Molecular Ecology (2009) 18, 4841–4853
doi: 10.1111/j.1365-294X.2009.04403.x
Turtle groups or turtle soup: dispersal patterns of
hawksbill turtles in the Caribbean
J. M. BLUMENTHAL,*† F. A. ABREU-GROBOIS,‡ T. J. AUSTIN,* A. C. BRODERICK,†
M . W . B R U F O R D , § M . S . C O Y N E , †– G . E B A N K S - P E T R I E , * A . F O R M I A , § P . A . M E Y L A N , * *
A . B . M E Y L A N †† and B . J . G O D L E Y †
*Department of Environment, Box 486, Grand Cayman KY1-1106, Cayman Islands, †Centre for Ecology and Conservation,
School of Biosciences, University of Exeter Cornwall Campus, Penryn TR10 9EZ, UK, ‡Laboratorio de Genética, Unidad
Académica Mazatlán, Instituto de Ciencias del Mar y Limnologı́a, Mazatlán, Sinaloa 82040, México, §Biodiversity and
Ecological Processes Research Group, School of Biosciences, Cardiff University, Cardiff CF10 3TL, UK, –SEATURTLE.ORG,
Durham, NC 27705, USA, **Natural Sciences, Eckerd College, St Petersburg, FL 33711, USA, ††Florida Fish and Wildlife
Conservation Commission, Fish and Wildlife Research Institute, St Petersburg, FL 33701, USA
Abstract
Despite intense interest in conservation of marine turtles, spatial ecology during the
oceanic juvenile phase remains relatively unknown. Here, we used mixed stock analysis
and examination of oceanic drift to elucidate movements of hawksbill turtles (Eretmochelys imbricata) and address management implications within the Caribbean. Among
samples collected from 92 neritic juvenile hawksbills in the Cayman Islands we detected
11 mtDNA control region haplotypes. To estimate contributions to the aggregation, we
performed ‘many-to-many’ mixed stock analysis, incorporating published hawksbill
genetic and population data. The Cayman Islands aggregation represents a diverse mixed
stock: potentially contributing source rookeries spanned the Caribbean basin, delineating a scale of recruitment of 200–2500 km. As hawksbills undergo an extended phase of
oceanic dispersal, ocean currents may drive patterns of genetic diversity observed on
foraging aggregations. Therefore, using high-resolution Aviso ocean current data, we
modelled movement of particles representing passively drifting oceanic juvenile
hawksbills. Putative distribution patterns varied markedly by origin: particles from
many rookeries were broadly distributed across the region, while others would appear to
become entrained in local gyres. Overall, we detected a significant correlation between
genetic profiles of foraging aggregations and patterns of particle distribution produced
by a hatchling drift model (Mantel test, r = 0.77, P < 0.001; linear regression, r = 0.83,
P < 0.001). Our results indicate that although there is a high degree of mixing across the
Caribbean (a ‘turtle soup’), current patterns play a substantial role in determining genetic
structure of foraging aggregations (forming turtle groups). Thus, for marine turtles and
other widely distributed marine species, integration of genetic and oceanographic data
may enhance understanding of population connectivity and management requirements.
Keywords: conservation genetics, hawksbill, marine turtle, migratory species, mixed stock,
ocean currents
Received 29 June 2009; revision received 21 September 2009; accepted 25 September 2009
Introduction
Many marine vertebrates have life cycles that span wide
spatiotemporal scales – complicating research and
Correspondence: Brendan J. Godley, Fax: 44 (0) 1326 253638;
E-mail: [email protected]
2009 Blackwell Publishing Ltd
management. Recently, molecular techniques have provided insights into patterns of migration and stock resolution in species ranging from fish (Millar 1987) to
porpoises (Andersen et al. 2001) and great whales
(Baker et al. 1999; Witteveen et al. 2004). Such stock resolution issues are particularly important when species
4842 J . M . B L U M E N T H A L E T A L .
are commercially valuable, cross geopolitical boundaries
or have been exploited to the point of endangerment.
Marine turtles exemplify these concerns: a long history
of commercial use has resulted in global declines
(Baillie et al. 2004) and management of marine turtle
populations is complex, as developmental and reproductive migrations may span the territorial waters of
several nations and the high seas (Bowen et al. 1995;
Bolten et al. 1998).
Like many other marine organisms (Mora & Sale
2002) most marine turtle species experience a phase of
oceanic dispersal, which in marine turtles is known as
the ‘lost year’ or ‘lost years’ (Carr 1987). This extended
period of drifting in concert with initially small size
and high mortality inhibits tracking movements of neonates from the time of entry into marine habitat off
natal beaches to the time of recruitment into neritic foraging grounds (Bolten et al. 1998; Lahanas et al. 1998;
Engstrom et al. 2002; Bowen et al. 2004; Luke et al.
2004). Although foraging grounds are typically inhabited by individuals originating from multiple nesting
beaches a behavioural barrier to genetic mixing at nesting beaches is maintained: generations of mature
females return to their natal regions to reproduce (Meylan et al. 1990). This ‘natal homing’ leads to genetic differentiation at mitochondrial loci of nesting populations
over time (hawksbills Eretmochelys imbricata: Broderick
et al. 1994; Bass et al. 1996; loggerheads Caretta caretta:
Encalada et al. 1998; Engstrom et al. 2002; green turtles
Chelonia mydas: Meylan et al. 1990; Allard et al. 1994).
Therefore, mitochondrial DNA haplotypes and haplotype frequencies characteristic of each region serve as
a form of ‘genetic tag,’ linking juveniles in mixed
aggregations with their specific nesting beach origins
(Norman et al. 1994).
In a method originally developed to resolve the origins of salmon stocks (Millar 1987), estimates of marine
turtle origin can be made using a maximum likelihood
(Pella & Milner 1987) or Bayesian (Pella & Masuda
2001) mixed stock analysis (MSA), where haplotype frequencies for genetically mixed aggregations are compared with potential source populations. MSA has
recently proven valuable in elucidating migratory patterns and range states in hawksbill (Bowen et al. 1996,
2007a; Bass 1999; Diaz-Fernandez et al. 1999; Troëng
et al. 2005; Velez-Zuazo et al. 2008), loggerhead (Bowen
et al. 1995; Laurent et al. 1998; Engstrom et al. 2002;
Maffucci et al. 2006) and green turtles (Lahanas et al.
1998; Luke et al. 2004). While mixed stock analysis has
traditionally been used to estimate contributions of
many potential source rookeries to a single foraging
ground (‘many-to-one’ analysis), a new approach has
recently been developed which more realistically estimates the contributions of many rookeries to many for-
aging grounds within a metapopulation context (‘manyto-many’ analysis; Bolker et al. 2007). Caribbean hawksbills represent ideal candidates for many-to-many
analysis: rookeries are sufficiently differentiated (Bass
et al. 1996), populations may be relatively contained in
the Caribbean region (Bowen et al. 1996) and data from
multiple mixed stocks have recently been published
(Bowen et al. 2007a; Velez-Zuazo et al. 2008).
Tagging and satellite tracking have demonstrated
that hawksbill developmental and reproductive migrations span the territorial waters of multiple jurisdictions (Meylan 1999; Horrocks et al. 2001; Troëng et al.
2005; Whiting & Koch 2006; van Dam et al. 2008) and
genetic research is beginning to elucidate links
between nesting beaches and foraging grounds (Bowen
et al. 1996, 2007a; Bass 1999; Diaz-Fernandez et al.
1999; Troëng et al. 2005; Velez-Zuazo et al. 2008). It
has also been determined that Caribbean hawksbill
foraging aggregations show shallow but significant
genetic structure (Bowen et al. 2007a). As in green turtles (Bass et al. 2006), this suggests that rather than the
oceanic phase being a homogeneous mixture (the ‘turtle soup’ model: Engstrom et al. 2002), dispersal is
non-random. However for many populations links
between nesting beaches and foraging areas have not
been identified (Bowen et al. 1996) and little is known
regarding movements of oceanic juveniles during the
lost year or years (Musick & Limpus 1997; Bolten
2003) and factors which drive dispersal during this
phase (Bowen et al. 2007a).
In previous genetic studies of green (Lahanas et al.
1998; Bass & Witzell 2000; Luke et al. 2004) and loggerhead turtles (Engstrom et al. 2002; Reece et al. 2006)
source population size and geographic distance have
proven inconsistent in determining recruitment from
rookeries to foraging grounds. For Caribbean hawksbills, these factors were significantly correlated with foraging aggregation genetic composition, but correlations
were weak and significant only for a small proportion
of individual foraging grounds (Bowen et al. 2007a).
Ocean currents have been postulated as playing a major
role in distribution of oceanic juvenile marine turtles
(Carr 1987) and recent studies have linked genetic composition of marine turtle foraging aggregations to ocean
current patterns (Luke et al. 2004; Okuyama & Bolker
2005; Bass et al. 2006; Carreras et al. 2006). However,
for hawksbills, attempts to account for the role of ocean
currents through incorporation of a current correction
factor (increasing or decreasing geographic distances in
a regression model by 10–20%) did not increase the
strength of correlations (Bowen et al. 2007a).
To evaluate population connectivity for marine organisms, empirical data must be linked with biophysical
modelling of oceanic dispersal (Botsford et al. 2009).
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4843
Generalized ocean current diagrams can aid in consideration of the role of ocean currents in marine turtle
dispersal (Bass et al. 2006; Carreras et al. 2006; Bowen
et al. 2007a) but this method falls short when current
patterns are complex. Trans-Atlantic drift has been
modelled for oceanic juvenile loggerheads (Hays &
Marsh 1997) and the availability of high-resolution
ocean current data for the Caribbean and other regions
now opens up the possibility of modelling dispersal of
oceanic juvenile hawksbills in dynamic ocean currents.
In this study, characterization of stock composition at
the Cayman Islands hawksbill foraging site was
extended by development of a hatchling drift model
incorporating source population sizes and regional
ocean current data to answer fundamental questions of
hawksbill ecology: (i) how are oceanic juvenile hawksbills distributed during the lost year or years? (ii) how
are foraging stocks geographically structured or regionally mixed? (iii) what role might ocean currents play in
determining these patterns?
Materials and Methods
Study area
The Cayman Islands are located in the Caribbean Sea,
approximately 240 km south of Cuba (Fig. 1). The
three low-lying islands are exposed peaks on the
Cayman Ridge formation, with nearly vertical slopes
extending to depths in excess of 2000 m on all sides
(Roberts 1994). Hawksbill nesting, while described as
abundant in historical records (Lewis 1940), appears to
have been largely extirpated in recent decades (Bell
et al. 2007). However, the islands host foraging aggregations of neritic juvenile hawksbill turtles, inhabiting
colonized pavement, coral reef and reef wall habitats
(Blumenthal et al. 2009a, b). For this study, two sampling sites were selected: Bloody Bay, Little Cayman
(19.7N, 81.1W) and western Grand Cayman (19.3N,
81.4W).
Capture and sampling
Juvenile hawksbills (20–60 cm straight carapace length)
were hand-captured in neritic foraging habitat and flipper and PIT tagged according to standard protocols
(Blumenthal et al. 2009b) to prevent repeated sampling
of individuals. Genetic sampling for this study was
undertaken year-round between 2000 and 2003. Skin
biopsies were obtained from a rear flipper with a sterile
4 mm biopsy punch and preserved in a solution of 20%
DMSO saturated with NaCl (Dutton 1996). Blood samples were collected from the dorsal cervical sinus
(Owens & Ruiz 1980) and preserved in lysis buffer
2009 Blackwell Publishing Ltd
Fig. 1 Distribution of hawksbill aggregations studied in the
Caribbean region. Circles represent locations of Caribbean
hawksbill rookeries: filled circles are genetically characterized.
Haplotype frequency data from Antigua, Barbados, Belize,
Costa Rica, Cuba, Mexico, Puerto Rico and the US Virgin
Islands were used in many-to-many mixed stock analysis;
rookeries in Venezuela are presently insufficiently characterized to permit inclusion. Triangles represent genetically characterized hawksbill foraging grounds also incorporated in mixed
stock analysis: Bahamas, Cayman, Cuba, Dominican Republic,
Mexico, Puerto Rico, Texas and the US Virgin Islands. Marine
ecoregions (1) Gulf of Mexico (2) western Caribbean (3) southwestern Caribbean (4) Greater Antilles (5) southern Caribbean
(6) eastern Caribbean (7) Bahamian. Source of published
genetic data: Antigua (Bass 1999), Barbados (Browne et al. in
press), Belize (Bass 1999), Cuba (Diaz-Fernandez et al. 1999),
Mexico (Bass 1999; Diaz-Fernandez et al. 1999), USVI (Bass
1999; Bowen et al. 2007a), Costa Rica (Troëng et al. 2005;
Bowen et al. 2007a) and Puerto Rico (Diaz-Fernandez et al.
1999; Velez-Zuazo et al. 2008).
(100mM Tris-HCl, pH 8, 100mM EDTA, pH 8, 10mM
NaCl and 1–2% SDS: Dutton 1996).
Molecular analysis
Following overnight digestion at 55 C with proteinase
K, DNA extraction was conducted via standard phenol
chloroform extraction (Milligan 1998), Qiagen DNEasy
tissue kit, or a modification of a protocol by Allen et al.
(1998) (Formia et al. 2006, 2007). Fragments of 486 or
550 base pairs (bp) were amplified via polymerase
chain reaction (PCR), using primer pairs TCR6 (Norman
et al. 1994) ⁄ L15926 (Kocher et al. 1989) or LTEi3 ⁄ HDEi1
(LTEi3
5¢-CCTAGAATAATCAAAAGAGAAGG-3¢;
HDEi1 5¢- AGTTTCGTTAATTCGGCAG-3¢; Abreu-Grobois et al. 2006) respectively. PCR reactions [PCR
Ready-To-Go Beads (Amersham) or 1.5 mM MgCl2, 1·
PCR buffer, 200 lM of each dNTP, 0.5 lM of each
primer, 0.5 U Invitrogen Taq DNA polymerase, 1 lL
4844 J . M . B L U M E N T H A L E T A L .
template DNA and sterile H2O to a volume of 10 lL)
were carried out in a Perkin Elmer GeneAmp PCR
system 9700 (3 min at 94 C, 35 cycles of: 45 s at
94 C, 30 s at 55 C and 1.5 min at 72 C, followed by
72 C for 10 min; Formia et al. 2006). PCR products
were cleaned with a Geneclean Turbo Kit (Qbiogene)
or QIAquick spin columns (Qiagen), sequenced in
both directions with a BigDye Terminator Cycle
Sequencing Kit v. 2.0 (Applied Biosystems), purified
via ethanol precipitation and run on an Applied Biosystems model 3100 automated DNA sequencer at
Cardiff University, or analysed at the University of
Florida Sequencing Core on an Amersham Pharmacia
Biotech MegaBACE 1000 capillary array DNA
sequencing unit. Negative controls (template-free PCR
reactions) were utilized throughout to test for contamination.
To identify haplotypes, sequences were aligned with
published hawksbill mtDNA sequences (Bass et al.
1996; Bowen et al. 1996; Bass 1999; Diaz-Fernandez
et al. 1999; Troëng et al. 2005; Bowen et al. 2007a;
Velez-Zuazo et al. 2008; Browne et. al. in press) in Sequencher 2.1.2 (Gene Codes Corp). Haplotype diversity
(h), nucleotide diversity (p, and population pairwise
FST values (10 000 permutations; Kimura 2-parameter
model, a = 0.50) were calculated in Arlequin v. 3.11
(Excoffier et al. 2005).
Mixed stock analysis
Weighted and unweighted many-to-many analyses (Bolker et al. 2007) were also conducted, incorporating all
previously published Caribbean hawksbill genetic data
(haplotype frequencies: Antigua (Bass 1999), Barbados
(Browne et al. in press; windward and leeward rookeries combined), Belize (Bass 1999), Cuba (Diaz-Fernandez
et al. 1999; all foraging grounds combined), Mexico
(Bass 1999; Diaz-Fernandez et al. 1999), USVI (Bass
1999; Bowen et al. 2007a), Costa Rica (Troëng et al.
2005; Bowen et al. 2007a) and Puerto Rico (Diaz-Fernandez et al. 1999; Velez-Zuazo et al. 2008). To enable comparison with previously published rookery haplotypes,
sequences were truncated to 384 bp for analysis. Source
population size (number of nests) was used as a strict
constraint in the ‘weighted’ many-to-many model
(Bolker et al. 2007), with population sizes taken from
Mortimer & Donnelly (2007).
Hatchling drift model
Movement of particles was modelled using high resolution ocean current data from Aviso (Archiving, Validation and Interpretation of Satellite Oceanographic data,
http://www.aviso.oceanobs.com). Aviso geostrophic
velocity vector (GVV) data (maps of absolute geostrophic velocities derived from maps of Absolute
Dynamic Topography combining all satellites, obtained
from http://www.aviso.oceanobs.com) have a spatial
resolution of 1 ⁄ 3 degree and a temporal resolution of
1 day. To model the movement of passively drifting
hawksbill hatchlings, post-hatchlings and oceanic juveniles, virtual particles were released from the coordinates of genetically characterized Caribbean hawksbill
rookeries locations (Fig. 1). To simulate peak hatchling
emergence, particles were released 60 days after the
peak of the nesting season at each location (nesting season data: Chacón 2004; Garduño-Andrade 1999; Moncada et al. 1999), with one particle released per day over a
60-day period and the resulting particle profiles
weighted by source population size (taken from Mortimer & Donnelly 2007). Drift of virtual particles was initiated approximately 50 km offshore from each rookery to
account for the phase of directed hatchling swimming
that occurs prior to beginning passive drift (Hasbún
2002; Chung et al. 2009a,b) and to bring particles into
the area covered by Aviso data, as nearshore currents
are too complex to map reliably. The u (eastward) and v
(northward) ocean current vector components were sampled at the location of each particle from the daily GVV
data file using a bilinear interpolation [Generic Mapping
Tools (GMT 4.11; Wessel & Smith 1991)]. The u and v
values (cm ⁄ s) were converted to m ⁄ day and the particle
advected that distance in the x and y direction, respectively, and a new location obtained for that particle. Particles that left the GVV field (i.e. pushed towards the
coastline) were removed from the model, as there is no
information on how an oceanic juvenile hawksbill would
behave in this situation. Also, while natural mortality
during the hatchling, post-hatchling and oceanic juvenile
phase is high, this factor was not incorporated into the
model. One location per day was saved for each particle
for analysis. All current modelling was carried out using
custom perl scripts, the geod program, part of the
PROJ.4 Cartographic Projections Library (http://
www.remotesensing.org/proj/) and the GMT package.
Comparison of the hatchling drift model with mixed
stock analysis
To examine the possible influence of ocean currents on
patterns of dispersal, we compared foraging aggregation genetic profiles (determined via many-to-many
analysis) with patterns of particle distribution (particle
profiles) produced by the hatchling drift model
(Table 1). To produce particle profiles, we used the Global Maritime Boundaries Database (GMBD) to delineate
Exclusive Economic Zones of nations containing genetically characterized hawksbill foraging aggregations.
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4845
Table 1 Comparison of genetic profiles from many-to-many analysis and particle profiles derived from the hatchling drift model
Cay
Anti
Barb
Belz
Cuba
Mexi
USVI
CR
PR
Tex
Bah
Cuba
PR
USVI
DR
Mex
MSA
Drift
MSA
Drift
MSA
Drift
MSA
Drift
MSA
Drift
MSA
Drift
MSA
Drift
MSA
Drift
6.0
46.1
0.7
21.9
13.2
3.0
0.5
8.7
13.1
40.3
0.0
45.7
0.0
0.3
0.6
0.0
1.9
3.3
0.6
2.4
85.3
2.0
0.3
4.1
0.0
0.0
0.0
0.0
100.0
0.0
0.0
0.0
5.2
22.7
0.8
26.1
28.4
3.7
0.5
12.6
13.2
25.9
0.0
0.0
46.6
0.7
0.5
13.0
2.7
15.7
0.9
44.3
10.5
2.4
0.7
22.9
4.3
13.5
0.0
64.6
17.1
0.2
0.3
0.0
5.3
47.4
1.5
8.0
8.5
8.1
1.4
19.7
18.5
28.6
0.0
0.0
0.0
8.5
0.0
44.4
10.2
27.7
0.7
23.0
14.1
2.3
0.4
21.6
50.2
21.2
0.0
0.0
0.0
26.4
0.0
2.1
4.4
65.7
2.1
10.1
7.7
3.9
1.5
4.7
23.1
35.3
0.0
0.0
0.0
4.0
0.0
37.6
6.7
5.8
0.7
5.8
71.6
2.5
0.4
6.3
3.8
13.8
0.6
0.0
81.3
0.2
0.3
0.0
Rows represent potential source rookeries – Antigua, Barbados, Belize, Cuba, Mexico, US Virgin Islands, Costa Rica and Puerto Rico;
columns represent genetically characterized foraging grounds – Cayman Islands, Texas, Bahamas, Cuba, Puerto Rico, US Virgin
Islands, Dominican Republic and Mexico (for sources of published genetic data see Fig. 1). Numbers are mean proportional
contribution from each rookery to each foraging ground, estimated via weighted many-to-many mixed stock analysis (MSA) and
proportion of particles from each rookery to enter each foraging ground during the first year of drifting, from the hatchling drift
model (Drift).
We then quantified the number of particles from each
source to enter these territorial waters during the modelled first year of drifting (though it should be noted that
a longer oceanic phase could occur; Musick & Limpus
1997). Euclidean distance matrices were calculated for
particle and genetic profiles (treating each estimate of
proportional contribution as a character) and compared
with a Mantel test (PopTools v. 3.0; Hood 2009). Additionally, to facilitate comparison with the results of
Bowen et al. (2007a), we investigated the relationship
between genetic and particle profiles with linear regression, regressing arcsine transformed proportions (GraphPad InStat 3.10, GraphPad Software).
Results
Genetic structure
Among 92 neritic juvenile hawksbills from the Cayman
Islands, we detected 11 mitochondrial DNA (mtDNA)
control region haplotypes: Ei-A1, Ei-A2, Ei-A3, Ei-A9c,
Ei-A11c, Ei-A18, Ei-A20, Ei-A24, EiA28, Ei-A29 and a
previously undetected haplotype, Ei-A72 (GenBank
accession number GQ925509). The Cayman Islands foraging aggregation represented a diverse mixed stock
(h = 0.72 ± 0.04; p = 0.01 ± 0.005). Haplotype frequencies from Grand Cayman and Little Cayman were not
significantly different (population pairwise FST = )0.01,
P = 0.83). Therefore results for the two study sites were
pooled for subsequent analyses.
Rookeries of origin
Using unweighted and weighted many-to-many analysis,
we estimated the contributions of genetically character 2009 Blackwell Publishing Ltd
ized Caribbean rookeries to the Cayman Islands foraging
aggregation. Shrink factors for all chains were <1.2,
indicating convergence. Results differed between the
weighted and unweighted models (Fig. 2a), but indicated contributions spanning the Caribbean region
(Fig. 2b). Confidence intervals were narrowest in the
weighted many-to-many analysis – suggesting that
this method may provide the most reliable estimate of
origins.
Hatchling drift model
Tracks of virtual particles were plotted over a 1-year period (Fig. 3). Distribution patterns varied markedly by
ecoregion (areas of oceanographic and biotic similarity;
Spalding et al. 2007): particles from Mexico were largely
entrained within the Gulf of Mexico (Fig. 3a) and a proportion of particles from the Greater Antilles and the
eastern Caribbean were entrained in the Bahamian ecoregion (Fig. 3b, c). Other particles from the Greater Antilles
and most particles from the southern, southwestern and
eastern Caribbean were transported eastward in the
Caribbean current (Fig. 3b–e) while distribution of particles from the western Caribbean was constrained by the
strong westward flowing Caribbean current (Fig. 3f).
Thus, overall a high level of mixing occurred in the
Caribbean basin, but dispersal varied by region (Fig. 3g).
Comparison of the hatchling drift model and mixed
stock analysis
A significant correlation (Mantel test, r = 0.77,
P < 0.001; linear regression r = 0.83, P < 0.001) was
detected between particle profiles (frequency of particles from each source entering foraging areas over
4846 J . M . B L U M E N T H A L E T A L .
Fig. 2 Genetic composition at the Cayman Islands foraging
aggregation. Among the 92 individuals sequenced, 11 haplotypes were detected: Ei-A1 (44); Ei-A2 (2); Ei-A3 (1); Ei-A9c (8);
Ei-A11c (17); Ei-A18 (1); Ei-A20 (1); Ei-A24 (11); Ei-A28 (4); EiA29 (1); Ei-A72 (2). (a) Using an unweighted many-to-many
model (white bars), estimated contributions (mean proportion,
2.5% and 97.5% confidence intervals) of genetically characterized hawksbills rookeries to the Cayman Islands foraging aggregation were Antigua (21.3%, 3.0–41.8%), Barbados (20.5%,
1.1–48.1%), Belize (8.7%, 0.4–24.9%), Cuba (12.6%, 0.6–34.1%),
Mexico (9.1%, 0.8–16.0%), USVI (11%, 0.4–27.6%), Costa Rica
(10.7%, 2.9–21.3%), Puerto Rico (6.1%, 0.8–16.6%). When
source population size was used as a constraint (grey bars), estimates were Antigua (6.0%, 0.7–15.5%), Barbados (46.1%, 20.6–
69.0%), Belize (0.7%, 0.0–2.7%), Cuba (21.9%, 3.7–43%), Mexico
(13.2%, 7.4–20.6%), USVI (3%, 0.1–9.4%), Costa Rica (0.5%,
0.0–2.0%), Puerto Rico (8.7%, 1.5–18.3%). (b) Arrows are scaled
in proportion to estimated mean contribution (weighted manyto-many) to indicate potential scale of recruitment.
365 days of drifting) and genetic profiles (many-tomany results for all genetically characterized Caribbean
foraging aggregations) (Fig. 4).
Discussion
Because hawksbill turtles spend the vast majority of
their lives at sea, for many populations links between
nesting beaches and foraging grounds are unknown.
To estimate origins of neritic juvenile hawksbill turtles
foraging in the Cayman Islands, we applied many-tomany modelling using data from this study and other
published genetic studies of Caribbean hawksbills. For
the Cayman Islands foraging aggregation potentially
contributing source rookeries spanned the Caribbean
basin – delineating a scale of recruitment of 200–
2500 km (straight-line distance) and highlighting the
complex spatial ecology of the species.
We then investigated the role of ocean currents as a
mechanism underlying patterns of dispersal. Bowen
et al. (2007a) previously tested the role of geographic
distance (both straight-line and modified in order to
account for a likely influence of ocean currents) in determining recruitment of hawksbills from nesting populations to foraging aggregations. Here, we compared
results of mixed stock analysis to particle profiles, which
take source population size, geographic distance and
oceanographic circulation patterns into account – an
important consideration as oceanic juveniles are unlikely
to follow a direct route between nesting beaches and foraging grounds. A strong correlation was detected
between patterns of stock composition at Caribbean
hawksbill foraging aggregations and patterns of particle
distribution at these sites – indicating that ocean currents likely play a substantial role in determining geographic patterns of genetic diversity, and that models of
oceanic drift may illustrate movements of oceanic juvenile turtles during the lost year or years. Putative dispersal patterns for oceanic juvenile hawksbills varied
regionally: particles from many rookeries were broadly
distributed across the Caribbean (a ‘turtle soup’), while
particles from other rookeries appeared to be regionally
constrained (forming ‘turtle groups’). Thus, because of
the complexities of regional and local current patterns,
some areas will be more diverse (turtles recruiting from
multiple jurisdictions) while others will experience
greater levels of proximate or local recruitment.
Ultimately, population resolution may allow demographic parameters such as abundance, survival, sex
ratio and growth to be linked across broad spatiotemporal scales and incorporated into population modelling
(Bowen et al. 2004; Reece et al. 2006). However, at present, the accuracy of mixed stock estimates is limited by
insufficient baseline data from nesting beaches (Troëng
et al. 2005). Indeed, detection of a previously unknown
haplotype at the Cayman Islands foraging aggregation
indicates that nesting beaches in the Caribbean or source
rookeries farther afield have been incompletely or insufficiently sampled. Longer periods of sampling and larger
sample sizes (Reece et al. 2006), standardized use of
primers which amplify larger mtDNA fragments (AbreuGrobois et al. 2006), and expanded geographic coverage
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4847
Fig. 3 Trajectories of modelled particles representing drifting oceanic stage juvenile turtles, released from hawksbill index beaches
in (a) Gulf of Mexico (Isla Holbox, Punta Xen and Las Coloradas Mexico) (b) Greater Antilles (Doce Leguas Cuba, Mona Island
Puerto Rico) (c) eastern Caribbean (Long Beach Antigua, Hilton Beach Barbados, Buck Island USVI) (d) southern Caribbean (Archipiélago Los Roques and Penı́nsula de Paria Venezuela) (e) southwestern Caribbean (Tortuguero Costa Rica) (f) western Caribbean
(Gales Point Belize) (g) depicts overall particle trajectory patterns in the wider Caribbean.
2009 Blackwell Publishing Ltd
4848 J . M . B L U M E N T H A L E T A L .
Fig. 4 Comparison of hatchling drift (black bars) with weighted (grey bars) and unweighted (white bars) many-to-many results for
the Cayman Islands and other genetically characterized foraging grounds (for sources of published genetic data see Fig. 1). Data are
mean proportion (±2.5–97.5% confidence intervals for many-to-many analysis).
(Engstrom et al. 2002) are priorities for improving estimates. Ocean current modelling may serve to identify
potential locations of regional foraging aggregations
and priorities for studies of unsurveyed source rookeries:
for example, particles from the southern Caribbean
(Venezuela) would appear to be widely dispersed, yet
rookeries are presently insufficiently characterized to
permit inclusion in mixed stock estimates.
Within the context of developing a hatchling drift
model, much remains to be resolved or refined regarding biological parameters (e.g. sizes of hawksbill source
populations, timing of peak hatching, duration of the
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4849
oceanic phase, behaviour of oceanic juveniles pushed
toward landmasses, and roles of active swimming and
orientation in hatchlings, post-hatchlings, and oceanic
juveniles) and modelling ocean currents (e.g. incorporating wind-induced components; Hays & Marsh 1997;
Girard et al. 2009). In this study, our hatchling drift
model was limited to determining exposure of foraging
sites to particles during the first year of drift. This is
useful as a measure of potential for recruitment from
each of the sources, as the timeframes within which
hawksbills begin actively swimming and recruit to foraging grounds is unknown. However, a longer oceanic
phase may occur (Musick & Limpus 1997), calling for
extended modelling of oceanic drift when such longterm high-resolution oceanographic data become available. Comparison of multiple years of Aviso data may
also be informative in determining inter-annual variations in recruitment and implications for management:
if, as in green turtles, there is temporal structuring in
the genetic composition of neritic juvenile foraging
aggregations, then marine turtle management plans
should not be based on the ‘snapshot’ of short-term
genetic studies (Bjorndal & Bolten 2008).
In coral reef fishes, behaviour of larvae has been
shown to influence population connectivity (Paris et al.
2007) but to date, it has proven difficult to separate the
role of oceanic currents from behaviour in determining
distribution of oceanic juvenile marine turtles (NaroMaciel et al. 2007). Under experimental conditions, loggerhead hatchlings have been shown to exhibit directed
swimming which is consistent with that which would
be needed to remain within oceanic gyres (Lohmann
et al. 2001) and in the wild, oceanic juvenile loggerheads may actively select foraging areas at locations
closely correlated with the latitudes of their natal beaches (Monzón-Argüello et al. 2009) and home toward
their natal regions at the conclusion of the oceanic stage
(Bowen et al. 2004). However, the ability to move
against prevailing currents appears to be linked to
increasing body size (Monzón-Argüello et al. 2009).
Oceanic juvenile loggerheads in downwelling lines
show minimal activity and positive buoyancy (Witherington 2002) and as hawksbill hatchlings at least in
some populations appear to be less active in the initial
days of dispersal than other marine turtle species
(Chung et al. 2009a) it is likely that small oceanic juvenile hawksbills are also relatively passive surface drifters. Therefore, while vertical movement of reef fish
larvae limits the application of models based on surface
current data (Fiksen et al. 2007) these models are especially applicable to marine turtles.
Occurrence of active orientation, particularly at the
end of the oceanic stage, may nevertheless explain
contributions of hawksbill rookeries against prevailing
2009 Blackwell Publishing Ltd
currents to foraging sites. Alternatively, unsurveyed
hawksbill rookeries may contribute to foraging stocks
(i.e. contributions assigned to a down-current source
may instead originate from an unsurveyed or incompletely surveyed up-current source). It is also possible
that a portion of Caribbean hawksbills could circle the
Atlantic before re-entering the Caribbean, following a
similar route to oceanic juvenile loggerheads – or perhaps a ‘short-cycle’ of the Atlantic could occasionally
occur where hawksbills might become entrained in
Atlantic eddies which more rapidly re-enter the Caribbean. In our comparisons between many-to-many
results and the hatchling drift model, estimated contributions from some rookeries were high using genetic
markers while the drift model predicted no contributions. These findings may be indicative of an Atlantic
cycle or short-cycle: for instance, Mexican contributions
to seemingly up-current foraging sites in the eastern
Caribbean may be explained by hawksbills cycling a
portion of the Atlantic before re-entering the Caribbean.
Integration of better biological data with heuristic
modelling of oceanic drift will assist in resolving migratory connectivity for marine turtles. Patterns of dispersal for oceanic juveniles may be confirmed by
documentation of stranding patterns (Meylan & Redlow
2006) and genetic studies (Carreras et al. 2006; Bowen
et al. 2007a). In surveys to date, magnitude of hawksbill
genetic diversity varies regionally – being lowest in the
Gulf of Mexico, where mixed stock analysis indicates
that oceanic juveniles originate almost exclusively from
Mexican rookeries (Bowen et al. 2007a), and higher near
the eastern Caribbean gyres and in foraging areas
passed by the Caribbean current. Sightings and strandings of oceanic phase hawksbills in Florida (Meylan &
Redlow 2006) and rarely on the east coast (Parker 1995)
support transport from the Caribbean through the
Florida Straits, and the presence of a neritic juvenile
hawksbill foraging aggregation in Bermuda (Meylan
et al. 2003) suggests some transport in the Gulf Stream.
However, further genetic sampling of stranded and
free-living oceanic hawksbills and genetic characterization of additional neritic foraging grounds is required.
Efforts to resolve movements are particularly timely,
as management of hawksbill turtles has been the subject
of considerable controversy in recent years. Hawksbills
are designated as ‘critically endangered’ by the IUCN
(2007) and listing on Appendix I of the Convention on
the International Trade in Endangered Species (CITES)
bars international trade among parties to the Convention. Nevertheless, commercially valuable hawksbill
products are still in high demand in some areas. Recent
debate has centred on conservation status and priorities,
scale and extent of transboundary movements (Bowen
& Bass 1997; Mrosovsky 1997) and sustainable use of
4850 J . M . B L U M E N T H A L E T A L .
mixed stocks (Bowen et al. 2007b; Godfrey et al. 2007;
Mortimer et al. 2007a,b). It has been argued that harvesting of hawksbills from foraging grounds should be
prohibited, as exploitation of mixed stocks could potentially impact rookeries on a regional scale – or alternatively, that sustainability of harvesting on hawksbill
foraging aggregations could be determined on a caseby-case basis (Godfrey et al. 2007).
Many-to-many analysis and hatchling drift models
facilitate an integrative, rookery-centric approach
(Bolker et al. 2007) in which contributions of both small,
vulnerable rookeries and large, regionally important
rookeries are evaluated. Oceanic juveniles from some
rookeries appear to be dispersed among multiple foraging grounds, while those from other rookeries appear to
be more locally constrained. This diversity of distribution represents both a challenge and an opportunity for
marine turtle management. Broadly dispersed stocks
are difficult to manage, yet threats are often geographically widely dispersed and their impacts on individual
stocks will depend on how they interact. In contrast, in
areas with greater levels of local or proximate recruitment, threats – or conservation efforts – may have a
concentrated effect (Bowen et al. 2004, 2005). Indeed,
Caribbean hawksbill rookeries have experienced varying population trends, ranging from near-extirpation
(Cayman Islands: Aiken et al. 2001) to dramatic increase
(Antigua: Richardson et al. 2006; Barbados: Beggs et al.
2007) – raising the possibility that variations in dispersal across an oceanographically complicated region
may be a factor in these differing trajectories.
While much remains to be determined regarding
movements and management needs of hawksbill turtles,
integration of many-to-many analysis and oceanographic data represents a promising approach. In this
study, we delineated links between regional management units by conducting many-to-many mixed stock
analysis across the range of Caribbean hawksbill rookeries and foraging aggregations, developed a tool with
which to illustrate a possible role of ocean currents in
determining distribution of oceanic juveniles, and highlighted gaps in knowledge and priorities for future
sampling. Overall, results facilitate a broader understanding of Caribbean hawksbill movements – filling an
important gap in knowledge of life-history and potentially informing regional management.
Acknowledgments
For invaluable logistical support and assistance with fieldwork,
we thank Cayman Islands Department of Environment
research, administration, operations and enforcement staff and
numerous volunteers. For advice on analysis, we thank J. Hunt
and W. Pitchers at University of Exeter, Cornwall campus.
Vincente Guzmán (CONANP-México) and Eduardo Cuevas
(Pronature-Penı́nsula de Yucatán A.C.) generously provided
information on population sizes of Campeche, Yucatán and
Quintana Roo rookeries. Work in the Cayman Islands, the UK
and the USA was generously supported by the European
Social Fund, the Darwin Initiative, the Department of Environment, Food & Rural Affairs (DEFRA), Eckerd College, the Foreign and Commonwealth Office for the Overseas Territories,
the Ford Foundation, the National Fish and Wildlife Foundation (NFWF) and the Turtles in the Caribbean Overseas Territories (TCOT) Project. We also acknowledge support to J.
Blumenthal (University of Exeter postgraduate studentship and
the Darwin Initiative). The manuscript was improved by the
input of three anonymous reviewers.
References
Abreu-Grobois FA, Horrocks JA, Formia A et al. (2006) New
mtDNA dloop primers which work for a variety of marine
turtle species may increase the resolution capacity of mixed
stock analyses. Poster presented at the 26th Annual
Symposium on Sea Turtle Biology and Conservation, Crete,
Greece, 2–8 April 2006. Available from http://www.
iucnmtsg.org/genetics/meth/primers/abreu_grobois_etal_
new_dloop_primers.pdf. Accessed 1 March 2008.
Aiken JJ, Godley BJ, Broderick AC, Austin T, Ebanks-Petrie G,
Hays GC (2001) Two hundred years after a commercial
marine turtle fishery: the current status of marine turtles
nesting in the Cayman Islands. Oryx, 35, 145–151.
Allard MW, Miyamoto MM, Bjorndal KA, Bolten AB, Bowen
BW (1994) Support for natal homing in green turtles from
mitochondrial DNA sequences. Copeia, 1994, 34–41.
Allen M, Engstrom AS, Meyers S et al. (1998) Mitochondrial
DNA sequencing of shed hairs and saliva on robbery caps:
sensitivity and matching probabilities. Journal of Forensic
Science, 43, 453–464.
Andersen LW, Ruzzante DE, Walton M et al. (2001)
Conservation genetics of harbour porpoises, Phocoena
phocoena, in eastern and central North Atlantic. Conservation
Genetics, 2, 309–324.
Baillie JEM, Hilton-Taylor C, Stuart SN (2004) IUCN red List of
Threatened Species. A Global Species Assessment. IUCN, Gland
Switzerland.
Baker CS, Patenaude NJ, Bannister JL, Robins J, Kato H (1999)
Distribution and diversity of mtDNA lineages among
southern right whales (Eubalaena australis) from Australia
and New Zealand. Marine Biology, 134, 1–7.
Bass AL (1999) Genetic analysis to elucidate the natural history
and behavior of hawksbill turtles (Eretmochelys imbricata) in
the Wider Caribbean: a review and reanalysis. Chelonian
Conservation and Biology, 3, 195–199.
Bass AL, Witzell WN (2000) Demographic composition of
immature green turtles (Chelonia mydas) from the east central
Florida coast: evidence from mtDNA markers. Herpetologica,
56, 357–367.
Bass AL, Good DA, Bjorndal KA et al. (1996) Testing models
of female reproductive migratory behavior and population
structure in the Caribbean hawksbill turtle, Eretmochelys
imbricata, with mtDNA sequences. Molecular Ecology, 5, 321–
328.
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4851
Bass AL, Epperly SP, Braun-McNeill J (2006) Green turtle
(Chelonia mydas) foraging and nesting aggregations in the
Caribbean and Atlantic: impact of currents and behavior on
dispersal. Journal of Heredity, 97, 346–354.
Beggs JA, Horrocks JA, Krueger BH (2007) Increase in hawksbill
sea turtle Eretmochelys imbricata nesting in Barbados, West
Indies. Endangered Species Research, 3, 159–168.
Bell CD, Solomon JL, Blumenthal JM et al. (2007) Monitoring
and conservation of critically-reduced marine turtle nesting
populations: lessons from the Cayman Islands. Animal
Conservation, 10, 39–47.
Bjorndal KA, Bolten AB (2008) Annual variation in source
contributions to a mixed stock: implications for quantifying
connectivity. Molecular Ecology, 17, 2185–2193.
Blumenthal JM, Austin TJ, Bell CD et al. (2009a) Ecology of
hawksbill turtles Eretmochelys imbricata on a western
Caribbean foraging area. Chelonian Conservation and Biology,
8, 1–10.
Blumenthal JM, Austin TJ, Bothwell JB et al. (2009b) Diving
behaviour and movements of juvenile hawksbill turtles
Eretmochelys imbricata on a Caribbean coral reef. Coral Reefs,
28, 55–65.
Bolker BM, Okuyama T, Bjorndal KA, Bolten AB (2007)
Incorporating multiple mixed stocks in mixed stock analysis:
‘many-to-many’ analyses. Molecular Ecology, 16, 685–695.
Bolten AB (2003) Variation in sea turtle life-history patterns:
neritic vs. oceanic developmental stages. In:The Biology of Sea
Turtles, Volume I (eds Lutz PL, Musick JA, Wyneken J). pp.
243–257, CRC Press, Boca Raton.
Bolten AB, Bjorndal KA, Martins HR et al. (1998) TransAtlantic developmental migrations of loggerhead sea turtles
demonstrated by mtDNA sequence analyses. Ecological
Applications, 8, 1–7.
Botsford LW, White JW, Coffroth M-A et al. (2009)
Connectivity and resilience of coral reef metapopulations in
marine protected areas: matching empirical efforts to
predictive needs. Coral Reefs, 28, 303–305.
Bowen BW, Bass AL (1997) Movement of hawksbill turtles:
what scale is relevant to conservation and what scale is
resolvable with mtDNA data? Chelonian Conservation and
Biology, 3, 440–442.
Bowen BW, Abreu-Grobois FA, Balazs GH et al. (1995) TransPacific migrations of the loggerhead sea turtle demonstrated
with mitochondrial DNA markers. Proceedings of the National
Academy of Sciences USA, 92, 3731–3734.
Bowen BW, Bass AL, Garcia A et al. (1996) The origin of
hawksbill turtles in a Caribbean feeding area as indicated by
genetic markers. Ecological Applications, 6, 566–572.
Bowen BW, Bass AL, Chow S et al. (2004) Natal homing in
juvenile loggerhead turtles (Caretta caretta). Molecular Ecology,
13, 3797–3808.
Bowen BW, Bass AL, Soares L, Toonen RJ (2005) Conservation
implications of complex population structure: lessons from
the loggerhead turtle (Caretta caretta). Molecular Ecology, 14,
2389–2402.
Bowen BW, Grand WS, Hillis-Starr Z et al. (2007a) Mixed-stock
analysis reveals the migrations of juvenile hawksbill turtles
(Eretmochelys imbricata) in the Caribbean sea. Molecular
Ecology, 16, 49–60.
Bowen BW, Grant WS, Hillis-Starr Z et al. (2007b) The
advocate and the scientist: debating the commercial
2009 Blackwell Publishing Ltd
exploitation of endangered hawksbill turtles. Molecular
Ecology, 16, 3514–3515.
Broderick D, Moritz C, Miller JD, Guinea ML, Prince JR,
Limpus CJ (1994) Genetic studies of the hawksbill turtle
(Eretmochelys imbricata): evidence for multiple stocks and
mixed feeding grounds in Australian waters. Pacific
Conservation Biology, 1, 123–131.
Browne DC, Horrocks JA, Abreu-Grobois FA (in press)
Population subdivision in hawksbill turtles nesting on
Barbados, West Indies, determined from mitochondrial DNA
control region sequences. Conservation Genetics, DOI 10.1007/
s10592-009-9883-3
Carr AF (1987) New perspectives on the pelagic stage of sea
turtle development. Conservation Biology, 1, 103–121.
Carreras C, Pont S, Maffucci F et al. (2006) Genetic structuring
of immature loggerhead sea turtles (Caretta caretta) in the
Mediterranean Sea reflects water circulation patterns. Marine
Biology, 149, 1269–1279.
Chacón D (2004) Caribbean Hawksbills – An Introduction to
Their Biology and Conservation Status. WWF-Regional
Program for Latin America and the Caribbean, San Jose,
Costa Rica.
Chung FC, Pilcher NJ, Salmon M, Wyneken J (2009a) Offshore
migratory activity of hawksbill turtle (Eretmochelys imbricata)
hatchlings, I. Quantitative analysis of activity, with
comparisons to green turtles (Chelonia mydas). Chelonian
Conservation and Biology, 8, 28–34.
Chung FC, Pilcher NJ, Salmon M, Wyneken J (2009b) Offshore
migratory activity of hawksbill turtle (Eretmochelys imbricata)
hatchlings II. Swimming gaits, swimming speed, and
morphological comparisons. Chelonian Conservation and
Biology, 8, 35–42.
van Dam RP, Diez CE, Balazs GH et al. (2008) Sex-specific
migration patterns of hawksbill turtles from Mona Island,
Puerto Rico. Endangered Species Research, 4, 85–94.
Diaz-Fernandez R, Okayama T, Uchiyama T et al. (1999)
Genetic sourcing for the hawksbill turtle, Eretmochelys
imbricata, in the northern Caribbean region. Chelonian
Conservation and Biology, 3, 296–300.
Dutton PH (1996) Methods for collection and preservation of
samples for sea turtle genetic studies. In: Proceedings of the
International Symposium on Sea Turtle Conservation Genetics
(eds Bowen BW, Witzell WN), pp. 17–24. National Technical
Information Service, Springfield, VA, USA, NOAA Technical
Memorandum NMFS-SEFSC-396.
Encalada SE, Bjorndal KA, Bolten AB et al. (1998) Population
structure of loggerhead turtle (Caretta caretta) nesting
colonies in the Atlantic and Mediterranean as inferred from
mitochondrial DNA control region sequences. Marine
Biology, 130, 567–575.
Engstrom TN, Meylan PA, Meylan AB (2002) Origin of
juvenile loggerhead turtles (Caretta caretta) in a tropical
developmental habitat in Caribbean Panama. Animal
Conservation, 5, 125–133.
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an
integrated software package for population genetics data
analysis. Evolutionary Bioinformatics Online, 1, 47–50.
Fiksen O, Jorgensen C, Kristiansen T, Vikebo F, Huse G (2007)
Linking behavioural ecology and oceanography: larval
behaviour determines growth, mortality and dispersal.
Marine Ecology Progress Series, 347, 195–205.
4852 J . M . B L U M E N T H A L E T A L .
Formia A, Godley BJ, Dontaine J-F, Bruford MW (2006)
Mitochondrial DNA diversity and phylogeography of
endangered green turtle (Chelonia mydas) populations in
Africa. Conservation Genetics, 7, 353–369.
Formia A, Broderick AC, Glen F et al. (2007) Genetic
composition of the Ascension Island green turtle rookery
based on mitochondrial DNA: implications for sampling and
diversity. Endangered Species Research, 3, 145–158.
Garduño-Andrade M (1999) Nesting of the hawksbill turtle,
Eretmochelys imbricata, in Rio Lagartos, Yucatan, Mexico,
1990–1997. Chelonian Conservation and Biology, 3, 281–285.
Girard C, Tucker AD, Calmettes B (2009) Post-nesting
migrations of loggerhead sea turtles in the Gulf of Mexico:
dispersal in highly dynamic conditions. Marine Biology, 156,
1827–1839.
Godfrey MH, Webb GJW, Manolis SC, Mrosovsky N (2007)
Hawksbill sea turtles: can phylogenetics inform harvesting?
Molecular Ecology, 16, 3511–3513.
Hasbún CR (2002) Observations on first day dispersal of
neonatal hawksbill turtles (Eretmochelys imbricata). Marine
Turtle Newsletter, 96, 7–10.
Hays GC, Marsh R (1997) Estimating the age of juvenile
loggerhead sea turtles in the North Atlantic. Canadian Journal
of Zoology, 75, 40–46.
Hood GM (2009) PopTools version 3.1.1. Available on the
internet. URL http://www.cse.csiro.au/poptools. Accessed
20 July 2008.
Horrocks JA, Vermeer LA, Krueger B et al. (2001) Migration
routes and destination characteristics of post-nesting
hawksbill turtles satellite-tracked from Barbados, West
Indies. Chelonian Conservation and Biology, 4, 107–114.
IUCN (2007) Red List of Threatened Species. Available from
http://www.iucnredlist.org. Accessed 1 April 2008.
Kocher TD, Thomas WK, Meyers A et al. (1989) Dynamics of
mitochondrial DNA evolution in animals: amplification and
sequencing with conserved primers. Proceedings of the
National Academy of Sciences of the USA, 86, 6196–6200.
Lahanas PN, Bjorndal KA, Bolten AB et al. (1998) Genetic
composition of a green turtle (Chelonia mydas) feeding
ground population: evidence for multiple origins. Marine
Biology, 130, 345–352.
Laurent L, Casale P, Bradai N et al. (1998) Molecular
resolution of marine turtle stock composition in fishery
bycatch: a case study in the Mediterannean. Molecular
Ecology, 7, 1529–1542.
Lewis CB (1940) The Cayman Islands and marine turtle.
In:Herpetology of the Cayman Island (ed. Grant C). pp. 56–65,
Bulletin of the Institute of Jamaica Science Series, Kingston.
Lohmann KJ, Cain SD, Dodge SA, Lohmann CM (2001)
Regional magnetic fields as navigational markers for sea
turtles. Science, 294, 364–366.
Luke K, Horrocks JA, LeRoux RA, Dutton PH (2004) Origins of
green turtle (Chelonia mydas) feeding aggregations around
Barbados, West Indies. Marine Biology, 144, 799–805.
Maffucci F, Kooistra WHCF, Bentivegna F (2006) Natal origin
of loggerhead turtles, Caretta caretta, in the neritic habitat off
the Italian coasts, Central Mediterranean. Biological
Conservation, 127, 183–189.
Meylan AB (1999) International movements of immature and
adult hawksbill turtles (Eretmochelys imbricata) in the Caribbean
region. Chelonian Conservation and Biology, 3, 189–194.
Meylan A, Redlow A (2006) Eretmochelys imbricata – Hawksbill
turtle. In: Biology and Conservation of Florida Turtles (ed.
Meylan PA). Chelonian Research Foundation, Lunenburg, MA,
USA, 3, 105–127.
Meylan AB, Bowen BW, Avise JC (1990) A genetic test of the
natal homing versus social facilitation models for green
turtle migration. Science, 248, 724–727.
Meylan PA, Meylan AB, Gray J, Ward J (2003) The hawksbill
turtle in Bermuda. In: Proceedings of the Twenty-Second Annual
Symposium on Sea Turtle Biology and Conservation (compiler,
Seminoff JA), p. 26. National Technical Information Service,
Springfield, VA, USA. NOAA Technical Memorandum
NMFS-SEFSC-503.
Millar RB (1987) Maximum likelihood estimation of mixed
stock fishery composition. Canadian Journal of Fisheries and
Aquatic Science, 44, 583–590.
Milligan BG (1998) Total DNA isolation. In:Molecular Genetic
Analysis of Populations, a Practical Approach (ed Hoelzel AR).
pp. 29–64, IRL Press, Oxford.
Moncada F, Carillo E, Saenz A, Nodarse G (1999)
Reproduction and nesting of the hawksbill turtle,
Eretmochelys imbricata, in the Cuban archipelago. Chelonian
Conservation and Biology, 3, 257–263.
Monzón-Argüello C, Rico C, Carreras C, Calbuig P, Marco A,
López-Jurado LF (2009) Variation in spatial distribution of
juvenile loggerhead turtles in the eastern Atlantic and
western Mediterranean sea. Journal of Experimental Marine
Biology and Ecology, 373, 79–86.
Mora C, Sale PF (2002) Are populations of coral reef fish open
or closed? Trends in Ecology and Evolution, 17, 422–428.
Mortimer JA, Donnelly M (2007) Marine Turtle Specialist
Group 2007 IUCN Red List Status Assessment Hawksbill
Turtle (Eretmochelys imbricata). Available from http://
www.iucn-mtsg.org/red_list/ei/index.shtml. Accessed 1
March 2008.
Mortimer JA, Donnelly M, Meylan AB, Meylan PA (2007a)
Critically endangered hawksbill turtles: molecular genetics
and the broad view of recovery. Molecular Ecology, 16, 3516–
3517.
Mortimer JA, Meylan PA, Donnelly M (2007b) Whose turtles
are they, anyway? Molecular Ecology, 16, 17–18.
Mrosovsky N (1997) Movement of hawksbill turtles – a
different perspective on the DNA data. Chelonian
Conservation and Biology, 3, 438–439.
Musick JA, Limpus CJ (1997) Habitat utilization and migration
in juvenile sea turtles. In:The Biology of Sea Turtle (eds Lutz
PL, Musick JA). pp. 137–163, CRC Press, Boca Raton.
Naro-Maciel E, Becker HJ, Lima EHSM, Marcovaldi MA,
DeSalle R (2007) Testing dispersal hypotheses in foraging
green sea turtles (Chelonia mydas) of Brazil. Journal of
Heredity, 98, 29–39.
Norman JA, Moritz C, Limpus CJ (1994) Mitochondrial DNA
control region polymorphisms: genetic markers for
ecological studies of marine turtles. Molecular Ecology, 3,
363–373.
Okuyama T, Bolker BM (2005) Combining genetic and
ecological data to estimate sea turtle origins. Ecological
Applications, 15, 315–325.
Owens DW, Ruiz GW (1980) New methods of obtaining blood
and cerebrospinal fluid from marine turtles. Herpetologica, 36,
17–20.
2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4853
Paris CB, Cherubin LM, Cowen RK (2007) Surfing, spinning, or
diving from reef to reef: effects on population connectivity.
Marine Ecology Progress Series, 347, 285–300.
Parker LG (1995) Encounter with a juvenile hawksbill turtle
offshore Sapelo Island, Georgia. Marine Turtle Newsletter, 71,
19–22.
Pella J, Masuda M (2001) Bayesian methods for analysis of
stock mixtures from genetic characters. Fishery Bulletin, 99,
151–167.
Pella J, Milner GB (1987) Use of genetic marks in stock
composition analyses. In:Population Genetics and Fishery
Managemen (eds Ryman N, Utter F). pp. 247–276, University
of Washington Press, Seattle.
Reece JS, Ehrhart LM, Parkinson CL (2006) Mixed stock
analysis of juvenile loggerheads (Caretta caretta) in Indian
River lagoon, Florida: implications for conservation
planning. Conservation Genetics, 7, 345–352.
Richardson JI, Hall DB, Mason PA et al. (2006) Eighteen years
of saturation tagging data reveal a significant increase in
nesting hawksbill sea turtles (Eretmochelys imbricata) on Long
Island, Antigua. Animal Conservation, 9, 302–307.
Roberts H (1994) Reefs and lagoons of Grand Cayman. In:The
Cayman Islands: Natural History and Biogeograph (eds Brunt
MA, Davies JE). pp. 75–104, Kluwer Academic Publishers,
Netherlands.
Spalding MD, Fox HE, Allen GR et al. (2007) Marine
ecoregions of the world: a bioregionalization of coast and
shelf areas. BioScience, 57, 573–583.
Troëng S, Dutton PH, Evans D (2005) Migration of hawksbill
turtles Eretmochelys imbricata from Tortuguero, Costa Rica.
Ecography, 28, 394–402.
Velez-Zuazo X, Ramos WD, van Dam RP et al. (2008) Dispersal,
recruitment and migratory behaviour in a hawksbill sea turtle
aggregation. Molecular Ecology, 17, 839–853.
Wessel P, Smith HF (1991) Free software helps map and
display data. Eos, Transactions of the American Geophysical
Union, 72, 441.
2009 Blackwell Publishing Ltd
Whiting SD, Koch AU (2006) Oceanic movement of a benthic
foraging juvenile hawksbill turtle from the Cocos (Keeling)
Islands. Marine Turtle Newsletter, 112, 15–16.
Witherington BE (2002) Ecology of neonate loggerhead turtles
inhabiting lines of downwelling near a Gulf Stream front.
Marine Biology, 140, 843–853.
Witteveen BH, Straley JM, von Ziegesar O, Steel D, Baker CS
(2004) Abundance and mtDNA differentiation of humpback
whales (Megaptera novaeangliae) in the Shumagin Islands,
Alaska. Canadian Journal of Zoology, 82, 1352–1359.
J.M.B. is a research officer at the Cayman Islands Department
of Environment, where her work centres on using spatial ecology to aid in understanding management requirements of sea
turtles and other marine species. G.E.-P. is the director of the
Department of Environment and T.J.A. is the deputy director
for the research section. M.S.C. coordinates a gobal sea turtle
network as director of seaturtle.org. F.A.A.-G.’s research on
conservation genetics spans marine turtle populations in Mexican and Wider Caribbean sites. P.A.M. is RR Hallin Professor
of Natural Science at Eckerd College; A.B.M. leads the marine
turtle research program for the Florida Fish and Wildlife Conservation Commission; together they study the reproductive
biology, ecology and migrations of sea turtles in the Western
Atlantic with long-term projects in Panama and Bermuda. A.F.
works on marine turtle conservation genetics in West Africa
and coordinates the Marine Turtle Partnership in Gabon.
M.W.B. is head of the Biodiversity and Ecological Processes
Group at Cardiff University. A.C.B. and B.J.G. coordinate the
Marine Turtle Research Group at a range of sites around the
world including Ascension Island, Northern Cyprus and the
UK Overseas Territories.