Technical Measures to Enhance Selectivity in Pelagic Fisheries
Model of the separator grid used in the trial (Swan Net-Gundry Ltd).
This project was undertaken under the Scottish Industry Science Partnership (SISP) in partnership with the Shetland Fishermen’s Association, and with the assistance of Seafish and the Marine Scotland Science Marine Laboratory in Aberdeen.
The principal aim of the project was to investigate the effectiveness of a selectivity grid in a pelagic trawl net as a means of reducing the proportion of smaller fish, and the bycatch of herring, in catches of mackerel.
Start Date and Duration
The project was undertaken during the Autumn 2007 mackerel fishing season.
The project was funded by the through the Scottish Industry Science Partnership (SISP).
It is considered that the size of mackerel encountered during the October 2007 season, together with the low bycatch levels that characterised the season, resulted in the full potential of the selectivity grids not being realised. A further development within the fleet was the widespread adoption of sampling identified shoals of mackerel for size using handline or automated jigging machines, prior to a decision being made on whether or not to shoot and target that shoal. This resulted in the avoidance of shoals that were, based on sampling, expected to result in a catch of mackerel with a smaller than desirable average weight (grams). However, it would have been these shoals where the effectiveness of the grid could have been best determined. Mackerel shoals were found to be plentiful and so quotas were caught in a
short space of time and in relatively few hauls and trips. The short duration of the season limited the amount of data that could be collected and the occurrence of foul hauls and gear damage further limited the amount of data available for analysis.
It was confirmed, from both underwater video footage and from obtaining samples in coverbags over the grids that both herring and mackerel were escaping through the grids. However an analysis of the available data suggested that the numbers of fish escaping were insufficient to significantly affect the retained catches of the vessel using the grid compared to the catches of the pair vessel. A smoother analysis indicated that the grids did not have any statistically significant effect on catch rates at any weight between the pair trawlers; however the low number of paired hauls meant that the comparison had little statistical power. Catches generally had very low bycatch levels during the season and no statistically significant difference in discard rates was found between the two pair vessels.
During the trip where video footage was obtained, only one of the hauls occurred during daylight hours and hence escapes through the grid on the bottom panel were not verified. Shoaling behaviour was observed almost immediately in mackerel that escaped through the grid. Escaping mackerel can also be described as showing strong swimming behaviour after they escaped. Mackerel sampled from the coverbags were significantly smaller than those retained in the codend.
It is concluded that although small mackerel and herring can and did escape through the grids, the data that was available from the October 2007 season did not show that there was a statistically significant difference in average sizes and weights in catches retained in the codends of vessels using the grid compared to control vessels. It is therefore not possible at this stage to assess the cost effectiveness of the grids. It is recommended that catches should be monitored over a longer period of time, and in both the autumn and spring seasons, before the true effectiveness of the grids can be fully ascertained.
Download the Project Report
Paul MacDonald (firstname.lastname@example.org)