Workshop on Methods to develop a swept-area based effort index (WKSABI)

The Workshop on methods to develop a swept-area based effort index (WKSABI) was held at ICES HQ, Copenhagen, on 8–9 January 2019 under the co-chairmanship of Kai Wieland, Den-mark, and Henrik Degel, Denmark. The workshop was attended by 12 participants representing seven different countries. The objectives of the workshop were to adopt and agree on swept-area based effort indices for trawl surveys available in DATRAS and define and describe MFSD indicators using these swept-area based indices.

Surveys considered where the North Sea International Bottom Trawl Survey, the Northeast Atlantic International Bottom Trawl Surveys and the Baltic International Trawl Survey. MFSD indicators considered were Marine Litter and the Large Fish Indicator (LFI). In total, eight presentations were given addressing the Terms of References of the workshop.

Based on the initial presentations on data availability, quality check and swept-area based results for the different surveys and MFSD indicators, caveats were described, and methodological strat-egies to overcome existing data issues were defined. Swept-area based survey indices require tow-by-tow information of towed distance and door or wing spread from all survey participants.

For several reasons, this information is not always available and missing values have to be estimated. The level of data availability for this purpose differed considerably between the surveys. Despite the use of standardized survey trawl, gear geometry between countries or within countries between years can vary substantially and it was therefore concluded that algorithms for estimating missing values should be as specific as pos-sible for a given survey, vessel, country, and year combination.

This is important in order to minimize bias and variability when including the imputed values for the calculation of swept-area. After these requirements are met and datasets have been completed, further analysis on the potential of swept-area based indices for abundance and its use in stock assessment can be un-dertaken.

Since door spread has been much more often observed than wing spread and is usually measured with a low variability than wing spread, swept are based on door spread is preferred if indices rather than absolute estimates are required. The calculation of the MFSD indicators Marine litter and LFI require the availability of swept-area and this is then usually based on wing spread.

Since direct observations of wing spread area missing for the majority of surveys and years, generic models across years and vessels have been used for estimating the missing values. This may be appropriate when trends rather than partic-ular year effects are needed. Although estimates of MFSD indicators suffer of the problems men-tioned above, missing checks of data quality on marine litter submissions or the application of ‘global’ length-weight relationships for species for which no specific observations are available in the calculation of e.g.

LFI may be more important here.