since dingoes are hypercarnivorous communal-living canids, a similar

genetic structure to that reported for wolves and African wild dogs can be expected in

the current study.

Morphometric measurements, colouration, skull allometry and genetic “purity” of

samples collected in the current study are analysed in this chapter. The relationships

between colour, morphometric measurements and “purity” are compared with data on

the genetic structure of the SGBMWHA dingo population. Objectives one and two

are investigated and the dingoes of the SGBMWHA are described.

2.2 Methods

Techniques used for capture and restraint, morphometric measurements and the

collection of samples are described in chapter one. This section describes the

techniques used to analyse collected data for morphometrics, skull allometry and

genetic “purity” and genetic structure of the population.

2.2.1 Animal handling and operation

Capture data for target and non target species were recorded in relation to trap effort.

Morphometric measurements were compared with dingo studies in the Eastern

Highlands of Victoria (Jones 1990), Victorian highlands, Central Australia and

Kakadu National Park (Corbett 2001). Although Thomson (1992a) did not publish

morphometric data, weights recorded by Thomson (1992a) exceeded those by Corbett


85

(2001) and Jones (1990) and were also used for comparison. Maximum, minimum or

average measurements of dingo populations were compared.

2.2.2 Skull measurements, genetic “purity” and population structure

Samples compared for genetic tests, specified in section 1.4.1.3, were separated and

sent to external laboratories for analysis. One lab tested for “purity” (University of

New South Wales, Sydney, Australia) and the other tested for relatedness of

individuals (Genetic Technologies Limited, Melbourne, Australia).

Use of two

laboratories assisted data interpretation for this genetic dataset.

Collection of skull specimens was not a focus in this research which employed

techniques to minimise disturbance to the population. Invasiveness was minimised by

taking ear biopsies, blood and hair or saliva samples from live animals for DNA

analyses. Genetic data were used for measurements of introgression between packs,

relatedness and comparisons with genetic “purity” data. Skulls were collected from

observed mortalities and measured in accordance with Newsome et al. (1980) for

comparison with “purity” scores from other studies. These are discussed further in

section 2.2.2.1.

2.2.2.1 Skull measurements

Skulls from deceased individuals were collected and prepared by slowly boiling in

water until all flesh was removed. Teeth that became dismantled were glued back into

the specific orifice with water soluble glue and the specimens were bleached in 5%

peroxide: water ratio. Measurements taken were compared with Newsome et al

(1980), Jones (1990), Woodall et al (1996) and Corbett (2001). Skull lengths were


86

measured on grid paper, similar to Plate 2.1d. Measurement of skull length was not

usually recorded in past research (with the exception of Woodall et al. 1996) although

head length is a common measure. Since populations of canid species will be subject

to different environmental conditions, skull measurement data were included for

comparison with traditional (skull allometry) and contemporary (genetic)

classifications of dingo “purity” ( cf. section 1.1.2 in chapter one and section 2.1 in the

current chapter for additional information on skull allometry).

2.2.2.2 Genetic “purity”

Wilton et al. (1999) and Wilton (2001) specified details of sample preparation for

DNA extraction, genetic markers used, PCR amplification and statistics employed for

“purity” tests. Ear biopsies, hair follicles, buccal swabs or FTA Cards were collected

and prepared in accordance with methods described by Wilton et al. (1999) and

Wilton (2001). Microsatellite loci typed and used in this study included AHT103,

AHT109, AHT125 (Holmes et al. 1993), FH2079, FH2138, FH2175, FH2199,

FH2247, FH2257, FH2293, FH2313 (Mellersh et al. 1997), CXX30, CXX109,

CXX402, CXX406, CXX410, CXX434, CXX460 (Ostrander et al. 1993), LEI008

(Mellersh et al. 1994), PEZ1 (US Patent 05874217), CPH2, M13TT and M13C19

(Wilton pers. comm. 2007). Wilton et al. (1999) and Wilton (2001) also used

additional, previously described canine microsatellite loci (Holmes et al. 1993, 1994;

Mellersh et al. 1994; Ostrander et al. 1994). Table 2.1 shows the origin of these loci.


87

Table 2.1: Origin of microsatellite loci reference samples typed by Wilton et al. (1999) and Wilton (2001) for “ purity ” tests.

Microsatellite loci Types of canines sampled Reference AHT103, AHT109, AHT125

Unrelated dogs from a variety of breeds including 11 miniature poodles, two Irish setters, one miniature long-haired daschund and 13 Bedlington terriers. The usefulness of these microsatellites varies from breed to breed due to in-breeding within particular breeds.

Holmes et al. (1993)

CXX30, CXX109, , CXX402, CXX406, CXX410, CXX434, CXX460

Mixed breed dogs sampled in Iowa, USA. Ostrander et al (1993)

LEI008 A full -sib, two generation Irish setter family and 30 unrelated dogs of various breeds.

Mellersh et al. (1994)

FH2079, FH2138, FH2175, FH2199, FH2247, FH2257, FH2293, FH2313

212 “purebred” and mixed breed dogs from the USA housed under protocols approved by an ethics committee.

Mellersh et al. (1997)

Most markers were chosen for their differences in allele frequencies between

domestic dogs and captive dingoes. Some were considered diagnostic for “purity” or

hybridisation and were used to estimate the proportion of domestic dog ancestry in

SGBMWHA dingoes (Wilton 2001). The loci M13TT and M13C19 have alleles

which differ by insertions of two base pairs in the captive dingoes and domestic dogs,

and the absence of a distinctive dingo allele is considered diagnostic of domestic dog

ancestry (Wil ton pers. comm. 2006). Diagnostic markers like M13TT are ideal for

portraying ancestry specific to closely related species. Fluoro-labelled primers

specific for the loci in Table 2.1 were obtained commercially (Research Genetics,

Proligo Australia, Lismore NSW, Australia, and Applied Biosystems, Foster City CA,

U.S.A.). Procedures used for polymerase chain reaction (PCR) amplification were

described by Wilton (2001). GENESCAN and GENOTYPER software (PE

Biosystems, Foster City CA, U.S.A.

) were used to size the PCR products relative to

the GS-500 size standard.

Frequencies of microsatellite alleles in reference captive dingo samples and reference

domestic dog samples (Wilton 2001) were used to estimate the likelihood that any


88

genotype would occur in either group. The combined likelihoods were based on all

the genetic types for each sample. Alleles (presence or absence) were inserted into an

algorithm to assess the likelihood their origins were in a domestic dog population, in a

captive dingo population, in a hybrid population (50% dog, 50% dingo) or in a three

quarter (3Q) dingo X domestic dog hybrid population (75% dingo, 25% domestic

dog). Probabilities were compared using a relative likelihood Log of odds (LOD)

logarithm to determine:

a) The probability the genetic type was from one “pure” dingo and one domestic

dog parent: P(dingo/dog);

b) The probability the genetic type was from one hybrid dingo and one domestic

dog parent: P(hyb/dog);

c) The probability the genetic type was from one “pure” dingo and one hybrid

dingo parent: P(dingo/hyb);

d) The probability the genetic type was from one “pure” dingo and one ¾ dingo:

¼ domestic dog hybrid parent: P(dingo/3Q).

NB: Points a - d are relevant to results Table 2.11

An increase in the number of tests performed increases the likelihood of detecting past

hybridisation events (Wilton 2001). In the current study, P(dingo/3Q) was divided by

the number of tests conducted to determine an average 3Q score and increase the

accuracy of “purity” estimates. Average 3Q scores in reference captive dingo groups

were greater than 0.58 (Nesbitt et al. 2000). This indicated any samples from “pure”

dingoes will score higher than 0.58, although 0.5 is considered a reasonable medium.

Some alleles were never, or very rarely, observed in reference captive dingo samples.

Genetic types that were at least ten times more common in dogs than dingoes were


89

defined as dog-like alleles. The presence of dog-like alleles in a sample is considered

indicative of domestic dog ancestry within the lineage. Figure 2.5 shows the

differences between calculated hybrid mixture proportions and reference alleles from

captive dingoes and domestic dogs using the LOD score.

0.00

0.20

0.40

0.60

0.80

1.00

0.00 0.25 0.50 0.75 1.00

Prop. rare

Pro

p. d

oglik

e

0.00

0.20

0.40

0.60

0.80

1.00

-1.00 -0.75 -0.50 -0.25 0.00 0.25Average 3Q

Pro

p. d

oglik

e

0.00

0.20

0.40

0.60

0.80

1.00

-1.00 -0.75 -0.50 -0.25 0.00 0.25

Average 3Q

Pro

p. r

are

Figure 2.5: Results of genetic analyses for calculated hybrid mixture proportions ( = Wilton et al. 1999 and Wilton 2001) with reference domestic dog genes ( = Wilton et al. 1999 and Wilton 2001) and reference captive dingo genes ( = Wilton et al. 1999 and Wilton 2001). Prop. doglike = Alleles ten times more common in domestic dog samples compared with captive dingo samples; Prop. rare = Proportion alleles rarely found in dingoes; Average 3Q = Likelihood the sample is from a ¾ dingo population.

From Figure 2.5, individual animals in a population can be designated as “pure”

dingoes, dingo X domestic dog hybrids or domestic dogs. Based on genotype

information from captive dingoes and relative frequencies of each genotype in the

domestic dog, data for hybrid animals were calculated. Average frequencies of alleles

shared between these two populations are then used to broadly classify animals as

having 25%, 35% or 50% domestic dog genes. Genotypes for assessment of

hybridisation in the present study were tested for 47 dingoes using 23 microsatellite

a. b.

c.


90

loci (Wilton et al. 1999; Nesbitt et al. 2000; Wilton 2001) to assign each animal to

one of seven categories outlined in Table 2.2.

Table 2.2: Assigned scores of dingo “purity” or percentage of hybridisation Score Description

1 “Pure” dingo - high average 3Q, no doglike alleles 2 Likely dingo - acceptable average 3Q in reference range 3 Likely hybrid with small amount of dog - average 3Q just below reference range >75% dingo 4 Less than 75% dingo - average 3Q less than 0 5 Less than 65% dingo - average 3Q less than -0.1 6 Less than 50% dingo - average 3Q less than -0.25 7 No detectable dingo ancestry - average 3Q less than -0.5

Addition of rare alleles only found in dingoes such as M13TT and rare alleles only

found in domestic dogs were included and assisted characterisation of the genotype

for score assignment. Due to the positive correlation between assigned “purity” score

and number of loci tested, as many loci as possible were examined for each sample.

Discernable differences existed in samples assigned scores with ten loci compared

with scores assigned by 20 loci when compared with the reference range for domestic

dogs and dingoes. To be certain a dingo was “pure” , the LOD had to score higher

than 0.5 within the reference range from captive dingoes.

2.2.2.3 Population structure

DNA was extracted from ear biopsies, hair follicles and FTA Cards using Qiagen

minikits as per the blood spot protocol in the Qiagen minikit handbook. Standard

conditions in the Qiagen multiplex master mix handbook were used for DNA

amplification using the Qiagen multiplex master mix and to set universal cycling

conditions for the thermocycler. Animals were genotyped with 22 canine

microsatellite markers (Table 2.3). STRUCTURE V. 2.2 (Pritchard et al. 2000), a

model based Bayesian clustering algorithm, was used to assign individuals to genetic

clusters. The parameters used were a burn in length of 50000 with 50000 Monte


91

Carlo Markov Chain (MCMC) repeats, the admixture ancestry model and the

correlated allele frequency model (F model). Falush et al. (2003) stated the F model

could assign individuals within sub populations accurately. The number of clusters

was determined both by calculating the posterior probability and using the Evanno

delta K criterion (Evanno et al. 2005). Clustering was determined without reference

to spatial data. Comprehensive details of the STRUCTURE V. 2.2 algorithms are also

available from Falush et al. (2003) and Falush et al. (2007). Inbreeding coefficients

(FIS), expected heterozygosity (HE), rarefacted allelic richness (A) and population

differentiation (FST

) were calculated in FSTAT V. 2.9.3 (Goudet 1995).

Table 2.3: Microsatell ite loci typed for relatedness tests and origin of microsatellite loci reference samples.

Microsatellite loci Types of canines sampled Reference AHT121 Public domain marker from 25

unrelated dogs of various breeds Holmes et al. (1995)

FH2001, FH2164, FH2289, FH2326, FH2611, FHC2010, FHC2054, FHC2079

Public domain markers from unrelated mixed breed and “purebred” domestic dogs, probably from USA

Francisco et al. (1996)

PEZ1, PEZ11, PEZ12, PEZ13, PEZ16, PEZ2, PEZ20, PEZ21, PEZ3, PEZ5, PEZ6, PEZ8, PEZ8_2

Commercially available MMI from ABI

Halverson, J., Dvorak, J., Stevenson, T. 1995microsatellite sequences for canine genotyping US Patent 05874217

2.3 Results

2.3.1 Animal handling and operation

This section provides results for capture rates and results from morphometric

measurements compared with other dingo studies. Data from trapping programs over

three breeding seasons (2005, 2006 and 2007) and whelping and rearing seasons,

2007, are presented in section 2.3.1.1.


92

2.3.1.1 Capture rates

In total, 47 dingoes were captured (31 males, 16 females) with two males (2.5 and

2.28) recaptured in different trapping programs and one female was trapped twice in

three nights during April 2006. Ten males and one female were trapped in Site 1

whilst 21 males and 15 females were trapped in Site 2. Twelve foxes and one cat

were euthenased as a result of trapping programs. Seventeen non-target species

caught and released were common wombats, brushtail possums, lace monitors

Varanus varius, one Australian wood duck Chenonetta jubata, one Australian raven

Corvus coronoides and one European rabbit (Table 2.4). Casualties incurred as a

result of being trapped were one dingo and one wood duck. The wood duck was

consumed by an unknown predator in the trap. The dingo suffered from acute capture

myopathy.

Table 2.4: Capture data for the five trapping programs April 2005 April 2006 March 2007 June 2007 Sept. 2007 Total Trap nights 321 353 115 136 266 1191 Dingoes 12 20 10 5 3 50 Foxes 6 4 1 1 0 12 Cats 0 0 1 0 0 1 Non-target species 2 3 2 4 6 17 Trap set-off 27 33 14 6 24 104

2.3.1.2 Morphometric measurements

Morphometric measurements from dingoes in this study are shown in Table 2.5.

Compared with previous studies (Table 2.6), dingoes of the SGBMWHA had longer

heads, smaller ears, smaller shoulder heights, smaller hind foot lengths, longer tail and

total lengths and similar weight. More detailed analysis was not possible because raw

data for the populations was not available. Results indicated there were no major

differences in morphometric measurements between isolated wild dingoes of the

south east of Australia and free ranging dingoes from other regions across the

continent.

Table 2.5: Morphometric data for dingoes of the SGBMWHA . All measurements are in mm, except weight (kg). Location Site 1 Site 2 Total Sex Male n Female n Male n Female n Male n Female n M+F n

Head length 232.2 ± 12.3 10 243 1 241.5 ± 9.7 21 224.7 ± 10.0 15 238.5 ± 11.3 31 225.8 ± 10.7 16 234.2 ± 12.6 47

Ear length 94.5 ± 9.0 4 - - 98.6 ± 6.3 17 96.2 ± 3.9 13 97.8 ± 6.8 21 96.2 ± 3.9 13 97.2 ± 5.9 34

Shoulder height 582.9 ± 24.6 10 567 1 585.9 ± 30.8 21 556.9 ± 22.9 15 584.9 ± 28.5 31 557.5 ± 22.3 16 575.6 ± 29.4 47

Hind foot width 45.4 ± 2.4 10 44.49 1 45.5 ± 3.5 21 42.5 ± 2.6 15 45.5 ± 3.1 31 42.6 ± 2.5 16 44.5 ± 3.2 47

Hind foot length - - - - 185.3 ± 9.9 11 180.5 ± 7.7 6 185.3 ± 9.9 11 180.5 ± 7.7 6 183.6 ± 9.2 17

Tail length 396.9 ± 105.7 10 346 1 433.4 ± 55.5 21 432.6 ± 53.7 14 421.6 ± 75.5 31 426.9 ± 56.4 15 423.3 ± 69.3 46

Total length 1317 ± 100.4 10 1270 1 1358.2 ± 81.7 21 1290.8 ± 81.5 15 1344.9 ± 88.7 31 1289.5 ± 78.9 16 1326.1 ± 88.7 47

Weight 16.1 ± 2.2 10 18 1 17.1 ± 2.0 21 14.2 ± 3.2 15 16.8 ± 2.1 31 14.4 ± 3.3 16 16 ± 2.8 47

No. of dingoes 10 1 21 15 31 16 47

Table 2.6: Morphometric data for comparison of dingoes from the SGBMWHA with dingo populations around Australia (Adapted from Corbett 2001). All measurements, except weight in kg, are in mm.

Location Victorian Highlands

(Corbett 2001) Eastern Highlands of Victoria (Jones 1990)

Central Australia (Corbett 2001)

Kakadu National Park (Corbett 2001) SGBMWHA

Sex Male Female Male Female Male Female Male Female Male Female

Head length 220.4 ± 5.0 214.9 ± 12.0 - - 220.8 ± 4.3 213.6 ± 2.9 226.0 ± 8.0 215.0 ± 4.7 238.5 ± 11.3 225.8 ± 10.7 Ear length* 105.8 ± 3.3 101.2 ± 3.2 102 ± 6 99 ± 6 103.9 ± 3.4 99.5 ± 3.0 101.9 ± 2.9 94.7 ± 4.2 97.8 ± 6.8 96.2 ± 3.9 Shoulder height 580.4 ± 25.9 562.9 ± 25.0 - - 587.4 ± 18.0 559.0 ± 13.0 592.9 ± 23.5 565.0 ± 14.1 584.9 ± 28.5 557.5 ± 22.3 Hind foot length * 191.8 ± 7.8 179.4 ± 6.2 187 ± 9 176 ± 8 188 ± 5.3 179.5 ± 4.6 198.8 ± 10.8 185.0 ± 6.6 185.3 ± 9.9 180.5 ± 7.7 Tail length 299.4 ± 59.1 289.2 ± 44.8 318 ± 25 300 ± 24 324.3 ± 13.5 311.8 ± 18.1 322.5 ± 24.8 312.9 ± 11.5 421.6 ± 75.5 426.9 ± 56.4 Total length 1244.7 ± 76.2 1218.8 ± 64.1 1291 ± 73 1227 ± 61 1207.0 ± 32.4 1168.5 ± 25.5 1288.9 ± 84.4 1271.1 ± 27.7 1344.9 ± 88.7 1289.5 ± 78.9 Weight 15.5 ± 2.0 14.7 ± 1.7 16.3 ± 2.4 14.0 ± 2.3 14.5 ± 1.5 12.4 ± 1.1 17.4 ± 1.9 15.2 ± 1.1 16.8 ± 2.1 14.4 ± 3.3 No. of dingoes 12 16 199 177 25 25 12 7 31 16

* Incorrect measurements for ear length and hind foot length for the SGBMWHA population are excluded so all measurements are comparable


94

Coat colours of captured dingoes in this study were black and tan (38.3%), sable

(31.9%), tan/ginger (23.4%) and patchy (6.4%) (Table 2.7). Straight black, straight

white and brindle colours were not recorded, however brindle may have been

observed on one occasion from field trials of passive camera traps (the stance of the

urinating dingo and the angle of the camera may have made sable colouration appear

brindle). Some dingoes with patchy colouration were observed but not captured. Of

the two “patchy” coloured dingoes captured, one was a combination of sable, tan and

white and the other was tan and white. Coat colours of dingoes of the SGBMWHA

were similar to dingoes described elsewhere and the higher proportion of black and

tan coloured dingoes is consistent with increased proportion of that colour in south

eastern Australia (Table 2.8).

Table 2.7: Percentage coat colours recorded from dingoes of the SGBMWHA Colour Site 1 (n = 11) Site 2 (n = 36) Total (n = 47) Black and tan 12.8 25.5 38.3 Sable 6.4 25.5 31.9 Tan/Ginger 4.3 19.1 23.4 Patchy 0.0 6.4 6.4

Table 2.8: Percentage coat colour for comparison of dingoes from the SGBMWHA with dingo populations around Australia in order of latitude (Adapted from Corbett 2001) Location Tan Black and tan Black White Sable Brindle Patchy n Kakadu 99 0 0 0 <1 0 <1 500 Central Australia 88 5 0 4 2 <1 <1 1320 Western Australia 72 15 0 7 0 3 4 256 SGBMWHA 23 38 0 0 31 0 6 47 Vic. highlands 43 26 3 <1 14 7 6 734

In addition to these data, 95.7% of the animals sampled had four white paws (4.3%

had two white paws), 97.9% had bushy tails (n = 46) and 72% had white, 13% had

black and 6.4% had red tail tips. Of captures, 74.5% were free of mange and 66% of

the captures were >2 years of age according to head length, weight, tooth eruption and


95

presence or absence of fleur-de-lis. Sex ratio from captured animals was 1.9: 1 male:

female. Male dingo 2.22 was cryptorchid.

2.3.2 Skull measurements, genetic “purity” and population structure

This section provides results from canonical scores (Newsome et al. 1980), “purity”

tests (Wilton et al. 1999; Wilton 2001) and population structure tests (Pritchard et al.

2000).

2.3.2.1 Skull measurements

Ten skulls were retrieved from deceased individuals and prepared according to the

technique specified in section 2.2.2.1. Measurements of each skull were compared

with dingo skull measurements from previous studies (Table 2.9; Jones 1990;

Woodall 1996; Corbett 2001) to assist with assessment of animal “purity”.

Table 2.9: Comparisons of average canonical measurements of dingoes from around Australia with dingoes from the SGBMWHA (± sd) (cf. section 1.1.2; Corbett 2001)

Corbett (2001) EHV (n = 338)

AUG (n = 41)

WQ (n = 15)

SEQ (n = 5)

FI (n = 21)

SGBMWHA (n = 10) Dingo Hybrid Dog

X 25.1 1 22.1 20.8 24.4 26.4 25.7 25.4 26.8 25.64 ± 2.13

X 60.3 2 60.1 62.8 63.1 62.0 60.7 66.7 63.0 59.03 ± 2.57

X 7.5 3 6.8 6.8 7.2 7.8 7.6 7.8 7.9 6.77 ± 0.33

X 9.5 4 9.4 9.8 10.1 9.7 10.0 10.7 10.4 9.41 ± 0.65

X 33.5 5 30.3 28.4 34.6 32.8 32.7 33.2 32.8 35.96 ± 1.97

X 11.6 6 10.7 10.2 12.4 12.6 12.5 12.4 12.1 12.08 ± 1.69

X 55.9 7 55.2 58.2 58.8 57.0 56.6 57.3 56.9 59.32 ± 2.40

X 54.6 8 49.8 50.5 53.0 55.6 54.9 56.9 55.6 54.27 ± 2.51

SL - - - - 196.6 193.4 183.6 193.4 196.7 ± 7.66

CA = Central Australia (Corbett 2001); EHV = Eastern Highlands of Victoria (Jones 1990); AUG

= Augathella; SEQ = South East Queensland; FI = Fraser Island (Woodall et al. 1996); X1 to X8

represent skull measurements for “purity” assessment (Newsome et al. 1980); SL = Skull length.


96

Canonical scores for “purity” assessment implied 80% of SGBMWHA dingoes were

“pure” (Table 2.10). Dingo “Yerranderie” also showed a “pure” (Score 2 = Likely

dingo) result from genetic tests (cf. section 2.3.2.2). The sable coat colour visually

implied a degree of hybridisation in this animal that was subsequently killed by a

private landholder.

Table 2.10: “ Purity ” of dingoes in the SGBMWHA according to skull scores (cf. section 1.1.2;

Corbett 2001)

Dingo “ Purity ” according to canonical analysis

Score “ Purity ”

Yerranderie 2.929 “Pure”

1.3 2.479 “Pure”

1.7 1.942 “Pure”

1.11 3.318 “Pure”

2.11 2.510 “Pure”

2.18 3.215 “Pure”

2.22 4.337 “Pure”

2.30 0.643 “Intermediate”

2.31 3.001 “Pure”

2.34 0.456 “Intermediate”

2.3.2.2 Genetic “purity”

All 47 captured individuals were tested for “purity” . In addition, dingo

“Yerranderie”, captured from a disused mining town half way between Sites 1 and 2,

was included in “purity” tests and returned the most “pure” results. Sample collection

and analysis was successful in all attempts except for dingo 1.3 where the ear biopsy

was lost upon release of this animal. DNA was instead extracted from hair follicles

from dingo 1.3 with no apparent difference to results. Relative likelihood scores for

“purity” are provided in Table 2.11. Discernable differences existed in samples

assigned scores from ten tests compared with scores assigned by 20 tests when

compared with the reference range for domestic dogs, captive dingoes and calculated

hybrid proportions.

Table 2.11: Results for examination of “purity” for all 50 samples collected from capture/release animals, one deceased individual and two recapture/release indivi duals.

ID P(dingo/dog) P(hyb/dog) P(dingo/hyb) P(dingo/3Q) Number

loci Level of “ purity ”

Average 3Q Doglike Rare

Prop doglike Score M13TT

Diagnostic for dog

1.1 8.13 13.57 -5.44 -5.50 19 < 50% dingo -0.29 6 0.26 0.16 6 1 3

1.2 9.45 13.39 -3.94 -4.09 18 < 65% dingo -0.23 7 0.36 0.19 5 0 1

1.3 3.58 4.21 -0.63 -0.58 8 < 75% dingo -0.07 1 0.31 0.06 4 1

1.4 11.48 16.17 -4.70 -4.81 21 < 65% dingo -0.23 8 0.36 0.19 5 0 1

1.5 7.91 14.50 -6.59 -6.38 21 < 50% dingo -0.30 7 0.40 0.17 6 2 5

1.6 8.24 12.56 -4.32 -3.68 21 < 65% dingo -0.18 7 0.38 0.17 5 1 2

1.7 17.58 15.92 1.66 -0.04 13 < 75% dingo 0.00 2 0.15 0.08 4 0 2

1.8 8.24 11.99 -3.75 -3.73 19 < 65% dingo -0.20 5 0.32 0.13 5 2 3

1.9 16.87 15.59 1.28 -0.33 11 < 75% dingo -0.03 1 0.23 0.05 4 0 1

1.10 11.34 15.15 -3.81 -4.22 16 < 50% dingo -0.26 6 0.28 0.19 6 0 3

1.11 6.71 12.10 -5.39 -4.91 15 < 50% dingo -0.33 7 0.40 0.23 6 1 4

Yerranderie 22.92 18.94 3.98 1.35 17 likely “pure” 0.08 0 0.21 0.00 2 0

2.1 17.64 18.45 -0.81 -2.16 17 < 65% dingo -0.13 4 0.26 0.12 5 0 2

2.2 23.93 22.38 1.54 -1.03 21 < 75% dingo -0.05 5 0.26 0.12 4 1 2

2.3 13.09 16.50 -3.41 -3.87 21 < 65% dingo -0.18 6 0.29 0.14 5 0 3

2.4 15.18 17.53 -2.35 -3.33 20 < 65% dingo -0.17 5 0.23 0.13 5 0 3

2.5 (2005) 16.76 18.59 -1.83 -2.79 19 < 65% dingo -0.15 4 0.29 0.11 5 0 3

2.5 (2006) 15.08 14.46 0.62 -0.91 13 < 75% dingo -0.07 3 0.19 0.12 4 0 2

2.6 15.72 17.96 -2.24 -3.20 20 < 65% dingo -0.16 6 0.28 0.15 5 1 2

2.7 10.47 15.14 -4.67 -4.94 17 < 50% dingo -0.29 6 0.32 0.18 6 0 3

2.8 15.68 18.09 -2.41 -3.30 14 < 65% dingo -0.24 5 0.32 0.18 5 0 2

2.9 11.63 13.24 -1.62 -2.39 10 < 65% dingo -0.24 3 0.20 0.15 5 0 1

2.10 3.96 12.59 -8.63 -7.98 18 < 50% dingo -0.44 10 0.31 0.28 6 0 6

2.11 6.03 11.22 -5.20 -4.43 14 < 50% dingo -0.32 6 0.46 0.21 6 1 3

2.12 16.04 17.05 -1.01 -2.44 15 < 65% dingo -0.16 4 0.23 0.13 5 0 4

2.13 19.20 19.79 -0.59 -2.26 17 < 65% dingo -0.13 4 0.26 0.12 5 0 4

2.14 8.38 12.48 -4.10 -4.49 14 < 50% dingo -0.32 7 0.25 0.25 6 0 6

2.15 15.76 17.07 -1.32 -2.54 16 < 65% dingo -0.16 4 0.25 0.13 5 0 2

2.16 15.21 15.46 -0.25 -1.26 14 < 75% dingo -0.09 2 0.29 0.07 4 0 1

2.17 19.21 18.41 0.80 -0.79 16 < 75% dingo -0.05 2 0.25 0.06 4 1 2

Table 2.11 (cont.): Results for examination of “ purity” for all 50 samples collected from capture/release animals, one deceased individual and two recapture/release individuals.

ID P(dingo/dog) P(hyb/dog) P(dingo/hyb) P(dingo/3Q) Number

loci Level of “ purity ”

Average 3Q Doglike Rare

Prop doglike Score M13TT

Diagnostic for dog

2.18 10.64 12.37 -1.73 -2.71 13 < 65% dingo -0.21 4 0.31 0.15 5 1 2

2.19 16.16 16.96 -0.81 -2.13 15 < 65% dingo -0.14 3 0.23 0.10 5 0 3

2.20 1.48 4.19 -2.71 -2.66 6 < 50% dingo -0.44 3 0.50 0.25 6 2

2.21 10.08 11.04 -0.96 -1.94 13 < 65% dingo -0.15 3 0.50 0.12 5 1

2.22 1.39 5.43 -4.05 -3.71 11 < 50% dingo -0.34 4 0.45 0.18 6 2

2.23 -2.84 2.52 -5.35 -4.16 12 < 50% dingo -0.35 5 0.46 0.21 6 3

2.24 3.67 7.35 -3.67 -3.46 13 < 50% dingo -0.27 4 0.38 0.15 6 2

2.25 -1.58 5.04 -6.62 -5.59 13 < 50% dingo -0.43 6 0.42 0.23 6 4

2.26 5.88 8.78 -2.90 -2.95 12 < 65% dingo -0.25 4 0.38 0.17 5 3

2.27 2.47 8.28 -5.80 -5.02 13 < 50% dingo -0.39 5 0.46 0.19 6 3

2.28 (2006) 1.20 5.27 -4.07 -3.52 9 < 50% dingo -0.39 4 0.50 0.22 6 2

2.28 (2007) 6.99 13.47 -6.48 -5.84 20 < 50% dingo -0.29 7 0.43 0.18 6 0 2

2.29 4.20 6.82 -2.62 -2.65 10 < 50% dingo -0.27 3 0.45 0.15 6 0

2.30 23.25 21.89 1.36 -0.59 19 < 75% dingo -0.03 2 0.21 0.05 4 0 2

2.31 10.29 11.48 -1.18 -1.83 15 < 65% dingo -0.12 2 0.43 0.07 5 1

2.32 2.32 11.28 -8.96 -7.63 19 < 50% dingo -0.40 8 0.47 0.21 6 0 4

2.33 15.10 16.16 -1.06 -2.38 17 < 65% dingo -0.14 4 0.32 0.12 5 1

2.34 19.74 19.68 0.06 -1.36 19 < 75% dingo -0.07 3 0.29 0.08 4 0 2

2.35 9.65 15.03 -5.39 -5.22 20 < 50% dingo -0.26 7 0.38 0.18 6 0 4

2.37 10.75 14.58 -3.83 -4.24 20 < 65% dingo -0.21 6 0.43 0.15 5 2 3


99

Dingo 2.5, one of the two individuals recaptured and tested for “purity” twice,

returned results that suggested a higher proportion of “pure” genes were prevalent

during 2006 when compared with 2005. Only 13 tests were run in 2006 (19 were run

in 2005) and this could be the reason for the observed difference. Dingo 2.28 in

contrast returned a “purity” score of 6 (less than 50% dingo) during 2006 and 2007,

although 9 tests were run during 2006 and 20 tests were run during 2007. These

examples provide an indication for accuracy of the test. Results from Table 2.11 are

graphed in Figure 2.5. Figure 2.5a compares sample data with proportion of doglike

alleles and proportion of rare alleles. Figure 2.5b compares sample data with

proportion of doglike alleles with average 3Q results, whilst Figure 2.5c compares

rare alleles with average 3Q results to provide the most reliable estimation for

“purity” . The light blue squares are genotypes of domestic dogs tested and the blue

diamonds are the reference dingo samples. Each cross in the graph represents one of

the dingoes from the SGBMWHA. Figure 2.6 shows three canid genetic groups

(domestic dog reference group; SGBMWHA samples; and captive bred dingo

reference group), none of which show extensive relatedness to one another because

they are distinct and isolated canid populations in themselves.


100

0.00

0.20

0.40

0.60

0.80

1.00

0.00 0.25 0.50 0.75 1.00

Prop. rare

Pro

p. d

oglik

e

0.00

0.20

0.40

0.60

0.80

1.00

-1.00 -0.75 -0.50 -0.25 0.00 0.25Average 3Q

Pro

p. d

oglik

e

0.00

0.20

0.40

0.60

0.80

1.00

-1.00 -0.75 -0.50 -0.25 0.00 0.25

Average 3Q

Pro

p. r

are

Figure 2.6: Results of genetic analyses for 47 capture/release, one deceased and two recapture/ release dingoes from 2004 – 2007. = Wilton et al. (1999) and Wilton (2001) reference domestic dog genes; = Wilton et al. (1999) and Wilton (2001) reference captive dingo genes; X = SGBMWHA dingoes; Prop. doglike = Alleles ten times more common in domestic dog samples compared with captive dingo samples; Prop. rare = Proportion alleles rarely found in dingoes; Average 3Q = Likelihood the sample is from a ¾ dingo population.

a.

b.

c.


101

One dingo, 2.1% of the population, was likely “pure” dingo, 16.7% contained less

than 25% domestic dog genes, 43.8% contained less than 35% domestic dog genes

and 37.5% contained less than 50% domestic dog genes (Table 2.12). These data

show that 100% of the population contained more genes related to captive dingoes

than domestic dogs.

Table 2.12: Percentage genetic “purity” of dingoes per site and in total in the SGBMWHA Level of genetic “ purity ” Site 1 n = 11 Site 2 n = 37 Total n = 48

“ Pure” dingo 0.0 0.0 0.0 Likely “pure” dingo 0.0 2.7 2.1

Likely hybrid 0.0 0.0 0.0 <75% dingo 27.3 13.5 16.7 <65% dingo 36.4 45.9 43.8 <50% dingo 36.4 37.8 37.5

No detectable dingo ancestry 0.0 0.0 0.0

The levels of introgression may be reflected by data on morphometrics and

colouration. Recorded colours of dingoes are similar around Australia and colours

that deviate from traditional ginger/tan, black and tan or white are considered a result

of introgression (Corbett 2001; Fleming et al. 2001). Table 2.13 presents colouration

data with “purity” estimates and morphometric measurements, and compares these

with expected dingo morphometric measurements specified in Table 1.2.


102

Table 2.13: Comparison of “purity” score with coat colour and morphometric measurements, of capture/release dingoes during 2004, 2005, 2006 and 2007. = measurement falls within range specified in section Table 1.2; X = Colour is not accepted for “pure” dingo; = measurement falls above range specified in Table 1.2; = measurement falls below range specified in Table 1.2; B+T = Black and tan colouration

Site.Dingo “ Purity ” Score

Colour <

Head length (mm)

Shoulder height (mm)

Total length (mm)

Weight (kg)

1.1 6 Sable X 235 597 1390 17 1.2 5 B+T 225 568 1145 14 1.3 4 Sable X 232 612 1326 18 1.4 5 B+T 243 567 1270 18 1.5 6 B+T 210 550 1264 13 1.6 5 B+T 235 560 1285 16 1.7 4 Sable X 235 548 1205 16 1.8 5 B+T 250 617 1490 20 1.9 4 Tan 250 585 1400 15 1.10 6 Tan 230 590 1365 18 1.11 6 B+T 220 600 1300 14 Yerranderie 2 Sable X 253 - - 1448 20 2.1 5 Sable X 213 514 1200 9.5 2.2 4 Tan 215 528 1159 10 2.3 5 Sable X 240 596 1411 15.5 2.4 5 Sable X 234 583 1275 14.5 2.5 4 Sable X 248 587 1311* - 18 2.6 5 Sable X 255 650 1420 17 2.7 6 B+T 215 542 1250 10.5 2.8 5 B+T 235 573 1330 16 2.9 5 B+T 230 562 1360 14 2.10 6 Tan 245 555 1465 17 2.11 6 Sable X 235 580 1400 17.5 2.12 5 Sable X 240 630 1540 18 2.13 5 Sable X 235 590 1340 18 2.14 6 Sable X 220 550 1230 14 2.15 5 B+T 220 556 1370 13 2.16 4 B+T 225 550 1300 14 2.17 4 Sable X 245 640 1460 20 2.18 5 Tan 235 610 1365 20 2.19 5 Sable X 210 530 1290 10 2.20 6 B&T 238 592 1340 18 2.21 5 Tan 250 565 1320 19.5 2.22 6 Sable X 255 602 1405 19 2.23 6 Patchy X 256 587 1357 15 2.24 6 Tan 227 572 1315 17.5 2.25 6 Patchy X 252 571 1450 16 2.26 5 Patchy X 237 570 1300 17.5 2.27 6 B&T 248 570 1350 16 2.28 6 B&T 246 570 1236 15 2.29 6 B&T 228 510 1215 12.5 2.30 4 B&T 220 540 1234 13 2.31 5 Sable X 236 595 1245 20.5 2.32 6 Sable X 230 565 1375 19.5 2.33 5 B&T 245 585 1423 17 2.34 4 B&T 235 575 1294 16 2.35 6 Sable X 227 572 1365 14.5 2.37 5 Tan 218 587 1185 17 <

* = Animals tail tip was damaged during capture. = See Table 2.2 for explanation of “purity” score.


103

Dingoes that matched the coat colour criteria scored as less “pure” than dingoes that

did not match the coat colour criteria. Sable dingoes scored between 2 and 6 and

presented a higher proportion of “purity” (5.3% of sable animals were “pure”; n = 1)

than tan and black and tan animals. Only 25% of tan dingoes scored 4 whilst 16.7%

of black and tan dingoes scored 4. Dingoes 1.4, 1.10, 1.11 and 2.7 that are within

dingo colour and morphometric measurements scored 5, 6, 6 and 6 respectively.

Table 2.14 relates “purity” with coat colours and Table 2.15 provides a cumulative

summary of dingoes which fall within or outside of colour, morphometric

measurements and “purity” criterions.

Table 2.14: Relationship of genetic “purity” scores with coat colour. “ Purity ” score 1 2 3 4 5 6 7 Total Tan n 0 0 0 2 3 3 0 8 Tan % 0 0 0 25 37.5 37.5 0 B+T n 0 0 0 3 8 7 0 18 B+T % 0 0 0 16.7 44.4 38.9 0 Sable n 0 1 0 4 8 6 0 19 Sable % 0 5.3 0 21.1 42.1 31.6 0 Patchy n 0 0 0 0 1 2 0 3 Patchy % 0 0 0 0 33.3 66.7 0

Table 2.15: Relationship of dingoes which do or do not match coat colour, morphometric and genetic “purity” criteria.

Dingoes which:

Site 1 (n=11)

Site 2 (n=37)

Total (n=48)

n % n % n % Match coat colour (Tan, B+T or white) 8 72.7 18 48.6 26 54.2 Do not match coat colour (Sable or patchy) 3 27.3 19 51.4 22 45.8 Match Morphometrics 4 36.4 3 8.1 7 14.6 Do not match morphometrics 7 63.6 34 91.9 41 85.4 Match morphometrics and coat colour 3 27.3 2 5.4 5 10.4 Do not match morphometrics and coat colour 8 72.7 35 94.5 43 89.6 Match morphometrics/ coat colour and are “pure” 0 0 0 0 0 0 Do not match morphometrics/ coat colour and are “pure” 0 0 1 2.7 1 2.1

Table 2.15 shows a lower proportion of Site 2 dingoes fall within dingo morphometric

range. In contrast to this, dingoes in Site 1 had a higher percentage of “purity” scores

believed to be valuable for preservation of “pure” dingo genes in wild populations

(Table 2.12). Although not as “valuable” as the dingo “Yerranderie” sampled


104

between Site 1 and Site 2, Site 1 dingoes fit morphometric criteria more often than

Site 2 (Table 2.15). .

2.3.2.3 Population structure

Microsatellite loci from 50 dingoes of the SGBMWHA, two dingoes from the

northern GBMWHA (NGBMWHA) and one dingo from a partially connected reserve

east of the SGBMWHA were compared. Two recaptured dingoes (dingo 2.5 and

dingo 2.28) from the SGBMWHA were re-analysed to test the technique making 55

the total number of samples tested for relatedness. Data from this test were consistent

and both recaptured individuals were removed from further analyses.

The Evanno method (Evanno et al. 2005) identified two genetic populations and the

posterior probability method identified eight (Figure 2.7). The calculation of

population genetic statistics (FIS, HE

and A) was based on two genetic populations

because the sample sizes of the eight populations were in some cases too small for

meaningful analysis.


105

a.

-600

-500

-400

-300

-200

-100

0

0 1 2 3 4 5 6 7 8 9 10In

pr-I

nprlo

v

K probability

b.

-3500

-3000

-2500

-2000

-1500

-1000

-500

0

1 2 3 4 5 6 7 8 9 10

K

Ln P

r(X

|K)

0

5

10

15

20

25

30

Δ K

Figure 2.7: Determining probability of K (a. K probably = 8) and identification of K = 2 (blue line/boxes) as the most probable number of genetic clusters using the least negative Ln Pr(X|K) or K = 8 (dashed red line) using the second order rate of change (ΔK). Values summarised from five runs of STRUCTURE and data in graph b. were formatted using methods employed by Crompton et al. (2008).

The FST value for the two genetic populations was FST = 0.1213 which was

significant at α = 0.05. Allelic richness ranged from K = 2.661 for genetic population

one (n = 13 dingoes) and K = 4.593 for genetic population two (n = 40 dingoes). FIS

values ranged from FIS = -0.151 for population one to FIS = 0.066 for population two.


106

Population one was not significantly inbred8 (P = 1) but was significantly outbred (P

= 0.0011). Alternatively population two was neither inbred (P = 0.0023) nor outbred

(P = 0.9989). Expected heterozygosity ranged from He = 0.46 in genetic population

one to He

= 0.66 in genetic population two. Despite being significantly outbred,

genetic population one had lower levels of expected heterozygosity (Wahlund effect)

and allelic richness than genetic population two. To assist interpretation of the

proceeding graphical representations of genetic population structure, Table 2.16

details which STRUCTURE number was assigned to each dingo and which population

the dingo descended from.

8 P sig. inbred alpha = 0.00114; P sig. outbred alpha = 0.00114


107

Table 2.16: Numbers assigned to dingoes by STRUCTURE V. 2.2. Site.Dingo Colour Sex “ Purity ” score STRUCTURE no. < Genetic population 1.1 Sable M 6 4 2 1.2 B+T M 5 6 2 1.3 Sable M 4 9 2 1.4 B+T F 5 3 2 1.5 B+T M 6 16 2 1.6 B+T M 5 12 2 1.7 Sable M 4 18 2 1.8 B+T M 5 21 2 1.9 Tan M 4 7 2 1.10 Tan M 6 20 2 1.11 B+T M 6 11 2 Yerranderie Sable M 2 31 2 2.1 Sable F 5 1 2 2.2 Tan F 4 5 2 2.3 Sable M 5 8 1 2.4 Sable M 5 2 2 2.5 Sable M 4 15 1 2.6 Sable M 5 14 2 2.7 B+T F 6 19 2 2.8 B+T F 5 17 2 2.9 B+T F 5 10 2 2.10 Tan F 6 22 1 2.11 Sable M 6 13 2 2.12 Sable M 5 23 2 2.13 Sable M 5 24 1 2.14 Sable M 6 25 1 2.15 B+T F 5 26 2 2.16 B+T F 4 27 2 2.17 Sable M 4 28 2 2.18 Tan M 5 29 2 2.19 Sable F 5 30 2 2.20 B&T M 6 36 2 2.21 Tan M 5 37 1 2.22 Sable M 6 38 1 2.23 Patchy M 6 39 1 2.24 Tan F 6 40 1 2.25 Patchy M 6 41 1 2.26 Patchy M 5 42 2 2.27 B&T M 6 43 2 2.28 B&T M 6 44 2 2.29 B&T M 6 45 2 2.30 B&T F 4 46 2 2.31 Sable F 5 47 1 2.32 Sable M 6 48 2 2.33 B&T M 5 49 2 2.34 B&T F 4 50 2 2.35 Sable F 6 52 1 2.37 Tan F 5 53 2 <

= See Table 2.2 for explanation of “purity” score; B+T = Black and tan

Figure 2.8 shows genetic population structure at K = 2. Mapping capture locations

according to genetic relatedness data in Figure 2.9 displays the general geographic

division of population one and population two, but also spatial overlap of genetic

groups (Figure 2.8).

Figure 2.8: Population one (red) and population two (green) identified using STRUCTURE at K = 2.


109

Figure 2.9: Representation of capture location and relatedness data for population one (red) and population two (green) within the SGBMWHA . Numbers and colours on the map are representations of Figure 2.8. Samples 32, 33, 34, 50 and 51 were collected outside the study site and are not included within this figure.

, 9, 6, 11, 16, 20, 12, 4, 3, 21, 7, 18 31, 2, 44, 42, 43, 30 27 13, 48, 17, 26, 45, 49 10 52, 29, 5, 38, 1, 19, 23, 37, 46 14, 47, 28 53, 39, 22, 24, 8, 15, 32, 25, 35, 36, 40, 41,


110

Population structure at K = 8 (Figure 2.10) shows how population two can be

segregated in to seven subpopulations. Figure 2.11 shows related individuals were

trapped in similar localities or during extraterritorial forays. On the contrary, 13.2%

of the animals showed ≤25% allele frequencies were representative to one group, with

the remaining 75% of their allele frequencies being from one, two or three additional

groups. This indicated 81.1% of the dingoes shared more than 75% of their alleles

with one subpopulation.

Figure 2.10: Representation of subpopulations from population one (orange) and population two (all other colours) using K = 8 in STRUCTURE. Colouration of animals from: the red group were black and tan and sable; the light green group were black and tan and sable; the blue group were tan and sable; the orange group were tan, sable and patchy; and the purple group were tan, black and tan and sable. Comparison of colouration with genetic relatedness indicated coat colour is an inherited trait per group or geographic location.


112

Figure 2.11: Representation of capture locations for subpopulations and relatedness data for population one (orange/grey) and population two (all other colours) within the SGBMWHA using K = 8. Numbers and colours on the map are representations of Figure 2.10. Samples 32, 33, 34, 50 and 51 were collected outside the study site and are not included within this figure.

9, 6, 11 16, 20. 12 4, 3, 21 7, 18 31, 2, 44, 42, 43, 30 27 13 48, 17, 26, 45, 49, 10 52, 29, 5, 38, 1, 19, 23, 37, 46, 14, 47, 28, 53, 39, 22, 24, 8, 15, 25, 35, 36, 40, 41,


113

Five individuals culled and sampled outside of the study sites were tested for

relatedness to the SGBMWHA samples. Samples 33 and 34 were collected from the

NGBMWHA and sample 50 was from a distant though connected reserve to the east

(Figure 2.12). Figure 2.12a shows how these five individuals were assigned to

populations at K = 2 (cf. Figure 2.8) and Figure 2.12b shows how these five

individuals were related, or unrelated, to groups identified using K = 8 (cf. Figure

2.10).


114

Figure 2.12: Genetic samples collected outside of the SGBMWHA study site. Data are relevant to Figures 2.8 and 2.9 (a.) and Figures 2.10 and 2.11 (b.). Yellow lines represent Site 1 (top) and Site 2 (bottom).

, 33, 34,

50, .51, ,32,

, 33, 34,

50, .51, ,32,

a.

b.


115

Genetic relatedness data for each individual tested for “purity” were applied to the

dingo “purity” graph (Figure 2.13) that best represented “purity” (cf. Figure 2.6c).

Objectives of this graph were to observe if related individuals clustered together and

to check whether some genetically related individuals showed higher “purity” values

than others. Related individuals did not cluster together and no genetically related

group of individuals was more “pure” than others. Methods of analysis can therefore

remain divided.

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

-1.00 -0.75 -0.50 -0.25 0.00 0.25

Average 3Q (Rare dingo alleles)

Rar

e do

g al

lele

s

Figure 2.13: Relationship between “purity” (coloured ‘♦’ represent ‘X’ marks in Figure 2.6c) and relatedness of SGBMWHA dingoes (♦, ♦, ♦, ♦, ♦, ♦ represent main colours from K = 8 relatedness Figure 2.10), with reference to the domestic dog reference group (), reference calculated hybrid mixture proportions () and captive dingo reference groups () from Wilton et al. (1999) and Wilton (2001). NB: Data from the SGBMWHA samples overlap in some instances.


116

2.4 Discussion

Comparative analyses of morphometric measurements, colouration, skull allometry

and genetic biotechnology testing of “purity” for dingoes captured in this study

showed that the methods available to ascribe “purity” to a population are conflicting.

Canonical analyses using the method described by Corbett (2001) showed 80% of

sampled skulls were from “pure” dingoes. However, morphometric measurements

and colouration data showed 89.6% of the population were outside dingo criteria

(Table 2.13). Further, genetic analyses of “purity” showed that one dingo, 2.1% of

the animals sampled, was genetically “pure” (score 2; likely pure), though outside

morphometric measurements and colour criteria. Alternatively, tests to genetically

identify population structure indicated all 47 dingoes could be assigned to one of eight

related groups. Groups were formed within discrete topographic areas and dispersal

by individuals was evident, as would be expected in a communal living

hypercarnivorous canid population Using all of the tests previously described to

characterise dingoes , each of the animals sampled in this study matched one or more

criteria for dingoes. As a result of analyses in this chapter, genetic population

structure and phenotype of dingoes in the SGBMWHA are described.

Being hypercarnivores, dingoes live communally to maintain total energy

requirements, and results presented are consistent with this supposition. Data

presented by Lehman et al. (1992) on wolves and Girman et al. (1997) on African

wild dogs are consistent with population structure results presented in the current

study (section 2.3.2.3). This interpretation is similar to the findings of Girman et al.

(1997) and Lehman et al. (1992), that a high level of relatedness is shared by most

individuals within the population. Genetic relatedness data consistently related to


117

capture location and could be interpreted as extraterritorial forays by some individuals

at the time of capture. These data could also be evidence of dispersal in the

SGBMWHA.

Observations of genetic pack structure in the SGBMWHA are also consistent with

data from behavioural studies of dingoes in northwest Western Australia (Thomson

1992d; Thomson et al. 1992a). Observations of groups of dingoes hunting eastern

grey kangaroos and wombats in the current study are also consistent with findings of

Thomson (1992d) and Thomson et al. (1992a). If genetic population structure was

not observed, dingoes of the SGBMWHA would merely be feral domestic dogs or

feral, wild-born primitive dogs from South East Asia in the Australian bush, as was

suggested by Boitani et al. (1995) in their study on feral dogs in central Italy.

Behavioural research on the social structure of the SGBMWHA genetic

subpopulations might validate these conclusions.

2.4.2 Population genetic structure of SGBMWHA d ingoes

Multilocus genotype data sampled from the population of dingoes within the

SGBMWHA assigned individuals to eight topographically distinct groups (c.f. Figure

4.6, chapter four), with some evidence of admixture between groups. Research on

population structure of captive-bred “pure” dingoes and wild dingoes has not been

published previously. Analyses of genetic relatedness data demonstrated that pack

structure can be maintained in scheduled dingo conservation habitats. This is

consistent with management objectives for dingoes in the SGBMWHA and also

consistent with amendments to the second wild dog control order 1998 (Anon. 2000).


118

Analyses of genetic structure firstly identified two groups consisting of highly related

individuals with minor overlap of genes. Population one was significantly outbred

which implies it may have started from very low population levels and slowly

recruited dispersing individuals over time. Admixture of genes from more than one

pack is extensive in some individuals in population two which split into six or seven

familial group clusters. Approximately 81.1% of sampled dingoes showed more than

75% microsatellite alleles were related to only one of seven or eight groups. This is

similar to research on domestic dog breeds where different breeds are genetically

distinct and individuals can be assigned to breeds on the basis of their genotype

(Parker et al. 2004). Similar to observations made by Parker et al. (2004), formation

of groups in K = 8 from K = 2 represent separate “adaptive radiations” of genetic

groups. Hence the reason for population structure, as defined by microsatellite tests,

to relate to capture location in most cases (Figures 2.9 and 2.11). The animals

captured outside their genetic range were usually captured within the neighbouring

genetic group, and the individuals that showed less relatedness to other dingoes

sampled could have immigrated into the study sites. No distinct clusters were

representative of capture locations and a bigger sampling effort encompassing the

whole area of the GBMWHA and adjacent protected areas would be beneficial to

understand genetic drift and dispersal patterns.

The two individuals from the NGBMWHA showed relatedness to the green group in

Figure 2.9 (K = 2), but in Figure 2.11 (K = 8) they were unrelated to all other dingoes.

Their presence in K = 2 suggests one of the founding populations (green group in

Figure 2.6) of dingoes for the current integrating population was radiating from the

centre of the SGBMWHA or colonised the SGBMWHA from the north. In addition,


119

non-relatedness of these two individuals to other genetic groups in K = 8 showed that

data on relatedness between individuals was consistent throughout the SGBMWHA.

From the samples collected outside of the study sites, animals 32 and 51 showed

relatedness to dingoes within Site 2. Sample 50, in contrast, showed relatedness to

admixed samples 6, 10, 20, 28, 35 and others, to a lesser extent, from the

SGBMWHA. The pink and yellow colouring in K = 8 also indicated that a solely

“pink” genetic group and a solely “yellow” genetic group existed but were not

sampled.

The current study had a limited sampling regime due to logistical constraints

associated with field work of this nature. Evanno et al. (2005) suggested that the

intensity of sampling regimes plays a role in the correct assignment of individuals to

groups. This appears to be true for this research. The minority genetic groups

detected (purple, pink and yellow groups within K = 8) did not display clear genetic

clusters when mapped according to capture location. These dingoes do display,

however, the prevalence of migration between groups that may infuse genetic

diversity into this population (Girman et al. 1997), minimise inbreeding and maximise

outbreeding as was shown by analyses using STRUCTURE. It would be impossible to

detect where red group numbers 13 and 19 in Site 2 (within K = 8) had travelled from

if genetic data of dingoes within Site 1 were not obtained. This is similar with pink

and yellow admixed dingoes 6, 20, 28, 35 and 50 within K = 8 too. These data

suggested that dingoes may have dispersed from the SGBMWHA south and east

through connected reserves. Alternatively dingoes dispersing from the protected areas

where sample 50 was collected could have entered the SGBMWHA and introgressed


120

with the existing population. Additional research on genetics of dingoes from

adjacent reserves or within the SGBMWHA is required to validate this information.

Observations of dispersal by male dingoes in the SGBMWHA were evident and are

discussed further in chapter four. Dingo 2.12 (dark blue group in K = 8) was filmed

hunting and feeding with dingoes 2.10, 2.24, 2.25 and 2.26 (orange group in K = 8).

This event was consistent with results indicating population one was significantly

outbred. Like research on dingoes by Thomson et al. (1992a) and wolves by Lehman

et al. (1992), alterations in group formation between K = 2 and K = 8 in this study

indicated that some groups may be derivatives from larger packs. Using new

techniques for intensive records for movement patterns (chapter four), the

combination of seasonal movement and dispersal data were consistent with these

results of genetic relatedness.

Use of STRUCTURE in the current study proved to be invaluable in assigning related

individuals to familial groups. This method of analysis9

9 Research exploring deficiencies in the ecological paradigm of “population” definitions (Waples and Gagiotti 2006), identification of genetic discontinuities in natural populations (Manel et al. 2007) and detection of population expansion and decline (Beaumont 1999) were reviewed but considered unnecessary for the objectives of this thesis.

, combined with other

biological measures, should be adopted in landscape scale wildlife research for

holistic management because it accounts for communal dynamics as well as

individual animal characteristics. Faubet et al. (2007) explored the use of

STRUCTURE and other Bayesian models for ecological, evolutionary and other

genetic based research questions. Their conclusion claimed Bayesian methods to be

very powerful and successful measures to infer population structure, which were

consistent with results presented in this study. Comparison of capture location and


121

assigned structure presented an opposite effect to that described by Pritchard et al.

(2000a) who stated that estimation of K was not biologically interesting. Estimation

of K in this study was biologically interesting and provided relevant information on

the dingo population being studied.

Microsatellite data to identify relatedness were used according to criteria specified by

Chambers and MacAvoy (2000) and Wayne et al. (2004). That is, microsatellites

were used to assess allelic expression from related individuals, not distant ancestral

populations. Perhaps research on population structure of captive-bred dingoes will

provide a good indication of relatedness amongst animals in captivity, or relationships

between morphometric measurements and colour. However, the relatedness of captive

bred dingoes and wild-born dingoes requires further research. Koskinen and

Predbacka (2000) discussed genetic differentiation and genetic affinities among five

breeds of Finnish domestic dogs. Large variation between breeds was suggested to be

a result of genetic isolation, random drift and strong artificial selection. Nesbitt et al.

(2000) showed that average 3Q for wild dingo populations in New England and Guy

Fawkes River National Parks in north eastern NSW did not cluster with domestic

dogs, but occasionally clustered with captive dingo populations tested by Wilton

(2001). Results of genetic “purity” testing in the current study showed SGBMWHA

dingoes consistently clustered between domestic dog and captive dingo average 3Q

scores. This may corroborate the hypothesis that captive-bred dingoes are a

genetically distinct group. However, results also showed levels of “purity” between

genetically related individuals varied from “likely pure”, to less than 50% dingo and

less than 75% dingo. The STRUCTURE data were interpreted in this study as a

standard measure because the program only compared like alleles found within groups


122

of the SGBMWHA dingo population. The “purity” data, however, compared

expected alleles found in captive dingo populations or domestic dog populations with

the SGBMWHA population to interpret a calculated hybrid proportion of admixture

which was inconsistent between related individuals. The “pure” canonical analyses,

variable morphometric measurements and colouration data showed that genetic

“purity” data were inconsistent between indicators of dingo “purity” in the

SGBMWHA population. Since related individuals did not show similar levels of

“purity” , it may be postulated that “purity” criteria require further review. Research

on gene silencing, artificial selection in captive dingo populations, genetic structure

within and between the captive bred dingo population and levels of inbreeding and/or

outbreeding may be good areas to commence such a study.

The most genetically “pure” dingo in the current study was sable, and outside of dingo

morphometric criterion though dingoes which did not match colour criteria were

never selected for “purity” tests (Wilton et al. 1999). This raises the question whether

dingo preservationists would accept a sable coloured dingo as “pure” if genetic tests

indicated that it was. For reliable analyses of genetic “purity” a comprehensive stud

book should be made publicly available for persons interested in breeding “pure”

dingoes. It is likely that dingo breeders currently select for behavioural and

phenotypic traits exhibited by their stock which will ultimately change genetic and

phenotypic structuring within the captive populations. The most “pure” dingoes

according to morphometric measurements and coat colour in the SGBMWHA were

assigned “purity” scores of 5, 6, 6 and 6 (Table 2.13). These data may be used to

postulate that the functionality of dingoes, phenotype and “purity” are separate

management issues.


123

Alternatively, Wilton et al. (1999) and Wilton (2001) provided exceptional criteria to

test for “purity” in captive dingo populations. Dingo preservation societies did not

have this ability prior to the studies by Wilton et al. (1999) and Wilton (2001) and

presumably, selected breeding criteria were based on colour, skull allometry or

perceptions of what a dingo should look and behave like in a captive population.

Captive dingoes can therefore be called “pure” by breed societies because the

phenotype is standardised and representative of the traditional anthropocentric image

of dingoes. Another alternative view is that captive-bred “pure” dingoes are

specifically that. Unlike all other studies on wild dingoes, captive-bred dingoes do

not have a naturally selected territory, do not hunt prey or perform any ecological

function. Captive dingoes have no opportunities for dispersal or to challenge

dominant animals for status within a pack unless breeders artificially create such

opportunities. Combining population structure data with movement data presented in

chapter four, the current study showed dispersal and communal living are natural

behaviours in the SGBMWHA dingo population.

Data presented in the current study demonstrated that allelic composition of

SGBMWHA dingoes did not coalesce strongly with allelic composition of captive-

bred dingoes. This does not necessarily mean that they were hybrids based on

calculations by Wilton et al. (1999). Results from analyses of genetic population

structure therefore provide a genetic basis for future analyses in the SGBMWHA.

Changes in allelic structure may occur over time. Without the genetic population

structure analyses in the current study, land managers would not have been able to

monitor such genetic changes if that is to become an objective for management.


124

Of the microsatellite loci tested only two were used for analyses of genetic results in

both datasets. Unlike “purity” tests which compared populations of domestic dogs

with dingoes, relatedness tests only compared DNA of animals from within the

population with each other. Interpretation of “purity” data could lead to the

conclusion that 97.9% of dingoes in the SGBMWHA were “hybrids”. It could also be

argued that “hybrids”, as defined by tests of genetic “purity”, require management for

conservation of remaining “pure” genes or to reintroduce dingoes with “pure” genes

into “hybrid” populations. Alternatively, interpretation of relatedness data indicated

that all dingoes in the SGBMWHA uphold an ecological structure, unlike “pure”

dingoes held in captivity, and the population should be maintained for the benefit of

the ecosystem. It is clear that options for genetically based dingo management are in

opposition to each other. Land managers need to consider the costs and benefits for

conservation of the SGBMWHA based on general management objectives referred to

in section 1.2.6.1, and data presented in proceeding chapters of the current study.

Chambers and MacAvoy (2000) suggested that microsatellite loci are very successful

at identifying genetic structure within and between species or between populations of

a single species. Data in the current study have shown that genetic sub-structuring

within a population of wild dingoes can be extensive within a relatively small area.

Genetic and phenotypic variation has also been observed in wild living fishes.

Research on Arctic charr Salvelinus alpinus, a salmonid fish species in a three lake

system with no barriers to movement in Scotland, showed genetic and phenotypic

sub-structuring in the population (Adams et al. 2006). A very high level of site

fidelity was the major conclusion for observed results (Adams et al. 2006). Similar

genetic sub-structuring was observed for polar bears Ursus maritimus (Crompton et


125

al. 2008) and Canadian Arctic Island Wolves (Carmichael et al. 2008). Since

European settlement in Australia, many dingo populations have had to maintain high

site fidelity due to habitat fragmentation and isolation (cf. section 1.1.2.1). Based on

evidence presented by Adams et al. (2006), genetic and phenotypic sub-structuring

between populations should be expected due to barriers of movement found between

protected areas. Direct latitudinal and longitudinal comparison of microsatellite loci

within and between dingo populations throughout Australia will be useful to

understand relatedness and variation between all dingo populations. As has been

stated, variation in colouration was observed between packs by Thomson (1992a) as

well as in the current study.

2.4.3 Skull allometry, coat colour and epigenetic effects

Wilton et al. (1999) and Wilton (2001) selected genetic samples based on physical

characteristics of dingoes but physical appearance was a potentially unreliable method

to infer “purity” in the current study. Sample discrimination such as this is an

example of observer bias that can occur in genetic studies that microsatellite loci

repeat arrays are subject to (Chambers and MacAvoy 2000). Table 2.8 in section

2.3.1.2 showed tan colouration decreased from the north (Kakadu) to the south of

Australia (Victoria) whilst black and tan colouration increased. Corbett (2001) related

this phenomenon to hybridisation rather than to geographic variation. Such biological

questions can now be answered by programs such as STRUCTURE. Nationwide

colour variations, for instance, could be adaptations to local environmental conditions.

Colour variations were shown between STRUCTURE groups in the current study

which was also based on territories (cf. chapter four). As stated previously, the range

of colouration in the SGBMWHA is the same as reported by Newsome and Corbett


126

(1985) and Jones (1990). In total in northern and central Australia, 4.2% of 388

dingoes were ginger with white spots and 0.6% had patchy ginger and white

colouration (Newsome and Corbett 1985). In south eastern Australia, 2.8% were

patchy ginger or ginger and white (Jones 1990). In the current study, 6% of 47

captures were patchy and all of the patchy coloured captures were from the same

genetically familial group. In all major studies that reported colouration of dingoes in

Australia, patchy was not an unusual occurrence and therefore should be reconsidered

in descriptions associated with “purity”. In addition to this, it can be hypothesised

from population structure data (cf. section 2.3.2.3) that limited breeding opportunities

for the post-European population of SGBMWHA dingoes, due to eradication

programs, may have resulted in coat colour variations. The most multicoloured

genetic group of dingoes (tan, sable and patchy) was population one, which was

significantly outbred. This is consistent with results reported by Trut (2001) and Trut

et al. (2004) when foxes were reared within an intense selection pressure

environment. If it only took six to eight generations for standard coat colour and skull

allometry of silver foxes to vary, it may only take eight generations for colour and

allometry of isolated dingo groups to vary. Results from the fox data showed

phenotypic variations within “pure” genetic stock. In respect to dingo “purity”,

variations in morphological phenotype may also arise under artificial and natural

selection conditions.

In the present study, dingoes that matched morphometric measurements and colour

criteria did not meet the criteria set for genetic “purity” estimates. Based on

appearance these animals were the most likely candidates for dingo conservation, yet

would probably be culled because they were less than 65% dingo. Coat colour and


127

total length for dingo “Yerranderie” were outside of traditional “dingo” criteria, as

was his head length, though his canonical score (2.93; Table 2.9) and genetic “purity”

score (2; Table 2.11) indicated he was the most “pure”. These examples confirm that

current measures for dingo “purity” remain imprecise, and further research is required

before conservation actions for preservation of wild dingoes are implemented.

Data presented in section 2.3.2 on skull measurements for dingoes from the

SGBMWHA indicated that 80% of these animals were adapted for predation

according to characteristics presented by Newsome et al. (1980). However, the two

dingoes that returned intermediate skull scores were juveniles, and Macintosh (1976)

showed that juveniles tended to have intermediate skull scores. Morphometric

measurements of SGBMWHA dingoes appeared to be similar to those of dingoes in

Victoria, Kakadu and Central Australia (c.f. Table 2.8). Observed differences in

colouration and skull allometry may be due to environmental influences but generally

appear to be genetic. Average measurements from Kakadu National Park (KNP) for

instance, suggested KNP dingoes were taller and heavier whilst SGBMWHA dingoes

are shorter, longer, and lighter. These slight variations could be attributable to

latitude, climate, diet and topography.

One of the male dingoes captured in the current study was cryptorchid.

Cryptorchidism is a congenital defect known to be inherited as an autosomal recessive

polygenic trait in domestic dogs (Yates et al. 2003). Male gonads form from stem

cells of the neural crest in the genital ridge (Yates et al. 2003) and pregnancy and

lactation are reported as mobilising pathways for EDC (Colborn et al. 1993). These

three biological features (stem cells, pregnancy and lactation) have been identified as


128

susceptible to epigenetic effects. As stated in the introduction to this chapter, there is

a dearth of research on epigenetic effects and developmental programming in the

family Canidae; yet such research would clearly be beneficial (Ruvinsky 2001).

This chapter identified an array of information from past studies that are consistent

with the hypothesis that the dingoes in the current study were communally living

hypercarnivorous canids. Effects of colonisation, such as hybridisation, on the

SGBMWHA dingoes and subsequent isolation for over a century are potentially

significant However, current molecular studies do not adequately address the

question of hybridisation in the wild due to use of reference material that may be

biased by sample design, similar to research on differences in skull allometry. Gilbert

(2002) demonstrated genetic and biological adaptations for survival (Mendelian or

epigenetic effects) in vertebrates and invertebrates. Such adaptations may account for

similar variations in dingo populations. Based on evidence presented in this chapter

dingoes are a primitive (breeds only once per year) communal living

hypercarnivorous canid.

The investigation of biological characteristics presented indicated that the definition

of a dingo should be redefined, as follows:

The dingo is defined as a primitive communal living hypercarnivorous

member of the family Canidae from Australia, with one annual breeding cycle,

and minor genotypic and phenotypic variations across their geographic range.

To address questions regarding the functional role of dingoes in the SGBMWHA,

analyses of diet are compared with annual patterns in activity in chapter three.

Chapter Three:

Interactions between sympatric predators and prey

Interactions between sympatric competitors and prey

130

Chapter Three: Interactions between

sympatric predators and prey

3.1 Introduction ..................................................................................... 131

3.2 Methods ............................................................................................ 132

3.2.1 Diet ........................................................................................................ 133

3.2.2 Activity and abundance ........................................................................ 135

3.2.3 Relationship between diet and activity ................................................ 137

3.3 Results .............................................................................................. 137

3.3.1 Diet ........................................................................................................ 137

3.3.2 Activity and abundance ........................................................................ 149

3.3.3 Relationship between diet and activity ................................................ 154

3.4 Discussion ......................................................................................... 164


131

3.1 Introduction

Newsome et al. (1983a) and Corbett and Newsome (1987) showed diet of dingoes

alternated with fluctuations in prey abundance. Corbett and Newsome (1987) also

showed prey switching was relative to changes in prey abundance and shifts in

weather patterns in central Australia.

Research on the diet of red foxes in Australia is similar to research on dingo diet (c.f.

chapter one) in that scats or stomach contents were analysed to assess food items

(Brunner et al. 1976; Friend 1978; Green and Osborne 1981; Lunney et al. 1990;

Meek and Triggs 1998; Green 2002; Paltridge 2002; Mitchell and Banks 2005).

Green and Osborne (1981) and Green (2002) reported diet in relation to prey

abundance and selective predation of broad toothed rats Mastocomys fuscus by red

foxes in the Snowy Mountains in south east Australia. Diet of red foxes differs from

dingoes due to: a) higher consumption rates of small mammals within the critical

weight range (CWR; 500g – 1500g) of Australian fauna; and b) greater diversity in

species consumed (Burbidge and Mackenzie 1989; May and Norton 1996).

Studies on dingo ecology have begun to focus on the role of the dingo as a higher

order predator (Daniels and Corbett 2003; Glen and Dickman 2005; Glen et al. 2007;

Johnson et al. 2007). These authors suggested that dingoes may aid the conservation

of CWR fauna by suppressing the abundance and activity of mesocarnivores (cats and

foxes) which opportunistically prey on species in the CWR. Mitchell and Banks

(2005) studied dietary and spatial interactions of wild dogs and foxes in the

GBMWHA and found a high degree of overlap in niche breadth and resource


132

partitioning. Negative associations between the competitors suggested that temporal

avoidance was plausible.

Chapter three describes the functional ecology of dingoes in the SGBMWHA between

2005 and 2007. Analyses of monthly and seasonal variations in diet and activity of

dingoes are assessed for consistent patterns over time, related to biological traits such

as breeding, whelping and rearing. Comparative tests of predator/predator and

predator/prey interactions showed dingoes may influence movements and foraging

effort of competitors and prey. Alternation in predation theory (Corbett and

Newsome 1987) is addressed and discussed in relation to the techniques used to

sample the diet of the dingoes in that study. The aims of the current study were to

determine the extent of interaction between predators and their prey and predators and

their competitors over time using techniques with minimum invasiveness to identify

the role of dingoes as predators in the SGBMWHA.

3.2 Methods

Field work was conducted for seven or eight days (3-4 days in each of Sites 1 and 2)

every month for a total of 26 months, commencing in March 2005 and finishing in

April 2007. Two weeks was the minimum time permitted between field trips pending

trips being postponed due to inclement weather or trapping programs (cf. chapter

one), collar retrieval programs (cf. chapter four) or alternative commitments

elsewhere. A minimum of two people was involved on each field trip.


133

3.2.1 Diet

Scats were regularly collected along roads and roadsides by researchers, or

opportunistically on rare occasions from less frequented locations off transects.

Whilst monitoring animal activity on sand plots (cf. section 3.2.2 below), each of the

25km transects were searched from the vehicle over four to six hours per day for two

days (half transect per day) or over three days when scats were abundant. The vehicle

was driven at walking pace (≤5km/h) whilst the driver searched the right hand-side of

the road and road side and the second person, sitting on the bonnet or walking in front

of the car, concentrated search efforts to the left hand side and centre of the road,

scanning the whole road area whenever feasible. Assistants were taught to focus on

or under small bushes and prominent rocks and were advised that samples may also

be found flat on the road or roadside.

Clear snap lock freezer bags turned inside-out were used as a glove to collect all scats.

Bags were stapled shut for sample security and safety from infection from zoonoses

such as hydatids. Locations were recorded using the GARMIN GPS 72 waypoint

system. Individual waypoints were written using black permanent markers on the bag

of each sample with the date of collection, predator type (dingo or fox) and transect

(Site 1 or Site 2).

Scats were classified as dingo or fox in the field and corrected when necessary during

analyses based on the presence of dingo or fox grooming hairs. Identifying features

for dingo scats included size, shape, odour and sometimes the presence of large bone

material. Fox faeces were identified using similar means but also included the

presence of tapered ends which dingo scats do not have. Odour of the scat was


134

incorporated where possible and most scats were assigned as either fox or dingo at the

time of collection. Size and shape were key determinants of species because scats

were often old and odourless at the time of collection. Scats were frozen stored prior

to being delivered or sent to an independant analyst (Barbara Triggs, Dead Finish,

Victoria Australia) for itemisation of dietary items using techniques specified by

Brunner et al. (2002). Major prey items were identified to species level where

possible. Occurrence of other matter such as vegetation, invertebrates, reptiles, birds

and unknown material were recorded as a percentage of the total scat sample. Dingo

and fox scats were separated according to site.

Interactions were analysed using Pearson’s correlation, univariate ANOVA and

ARSIN in SPSS. The assumption of normal distribution was checked using P-P plot

and homogeneity of variance using Levene’s test of equality of error variances.

Differences between means were tested using Ryan’s Q test for assumed equal

variances. Factor analysis was used to observe the pattern of correlations between 13

prey species categories. Prey items were assigned to one of twelve categories to test

significance for importance of prey items in the diet of the predators, seasonal

variations in prey species consumed, and differences between prey consumed by

dingoes and foxes. Prey items were grouped as:

1. Swamp wallaby;

2. Macropod species;

3. Medium ground dwelling

native species;

4. Medium ground dwelling

exotic species;

5. Brushtail possum;

6. Other arboreal species;

7. Small ground dwelling

mammals;

8. Other

9. Vegetative material;

10. Avifauna;

11. Herpetofauna; and

12. Insect


135

To graphically present changes in prey items consumed over time, total scats with

prey item y during month m were divided by total scats collected (t) during month m.

This equation has been coined the Predator Diet Index (PDI) and can be represented

as:

PDI = y (during m) / t (during m)

This formula standardised the diet data so they were comparable with the Passive

Activity Index (PAI) used to determine changes in animal activity (Allen et al. 1996).

3.2.2 Activity and abundance

Twelve tonnes of screened sand were stored on site and distributed on transects as

sand plots three to four centimetres deep and 1-1.5 metre wide strips from gutter to

gutter across fire trails in areas less susceptible to runoff and erosion. These

dimensions were necessary to record gait and track pattern (walking, trotting,

hopping, running or bounding) when identification of individual tracks became

difficult (Catling et al. 2000; Catling et al. 2005). Individual sand plots were placed

approximately every kilometre for 25km on both transects. Imported sand was

required because local landform was generally barren, hard and rocky and inadequate

for recording animal prints successfully. Sand plots were replenished when necessary

after rain events, road works by land management authorities or general

compaction/sand loss over time.

Sand plots were raked so ridges were created in the sand when the sites were entered

and every day after tracks had been recorded. The PAI (Allen et al. 1996) required

numbers of tracks across a plot to be counted. Numbers of tracks on each sand plot


136

were counted by comparing entry and exit points of spoor. If, for instance, dingo

footprints were over-abundant on one plot, the tracks in the centre of the plot were

ignored and the points where spoor entered and departed the plot were counted. If the

sum of the prints entering and departing the plot was odd (more tracks departing the

plot than entering the plot or vice versa), the higher number was used for the analyses

because the PAI is a measure of activity opposed to a population census. Total tracks

counted were divided by the total number of plot nights to determine the PAI. The

PAI can be represented as PAI = t / n where t is the total number of tracks counted per

plot per species and n is the total number of plot nights.

Abundance indices were calculated using methods specified by Catling et al. (2005).

The Abundance Rating Formula (ARF) requires presence or absence of spoor of each

species on a sand plot divided by the total number of plot nights. The ARF can be

represented as ARF = p / n x 100 where p is the total number of plots with spoor of

each species and n is the total number of plot nights. Activity and relative abundance

indices were determined for all large and medium ground dwelling vertebrates.

Difficulties were encountered when attempting to distinguish between spoor of red-

necked wallaby, wallaroo Macropus robustus and swamp wallaby. Differences

between tracks of these species have been noted (Triggs 1996), however, wallabies in

this study were grouped (Catling et al. 2005). Tracks of these species were therefore

pooled as “wallaby” for activity and relative abundance indices.


137

3.2.3 Relationship between diet and activity

To compare seasonal movements with seasonal shifts in prey consumption, indices

calculated for the PDI and the PAI were graphed simultaneously to observe trends

over time. Correlations were performed in SPSS between raw track counts, PAI

indices and ARF indices of dingoes, foxes and cats to determine the effects of activity

or relative abundance of one predator on another. Raw track counts, PAI indices,

ARF indices and total scats collected for dingoes and foxes were correlated separately

with raw track counts, PAI indices and ARF indices for “wallaby” species, eastern

grey kangaroos, brushtail possums, European rabbits and common wombats because

they were among the most common dietary items of the dingoes. Also included in the

matrix were data of the raw occurrence of each prey species as a dietary item in total

scats collected (per predator) and the PDI indices (per predator).

3.3 Results

3.3.1 Diet

A total of 2545 predator scats were removed from both transects over 26 months

(Figure 3.1). Combining site totals, 1576 were dingo (Site 1: 862; Site 2: 714) and

969 were fox scats (Site 1: 695; Site 2: 274). Due to the exclusion of 94 fresh faecal

samples from dingo and/or fox diet analyses (frozen, but not used, for genetic tests),

details for 833 dingo and 692 fox samples from Site 1 and 656 dingo and 270 fox

faeces from Site 2 are included in the analyses (n = 2451). Scat samples without hairs

(7.5% total; 2.7% dingo; 15.1% fox) contained proportions of vegetative, bird, reptile

or insect material whilst 9% (13.6% of dingo; 1.8% of fox) of all analysed samples

contained predominantly bone fragments. From 27 identified mammalian species,


138

dingo samples contained 20 species and fox samples contained 22 species (Table 3.1).

Seven identified mammalian prey were introduced species.

a. b.

0

20

40

60

80

100

120

M2005

A M J J A S O N D J2006

F M A M J J A S O N D J2007

F*MA^

0

20

40

60

80

100

120

M2005

A M J J A S O N D J2006

F M A M J J A S O N D J2007

F*MA^

Figure 3.1: Changes in number of scats collected per month for 26 months at Site 1 (a.) and Site 2 (b.) for dingoes (orange columns) and foxes (blue columns). * Field trips were cancelled due to wet weather in February 2007. ̂ Scats were only collected on one day in Site 2 in April 2007.

Red-necked wallaby was only recorded on two occasions in fox scats and wallaroo

was never recorded as an item of diet. As a result, PAI and ARF indices include data

from the three “wallaby’’ species but swamp wallaby is the only wallaby used in

dietary analyses so caution is required when interpreting correlation analyses.

Table 3.1: Total count (n) and percentage (%) of total prey groups and items observed in dingo and fox scats over 26-months during 2005, 2006 and 2007 from the SGBMWHA . ̂ = Prey group is the grouping for proceeding statistical tests; * = Introduced species; <

Prey groups^

= Native or introduced species Prey items Dingo Fox Total

Common name Genus/species na me Site 1 Site 2 Site 1 Site 2 N % N % N % N % N %

Swamp wallaby Swamp wallaby Wallabia bicolor 457 43.2 278 35.1 215 20.0 36 8.5 986 29.5

Macropus sp. Eastern grey kangaroo Macropus giganteus 16 1.5 100 12.6 17 1.6 19 4.5 152 4.5

Red-necked wallaby Macropus rufrogriseus 0 0.0 0 0.0 2 0.2 0 0.0 2 0.1

Medium ground dwelling native species Common wombat Vombatus ursinus 30 2.8 28 3.5 16 1.5 2 0.5 76 2.3

Short-beaked echidna Tachyglossus aculeatus 4 0.4 1 0.1 0 0.0 1 0.2 6 0.2

Medium ground dwelling exotic species European rabbit* Oryctolagus cuniculus 42 4.0 20 2.5 47 4.4 17 4.0 126 3.8

Brushtail possum Brushtail possum Trichosurus vulpecula 146 13.8 62 7.8 211 19.7 39 9.2 458 13.7

Other arboreal species Common ringtail possum Pseudocheirus peregrinus 12 1.1 5 0.6 28 2.6 4 0.9 49 1.5

Sugar Glider Petaurus breviceps 17 1.6 1 0.1 22 2.1 2 0.5 42 1.3

Greater Glider Petauroides volans 11 1.0 1 0.1 10 0.9 0 0.0 22 0.7

Yellow bellied glider Petaurus australis 1 0.1 0 0.0 0 0.0 0 0.0 1 0.0

Feathertail glider Acrobates pygmaeus 0 0.0 0 0.0 1 0.1 1 0.2 2 0.1

Eastern pygmy possum Cercartetus nanus 0 0.0 0 0.0 2 0.2 0 0.0 2 0.1

Small ground dwelling mammals House mouse* Mus musculus 4 0.4 15 1.9 10 0.9 61 14.3 90 2.7

Brown antechinus Antechinus stuartii 0 0.0 0 0.0 13 1.2 2 0.5 15 0.4

Dusky antechinus Antechinus swainsonii 0 0.0 0 0.0 3 0.3 0 0.0 3 0.1

Common dunnart Sminthopsis murina 3 0.3 0 0.0 14 1.3 3 0.7 20 0.6

Bush Rat Rattus fuscipes 1 0.1 2 0.3 5 0.5 3 0.7 11 0.3

Black rat* Rattus rattus 2 0.2 3 0.4 3 0.3 3 0.7 11 0.3

Rat species Rattus sp. < 1 0.1 1 0.1 2 0.2 1 0.2 5 0.1

Other Microbat Genus & species unknown 1 0.1 0 0.0 0 0.0 0 0.0 1 0.0

Grey headed flying fox Pteropus poliocephalus 0 0.0 0 0.0 1 0.1 0 0.0 1 0.0

Cattle* Bos taurus 0 0.0 2 0.3 0 0.0 0 0.0 2 0.1

Sheep* Ovis aries 0 0.0 1 0.1 0 0.0 0 0.0 1 0.0

Pig* Sus scrofa 2 0.2 0 0.0 0 0.0 0 0.0 2 0.1

Dingo/dog< Canis lupus sp. (prey) 3 0.3 8 1.0 1 0.1 4 0.9 16 0.5

Platypus Ornithorhynchus anatinus 0 0.0 0 0.0 1 0.1 0 0.0 1 0.0

Undetermined species - 5 0.9 5 2.0 12 1.1 7 1.6 29 0.9

Vegetative material - - 137 12.9 177 22.3 129 12.0 61 14.3 504 15.1

Avifauna - - 41 3.9 25 3.2 102 9.5 39 9.2 207 6.2

Herpetofauna - - 12 1.1 2 0.3 38 3.5 8 1.9 60 1.8

Insect - - 105 9.9 55 6.9 169 15.8 113 26.5 442 13.2

Total items 1053 792 1073 426 3344

Total samples analysed 833 656 692 270 2451


140

Most commonly eaten prey identified in dingo faeces was swamp wallaby in both

sites (Site 1: 43.2% Site 2: 35.1%). Second most common prey from Site 1 samples

was brushtail possum (Site 1: 13.8%; Site 2: 7.8%) and Site 2 was eastern grey

kangaroo (Site 1: 1.5%; Site 2: 12.6%), however, vegetative material was a higher

proportion in Site 2 (Site 1: 12.9%; Site 2: 22.3%). Other mammalian contributions

to the diet included European rabbit (Site 1: 4%; Site 2: 2.5%) and common wombat

(Site 1: 2.8%; Site 2: 3.5%). Presence of avifauna (Site 1: 3.9%; Site 2: 3.2%) in

faecal material was exclusive in only 2.4% (n = 41) from Site 1 and 8% (n = 25) of

samples containing avifauna from Site 2, while 95.5% (n = 66) of samples containing

avifauna also had other items identified in the scats. Insect presence (Site 1: 9.9%;

Site 2: 6.9%) was similar, with 95.6% (n = 160) occurring with other prey items.

Fox samples indicated most commonly eaten prey were also swamp wallabies (Site 1:

20.0% Site 2: 8.5%) and brushtail possums (Site 1: 19.7%; Site 2: 9.2%) though the

second highest proportion of prey for foxes in Site 2 was house mouse (Site 1: 0.9%;

Site 2: 14.3%). Occurrence of eastern grey kangaroo (Site 1: 1.6%; Site 2: 4.5%) and

European rabbit (Site 1: 4.4%; Site 2: 4.0%) was relatively common. Vegetative

materials were consistent (Site 1: 12.0%; Site 2: 14.3%) whilst avifauna (Site 1: 9.5%;

Site 2: 9.2%) and insect (Site 1: 15.8%; Site 2: 26.5%) samples were high when

compared with occurrence of mammalian prey. Avifauna presence was exclusive in

16.3% (n = 141) of samples containing avifauna and 18.8% of all fox samples (n =

1499) contained insect. Fox was never observed as a prey item within fox or dingo

scat samples, however, Canis sp. hair was observed in fox (n = 5) and dingo (n = 11)

faeces as an item of diet. Observations of Canis sp. hairs in a scat were classified as

grooming hairs (few) or consumption/predation (many) based on quantity.


141

European rabbit (3.8%) and house mouse (2.7%) were the highest contributing

introduced species in diet. Low occurrence of pig, cattle and one sample with sheep

indicated that they are not regular prey species in this study area. Figure 3.1 shows

proportions of prey groups (cf. Table 3.1). Swamp wallaby was staple prey for

dingoes in both sites and there was variation between consumption of brushtail

possum in Site 1 and Macropod species in Site 2 by dingoes. Compared with dingoes,

foxes appeared less selective of prey, with more variation and opportunism.

a. c.

b. d.

Figure 3.2: Proportion of prey for dingoes in Site 1 (a.) and Site 2 (b.) and foxes in Site 1 (c.) and Site 2 (d.) of the SGBMWHA ( Swamp wallaby; Macropod species; Medium ground dwelling native species; European rabbit; Brushtail possum; Other arboreal species; Small ground dwelling mammals; Other/undetermined species; Vegetative material; Avifauna; Reptile and Insect). Refer to Table 3.1 for percentage values.


142

Figures 3.3a-d show periodical or seasonal variation in prey consumed by dingoes and

foxes. There were significant differences in the number of scats with swamp wallaby

remains for dingo and fox in breeding seasons of 2005, 2006 and 2007 (F1, 1 =

1265.57, P = 0.018, F1, 1 = 573.39, P = 0.027, F1, 1 = 10281.94, P = 0.006 for 2005,

2006 and 2007 season respectively). There were no significant differences in the

number of scats with swamp wallaby remains for dingo and fox in whelping seasons

of 2005 and 2006 (F1, 1 = 12.82, P = 0.173, F1, 1 = 24.32, P = 0.127 for 2005 and 2006

seasons respectively). There were significant differences in the number of scats with

swamp wallaby remains for dingo and fox during rearing season in 2006 (F1, 1 =

506.209, P = 0.028) but there were no significant differences in 2005 (F1, 1 = 4.479, P

= 0.281) or in exploratory season 2005 and 2006 (F1, 1 = 63.277, P = 0.080, F1, 1

a. c.

=

1.882, P =0.401).

0%

20%

40%

60%

80%

100%

Per

cent

vol

ume

0%

20%

40%

60%

80%

100%

Per

cent

vol

ume

b. d.

0%

20%

40%

60%

80%

100%

B05W05R05 E06 B06W06R06 E07 B07Season

Per

cent

vol

ume

0%

20%

40%

60%

80%

100%

B05 W05R05 E06 B06W06R06 E07 B07Season

Per

cent

vol

ume

Figure 3.3: Changes in the proportion of prey consumed by dingoes during breeding (B), whelping (W), rearing (R) and exploratory (E) biological seasons in Site 1 (a.) and Site 2 (b.) and foxes in Site 1 (c.) and Site 2 (d.) of the SGBMWHA during 2005 (05), 2006 (06) and 2007 (07) ( Swamp wallaby; Macropod species; Medium ground dwelling native species; European rabbit; Brushtail possum; Other arboreal species; Small ground dwelling mammals; Other/undetermined species; Vegetative material; Avifauna; Reptile and Insect)


143

Of the remaining prey categories, there were significant differences in the number of

scats with vegetative material in the 2005 exploratory season (F1, 1 = 350.078, P =

0.034), with insects during the 2006 exploratory season (F1, 1 = 206.087, P = 0.044)

and with other arboreal species during breeding 2006 (F1, 1

= 136.277, P = 0.054).

There were no significant differences in diet between dingoes and foxes during any

other season for Macropus sp., medium ground dwelling native species, European

rabbit, brushtail possum, other arboreal species, small ground dwelling mammals,

other/undetermined species, vegetative material, avifauna, reptile and insect.

There were significant differences in elements of dingo diet between years and also

between biological seasons (Breeding 2005, 2006 and 2007 F11, 11 = 8.973, P = 0.001,

F11, 11 = 15.083, P = 0.000, F11, 11 = 6.084, P = 0.003; whelping 2005 and 2006 F11, 11

= 7.187, P = 0.001, F11, 11 = 5.727, P = 0.004; rearing 2005 and 2006 F11, 11 = 5.271, P

= 0.005, F11, 11 = 8.278, P = 0.001; and exploratory 2005 and 2006 F11, 11 = 6.486, P =

0.002, F11, 11 = 8.016, P = 0.001 respectively). Ryan’s Q test showed that the

proportion of wallaby in dingo scats was consistently significantly higher than most

other prey categories (Figure 3.4a-d). Insect and vegetation during breeding 2005

(Figure 3.4a) and vegetation in whelping seasons 2005 and 2006 (Figure 3.4b) were as

significant as swamp wallaby as dietary items in these seasons. Ryan’s Q test

indicated that dingo diet diversified during rearing seasons (Figure 3.4c) for 2005

when all prey categories except reptile significantly contributed to diet and 2006 when

all prey categories except small ground dwelling mammals significantly contributed

to diet. Swamp wallaby was the major dietary item during both exploratory seasons

(Figure 3.4d) and all other prey categories occurred as items of diet much less

frequently compared with other biological seasons.


144

0.00

0.20

0.40

0.60

0.80O

ccur

renc

e of

pre

y ca

tego

ries

a

ab ab

x

yzyz

y

c bcbcbcbc

bcbcbcbc

yz

yzyz

zz

zzz

a.

0.00

0.20

0.40

0.60

0.80

Occ

urre

nce

of p

rey

cate

gorie

s

a

ab

bc

bc bc bcbc

bc

bc

bc

cc

x

xy

yy

yy

y

yy

yy

y

b.

0.00

0.20

0.40

0.60

0.80

Occ

urre

nce

of p

rey

cate

gorie

s

a

bcc

aab

abc abcabc

abcabc abc

abc

x

yz xyz

yzyz yz

yzyz yz

yzyz

z

c.

0.00

0.20

0.40

0.60

0.80

1 2 3 4 5 6 7 8 9 10 11 12

Prey category

Occ

urre

nce

of p

rey

cate

gorie

s

a

b

b

b

bbbbb

bb y

yyy

yyyy

yyy

x

b

d.

Figure 3.4: Significance of dietary items for dingoes during breeding (a.), whelping (b.), rearing (c.) and exploratory (d.) biological seasons for 2005 (■; a, b and c) and 2006 (■; x, y and z). Occurrence of the same letter in relevant columns indicates lack of significant difference.


145

There were significant differences in elements of fox diet during breeding seasons

2005 (F11, 11 = 4.084, P = 0.014), 2006 (F11, 11 = 3.487, P = 0.025) and 2007 (F11, 11 =

6.272, P = 0.003). There were no significant differences in fox diet during whelping

2005 (F11, 11 = 2.146, P = 0.111) though there were significant differences in diet

during whelping 2006 (F11, 11 = 3.189, P = 0.033). Although data are similar, no

significant differences in diet were evident for foxes during rearing season 2005 or

2006 (F11, 11 = 2.416, P = 0.080, F11, 11 = 2.398, P = 0.081). Differences in fox diet

for exploratory seasons 2005 and 2006 were highly significant (F11, 11 = 13.159, P =

0.000, F11, 11

= 4.280, P = 0.012 respectively) and Ryan’s Q test indicated that diet

was dominated by insects (Figure 3.5d) in both years. Ryan’s Q test showed that

proportion of insect in diet was also significantly higher during breeding 2005 and

2006 (Figure 3.5a). In general, Figure 3.5a-d shows the opportunistic nature of foxes

as a mesocarnivore when compared with dingoes (Figures 3.4a-d) and the selective

predation by dingoes as a hypercarnivore.


146

0.00

0.20

0.40

0.60

0.80

1.00

1.20

Occ

urre

nce

of p

rey

cate

gory

a

ab

ab

ab

ab ab ab

bb

b

b

x

xy

xyxyxy xy

xy

a.

ab

xyxy

xy

yy

0.00

0.20

0.40

0.60

0.80

1.00

1.20

Occ

urre

nce

of p

rey

cate

gory

a

a

a

aa

a

a

a

aa

a

a x

x

xx

x

x

x

x

xxx

x

b.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

Occ

urre

nce

of p

rey

cate

gory

aaa

a

a

a

aa

x

xx

x

x

x

x x

c.

a

a

a

ax

xxx

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1 2 3 4 5 6 7 8 9 10 11 12Prey category

Occ

urre

nce

of p

rey

cate

gory

a

b

bcbcbc

bc

bcbcbc

bc

bc

c

x

xy

xy

xyxy

y yy

xx

y

d.

y

Figure 3.5: Significance of dietary items for foxes during breeding (a.), whelping (b.), rearing (c.) and exploratory (d.) biological seasons for 2005 (■; a, b and c) and 2006 (■; x, y and z). Occurrence of the same letter in relevant columns indicates lack of significant difference.


147

Mean occurrences of each prey category for dingoes are compared with mean

occurrences of each prey category for foxes per season in Figures 3.6a-l. These

graphs illustrate the significance of dietary differences for dingoes and foxes in the

SGBMWHA.

a. Swamp wallaby d. European rabbit

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

+++ ++

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

b. Macropod species e. Brushtail possum

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

c. Medium ground dwelling native species f. Other arboreal species

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

+

Figure 3.6: Comparison of mean occurrence of prey categories in diet of dingoes (■) with diet of foxes (■) in the SGBMWHA within breeding (B), whelping (W), rearing (R) and exploratory (E) biological seasons during 2005 (05), 2006 (06) and 2007 (07). Significant differences between dingo and fox diet at P < 0.05 (+) and differences approaching significance P < 0.1 (^) are indicated above the seasons when differences were observed.


148

g. Small ground dwelling mammals’ j. Avifauna

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

^ ^

h. Other k. Reptiles

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

^

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts^

i. Vegetative material l. Insects

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

+

0

0.2

0.4

0.6

0.8

1

1.2

B05 B06 B07 W05 W06 R05 R06 E05 E06

Pro

port

ion

occu

rren

ce in

sca

ts

+

Figure 3.6 (cont.): Comparison of mean (± sd) occurrence of prey categories in diet of dingoes (■) with diet of foxes (■) in the SGBMWHA within breeding (B), whelping (W), rearing (R) and exploratory (E) biological seasons during 2005 (05), 2006 (06) and 2007 (07). Significant differences between dingo and fox diet at P < 0.05 (+) and differences approaching significance P < 0.1 (^) are indicated above the seasons when differences were observed.

Observations of prey selection by dingoes and foxes over time are displayed using the

proportional PDI in Figures 3.7a-d. These graphs also clearly demonstrate the

opportunistic nature of foxes as a mesocarnivore when compared with the selective

predation by dingoes as a hypercarnivore.


149

a. c.

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

PD

I

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

PD

I

b. d.

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PD

I

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PD

I

Figure 3.7: Use of the Predator Diet Index to illustrate annual changes in the monthly occurrence of swamp wallaby and brushtail possum ▲ within dingo diet at Site 1 (a.) and Site 2 (b.) and fox diet at Site 1 (c.) and Site 2 (d.) from analyses of hair and bone fragments in faecal material collected over 26 months. * Only two sand plot nights observed. Field trips were cancelled due to wet weather in February 2007 and scats were only collected on one day in Site 2 in April 2007.

3.3.2 Activity and abundance

There were 3700 plot nights comprising 1807 dingo tracks (Site 1: 759; Site 2: 1048),

1805 fox tracks (Site 1: 986; Site 2: 819) and 377 cat tracks (Site 1: 262; Site 2: 115).

European rabbit tracks were most common with 2574 tracks (Site 1: 1258; Site 2:

1316) though this may be due to their method of foraging. Of native prey species,

there were 1583 eastern grey kangaroo tracks (Site 1: 162; Site 2: 1421), 953 wallaby

tracks (Site1: 570; Site 2: 383), 2097 wombat tracks (Site 1: 623; Site 2: 1474), 269

brushtail possum tracks (Site 1: 57; Site 2: 212) and 303 tracks of Varanid sp. (Site 1:

92; Site 2: 211). Records of native prey species may also be affected by their method

of foraging or seasonal torpor in the case of Varanid sp. Other tracks recorded but

excluded from analyses included humans, frogs, echidnas, small mammals, feral


150

cattle, feral pigs and bird species. Total tracks counted were 4769 in Site 1 and 6999

in Site 2 (n = 11768) tracks of various species spanning over 26 months.

Passive activity indices are compared with relative abundance rating indices for

dingoes, foxes and cats in Figure 3.7. Periods of high and low activity occurred

concurrently with periods of high and low relative abundance. These data indicated

that activity is relative to abundance or vice versa, and abundance or activity on sand

plots fluctuated annually. Presentation of PAI with ARF showed the ARF merely

reflected changes in activity patterns. However, the method has been of value in

measuring before and after effects of pest control programs.

a. b.

05

10152025303540

AR

F

0

0.2

0.4

0.6

0.8

1

1.2

PA

I

05

1015202530354045

AR

F

00.20.40.60.811.21.4

PA

I

c. d.

05

10152025303540

AR

F

0

0.2

0.4

0.6

0.8

1

1.2

PA

I

05

10152025303540

AR

F

00.20.40.60.811.21.4

PA

I

e. f.

05

10152025303540

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

0

0.2

0.4

0.6

0.8

1

1.2

PA

I

05

10152025303540

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

00.20.40.60.811.21.4

PA

I

Figure 3.8: Comparison of monthly “relative abundance” using the Abundance Rating Formula (ARF) (/orange) with monthly “activity” using the Passive Activity Index (PAI) (/blue) as determined using the sand plot technique for Site 1 dingoes (a.), Site 2 dingoes (b.), Site 1 foxes (c.), Site 2 foxes (d.), Site 1 cats (e.) and Site 2 cats (f.) between March 2005 and April 2007. * Only two sand plot nights observed.


151

In Figure 3.8 the monthly activity of dingoes, foxes and cats in Site 1 and Site 2 are

compared, using the sand plot technique and the PAI. Standard deviations have been

excluded to maintain visual clarity on the graphs. Trends in activity can be related to

biological seasons of the canids but cat activity shows no clear trends. Dingo ARF

negatively correlated with cat ARF (r = -0.289, P = 0.038). These data indicate cats

either avoided dingoes on roads or dingoes limited cat activity and abundance. No

significant relationship between dingoes and foxes or foxes and cats was observed

using Pearson, Spearman or Kendall correlation analyses.

a. Dingo activity b. Fox activity

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

c. Cat activity

0

0.1

0.2

0.3

0.4

0.5

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

Figure 3.9: Monthly activity (PAI) of dingoes (a.), foxes (b.) and cats (c.) as determined using the sand plot technique and the passive activity index (PAI) in Site 1 () and Site 2 () from March 2005 until April 2007. (February to May ▬ = exploratory season (not included for cats); May to August ▬ = breeding season; August to November ▬ = whelping season; November to February ▬ = rearing season). * Only two sand plot nights observed.

Relationships between dingo and fox activity, dingo and cat activity and fox and cat

activity for each site over 26 months are shown in Figure 3.10. Figure 3.10c and


152

Figure 3.10d show slight increase in cat activity during periods of low dingo activity

and relative decrease during periods of high dingo activity.

a. b.

0

0.2

0.4

0.6

0.8

1

1.2

M*2005

M J S* N J2006

M M J S* N J2007

M

PA

I

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

PA

I

c. d.

0

0.2

0.4

0.6

0.8

1

1.2

M*2005

M J S* N J2006

M M J S* N J2007

M

PA

I

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

PA

I

e. f.

0

0.1

0.2

0.3

0.4

0.5

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

0

0.1

0.2

0.3

0.4

0.5

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

Figure 3.10: Comparison of changes in PAI for Site 1 dingoes with Site 1 foxes (a.), Site 2 dingoes with Site 2 foxes (b.), Site 1 dingoes with Site 1 cats (c.), Site 2 dingoes with Site 2 cats (d.), Site 1 foxes with Site 1 cats (e.) and Site 2 foxes with Site 2 cats (f.). = dingoes; = foxes; and = cats. * Only two sand plot nights observed.

No significant relationships between medium-sized vertebrate prey tracks, activity, or

abundance with fox and cat activity were observed. Dingo PAI, ARF and raw track

counts were associated with the activity, abundance and raw occurrence of brushtail


153

possum, wombat and kangaroo on roads in this study area. Combining data from both

sites, raw kangaroo tracks, PAI and ARF for kangaroos were significantly correlated

with raw dingo tracks (r = 0.440, P = 0.001) dingo PAI (r = 0.412, P = 0.002) and

dingo ARF (r = 0.468, P = 0.000). This indicated kangaroo activity increased when

dingo activity increased (Figure 3.11).

a. b.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

Figure 3.11: Relationship between eastern grey kangaroo PAI () and dingo PAI () for Site 1 (a.) and Site 2 (b.). * Only two sand plot nights observed.

Positive relationships between raw dingo tracks, PAI and ARF for dingoes and raw

wombat tracks (r = 0.339, P = 0.014), wombat PAI (r = 0.297, P = 0.033) and

wombat ARF (r = 0.413, P = 0.002) were observed when data from both sites was

combined. This indicated that wombat activity increased when dingo activity

increased (Figure 3.12).

a. b.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PA

I

Figure 3.12: Relationship between common wombat PAI () and dingo PAI () for Site 1 (a.) and Site 2 (b.). * Only two sand plot nights observed.


154

There was a positive significant relationship between ARF for dingoes and brushtail

possum ARF (r = 0.300, P = 0.031) when data from both sites was combined. This

indicated possum abundance increased simultaneously with increased dingo activity

and abundance (Figure 3.13).

a. b.

05

1015202530354045

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

05

1015202530354045

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

Figure 3.13: Relationship between brushtail possum ARF () and dingo ARF () in Site 1 (a.) and Site 2 (b.). * Only two sand plot nights observed.

3.3.3 Relationship between diet and activity

Table 3.2 explains acronyms for the Pearson correlation analysis between activity of

medium-large sized vertebrates and their occurrence in diet for dingoes and foxes.

Significant results from these analyses are highlighted in Tables 3.3 and 3.4 and

presented in context with predator-prey relationships.


155

Table 3.2: Explanation of acronyms for Tables 3.3 and 3.4

Acronym Explanation DAR Dingo spoor on sand plots as raw tracks DAP Dingo spoor on sand plots as PAI DAF Dingo spoor on sand plots as ARF TDS Total dingo scats FAR Fox spoor on sand plots as raw tracks FAP Fox spoor on sand plots as PAI FAF Fox spoor on sand plots as ARF TFS Total Fox Scats WAR Wallaby spoor on sand plots as raw tracks WAP Wallaby spoor on sand plots as PAI WAF Wallaby spoor on sand plots as ARF WDR Wallaby occurrence in diet as raw counts WDI Wallaby occurrence in diet as PDI KAR Kangaroo spoor on sand plots as raw tracks KAP Kangaroo spoor on sand plots as PAI KAF Kangaroo spoor on sand plots as ARF KDR Kangaroo occurrence in diet as raw counts KDI Kangaroo occurrence in diet as PDI BPAR Brushtail possum spoor on sand plots as raw tracks BPAP Brushtail possum spoor on sand plots as PAI BPAF Brushtail possum spoor on sand plots as ARF BPDR Brushtail possum occurrence in diet as raw counts BPDI Brushtail possum occurrence in diet as PDI ERAR European rabbit spoor on sand plots as raw tracks ERAP European rabbit spoor on sand plots as PAI ERAF European rabbit spoor on sand plots as ARF ERDR European rabbit occurrence in diet as raw counts ERDI European rabbit occurrence in diet as PDI WOAR Wombat spoor on sand plots as raw tracks WOAP Wombat spoor on sand plots as PAI WOAF Wombat spoor on sand plots as ARF WODR Wombat occurrence in diet as raw counts WODI Wombat occurrence in diet as PDI PC Pearson Correlation Sig. Significance

Table 3.3: Pearson correlation coefficients for comparative analyses of activity/abundance records and occurrence of medium sized vertebrate prey in dingo diet (n = 52). Bold numbers indicate two tailed significance at the 0.01 and 0.05 levels or correlations approaching significance (0.06-0.1)

Table 3.4: Pearson correlation coefficients for comparative analyses of activity/abundance records and occurrence of medium sized vertebrate prey in fox diet (n = 52). Bold numbers indicate two tailed significance at the 0.01 and 0.05 levels or correlations approaching significance (0.06-0.1)


158

Presentation of results below are based on those which are significant (P < 0.05) or

approaching significance (P < 0.1). Faecal abundance increased when dingo and fox

activity decreased and faecal abundance decreased when dingo and fox activity

increased (Figure 3.14). The correlation between TDS and DAP was negative and

approaching significance. TFS and FAP was, however, a positive correlation though

not significant.

a. c.

020406080

100120140

Num

ber

of s

cats

00.20.40.60.811.21.4

PA

I

0

20406080

100120140

Num

ber

of s

cats

00.20.40.60.811.21.4

PA

I

b. d.

020406080

100120140

M(2005)

M J S* N J(2006)

M M J S* N J(2007)

M

Month

Num

ber

of s

cats

00.20.40.60.811.21.4

PA

I

020406080

100120140

M(2005)

M J S* N J(2006)

M M J S* N J(2007)

M

Month

Num

ber

of s

cats

00.20.40.60.811.21.4

PA

I

Figure 3.14: Comparison of number of scats collected per month (columns) and passive activity (lines) per month for Site 1 (a.) and Site 2 (b.) dingoes and Site 1 (c.) and Site 2 (d.) foxes of the SGBMWHA . * Only two sand plot nights observed.

Significant negative correlations between KAP and KAF and WOAP and WOAF with

the occurrence of kangaroo and wombat in TFS indicated their occurrence as dietary

items for foxes is unrelated to their activity. DAF was significantly related to KDI

which suggested that the relative abundance of dingoes affected predation of

kangaroos (Figure 3.15).


159

a. b.

0

1020

30

4050

60

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

0

0.10.2

0.3

0.40.5

0.6

PD

I

0102030

405060

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

AR

F

00.10.2

0.30.40.50.6

PD

I

Figure 3.15: Relationship between eastern grey kangaroo PDI (; green) and dingo ARF (; black line) for Site 1 (a.) and Site 2 (b.). * Only two sand plot nights observed. = dingoes;

BPDR and BPDI from dingo scats was significantly negatively correlated with DAR

and DAP. These relationships showed possum increased as a dietary item when dingo

activity on roads decreased (Figure 3.16). BPDI was the only PDI positively

significantly correlated with TDS.

a. b.

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PD

I / P

AI

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

PD

I / P

AI

Figure 3.16: Relationship between brushtail possum PDI (; purple) and dingo PAI (; black line) for Site 1 (a.) and Site 2 (b.). * Only two sand plot nights observed.

Figure 3.17 shows the variation in prey selection of swamp wallabies, eastern grey

kangaroos and brushtail possums by dingoes during biological seasons. Brushtail

possum and eastern grey kangaroo increased as diet items when dingoes were

breeding and whelping. Predation of swamp wallaby in both sites increased with

increased dingo activity during rearing and exploratory seasons. Swamp wallaby PDI

has been included for comparison of brushtail possum PDI and eastern grey kangaroo

PDI because it is the most common dietary item over time.


160

a. c.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Din

go P

DI

0

0.2

0.4

0.6

0.8

1

1.2

1.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Din

go P

DI

b. d.

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

Din

go P

DI

00.20.40.60.8

11.21.4

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

Din

go P

DI

Figure 3.17: Relationship between dingo PAI, dingo biological seasons and swamp wallaby PDI () from dingo scats from Site 1 (a.) with eastern grey kangaroo PDI () from dingo scats, Site 2 (b.) eastern grey kangaroo PDI () from dingo scats, Site 1 (c.) brushtail possum PDI () from dingo scats and Site 2 (d.) brushtail possum PDI () from dingo scats. Swamp wallaby PDI from dingo scats was included in all graphs because it was the most consistent dietary item. Red = exploratory; black = breeding; green = whelping; and blue = rearing seasons for dingoes. Due to a small sample size for faecal material collected during February 2007, February 2007 and March 2007 were grouped. * Only two sand plot nights observed.

It is clear from these data that the PDI can be used in conjunction with the PAI to

illustrate relationships between changes in the frequency of dietary items relative to

biological seasons of the predators. Figure 3.7 showed how foxes opportunistically

consumed swamp wallabies and brushtail possums. Figure 3.8 showed fox activity in

the SGBMWHA appeared to correspond with their biological seasons. Figure 3.18

shows no clear trends between the PDI for swamp wallaby, eastern grey kangaroo and

brushtail possum in fox scats and the PAI for fox activity in Site 1 and Site 2.


161

a. c.

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Fox

PD

I

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Fox

PD

I

b. d.

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

Fox

PD

I

0

0.2

0.4

0.6

0.8

1

M*2005

M J S* N J2006

M M J S* N J2007

M

Month

Fox

PD

I

Figure 3.18: Relationship between fox PAI, fox biological seasons and swamp wallaby PDI () from fox scats sampled in Site 1 (a.) with eastern grey kangaroo PDI () from fox scats, Site 2 (b.) eastern grey kangaroo PDI () from fox scats, Site 1 (c.) brushtail possum PDI () from fox scats and Site 2 (d.) brushtail possum PDI () from fox scats. Swamp wallaby PDI from fox scats was included in all graphs because it was the most consistent dietary item. Red = exploratory; black = breeding; green = whelping; and blue = rearing seasons for foxes. Due to a small sample size for faecal material collected during February 2007, February 2007 and March 2007 were grouped. * Only two sand plot nights observed. Since the study sites had unlimited water resources and modified habitat with roads

and ex-pastoral cleared land it is difficult to test for relationships between predator

dietary shifts and fluctuations in activity or relative abundance of prey. In some

instances, prey activity/abundance was significantly correlated with the occurrence of

other prey species as dietary items. WAR, WAP and WAF had a significant

relationship with the WPDI for dingo scats (Table 3.3) showing that wallaby activity

was positively correlated with the occurrence of swamp wallaby as a prey item.

WPDI for fox scats was also significantly correlated with WAR and WAP (Table

3.4). There was a negative correlation between the WDR in TDS and WPDI with

KAP. This suggested kangaroo activity was lowest when dingoes were consuming

swamp wallabies or alternatively, dingoes changed to consuming swamp wallaby


162

when kangaroos were less active. Similarly there was a significant negative

correlation for occurrence of swamp wallaby in TFS involving KAR, KAP and KAF.

WPDI for fox diet showed significant negative correlation with KAP and KAF.

WOAR, WOAP and WOAF showed negative correlation with raw occurrence of

swamp wallaby as a prey item with TFS and WPDI for fox diet was significantly

negatively correlated with WOAP and WOAF. WOAP was negatively correlated

with WDR in dingo scats. WDR in dingo scats positively correlated with WODR,

ERDR and BPDR, as did BPDI.

KDR in dingo diet was positively correlated with KAF. KDI positively correlated

with KAR, KAP and KAF. KDR in dingo diet also showed positive correlation with

WOAR, WOAP and WOAF, as did KDI. Kangaroo in TFS was negatively correlated

with WAR and WAP indicating kangaroo remains were found less frequently in fox

scats when wallaby activity was high and when a high proportion of dingo scats

contained swamp wallaby remains.

BPAF negatively correlated with BPDR and BPDI. BPDR in TDS and TFS

negatively correlated with KAR, KAP and KAF. BPDI from dingo scats also

negatively correlated with KAR, KAP and KAF but BPDI from fox scats showed no

relationship with activity of other prey species. WOAP was negatively correlated

with BPDR in dingo and fox scats, as was BPDR in fox scats with WOAF.

ERDR in fox scats had a negative relationship with KAR, KAP and KAF, WOAP and

WOAF. WODR in fox scats related negatively with KAR, KAP and KAF. WOAP

and WOAF were also negatively correlated with WODR in fox scats. The Pearson

correlation values for relationships between PAI and PDI for each species are

presented in Figure 3.19.


163

a.

b.

Figure 3.19: Relating dingo (a.) and fox (b.) activity with prey activity and the occurrence of swamp wallaby (), eastern grey kangaroo (), brushtail possum (), European rabbit () and common wombat () as dietary items.

Figure 3.19 summarises the relationship between predator activity and the occurrence

of prey species in scats with prey activity and the occurrence of prey species in scats.

The amount of kangaroo consumed increased, for instance, during periods of high

dingo activity on roads and the amount of brushtail possum consumed increased

during periods of low dingo activity on roads. Similarly, consumption of swamp

wallaby increased when more wallaby tracks were being recorded on sand plots and


164

consumption of kangaroo increased when fewer wallabies were being recorded on

sand plots.

3.4 Discussion

In contrast to many dietary studies of the dingo in Australia, this study showed

variation in dingo diet when data were analysed according to breeding, whelping,

rearing and exploratory seasons. Comparison of breeding season 2005 with breeding

season 2006, whelping season 2005 with whelping season 2006, rearing season 2005

with rearing season 2006 and exploratory season 2005 with exploratory season 2006

showed dingo diet significantly varied between seasons, though not within seasons.

Inconsistent patterns of seasonality in fox diet in this study over time provided

evidence consistent with the hypothesis that the dingo is a hypercarnivorous predator

in this ecosystem and the fox is an opportunistic mesocarnivore. These results are

also consistent with other research on fox diet in Australia (Green and Osborne 1981;

Lunney et al. 1990; May and Norton 1996; Meek and Triggs 1998; Paltridge 2002;

Mitchell and Banks 2005). With the exception of consumption of swamp wallabies,

dingo and fox diets generally overlapped within seasons. Mitchell and Banks (2005)

made similar observations between dingoes and foxes in the GBMWHA. Given that

fox diet was used to show differences between functional ecology of mesocarnivores

and hypercarnivores, aspects of fox diet will not be discussed in great detail. Dietary

overlap, such as predation on eastern grey kangaroos, almost certainly indicated

scavenging activity of foxes following dingo predation though data presented by

Banks et al. (2000) suggested that foxes may limit recruitment in kangaroo

populations.


165

Data in the current study showed numerous positive and negative correlations

between diet and activity of dingoes, foxes and prey species (Tables 3.3 and 3.4), and

demonstrated that the functional role of dingoes in the SGBMWHA is as defined in

section 1.3. Predation of some prey species (swamp wallaby, eastern grey kangaroo,

brushtail possum, European rabbit and common wombat) appeared to be related to

biological cycles of dingoes but this could not be tested because it was beyond the

scope of this study and more data on prey populations is required. Seasonal prey

consumption was repeated over 24 months. Whether the changes in passive activity

or relative abundance of prey species was a response to dingo activity/relative

abundance, or dingo hunting strategies were a response to changes in prey populations

is equivocal. These data showed that seasonal changes in predator/prey interactions

occurred and that the dingoes maintained a functional role in the SGBMWHA

between 2005 and 2007.

As found in other studies in eastern and south eastern Australia (Newsome et al.

1983a, b; Robertshaw and Harden 1985; Corbett 2001), swamp wallaby was the

dominant prey of dingoes. In contrast to other research however, this study showed

variation in the amount of swamp wallaby consumed over time. The relationship

between raw activity of wallabies and raw occurrence of swamp wallaby as prey

indicated that predation on swamp wallaby was related to wallaby activity or relative

abundance and not just as a preferred prey. Although red-necked wallabies, swamp

wallabies and wallaroos were commonly seen, and recognition of their spoor was

subjective, the significant correlation between swamp wallaby PDI and wallaby

activity suggests data on wallaby activity is consistent with the success of swamp

wallaby predation by dingoes. Activity of wallabies was higher at Site 1 than Site 2,

though this may be due to consistent water points near favourable habitat and


166

vegetation disturbance due to annual maintenance of fire trails via slashing in Site 1.

Compared with other studies in less modified habitats significant relationships

between prey activity/abundance and their occurrence as a dietary item were

observed. Newsome et al. (1983a) reported similar findings and suggested that

preference of these frequently disturbed habitats by macropods may alter dingo

activity on roads because the infrastructure creates access and a reliable source of

food. Alternatively, data presented in chapter four shows dingoes use prominent

landscape features (ridges) for marking territories, which is where most roads were

constructed.

Robertshaw and Harden (1985) identified swamp wallaby as the major contributor to

dingo diet in north east NSW. In their study, swamp wallaby was found in 30.5% of

1993 dingo scats. Robertshaw and Harden (1986) also found that the number of

dingo scats collected on fire trails peaked in winter months and receded in summer

months. In contrast to the current study, bush rat produced the second highest

proportion of diet at 12.24%, red-necked wallaby third at 11.14% and Trichosurus sp.

fourth at 6.92% in total scats (Robertshaw and Harden 1985). Red-necked wallaby

was not observed in dingo scats in the SGBMWHA from 2005-2007 however,

Mitchell and Banks (2005) reported red-necked wallaby in 6.7% of scats in the

GBMWHA.

Brushtail possum and eastern grey kangaroo were the second most frequently

occurring prey items at Site 1 and Site 2 respectively. The latter was a minor

contributor to diet of both canid species in Site 1, surroundings of which were

characterised by steep sometimes densely vegetated moist gullies, ephemeral creeks

and one major river. In comparison, kangaroo was a larger part of the preferred diet


167

for dingoes in Site 2 which was characterised by regenerating ex-pastoral cleared

areas and open woodland habitat located alongside a major river. This increase in

kangaroo predation indicated that variation of prey preference between sites is

probably due to differences in habitat and density of prey.

Sheep and cattle were observed on rare occasions as dietary items in the current study.

Records of illegal poaching and pig hunting activity by trespassers with domestic

dogs in the field sites may explain these species in collected faecal material. Wool

may also have been scavenged from the relic wool shed at an ex-pastoral lease. A

feral wild cattle population is also present in the study site and cattle were probably

scavenged as carrion.

Newsome et al. (1983b) suggested that there are problems when nominating primary,

secondary or alternative prey for the dingo. Primary prey in the current study was

consistently swamp wallaby. Brushtail possum could be considered secondary prey

because it occurred more than eastern grey kangaroo when sites were combined. If

not, primary, secondary or alternative, prey may depend on the pack, habitat and prey

density within the territory of each dingo pack. Chapter one showed prey preference

varied based on prey availability from the south east to the north west of Australia.

Secondary prey items were notably different between Site 1 and Site 2 in the current

study. Alternatively, prey may be nominated as staple, supplementary, opportune or

scavenged (Newsome et al. 1983a). Nominating staple, supplementary, opportune or

scavenged dietary items is probably only applicable in current data sets since diet may

shift dramatically, for instance, in the event of a fire (Newsome et al. 1983a) or shifts

in the weather patterns such as drought or high rainfall and in response to prey

availability (Corbett and Newsome 1987). From the current data, swamp wallaby


168

may be the staple prey whilst brushtail possum and kangaroo can be identified as

supplementary prey. Predation on other prey appeared to be opportunistic based on

consistency or inconsistency in observed trends. Depending on the search method

employed by the predator such as adaptive foraging (Abrams 1992) or Lévy foraging

(Atkinson et al. 2002), all prey items may be opportunistic and relative to abundance.

Corbett and Newsome (1987) used evidence from a seven year study on wild dingo

ecology in central Australia to introduce the concept of ‘alternation of predation’.

Their hypothesis was that dingoes alternated between large and small to medium sized

prey during periods of drought and wet or “flush” periods respectively. They

proposed that this interaction with prey and the environment was consistent with all

models of foraging behaviours. One major difference in the study by Corbett and

Newsome (1987), compared with the current study, is that they determined diet from

contents of 386 stomachs. Dingoes from their study area were culled to collect

samples, whereas data presented in the current study were collected using non-

invasive techniques based on identifying prey remains in scats. In their study, 58

stomachs were sampled during the first sampling period, followed by 49 in the

second. Later sample sizes remained below 35 dingo stomachs per period with the

exception of one year, when 42 stomachs were collected after the false break in the

drought (sample sizes: 58, 49, 18, 31, 23, 34, 35, 30, 35, 42, 18 and 13 over seven

years). Their sampling technique which successively removed dingoes from the

population would have changed the age, stability and pack structure of dingoes within

their population following repeated regular sampling. The ability of some individuals

to hunt in pack formation and to take large prey may also have been compromised.

Thomson et al. (1992a) suggested that perturbations in dingo social systems enabled

solitary dingoes to establish a territory in areas no longer defended by packs, which


169

also implied neighbouring pack ranges could have expanded. Allen and Gonzalez

(1998) showed predation on livestock increased following dingo control campaigns

which may have disrupted the dingo social system. Corbett and Newsome (1987)

were possibly sampling dingoes from an unstable population. Dingo abundance in

their study appeared to have been reduced after the first two sample periods since

numbers of stomachs sampled, or numbers of dingoes available to be shot or trapped,

did not approach the same sample size for four years. Proceeding the fourth year

when 42 samples were collected, dingo numbers had obviously decreased, having two

of the three lowest sampling events in seven years. Corbett and Newsome (1987),

therefore, studied variations in the feeding ecology of dingoes during seven years of

high, human induced dingo mortality during climate-induced fluctuations in prey

populations. Results presented in the current study were not affected by human

induced mortality of either predator or prey species.

Interpretation of changes in dingo abundance and changes in dingo diet reported by

Newsome et al. (1983a) may have also been affected by uncontrollable changes in the

landscape during their study. The research site was affected by wildfire, and the

ecology of the landscape was in a regeneration phase during data collection.

Hypothetically, measures of dingo “abundance” may have increased because scent

posts would have been consumed by the fire and packs would need to remark their

territories. The dingo population may have increased to aid territory defence though it

is more likely that the levels of dingo activity increased whilst territories and packs

were reforming. Data in the present study were consistent with these claims and

showed dingo abundance and activity were directly related. Predation of macropods

also increased post-fire when dingo “abundance” increased, consistent with

observations of macropod predation in the SGBMWHA. Pending reformation of


170

dingo social systems and the kangaroo population in the study by Newsome et al.

(1983a), one may hypothesise macropod predation would revert to similar seasonal

patterns as those observed in the SGBMWHA. Newsome et al. (1983a) also showed

that dingoes in Nadgee Nature Reserve in south eastern Australia maintained the

equilibrium of the ecosystem by suppressing populations of herbivores during

instability in the environment. Again, data presented for the current study were not

affected by natural phenomena such as fire, flood, or climatic fluctuations and dietary

shifts observed appeared to relate to biological seasons of dingoes. Drought could

have affected abundance of some prey species in the SGBMWHA during this study

however, water sources could be located on transects and within 1km of both transects

at all times of the year.

Results presented in the current study are different to past studies (Newsome et al.

1983a; Corbett and Newsome 1987) due to the intensity of sampling regimes, and

differences in analysis of results compared with Robertshaw and Harden (1985).

Newsome et al. (1983a) and Corbett and Newsome (1987) compared data from

analyses of diet and activity with fluctuations in prey populations subject to

environmental perturbations. The current study, in contrast, showed monthly and

seasonal shifts in prey preference of dingoes related to dingo biological seasons.

Shepherd (1981) documented dingo predation on red kangaroos during January and

March in Sturt National Park, NSW. In total, 83 red kangaroos were killed within

150m of a water source in that study. Relating these data with results presented in the

current study implies the dingoes studied by Shepherd (1981) may have been training

juveniles how to hunt kangaroos, especially because the predation rate decreased by

March when dingo breeding season commences. Shepherd (1981) also stated that the


171

pattern of predation appeared to be normal based on the behaviours exhibited by

juvenile and female kangaroos when compared with male kangaroos. Robertshaw

and Harden (1985) showed significant differences in the composition of diet between

May-October when the proportion of scats containing non-macropod species

increased and November-April when proportion of scats containing macropods

increased. These observations are directly comparable with data shown in the current

study when brushtail possum increased in the diet of dingoes between June and

November in 2005 and 2006 and swamp wallaby increased in dingo diet between

December and March in 2005 and 2006. Thus, as stated by Robertshaw and Harden

(1985), diet changed during the annual reproductive cycle of dingoes and subsequent

changes in the hunting strategy, and subsequent changes in the age structure of the

population, as reported by Allen and Gonzalez (1998).

The non-binary PAI and binary ARF were used in the current study to investigate

seasonal changes in activity or relative abundance over time. Blaum et al. (2008)

tested the efficacy of the PAI and two index calculations for small carnivores in the

southern Kalahari, South Africa. Correlations between longitudinal spoor counts on

transects (non-binary) combined with across transect intrusions (binary) showed non-

binary abundance indices positively correlated with binary abundance indices when

carnivore abundance was low. However, in a high density population, the binary

technique proved less useful to monitor trends. Compared with Allen et al. (1996),

the non-binary PAI proved to be more effective, based on the sensitivity of the

technique to monitor spoor, than non-binary indices calculated with data from active

(sand plots with meat or fatty acid scent) sand plot stations. In the current study, both

the PAI (total tracks per plot) and the ARF (presence/absence per plot) successfully


172

measured trends in activity and relative abundance that were consistent with the

biological seasons of dingoes. These measurement techniques are therefore useful to

monitor trends for adaptive management of the SGBMWHA dingo population.

Triggs (1996) stated reptilian scats can be difficult to differentiate from certain

mammals and some reptile scats may have been misidentified as dingo or fox scats in

preceding analyses or in any other study of canid diet in Australia. Past studies on

dingo and fox diet did not mention methods used to differentiate between scats of

reptilian and mammalian fauna. Inclusion of reptilian scats could be expected to

affect records of all prey groups because they scavenge carcasses. Furthermore, most

Varanid sp. are semi-arboreal animals so records of avifauna or arboreal mammals in

canid diet may be due to occasional false identification of faecal pellets. In addition,

swamp wallaby may not decrease so dramatically as a prey item when pups are

whelped but changes in the proportion of occurrence may stem from changes in the

proportion of material regurgitated by adult dingoes to pups. If pups were not of the

age to travel great distances away from the den, then records of diet during whelping

seasons may be questionable. To validate queries such as these will require controlled

experiments to compare regurgitated material with defecated material during

whelping seasons. Observations of two den sites during whelping season in the

SGBMWHA showed eastern grey kangaroo carcasses were numerous near the den

sites. These records are consistent with seasonal occurrence of kangaroo remains in

scats and suggested dingoes either predate kangaroos in areas accessible to pups, drag

carcasses to den sites, or herd kangaroos to den sites to kill them near the den.

Newsome et al. (1983a) and Corbett and Newsome (1987) discussed whether or not

dingoes controlled or regulated either feral cat and fox populations, or both, in

Australian ecosystems, thereby aiding the conservation of native Australian


173

marsupials (Glen and Dickman 2005; Glen et al. 2007; Johnson et al. 2007). Results

from the current study showed dingoes had no apparent affect on fox abundance or

activity in the SGBMWHA but may have either limited recruitment in cat populations

or caused avoidance behaviour by cats of dingoes on roads. Evidence from diet data

showed that dingoes selectively predate in contrast to opportunistic predation by

foxes. In this study, dingoes consumed fewer species and foxes consumed a higher

proportion of small ground dwelling mammals than dingoes. Fox predation in the

SGBMWHA was consistent with previous studies that indicated foxes have

detrimental effects on populations of small mammals (Green and Osborne 1981;

Burbidge and McKenzie 1989; May and Norton 1996; Green 2002). To more fully

understand the role of the dingo in the conservation of Australian ecosystems, a

holistic, landscape scale analysis is required because canid species are adaptable to a

variety of ecosystems (Macdonald and Sillero-Zubiri 2004b). The role of dingoes

may, therefore, be broader than the conservation of an ecosystem, but in the

conservation of ecosystems throughout a landscape.

Results from the current study showed how dingoes functioned at a landscape scale.

Wallaby activity was positively correlated with the occurrence of swamp wallaby as a

prey item for dingoes and foxes. The negative correlation between occurrence of

swamp wallaby in dingo diet with kangaroo activity on roads suggested that decreased

activity of kangaroos occurred simultaneously with increased foraging efforts by

dingoes on swamp wallabies or vice versa. Kangaroo activity was higher when

kangaroo increased as an item of diet when dingo activity on roads increased. These

data suggested that dingoes may influence kangaroo movement patterns, or kangaroo

abundance and activity influences dingo activity.


174

When kangaroo activity increased with dingo activity and the occurrence of kangaroo

in dingo scats, swamp wallaby decreased as an item of fox diet. This was presumably

relative to changes in dingo hunting strategies and the availability of carrion to foxes.

Increased activity by wombats somehow related with swamp wallaby not being

targeted by foxes or dingoes as frequently and wombat activity increased when

kangaroo activity increased. However, swamp wallaby as an item of diet positively

correlated with wombat, rabbit and brushtail possum as items of dingo diet.

Kangaroo in dingo diet significantly correlated with kangaroo abundance suggesting

predation on kangaroos increased with growth in the kangaroo population. In contrast

wombat activity and relative abundance increased when kangaroos were being

targeted by dingoes. Brushtail possum abundance was greater when brushtail

possums were not being targeted by dingoes and predation of possums was less when

kangaroo activity and kangaroo abundance increased. Evidently this may be because

greater abundance or activity of kangaroos correlated with increased predation of

kangaroos by dingoes. Similarly, wombat activity or relative abundance increased

when dingoes or foxes were concentrating on brushtail possums. These data showed

shifts in activity, abundance and diet of a hypercarnivorous predator may affect

activity and abundance of sympatric predators and their prey in forested, water rich

environments.

In terms of dingo biology and ecosystem function, this study indicated:

• Although activity patterns of most medium sized terrestrial vertebrates in this

study area could have been due to natural fluctuations, they were significantly

correlated with the biological seasons of the dingo;


175

• Foxes may have benefited from dingo predation (carrion) in this study area

though comparative research of fox ecology in a dingo-free study site is

required to test whether dingoes limited recruitment in the SGBMWHA fox

population;

• Predatory behaviours of dingoes may provide periods for vegetation to

regenerate from foraging by herbivores in some habitats when, dingoes shift

predation from kangaroos or brush tail possums to swamp wallabies and vice

versa; and

• Feeding ecology of dingoes changed with the reproductive needs or age

structure of the dingo population, as hypothesised by Allen and Gonzalez

(1998), suggesting adult dingoes train juvenile dingoes how to hunt and

survive between whelping and breeding seasons, on an annual basis.

This chapter has detailed some elements of the function of dingoes in this study area.

Dingoes may facilitate ecosystem processes within this landscape through natural

repeated predatory behaviours during the identified biological seasons of the dingo.

Additional positive or negative effects between dingoes with competitors or prey may

be present but outside the scope of the current study. No effects of mesopredator

release were tested for, for instance, so the positive or negative effect of dingoes on

fox and cat populations in this ecosystem cannot be deduced. This is also the case for

populations of other introduced pests that were not tested for in this study, such as

pigs and goats. To describe the functional ecology of dingoes in this study area in

more detail, analyses of home range and a description of short and annual movement

patterns of some genetically related individuals are analysed in chapter four.

Chapter Four:

Intraspecific variation in spatial organisation, movements and

activity

Spatial organisation, movements and activity

177

Chapter Four: Intraspecific variation in

spatial organisation, movements and

activity 4.1 Introduction ..................................................................................... 178

4.1.1 Social systems of the Canini tribe ........................................................ 180

4.1.2 Effect of social systems on movement patterns ................................... 182

4.1.3 Activity, movement, dispersal and home range of dingoes ................. 184

4.2 Methods ............................................................................................ 184

4.2.1 Home range .......................................................................................... 187

4.2.2 Patterns of movement ........................................................................... 189

4.2.3 Patterns of activity ................................................................................ 190

4.2.4 Observations of mortality and dispersal .............................................. 191

4.3 Results .............................................................................................. 191

4.3.1 Home range .......................................................................................... 196

4.3.2 Patterns of movement ........................................................................... 203

4.3.3 Patterns of activity ................................................................................ 211

4.2.4 Observations of mortality and dispersal .............................................. 218

4.4 Discussion ......................................................................................... 220

4.4.1 Social systems of the Canini tribe ........................................................ 224

4.4.2 Effect of social systems on movement patterns ................................... 226

4.4.3 Activity, movement, dispersal and home range of dingoes ................. 230


178

4.1 Introduction

Radio telemetry has previously been used to monitor movements of canids and other

wide-ranging carnivores (for example: wolves, Ciucci et al. 1997; coyotes, Gantz and

Knowlton 2005; lions Panthera leo, Hemson et al. 2005; bears Ursus americanus,

Kindall and Van Manen 2007). Studies generally tested hypotheses associated with

home range size, home range use and patterns in activity and movement, including

dispersal. Some research integrated analyses of behaviour to understand mechanisms

for observed patterns in movement (Bekoff and Wells 1981; Laundré and Keller

1981; Thomson 1992a, b, c, d; Thomson et al. 1992 a, b). Gregarious behaviours

have been implicated as stimuli for altering feeding behaviour and patterns in activity,

such as those described in chapter three, and general movement patterns (cf.

Macdonald and Sillero-Zubiri 2004b). Laundré and Keller (1981) used behavioural

data to show that the home range of coyotes consisted of core areas where animals

spent extended time periods resting or hunting. A third type of behaviour was also

observed periodically when coyotes ranged to areas that surrounded or were adjacent

to core areas (Laundré and Keller 1981). It was hypothesised that this behaviour was

used in the re-establishment of territory boundaries, investigation of neighbouring

groups or as a second form of hunting behaviour. Sillero-Zubiri and Macdonald

(1998) partially supported these observations using behavioural data from Ethiopian

wolves Canis simensis and termed it “border patrol”.

An underlying objective in many aspects of biological research has been to determine

patterns in resource selection by animals (Samuel and Green 1988). If an animal or

group of animals can secure a core area with optimal habitat, which includes access to

reliable water and prey or food resources, then their movements will revolve around


179

foraging and the maintenance of the optimal habitat (Kenward 2001). This is often

shown in results of home range studies though attaining sufficient data to assess

patterns of optimal habitat maintenance can be time consuming (Laundré and Keller

1981; Harden 1985). Core areas would usually be maintained by carnivores by using

vocalisation, olfactory signalling and other behavioural gestures to caution or inform

sympatric competitors of their state. Depending on the reproductive cycle and the

ambition of an individual or group to secure resources in an optimal habitat, territories

need to be defined and may need to be defended (Asa et al. 1985; Sillero-Zubiri and

Macdonald 1998; Allen et al. 1999). Residents have to ensure neighbouring and lone

competitors can identify territories and are made aware that they have to contest for

resources of a core area or evade conflict.

On these grounds, the rationale for cyclical movement patterns can be better

understood. If canids maintain scent posts at regular intervals, neighbouring

individuals or groups are kept informed of changes within social systems (Sillero-

Zubiri and Macdonald 1998; Allen et al. 1999). Changes in the frequency of scent

post maintenance could indicate opportunities to enter a core area and challenge

residents for territory or to join a pack to secure resources and increase gene flow.

Seasonal activity patterns and changes in feeding behaviour shown in chapter three

appear to be related to seasonal changes in territory maintenance, patterns of

movement and patterns of activity. The relationships between packs and their

territories are investigated in chapter four. Data are interpreted and discussed in

relation to identified gregarious behaviours of other hypercarnivorous canid species.


180

4.1.1 Social systems of the Canini tribe

Macdonald et al. (2004) comprehensively reviewed canid social systems. The two

overarching questions for this review were:

1. Why do some canids live in groups, while others do not, and what shapes their

societies?

2. How do body size and associated energy requirements relate to prey size and

home range size?

In a study of energy requirements of terrestrial carnivores, Carbone et al. (1999)

developed a series of mathematical models known as the net rate model analysis.

This analysis, later validated by Macdonald et al. (2004), was based on correlations

between carnivore body mass and their most common prey. The net rate model

analysis showed that canids heavier than 21.5kg could not be sustained by small prey,

and shifted to larger prey with equal or greater body mass at this point. The mean live

weight of dingoes in the current study is less than 21.5kg (cf. chapters one and two),

though based on diet analysis of prey species, dingoes can be classed as a

hypercarnivorous canid. This classification is supported by Shepherd (1981) because

the predation rate of dingoes on red kangaroos (0.38kg prey per kg predator per day)

exceeds the predation rate of wolves on moose (0.13kg prey per kg predator per day),

and wolves are hypercarnivorous canids (Macdonald and Sillero-Zubiri 2004b).

Based on the amount of energy required to survive, time spent resting and time spent

hunting are identified as the behaviours which ultimately shape canid societies

(Macdonald et al. 2004). Since hypocarnivorous and mesocarnivorous canids only

require small to medium sized prey to meet daily energy requirements they can forage


181

opportunistically alone. In contrast, hypercarnivorous canids are required to live

communally to sustain adequate metabolic needs for hunting and resource defence.

Resources that require defence for optimal foraging include food, water and habitat

within optimal space. Discussion of optimal space requirements, or home range size,

for canids by Macdonald et al. (2004) took two directions. If resources are randomly

distributed through a home range, an increase in group size demands an increase in

range size to retain an adequate resource base. If resources are not randomly

dispersed, home range size will not be related to group size (Macdonald et al. 2004)

but is more likely to be related to resource abundance within the optimal space.

Concern regarding why some home ranges become smaller as group size increases

was raised, though it was unclear whether Macdonald et al. (2004) were considering

the home range to be the expected size for the calculated energy required for the

group or the area traversed daily by the group.

The area traversed daily by the group will be different to the expected or required

home range size because the strength of neighbouring groups and their area traversed

daily will limit opportunities for range expansion. Mech (1970) stated that prey

resources for wolves, such as deer herds, were generally found in the interstices of

territories. Alternatively, in arid Australia, neighbouring dingo groups have been

reported to howl as they approach a water source to minimise conflicts between packs

(Corbett 2001). Therefore, wolves share resources with neighbouring wolves and

dingoes share resources with neighbouring dingoes in territory interstices. If

resources within interstices are rich and, by design of territories, shared by

neighbouring groups, then the areas traversed daily by each group must contain

optimal resources for that group. The canids in question should therefore be able to


182

increase the size of their pack within their home range until conflict between

conspecifics becomes intolerable.

Social structure in carnivores forms by tolerating or not tolerating conspecifics

(Laundré and Keller 1981; Creel and Macdonald 1995; Gese et al. 1996).

Mechanisms for tolerance are related to prey availability, dispersal opportunities

(space), strength in numbers, vulnerability to predation and/or reproductive success.

Benefits of group living generally include cooperative breeding, increased hunting

success and protection of kills, territory and young from interspecific competitors

(Macdonald et al. 2004). Disadvantages of group living are intraspecific competition

and reproductive suppression (Macdonald et al. 2004). This appears to be why the

social systems of the subfamily Caninae affect their patterns of movement within and

outside their home range.

4.1.2 Effect of social systems on movement patterns

Stemming from the benefits of group living, one can expect movement patterns of

hypercarnivorous canids to revolve around foraging, maintenance of territory,

breeding and rearing young. Alternatively, if intrapack competition is so great that

the costs of remaining with a group outweigh the benefits to fitness of dispersal,

establishing a new range or joining a neighbouring group, analyses of movement

patterns will reveal changes in sociality. Perturbations in social systems of dingoes

have previously been observed by Thomson et al. (1992a) for dingoes in Western

Australia. Macdonald et al. (2004) discussed dispersal in a similar sense, whereby

benefits and disadvantages to fitness varied with circumstances per site. A resource

saturated site, potentially similar to the SGBMWHA where there are no limits to


183

water or food may facilitate establishment of larger social groups and delayed

dispersal (Messier and Barrette 1982). Alternatively, a resource saturated site may

produce more closely related groups in discrete territories over a broader landscape,

similar to results presented in section 2.3.2.3. When populations or groups exceed the

maximum threshold for the optimal space, intraspecific conflict will dictate death,

dispersal or changes in social structure.

In canids with mean live weight >13kg, a sex-ratio biased towards males can be

expected because males use less resources (do not bear young) and provide more

assistance in territory defence and hunting (Moehlman 1986; Macdonald and Sillero-

Zubiri 2004b). Consequently, emigration of females is hypothesised to occur more

frequently since they do bear young and place additional stress on the social system

(such as require helpers) and resources. Results presented in chapter two of the

current study showed that the mean live weight of the dingoes sampled from the

SGBMWHA population was 16kg and the sex ratio of captures was male biased at

1.9: 1 males: female (c.f. footnote 3, p. 54). This is the same ratio observed by

Sillero-Zubiri et al. (2004) for Ethiopian wolves which generally weigh between 11kg

and 19kg. Therefore, dispersal and transience in this population can be expected to be

sex biased towards females. Macdonald et al. (2004) however could not corroborate

sex bias in dispersal of individuals from large canid populations. Moehlman (1986;

1989) discussed dispersal as though it was a choice, between disperse and try to

survive or stay and be killed, and suggested dispersal largely depended on resource

availability.


184

4.1.3 Activity, movement, dispersal and home range of dingoes

Studies on dingo movement patterns and home range determination have been

conducted using VHF telemetry (Harden 1985; Thomson 1992b; d; Thomson et al.

1992a; b; Corbett 2001) and more recently, GPS telemetry (Allen 2006). Research on

dingo movement patterns was introduced in chapter one, though studies assessing

space utilization by dingoes are scarce. Short term movements of nine dingoes in an

urban Queensland environment using GPS technology showed mean home range was

18.14km2 (± 11.12 se) between October 2005 and June 2006 (Allen 2006). Studies in

mountainous regions of southern New South Wales reported mean home range of

“wild dogs” was 97.8km2 (Claridge and Mills, unpublished data). These data are

comparable with a preliminary investigation by Purcell et al. (2006; section 4.3.1 of

this thesis), where the mean home range for three dingoes (excluding a fourth dingo

which travelled extraterritorially) in the SGBMWHA was 49.5km2

(± 16.4 se).

Home range, patterns of movement, activity and patterns in observed mortality of

dingoes in the SGBMWHA are described in this chapter. Differences between

biological seasons are a primary focus so that changes in the functional ecology of

dingoes in this ecosystem throughout the year can be discussed and compared with

seasonal changes in diet and abundance shown in chapter three.

4.2 Methods

Dingoes were captured as described in chapter one. Six GPS data logging collars

programmed to store data on seasonal movements were deployed during breeding

season (April) 2005 and 2006 (n = 12 over two years). GPS receivers were mounted


185

on leather collars fitted with automatic, time-release drop-off mechanisms (Sirtrack®

,

Havelock North, New Zealand). A VHF transmitter was also built in to aid recovery

of collars. The final weight of each collar was 350 grams and GPS data were

recorded using datum WGS-84 on a Trimble GPS.

Collars were programmed to store one waypoint on the hour for three hours, followed

by an eight hour interval to obtain a cyclic data set. Eight hour cycles were chosen to

promote statistical independence of points for home range analyses while obtaining

data for different times of the day. Three hourly waypoints enhanced the probability

that at least one location was obtained each sampling period, enabled overlap for

analyses of movement patterns and conserved battery power so that GPS locations

could be recorded for a period of at least one year. The automatic release mechanism

was programmed to release 13 months after deployment. Annual GPS collars were

deployed on five adult males, four adult females, one juvenile male and two juvenile

female dingoes. Collar retrieval was expedited using a fixed wing aircraft with wing

mounted Telonics Receiving Antennas (Aircraft) connected to a Telonics TR-5

telemetry scanning receiver with TAC-2-RLB Antenna Control Unit, and a helicopter

using a three element hand-held Yagi antenna and Telonics TR-4 receiver to obtain

general locations of the collars. Located collars were recovered from the study site on

foot by search teams with hand held Yagi antennas and TR-4 receivers.

VHF collars were also deployed during the 2005 and 2006 trapping events to monitor

dispersal and mortality through resighting methods. Animals that were not being

tracked with VHF or GPS telemetry collars were collared with a 2cm wide coloured

nylon domestic dog collar with metal buckles. Excess nylon was fastened to the

collar with strong tape. If a captured dingo had specific individual coat colours it was


186

considered marked and a collar was not outfitted. Resighting methods included ad

hoc observations and photographs using TrailMAC motion-sensing digital camera

traps described in chapter one.

During breeding season (March) 2007, five GPS collars programmed to collect data

on short term movements over 52 days were deployed. These collars were set to store

one location data point every ten minutes. Attempts were made to deploy GPS data

logging collars fitted with Argos satellite transmitting technology10

and VHF

transmitters for a three year period at the same time, however, there were

manufacturing difficulties and the trip was rescheduled to whelping (June) 2007. Five

collars were deployed during whelping 2007 though the technology failed and these

collars did not transmit or log any data and became basic VHF collars. Three

GPS/Argos hybrid collars were successfully deployed at the start of rearing season

(September) 2007. Although data from the eight individuals collared with GPS/Argos

collars are not included in this chapter, ancillary data such as their relatedness to other

dingoes was. Efforts to spread the GPS collars across the start, middle and end of

each transect were made during trapping programs to increase the chance of

deploying GPS collars in different packs.

Resighting data from field observations and images from passive camera traps

revealed patterns of movement (including dispersal and extraterritorial forays) and

survival. Records were included in profiles for encounter histories with known

individuals and to identify structure of social units. Data on seasonality of sightings

and group movements were obtained by counting the occasions when a group of

dingoes were encountered. Interpretation of lone dingo sightings is less likely to be

10 Collars transmit data via satellite which can be downloaded using the internet.


187

reliable because there was no certainty that the passive camera traps would

photograph all dingoes travelling as a group. Similarly, observations of individual

dingoes, when driving along transects, could not be reliably used as an indicator of

pack movements.

4.2.1 Home range

In this study, the term home range was defined as the area traversed by an animal in

its daily movements (Burt 1943) over a defined period of time. A core area of activity

is defined as a selected area which contains home site, refuges and dependable food

sources (Burt 1943). Only GPS data logging collars for the seasonal and fine scale

movements were used to determine the home range of individuals. To obtain seasonal

data sets on home range, GPS collars were programmed to log one GPS point on the

hour for three hours every eight hours for a cyclic pattern of locations over a period of

12 to 13 months. This was done to minimise bias of night and day-time locations.

Based on the justification presented by Allen (2006) regarding the accuracy of the

GPS data logging collars he deployed in Queensland, all points stored after

deployment were included in major analyses. GPS data with Horizontal Dilution of

Precision (HDOP) values greater than three would have been removed from major

analyses if research questions were concerned with habitat preference. The HDOP

value is a measure of the quality of the GPS coordinates between satellites and values

<4 approximately represent a 10m error circumference of the location on the ground

(Allen 2006; Sargisson, pers. comm., Sirtrack®, 2008). HDOP values present an

estimate of horizontal (2-Dimensional) accuracy which is more applicable when using

2D mapping programs such as those in this study. Graves and Waller (2006) provided


188

a comprehensive review of GPS telemetry systems and indicated that the hypotheses

being tested will dictate the detail of the analyses regarding error and bias of GPS

coordinates. D’Eon et al. (2002) researched GPS radiotelemetry error in mountainous

terrain to test general positional dilution of precision (PDOP) fixes. Coelho et al.

(2007) reported the success of obtaining accurate 2D and 3D DOP values without

specifying whether the DOP was positional, vertical or horizontal. The use of 2D and

3D data indicated they assessed the accuracy of PDOP fixes on GPS collared maned

wolves Chrysocyon brachyurus. PDOP values are combined values from vertical

dilution of precision (VDOP) with HDOP. VDOP values are inherently inaccurate

because the occurrence of satellites, or a satellite, in the vertical plane (directly

overhead) is relatively low (Sargisson, pers. comm., Sirtrack®, 2008). Since this

study was concerned only with patterns of movement and not with habitat selection or

habitat use, GPS error observed in HDOP values did not require detailed analysis.

The accuracy of the GPS collars is presented in Table 4.1. GPS locations were so

numerous and accurate that independence of points and autocorrelation of data did not

require validation for results. Application of mathematical analyses increased the

error rate in the data and only seemed necessary for comparison with past studies.

Fixed kernel home range estimates (90% and 50%) and minimum convex polygon

(MCP) home range estimates (100%, 95% and 50%) were calculated using the

XTools Pro Extension software (Data East, LLC, 2005) and the Hawth’s Tools

Extension in ArcView® Geographic Information System (GIS). Due to the accuracy

of the GPS data, raw points were also mapped for comparison of analyses.


189

Table 4.1: Summary of GPS data precision from dingoes collared with annual GPS collars (Table A) and dingoes collared with short term GPS collars (Table B) Table A Thirteen month collars Site.Dingo 1.2 1.4 1.7 2.1 2.2 2.4 2.10 Age at capture J A A J J J A Sex ♂ ♀ ♂ ♀ ♀ ♂ ♀ GPS days 136 403 178 337 130 384 313 Total fixes (expected fixes)

802 (890)

2238 (2638)

857 (1165)

2060 (2206)

834 (850)

2302 (2513)

1669 (2049)

Proportion expected fixes 0.90 0.85 0.74 0.93 0.98 0.92 0.81 Mean fixes/day (range f/d)

5.89 (6-8)

5.55 (6-8)

4.81 (6-8)

6.11 (6-8)

6.42 (6-8)

5.99 (6-8)

5.33 (6-8)

Mean HDOP 3.3 3.4 3.7 3.2 2.8 3.0 3.2 Median HDOP 2.7 2.8 2.7 2.6 2.3 2.4 2.3 Fixes HDOP<3 512 1328 520 1314 602 1585 1159 Proportion HDOP<3 0.64 0.59 0.61 0.64 0.72 0.69 0.69 Fixes HDOP>3 290 910 337 746 232 717 510 Proportion HDOP>3 0.36 0.41 0.39 0.36 0.28 0.31 0.31 Fixes HDOP>10 22 43 51 33 12 31 87 Proportion HDOP>10 0.03 0.02 0.06 0.02 0.01 0.01 0.05

Table B Short term collars (52-54 days during breeding season) Site.Dingo 2.20 2.21 2.22* 2.23 2.24 2.25 Age at capture A A A J J A Sex ♂ ♂ ♂ ♂ ♀ ♂ GPS days 54 54 2 54 52 52 Total fixes (expected fixes)

7350 (7776)

7534 (7488)

108 (288)

7524 (7488)

7199 (7776)

6978 (7776)

Proportion expected fixes 0.95 0.97 0.38 0.97 0.96 0.93 Mean fixes/day (range f/d)

136.11 (144)

139.51 (144)

54 (144)

139.33 (144)

138.44 (144)

134.19 (144)

Mean HDOP 3.9 3.7 5.1 3.2 3.6 3.1 Median HDOP 2.9 2.7 3.7 2.6 2.6 2.5 Fixes HDOP<3 4009 4450 43 4757 4330 4578 Proportion HDOP<3 0.55 0.59 0.40 0.63 0.60 0.66 Fixes HDOP>3 3341 3084 65 2767 2869 2400 Proportion HDOP>3 0.45 0.41 0.60 0.37 0.40 0.34 Fixes HDOP>10 481 419 16 114 382 101 Proportion HDOP>10 0.07 0.06 0.15 0.02 0.05 0.01

* Deceased individual excluded from further analyses and discussion.

4.2.2 Patterns of movement

GPS and VHF collars and resighting data were used to determine movement patterns

for individuals and groups. Movement patterns assessed were seasonal variations in

home range size and core area size, territory maintenance, monthly distribution of

GPS locations around 50% MCP core areas, extraterritorial forays and dispersal.


190

Seasons were based around those specified in chapters one and three for breeding,

whelping, rearing and exploratory movements.

4.2.3 Patterns of activity

Activity i s generally defined as the rate of movement between successive locations.

Patterns of activity were subsequently assessed as means from speed travelled in

metres/hour (m/hr) and distance moved between successive GPS locations. Mean

speed and distance between points of annual GPS collars were calculated per month.

Activity of individuals derived from GPS monitoring was compared with activity

derived from sand plot data to assess whether the sand plot activity indices reflected

real activity patterns.

To classify activity patterns from short term collar data, “resting” movements were

defined as those <199m/hr, “walking” were movements between 200-1999m/hr,

“trotting” movements were those between 2000-5999m/hr and speeds >6000m/hr

were classed as “running”. Mean speed (m/hr) of movement was subsequently

calculated every ten-minutes; for example between 01:00-01:09 and 01:10-01:19, and

every hour such as 01:00-01:59. Line graphs of time versus speed were created for

each day and each dingo wearing short term GPS collars. The speed axis was

standardised to the maximum speed observed for each dingo for easier comparison of

activity patterns between consecutive days. Pie graphs were used to compare

proportions of time spent resting with time spent walking, trotting or running per

dingo for the duration of collar deployment and on separate days when large

variations in activity were observed. One male from each of the three adjacent packs

was selected to assess variation between packs. Their activity was divided into four


191

categories (majority resting, low activity, crepuscular activity and high activity) and

graphed for visual comparison.

4.2.4 Observations of mortality and dispersal

Extraterritorial forays, dispersal and mortality were monitored and recorded per

month and season. Observations of dingo groups and pregnant bitches were included

in this section to note any trends that may have provoked the dispersal or death of

individuals.

4.3 Results

Twelve dingoes (three females, nine males) were captured in April 2005 and 18

dingoes (seven females, eleven males) were captured in April 2006 (c.f. Table 2.13 for

data on each individual). In Site 2 during 2007, eight males and one female were

trapped in March, two males and three females were trapped in June and one male and

two females were trapped in September. Twelve GPS collars (Site 1: two males and

one female in 2005 and three males in 2006; Site 2: two females and one male in

2005; three females in 2006) were outfitted to dingoes for analysis of annual

movement patterns. Five GPS collars were deployed on four males and one female

during March 2007 for analysis of fine scale movement patterns in Site 2. Of the

remaining captures, eight were collared with a coloured domestic dog collar for

resighting data and four were not collared because it was unnecessary during the latter

stages of this study. Ten VHF collars (three per site during 2005 and two per site

during 2006) were deployed to test for patterns of dispersal and site fidelity.


192

Five of the six 2005 seasonal GPS collars were recovered and two of the six 2006

seasonal GPS collars were recovered. The time release mechanism of the sixth 2005

collar failed to release and this individual (dingo 1.1) was last tracked within

relatively close proximity to his area of capture during 2007. Of the 2006 GPS data

logging collars, dingoes 1.10 and 2.7 could not be found during aerial telemetry

surveys. Dingo 1.8 was found with a mortality signal during one fixed wing aerial

survey but not on a second or third aerial survey with a helicopter and a fixed wing

aircraft respectively. The time release mechanism on dingo 2.8 from 2006 captures

failed to release. This female was found outside of the area where she was commonly

encountered during one fixed wing aerial survey and has not been tracked since.

One of the three 2005 recoveries from Site 2 malfunctioned and released five months

after deployment. The second was found within the core area of dingo 2.4 after

extensive explorations east and west of his core area during the 2006 breeding season.

The third collar recovered from Site 2 was found within its recorded range but had

ceased logging data two months earlier than expected. One of the two 2005

recoveries from Site 1 had torn between the time release mechanism and the GPS

store on board unit during September 2005. This collar ceased logging points in

November 2005 after deployment in April 2005. Reasons for the malfunction and the

tear could not be determined but the dingo was presumed dead due to the severity of

the tear. The collar fitted to dingo 1.4 stored data until it released in June 2006

although it was programmed for release in May 2006. Although this collar appeared

to be undamaged, water had infiltrated the data logger via hair line cracks in the

plastic casing. Of the 2006 GPS collars recovered, one was recovered after dingo 1.7

had died within his range. The second, fitted to dingo 2.10, was found within five

metres of the short term GPS collar deployed on dingo 2.24 who was genetically


193

related to dingo 2.10. Their territories overlapped and both animals were often

observed together and in a pack including dingoes 2.12, 2.14, 2.26 and 2.25.

In addition to dingo 2.24, the four other short term GPS collars were recovered. The

collar from dingo 2.23 was recovered within his range. Data recovered from the

collar worn by dingo 2.25 showed range overlap with dingoes 2.10 and 2.24.

However, this collar released at the end of an extraterritorial foray by dingo 2.25

during the breeding season and was recovered from property bordering the protected

area. The property owners indicated they had not had any problems with dingoes or

wild dogs for some time. The ranges of the fourth (dingo 2.20) and fifth (dingo 2.21)

also overlapped and were recovered within 50m of each other in a gully within their

50% core area of activity.

Of the dingoes with VHF collars deployed in Site 1 during 2005, dingo 1.3 died near

his point of capture. Dingo 1.5 left Site 1 and was tracked on four occasions in gullies

below the urban town of Bullaburra on the northern side of the Cox’s River arm of

Lake Burragorang, 24km northeast of his capture location. Dingo 1.6 was last tracked

during 2007 within Site 1, showing site fidelity. Of the two with VHF collars

deployed in Site 1 during 2006, dingo 1.9 was genetically related to seven dingoes

captured in Site 2 (Figure 2.9), and was tracked via aerial telemetry on three occasions

between August and November 2006 relatively close to his point of capture in Site 1.

This suggested that he was foraging extraterritorially or dispersing when captured.

The second juvenile male (dingo 1.11) had been seen by park rangers 26km west from

his point of capture. This individual was found deceased within two weeks of the

sighting.


194

Two of the three VHF collars deployed in Site 2 during 2005 began to malfunction

during 2006. Dingo 2.5 outfitted with the third VHF collar during 2005 was

recaptured in 2006 and was tracked during 2007 and 2008 within Site 2. An adult

male (dingo 2.11) and a juvenile female (dingo 2.9) were captured and outfitted with

VHF collars during 2006. Dingo 2.9 was genetically related to other dingoes caught

within the vicinity and was frequently tracked up to 2007. Dingo 2.11 was genetically

related to dingoes from Site 1 and found deceased in October 2006 close to his

capture location. Encounter histories and observations made between April 2005 and

July 2008 for all captures and three frequently observed non-captures are shown in

Figure 4.1.


195

Figure 4.1: Encounter history with known dingoes from April 2005 until July 2008. Ã = Age; A = adult; J = Juvenile; Straight lines represent captured dingoes collared with VHF transmitters (including GPS and ARGOS with VHF transmitters) and start in the month the dingo was tr apped and finish in the month the collar failed or dropped off; Broken lines start and end during months of first and last encounters with identified captures and non-captures, based on resighting information including all records from passive camera traps; Δ = Alive when last encountered but fate is unknown; Colours indicate genetic relatedness at K = 8 (Genetic relatedness data for dingoes BW, Gn and Mk are unknown); C = Collar failed expected deployment; C1 = VHF failed; C2 = GPS failed; C3 = ARGOS failed; Ct = Time release failed; C? = Potential failure of GPS and/or VHF; X? = Probable death; X = Known death; D = Known dispersal; D? = Probable dispersal; E = Encountered outside of known range (extraterritorially); R = Recaptured; #, ^, $, *, ~, &, %, < = Dingoes which were observed together, caught or observed within 1km of each other on the same day; L = dingo was observed alone but does not include lone photographs from passive camera traps; > = Alive and part of long term ARGOS research; Pa = dingo was observed in a pack; Pā = dingo was observed in a pack with pups; α = Dominant during encounter; H? = Possible helper (rank >β); β = Subordinate during encounter; Pr = Dingo was pregnant when encountered; X1 = Dingo was not included in the mortality tally because his death was due to capture myopathy; BW = Black and white dingo that was never captured but observed for the duration of the study; Gn and Mk = Known individuals occasionally observed together, with pups, alone and with dingo 2.2.


196

4.3.1 Home range

Seven of twelve annual GPS collars (58.3%) were recovered. All five fine scale

collars were recovered. Two of the seven annual collars dropped off 136 and 130

days after deployment. Another two ceased storing location data at days 337 and 313.

Of the remaining three, one logged data 384 days after deployment, one logged data

403 days after deployment and the third stored data points until recovered from the

deceased individual, 178 days after deployment. In all, 10,762 GPS locations for

seven dingoes were used in analyses of annual movement data. Of the short term

collars, 36,58511

GPS locations for five dingoes were used in analyses of home range

and movements during the 2007 breeding season.

The HDOP for dingoes with annual collars in Site 1 was significantly different from

dingoes with annual collars in Site 2 (F1, 5 = 8.15, P = 0.04). There was no difference

between HDOP for annual collars and short term collars independent of site (F1, 10 =

2.5, P = 0.14). However, there was a difference in HDOP between annual collars and

short term collars in Site 2 (F1, 7

= 5.7, P = 0.05).

GPS locations are shown on an orthophoto with an image of the collared individual

(Figure 4.2). Mean home range size for 95% MCP was 37.7km2 (± 8.8 se) and

34.2km2 (± 8.2 se) for 90% kernel contours. Mean size for 50% core areas were

7.9km2 (± 1.8 se) using 50% MCP and 5.9km2

11 Locations for dingo 2.22 have been excluded from analyses because he died from capture myopathy and location and home range data will skew results due to bias from HDOP values resulting from a downward facing antenna.

(± 1.4 se) for 50% kernel contours (n =

47,347 locations from 12 dingoes).


197

Figure 4.2: Separate maps of total data points from retrieved GPS collars for each dingo and all dingoes on one map (continued overleaf).


198

Figure 4.2 (Cont.): Separate maps of total data points from retrieved GPS collars for each dingo, and all GPS collared dingoes on one map.


199

As reported by Craighead et al. (1973) and Harden (1985), mapping raw data was

more representative of the movement of an animal compared with mathematical

analyses of data. Figures 4.3 and 4.4, for instance, show dingoes which have a

dispersed range due to extraterritorial forays within breeding seasons of 2006 and

2007 respectively.

Figure 4.3: Comparison between 100% MCP (outer red line) 95% MCP (middle red line), and 50% MCP (inner red line) with 90%-50% kernel contours (green lines) and raw data points (blue dots) for male dingo 2.4 during breeding season 2006. Pink shading represents Lake Burragorang and associated tributaries. See section 5.3.2 for details on his breeding season forays.

a. b. Figure 4.4: Home range of dingo 2.25 including (a.) and excluding (b.) extraterritorial movements. Both figures show 100% MCP (outer blue line), 95% MCP (middle blue line) and 50% MCP (inner blue line) with 90%-50% kernel contours (grey lines) and raw data points (yellow dots) for male dingo 2.25 during breeding season 2007. Pink shading represents the southern most section of Lake Burragorang (also visible in Figure 4.3) and associated tributaries.


200

The assumption of normal distribution for annual mean home range was checked

using P-P plot and homogeneity of variance using Levene’s test of equality of error

variances. Annual mean home range did not differ between years (F2, 9 = 2.4, P =

0.15), sex (F1, 10 = 0.2, P = 0.65) or sites (F1, 5 = 0.2, P = 0.68) for all of the dingoes

tested. Data for males and females were, therefore, combined per site for further

analysis. No statistical differences were observed between 95% MCP home range

size for annual and short term collars (F1, 10

= 3.0, P = 0.11).

Based on the density and accuracy of the GPS locations, mathematical home range

estimation was performed for comparison with past studies and between individuals

within this study. Home range estimations using 95% and 50% MCP or 90% and

50% kernel contours are shown in Table 4.2. The shape of each MCP and kernel

contour can be seen in Figure 4.5 (A. 2005 collars; B. 2006 collars; C. 2007 [short

term] collars). With few exceptions, boundaries were either clearly demarcated or

wholly overlapped; the latter representing individuals from the same packs (cf. Figure

5.1 for relatedness of individuals).

Table 4.2: Comparison of 95% MCP with 90% kernel contours and 50% MCP with 50% kernel contours for 12 dingoes in the SGBMWHA (cf. Tables 2a and 2b for technical details).

Site.Dingo Sex MCP 95 MCP 50 Kernel 90 Kernel 50 Annual collars 1.2 ♂ 49.99 9.35 48.64 9.71

1.4 ♀ 75.00 18.17 60.19 16.15 1.7 ♂ 42.92 8.79 49.92 9.59 2.1 ♀ 45.71 8.65 30.40 6.58 2.2 ♀ 13.28 6.17 12.91 2.84 2.4 ♂ = 103.80 12.75 86.93 9.38 2.10 ♀ 15.75 3.13 12.20 3.30

Short term collars 2.20 ♂ 14.76 1.63 11.78 1.10 2.21 ♂ 13.08 1.57 10.76 0.96 2.23 ♂ 6.67 1.37 4.00 0.89 2.24 ♀ 13.21 4.14 10.45 2.12 2.25 ̂ ♂ 57.70 18.91 72.00 8.29

37.7 7.9 34.2 5.9

se 8.8 1.8 8.2 1.4

^ Data includes foray; = Excludes breeding season foray (outside of 95% MCP)


201

a. b.

c. d. e.

Figure 4.5: Comparison of 95% Minimum Convex Polygons (a.) with 90% kernel contours (b.) for all dingoes (c. 2005 collars; d. 2006 collars; e. 2007 (short term) collars in relation to Lake Burragorang and associated tributaries). ▬ = Dingo 1.2; ▬ = Dingo 1.4; ▬ = Dingo 1.7; ▬ = Dingo 2.1; ▬▬▬ = Dingo 2.2; ▬ = Dingo 2.4; ▬ = Dingo 2.10; ▬ = Dingo 2.20; ▬ = Dingo 2.21; ▬ = Dingo 2.23; ▬▬▬ = Dingo 2.24; ▬ = Dingo 2.25.

The overlapping ranges for dingoes 2.20 and 2.21, and dingoes 2.10, 2.24 and 2.25,

indicated that apart from extraterritorial explorations, the ranges for dingoes 1.2, 1.4,

1.7 and 2.4 were representative of dingo groups in those areas. Observations of dingo


202

2.2 with adults and pups supported this though dingo 2.1 did not follow the same

pattern. She was only observed alone and her range was within the interstices of

dingoes 2.2, 2.4 and 1.7. Based on these data, she may be considered a loner for the

duration of the current study and analysis of variance suggested her 50% fixed kernel

home range was significantly different than other females from Site 2 (F1, 2 = 31.09, P

= 0.03). Excluding the range of dingo 2.1 from female home range data for

comparison between sites showed the home range of dingo 1.4 was significantly

different to that of female dingoes 2.2, 2.10 and 2.24 (F1, 2 = 1331.99, P = 0.001).

There was less difference, however, when including dingo 2.1 (F1, 3 = 8.9, P = 0.06).

Such difference in home range between individuals and sites was consistent with

previously presented data suggesting that home range is related to resource abundance

per site or the social status of individuals and every site and dingo should be assessed

individually. Recent data (Mulley and Purcell, unpublished) have shown that the

range of dingo 2.1 during 2005 is similar to the range of female dingo 2.35, collared

with an Argos satellite collar during 200712

. Tissue samples from dingoes 2.1 and

2.35, however, showed that they were genetically unrelated.

Distinct spatial separation of 50% core areas of activity for most dingoes was evident

(Figure 4.6) for those outfitted with both annual and short term GPS collars. These

rarely overlapped and may only do so due to the analysis techniques or changes in the

distributions of packs. This may also be the case in some circumstances of 90% or

95% home range estimates. Home range areas were largely determined by

topographical features such as ridgelines and rivers demarcating territory borders.

12 Argos home range data has been excluded from the present study because it is part of a longer term study (three years).


203

Figure 4.6: Spatial separation of 50% MCP core areas of activity and their relationship with topography, Lake Burragorang and associated tributaries (pink shading/ lines). ▬ = Dingo 1.2; ▬ = Dingo 1.4; ▬ = Dingo 1.7; ▬ = Dingo 2.1; ▬▬▬ = Dingo 2.2; ▬ = Dingo 2.4; ▬ = Dingo 2.10; ▬ = Dingo 2.20 (mostly behind dingo 2.21); ▬ = Dingo 2.21; ▬ = Dingo 2.23; ▬▬▬ = Dingo 2.24; ▬ = Dingo 2.25. NB: Dingoes 2.1 and 2.2 show overlap in 2005 however dingo 2.23 was collared for the short term study in 2007 and may not be representative of 2005 home range data.

4.3.2 Patterns of movement

For the dingoes outfitted with a 13 month GPS collar, changes in the size of home

ranges using 100% and 50% MCP data were analysed by biological season (Table

4.3). Mean 100% MCP home range for whelping season was 41.2km2 (± 8.2 se; 50%


204

MCP = 6.9 ± 2.5km2 se; n = 7), rearing season was 37.6km2 (± 11.7 se; 50% MCP =

7.4 ± 3.3km2 se; n = 4), exploratory season was 47km2 (± 11.6 se; 50% MCP = 9.3 ±

2.5km2 se; n = 4) and breeding season was 93.1km2 (± 53.2 se; 50% MCP = 11.7 ±

3.7km2 se; n = 7). Excluding extraterritorial movements of dingo 2.4 during the

breeding season, because this was an outlying event, the mean 100% MCP from

March-May for dingoes of the SGBMWHA was 40.7 km2 (±10.3 se). No differences

in mean home range size were observed between males and females during whelping

(100% MCP: F1, 5 = 0.76, P = 0.423; 50% MCP: F1, 5 = 0.31, P = 0.87) and breeding

seasons (100% MCP: F1, 5 = 2.07, P = 0.21; 50% MCP: F1, 5 = 4.325, p = 0.092)

although 50% MCP data between males and females were approaching significance

during the breeding season. Gender differences were not tested for rearing and

exploratory seasons due to insufficient data for males. No differences were observed

between seasons for Sites 1 and 2 combined (100% MCP: F3, 18 = 0.614, P = 0.615;

50% MCP: F3, 18

= 0.532, P = 0.67). Table 4.3 and Figure 4.7 show 50% core areas

usually increased after whelping into the rearing, exploratory and breeding seasons

with the exception of dingo 1.4.

Table 4.3: Comparison of 100% MCP area (km2) with 50% MCP area (km2

Dingo

) during whelping (Wh.), rearing (Re.), exploratory (Ex.) and breeding (Br.) biological seasons from dingoes outfitted with annual GPS collars.

1.2 ♂ 1.4 ♀ 1.7 ♂ 2.1 ♀ 2.2 ♀ 2.4 ♂ 2.10 ♀ MCP 100 50 100 50 100 50 100 50 100 50 100 50 100 50 Wh. 40.4 6.3 71.9 20.9 46.5 8.3 38.4 3.5 14.3 3.9 61.9 4.7 14.6 1.3 Re. - - 69.7 17.1 - - 32.8 4.3 - - 34.5 5.1 13.5 3.0 Ex. - - 69.7 15.5 - - 54.0 9.2 - - 49.5 9.2 14.8 3.5 Br. 57.4 29.8 69.9 14.4 58.2 9.0 35.7 5.5 9.2 1.6 407.8 18.0 13.9 3.9


205

a.

0

10

20

30

40

50

60

70

80km

2

b.

0

10

20

30

40

50

60

70

80

1.2 1.4 1.7 2.1 2.2 2.4 2.10

Dingo

km2

< (407.76)

Figure 4.7: Comparison of 50% MCP home range estimates (a.) with 100% MCP home range estimates (b.) for seasonal differences ( Whelping; Rearing; Exploratory ; Breeding). ^ = 100% MCP was 407.76km2

but the column was reduced so this graph was not distorted.

For males, extraterritorial movements were only detected with GPS collar data during

the breeding season. Range expansion during the breeding season was observed for

female dingo 2.1 which overlapped neighbouring territories but not for other females

monitored. However, the movement data for the females with GPS collars that were


206

not recovered (cf. beginning of results section 4.3) showed that females do travel

extraterritorially, though during whelping and rearing seasons, and not the breeding

season like males. Figure 4.8 shows how movements become consolidated within the

50% MCP core area during whelping season (June, July and August) for males and

females. It is apparent from these graphs that whelping for female dingoes 1.4 and

2.10 was likely to have occurred in June because movements, as represented by data

points, outside of the 50% MCP were minimal. Based on a 63 day gestation period,

they probably conceived in late March or early April. In contrast, females associated

with male dingo 2.4 appear to have whelped in August 2005. Data points from dingo

2.4 suggest he avoided the mean 50% MCP core area from March to May when

extraterritorial explorations increased, consistent with the breeding season. The data

in Figure 4.8 are comparable with data presented on activity and abundance in chapter

three. Dingo activity on sand plots was lower between June and September in 2005

and 2006 when dingoes 1.4, 2.4 and 2.10 appeared to be restrained by whelping

activities. Extraterritorial movements for dingo 2.4 were extensive during the

breeding season (Mar.-May; Figure 4.8). On all forays, he returned to his 50% core

area although data suggested he did not travel through it. His March foray showed

that his extraterritorial movements followed topographic features such as cliffs,

gullies and watercourses. These landscape features coincided with interstices between

pack territories in Site 2 when he re-entered Site 2 (Figure 4.9).


207

a. b.

Figure 4.8: Monthly GPS locations (dots) around 50% MCP for female dingoes 1.4 (a.) and 2.10 (b.) and male dingo 2.4 (c.) which highlight movement patterns associated with whelping (June-Aug.), rearing (Sept.-Nov.), exploratory (Dec.-Feb.) and breeding (Mar.-May) seasons. Pink shading represents Lake Burragorang and associated tributaries.

c.


208

a.

b.

Figure 4.9: Representation of the potential movements of dingo 2.4 (b.) in comparison with how movements are presented using GIS (a.) during his extraterritorial foray from 15th-18th March 2006. This depicts how gullies and ridgelines were used during navigation through a landscape and how movement corridors may be created by territories of adjacent groups, providing one reason for dingo packs to partake in border patrol activities. Total distance travelled≈ 120km. S = General start /end of extraterritorial foray; Black triangles = direction of movement and locations of GPS fixes; Black circles = locations of GPS fixes; 50% MCP core areas = ▬ for Dingo 1.7; ▬ for Dingo 2.1; ▬▬▬ for Dingo 2.2; ▬ for Dingo 2.4; ▬ for Dingo 2.10; ▬ = Dingo 2.21.


209

Simulated data of movements by dingo 2.4 in Figure 4.9b are consistent with dispersal

movement patterns of dingo 2.25 shown in Figure 4.10. Data show dingo 2.25 based

forays around one area on the border of his 95% MCP before traversing through his

range and commencing dispersal. At the conclusion of the foray, data indicated dingo

2.25 once again found an area to base his forays from.

Figure 4.10: Extraterritorial movements of dingo 2.25 during breeding season 2007. ▬ and ▬▬▬ = Pre dispersal and post dispersal forays from a base; ▬ and ▬ = One day of movement; Black polygon represents the area he rested for six days; pink line represents tributary to Lake Burragorang.


210

Movement patterns of dingoes 2.20 and 2.21 provided a good example of border

patrol behaviour during breeding season 2007 (Figure 4.11). Both individuals

circumnavigated the border of their home range within two-three days. This was the

only movement event of its nature by dingoes 2.20 and 2.21 during the 54 days of

collar data and the event occurred on days 40-42 for each. Other border patrols

observed for this pair targeted sections of the border, similar to movement of dingo

2.21 on day 42.

Figure 4.11: Movements of dingoes 2.21 (a.) and 2.20 (b) around their 50% MCP core area (black circle) during an identified border patrol. Each of the three days are represented by a different colour (Day 1 = blue for 2.21; red for 2.20; Day 2 = green for 2.21; yellow for 2.20; Day 3 = yellow for 2.21; white for 2.20). Orange and grey dots represent GPS locations, pink line represents tributary to Lake Burragorang. NB: Dingo 2.21 had completed the circuit one day earlier than dingo 2.20 and proceeded to foray or re-patrol the northern border on the third day. This is also an example of a normal patrol of one segment of the home range where the individual returns directly to the core area opposed to circumnavigating the core area.

a.

b.

0 2

Kilometres


211

4.3.3 Patterns of activity

Comparison of sand plot activity data with activity patterns determined using mean

distance between GPS locations showed no relationship, with the exception of dingo

2.4 during the breeding season (Figure 4.12).

a.

0200400600800

1000120014001600

A M J J A S O N D J F M A M J J A S O N D J F

2005 2006 2007

Month / Year

Mea

n di

stan

ce (

m)

betw

een

GP

S lo

catio

ns

0

0.2

0.4

0.6

0.8

1

1.2

1.4

PA

I

b.

0200400600800

1000120014001600

A M J J A S O N D J F M A M J J A S O N D J F

2005 2006 2007

Month / Year

Mea

n di

stan

ce (

m)

betw

een

GP

S lo

catio

ns

0

0.2

0.4

0.6

0.8

1

1.2

1.4

PA

I

Figure 4.12: Comparison of sand plot activity (― Site 1; ― Site 2) with activity determined using distance travelled between GPS locations for Site 1 (a; green line = dingo 1.2; light blue line = dingo 1.4; brown line = dingo 1.7) and Site 2 (b; pink line = dingo 2.1; orange line = dingo 2.2; black line = dingo 2.4; red line = dingo 2.10) annual collars.

Monthly means of activity using speed (time/distance between points) or distance

between points showed March and April were the most active months in Site 1 and

Site 2 respectively (Figure 4.13). Analysis of variance between seasons for speed

travelled and distance between points were not significant. Between sex differences


212

were approaching significance for distance between points (F1, 24 = 2.91, P = 0.101)

but not for speed (F1, 24 = 1.278, P = 0.269). Data for males and females were tested

for differences in distances travelled between points and the analysis showed they

approached normal distribution after X2 was checked using P-P plot (F1, 24 = 3.84, P =

0.062). The combined means showed March was the month when speed or distances

travelled between points was highest. Following the peak in activity, May was

generally the month with least movement and movement slowly increased up until

February when it became reduced before the peak in March.

Figure 4.13: Mean monthly activity plots for all dingoes collared with annual GPS collars. 1a: Mean distance between GPS locations for Site 1 (n = 3); 1b: Mean distance between GPS locations for Site 2 (n = 4); 1c: Mean distance between GPS locations for all dingoes (n = 7); 2a: Mean speed (m/hr) between GPS locations for Site 1 (n = 3); 2b: Mean speed (m/hr) between GPS locations for Site 2 (n = 4); 2c: Mean speed (m/hr) between GPS locations for all dingoes (n = 7).


213

Figure 4.14 shows dingo movements were generally crepuscular during the 2007

breeding season, though there was more activity during daylight hours, from short

term collar data. Two males from adjacent groups (dingoes 2.23 and 2.25) behaved

differently to two males from the same group (dingoes 2.20 and 2.21). Female dingo

2.24 and male dingo 2.25 also behaved very differently although they were from the

same group and were genetically related. Movements of dingo 2.25 may be related to

a pre-dispersal pattern of activity. Daily activity generally began at 06:00hrs and

finished by 20:00hrs for all dingoes wearing short term collars. A period of less

activity was observed between 10:00hrs and 14:00hrs.

a. b.

0

200

400

600

800

1000

1200

0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00

Time

m/h

r

0

100

200

300

400

500

600

700

800

900

1000

0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00

Time

m/h

r

c.

0

100

200

300

400

500

600

700

800

900

1000

0:00 6:00 12:00 18:00

Time

m/h

r

23:50

Figure 4.14: Mean speed (± se) travelled by four male dingoes (black line = dingo 2.20; orange line = dingo 2.21; light blue line = dingo 2.23; green line = dingo 2.25) and one female dingo (pink line = dingo 2.24) every hour (a. individual mean; b. total mean) and every 10 minutes (c. total mean) between stored GPS locations in Site 2 between 23rd (dingoes 2.20, 2.21 and 2.23) and 25th (dingoes 2.24 and 2.25) April 2007 and 15th May 2007.


214

Short term movements for each individual varied between diurnal, nocturnal and

crepuscular activity. Hence, dingo activity in the SGBMWHA may therefore be

classed as cathemeral (irregular) though not to the strict definition of the word. An

overall crepuscular pattern was observed in all dingoes, with days of long rest periods

and days of increased activity also observed.

Dingoes 2.20 and 2.21

This genetically related male pair was more active than the males from adjacent

packs. Figure 4.15 shows the proportions of time spent at the various speeds for these

individuals. Dingo 2.20 was resting for approximately half of his time and reached

speeds classed as running only five times in 54 days. He was generally more active

during the daytime with a few days showing nocturnal activity (Days 8, 14 and 40).

Increased night time activity was also observed on days 14 and 40 for dingo 2.21.

There was a full moon on days 12 and 41 and these may be associated with nights of

increased activity on days 8, 14 and 40. The activity of 2.20 and 2.21 ranged from

resting almost all day with slight crepuscular peaks (day 31) to being active all day

and even into the night (day 40) (Figure 4.16). Throughout the 54 day observation

period, multi day periods of increased activity were frequent, although single days of

increased activity were more common for both males.

a. b.

Figure 4.15: Proportion of time spent resting (0-199m/hr; ■; a. = 51%; b. = 54%), walking (200-1999m/hr; □; a. 42%; b. 39%), trotting (2000-5999m/hr; ■; a. = 7%; b. = 7%) and running (>6000m/hr) for dingoes 2.20 (a.) and 2.21 (b.)


215

a. b.

0

1000

2000

3000

4000

5000

6000

7000

0:05

2:43

4:13

5:42

7:12

8:41

10:1

1

11:4

0

13:1

0

14:5

0

16:1

9

17:4

9

19:1

9

20:5

8

22:4

7

Time

Spe

ed (m

/hr)

0

10002000

3000

4000

50006000

7000

0:01

1:51

3:40

5:30

7:19

9:08

10:5

8

12:4

8

14:3

7

16:2

6

18:1

6

20:2

5

22:1

4

Time

Spe

ed (

m/h

r)

c. d.

01000200030004000500060007000

0:00

1:59

3:59

5:58

7:58

9:57

11:5

7

14:0

7

16:0

5

18:0

5

20:0

4

22:0

4

Time

Spe

ed (

m/h

r)

01000200030004000500060007000

0:07

2:07

4:06

6:06

8:05

10:0

4

12:0

3

14:0

3

16:0

2

18:0

2

20:0

1

22:0

1

Time

Spe

ed (

m/h

r)

Figure 4.16: Activity patterns on day 31 (top) of the study showing minimal activity and day 40 of (bottom) of the study showing increased activity for dingoes 2.20 (a. and c.) and 2.21 (b. and d.).

Corresponding with the increased activity observed on the 40th

day of the study was

an apparent border patrol by both individuals (Figure 4.11). The approximate

distance travelled on the first day was 16.97km followed by 11.2km on the second.

The average speed was 810m/hr with a maximum speed of 9300m/hr and 7400m/hr

on the first and second day respectively. For dingo 2.20 the patrol lasted three days

but for dingo 2.21 the patrol only lasted two days. On the third day, dingo 2.21

travelled back to the northern border before returning to his core area. This is an

example of other observed border patrols where only a segment of the border was

traversed, opposed to the circumnavigation observed in Figure 4.11.

Dingo 2.23

This individual was less active, with 66% of his time spent at less than 200m/hr and

only 2% at over 2000m/hr (Figure 4.17).


216

a. b.

0

1000

2000

3000

4000

5000

6000

7000

0:02

1:42

4:51

6:30

8:10

9:49

11:3

9

13:1

9

14:5

8

16:3

7

18:1

7

19:5

7

21:4

6

23:2

5

Time

Spe

ed (

m/h

r)

Figure 4.17: Proportion of time spent resting (0-199m/hr; ■; 66%), walking (200-1999m/hr; □; 32%), trotting (2000-5999m/hr; ■; 2%) and running (>6000m/hr) (a.) and example of low activity exhibited by dingo 2.23 (b.).

The 90% fixed kernel estimate for dingo 2.23 showed that his home range was only

4km2, which was less than half the size of the other dingoes in this short term study

(range 10.45-11.78km2

, n = 4). The smaller home range may be the cause of GPS

data showing low activity patterns because the distance between points will also halve

in length. Aside from some increased nocturnal activity on four days, dingo 2.23 did

not have many days of increased activity and even had a sequence of nine consecutive

days of resting for more than 70% of the day. No patterns of movement throughout

the home range were observed for this animal for the duration of the study.

Dingo 2.24

Dingo 2.24 was the only female collared as part of the short term movements study.

Her movement patterns showed crepuscular trends, with two distinct peaks (cf. Figure

4.14). She also had more nocturnal movements than the male dingoes. Trends

showed nocturnal activities first occurred every five days, followed by eight days with

no nocturnal activity. Following this were two (in three) nights with increased

nocturnal activity, followed by another nine days without nocturnal activity and three

consecutive nights of nocturnal activity.


217

Dingo 2.25

This male was the least active of all dingoes in the short term study, resting 67% of

his time and only exceeded 6000m/hr once in the 52 days of monitoring. No obvious

overall trends of activity were observed, although diurnal patterns were prevalent.

Increased nocturnal activity was observed, but particularly during very early mornings

on days 13, 25, 26, 47, 48 and 52. Genetically related dingo 2.24, in comparison, was

more active at night on days 2, 8, 14, 18, 27, 29, 39, 40 and 41. Dingo 2.25 moved

out of his territory and rested for six days before travelling extraterritorially on day

38. Pace of travel then increased for four days before the dingo arrived at property

bordering the protected area. Most of his time was spent resting within 500m of a

residence until the time release mechanism triggered. Movements during these three

days generally revolved around a densely vegetated peak with the exception of a foray

on one night around the periphery of a paddock.

Movement patterns for dingoes 2.21, 2.23 and 2.25 showed substantial variations in

mean patterns of activity (Figure 4.18). Dingo 2.25 dispersed and dingo 2.21 was

observed circumnavigating his 95% home range.

0%

20%

40%

60%

80%

100%

2.21 2.23 2.25Dingo

% ti

me

Figure 4.18: Percentage of time spent resting (black), in low activity (white), in crepuscular activity (grey) and in increased (diurnal and nocturnal, short and long distance) activity (diagonal lines) for dingoes 2.21, 2.23 and 2.25 who lived in territories adjacent to one another.


218

4.2.4 Observations of mortality and dispersal

Of the 47 dingoes captured throughout this study, ten died (one was due to capture

myopathy) and eight potentially dispersed or travelled extraterritorially (cf. Figure

4.1). Extraterritorial forays and dispersal could not be separated since one individual

forayed extraterritorially and returned to his territory on more than one occasion.

Dispersal and extraterritorial forays were therefore grouped because some collars

provided incomplete records of extraterritorial events. The time release mechanism

on a short term collar, for instance, released during an extraterritorial foray and it is

not known whether this dingo also returned to his territory or if he continued to travel

extraterritorially. The proportions of males and females potentially dispersing were

both at 13%. A higher proportion of females (13%) forayed extraterritorially

compared with males (10%). In total, 19% of captured males and females died

(excluding the case of capture myopathy), 11% travelled extraterritorially, and 13%

may have dispersed. Seasonal and monthly observations recorded during field work

or from other data are shown in Table 4.4. All females that died or dispersed did so at

the end of whelping or in the rearing season and males generally had two peaks, one

during the breeding season and another during whelping season. In contrast, sightings

or camera observations of groups were lowest during whelping (5) and increased

during rearing (7), exploratory (10) and eventually peaked during breeding season

(14). These data are consistent with seasonal changes in dingo biology using PAI and

ARF indices in chapter three.


219

Table 4.4: Total observations of deaths, potential dispersal and extraterritorial forays, groups and pregnant females

Breeding Whelping Rearing Exploratory M A M J J A S O N D J F Total

♂ - - 1 - - 1 - 2 1 1 - - 6 Deaths ♀ - - - - - - 3 - - - - - 3

T - - 1 - - 1 3 2 1 1 - - 9 Dispersal ♂ 1 1 1 - - - 2 - 1 - - - 6

or ♀ - - - - - 1 - 1 - - - - 2 Extraterritorial T 1 1 1 - - 1 2 1 1 - - - 8

Group s T 5 5 4 3 - 2 1 2 4 2 6 2 36 Pregn ant obs. ♀ - 2 - 1 - - - - - - - - 3

Mapping the dispersal and mortality data of captured females (Figure 4.19) showed

four of the five observations were from genetically related females in the same area

(cf. Figure 4.1 for relatedness data). One of the collected skulls (dingo 2.34) had a

hole the size and shape of a canine tooth in the top, which implied intraspecific

predation may have been the cause.

Figure 4.19: Observations of extraterritorial movements and mortality of female dingoes 2.8 (blue), 2.15 (purple), 2.30 (red), 2.31 (green) and 2.34 (brown). Numbers 1, 2 or 3 = first second and last sighting respectively; X =Found deceased.


220

4.4 Discussion

The locational data collected in this study enabled quantification of short term daily

movements, medium term seasonal movements and longer term annual movement

patterns with great precision when compared with studies using VHF telemetry. The

GPS data logging collars and the duty cycles used in this study proved ideal for

investigating annual, seasonal and monthly home range size, movement patterns and

patterns of activity. The extent and quality of data collected exceeded that deliverable

by conventional telemetry techniques. Obtaining daily locations using VHF telemetry

would have been too laborious and costly to sustain the use of a vehicle and salary of

the tracker(s). In addition, the effect of erroneous data from triangulation of signal

bearings received from VHF transmitters (Mech and Barber 2002) is negated with

GPS telemetry. Using GPS systems, daily locations of dingoes are almost guaranteed,

barring technical failures.

Data from GPS collars used in this study can also be used as an outline of times to use

VHF telemetry if it is the only available technology. Some of the collars were so

successful that behavioural traits of communal living in hypercarnivorous canid

populations were observed. Land managers can easily assess the value of declaring

dingo conservation habitat inside protected areas from the data presented. Pack

territories were as clearly demarcated as some of the genetic relatedness clusters

shown in maps in chapter two, but more than that, they directly related to the genetic

clusters. Sociality was observed and territory boundaries were distinctly associated

with prominent landscape features and movement corridors for extraterritorial

foraging or dispersal. The precision of these collars showed how the dingo movement


221

pattern system operated between conspecifics and intraspecifics that had never been

subjected to large scale dingo control campaigns.

Due to the consistency and accuracy of HDOP values in this study, no points were

lost or labelled as outliers in these data. Significant differences in HDOP values

between the 13 month collars and the short term collars may be due to differences

between years, seasons, duration of deployment, age of the collars and improvements

in technology. Concern about losing outlying GPS locations when analysing GPS

based data due to inaccurate DOP values has been expressed (Bandeira de Melo et al.

2007) though outlying locations in MCP analyses may not be crucial for inclusion.

When raw location GPS data points were plotted within MCP analyses, “gaps” or

“holes” between GPS data points were visible within the polygon. These gaps

generally appeared to be negligible, however, may have made the home range

estimates larger than they should have been. Inclusion of MCP analyses was only

necessary for comparing home range size with past studies. In this study, the volume

of points used to calculate the MCP or the kernel estimates far exceeded the number

of points used in past research where mathematical analyses of home range were

essential to account for errors in the home range estimation technique. In contrast,

mathematical analysis of this dataset proved GPS location data of this calibre can

stand alone to represent the home range and movements of animals.

Problems encountered with use of GPS technology included the failure of time-release

mechanisms and failure of the Argos collars deployed during June 2007 to transmit

and log data. Although a large number of points were obtained for each animal in the

current study, intensive radiotelemetry techniques used by Harden (1985) were only

able to obtain locations of dingoes in rugged terrain on 51% of the days the collars


222

were deployed. Although the accuracy of all GPS locations on some days for the 13

month collars were greater than a HDOP value of three, the range of fixes per day was

6-8 locations and the lowest proportion of expected fixes for the length of deployment

was 74% (Table 4.1). Tracking animals using a mounted VHF system in a helicopter

or light aircraft (cf. Thomson 1992d) to obtain a similar dataset is not cost effective

although the use of aircraft was essential to retrieve the timed-release collars in this

terrain. Inability to aerially survey a larger area and locate collars of individuals that

may have dispersed further than expected may be one reason some collars were not

found or salvaged in the current study.

GPS data logging collars used had more accuracy to the metre compared with Argos

satellite telemetry systems (Anon. 2006). This is an essential component when

investigating habitat selection and resource partitioning with GPS (Di Orio et al.

2003). Argos satellite telemetry has better application for studies with questions

concerned with dispersing or wide ranging animals, and should be used in future

studies of dingoes in the SGBMWHA. Both GPS and satellite collar systems are

limited by battery power and the power consumption of the duty cycle. Discussing

project parameters with collar manufacturers contributed to maximising the benefits

of the technology, customised the duty cycles to suit the research questions and also

provided scope to attain ancillary data for further questioning. GPS collar duty cycles

selected for the current study only allowed annual movement patterns to be monitored

when periods of longer storage time would have been advantageous to explore

questions across years.

Observations of dingo movements from VHF telemetry were useful for some

information of site fidelity or behaviours such as dispersal but lacked accuracy and


223

consistency. This may have been related to the rugged terrain and difficulty in finding

locations for triangulation. Resighting data derived from dingoes outfitted with

coloured nylon dog collars also proved useful if the dingoes were resighted but luck

played a major role. The death of dingo 2.18, for instance, was reported by staff of

the protected area engaged on routine tasks. Extraterritorial movements of female

dingo 2.15 would not have been noticed if a passive camera trap was not

coincidentally set up in her path. The use of various techniques, such as VHF

telemetry, ad hoc sightings and passive camera traps during this study validated some

observations but the total data collected using such methods had limited scope for

analyses. Irregularity in VHF and observational data for instance were often due to

the movements of the dingoes in this rugged terrain, rather than faulty data collection

techniques.

Population data presented in Figure 4.1 was one of the more useful methods of

analysis to identify and quantify general patterns in social systems, dispersal and

extraterritorial movements and survival. One obvious pattern is that genetically

related dingoes were trapped around the same time. Observations of genetically

related captures between dingoes 2.13 to 2.34 showed that the succession of captures

was a pair from the orange group, a pair from the green group, a pair from the blue

group and a pair from the purple group, followed by five individuals from the orange

group and then four individuals from the green group. This likely indicates how

packs moved into the interstices of their territories (where the traps were set)

consecutively every two to three days, perhaps even after trapped and released

dingoes had retreated to protective cover. In addition, although dingoes 2.22 and 2.23

were genetically related to dingoes 2.24, 2.25 and 2.26, they maintained different

territories. The former were trapped on the same day as dingoes 2.20 and 2.21, and


224

the latter were trapped two days after dingoes 2.20 and 2.21 in the same traps.

Thomson (1992d) stated that encounters between neighbouring dingo groups were

extremely rare and concluded, based on coat colour that packs consisted of closely

related individuals. Conducting further analyses on the effects of movement by dingo

groups on each other will be useful for management, especially if removal is the

method of “managing” a population. Other data shown in Figure 4.1 included the

observation of related individuals in group hunting activities (dingoes 2.10, 2.12, 2.14,

2.24, 2.25 and 2.26), capture of unrelated individuals together (dingoes 1.2 and 1.3,

and 2.31 and 2.32) and trends in female mortality and potential dispersal during

whelping and rearing seasons that were analysed further in section 4.2.4. These data

are comparable with observations of other communal living canids (cf. Moehlman

1986; Macdonald and Sillero-Zubiri 2004b).

4.4.1 Success of GPS collars to show core areas of activity

There were two factors apparent from the presentation of home ranges observed

during this study. The first was the success of the GPS collars to define a home range

without mathematical analyses. Simple presentation of GPS locations has the ability

to depict travel routes, rendezvous sites, spatial separation and overlap of territories,

interactions between individuals and more importantly, identification of core areas

(including den sites), all of which are outcomes of this study. GPS data clearly

displayed crucial differences between movements within core areas, territory

maintenance (such as border patrols) and exploratory behaviour, as described by

Harden (1985). Home range terminology, therefore, needs to be redefined to suit

changes in technology as they occur. If exploratory activity, where dingoes travel

between or through other dingo territories and return to their home range, were to be


225

considered a part of a home range then it might be better termed as a “dispersed

range” (Purcell et al. 2006). A 50% kernel can remain defined as the core area of

activity and the 95% kernel can remain defined as the home range (Purcell et al.

2006). The dispersed range of dingo 2.4, observed during the breeding season, was

95.6% larger than his core area during the breeding season. Exploration efforts of this

male decreased at the conclusion of the breeding season when he returned to his core

area of activity, where the collar released (Figure 4.8).

The second factor depicted from these data was the identification of core areas. The

relationship of individuals and, in some instances, packs with their core area indicated

core areas are a highly valued part of a territory. Sillero-Zubiri and Macdonald (1998)

analysed behavioural data on territory defence and inter pack interactions in Ethiopian

wolves and suggested that intruding Ethiopian wolves identified territories by the

frequency of scent marks (cf. Bowen and Cowan 1980). Strong site fidelity of

dingoes to their core areas would suggest the frequency of scent marks or strength of

scent must increase to a point where it is beyond doubt to the intruder(s) that they

have entered the territory of a pack. The only core area which overlapped with a core

area of a neighbouring group was that of dingo 2.1. Data presented in section 4.3.1

identified dingo 2.1 as an omega, or a loner, so the overlapping core area may be

representative of a range for a lone dingo because it overlapped with a core area of an

unrelated dingo (female dingo 2.2). In comparison with observations by Thomson

(1992d), social organisation of the animals under study has to be taken into

consideration when attempting to interpret home range size and space utilisation by

communal living canids.


226

Reasons for the selection of a core area by a pack are subject to debate in this study

area because there are no known limits in resources. Thomson (1992d) suggested that

food availability was the determinant of the size of a core area for dingoes in Western

Australia. This is consistent with hypothesis where home range size is related to

resource abundance within the optimal space (Macdonald et al. 2004). The Resource

Dispersion Hypothesis (Carr and Macdonald 1986), however, is not applicable in the

current study. Data in chapter three showed that dingo diet fluctuated in relation to

the identified biological seasons of dingoes, independent of resource abundance or

climatic events. Prey activity coincidentally correlated and fluctuated with dingo

activity. If food, water and habitat were homogenous and maximum pack size or

threshold was achieved, then the home range and movements of dingoes in this study

were only limited by intraspecific competition or variations of social systems. That is,

this population showed traits similar to other findings on the tolerance of conspecifics

(Laundré and Keller 1981; Creel and Macdonald 1995; Gese et al. 1996) via

observations of extraterritorial foraging and mortality of males and females (Table

4.4; Figure 4.19). These data suggested tolerance of conspecifics was low during

periods of breeding and whelping.

4.4.2 Effect of social systems on movement patterns

Lindstrom (1986) used the Territory Inheritance Hypothesis (TIH) to show that spatial

or temporal variations in food supply did not affect territory ownership. Instead,

territory is defended by the dominant breeders until they weaken or die and is then

inherited by the next fittest progeny. Intolerance of conspecifics in the current study

indicated that packs within the SGBMWHA were stable and functional. The TIH

seems generalised but more applicable in this population of carnivores and study site


227

because resources were generally homogenous and no presuppositions about food

resources were required. Boitani et al. (1995) made similar observations when

studying a feral dog population in central Italy. The feral dogs lived directly between

rubbish dumps and human settlements which provided both food resources and stray

dogs for recruitment into the population. The underlying premise that territories will

be inherited by relations does not fit all observations by Boitani et al. (1995) though

the abundant resource supply in their study and the current study are fundamentally

comparable. Overlap of 95% home range estimates also suggested that interstices of

territories and associated resources were habitually shared between neighbouring

dingo packs. These observations were consistent with interactions between wolf

packs and moose herds (Mech 1970), and dingo packs and water resources (Corbett

2001).

The movement data were consistent with the conclusions from chapter three that

modern day dingoes remain primitive hypercarnivorous canids characterised by one

annual breeding cycle. Further, research by Boitani et al. (1995) showed feral dogs

continued to breed and attempted to rear young biannually in a resource saturated site.

All movement patterns for males and females in the current study showed core areas

reduced in size during whelping and slowly increased in size as the pups aged. Speed

and distance between points also increased as the pups developed until the next

whelping season. The time periods selected for biological seasons were not

particularly fitting for movements of female dingo 1.4 or general activity patterns on

roads for dingoes in Site 1. This does not however, affect the outcomes of this

research because the distribution of resources and contrast in terrain (cf. section 1.2.7)

between Site 1 and Site 2 are obvious, and differences in trends were expected.

Breeding trends observed in both sites remained annual, opposed to biannual as has


228

been speculated for “hybrid dingoes” or “wild dogs” (cf. Jones and Stevens 1988;

Corbett 2001; Daniels and Corbett 2003). Dingoes have however been observed to

raise two litters from each of two females within one pack in Western Australia

(Thomson et al. 1992a). Van Ballenberghe (1983) reported that where one wolf pack

raised two litters, these litters were always borne by different mothers. There is no

evidence that dingoes in the SGBMWHA can either raise two litters within one year

or that females go into oestrus twice, even with abundant resources. On the contrary,

data from the current study (Figure 3.2) was consistent with findings by Corbett

(1988) who reported intraspecific predation in captive dingo packs and theories of

reproductive suppression by communal-living canids (Moehlman 1986; 1989; Geffen

et al. 1996; Macdonald and Sillero-Zubiri 2004b). This chapter provided more

evidence consistent with the hypothesis of this study that movement patterns

demonstrate the functional role of dingoes in the SGBMWHA by showing their

discrete territories and seasonal patterns of extraterritorial explorations.

No significant differences or biases were observed between males and females for

extraterritorial movements or mortality. Observations of mortality and dispersal, like

movement patterns, increased during the whelping and rearing seasons when the

breeding pair may have been concerned about space and resources for their young.

Females in Site 2, for example, dispersed or died in the 2005 and 2006 whelping and

rearing seasons.

The family Canidae was introduced in chapter one as landscape specialists

(Macdonald and Sillero-Zubiri 2004b). Merrill and Mech (2003) suggested that

wolves may have complex memories of landscape features. Dingoes in this study

showed how ridges and gullies were utilized when foraging and patrolling borders.


229

The border patrol by dingoes 2.20 and 2.21 and the extraterritorial foray of dingoes

2.4 and 2.25 are consistent with observations by previous authors.

Analyses of 95% MCP data showed dingoes generally maintained peripheral borders

which encompassed their 50% core area all year round. Merrill and Mech (2003) also

showed that one nursing female wolf continued to patrol the boundary of her territory

in early stages of rearing. Although specific details of nursing females were not

collected in this study, all females maintained movement patterns to 95% MCP home

range borders around their 50% MCP core areas of activity in every season. This

indicated that maintenance of scent posts may be necessary for dingoes to: a) signal to

potential intruders that the pack remains functional; b) to protect their young from

interspecific and intraspecific predation; and c) maintain olfactory communication

with neighbouring groups (Laundré and Keller 1981).

Seasonal differences in the size of home range for each season were not significant

though centres of activity and 95% home ranges varied in shape and location. Gender

differences in mean home range size were expected during whelping season but this

was not the case. Males and females generally showed strong site fidelity all year.

Dingoes 1.4, 1.7, 2.1 and 2.4 always returned to their 50% MCP after extraterritorial

forays in breeding seasons and this may be the case for all dingoes which do not die or

permanently relocate to a new territory at the conclusion of extraterritorial excursions.

Information on social organisation and behaviour would greatly assist interpretation

of data derived from GPS collars. Harden (1985) made similar suggestions to

improve movement data collected using VHF telemetry and Thomson (1992a, b, c, d)

and Thomson et al. (1992a, b) showed the benefits of studying behaviour whilst


230

studying movement patterns. Studies on dingo activities and behaviours during

border patrol movements, such as that exhibited by dingoes 2.20 and 2.21 and

reported by Sillero-Zubiri and Macdonald (1998) should be a priority for future

research to assist land managers. If dingoes 2.20 and 2.21 were killed during their

border patrol, would their former territory be invaded by dingoes from adjacent

packs? Could the remaining members of their pack defend their territory? If not,

would they disperse and attempt to become established elsewhere or die? Research to

improve dingo “management” has never accounted for the effects on dingo socio-

biology, nor has it considered the impact of dingo control on Australian ecosystems.

Research assessing the effects of secondary poisoning or the effects of traps on non-

target animals has been undertaken (Meek et al. 1995; Fleming et al. 1998; Twigg et

al. 2000; Eldridge et al. 2000; Körtner and Watson 2005; Claridge et al. 2006;

Claridge and Mills 2007) but these were generally in response to public concern

regarding animal welfare issues or to review products for improved dingo

“management”.

4.4.3 Activity, movement, dispersal and home range of dingoes

Animal movement data collected during this study have been used to describe

characteristics of the patterns of movement of the dingoes in the SGBMWHA. The

home range sizes observed were within the limits of home ranges reported by past

studies on dingo movements, which varied between 9km2 and 27km2 in north eastern

NSW (Harden 1985) and 37km2 and 84.8km2 in Western Australia (Thomson 1992d).

Expressing isopleths of data in this study revealed clearly demarcated boundaries

within 90% kernel contours. These raised the questions of how many individuals live

within a pack home range, how many packs live within the SGBMWHA and what


231

their interactions are. If each pack within a mean home range of 37.7km2 consisted of

five dingoes (aged >1.5 years) and four pups during whelping season, the estimated

upper limit for the SGBMWHA dingo population is approximately 467 dingoes

distributed through 58-59 packs annually. This model has inherent flaws because it

does not and cannot account for such factors like social cohesion of groups, survival

estimates, habitat variation (dingo home range in Site 1 is generally double the size of

those in Site 2) and selective management of the population across vast area of the

SGBMWHA. If 61.8km2

(mean for Site 1, n = 3) was considered the mean home

range of dingoes for the entire SGBMWHA then approximately 285 dingoes are

distributed through 34-35 packs. An average of the two estimates is calculated as 376

dingoes, distributed through 46-47 packs in the SGBMWHA. The outcomes from

extrapolating these data are varied. Applying the population estimates to review

current dingo management practices indicates implemented programs on the periphery

of the park are effective. In the four years of this study, approximately 40 dingoes

were “managed” by means other than baiting (Glover, pers. comm. Moss Vale Rural

Lands Protection Board). The low number of troublesome animals that needed to be

culled on the edge of park boundaries is consistent with data from this study, which

showed that only 1 out of 35 dingoes (2.9%) collared with a VHF transmitter (12

annual GPS collars; 5 short term collars; 10 VHF collars; and 8 Argos/GPS collars)

was observed outside of SGBMWHA boundaries.

Harden (1985) also showed that dingoes did not travel from the core of protected

areas to kill livestock. The common theme for dingo migration from this and other

studies then is reproduction. Controlling oestrous cycles of domestic dogs living

adjacent to protected areas could reduce dingo movements around the periphery of a

park when dingoes are actively seeking a mate. This would minimise interactions


232

between dingoes and livestock and minimise the effects of dispersal sinks which are

created when dingoes are controlled using the ‘buffer zone’ strategy specified by

Thomson (1992d).

Female dingoes were not observed outside of their common range until the whelping

and rearing seasons when their mortality rate appeared to increase, coincidentally

when pups begin to forage. Male dingoes in this study were not observed outside of

their home range until the breeding season. Interaction with farmland was limited to

one individual male who took ten days to travel to the border of the SGBMWHA.

Relating the data to annual dingo biological cycles showed that extraterritorial forays

may have been motivated by instinct to increase opportunities to mate or potentially to

escape intraspecific predation. Dingo 2.4 travelled an approximate route greater than

50km from the most eastern GPS locations of his extraterritorial forays, west, to a

property housing and breeding captive dingoes when the GPS collar logged

approximately 17 locations within 3km of the property (Figures 4.2 and 4.3; the

property was not identified for privacy). At the time, between 18 and 22 dingoes were

thought to be housed on the property. If 7-10 of these dingoes were females in oestrus

surrounded by 10-12 males howling from a captive location then it could be assumed

many more wild dingoes within the SGBMWHA would probably be attracted to the

property.

Further research using a larger sample size of dingoes across the whole GBMWHA

would greatly assist future management of this dingo population, and adjoining rural

lands and urban areas. To maintain the integrity of the research, park boundaries and

domestic dogs from properties adjacent to the protected area would have to be

included in the study. Apart from research by Meek (1999), movements of free-


233

ranging domestic dogs have been seldom studied yet domestic dogs are known to

predate livestock and native fauna. Boitani et al. (1995) discussed outcomes of

studies on feral dogs whereby feral dogs were ‘merely a nuisance’ to land managers.

On the contrary, free ranging pet domestic dogs, similar to those studied by Meek

(1999), were reported to kill three calves in the study by Nesbitt (1975), cited by

Boitani et al. (1995), and domestic dogs were also observed killing livestock in Italy

by Boitani et al. (1995).

Patterns of movement observed and discussed by Harden (1985) and Thomson

(1992d) were general and similar to preceding discussion in this chapter. Thomson

(1992d) showed that pup rearing was the most important seasonal factor influencing

dingo movements. Harden (1985) showed that dingo movements were short and

interspersed with rest periods, and these observations are consistent with data shown

in the current study. Based on this and the fact that dingoes in this study rarely

reached running speed (>6000m/hr) it can be hypothesised that dingoes of the

SGBMWHA hunt by stealth rather than pursuit. This was also hypothesised by

Harden (1985) for dingoes in the North Eastern Tablelands of NSW and expected in

this study given the terrain and habitats of both study areas were mountainous and

forested respectively.

Movement and activity patterns in this study were examined in greater detail than in

previous research and with much better accuracy. Harden (1985) reported that

dingoes were neither nocturnal nor exhibited crepuscular patterns which implied they

were cathemeral. Substantial amounts of time were spent resting by dingoes in the

current study, consistent with the hypothesis that dingoes are hypercarnivorous

canids. General activity of dingoes in the SGBMWHA was markedly different on


234

some days compared with others. On average one day in eight had more nocturnal

activity than the others, but during most 24hr periods the majority of time was spent

resting (section 4.3.3). Furthermore, patterns in activity between days were irregular.

In one instance an eight day gap was observed between nights with increased activity

and then the dingo was exhibiting nocturnal behaviours for two of the three

proceeding nights. It seems then that patterns in dingo activity are very variable and

external factors such as weather, moonlight, temperature and breeding status need to

be incorporated in future analyses. There was evidence that some dingoes were more

active when the moon was full, but movement patterns for other dingoes showed no

association. The proposition that moon phases affect dingo movements requires

further testing in subsequent studies. Observations of activity in past studies were

also affected by the sampling regime set by the researchers. Allen (2006) did not

examine daytime activity because dingoes had previously been reported as nocturnal

or crepuscular (Thomson 1992b). The current study of patterns of activity showed: a)

dingoes were more diurnal than nocturnal in this study site; b) that patterns across

consecutive days are not predictable; and c) that patterns varied per dingo. Activity

patterns of dingoes 2.23 and 2.25 appeared similar. However, they were from

different dingo groups and dingo 2.25 left on an extraterritorial foray. Therefore, if

pre-dispersal movement patterns of dingo 2.25 were an indication that he was

preparing to travel extraterritorially, the duration of the current study may have been

too short to make dispersal observations of dingo 2.23. Alternatively, it may be

hypothesised that dingoes 2.23 and 2.25 were of similar rank in their packs and dingo

2.25 chose to seek opportunities elsewhere.

Merrill and Mech (2003) set GPS collars to either store one location between every 15

minutes or in some cases every three hours. They identified that peaks in activity may


235

have been missed due to the sampling regime selected by previous researchers and

referenced conflicting studies which stated wolves were primarily nocturnal. The

GPS collar duty cycle selected (every two hours) by Bandeira de Melo et al. (2007)

was more accurate at identifying daily activity trends in maned wolves. In addition,

their data were consistent with hypotheses that maned wolves were more nocturnal

and crepuscular than diurnal or cathemeral. Determining an appropriate GPS collar

duty cycle that will collect data to answer the questions being tested is a matter of

getting the balance right between the frequency of locations and the duration of the

study. In this study, the data from collars deployed for annual collection of data

showed many movements were being missed, such as the details of the extraterritorial

foray by dingo 2.4. Short term ten-minute duty cycles were selected in a following

experiment to answer questions related to diurnal movements of dingoes in the

breeding season. This duty cycle was very successful at showing details of such

forays except the length of collar deployment was sacrificed in the process because of

limits to battery power. The data collected gave new insights into dingo behaviour in

the SGBMWHA population and diurnal activity data collected throughout other

biological seasons would be helpful in better understanding the behaviour of these

animals. Nevertheless, GPS collars will always be limited in scope if they remain

dependent on battery power. Future studies will hopefully fill in knowledge gaps and

fabricate new techniques to improve the output of store on board GPS data units. The

Argos satellite/GPS store on board hybrid collars deployed in September 2007 for

data on long term movement patterns are one such adaptation except they also remain

limited by battery power.

On the selection of a core area of activity, dingoes in this study appeared to inhabit an

area which provided optimal resources (food, water, protective habitat) sufficient to


236

maintain daily energy requirements. Dingoes may, therefore, concentrate their

activities towards territory maintenance for inheritance by their next fittest progeny in

the core area of activity. Patterns of movement obtained from GPS collar data were

expected to increase in dimensions, with increasing age of the pups, though this was

not observed using selected methods of analysis. Instead, when pups were too young

to be introduced to the borders, such as during the whelping season, 95% MCP

territories were maintained. Spoor counts on sand plots suggested that borders

appeared to be patrolled by one to three adults instead of numerous dingoes during

these periods. This observation coincided with reduced activity on sand plots and

increased defecation rates on transects. Depending on the location of sand plots or

transects within interstices of territories or the core area of a pack, abundance of spoor

and/or scat during certain seasons may indicate which parts of a transect pass through

core areas and which parts of a transect are a territory border. Figure 3.14 showed

there were consistent levels of activity on road infrastructure and the number of dingo

scats and tracks per month were approaching significant negative correlation (cf.

section 3.3.3). Sand plots on known or constructed dingo travel routes such as roads

proved to be very useful to monitor changes in the frequency of scent post

maintenance over time though were of limited value in determining movements of

packs during some biological seasons. Peaks in passive activity indices were peaks

because extraterritorial foraging increased during the breeding season and dingoes

which reside in core areas may have needed to increase the frequency of olfactory

signals to deter extraterritorial foragers (intruders) or attract unrelated helpers for the

following year (cf. chapter three). Measures of relative abundance and passive

activity using sand plots therefore increased with the age of the pups and were lowest

when pups were youngest.


237

Randomisation of sand plot locations is therefore essential because sand plots in core

areas may show constant high activity. If sand plots were located on a landscape

feature which separated two packs then these data will be biased by the activities of

the packs using that feature. Low activity levels on sand plots in Site 1, a long ridge

between two river gullies, in one month may have high sand plot activity on a parallel

transect that is on the river running through the core area in the same month. Changes

in diet were therefore related to changes in activity because raw GPS data points from

collared dingoes were condensed due to whelping activities and various prey species

may not be available to predate within the core area of activity. The negative

correlation observed between scats collected per month and dingo tracks observed on

roads per month were consistent with data that showed 95% MCP borders were well

maintained when general activities were condensed to smaller 50% MCP core areas

during whelping and rearing seasons. It can be hypothesised then that if frequency of

dispersal and intraspecific predation or aggression increases during the breeding and

whelping seasons then monitoring dispersal and intraspecific predation may indicate

whether or not the dingo population in question is functional. Dispersing dingoes are

essential for reducing genetic drift and to repopulate areas devoid of dingoes or

affected by baiting/ control campaigns, to become helpers for packs that suffered poor

reproduction rates or high predation of offspring. Maintenance of the dominant pair

to control order in their pack, which consequently maintains order between the

surrounding packs within a protected area (if each dominant pair is maintaining order

in each of their own packs) ensures resources are not over exploited in core areas and

territories. Evidently, dingoes are self-regulated dependant on resource availability

and may not need to be “managed” or “controlled” as drastically as they currently are

in some areas.

Chapter Five:

Effects of cultural transmission and a proposal for management

© Jon Reid, Fairfax.

Effects of cultural transmission

239

Chapter Five: Effects of cultural

transmission and the proposal for

management

5.1 Introduction ..................................................................................... 240

5.2 Cultural transmission ..................................................................... 242

5.3 Proposal for management .............................................................. 248

5.4 Conclusions ...................................................................................... 253


240

This study has investigated the ecological role of dingoes as a top order predator in a

regenerating Australian landscape. As was hypothesised in chapter one, measures of

diet, abundance, and activity and movement patterns showed repeatable behaviours,

significant relationships and strong trends that indicated dingoes have a functional

role in this landscape. Functional role of dingoes was defined in section 1.3 as the

dynamic responses of dingoes to immediate factors of the environment including

changes in populations of sympatric (competing and prey) species (Krebs 2001). The

dingoes studied in the SGBMWHA showed that they either caused feral cats and prey

species to be more or less active at certain times of the year or that they responded to

periods of higher or lower activity by prey species and feral cats. Dingoes also

showed that movements within or outside of their territory were effected by

conspecifics within their own pack or neighbouring packs.

5.1 Introduction

Recent studies have provided evidence consistent with the hypothesis that dingoes

have a functional role in Australian ecosystems. Ritchie and Johnson (2009) reviewed

literature on the function of apex vertebrate predators and concluded that they have

significant effects on ecosystem processes and organisation of communities. Johnson

and VanDerWal (2009) re-examined data on abundance of dingoes and foxes in NSW

and showed that dingoes, or wild dogs, had a negative affect on fox populations.

Wallach et al. (2008) alternatively showed that some threatened species in South

Australia were surviving in the presence of dingoes. One conclusion from their study

was that dingoes may have been functioning as a top order predator in areas where

they were presumed rare due to continued lethal control, and that the dingoes may


241

have been suppressing foxes. Dickman et al. (2009) also reviewed numerous studies

and discussed data that were consistent with the hypothesis that dingoes have a

functional role in Australian landscapes. These studies presented the concept that

dingoes perform a functional role in Australian landscapes and the data presented in

the current study provides evidence of that role.

Dickman et al. (2009) emphasised that if Australian land management organisations

and livestock enterprises did not attempt to coexist with dingoes, then there will be

serious ramifications for: a) ecosystem processes and functioning; b) overgrazing by

native species and introduced hard hoofed domestic and feral livestock; c) continued

spread of noxious weeds; d) overuse of poisons to control pest species; and e) ongoing

salinisation of remaining soil. Other threatening processes not discussed were the use

of poisons to control weeds, desertification and acidification of ecosystems and the

contribution of livestock to anthropogenic climate change (Anon. 1996; 2001d;

Griffiths 2001; Lunney 2001; Quinn 2001; Beeton et al. 2006; Steinfield et al. 2006).

In addition, both movements of dingoes and livestock depredation were recently

shown to increase following dingo control programs in South Australia (Allen and

Miller 2009), consistent with similar studies in Queensland (Allen and Gonzalez

1998). Using preliminary data, Allen and Miller (2009) stated that dingo control did

not appear to reduce calf losses or return an economic benefit for the livestock

enterprise. Lethal management of dingoes may therefore create a risk to livestock

enterprises, rather than be a plausible option for dingo control.


242

The above observations are important for many reasons but most notably, they

implied that dingoes are important for sustainable management of Australian

landscapes. Management of dingoes is becoming a national issue, rather than a local

or state wide control program and it is beyond the debate of dingo “purity”. Ritchie

and Johnson (2009) stated that population genetics and its connection with social

structure and behaviour is where the associated potential ecosystem benefits exist.

The current study in the SGBMWHA is the first to show the connection between

population genetics, social structure and behaviour exists (especially seasonal

predatory behaviours and training of juvenile dingoes by adult dingoes). Results from

the current study may be viewed as a model of dingo social organisation and

behaviours when they are living in optimal wild conditions. Maintenance of pack

structure, by not exercising routine control over the SGBMWHA, or any other

protected area for that matter, is arguably the way that dingoes should be managed.

Much of the study was conducted in a part of the SGBMWHA which was used for

agricultural purposes for over a century prior to being subsumed as a protected area

for water storage to supply the city of Sydney. The following discussion takes

account of a range of biological and cultural parameters to aid in development of a

holistic and adaptive proposal for future management of dingoes in this site of

international significance. To set the context, some aspects of Australian culture

related to perceptions of dingoes are revisited.

Stories of the dingo generally take two forms which depend on the story teller. One

form is a romanticised narrative from people impressed by the wild heart of the

5.2 Cultural transmission


243

Australian bush. The other form of story is a yarn by a classic “Aussie battler” at war

with a cunning, enigmatic enemy and usually ends or begins in bloodshed and a

trophy pelt or scalp. Marcus (1989) and Parker (2007) discussed dingo stories

preserved by writings and artwork to explain what it may be like to be a dingo, or to

be living in Australia with dingoes. One common theme in both studies, although it is

never suggested, is that the iconic Australian dingo is a scapegoat. Part of the public

reaction to the cry of Lindy Chamberlain on the 17th

August 1980 at an Uluru camp

ground, “The dingo’s got my baby!” , may have been because they felt the dingo was

being made a scapegoat again. Marcus (1989) stated that, regardless of how the baby

Azaria Chamberlain died, Lindy Chamberlain was incarcerated by the resultant media

and public outcry for being an uncaring parent. Similarly, public outcry could inhibit

pastoralist activities for saying something such as “Dingoes [or wild dogs] killed my

sheep!”

Dingoes are an iconic Australian mammal yet the largest predator proof fence in the

world was constructed to limit their distribution for the benefit of the agricultural

sector, without exploration of the impact that exclusion of dingoes might have on

semi-arid Australian ecosystems. Early Australian development relied heavily on the

sheep industry and protection of sheep was seen to be more important than the

protection of dingoes. Declines in biodiversity including the extinction of 18

Australian native mammal species that represented over half of the global mammal

extinctions in 200 years (Johnson and Wroe 2003; Johnson et al. 2007) could be

attributed to the rise of agriculture and the neglect of ecosystem function. Johnson et

al. (2007) showed that high-density dingo populations may aid the conservation of

native Australian marsupials. It seems strange then that public outcry against the


244

suppression of dingoes has not questioned the validity of the dingo fence and dingo

“control” programs. Marcus (1989) and Parker (2006; 2007) stated that livestock

farmers are protected by such public outcry due to Australian Frontier Nationalism

(AFN), a culturally inherited trait of many Australians that protects “A ussie battlers”

using the prejudices inherited from European colonisation.

During the current study, a male/female dingo pair and two of their pups were

loitering on and around the grounds of a high school and the statement was: “Dingoes

are hanging around the school yards and stealing the children’s lunches”. The male

and pups were inevitably “controlled” by staff from land management organisations

because they were a threat to the safety of the students. Response to an actual fatal

incident on Fraser Island during 2001 was lethal management of some dingoes

followed by trials of non-lethal techniques under the Dingo Management Strategy

formulated in response to the event (Edgar et al. 2007). Prior to the implementation

of “management”, the first response to any statement blaming a dingo or dingoes for

prevailing circumstances or events should be “Why?” For instance, ascertaining

reasons why or how a dingo would be prompted to seize a baby from within a tent in a

camping ground, could be related to cultural transmission of behaviours in an area

where dingoes have been habituated to obtaining food in camping grounds. This is

similar to dingoes foraging in school grounds or livestock enterprises located adjacent

to dingo habitat. The dingo is innately a predator first and this should be considered

in development of management strategies for landscapes where dingoes and humans

intersect.


245

Canids learn behaviours through cultural transmission (Baenninger 1978; Boitani et

al. 1995; Borgerhoff Mulder et al. 2006). Cultural transmission occurs when

behavioural traits are inherited (vertical transmission) from parent populations or

copied (horizontal transmission) from neighbouring populations (Cavilli-Sforza and

Feldman 1981; Borgerhoff Mulder et al. 2006). Slabbert and Rasa (1997) and

Pongrácz et al. (2003) demonstrated how domestic dogs learned behaviours by

watching their mothers, conspecifics or a human equivalent. Both studies showed that

asocial (individual) and social learning abilities contributed to increased fitness.

Observational learning thus becomes the nexus for cultural transmission of

behavioural traits such as predation on livestock, predation on native species or

scavenging from human settlements.

The above observations imply that the dingo at Uluru learnt, through observation, that

exploring into a tent would result in the opportunity of taking food. Dingoes have

been known to become habituated to finding food items around camp sites (Montagu

1942) and this may have been reinforced by campers willingly offering food items to

dingoes at the time of the event. Pongrácz et al. (2003) showed that problem solving

behaviours in domesticated dogs involved complex social learning interactions and

individual experiences. In contrast, wolves are naturally skilled problem solvers,

highly social predators with well-organised social hierarchy who depend on vertical

transmission to acquire behavioural traits (Pongrácz et al. 2003). Frank (1980)

compared trainability, learning and intelligence of a malamute, a malamute-wolf

hybrid and a wolf, and concluded that wolves have a “duplex” information processing

system of cognitive processing and instinct because of their wild heritage. The

malamute, in their study, failed to unlock a latch after six years of watching the


246

researchers enter and leave the enclosure. However, the malamute-wolf hybrid

performed the task within two weeks and the older female wolf unlocked the latch

after watching the hybrid do it once. In addition, the hybrid, having received no

obedience training, performed the obedience routines being taught to the malamute

over six days, instantaneously. Indeed it seems as if wolves will learn behaviours

from conspecifics or neighbours much faster than they would from humans.

Allen and Gonzalez (1998) explored reasons why livestock losses increased following

dingo baiting programs. They suggested the effect was caused by lost social ties in

dingo groups and changes in the age structure of the dingo population, but how does a

juvenile dingo know livestock are prey? Livestock predation may result from the

collective influence of asocial learning, opportunities, and motivation (hunger c.f.

Baenninger 1978) but horizontal transmission cannot be overlooked. Heyes and Ray

(2000) indicated imitation behaviour was distinct from asocial and social learning.

Analogous with natural selection and behavioural adaptation, they suggested that

imitation behaviour incited cultural evolution which inherently caused imitation

behaviours.

Marcus (1989) and Parker (2007) stated that domestic working dogs are part of the

moral code and right of passage for livestock farmers in Australia. They serve many

functions, one of which is to herd livestock. Although herding behaviours may be

instinctive, research presented above implied that observational learning may be a

more instinctive behaviour than the imitation or action of behaviours in domestic dogs

and wolves. This herding task could be observed and learnt by dingoes, especially

juvenile loners with deceased parents, or others with limited social ties, as postulated


247

by Pongrácz et al. (2003). Considering dingoes are predators they may have innate

herding instincts and their cryptic habits are an indication that working dogs may not

detect their presence easily. Sorenson (1908), cited by Parker (2007), was a livestock

farmer who described “dingoes droving sheep”, which is consistent with imitation of

sheep herding behaviours reflected upon by Corbett (2001) and Parker (2007).

Current dingo “management” practices of creating a dispersal sink or “buffer zone”

(Thomson et al. 1992b) adjacent to livestock enterprises is the most likely cause for

observational learning by roaming dingoes because optimal habitat is made available

for habitation. Allen (pers. comm. 2008) showed, using satellite/GPS telemetry, that a

lone dispersing dingo rapidly inhabited a recently baited “buffer zone” in Queensland.

Imitation of domestic working dog herding behaviours by dingoes toward livestock

may be the reason behind some predation on livestock. Some dingoes may even be

attracted to livestock properties because of a female domestic dog in oestrus during

the breeding season.

Parker (2007) concluded that the confrontation between humans and dingoes exists

because of the flaws of humans. Australian frontier nationalism is one of those flaws,

a culturally inherited patriotic trait of many Australians (Marcus 1989). Apparently

the Australian bush was a place for men, being too scary and too dangerous for

women. Marcus (1989) suggested that this culture is the reason Lindy Chamberlain

was imprisoned and it is this same culture that imprisons dingoes. The dingo could

have taken Azaria, but intentionally or not, Azaria was taken to the dingo. In the

same sense, dingoes may predate livestock, but intentionally or not, livestock are

taken into areas that are traditional dingo habitat. These and other causes of conflict


248

between dingoes and humans are the responsibility of humans and not dingoes. This

understanding is vital for developing a novel approach for dingo management.

The current management system in the SGBMWHA is serviceable; however, dingoes

in some areas continue to arise as “problem” animals and a lobby for increased

control efforts is sustained. Rogue dingoes generally arise in the south east region of

the SGBMWHA where control efforts are more forceful as a result. Landholders are

regularly consulted through the Oberon Pest Animal Management Committee and

arising problems are rapidly identified, prioritised and managed. The below proposal

for improved management intends to promote coexistence between farming

commodities and the preservation of natural phenomena.

5.3 Proposal for management

It is now understood that hypercarnivorous canids require communal living to learn

behaviours. Chapter two showed the population in this study site consisted of closely

related genetic social units. The functional ecology of this population was revealed

through analysis of seasonal changes in activity and prey preference in chapter three

as a consequence of maintaining population structure in the core of this protected

area. Chapter four showed social groups are restricted to core areas of activity for ten

months of the year. Range expansion and extraterritorial forays occurred in two of

three months during breeding season but expansions generally remained within the

boundaries of the park. Seasonal variations in spatial organisation were consistent

with the seasonal variations in prey preference and patterns of activity, showing this

dingo population had a functional role in the SGBMWHA.


249

Destruction of native fauna due to community pressure does not appear to be in the

best interest to achieve the goals set by land management organisations for a

universally significant World Heritage Area. The best practice method of

management then is to cease attempting to “control” the dingo population because it is

self regulatory. Instead, management should review farming practices on properties

affected by dingoes adjacent to the GBMWHA to minimise interactions between

dingoes and livestock. Circumstances to be reviewed are those which may attract a

dingo to a property and those circumstances which may deter a dingo from a property.

If there are more attractants than deterrents then it is not a land management problem

but a farm management problem.

Attractants may include but are not limited to:

1. The presence of domestic dogs – Are they male/female and are they sterile?

How many are there? Are they restrained if they are not being worked?

2. The type, abundance and prevalence of preferred native fauna and water

resources on the property.

3. The type of livestock farmed and their housing. Are livestock left in paddocks

adjacent to bushland during peak dingo activity periods of a day or season?

4. Attractive free feeds and baiting programs.

5. Presence of the domestic dogs of dingo/wild dog trappers used on land

adjacent to the property.

6. Presence of a dispersal sink/buffer zone adjacent to the property.

7. Use of livestock and wildlife corpses, offal and household wastes.


250

Possible deterrents may include but are not limited to:

1. Removal of attractants for dingoes, identified by landscape managers.

2. Use of guard animals in livestock herds, such as alpacas or sterile guard dogs,

to ward off predators.

3. Moving vulnerable livestock into safer areas at night or during periods of high

dingo activity.

4. Use of electric fences and/or fladry tape on property perimeters or around

livestock safe zones.

In relation to data presented in the current study, this form of management can work

in two stages dependant on the movement patterns of local dingo groups. From June

until November, encompassing low periods of dingo activity (whelping and rearing

seasons) in the SGBMWHA, surrounding properties can be audited for dingo

attractants and deterrents. These can be ameliorated in time for the exploratory and

breeding season when range expansion of dingoes is more extensively documented.

The following six months (December – May) of higher dingo activity and stronger

social ties can then be monitored for comparison with the preceding six months.

However, during this period, livestock producers could alter methods of managing

their livestock by learning more about the biological seasons of dingoes in their area

and adopting practices and breeding regimes to reduce the risk of attack to livestock.

Moving the stock closer to the residence or into a dingo proof section of the enterprise

would be recommended at prescribed times of the year. Public property, including

school yards and campgrounds, on the contrary, will require more regular removal of

dingo attractants, especially food and litter.


251

Land management organisations would need to provide collaborating pastoral

enterprises with quality assurance that they meet the requirements of the audit. Any

subsequent loss of livestock should result in a consultation to ascertain reasons for

livestock losses, amending farm management practices if necessary and targeted

management (most likely euthanasia using a firearm) of the rogue dingo or dingoes

that have learnt to attack livestock. Using this method may reduce expenditure on

widespread dingo control campaigns, even if livestock producers were compensated

or subsidised for their livestock lost to dingo attacks. Baker et al. (2008) provided a

comprehensive review of predator management techniques and suggested the majority

of lost livestock in Australia were not due to predation but other biological factors

which also need to be accounted for during such audits.

Collaboration between scientists, farmers and land managers is integral to the success

of the management technique outlined, and has the ability to influence change. This

form of management has many levels and topics that cannot be entirely addressed by

the current study. The most important concept in contention is that many tiers of

culture will have to accommodate change and evolve to a new level. Fortunately this

cultural evolution can be based on objective scientific measurement for the

GBMWHA, and other areas where dingo populations occur, rather than on subjective

judgement, anecdote, and the resultant cultural transmission. Whole government

organisations will need to review and alter policy to suit the needs of the environment

in harmony with the needs of the populace. In the context of sustainability and

ecological sustainable development, the long term benefits will outweigh the short

term gains.


252

Parallels of natural resource management exist between Australia and America. The

recent Australian conservation values for dingoes appear to have changed as a

consequence of the American conservation values for wolves. Wolf conservation

techniques in the United States of America have recently changed. One case is the

reintroduction of wolves to Yellowstone National Park (YNP), Wyoming (White and

Garrott 1995). The ecology of YNP had been thrown off balance without wolves

because residing herbivores, such as elk Cervus elaphus and other deer species, were

increasing beyond stable population levels. Mech (1970) listed numerous factors

about the value of wolves in American ecosystems concluding that wolves are one of

the top down regulating mechanisms that maintain biological stability. Since

reintroduction, the wolf population has grown to the point that wolf management

needs to be re-evaluated due to increasing attacks on livestock on ranches adjoining

YNP. More to the point, a level of ecosystem stability was regained.

Dingo management in NSW seems to be influenced by the YNP wolf reintroduction

strategy. The submission for a second wild dog control order under the Rural Lands

Protection Act 1998 was internally generated and prepared by land management

organisations to declare some public lands in NSW important for the survival of

dingoes. This move marked the beginning of a cultural revolution recognising the

importance of the dingo to Australians and the biological stability of Australian

ecosystems. In the current structure of organisations charged with the responsibility

to manage Australian landscapes, there appears to be bifurcation between the decision

making process and the implementation of management decisions. The GBMWHA,

for instance, includes reserves spanning 1,032,649ha of universally significant

ecosystems and a global community of interest groups which have universal


253

ecological values. Local management can sometimes be diverted to servicing the

needs of local communities whilst attempting to maintain the values outlined by

decision makers. The goals of interest groups may conflict and adaptive local

management practices may outweigh long term strategic planning. On one front, the

globally iconic dingo is an essential component of the GBMWHA. On the local level,

dingoes are often seen as vermin that require control.

Lodgement of the second wild dog control order after the reintroduction of wolves to

YNP does not appear to be a coincidence. More so it appears to be a case of culture

being horizontally transmitted to Australia from America, demonstrating holistic

concern for the integrity of Australian ecosystems. The YNP study showed that

overpopulating ecosystems with herbivores not only has detrimental impacts on the

ecosystem, but also that detrimental impacts can be naturally controlled. The Murray-

Darling Basin (MDB) in Australia may be one such ecosystem. Dingoes have been

excluded from this basin since 1885 by the longest dingo proof fence in the world for

livestock farmers to maximise economic benefits from their livestock. Hatton et al.

(1993), Pierce et al. (1993) and Walker et al. (1993) highlighted some effects of

agriculture in the MDB and a federal commission has since been established to

manage the unstable ecosystem. The removal of the dingo fence to rejuvenate the

MDB has been suggested (Newsome 2001), however, in its present state with low

water flows and saline soils, a reduction in herbivore abundance would be ineffective

without landscape remediation work by humans. Rehabilitation of the natural

ecosystem and reduction of unsustainable agricultural practices should be a focus for

5.4 Conclusions


254

adaptive management of the MDB and other Australian landscapes. Such a strategy

may lead to reintroduction of dingoes to maintain biological stability (Glen and

Dickman 2005; Glen et al. 2007; Johnson et al. 2007; Wallach and O’Neill 2008).

Claridge and Hunt (2008) raised alternative arguments to Glen et al. (2007), regarding

dingo reintroduction, due to the changing nature of “wild dog” populations. Evidence

presented in this thesis showed “wild dog” populations may not actually be changing

in nature but landscape management practices should. Claridge and Hunt (2008) drew

on numerous sources that showed dingoes and/or “wild dogs”, in comparison with

foxes, had a negligible effect on populations of native Australian endangered species.

Johnson and Wroe (2003) also suggested alternative hypotheses to competition with

dingoes, which resulted in the exclusion of endemic Tasmanian species from the

mainland of Australia. The only apparent reason to control dingoes, therefore,

appears to be for the protection of livestock enterprises. Berger (2006) showed that

government-subsidised predator control failed to prevent the decline in the sheep

industry in the United States of America. Evidence presented instead suggested other

economic variables influenced the viability of livestock industries. Results presented

by Berger (2006) were therefore consistent with conclusions of the current study:

Predators are blamed for the faults of humans.

Regardless, the process of cultural evolution in Australia has commenced. Climate

change is seen as an impending doom and many Australians are seeking reasons why.

Steinfield et al. (2006) detailed how significant the impacts of livestock production

systems are on the global climate. Cultural traits, like those associated with climate

change, are subject to intra and intergenerational variations. Traits may conflict with


255

one another but favourable heritable variants accumulate and culturally dominate over

time (Borgerhoff Mulder et al. 2006). The land management organisations and

landholders of the GBMWHA can show by example that dingoes and contemporary

Australian people can coexist in a sustainable manner. It is highly recommended that

adaptive management to promote coexistence with dingoes be publicly trialled for the

GBMWHA. The example of dingoes may achieve other cultural shifts in thinking for

future management of Australian landscapes and ecosystems. Perhaps wild canids

such as dingoes and wolves will attain the same status as sharks internationally, so

that they can continue to perform their key functional role as top order predators in

unregulated ecosystems. Such a change in culture may be essential if pastoral

industries and natural ecological systems are to become sustainable in Australia.

References

256

Abrams, P. A. (1992) Adaptive foraging by predators as a cause of predator-prey cycles. Evolutionary Ecology, 6, 56-72.

6.0 References

Adams, C. E., Hamilton, D. J., Wilson, I. M. A. J., Grant, A., Alexander, G., Waldron, S., Snorasson, S. S., Ferguson, M. M. and Skúlason, S. (2006) Does breeding site fidelity drive phenotypic and genetic sub-structuring of a population of Arctic charr? Evolutionary Ecology Research, 20, 11-26.

Allen, B. L. (2006) The spatial ecology and zoonoses of urban dingoes - A preliminary investigation, University of Queensland, Honours Thesis.

Allen, B.L. and Miller, H.A. (2009) The biodiversity benefits and the production costs of dingoes in the arid zone: Summary of research results for 2008, South Australian Arid Lands Natural Resources Management Board, Port Augusta.

Allen, J. J., Bekoff, M. and Crabtree, R. L. (1999) An observational study of coyote (Canis latrans) scent-marking and territoriality in Yellowstone National Park. Ethology, 105, 289-302.

Allen, L. (2000) Measuring predator control effectiveness: reducing numbers may not reduce predator impact IN: Salmon, T. P. and Crabb, A. C. (Eds.) Proceedings of the 19th Vertebrate Pest Conference., University of California Davis, San Diego.

Allen, L., Engeman, R. M. and Kruper, H. (1996) Evaluation of three relative abundance indices for assessing dingo populations. Wildlife Research, 23, 197-206.

Allen, L. and Sparkes, E. C. (2001) The effect of dingo control on sheep and beef cattle in Queensland. Journal of Applied Ecology, 38, 76-87.

Allen, L. R. and Gonzalez, A. (1998) Bating reduces dingo numbers, changes age structures yet often increases calf losses. 11th Australian Vertebrate Pest Conference, Bunbury, Western Australia

Anon. (1978) Australian vertebrate pest control conference 1978: working papers, Australian Vertebrate Pest Control Conference Proceedings, A.C.T., Canberra.

References

257

Anon. (1995) Proposed Warragamba Flood Mitigation Dam Environmental Impact Statement, ERM Mitchell McCotter Pty. Ltd., Sydney.

Anon. (1996) Australia: State of the Environment 1996, Independent Report to the Commonwealth Minister for the Environment,

CSIRO Publishing, Canberra.

Anon. (2000) Submission for proposed amendments to the Rural Lands Protection Act 1998 Wild Dog Control, NSW National Parks and Wildlife Service, State Forests of NSW, Department of Land and Water Conservation and Sydney Catchment Authority, Sydney.

Anon. (2001a) Blue Mountains National Park Plan of Management, NSW National Parks and Wildlife Service, Hurstville.

Anon. (2001b) Kanangra-Boyd National Park Plan of Management, NSW National Parks and Wildlife Service, Hurstville.

Anon. (2001c) Nattai Reserves: Nattai National Park, Bargo State Recreation Area, Burragorang State Recreation Area, Nattai State Recreation Area, and Yerranderie State Recreation Area Plan of management, NSW National Parks and Wildlife Service, Hurstville.

Anon. (2001d) Australia State of the Environment 2001, Independent Report to the Commonwealth Minister for the Environment and Heritage, CSIRO Publishing on behalf of the Department of the Environment and Heritage, Canberra.

Anon. (2006) GPS tracking systems: state of the art GPS collars for wildlife tracking, wildlife monitoring systems, [Online] Available: http://www.environmental-studies.de/Argos/A-0.html [Accessed: 10.08.2006]

Anon. (2007) Australian Dingo, [Online] Available: http://www.ankc.org.au/home/breeds_details_print.asp?bid=103 [Accessed: 12.12.2007]

Anon. (2008) 2008 IUCN Red List of Threatened Species, Canis lupus ssp. dingo [Online] Available: www.iucnredlist.org [Accessed: 28.10.2008]

http://www.environmental-studies.de/Argos/A-0.html�

http://www.environmental-studies.de/Argos/A-0.html�

http://www.ankc.org.au/home/breeds_details_print.asp?bid=103�

http://www.iucnredlist.org/�

References

258

Asa, C. S., Mech, L. D. and Seal, U. S. (1985) The use of urine, faeces, and anal-gland secretions in scent-marking by a captive wolf (Canis lupus) pack. Animal Behaviour, 33, 1034-1036.

Ashe, A. and Whitelaw, E. (2006) Another role for RNA: a messenger across generations. TRENDS in Genetics, 23, 8-10.

Atchley, W. R., Logsdon, T., Cowley, D. E. and Eisen, E. J. (1991) Uterine effects, epigenetics, and postnatal skeletal development in the mouse Evolution, 45, 891-909.

Atkinson, R. P. D., Rhodes, C. J., Macdonald, D. W. and Anderson, R. M. (2002) Scale-free dynamics in the movement patterns of jackals Oikos, 98, 134-140.

Baenninger, R. (1978) Some aspects of predatory behaviour. Aggressive Behaviour, 4, 287-311.

Baker, P. J., Boitani, L., Harris, S., Saunders, G. and White, P. C. L. (2008) Terrestrial carnivores and human food production: impact and management. Mammal Review, 38, 123-166.

Bandeira de Melo, L. F., Lima Sábato, M. A., Vaz Magni, E. M., Young, R. J. and Coelho, C. M. (2007) Secret lives of maned wolves (Chrysocyon brachyurus Illiger 1815): As revealed by GPS tracking collars. Journal of Zoology, 271, 27-36.

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

Bannerman, S. M. (1990) Soil landscapes of the Penrith 1:100 000 sheet, Soil Conservation Service of NSW, Sydney.

Baverstock, P. and Green, B. (1975) Water recycling in lactation. Science, 187, 657-658.

Beeton R. J. S., Buckley K. I., Jones G. J., Morgan D., Reichelt R. E. and Trewin D. (2006), Australia State of the Environment 2006, Independent report to the Australian Government Minister for the Environment and Heritage, Department of the Environment and Heritage, Canberra.

References

259

Bekoff, M. and Wells, M. C. (1981) Behavioural budgeting by wild coyotes: The influence of food resources and social organization. Animal Behaviour, 29, 794-801.

Berger, K. (2006) Carnivore-Livestock Conflicts: Effects of subsidized predator control and economic correlates on the sheep industry. Conservation Biology, 20, 751-761.

Blaum, N., Engeman, R. M., Wasiolka, B. and Rossmanith, E. (2008) Indexing small mammalian carnivores in the southern Kalahari, South Africa. Wildlife Research, 35, 72-79.

Blewitt, M. E., Chong, S. and Whitelaw, E. (2004) How the mouse got its spots. TRENDS in Genetics, 20, 550-554.

Boitanni, L., Francisci, F., Ciucci, P. and Andreoli, G. (1995) Population biology and ecology of feral dogs in central Italy. IN: Serpell, J. and Barrett, P. (Eds.) The domestic dog: Its evolution, behaviour, and interactions with people, Cambridge University Press, Cambridge.

Borgerhoff Mulder, M., Nunn, C. L. and Towner, M. C. (2006) Cultural macroevolution and the transmission of traits. Evolutionary Anthropology, 15, 52-64.

Bowen, W. D. and Cowen, I. M. (1980) Scent marking in coyotes. Canadian Journal of Zoology, 58, 473-480.

Brunner, H., Amor, R. L. and Stevens, P. L. (1976) The use of predator scat analysis in a survey at Dartmouth in north-eastern Victoria Australian Wildlife Research, 3, 85-90.

Brunner, H., Triggs, B. and Ecobyte, P. L. (2002) Hair ID: An interactive tool for identifying Australian mammalian hair. CD ROM, CSIRO Publishing, Collingwood.

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

References

260

Burrows, N. D., Algar, D., Robinson, A. D., Sinagra, J., Ward, B. and Liddelow, G. (2003) Controlling introduced predators in the Gibson Desert of Western Australia. Journal of Arid Environments, 55, 691-713.

Burt, W. H. (1943) Territoriality and home range concepts as applied to mammals. Journal of Mammalogy, 24, 346-352.

Caplan, A. I. (1991) Mesenchymal stem cells Journal of Orthopaedic Research, 9, 641-650.

Carbone, C., Mace, G. M., Roberts, S. C. and Macdonald, D. W. (1999) Energetic constraints on the diet of terrestrial carnivores. Nature, 402, 286-288.

Carmichael, L. E., Krizan, J., Nagy, J. A., Dumond, M., Johnson, D., Veitch, A., Strobeck, C. (2008) Northwest passages: conservation genetics of Arctic Island wolves. Conservation Genetics, 9, 879-892.

Carr, G. M. and Macdonald, D. W. (1986) The sociality of solitary foragers: a model based on resource dispersion. Animal Behaviour, 34, 1540-1549.

Catling, P. C. (1979) Seasonal variation in plasma testosterone and the testis in captive male dingoes, Canis familiaris dingo. Australian Journal of Zoology, 27, 939-944.

Catling, P. C., Corbett, L. K., Westcott, M. (1991) Age determination in the dingo and crossbreeds. Wildlife research 18, 75-83.

Catling, P. C., Burt, R. J. and Forrester, R. I. (2000) Models of the distribution and abundance of ground-dwelling mammals in the eucalypt forests of north-eastern New South Wales in relation to habitat variables. Wildlife Research, 27, 639-654.

Catling, P. C., Reid, A. and Claridge, A. W. (2005) The Sand Plot Technique: Sampling Ground-Dwelling Vertebrates. NSW Department of Environment and Conservation and CSIRO Sustainable Ecosystems, Canberra.

Cavilli-Sforza, L. L. and Feldman, M. W. (1981) Cultural transmission and evolution: A quantitative approach. Princeton University Press, Princeton, N.J.

References

261

Chambers, G. K. and MacAvoy, E. S. (2000) Microsatellites: Consensus and controversy. Comparative Biochemistry and Physiology Part B, 126, 455-476.

Choi, K. C. and Jeung, E. B. (2003) The biomarker and endocrine disruptors in mammals. Journal of Reproduction and Development, 49, 337-345.

Ciucci, P., Boitani, L., Francisci, F. and Andreoli, G. (1997) Home range, activity and movements of a wolf pack in central Italy. Journal of Zoology, 243, 803-819.

Claridge, A. W. and Hunt, R. (2008) Evaluating the role of the dingo as a trophic regulator: Additional practical suggestions. Ecological Management and Restoratrion, 9, 116-119.

Claridge, A. W. and Mills, D. J. (2007) Aerial baiting for wild dogs has no observable impact on spotted-tailed quolls (Dasyurus maculatus) in a rainshadow woodland. Wildlife Research, 34, 116-124.

Claridge, A. W., Murray, A. J., Dawson, J., Poore, R., Mifsud, G. and Saxon, M. J. (2006) The propensity of spotted-tailed quolls (Dasyurus maculatus) to encounter and consume non-toxic meat baits in a simulated canid control program. Wildlife Research, 33, 85-91.

Coelho, C. M., Bandeira de Melo, L. F., Lima Sábato, M. A., Rizel, D. N. and Young, R. J. (2007) A note on the use of GPS collars to monitor wild maned wolves Chrysocyon brachyurus (Illiger 1815) (Mammalia, Canidae). Applied Animal Behaviour Science, 105, 259-264.

Colborn, T., Saal, F. S. V. and Soto, A. M. (1993) Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environmental Health Perspectives, 101, 378-384.

Corbett, L. K. (1988) Social dynamics of a captive dingo pack: population regulation by dominant female infanticide. Ethology, 78, 177-198.

Corbett, L. K. (2001) The dingo in Australia and Asia. J.B. Books Pty Ltd., Marleston.

Corbett, L. K. and Newsome, A. E. (1987) The feeding ecology of the dingo. III. Dietary relationships with widely fluctuating prey populations in arid Australia: an hypothesis of alternation of predation. Oecologia, 74, 215-227.

References

262

Craighead, J. J., Frank C. Craighead, J., Ruff, R. L. and O'Gara, B. W. (1973) Home range and activity patterns of non migratory elk of the Madison Drainage herd as determined by biotelemetry. Wildlife Monographs, 33, 1-50.

Creel, S. R. and Macdonald, D. W. (1995) Sociality, group size, and reproductive suppression among carnivores. Advances in the Study of Behaviour, 24, 203-257.

Crews, D., Willingham, E. and Skipper, J. K. (2000) Endocrine disruptors: present issues, future directions. The Quarterly Review of Biology, 75, 243-260.

Crompton, A. E., Obbard, M. E., Petersen, S. D. and Wilson, P. J. (2008) Population genetic structure in polar bears (Ursus maritimus) from Hudson Bay, Canada: implications of future climate change. Biological Conservation, 141, 2528-2539.

D'Eon, R. G., Serrouya, R., Smith, G. and Kochanny, C. O. (2002) GPS telemetry error and bias in mountainous terrain. Wildlife Society Bulletin, 30, 430-439.

Daniels, M. J. and Corbett, L. K. (2003) Redefining introgressed protected mammals: when is a wildcat a wild cat and a dingo a wild dog? Wildlife Research, 30, 213-218.

Davis, E. (2001) Legislative issues relating to control of dingoes and other wild dogs in New South Wales 1. Approaches to future Management. IN: Dickman, C. R. and Lunney, D. (Eds.) A Symposium on the Dingo. Royal Zoological Society of New South Wales, Mosman.

Di Orio, A. P., Callas, R. and Schaefer, R. J. (2003) Performance of two GPS telemetry collars under different habitat conditions. Wildlife Society Bulletin, 31, 372-379.

Dickman, C. R. and Lunney, D. (Eds.) (2001) A symposium on the Dingo, Royal Zoological Society of New South Wales, Mosman.

Dickman, C., Glen, A. and Letnic, M. (2009) Reintroducing the dingo: Can Australia’s conservation wastelands be restored? IN: Hayward, M.W. and Somers, M. J. (Eds.) Reintroduction of top-order predators, Blackwell Publishing, Oxford.

References

263

Edgar, J. P., Appleby, R. G. and Jones, D. N. (2007) Efficacy of an ultrasonic device as a deterrent to dingoes (Canis lupus dingo): a preliminary investigation. Journal of Ethology, 25, 209-213.

Edwards, M. G., de Prue, N. D., Shakeshaft, B. J. and Crealy, I. V. (2000) An evaluation of two methods of assessing feral cat and dingo abundance in central Australia. Wildlife Research, 27, 143-149.

Eldridge, S. R., Berman, D. M. and Walsh, B. (2000) Field evaluation of four 1080 baits for dingo control. Wildlife Research, 27, 495-500.

Eldridge, S. R., Shakeshaft, B. J. and Nano, T. J. (2002) The impact of wild dog control on cattle, native and introduced herbivores and introduced predators in central Australia, unpublished, Final administrative report to the Bureau of Rural Sciences.

Elledge, A. E., Leung, L. K.-P., Allen, L. R., Firestone, K. and Wilton, A. N. (2006) Assessing the taxonmic status of dingoes Canis familiaris dingo for conservation Mammal Review, 36, 142-156.

Evanno, G., Regnaut, S. and Goudet, J. (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology, 14, 2611-2620.

Falush, D., Stephens, M. and Pritchard, J. K. (2003) Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics, 164, 1567-1587.

Falush, D., Stephens, M. and Pritchard, J. K. (2007) Inference of population structure using multilocus genotype data: dominant markers and null alleles. Molecular Ecology, technical article, 1-5.

Faubet, P., Waples, R. S. and Gaggiotti, O. E. (2007) Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Molecular Ecology, 16, 1149-1166.

Fleming, P. J. S. (2001) Legislative issues relating to control of dingoes and other wild dogs in New South Wales II: historical and technical justifications for current policy. IN: Dickman, C. R. and Lunney, D. (Eds.) A Symposium on the Dingo, Royal Zoological Society of New South Wales, Mosman.

References

264

Fleming, P. J. S., Allen, L. R., Berghout, M. J., Meek, P. D., Pavlov, P. M., Stevens, P. L., Strong, K., Thompson, J. A. and Thomson, P. C. (1998) The performance of wild-canid traps in Australia: efficiency, selectivity and trap-related injuries. Wildlife Research, 25, 327-338.

Fleming, P. J. S., Allen, L. R., Lapidge, S. J., Robley, A., Saunders, G. R. and Thomson, P. C. (2006) A strategic approach to mitigating the impacts of wild canids: proposed activities of the Invasive Animals Cooperative Research Centre. Australian Journal of Experimental Agriculture, 46, 753-762.

Fleming, P. J. S., Corbett, L. K., Harden, R. and Thomson, P. C. (2001) Managing the impacts of dingoes and other wild dogs. Bureau of Rural Sciences, Canberra.

Fleming, P. J. S., Thompson, J. A. and Nicol, H. I. (1996) Indices for measuring the efficacy of aerial baiting for wild dog control in north-eastern New South Wales. Wildlife Research, 23, 665-674.

Francisco, L. V., Langston, A. A., Mellersh, C. S., Neal, C. L. and Ostrander, E. A. (1996) A class of highly polymorphic tetranucleotide repeats for canine genetic mapping. Mammalian Genome, 7, 359-362.

Frank, H. (1980) Evolution of canine information processing under conditions of natural and artificial selection. Zeitschrift für Tierpsychologie, 53, 389-399.

Freeman, C. (2005) Is this picture worth a thousand words? An analysis of Harry Burrell's photograph of a thylacine with a chicken. Australian Zoologist, 33, 1-16.

Friend, G. R. (1978) A Comparison of Predator Scat Analysis with Conventional Techniques in a Mammal Survey of Contrasting Habitats in Gippsland, Victoria. Australian Wildlife Research, 5, 75-83.

Gantz, G. F. and Knowlton, F. F. (2005) Seasonal activity areas of coyotes in the Bear River Mountains of Utah and Idaho. Journal of Wildlife Management, 69, 1652-1659

Geffen, E., Gompper, M. E., Gittleman, J. L., Luh, H.-K., MacDonald, D. W. and Wayne, R. K. (1996) Size, life-history traits, and social organization in the canidae: a re-evaluation. The American Naturalist, 147, 140-160.

References

265

Gese, E. M., Ruff, R. L. and Crabtree, R. L. (1996) Social and nutritional factors influencing the dispersal of resident coyotes. Animal Behaviour, 52, 1025-1043.

Gilbert, S. F. (2002) The genome in its ecological context: philosophical perspectives on interspecies epigenesis. Annals New York Academy of Sciences, 981, 202-218.

Girman, D. J., Mills, M. G. L., Geffen, E. and Wayne, R. K. (1997) A molecular genetic analysis of social structure, dispersal, and interpack relationships of the African wild dog (Lycaon pictus). Behavioural Ecology and Sociobiology, 40, 187-198.

Glen, A. S. and Dickman, C. R. (2005) Complex interactions among mammalian carnivores in Australia, and their implications for wildlife management Biological Review, 80, 1-15.

Glen, A. S., Dickman, C. R., Soulé, M. E. and Mackey, B. G. (2007) Evaluating the role of the dingo as a trophic regulator in Australian ecosystems. Austral Ecology 32, 492-501.

Goldberg, A. D., Allis, C. D. and Bernstein, E. (2007) Epigenetics: a landscape takes shape. Cell, 128, 635-638.

Goudet, J. (1995) FSTAT (vers. 1.2): a computer program to calculate F-statistics. Journal of Heredity, 86, 485-486.

Graves, T. A. and Waller, J. S. (2006) Understanding the causes of missed global positioning system telemetry fixes. Journal of Wildlife Management, 70, 844-851.

Green, B. F. and Catling, P. C. (1977) The biology of the dingo. IN: Messel, H. and Butler, S. T. (Eds.) Australian Animals and their Environment, Shakespeare Head Press, Sydney.

Green, K. (2002) Selective Predation on the broad-toothed rat, Mastacomys fuscus (Rodentia: Muridae), by the introduced red fox, Vulpes vulpes (Carnivora: Canidae), in the Snowy Mountains, Australia. Austral Ecology, 27, 353-359.

References

266

Green, K. and Osborne, W. S. (1981) The diet of foxes, Vulpes vulpes (L.), in relation to abundance of prey above the winter snowline in New South Wales. Australian Wildlife Research, 8, 349-360.

Griffiths, T. (2001) One hundred years of environmental crisis Australian Rangelands Journal 23, 5-14.

Harden, R. H. (1985) The ecology of the dingo in north-eastern New South Wales I: movements and home range. Australian Wildlife Research, 12, 25-37.

Hatton, T. J., Pierce, L. L. and Source, J. W. (1993) Ecohydrological changes in the Murray-Darling Basin II: development and tests of a water balance model. The Journal of Applied Ecology, 30, 274-282.

Hemson, G., Johnson, P., South, A., Kenward, R., Ripley, R. and McDonald, D. (2005) Are kernels the mustard? Data from global positioning system (GPS) collars suggests problems for kernel home-range analyses with least-squares cross-validation. Journal of Animal Ecology, 74, 455-463

Henikoff, S. and Matzke, M. A. (1997) Exploring and explaining epigenetic effects. TRENDS in Genetics, 13, 293-295.

Herring, S. W. (1993) Formation of the vertebrate face: epigenetic and functional influences. American Zoologist, 33, 472-483.

Heyes, C. M. and Ray, E. D. (2000) What is the significance of imitation in animals? Advances in the study of behaviour, 29, 215-245.

Holmes, N. G., Dickens, H. F., Parker, H. L., Binns, M. M., Mellersh, C. S. and Sampson, J. (1995) Eighteen canine microsatellites. Animal Genetics, 26, 132-133.

Holmes, N. G., Mellersh, C. S., Humphreys, S. J., Binns, M. M., Holliman, J., Curtis, R. and Sampson, J. (1993) Isolation and characterization of microsatellites from the canine genome. Animal Genetics, 24, 289-292.

Hulst, F. (2008) Dingoes IN: Vogelnest, L. and Woods, R. (Eds.) Medicine of Australian Mammals CSIRO Publishing, Collingwood.

References

267

Johnson C. N. and VanDerWal J. (2009) Evidence that dingoes limit abundance of a mesopredator in eastern Australian forests Journal of Applied Ecology, 46, 641–646

Johnson, C. N. and Wroe, S. (2003) Causes of extinction of vertebrates during the Holocene of mainland Australia: arrival of the dingo, or human impact? The Holocene 13, 941-948.

Johnson, C. N., Isaac, J. L. and Fisher, D. O. (2006) Rarity of a top predator triggers continent-wide collapse of mammal prey: dingoes and marsupials in Australia. Proceedings of the Royal Society B: Biological Sciences, 274, 341-346.

Jones, E. (1990) Physical Characteristics and taxonomic status of wild canids, Canis familiaris, from the eastern highlands of Victoria. Australian Wildlife Research, 17, 69-81.

Jones, E. (2009) Hybridisation between the dingo, Canis lupus dingo, and the domestic dog, Canis lupus familiaris, in Victoria: a critical review Australian Mammalogy, 31, 1-7

Jones, E. and Stevens, P. L. (1988) Reproduction in wild canids, Canis familiaris, in the eastern highlands of Victoria. Australian Wildlife Research, 15, 385-394.

Kelce, W. R., Gray, L. E. and Wilson, E. M. (1998) Antiandrogens as environmental endocrine disruptors. Reproduction, Fertility and Development, 10, 105-111.

Kenward, R. E. (2001) A manual for wildlife radio tagging, Academic Press, London.

Kiernan, K., Jones, R. and Ranson, D. (1983) New evidence from Fraser Cave for glacial age man in south-west Tasmania Nature 301, 28 - 32

Kindall, J. L. and Van Manen, F. T. (2007) Identifying habitat linkages for American black bears in North Carolina, USA. Journal of Wildlife Management, 71, 487-495.

King, D. P. (1994) Soil landscapes of the Katoomba 1:100 000 sheet (Mount Tomah, Woodford, Jenolan Caves, Hampton). Department of Conservation and Land Management, Sydney.

References

268

Kleiman, D. (1968) Reproduction in the Canidae. International Zoo Yearbook, 8, 3-8.

Kohen, J. L. (1995) Aboriginal Environmental Impacts University of New South Wales Press, Sydney.

Körtner, G. and Watson, P. (2005) The immediate impact of 1080 aerial baiting to control wild dogs on a spotted-tailed quoll population. Wildlife Research, 32, 673-680

Koskinen, M. T. and Bredbacka, P. (2000) Assessment of the population structure of five Finnish dog breeds with microsatellites. Animal Genetics, 31, 310-317.

Laundré, J. W. and Keller, B. L. (1981) Home-range use by coyotes in Idaho. Animal Behaviour, 29, 449-461.

Lehman, N., Clarkson, P., Mech, L. D., Meier, T. J. and Wayne, R. K. (1992) A study of the genetic relationships within and among wolf packs using DNA fingerprinting and mitochondrial DNA. Behavioural Ecology and Sociobiology, 30, 83-94.

Leonard, J. A., Wayne, R. K., Wheeler, J., Valadez, R., Guillén, S. and Vilà, C. (2002) Ancient DNA evidence for old world origin of new world dogs. Science, 298, 1613-1616.

Lindström, E. (1986) Territory inheritance and the evolution of group-living in carnivores. Animal Behaviour, 34, 1825-1835.

Lunney, D. (2001) Causes of the extinction of native mammals of the western division of New South Wales: An ecological interpretation of the nineteenth century historical record Australian Rangelands Journal 23, 44-70.

Lunney, D., Triggs, B., Eby, P. and Ashby, E. (1990) Analysis of scats of dogs Canis familiaris and foxes Vulpes vulpes (Canidae: Carnivora) in coastal forests near Bega, New South Wales. Australian Wildlife Research, 17, 61-68.

Lyon, M. F. (1993) Epigenetic inheritance in mammals. Trends in Genetics, 9, 123-128.

References

269

Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids, Oxford University Press, Oxford.

Macdonald, D. W., Creel, S. and Mills, M. G. L. (2004) Society. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids, Oxford University Press, Oxford.

Macdonald, D. W. and Sillero-Zubiri, C. (2004a) Conservation. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids, Oxford University Press, Oxford.

Macdonald, D. W. and Sillero-Zubiri, C. (2004b) Dramatis personae. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canid, Oxford University Press, Oxford.

Macqueen, A. (1993) Blue Mountains to Bridgetown: The Life and Journeys of Barrallier 1773—1853. Published by the author, Springwood, Australia.

Marcus, J. (1989) Prisoner of discourse: the dingo, the dog and the baby. Anthropology Today, 5, 15-19.

Marsack, P. and Campbell, G. (1990) Feeding behaviour and diet of dingoes in the Nullarbor Region, Western Australia. Australian Wildlife Research, 17, 349-357.

May, S. A. and Norton, T. W. (1996) Influence of fragmentation and disturbance on the potential impact of feral predators on native fauna in Australian forest ecosystems. Wildlife Research, 23, 387-400.

Macintosh, N. W. G. (1976) The origin of the dingo: an enigma IN: Fox, M. W. (Ed.) The Wild Canids: Their Systematics, Behavioural Ecology and Evolution Dogwise Publishing, Washington.

Mech, L. D. (1970) The Wolf: The ecology and behaviour of an endangered species. The Natural History Press, New York.

Mech, L. D. and Barber, S. M. (2002) A critique of wildlife radio-tracking and its use in national parks. Northern Prairie Wildlife Research Centre and University of Minnesota, Jamestown.

References

270

Meek, P. D. (1999) The movement, roaming behaviour and home range of free-roaming domestic dogs, Canis lupus familiaris, in coastal New South Wales. Wildlife Research, 26, 847-856.

Meek, P. D., Jenkins, D. J., Morris, B., Ardler, A. J. and Hawksby, R. J. (1995) Use

of two human leg-hold traps for catching pest species. Wildlife Research, 22, 733-739.

Meek, P. D. and Triggs, B. (1998) The Food of foxes, dogs and cats on two peninsulas in Jervis Bay, New South Wales. Proceedings of the Linnean Society NSW, 120, 117-127.

Mellersh, C., Holmes, N., Binn, M. and Sampson, J. (1994) Dinucleotide repeat polymorphisms at four canine loci (LEI 003, LEI 007, LEI 008 and LEI 015). Animal Genetics, 25, 125.

Mellersh, C. S., Langston, A. A., Acland, G. M., Fleming, M. A., Ray, K., Wiegand, N. A., Francisco, L. V., Gibbs, M., Aguirre, G. D. and Ostrander, E. A. (1997) A linkage map of the canine genome. Genomics, 46, 326-336.

Merrill, S. B. and Mech, L. D. (2003) The usefulness of GPS telemetry to study wolf circadian and social activity. Wildlife Society Bulletin, 31, 947-960.

Messier, F. and Barrette, C. (1982) The social system of the coyote (Canis latrans) in a forested habitat Canadian Journal of Zoology, 60, 1743-1753.

Mitchell, B. D. and Banks, P. B. (2005) Do wild dogs exclude foxes? Evidence for competition from dietary and spatial overlaps. Austral Ecology, 30, 581-591.

Moehlman, P. D. (1986) Ecology of cooperation in canids. IN: Rubinstein, D. I. and Wrangham, R. W. (Eds.) Ecological aspects of social evolution: birds and mammals. Princeton University Press, Princeton, N.J.

Moehlman, P. D. (1989) Intraspecific variation in canid social systems. IN: Gittleman, J. L. (Ed.) Carnivore behaviour, ecology and evolution. Cornell University Press, Ithaca, N.Y.

Moehlman, P. D. and Hofer, H. (1997) Cooperative breeding, reproductive suppression and body mass in canids. IN: Solomon, N. G. and French, J. A. (Eds.) Cooperative breeding in mammals. Cambridge University Press, Cambridge, U.K.

References

271

Monk, M. (1990) Variation in epigenetic inheritance. TRENDS in Genetics, 6, 110-114.

Montagu, M. F. A. (1942) On the origin of the domestication of the dog. Science, 96, 111-112.

Morey, D. F. (1992) Size, shape and development in the evolution of the domestic dog. Journal of Archaelogical Science, 19, 181-204.

Nakada, M. and Lambeck, K. (1989) Late Pleistocene and Holocene sea-level change in the Australian region and mantle rheology. Geophysical Journal 96. 497-517

Nesbitt, B., Wilton, A. N. and Jenkins, D. J. (2000) Dingoes of the New England and Guy Fawkes National Parks of north east New South Wales - are there any left out there? IN: Balough, S. (Ed.) NSW Pest Animal Control Conference - Practical Solutions to Pest Management Problems. Orange, NSW.

Nesbitt, W. H. (1975) Ecology of a feral dog pack on a wildlife refuge. IN: Fox, M. W. (Ed.) The Wild Canids, Van Nostrand Reinhold, New York.

Newsome, A. E. (2001) The biology and ecology of the dingo IN: Dickman, C. R. and Lunney, D. (Eds.) A Symposium on the Dingo., The Royal Zoological Society of New South Wales, Mosman.

Newsome, A. E., Catling, P. C. and Corbett, L. K. (1983a) The feeding ecology of the dingo II. Dietary and numerical relationships with fluctuating prey populations in south-eastern Australia Australian Journal of Ecology, 8, 345-366.

Newsome, A. E. and Coman, B. J. (1989) Canidae. IN: Dalton, D. W. and Richardson, B. J. (Eds.) Fauna of Australia. Mammalia, Australian Government Publishing Service, Canberra.

Newsome, A. E. and Corbett, L. K. (1982) The identity of the dingo II.* hybridization with domestic dogs in captivity and in the wild. Australian Journal of Zoology, 30, 365-374.

Newsome, A. E. and Corbett, L. K. (1985) The identity of the dingo III.* the incidence of dingoes, dogs and hybrids and their coat colours in remote and settled regions of Australia. Australian Journal of Zoology, 33, 363-375.

References

272

Newsome, A. E., Corbett, L. K. and Carpenter, S. M. (1980) The identity of the dingo I. morphological discriminants of dingo and dog skulls. Australian Journal of Zoology, 28, 615-625.

Newsome, A. E., Corbett, L. K., Catling, P. C. and Burt, R. J. (1983b) The feeding ecology of the dingo. I. stomach contents from trapping in south-eastern Australia, and non-target wildlife also caught in dingo traps. Australian Wildlife Research, 10, 477-486.

Oakman, B. (2001) The problem with keeping dingoes as pets and conservation. IN: Dickman, C. R. and Lunney, D. (Eds.) A Symposium on the Dingo, The Royal Zoological Society of New South Wales, Mosman.

Ostrander, E. A., Sprague Jr., G. F. and Rine, J. (1993b) Identification and characterization of dinucleotide repeat (CA)n

markers for genetic mapping in dog. Genomics, 16, 207-213.

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

Parker, H. G., Kim, L. V., Sutter, N. B., Carlson, S., Lorentzen, T. D., Malek, T. B., Johnson, G. S., DeFrance, H. B., Ostrander, E. A. and Kruglyak, L. (2004) Genetic structure of the purebred domestic dog. Science, 304, 1160-1163.

Parker, M. (2007) The cunning dingo. Society and Animals, 15, 69-78.

Parker, M. A. (2006) Bringing the dingo home: discursive representations of the dingo by Aboriginal, colonial and contemporary Australians. PhD Thesis, University of Tasmania.

Pierce, L. L., Walker, J., Dowling, T. I., McVicar, T. R., Hatton, T. J., Running, S. W. and Coughlan, J. C. (1993) Ecohydrological changes in the Murray-Darling Basin. III . a simulation of regional hydrological changes. The Journal of Applied Ecology, 30, 283-294.

Pongrácz, P., Miklósi, Á., Kubinyi, E., Topáli, J. and Csányi, V. (2003) Interaction between individual experience and social learning in dogs. Animal Behaviour, 65, 595-603.

References

273

Pritchard, J. K., Stephens, M. and Donnelly, P. (2000a) Inference of population structure using multilocus genotype data. Genetics, 155, 945-959.

Pritchard, J. K., Stephens, M., Rosenberg, N. A. and Donnelly, P. (2000b) Association mapping in structured populations. American Journal of Human Genetics, 67, 170-181.

Purcell, B. V., Mulley, R., Close, R. and Fleming, P. (2006) Use of GPS collars for tracking wild dogs. Queensland Pest Animal Symposium Proceedings, Toowoomba.

Quinn, M. (2001) Rights to the rangelands: European contests of possession in the early 20th

century Australian Rangelands Journal 23: 5-14.

Rakyan, V. K. and Beck S. (2006) Epigenetic variation and inheritance in mammals Current Opinion in Genetics and Development. 16, 573-577.

Rakyan, V. and Whitelaw, E. (2003) Transgenerational epigenetic inheritance. Current Biology, 13.

Ritchie, E. G. and Johnson, C. N. (2009) Predator interactions, mesopredator release and biodiversity conservation, Ecology Letters 12, 1–18

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

Robertshaw, J. D. and Harden, R. H. (1986) The ecology of the dingo in north-eastern New South Wales IV. Prey selection by dingoes, and its effect on the major prey species, the swamp wallaby, Wallabia bicolor (Desmarest). Australian Wildlife Research, 13, 141-163.

Roemer, I., Reik, W., Dean, W. and Klose, J. (1997) Epigenetic inheritance in the mouse. Current Biology, 7, 277-280.

Rosen, S. (1995) Losing Ground: An Environmental History of the Hawkesbury-Nepean Catchment. Hale and Iremonger Pty. Ltd., Sydney

Ruvinsky, A. (2001) Developmental Genetics. IN: Ruvinsky, A. and Sampson, J. (Eds.) The Genetics of the Dog. CABI Publishing, Wallingford.

References

274

Samuel, M. D. and Green, R. E. (1988) A revised test procedure for identifying core areas within the home range. The Journal of Animal Ecology, 57, 1067-1068.

Savolainen, P., Leitner, T., Wilton, A. N., Matisoo-Smith, E. and Lundeberg, J. (2004) A detailed picture of the origin of the Australian dingo, obtained from the study of mitochondrial DNA. Proceedings of the National Academy of Sciences of the United States of America, 101, 12387-12390.

Savolainen, P., Zhang, Y., Luo, J., Lundeberg, J. and Leitner, T. (2002) Genetic evidence for an east Asian origin of domestic dogs. Science, 298, 1610-1613.

Scott, J. P. (1968) Evolution and domestication of the dog. Evolutionary Biology. 2, 243-75.

Shepherd, N. C. (1981) Predation of red kangaroos, Macropus rufus, by the dingo, Canis familiaris dingo (Blumenbach), in north-western New South Wales. Australian Wildlife Research, 8, 255-262.

Sillero-Zubiri, C. and Macdonald, D. W. (1998) Scent-marking and territorial behaviour of Ethiopian wolves Canis simensis Journal of Zoology, 351-361.

Sillero-Zubiri, C., Reynolds, J. and Novaro, A. J. (2004) Management. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids. Oxford University Press, Oxford.

Slabbert, J. M. and Rasa, O. A. E. (1997) Observational learning an acquired maternal behaviour pattern by working dog pups: an alternative training method. Applied Animal Behaviour Science, 53, 309-316.

Sponenberg, D. P. and Rothschild, M. F. (2001) Genetics of Coat Colour and Hair Texture. IN Ruvinsky, A. and Sampson, J. (Eds.) The Genetics of the Dog. CABI Publishing, Wallingford.

Steinfield, H., Gerber, P., Wassenaar, T., Catel, V., Rosales, M. and de Haan, C. (2006) Livestock's long shadow: environmental issues and options. Food and agriculture organisation of the United Nations, Rome.

Thompson, J. A., Fleming, P. J. S. and Heap, E. W. (1990) The accuracy of aerial baiting for wild dog control in New South Wales. Australian Wildlife Research, 17, 209-217.

References

275

Thomson, P. C. (1986) The effectiveness of aerial baiting for the control of dingoes in north-western Australia. Australian Wildlife Research, 13, 165-176.

Thomson, P. C. (1992a) The behavioural ecology of dingoes in north-western Australia. I. The Fortescue River study area and details of captured dingoes. Wildlife Research, 19, 509-518.

Thomson, P. C. (1992b) The behavioural ecology of dingoes in north-western Australia. II. Activity patterns, breeding season and pup rearing. Wildlife Research, 19, 519-530.

Thomson, P. C. (1992c) The behavioural ecology of dingoes in north-western Australia. III. Hunting and feeding behaviour, and diet. Wildlife Research, 19, 531-541.

Thomson, P. C. (1992d) The behavioural ecology of dingoes in north-western Australia. IV. Social and spatial organisation, and movements. Wildlife Research, 19, 543-563.

Thomson, P. C. and Rose, K. (1992) Age determination of dingoes from characteristics of canine teeth. Wildlife Research, 19, 597-599.

Thomson, P. C., Rose, K. and Kok, N. E. (1992a) The behavioural ecology of dingoes in north-western Australia. V. Population dynamics and variation in the social system. Wildlife Research, 19, 565-584.

Thomson, P. C., Rose, K. and Kok, N. E. (1992b) The behavioural ecology of dingoes in north-western Australia. VI. Temporary extraterritorial movements and dispersal. Wildlife Research, 19, 585-595.

Triggs, B. (1996) Tracks, scats and other traces: a field guide to Australian mammals. Oxford University Press, Melbourne.

Trut, L. N. (2001) Experimental studies of early canid domestication. IN: Ruvinsky, A. and Sampson, J. (Eds.) The Genetics of the Dog. CABI Publishing, Wallingford.

Trut, L. N., Plyusnina, I. Z. and Oskina, I. N. (2004) An experiment on fox domestication and debatable issues of evolution of the dog. Russian Journal of Genetics, 40, 644-655.

References

276

Twigg, L. E., Eldridge, S. R., Edwards, G. P., Shakeshaft, B. J., dePrue, N. D. and Adams, N. (2000) The longevity and efficacy of 1080 meat baits used for dingo control in central Australia. Wildlife Research, 27, 473-481.

Van Ballenberghe, V. (1983) Two litters raised in one year by a wolf pack. Journal of Mammalogy, 64, 171-173.

Van Speybroeck, L., De Waele, D. and Van De Vijver, G. (2002) Theories in early embryology: close connections between epigenesis, preformationism, and self-organization. Annals of the New York Academy of Sciences, 981, 7-49.

Van Valkenburgh, B. (2007) Déjà vu: the evolution of feeding morphologies in the Carnivora. Itegrative and Comparative Biology, 47, 147-163.

Vilà, C., Savolainen, P., Maldonado, J. E., Amorim, I. R., Rice, J. E., Honeycutt, R. L., Crandall, K. A., Lundeberg, J. and Wayne, R. K. (1997) Multiple and ancient origins of the domestic dog. Science, 276, 1687-1689.

Walker, J., Bullen, F. and Williams, B. G. (1993) Ecohydrological changes in the Murray-Darling Basin. I. the number of trees cleared over two centuries The Journal of Applied Ecology, 30, 265-273.

Wallach, A. D., Murray, B. and O’Neill, A. J. (2008) Can threatened species survive where the top predator is absent? Biological conservation, 142, 43–52

Wallach, A. D. and O’Neill, A. J. (2008) Persistence of endangered species: is the dingo the key? Final report to Department of Environment and Heritage, Adelaide.

Wang, X., Tedford, R. H., Van Valkenburgh, B. and Wayne, R. K. (2004) Ancestry. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids. Oxford University Press, Oxford.

Wayne, R. K. (1986) Cranial morphology of domestic and wild canids: the influence of development on morphological change. Evolution, 40, 243-261.

Wayne, R. K. (2001) Consequences of domestication: morphological diversity of the dog. IN: Ruvinsky, A. and Sampson, J. (Eds.) The Genetics of the Dog. CABI Publishing, Wallingford.

References

277

Wayne, R. K., Geffen, E., Girmin, D. J., Koepfli, K. P., Lau L.M., Marshall C. (1997) Molecular systematics of the Canidae. Systematic Biology, 4, 622–653.

Wayne, R. K., Geffen, E. and Vilà, C. (2004) Population genetics. IN: Macdonald, D. W. and Sillero-Zubiri, C. (Eds.) Biology and Conservation of Wild Canids. Oxford University Press, Oxford.

Wayne, R. K. and Ostrander, E. A. (1999) Origin, genetic diversity, and genome structure of the domestic dog. BioEssays, 21, 247-257.

Wayne, R. K. and Ostrander, E. A. (2007) Lessons learned from the dog genome. TRENDS in Genetics, 23, 557-567.

Wayne, R. K. and Vilà, C. (2001) Phylogeny and Origin of the Domestic Dog. IN: Ruvinsky, A. and Sampson, J. (Eds.) The Genetics of the Dog. CABI Publishing, Wallingford.

White, P. J. and Garrott, R. A. (1995) Northern Yellowstone elk after wolf restoration. Wildlife Society Bulletin, 33, 942-955.

Whitehouse, S. J. O. (1977) The diet of the dingo in Western Australia. Australian Wildlife Research, 4, 145-150.

Wilton, A. N. (2001) DNA methods of assessing dingo purity. IN: Dickman, C. R. and Lunney, D. (Eds.) A Symposium on the Dingo. Royal Zoological Society of New South Wales, Mosman.

Wilton, A. N., Steward, D. J. and Zafiris, K. (1999) Microsatellite variation in the Australian dingo. Journal of Heredity, 90, 108-111.

Woodall, P. F., Pavlov, P. and Twyford, K. L. (1996) Dingoes in Queensland, Australia: skull dimensions and the identity of wild canids. Wildlife Research, 23, 581-587.

Wroe, S., Clausen, P., McHenry, C., Moreno, K. and Cunningham, E. (2007) Computer simulation of feeding behaviour in the thylacine and dingo as a novel test for convergence and niche overlap. Proceedings of the Royal Society B: Biological Sciences, 274, 2819-2828

References

278

Yates, D., Hayes, G., Heffernan, M. and Beynon, R. (2003) Incidence of cryptorchidism in dogs and cats. The Veterinary Record, 152, 502-504.

Zacharewski, T. (1998) Identification and assessment of endocrine disruptors: limitations of in vivo and in vitro assays. Environmental Health Perspectives, 106, 577-582.

Zrzavý, J. and Ricánková, V. (2004) Phylogeny of the recent Canidae (Mammalia, Carnivora): relative reliability and utility of morphological and molecular datasets. Zoologica Scripta, 33, 311-333.

Date post:	24-Oct-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Order in the Pack - Western Sydney

Documents