Bayesian hierarchical surplus production model of whelk (Buccinum undatum) in Icelandic waters.

No abstracts are to be cited without prior reference to the author.

Hierarchical Bayesian methods are suitable for estimating reference points for data-poor stocks, as they can borrow strength from fisheries with more information. That is the case in the restricted whelk (Buccinum undatum) fishery in Iceland, which initiated in late 1990 with annual catches from almost zero to 1300 t. Recent recommendations have been based on average historical catch and logbook statistics. To get better estimate of whelk population dynamics, we used monthly-scaled CPUE data as an index of biomass in a Bayesian Schaefer surplus production model. Two models were compared: the first treated the fishery as one unit (“Simple” model), whereas the second analysed the fishery as being composed of 6 spatial sub-units (Hierarchical model). We imposed 3 different levels of fishing mortality to analyse the risk of high population depletion and also compared results when catches were evenly distributed over sub-units compared to status quo catch distributions. The hierarchical model seems better because it yields more flexibility to include trends among subareas that are not similar and it was not biased like the simple model. The results yield population parameter distributions that can be used as prior information for assessments of whelk elsewhere.