Length-selective retention of walleye pollock, Theragra chalcogramma, by midwater trawls

Midwater trawls are commonly used during acoustic surveys of fish abundance to determine species and length compositions of acoustically sampled fish aggregations. As trawls are selective samplers, catches can be unrepresentative of sampled populations and lead to biased abundance estimates. Length-dependent retention of walleye pollock was estimated using small recapture nets, so-called pocket nets, attached to the outside of the trawl. Experimental haul sets comprising eight hauls each were conducted in the Gulf of Alaska in 2007 and 2008 and in the eastern Bering Sea (EBS) in 2007. Pocket-net catches were then modelled by fitting parameters for selectivity and escapement location along the trawl. Within-and between-haul variability was jointly estimated using hierarchical Bayesian methods. There was significant undersampling of juvenile (<25 cm) pollock, with the length-at-50%-retention (L(50)) estimated between 13.5 and 26.1 cm among haul sets. In the EBS set, L(50) values were correlated with light level, escapement being greater at night. Trawl selectivity may be a significant source of error in acoustic-survey estimates of the abundance of pollock.