The effect of net spread on the capture efficiency of a demersal survey trawl used in the eastern Bering Sea

Abstract

Abstract Relative abundance estimates generated from the National Marine Fisheries Service's annual groundfish trawl surveys in the eastern Bering Sea shelf are based on mean catch per area swept. In the calculation of these estimates, the capture efficiency is assumed to be independent of net spread, which, in turn, increases with depth throughout the depth range of the survey area, and also varies substantially with bottom type and currents. The validity of this assumption was tested using a linear model of catch-per-unit-towed distance versus net spread and depth that was fit to 14 years of survey data for seven species. Catch rates decreased significantly with increasing net spreads for six of the seven species despite the larger area swept at greater net spreads. The magnitude of the downward trends varied considerably by species. Purple sea star (Asterias amurensis) and northern rock sole (Lepidopsetta polyxystra) exhibited the strongest negative relationships between catch rate and net spread, while Alaska plaice (Pleuronectes quadrituberculatus) exhibited the weakest relationship. We attribute this decrease in catch rate to a decrease in capture efficiency. For the benthic species we consider, the decrease in capture efficiency is likely due to increased escapement under the footrope, because calibrated bottom contact sensors indicate that footrope clearance with the bottom increases with net spread. The current assumption of constant capture efficiency at all net spreads may result in biased size distributions and sex ratios for species displaying size and sex segregation by depth. This assumption may also have negative implications for the assessment of highly patchily distributed species as well as those whose depth distribution changes with environmental conditions. Controlling trawl geometry by means of restrictor cables on the warps is a feasible method to address this problem.