Between-haul Variation Research Articles

The influence of twine thickness on the size selectivity of polyamide codends in a Mediterranean bottom trawl

We investigated the effect of twine thickness on the codend selectivity of Mediterranean bottom trawls. Two codends having meshes with the same opening (ca. 44 mm) but different twine thicknesses (Light PA: R3644tex, Ø 2.38 mm and Heavy PA: R5312tex, Ø 2.89 mm) were fished daily and alternately on the same trawl. Selectivity was measured using the covered codend technique, the cover being supported by circular hoops. The results were analysed taking into account the between-haul variation in selectivity. They demonstrated a substantial reduction in selectivity with increased twine size. For all the species considered, the lengths at 50% retention (L50) indicate that an increase in twine thickness from 2.38 mm (Rtex 3644) to 2.89 mm (Rtex 5312) reduces selectivity by 20–31%. Therefore, the twine thickness of the codend netting played an important role in selectivity. It is thus essential to include this factor in EU legislation aimed at improving codend selectivity in Mediterranean bottom trawl fisheries, where the recent trend towards using thicker twines threatens to cancel the benefit of the effect of mesh size increases or that of the square-mesh codend, allowing the present high rate of discarding small demersal fish to continue.

Theoretical study of the between-haul variation of haddock selectivity in a diamond mesh cod-end

Abstract We use PRESEMO, an individual-based structural model of cod-end selection, to investigate some of the mechanisms that may give rise to what has been termed ‘uncontrolled’ between-haul variation. This is the variation in selection that can occur from haul-to-haul even though there may have been no changes to the cod-end. We demonstrate how haul-to-haul variation of the population density, the population structure and the spatial distribution of both the target and by-catch species may lead to between-haul variation of cod-end selection. We also show that for reasonable levels of variation of these parameters PRESEMO can simulate selection estimates that are consistent with and have comparable levels of between-haul variation as those obtained experimentally.

Effect of PA and PE material on codend selectivity in Turkish bottom trawl

This study investigates the differences in the selective properties of a 40 mm nominal PE, 36 and 44 mm nominal PA codends. The experiments were carried out in Izmir Bay in the Eastern Aegean Sea between 4 February 2002 and 22 March 2002. A conventional bottom trawl was operated onboard R/V Egesüf. Selectivity data were collected for red mullet ( Mullus barbatus), annular sea bream ( Diplodus annularis), picarel ( Spicara smaris) and common pandora ( Pagellus erythrinus) by using hooped covered codend method. Selectivity parameters were obtained by using logistic equation with the maximum likelihood method and by taking into account between-haul variation. Three codends were tested in a total of 25 hauls of which 18 were accepted valid for picarel and 23 valid for the other three species. In general, the PA material codends produced much higher L 50 values than the PE codend. However, the SRs of the three tested codends were very close to each other. L 50’s of 40 mm PE, 36 mm PA and 44 mm PA codends, respectively, were found as 8.8, 8.4 and 10.3 cm for annular sea bream; 10.8, 12.4 and 13.8 cm for common pandora. There were not enough red mullet and picarel retained in the 44 mm PA codend to estimate the selectivity parameters. Due to insufficient amount of retained fish selection parameters of 44 mm PA codend could not be obtained for red mullet and picarel. L 50’s of PE 40 mm and PA 36 mm codends, respectively, were 10.7 and 12.8 cm for red mullet and 12.4 and 13.3 cm for picarel. It is thought that 40 mm PE netting, which is commercially used, has rather poor selectivity. On the other hand, 44 mm PA codend causes the loss of marketable fish. Amongst the three tested codends 36 mm PA provides the most appropriate selection.

Size selectivity for four fish species of the deep groundfish assemblage off the Portuguese southwest coast: evidence of mesh size, mesh configuration and cod end catch effects

The effects were evaluated of an increase in cod end mesh size from 65 to 70 and 80 mm and change of mesh configuration from 65 mm diamond to 65 mm square mesh on the size selectivity of hake ( Merluccius merluccius), horse mackerel ( Trachurus trachurus), four-spot megrim ( Lepidorhombus boscii) and blue whiting ( Micromesistius poutassou) captured on the upper continental slope off the Portuguese southwest coast, at depths from approximately 200–400 m. A number of individual hauls were analysed in the diamond mesh cod ends, for hake and blue whiting, and in the square mesh cod end, for horse mackerel, and mean selection curves were estimated for these cod ends taking into account between-haul variation. For the four-spot megrim, the selectivity estimates were based on pooled data for all the cod ends tested. Selectivity models are proposed for hake and blue whiting in which positive effects of both the increase in mesh size and in cod end catch were estimated for L 50, while SR was only affected by mesh size. For horse mackerel a positive effect of cod end catch was estimated for SR in the square mesh cod end. The selectivity was greatly affected by the change in mesh configuration for all species with the exception of the four-spot megrim, for which only mesh size was found to affect selectivity.

Using mixed models to combine smooth size-selection and catch-comparison curves over hauls

Parametric size-selection curves are often combined over hauls to estimate a mean selection curve using a mixed model in which between-haul variation in selection is treated as a random effect. This paper shows how the mixed model can be extended to estimate a mean selection curve when smooth nonparametric size-selection curves are used. The method also estimates the between-haul variation in selection at each length and can model fixed effects in the form of the different levels of a categorical variable. Data obtained to estimate the size-selection of dab by a Nordmøre grid are used for illustration. The method can also be used to provide a length-based analysis of catch-comparison data, either to compare a test net with a standard net or to calibrate two research survey vessels. Haddock data from an intercalibration exercise are used for illustration.

Size selectivity of the Spisula solida dredge in relation to tooth spacing and mesh size

A study was undertaken on the northwestern coast of Portugal to evaluate the selectivity effect of both tooth spacing and mesh size on the catch of the white clam ( Spisula solida). Two typical clam dredges, equipped with different tooth spacing and mesh size, were towed side by side. Three tooth spacing (20, 40 and 60 mm) and three mesh sizes (35, 40 and 50 mm) were investigated. The dredge selectivity experiments were performed by attaching a cover bag with a 20 mm mesh to the gear. Individual hauls were analysed by the SELECT model. The Fryer’s mixed model of between-haul variation was also applied to data. The results showed that the space between teeth does not have an effect on selectivity. Tooth length was directly related to the dredge’s capture efficiency. The only factor that contributed to dredge size selection was mesh size. The 50% retention lengths estimated for mesh sizes 30, 40 and 50 mm were 25.68, 26.35 and 31.94 mm, respectively. For all mesh sizes, very narrow selection ranges were observed. The most appropriate mesh size to be enforced within the Portuguese northwestern S. solida fishery should be 40 mm.

Size selectivity of diamond and square mesh cod ends for four by-catch species in the crustacean fishery off the Portuguese south coast

The effects of an increase in cod end mesh size from 55 to 60 and 70 mm and a change of mesh configuration from 55 mm diamond to 55 mm square mesh on the size selectivity of four by-catch species (the red shrimp Aristeus antennatus, the European hake Merluccius merluccius, the horse mackerel Trachurus trachurus and the blue whiting Micromesistius poutassou) commonly captured in the crustacean fishery off the Portuguese south coast, were evaluated. Selectivity parameters for blue whiting, the most abundant species in the catches, were estimated taking into account between-haul variation, while for the remaining species, captured in much lower quantities, the selectivity estimates were based on pooled data by length class for all hauls within the same cod end. Length at 50% retention, L 50, was found to increase with mesh size and with the change in mesh configuration for all the studied species. For blue whiting trawling depth and cod end catch were found to play a role in between-haul variation by increasing L 50 as well. The results suggest that an increase in the current minimum mesh size of 55–70 mm would be advisable to be compatible with the minimum landing sizes (MLSs) of 29 mm carapace length and 27 cm total length for red shrimp and hake, respectively, while it would greatly reduce the amount of discards, particularly those for blue whiting, that accounted for approximately 50% of the total catch weight. Horse mackerel was the only species for which the use of a larger mesh size would result in a significant escapement of individuals above the MLS of 15 cm.

Size selectivity of diamond and square mesh cod ends for rose shrimp ( Parapenaeus longirostris) and Norway lobster ( Nephrops norvegicus) off the Portuguese south coast

The effects of an increase in cod end mesh size from 55 to 60 and 70 mm and a change of mesh configuration from diamond to square mesh on the size selectivity for rose shrimp Parapenaeus longirostris and Norway lobster Nephrops norvegicus captured off the Portuguese south coast were evaluated. The results were analysed taking into account between-haul variation in selectivity, and indicate a significant increase in L 50 for rose shrimp with an increase in mesh size or with the use of a square mesh cod end, while for Norway lobster only mesh configuration was found to affect this parameter. Two other important external variables were identified; the trawling depth and the cod end catch, which influence between-haul variation, by increasing the selection range for rose shrimp and Norway lobster, respectively. The results obtained suggest that an increase in the current minimum mesh size of 55 mm would be advisable for rose shrimp in order to respect the minimum landing size of 24 mm carapace length presently established for this species. Moreover, trawling for rose shrimp should be avoided at depths above 200 m, in order to avoid catches consisting almost exclusively of juveniles. Such an increase in mesh size would have a minor impact in terms of losses of individuals above the minimum landing size for Norway lobster and would contribute to reducing the amount of discards in this fishery.

Estimating the size-selection curves of towed gears, traps, nets and hooks

A general statistical methodology for analysis of selectivity data from any type of fishing gear is presented. This formal statistical modelling of selectivity is built on explicit definitions of the selection process and specification of underlying assumptions and limitations, and this gives the resulting estimates of gear selectivity (and possibly fishing power) a clear interpretation. Application of the methodology to studies using subsampled catch data and to towed gears having windows or grids is outlined, and examples applied to passive nets and towed gears are presented. The analysis of data from replicate deployments is covered in detail, with particular regard to modelling the fixed and random effects of between-haul variation. Recent developments on the design of selectivity experiments are introduced and demonstrated.

A model of between-haul variation in selectivity

A model of between-haul variation in selectivity