Search

On 30th March 2022, all Russian participation in ICES was suspended. Although the announcement of the suspension stressed the role of ICES as a “multilateral science organization”, this suspension applied not only to research activities but also to the ICES work providing fisheries advice for the sustainable management of fish stocks and ecosystems. As a result of the suspension, it is not possible to run ICES stock assessments or provide ICES advice for the Barents Sea stocks of NEA cod, NEA haddock, Sebastes mentella or Greenland Halibut, as management and data collection for these stocks are shared between Norway and Russia. There are therefore no AFWG stock assessments for these stocks this year. This is especially unfortunate as NEA cod is currently declining, and updated assessments are required to ensure an appropriate management response. It is to be hoped that the political decision to exclude Russia from the ICES advice process which underlies our sustainable fisheries management does not lead to mismanagement of the shared stocks and the consequent ecological harm.

The REDUS project (2016-2020) has been a strategic project at the Institute of Marine Research (IMR) aimed at quantifying and reducing the uncertainty in data-rich and age-structured stock assessments (e.g., cod, herring, haddock, capelin). Work was organized in four topical work-packages: Fisheries-dependent (catch) surveys and assessment modeling (WP1), Fishery-independent (scientific) surveys (WP2), Evaluating and testing of long-term management strategies (WP3), and Communication of uncertainty, dissemination of project results and capacity building (WP4). The Norwegian Computing Center (NR) was contracted in as a strategic partner in statistical modeling and analysis, contributing mainly to WP1 and WP2, but found the research of fundamental interest therefore also allocating internal (NR) funding to develop the statistical science base of several of the methods.

A Benchmark Workshop for Demersal species (WKDEM), chaired by External Chair Richard D.M. Nash, UK and ICES Chair Daniel Howell, Norway and attended by two invited external experts Amy Schueller, US and Robert Boenish, US met at ICES, HQ, Copenhagen 9–13 December 2019 for a data evaluation meeting and at ICES HQ, Copenhagen, Denmark, for a Benchmark meeting 10–14 February 2020. In addition, two WebEx meetings were convened (23rd October 2019 and 23rd February 2020) to discuss data issues and logistics related to the work plan prior to the data evaluation meeting and the Benchmark.Four demersal stocks were Benchmarked in WKDEM, Haddock (Melanogrammus aeglefinus) in subareas 1 and 2 (Northeast Arctic) (Had.27.1–2), Cod (Gadus morhua) in Division 6.a (West of Scotland) (Cod.27.6a), Whiting (Merlangius merlangus) in Division 6.a (West of Scotland) (Whg.27.6a) and Whiting in Division 3.a (Skagerrak and Kattegat) (Whg.27.3a).The 2020 benchmark for Northeast Arctic Haddock was proposed due to the poor retrospective pattern (annually changing perception of the stock) for the assessment in 2018. The final SAM model was based on the updated dataset and a new SAM setting allowing for plus groups in surveys. The final model had a lower AIC and more parameters compared to the configuration used in the previous assessment with the same data as the final model. Some changes were made to the prediction models, mainly to reflect the current perception of the stock dynamics. The reference points were not changed.The 2020 benchmark for 6.a cod was proposed initially due to uncertainties in the catch data and the selectivity patterns in the catch and survey data. In addition, a separate analysis of the stock estimated a different trend in fishing mortality and higher biomass, which lead to an inter-benchmark in February 2019. A number of alternative assessment methods were considered with a decision to switch from the TSA model to SAM. An improved approach to estimating area-misreported landings was implemented, and maturity and natural mortality values were modified. The survey data included in the final assessment are consistent (in terms of year and age range) with those used previously. The reference points were updated having been derived following the ICES guidelines and using EqSim. There was some concern that an alternative model, FAFmodel, along with an extension to allow natural mortality to be estimated by the model (“TVM” model) provided a different perception of the stock dynamics. The latter model was not presented for evaluation prior to the Benchmark and as such needs to be evaluated as to its efficacy for assessment of this stock in the future.The 2020 benchmark for 6.a whiting was proposed because the assessment was influenced partly by incorrect reporting of landings data (species and quantity) from the past which directly affected the perception of the stock. There were newly available catch data and reworked survey indices (including the combined Irish and Scottish Q4 survey), plus reworked maturity and natural mortality estimates were available. It became clear that running TSA with the new data and changed survey configuration resulted in poorly converged optimisation runs. An age-aggregated stochastic Surplus Production in Continuous Time (SPiCT) model was tried but the modelled stock level was highly uncertain, although the trends were considered more robust. The result was to reclassify 6.a whiting to category 3 using the SPiCT model to give trends, according to the ICES guidelines for data-limited stocks.The 2020 benchmark for 3.a whiting was proposed because it was considered a category 5 stock with no advice and more information is available (time-series of catches (landings and discards), a new fisheries independent biomass index that combines all relevant scientific surveys in the area) that were not being used but could be utilised to infer stock status and provide better catch advice. The outcome was to reclassify the stock to category 3 and provide advice using the “2-over-3” trend based approach with the new biomass index. An attempt to provide an analytical assessment for the stock using the surplus production model (SPiCT) was unsuccessful.