Saturday, 14 November 2020

The ecological disturbance caused by fishing in the North Sea.

Theoretical models linking species diversity to disturbance consider disturbance to be the alteration of population mortality rates caused by the specified perturbation. Current knowledge of the ecological disturbance caused by fishing to marine fish and benthic invertebrate communities is reviewed. This review considers the various components of mortality caused by fishing (eg landings, discards, trawl escapees, etc) and examines the information currently available, or required, to determine the importance of each component in assessing the mortality of fish and benthos caused by fishing.

Landings data provide an estimate of part of the mortality of fish caused by fishing activity in the North Sea. Thus international landings databases were compiled for seven key target species by fishing gear, rectangle and year over the period 1997 to 2004.

Data were provided by theUK (England, Wales, Northern Ireland and Scotland), The Netherlands, Germany and Norway. Data were aggregated to the lowest common gear category.  Annual variation at the North Sea scale in the data supplied by these countries very closely matched similar data supplied to ICES for use in the annual stock assessment process, confirming the veracity of the MAFCONS dataset.

Landings data included in the MAFCONS data base represented a high proportion of totalNorth Sea landings, suggesting that fishing by these nations accounted for most of the fishing disturbance taking place in the North Sea. Of the seven main species, saithe and cod were the two with significant proportions of total North Sea landings taken by countries outside the MAFCONS consortium.

Spatial distributions of landings by all gears are reported by year, and by gear and country for two separate time periods.For components of the marine ecosystem beyond the targeted commercial species, landings data provide little indication of fishing disturbance  For these species, fishing disturbance must be modelled based on data that quantify levels of fishing activity; fishing effort data.

To do this, international fishing effort data bases were compiled for most countries participating in the MAFCONS project. Again these data were compiled by ICES rectangle, gear category, and year over the period 1997 to 2004 in terms of hours spent fishing.

For two countries, Scotland and the Netherlands, data, in terms of hours-effort, were not available and had to be modelled on the basis of days-absent from port.No landings or effort data were available for Belgian, Danish, French, and Swedish fishing vessels, countries that all prosecute significant fisheries in the North Sea. Landings and effort by these countries were therefore modelled.

Annual variation and spatial trends in these modelled data were examined and assessed by comparison with all available information.Generally the modelled output was deemed acceptable.

It was therefore considered advisable to include these data in the MAFCONS data base so as to examine variation in all fishing activity taking place in the North Sea, rather than simply just ignoring the significant components of fishing activity associated with countries that were not part of the MAFCONS consortium. MAFCONS reported and modelled total annual landings data closely matched landings data reported by ICES, suggesting that the MAFCONS database accurately reflected the actual fishing situation in the North Sea.

Total landings in the North Sea fell by 40% over the period1997 to 2004.

When this overall reduction in landings was taken into account, spatial distributions in the total landings of each species, and in the landings of each species by gear, were relatively consistent over time.

Total fishing effort declined by 28% since 1999, with some countries (eg England and Scotland) more affected than others. 

  • Beam trawling was primarily asouthern North Sea activity. 
  • Otter trawling occurred throughout the North Sea but was more prevalent in the northern North Sea.
  • Seine netting occurred mainly in the northern North Sea.

These distributions were consistent over time.Models were developed for both the fish and benthic invertebrate communities to convert the observed patterns of fishing activity to indices of fishing mortality. The benthic invertebrate model utilised recent meta-analysis studies to determine “per fishing event” mortality rates for various benthic invertebrate fauna.

Recent studies of the Dutch beam trawl fleet have suggested that at small spatial scale, the distribution of fishing activity follows a Poisson distribution.

Knowing tow velocity, tow duration, and gear width for each fishing metier, the mean fishing frequency in each rectangle can be determined from the fishing effort statistics. Application of the Poisson distribution then enables the distribution of fishing frequencies in 900small sub-units of each rectangle to be determined, which, given the “per fishing event” mortality rates, allows the total benthic mortality in each ICES rectangle to be estimated.

Rather than being linear, the relationships between fishing effort and mortality were strongly asymptotically curvilinear. The model was applied to the observed fishing effort data assuming a “generic”benthic invertebrate community and using actual observed epibenthic species abundance data.The model indicated that rectangles of moderate fishing effort had mortality rates almost as hig has high fishing effort rectangles.

Consequently the area of high benthic impact from fishing was more widespread that might have been suggested by the fishing effort data alone.The fish mortality model utilised swept area estimates combined with estimates of local abundance (density), and making assumptions about catchability, determined the likely number of fish taken in each fishing event.

The number of fishing events in each rectangle was estimated from the fishing effort statistics, knowing mean tow duration for each type of fishing activity. The model was used to determine annual rates of fishing mortality for each species recorded in the Dutch beam trawl survey. Spatial distributions in mortality were presented.Estimates of fishing disturbance to the fish community were also derived directly from the landings data.

Estimates of discard levels for each commercial species are provided each yearby the stock assessments. These were used to raise landings to total catch, with corrections applied to account for the non-target species. Total catches in each ICES rectangle were converted to “exploitation” rate indices by dividing them by estimates of the abundance of fish present.

To read the full project paper download from here