Working Group on Improving Survey Data for Analysis and Advice (WGISDAA; outputs from 2020 meeting)

The Working Group on Improving use of Survey Data for Assessment and Advice (WGISDAA) provides a forum where survey practitioners, statisticians and stock assessment scientists can share expertise and knowledge to advance the science of fisheries surveys. The working group serves to bring together data collection and data users in a supportive and cooperative environment and to improve communication and coordination across survey groups in the ICES network.

The working group developed a strategy for making decisions on survey design and implementation that maximises the benefits of long-term survey data and minimises the impacts of enforced changes to survey design and implementation. WGISDAA concluded that the increasing use of model-based indices is beneficial in dealing with change in data collection but requires an approach that integrates across data collection and data use to control overall uncertainty in indices.

This is especially important in cases where multiple objectives are addressed by a single survey but also across different surveys. Existing practices for decision making are now documented and gaps in the strategy identified. Further workshops are being planned to develop the tools that quantify the trade-offs between different objectives and establish useful estimates of uncertainty for different survey designs / data collections.

WGISDAA assisted The Working Group on Mackerel and Horse Mackerel Egg Surveys (WGMEGS), The International Bottom Trawl Survey Working Group (IBTSWG), The Working Group on Beam Trawl Surveys (WGBEAM) and The Working Group on Surveys on Ichthyoplankton in the North Sea and adjacent Seas (WGSINS) with survey design questions, most notable was the cooperation with WGMEGS on developing an improved spatial interpolation for the egg abundance data given the considerable changes in the in spawning distribution and the associated consequences to data consistency due to organisational difficulties in providing adequate survey coverage.

A spatio-temporal annual generalized additive model (GAM) of egg distribution was developed which allowed for much more appropriate interpolation between samples and indicated that the total egg production and hence the standing stock biomass (SSB) was likely to be considerably smaller than previously thought.

However, the temporal trend in production which matters most to the stock assessment was largely unaffected by the improved model except for the 2013 data point. Sensitivity analysis of the egg production calculation suggests that other factors in the index particularly temporal changes in fecundity or changes in catchability in the trawl survey are more likely causes of the current stock assessment difficulties.

The group worked with stock assessors for North Sea cod, West of Scotland whiting and Celtic Sea gadoids. As for the mackerel assessment, cooperation was essential for resolving disparities in the information provided by different data sources. The group helped develop more appropriate combined indices that took account of changes in spatial distribution of the stock and the survey effort.

For North Sea cod however, this still proved to be a poor explanation of the underlying dynamics and further investigation suggested that there has been increased mixing with an adjacent population of cod particularly at the older ages. A more appropriate method to deal with this failure of the closed population assumption is being developed.