Skip to main content
SLU publication database (SLUpub)

Abstract

The Excluder, a net-based sorting device, has been introduced in the North Sea industrial trawl fishery to reduce unwanted bycatch. This study evaluated the sorting performance of the Excluder by combining traditional catch composition analysis with automated video analysis of underwater recordings. A total of 31 hauls were observed across two years (2021 and 2022), using trawls with and without the Excluder. Underwater cameras recorded fish behaviour and escapement through the Excluder's exit opening, and an object detection model was trained to estimate number of escapees across four taxonomic groups. Results showed that the Excluder allowed nearly complete escapement of demersal species such as gadoids, flatfish, and gurnards, with estimated sorting efficiencies approaching 100 %. Pelagic species (herring and mackerel) were retained to a larger extent and frequently exhibited signs of injury, raising concerns about post-release survival. Although the automated video analysis likely underestimated true escapee numbers due to visual occlusion during mass exits, the method provided valuable behavioural and ecological insights and proved effective for assessing gear performance without the need to handle live fish and bring them out of their natural habitat.

Keywords

Selective fishing; Computer vision; Automated video analysis; Sorting efficiency; Excluder; Sandeel

Published in

Fisheries Research
2026, volume: 294, article number: 107666
Publisher: ELSEVIER

SLU Authors

UKÄ Subject classification

Fish and Wildlife Management

Publication identifier

  • DOI: https://doi.org/10.1016/j.fishres.2026.107666

Permanent link to this page (URI)

https://res.slu.se/id/publ/146150