Ref. Ares(2018)3458869 - 29/06/2018
Ref. Ares(2019)2387732 - 04/04/2019
Increasing the survival of discards in North
Sea pulse-trawl fisheries
Authors: Pieke Molenaar and Edward Schram
Wageningen University &
Research Report C038/18
Increasing the survival of discards in
North Sea pulse-trawl fisheries
Author(s):
Pieke Molenaar and Edward Schram
Publication date: 14 May 2018
Wageningen Marine Research
IJmuiden, May 2018
Wageningen Marine Research report C038/18
Pieke Molenaar and Edward Schram
,2018. Increasing the survival of discards in North Sea pulse trawl
fisheries. Wageningen, Wageningen Marine Research (University & Research centre), Wageningen
Marine Research report C038/18. 39 pp.
Client:
Visned
Attn.: Wouter van Broekhoven
Postbus 59
8320 AB Urk
The Netherlands
European Union, European Maritime and Fisheries Fund (EMFF)
This report can be downloaded for free from
https://doi.org/10.18174/449808
Wageningen Marine Research provides no printed copies of reports
Wageningen Marine Research is ISO 9001:2008 certified.
Photo cover: Edward Schram
© 2018 Wageningen Marine Research Wageningen UR
Wageningen Marine Research
The Management of Wageningen Marine Research is not responsible for resulting
institute of Stichting Wageningen damage, as well as for damage resulting from the application of results or
Research is registered in the Dutch research obtained by Wageningen Marine Research, its clients or any claims
traderecord nr. 09098104,
related to the application of information found within its research. This report
BTW nr. NL 806511618
has been made on the request of the client and is wholly the client's property.
This report may not be reproduced and/or published partially or in its entirety
without the express written consent of the client.
A_4_3_2 V27
2 of 39 | Wageningen Marine Research report C038/18
Contents
Preface
4
Summary
5
1
Introduction
7
2
Materials and Methods
8
2.1 Experimental design
8
2.1.1 Ethics statement
8
2.1.2 Outline of the experiments
8
2.1.3 Sea trips
8
2.1.4 Treatment 1: Water fil ed hopper
9
2.1.5 Treatment 2: Short hauls
11
2.1.6 Treatment 3: Knotless cod-end
11
2.1.7 Control-fish
12
2.2 Assessment of fish condition and monitoring of survival
13
2.3 Experimental facilities
13
2.4 Data analysis
14
3
Results
16
3.1 Survival of control-fish
16
3.2 The effect of a water fil ed hopper on discards survival
16
3.2.1 Main effect of a water fil ed hopper
16
3.2.2 Effects of sea trips and vessels on the effect of the water fil ed hopper
17
3.2.3 Effect of a water fil ed hopper on fish condition
18
3.2.4 Interactive effects of fish condition and a water fil ed hopper on discards
survival probability
18
3.3 The effect of short hauls on discards survival probability
19
3.3.1 Main effects of short hauls
19
3.3.2 Effect of short hauls on fish condition
20
3.3.3 Interactive effects of fish condition and short hauls on discards survival
probability
20
3.4 Effect of a knotless cod-end on discards survival probability
21
4
Discussion
22
4.1 General
22
4.2 Water fil ed hopper
22
4.3 Short hauls and knotless cod-end
24
5
Conclusions and recommendations
25
6
Acknowledgements
26
7
Quality Assurance
27
References
28
Justification
29
Annex 1: Survival per trip
30
Wageningen Marine Research report C03/18| 3 of 39
Preface
The project ‘Survival of flatfish and ray discards’ investigates four topics related to flatfish and ray
discards survival in the 80 mm pulse-trawl fisheries in the North Sea: 1. Discards survival of
undersized plaice, sole, turbot, bril , thornback ray and spotted ray in conventional pulse-trawl
fisheries, 2. Measures to increase discards survival, 3. Factors affecting discards survival and 4. The
use of vitality index scores as a proxy for discards survival.
Each topic wil be reported separately and the current report is the second in the series of four reports
delivered by the project.
Al research data for this project were collected during nine sea trips with three commercial pulse-
trawlers. Utilization of methods and research data partly overlaps among the four topics. In addition,
each report can be read independently from the other reports in the series. Consequently the
description of methods and reporting of data partly overlaps in the four reports.
In a later stage, parts of the results presented in these four reports will be submitted for publication in
peer-reviewed scientific journals. These four reports should be considered as pre-publications of final
results.
The project was commissioned by VISNED and received financial support from the European Maritime
and Fisheries Fund (EMFF) of the European Union.
May, 2018.
4 of 39 | Wageningen Marine Research report C038/18
Summary
Measures to increase discard survival in the 80 mm pulse-trawl fisheries were assessed under
commercial fishing conditions using plaice (
Pleuronectes platessa) as model species. Measures tested
were a water fil ed hopper (8 sea trips), short hauls (90 instead of 120 min, 4 sea trips) and a
knotless cod-end (1 sea trip) with undersized plaice. In total nine sea trips were performed with three
different commercial pulse-trawlers. Sea trips were spread over the year to account for potential
seasonal variation in discards survival. Additional trials with a knotless cod-end (one sea trip) and
water fil ed hopper (2 sea trips) were performed with undersized sole. Effects were assessed by
comparing survival to conventional conditions (dry hopper, conventional haul duration, conventional
cod-end) for which data were col ected from the same or subsequent hauls.
Al test-fish were randomly col ected from the end of the sorting belt at both the start and end of the
catch-sorting process from multiple hauls per sea trip. Reflex impairment and damages were assessed
for al test-fish and summarized in a vitality index score. Test-fish were housed on-board in custom-
built monitoring units containing 16 (24L) tanks with five fish each tank. Tank water was continuously
renewed with sea water at a rate of at least two tank volumes per hour to maintain proper water
quality. Survival was monitored and dead fish were removed upon detection. Upon arrival in the
vessel’s home port, monitoring units were road transported to the laboratory to continue survival
monitoring for two more weeks. Total monitoring period ranged from 15 to 18 days among test-fish
depending on the day of col ection at sea. In the laboratory tank bottoms were covered with coarse
sand and fish were fed natural food. In total 558 plaice from conventional fisheries (ca. 60 per sea
trip) were col ected, 478 plaice for the water fil ed hopper treatment (ca. 60 per sea trip), 200 plaice
from short hauls (ca. 40 and 60 each in two sea trips) and 60 plaice from the knotless cod-end.
Control-fish of the same species, in good condition and not exposed to the fishing gear were deployed
during al sea trips (circa 30 control plaice and circa 15 control sole per sea trip). Control-fish were
handled and tagged as test-fish to separate fisheries related mortality from mortality caused by the
experimental procedures.
Discards survival probabilities and their 95% confidence intervals (CI) were estimated from counts of
surviving fish at the end of the monitoring period. For al sea trips combined, no significant effect of a
water fil ed hopper on plaice discards survival probability could be detected with 16% (95%CI 12-
19%) for the conventional dry hopper and 20% (95%CI 15-25%) for the water fil ed hopper. Within
the individual sea trips, a significantly higher survival probability for plaice discards from the water
fil ed hopper was found for three sea trips. In three other sea trips a lower survival of plaice discards
was detected for the water fil ed hopper, although the difference with the dry hopper was not
significant. Given this observation, it cannot be entirely excluded that the water fil ed hopper can also
have a negative effect on discards survival. For sole discards the effect of a water fil ed hopper was
tested during two sea trips only, yielding a higher survival probability of sole discards for the water
fil ed hopper (14%, 95%CI 10-21%) compared to the dry hopper (5%, 95%CI 2-10%).
Deployment of a water fil ed hopper results in a shift towards a better condition of the discarded fish.
Despite this effect, the total proportion of fish in good condition within catches remained smal . We
therefore recommend to prioritize measures aimed at improving fish condition in the trawl to increase
discards survival chances.
For al sea trips combined, no effect of short (90 instead of 120 min) hauls on discards survival
probability could be detected: survival probabilities for plaice discards were equal at 11% (95% CI 8-
15%) for both short and conventional hauls. No effect of a knotless cod-end on plaice and sole
discards survival probability could be detected.
In conclusion, deployment of a water fil ed hopper does not result in higher survival probability for
plaice discards than a conventional dry hopper in year-round pulse-trawl fisheries. However, it is clear
Wageningen Marine Research report C03/18| 5 of 39
that for individual trips the deployment of a water fil ed hopper can result in an increase of survival
chances of discarded plaice, but as it seems only under certain specific, yet to be established,
conditions. In addition, it cannot be excluded at this point that under certain conditions a water fil ed
hopper may have a negative effect on discards survival. For sole a positive effect of the water fil ed
hopper on discards survival was detected. However, since sole was tested during two sea trips only,
the current findings may not be representative for year-round fisheries and the positive effect may be
specific for the conditions that prevailed during the two trips. Survival probability of plaice and sole
discards cannot be increased by reducing haul duration from 120 to 90 min or using a knotless cod-
end.
6 of 39 | Wageningen Marine Research report C038/18
1
Introduction
Demersal pulse-trawl fisheries in the North Sea is a mixed fishery that mainly targets Dover sole
(
Solea solea) and plaice (
Pleuronectes platessa). Undersized and over quota fishes and species with no
market value are discarded. By 2019 this practise of discarding wil be restricted for al quota
regulated species by the implementation of a landing obligation under the Common Fisheries Policy
(European Union, 2013). As a result of this legislation fishermen wil be forced to land al undersized,
damaged and marketable fish of species under quota management, also referred to as a landing
obligation (LO). However, this landing obligation al ows exemptions for species which according to the
best available scientific advice have a high survival rate when released into the sea, taking into
account gear characteristics, fishing practices and the ecosystem.
Previous work on the survival of discards from pulse-trawl fisheries resulted in survival probability
estimates of 15% (95%CI: 11-19%) for plaice and 29% (95%CI:24-35%) for sole (Van der Reijden et
al., 2017). More recently, we reported a very similar discards survival probability estimate of 14%
(95%CI 11-18%) for plaice (Schram and Molenaar, 2018). For undersized sole we reported a slightly
lower discards survival than Van der Reijden et al. (2017) of 19% (95%CI 13-28%).
Several measures aimed at increasing the post-capture survival of fish when released into the sea
have been explored with promising results. Reducing haul duration from 100-130 min to 60-70 min
was found to promote the survival of plaice discards but not sole discards (Van der Reijden et al.,
2017). Haul duration reduction is preferably limited as for its significant operational impact on-board
fishing vessels and effect on total fishing time per trip. Preliminary work on the effect of a water fil ed
hopper instead of the common practise of discharging catches from the cod-end into a dry hopper
suggested an increase of the survival of plaice discards (Van Marlen et al., 2016), although data were
insufficient to draw final conclusions.
This study therefore assessed the effect of a reduction of haul duration to 90 min and a water fil ed
hopper on survival of discards in the 80 mm pulse-trawl fisheries. In addition, the effect of a knotless
cod-end was tested. These measures (treatments) were implemented at three commercial pulse-
trawlers during nine sea trips. Their effects were assessed by comparing survival of undersized plaice
and sole discards col ected from modified and conventional fisheries.
Wageningen Marine Research report C03/18| 7 of 39
link to page 9 link to page 9 link to page 9 link to page 10 link to page 13
2
Materials and Methods
2.1
Experimental design
2.1.1
Ethics statement
The treatment of the fish was in accordance with the Dutch animal experimentation act, as approved
by ethical committees (Experiment 2017 D0012.002)
2.1.2
Outline of the experiments
Measures aimed at increasing discards survival were assessed by comparing discards survival between
modified (one of the measures implemented) and conventional fisheries and catch processing
practices. Fish were col ected during nine sea trips with three commercial pulse-trawlers and three
trips per pulse-trawler. Within each sea trip, fish were col ected from multiple hauls to account
potential for variation in discards survival among hauls. The typical number of hauls was 40 to 50 per
sea trip. Survival monitoring started during the sea trip and was continued on land for 14 days after
the fish had been transferred to the laboratory. Total survival monitoring time ranged from 15 to 18
days after col ecting test-fish at sea. Measures tested for their effects on discards survival included the
treatments; short hauls (S), a water fil ed hopper (W) and a knotless cod end (K). An overview of
treatments per sea trip and the number of test-fish col ected is provided in
Table 1.
Table 1
Overview of sea trips and fish sampling: total number of test-fish col ected and control-fish
deployed per species and sea trip and treatment and the survival of the control-fish
Trip Vessel Treatments
# Plaice collected per # Sole collected per
Survival of control-fish
tested (R, C,
treatment
treatment
W, S, K)*
R
C
W
S
K
R
C
W
K
Plaice
Sole
1
1
R, C, W, S
35 60 60
40
-
-
-
-
-
100%
-
2
2
R, C, W, S
30 60 59
60
-
-
-
-
-
97%
-
3
3
R, C, W, S
30 60 60
60
-
-
-
-
-
100%
-
4
3
R, C, W, S
30 59 59
40
-
-
-
-
-
90%
-
5
1
R, C, K
33 80
-
-
60 10 33
-
31
30%
90%
6
3
R, C, W
30 60 60
-
-
15 30 30
-
100%
100%
7
2
R, C, W
30 60 60
-
-
15 30 30
-
72%
100%
8
1
R, C, W
30 58 60
-
-
-
-
-
-
72%
-
9
2
R, C, W
29 59 60
-
-
-
-
-
-
93%
-
Total/ overall
277 576 478 200 60 40 93 60
31
84%
97%
* R= controls, C=Conventional, W=Water fil ed hopper, S= Short 90 minute haul, K= knotless cod-end.
2.1.3
Sea trips
Al nine sea trips were conducted in the Southern North Sea according to the regular commercial
practices of the pulse-trawlers. Sea trips typical y started on Mondays around 0:00 and ended on
Fridays around 4:00. Sea trips were spread out over the year
(Table 2) to account for the potential
effect of varying fishing conditions throughout the year on discards survival (Van der Reijden et al.,
2017). For each haul during a sea trip the operational and environmental conditions were recorded by
the skipper. Locations and conditions during the sea trips are presented in
Table 2 an
d Figure 1 .
Vessel and gear specifics are presented in
Table 3.
8 of 39 | Wageningen Marine Research report C038/18
link to page 9 link to page 9
Table 2 Conditions during the sea trips
Trip
Vessel Year Month Week
Temperature
Wind
Wave
Catch
Haul
Fishing
(°C)
speed height processing duration depth
Air
Water
(Bft)
(m)
(min)
(min)
(m)
1
1
2017
May
18
-
9-12
2-5
0.5-2.0
25
85-135
26-28
2
2
2017
May
21
14-18
12-13
1-4
0.2-0.5
24
90-120
39-50
3
3
2017
June
24
15-20
14-15
1-2
0.1-0.5
18
90-125
22-25
4
3
2017
July
28
15-21
16-17
1-6
0.2-1.0
18
90-120
28-40
5
1
2017
Sept
36
15-18
18
4-5
0.5-1.0
21
120-130 26-37
6
3
2017
Oct
44
13-15
13-15
3-4
1.0-1.2
20
120-130 30-31
7
2
2017
Dec
49
6-9
11-12
3-5
1.0-2.0
33
120
37-50
8
1
2018
Jan
4
8
6-7
5-6
1.0-1.7
33
120
28-35
9
2
2018
Feb
8
6
7-8
3-4
0.5-1.0
24
110-120 40-45
Figure 1. Fishing locations trials per sea trip
2.1.4
Treatment 1: Water filled hopper
The effect of a water fil ed hopper on discards survival was tested on model species plaice during eight
sea trips (al except trip 5,
Table 1) to cover fishing conditions as they prevail year-round. Test with
sole were conducted during two sea trips(trips 6 and 7,
Table 1) to gain some insight in the effect of
the water fil ed hopper on other species than the model species.
Several modifications were applied to one of the two hoppers on each participating vessel to be able to
operate it as a water fil ed hopper. The 1.5 to 2 m tal water canon/hose that is used to flush catch
from the hopper to the conveyer belt was replaced by one or multiple water inlets located 10-20 cm
Wageningen Marine Research report C03/18| 9 of 39
link to page 12
above the hopper bottom to reduce mechanical water pressure on the catch during processing.
Additional to the water hose modifications, three step cascade flow control devices were applied in the
hopper’s opening to the conveyor belt to maintain water levels in the hopper and enable stepwise
catch processing. Before hauling the cod-end on-board, the modified hopper was filled with ample
seawater to maintain a water level of 40 to 50 cm. As a result the catch was discharged in water
instead of on the metal surface of the hopper bottom. After discharging the catch, the hopper was
continuously supplied with seawater to maintain water and oxygen levels in the modified hoppers.
Stepwise catch processing was applied for the modified hoppers to prevent catch piling up on the
conveyor belt and maximise time spent in the water fil ed hopper. In addition, the stepwise processing
results in most of the fish being processed while the majority of benthic animals and debris remain in
the water fil ed hopper and are processed last only after removing the third cascade.
The water fil ed hopper treatment (W) was instal ed at one side of each trawler. The other hopper was
not fil ed with water according to conventional practices and served as control treatment (C) (not to be
confused with control-fish, see 2.2.7). For each haul the starboard and port side catches were kept
and processed separately and thus appeared as two separate batches on the sorting belt to al ow for
test-fish col ection for both treatments. Al test-fish were randomly col ected from the end of the
sorting belt
(Figure 1) from six (plaice) or two (sole) hauls per sea trip to account for potential
variation in fishing conditions and discards survival among hauls. To obtain representative samples
from hauls, the potential effects of processing time on discards survival (Benoit et al., 2013) were
accounted for by col ecting test-fish in equal numbers at both the start and the end of the catch-
sorting process of each haul. The processing sequence of the two hoppers was alternated between
hauls to obtain an equal average catch-processing time across the col ected test-fish for both
treatments. For each sampled haul, the time the catches were discharged in the hoppers as well as
the time of col ection of individual fish were recorded to determine catch-processing time per
individual fish.
Plaice were col ected during eight sea trips from six hauls per trip. Each haul ten fish per treatment
were col ected, five at the start and five at the end of the catch-sorting process for each hopper. This
resulted in a total of 60 plaice per treatment for each sea trip and 478 plaice per treatment for the
entire experiment.
Sole were col ected during two sea trips from two hauls per trip. Each haul 15 fish per treatment were
col ected, resulting in a total of 30 sole per treatment for each sea trip and 60 sole per treatment for
the entire experiment.
10 of 39 | Wageningen Marine Research report C038/18
link to page 9 link to page 9 link to page 12 link to page 9 link to page 9 link to page 9 link to page 12
Figure 1. Schematic drawing of semi-automatic catch processing line on board of a pulse trawler. Al
fish col ected from the catch for the survival experiment are col ected at the location marked with
‘sample location’.
2.1.5
Treatment 2: Short hauls
The effect of short hauls (S, circa 90 min.) on discards survival was tested on plaice during the first
four sea trips
(Table 1, page
8). Test-fish were col ected at the start and the end of the catch-sorting
process at the end of the sorting belt
(Figure 1). Plaice from hauls of conventional duration (circa 120
min) were collected the haul before or after the short haul during the same sea trip served as
conventional reference treatment. These controls for short hauls are the same test-fish (C) as
col ected as conventional ‘controls’ for the water fil ed hopper treatment (see 2.2.4,
Table 1). During
trips 1 and 4, 40 undersized plaice were sampled from two short hauls (20 per haul). During trips 2
and 3, 60 undersized plaice were sampled from three short hauls (20 per haul), resulting in a total of
200 test-fish for the short haul treatment in the entire experiment.
2.1.6
Treatment 3: Knotless cod-end
The effect of a knotless cod-end on discards survival was tested on plaice and sole during one sea trip.
Based on this single test it seemed clear that the knotless cod-end was not a breakthrough measure
with a large positive effect on discards survival. Testing of the knotless cod-end was consequently not
continued after one sea trip. The knotless cod-end (K) was connected to the starboard trawl and
tested during sea trip 5
(Table 1, page
8). The conventional cod-end at the port side served as
conventional ‘control’ treatment (C). The knotless cod-end was deployed for in total six hauls. During
knotless cod-end trials the hoppers were used according to regular practice; not filled with water. For
each haul the starboard and port side catches were kept, processed and sampled separately from the
end of the sorting belt
(Figure 1). The order in which the catches per treatment were processed and
sampled was alternated between each sampled haul to obtain an equal number of ‘first processed’
catches for both treatments. From each haul 10 undersized plaice per treatment (C and K) were
randomly col ected, five at the start and five at the end of the catch-sorting process, resulting in a
total of 60 plaice per treatment. Sole were only sampled from two hauls with 14 and 16 fish col ected
per treatment and haul.
Wageningen Marine Research report C03/18| 11 of 39
link to page 9 link to page 9 link to page 9
Table 3 Vessel and gear specifics
Specifics
Vessel 1
Vessel 2
Vessel 3
Engine power
1471
1430
1470
(Kw)
Gear
Sumwing pulse
Sumwing pulse
Sumwing pulse
Number of gears 2
2
2
Fishing speed
4.8
4.8
4.9
(kn)
Beam (wing)
Width (m)
12
12
12
Length (m)
1.1
1.1
1.1
Total weight (kg) 2600
2740
2300
False ground rope
Type
Rubber discs
Rubber discs
Rubber discs
Length (m)
11.7
11
11.8
Diameter (mm)
220
120
120
Total weight (kg) 110
140
80
Electrodes
Number
22
24
26
Type
HFK
HFK
HFK
Total length (m)
7.5
7.2
7.4
Distance between 40.0
42.5
45.0
electrodes (cm)
Length electrodes 3.0
3.2
4.4
on seabed (pulse
field) (m)
Conductor elements
Number
11
10
12
Diameter (mm)
35
28
33
Length (mm)
130
130
134
Distance between 220
210
200
elements (mm)
Pulse
Power (kW/m)
6.0
5.3
7.3
Width (µs)
340
390
330
Frequency (Hz)
60
45
60
Peak voltage over 60
60
60
electrode (V)
Maximum
1.2
1.3
1.7
exposure to pulse
field (s)
Trawl
Total length (m)
34
30
34
Mesh size cod-end 80
80
80
(mm)
Twine cod-end
Double knotted
Double knotted
Double knotted
Twine thickness
4
3
3
(mm)
2.1.7
Control-fish
In each of the nine sea trips, control plaice and sole (R in
Table 1, page
8) were deployed to separate
potential effects of the experimental procedures on mortality from fisheries induced mortality. During
each of the nine sea trips, control-fish were transported from the research facilities to the vessel and
taken on-board of the pulse-trawler. At the vessel control-fish were stored on deck in aerated 600L
tanks with regularly renewed surface seawater. Only fish in visual y observed good condition, wel fed
and without visible injuries, were selected for use as control-fish. Control-fish were exposed to the
exact same experimental procedures as the test-fish, including vitality assessment, tagging and
housing in the monitoring units throughout the experiments. The number of control-fish deployed was
approximately 30% of the number of test-fish per species
(Table 1)
12 of 39 | Wageningen Marine Research report C038/18
link to page 16 link to page 16 link to page 16 link to page 16 link to page 16
Control-fish were obtained by commercial shrimp and pulse beam trawlers (<221kW) which had been
requested to col ect least damaged and undersized fish from short hauls. Control-fish were also
col ected during the sea trips with the pulse-trawlers for use in subsequent sea trips. In both cases
col ected fish were stored on-board in 600L containers fil ed with surface seawater which was aerated
and regularly exchanged to maintain proper water quality. Prior to their use as control-fish were
stored in tanks placed in a climate control ed room for at least three weeks. During this period,
fisheries induced mortality levelled out while surviving fish could recover from injuries and regain good
condition. Tanks with candidate control-fish were inspected daily for mortalities which were removed
upon detection. During storage, fish were fed daily with live polychaete worms (
Nereis spp) and dead,
uncooked brown shrimps (
Crangon crangon) to visual y observed satiation.
2.2
Assessment of fish condition and monitoring of
survival
After col ection from the sorting belt, test-fish were temporarily stored in 105L holding containers fil ed
with seawater. The seawater in the holding containers was regularly renewed to maintain sufficient
dissolved oxygen levels during storage. Upon completion of fish col ection, fish were sequential y taken
from the holding containers to measure total length (TL: in cm below) and for vitality assessment and
tagging. Fish were taken randomly from the holding containers in case more than the required number
of fish had been col ected. Vitality status of each individual fish was assessed by scoring vitality index
class, external damage and reflex impairment as described by Van der Reijden et al. (2017) and
summarized in
Table 4. For thornback and spotted ray the protocols for external damage and reflex
impairment scores in flatfish by Van der Reijden et al. (2017) were adapted
(Table 4, page
15). Individual fish were tagged with Trovan Unique glass transponders (type ID100) to al ow for
identification of individuals throughout the experiments. Transponders were injected subcutaneously
just behind the head using the IID100E injector. Upon completion of the vitality assessment and
tagging, live fish were placed in 24 L tanks (see
Experimental facilities) with a maximum of five fish
per tank. Fish that were dead (defined as the absence of Head-complex,
Table 4) at the moment of
vitality assessment were recorded as dead at time zero. Dead fish were stored on ice and not replaced
by live individuals.
Monitoring of survival and experimental conditions started after the first fish had been placed in the
monitoring units. Al tanks containing fish were inspected every 12 hours on-board and every 24 hours
after transfer to the laboratory. Tanks were inspected for mortalities through or by lifting the
transparent lid of the tanks by visual observation of fish movement. In case any mortalities were
suspected to be present, these individuals were gently touched with a blunt plastic probe to provoke a
behavioural response. Fish that showed no response were manual y removed from the tank and dead
was confirmed by visual observation of a 15 seconds absence of gil plate/spiracle movement in water
and the ‘head complex’ reflex
(Table 4). Lethargic fish were not removed. Dissolved oxygen
concentration and saturation and water temperature were measured (Hach Lange Multimeter). Water
flows to the tanks were increased if oxygen saturation was below 80%.
2.3
Experimental facilities
Al test-fish col ected during sea trips and control-fish were housed in four custom-built monitoring
units instal ed on-board of the vessels. Each unit consisted of a stainless steel framework which holds
16 24L tanks (60 cm L x 40 cm W x 12 cm H), resulting in a total capacity of 64 tanks on a vessel.
Each tank was equipped with an individual water supply. A central pump instal ed on the vessel
continuously supplied surface seawater to the tanks. The water intake of this pump was approximately
2 meters below sea surface. Water flow rates to the tanks were instal ed at approximately two tank
volumes per hour (1-1.5L-1 min) to maintain proper water quality. Tanks were covered with
transparent lids to limit water losses by sloshing while al owing for visual inspection of the fish. Upon
return of the vessels in their home ports, the entire units were off-loaded and transported to the
Wageningen Marine Research report C03/18| 13 of 39
laboratory by road in a temperature control ed truck. Transport time ranged from one to three hours
depending on the home port of the vessel. During transport each unit was placed inside a pumping
tank partly fil ed with seawater and equipped with a submerged pump to supply water to each fish
tank in the unit. Fish tanks discharged their effluents in the pumping tank, al owing for recirculation
and aeration of the water. Upon arrival at the laboratory the fish tanks were manual y stacked in
racks. Al tanks were connected to a single water recirculation system consisting of a 440 L pumping
tank and a 330 L trickling filter. Total system volume was approximately 3.2 m3 and continuously
renewed with filtered water from the Eastern Scheldt at a rate of 8.6 m3/d. Al tanks were placed in a
temperature control ed room with its temperature set at the actual North Sea surface water
temperature at the time of test-fish col ection. In the laboratory, al tanks were supplied with coarse
sand as bottom substrate and the fish were fed daily to visual y observed satiation with polychaete
worms (
Nereis spp) and uncooked brown shrimps (
Crangon crangon). On-board, bottom substrate
was not applied as in combination with the inevitable rocking of the vessels, sand would probably
result in injuries through abrasion of the fish. Fish were not fed on-board as in our experience from
previous discards survival studies they do not restart feeding until several days after catching while
uneaten feed in the tanks would compromise water quality.
2.4
Data analysis
Survival, fish condition and sampling related time data were al collected at the level of the individual
fish. Fish were either dead or alive at the end of the survival monitoring period.
For each fish that died during the course of survival monitoring, the survival time was recorded as the
time (h) since col ection from a catches. Survival curves presenting the development over time of
survival within a group, were estimated using the non-parametric Kaplan-Meier estimator (Kaplan and
Meier, 2012).
The effect of a water fil ed hopper on discards survival was tested by comparing estimates for discards
survival probabilities for the water fil ed and dry hopper for al sea trips combined and the individual
sea trip separately. Survival probabilities per treatment were estimated and tested for significant
differences by multilevel linear logistic regression with sea trips, hauls and individual fish as
subsequent levels. To account for imbalances in the number of observations per sea trip and give sea
trips equal weight in the analysis, the contribution of each individual fish was weighed according to the
number of test-fish col ected per sea trip.
The effect of short hauls and the knotless cod end on discards survival was tested as described for the
effect of the water fil ed hopper, considering only those sea trips during which the measures were
tested.
Interactive effects of a water fil ed hopper and fish condition or vessel on discards survival of plaice
were tested for al sea trips combined by comparing the estimates for the discards survival
probabilities for the groups formed by
treatment*vitality index score and
treatment*vessel. Survival
probabilities per treatment combination were estimated and tested for significant differences by
multilevel linear logistic regression with sea trips, hauls and individual fish as subsequent levels.
Effects of the water fil ed hopper and the short hauls on fish condition were tested by comparing the
probability of assigning a test-fish to one of the four vitality index scores among treatment levels.
Probabilities of assigning test-fish to one of the four vitality index scores were estimated and tested
for differences among the hopper treatments and haul durations by multilevel linear logistic regression
with sea trips, hauls and individual fish as subsequent levels.
Al estimates and 95% confidence intervals for the odds ratios were back transformed in survival
percentages per treatment or group. A least significance difference (LSD) post-hoc analysis was used
to estimate the level of significance between treatments or groups in case a significant effect was
detected. Al tests were performed at the level of species, plaice or sole. In al cases the fiducial limit
was set at 5%.
14 of 39 | Wageningen Marine Research report C038/18
Table 4 Description of criteria to score vitality status.
Vitality index
Class
Description
A
Fish lively, no visible signs of loss of scale or mucus layer.
B
Fish less lively, minor lesions and some scales missing, mucus layer
affected up to 20% of skin surface area, some point haemorrhaging on the
blind side.
C
Fish lethargic, intermediate lesions and some patches without scales,
mucus layer affected up to 50% of skin surface area, several point
haemorrhaging on the blind side.
D
Fish lethargic or dead, clear head haemorrhaging, major lesions and
patches without scales, mucus layer affected for more than 50% of the
skin surface area, significant point haemorrhaging on the blind side.
External damage scores
Damage
Description (1 = present; 0 = absent)
Fin or wings
Fins are damaged or split (including tail fin). Wings in case of rays.
>50%
Damage to skin surface, scale or mucus layer at more than 50% of the
dorsal body surface.
Head haemorrhages
Presence of a haemorrhage in the head of the fish
Hypodermic haemorrhages Presence of a hypodermic haemorrhage
Intestines
Intestines are protruding or are visible through damaged body tissue of
the fish.
Wound
Presence of a wound such that flesh is visible.
Reflex impairment scores Reflex
Description (1 = impaired; no (clear) response within 5 s of observation; 0
= unimpaired; obvious response within 5 s).
Body flex
Fish is held on the palm of the hand with its ventral side up in the air. Fish
actively tries to move head and tail towards each other or wriggle out of
the hand.
Righting
Fish is held on the fingers of two hands with the dorsal side touching the
water surface. When released the fish actively rights itself under water.
Evasion
Fish is held underwater in an upright position by supporting its ventral side
with the fingers and its dorsal side with the thumbs. When the thumbs are
lifted the fish actively swims away.
Stabilize
Untouched fish tries to find a stable position flat on the bottom by
rhythmic and swift movement of the fins and/or body.
Tail grab
Fish is gently held by the tailfin between the thumb and index finger. Fish
actively struggles free and swims away.
Head complex
Fish moves its operculum or mouth during 5 s of observation while laying
undisturbed under water.
Wageningen Marine Research report C03/18| 15 of 39
link to page 9 link to page 9 link to page 18 link to page 17 link to page 18 link to page 17
3
Results
3.1
Survival of control-fish
The survival of the control-fish for plaice and sole is presented in
Table 1 (page
8). The survival of
control-fish for sole was ≥90% for the three sea trips in which test with sole were done. The survival
of control-fish for plaice is ≥93% in al sea trips except trips 5, 7 and 8.
3.2
The effect of a water filled hopper on discards survival
3.2.1
Main effect of a water filled hopper
The survival curves based on eight sea trips for plaice discards col ected from the dry and the water
fil ed hopper and control-fish are presented in
Figure 2, survival curves per sea trip are presented in
Annex 1. Except for the immediate mortality of test-fish at col ection (t=0), the development over
time of survival is very similar for both treatments.
For al sea trips combined, no significant effect a water fil ed hopper with water on plaice discards
survival probability could be detected (multilevel linear logistic regression, p = 0.14,
Table 5). The overal survival curves based on two sea trips for sole discards col ected from the dry and the
water fil ed hopper and control-fish are presented in
Figure 3. Similar to our observations for plaice,
the development over time of survival is very similar for both treatments apart from the initial
mortality of test-fish at col ection (t=0). A significant effect a water fil ed hopper with water on sole
discards survival probability was detected (multilevel linear logistic regression p <0.0001
, Table 5).
This effect seems mainly present in sea trip 7 as in sea trip 6 most fish died for both treatments
(Annex 1).
Table 5 The effect of a water fil ed hopper on the estimates and 95% confidence intervals (CI) for
plaice discards survival probability for al sea trips combined and for the individual sea trips. Survival
probability estimates differed between the dry and water fil ed hopper in sea trips 2, 4 and 9 (p<
0.05).
Species Sea trip
Dry hopper
Water filled hopper
p-value
Estimate 95% CI LL 95% CI UL Estimate 95% CI LL 95% CI UL
Sole
6 and 7
5%
2%
10%
14%
10%
21%
<0.0001
Plaice
All*
16%
12%
19%
20%
15%
25%
0.14
1
15%
9%
24%
18%
11%
30%
0.69
2
15%
11%
21%
29%
25%
33%
0.0009
3
12%
6%
22%
15%
4%
44%
0.77
4
3%
1%
12%
10%
6%
17%
0.03
6
22%
11%
38%
18%
7%
41%
0.74
7
20%
12%
32%
10%
3%
26%
0.17
8
17%
7%
38%
12%
4%
32%
0.26
9
20%
12%
33%
45%
24%
68%
0.01
*)All sea trips except no. 5.
16 of 39 | Wageningen Marine Research report C038/18
link to page 17
Figure 2. Overal survival of plaice discards over time (n = 8 sea trips) for the conventional dry hopper
(Conventional), for the water fil ed hopper (Water) and control-fish (Control). In the figures X
represent fish that is alive at the end of the experiment, O represent fish that died due to other causes
than fishing mortality (e.g. technical failures) and were excluded from the experiment after O.
Figure 3. Overal survival of sole discards over time (n = 2 sea trips) for the conventional dry hopper
(Conventional), for the water fil ed hopper (Water) and control-fish (Control). In the figures X
represent fish that is alive at the end of the experiment, O represent fish that died due to other causes
than fishing mortality (e.g. technical failures) and were excluded from the experiment after O.
3.2.2
Effects of sea trips and vessels on the effect of the water filled hopper
Within the individual sea trips, significant differences in discards survival probabilities were detected in
sea trips 2, 4 and 9
(Table 5). In al these cases the highest survival probability was observed in the
water fil ed hopper. In al other sea trips during which the water fil ed hopper was tested, the survival
of plaice discards was not significantly different for the water fil ed and the dry hopper.
Wageningen Marine Research report C03/18| 17 of 39
link to page 19 link to page 19 link to page 19 link to page 19 link to page 16 link to page 20
Although the mean plaice discards survival observed for the water fil ed hopper of vessel 2 is nearly
double the survival observed for the other two vessels, no interactive effects of vessel and hopper
treatment on plaice discards survival probability were detected (multilevel linear logistic regression
Ptreatment x Vessel = 0.83,
Table 6). The effect of the water hopper on discards survival probability
apparently was not influenced by the vessel it was tested on.
Table 6 Vessel effects on the effect of a water fil ed hopper on plaice
discards survival probability.
Treatment
Vessel
Discards survival probability
Estimate
95% CI LL 95% CI UL
Dry hopper
1
16%
9%
26%
2
19%
14%
24%
3
12%
8%
19%
Water fil ed hopper
1
15%
9%
25%
2
28%
20%
37%
3
15%
8%
25%
Ptreatment x Vessel
0.83
3.2.3
Effect of a water filled hopper on fish condition
The probability of assigning a test-fish to one of the four vitality index scores is presented for the dry
hopper and the water fil ed hopper in
Table 7. Significant differences in probability estimates were
detected between the two hopper treatments. For plaice, the probability to be assigned a vitality index
score A was higher while the probability to be assigned a vitality index score D was lower for test-fish
col ected from the water filled hopper (Multilevel linear logistic regression,
Table 7). A similar shift
towards a better fish condition was observed among the sole col ected from the water filled hopper,
but only the higher probability to be assigned a vitality index score B was significant
(Table 7). Condition of test-fish, expressed by an individual vitality class score A, B, C or D
(Table 4), was
positively affected by deployment of a water fil ed hopper.
Table 7 Probability of vitality index scores per hopper treatment for plaice and sole. Probabilities per
vitality class (rows) differ among hopper treatments in case p < 0.05 (Multilevel linear logistic
regression).
Species
Vitality
p-value
Probability of vitality index scores
index score
Dry hopper
Water filled hopper
Plaice
A
8%
13%
0.003
B
27%
30%
0.35
C
31%
31%
0.88
D
34%
26%
0.001
Sole
A
5%
11%
0.13
B
28%
34%
<0.001
C
47%
42%
0.61
D
20%
13%
0.21
3.2.4
Interactive effects of fish condition and a water filled hopper on discards
survival probability
No interactive effects of fish condition and hopper treatment on survival probability of plaice discards
were detected (Multilevel linear logistic regression PTreatment x vitality index score =0.44,
Table 8). The survival
probability for plaice discards per vitality class did not differ between the dry hopper and water hopper
treatment. The effect of a water fil ed hopper on the survival probability of plaice discards is not
affected by fish condition. A significant main effect on fish condition on discards survival probability
was detected (Multilevel linear logistic regression P vitality index score <0.001, data not shown).
18 of 39 | Wageningen Marine Research report C038/18
link to page 20 link to page 21 link to page 21
Table 8 Discard survival probability estimates for plaice per vitality index class for the dry and water
fil ed hopper treatments. No interactive effects of fish condition (defined by vitality index score) and
hopper treatment on survival probability were detected.(Multilevel linear logistic regression PTreatment x
vitality index scor e =0.44 ).
Vitality
Discards survival probability
index score
Dry hopper
Water filled hopper
Estimate
95% CI LL 95% CI UL
Estimate 95% CI LL 95% CI UL
A
59%
44%
73%
59%
46%
71%
B
31%
24%
39%
27%
18%
37%
C
4%
2%
10%
9%
5%
15%
D
4%
2%
8%
4%
1%
12%
3.3
The effect of short hauls on discards survival
probability
3.3.1
Main effects of short hauls
The effect of short hauls on plaice discards survival was tested during the first four sea trips (Annex 1,
Figure 4). For al four sea trips combined, short hauls did not affect plaice discards survival (Multilevel
linear logistic regression p= 0.77,
Table 9). Within each of the four individual sea trips, a significant
difference in discards survival probability between the short and conventional haul duration was
detected only in the second sea trip (Multilevel linear logistic regression p= 0.002,
Table 9). In this
sea trip survival probability was highest among test-fish col ected from hauls with a conventional
duration.
Figure 4. Mean survival of plaice discards over time (n = 4 sea trips) for conventional (120min) and
short hauls (90min). In the figures X represent fish that is alive at the end of the experiment, O
represent fish that died due to other causes than fishing mortality (e.g. technical failures) and were
excluded from the experiment after O.
Wageningen Marine Research report C03/18| 19 of 39
link to page 21 link to page 21 link to page 16 link to page 21 link to page 21
Table 9 The effect of short hauls on the estimates and 95% confidence intervals for plaice discards
survival probability for al sea trips combined and for the individual sea trips. Per row, estimates differ
between the short and conventional hauls when p< 0.05.
Sea trip Conventional haul (120 min)
Short haul (90 min)
p-value
Estimate 95% CI LL 95% CI UL Estimate 95% CI LL 95% CI UL
1 to 4
11%
8%
15%
11%
8%
15%
0.77
1
15%
9%
23%
17%
14%
22%
0.54
2
15%
11%
20%
7%
5%
11%
0.002
3
12%
6%
22%
10%
4%
21%
0.77
4
3%
1%
11%
8%
5%
12%
0.24
3.3.2
Effect of short hauls on fish condition
The probability of assigning a test-fish to one of the four vitality index scores is presented for the short
and conventional hauls in
Table 10. No differences in probability estimates were detected between the
two haul duration treatments (Multilevel linear logistic regression,
Table 10). Condition of test-fish,
expressed by an individual vitality class score A, B, C or D
(Table 4), was not affected by the haul
duration treatment.
Table 10 Probability of vitality index scores for plaice for short (90 min) and conventional (120 min)
hauls. Probabilities per vitality class (rows) differ among haul durations in case p < 0.05 (Multilevel
linear logistic regression).
Species
Vitality
Probability of vitality index scores
index score
Conventional haul (120 min)
Short haul (90 min)
p-value
Plaice
A
8%
6%
0.37
B
28%
24%
0.30
C
34%
35%
0.79
D
30%
35%
0.37
3.3.3
Interactive effects of fish condition and short hauls on discards survival
probability
Discard survival probability estimates for plaice per vitality class are presented for short and
conventional hauls in
Table 11. No interactive effects of fish condition and haul duration on survival
probability of plaice discards were detected (Multilevel linear logistic regression PTreatment x vitality index score
=0.42,
Table 11). The survival probability for plaice discards per vitality class did not differ between
short and conventional hauls. The effect of haul duration on the survival probability of plaice discards
is not affected by fish condition.
Table 11 Discard survival estimates for conventional and short hauls per vitality class. No interactive
effects of fish condition (defined by vitality index score) and haul duration on plaice discard survival
probability of plaice discards were detected (Multilevel linear logistic regression PTreatment x hau l duration
=0.42 ).
Vitality
Discards survival probability
index score
Conventional haul (120 min)
Short haul (90 min)
Estimate
95% CI LL 95% CI UL
Estimate 95% CI LL 95% CI UL
A
62%
42%
79%
78%
58%
90%
B
30%
23%
37%
20%
10%
34%
C
3%
2%
7%
4%
1%
11%
D
0%
0%
0%
0%
0%
0%
20 of 39 | Wageningen Marine Research report C038/18
link to page 22 link to page 22 link to page 22
3.4
Effect of a knotless cod-end on discards survival
probability
Estimates for plaice and sole discards survival probability for the conventional and knotless cod-end
are presented
Table 12. For plaice no effect of the cod-end on discards survival probability could be
detected (Multilevel linear logistic regression p = 0.85,
Table 12). A significant effect of the type of
cod-end was detected for sole discards (Multilevel linear logistic regression p < 0.001,
Table 12), based on one surviving test-fish col ected from the conventional cod-end and no survival among the
test-fish col ected from the knotless cod-end.
Table 12 The effect of a knotless cod-end on the estimates and 95% confidence intervals for plaice
and sole discards survival probability.
Species Sea trip
Conventional cod-end
Knotless cod-end
p-value
Estimate 95% CI LL 95% CI UL Estimate 95% CI LL 95% CI UL
Sole
5
3%
1%
13%
0%
0%
0%
<0.001
Plaice
5
1%
0%
9%
2%
0%
11%
0.85
Wageningen Marine Research report C03/18| 21 of 39
4
Discussion
4.1
General
The effect of three measures aimed at increasing survival of discards in pulse-trawl fisheries were
investigated. The survival monitoring periods of 15 to 18 days were sufficiently long as mortality had
levelled out in al cases before survival monitoring was terminated.
Control-fish were deployed to detect any mortality potential y caused by the experimental procedures
instead of being fisheries induced. Survival among control-fish was consistently high at 84% for plaice
over al nine sea trips and >90% for sole over three trips. The lower survival among plaice control-fish
is caused by three trips with lower than 90% survival. We attribute the low control-fish survival of
30% in sea trip 5 to the poor state the control plaice were in prior to the sea trip rather than the
experimental procedures, even though this not reflected by the vitality index scores of these fish. The
water fil ed hopper was not tested in this sea trip, so our conclusions on the most important measure
to increase discards survival in this study remain entirely unaffected by the low survival among
control-fish in sea trip 5.
We cannot explain the lower survival (72%) among plaice control-fish in sea trips 7 and 8. However,
since the mortality among control-fish started after the mortality in the test-fish had already
stabilized, we do not attribute this mortality to the experimental procedures at sea. Although unlikely,
we cannot entirely exclude that the experimental procedures caused some additional mortality on top
of the fisheries induced mortality, especial y among test-fish for plaice. Since survival probability
estimates were not correct in case control-fish survival was < 100%, the presented survival
probability estimates may be slight underestimations. On the other hand, it cannot be excluded that
the presented survival probabilities are slight overestimations because potential post-discarding
predation by sea birds and other species was not incorporated in the experiment, although it is
unknown to what extent the discarded fish are preyed upon when discarded.
4.2
Water filled hopper
The effect of a water fil ed hopper on discards survival was tested on plaice during eight sea trips and
on sole during two sea trips. For al sea trips combined, deployment of a water fil ed hopper resulted in
a significant increase in survival probability for sole discards but not for plaice discards. Although the
effect was significant over both trips for sole, only one trip showed a strong increase in survival
probability; in the other trip nearly al sole died. Prior to the experiment, we hypothesized that
measures that result in an increased survival in plaice discards, also benefits discards survival in other
species. The actual increase in survival then remains to be quantified for each individual species. From
this study it appears that a significant effect in one species does not necessarily imply that other
species also significantly benefit from the same measure. Although a generic positive effect of the
water fil ed hopper was not detected, survival was higher among plaice col ected from the water fil ed
hopper for three sea trips. In sea trip 9, the plaice survival probability was at 45% even 25% higher
than the survival probability observed for the dry hopper. It is seems that a positive effect of a water
fil ed hopper on plaice discards survival occurs under certain circumstances. Indeed, the conditions
during the three sea trips that yielded a positive effect where comparable with low wind speeds and
limited waves heights, and during sea trip 9 the water temperature was among the lowest in the
experiment. Establishing the extent to which an effect of the water fil ed hopper on discards survival
interacts with environmental conditions as wel as other factors is beyond the scope of the current
study but wil be the objective of our further studies using the data col ected in the current
experiment.
It should be noted that sole was only tested in two sea trips and it cannot be excluded that, in line
with our observations on plaice, more tests with sole may results in absence of significant effects in
some of the additional sea trips. Clearly, it remains to be established whether the positive effect of a
22 of 39 | Wageningen Marine Research report C038/18
water fil ed hopper on sole discards survival probability is generic or specific for the conditions in which
the current tests took place. Surprisingly, for sea trip 6 in which a positive effect for sole was found,
no significant effect of the water fil ed hopper on plaice discards survival was detected. This suggests
that the conditions under which a water fil ed hopper is effective, vary among fish species.
We previously established that survival probability increases with improving condition of the discarded
fish (Schram and Molenaar, 2018). To investigate mechanisms that underlie the effect of a water fil ed
hopper, we looked into the effect of this measure on the condition of discarded fish. The larger
proportion of fish with an vitality index score A combined with a lower proportion of score D among
plaice col ected from the water fil ed hopper compared to the dry hopper, indeed indicates a shift
towards a better fish condition as a result of deploying the water fil ed hopper. We also compared the
survival per vitality index scores (classes A, B C and D) for a water fil ed and dry hopper but detected
no interactive effects. Clearly none of the vitality index score classes particularly benefits from using a
water fil ed hopper instead of a dry hopper. This notion combined with the shift towards a better fish
condition among fish from the water fil ed hopper, suggests that the smal positive effect of the water
fil ed hopper on discards survival acts through prevention of deterioration of fish condition during cod-
end discharging and catch processing. The proportion of fish in good condition (score A) remains
however low, 13% over al sea trips, among plaice col ected from the water fil ed hopper. This
indicates that using a water fil ed hopper only marginal y prevents deterioration of fish condition. It
seems that in general fish condition is mainly determined by the preceding capture process rather
than the catch-sorting process. In that case the scope to improve fish condition and subsequently
increase survival by measures implemented in the catch-sorting process is smal . However, focussing
on sea trip 9 for which a difference in discards survival probability as large as 25% was detected
between the water fil ed and dry hopper, reveals a concurring difference in the proportion of fish in
good and bad condition. For this trip 40% of the fish from the water fil ed hopper scored either a
vitality index A or B while this proportion was only 15% among the fish from the dry hopper.
Assuming that there is no variation in fish condition between port- and starboard side catches from
the same hauls; fish from both nets arrive on deck in identical condition that cannot improve during
catch-sorting, it seems clear that during this trip the water filled hopper prevented to a large extent
the deterioration of fish condition during the catch-sorting process.
It should be noted that the water fil ed hopper does not consistently yield higher discards survival than
the dry hopper. For three out of the eight trips during which the water fil ed hopper was tested, higher
survival, although not significantly, was observed for the dry hopper. This may be part of the reason
why for al trips combined, no significant effect of the water fil ed hopper could be detected. It further
strengthens the notion that an effect of a water fil ed hopper on plaice discards survival, either
positive or negative, depends on the fishing conditions. The possibility that the water fil ed hopper can
also have a negative effect on discards survival chances should be taken into consideration. Negative
effects of a water fil ed hopper on discards survival may act through depletion of dissolved oxygen
since fish are more prone to suffocation in water with low levels of dissolved oxygen than when
exposed to air outside the water. Unfortunately we did not systematical y measure dissolved oxygen
levels during deployment of the water fil ed hopper and thus have no data that could corroborate this
notion. Sloshing of the water in the hopper, observed during heavy seas, may be another possible
mechanism underlying a negative effect of the water fil ed hopper on discards survival. Although not
systematical y observed nor measured, sloshing the water in the hopper seems to increase physical
impacts of fish with the hopper itself as well as other fish, benthic organisms and debris in the catch.
It is not unlikely that increased physical impacts lead to deterioration of fish condition and
subsequently lower survival probability of discarded fish. Interaction with catch composition, e.g. the
amount of pebbles in the catch, then seems likely. Cumulative effects of sloshing and oxygen
depletion are unlikely as sloshing would promote oxygenation of the water in the hopper.
Summarizing, it is clear that deployment of a water fil ed hopper can result in a large increase of
survival chances of discarded fish compared to the dry hopper. However, it seems that a water fil ed
hopper only effectively increases discards survival and fish condition under certain specific, yet to be
established, conditions. In addition, potential negative effects of a water fil ed hopper on discards
survival cannot be excluded at this point and should not be neglected. Again, it seems that the effect
depends on the conditions.
Wageningen Marine Research report C03/18| 23 of 39
4.3
Short hauls and knotless cod-end
The effect of reduced haul duration, 90 instead of 120 min, on discards survival probability was
investigated during four sea trips. We predicted shorter hauls to result higher discards survival
through a reduction of capture process related physical impacts on the fish in the cod-ends. Such
reduction in physical impacts could be associated with a reduced retention time of fish in the cod-end
as wel as smal er catches. Surprisingly, for the four sea trips combined, no effect of shorter hauls on
discards survival could be detected. In one sea trip, the shorter hauls even resulted to a lower survival
among plaice discards. For the other three sea trips no effect was detected. In line with the absence of
an effect of short hauls on discards survival, no effect of short hauls on fish condition was detected.
Hauls as short as approximately 60 min were previously found to increase survival of plaice discards in
the same fishery (Van der Reijden et al., 2017). This suggests that a reduction of haul duration from
120 to 90 min is insufficient to sort such effect on discards survival. The relatively smal reduction of
the haul duration in our study was deliberately chosen as hauls of 90 min may stil be practical y
feasible for the vessel’s crew if proven to be very effective in increasing discards survival. While
reduction of haul duration below 90 min may lead to a higher increase in discards survival, practical
implementation is unrealistic in view of the potential reduction of catches due to the reduction of
effective fishing time (by approximately 17%) combined with an almost double workload for the crew.
A knotless cod-end was tested during one sea trip only. Survival of plaice and sole discards col ected
from the knotless cod-end as wel as the conventional cod-end which was deployed in the same hauls
was very low. In fact, only one of the sole col ected from the conventional cod-end survived while
those col ected from the knotless cod-end al died. Catch composition and environmental conditions
could have contributed to this low survival. Nevertheless, based on this single test it seemed clear that
the knotless cod-end was not a breakthrough measure with a large positive effect on discards survival.
Testing of the knotless cod-end was consequently not continued.
24 of 39 | Wageningen Marine Research report C038/18
5
Conclusions and recommendations
Deployment of a water fil ed hopper does not result in higher survival probability for plaice discards
than a conventional dry hopper in year-round pulse-trawl fisheries. However, it is clear that for
individual trips the deployment of a water fil ed hopper instead of a dry hopper can result in an
increase of survival chances of discarded plaice, but as it seems only under certain specific, yet to be
established, conditions. In addition, it cannot be excluded at this point that under certain conditions a
water fil ed hopper has a negative effect on discards survival. Our further analysis of data col ected
during the current study is expected to provide more insight in conditions that result in either
increased or decreased discards survival. It is expected that such insights contribute to optimization of
the use of the water fil ed hopper for the benefit of discards survival chances.
The condition of the individual fish does not interact with the effect of the water fil ed hopper on
discards survival chances; none of the vitality index score classes particularly benefited from using a
water fil ed hopper instead of a dry hopper. However, deployment of a water fil ed hopper does result
in a shift towards a better condition of the discarded fish. This is important because fish in good
condition have higher survival chances when discarded than fish in poor conditions. Despite the use of
a water fil ed hopper, the total proportion of fish in good condition within a catch remains smal . It
should be noted that when conditions during trawling result in in poor condition for a large part of the
fish that arrive on deck, fish condition and survival chances cannot be regained by on-board
measures. We therefore recommend to prioritize measures aimed at improving fish condition in the
trawl to increase discards survival chances.
The two other measures tested showed no potential to increase discards survival chances. Plaice
discards survival probability is not increased by a reduction of haul duration from 120 to 90 min. It is
unlikely that deployment of a knotless cod-end results in a higher survival probability for plaice
discards than a conventional cod-end.
Wageningen Marine Research report C03/18| 25 of 39
6
Acknowledgements
This study was commissioned by VISNED, The Netherlands. This study was partly funded by the
European Union, European Maritime and Fisheries Fund (EMFF). The authors would like to thank the
fol owing persons and organisations for their indispensable contributions to this project. The owners,
skippers and crews of the UK33, TX3 and GO23 for welcoming researchers on board of their vessels
and enabling research at sea. The skippers and crews of the TH10 and OD3 for col ecting control-fish
at sea. Richard Martens and Wouter van Broekhoven for project management. Pim van Dalen, Ainhoa
Blanco, Ad van Gool, Emiel Brummelhuis and Yoeri van Es for al the practical work related to the
col ection of control-fish, preparation of sea trips and survival monitoring in the laboratory. Van Wijk
instal aties en constructies BV, Maaskant Shipyards Stel endam BV and Visserij Coöperatie Urk (VCU)
for preparing and instal ing the technical instal ations at the vessels required for the research. Ewout
Blom, Nathalie Steins, Joe Freijser (De Aquanoom), Raoul Kleppe (Wageningen University) and Pim
Boute (Wageningen University) for taking part in the sea trips. Mulder Transport BV for transporting
the survival monitoring units between vessels and the laboratory. Schreuder Transport BV and
Steketee BV for transporting equipment and control-fish to the vessels. Tom Catchpole (CEFAS) for
guidance with the protocols for reflex scoring in rays. Sebastian Uhlmann (ILVO) for providing training
in reflex scoring in turbot and brill. Jan Jaap Poos, Sander Glorius and Pepijn de Vries for their
assistance in data analysis.
26 of 39 | Wageningen Marine Research report C038/18
7
Quality Assurance
Wageningen Marine Research utilises an ISO 9001:2008 certified quality management system
(certificate number: 187378-2015-AQ-NLD-RvA). This certificate is valid until 15 September 2018. The
organisation has been certified since 27 February 2001. The certification was issued by DNV
Certification B.V.
Furthermore, the chemical laboratory at IJmuiden has NEN-EN-ISO/IEC 17025:2005 accreditation for
test laboratories with number L097. This accreditation is valid until 1th of April 2021 and was first
issued on 27 March 1997. Accreditation was granted by the Council for Accreditation. The chemical
laboratory at IJmuiden has thus demonstrated its ability to provide valid results according a
technical y competent manner and to work according to the ISO 17025 standard. The scope (L097) of
de accredited analytical methods can be found at the website of the Council for Accreditation
(www.rva.nl).
On the basis of this accreditation, the quality characteristic Q is awarded to the results of those
components which are incorporated in the scope, provided they comply with al quality requirements.
The quality characteristic Q is stated in the tables with the results. If, the quality characteristic Q is
not mentioned, the reason why is explained.
The quality of the test methods is ensured in various ways. The accuracy of the analysis is regularly
assessed by participation in inter-laboratory performance studies including those organized by
QUASIMEME. If no inter-laboratory study is available, a second-level control is performed. In addition,
a first-level control is performed for each series of measurements.
In addition to the line controls the fol owing general quality controls are carried out:
Blank research.
Recovery.
Internal standard
Injection standard.
Sensitivity.
The above controls are described in Wageningen Marine Research working instruction ISW 2.10.2.105.
If desired, information regarding the performance characteristics of the analytical methods is available
at the chemical laboratory at IJmuiden.
If the quality cannot be guaranteed, appropriate measures are taken.
Wageningen Marine Research report C03/18| 27 of 39
References
Benoît, H.P., Plante, S., Kroiz, M., Hurlbut, T. 2013. A comparative analysis of marine fish species
susceptibilities to discard mortality: effects of environmental factors, individual traits, and phylogeny.
ICES Journal of Marine Science 70 (1, 1), 99–113.
Enever, R., Catchpole, T.L., El is, J.R., Grant, A. 2009. The survival of skates (Rajidae) caught by
demersal trawlers fishing in UK waters. Fisheries Research 97 (1–2).
European Union. 2013. REGULATION (EU) No 1380/2013 OF THE EUROPEAN PARLIAMENT AND OF
THE COUNCIL of 11 December 2013 on the Common Fisheries Policy, amending Council Regulations
(EC) No 1954/2003 and (EC) No 1224/2009 and repealing Council Regulations (EC) No 2371/2002
and (EC) No 639/2004 and Council Decision 2004/585/EC. Official Journal of the European Union,
L354/22
Kaplan, E.L., Meier, P. 2012. Nonparametric Estimation from Incomplete Observations, Journal of the
American Statistical Association, 53:282, 457-481,
Schram, E., Molenaar, P. 2018. Discards survival probabilities of flatfish and rays in North Sea pulse-
trawl fisheries. Wageningen Marine Research Report
C037/18.
. 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., Molenaar, P., Van der Reijden, K.J., Goudswaard, P.C., Bol, R.A., Glorius, S.T.,
Theunynck, R., Uhlmann, S.S. 2016. Overleving van discard platvis – Vaststel en en verhogen.
IMARES rapport C180/15, IMARES Wageningen UR.
28 of 39 | Wageningen Marine Research report C038/18
Justification
Report C038/18
Project Number: 4311400003
The scientific quality of this report has been peer reviewed by a col eague scientist and a member of
the Management Team of Wageningen Marine Research
Approved:
Dr. ir. N.A. Steins
Programme Manager
Signature:
Date:
15 May 2018
Approved:
Dr. ir. T.P. Bult
Director
Signature:
Date:
15 May 2018
Wageningen Marine Research report C03/18| 29 of 39
Annex 1: Survival per trip
Trip 1
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
1
UK33
2017
May
18
-
9-12
2-5
0.5-2.0
25
85-135
26-28
Survival of plaice
#Control
# Test
# Water # Short
100%
15%
18%
11%
30 of 39 | Wageningen Marine Research report C038/18
Trip 2
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
2
GO23
2017
May
21
14-18
12-13
1-4
0.2-0.5
24
90-120
39-50
Survival of Plaice
#Control
# Test
# Water # Short
97%
15%
29%
17%
Wageningen Marine Research report C03/18| 31 of 39
Trip 3
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
3
TX3
2017
June
24
15-20
14-15
1-2
0.1-0.5
18
90-125
22-25
Survival of Plaice
#Control
# Test
# Water # Short
100%
12%
15%
10%
32 of 39 | Wageningen Marine Research report C038/18
Trip 4
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
4
TX3
2017
July
28
15-21
16-17
1-6
0.2-1.0
18
90-120
28-40
Survival of Plaice
#Control
# Test
# Water # Short
90%
3%
10%
8%
Wageningen Marine Research report C03/18| 33 of 39
Trip 5
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
5
UK33
2017
Sept
36
15-18
18
4-5
0.5-1.0
21
120-130
26-37
Survival of Plaice
Survival of Sole
#Control
# Test
# Knotless
#Control
# Test
# Knotless
30%
1%
2%
90%
3%
0%
34 of 39 | Wageningen Marine Research report C038/18
Trip 6
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration
depth (m)
e (°C)
e (°C)
(min)
(min)
6
TX3
2017
Oct
44
13-15
13-15
3-4
1.0-1.2
20
120-130
30-31
Survival of Plaice
Survival of Sole
#Control
# Test
# Water
#Control
# Test
# Water
100%
22%
18%
100%
10%
25%
Wageningen Marine Research report C03/18| 35 of 39
Trip 7
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
7
GO23
2017
Dec
49
6-9
11-12
3-5
1.0-2.0
33
120
37-50
Survival of Plaice
Survival of Sole
#Control
# Test
# Water
#Control
# Test
# Water
72%
20%
10%
100%
0%
3%
36 of 39 | Wageningen Marine Research report C038/18
Trip 8
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration depth (m)
e (°C)
e (°C)
(min)
(min)
8
UK33
2018
Jan
4
8
6-7
5-6
1.0-1.7
33
120
28-35
Survival of Plaice
#Control
# Test
# Water
72%
17%
12%
Wageningen Marine Research report C03/18| 37 of 39
Trip 9
Trip
Vessel Year Month Week
Air
Water
Wind speed
Wave
Catch
Haul
Fishing
temperatur temperatur
(Bft)
height (m) processing duration
depth (m)
e (°C)
e (°C)
(min)
(min)
9
GO23
2018
Feb
8
6
7-8
3-4
0.5-1.0
24
110-120
40-45
Survival of Plaice
#Control
# Test # Water
93%
20%
45%
38 of 39 | Wageningen Marine Research report C038/18
Wageningen Marine Research
Wageningen Marine Research is the Netherlands research institute
T +31 (0)317 48 09 00
established to provide the scientific support that is essential for developing
E: xxxxxxxxxxxxxxx@xxx.xx
policies and innovation in respect of the marine environment, fishery
www.wur.eu/marine-research
activities, aquaculture and the maritime sector.
Visitors’ address
Wageningen University & Research:
•
Ankerpark 27 1781 AG Den Helder
is specialised in the domain of healthy food and living environment.
•
Korringaweg 7, 4401 NT Yerseke
•
Haringkade 1, 1976 CP IJmuiden
The Wageningen Marine Research vision
‘To explore the potential of marine nature to improve the quality of life’
The Wageningen Marine Research mission
•
To conduct research with the aim of acquiring knowledge and offering
advice on the sustainable management and use of marine and coastal
areas.
• Wageningen Marine Research is an independent, leading scientific
research institute
Wageningen Marine Research is part of the international knowledge
organisation Wageningen UR (University & Research centre). Within
Wageningen UR, nine specialised research institutes of the Stichting
Wageningen Research Foundation have joined forces with Wageningen
University to help answer the most important questions in the domain of
healthy food and living environment.
Wageningen Marine Research report C03/18| 39 of 39
Document Outline