Services on Demand
- Cited by SciELO
- Access statistics
- Cited by Google
- Similars in SciELO
- Similars in Google
On-line version ISSN 0718-560X
Lat. Am. J. Aquat. Res. vol.40 no.1 Valparaíso Mar. 2012
Lat. Am. J. Aquat. Res., 40(1): 124-133, 2012
Relative efficiency of square-mesh codends in an artisanal fishery in southern Brazil
Eficiencia relativa de copos de malla cuadrada en una pesquería artesanal del sur de Brasil
Catarina N.S. Silva1,2, José H. Dias1, André P. Cattani1 & Henry L. Spach1
1Centro de Estudos do Mar, Av. Beira Mar, s/n, 83255-000 Pontal do Sul, Paraná, Brazil.
2School of Biological Sciences, Victoria University of Wellington, P.O. Box 600 Wellington, New Zealand.
ABSTRACT. The relative efficiency of two square-mesh codend designs with the same circumference (~2 m) but different mesh sizes and materials (32 mm polyethylene-PE and 30 mm polyamide-PA) was tested in an artisanal shrimp trawl fishery in Paraná, southern Brazil. The two square-mesh codends were hauled alternately with a 26 mm diamond-shaped mesh codend (control) in a twin gear configuration. Although not significant, the mean numbers of total bycatch were reduced by 16.6 and 10.0% with the 32 and 30 mm square-mesh codends, respectively. The results indicate significant improvement in size selectivity for some species (Xiphopenaeus kroyeri and Stellifer rastrifer). The operational changes tested can be a suitable technical solution for reducing the capture of immature organisms in the shrimp trawl fishery in Paraná, Brazil.
Keywords: discards, shrimp fishery, bycatch reduction devices, square-mesh, size selection, fisheries management, southern Brazil.
RESUMEN. La eficiencia relativa de dos diseños de copos con malla cuadrada con la misma circunferencia (~2 m) pero diferentes tamaños de malla y materiales (32 mm de polietileno-PE y 30 mm de poliamida PA) ha sido evaluada en una pesquería de arrastre artesanal de camarón en Paraná, sur de Brasil. Los dos copos de malla cuadrada fueron arrastrados alternativamente con un copo de malla de 26 mm en forma de diamante (control) en una configuración de doble arrastre. Aunque no es significativo, el número medio de la captura incidental total se redujo en 16,6 y 10,0% en los copos de malla cuadrada de 32 y 30 mm, respectivamente. Los resultados indican que se logra una significativa mejora en la selectividad por tamaño medio de algunas especies (Xiphopenaeus kroyeri y Stellifer rastrifer). Los cambios operacionales evaluados pueden representar una solución técnica adecuada para reducir las capturas de organismos inmaduros en la pesquería de arrastre de camarón en Paraná, Brasil.
Palabras clave: descarte, pesca del camarón, dispositivos de reducción de capturas incidentales, malla cuadrada, selección por tamaño, gestión pesquera, sur de Brasil.
harvested each year in Brazil (IBAMA, 2005). Bottom trawling for shrimp is one of the most common fishing practices in the southern region of Brazil (Andriguetto-Filho et al., 2009) and like most penaeid-trawl fisheries worldwide, owing to the small meshes used, it is responsible for the retention of large quantities of unwanted organisms (collectively named 'bycatch'; Hall, 1996) that are consequently discarded.
Southern Brazilian artisanal trawl fisheries are noteworthy for their large catches of seabob shrimp (Xiphopenaeus kroyeri). This penaeid species, globally ranked among the top five wild-caught penaeids (Gillett, 2008), has typically 10- to 30-mm carapace length CL and accounts for more than 5,000 ton Southern Brazilian artisanal trawlers discard all of their bycatch, which mainly comprises small teleosts (e.g. sciaenids) and brachyurids (Silva et al., 2011).
In recent years, concerns about the impact on stocks of important bycatch species by penaeid trawls have resulted in efforts at improving selectivity, as a part of an ecosystem-based approach to fisheries management (Cochrane, 2002). An inexpensive and simple modification to improve the selectivity of some penaeid trawls involves changing only the configuration of mesh in the codend by hanging conventional diamond-shaped mesh on the bar -termed "square-mesh" (e.g. Broadhurst & Kennelly, 1994; Macbeth et al., 2004). Codends made entirely by square meshes are bycatch reduction devices (BRDs) that separate species based on their size -mechanical separation.
Most of the studies describing the development of BRDs for penaeid trawls have originated from developed countries (Broadhurst, 2000). Gillett (2008) states that landed bycatch tend to be much higher in poor tropical countries than in developed countries. Since developing countries greatly contribute towards total global bycatch estimates (Kelleher, 2005), it is of the greatest importance to develop and apply BRDs in these fisheries.
Our aim in this paper was to conduct an experiment with two designs of square-mesh codends under normal fishing operations in southern Brazil to investigate their potential for reducing catches of unwanted organisms, while maintaining the catches of the retained shrimps. This study is the first of its kind in a Brazilian artisanal trawl shrimp fishery.
MATERIALS AND METHODS
The experiment was done during four fishing days in February 2010 off the coast of Paraná, Brazil (25°40'S, 48°30'W) (Fig. 1), using a fibreglass canoe (10 m) powered by a single cylinder 16 kW diesel engine. The two trawls were constructed from 26 mm polyamide-PA mesh throughout the wings and body with 8.8 mm headline and foot rope (see Silva et al., 2011 for a description of the gear and vessels). Zippers (Burazchi S146R) measuring 1.5 m in length were attached to the posterior trawl body to facilitate changing codends. The tow duration was chosen based on conventional fishing practice in the region which is between 30 and 60 min. All hauls were 30 min in duration; done between 07:00 and 13:00 h across sandy bottoms (7-16 m) at ~0.5-0.8 m s-1 and manoeuvred by hand. The locations and directions of the tows were decided by local fishers to guarantee conventional fishing practices.
|Figure 1. Map showing the location of the coast of Paraná and the area trawled during the experiment. |
Figura 1. Mapa mostrando la localización de la costa de Paraná y el área de arrastre durante el experimento.
A control diamond-shaped mesh and two square-mesh codends were constructed for the experiment (Fig. 2). The control codend represented existing conventional 26 mm polyamide-PA diamond-mesh designs (1.5 mm twine diameter) used in the artisanal penaied-trawl fishery of southern Brazil. The two treatment codends were made of 32 mm polyethylene-PE (4.0 mm twine diameter) and 30 mm polyamide-PA (1.5 mm twine diameter) square-mesh. The two square-mesh designs were alternately compared against the conventional diamond-mesh codend, rigged in a twin-gear configuration (one on each side of the vessel). Three replicates were done on each day providing a total of 12 hauls for each paired comparison, randomly allocated to each trawl (to eliminate any trawl bias).
Figure 2. Diagrammatic representation of a) the control 26 mm diamond-shaped mesh polyamide-PA codend, b) the 32 mm square-mesh (22.5 mm outside bar) polyethylene-PE codend, and c) the 30 mm square-mesh (16 mm outside bar) polyamide-PA codend used in the experiment (T: transversals, B: bars, N: normals).
Figura 2. Diagrama de a) copo control de malla diamante de 26 mm de poliamida-PA, b) copo de malla cuadrada de 32 mm (22.5 mm barra externa) de polietileno-PE y c) copo de malla cuadrada de 30 mm (16 mm barra externa) de poliamida-PA usados en los experimentos (T: transversales, B: barras, N: normales).
After each haul, the contents of each codend were emptied into separate trays and the total weights of X. kroyeri, teleosts and brachyurids (and therefore total bycatch) were recorded onboard. The total bycatch and a subsample (n = 50) of X. kroyeri were then collected from each codend, stored on ice, separated by species, measured (to the nearest 0.5 cm) and weighted in the laboratory.
After preliminary tests for normality (Shapiro-Wilk's W test) and for homogeneity of variances (Levene's test), all variables were analysed with two-tailed, paired t-tests (P < 0.05), when necessary data were ln(x + 0.1) transformed. Size frequencies of X. kroyeri and two abundant key teleosts (Stellifer rastrifer and Selene setapinnis) were combined across all tows and compared between the square-mesh and their respective control codends using two-sample Kolmogorov-Smirnov tests (P < 0.01). Mean sizes of each species (Carapace length (CL) for X. kroyeri and Total length (TL) for S. rastrifer and S. setapinnis) were estimated for each haul and compared between control and square-mesh codends using two-tailed, paired t-tests (P < 0.05).
Seabob shrimp comprised approximately 75% (in number) of the total catches from the three codends. In total, 46 species (38 teleosts, 1 elasmobranch and 7 crustaceans) were recorded, although more than 70% of the bycatch (in number) comprised only two brachyurids (Callinectes ornatus and Hepatus pudibundus) and two teleosts (Stellifer rastrifer and Selene setapinnis) (Table 1). The species composition was consistent with that typically observed in the fishery.
Table 1. Numbers (n) of organisms caught with the control and square-mesh codends during the four days of experiment and their scientific and common names.
Tabla 1. Números (n) de los organismos capturados con los copos controle y de malla cuadrada al largo de los cuatro días de experimento y sus respectivos nombres científico y común.
There were no significant differences in terms of weight and number of bycatch retained between control and square-mesh codends. However, compared to the control, the 32 and 30 mm square-mesh codends retained fewer total bycatch by 16.6 and 10.0%, respectively (P > 0.05; Fig. 3b, Table 2). The catches of X. kroyeri were not significantly reduced, despite the 32 and 30 mm square-mesh codends having reduced the weights by 4.0 and 3.4% and the numbers by 6.7 and 9.6%, respectively (Fig. 3a, Table 2). Although not significant, the 32 and 30 mm square-mesh codends also reduced the numbers of total fish by 24.0 and 29.9% (Fig. 3c, Table 2). Compared to the control, the 32 mm square-mesh codend caught fewer S. rastrifer and S. setapinnis (11.5 and 21.4%, respectively) and reduced the weight of S. setapinnis by 26.6%, while the 30 mm square-mesh codend caught fewer S. rastrifer (in terms of weight and number, 14.4 and 40.9%) (Figs. 3d-3e). Further, none of the square-mesh codends were effective in terms of reducing the catches of brachyurids (Table 2).
Table 2. Summary of two-tailed paired t-test comparing catches of X. kroyeri, total bycatch, brachyurids, total fish, S. rastrifer and S. setapinnis and mean sizes of X. kroyeri, S. rastrifer and S. setapinnis between control and square-mesh codends (32 and 30 mm). n = 12. ** Significant differences at P < 0.01. * Significant differences at P < 0.05, n: number, w: weight, CL: carapace lenght, TL: total length.
Tabla 2. Resumen del teste t apareado con dos colas comparando las capturas de X. kroyeri, bycatch total, brachyurids, total de peces, S. rastrifer y S. setapinnis y los tamaños medios de X. kroyeri, S. rastrifer y S. setapinnis entre los copos control y de malla cuadrada de 32 y 30 mm. n = 12. ** Diferencia significativa con P < 0,01. * Diferencia significativa con P < 0,05; n: número, w: peso, CL: longitud del caparazón, TL: longitud total.
Two-sample Kolmogorov-Smirnov tests detected significant differences in the size-frequency distributions of X. kroyeri and S. rastrifer (P < 0.01; Fig. 4a-4b) between the control and the 32 and 30 mm square-mesh codends. There were no significant differences in size-frequencies of S. setapinnis between the control and square-mesh codends (Kolmogorov-Smirnov test, P > 0.05, Fig. 4c).
Figure 4. Size-frequency distributions of a) Xiphopenaeus kroyeri, b) Stellifer rastrifer, and c) Selene setapinnis captured with the control and square-mesh codends (32 and 30 mm) during the experiment. ** Significant differences at P < 0.01.
Figura 4. Distribuciones de frecuencia de tamaño de a) Xiphopenaeus kroyeri, b) Stellifer rastrifer, y c) Selene setapinnis capturados con los copos control y de malla cuadrada (32 y 30 mm) durante el experimento. ** Diferencia significativa con P < 0,01.
Compared to the control, the 32 mm codend caught individuals of X. kroyeri with a significantly higher mean carapace length (P < 0.05, Table 2, Fig. 5a). Both square-mesh codends (32 and 30 mm) caught significantly bigger individuals of S. rastrifer in terms of mean total length, comparing to the control codend (P < 0.01, Table 2, Fig. 5b). No other significant differences in terms of mean size between the control and square-mesh codends were detected (P > 0.05, Table 2, Figs. 5a and 5c).
Figure 5. Mean sizes ± SE 30 min-1 (between hauls) of: a) Xiphopenaeus kroyeri, b) Stellifer rastrifer; and c) Selene setapinnis captured with the control and square-mesh codends (32 and 30 mm). ** Significant differences at P < 0.01, * Significant differences at P < 0.05.
Figura 5. Tamaños medios ± ES 30 min-1 (entre arrastres) de: a) Xiphopenaeus kroyeri, b) Stellifer rastrifer; y c) Selene setapinnis capturados con los copos control y de malla cuadrada (32 y 30 mm). ** Diferencia significativa con P < 0,01, * Diferencia significativa con P < 0,05.
Although the reduction of total bycatch in terms of number and weight was not significant, the results of this experiment tend to support the wide utility of square-mesh codends for releasing small fish and penaeids from trawls, as shown by Bahamon et al. (2006), Macbeth et al. (2007) and Broadhurst et al. (2010). Both the 32 and 30 mm square-mesh codends performed similarly without significantly affecting the catches of Xiphopenaeus kroyeri.
Comparing with the control, the 30 mm codend caught fewer total fish and Stellifer rastrifer than the 32 mm square-mesh codend (Figs. 3c-3d). Such an apparent negative relationship between selection and mesh size may have been caused because the thick twine diameter of the 32 mm codend resulted in the meshes being more closed and maybe more visible and therefore dissuading escape attempts (Sala et al., 2007). This result may also reflect species-specific behaviour, swimming ability and twine bending stiffness (Herrmann & O'Neill, 2006).
Due to the small sizes of the organisms caught in the fishery, the square-mesh codends had no substantial benefit in terms of reducing the weight of total bycatch; however, although not significant, the reduction in the number of unwanted organisms caught was appreciable. Moreover, there was evidence of an associated dependence of species morphology with the mechanisms by which the square-mesh codends select the catches. Specifically, compared to the conventional codend, both the square-mesh designs contributed to some improvement towards trawl performance which was due to the escape of small individuals of Stellifer rastrifer. The significant size-selectivity of S. rastrifer may be possibly due to their relatively fusiform body, while S. setapinnis has a strongly compressed body (Figs. 4b-4c; Figs. 5b-5c).
Other studies have shown the utility of square-mesh codends to increase and maintain lateral mesh openings in demersal trawls (Broadhurst et al., 2004, 2010; Macbeth et al., 2007). Due to this fact, the square meshes allow more small fishes and penaeids to escape than the diamond-meshes. In this study, the 30 and 32 mm square-mesh codends caught individuals of S. rastrifer with a significant higher mean total length than individuals from the control (Fig. 5b); similarly, the 32 mm square-mesh codend caught seabob shrimp with a significant higher mean carapace length than the control codend (Fig. 5a) which indicates the escape of small fishes and shrimps through the square meshes. Moreover, comparing to the diamond-mesh codends, the square-mesh codends are the best alternative in terms of minimising the damage of escapees, as proved by Farmer et al. (1998).
As observed in many of the world's shrimp-trawl fisheries (e.g. Liggins et al., 1996; Kennelly et al., 1998; Sala et al., 2008) fish that comprise bycatches are usually quite small (< 20 cm long) and often juveniles. Seabob shrimp is a smaller species of penaeid (usually < 10 cm long) and, as bycatch fish species have a similar size it may be possible to reduce more effectively bycatches by a behavioural separating mechanism. For example, a square-mesh panel "strategically" positioned in the codend separates species based on the characteristic escape response of fishes to trawls (see Broadhurst, 2000 for a review).
Like other studies (e.g. Macbeth et al., 2005) the data suggest that the codends made of diamond-mesh used in the fishery are less selective than the square-mesh codends tested and, since the catches of seabob shrimp were not affected, it should be feasible to use larger sizes of mesh. Further research is required, therefore, to explore the utility of different operational changes in the fishing gear configurations in the fishery. Additionally, future experiments should be replicated under different operational conditions (e.g. longer hauls) and in different periods of the year to encompass changes in the size structure of the species in order to modify the codends design and improve the efficiency of the artisanal penaeid trawl nets. Additionally, given that there were few (if any) negative impacts to the fishing operation associated with using square-mesh codends and the netting materials are locally available, enforcing the use of square-mesh codends in the southern Brazilian artisanal penaeid fishery would be a suitable technical solution to reduce the capture of immature fish. Assuming most of these escaping individuals survive (Broadhurst et al., 2006), such a management option should translate to a reduction in impacts on the stocks of key species.
We thank "Laboratório de Biologia de Peixes, Centro de Estudos do Mar", CAPES "Coordenação de Aperfeiçoamento de Pessoal de Nível Superior" for financial support, Matt Broadhurst for his invaluable advice and assistance, Andréia Schwingel and Bianca Budel for their technical assistance and Jair Crisanto da Silva, Florismar Santana da Silva, Ismail Santana da Silva, Atair Santana da Silva and Alex Nascimento da Silva for the use of their vessels.
Andriguetto-Filho, J.M., R. Krul & S. Feitosa. 2009. Analysis of natural and social dynamics of fishery production systems in Paraná, Brazil: implications for management and sustainability. J. Appl. Ichthyol., 25: 277-286. [ Links ]
Bahamon, N., F. Sardà & P. Suuronen. 2006. Improvement of trawl selectivity in the NW Mediterranean demersal fishery by using a 40 mm square mesh codend. Fish. Res., 81: 15-25. [ Links ]
Broadhurst, M.K. 2000. Modifications to reduce bycatch in prawn trawls: a review and framework for development. Rev. Fish Biol. Fish., 10: 27-60. [ Links ]
Broadhurst, M.K. & S.J. Kennelly. 1994. Reducing the by-catch of juvenile fish (mulloway Argyrosomus hololepidotus) using square-mesh panels in codends in the Hawkesbury River prawn-trawl fishery. Fish. Res., 19: 321-331. [ Links ]
Broadhurst, M.K., R.B. Millar & C.P. Brand. 2010. Diamond- vs. square-mesh codend selectivity in southeastern Australian estuarine squid trawls. Fish. Res., 102: 276-285. [ Links ]
Broadhurst, M.K., P. Suuronen & A. Hulme. 2006. Estimating collateral mortality from towed fishing gear. Fish Fish., 7: 180-218. [ Links ]
Broadhurst, M.K., R.B. Millar, S.J. Kennelly, W.G. Macbeth, D.J. Young & C.A. Gray. 2004. Selectivity of conventional diamond- and novel square-mesh codends in an Australian estuarine penaeid-trawl trawl fishery. Fish. Res., 67: 183-194. [ Links ]
Cochrane, K.L. 2002. A fishery manager's guidebook. Management measures and their application. FAO Fish. Tech. Paper, 424: 231 pp. [ Links ]
Farmer, M.J., D.T. Brewer & S.J.M. Blaber. 1998. Damage to selected fish species escaping from prawn trawl codends: a comparison between square-mesh and diamond-mesh. Fish. Res., 38: 73-81. [ Links ]
Gillett, R. 2008. Global study of shrimp fisheries. FAO Fish. Tech. Paper, 475: 359 pp. [ Links ]
Hall, M.A. 1996. On bycatches. Rev. Fish Biol. Fish., 6: 319-352. [ Links ]
Herrmann, B. & F.G. O'Neill. 2006. Theoretical study of the influence of twine thickness on haddock selectivity in diamond mesh-codends. Fish. Res., 80: 221-229. [ Links ]
Instituto Brasileiro do Meio Ambiente e dos Recursos Renováveis (IBAMA). 2005. Estatística da Pesca 2005. Brasil. Grandes regiões e unidades de federação. IBAMA, Brasília, 137 pp. [ Links ]
Kelleher, K. 2005. Discards in the world's marine fisheries. An update. FAO Fish. Tech. Paper, 470: 131 pp. [ Links ]
Kennelly, S.J., G.W. Liggins & M.K. Broadhurst. 1998. Retained and discarded by-catch from oceanic prawn trawling in New South Wales. Fish. Res., 36: 217-236. [ Links ]
Liggins, G.W., S.J. Kennelly & M.K. Broadhurst. 1996. Observer-based survey of by-catch from prawn trawling in Botany Bay and Port Jackson, New South Wales. Mar. Freshw. Res., 47: 877-88. [ Links ]
Macbeth, W.G., M.K. Broadhurst & R.B. Millar. 2004. The utility of square mesh to reduce bycatch in Hawkesbury River prawn trawls. Ecol. Manage. Restor., 5(3): 210-213. [ Links ]
Macbeth, W.G., M.K. Broadhurst & R.B. Millar. 2005. Fishery-specific differences in the size selectivity and catch of diamond- and square-mesh codends in two Australian penaeid seines. Fish. Manag. Ecol., 12: 225-236. [ Links ]
Macbeth, W.G., R.B. Millar, M.K. Broadhurst, C.W. Hewitt & M.E.L. Wooden. 2007. Intra-fleet variability in the size selectivity of a square-mesh trawl codend for school prawns (Metapenaeus macleayi). Fish. Res., 27: 92-98. [ Links ]
Sala, A., A. Lucchetti & G. Buglioni. 2007. The influence of twine thickness on the size selectivity of polyamide codends in a Mediterranean bottom trawl. Fish. Res., 83: 192-203. [ Links ]
Sala, A., A. Lucchetti, C. Piccinetti & M. Ferretti. 2008. Size selection by diamond- and square-mesh codends in multi-species Mediterranean demersal trawl fisheries. Fish. Res., 93: 8-21. [ Links ]
Silva, C.N.S., M.K. Broadhurst, A. Schwingel, J.H. Dias, A.P. Cattani & H.L. Spach. 2011. Refining a Nordm0re-grid for a Brazilian artisanal penaeid-trawl fishery. Fish. Res., 109: 168-178. [ Links ]
Received: 24 January 2011; Accepted: 9 January 2012.
Corresponding author: Catarina N.S. Silva (firstname.lastname@example.org)