Content uploaded by Fabrice R. A. Jaine
Author content
All content in this area was uploaded by Fabrice R. A. Jaine on Sep 10, 2014
Content may be subject to copyright.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 510: 73– 86, 2014
doi: 10.3354/meps10910 Published September 9
INTRODUCTION
Understanding patterns of habitat use in vulnera-
ble, large and highly mobile marine species is crucial
to implementing effective, spatially explicit manage-
ment strategies (Block et al. 2005, Graham et al.
2012). This is especially true for plankton-feeding
elasmobranchs that depend on their ability to locate
minute and diffuse prey in a vast, dynamic and
changing ocean (Richardson & Schoeman 2004, Sims
et al. 2006). Planktivorous elasmobranchs occur at
low sub-population levels and for limited periods at
© The authors 2014. Open Access under Creative Commons by
Attribution Licence. Use, distribution and reproduction are un -
restricted. Authors and original publication must be credited.
Publisher: Inter-Research · www.int-res.com
*Corresponding author: fabrice@marinemegafauna.org
Movements and habitat use of reef manta rays off
eastern Australia: offshore excursions, deep diving
and eddy affinity revealed by satellite telemetry
F. R. A. Jaine1,2, 3,*, C. A. Rohner1,2,3, S. J. Weeks1, L. I. E. Couturier2,4,
M. B. Bennett4, K. A. Townsend5,6, A. J. Richardson2, 7
1Biophysical Oceanography Group, School of Geography, Planning and Environmental Management,
The University of Queensland, St Lucia, Queensland 4072, Australia
2Climate Adaptation Flagship, CSIRO Marine and Atmospheric Research, Ecosciences Precinct, Dutton Park,
Queensland 4102, Australia
3Manta Ray and Whale Shark Research Centre, Marine Megafauna Foundation, Praia do Tofo, Inhambane, Mozambique
4School of Biomedical Sciences, and 5School of Biological Sciences, The University of Queensland, St Lucia, Queensland
4072, Australia
6Moreton Bay Research Station, The University of Queensland, Dunwich, Queensland 4183, Australia
7Centre for Applications in Natural Resource Mathematics, School of Mathematics and Physics,
The University of Queensland, St Lucia, Queensland 4072, Australia
ABSTRACT: Manta rays (Manta spp.) are plankton-feeding elasmobranchs classified as vulnera-
ble to extinction on the IUCN Red List for Threatened Species. Despite increasing public and sci-
entific interest in manta rays, major knowledge gaps concerning their movement ecology and dis-
persal capabilities remain. Here, we used pop-off satellite-linked archival transmitting tags to
examine the horizontal movements and habitat use patterns of reef manta rays (M. alfredi) depart-
ing Lady Elliot Island in the southern Great Barrier Reef, Australia. Tagged individuals moved
across a latitudinal range of 1035 km, travelling up to 2441 km in 118 d, diving down to 294.5 m
and venturing up to 155 km off the continental shelf. Using random walk simulations, we showed
that manta rays spent significantly more time in an offshore region characterised by the mesoscale
cyclonic Capricorn Eddy than would be expected by chance. A behaviour-switching state-space
model suggested this area to be an important foraging ground for M. alfredi off eastern Australia.
We document the movements of 1 individual using offshore waters between 2 known aggregation
regions off eastern Australia. Reef manta rays thus not only occupy inshore continental shelf and
shelf-edge waters but also use offshore environments to exploit productive hotspots and travel
long distances. Our findings highlight the need to better understand their movement ecology for
effective management.
KEY WORDS: Eddy · East Australian Current · Manta alfredi · Movements · Oceanography ·
Random walk · Satellite tracking · State-space analysis
O
PEN
PEN
A
CCESS
CCESS
Mar Ecol Prog Ser 510: 73–86, 2014
inshore aggregation sites, where they can be readily
observed. Therefore, obtaining daily data on their
behaviour and identifying the driving forces behind
their distributions are major challenges. Advances in
the fields of bio-logging science, geographic infor-
mation systems and ecological modelling have led to
the increasing use of animal-attached sensors to
remotely examine the movements, behaviour, physi-
ology and/or biophysical habitat of a wide range of
marine species (Cooke et al. 2004, Ropert-Coudert et
al. 2009, Costa et al. 2012). Such studies have shed
new light on the ecology of marine predators, unrav-
elling a variety of behaviours ranging from localised
movements made in relation to foraging opportuni-
ties (Sims et al. 2006, Papastamatiou et al. 2012) to
larger-scale migrations (Bonfil et al. 2005, Block et al.
2011). In most cases, movements are driven by the
availability of food resources (Zerbini et al. 2006,
Anderson et al. 2011), species-specific physiologies
(Pillans 2006) or the need to reproduce (Bonfil et al.
2005, Skomal et al. 2009).
Previous tracking studies have highlighted the im -
portance of productive regions or features, such as
oceanographic fronts and mesoscale eddies, in pro-
viding foraging opportunities (Polovina et al. 2000,
Bailleul et al. 2010) for marine species, including
plankton-feeding sharks (Sims & Quayle 1998, Sims
et al. 2003). Planktivorous elasmobranchs must
acquire sufficient energy from minute and diffuse
prey. As a result, bottom-up processes, whereby phy -
sical oceanographic features act to concentrate prey
items in specific water bodies, are likely to influence
predator distributions and their behavioural deci-
sions. For instance, basking sharks Ceto rhinus maxi -
mus congregate and feed on zooplankton blooms in
frontal areas of the northeastern Atlantic (Sims &
Quayle 1998, Sims et al. 2003). However, the disper-
sal capabilities, behavioural ecology and ha bitat use
patterns of other planktivorous elasmobranchs, such
as manta rays (Manta spp.), are currently not well
understood.
Manta rays are the largest of the batoid fishes and
have a circumglobal distribution in tropical and sub-
tropical waters (Marshall et al. 2009, Couturier et al.
2012). They are classified as vulnerable to extinction
on the IUCN Red List of Threatened Species (Mar-
shall et al. 2011a,b). Despite increasing public and
scientific interest in manta rays, major knowledge
gaps remain in their dispersal abilities, migratory
ecology and drivers for their observed distributions.
Until recently, most of the movement data available
for manta rays had been derived from re-sightings
of photographically identified individuals over rela-
tively long time periods (e.g. Kashiwagi et al. 2010,
Couturier et al. 2011) or from acoustic telemetry data
(e.g. Dewar et al. 2008, Deakos et al. 2011). Such
technologies provided the first insights into the
movement ecology of manta rays, highlighting diur-
nal visitations of reef manta rays Manta alfredi to
particular inshore sites (Dewar et al. 2008, Marshall
2008), movements between aggregation sites up to
500 km apart (Kashiwagi et al. 2010, Couturier et al.
2011) and seasonal migratory patterns in some
regions (Anderson et al. 2011, Couturier et al. 2011).
More recent telemetry studies have, for the first time,
documented the movement patterns of both giant
manta rays M. birostris and reef manta rays M.
alfredi foraging in shallow habitats off the Yucatan
Peninsula, Mexico, and in the Line Islands of the cen-
tral Pacific, respectively (Graham et al. 2012, Papas-
tamatiou et al. 2012). Such results have highlighted
the importance of productivity blooms in inshore,
coastal and coral reef ecosystems as key drivers for
the spatial distributions and foraging habitats of
manta rays (Graham et al. 2012, Papastamatiou et al.
2012).
In eastern Australia, M. alfredi occur at various
localities along the coast, with some individuals sea-
sonally migrating between sites up to 500 km apart
(Couturier et al. 2011). One major aggregation site
for M. alfredi is Lady Elliot Island (LEI) (Couturier et
al. 2011, Jaine et al. 2012), a small coral cay in the
southern Great Barrier Reef (GBR), located only a
few kilometres from the continental shelf edge (see
Fig. 1A). Here, M. alfredi occur year-round; they peak
in austral autumn and winter, coincident with en -
hanced local productivity and foraging activity (Jaine
et al. 2012). The presence of the mesoscale cyclonic
Capricorn Eddy, which forms in the lee of the bathy -
metry off LEI because of the variability in strength of
the southward-flowing East Australian Current (EAC)
(Weeks et al. 2010), is known to trigger upwelling of
cool, nutrient-rich sub-surface waters onto the shelf
and around the Capricorn-Bunker reefs (Kleypas &
Burrage 1994, Weeks et al. 2010). Previously, the
eddy has been suggested as an important driver of
reef manta ray occurrence in the area and, more
specifically, at LEI (Jaine et al. 2012). There, passive
acoustic telemetry showed that individual manta rays
can typically be observed daily over short time scales
(i.e. up to 23 d) and may then leave the surveyed area
for extended periods of time (i.e. weeks to months)
before returning to the site (Couturier 2013). To date,
it is unknown where they go and what they do
when not around LEI. In addition, it is unclear how
M. alfredi disperse along the eastern Australian
74
Jaine et al.: Movements of reef manta rays
seaboard when undertaking long-distance move-
ments (>300 km) and whether they commonly utilise
specific migratory corridors, as observed in other
species off eastern Australia (Bansemer & Bennett
2011, Smith et al. 2012) and in other regions (Zerbini
et al. 2006, Campana et al. 2011).
Here, we used satellite telemetry to explore the
horizontal movements of individual M. alfredi de -
parting LEI and gain insight into their movement
patterns and habitat use in eastern Australia. We
used random walk model simulations and a 2-state
behaviour-switching state-space model (SSM) to
test the hypotheses that (1) manta ray distributions
within the southern GBR region are influenced by
the presence of the nearby Capricorn Eddy and (2)
movements of M. alfredi along the eastern Aus-
tralian seaboard are directed towards productive
regions. Our results suggest that reef manta rays
not only occupy inshore continental shelf and shelf-
edge waters but also are capable of using offshore
environments to exploit productive hotspots or
travel long distances.
MATERIALS AND METHODS
Manta ray tagging
Fieldwork was conducted at LEI (23° 06’ S,
152° 42’ E) in the southern GBR. Ten Manta alfredi
were fitted with pop-off satellite-linked archival
transmitting (PSAT) tags during austral winter 2010
(n = 2), summer 2010-2011 (n = 2) and winter 2011
(n = 6; Table 1). Prior to tag deployment, each indi-
vidual was identified, its size (disc width, WD) was
estimated and its sex was determined using con-
ventional manta ray photographic identification
and laser photogrammetry techniques (Deakos
2010, Couturier et al. 2011, Marshall & Pierce
2012). Tags were deployed on free-swimming indi-
viduals while free diving using a 320 kg Dyneema
braid leader and an umbrella-shaped plastic dart
inserted into the dorsal musculature, away from the
body cavity, with a customised tagging pole. Eight
manta rays (M1−M8) were equipped with Mk10
pop-up archival transmitting (Mk10-PAT) tags
(Wildlife Computers), and manta rays 9 and 10 (M9
and M10) were equipped with Standard Rate X-
Tags (Microwave Telemetry) (Table 1). Most tagged
individuals were re-sighted within a few minutes
post tag deployment, having resumed their previous
foraging or cleaning activity and appearing to be
unaffected by the devices.
Tag programming and geolocation
Mk10-PAT tags were programmed to record ambi-
ent light levels, sea temperature and pressure (to
allow for calculations of swimming depth) at 30 s
intervals and detach from the individual after 90 to
120 d. Depending on the tags, the binned data to
be transmitted to the Argos satellite system (www.
argos- system.org) upon release were summarised
over time intervals of 6 or 12 h. By default, X-Tags
recorded the same parameters every 2 min; however,
based on the total deployment period of 120 d, the
transmitted data were summarised into 15 min bins.
Upon release, 8 of the 10 deployed tags transmitted
the summarised data successfully. Tags 66701 and
18379, deployed on M3 and M9, respectively, failed
to transmit to the Argos system and were thus consid-
75
Manta Tag Sex Size Tag type Pop-off Date (dd/mm/yy) Duration (d) Tag
no. no. (WD, m) location Tagging Pop-off Planned Realised retrieved
M1 47726 M 3.50 Mk10-PAT 24°08' S, 153°16' E 27/06/10 30/08/10 120 65 Yes
M2 60518A F 3.50 Mk10-PAT 22°25' S, 151°39' E 28/06/10 5/10/10 120 99 Yes
M3 66701 F 3.75 Mk10-PAT – 5/01/11 – 120 – –
M4 66702 F 3.50 Mk10-PAT 23°28' S, 151°26' E 5/01/11 11/02/11 120 37 No
M5 66700 M 3.25 Mk10-PAT 23°40' S, 152°27' E 19/06/11 17/09/11 90 90 No
M6 66703 F 3.75 Mk10-PAT 23°30' S, 152°48' E 19/06/11 15/10/11 120 118 Yes
M7 66705 F 4.25 Mk10-PAT 22°45' S, 151°41' E 19/06/11 14/10/11 120 117 Yes
M8 60518B F 4.00 Mk10-PAT 22°22' S, 153°30' E 20/06/11 18/09/11 90 90 Yes
M9 18379 M 3.75 X-Tag – 20/06/11 – 120 – –
M10 18380 F 4.25 X-Tag 24°09' S, 152°37' E 20/06/11 18/10/11 120 120 Yes
Table 1. Information on the deployment of pop-off archival satellite-transmitting tags on photographically identified, sexed
and measured Manta alfredi. Tagging location was 23°06’ S, 152° 42’ E for all mantas. WD= disc width
Mar Ecol Prog Ser 510: 73–86, 2014
ered as ‘lost’. An additional 6 tags washed ashore
over a wide geographic range, from Maroochydore,
Sunshine Coast (Queensland, 26.6° S, 153.9° E), to
Shoalwater Bay (Queensland, 22.4° S, 150.7° E), and
were physically retrieved by members of the public,
allowing 100% of the higher-resolution archived raw
data to be examined (Table 1). Transmissions from
the remaining 2 tags (66 700 and 66 702) enabled
varying amounts of data to be obtained and analysed
(921 and 1565 messages, respectively).
Daily manta ray positions were estimated using the
‘Track&Loc’ geolocation filter developed at Collecte
Localisation Satellites (CLS, www.cls.fr), France. The
algorithm relies on an SSM to represent process
(movement) and observation uncertainty. Under -
water positioning is achieved using an Ensemble
Kalman filter applied to light-level measurements,
with sea surface temperature (SST) and bathymetry
data used to better constrain the tracks (Royer et al.
2005, Nielsen et al. 2006, Nielsen & Sibert 2007,
Royer & Lutcavage 2008, 2009). This process allowed
for the reconstruction of movement tracks based on
daily position estimates. Tracks were then plotted
and analysed for patterns of space utilisation, in
ArcGIS 10 (ESRI, www.esri.com).
Eddy affinity simulations
To test the hypothesis that manta ray distributions
within the southern GBR region are influenced by
the presence of the cyclonic Capricorn Eddy, we
used random walk model simulations to compare the
proportion of time spent in the eddy region by ‘real’
tracked manta rays and model manta rays. For each
tag, random movements of model manta rays were
simulated in R (Ihaka & Gentleman 1996) using the
SDMTools package, such that each model ray fea-
tured the same starting location and total track
length as the respective tracked manta ray. Distances
between successive daily positions, termed step-
lengths, were randomly chosen from the real manta
ray’s step-length frequency distribution, and a ran-
dom turn angle drawn from a uniform distribution
was selected at the end of each step. Each step of
the model manta rays was validated against a high-
resolution (i.e. 100 m) digital bathymetry map gene -
rated using the gbr100 dataset (Beaman 2010) to
preclude model rays from crossing land. The re-
orientation angle was replaced if the prior step was
rejected. For each tag, movement tracks for 1000
model manta rays were simulated, and the propor-
tion of time (in days) spent within the Capricorn Eddy
region was recorded and compared to that of the real
manta ray.
Behavioural analysis
A behaviour-switching SSM was fitted to move-
ment data collected for real manta rays to discrimi-
nate behavioural activity from movement patterns
and examine behavioural hotspots within the region
of interest. Behaviour-switching SSMs have previ-
ously been used to successfully infer the ‘hidden’
behavioural state of an animal based on movement
properties such as turn angles, step-lengths and
autocorrelation in speed and direction derived from
tracking data (Jonsen et al. 2005, Breed et al. 2009,
2012). Here, we fitted the 2-state switching corre-
lated random walk model originally described by
Jonsen et al. (2005) and refined in Breed et al. (2009)
to our Manta alfredi data. This model has been suc-
cessfully applied to various guilds of marine species
to discriminate ‘transiting’ from ‘foraging’ behav-
ioural states (e.g. pinnipeds: Breed et al. 2009, sea
turtles: Maxwell et al. 2011, cetaceans: Bailey et al.
2009 and giant manta rays M. birostris: Graham et al.
2012). We implemented the model in R and Win-
BUGS (www.winbugs-development.org.uk). To fit
the model, 2 Markov Chain Monte Carlo simulations
were computed for 10 000 iterations, with a ‘burn-in’
factor of 7000 and ‘thinning’ of 5, leaving 600 sam-
ples per chain as output to estimate each model
parameter (see Breed et al. 2009). Model output was
examined for differences between the autocorrela-
tion parameter γand the behavioural state transition
parameter αof the 2 states, indicative of true differ-
entiation of the associated movement patterns.
RESULTS
Over the course of this study, tags remained
attached to Manta alfredi for a mean period of 92 d
(± 29 SD, range 37 to 120 d). Five of the 8 tags that
successfully transmitted to the Argos system re -
ported on or near the programmed pop-up date,
with 3 detaching prematurely for unknown reasons
(Table 1). Overall, 5874 transmissions were received
by the Argos system, with raw light-level data avail-
able for the 6 recovered tags. Geolocation estimates
ranged in accuracy between 7.3 and 75.5 km2(me -
dian = 18.4 km2± 23.1 SD), depending on tag deploy-
ment duration and spatial extent of the movements
recorded.
76
Jaine et al.: Movements of reef manta rays
Spatial dynamics
Tagged Manta alfredi were tracked over 736 d,
with movements recorded across a latitudinal range
of 1035 km and with a mean track length of 1169 km
± 640 SD (Fig. 1B). Track length varied among indi-
viduals, with M6 travelling the farthest (2441 km in
118 d), whereas M4 moved only 314 km in 37 d
(Table 2). Female M. alfredi (n = 6) dispersed farther
than males (n = 2) (Fig. 1C), despite no significant dif-
ference in their respective mean track lengths (t =
−0.99, p = 0.38) and mean speeds (t = −0.59, p = 0.58).
Median speed across all tags, derived from daily
move-steps, was 0.4 km h−1 ± 0.5 SD, and maximum
speed recorded was 3.5 km h−1 for M8 (Fig. 1D). Esti-
mated swimming speeds across all tags were typi-
cally slow (i.e. <0.75 km h−1) near LEI and directly off
the shelf but increased when individuals dispersed to
other regions. Daily maximum diving depths re cor -
ded by the PSAT tags revealed that the tracked rays
dived down to 294.5 m (maximum depth range 56 to
294.5 m), with greater maximum depths logged off
the shelf near LEI (Fig. 1E).
Seven of the 8 tracked M. alfredi remained within
the GBR region, while M10 moved off the shelf in
waters over 2000 m deep and travelled southward to
28° S and back in 120 d (Fig. 1B). The southward leg
of the journey was covered at ~24 km d−1 (Fig. 1D).
M10 then spent ~50 d near the shelf edge off More-
ton Island and North Stradbroke Island (NSI). Start-
ing on Day 82, M10 moved 155 km eastward off the
shelf before turning northward and broadly retracing
its earlier southward movement, until the tag de -
tached near LEI, coincident with the detection of the
Capricorn Eddy and shelf intrusions in the satellite
signal (Fig. 2). The other manta rays tagged at LEI
(M1, M2 and M4−M8) all moved directly off the shelf
near the tagging site, where they remained for a
period of 50.7 d ± 28.3 SD before moving back onto
the shelf and dispersing farther within the GBR. M1,
M2 and M4−M8 moved up to 520 km from LEI (mean
farthest linear distance from tagging site was 247 km
± 149 SD in 77 d ± 16 SD).
Activity hotspots and eddy affinity
Movement tracks recorded across all tags (n = 8)
highlighted a major activity hotspot off the shelf near
LEI (Fig. 1B), where the mantas spent 58.5% (range
2.5 to 87.5%) of their time. There was a second hot -
spot off the shelf ~100 km east of Moreton Island
based on movement data for M10 only, which spent a
considerable amount of time in this area (42% of total
track).
Results from the random walk simulations varied
among tags (Fig. 3). Only 2 of the 8 simulations were
significant, where real manta rays spent significantly
more time in the eddy region than the 1000 models
(Table 3). However, since each simulation can be
regarded as an independent test of whether manta
rays spent more time than random in the eddy,
results from the 8 tests, each with 1000 model itera-
tions, were combined using a proportion test. To -
gether, these simulations revealed that real manta
rays spent significantly more time in the eddy region
than model rays (p < 0.001).
Behaviour
The 2-state behaviour-switching SSM discrimi-
nated transiting and foraging activity from the track-
ing data (Fig. 1F). Of the total tracking positions (n =
77
Manta Min. horizon- Track Max. Step-length Speed (km d−1) Time (%) Mean speed (km h−1)
no. tal displace- length depth (km d−1) Median Max. Fora- Tran- Fora- Tran-
ment (km) (km) (m) Mean Max. ging siting ging siting
M1 63 523 283 8.2 31.9 0.3 1.3 100.0 0.0 0.3 –
M2 218 1023 167 10.3 34.7 0.3 1.4 76.8 23.2 0.3 0.7
M4 137 314 56 8.5 61.0 0.2 2.5 78.4 21.6 0.2 0.9
M5 53 1089 96 12.0 63.6 0.4 2.7 82.4 17.6 0.5 0.6
M6 69 2441 140 20.7 75.5 0.6 3.1 41.5 58.5 0.6 1.2
M7 183 1243 196 10.6 43.7 0.3 1.8 66.7 33.1 0.4 0.7
M8 210 1365 294.5 15.2 84.4 0.6 3.5 26.7 73.3 0.5 0.8
M10 7 1351 97.2 11.1 31.6 0.4 1.3 0.0 100.0 – 0.5
Table 2. Movement metrics for 8 Manta alfredi tracked by pop-off archival satellite-transmitting tags. Minimum horizontal dis-
placement = distance between tag deployment and detachment locations. Speed is estimated from daily move step-lengths.
Foraging and transiting behaviours are inferred from state-space analysis of movement data
78
Jaine et al.: Movements of reef manta rays
736), 53.5% were determined as the ‘foraging’ state,
while the ‘transiting’ state made up the remaining
46.5%. Results revealed important foraging activity
across all but one (M10) track in the eddy region, off-
shore of LEI. The foraging behavioural state made up
~95% of locations recorded in the eddy region. A
secondary foraging location off the Swain Reefs was
also highlighted based on tracking data for M8,
which made up 3.6% of total locations across all tags.
DISCUSSION
Using satellite telemetry and spatial analyses, we
showed that reef manta rays not only occupy inshore
continental shelf and shelf-edge waters but also use
offshore environments to exploit productive hotspots
or undertake long-distance movements. All but 1
manta ray tagged in the southern GBR spent a sub-
stantial amount of time in the Capricorn Eddy region,
off the continental shelf. A behavioural analysis of
the tracking data further suggested that they may
use this area as a foraging ground. Using offshore
waters, 1 manta ray travelled 520 km southward to
another known aggregation region before returning
to the tagging region.
Spatial dynamics
Reef manta rays tagged in the southern GBR
moved across a wide geographical area along the
eastern Australian coastline, covering a latitudinal
range of 1035 km (between 20° S and 28° S). Al -
though minimum horizontal displacements (i.e. dis-
tances between tagging and pop-off locations) were
relatively small and suggestive of localised move-
ments of individuals (Table 2), track reconstruction
revealed more extensive movements. Reef manta
rays travelled up to 2441 km in 118 d and dispersed
as far as 520 km from the tagging site, occupying
both regional shelf and offshore waters. This is, to
date, one of the largest directional movements
recorded and the first horizontal movement tracks
obtained for Manta alfredi using satellite telemetry.
Tracked reef manta rays displayed some degree of
affinity to the southern GBR. Despite recording some
of the longest movement tracks and moving the far-
thest away from LEI, M6 and M10 eventually
returned to the southern GBR towards the end of
their respective tagging periods. The 6 other tagged
individuals remained within the southern GBR
region during their tracking periods. These results
suggest some degree of fidelity to the region,
79
Fig. 1. Study region showing (A) typical regional oceanographic setting (Moderate Resolution Imaging Spectroradiometer
[MODIS] sea surface temperature, June 2010), marked by the southward flow of the core East Australian Current (EAC) and
the cyclonic Capricorn Eddy (CE) forming in the lee of the shelf topography; (B) movement tracks for Manta alfredi tagged at
Lady Elliot Island (n = 8), with daily position estimates (filled circles); (C) spatial dynamics of male (n = 2) and female (n = 6)
M. alfredi; (D) swimming speed estimated from daily position estimates; (E) daily maximum diving depth for tracked M. alfredi
(n = 8); and (F) behaviour inferred from state-space analysis of the tracking data. Blue shading in (B−F) indicates water depth,
where lightest colours are depths < 200 m, and darkest colours are depths > 2000 m
Manta Days in eddy Time in eddy Successful model Unsuccessful model p-value
ray region region (%) simulations simulations
M1 55 84.6 101 899 0.101
M2 65 65.6 154 846 0.154
M4 32 86.5 92 908 0.092
M5 79 87.8 16 984 0.016
M6 55 46.6 40 960 0.040
M7 71 60.7 125 875 0.125
M8 30 33.3 328 672 0.328
M10 3 2.5 647 353 0.647
Total 390 58.5 1503 6497 <0.001
Table 3. Time spent by each tracked Manta alfredi in the Capricorn Eddy region and output from the random walk simulations for
each track, including numbers of successes, failures and resulting p-value for the test comparing the proportion of time spent in
the eddy by tracked and model manta rays. Successful and unsuccessful simulations refer to the numbers of model rays that
successfully (or unsuccessfully) spent as much time in the eddy region as real manta rays
Mar Ecol Prog Ser 510: 73–86, 2014
which supports previous findings by Couturier et al.
(2011). Using photographic identification techniques,
Couturier et al. (2011) documented dispersal by
M. alfredi to other known aggregation sites along
the eastern Australian seaboard, likely associated
with seasonal migratory movements. Similarly, van
Duinkerken (2010) identified the dispersal of individ-
ual M. alfredi along the southern Mozambican coast.
There, acoustically tagged M. alfredi typically showed
high fidelity to particular sites along the coast and
occasionally dispersed throughout the entire 95 km
long acoustic array. Long-distance movements and
80
Fig. 2 (this page and next page). Spatial and temporal dynamics of Manta alfredi M10 in relation to the oceanographic condi-
tions. Moderate Resolution Imaging Spectro radio me ter (MODIS) sea surface temperatures (left column panels) and chl acon-
centrations (right column panels) are presented for movements of M10 (white track) tracked between (A) June 19 and July 18,
2011, (B) July 19 and August 18, 2011, (C) September 19 and October 18, 2011, and (D) November 19 and December 18, 2011.
Data are presented at monthly intervals for clarity only, although dy namics at finer temporal resolutions have been in -
vestigated and did not differ. ‘CE’ marks the presence of the Capricorn Eddy feature in the satellite signal, and white arrows
denote intrusions of upwelled oceanic wa ters onto the shelf. EAC = East Australian Current
Jaine et al.: Movements of reef manta rays
migrations are typically attributed to the need to
reproduce and exchange genetic material between
members of separate populations (Bonfil et al. 2005,
Skomal et al. 2009) or the search for abundant food
resources (Zerbini et al. 2006, Anderson et al. 2011).
The question of sex-biased dispersal in M. alfredi
could not be assessed here because of our limited
data, despite tagged females noticeably dispersing
farther than the 2 males.
The mean and maximum speeds observed here,
derived from daily distances travelled, were similar
to those observed in other M. alfredi sub-popula-
tions (e.g. van Duinkerken 2010). Despite the low
spatial and temporal resolution of the collected
data, results suggested that reef manta rays can
travel distances of up to ~85 km d−1, averaging
speeds of ~3.5 km h−1. In the future, the use of
additional sensors (e.g. accelerometers) will help
collect more accurate, higher-resolution swimming
speed data, which may provide additional insights
into the movement ecology of these planktivorous
elasmobranchs.
81
Fig. 2 (continued)
Mar Ecol Prog Ser 510: 73–86, 2014
Eddy affinity and offshore foraging
The offshore Capricorn Eddy region was the pri-
mary site of occupancy for most satellite-tagged rays.
Mesoscale eddies, such as the Capricorn Eddy, stim-
ulate and redistribute biological production in the
ocean, thus creating attractive pelagic habitats for
free-ranging, higher trophic level marine organisms
(Chelton et al. 2011, Godø et al. 2012). Our results
suggest that manta rays may exploit offshore meso -
scale eddies for foraging purposes and corroborate
several other studies that showed the importance of
mesoscale eddies as offshore foraging grounds for a
variety of marine species (e.g. seabirds: Weimers -
kirch et al. 2004, Cotté et al. 2007; pinnipeds: Bailleul
et al. 2010, Dragon et al. 2010; cetaceans: Woodworth
et al. 2012).
Cyclonic eddies are known to enhance nutrient
enrichment and subsequent primary production in
subtropical western boundary systems (Falkowski et
al. 1991, McGillicuddy et al. 1998, 2007, Weeks et al.
2010). In the case of the Capricorn Eddy, turbulent
momentum exchanges with the strong southward-
flowing EAC that flows along the edge of the GBR
shelf produce a lateral stress on the mass of water in
the lee of the Swain Reefs. This drives cyclonic
(clockwise) and radially outward eddy circulation,
leading to the upwelling of cooler, nutrient-enriched
water in the centre of the eddy (Bakun 1996, Weeks
et al. 2010). The resulting upwelled waters flow
82
Fig. 3. Representative unsuccessful (M5) and successful (M10) random walk simulations. Panels in the first column, ’observed
track’, present the movement track (red line) and daily position estimates (filled circles) for each individual manta ray. The second
column, ‘simulation’, shows the geographically unconstrained 1000 random walks (blue) generated for each respective Manta
alfredi (red) from the step-length frequency distribution and the typical Capricorn Eddy region (black circle) (x- and y-axes are
degrees of longitude and latitude, respectievly). The third column, ‘time in eddy region’, indicates the proportion of time spent in
the eddy region by 1000 unconstrained random walks (white bars) compared to that of each respective M. alfredi (red line)
Jaine et al.: Movements of reef manta rays
coastward in the near-surface upper layer, eventu-
ally accumulating near LEI and the Capricorn-
Bunker reefs. Interestingly, the primary activity area
of tagged and presumed foraging Manta alfredi in
this study was the southeastern portion of the eddy
region, most likely because of the consistently en -
hanced productivity in this area triggered by eddy
dynamics.
Reef manta rays tagged in the southern GBR spent
extended periods off the shelf, undertaking deep
dives, presumably foraging, and occasionally moving
back onto the shelf. This result supports findings
from Braun et al. (2014), who documented offshore
deep-diving behaviour of satellite-tracked M. alfredi
directly adjacent to shallow coral reefs. Together,
these results challenge previous assumptions that
reef manta rays primarily rely on inshore productiv-
ity blooms as foraging habitats and highlight that
regional and mesoscale oceanographic investiga-
tions are needed to better understand habitat use
patterns of these large, free-ranging planktivores.
Moreover, a recent study by Couturier et al. (2013)
suggested that M. alfredi in eastern Australia do
not predominantly rely on near-surface zooplankton
prey during coastal feeding events and suggested
that deep and/or demersal zooplankton may be an
important part of the diet. The movement patterns
and depths recorded by tagged M. alfredi in this
study, suggestive of extensive off-shelf foraging and
greater diving activity in a region of well-
documented upwelling (Kleypas & Burrage 1994,
Weeks et al. 2010), may thus provide additional
insights into the foraging ecology of the species in
eastern Australia. More detailed analyses of the
recovered high-resolution depth data will help fur-
ther explore the diving behaviour and vertical habi-
tat use of M. alfredi.
Movements between aggregation regions
A census for manta ray sightings in eastern Aus-
tralia provided the first insight into the geographical
distribution and movement patterns of Manta alfredi
in the region (Couturier et al. 2011). Results from the
photographic survey revealed that M. alfredi is pres-
ent along ~3000 km of coastline, from Torres Strait in
northern Queensland to Sydney in central New
South Wales, and highlighted several inshore loca-
tions as key aggregation sites for the species (Cou-
turier et al. 2011). More importantly, some M. alfredi
appear to migrate between waters off NSI, Byron Bay
or the Solitary Islands in spring-summer (mid-
October to mid-April) and the southern GBR in
autumn-winter (Couturier et al. 2011, Jaine et al.
2012). Here, we found that 1 reef manta ray moved
southward from LEI to near NSI and returned to the
southern GBR in 120 d, providing the first movement
track for M. alfredi travelling between known aggre-
gation regions along the eastern Australian sea -
board.
The movement patterns of this 1 individual indi-
cated that reef manta rays can use offshore waters,
including the EAC flow, to travel up and down the
eastern coast of Australia (Fig. 2). Despite the single
observation here, 35 individual M. alfredi have been
repeatedly re-sighted previously between LEI and
the southern sites of NSI, Byron Bay and the Solitary
Islands (Couturier et al. 2011), and it is thus likely
that more individuals undertake similar movements
because of the variability and seasonality in the EAC
flow. For example, during summer reef manta rays
are common in the southern portion of the coast
(Couturier et al. 2011), coincident with increased
EAC intensity and subsequent enhanced biological
productivity (Oke & Middleton 2000, Roughan &
Middleton 2004, Nieblas et al. 2009). In winter, when
the EAC flow is weaker and the core EAC waters do
not extend as far south, reef manta ray numbers peak
in the southern GBR (Jaine et al. 2012), coincident
with a clear satellite signal of the Capricorn Eddy
and associated enhanced productivity (Kleypas &
Burrage 1994, Weeks et al. 2010).
Although limited to 1 individual, the observed
movements of M10, which travelled between the
southern GBR and NSI, could suggest that enhanced
productivity off these sites may act as a driver for
M. alfredi migrations. The individual, tracked over
winter-spring 2011, moved to spend a considerable
amount of time off the shelf-edge near NSI. Although
the behavioural SSM analysis revealed transiting-
only behaviour for this individual, most likely
because of the very directional nature of its move-
ment patterns, the extended period spent off the
shelf edge off NSI suggests otherwise. Despite anec-
dotal records of M. alfredi foraging in NSI coastal
waters, the NSI study site is primarily known as a
popular manta ray cleaning station. However, while
no published study has yet documented upwelling
frequency and productivity in the area, in situ
oceanographic data clearly show that upwelling
events occur and are most pronounced with in -
creased EAC intensity in spring-summer (S. J. Weeks
pers. obs.). The individual slowed down when
approaching the shelf-edge region off NSI and spent
the following ~50 d in the area before moving farther
83
Mar Ecol Prog Ser 510: 73–86, 2014
offshore and eventually returning to the southern
GBR. While remotely sensed SSTs and chl aconcen-
trations for the corresponding period do not show
strong inshore EAC flow or sharp fronts off NSI, typ-
ically indicative of inshore upwelling processes, the
departure from and subsequent return to LEI appear
to coincide with increased intensity of the Capricorn
Eddy and shelf intrusions, as seen in the satellite sig-
nal (Fig. 2). It is thus possible that this individual may
have been searching for other productive hotspots
along the eastern Australian seaboard.
CONCLUSIONS
This study presents the first satellite-derived move-
ment dataset for the reef manta ray Manta alfredi in
Australia. In addition to showing that M. alfredi move
across relatively large scales (i.e. > 500 km), this
study highlights that reef manta rays spend consider-
able time in offshore pelagic waters. Our results
show that eastern Australian manta rays have an
affinity for a relatively stable mesoscale cyclonic
eddy, which may comprise an important foraging
ground for the species. Further, we document the
movements of 1 individual between 2 popular aggre-
gation regions off the eastern Australian seaboard
using offshore waters. Together, our findings suggest
that M. alfredi may commonly use offshore waters to
undertake movements between aggregation sites
and exploit ephemeral productivity hotspots. Al -
though threats to M. alfredi are comparatively low in
eastern Australia, these findings highlight the need
for enhanced knowledge about the movements of
manta rays to be considered when implementing re -
gional management strategies.
Acknowledgements. We thank Dr. J. Brunnschweiler and
Dr. C. Huveneers for providing feedback on earlier versions
of the manuscript. We also thank Dr. G. Breed for help with
interpretation of the output from the behaviour-switching
SSM, as well as the ‘Track&Loc’ team at CLS, Argos, France,
for assistance with geolocation analysis and PSAT tag data
monitoring. We gratefully acknowledge the NASA Ocean
Biology Processing Group for provision of Moderate Resolu-
tion Imaging Spectroradiometer satellite data. We are grate-
ful to the various helpers who contributed to the fieldwork,
including C. Garraway, A. Rakovski, R. Jeffery, K. Fiora, C.
Ansell, R. Cheseldene-Culley, A. Donnelly and M. Roma -
novski. This study was supported by Australian Research
Council Linkage Projects grant LP110100712, a research
grant from the PADI Foundation, Sibelco, an anonymous
donor, Earthwatch Institute Australia and Lady Elliot Island
Eco Resort. Fieldwork was conducted under Great Barrier
Reef Marine Park permit G09/29853.1 and ethics approval
SBMS/071/08/SEAWORLD.
LITERATURE CITED
Anderson RC, Adam MS, Goes JI (2011) From monsoons to
mantas: seasonal distribution of Manta alfredi in the
Maldives. Fish Oceanogr 20: 104−113
Bailey H, Mate BR, Palacios DM, Irvine L, Bograd SJ, Costa
DP (2009) Behavioural estimation of blue whale move-
ments in the Northeast Pacific from state-space model
analysis of satellite tracks. Endang Species Res 10:
93−106
Bailleul F, Cotté C, Guinet C (2010) Mesoscale eddies as for-
aging area of a deep-diving predator, the southern ele-
phant seal. Mar Ecol Prog Ser 408: 251−264
Bakun A (1996) Patterns in the ocean: ocean processes and
marine population dynamics. University of California
Sea Grant, San Diego, California, USA, in cooperation
with Centro de Investigaciones Biologicas de Noroeste,
La Paz, Baja California Sur
Bansemer CS, Bennett MB (2011) Sex- and maturity-based
differences in movement and migration patterns of grey
nurse shark, Carcharias taurus, along the eastern coast
of Australia. Mar Freshw Res 62: 596−606
Beaman RJ (2010) Project 3DGBR: a high-resolution depth
model for the Great Barrier Reef and Coral Sea. Marine
and Tropical Sciences Research Facility, Project 2.5i.1a
Final Report, Cairns, Australia, available at www.
deepreef.org/bathymetry/65-3dgbr-bathy.html
Block BA, Teo SLH, Walli A, Boustany A and others (2005)
Electronic tagging and population structure of Atlantic
bluefin tuna. Nature 434: 1121−1127
Block BA, Jonsen ID, Jorgensen SJ, Winship AJ and others
(2011) Tracking apex marine predator movements in a
dynamic ocean. Nature 475: 86−90
Bonfil R, Meyer M, Scholl M, Johnson R and others (2005)
Transoceanic migration, spatial dynamics, and popula-
tion linkages of white sharks. Science 310: 100−103
Braun CD, Skomal GB, Thorrold SR, Berumen ML (2014)
Diving behavior of the reef manta ray links coral reefs
with adjacent deep pelagic habitats. PLoS One 9(2),
e88170
Breed GA, Jonsen ID, Myers RA, Bowen WD, Leonard ML
(2009) Sex-specific, seasonal foraging tactics of adult
grey seals (Halichoerus grypus) revealed by state−space
analysis. Ecology 90: 3209−3221
Breed GA, Costa DP, Jonsen ID, Robinson PW, Mills-
Flemming J (2012) State-space methods for more com-
pletely capturing behavioral dynamics from animal
tracks. Ecol Model 235-236: 49−58
Campana SE, 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
Chelton DB, Gaube P, Schlax MG, Early JJ, Samelson RM
(2011) The influence of nonlinear mesoscale eddies on
near-surface oceanic chlorophyll. Science 334: 328−332
Cooke SJ, Hinch SG, Wikelski M, Andrews RD, Kuchel LJ,
Wolcott TG, Butler PJ (2004) Biotelemetry: a mechanistic
approach to ecology. Trends Ecol Evol 19: 334−343
Costa DP, Breed GA, Robinson PW (2012) New insights into
pelagic migrations: implications for ecology and conser-
vation. Annu Rev Ecol Evol Syst 43: 73−96
Cotté C, Park YH, Guinet C, Bost CA (2007) Movements of
foraging king penguins through marine mesoscale
eddies. Proc Biol Sci 274: 2385−2391
Couturier LIE (2013) Population ecology and biology of the
84
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Jaine et al.: Movements of reef manta rays
reef manta ray Manta alfredi in eastern Australia. PhD
thesis, School of Biomedical Sciences, The University of
Queensland, Brisbane, available at http://espace.library.
uq. edu.au/ view/UQ:302657
Couturier LIE, Jaine FRA, Townsend KA, Weeks SJ,
Richardson AJ, Bennett MB (2011) Distribution, site
affinity and regional movements of the manta ray, Manta
alfredi (Krefft, 1868), along the east coast of Australia.
Mar Freshw Res 62: 628−637
Couturier LIE, Marshall AD, Jaine FRA, Kashiwagi T and
others (2012) Biology, ecology and conservation of the
Mobulidae. J Fish Biol 80: 1075−1119
Couturier LIE, Rohner CA, Richardson AJ, Marshall AD and
others (2013) Stable isotope and signature fatty acid
analyses suggest reef manta rays feed on demersal zoo-
plankton. PLoS ONE 8: e77152
Deakos MH (2010) Paired-laser photogrammetry as a simple
and accurate system for measuring the body size of free-
ranging manta rays Manta alfredi. Aquat Biol 10: 1−10
Deakos MH, Baker JD, Bejder L (2011) Characteristics of a
manta ray Manta alfredi population off Maui, Hawaii,
and implications for management. Mar Ecol Prog Ser
429: 245−260
Dewar H, Mous P, Domeier M, Muljadi A, Pet J, Whitty J
(2008) Movements and site fidelity of the giant manta
ray, Manta birostris, in the Komodo Marine Park, Indone-
sia. Mar Biol 155: 121−133
Dragon AC, Monestiez P, Bar-Hen A, Guinet C (2010) Link-
ing foraging behaviour to physical oceanographic struc-
tures: southern elephant seals and mesoscale eddies east
of Kerguelen Islands. Prog Oceanogr 87: 61−71
Falkowski PG, Ziemann D, Kolber Z, Bienfang PK (1991)
Role of eddy pumping in enhancing primary production
in the ocean. Nature 352: 55−58
Godø OR, Samuelsen A, Macaulay GJ, Patel R and others
(2012) Mesoscale eddies are oases for higher trophic
marine life. PLoS ONE 7: e30161
Graham RT, Witt MJ, Castellanos DW, Remolina F, Maxwell
S, Godley BJ, Hawkes LA (2012) Satellite tracking of
manta rays highlights challenges to their conservation.
PLoS ONE 7: e36834
Ihaka R, Gentleman R (1996) R: a language for data analysis
and graphics. J Comput Graph Stat 5: 299−314
Jaine FRA, Couturier LIE, Weeks SJ, Townsend KA, Bennett
MB, Fiora K, Richardson AJ (2012) When giants turn up:
sighting trends, environmental influences and habitat
use of the manta ray Manta alfredi at a coral reef. PLoS
ONE 7: e46170
Jonsen ID, Flemming JM, Myers RA (2005) Robust state-
space modeling of animal movement data. Ecology 86:
2874−2880
Kashiwagi T, Ito T, Sato F (2010) Occurrences of reef manta
ray, Manta alfredi, and giant manta ray, M. birostris, in
Japan, examined by photographic records. Rep Jpn Soc
Elasmobranch Stud 46: 20−27
Kleypas J, Burrage D (1994) Satellite observations of circula-
tion in the southern Great Barrier Reef, Australia. Int J
Remote Sens 15: 2051−2063
Marshall AD (2008) Biology and population ecology of
Manta birostris in southern Mozambique. PhD thesis,
School of Biomedical Sciences, The University of
Queens land, Brisbane, available at http: //espace. library.
uq. edu. au/ view/ UQ: 160974
Marshall AD, Pierce SJ (2012) The use and abuse of photo-
graphic identification in sharks and rays. J Fish Biol 80:
1361−1379
Marshall A, Compagno L, Bennett M (2009) Redescription of
the genus Manta with resurrection of Manta alfredi
(Krefft, 1868) (Chondrichtyes; Myliobatoidei; Mobuli-
dae). Zootaxa 2031: 1−28
Marshall AD, Kashiwagi T, Bennett MB, Deakos M and oth-
ers (2011a) Manta alfredi. In:
IUCN Red List of Threat-
ened Species version 2012-1 (accessed on February 15,
2013)
Marshall A, Bennett M, Kodja G, Hinojosa-Alvarez S and
others (2011b) Manta birostris. In: IUCN Red List of
Threatened Species version 2012-1 (accessed on Febru-
ary 15, 2013)
Maxwell SM, Breed GA, Nickel BA, Makanga-Bahouna J
and others (2011) Using satellite tracking to optimize
protection of long-lived marine species: olive ridley sea
turtle conservation in central Africa. PLoS ONE 6:e19905
McGillicuddy DJ, Robinson AR, Siegel DA, Jannasch HW
and others (1998) Influence of mesoscale eddies on new
production in the Sargasso Sea. Nature 394: 263−266
McGillicuddy DJ, Anderson LA, Bates NR, Bibby T and
others (2007) Eddy/wind interactions stimulate extra -
ordinary mid-ocean plankton blooms. Science 316:
1021−1026
Nieblas AE, Sloyan BM, Hobday AJ, Coleman R, Richardson
AJ (2009) Variability of biological production in low
wind-forced regional upwelling systems: a case study off
southeastern Australia. Limnol Oceanogr 54: 1548−1558
Nielsen A, Sibert J (2007) State−space model for light-based
tracking of marine animals. Can J Fish Aquat Sci 64:
1055−1068
Nielsen A, Bigelow KA, Musyl MK, Sibert J (2006) Improv-
ing light-based geolocation by including sea surface
temperature. Fish Oceanogr 15: 314−325
Oke PR, Middleton JH (2000) Topographically induced
upwelling off eastern Australia. J Phys Oceanogr 30:
512−531
Papastamatiou YP, DeSalles PA, McCauley DJ (2012) Area-
restricted searching by manta rays and their response to
spatial scale in lagoon habitats. Mar Ecol Prog Ser 456:
233−244
Pillans RD (2006) The physiological ecology of the bull shark
Carcharhinus leucas in the Brisbane River. PhD thesis,
School of Integrative Biology, The University of Queens-
land, Brisbane, available at http: //espace.library.uq.
edu.au/ view/ UQ: 158284
Polovina JJ, Kobayashi DR, Parker DM, Seki MP, Balazs GH
(2000) Turtles on the edge: movement of loggerhead tur-
tles (Caretta caretta) along oceanic fronts, spanning
longline fishing grounds in the central North Pacific,
1997−1998. Fish Oceanogr 9: 71−82
Richardson AJ, Schoeman DS (2004) Climate impact on
plankton ecosystems in the northeast Atlantic. Science
305: 1609−1612
Ropert-Coudert Y, Beaulieu M, Hanuise N, Kato A (2009)
Diving into the world of biologging. Endang Species Res
10: 21−27
Roughan M, Middleton JH (2004) On the East Australian
Current: variability, encroachment, and upwelling. J
Geophys Res 109: C07003, doi:10.1029/2003JC001833
Royer F, Lutcavage M (2008) Filtering and interpreting loca-
tion errors in satellite telemetry of marine animals. J Exp
Mar Biol Ecol 359: 1−10
Royer F, Lutcavage M (2009) Positioning pelagic fish from
sunrise and sunset times: complex observation errors call
85
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Mar Ecol Prog Ser 510: 73–86, 2014
for constrained, robust modeling. In: Nielsen JL, Fragoso
N, Lutcavage M, Arrizabalaga H, Hobday A, Sibert J
(eds) Tagging and tracking of marine animals with
electronic devices. Reviews: methods and technologies
in fish biology and fisheries. Springer, Dordrecht,
p 323−341
Royer F, Fromentin JM, Gaspar P (2005) A state-space
model to derive bluefin tuna movement and habitat from
archival tags. Oikos 109: 473−484
Sims D, Quayle V (1998) Selective foraging behaviour of
basking sharks on zooplankton in a small-scale front.
Nature 393: 460−464
Sims DW, Southall EJ, Richardson AJ, Reid PC, Metcalfe JD
(2003) Seasonal movements and behaviour of basking
sharks from archival tagging: no evidence of winter
hibernation. Mar Ecol Prog Ser 248: 187−196
Sims DW, Witt MJ, Richardson AJ, Southall EJ, Metcalfe JD
(2006) Encounter success of free-ranging marine preda-
tor movements across a dynamic prey landscape. Proc R
Soc Lond B Biol Sci 273: 1195−1201
Skomal GB, Zeeman SI, Chisholm JH, Summers EL, Walsh
HJ, McMahon KW, Thorrold SR (2009) Transequatorial
migrations by basking sharks in the western Atlantic
Ocean. Curr Biol 19: 1019−1022
Smith JN, Grantham HS, Gales N, Double MC, Noad MJ,
Paton D (2012) Identification of humpback whale breed-
ing and calving habitat in the Great Barrier Reef. Mar
Ecol Prog Ser 447: 259−272
van Duinkerken DI (2010) Movements and site fidelity of the
reef manta ray, Manta alfredi, along the coast of southern
Mozambique. Masters thesis, Utrecht University
Weeks SJ, Bakun A, Steinberg CR, Brinkman R, Hoegh-
Guldberg O (2010) The Capricorn Eddy: a prominent
driver of the ecology and future of the southern Great
Barrier Reef. Coral Reefs 29: 975−985
Weimerskirch H, Le Corre M, Jaquemet S, Potier M, Marsac
F (2004) Foraging strategy of a top predator in tropical
waters: great frigatebirds in the Mozambique Channel.
Mar Ecol Prog Ser 275: 297−308
Woodworth PA, Schorr GS, Baird RW, Webster DL and
others (2012) Eddies as offshore foraging grounds for
melon-headed whales (Peponocephala electra).Mar
Mamm Sci 28: 638−647
Zerbini AN, Andriolo A, Heide-Jørgensen MP, Pizzorno JL
and others (2006) Satellite-monitored movements of
humpback whales Megaptera novaeangliae in the
Southwest Atlantic Ocean. Mar Ecol Prog Ser 313:
295−304
86
Editorial responsibility: Nicholas Tolimieri,
Seattle, Washington, USA
Submitted: June 11, 2013; Accepted: June 10, 2014
Proofs received from author(s): August 29, 2014
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
