Precision in estimates of density and biomass of Norwegian spring-spawning herring based on acoustic surveys

The abundance and biomass of the Norwegian spring-spawning herring (NSSH) stock are assessed annually using a virtual population analysis (VPA) applied to catch-at-age data from the fishery and fishery-independent abundance indices derived from research vessel surveys for calibration (tuning'). The most important of these surveys is the International Ecosystem Survey in the Nordic Seas (IESNS) and this is highly influential for the outcome of the assessment. Until now, the abundance indices from the IESNS have been reported without any measure of uncertainty. In this study, the sampling errors associated with the density estimates of NSSH from the IESNS for the years 2009-2012 are estimated using design-based survey sampling theory. The annual survey estimates of number and biomass of herring per square nautical mile (nm(2)) were relatively precise for all age groups combined, with relative standard error (RSE) ranging from 8% to 15%, while for each individual age group (3-12) the RSE was less than 30% for most years. For age groups 1 and 2, and all age groups older than 12 years, the density estimates are highly imprecise, with RSE ranging from 30% to 100%. The precision in estimated density provided here indicates that the time series from the survey can be used to study trends in overall abundance and biomass of the stock. It is recommended that statistical assessment models that can account for sampling errors in input data from survey indices by age group and catch-at-age be tested in the assessments of the NSSH stock.