Why has the 2025 advice for Northern Shelf Cod changed?
The advice has been changed due to an error detected in the bespoke multi-stock stochastic forecasting model used for the advice in 2024 and 2025. This does not affect the stock assessment model fit; only the forecast is impacted.
A corrected version of the forecasting model has now been developed and validated. New forecasts providing revised catch options for 2025 were reviewed by experts and approved by the Advisory Committee (ACOM).
Why has the catch advice been reduced?
The combined catch advice for all three substocks (Northwestern, Viking, and southern) is now 15 511 tonnes, revised down from 19 321 tonnes.
The main reason for this lower forecast is that the model predicts higher levels of natural mortality. As a result, fishing must be reduced even more—especially for the southern substock—to help rebuild it to a safe population level above the biomass limit, Blim.
How was the error detected?
ICES Mixed Fisheries Working Group (WGMIXFISH) independently reproduces the single stock catch forecasts with their mixed-fisheries forecast model as a routine quality control check.
During their meeting this October, WGMIXFISH identified discrepancies in the forecasting parameters for Northern Shelf cod. Experts from both WGMIXFISH and the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak (WGNSSK) collaborated to investigate the cause of the issue.
Why wasn't this problem detected previously?
The Northern Shelf cod stock underwent a benchmark review in 2023. ICES uses the benchmark process to peer-review and integrate new scientific knowledge, data, analyses, and assessment methods for use in the provision of recurrent advice in response to annual requests.
The outcome of that benchmark was a multi-stock assessment model which estimates the dynamics of three substocks separately. This improved the consistency of the assessment. Previously, each annual update tended to revise spawning stock biomass estimates downward and increase fishing mortality estimates. However, the new approach has introduced complex challenges for providing advice and managing the stock.
As part of the benchmark process, a new multi-stock stochastic forecasting model was developed. An error was found in the model that caused it to revert to long-term mean values for natural mortality, maturity, and stock weights more quickly than intended. This issue wasn’t detected sooner because those forecast parameters were difficult to retrieve from the model.
What process improvements will ICES make to ensure this won't happen again?
Quality assurance is central to ICES advice and a key part of our Advisory Plan. ICES is committed to continuously improving this process. The forecasting model for Northern Shelf cod now includes key diagnostic parameters as standard outputs. Moving forward, we will enhance the expert group audit process to include independent catch forecasts, such as those by WGMIXFISH. This will be made easier with broader adoption and improved standardization of tools and diagnostics in ICES Transparent Assessment Framework (TAF).
Has the basis for the advice changed?
The basis for the advice has not changed. The advised reduction in fishing mortality for the Northwestern and Viking substocks has increased from a 61% reduction to a 65% reduction to ensure consistency with the fishing mortality reduction advised for the southern substock.
