AGE, GROWTH, MORTALITY, REPRODUCTION, AND EXPLOITATION RATES FOR FISHERY MANAGEMENT OF GREY MULLET SPECIES IN THE KÖYCEĞIZ LAGO- ON–ESTUARY (MEDITERRANEAN COAST)

Background. The Mugilidae is a widely distributed family in the tropical, subtropical, and temperate waters. These fish species have a global economic value because of the high quality of their flesh and caviar. This study provides new data on the age, growth, and reproduction parameters of commercially exploited grey mullets from the Köyceğiz Lagoon, Turkey, namely golden grey mullet, Chelon auratus (Risso, 1810), leaping mullet, Chelon saliens (Risso, 1810), thicklip grey mullet, Chelon labrosus (Risso, 1827), and flathead grey mullet, and Mugil cephalus Linnaeus, 1758. The obtained results are intended for fisheries management of golden mullets in the area. Materials and methods. A total of 1195 fish specimens were collected from the Köyceğiz Lagoon (south-western Anatolia). The mullets were caught monthly, between January 2017 and December 2017, using fish barrier, trammel net, beach seine, and cast-net. The fish age was determined from sagittal otoliths. Growth parameters were determined by applying the von Bertalanffy growth function. Reproduction period, mortality and exploitation, relative yield per recruit (Y′/R), and biomass per recruit (B′/R) were determined. Results. The most frequent mullet age groups were 3+ and 4+ (for M. cephalus and C. labrosus) and 4+ (for C. auratus and C. saliens). The following von Bertalanffy’s growth models were calculated: Lt = 58.78(1 – e –0.163(t


INTRODUCTION
The Köyceğiz Lagoon (south-western Anatolia) is one of the most important active lagoon fishing areas in Turkey. It covers 5400 ha of open water and 1150 ha of marsh delta and is connected to the sea thorough a 14-km long canal. The width of the canal varies between 5 and 70 m and its depth between 1 and 6 m (Buhan 1998). Grey mullets (Mugilidae) are the most important commercial fish species in the Köyceğiz Lagoon. There are five grey mullet species in the Köyceğiz Lagoon, namely: golden grey mullet, Chelon auratus (Risso, 1810), leaping mullet, Chelon saliens (Risso, 1810), thicklip grey mullet, water bodies (Arruda et al. 1991, Hotos et al. 2000, Hoşsucu 2001a, Fazli et al. 2008a, Kraljević et al. 2011, Saoudi and Aoun 2014, Tulkani 2017, Panda et al. 2018.
For the sustainable management of fish stocks, information is needed on their age and growth, mortality, and exploitation rates. This study provides new data on selected biological parameters of commercially caught grey mullets in the Köyceğiz Lagoon required for proposing some targeted reference points for its management.

MATERIAL AND METHODS
The fish samples were collected monthly using a fish barrier, trammel net, beach seine, and cast-net in the Köyceğiz Lagoon, Turkey between January 2017 and December 2017 (Reis and Ateş 2019) (Fig. 1). The fish samples were brought to the laboratory and were taxonomically identified according to Thomson (1997). Total length (TL) was measured to the nearest 0.1 cm, and body weight (W) was determined with a precision balance (0.01 g). The sex of all specimens was recorded by macroscopic examination of the gonads as female, male, or immature. The sex ratio of the studied grey mullet species was analyzed using the Chi-square test (χ 2 ).
For aging, the gill cavity of the fish was opened and the otoliths were removed with forceps and cleaned from waste materials in Petri dishes containing 90% ethyl alcohol. The otoliths were then stored in numbered Eppendorf tubes for age determination. After marking the centers of otoliths under the microscope, they were broken from the marked places using thumb and index fingers (Skurdal et al. 1985). Broken otoliths were burned in a spirit stove until they were brown (Aprahamian 1988). For the age determination, the burned otoliths were placed on the tack it with their broken surfaces facing up and glycerin was dropped to reveal the age rings (Fig. 2) and examined under a stereomicroscope (Christensen 1964). Otoliths of each fish were read 3 times by the researchers, and reading for a given fish otolith was accepted only when 2 readings agreed.
Growth parameters were investigated by applying the von Bertalanffy (1938) growth function as follows where L t is the length at age t, W t is the weight at age t, L ∞ is asymptotic length, W ∞ is asymptotic weight, K is the growth coefficient, and t 0 is the hypothetical age at which length is equal to zero.
The growth performance index (Ø′), to compare the growth parameters obtained in the presently reported study with those reported by other authors for the same species, was calculated by the equation of Pauly and Munro (1984) Beverton and Holt's (1956) equation to obtain the total mortality coefficient Z as where Ĺ is the mean length of fish of length Ľ and longer, while Ľ is the lower limit of the length class of highest frequency.
The natural mortality coefficient (M) was calculated using the formula of Djabali et al. (1993) as where L ∞ is the asymptotic length and K is the growth coefficient.
The fishing mortality coefficient (F) was computed as while the exploitation rate E was computed from the formula of Gulland (1971) In this study the relative yield per recruit (Y′/R) and relative biomass per recruit (B′/R) models, developed by Beverton and Holt (1966) and incorporated in FISAT II software (Gayanilo et al. 2005), were used to evaluate the stock of grey mullets.
The gonadosomatic index (GSI) was calculated monthly following the formula of Avşar (1998) where W G is the gonad weight, and W T is the total fish weight.

RESULTS
As of 2003, the amount of fishing has changed between 169-633 tons per year in the last fifteen years and has been determined as mean 348 tons per year in Köyceğiz lagoon. Based on the fishing amounts of the Köyceğiz Lagoon in the last 15 years, mullet fishing has the highest ratio with 85.9%. This is followed by eel fishing with a rate of 1.2% and sea bass fishing with a rate of 0.9%. However, eel fishing has decreased considerably in recent years and it is determined as 0.06% in 2017. Sex ratio. During the sampling period, 1195 individuals were collected, in this number 476 (39.8%) representing Chelon auratus, 291 (24.3%) M. cephalus, 279 (23.3%) Chelon labrosus, and 149 (12.5%) Chelon saliens. Female:male ratios of C. auratus, M. cephalus, C. labrosus, and C. saliens were 1:0.60, 1:0.47, 1:0.58, and 1:0.52, respectively. The χ 2 test revealed that there were significant differences between the female and male for sex ratio of all studied species (χ 2 = 28.38, df = 1, for C. auratus; χ 2 = 28.38, df = 1, for M. cephalus; χ 2 = 28.38, df = 1, for C. labrosus; χ 2 = 28.38, df = 1, for C. saliens; P < 0.05). Age and length composition. It was determined that the age composition of C. auratus, M. cephalus, C. labrosus, and C. saliens individuals ranged within 0 + -5 + , 0 + -7 + , 0 + -6 + , and 0 + -5 + , respectively. The most frequent age groups were 4 + (for C. auratus and C. saliens) and 3 + and 4 + (for M. cephalus and C. labrosus). The mean total length of C. auratus, M. cephalus, C. labrosus, and C. saliens was determined as 27.9, 30.2, 25.1, and 24.0 cm, respectively. The mean length, mean weight, number of fish, and the standard deviations corresponding to the age groups of C. auratus, M. cephalus, C. labrosus, and C. saliens are given in Table 1. Growth parameters. The constants of the von Bertalanffy's growth model were calculated (Table 2) yielding the following equations for growth in length and weight: C. auratus L t = 58.78(1 -e -0.163(t + 0.0195) ) W t = 1501.20(1 -e -0.163(t + 0.0195) ) M. cephalus  The total length and weight values are mean ± standard deviation; n = the number of fish sampled.

Reproduction.
In this study, the monthly mean values of the gonadosomatic index of female individuals for C. auratus, M. cephalus, C. labrosus, and C. saliens were calculated. The highest value amounting to 13.85 was found for C. auratus in December (14.9°C). It was followed by 13.46 for M. cephalus in July (30.0°C), 7.70 for C. labrosus in January (13.1°C), and 10.85 for C. saliens in May (24.0°C). The above-mentioned values suggest that the spawning periods of C. auratus, M. cephalus, C. labrosus, and C. saliens extend from October to January, from June to September, from December to March, and from April to July, respectively (Fig. 3). L ∞ = asymptotic length, W ∞ = asymptotic weight, K = growth coefficient, t 0 = hypothetical age, Ø′ = growth performance index, Z = total mortality, M = natural mortality, F = fishing mortality, E = exploitation rate, L c = length at first capture, E max = maximum sustainable level of exploitation, E 0.1 = the level of exploitation at which the marginal increase in yield per recruit reaches 10%, E 0.5 = the exploitation level that will result in a reduction of the unexploited biomass by 50%.

DISCUSSION
The annual total catch efficiency of the Köyceğiz Lagoon was between 26-97 kg per ha per year and its mean value for the last fifteen years was 53 kg per ha per year (based on face to face interview). In a study carried out by Buhan (1998) in the Köyceğiz Lagoon, the reported catch efficiency values were between 27-80 kg per ha per year. The mean catch efficiency of the Homa Lagoon was found as 20.83 kg per ha per year by Acarlı (unpublished * ). In a different study, carried out in the Muni Lagoon, the catch efficiency was reported as 125-250 kg per ha per year by Koranteng et al. (2000). The total of the lagoons of Turkey has been reported as 20-50 kg per ha per year, while in other Mediterranean countries it reached 56 kg per ha per year (Crivelli 1992  Sex ratio. The female:male ratios of the presently reported study agree with the results reported in the Gulf of Gabes for C. auratus (see Abdallah et al. 2013); in the Homa Lagoon for M. cephalus (Acarlı unpublished * ), while the ratio of females was lower in the Sinop-Samsun coast of the Black Sea for C. auratus (see Bilgin et al. 2006); in the Homa Lagoon for C. saliens (see Acarlı unpublished * ) and in the Homa Lagoon for C. labrosus (see Akyol 1999). El-Zarka and El-Sedfy (1970) reported that the sex difference was due to the age and size of maturity. Also, Brusle (1981) reported that conditions such as heat and cold resistance and breeding migrations affect the female:male ratio in a population.
Age and length composition. Bilgin et al. (2006) reported  Table 3. Some differences were observed in age groups of the species under study when compared to previous researches. These differences may be due to the sampling method, fishing activity, feeding habitats, population density, and the ecological conditions of water bodies. Growth parameters. The value of L ∞ for C. auratus, determined in presently reported study, was smaller than that from the Caspian Sea (Fazli et al. 2008a) and higher than that from the Mirna estuary (Kraljević et al. 2011). The L ∞ value of M. cephalus obtained in this study was smaller than that from the Bardawil Lagoon (El-Ganainy et al. 2002) and higher than that from the Bafra fish lakes (Yılmaz and Polat 2011). Moura and Gordo (2000) reported a smaller L ∞ value for C. labrosus, whilst Richter (1995) reported higher L ∞ value compared to the presently reported study. Balık et al. (2011) reported a smaller L ∞ value for C. saliens in the Beymelek Lagoon compared to this research. The value of W ∞ was found 292.26 g for C. auratus in the Bitter Lakes (Mehanna, 2004), in contrast the value of W ∞ was found 1501.20 g in this study. Ibáñez Aguirre et al. (1999) reported a higher W ∞ value for M. cephalus compared to the present study. Koutrakis and Sinis (1994) reported a smaller W ∞ value for C. labrosus and C. saliens compared to the presently reported study. The growth coefficient values (K) of studied species were generally lower than compared to the results of different authors (Table 4). The mean growth performance index (Ø′) value of C. auratus, M. cephalus, C. labrosus, and C. saliens was reported as 2.693, 2.996, 2.799, and 2.540, respectively (Ibáñez 2016). The growth parameters (L ∞ , K, t 0 ) and growth performance index (Ø′) obtained in this study are also compared by different researches in the other water bodies (Table 4). Ma et al. (2010) reported that different age compositions may be causes of differences of the estimated parameters in different study areas. Kennedy and Fitzmaurice (1969) reported that the different growth coefficients found in different regions were due to differences in water temperature and this is because grey mullets spend most of their lifespan in shallow inshore waters, where the temperature is influenced more by local conditions than by temperature of the open sea, which is more stable.
Reproduction. The spawning periods of studied species are in agreement with other studies on the spawning periods of these species in different areas (Hoşsucu 2001b, Patimar 2008, Abdallah et al. 2013. In this study, spawning periods of the studied species were compared to the other researchers in Table 5. We assume that the physical parameters of the water, which differ from region to region, affect the spawning periods, which are found different from the results of this study. Sagi and Abraham (1984) reported that the water temperature and salinity effect reproduction periods. Whereas, Brusle (1981) reported that grey mullets reproduction in different geographic regions at different times of the year. Mortality and exploitation rates. Fishing mortality (F) and natural mortality (M) contribute to the total mortality (Z). According to Barry and Tegner (1990), the predominance of growth on mortality can be perceived by the ratio Z:K being lower than 1; a ratio higher than 1 means that the stock is collapsing; if the ratio is equal to 1, the population is in a steady state and if this proportion is much higher than 2, the stock is overexploited. The ratio Z:K was 5.03 for C. auratus, 5.56 for M. cephalus, 4.46 for C. labrosus, and 3.62 for C. saliens and these results show overexploited of the studied species in the Köyceğiz Lagoon. The exploitation rates calculated in the presently reported study agree with the previous studies (Buhan 1998, Mehanna 2004, Hotos et al. 2019. Gulland (1971) reported that the rate of exploitation for the fish stock should be 0.5 (F = M). According to this result, it is inevitable that the fish stocks have a fishing pressure in the Köyceğiz Lagoon and that stocks will reach the level that will be exhausted. For the sustainable management of the grey mullet stocks in the Köyceğiz Lagoon, some of the mature grey mullets that enter the fish barriers should be left to the sea.  C. auratus (L c = 28.00 cm), M. cephalus (L c = 26.86 cm), and C. labrosus (L c = 23.73 cm) was smaller than the length of first sexual maturation (L m = 34.0 cm, L m = 35.4 cm and L m = 29.5 cm, respectively, Froese and Pauly 2019). Due to the harvesting pre-spawning fishes, a greater reduction may be in the fishing in the near future. For sustainable grey mullet fishing, it is of great importance to give each fish a chance to reproduction at least once in its lifecycle, and therefore the length at first capture (L c ) should be bigger than the length at first sexual maturation (L m ). Presently reported study n = number of fish sampled, , L ∞ = asymptotic length, K = growth coefficient, t 0 = hypothetical age, Ø′ = growth performance index, C = calculated from the L ∞ and K values of the published data. The relative yield per recruit (Y′/R) analysis results for grey mullet species in the Köyceğiz Lagoon has shown that additional fishing effort would provide very little additional catch, this means no economic return. Also, the results of biomass per recruit (B′/R) analysis showed that the increase in exploitation rate causes a sharply declined in Biomass per recruit (B′/R). It could be concluded that the grey mullet stocks are in a situation of overexploitation in the Köyceğiz Lagoon. For the management implications of the assessment, the present level of exploitation rate should be decreased by about 47.5, 49, 45.9, and 41.45 percentage points for C. auratus, M. cephalus, C. labrosus, and C. saliens, respectively to maintain sufficient spawning biomass for recruitment. This can be realized by reducing the number of fishing days and allow some of the captured fish to be released from the barriers and migrate to the sea.

CONCLUSION
As a result, growth parameters provide some indication of resource utilization and the effectiveness of management strategies. When age and growth were evaluated in combination, it can be easier to understand the relation between population size and biomass. This understanding is the basis of modern fisheries resource allocation and management. Fisheries management should be designed on biological data to understand the status and to manage fish stocks. The Köyceğiz Lagoon is an important fishing area in Turkey. This study provides information related to age, growth, mortality, reproduction, and exploitation rates of the grey mullet species from the Köyceğiz Lagoon. The results of the study may be used for fisheries researches, management, and conservation in the Köyceğiz Lagoon. In addition, due to activities such as fishing pressure, environmental pollution, and tourism intense, fisheries management policies should be implemented to ensure optimum and sustainable use of the Köyceğiz Lagoon immediately.