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
Initialisms and abbreviations
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
Initialisms and abbreviations
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
Initialisms and abbreviations
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
Initialisms and abbreviations
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
Initialisms and abbreviations
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.
Dingo ecology in the Southern Greater Blue Mountains World Heritage Area
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
Dingo ecology in the Southern Greater Blue Mountains World Heritage Area
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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).
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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)
Dingo ecology, study site and research framework
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).
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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).
Dingo ecology, study site and research framework
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%).
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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).
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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).
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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.
Dingo ecology, study site and research framework
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?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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)
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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)
What is a “pure” dingo?
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.
What is a “pure” dingo?
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).
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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,
What is a “pure” dingo?
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.
What is a “pure” dingo?
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,
What is a “pure” dingo?
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).
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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).
What is a “pure” dingo?
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,
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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.
What is a “pure” dingo?
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.
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
What is a “pure” dingo?
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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,
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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)
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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)
Interactions between sympatric competitors and prey
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).
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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
Interactions between sympatric competitors and prey
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.
Interactions between sympatric competitors and prey
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;
Interactions between sympatric competitors and prey
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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)
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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%
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.)
Spatial organisation, movements and activity
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).
Spatial organisation, movements and activity
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.
Spatial organisation, movements and 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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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-
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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
Spatial organisation, movements and activity
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.
Spatial organisation, movements and activity
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
Effects of cultural transmission
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
Effects of cultural transmission
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.
Effects of cultural transmission
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
Effects of 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
Effects of cultural transmission
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.
Effects of cultural transmission
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
Effects of cultural transmission
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
Effects of cultural transmission
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
Effects of cultural transmission
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.
Effects of cultural transmission
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.
Effects of cultural transmission
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.
Effects of cultural transmission
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.
Effects of cultural transmission
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
Effects of cultural transmission
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
Effects of cultural transmission
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
Effects of cultural transmission
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]
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.