Trawl cod-end mesh selectivity for some fishes of North-Western Australia

Abstract A cod-end mesh selectivity experiment was conducted using the covered cod-end technique on the continental shelf of northern Australia. Cod-end (stretched) mesh sizes of 40, 51, 55, 70 and 100 mm were tested with a 30-mm-mesh cover bag. Mesh selection parameters were estimated for 28 species by fitting a logistic curve to the size-specific escapement data, and a gross selection rate was calculated for an additional 27 species. Some species (notably Lutjanus spp., Pristipomoides spp. and Plectorhynchus pictus ) were largely retained by all cod-end mesh sizes tested, while others (notably Loligo spp.) largely escaped all the cod-ends tested. The common assumption that the selection factor does not alter with mesh size was accepted for Nemipterus nematopus, Parupeneus pleurospilus and Lutjanus vittus , but rejected for Saurida undosquamis . Dependence of the selection factor upon mesh size for S. undosquamis was due to a non-zero intercept in the relationship between fish size at 50% escapement and mesh size. The general implications of dependence of selection factor on mesh size are discussed.