Assessment of the shery, growth and mortality characteristics of the cassava croaker, Pseudotolithus senegalensis (Sciaenidae) from coastal waters of Ghana

This paper reports on the shery and some population parameters of the cassava croaker (Pseudotolithus senegalensis) at Anlo Beach near Shama (5° 01' N, 1°37' W) in the Western Region of Ghana. The study was conducted from July 2012 to June 2013 to contribute to information on the dynamics of demersal stocks in the country. Beach seining was the prime shing method in the area with the shing season spanning July to March. Catch per unit effort (CPUE), determined as catch per net haul per day was 9.5 kg per haul in July 2012, 110 kg per haul in January-February 2013 and 47 kg per haul in March 2013. The species was absent from the shery from April 2013 to June 2013. Monthly catches varied from 2,295 kg to 31,500 kg, and specimens of the sh measured 7.5 cm TL to 105.4 cm TL and weighed 2.9 g to 7,634 g. The length-weight relationship of the population was described by the equation: BW = 0.005TL 3.04 (r = 0.99), and grouping the sh at 5 cm class intervals showed modal lengths at 15.0 – 19.9, 35.0 – 39.9 and 60.0 – 64.9 cm TL. Analysis of the length-frequency distribution using the ELEFAN programme indicated an asymptotic length (L ∞ ) of 110.3 cm TL, a growth coecient (K) of 0.2 yr -1 , and total (Z), natural (M) and shing (F) mortality coecients of 0.68yr -1 , 0.42 yr -1 , and 0.26 yr -1 respectively, for the population. Longevity of the sh was estimated to be 15 years. A low exploitation level of the stock is suggested by the calculated exploitation ratio, E= 0.38. These population characteristics are compared with those of other West African P. senegalensis populations.


Introduction
Croakers (Family Sciaenidae) are widely distributed along the west coast of Africa from Morocco to Angola (Paugy et al., 2003), and have also been reported from Namibia (Chao and Trewavas, 1990). The cassava croaker Pseudotolithus senegalensis is an economically important demersal sh in Ghana (Koranteng, 1997) as in the sheries of other West African countries, e.g. Cameroon (Djama and Pitcher, 1989), Nigeria (Anyanwu and Kusemiju, 1990) Benin (Sossoukpe et al., 2013a) and Liberia (Wehye et al., 2017). According to Nunoo and Nascimento (2015), P. senegalensis is the commonest sciaenid in the Gulf of Guinea representing 70% of the biomass of this group.
Stock assessment survey records show that the biomass of the Sciaenidae declined by 37% from 2000 to 2006, and mean body size of species has also declined from 50 cm TL in 1989-90 to [25][26][27][28][29][30] cm TL in recent times (Nunoo and Nascimento, 2015). Based on these records and other considerations across the West African sub-region, Nunoo (2015) and Nunoo and Nascimento (2015) suggested that P. senegalensis and P. typus be considered as threatened species. Information on current trends in biomass changes is however not available but reductions in landings and body size of Pseudotolithus spp. have been reported in Cameroon (Djama and Pitcher, 1989), Nigeria (Anyanwu and Kusemiju, 1990) and Benin (Sossoukpe et al., 2013a). Considering the high incidence of juvenile cassava croakers in beach seine landings from Cape Coast waters (Ghana), Blay, Awittor and Agbeko (2006) cautioned the likely negative effect of this shery on recruitment of juveniles into the adult stocks.
The study was conducted at Anlo Beach in Shama District (Ghana) where beach seine shing is the main livelihood of the people. Cast netting, gill netting, and use of crustacean pots are practiced on a smaller scale in the nearby River Pra Estuary. This study therefore examined the shery, growth, mortality and exploitation of P. senegalensis, the commonest species in the catches from the area to provide requisite information for managing the stock.

2.1Study area
The study was conducted at Anlo Beach (5° 01' N, 1°37' W), a small village located on a strip of land between the ocean and the wetland associated with the River Pra estuary near Shama in the Western Region of Ghana.

2.2Collection of shery data
Data on the beach seine shery for cassava croaker P. senegalensis were recorded monthly from July 2012 to June 2013. The beach seine nets used measured 100 m to 800 m in length, with about 70% of the 31 nets encountered measuring averagely 200m, and comprised 3 to 5 meshes with stretched mesh sizes ranging from 20 mm to 50 mm; the smallest of which is at the codend. The catch (kg) from the haul of a seine net (usually operated by a group comprising 25 to 55 persons depending on the size of the net), was recorded. Due to the variation in the sizes of the nets, catch per unit effort (CPUE) which was calculated as catch per haul was adjusted to catch per haul of a 200m net deployed. Since the management regime in the community allowed for deployment of consistently 11 nets per day although occasionally up to 13, the mean CPUE was calculated for 11 groups in a day. The total daily catch was determined as the sum of the amount of sh caught by all the seine net groups, and monthly sh catch as the total weight of sh caught in 26 shing days.

2.3Determination of growth and mortality parameters
Monthly samples of sh were obtained from commercial catches from July 2012 to March 2013, and specimens were measured for total length (TL) and body weight (BW) to the nearest 0.1 cm and 0.1 g respectively. The length-weight relationship of the species was established by the least-squares method, and monthly length-frequency distribution was analyzed at 5 cm class intervals. Estimates of the growth coe cientK) and the maximum theoretical length (L ∞) were obtained from the ELEFAN programme (Gayanilo et al., 2005) in the FiSAT II software. The parameter t 0 which refers to the age at which the length of the sh is zero (Gulland, 1983) was estimated from Pauly's (1983) equation: log 10 t 0 = -0.3922-0.2752 log 10 L ∞ -1.038 log 10 K Longevity (t max ) of the population was determined as t max = 3/K (Pauly, 1983), and growth performance index ( ) as log 10 K+2 log 10 L ∞ (Moreau et al., 1986). Total mortality (Z) of the population was determined from the catch curve, a plot of the natural logarithm of the numerical proportion of sh belonging to different ages caught by the shing gear against their relative ages. Natural mortality (M) was estimated from Pauly's (1980a) empirical equation: log 10 M = -0.0066 -0.279 log 10 L ∞ +0.6543 log 10 K + 0.4634 log 10 T, where T is the mean annual sea surface temperature of the study area. The shing mortality coe cient (F) was derived from the equation, Z = F + M (Ricker, 1975), and exploitation ratio (E) as E = F /Z (Gulland, 1971).

3.1The shery
Beach seining was the predominant shing method at Anlo Beach, a community located near the mouth of the River Pra estuary. Depending on net size, thirty-one groups each comprising 25 to 55 persons shed in the area, however ten to fteen groups, but regularly eleven groups operated each day as local regulations restricted the number of seine nets in operation at a time. Each group did one haul a day. Fishing occurred throughout the week except on Tuesday, the traditional rest day for shermen in the country resulting in about 26 shing days a month. The shing season commenced from July and ended in March after which shing was sporadic, by a few individuals for subsistence. Interaction with members of the community revealed this is an annual practice as most shers go into farming between April and June as an alternate livelihood activity because the low catches at the time make shing less pro table.
Monthly catches increased from 2,295 kg in July 2012 to a peak of 31,500 kg in January-February 2013 and declining to 11,464 kg in March (Fig. 1a). The CPUE also increased from 9.

3.2Length-weight relationship
The P. senegalensis specimens sampled measured 7.5 cm TL to 105.4 cm TL, and weighed 2.9 g to 7,634 g. The croaker population showed isometric growth (i.e. the coe cient b was not signi cantly different from 3.0) in the length-weight relationship described by the equation BW = 0.0054TL 3.04 (r = 0.99; t = 1.24, P > 0.05) (Fig. 2). Isometric growth is inferred if a sh increases proportionally in length, width and depth, while a disproportionate increase in these dimensions is indication of allometric growth (Pauly, 1984).

3.3Length-frequency distribution
The length-frequency distribution of P. senegalensis (Fig. 3) showed three size groups with modes at the15.0-19.9, 35.0-39.9 and 60.0-64.9 cm TL classes. Figure 4 shows the monthly length-frequency distribution of the species tted with the von Bertalanffy growth curve. In July, August and September 2012 the primary mode of the distributions was 35 cm TL.
This increased to 60 cm TL in January 2013 and was maintained in the February and March 2013 samples, suggesting that the samples comprised immigrants, and this may explain the departure of the modes from the growth curve during the period.

3.4Growth and mortality parameters
Analysis of the length-frequency data using the ELEFAN software gave growth parameters of the population as K = 0.2 yr -1 , and L ∞ = 110.3 cm TL. Substituting these parameters in Pauly's (1983) empirical equation resulted in an estimated t 0 of -0.59 yr. Hence on yearly basis, growth of the cassava croaker population may be described by the von Bertalanffy equation: where L t is the length of sh at age t (Fig. 5). The sh in this population were estimated to have a longevity (t max ) of 15 years.
The total mortality (Z) was detrmined only for sh that were fully expoited from the length-converted catch curve for the P. senegalensis population (Fig. 6). Regression analysis in the catch curve excluded data points on the ascending portion of the curve representing sh that were not fully recruited into the shery, and those on the descending part corresponding to sh in the older age groups with low occurrence in the population. Also precluded were from the analysis were the backward-projected points representing small size sh that would have been retained in the net in the absence gear selection and/or incomplete recruitment. The total mortality coe cient (Z) determined from the slope of the regression line for fully exploited sh was 0.68 yr -1 , and the natural mortality coe cient (M) calculated from Pauly's Therefore, the shing mortality coe cient (F = Z -M) was 0.26 yr -1 , and the resulting exploitation ratio (E) was 0.38.

Discussion
The beach seine shery at Anlo Beach in the Shama district of Ghana exploits several species of which the cassava croaker, Pseudotolithus senegalensis is the most important, economically. The close proximity of the stock to the River Pra estuary could be attributed to availability of its prey species, mainly sh and shrimps (Blay et al., 2006) in the nearshore waters. Furthermore, sciaenids are reported (Ssentongo and Ansa-Emmin, 1986) (Pauly et al., 2000), which supports the current low exploitation ratio of the P. senegalensis stock. This apparent low exploitation could be attributed to strict compliance of the local regulation whereby daily shing is rotated among the thirty-one shing companies which operate in the area.

Conclusion
From the foregoing it may be concluded that the Anlo Beach P. senegalensis population has better growth characteristics than other West African populations, perhaps due to its unique location near the River Pra estuary where food is likely abundant. The level of exploitation of the P. senegalensis stock is adequate to sustain the shery, hence, the current shing regime in the area should be maintained while encouraging voluntary compliance of local shery management regulations.  Figure 1 Monthly (a) total catch, and (b) mean CPUE of Pseudotholithus senegalensis at Anlo Beach