Report of the Study Group on the Bycatch of Salmon in Pelagic Trawl Fisheries (SGBYSAL) [9-12 March 2004 Bergen, Norway]

Contributors: Marianne Holm, Svein A. Iversen, Jens Christian Holst

Section 1. Gives terms of reference, lists the six participants from the three nations represented at the meeting (Ireland, Norway and Russia). Background for the Study Group was the observed large number of post-smolts taken together with large catches of mackerel in Norwegian research surveys in the Norwegian Sea (div. during June-August since the start of dedicated salmon surveys in 1995. These observations gave rise to an apprehension that the large commercial fisheries for mackerel in these areas might heavily intercept the post-smolt cohorts mowing northwards during the summer months. Russian observers on board commercial mackerel trawlers did, however, detect only negligible amounts of post-smolts in the catches screened, resulting in a large discrepancy in the estimates of post-smolts taken if the observed ratios were used when scaling up with the total catch in these areas (from 60 to > 1 million post-smolts taken). Based on quarterly catch data, the overlap between post-smolts and the fisheries in the Norwegian Sea was seemingly high, but the absence of disaggregated data (per week and statistical rectangle) impeded an assessment of the true overlap of post-smolts. When the Working Group on North Atlantic Salmon (WGNAS) in 2003 received indications that also the herring fisheries occurring in August in northerly areas of ICES areas might intercept adult salmon, the necessity to pool data from pelagic Working Groups and WGNAS in order to progress on these issues became apparent. Section 2. An overview of the marine distribution of salmon in the North East Atlantic is given. Some areas of denser distribution of post-smolts have been identified (1995–1997) in mid- June on the shelf areas west of UK-Scotland (IVa) and in the strong surface currents in the Faroe-Shetland Channel (VIa west). Another high-density migration area of post-smolt cohorts has been identified west and north of the Vøring Plateau in the Norwegian Sea (IIa). Distribution charts for post-smolts broken into weeks 16–20, 21–26, 27–31 and 32–36 illustrate the northward progression of post-smolt cohorts predominantly of Irish and UK origin (assessed from smolt ages read from scales and tagged fish), but it also shows areas where the knowledge basis is weak due to lesser number of cruises performed (Figure 2.2.2). I t is also pointed out that due to deficient knowledge of the distribution of pre-adults and adult fish, it is difficult to evaluate the degree of overlap salmon distribution and major fisheries occurring in the Norwegian Sea and the transition areas for homing salmon. Section 3. Describes the major pelagic fisheries in the Norwegian Sea, the North Sea and areas west and south of UK and Ireland. Disaggregated data for landings to the UK and Germany enabled a closer study of mackerel and herring fisheries in the western (VIa) and northern North Sea areas (IVa) per week and statistical rectangle. Possible areas of interception were detected, but due to lack of information on salmon distribution, an assessment of the impact on salmon could not be performed. In addition the catches are rather small at the time when the salmon are thought to move through these areas. For the other nations no disaggregated data were provided, and the Study Group used quarterly catch records from ICES working group reports. Due to lack of data, a complete overview of boats and gear types used could not be made. Section 4. The Study Group received some information on whether salmon have been detected in the screening of catches performed by various countries and this is summarized in Table 4.2.1. Some methods of screening catches for bycatches are described in more detail. Information of salmon registered from logbooks from catches landed in the Netherlands was provided to the Study Group. A large proportion of these “salmon” may be sea trout due to lack of proper headings to enable discrimination between different salmonid species. The bycatches are small but consistently occurring, and the largest bycatches have been reported from various types of trawl fisheries. Also in Iceland adult salmon have been observed in various trawl fisheries. These records indicate that bycatches of adult salmon may be more frequently occurring than previously thought. An EU regulation launched in 1998 rendered landing of salmon from non-static gear illegal and after that the Dutch recorded bycatches have gone down from more than1000 kg to less than 20 kg annually. The Study Group evaluated the advantages and constraints of different on board and land based screening methods and concluded that observer based screening on board following different protocols for different species is the most effective way of screening. Screening discards from filleting factories should also be explored as a source of information. Section 5. The Study Group explored analytical methods to allow catch rates of salmon in research surveys to be extrapolated to catch rates in commercial fisheries. Comparing research results form surveys and observer based screening on board, the Group considered that the best method of estimating bycatches of salmon would be derived from direct observation on board commercial pelagic fishing vessels performed according to agreed protocols. These estimates would be based on consistent gear types and fishing methods and would not require any assumptions about the transferability of research catches. A range of estimates of potential post-smolt bycatch based on the data available at the meeting were developed (Table 5.1) as illustration of the likely ranges of values obtained by applying different methods to the mackerel fishery in the Norwegian Sea. The purpose was to assess the performance of the various methods. There are obvious risks of over- or underestimations depending on which sources used for scaling up with commercial catch. These need to be discussed further when more adequate data sets are available. Section 6. The Study Group was presented with a method for estimating the progress in time and space of post-smolts cohorts in the Norwegian Sea, the “smolt passage model”. Based on data from research surveys on the north – south extension of the post-smolt cohorts, the minimum time the main concentration of post-smolts migration through a fishing area can be estimated at different anticipated migration speeds (1–2 body lengths s-1), and the time of overlap between fisheries and post-smolts can be assessed. The model is based on a rather limited set of data today, and needs further refinement when more data become available, but the group considers it a useful tool assessing and minimizing the risk of post-smolts being intercepted by the commercial fishery in the area of passage. Section 7. A set of ten recommendations for following up is given. These pertain to continuation of screening research and commercial catches for salmon, on board observer based screening of commercial catches and establishment of protocols for screening of salmon bycatches in catches of different species, screening of discards at filleting factories, development of methods of estimating of salmon post-smolt bycatches primarily via observer screening programmes on commercial fishing vessels. The application of a range of bycatch estimates to known data on salmon abundance and survival trends in the stocks in question is recommended to determine whether crude levels of potential bycatch can account for recent changes in abundance or survival at sea. Under a range of bycatch rate scenarios the scale and nature of any tagging programme that would be required to yield reliable estimates of bycatch should be determined. It is not considered appropriate to reconvene the SGBYSAL until disaggregated catch data for the mackerel fishery in the Norwegian Sea and for other fisheries and areas where such data are lacking become available.