What do pelagic freezer-trawlers discard? - ICES Journal of Marine ...

605

What do pelagic freezer-trawlers discard? Lisa Borges, Olvin A. van Keeken, Aloysius T. M. van Helmond, Bram Couperus, and Mark Dickey-Collas Borges, L., van Keeken, O. A., van Helmond, A. T. M., Couperus, B., and Dickey-Collas, M. 2008. What do pelagic freezer-trawlers discard? – ICES Journal of Marine Science, 65: 605 – 611.

Estimates of discards by the Dutch pelagic freezer-trawler fleet were derived based on data from observers on board commercial vessels. In all, 27 fishing trips of duration 2– 5 weeks each were sampled between 2002 and 2005, covering the North Sea and the western waters of the British Isles. Different methods to estimate discards were compared, and raising these estimates by number of trips or by total landings did not greatly influence the annual estimates of total discards. Some 30 000 t of fish were discarded annually by the fleet, the commercial target species mackerel, herring, and horse mackerel being discarded most, with 41%, 19%, and 15% of the total discards by weight, respectively. The most commonly discarded non-commercial species was boarfish, accounting for 5% of total discards. Slippage accounts for 10% of all discards, and the most common species slipped is herring. There is extensive discarding of juvenile mackerel in the horse mackerel fishery, and to a lesser extent highgrading of mackerel in the same fishery. Keywords: discards, herring, horse mackerel, raising, slippage. Received 6 June 2007; accepted 2 March 2008; advance access publication 3 April 2008. L. Borges, O. A. van Keeken, A. T. M. van Helmond, B. Couperus and M. Dickey-Collas: Wageningen-IMARES, Institute for Marine Resources and Ecosystem Studies, PO Box 68, 1970 AB IJmuiden, The Netherlands. Correspondence to M. Dickey-Collas: tel: þ31 255 564684; fax: þ31 255 564644; e-mail: [email protected]

Introduction Discarding in pelagic fisheries is more sporadic than in demersal fisheries. This is because of the nature of pelagic fishing, in which boats pursue schooling fish, take hauls with a low diversity of species and sizes, and consequently often show wide fluctuations in the rates of discarding (from 0% to 100%; ICES, 2006a). Many of the world’s fisheries for pelagic fish produce material for fishmeal and for bait (FAO, 2007), whereas Dutch freezer-trawlers fish exclusively for human consumption and process the fish on board. Therefore, the catch by freezer-trawlers is sorted by species, size, and quality. Discarding itself may be caused by limits on quotas, a low market value of some species, or species size (or quality) in catches. Only a few estimates of the extent of discarding of pelagic species in pelagic and demersal fisheries have been published. Discard percentages of pelagic species from demersal fisheries were estimated at 3 –7% (otter trawl and beam trawlers; Borges et al., 2005a) of the total catch by weight, and from pelagic fisheries at 1 –11% (pelagic trawl; Berrow et al., 1998; Pierce et al., 2002; Palsson, 2005; Hofstede and Dickey-Collas, 2006). Here, we define the term slippage as the act of discarding fish before sorting, i.e. the catch or the proportion of the catch not brought on board for processing. Estimates of slippage have only been published for the Portuguese purse-seine fishery targeting sardine, with values at around 60% of the total catch (Stratoudakis et al., 2002). However, most of the published estimates were raised using catch or effort variables, a method open to criticism (ICES, 2007a). The estimators normally used in demersal fisheries to raise sampled discard data to population levels, such as effortand catch-related variables, may not be applicable to pelagic # 2008

fisheries because effort may not relate in a linear manner to catch, and catches are often themselves underestimated (ICES, 2007a). We therefore also address the differences between raising methods for discards in pelagic fisheries, though could find no published studies on the issue. The Dutch pelagic freezer-trawler fleet contains some of the biggest fishing vessels in the world. The fleet consists of 14 vessels, of which six normally operate in European waters. The vessels average 107-m long and have an average engine horsepower of 7900. The height of the pelagic trawls deployed varies between 30 and 60 m, and the horizontal spread of the wings between 80 and 120 m. Tow duration is normally 4 h and trip duration some 2 –5 weeks. The fleet operates mainly in the Northeast Atlantic, off West Africa, and occasionally in the southern hemisphere. In European waters, the fleet targets mainly herring (Clupea harengus), mackerel (Scomber scombrus), horse mackerel (Trachurus trachurus), and blue whiting (Micromesistius poutassou). Dutch-owned freezer-trawlers also operate in European waters under German, UK, and French flags. The main objective of our study was to estimate the level of discarding by the Dutch pelagic freezer-trawler fleet using an observer programme, commercial and non-commercial species being recorded to investigate the range of species being taken. Simultaneously, we made an effort to evaluate the potentially best raising procedure.

Methods The Dutch pelagic discard sampling programme is a scheme targeted at commercial vessels in European waters using on-board observers (Figure 1) accommodated voluntarily by the fishing

International Council for the Exploration of the Sea. Published by Oxford Journals. All rights reserved. For Permissions, please email: [email protected]

Downloaded from https://academic.oup.com/icesjms/article-abstract/65/4/605/640309 by guest on 03 November 2017

606

L. Borges et al. Table 1. Sampling levels of the Dutch discard observers programme (Sampled) and the pelagic freezer-trawler fleet activity (Fleet) per year.

Figure 1. Map of the study area showing the location of hauls of the Dutch pelagic discard sampling programme sampled between 2002 and 2005.

fleet. Each observer records information covering position, depth, date and time, total catch composition, and total weight of all hauls (from the ship’s log). Discard estimates are made from as many hauls as possible. The process of estimating discards per haul has two stages: first, by estimating the percentage of discards, and second, by collecting samples of the discards and the unsorted catch. The total percentage of discarded fish by weight is estimated by comparing the rate of processing against the rate of discarding through the processing belts. Traditionally, a discard sample is collected from the gutter of the processing belt and sorted for species and length distribution. Another sample is taken from the unsorted catch. Apart from the routinely sorted discards, a part of the catch is sometimes released before the catch has been sorted. This is called “slippage”. A catch (or a proportion of it) can be pumped directly from the chilling tanks out to sea, or the codend of the net may be opened although the net is still in the water (Table 1). Slippage estimates were made by eye by the on-board observer, and its species and length composition was determined directly by taking a sample from the hold. In case of slippage from the codend, the composition of the catch was assumed to be similar to the preceding or following haul in the same area. An analysis of this assumption was carried out by comparing visually the catch composition of a haul’s slippage from the chiller tanks, where a sample was collected, with the catch composition of the preceding or the following haul in the same ICES rectangle. We found no major differences in species composition between the hauls where slippage was observed and in hauls in the same vicinity. However, we are aware that this assumption dominates the results for slippage, and realize that better methods need to be developed and included in subsequent studies. The observer programme started in 2002 with four trips sampled, increasing to 12 trips in 2005 (Figure 1). In all, 27 trips and 904 hauls were sampled for this analysis, some 4 –9% of the fleet activity during more than 3 years of this study. The discards sampled were raised at a haul level by multiplying the fraction of sampled discarded weight to total discarded weight


Parameter 2002 2003 2004 2005 Sampled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of vessels 4 5 6 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of trips 4 5 6 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of trips with 4 5 5 6 slippage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of hauls 150 198 207 349 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of hauls with 9 9 9 13 slippage (from net) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of hauls with 12 15 3 7 slippage (from chiller tanks) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Total landings (t) 7864 15 085 13 859 28 396 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Fleet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of vessels 14 14 13 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Number of trips 104 139 140 137 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Total landings (t) 180 040 228 749 290 571 297 607

in that haul. Total discards per trip (the sum of all hauls) were raised to population level by multiplying by the total number of trips made by the fleet per year (ratio estimator; see Borges et al., 2005b, for formulae). The landings (the sum of all species) were also used as a raising factor for the sampled discards, and were then compared with the discards raised by trip. It was clear during the analysis that the coarse resolution of the sampling meant that obtaining robust estimates of discarding by species by area was impossible, so the analysis was kept at the level of total discard by area, or by species. This does make these estimates difficult to use directly in a stock assessment which, using current methods, requires estimates of discards by age, by year, by stock, and by fishing area. Just adding poorly estimated estimates of discarded fish to the input of a stock assessment will probably reduce the precision of the stock assessment, though it may well reduce any bias (see Dickey-Collas et al., 2007), so it is important to estimate the precision of the estimates of discards.

Results The Dutch pelagic industrial trawl fishery discards annually some 30 000 t of fish (Table 2), 10% of its catch, or 1.3 billion individual fish (Table 3). Slippage accounts for just 10% of the total catch discarded (Table 2). In theory, an estimator using different auxiliary variables, but applied to the same data, should produce similar results. Both raising methods (by trip and by catch) gave similar results, but raising by trip gave more consistent results (between variables within years), but less precision (Figure 2). We note that raising with landings gave more precise results but was more sensitive in anomalous years, particularly where there were small sample sizes. In 2003, one of the five fishing trips sampled accounted for very little discarding, which caused high variability in the estimates for that year (Figure 2). Moreover, there is evidence of landing misreporting by area by all boats that fish for pelagic fish in northern Europe, and to some extent by species in these fisheries (ICES, 2006a, 2007b), so raising by trip was considered to be a more robust procedure. The discard estimates obtained

607

What do pelagic freezer-trawlers discard? Table 2. Total biomass discarded, with and without slippage, by the Dutch pelagic freezer-trawler fleet annually, raised to population levels using total numbers of trips and total landings (sum of all species). Year

Landings

Trips

t CV t CV With slippage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 32 456.8 22.3 29 679.3 39.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 27 641.8 65.6 39 275.8 39.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 23 061.8 17.9 21 064.7 27.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 33 395.1 16.4 30 373.2 23.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . Without slippage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 28 078.1 23.5 25 675.3 40.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 26 233.1 67.8 37 274.2 41.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 21 317.0 21.4 19 471.0 27.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 32 051.8 16.6 29 151.4 24.8 Slippage is the act of discarding fish before sorting, i.e. after being in the net but not brought on board, or after being in the chiller tanks.

Table 3. Total numbers discarded, with and without slippage, by the Dutch pelagic freezer-trawler fleet per area and year (raised using total numbers of trips). Year

Norwegian Sea

North Sea

Millions of fish

Millions of fish

CV

Western waters CV

Millions of fish

CV

With slippage . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 0* – 169 – 1219 56.2 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 26 – 581 – 1218 84.2 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 No data 214 95.7 539 35.3 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 43 92.1 166 84.7 1116 53.3 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . Without slippage . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 0* – 117 – 1054 56.1 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 No data 460 – 1213 84.6 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 26 – 136 94.1 509 40.0 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 43 92.1 146 82.8 1084 55.0 *Actually 0.01 ¼ 10 000 fish. –Only one trip sampled.

by area raised by landings differ substantially from estimates derived by trip. Using landings as the raising variable results in 20% more discards being estimated in western waters than in the North Sea, suggesting that 20% of the landings from the North Sea are misreported as bein taken in western waters. Consequently, for this reason too, all the results presented hereafter are based on discards raised by trip. An analysis of the total discards by area shows that the Dutch pelagic freezer-trawler fleet discards more in western waters (54% of the total catch), because there is more fishing activity there (Table 4). However, slippage occurs mainly when the fleet is operating in the North Sea, where slippage accounts for 13% of total discards by weight (25% by number), compared with just 5% slippage by weight (6% by number) in western waters. The CVs of the area estimates of discarding (35 –95%; Table 4) suggest that discards are currently too poorly estimated


Figure 2. Total numbers discarded by the Dutch pelagic freezer-trawler fleet annually, raised to population levels using total numbers of trips and total landings (sum of all species). Vertical lines represent the standard errors of the estimates.

Table 4. Total biomass discarded, with and without slippage, by the Dutch pelagic freezer-trawler fleet per area and year (raised using the total number of trips). Year

Norwegian Sea

North Sea

Western waters

t CV t CV t CV With slippage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 1.3 – 11 894.7 – 18 901.0 55.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 1021.5 – 27 047.1 – 19 695.4 63.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 No data 9082.1 92.5 15 665.5 25.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 2150.9 84.8 7698.8 69.9 19 842.2 30.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . Without slippage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2002 1.3 – 9786.7 – 16 501.0 55.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2003 1021.5 – 23 567.1 – 19 185.4 65.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2004 No data 7235.9 91.0 15 410.5 27.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . 2005 2150.9 84.8 7464.2 69.0 18 861.2 32.5 –Only one trip sampled.

to be further partitioned to numbers-at-age by stock, so cannot provide further information to improve the precision of stock assessments (Dickey-Collas et al., 2007). As expected, the diversity of discards is low, overall some 29 species, reflecting a low diversity in the catch (Table 5). During the 4 years investigated, 11 species were caught in just 1 year of the 4. Of the total of 30 000 t discarded annually, some 12 200 t were mackerel, 5600 t were herring, and 4600 t were horse mackerel. Blue whiting, sardine, and boarfish (Caprus aper) made a minor contribution to pelagic discards compared with the first three species (2400 t of blue whiting, and 1500 t each of sardine and boarfish; Figure 3). Slippage accounted for 18% of the herring discards and 17% of the horse mackerel discards, though it was extremely variable (from 0% to 43%), just 4% and 2% of blue whiting and mackerel, respectively. In terms of numbers discarded, mackerel was the most discarded species, followed by horse mackerel and boarfish, then herring, blue whiting, and sardine (Figure 4). By comparing the length frequencies of landed to discarded fish, it is clear that herring and blue whiting are not selected and discarded for length reasons, but that in sorting and processing

608

L. Borges et al.

Table 5. Average annual discards, raised using total numbers of trips, of all species discarded by the Dutch pelagic freezer-trawler fleet between 2002 and 2005. Species

Discard biomass

Discard numbers

t CV Thousands CV Engraulis 6.7 45 encrausilae . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Anglerfish Lophius spp. 7.7 42 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Blackfish Centrolophus 11.5 16 niger . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Blue Micromesistius 2402.5 65.4 27 278 64.4 whiting poutassou . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Boarfish Caprus aper 1540.8 48.7 54 599 53.5 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Conger Conger conger 0.8 1 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Golden Sebastes sp. 22.0 122.2 54 111.7 redfish . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Greater Argentina silus 661.8 76.4 2 966 78.3 argentine . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Greater Phycis phycis 6.9 3 forkbeard . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Grey Eutrigla 65.1 49.6 334 98.2 gurnard gurnardus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Haddock Melanogrammus 446.8 64.9 1777 74.9 aeglefinus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Hake Merluccius 61.9 92.1 139 59.3 merluccius . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Herring Clupea harengus 5672.0 49.8 30 595 69.4 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Horse Trachurus 4670.1 87.2 56 195 98.3 mackerel trachurus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . John Dory Zeus faber 153.5 154 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Ling Molva molva 1.7 1 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Lumpsucker Cyclopterus 11.8 13.5 14 13.1 lumpus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Mackerel Scomber 12 221.0 39.1 88 857 82.1 scombrus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Norway Trisopterus 14.2 142 pout esmarkii . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Pollack Pollachius 14.1 8 pollachius . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Saithe Pollachius virens 173.9 103.2 102 99.7 . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Sardine Sardina 1514.2 93.8 15 850 98.7 pilchardus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Seabass Dicentrarchus 5.0 3 labrax . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Silver Pterycombus 3.4 58.6 5 41.0 pomfret brama . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . 1469.4 141.3 4409 141.3 Striped red Mullus surmuletus mullet . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Tub Trigla lucerna 8.5 136.6 65 138.3 gurnard . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . White Pagellus sp. 49.7 131.8 395 135.4 seabream . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Whiting Merlangius 92.8 78.3 330 73.6 merlangus . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . Witch Glyptocephalus 0.2 1 cynoglossus Anchovy

CV is the coefficient of variation between years. CV values are missing where the species was discarded in just 1 year.


mackerel, discarding of small fish is common (Figure 5). Further analysis shows that most small mackerel (,25 cm) are caught from July to December, when the target species of the fleet is traditionally horse mackerel. Earlier in the year, there is evidence of highgrading of mackerel, mainly around the minimum landing size (Figure 5). By weight, the proportion of the total catch discarded was 32%, 6%, and 8% for mackerel, herring, and horse mackerel, respectively, whereas by number of fish, the proportions of herring and horse mackerel were similar but half the mackerel catch was discarded.

Discussion During our study period, the Dutch pelagic freezer-trawler fleet caught high volumes of a few species, but it did not discard a large proportion of its catch. Some 10% of the catch was discarded, showing that it is a selective fishery for its targeted species. Note that discarding is not illegal in the EU, whereas misreporting of landings and the landing of undersized fish is. The fishery is highly seasonal, the target species changing through the year. Most discards are of three target species, but also of bycatch species, because the definition of a bycatch species changes with season/area (e.g. herring caught during the mackerel season; mackerel caught in the season when horse mackerel are caught). Discarding behaviour is the result of an interaction between quota limits, target species, market price for fish, and fish size and quality. It has been argued before (ICES, 2006a, 2007b) that any assessment of the discard behaviour of pelagic freezer-trawlers must take into account the seasonality of the targeting of all species, and although it was not analysed comprehensively here, our study does suggest that a single-stock approach would not provide as much insight as an entire fleet-based approach. This is the first documentation of highgrading of mackerel. There is a great price differential between small and large mackerel, more so than for any of the other target species of the pelagic fleet. The Netherlands has a relatively small quota for mackerel (9% of the EU quota) compared with its North Sea herring and horse mackerel quotas (25% and 29% of the EU quota, respectively), so there is a strong financial incentive to discard mackerel that will not provide the best price when landed. Some of the mackerel discarded are close to the minimum landing size. In terms of a fisher’s short-term gain, it would be preferable to use a small quota to catch large fish of good quality, for which one would get the best price. Therefore, mackerel discarding by this Dutch fleet is a product of management by a single-species TAC of a fleet that has multispecies targets and whose quota on each species is not in proportion to their relative catchabilities. The bycatch of small mackerel in the season when horse mackerel are being targeted suggests that the horse-mackerel fishery operates in an area with a great abundance of juvenile mackerel. Most mackerel caught then are indeed under the minimum landing size, so cannot be landed, with clear implications for managing this fishery sustainably. Although raising sampled discards by total landings resulted in greater precision of the final estimates than raising by number of trips, raising by total landings also resulted in greater sensitivity to anomalous outliers in the dataseries. Moreover, the fisheries on herring and mackerel in the EU have a long track-record of misreporting landings, either in terms of quantity landed or in terms of the area where the catch was taken (ICES, 2006a). This makes any decision to raise discards by official landings questionable. Therefore, despite the loss of precision, we conclude that

What do pelagic freezer-trawlers discard?

609

Figure 3. Biomass discarded by the Dutch pelagic freezer-trawler fleet annually (raised using total numbers of trips) for the six most discarded species. Vertical lines represent the standard errors of the estimates.

Figure 4. Numbers of fish discarded by the Dutch pelagic freezer-trawler fleet annually (raised using total numbers of trips) for the six most discarded species. Vertical lines represent the standard errors of the estimates.

raising by trip would be the most defensible method of estimating discards by this fleet. Slippage is thought of as an important component of discarding by pelagic trawlers and purse-seiners. The problems in estimating slippage are regularly used as an excuse not to estimate discards. In the current study, observers were asked to approximate what was being lost from the net if it was not brought on board. Although such estimates may be only a crude approximation, our results suggest that the volume of slippage is low, at 3000 t year21, mainly in the North Sea. However, anecdotal information does suggest that skippers are on “good behaviour” when observers are aboard, particularly about slippage.


Therefore, there may well be an additional bias in these estimates that is difficult to quantify, so these estimates of slippage should be viewed as a minimum. Herring and horse mackerel are the species most subjected to slippage, but the reasons for this are unclear. There may also be slippage when larger quantities of boarfish or horse mackerel are caught mixed with other fish, because such spiny fish lower the quality of the catch in the chilling tanks. As mentioned above, there are very few documented studies of the discarding of pelagic fish, and although some estimates are included in stock assessments (from Germany and the UK; ICES, 2006a, 2007b), such estimates are not raised to the total catch or the total number of trips by the fleet, so are not

610

L. Borges et al.

Figure 5. Length frequencies of discarded (filled histograms) and landed fish (white histograms) by the Dutch pelagic freezer-trawler fleet between 2002 and 2005 (summed over years and all areas), for the six most discarded species. Note that although sardine are landed (2400 t year21) into the Netherlands by the Dutch fleet, the landings are not sampled. Note that the y-axis values are different between species. For mackerel, those discarded from January to June are shown separately from those discarded between July and December.

comparable with this study. It is often assumed that adding extra data will improve stock assessments, but care needs to be taken when those data are poorly estimated, as is the case here (Cotter et al., 2002; Rochet et al., 2002; Punt et al., 2006). Although they did not raise the estimates of discarding to the total fleet, Pierce et al. (2002) found little evidence of discarding (,10%) and a similar species mix to that documented here. Berrow et al. (1998) reported low discard rates (4.7%) from the herring fishery in the Celtic Sea, and a similar mix of non-target species as seen in this study. In terms of herring and blue whiting, the estimates from the current study are not that dissimilar to the estimates from the same fleet off West Africa when fishing for Sardinella (Hofstede and Dickey-Collas, 2006). The fishery of West Africa is not limited by quota, and there is no major reward in terms of the size of the fish caught.


Pelagic fisheries such as this Dutch freezer-trawler fleet are often described as relatively clean in terms of non-commercial bycatch (ICES, 2006b). The freezer-trawlers appear to catch regularly 18 species in small quantities (Table 5), most of them also exploited commercially by other fleets. The boarfish is the most commonly caught non-commercial fish (1500 t year – 1), though the impact of this catch on the population of boarfish is unknown. We only analysed fish species caught by the freezertrawlers, and although we conclude that discarding is generally low, we did not analyse the bycatch of charismatic species such as dolphins, turtles, whales, birds, and sharks. Zeeberg et al. (2006) reported frequent catches of sea mammals and sharks by Dutch pelagic freezer-trawlers off Northwest Africa. For those species, of course, a catch of only a few individuals may be socially and/or conservationally unacceptable.

What do pelagic freezer-trawlers discard?

Acknowledgements The study was funded through the Discard Project, a project co-funded by the Department of Fisheries of the Netherlands Ministry of Agriculture, Nature and Food Quality, and the European Commission (DG Fish). We thank Ciaran Kelly and an anonymous referee for their constructive comments.

References Berrow, S. D., O’Neill, M., and Brogan, D. 1998. Discarding practices and marine mammal by-catch in the Celtic Sea herring fishery. Biology and Environment: Proceedings of the Royal Irish Academy, 98B: 1 – 8. Borges, L., Rogan, E., and Officer, R. 2005a. Discarding by the demersal fishery in the waters around Ireland. Fisheries Research, 76: 1 – 13. Borges, L., Zuur, A. F., Rogan, E., and Officer, R. 2005b. Choosing the best sampling unit and auxiliary variable for discards estimations. Fisheries Research, 76: 29– 39. Cotter, A. J. R., Course, G., Buckland, S. T., and Garrod, C. 2002. A PPS sample survey of English fishing vessels to estimate discarding and retention of North Sea cod, haddock, and whiting. Fisheries Research, 55: 25– 35. Dickey-Collas, M., Pastoors, M. A., and van Keeken, O. A. 2007. Precisely wrong or vaguely right: simulations of the inclusion of noisy discard data and trends in fishing effort on the stock assessment of North Sea plaice. ICES Journal of Marine Science, 64: 1641– 1649. FAO. 2007. The state of world fisheries and aquaculture in 2006. FAO, Rome. 180 pp. Hofstede, R., and Dickey-Collas, M. 2006. An investigation of seasonal and annual catches and discards of the Dutch pelagic freezertrawlers in Mauritania, Northwest Africa. Fisheries Research, 77: 184– 191.


611 ICES. 2006a. Report of the Working Group on the Assessment of Mackerel, Horse Mackerel, Sardine and Anchovy. ICES Document CM 2006/ACFM: 36. 650 pp. ICES. 2006b. Report of the ICES Advisory Committee on Fisheries Management, Advisory Committee on the Marine Environment and Advisory Committee on Ecosystems 2006. ICES Advice, 1 – 10. 1680 pp. ICES. 2007a. Report of the Working Group on Discard Raising Procedures, 6 – 9 February, San Sebastian, Spain. ICES Document CM 2007/ACFM: 06. 56 pp. ICES. 2007b. Report of the Herring Assessment Working Group for the Area South of 628N. ICES Document CM 2007/ACFM: 11. 554 pp. Palsson, O. K. 2005. An analysis of by-catch in the Icelandic blue whiting fishery. Fisheries Research, 73: 135 –146. Pierce, G. J., Dyson, J., Kelly, E., Eggleton, J. D., Whomersley, P., Young, I. A. G., Santos, M. B. et al. 2002. Results of a short study on by-catches and discards in pelagic fisheries in Scotland (UK). Aquatic Living Resources, 15: 327– 334. Punt, A. E., Smith, D. C., Tuck, G. N., and Methot, R. D. 2006. Including discard data in fisheries stock assessments: two case studies from south-eastern Australia. Fisheries Research, 79: 239– 250. Rochet, M-J., Pe´ronnet, I., and Trenkel, V. M. 2002. An analysis of discards from the French trawler fleet in the Celtic Sea. ICES Journal of Marine Science, 59: 538 – 552. Stratoudakis, Y., and Marçalo, A. 2002. Sardine slipping during purseseining off northern Portugal. ICES Journal of Marine Science, 59: 1256– 1262. Zeeberg, J., Corten, A., and DeGraaf, E. 2006. Bycatch and release of pelagic megafauna in industrial trawler fisheries off Northwest Africa. Fisheries Research, 78: 186 – 195. doi:10.1093/icesjms/fsn050