Ref. Ares(2018)3458869 - 29/06/2018
Ref. Ares(2019)2387732 - 04/04/2019
Survival of plaice caught and discarded by Belgian beam trawlers
Confidential internal nota requested by ir. Marc Welvaert
Vlaamse overheid
Departement Landbouw & Visserij
Afdeling Kennis Kwaliteit en Visserij
Dienst Zeevisserij
Authors: Sebastian Uhlmann, Bart Ampe, Noémi Van Bogaert, Els Vanderperren, Els Torreele, Hans Polet
Institute for Agricultural, Fisheries and Food Research
Animal Sciences Unit - Fisheries and Aquatic Production
Ankerstraat 1
8400 Oostende
www.ilvo.vlaanderen.be
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 2
Disclaimer
This report cannot be published or disseminated publicly without the approval of the authors. The
data in this report are still under revision and should be interpreted with care.
ILVO Animal Sciences does not accept any liability for damage caused by the use of research
findings and/or opinions supplied by or on behalf of ILVO. The information is provided “as is”.
ILVO makes no warranty, express or implied, regarding the accuracy or completeness of the
information. In no event shall ILVO be liable for (a) any damages (including, without limitation,
damages for loss of profits, business interruption, loss of programs or information, and the like)
arising out of the application, abuse or inadequate use of the information provided by ILVO (b)
any claim attributable to errors, omissions, or other inaccuracies in the information provided by
ILVO, or (c) any claim by any third party.
2
link to page 4 link to page 4 link to page 7 link to page 7 link to page 7 link to page 8 link to page 9 link to page 9 link to page 9 link to page 9 link to page 10 link to page 11 link to page 11 link to page 12 link to page 15 link to page 17 link to page 21 link to page 21
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 3
Contents
1
Overview of the Belgian beam-trawl fishery ............................................................................................................... 4
1.1
Introduction ......................................................................................................................................................................... 4
1.2
Data for 2017 ....................................................................................................................................................................... 7
1.3
Managing discards ........................................................................................................................................................... 7
1.3.1
Landing Obligation ............................................................................................................................................... 7
1.3.2
Selective discard measures .............................................................................................................................. 8
2
Survival research .......................................................................................................................................................................... 9
2.1
Methodology ........................................................................................................................................................................ 9
2.1.1
Selection of participating vessels ................................................................................................................ 9
2.1.2
At-sea sampling and monitoring ................................................................................................................. 9
2.1.3
Statistical analyses .............................................................................................................................................. 10
2.2
Results & Discussion ...................................................................................................................................................... 11
2.2.1
Immediate mortality ........................................................................................................................................... 11
2.2.2
Delayed mortality ................................................................................................................................................ 12
2.2.3
Predicted and total mortality ...................................................................................................................... 15
3
Appendix ..........................................................................................................................................................................................17
4
Acknowledgements ................................................................................................................................................................... 21
5
References ...................................................................................................................................................................................... 21
3
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 4
1
Overview of the Belgian beam-trawl fishery
1.1
Introduction
While the Belgian fishing fleet is relatively small, its activity is widespread fishing in both the
North Sea and the English Channel as well as the Western waters and the Bay of Biscay (ILVO,
2016). Over the past 60 years the number of vessels has continuously decreased and since 1990,
also total engine power and tonnage has decreased. In 2008, total landings were at a minimum.
Since then landings have gradually increased again (ILVO, 2016). In 2016, Belgian vessels landed a
total of 24.583 tonnes in Belgian and foreign harbours (Departement Landbouw & Visserij, 2017).
After years of steady increase, this amount is now comparable to what was landed 15 years ago.
European plaice (Pleuronectes platessa) and Common sole (Solea solea) dominate the landings,
namely 8.946 tonnes and 2.481 tonnes, respectively (Figure 1A, Departement Landbouw & Visserij,
2017). However, in terms of value, sole is clearly the main target species of the Belgian fleet (Figure
1B). A considerable proportion of the Belgian landings (7.870 tonnes) is sold in foreign harbours,
mainly to the Netherlands, the UK and France. Most of these foreign landings are crustaceans and
molluscs, particularly, langoustine (Nephrops norvegicus, 671.131 kg), brown shrimp (Crangon
crangon, 658.210 kg) and scallops (Pecten maximus, 414.711 kg, Departement Landbouw & Visserij,
2017).
The Belgian fishing fleet consists of vessels of either ≤ or > 221 kW engine power and is therefore
segregated into a small (KVS) and large (GVS) fleet segment, respectively. These two segments are
further subdivided according to the type of fishing gear (ILVO, 2016). Both the small and large
segments of the Belgian fishing fleet are dominated by beam trawlers (Figure 2). Among the 71
licensed and active Belgian beam trawlers in 2017, 37 are classified into a small (≤221 kW), and 34
into a large (>221 kW) engine power segment (Departement Landbouw & Visserij, personal
communication Martine Velghe, 26-3-2018). Each of these fleets has a distinct fishing pattern: for
instance, coastal vessels of the small fleet segment spend ≤48 hours at sea fishing, targeting
demersal fish between March and June (70-99 mm mesh size), off the Belgian coast in the Southern
North Sea (Figure 3).
In conclusion, the Belgian fishing fleet is highly diverse with respect to its activity ranging across a
large geographical area which is characterized by different seafloor substrate types (Figure 4).
4
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 5
Figure 1 Top 10 fish species with respect to A) landed weight (in tonnes) and B) value (in x.1000
EUR) and for the Belgian fisheries in 2016 (Modified from Figures 6 & 11 in “Aanvoer & Besomming
2016”, Departement Landbouw & Visserij).
Composition of the Belgian fishing fleet
9%
2% 2%
Beam trawl
Bottom otter trawl
Gillnet
Others
88%
Figure 2 Share of the most important fishing techniques used in the Belgian fisheries (ILVO, 2012).
5
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 6
Figure 3 Heat map of the number of fishing hours (based on filtered VMS pings of Belgian beam
trawlers (mesh size 70-99 mm) in 2017.
6
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 7
1.2
Data for 2017
In 2017, Belgian beam trawlers spent most effort (i.e. number of fishing hours) in ICES-area 7d
(Figure 3): an area mainly characterized by rocks and hard substrate (Figure 4). The Eastern English
Channel (7d) was attributed to the second highest fishing quota (in tonnes) during the past three
years after ICES-area 4 (North Sea) (Appendix Table 1). Table 1 (Appendix) also shows the weight-
based (in tonnes) discard ratios (DR) for plaice in five different ICES sub-Divisions for two beam-
trawl métiers with different mesh sizes. The discard ratio is highest for beam trawling with 70-99
mm mesh size in ICES sub-Division 7a (DR = 0.57). In ICES sub-Divisions 4 and 7d the highest numbers
of discards are between 0-3 years, most landed fish are between 4 and 7 years old.
30000
25000
sr 20000
Till
hou g
Seabed
15000
Sand to muddy sand
Fishin
Rock or other hard substrata
10000
Mud to sandy mud
Mixed sediment
Coarse sediment
5000
0
3a
4b
4c
7a
7d
7e
7f
7g
7h
ICES-area
Figure 4 Mean number of fishing hours per substrate type and per ICES sub-Division between 2015
and 2017.
1.3
Managing discards
1.3.1
Landing Obligation
Discarding is a widely recognized problem in many fisheries, particularly in multi-species trawl
fisheries (Kelleher, 2005). Discards have become more important in the public eye in Europe with
increasing public awareness to ocean conservation, with the intensification of overexploitation of
fisheries resources, and, recently, with concerns raised about widespread discarding of commercial
species (Borges 2015). To address the discarding problem in European (EU) waters, a Landing
Obligation (LO) was introduced as part of the reformed EU Common Fisheries Policy (CFP,
European Union, 2013).
7
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 8
The LO requires all catches of regulated commercial species on-board to be landed and counted
against quota. This radical change in fisheries management aims to improve fishing behaviour
through avoidance of unwanted catches and also through improvements in selectivity. However,
discarding of fish and crustaceans at sea may still be allowed if it can be scientifically proven that
they stand a high chance to survive the capture-and-discarding process (high survival exemption).
In response to a European Commission request, an ICES working group was established to provide
guidance on how to quantify this (unaccounted) survival or mortality probability.
1.3.2
Selective discard measures
To improve avoidance, selectivity and survival of unwanted bycatch/discard species, different
measures may be implemented. The effectivity of these measures varies mainly according to the
type of fishery, gears used, and species of interests. For the Belgian fleet, advances in the reduction
of ecological impact in the past decade were particularly resulting from fleet reduction and gear
modifications and/or replacements to reduce fuel consumption (Depestele, 2015). Gear
modifications focused on reducing fuel consumption by reducing seabed interactions, for instance
by using lighter beam trawls, Sumwings or outrigger trawls (Depestele, 2015). Gear modifications
to increase selectivity were during this period not introduced, except for a panel of larger mesh
sizes in the top panel (Departement Landbouw & Visserij, 2013). This “Flemish panel” is a 3-m long,
large mesh (120-mm) panel in front of the codend to reduce the retention of <MCRS sole (Bayse &
Polet, 2015). Since January 2016 all Belgian beam trawlers are obliged to use this panel to allow
escape of undersized sole (Bayse & Polet, 2015). Additionally, when targeting brown shrimp
(Crangon crangon), coastal vessels are obliged to use bycatch reduction panels (BRD, named
“zeeflap”) between Dec 1 and May 31 with 16-31 mm nets. Such panels separate and exclude
unwanted bycatch of invertebrate and fish species from the shrimp catch. Another modification,
is a wing profile to replace conventional beams (termed “seewing”). The most promising
modification to both reduce discards, reduce fuel consumption, seabed impacts and in turn
increase catch efficiency is the pulse trawl, which is currently banned in Europe and only used
based on experimental exemption licenses. After conducting a comparative fishing experiment,
Van Marlen et al. (2014) showed that pulse fishing resulted in fewer discards (57%) compared to
conventional beam trawls, including 62% undersized plaice and 80% benthic invertebrates (Van
Marlen et al. 2014).
8
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 9
2
Survival research
2.1
Methodology
2.1.1
Selection of participating vessels
Commercial vessel operators were invited via an open letter to participate in the survival project
in 2015/2016 this study. Out of the pool of volunteers, a selection was made based on the following
criteria: i) sufficient space and suitable infrastructure on board (i.e. copper-free deck water pumps
with a guaranteed continuous > 25 L discharge volume) to accommodate two scientific observers
and their equipment; ii) fishing gear configuration typical for Belgian beam trawlers; iii) activity
within a management-relevant fishing area; and iv) willingness by the crew to co-operate. Vessel
owners were compensated for any costs accrued due to loss of fishing time and/or catch. For
three extra trips in 2017/2018, vessels were selected based on their activity patterns and voluntary
commitment.
Five trips were done with a commercial coastal Belgian beam trawler (coastal fleet segment)
and five trips with a Eurocutter (small fleet segment) targeting sole in waters of the Southern
North Sea and English Channel between November 2014 and September 2015. Both the coastal and
Eurocutter beam trawlers were conventionally rigged with two 4-m beam trawls with chain mats
(each weighing ~1200 kg). Five trips were done with three vessels from the >221 kW segment fishing
in the North Sea (two trips in July/August 2015), English Channel and Celtic Sea (three trips in
October/November 2017 and January 2018) with beam trawls between 11 and 12 m in length. Two
vessels used conventional chain-mat gears and the other vessel a sumwing with an estimated
single-beam gear weight of 6000 kg and 4500 kg, respectively. All vessels used 80 mm diamond
mesh codends.
2.1.2
At-sea sampling and monitoring
The following technical, environmental and biological data were recorded for each trawl: start
and end positions; trawl duration; wind force; average water depth; and catch weight. Catch
weights of plaice were estimated in kg and where necessary ‘number of boxes’ converted to
weights by a factor of 35 kg. Additional measurements such as maximum wave heights, air and
surface seawater temperatures were received from weather stations “Trapegeer”, “Westhinder”,
and “Wandelaar”, respectively (Flemish Government, 2015). A sample of plaice discards from these
fifteen commercial Belgian beam-trawl trips was assessed for reflex responsiveness severity of
injury types to establish an individual’s vitality when it arrived on deck (see Uhlmann et al., 2016
for details). Where possible all undersized plaice were collected inside a water-filled 244-L white
PVC container positioned at the end of the conveyor belt. At the end of the sorting process, >30
plaice were randomly selected. Ten of which were assessed for their reflexes and injuries as
described below, and from another 20 the ratio of alive and dead was determined. An individual
that did not respond to its tailfin being grabbed and which did not show any breathing movements
of the operculum and/or mouth was considered to be dead. A sub-sample of fish were monitored
in both water-filled on-board and lab-based containers (for delayed mortality up to 14 days).
9
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 10
For the quantification of delayed mortality, batches of between five undersized, < 27 cm MLS
plaice (22 ± 3 cm TL, Table Appendix 2) were collected from the sorting conveyor at the beginning,
mid- and end points of the sorting process (see Uhlmann et al., 2016, for details) except for an
extra batch collected straight from the hopper (a large container holding the discharged catch
after the codend was emptied), placed into a 10-l ambient seawater-filled PVC bucket and within
~5 min tested for their reflexes. Air exposure for each fish was expressed as the minutes it spent
on deck before being placed inside a water-filled bucket plus one third of the handling time during
reflex testing (because 1/3 of the reflex tests were done in air out of the water-filled container).
Prior to a commercial trip, fish were sourced from the R/V Simon Stevin, transferred to 124-L
monitoring containers and checked daily for any mortality. Between 10 and 20 individuals were
brought on-board for a trip and considered as controls.
Both controls and caught-and-discarded fish were assessed for their reflex responsiveness and
injuries. A response to a reflex stimulus was scored as present (unimpaired, 0) when clearly visible,
or absent (impaired, 1) when not visible, weak or in doubt within 3 s of observation. Reflex and
injury assessments took approximately 30 seconds per fish. Each plaice was scored for the body
flex, righting, head complex, evasion, stabilize, and tail grab reflexes, and a presence/absence of
injury (e.g. extent of point bleeding and bruising around head or body; scored on a 4-point
categorical scale). These injuries were associated with mortality in earlier work (Depestele et al.,
2014a, J. Depestele, pers. comm.). Fish that were unresponsive to all reflex tests were considered to
be dead and registered as immediate mortality. All reflex-tested fish were length-measured to the
nearest cm of TL and all alive fish T-bar (29 x 8 mm) anchor tagged with Bano’k© guns in the
dorsal musculature following McKenzie et al. (2012). All tagged plaice were ‘discarded’ into stacks
of independently arranged, water-filled, onboard 30-l monitoring containers (60 cm L x 40 cm W
x 12 cm H; Figure 1) and at the end of a trip transferred within <2 h to laboratory-based, 124-l
monitoring containers for 14 days of at least daily monitoring. Fish were offered defrosted brown
shrimp (Crangon crangon) as food at <5% of their biomass after 7 days of monitoring. Any food
remains and/or dead fish were removed and the time to mortality noted. Fish were monitored
three times per day within the first and daily within the second week of monitoring. At each
monitoring interval, the status of a fish was classified as either alive (0) or dead (1). The
corresponding impairment score for each fish was calculated as the mean score of impaired
reflexes (and present injuries – in experiment 3).
2.1.3
Statistical analyses
Data were analyzed separately for immediate and delayed mortality. Immediate mortality was
analyzed as a function of independent biological, technical and environmental factors such as
single gear weight, beam length, gear deployment duration, total catch weight, sea surface
temperature, air exposure, sorting and handling durations, and sediment type, using generalized
logistic regression (GLM). Based on correlations, if any, among the explanatory variables and
univariate GLM analyses, a selection was made on which factors to include for a full candidate
model.
10
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 11
For the analysis of delayed mortality, non-parametric Kaplan-Meier curves were used to plot
survival as a function of time to illustrate when mortality began to level off and to explore which
biological, technical or environmental factors may contribute towards mortality. To explore which
of the status indicator scores (i.e., damage class, reflex index, reflex and injury index, individual
reflexes) may be associated with delayed mortality, logistic regression models were used to model
the probability of dying (within 9 days of monitoring) as a linear function of the above explanatory
variables. Univariate analyses of variance (type III ANOVA) were used to test the null hypothesis
(H0) of no differences between the probability of dying, at 9d of monitoring, of fish at each status
indicator’s score level. To select the most discriminating combination of (non-collinear) reflexes,
the % dead fish out of the total which were reflex tested as either un-/ or impaired for a given
reflex were plotted against the ratio for when another given reflex was either un-/ or impaired to
graphically depict whether the risk of dying was generally greater when a reflex became impaired
and whether this held true, also when another reflex was already recorded as impaired. To validate
the logistic regression model which included the number of absent body flex, righting and evasion
reflexes as one of the best explanatory variables, a training set was built from all observations,
except those from a given trip. Predicted average survival probabilities per number of absent reflex
were compared to observed probabilities.
Deviations between 1 and 2% were considered to be a good fit. Based on the validated
relationships between reflex impairment (i.e., the number of absent body flex, righting and evasion
reflexes and also reflex impairment index) and mortality, average survival probability was
predicted using logistic regression for those trips were undersized plaice were scored for reflex
impairment and injury, but not monitored in captivity for recording any delayed mortality. Model
parameters were estimated based on a training set which comprised all reflex observations of fish
that were monitored for delayed mortality (up to 9 days). Based on these model parameter
estimates, mortality was predicted for all the reflex-scored fish sampled during the trips where no
monitoring for delayed mortality took place. Total mortality was calculated based on the
proportion of fish that were dead immediately on-board plus the remaining number of fish from
the total (1-immediate mortality) multiplied by either the observed or predicted delayed mortality
proportion.
2.2
Results & Discussion
2.2.1
Immediate mortality
From 147 trawls and 15 trips of five Belgian beam trawlers a total of 4815 undersized plaice (24
± 5 cm, mean ± SD) were assessed for their on-board immediate mortality status. Immediate
mortality was low for most trips, except those with vessels from the >221 kW fleet segment (ranging
between 8 and 62%). Immediate mortality was greater during monitored trips in the summer
compared with autumn and winter (Table 1c). Due to collinearity with gear deployment duration
and total catch weight, sorting duration was excluded from the analysis. Immediate mortality was
positively related with gear deployment duration (broken line in the slope from >80 min; p < 0.001)
and sea surface temperature (p < 0.1), with an adjusted R-square of 37%.
11
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 12
To test whether the assumptions and choice of statistical model were appropriate, the residuals
were examined through a frequency distribution (Figure 5B) and a QQ-plot (Figure 5C). Figure 5B
shows that residuals are normally distributed, although slightly skewed to the right (Figure 5B).
Small departures from the straight line in the QQ-plot are common and were expected (Figure
5C)., Hence, based on these Figures, it was concluded that the assumption for normality was met.
Figure 5 Average immediate mortality (%) per monitored trawl in relation to A) gear deployment
duration (in min). Dashed red line indicate 95% confidence interval around the smoother
regression line (red); green line corresponds to a linear regression line. B) frequency distribution
of the residuals illustrating for normality; C) QQ plot of the residuals for validation.
2.2.2 Delayed mortality
To quantify any delayed mortality, 908 plaice (from 53 deployments; 22.6 ± 3.1 cm TL, mean
± SD) were assessed for injury and reflex impairment and subsequently ‘discarded’ into holding
containers. Of these, 236 and 672 were randomly selected from short (<40 min) and conventional
trawls (~60 min gear deployment duration for the <221 kW vessels; or ~140 min for the >221 kW
vessels), respectively. Except for two trips, fish were monitored, alongside 160 control plaice, for
an average of 15 days until no more mortality attributable to the catch-and-discarding process
was observed (Table 1a-c). Non-parametric Kaplan-Meier curves were neatly segregated for reflex,
reflex and injury indices, trawl duration, sorting duration, wave height, sea surface temperature,
and total catch suggesting that these variables are potentially associated with delayed survival
(Figure 6). For sediment type, if catches were scored as having >25% in volume stones or sand or
both present, survival probability for fish was lower.
12
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 13
Figure 6 Non-parametric Kaplan-Meier survival probability estimates based on monitoring over
time (between 9 and 20 days) (%) at a corresponding level of an potentially explanatory variable
such as reflex impairment index (RAMP.score, a), reflex impairment and injury index
(RAMPINJ.score, b), trawl duration (min, c), sorting duration (min, d), wave height (m, e), surface
seawater temperature (°C, f), sediment type (0 : < 25% stones and/or < 25% sand; 1 : > 25% sand; 2
: > 25% stones; 3 : > 25% sand and stones), and total catch (kg, g), Crosses indicate censored values.
13
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 14
Based on the univariate analyses of variance (type III ANOVA) models, there was a significant
relationship between delayed mortality and several status indicators (Table 1). The number of
absent body flex, righting and evasion reflexes were positively related in a linear relationship with
delayed mortality (Figure 7).
Table 1 Summary of univariate analyses of variance (type III ANOVA) models used to investigate the
effects of status indicator scores (i.e., damage class, reflex index, reflex and injury index) on delayed
mortality probability. ***p < 0.001; **p < 0.01; *p < 0.05. DF = degrees of freedom.
VARIABLE AND LEVEL
CHI-SQUARE
DF
Bellybend reflex
1.848
1
Righting
39.759***
1
Headcomplex
0.274
1
Evasion
62.381***
1
Stabilize
27.994***
1
Tailgrab
3.851*
1
No. absent reflexes
51.921***
1
No. absent reflexes (reduced)
67.361***
1
Headbruises
56.936***
1
Bodybruises
27.986***
1
Point bleeding – head
24.63***
1
Point bleeding – body
11.066***
1
Damage class
122.61***
3
Reflex index
60.357***
1
Reflex and injury index
123.42***
1
Based on non-parametric Kaplan-Meier models, it was estimated that at vessel level (each
participating vessel represented a different Belgian beam-trawl fleet segment) that survival of
plaice discarded from conventional trawls ranged between 43-57%, 10-26%, 3-5% (95% confidence
interval) for trips of the coastal (trips 1-5), small (trips 6-10) and large fleet segment (trips 11 and
12), respectively. These data are representative for the specific conditions during the sampled trips.
14
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 15
Figure 7 Validated relationship between the number of absent reflexes (body flex, righting and
evasion) observed among discarded plaice and their post-capture mortality rate.
2.2.3 Predicted and total mortality
Based on non-parametric Kaplan-Meier models, it was estimated previously (Uhlmann et al.
2016) that at vessel level (each participating vessel represented a different Belgian beam-trawl fleet
segment), survival of plaice discarded from conventional trawls ranged between 43-57%, 10-26%,
3-5% (95% confidence interval) for trips of the coastal (trips 1-5), small (trips 6-10) and large fleet
segment (trips 11 and 12), respectively (Figure 8). The majority of fisheries-related mortality was
observed within the first five days of monitoring. These data are representative for the specific
conditions during the sampled trips.
Based on the logistic prediction model and i) the observed immediate and delayed (monitored
for up to 9 days) mortality estimates from 10 Belgian beam-trawl trips1 and ii) the observed
immediate and predicted delayed mortality from three recent trips, it was estimated that for 50%
of the monitored trips2, mortality probability may have ranged between 24 and 65% or in other
words survival may have ranged between 35 and 76%. Hence, for 50% of the monitored trips, the
survival percentage falls within the range of 35-76%.
1Two trips were excluded from this analysis because the monitoring period was truncated after three days
and so fisheries-related mortalities were not observed until asymptote.
2This 50% is the interquartile range: i.e; the % of observations that lies between the 25 and 75 percentile.
Considering that three estimates of the >221 kW vessel trips are predictions which are based on input data
stemming in the majority of <221 kW vessels.
15
V e r s i o n : 2 D a t e : 0 3 - 0 4 - 2 0 1 8 I L V O | 16
Figure 8 Non-parametric Kaplan-Meier survival probability estimates based on monitoring over
time (between 9 and 33 days) averaged over all conventional trawls done during monitored trips
for each of the participating vessels (between 2014 and 2016).
Table 2* Non-parametric Kaplan-Meier survival estimates (mean ± SE) of discarded plaice at the
end of the monitoring period per trip and trawl duration (short/experimental vs conventional
gear deployment durations). NA = not available. For the three trips in 2017/2018, total mortality
was estimated based on the validated relationship between the number of absent reflexes and
delayed mortality observed from the other trips.
SURVIVAL PROBABILITY (%)
TRIP ID
MONTH
YEAR MONITORING- PERIOD (DAYS)
SHORT
CONVENTIONAL CONTROL
1
November
2014
20
55.0 ± 4.8
59.2 ± 3.5
NA
2
December
2014
22
80.0 ± 17.9
14.9 ± 5.2
100.0
3
February
2015
14
84.2 ± 8.4
92.5 ± 4.1
100.0
4
March
2015
3
80.0 ± 8.9
72.5 ± 7.1
100.0
5
March
2015
3
95.0 ± 4.9
54.8 ± 8.9
100.0
6
April
2015
17
60.0 ± 10.9
7.7 ± 4.3
100.0
7
June
2015
23
80.0 ± 8.9
50.0 ± 7.9
100.0
8
July
2015
14
0.0
3.5 ± 2.4
90.0 ± 2.7
9
July
2015
8
15.0 ± 8.0
3.9 ± 0.8
NA
10
August
2015
8
11.8 ± 8.0
4.2 ± 0.9
NA
11
September 2015
34
85.0 ± 7.9
35.0 ± 7.5
NA
12
September 2015
15
25.0 ± 9.6
17.9 ± 6.2
95.0 ± 5.0
Predicted survival based on the validated reflex impairment-survival relationship:
13
October
2017
NA
NA
58
NA
14
November
2017
NA
NA
35
NA
15
December
2018
NA
NA
66
NA
*Disclaimer: The data in this Table are still under revision and should be interpreted with care.
16
3
Appendix
Appendix Table 1* An overview of the fishery statistics for plaice (year 2017) and literature estimates for survival for plaice per ICES-area and geographical
region, the number of trips and fishing hours of the Belgian beam-trawl fishery (métiers ≤221kW and >221kW) are presented. For each area, quotum (in tonnes),
landings, discards and catch (x 1000, summed for all age classes) and the discard ratio (only for 70-99mm mesh sizes) are presented. Mean delayed survival (±
Standard Deviation, SD) and mean observation (captive) period (hours±SD) were obtained from literature. NA = not available.
AREA
REGION TRIPS
FISHING HOURS
QUOTUM
LANDINGS
DISCARDS
CATCH
DISCARD RATIO
MEAN DELAYED SURVIVAL (± SD)
MEAN OBSERVATION PERIOD (± SD)
4B
139
12964
0.28 ±0.21 (Kelle 1976)
186 (Kelle 1976)
0.51 ± 0.35 (Berghahn et al. 1992)
79.56 ± 11.46 (Berghahn et al. 1992)
North
4C
Sea
8409,16
6189,572
21679,85
27869,42
0.49
0.28 ± 0.27 (Van Beek et al. 1990)
NA (Van Beek et al. 1990)
678
10944
0.42 (Catchpole et al. 2015)
120 (Catchpole et al. 2015)
0.146 (Van der Reijden et al. 2017)
600 (Van der Reijden et al. 2017)
7A
Irish Sea
69
3023
90
NA
NA
NA
0.57
NA
NA
7D
647
48320
5584,556
8819,743
14404,299
0.35
2572,93
7E
151
7029
NA
NA
NA
0.14
Channel
0.44 ± 0.19 (Revill et al. 2013)
84.20 ± 27.28 (Revill et al. 2013)
7H
72
1979
NA
NA
NA
0.58 ± 0.19 (Catchpole et al. 2015)
182.33 ± 142 (Catchpole et al. 2015)
11,20
NA
7J
6
58
NA
NA
NA
7F
Celtic
204
14082
NA
NA
NA
Sea
190,01
0.54
NA
NA
7G
248
19266
NA
NA
NA
8A
Gulf of
15
1113
NA
NA
NA
Biscay
5
NA
NA
NA
8B
61
7191
NA
NA
NA
*Disclaimer: The mean delayed survival estimates that are listed in this Table come from studies that applied normal commercial catching and sorting practices (Berghahn
et al. 1992, Catchpole et al. 2015, Kelle 1976, Revill et al. 2013, Van Beek et al. 1990). As can be observed from these estimates in Table 1 (Appendix), the mean (delayed) discard
survival for plaice ranges between 0.28 and 0.58 for the North Sea and English Channel, with large standard deviations. The listed studies are characterized by a wide array
of different technical (e.g. gears, métiers) and environmental (e.g. area, temperature) parameters. Additionally, immediate mortality has not been taken into account. Hence,
an estimate of total mortality cannot be provided. The mean captive observation period ranges between 79,56 hours (±3 days) and 186 hours (±7 days). This enormous
variability makes generalizations extremely difficult. It is clear from Table 1 that data are still lacking for many ICES-areas and that more research is needed to feel these gaps.
Appendix Table 2a Summary of mean ± SD key technical, environmental, and biological variables collected during each monitored trip of a commercial
coastal beam trawler with < 221 kW engine power, 4-m beam- length and 1300 kg single gear weight (n observations). NA = not available.
TYPE
VARIABLE
TRIP 1
TRIP 2
TRIP 3
TRIP 7
TRIP 11
Month of year
November ´14
December ´14
February ´15
June ´15
September ´15
ICES sub-Division
4c
4c
4c
4c
4c
Total no. of deployments
16
15
15
15
15
Short deployments sampled
2
1
1
1
1
General
No. plaice sampled
30
10
19
20
20
Conventional deployments sampled
4
5
2
2
2
No. plaice sampled
50
47
40
40
40
% immediate mortality
0
0
0
0
0
Depth (m)
8.8 ± 2.7 (16)
16.1 ± 4.1 (15)
9.6 ± 3.6 (15)
9.7 ± 3.6 (15)
7.8 ± 4.2 (15)
Duration (min)
44.9 ±12.9 (16)
51.2 ± 9.1 (15)
52.3 ± 9.6 (15)
42.9 ± 14.4 (15)
48.4 ± 10.8 (15)
Hopper
4.8 ± 1.6
7.2 ± 2.5
5.9 ± 0.9
3.5 ± 0.9
4.4 ± 2.2
Technical
Begin of sorting
8.3 ± 2.3
15.2 ± 3.2
7.0 ± 3.0
7.7 ± 3.0
6.0 ± 1.8
Mid
16.5 ± 3.3
19.6 ± 0.2
10.0 ± 2.7
9.2 ± 2.7
7.7 ± 1.3
End
NA
NA
12.7 ± 2.4
10.7 ± 2.4
9.7 ± 1.3
Wind force (Bft)
2.1 ± 0.7 (16)
4.8 ± 1.4 (15)
2.6 ± 0.8 (15)
2.7 ± 0.8 (15)
3.9 ± 0.8 (15)
Wave height (cm)
62.4 ± 9.9 (9)
106.8 ± 45.9 (5)
50.6 ± 14.2 (12)
36.4 ± 6.01 (4)
117.4 ± 18.9 (3)
Environmental
Air temperature (°C)
10.9 ± 0.1 (6)
8 ± 0.4 (6)
8.1 ± 0.4 (3)
12.8 ± 0.2 (4)
17.0 ± 0.2 (3)
Seawater temperature (°C)
11.7 ± 0.1 (6)
7 ± 0.1 (6)
5.3 ± 0.1 (3)
14.4 ± 0.2 (4)
16.9 ± 0.1 (3)
Total catch (kg)
2012.3 ± 1854.8 (6)
NA
481.3 ± 440.9 (3)
333.5 ± 233.7 (4)
349.1 ± 105.1 (3)
Biological
TL of plaice (cm)
21.5 ± 3.7 (79)
19.5 ± 4.2 (57)
23.6 ± 3.2 (59)
20.6 ± 2.8 (60)
23.0 ± 3.1 (60)
TL of plaice (cm) – controls
NA
22.6 ± 4.0 (20)
23.2 ± 2.9 (20)
NA (20)
NA
Appendix Table 2b Summary of mean ± SD key technical, environmental, and biological variables collected during each monitored trip of a commercial
Eurocutter beam trawler with < 221 kW engine power, 4-m beam- length and 2000 kg single gear weight (n observations). NA = not available.
TYPE
VARIABLE
TRIP 4
TRIP 5
TRIP 6
TRIP 8
TRIP 12
For all plaice Month of year
March ’15
March ’15
April ’15
July ’15
September ’15
ICES sub-Division
7d
7d
7d
7d
7d
Total no. of deployments
52
55
47
43
45
Sampled and monitored for survival Short deployments sampled
1
1
1
1
1
No. plaice sampled
20
20
20
20
20
General
Conventional deployments sampled
2
2
2
3
2
No. plaice sampled
40
40
39
58
39
% immediate mortality
<1
<1
3
2
0
Sampled for reflexes Conventional deployments sampled
4
4
9
5
2
No. plaice sampled
79
80
206
88
34
% immediate mortality
0
0
0
0
0
Depth (m)
43.1 ± 12.4 (7)
36.3 ± 4.4 (7)
36.3 ± 15.9 (12)
35.8 ± 6.5 (9)
31.8 ± 7.6 (5)
Duration (min), conventional
75.7 ±11.8 (6)
72.6 ± 8.2 (6)
86.4 ± 9.6 (11)
69.2 ± 9.4 (8)
69.7 ± 10.5 (4)
Duration (min), short
25.0 (1)
26.0 (1)
25.0 (1)
36.0 (1)
24.0 (1)
Hopper
3.4± 1.5
1.4 ± 0.6
2.2 ± 0.9
2.9 ± 0.9
1.6 ± 1.0
Technical
Begin of sorting
5.9 ± 1.8
5.3 ± 1.8
4.6 ± 1.6
4.7 ± 0.7
2.9 ± 1.1
Mid
8.3 ± 1.8
6.7 ± 1.6
5.7 ± 1.5
6.3 ± 0.8
4.8 ± 1.2
End
10.3 ± 2.3
7.8 ± 1.9
7.2 ± 1.8
7.9 ± 1.1
6.5 ± 1.5
Sorting time (min)
18.6 ± 5.6 (7)
12.4 ± 2.6 (7)
10.8 ± 2.1 (12)
14.9 ± 5.1 (9)
7.4 ± 1.7 (5)
Wind force (Bft)
3.0 ± 0.8 (7)
3.4 ± 0.5 (7)
2.0 ± 1.9 (12)
3.6 ± 0.7 (9)
5.8 ± 0.8 (5)
Wave height (cm)
67.9 ± 31.3 (7)
57.1 ± 18.9 (7)
20.8 ± 25.7 (12)
37.8 ± 18.6 (9)
190.0 ± 41.8 (5
Wave height (cm)
67.9 ± 31.3 (7)
57.1 ± 18.9 (7)
20.8 ± 25.7 (12)
37.8 ± 18.6 (9)
190.0 ± 41.8 (5)
Air temperature (°C)
7.6 ± 1.7 (7)
8.6 ± 1.2 (7)
10.1 ± 1.7 (12)
20.3 ± 3.0 (9)
15.5 ± 1.5 (5)
Environmental
Seawater temperature (°C)
8.1 ± 0.2 (7)
8.1 ± 0.2 (7)
9.9 ± 0.2 (12)
16.2 ± 0.7 (9)
16.6 ± 0.1 (5)
Total catch (kg)
466.2 ± 270.2 (7) 515.1 ± 299.9 (7) 291.7 ± 296.9 (12) 726.7 ± 481.6 (9) 270.0 ± 163.1 (5)
TL of plaice (cm)
22.6 ± 3.1 (138)
22.4 ± 3.0 (170)
22.8 ± 2.8 (256)
22.7 ± 2.4 (196)
25.9 ± 2.1 (123)
TL of plaice (cm) – controls
22.6 ± 2.4 (20)
NA (20)
NA (20)
NA (20)
NA (20)
Appendix Table 2c Summary of mean ± SD key technical, environmental, and biological variables collected during each monitored trip of three
commercial beam trawlers (n observations). NA = not available
TYPE
VARIABLE
TRIP 9
TRIP 10
TRIP 13
TRIP 14
TRIP 15
Month
July ’15
August ’15
October ’17
November ’17 January ’18
ICES sub-Division
4b,c
4b,c
7e,h,g
7d
7d,e
For all plaice Beam length (m)
11.4
11.4
12
11
11.4
Single gear weight (kg)
5750
5750
5500
4500
5750
Total no. of deployments
88
61
42
36
54
General
Short deployments sampled
1
1
0
0
0
No. plaice sampled
20
17
NA
NA
NA
Sampled and monitored for survival Conventional deployments sampled 6
8
0
0
0
No. plaice sampled
114
126
NA
NA
NA
Conventional deployments sampled 12
8
11
29
11
Sampled for reflexes No. plaice sampled
951
694
154
794
366
Depth (m)
63.4 ± 15.1 (19)
61.0 ± 16.7 (17)
67.7 ± 7.1 (11)
42.4 ± 7 (35)
62.0 ± 4.4 (42)
Duration (min), conventional
132.3 ±17.5 (18)
141.3 ± 18.5 (16)
164.7 ± 5.8 (8) 165 ± 8 (29)
145.4 ± 19.9 (54)
Duration (min), short
31.0 (1)
24.0 (1)
n/a
n/a
n/a
Hopper
1.5 ± 1.5
2.6 ± 3.1
4.5 ± 2.2 (40) 3.1 ± 1.9 (139)
1.4 ± 0 (51)
Technical
Begin of sorting
5.5 ± 1.6
6.5 ± 2.7
11.3 ± 5.4 (39) 11.6 ± 4.7 (251) 7.3 ± 1 (51)
Mid
9.2 ± 2.3
10.3 ± 3.1
14.7 ± 5.4 (40) 25.1 ± 6.4 (322) 14.9 ± 6.5 (216)
End
14.0 ± 4.4
15.9 ± 5.6
19.8 ± 6.5 (35) 34.9 ± 3.5 (82 28.4 ± 5.9 (99)
Sorting time (min)
45.9 ± 10.9 (19)
33.5 ± 7.3 (17)
23.1 ± 5.5 (8)
36 ± 6.7 (29)
38.6 ± 9.9 (11)
Wind force (Bft)
2.4 ± 1.7(19)
2.6 ± 1.1 (17)
5.9 ± 0.9 (11)
4.1 ± 1.5 (35)
5.5 ± 1.5 (42)
Wave height (cm)
43.2 ± 39.0 (19)
53.5 ± 27.6 (17)
2.0 ± 0.6 (11)
0.3 ± 0 (6)
160.9 ±110.2 (42)
Environmental
Air temperature (°C)
17.8 ± 4.1 (19)
16.4 ± 2.2 (17)
13.4 ± 0.9 (11) 9.7 ± 1 (35)
7.2 ± 1.6 (42)
Seawater temperature (°C)
16.0 ± 0.3 (19)
16.2 ± 0.6 (17)
15.8 ± 0.2 (11) 10.6 ± 0.2 (35) 7.3 ± 0.4 (42)
Total catch (kg)
2600.0 ± 1032.8 (19) 1665.9 ± 487.6 (17) 643 ± 111 (11)
NA
1127 ± 119 (11)
Biological
TL of plaice (cm)
23.9 ± 2.3 (546)
23.6 ± 2.4 (511)
25.9 ± 1.9 (154) NA
23.9 ± 2.4 (366)
TL of plaice (cm) – controls
NA
NA
NA
NA
NA
4
Acknowledgements
The authors would like to thank Jochen Depestele for carefully reading and reviewing this report.
Additionally, we thank Martine Velghe (Departement Landbouw & Visserij) for her help in providing
recent data for plaice fisheries and Bart Vanelslander for the VMS maps and effort statistics.
5
References
Berghahn, R., Waltemath, M., & Rijnsdorp, A. D. (1992). Mortality of fish from the by‐catch of shrimp
vessels in the North Sea. Journal of Applied Ichthyology, 8(1‐4), 293-306. Harris, R. R., &
Bayse, S.M., Polet, H., 2015. Evaluation of a large mesh extension in a Belgian beam trawl to reduce
the capture of sole (Solea solea). ILVO-report. 12pp.
Borges, L. (2015). The evolution of a discard policy in Europe. Fish and fisheries, 16(3), 534-540.
Catchpole, T., Randall, P., Forster, R., Santos, A. R., Armstrong, F., Bendall, V., & Maxwell, D. (2015).
Estimating the discard survival rates of selected commercial fish species (plaice-Pleuronectes
platessa) in four English fisheries. Cefas report.
Departement Landbouw & Visserij (2017). De Belgische visserij 2016 – Aanvoer en besomming.
Departement Landbouw & Visserij, Belgium. Depot no. D/2016/3241/140.
Depestele, J., Desender, M., Benoît, H. P., Polet, H., and Vincx, M. 2014. Short-term survival of discarded
target fish and non-target invertebrate species in the "eurocutter" beam trawl fishery of the
southern North Sea. Fisheries Research, 154: 82-92.
Depestele, J. 2015, The fate of discards from marine fisheries. PhD thesis Ghent University. Gent,
Belgium. 286pp.
Kelle, W. (1976) Sterblichkeit untermafliger Plattfische im Beigng der Garnelenfischerei. Ber. dt. wiss.
Kommn. Meeresforsch, 25 (1/2): 77-89.
Kelleher, K. Discards in the world’s marine fisheries. An update. FAO Fisheries Technical Paper. No.
470. Rome, FAO. 2005. 131p.
ILVO (2016). Evolution of the Belgian fisheries. Authors: Heleen Lenoir, Hans Polet, Els Vanderperren,
Arne Kinds, Kim Sys.
ILVO (2017). Authors: Sebastian Uhlman, Bart Vanelslander, Els Torreele, Els Vandeperre, Hans Polet.
Advice document to inform any request under article 15.4(b) of Regulation (EU) 1380/2013 to
exempt from the Landing Obligation European plaice (Pleuronectes platessa) under MCRS
(<27cm in length) caught by beam-trawl gears (TBB).
Methling, C., Skov, P. V., & Madsen, N. (2017). Reflex impairment, physiological stress, and discard
mortality of European plaice Pleuronectes platessa in an otter trawl fishery. ICES Journal of
Marine Science, 74(6), 1660-1671.
Morfin, M., Kopp, D., Benoît, H. P., Méhault, S., Randall, P., Foster, R., & Catchpole, T. (2017). Survival
of European plaice discarded from coastal otter trawl fisheries in the English Channel. Journal
of environmental management, 204, 404-412.
Revill, A. S., Broadhurst, M. K., & Millar, R. B. (2013). Mortality of adult plaice, Pleuronectes platessa
and sole, Solea solea discarded from English Channel beam trawlers. Fisheries research, 147, 320-
326.
Uhlmann, S. S., Theunynck, R., Ampe, B., Desender, M., Soetaert, M., & Depestele, J. (2016). Injury, reflex
impairment, and survival of beam-trawled flatfish. ICES Journal of Marine Science, 73(4), 1244-
1254.
Van Beek, F. A., Van Leeuwen, P. I., & Rijnsdorp, A. D. (1990). On the survival of plaice and sole
discards in the otter-trawl and beam-trawl fisheries in the North Sea. Netherlands Journal of
Sea Research, 26(1), 151-160.
van der Reijden, K. J., Molenaar, P., Chen, C., Uhlmann, S. S., Goudswaard, P. C., & van Marlen, B.
(2017). Survival of undersized plaice (Pleuronectes platessa), sole (Solea solea), and dab (Limanda
limanda) in North Sea pulse-trawl fisheries. ICES Journal of Marine Science, 74(6), 1672-1680.
Van Marlen, B., Wiegerinck, J. A. M., van Os-Koomen, E., & Van Barneveld, E. (2014). Catch comparison
of flatfish pulse trawls and a tickler chain beam trawl. Fisheries Research, 151, 57-69.