Vol. 32: 251–264, 2017 ENDANGERED SPECIES RESEARCH
Published March 14
https://doi.org/10.3354/esr00809 Endang Species Res
OPEN
 ACCESS
Connectivity, population structure, and
 conservation of Ecuadorian green sea turtles
Jaime A. Chaves1,2,*, Micaela Peña3, Jhonnattan A. Valdés-Uribe1, 
Juan Pablo Muñoz-Pérez1,2,3, Felipe Vallejo3, Maike Heidemeyer4,5, 
Omar Torres-Carvajal6
1Universidad San Francisco de Quito, Colegio de Ciencias Biológicas y Ambientales, Diego de Robles y Av. Interoceánica, 
Cumbayá, Quito, Ecuador
2Galápagos Science Center, Puerto Baquerizo Moreno, Galápagos, Ecuador
3Fundación Equilibrio Azul, PO Box 17116025, Quito, Ecuador
4Centro de Investigación en Biología Celular y Molecular CIBCM, Universidad de Costa Rica, Costa Rica
5Centro de Restauración de Especies Marinas Amenazadas CREMA, Tibas, Costa Rica
6Museo de Zoología, Escuela de Biología, Pontificia Universidad Católica del Ecuador, Avenida 12 de Octubre y Roca, 
Apartado 17-01-2184, Quito, Ecuador
ABSTRACT: Studies of highly migratory species that increase our understanding of the dynamics
of genetic diversity, migratory routes, and genetic connectivity are essential for informing conser-
vation actions. Genetic data for green turtles Chelonia mydas from Ecuador have only been avail-
able from Galápagos Islands (GPS) rookeries, but not from foraging aggregations. Furthermore,
green turtles from habitats associated with mainland Ecuador (Machalilla National Park; MNP)
have not been sampled. To assess the genetic relationships between nesting and foraging aggre-
gations from these 2 regions and other regional populations, the mitochondrial DNA (mtDNA)
control region was sequenced from 133 turtles. Conventional FST (haplotype frequency) and ΦST
(sequence-based) values were low and non-significant between Ecuadorian rookeries, suggesting
high connectivity between these sites located ca. 1000 km apart. Mixed stock analysis (MSA) indi-
cated a dominant (>94%) GPS-MNP contribution to both foraging grounds, with small and nearly
negligible contributions from other rookeries in the region (e.g. Costa Rica and Mexico). While
orphan haplotypes were not included in the MSA because their rookery of origin is not known,
their close genetic relationships to Western and Central Pacific mtDNA clades suggests that a rel-
atively large percentage of turtles at the combined foraging sites (>10%) have been involved in
transoceanic migration events. The genetic links between GPS and MNP C. mydas nesting popu-
lations revealed by our study highlight the need to incorporate the nesting populations from
coastal Ecuador in more comprehensive future conservation planning.
KEY WORDS:  Chelonia mydas · Galápagos · Machalilla · Connectivity · Ecuador · Conservation ·
Mixed stock analysis · Phylogenetics
INTRODUCTION for the species’ conservation. The use of genetic in- 
formation provides a comprehensive account of pop-
Phylogeographic studies of highly vagile species ulation structure, connectivity, and overall demo-
are important to an understanding of broad patterns graphic patterns of populations (Bowen & Karl 2007,
of dispersal, as well as to identify important regions Seminoff et al. 2008, Shamblin et al. 2012, Joseph et
© The authors 2017. Open Access under Creative Commons by
*Corresponding author: jaimechaves76@gmail.com Attribution Licence. Use, distribution and reproduction are un -
restricted. Authors and original publication must be credited. 
Publisher: Inter-Research · www.int-res.com
252 Endang Species Res 32: 251–264, 2017
al. 2016). Obtaining genetic information, however, is has the advantage over the traditional mark-recap-
parti cularly challenging in marine environments ture techniques in that it can be applied to individu-
where physical barriers for gene flow are less evident als belonging to both genders and all size classes
and dispersal potential is high (Ward et al. 1994). (Jensen 2010, Jensen et al. 2013, 2016), can involve
Geographically dispersed populations can impose modest sample sizes, and requires only a relatively
logistical and financial limitations on sufficient sam- short period of time to complete (Read et al. 2014).
pling of representative breeding and feeding popu - Despite numerous studies using MSA in marine tur-
lations. For instance, the circumtropical green turtle tles (Bass & Witzell 2000, Engstrom et al. 2002, Dut-
Chelonia mydas has the widest and most continuous ton et al. 2008, Jensen 2010, Nishizawa et al. 2013,
distribution of all hard-shelled sea turtle species Jensen et al. 2016, Joseph et al. 2016), important
(Vieyra 2006). Previous work on C. mydas shows that sampling gaps from Eastern Pacific (EP) populations
reg ional genetic structure has been shaped by both remain, including the breeding populations nesting
intrinsic (behavior and ecology) and extrinsic (geo- on mainland Ecuador. Furthermore, limited informa-
graphic, climate, and oceanography) events (Semi- tion exists on the levels of connectivity between for-
noff et al. 2008, Roden et al. 2013, Dutton et al. 2014a, aging and nesting habitats, with only a few MSAs for
Hamabata et al. 2014). For example, current levels of the region (Amorocho et al. 2012, Heidemeyer 2015).
genetic diversity are potentially the result of glacial We examined a previously unstudied rookery of
and inter-glacial paleo-oceanic periods that created green turtles from the Pacific coast of mainland
opportunities for the establishment of new migratory Ecuador at Machalilla National Park (MNP) in order
routes (Formia et al. 2006, Hamabata et al. 2014). to fill a void of genetic information from that region.
Genetic differentiation between Pacific and Atlantic The MNP contains the largest reported breeding
basin populations is most likely the consequence of population of green turtles along the coast of Ecua -
the formation of the Isthmus of Panama that imposed dor, with the greatest number recorded at Isla de la
a barrier to gene flow between these oceans (Bowen Plata (ca. 470 total nests reported from nesting sea-
et al. 1992, Bowen & Karl 2007, Naro-Maciel et al. sons 2008 to 2010) and fewer nests per area
2008, Wallace et al. 2010, Duchene et al. 2012). recorded from coastal sites at Drake Bay (n = 73
Defining genetic clusters of green turtles across their nests) and La Playita (n = 7 nests) (Peña-Mosquera
broad geographic range has allowed delineation of 2010, Anhalzer et al. 2012). Unfortunately, green
Management Units (MUs) for conservation in several turtles at MNP face important conservation chal-
regions (Moritz 1994, Dethmers et al. 2006, Dutton et lenges (e.g. presence of feral dogs, illegal nest har-
al. 2014a,b) within broader geographic regional MUs vesting, urbanization, light pollution, beach motor
(see Wallace et al. 2010). Nonetheless, many green traffic, garbage) as do olive ridleys Lepidochelys oli-
turtle rookeries along the coast of the southeast vacea and hawksbill turtles Eretmochelys imbricata
Pacific Ocean remain understudied, which limits that also nest in this area (Peña-Mosquera 2010,
both our understanding of evolutionary pathways Mast et al. 2016). Previous genetic work in Ecuador
and the identification of potentially new MUs. focused solely on females nesting on the Galápagos
The life history of the green turtle involves an early Islands (Dutton et al. 2014a, Naro-Maciel et al.
oceanic epipelagic phase lasting several years (Carr 2014) but did not assay the foraging aggregations in
1987) followed by a less vagile stage that is associ- the waters off these islands. Galápagos rookeries
ated with neritic foraging grounds (Musick et al. (GPS) apparently constitute a single individual gen -
1997, Luschi et al. 2003) that commonly host individ- etic stock (the Galápagos MU; Dutton et al. 2014a)
uals from a mixture of rookeries (Moncada et al. that has very limited connectivity to other rookeries
2006, Bowen & Karl 2007). Green turtles show philo - across the EP. Some rookeries that are separated by
patry (Mayr 1963) and eventually return to their natal a distance of approximately 500 km or more are
beaches, which can be hundreds or thousands of commonly considered independent genetic stocks
kilometers from their foraging areas, for reproduc- (Dethmers et al. 2006, Browne et al. 2010, Dutton et
tion (Lohmann et al. 1997, Bolten 2003). Reproduc- al. 2014a, Shamblin et al. 2015). Thus the distance
tive stocks can be identified through the use of mito- between MNP and GPS rookeries (ca. 1000 km)
chondrial DNA (mtDNA) sequences (Bowen et al. would suggest that these populations should be
1992). Moreover, given the genetic distinctiveness of genetically independent (i.e. distinguishable genetic
nesting stocks, their relative contributions to a given stock and possibly new MU), although no genetic
foraging ground can be estimated with mixed stock data are currently available that confirm a lack of
analysis (MSA) (Bolker et al. 2003). The MSA method connectivity be tween the rookeries.
Chaves et al.: Green sea turtles in Ecuador 253
In this study, we used mtDNA sequences to (1) de - turtles along the coast of Ecuador (Peña- Mosquera et
termine the genetic relationships and connectivity al. 2009, Peña-Mosquera 2010) and the Pacific coast
between green turtle rookeries in MNP and GPS, (2) of South America (Mast et al. 2016). Nesting aggre-
evaluate genetic composition at foraging grounds in gations were sampled at beaches of La Playita on the
MNP and GPS, (3) estimate the origin of orphan mainland and of Isla de la Plata, and foraging aggre-
haplo types (i.e. unknown rookery of origin) from gations were sampled around Isla de la Plata and off
phylo genetic trees, and (4) assess the extent of the the coast of Puerto López on the mainland. Samples
Galápagos MU in the Southeast Pacific. from GPS were collected from foraging aggregations
around San Cristóbal, Espa ñola, Floreana, Darwin,
and Wolf Islands (Fig. 1).
MATERIALS AND METHODS We obtained tissue samples (skin biopsies) for
DNA extraction from a total of 133 green turtles: 43
Study area and data collection from foraging aggregations at Puerto López and Isla
de la Plata, herein considered as a single foraging ag -
Samples were collected from both nesting and for- gregation given their geographic proximity (Fig. 1),
aging aggregations along continental-coastal Ecua- 29 from nesting aggregations at MNP beaches (7
dor at MNP, Manabí province, western Ecuador. The from La Playita and 22 from Isla de la Plata), and 61
MNP (01° 31’ 00”S, 80°42’ 00”W) has a land area of from different foraging aggregations in the Galápa-
ca. 551 km2 and extends 2 nautical miles (ca. 3.7 km) gos Islands. Turtles were captured by dir ect swim-
along the coast and around Isla de la Plata (Peña- and-catch or while monitoring reproductive females
Mosquera et al. 2009). Isla de la Plata (1°16’ 55”S, during nesting activities at beaches. Morphometric
81° 3’ 84”W) is a 14 km2 continental island located measurements were taken with a flexible measuring
40 km from mainland Ecuador and known for pre- tape: curved carapace length (CCL), curved cara-
senting the highest concentrations of nesting green pace width (CCW), and total tail length (TTL) as per
standard methods. We used a power
hang scale (Do-All Outdoors) to ob- 
tain mass (± 0.5 kg) of turtles at the
foraging grounds. Individuals were
uniquely marked using Iconel tags
placed on flippers, thereby avoiding
pseudoreplication that would result
from duplicate sampling of indivi -
duals. Turtles sampled at foraging
grounds were returned to the site of
capture. Because the sex of an imma-
ture green turtle cannot be deter-
mined with confidence on the basis of
an external examination (Bolten &
Bjorndal 1992), we used a cut-off size
of 70 cm CCL to discriminate imma-
ture from mature individuals (Amoro-
cho et al. 2012). This is consistent
with the mean nesting size of repro-
ductive females at GPS, where 2
modal values of 80 and 95 cm CCL
have been recorded, with the small-
est nesting female documented for
Fig. 1. Location of study sites in Ecuador. Boxes correspond to Galápagos the archipelago at 60.7 cm CCL
(GPS) and Machalilla National Park (MNP) sampling locations, separated by (Zárate 2002). The sex of green turtles
ca. 1000 km. Gray area in MNP corresponds to the protected zone of the park with CCL >70 cm was determined
(including Isla de la Plata, located ca. 40 km from the mainland). Names with
nesting and foraging numbers represent locations and number of individuals based on sexual dimorphism presen -
sampled. Names with asterisks correspond to sites sampled by Dutton et al. ted in Chelonia mydas (Wibbels 1999,
(2014a) used in this study Wyneken et al. 2013). Thereafter,
254 Endang Species Res 32: 251–264, 2017
mature adult males were categorized as individuals on several estimates of genetic differentiation: haplo-
presenting tail measurements of 25 cm (or longer) type frequencies, haplotype diversity (h), and nu -
(Hamann et al. 2006). Skin biopsies were preserved cleotide diversity (π). Population structure was tested
in 90% ethanol in the field, and stored at −80°C. All by conducting analysis of molecular variance (AMO- 
of the sequences for GPS rookeries inc orporated into VA), pairwise FST comparisons, and pairwise exact
this study (n = 126: 45 from Las Bac has on Santa Cruz test of population differentiation. Both sequence-
Island and 81 from Las Salinas on Baltra Island) were based ΦST and conventional FST distance measures
downloaded from GenBank and correspond to the (10 000 permutations) (Reynolds et al. 1983) were
work by Dutton et al. (2014a). used to calculate within- and among-population di -
versity in Arlequin v. 2.0 (Schneider et al. 2000). The
same steps were taken to compare foraging aggrega-
Laboratory analysis tions in MNP and GPS. Because Western Pacific (WP)
haplotypes were found frequently in the foraging
Mitochondrial DNA (mtDNA) was isolated from tis- aggregations, we replicated these anal yses removing
sue using a Qiagen DNeasy® Blood & Tissue Kit fol- WP haplotypes.
lowing the manufacturer’s protocol, and DNA quality
and concentration were measured with Nanodrop
1000 (ThermoScientific). Primers LTEi9 (5’-GAA Phylogenetic analysis
TAA TCA AAA GAG AAG G-3’) and H950 (5’-GTC
TCG GAT TTA GGG GTT T-3’) (Abreu-Grobois et al. The best fitting model of evolution for the locus in
2006) were used in a polymerase chain reaction the studied populations (TrN+I) was determined with
(PCR) to amplify a ~832 bp fragment of the control re- jModelTest v.0.1.1 (Posada 2008) via Akaike’s infor-
gion. PCR was performed in 25 to 50 µl total vol umes mation criterion (AIC, Burnham & Anderson 2002). A
with cocktail concentrations of 1× buffer, 1.25 mM of phylogenetic reconstruction primarily to investigate
each deoxynucleoside triphosphate (dNTPs), 1.5 mM the origin of several orphan haplotypes found in
MgCl2, 10 µM of each primer, 1.25 U Taq poly - Ecuadorian foraging aggregations was conducted
merase, and ~40 ng of template DNA. Amplification using Bayesian inference in MrBayes v.3.2 (Ronquist
was run under an initial denaturation at 94−96°C for & Huelsenbeck 2003), with 2 runs of 4 simultaneous
3 min, followed by 35 to 45 cycles of denaturation at Markov chains, each for 10 million generations, and
94−96°C for 30 s, annealing of primers at 55°C for sampling of parameters and trees every 1000 gener-
30 s, and extension of primers and elongation at 72°C ations. Convergence of the chains was confirmed
for 10 min. To detect PCR products, 3 µl aliquots of using TRACER v.1.4 (Rambaut & Drummond 2007)
product were run on 1% agarose gels stained with and AWTY (Nylander et al. 2008), followed by a 25%
SYBR® Safe (Invitrogen) and visualized with Biorad burn-in of all retrieved trees to calculate posterior
Gel Doc XR (BIORAD) for photo-documentation. PCR probabilities in a 50% majority-rule consensus tree.
products were purified with 20% ExoSap reactions We used Natator depressus (GenBank accession no.
following the manufacturer’s protocol. All samples U40662) as an outgroup (Duchene et al. 2012), as
were sequenc ed by Macrogen (Seoul). well as additional sequences from GenBank from EP
rookeries that nested at Michoacán, Revillagigedos
Islands, and Costa Rica, from Central Pacific (CP)
Statistical analysis rookeries in Hawaii (Dutton et al. 2014a, Heidemeyer
2015), and from WP populations nesting on American
Sequences were aligned using Geneious Basic Samoa, the Republic of the Marshall Islands, and
v.8.0.5 (www.geneious.com, Kearse et al. 2012) un - Ulithi Atoll-Yap (Dutton et al. 2014b). We also recon-
der default settings for Geneious Alignment against structed a time-calibrated tree under a relaxed-clock
reference GenBank haplotypes (Table S1 in the Sup- framework in BEAST v.1.7.5 (Drummond & Rambaut
plement at www. int-res. com/a rticles/ suppl/n 032p 251 2007), after rejection of a global molecular clock
_ supp. pdf). Novel haplotypes were assigned the stan- (likelihood ratio test, Felsenstein 1981). The tree was
dard nomenclature CmP for C. mydas, with a suffix calibrated using 0.01751 substitutions site−1 lineage−1
added to denote the sampling locality of the haplo- million yr−1 as the rate of evolution of the mtDNA
type (see Dutton et al. 2008, 2014a). Genetic compar- control region of green turtles (Formia et al. 2006).
isons between the Ecuadorian rookeries and foraging Our analyses were performed for 10 million genera-
aggregations (GPS and MNP) were assessed based tions using a random starting tree (tree prior specia-
Chaves et al.: Green sea turtles in Ecuador 255
tion: Yule process). Stationarity, high effective sam- were run for 8 chains from MNP foraging grounds and
ple sizes (ESS > 200), and 95% highest posterior den- 5 chains from GPS foraging aggregations. Each of the
sity intervals (HPDs) were evaluated for all para- chains was given a different starting point, and the
meters in TRACER (Rambaut & Drummond 2007). A number of chains corresponded to the potential
consensus tree with divergence times was obtained source origins used for MNP and GPS (see above).
for unique green turtle sequences after discarding Posterior distribution was calculated using a burn-in
the first 25% of sampled trees. Finally, to represent of 25 000 runs, and the shrink factor of Gelman & Ru-
intraspecific phylogenies and geographic distribu- bin (1992) was calculated to test if chains had con-
tion of genetic diversity of Ecuadorian haplotypes in verged. We excluded individuals with haplotypes not
relation to the EP region, a minimum-spanning net- found in any nesting rookery (orphan haplotypes) be-
work of absolute distances between mtDNA haplo- cause they were uninformative.
types was constructed using TCS v.1.2.2 (Clement et
al. 2000) and Arlequin v. 2.0 (Schneider et al. 2000)
using all published nesting haplotypes from Dutton RESULTS
et al. (2014a) (see Fig. 3).
Sequence diversity in Ecuador
MSA of Ecuadorian foraging populations Genetic analysis of the 133 Ecuadorian samples re- 
solved a total of 14 haplotypes. From the 29 individu-
The contributions of distinct nesting aggregations to als sampled in the MNP rookery, only 4 haplotypes
the MNP and GPS green turtle foraging grounds were were recovered (CmP4.1, CmP4.4, CmP4.6, and
estimated with Bayesian MSA using the program CmP4.7), the same as those reported in GPS and
Bayes (Pella & Masuda 2001). Source haplotype fre- other nesting individuals in the EP (Table 1) (Mi -
quencies were taken from published nesting stock choacán and Costa Rica; Dutton et al. 2014a). When
characterizations from the EP sites Michoacán, Revil- combining our results with those from the previously
lagigedos (Mexico), Nombre de Jesús (Costa Rica), published literature, these 4 haplotypes were found
and Galápagos, as well as the CP Hawaiian archipel- to be the most common in the Ecuadorian rookeries,
ago (Table S1). Given that one of the haplotypes sam- occurring in 143 of 259 individuals for GPS beaches,
pled in the MNP has an identified origin in WP rook- and 79 of 259 individuals for both foraging grounds
eries (CmP22.1; Dutton et al. 2014b), haplotypes from (Table 1). Only 3 haplotypes previously reported in
American Samoa, Marshall Islands, and Ulithi Atoll GPS rookeries were not found in this study (see Dut-
were also included in the MSA. The MSA was con- ton et al. 2008). One of the haplotypes found in nest-
ducted using both uniform (flat: equal values for each ing individuals in continental Ecuador (Cmp4.1) has
stock) and informed (weighted: different values) also been reported in other foraging aggregations
priors (Table S2 in the Supplement). Using informed across the EP (Japan: Hamabata et al. 2014) and CP
priors allows incorporating knowledge of the species (Palmyra Atoll; Naro-Maciel et al. 2014). The other 10
to provide a stronger and more biologically meaning- haplotypes recovered in this study were found only
ful result in cases where genetic structure is weak at foraging aggregations: 3 only found in MNP, 4 only
(Jensen et al. 2013). We used ‘rookery population size’ found in GPS, and 3 were shared between these 2
and ‘geographic distance’ from sources as weighted foraging populations (Table 1). Three out of the 10
priors, as these are widely used in MSA and influence haplotypes have also been previously reported for
the estimated composition of the foraging grounds GPS rookeries (Dutton et al. 2014a). Interestingly, 1
(Proietti et al. 2012). ‘Geographic distance’ prior was individual sampled at the MNP foraging aggrega-
set for each stock as the distance from the putative tions carried haplotype CmP22.1, which is known to
stock to the foraging area, over the sum of distances occur at nesting rookeries in the CP (see below; Dut-
from all stocks to the foraging area. These distances ton et al. 2014a). Therefore, 6 out of the 10 haplo-
were obtained using Google Earth. ‘Population size’ types were categorized as ‘orphan’ as they have not
prior was set as the proportion of the size of each stock been reported in any rookery, including 3 new haplo-
to the total population size derived by the summation types reported here for the first time (CmP93.2,
of all stock abundances included in the analysis CmP4.19, and CmP112.1; GenBank accession num-
(Table S2). For each of the 3 models (uniform, geo- bers KY350180, KX499513, and KY350181, respec-
graphic distance, and rookery population size), a total tively). Although CmP112.1 and CmP4.19 have been
of 50 000 Markov Chain Monte Carlo generations previously identified in foraging individuals in GPS
256 Endang Species Res 32: 251–264, 2017
Table 1. mtDNA haplotype profiles in Ecuadorian green turtle Chelonia mydas habitats along with their GenBank accession
numbers. Haplotype names follow conventional nomenclature. ‘GPS nesting’ column corresponds to Galápagos (GPS) rookeries
data from Dutton et al. (2014a); sources for the other data are listed in the ‘Reference’ column. MNP: Machalilla National Park
Haplotype MNP MNP GPS GPS Total GenBank Reference
foraging nesting foraging nesting acc. no.
CmP4.1 11 3 13 26 53 KC306666 Dutton et al. (2014a)
CmP4.4 2 2 2 6 12 KC306665 Dutton et al. (2014a)
CmP4.6 18 12 16 51 97 KC306647 Dutton et al. (2014a)
CmP4.7 5 12 12 31 60 KC306660 Dutton et al. (2014a)
CmP4.9 0 0 0 1 1 KC306643 Dutton et al. (2014a)
CmP4.11 0 0 0 1 1 KC306661 Dutton et al. (2014a)
CmP4.15a 0 0 1 0 1 KX499506 Dutton et al. (2014a)
CmP15.1 0 0 1 3 4 KC306649 Dutton et al. (2014a)
CmP17.1 1 0 2 4 7 KC306648 Dutton et al. (2014a)
CmP24.1 2 0 0 1 3 KC306646 Dutton et al. (2014a)
CmP22.1 1 0 0 0 1 KF311747 Dutton et al. (2014b)
CmP93.1 0 0 0 2 2 FJ917194 Dutton et al. (2014a)
CmP93.2b 1 0 1 0 2 KY350180 This study
CmP97.1a 1 0 9 0 10 FJ917198 P. H. Dutton et al. (unpubl.)
CmP94.1a 1 0 0 0 1 FJ917193 P. H. Dutton et al. (unpubl.)
CmP112.1b 0 0 3 0 3 KY350181 This study
CmP4.19b 0 0 1 0 1 KX499513 This study
Total 43 29 61 126 259
aOrphan haplotypes reported in other foraging aggregations; bNovel orphan haplotypes reported here for the first time
(P. H. Dutton et al. pers. comm.) and Costa Rica (M. The Ecuadorian nesting aggregation at MNP had a
Heidemeyer unpubl.), these sequences have not pre- moderate h of 0.665 and low π of 0.002 (Table 2), indi-
viously been published. A fourth orphan haplotype, cating the presence of genetically similar haplotypes.
CmP4.15, recently re ported in Gorgona Islands in In contrast, these estimates were substantially higher
Colombia (M. Sanchez pers. comm.), was found in at both MNP and GPS foraging aggregations, where
GPS foraging aggregations. The last 2 orphan haplo- h was 0.749 and 0.831, respectively, indicating the
types in MNP and GPS foraging grounds (CmP97.1 abundance of haplotypes with origins other than
and CmP94.1) were rep orted as orphan in the CP Ecuadorian nesting grounds. Nucleotide diversity at
(Palmyra Atoll: Naro-Maciel et al. 2014). the MNP foraging ground was moderate, with π =
As mentioned previously, 4 haplotypes were the 0.004, and almost 4 times higher at the GPS foraging
most dominant types across all habitats. At the aggregation, with π = 0.013, which reflects the abun-
MNP nesting aggregations, haplotypes
CmP4.6 and CmP4.7 were present at high
frequencies, each with 12 individuals out Table 2. Mitochondrial control region sequence diversity for Ecuadorian
green turtles Chelonia mydas. Sample size (n), number of haplotypes (H),
of the 29 sampled (41.3%). In GPS nesting nucleotide diversity (π, with standard deviation, SD), and haplotype di-
individuals, the 3 haplotypes with the versity (h, with SD) values are shown. Data for Galápagos (GPS) rook-
highest frequencies were CmP4.6 (n = 51; eries were taken from Dutton et al. (2014a). Values for GPS and
40.5%), CmP4.7 (n = 31; 24.6%), and Machalilla National Park (MNP) foraging aggregations (bottom row for
CmP4.1 (n = 26; 20.6%) out of the 126 indi- each location) correspond to results without Western Pacific haplotypes
viduals analyzed by Dutton et al. (2014a).
The MNP foraging aggregations were Location n H π SD h SD
dominated by haplotypes CmP4.6 and MNP rookery 29 4 0.002 0.0014 0.665 0.0507
CmP4.1 (n = 18; 41.8% and n = 11; 25.5%, MNP foraging aggregations 43 10 0.004 0.0027 0.749 0.0511
respectively). Finally, GPS foraging indi- 41 8 0.002 0.0017 0.731 0.0505
viduals (n = 61) presented haplotypes GPS rookeries 126 10 0.001 0.0010 0.734 0.0234
CmP4.6, CmP4.1, and CmP4.7 in similar GPS foraging aggregations 61 11 0.013 0.0069 0.831 0.0225
frequencies (n = 16, 26.2%; n = 13, 21.3%; 49 9 0.002 0.0014 0.773 0.0306
and n = 12, 19.6%, respectively).
Chaves et al.: Green sea turtles in Ecuador 257
Table 3. Pairwise FST, ΦST, and Fisher’s exact test for population differentiation affinity of Cmp4.19 (Dutton et al.
between: (1) Galápagos (GPS) and Machalilla National Park (MNP) rookeries, 2014a) and Cmp4.15 from foraging
(2) GPS and MNP foraging aggregations, and (3) GPS and MNP foraging
a ggregations without Western Pacific (WP) haplotypes (no WP). Significance aggregations from GPS to haplotypes
values from Arlequin were at the 0.05 level CmP93.2 and CmP4.6 (Dutton et al.
2014a) from rookeries at GPS, MNP,
F Φ Exact test Michoacán, and Costa Rica. AlthoughST ST
haplotype Cmp4.6 was found in the
Rookeries 0.0073 (p > 0.05) 0.052 (p > 0.05) 0.738 (p > 0.05) EP Clade III by Dutton et al. (2014a),
Foraging 0.0097 (p > 0.05) 0.025 (p > 0.05) 0.106 (p > 0.05) our phylog enetic reconstruction placed
Foraging (no WP) 0.014 (p > 0.05) 0.0643 (p = 0.018) 0.600 (p > 0.05)
it within samples that correspond to
EP Clade II. Surprisingly, 3 hap -
lotypes found in GPS and MNP
dance of genetically distant haplotypes endemic to f oraging aggregations (CmP97.1, CmP112.1, and
the WP (see below; Table 2). These values were CmP22.1) corresponded to haplotypes found in CP
much lower (π = 0.002) when we removed WP haplo- and WP. Haplotype CmP22.1 has been found in nest-
types from the analysis (Table 2). ing rookeries from American Samoa, the Marshall
Islands, and Ulithi Atoll (Dutton et al. 2014b), while
the orphan haplotypes CmP97.1 and CmP112.1 are
Population structure phylogenetically related to WP rookeries (Fig. 2)
(Dutton et al. 2014a, Naro-Maciel et al. 2014, P. H.
There was no significant genetic differentiation be - Dutton unpubl.). The last 3 haplotypes were found in
tween the 2 Ecuadorian nesting populations (Table 3), 19.7% of GPS foraging grounds and in 4.6% of the
suggesting genetic connectivity between GPS and MNP foraging aggregation.
MNP (FST = 0.0073, p > 0.05; ΦST = 0.052, p > 0.05). The Bayesian calibrated tree supports the origin of
Foraging aggregations without WP haplotypes fol- green turtles in the Pliocene ca. 4.40 million yr (HPD
lowed the same pattern as nesting populations of lit- = 2.70−6.37; stem-based estimate) falling within the
tle genetic differentiation (FST = 0.0097, p > 0.05; ΦST range proposed using whole mitogenone calibration
= 0.025, p > 0.05), with most of the variation ex - (Duchene et al. 2012). Our tree also suggests a young
plained within populations (AMOVA). When WP origin (<0.1 million yr) for all haplotypes found in
haplotypes were incorporated into the analysis, only Ecuadorian rookeries.
sequence-based ΦST distance measures were signifi- The haplotype network of Ecuadorian rookeries
cant (FST = 0.014, p > 0.05; ΦST = 0.0643, p = 0.018), (MNP and GPS; Fig. 3) showed most nesting females
with a small proportion of variance (~5%) explained bearing the aforementioned common haplotypes
among groups. (CmP4.6, CmP4.7, and CmP4.1; see also Table 1),
and few individuals characterized by less common
haplotypes that had connections of 1 bp in a star-like
Phylogenetic and phylogeographic analysis fashion. The 3 common haplotypes were also com-
monly found in several individuals from Michoacán
To determine phylogenetic affinities and origins of and Revillagigedo when pooled with data from Gen-
new orphan haplotypes, the Bayesian reconstruction Bank (see Dutton et al. 2014a).
included the 17 haplotypes from Ecuador plus addi-
tional haplotypes from GenBank. Ecuadorian haplo-
types from nesting individuals were nested within 2 MSA
previously recognized clades (Fig. 2): (1) CP-EP
Clade (Dutton et al. 2014a) and (2) a WP clade (Deth- MSA was performed using only haplotypes with
mers et al. 2006, Naro-Maciel et al. 2014). The CP-EP known origin. Because MNP and GPS nesting popu-
clade includes 3 subclades with samples from Ha - lations were not significantly different (population
waii, Revillagigedo, Michoacán, Costa Rica, Galápa- genetic results), these 2 stocks were pooled into a
gos (Dutton et al. 2014a), and MNP in coastal single GPS-MNP entity for subsequent analyses. The
Ecuador. The lack of support in several clusters with - MSA was run using uniform and weighted priors
in the CP-EP clades failed to recover the mono- (population size, distance from source rookery); how-
phyletic clades as reported by Dutton et al. (2014a). ever, weighted priors are only useful for estimating
Nevertheless, the Bayesian tree supports the genetic source contributions when the estimates using uni-
258 Endang Species Res 32: 251–264, 2017
Fig. 2. Phylogenetic reconstruction of green turtle Chelonia mydas mitochondrial DNA control region haplotypes combining
sequences from this study and those from Dutton et al. (2014a). Only haplotypes found in Ecuador are labeled with standard
nomenclature. Boxes with different shading next to haplotypes indicate their presence in foraging aggregations in the GPS
(black squares) or MNP (dark grey squares), or in rookeries in GPS (white squares) or MNP (light grey squares). Orphan hap-
lotypes are marked with a star. Mean marginal means for age estimates are shown as blue bars on nodes. Scale at bottom is in
millions of years. Haplotype names follow Table 1. Terminal branches were collapsed (triangles) and outgroups removed for 
better visualization. PP: posterior probability
form priors (based on stock haplotype
frequencies only) fail to provide Table 4. Mixed stock analysis estimated contributions of Pacific green turtle
Chelonia mydas stocks to Machalilla National Park (MNP) and Galápagos
explanatory results. In our case, the (GPS) foraging aggregations. Results are shown for flat-prior analysis only
uniform prior provided a reasonable (mean and 95% confidence intervals are indicated). Estimated contributions 
estimate on the contributions by the >5% are shown in bold; blank cells indicate no data
major stocks to the analyzed foraging
grounds, on which we fo cus hereafter. Stock MNP GPS
The estimated cont ributions indicate Mean CI (95%) Mean CI (95%)
a predominance of the combined
Revillagigedo (Mexico) 0.41 0.00−3.71 0.56 0.00−4.14
GPS-MNP stock to both foraging Michoacán (Mexico) 1.02 0.00−8.83 1.26 0.00−9.29
grounds with average contributions Nombre de Jesús (Costa Rica) 0.77 0.00−6.72 0.78 0.00−5.95
of around 96.5% (95% to GPS and GPS-MNP 94.89 72.95−96.64 96.98 87.33−99.97
98% to MNP; Table 4, Fig. S1 in Hawaii 0.69 0.00−2.84 0.42 0.00−3.15
American Samoa 1.16 0.00−11.15
the Supplement at www. int-res.c om/ Ulithi Atoll 1.16 0.00−10.96
articles/s uppl/n 032 p251_ s upp. pdf). Marshall Islands 1.13 0.00−10.68
The Michoacán source ap peared to
Chaves et al.: Green sea turtles in Ecuador 259
Fig. 3. Haplotype network of
nes ting green turtle Chelonia
mydas mitochondrial control re-
gion haplotypes found in the
Eastern Pacific. Each circle cor-
responds to a unique haplotype,
the size is proportional to the
number of sampled individuals
carrying that haplotype over all
data used (this study + Dutton et
al. 2014a), and the connections
represent 1 base pair (bp) dif -
ference unless specified other-
wise (number in parentheses).
Dashed lines correspond to al-
ternative haplotype connections,
and dotted lines depict connec-
tions of 30 bp. Haplotype names
reported in Ecuador follow
Table 1. MNP: Machalilla Na-
tional Park, WP: Western Pacific
provide a small contribution of 1.3 and 1.0% to GPS and CmP 22.1) (Fig. S2) corresponded to turtles with
and MNP, respectively. For MNP, a minimal contri- yellow morphotypes; all 4 immature/juvenile individ-
bution of 1.2 and 1.1% was estimated to come from uals for GPS (mean CCL = 53.75 cm, range = 44−76)
American Samoa and Micronesia (Ulithi Atoll and and 1 female in MNP (CCL = 73 cm). Nesting females
Marshall Islands), respectively. The estimated contri- in MNP presented a mean CCL of 86.52 cm (n = 27;
butions to both Ecuadorian foraging grounds from SD = 5.25 cm, range = 77.3−98.0 cm).
Revillag igedo Archipelago, Nombre de Jesús (Costa
Rica), and Hawaiian rookeries were negligible (mean
of ≤1.0%). Given that, with the exception of the DISCUSSION
Ecuadorian sources, all estimates include 0 within
their confidence limits, it is likely that only the Rookeries
Ecuadorian rookeries contribute to the foraging sites
under study (Table 4). The genetic similarity found between coastal Ecua- 
dor (MNP) and Galapagos rookeries (GPS) suggests
that these 2 nesting assemblages are behaving as a
Morphological diversity of foraging individuals single panmictic stock. Both nesting sites share
unique haplotypes (CmP4.4 and CmP4.7), and shared
Size class histograms show a substantial number of haplotypes (CmP4.1 and CmP4.6) found in green tur-
immature individuals in both foraging grounds. Based tle nesting beaches across the EP (Table S1: Hawaii,
on the cut-off size of 70 cm to distinguish immature Michoacán, and Costa Rica). Although green turtle
from mature individuals (see ‘Materials and methods’), rookeries that are more than 500 km apart may be-
we recorded a total of 18 immature/juveniles (<70 cm have as independent stocks (Dethmers et al. 2006,
CCL) and 43 adults (>70 cm CCL), of which 20 were Bowen & Karl 2007, Dutton et al. 2014b, Shamblin et
females and 23 males (Fig. S2 in the Supplement). On al. 2015), the GPS and MNP rookeries are genetically
the other hand, MNP foraging individuals were domi- similar despite the fact that they are about 1000 km
nated by immature/juvenile turtles (n = 26). Remain- apart. However, our results were not unexpected
ing adult individuals sampled were 11 females and 6 given that a similar pattern was found between GPS
males (Fig. S2). In both fora ging aggregations, all WP and Mexican rookeries using nuclear microsatellites
haplotype-bearing individuals (CmP 97.1, CmP 112.1, and mtDNA data over larger distances than those pre-
260 Endang Species Res 32: 251–264, 2017
sented here (Dethmers et al. 2006, Dutton et al. 2008). and MNP foraging grounds (GPS π = 0.014 vs. MNP π
We hypothesize that this finding might be ex plained = 0.005) clearly reflects this pattern where apparent
by at least 3 scenarios: (1) high levels of ongoing gene WP haplotypes were more abundant in GPS waters
flow between GPS and MNP females, (2) recent colo- (18%) than in MNP (<5%) (Table S2). Furthermore,
nization of nesting females to either of these beaches only the nucleotide-based measure of genetic dis-
with little time for genetic differentiation (i.e. evolu- tance (and not haplotype frequency measures) was
tionary connectivity), or (3) insufficient resolution us- significant between these regions, consistent with
ing currently available genetic markers. Although the presence of highly divergent haplotypes found in
none of these were explicitly tested here, our time- GPS foraging individuals. The observed differences
calibrated tree coincides with a very recent origin for were non-significant when removing these WP hap-
the region’s haplotypes (<0.1 million yr). Furthermore, lotypes from the analysis, which is clear evidence of
the star-like haplotype network for MNP and GPS their influence inflating values of π, thus increasing
rookeries (Fig. 3) revealed little genetic struc ture. the genetic composition of the foraging population in
This result is indicative of recent shared an cestry or GPS. The orphan haplotype CmP97.1, closely related
recent divergence of several endemic haplot ypes sep- to foraging haplotypes from Palmyra Atoll (Naro-
arated by 1 bp from more common and widely distrib- Maciel et al. 2014), was found in 12 of 61 (19.67%)
uted haplotypes, ultimately suggesting recent colo- turtles from GPS foraging grounds, and was previ-
nization (see Avise 2000). A common limitation of the ously also reported from foraging grounds off Gorg-
use of mtDNA in marine turtle studies in general is the ona Island in Colombia (Amorocho et al. 2012). The
difficulty to detect population structure at fine scales orphan and new haplotype CmP112.1, which is
(reviewed by Bowen & Karl 2007). phylog enetically related to WP rookeries, was found
We recognize that a more accurate estimate of gene at much lower frequency (4.91%, 3 out of 61) in the
flow between highly similar rookeries from MNP and GPS foraging grounds (2 immature and 1 mature
GPS could require a combination of both nuclear and female; Fig. S2). While their exact origins remain
mitochondrial DNA (bi-parental and female-mediated unknown, Ecuadorian individuals carrying these 2
gene flow) markers. Alternatively, the use of mito- orphan haplotypes exhibit the ‘yellow morphotype’
chondrial short tandem repeats (Tiko chinski et al. characteristic of WP green turtles (see Zárate 2012),
2012) or mitogenomic sequencing could help reveal which offers a possible clue to their origins. The ap -
cryptic differentiation among Ecuadorian populations, parent orphan haplotypes, along with the previously
as found in Caribbean rookeries (Shamblin et al. reported CmP22.1 from American Samoa, Marshall
2012). Despite the limitations, these are robust results Islands, and Ulithi Atoll (Dutton et al. 2014b), repre-
which suggest close genetic links bet ween the sent evidence for trans-Pacific associations, suggest-
Ecuadorian stocks in spite of their geographic separa- ing that trans-oceanic links among Pacific green tur-
tion that could be used as the first step in redefining tles are more widespread than previously thought for
MUs for these populations. At a reg ional scale, and the species but not unique among sea turtles (Caretta
besides the sharing of common haplotypes between caretta: Bowen et al. 1995, Dermochelys coriacea:
MNP and GPS, the haplotype network showed that Dutton et al. 2000). A simple visualization of passive
MNP nesting females shared haplotypes with individ- drift movements by oceanic currents (adrift.org.au)
uals breeding in Mexico and Costa Rica (CmP4.6 and from putative CP and WP sites provide possible
CmP4.7; Fig. 3). The MNP is apparently an important routes to EP regions (e.g. GPS, MNP, and Gorgona;
nesting area for green turtles that share genetic ties Amorocho et al. 2012). Thus, dispersal of individuals
with other rookeries along the EP. Thus, our study ex- carrying WP haplotypes could be facilitated towards
pands the geographic distribution of these haplotypes eastern regions either via the South Equatorial cur-
in the EP to the newly sampled MNP beaches. rent in combination with the Humboldt Current from
the south, or the North Equatorial current in combi-
nation with El Niño from the north. Previous satellite
Foraging grounds telemetry studies (C. caretta, D. coriacea, Eretmo- 
chelys imbricata: Mast et al. 2016) and tag recovery
Foraging grounds at MNP and GPS exhibited rela- from hatchlings captured in longline fisheries (C.
tively high π and h diversity, characteristic of habitats caretta; Boyle et al. 2009) have identified the impor-
with a mixture of diverse local haplotypes (high h) tance of these dynamic oceanic currents in sea turtle
with highly divergent haplo types (high π). The lar - movements. Thus, these currents could not only
gest difference in nucleotide diversity between GPS explain the presence of trans-Pacific haplotypes in
Chaves et al.: Green sea turtles in Ecuador 261
the EP, but could also be responsible for the genetic MUs. On the other hand, foraging grounds in this
patterns of connectivity reported here between GPS study are found at the easternmost edge of the geo-
and MNP at a much smaller geographic scale. graphic range for green turtles in the EP, which may
In a regional comparison, and excluding divergent explain the limited exchange with other rookeries.
WP haplotypes from our analyses, the nucleotide Nevertheless, this pattern does not exclude the possi-
diversity in GPS foraging grounds (π = 0.002) is with- bility that individuals from local rookeries may feed
in the ranges found at other previously studied forag- at multiple, common foraging grounds. For example,
ing aggregations, such as the ones in CP (0.002− foraging turtles from GPS are mostly from rookeries
0.003, Hawaii; Dutton et al. 2008), Caribbean (0.003, in GPS, but turtles from GPS rookeries could also use
Costa Rica; Bjorndal et al. 2005), and Atlantic foraging grounds in GPS, MNP, and Gorgona Island
(0.002−0.006, Brazil; Naro-Maciel et al. 2006). How- and coastal areas in Central America, as supported
ever, when incorporated, the WP haplotypes inflate by post-nesting satellite-tacked individuals from
this population genetic metric (π = 0.013), placing the GPS (Seminoff et al. 2008). This is consistent with
GPS aggregation within much higher ranges like their life histories, since no breeding takes place at
those found at EP foraging aggregations (0.011, foraging habitats. The lack of genetic differentiation
Colombia; Amorocho et al. 2012). This result high- combined with the MSA results supports a high con-
lights the importance of GPS foraging grounds con- nectivity between MNP and GPS, and suggests the
tributing to the genetic diversity of green turtles in existence of an important marine corridor connecting
the EP. One such case is the orphan CmP4.15, a rare these 2 regions in Ecuador. Amorocho et al. (2012)
haplotype found in 1 individual in GPS foraging found that a high percentage of Gorgona green tur-
grounds and phylogenetically linked to haplotypes tles in Colombia have their origins in GPS nesting
found in rookeries from Ecuador and Mexico. Recent beaches. These findings together provide evidence
reports from Gorgona Island in Colombia have found for yet another axis of marine gateways in this region
this same haplotype in other foraging individuals (M. with important conservation implications.
Sanchez pers. comm.), thus indicating a broad distri-
bution regardless of its source. Further sampling on
rookeries along the EP coast of South, Central, and Implications for management and conservation
North America will be important to identify its origin.
Furthermore, the number of orphan haplotypes clus- Our study contributes information on the genetic
tering within EP clades reported here (CmP4.19, diversity and connectivity of green turtles found in an
CmP93.2 about 3% of all foraging turtles) and else- important region of Ecuador (MNP) that had been
where (CmP94.1; Dutton et al. 2014a) could suggest unstudied. The existing data from GPS rookeries and
either the presence of many genetically uncharacter- newly published breeding haplotypes from MNP
ized nesting stocks in the EP region, or the presence could be used to reassess the composition of green
of rare local haplotypes that have not been detected turtles in the southern range of the EP. The evidence
due to small sample sizes studied at rookeries. implies that green turtles born in GPS utilize MNP as
foraging grounds and vice versa, and hence conser-
vation efforts in both areas are required to protect
MSA these 2 genetic stocks. On the basis of the present
results, it would be justifiable to extend the previ-
Our MSA results showed a predominance of a ously defined Galápagos MU (Dutton et al. 2014a) to
GPS-MNP contribution to both foraging grounds include the continental Ecuador rookeries (MNP). A
(93.8% to GPS and 94.1% to MNP), revealing a very series of per vasive anthropogenic threats including
high degree of connectivity between rookery and for- loss of habitat, coastal development, effect of fishery
aging grounds in Ecuador. High connectivity among by-catch, poaching, boat strikes, and climate change
regional green turtles has also been documented are thought to determine marine turtle survival in the
within the Hawaiian MU, where most of the haplo- region. Although both nesting sites and waters are
types found in foraging aggregations have their ori- currently protected under Ecuadorian National Park
gins at proximate Hawaiian rookeries (Dutton et al. regulations, the potential impacts from these activi-
2008). Extremely low contributions to Ecuadorian for- ties are more evident in MNP (Peña-Mosquera 2010).
aging grounds from regional stocks (e.g. Costa Rica Including MNP populations in the Galápagos MU
and Mexico), could suggest a pattern for many forag- would allow for rapid implementation of conserva-
ing grounds along the EP composed mostly of local tion strategies for coastal populations of green turtles
262 Endang Species Res 32: 251–264, 2017
using the better consolidated conservation scope A, Rees AF, Williams K (comps) Book of abstracts. 26th
applied to green turtles in the iconic Galápagos Annu Symp Sea Turtle Biology. Island of Crete, Greece,
3–8 April 2006. International Sea Turtle Society, Athens,
Islands. Another possibility is to extend the current p 179 (Abstract only)
protection of the EP Tropical Marine Corridor to in - Amorocho DF, Abreu-Grobois FA, Dutton PH, Reina RD
corporate MNP within this axis of international con- (2012) Multiple distant origins for green sea turtles ag -
servation efforts, a suggestion that stems from the gregating off Gorgona Island in the Colombian Eastern
important connectivity shown between MNP, GPS, Pacific. PLOS ONE 7:e 31486
Anhalzer G, Ferrin JL, Baquero A, Vallejo F, Muñoz JP, Vera
and Gorgona in Colombia (Amorocho et al. 2012), F Chalen X (2012) Machalilla National Park, a critical
which should encourage cooperative national and nesting site for hawksbill (Eretmochelys imbricata) and
international efforts to reduce anthropogenic effects green (Chelonia mydas) sea turtles in the eastern Pacific.
primarily in MNP. This habitat is valuable regionally In: Jones TT, Wallace BP (comps) Proc 31st Annu Symp
Sea Turtle Biology and Conservation, 10–16 April 2011,
not only for green turtles but for multiple marine tur- San Diego, CA. NOAA Tech Memo NMFS-SEFSC-631.
tle species, including the Critically Endangered NOAA Southeast Fisheries Science Center, Miami, FL,
hawksbill. More broadly, and since a significant por- p 208−209 (Abstract only)
tion of the turtles at foraging sites in Ecuadorian Avise JC (2000) Phylogeography: the history and formation
of species. Harvard University Press, Cambridge, MA
waters app ear to be of WP origin, our study stresses Bass AL, Witzell WN (2000) Demographic composition of
the importance of the conservation value for this immature green turtles (Chelonia mydas) from the east
region by extending conservation attention across central Florida coast: evidence from mtDNA markers.
the Pacific, reaching the opposite side of the ocean Herpetologica 56:3 57−367
basin. Finally, and given that the origin of many Bjorndal KA, Bolten AB, Troëng S (2005) Population struc-
ture and genetic diversity in green turtles nesting at Tor-
orphan foraging individuals in the region remains tuguero, Costa Rica, based on mitochondrial DNA con-
unresolved, we re commend exploration of potential trol region sequences. Mar Biol 147: 1449−1457
rookery sites north and south of MNP to improve Bolker B, Okuyama T, Bjorndal K, Bolten A (2003) Sea turtle
conservation efforts for green turtles along the EP. stock estimation using genetic markers: accounting for
sampling error of rare genotypes. Ecol Appl 13:7 63−775
Bolten AB (2003) Variation in sea turtle life history patterns: 
neritic vs. oceanic developmental stages. In: Lutz PL,
Acknowledgements. We thank the Tortuga Negra-Galápa- Musick J, Wyneken J (eds) The biology of sea turtles, Vol
gos project and all the people involved, and the following 2. CRC Press, Boca Raton, FL, p 243−257
institutions for their support: Galápagos Science Center, Bolten AB, Bjorndal KA (1992) Blood profiles for a wild pop-
Universidad San Francisco de Quito, GAIAS, Parque ulation of green turtles (Chelonia mydas) in the southern
Nacional Galápagos, Parque Nacional Machalilla, Univer- Bahamas: size-specific and sex-specific relationships.
sity of North Carolina at Chapel Hill, Equilibrio Azul, Ponti- J Wildl Dis 28:4 07−413
ficia Universidad Católica del Ecuador, SharkSky-Galápa- Bowen B, Karl S (2007) Population genetics and phylogeo -
gos, and James Cook University. This work was possible graphy of sea turtles. Mol Ecol 16: 4886−4907
under permits authorized by the Ministerio del Ambiente Bowen BW, Meylan AB, Ross JP, Limpus CJ, Balazs GH,
(005-DRM-MA, 011-DRM-MA) and the Parque Nacional Avise JC (1992) Global population structure and natural
Galápagos (PC-20-14). We thank the following people for history of the green turtle (Chelonia mydas) in terms of
their valuable support: Daniela Alarcón-Ruales, Jason G. matriarchal phylogeny. Evolution 46: 865−881
Castaneda, Carlos F. Mena, Stephen Walsh, Andrés Bowen BW, Abreu-Grobois FA, Balazs GH, Kamezaki N,
Baquero, Maximilian Hirschfeld, Judith Denkinger, Leandro Limpus CJ, Ferl RJ (1995) Trans-Pacific migrations of the
Vaca-Pita, Daniela Matheus-Holland, Andrés Leon, Cori loggerhead turtle (Caretta caretta) demonstrated with
Lopazanski, Cristian Quintero, Federico Idrovo-Bermeo, mitochondrial DNA markers. Proc Natl Acad Sci USA 92: 
Juan García, Jorge Torres, Maryuri Yépez, Galo Quezada, 3731−3734
Silvya Sotamba, Sofia Tacle, Danilo Silva, and Ernesto Boyle MC, FitzSimmons NN, Limpus CJ, Kelez S, Velez-
Briones. Antonio León, Maria de Lourdes Torres, and Ga brie - Zuazo X, Waycott M (2009) Evidence for transoceanic
la Castillo provided laboratory support. We are grateful to migrations by loggerhead sea turtles in the southern
Alexander Gaos, Michael Jensen, Ryan Harrigan, and John Pacific Ocean. Proc R Soc Lond B Biol Sci 276:1 993−1999
Rowe for comments that greatly improved the manuscript. Browne DC, Horrocks JA, Abreu-Grobois FA (2010) Popula-
Funds were provided by Ecofondo, Pontificia Universidad tion subdivision in hawksbill turtles nesting on Barbados,
Católica del Ecuador, SENESCYT, the USFQ Chancellor West Indies, determined from mitochondrial DNA con-
Grant, and the Galápagos Science Center Research Grant. trol region sequences. Conserv Genet 11:1 541−1546
Burnham KP, Anderson DR (2002) Model selection and mul-
timodel inference: a practical information-theoretic
LITERATURE CITED Approach. Springer-Verlag, New York, NY
Carr A (1987) New perspectives on the pelagic stage of sea
Abreu-Grobois A, Horrocks J, Formia A, Dutton P and oth- turtle development. Conserv Biol 1: 103−121
ers (2006) New mtDNA Dloop primers which work for a Clement M, Posada D, Crandall KA (2000) TCS: a computer
variety of marine turtle species may increase the resolu- program to estimate gene geneologies. Mol Ecol 9: 
tion of mixed stock analyses. In: Frick M, Panagopoulou 1657−1659
Chaves et al.: Green sea turtles in Ecuador 263
Dethmers KE, Broderick D, Moritz C, Fitzsimmons NN and Spatial and temporal genetic variation among size
others (2006) The genetic structure of Australasian green classes of green turtles (Chelonia mydas) provides infor-
turtles (Chelonia mydas): exploring the geographical mation on oceanic dispersal and population dynamics.
scale of genetic exchange. Mol Ecol 15: 3931−3946 Mar Ecol Prog Ser 543:2 41−256
Drummond A, Rambaut A (2007) BEAST: Bayesian evolu- Joseph J, Nishizawa H, Arshaad WM, Kadir SAS and others
tionary analysis by sampling trees. BMC Evol Biol 7: 214 (2016) Genetic stock compositions and natal origin of
Duchene S, Frey A, Alfaro-Núñez A, Dutton PH, Gilbert green turtle (Chelonia mydas) foraging at Brunei Bay.
MTP, Morin PA (2012) Marine turtle mitogenome phylo- Glob Ecol Conserv 6: 16−24
genetics and evolution. Mol Phylogenet Evol 65: 241−250 Kearse M, Moir R, Wilson A, Stones-Havas S and others
Dutton PH, Balazs G, Dizon A, Barragan A (2000) Genetic (2012) Geneious Basic: an integrated and extendable
stock identification and distribution of leatherbacks in desktop software platform for the organization and
the Pacific: potential effects on declining populations. In: analysis of sequence data. Bioinformatics 28: 1647−1649
Abreu-Grobois FA, Briseno-Duenas R, Marquez-Millán Lohmann KJ, Witherington BE, Lohmann CMF, Salmon M
R, Sarti-Martinez L (eds) Proc 18th International Sea Tur- (1997) Orientation, navigation, and natal beach homing
tle Symp. Tech Memo NMFS-SEFSC 436, US Dept Com- in sea turtles. In: Lutz P, Musick J (eds) The biology of
merce, NOAA, Miami, FL, p 38−39 (Abstract only) sea turtles. CRC Press, Boca Raton, FL, p 107–136
Dutton PH, Balazs GH, LeRoux RA, Murakawa SKK, Zarate Luschi P, Hays GC, Papi F (2003) A review of long distance
P, Martínez LS (2008) Composition of Hawaiian green movements by marine turtles, and the possible role of
turtle foraging aggregations: mtDNA evidence for a dis- ocean currents. Oikos 103: 293−302
tinct regional population. Endang Species Res 5:3 7−44 Mast RB, Hutchinson BJ, Villegas PE (eds) (2016) State of
Dutton PH, Jensen MP, Frey A, LaCasella E and others the World’s Sea Turtles (SWOT) Report, Book XI. Special
(2014a) Population structure and phylogeography reveal Feature - South America. Oceanic Society, Ross, CA
pathways of colonization by a migratory marine reptile Mayr E (1963) Animal species and evolution. Harvard Uni-
(Chelonia mydas) in the central and eastern Pacific. Ecol versity Press, Cambridge, MA
Evol 4: 4317−4331 Moncada F, Abreu-Grobois FA, Muhlia-Melo A, Bell C and
Dutton PH, Jensen MP, Frutchey K, Frey A and others others (2006) Movement patterns of green turtles (Chelo-
(2014b) Genetic stock structure of green turtle (Chelonia nia mydas) in Cuba and adjacent Caribbean waters
mydas) nesting populations across the Pacific Islands. inferred from flipper tag recaptures. J Herpetol 40: 22−34
Pac Sci 68: 451−464 Moritz C (1994) Applications of mitochondrial DNA analysis
Engstrom TN, Meylan PA, Meylan AB (2002) Origin of in conservation: a critical review. Mol Ecol 3: 401−411
 juvenile loggerhead turtles (Caretta caretta) in a tropical Musick JA, Limpus CJ, Lutz P, Musick J (1997) Habitat uti-
dev elopmental habitat in Caribbean Panama. Anim lization and migration in juvenile sea turtles. In: Lutz P,
 Conserv 5: 125−133 Musick J (eds) The biology of sea turtles. CRC Press,
Felsenstein J (1981) Evolutionary trees from DNA Boca Raton, FL, p 137−163
sequences: a maximum likelihood approach. J Mol Evol Naro-Maciel E, Becker JH, Lima EHSM, Marcovaldi MÂ,
17:3 68−376 DeSalle R (2006) Testing dispersal hypotheses in forag-
Formia A, Godley B, Dontaine JF, Bruford M (2006) Mito- ing green sea turtles (Chelonia mydas) of Brazil. J Hered
chondrial DNA diversity and phylogeography of endan- 98: 29−39
gered green turtle (Chelonia mydas) populations in Naro-Maciel E, Le M, FitzSimmons NN, Amato G (2008)
Africa. Conserv Genet 7: 353−369 Evolutionary relationships of marine turtles: a molecular
Gelman A, Rubin DB (1992) Inference from iterative simula- phylogeny based on nuclear and mitochondrial genes.
tion using multiple sequences. Stat Sci 7: 457−472 Mol Phylogenet Evol 49: 659−662
Hamabata T, Kamezaki N, Hikida T (2014) Genetic struc- Naro-Maciel E, Gaughran SJ, Putman NF, Amato G and oth-
ture of green turtle (Chelonia mydas) peripheral popula- ers (2014) Predicting connectivity of green turtles at
tions nesting in the northwestern Pacific rookeries: evi- Palmyra Atoll, central Pacific: a focus on mtDNA and dis-
dence for northern refugia and postglacial colonization. persal modelling. J R Soc Interface 11:2 0130888
Mar Biol 161: 495−507 Nishizawa H, Naito Y, Suganuma H, Abe O and others
Hamann M, Schäuble CS, Simon T, Evans S (2006) Demo- (2013) Composition of green turtle feeding aggregations
graphic and health parameters of green sea turtles Che- along the Japanese archipelago: implications for
lonia mydas foraging in the Gulf of Carpentaria, Aus- changes in composition with current flow. Mar Biol 160: 
tralia. Endang Species Res 2: 81−88 2671−2685
Heidemeyer M (2015) Orígenes natales y migratorios de la Nylander JAA, Wilgenbusch JC, Warren DL, Swofford DL
agregación de tortuga negra (Chelonia mydas agassizii) (2008) AWTY (are we there yet?): a system for graphical
en el hábitat de alimentación de la Isla del Coco basado exploration of MCMC convergence in Bayesian phyloge-
en análisis de ADN, bioquímicos y tecnología satelital. netics. Bioinformatics 24: 581−583
MSc thesis, Universidad de Costa Rica, San José Pella J, Masuda M (2001) Bayesian methods for analysis of
Jensen M (2010) Assessing the composition of green turtle stock mixtures from genetic characters. Fish Bull 99: 
(Chelonia mydas) foraging grounds in Australasia using 151−167
mixed stock analyses. PhD dissertation, University of Peña-Mosquera MP (2010) Diseño participativo de un plan
Canberra de acción para la recuperación y conservación de las
Jensen M, FitzSimmons N, Dutton P (2013) Molecular genet- t ortugas marinas en el Parque Nacional Machalilla,
ics of sea turtles. In: Wyneken J, Lohmann K, Musick J Manabí, Ecuador. MSc thesis, Universidad Internacional
(eds) Biology of sea turtles, Book III. CRC Press, Boca Menéndez Pelayo, Madrid
Raton, FL, p 134–154 Peña-Mosquera MP, Baquero-Gallegos AB, Muñoz-Pérez
Jensen MP, Bell I, Limpus CJ, Hamann M and others (2016) JM, Puebla-Jiménez F, Alvarez V, Chalen-Noroña X
264 Endang Species Res 32: 251–264, 2017
(2009) El Parque Nacional Machalilla: zona crítica de southern Greater Caribbean green turtle rookeries. Mol
anidación para la tortuga carey (Eretmochelys imbricata) Ecol 21: 2330−2340
y verde (Chelonia mydas) en el Ecuador y el Pacífico Ori- Shamblin BM, Bagley DA, Ehrhart LM, Desjardin NA and
ental. Temporadas 2007-2009. www.equilibrioazul.org/ others (2015) Genetic structure of Florida green turtle
documentos/P ublicaciones/Pe%F1a%20et%20al_2 009_ rookeries as indicated by mitochondrial DNA control
Anidacion_PNM_Equilibrioazul.pdf region sequences. Conserv Genet 16:6 73−685
Posada D (2008) jModelTest: phylogenetic model averaging. Tikochinski Y, Bendelac R, Barash A, Daya A, Levy Y, Fried-
Mol Biol Evol 25:1 253−1256 mann A (2012) Mitochondrial DNA STR analysis as a tool
Proietti MC, Reisser JW, Kinas PG, Kerr R, Monteiro D, for studying the green sea turtle (Chelonia mydas)
Marins LFF, Secchi ER (2012) Green turtle (Chelonia p opulations: the Mediterranean Sea case study. Mar
mydas) mixed stocks in the western South Atlantic, as Genomics 6:1 7−24
revealed by mtDNA haplotypes and drifter trajectories. Vieyra M (2006) A molecular approach to understanding the
Mar Ecol 447:1 −41 importance of olfaction in the life history of sea turtles: 
Rambaut A, Drummond AJ (2007) Tracer v1.4. http: //beast. turtle olfactory receptor gene phylogeny, allelic diversity
bio.ed.ac.uk/Tracer and evidence for selection. PhD dissertation, University
Read TC, Wantiez L, Werry JM, Farman R, Petro G, Limpus of South Carolina, Columbia, SC
CJ (2014) Migrations of green turtles (Chelonia mydas) Wallace BP, DiMatteo AD, Hurley BJ, Finkbeiner EM and
between nesting and foraging grounds across the Coral others (2010) Regional management units for marine tur-
Sea. PLOS ONE 9: e100083 tles: a novel framework for prioritizing conservation and
Reynolds J, Weir BS, Cockerham CC (1983) Estimation of research across multiple scales. PLOS ONE 5: e15465
the coancestry coefficient: basis for a short-term genetic Ward RD, Woodwark M, Skibinski DOF (1994) A compari-
distance. Genetics 105:7 67−779 son of genetic diversity levels in marine, freshwater, and
Roden SE, Morin PA, Frey A, Balazs GH, Zarate P, Cheng IJ, anadromous fishes. J Fish Biol 44: 213−232
Dutton PH (2013) Green turtle population structure in the Wibbels T (1999) Diagnosing the sex of sea turtles in forag-
Pacific: new insights from single nucleotide poly mor- ing habitats. In: Eckert KL, Bjorndal KA, Abeu-Grobois
phisms and microsatellites. Endang Species Res 20: FA, Donnelly M (eds) Research and management tech-
227−234 niques for the conservation of sea turtles. Publication No.
Ronquist F, Huelsenbeck JP (2003) MRBAYES 3: Bayesian 4 1−5, Book 4. IUCN/SSC Marine Turtle Specialist
phylogenetic inference under mixed models. Bioinfor- Group, Washington, DC, p 139−144 
matics 19: 1572−1574 Wyneken J, Lohmann KJ, Musick JA (eds) (2013) The bio- 
Schneider S, Roessli D, Excoffier L (2000) Arlequin v.2.0: a logy of sea turtles, Vol III. CRC Press, Boca Raton, FL
software for population genetics data analysis. Genetics Zárate P (2002) Evaluación de la actividad de anidación de
and Biometry Laboratory, University of Geneva, Geneva la tortuga verde Chelonia mydas, en las islas Galápagos
Seminoff JA, Zárate P, Coyne M, Foley DG, Parker D, Lyon durante la temporada 2001. Fundación Charles Darwin,
BN, Dutton PH (2008) Post-nesting migrations of Galápa- Puerto Ayora, Santa Cruz, Galápagos
gos green turtles Chelonia mydas in relation to oceano- Zárate P (2012) Offshore oasis: ecology of sea turtles at
graphic conditions: integrating satellite telemetry with re- oceanic islands of the eastern Pacific. In: Seminoff JA,
motely sensed ocean data. Endang Species Res 4: 57−72 Wallace BP (eds) Sea turtles of the eastern Pacific: ad -
Shamblin BM, Bjorndal KA, Bolten AB, Hillis-Starr ZM, vances in research and conservation. Arizona-Sonora
Lundgren IAN, Naro-Maciel E, Nairn CJ (2012) Mito - Desert Museum Studies in Natural History, p 63–87. The
genomic sequences better resolve stock structure of University of Arizona Press, Tucson, AZ
Editorial responsibility: Jeffrey Seminoff, Submitted: May 13, 2016; Accepted: January 7, 2017
La Jolla, California, USA Proofs received from author(s): March 1, 2017