Diel vertical migration and shoaling heterogeneity in Atlantic redfish: effects on acoustic and bottom-trawl surveys

A series of experiments comparing acoustic and bottom-trawl surveys was conducted on Atlantic redfish (Sebastes spp.) on the edge of the Green and Grand Banks of Newfoundland, Canada. Redfish were on or near bottom by day and migrated vertically in the water column at night. In an attempt to account for biases attributable to the presence of fish in the near-bottom dead zone (DZ), a correction factor was applied based on density values measured within the first few metres above the detected bottom. Acoustic densities within increasing range increments above the bottom were compared with densities estimated from the trawl catch. Swept area was calculated using both the trawl's wing spread and door spread as proxies for the minimum and maximum fishing widths. Uncorrected acoustic densities were significantly higher during the night than during the day. No significant day/night differences for the entire water column were observed after DZ corrections. Close agreement between acoustic and trawl densities was obtained by integrating within the first 10 to 20 m off the bottom, with or without the DZ corrections, for both day and night experiments, but regression slopes differed. Trawl catchability appeared to be density-dependent at night, being higher at lower fish densities. Daytime acoustic estimates were more variable than those made at night, as indicated by consecutive passes of several transects and CVs of density (means of 131% during day, 35% at night). We conclude that acoustic measurements made at night provide the most reliable and least variable density estimates, and make recommendations for surveys