Content uploaded by Stefania Gaspari
Author content
All content in this area was uploaded by Stefania Gaspari on Jun 28, 2022
Content may be subject to copyright.
Content uploaded by Stefania Gaspari
Author content
All content in this area was uploaded by Stefania Gaspari on Mar 30, 2018
Content may be subject to copyright.
Population genetic structure of common bottlenose dolphins
(Tursiops truncatus) in the Adriatic Sea and contiguous regions:
implications for international conservation
STEFANIA GASPARI
a,
*, DRAŠKO HOLCER
b,c
, PETER MACKELWORTH
c
, CATERINA FORTUNA
d
,
ALEXANDROS FRANTZIS
e
, TILEN GENOV
f
, MORGANA VIGHI
a
, CHIARA NATALI
a
, NIKOLINA RAKO
c
,
ELISA BANCHI
a
, GUIDO CHELAZZI
a
and CLAUDIO CIOFI
a
a
Department of Biology, University of Florence, Florence, Italy
b
Department of Zoology, Croatian Natural History Museum, Zagreb, Croatia
c
Blue World Institute of Marine Research and Conservation, Veli Lošinj, Croatia
d
Italian National Institute for Environmental Protection and Research, Rome, Italy
e
Pelagos Cetacean Research Institute, Vouliagmeni, Greece
f
Morigenos - Slovenian Marine Mammal Society, Piran, Slovenia
ABSTRACT
1. Habitat diversity plays a significant role in shaping the genetic structure of cetacean populations. However, the
processes involved in defining the genetic differentiation of these highly mobile marine mammals are still largely
unknown.
2. Levels of genetic differentiation and dispersal patterns of common bottlenose dolphins (Tursiops truncatus)
were assessed in the north-eastern Mediterranean Sea, with a focus on the Adriatic Sea. This is a region
characterized by diverse marine ecosystems and high levels of human-induced habitat degradation.
3. Although this species seems almost uniformly distributed throughout the Adriatic Basin, genetic evidence
rejected the hypothesis of a single stock. Pairwise estimates of genetic differentiation at 12 microsatellite loci,
and mitochondrial DNA (entire control region, 920bp), revealed diverse levels of genetic differentiation among
five putative populations from the Tyrrhenian Sea to the Aegean Sea.
4. A fine-scale genetic structure was recorded within the Adriatic Sea, where females appear to be the principal
gene flow mediators. The assessment of recent migration rates indicates a relatively high level of gene flow from the
North Adriatic towards adjacent areas.
5. Indication of a fine-scale population structure across the Adriatic Sea is a factor to be carefully considered in
the emerging marine management scenario set by the implementation of the EU Marine Strategy Framework
Directive (2008/56/CE), particularly when it comes to assessing and managing direct mortality caused by human
activities (e.g. fisheries or maritime traffic). A good knowledge of population structure at the basin level is also
fundamental for the identification of potential Adriatic Special Areas of Conservation for the bottlenose dolphin
under the Habitats Directive (Council Directive 92/43/EEC).
Copyright #2013 John Wiley & Sons, Ltd.
Received 22 April 2013; Revised 29 August 2013; Accepted 20 September 2013
*Correspondence to: S. Gaspari, Department of Biology, University of Florence, Via Madonna del Piano 6, 50019 Sesto Fiorentino (FI), Italy. Email:
stefania.gaspari@unifi.it, stefaniagaspari@gmail.com
Copyright #2013 John Wiley & Sons, Ltd.
AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS
Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
Published online in Wiley Online Library
(wileyonlinelibrary.com). DOI: 10.1002/aqc.2415
KEY WORDS: cetacean conservation; Tursiops truncatus; Mediterranean Sea; Adriatic Sea; population structure; gene flow
INTRODUCTION
Several cetacean species have greater population
structure than would be expected over relatively
small geographic scales. This is predominantly due
to their demographic history, habitat association
and foraging behaviour (Hoelzel et al., 2002).
Furthermore, dispersal may be limited by
oceanographic processes such as salinity and
temperature gradients (Fullard et al., 2000;
Jørgensen et al., 2005; Natoli et al., 2005). The
common bottlenose dolphin (Tursiops truncatus)
shows strong genetic structure among populations
across its worldwide range (Hoelzel et al., 1998;
Natoli et al., 2004). This pattern is not always
associated to geographical distance and appears to
be highly dependent on the type of environment
that the constituent individuals inhabit. The
identification of genetic discontinuities is critical
when evaluating processes affecting the distribution
of genetic variation both within and among
populations. This is particularly important for
marine species, such as Tursiops truncatus,thatare
often discretely distributed, but can nevertheless be
genetically structured.
One previous population genetic study on
common bottlenose dolphins was conducted on a
broad geographic scale from the eastern North
Atlantic to the Black Sea. It reported a general
pattern of genetic divergence between the east and
west Mediterranean basins (Natoli et al., 2005).
Common bottlenose dolphins have been studied in
only relatively limited parts of the Mediterranean,
therefore wide areas remain largely unexplored,
particularly in the eastern basin (see Bearzi et al.,
2009 for a review). Moreover, the distinction
between inshore and offshore populations has yet
to be defined, especially in the Adriatic Sea, which
is characterized by very diverse marine coastal
habitats within a relatively small geographic area
(Cushman-Roisin et al., 2010).
The common bottlenose dolphin, living mainly in
coastal areas, faces numerous human-induced
threats and has therefore attracted the attention of
both national and international conservation
authorities (Bearzi et al., 2009). The European
Habitats Directive (Council Directive 92/43/EEC)
lists Tursiops truncatus in Annex II and requires
European Union (EU) member states to establish
Special Areas of Conservation where populations
are resident. This species is also listed in Annex
IV, which demands national authorities to
monitor the status of extant populations and
human-induced mortality. Moreover, the Marine
Strategy Framework Directive (MSFD, Directive
2008/56/EC) requires all EU member states to
assess and monitor the status of the marine
environment in relation to the main pressures
caused by human activities in order to achieve a
‘Good Environmental Status’by 2020. Core
aspects of this framework are the development of
monitoring and management programmes at the
regional level. This includes the assessment of the
status of cetacean species and analysis of
population genetic structure (MSFD indicator
1.3.2) that may help clarify migratory routes and
identify distinct units for conservation (European
Commission, 2011). This is of particular importance
in the Adriatic Sea, a semi-enclosed basin that
represents one of the four Mediterranean MSFD
sub-regions and is affected, particularly on the
north-west coast, by human encroachment and
water pollution.
This study evaluated whether common bottlenose
dolphins living in the Adriatic Sea represent a
genetically distinct unit and/or show patterns of
population genetic structure. Dispersal routes were
also assessed across the Adriatic, Ionian and
Aegean Seas, and the non-contiguous area of the
Tyrrhenian Sea.
MATERIALS AND METHODS
Study area
This study was conducted in the Adriatic, Ionian,
Aegean and Tyrrhenian Seas (Figure 1). The Adriatic
Sea shows clear differences in coastal and submarine
topography along its longitudinal and transverse
axes, and it is divided into three sub-basins
(Artegiani et al., 1997; Cushman-Roisin et al., 2010).
S. GASPARI ET AL.
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
The northern section is shallow (average depth of
35 m) and extends south-eastwards to the 100 m
bathymetric contour. The central Adriatic is a
transition sub-basin with an average depth of 140 m
and extends as far as the 170 m deep Pelagosa sill.
The southern sub-basin is characterized by a wide
depression, more than 1200 m deep and is divided
from the Ionian Sea by the Otranto sill. The west
coast of the Adriatic Sea is fairly linear, sandy
with gentle slopes, while the east coast is irregular,
has many islands and a rocky, steeply sloping
bathymetry. Despite the north-west–south-east
topographic division, Artegiani et al. (1997) and
Cushman-Roisin et al. (2010) considered the North
Adriatic sub-basin dynamically independent from
the Central and South Adriatic sections. Therefore,
samples were collected from the central and
southern sections of the basin and pooled as the
Central-South Adriatic sampling area.
Sample collection and genetic analysis
Tissue samples from 89 adult common bottlenose
dolphins were collected between 1992 and 2009
from stranded animals (N= 69) and free-ranging
specimens (N= 20) in the Adriatic Sea (43 stranded
and 20 free-ranging), and in the Ionian (N=6
stranded), Aegean (N= 6 stranded) and Tyrrhenian
(N= 14 stranded) Seas. Biopsy samples from
free-ranging animals were collected from an average
of one individual per social group encountered
during surveys. Samples from stranded dolphins
were collected opportunistically from isolated
individuals at different times of the year.
DNA was extracted with phenol/chloroform and
ethanol precipitation from tissue samples preserved
in salt-saturated 20% DMSO. Samples were
genotyped at 12 microsatellite loci, including
EV37Mn and EV14Pm (Valsecchi and Amos,
1996), TtruGT6 (Caldwell et al., 2002), D08
(Shinohara et al., 1997) and Ttr04, Ttr11, Ttr19,
Ttr34, Ttr58, Ttr63, TtrRH1 and TtrRC12 (Rosel
et al., 2005). To ensure accuracy in genotyping and
to standardize allele sizing for each locus, about
30% of the samples were re-amplified as controls.
Microsatellite genotypes were screened for duplicate
sampling using the Excel Microsatellite Toolkit
3.1.1 (Park, 2001) and tested for scoring errors due
to allelic dropout, null alleles and stuttering with
Micro-Checker 2.2.3 (Van Oosterhout et al., 2004).
Aegean
Ionian
Tyrrhenian
Adriatic
NORTH AFRICA
EUROPE
1234567891011121314151617181920212223242526272829
1212 8
11121 1 3 1
2
12 3 2 1
5321 111
12 111
11111
North Adriatic
Tyrrhenian
Ionian
Aegean
Central-South Adriatic
Haplotype
Figure 1. Map of the study area. Samples were obtained from stranded and free-ranging bottlenose dolphins in the Tyrrhenian, Adriatic, Ionian and
Aegean Seas (dashed circles). Dotted line in the Adriatic Sea shows approximate limit between north and central-south basins. Number of haplotypes
is reported for each sampling area.
POPULATION STRUCTURE OF TURSIOPS TRUNCATUS IN THE EAST MEDITERRANEAN
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
The mitochondrial DNA (mtDNA) entire
control region was amplified and sequenced
using the light-strand primer TURCRL5483
(5′- GGTCTTGTAAACCGGAAAAGG - 3′)
and the heavy-strand primer TURCRH6379
(5′- GCAGACTTACACATGCAAGCA - 3′)
designed specifically on the threonine tRNA and
12S rRNA genes, respectively. Primer numbers refer
to the 3′base of the published T.truncatus
mitochondrial genome sequence (Xiong et al., 2009).
Microsatellite alleles and cycle sequencing
reactions were resolved on an Applied Biosystems
3100 genetic analyser. Raw sequence chromatographs
from both strands were edited and aligned using
CodonCode Aligner 3.0.1 (CodonCode Corporation).
The consensus sequence consisted of 920 nucleotides
corresponding to the entire common bottlenose
dolphin mtDNA control region sequence.
Analysis of genetic variation and population structure
Allele diversity, observed heterozygosity, and
unbiased gene diversity were assessed using
GenAlEx 6.5 (Peakall and Smouse, 2012), and
tested for departure from Hardy–Weinberg
equilibrium (HWE) using the Markov chain
randomization implemented in Genepop 4.1
(Rousset, 2008). Allelic richness was calculated to
account for variation in sample size using the
rarefaction method implemented in Fstat 2.9.3.2
(Goudet, 1995). Haplotype and nucleotide
diversity for mtDNA sequences and population
comparisons based on F-statistics for mtDNA
and microsatellite loci were calculated using
Arlequin 3.5 (Excoffier and Lischer, 2010).
Genetic differentiation among populations was also
evaluated using principal component analysis
(PCA) on multilocus genotypes implemented in
GenAlEx 6.5. This analysis transforms a number
of correlated variables (the alleles) into a smaller
number of uncorrelated variables (the principal
components) to best represent the original
relationships among sampling sites and alleles. A
Mantel test for matrix correspondence and a spatial
autocorrelation analysis were also performed using
GenAlEx 6.5 to compare patterns of genetic
variability as a function of geographic distances in
the Adriatic Sea. Pairwise genetic distance matrix
was calculated following Peakall et al. (1995) and
Smouse and Peakall (1999). Pairwise geographic
distances were calculated based on geographic
coordinates of individual sampling locations. The
spatial autocorrelation coefficient of genetic
distance (r) was calculated for 10 geographic
distance classes of 20 km and 95% confidence
intervals were generated based on 1000 bootstrap
replicates. Significance of Mantel test was based on
1000 permutations. The Bayesian clustering method
of Structure 2.3.3 (Pritchard et al., 2000; Falush
et al., 2003) was implemented to estimate patterns
of genetic structure. Markov Chain Monte Carlo
(MCMC) runs were conducted without prior
population information to assess the most
appropriate number (K) of populations needed for
interpreting the observed genotypes. The analysis
was conducted for Kvalues ranging from one to
seven (number of sampling areas plus two) using a
burn-in period of 100 000 iterations followed by
runs of 10
6
steps. Given the contiguous geographic
range and the probability of gene flow, the
admixture and the correlated frequency models
were chosen. The LOCPRIOR option (Hubisz
et al., 2009) was used to include sampling locations
information and assist clustering at lower levels of
divergence or with fewer data, as was the case with
the Ionian and Aegean sample set. The mean
likelihood L(K) was calculated over 10 runs for each
K. The mean difference was assessed between
successive likelihood values of K,L′(K) and the
absolute value of the difference between successive
values of L′(K), |L′′(K)|. The modal value of ΔK
(the mean of |L′′(K)| averaged over 10 runs
divided by the standard deviation of L(K)) was
estimated using Structure Harvester (Earl and
vonHoldt, 2012) and it was taken as the most likely
number of populations Kas described in Evanno
et al. (2005). The Kvalue was then used as prior
information to estimate the probability that an
individual belongs to a given population. The
program BayesAss (Wilson and Rannala, 2003) was
used to estimate bi-directional recent migration
rates among populations. The program implements
a Bayesian procedure using MCMC procedure
allowing for deviations from Hardy–Weinberg
equilibrium. MCMC runs of 10 × 10
6
iterations
were conducted and sampled every 1000 iterations
S. GASPARI ET AL.
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
to infer posterior probability distribution using
default Delta values. The Bayesian coalescent
approach implemented in Migrate 3.4.2 was also
applied to estimate average mutation-scaled
effective migration rates (M) between sampling
areas assuming migration-drift equilibrium (Beerli,
2006). An infinite allele microsatellite mutation
model (IAM) with constant mutation rate, an
F
ST
based measure as starting parameter and an
uniform prior distribution were used. The Bayesian
run consisted of one long chain with 5 × 10
6
parameter values sampled every 100 iterations and a
burn-in of 10 000 genealogies discarded at the
beginning of each chain.
RESULTS
Measures of genetic diversity
Common bottlenose dolphins showed a relatively
high degree of genetic diversity. Mean number
of alleles per sampling area varied from 4.9 to 9
and allelic richness was similar among areas,
ranging from 4.1 to 4.3. Expected and observed
heterozygosities were 0.764 ± 0.005SE and
0.692 ± 0.023SE, respectively (Table 1). There was
no evidence of scoring errors due to stuttering,
large allele dropout or null alleles. However,
significant departure from Hardy–Weinberg
equilibrium was recorded in the North Adriatic
and Tyrrhenian seas after applying sequential
Bonferroni correction. Deviation from equilibrium
was detected at four loci for the Adriatic and
Tyrrhenian populations, suggesting possible
population substructure. Population comparison
analyses were run removing those loci that
showed heterozygosity deficiency for one or more
populations (Supplementary material, Table S1).
However, patterns of population divergence did
not change significantly and therefore the original
set of 12 loci was maintained. Sequencing of
the mitochondrial DNA control region revealed a
total of 29 haplotypes with five to 12 haplotypes
per sample area (Table 1). Mean number of
pairwise differences was 9.8 ± 0.9SE with the
lowest value (6.7 ± 1.5SE) recorded in the
Tyrrhenian Sea and the highest (11.8 ± 2.8SE)
reported for the Aegean Sea. Haplotypic diversity
was generally high, ranging from 0.67 in the
Tyrrhenian to 1 in the Aegean Sea. Twenty-four
unique haplotypes were recorded. The Aegean Sea
was represented by five samples and each revealed
a unique haplotype. The Ionian and Tyrrhenian
Sea had 60% and 80% unique haplotypes,
respectively. The North and Central-South Adriatic
shared three haplotypes, and the most common
sequence was shared by 25 dolphins in the
Tyrrhenian and Adriatic Sea (Figure 1).
Population structure
Genetic divergence among common bottlenose
dolphins from the five sampling areas of the
Mediterranean Sea at microsatellite loci and
mtDNA is reported in Table 2. Microsatellite data
detected significant differentiation among all
putative populations except for the Aegean Sea,
which showed a low level of genetic divergence
from the Ionian and Adriatic Sea. A low but
significant F
ST
wasalsoestimatedbetweentheNorth
and Sentral-South Adriatic areas. This pattern of
genetic structuring was inpartcorroboratedby
Table 1. Microsatellite loci and mitochondrial DNA diversity measures for bottlenose dolphins from five sampling regions across the Mediterranean
basin. N, sample size; A, allele diversity; A
R
, allelic richness; H
E
, mean expected heterozygosity; H
O
, mean observed heterozygosity; k, number of
haplotypes; S, number of segregating sites; h, haplotype diversity; π, nucleotide diversity. Standard error values in parenthesis
Microsatellites Mitochondrial DNA
Sampling site NA A
R
H
E
H
O
NkShπ(× 10
-2
)
North Adriatic 39 9.0 (1.0) 4.3 (0.2) 0.77 (0.02) 0.71 (0.03)** 29 12 43 0.82 (0.02) 1.07 (0.09)
Central-South Adriatic 24 8.1 (0.9) 4.4 (0.2) 0.78 (0.02) 0.77 (0.03)* 16 8 23 0.87 (0.02) 0.95 (0.15)
Ionian 6 4.9 (0.4) 4.2 (0.3) 0.76 (0.03) 0.68 (0.05) 6 5 26 0.93 (0.05) 1.26 (0.32)
Aegean 6 4.9 (0.5) 4.3 (0.3) 0.76 (0.04) 0.63 (0.04) 5 5 24 1.00 (0.06) 1.28 (0.37)
Tyrrhenian 14 6.1 (0.5) 4.1 (0.2) 0.75 (0.03) 0.67 (0.05)** 14 5 20 0.67 (0.06) 0.73 (0.11)
*and **denote significant heterozigosity deficiency (P<0.05 and P<0.01, respectively) before Bonferroni correction.
POPULATION STRUCTURE OF TURSIOPS TRUNCATUS IN THE EAST MEDITERRANEAN
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
principal component analysis (PCA) (Figure 2). The
first three components of PCA explained 88% of the
total inertia, with components 1 and 2 explaining
44% and 25% of the variation, respectively. The
Mantel test found a weak correlation between
geographic and genetic distance in the Adriatic Sea
(Rxy = 0.009, P= 0.048). Spatial autocorrelation
analysis revealed no significant correlation. Analysis
of mtDNA sequences partially supported these
results. Significant differentiation was, in fact,
recorded between the North Adriatic and the
Aegean sampling locations and no significant
divergence was recorded between the Tyrrhenian
and the Adriatic Sea. Genetic comparison
between West and East Adriatic animals was
highly significant at nuclear DNA loci (P<0.001)
but not significant for mitochondrial DNA
sequence comparison.
Mean values of the log likelihood of the data
estimated using the Bayesian clustering approach
did not provide a Kvalue with a significantly high
posterior probability. Conversely, using the statistic
ΔKbased on the rate of change in the log
probability of the data between successive Kvalues,
a modal value of ΔK= 21.1 was found for K=5.
This value was used as prior population
information for calculating the posterior probability
of individual assignment. However, different
proportions of dolphin multilocus genotypes were
assigned to different clusters and no strong pattern
of population structure could be detected. High
rates of gene flow were observed from the North
Adriatic to the other regions (Table 3). Migration
rates among the Central-South Adriatic, Ionian,
Aegean and Tyrrhenian seas appeared to be
negligible. Comparison between the West and East
Adriatic coastal areas revealed a pattern of gene
flow from west to east. For the West Adriatic, the
proportions of resident individuals and migrants
from the East Adriatic were 0.970 and 0.029,
respectively. For the East Adriatic, the proportions
of residents and migrants from the West Adriatic
were 0.680 and 0.320, respectively. The coalescent
approach recovered relatively high migration rates
from the Adriatic to the Ionian Sea and between the
Ionian and the Aegean Sea (range of posterior
distribution values of M: 93.1–133.6). Minor rates
of gene flow were instead recorded among other
possible migration routes (Mvalues from 2.9 to 55.1).
DISCUSSION
Results from this study revealed different degrees of
genetic differentiation across the study area. An
overall pattern of population structuring was
detected by principal component analysis and, to a
minor extent, by a Bayesian clustering approach.
Pairwise comparisons based on F-statistics showed
Table 2. Genetic differentiation based on F-statistics for the bottlenose dolphin. F
ST
values are reported on the left triangulation matrix for
microsatellite loci (below diagonal) and mtDNA sequences (above diagonal)
Sampling site North Adriatic Central-South Adriatic Ionian Aegean Tyrrhenian
North Adriatic ―–0.003 0.131(*) 0.112(*) 0.021
Central-South Adriatic 0.009(*) ―0.080(*) 0.073 0.057
Ionian 0.026(*) 0.025(*) ―0.034 0.221(**)
Aegean 0.010 0.011 0.015 ―0.200(*)
Tyrrhenian 0.051(**) 0.033(**) 0.042(*) 0.043(*) ―
Significance of fixation indices at 5% (*) and 1% (**) levels was tested by 10 000 permutations.
Aegean
Ionian
North Adriatic Central-South Adriatic
Tyrrhenian
PC1
PC2
Figure 2. Principal component analysis of common bottlenose dolphin
multilocus genotypes from the five study areas (North Adriatic,
Central-South Adriatic, Ionian, Aegean, and Tyrrhenian Seas).
S. GASPARI ET AL.
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
significant differences in nuclear DNA between all
areas except between the Aegean Sea and the other
east Mediterranean sampling sites. Mitochondrial
DNA data revealed a strong divergence between the
geographically distant sampling sites of the North
Tyrrhenian Sea and Aegean Sea, but did not show a
clear separation between the Tyrrhenian Sea and
the Adriatic Sea. These results are in general
agreement with the previous study conducted by
Natoli et al. (2005) in which an overall pattern of
genetic differentiation was found between the west
and east Mediterranean basins, and mtDNA
sequence divergence was weaker than nuclear DNA
data. Lack of genetic differentiation among the
Aegean, Ionian and Central-South Adriatic in both
mitochondrial and nuclear markers mirrored the
general south-eastwardly pattern of gene flow
detected by the Bayesian inference methods.
Further subdivision was found within the
Adriatic Sea, where common bottlenose dolphins
revealed a fine-scale genetic structure at nuclear
DNA markers, showing a differentiation between
north and central-south sub-basins, as well as
between the west and east coasts. This subdivision
seems to reflect the physiographic differences found
along both latitudinal and longitudinal axes of the
Basin. The Adriatic Sea presents very diverse
oceanographic features between the sandy and
fairly linear west coast and the karst rocky and
rugged east coast. Important environmental
differences are also present between the North and
Central-South Adriatic sub-basins, which are
characterized by different depth gradients and
water mass circulations (Artegiani et al., 1997).
Differences in marine habitats and resources
could be among the mechanisms by which this
differentiation has evolved and is maintained.
Different degrees of resource specialization, with
respect to prey and specific habitats, have been
documented in bottlenose dolphins by Barros and
Wells (1998), Connor (2000), and Sargeant et al.
(2005). Combined results between genetics,
photo-identification and stable isotope analyses
suggest, as a preliminary hypothesis, that factors
such as local site fidelity and/or physiographic
features, rather than prey specialization, are the
mechanisms maintaining the observed population
structure. In fact, stable isotopes analyses conducted
in the central Adriatic shows that common
bottlenose dolphins readily shift prey, probably
dependent on which prey is available (Holcer, 2012).
The genetic structure of bottlenose dolphin
populations at a relatively small geographic
scale was also suggested by Krützen et al.(2004)at
Shark Bay, Australia, by Sellas et al. (2005) in the
western North Atlantic, and recently by Ansmann
et al. (2012) for inshore bottlenose dolphins
(Tursiops aduncus) in Moreton Bay, Australia.
These studies and our data imply a broader,
worldwide pattern of small-scale differentiation for
this genus. Photo-identification data have suggested
that common bottlenose dolphins of the Adriatic
Sea are structured in putative local populations
(Fortuna, 2006; Genov et al., 2008, 2009; Holcer,
2012; Pleslićet al., in press), corroborating the results
of this study. For instance, in the North Adriatic, no
matches were found between individuals from the
Gulf of Trieste (Italy, Slovenia and Croatia) and the
Table 3. Means and 95% confidence intervals of the posterior probabilities for migration rates between bottlenose dolphin populations. Sampling sites
listed in the rows represent populations of origin of migrants. Recipient populations are listed in the columns
Recipient population
Population of origin North Adriatic Central-South Adriatic Ionian Aegean Tyrrhenian
North Adriatic 0.981 0.298 0.219 0.189 0.273
(0.951, 0.997) (0.249, 0.327) (0.103, 0.307) (0.073, 0.298) (0.200, 0.324)
Central-South Adriatic 0.004 0.679 0.023 0.026 0.015
(0.000, 0.025) (0.667, 0.711) (0.000, 0.097) (0.000, 0.107) (0.000, 0.065)
Ionian 0.005 0.007 0.711 0.026 0.012
(0.000, 0.022) (0.000, 0.034) (0.667, 0.810) (0.000, 0.110) (0.000, 0.052)
Aegean 0.004 0.006 0.023 0.728 0.012
(0.000, 0.022) (0.000, 0.032) (0.000, 0.101) (0.669, 0.858) (0.000, 0.053)
Tyrrhenian 0.004 0.007 0.023 0.029 0.688
(0.000, 0.020) (0.000, 0.037) (0.000, 0.102) (0.000, 0.113) (0.667, 0.735)
POPULATION STRUCTURE OF TURSIOPS TRUNCATUS IN THE EAST MEDITERRANEAN
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
Kvarnerićarchipelago, Croatia, only 120–150 km
apart (Genov et al., 2008, 2009). Similarly, long-term
studies based on photo-identification show an
even finer-scale population structure in this
region, reporting very little exchange among local
populations along the Croatian coast (Holcer,
2012). Conversely, some individuals were found in
both catalogues of the Kvarnerićand Kornati
archipelagos (North Adriatic), approximately
80–100 km apart (Fortuna, 2006; Kammigan
et al., 2008). Data from the Vis and Lastovo
archipelagos (Central-South Adriatic), 80–150 km
south of Kornati archipelago, (Holcer, 2012) did
not show any match with other sites in the region.
The use of stranded samples may lead to an
underestimate of genetic differentiation, particularly
when population structure is assessed over a
relatively small geographic scale (Bilgmann et al.,
2011). This may render difficult the identification of
management units for conservation. Although the
majority of tissue samples collected for this study
were obtained from stranded specimens, the results
did show evidence of population structure in the
Adriatic Sea, which was also strengthened by a
strong match with photo-identification and stable
isotopes data.
When considering the discrepancies between
nuclear and mitochondrial DNA the data may be
influenced by sex-biased gene flow. Differential
dispersal of males and females can have a major
influence on the distribution of maternally and
bi-parentally inherited genes in bottlenose dolphin
populations (Hoelzel et al., 1998). Diverse dispersal
behaviours may be adopted in different regions. In
some populations, male bottlenose dolphins disperse
more often and further than females (Krützen et al.,
2004; Möller and Beheregaray, 2004), while in other
populations there are no significant differences in
dispersal between the sexes (Natoli et al., 2005).
Unlike the nuclear DNA data, the mitochondrial
DNA sequences comparisons recovered no
differences among sampling sites in the Adriatic
Sea, suggesting that female bottlenose dolphins may
have an important role in mediating gene flow
across the basin.
The absence of a clear assignment of individual
multilocus genotypes to distinct genetic clusters,
despite the Bayesian clustering approach
distinguishing five partitions, may be due to
limitations of the software in detecting genetic
differentiation when F
ST
values are low (Latch et al.,
2006). The existence of both resident and visiting
individuals, as indicated by photo-identification
studies (Genov et al., 2008; Pleslićet al., in press)
may be an additional or concomitant cause of why
consistent assignment to distinct clusters could not
be detected. In particular, a significant proportion
of either migrants or individuals of mixed ancestry
may result in a degree of gene flow that potentially
obscures population structure.
Analysis of recent migration rates recorded a
significant rate of gene flow from the North
Adriatic towards the other study areas, and
negligible movements of individual dolphins
between all other sampled areas. This study
supports the findings that there are differences in
movement patterns of resident and visiting animals
according to the data derived from different
projects in the North Adriatic (Fortuna, 2006;
Genov et al., 2008, 2009; Kammigan et al., 2008;
Holcer, 2012; Pleslićet al., in press). The
Kvarnerićbottlenose dolphin population showed
up to 30% rate of non-random temporary
emigration (Fortuna, 2006), while in the Vis and
Lastovo archipelagos Holcer (2012) described
areas with locally defined groups mixing with
transient individuals. Similarly, 48% of about 100
identified individuals in the Gulf of Trieste were
recorded only once and therefore considered
transient or visiting dolphins (Genov et al., 2008).
Conservation implications
The Adriatic Sea common bottlenose dolphins
inhabit an environment greatly affected by human
activities, including intensive fishery, gas and oil
exploitation, maritime traffic, tourism and chemical
pollutants. These pressures, particularly fishery
bycatch, may have a strong, adverse impact on
population viability and need to be carefully
assessed and managed at scales that are consistent
with the population structure of bottlenose dolphins
(Fortuna et al., 2010). Indeed, scientific assessment
of population structure, distribution and status,
identification of threats affecting a specie’sviability
and implementation of conservation and
S. GASPARI ET AL.
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
management measures should not prescind from
an international and coordinated plan of actions.
The Mediterranean ‘subpopulation’(sensu IUCN)
of Tursiops truncatus is listed as Vulnerable (A2cde)
in the IUCN Red List (http://www.iucnredlist.
org/details/16369383/0). The IUCN assessment
noted that ‘the listing of Mediterranean common
bottlenose dolphins as a single subpopulation should
not be interpreted to mean there is no further
subpopulation structure within the region’.This
study provides evidence of fine-scale population
structure and dispersal patterns in the Adriatic Sea
and the eastern Mediterranean basin and supports
the necessity of identifying separate areas for
conservation. Genetic divergence between the North
and Central-South Adriatic dolphin populations
strengthens the case for conservation actions
targeting different sites. Population management
actions should also consider how the impact of
human activities differs across geographically distinct
areas. The results, describing possible south-eastward
migration routes from the North Adriatic,
advocate the development of an international
network of marine protected areas and connecting
corridors (Bearzi et al., 2011). The Convention for
the Protection of the Marine Environment and the
Coastal Region of the Mediterranean (Barcelona
1995) considers the development of protected
areas for cetacean species in the Adriatic basin.
Croatia joined the EU in 2013, and although there
is no control over marine regions currently
under the jurisdiction of Bosnia-Herzegovina,
Montenegro and Albania, the application of these
countries to become EU members would require a
certain level of cooperation. In this instance, the
development of the EU Natura 2000 network of
the Habitats Directive (Directive 92/43/EEC) and
the extension of the Pan European Ecological
Network of the Convention on the conservation
of European wildlife and natural habitats (Bern,
1979) to non-EU States, provide the instruments
to coordinate international conservation strategies
within the Adriatic Sea (see Mackelworth et al.,
2011, 2013 for a review; Genov et al., 2012). In
addition, the implementation of the Marine
Strategy Framework Directive (MSFD) (Directive
2008/56/CE) and its monitoring activities (Article
11) calls for the cooperation between EU and
third party States in regions of shared marine
waters (Article 13). This is of particular importance
as the Adriatic Sea is one of only four defined
sub-regions of the Mediterranean basin and it is
considered to be the future pilot area for the
development of marine spatial planning which
includes the definition of MPAs. While there are
clear arguments for the development of spatial
conservation measures for the Adriatic Sea, the
paucity of data available requires that further
genetic research is undertaken to help define these
regionsinthenearfuture,especiallytohelpinthe
effective evaluation of fishery-induced mortality at
the level of ‘units-to-conserve’(Taylor et al.,
2010). The MSFD requires member states to work
together to establish programmes for monitoring
the ‘Good Environmental Status’of shared waters
by July 2014. This would be an excellent
timeframe for the development of further genetic
research in the Adriatic region.
ACKNOWLEDGEMENTS
Thanks to The Marine Mammal Tissue Bank,
University of Padua, Carola Vallini, ARCHE,
Marco Affronte, Fondazione Cetacea, Paola Pino
d’Astore; Petros Lyberakis, Natural History
Museum of Crete, Elisabeth Dimitra, University
of the Aegean, Joan Gonzalvo, Tethys Research
Institute, Christos Delistathis, Spyros Eleftheriou
and Letizia Marisili. We are also grateful to Scott
Baker and two anonymous referees for useful
comments on an early version of the manuscript.
This study was funded by the Italian Ministry of
Agriculture Food and Forestry Policies (MPAAF).
REFERENCES
Ansmann IC, Parra GJ, Lanoyon JM, Seddon JM. 2012.
Fine-scale genetic population structure in mobile marine
mammal: inshore bottlenose dolphins in Moreton Bay,
Australia. Molecular Ecology 21: 4472–4485.
Artegiani A, Bregant D, Paschini E, Pinardi N, Raicich F,
Russo A. 1997. The Adriatic Sea general circulation. Part I:
air-sea interactions and water mass structure. Journal of
Physical Oceanography 27: 1492–1513.
Barros NB, Wells RS. 1998. Prey and feeding patterns of
resident bottlenose dolphins (Tursiops truncatus) in Sarasota
Bay, Florida. Journal of Mammalogy 79: 1045–1059.
POPULATION STRUCTURE OF TURSIOPS TRUNCATUS IN THE EAST MEDITERRANEAN
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
Bearzi G, Fortuna CM, Reeves RR. 2009. Ecology and
conservation of common bottlenose dolphins Tursiops
truncatus in the Mediterranean Sea. Mammal Review 39:92–123.
Bearzi G, Bonizzoni S, Gonzalvo J. 2011. Mid-distance
movements of common bottlenose dolphins in the coastal
waters of Greece. Journal of Ethology 29: 369–374.
Beerli P. 2006. Comparison of Bayesian and maximum-likelihood
inference of population genetic parameters. Bioinformatics 22:
341–345.
Bilgmann K, Möller LM, Harcourt RG, Kemper SM,
Beheregaray LB. 2011. The use of carcasses for the analysis
of cetacean population genetic structure: a comparative
study in two dolphin species. PLoSOne 6: e20103.
Caldwell CM, Gaines MS, Hughes CR. 2002. Eight
polymorphic microsatellite loci for bottlenose dolphin and
other cetacean species. Molecular Ecology Notes 2: 393–395.
Connor RC. 2000. Group living in whales and dolphins. In
Cetaceans Societies:Field Studies of Dolphins and Whales,
Mann J, Connor RC, Tyack PL, Whitehead H (eds). The
University of Chicago Press: Chicago, IL; 199–218.
Cushman-Roisin B, GačićM, Poulain P-M, Artegiani A. 2010.
Physical Oceanography of the Adriatic Sea.Past,Present and
Future. Kluwer Academic Publishers: Dordrecht.
Earl DA, vonHoldt BM. 2012. STRUCTURE HARVESTER
a website and program for visualizing STRUCTURE output
and implementing the Evanno method. Conservation Genetics
Resources 4: 359–361.
European Commission. 2011. Commission Staff Working
Paper. Relationship between the initial assessment of
marine waters and the criteria for good environmental
status. SEC(2011) 1255 final. Downloadable at: http://ec.
europa.eu/environment/marine/pdf/
SEC_2011_1255_F_DTS.pdf
Evanno G, Regnaut S, Goudet J. 2005. Detecting the number
of clusters of individuals using the software STRUCTURE
a simulation study. Molecular Ecology 14: 2611–2620.
Excoffier L, Lischer HEL. 2010. Arlequin suite ver 3.5: a new
series of programs to perform population genetic analysis
under Linux and Windows. Molecular Ecology Resources
10: 564–567.
Falush D, Stephens M, Pritchard JK. 2003. Inference of
population structure using multilocus genotype data:
Linked loci and correlated allele frequencies. Genetics 164:
1567–1587.
Fortuna CM. 2006. Ecology and conservation of bottlenose
dolphins (Tursiops truncatus) in the north-eastern Adriatic
Sea. PhD thesis, University of St. Andrews, UK.
Fortuna CM, Vallini C, Filidei E, Ruffino M, Consalvo I, Di
Muccio S, Gion C, Scacco U, Tarulli E, Giovanardi O,
Mazzola A. 2010. By-catch of cetaceans and other species
of conservation concern during pair trawl fishing operations
in the Adriatic Sea (Italy). Chemistry and Ecology 26:65–76.
Fullard KJ, Early G, Heide-Jørgensen MP, Bloch D,
Rosing-Asvid A, Amos W. 2000. Population structure of
long-finned pilot whales in the North Atlantic: a correlation
with sea surface temperature? Molecular Ecology 9:949–958.
Genov T, Kotnjek P, Lesjak J, Hace A, Fortuna CM. 2008.
Bottlenose dolphins (Tursiops truncatus) in Slovenian and
adjacent waters (northern Adriatic Sea). Annales, Series
Historia Naturalis 18: 227–244.
Genov T, Wiemann A, Fortuna CM. 2009. Towards
identification of the bottlenose dolphin (Tursiops truncatus)
population structure in the north-eastern Adriatic sea:
preliminary results. Varstvo narave 22:73–80.
Genov T, Hace A, Centrih T, Kotnjek P. 2012. Slovenian
experience: from no dolphins to dolphin Special Area of
Conservation. ECS/ASCOBANS/ACCOBAMS workshop
on the EU Habitats Directive and its implementation in
relation to cetaceans. European Cetacean Society, Galway,
Ireland.
Goudet J. 1995. FSTAT Version 1.2: a computer program to
calculate F-statistics. Journal of Heredity 86: 485–486.
Hoelzel AR, Potter CW, Best PB. 1998. Genetic differentiation
between parapatric ’nearshore’and ’offshore’populations of
the bottlenose dolphin. Proceedings of the Royal Society of
London. Series B 265: 1177–1183.
Hoelzel AR, Natoli A, Marilyn E, Dahlheim ME, Olavarria C,
Baird RW, Black NA. 2002. Low worldwide genetic diversity
in the killer whale (Orcinus orca): implications for
demographic history. Proceedings of the Royal Society of
London. Series B 269: 1467–1473.
Holcer D. 2012. Ecology of the common bottlenose dolphin,
Tursiops truncatus (Montagu, 1821) in the Central Adriatic
sea. Faculty of Sciences, University of Zagreb, Croatia.
Hubisz MJ, Falush D, Stephens M, Pritchard JK. 2009.
Inferring weak population structure with the assistance of
sample group information. Molecular Ecology Resources 9:
1322–1332.
Jørgensen HBH, Hansen MM, Bekkevold D, Ruzzante DE,
Loeschcke V. 2005. Marine landscapes and population
genetic structure of herring (Clupea harengus L.) in the
Baltic Sea. Molecular Ecology 14: 3219–3234.
Kammigan IC, Bräger S, Hennig V, Wiemann A, Impetuoso A.
2008. Ecology of bottlenose dolphins (Tursiops truncatus)in
the Kornati National Park, Croatia: population estimation,
group composition and distribution. In Marine Mammals in
Time:Past,Present and Future, Pierce GJ, Philips E, Lick R
(eds). 22nd Annual Conference of the European Cetacean
Society, European Cetacean Society, Egmond aan Zee: The
Netherlands.
Krützen M, Sherwin WB, Berggren P. 2004. Population
structure in an inshore cetacean revealed by microsatellites
and mtDNA analyses: bottlenose dolphins (Tursiops sp.) in
Shark Bay, Western Australia. Marine Mammal Science 20:
28–47.
Latch E, Dharmarajan G, Glaubitz J, Rhodes OE. 2006.
Relative performance of Bayesian clustering software for
inferring population structure and individual assignment at
low levels of population differentiation. Conservation
Genetics 7: 295–302.
Mackelworth P, Holcer D, JovanovićJ, Fortuna C. 2011.
Marine conservation and accession, the future for the
Croatian Adriatic. Environmental Management 47: 644–655.
Mackelworth P, Holcer D, Fortuna C. 2013. Unbalanced
governance: the Cres-Losinj Special Marine Reserve a
missed opportunity. Marine Policy 41: 126–133.
Möller LM, Beheregaray LB. 2004. Coastal bottlenose
dolphins from southeastern Australia are Tursiops aduncus
according to sequences of the mitochondrial DNA control
region. Marine Mammal Science 17: 249–263.
Natoli A, Peddemors VM, Hoelzel AR. 2004. Population
structure and speciation in the genus Tursiops based on
microsatellite and mitochondrial DNA analyses. Journal of
Evolutionary Biology 17: 363–375.
S. GASPARI ET AL.
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)
Natoli A, Birkun A, Aguilar A, Lopez A, Hoelzel AR. 2005.
Habitat structure and the dispersal of male and female
bottlenose dolphins (Tursiops truncatus). Proceedings of the
Royal Society of London. Series B 272: 1217–1226.
Park SDE. 2001. Trypanotolerance in West African cattle and
the population genetic effects of selection. PhD thesis,
University of Dublin, Republic of Ireland.
Peakall R, Smouse PE. 2012. GenAlEx 6.5: genetic analysis in
Excel. Population genetic software for teaching and
research –an update. Bioinformatics 28: 2537–2539.
Peakall R, Smouse PE, Huff DR. 1995. Evolutionary
implications of allozyme and RAPD variation in diploid
populations of dioecious buffalograss Buchloe dactyloides.
Molecular Ecology 4: 135–147.
PleslićG, Rako N, Mackelworth P, Wiemann A, Holcer D, Fortuna
C. (in press). The abundance of common bottlenose dolphins
(Tursiops truncatus) in the former marine protected area of the
Cres-Lošinj archipelago, Croatia. Aquatic Conservation:Marine
and Freshwater Ecosystems. doi:10.1002/aqc.2416.
Pritchard JK, Stephens M, Donnelly PJ. 2000. Inference of
population structure using multilocus genotype data.
Genetics 155: 945–959.
Rosel PE, Forgetta V, Dewar K. 2005. Isolation and
characterization of twelve polymorphic micosatellite markers
in bottlenose dolphins (Tursiops truncatus). Molecular
Ecology Resources 5:830–833.
Rousset F. 2008. GENEPOP’007: a complete re-implementation
of the GENEPOP software for Windows and Linux. Molecular
Ecology Resources 8:103–106.
Sargeant BL, Mann J, Berggren P, Krützen M. 2005.
Specialization and development of beach hunting, a rare
foraging behavior, by wild bottlenose dolphins (Tursiops sp.).
Canadian Journal of Zoology 83:1400–1410.
Sellas AB, Wells RS, Rosel PE. 2005. Mitochondrial and
nuclear DNA analyses reveal fine scale geographic structure
in bottlenose dolphins (Tursiops truncatus) in the Gulf of
Mexico. Conservation Genetics 6: 715–728.
Shinohara M, DomingoRoura X, Takenaka O. 1997.
Microsatellites in the bottlenose dolphin Tursiops truncatus.
Molecular Ecology 6: 695–696.
Smouse PE, Peakall R. 1999. Spatial autocorrelation analysis
of individual multiallele and multilocus genetic structure.
Heredity 82: 561–573.
Taylor BL, Martien K, Morin P. 2010. Identifying units to
conserve using genetic data. In Marine Mammal Ecology
and Conservation:A Handbook of Techniques, Boyd IL,
Bowen WD, Iverson SJ (eds). Oxford University Press: New
York; 306–324.
Valsecchi E, Amos W. 1996. Microsatellites markers for the study
of cetacean populations. Molecular Ecology 5:151–156.
Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P.
2004. MICRO-CHECKER: software for identifying and
correcting genotyping errors in microsatellite data.
Molecular Ecology Notes 4: 535–538.
Wilson GA, Rannala B. 2003. Bayesian inference of recent
migration rates using multilocus genotypes. Genetics 163:
1177–1191.
Xiong Y, Brandley MC, Xu SX, Zhou KY, Yang G. 2009. Seven
new dolphin mitochondrial genomes and a time-calibrated
phylogeny of whales. BMC Evolutionary Biology 9:20.
SUPPORTING INFORMATION
Additional supporting information may be found in
the online version of this article at the publisher’s
web site:
Table S1. Observed and expected heterozygosity per
locus per putative populations (four loci were
excluded).
POPULATION STRUCTURE OF TURSIOPS TRUNCATUS IN THE EAST MEDITERRANEAN
Copyright #2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)