Title: Dynamic production model for shrimp stock assessment: a case study of northern shrimp (Pandalus borealis) in Hunafloi, Icelandic waters

The northern shrimp (Pandalus borealis) fishery in the Húnaflói area developed in the late last century but collapsed in 1999 because of the increasing of cod (Gadus morhua) abundance. This study aims to estimate the shrimp biomass before the collapse and predict the equilibrium yield and spawning stock biomass if the shrimp stock recovers. A length distribution model, a dynamic production model, and a yield per recruit model were used with the input data on length and weight of shrimp, survey abundance index from the annual spring and autumn surveys and catch data from the commercial shrimp fishery. The maximum likelihood criterion and least-squares criterion were used for fitting the models. The size of Pandalus borealis in the Húnaflói area was small compared to other areas such as the Flemish Cap, Barents Sea or Jan Mayen. The growth parameters were in the range of the others with L∞ = 24.9 mm (carapace length), K = 0.29 and t0 = -0.33. The shrimp biomass was estimated with a dynamic production model 1989 and 1999 by tuning it with the abundance index from the surveys and the commercial catch. The average biomass was estimated at about 10,000 tonnes. The expected yield and spawning stock biomass were obtained from the prediction model assuming a recovery of shrimp stock in Húnaflói. An equilibrium yield and spawning stock biomass were also modelled in order to estimate the sustainable harvest levels in the future. Both models agreed on a long term annual yield of 2500 tonnes.