Fig 1.
Study area, hauls made during the sea trials, and base harbour of each vessel.
AUD: Audace, RIM: Rimas, JOA: Joacchì, AST: Astuzia, GLA: Gladiatore, PAL: Palestini, TAR: Tarantini.
Table 1.
Main characteristics of the trawlers used in the sea trials.
AUD: Audace, RIM: Rimas, JOA: Joacchì, AST: Astuzia, GLA: Gladiatore, PAL: Palestini, TAR: Tarantini. LOA: Length overall, GT: Gross Tonnage.
Fig 2.
The flexible Turtle Excluder Device (TED).
a) Position of the TED in relation to the codend. b) Specification of the high-strength, flexible plastic TED used in the trials. Measurements given in mm. c) Technical drawing of the TED rigging (lateral view). GF: Guide Funnel. AB and AN, types of net cuts (AB is a parallel cut to a line of sequential mesh bars; AN is a cut perpendicular to the general course of the net; figures indicate mesh numbers). The mean grid angle recorded during sampling is also reported.
Table 2.
Operating conditions and catch per unit effort based on weight (kg h-1) for each vessel during the sea trials.
SE: standard error, ns: not significant. P value: Wilcoxon test. AUD: Audace, RIM: Rimas, JOA: Joacchì, AST: Astuzia, GLA: Gladiatore, PAL: Palestini, TAR: Tarantini.
Fig 3.
Mean commercial, discard and debris CPUEW of TED and CTRL nets per vessel.
Bars: standard error. AUD: Audace, RIM: Rimas, JOA: Joacchì, AST: Astuzia, GLA: Gladiatore, PAL: Palestini, TAR: Tarantini.
Table 3.
Model selection for the catch rates.
In bold between square brackets: the best model explained by the independent variables. The estimates of the factors obtained by the regression are also reported with the respective standard errors, degree of freedom and t values.
Table 4.
Raw means and standard errors (SE) for commercial, discard and debris catches.
The significant results (except for factor “net” of the commercial catch) of the Tukey’s pairwise of the selected models are also reported.
Fig 4.
Catch length-frequency distributions of each commercially important species per vessel.
AUD: Audace, RIM: Rimas, JOA: Joacchì, AST: Astuzia, GLA: Gladiatore, PAL: Palestini, TAR: Tarantini.
Fig 5.
Pooled catch length-frequency distribution of each commercially important species (all vessels).
Fig 6.
Proportions of the total catches of the TED nets according to the GLMM.
The main diagrams show the model for all vessels, whereas the smallest diagrams show the model applied to each trawler. Interpretation: a value of 0.5 indicates an even catch between the TED and the CTRL nets, whereas a value of 0.25 indicates that the TED net caught 25% of the total fish of that length and the CTRL net caught 75%. Shaded area: 95% confidence interval.
Table 5.
GLMM parameter estimates of the catch selectivity logistic models.
TED No. Ind.: TED number of individuals used for the analysis; CTRL No. Ind.: control number of individuals; SE: standard error.