Population Dynamics, Fishery, and Exploitation Status of Norway Lobster ( Nephrops norvegicus ) in Eastern Mediterranean

: Norway lobster ( Nephrops norvegicus ) is one of the most valuable shellﬁsh resources in the Mediterranean and the northeast Atlantic. Sustainable management of its ﬁshery require long-term scientiﬁc monitoring of its stocks. Population dynamics, ﬁshery, and exploitation status of the Norway lobster were studied over a long time series (between 2008 and 2014) in a semi enclosed gulf in Easter Mediterranean (Pagasitikos Gulf). Sampling was carried out using a commercial bottom trawl with a codend mesh size of 28 mm in three different regions within the Gulf. Sex was identiﬁed macroscopically and all individuals were measured and weighed. Growth and Age groups were investigated by analysis of length–frequency distributions of the Carapace length (CL) for both sexes. Fishing, Natural and Total Mortality, along with Exploitation rate of the total stock were estimated by the FiSAT software package. Females were further examined and classiﬁed based on their maturity stage and the size at onset of sexual maturity was estimated by Probit analysis. Temporal differences in mean CL were detected for both sexes, with a tendency of lower values each year, along with a respective increase in Fishing Mortality and Exploitation Rate, indicating the effects of ﬁshing pressure over the population. Size at onset of sexual maturity was estimated at 32.6 mm, higher than the ofﬁcial mesh sizes for creels (28 mm) and gillnets (26 mm). According to the analyses, the local Norway lobster stock is over-exploited, which should be taken under consideration by the regional policy makers.


Introduction
The Norway lobster Nephrops norvegicus, Linnaeus 1758, (hereafter referred to as "Nephrops") is a common decapod crustacean inhabiting the muddy bottoms of the European continental shelves and slopes, at depths from 10 to 800 m. It is found throughout the northeast Atlantic and also in the western Mediterranean, the Adriatic, and the Aegean seas, with its distribution being limited by the extent of suitable muddy sediment in which animals construct their burrows [1].
Nephrops is fished wherever it is found in exploitable quantities and is one of the commercially most important demersal species in the north-east Atlantic [2]. Across Europe, it is considered an important target species in 34 Functional Management Units (FUs), with the UK and Ireland sharing the majority of landings from the North Atlantic stocks, followed by France, Iceland, Denmark, and Sweden [3]. Global capture production has reached a plateau and seems to fluctuate around 60,000 tonnes for the last 30 years [4].
Management of Nephrops stocks, just like all other fisheries in the European Union (EU), derives from a mixture of EU regulations (Common Fisheries Policy-CFP) and individual National laws. The proposed managing measures, such as the minimum landing size (MLS), total allowable catches (TACs) or maximum sustainable yield (MSY), are decided following biological scientific advices provided by the International Council for the Exploration of the Sea (ICES) [3,5,6]. The challenge which most management policies face is that, at least in the case of the Mediterranean Sea, they are not often successfully enforced and they are not revised according to the current state of stocks [7]. Since the marine ecosystems are constantly changing due to environmental factors and, mostly, pressure from human exploitation, fisheries and their stocks do not stay unaffected. In that context, fisheries management should not be based on just recent studies, but rather on long time series data in order to better understand the temporal changes in species abundance, distribution and biology [8].
Due to its importance and high economic value as a commercial species in the Atlantic and North Sea [2,3,9], Nephrops has been the subject of several biological and fishery studies , leading to the collection of long time series data for that region. In regard to the Mediterranean Sea, most of the studies have focused on the Western region and the Adriatic sea [7,8,13,15,[30][31][32][33][34][35][36][37][38], with the majority of them within the framework of the EU research project "MEDITS" [39], contrary to even fewer studies in the Eastern Mediterranean and the Aegean Sea [40][41][42][43][44]. The aim of the present study was to fill that knowledge gap in the Eastern Mediterranean, since no such studies were found for the region. The goals were to study the local stock and estimate its exploitation status over a long time series, in order to assess the potential impact of the fishery to the population dynamics of Nephrops.

Sampling Area
Pagasitikos Gulf is a semi-enclosed bay, with a maximum depth of approximately 100 m and is an area of high interest for Nephrops, as this is the shallowest occurrence (<100 m depth) of the species in high quantities, in Greek waters. Sediments below 60 m are dominated by soft clay muds allowing full development of sedimentary burrowing communities [40]. Trawling is prohibited inside the gulf due to an official ban, established in 1966, as a conservational measure. However, artisanal fishing with bottoms nets and baited creels is still allowed, with seasonal closures in June-August for bottom nets and May-July for baited creels [44].

Trawl Survey
The survey design consisted of three distinct quadrants: at the western, center, and eastern part of the gulf, each spanning over an area of approximately 12 km 2 (Figure 1), covering the range of previously recorded Nephrops burrow densities [40]. Samplings were carried for 2 consecutive days during the first week of June each year, in order to coincide with the maximum availability of female Nephrops on the sediment surface, which is estimated between April and August [41]. Every sampling day, three hauls were conducted, one in each quadrant, adding to a total of 42 hauls, over a 7-year period (2008-2014).
The bottom trawl was a traditional Greek commercial trawl with a codend mesh size of 28 mm (stretched). Speed of tow was approximately 2 knots. Starting position and depth were recorded when the trawl reached the bottom with wires fixed and then, position and depth were noted every 5 min until hauling. Hauling duration was approximately 30 min. Nephrops from each trawl haul were separated from the rest of the catch, placed in labeled polystyrene boxes packed with ice flakes and, at the end of the day, transported to the Laboratory for further processing.

Measurements
In the laboratory, samples were removed from the boxes, the ice flakes were washed away from the body and only individuals without missing parts were selected to be used for size and growth measurements. Wet body weight (BW) was measured to the nearest 0.01 g. Carapace length (CL) was measured with a digital caliper to the nearest 0.01 mm, as the distance between the post-orbital margin and the medial posterior border of the carapace. Individuals with CL < 20 mm (minimum landing size, MLS, Council Regulation (EC) No 1967/2006) were still included in the data and were considered as "recruits" [32].
Sex was determined macroscopically by observing the distinctive first pair of pleopods, which for males is harder and larger. Sex ratio (SR) was calculated as the proportion of females in the whole sample (SR = F/F + M) and tested for significant difference from the expected sex ratio (SR = 0.5, i.e., equal proportion of two sexes) using the chi-squared test.

Data Analysis
The null hypothesis of no significant differences in CL and BW values between sexes was tested using Student's t-test and since significant differences were detected, following the recommendation by ICES, further analyses were carried out separately for each sex [7]. Spatial (among regions) and temporal (among years) variations in CL and BW values were tested using one-way ANOVA, while individual group differences were assessed by the Fisher's LSD procedure. The Kolmogorov-Smirnov two sample test was used to detect any difference between female and male size distributions (DN, maximum distance between the cumulative distributions) and a Mann-Whitney test was used to compare medians (U statistic) [45]. The biometric relationship between CL and BW was estimated with the

Measurements
In the laboratory, samples were removed from the boxes, the ice flakes were washed away from the body and only individuals without missing parts were selected to be used for size and growth measurements. Wet body weight (BW) was measured to the nearest 0.01 g. Carapace length (CL) was measured with a digital caliper to the nearest 0.01 mm, as the distance between the post-orbital margin and the medial posterior border of the carapace. Individuals with CL < 20 mm (minimum landing size, MLS, Council Regulation (EC) No 1967/2006) were still included in the data and were considered as "recruits" [32].
Sex was determined macroscopically by observing the distinctive first pair of pleopods, which for males is harder and larger. Sex ratio (SR) was calculated as the proportion of females in the whole sample (SR = F/F + M) and tested for significant difference from the expected sex ratio (SR = 0.5, i.e., equal proportion of two sexes) using the chi-squared test.

Data Analysis
The null hypothesis of no significant differences in CL and BW values between sexes was tested using Student's t-test and since significant differences were detected, following the recommendation by ICES, further analyses were carried out separately for each sex [7]. Spatial (among regions) and temporal (among years) variations in CL and BW values were tested using one-way ANOVA, while individual group differences were assessed by the Fisher's LSD procedure. The Kolmogorov-Smirnov two sample test was used to detect any difference between female and male size distributions (DN, maximum distance between the cumulative distributions) and a Mann-Whitney test was used to compare medians (U statistic) [45]. The biometric relationship between CL and BW was estimated with the exponential model (BW = a × CL b ), using linear regression analysis, while a t-test was applied to check for allometry (H 0 :b = 3) [46].
Length-frequency distributions (LFD) were constructed per 1-mm size class for each sex [12]. In order to identify mean lengths for the age groups present in the length distributions, the Modal Progression Analysis (MPA) of the "FiSAT II" software package (version 1.2.2, FAO, Rome, Italy) was used, by applying the Bhattacharya and NORMSEP analyses on the constructed LFD. All identified age groups were taken into account for analyses, but only the ones with n > 30 and Separation Index (SI) > 2 were considered "representative" [32]. Mean lengths-at-age results were used to estimate the parameters of the von Bertalanffy growth function (VBGF), with a non-linear regression analysis [13].
Total mortality (Z, years −1 ) and natural mortality (M, year −1 ) were calculated as the respective average of Z and M values estimated by specific methods incorporated in FiSAT II (detailed explanation of the formulas, in Kevrekidis and Thessalou-Legaki [47]). Fishing mortality (F, year −1 ) was estimated as the difference between Z and M (F = Z − M), while the exploitation rate (E) was calculated as the ratio of fishing to total mortality (E = F/Z) [47].
Females were further examined and individuals were classified into 6 maturity stages, based on the macroscopic examination of the colors from the ovaries and the eggs, following the scale proposed by Mente et al. [42]: stage 1, white-immature; stage 2, green, expands up to the cephalothorax-maturing; stage 3, dark green, expands down into the abdomen-fully mature; stage 4, white-ovigerous female carrying green eggs on its pleopods; stage 5, white-ovigerous female carrying brown eggs on its pleopods; stage 6, white-eggs are released with brown filaments on legs of the adult females. For the estimation of the size at onset of sexual maturity (SOM), females from stages 3-6 were considered as "mature". A Probit analysis was used and a logarithmic curve was fitted to the data so that the CL value at which 50% of females are in reproductive condition (L50) for SOM could be estimated [42].

Population Structure
From the 42 hauls, the total number of measured Nephrops individuals was 2954 males and 4362 females (n = 7316). Trawling depth ranged between 50 and 70 m, with hauls within the West and East parts ranging between 50 and 60 m, while the Central part had a wider range, between 55 and 70 m. The West part of the gulf always yielded the lowest catch in the total number of individuals (312 males-429 females), followed by the Central region (1194 males-1766 females), with the Eastern part having the highest abundance (1448 males-2167 females). In terms of years, the lowest catch in absolute numbers was in 2008 (362 males-517 females) and the highest in 2010 (508 males-869 females). Considering the total biomass of the catch, the West part yielded the lowest (23.56 kg), followed by the Central (82.33 kg), whereas the Eastern part yielded a total 94.73 kg. In terms of years, the lowest value was 2011 (23.15 kg) and the one with the highest was 2010 (37.18 kg) ( Figure 2).
No spatial differences in CL were detected among the three regions, both for males (F 2, 2953 = 2.54, p = 0.0792) and females (F 2, 4361 = 1.35, p = 0.0533), whereas significant differences were detected in BW values, both for males (F 2, 2953 = 9.98, p < 0.05) and females (F 2, 4361 = 54.99, p < 0.05). On the contrary, temporal differences were detected for CL and BW values among the years of sampling, for both sexes (Figure 3). Mean size (CL ± SD) and weight (BW ± SD) of males was 34.89 ± 6.75 mm (range 14-64 mm) and 31.52 ± 18.78 g (range 3-125 g), while that of females was 32.79 ± 4.99 mm (range 14-53 mm) and 24.65 ± 10.68 g (range 3-89 g) respectively. Sex ratio was found to be significantly different from the expected equilibrium (0.50) in all samplings (p < 0.05), always in favor of the females (range 0.54-0.65). Females were more abundant in smaller CL size classes, but their proportion decreased significantly with the increase in CL size ( Figure 4). always in favor of the females (range 0.54-0.65). Females were more abundant in smaller CL size classes, but their proportion decreased significantly with the increase in CL size ( Figure 4).     The biometric relationship between CL and BW was negatively allometric in all cases, for both sexes, indicating faster axonometric growth (body length) than volumetric growth (body mass). High values of the correlation coefficient are indicators of a strong relationship between the variables, while low p-values of the lack-of-fit test indicated that the selected model appeared to be more accurate for the males rather than the females (Table 1).

Age Determination and Maturity
Length-frequency distribution (LFD) differed between sexes with males exhibiting a wider size range (14.08-63.40 mm) than the females (14.12-52.82 mm) (Kolmogorov-Smirnov DN = 0.168, p < 0.05), with also a significantly higher median size (34.56 mm) than The biometric relationship between CL and BW was negatively allometric in all cases, for both sexes, indicating faster axonometric growth (body length) than volumetric growth (body mass). High values of the correlation coefficient are indicators of a strong relationship between the variables, while low p-values of the lack-of-fit test indicated that the selected model appeared to be more accurate for the males rather than the females (Table 1).
In a few cases, 1-2 of the identified age groups were below the MLS threshold, but they were not considered as representative. The LFD of the whole stock (pooled data from all samplings) identified eight age groups for each sex, with five being representative for female and six for male Nephrops ( Figure 5) females (33.07) (Mann-Whitney U = 7.61 × 10 6 , p < 0.05). MPA estimations identified 6-8 age groups every year for both sexes, but usually only 4-5 were considered representative. In a few cases, 1-2 of the identified age groups were below the MLS threshold, but they were not considered as representative. The LFD of the whole stock (pooled data from all samplings) identified eight age groups for each sex, with five being representative for female and six for male Nephrops ( Figure 5) All maturity stages were observed ( Figure 6), with stages 2 and 3 being more prominent through all years, contrary to low numbers of berried females, except for 2012, when a large number of stage 4 females were captured. The estimated size at 50% of maturity was estimated from the pooled data of female Nephrops and was 32.6 mm, as estimated by Probit analysis (Figure 7). All maturity stages were observed ( Figure 6), with stages 2 and 3 being more prominent through all years, contrary to low numbers of berried females, except for 2012, when a large number of stage 4 females were captured. The estimated size at 50% of maturity was estimated from the pooled data of female Nephrops and was 32.6 mm, as estimated by Probit analysis (Figure 7).

Growth Parameters and Mortality
The yearly parameters of the VBGF and mortality values, for each sex are presented in Table 2

Discussion
Populations of Nephrops are known to differ in terms of distribution, density, or growth rates, even in small spatial scales. Such degree of variation could be attributed to the respective variability in several local environmental factors (e.g., temperature, hydro geography, and sediment particle size) [2,12,31]. In the case of Pagasitikos Gulf, popula-

Discussion
Populations of Nephrops are known to differ in terms of distribution, density, or growth rates, even in small spatial scales. Such degree of variation could be attributed to the respective variability in several local environmental factors (e.g., temperature, hydro geography, and sediment particle size) [2,12,31]. In the case of Pagasitikos Gulf, population density (in terms of total catch) varied between the different regions, with a distinct geographical pattern. Low densities were recorded in the West contrary to higher abundance in the Central-East region. This distinction coincides with the local water circulation pattern of Pagasitikos, which includes a cyclone in the West and an anti-cyclone in the East. This unique pattern seems to alter certain functions or processes with the local coastal ecosystems (e.g., nutrient flow, hydrodynamic regime, genetic flow, etc.), thus affecting the composition or structure of several coastal communities in the area [48][49][50][51]. In the case of Nephrops, local hydrographic features could interfere with the settlement of post larval juveniles or change sediment composition, thus effecting burrow densities [1,40,52].
Reproductive cycle of Nephrops varies throughout the year, seemingly affected by factors such as geographic distribution, depth, water temperature, even annual weather patterns and climate [53]. In the Mediterranean Sea, developmental rate and incubation period were found to be approximately 6 months [33] contrary to the Irish Sea, where they were found around 10 months [10] and the Icelandic waters, where they were found to be over 14 months [23]. During their incubation period, females are less active, spending more time in their burrows, thus making them harder to catch [11], and thus making the sampling of females less representative of the actual stock [7]. This might be the case in several studies [7,11,12,19,20], where more males where captured than females, probably because samplings were carried out during winter. In our study, the choice to schedule the trawling surveys early in June, lead to a sex ratio of over 50%, in favor of females, every year. This is an important finding for managing authorities to consider when planning new policies, but also as a reference for future research studies of Nephrops, especially in Eastern Mediterranean.
Growth rates, estimated via the von Bertalanffy Growth Function (VBGF), were found to be slightly different between sexes, which is to be expected, since Nephrops males grow to larger size but in slower rates [2]. Values of K (growth rate) and CL inf (maximum length) vary strongly among several studies in the Mediterranean and the North Atlantic [7,[12][13][14]20,25,30,32], with no apparent geographic pattern. Most studies estimated growth rate of Nephrops between 0.1 and 0.2 years −1 , which comes to an agreement with the values obtained in our study. Spatial and temporal variation in growth can be attributed to the different environmental factors such as population density, hydrological conditions, sediment type [12,31] or fishing pressure [54]. It is worth noting that the very high growth rates estimated in Portugal [20] were attributed to the low fishing pressure on Nephrops in the area which positively affected local population density.
Age estimation within a stock is an essential resource in the effective management of commercial species, in conjunction with the estimation of size at the onset of sexual maturity (SOM), which is particularly important in understanding the reproductive strategy and fitness of a species as a determinant of reproductive output [16]. SOM was estimated 32.6 mm which falls within the value range reported in other studies [20,30,32,55]. By assuming that the first representative mode of the MPA groups corresponds to an absolute age of 3.5 years, the estimated SOM of this study is estimated between 5 and 6 years, which is similar to other studies [2,11,13,30,32,56,57]. Since mesh sizes in fishing gear are set within the framework of an effective management policy by official bodies (e.g., EU, Central Government, Local Authorities), in order to reduce the capture of immature individuals [52], it is important to note that the regional legal minimum mesh size for creels (28 mm) and gillnets (26 mm) for Nephrops fisheries in Greece [44] is lower than the estimated SOM of this study. This difference should be put under consideration by the regional policy makers and stakeholders in Nephrops fishery, since it could pose as a potential threat in the long-term viability of the stocks. In general, size-frequency growth models have proven useful in several analyses in crustaceans, even if they have some limitations [58]. They usually are more reliable when applied on short-lived species, because specimens of older age with slower growth rates might be grouped with younger ones. Especially in Nephrops, after the onset of sexual maturity, growth rate slows down and the identification of age cohorts becomes difficult [2]. Nevertheless, these models have proven to be very convenient, easy to apply and require very little resources for the estimation of the age structure of species which lack "age-recording" structures (e.g., otoliths); therefore, they will "remain the prime source of information on longevity in decapod crustaceans, at least until something better comes along" [58].
Very few studies have focused on the exploitation of the Nephrops fishery by estimating natural and fishing mortality [7,30,35,55]. In general, natural mortality values are lower than the respective values for females, usually because males are captured in larger quantities than females due to the cryptic behavior of the latter, but also owning to the differences in the methods used to calculate mortalities [35,47,59]. In the case of the Nephrops fishery in Pagasitikos, fishing mortality was increasing every year with the respective increase in the exploitation rate (E), for both sexes. On the other hand, average values of Carapace length (CL) for both sexes were decreasing, which is an indication of fishing pressure [35]. Furthermore, values of E were above 0.5, with an increasing trend, reaching up to 0.8 for males and 0.7 for females, marking the local fishery as over-exploited. Even though, total catch (in terms of number of individuals or total biomass) did not fluctuate much between the sampling years, this should not be considered as a sign of potential resilience of the species to overexploitation. The biology of Nephrops (burrowing behavior, absence of juveniles in catches, wide trophic spectrum) might explain this resistance but does not change the fact that Nephrops is a sensitive species to trawl disturbance, and thus management and conservation measures should be taken to minimize the impact on fisheries and stocks from trawl gear, mostly, but also other fishing gear, in general [7,17].
In conclusion, the results from our study identified temporal and spatial changes in the population dynamics of Nephrops in the Pagasitikos gulf, which can be related to fishing pressure. Still, in order to improve the tools for better stock assessment and to provide better insight for the management and conservation of Nephrops fisheries, more focused research is required in the future. Effort should be directed in the better understanding of the life-cycle of the species in local and regional scale, as well as improving the available tools for data collecting by integrating new technologies and non-destructive methods (e.g., ROV imaging and GIS mapping).