Capelin distribution in winter 2004-2017: spatiotemporal correlation between density estimates from different sampling methods
MetadataShow full item record
The winter distribution of Barents Sea capelin is highly dynamic in space and time, as this is the time of year when mature individuals separate from the immatures and start their spawning migration to the Norwegian and Russian coasts. Due to the difficulty of sampling the entire distribution at this time of year, the stock assessment of capelin is based on data from the ecosystem survey in autumn when capelin has a more favourable distribution for acoustic abundance estimation. A stock projection model is then used to predict how much of the stock will be left and available for the fishery the coming winter. It has long been discussed if a winter survey could be used to improve accuracy of the stock prediction by estimating the maturing part of the stock just before the fishery starts. Estimations based on experimental winter surveys for capelin have so far not been able to achieve this goal. To improve on future attempts, more information on capelin distribution in winter is needed. Here we explore data on capelin distribution collected during the winter survey of the Barents Sea in 2004-2017, a survey primarily collecting data on cod and haddock. Using data from acoustics, demersal trawls, and cod stomachs, we study changes in capelin distribution patterns and correlations between densities from the different data sources during the January-March survey period, with the overarching aim of evaluating the reliability of data sources for capelin investigations. All sampling methods except acoustics showed that capelin shifted toward the south from January-March, and the correlations between capelin densities from different sampling methods varied substantially, reflecting this migration. We found support for the hypothesis that capelin may migrate in the acoustic dead zones, in particularly in the bottom zone, as they move closer to the coast. The discrepancy between sampling methods and variability in the data increased later in the period when capelin shifted south. We recommend further analysis on the vertical distribution of capelin to examine the possibility of combining demersal trawl and acoustic data for abundance estimation of maturing capelin.