Refining a Nordmøre-grid for a Brazilian artisanal penaeid-trawl fishery

Two experiments were done to assess and refine the Nordmøre-grid in a canoe-trawl fishery targeting Xiphopenaeus kroyeri off southern Brazil. During experiment 1, a codend containing a small (1881cm ) Nordmøre-grid (24-mm bar spaces) retained significantly less total bycatch (mean predicted weight reduced by 56%), brachyurids (by 79%) and teleosts (by 50%) than a conventional configuration (control), without significantly affecting the weights or numbers of X. kroyeri, although mean predicted weights were 12% lower. During experiment 2, the bar spaces were subsequently maintained, but the grid area was increased by 1.4 while the extension section mesh size was reduced by 1.3, and the impacts of different materials (hollow vs. solid aluminium rod), bar diameter (10 vs. 16mm) and presence/absence of a guiding panel were investigated. Most of these technical changes had minimal effects, with all grid configurations maintaining catches of X. kroyeri (at between -13 and +22% of the control) and significantly reducing total bycatch (up to 54%) and brachyurids (up to 90%). But, unlike the small grid, none of the large grids significantly reduced the catches of teleosts. This result was attributed to the increased distance required by fish smaller than the bar spaces to swim out of the escape exit. While ongoing refinements, including narrower bar spaces, could improve the performance of the Nordmøre-grid, the results justify the adoption of the generic configuration as a means for improving selectivity in this fishery