Identifying nonproportionality of fishery-independent survey data to estimate population trends and assess recovery potential for cusk (Brosme brosme)

Cusk (Brosme brosme) was designated as "threatened" by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) in 2003, based on an estimated 93% decline between 1970 and 2001 from the Fisheries and Oceans Canada (DFO) Scotian Shelf summer bottom trawl survey index. We combined this index with a fishery-dependent longline index in a Bayesian surplus production state-space model to estimate population trends and the recovery potential of western Scotian Shelf cusk. We tested for index nonproportionality using a power curve function in the observation model and identified potential hyperdepletion for cusk in the trawl survey index. We estimate a 59% decline in cusk biomass between 1970 and 2001, and a 64% decline from 1970 to 2007. Although population projections indicate the current landing limit should lead to population recovery, robustness tests suggest the biomass projections and recovery time lines are overly optimistic. Simulations showed that incorporating multiple indices increases power to recapture model parameters and failure to account for index nonproportionality results in biased parameter estimates. We suggest that nonproportionality of fishery-independent indices must be considered when determining the population status of data-poor species.