101

J. Black Sea/Mediterranean Environment Vol. 23, No. 2: 101-120 (2017)

RESEARCH ARTICLE Encounter rate and relative abundance of bottlenose dolphins and distribution modelling of main cetacean species in the North Aegean Sea (Greece) Cristina B. Milani 1,2*, Adriana Vella1, Pavlos Vidoris2, Aristidis Christidis2, Emmanouil Koutrakis2, Georgos Sylaios3, Argyris Kallianiotis2

1Conservation Biology Research Group, University of Malta, Msida MSD 2080 MALTA 2Fisheries Research Institute of Kavala, 64007 Nea Peramos, Kavala, GREECE 3Department of Environmental Engineering, Democritus University of Thrace, Xanthi, GREECE *Corresponding author: crismilani13@hotmail.com Abstract The encounter rate, relative abundance and distribution of bottlenose dolphins were investigated as part of a first detailed study on cetacean populations in the Greek North Aegean Sea. Sighting data were recorded between 2005 and 2013 through dedicated scientific marine surveys, covering 14,701 km. The line-transect sampling method was used to estimate relative abundance, using Distance software 6.0. Bottlenose dolphins presented an overall encounter rate of 0.46 groups/100km (2.5 dolphin/100km) and a mean group size of 3.45 (SE=0.38). The yearly relative abundance in the area was estimated after post-stratification at 377 individuals (95% C.I.=289–465; CV=18.37). Environmental modeling has been proposed for the main species of the study area using GAM and PCA; dolphin sightings were then correlated to eight environmental variables (distance from the coast, depth, slope, gradient of salinity, median salinity, gradient of temperature, median temperature and mean current). Bottlenose dolphin sightings were significantly correlated to depth and median temperature. This research presents the first data on encounter rate, relative abundance for the species and species distribution in relation to environmental variables in study area which are essential for effective conservation and management considerations. Key words: DISTANCE sampling analysis, Generalized Additive Modelling (GAM), Marine Protected Areas, North Aegean Sea, Principal Component Analysis (PCA), Tursiops truncatus Received: 24.05.2017, Accepted: 05.08.2017

102

Introduction The North Aegean Sea is very rich in nutrients which reach it from numerous rivers (Sylaios et al. 2005) and from the Black Sea (Poulos et al. 1997). ACCOBAMS (The Agreement on the Conservation of Cetaceans in the Black Sea, Mediterranean Sea and contiguous Atlantic area) has reported the North Aegean Sea as one of the most important areas for cetacean species in the Mediterranean Sea and has proposed a Marine Protected Area from the Chalcidice Peninsula up to the Turkish border, including some Greek islands (ACCOBAMS 2004, 2007, 2010). The ecological importance of the region is confirmed at a national level, by the establishment of three new NATURA 2000 protected areas in the North Aegean Sea (Kassandra Peninsula and Month Athos in Chalcidice and the island of Lemnos) and the widening of two existing areas (Evros Delta and Nestos Delta and surrounding areas) (Kotzageorgis et al. 2015). The presence of the bottlenose dolphins (Tursiops truncatus Linnaeus, 1758) in the North Aegean Sea is documented by several independent studies involving opportunistic sighting reports, fishermen interviews, stranding records, diet analysis and hydrophone recordings (Casale et al. 1999; Mitra et al. 2001, 2004; Frantzis 2009; Frantzis et al. 2003; Milani et al. 2011, 2012, 2013, 2017; Ryan et al. 2014). However, quantitative data on encounter rate and relative abundance, as well as distribution analyses of cetacean species for this region were missing, previous to this research. These data are useful for management and conservation decisions based on scientific knowledge. In order to determine the best future protected area for cetaceans, population parameters, such as abundance, distribution, habitat use and home range of each species, need to be investigated (Fayos et al. 2001). The work presented here is part of a wider research effort which investigated the cetacean fauna of the northern part of the North Aegean Sea using several methodologies, such as marine surveys, diet analysis, fishermen interviews and evaluation of conflicts with fisheries (Milani 2015, 2017; Milani et al. submitted). The research presented here focuses on the encounter rate, relative abundance and distribution of bottlenose dolphins inhabiting the study area. Materials and Methods Survey design and sighting data Dedicated marine surveys were undertaken between October 2005 and August 2013 in an area of approximately 2000 km2 in the Gulf of Kavala and around Thassos, in the northern portion of the North Aegean Sea (Figure 1).

103

Figure 1. Study area in the Gulf of Kavala and Thassos. The research area is delimited by red lines and the coast. Upper right box: the map of the whole study area with the

equal-spaced zig-zag transects designed by DISTANCE. A 14 m research vessel with two engines of 370 Hp each was equipped with a Global Positioning System (GPS), an OLEX AS sea bottom depth-finder, a SIMRAD ES60 eco-sounder, a radio VHF, a marine reticule binocular Konus Tornado 7x50 and a Canon Powershot Pro1 digital camera fitted with a 300 mm lens. The boat surveys were conducted with a crew of 2 to 4 people and a minimum of 2 trained observers, one of whom was a researcher of this project. While the boat was underway, the researchers observed the sea around the boat, using both the naked eye and binocular 7x50, for dolphin schools ahead and on each side of the boat. The vessel was used to run standard line transects which sampled the research area throughout the four seasons. Transects followed a systematic design with Equal Spaced zigzag lines for a more uniform coverage, avoiding double-counting of cetacean groups, following authors’ indications (Buckland et al. 2010). Transect covered a total length of 500km and was performed in total once a season during the study period. For practical reasons, the research area was divided in subareas where part of transect was performed separately, but always within few days, in order to assure the coverage of the whole area in the minimum time and within the season.

AEGEAN SEA

104

Sighting effort was measured as the numbers of kilometres travelled on effort. Survey data were defined ‘on effort’ when the following conditions applied: 1) daylight and good visibility (at least 2nm); 2) sea state ≤ 3 Beaufort with no swell; 3) at least two experienced observers scanning the sea surface; 4) eye elevation of approximately 1.5-2.5 m for both observers; and 5) survey speed between 7 and 10 knots (Bearzi et al. 2005). Following standard protocols for cetacean sightings (Forcada and Hammond 1998; Dawson et al. 2004, 2008; Cañadas 2006; SCANS-II protocol of ASCOBANS 2008; Cañadas et al. 2010; MacLeod et al. 2010; Dick and Hines 2011; Hammond et al. 2013), once an animal or group was detected, primary data were taken, such as name of the observer, date and time, GPS position, sea state and weather, angle of the detected group on the trackline, estimated radial distance, species and number of individuals (best estimate), initial behaviour with time interval recording every 5 minutes (Shane 1990), association with sea birds and boat presence. All data recorded, except for behaviour and association with birds and boats were used in the present work. Statistical analysis For visual data representation, the software Ocean View was used for the distribution maps of the species in the study area. The maps generated by the program show the total sighting frequency of bottlenose dolphins in the research area, related to search effort. The scales represent the number of cetacean groups detected in a grid divided to the number of times that the grid was sampled. Sighting frequency is presented according to season. Encounter rates were calculated as the total number of animals and groups encountered per kilometer and per hour of research effort. Data are presented according to season and cumulative effort during each season. Univariate test of Significance and Kruskal-Wallis ANOVA test were performed to see if the values of encounter rates among seasons were significant. For relative abundance estimation the DISTANCE 6.0 release 2 program using the conventional distance sampling (CDS) and the multiple covariate distance sampling (MCDS) methods were used (Buckland et al. 2010; Thomas et al. 2010). The study area was divided into 162 grid cells, with a cell resolution of 2 minutes latitude by 2 minutes longitudes each. A detection function was estimated from distance data, using all sightings made on effort since 2005. Kolmogorov-Smirnov test and Cramer-von Mises family test were used to test goodness of fit of the detection function. Analyses were run both with no truncated data and with perpendicular distance data right truncated at 5%, in order to identify the best set of analyses, following the recommendation of Buckland et al. (2001). The selection of the best detection function was made using Akaike’s Information Criterion (AIC) and Goodness of fit (GOF) tests (Buckland et al. 2004). Estimate values of relative abundance of groups and of group size were produced over the

105

grid cells of the study area, according to the values of the detection function used in the final model. The estimates of abundance for the species was calculated for the total area. Statistical relationships between bottlenose dolphins’ distributions and environmental parameters have been investigated through regression models (principal component analysis PCA and generalized additive modeling GAM) (Hastie and Tibshirani 1990; Schimek 2000; Clarke and Warwick 2001). Principal component analysis (PCA) is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. The definition of principal components is in terms of successively maximizing the variance of sample points projected along each axis, with the variance decreasing from PC1 to PC2 and eventually to PC3. The values of these variances represent the measure of the amount of information contained in each axis (Clarke and Warwick 2001). Generalized additive model (GAM) was used in order to predict dolphin abundance using a series of eight environmental variables. The purpose of generalized additive models is to maximize the quality of prediction of a dependent variable Y from various distributions, by estimating unspecific (non-parametric) functions of the predictor variables which are connected to the dependent variable via a link function (Oedekoven et al. 2014). The parameters utilized in these analyses were distance from the coast, depth, slope, salinity gradient (dS/dD), median salinity, temperature gradient (dT/dD), median temperature and mean current. Data regarding the distance from the coast were extrapolated a posteriori from the GPS position, using a map from the Hydrographic Service of the Greek Navy and the program Google Earth. Slope was calculated by the Euclidean distance between the depth of the grid where sighting took place and the depth of the previous grid. The salinity gradient (dS/dD), median salinity, temperature gradient (dT/dD), median temperature and mean current were obtained using the European platform MyOcean, a concerted and integrated pan-European capacity for ocean monitoring and forecasting, which includes data from a hydrodynamic and a mathematical model, interconnected between them. For the regression models, software STATISTICA 10 (2010) and PRIMER 6 (Clarke and Gorley 2006) were used. In GAM, the smoothed values of each environmental variable were plotted against the residuals after removing the effect of all other environmental variables (Hastie and Tibshirani 1990). Analysis of variance (ANOVA) was used to test statistically significant differences in PCA and R-square value and deviance statistic was computed on GAM (Oedekoven et al. 2014) to identify which environmental parameters are effecting the distribution of dolphins.

106

Results Bottlenose dolphin distribution and abundance A total of 14,701 km was surveyed on effort between autumn 2005 to summer 2013 in the study area throughout the different seasons of the year. In the whole period of the study cumulative effort (km searched) was dissimilar across years and seasons with less research effort spent during winter, due to bad weather conditions. A total of 447 surveys were undertaken and several cetacean species were detected. Here only the data for bottlenose dolphins are presented: 68 sightings of bottlenose dolphins, in total 368 individuals, were detected during dedicated marine surveys. Since the survey effort increased in the last three years of the study (2011, 2012 and 2013), the datasets were pooled into two strata, accordingly with equal effort: 2005-2010 and 2011-2013, to investigate any changes in abundance between these two periods (Table 1).

Table 1. Summary of bottlenose dolphin sightings during the years 2005-2013

Effort (km) Sightings Individuals

observed

Mean group size (SE)

ER groups (SE)

ER individuals

(SE)

2005-2010 6850.5 21 188 5.568 (2.553)

0.0031 (0.003)

0.027 (0.002)

2011-2013 7850.6 47 180 2.835 (0.799)

0.0060 (0.002)

0.023 (0.001)

Total 14701.2 68 368 3.447 (0.375)

0.0046 (0.001)

0.025 (0.004)

ER: encounter rate; SE: Standard Error Observations of bottlenose dolphins divided per season and according to survey effort are shown in Table 2. The Ocean View maps of the species distribution for the whole period and according to seasons are showed in Figure 2. Bottlenose dolphins were present in the entire study area with prevalence along the western coast of the mainland and around Thassos. Table 2. Total effort in km (in hours), number of groups (number of individuals), mean

group size and encounter rates (ER) for groups and for individuals (per km) for bottlenose dolphins observed in the study area during the period 2005-2013 per season

(Kruskal-Wallis ANOVA by Ranks H = 2.323; p = 0.128 per groups; H = 4.734; p = 0.030 per individuals; N = 68).

Effort in km (h)

N. of groups (individuals)

Mean group size (SE)

ER groups (SE)

ER indiv. (SE)

Spring 2,632 (256.5) 18 (111) 6.17 (1.90) 0.007 (0.004) 0.042 (0.06) Summer 8,104 (925) 23 (118) 5.13 (1.41) 0.003 (0.002) 0.015 (0.02) Autumn 2,867 (312.5) 14 (62) 4.43 (1.43) 0.005 (0.003) 0.022 (0.03) Winter 1,326 (93) 13 (77) 6.33 (1.48) 0.009 (0.005) 0.057 (0.04)

107

Figure 2. Densities of bottlenose dolphins in the study area during the period 2005-2013 a) total records; b) spring; c) summer; d) autumn; e) winter. The values of density are expressed in relation to effort (as number of groups detected in a grid/number of times

that the square was sampled).

Detection function, estimated distribution, abundance and trend For the estimation of the detection function, boat distance was truncated at 1,200m after visual inspection of the data, due to limitation that occurred on the field in the calculation of higher distance with the reticule binocular. Sixty-six

108

sightings of bottlenose dolphins were used for analysis and data for all years were pooled. Forty-eight models were fitted by the Conventional Distance Sampling method on Distance software. The best fitting model for the detection function was selected based on it AIC value; it was found to be a Half-normal key function with cosine series expansion and zero adjustment terms. The goodness of the fit of the distribution function was tested using the following two tests: Kolmogorov-Smirnov test = 0.5413; Cramer-von Mises family tests for goodness of fit = 0.700 <p< 0.800. Both tests present quite high values of p, required for a good fit of the detection function. Figure 3 shows the fit of the empirical distribution function and the observed frequency at given distances, pooled over all covariates, and the fitted half-normal function.

Figure 3. Perpendicular distance distribution, pooled over all covariates (histograms)

which represent the sightings of bottlenose dolphins, and the fitted half-normal detection function (line)

Analyses of data against time were done both with the pooled and estimated abundance and density are shown in Table 3. The yearly relative abundance of bottlenose dolphins in the whole area was estimated after post-stratification.

Table 3. Parameter estimation for Tursiops truncatus for 2005-2013

Parameter Point Estimate

Standard Error % C.V. 95% Conf. Interv.

DS 0.0081 0.0012 15.03 0.0060 0.011 E (S) 5.26 0.56 10.57 4.27 6.49

D 0.042 0.0078 18.37 0.030 0.061 N 3013 553.47 18.37 2106 4308 Ā 377 88.81 18.37 289 465

DS: estimate of density of clusters, E(S): estimate of expected value of cluster size, D: estimates of density of dolphins, N: estimates of number of dolphins in the whole area, Ā: estimate yearly relative abundance of dolphins in the whole area, C.V.: Coefficient of Variation

109

Correlation with environmental parameters The correlation between dolphin abundance and environmental parameters was investigated in the frame of a wider research on dolphin species in the study area. The analysis was run together for the two most abundant species (bottlenose dolphins T. truncatus and short-beaked common dolphins Delphinus delphis), through PCA and GAM. For PCA, the analyses were possible for 98 dolphin sightings (64 observations of bottlenose dolphins and 34 of common dolphins). After transformation and normalization of the data, PCA indicates two principal axes with eigenvalues greater than 1, which explain 64.2% of the total variation (PC1 explains 40.2% and PC2 24% of the total variation). In the PCA, the coefficients in the linear combinations of environmental variables presenting higher values were: salinity gradient (-0.503), median salinity (0.415), temperature gradient (-0.470) and median temperature (-0.449) for PC1 and distance from the coast (-0.695) and slope (-0.659) for PC2. Figure 4 indicates that the first principal axis differentiates sites of warm (summer) from cold (winter) and middle season, which includes both spring and autumn that the analysis pooled together, due to their similar temperature. The second principal axis PC2 differentiates the sites close to the coast with low values of slope from those which are away from the coast with high values of slope.

Figure 4. PCA plot of 98 sites of dolphin observations

Vector plot indicates the contribution of environmental parameters to PCs. The sites of warm (W), cold (C) and middle (M) season are also indicated.

Analysis of variance of the environmental parameters on the first principal axis, PC1, for two factors, (a) species (two dolphin species: bottlenose dolphin and common dolphin) and (b) season (three seasons: warm, cold and mid season) indicated statistically significant differences for the factor season (p=0.0001) but not for the factor species (p=0.250). On the contrary, analysis of variance of the same parameters of the second principal axis, PC2, for two factors, (a) species

110

(two dolphin species: bottlenose dolphin and common dolphin) and (b) season (three seasons: warm, cold and middle) indicated that the scores show statistically significant differences for the factor species (p=0.0001) but not for the factor season (p=0.0879). For bottlenose dolphins, the mean values for the eight environmental parameters were calculated and presented in Table 4.

Table 4. Mean values of the environmental parameters for the bottlenose dolphin sightings

Mean SD Distance from coast 2448.9 (m) 1864.4 Depth 33.36 (m) 19.92 Slope 1.32 (m) 1.15 dS/dD 0.0348 (psu/m) 0.0419 Salinity (median) 37.70 (psu) 0.85 dT/dD 0.0744 (oC/m) 0.10 Temperature (median) 17.29 (oC) 3.82 Current (mean) 0.0244 (m/s) 0.0249

SD: standard deviation The analysis of variance was used to test the difference of bottlenose dolphin abundance between seasons as well as between two temporal periods (2006-2010 and 2011-2013). The analysis indicated statistically significant difference of abundance between seasons (p=0.0078) but not between study periods (p=0.0673). The mean values of abundance for the warm and cold season were 2.810 and 5.167, respectively. Table 5. Fit summary for GAM. The distribution used was Poisson with logarithmic link

function. Extreme values (50 individuals) were excluded. The response is the number of individuals. Significant values are in bold.

Degr. of - freedom

GAM - coef.

Standard - Error

Std. – Score

Non-Linear - p-value

Intercept 1.000 -9.0842 4.2925 -2.1163 Distance coast 4.041 0.0001 0.00005 1.9083 0.6017 Depth 3.998 -0.0006 0.0043 -0.1361 0.0069 Slope 4.029 -0.0765 0.0773 -0.9891 0.1320 dS/dD 3.986 -0.8035 4.3461 -0.1849 0.0782 S. median 4.024 0.2635 0.1100 2.3958 0.8730 dT/dD 4.014 -2.7713 1.5954 -1.7370 0.7901 T. median 4.092 0.0379 0.0247 1.5329 0.0416 Curr.Mean 4.035 1.7913 3.7244 0.4810 0.0855

SD: standard deviation

Generalized additive model (GAM) was used in order to predict the dolphins’ abundance by the eight environmental variables mentioned above. Summary statistics for the GAM analysis presented an R square value of 67.28%, which indicates the goodness of fit of the Poisson distribution with logarithmic link function that was used. The fit summary of the analysis indicates that, among the eight parameters evaluated, depth and median temperature are influencing the

111

bottlenose dolphin abundance (Table 5). The plots of the partial residuals distributions according to the two variables are presented in Figure 5. In the first figure, the adjusted dependent variable (occurrence of bottlenose dolphins) seems to first decrease and then increase in relation to depth, showing maximum values between 23m and 55m. For median temperature, one clear trend can be observed: during the cold season, the adjusted dependent variable seems to increase in relation to temperature, showing maximum values around 16°C, indicating mild winter temperatures, while during the warm season and especially the mid-season, the confidence band is too wide to allow us to extrapolate conclusions.

Figure 5. Spline line and 95% confidence band for depth (a) and median temperature (b).

Response: number of individuals. Extreme values (50 individuals) were excluded. C=cold, W=warm, M=mid-season.

On y-axis the Partial residual distributions are represented. Discussion Knowledge on the population parameters, such as overall and seasonal encounter rates, abundance and distribution patterns, provide useful information on the population status and allow the assessment of anthropogenic impacts on the species. This information is essential for the application of wildlife conservation and management efforts. For this purpose, the present work focused on these parameters for the bottlenose dolphins inhabiting the North Aegean Sea. This study has provided the first relative abundance estimate and distribution analysis of bottlenose dolphins in the Thracian Sea. The results of this study points out that the bottlenose dolphin was commonly encountered species along the continental shelf and was observed in all seasons; its presence in the study area can be considered permanent and independent of seasonal fluctuation, in terms of density and habitat use, reflecting the probable presence of a resident population. Bottlenose dolphin distribution was recorded in the entire study area from Cape Pirgos up to Nestos Delta and around Thassos with a higher distribution close to the mainland, to north Thassos and in the western part of the study area, out of

112

Cape Pirgos. Moreover, the bottlenose dolphin numbers did not show evidence of decline during the study period. The gap in knowledge for bottlenose dolphin’s encounter rate in the Gulf of Kavala and adjacent regions prior to this study does not allow for the comparison with previous opportunistic data. However, the encounter rates (both overall and seasonal) were generally similar to the values recorded in other areas in Greece, such as Kalamos (Bearzi et al. 1997; 2008; Petroselli 2006) and the Evoikos Gulf, where Bonizzoni et al. (2014) estimate the presence of 109 animals in an area of 960km2, but they were lower if compared with other areas of the Eastern Mediterranean Sea, such as in Antalya Bay (Baş et al. 2016), Amvrakikos Gulf (Bearzi et al. 2008) or along the Israeli coast (Scheinin 2010). However, comparison of encounter rates of the same species in different habitats should be considered with caution, due to differences in habitat characteristics and sampling design. In literature, the correlation of dolphin sightings with environmental parameters shows that among the variables found to influence dolphin distribution, prey distribution represents the main one (Cañadas 2006). The latter is reported to be affected by oceanographic factors, such as temperature, salinity, currents, which in turn affect phytoplankton and zooplankton followed by fish, cephalopods and crustaceans, representing dolphin prey (MacLeod et al. 2007; Pirotta et al. 2011; Milani et al. 2017). Prey is also distributed in certain preferred habitats, therefore other geographic factors, such as depth, slope and distance from the coast, could influence their presence (Cañadas et al. 2005; Panigada et al. 2008). In the current study, the species was found to have a preference for shallow waters between 20-50 m on the continental shelf, mainly between 500 to 4500 m from the coast. According to habitat preference of the species (Cañadas et al. 2005), bottlenose dolphin is normally recorded in near-shore waters (Silva et al. 2014) inside the isobaths of 200 m (Cañadas et al. 2005; Viddi et al. 2010; Gnone et al. 2011). Regarding median temperature, the species showed a preference to moderate temperature, avoiding extreme temperatures both in winter and in summer. In other studies the bottlenose dolphin was recorded at a range of temperatures both in winter and in summer, indicating that the species could be found in areas where the temperature is very variable (Bassos-Hull et al. 2013). Similar distribution has been found in other regions of the Mediterranean Sea (Vella 1999, 2005; Fayos et al. 2000; Cañadas et al. 2002; Forcada et al. 2004; Bearzi et al. 2005; Cañadas 2006; Stockin et al. 2006). Therefore, bottlenose dolphins in the northern North Aegean Sea preferred coastal zones with low depth and moderate temperature fluctuations. A seasonal change in distribution and density was observed in the results, characterized by a higher density in winter and early spring, in spite of the greater research effort during summer. The increase in winter and spring sightings can be an effect of biotic and abiotic factors resulting in predictable shifts of seasonal

113

prey distribution (Holt 2003). The North Aegean Sea represents a small “oasis” of high productive waters surrounded by oligotrophic water typical of the Eastern Mediterranean basin (Coll et al. 2010). It is rich in nutrients brought by the numerous rivers outflowing into it, especially during winter and spring, which sustains several species of epipelagic and benthopelagic fish (Kallianiotis et al. 2004; Somarakis et al. 2011). Research analysis conducted on bottlenose dolphin in the study area, has found that the species actually feed on epipelagic and benthopelagic species (Milani 2015; Milani et al. 2017), therefore the increased presence of bottlenose dolphins during spring could be due to the higher food availability offered during this period of the year. The results of multi-year surveys in the northern Greek North Aegean Sea confirm bottlenose dolphin presence and reveal the distribution of this cetacean species also in relation to certain environmental variables. The northern portion of the Greek North Aegean Sea, specifically the coastal waters of the Gulf of Kavala and North Thassos, are important bottlenose dolphin habitats. Bottlenose dolphins are an indicator species and they reflect the effects of human disturbance regime or the efficacy of efforts to mitigate such disturbance effects thus can be useful during management strategies (Viale 1994; Wilson et al. 2004; Pinn 2010). Conservation measures for bottlenose dolphins and their habitat may include the creation of one or more marine reserves or marine parks, as proposed by many international and national entities considering marine conservation in the Mediterranean Sea (Agardy 1997; Wilson et al. 2004; UNEP-MAP-RAC/SPA 2008, 2010; Vella 2009, 2016; CIESM 2004, 2010; Hoyt 2011; OCEANA 2014). Due to the large areas of conservation required for this species, their conservation areas would allow marine biodiversity and habitats to benefit from such an umbrella and flagship species. Acknowledgement We acknowledge the survey volunteers and Dr. Nikolaos Kamidis for Ocean View maps. References ACCOBAMS (2004) Report of the Second Meeting of the Parties to ACCOBAMS. Palma de Maillorca, Spain 9-12 November 2004. 180 pp. ACCOBAMS (2007) Report of the Third Meeting of the Contracting Parties to ACCOBAMS, Dubrovnik, Croatia 22-25 October 2007. 350 pp. ACCOBAMS (2010) Rapport de la quatrieme reunion des parties contractantes a l’ACCOBAMS. Monaco, 9-12 Novembre 2010. 399 pp. Agardy, T. (1997) Marine Protected Areas and Ocean Conservation. Academic Press and R.G. Landes Company, Austin. 244 pp.

114

ASCOBANS (2008) Small Cetaceans in the European Atlantic and North Sea (SCANS-II). Final Report, 15th ASCOBANS Advisory Committee Meeting Document AC15/Doc.21(S). UN Campus, Bonn, Germany, 31 March-3 April 2008. 54 pp. Baş, A. A., Erdogan, M. A., Morris, N. R. C., Yeoman, K., Humphrey, O., Gaggioli, E., Roland, C. (2016) Seasonal encounter rates and residency patterns of an unstudied population of bottlenose dolphin (Tursiops truncatus) in the northwestern Levantine Sea, Turkey. Hyla 2016(1): 1-13. Bassos-Hull, K., Perrtree, R., Shepard, C., Schilling, S., Barleycorn, A., Allen, J., Balmer, B., Pine, W., Wells, R. (2013) Long-term site fidelity and seasonal abundance estimates of common Bottlenose dolphins (Tursiops truncatus) along the southwest coast of Florida and responses to natural perturbations. Journal of Cetacean Research and Management 13: 19-30. Bearzi, G., Notarbartolo di Sciara, G., Politi, E. (1997) Social ecology of bottlenose dolphins in the Kvarneric (Northern Adriatic Sea). Marine Mammal Science 4: 650-668. Bearzi, G., Politi, E., Agazzi, S., Bruno, S., Costa, M., Bonizzoni, S. (2005) Occurrence and present status of coastal dolphins (Delphinus delphis and Tursiops truncatus) in the eastern Ionian Sea. Aquatic Conservation: Marine and Freshwater Ecosystems 15: 243-257. Bearzi, G., Agazzi, S., Bonizzoni, S., Costa, M., Azzellino, A. (2008) Dolphins in a bottle: abundance, residency patterns and conservation of bottlenose dolphins Tursiops truncatus in the semi-closed eutrophic Amvrakikos Gulf, Greece. Aquatic Conservation: Marine and Freshwater Ecosystem 18: 130-146. Bonizzoni, S., Santostasi, N.L., Würsig, B., Bearzi, G. (2014) Bottlenose dolphin abundance in the Northern Evoikos Gulf, Greece. In: Abstracts of the 28th Annual Conference of the European Cetacean Society 2014, 5-9 April 2014, Liege Belgium, p.104. Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., Borchers, D.L., Thomas, L. (2001) Introduction to Distance Sampling: Estimating Abundance of Biological Populations. Oxford University Press, Oxford, UK. 432 pp. Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., Borchers, D.L., Thomas, L. (2004) Advanced Distance Sampling. Oxford University Press, Oxford, UK. 416 pp. Buckland, S.T., Laake, J.L., Borchers, D.L. (2010) Double-observer line transect methods: levels of independence. Biometrics 66: 169-177.

115

Cañadas, A. (2006) Towards conservation of dolphins in the Alboran Sea. PhD Thesis - European Doctorate. Universidad Autonoma de Madrid. 329 pp. Cañadas, A., Burt, M. L., Macleod, K., Rogan, E., Santos, B., Vázquez, J. A. Hammond, P. S. (2010) CODA Cetacean Offshore Distribution and Abundance. Appendix II. Model-based estimates of cetacean abundance in offshore European Atlantic waters. Final report. 46 pp. Cañadas, A., Sagarminaga, R., García-Tiscar, S. (2002) Cetacean distribution related with depth and slope in the Mediterranean waters off southern Spain. Deep Sea Research I 49: 2053-2073. Cañadas, A., Sagarminaga, R., De Stephanis, R., Urquiola, E., Hammond, P. (2005) Habitat preference modelling as a conservation tool: proposals for marine protected areas for cetaceans in southern Spanish waters. Aquatic Conservation: Marine and Freshwater Ecosystems 15: 495-521. Casale, M., Milani, C., Kallianiotis, A. (1999) Preliminary survey on the interactions between local populations of Delphinus delphis and Tursiops truncatus and coastal fishery in north-eastern Aegean Sea (Greece). European Research on Cetaceans 13: 100. CIESM (2004) Investigating the roles of cetaceans in marine ecosystems. CIESM Workshop. Monograph n°25. Monaco, 144 pp. CIESM (2010) Marine Peace Parks in the Mediterranean – a CIESM proposal. Workshop Monographs, n°41. Siracusa (Italy) 18 - 20 November 2010, 128 pp. Clarke, K. R., Gorley, R. N. (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. Clarke, K. R., Warwick, R. M. (2001) Change in Marine Communities: An Approach to Statistical Analysis and Interpretation. 2nd edition. PRIMER-E Plymouth, UK. 172 pp. Coll, M., Pirrodi, C., Steenbeek, J., Kaschner, K., Lasram, F. B. R., Aguzzi, J., Ballestreros, E., Bianchi, C. N., Corbera, J., Dailianis, T., Danovaro, R. (2010) The biodiversity of the Mediterranean Sea: estimates, patterns, and threats. Plos One 5(8): e11842. Dawson, S., Slooten, E., Du Fresne, S., Wade, P., Clement, D. (2004) Small boat surveys for coastal dolphins: line-transect surveys for Hector’s dolphins (Cephalorhynchus hectori). Fisheries Bulletin 102(3): 441-451.

116

Dawson, S., Wade, P., Slotten, E., Barlow, J. (2008) Design and field methods for sighting surveys of cetaceans in coastal and riverine habitats. Mammal Reviews 38(1): 19-49. Dick, D. M., Hines, E. M. (2011) Using distance sampling techniques to estimate Bottlenose dolphin (Tursiops truncatus) abundance at Turneffe Atoll, Belize. Marine Mammal Science 27(3): 606-621. Fayos, J.A., Cañadas, A., Sagarminaga, R. (2000) A comparative approach of the movement patterns and orientation of common dolphins (Delphinus delphis) and bottlenose dolphins (Tursiops truncatus) in the north-eastern Alboran Sea. In: Proceedings of the 14th Annual Conference of the European Cetacean Society, Ireland, 2-5 April, Cork, pp. 130-133. Fayos, J.A., Cañadas, A., Sagarminaga, R. (2001) Habitat use and general behaviour analysis as a tool for the designation of marine protected areas. In: Proceedings of the 15th Annual Conference of the European Cetacean Society, Rome, Italy, 6-10 May 2001, pp. 223-224. Forcada, J., Gazo, M., Aguilar, A., Gonzalvo, J., Fernández-Contreras, M. (2004) Bottlenose dolphin abundance in the NW Mediterranean: addressing heterogeneity in distribution. Marine Ecology Progress Series 275: 275-287. Forcada, J., Hammond, P. S. (1998) Geographical variation in density and numbers of striped and common dolphins of the western Mediterranean. Journal of Sea Research 39(3-4): 31-325. Frantzis, A. 2009. Cetaceans in Greece: Present Status of Knowledge. Initiative for the Conservation of Cetaceans in Greece, Athens, Greece. 94 pp. Frantzis, A., Alexiadou, P., Paximadis, G., Politi, E., Gannier, A., Corsini-Foka, M. (2003) Current knowledge of the cetacean fauna of the Greek Seas. Journal of Cetacean Research and Management 5(3): 219-232. Gnone, G., Bellingeri, M., Dhermain, F., Dupraz, F., Nuti, S., Bedocchi, D., Moulins, A., Rosso, M., Alessi, J., Mac Crea, R. S., Azzellino, A., Airoldi, S., Portunato, N., Laran, S., David, L., Di Meglio, N., Bonelli, P., Montesi, G., Trucchi, R., Fossa, F., Wurtz, M. (2011) Distribution, abundance, and movements of the bottlenose dolphin (Tursiops truncatus) in the Pelagos Sanctuary MPA (north‐west Mediterranean Sea). Aquatic Conservation: Marine and Freshwater Ecosystems 21: 372-388. Hammond, P. S., Macleod, K., Berggren, P., Borchers, D. L., Burt, L., Cañadas, A., Desportes, G., Donovan, G. P., Gilles, A., Gillespie, D., Gordon, J., Hiby, L., Kuklik, I., Leaper, R., Lehnert, K., Leopold, M., Lovell, P., Øienm, N., Paxton.

117

C. G. M., Ridoux, V., Rogan, E., Samarra, F., Scheidat, M., Sequeira, M., Siebert, U., Skov, H., Swift, R., Tasker, M. L., Teilmann, J., Van Canneyt, O., Vázquez, J. A. (2013) Cetacean abundance and distribution in European Atlantic shelf water to inform conservation and management. Biological Conservation 164: 107-122. Hastie, T.J., Tibshirani, R.J. (1990) Generalized Additive Models. Chapman & Hall/CRC Monographs on Statistics & Applied Probability. 339 pp. Holt, R. D. (2003) On the evolutionary ecology of species’ ranges. Evolutionary Ecology Research 5: 159-178. Hoyt, E. (2011) Marine Protected Areas for Whales, Dolphins and Porpoises. A World Handbook for Cetacean Habitat Conservation and Planning. 2nd edition. Earthscan, London. 464 pp. Kallianiotis, A., Vidoris, P., Sylaios, G. (2004) Species assemblages and fishing geographical sub areas in North Aegean Sea, Greece. Fisheries Research 68: 171-187. Kotzageorgis, G., Mantzavelas, A., Hadjicharalambous, E., Defingou, M., Gioutlakis, M., Papagrigoriou, S., Alexandridou, E. (Eds.) (2015) Deliverable C7: New Natura 2000 areas or Natura 2000 sites proposal with digital or printed maps, filled in Standard Data Forms and new updated Natura 2000 database (Access 2003) with data from the new proposed areas. YΡEN, Athens, Joint venture of Enveco S.A., Omikron S.A., Karolidis Theodoros, Fyselias Spyridon, Consultant EKVY. 124 pp. MacLeod, K., Burt, M.L., Cañadas, A., Rogan, E., Santos, B., Uriarte, A., Van Canneyt, O., Vázquez, J.A., Hammond, P.S. (2010) CODA Cetacean Offshore Distribution and Abundance. Appendix I. Design-based estimates of cetacean abundance in offshore European Atlantic waters. Final report. 16 pp. MacLeod, C. D., Weir, C. R., Pierpoint, C., Harland, E. J. (2007) The habitat preferences of marine mammals west of Scotland (UK). Journal of the Marine Biological Association of the U.K. 87: 157-164. Milani, C. (2015) Cetaceans and their interactions with coastal fisheries in the Gulf of Kavala, Eastern Mediterranean Sea. PhD Dissertation at the University of Malta, Department of Biology, Research group of Conservation Biology. 384 pp. Milani, C., Vella, A., Vidoris, P., Christidis, A., Kallianiotis, A. (2011) Diet of five cetacean species in the North Aegean Sea (Greece) In: Abstracts of the 25th Conference of the European Cetacean Society, Cadiz (Spain), 21-23 March 2011, p. 247.

118

Milani, C., Vella, A., Vidoris, P., Christidis, A., Kallianiotis, A. (2012) Dolphin species’ diet, fisheries association and net damage in coastal fisheries of the north Aegean Sea (Greece). In: Abstracts of the 26th Conference of the European Cetacean Society. Galway (Ireland), 26-28 March 2012. p. 299. Milani, C., Vella, A., Vidoris, P., Christidis, A., Koutrakis, E. (2013) Cetacean distribution in the Thracian Sea (Greece) related with fishing activities. Abstracts of the 27th Annual Conference on the European Cetacean Society, Setubal (Portugal), 8-10 April 2013, p. 170. Milani, C., Vella, A., Vidoris, P., Christidis, A., Koutrakis, E., Kallianiotis, A. (2017) Cetacean strandings and diet analyses in the North Aegean Sea (Greece). Journal of the Marine Biological Association of the UK: 1-18. doi:10.1017/S0025315417000339. Milani, C., Vella, A., Vidoris, P., Christidis, A., Koutrakis, E., Kallianiotis, A. (Submitted). Abundance and distribution of the Common dolphin, Delphinus delphis, in the North Aegean Sea (Greece). Aquatic Conservation: Marine and Freshwater Ecosystems. Special Issue Common dolphins. Mitra, S., Koutrakis, E., Milani, C. (2001) Cetacean interaction with small-scale coastal fisheries: implications for conservation and damage limitation. In: Proceedings of the 15th Conference of the European Cetacean Society, Rome (Italy) 1, p.39. Mitra, S., Koutrakis, E., Clark, T., Milani, C. (2004) Cetacean interaction with small scale coastal fisheries: implications for conservation and damage limitation in the Northern Aegean, Greece. European Research on Cetacean 15: 172-176. Oedekoven, C. S., Mackenzie, M. L., Scott-Hayward, L. A. S., Rexstad, E. (2014) Statistical Modelling of Seabird and Cetacean data: Literature Review. Marine Scotland Science, Report 05/14, 31 pp. OCEANA. (2014) MPAs in the Mediterranean Sea. Addressing opportunities to face the protection gap. First working group on MPAs, SAC, SCMEE. Bar, Montenegro, 3rd February 2014, 39 pp. Panigada, S., Zanardelli, M., MacKenzie, M., Donovan, C., Mélin, F., Hammond, P. S. (2008) Modelling habitat preferences for fin whales and striped dolphins in the Pelagos Sanctuary (Western Mediterranean Sea) with physiographic and remote sensing variables. Remote Sensing of Environment 112: 3400-3412. Petroselli, A. (2006) Habitat use and distribution of short-beaked common dolphins Delphinus delphis around the island of Kalamos, Greece (eastern Ionian Sea). MPhill Thesis, University of Wales, Bangor, UK, 108 pp.

119

Pinn, E. H. (2010). Healthy and Biologically Diverse Seas Evidence Group Technical Report Series: Evaluation and gap analysis of current and potential indicators for Cetaceans. Joint Nature Conservation Committee. Peterborough 2010. 69 pp. Pirrotta, E., Matthiopoulos, J., MacKenzie, M., Scott-Hayward, L., Rendell, L. (2011) Modelling sperm whale habitat preference: a novel approach combining transect and follow data. Marine Ecology Progress Series 436: 257-272. Poulos, S. E., Drakopoulos, P. G., Collins, M. B. (1997) Seasonal variability in sea surface oceanographic conditions in the Aegean Sea (Eastern Mediterranean): an overview. Journal of Marine Systems 13(1-4): 225-244. Ryan, C., Boisseau, O., Cucknell, A.C., Romagosa, M., Moscrop, A., McLanagan, R. (2014) Comparison of cetacean distribution in the Aegean and Levantine Seas with ambient noise and shipping density. In Abstracts of the 28th Annual Conference of the European Cetacean Society, Liege, Belgium, 5-9 April 2014, p. 222. Scheinin, A.P. (2010) The population of bottlenose dolphins (Tursiops truncatus), bottom trawl catch trends and the interaction between the two along the Mediterranean continental shelf of Israel. PhD Thesis. University of Haifa, Israel. 142 pp. Schimek, M. G. (2000) Estimation and inference in partially linear models with smoothing splines. JSPI 91: 525-540. Shane, S.H. (1990) Behavior and Ecology of the Bottlenose Dolphin at Sanibel Island, Florida. In: The Bottlenose Dolphin (eds., S. Leatherwood, R. R. Reeves), San Diego Academic Press Inc., pp. 245-266. Silva, M.A., Prieto, R., Cascão, I., Seabra, M. I., Machete M., Baumgartner M. F. and Santos R. S. (2014). Spatial and temporal distribution of cetaceans in the mid-Atlantic waters around the Azores. Marine Biology Research 10(2): 123-137. Somarakis, S., Isari, S., Machias, A. (2011) Larval fish assemblages in coastal waters of central Greece: reflections of topographic and oceanographic heterogeneity. Scientia Marina 75(3): 605-618. Stockin, K.A., Weir, C.R.. Pierce, G.J. (2006) Examining the importance of Aberdeenshire (UK) coastal waters for North Sea bottlenose dolphins (Tursiops truncatus). Journal of the Marine Biological Association of the U.K. 87: 327-338.

120

Sylaios, G., Stamatis, N., Kallianiotis, A., Vidoris, P. (2005) Monitoring water quality and assessment of land-based nutrient loadings and cycling in Kavala Gulf. Water Resources Management 19(6): 713-735. Thomas, L., Buckland, S.T., Rexstad, E.A., Laake, J.L., Strindberg, S., Hedley, S.L., Bishop, J.R.B., Marques, T.A., Burnham, K.P. (2010) Distance software: design and analysis of distance sampling surveys for estimating population size. Review in Journal of Applied Ecology 47: 5-14. UNEP-MAP.-RAC/SPA. (2008) Procedure for the Revision of the Areas Included in the SPAMI List. Tunis, 18 pp. UNEP-MAP-RAC/SPA. (2010) Specially Protected Areas in the Mediterranean Assessment and Perspectives. (eds., Ben Haj, S., Ben Nakhla, L., Ouerghi, A., Rais, C.), Tunis, pp. 1-36. Vella, A. (1999) Cetacean research and conservation around the Maltese islands. European Research on Cetaceans 13: 274. Vella, A. (2005) Common dolphins (Delphinus delphis) status in the central and southern Mediterranean around the Maltese Islands. In: Common dolphins: Current Research, threats and issues. (eds., K. Stockin, A.Vella, P. Evans). ECS newsletter no. 45 - special issue July 2005, pp. 8-16. Vella, A. (2009) Sustainable Fisheries in the Mediterranean. Study and Publication commissioned by the Regional Activity Centre for Cleaner Production (CP/RAC), Barcelona, Spain (available from: http://www.cprac.org/en/downloads/documents/sector-studies). Vella, A. (2016) Resolving Bottlenose dolphin – fisheries association problems in Maltese waters, Central Mediterranean. Rapp. Comm. int. Mer Medit. 41:499. Viale, D. (1994) Cetaceans as indicators of a progressive degradation of Mediterranean water quality. International Journal of Environmental Studies 45 (3-4): 183-198. Viddi, F. A., Hucke-Gaete, R., Torrses-Florez, J. P., Ribeiro, S. (2010) Spatial and seasonal variability in cetacean distribution in the fjords of northern Patagonia, Chile. ICES Journal of Marine Science 67: 959-970. Wilson, B., Reid, R. J., Grellier, K., Thompson, P. M., Hammond, P. S. (2004) Considering the temporal when managing the spatial: a population range expansion impacts protected areas-based management for bottlenose dolphins. Animal Conservation 7: 331-338.