Seeing Spots: Photo-identification as a regional tool for whale shark identification

Article (PDF Available) · January 2010with 109 Reads
Abstract
The identification of individual animals over temporal and spatial scales can provide robust estimates of population size and distribution. While marker tagging can provide an option to achieve this, it can be problematic both in terms of tag loss and the associated difficulties and effects of attaching the tags. Photo-identification of distinctive characteristics which remain stable over time has replaced tagging in some species but usage at regional scales has been hampered by a lack of standardisation of matching methods. We describe the use of a semi-automated computer program (I 3 S) for matching the spot patterns of whale sharks, Rhincodon typus, in the Seychelles aggregation and compare this to images captured from other areas in the Western Indian Ocean. Sharks totalling 443 individuals were uniquely identified in the Seychelles from 2001 – 2009, 109 of which were seen in multiple years. Conventional open mark-recapture models for 2004 – 2009 gave an abundance estimate of 469 to 557 sharks (95% C.I.). I 3 S digital fingerprints were shared with researchers in Djibouti, Mozambique, and Tanzania and, while no matches were found between locations, the ease with which regional comparisons were made will help to define whether the shark populations in these areas are distinct, enabling long-term and broad-scale regional comparisons.
Seeing Spots: Photo-identication as a Regional Tool for
Whale Shark Identication
Katie Brooks1, David Rowat1, Simon J. Pierce2, Daniel Jouannet3 and Michel Vely3
1Marine Conservation Society Seychelles, Box 384, Victoria, Seychelles;
2Marine Megafauna Foundation, Tofo Beach, Mozambique;
3Megaptera, 23 Rue Alexandre Dumas, 75011, Paris, France.
Keywords: Whale shark, Rhincodon typus, Indian Ocean, photo-identication,
mark-recapture, population estimation.
AbstractThe identication of individual animals over temporal and spatial scales
can provide robust estimates of population size and distribution. While marker
tagging can provide an option to achieve this, it can be problematic both in terms
of tag loss and the associated difculties and effects of attaching the tags. Photo-
identication of distinctive characteristics which remain stable over time has replaced
tagging in some species but usage at regional scales has been hampered by a lack
of standardisation of matching methods. We describe the use of a semi-automated
computer program (I3S) for matching the spot patterns of whale sharks, Rhincodon
typus, in the Seychelles aggregation and compare this to images captured from other
areas in the Western Indian Ocean. Sharks totalling 443 individuals were uniquely
identied in the Seychelles from 2001 – 2009, 109 of which were seen in multiple
years. Conventional open mark-recapture models for 2004 – 2009 gave an abundance
estimate of 469 to 557 sharks (95% C.I.). I3S digital ngerprints were shared with
researchers in Djibouti, Mozambique, and Tanzania and, while no matches were
found between locations, the ease with which regional comparisons were made will
help to dene whether the shark populations in these areas are distinct, enabling
long-term and broad-scale regional comparisons.
Western Indian Ocean J. Mar. Sci. Vol. 9, No. 2, pp. 185-194, 2010
© 2010 WIOMSA
Corresponding Author: DR
E-mail: david@mcss.sc
INTRODUCTION
Whale sharks (Rhincodon typus) are pan-
oceanic planktivores that were rst described
from a specimen captured in the Western Indian
Ocean in 1828 (Smith, 1828). They are listed by
the IUCN as Vulnerable (VU A1bd+2d ) based
on observed reductions in landings, actual
levels of exploitation and because further
population decline is deemed likely to occur if
directed sheries remain unmanaged (IUCN,
2009). In November 1999, the whale shark
was added to Appendix II of the Convention on
Migratory Species (CMS) as “a species whose
conservation status would benet from the
implementation of international co-operative
Agreements”. The species was also listed in
Appendix II of the Convention on International
Trade in Endangered Species (CITES) in
November 2002.
Despite broad international interest in the
conservation status of whale sharks, relatively
little is known about regional population sizes
or trends in abundance. Some local declines in
the Indian Ocean have been linked to targeted
sheries (Anderson & Ahmed, 1993; Hanfee
2001; Theberge & Dearden, 2006). In Western
Australia, a decline in shark abundance and
average body length based on tourism operator-
collected data was attributed to shing in other
areas of this population’s range (Bradshaw et
al., 2007 & 2008); however, this decline has
been debated (Holmberg et al., 2008, 2009).
The need to quantify the population abundance
of whale sharks at both local and regional
(oceanic) scales thus remains a priority for
conservation management.
In the Western Indian Ocean, whale sharks
are known to aggregate seasonally around
the islands of the Seychelles (Rowat, 1997;
Fowler, 2000; Rowat & Gore, 2007 ). The
temporal and spatial extent of their distribution
has been monitored intensively since 2001 and
it has been found that the population comprises
a mixture of both site-faithful and migratory
individuals (Rowat et al., 2008; Rowat et
al., 2009a). Seasonal aggregations are also
known to occur in various other coastal sites
throughout the Indian Ocean, specically
Djibouti (Rowat et al., 2006), Madagascar
(Jonahson & Harding, 2007), the Maldives
(Anderson & Ahmed, 1993), Mozambique
(Speed et al., 2008), South Africa (Cliff et al.,
2007), Tanzania (Mahingika & Potenski, 2009)
and Western Australia (Meekan et al., 2006).
Whale sharks need to be uniquely
identied to monitor demographics and
estimate population numbers through
Catch-Mark-Recapture (CMR) modelling
techniques. Sharks can be uniquely identied
by spot patterns on their skin, the area
posterior to the fth gill slit being particularly
suited to this purpose (Arzoumanian et al.,
2005; Speed et al., 2007). With the advent of
digital photography, underwater photography
has become cheaper and easier; digital
image les are also readily manipulated
which has promoted the development and
use of photo-identication software. Once
individuals are uniquely identied, their re-
sighting in subsequent years can be used to
develop population abundance estimates. On
a regional scale, comparisons can be made
between different sites to see if individuals
frequent multiple aggregation sites, thereby
providing an indirect examination of regional-
scale migrations and potentially enabling
large-scale population estimates.
METHODS
Study area
Seychelles
The study area has been previously described
(Rowat et al., 2009a, b). Briey, the granitic
islands of Seychelles are situated on a shallow
continental plateau at 4° S and 55° E in the path
of the westward owing Southern Equatorial
Current in the Western Indian Ocean (New
et al., 2005). From June to October, seasonal
winds blow from the southeast, resulting
in localised primary productivity and the
appearance of whale sharks and other
planktivores such as manta and devil rays
(Manta birostris and Mobula spp.). The study
area was the coastal zone around the island
of Mahe extending to a maximum of 4 km
offshore.
Regional Data
Whale shark photo-ID data were exchanged
with other research programmes operating
in Djibouti, Mozambique and Tanzania to
facilitate regional comparisons.
Identication Studies - Seychelles
Aerial surveys were undertaken from a delta-
wing micro-light aircraft (Aquilla II, Solo
Wings, South Africa) by experienced pilots
trained in aerial survey techniques. Survey
teams were directed to individual sharks by
radio communication with the spotter aboard
the aircraft. During the peak season, aerial and
boat surveys were carried out on a daily basis,
conditions permitting (Rowat et al., 2009b).
186 K. BROOKS ET AL
Wherever possible, sharks encountered
were sexed by the presence (in males) or
absence (in females) of pelvic claspers, sizes
were estimated by an experienced observer,
and any prominent scars or features were
noted and photographed for identication
purposes. From 2001 – 2004, sharks were
opportunistically tagged with marker tags
(Rowat et al., 2009a). The focal area for
photo-identication was the area posterior
to the gill slits where the spot patterns of
the sharks have been found to be unique to
each individual (Arzoumanian et al., 2005;
Meekan et al., 2006; Speed et al., 2007). The
patterns on the left and right of each shark,
however, are different (Speed et al., 2007)
and therefore both sides were photographed
to prevent duplicate entries in the database.
From 2001 to 2004, photographs were
taken opportunistically for potential photo-
identication; from 2004, affordable digital
underwater cameras increased the number of
images collected.
Digital images were matched using the
computer program I3S (van Tienhoven et al.,
2007), which is an effective tool for semi-
automated photo-identication of whale
sharks (Speed et al., 2007). I3S allows the user
to ’ngerprint’ the spot patterns on the skin of
a whale shark and compare these to similarly
ngerprinted images in the database to see if
the shark has been previously photographed.
The images of the area behind the gill slit
were opened with the I3S program and three
reference points were plotted at (1) the top of
the fth gill slit, (2) the posterior-most point
where the pectoral reaches on the body and
(3) the bottom of the fth gill slit. Specifying
these reference points allowed the program
to re-scale or rotate images as required to
ensure standardised comparisons regardless
of photographer orientation or distance from
the shark. The spots on the shark’s anks
were then marked, allowing the program to
calculate the position of each marked spot
relative to the reference points and to compare
the marked spot’s position, through linear
transformation, with its potential ‘pair’ on
each image in the database (van Tienhoven
et al., 2007). A ‘score’ was derived from
the sum of the distances between the paired
spots divided by the number of pairs. The
program presented to the user a list of the
top fty highest-ranked matches and the user
visually analysed these matches, considering
differences in spot selection, other patterning
and scarring, to conrm the nal selection.
A database of sightings for each individual
shark was compiled for tagged individuals
and those with I3S identities. This was used
to create a combined inter-annual history for
CMR models to estimate population size.
Photo-identities of both the left and right side
were not available for all individuals with I3S
identities and so left-side identities were used
because these were more common than those
for the right-side (381 cf 330).
Population estimates were made using
conventional CMR modelling software. To
estimate the population, assuming a closed
population with no net immigration or
emigration (demographic closure), we used
the program CAPTURE (Otis et al., 1978).
This provided goodness-of-t tests for each
model and the program selected the most
probable model(s) for the dataset.
For estimation of population size using
open population models that do not assume
demographic closure, we used the Cormack-
Jolly-Seber (CJS) model (Schwarz &
Arnason, 1966) in the POPAN option in the
program MARK (White & Burnham, 1999).
The POPAN option in MARK does not offer
a bootstrap goodness-of-t, so a recaptures-
only (CJS) analysis was run in MARK, using
the same data to allow a bootstrap goodness-
of-t to the model.
Identication Studies - Regional Data
Whale shark photo-ID data obtained from
other organisations within the Indian Ocean
were examined and re-processed where
necessary for comparison using I3S. The
ngerprinted images were then compiled
with those from the Seychelles into a
regional database to see if any of the sharks
were observed at multiple sites.
Seeing Spots: Photo-Identication as a Regional Tool for Whale Shark Identication 187
RESULTS
Identication Studies - Seychelles
Population Demographics
A total of 443 individual sharks were
identied using photo ID from 2001-2009.
Of these, the sex of 337 individuals was
established, 278 (82.5%) being male and 59
(17.5%) female. The mean size of individuals
identied each year was 5.8 m 1.2 SD),
with the highest size class frequency being
the 5 7.5 m range (51%) followed by the
<5 m range (38%) (Fig. 1a). However, there
were very few sharks recorded below 4 m
(N=23) and, while there were several >7.5 m
(N=59), there were very few >10 m (N=2).
There was some variation in size between
years and between sexes (Fig. 1b) but this
was not signicant.
188 K. BROOKS ET AL
The annual number of photo IDs collected
rose from 2004, as did the percentage of re-
sighted sharks (Table 1). Overall, 109 (24.6%)
of the 443 individuals have been sighted in
multiple years. Two sharks have been seen
in ve different years, 11 in four years, 22 in
three years and 74 in two years. The longest
time-span of sightings was nine years: four
sharks rst seen in 2001 were also seen 2009
as well as in intervening years. The spot
patterns did not change during this period.
Population estimation
Previous attempts to estimate abundance with
re-sighting data from marker tags produced
estimates with a very high error, largely
due to tag loss or deterioration (Rowat et
al., 2009a). Only 34 photo identities were
captured in 2001 - 2003, so population
estimates were only made using the photo ID
data from 2004 – 2009.
Fig. 1a) Size frequency distribution and b) mean shark length
(with standard error) in Seychelles whale shark aggregations in
2001-2009.
0
1
2
3
4
5
6
7
8
2001 2002 2003 2004 2005 2006 2007 2008 2009
Length (m)
Male
Female
Unsexed
a)
b)
Closed population
models generated using
the program CAPTURE
on the latter photo-
identication data indicated
that a model allowing for
variation in the records
due to time, behaviour and
heterogeneity {m(tbh)} was
the most appropriate, but no
population estimator was
available for this model;
also, the data violated the
assumption of closure (Z =
-5.395; P<0.001).
In the program MARK,
candidate models are
ranked by the likelihood of
the goodness-of-t of the
data (c) to the individual
models based on Akaike’s
Information Criterion (AIC)
values and weights (Akaike,
1973). Bootstrap goodness-
of-t is not available with the
POPAN model and we thus
rst modelled the data with
Seeing Spots: Photo-Identication as a Regional Tool for Whale Shark Identication 189
the recaptures-only model that has this option.
The model {Φ(.) p(t)} with time-dependent
variability was ranked the highest. However,
the bootstrap goodness-of-t simulation
yielded some evidence of over-dispersion (p =
0.022), indicating that the probability of capture
was not uniform. Using the routine provided
within MARK, the calculated over-dispersion
was ĉ =1.253. Adjusting the AIC accordingly
within the POPAN open population models, an
abundance estimate of 469 to 557 sharks (95%
C.I., S.E. = 22) was ranked the highest. This
was based on the constant model {Φ(.) p(.)β
(.)N(.)} (Table 2), with a high probability of
capture; however, the level of entry into the
population could not be estimated.
Identication Studies - Regional)
Population Demographics
A total of 1069 individual sharks were identied
in the western Indian Ocean, Seychelles,
Djibouti, Mozambique and Tanzania. This
dataset includes individuals identied in 2009
in Djibouti, Mozambique and Seychelles. In
all three of these aggregations, the number
of males was much greater than females, 76-
83% of the individuals being male (Table 3).
In the Djibouti aggregation, it was possible
to generate a frequency distribution of size
classes (Fig. 2) that indicated that 81%
(n=133) of the individuals identied were
between 3 m and 5 m, while a further 15%
(n=25) were <3 m.
Population distribution
There were no photo-identication matches
between the different geographic sites. In
Tanzania, 66 whale sharks had been identied
by researchers with marker tags over a three
year period (Mahingika & Potenski, 2009);
however, only three sharks could be used
for photo-identication using I3S, none of
which matched any of the other sharks in
the combined regional data set. None of the
sharks tagged in Tanzania were sighted during
monitoring activities in the Seychelles.
Table 1. Whale shark photo-identication records for Seychelles for 2001- 2009 with details of new
records and re-sightings from previous years.
2001 2002 2003 2004 2005 2006 2007 2008 2009
Sharks identied (N) 15 0 24 19 114 186 88 68 88
New identications 15 0 23 19 108 146 49 37 46
Old identications 0 0 1 0 6 40 39 31 42
% Re-sightings 0% 0% 4% 0% 5% 21% 44% 46% 48%
Table 2. Seychelles whale shark population estimates and parameters derived from Cormack-Jolly-Seber
open population model (POPAN option) for the combined photo-identication and tag data for 2004 - 2009,
with estimates of apparent survival (Φ),capture probability (p), probability of entry into the population (β)
and population size (N)
Real Function Parameters of {Φ(.) p(.)β (.) N(.)}
Parameter Estimate Standard error 95% Condence interval
Lower Upper
1: Φ 0.382 0.021 0.343 0.423
2: p 0.739 0.024 0.690 0.783
3: β 1.000 <0.001 <0.001 1.000
4: N 506.218 22.11 469.299 556.863
DISCUSSION
Although monitoring techniques and intensities
differed substantially between the locations
considered in this study, the results show that
the use of standardised photo-identication
protocols and software processing enables
implementation of inter-site comparisons on
a regional scale. While this is the rst attempt
at a regional comparison, and there may be
further regional photo-identities that can be
included, to date, photo-matching has yet to
show movement of sharks away from the
aggregation found in the Seychelles.
All of the aggregations at the sites included
in this comparison were dominated by
immature male sharks. In excess of 75% of the
population were males, with an average size
of <8 m in each of the aggregations. Analysis
of size frequency classes of Seychelles sharks
showed that there has been little variation in
the sizes of sharks reported and that there are
few small juvenile or adult-sized sharks of >8
m (Fig. 1a). Thus, it appears that, as the sharks
reach adult size, they leave this aggregation.
In comparison, in the Djibouti aggregation,
the average size of sharks was 3.7 m, with
15% of identied individuals being <3 m,
only 5% between 5-7.5 m and no larger
sharks; as such, this may indicate that there is
a further size segregation between neonatal-
sized individuals (1-2 m) and those found in
the coastal aggregations (4-8 m).The average
size and sex ratio of sharks in Seychelles
were similar to those reported in Australia
(Bradshaw et al., 2007), Belize (Heyman et
al., 2001), Mozambique (Simon Pierce pers.
comm.) and Maldives (Riley et al., 2010).
These sites all have juvenile male-dominated
populations. This poses the question as to the
location of the adult sharks.
The problem of poor retention of
conventional tags on this species has previously
been noted (Graham & Roberts 2007). Photo-
identication has provided an estimate of tag
retention time and shown the effects this has
on the estimation of population abundance
(Rowat et al., 2009a). Within the Seychelles
aggregation, the addition of a further two years
of photo-identication data has conrmed the
order of magnitude of the previous population
estimate, with lower margins of error: in 2004
2007 the estimate was 348-488 (95% C.I.,
S.E. = 34) compared to 472 – 561 (95% C.I.,
S.E.= 22) in 2004 – 2009 (Rowat et al., 2009a).
This indicates that, in absolute terms, a small
population of whale sharks is using Seychelles
waters. As with previous estimates, the tests for
closure of the population were violated and the
rate of entry into the population could not be
estimated. These ndings are corroborated by
the results of satellite tracking studies that have
shown that whale sharks move considerable
distances away from the Seychelles (Rowat &
Gore, 2007). However, the regional comparison
made here indicated that there were no matches
between the Seychelles, Djibouti, Mozambique
190 K. BROOKS ET AL
Table 3. Basic population demographics for whale sharks photo-identied in three regional aggregations.
Sharks (N) % Male % Female Mean size (m)
Seychelles 443 83 17 5.7
Mozambique 366 76 24 6.5
Djibouti 257 83 17 3.7
Fig. 2. Size frequency distribution of whale
sharks in the Djibouti aggregation in 2009.
and Tanzania populations, suggesting that
the major “known” aggregation sites in the
Western Indian Ocean are not sharing whale
sharks.
Recent work on the genetic diversity of
whale sharks based on haplotype frequency of
complete mitochondrial DNA control regions
has shown little evidence of geographical
clustering (Castro et al., 2007). This is
corroborated by microsatellite studies of
specimens from the Caribbean, Pacic and
Indian Oceans (Schmidt et al., 2009). There
was some evidence of separation between
Atlantic and Indian Ocean samples, but not
between Indian Ocean and Pacic samples
(Castro et al., 2007). Although sample sizes in
both these studies were relatively small, these
ndings support those of satellite tracking that
show widespread movements of sharks away
from the Seychelles (Rowat & Gore, 2007).
This would tend to promote interbreeding,
at least on ocean-basin scales, leading to
low levels of genetic differentiation between
regions. However, the high re-sighting rate
does indicate at least seasonal philopatric
behaviour.
Of note, most of the samples tested in both
these genetic studies were taken from known
aggregation sites and, where recorded, the sharks
ranged from 2.5 - 13.5 m in length, the average
being 6.25 m (Schmidt et al., 2009). It has been
suggested that the low genetic diversity and lack
of structure between geographically separated
populations is an indication of high maternal
gene ow caused by movement of breeding
females (Bradshaw, 2007). This situation may,
however, be complicated by a bias in sampling
mainly immature individuals at what are almost
certainly feeding aggregation sites (Heyman et
al., 2001; Nelson & Eckert, 2007; Bradshaw et
al., 2007): immature individuals from different
breeding populations may aggregate at these
feeding sites, thereby masking population
separation.
The aforementioned genetic studies
both attempted to estimate the effective
(i.e. breeding) population size based on
generational mutation rates. These estimates
ranged from 27,401–179,794 (Schmidt et al.,
2009) to 119,000–238,000 females (Castro
et al., 2007), although the authors of both
studies urged caution in using these values
because of the assumptions they made and
the small sample sizes. These estimates of the
global population appear to be at odds with the
population estimate presented here and one for
Ningaloo Reef in Western Australia (Meekan
et al., 2006), and strongly suggest that transient
feeding aggregations do not comprise the
only or even the principle communities of this
species (Castro et al., 2007).
These ndings therefore reinforce the
importance of implementing more formal
population monitoring in other areas, both
within the region and globally. Presently,
there are very few locations where adult
sharks or pregnant females are found and
very little is known of small sharks under
3 m. Until such time that adult (breeding)
groups are identied and persistent questions
regarding their life-history are answered, in-
depth and consistent long-term monitoring of
shallow-water aggregations of these sharks
is one of the only ways to estimate the status
of the species. The results also emphasise
the need for an ocean-wide approach to the
conservation and management of this, the
largest extant shark in the world.
Photo-identication can play a useful
role in answering some of these questions.
However, due to the fact that whale sharks
are slow-growing and are known to frequent
particular aggregation sites for long periods,
photo-identication cannot be used in
isolation as it appears that once they leave
these aggregations, the sharks are seldom
seen again. These methods need to be used
on a long-term basis and in conjunction
with other monitoring methods, such as
aerial surveys and satellite tagging, to obtain
information regarding whale shark migrations
and behaviour away from the aggregation
sites. Similarly, genetic studies may yet
show regional population and even familial
relationships if carried out at sufcient
intensity. Monitoring needs to be expanded
regionally and, in particular, to areas known
to have different population demographics.
Seeing Spots: Photo-Identication as a Regional Tool for Whale Shark Identication 191
192 K. BROOKS ET AL
Data currently being collected in Djibouti
may, in years to come, show if the smaller
sharks there join other aggregations of larger
sharks when they mature or whether there
are as yet unknown aggregations in the area.
Photo-identities can be captured by personnel
with minimal levels of training, which
broadens the opportunities for data capture.
As long as suitable photos are taken, identities
can be established, thus enabling the public
to participate in whale shark identication
programmes and promote broad-scale
regional comparisons.
AcknowledgmentsOur thanks to the interns
and volunteers of the Marine Conservation
Society Seychelles for their hard work, to
the general public and diving community
in Mozambique who assisted with data
collection, and to the staff and volunteers of
Megaptera who assisted in Djibouti. Thanks
also to Corey Bradshaw and Mark Meekan for
assistance with population estimate modelling.
Funding was provided by the Save Our
Seas Foundation, the Global Environmental
Facility (under the Seychelles Marine
Ecosystem Management Project), and through
public donations. This programme is run with
permission from the Ministry of Environment
and Natural Resources, Seychelles, and in
accordance with the “Wild Animals (whale
shark) Protection Regulations, 2003”.
Research work in Mozambique was supported
by Casa Barry Lodge, Project AWARE
Foundation (International), Rufford Small
Grants, Idea Wild and Tofo Scuba. Thanks
to Andrea Marshall, Andrew Currey and
Nobina Morimoto for their assistance with
the collection and processing of this dataset.
Activities in Djibouti were partially supported
by the Fondation Nature et Découvertes.
The manuscript was greatly enhanced by the
comments of an anonymous reviewer.
REFERENCES
Akaike H, 1973 Information theory as an
extension of the maximum likelihood
principle. In: Petrov B, Csaki F (eds)
Proceedings of the Second International
Symposium on Information Theory,
Budapest, Hungary, pp. 267-281.
Anderson RC, Ahmed H (1993) Shark
sheries of the Maldives, Ministry of
Fisheries and Agriculture, Maldives,
and FAO, Rome, 81pp
Arzoumanian Z, Holmberg J, Norman
B (2005) An astronomical pattern-
matching algorithm for computer-
aided identication of whale sharks
Rhincodon typus. Journal of Applied
Ecology 42: 999-1011
Bradshaw CJA (2007) Swimming in the deep
end of the gene pool - global population
structure of an oceanic giant. Molecular
Ecology 16: 5111-5113
Bradshaw CJA, Mollet HF, Meekan MG
(2007) Inferring population trends for
the world’s largest sh from mark-
recapture estimates of survival. Journal
of Animal Ecology 76: 480-489
Bradshaw CJA, Fitzpatrick BM, Steinberg
CC, Brook BB, Meekan MG (2008)
Decline in whale shark size and
abundance at Ningaloo Reef over the
past decade: The world’s largest sh is
getting smaller. Biological Conservation
141: 1894-1905
Castro ALF, Stewart BS, Wilson SG, Hueter
RE, Meekan MG, Motta PJ, Bowen
BW, Karl SA (2007) Population
genetic structure of Earth’s largest sh,
the whale shark (Rhincodon typus).
Molecular Ecology 16: 5183-5192
Cliff G, Anderson-Reade MD, Aitken AP,
Charter GE, Peddemors VM (2007)
Aerial census of whale sharks on the
northern KwaZulu-Natal coast, South
Africa. Fishery Research 84: 41-46
Seeing Spots: Photo-Identication as a Regional Tool for Whale Shark Identication 193
Fowler S (2000) Whale Shark, Rhincodon
typus, policy and research scoping
study. Nature Conservation Bureau,
Newbury, UK. 27pp
Graham R, Roberts CM (2007) Assessing the
size, growth rate and structure of a seasonal
population of whale sharks (Rhincodon
typus Smith 1828) using conventional
tagging and photo identication. Fishery
Research 84: 71-80
Hanfee F (2001) Trade in whale shark and its
products in the coastal state of Gujarat,
India. TRAFFIC, New Delhi, India.
48pp
Heyman W, Graham R, Kjerfve B, Johannes
RE (2001) Whale sharks Rhincodon
typus aggregate to feed on sh spawn in
Belize. Marine Ecology progress Series
215: 275-282
Holmberg J, Norman B, Arzoumanian Z
(2008) Robust, comparable population
metrics through collaborative photo-
monitoring of whale sharks Rhincodon
typus. Ecological Applications 18: 222-
233
Holmberg J, Norman B, Arzoumanian Z
(2009) Estimating population size,
structure, and residency time for
whale sharks Rhincodon typus through
collaborative photo-identication.
Endangered Species Research 7: 39-53
IUCN (2009) IUCN Red List of Threatened
Species. Version 2009.1, International
Union for the Conservation of
Nature and Natural Resources. www.
iucnredlist.org. last accessed 18.09.09
Jonahson M, Harding S (2007) Occurrence
of whale sharks (Rhincodon typus) in
Madagascar. Fishery Research 84: 132-
135
Mahingika H, Potenski M (2009) Whale shark
monitoring via observational surveys
in Maa Island, Tanzania. . In: Sixth
Scientic Symposium of WIOMSA,
Reunion
Meekan MG, Bradshaw CJA, Press M,
McLean C, Richards A, Quasnichka S,
Taylor JA (2006) Population size and
structure of whale sharks (Rhincodon
typus) at Ningaloo Reef, Western
Australia. Marine Ecology Progress
Series 319: 275-285
Nelson J, Eckert S (2007) Foraging ecology
by whale sharks (Rhincodon typus)
within Bahia de los Angeles, Baja
California Norte, Mexico. Fishery
Research 84: 47-64
New A, Stanseld K, Smythe-Wright D,
Smeed A, Evans A, Alderson A (2005)
Physical and biochemical aspects of
the ow across the Mascarene plateau
in the Indian Ocean. Philosophical
Transactions of the Royal Society A
363: 151-168
Otis DL, Burnham KP, White GC (1978)
Statistical inference for capture data
on closed animal populations. Wildlife
Monographs 62: 1-135
Pollock KH, Marsh H, Lawler IR,
Alldredge MW (2006) Estimating
animal abundance in heterogenous
environments: an application to aerial
surveys for dugongs. Journal of Wildlife
Management 70: 255-262
Riley MJ, Hale MS, Harman A, Rees
RG, 2010. Analysis of whale shark
Rhincodon typus aggregations near
South Ari Atoll, Maldives Archipelago.
Aquatic Biology 8: 145-150
Rowat D (1997) Seychelles whale shark
tagging project - pilot project report.
Phelsuma, Nature Protection Trust of
Seychelles 5: 77-80
Rowat D, Meekan MG, Engelhardt U,
Pardigon B, Vely M (2006) Aggregation
of juvenile whale shark (Rhincodon
typus) in the Gulf of Tadjoura, off
Djibouti. Environmental Biology of
Fishes. 80: 465-472
194 K. BROOKS ET AL
Rowat D, Gore M (2007) Regional scale
horizontal and local scale vertical
movements of whale sharks in the
Indian Ocean off Seychelles. Fishery
Research 84: 32-40
Rowat D, Speed CW, Meekan MG, Gore
M, Bradshaw CJA (2009a) Population
abundance and apparent survival
estimates of the Seychelles whale shark
aggregation. Oryx. 43: 591-598
Rowat D, Gore M, Meekan MG, Lawler IR,
Bradshaw CJA (2009b) Aerial survey as
a tool to estimate whale shark abundance
trends. Journal of Experimental Marine
Biology and Ecology 368: 1-8
Schmidt JV, Schmidt LV, Ozer F, Ernst RE,
Feldheim KA, Ashley MV, Levine M
(2009) Low genetic differentiation
across three major ocean populations of
the whale shark, Rhincodon typus. PLoS
One. 4: e4988
Schwarz CJ, Arnason AN (1966) A general
methodology for the analysis of
capture-recapture experiments in open
populations. Biometrics. 52: 860-873
Smith A, (1828). Description of new, or
imperfectly known objects of the animal
kingdom, found in the south of Africa.
African Commercial Advertiser 3pp
Speed CW, Meekan MG, Bradshaw CJA
(2007) Spot the match – wildlife photo-
identication using information theory.
Frontiers in Zoology 4: 1184-1182
Theberge MM, Dearden P (2006) Detecting a
decline in whale shark Rhincodon typus
sightings in the Andaman Sea, Thailand,
using ecotourist operator-collected data.
Oryx. 40: 337-342
White GC, Burnham KP (1999) Program
MARK: survival estimation from
populations of marked animals. Bird
Study. 46: 120-13
  • ... Large-scale population genetics studies on whale sharks have found no defined structure within the Indo-Pacific region, indicating that such aggrega- tions are broadly connected over evolutionary time scales (Schmidt et al. 2007). However, over shorter periods, connectivity studies using photo-identifica- tion ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) and satellite tags (Berumen et al. 2014, Vignaud et al. 2014, Robinson et al. 2017) in this region have found minimal connectivity between these areas. ...
    ... Photo-ID is routinely used for monitoring whale shark population structure, abundance and connec- tivity (Graham & Roberts 2007, Holmberg et al. 2009, Brooks et al. 2010, Norman et al. 2017). The unique and stable skin colouration pattern of whale sharks ( Arzoumanian et al. 2005, Marshall & Pierce 2012 allows individual sharks to be identified and re-iden- tified over decadal time-scales ( Norman et al. 2017). ...
    ... While this study only considers 3 of the several known whale shark aggregations in the Indian Ocean, broader photo-ID studies ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) have similarly found minimal evidence for connectivity of juvenile and sub-adult whale sharks among coastal aggregations in the region, although Andrzejaczek et al. (2016) noted the high sampling effort required to state this with confidence. Sequeira et al. (2013) also postulated that separate whale shark subpopulations, respectively, may exist in (1) the southern and central Western Indian Ocean, and (2) the northern Western Indian Ocean and Arabian Sea region. ...
    Article
    Full-text available
    Assessing the movements and connectivity of whale sharks Rhincodon typus through their range is difficult due to high individual mobility and limited knowledge of their behaviour following dispersal from coastal aggregation sites. Here, we use a large set of photo-identification and stable isotope data (δ15N and δ13C) to test the assumption that sharks frequenting aggregation sites in Mozambique, Tanzania, and Qatar are a mixed stock, as inferred by genetic data. Photo-identification revealed negligible connectivity among aggregation sites and none between the southern and central areas of the Western Indian Ocean (Mozambique and Tanzania) and the Arabian Gulf (Qatar). Sight−resight data indicated that shark movements at each site could be best represented by a model that included emigration, re-immigration, and some mortality or permanent emigration. Although there was high individual variation in the isotope profiles of sharks from each location, comparison with latitudinal isotope data suggests that sharks had shown site fidelity to within a few hundred kilometres of each study area over the period of isotopic integration. Given the Endangered status of whale sharks and regional differences in anthropogenic threat profiles, further studies — and conservation assessment efforts — should consider the possibility that whale shark subpopulations exist over smaller geographical scales than previously documented.
  • ... Large-scale population genetics studies on whale sharks have found no defined structure within the Indo-Pacific region, indicating that such aggrega- tions are broadly connected over evolutionary time scales (Schmidt et al. 2007). However, over shorter periods, connectivity studies using photo-identifica- tion ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) and satellite tags (Berumen et al. 2014, Vignaud et al. 2014, Robinson et al. 2017) in this region have found minimal connectivity between these areas. ...
    ... Photo-ID is routinely used for monitoring whale shark population structure, abundance and connec- tivity (Graham & Roberts 2007, Holmberg et al. 2009, Brooks et al. 2010, Norman et al. 2017). The unique and stable skin colouration pattern of whale sharks ( Arzoumanian et al. 2005, Marshall & Pierce 2012 allows individual sharks to be identified and re-iden- tified over decadal time-scales ( Norman et al. 2017). ...
    ... While this study only considers 3 of the several known whale shark aggregations in the Indian Ocean, broader photo-ID studies ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) have similarly found minimal evidence for connectivity of juvenile and sub-adult whale sharks among coastal aggregations in the region, although Andrzejaczek et al. (2016) noted the high sampling effort required to state this with confidence. Sequeira et al. (2013) also postulated that separate whale shark subpopulations, respectively, may exist in (1) the southern and central Western Indian Ocean, and (2) the northern Western Indian Ocean and Arabian Sea region. ...
    Article
    Assessing the movements and connectivity of whale sharks Rhincodon typus through their range is difficult due to high individual mobility and limited knowledge of their behaviour following dispersal from coastal aggregation sites. Here, we use a large set of photo-identification and stable isotope data (δ¹⁵N and δ¹³C) to test the assumption that sharks frequenting aggregation sites in Mozambique, Tanzania, and Qatar are a mixed stock, as inferred by genetic data. Photo-identification revealed negligible connectivity among aggregation sites and none between the southern and central areas of the Western Indian Ocean (Mozambique and Tanzania) and the Arabian Gulf (Qatar). Sight−resight data indicated that shark movements at each site could be best represented by a model that included emigration, re-immigration, and some mortality or permanent emigration. Although there was high individual variation in the isotope profiles of sharks from each location, comparison with latitudinal isotope data suggests that sharks had shown site fidelity to within a few hundred kilometres of each study area over the period of isotopic integration. Given the Endangered status of whale sharks and regional differences in anthropogenic threat profiles, further studies — and conservation assessment efforts — should consider the possibility that whale shark subpopulations exist over smaller geographical scales than previously documented.
  • ... Where they go, and the underlying drivers of this rapid turnover, remain uncertain. Although whale sharks are also seen in nearby Tanzania, Seychelles and Djibouti, photo-identification has shown limited connectivity among those sites ( Norman et al., 2017;Brooks et al., 2010;Andrzejaczek et al., 2016). Despite their well-documented ability to move long distances (Hueter, Tyminski & De la Parra, 2013;Hearn et al., 2016), including from Praia do Tofo ( Brunnschweiler et al., 2009), in the Indian Ocean there have been few examples of whale sharks being re-sighted outside the geographic region where they were first identified ( Norman et al., 2017). ...
    ... obs., 2015), and entanglements are commonly reported ( Speed et al., 2008;S Pierce, 2017, unpublished data). Whale sharks are a valuable focal species in marine tourism off Praia do Tofo and adjacent areas ( Pierce et al., 2010;Tibiriçá et al., 2011;Haskell et al., 2015). The species received formal protection in Mozambique and, separately, were listed on Appendix I of the Convention of Migratory Species-which requires prohibition of take by signatory countries (which includes Mozambique)-during 2017. ...
    ... Hence, while our tracks were relatively short and did not span the whole year, the general pattern may apply throughout the year. The narrow shelf waters around Praia do Tofo were a preferred habitat for whale sharks in the region in our study, which is further corroborated by photo-identification and tourism studies ( Pierce et al., 2010;Haskell et al., 2015;Rohner et al., 2015b). However, our tagging data also show that the core use area for whale sharks in Mozambique is larger than previously reported, and larger than in some other, more defined whale shark aggregations that exploit specific and localised ephemeral prey sources or biological events ( Heyman et al., 2001;Robinson et al., 2013;Rohner et al., 2015a). ...
    Article
    Full-text available
    The whale shark Rhincodon typus is an endangered, highly migratory species with a wide, albeit patchy, distribution through tropical oceans. Ten aerial survey flights along the southern Mozambican coast, conducted between 2004–2008, documented a relatively high density of whale sharks along a 200 km stretch of the Inhambane Province, with a pronounced hotspot adjacent to Praia do Tofo. To examine the residency and movement of whale sharks in coastal areas around Praia do Tofo, where they may be more susceptible to gill net entanglement, we tagged 15 juveniles with SPOT5 satellite tags and tracked them for 2–88 days (mean = 27 days) as they dispersed from this area. Sharks travelled between 10 and 2,737 km (mean = 738 km) at a mean horizontal speed of 28 ± 17.1 SD km day ⁻¹ . While several individuals left shelf waters and travelled across international boundaries, most sharks stayed in Mozambican coastal waters over the tracking period. We tested for whale shark habitat preferences, using sea surface temperature, chlorophyll- a concentration and water depth as variables, by computing 100 random model tracks for each real shark based on their empirical movement characteristics. Whale sharks spent significantly more time in cooler, shallower water with higher chlorophyll- a concentrations than model sharks, suggesting that feeding in productive coastal waters is an important driver of their movements. To investigate what this coastal habitat choice means for their conservation in Mozambique, we mapped gill nets during two dedicated aerial surveys along the Inhambane coast and counted gill nets in 1,323 boat-based surveys near Praia do Tofo. Our results show that, while whale sharks are capable of long-distance oceanic movements, they can spend a disproportionate amount of time in specific areas, such as along the southern Mozambique coast. The increasing use of drifting gill nets in this coastal hotspot for whale sharks is likely to be a threat to regional populations of this iconic species.
  • ... Where they go, and the underlying drivers of this rapid turnover, remain uncertain. Although whale sharks are also seen in nearby Tanzania, Seychelles and Djibouti, photo-identification has shown limited connectivity among those sites ( Norman et al., 2017;Brooks et al., 2010;Andrzejaczek et al., 2016). Despite their well-documented ability to move long distances (Hueter, Tyminski & De la Parra, 2013;Hearn et al., 2016), including from Praia do Tofo ( Brunnschweiler et al., 2009), in the Indian Ocean there have been few examples of whale sharks being re-sighted outside the geographic region where they were first identified ( Norman et al., 2017). ...
    ... obs., 2015), and entanglements are commonly reported ( Speed et al., 2008;S Pierce, 2017, unpublished data). Whale sharks are a valuable focal species in marine tourism off Praia do Tofo and adjacent areas ( Pierce et al., 2010;Tibiriçá et al., 2011;Haskell et al., 2015). The species received formal protection in Mozambique and, separately, were listed on Appendix I of the Convention of Migratory Species-which requires prohibition of take by signatory countries (which includes Mozambique)-during 2017. ...
    ... Hence, while our tracks were relatively short and did not span the whole year, the general pattern may apply throughout the year. The narrow shelf waters around Praia do Tofo were a preferred habitat for whale sharks in the region in our study, which is further corroborated by photo-identification and tourism studies ( Pierce et al., 2010;Haskell et al., 2015;Rohner et al., 2015b). However, our tagging data also show that the core use area for whale sharks in Mozambique is larger than previously reported, and larger than in some other, more defined whale shark aggregations that exploit specific and localised ephemeral prey sources or biological events ( Heyman et al., 2001;Robinson et al., 2013;Rohner et al., 2015a). ...
    Article
    The whale shark Rhincodon typus is an endangered, highly migratory species with a wide, albeit patchy, distribution through tropical oceans. Ten aerial survey flights along the southern Mozambican coast, conducted between 2004–2008, documented a relatively high density of whale sharks along a ~200 km stretch of the Inhambane Province, with a pronounced hotspot adjacent to Praia do Tofo. To examine the residency and movement of whale sharks in coastal areas around Praia do Tofo, where they may be more susceptible to gill net entanglement, we tagged 15 juveniles with SPOT5 satellite tags and tracked them for 2–88 days (mean = 27 days) as they dispersed from this area. Sharks travelled between 10 and 2,737 km (mean = 738 km) at a mean horizontal speed of 28 ± 17.1 SD km day ⁻¹ . While several individuals left shelf waters and travelled across international boundaries, most sharks stayed in Mozambican coastal waters over the tracking period. We tested for whale shark habitat preferences, using sea surface temperature, chlorophyll- a concentration and water depth as variables, by computing 100 random model tracks for each real shark based on their empirical movement characteristics. Whale sharks spent significantly more time in cooler, shallower water with higher chlorophyll- a concentrations than model sharks, suggesting that feeding in productive coastal waters is an important driver of their movements. To investigate what this coastal habitat choice means for their conservation in Mozambique, we mapped gill nets during two dedicated aerial surveys along the Inhambane coast and counted gill nets in 1,323 boat-based surveys near Praia do Tofo. Our results show that, while whale sharks are capable of long-distance oceanic movements, they can spend a disproportionate amount of time in specific areas, such as along the southern Mozambique coast. The increasing use of large-mesh gill nets in this coastal hotspot for whale sharks is likely to be a threat to regional populations of this iconic species.
  • ... Limited data are available on whale shark move- ments within the WIO, although genetic data support a single subpopulation within the Indo-Pacific (Castro et al. 2007, Schmidt et al. 2009, Vignaud et al. 2014). However, international photo-identification comparisons have shown limited connectivity be - tween the known feeding areas in the region, which include Djibouti, the Maldives, Mozambique, the Seychelles, South Africa and Tanzania ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) along with the Arabian Gulf and Gulf of Oman ( Robinson et al. 2016). The few published satellite tracks from the WIO have not shown significant interchange between these feeding areas ( Gifford et al. 2007, Rowat et al. 2007, Brunnschweiler et al. 2009, Rohner et al. 2018). ...
  • ... Limited data are available on whale shark move- ments within the WIO, although genetic data support a single subpopulation within the Indo-Pacific (Castro et al. 2007, Schmidt et al. 2009, Vignaud et al. 2014). However, international photo-identification comparisons have shown limited connectivity be - tween the known feeding areas in the region, which include Djibouti, the Maldives, Mozambique, the Seychelles, South Africa and Tanzania ( Brooks et al. 2010, Andrzejaczek et al. 2016, Norman et al. 2017) along with the Arabian Gulf and Gulf of Oman ( Robinson et al. 2016). The few published satellite tracks from the WIO have not shown significant interchange between these feeding areas ( Gifford et al. 2007, Rowat et al. 2007, Brunnschweiler et al. 2009, Rohner et al. 2018). ...
    Article
    Whale sharks Rhincodon typus, the world's largest fish, are routinely sighted offthe northwest coast of Madagascar, particularly offthe island of Nosy Be. Dedicated whale shark tourism has been developing in the area since 2011. During our first dedicated survey, from September to December 2016, we photo-identified 85 individual whale sharks ranging from 3.5 to 8 m in total length (all juveniles). None had been previously identified from surrounding countries. We tagged 8 sharks with tethered SPOT5 tags in October 2016, with tracking durations of 9 to 199 d. Kernel density plots showed that the main activity hotspot for tagged sharks was around the Nosy Be area. Three individuals were resighted back at Nosy Be in late 2017 after having lost their tags. A secondary hotspot was identified offPointe d'Analalava, 180 km southeast of Nosy Be. Five sharks swam offthe shelf into the northeastern Mozambique Channel, between Madagascar and Mayotte, and one of these continued to near the Comoros islands. Two sharks swam to southern Madagascar, with minimum track distances of 3414 and 4275 km. The species is presently unprotected in Madagascar, although a small proportion of the high-use area we identified in this study is encompassed within 2 marine protected areas adjacent to Nosy Be. Whale sharks are globally endangered and valuable to the local economy, so there is a clear rationale to identify and mitigate impacts on the sharks within the 2 hotspots identified here.
  • ... Additionally, the high diversity of habitats ( Miller, Westphalen, Jolley, & Brayford, 2009) and differences in environmental conditions found in Coffin Bay ( Kämpf & Ellis, 2015) likely usually only a few individuals from a population can be studied, resulting in ranging patterns that may not be representative of the entire population ( Castro et al., 2014;Irvine et al., 2014). Photo-ID is a noninvasive mark-recapture technique that has been applied to study the fidelity and space use patterns of several species, including highly mobile marine animals such as sharks ( Brooks, Rowat, Pierce, Jouannet, & Vely, 2010;Domeier & Nasby-Lucas, 2007), whales ...
    Article
    Full-text available
    Information on site fidelity and ranging patterns of wild animals is critical to understand how they use their environment and guide conservation and management strategies. Delphinids show a wide variety of site fidelity and ranging patterns. Between September 2013 and October 2015, we used boat-based surveys, photographic identification, biopsy sampling, clustering analysis, and geographic information systems to determine the site-fidelity patterns and representative ranges of southern Australian bottlenose dolphins (Tursiops cf. australis) inhabiting the inner area of Coffin Bay, a highly productive inverse estuary located within Thorny Passage Marine Park, South Australia. Agglomerative hierarchical clustering (AHC) of individuals’ site-fidelity index and sighting rates indicated that the majority of dolphins within the inner area of Coffin Bay are “regular residents” (n = 125), followed by “occasional residents” (n = 28), and “occasional visitors” (n = 26). The low standard distance deviation indicated that resident dolphins remained close to their main center of use (range = 0.7–4.7 km, X ± SD = 2.3 ± 0.9 km). Representative ranges of resident dolphins were small (range = 3.9–33.5 km², X ± SD = 15.2 ± 6.8 km²), with no significant differences between males and females (Kruskal–Wallis, χ² = 0.426, p = .808). The representative range of 56% of the resident dolphins was restricted to a particular bay within the study area. The strong site fidelity and restricted ranging patterns among individuals could be linked to the high population density of this species in the inner area of Coffin Bay, coupled with differences in social structure and feeding habits. Our results emphasize the importance of productive habitats as a major factor driving site fidelity and restricted movement patterns in highly mobile marine mammals and the high conservation value of the inner area of Coffin Bay for southern Australian bottlenose dolphins.
  • ... Photo-identification (photo-id) makes use of unique, naturally occurring marks, eliminating the need to physically capture or tag the organism [1]. Many species have distinctive, easily identifiable marks, for example, cheetahs are commonly recognized through their pelage spots [2], whisker spot patterns for polar bears [3], pigmentation spots for whale sharks [4], scales of eastern water dragons [5], body patterns of jewelled geckos [6]. Photo-id technique is popular for identification of cetaceans, such as dolphins [7] because it is considered a non-invasive and cost-effective approach; but it requires the animal's fin to show some type of permanent damage around the edge, such as nicks, tears and notches (see Figure 1). ...
  • Thesis
    Full-text available
    Five of the world’s seven species of sea turtles have been documented to use Mozambican habitats. While they are thought to be extensively distributed throughout Mozambican coastal waters, and the offshore waters of the Mozambican Channel, little is known about these populations. Specifically, information about the state and structure of sea turtle populations (population size estimates, species composition, age class distribution, movements of animals into and out of the study area, residency, habitat use and preferences) was scarce or non-existent for Mozambique. Therefore, my research adopted several complementary research techniques to increase knowledge on sea turtles in their foraging grounds and their exposure to human impacts. The major research aim of my thesis is to understand factors related to the distribution, abundance and use of sea turtle populations within the Inhambane region, Mozambique, and use this knowledge to inform and improve conservation and management efforts. In Chapters 2 and 3, I explore the use of citizen science and photo- identification (photo-ID) as tools to facilitate the collection of data on turtles encountered in-water. I found that citizen science is a useful tool for collecting basic biological information, particularly when coupled with photo-ID encounters. While the quality of dive log records was improved by having a few well-trained and consistent contributors, the photo-ID database benefitted from broadened public involvement. Results from the generalised linear modelling of the dive log data (Chapter 3) suggested that sightings and abundance of turtles were influenced by environmental conditions. It was also evident that factors such as visibility and diving depth lead to availability and perception bias in the citizen scientists’ records. It is important to be aware of such biases since they reflect physical environmental diving conditions rather than habitat or behavioural predictors that influence sea turtles. Overall, citizen science coupled with photo-ID datasets provided the first details of Mozambique’s foraging sea turtle populations. In Chapter 3, I described the use of coastal reefs by green and loggerhead sea turtles in Inhambane, Mozambique. Based on population models from the photo-ID dataset, both green and loggerhead populations were small but present year-round. Regardless of species, sea turtles favoured coastal nearshore waters and relatively shallow reef systems, which make them vulnerable to interaction with small scale fisheries (SSF). Impacts of SSF is unlikely to be consistent between species or age-classes. My findings suggest that the long-term residency of late-stage juvenile greens in these nearshore and shallow habitats make them most vulnerable to interactions with SSF. In Chapter 4, I investigated the prevalence of illegal take of sea turtles in a coastal region of Mozambique – the Tofo area, Inhambane Province - and conducted a national-scale literature review. Transect- based sampling in the sand dunes demonstrated that the Tofo area and greater Inhambane peninsula are a hotspot for take of sea turtles. The literature review documented year-round take of sea turtles to occur through much of Mozambique. Use of sea turtles focused on their meat, and it was rare to detect more than an empty carapace or old bones. Small scale fisheries interact with turtles in their favoured habitats (close to reefs systems) in coastal waters, particularly in the south of the primary study area, between Praia do Tofinho and Praia de Rocha. Based on interviews with fishers, the opportunistic take of turtles is prevalent and widespread (Chapter 5) in Inhambane Province. A targeted marine megafauna multi-species fishery exists in the study area. Widespread use of gillnets and long-lines occurs and these fishing gears are favoured because of their non-selectivity and ability to capture turtles and other species. Sea turtle capture in these fisheries is neither bycatch nor accidental. Interviews with fishers (Chapter 5) indicated that the motives and drivers influencing fishers to illegally take sea turtles were variable between communities and individuals. In the two fishing communities surveyed, opportunities for alternative livelihoods were lacking or insufficient to supplement or replace their reliance on fishing activities. Five of the six major drivers identified in Chapter 5 reflect Mozambique’s low socio- economic status. Similarities were evident between the drivers and motives of illegal take of turtles and the terrestrial mammal bushmeat hunting and trade. The majority of fishers had multiple motives for participating in illegal take of sea turtles. Awareness of turtle protection laws amongst fishers was high, although compliance was low. This suggests that simple campaigns to increase awareness of turtle legislation will have little impact in deterring illegal take. Future conservation efforts will need to address food security, livelihoods options and aim to minimise the number of motives an individual fisher or community may have to participate in illegal take. I solicited opinions from local experts to quantitatively rank threats and investigate the context of conservation and management efforts underway in Mozambique (Chapter 6). Consensus of expert opinions revealed the most pressing threats to sea turtles were fisheries-related (bycatch from commercial trawling, SSF bycatch and hunting of nesting turtles). The top- ranking threat, bycatch within the commercial shallow-water prawn trawl industry, could be easily mitigated with effective implementation of pre- existing Turtle Exclusion Device legislation. This is not the case for the other two threats, which given their nature are likely to involve extensive changes/improvements to living standards and Mozambique’s overall socio-economic status. Compliance with sea turtle legislation was weak throughout the country and experts identified improving enforcement efforts as critical. Parallels are evident among the issues that hamper the conservation and management of terrestrial megafauna and marine megafauna. A holistic process will be required to solve large-scale issues (e.g. governance, corruption, compliance) and strengthen overall biodiversity conservation. Given the extremely limited funding allocated to the conservation of sea turtles, the marine environment and limited access to skilled people and resources, a prioritised list of management actions for sea turtle hotspot areas is necessary. I conclude this study by discussing my key findings relating to the sea turtle populations using the Tofo area, the impacts they face and how and where conservation management efforts could be strengthened. I also suggest specific priorities for future research to enhance knowledge of sea turtle populations, socio-economic understanding of SSF and alternative livelihoods. A balance needs to be struck between the environment, economic development and social and cultural values of coastal people in order to achieve sustainable growth whilst preserving marine biodiversity and improving living standards.
  • Article
    Full-text available
    Genetic and modelling studies suggest that seasonal aggregations of whale sharks (Rhincodon typus) at coastal sites in the tropics may be linked by migration. Here, we used photo-identification (photo-ID) data collected by both citizen scientists and researchers to assess the connectedness of five whale shark aggregation sites across the entire Indian Ocean at timescales of up to a decade. We used the semi-automated program I3S (Individual Interactive Identification System) to compare photographs of the unique natural marking patterns of individual whale sharks collected from aggregations at Mozambique, the Seychelles, the Maldives, Christmas Island (Australia) and Ningaloo Reef (Australia). From a total of 6519 photos, we found no evidence of connectivity of whale shark aggregations at ocean-basin scales within the time frame of the study and evidence for only limited connectivity at regional (hundreds to thousands of kilometres) scales. A male whale shark photographed in January 2010 at Mozambique was resighted eight months later in the Seychelles and was the only one of 1724 individuals in the database to be photographed at more than one site. On average, 35% of individuals were resighted at the same site in more than one year. A Monte Carlo simulation study showed that the power of this photo-ID approach to document patterns of emigration and immigration was strongly dependent on both the number of individuals identified in aggregations and the size of resident populations.
  • Conference Paper
    In this paper it is shown that the classical maximum likelihood principle can be considered to be a method of asymptotic realization of an optimum estimate with respect to a very general information theoretic criterion. This observation shows an extension of the principle to provide answers to many practical problems of statistical model fitting.
  • Article
    Full-text available
    In this paper, we analyse long-term whale shark Rhincodon typus sightings collected by ecotourist operators and evaluate the validity of conclusions drawn from the data for scientific and conservation purposes. To date information about the basic ecology and movements of whale sharks is sparse, and only recently has the species received global conservation attention. A dive company in Phuket, Thailand, documented whale shark sightings in the Andaman Sea for 10 years along 300 km of coastline. Whale shark sightings, corrected for effort, dropped by 96% between 1998 and 2001. Combining the seasons from 1992 to 1998, the number of whale shark sightings increased significantly from October to May. The sizes of sharks observed suggest that the majority were juveniles. We discuss the limitations of using ecotourist operators as non-specialist volunteers for data collection but conclude that their use can be beneficial for long-term, broad geographic studies such as this.
  • Article
    Full-text available
    MARK provides parameter estimates from marked animals when they are re-encountered at a later time as dead recoveries, or live recaptures or re-sightings. The time intervals between re-encounters do not have to be equal. More than one attribute group of animals can be modelled. The basic input to MARK is the encounter history for each animal. MARK can also estimate the size of closed populations. Parameters can be constrained to be the same across re-encounter occasions, or by age, or group, using the parameter index matrix. A set of common models for initial screening of data are provided. Time effects, group effects, time x group effects and a null model of none of the above, are provided for each parameter. Besides the logit function to link the design matrix to the parameters of the model, other link functions include the log—log, complimentary log—log, sine, log, and identity. The estimates of model parameters are computed via numerical maximum likelihood techniques. The number of parameters that are estimable in the model are determined numerically and used to compute the quasi-likelihood AIC value for the model. Both the input data, and outputs for various models that the user has built, are stored in the Results database which contains a complete description of the model building process. It is viewed and manipulated in a Results Browser window. Summaries available from this window include viewing and printing model output, deviance residuals from the model, likelihood ratio and analysis of deviance between models, and adjustments for over dispersion. Models can also be retrieved and modified to create additional models. These capabilities are implemented in a Microsoft Windows 95 interface. The online help system has been developed to provide all necessary program documentation.
  • Article
    Full-text available
    Effective approaches for the management and conservation of wildlife populations require a sound knowledge of population demographics, and this is often only possible through mark-recapture studies. We applied an automated spot-recognition program (I3S) for matching natural markings of wildlife that is based on a novel information-theoretic approach to incorporate matching uncertainty. Using a photo-identification database of whale sharks (Rhincodon typus) as an example case, the information criterion (IC) algorithm we developed resulted in a parsimonious ranking of potential matches of individuals in an image library. Automated matches were compared to manual-matching results to test the performance of the software and algorithm. Validation of matched and non-matched images provided a threshold IC weight (approximately 0.2) below which match certainty was not assured. Most images tested were assigned correctly; however, scores for the by-eye comparison were lower than expected, possibly due to the low sample size. The effect of increasing horizontal angle of sharks in images reduced matching likelihood considerably. There was a negative linear relationship between the number of matching spot pairs and matching score, but this relationship disappeared when using the IC algorithm. The software and use of easily applied information-theoretic scores of match parsimony provide a reliable and freely available method for individual identification of wildlife, with wide applications and the potential to improve mark-recapture studies without resorting to invasive marking techniques.