Comparison of the relative efficiencies of stereo-BRUVs and traps for samplingtropical continental shelf demersal fishes

Abstract

The sampling efficiencies of commercial standard fish traps and baited remote underwater stereo-video systems (stereo-BRUVs) were compared by examining the diversity and relative abundance of tropical demersal fish that each method sampled on the north-western shelf (40–60 m) of Western Australia. Stereo-BRUVs recorded many more species (91 species from 32 families) than commercial fish traps (30 species and 15 families). Stereo-BRUVs also sampled many more individuals (mean 36.55±5.91 SE) than fish traps (mean 12.30±1.40 SE). This suggests stereo-BRUVs would be more capable of detectingchanges in the relative abundance of species over time. Data from four commercially important species (Epinephelus bilobatus, Epinephelus multinotatus, Lethrinus punctulatus and Lutjanus russelli) revealed that stereo-BRUVs had much greater statistical power to detect change than an equivalent number of samples from fish traps. In contrast, fish traps had a greater statistical power to detect change for a fifth target species, Lutjanus sebae. For two commonly sampled species, Abalistes stellatus1 and Lethrinus punctulatus,stereo-BRUVs sampled a smaller mean length than fish traps while for a third species, Lutjanus sebae, stereo-BRUVs recorded a larger mean length. The length frequencies for these species were not significantly different between methods, although stereo-BRUVs sampled a much larger range of lengths than was captured in traps. This study demonstrates that stereo-BRUVs are potentially a much more powerful technique than fish traps for assessing species richness, relative abundance and size structure inmulti-species fisheries in north-western Australia.