Species composition, seasonality and biological characteristics of Western Ghana’s elasmobranch fishery

https://doi.org/10.1016/j.rsma.2022.102338Get rights and content

Highlights

  • Prionace glauca and Raja parva were the dominant shark and ray species, respectively.

  • The greatest abundance and species richness for all the sites was documented in the minor rainy season (August to November).

  • There was variation in sizes on the common shark and ray species landed.

  • Sexual segregation was typical among most of the common elasmobranch species landed.

Abstract

Despite the fact that elasmobranchs are now targeted in artisanal fisheries in the wake of continuous declines of teleosts in Ghana, data on any aspects of these fisheries are poor. This study aims to document the spatial and temporal variation, and biological composition of elasmobranchs in five key elasmobranch-dominated fishing communities in Western Ghana, which include Adjoa, Axim, Busua, Dixcove and Shama. A total of 2157 elasmobranchs comprising 1414 sharks and 743 rays, belonging to 34 species in 15 families, were recorded over the course of 9 months. Prionace glauca (81.2%) and Raja parva (29.3%) were the dominant shark and ray species, respectively. Shama had the highest mean shark catch (33.3 ± S.D 29.7), followed by Dixcove (19.1 ± 14.7) and Axim (18.2 ± 12.4). However, Axim exhibited the greatest mean ray catch (25.8 ± 10.6), followed by Adjoa (10.9 ± 3.8) and Busua (9.6 ± 5.4). The greatest relative catch and species richness for all the sites was documented in the minor rainy season (August to November). Sexual segregation was typical among most of the common elasmobranch species landed. This study serves as a baseline for monitoring future changes in the artisanal shark and ray fisheries in Western Ghana.

Introduction

Elasmobranchs (sharks and rays) are among the most threatened of the world’s vertebrate groups (Bräutigam et al., 2015, Dulvy et al., 2021a, Dulvy et al., 2021b). Elasmobranchs are characterized by long generation times, late sexual maturity, low reproductive and slow growth rates (Walker, 1998, Dulvy and Forrest, 2010, Dulvy et al., 2017), and, as such, these life history traits make them particularly vulnerable to overfishing and population decline (D’Alberto et al., 2019, Dulvy et al., 2014). Other characteristics such as site fidelity, low global abundance, complex migration patterns and spatial segregation of sexes can also increase the vulnerability of elasmobranchs to overfishing (Heupel and Simpfendorfer, 2005, Sims, 2005a, Sims, 2005b). As apex predators in marine ecosystems, they facilitate nutrient cycling and serve as indicators of ocean health (Burkholder et al., 2013). Thus, the loss of elasmobranchs could have negative cascading impacts for ecosystem function and structure, with consequences across multiple spatiotemporal scales (Dulvy et al., 2000, Temple et al., 2019).

Species responses to changes in their environment are important in understanding their movement, distribution and abundance in marine habitats. This is increasingly important given the growing human-induced environmental impacts on marine ecosystems globally. For example, abiotic factors are known to be important drivers of shark and ray movement, which invariably influence their distribution (Schlaff et al., 2014), while temperature influences shark and ray populations at a variety of temporal scales (Schlaff et al., 2014). Movement to locate a spatially preferred water temperature range has been linked to foraging and reproductive strategies among sharks and rays (Sims et al., 2006, Speed et al., 2012, Thums et al., 2013), and both severe weather events and seasonal temperature changes influence shark movement (Carlisle and Starr, 2009, Matich and Heithaus, 2012). For instance, the bat ray Myliobatis californica, leopard shark Triakis semifasciata, and brown smooth-hound shark Mustelus henlei were observed emigrating from Tomales Bay, California in response to a seasonal decrease in water temperature (Hopkins and Cech, 2003). Further, Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c observed variations in catch composition between spring and summer which was linked to temporal temperature differences in the Gulf of California.

Elasmobranch fisheries provide employment, sustenance and income through the sale of meat and fins for some of the poorest rural coastal communities (Bonfil, 1994). The high demand and price of shark fins and meat are the major drivers for the proliferation in shark fisheries across the world (Fowler et al., 2005). Artisanal elasmobranch fisheries have increased in size and geographical extent in recent years, which has exerted pressure on elasmobranchs through both target and incidental catches, resulting in notable population declines (Haque et al., 2021). These threats facing elasmobranchs are further exacerbated by the continuous depletion of their prey species and the degradation and pollution of their habitats (Field et al., 2009, Gallagher et al., 2012). However, very little is known about how these effects are manifested in many regions, particularly in West Africa, which is facing dramatic marine ecosystem degradation, and where data collection and monitoring of the marine environment is lacking (Dulvy et al., 2017).

In Ghana, elasmobranch landings began in the late 1950s and peaked in 1975 with 11,478 tons (FAO, 2017). In the last decade, the total reported elasmobranch catches fluctuated considerably, increasing intermittently on average by 2,000 tons per year and peaked up to 10,104 tons in 2013 (Fig. 1), contributing up to 31.4% of the reported Eastern Central Atlantic catch (FAO, 2021). In 2019, however, the catch estimate trends indicate a sharp decline of sharks landed in Ghana to an average of 1,292 tons (Fig. 1), which comprises 9.3% of the total elasmobranch catch in Eastern Central Atlantic (FAO, 2021). This trend corroborates a recent study and report from fishers in Ghana that indicates a sharp decline in elasmobranch catches, which suggests the elasmobranch fishery is overexploited (Agyeman et al., 2021; Seidu et al. under review). The decline in elasmobranch stocks may impede the livelihoods of dependent coastal communities in Ghana. This also has implications on food security, as a significant number of people depend on sharks and rays for their protein requirements in the wake of continuous declines of teleosts (MoFAD, 2015, Seidu et al., 2022). Thus, there is a pressing need to devise immediate and effective strategies to support the sustainable management of Ghana’s elasmobranch fisheries. However, the lack of key baseline information on species-specific catch and size structure of elasmobranchs hampers the development of such management strategies. The only available fisheries data reported by the Ministry of Fisheries Scientific Survey Division to the Food and Agriculture Organization (FAO) is taxonomically aggregated as sharks and rays; a reporting scheme that gives little information about species composition and ignores the distinction between sizes of shark catches, and hinders the evaluation of trends in catches and abundance of species (FAO, 2017). Improving fisheries catch and size composition data are essential in developing evidence-based management strategies to safeguard the future sustainability of elasmobranch fisheries and the livelihoods of dependent Ghanaian coastal communities.

Accordingly, this study aims to improve fisheries catch data on elasmobranchs by investigating species-specific spatial and seasonal variability, as well as the size and sex composition of common elasmobranch species in landings in Western Ghana. Three research questions were addressed regarding elasmobranchs landed in the various fishing communities: (i). How does elasmobranch landed and species richness differ among the various fishing communities in Western Ghana? (ii). How does elasmobranch landed and species richness vary among seasons in Western Ghana? (iii). What are the various sizes and sex structures of the most frequently landed species in Western Ghana?

Ghana is a West African nation bordered by Burkina Faso to the north, Republic of Togo to the east, Republic of Côte d’Ivoire (Ivory Coast) to the west, and Gulf of Guinea to the south. Ghana’s coastline is approximately 550 km long and is interspersed with about 90 lagoons and associated wetlands. Ghana’s coastal zone covers 6.5% of the land area, which is inhabited by a quarter of the population (deGraft-Johnson et al., 2010). The marine resources of Ghana include over 347 fish species, belonging to 82 taxonomic families (deGraft-Johnson et al., 2010).

A total of 186 fishing villages and 302 landing beaches have been recorded in Ghana during the 2013 national marine canoe framework survey (MoFAD, 2013). However, there are 76 fishing communities in Western Region (MoFAD, 2013), the focus area of this study. We identified a total of 13 fishing communities landing elasmobranchs in the study region; however we conducted the study in only five coastal communities namely, Axim, Adjoa, Busua, Dixcove, and Shama (Fig. 2). The communities were chosen based on three main reasons: (i) fishing is exclusive to artisanal fishers, although there are also industrial fleets that catch sharks and rays in Ghana; (ii) sharks and/or rays formed a significant component of the artisanal catch; and (iii) fishers were willing to cooperate with the researchers for both landing and interview data. These communities fall within three administrative districts, namely Nzema East Municipality, Ahanta West District and Shama District (Table 1). These communities are located along the West Coast (95 km), which extends from the Ghana-Côte d’Ivoire border to the Ankobra River estuary.

Western Region is approximately 21,391 km2, about 10% of Ghana’s total land area, with a 202 km long coastline (deGraft-Johnson et al., 2010). The coastline of Western Ghana is characterized mainly by stone sea walls and gabions, curved sandy beaches interspersed with nearshore rocky bottoms. Coastal materials are generally heterogeneous, which are characterized by coastal rock types ranging from granitoids, shales, sandstones and soils (Boye, 2015). The climate of the region shows an average significant wave height of about 1.2 m and a relative period ranging between 10 and 12 s (Boye, 2015). The prevailing wave direction approaches the Ghanaian coastline from south–south west with a local sea level rise rate of about 3 mm/year (Appeaning-Addo et al., 2008). The region is characterized by relatively low tides with neap and spring tidal ranges recorded as 0.53 m and 1.3 m respectively (Wiafe et al., 2013) and a mean tidal range of 0.6 m (Boye, 2015). The climate of the region is described as equatorial with significant variation in spatial distribution in precipitation. Western Region is also the wettest part of the country with a mean annual rainfall of 2,083 mm. The mean annual temperature range along the coast is 26 °C to 28 °C, but shows strong seasonal differences in August (21 °C to 22 °C) and 24 °C to 28 °C in April. The marine environment is characterized by two seasonal upwelling events, namely the major and minor upwelling. The major upwelling occurs from either late June or early July to late September or early October, while a minor upwelling occurs between January and March. These upwelling events have a considerable influence on both the local and sub-regional fisheries (Boye, 2015).

The gear types predominantly used in the study communities include drift gillnets, bottomset gillnets, longlines, handlines, trolling lines, purse seine nets and ring nets. Sharks and rays are mostly caught with two gear types; drift gillnets complemented with longlines and bottom-set gillnets, respectively. Fishers in Shama, Dixcove and Axim mostly use drift gillnets, while fishers in Adjoa and Busua predominantly use bottomset gillnets for their fishing operations (Seidu per. Obs.). Both types of gears are made of monofilament fishing line. The gears are deployed and retrieved manually by the fishers.

The drift gillnet gear is classified into two sizes, namely large and small. Fishers using the drift gillnet gears mostly fish in oceanic habitats where they target pelagic species such as sharks, sail fish, tuna, dolphin fish and sword fish, amongst others. The large drift gillnet gears normally range from 100–120 m in length and 4–8 m in depth. It has several mesh sizes which comprise either 181/2, 151/2, 151/4, 121/2 and 121/4 inches. The small drift gillnet gears range from 80–100 m in length and 3.5–4.6 m in depth. These mesh sizes comprise either 121/4, 91/2 and 91/4 inches.

The size of the bottomset gillnets range from 90–180 m in length and 1.3–2.8 m in depth. The net is made of polyethylene material which can be joined together to get any required length and depth. It has mesh sizes of 91/4, 91/5, 92/5, and 91/2 inches. The bottomset gillnets are used in coastal habitats and to target demersal fish such as rays, guitarfish, cassava fish and anchovy, amongst others.

Wooden canoes are the only vessels used by artisanal fishers in Ghana. Artisanal elasmobranch fishers in the study communities use three types of canoe; large, medium and small. The large (16–25 m long and 2–4 m wide) and medium canoe (9–15 m long and 1–2 m wide) are predominantly used in Axim, Dixcove and Shama, while the small canoes (4–8 m long and 1–2 m wide) are mostly used in the Adjoa and Busua communities. Canoe crews range from four to eight fishers for the large and medium operated canoes and three to six fishers for the small canoes. Fishing distances from landing sites to fishing grounds range from 129–290 km for large motorized wooden canoes (in most cases not exceeding 177 km) and from 32–80 km for medium-sized canoes. The large canoes are well equipped for fishing trips lasting 5–8 days. The medium canoes are made up from a single dug out log from the wawa tree, Triplochiton scleroxylon. The medium-sized canoes are equipped for fishing trips lasting between two and four days. Both large and medium sized canoes operating with drift gillnets are equipped with a water tank, flash light or reflectors, Global Position System (GPS) units, a compass, fire extinguisher, tool box and life jackets for crew members. The canoes mostly use outboard motors with an engine capacity of 15 Horse Power (HP), 25 HP, 30 HP, or 40 HP. Details of the fishing operations and typology of the various types of canoe mostly used in the fishing communities of Ghana are provided in Table 2.

Tri-weekly data collection was conducted between February 2020 and October 2020, by the first author and trained local volunteers. Data collection was conducted randomly between 7:00 and 17:00 and lasted between six and eight hours per visit. Sharks and rays were sampled for 28 days in the minor rainy season (August–October), 45 days in major rainy season (May–July) and 35 days in the dry season (February–April). Specimens encountered were identified to the lowest possible taxonomic resolution using keys from Séret (2006), Compagno (2002) and Séret (2016); Batoid in Food and Agriculture Organization catalogue). At each landing site in this study, the total elasmobranch catch was recorded and summed by season and by community to determine the variation in species composition across landing sites. The total length (TL) to the nearest cm was measured as the distance between the tip of the snout and the tip of the upper caudal fin in a relaxed position for sharks, guitarfish and wedgefish, and disc width (DW) for all other rays. Sex, and in some cases maturity stage, of males were recorded based on the presence/absence of claspers, which are visible from an early stage of development on the inside edge of the pelvic fins (Capapé and Zaouali, 1994). Clasper length and degree of calcification was measured as an indicator of sexual maturity in males (Jabado et al., 2016) and were only reported for the most frequent species landed. Specimens were considered as immature if claspers were short and flexible. Males characterized by partially calcified claspers were categorized as immature, while specimens exhibiting elongated and calcified claspers were considered fully matured adults. It was not possible to determine the maturity stage of female specimens.

Species relative catch (RC) in the study sites was obtained first by combining all of the minor rainy, major rainy, and dry season records of each sampling sites, to explain variation in elasmobranch assemblages according to the five sampling sites (Axim, Adjoa, Busua, Dixcove, and Shama communities) using the formula below: Relativecatch(RC)=Total number of catch per elasmobranch speciesTotal number of catch of all elasmobranch species sampled×100%

Species richness was estimated simply as the number of species observed in the various communities or seasons. The mean shark or ray catch per community was computed as the total number of shark or ray species landed in each community divided by the number of days visited. Variability in species richness and catch among the various communities were tested using a Kruskal–Wallis test and further compared with Bonferroni pairwise tests if there were significant differences among the communities.

Shark and ray species richness and catch were compared across seasons. Similarly, mean shark or ray catch per season was computed as the total number of shark or ray species caught during a specific season divided by the number of days visited or monitored during that season. To correct for unequal sampling among the seasons, we employed the Margalef index (D) of species diversity (Colwell and Coddington, 1994). The Margalef index (D) includes the total number of individuals (N), and estimates the total number of species (S) for a given number of individuals (Margalef, 1968): D=S1InNWe computed D for the various seasons and used this as our estimate of local biodiversity (Marshall et al., 2006). We transformed (using log, square root) the dataset for the various seasons. A one-way ANOVA test was then used to test for significant differences for D among the seasons, since a priori test for normality and equality of variance, using Shapiro–Wilk test (Zar, 2010) and Bartlett’s Test (Arsham and Lovric, 2011) respectively, showed that the dataset was normally distributed and of equal variance. However, a Kruskal–Wallis test was applied to test for differences in shark and ray catch among the seasons, due to the non-normal nature of the catch dataset. Jaccard’s similarity index was employed to determine the similarity percentages of species composition shared between the various landing sites (Magurran, 2003).

All measured specimens were used to determine size compositions and sex ratios for each species landed. For all species with more than 50 measured specimens (regarded as common species), parametric or non-parametric approaches were applied to examine the potential differences in size composition between the sexes. We applied Shapiro–Wilk and two tailed variance ratio (F) tests to test for normality and equality of variance, respectively, of the size data (Zar, 2010). When the test revealed that the data were normally distributed and of equal variance, two-tailed t-tests were employed to test the hypothesis that mean sizes of females and males did not differ significantly at an alpha level of 0.05 for the species. Size data that did not meet these assumptions were evaluated using two-tailed non-parametric Mann–Whitney U tests. Additionally, Chi-square statistics with Yates correction for continuity (Zar, 2010) was used to test the hypothesis of parity in the sex ratios of all common species at a 95% confidence level. All statistical tests were performed using PAST version 3.12 (Hammer et al., 2001).

All shark and ray specimens examined in this study were landed by artisanal fisheries in the study communities and were already dead upon inspection. Permission to conduct the survey was granted by the Western Regional Fisheries Commission and the Department of Wildlife and Range Management of the Kwame Nkrumah University of Science and Technology, Ghana. No other authorization was required to undertake this study in Ghana.

A total of 2,157 elasmobranchs, comprising 1,414 sharks and 743 rays, belonging to 34 species in 15 families, were recorded during the 108 survey visits (Table 3). The 34 recorded species comprised 20 shark and 14 ray species belonging to seven shark and eight ray families. Carcharhinidae (n=8 species) exhibited the most diverse shark family that were landed, while the families Dasyatidae and Mobulidae, with three species each, represented the most diverse ray families. The family Carcharhinidae dominated landings (89.4% of all shark catch), followed by family Lamnidae, comprising 5.7% of all shark catch. Among the rays, Dasyatidae (30.8%), Rajidae (29.3%), and Mobulidae (27.2%) were the most important families across the study communities. The most dominant shark species was Prionace glauca, with an estimated 81.2% relative catch across the study communities, followed by Isurus oxyrinchus (4.7%) and Carcharhinus brevipinna (2.8%). Rarer species such as Negaprion brevirostris and Rhincodon typus were recorded only once in the study communities (Table 3). The most dominant ray species were Raja parva, which constituted 29.3% of the relative catch, followed by Fontitrygon spp. (which comprised both Fontitrygon margaritella and Fontitrygon margarita, 27.3%), and Mobula tarapacana (18.3%) (Table 3).

From the 34 species recorded, Shama and Axim exhibited the highest number of shark species, with 17 species each (Table 3). No shark species were recorded in Busua. Axim and Adjoa were the communities richest in ray species with 11 species each, followed by Busua (nine ray species). Shama displayed the lowest ray species richness with only two species recorded (Table 4). There was variability in the number of shark and ray species landed among the communities. Shama had the highest estimated mean shark catch (33.3 ± S.D 29.7), followed by Dixcove (19.1 ± 14.7) and Axim (18.2 ± 12.4). Axim exhibited the greatest mean ray catch (25.8 ± 10.6), followed by Adjoa (10.9 ± 3.8) and Busua (9.6 ± 5.4) (Fig. 3). There was a significant difference in the catch landings of shark species (K=11.37, p=0.009) among the study communities. Adjoa community differed significantly in the Bonferroni pairwise comparison with Axim (p=0.001), Dixcove (p=0.013) and Shama (p=0.002). Conversely, pairwise tests between Axim and Dixcove (p=0.450) and Shama (p=0.201), as well as Dixcove and Shama (p=0.967) did not differ significantly. However, there was no statistically significant difference in the ray landed among the study communities (K=4.24, p=0.359).

Axim and Shama had the greatest similarity in shark species composition with 79% similarity between these communities. Dixcove was similar to Axim and Shama communities in terms of shark species composition (Ds = 63%) (Table 4). Similarly, shark community composition of Adjoa was slightly more similar to Axim (Ds = 24%) than it was to Dixcove (Ds = 29%).

For the ray species, there were only two records in Shama and hence this study site was not included in the final analysis of the Jaccard’s similarity index. Adjoa and Busua had the greatest similarity in ray species composition (Ds = 64%). Adjoa was much more dissimilar to Dixcove in terms of species composition: (Ds = 32%). Conversely, the community composition of rays in Dixcove was slightly more similar to Busua (Ds = 42%) than it was to Axim (Ds = 40%) (Table 4).

The minor rainy season exhibited the highest shark species richness (D=2.91), followed by major rainy (D=2.44) and dry season (D=1.99). Conversely, the dry season had the greatest estimated ray species richness (D=2.02) compared to the minor rainy (D=1.93) and major rainy season (D=1.87). Despite these intriguing trends in the data, there was no significant difference in the estimated species richness of shark species (F=0.48, df = 2, p=0.63) and ray species (F=1.96, df = 2, p=0.18) across seasons.

The minor rainy season supported the highest shark catch (34.1 ± S.D. 26.7). It was followed by the dry season (21.1 ± 18.8) and major rainy season (15.6 ± 11.4) (Fig. 3). There was significant variation in the catch landed of shark species across seasons (K=5.83, p=0.05). Bonferroni pairwise comparisons revealed a significant difference between the minor rainy and dry season (p=0.013). The major rainy season did not differ significantly in the pairwise comparison with the minor rainy season (p=0.237) and dry season (p=0.295).

Conversely, ray species catch was highest in the dry season (23.1 ± 9.9) as compared to the minor rainy season (21.2 ± 8.7) and major rainy season (8.7 ± 3.6) (Fig. 3). There was no significant difference in ray species catch among seasons (K=2.68, p=0.07).

Shark catch in the landing sites differed seasonally by community. Shama had higher shark landings in the minor rainy season (n=356) compared to the major and the dry season combined (Fig. 4). Conversely, Dixcove exhibited the greatest number of shark landings in the dry season (n=161). Only four shark specimens were recorded in Adjoa in the major rainy season, while no specimens were recorded in the minor rainy and dry seasons. Ray species landings were similar in the minor rainy (n=173) and dry seasons (n=172) in Axim community, but very low in the major rainy season (n=42) (Fig. 4). Adjoa and Busua had the highest number of ray landings in the dry season, while Dixcove and Shama exhibited the greatest shark catch in the minor rainy season.

Noteworthy shark species occurrence included one record of N. brevirostris at Dixcove in the minor rainy season; one record of R. typus, which was only found in Shama in the minor rainy season; and three records of L. nasus which was only encountered in the minor rainy season in Axim. The ray species with the lowest record was Pteroplatytrygon violacea, which was only recorded in Adjoa and Busua in the minor rainy and dry season, respectively.

All the shark and ray species recorded were below their maximum known total length and disc width, respectively, with the exception of M. thurstoni (Table 5). The majority of specimens of P. glauca and I. oxyrinchus were greater than 180 cm TL. These specimens were identified as mature based on clasper calcification and recorded total lengths which were above the reported size of sexual maturity for the respective species (Table 5).

Sampled landings of 1,098 specimens of P. glauca were dominated by females, representing 57% of the total landings (Table 5). Specimens ranged from 21–360 cm TL, and the average size of males (231.4 ± 45.3) and females (221.1 ± 39.1) was significantly different in the landings (Mann–Whitney​ U=1.202 E05, p < 0.0001). The sex ratio of females to males (1.3: 1) departed from 1:1 parity (χ2 = 20.49, p < 0.0001). There was a concentration of female and male specimens in the 200–219 cm, 220–239 cm and 240–259 cm size classes. Significant differences in size class distribution of P. glauca was observed (χ2 = 46.94, df = 17, p < 0.0001), with most of the larger size classes comprising females. Both mature female (89%) and male (92%) specimens had higher sizes than the minimum reported size at maturity. However, based on clasper calcification, only 63% of male specimens were matured.

A total of 65 specimens ranging from 70–343 cm TL were documented in the various landing sites (Table 5). The largest specimens were represented by females, while males constituted the smallest specimens. Although not significant differences (χ2 = 0.75, p = 0.385) the observed sex ratio of females to males (1.2: 1) was, however, larger than the 1:1 parity of sexes. Mean size of females (191.2 ± 52.4) was greater than that of males (172.1 ± 50.9), but did not differ significantly (t=1.48, p=0.143). Female specimens were common in the 180–199 cm TL size class, while the male size frequency distribution was bimodal, concentrated at 180–199 cm and 200–219 cm TL size classes (Fig. 5). Similarly, size class distributions did not differ significantly between females and males (χ2 = 8.29, df = 11, p = 0.688). Most male specimens (69%) were larger than the minimum reported size at maturity. However, based on clasper calcification, 76% of male specimens were matured.

The observed size range of 214 specimens of R. parva was 21–63 cm DW. Female R. parva outnumbered males (1.6: 1), comprising 62% of the total number of specimens recorded, which significantly departed from the 1:1 sex ratio (χ2 = 11.68, p < 0.0001). Likewise, the mean size of females (31.5 ± 9.0) was significantly larger than males (28.2 ± 4.1) (t=3.102, p=0.002). Both females and males of 20–29 cm DW size class were common in landings (Fig. 6). The female size class distribution differed significantly from that of the males (χ2 = 17.34, df = 4, p=0.0017). Mature male (38%) specimens had higher sizes than the minimum reported size at maturity. However, 52% of male specimens were matured based on clasper calcification.

A total of 156 specimens of Fontitrygon spp. contained significantly more females than males (3.2: 1) (χ2 = 43.10, p < 0.0001). Landings constituted 76% females, ranging from 17–52 cm DW and representing the smallest and largest specimens (Table 5). Average size of specimens did not differ significantly between females (26.1 ± 6.3) and males (26.7 ± 7.2) (Mann–Whitney U=192, p=0.233). Most specimens of both females and males were distributed in the 20–29 cm DW size class (Fig. 5). The female and male size classes was similar in the frequency distribution (χ2 = 2.241, df = 4, p = 0.691).

The size composition of 139 specimens of M. tarapacana ranged from 38–114 cm DW, with males representing both the smallest and largest specimens (Table 5). Landings comprised a higher proportion of females (73%) but consisted largely of specimens between 220 and 239 cm DW (Fig. 6). However, size class distributions were significantly different between female and male specimens (χ2 = 23.591, df = 14, p = 0.049). The average size of specimens did not differ significantly between females and males (Mann–Whitney U=1904, p=0.960). The observed sex ratio of females to males (2.7:1) differed significantly (χ2=28.55, p < 0.0001). Both mature female (24%) and male (55%) specimens had higher sizes than the minimum reported size at maturity. However, based on clasper calcification, only 46% of male specimens were matured.

This fishery-dependent assessment is the first comprehensive species-specific study of Ghanaian elasmobranchs. It provides a major increase in our understanding on temporal and spatial variation, as well as species and size composition of both sharks and rays in the artisanal elasmobranch fisheries of Western Ghana. This is despite its limitations in terms of spatial coverage compared with other recent studies on elasmobranchs in the West Africa sub-region (e.g., (Diop and Dossa, 2011)). Carcharhinidae was the dominant shark family in terms of species diversity and abundance in catch landings. Several studies have demonstrated that the family Carcharhinidae is prevalent in artisanal elasmobranch fisheries, with certain species within this family dominating shark landings in the sub-Saharan African region (Robinson and Sauer, 2013, Kiilu et al., 2019), the Arabian Gulf (Jabado et al., 2014), Red Sea (Spaet and Berumen, 2015), and in North America (Beerkircher et al., 2002, Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c, Cartamil et al., 2011). The most abundant species documented in this study, comprising Prionace glauca, Isurus oxyrinchus, and Carcharhinus brevipinna, are widespread in the sub-Saharan Africa region and dominate the landings in maritime countries where they occur (Temple et al., 2019). Dasyatidae was the dominant ray family in landings, which corresponds with other studies in the sub-region (Moore et al., 2019). Further, both Carcharhinidae and Dasyatidae are among the most speciose families, and thus, may often be the dominant families mostly targeted. The low number of Rhincodon typus and Negaprion brevirostris landed may be due to these species being migratory in nature (Fowler, 2014) and thus only occurring in Ghanaian waters during certain times of the year. This current study has confirmed the presence of 34 elasmobranch species in the Ghanaian artisanal fishery, with sharks comprising higher species richness than the rays. This finding corroborates with Temple et al. (2019) who identified a total of 29 shark and 23 ray species across Kenya, Zanzibar, and northern Madagascar. Conversely, a higher species diversity of rays compared to sharks was documented in artisanal fisher landings in Madagascar (Robinson and Sauer, 2013).

The difference in shark landed among the study communities may be partially explained by the variability in fishing grounds utilized by fishers in these communities. Most fishers in Dixcove, Axim, and Shama use large canoes with modern equipment such as GPS, which aid them in travelling further distances to target sharks in the oceanic habitat. This likely explains the higher catch of sharks recorded in these two communities. In comparison, fishers in Busua and Adjoa tend to fish in the coastal zone and mostly embark on short fishing trips. The difference in fishing between coastal and oceanic environments means that these areas exhibit different habitats and environmental conditions, and thus attract different species assemblages. Fishing methods or gears used may also play a key role in the differences in species richness among these communities. It was observed that fishers in Shama, Dixcove, and Axim mostly used drift gillnet gears and hence targeted sharks and other pelagic species, resulting in shark diversity and abundance being higher in these communities. Conversely, most fishers in Adjoa and Axim utilized bottomset gillnets with the aim of targeting demersal fishes; thus leading to high numbers of rays in the total catch in these two communities.

In contrast to studies in Sonora, Mexico    (Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c) and in the Arabian region (Henderson et al., 2007), seasonal variations in the number of shark and ray species (species richness) in this study was not evident. However, the abundance of shark and ray landings varied seasonally, which is consistent with studies in the Arabian Gulf (Jabado et al., 2014, Spaet and Berumen, 2015). The variations in shark catch between the minor rainy season and dry season may be partially explained by the higher fishing effort in terms of number of fishing trips and distance covered by fishers within the minor rainy season. Further, the minor rainy season coincides with the bumper harvest season, whereby offshore upwelling occurs, leading to nutrient rich waters which favour high diversity and abundance of elasmobranch species and other sympatric fish species. Fishers in Ghana are reported to harvest both small and large pelagic fishes, whose numbers peak during the offshore upwelling periods (Boye, 2015); hence the high catch rate of sharks during this season.

The data confirmed sex- and size-based differences among some common elasmobranch species in the artisanal catch landings. Sexual segregation has been reported in sharks and rays (Sims, 2005a, Sims, 2005b, Cartamil et al., 2011) and was evident for some shark and ray species in this study. The sex ratio of P. glauca differed significantly and echoes the findings of studies in North America (Beerkircher et al., 2002, Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c, Cartamil et al., 2011). Among the primary shark species, more female P. glauca and I. oxyrinchus were evident in the catch landings. In contrast to this finding, more male P. glauca (Beerkircher et al., 2002, Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c) have been observed in landings in North America. Ray species common in the artisanal catch landings in this study included Fontitrygon spp. (encompassing Fontitrygon margaritella and F. margarita), M. tarapacana, and R. parva, all of which had significant differences in sex ratios with females being dominant in the landings. Species in the family Dasyatidae have been reported to undergo seasonal movements, and are especially abundant in embayments during warm water periods, often for breeding and nursery functions (Villavicencio-Garayzar, 1993, Smith et al., 2007). Furthermore, during warm periods gravid females of Dasyatidae tend to aggregate in nearshore and insular waters for more protracted periods to pup and breed (Márquez-Farías, 2007, Bizzarro et al., 2009c) and may be targeted by fishers operating in these habitats, thus contributing to their high number in catch landings. However, caution needs to be taken in interpreting sexual segregation of data from elasmobranch landings, as differences may be a result of a number of confounding factors such as gear bias and natural segregation, which have been also reported for various elasmobranch species (Sims, 2005a, Sims, 2005b, Bizzarro et al., 2009a, Bizzarro et al., 2009b, Bizzarro et al., 2009c).

Different sizes of sharks were observed among species common in the artisanal catch landings in this study. In contrast to Moore et al. (2012), landings of I. oxyrinchus, and P. glauca consisted of several medium to large specimens larger than 100 cm TL. None of the specimens of common shark species landed approached the maximum reported sizes according to the IUCN Red list (2021). However, most of the shark species landed comprised several size classes, which were larger than the minimum size at sexual maturity. Similarly, size classes of rays exploited in the study communities varied among species. Landings of M. tarapacana largely comprised relatively small individuals, which were lower than the reported minimum sizes of maturity (Marshall et al., 2019a, Marshall et al., 2019b). Landings of M. thurstoni were composed of several large specimens, which were larger than the smallest size at sexual maturity. Specimens of M. thurstoni, however, exceeded the currently published maximum size for this species (220 cm DW; (Marshall et al., 2019a, Marshall et al., 2019b)). Similarly, many specimens of R. irvinei comprised larger sizes than the minimum reported size of sexual maturity (Séret and Valenti, 2009). A specimen of R. irvinei of 98 cm TL almost approached the maximum reported size for this species, which is 100 cm TL (Séret, 2006, Séret, 2016).

The study found that P. glauca and R. parva were the most abundant shark and ray species landed, respectively. The number of shark and ray species landed varied among the study communities, with Shama exhibiting the highest mean shark abundance. The minor rainy season exhibited the highest shark species richness and abundance, while the ray species abundance was highest in the dry season. All shark and ray species observed in landings were below their maximum reported total length and disc width, with the exception of M. thurstoni. Furthermore, there was sexual segregation among most of the common shark and ray species, with females dominating the landings.

The general lack of information on species-specific elasmobranch catch in the sub-region makes it difficult to ascertain the current status of elasmobranch populations and the extent to which they have declined, which calls for further investigation. Although the information from this study does not detail the status of elasmobranch stocks, they do shed light on artisanal elasmobranch fisheries and provide baseline data that are indispensable in future monitoring and for the sustainable management of elasmobranch stocks in Ghana. Further, information on the spatial, seasonal, and size composition of elasmobranch fauna are useful for informing future decisions regarding size and catch limits, particularly for the most threatened species, in addition to informing the design of spatially and/or temporally restricted marine protected areas.

Despite the significant contribution of elasmobranch fisheries in Ghana, there are no specific shark fishery regulations for the sustainable exploitation of these species in national waters. Although the Food and Agriculture Organization’s International Plan of Action for the conservation and management of sharks calls on all maritime states to develop their own tailored National Plan of Action (Fowler et al., 2005), there has not been any efforts in Ghana to develop such regulations, owing to the general lack of species-specific data.

Of the 20 shark and 14 ray species that were recorded in this study, 18 shark (90%) and 7 ray (50%) species are classified as threatened (i.e., Critically Endangered, Endangered or Vulnerable) on the IUCN Red List of Threatened Species (IUCN, 2021), and as such face a high to extremely high likelihood of extinction in the wild. Furthermore, C. carcharias, C. longimanus, M. tarapacana, M. thurstoni and R. typus are all listed on Appendix I of the Convention on the Conservation of Migratory Species of Wild Animals (CMS). The CMS text and Appendices are legally binding on CMS Parties, and considering that Ghana is signatory to this Convention, these species should thus receive strict protection in Ghana. Furthermore, all these species, in addition to A. superciliosus, A. vulpinus, G. cemiculus, I. oxyrinchus, I. paucus, S. lewini, S. mokarran and S. zygaena are all listed on CITES Appendix II, and therefore international trade is restricted on these species. However, implementation of these conventions is impeded by the general lack of species specific information and identification.

The fact that over 70% of the shark and ray species recorded in this study were of global conservation concern and were listed as threatened on the IUCN Red List of Threatened Species (IUCN, 2021), and that the presence of CMS Appendix I-listed species were recorded in artisanal fisher catches, is concerning, and highlights the need for their adequate protection in Ghanaian waters. This study provides a precursor for the development of such policy and further serves as a baseline for monitoring future changes in artisanal shark fisheries in Ghana. In order to assess the regional conservation status of shark and ray species, future research should focus on the life history traits of the various species recorded. Further, steps should be taken to invest in education and research, capacity building and awareness creation regarding the sustainable exploitation of elasmobranchs in Ghana.

Section snippets

CRediT authorship contribution statement

Issah Seidu: Conceptualization, Investigation, Methodology, Formal analysis, Funding acquisition, Writing – original draft, Writing – review & editing. David van Beuningen: Supervision, Methodology, Writing – original draft, Writing – review & editing. Lawrence K. Brobbey: Supervision, Data curation, Writing – review & editing. Emmanuel Danquah: Supervision, Data curation, Writing – review & editing. Samuel K. Oppong: Methodology, Writing – review & editing, Supervision, Formal analysis.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Special thanks go to the Zoological Society of London EDGE of Existence Program team members for their training, advice and mentoring. Our heartfelt appreciation to Prof. Nicholas Dulvy for his immense contribution, mentoring, advice and guidance towards the successful completion of this study. Finally, the authors appreciation goes to Moro Seidu, Bukari Saphianu, Paul Tehoda, Emmanuel Amoah, Clement Sullibie Saagulo Naabeh and Adomako Ohene as well as the local volunteers Isaac Assefuah,

References (98)

  • TempleA.J. et al.

    Marine megafauna catch in southwestern Indian Ocean small-scale fisheries from landings data

    Biol. Conserv.

    (2019)
  • AgyemanN.A. et al.

    Illegal, unreported, and unregulated Fisheries threatening shark conservation in african waters revealed from high levels of shark mislabeling in ghana

    Genes

    (2021)
  • ArshamH. et al.

    Bartlett’s test

    Int. Encyclop. Statist. Sci.

    (2011)
  • BeerkircherL.R. et al.

    Characteristics of shark bycatch observed on pelagic longlines off the southeastern United States, 1992–2000

    Mar. Fish. Rev.

    (2002)
  • BizzarroJ.J. et al.

    Activities and catch composition of artisanal elasmobranch fishing sites on the eastern coast of baja california sur

    Mexico. Bull. South. Calif. Acad. Sci.

    (2009)
  • BonfilR.

    Overview of World Elasmobranch Fisheries (No.341)

    (1994)
  • BoyeC.B.

    Causes and Trends in Shoreline Change in the Western Region of Ghana

    (2015)
  • BräutigamA. et al.

    Global priorities for conserving sharks and rays: A 2015–2025 strategy

    (2015)
  • BurkholderD.A. et al.

    Patterns of top-down control in a seagrass ecosystem: could a roving apex predator induce a behaviour-mediated trophic cascade?

    J. Anim. Ecol.

    (2013)
  • CapapéC. et al.

    Distribution and reproductive biology of the blackchin guitarfish, rhinobatos cemiculus (pisces: Rhinobatidae), in Tunisian waters (central mediterranean)

    Mar. Freshwater Res.

    (1994)
  • CarlisleA.B. et al.

    Habitat use, residency, and seasonal distribution of female leopard sharks Triakis semifasciata in Elkhorn Slough, California

    Mar. Ecol. Prog. Ser.

    (2009)
  • CarlsonJ. et al.

    Negaprion brevirostris

  • ColwellR.K. et al.

    Estimating terrestrial biodiversity through extrapolation

    Philos. Trans. R. Soc. B

    (1994)
  • CompagnoL.J.

    Sharks of the world: an annotated and illustrated catalogue of shark species known to date (no.1)

  • CRC

    Coastal Resources Center. Report on characterization of coastal communities and shoreline environments in the Western Region of Ghana. Integrated Coastal and Fisheries Governance Initiative for the Western Region of Ghana. Coastal Resources Center, University of Rhode Island

    (2010)
  • deGraft-JohnsonK.A.A. et al.

    Biodiversity threats assessment of the western region of Ghana

  • DiopM. et al.

    30 Years of shark fshing in West Africa

  • DulvyN.K. et al.

    Gymnura sereti

  • DulvyN.K. et al.

    Chapter 17: Life histories, population dynamics, and extinction risks

  • DulvyN.K. et al.

    Extinction risk and conservation of the world’s sharks and rays

    eLife

    (2014)
  • DulvyN.K. et al.

    Fishery stability, local extinctions, and shifts in community structure in skates

    Conserv. Biol.

    (2000)
  • D’AlbertoB.M. et al.

    Population productivity of shovelnose rays: inferring the potential for recovery

    PLoSONE

    (2019)
  • FAO

    Global Capture Production 1950-2015

    (2017)
  • FAO

    Fisheries and aquaculture software

  • FerreiraL.C. et al.

    Galeocerdo cuvier

  • FowlerS.

    The Conservation Status of Migratory Sharks

    (2014)
  • FowlerS.L. et al.

    Sharks, rays and chimaeras: The status of the chondrichthyan fishes

  • GallagherA.J. et al.

    Ecological risk assessment and its application to elasmobranch conservation and management

    J. Fish Biol.

    (2012)
  • Ghana Statistical Services

    2010 Population and housing census: district analytical report. ghana statistical services

    (2014)
  • HammerO. et al.

    PAST: Paleontological statistics software package for education and data analysis, version 2.03

    Paleontologia Electronica

    (2001)
  • HaqueA.B. et al.

    Evaluating artisanal fishing of globally threatened sharks and rays in the bay of Bengal, Bangladesh

    PLoS One

    (2021)
  • HeupelM.R. et al.

    Quantitative analysis of aggregation behavior in juvenile blacktip sharks

    Mar. Biol.

    (2005)
  • HopkinsT.E. et al.

    The influence of environmental variables on the distribution and abundance of three elasmobranchs in Tomales bay, California

    Environ. Biol. Fishes

    (2003)
  • IUCN

    The IUCN red list of threatened species, 2021

    (2021)
  • JabadoR.W. et al.

    Biological data from sharks landed within the United Arab Emirates artisanal fishery

    Afr. J. Mar. Sci.

    (2016)
  • JabadoR.W. et al.

    Shark diversity in the arabian/Persian gulf higher than previously thought: insights based on species composition of shark landings in the United Arab Emirates

    Mar. Biodiv.

    (2014)
  • JabadoR.W. et al.

    Zanobatus maculatus

  • JabadoR.W. et al.

    Mustelus mustelus

  • JabadoR.W. et al.

    Raja parva

  • Cited by (0)

    View full text