Ovegård, Mikael
- Institutionen för akvatiska resurser (SLU Aqua), Sveriges lantbruksuniversitet
Sammanfattning
Trawl fisheries are generally non-selective and often result in unwanted bycatches of juvenile fish, non-quota species, and protected marine life. The ecological consequences of this bycatch include direct mortality, habitat damage, and disruptions to marine food webs. Strategies for reduction of unwanted bycatch fall into two main categories: policy instruments and gear modifications. This essay focuses on the latter, particularly gear modifications that enhance size and species selectivity. Scientific evaluation of the effect of gear modifications on the catch composition in trawls has traditionally been conducted through either paired gear experiments, where the catch of the test gear is compared to the catch of a baseline/control gear, or by recapture of the escapees (e.g. by using codend covers or collection bags). One of the main drawbacks of these methods, apart from that they generally are expensive and difficult to conduct, is that it normally leads to the death of the unwanted catch. New approaches utilizing artificial intelligence and computer vision could potentially reshape the methodologies used in this field. Underwater camera systems can now monitor fish behaviour in situ, while machine learning models automatically detect, classify, and measure fish without physical handling. These technologies reduce experimentally induced mortality, improve data quality, and allow for scalable, non-invasive evaluation of gear modifications. Ultimately, the integration of artificial intelligence into fisheries science supports the broader aims of ecosystem-based fisheries management by providing insights into species-specific selectivity and escapee vitality. This has the potential to improve science-based decision making and promotes sustainable exploitation of marine resources in alignment with the United Nations Sustainable Development Goals.
Nyckelord
Bycatch; commercial fishing; gear selectivity; computer vision; ecosystem-based fisheries management
Publicerad i
Aqua introductory research essay
2026, nummer: 2026:2
Utgivare: Department of Aquatic Resources, Swedish University of Agricultural Sciences
SLU författare
UKÄ forskningsämne
Vilt- och fiskeförvaltning
Publikationens identifierare
- DOI: https://doi.org/10.54612/a.2tjuf8obt9
- eISBN: 978-91-8124-187-7
Permanent länk till denna sida (URI)
https://res.slu.se/id/publ/146279