Capelin in the Barents Sea

The management of Barents Sea capelin is based on two different tools. At the meeting after the yearly survey the quota corresponding to the harvest rule is calculated with the spreadsheet CapTool, which gets it's parameters from estimations done with the model Bifrost.

Present and possible future harvest rules

The present harvest rule dictates that there shall be a maximum of 5% probability for the spawning stock by April 1 to be smaller than 200 000 tonnes. The Norwegian-Russian Fishery Commission will by 2015 rethink harvest rules for the Barents Sea. An alternative could be to base the rule on MSY considerations, which necessitates thorough work with recruitment models, as new harvest rules must be tested by long-term simulations. Also, the Precautionary Approach could be implemented differently, like using the spawning stock biomass in 1989 (which was rather small but yielding a very good year class), demanding a low probability for low SSB 3 years in a row, or using prognosed recruitment taking into account herring (which could be inferred from the measured 0-group index) instead of the spawning stock biomass directly.

Biological assumptions and ongoing work

The ongoing work on the models is summarized at milestone meetings, primarily between the collaborating Russian and Norwegian scientists.

The modelling underlying present (autumn 2013) management relies on a few fundamental assumptions that, however, can be relaxed.

Problems with the biological assumptions

Mature cod does not feed on capelin during January-March

The rationale for this assumption has been that the mature cod to a large extent is out of the capelin migration area, as it is underway to it's spawning grounds at Lofoten. Also, because it is on a spawning migration, it will not feed to any appreciable degree. If these assumptions are invalidated appreciably, it may have serious consequences for today's management of capelin, as the mature stock of cod is large and growing.

Problematic assumption 1: cod does not feed on immature capelin during January-March

The rationale for this assumption has been that there traditionally has been little overlap between cod and immature capelin, as the immature capelin during this period. Recent observations suggest that this assumption might be violated.

Problematic assumption 2: mean monthly natural mortality for mature capelin October-December can be taken from the estimated mean natural mortality for immature capelin

We have seen that there have been full overlap between cod and capelin during September, with substantial consumption of capelin by cod. At the same time, the natural mortality over the year is not particularly high, This may point to the assumption above being invalidated. The obvious way to deal with this problem is to model the natural mortality exerted by cod predation explicitly, as it is being done for January-March.

Revised model

The model is being revised to meet the above challenges. Consumption of capelin by cod the year around is modelled and different model riggings is being investigated to meet new estimation problems.

Some biological background

Consumption by cod