Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Spatial ecology of a top-order marine predator,
the tiger shark (Galeocerdo cuvier)
by
Luciana Cerqueira Ferreira
BSc Univ. Federal do Rio Grande, MSc Univ. Federal de Pernambuco
This thesis is presented for the degree of
Doctor of Philosophy of
The University of Western Australia
Faculty of Science
School of Animal Biology
2017
iii
Declaration of authorship
I declare that this thesis is my own composition, all sources have been acknowledged and my
contribution is clearly identified in the thesis. This thesis has been substantially completed
during the course of enrolment at the University of Western Australia and has not previously
been accepted for a degree at any tertiary education institution. I confirm that for any work in
this thesis that has been co-published with other authors, I have the permission of all co-authors
to include this work in my thesis, and include a signed declaration.
Luciana Cerqueira Ferreira
Date: 12/05/2017
v
Abstract
Top-order predators are large consumers that occupy the highest trophic level of food-
webs. Through both consumptive and behavioural impacts on prey, these animals play
important roles in the structure, stability and resilience of communities and ecosystems.
In marine environments, top-order predators are often highly mobile and can display
large-scale (1000s km) seasonal migrations. This poses a major challenge for
conservation of these animals, because common approaches, such as the establishment
of marine protected areas, rarely encompass the scale and extent of their movement
patterns. Furthermore, broad-scale migrations make it likely that top-order predators
will encounter multiple human threats across their range and as a consequence,
populations of many species have undergone rapid declines. Identification of the key
environmental drivers of movement patterns and habitat use by top-order predators is
thus urgently required in order to understand the scale and consequences of
anthropogenic threats that these sharks now face. Here, I address these issues for the
tiger shark (Galeocerdo cuvier), a keystone top-order consumer in tropical and warm-
temperate marine ecosystems.
In Chapter 2, I reviewed the conservation status of tiger shark according to the
guidelines for the International Union for the Conservation of Nature (IUCN) Red List.
I summarised information available on its biology, ecology and threats to populations,
and indicated gaps in the ecological knowledge of the species. This assessment formally
updated the status of the species for this species as “Near Threatened” IUCN Red List.
In Chapter 3, I described and analyse spatial patterns in movement of tiger sharks along
the coastline of Western Australia as derived from satellite telemetry. Using a Brownian
Bridge movement kernel method to analyse tracks, I showed that the extensive
movements of tiger sharks (thousands of km) can encompass tropical and cool-
temperate environments across international maritime boundaries that offer varying
levels of fisheries protection for the species. Individual variability was high with respect
to levels of residency, space use and environmental preferences in each climate zone.
In order to describe spatial variability in diet and trophic ecology of tiger sharks, I used
the analysis of stable isotopes from multiple tissues (dermis, muscle, blood) of sharks
collected from both the western and eastern coasts of Australia (Chapter 4). This
continental-scale analysis of isotopic signatures demonstrated that the species resides
vi
near the top of the food chain, but is a generalist with a very broad diet. I showed that
tiger sharks explored multiple resources throughout their distribution that led to changes
in the position of the species within the food chain in different regions. Finally, in
Chapter 5, I compiled and analysed a global dataset of tiger shark tracks across multiple
ocean basins to identify the key environmental drivers of movement and habitat use at
the scale of the range of the entire species. I applied Biased Random Bridge methods to
generate monthly utilization distributions for all individual sharks and overcome issues
of data quality and sparseness, and then applied generalised additive mixed-models to
identify the key drivers of migratory movements and large-scale patterns of space
utilisation. Sea surface temperature and water depth were important predictors of space
utilization for all sizes and sexes of sharks and at all locations, but patterns varied
greatly at both temporal and spatial scales. Furthermore, these environmental predictors
explained only small amounts of variance in habitat utilisation. Geographically
weighted regression was used to identify areas within the species range where
environmental variables failed to explain space utilisation, allowing the generation of
alternative hypotheses to explain patterns of habitat use and the spatial targeting of
future research.
Combined, these findings demonstrated the value of using large datasets pooled across
studies and locations to provide insights into the spatial ecology of wide-ranging marine
predators. My analyses showed that tiger sharks are cosmopolitan and generalist
predators that demonstrate high levels of intra-specific variation in their use of space,
movement and trophic ecology. This plasticity in spatial and trophic ecology is possibly
a key element of their success as a top-order marine predator and might provide some
resilience to human threats. However, it also highlights the difficulties involved in
establishing effective conservation strategies in order to manage and protect the species.
vii
Acknowledgements
A sincere and heartfelt thank you goes to my supervisors Dr Mark Meekan, Dr Michele
Thums and Prof Jessica Meeuwig, for their support and guidance. Prof Jessica thanks
for giving me the opportunity to pursue this work as part of your team and sharing you
insights into marine conservation. Mark, thank you for your indispensable mentorship,
inputs and edits to my work, excitement about my results, motivational chats about
science and life as scientist, incentive to share my science and to attend as many
conferences as I could, and for pushing me to “finish that paper!”. Michele, I cannot
begin to thank the incredible help and guidance you have given me throughout this
process. Thanks for sharing your expertise in quantitative modelling, animal and
movement ecology, R and stats, but most importantly for the endless support and
sincere friendship. Your supervision has helped me develop into the scientist I am today
and I deeply appreciate the opportunities you and Mark have given me.
Many collaborators and co-authors have had a great input in this thesis. I am
enormously grateful for the help and contribution of Dr Ben Radford in developing the
analytical approach for my last chapter, his inputs and unending enthusiasm about my
research, which was a much needed motivation during data analysis. I am grateful to Dr
Becky Fisher for always finding time to help me with my many R and statistics
questions. I am also grateful to my collaborator and co-authors Dr Rory McAuley, Dr
Michael Heithaus, Dr Fabio Hazin, Dr John Stevens, Dr Neil Hammerschlag, Dr Adam
Barnett, Dr Bonnie Holmes, Dr André Afonso, Dr Kátya Abrantes, Dr Brad Wetherbee,
Dr Mahmood Shivji, Dr Robert Nowicki, Dr Jeremy Vaudo, Dr Derek Burkholder and
Dr Julian Pepperell, for providing tracking data and samples, facilitating access to lab
equipment and valuable suggestions for my papers. Thanks to Dr Michele Thums, Dr
Mark Meekan, Dr Ana Sequeira, Dr Carlos Duarte, Dr Graeme Hays, and all
participants of the Marine Megafauna Analytical Program. Thanks also to Dr Colin
Simpfendorfer and Dr Michelle Heupel, for the support in the preparation of the IUCN
assessment of tiger sharks, facilitating my attendance in the Australian Shark Specialist
Group (SSG) workshop, for providing ideas and samples, and always finding time to
chat about my PhD. My most sincere gratitude goes to the Graduate Research
Coordinators in the School of Animal Biology Prof Leigh Simmons and Prof Joseph
Tomkins for your wise advice and support during challenging moments in the PhD.
viii
I would not have been able to produce this thesis without the generous support of the
institutions that have provided funding and field assistance in obtaining data for this
research. My gratitude goes to the Conselho Nacional de Desenvolvimento Científico e
Tecnológico for my PhD scholarship, and the UWA Oceans Institute and Australia
Institute of Marine Science for providing a top-up scholarship. I am also grateful to the
Royal Zoological Society of New South Wales for providing funding through the Paddy
Pallin Science Grant and to Save Our Seas, the Department of Economic Development
and Tourism of KwaZulu-Natal South Africa, the American Elasmobranch Society, the
Postgraduate Student Association at UWA, the Graduate Research Office (UWA) and
the Australian Institute of Marine Science for travel support. Thanks to OCEARCH, in
special to Chris Fischer and Fernanda Ubatuba, for the field work support and tags, and
the hard-working crew that made the WA Ningaloo Expedition a unique experience.
Receiver data was sourced though the Australian Animal Tracking and Monitoring
Systems (AATAMs). The WA Department of Fisheries provided tracking data,
laboratory space and technical support for some of my analysis.
Thanks to my dear friends and fellow marine geeks Bev, Joyce, Phillipa (aka Pipi),
Tom, Gabe, Milly, Dani, Al, Conrad, Charli, Mike, Sammy and Emily. Thank for the
fun times and the many chats about our research that would have bored anyone outside
the “PhD world”. Pipi, Milly and Sammy, I am truly grateful for all your help with the
formatting of this thesis, you are rock stars! Bev, my fieldwork buddy and coffee addict,
going through this process with you made it so much easier and fun. Joyce, thank you
for always trying to keep me levelled-head and providing sensible advice when I needed
and being such a great company during the final months of PhD. Pipi, thanks for
making my days so much more fun, for the daily doses of uncontrollable laughter and
for helping me get through the craziness of the final stages of thesis writing. You are an
awesome office mate and very dear friend. I will never forget the great moments I had
living in the Subi house with my two big brothers, Tom and Gabe. Thanks for being
awesome housemates and true friends.
Many thanks to all my non-marine-scientist friends Laura, Paddy, Lily, Tash, Phil,
Renee, Siobhan, Nick, Jackson, Mike, Faye and Robin, for always reminding me that
life is not always about sharks. Laura, huge thanks for your support and friendship, and
for all the hilarious moments, blue parties and home-cooked meals. To my oldest group
of amazing friends Marina, Caio, João, Cris and Dai, I’ve known you for a life-time and
no matter where my crazy profession took me you have been the most loyal of friends.
ix
Thanks for always understanding (sometimes a bit reluctantly) that I could not always
be present, for your unconditional friendship and always being there through all the
good and rough times.
Profound thanks to everyone at the Perth AIMS office for your support, assistance and
encouragement. Olwyn, I am immensely thankful for your constant support and for
always making sure I was ok and that things were running smoothly with the PhD, and
for helping to fix them when they weren’t. Kim Brooks, thanks for your unique sense of
humour, for bringing me on the whale shark trip and honest friendship. Most sincere
thanks to Louise, Kat and Dr James Gilmore for the constant support, entertaining
conversations, and always taking the time to check if I was OK when the PhD was
stressful.
Finally, an unmeasurable thank you to my family, my mom Lucia, my dad Marcos, my
sister Dea and my brother-in-law Rafa (and obviously all the furry family members).
None of this would be possible without your love and support. You have always
allowed me to dream big and reminded me that you believe I can reach all my goals,
even when it seemed too hard. Thanks for encouraging me to pursue my dreams despite
having to endure the “saudades” and long distance. I hope I have made you proud. I am
also thankful for all the support my grandparents have given me, in particular my
grandpa Hiram. “Vô”, you helped me develop the thirst for knowledge that has driven
me to seek a career in science, and I will miss you forever.
xi
Publication from this thesis
Chapter 2
Ferreira LC, Simpfendorfer CA. In review. Galeocerdo cuvier, Tiger shark. The IUCN
Red List of Threatened Species.
Chapter 3
Ferreira LC, Thums M, Meeuwig JJ, Vianna GMS, Stevens J, McAuley R, Meekan MG
(2015) Crossing latitudes—long-distance tracking of an apex predator. Plos One, 10(2):
e0116916.
Chapter 4
Ferreira LC, Thums M, Heithaus MR, Barnett A, Abrantes K, Holmes B, Zamora LM,
Frisch AJ, Pepperell J, Burkholder D, Vaudo J, Nowicki R, Meeuwig J, Meekan MG. In
Review. The trophic role of a large marine predator, the tiger shark Galeocerdo cuvier.
Scientific Reports.
Chapter 5
Ferreira LC et al. (In preparation) Scale and habitat-dependent movement behaviour of
a large marine predator, the tiger shark Galeocerdo cuvier. Ecology.
xiii
Statement of candidate contributions
This thesis is presented as a series of four manuscripts in journal formats, as well as the
general introduction and general discussion sections. These papers were developed by
my own ideas and hypotheses with inputs from my supervisors.
Fieldwork for Chapter 4 and 5 was supported by the Paddy Pallin Science Grant (Royal
Zoological Society of NSW), The University of Western Australia, the Australian
Institute of Marine Science and OCEARCH. Tuition stipend was provided by the
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Brazil) and a
RCA Top-up scholarship by The UWA Oceans Institute and Australian Institute of
Marine Science (AIMS).
Chapter 2 was facilitated by Dr Colin Simpfendorfer (James Cook University) and the
IUCN Shark Specialist Group (SSG) that supported travel to attend the SSG Australian
Regional Workshop. Existing tracking data for Chapter 3 was provided by Dr Mark
Meekan (Australian Institute of Marine Science), Dr Rory McAuley (Department of
Fisheries, Government of Western Australia) and Dr John Stevens (CSIRO,
Commonwealth Scientific and Industrial Research Organisation).
I collected samples at Ningaloo Reef for Chapter 4. Samples from other locations were
obtained from Dr Adam Barnett (James Cook University), Dr Bonnie Holmes
(University of Queensland), Dr Julian Pepperell (Pepperell Research) and the
Queensland Shark Control Program. Dr Michael Heithaus (Florida International
University) provided existing isotope data for Shark Bay. I deployed tags at Ningaloo
Reef for Chapter 5 and existing tag data was provided by the by Dr Mark Meekan, Dr
Rory McAuley, Dr John Stevens, Dr Neil Hammerschlag (University of Miami), Dr
Fabio Hazin (Universidade Federal Rural de Pernambuco), Dr Richard Fitzpatrick
(Biopixel) and Dr Michael Heithaus.
The analyses described in all data chapters were carried by myself and all chapters were
written by me with feedback from Prof Jessica Meeuwig (University of Western
Australia, Chapters 1-3, 6), Dr Mark Meekan (all chapters), Dr Michele Thums (all
chapters), Dr Colin Simpfendorfer (Chapter 2), Dr Gabriel Vianna (Rare Brasil, Chapter
3), Dr Rory McAuley (Chapter 3), Dr John Stevens (Chapter 3), and Dr Ben Radford
(AIMS, Chapter 5)
xv
Co-author authorisation
By signing below, co-authors agree to the listed publication being included in the
candidate’s thesis and acknowledge that the candidate is the primary author, i.e.
contributed greater than 50% of the content and was primarily responsible for the
planning, execution and preparation of the work for publication.
Publication title: Crossing latitudes—long-distance tracking of an apex predator.
Dr. Michele Thums
(Australian Institute of Marine Science)
Prof Jessica Meeuwig
(University of Western Australia)
Dr Gabriel Vianna
(Rare Brasil)
Dr John Stevens
(CSIRO Marine and Atmospheric
Research)
Dr Rory McAuley
(Western Australia Department of
Fisheries)
Dr Mark Meekan
(Australian Institute of Marine Science)
Publication title: The trophic role of a large marine predator, the tiger shark
Galeocerdo cuvier (In Review)
Dr. Mark Meekan, signed in name of co-authors
(Australian Institute of Marine Science)
xvii
Table of Contents
Declaration of authorship ................................................................................................ iii
Abstract…………………… ....................................................................... …………….v
Acknowledgements ........................................................................................................ vii
Publication from this thesis ............................................................................................. xi
Statement of candidate contributions ............................................................................ xiii
Co-author authorisation .................................................................................................. xv
Table of Contents ......................................................................................................... xvii
List of Figures ............................................................................................................... xxi
List of Tables................................................................................................................ xxv
Chapter 1 General Introduction ................................................................................. 27
1.1 Top-order predators .................................................................................. 27
1.2 Movement ecology ................................................................................... 30
1.3 Challenges for the conservation of large marine predators ...................... 35
1.4 The tiger shark .......................................................................................... 37
1.5 Aims .......................................................................................................... 43
1.6 Thesis outline ............................................................................................ 43
Chapter 2 Review of tiger shark ecology and conservation status ............................ 45
2.1 Introduction ............................................................................................... 45
2.2 Distribution ............................................................................................... 45
2.3 Population trends ...................................................................................... 46
2.4 Life-history and ecology ........................................................................... 51
2.5 Threats ...................................................................................................... 53
2.5.1 Shark Fisheries .......................................................................................... 54
2.5.2 Non-shark commercial fisheries ............................................................... 55
2.5.3 Recreational fisheries ................................................................................ 56
2.5.4 Artisanal and IUU Fisheries ..................................................................... 56
2.5.5 Shark Control Programs ............................................................................ 57
2.6 Assessment Rationale ............................................................................... 58
Chapter 3 Crossing latitudes - long-distance tracking of an apex predator ............... 61
3.1 Abstract ..................................................................................................... 61
3.2 Introduction ............................................................................................... 61
xviii
3.3 Material and Methods ............................................................................... 64
3.4 Results ....................................................................................................... 67
3.5 Discussion ................................................................................................. 74
3.6 Acknowledgments .................................................................................... 78
3.7 Supporting information ............................................................................. 79
Chapter 4 Diet and trophic role of a large marine predator, the tiger shark
Galeocerdo cuvier .................................................................................... 81
4.1 Abstract ..................................................................................................... 81
4.2 Introduction ............................................................................................... 82
4.3 Methods .................................................................................................... 84
4.3.1 Sample collection ...................................................................................... 84
4.3.2 Data analysis ............................................................................................. 87
4.4 Results ....................................................................................................... 91
4.4.1 Comparisons of stable isotopic signatures among locations .................... 91
4.4.2 Comparisons of isotopic signatures within locations ............................... 93
4.4.3 Comparison of tissues among locations ................................................... 96
4.4.4 Diet stability ............................................................................................ 102
4.5 Discussion ............................................................................................... 103
4.6 Acknowledgements ................................................................................. 109
4.7 Supporting information ........................................................................... 110
Chapter 5 Scale and habitat-dependent movement behaviour of a large marine
predator, the tiger shark Galeocerdo cuvier ........................................... 111
5.1 Abstract ................................................................................................... 111
5.2 Introduction ............................................................................................. 112
5.3 Methods .................................................................................................. 114
5.3.1 Telemetry ................................................................................................ 114
5.3.2 Environmental data ................................................................................. 116
5.3.3 Identification of areas of high use .......................................................... 116
5.3.4 Identification of migratory behaviour ..................................................... 117
5.3.5 Modelling movement behaviour ............................................................. 118
5.3.6 Assessment of spatial variability in model predictions .......................... 121
5.4 Results ..................................................................................................... 122
5.4.1 Drivers of high use of habitats ................................................................ 126
5.4.2 Drivers of migration ............................................................................... 127
5.4.3 Assessment of spatial variability in model predictions .......................... 131
5.5 Discussion ............................................................................................... 134
5.6 Acknowledgements ................................................................................. 138
xix
5.7 Supporting information ........................................................................... 139
Chapter 6 General Discussion ................................................................................. 145
6.1 Variability in movement behaviour ........................................................ 145
6.2 Drivers of movement and habitat use ..................................................... 146
6.3 Low predictability of models .................................................................. 147
6.4 Implications for conservation ................................................................. 151
6.5 Future directions ..................................................................................... 153
6.6 Concluding remarks ................................................................................ 155
Bibliography… ............................................................................................................. 157
Appendix A… .............................................................................................................. 191
Appendix B… .............................................................................................................. 192
Appendix C… .............................................................................................................. 193
xxi
List of Figures
Figure 1.1 The tiger shark. ............................................................................................ 38
Figure 2.1 Distribution range of tiger sharks (blue). ..................................................... 46
Figure 2.2 Relative abundance (calculated as an index of standardised catch) for
the Shark Bottom Longline Observer Program (SBLOP, blue
diamonds) and longline survey of the Northeast Fisheries Science
Centre (NEFSC, green triangles) and SEFSC Longline (MS longline,
red squares). Figure produced by J. Carlson (pers. comm.). ..................... 48
Figure 2.3 CPUE (number of sharks gear-1
day-1
) by year and size (small < 3 m
and large > 3 m) of sharks for drumlines (A) and nets (B) by the
QSCP between 1993 and 2010. Figure extracted from Holmes et al.
(2012). ....................................................................................................... 49
Figure 2.4 Locations of the Queensland Shark Control Program with details of
gear (1993-2010), from Holmes et al. (2012). .......................................... 50
Figure 2.5 Catch per unit of effort of tiger sharks by the New South Wales Shark
Meshing Program from 1950-2010 modified from Reid et al. (2011).
Dashed line indicates Loess estimates. ..................................................... 51
Figure 2.6 Annual catch per unit of effort of tiger sharks in nets from the
KwaZulu-Natal beach protection program between 1978 and 2003.
Dash line represents the slope of a linear regression. Figure adapted
form Dudley & Simpfendorfer (2006). ..................................................... 51
Figure 2.7 Summary table of the criteria (A - E) used to evaluate Red List status
(IUCN 2012). ............................................................................................ 60
Figure 3.1 Movement patterns of tiger sharks in Western Australia. Maps show
location uplinks of 8 tiger sharks. Triangles indicate tagging location
and grey polygons indicate Commonwealth Marine Reserves. ................ 68
Figure 3.2 Home range of tiger sharks. The utilisation distribution calculated
using the Brownian Bridge kernel method. Black line represents the
50% Brownian Bridge home range distribution. ...................................... 69
Figure 3.3 Time spent in each temperature bin. Plots show the percentage of time
spent within the specified temperature bins for each tiger shark (ID
number in the top right hand corner of each plot). ................................... 71
Figure 3.4 . Latitudinal variation of temperature profiles for Shark 5. Percentage
of time spent in each temperature bin at each region of Western
Australia for Shark 5. ................................................................................ 72
Figure 3.5 Time spent in each depth bin. Percentage of time spent in each depth
bin for two tiger sharks tagged with SPLASH tags. ................................. 72
Figure 3.6 Vertical profiles of water temperature. Plots show temperature profiles
recorded by Argo floats of the north (17.6°S 117.6°E) and south
(35.78°S 119.5°E) sections of the WA coast. Vertical lines represent
the minimum temperature reported by the shark’s satellite
transmitter and horizontal lines represent the estimated diving depth. ..... 72
xxii
Figure 3.7 Generalised linear model for Shark 5. Generalised linear model (GLM)
predicted probabilities (solid line) of a shark being in a 25%
utilisation distribution in relation to (a) water depth, (b) water
temperature in the north and (c) in the south. Dotted lines show the
standard error. ........................................................................................... 73
Figure 4.1 Map showing the study locations across Australia and sampling sites
within each region in colour coded triangles. Circles represent
relative and total sample size for each dataset. GBR = Great Barrier
Reef, QSCP = Queensland Shark Control Program, NSW = New
South Wales. Map was created with ArcGis 10.3
(http://www.esri.com/). ............................................................................. 85
Figure 4.2. Mean and standard deviation stable isotopic composition of each
tissue for each of the locations sampled (SB= Shark Bay, GBR =
Great Barrier Reef, NSW =New South Wales, QSCP = Queensland
Shark Control Program, SA = South Africa) for A) slow tissues
(muscle, dermis, fin) and B) multiple tissues (muscle, dermis, fin,
red blood cells, plasma), for sites where multiple tissues were
collected in this study. Data from GBR1, Reunion Island and South
Africa obtained from published studies (Hussey et al. 2015b; Frisch
et al. 2016; Trystram et al. 2016) ............................................................. 92
Figure 4.3. Mean isotope values (± SD) of slow turnover tissues (muscle, dermis,
fin) of tiger sharks, other predators, consumers and primary
producers at Ningaloo Reef, Shark Bay, Great Barrier Reef and New
South Wales. Community data for each location was obtained from
Davenport & Bax (2002), Arthur et al. (2008), Revill et al. (2009)
Vaudo & Heithaus (2011), Speed et al. (2012), Wyatt et al. (2012),
Thomson et al. (2012), Heithaus et al. (2013), Frisch et al. (2014,
2016), Marcus, unpublished data. Red = sharks, orange = rays,
blue/golden = teleosts, purple = turtles, magenta = penguins, grey =
seals, maroon = cephalopods, light green = producers, dark green =
zooplankton. .............................................................................................. 95
Figure 4.4. Muscle. Standard ellipse areas for analysis of muscle tissue corrected
for sample size (SEAc) (A) and Bayesian Standard ellipse areas
(SEAb) (B), representing the 50, 75 and 95% credible intervals in
decreasing order of box size, with SEAb mode indicated by a black
circle and SEAc by a triangle. Partial dependence plots of the
relationship between δ13
C and the explanatory variables in the top-
ranked model, total length (C) latitude and sex and their interaction
(D). Points represent residuals and shaded areas show the 95%
confidence intervals. ................................................................................. 97
Figure 4.5. Dermis. Standard ellipse areas for analysis of dermal tissue corrected
for sample size (SEAc) (A) and Bayesian Standard ellipse areas
(SEAb) (B), representing the 50, 75 and 95% credible intervals in
decreasing order of box size, with SEAb mode indicated by a black
circle and SEAc by a triangle. Partial dependence plots of the
relationship between δ13
C (C) and δ15
N (D), and the explanatory
variables in the top-ranked models; total length (C) and location (D).
Points represent residuals and shaded areas show the 95%
confidence intervals. ................................................................................. 98
xxiii
Figure 4.6. Red Blood Cells. Standard ellipse areas for analysis of red blood cells
corrected for sample size (SEAc) (A) and Bayesian Standard ellipse
areas (SEAb) (B), representing the 50, 75 and 95% credible
intervals in decreasing order of box size, with SEAb mode indicated
by a black circle and SEAc by a triangle. Partial dependence plots of
the relationship between δ13
C and the explanatory variables in the
top-ranked model; total length (C) and location (D). Points represent
residuals and shaded areas show the 95% confidence intervals. ............ 100
Figure 4.7. Plasma. Standard ellipse areas for analysis of plasma corrected for
sample size (SEAc) (A) and Bayesian Standard ellipse areas
(SEAb)(B), representing the 50, 75 and 95% credible intervals in
decreasing order of box size, with SEAb mode indicated by a black
circle and SEAc by a triangle. Partial dependence plots of the
relationship between δ13
C and the explanatory variables in the top-
ranked model; total length and location. Points represent and shaded
areas show the 95% confidence intervals respectively. .......................... 102
Figure 4.8. Relationship between δ13
C values of paired tissues of fast (plasma)
and intermediate (red blood cells) turnover rates (A). Individuals
with a difference between tissues (calculated from resident sharks)
larger than mean differences Δ13
C are highlighted in filled symbols.
Open symbols represent individuals in steady state (stable). Solid
line represents the relationship between paired tissues and hashed
lines enclose the area within ± mean Δ13
C. Partial dependence plots
for the relationship between the stability of diet (probability of
steady state) and the explanatory variable in the top-ranked model;
location (B). Numbers indicate sample sizes. ......................................... 103
Figure 5.1 Telemetry data included in the analyses, colour coded by individual
shark tag ID, from Hawaii (Pacific Ocean) (top left), North Atlantic
(top right), South Atlantic (left), Great Barrier Reef, Australia
(Pacific Ocean) (middle) and Indian Ocean, Australia (bottom right). .. 120
Figure 5.2 Utilisation densities combined and averaged for each month for each
grid cell for all individuals in each regional dataset and filtered to the
0.75UD. Colour scale represents the average utilisation of each cell
and grey contours indicate the 500 m isobath. ........................................ 123
Figure 5.3 Conditional plots of female utilisation density relative to the
explanatory variables in the top ranked mixed model for each region
from the suite of models used to explain the response variable. Solid
curves are the model fit and shaded areas indicate the 95%
confidence intervals. Tick marks on the x-axis indicate the
distribution of observations. Plots a, c, e, g are fits for bathymetry for
North Atlantic, Indian Ocean, Pacific (Great Barrier Reef) and South
Atlantic, respectively; b, d, f, h are fits for sea surface temperature
for North Atlantic, Indian Ocean, Pacific Ocean (Hawaii) and South
Atlantic, respectively; i-j are fits for standard deviation of elevation
and aspect in the North Atlantic. ............................................................ 130
Figure 5.4 Conditional plots of male utilisation density relative to the explanatory
variables in the top ranked mixed model from the suite of models
used to explain the response variable. Solid curves are the model fit
and shaded areas indicate the 95% confidence intervals. Tick marks
on the x-axis indicate the distribution of observations. .......................... 131
xxiv
Figure 5.5 Conditional plots of probability of migration relative to the
explanatory variables in the top ranked mixed model from the suite
of models used to explain the response variable. Solid curves are the
model fit and shaded areas indicate the 95% confidence intervals.
Tick marks on the x-axis indicate the distribution of observations. ....... 132
Figure 5.6 GWR-derived local coefficient estimates for the relationship between
predicted and observed values from the top ranked mixed model
explaining the drivers of habitat use for females. A-B = North
Atlantic Ocean and B is zoomed in on the box in A, C-D = Indian
Ocean and D is zoomed in on the box in C, E = Pacific Ocean
around Hawaii, F = Great Barrier Reef, G = South Atlantic. Grey
isolines represent bathymetry. Red and orange show areas where
observed UD was higher than predicted (see detail in B and D) and
blue and green show area where observed UD was lower than
predicted. ................................................................................................ 133
Figure 5.7 GWR-derived local coefficient estimates for the relationship between
predicted and observed values from the top ranked mixed model
explaining the drivers of habitat use for males. A-B = North Atlantic
Ocean, B = Indian Ocean, C = Great Barrier Reef, D = South
Atlantic. Grey isolines represent bathymetry. Red and orange show
areas where observed UD was higher than predicted and blue and
green show area where observed UD was lower than predicted. ........... 134
Figure S 3.1. Sea Surface Temperature map (smoothed OISST data) for 16th May
2009 with location uplinks for Shark 5. .................................................... 79
Figure S 4.1. Location of acoustic receivers (grey and black circles) around
Ningaloo Reef, Australia. Black circles represent receivers where
tiger sharks were detected. Star indicates tagging location. ................... 110
Figure S 5.1 Representative example of a tiger shark track (black points) analysed
with Biased Random Bridges showing both residency (points within
the 0.75 UD (red lines) and migration (points outside the 0.75 UD). .... 143
Figure S 5.2 Figure extracted from: A) (Bouchet 2015) indicating hotspots in fish
abundance off the coast of Western Australia, B)Figure re extracted
from (Double et al. 2014) showing areas of high occupancy by
pygmy blue whales off the coast of Western Australia. ......................... 143
xxv
List of Tables
Table 1.1 Range and drivers of movement patterns displayed by large epipelagic
top-order sharks and teleosts. .................................................................... 34
Table 1.2 Tiger shark tagging programs with the number of tags deployed, sex
ratio of animals tagged, size range of sharks tagged and information
provided by telemetry devices. ................................................................. 41
Table 2.1 Summary of long-term trends of catch rates of tiger sharks based on
regional datasets. ....................................................................................... 47
Table 2.2 Average CPUE(number of sharks gear-1
day-1 x1000) for drumlines
from the QSCP for 1993 and 2010 from Holmes et al. (2012) and
projected CPUE for three generation lengths (Projection 1 =
generation length of 17.5 years, Projection 2 = generation length of
22.5 years). ................................................................................................ 49
Table 3.1. Details of satellite transmitter deployments on tiger sharks. ....................... 66
Table 3.2 Ranked Generalised Linear Models with bootstrap sampling of the
probability of a shark being in a 25% utilisation distribution
explained by bathymetry (Depth) and region of the coast (Region). ........ 70
Table 3.3 Ranked Generalised Linear Models with bootstrap sampling of the
probability of a shark being in a 25% utilisation distribution
explained by sea surface temperature (Temperature) at the north and
south coast separately. .............................................................................. 70
Table 4.1. Mean and standard deviation (SD) stable isotopic composition of tiger
sharks sampled in Queensland (QSCP), New South Wales (NSW)
the Great Barrier Reef (GBR) and Ningaloo and Shark Bay for each
sex (F = female, M= male). δ15
N and δ13
C given in parts per mill
(‰). N= number of samples, TL = mean total length (cm) ± SD. ........... 86
Table 4.2. Isotopic niche area described by Standard Ellipse Area SEAc (‰),
Bayesian SEAb and parameters; N= number of samples, TA = total
area of the convex hull, CD = mean centroid distance and standard
deviation, E = eccentricity, θ = angle of ellipse and CI = confidence
interval. F = Females, M = Males, QLD = Queensland Shark Control
Program and GBR, GBR = Great Barrier Reef and NSW = New
South Wales. ............................................................................................. 94
Table 4.3. Ranked generalised linear models of δ13
C and δ15
N for each tissue. The
top six models for each response are shown; values in bold indicate
the top ranked model (or the most parsimonious model). Degrees of
freedom (df), Sample corrected Akaike’s Information Criterion
(AICc), change in AICc relative to the model with the lowest AICc
value (ΔAICc), relative AICc weight (wAICc) and percent deviance
explained (DE%). Lat = latitude, TL = total length (cm). ........................ 99
xxvi
Table 4.4. Ranked binomial generalised linear models of stability of diet for
sharks (stable with the local food web), using paired tissue values of
δ13
C from red blood cells (RBC)-plasma comparisons. All models
are shown, values in bold indicate the most parsimonious model.
Sample corrected Akaike’s Information Criterion (AICc), change in
AICc relative to the model with the lowest AICc value (ΔAICc),
relative AICc weight (wAICc) and percent deviance explained
(DE%). TL= total length (cm). ............................................................... 103
Table 5.1 Description of all environmental, temporal and biological predictors for
the analysis of monthly utilisation distribution. ..................................... 124
Table 5.2 Predictors used to model migratory movements of tiger sharks. ................ 124
Table 5.3 Summary tracking datasets including information of tagging location,
number of tags deployed, track duration and size range of sharks
tagged. Mean values and SD are presented for total length (TL) and
track duration. ......................................................................................... 125
Table 5.4 The top ranked generalised additive mixed models from the suite of
models used to explain monthly utilisation distributions of female
and male tiger sharks. SST = sea surface temperature, Std = standard
deviation of elevation. Shown are the percent deviance explained
(%DE), Akaike’s information criterion (AIC), and AIC weight
(wAIC). ................................................................................................... 127
Table 5.5 The top three ranked generalised additive mixed models with 1000
bootstraps sampling for the suite of models used to explain the
probability of migration of tiger sharks. Shown for each model are
the median, 25th
and 75th
quantile Akaike’s information criterion
corrected for small samples (AICc), AICc weight (wAICc) and the
adjusted R-squared (Adj R2). The top ranked model is in bold. ............. 129
Table S 4.1 Details of tiger sharks tagged at Ningaloo Reef in April-May 2015,
their duration of detection (until last receiver download) by receivers
stations at Ningaloo and their behavioural classification (resident or
non-resident). .......................................................................................... 110
Table S 5.1 Details of each of the deployments of telemetry devices on tiger
sharks. ..................................................................................................... 139
27
Chapter 1 General Introduction
Top-order predators play keystone roles in the structure, stability and resilience of
communities and ecosystems (Mills et al. 1993) by both consuming prey (Tegner &
Levin 1983; Werner et al. 1983; Estes & Duggins 1995; Creel & Christianson 2008)
and influencing their behaviour (Lima & Dill 1990; Schmitz et al. 1997; Lima 1998;
Kie 1999; Estes et al. 2001; Wirsing & Ripple 2011). Such predators are often capable
of moving long distances (100-1000s km; Harestad & Bunnel 1979; Hays & Scott
2013), creating linkages between multiple ecosystems (McCauley et al. 2012). This
movement also brings them into contact with multiple human threats such as habitat
loss and fragmentation, pollution and overexploitation (Ritchie & Johnson 2009; Heupel
et al. 2014). As a result, top-order predators have been disproportionately and
negatively affected by anthropogenic impacts in both terrestrial and marine
environments. The removal of top-order predators can result in major shifts in the
structure and function of communities, reducing the resilience of ecosystems to human
impacts and climate change (Heithaus et al. 2008a, 2014; Beschta & Ripple 2009; Estes
et al. 2011; Ruppert et al. 2013; Atwood et al. 2015).
In this chapter, I examine the role of top-order predators in the trophic ecology of
ecosystems and show why the study of movement ecology is essential for the
development of an understanding of the ecology of these species. I describe the current
research challenges for these highly mobile predators and outline the rationale for my
choice of study species for this thesis.
1.1 Top-order predators
The terms “apex” and “top-order predator” encompass carnivores that occupy the
highest level of the trophic ladder (Estes et al. 2001; Wallach et al. 2015). These
animals are often large-bodied and specialized hunters (Ritchie & Johnson 2009), invest
in K-selected reproductive strategies and parental care, have large territories and display
complex social behaviours (Wallach et al. 2015). In terrestrial systems, carnivorous
mammals typically occupy these roles. Tigers (Panthera tigris), for example, are large,
solitary, territorial carnivores that are specialized forest-edge hunters of large ungulates
and preferentially predate on wild boar (Sus scrofa) and sambar deer (Rusa unicolor)
(Smith et al. 1989; Hayward et al. 2012). The species displays high investment in a low
28
numbers of offspring, with litters of females averaging only 1 - 5 cubs that are
dependent on their mother for up to two years (Sankhala 1967; Singh et al. 2013).
Carnivorous mammals can also act as top-order predators in some marine ecosystems.
In the Antarctic, leopard seals (Hydrurga leptonyx) are strongly territorial, solitary
predators whose predation on Adelie penguins (Pygoscelis adeliae) is a significant
structuring force of penguin colonies (Ainley et al. 2005). However, in marine
environments, top-order roles are more often occupied by fishes, notably sharks,
billfishes and tunas. These taxa differ from their mammalian counterparts by having
greater reproductive output (for example, the broadcast spawning of many hundreds of
thousands of tiny, pelagic eggs in tunas) and having little or no investment in parental
care. They also exhibit ontogenetic shifts in diet, habitat use and behaviour that
complicate their classification as top-order predators throughout their life cycle. For
example, white sharks (Carcharodon carcharias) use coastal habitats and have a diet of
fish as juveniles and are susceptible to predation by other species (e.g. orca, Orcinus
orca; Pyle et al. 1999). As adults, they target marine mammals and use offshore habitats
(Bruce et al. 2006; Estrada et al. 2006; Weng et al. 2007b; a; Bruce & Bradford 2008;
Carlisle et al. 2012; Hoyos-Padilla et al. 2016). Thus, the size of predatory fishes has a
strong influence both on their prey and on whether they have a role as a top-order
carnivore or lower, secondary-order consumer within a food chain. For many large (>3
m length) sharks such as the white (Carcharodon carcharias), tiger (Galeocerdo cuvier)
and hammerhead (Sphyrna sp.), only the adult life history stages can be classified as
truly top-order predators (Heupel et al. 2014).
Many top-order predators preferentially target specific types of prey (Hayward et al.
2011, 2012; Lyngdoh et al. 2014), although feeding strategies vary both among and
within species and can range from specialist to generalist. Dietary specialization is often
strongly associated with the availability of resources (Cherel et al. 2007; Woo et al.
2008). Populations of a species that have a generalist diet may be composed of multiple,
generalist individuals or of sub-groups or even individuals, each of which is specialized
on different types of prey (or some combination of the two options) (Van Valen 1965;
Bolnick et al. 2002; Bearhop et al. 2004; Urton & Hobson 2005). For example, orcas
are described as a generalist predator, hunting a diverse range of marine prey including
fishes, marine mammals and even large sharks, however, different populations exhibit
high levels of specialization on particular types of prey that reflect their availability in
29
the environment (Hoelzel 1991; Pauly et al. 1998b; Pyle et al. 1999; Saulitis et al. 2000;
Ford et al. 2010).
Irrespective of their degree of specialisation, predators regulate the structure,
composition and diversity of prey communities directly by consumption and indirectly
through evoking predator risk effects in prey. Risk effects include shifts in the types of
habitat used by prey to reduce the likelihood of predation, altering the location of
foraging to reduce detection and/or facilitate evasion of predators and increased
investment in vigilance. These behaviours are costly to prey, since time spent avoiding
predators will result in a reduction in activities that enhance fitness such as foraging,
territory defence or reproduction (Brown et al. 1999; Ripple & Beschta 2004; Wirsing
& Ripple 2011). Behavioural responses to risk effects have been recorded and described
in a variety of terrestrial systems (Holmes 1991; Schmitz et al. 1997; Lima 1998;
Brown et al. 1999; Kie 1999; Sinclair et al. 2003; Creel & Christianson 2008; Thaker et
al. 2011), such as the effects of grey wolves (Canis lupus) on the grazing rate, habitat
selection and group dynamics of elk (Cervus elaphus) (Ripple & Beschta 2004; Creel et
al. 2005; Fortin et al. 2005) and the behavioural responses of ungulates under the threat
of lions (Panthera leo) in the African savanna (Sinclair & Arcese 1995; Valeix et al.
2009a,b; Thaker et al. 2010; Creel et al. 2014). Risk effects have been the subject of a
number of studies in aquatic communities (Tegner & Levin 1983; Werner et al. 1983;
Lima & Dill 1990; Lima 1998). However, there is relatively little information available
about the role of large top-order teleosts and sharks in creating risk effects in marine
environments, with the exception of the well-documented behavioural responses of
large herbivores such as turtles and dugongs to the threat of predation by tiger sharks in
seagrass habitats (Wirsing et al. 2007a,b, Heithaus et al. 2007b, 2009).
Although predator risk effects are exhibited by prey in both terrestrial and marine
ecosystems (Wirsing & Ripple 2011), the physical properties of marine environments
have some unique elements that can structure these interactions, such as three-
dimensional distributions of species, greater spatial structure of populations and
“openness” of the environment, stronger and more complex trophic relationships, and
the presence of pronounced ontogenetic shifts in trophic relationships (Carr et al. 2003).
Additionally, top-order predators in nearly all terrestrial ecosystems (with perhaps the
exception of the Arctic) are now constrained to only a fraction of their historical range
due to habitat loss and indiscriminate hunting as a result of the encroachment of humans
and livestock (see review by Ripple et al. 2014). Despite markedly reduced abundances,
30
in marine systems many top-order predators still occupy much of their original ranges,
although the extent and impacts of anthropogenic threats in these ecosystems are
growing rapidly (McCauley et al. 2015). Conservation and management strategies for
these wide-ranging and highly mobile species in marine ecosystems need to
accommodate their complex life-histories, the dynamics of oceanic habitats and their
spatio-temporal overlap with various human threats (Hooker et al. 2011).
1.2 Movement ecology
Movement is a change in the location of an individual in space over time (Nathan et al.
2008) and is a major component of some part of the life history of almost every
organism (Holyoak et al. 2008). Movement has implications for the individual,
populations, communities, ecosystems and, ultimately, evolution (Turchin 1991;
Webster et al. 2002; Dingle & Drake 2007). The decision to move is related to the
organism’s internal state (physiology, locomotory and navigation capabilities; Nathan et
al. 2008), but is influenced by external environmental factors such as the landscape,
weather, physical processes, resource availability, and other organisms that can act as
competitors, predators or mates. Movement can thus be considered as the result of an
interaction between environmental and internal factors and is driven by ecological
processes at various spatial and temporal scales (Crist et al. 1992; Nathan et al. 2008).
Animal movement can be classified into a wide range of modes depending on the
definition of the goals, motivation or temporal dynamics of the processes(e.g. foraging,
dispersion, nomadic, migration), but most frequently is connected to the search for and
use of resources (Dingle & Drake 2007; Nathan et al. 2008; Sims et al. 2008).
Resources in environments are spatially heterogeneous, resulting in unequal or patchy
distributions that drive the dispersion and foraging behaviours of organisms (Turchin
1991; Johnson et al. 1992). Moreover, this patchiness is argued to have a hierarchical
spatial structure, where a patch at a larger spatial scale has nested, internal structures of
heterogeneity at increasingly finer scales (Kotliar & Wiens 1990). For example, small
pelagic fishes, the prey of many larger marine predators, occur in high density schools
at small scales (m - km) that are aggregated into patches associated with mesoscale
oceanographic features at intermediate scales (10 - 100s km), which occur on migration,
feeding or spawning grounds at large (100 - 1000s km) spatial scales (Fauchald 1999;
Fauchald & Tveraa 2006). As a result of these patchy distributions of prey, in the
absence of knowledge of the distribution of resources, predators must optimize their
31
movement in order to maximize the probability of encountering food (Fauchald 1999;
Fauchald & Tveraa 2003; Bartumeus et al. 2005). According to theories of optimal
search strategies, animals will display more directed, straight-line movements in areas
with a low density of resources, whereas in areas of high density of prey they will slow
down and increase the frequency of turning angles to maximize the probability of prey
encounter (Kareiva & Odell 1987). Without previous knowledge of the distribution of
resources, predators may adapt their search and movement patterns to optimise prey
encounter (Sims et al. 2008). A specialised type of random walk known as Lévy walks
(Viswanathan et al. 2000; Bartumeus et al. 2005; Sims et al. 2008) appears to describe
probabilistic search patterns of some species consisting of numerous discrete steps that
are linked by rarer longer steps, which are repeated across multiple scales. However,
much controversy exists around the validity of Lévy foraging hypothesis to describe
optimal search in comparison with cue-driven Brownian-type motion (Benhamou &
Collet 2015), where animals would move in a straight-line random direction until prey
is found, and then switching to a smaller steps described by a Brownian walk
(Benhamou 2007).
Environmental conditions and habitat characteristics create much of the patchiness
within the marine environment. This occurs through the influence on movement of
currents and wind (Weimerskirch et al. 2000; Fritz et al. 2003; Chapman et al. 2011);
variation in optimal ranges of salinity, temperature and dissolved oxygen (Jonsen et al.
2007; Heupel & Simpfendorfer 2008; Espinoza et al. 2011; Bestley et al. 2013) and/or
by physical features that may have an aggregating function such as thermal fronts,
upwelling, eddies, canyons and shelf breaks (Klimley & Butler 1988; Royer et al. 2004;
Croll et al. 2005; Cotte et al. 2007; Doniol-Valcroze et al. 2007; Scales et al. 2014;
Bouchet 2015; Bouchet et al. 2015; Queiroz et al. 2016). In turn, this patchiness also
affects the movements of predators (Humphries et al. 2010). For example, the thermal
structure of the water column shapes the vertical distribution of the sardine (Sardinops
sagax) and as a result it influences the vertical movements of their predators, the Pacific
bluefin tuna (Thunnus orientalis) (Kitagawa et al. 2004, 2007). Relationships between
such environmental drivers and movement of animals is largely scale dependent, both
spatial and temporally (Guinet et al. 2001; Pinaud & Weimerskirch 2005; Fryxell et al.
2008; Avgar et al. 2013). For instance, at a large scale (1000s km) the habitat use of
yellow-nose albatrosses (Thalassarche carteri) is related to bathymetry (a proxy for
ocean and shelf habitats) and sea surface temperatures (SST) associated with subtropical
32
waters, but at mesoscales (10 – 100s km) habitat use is associated with chlorophyll-a
and sea-surface height anomalies that are indicative of mesoscale oceanic features such
as eddies (Pinaud & Weimerskirch 2005).
The patchy nature of marine environments constrains the movements of animals. The
area within which animals move continuously for feeding, resting and reproduction is
termed the home range (Burt 1943). As an statistical concept, home range is define as a
fixed percentage (commonly 95%, 50%, 25%) of the confidence region obtained from
the relative frequency distribution of an animal’s location, usually two dimensional,
over time (Worton 1987). However, as noted above, how much an individual utilises
specific areas within its home range will vary temporally, and according to resources
available and habitat quality (Benhamou 2011; Benhamou & Riotte-Lambert 2012).
Areas of high use will be related to accessibility of food, but must also offer
environmental conditions that are within an optimal range. For example, dusky sharks
(Carcharhinus obscurus) in the Gulf of Mexico, although highly mobile, spend the
majority of their time within waters off the continental shelf where water temperatures
are between 24-28°C (Hoffmayer et al. 2014).
Seasonal events such as migration may break the stability of an animal’s use of space
(Bunnefeld et al. 2011). Similar to ‘animal movement’, the term “migration” is used to
describe a large array of movement types, from vertical movements in the water column
to seasonal largescale movements of several species, and to describe movements from
individual to population levels (Dingle & Drake 2007; Nathan et al. 2008). However, it
generally refers to largescale movements occurring as a result of seasonal variability in
the distribution of resources (Alerstam et al. 2003; Dingle & Drake 2007) or spatial
segregation between breeding and non-breeding areas (Webster et al 2002). More than
12% of vertebrates make long-distance (100-1000s km) migrations throughout the
continents and oceans, however, in general these are more common in swimming and
flying vertebrates as a result of the lower cost of locomotion (Robinson et al. 2009). In
marine systems environmental conditions have been identified as major element of
migratory movements. For instance, migration of juvenile bluefin tuna (Thunnus
thynnus) to the waters of the northern California coast during spring is associated with
warmer ocean temperatures and reduced wind stress, which also results in higher prey
availability (Kitagawa et al. 2007).
33
Large marine predators, such as epipelagic sharks and tunas, are found throughout the
world’s oceans and considerable information is available on the horizontal movements
of some of the most wide-ranging species (Table 1.1). These top-order predators display
a mix of resident behaviours interspersed with extensive pelagic movements.
Environmental variables, such as sea surface temperature (SST) and chlorophyll-a, have
been identified as some of the major drivers of movement patterns for these large
predators (Block et al. 2005, 2011; Queiroz et al. 2016). However, for some of these
species, drivers have only been defined for a limited region within their distribution
range, and the motivation behind large scale movements (foraging, breeding) has yet to
be determined. Defining environmental and habitat preferences of mobile predators
during residency and along migration pathways allows key habitats within their
distribution to be identified (Cooke 2008; Block et al. 2011). This information is also a
key step for ecosystem-based fisheries management (Pikitch et al. 2004) such as
dynamic ocean management (Hobday et al. 2014; Maxwell et al. 2015), where
management adapts to the spatial and temporal dynamics of the ocean based on the
integration of real-time biological, oceanographic and socio-economic data (Maxwell et
al. 2015). The identification of preferred environmental conditions and association with
oceanographic features (i.e. mesoscale features such as upwelling areas, fronts and
eddies) could be used to inform fisheries and enhance dynamic management actions for
highly mobile species (Hobday & Hartog 2014; Lewison et al. 2015).
34
Table 1.1 Range and drivers of movement patterns displayed by large epipelagic top-order sharks and teleosts.
Species Common
name
Maximum
size (cm)
Range of movements Drivers of movements References
Thunnus thynnus bluefin tuna > 450 Seasonal residency, migration to
breeding grounds, longitudinal and
transatlantic movement to foraging
grounds
SST**, chlorophyll-a,
body size, population
(Lutcavage et al. 2000; Block et al.
2005, 2011; Walli et al. 2009;
Galuardi et al. 2010; Galuardi &
Lutcavage 2012)
Carcharodon carcharias white shark > 600 Residency and site fidelity to coastal
areas, philopatry, long-distance pelagic
movements, trans-oceanic migration
foraging, mating and
breeding (hypothesised),
SST**
(Boustany et al. 2002; Bruce et al.
2006; Weng et al. 2007b,a, Domeier &
Nasby-Lucas 2008, 2013; Bonfil et al.
2010; Jorgensen et al. 2010; Block et
al. 2011)
Galeocerdo cuvier tiger shark > 500 Residency and restricted movements in
coastal habitats, oceanic islands and
around coral reefs, seasonal migrations
and residency, long-distance latitudinal
and pelagic movements
SST, bathymetry, season,
chlorophyll-a
(Holland et al. 1999; Heithaus et al.
2007c; Meyer et al. 2009;
Hammerschlag et al. 2012; Hazin et al.
2013; Papastamatiou et al. 2013;
Werry et al. 2014; Holmes et al. 2014;
Ferreira et al. 2015; Lea et al. 2015)
Isurus oxyrinchus mako shark 400 Extensive offshore movements and site
fidelity to slope and shelf edge, seasonal
latitudinal movements,
SST, SST gradients,
chlorophyll-a
(Musyl et al. 2011; Block et al. 2011;
Rogers et al. 2015; Queiroz et al.
2016; Vaudo et al. 2016)
Carcharhinus
longimanus
oceanic
whitetip
350 Long-distance pelagic movements, some
indication of philopatry, trans-equatorial
migration*
SST** (Kohler et al. 1998; Musyl et al. 2011;
Carlson & Gulak 2012; Howey-Jordan
et al. 2013)
Sphyrna lewini great
hammerhead
< 350 Residency and site fidelity in continental
shelfs and islands, occasional offshore
movements
SST gradients, water
temperature, currents
(Klimley 1993; Bessudo et al. 2011;
Hammerschlag et al. 2011b; Ketchum
et al. 2014a,b; Queiroz et al. 2016)
Prionace glauca blue shark 320 Overwintering around Gulf Stream and
Sargasso Sea (central North Atlantic),
long-distance pelagic movements, trans-
oceanic migrations*
SST, foraging (Kohler et al. 2002; Queiroz et al.
2005, 2016; da Silva et al. 2010;
Campana et al. 2011; Block et al.
2011; Vandeperre et al. 2014)
* Information obtained from tag-recapture data.
** Driver only identify for the North Pacific by Block et al. 2011a
SST = sea surface temperature
35
1.3 Challenges for the conservation of large marine predators
Recent analyses show that body size is one of the strongest predictors of the risk of
human-driven extinction in the modern oceans, with large-bodied animals suffering the
most elevated levels of threat (Payne et al. 2016) because they are preferentially
targeted by human hunting (Pauly et al. 1998a; Pauly & Palomares 2005). Top-order
marine predators are typically large animals and this trait, combined with their life
history strategies and wide-ranging movements has led to declines in many species so
that they are now classified as Vulnerable, Threatened or Endangered by the
International Union for the Conservation of Nature (IUCN) Red list guidelines (IUCN
2012).
The mobility of marine predators such as large sharks presents one of the greatest
challenges for conservation, as is the case for many other species (Shuter et al. 2011;
Runge et al. 2014; Campana 2016). For some species, nomadic movement behaviours
might be predominant, in which home ranges, and breeding timing and areas are not
predictable (Dingle & Drake 2007). Nomadic or migratory behaviour brings these
animals into contact with multiple threats and at the same time, it hinders our ability to
assess impacts on their populations (Shuter et al. 2011; Runge et al. 2014; McCauley et
al. 2015). Mobile species are extremely vulnerable when large portions of a population
migrate to restricted breeding or wintering locations, as is the case for many taxa
including fishes, reptiles and pelagic birds (Webster et al. 2002) or when migrations
traverse socio-political boundaries that have varying levels of protection as occurs for
many sharks and fishes (Martin et al. 2007; Berger et al. 2008; Hooker et al. 2011;
Dallimer & Strange 2015). For terrestrial species, a common response to ensure the
conservation of mobile species has been a focus on the protection of migration corridors
(Berger 2004; Berger et al. 2008), but for marine predators, particularly in the open
ocean, such essential habitats are much harder to identify and delineate.
Significant efforts have been directed towards the protection of “hotspots” of
biodiversity through the designation of Marine Protected Areas (MPAs) (Mittermeier et
al. 1998; Médail & Quézel 1999; Myers et al. 2000; Hughes et al. 2002; Roberts et al.
2002) and these have become a key strategy for ecosystem-based conservation and
management (Hooker et al. 2011; Abecasis et al. 2014; Lascelles et al. 2014; O’Leary
et al. 2016). In recent years, the mean size of MPAs has increased greatly, however
most are still too small relative to the scale of movements and home ranges of many
36
top-order predators such as sharks to ensure the protection of all life history stages
(Heupel et al. 2015; McCauley et al. 2015). In such species, individuals can be spread
over many thousands of kilometres, which prevents protection of the entire population
(Hooker et al. 1999; Chapman et al. 2005; Game et al. 2009; Hays & Scott 2013). If
MPAs are to be an effective strategy for the conservation of large predators, they must
target areas of higher vulnerability such as mating and foraging grounds, migration
corridors, or nursery areas (Heupel et al. 2007; Lascelles et al. 2014; Pendoley et al.
2014; Hays et al. 2014). Dynamic ocean management has also indicated great potential
benefits to managing highly mobile species or processes (Lewison et al. 2015; Maxwell
et al. 2015; Dunn et al. 2016). Real-time tracking has been suggested as a useful tool for
several migratory species (Hobday & Hartmann 2006; Hobday & Hartog 2014; Hazen
et al. 2016). For example, a management tool “WhaleWatch” has been developed for
blue whales (Balaenoptera musculus) in the California Current by combining tracking
data with distribution models and the collection of environmental data to develop near-
real-time habitat prediction tool (Hazen et al. 2016). Optimisation of these strategy
requires a clear understanding of the relationships between movement patterns,
environmental drivers and habitat preferences of these animals (Bailey & Thompson
2009; Game et al. 2009; Hooker et al. 2011; Abecasis et al. 2014). For many species,
particularly large sharks, we still lack much of this data.
One of the key issues hampering our understanding of the relationship between
movement patterns and environmental drivers of marine predators has been limitations
on the numbers and components of populations animals tracked in tagging programs.
Over the last three decades, satellite tags have provided unprecedented insights into the
movement behaviours of marine predators and their relationships to environmental cues
(Biuw et al. 2007; Block et al. 2011; Schlaff et al. 2014; Hussey et al. 2015a).
However, due to the expense of this technology and the logistics involved in capturing
animals for tagging (Hussey et al. 2015a) most studies of large marine predators rarely
tag more than a few dozen individuals, unless the species is targeted by fisheries, as is
the case for tunas (Block et al. 2005). Moreover, the ontogenetic stage and sex of
animals tagged is rarely representative of the entire population due to differential
accessibility of different demographic components during sampling. This has limited
our understanding of patterns across species and populations. However, the combination
of information from multiple studies of the same species, or from studies of multiple
species within a single region, has begun to overcome this issue, allowing
37
generalisations of spatial patterns of habitat use and drivers of movement on a
population level and even for entire communities. For example, in the Pacific Ocean,
the Tagging of Pacific Predators (TOPP) program compiled tracking data for 23 species
and was able to describe how various taxa including tunas, sharks, cetaceans, seabirds
and marine turtles were associated with the California Current Large Marine Ecosystem,
where seasonal movements were associated with changes in ocean temperature and
measures of productivity (Block et al. 2011). Similar studies in other marine ecosystems
are urgently required, given that many of the threats facing marine systems such as
climate change and ocean acidification are global in scope (Hoegh-Guldberg & Bruno
2010; Hazen et al. 2013; Jones & Cheung 2015). Due to the importance of
environmental cues for migratory patterns, climate change is predicted to affect the
distribution and abundance of marine fishes worldwide (Perry et al. 2005; Jones &
Cheung 2015). However, most of our knowledge about the effects of climate change on
migratory patterns of animals is related to birds, with large gaps in our knowledge of
long distance movement and potential human impact in other groups (Robinson et al.
2009). Detailed information about the environmental drivers of movement is necessary
to understand the impacts of projected climate changes, predict what areas and which
species are at greater risk, and to plan conservation strategies.
1.4 The tiger shark
Tiger sharks (Galeocerdo cuvier) (Figure 1.1) are distributed worldwide in most of the
tropical, subtropical and temperate habitats of oceanic and coastal ecosystems (Randall
1992). They are categorised as a top-order predator in tropical marine ecosystems and
can exert strong structuring forces in prey communities through predator risk effects.
For example, in the shallow seagrass habitats of Western Australia, green turtles
(Chelonia mydas), dugongs (Dugong dugon) and cormorants (Phalacrocorax varius)
often select habitats that are less profitable for foraging, but reduce the chance of
interactions with tiger sharks at times when these predators are abundant (Heithaus
2005; Wirsing et al. 2007a,b, Heithaus et al. 2007b, 2008b). The impact of tiger sharks
on the behaviour of prey species in these seagrass communities is well established and
is frequently cited as an example of risk effects in marine predator-prey interactions
(Heithaus et al. 2008b; Wirsing & Ripple 2011). However, tiger sharks exhibit a
generalist diet (Stevens & McLoughlin 1991; Simpfendorfer 1992; Lowe et al. 1996;
Simpfendorfer et al. 2001) and also often feed at lower trophic levels in food chains or
38
scavenge on dead prey. Despite their importance as top-order predators in some
habitats, such as the seagrass system described above, this role may not be duplicated
across all the marine environments they traverse. For example, analyses of
specialization of diet and trophic niche using stable isotopes have shown that tiger
sharks are indeed true generalists (Matich et al. 2011; Trystram et al. 2016) and
although they occupy apex levels of seagrass and reef-associated food webs (Heithaus et
al. 2013; Frisch et al. 2016), the species feeds at the lowest trophic level within the
assemblage of large sharks off coastal reefs in South Africa (Hussey et al. 2015b). Such
conflicting results imply that the movement patterns, habitat use, diet and trophic role of
tiger sharks are intertwined and not necessarily consistent throughout the species range.
To disentangle these factors, and others such as sex and ontogenetic stage that might
also influence the role of these sharks (MacNeil et al. 2005; Borrell et al. 2011; Carlisle
et al. 2014; Munroe et al. 2015), comprehensive data sets are required that document
patterns of movement, habitat use and diet across multiple spatial and temporal scales.
Figure 1.1 The tiger shark.
Tiger sharks have a global distribution and display high utilisation of coastal systems
(Heithaus et al. 2002). As a result, a number of research programs have investigated
movement patterns and habitat use of tiger sharks throughout most of the species range
(Table 1.2). These sharks have been tagged with a wide range of tags (acoustic, PAT,
SPOT, accelerometers) and both vertical and horizontal movements have been
described. Typically, tagging studies of tiger sharks show great individual variability in
the use of space (Holland et al. 1999; Heithaus et al. 2007c; Fitzpatrick et al. 2012;
Hammerschlag et al. 2012; Hazin et al. 2013; Papastamatiou et al. 2013; Holmes et al.
2014; Lea et al. 2015). Sharks tagged in seagrass habitats off Shark Bay, Western
Australia, exhibited either restricted movements near the area where they were tagged,
remaining within the coastal embayment and surrounds, or moved to oceanic waters
39
(Heithaus et al. 2007c). The same pattern of individual variation in residency and
migration has been recorded for tiger sharks in waters off Hawaii and the eastern coast
of the United States (Hammerschlag et al. 2012; Papastamatiou et al. 2013). Some
individuals also displayed large seasonal migrations (1000s km) in the North Atlantic
and off both the east (Holmes et al. 2014; Lea et al. 2015) and west coasts (Ferreira et
al. 2015) of Australia. However, these studies tagged mostly female sharks and many
satellite tag deployments were short, typically in the order of only a few weeks or
months. Longer durations of tagging have been achieved using acoustic tags, however
detection rates are typically low (Meyer et al. 2009, 2016; Papastamatiou et al. 2010;
Werry et al. 2014). Furthermore, only two studies have modelled tracking information
in order to define the drivers of movement and residency for tiger sharks (Fitzpatrick et
al. 2012; Papastamatiou et al. 2013). Although tiger sharks have been tagged in many
sites across their distribution range and in different habitats, the explanation for the
female bias in most datasets is still unresolved. This bias can be a result of the habitats
being sampled (warm, coastal habitats); however, Lea et al. (2015) described the
seasonal use of coral reef systems in the Caribbean by male tiger sharks
Since most studies have tagged females, little is also known about critical habitats and
movement patterns of the other components of the population, particularly males,
juveniles and neonates (Driggers et al. 2008). Moreover, contrasting evidence of
ontogenetic habitat shifts have been reported, with sub-adult sharks believed to forage
over large areas in some regions (Meyer et al. 2009) but showing limited movement in
others (Werry et al. 2014; Afonso & Hazin 2015). Although most sharks have a K-
selected life history (Stevens et al. 2000; Frisk et al. 2001; Dulvy et al. 2014a), tiger
sharks produce litters of up to 80 pups (Stevens & McLoughlin 1991; Simpfendorfer
1992; Whitney & Crow 2007). In contrast, other large predatory sharks often produce a
limited number of embryos, usually less than 10 (Gilmore 1993), such as white sharks
(litter sizes of only 2-10 pups; Francis 1996) and the grey nurse or sand tiger shark
(Carcharias taurus), which produces only two embryos (Pollard et al. 1996). Tiger
sharks also display extreme rapid growth in their early life history stages (Branstetter et
al. 1987; Natanson et al. 1999; Wintner & Dudley 2000; Afonso et al. 2012), doubling
their size at birth in the first year of life. These biological traits suggest that tiger sharks
might have higher population growth rates and be more resilient to fishing pressure than
most sharks that are top-order predators (Simpfendorfer 2009). However, they are a key
target in coastal shark control programs (Dudley & Simpfendorfer 2006; Green et al.
40
2009; Simpfendorfer et al. 2010; Reid et al. 2011) and components of catches in many
artisanal and unregulated fisheries (Field et al. 2009; Moir Clark et al. 2015). At present
the limited information available on trends in population sizes suggests that abundances
are declining, leading to tiger sharks being classified as “Near Threatened” by the IUCN
(Chapter 2).
The progress made by tracking studies that have pooled information from different
sampling sites across an ecosystem (Costa et al. 2010a; Block et al. 2011; Raymond et
al. 2015) or species range (Biuw et al. 2007; Hindell et al. 2016) demonstrate a means
to move towards a better understanding of key aspects of the ecology of marine
predators. These studies show the value of large datasets in defining environmental
preferences and critical habitats for populations and communities in areas such as the
California Current Large Marine Ecosystem in the North Pacific Ocean (Block et al.
2011) and the Antarctic continental shelf, an important foraging ground for southern
elephant seals (Mirounga leonina) (Hindell et al. 2016). For tiger sharks, the pooling of
key data sets could advance our understanding of the movement and feeding ecology of
the species: firstly, tracking studies over locations distributed across the species range
and composed of different parts of the populations (males and females sub-adults etc.);
and secondly, dietary information across a range of habitats and environments. The
latter will allow a better understanding of the role of food availability as one of the
major biological drivers of movement patterns.
41
Table 1.2 Tiger shark tagging programs with the number of tags deployed, sex ratio of animals tagged, size range of sharks tagged and information provided by
telemetry devices.
Tagging
program
Ocean
basin
N
Tags Tag type
Sex
ratio
(F:M)
Total
length
(cm)
Track
duration
(days)
Distance
(km)
Temperature
range (°C) Data collected Aims Reference
Hawaii Pacific 140 Acoustic,
PSAT, SPOT,
accelerometers
2.1:1a 99 - 464 <1-
464b,161
c
1000s 5 - 30 Presence, location,
depth,
temperature,
swimming
direction, stroke
frequency, video,
acceleration,
speed
Diel movements,
horizontal and vertical
movement, depth-
temperature profile, yo-
yo diving, foraging
behaviour, residency,
utilization distribution
and migration (and
drivers), site fidelity
(Tricas et al.
1981; Holland et
al. 1999; Meyer
et al. 2009, 2010,
Papastamatiou et
al. 2010, 2013;
Nakamura et al.
2011)
East
Australia
Pacific 17 PSAT, SPOT 5.7:1 150 - 350 6 - 408 15791d 6-29.5 Location, depth,
temperature, light-
level (geolocation)
Depth - temperature
profile, horizontal
movements, habitat use,
residency
Holmes et al.
2014
Raine
Island
Pacific 9 SPOT 3.5:1 288 - 350 16 - 356 - <12 - >30 Location,
temperature, depth
Area-restricted search,
horizontal movements,
relation with turtle
nesting season
Fitzpatrick et al.
2012
Coral Sea Pacific 34 PSAT, SPOT,
Acoustic
3.4:1 154 - 390 1 - 255b;
210c
1141 5.6 - 32.2 Presence, depth,
temperature,
location
Site fidelity, migration,
diving, habitat use,
depth- temperature
profile
Werry et al. 2014
Florida
and
Bahamas
North
Atlantic
25 SPOT 11.5:1 184 - 403 26 - 297 3500 - location Horizontal movements,
residency, activity space,
impacts of provisioning
ecotourism
Hammerschlag et
al. 2012
Challenger
Bank
North
Atlantic
24 SPOT 0.2:1 173 -369 < 45 -
>1000
42996 18 -26e location Migration, seasonal
horizontal movements,
Lea et al. 2015
42
philopatry, straightness
of movement
U.S.
Virgin
Islands,
Bermuda
North
Atlantic
14 PSAT 0.5:1 210 - 305 9 - 184 1354 6.3 - 29.1 Depth,
temperature, light-
level (geolocation)
Horizontal and vertical
movements, depth-
temperature profile,
depth behavioural type
Vaudo et al. 2014
Recife South
Atlantic
21 PSAT,
miniPAT,
Acoustic
1.6:1 120 - 295 6 - 159 1156f 4 - 25 Depth,
temperature, light-
level (geolocation)
release survival, vertical
and horizontal
movements patterns,
shark bite mitigation
Hazin et al.
2013a; Afonso &
Hazin 2014, 2015
Ningaloo
Reef
Indian 8 SPOT,
SPLASH
4.5:1 145 - 333g 7 - 517 > 4000 6-33 Location, depth,
temperature
Residency and drivers,
home range area, large-
scale movements
Ferreira et al.
2015
Shark Bay Indian 57 SPOT,
acoustic,
crittercam
1.7:1 230 - 401 <1 - 99 505h - Presence, video,
depth, location,
Site fidelity, horizontal
and vertical movements,
habitat use, foraging
behaviour
Heithaus 2001a,
Heithaus et al.
2001, Heithaus,
Dill, et al. 2002,
Heithaus,
Wirsing, et al.
2007 a Sex ratio of reported numbers, sex ratio of acoustic tagged sharks was up to 6.3:1 (Papastamatiou et al. 2013)
b Acoustic tag
c Satellite tag
d track length (km)
e Remote sensed sea surface temperature
f Minimum distance travelled
g Fork length (cm)
h A movement of 8000 km was registered by a single location
43
1.5 Aims
The overall goal of my thesis is to describe the movement ecology of tiger sharks
(Galeocerdo cuvier) and identify the drivers of movement patterns and habitat use
across a range of spatial and temporal scales. My research aims to:
1) Outline the current understanding of the biology and population trends, and assess
the conservation status of the species.
2) Describe large-scale movements, residency and environmental preferences of tiger
sharks along the coastline of Western Australia.
3) Identify spatial patterns of diet and trophic ecology of different sexes and sizes of
tiger sharks in Australia. Describe the variability and stability in diets in relation to local
food-webs at different temporal scales.
4) Describe the movement patterns and habitat use of tiger sharks at a global scale and
use novel statistical approaches to identify the potential physical and biological factors
that drive these patterns.
In line with these aims, my thesis is organised into an introductory literature review,
four research chapters, and the findings are drawn together in a final discussion chapter.
1.6 Thesis outline
In Chapter 2 I review the global literature on the distribution, population, ecology and
threats to tiger shark populations. I then use this information to re-assess the
conservation status of the species. This review followed the category guidelines of the
International Union for Conservation of Nature (IUCN) and will update the IUCN Red
List assessment for tiger sharks.
Chapter 3 describes depth and temperature preferences associated with movements and
residency of tiger sharks along the coastline of Western Australia. Movement-based
home range analyses were applied to satellite tracking data to define areas of residency
and space used by individual sharks. Mixed models were then fitted to the relationship
between residency and environmental variables (depth and sea surface temperature). In
this chapter, I show that, as top-order predators, tiger sharks are able to move extensive
44
distances, linking tropical and temperate ecosystems and that their environmental
preferences vary among different regions (north/south) of the coast of Western
Australia.
An important motivation behind predator movement is the distribution of prey.
Knowledge of spatial changes in diet is likely to be key in understanding the function of
habitats used by marine predators. In Chapter 4, I use stable isotope analysis to
characterise diet and trophic ecology of tiger sharks throughout Australia and identify
factors driving spatial and temporal variation in isotopic signatures of tissues. I discuss
diet in the context of the differential use of multiple habitats and food webs by tiger
sharks across their range in northern Australia.
In Chapter 5, I compile and analyse a global dataset of tiger shark tracks across the
Atlantic, Indian and Pacific oceans to search for and identify environmental drivers of
movement and habitat use for tiger sharks at the scale of the entire range of the species.
Using a novel analytical approach, I examine how the influence of environmental
factors varies across different ocean basins and between sexes, and identify areas where
environmental variables did not explain habitat use. These areas can then be targeted for
future study.
I integrate and synthesise the findings of my thesis in Chapter 6. I discuss how insights
into the movement, habitat use and diet of tiger sharks across multiple spatial and
temporal scales can fill some of the gaps in our knowledge on the ecology of the
species. Using this information, I examine the role of the species as a top-order
predator, how the behaviours and traits of the species compare to other top-order
predators and the implications for the resilience of tiger sharks in marine environments.
Finally, I evaluate the benefit of better understanding of the drivers of movement
ecology of marine predators for conservation planning.
45
Chapter 2 Review of tiger shark ecology and
conservation status
2.1 Introduction
Globally, shark populations have shown accelerated declines due to overfishing and
habitat degradation (Dulvy et al. 2008; Ward-Paige et al. 2010; McCauley et al. 2015;
Campana 2016), with one quarter of shark and ray species now classified as
“Threatened” by the International Union for Conservation of Nature (IUCN) Red List
(Dulvy et al. 2014a). The IUCN Red List assessments are regarded as reliable tools to
determine the conservation status of species because they use a comprehensive and
scientifically rigorous approach to estimate extinction risk (Rodrigues et al. 2006). The
goal of the assessment is to provide information and analyses on the status, trends and
threats to species in order to inform conservation actions. Current assessments of sharks
have been undertaken during regional Red List workshops coordinated by the IUCN
Shark Specialist Group (SSG) in the past 5 - 10 years (Dulvy et al. 2014a), with the last
assessment for tiger sharks occurring in 2009 (Simpfendorfer 2009). In this chapter, I
update the IUCN assessment for tiger sharks, reviewing their current conservation status
and indicating the research needs for the future using the guidelines of the IUCN Red
List (IUCN 2012), with some adaptations appropriate for a thesis chapter.
2.2 Distribution
Tiger sharks (Galeocerdo cuvier) have a worldwide distribution in tropical and warm
temperate oceans, ranging between the latitudes of approximately 40°N - 36°S (Figure
2.1). Compagno (1984) and Randall (1992) describe the distribution across ocean basins
as follows:
- In the western Atlantic, they range from the USA (from Cape Cod, Massachusetts) to
Uruguay, including the Gulf of Mexico, Bermuda and islands of the Caribbean. In the
eastern Atlantic, they are found along the West African coast, from Morocco to Angola,
including the Canary Islands and at isolated islands such as Fernando de Noronha in the
South Atlantic.
- Tiger sharks are also reported from locations throughout the Pacific Ocean. In the
eastern Pacific they range from southern California to Peru, including the Galapagos
46
and Revillagigedo Islands. In the western Pacific they occur off the coasts of Japan, east
China, Australia and northern New Zealand; as well as in the western, central Pacific in
Palau, east to Solomon, Marshall and Hawaiian Islands, French Polynesia and other
isolated atolls.
- In the Indian Ocean, tiger sharks are found all along the east coast of Africa, including
remote localities such as Réunion Island and the British Indian Ocean Territory, north to
the Red Sea, and throughout the tropical Indian Ocean off the coasts of Pakistan, India,
Sri Lanka, Thailand, Vietnam, Southern China, the Indo-Pacific region and most of
western Australia.
Tiger sharks also appear seasonally in cool temperate waters, most likely following
warmer currents with reports from the United Kingdom (Compagno 1984) and Iceland
(Matsumoto et al. 2005), the southern coasts of New South Wales and Western
Australia (Pepperell 1992; Holmes et al. 2014; Ferreira et al. 2015) and South Africa
(Dicken & Hosking 2009).
Figure 2.1 Distribution range of tiger sharks (blue).
2.3 Population trends
Long-term trends of catch rates of tiger sharks are only available from limited sources
including regional shark control programs and fisheries observer and fisheries-
independent longline survey datasets for the US east coast and Gulf of Mexico ( Table
2.1). Tiger sharks are not usually targeted by large commercial fisheries, but occur as
by-catch in low numbers. As a result, captures are often not recorded and there is a lack
47
of adequate data for stock assessment, definition of population structure and
documentation of trends in demography of the species.
Table 2.1 Summary of long-term trends of catch rates of tiger sharks based on regional
datasets.
In the northwest Atlantic, the bottom longline fishery for sharks is active along the east
coast of the US and Gulf of Mexico (Gulak et al. 2013), and the Northeast Fisheries
Science Centre (NEFSC) conducts fisheries-independent surveys of large and small
coastal sharks in waters off Florida to the Mid-Atlantic (http://nefsc.noaa.gov/nefsc/
Narragansett/sharks/survey.html). Data from NEFSC surveys conducted from 1960 -
2000 showed an upward trend in the relative abundance of tiger sharks, with an increase
of 6.7% per year since the 1980s. Because relative abundance is calculated as an index
of actual or absolute abundance based on standardised catches (Blackhart et al. 2006),
these trends reflect changes in CPUE (catch per unit effort) rather than being a
consequence of an increase in effort. Similarly, the Shark Bottom Longline Observer
Program (SBLOP) also showed an increase in relative abundance of the species of 8%
per year between 1994 and 2013. However, the Pelagic Observer Program from the
Southeast Fisheries Science Centre (SEFSC) showed no change in abundance for the
same period (Figure 2.2) (J. Carlson, pers. comm.) and previous estimates based on the
standardised CPUE time series of logbook data for the U.S. pelagic longline fleets in the
northwest Atlantic suggested a decline of 61 - 65% in abundance of tiger sharks (Baum
et al. 2003). These estimates of declines might be biased due to analytical issues
(Burgess et al. 2005)
Region Pop trend Period Source Metric used
NW Atlantic Increasing 1994-2013 Northeast Fisheries Science
Centre
Relative
abundance
Queensland Declining 1993-2010 Queensland Shark Control
Program
CPUE
(projection)
New South Wales Declining 1998-2007 NSW Shark Meshing Program CPUE
South Africa Increasing 1978-2003 KwaZulu-Natal Beach
Protection Program
CPUE
48
Figure 2.2 Relative abundance (calculated as an index of standardised catch) for the Shark
Bottom Longline Observer Program (SBLOP, blue diamonds) and longline survey of the
Northeast Fisheries Science Centre (NEFSC, green triangles) and SEFSC Longline (MS
longline, red squares). Figure produced by J. Carlson (pers. comm.).
In Australia, the Queensland Shark Control Program (QSCP) has used a combination of
nets and drumlines adjacent to popular swimming beaches to target large sharks since
1962 (Paterson 1990). Catch per unit effort (defined as sharks net−1
day−1
or sharks
drumline−1
day−1
) of tiger sharks showed an overall declining trend for the QSCP
between 1993 and 2010 (Figure 2.3), with the most significant decline in the southern
region (Holmes et al. 2012). A separate analysis for nine beaches in Cairns and
Townsville, northern Queensland (Figure 2.4), showed that annual, standardised catch
rates (annual catch was modelled including effort by beach in all models to account for
changes in effort over time and generate an index equivalent to CPUE) increased up
until the 1980s and then decreased to approximately 66% of their original level in the
2000s (Simpfendorfer et al. 2010). Based on the CPUE for 1993 and 2010 from Holmes
et al. (2012), and a generation length of 17.5 to 22.5 years for tiger sharks (Natanson et
al. 1999; Holmes et al. 2015), a prediction of population change within three
generations was calculated according to the rules for estimating reductions by the Red
List guidelines (IUCN 2012). Generation length was calculated according to the
formula: Age at first maturity + (Maximum age - Age at first maturity/2). The
prediction considered CPUE values from 1993 and 2010 as “Past” population sizes
assuming a regression with an exponential rate of decline. The prediction showed that
49
tiger shark CPUE will undergo a decline of 90 - 96% for drumlines and of 94-99% for
nets over the next 30-45 years in relation to the CPUE for the QSCP in 1993 (Figure
2.3). These predictions should nevertheless be treated with caution, as they do not
account for fluctuations in CPUE or demographic characteristics of the tiger shark
population.
Figure 2.3 CPUE (number of sharks gear-1
day-1
) by year and size (small < 3 m and large > 3 m)
of sharks for drumlines (A) and nets (B) by the QSCP between 1993 and 2010. Figure extracted
from Holmes et al. (2012).
Table 2.2 Average CPUE(number of sharks gear-1
day-1 x1000) for drumlines from the QSCP
for 1993 and 2010 from Holmes et al. (2012) and projected CPUE for three generation lengths
(Projection 1 = generation length of 17.5 years, Projection 2 = generation length of 22.5 years).
Population Past 1 (1993) Past 2 (2010) Projection 1 (2044) Projection 2 (2060)
QSCP Drumlines 0.631 0.430 0.044 0.015
QSCP Nets 0.244 0.187 0.010 0.003
50
Figure 2.4 Locations of the Queensland Shark Control Program with details of gear (1993-
2010), from Holmes et al. (2012).
In southeast Australia, the New South Wales Shark Meshing Program (SMP) has
operated since 1937 (Reid et al. 2011), with marked fluctuations in tiger shark CPUE
seen in the first four decades of the program (Figure 2.5). The overall CPUE has not
changed significantly between 1950 and 2010 (R2 = 0.048, p = 0.0937, Reid et al.
2011), although a downward trend in CPUE was observed in the last two decades
(Figure 2.5). The size-frequency distribution of tiger sharks caught by the program has
changed over 60 years (1998 - 2007) with a significant reduction in the modal size of
tiger sharks from 3.5 to 3.0 m. The apparent decline in CPUE over the last 20 years
combined with the decrease in the proportion of large individuals suggests changes in
the population structure, and raises concern on impacts the program is having on the
tiger shark population off New South Wales (Reid et al. 2011).
In contrast, the KwaZulu-Natal beach protection program in South Africa has deployed
large-mesh gillnets off popular swimming beaches since 1952, and has found an annual
increase of 3% in the CPUE of tiger sharks between 1978 - 2003 (Dudley &
Simpfendorfer 2006) (Figure 2.6).
51
Figure 2.5 Catch per unit of effort of tiger sharks by the New South Wales Shark Meshing
Program from 1950-2010 modified from Reid et al. (2011). Dashed line indicates Loess
estimates.
Figure 2.6 Annual catch per unit of effort of tiger sharks in nets from the KwaZulu-Natal beach
protection program between 1978 and 2003. Dash line represents the slope of a linear
regression. Figure adapted form Dudley & Simpfendorfer (2006).
2.4 Life-history and ecology
Tiger sharks occur throughout tropical and subtropical waters on continental shelves,
coral reefs, offshore islands and atolls. Tiger sharks are encountered from very shallow
waters (Heithaus et al. 2002) to open ocean areas (Polovina & Lau 1993; Meyer et al.
2009; Hammerschlag et al. 2012; Werry et al. 2014; Holmes et al. 2014; Ferreira et al.
2015; Lea et al. 2015; Domingo et al. 2016; Queiroz et al. 2016). Vertical movements
have been recorded to depths of up to 1,136 m (Werry et al. 2014); however they
52
typically spend approximately 90% of time within the first 50 - 100 m of the water
column (Meyer et al. 2010; Fitzpatrick et al. 2012; Vaudo et al. 2014; Afonso & Hazin
2015).
Relative to other requiem sharks, family Carcharhinidae, tiger sharks show higher
intrinsic rates of population increase based on life history parameters (Dudley &
Simpfendorfer 2006). Size at birth has been estimated to be 50 - 90 cm (Stevens &
McLoughlin 1991; Simpfendorfer 1992; Winter & Dudley 2000; Whitney & Crow
2007) and growth rates during early life stages are thought to be high (36 - 118.4 cm yr-
1) (Branstetter et al. 1987; Natanson et al. 1999; Kneebone et al. 2008; Afonso et al.
2012). In comparison, the grey nurse shark (Carcharias taurus) is estimated to grow
between 14.5 - 18.5 cm per year in early life stages (Goldman et al. 2006). Maximum
total length attained by tiger sharks is estimated to be 6 m but individuals larger than 5
m are rarely seen (Compagno 1984; Randall 1992). The maximum age and generation
length are estimated to be 27 - 33 years and 17 - 22.5 years respectively. Tiger sharks
reach maturity between 5 - 13 years (Branstetter et al. 1987; Smith et al. 1998;
Natanson et al. 1999; Wintner & Dudley 2000; Meyer et al. 2014; Holmes et al. 2014),
with females maturing at sizes (total lengths) between 274 - 345 cm and males around
250 - 305 cm (Compagno 1984; Stevens 1984; Stevens & McLoughlin 1991; Randall
1992; Simpfendorfer 1992; Wintner & Dudley 2000; Whitney & Crow 2007; B.
Holmes, in press) although estimates of size and age at maturity vary greatly among
locations. The high level of separation between tiger shark in the Indo-Pacific and
western Atlantic (Bernard et al. 2016) suggests that hat these dissimilarities in estimates
of size and age at maturity may be due to differential population parameters among
ocean basins.
Several studies have reported aspects of tiger shark reproduction, but the complete
reproductive cycle has only been described for Hawaii (Whitney & Crow 2007). In this
location, the cycle occurs triennially, with mating thought to occur in the boreal winter
between January and February and females storing sperm in the oviducal gland while
oocytes develop. Ovulation occurs in June to July with pupping during the boreal
autumn between late September and early October of the following year after a 15 - 16
month gestation (Whitney & Crow 2007). Conflicting information on the seasonality of
reproduction is available for other parts of the world. For example, in Australia, a
female was sighted bearing fresh mating wounds during the austral autumn in early May
(LC. Ferreira, pers. obs.) and pupping is reported to occur during the austral summer
53
(Stevens & McLoughlin 1991; Simpfendorfer 1992). The number of embryos in a litter
ranges from 3 - 82 and litters have on average 26 - 33 embryos (Stevens & McLoughlin
1991; Simpfendorfer 1992; Whitney & Crow 2007). Little is also known about pupping
areas for the species. In the North Atlantic tiger sharks do not appear to use specific
areas for pupping that can be defined as nurseries, because parturition takes place over
an extensive area of shelf waters <100 m deep between the Gulf of Mexico and the
northwest Atlantic, with higher abundances of neonates recorded between the
longitudes of 30° - 33°N (Driggers et al. 2008). No information on pupping areas exists
for the Indian or Pacific Oceans and rates of juvenile mortality remain unknown.
Telemetry data show that tiger shark movements vary greatly among individuals, with
some individuals displaying elevated levels of residency in discrete areas and others
displaying extensive movements of thousands of kilometres between tropical and
temperate ecosystems, as well as seasonal migrations to open oceanic waters (Kohler et
al. 1998; Hammerschlag et al. 2012; Papastamatiou et al. 2013; Werry et al. 2014;
Holmes et al. 2014; Ferreira et al. 2015; Lea et al. 2015). Although many tracking
programs exist globally, the likely drivers of movements (e.g. prey abundance,
temperature, bathymetry, productivity, ocean currents, etc.) remain mostly unknown. In
Western Australia, residency of tiger sharks is associated with shallow depths and water
temperatures above 23°C (Ferreira et al. 2015), and in Hawaii inter-island movements
are associated with season (late summer/early spring), sea surface temperatures of 23° -
26°C and chlorophyll-a concentrations above 0.12 mg m-3
(Papastamatiou et al. 2013).
Variability in movement behaviour may be due to partial migrations, where only part of
the population performs large-scale movements while the remainder shows more
resident behaviour (Papastamatiou et al. 2013). Tiger sharks also seem to show
ontogenetic changes in habitat use and range of movement, with smaller individuals
restricting their movements to coastal areas (Werry et al. 2014; Afonso & Hazin 2015).
2.5 Threats
Tiger sharks are caught as secondary target species in multi-species shark fisheries and
are also retained as bycatch in commercial, small-scale and artisanal fisheries where
they are valued for their high quality flesh, fins, skin, liver oil and cartilage. The fins of
tiger sharks are a common component of the Hong Kong fin trade (Clarke et al. 2006).
The species has been increasingly exploited by fisheries since the 1950s largely due to
the increasing demand for shark fins (Ward-Paige et al. 2010). Tiger sharks are also
54
subject to illegal, unreported and unregulated (IUU) fishing operations and are regularly
caught in recreational fisheries and targeted by shark control programs in Australia
(Paterson 1990; Reid et al. 2011), South Africa (Dudley & Simpfendorfer 2006; Cliff &
Dudley 2011), and since 2014 at Réunion Island.
2.5.1 Shark Fisheries
Catches of tiger sharks in multi-species shark fisheries have been documented in a
number of regions including the western Atlantic (Hoey & Casey 1986; Berkeley &
Campos 1988; Bonfil 1994; Morgan et al. 2009; Carlson et al. 2012), Brazil (Bonfil
1994), Australia (Stevens et al. 1982; Lyle et al. 1984; Macbeth et al. 2009; Tillett et al.
2012), Papua New Guinea (Kumoru 2003), Peru (Gonzalez-Pestana et al. 2014),
Taiwan (Bonfil 1994), India (Bineesh et al. 2014) and Saudi Arabia (Spaet & Berumen
2015). Tiger sharks are also a common, non-target component of the US east coast/Gulf
of Mexico commercial shark bottom longline fishery that targets large coastal species
(Carcharhinus plumbeus and Carcharhinus limbatus), where the species accounts for 8-
36% of the catch (Morgan et al. 2009; Carlson et al. 2012). Tiger shark mortality rate is
estimated at 30.9% with only 14% of the catch being landed and the remaining used as
bait or discarded. Of sharks caught alive (69%) almost half are released upon capture
(Morgan et al. 2009). Most of the captured individuals are juveniles and sub-adults as
the average size of individuals caught by this fishery is 113.9 cm for females and 112.1
cm for males, with sizes ranging from 40 - 336 cm (J. Carlson pers. comm.).
Tiger sharks are relatively common in the Indonesian shark fishery and contributed
5.2% of the total biomass of sharks in catches between 2001 and 2006 (White 2007). In
Australia, tiger sharks are also targeted by commercial shark fisheries in northern New
South Wales and in Western Australia. Tiger sharks compose 3.12 t and 5.9% of total
catch of the Ocean Trap and Line (OTL) Fishery in northern New South Wales.
However, catches by this fishery are probably smaller than landings from recreational
fisheries in that same area (Park 2007; Macbeth et al. 2009). In the Western Australia
Tropical Shark Fishery, tiger sharks were caught as a secondary target species and
annual catches averaged approximately 41 t between 2000 and 2004. However, this
fishery was closed in late 2005 due to indication of overfishing the target species,
sandbar sharks (Carcharhinus plumbeus) (Department of Fisheries 2005; Department of
Fisheries Western Australia 2006). Tiger sharks are also caught in the seamount gillnet
55
and longline fishery off the west coast of India but species-specific catches are not
recorded (Bineesh et al. 2014).
2.5.2 Non-shark commercial fisheries
Tiger sharks (along with many other sharks) are taken as bycatch in a variety of
commercial fisheries that target large teleosts such as tuna and swordfish (ICCAT
2014). They are typically caught in small numbers compared to pelagic sharks that are
considered to be of high value, and, as a result, catch is often not reported or only
recorded in very coarse resolution. Catch of tiger sharks is reported for datasets that
encompass complete ocean basins (International Commission for the Conservation of
Atlantic Tunas, ICCAT) or for national longline pelagic fisheries (US longline) and for
fisheries that act within the exclusive economic zone of a country (Panama, Australia).
Tiger sharks are occasionally caught in the international tuna longline fishery in the
Atlantic Ocean regulated by ICCAT, where an average of 16 t was landed per year
between 1982 and 2013 (ICCAT database). Although the overall catch (in tons as effort
data is not available for the computation of CPUE) of tiger sharks is low, it has shown
an upward trend in the last decade, driven mostly by increased catch from US, Brazil
and Netherlands (ICCAT database). Between 2007 and 2013, annual landings by tuna
longline boats regulated by the ICCAT averaged 54 t (ICCAT database, P. de Bruyn,
pers. comm.). Tiger sharks also represented less than 10% of the total bycatch of all
shark species in each of the management zones of the U.S. pelagic longline fishery
(Mandelman et al. 2008). In the Panama longline fishery, tiger sharks represented 1.6%
of the estimated catch of 109,500 t of various sharks since the mid-1980s (Harper et al.
2014). In the central and western Pacific, tiger sharks are caught in low numbers (0.025
sharks per set of longline; P. Williams, pers. comm.) by tuna longline boats from most
countries fishing in that region (PNG, Australia, Kiribati, New Caledonia, Hawaii,
Solomon Islands, Fiji, Vanuatu, Federated States of Micronesia, Tonga, French
Polynesia, Palau). Tiger sharks are bycatch (27 t in 2012) in the Offshore Net and Line
Fishery of northern Australia (Lyle et al. 1984; Northern Territory Department of
Primary Industry and Fisheries 2012) and the species is also taken in the Southern and
Western Demersal Gillnet and Demersal Longline Fishery in Western Australia, with a
catch of 112 t reported in 2005/2006 (McAuley & Leary 2008). Tiger sharks also
constitute bycatch in squid, fish and crustacean trawl fisheries, although normally in
small numbers and there are few records of landings for these fisheries. The Australian
Commonwealth Trawl Sector reported a total catch of tiger sharks of 4.7 t from 2004 -
56
2011. Tiger sharks are caught occasionally in the longline fisheries of Costa Rica
(CPUE = 0.006 shark hook-1
×1000, Dapp et al. 2013), Mozambique (Sousa 2012) and
Saudi Arabia (<1% of all sharks, Spaet & Berumen 2015) and by purse seiners in the
Indian Ocean (Chassot et al. 2014) with no data available on catches of tiger sharks for
most regions.
2.5.3 Recreational fisheries
Tiger sharks are a common target of recreational fisheries in the United States (east
coast), Australia, South Africa and in the Gulf of Mexico. Approximately 96% of the
catch in US waters are released alive and many shark fishing tournaments have become
catch and release. Survival rates of tiger sharks are assumed to be high in catch-and-
release activities based on blood chemistry analyses (Gallagher et al. 2014). In the game
fishery off the southeast coast of Australia (New South Wales Gamefish Tournament),
tiger sharks comprised 10% of catches from 1961 to 1990 with captured individuals
ranging in mass from 21 to 560 kg (Pepperell 1992). Between 1993 - 2005, tiger shark
catches in that game fishery were approximately 8 t year-1
, comprising approximately 1-
5% of the total catch for all species and 10-30% of all sharks in that period (Park 2007).
The largest shark caught in this tournament weighed 626 kg in 1939. A decline in
CPUE from approximately 1.2 to 0.5 shark vessel-day-1
has been recorded in New
South Wales game fishing tournaments between 1993/94 and 2004/05, even though
fishers still target tiger sharks during the tournaments; thus suggesting a decline in the
population (Park 2007). Tiger sharks are probably caught by recreational fishermen in
many countries, and the tournaments documented above and recreational fishing is
likely to account for significant mortality of populations in coastal waters of other
countries, although catches and potential impacts remain unmonitored.
2.5.4 Artisanal and IUU Fisheries
Artisanal fisheries and IUU fisheries are also likely to be catching tiger sharks, however
information about landings from these fisheries is scarce as they remain mostly
unmonitored, not allowing further analysis of catch rates and trends. Tiger sharks are
commonly caught in artisanal fisheries in the tropics and subtropics, including in
Mexico ( Cartamil et al. 2011), Panama (Harper et al. 2014), Brazil (Bornatowski et al.
2014; Pimenta et al. 2014) and African countries ( Food and Agriculture Organization
of the United Nations 2014) but information on catch rates is extremely limited. In
57
Bangladesh, landings of tiger sharks of 4.48 t represent an average 1.36% of the total
shark catch (Jit et al. 2014). Although gear limitations probably preclude the capture of
large individuals in artisanal fisheries, catches are often not reported and it is not
possible to fully quantify the impact that these fisheries may have on the population
status of tiger sharks.
Tiger sharks compose 19% of the total shark biomass and 7.4% of total catch in
numbers from Indonesian and Taiwanese IUU fishing vessels (Marshall 2011). There is
also evidence that tiger sharks have been overfished by illegal Indonesian fishing boats
at Ashmore, Cartier and Scott Reefs in northern Australian waters. This area had been
targeted by Indonesian fishermen since the 1800s but fishing was banned in 1988 at
Ashmore Reef and 2000 at Cartier Reef (Meekan et al. 2006). A study using baited
remote cameras showed an absence of tiger sharks at reefs historically fished despite the
18 and 6 years of protection, respectively, whereas they were recorded as present at
nearby atolls that had been always protected from fishing (Meekan et al. 2006).
2.5.5 Shark Control Programs
Shark control programs are established on the basis that by reducing the abundance of
large shark species, human safety will be increased. The evidence as to whether such
programs reduce the probability of shark/human interactions is inconclusive. In
Australia, the Queensland Shark Control Program (QSCP) captured 4,757 tiger sharks
between 1993 and 2010 (Holmes et al. 2012) and the species represented approximately
10 to 30% of total catch in the northern locations of the QSCP between 1964 - 2007
(Simpfendorfer et al. 2010). This pattern is most likely due to the replacement of nets
with drumlines, which are particularly effective in catching tiger sharks (Simpfendorfer
et al. 2010). In southeast Australia, tiger sharks were commonly captured in low
numbers by the New South Wales (NSW) Shark Meshing Program with a total of 352
sharks caught between 1950 and 2008, representing approximately 10% of the common
shark species caught by the program in each of its locations (Reid et al. 2011). The
NSW program also showed marked fluctuations in CPUE of tiger sharks, with peaks
occurring approximately every 10 years (Green et al. 2009; Reid et al. 2011). In South
Africa, the KwaZulu-Natal beach protection program has caught an average of 48.7 (±
13.3 SD) tiger sharks annually from 1978 and 2003, which represents a catch rate much
lower when compared to dusky sharks Carcharhinus obscurus (232.7 ± 113.3) and sand
tiger sharks Carcharhinus taurus (194.4 ± 101.8) (Dudley & Simpfendorfer 2006).
58
Since 1989, all sharks caught alive by the KwaZulu-Natal program are released;
however, mortality of tiger sharks in nets is 27% of captures (Dudley & Simpfendorfer
2006).
2.6 Assessment Rationale
In summary, tiger sharks are a large, highly mobile top order predator with a worldwide
distribution throughout the world’s tropical and warm temperate oceans. Growth and
reproductive rates are known for some areas within the species distribution range and
are comparatively higher than rates for other large sharks. These biological
characteristics suggest that tiger sharks can support some levels of exploitation.
However, the triennial reproductive cycle suggested for the species could greatly reduce
its ability to recover from fishing pressure. Moreover, we still lack information on
natural mortality for the species, especially in terms of juvenile survivorship. Tiger
sharks are caught by several commercial, recreational and artisanal fisheries as a target
species or as bycatch. However, records on the actual catch of tiger sharks by many
fisheries globally are still mostly under-reported or unknown. The species is also one of
the major components in catches from shark control programs that target large shark
species in Australia, South Africa and Reunion Island.
My assessment of the tiger shark remains as “Near Threatened”. The species is close to
being classified as Threatened under criterion A3bd (Figure 2.6, IUCN 2012). The basis
for this assessment is: projected declines in CPUE for eastern Australia that indicated
potential significant reduction in the population in that region. However, tiger sharks
have a wide-ranging distribution, and although they are caught in a variety of fisheries
that represent a major threat for the species, population estimates for the species are
scarce. Biological traits of tiger sharks such as rapid growth rates, relatively high
reproductive output, moderate rates of intrinsic population increase and low estimated
fishing mortality indicates resilience to fishing but also creates vulnerabilities,
particularly regarding the three year reproductive cycle and the lack of information on
juvenile mortality. Information on stock assessment, population structure and overall
population trends for the tiger sharks are still lacking for much of the species
distribution and available trends in CPUE that could be used for this assessment were
based on three long-term datasets (north-eastern Australia, South Africa and the
Northwest Atlantic). The dataset from the Queensland Shark Control Program in
Australia is the longest available for the species and allowed projections of trends in
59
CPUE to be calculated, which suggested a marked decrease of around 95% in CPUE
over the length of three generations. However, these projections should be interpreted
with caution, as they are based on regression relationships that do not account for
demographic metrics or the fluctuation in the catch rate and considering that the time
series used in the predictions is relatively short.
Evidence of regional declines is compounded by the continuing exploitation of tiger
sharks by commercial, recreational and unregulated fisheries. Catches of the species
remain unrecorded by many fisheries and the low numbers of sharks caught by
regulated, commercial fisheries do not allow for population trends to be calculated at
this time, despite raising concern over population trajectories. Projections from the
KwaZulu-Natal beach protection program in South Africa suggest an increase in
populations over a period of 25 years. The program releases sharks that are caught alive,
which might have contributed to the recorded increase in CPUE. Although the rate of
increase in South Africa is lower than the projected decline seen in Queensland, it
provides estimates of population trends from standardised catches and suggests that the
meshing program is not impacting the tiger shark population of southern Africa. In the
northwest Atlantic, increases of 6 - 8% per year in relative abundance (index of absolute
abundance based on standardised catch) were also recorded by the Southeast Fisheries
Science Centre and the Northeast Fisheries Science Centre (NOAA) for the bottom
longline shark fishery and fisheries independent longline surveys that are active
throughout the Gulf of Mexico and US eastern coastline. However, this rate of increase
should be treated with caution, since the time series of data has a much shorter temporal
range than data from Queensland.
It is recommended that further research is required to accurately assess exploitation
rates by fisheries and to define population size, structure and trends throughout the
species range. It is also recommended that the trends from shark control programs in
Australia should continue to be monitored in relation to the Red List criteria (A3bd) and
the projections made by this assessment. Continuing declines in Queensland and New
South Wales may possibly determine if these populations have attained a level of threat
that would recommend the update of the status assessment to “Vulnerable”. Recent
evidence based on global genetics of tiger sharks indicated separation between the
populations in the western Atlantic and Indo-Pacific (Bernard et al. 2016). Based on
these results, it is also recommended that tiger shark populations from different ocean
basins and regions should be assessed separately by the IUCN. Regional assessments
60
could consider CPUE trends reported from the northwest Atlantic and South Africa
separately from the trend in Australia and will potentially better reflect the threat level,
population trends and conservation status for each population of tiger sharks.
Figure 2.7 Summary table of the criteria (A - E) used to evaluate Red List status (IUCN 2012).
61
Chapter 3 Crossing latitudes - long-distance
tracking of an apex predator
3.1 Abstract
Tiger sharks (Galeocerdo cuvier) are apex predators occurring in most tropical and
warm temperate marine ecosystems, but we know relatively little of their patterns of
residency and movement over large spatial and temporal scales. We deployed satellite
tags on eleven tiger sharks off the north-western coast of Western Australia and used the
Brownian Bridge kernel method to calculate home ranges and analyse movement
behaviour. One individual recorded one of the largest geographical ranges of movement
ever reported for the species, travelling over 4000 km during 517 days of monitoring.
Tags on the remainder of the sharks reported for shorter periods (7 - 191 days). Most of
these sharks had restricted movements and long-term (30 - 188 days) residency in
coastal waters in the vicinity of the area where they were tagged. Core home range areas
of sharks varied greatly from 1166.9 to 634,944 km2. Tiger sharks spent most of their
time in water temperatures between 23° - 26°C but experienced temperatures ranging
from 6°C to 33°C. One shark displayed seasonal movements among three distinct home
range cores spread along most of the coast of Western Australia and generalized linear
models showed that this individual had different patterns of temperature and depth
occupancy in each region of the coast, with the highest probability of residency
occurring in the shallowest areas of the coast with water temperatures above 23°C.
These results suggest that tiger sharks can migrate over very large distances and across
latitudes ranging from tropical to the cool temperate waters. Such extensive long-term
movements may be a key element influencing the connectivity of populations within
and among ocean basins.
3.2 Introduction
Throughout human history, top-order predators have been disproportionately and
negatively affected by anthropogenic activities, both directly through human behaviour
such as hunting and indirectly by alteration of habitat and depletion of their food
resources (Jackson et al. 2001; Sydeman et al. 2006; Beschta & Ripple 2009; Estes et
al. 2011). We now know that the ecological impacts of our elimination of top-order
predators can be severe, leading to shifts in the composition and resilience of
62
ecosystems through processes such as mesopredator release and trophic cascades (Pace
et al. 1999; Heithaus et al. 2008a; Estes et al. 2011). Globally, the removal of sharks
due to fishing has accelerated rapidly in recent decades, with the result that many
species are now threatened with or vulnerable to extinction in many regions (Baum et
al. 2003; Dulvy et al. 2008, 2014a; Field et al. 2009; Ward-Paige et al. 2010; Ruppert et
al. 2013). There is a growing recognition of the importance of sharks as keystone
species in the structuring of marine ecosystems through their influence on species
composition, biomass and the trophic roles of prey assemblages (Estes et al. 2001;
Johnson et al. 2007; Heithaus et al. 2010; Ruppert et al. 2013). However, we remain
largely unaware of some of the most basic aspects of the ecology of many species,
including movement patterns and habitat requirements. As top-order predators, sharks
tend to have large body sizes (Heupel et al. 2014) and thus generally require large areas
in which to forage. For this reason, they are likely to undergo long distance (10 - 1000s
km) movements that could bring them into contact with multiple habitats, ecosystems
and anthropogenic threats (Harestad & Bunnel 1979; McCauley et al. 2012; Heupel et
al. 2014).
Tiger sharks (Galeocerdo cuvier) are one of the largest sharks, growing to over 5 m
(Compagno 1984; Randall 1992; Holmes et al. 2012) and the species is both an apex
predator and scavenger that occurs in most tropical and warm-temperate marine
ecosystems. They feed on a wide array of prey (Grudger 1949; Clark & von Schmidt
1965; Stevens 1984; Stevens & McLoughlin 1991; Randall 1992; Simpfendorfer 1992;
Lowe et al. 1996; Heithaus 2001; Simpfendorfer et al. 2001), on which they exert both
lethal and behavioural risk effects (Heithaus et al. 2007b). Globally, anthropogenic
threats to populations of tiger sharks include commercial fisheries (Polovina & Lau
1993; Bonfil 1997; Castillo-Geniz et al. 1998; Baum et al. 2003; Handley 2010) and
illegal, unreported and unregulated (IUU) fishing (Field et al. 2009). Moreover, they
have been targeted by shark control programs as a species potentially dangerous to
humans (Paterson 1990; Wetherbee et al. 1994; Dudley 1997; Dudley & Simpfendorfer
2006; Hazin et al. 2008; Green et al. 2009) and evidence of declines in populations of
tiger sharks have been reported by beach meshing programs in some areas of Australia
(Reid et al. 2011; Holmes et al. 2012). However, while the species is classified as “Near
Threatened” by the International Union for the Conservation of Nature (IUCN) due to
evidence of declines in some populations (Simpfendorfer 2009), broad-scale trends in
abundances are still unknown.
63
Given the anthropogenic threats to these apex predators, information on their movement
behaviour, particularly over large spatial and temporal scales (100s - 1000s of km,
months to years) is essential for the development of appropriate management and
conservation strategies (Sims 2010; Block et al. 2011; Papastamatiou & Lowe 2012;
Drymon et al. 2013). To date, most information on the horizontal movements of tiger
sharks is available from SPOT and PAT satellite tracking studies conducted at relatively
small temporal scales (<1 year; e.g. (Heithaus et al. 2007c; Meyer et al. 2010;
Fitzpatrick et al. 2012; Hazin et al. 2013; Werry et al. 2014)). The information provided
by SPOT transmitters is usually limited by the relatively short intervals that sharks
spend on the surface, which results in low numbers of location estimates, obtained at
irregular intervals and typically with low spatial resolution (Heithaus et al. 2007c;
Fitzpatrick et al. 2012). Additionally, physical damage, biofouling and premature
shedding of tracking devices are also widely reported (Kerstetter et al. 2004; Heithaus et
al. 2007c; Hays et al. 2007; Meyer et al. 2010), resulting in short and sparse location
data sets. For PAT tags in particular, deployment periods are commonly much shorter
than programmed due to tag damage and biofouling that may cause premature release of
the tag. Longer-term tracking studies have used passive acoustic telemetry to monitor
movements of tiger sharks (e.g. (Holland et al. 1999; Meyer et al. 2009; Papastamatiou
et al. 2013; Werry et al. 2014)), but these are limited by the detection range of receivers
and the number and scale of receiver arrays (Heupel et al. 2006). Overall, these studies
have shown high individual variability in movement patterns of tiger sharks, with some
degree of residency in particular habitats interspersed with occasional forays into the
open ocean (Heithaus et al. 2007c; Meyer et al. 2010; Werry et al. 2014).
Due to the short duration of most tracking datasets it is unclear whether the wider
movements across scales of 1000s of km, typical of other large sharks such as basking
(Cetorhinus maximus), white (Carcharodon carcharias) and whale sharks (Rhincodon
typus) (Boustany et al. 2002; Bonfil et al. 2005; Bruce et al. 2006; Wilson et al. 2007;
Skomal et al. 2009; Farmer & Wilson 2011; Domeier & Nasby-Lucas 2013) might also
exist for tiger sharks. Our study reports on the results of the deployment of eleven
satellite transmitters on tiger sharks, one of which produced one of the longest duration
tracks ever recorded for the species (517 days). In combination with the results from the
other deployments, we examined horizontal and vertical movements of tiger sharks and
quantified habitats of high and low use. We hypothesise that tiger sharks display a mix
64
of both restricted and transient movement, and that residency patterns will be driven by
water temperature and bathymetry.
3.3 Material and Methods
Ethics statement
This project was conducted under permit number SF6104, WA Fisheries permit 2007–
30–32, and ethics approvals A07035 (Charles Darwin University Ethics Committee)
and DPIW 7/2007–08.
Tiger sharks were caught using longlines off Ningaloo Reef, Western Australia, in June
2007, August 2008 and May-June 2010. Between 118 and 350 hooks were set at
approximately 10 m intervals along one to five demersal longlines deployed on the
seaward side of the reef each day. Longlines were usually deployed at dawn, with a few
deployments at dusk. Short soak times of between 2.2 h to 5.2 h were used to maximise
the survival rates of captured sharks. One shark (Shark 7) was caught by rod and reel
between longline sets. Hooked sharks were brought on deck or restrained in a stretcher
at the stern of the vessel, measured and sexed. Eleven tiger sharks were instrumented
with fin-mounted satellite-linked transmitters (SPOT4, SPOT5 or SPLASH tags,
Wildlife Computers, Redmond, Washington, USA); however, three did not report any
data after they were deployed (Table 3.1). All reporting transmitters relayed satellite
positions via the ARGOS satellite network and time-at-temperature histograms in 14
user-defined temperature ranges from 0°C to 60° (± 0.2°C). Two SPLASH tags (Table
3.1) also relayed summaries of time-at-depth with user-defined depth ranges of 0-600 m
and 0-800 m (± 0.5 m).
Position estimates were provided by ARGOS with an associated error (Location Class
(LC) of 3 (<250 m), 2 (250–500 m), 1 (500–1500 m), 0 (>1500 m), A and B (not
specified), www.argos-system.org). A small amount (0.5%) of location points were
excluded or substituted by the secondary location estimate reported by ARGOS because
they were obviously erroneous, i.e. they were well beyond the bounds of possible
distances the shark could have travelled based on both earlier and later location
estimates for the track. An analysis of the travel speed found that these erroneous
locations required a travel speed of >1000 km per day. More advanced filtering methods
were attempted, such as a Bayesian state-space model; however, the models did not
converge, probably due to the sparseness of the data.
65
We calculated the home range of tiger sharks using the Brownian Bridge kernel method
using the adehabitatHR package in R software V2.15.3 (Calenge 2011; R Core Team
2013). This method takes into account not only the shark locations, but also the path
travelled by the animal between successive locations (Bullard 1991; Horne et al. 2007)
by applying a conditional random walk to model the expected path between locations.
Two smoothing parameters were set: sig1, which controlled the width of the “bridge"
connecting successive positions (this is the Brownian Bridge motion variance parameter
- BMV); and sig2, which was related to the imprecision of the positions. Values of sig1
were chosen using the function liker that implemented the maximum likelihood
approach developed by Horne (Horne et al. 2007) and sig2 was set at 1.9 km, the
median ARGOS location error across pooled position classes (Costa et al. 2010b).
Tracks from Sharks 1 and 2 were removed from the analysis due to their very short
length (7 and 14 days respectively). Because of the possibility of effects of tagging on
behaviour, we assumed that the first two weeks might not be representative of the
shark’s natural behaviour (Afonso & Hazin 2014). Location estimates from Sharks 3
and 8 were particularly sparse (0.18 and 0.04 locations per day respectively), and Shark
3 had a very high Brownian Bridge motion variance parameter compared to the scale of
the shark’s movements (sig1 = 822.5). Consequently, these tracks were not included in
further analyses of home range.
The proportion of observations in each temperature and depth bin was calculated from
time-at-temperature and time-at-depth histograms provided by the tag to determine
thermal and depth range of all sharks. For one shark (Shark 5), time-at-temperature
histograms associated with latitude/longitude information were used to assess variation
in the proportion of time at temperature. Water temperature profiles were constructed
from data downloaded from IMOS floats (IMOS, http://imos.aodn.org.au/imos/) located
in the vicinity of position uplinks within a week of the time the uplinks were recorded
(South coast of WA: 35.378°S, 119.531°E; North coast of WA: 17.604°S, 117.657°E).
Maximum possible diving depth of Shark 5 was estimated by comparing the minimum
temperature registered by the shark’s tag with profiles of water temperature from IMOS
floats in the same region where the shark was resident. The maximum depth of descent
was assumed to be the greatest depth in the water temperature profile where the
minimum temperatures reported by the tag and those of the water temperature profile
were the same. For the track with multiple home range cores (Shark 5), movement
patterns were categorised as within and outside the 25% utilisation distribution. we then
66
used generalised linear models with a binomial distribution and a logit link function to
assess the relationship between the probability of the shark being in a home range core
and water temperature, bathymetry and region of the WA coast (north - latitude < 24°S
and south - latitude >24°S). We were not able to fit all three explanatory variables in
one model as there was no overlap of the temperature ranges between the two regions.
Consequently, we fitted a model to examine the probability of being in a home range
core in relation to bathymetry and region and two separate models (using the data from
north and south coasts, respectively) to examine the probability of the shark being in a
home range core in relation to sea surface temperature. To address the autocorrelation
present in the data we used a matched-block bootstrap sampling for all models with
replacement procedure (Carlstein et al. 1998; Politis & White 2004) that resampled
blocks of data randomly and then recombined them in a random order, creating a
bootstrapped dataset that minimized the effect of autocorrelation (Carlstein et al. 1998;
Politis & White 2004; Patton et al. 2009). Model fitting was applied to 100
bootstrapped samples and model selection used the sample-corrected Akaike’s
information criterion (AICc), AICc weight (wAICc), and percent deviance explained
(%DE) (Burnham & Anderson 2004; Vianna et al. 2013). Bathymetry data with a grid
resolution of 2’ from ETOPO1 database hosted by the NOAA was obtained by the R
software package marmap (Pante & Simon-Bouhet 2013). Daily Sea Surface
Temperature was obtained through the daily Optimum Interpolation Sea Surface
Temperature (OISST) analysis (Reynolds et al. 2007) on a 0.25 degree
latitude/longitude grid from NOAA's National Climatic Data Centre
(ftp://eclipse.ncdc.noaa.gov/pub/OI-daily-v2/NetCDF).
Table 3.1. Details of satellite transmitter deployments on tiger sharks.
ID Fork
length
(cm)
Sex Date
deployed
Location Type Duration of
transmission
(days)
Locations day-1
(mean ± SD)
1 145 F 19/06/2007 23.32°S 113.71°E SPOT 7 0.75 (±1.16)
2 276 F 21/06/2007 22.97°S 113.76°E SPLASH 14 0.92 (±1.50)
3 179 F 17/08/2008 21.86°S 113.96°E SPOT 105 0.18 (±0.43)
4 214 F 19/08/2008 22.38°S 113.73°E SPOT 70 0.54 (±0.83)
5 222 F 19/08/2008 22.42°S 113.71°E SPOT 517 0.43 (±0.95)
6 333 M 27/05/2010 23.07°S 113.70°E SPOT 191 1.86 (±1.66)
7 224 M 2/06/2010 21.58°S 114.52°E SPOT 38 0.60 (±0.72)
8 240 F 30/05/2010 22.27°S 113.72°E SPLASH 154 0.04 (±0.19)
9 155 F 19/06/2007 23.35°S 113.72°E SPOT Did not report
10 252 F 21/06/2007 22.97°S 113.76°E SPLASH Did not report
11 254 F 20/08/2008 22.66°S 113.60°E SPLASH Did not report
67
3.4 Results
The time that data was received from the tagged sharks varied greatly among
individuals (7 to 517 days; Table 3.1). 49% of locations were from ARGOS Location
Class B, 20% were from Location Class A and only 31% of locations were within
Location Classes 3-0. Four satellite transmitters provided data for less than 100 days but
even the longest deployments had long periods of no transmission. For example, Shark
5 had a period of 118 days between May and September 2009 when no location data
were transmitted, while Shark 8 did not transmit data between July and October 2010.
Transmitters deployed on three sharks (1, 2 and 7) provided data for only 7, 14 and 38
days respectively. During this time they occupied waters with average depths of 42.2 m
(± 26.8 m SD) and their movements were restricted to the vicinity of where they were
initially tagged (Figure 3.1a) on the shelf. Three sharks (3, 4 and 6; Figure 3.1b-c)
stayed within shelf waters that averaged 117.5 m deep (± 113.0 m SD). Shark 6’s
movements were restricted to a relatively small area (1166.9 km2) off the Ningaloo Reef
for six months (Figure 3.1c). Shark 8 moved 303.4 km from the point of first capture
and tagging into waters over 5000 m deep, where the last data transmission was
received (Figure 3.1b). There were six position estimates in this area and the data were
very limited for the duration of tracking (154 days between tagging and the last uplink).
One shark (Shark 5) ranged over 4000 km and apart from one period of approximately
three and a half months, provided relatively frequent transmissions over 517 days of
monitoring (Figure 3.1d).
68
Figure 3.1 Movement patterns of tiger sharks in Western Australia. Maps show location uplinks
of 8 tiger sharks. Triangles indicate tagging location and grey polygons indicate Commonwealth
Marine Reserves.
The kernel utilisation distributions of all sharks indicated movement between coastal
regions and islands or atolls off the north coast of Western Australia, with predominant
use of coastal waters (Figure 3.2). The 50% kernel utilisation distributions varied greatly
in area from 1166.9 km2 to 634,944 km
2 among sharks (Figure 3.2). Overall, a total of
56.7% of all locations received from the satellite transmitters were within the
Commonwealth Marine Reserve network. All sharks had some locations inside marine
reserves and six of the sharks had 50% or more locations inside reserves (Figure 3.1).
69
Figure 3.2 Home range of tiger sharks. The utilisation distribution calculated using the
Brownian Bridge kernel method. Black line represents the 50% Brownian Bridge home range
distribution.
Shark 5, a 222 cm female, was monitored for 517 days and ranged from 10.4°S to
35.8°S of latitude and from 113.0°E to 124.1°E of longitude (Figure 3.1d). The
transmitter was deployed on this shark at Ningaloo Reef and the shark then moved
along the 500 m bathymetric contour to the Rowley Shoals and Kimberley region. It
then made a path to Sumba Island, Indonesia, and returned, crossing ocean depths of 5
km and covering a distance of more than 1000 km in 2 weeks. In December 2008 the
shark moved south, traveling to waters off Jurien Bay and Perth between January-
February 2009. Between April and May 2009 the shark rounded Cape Leeuwin with
transmissions clustering off Albany. After a period of no transmissions (118 days), the
tag then started to transmit again in September when the shark moved towards the north,
returning again to Perth/Jurien Bay in January 2010. The shark had three distinct areas
of 50% Brownian Bridge kernel utilisation that included a large area off the north coast
of Western Australia, another off Jurien Bay and Perth and a smaller one off the coast of
Albany (Figure 3.2).
70
Table 3.2 Ranked Generalised Linear Models with bootstrap sampling of the probability of a shark being in a 25% utilisation distribution explained by
bathymetry (Depth) and region of the coast (Region).
Maximum log-likelihood (LL), degrees of freedom (df), Akaike’s information criterion corrected for small samples (AICc), AICc weight (wAICc), the %
deviation explained (%DE), lower quantile LCI (25th quantile) (25%) for LL.25, wAICc.25, AICc.25 and %DE.25 and upper quantile (75
th quantile) for
LL.75, wAICc.75, AICc.75 and %DE.75.
Table 3.3 Ranked Generalised Linear Models with bootstrap sampling of the probability of a shark being in a 25% utilisation distribution explained by
sea surface temperature (Temperature) at the north and south coast separately.
Maximum log-likelihood (LL), degrees of freedom (df), Akaike’s information criterion corrected for small samples (AICc), AICc weight (wAICc), and
the % deviation explained (%DE), lower quantile LCI (25th quantile) (25%) for LL.25, wAICc.25, AICc.25 and %DE.25 and upper quantile (75
th
quantile) for LL.75, wAICc.75, AICc.75 and %DE.75.
Model LL df AICc wAICc %DE LL.25 LL.75 wAICc.25 wAICc.75 AICc.25 AICc.75 %DE.25 %DE.75
Home ˜ Depth -82.456 2 168.967 0.508 34.49 -124.637 -101.77 0.001 0.317 207.595 253.328 5.75E+00 20.347
Home ˜ Depth +
Region -82.010 3 170.131 0.284 34.84 -122.153 -97.006 0.167 0.5143 200.125 250.417 6.85E+00 23.756
Home ˜
Depth*Region -81.283 4 170.751 0.208 35.42 -117.994 -95.781 0.221 0.649 199.749 244.173 9.53E+00 25.453
Home ˜ Region -125.803 2 255.661 0.000 0.05 -134.792 -118.812 0 0.001 241.678 273.639 2.98E-01 4.687
Home ˜ 1 (Null) -125.863 1 253.744 0.000 0.00 -137.028 -124.8 0 0 251.619 276.075 -1.91E-14 0
Model LL df AICc wAICc %DE LL.25 LL.75 wAICc.25 wAICc.75 AICc.25 AICc.75 %DE.25 %DE.75
North
Home ˜ Temperature -94.845 2 193.756 0.999 9.67 -101.645 -80.5813 0.748725 1 165.2297 207.3572 0 0
Home ˜ 1 (Null) -111.514 1 212.024 0.000 0.00 -111.514 -91.553 0 0.251 185.128 225.05 2.184 25.802
South
Home ˜ Temperature -16.535 2 37.337 1 44.54 -
13.75475 0 1 1 4.267 31.7765 0 0
Home ˜ 1 (Null) -29.812 1 61.711 0.000 0.00 -32.895 -29.812 0 0 61.711 67.877 30.254 57.026
71
Figure 3.3 Time spent in each temperature bin. Plots show the percentage of time spent within
the specified temperature bins for each tiger shark (ID number in the top right hand corner of
each plot).
Temperatures experienced by sharks ranged from 6°C to 33°C. Overall, tiger sharks
spent most of their time in temperatures between 23°-26°C (Figure 3.3). The shark with
the widest latitudinal range of movements (Shark 5) also experienced the greatest range
of temperatures (from 10°-33°C), but spent approximately 78% of time in temperatures
between 23°-29°C. It also experienced different ranges of temperature in each home
range (Figure 3.4). Off the Kimberley and Pilbara coasts (10°-24°S of latitude) the
72
temperature range experienced by the shark was much greater (10°-33°C) than when the
shark visited the south coast off Albany (35°S of latitude; 18°-23°C).
Figure 3.4 . Latitudinal variation of temperature profiles for Shark 5. Percentage of time spent
in each temperature bin at each region of Western Australia for Shark 5.
Figure 3.5 Time spent in each depth bin. Percentage of time spent in each depth bin for two
tiger sharks tagged with SPLASH tags.
Figure 3.6 Vertical profiles of water temperature. Plots show temperature profiles recorded by
Argo floats of the north (17.6°S 117.6°E) and south (35.78°S 119.5°E) sections of the WA
coast. Vertical lines represent the minimum temperature reported by the shark’s satellite
transmitter and horizontal lines represent the estimated diving depth.
73
Time-at-depth histograms of the two sharks with SPLASH tags showed that these
individuals differed greatly in depth range (Figure 3.5). The first (Shark 2) did not dive
deeper than 150 m, with 27% of observations between the surface and 5 m and 97% in
water depths up to 75 m. Locations for this shark were all close to the coast in areas
around the 100 m isobath. The second (Shark 8), which moved to pelagic waters
experienced a greater range of depths. Though this shark spent 57% of time between the
surface and 10 m, it had a maximum recorded depth of 400 m, with approximately 20%
of time spent at 150 m.
Water temperatures reported by Shark 5, which was tracked for 517 days, were used to
estimate diving depths. Analyses of the vertical profiles of water temperature in each
region where location data were recorded indicated that the shark was diving up to 380
m in tropical latitudes, but not descending below 100 m in temperate latitudes (Figure
3.6). Sea surface temperatures were consistent with a strong Leeuwin Current that
supplied a warm mass of water around the Cape Leeuwin and Albany coasts with
temperatures of 19°-23°C during the time locations were registered in this region
(Figure S 3.1). Here, the shark spent 82% of time in waters between 21-23°C (Figure
3.4), indicating that it remained relatively close to the surface.
Figure 3.7 Generalised linear model for Shark 5. Generalised linear model (GLM) predicted
probabilities (solid line) of a shark being in a 25% utilisation distribution in relation to (a) water
depth, (b) water temperature in the north and (c) in the south. Dotted lines show the standard
error.
74
The highest ranking model describing whether the shark remained in a 25% home range
core (i.e. remained resident and did not switch to transit behaviour), included water
depth only, and explained 34.5% of the deviance in the data set (Table 3.2). There was a
negative relationship between the probability of the shark being in a 25% home range
core and water depth, meaning the shark was more likely to be resident (in a home
range core) in the shallower areas and less likely to remain resident in deeper waters
(Figure 3.7). Even though the tiger shark crossed areas over 5000 m deep while
migrating, it had low probability of being in a home range core in deeper water (> 1000
m). Temperature explained 9.7% of the deviance in the probability of Shark 5 being in a
home range core on the north coast and 44.5% in the south (Table 3.3). In both regions,
the shark had a greater than 80% chance of being in a home range core in temperatures
of 23-24 °C, suggesting that this was a preferred temperature in both regions (Figure
3.7).
3.5 Discussion
Tiger sharks are typically considered to be residents of tropical and warm temperate
habitats (Randall 1992), but individuals appear on a seasonal basis in temperate or cool
temperate waters (Last & Stevens 1994). The most extreme examples of this pattern are
the occasional records of tiger shark catches in the waters of countries such as Iceland
(Matsumoto et al. 2005) and far off the south coast of New South Wales in Australia
(Pepperell 1992). Such occurrences seem to be related to the influx of warm waters
brought by western boundary currents such as the Gulf Stream and the East Australian
Current (Schmitz & McCartney 1993; Reid 1994; Cresswell 2000). Similarly, in
Western Australia the Leeuwin Current flows southward bringing warm, low salinity
water along the shelf into the cool temperate environments off Albany (35°S) (Smith et
al. 1991). In our study, one shark moved into temperate waters twice, both times during
the summer months of January-February. This implies that such behaviour is most
likely to be the result of directed and seasonal migrations, rather than simply a
haphazard event. Other evidence to support this idea of regular southward movements
of tiger sharks during the austral summer comes from data from beach protection
programs, which show that tiger sharks are more common off the coast of New South
Wales, a temperate region in the southeast of Australia, during summer (Krogh 1994;
Reid et al. 2011), when the warm East Australian Current flows southward. A recent
tagging study found that a tiger shark moved to warmer waters off Queensland during
winter and went south to 37°S latitude during the austral summer (Holmes et al. 2014).
75
Similarly, the abundance of tiger sharks in the Aliwal Shoals off South Africa increases
during summer months (Dicken & Hosking 2009).
The shark we tracked for 517 days had home range core areas in both the tropics (15°-
20°S) and the cool-temperate coast off Albany (35°S). The movement of this shark to
temperate waters off Perth occurred twice in consecutive years, both times during
January. Overall, this track was the longest and covered one of the greatest range of
latitudes recorded by satellite telemetry for the species. However, such movements are
not entirely without precedent. For example, a fisheries tagging study recorded tiger
sharks travelling more than 3,400 km in the Atlantic (Kohler et al. 1998) and recent
satellite tagging studies have shown tiger sharks swimming distances of over 1000 km
in the South Pacific (Werry et al. 2014; Holmes et al. 2014), and as far as 3500 km from
the tagging site in the North Atlantic (Hammerschlag et al. 2012). A study by Heithaus
et al. (2007b) recorded a potential movement of a tiger shark of 8000 km from the point
where it was tagged in Shark Bay, WA to the coast of South Africa. However, this
result was inferred from a single low-quality position fix that could not confirm if the
tag was still attached to the shark or determine if the position estimate was simply an
artefact of the position estimation algorithm. In our study, associated water temperature
data reported by the tag and multiple position fixes showed that the track of the shark
was both reliable and that the tag remained deployed on the shark. In terms of
geographic scale, the movements of this tiger shark were comparable to those of white
sharks (Carcharodon carcharias), which in certain regions appear to display seasonal
residency at cool temperate coastal locations, mostly in areas with high abundances of
pinnipeds (Bruce 1992; Klimley et al. 2001a; Domeier & Nasby-Lucas 2007; Johnson
et al. 2009), interspersed with long oceanic migrations into the tropics (Boustany et al.
2002; Bonfil et al. 2005; Bruce et al. 2006; Domeier & Nasby-Lucas 2008).
The durations of residency and distances of migration of tiger sharks recorded by our
study appear to be at least partly related to the length of tag deployment. When tags
reported for a short time, movements of sharks tended to be restricted to one home
range core within a relatively small area. However, longer-term tag deployments
revealed a very different pattern, where periods of residency were interspersed with
long distance (100 -1000s km) movements that appeared both directed and predictable.
However, some individuals appeared highly resident despite relatively long tagging
records (up to 6 months) suggesting that two forms of movement may be present. In our
study, home range cores (averaging 4,474.1 km2, excluding Shark 5) were consistent
76
with earlier work that has shown movement patterns of tiger sharks off the east coast of
Australia (Holmes et al. 2014) and the Florida coast (Hammerschlag et al. 2012).
Similar to our results, most other studies of tiger sharks have also shown that the degree
of residency varies greatly among individuals. Fitzpatrick et al. (2012) found that some
sharks remained in the area of Raine Island on the Great Barrier Reef throughout the
year, while others ventured into the Coral Sea and to the Torres Strait and northern
Great Barrier Reef. Werry et al (Werry et al. 2014) found that tiger sharks could reside
year-round at the atolls of the Chesterfield Reefs, with other sharks venturing over 1000
km into the open ocean. Meyer et al. (2010) found that some tiger sharks were resident
on the French Frigate Shoals of Hawaii throughout the year, while others were recorded
at this locality only when fledging seabirds were available as a food source during
summer. It has been suggested that variations in movement behaviour of tiger sharks
might be due to partial migrations, where only a part of the population would be
transient and perform large scale movements while the other part would show resident
behaviour (Papastamatiou et al. 2013; Werry et al. 2014). Although our study was
comprised of a small sample size, recent tagging studies in Florida, Bahamas, Australia
and New Caledonia have consistently demonstrated that a few individuals can move
over large distances while most of the remainder sharks show more restricted residency
patterns (Hammerschlag et al. 2012; Werry et al. 2014; Holmes et al. 2014). Analyses
involving larger datasets are needed to understand the characteristics of individuals that
show these partial and/or complete migrations, to determine their prevalence and to
identify the drivers that lead to differential patterns of movement.
Despite the number of satellite and acoustic telemetry studies now describing the
horizontal movements of tiger sharks (approximately 10 to date) it is difficult to
determine if there are consistent patterns in the life stage or sex of individuals that are
likely to remain resident or migrate. Meyer et al. (2009) suggested that juveniles were
more likely to display broad-scale patterns of movement than other components of the
population, a pattern also recorded in Northeast Brazil (Hazin et al. 2013). The
individual that we recorded moving the greatest distance on the WA coast was also a
sub-adult (female). In contrast, Papastamatiou et al. (2013) found that adult females
were the most likely to migrate among the islands of Hawaiian Archipelago, a pattern
they attributed to the movement of females to pupping sites. However, these researchers
mostly relied on records generated from acoustic tracking and a receiver array in coastal
waters, limiting the possibility of recording movements of sharks into the open ocean.
77
The tagged tiger sharks showed preferences for temperatures between 23° and 26°C,
consistent with surface waters off Ningaloo Reef (Taylor & Pearce 1999), although the
sharks experienced temperatures as low as 6°C. These lower temperatures suggest that
sharks were descending below the thermocline (± 100-200 m depths at Ningaloo Reef;
Rousseaux et al. 2012). Data from a SPLASH tag deployed on Shark 8 recorded
occasional transits to depths of 400 m, although time-at-depth histograms showed a
preference by this shark for shallow waters (65% of time in depths < 50 m) within the
mixed layer of the water column, similar to patterns reported by other studies (Holland
et al. 1999; Meyer et al. 2010; Nakamura et al. 2011; Fitzpatrick et al. 2012; Hazin et
al. 2013; Werry et al. 2014; Vaudo et al. 2014). For five sharks, location estimates at
Ningaloo Reef with depths <15 m suggested that some sharks spent time in the shallow
lagoon.
Shark 5 experienced a wide range of temperatures within a home range in the tropics,
suggesting that it used the water column to depths of 380 m, well below the thermocline
in this region (Rousseaux et al. 2012). When the shark moved to the south coast off
Albany, it remained in waters above 100 m deep, implying that its vertical movements
may have been constrained by water temperatures when in the cooler waters at more
southern latitudes.
The long-term residency of tiger sharks in limited areas implies that they may have
strong structuring effects on those ecosystems. A study conducted for 15 years
(Heithaus et al. 2012) found that the presence of tiger sharks at Shark Bay had major
influences on the behaviour, movement and feeding of prey such as dolphins, turtles and
dugongs, but beyond this largely seagrass habitat, we have little idea of the role of these
predators in environments such as coral or temperate reef systems. In addition to
quantifying the residency patterns of tiger sharks, our study showed large-scale
movements that have implications for conservation, since this behaviour may take them
across management and national boundaries. In our study, more than half of the
locations provided by satellite transmitters were within the Commonwealth Marine
Reserves network, although our results also support the suggestion that such regional
scale management zones are likely to only provide temporary protection for some parts
of the population or life history stages (Hooker et al. 1999; Chapman et al. 2005;
Heupel et al. 2014; Werry et al. 2014). The movement of sharks from Australia to
Indonesian waters also shows that conservation of these animals will also depend on
international cooperation to mitigate anthropogenic threats to the resilience of the
78
species, such as the large IUU shark fishery in Australia’s northern waters and small-
scale shark fishing industry in Indonesia to the north (Field et al. 2009).
Future research efforts will be aided by technological developments of satellite
transmitters and attachment methods. One such development, the FastlocTM
GPS,
already exists, however is unavailable on fin-mounted devices. Fastloc tags are capable
of acquiring the data required for a location fix in a much shorter period of time and
with greater location accuracy than other types of satellite tags, improving both the
frequency of location estimates and the accuracy of position fixes. Ultimately, there are
two key goals that must be achieved in order to obtain a better understanding of the
movement ecology of tiger sharks: firstly, improved attachment techniques for tags that
allow both frequent uploads of position data and long deployments (> 1 year) and
secondly, tagging of a larger number of animals and a wider range of life history stages
of both sexes. Our study shows that these advances will be necessary to gain a better
appreciation of the role of these animals in the ecology of marine ecosystems.
3.6 Acknowledgments
We acknowledge the Conselho Nacional de Desenvolvimento Científico e Tecnológico,
Brazil.
79
3.7 Supporting information
Figure S 3.1. Sea Surface Temperature map (smoothed OISST data) for 16th May 2009 with
location uplinks for Shark 5.
81
Chapter 4 Diet and trophic role of a large
marine predator, the tiger shark
Galeocerdo cuvier
4.1 Abstract
Stable isotope analysis of tissues reveals diet, trophic niche and food-web interactions
of animals over a range of temporal scales. Multiple tissues (muscle, fin, dermis, blood)
of tiger sharks were sampled at Ningaloo Reef and Shark Bay in Western Australia and
from the Great Barrier Reef (GBR), southern Queensland and the central New South
Wales (NSW) coasts in eastern Australia. Stable isotope analysis was used to
investigate the effects of location, size and sex of sharks on diet, trophic ecology and
role among these locations. Sharks sampled in Shark Bay and on the GBR had a diet
that was based on seagrass food chains, whereas sharks at Ningaloo Reef had diet that
was transitional between pelagic and seagrass chains. Sharks collected in NSW had a
diet based on pelagic food chains with an isotopic signature consistent with earlier
studies of the species at Réunion Island and in South Africa. Tiger sharks were located
at the top of the food-chain at Shark Bay and on the Great Barrier Reef (GBR);
however, these predators did not occupy apex roles at Ningaloo Reef or in New South
Wales. Bayesian standard ellipses were used to assess niche occupancy and generalised
linear models were used to test the effect of shark size, location and sex on δ13
C and
δ15
N values derived from the results of isotope analyses of each tissue. Analysis of
muscle samples suggested that niche width of female sharks in Queensland was broader
than that of males in the same locality and the niche widths of both sexes in NSW. The
diet of tiger sharks became more enriched in 13
C with increasing body size and towards
tropical latitudes consistent with a diet shifted towards large marine herbivores.
Analysis of other tissues (dermis, red blood cells and plasma) also showed a distinct
trophic niche of sharks at Ningaloo Reef, where they exhibited lower levels of δ13
C
values than at Shark Bay or on the GBR. Diets of sharks at Ningaloo were more stable
in the short term (weeks) than at Shark Bay or on the GBR. The large amount of spatial
variation in δC13
values indicated that diet and trophic role seems to be context and
habitat dependent, and reflects variations in food-web complexity, the availability of
resources and body size of tiger sharks.
82
4.2 Introduction
Marine megafauna such as large sharks, cetaceans, billfishes and tunas typically occupy
upper trophic levels in marine food webs. Although their direct predation on - and
induced anti-predator behaviours (risk effects) in - their potential prey can structure
marine ecosystems (Heithaus et al. 2010; Wirsing & Ripple 2011; Heupel et al. 2014),
it is often difficult to quantify their trophic interactions. First, top-order predators are a
relatively rare component of food-webs and documenting their feeding behaviours
within a three-dimensional environment presents some unique logistic obstacles that
still challenge available technology (Hays et al. 2016). Second, the prey targeted by
many top-order species may not be consistent in space and time. Most predators can
switch among potential prey taxa depending on context and availability in order to
optimise energy gain from foraging behaviour (Beck et al. 2007; Thums et al. 2011).
Difficulties involved in the direct observation of feeding behaviour of marine
megafauna have led to the use of alternative techniques to provide insights into the
process of foraging. One of the most common of these is the analysis of stable isotopes
of carbon and nitrogen of consumers’ tissues, which can provide information on diet
(Hobson & Clark 1992; Hussey et al. 2011), trophic niche (Bearhop et al. 2004;
Newsome et al. 2007), trophic interactions among different species (Churchill 2015;
Kiszka et al. 2015), and migratory movements (Hobson 1999; Phillips et al. 2009).
Ratios of nitrogen isotopes (15
N/14
N = δ15
N) indicate the trophic position of a predator
in the food web and the δ15
N composition of its prey, whereas ratios of carbon (13
C/12
C
= δ13
C) indicate the basal source of carbon in the food chain (Post 2002). Together, δ15
N
and δ13
C can provide a general understanding of the structure of a food web (Vander
Zanden & Rasmussen 2001).
In marine ecosystems, values of δ13
C in primary producers are indicative of
environmental (inshore to offshore) and carbon source (benthic to pelagic) gradients
(Hobson et al. 1994; Clementz & Koch 2001). However, for highly mobile and slow-
growing predators such as some large sharks, the interpretation of stable isotopic values
must consider the time taken to incorporate isotopes into different tissues (Hobson &
Clark 1992; Malpica-Cruz et al. 2012). Blood plasma and liver have rapid turnover rates
due to their high catabolic rate and represent a short-term (days to weeks) integration of
a predator’s diet (Meyer et al. 2010). In contrast, muscle, red blood cells, cartilage and
fin have slower turnover rates and provide long-term (months to years) integration of
83
dietary information (Matich et al. 2010; Malpica-Cruz et al. 2012). This means that the
isotopic composition measured from different tissues within the same individual can
resolve dietary information at different time scales (MacNeil et al. 2005; Madigan et al.
2015) and thus assess individual strategies of dietary specialisation (Matich et al. 2011).
Because isotopic signatures integrate diet over time and space (Carlisle et al. 2012),
changes in composition in different tissues of mobile marine predators can “track” their
use of isotopically distinct environments at different time scales (Hobson 1999). This
attribute has been used successfully to define shifts in diets of marine megafauna that
relate to changes in use of habitats (Hatase et al. 2002; Cherel et al. 2007).
Here, stable isotope analyses were used to examine the trophic role of tiger sharks
(Galeocerdo cuvier). These apex predators are highly mobile, large-bodied sharks
distributed globally in tropical, sub-tropical and temperate regions (Randall 1992;
Domingo et al. 2016). Based on their diet, size and lack of predators, adult tiger sharks
(>3 m) are classified as apex predators (Heupel et al. 2014). In at least some contexts,
they play an important role in structuring the environments they inhabit (Heithaus et al.
2002; Burkholder et al. 2013). Tiger sharks have a diverse diet, feeding on prey from
many trophic levels including invertebrates, teleosts, sea snakes, large marine
herbivores such as sea turtles and sirenians, seabirds and other (Stevens & McLoughlin
1991; Simpfendorfer 1992; Lowe et al. 1996; Heithaus 2001; Simpfendorfer et al. 2001;
Trystram et al. 2016). The species also exhibits high individual variability in movement
patterns with some evidence of ontogenetic expansion in habitat utilisation (Werry et al.
2014; Holmes et al. 2014; Afonso & Hazin 2015; Ferreira et al. 2015; Lea et al. 2015).
It is likely, therefore, that their role in structuring marine ecosystems will vary both in
space and with ontogenetic stage. Sampling over broad spatial and temporal scales and
multiple habitats could allow incorporation of stable isotope data into studies of the
spatial ecology (movement patterns, space use and habitat preferences) of these
predators, and provide novel insights into both the motivations behind patterns (prey
preference and foraging locations) and the role of these sharks within the various marine
habitats that their movements encompass.
In this study, we compare the stable isotope composition of multiple tissues of tiger
sharks collected at locations spread across their range in Australian waters. Sampling
encompassed tropical and warm-temperate environments to provide insights into the
trophic ecology and niche of the species across multiple habitats. This study also sought
84
evidence for the influence of biological (size, sex) and spatial (latitude, location) factors
that could influence the stable isotopic composition of tiger sharks and how that
relationship changed at both short (days to weeks) and long (months to years) time
scales by analysing tissues with different turnover rates.
4.3 Methods
4.3.1 Sample collection
Samples were collected under permit numbers: 2563 (WA Department of Fisheries),
SF010311 (WA Department of Parks and Wildlife); G14/36467.1 and G14/37133.1
(Great Barrier Reef Marine Park Authority); 100541, 165491 and 56095 (NSW
Department of Primary Industries and Fisheries); QS2009/GS001, QS2010/MAN26 and
QS2010/GS059 (QLD Department of Environment and Resource Management). All
methods and experimental protocols were performed in accordance with approved
guidelines by Animal Ethics Committees from the University of Western Australia
(RA/3/100/1209), James Cook University (A1974) and University of Queensland
(CMS/300/08/DPI/SEAWORLD and CMS/326/11/DPI).
Tiger sharks were sampled from Ningaloo Reef and Shark Bay in Western Australia,
the Great Barrier Reef (GBR), and the coast of Queensland and deeper continental shelf
waters (> 100 m) of New South Wales (NSW) (Figure 4.1). Two different data sets
were available from Queensland, one composed of muscle samples collected between
2005 and 2014 from sharks captured by the Queensland Shark Control Program (QSCP)
between latitudes of 28°S and 10°S (hereafter the QSCP dataset), and another of
multiple tissues (muscle, dermis, blood) from individual sharks collected at different
locations throughout the Great Barrier Reef (hereafter the GBR dataset). The majority
of the individuals sampled in the QSCP were captured in coastal waters of southern
Queensland, below the latitude of the southern extremity of the GBR. QSCP and GBR
data sets were not pooled due to the different analytical methods applied to each (see
below).
Sample collection was a collaborative effort across multiple research programs in
Australia and as a consequence, the number of samples for each location, sex, types of
tissues collected and sampling methods varied among locations. For the sampling at
Ningaloo Reef and the GBR, tiger sharks were caught using drumlines inside the reef
lagoon and led onto a submerged platform. The platform was raised clear of the water
85
and the shark was restrained by a tail rope. A hose with a continuous flow of
oxygenated seawater was inserted in the mouth of the animal. The shark was then
measured and sexed and fitted with electronic tags prior to release (described below). A
small sample (3 g) of white dorsal muscle and dermis were collected with a scalpel or
an 8-mm diameter tissue punch, about 5 cm lateral to the first dorsal fin. Blood samples
of 5 ml were collected with an 18-gauge needle from the caudal vein. The blood was
separated into plasma and red blood cells with a portable centrifuge (6500 rpm) and all
samples were immediately frozen in either liquid nitrogen (-80°C) or -20°C freezers
after collection. Dermal denticles were removed from dermal samples prior to analysis.
Stable isotope data from sharks captured in Shark Bay were collected by a multi-year
program (Heithaus et al. 2012, 2013) and added to the dataset. For field and sampling
procedures for sharks captured in Shark Bay see Heithaus (2001) and Wirsing et al.
(2006). For sharks caught by the QSCP, muscle samples were collected from the
musculature adjacent to frozen vertebrae. In NSW, shark muscle was collected from the
dorsal region next to the first dorsal fin at recreational game-fishing competitions and
frozen immediately for storage. Sampling procedures for Queensland and NSW are
described by Holmes et al. (2015).
Figure 4.1 Map showing the study locations across Australia and sampling sites within each
region in colour coded triangles. Circles represent relative and total sample size for each dataset.
GBR = Great Barrier Reef, QSCP = Queensland Shark Control Program, NSW = New South
Wales. Map was created with ArcGis 10.3 (http://www.esri.com/).
86
Table 4.1. Mean and standard deviation (SD) stable isotopic composition of tiger sharks sampled in Queensland (QSCP), New South Wales (NSW) the Great
Barrier Reef (GBR) and Ningaloo and Shark Bay for each sex (F = female, M= male). δ15
N and δ13
C given in parts per mill (‰). N= number of samples, TL = mean
total length (cm) ± SD.
* n=1
Location Sex N TL Muscle Dermis Fin Red Blood Cells Plasma
δ15
N δ13
C δ15
N δ13
C δ15
N δ13
C δ15
N δ13
C δ15
N δ13
C
QSCP F 36 207 ± 84.8 12.4 ± 1.0 -16.8 ± 0.7
M 12 161 ± 37.3 12.2 ± 1.1 -17.4 ± 0.5
NSW F 17 322 ± 50.9 12.6 ± 0.8 -18.2 ± 0.6
M 25 345 ± 39.5 12.3 ± 0.8 -18.3 ± 0.8
GBR F 14 296 ± 48.3 11.6 ± 0.9 -14.2 ± 1.6 11.3 ± 0.5 -12.0 ± 1.3 11.0 ± 0.3 -13.5± 1.9 11.5 ± 1.1 -13.4 ± 1.9
M 2 232 ± 235 12.8 ± 0.7 -16.2 ± 0.1 11.9 ± 0.2 -13.8 ± 0.1 11.9 ± 0.5 -15.2 ± 1.1 12.3 ± 1.3 -15.7 ± 0.0
Ningaloo F 20 348 ± 46.3 12.0 ± 0.4 -13.9 ± 1.4 11.6 ± 0.3 -16.6 ± 1.6 11.0 ± 0.5 -16.7 ± 1.2
Shark Bay F 124 280 ± 59.2 11.8 ± 0.7 -16.6 ± 1.3 11.4 ± 0.5 -13.2 ± 1.6 11.6 ± 0.9 -15. ± 1.7
M 23 268 ± 52.6 12.0 ± 0.6 -11.9 ± 1.4 11.0* -13.1* 12.0 ± 1.9 -15.3 ± 1.7
87
Samples from Ningaloo Reef, the GBR, the QSCP and NSW were analysed by the
Western Australian Biogeochemistry Centre at the University of Western Australia.
Samples were freeze dried for 48 h and homogenised using a Mixer Mill MM 200 with
6.4-mm ball bearings. Samples were analysed using a continuous flow system in a Delta
V Plus mass spectrometer coupled to a Thermo Flush 1112 via Conflo IV (Thermo-
Finnigan/Germany). Stable isotope ratios were expressed as δ13
C and δ15
N and reported
in parts per mil (‰) relative to the standard reference Vienna Pee Dee Belemnite
(VPDB) for δ13
C and atmospheric N2 for δ15
N. Samples were standardised against
primary analytical standards from the International Atomic Energy Agency, Vienna
(δ13
C: NBS22, USGS24, NBS19, LSVEC; δ15
N: N1, N2, USGS32 and laboratory
standards). The analytical precision (standard deviation of mean values) of both carbon
and nitrogen isotopes was 0.1‰. Samples had mean C:N ratios of 2.5 ± 0.5 (mean ±
SD) for samples analysed at the Western Australian Biogeochemistry Centre, and
because elasmobranchs are reported to have low lipid content with relatively small
changes to δ13
C values after lipid extraction, we did not correct δ13
C values for the
effects of lipids (Post et al. 2007; Hussey et al. 2010; Vaudo & Heithaus 2011; Speed et
al. 2012; Heithaus et al. 2013). Samples from Shark Bay were analysed at Yale
University and Florida International University (Matich et al. 2011; Heithaus et al.
2013) with analytical precision of 0.10 - 0.19‰ for δ13
C and 0.02 - 0.08‰ for δ15
N
(Heithaus et al. 2013) and showed low C:N ratios (2.45 ± 0.39).
4.3.2 Data analysis
Stable isotopic composition
In order to assess variations in the isotopic signature of tiger sharks among widely
distributed locations, isotopic signatures of muscle samples previously described by
studies in South Africa (Hussey et al. 2015), the Great Barrier Reef (Frisch et al. 2016)
and Reunion Island (Trystram et al. 2016) were plotted against the average values of
δ15
N and δ13
C of slow and intermediate tissues (dermis, muscle, and fin) from this
study. Dermis is likely to have faster turnover rates than muscle (Carlisle et al. 2012; Li
et al. 2016a), however, in the absence of muscle samples at Ningaloo, dermis was
considered as the longer-term representation of diet. Dermis is a poorly vascularised
structural tissue that may reflect longer-term (months) signals in diet (Zamora et al.
2016) when compared to tissues that are metabolically active such as blood (Kim et al.
2012). To assess the trophic position of tiger sharks in each ecosystem, the stable
88
isotope composition of each tissue (muscle, dermis, red blood cells and plasma) for all
sharks at each location was also plotted against that of taxa from different trophic
positions from each location (Ningaloo Reef, Shark Bay, GBR and NSW), with values
extracted from published literature. Additional isotope data from muscle tissue of four
tiger sharks sampled on the GBR (Lizard Island and southern GBR) between 2013 and
2014, and analysed at James Cook University, were combined with isotopic values from
the literature for the comparison with the values obtained in the present study. As earlier
studies were analysed in different laboratories with varying sample handling and
processing protocols (Bessey & Vanderklift 2014), these comparisons were made
visually and no statistical analysis was attempted.
Isotopic niche
Isotopic niche space occupied by tiger sharks in each location in Australia was
measured by calculating standard ellipse areas corrected for sample size (SEAc), the
SEAc parameters of eccentricity (E) and theta (θ), the Bayesian estimation of SEA
(SEAb) and the overlap between paired groups as a probability index with values
between 1 and 0 (Jackson et al. 2011). The Bayesian approach used vague and non-
informative priors and was built with 2 Markov chains with 20,000 steps per chain, a
discarded burn-in of 1,000 iterations and a thinning interval of 10. This produced a
range of SEAb with 95% credible intervals that allowed for direct probabilistic
interpretation of pairwise comparisons between groups that could be interpreted as the
probability of the Bayesian ellipse area of one group being larger than another (Jackson
et al. 2011; Reid et al. 2016). The total area (TA) occupied was calculated as the areas
of the convex hull that incorporated all individuals and represents a measure of niche
width and reflects the isotopic diversity of a group (Layman et al. 2007; Vaudo &
Heithaus 2011). The mean distance to centroid (CD) was calculated as the distance in
isotopic space of each individual to the mean of all individuals and represented a
measure of the average trophic diversity within a group (Vaudo & Heithaus 2011). As
eccentricity (E) was related to the variance on the x (δ13
C) and y (δ15
N) axes and to the
circularity of the SEAc, high values of E were representative of ellipses that were
stretched either in the x or y-axes. Theta was a measure of the inclination of the ellipse
and is reported here in radians but was converted to the angular range between -90 to
90, with positive and negative values indicating the direction of inclination of standard
ellipses (Jackson et al. 2011; Reid et al. 2016). Values for each tissue were analysed
separately using the R packages siar (Parnell & Jackson 2013) and SIBER (Jackson et
89
al. 2016). Sexes were considered as separate groups within a region for datasets that
contained a representative number of samples (n > 10) for both male and female sharks.
Spatial patterns of isotopic composition of multiple tissues
Generalised linear models (GLMs) with a Gaussian distribution were used to assess if
δ13
C and δ15
N varied spatially (by location or latitude), and in relation to sex and size of
sharks and at different temporal scales by using tissues with different turnover rates.
Data from each tissue (muscle, dermis, red blood cells and plasma) was modelled
separately using both δ13
C and δ15
N as response variables and a set of explanatory
variables that were dependent on the data available for each tissue. The set of predictors
varied among tissues and was selected based on datasets available (e.g. sex was only
used as predictor if the number of males was sufficient, whereas season was only
included in the analysis of datasets of muscle). For muscle, the stable isotope
composition was analysed with regard to the latitude, season (austral summer, autumn,
winter and spring), sex, total length (TL, in cm) and interaction terms between sex and
latitude, and sex and TL. For dermis, stable isotopes were assessed against the
categorical variable (location) rather than the continuous variable (latitude), due to
sampling locations on the west and east coast of Australia having similar latitudes. Total
length and an interaction term between location and TL were also included. For red
blood cells and plasma, variables analysed included location, TL, and the interaction
terms between TL and location. Fin samples were only available from Shark Bay and,
therefore, these were not included in this analysis. All potential combinations of these
predictor variables were fit with the package MUMIn (Barton 2013). Model selection
was done by ranking the models by Akaike’s Information Criterion corrected for sample
size (AICc) and the AICc weight (wAICc) which varies from 0 (no support) to 1
(complete support). When multiple models were ranked equally (models within 2 AICc
units), the most parsimonious model was selected (the model containing the lowest
number of explanatory variables). Goodness of fit was assessed by the percentage
deviance explained (%DE). Conditional plots for each variable in the top-ranked model
were plotted using the R package ggplot2 and visreg (Breheny & Burchett 2016).
Diet Stability
Tiger sharks at Ningaloo Reef were also tagged internally with a V16TP or V16
acoustic transmitters (Vemco Ltd., Canada) (Table S 4.1). Upon capture, sharks were
restrained and rolled to expose their ventral side. A 4 cm incision was made along the
90
centre line of the abdomen, an acoustic tag was inserted and the incision was closed
with three stiches in the muscle and skin using surgical sutures. Sharks tagged with an
acoustic transmitter were monitored by receiver arrays around Ningaloo Reef (Figure S
4.1) deployed by the Australian Animal Tracking and Monitoring System. This tracking
allowed sharks that were resident at Ningaloo Reef to be identified so that their isotopic
composition could be used as a baseline value for analyses. An earlier study at Ningaloo
Reef demonstrated that tiger sharks either resided in the areas where they were tagged
for up to 6 months or departed and ranged 1000s of km along the coast of Western
Australia (Ferreira et al. 2015). Consequently, we assumed that individuals that
remained resident after tagging and tissue sampling were likely to have been resident at
Ningaloo Reef prior to capture. As receiver downloads were sporadic during the
duration of the study, sharks were defined as residents if they were detected by the
receiver arrays for at least five successive months after tagging and prior to the last
download of receivers (Table S 4.1), again consistent with residency times recorded by
earlier studies (Ferreira et al. 2015). As blood has an isotopic turnover rate of months
(Logan & Lutcavage 2010), the signature of this tissue was used as a baseline for
resident sharks.
Stability of diet was assessed based on differences in δ13
C between paired tissues (red
blood cells and plasma) for the locations where multiple tissues were collected from the
same sharks (Ningaloo Reef, Shark Bay and GBR). Plasma has a faster isotopic
turnover rate than red blood cells (a half-life of 32 vs. 60 days, Kim et al. 2012) than red
blood cells (~60 days, Kim et al. 2012). The contrasting turnover rates of these tissues
allowed for comparison of isotopic values within an individual, giving an insight into
the temporal stability of diet (Matich et al. 2011; Malpica-Cruz et al. 2013). Preliminary
analysis of this dataset and previous studies of tiger sharks in Shark Bay showed greater
inter-tissue differences in δ13
C than in δ15
N (Matich et al. 2010, 2011), therefore, this
analysis only considered δ13
C values. Linear regression was fitted to the paired values
of δ13
C for sharks from Ningaloo Reef, Shark Bay and the GBR in order to describe the
consistency of differences in δ13
C between tissues and to quantify the inter-tissue
relationship, as a measure of tissue discrimination differences (Matich et al. 2010;
Matich & Heithaus 2014; Davis et al. 2015). As paired tissue discrimination shows
stable relationships in other species including seabirds, mako (Isurus oxyrinchus) and
white sharks (Carcharodon carcharias) (Quillfeldt et al. 2008; Malpica-Cruz et al.
2013), it can be used to compare stability of diet in samples from various locations by
91
identifying individuals that are not in a steady state. AIC values were compared for the
regression of δ13
C between red blood cells and plasma against a null model. If there was
strong support for the regression model and the R2 value was high (>0.8), the
relationship was considered significant and stable. Stable relationships between Δδ13
C
values of paired tissues suggest a stable discrimination of differences among locations
and validate use of the fitted line of the regression and δ13
C differences to investigate
dietary stability in individual sharks. Since only a relationship with strong support was
used among different locations, mean values of the paired tissue differences of resident
sharks at Ningaloo Reef were used as a proxy for the δ13
C for discrimination between
red blood cells and plasma (Δδ13
C). Paired tissue differences for each individual
(Δδ13
C) were then plotted over the fitted line and interval (± mean Δδ13
C) around it.
Data points that plotted more than Δδ13
C above the fitted line indicated plasma values
with higher δ13
C values than predicted by the model and data points that plotted less
than Δδ13
C below the fitted line indicated plasma values with lower δ13
C than predicted
by the model (Matich & Heithaus 2014; Davis et al. 2015). Values of plasma outside the
± mean Δδ13
C interval area around the fitted line, therefore, indicated a recent shift in
diet. Values within the ± mean Δδ13
C interval around the fitted line were considered as
“stable” and were assigned a value of 1. Values that were either higher or lower than the
± mean Δδ13
C interval were considered “non-stable” and assigned a value of 0.
Generalised linear models with a binomial distribution were used to test the probability
of sharks having a stable diet in relation to total length and location. All potential
combinations of the predictor variables (location, total length and an interaction term
with total length and location) were used and model selection was as described
previously.
4.4 Results
4.4.1 Comparisons of stable isotopic signatures among
locations
A comparison of slow-turnover tissues (dermis, fin and muscle) shows a spread of mean
isotopic signatures that indicated variation in the diet of tiger sharks across both
latitudes and shelf positions (inshore/offshore habitats) (Figure 4.2a). ). Most of the
variation occurred along the δ13
C axis. Individuals from Shark Bay had higher δ13
C,
whereas sharks from the QSCP, NSW and Reunion Island had lower δ13
C. Sharks
sampled on the GBR had intermediate δ13
C values and were very similar to those of
92
muscle tissue collected and analysed by other studies in the same locality (Frisch et al.
2016; Figure 4.2a). However, δ13
C values of dermis from GBR sharks were high and
overlapped with those of Shark Bay individuals. There was limited variation in mean
δ15
N values of sharks among localities, with the exception of tiger sharks in South
Africa, which had higher δ15
N values (Figure 4.2a).
Figure 4.2. Mean and standard deviation stable isotopic composition of each tissue for each of
the locations sampled (SB= Shark Bay, GBR = Great Barrier Reef, NSW =New South Wales,
QSCP = Queensland Shark Control Program, SA = South Africa) for A) slow tissues (muscle,
dermis, fin) and B) multiple tissues (muscle, dermis, fin, red blood cells, plasma), for sites
where multiple tissues were collected in this study. Data from GBR1, Reunion Island and South
Africa obtained from published studies (Hussey et al. 2015b; Frisch et al. 2016; Trystram et al.
2016)
The collection of multiple tissues from the same animals in some localities allowed the
isotopic composition of both fast (blood, plasma) and slow (muscle, dermis and fin)
turnover tissues to be compared (Figure 4.2b). Once again, there was greater variation in
δ13
C than δ15
N values. Fast tissues tended to have lower δ13
C compared to slow
turnover tissues collected at the same locality. At Ningaloo, red blood cells and plasma
had lower δ13C compared to dermis (Figure 4.2b), whereas at Shark Bay, plasma had a
lower δ13
C compared to red blood cells and blood cells had lower δ13
C compared to fin
(Figure 4.2b). Analyses of the GBR samples revealed a somewhat different pattern, with
blood and plasma having lower δ13
C compared to dermis, but muscle samples having
similar δ13
C values to these fast turnover tissues (Figure 4.2b).
93
4.4.2 Comparisons of isotopic signatures within
locations
Stable isotopic composition of tiger sharks was plotted with those of other fishes, sharks
and taxa from differing trophic levels at each location (Figure 4.3). Only values for slow
turnover tissues of sharks were included in these plots since signatures for other
vertebrates were largely derived from the analysis of muscle (Figure 4.3Only values for
slow turnover tissues muscle and dermis) of sharks were included in these plots since
stable isotopic compositions for other vertebrates were largely derived from the analysis
of muscle. At Ningaloo, tiger sharks were identified as upper-order predators, but did
not display the highest average δ15
N values in the shark community. Two species of reef
sharks (grey reef, Carcharhinus amblyrhynchos and blacktip reef, Carcharhinus
melanopterus) had higher δ15
N (Speed et al. 2012) in comparison to tiger sharks,
implying a higher trophic position and had higher δ13
C (Figure 4.3a). In Shark Bay,
tiger shark muscle had higher δ15
N than all other sharks at this location (Vaudo &
Heithaus 2011; Heithaus et al. 2013) (Figure 4.3b). On the GBR, tiger sharks had higher
δ15
N compared to other sharks at this location, with the exception of C. obscurus, the
dusky shark. This pattern was consistent across tissue types (dermis and muscle) of
sharks sampled on the GBR (Figure 4.3c). Muscle tissue from sharks collected by the
QSCP displayed similar δ15
N values as dermis and muscle tissue from animals collected
on the GBR, but had lower δ13
C, suggesting a diet based on a more pelagic food chain
for sharks sampled in the QSCP than on the GBR (Figure 4.3c). The isotopic
composition of muscle from tiger sharks collected on the NSW coast was also
consistent with a diet based on pelagic food chains or different basal sources of carbon.
In contrast to both the GBR and Shark Bay, the stable isotopic composition of
Australian fur-seals (Arctocephalus pusillus doriferus) (Davenport & Bax 2002) and a
number of large pelagic fishes including tunas and mako sharks (Isurus oxyrinchus)
(Revill et al. 2009) had higher δ15
N than those of tiger sharks, suggesting that these
species feed at higher trophic levels. The isotopic composition of tiger shark tissues in
this location was similar to those of billfishes (Figure 4.3d).
94
Table 4.2. Isotopic niche area described by Standard Ellipse Area SEAc (‰), Bayesian SEAb
and parameters; N= number of samples, TA = total area of the convex hull, CD = mean centroid
distance and standard deviation, E = eccentricity, θ = angle of ellipse and CI = confidence
interval. F = Females, M = Males, QLD = Queensland Shark Control Program and GBR, GBR
= Great Barrier Reef and NSW = New South Wales.
Location N SEAc SEAb TA CD ± SD E θ 95% CI
Muscle
F NSW 18 1.16 1.36 4.52 0.55 ± 0.23 0.74 60.6 0.85 - 2.26
M NSW 25 2.06 1.83 7.26 0.74 ± 0.27 0.68 51.5 1.17 - 2.71
F QLD 45 3.79 4.23 15.32 1.34 ± 0.17 0.88 -28.9 3.14 - 6.68
M QLD 14 2.38 1.01 5.25 0.84 ± 0.25 0.77 -72.8 0.57 - 1.82
Dermis
Ningaloo 20 1.61 1.7 5.04 0.22 ± 0.09 0.97 -6.4 1.00 - 2.51
GBR 1.42 1.54 23.33 1.21 ± 0.73 0.98 -18.9 0.89 - 2.70
Red Blood Cells
Ningaloo 17 1.78 1.57 5.05 0.80 ± 0.69 0.98 2.7 0.99 - 2.69
Shark Bay 25 2.48 2.33 7.01 1.02 ± 0.93 0.95 -9.2 1.55 - 3. 47
GBR 14 2.58 2.33 5.61 0.89 ± 0.46 0.98 -7.4 1.30 - 4.08
Plasma
Ningaloo 19 1.85 1.66 5.04 1.39 ± 0.69 0.93 -6.6 1.06 - 2.70
Shark Bay 45 5.65 5.48 23.33 0.93 ± 0.43 0.79 -3.6 3.97 - 7.38
GBR 14 6.72 5.93 15.07 1.09 ± 0.24 0.86 -12.9 3.30 - 10.40
95
Figure 4.3. Mean isotope values (± SD) of slow turnover tissues (muscle, dermis, fin) of tiger sharks, other predators, consumers and primary producers at Ningaloo
Reef, Shark Bay, Great Barrier Reef and New South Wales. Community data for each location was obtained from Davenport & Bax (2002), Arthur et al. (2008),
Revill et al. (2009) Vaudo & Heithaus (2011), Speed et al. (2012), Wyatt et al. (2012), Thomson et al. (2012), Heithaus et al. (2013), Frisch et al. (2014, 2016),
Marcus, unpublished data. Red = sharks, orange = rays, blue/golden = teleosts, purple = turtles, magenta = penguins, grey = seals, maroon = cephalopods, light
green = producers, dark green = zooplankton.
96
4.4.3 Comparison of tissues among locations
Muscle
Differences in isotopic niches were assessed separately for each sex. For this analysis,
muscle samples from the QSCP and the GBR were pooled (QLD; Figure 4.4 Female
tiger sharks in QLD had a larger niche and higher trophic diversity than all other
combinations of sex and location, based on the distribution of SEAb and TA (Table 4.2;
Figure 4.4a-b). The SEAc parameter for females from QLD showed that the ellipse for
this group was stretched with dispersion both on the δ13
C axis and the δ15
N axis
(E=0.88, θ= -28.87°; Figure 3a, Table 1). In contrast, male sharks in QLD waters had a
greater dispersion over the δ15
N axis (E=0.77, θ= -72.85°) than the δ13
C axis as did both
males and females from NSW (E=0.68 and 0.74, θ= 51.5 and 60.6, respectively). Sharks
in NSW waters had a large overlap between sexes (0.68) and little difference in niche
area (probability = 0.29). Females and males in QLD displayed high niche overlap
(0.60), but little overlap with tiger sharks from NSW waters. Females in NSW showed
the lowest trophic diversity and total area, despite having a larger SEAb than males in
QLD (Table 4.2).
The generalised linear model that included latitude, sex, total length and the interaction
between latitude and sex had the highest support (wAICc = 0.76) and explained 65.3%
of the deviance in the response (Table 4.3). This model indicated a shift to higher δ13
C
from smaller to larger sharks and towards lower latitudes for both sexes, trends that
were more pronounced for females than males (Figure 4.4c-d In the analysis of the δ15
N
data set, the four top-ranked models were all within two AICc points of the most
parsimonious model (i.e. with the least number of parameters) including latitude only.
The model indicated lower δ15
N values towards lower latitudes, but explained only
7.6% of the deviance in the response.
97
Figure 4.4. Muscle. Standard ellipse areas for analysis of muscle tissue corrected for sample
size (SEAc) (A) and Bayesian Standard ellipse areas (SEAb) (B), representing the 50, 75 and
95% credible intervals in decreasing order of box size, with SEAb mode indicated by a black
circle and SEAc by a triangle. Partial dependence plots of the relationship between δ13
C and the
explanatory variables in the top-ranked model, total length (C) latitude and sex and their
interaction (D). Points represent residuals and shaded areas show the 95% confidence intervals.
Dermis
Dermis was only collected from sharks captured at Ningaloo Reef and the GBR
and data from both sexes were pooled for analysis due to low numbers of males in
samples (n=2). Samples from sharks on the GBR had greater dispersion over the δ15
N
axis (θ Ningaloo Reef = -6.4, θ GBR=-18.90) than those at Ningaloo (Figure 4.5a-b).
Although the size of the SEAb area was similar in both locations, niches were mostly
distinct with sharks in GBR showing a larger TA and trophic diversity (Figure 4.5a,
Table 4.2) with only 0.2 probability of overlap. Of the three models selected to describe
variation in δ13
C values, the most parsimonious included only total length (wAICc =
0.28) and explained 28% of the deviance (Table 4.3). Model predictions indicated a
shift to lower δ13
C with increasing total length (TL) of sharks (Figure 4.5c). For δ15
N
data sets, the three top-ranked models were within two AICc points, with the most
parsimonious model including location only and explaining 32% of the deviance.
98
Dermis tissue of sharks at Ningaloo Reef had higher δ15
N than those of sharks from the
GBR (Figure 4.5d).
Figure 4.5. Dermis. Standard ellipse areas for analysis of dermal tissue corrected for sample
size (SEAc) (A) and Bayesian Standard ellipse areas (SEAb) (B), representing the 50, 75 and
95% credible intervals in decreasing order of box size, with SEAb mode indicated by a black
circle and SEAc by a triangle. Partial dependence plots of the relationship between δ13
C (C) and
δ15
N (D), and the explanatory variables in the top-ranked models; total length (C) and location
(D). Points represent residuals and shaded areas show the 95% confidence intervals.
99
Table 4.3. Ranked generalised linear models of δ13
C and δ15
N for each tissue. The top six
models for each response are shown; values in bold indicate the top ranked model (or the most
parsimonious model). Degrees of freedom (df), Sample corrected Akaike’s Information
Criterion (AICc), change in AICc relative to the model with the lowest AICc value (ΔAICc),
relative AICc weight (wAICc) and percent deviance explained (DE%). Lat = latitude, TL = total
length (cm).
Response Model df AICc Δ AICc wAICc %DE
Muscle
δ13
C lat + sex+ TL+TL × lat 6 237.9 0.00 0.76 65.29
δ13
C lat + sex + TL 5 241.4 3.57 0.13 63.01
δ13
C lat + sex + lat × sex 5 243.5 5.61 0.05 62.18
δ13
C lat + sex + TL + sex × TL 6 243.7 5.81 0.04 63.03
δ13
C lat + season + sex + TL 8 246.6 8.69 0.01 63.77
δ13
C lat + TL 4 247 9.13 0.00 59.74
δ15
N lat 3 250.5 0.00 0.34 7.65
δ15
N lat + sex + lat × sex 5 251.9 1.38 0.17 10.66
δ15
N lat + sex 4 252.6 2.10 0.12 7.74
δ15
N lat + TL 4 252.7 2.15 0.12 7.69
δ15
N lat + sex + TL 6 254.2 3.62 0.05 10.71
δ15
N lat + sex + TL + lat × sex 5 254.8 4.31 0.04 7.77
Dermis
δ13
C location + TL 4 121.8 0.00 0.42 34.85
δ13
C TL 3 122.7 0.86 0.28 27.93
δ13
C location 3 123.8 1.96 0.16 25.54
δ13
C location + TL + TL × location 5 124.0 2.22 0.14 35.89
δ15
N location 3 49.5 0.00 0.40 31.82
δ15
N location + TL + TL × location 5 49.8 0.23 0.36 41.34
δ15
N location + TL 4 50.8 1.23 0.21 34.42
δ15
N TL 3 55.3 5.76 0.02 19.22
Red blood cells
δ13
C location + TL 5 212.9 0.00 0.86 55.62
δ13
C location+ TL + TL × location 7 216.6 3.73 0.13 56.72
δ13
C location 4 222.3 9.37 0.01 45.23
δ13
C TL 3 230.7 17.84 0.00 33.58
δ15
N location 4 78.6 0.00 0.72 15.93
δ15
N Location + TL 5 81.0 2.41 0.22 15.93
δ15
N TL 3 84.5 5.84 0.04 2.75
δ15
N location + TL + TL × location 7 85.9 7.25 0.02 16.37
Plasma
δ13
C location + TL 5 297.2 0.00 0.83 33.38
δ13
C location+ TL + TL × location 7 300.8 3.69 0.13 34.29
δ13
C location 4 303.3 6.16 0.04 25.76
δ13
C TL 3 307.2 10.06 0.00 19.70
δ15
N location 4 218.2 0.00 0.66 8.66
δ15
N location + TL 5 220.5 2.26 0.21 8.69
δ15
N TL 3 222.2 3.95 0.09 1.13
δ15
N location +TL + TL × location 7 224.3 6.04 0.03 9.83
100
Red Blood Cells
Red blood cells were obtained from female sharks at Ningaloo and from both
males and females at Shark Bay and on the GBR. Samples from different sexes were
pooled within locations due to the very low numbers of males (3 in total). Plots of
isotopic niches of sharks from all locations had ellipses stretched across the x-axis with
high variation in δ13
C values (Figure 4.6a, Table 4.2). Sharks at Ningaloo Reef had a
more distinct, smaller and less diverse niche than in Shark Bay and on the GBR based
on SEAb, TA and CD (probability=0.86 and 0.85, respectively), who were more similar
to each other (Figure 4.6a-b). In the analysis of the δ13
C data set there was one top-
ranked model, which included total length and location and explained 56% of the
deviance (Table 4.3 Red blood cells had lower δ13
C with increasing size and higher δ13
C
at Shark Bay and on the GBR than at Ningaloo (Figure 4.6c-d). The top-ranked model
for the δ15
N dataset only included location and explained 16% of the deviance,
indicating marginally higher δ15
N at Ningaloo (δ15
N = 11.6 ±0.1‰) in comparison to
individuals from Shark Bay (δ15
N = 11.4 ±0.1‰) and the GBR (δ15
N = 11.11 ±0.12‰).
Figure 4.6. Red Blood Cells. Standard ellipse areas for analysis of red blood cells corrected for
sample size (SEAc) (A) and Bayesian Standard ellipse areas (SEAb) (B), representing the 50,
75 and 95% credible intervals in decreasing order of box size, with SEAb mode indicated by a
black circle and SEAc by a triangle. Partial dependence plots of the relationship between δ13
C
and the explanatory variables in the top-ranked model; total length (C) and location (D). Points
represent residuals and shaded areas show the 95% confidence intervals.
101
Plasma
Samples used for determination of niche area were collected from female sharks at
Ningaloo Reef and from both sexes at Shark Bay and the GBR. Again, data were pooled
for analysis between sexes within locations due to limited sample sizes. Sharks from the
GBR and Shark Bay had significantly greater SEAb and TA areas than those at
Ningaloo Reef (probability=1 for both). However, sharks at Ningaloo Reef showed
higher trophic diversity than in Shark Bay and on the GBR based on CD (Table 4.2). At
all locations the dispersion of stable isotopic composition occurred on both axes (θ= -
6.6 Ningaloo, θ= -3.6 Shark Bay, θ= -13.9 GBR, Table 4.2, Figure 4.7a-b) and no niche
overlap was evident in samples from Ningaloo Reef and the GBR, but there was an
overlap of niche area with individuals from Shark Bay (0.30). The model including
location and total length had majority support (wAICc = 0.83) and explained 33% of the
deviance in the response (Table 4.3). The model predictions showed lower δ13
C with
increasing body length, with the lowest δ13
C value recorded for sharks at Ningaloo Reef
(Figure 4.7c-d). For δ15
N values, the top-ranked model only included location and had
the majority of support (wAICc = 0.66), but explained only 9% of the deviance in the
data set.
102
Figure 4.7. Plasma. Standard ellipse areas for analysis of plasma corrected for sample size
(SEAc) (A) and Bayesian Standard ellipse areas (SEAb)(B), representing the 50, 75 and 95%
credible intervals in decreasing order of box size, with SEAb mode indicated by a black circle
and SEAc by a triangle. Partial dependence plots of the relationship between δ13
C and the
explanatory variables in the top-ranked model; total length and location. Points represent and
shaded areas show the 95% confidence intervals respectively.
4.4.4 Diet stability
Six of the 18 tiger sharks tagged at Ningaloo Reef were considered to be
residents as they were detected by receiver arrays at Ningaloo Reef (Table S 4.1for five
or more months after tagging and tissue sampling. A mean Δδ13
C value of -1.26 ‰ for
the difference between red blood cell and plasma of resident sharks was considered as a
threshold for the comparison of paired tissues that defined sharks as having stable diets.
Due to a lack of tagging data from Shark Bay and the GBR in this study, the threshold
calculated for Ningaloo was used in analyses for three locations.
Linear regressions between δ13
C values from red blood cells and plasma from the same
individuals had strong support when compared to a null model (wAICc =1, 0.82 %DE)
and the fitted line was then used for the identification of stability of diet (Figure 4.8a).
The most parsimonious binomial generalised linear model (wAICc =0.33, Table 4.4) to
explain the probability of diet stability over one or two months only included location
and explained 16% of the deviance. The predictions from this model showed that sharks
103
at Ningaloo had a higher probability of having a stable diet than those sampled on the
GBR and at Shark Bay (Figure 4.8b).
Table 4.4. Ranked binomial generalised linear models of stability of diet for sharks (stable with
the local food web), using paired tissue values of δ13
C from red blood cells (RBC)-plasma
comparisons. All models are shown, values in bold indicate the most parsimonious model.
Sample corrected Akaike’s Information Criterion (AICc), change in AICc relative to the model
with the lowest AICc value (ΔAICc), relative AICc weight (wAICc) and percent deviance
explained (DE%). TL= total length (cm).
Response Model df AICc Δ AICc wAICc %DE
Stability location 3 74.8 0.00 0.61 9.87
Stability Location + TL 4 76.2 1.40 0.30 11.10
Stability TL 2 79.9 5.11 0.05 0.17
Stability Location + TL + TL × location 6 80.1 5.28 0.04 12.51
Figure 4.8. Relationship between δ13
C values of paired tissues of fast (plasma) and intermediate
(red blood cells) turnover rates (A). Individuals with a difference between tissues (calculated
from resident sharks) larger than mean differences Δ13
C are highlighted in filled symbols. Open
symbols represent individuals in steady state (stable). Solid line represents the relationship
between paired tissues and hashed lines enclose the area within ± mean Δ13
C. Partial
dependence plots for the relationship between the stability of diet (probability of steady state)
and the explanatory variable in the top-ranked model; location (B). Numbers indicate sample
sizes.
4.5 Discussion
Our analysis of stable isotopes suggested that the functional role of tiger sharks varied
among different marine habitats within the range of the species distribution along the
tropical and temperate coasts of Australia. Typically, there was less variation in δ15
N
than δ13
C values in slow turnover tissues (muscle, dermis and fin), indicating that the
species remains close to the top of food webs at the body sizes we sampled, but that the
104
ultimate sources of nutrition varied among localities and were dependent on body size,
habitat and food-web composition.
Sharks sampled in the enclosed seagrass habitats of Shark Bay on the west coast and the
large, shallow lagoon of the GBR showed the strongest evidence of long-term diets
based in coastal-associated food-webs originating with seagrass. Dugong and turtles are
likely prey in these localities (Heithaus 2001; Heithaus et al. 2013). At higher latitudes,
such as the coast of NSW, mean stable isotopic composition indicated a diet more
closely linked to pelagic food chains. However, isotopic composition also appeared to
be affected by an interaction between latitude and oceanographic setting. Most of the
samples collected by QSCP were obtained from beaches south of the GBR, which are
exposed to prevailing oceanic swells, upwelling and open waters across the shelf (Oke
& Middleton 2000). Stable isotopic compositions in this locality were similar to NSW,
where equivalent oceanographic conditions prevail (Tranter et al. 1986; Suthers et al.
2011), , but were also similar to samples from Reunion Island, an exposed and isolated
island in the Indian Ocean that does not feature an extensive shelf or abundant seagrass
(Trystram et al. 2016). In NSW, sharks were collected in deeper, pelagic waters at the
edge of continental shelf during game-fishing competitions. These results suggest that
lower δ13
C was not just a function of latitude but also the oceanographic setting and
proximity of pelagic food chains. When plotted in the isotopic space, values of δ13
C of
sharks sampled at Ningaloo Reef were between pelagic and seagrass food chains, as
might be expected given that this location occurs at the narrowest point of the
continental shelf along the Australian coast and so is strongly influenced by the
Leeuwin Current, seasonal upwelling and the oceanographic conditions of the open
ocean (Taylor & Pearce 1999; Harris et al. 2005; Rousseaux et al. 2012). Although
seagrass beds exist at Ningaloo Reef in the shallow lagoon that extends one to two
kilometres behind the reef crest to the shore (Cassata & Collins 2008), these are
relatively small in area compared to those in the inter-reefal lagoon of the GBR or at
Shark Bay, where they are a dominant type of benthic habitat (Walker et al. 1988;
Wachenfeld et al. 1998; Carruthers et al. 2002).
The average δ15
N for tiger sharks in the present study differed greatly from those
calculated by Hussey et al. (2015b), who obtained samples from sharks caught in bather
protection programs along beaches off the Natal coast of South Africa. Hussey et al.
(2015b) reported δ13
C values for tiger sharks that were consistent with samples from the
exposed sub-tropical coastlines of the present study (the QSCP and NSW datasets), but
105
δ15
N values were higher relative to all sampling locations in Australia. It is possible that
the difference seen in South Africa may have been a reflection of differences in isotopic
baselines and composition of the food-web, nutrient inputs or and/or sample treatment
between these locations (Spies et al. 1989; Graham et al. 2010; Li et al. 2016b)
Tiger sharks are a wide-ranging species and tracking studies show that these animals
move between tropical and temperate environments and between coastal habitats and
the open ocean off the continental shelf. Our results suggest that the role of tiger sharks
within food webs and their classification as an apex predator varies as they move across
these environmental gradients. Tiger sharks occupied the apex position of the food-
webs in seagrass habitats and shallow lagoons of Shark Bay and the GBR, where nearly
all other shark species for which isotopic data were available had lower δ15
N values. In
contrast, at Ningaloo Reef, other, smaller sharks such as grey reef (Carcharhinus
amblyrhynchos; adult size 1.9 m total length, Wetherbee et al. 1997) and blacktip reef
(C. melanopterus; adult size 1.8 m total length, Compagno 1984; Last & Stevens 1994),
had higher δ15
N (Speed et al. 2012) than tiger sharks. These differences may reflect the
feeding by reef sharks on mesopredator teleosts whose productivity is driven by
complex coral reef food chains, whereas the diet of tiger sharks can also include species
such as herbivorous teleosts, turtles and sirenians, which feed at a more basal level of
the food web. As a result of the large home range areas of tiger sharks off the northwest
coast of Australia (Ferreira et al. 2015), it is also possible that sharks were feeding in
different habitats or offshore areas (i.e. pelagic food-webs) with a lower stable isotopic
baseline (Sherwood & Rose 2005) which could explain the differences between the
stable isotopic composition of tiger sharks and the more resident reef shark community
at Ningaloo Reef (Speed et al. 2016).
In NSW, large oceanic fishes such as tunas and other sharks including makos (Isurus
oxyrinchus and I. paucus) (Revill et al. 2009) had higher δ15
N values than tiger sharks,
suggesting that they fed at higher trophic levels, which is supported by the high
proportion of other predatory teleosts reported in the stomach contents of these species.
This was also the case for Australian fur-seals (Arctocephalus pusillus doriferus), which
feeds on the continental shelf (Kirkwood et al. 2006). A lower trophic level for tiger
sharks has also been reported in the coastal waters of warm temperate environments in
southern Africa, where tiger sharks occupy the lowest trophic position within the
community of large sharks (Hussey et al. 2015b). Our results suggest that when tiger
sharks move into pelagic and offshore habitats where other large teleost and sharks are
106
present, these predators will occupy higher trophic levels than tiger sharks by having
more specialised diets largely focused on fishes that are tertiary consumers or higher.
Because no community isotopic data was available for the southern Queensland coast, it
was not possible to determine the extent to which tiger sharks occupied the highest
trophic level in these environments. However, the similarity of oceanographic
conditions in this locality to the coast of NSW and the migrations of tiger sharks along
much of the eastern coast of Australia suggests (Holmes et al. 2014) that they are likely
to have similar trophic roles in both locations.
The position of tiger sharks in a food chain may also reflect sampling bias of other
predators within these systems. For example, Vaudo & Heithaus 2011 and Heithaus et
al. 2013 comprehensively sampled the fauna of elasmobranchs and dolphins in shallow
banks and sandflats habitats of Shark Bay and these shallow habitats might not be
suitable for many large pelagic fishes such as tunas. The classification of tiger sharks as
the highest trophic level predators in these habitats thus seems appropriate because all
other species that might act in this role had been sampled. At Ningaloo Reef, sampling
of the shark and predatory fish community was limited to reef environments, so that the
larger pelagic species that might occupy higher trophic positions such as tunas could not
be included in plots. With the exception of Shark Bay, where samples were collected
over 6 years, there was also a lack of temporal context for sampling in Ningaloo and
GBR localities. It may be possible that the diet and role of tiger sharks could vary across
both seasons and years, as is the case with other marine predators (MacNeil et al. 2005;
Cherel et al. 2007; Carlisle et al. 2012), although such variation may be limited, given
the fact that the composition of samples in terms of δ15
N values was very similar across
all locations and that the mean stable isotopic composition from muscle tissues of tiger
sharks analysed by earlier studies on the GBR (Frisch et al. 2016) were nearly identical
to those presented here (Figure 4.3c).
There was some evidence that size affected the diets of tiger sharks, although patterns
were inconsistent among tissues. Muscle, a slow turnover tissue, provided the largest
data set available for analysis and showed higher δ13
C with increasing size (total length)
suggesting a diet increasingly shifted towards prey from seagrass habitats. This is likely
due to larger sharks having a greater capability to attack and subdue large marine
herbivores such as turtles and dugong than smaller, younger sharks. It is also consistent
with the results of studies of the stomach contents of tiger sharks that show a change in
diet towards larger prey with increasing size (Simpfendorfer 1992; Lowe et al. 1996;
107
Heithaus 2001). There was also a shift to higher δ13
C at lower latitudes where these
large herbivores such as turtles and dugongs are more abundant. In addition, plots of
isotopic niches calculated from muscle tissues and generalised linear models suggested
that there was some effect of sex on diet. Female sharks had a larger trophic niche
(SEAb, TA and CD) and showed greater variability in δ13
C values, a pattern that may be
due to high physiological and energetic demands associated with reproductive cycles
(Sulikowski et al. 2016) which may require a more adaptable and opportunistic diet
than male sharks. However, analysis of dermis tissue revealed an opposite pattern to
muscle, with decreasing δ13
C values with increasing body size. This result must be
treated with caution because sample sizes of dermis were low (n = 34) and the range in
total lengths of the animals from which they were obtained were truncated compared to
analyses of muscle. Furthermore, with the exception of a single individual, dermis
samples of the largest sharks (>350 cm TL) were only collected from Ningaloo Reef
and these clustered at one end of the plot, thus may have strongly weighted the
relationship with size. Plots of the niche areas for all tissues did, however, separate
samples from Ningaloo Reef from those collected on the GBR, suggesting a diet shifted
towards more planktonic food webs at the former locality.
Analysis of fast-turnover tissues (red blood cells and plasma) also displayed a decrease
in δ13
C values with size, contrasting with results for muscle. In this case the shift in the
relationship with size could represent a seasonal or short term (months) component to
the diet that differed between smaller and larger sharks. Given that blood samples were
only taken within a time window of a few months of the year, even where sampling
extended over multiple years (as was the case at Shark Bay), it is difficult to determine
how often this trend with size was likely to occur. Notably, the effect of location evident
from the analysis of muscle tissue was also present in these tissues, with the trophic
niches of sharks from the GBR and Shark Bay largely overlapping and sharks at
Ningaloo having a mostly separated trophic niche with lower δ13
C values. The analysis
of plasma suggested that sharks sampled in Shark Bay and the GBR had much wider
trophic niches than those at Ningaloo, consistent with the stability analysis that also
suggested that diets of sharks at Ningaloo were more stable in short term (weeks-
months) than those of sharks on the GBR and at Shark Bay. The low probability of
sharks showing a stable diet in Shark Bay and the GBR suggests high variability in diet
across these short temporal scales and supports the idea that tissues with slower rates of
turnover (muscle, dermis, and fin) may provide a better picture of the overall diet of
108
tiger sharks. Nevertheless, results from the analysis of diet stability should be
interpreted with caution as residency was defined on the basis tracking data that was
only collected from sharks at Ningaloo Reef.
Overall, the local environment appeared to be a better determinant of the diet of tiger
sharks than latitude, despite the ability of the species to migrate long distances from the
tropics to cool temperate environments (Holmes et al. 2014; Ferreira et al. 2015).
Sharks sampled at locations that were separated by <300 km such as Ningaloo and
Shark Bay displayed very distinct stable isotopic compositions, which is unexpected
considering the movement patterns of these sharks between those locations (Heithaus et
al. 2007c; Ferreira et al. 2015). In contrast, stable isotopic composition of sharks
sampled from the mid-NSW coast were similar to those collected in central and
southern QLD, around 1000 km to the north, which may reflect seasonal movements
along that coastline (Holmes et al. 2014). The isotopic composition of sharks in more
temperate and pelagic habitats in Australia were also similar to sharks sampled at
Reunion Island (Trystram et al. 2016), an isolated atoll in the far western Indian Ocean.
Environment also appeared to determine the degree to which the species acted as an
apex predator, although in some localities (notably Ningaloo Reef and southern
Queensland) more comprehensive sampling of other elements of the fauna of pelagic
consumers is required to accurately place tiger sharks within the trophic pyramid.
The description of the species as a generalist predator (Simpfendorfer 1992; Lowe et al.
1996; Simpfendorfer et al. 2001; Matich et al. 2011; Trystram et al. 2016) 39
appears
justified with stable isotopic compositions being habitat-dependent and ranging from
seagrass (Shark Bay) to planktonic food chains (NSW coast) or a mixture of both
(Ningaloo Reef). Given that individuals tagged at Shark Bay and Ningaloo have
dispersed across the north-western coast of Australia (Heithaus et al. 2007c; Ferreira et
al. 2015) and sharks tagged in NSW have dispersed to the tropical Queensland coast
(Holmes et al. 2014), it appears that the degree of specialisation by an individual shark
on prey at any one time is likely to be context-dependent and reflect both resource
availability and to some extent, body size. Although tiger sharks at Ningaloo Reef
showed greater trophic diversity based on tissues with fast turnover (e.g. plasma), they
also had a greater stability of diet than sharks in Shark Bay and on the GBR. This might
suggest that tiger sharks at Ningaloo Reef have wider diet than those in Shark Bay and
on the GBR, but are more likely to remain within the same habitat, at least over short
time frames (weeks - months).
109
Our results emphasise the flexibility of the trophic ecology and role of tiger sharks
throughout their distribution range in Australia. Generalist predators such as tiger sharks
are able to explore multiple habitats and food-webs, likely adapting foraging strategies
to specific prey in different habitats. This generalist behaviour may both drive or be
driven by scale- and habitat-dependent space use by the species. However, separating
the cause and consequence of the relationship between movement and trophic ecology
for large, highly mobile, top-level predators is still challenging (Hays et al. 2016). An
important next step to better understand the relation between movement strategies and
feeding behaviour of these sharks is offered through the combination of diet analysis
and the information collected by high resolution tags (Fastloc GPS, camera tags and
accelerometers) that could provide a more comprehensive picture of animal movement
behaviours and patterns of prey search, and how they change in different habitats and
over time, so that can then be matched against animal trophic ecology.
4.6 Acknowledgements
Funding for this work was provided by the Royal Zoological Society of New South
Wales through the Paddy Pallin Science Grant, the University of Western Australia, the
Australian Institute of Marine Science, and the Conselho Nacional de Desenvolvimento
Científico e Tecnológico (CNPq, Brazil). The authors also acknowledge Chris Fischer
and the crew of the MV OCEARCH for the aid with logistics and field work, the
Australian Animal Tracking and Monitoring System (AATAMS) for providing receiver
data, and the Queensland Shark Control Program and the New South Wales
Gamefishing Association for providing samples.
110
4.7 Supporting information
Table S 4.1 Details of tiger sharks tagged at Ningaloo Reef in April-May 2015, their duration of
detection (until last receiver download) by receivers stations at Ningaloo and their behavioural
classification (resident or non-resident).
Shark ID Sex TL (cm) Monitoring
(months) Behaviour
1 F 373 3 Non-resident
2 F 319 1 Non-resident
3 F 372 2 Non-resident
4 F 391 1 Non-resident
5 F 407 2 Non-resident
6 F 285 6 Resident
7 F 363 0 Non-resident
8 F 304 5 Resident
9 F 390 6 Resident
10 F 390 3 Non-resident
11 F 303 3 Non-resident
12 F 405 0 Non-resident
13 F 322 5 Resident
14 F 352 0 Non-resident
15 F 268 5 Resident
16 F 290 0 Non-resident
17 F 393 1 Non-resident
18 F 282 6 Resident
Figure S 4.1. Location of acoustic receivers (grey and black circles) around Ningaloo Reef,
Australia. Black circles represent receivers where tiger sharks were detected. Star indicates
tagging location.
111
Chapter 5 Scale and habitat-dependent
movement behaviour of a large
marine predator, the tiger shark
Galeocerdo cuvier
5.1 Abstract
Tiger sharks are top-order marine predators found throughout the world’s tropical and
temperate oceans. Although many tracking datasets are available for the species, we still
lack information on the major processes that structure movement patterns and habitat
use at large spatial scales. This study combined multiple tracking datasets (108 tracks)
from locations in the Atlantic, Pacific and Indian oceans in order to define the major
environmental predictors of movement and space use of tiger sharks throughout most of
the distribution range of the species. Movement-based kernel methods and generalised
additive mixed models were applied to define predictors of utilisation distributions
(UD), and migratory movements. Geographically weighted regression (GWR) was then
used to test for spatial non-stationarity in the relationship between values predicted from
the models and those generated by kernel methods. Bathymetry and sea surface
temperature (SST) were identified as the major drivers of utilisation distributions and
migration patterns of tiger sharks of both sexes in different ocean basins, however, the
strength of the relationship with these drivers varied greatly among regions and the
explanatory power was relatively low (15% DE). This was due to spatial non-
stationarity, showing that the relationships changed across space. Although this may
have been due to missing biological covariates in models, notably food availability,
spatial resolution of both response and explanatory variables appear to be the main
driver. The combined results of additive mixed models and GWR show that the
relationships between sharks and the environment occurred at finer scales on the shelf
than in the open ocean, suggesting that movement patterns were both scale- and habitat-
dependent. This implies that the physical processes that aggregate food occur at
different spatial scales in each of these habitats and the analysis indicates the locations
where further research might reveal these underlying patterns.
112
5.2 Introduction
Large fishes and sharks, reptiles, cetaceans and pinnipeds (marine megafauna) tend to
act as top-order consumers and commonly occupy keystone roles in the trophic ecology
of marine ecosystems (Estes et al. 2016). The relative rarity of these top-order species in
food chains and their typically slow life history traits make them vulnerable to growing
anthropogenic threats (climate change, over-harvesting, pollution, habitat loss etc.) now
present throughout the world’s oceans (Stevens et al. 2000; Schipper et al. 2008; Hazen
et al. 2013; Worm et al. 2013; McCauley et al. 2015). As a consequence, the abundance
of many species has declined, so that they are now classified as Vulnerable, Threatened
or Endangered by the International Union for the Conservation of Nature (IUCN)
(Dulvy et al. 2014b).
Despite the importance of marine megafauna, management and conservation strategies
for this group are complicated by a lack of understanding of movement patterns
(Hooker et al. 2011; Hays et al. 2016). Although we know that most can range over
relatively large distances (100–1000s km), many of the key drivers of these behaviours
have not yet been quantified. In part, this is due to the difficulty of monitoring
movement patterns of such animals in the open ocean. Satellite-linked tags offer an
effective means to track marine megafauna (Hussey et al. 2015a; Hays et al. 2016),
however this technology is expensive and researchers must usually deal with significant
logistic hurdles in order to attach these tags to animals. As a result, most studies tend to
deploy relatively few tags, so that low sample sizes preclude the population or species-
level analyses of movement patterns necessary for the evaluation and optimisation of
conservation strategies and recovery plans (Hays et al. 2016).
Recent collaborative initiatives that have facilitated data sharing (e.g. TOPP, CLIOTOP,
Marine Megafauna Movement Synthesis Group, MEOP) have now enabled researchers
to overcome some of these issues (Costa et al. 2010a; Block et al. 2011; Hindell et al.
2016). For example, the pooling of the results of tagging of southern elephant seals
(Mirounga leonina) among studies has provided high-resolution information about the
physical environment occupied by the species throughout the entire Southern Ocean
(Biuw et al. 2007; Hindell et al. 2016). Nevertheless, such studies remain exceptions
and limitations on our ability to collect replicated, multi-year tracking data still impedes
a comprehensive understanding of the long-term movement behaviour and migration
patterns of many species of megafauna (Lea et al. 2015; Hays et al. 2016).
113
The tiger shark (Galeocerdo cuvier) is a large marine predator that typifies these
problems. In addition to the difficulties of deploying satellite tags, these sharks spend
limited time at the surface, restricting the opportunities for tags to broadcast location
data to satellites. Consequently, location estimates along tracks tend to be both sparse
and patchy, even when long (months–years) deployments of tags are achieved (Pinaud
2008; Meyer et al. 2009; Hammerschlag et al. 2012; Papastamatiou et al. 2013; Werry
et al. 2014; Ferreira et al. 2015; Lea et al. 2015). This complicates the analysis of
movement patterns, information that is urgently required given that there is increasing
concern over the conservation status and management of the species due to
unsustainable demand for shark products in some markets, compounded by high rates of
bycatch of sharks in many fisheries (Pauly et al. 2005; Lewison et al. 2014; Dulvy et al.
2014b). Such threats and uncertainty in population trajectories have led to designation
of the species as “Near Threatened” by the IUCN (Ferreira et al. in review).
Here, these issues were addressed using two approaches. Firstly, the results of a large
number of tracking studies of tiger sharks were pooled to form a large, global data set so
that general patterns in movement behaviour could be identified across the entire range
of the species. Secondly, a novel combination of statistical approaches was applied to
identify the drivers of movement patterns and to disentangle the biological and physical
factors likely to structure this process. Although the focus of the work was on tiger
sharks, these techniques are applicable to a wide range of marine megafauna, given that
sparse data with low sample sizes are common issues for most tracking studies. Firstly,
the Biased Random Bridge (BRB) kernel method (Benhamou 2011) was used to
estimate monthly utilisation distributions (UD) from the tracks of each shark. This
method differs from traditional kernel approaches in that it incorporates movement
trajectories between locations and aspects of the landscape, in addition to accounting for
the temporally auto-correlated nature of animal trajectories (Benhamou 2011).
Utilisation distributions created by BRB analyses were then used as response variables
in generalised additive mixed models that examined the influence of a variety of
physical variables (sea surface temperature (SST), depth, current direction etc.) in
driving migration and high use of particular areas. Finally, geographically weighted
regression (GWR) (Fotheringham et al. 1998) was used to identify non-stationarity in
the relationships between these variables, showing areas where there was higher or
lower utilisation than predicted by the model. This approach is a new concept for
marine species, but has been used for studies of species distribution and occurrence in
114
the terrestrial realm (Foody 2004; Osborne et al. 2007; Powney et al. 2010; Eiserhardt
et al. 2011).
Tiger sharks are an ideal model species for this work for a number of reasons. They are
a large (up to 5 m total length) and mobile (Heithaus et al. 2007c; Meyer et al. 2009;
Hammerschlag et al. 2012; Hazin et al. 2013; Werry et al. 2014; Ferreira et al. 2015;
Lea et al. 2015) predator present throughout the world’s tropical and warm temperate
oceans (Compagno 1984; Randall 1992). As top-order predators, they are iconic and
charismatic and have been the subject of a number of tracking studies world-wide,
although these tend to have limited sample sizes or have been restricted in spatial and/or
temporal scope (Meyer et al. 2009; Hazin et al. 2013; Werry et al. 2014; Holmes et al.
2014; Ferreira et al. 2015). Sampling by these studies also tends to be biased towards
female sharks (but see Lea et al. 2015), possibly reflecting easier accessibility of this
sex for tagging. However, the number and diversity of localities where tagging has
occurred offers the opportunity for pooled data sets to provide a global picture of
movement behaviours of these predatory sharks. We combine the results of tracking
studies from eight localities across the Indian, Pacific and Atlantic oceans and using this
pooled data set, we seek to identify the key physical and biological drivers of movement
patterns at global scales. We hypothesise that the drivers of high use of areas and
migration will be different, thus we identify these components in the data and analyse
them separately as response variables against a range of environmental variables and
deployment location. We also hypothesise that the variability of movement patterns
reported previously for this species might be related to sex and ontogeny and thus we
examine the importance of total length (as a proxy for age) and sex as explanatory
variables in the models.
5.3 Methods
5.3.1 Telemetry
Telemetry data from 130 adult and sub-adult tiger sharks of both sexes, tagged at
Ningaloo Reef (13 sharks), Shark Bay (9), Great Barrier Reef (12), USA (Florida and
the Bahamas - 71), Hawaii (3) and Brazil (Recife - 18; Fernando de Noronha Island - 4)
between 2008 and 2015 were combined to create a global database (Figure 5.1). Several
types of telemetry tags were used, including ARGOS satellite transmitting tags (SAT;
mostly Wildlife Computers SPOT and SPLASH), and Pop-up archival tags (PAT;
115
mostly Wildlife Computers PSAT Mk10 and miniPAT; see Table S1). Sharks were
caught by drumlines, longlines and rod-and-reel and tags were attached to the first
dorsal fin using either corrodible bolts or by a 2.0 mm polyamide monofilament line
passed through a hole in the anterior region of the fin. All procedures were approved by
the Animal Ethics Committees of the organisations that led the field work. For more
details on these methods see Heithaus et al. (2007c), Fitzpatrick et al. (2012),
Hammerschlag et al. (2012), Hazin et al. (2013), Ferreira et al. (2015).
Location estimates were obtained from either the Argos satellite network (SPOT and
SPLASH) or via light-based geolocation (PAT) (Hill & Braun 2001). For SAT tags, all
Argos Location Classes (3, 2, 1, 0, A, B, Z) were included in the analysis. Raw
locations from Argos were filtered according to Ferreira et al. (2015), with extreme
location estimates (e.g. points requiring movement rate of > 1000 km a day) either
removed or replaced by the secondary locations reported by Argos. Tracks shorter than
30 days or with very sparse locations (e.g. 5 location estimates in >100 days) were not
included in the analysis. This gave a total sample size of 105 individual tiger shark
tracks for analysis. We attempted to use the BSAM package in R (Jonsen et al. 2016) to
fit state-space models to account for location error, however these models did not
converge due to sparse location data.
Location estimates for tiger sharks tagged with PAT tags were calculated from light
level data recorded and stored on board the tag. The data was relayed via the Argos
Network after detachment of the tag from the shark or downloaded from the archive
when the tag was recovered. Locations were estimated from light data using the
software WC-GPE: Global Position Estimator Program (Wildlife Computers Inc.). The
most probable track was reconstructed using the TrackIt state-space model with a
Kalman filter (Nielsen & Sibert 2007) in R (R Core Team 2013). Sea surface
temperature (SST) was applied to the model to improve the accuracy of location
estimates and reduce confidence intervals (Lam et al. 2010). However, the majority of
tiger sharks tracked with PAT tags remained within the equatorial Atlantic Ocean (Fig
1) where SSTs are relatively uniform and cloud cover is high. This led to model under-
performance and convergence (Nielsen & Sibert 2007) so that SST corrections were not
incorporated in some tracks.
116
5.3.2 Environmental data
Data for SST and ocean currents (direction and magnitude) were obtained using the
Marine Geospatial Ecology Tools (MGET) for ArcGis10.3 (Roberts et al. 2010). Daily
and monthly average SSTs were generated by the Moderate Resolution Imaging
Spectroradiometer (MODIS), Aqua satellite Level 3 with a 4 km resolution. Daily data
on direction and magnitude of ocean currents was derived from the Hybrid Coordinate
Ocean Model with a 1/12 degree resolution. Bathymetry data was obtained from the
General Bathymetry Chart of the Oceans Gebco15 database in a 30 arc-second
resolution grid (http://www.gebco.net). A suite of bathymetric covariates were
calculated from this data that have been previously associated with distributions of
marine fauna (Bouchet et al. 2015; Yates et al. 2016) and these were used as a proxy for
habitat complexity. Bathymetric predictors were calculated in ArcGis 10.3 using digital
terrain analysis with fixed window sizes (Holmes et al. 2008) and a resolution of 1 km
to match the bathymetry dataset (Table 5.1).
5.3.3 Identification of areas of high use
In order to identify areas of high use through time, utilisation distributions were
calculated monthly across a 4 km square grid for each shark using the Biased Random
Bridge (BRB) kernel method (monthly UD values were calculated to account for the
dynamic nature of some covariates (e.g. SST). Similar to the Brownian Bridge kernel
method (Bullard 1991), the time spent between successive locations and the correlation
between locations is taken into account in the calculation of utilisation distributions.
However, the BRB approach also includes a component of advection in the trajectory so
that movement from one location to the next is considered to be biased towards the next
location. This is an important distinction, because successive locations in the track of an
animal are typically auto-correlated, where the next is bound both to the previous
starting location and the capability of an animal for locomotion within the given time
period (Papworth et al. 2012). The advection component of the analysis affects the
orientation and shape of the bridge, with stronger advection generating longer and
narrower bridges (Benhamou 2011). An upper time limit between successive locations
is also used as a threshold value in the kernel bridging computation to ensure that steps
longer than this maximum are not included in the analysis. Here, a maximum threshold
of 5 days was imposed, based on the average rate of transmission from the tiger shark
117
tags, as it was assumed that steps longer than this would result in unrealistic estimations
of utilisation distributions.
The BRB method also includes a component related to the uncertainty of locations and
the mean values for the track of the error associated with each ARGOS Location Class
(LC 3 = 270 m, LC 2= 540 m, LC 1 = 1330 m, LC 0 = 5179 m, A = 8072 m, B = 11484
m) were used for this analysis (Hays et al. 2001a; Hazel 2009). For geolocations,
location uncertainty was estimated from the variance of location estimates provided by
the TrackIT model. All analyses for the BRB used the adehabitatHR package (Calenge
2011) in the software R (R Core Team 2013).
The monthly utilisation distributions were then converted into point data, where each
point contained the UD value for each 4 km grid cell, and the values were normalised to
values between 0 and 1, so that the gridded UD value for each cell became the
probability of shark use of a grid cell. Utilisation distribution is a similar concept to that
of home range (Powell 2000), where home range cores of an animal typically range
from 25% to 95% with the lower values containing the highest density of points.
However, as a result of the normalisation, utilisation distribution values here were
represented as the inverse of home range (i.e. 0.75 UD represented the 25% HR core,
whereas 0.25 UD represented the 75% HR core). This produced distributions of values,
more useful as a response variable in subsequent models; for example a 25% HR core
corresponds to a distribution of values between 0.75-1 UD. The larger the value of the
UD for a cell in the grid, the greater the use of that cell by the shark. Normalised values
of monthly utilisation distributions for males were log-transformed to improve
distribution of the model residuals.
In order to identify general areas of high use for tiger sharks in each ocean basin, these
normalised gridded monthly utilisation distribution of values between 0.75-1 UD for all
sharks in each region were also averaged, so that each grid cell represented the average
UD of all sharks that utilised that cell.
5.3.4 Identification of migratory behaviour
In order to examine the environmental drivers of shark migratory behaviour, periods of
migration within tracks were identified, as long, directional, relatively straight,
extensive movement (Dingle & Drake 2007; Lea et al. 2015), as opposed to residency,
which was defined as a cluster of locations in a defined area (Figure S 5.1). To
118
quantitatively identify these behaviours, utilisation distributions were calculated for the
entire track of each shark using BRB. Tracks were then interpolated to a single location
estimate per day with the R package crawl (Johnson 2014) in order to match the
resolution of the environmental data. Gaps in the data longer than 20 days were not
interpolated to avoid unrealistic interpolated paths (Block et al. 2011; Queiroz et al.
2016). The interpolated track from each individual shark was then overlaid on the 0.75
utilisation distribution indicating a high degree of residency (Ferreira et al. 2015). All
locations inside the 0.75 UD were categorised as resident and assigned a value of 0 and
locations outside the 0.75 UD categorised as migratory and assigned a value of 1
(Figure S 5.1). Resident was defined as the behaviour associated with high-use areas,
whereas migrant is the straight-line behaviour associated with large-scale movement.
5.3.5 Modelling movement behaviour
Habitat use
As tagging was skewed towards females (82% of tagged individuals) UD was analysed
separately for each sex to avoid over-estimating parameters for males. Due to the large
size of the dataset for females (4,211,047 data points), the normalised UD dataset was
filtered by removing values below 0.25 UD. These values represented the grids with
very low probability of use that were present in large numbers as a result of the shape of
a Brownian Bridge (Bullard 1991), but provided only low explanatory power of the
drivers of residency. The dataset for females was also analysed separately for each
region: North Atlantic, South Atlantic, Indian and Pacific oceans (Hawaii and Great
Barrier Reef were analysed separately) due to differences in the range of some of the
environmental variables among regions. Utilisation density (normalised for females,
normalised and log-transformed for males) was used as the dependent variable in a suite
of generalised additive mixed models with a Gamma distribution with an inverse link
function, with individual shark as the random effect using the R function bam (gam for
large datasets) from the R package mgcv (Wood 2016). A set of biological, temporal
and environmental variables was assembled to determine the important predictors of
high use of habitat by tiger sharks (Table 5.1). Exploratory analyses identified any
collinearity within the set of bathymetric covariates and only one was selected for use in
the models when they were highly correlated (Pearson’s r > 0.8). Models with all
possible combinations of selected predictor variables were fitted then compared and
ranked according to Akaike’s Information Criterion (AIC), and by the AIC weight
119
(wAIC). The wAIC varies from 0 (no support) to 1 (complete support). The amount of
variance (percentage deviance) in the response variable explained by each of the
candidate models was used as a measure of goodness-of-fit to the data (Anderson &
Burnham 2002; Burnham & Anderson 2004). We considered all models within 2 AIC
units of the top-ranked model, using the principal of parsimony to select the final
model. Conditional plots of UD relative to each explanatory variable in the top-ranked
models were generated using the package visreg (Breheny & Burchett 2016).
120
Figure 5.1 Telemetry data included in the analyses, colour coded by individual shark tag ID, from Hawaii (Pacific Ocean) (top left), North Atlantic (top right), South
Atlantic (left), Great Barrier Reef, Australia (Pacific Ocean) (middle) and Indian Ocean, Australia (bottom right).
121
Migration
The drivers of migration were also examined using generalised additive mixed models
with a binomial distribution, using the package gamm4 (Wood 2016). The probability of
migration (migrant = 1, resident = 0) was the dependent term in the models with a set of
environmental predictors (Table 5.2) and a random effect of individual shark. Because
direction of ocean current was a circular variable, it was modelled with a cyclic cubic
regression spline, whereas a regular cubic regression spline with shrinkage was applied
to the other (non-circular) covariates. The number of predictors in each model was set to
two or less and basis dimension “k” to five to avoid over-fitting. Autocorrelation was
addressed by using a matched-block bootstrap sampling with replacement procedure
(Carlstein et al. 1998; Patton et al. 2009). Model fitting was applied to 1000
bootstrapped samples and model selection was as described above, but with upper and
lower quartiles also presented.
5.3.6 Assessment of spatial variability in model
predictions
Geographically Weighted Regression (GWR) is a tool used to explore spatially varying
relationships between variables, termed spatial non-stationarity (Fotheringham et al.
1998). A global regression model applied to spatial data will estimate a single set of
parameters for the relationship between the response and explanatory variables.
However, the presence of spatial non-stationarity alters the relationship between
variables in space, limiting the descriptive and predictive power of a model (Foody
2004). Non-stationarity might occur due to methodological issues (model
misspecification, missing variables and/or their use at inappropriate temporal or spatial
scales), spatially auto-correlated data and, in this case, potential differences in
movement behaviour of animals among locations. Geographically Weighted Regression
accounts for these issues, allowing parameters to be estimated locally by fitting a
regression equation to any point in space to explore local spatial relationships.
Consequently, it allows relationships to vary over space so that localised model
parameters are estimated around each regression point. Geographical weighting is
implemented on neighbouring observations according to a spatial kernel function that is
then used in the localised model. The weighting of observations decreases as the
distance from the centre of the kernel increases and the area or number of neighbouring
122
observations around the centre of the kernel used in the weighting function is controlled
by the bandwidth, which ultimately defines the scale of the analysis (Foody 2004).
Geographically Weighted Regression has been used as a tool to explore localised
variation in the relationships of variables used in a global model such as a logistic
regression, generalised linear models or generalised additive models (Foody 2004;
Osborne et al. 2007; Windle et al. 2009; Miller & Hanham 2011). However, in contrast
to previous studies, GWR was used here to identify the presence of non-stationarity in
the relationship between the predicted values of UD (from the top-ranked models)
against the observed value of UD. The ‘global model’ was thus a regression between
predicted and observed values from each top-ranked model in the analysis of drivers of
high habitat use. Effectively, the analysis determined if the environmental drivers
identified in top-ranked models adequately explained areas of high use and if this
relationship varied spatially. A Gaussian kernel was used in the geographical weighting
so that all data points contributed to the regression, but points spatially distant from the
centre of the kernel were inversely weighted according to a Gaussian curve. An adaptive
bandwidth was also used in the weighting function (e.g. an adaptive distance of the
kernel bandwidth) that ensured that sufficient local information was used for each local
model in an irregular sampling configuration. Automated optimal bandwidth selection
was determined by minimising AIC. The presence of spatial non-stationarity was
evaluated by the range in the local parameters of GWR coefficients where large
variations of values and the presence of highly positive or negative clusters suggested
non-stationarity. All GWR analyses were done in ArcGIS 10.3 software.
5.4 Results
The telemetry dataset provided a total of 14,357 tracking days of tiger sharks with
average track duration of 136 ± 141 days for females and 85 ± 76 days for males
(Figure 5.1, Table 5.3). Females were the dominant sex tagged in all regions (82.3% of
all tracks) and the majority of male tiger sharks in the sample (82%) were tagged in the
Atlantic Ocean. Female sharks ranged in total length from 120 - 407 cm (average 289 ±
68 cm) whereas males ranged from 130 - 395 cm (average 231 ± 72 cm). Sharks tagged
in the South Atlantic appeared to be smaller (average TL of 186 ± 54 cm) than in the
Indian (mean 314 ± 55 cm), North Atlantic (mean 288 ± 68 cm) and Pacific (319 ± 52
cm) oceans (Table 5.3).
123
Figure 5.2 Utilisation densities combined and averaged for each month for each grid cell for all
individuals in each regional dataset and filtered to the 0.75UD. Colour scale represents the
average utilisation of each cell and grey contours indicate the 500 m isobath.
124
Table 5.1 Description of all environmental, temporal and biological predictors for the analysis of monthly utilisation distribution.
Table 5.2 Predictors used to model migratory movements of tiger sharks.
Predictor Resolution Description
Depth 30-arc second Water depth (m) in relation to bathymetry extracted from Gebco15
Aspect 1 km Direction of the steepest slope in degrees (0-360°) on a 3 x 3 cell area
Slope 1 km First derivative of elevation: Average change in elevation calculated on a 3 x 3 cell area. Calculated in degrees (0-360)
Curvature 1 km Combined index of profile (hill cross-section) and plan (contour lines) curvatures
Standard Deviation 1 km Standard deviation of elevation on a 3 x 3 cell area
Plan curvature 1 km Second derivative of elevation: concavity/convexity perpendicular to the slope, calculated on a 3 x 3 cell area
Profile curvature 1 km Second derivative of elevation: concavity/convexity parallel to the slope, calculated on a 3 x 3 cell area
Hypsometric index 1 km Indicator of whether a cell is a high or low point within the local neighbourhood
Local relief (Range) 1 km Maximum minus the minimum elevation in a local neighbourhood
Sea Surface Temperature 4 km Monthly SST(°C) generated from the Moderate Resolution Imaging Spectroradiometer (MODIS), Aqua satellite Level 3
Month 1 - 12 Month the utilisation distribution was calculated
Total length - Total length of the shark (cm)
Predictor Resolution Description
Depth 30-arc second Water depth (m) in relation to bathymetry extracted from Gebco15
Sea Surface
Temperature
4 km Daily SST (°C)extracted from the Moderate Resolution Imaging Spectroradiometer
(MODIS), Aqua satellite Level 3
Ocean current direction 1/12 degree Daily direction in degrees (0-360°) derived from the Hybrid Coordinate Ocean Model -
HYCOM
Ocean current speed 1/12 degree Daily absolute magnitude of water velocity (m/s) derived from the Hybrid Coordinate Ocean
Model - HYCOM
125
Table 5.3 Summary tracking datasets including information of tagging location, number of tags deployed, track duration and size range of sharks tagged. Mean
values and SD are presented for total length (TL) and track duration.
Ocean Tagging locations Tag Types Deployment
years Sex
N
tags TL range Duration
Indian Ningaloo, Shark
Bay SPOT, SPLASH 2007-2015
F 20 312.3 ±53.5 177.95 ±143.4
M 2 395, 271 38, 191
Pacific Hawaii, Great
Barrier Reef
SPOT,
SPLASH,
Telonics ST18
2003-2013
F 13 306.3 ±59.8 129.5 ±116.6
M 2 288,292 63, 231
North
Atlantic
Florida, The
Bahamas SPOT, uCricket 2009-2015
F 60 296.1 ±58.3 141.6 ± 155.6
M 11 247.8 ±56.5 76.1 ± 57.05
South
Atlantic
Recife, Fernando
de Noronha Island
PAT-Mk10,
MiniPAT 2008-2013
F 15 198.7 ±60.3 58.8 ± 38.8
M 7 159.6 ±25.7 45.57 ±2 4.1
126
Scale of movement and habitats varied greatly among individual sharks and regions.
Wider ranging tracks were recorded in the North Atlantic, the Indian Ocean and off
Hawaii, where large-scale (100s - 1000s km) movements along the coast and/or into the
open ocean were observed (Figure 5.1). Tracks from the South Atlantic were obtained
from PSAT deployments that provided only low resolution estimates of geolocation and
tracking durations that were never >120 days. This reduced the likelihood of identifying
patterns of high habitat use and migratory behaviour for sharks in this region.
In the North Atlantic, average utilisation distributions for all sharks combined showed
areas of high use around the tagging site off the Bahamas, but also in the Gulf of
Mexico, in the waters of North Carolina and around the edge of the Gulf Stream (41°N)
(Figure 5.2a-c). In Western Australia, high use areas were associated with tagging
locations (Ningaloo Reef and Shark Bay, Figure 5.2d), while in Hawaii the sharks
mostly used oceanic areas north of the islands (Figure 5.2e). In the South Atlantic, an
area of high use was identified in the equatorial region off the coast of northern Brazil
(Figure 5.2f), whereas on the Great Barrier Reef, sharks had small areas of high use
(Figure 5.2g).
5.4.1 Drivers of high use of habitats
The model including depth, SST, aspect and standard deviation of elevation had 100%
support (wAIC = 1) out of the suite of models fitted to explain utilisation distribution of
females in the North Atlantic (Table 5.4), whereas for the South Atlantic and Indian
Ocean the model with depth and SST only had 100% support (wAIC= 1) (Table 5.4).
For Hawaii, the model with SST only had majority support (wAIC = 0.93) and for the
Great Barrier Reef, the model with depth only had complete support (wAIC = 1) (Table
5.4). In the North Atlantic, females showed an elevated utilisation of oceanic waters
(>3,000 m deep) beyond the continental shelf and in temperatures below 22°C (Figure
5.3a-b). Higher values of UD were also associated with bathymetric aspects in north and
northeast directions (350° and 50°), and areas with minor deviations of elevation
(Figure 5.3i-j). In the Indian Ocean, residency was greater in water temperature
between 22°-24°C and oceanic areas around 3000 m deep (Figure 5.3c,d), with some
preference for depths around 250 m. In the Pacific Ocean around the Great Barrier Reef
(Pacific Ocean), utilisation increased with increasing depths (>1000 m Figure 5.3e). For
sharks in Hawaii, higher values of UD were related to temperatures between 24°-25°C
127
(Figure 5.3f). In the South Atlantic, sharks showed both preference for open waters
>4000 m and the shallow shelf, and waters around 25°C (Figure 5.3g-h).
Table 5.4 The top ranked generalised additive mixed models from the suite of models used to
explain monthly utilisation distributions of female and male tiger sharks. SST = sea surface
temperature, Std = standard deviation of elevation. Shown are the percent deviance explained
(%DE), Akaike’s information criterion (AIC), and AIC weight (wAIC).
For males, the model including depth and sea surface temperature had complete support
(wAIC = 1) and explained 14.4% of the total deviance in the data (Table 5.4). Males
exhibited high utilisation of habitats associated with shallow depths, in addition to sea
surface temperatures of around 26°C or above 30°C (Figure 5.4), which are
representative of coastal and shelf environments.
5.4.2 Drivers of migration
Migration was identified in tracks from female sharks tagged in the North Atlantic,
Indian and Pacific oceans. As the tracks from the region of the South Atlantic showed
no evidence for this behaviour they were excluded from this analysis. Large-scale
movements by sharks to northern areas of the North Atlantic were observed during
summer, with sharks reaching the margins of the Gulf Stream, whereas movements
south to temperate regions were observed in the Indian Ocean during austral autumn
(Figure 5.5). A repeated annual migration to the south was only observed for one shark
in the Indian Ocean (T5, Table S 5.1).
Model % DE AIC wAIC
Females
North Atlantic
Depth + SST+Aspect +Std 11.9 -71047.5 1
South Atlantic
Depth+ SST 7.94 -69164.7 1
Indian Ocean
Depth + SST 14.4 -23959.6 1
Hawaii
SST
Depth
7.74
7.39
-910.2
-905.2
0.93
0.07
Great Barrier Reef
Depth 12.6 -2636.317 1
Males
Depth + SST 14.4 757422.6 1
128
The top ranked model to explain the probability of tiger sharks being in migratory mode
contained the variables depth and sea surface temperature, with 68% support and
accounting for 7% of the total deviance explained (Table 5.5). The model predictions
showed a high probability of migration across the full range of environmental variables
(>0.5, Figure 5.5). Migration was more likely to occur in oceanic areas over depths
>1000 m (Figure 5.5). Increasing probability of migration was also related to water
temperatures from both tropical and subtropical-temperate regions with likelihood of
migration being highest in water temperatures above 28°C and below 22° C (Figure
5.5).
129
Table 5.5 The top three ranked generalised additive mixed models with 1000 bootstraps sampling for the suite of models used to explain the probability of migration
of tiger sharks. Shown for each model are the median, 25th and 75
th quantile Akaike’s information criterion corrected for small samples (AICc), AICc weight
(wAICc) and the adjusted R-squared (Adj R2). The top ranked model is in bold.
Model AICc wAICc % DE AICc.25 AICc.75 wAICc.25 wAICc.75 Adj R2.25 Adj R
2.75
Depth + SST 2689.06 0.68 0.07 2553.99 2833.11 0.04 1.00 0.04 0.11
Depth + Current direction 2713.43 0.18 0.01 2590.20 2852.38 0 0.10 0.00 0.10
Depth + Current speed 2718.00 0.14 0.04 2591.91 2856.18 0 0.02 0.03 0.10
130
Figure 5.3 Conditional plots of female utilisation density relative to the explanatory variables in
the top ranked mixed model for each region from the suite of models used to explain the
response variable. Solid curves are the model fit and shaded areas indicate the 95% confidence
intervals. Tick marks on the x-axis indicate the distribution of observations. Plots a, c, e, g are
fits for bathymetry for North Atlantic, Indian Ocean, Pacific (Great Barrier Reef) and South
Atlantic, respectively; b, d, f, h are fits for sea surface temperature for North Atlantic, Indian
Ocean, Pacific Ocean (Hawaii) and South Atlantic, respectively; i-j are fits for standard
deviation of elevation and aspect in the North Atlantic.
131
5.4.3 Assessment of spatial variability in model
predictions
The local coefficient estimates produced by the GWR indicated the presence of spatial
non-stationarity in all regions for UDs of females (Figure 5.6a-g) but not for males,
with the exception of the Great Barrier Reef (Figure 5.7a-d). This was shown by the
large variation of GWR estimates for females, indicating the presence of spatial
variability in localised relationships, particularly along the east coast of USA (Figure
5.6a,b) and off the coast of Western Australia (Figure 5.6c,d). By mapping the GWR
coefficient estimates it was possible to identify areas where the observed values of UD
were much higher or lower than the values predicted by the models. For females in the
North Atlantic, a core of strongly positive GWR estimates (utilisation higher than
predicted by the top-ranked model) were identified along the Florida and North Carolina
coasts and in the Bahamas (Figure 5.6a,b). Similarly, localised areas of strongly positive
values were also seen on the northwest coasts of Western Australia around the latitudes
19°-20°S and 24°S (Figure 5.6c,d). Small areas of positive values were also found along
the Great Barrier Reef (Figure 5.6f). The presence of higher-than-predicted values
suggests that other localised factors were accounting for high use in these areas. For
females in the South Atlantic, GWR local parameters showed a particularly large
variation with a few cores of highly negative and one area with highly positive values
(Figure 5.6g).
Figure 5.4 Conditional plots of male utilisation density relative to the explanatory variables in
the top ranked mixed model from the suite of models used to explain the response variable.
Solid curves are the model fit and shaded areas indicate the 95% confidence intervals. Tick
marks on the x-axis indicate the distribution of observations.
132
Figure 5.5 Conditional plots of probability of migration relative to the explanatory variables in
the top ranked mixed model from the suite of models used to explain the response variable.
Solid curves are the model fit and shaded areas indicate the 95% confidence intervals. Tick
marks on the x-axis indicate the distribution of observations.
133
Figure 5.6 GWR-derived local coefficient estimates for the relationship between predicted and
observed values from the top ranked mixed model explaining the drivers of habitat use for
females. A-B = North Atlantic Ocean and B is zoomed in on the box in A, C-D = Indian Ocean
and D is zoomed in on the box in C, E = Pacific Ocean around Hawaii, F = Great Barrier Reef,
G = South Atlantic. Grey isolines represent bathymetry. Red and orange show areas where
observed UD was higher than predicted (see detail in B and D) and blue and green show area
where observed UD was lower than predicted.
134
Figure 5.7 GWR-derived local coefficient estimates for the relationship between predicted and
observed values from the top ranked mixed model explaining the drivers of habitat use for
males. A-B = North Atlantic Ocean, B = Indian Ocean, C = Great Barrier Reef, D = South
Atlantic. Grey isolines represent bathymetry. Red and orange show areas where observed UD
was higher than predicted and blue and green show area where observed UD was lower than
predicted.
5.5 Discussion
Depth and sea surface temperature were identified as key physical variables structuring
patterns of both high habitat use and migration patterns of tiger sharks. However, the
relationships between these variables and shark movements were inconsistent among
regions. Because tiger sharks mostly occupy the surface layers of the water column
(Vaudo et al. 2014), water depth at a location provided an indication of the use of shelf
(depth <200 m) and oceanic waters. In the North Atlantic, many of the female sharks
tagged off Florida and the Bahamas displayed tracks that followed the Gulf Stream and
had excursions into the Atlantic Ocean basin, thousands of kilometres from the coast.
This was reflected in the results of the mixed models, which showed preferences for the
use of habitats in very deep water (5000-6000 m). In contrast, females tagged in
Australia at Ningaloo Reef and Shark Bay tended to reside in waters closer to the coast
just beyond the shelf break, so that conditional plots from the mixed models showed
preferences of tiger sharks for shallower oceanic habitats (3000 m) than in the North
135
Atlantic. On the Great Barrier Reef females were tagged at the edge of the shelf and
displayed residencies both off the edge of the shelf and in deeper water of the Coral Sea.
Similarly, relationships between utilisation distribution and water temperatures also
varied among regions. Females in the North Atlantic and Indian oceans showed a
preference for cooler waters of <20°C and 23°C, respectively, whereas females off
Hawaii and in the South Atlantic preferred temperatures between 25-26°C. In contrast,
males showed greater use of shelf waters and selected temperatures of 25-26°C or above
30°C. Depth and sea surface temperature were also identified as the main drivers of
migratory movements of female sharks, with these occurring beyond continental shelves
and in water temperatures either below 22°C, or above 26°C. No migratory movements
could be identified in the tracks of male tiger sharks. Ocean currents have been
previously suggested as a possible driver of shark migration (Hazin et al. 2013; Holmes
et al. 2014; Ferreira et al. 2015; Lea et al. 2015). However, our mixed models did not
identify current speed or direction as important predictors in the top-ranked model,
despite the fact that long-term (several months–years) tracks of sharks in both the North
Atlantic and off the coast of Western Australia showed excursions of sharks from
tropical waters into temperate regions that followed the path of principal boundary
currents (the Gulf Stream in the North Atlantic and the Leeuwin Current in the Indian
Ocean). A lack of any easily detectable influence of direction and speed of water flow in
these long-distance movements of tiger sharks might be due to animals undertaking
vertical movements during migration to take advantage of counter-current flows at
depth or for behavioural thermoregulation (Thums et al. 2012), although such behaviour
has yet to be fully investigated for the species (Nakamura et al. 2011; Iosilevskii et al.
2012). These phenomena could mask simple relationships between migration patterns
and current systems measured by remote sensing at the ocean surface.
Despite amassing the largest tracking data set yet assembled, our models explained
relatively little of the variation in patterns of high utilisation and migration of tiger
sharks. At best, physical predictors of water depth and SST explained less than 15% of
the variance in utilisation distributions for both female and male sharks. The GWR
highlighted that spatial non-stationarity existed in the relationship between utilisation
distributions and values of UD predicted by top-ranked mixed models for each ocean
basin, thereby showing that relationships vary across space, accounting for the low
explanatory power of the models (Osborne & Suárez-Seoane 2002). Overall, the GWR
showed that the models predictions were consistent with most areas of high utilisation
136
by tiger sharks, reinforcing that water temperatures and depths were reasonable
predictors of spatial patterns of habitat use at least in open ocean environments. The
non-stationarity identified for females was most pronounced in the Atlantic and Indian
oceans, as evidenced by the large range in GWR coefficient estimates. The GWR
highlighted a number of areas where the observed UDs were much higher or lower than
those predicted by our models. Two areas at the shelf edge and the mid-shelf off the
coast of Western Australia, around latitudes of 19°-21°S (Figure 5.6d), had strongly
positive GWR estimates (observed values of utilisation greater than predicted by the
models). These matched hotspots of abundances of pelagic fishes documented in earlier
studies by Bouchet (2015) (Figure S 5.2). Another zone of strongly positive GWR
estimates was present off the coast from 24-25°S, in an area of high occupancy for
pygmy blue whales (Balaenoptera musculus brevicauda) (Double et al. 2014) (Figure S
5.2). The coincidence of these distributions suggest that prey availability may indeed
account the spatial non-stationarity found in this area. However, many tracks had
relatively short duration and, therefore, areas of high use indicated by GWR may also
have been a result of local residency demonstrate by sharks tracked for shorter periods.
The GWR also identified areas on the shelf on the eastern coast of the US where the
observed UD was higher than model predictions (26°-27°N and 34°-36°N). It is notable
that these areas of high use for tiger sharks were also areas of high use for other coastal
sharks. For these coastal species, residency and movement patterns were strongly
correlated with water temperatures, relationships that were documented using data
collected at relatively fine spatial scales by oceanographic buoys and loggers deployed
across the study area (Kessel et al. 2014; Kajiura & Tellman 2016). A similar approach
may be necessary to fully describe residency and movement patterns of tiger sharks in
these habitats. In the South Atlantic, the problem of non-stationarity was probably
driven by the low resolution of both tracking and environmental data. This was due to
the reconstruction of tracks from geolocations using light data, which have large errors
associated with position estimates (Lam et al. 2010). As a consequence, maps of GWR
coefficient estimates in the South Atlantic showed a very broad spread of values across
the sampling area.
For the North Atlantic and Indian Ocean data sets, it appears that the reasons for spatial
non-stationarity are also a result of scale, but in more nuanced way. The results of the
GWR suggested that SST and depth could accurately predict the use of oceanic habitats
in the North Atlantic and Indian Ocean, probably because tiger sharks in oceanic
137
habitats are operating at large and mesoscales (10s -100s km) that can be identified at
the scale of resolution of the environmental data sets provided by satellite remote
sensing. However, in shelf waters tiger shark movements and the processes driving their
utilisation patterns are likely to operate at far smaller spatial scales, where topographic
features such as canyons, banks and reefs or habitats such as estuaries and seagrass
meadows may locally enhance productivity of food chains that sustain the targets of
feeding by tiger sharks (Heithaus et al. 2002, 2007a; Morato et al. 2010; Bouchet 2015).
The environmental signals of these habitats may be very difficult to identify in the low-
resolution data provided by remote sensing.
It has been suggested that movements of tiger sharks are mainly driven by prey
distribution (Heithaus et al. 2002), but to date this variable has not been included in any
models that seek to predict movement patterns. This is largely due to the difficulties of
collecting data sets of prey availability at scales comparable to the spatial and temporal
patterns of movement of these predators and the issue of matching predator and prey
fields is recognised as a major problem in the field of ecology of marine megafauna
(Hindell 2008; Hays et al. 2016). For tiger sharks, describing the prey field is further
complicated by the fact that the species has a very broad diet and is both an top-order
predator and scavenger depending on the habitat, prey field and opportunities that are
presented within the environment (Lowe et al. 1996; Simpfendorfer et al. 2001;
Gallagher et al. 2011; Hammerschlag et al. 2016). This makes identification of potential
prey and mapping of their distributions within the environment a very difficult task.
The finding that this scale-dependent behaviour may be causing models to have low
predictive power to describe utilisation patterns is perhaps not surprising, given that
tiger sharks are recognised as a generalist predator (Lowe et al. 1996; Simpfendorfer et
al. 2001; Matich et al. 2011; Trystram et al. 2016). This scale-dependent pattern
combined with the high individual variability found in the behaviour of tiger sharks may
have been responsible for the low explanatory power of the environmental variables
tested to describe their utilisation patterns. In terrestrial systems many predators exhibit
preferential patterns of prey choice where they specialise on a specific type of prey that
inhabit particular habitats (Hayward et al. 2007, 2012; Lyngdoh et al. 2014). For
example, lions (Panthera leo) have been shown to actively and cognitively decide to
hunt preferred prey, increasing the chance of encounter by choosing to forage in habitats
with higher abundances of those prey (Hayward et al. 2011). In the marine
environment, although orcas (Orcinus orca) can be considered as a generalist species,
138
different populations of killer whales adapt their hunting and behavioural strategies to
target preferred prey in the habitat where they are most abundant (Hoelzel 1991, Saulitis
et al. 2000, Ford & Ellis 2006, Ford et al. 2010, and references within). In contrast, tiger
sharks do not display high levels of specialisation in diet and seem to shift prey
preference across their range, seasonally targeting areas with high abundances of prey
(Simpfendorfer et al. 2001; Lowe et al. 2006). These generalist predators will thus
require multiple foraging strategies to capture prey that will be pursued across multiple
habitats, resulting in scale and habitat-dependent space use.
Scale-dependent movements of marine predators are likely to be accompanied by local
adaptation of foraging strategies according to prey type and availability. Identifying
these behaviours is challenging (Austin et al. 2006), particularly when tracking data is
very patchy, as is typically the case for tiger sharks. Location estimates in tracks are
often too sparse to extract horizontal displacement metrics that can be used in
movement-based models for the identification of foraging behaviour (Fauchald &
Tveraa 2003; Jonsen et al. 2013, but see Fitzpatrick et al 2015). The use of GWR
provides an understanding of movement patterns from such data sets because it
identifies where and how the modelled relationships change across variable space and
the potential reason as to why this occurs. Here, it seems likely that tiger sharks respond
to environmental signals that occur at smaller spatial scales on the continental shelf than
the open ocean, thus indicating where fine-scale studies of the use of habitats should be
targeted. Given that sparse tracks and scale-dependent patterns of movement are a
frequent outcome of many studies of large marine predators (Bradshaw et al. 2004a;
Austin 2007; Bonfil et al. 2009; Hammerschlag et al. 2011a), the combined approach of
additive mixed models and GWR has the potential for widespread application to the
field of movement ecology.
5.6 Acknowledgements
Funding for this work was provided by Australian Institute of Marine Science, the
University of Western Australia, and the Conselho Nacional de Desenvolvimento
Científico e Tecnológico (CNPq, Brazil). The authors acknowledge Chris Fischer and
the crew of the MV OCEARCH for the aid with logistics and field work for the tracks
deployed at Ningaloo Reef in April-May 2015. The Save our Seas Foundation also
provided funding support for this work.
139
5.7 Supporting information
Table S 5.1 Details of each of the deployments of telemetry devices on tiger sharks.
*TL estimated from fork length using the equation by Stevens & McLoughlin (1991)
ID Transmitter Sex Total length (cm) Tag Type Deployment date Tagging location Region Duration (days)
T1 62346 F 180.75* SPOT 19/06/07 Ningaloo Reef Indian Ocean 7
T2 62343 F 330.27* SPLASH 21/06/07 Ningaloo Reef Indian Ocean 14
T3 83858 F 219.56* SPOT 17/08/08 Ningaloo Reef Indian Ocean 105
T4 83857 F 259.51* SPOT 19/08/08 Ningaloo Reef Indian Ocean 70
T5 83859 F 268.64* SPOT 19/08/08 Ningaloo Reef Indian Ocean 517
T6 93304 M 395.34* SPOT 27/05/10 Ningaloo Reef Indian Ocean 191
T7 93302 M 270.92* SPOT 02/06/10 Ningaloo Reef Indian Ocean 38
T8 93299 F 289.18* SPLASH 30/05/10 Ningaloo Reef Indian Ocean 154
T9 146703 F 303 SPOT 24/04/15 Ningaloo Reef Indian Ocean 117
T10 146710 F 363 SPOT 25/05/15 Ningaloo Reef Indian Ocean 251
T11 146711 F 407 SPOT 24/04/15 Ningaloo Reef Indian Ocean 205
T12 146714 F 319 SPOT 23/04/15 Ningaloo Reef Indian Ocean 57
T13 149046 F 390 SPOT 27/04/15 Ningaloo Reef Indian Ocean 267
T14 105171 F 309 SPOT 21/02/11 Shark Bay Indian Ocean 90
T15 105169 F 330 SPOT 27/02/11 Shark Bay Indian Ocean 224
T16 105173 F 305 SPOT 14/03/11 Shark Bay Indian Ocean 196
T17 105168 F 317 SPOT 14/03/11 Shark Bay Indian Ocean 292
T18 105174 F 320 SPOT 17/03/11 Shark Bay Indian Ocean 99
T19 105175 F 360 SPOT 17/03/11 Shark Bay Indian Ocean 24
T20 105167 F 306 SPOT 17/03/11 Shark Bay Indian Ocean 85
T21 105172 F 362 SPOT 15/04/11 Shark Bay Indian Ocean 482
T22 105170 F 307 SPOT 24/04/11 Shark Bay Indian Ocean 303
T23 54738 F 350 SPOT4 23/11/04 Great Barrier Reef Pacific Ocean 16
T24 62849 M 288 SPOT5 8/12/06 Great Barrier Reef Pacific Ocean 63
T25 54739 F 330 SPLASH 10/12/07 Great Barrier Reef Pacific Ocean 42
T26 62848 F 295 SPLASH 13/12/07 Great Barrier Reef Pacific Ocean 42
T27 79974 M 292 SPLASH 16/12/07 Great Barrier Reef Pacific Ocean 231
T28 79973 F 300 SPLASH 26/11/06 Great Barrier Reef Pacific Ocean 20
140
T29 79972 F 296 SPLASH 10/12/07 Great Barrier Reef Pacific Ocean 209
T30 79975 F 368 Telonics ST18 21/11/03 Great Barrier Reef Pacific Ocean 120
T31 72587 F 350 SPOT 3/12/06 Great Barrier Reef Pacific Ocean 43
T32 79984 F 311 SPOT 25/03/13 Great Barrier Reef Pacific Ocean 370
T33 79985 F 331 SPOT 25/03/13 Great Barrier Reef Pacific Ocean 110
T34 79979 F 308 SPOT 1/12/13 Great Barrier Reef Pacific Ocean 136
T35 111549 F 285 SPOT 10/04/11 Hawaii Pacific Ocean 21
T36 111547 F 405 SPOT 5/09/11 Hawaii Pacific Ocean 350
T37 111543 F 380 SPOT 8/09/11 Hawaii Pacific Ocean 141
T38 34107 F 256 SPOT 6/11/09 Florida North Atlantic Ocean 206
T39 34029 F 256 SPOT 25/05/10 U.S.A North Atlantic Ocean 191
T40 34020 M 262 SPOT 26/05/10 U.S.A North Atlantic Ocean 40
T41 34021 F 250 SPOT 26/05/10 U.S.A North Atlantic Ocean 23
T42 33992 F 201 SPOT 26/05/10 U.S.A North Atlantic Ocean 33
T43 55494 F 244 SPOT 26/05/10 U.S.A North Atlantic Ocean 95
T44 55495 F 295 SPOT 09/06/10 U.S.A North Atlantic Ocean 128
T45 68477 M 201 SPOT 10/06/10 U.S.A North Atlantic Ocean 127
T46 34203 F 335 SPOT 29/10/10 U.S.A North Atlantic Ocean 46
T47 68471 F 245 SPOT 13/11/10 U.S.A North Atlantic Ocean 26
T48 105594 F 375 SPOT 29/01/11 U.S.A North Atlantic Ocean 8
T49 68529 F 325 SPOT 19/02/11 U.S.A North Atlantic Ocean 556
T50 68554 F 403 SPOT 19/02/11 U.S.A North Atlantic Ocean 184
T51 68494 F 365 SPOT 19/02/11 U.S.A North Atlantic Ocean 191
T52 68485 F 335 SPOT 19/02/11 U.S.A North Atlantic Ocean 94
T53 105599 F 340 SPOT 19/02/11 U.S.A North Atlantic Ocean 94
T54 68555 F 286 SPOT 20/02/11 U.S.A North Atlantic Ocean 254
T55 68556 F 322 SPOT 20/02/11 U.S.A North Atlantic Ocean 418
T56 68495 F 325 SPOT 20/02/11 U.S.A North Atlantic Ocean 231
T57 68496 F 280 SPOT 20/02/11 U.S.A North Atlantic Ocean 216
T58 68488 F 295 SPOT 20/02/11 U.S.A North Atlantic Ocean 830
T59 68486 F 320 SPOT 20/02/11 U.S.A North Atlantic Ocean 99
T60 105595 F 325 SPOT 20/02/11 U.S.A North Atlantic Ocean 39
T61 106660 M 206 SPOT 22/02/11 U.S.A North Atlantic Ocean 52
T62 106661 F 320 SPOT 10/04/11 U.S.A North Atlantic Ocean 59
T63 98332 F 184 SPOT 20/09/11 U.S.A North Atlantic Ocean 85
141
T64 108064 F 175 SPOT 12/11/11 U.S.A North Atlantic Ocean 382
T65 112979 F 325 SPOT 9/12/11 U.S.A North Atlantic Ocean 2
T66 112988 F 305 SPOT 12/12/11 U.S.A North Atlantic Ocean 379
T67 112981 F 332 SPOT 14/12/11 U.S.A North Atlantic Ocean 7
T68 113537 F 346 SPOT 14/12/11 U.S.A North Atlantic Ocean 392
T69 113536 F 305 SPOT 14/12/11 U.S.A North Atlantic Ocean 145
T70 113534 F 321 SPOT 15/12/11 U.S.A North Atlantic Ocean 32
T71 115907 M 200 SPOT 16/12/11 U.S.A North Atlantic Ocean 32
T72 115906 M 200 SPOT 2/02/12 U.S.A North Atlantic Ocean 40
T73 112994 M 232 SPOT 27/05/12 U.S.A North Atlantic Ocean 1
T74 112993 F 232 SPOT 22/07/12 U.S.A North Atlantic Ocean 1
T75 112987 F 296 SPOT 22/07/12 U.S.A North Atlantic Ocean 153
T76 112986 F 248 uCricket 22/07/12 U.S.A North Atlantic Ocean 323
T77 115915 F 215 uCricket 23/07/12 U.S.A North Atlantic Ocean 31
T78 112991 F 260 uCricket 04/09/12 U.S.A North Atlantic Ocean 9
T79 115912 F 297 SPOT 06/09/12 U.S.A North Atlantic Ocean 104
T80 115913 F 157 SPOT 6/09/12 U.S.A North Atlantic Ocean 135
T81 112982 F 196 uCricket 8/09/12 U.S.A North Atlantic Ocean 43
T82 119440 F 206 SPOT 12/09/12 U.S.A North Atlantic Ocean 43
T83 111551 F 180 SPOT 18/03/13 U.S.A North Atlantic Ocean 29
T84 130582 F 220 SPOT 20/04/13 U.S.A North Atlantic Ocean 30
T85 130985 F 289 SPOT 24/05/13 U.S.A North Atlantic Ocean 291
T86 129952 F 360 SPOT 1/06/13 U.S.A North Atlantic Ocean 26
T87 130583 F 368 SPOT 05/08/13 U.S.A North Atlantic Ocean 25
T88 130986 F 273 SPOT 17/10/13 U.S.A North Atlantic Ocean 26
T89 130990 F 357 SPOT 17/10/13 U.S.A North Atlantic Ocean 147
T90 132795 M 356 SPOT 17/10/13 U.S.A North Atlantic Ocean 175
T91 133719 F 378 SPOT 17/10/13 U.S.A North Atlantic Ocean 10
T92 133720 F 307 SPOT 17/10/13 U.S.A North Atlantic Ocean 174
T93 133721 F 344 SPOT 17/10/13 U.S.A North Atlantic Ocean 131
T94 133722 F 373 SPOT 17/10/13 U.S.A North Atlantic Ocean 19
T95 133723 F 294 SPOT 17/10/13 U.S.A North Atlantic Ocean 170
T96 129957 M 230 SPOT 17/10/13 U.S.A North Atlantic Ocean 65
T97 137737 F 300 SPOT 13/11/13 U.S.A North Atlantic Ocean 9
T98 134992 F 199 SPOT 8/02/14 U.S.A North Atlantic Ocean 40
142
T99 134993 M 242 SPOT 21/03/14 U.S.A North Atlantic Ocean 41
T100 137335 F 366 SPOT 22/03/14 U.S.A North Atlantic Ocean 117
T101 137337 F 352 SPOT 12/05/14 U.S.A North Atlantic Ocean 186
T102 137336 F 383 SPOT 12/05/14 U.S.A North Atlantic Ocean 251
T103 137738 F 296 SPOT 13/05/14 U.S.A North Atlantic Ocean 26
T104 130580 F 300 SPOT 11/07/14 U.S.A North Atlantic Ocean 27
T105 146598 M 245 SPOT 14/11/14 U.S.A North Atlantic Ocean 162
T106 144397 F 270 SPOT 16/11/14 U.S.A North Atlantic Ocean 209
T107 144019 M 352 SPOT 15/03/15 U.S.A North Atlantic Ocean 102
T108 144020 F 356 SPOT 26/04/15 U.S.A North Atlantic Ocean 332
T109 73058 M 131 Mk10 28/06/08 Brazil South Atlantic Ocean 31
T110 78010 M 193 Mk10 24/07/09 Brazil South Atlantic Ocean 12
T111 90816 F 128 Mk10 01/06/10 Brazil South Atlantic Ocean 75
T112 78012 M 150 Mk10 07/08/10 Brazil South Atlantic Ocean 79
T113 97760 F 190 Mk10 21/12/10 Brazil South Atlantic Ocean 26
T114 78008 F 295 Mk10 06/01/11 Brazil South Atlantic Ocean 79
T115 98008 M 190 Mk10 08/02/11 Brazil South Atlantic Ocean 23
T116 78004 F 120 Mk10 05/03/11 Brazil South Atlantic Ocean 70
T117 97758 F 270 Mk10 22/03/11 Brazil South Atlantic Ocean 118
T118 78012a M 130 Mk10 11/07/11 Brazil South Atlantic Ocean 52
T119 98005 M 170 Mk10 11/07/11 Brazil South Atlantic Ocean 62
T120 97762 F 134 Mk10 25/07/11 Brazil South Atlantic Ocean 53
T121 98006 M 153 Mk10 14/08/11 Brazil South Atlantic Ocean 60
T122 98007 F 172 Mk10 23/08/11 Brazil South Atlantic Ocean 10
T123 112679 F 212 MiniPAT 01/08/12 Brazil South Atlantic Ocean 123
T124 112682 F 240 MiniPAT 01/08/12 Brazil South Atlantic Ocean 124
T125 112678 F 238 MiniPAT 02/08/12 Brazil South Atlantic Ocean 13
T126 112681 F 310 MiniPAT 12/08/12 Brazil South Atlantic Ocean 57
T127 52663 F 180 Mk10 24/08/12 Brazil South Atlantic Ocean 32
T128 112683 F 156 MiniPAT 09/09/12 Brazil South Atlantic Ocean 124
T129 122811 F 184 MiniPAT 31/08/13 Brazil South Atlantic Ocean 60
T130 78012b F 151 MiniPAT 31/08/213 Brazil South Atlantic Ocean 11
143
Figure S 5.1 Representative example of a tiger shark track (black points) analysed with Biased
Random Bridges showing both residency (points within the 0.75 UD (red lines) and migration
(points outside the 0.75 UD).
Figure S 5.2 Figure extracted from: A) (Bouchet 2015) indicating hotspots in fish abundance
off the coast of Western Australia, B)Figure re extracted from (Double et al. 2014) showing
areas of high occupancy by pygmy blue whales off the coast of Western Australia.
145
Chapter 6 General Discussion
An understanding of movement ecology is essential for the conservation and
management of mobile, top-order predators in marine systems. In this thesis I examined
how environment and diet influenced movement patterns and habitat use of tiger sharks
at multiple spatial and temporal scales. Firstly, I compiled available information on the
biology and ecology of tiger sharks and identified current threats to the species in order
to update their conservation status for the International Union for Conservation of
Nature (IUCN) Red List (Chapter 2). I then examined patterns of movement of the
species along the coast of Western Australia and showed that tiger sharks made
extensive movements between tropical and temperate ecosystems and exhibited high
individual variability in distances travelled and degree of residency within a habitat
(Chapter 3). Isotopic analyses (Chapter 4) showed that these large-scale movements
were accompanied by changes in diet, trophic niche and role across multiple habitats at
a continental scale. Finally, I analysed a large dataset of satellite tracking from locations
in the Indian, Pacific and Atlantic Ocean basins and identified bathymetry and
temperature as key drivers of movements and habitat use by tiger sharks at global scales
(Chapter 5). The relationship between these physical variables and movement patterns
varied among ocean basins and had low predictive power to explain patterns of
utilisation of habitats due to their variability in space (spatial non-stationarity). A
common thread across all my chapters was the high degree of variation in behaviour
that occurred among individual sharks, likely as a consequence of their role as a
generalist top-order predator in marine ecosystems.
6.1 Variability in movement behaviour
Why are movement patterns of these large sharks so variable? One key reason might be
that such species are not tied to any fixed point in the marine habitat during their
ontogeny, such as discrete nursery areas that are commonly used by several sharks
(Heupel et al. 2007) or philopatry to specific nesting beaches by sea turtles (Brothers &
Lohmann 2015). For example, taxa that have portions of their life cycle associated with
land, either for breeding or moulting, such as most pinnipeds, marine reptiles and
seabirds, typically show a high degree of predictability in movement patterns and
habitat use (Shaffer et al. 2006; Broderick et al. 2007; Egevang et al. 2010; Fitzpatrick
et al. 2012; Raymond et al. 2015) because these behaviours are associated with the
146
regular return to natal beaches, islands and nesting grounds in order to breed. Southern
elephant seals (Mirounga leonina), for instance, exhibit site fidelity to breeding colonies
and foraging grounds, with many individuals returning to the same haul-out site after
extended foraging trips (Jonker & Bester 1998; Bradshaw et al. 2004b; Fabiani et al.
2006; Mulaudzi et al. 2008). This highly predictable pattern of movement has allowed
key environmental and habitat requirements to be identified throughout the species
range (Biuw et al. 2007; Hindell et al. 2016).
In contrast, tiger sharks and other top-order predatory fishes including mako (Isurus
sp.), thresher (Alopias sp.), billfishes and tunas rarely display such predictable
behaviours. This is probably a reflection of the nomadic nature of their ecology with no
easily identifiable specific start or end point to their migrations, combined with a lack of
association with static points in the environment such as land masses, beaches or ice
caps where breeding colonies occur. As a consequence, these animals display a high
degree of individual variability in movement patterns and space use (Boustany et al.
2002; Walli et al. 2009; Carlisle et al. 2012) and a plasticity in behaviour and flexibility
in habitat associations that hampers our ability to identify the drivers of migration and
residency.
6.2 Drivers of movement and habitat use
The Biased Random Bridges (BRB) analytical technique provided a means to analyse
and visualise tracking data for tiger sharks, while incorporating inherent autocorrelation
in the computation of utilisation distributions and addressing gaps in the dataset
(Bullard 1991; Benhamou 2011; Benhamou & Riotte-Lambert 2012). The description of
utilisation distributions allowed me to assess if residency was potentially associated
with foraging (Chapter 3) and to identify the environmental drivers of patterns of high
utilisation of some habitats (Chapter 5). In Chapter 3, I found that residency at regional
spatial scales was associated with bathymetry and sea surface temperatures (SST), and
was more likely to occur in shallow coastal waters with temperatures above 23°C. The
same environmental variables drove patterns of residency of tiger sharks in other ocean
basins (Chapter 5). Although the distribution and habitat use of a wide range of fauna
has been associated with topographic covariates (McConnell et al. 1992; Woodley &
Gaskin 1996; Sjöberg & Ball 2000; Bouchet et al. 2015), tiger sharks are an epipelagic
species that spend most of their time in the uppermost surface layer of the water
column, above the thermocline (Meyer et al. 2010; Fitzpatrick et al. 2012; Hazin et al.
147
2013; Vaudo et al. 2014; Afonso & Hazin 2015). In my study, bathymetry acted as a
surrogate measure of the use of oceanic and shelf habitats for the species (Chapter 3 and
5). As ectotherms, elasmobranchs are strongly influenced by changes in water
temperature (Crawshaw 1977; Bernal et al. 2012), implying that these animals are likely
to be constantly searching for optimal environmental conditions (Klimley & Butler
1988; Block et al. 1997, 2011; Bestley et al. 2013; Kessel et al. 2014; Schlaff et al.
2014). Despite the fact that water temperatures was a predictor of migration, it seems
unlikely that this behaviour was a search for optimal conditions since tiger sharks
selected temperatures during migrations to both tropical and temperate regions that were
outside the thermal range they preferred while resident (Chapters 3 and 5).
6.3 Low predictability of models
Although SST and bathymetry were selected as major predictors of migration, residency
and areas of high use, these explained only a small proportion of the variance in
movement behaviour (Chapters 3 and 5). In Chapter 3, this result might have been an
outcome of relatively low sample size. However, in Chapter 5, I gathered and analysed
the largest global dataset of tiger shark tracks available to date. Despite the large
number of tracks, generalised additive mixed models still explained only small amounts
of variance in movement and had low power to predict the utilisation distributions (UD)
of tiger sharks. To understand why this was the case, I then applied geographically
weighted regression (GWR) to determine if patterns of utilisation were changing in
space and causing non-stationarity in the relationship between model predictions and
observed values. The GWR calculated localised coefficients for a regression by
allowing the relationship between variables to vary in space and subsequently mapped
the range of coefficients in a spatial context (Brunsdon et al. 1996; Fotheringham et al.
1998). This allowed me to determine the presence of non-stationarity in the relationship
and to identify areas where observed values of UD where much higher or lower than
predicted by mixed models (Chapter 5). When non-stationarity is detected, there are a
number of possible explanations including, model misspecification, missing variables,
autocorrelation or inappropriate spatial scales of measurement (resolution) (Brunsdon et
al. 1996; Fotheringham et al. 1998; Foody 2004; Osborne et al. 2007). Autocorrelation
was incorporated in the computation of utilisation distributions (Benhamou 2011) and
so was not an issue here, and the consistent selection of the same predictors (depth and
SST) from the suite of fitted models suggested that principal environmental drivers of
movement had been identified. There were obvious issues with resolution of data in
148
tracks from the South Atlantic where geolocations were associated with large errors
(Lam et al. 2010) that most likely drove non-stationarity. However, in the Atlantic and
Indian oceans, where non-stationarity was also identified, (as evidenced by a large range
in estimates of GWR coefficients), location estimates were obtained from satellite
telemetry and had less error. For these regions, areas where values of utilisations were
higher than predicted by the models seemed to be related to two main issues: missing
covariates in the model (such as food availability) and changes in the scale of movement
behaviour by tiger sharks between oceanic and coastal habitats.
In the Indian Ocean, off the coast of Western Australia, there were areas that had greater
than predicted utilisation giving highly positive values of GWR estimates. These areas
overlapped with hotspots for abundances of pelagic fish and occupancy by pygmy blue
whales (Balaenoptera musculus brevicauda) (Double et al. 2014; Bouchet 2015),
suggesting that prey availability may be accounting for some of the variation not
explained by the model and the resulting non-stationarity. In the North Atlantic, patterns
of higher than predicted UD values were found in areas with high average utilisation by
all sharks off the eastern coast of the USA. This suggests that our models properly
explained high use of oceanic areas, but failed to predict use of coastal habitats. In the
open ocean, the influence of large and mesoscale scale oceanographic features such as
large patterns of SST fields and fronts (Ullman & Cornillon 1999; Fryxell et al. 2008;
Belkin et al. 2009) can be identified given the resolution of the environmental data
provided by remote sensing. Here, mixed models identified a preference for deeper and
colder waters, probably associated with high use of areas around the Gulf Stream as
seen in previous studies (Lea et al. 2015; Queiroz et al. 2016). The association of tiger
sharks and thermal fronts in the North Atlantic (Queiroz et al. 2016) suggests that
mesoscale features such as eddies and coastal frontals may be important habitats for the
species, and could explain some of the patterns found here. Whereas, complex shelf
habitats influenced by topography and interactions with coastal oceanographic
processes would likely have higher explanatory power for areas of high use near the
coast if a finer-resolution spatial scale had been used in the analysis. My results suggest
that non-stationarity was likely to be a result of scale-dependent patterns in
environmental signals and the behaviour of tiger sharks, causing the models to have low
predictive power to describe utilisation patterns in shelf environments. The
environmental signals created by mesoscale oceanic features and coastal features will be
difficult to describe using remotely sensed data ad with a global dataset, however in-situ
149
environmental data such as provided by CTD-SRL tags combined with high resolution
tracking data (such as provided by Fastloc GPS tags and Diary tags) would likely be
necessary to define and explain those scale-dependent associations.
Tiger sharks are generalist predators that explore multiple food-webs (Chapter 4), likely
with foraging strategies adapted to specific prey in different habitats, a trait that may
drive scale- and habitat-dependent space use. Identifying scale-dependent movements of
top-predators can be challenging (Austin et al. 2006), especially when tracking data has
low resolution (Bradshaw et al. 2007), as is typically the case for tiger sharks. Although
my findings show that large scale patterns of SST and bathymetry can explain use of
oceanic areas, the application of GWR has allowed me to identify, for the first time that
the relationship between space utilisation by tiger sharks and environment operates at
finer scale in shelf than in oceanic habitats. The limited predictability of environmental
preferences (Chapter 5) and highly generalist behaviour (Chapter 4) currently preclude
our understanding of the potential effects of large-scale climate change in the
distribution and resiliency of tiger sharks. The prediction of movements and habitat use
for the species will likely be only possible in habitat-dependent contexts (e.g. use of
specific features or habitats in coastal and oceanic environments), when considering
different sections of the populations separately, i.e. movement of females to potential
mating and pupping grounds (Papastamatiou et al. 2013) for example. Spatial patterns
in trophic ecology
Variability in movement patterns of tiger sharks was paralleled by variability in diet.
Comparisons of isotopic signatures of sharks from locations on both the west and east
coasts of Australia suggested that tiger sharks have a diet that is size, context and
habitat specific. Signatures from muscle tissue showed that larger sharks fed on a diet
that had high values of δ13
C, consistent with these animals targeting large herbivores
such as dugong (Dugong dugon) and turtles that are part of a seagrass food chain.
Smaller tiger sharks had lower values of δ13
C, probably reflecting the more limited
ability of these size classes to subdue these large herbivores and thus resulting in diets
that have a larger component of smaller species such as fishes. These patterns were
consistent with previous studies of stomach contents (Simpfendorfer 1992; Lowe et al.
1996; Heithaus 2001).
Diets of tiger sharks also varied by location. The isotopic signatures of sharks sampled
on the Great Barrier Reef (GBR) and in Shark Bay were characteristic of a diet based on
150
a seagrass food chain, whereas sharks sampled on the coast of New South Wales (NSW)
had signatures indicative of a plankton-based food chain. Signatures of sharks sampled
at Ningaloo Reef were characterised by a diet based both on planktonic and benthic
food-webs. Despite the large differences in communities and habitats between Shark
Bay (Walker et al. 1988; Heithaus 2004; Vaudo & Heithaus 2011; Heithaus et al. 2013)
and on the GBR (Wachenfeld et al. 1998; Carruthers et al. 2002; Espinoza et al. 2014),
tiger sharks still remained at the top of the food web in both locations, showing that in
these tropical ecosystems, they acted as top-order predators (Heithaus et al. 2008a;
Wirsing & Ripple 2011). This did not appear to be the case off the temperate coast of
NSW (32° - 35°S) where seals, other oceanic sharks and tunas rather than tiger sharks
were located in the top-order positons of the food web (Davenport & Bax 2002; Revill
et al. 2009) and may have a larger role in structuring communities in this habitat.
The shift in isotopic signatures of tiger sharks between these tropical and temperate
environments was not solely due to latitude, it also appeared to be driven by the
oceanographic setting of the environment where these sharks occur. The signatures of
sharks sampled in offshore environments of NSW were very similar to those of sharks
sampled on the isolated reefs of Reunion Island in the tropical western Indian Ocean
(Trystram et al. 2016). Both localities are exposed to the open ocean and these
signatures probably reflect the accessibility of planktonic food chains. Similarly, at
Ningaloo Reef the continental shelf is very narrow, so that boundary currents occur
within 5 - 10 km of the coast accounting for the use of mixed food chains revealed by
signatures.
These results emphasise the flexibility of the diet and trophic role of tiger sharks.
Whether such traits are a cause or a consequence of their wide-ranging movements is
difficult to determine. These sharks are not simply predators, as opportunistic
scavenging is also an important feeding behaviour (Gallagher et al. 2011; Bornatowski
et al. 2012a,b; Clua et al. 2013; Hammerschlag et al. 2016) and may make an important
contribution to energy intake. In some localities, such as at Raine Island, where dead or
dying sea turtles make a significant part of seasonal diets (Hammerschlag et al. 2016), it
will be very difficult to distinguish between active predation and scavenging, since
these will give an equivalent isotopic signal.
Although there is no evidence that tiger sharks display the hunting patterns of other top-
order marine predators, such as the stalking strategy of white sharks (Klimley 1994;
151
Klimley et al. 2001b; Huveneers et al. 2015) or the cooperative group hunting
behaviour of killer whales (Orcinus orca) (Hoelzel 1991; Saulitis et al. 2000) and their
isotopic signatures place them at the lowest trophic position of assemblages of large
sharks in some coastal environments (e.g. Hussey et al. 2015b), few species are likely to
represent a predation threat for large tiger sharks (Heupel et al. 2014), with the possible
exception of white sharks (Carcharodon carcharias) and killer whales (Pyle et al.
1999).
6.4 Implications for conservation
Biological characteristics such as high growth rates, relatively low fishing mortality
(Chapter 2) combined with the high variability in movement and space use (Chapter
3,5) and flexibility in feeding preferences and trophic role (Chapter 4), suggests that
tiger sharks could potentially display greater resilience to fishing pressure and human-
induced changes in habitats, including climate change (Hazen et al. 2013), compared to
other large marine predators. However, a key issue hampering the quantification of
resilience to threats and evaluation of the conservation status of tiger sharks is a lack of
data on population structure and dynamics, and catch trends of fisheries for the species
at both regional and global scales (Chapter 2). This combination of knowledge gaps and
somewhat constraining reproductive traits (i.e. producing many pups but having a
triennial reproductive cycle) is reflected in the species classification as “Near
Threatened” in the IUCN Red List. Recently, a global analysis of tiger shark genetics
showed significant genetic separation between sharks from the western Atlantic and
Indo-Pacific, with limited gene flow and different population structures among different
ocean basins (Bernard et al. 2016). Although insufficient to be designated as sub-
species (Bernard et al. 2016), these distinct genetic and population structures suggests
that tiger sharks from different ocean basins should be assessed separately by the IUCN.
New regional assessments could facilitate the identification of threats and population
trends, and possibly result in different, but more appropriate, classifications of
conservation status.
Mounting evidence of the key role of top predators in regulating ecosystem health
(Heithaus et al. 2008a; Beschta & Ripple 2009; Estes et al. 2011; Ruppert et al. 2013;
Atwood et al. 2015), combined with the severe impacts of human activities to their
populations (Dulvy et al. 2008, 2014; Schipper et al. 2008; McCauley et al. 2015;
Payne et al. 2016) highlight the importance of understanding how predators relate to the
152
environment if we are to create appropriate conservation measures to remediate and halt
the consequences of their removal from ecosystems worldwide. Tiger sharks are top-
order predators in tropical and seagrass habitats (Chapter 4) where they have been
shown to regulate grazing on seagrass by communities of large herbivores (Burkholder
et al. 2013). Their removal could result in a reduction of the control of grazing by
herbivores, which in turn could lead to declines in seagrass productivity and eventually
result in the collapse of this habitat (Heithaus et al. 2014; Atwood et al. 2015). Marine
reserves in the Great Barrier Reef, Ningaloo Reef and Shark Bay in Australia can
provide limited spatial protection for mobile species (Preen et al. 1997; McCook et al.
2010; Speed et al. 2010; Heupel et al. 2015). However, due to the very wide-ranging
nature of tiger sharks (Chapter 3 and 5) such reserves will not provide protection for
individuals throughout their life history. The tracks shown in Chapter 5 show that some
sharks routinely moved across 1000s of km of open ocean and coastline and across the
territorial waters of different countries, each with different protections, fishery
regulations and conservation strategies. Protection for these animals can even differ
within the territorial waters of the same country. Thus, it may make sense to aim to
identify and focus conservation strategies on those key areas that serve as foraging or
breeding habitat and to unify and strengthen management approaches to fisheries that
capture these sharks on a regional basis. In the case of tiger sharks in Western Australia,
large marine parks such as Ningaloo Reef and Shark Bay seem to offer sufficient
protection for the species despite the varied classes of zoning present, and modification
of zoning within these reserves would unlikely change the level of protection offered for
the species. However, in the North Atlantic areas of high use by large sharks overlap
with highly exploited fishing grounds in the North Atlantic Current – Labrador Current
convergence zone, and the Mid-Atlantic Ridge, indicating the necessity of ocean-scale
management of fisheries (Queiroz et al. 2016). Therefore, the identification of
environmental and habitat preferences for tiger sharks, and other large migratory sharks,
in areas where no protection currently exists will be a major step for the development of
alternative management actions (fisheries regulation, catch quotas, dynamic ocean
management) where the development of large-scale marine reserves might not be
feasible. Additional tagging effort and larger spread across the population (i.e. also
including juveniles and males) that allow the identification of movements related to
reproduction and habitats associated with pupping and juvenile grounds, combined with
the use of analytical approaches such as geographically weighted regression could aid
this goal.
153
6.5 Future directions
The need for more tracking data to better describe movement patterns derived from all
components of the population of sharks was identified as a key goal for conservation
strategies. Poor monitoring of catches in commercial and artisanal fisheries that prevent
stock assessments and definition of long-term population trends have been identified as
one of the major issues regarding management and conservation of tiger sharks (Chapter
2). Management strategies that integrate near-real-time tracking and predictive
modelling of habitat and environmental preferences are able to account for the dynamic
nature of species and oceanographic features (Hobday & Hartmann 2006; Hobday et al.
2009; Hobday & Hartog 2014; Dunn et al. 2016; Hazen et al. 2016) and offer great
potential for management of highly mobile marine predators.
To date, nearly all studies of tiger sharks have focused on horizontal movements and
environmental variables obtained by satellite remote sensing techniques from the
superficial layers of the water column (Chapter 3 and 5). However, tiger sharks occupy
a three dimensional environment where movements also occur in the vertical plane
often to depths well beyond the thermocline (Nakamura et al. 2011; Iosilevskii et al.
2012; Chapter 3). I have presented evidence of differential utilisation of the water
column between tropical and temperate environments (Chapter 3) and that diet and
trophic role of tiger sharks also change between those ecosystems (Chapter 4), which
reinforces the need for additional efforts to deploy tags that can be used to link
horizontal and vertical movements of individuals across multiple ecosystems. Many
predators utilise water column and vertical movements in their hunting strategy
(Klimley 1994) or forage while diving (Gallon et al. 2013). Defining the link between
horizontal and vertical displacements of tiger sharks, how they relate to and vary
according to environmental conditions, could resolve much of the lack of predictability
and ‘noise’ associated with ecological models applied to tracking data.
Studies that provide long-term and high-resolution tracking of tiger sharks are rare,
likely due to tag failure, shedding, damage and biofouling (Kerstetter et al. 2004;
Heithaus et al. 2007c; Hays et al. 2007; Meyer et al. 2010), which results in patchy and
sparse position estimates and short track durations. Some of the tracks presented in this
thesis were considerably longer than many previously obtained with satellite tags for the
species (Heithaus et al. 2007c; Fitzpatrick et al. 2012; Hazin et al. 2013; Papastamatiou
et al. 2013) and were able to describe extensive latitudinal movements (Chapter 3) and a
154
larger utilisation of areas beyond the shelf break than previously recorded for the
species (Chapter 5). However, long, high quality tracks represented only a small number
of the tracks gathered in this thesis. For non-air-breathing marine taxa the rate of uplink
of satellite tags and the use of cutting-edge technology is somewhat limited (Hussey et
al. 2015a; Hays et al. 2016). For exemple, CTD-SRL tags (Boehme et al. 2009) have
provided great advances in the understanding of habitat use of marine mammals (Biuw
et al. 2007) and could aid in overcoming some of the issues with data resolution and
uplink rates of satellite tracking of sharks. However, a towed version that could be
deployed on sharks is still not available. Studies using this tag could accelerate our
understanding of relationships between locomotion, foraging and the physical
environment inhabited by tiger sharks.
My results showed that predicting drivers of movement of tiger sharks is challenging,
most likely because sharks are possibly targeting diverse resources and habitats in
different ecosystems (Chapter 3 - 4). At fine spatial scales, the deployment of multi-
sensor and camera tags such as Customized Animal Tracking devices (CATS-Cam tags)
across multiple habitats would allow the identification of inter-specific interactions and
feeding events (Heithaus et al. 2002; Nakamura et al. 2011), and illuminate the decision
process behind feeding behaviour and prey selection, while linking movement patterns
to the role of sharks within specific habitats. At large spatial scales, GWR could be used
to test predictions associated with habitat use and to identify areas where general
relationships defined by physical parameters fail to explain patterns of use (Chapter 5).
The overlap found between patterns of GWR estimates and areas of high utilisation by
multiple species aids the targeting of future research efforts in order to identify factors
structuring habitat use by tiger sharks. New satellite technologies and the expansion of
the tracking data to other sections of the population, such as juveniles and males, would
allow the identification ecologically important habitats yet to be defined that warrant
special attention for conservation. Tagging programs could also potentially provide the
real-time tacking of animals that could feed into dynamic ocean management actions.
The analytical methods utilised in this thesis showed great potential applicability to
tracking data from a variety of taxa of marine megafauna, particularly when predictive
and explanatory power of models are low. Similarly, the technological development of
tracking systems and a better understanding of variability in species’ behaviours have
also been flagged as a major necessity for directing future research in marine megafauna
movement (Hays et al. 2016).
155
6.6 Concluding remarks
This thesis demonstrated how increasing sample numbers and the use of large datasets
can provide new insights into the ecology of a cosmopolitan marine predator. My
results revealed that tiger sharks use oceanic habits more often than previously reported
for the species, with both movement and residency being influenced by sea surface
temperature and bathymetry. Large-scale movements displayed by tiger sharks imply
that they act as an important link between different habitats, which is also supported by
their context specific trophic ecology. My results also showed how the combination of
data obtained from tracking and stable isotopes analysis of multiple tissues can be used
for defining large scale habitat use of top-order predators, despite some clear limitations
regarding data resolution and temporal scale of sampling. Although ecological
modelling of sparse tracking datasets and coarse resolution environmental data are
problematic, methods to generate utilisation distributions that can overcome these issues
showed great value and possible broad application for tracking of other large marine
taxa. Moreover, I provided a novel method to test explanatory power of models, identify
when and where model predictions inadequately explain spatial patterns, and guide
future efforts in areas where the compilation of high resolution and multispecies
datasets should be targeted. This is essential for understanding the spatial ecology and
conservation of mobile top-order marine predators such as tiger sharks.
157
Bibliography
Abecasis, D., Afonso, P. & Erzini, K. (2014). Combining multispecies home range and
distribution models aids assessment of MPA effectiveness. Marine Ecology
Progress Series, 513, 155–169.
Afonso, A.S. & Hazin, F.H.V. (2014). Post-release survival and behavior and exposure
to fisheries in juvenile tiger sharks, Galeocerdo cuvier, from the South Atlantic.
Journal of Experimental Marine Biology and Ecology, 454, 55–62.
Afonso, A.S. & Hazin, F.H. V. (2015). Vertical movement patterns and ontogenetic
niche expansion in the tiger shark , Galeocerdo cuvier. PLoS ONE, 10, 1–26.
Afonso, A.S., Hazin, F.H. V., Barreto, R.R., Santana, F.M. & Lessa, R.P. (2012).
Extraordinary growth in tiger sharks Galeocerdo cuvier from the South Atlantic
Ocean. Journal of fish biology, 81, 2080–5.
Ainley, D.G., Ballard, G., Karl, B.J. & Dugger, K.M. (2005). Leopard seal predation
rates at penguin colonies of different size. Antarctic Science, 17, 335.
Alerstam, T., Hedenstro, A. & Susanne, A. (2003). Long-distance migration: evolution
and determinants. OIKOS, 103, 247–260.
Anderson, D.R. & Burnham, K.P. (2002). Avoiding pitfalls when using information-
theoretic methods. Journal of Wildlife Management, 66, 912–918.
Arthur, K.E., Boyle, M.C. & Limpus, C.J. (2008). Ontogenetic changes in diet and
habitat use in green sea turtle (Chelonia mydas) life history. Marine Ecology
Progress Series, 362, 303–311.
Atwood, T.B., Connolly, R.M., Ritchie, E.G., Lovelock, C.E., Heithaus, M.R., Hays,
G.C., Fourqurean, J.W. & Macreadie, P.I. (2015). Predators help protect carbon
stocks in blue carbon ecosystems. Nature Climate Change, 5, 1038–1045.
Austin, M. (2007). Species distribution models and ecological theory: A critical
assessment and some possible new approaches. Ecological Modelling, 200, 1–19.
Austin, D., Bowen, W.D., McMillan, J.I. & Iverson, S.J. (2006). Linking movement ,
diving , and habitat to foraging success in a large marine predator. Ecology, 87,
3095–3108.
Avgar, T., Mosser, A., Brown, G.S. & Fryxell, J.M. (2013). Environmental and
individual drivers of animal movement patterns across a wide geographical
gradient. Journal of Animal Ecology, 82, 96–106.
Bailey, H. & Thompson, P.M. (2009). Using marine mammal habitat modelling to
identify priority conservation zones within a marine protected area. Marine
Ecology Progress Series, 378, 279–287.
Barton, K. (2013). Package ‘MuMIn’: Multi-model inference. R package version 1.15.6.
Bartumeus, F., Da Luz, M.G.E., Viswanathan, G.M. & Catalan, J. (2005). Animal
search strategies: a quantitative random walk analysis. Ecology, 86, 3078–
3087.
Baum, J.K., Myers, R.A., Kehler, D.G., Worm, B., Harley, S.J. & Doherty, P. A.
(2003). Collapse and conservation of shark populations in the Northwest Atlantic.
Science, 299, 389–392.
Bearhop, S., Adams, C.E., Waldron, S., Fuller, R.A. & Macleod, H. (2004).
158
Determining trophic niche width: a novel approach using stable isotope analysis.
Journal of Animal Ecology, 73, 1007–1012.
Beck, C. A., Iverson, S.J., Bowen, W.D. & Blanchard, W. (2007). Sex differences in
grey seal diet reflect seasonal variation in foraging behaviour and reproductive
expenditure: evidence from quantitative fatty acid signature analysis. The Journal
of animal ecology, 76, 490–502.
Belkin, I.M., Cornillon, P.C. & Sherman, K. (2009). Fronts in large marine ecosystems.
Progress in Oceanography, 81, 223–236.
Benhamou, S. (2007). How many animals really do the Lévy walk? Ecology, 88(8),
1962–1969.
Benhamou, S., & Collet, J. (2015). Ultimate failure of the Lévy foraging hypothesis :
Two-scale searching strategies outperform scale-free ones even when prey are
scarce. Journal of Theoretical Biology, 387, 221–227.
Benhamou, S. (2011). Dynamic approach to space and habitat use based on biased
random bridges. PloS one, 6, e14592.
Benhamou, S. & Riotte-Lambert, L. (2012). Beyond the utilization distribution:
Identifying home range areas that are intensively exploited or repeatedly visited.
Ecological Modelling, 227, 112–116.
Berger, J. (2004). The last mile: How to sustain long-distance migration in mammals.
Conservation Biology, 18, 320–331.
Berger, J., Young, J.K. & Berger, K.M. (2008). Protecting migration corridors:
Challenges and optimism for Mongolian saiga. PLoS Biology, 6, 1365–1367.
Berkeley, S. & Campos, W. (1988). Relative abundance and fishery potential of pelagic
sharks along Florida’s east coast. Marine Fisheries Review, 50, 9–16.
Bernal, D., Carlson, J., Goldman, K. & Lowe, C. (2012). Energetics, metabolism, and
endothermy in sharks and rays. Biology of sharks and their relatives (eds J.
Carrier, J. Musick & M. Heithaus), pp. 211–237. CRC Press, Boca Raton.
Bernard, A.M., Feldheim, K.A., Heithaus, M.R., Wintner, S.P., Wetherbee, B.M. &
Shivji, M.S. (2016). Global population genetic dynamics of a highly migratory,
apex predator shark. Molecular Ecology.
Beschta, R.L. & Ripple, W.J. (2009). Large predators and trophic cascades in terrestrial
ecosystems of the western United States. Biological Conservation, 142, 2401–
2414.
Bessey, C. & Vanderklift, M.A. (2014). Drying method has no substantial effect
on δ15
N or δ13
C values of muscle tissue from teleost fishes. Rapid
Communications in Mass Spectrometry, 28, 265–273.
Bessudo, S., Soler, G.A., Klimley, P.A., Ketchum, J. & Arauz, R. (2011). Vertical and
horizontal movements of the scalloped hammerhead shark (Sphyrna lewini) around
Malpelo and Cocos islands (tropical eastern Pacific) using satellite telemetry.
Boletin investigaciones marinas costeras, 40, 91–106.
Bestley, S., Jonsen, I.D., Hindell, M. A., Guinet, C. & Charrassin, J.B. (2013).
Integrative modelling of animal movement: incorporating in situ habitat and
behavioural information for a migratory marine predator. Proceedings of The
Royal Society B: Biological sciences, 280, 2012–2262.
Bineesh, K.K., Akhilesh, K. V, Abdussamad, E.M. & Prakasan, D. (2014). Seamount
associated fishery of south-west coast of India - a preliminary assessment. Indian
159
Journal of Fisheries, 61, 29–34.
Biuw, M., Boehme, L., Guinet, C., Hindell, M., Costa, D., Charrassin, J.B., Roquet, F.,
Bailleul, F., Meredith, M., Thorpe, S., Tremblay, Y., McDonald, B., Park, Y.-H.,
Rintoul, S.R., Bindoff, N., Goebel, M., Crocker, D., Lovell, P., Nicholson, J.,
Monks, F. & Fedak, M. A. (2007). Variations in behavior and condition of a
Southern Ocean top predator in relation to in situ oceanographic conditions.
Proceedings of the National Academy of Sciences of the United States of America,
104, 13705–10.
Blackhart, K., Stanton, D.G. & Shimada, A.M. (2006). NOAA Fisheries Glossary:
NOAA technical memorandum NMFS-F/SPO-69, Office of Science and
Technology. URL: http://www.st.nmfs.gov/st4/documents/fishglossary.pdf.
Block, B.A., Jonsen, I.D., Jorgensen, S.J., Winship, A. J., Shaffer, S. A., Bograd, S.J.,
Hazen, E.L., Foley, D.G., Breed, G. A., Harrison, A.L. A.L., Ganong, J.E.,
Swithenbank, A., Castleton, M., Dewar, H., Mate, B.R., Shillinger, G.L., Schaefer,
K.M., Benson, S.R., Weise, M.J., Henry, R.W., Costa, D.P., A.L. Harrison,
Ganong, J.E., Swithenbank, A., Castleton, M., Dewar, H., Mate, B.R., Shillinger,
G.L., Schaefer, K.M., Benson, S.R., Weise, M.J., Henry, R.W. & Costa, D.P.
(2011). Tracking apex marine predator movements in a dynamic ocean. Nature,
475, 86–90.
Block, B.A., Keen, J.E., Castillo, B., Dewar, H., Freund, E. V., Marcinek, D.J., Brill,
R.W. & Farwell, C. (1997). Environmental preferences of yellowfin tuna (Thunnus
albacares) at the northern extent of its range. Marine Biology, 130, 119–132.
Block, B., Teo, S., Walli, A., Boustany, A., Stokesbury, M.J.W., Farwell, C., Weng, K.,
Dewar, H. & Williams, T. (2005). Electronic tagging and population structure of
Atlantic bluefin tuna. Nature, 434, 109–112.
Boehme, L., Lovell, P., Biuw, M., Roquet, F., Nicholson, J., Thorpe, S.E., Meredith,
M.P. & Fedak, M. (2009). Technical note: Animal-borne CTD-satellite relay data
loggers for real-time oceanographic data collection. Ocean Science, 5, 685–695.
Bolnick, D.I., Yang, L.H., Fordyce, J.A., Davis, J.M. & Svanbäck, R. (2002).
Measuring individual-level resource specialization. Ecology, 83, 2936–2941.
Bonfil, R. (1994). Overview of world elasmobranch fisheries. FAO Fisheries Technical
Paper, 341. Rome. 119 pp.
Bonfil, R. (1997). Status of shark resources in the southern Gulf of Mexico and
Caribbean: implications for management. Fisheries Research, 29, 101–117.
Bonfil, R., Johnson, R., Brien, S.O., Oosthuizen, H., Swanson, S., Kotze, D. &
Paterson, M. (2005). Transoceanic migration , spatial dynamics , and population
linkages of white sharks. 100, 100–103.
Bonfil, R., Francis, M.P., Duffy, C., Manning, M.J. & O’Brien, S. (2010). Large-scale
tropical movements and diving behavior of white sharks Carcharodon carcharias
tagged off New Zealand. Aquatic Biology, 8, 115–123.
Bornatowski, H., Braga, R.R. & Vitule, J.R.S. (2014). Threats to sharks in a developing
country: The need for effective and simple conservation measures. Natureza a
Conservação, 12, 11–18.
Bornatowski, H., Heithaus, M.R., Batista, C.M.P. & Mascarenhas, R. (2012a). Shark
scavenging and predation on sea turtles in northeastern Brazil. Amphibia-Reptilia,
33, 495–502.
160
Bornatowski, H., Wedekin, L.L., Heithaus, M.R., Marcondes, M.C.C. & Rossi-Santos,
M.R. (2012b). Shark scavenging and predation on cetaceans at Abrolhos Bank,
eastern Brazil. Journal of the Marine Biological Association of the United
Kingdom, 92, 1767–1772.
Borrell, A., Aguilar, A., Gazo, M., Kumarran, R.P. & Cardona, L. (2011). Stable
isotope profiles in whale shark (Rhincodon typus) suggest segregation and
dissimilarities in the diet depending on sex and size. Environmental Biology of
Fishes, 92, 559–567.
Bouchet, P.J. (2015). Submarine topography as a predictor of mobile predator hotspots:
Relevance and applications to conservation planning in the pelagic ocean. PhD
Thesis. University of Western Australia,Perth.
Bouchet, P.J., Meeuwig, J.J., Salgado Kent, C.P., Letessier, T.B. & Jenner, C.K. (2015).
Topographic determinants of mobile vertebrate predator hotspots: Current
knowledge and future directions. Biological Reviews, 90, 699–728.
Boustany, A.M., Davis, S.F., Pyle, P., Anderson, S.D., Le Boeuf, B.J. & Block, B.A.
(2002). Expanded niche for white sharks. Nature, 415, 35–36.
Bradshaw, C.J.A., Higgins, J., Michael, K.J., Wotherspoon, S.J. & Hindell, M.A.
(2004a). At-sea distribution of female southern elephant seals relative to variation
in ocean surface properties. ICES Journal of Marine Science, 61, 1014–1027.
Bradshaw, C.J.A., Hindell, M.A., Sumner, M.D. & Michael, K.J. (2004b). Loyalty pays:
Potential life history consequences of fidelity to marine foraging regions by
southern elephant seals. Animal Behaviour, 68, 1349–1360.
Bradshaw, C.J.A., Sims, D.W. & Hays, G.C. (2007). Measurement error causes scale-
dependent threshold erosion of biological signals in animal movement data.
Ecological Applications, 17, 628–638.
Branstetter, S., Musick, J.A. & Colvocoresses, J.A. (1987). A comparison of the age and
growth of the tiger shark, Galeocerdo cuvieri, from off Virginia and from the
northwestern Gulf of Mexico. Fishery Bulletin, 85, 269–279.
Breheny, P. & Burchett, W. (2016). Package ‘visreg’ : Visualization of Regression
Models. R package version 2.3-0.
Broderick, A.C., Coyne, M.S., Fuller, W.J., Glen, F. & Godley, B.J. (2007). Fidelity and
over-wintering of sea turtles. Proceedings of the Royal Society B, 274, 1533–1538.
Brothers, J.R. & Lohmann, K.J. (2015). Evidence for geomagnetic imprinting and
magnetic navigation in the natal homing of sea turtles. Current Biology, 25, 392–
396.
Brown, J., Laundre, J. & Gurung, M. (1999). The ecology of fear : Optimal foraging,
game theory, and trophic interactions. Journal of Mammalogy, 80, 385–399.
Bruce, B. (1992). Preliminary observations on the biology of the white shark,
Carcharodon carcharias, in South Australian waters. Marine and Freshwater
Research, 43, 1–11.
Bruce, B.D. & Bradford, R.W. (2008). Spatial dynamics & habitat preferences of
juvenile white sharks – identifying critical habitat and options for monitoring
recruitment. Final Report. CSIRO Marine and Atmospheric Research, Hobart.
Bruce, B.D., Stevens, J.D. & Malcolm, H. (2006). Movements and swimming behaviour
of white sharks (Carcharodon carcharias) in Australian waters. Marine Biology,
150, 161–172.
161
Brunsdon, C., Fotheringham, A.S. & Charlton, M.E. (1996). Geographically Weighted
Regression: A method for exploring spatial nonstationarity. Geographical
Analysis, 28, 281–298.
Bullard, F. (1991). Estimating the home range of an animal, a browniam bridge
approach. M.Sc. Thesis, University of North Carolina.
Bunnefeld, N., Börger, L., Van Moorter, B., Rolandsen, C.M., Dettki, H., Solberg, E.J.
& Ericsson, G. (2011). A model-driven approach to quantify migration patterns:
Individual, regional and yearly differences. Journal of Animal Ecology, 80, 466–
476.
Burgess, G.H., Beerkircher, L.R., Cailliet, G.M., Carlson, J.K., Cortes, E., Goldman,
K.J., Grubbs, R.D., Musick, J.A., Musyl, M.K. & Simpfendorfer, C.A. (2005). Is
the collapse of shark populations in the Northwest Atlantic Ocean and Gulf of
Mexico real? Fisheries, 30, 19–26.
Burkholder, D.A., Heithaus, M.R., Fourqurean, J.W., Wirsing, A. & Dill, L.M. (2013).
Patterns of top-down control in a seagrass ecosystem: could a roving apex predator
induce a behaviour-mediated trophic cascade? Journal of animal ecology, 82,
1192–1202.
Burnham, K.P. & Anderson, D.R. (2004). Multimodel inference: understanding AIC
and BIC in model selection. Sociological Methods & Research, 33, 261–304.
Burt, W. (1943). Territoriality and home range concepts. Journal of mammalogists, 24,
346–352.
Calenge, C. (2011). Home range estimation in R: the adehabitatHR Package.
Campana, S.E. (2016). Transboundary movements, unmonitored fishing mortality, and
ineffective international fisheries management pose risks for pelagic sharks in the
Northwest Atlantic. Canadian Journal of Fisheries and Aquatic Sciences, 73,
1599–1607.
Campana, S.E., Dorey, A., Fowler, M., Joyce, W., Wang, Z., Wright, D. & Yashayaev,
I. (2011). Migration pathways, behavioural thermoregulation and overwintering
grounds of blue sharks in the Northwest Atlantic. PLoS ONE, 6: e16854.
Carlisle, A.B., Goldman, K.J., Litvin, S.Y., Madigan, D.J., Bigman, J.S., Swithenbank,
A.M., Kline, T.C. & Block, B.A. (2014). Stable isotope analysis of vertebrae
reveals ontogenetic changes in habitat in an endothermic pelagic shark.
Proceedings of the Royal Society B: Biological Sciences, 282, 1–9.
Carlisle, A.B., Kim, S.L., Semmens, B.X., Madigan, D.J., Jorgensen, S.J., Perle, C.R.,
Anderson, S.D., Chapple, T.K., Kanive, P.E. & Block, B.A. (2012). Using stable
isotope analysis to understand the migration and trophic ecology of northeastern
Pacific white sharks (Carcharodon carcharias). PloS one, 7, e30492.
Carlson, J.K. & Gulak, S.J.B. (2012). Habitat use and movements patterns of oceanic
whitetip, bigeye thresher and dusky sharks based on archival satellite tags.
Collected Volumes Science Papers ICCAT, 68, 1922–1932.
Carlson, J.K., Hale, L.F., Morgan, A. & Burgess, G. (2012). Relative abundance and
size of coastal sharks derived from commercial shark longline catch and effort
data. Journal of Fish Biology, 80, 1749–1764.
Carlstein, E., Do, K., Hall, P., Hesterberg, T. & Künsch, H.R. (1998). Matched-block
bootstrap for dependent data. Bernoulli, 4, 305–328.
Carr, M., Neigel, J., Estes, J., Andelman, S., Warner, R.R. & Largier, J.L. (2003).
162
Comparing marine and terrestrial ecosystems : Implications for the design of
coastal marine reserves. Ecological Applications, 13, 90–107.
Carruthers, T.J.B., Dennison, W.C., Longstaff, B.J., Waycott, M., Abal, E.G.,
McKenzie, L.J. & Lee Long, W.J. (2002). Seagrass habitats of northeast Australia:
Models of key processes and controls. Bulletin of Marine Science, 71, 1153–1169.
Cartamil, D., Santana-Morales, O., Escobedo-Olvera, M., Kacev, D., Castillo-Geniz, L.,
Graham, J.B., Rubin, R.D., Sosa-Nishizaki, O. & Rochin-Alamillo, A. (2011). The
artisanal elasmobranch fishery of the Pacific coast of Baja California, Mexico.
Fisheries Research, 108, 393–403.
Cassata, L. & Collins, L.B. (2008). Coral reef communities, habitats, and substrates in
and near santuarty zones of Ningaloo Reef Marine Park. Journal of Coastal
Research, 24, 139–151.
Castillo-Geniz, J.L., Marquez-Farias, J.F., Rodrigues de la Cruz, M.C., Cortes, E. & Cid
del Prado, A. (1998). The Mexican artisanal shark fishery in the Gulf of Mexico:
towards a regulated fishery. Marine and Freshwater Research, 49, 611–620.
Chapman, J.W., Klaassen, R.H.G., Drake, V.A., Fossette, S., Hays, G.C., Metcalfe,
J.D., Reynolds, A.M., Reynolds, D.R. & Alerstam, T. (2011). Animal orientation
strategies for movement in flows. Current Biology, 21, R861–R870.
Chapman, D.D., Pikitch, E.K. & Babcock, E. (2005). Marine reserve design and
evaluation using automated acoustic telemetry : A case-study involving coral reef-
associated sharks in the Mesoamerican Caribbean. Marine Technology Society
Journal, 39, 42–55.
Chassot, E., Floch, L., Dewals, P., Damiano, A., Cauquil, P. & Chavance, P. (2014).
Fishing activities of the French and associated flags purse seiners targeting tropical
tunas in the Indian Ocean ( 1981-2013 ). IOTC-2014-WPTT16-XX.
Cherel, Y., Hobson, K.A., Guinet, C. & Vanpe, C. (2007). Stable isotopes document
seasonal changes in trophic niches and winter foraging individual specialization in
diving predators from the Southern Ocean. Journal of Animal Ecology, 76, 826–
836.
Churchill, D.A. (2015). Investigating trophic interactions of deep-sea animals (sharks,
teleosts, and mobile scavengers ) in the Gulf of Mexico using stable isotope
analysis. Ph.D. Thesis, Florida International University, Miami.
Clark, E. & von Schmidt, K. (1965). Sharks of the central gulf coast of Florida. Bulletin
of Marine Science, 15, 13–83.
Clarke, S.C., McAllister, M.K., Milner-Gulland, E.J., Kirkwood, G.P., Michielsens,
C.G.J., Agnew, D.J., Pikitch, E.K., Nakano, H. & Shivji, M.S. (2006). Global
estimates of shark catches using trade records from commercial markets. Ecology
Letters, 9, 1115–1126.
Clementz, M. & Koch, P. (2001). Differentiating aquatic mammal habitat and foraging
ecology with stable isotopes in tooth enamel. Oecologia, 129, 461–472.
Cliff, G. & Dudley, S.F.J. (2011). Reducing the environmental impact of shark-control
programs: a case study from KwaZulu-Natal, South Africa. Marine and
Freshwater Research, 62, 700.
Clua, E., Chauvet, C., Read, T., Werry, J.M. & Lee, S.Y. (2013). Behavioural patterns
of a tiger shark (Galeocerdo cuvier) feeding aggregation at a blue whale carcass in
Prony Bay, New Caledonia. Marine and Freshwater Behaviour and Physiology,
163
46, 1–20.
Compagno, L.J. V. (1984). Sharks of the world. An annotated and illustrated catalogue
of shark species know to date., Vol.4n. FAO species catalogue for fishery
purposes. Food and Agriculture Organisation of the United Nations, Rome.
Cooke, S.J. (2008). Biotelemetry and biologging in endangered species research and
animal conservation: Relevance to regional, national, and IUCN Red List threat
assessments. Endangered Species Research, 4, 165–185.
Costa, D.P., Block, B. A., Bograd, S.J., Fedak, M. A. & Gunn, J.S. (2010a). TOPP as a
marine life observatory: Using electronic tags to monitor the movements,
behaviour and habitats of marine vertebrates. Proceedings of OceanObs’09:
Sustained Ocean Observations and Information for Society (Vol 2.), 190–198.
Costa, D.P., Robinson, P.W., Arnould, J.P.Y., Harrison, A.L., Simmons, S.E., Hassrick,
J.L., Hoskins, A.J., Kirkman, S.P., Oosthuizen, H., Villegas-Amtmann, S. &
Crocker, D.E. (2010b). Accuracy of ARGOS locations of pinnipeds at-sea
estimated using Fastloc GPS. PLoS ONE, 5, e8677.
Cotte, C., Park, Y.H., Guinet, C. & Bost, C.A. (2007). Movements of foraging king
penguins through marine mesoscale eddies. Proceedings of the Royal Society B-
Biological Sciences, 274, 2385–2391.
Crawshaw, L.I. (1977). Physiological and behavioral reactions of fishes to temperature
change. Journal of the Fisheries Research Board of Canada, 34, 730–734.
Creel, S. & Christianson, D. (2008). Relationships between direct predation and risk
effects. Trends in ecology & evolution, 23, 194–201.
Creel, S., Schuette, P. & Christianson, D. (2014). Effects of predation risk on group
size, vigilance, and foraging behavior in an African ungulate community.
Behavioral Ecology, 25, 773–784.
Creel, S., Winnie, J., Maxwell, B., Hamlin, K. & Creel, M. (2005). Elk alter habitat
selection as an antipredator response to wolves. Ecology, 86, 3387–3397.
Cresswell, G. (2000). Currents of the continental shelf and upper slope of Tasmania.
Papers and Proceedings of the Royal Society of Tasmania, 133, 21–30.
Crist, T., Guertin, D.S., Wiens, J.A. & Milne, B.T. (1992). Animal movement in
heterogeneous landscapes: An experiment with Eleodes beetles in shortgrass
prairie. Functional Ecology, 6, 219–226.
Croll, D.A., Marinovic, B., Benson, S., Chavez, F.P., Black, N., Ternullo, R. & Tershy,
B.R. (2005). From wind to whales: trophic links in a coastal upwellling system.
Marine Ecology Progress Series, 289, 117–130.
Dallimer, M. & Strange, N. (2015). Why socio-political borders and boundaries matter
in conservation. Trends in Ecology and Evolution, 30, 132–139.
Dapp, D., Arauz, R., Spotila, J.R. & O’Connor, M.P. (2013). Impact of Costa Rican
longline fishery on its bycatch of sharks, stingrays, bony fish and olive ridley
turtles (Lepidochelys olivacea). Journal of Experimental Marine Biology and
Ecology, 448, 228–239.
Davenport, S.R. & Bax, N.J. (2002). A trophic study of a marine ecosystem off
southeastern Australia using stable isotopes of carbon and nitrogen. Canadian
Journal of Fisheries and Aquatic Sciences, 59, 514–530.
Davis, J.P., Pitt, K.A., Fry, B. & Connolly, R.M. (2015). Stable isotopes as tracers of
residency for fish on inshore coral reefs. Estuarine, Coastal and Shelf Science,
164
167, 368–376.
Department of Fisheries of Western Australia (2005). Application to the Australian
Government Department of Environment and Heritage on the Western Australian
Tropical Shark Fisheries (covering the Western Australia North Coast Shark
Fishery and the Joint Authority Northern Shark Fishery). Department of Fisheries,
Western Australia.
Department of Fisheries Western Australia (2006). Assessment of the Western
Australian Tropical Shark Fisheries. Department of the Natural Environment and
Heritage, Western Australia.
Dicken, M.L. & Hosking, S.G. (2009). Socio-economic aspects of the tiger shark diving
industry within the Aliwal Shoal Marine Protected Area, South Africa. African
Journal of Marine Science, 31, 227–232.
Dingle, H. & Drake, V.A. (2007). What Is Migration? BioScience, 57, 113–121.
Domeier, M.L. & Nasby-Lucas, N. (2007). Annual re-sightings of photographically
identified white sharks (Carcharodon carcharias) at an eastern Pacific aggregation
site (Guadalupe Island, Mexico). Marine Biology, 150, 977–984.
Domeier, M. & Nasby-Lucas, N. (2008). Migration patterns of white sharks
Carcharodon carcharias tagged at Guadalupe Island, Mexico, and identification of
an eastern Pacific shared offshore foraging area. Marine Ecology Progress Series,
370, 221–237.
Domeier, M.L. & Nasby-Lucas, N. (2013). Two-year migration of adult female white
sharks (Carcharodon carcharias) reveals widely separated nursery areas and
conservation concerns. Animal Biotelemetry, 1, 1–9.
Domingo, A., Coelho, R., Cortes, E., Garcia-Cortes, B., Mas, F., Mejuto, J., Miller, P.,
Ramos-Cartelle, A., Santos, M.N. & Yokawa, K. (2016). Is the tiger shark
Galeocerdo cuvier a coastal species? Expanding its distribution range in the
Atlantic Ocean using at-sea observer data. Journal of Fish Biology, 88, 1223–1228.
Doniol-Valcroze, T., Berteaux, D., Larouche, P. & Sears, R. (2007). Influence of
thermal fronts on habitat selection by four rorqual whale species in the Gulf of St.
Lawrence. Marine Ecology Progress Series, 335, 207–216.
Double, M.C., Andrews-Goff, V., Jenner, K.C.S., Jenner, M.N., Laverick, S.M.,
Branch, T.A. & Gales, N.J. (2014). Migratory movements of pygmy blue whales
(Balaenoptera musculus brevicauda) between Australia and Indonesia as revealed
by satellite telemetry. PLoS ONE, 9, 1–11.
Driggers, W., Ingram, G., Grace, M., Gledhill, C., Henwood, T., Horton, C. & Jones, C.
(2008). Pupping areas and mortality rates of young tiger sharks Galeocerdo cuvier
in the western North Atlantic Ocean. Aquatic Biology, 2, 161–170.
Drymon, J.M., Carassou, L., Powers, S.P., Grace, M., Dindo, J. & Dzwonkowski, B.
(2013). Multiscale analysis of factors that affect the distribution of sharks
throughout the northern Gulf of Mexico. Fishery Bulletin, 111, 370–380.
Dudley, S.F.J. (1997). A comparison of the shark control programs of New South Wales
and Queensland (Australia) and KwaZulu-Natal (South Africa). Ocean & Coastal
Management, 34, 1–27.
Dudley, S.F.J. & Simpfendorfer, C.A. (2006). Population status of 14 shark species
caught in the protective gillnets off KwaZulu–Natal beaches, South Africa, 1978–
2003. Marine and Freshwater Research, 57, 225.
165
Dulvy, N.K., Baum, J.K., Clarke, S., Compagno, L.J. V, Cortes, E., Domingo, A.,
Fordham, S., Fowler, S., Francis, M.P., Gibson, C., Martinez, J., Musick, J.A.,
Soldo, A., Stevens, J.D. & Valenti, S. (2008). You can swim but you can’t hide :
the global status and conservation of oceanic pelagic sharks and rays. Aquatic
Conservation: Marine and Freshwater Ecosystems, 18, 459–482.
Dulvy, N.K., Fowler, S.L., Musick, J.A., Cavanagh, R.D., Kyne, P.M., Harrison, L.R.,
Carlson, J.K., Davidson, L.N.K., Fordham, S. V., Francis, M.P., Pollock, C.M.,
Simpfendorfer, C.A., Burgess, G.H., Carpenter, K.E., Compagno, L.J.V., Ebert,
D.A., Gibson, C., Heupel, M.R., Livingstone, S.R., Sanciangco, J.C., Stevens, J.D.,
Valenti, S. & White, W.T. (2014a). Extinction risk and conservation of the world’s
sharks and rays. eLife, 3, e00590–e00590.
Dulvy, N.K., Fowler, S.L., Musick, J.A., Cavanagh, R.D., Kyne, P.M., Harrison, L.R.,
Carlson, J.K., Davidson, L.N., Fordham, S. V., Francis, M.P., Pollock, C.M.,
Simpfendorfer, C. A., Burgess, G.H., Carpenter, K.E., Compagno, L.J., Ebert, D.
A., Gibson, C., Heupel, M.R., Livingstone, S.R., Sanciangco, J.C., Stevens, J.D.,
Valenti, S. & White, W.T. (2014b). Extinction risk and conservation of the world’s
sharks and rays. eLife, 3, e00590–e00590.
Dunn, D.C., Maxwell, S.M., Boustany, A.M. & Halpin, P.N. (2016). Dynamic ocean
management increases the efficiency and efficacy of fisheries management.
Proceedings of the National Academy of Sciences , 113, 668–673.
Egevang, C., Stenhouse, I.J., Phillips, R.A., Petersen, A., Fox, J.W. & Silk, J.R.D.
(2010). Tracking of Arctic terns Sterna paradisaea reveals longest animal
migration. Proceedings of the National Academy of Sciences of the United States of
America, 107, 2078–2081.
Eiserhardt, W.L., Bjorholm, S., Svenning, J.C., Rangel, T.F. & Balslev, H. (2011).
Testing the water-energy theory on american palms (Arecaceae) using
geographically weighted regression. PLoS ONE, 6:e27027.
Espinoza, M., Cappo, M., Heupel, M.R., Tobin, A.J. & Simpfendorfer, C.A. (2014).
Quantifying shark distribution patterns and species-habitat associations:
Implications of marine park zoning. PLoS ONE, 9, e106885.
Espinoza, M., Farrugia, T.J. & Lowe, C.G. (2011). Habitat use, movements and site
fidelity of the gray smooth-hound shark (Mustelus californicus Gill 1863) in a
newly restored southern California estuary. Journal of Experimental Marine
Biology and Ecology, 401, 63–74.
Estes, J., Crooks, K. & Holt, R. (2001). Ecological role of predators. Encyclopedia of
Biodiversity, 4, 857–878.
Estes, J.A. & Duggins, D.O. (1995). Sea otters and kelp forests in Alaska: Generality
and variation in a community ecological paradigm. Ecological Monographs, 65,
75–100.
Estes, J.A., Heithaus, M., McCauley, D.J., Rasher, D.B. & Worm, B. (2016).
Megafaunal impacts on structure and function of ocean ecosystems. Annual Review
of Environment and Resources, 41, 83–116.
Estes, J.A., Terborgh, J., Brashares, J.S., Power, M.E., Berger, J., Bond, W.J.,
Carpenter, S.R., Essington, T.E., Holt, R.D., Jackson, J.B.C., Marquis, R.J.,
Oksanen, L., Oksanen, T., Paine, R.T., Pikitch, E.K., Ripple, W.J., Sandin, S. A.,
Scheffer, M., Schoener, T.W., Shurin, J.B., Sinclair, A.R.E., Soulé, M.E., Virtanen,
166
R. & Wardle, D. A. (2011). Trophic downgrading of planet Earth. Science, 333,
301–306.
Estrada, J. A., Rice, A.N., Natanson, L.J. & Skomal, G.B. (2006). Use of isotopic
analysis of vertebrae in reconstructing ontogenetic feeding ecology in white sharks.
Ecology, 87, 829–834.
Fabiani, A., Galimberti, F., Sanvito, S. & Hoelzel, A.R. (2006). Relatedness and site
fidelity at the southern elephant seal, Mirounga leonina, breeding colony in the
Falkland Islands. Animal Behaviour, 72, 617–626.
Farmer, B.M. & Wilson, S.K. (2011). Diet of finfish targeted by fishers in North West
Australia and the implications for trophic cascades. Environmental Biology of
Fishes, 91, 71–85.
Fauchald, P. (1999). Foraging in a hierarchical patch system. The American Naturalist,
153, 603–613.
Fauchald, P. & Tveraa, T. (2006). Hierarchical patch dynamics and animal movement
pattern. Oecologia, 149, 383–395.
Fauchald, P. & Tveraa, T. (2003). Using first-passage time in the analysis of area-
restricted search and habitat selection. Ecology, 84, 282–288.
Ferreira, L.C., Thums, M., Meeuwig, J.J., Vianna, G.M.S., Stevens, J., McAuley, R. &
Meekan, M.G. (2015). Crossing latitudes—Long-distance tracking of an apex
predator. Plos One, 10, e0116916.
Field, I.C., Meekan, M.G., Buckworth, R.C. & Bradshaw, C.J.A. (2009). Protein mining
the world’s oceans: Australasia as an example of illegal expansion-and-
displacement fishing. Fish and Fisheries, 10, 323–328.
Fitzpatrick, R., Thums, M., Bell, I., Meekan, M.G., Stevens, J.D. & Barnett, A. (2012).
A comparison of the seasonal movements of tiger sharks and green turtles provides
insight into their predator-prey relationship. PloS one, 7, e51927.
Food and Agriculture Organization of the United Nations. (2014). Preparation of
management plans for selected fisheries. . FAO EAF-Nansen Project Report, 23,
A. (eds. K.A. Koranteng, M.C. Vasconcellos & B.P. Satia). Rome.
Foody, G.M. (2004). Spatial nonstationarity and scaledependency in the relationship
between species richness and environmental determinants for the sub-Saharan
endemic avifauna. Global Ecology and Biogeography, 13, 315–320.
Ford, J.K.B. & Ellis, G.M. (2006). Selective foraging by fish-eating killer whales
Orcinus orca in British Columbia. Marine Ecology Progress Series, 316, 185–199.
Ford, J.K.B., Ellis, G.M., Olesiuk, P.F. & Balcomb, K.C. (2010). Linking killer whale
survival and prey abundance: Food limitation in the oceans’ apex predator?
Biology letters, 6, 139–142.
Fortin, D., Beyer, H.L., Boyce, M.S., Smith, D.W., Duchesne, T. & Mao, J.S. (2005).
Wolves influence elk movements: Behavior shapes a trophic cascade in
Yellowstone National Park. Ecology, 86, 1320–1330.
Fotheringham, A., Charlton, M.E. & Brunsdon, C. (1998). Geographically weighted
regression: A natural evolution of the expansion method for spatial data analysis.
Environment and Planning A, 30, 1905–1927.
Francis, M. (1996). Observation on a pregnant white shark with a review of
reproductive biology. Great white sharks: the biology of Carcharodon carcharias.
(eds P.A. Klimley & D.G. Ainley), pp. 157–172. Academic Press, San Diego.
167
Frisch, A.J., Ireland, M. & Baker, R. (2014). Trophic ecology of large predatory reef
fishes : energy pathways, trophic level, and implications for fisheries in a changing
climate. Marine Biology, 161, 61–73.
Frisch, A.J., Ireland, M., Rizzari, J.R., Lönnstedt, O.M., Magnenat, K.A., Mirbach, C.E.
& Hobbs, J.P.A. (2016). Reassessing the trophic role of reef sharks as apex
predators on coral reefs. Coral Reefs, 1–14.
Frisk, T.J., Miller, T.M. & Fogart, M.J. (2001). Estimation and analysis of biological
parameters in elasmobranch fishes: a comparative life history study. Can. J. Fish.
Aquat. Sci., 58, 969–981.
Fritz, H., Said, S. & Weimerskirch, H. (2003). Scale-dependent hierarchical adjustments
of movement patterns in a long-range foraging seabird. Proceedings. Biological
sciences / The Royal Society, 270, 1143–1148.
Fryxell, J.M., Hazell, M., Börger, L., Dalziel, B.D., Haydon, D.T., Morales, J.M.,
McIntosh, T. & Rosatte, R.C. (2008). Multiple movement modes by large
herbivores at multiple spatiotemporal scales. Proceedings of the National Academy
of Sciences of the United States of America, 105, 19114–19119.
Gallagher, A.J., Jackson, T. & Hammerschlag, N. (2011). Occurrence of tiger shark
(Galeocerdo cuvier) scavenging on avian prey and its possible connection to large-
scale bird die-offs in the Florida Keys. Florida Scientist, 74, 264–269.
Gallagher, A.J., Serafy, J.E., Cooke, S.J. & Hammerschlag, N. (2014). Physiological
stress response, reflex impairment, and survival of five sympatric shark species
following experimental capture and release. Marine Ecology Progress Series, 496,
207–218.
Gallon, S., Bailleul, F., Charrassin, J.B., Guinet, C., Bost, C.A., Handrich, Y. & Hindell,
M. (2013). Identifying foraging events in deep diving southern elephant seals,
Mirounga leonina, using acceleration data loggers. Deep-Sea Research Part II:
Topical Studies in Oceanography, 88–89, 14–22.
Galuardi, B. & Lutcavage, M. (2012). Dispersal routes and habitat utilization of juvenile
Atlantic bluefin tuna, Thunnus thynnus, tracked with mini PSAT and archival tags.
PloS one, 7, e37829.
Galuardi, B., Royer, F., Golet, W., Logan, J., Neilson, J. & Lutcavage, M. (2010).
Complex migration routes of Atlantic bluefin tuna (Thunnus thynnus) question
current population structure paradigm. Canadian Journal of Fisheries and Aquatic
Sciences, 67, 966–976.
Game, E.T., Grantham, H.S., Hobday, A.J., Pressey, R.L., Lombard, A.T., Beckley,
L.E., Gjerde, K., Bustamante, R., Possingham, H.P. & Richardson, A.J. (2009).
Pelagic protected areas: The missing dimension in ocean conservation. Trends in
ecology & evolution, 24, 360–9.
Gilmore, R.G. (1993). Reproductive biology of lamnoid sharks. Environmental Biology
of Fishes, 38, 95–114.
Goldman, K.J., Branstetter, S. & Musick, J.A. (2006). A re-examination of the age and
growth of sand tiger sharks, Carcharias taurus, in the western North Atlantic: The
importance of ageing protocols and use of multiple back-calculation techniques.
Environmental Biology of Fishes, 77, 241–252.
Gonzalez-Pestana, A., Kouri J., C. & Velez-Zuazo, X. (2014). Shark fisheries in the
Southeast Pacific: A 61-year analysis from Peru. F1000Research, 3.
168
Graham, B., Koch, P., Newsome, S., McMahon, K.W. & Aurioles, D. (2010). Using
isoscapes to trace the movements and foraging behavior of top predators in oceanic
ecosystems. Isoscapes: Understanding movement, pattern, and process on Earth
through isotope mapping (eds J.B. West, G.J. Bowen, T.E. Dawson & K.P. Tu),
pp. 299–318. Springer.
Green, M., Ganassin, C. & Reid, D.D. (2009). Report into the NSW Shark Meshing
(Bather Protection) Program. New South Wales Department of Primary Industries
Fisheries Conservation and Aquiculture Branch. New South Wales
Grudger, E.W. (1949). Natural history notes on tiger sharks, Galeocerdo tigrinus,
caught at Key West, Florida, with emphasis on food and feeding habits. Copeia, 1,
39–47.
Guinet, C., Dubroca, L., Lea, M.A., Goldsworthy, S., Cherel, Y., Duhamel, G.,
Bonadonna, F. & Donnay, J.P. (2001). Spatial distribution of foraging in female
antarctic fur seals Arctocephalus gazella in relation to oceanographic variables: A
scale-dependent approach using geographic information systems. Marine Ecology
Progress Series, 219, 251–264.
Gulak, S.J.B., Enzenauer, M.P. & Carlson, J.K. (2013). Characterization of the shark
and reef fish bottom longline fisheries: 2012. NOAA Technical Memorandum
NMFS-SEFSC-652, 42.
Hammerschlag, N., Bell, I., Fitzpatrick, R., Gallagher, A.J., Hawkes, L.A., Meekan,
M.G., Stevens, J.D., Thums, M., Witt, M.J. & Barnett, A. (2016). Behavioral
evidence suggests facultative scavenging by a marine apex predator during a food
pulse. Behavioral Ecology and Sociobiology.
Hammerschlag, N., Gallagher, A.J. & Lazarre, D.M. (2011a). A review of shark satellite
tagging studies. Journal of Experimental Marine Biology and Ecology, 398, 1–8.
Hammerschlag, N., Gallagher, A.J., Lazarre, D.M. & Slonim, C. (2011b). Range
extension of the endangered great hammerhead shark Sphyrna mokarran in the
Northwest Atlantic: Preliminary data and significance for conservation.
Endangered Species Research, 13, 111–116.
Hammerschlag, N., Gallagher, A.J., Wester, J., Luo, J. & Ault, J.S. (2012). Don’t bite
the hand that feeds: assessing ecological impacts of provisioning ecotourism on an
apex marine predator. Functional Ecology, 26, 567–576.
Handley, A.J. (2010). Fishery Status Reports 2009. Northern Territory Government
Department of Resources. Fishery Report No. 104
Harestad, A.S. & Bunnel, F.L. (1979). Home range and body weight-A reevaluation.
Ecology, 60, 389–402.
Harper, S., Guzmán, H.M., Zylich, K. & Zeller, D. (2014). Reconstructing Panama’s
total fisheries catches from 1950 to 2010: Highlighting data deficiencies and
management needs. Marine Fisheries Review, 76, 51–65.
Harris, P., Heap, A., Passlow, V., Sbaffi, L., Fellows, M., Porter-Smith, R., Buchanan,
C. & Daniell, J. (2005). Geomorphic Features of the Continental Margin of
Australia. Geoscience Australia 2003/30, 1-142.
Hatase, H., Takai, N., Matsuzawa, Y., Sakamoto, W., Omuta, K., Goto, K., Arai, N. &
Tateki, F. (2002). Size-related differences in feeding habitat use of adult female
loggerhead turtles Caretta caretta around Japan determined by stable isotope
analyses and satellite telemetry. Marine Ecology Progress Series, 233, 273–281.
169
Hays, G.C., Åkesson, S., Godley, B.J., Luschi, P. & Santidrian, P. (2001). The
implications of location accuracy for the interpretation of satellite-tracking data.
Animal Behaviour, 61, 1035–1040.
Hays, G.C., Bradshaw, C.J.A., James, M.C., Lovell, P. & Sims, D.W. (2007). Why do
Argos satellite tags deployed on marine animals stop transmitting? Journal of
Experimental Marine Biology and Ecology, 349, 52–60.
Hays, G.C., Ferreira, L.C., Sequeira, A.M.M., Meekan, M.G., Duarte, C.M., Bailey, H.,
Bailleul, F., Bowen, W.D., Caley, M.J., Costa, D.P., Eguíluz, V.M., Fossette, S.,
Friedlaender, A.S., Gales, N., Gleiss, A.C., Gunn, J., Harcourt, R., Hazen, E.L.,
Heithaus, M.R., Heupel, M., Holland, K., Horning, M., Jonsen, I., Kooyman, G.L.,
Lowe, C.G., Madsen, P.T., Marsh, H., Phillips, R.A., Righton, D., Ropert-Coudert,
Y., Sato, K., Shaffer, S.A., Simpfendorfer, C.A., Sims, D.W., Skomal, G.,
Takahashi, A., Trathan, P.N., Wikelski, M., Womble, J.N. & Thums, M. (2016).
Key Questions in Marine Megafauna Movement Ecology. Trends in Ecology and
Evolution, 31, 463–475.
Hays, G.C., Mortimer, J. A., Ierodiaconou, D. & Esteban, N. (2014). Use of long-
distance migration patterns of an endangered species to inform conservation
planning for the world’s largest Marine Protected Area. Conservation Biology, 28,
1636–1644.
Hays, G.C. & Scott, R. (2013). Global patterns for upper ceilings on migration distance
in sea turtles and comparisons with fish, birds and mammals. Functional Ecology,
27, 748–756.
Hayward, M.W., Hayward, G.J., Tambling, C.J. & Kerley, G.I.H. (2011). Do lions
Panthera leo actively select prey or do prey preferences simply reflect chance
responses via evolutionary adaptations to optimal foraging? PLoS ONE, 6: e23607.
Hayward, M.W., Jedrzejewski, W. & Jedrzejewska, B. (2012). Prey preferences of the
tiger Panthera tigris. Journal of Zoology, 286, 221–231.
Hayward, M.W., O’Brien, J. & Kerley, G.I.H. (2007). Carrying capacity of large
African predators: Predictions and tests. Biological Conservation, 139, 219–229.
Hazel, J. (2009). Evaluation of fast-acquisition GPS in stationary tests and fine-scale
tracking of green turtles. Journal of Experimental Marine Biology and Ecology,
374, 58–68.
Hazen, E.L., Jorgensen, S., Rykaczewski, R.R., Bograd, S.J., Foley, D.G., Jonsen, I.D.,
Shaffer, S. A., Dunne, J.P., Costa, D.P., Crowder, L.B. & Block, B. A. (2013).
Predicted habitat shifts of Pacific top predators in a changing climate. Nature
Climate Change, 3, 234–238.
Hazen, E.L., Palacios, D.M., Forney, K.A., Howell, E.A., Becker, E., Hoover, A.L.,
Irvine, L., DeAngelis, M., Bograd, S.J., Mate, B.R. & Bailey, H. (2016).
WhaleWatch: A dynamic management tool for predicting blue whale density in the
California Current. Journal of Applied Ecology, n/a-n/a.
Hazin, F.H., Afonso, A.S., De Castilho, P.C., Ferreira, L.C. & Rocha, B.C. (2013).
Regional movements of the tiger shark, Galeocerdo cuvier, off Northeastern
Brazil: inferences regarding shark attack hazard. Anais da Academia Brasileira de
Ciências, 85, 1053–1062.
Hazin, F.H. V, Burgess, G.H. & Carvalho, F.C. (2008). A shark attack outbreak off
Recife , Pernambuco, Brazil: 1992-2006. Bulletin of Marine Science, 82, 199–212.
170
Heithaus, M.R. (2004). Fish communities of subtropical seagrass meadows and
associated habitats in Shark Bay , Western Australia. Bulletin of Marine Science,
75, 79–99.
Heithaus, M.R. (2005). Habitat use and group size of pied cormorants (Phalacrocorax
varius) in a seagrass ecosystem: Possible effects of food abundance and predation
risk. Marine Biology, 147, 27–35.
Heithaus, M.R. (2001). The biology of tiger sharks, Galeocerdo cuvier, in Shark Bay,
Western Australia: Sex ratio, size distribution, diet, and seasonal changes in catch
rates. Environmental Biology of Fishes, 61, 25–36.
Heithaus, M.R., Alcoverro, T., Arthur, R., Burkholder, D. A., Coates, K. A.,
Christianen, M.J.A., Kelkar, N., Manuel, S. A., Wirsing, A.J., Kenworthy, W.J. &
Fourqurean, J.W. (2014). Seagrasses in the age of sea turtle conservation and shark
overfishing. Frontiers in Marine Science, 1, 1–6.
Heithaus, M.R., Burkholder, D., Hueter, R.E., Heithaus, L.I., Pratt, H.L. & Carrier, J.C.
(2007a). Spatial and temporal variation in shark communities of the lower Florida
Keys and evidence for historical population declines. Canadian Journal of
Fisheries and Aquatic Sciences, 64, 1302–1313.
Heithaus, M.R., Dill, L.M., Marshall, G.J. & Buhleier, B. (2002). Habitat use and
foraging behavior of tiger sharks (Galeocerdo cuvier) in a seagrass ecosystem.
Marine Biology, 140, 237–248.
Heithaus, M.R., Frid, A., Vaudo, J.J., Worm, B. & Wirsing, A.J. (2010). Unraveling the
ecological importance of elasmobranchs. Sharks and their relatives II: biodiversity,
adaptive physiology, and conservation (eds J.C. Carrier, J.A. Musick & M.R.
Heithaus), pp. 612–638. CRC Press, Boca Raton.
Heithaus, M.R., Frid, A., Wirsing, A.J., Dill, L.M., Fourqurean, J.W., Burkholder, D.,
Thomson, J. & Bejder, L. (2007b). State-dependent risk-taking by green sea turtles
mediates top-down effects of tiger shark intimidation in a marine ecosystem.
Journal of Animal Ecology, 76, 837–44.
Heithaus, M.R., Frid, A., Wirsing, A.J. & Worm, B. (2008a). Predicting ecological
consequences of marine top predator declines. Trends in ecology & evolution, 23,
202–210.
Heithaus, M., Marshall, G., Buhleier, B. & Dill, L. (2001). Employing crittercam to
study habitat use and behavior of large sharks. Marine Ecology Progress Series,
209, 307–310.
Heithaus, M., Vaudo, J., Kreicker, S., Layman, C., Krützen, M., Burkholder, D.,
Gastrich, K., Bessey, C., Sarabia, R., Cameron, K., Wirsing, A., Thomson, J. &
Dunphy-Daly, M. (2013). Apparent resource partitioning and trophic structure of
large-bodied marine predators in a relatively pristine seagrass ecosystem. Marine
Ecology Progress Series, 481, 225–237.
Heithaus, M.R., Wirsing, A.J., Burkholder, D., Thomson, J. & Dill, L.M. (2009).
Towards a predictive framework for predator risk effects: The interaction of
landscape features and prey escape tactics. The Journal of animal ecology, 78,
556–562.
Heithaus, M.R., Wirsing, A.J. & Dill, L.M. (2012). The ecological importance of intact
top-predator populations: A synthesis of 15 years of research in a seagrass
ecosystem. Marine and Freshwater Research, 63, 1039–1050.
171
Heithaus, M.R., Wirsing, A.J., Dill, L.M. & Heithaus, L.I. (2007c). Long-term
movements of tiger sharks satellite-tagged in Shark Bay, Western Australia.
Marine Biology, 151, 1455–1461.
Heithaus, M.R., Wirsing, A.J., Thomson, J.A. & Burkholder, D.A. (2008b). A review of
lethal and non-lethal effects of predators on adult marine turtles. Journal of
Experimental Marine Biology and Ecology, 356, 43–51.
Heupel, M.R., Carlson, J.K. & Simpfendorfer, C.A. (2007). Shark nursery areas:
Concepts, definition, characterization and assumptions. Marine Ecology Progress
Series, 337, 287–297.
Heupel, M.R., Knip, D.M., Simpfendorfer, C. A. & Dulvy, N.K. (2014). Sizing up the
ecological role of sharks as predators. Marine Ecology Progress Series, 495, 291–
298.
Heupel, M.R., Semmens, J.M.B. & Hobday, A.J. (2006). Automated acoustic tracking
of aquatic animals: scales, design and deployment of listening station arrays.
Marine and Freshwater Research, 57, 1–13.
Heupel, M.R. & Simpfendorfer, C.A. (2008). Movement and distribution of young bull
sharks Carcharhinus leucas in a variable estuarine environment. Aquatic Biology,
1, 277–289.
Heupel, M.R., Simpfendorfer, C.A., Espinoza, M., Smoothey, A.F., Tobin, A. &
Peddemors, V. (2015). Conservation challenges of sharks with continental scale
migrations. Frontiers in Marine Science, 2, 1–7.
Hill, R.D. & Braun, M.J. (2001). Geolocation by light level the next step: Latitude.
Electronic tagging and tracking in marine fisheries reviews: Methods and
technologies in fish biology and fisheries (eds J. Sibert & J. Nielsen), pp. 315–330.
Kluwer Academic Press, Dordrecht.
Hindell, M.A. (2008). To breathe or not to breathe: Optimal strategies for finding prey
in a dark, three-dimensional environment. Journal of Animal Ecology, 77, 847–
849.
Hindell, M.A., Mcmahon, C.R., Bester, M.N., Boehme, L., Costa, D., Fedak, M.A.,
Guinet, C., Herraiz-borreguero, L., Harcourt, R.G., Huckstadt, L., Kovacs, K.M.,
Lydersen, C., Mcintyre, T., Muelbert, M., Paterson, T., Roquet, F., Williams, G. &
Charrassin, J.-B. (2016). Circumpolar habitat use in the southern elephant seal:
Implications for foraging success and population trajectories. Ecosphere, 7, 1–27.
Hobday, A.J., Flint, N., Stone, T. & Gunn, J.S. (2009). Electronic tagging data
supporting flexible spatial management in an Australian longline fishery. Tagging
and tracking of marine animals with electronic devices, pp. 381–403. Springer
Netherlands.
Hobday, A.J. & Hartmann, K. (2006). Near real-time spatial management based on
habitat predictions for a longline bycatch species. Fisheries Management and
Ecology, 13, 365–380.
Hobday, A.J. & Hartog, J.R. (2014). Derived ocean features for dynamic ocean
management. Oceanography, 27, 134–145.
Hobday, A.J., Maxwell, S.M., Forgie, J., Mcdonald, J., Darby, M., Sesto, K., Bailey, H.,
Bograd, S.J., Briscoe, D.K., Costa, D.P., Cowder, L.B., Dunn, D.C., Fossette, S.,
Halpin, P.N., Hartog, J.R., Hazen, E.L., Lascelles, B.G., Lewison, R.L., Poulos, G.
& Powers, A. (2014). Dynamic Ocean Management: Integrating scientific and
172
technological capacity with law, policy and management. Stanford Environmental
Law Journal, 33, 125–165.
Hobson, K.A. (1999). Tracing origins and migration of wildlife using stable isotopes: A
review. Oecologia, 120, 314–326.
Hobson, K.A. & Clark, R. (1992). Assessing avian diets using stable isotopes :
Turnover of 13
C in tissues. The Condor, 94, 181–188.
Hobson, K.A., Piattt, J.F. & Pitocchellij, J.A.Y. (1994). Using stable isotopes to
determine seabird trophic relationships. Journal of Animal Ecology, 63, 786–798.
Hoegh-Guldberg, O. & Bruno, J.F. (2010). The impact of climate change on the world’s
marine ecosystems. Science, 328, 1523–1528.
Hoelzel, A.R. (1991). Killer whale predation on marine mammals at Punta Norte,
Argentina; Food sharing, provisioning and foraging strategy. Behavioral Ecology
and Sociobiology, 29, 197–204.
Hoey, J. & Casey, J. (1986). Shark longline fisheries: Gear and production
characteristics. Proceedings of the 37th Annual Gulf & Caribbean Fisheries
Institute (eds F. Williams), 169-186. Cancun, Mexico.
Hoffmayer, E.R., Franks, J.S., Driggers III, W.B., McKinney, J.A., Hendon, J.M. &
Quattro, J.M. (2014). Habitat, movements and environmental preferences of dusky
sharks, Carcharhinus obscurus, in the northern Gulf of Mexico. Marine Biology,
161, 911–924.
Holland, K.N., Wetherbee, B.M., Lowe, C.G. & Meyer, C.G. (1999). Movements of
tiger sharks (Galeocerdo cuvier) in coastal Hawaiian waters. Marine Biology, 134,
665–673.
Holmes, W.G. (1991). Predator risk affects foraging behaviour of pikas: observational
and experimental evidence. Animal Behaviour, 42, 111–119.
Holmes, K.W., Van Niel, K.P., Radford, B., Kendrick, G.A. & Grove, S.L. (2008).
Modelling distribution of marine benthos from hydroacoustics and underwater
video. Continental Shelf Research, 28, 1800–1810.
Holmes, B.J., Peddemors, V., Gutteridge, A., Geraghty, P., Chan, R., Tibbetts, I. &
Bennett, M.B. (2015). Age and growth of the tiger shark Galeocerdo cuvier off the
east coast of Australia. Journal of Fish Biology, 87, 422–448.
Holmes, B.J., Pepperell, J.G., Griffiths, S.P., Jaine, F.R.A., Tibbetts, I.R. & Bennett,
M.B. (2014). Tiger shark (Galeocerdo cuvier) movement patterns and habitat use
determined by satellite tagging in eastern Australian waters. Marine Biology, 161,
2645–2658.
Holmes, B.J., Sumpton, W.D., Mayer, D.G., Tibbetts, I.R., Neil, D.T. & Bennett, M.B.
(2012). Declining trends in annual catch rates of the tiger shark (Galeocerdo
cuvier) in Queensland, Australia. Fisheries Research, 129–130, 38–45.
Holyoak, M., Casagrandi, R., Nathan, R., Revilla, E. & Spiegel, O. (2008). Trends and
missing parts in the study of movement ecology. Proceedings of the National
Academy of Sciences of the United States of America, 105, 19060–19065.
Hooker, S.K., Cansadas, A., Hyrenbach, K.D., Corrigan, C., Polovina, J.J. & Reeves,
R.R. (2011). Making protected area networks effective for marine top predators.
Endangered Species Research, 13, 203–218.
Hooker, S.K., Whitehead, H. & Gowans, S. (1999). Marine protected area design and
the spatial and temporal distribution of cetaceans in a submarine canyon.
173
Conservation Biology, 13, 592–602.
Horne, J.S., Garton, E.O., Krone, S.M. & Lewis, J.S. (2007). Analyzing animal
movements using Brownian bridges. Ecology, 88, 2354–63.
Howey-Jordan, L.A., Brooks, E.J., Abercrombie, D.L., Jordan, L.K.B., Brooks, A.,
Williams, S., Gospodarczyk, E. & Chapman, D.D. (2013). Complex movements,
philopatry and expanded depth range of a severely threatened pelagic shark, the
oceanic whitetip (Carcharhinus longimanus) in the Western North Atlantic. PLoS
ONE, 8.
Hoyos-Padilla, E.M., Klimley, A.P., Galván-Magaña, F. & Antoniou, A. (2016).
Contrasts in the movements and habitat use of juvenile and adult white sharks
(Carcharodon carcharias) at Guadalupe Island, Mexico. Animal Biotelemetry, 4,
14.
Hughes, T.P., Bellwood, D.R. & Connolly, S.R. (2002). Biodiversity hotspots, centers
of endemicity, and the conservation of coral reefs. Ecology Letters, 5, 775–784.
Humphries, N.E., Queiroz, N., Dyer, J.R.M., Pade, N.G., Musyl, M.K., Schaefer, K.M.,
Fuller, D.W., Brunnschweiler, J.M., Doyle, T.K., Houghton, J.D.R., Hays, G.C.,
Jones, C.S., Noble, L.R., Wearmouth, V.J., Southall, E.J. & Sims, D.W. (2010).
Environmental context explains Lévy and Brownian movement patterns of marine
predators. Nature, 465, 1066–1069.
Hussey, N.E., Brush, J., McCarthy, I.D. & Fisk, A.T. (2010). δ15
N and δ13
C diet-tissue
discrimination factors for large sharks under semi-controlled conditions.
Comparative Biochemistry and Physiology - A Molecular and Integrative
Physiology, 155, 445–453.
Hussey, N.E., Dudley, S.F.J., Mccarthy, I.D., Cliff, G. & Fisk, A.T. (2011). Stable
isotope profiles of large marine predators : viable indicators of trophic position,
diet, and movement in sharks? Canadian Journal of Fisheries and Aquatic
Sciences, 68, 2029–2045.
Hussey, N.E., Kessel, S.T., Aarestrup, K., Cooke, S.J., Cowley, P.D., Fisk, A.T.,
Harcourt, R.G., Holland, K.N., Iverson, S.J., Kocik, J.F., Mills Flemming, J.E. &
Whoriskey, F.G. (2015a). Aquatic animal telemetry: A panoramic window into the
underwater world. Science (New York, N.Y.), 348, 1255642-.
Hussey, N.E., MacNeil, M.A., Siple, M.C., Popp, B.N., Dudley, S.F.J. & Fisk, A.T.
(2015b). Expanded trophic complexity among large sharks. Food Webs, 4, 1–7.
Huveneers, C., Holman, D., Robbins, R., Fox, A., Endler, J.A. & Taylor, A.H. (2015).
White sharks exploit the sun during predatory approaches. The American
Naturalist, 185, 562–570.
ICCAT. (2014). Report of the standing committee on research and statistics (SCRS).
International Commission for the Conservation of Atlantic Tuna, Madrid,
Spain.
Iosilevskii, G., Papastamatiou, Y.P., Meyer, C.G. & Holland, K.N. (2012). Energetics of
the yo-yo dives of predatory sharks. Journal of theoretical biology, 294, 172–81.
IUCN Species Survival Commission. (2012). IUCN Red List Categories and Criteria:
Version 3.1. IUCN, Gland, Switzerland and Cambridge, UK.
Jackson, A.L., Inger, R., Parnell, A.C. & Bearhop, S. (2011). Comparing isotopic niche
widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses
in R. Journal of Animal Ecology, 80, 595–602.
174
Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K. A., Botsford, L.W., Bourque,
B.J., Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J. A., Hughes, T.P., Kidwell,
S., Lange, C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S.,
Tegner, M.J. & Warner, R.R. (2001). Historical overfishing and the recent collapse
of coastal ecosystems. Science, 293, 629–37.
Jackson, A.A., Parnell, A. & Jackson, M.A. (2016). Package ‘ SIBER ’: Stable Isotope
Bayesian Ellipses in R. R package version 2.0.2.
Jit, R.B., Ali, M.H., Singha, N.K., Rahman, G. & Alam, F. (2014). Sharks and rays
fisheries of the Bay of Bengal at the landing centers of Chittagong and Cox’s
Bazar, Bangladesh. Global Journal of Science Frontier Research: D, 14, 79–89.
Johnson, D.S. (2014). Fit continuous-time correlated random walk models to animal
movement data. R package version 1.4-1
Johnson, R., Bester, M.N., Dudley, S.F.J., Oosthuizen, W.H., Meÿer, M., Hancke, L. &
Gennari, E. (2009). Coastal swimming patterns of white sharks (Carcharodon
carcharias) at Mossel Bay, South Africa. Environmental Biology of Fishes, 85,
189–200.
Johnson, C.N., Isaac, J.L. & Fisher, D.O. (2007). Rarity of a top predator triggers
continent-wide collapse of mammal prey: dingoes and marsupials in Australia.
Proceedings. Biological sciences / The Royal Society, 274, 341–346.
Johnson, A. R., Wiens, J. a., Milne, B.T.I. & Crist, T.O. (1992). Animal movements and
population dynamics in heterogeneous landscapes. Landscape Ecology, 7, 63–75.
Jones, M.C. & Cheung, W.W.L. (2015). Multi-model ensemble projections of climate
change effects on global marine biodiversity. ICES Journal of Marine Science, 72,
741–752.
Jonker, F.C. & Bester, M.N. (1998). Seasonal movements and foraging areas of adult
southern elephant seals, Mirounga leonina, from Marion Island. Antarctic Science,
10, 21–30.
Jonsen, I.D., Basson, M., Bestley, S., Bravington, M.V., Patterson, T.A., Pedersen,
M.W., Thomson, R., Thygesen, U.H. & Wotherspoon, S.J. (2013). State-space
models for bio-loggers: A methodological road map. Deep Sea Research Part II:
Topical Studies in Oceanography.
Jonsen, I., Bestley, S., Wotherspoon, S. & Summer, M. (2016). Bayesian state-space
models for animal movement.
Jonsen, I.D., Myers, R.A. & James, M.C. (2007). Identifying leatherback turtle foraging
behaviour from satellite telemetry using a switching state-space model. Marine
Ecology Progress Series, 337, 255–264.
Jorgensen, S.J., Reeb, C.A., Chapple, T.K., Anderson, S., Perle, C., Van Sommeran,
S.R., Fritz-Cope, C., Brown, A.C., Klimley, A.P. & Block, B.A. (2010). Philopatry
and migration of Pacific white sharks. Proceedings of the Royal Society B, 277,
679–688.
Kajiura, S.M. & Tellman, S.L. (2016). Quantification of massive seasonal aggregations
of blacktip sharks (Carcharhinus limbatus) in Southeast Florida. PloS one, 11,
e0150911.
Kareiva, P. & Odell, G. (1987). Swarms of predators exhibit ‘preytaxis’ if individual
predators use area-restricted search. The American Naturalist, 130, 233–270.
Kerstetter, D.W., Polovina, J.J. & Graves, J.E. (2004). Evidence of shark predation and
175
scavenging on fishes equipped with pop-up satellite archival tags. Fishery Bulletin,
102, 750–756.
Kessel, S.T., Chapman, D.D., Franks, B.R., Gedamke, T., Gruber, S.H., Newman, J.M.,
White, E.R. & Perkins, R.G. (2014). Predictable temperature-regulated residency,
movement and migration in a large, highly mobile marine predator (Negaprion
brevirostris). Marine Ecology Progress Series, 514, 175–190.
Ketchum, J.T., Hearn, A., Klimley, A.P., Espinoza, E., Peñaherrera, C. & Largier, J.L.
(2014a). Seasonal changes in movements and habitat preferences of the scalloped
hammerhead shark (Sphyrna lewini) while refuging near an oceanic island. Marine
Biology, 161, 755–767.
Ketchum, J.T., Hearn, A., Klimley, A.P., Peñaherrera, C., Espinoza, E., Bessudo, S.,
Soler, G. & Arauz, R. (2014b). Inter-island movements of scalloped hammerhead
sharks (Sphyrna lewini) and seasonal connectivity in a marine protected area of the
eastern tropical Pacific. Marine Biology, 161, 939–951.
Kie, J. (1999). Optimal foraging and risk of predation: Effects on behavior and social
structure in ungulates. Journal of Mammalogy, 80, 1114–1129.
Kim, S.L., del Rio, C.M., Casper, D. & Koch, P.L. (2012). Isotopic incorporation rates
for shark tissues from a long-term captive feeding study. Journal of Experimental
Biology, 215, 2495–2500.
Kirkwood, R., Lynch, M., Gales, N., Dann, P. & Sumner, M. (2006). At-sea movements
and habitat use of adult male Australian fur seals (Arctocephalus pusillus
doriferus). Canadian Journal of Zoology, 84, 1781–1788.
Kiszka, J.J., Aubail, A., Hussey, N.E., Heithaus, M.R., Caurant, F. & Bustamante, P.
(2015). Plasticity of trophic interactions among sharks from the oceanic south-
western Indian Ocean revealed by stable isotope and mercury analyses. Deep-Sea
Research Part I: Oceanographic Research Papers, 96, 49–58.
Kitagawa, T., Boustany, A.M., Farwell, C.J., Williams, T.D., Castleton, M.R. & Block,
B.A. (2007). Horizontal and vertical movements of juvenile bluefin tuna (Thunnus
orientalis) in relation to seasons and oceanographic conditions in the eastern
Pacific Ocean. Fisheries Oceanography, 16, 409–421.
Kitagawa, T., Kimura, S. & Nakata, H. (2004). Diving behavior of immature, feeding
Pacific bluefin tuna (Thunnus thynnus orientalis) in relation to season and area: the
East China Sea and the Kuroshio – Oyashio transition region. Ocean Research, 13,
161–180.
Klimley, A.P. (1993). Highly directional swimming by scalloped hammerhead sharks,
Sphyrna lewini, and subsurface irradiance, temperature, bathymetry, and
geomagnetic field. Marine Biology, 117, 1–22.
Klimley, P. (1994). The predatory behavior of the white shark. American Scientist, 82,
122–133.
Klimley, P.A., Le Boeuf, B.J., Cantara, K.M., Davis, S.F. & Van Sommeran, S.
(2001a). Radio-acoustic positioning as a tool for studying site-specific behavior of
the white shark and other large marine species. Marine Biology, 138, 429–446.
Klimley, A. P., Le Boeuf, B.J., Cantara, K.M., Richert, J.E., Davis, S.F., Van
Sommeran, S. & Kelly, J.T. (2001b). The hunting strategy of white sharks
(Carcharodon carcharias) near a seal colony. Marine Biology, 138, 617–636.
Klimley, A.P. & Butler, S.B. (1988). Immigration and emigration of a pelagic fish
176
assemblage to seamounts in the Gulf of California related to water mass
movements using satellite imagery. Marine Ecology Progress Series, 49, 11–20.
Kneebone, J., Natanson, L.J., Andrews, A.H. & Howell, W.H. (2008). Using bomb
radiocarbon analyses to validate age and growth estimates for the tiger shark,
Galeocerdo cuvier, in the western North Atlantic. Marine Biology, 154, 423–434.
Kohler, N.E., Casey, J.G. & Turner, P.A. (1998). NMFS Cooperative shark tagging
program, 1962-93: An atlas of shark tag and recapture data. Marine Fisheries
Review, 60, 1–87.
Kohler, N.E., Turner, P.A., Hoey, J.J., Natanson, L.J. & Briggs, R. (2002). Tag and
recapture data for three pelagic shark species: blue shark (Prionace glauca),
shortfin mako (Isurus xyrinchus), and porbeagle (Lamna nasus) in the North
Atlantic Ocean. Collected Volumes Science Papers ICCAT, 54, 1231–1260.
Kotliar, N.B. & Wiens, J.A. (1990). Multiple scales of patchiness and patch structure: A
hierarchical framework for the study of heterogeneity. OIKOS, 59, 253–260.
Krogh, M. (1994). Spatial, seasonal and biological analysis of sharks caught in the New
South Wales Protective Beach Meshing Programme. Australian Journal of Marine
and Freshwater Research, 45, 1087–1106.
Kumoru, L. (2003). The Shark Longline Fishery in Papua New Guinea. Report prepared
for Billfish and bycatch research group at the 176th Meeting of the Standing
Committee on Tuna and Billfish, Mooloolaba, Australia.
Lam, C., Nielsen, A. & Sibert, J. (2010). Incorporating sea-surface temperature to the
light-based geolocation model TrackIt. Marine Ecology Progress Series, 419, 71–
84.
Lascelles, B., Notarbartolo Di Sciara, G., Agardy, T., Cuttelod, A., Eckert, S., Glowka,
L., Hoyt, E., Llewellyn, F., Louzao, M., Ridoux, V. & Tetley, M.J. (2014).
Migratory marine species: Their status, threats and conservation management
needs. Aquatic Conservation: Marine and Freshwater Ecosystems, 24, 111–127.
Last, P.R. & Stevens, J.D. (1994). Sharks and rays of Australia. CSIRO, Melbourne.
Layman, C.A., Quattrochi, J.P., Peyer, C.M. & Allgeier, J.E. (2007). Niche width
collapse in a resilient top predator following ecosystem fragmentation. Ecology
Letters, 10, 937–944.
Lea, J.S.E., Wetherbee, B.M., Queiroz, N., Burnie, N., Aming, C., Sousa, L.L.,
Mucientes, G.R., Humphries, N.E., Harvey, G.M., Sims, D.W. & Shivji, M.S.
(2015). Repeated, long-distance migrations by a philopatric predator targeting
highly contrasting ecosystems. Scientific Reports, 5, 11202.
Lewison, R.L., Crowder, L.B., Wallace, B.P., Moore, J.E., Cox, T., Zydelis, R.,
McDonald, S., DiMatteo, A., Dunn, D.C., Kot, C.Y., Bjorkland, R., Kelez, S.,
Soykan, C., Stewart, K.R., Sims, M., Boustany, A., Read, A.J., Halpin, P., Nichols,
W.J. & Safina, C. (2014). Global patterns of marine mammal, seabird, and sea
turtle bycatch reveal taxa-specific and cumulative megafauna hotspots.
Proceedings of the National Academy of Sciences of the United States of America,
111, 5271-5276.
Lewison, R., Hobday, A.J., Maxwell, S., Hazen, E., Hartog, J.R., Dunn, D.C., Briscoe,
D., Fossette, S., O’Keefe, C.E., Barnes, M., Abecassis, M., Bograd, S., Bethoney,
N.D., Bailey, H., Wiley, D., Andrews, S., Hazen, L. & Crowder, L.B. (2015).
Dynamic Ocean Management: Identifying the critical ingredients of dynamic
177
approaches to ocean resource management. BioScience, 65, 486–498.
Li, Y., Hussey, N.E. & Zhang, Y. (2016a). Quantifying ontogenetic stable isotope
variation between dermis and muscle tissue of two pelagic sharks. Aquatic Biology,
25, 53–60.
Li, Y., Hussey, N.E., Zhang, Y. & Dai, X. (2016b). Urea and lipid extraction treatment
effects on δ15
N and δ13
C values in pelagic sharks. Rapid Communications in Mass
Spectrometry, 30, 1–8.
Lima, S.L. (1998). Nonlethal effects in the ecology of predator-prey interactions.
BioScience, 48, 25–34.
Lima, S.L. & Dill, L.M. (1990). Behavioral decisions made under the risk of predation:
A review and prospectus. Canadian Journal of Zoology, 68, 619–640.
Logan, J.M. & Lutcavage, M.E. (2010). Stable isotope dynamics in elasmobranch
fishes. Hydrobiologia, 644, 231–244.
Lowe, C.G., Wetherbee, B.M., Crow, G.L. & Tester, A.L. (1996). Ontogenetic dietary
shifts and feeding behavior of the tiger shark, Galeocerdo cuvier, in Hawaiian
waters. Environmental Biology of Fishes, 47, 203–211.
Lowe, C.G., Wetherbee, B.M. & Meyer, C.G. (2006). Using acoustic telemetry
monitoring techniques to quantify movement patterns and site fidelity of sharks
and giant trevally around French Frigate Shoals and Midway Atol. Atoll Research
Bulletin, 543, 281–303.
Lutcavage, M.E., Brill, R.W., Skomal, G.B., Chase, B.C., Goldstein, J.L. & Tutein, J.
(2000). Tracking adult North Atlantic bluefin tuna (Thunnus thynnus) in the
northwestern Atlantic using ultrasonic telemetry. Marine Biology, 137, 347–358.
Lyle, J., Pyne, R., Hooper, J. & Croaker, S. (1984). North Australia’s multi-species
shark fishery: A preparatory evaluation of the development of a shark fishing
industry in Northern Territory waters. North Australia’s Multi-Species Shark
Fishery, 1, 1–36.
Lyngdoh, S., Shrotriya, S., Goyal, S.P., Clements, H., Hayward, M.W. & Habib, B.
(2014). Prey preferences of the snow leopard (Panthera uncia): Regional diet
specificity holds global significance for conservation. PLoS ONE, 9, 1–11.
Macbeth, W.G., Geraghty, P.T., Peddemors, V.M. & Gray, C.A. (2009). Observer-
based study of targeted commercial fishing for large shark species in waters off
northern New South Wales. Final report to the Northern Rivers Catchment
Management Authority, Project No. IS8-9-M-2. Industry and Investment NSW-
Fisheries final report series 114, Sydney.
MacNeil, M., Skomal, G. & Fisk, A. (2005). Stable isotopes from multiple tissues
reveal diet switching in sharks. Marine Ecology Progress Series, 302, 199–206.
Madigan, D.J., Brooks, E.J., Bond, M.E., Gelsleichter, J., Howey, L.A., Abercrombie,
D.L., Brooks, A. & Chapman, D.D. (2015). Diet shift and site-fidelity of oceanic
whitetip sharks Carcharhinus longimanus along the Great Bahama Bank. Marine
Ecology Progress Series, 529, 185–197.
Malpica-Cruz, L., Herzka, S.Z., Sosa-Nishizaki, O. & Escobedo-Olvera, M.A. (2013).
Tissue-specific stable isotope ratios of shortfin mako (Isurus oxyrinchus) and white
(Carcharodon carcharias) sharks as indicators of size-based differences in
foraging habitat and trophic level. Fisheries Oceanography, 22, 429–532.
Malpica-Cruz, L., Herzka, S.Z., Sosa-Nishizaki, O., Lazo, J.P. & Trudel, M. (2012).
178
Tissue-specific isotope trophic discrimination factors and turnover rates in a
marine elasmobranch: empirical and modeling results. Canadian Journal of
Fisheries and Aquatic Sciences, 69, 551–564.
Mandelman, J.W., Cooper, P.W., Werner, T.B. & Lagueux, K.M. (2008). Shark bycatch
and depredation in the U.S. Atlantic pelagic longline fishery. Reviews in Fish
Biology and Fisheries, 18, 427–442.
Marshall, L.J. (2011). The fin blue line: Quantifying fishing mortality using shark fin
morphology. PhD Thesis, University of Tasmania, Tasmania.
Martin, T.G., Chades, I., Arcese, P., Marra, P.P., Possingham, H.P. & Norris, D.R.
(2007). Optimal conservation of migratory species. PLoS ONE, 2, e751.
Matich, P. & Heithaus, M.R. (2014). Multi-tissue stable isotope analysis and acoustic
telemetry reveal seasonal variability in the trophic interactions of juvenile bull
sharks in a coastal estuary. Journal of Animal Ecology, 83, 199–213.
Matich, P., Heithaus, M.R. & Layman, C.A. (2011). Contrasting patterns of individual
specialization and trophic coupling in two marine apex predators. The Journal of
animal ecology, 80, 294–305.
Matich, P., Heithaus, M.R. & Layman, C.A. (2010). Size-based variation in intertissue
comparisons of stable carbon and nitrogen isotopic signatures of bull sharks
(Carcharhinus leucas) and tiger sharks (Galeocerdo cuvier). Canadian Journal of
Fisheries and Aquatic Sciences, 67, 877–885.
Matsumoto, T., Saito, H. & Miyabe, N. (2005). Report of observer program for
Japanese tuna longline fishery in the Atlantic Ocean from August 2003 to January
2004. Collected Volumes Science Papers ICCAT, 58, 1694–1714.
Maxwell, S.M., Hazen, E.L., Lewison, R.L., Dunn, D.C., Bailey, H., Bograd, S.J.,
Briscoe, D.K., Fossette, S., Hobday, A.J., Bennett, M., Benson, S., Caldwell, M.R.,
Costa, D.P., Dewar, H., Eguchi, T., Hazen, L., Kohin, S., Sippel, T. & Crowder,
L.B. (2015). Dynamic ocean management: Defining and conceptualizing real-time
management of the ocean. Marine Policy, 58, 42–50.
McAuley, R. & Leary, T. (2008). Demersal gillnet and longline fisheries status report.
State of the fisheries report, 9, 225-229.
McCauley, D.J., Pinsky, M., Palumbi, S., Estes, J.A., Joyce, F. & Warner, R.R. (2015).
Marine defaunation: Animal loss in the global ocean. Science, 347.
McCauley, D.J., Young, H.S., Dunbar, R.B., Estes, J.A., Semmens, B.X. & Micheli, F.
(2012). Assessing the effects of large mobile predators on ecosystem connectivity.
Ecological Aplications, 22, 1711–1717.
McConnell, B.J., Chambers, C. & Fedak, M.A. (1992). Foraging ecology of southern
elephant seals in relation to the bathymetry and productivity of the Southern-
Ocean. Antarctic Science, 4, 393–398.
McCook, L.J., Ayling, T., Cappo, M., Choat, J.H., Evans, R.D., De Freitas, D.M.,
Heupel, M., Hughes, T.P., Jones, G.P., Mapstone, B., Marsh, H., Mills, M.,
Molloy, F.J., Pitcher, C.R., Pressey, R.L., Russ, G.R., Sutton, S., Sweatman, H.,
Tobin, R., Wachenfeld, D.R. & Williamson, D.H. (2010). Adaptive management
of the Great Barrier Reef: A globally significant demonstration of the benefits of
networks of marine reserves. Proceedings of the National Academy of Sciences of
the United States of America, 107, 18278–18285.
Médail, F. & Quézel, P. (1999). Biodiversity hotspots in the Mediterranean basin:
179
Setting global conservation priorities. Conservation Biology, 13, 1510–1513.
Meekan, M., Cappo, M., Carleton, J. & Marriott, R. (2006). Surveys of shark and fin-
fish abundance on reefs within the MOU74 box and Rowley Shoals using Baited
Remote Underwater Video Systems. Report prepared for the Australian
Government Department of the Environment and Heritage, Australian Institute of
Marine Science, Townsville.
Meyer, C.G., Anderson, J.M., Coffey, D., Hutchinson, M.R., Royer, M. A. & Holland,
K.N. (2016). Spatial dynamics of tiger sharks (Galeocerdo cuvier) around Maui
and Oahu. Report prepared for the Hawaii Department of Land and Natural
Resources, Manoa.
Meyer, C., Clark, T., Papastamatiou, Y., Whitney, N. & Holland, K. (2009). Long-term
movement patterns of tiger sharks Galeocerdo cuvier in Hawaii. Marine Ecology
Progress Series, 381, 223–235.
Meyer, C.G., O’Malley, J.M., Papastamatiou, Y.P., Dale, J.J., Hutchinson, M.R.,
Anderson, J.M., Royer, M.A. & Holland, K.N. (2014). Growth and maximum size
of tiger sharks (Galeocerdo cuvier) in Hawaii. PloS one, 9, e84799.
Meyer, C.G., Papastamatiou, Y.P. & Holland, K.N. (2010). A multiple instrument
approach to quantifying the movement patterns and habitat use of tiger
(Galeocerdo cuvier) and Galapagos sharks (Carcharhinus galapagensis) at French
Frigate Shoals, Hawaii. Marine Biology, 157, 1857–1868.
Miller, J.A. & Hanham, R.Q. (2011). Spatial nonstationarity and the scale of species–
environment relationships in the Mojave Desert, California, USA. International
Journal of Geographical Information Science, 25, 423–438.
Mills, L.S., Soulé, M.E. & Doak, D.F. (1993). The Keystone-species concept in ecology
and conservation. BioScience, 43, 219–224.
Mittermeier, R.A., Myers, N., Thomsen, J.B., da Fonseca, G.A.B. & Olivieri, S. (1998).
Biodiversity hotspots and major tropical wilderness areas: Approaches to setting
conservation priorities. Conservation Biology, 12, 516–520.
Moir Clark, J., Duffy, H., Pearce, J. & Mees, C. (2015). Update on the catch and
bycatch composition of illegal fishing in the British Indian Ocean Territory (BIOT)
and a summary of abandoned and lost fishing. Report prepared for the IOTC,
London.
Morato, T., Hoyle, S.D., Allain, V. & Nicol, S.J. (2010). Seamounts are hotspots of
pelagic biodiversity in the open ocean. Proceedings of the National Academy of
Sciences of the United States of America, 107, 9707–9711.
Morgan, A., Cooper, P.W., Curtis, T. & Burgess, G.H. (2009). Overview of the U.S.
East coast bottom longline shark fishery, 1994-2003. Marine Fisheries Review, 71,
23–38.
Mulaudzi, T.W., Hofmeyr, G.J.G., Bester, M.N., Kirkman, S.P., Pistorius, P.A., Jonker,
F.C., Makhado, A.B., Owen, J.H. & Grimbeek, R.J. (2008). Haulout site selection
by southern elephant seals at Marion Island. African Zoology, 43, 25–33.
Munroe, S.E.M., Heupel, M.R., Fisk, A.T., Logan, M. & Simpfendorfer, C.A. (2015).
Regional movement patterns of a small-bodied shark revealed by stable-isotope
analysis. Journal of Fish Biology, 86, 1567–1586.
Musyl, M.K., Brill, R.W., Curran, D.S., Fragoso, N.M., McNaughton, L.M., Nielsen,
A., Kikkawa, B.S. & Moyes, C.D. (2011). Postrelease survival, vertical and
180
horizontal movements, and thermal habitats of five species of pelagic sharks in the
central Pacific Ocean. Fishery Bulletin, 109, 341–368.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B. & Kent, J. (2000).
Biodiversity hotspots for conservation priorities. Nature, 403, 853–858.
Nakamura, I., Watanabe, Y., Papastamatiou, Y., Sato, K. & Meyer, C. (2011). Yo-yo
vertical movements suggest a foraging strategy for tiger sharks Galeocerdo cuvier.
Marine Ecology Progress Series, 424, 237–246.
Natanson, L.J., Casey, J.G., Kohler, N.E. & Colket, T. (1999). Growth of the tiger
shark, Galeocerdo cuvier, in the western North Atlantic based on tag returns and
length frequencies; and a note on the effects of tagging. Fishery Bulletin, 97, 944–
953.
Nathan, R., Getz, W.M., Revilla, E., Holyoak, M., Kadmon, R., Saltz, D. & Smouse,
P.E. (2008). A movement ecology paradigm for unifying organismal movement
research. Proceedings of the National Academy of Sciences of the United States of
America, 105, 19052–19059.
Newsome, S., del Rio, C., Bearhop, S. & Phillips, D. (2007). A niche for isotopic
ecology. Frontiers in Ecology and the Environment, 5, 429–436.
Nielsen, A. & Sibert, J.R. (2007). State – space model for light-based tracking of marine
animals. Canadian Journal of Fishery and Aquatic Sciences, 64, 1055–1068.
Northern Territory Department of Primary Industry and Fisheries. (2012). Offshore net
and line fishery status report 2010: Department of Environment conditions and
recommendations for the World Trade Organisation. Northern Territory, Australia
O’Leary, B.C., Winther-Janson, M., Bainbridge, J.M., Aitken, J., Hawkins, J.P. &
Roberts, C.M. (2016). Effective coverage targets for ocean protection.
Conservation Letters, doi:10.1111/conl.12247.
Oke, P.R. & Middleton, J.H. (2000). Topographically induced upwelling offeastern
Australia. Journal of Physical Oceanography, 30, 512–531.
Osborne, P.E., Foody, G.M. & Suárez-Seoane, S. (2007). Non-stationarity and local
approaches to modelling the distributions of wildlife. Diversity and Distributions,
13, 313–323.
Osborne, P.E. & Suárez-Seoane, S. (2002). Should data be partitioned spatially before
building large-scale distribution models? Ecological Modelling, 157, 249–259.
Pace, M., Cole, J., Carpenter, S. & Kitchell, J. (1999). Trophic cascades revealed in
diverse ecosystems. Trends in ecology & evolution, 14, 483–488.
Pante, E. & Simon-Bouhet, B. (2013). Marmap: A package for importing, plotting and
analyzing bathymetric and topographic data in R. PloS one, 8, e73051.
Papastamatiou, Y.P., Itano, D.G., Dale, J.J., Meyer, C.G. & Holland, K.N. (2010). Site
fidelity and movements of sharks associated with ocean-farming cages in Hawaii.
Marine and Freshwater Research, 61, 1366–1375.
Papastamatiou, Y.P. & Lowe, C.G. (2012). An analytical and hypothesis-driven
approach to elasmobranch movement studies. Journal of fish biology, 80, 1342–60.
Papastamatiou, Y.P., Meyer, C.G., Carvalho, F., Dale, J.J., Hutchinson, M.R. &
Holland, K.N. (2013). Telemetry and random-walk models reveal complex patterns
of partial migration in a large marine predator. Ecology, 94, 2595–2606.
Papworth, S.K., Bunnefeld, N., Slocombe, K. & Milner-Gulland, E.J. (2012).
Movement ecology of human resource users: Using net squared displacement,
181
biased random bridges and resource utilization functions to quantify hunter and
gatherer behaviour. Methods in Ecology and Evolution, 3, 584–594.
Park, T. (2007). NSW gamefish tournament monitoring – angling research tournament
monitoring program, 94. NSW Department of Primary Industries, Cronulla.
Parnell, A. & Jackson, A. (2013). SIAR: Stable Isotope Analysis in R. R package
version 4.0.2.
Paterson, R. (1990). Effects of long-term anti-shark measures on target and non-target
species in Queensland, Australia. Biological Conservation, 52, 147–159.
Patton, A., Politis, D.N. & White, H. (2009). Correction to ‘Automatic block-length
selection for the dependent bootstrap’ by D. Politis and H. White. Econometric
Reviews, 28, 372–375.
Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres Jr, F. (1998a). Fishing
down marine food webs. Science, 279, 860–863.
Pauly, D. & Palomares, M. (2005). Fishing down marine food web: It is far more
persuasive than we thought. Bulletin of Marine Science, 76, 197–211.
Pauly, D., Trites, A., Capuli, E. & Christensen, V. (1998b). Diet composition and
trophic levels of marine mammals. ICES Journal of Marine Science, 55, 467–481.
Pauly, D., Watson, R. & Alder, J. (2005). Global trends in world fisheries: Impacts on
marine ecosystems and food security. Philosophical Transactions of the Royal
Society B, 360, 5–12.
Payne, J.L., Bush, A.M., Heim, N.A., Knope, M.L. & McCauley, D.J. (2016).
Ecological selectivity of the emerging mass extinction in the oceans. Science, 353,
1284–1286.
Pendoley, K.L., Schofield, G., Whittock, P.A., Ierodiaconou, D. & Hays, G.C. (2014).
Protected species use of a coastal marine migratory corridor connecting marine
protected areas. Marine Biology, 161, 1455–1466.
Pepperell, J.G. (1992). Trends in the distribution, species composition and size of sharks
caught by gamefish anglers off south-eastern Australia, 1961-90. Australian
Journal of Marine and Freshwater Research, 43, 213–225.
Perry, A., Low, P., Ellis, J. & Rynolds, J. (2005). Climate change and distribution shifts
in marine fishes. Science, 308, 1912–1915.
Phillips, R.A., Bearhop, S., McGill, R.A.R. & Dawson, D.A. (2009). Stable isotopes
reveal individual variation in migration strategies and habitat preferences in a suite
of seabirds during the nonbreeding period. Oecologia, 160, 795–806.
Pikitch, E.K., Santora, C., Babcock, E.A., Bakun, A., Bonfil, R., Conover, D.O.,
Dayton, P., Doukakis, P., Fluharty, D., Heneman, B., Houde, E.D., Link, J.,
Livingston, P.A., Mangel, M., McAllister, M.K., Pope, J. & Sainsbury, K.J.
(2004). Ecosystem-based fishery management. Science, 305, 346 LP-347.
Pimenta, E.G., Vieira, Y.C., Rodrigues, T. & Amorim, A.F. De. (2014). Analysis of
small-size fleet fishery based in Cabo Frio city, Rio de Janeiro-Brazil (2003-2012).
Collected Volumes Science Papers ICCAT, 70, 2856–2867.
Pinaud, D. & Weimerskirch, H. (2005). Scale-dependent habitat use in a long-ranging
central place predator. Journal of Animal Ecology, 74, 852–863.
Politis, D.N. & White, H. (2004). Automatic block-length selection for the dependent
bootstrap. Econometric Reviews, 23, 53–70.
Pollard, D., Lincoln Smith, M. & Smith, A. (1996). The biology and conservation status
182
of the grey nurse shark (Carcharias taurus Rafinesque 1810) in New South Wales,
Australia. Aquatic, 6, 1–20.
Polovina, J.J. & Lau, B.B. (1993). Temporal and spatial distribution of catches of tiger
sharks, Galeocerdo cuvier, in the pelagic longline fishery around the Hawaiian
Islands. Marine Fisheries Review, 55, 1–3.
Post, D. (2002). Using stable isotopes to estimate trophic position: Models, methods,
and assumptions. Ecology, 83, 703–718.
Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J. & Montaña,
C.G. (2007). Getting to the fat of the matter: Models, methods and assumptions for
dealing with lipids in stable isotope analyses. Oecologia, 152, 179–189.
Powell, R. a. (2000). Animal home ranges and territories and home range estimators.
Research Techniques in Animal Ecology: Controversies and consequences (eds L.
Boitani & T.K. Fuller), pp. 65–110. Columbia University Press, New York.
Powney, G.D., Grenyer, R., Orme, C.D.L., Owens, I.P.F. & Meiri, S. (2010). Hot, dry
and different: Australian lizard richness is unlike that of mammals, amphibians and
birds. Global Ecology and Biogeography, 19, 386–396.
Preen, A.R., Marsh, H., Lawler, I.R., Prince, R.I.T. & Shepherd, R. (1997). Distribution
and abundance of dugongs, turtles, dolphins and other megafauna in Shark Bay,
Ningaloo Reef and Exmouth Gulf, Western Australia. Wildlife Research, 24, 185–
208.
Pyle, P., Schramm, M.J., Keiper, C. & Anderson, S. (1999). Predation on a white shark
(Carcharodon carcharias) by a killer whale (Orcinus orca). Marine Mammal
Science, 15, 563–568.
Queiroz, N., Humphries, N.E., Mucientes, G., Hammerschlag, N., Lima, F.P., Scales,
K.L., Miller, P.I., Sousa, L.L., Seabra, R. & Sims, D.W. (2016). Ocean-wide
tracking of pelagic sharks reveals extent of overlap with longline fishing hotspots.
Proceedings of the National Academy of Sciences, 113, 1582–1587.
Queiroz, N., Lima, F.P., Maia, A., Ribeiro, P.A., Correia, J.P. & M. Santos, A. (2005).
Movement of blue shark, Prionace glauca, in the north-east Atlantic based on
mark–recapture data. Journal of the Marine Biological Association of the United
Kingdom, 85, 1107–1112.
Quillfeldt, P., Bugoni, L., McGill, R.A.R., Masello, J.F. & Furness, R.W. (2008).
Differences in stable isotopes in blood and feathers of seabirds are consistent
across species, age and latitude: Implications for food web studies. Marine
Biology, 155, 593–598.
R Development Core Team (2013). R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna.
Randall, J.E. (1992). Review of the biology of the tiger shark (Galeocerdo cuvier).
Australian Journal of Marine and Freshwater Research, 43, 21–31.
Raymond, B., Lea, M.A., Patterson, T., Andrews-Goff, V., Sharples, R., Charrassin,
J.B., Cottin, M., Emmerson, L., Gales, N., Gales, R., Goldsworthy, S.D., Harcourt,
R., Kato, A., Kirkwood, R., Lawton, K., Ropert-Coudert, Y., Southwell, C., van
den Hoff, J., Wienecke, B., Woehler, E.J., Wotherspoon, S. & Hindell, M.A.
(2015). Important marine habitat off east Antarctica revealed by two decades of
multi-species predator tracking. Ecography, 38, 121–129.
Reid, J.L. (1994). On the total geostrophic circulation of the North Atlantic Ocean:
183
Flow patterns, tracers, and transports. Progress in Oceanography, 33, 1–92.
Reid, D.D., Robbins, W.D. & Peddemors, V.M. (2011). Decadal trends in shark catches
and effort from the New South Wales, Australia, Shark Meshing Program 1950-
2010. Marine and Freshwater Research, 62, 676–693.
Reid, W.D.K., Sweeting, C.J., Wigham, B.D., McGill, R.A.R. & Polunin, N.V.C.
(2016). Isotopic niche variability in macroconsumers of the East Scotia Ridge
(Southern Ocean) hydrothermal vents: What more can we learn from an ellipse?
Marine Ecology Progress Series, 542, 13–24.
Revill, A.T., Young, J.W. & Lansdell, M. (2009). Stable isotopic evidence for trophic
groupings and bio-regionalization of predators and their prey in oceanic waters off
eastern Australia. Marine Biology, 156, 1241–1253.
Reynolds, R.W., Smith, T.M., Liu, C., Chelton, D.B., Casey, K.S. & Schlax, M.G.
(2007). Daily high-resolution-blended analyses for sea surface temperature.
Journal of Climate, 20, 5473–5496.
Ripple, W.J. & Beschta, R.L. (2004). Wolves and the ecology of fear: Can predation
risk structure ecosystems? BioScience, 54, 755.
Ripple, W.J., Estes, J.A., Beschta, R.L., Wilmers, C.C., Ritchie, E.G., Hebblewhite, M.,
Berger, J., Elmhagen, B., Letnic, M., Nelson, M.P., Schmitz, O.J., Smith, D.W.,
Wallach, A.D. & Wirsing, A.J. (2014). Status and ecological effects of the world’s
largest carnivores. Science, 343, 1241484.
Ritchie, E.G. & Johnson, C.N. (2009). Predator interactions, mesopredator release and
biodiversity conservation. Ecology Letters, 12, 982–998.
Roberts, J.J., Best, B.D., Dunn, D.C., Treml, E.A. & Halpin, P.N. (2010). Marine
Geospatial Ecology Tools: An integrated framework for ecological geoprocessing
with ArcGIS, Python, R, MATLAB, and C++. Environmental Modelling and
Software, 25, 1197–1207.
Roberts, C.M., McClean, C., Veron, J., Hawkins, J., Allen, G., McAllister, D.,
Mittermeier, C.G., Schueler, F.W., Spalding, M., Wells, F., Vynne, C. & Werner,
T. (2002). Marine biodiversity hotspots and conservation priorities for tropical
reefs. Science, 295, 1280–1284.
Robinson, R.A., Crick, H.Q.P., Learmonth, J.A., Maclean, I.M.D., Thomas, C.D.,
Bairlein, F., Forchhammer, M.C., Francis, C.M., Gill, J.A., Godley, B.J., Harwood,
J., Hays, G.C., Huntley, B., Hutson, A.M., Pierce, G.J., Rehfisch, M.M., Sims,
D.W., Santos, M., Sparks, T.H., Stroud, D.A. & Visser, M.E. (2009). Travelling
through a warming world: climate change and migratory species. Endangered
Species Research, 7, 87–99.
Rodrigues, A.S.L., Pilgrim, J.D., Lamoreux, J.F., Hoffmann, M. & Brooks, T.M.
(2006). The value of the IUCN Red List for conservation. Trends in Ecology and
Evolution, 21, 71–76.
Rogers, P.J., Huveneers, C., Page, B., Goldsworthy, S.D., Coyne, M., Lowther, A.D.,
Mitchell, J.G. & Seuront, L. (2015). Living on the continental shelf edge: Habitat
use of juvenile shortfin makos Isurus oxyrinchus in the Great Australian Bight,
southern Australia. Fisheries Oceanography, 24, 205–218.
Rousseaux, C.S.G., Lowe, R., Feng, M., Waite, A.M. & Thompson, P.A. (2012). The
role of the Leeuwin Current and mixed layer depth on the autumn phytoplankton
bloom off Ningaloo Reef, Western Australia. Continental Shelf Research, 32, 22–
184
35.
Royer, F., Fromentin, J.M. & Gaspar, P. (2004). Association between bluefin tuna
schools and oceanic features in the western Mediterranean. Marine Ecology
Progress Series, 269, 249–263.
Runge, C.A., Martin, T.G., Possingham, H.P., Willis, S.G. & Fuller, R.A. (2014).
Conserving mobile species. Frontiers in Ecology and the Environment, 12, 395–
402.
Ruppert, J.L.W., Travers, M.J., Smith, L.L., Fortin, M.J. & Meekan, M.G. (2013).
Caught in the middle: combined impacts of shark removal and coral loss on the
fish communities of coral reefs. PloS ONE, 8, e74648.
Sankhala, K.S. (1967). Breeding behaviour of the tiger Panthera tigris in Rajasthan.
International Zoo Yearbook, 7, 133–147.
Saulitis, E.V.A., Matkin, C.O., Barrett-Lennard, L.G., Heise, K.A. & Ellis, G.M.
(2000). Foraging strategies of sympatric killer whale (Orcinus orca) populations in
Prince William Sound, Alaska. Marine Mammal Science, 16, 94–109.
Scales, K.L., Miller, P.I., Embling, C.B., Ingram, S.N., Pirotta, E. & Votier, S.C.
(2014). Mesoscale fronts as foraging habitats: Composite front mapping reveals
oceanographic drivers of habitat use for a pelagic seabird. Journal of the Royal
Society Interface, 11, 20140679.
Schipper, J., Chanson, J.S., Chiozza, F., Cox, N.A., Hoffmann, M., Katariya, V.,
Lamoreux, J., Rodrigues, A.S.L., Stuart, S.N., Temple, H.J., Baillie, J., Boitani, L.,
Jr, T.E.L., Mittermeier, R.A., Smith, A.T., Absolon, D., Aguiar, J.M., Amori, G.,
Bakkour, N., Baldi, R., Berridge, R.J., Bielby, J., Black, P.A., Blanc, J.J., Brooks,
T.M., Burton, J.A., Butynski, T.M., Catullo, G., Garshelis, D.L., Gates, C.,
Gimenez-dixon, M., Gonzalez, S., Gonzalez-maya, J.F., Good, T.C., Hammerson,
G., Hammond, P.S., Happold, D., Happold, M., Hare, J., Harris, R.B., Hawkins,
C.E., Haywood, M., Heaney, L.R., Hedges, S., Helgen, K.M., Hilton-taylor, C.,
Hussain, S.A., Ishii, N., Jefferson, T.A., Jenkins, R.K.B., Johnston, C.H., Keith,
M., Kingdon, J., Knox, D.H., Kovacs, K.M., Langhammer, P., Leus, K., Lewison,
R., Lichtenstein, G., Lowry, L.F., Macavoy, Z., Medellín, R.A., Medici, P., Mills,
G., Moehlman, P.D., Molur, S., Mora, A., Nowell, K., Oates, J.F., Olech, W.,
Oliver, W.R.L., Oprea, M., Patterson, B.D., Perrin, W.F., Polidoro, B.A., Pollock,
C., Powel, A., Protas, Y., Racey, P., Ragle, J., Ramani, P., Rathbun, G., Reeves,
R.R., Reilly, S.B., Iii, J.E.R., Rondinini, C., Rosell-ambal, R.G., Rulli, M.,
Rylands, A.B., Savini, S., Schank, C.J., Sechrest, W., Self-sullivan, C., Shoemaker,
A., Sillero-zubiri, C., Silva, N. De, Smith, D.E., Taylor, B.L., Timmins, R., Tirira,
D.G., Tognelli, M.F., Tsytsulina, K., Veiga, L.M., Vié, J., Williamson, E.A.,
Wyatt, S.A., Xie, Y. & Young, B.E. (2008). The status of the world’s land and
marine mammals: Diversity, threat and knowledge. Science, 322, 225–230.
Schlaff, A.M., Heupel, M.R. & Simpfendorfer, C.A. (2014). Influence of environmental
factors on shark and ray movement, behaviour and habitat use: A review. Reviews
in Fish Biology and Fisheries, 24, 1089–1103.
Schmitz, O.J., Beckmann, A.P. & O’Brien, K.M. (1997). Behaviorally mediated trophic
cascades : effects of predation risk on food web interactions. Ecology, 78, 1388–
1399.
Schmitz, W.J. & McCartney, M.S. (1993). On the North Atlantic circulation. Reviews of
185
Geophysics, 31, 29–49.
Shaffer, S.A., Tremblay, Y., Weimerskirch, H., Scott, D., Thompson, D.R., Sagar, P.M.,
Moller, H., Taylor, GA., Foley, D.G., Block, B.A. & Costa, D.P. (2006). Migratory
shearwaters integrate oceanic resources across the Pacific Ocean in an endless
summer. Proceedings of the National Academy of Sciences of the United States of
America, 103, 12799–12802.
Sherwood, G.D. & Rose, G. a. (2005). Stable isotope analysis of some representative
fish and invertebrates of the Newfoundland and Labrador continental shelf food
web. Estuarine, Coastal and Shelf Science, 63, 537–549.
Shuter, J.L., Broderick, A.C., Agnew, D.J., Jonzen, N., Godley, B.J., Milner-Gulland,
E.J. & Thirgood, S. (2011). Conservation and management of migratory species.
Animal migration: A synthesis (eds E.J. Milner-Gulland, J.M. Fryxell & A.R.E.
Sinclair), pp. 172–206. Oxford Universoty Press, New York.
da Silva, C., Kerwath, S.E., Wilke, C.G., Meÿer, M. & Lamberth, S.J. (2010). First
documented southern transatlantic migration of a blue shark Prionace glauca
tagged off South Africa. African Journal of Marine Science, 32, 639–642.
Simpfendorfer, C.A. (1992). Biology of tiger sharks (Galeocerdo cuvier) caught by the
Queensland Shark Meshing Program off Townsville, Australia. Marine and
Freshwater Research, 43, 33.
Simpfendorfer, C.A. (2009). Galeocerdo cuvier. IUCN Red List of Threatened Species.
URL http://www.iucnredlist.org/details/39378/0
Simpfendorfer, C.A., Goodreid, A.B. & McAuley, R.B. (2001). Size, sex and
geographic variation in the diet of the tiger shark, Galeocerdo cuvier, from
Western Australian waters. Environmental Biology of Fishes, 67, 37–46.
Simpfendorfer, C.A., de Jong, S.K. & Sumpton, W. (2010). Long-term trends in large
shark populations from inshore areas of the Great Barrier Reef World
Heritage Area: results from the Queensland Shark Control Program. Marine
and Tropical Sciences Research Facility Transition Program Report,
Townsville.
Sims, D.W. (2010). Tracking and Analysis Techniques for understanding free-ranging
shark movements and behavior. Sharks and their relatives II: biodiversity,
adaptive physiology, and conservation (eds J.C. Carrier, J.A. Musick & M.R.
Heithaus), pp. 351–386. CRC Marine Biology Series, Boca Raton.
Sims, D.W., Southall, E.J., Humphries, N.E., Hays, G.C., Bradshaw, C.J.A., Pitchford,
J.W., James, A., Ahmed, M.Z., Brierley, A.S., Hindell, M.A., Morritt, D., Musyl,
M.K., Righton, D., Shepard, E.L.C., Wearmouth, V.J., Wilson, R.P., Witt, M.J. &
Metcalfe, J.D. (2008). Scaling laws of marine predator search behaviour. Nature
Letters, 451, 1098–1102.
Sinclair, A.R.E. & Arcese, P. (1995). Population consequences of predation-sensitive
foraging: The Serengeti wildebeest. Ecology, 76, 882–891.
Sinclair, A.R.E., Mduma, S. & Brashares, J.S. (2003). Patterns of predation in a diverse
predator-prey system. Nature, 425, 288–290.
Singh, R., Majumder, A., Sankar, K., Qureshi, Q., Goyal, S.P. & Nigam, P. (2013).
Interbirth interval and litter size of free-ranging Bengal tiger (Panthera tigris
tigris) in dry tropical deciduous forests of India. European Journal of wildlife
researh, 59, 629–636.
186
Sjöberg, M. & Ball, J.P. (2000). Grey seal, Halichoerus grypus, habitat selection around
haulout sites in the Baltic Sea: Bathymetry or central-place foraging? Canadian
Journal of Zoology, 78, 1661–1667.
Skomal, G.B., Zeeman, S.I., Chisholm, J.H., Summers, E.L., Walsh, H.J., McMahon,
K.W. & Thorrold, S.R. (2009). Transequatorial migrations by basking sharks in the
western Atlantic Ocean. Current biology, 19, 1019–22.
Smith, J.L.D., McDougal, C. & Miquelle, D. (1989). Scent marking in free-ranging
tigers, Panthera tigris. Animal Behaviour, 37, 1–10.
Smith, S.E., Au, D.W. & Show, C. (1998). Intrinsic rebound potentials of 26 species of
Pacific sharks. Marine and Freshwater Research, 49, 663–678.
Smith, R.L., Huyer, A., Godfrey, J.S. & Church, J.A. (1991). The Leeuwin Current off
Western Australia, 1986–1987. Journal of Physical Oceanography, 21.
Sousa, B.P. (2012). Sharks caught in Mozambican waters. 8th Session of the Working
Party on Ecosystems and Bycatch, IOTC, Cape Town.
Spaet, J.L.Y. & Berumen, M.L. (2015). Fish market surveys indicate unsustainable
elasmobranch fisheries in the Saudi Arabian Red Sea. Fisheries Research, 161,
356–364.
Speed, C., Field, I., Meekan, M. & Bradshaw, C. (2010). Complexities of coastal shark
movements and their implications for management. Marine Ecology Progress
Series, 408, 275–293.
Speed, C.W., Meekan, M.G., Field, I.C., Mcmahon, C.R., Abrantes, K. & Bradshaw,
C.J.A. (2012). Trophic ecology of reef sharks determined using stable isotopes and
telemetry. Coral reefs, 31, 357–367.
Speed, C.W., Meekan, M.G., Field, I.C., McMahon, C.R., Harcourt, R.G., Stevens, J.D.,
Babcock, R.C., Pillans, R.D. & Bradshaw, C.J.A. (2016). Reef shark movements
relative to a coastal marine protected area. Regional Studies in Marine Science, 3,
58–65.
Spies, R., Kruger, H., Ireland, R. & Rice, D. (1989). Stable isotope ratios and
contaminant concentrations in a sewage-distorted food web. Marine Ecology
Progress Series, 54, 157–170.
Stevens, J.D. (1984). Biological observations on sharks caught by sport fishermen off
New South Wales. Australian Journal of Marine and Freshwater Research, 35,
573–590.
Stevens, J.D., Bonfil, R., Dulvy, N.K. & Walker, P.A. (2000). The effects of fishing on
sharks, rays, and chimaeras (chondrichthyans), and the implications for marine
ecosystems. ICES Journal of Marine Science, 57, 476–494.
Stevens, J., Davis, T. & Church, A. (1982). NT shark gillnetting survey shows potential
for Australian fishermen. Australian Fisheries, 41, 39–43.
Stevens, J.D. & McLoughlin, K.J. (1991). Distribution, size and sex composition,
reproductive biology and diet of sharks from Northern Australia. Australian
Journal of Marine and Freshwater Research, 42, 151–199.
Sulikowski, J.A., Wheeler, C.R., Gallagher, A.J., Prohaska, B.K., Langan, J.A. &
Hammerschlag, N. (2016). Seasonal and life-stage variation in the reproductive
ecology of a marine apex predator, the tiger shark Galeocerdo cuvier, at a
protected female-dominated site. Aquatic Biology, 24, 175–184.
Suthers, I.M., Young, J.W., Baird, M.E., Roughan, M., Everett, J.D., Brassington, G.B.,
187
Byrne, M., Condie, S.A., Hartog, J.R., Hassler, C.S., Hobday, A.J., Holbrook, N.J.,
Malcolm, H.A., Oke, P.R., Thompson, P.A. & Ridgway, K. (2011). The
strengthening East Australian Current, its eddies and biological effects-An
introduction and overview. Deep-Sea Research Part II: Topical Studies in
Oceanography, 58, 538–546.
Sydeman, W.J., Brodeur, R.D., Grimes, C.B., Bychkov, A.S. & McKinnell, S. (2006).
Marine habitat ‘hotspots’ and their use by migratory species and top predators in
the North Pacific Ocean: Introduction. Deep Sea Research Part II: Topical Studies
in Oceanography, 53, 247–249.
Taylor, J.G. & Pearce, A.F. (1999). Ningaloo Reef currents: implications for coral
spawn dispersal, zooplankton and whale shark abundance. Journal of the Royal
Society of Western Australia, 82, 57–65.
Tegner, M. & Levin, L. (1983). Spiny lobsters and sea urchins: Analysis of a predator-
prey interaction. Journal of Experimental Marine Biology and Ecology, 73, 125–
150.
Thaker, M., Vanak, A.T., Owen, C.R., Ogden, M.B., Niemann, S.M. & Slotow, R.
(2011). Minimizing predation risk in a landscape of multiple predators: Effects on
the spatial distribuition of African ungulates. Ecology, 92, 398–407.
Thaker, M., Vanak, A.T., Owen, C.R., Ogden, M.B. & Slotow, R. (2010). Group
dynamics of zebra and wildebeest in a woodland savanna: Effects of predation risk
and habitat density. PLoS ONE, 5, e12758.
Thomson, J., Heithaus, M., Burkholder, D., Vaudo, J., Wirsing, A. & Dill, L. (2012).
Site specialists, diet generalists? Isotopic variation, site fidelity, and foraging by
loggerhead turtles in Shark Bay, Western Australia. Marine Ecology Progress
Series, 453, 213–226.
Thums, M., Bradshaw, C.J.A. & Hindell, M.A. (2011). In situ measures of foraging
success and prey encounter reveal marine habitat-dependent search strategies.
Ecology, 92, 1258–1270.
Thums, M., Meekan, M., Stevens, J., Wilson, S. & Polovina, J. (2012). Evidence for
behavioural thermoregulation by the world’s largest fish. Journal of the Royal
Society, Interface / the Royal Society, 2–6.
Tillett, B.J., Field, I.C., Bradshaw, C.J.A., Johnson, G., Buckworth, R.C., Meekan,
M.G. & Ovenden, J.R. (2012). Accuracy of species identification by fisheries
observers in a north Australian shark fishery. Fisheries Research, 127–128, 109–
115.
Tranter, D., Carpenter, D. & Leech, G. (1986). The coastal enrichmennt effects of the
East Australian Current eddy field. Deep Sea Research Part I: Oceanographic
Research Papers, 33, 1705–1728.
Tricas, T.C., Taylor, L.R. & Naftel, G. (1981). Diel behaviour of the tiger shark
Galeocerdo cuvier at French Frigate Shoals, Hawaiin Islands.pdf. Copeia, 4, 904–
908.
Trystram, C., Rogers, K., Soria, M. & Jaquemet, S. (2016). Feeding patterns of two
sympatric shark predators in coastal ecosystems of an oceanic island. Canadian
Journal of Fisheries and Aquatic Sciences, cjfas-2016-0105.
Turchin, P. (1991). Translating foraging movements in heterogeneous environments
into the spatial distribution of foragers. Ecology, 72, 1253–1266.
188
Ullman, D.S. & Cornillon, P.C. (1999). Satellite-derived sea surface temperature fronts
on the continental shelf off the northeast U.S. coast. Journal of Geophysical
Research, 104, 23.459-23.478.
Urton, E.J.M. & Hobson, K.A. (2005). Intrapopulation variation in gray wolf isotope
(δ15
N and δ13
C) profiles: Implications for the ecology of individuals. Oecologia,
145, 317–326.
Valeix, M., Fritz, H., Loveridge, A.J., Davidson, Z., Hunt, J.E., Murindagomo, F. &
Macdonald, D.W. (2009a). Does the risk of encountering lions influence African
herbivores at waterholes? Behavioral Ecology and Sociobiology, 63, 1483–1494.
Valeix, M., Loveridge, A.J., Chamaille-Jammes, S., Davidson, Z., Murindagomo, F.,
Fritz, H. & Macdonald, D.W. (2009b). Behavioral adjustments of African
herbivores to predation risk by lions: Spatiotemporal variations influence habitat
use. Ecology, 90, 23–30.
Van Valen, L. (1965). Morphological variation and width of ecological niche. The
American Naturalist, 99, 377–390.
Vandeperre, F., Aires-da-Silva, A., Santos, M., Ferreira, R., Bolten, A.B., Serrao
Santos, R. & Afonso, P. (2014). Demography and ecology of blue shark (Prionace
glauca) in the central North Atlantic. Fisheries Research, 153, 89–102.
Vander Zanden, M.J. & Rasmussen, J.B. (2001). Variation in δ15
N and δ13
C trophic
fractionation: Implications for aquatic food web studies. Limnology and
Oceanography, 46, 2061–2066.
Vaudo, J. & Heithaus, M. (2011). Dietary niche overlap in a nearshore elasmobranch
mesopredator community. Marine Ecology Progress Series, 425, 247–260.
Vaudo, J.J., Wetherbee, B.M., Harvey, G., Nemeth, R.S., Aming, C., Burnie, N.,
Howey-Jordan, L.A. & Shivji, M.S. (2014). Intraspecific variation in vertical
habitat use by tiger sharks (Galeocerdo cuvier) in the western North Atlantic.
Ecology and Evolution, 1–19.
Vaudo, J.J., Wetherbee, B.M., Wood, A.D., Weng, K., Howey-Jordan, L.A., Harvey,
G.M. & Shivji, M.S. (2016). Vertical movements of shortfin mako sharks Isurus
oxyrinchus in the western North Atlantic Ocean are strongly influenced by
temperature. Marine Ecology Progress Series, 547, 163–175.
Vianna, G.M.S., Meekan, M.G., Meeuwig, J.J. & Speed, C.W. (2013). Environmental
influences on patterns of vertical movement and site fidelity of grey reef sharks
(Carcharhinus amblyrhynchos) at aggregation sites. PloS one, 8, e60331.
Viswanathan, G. M., Afanasyev, V., Buldyrev, S. V, Havlin, S., da Luz, M. G. E.,
Raposo, E. P., & Stanley, H. E. (2000). Lévy flights in random searches. Physica
A, 282, 1–12.
Wachenfeld, D.R., Oliver, J. & Morrissey, J.I. (1998). State of the Great Barrier Reef
World Heritage Area 1998. Great Barrier Reef Marine Park Authority, Townsville.
Walker, D.I., Kendrick, G.A. & Mccomb, A.J. (1988). The distribution of seagrass
species in Shark Bay, Western Australia, with notes on their ecology. Aquatic
Biology, 30, 305–317.
Wallach, A.D., Izhaki, I., Toms, J.D., Ripple, W.J. & Shanas, U. (2015). What is an
apex predator? Oikos, 124, 1453–1461.
Walli, A., Teo, S.L.H., Boustany, A., Farwell, C.J., Williams, T., Dewar, H., Prince, E.
& Block, B.A. (2009). Seasonal movements, aggregations and diving behavior of
189
Atlantic bluefin tuna (Thunnus thynnus) revealed with archival tags. PLoS ONE, 4.
Ward-Paige, C.A., Mora, C., Lotze, H.K., Pattengill-Semmens, C., McClenachan, L.,
Arias-Castro, E. & Myers, R.A. (2010). Large-scale absence of sharks on reefs in
the greater-Caribbean: A footprint of human pressures. PloS one, 5, e11968.
Webster, M.S., Marra, P.P., Haig, S.M., Bensch, S. & Holmes, R.T. (2002). Links
between worlds: Unraveling migratory connectivity. Trends in Ecology and
Evolution, 17, 76–83.
Weimerskirch, H., Guionnet, T., Martin, J., Shaffer, S.A. & Costa, D.P. (2000). Fast and
fuel efficient? Optimal use of wind by flying albatrosses. Proceedings of the Royal
Society B, 267, 1869–1874.
Weng, K.C., Boustany, A.M., Pyle, P., Anderson, S.D., Brown, A. & Block, B.A.
(2007a). Migration and habitat of white sharks (Carcharodon carcharias) in the
eastern Pacific Ocean. Marine Biology, 152, 877–894.
Weng, K.C., Sullivan, J.B.O., Lowe, C.G., Winkler, C.E., Dewar, H. & Block, B.A.
(2007b). Movements, behavior and habitat preferences of juvenile white sharks
Carcharodon carcharias in the eastern Pacific. Marine Ecology Progress Series,
338, 211–224.
Werner, E.E., Gilliam, J.F., Hall, D.J. & Mittelbach, G.G. (1983). An experimental test
of the effects of predation risk on habitat use in fish. Ecology, 64, 1540–1548.
Werry, J.M., Planes, S., Berumen, M.L., Lee, K. a., Braun, C.D. & Clua, E. (2014).
Reef-fidelity and migration of tiger sharks, Galeocerdo cuvier, across the Coral
Sea (A.P. Klimley, Ed.). PLoS ONE, 9, e83249.
Wetherbee, B.M., Crow, G.L. & Lowe, C.G. (1997). Distribution, reproduction and diet
of the gray shark Carcharhinus amblyrhinchus in Hawaii. Marine Ecology
Progress Series, 151, 181–189.
Wetherbee, B.M., Lowe, C.G. & Crow, G.L. (1994). A review of shark control in
Hawaii with recommendations for future research. Pacific Science, 48, 95–115.
White, W.T. (2007). Catch composition and reproductive biology of whaler sharks
(Carcharhiniformes: Carcharhinidae) caught by fisheries in Indonesia. Journal of
Fish Biology, 71, 1512–1540.
Whitney, N.M. & Crow, G.L. (2007). Reproductive biology of the tiger shark
(Galeocerdo cuvier) in Hawaii. Marine Biology, 151, 63–70.
Wilson, S.G., Stewart, B.S., Polovina, J.J., Meekan, M.G., Stevens, J.D. & Galuardi, B.
(2007). Accuracy and precision of archival tag data: A multiple-tagging study
conducted on a whale shark (Rhincodon typus) in the Indian Ocean. Fisheries
Oceanography, 16, 547–554.
Windle, M., Rose, G.A., Devillers, R. & Fortin, M.-J. (2009). Exploring spatial non-
stationarity of fisheries survey data using geographically weighted regression
(GWR): An example from the Northwest Atlantic. ICES Journal of Marine
Science, 67, 1–10.
Wintner, S.P. & Dudley, S.F.J. (2000). Age and growth estimates for the tiger shark,
Galeocerdo cuvier, from the east coast of South Africa. Marine and Freshwater
Research, 51, 43–53.
Wirsing, A.J., Heithaus, M.R. & Dill, L.M. (2007a). Fear factor : Do dugongs (Dugong
dugon) trade food for safety from tiger sharks (Galeocerdo cuvier)? Oecologia,
153, 1031–1040.
190
Wirsing, A.J., Heithaus, M.R. & Dill, L.M. (2007b). Living on the edge: dugongs prefer
to forage in microhabitats that allow escape from rather than avoidance of
predators. Animal Behaviour, 74, 93–101.
Wirsing, A.J., Heithaus, M.R. & Dill, L.M. (2006). Tiger shark (Galeocerdo cuvier)
abundance and growth in a subtropical embayment: evidence from 7 years of
standardized fishing effort. Marine Biology, 149, 961–968.
Wirsing, A.J. & Ripple, W.J. (2011). A comparison of shark and wolf research reveals
similar behavioral responses by prey. Frontiers in Ecology and the Environment, 9,
335–341.
Woo, K.J., Elliott, K.H., Davidson, M., Gaston, A.J. & Davoren, G.K. (2008).
Individual specialization in diet by a generalist marine predator reflects
specialization in foraging behaviour. Journal of Animal Ecology, 77, 1082–1091.
Wood, S. (2016). Mixed GAM Computation Vehicle with GCV/AIC/REML
Smoothness Estimation. R package version 1.8-15.
Woodley, T.H. & Gaskin, D.E. (1996). Environmental characteristics of North Atlantic
right and fin whale habitat in the lower Bay of Fundy, Canada. Canadian Journal
of Zoology, 74, 75–84.
Worm, B., Davis, B., Kettemer, L., Ward-Paige, C.A., Chapman, D., Heithaus, M.R.,
Kessel, S.T. & Gruber, S.H. (2013). Global catches, exploitation rates, and
rebuilding options for sharks. Marine Policy, 40, 194–204.
Worton, B.J. (1987). A review of models of home range for animal movement.
Ecological Modelling, 38, 277–298.
Wyatt, A.S.J., Waite, A.M. & Humphries, S. (2012). Stable isotope analysis reveals
community-level variation in fish trophodynamics across a fringing coral reef.
Coral Reefs, 31, 1029–1044.
Yates, K.L., Mellin, C., Caley, M.J., Radford, B.T. & Meeuwig, J.J. (2016). Models of
marine fish biodiversity: Assessing predictors from three habitat classification
schemes. PloS ONE, 11, e0155634.
Zamora, L.M., Virtue, P., Pethybridge, H.R., Meekan, M.G., Thums, M. & Nichols,
P.D. (2016). Intraspecific variability in diet and implied foraging ranges of whale
sharks at Ningaloo Reef, Western Australia. Marine Ecology Progress Series, 554,
115–128.
191
Appendix A
192
Appendix B
193
Appendix C