Pelagic fish assemblages assessed using mid-water baited video: standardising fish counts using bait plume size

Abstract

We used a novel mid-water baited remote underwater video (BRUV) technique to sur- vey pelagic and mid-water fish assemblages at Lord Howe Island Marine Park, off eastern Australia (31.5° S, 159.1° E). We found differences in assemblages among regions separated on a scale of ~10s of kilometres. Of the environmental variables measured in the present study (depth, temperature, water current speed), current speed had the greatest influence on the structure of pelagic assem- blages. Carcharhinus galapagensis, Seriola rivoliana and Scomber australasicus had similar distribu- tions and were associated with low flow environments. Seriola lalandi was associated with the higher flow environments sampled during our surveys, although current speeds in the sample area were rel- atively slow (0.03 to 0.34 m s -1 ; mean 0.17 m s -1 ). Relationships between pelagic assemblages and current speed were detected when abundances were calculated using an area-based approximation for plume dispersal, but were not detected using a linear approximation, or when current was assumed constant across all BRUV deployments. Our results demonstrate the difficulty in separating the dual effects of current speed in BRUV studies: the effect of current on bait plume size and associ- ated estimates of fish abundance, and the role of current as an aspect of preferred pelagic fish habi- tat. We recommend the use of area-based approximations of plume dispersal and the inclusion of accurate estimates of current speed as a statistical covariate when interpreting mid-water BRUV data. Our results indicate that oceanographic characteristics can be important for pelagic fish habitat selection over the scale of kilometres and should be considered during marine protected area zoning.