Optimisation of baited remote underwater video sampling designs for estuarine fish assemblages

Baited remote underwater video stations (BRUVS) are used for monitoring fish assemblages and assessing management effectiveness in reef environments but are infrequently used in estuaries. A review of the BRUVS literature found that most adopted sampling designs from other studies were rarely designed from pilot studies. This potentially compromises their value for monitoring natural and anthropogenic variation. The aims of this study were: (i) to assess the suitability of BRUVS for sampling fishes in estuarine habitats (seagrass beds and unvegetated sediments) and (ii) to develop an optimal and cost effective sampling methodology for each habitat. Fishes in both habitats were sampled independently using BRUVS with soak times of 30, 60, 90 min (n = 4). Thirty five species of fishes were recorded including 18 species of economic importance. Mean number of species, mean total Max N and mean Max N of species did not differ among soak times. Precision was generally greater in seagrass and in both habitats it improved with increasing soak time. Bootstrapping revealed that greater improvements in precision occurred from increasing soak time rather than increasing replication. A sampling design with n = 5 replicates of 90 min soak time was optimal for most variables. This sampling effort is greater than many current applications of BRUVS. The results highlight the importance of pilot studies to optimise sampling methods and develop cost effective and statistically-robust monitoring programs.