Regional management units for marine turtles: a novel framework for prioritizing conservation and research across multiple scales.

Abstract

Resolving threats to widely distributed marine megafauna requires definition of the geographic distributions of both the threats as well as the population unit(s) of interest. In turn, because individual threats can operate on varying spatial scales, their impacts can affect different segments of a population of the same species. Therefore, integration of multiple tools and techniques--including site-based monitoring, genetic analyses, mark-recapture studies and telemetry--can facilitate robust definitions of population segments at multiple biological and spatial scales to address different management and research challenges.
To address these issues for marine turtles, we collated all available studies on marine turtle biogeography, including nesting sites, population abundances and trends, population genetics, and satellite telemetry. We georeferenced this information to generate separate layers for nesting sites, genetic stocks, and core distributions of population segments of all marine turtle species. We then spatially integrated this information from fine- to coarse-spatial scales to develop nested envelope models, or Regional Management Units (RMUs), for marine turtles globally.
The RMU framework is a solution to the challenge of how to organize marine turtles into units of protection above the level of nesting populations, but below the level of species, within regional entities that might be on independent evolutionary trajectories. Among many potential applications, RMUs provide a framework for identifying data gaps, assessing high diversity areas for multiple species and genetic stocks, and evaluating conservation status of marine turtles. Furthermore, RMUs allow for identification of geographic barriers to gene flow, and can provide valuable guidance to marine spatial planning initiatives that integrate spatial distributions of protected species and human activities. In addition, the RMU framework--including maps and supporting metadata--will be an iterative, user-driven tool made publicly available in an online application for comments, improvements, download and analysis.

Full text

Page 1

Regional Management Units for Marine Turtles: A Novel
Framework for Prioritizing Conservation and Research
across Multiple Scales
Bryan P. Wallace1,2,3*, Andrew D. DiMatteo1,4, Brendan J. Hurley1,2, Elena M. Finkbeiner1,3, Alan B.
Bolten1,5, Milani Y. Chaloupka1,6, Brian J. Hutchinson1,2, F. Alberto Abreu-Grobois1,7, Diego
Amorocho1,8, Karen A. Bjorndal1,5, Jerome Bourjea1,9, Brian W. Bowen1,10, Raquel Brisen ˜o Duen ˜as1,11,
Paolo Casale1,12,13, B. C. Choudhury1,14, Alice Costa1,15, Peter H. Dutton1,16, Alejandro Fallabrino1,17,
Alexandre Girard1,18, Marc Girondot1,19, Matthew H. Godfrey1,20, Mark Hamann1,21, Milagros Lo ´pez-
Mendilaharsu1,22,23, Maria Angela Marcovaldi1,22, Jeanne A. Mortimer1,24, John A. Musick1,25, Ronel
Nel1,26, Nicolas J. Pilcher1,27, Jeffrey A. Seminoff1,28, Sebastian Troe ¨ng1,2,29,30, Blair Witherington1,31,
Roderic B. Mast1,2
1International Union for Conservation of Nature (IUCN)/SSC Marine Turtle Specialist Group – Burning Issues Working Group, Arlington, Virginia, United States of America,
2Global Marine Division, Conservation International, Arlington, Virginia, United States of America, 3Center for Marine Conservation, Duke University, Beaufort, North
Carolina, United States of America, 4Marine Geospatial Ecology Laboratory, Duke University, Durham, North Carolina, United States of America, 5Department of Biology,
Archie Carr Center for Sea Turtle Research, University of Florida, Gainesville, Florida, United States of America, 6Ecological Modelling Services, Pty Ltd, University of
Queensland, Brisbane, Australia, 7Unidad Acade ´mica Mazatla ´n, Instituto de Ciencias del Mar y Limnologı ´a, Universidad Nacional Auto ´noma de Me ´xico, Mazatla ´n, Sinaloa,
Me ´xico, 8Centro de Investigacio ´n para el Medio Ambiente y Desarrollo, Cali, Colombia, 9Laboratoire Ressources Halieutiques, IFREMER, Ile Reunion, France, 10Hawaii
Institute of Marine Biology, Kaneohe, Hawaii, United States of America, 11Banco de Informacio ´n sobre Tortugas Marinas (BITMAR), Unidad Mazatla ´n, Instituto de Ciencias
del Mar y Limnologı ´a, Universidad Nacional Auto ´noma de Me ´xico, Mazatla ´n, Sinaloa, Me ´xico, 12Department of Biology and Biotechnology ‘‘Charles Darwin,’’ University of
Rome ‘‘La Sapienza,’’ Rome, Italy, 13World Wildlife Fund (WWF) Mediterranean Turtle Programme, World Wildlife Fund-Italy, Rome, Italy, 14Department of Endangered
Species Management, Wildlife Institute of India, Dehradun, Uttarakhand, India, 15World Wildlife Fund-Mozambique, Maputo, Mozambique, 16Southwest Fisheries
Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration (NOAA), La Jolla, California, United States of America, 17Karumbe ´,
Montevideo, Uruguay, 18Association RENATURA, Albens, France, and Pointe-Noire, Congo, 19Laboratoire d’Ecologie, Syste ´matique et Evolution, Universite ´ Paris-Sud,
Orsay, France, 20North Carolina Wildlife Resources Commission, Beaufort, North Carolina, United States of America, 21School of Earth and Environmental Sciences, James
Cook University, Townsville, Australia, 22Projeto Tamar-ICMBio/Fundac ¸a ˜o Pro Tamar, Salvador, Bahı ´a, Brazil, 23Department of Ecology, Institute of Biology, Universidade
do Estado do Rio de Janeiro, Rio de Janeiro, Brazil, 24Department of Biology, University of Florida, Gainesville, Florida, United States of America, 25Virginia Institute of
Marine Sciences, College of William and Mary, Gloucester Point, Virginia, United States of America, 26School of Environmental Sciences, Nelson Mandela Metropolitan
University, Summerstrand Campus, South Africa, 27Marine Research Foundation, Sabah, Malaysia, 28Marine Turtle Ecology and Assessment Program, Southwest
Fisheries Science Center, NOAA-National Marine Fisheries Service, La Jolla, California, United States of America, 29Department of Animal Ecology, Lund University, Lund,
Sweden, 30Scientific Advisory Committee, Sea Turtle Conservancy, Gainesville, Florida, United States of America, 31Florida Fish and Wildlife Conservation Commission,
Melbourne Beach, Florida, United States of America
Abstract
Background: Resolving threats to widely distributed marine megafauna requires definition of the geographic distributions
of both the threats as well as the population unit(s) of interest. In turn, because individual threats can operate on varying
spatial scales, their impacts can affect different segments of a population of the same species. Therefore, integration of
multiple tools and techniques — including site-based monitoring, genetic analyses, mark-recapture studies and telemetry
— can facilitate robust definitions of population segments at multiple biological and spatial scales to address different
management and research challenges.
Methodology/Principal Findings: To address these issues for marine turtles, we collated all available studies on marine
turtle biogeography, including nesting sites, population abundances and trends, population genetics, and satellite
telemetry. We georeferenced this information to generate separate layers for nesting sites, genetic stocks, and core
distributions of population segments of all marine turtle species. We then spatially integrated this information from fine- to
coarse-spatial scales to develop nested envelope models, or Regional Management Units (RMUs), for marine turtles globally.
Conclusions/Significance: The RMU framework is a solution to the challenge of how to organize marine turtles into units of
protection above the level of nesting populations, but below the level of species, within regional entities that might be on
independent evolutionary trajectories. Among many potential applications, RMUs provide a framework for identifying data
gaps, assessing high diversity areas for multiple species and genetic stocks, and evaluating conservation status of marine
turtles. Furthermore, RMUs allow for identification of geographic barriers to gene flow, and can provide valuable guidance
to marine spatial planning initiatives that integrate spatial distributions of protected species and human activities. In
addition, the RMU framework — including maps and supporting metadata — will be an iterative, user-driven tool made
publicly available in an online application for comments, improvements, download and analysis.
PLoS ONE | www.plosone.org1December 2010 | Volume 5 | Issue 12 | e15465

Page 2

Citation: Wallace BP, DiMatteo AD, Hurley BJ, Finkbeiner EM, Bolten AB, et al. (2010) Regional Management Units for Marine Turtles: A Novel Framework for
Prioritizing Conservation and Research across Multiple Scales. PLoS ONE 5(12): e15465. doi:10.1371/journal.pone.0015465
Editor: Steven J. Bograd, NOAA/NMFS/SWFSC, United States of America
Received August 2, 2010; Accepted October 1, 2010; Published December 17, 2010
This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration which stipulates that, once placed in the public
domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
Funding: This study was funded by the National Fish and Wildlife Foundation and the Offield Family Foundation. These funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the manuscript. MYC is employed by a commercial company that provides ecological modelling
services, and his involvement in this study was partially supported by this company. However, this support in no way biased his contributions to all aspects of this
study, the overall process or resulting products generated by this study.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: b.wallace@conservation.org
Introduction
Geospatial characterization of commercially important or
conservation-dependent marine species provides crucial input for
resource management in multi-use situations, as is currently
described by ecosystem-based marine spatial planning [1]. In
particular, linking impacts of various threats to widely distributed
marine megafauna populations (e.g. mammals, birds, turtles,
sharks) requires description of overlaps between threats and the
population segment of interest. Furthermore, widespread marine
species often exhibit inter-population variation in life history traits
and population dynamics that warrant population-specific man-
agement schemes [2,3].
However, resolution of population units for conservation is not
always straightforward, and requires clear understanding of
conservation objectives, as well as the natural history of – and
threats to – the species of interest [4]. For example, evolutionarily
significant units (ESUs) were first described to include sufficient
genetic diversity to retain evolutionary potential, and thus address
long-term conservation issues as well as historical population
trends [5,6]. In contrast, population segments or management
units (MUs) are functionally independent (i.e. exhibit distinct
demographic processes), can be characterized using various tools
or indicators, such as genetic markers, life history traits, behavior,
or morphology, and are appropriate short-term targets for
conservation [5]. A major challenge to prioritization schemes
arises when multiple population segments meet the criteria of a
MU, each deserving specially designed conservation strategies.
Six of the seven marine turtle species are categorized as
Vulnerable, Endangered, or Critically Endangered globally by the
IUCN Red List of Threatened Species [7], but threats on regional scales
can differentially affect life-stages of the same populations. Similar
to other long-lived marine vertebrates, marine turtles occupy
broad geographic ranges including separate breeding and feeding
areas utilized by adults, and in some cases geographically distinct
ontogenetic habitats for immature life stages, with different levels
of population overlap at each stage [8]. Furthermore, marine
turtles have complex population structures characterized by
female nesting site fidelity, male-mediated gene flow, and
population overlap during migrations and in developmental
habitats, with the degree of genetic population structuring
increasing with life stages (see [9] for review). Understanding the
complex relationships among various nesting sites, nurseries, and
foraging areas, particularly in the context of variation in
environmental conditions, is crucial to quantifying population-
level impacts of anthropogenic threats, as well as to designing
effective conservation responses to these threats [10–12]. Howev-
er, these complexities in population structures, habitat use,
environmental factors, and life-stage-specific threats have con-
founded the definition of marine turtle MUs.
Molecular genetic analyses are often used to describe population
structures and, by extension, to define MUs [4,5,13]. Due to
complex population structure and life history of marine turtles,
different molecular analyses can be applied to determine genetic
stock structure for different demographic segments of a popula-
tion. Maternally inherited mitochondrial DNA (mtDNA) is useful
for resolving nest site fidelity and homing behavior [14,15]. In
addition to defining nesting populations (i.e. one level of MUs), this
genetic marker is also useful for resolving maternal origin of both
males and females at various life stages and feeding habitats [9].
Nesting females typically demonstrate philopatry to nesting
areas, and both males and females can be philopatric to breeding
areas adjacent to a nesting beach [16]. However, males do not
restrict mating efforts to their ancestral breeding area, and
apparently copulate with females in coastal feeding habitat or
migratory corridors as well, where they encounter females from
other regional nesting populations. The result, registered in
biparentally-inherited nuclear DNA (nDNA), is that regional
nesting colonies are connected by this male-mediated gene flow
[2,17–18]. Because nDNA reflects contributions of both males and
females, analyses of nDNA markers (e.g. microsatellites) can
resolve breeding or reproductive stocks that encompass multiple
mtDNA-defined nesting stocks [9,19].
Genetic analyses clearly contribute to the resolution of MUs,
but are not always sufficient for this purpose; such analyses can
indicate that genetic population structure exists, but usually
cannot resolve the boundaries of that structure [4,20]. Unique
identifier tags applied to marine turtles at various life stages and
in various habitats via mark-recapture monitoring programs can
illuminate migration routes and connectivity of individual
animals between habitats [21–23]. Furthermore, the advent of
electronic tracking technology, specifically satellite telemetry and
remote sensing, has facilitated an exponential increase in
understanding of marine turtle movements, behaviors, biology,
and conservation concerns (see [24] for review). Therefore,
integration of multiple tools and techniques, including site-based
monitoring (e.g. nesting beaches, foraging areas), genetic
analyses, mark-recapture studies, and telemetry, especially when
supplemented with information on threats and influence of
environmental conditions [25], can facilitate robust definitions of
MUs for marine turtles at multiple scales to address different
management and research challenges.
In response to these issues, we compiled, collated, and
georeferenced available information on marine turtle biogeogra-
phy – including individual nesting sites, genetic stocks, and
geographic distributions based on monitoring research – to
develop multi-scale Regional Management Units (RMUs). These
RMUs spatially integrate sufficient information to account for
complexities in marine turtle population structures, and thus
provide a flexible, dynamic framework for evaluating threats,
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org2December 2010 | Volume 5 | Issue 12 | e15465

Page 3

identifying high diversity areas, highlighting data gaps, and
assessing conservation status of marine turtles.
Methods
The IUCN’s Marine Turtle Specialist Group convened the
Burning Issues Working Group (MTSG-BI) for two meetings
(August 2008 and September 2009) of marine turtle experts from
around the world who represented government agencies, non-
governmental organizations, and academic institutions. The
primary objective of these meetings was to develop a process for
evaluating and prioritizing the conservation status of marine turtle
populations worldwide. In this context, the MTSG-BI developed
Regional Management Units (RMUs) as the framework for this
evaluation and prioritization process.
To generate RMUs for marine turtles, we collated and
georeferenced available data from more than 1,200 papers,
reports, abstracts, and other sources (available for download at
http://tinyurl.com/29w4kbf), extracting information on nesting
sites, population genetics, tag returns, and satellite telemetry, as
well as other relevant natural history and biogeography. We then
spatially integrated this information in ArcMap 9.3 (ESRI,
Redlands, CA USA) from fine (i.e. points with geographic
coordinates for nesting sites) to coarse (i.e. polygon shapefiles for
distributions) spatial scales. We constructed separate layers
according to distinct biological/spatial scales, including a layer
for nesting sites (i.e. individual nesting beaches), and two layers for
genetic stocks (i.e. a layer each for maternally inherited mtDNA
and biparentally inherited nDNA, respectively). Finally, all finer-
scale levels were nested within the known core geographic
distributions of these population units, according to satellite
telemetry and tag return data. In this way, all nesting sites that
were sampled and analyzed for genetic studies were represented
within genetic stock layers, and defined genetic stocks were
represented within RMU layers, such that information shared
across scales was retained in each relevant layer. We also compiled
available information on population sizes (i.e. annual numbers of
nesting females) and trends (i.e. change in annual numbers of
nesting females over time) at each spatial/biological scale.
Although considerable uncertainty exists in estimates of population
sizes and trends for marine turtles among sites globally, we did not
attempt to standardize or improve estimates; we urge caution in
interpretation of these metadata. This ‘‘nested envelope’’ ap-
proach allowed for metadata to be arranged within and across
biologically defined spatial scales.
Nesting Sites
We defined a nesting site as a beach or beaches with confirmed
marine turtle nesting activity monitored or analyzed as a singular
site by the groups or individuals providing or publishing the
nesting data. To compile and georeference nesting sites globally
for all species, we used the State of the World’s Sea Turtles –
SWOT (www.seaturtlestatus.org) database, which relies on a
global network of researchers who voluntarily contribute annual
nesting data. We augmented the SWOT database with published
information. We filtered all nesting sites to distinguish sites with
confirmed, quantified nesting activity (i.e. counts) from those
without quantified counts since 2000. Across species globally, we
compiled more than 4,200 nesting sites, ranging from 30 for
Kemp’s ridleys (Lepidochelys kempii) to 1,337 for hawksbill turtles
(Eretmochelys imbricata) (Table 1; Figs. 1, 2, 3, 4, 5, 6, 7, panel A). In
accordance with data sharing protocols, metadata for SWOT
nesting sites are not provided here. However, all SWOT nesting
data can be viewed at http://seamap.env.duke.edu/swot, and a
complete list of SWOT data providers is provided in Appendix S1.
Genetic Stocks
We compiled and georeferenced genetic stock information from
available mtDNA and nDNA studies by constructing shapefiles
that included all individual nesting sites upon which the genetic
sampling and analyses were based. We identified 87 distinct
mtDNA and 28 nDNA reported genetic stocks globally across
marine turtle species (Table 1; Figs. 1, 2, 3, 4, 5, 6, 7, panels B and
C; Appendix S2). While recognizing the diversity of analytical
techniques – particularly the number and types of genetic markers,
sample sizes, and statistical approaches – used to determine
genetic stocks of marine turtles, we accepted genetic stock
definitions as described by the original sources, rather than
attempting to standardize or prioritize methodologies or interpre-
tations. However, in cases where more recent studies clarified or
contradicted earlier studies, we based our dataset on the results of
the updated studies.
Regional Management Units
We generated polygons representing RMUs for all species of
marine turtles, or geographically explicit population segments
based on geographic boundaries to distributions derived from
studies on genetics, tag returns, satellite telemetry, and other data
(Figs. 1, 2, 3, 4, 5, 6, 7, panel D; Table S1; Appendix S2). RMUs
were meant to encompass multiple nesting sites, mtDNA-defined
Table 1. Summary of number of nesting sites, genetic stocks, and RMUs identified for all marine turtle species.
speciesno. nesting sites no. mtDNA stocksno. nDNA stocks
no. RMUs
(no. putative RMUs)
Caretta caretta 626158 10 (1)
Chelonia mydas935339 17
Demochelys coriacea652967
Eretmochelys imbricata1,33719213 (7)
Lepidochelys kempii301*1*1
Lepidochelys olivacea426628 (1)
Natator depressus2874 ND2
TOTAL4,293872858 (9)
ND: no data. Putative RMUs were described for regions with available nesting data but no associated data on genetics or distributions. (
*denotes that although nesting sites have not been sampled for genetics, all L. kempii individuals are presumed to belong to same stock and a single RMU.)
doi:10.1371/journal.pone.0015465.t001
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org3 December 2010 | Volume 5 | Issue 12 | e15465

Page 4

Figure 2. Multi-scale Regional Management Units for green turtles Chelonia mydas. Individual maps are presented for A) global nesting
sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic stocks; C) nuclear
genetic (nDNA) stocks; and D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have no count values
reported since 2000 are represented by black squares, whereas nesting sites for which count data are available are represented by a colored circle.
For genetic stock maps (Figs. 2B and C), each symbol represents a different site that was sampled for genetic analyses, while each color represents a
distinct genetic stock. Data shown are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g002
Figure 1. Multi-scale Regional Management Units for loggerhead turtles Caretta caretta. Individual maps are presented for A) global
nesting sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic stocks; C)
nuclear genetic (nDNA) stocks; and D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have no count
values reported since 2000 are represented by black squares, whereas nesting sites for which count data are available are represented by a colored
circle. For genetic stock maps (Figs. 1B and C), each symbol represents a different site that was sampled for genetic analyses, while each color
represents a distinct genetic stock. Putative RMUs (see text for description) are noted by asterisks in the figure legends, and are colorless. Data shown
are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g001
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org4December 2010 | Volume 5 | Issue 12 | e15465

Page 5

Figure 4. Multi-scale Regional Management Units for hawksbill turtles Eretmochelys imbricata. Individual maps are presented for A) global
nesting sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic stocks; C)
nuclear genetic (nDNA) stocks; and D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have no count
values reported since 2000 are represented by black squares, whereas nesting sites for which count data are available are represented by a colored
circle. For genetic stock maps (Figs. 4B and C), each symbol represents a different site that was sampled for genetic analyses, while each color
represents a distinct genetic stock. Putative RMUs (see text for description) are noted by asterisks in the figure legends, and are colorless. Data shown
are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g004
Figure 3. Multi-scale Regional Management Units for leatherback turtles Dermochelys coriacea. Individual maps are presented for A)
global nesting sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic
stocks; C) nuclear genetic (nDNA) stocks; and D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have
no count values reported since 2000 are represented by black squares, whereas nesting sites for which count data are available are represented by a
colored circle. For genetic stock maps (Figs. 3B and C), each symbol represents a different site that was sampled for genetic analyses, while each color
represents a distinct genetic stock. Data shown are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g003
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org5 December 2010 | Volume 5 | Issue 12 | e15465

Page 6

Figure 5. Multi-scale Regional Management Units for olive ridley turtles Lepidochelys olivacea. Individual maps are presented for A) global
nesting sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic stocks; C)
nuclear genetic (nDNA) stocks; and D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have no count
values reported since 2000 are represented by black squares, whereas nesting sites for which count data are available are represented by a colored
circle. For genetic stock maps (Figs. 5B and C), each symbol represents a different site that was sampled for genetic analyses, while each color
represents a distinct genetic stock. Putative RMUs (see text for description) are noted by asterisks in the figure legends, and are colorless. Data shown
are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g005
Figure 6. Multi-scale Regional Management Units for Kemp’s ridley turtles Lepidochelys kempii. Individual maps are presented for A)
global nesting sites and in-water distributions (global distributions for each species represented by hatching); and D) Regional Management Units
(shown with nesting sites). Nesting sites that were unquantified or have no count values reported since 2000 are represented by black squares,
whereas nesting sites for which count data are available are represented by a colored circle. Putative RMUs (see text for description) are noted by
asterisks in the figure legends, and are colorless. Note: There are no genetics maps (Panels B or C) for L. kempii because all individuals are presumed to
belong to same stock and RMU. Data shown are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g006
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org6 December 2010 | Volume 5 | Issue 12 | e15465

Page 7

nesting populations, and nDNA-defined breeding populations to
reflect shared geographic distribution among conspecific marine
turtles in the same region. Thus, RMUs do not represent complete
geographic distributions of species on global or regional scales, but
rather distributions that are anchored to landmasses by known
nesting site(s) and/or genetic stock origins and defined by
biogeographical information.
Specifically, we defined the boundaries of RMU polygons by
generating termini directly from published satellite tracks, tag
returns, or other data sources. These polygons were then digitized
or imported into ArcGIS 9.3 either using the georeference tool in
the ArcMap Toolbox, or based on text descriptions. In the absence
of sufficient information, boundaries were further refined by
MTSG experts. To allow clear distinctions between RMUs and
complete global distributions, we also generated global distribution
polygons for all species (Figs. 1, 2, 3, 4, 5, 6, 7, panel A; Appendix
S2). These global distributions are coarse geographical represen-
tations of documented occurrence patterns – bounded by
maximum extents – for each species. This process was generally
similar to that used to generate species range maps for other
taxonomic groups, such as those produced by the IUCN Red List of
Threatened Species
(http://www.iucnredlist.org/technical-docu-
ments/spatial-data). We acknowledge that this distinction means
that RMU designations might not address threats occurring
outside RMU boundaries but within the distribution of a species,
but RMUs nonetheless encompass known core habitats and life
stages. However, the important point is that whereas global
distributions simply display broad geographic ranges for each
species (including RMUs), the RMUs themselves provide refined
spatial guidance for conservation strategies.
Olive ridley turtles (Lepidochelys olivacea) (along with their
congener, Kemp’s ridleys), are unique among marine turtles for
their polymorphic nesting behavior; i.e. synchronous mass nesting
at particular beaches (termed arribadas) and disperse, asynchronous
or solitary nesting at the rest of the species’ nesting beaches.
Therefore, we established separate RMUs for arribadas and solitary
nesters in regions where both behaviors occur (i.e. East Pacific
Ocean, East Indian Ocean). Although the geographic boundaries
of these RMUs are identical within a region, this dichotomy
accurately reflects differences in population abundances and
trends between the two behaviors [26].
In regions where nesting sites were known for certain species,
but no other biological information (e.g. genetics or distributions)
was available, we developed ‘putative RMUs’ so that no region-
species combination was excluded. As with all RMUs, these
putative RMUs will require modification as new information
becomes available, but in the meantime, they represent obvious
research and reporting priorities.
Results and Discussion
We identified 58 RMUs among the seven marine turtle species
worldwide, ranging from a single RMU for Kemp’s ridleys to 17
RMUs for green turtles (Chelonia mydas) (Figs. 1, 2, 3, 4, 5, 6, 7;
Table S1, Appendix S2). The RMU framework is essentially a set
of nested envelope models intended for multiple research and
management applications. However, the efficacy of applications
using RMUs depends on the accuracy and quality of the data
contained in the files; data exist that we were unable to acquire
and incorporate, and current designations of genetic stocks and
geographic distributions are subject to change with new and
improved analyses. For example, genetic analyses using new
markers, larger sample sizes, and broader geographic sampling
could reveal new or more nuanced stock structures, especially
because insufficient sampling is always a major detractor to
explanatory power of these analyses [20,27,28]. Furthermore,
because RMU boundaries are sometimes based on reports that
contain relatively few localizations, some of which might be species
Figure 7. Multi-scale Regional Management Units for flatback turtles Natator depressus. Individual maps are presented for A) global
nesting sites and in-water distributions (global distributions for each species represented by hatching); B) mitochondrial (mtDNA) genetic stocks; and
D) Regional Management Units (shown with nesting sites). Nesting sites that were unquantified or have no count values reported since 2000 are
represented by black squares, whereas nesting sites for which count data are available are represented by a colored circle. For genetic stock maps
(Fig. 7B), each symbol represents a different site that was sampled for genetic analyses, while each color represents a distinct genetic stock. Note:
There is no map for N. depressus nDNA stocks (panel C, due to lack of available data). Data shown are contained in metadata tables (Appendix S2).
doi:10.1371/journal.pone.0015465.g007
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org7 December 2010 | Volume 5 | Issue 12 | e15465

Page 8

misidentifications [29], improved data quality will help to resolve
the geographic extents of RMUs.
Along these lines, because the metadata we compiled is derived
from publicly available sources, the RMUs themselves will change
with new, refined data. To facilitate iterative improvement of
RMUs, we have made all map files (Figs. 1, 2, 3, 4, 5, 6, 7) and
metadata associated with each layer (Appendix S2) available for
comments and suggested edits, as well as download and analysis (in
accordance with relevant data sharing protocols) by users within
the OBIS-SEAMAP framework (Ocean Biogeographic Informa-
tion System-Spatial Ecological Analysis of Megavertebrate Popu-
lations [30]; http://seamap.env.duke.edu/swot). Thus, the appli-
cability of RMUs to marine turtle conservation and research will
be user-dependent, relying on collaboration among users in the
community to maintain up-to-date, accurate files.
An important caveat inherent in the RMUs is that we are
largely unable to differentiate between true absences (i.e. a species
is not identified or is no longer present in an area despite thorough
search effort) and absence due to lack of monitoring or reporting.
Clearly, distributions of nesting sites and other types of
information are biased by areas of high monitoring effort and
reporting (e.g. Wider Caribbean region; [31]). In this vein, the
maps and RMUs generally could be used in gap analyses to
identify areas toward which enhanced census efforts should be
directed in order to improve inter-regional comparisons of marine
turtle distribution patterns. For example, the distribution of
putative RMUs illustrates gaps in scientific understanding of
marine turtle biogeography in much of the Indian Ocean, and
biogeography of hawksbill turtles in particular (Figs. 1, 2, 3, 4, 5, 6,
7).
In addition to identifying data gaps, a major advantage of the
RMU approach is the potential to connect impacts of particular
threats to biologically relevant units and their associated
demographic characteristics. For example, because marine turtle
populations occur in terrestrial and marine habitats, the RMU tool
provides the ability to overlay the geographic extent of not only
nesting beach threats along a particular coastline (e.g. coastal
development) to the particular marine turtle nesting sites and
stocks that would be impacted, but also threats across a wider
ocean area (e.g. fisheries bycatch) that could impact several nesting
stocks and broader population units simultaneously [12]. Further-
more, this system allows identification of spatial overlaps among
RMUs – within and among species – which are important areas
for conservation because threats impacting multiple RMUs might
warrant different management attention than threats acting on a
single RMU. Ultimately, although RMUs will not be equally
valuable to conservation efforts in all regions, this framework is
intended for setting conservation priorities at different levels of
spatial and biological organization for marine turtles on a global
scale. Below, we outline other potential applications of RMUs to
marine turtle conservation and research.
Potential applications of RMUs
Identification
Characterization of heavily-utilized areas for marine turtles at
different spatial scales is fundamental for creating effective
conservationstrategies[12,32–34].
application of RMUs is to identifying important geographic
areas for marine turtle populations in terms of determination of
presence, density, and richness. For example, geographic regions
that host high densities of nesting sites, possibly of multiple species
and/or genetic stocks, could merit investment of conservation
resources. Demographic information included in the RMU
essentialhabitats for marineturtles.
Onestraightforward
metadata could also be incorporated in these evaluations to
account for abundance and trends among sites and regions.
Likewise, areas of overlap among genetic stocks or geographic
distributions, as well as regional variation in population trends at
various spatial or biological scales could also be identified using
RMUs. Although reproduction areas are relatively discrete
geographically and genetically, foraging areas for marine turtles
of all life stages host con-specific individuals from multiple stocks
and geographic locations [2,10,28,35–38]. Therefore, character-
izing the connectivity among multiple nesting sites and multiple
foraging areas – i.e. ‘many-to-many’ relationships [10] – is
necessary for holistically assessing demographic trends and
conservation effectiveness [38,39].
With this in mind, one next step for refining RMUs would be to
expand on the genetic stocks based on nesting sites by spatially
characterizing at-sea mixed-stock foraging or developmental areas
[9,10,28,40,41]. Moreover, weighting distribution layers according
to a relative measure of proportional habitat use (e.g. kernel
densities, home ranges) – incorporating both spatial and temporal
information – to distinguish among high-use areas and fringe
habitats would be another useful extension of the RMU
framework. Although incorporating and mapping this information
would be challenging given data currently available in published
research studies, this step would dramatically improve the ability
of RMUs to facilitate identification of marine turtle habitats in
relation to other georeferenced information of threats or
environmental factors.
Improveddefinition of genetic
Despite the emphasis on genetic resolution of MUs, reliance on
population structure derived from such analyses can be inadequate
for management [4,42]. Failure to detect population units with
genetic markers does not necessarily mean that no management-
relevant structuring exists. For example, a newly colonized nesting
beach might host turtles that are genetically indistinguishable from
the parent (source) population, yet the new nesting cohort is
geographically isolated from the parent population. These
limitations highlight the need for alternative types of information
to make informed MU designations [20].
Along these lines, several studies have proposed or identified
geographic or environmental barriers to migration of marine
turtles among nesting sites or foraging areas that appear to result
in significant population (but not necessarily genetic) differentia-
tion. The distance between nesting sites that results in isolated
nesting stocks appears to vary within [2,43] and among species
[13,44]. Based on identification of 17 distinct green turtle genetic
stocks among 27 nesting sites sampled in Australasia, Dethmers et
al. [13] proposed that nesting sites separated by more than 500 km
are likely to host isolated nesting populations; in contrast,
undifferentiated leatherback (Dermochelys coriacea) genetic stocks
span thousands of kilometers of coastlines within geographic
regions [44]. Furthermore, recent studies have described spatial
distributions of genetic stocks in foraging and developmental areas
at-sea based on local and regional ocean current patterns in the
Wider Caribbean [28,37], Atlantic Ocean [40], Mediterranean
Sea [45], and Pacific Ocean [46,47].
Although barriers to gene flow have not been defined in most
cases, these studies illustrate geographic and environmental
influences on marine turtle population structures that can be
tested within the RMU framework, because nesting sites as well as
known genetic stocks are georeferenced. Thus, by applying
oceanographic or other physical information as well as relatively
simple distance buffering analyses to the RMU files, researchers
can test hypotheses about spatial distributions of genetic stocks
within and across geographic regions. Resulting distributions of
stockdistributions.
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org8December 2010 | Volume 5 | Issue 12 | e15465

Page 9

population segments would allow more flexibility in defining MUs
than those derived solely from quantitative analyses of genetic
differentiation, and would also provide clear objectives for further
analyses.
Conservation status assessments.
population status, including population sizes and trends, as well
as threats and relevant biological information, is a prerequisite
for prioritizing conservation resources. However, because
defining the relevant biological or geographic scale at which to
conduct assessments and make recommendations is always a
fundamental question in these processes, RMUs are a valuable
resource to guide how populations of marine turtle species are
assessed.
For example, the IUCN Red List of Threatened Species represents the
only globally accepted system for evaluating extinction risk for
species, but the IUCN/SSC Marine Turtle Specialist Group
(MTSG) has debated the utility and validity of a global
classification system for marine turtles, and has advocated for
regional assessments (see [48] for review). A survey of MTSG
members from 23 countries revealed that nearly 90% of
respondents believed that regional assessments should use either
‘new MTSG criteria’ or ‘flexible, non-standardized criteria’ [48].
However, there was less consensus among members about the
appropriate population segment upon which to base regional
assessments, with nearly 50% of respondents stating ‘‘by
geographic region,’’ ,30% stating ‘‘by genetic stocks,’’ and
,20% stating ‘‘by nesting sites,’’ illustrating the challenges
inherent in defining management units for marine turtles.
Utilizing the RMU framework – which contains information at
each of those spatial and biological scales – within or in
conjunction with the Red List assessment process could address
some of the controversy within the MTSG regarding the
application of a single global listing for geographically variable
marine turtle species by providing regional units for assessment
[48,49].
Another example of incorporating population differences in
conservation status assessments is the listing process under the U.S.
Endangered Species Act (ESA). The ESA provides for designation
and listing of Distinct Population Segments (DPS) of a species
based on ‘discreteness’ and ‘significance’ of that segment [50,51].
In a recent Biological Review of loggerhead turtles (Caretta caretta)
for purposes of evaluating the species’ current Threatened listing
status on ESA, nine DPSs were defined, with distinct recommen-
dations for each DPS [52]. Not surprisingly, the process by which
DPSs are defined is very similar to that used to define RMUs, as
both depend on biogeographical information from genetics,
distribution, movements, and demographics. There was nearly
complete agreement between the loggerhead DPSs and our RMU
designations; the two schemes essentially differed only in how
putative RMUs are handled [52] (Table 1). This result provides
further support for the validity of the RMU framework for intra-
specific conservation assessments.
It is important to note that in geographic regions where there
are existing systems for effectively identifying marine turtle
population segments to which to target conservation efforts, such
aswell-definedbreeding populations
[13,16,18,21–23], RMUs will be of limited utility. In contrast,
RMUs will be most appropriately applied to areas hosting several
populations or stocks, possibly of multiple species, especially in
regions with relatively little available information. Thus, RMUs
are designed to be coarse yet flexible enough to be applicable
anywhere in the world, rather than being restricted to areas with
the best available information, but can be refined in the future
when new information becomes available.
Evaluation of species or
ineasternAustralia
Marine spatial planning and applications to similar
species with complex life histories.
function, especially in areas where multiple activities by multiple
nations occur, ecosystem-based marine management approaches
should be designed to preserve marine biodiversity, keystone
species, and biological connectivity among marine habitats [1,53].
Thus, detailed characterizations of distributions and natural
histories of key species (e.g. keystones, top predators, ecosystem
engineers) are instrumental to guiding exercises in marine spatial
planning [54].
RMUs provide explicit, ecologically based spatial and demo-
graphic information about geographic distributions of marine
turtle populations that could be integrated with other georefer-
enced layers, including human activities subject to management
(e.g. fisheries operations, hydrocarbon extraction, coastal devel-
opment, shipping, etc.). Furthermore, the RMU concept could be
applied easily to multi-scale biogeographical characterization of
other marine megafauna species with similar life history traits and
broad, complex distributions. As with marine turtles, identification
of high-use habitats, connectivity between breeding and feeding
areas, as well as overlapping distributions of distinct populations is
extremely important for designing appropriate management
schemes for these species [55–58].
Conclusions.
Thenovel
available biogeographical information on globally distributed
and imperiled marine turtles into a multi-scaled, geospatial tool
for which we envision numerous pertinent applications for marine
turtle conservation and research. For example, the MTSG
Burning Issues Working Group is utilizing RMUs as the basis
for status assessments in a developing process of global
conservation priority setting for marine turtles. The RMU
classification system is consistent with endangered species laws in
the United States and elsewhere, and could provide the IUCN-
MTSG with a way forward for regional evaluations of extinction
risk.
Although species are predominantly used as the currency for
evaluating and prioritizing conservation efforts (e.g. IUCN Red List,
Alliance for Zero Extinction, Conservation ‘Hotspots’), in the case
of globally distributed species like marine turtles, regional
population segments occupy distinct ecological roles and thus
merit conservation attention, because extinction of an RMU
would represent the loss of the species’ ecological role within an
entire region and ecosystems therein [59]. By defining populations
according to ecological characteristics, the RMU approach implies
the inherent importance of each RMU as an independent
conservation unit, which species-focused conservation approaches
might fail to recognize.
We emphasize that the resolution of RMUs does not detract
from the treatment of nesting populations as management units.
Abundant historical, mark-recapture, and genetic data indicate
that nesting populations will rise and fall as independent
demographic units. The value of RMUs is in the recognition of
regional groupings of marine turtles that overlap in feeding and
nursery habitats, exchange genetic material, and face common
threats at sea. Therefore, the RMU framework is a solution to the
challenge of how to organize marine turtles into units of
protection above the level of nesting populations, but below the
level of species, within regional entities that might be on
independent evolutionary trajectories. Finally, the RMU system
is neither static nor proprietary, but rather is iterative and user-
driven. We encourage broad and creative engagement with and
application of this tool, whose long-term accuracy and efficacy
will rely on updates, edits, and improvements arising from user
interactions.
To optimize ecosystem
RMUframeworksynthesizes
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org9December 2010 | Volume 5 | Issue 12 | e15465

Page 10

Supporting Information
Table S1
marine turtles worldwide, including number of nesting sites and
genetic stocks contained within each RMU.
(DOC)
Summary of Regional Management Units (RMUs) for
Appendix S1
World’s Sea Turtles data providers.
(XLS)
Complete list of SWOT – The State of the
Appendix S2
generate Regional Management Units.
(XLS)
Metadata associated with each layer synthesized to
Acknowledgments
This work resulted from meetings of the IUCN/SSC Marine Turtle
Specialists Group’s Burning Issues Working Group, which consisted of
representatives of governmental agencies, NGOs, and academia from
around the world. Special recognition goes to SWOT Team members and
partners who contributed information to the SWOT database and to this
initiative. We thank MTSG members and others who contributed valuable
comments, including GC Hays and an anonymous reviewer.
Author Contributions
Conceived and designed the experiments: BPW ADD BJH EMF ABB
MYC BJH FAAG DA KAB JB BWB RBD PC BCC AC PHD AF AG MG
MHG MH MLM MAM JAM JAM RN NJP JAS ST BW RBM.
Performed the experiments: BPW ADD BJH EMF. Analyzed the data:
BPW ADD BJH EMF. Contributed reagents/materials/analysis tools:
BPW ADD BJH EMF ABB MYC BJH FAAG DA KAB JB BWB RBD PC
BCC AC PHD AF AG MG MHG MH MLM MAM JAM JAM RN NJP
JAS ST BW RBM. Wrote the paper: BPW ADD BJH EMF ABB MYC
BJH FAAG DA KAB JB BWB RBD PC BCC AC PHD AF AG MG
MHG MH MLM MAM JAM JAM RN NJP JAS ST BW RBM.
References
1. Foley MM, Halpern BS, Micheli F, Armsby MH, Caldwell MR, et al. (2010)
Guiding ecological principles for marine spatial planning. Mar Pol 34: 955–966.
2. Bowen BW, Bass AL, Soares L, Toonen RJ (2005) Conservation implications of
complex population structure: lessons from the loggerhead turtle (Caretta caretta).
Mol Ecol 14: 2389–2402.
3. Suryan RM, Saba VS, Wallace BP, Hatch SA, Frederiksen M, et al. (2009)
Environmental forcing on life history strategies: Multi-trophic level response at
ocean basin scales. Prog Oceanogr 81: 214–218.
4. Taylor BL, Dizon AE (1999) First policy then science: why a management unit
based solely on genetic criteria cannot work. Mol Ecol 8: S11–S16.
5. Moritz C (1994) Defining ‘evolutionary significant units’ for conservation.
Trends in Ecology & Evolution 9: 373–375.
6. Ashley MV, Willson MF, Pergams ORW, O’Dowd DJ, Gende SM, et al. (2003)
Evolutionarily enlightened management. Biol Conserv 111: 1150–123.
7. IUCN (2010) IUCN Red List of Threatened Species. Version 2010.1. www.
iucnredlist.org. Downloaded on 11 April 2010.
8. Musick JA, Limpus CJ (1997) Habitat utilization and migration in juvenile sea
turtles. In: Lutz PL, Musick JA, eds. The Biology of Sea Turtles. Boca RatonFL:
CRC Press. pp 137–164.
9. Bowen BW, Karl SA (2007) Population genetics and phylogeography of sea
turtles. Mol Ecol 16: 4886–4907.
10. Bolker BM, Okuyama T, Bjorndal KA, Bolten AB (2007) Incorporating multiple
mixed stocks in mixed stock analysis: ‘many-to-many’ analyses. Mol Ecol 16:
68–695.
11. Wallace BP, Saba VS (2009) Environmental and anthropogenic impacts on
intra-specific variation in leatherback turtles: opportunities for targeted research
and conservation Endangered Species Research 7: 1–11.
12. Hamann M, Godfrey MH, Seminoff JA, Arthur K, Barata PCR, et al. (2010)
Global research priorities for sea turtles: informing management and
conservation in the 21st century. Endangered Species Research 11: 245–269.
13. Dethmers KEM, Broderick D, Moritz C, FitzSimmons NN, Limpus CJ, et al.
(2006) The genetic structure of Australasian green turtles (Chelonia mydas):
exploring the geographical scale of genetic exchange. Mol Ecol 15: 3931–3946.
14. 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.
15. Bowen BW, Bass AL, Chow S-M, Bostrom M, Bjorndal KA, et al. (2004) Natal
homing in juvenile loggerhead turtles (Caretta caretta). Mol Ecol 13: 3797–3808.
16. FitzSimmons NN, Limpus CJ, Norman JA, Goldizen AR, Miller JD, et al.
(1997a) Philopatry of male marine turtles inferred from mitochondrial DNA
markers. Proc Nat Acad Sci 94: 8912–8917.
17. Karl SA, Bowen BW, Avise JC (1992) Global population genetic structure and
male-mediated gene flow in the green turtle (Chelonia mydas): RFLP analyses of
anonymous nuclear loci. Genetics 131: 163–173.
18. FitzSimmons NN, Moritz C, Limpus CJ, Pope L, Prince R (1997b) Geographic
structure of mitochondrial and nuclear gene polymorphisms in Australian green
turtle populations and male-biased gene flow. Genetics 147: 1843–1854.
19. Carreras C, Pascual M, Cardona LM, Aguilar A, Margaritoulis D, et al. (2007)
The genetic structure of the loggerhead sea turtle (Caretta caretta) in the
Mediterranean as revealed by nuclear and mitochondrial DNA and its
conservation implications. Cons Gen 8: 761–775.
20. Martien KK, Taylor BT (2003) Limitations of hypothesis-testing in defining
management units for continuously distributed species. J Cet Res Manag 5:
213–218.
21. Limpus CJ, Miller JD, Parmenter CJ, Reimer D, McLachlan N, et al. (1992)
Migration of green (Chelonia mydas) and loggerhead (Caretta caretta) turtles to and
from eastern Australian rookeries. Wildlife Res 19: 347–358.
22. Limpus CJ, Miller JD, Parmenter CJ, Limpus DJ (2003) The green turtle,
Chelonia mydas, population of Raine Island and the Northern Great Barrier Reef:
1843-2001. Memoirs Queensland Museum 49: 349–440.
23. Limpus CJ (2008) A biological review of Australian Marine turtles: Green turtle
Chelonia mydas (Linnaeus), Queensland Environmental Protection Agency, 96.
24. Godley BJ, Blumenthal JM, Broderick AC, Coyne MS, Godfrey MH, et al.
(2008) Satellite tracking of sea turtles: Where have we been and where do we go
next? Endangered Species Research 4: 3–22.
25. Godley BJ, Barbosa C, Bruford M, Broderick AC, Catry P, et al. (2010)
Unravelling migratory connectivity in marine turtles using multiple methods.
J Appl Ecol;doi: 10.1111/j.1365-2664.2010.01817.x.
26. Abreu-Grobois FA, Plotkin PT (2007) IUCN Red List Status Assessment of the
olive ridley sea turtle (Lepidochelys olivacea). IUCN/SSC-Marine Turtle Specialist
Group, 39.
27. Formia A, Broderick AC, Glen F, Godley BJ, Hays GC, 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.
28. Blumenthal JM, Abreu-Grobois FA, Austin TJ, Broderick AC, Bruford MW,
et al. (2009) Turtle groups or turtle soup: dispersal patterns of hawksbill turtles in
the Caribbean. Mol Ecol 18: 4841–4853.
29. Frazier J (1985) Misidentifications of sea turtles in the East Pacific: Caretta caretta
and Lepidochelys olivacea. J Herp 19: 1–11.
30. Halpin PN, Read AJ, Fujioka E, Best BD, Donnelly B, et al. (2009) OBIS-
SEAMAP The World Data Center for Marine Mammal, Sea Bird, and Sea
Turtle Distributions. Oceanography 22: 104–115.
31. Dow W, Eckert KL, Palmer M, Kramer P (2007) An Atlas of Sea Turtle Nesting
Habitat for the Wider Caribbean Region. The Wider Caribbean Sea Turtle
Conservation Network and The Nature Conservancy. WIDECAST Technical
Report No. 6, Beaufort, NC, USA, 267.
32. Dryden J, Grech A, Moloney J, Hamann M (2008) Re-zoning of the Great
Barrier Reef World Heritage Area: does it afford greater protection for marine
turtles? Wildlife Res 35: 477–485.
33. Witt MJ, Broderick AC, Coyne MS, Formia A, Ngouessono S, et al. (2008)
Satellite tracking highlights difficulties in the design of effective protected areas
for Critically Endangered leatherback turtles Dermochelys coriacea during the inter-
nesting period. Oryx 42: 296–300.
34. Shillinger GL, Palacios DM, Bailey H, Bograd SJ, Swithenbank AM, et al.
(2008) Persistent leatherback turtle migrations present opportunities for
conservation. PLoS Biol 6: e171.
35. Bolten AB, Bjorndal KA, Martins HR, Dellinger T, Biscoito MJ, et al. (1998)
Transatlantic development migrations of loggerhead sea turtles demonstrated by
mtDNA sequence analysis. Ecol App 8: 1–7.
36. Lahanas PN, Bjorndal K, Bolten A, Encalada SE, Miyamoto MM, et al. (1998)
Genetic composition of a green turtle (Chelonia mydas) feeding ground population:
evidence for multiple origins. Mar Biol 130: 345–352.
37. 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. J Hered 97: 346–354.
38. Bjorndal K, Bolten A (2008) Annual variation in source contributions to a mixed
stock: implications for quantifying connectivity. Mol Ecol 17: 2185–2193.
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org 10December 2010 | Volume 5 | Issue 12 | e15465

Page 11

39. Bjorndal KA, Bolten AB, Troeng S (2005) Population structure and genetic
diversity in green turtles nesting at Tortuguero, Costa Rica, based on
mitochondrial DNA control region sequences. Mar Biol 147: 1449–1457.
40. Monzon-Arguello C, Lopez-Jurado L, Rico C, Marco A, Lopez P, Hays G, Lee P
(2010) Evidence from gnetic and Lagrangian drifter data for transatlantic
transport of small juvenile green turtles. J Biogeogr 37: 1752–1766.
41. Bass AL, Epperly S, Braun-McNeill J (2004) Multi-year analysis of stock
composition of a loggerhead turtle (Caretta caretta) foraging habitat using
maximum likelihood and Bayesian methods. Cons Gen 5: 783–796.
42. Waples RS, Punt AE, Cope JM (2008) Integrating genetic data into
management of marine resources: how can we do better? Fish and Fisheries
9: 423–449.
43. Lahanas PN, Miyamoto MM, Bjorndal KA, Bolten AB (1994) Molecular
evolution and population genetics of Greater Caribbean green turtles (Chelonia
mydas) as inferred from mitochondrial DNA control region sequences. Genetica
94: 57–67.
44. Dutton PH, Bowen BW, Owens DW, Barragan A, Davis SK (1999) Global
phylogeography of the leatherback turtle (Dermochelys coriacea). J Zool (Lond) 248:
397–409.
45. Carreras C, Pont S, Maffucci F, Pascual M, Barcelo ´ A, et al. (2006) Genetic
structuring of immature loggerhead sea turtles (Caretta caretta) in the
Mediterranean Sea reflects water circulation patterns. Mar Biol 149: 1269–1279.
46. Bowen BW, Abreu-Grobois FA, Balazs GH, Kamezaki N, Limpus CJ, et al.
(1995) Trans-Pacific migrations of the loggerhead turtle (Caretta caretta)
demonstrated with mitochondrial DNA markers. Proc Nat Acad Sci 92:
3371–3734.
47. Boyle MC, Fitzsimmons NN, Limpus CJ, Kelez S, Velez-Zuazo X, et al. (2009)
Evidence for transoceanic migrations by loggerhead turtles in the southern
Pacific Ocean. Proc Roy Soc B 276: 1993–1999.
48. Seminoff JA, Shanker K (2008) Marine turtles and IUCN Red Listing: A review
of the process, the pitfalls, and novel assessment approaches. J Exp Mar Biol
Ecol 356: 52–68.
49. Mrosovsky N (2003) Predicting Extinction: Fundamental Flaws In IUCN’s Red List
System, Exemplified By The Case Of Sea Turtles. http://members.seaturtle.org/
mrosovsky/.
50. Waples RS (1995) Evolutionary significant units and the conservation of
biological diversity under the Endangered Species Act. Evolution and the Aquatic
Ecosystem: Defining Unique Units in Population Conservation (ed. JL Nielsen), pp. 8-27.
American Fisheries Society Symposium 17, American Fisheries Society,
Bethesda, MD.
51. U.S. Fish and Wildlife Service and National Marine Fisheries Service (1996)
Policy regarding the recognition of distinct vertebrate population segments
under the Endangered Species Act. Federal Register 61(26): 4722–4725.
February 7, 1996.
52. Conant TA, Dutton PH, Eguchi T, Epperly SP, Fahy CC, et al. (2009)
Loggerhead sea turtle (Caretta caretta) 2009 status review under the U.S.
Endangered Species Act. Report of the Loggerhead Biological Review Team to
the National Marine Fisheries Service, August 2009. 222 p.
53. Bowen BW (1999) Preserving genes, ecosystems, or species? Healing the
fractured foundations of conservation policy. Mol Ecol 8: S5–S10.
54. Crowder LB, Norse E (2008) Essential ecological insights for marine ecosystem-
based management and marine spatial planning. Mar Pol 32: 772–778.
55. Block BA, Teo SLH, Walli A, Boustany A, Stokesbury MJW, et al. (2005)
Electronic tagging and population structure of Atlantic bluefin tuna. Nature 434:
1121–1127.
56. Foote AD, Simila ¨ T, Vı ´kingsson GA, Stevick PT (2009) Movement, site fidelity
and connectivity in a top marine predator, the killer whale. Evol Ecol;DOI:
10.1007/s10682-009-9337-x.
57. Yorio P (2009) Marine protected areas, spatial scales, and governance:
implications for the conservation of breeding seabirds. Cons Lett 2: 171–178.
58. Jorgensen SJ, Reeb CA, Chapple TK, Anderson S, Perle C, et al. (2010)
Philopatry and migration of Pacific white sharks. Proc of the Roy Soc B 277:
679–688.
59. Bjorndal KA, Jackson J (2003) Roles of sea turtles in marine ecosystems:
reconstructing the past. In: Lutz PL, Musick JA, Wyneken J, eds. The Biology of
Sea Turtles. Boca RatonFL: CRC Press 2: 259–274.
Regional Management Units for Marine Turtles
PLoS ONE | www.plosone.org 11December 2010 | Volume 5 | Issue 12 | e15465