ICES Journal of
Marine Science
ICES Journal of Marine Science (2015), 72(3), 841– 850. doi:10.1093/icesjms/fsu162
Original Article
Can permanent closures of nearshore areas reduce the proportions
of undersized fish in the Icelandic longline fishery?
Björn Björnsson*, Jón Sólmundsson, and Ólafur K. Pálsson
Marine Research Institute, PO Box 1390, Skúlagata 4, 121 Reykjavı́k, Iceland
*Corresponding author: tel: +354 575 2045; fax: +354 575 2001; e-mail: [email protected]
Björnsson, B., Sólmundsson, J., and Pálsson, Ó. K. Can permanent closures of nearshore areas reduce the proportions of undersized
fish in the Icelandic longline fishery? – ICES Journal of Marine Science, 72: 841 – 850.
Received 23 June 2014; revised 27 August 2014; accepted 29 August 2014; advance access publication 30 September 2014.
In recent years, about one-third of the cod (Gadus morhua) and haddock (Melanogrammus aeglefinus) catches in Icelandic waters have been
obtained with longline. Although longlining has been regarded as a conservation-oriented fishing method, a serious drawback of this fishing
gear is the high catches of undersized fish. Our purpose was to locate areas where catches of undersized fish are high and consider if it may be feasible
to close them permanently for longlining. Extensive length measurements used in the analysis were made by official inspectors on-board longliners
during 2005 –2013 in the main fishing area around Iceland. We found that the percentage of undersized cod (,55 cm total length) and haddock
(,45 cm) in those samples decreased in relation to both depth and distance from shore. Our results suggest that permanent closures of large
nearshore areas for longlining is a feasible option to reduce catch of undersized cod and haddock.
Keywords: cod, distance from shore, fishery closure, haddock, length distribution, longline.
Introduction
The optimum harvesting size of the fish in a year class has been
defined as the one at which the instantaneous rates of growth and
natural mortality are equal (Ricker, 1945). Although harvesting at
optimum size would be possible in fish farming, it is not possible
in a fishery of a wild stock in which cropping must be spread over
a period. The lower the fishing effort, the broader the range of
sizes that must be taken to obtain optimal yield (Ricker, 1975).
Different fish species require different harvesting strategies based
on their life history biology (Beverton and Holt, 1957).
The large potential economic benefits of sparing young fish by
increasing mesh size has been outlined for Northeast Arctic cod
(Gadus morhua) (Diekert et al., 2010a, b). However, other
workers have pointed out the possible negative ecological and evolutionary consequences of size-selective harvesting (Fenberg and
Roy, 2008; Stephen et al., 2011). Thus, in fish stocks that are
heavily exploited with size-selective gear, such as the trawl, the genotype that matures at a small size has a greater chance of producing
offspring than the one that matures at a large size. As this may, in
time, reduce the genetic growth potential of a stock, an alternate
strategy has been proposed based on harvesting mainly intermediate
lengths of fish, sparing both young and old fish (Gwinn et al., 2013).
For the Icelandic cod fishery, there have been strict measures to
spare both young and old age classes. A number of regulations have
# International
been issued to reduce the catch of small fish, e.g. minimum mesh
size in bottom trawls, bottom seines, and gillnets; sorting grids in
trawls; and temporary and long-term area closures on nursery
grounds (Schopka, 2007). The following measures imposed in
Iceland have both directly and indirectly increased the survival of
large cod: prohibition of purse-seines in catching spawning cod (in
1970), total allowable catch (in 1976), individual transferable quota
(ITQ) (in 1984), 10–21 day closures of spawning areas in April (in
1992), prohibition of large mesh sizes (.8 in.) in gillnets (in 2006),
and harvest control rule (catching 25 and 20% of fishable stock per
year, implemented in 1996 and 2008, respectively) (Thorsteinsson,
1980; Schopka, 2007; Marine Research Institute, 2014).
Longlining has been regarded as a size-selective and conservationoriented fishing method (Bjordal, 1989). Compared with trawling,
longlining causes less damage to the seabed and benthic fauna,
causes less scraping and bruising of the catch, and uses less energy
(Løkkeborg and Bjordal, 1992; Rotabakk et al., 2011). Although
a higher percentage of large cod is caught with longline than
bottom trawl within a given area (Huse et al., 2000), it can be
more difficult in longlining than in trawling to avoid catches of
undersized fish. Sufficiently, large mesh size in trawls allows small
fish to escape, whereas conventional longline hooks catch both
small and large fish. For this reason, attempts have been made to
modify the longline to make it more size selective.
Council for the Exploration of the Sea 2014. All rights reserved.
For Permissions, please email: [email protected]
842
Fewer small fish may be caught if the size of the hook and/or bait
is increased, but this usually leads to a reduction in the total catch of
larger fish (Løkkeborg and Bjordal, 1992; Halliday, 2002). As larger
bait increases fishing cost, experiments with large inedible bait have
been carried out to reduce the catch of small fish (Løkkeborg and
Bjordal, 1995; Huse and Soldal, 2000), but with relatively limited
success. Also, due to larger fish having larger feeding ranges than
small fish, the mean size of fish caught with longline can be increased
by increasing the spacing between hooks (Hamley and Skud, 1978).
However, this reduces the maximum number of fish that can be
caught with one longline set and thus will increase fishing cost
(Bjordal, 1989; Løkkeborg and Bjordal, 1992).
A temporary closure system has been in force in Icelandic waters
since 1976 with the objective to protect juvenile fish and hence
reduce discards (Schopka, 2007). According to government regulation, the Marine Research Institute has the authority to close
fishing grounds temporarily for certain gear types if the proportion
of small fish in the catch exceeds certain limits. Generally, a given
fishing area is closed for 2 weeks, usually due to large proportions
of juvenile cod and/or haddock (Melanogrammus aeglefinus) in
the catch, based on length measurements by official inspectors.
The common procedure is that a relatively restricted area is closed
if .25% of cod are ,55 cm (total length) and/or .30% of haddock
are ,45 cm. There are other criteria used for other species. These
temporary closures are quite common, especially in the longline
fishery. If a given area is closed repeatedly, it could be closed permanently for a given fishing gear for months or years, but such measures
are uncommon.
Despite the above-mentioned attempts to reduce the catch of juvenile fish, the proportion of undersized cod and haddock in longline catches is still high in Icelandic waters. Therefore, permanent
closures of areas where a high percentage of juvenile fish persists
might be a useful strategy in reducing the catch of undersized fish.
Although fishing captains have incentives to set their longlines in
areas with high catches of large fish, they may decide to fish in
areas with high catches of small fish if the number of large fish is
sufficiently high. Sometimes they may inadvertently set their lines
in areas with mainly small fish. At other times, due to weather conditions or other constraints, they may deliberately set their lines in
areas known to contain mainly small fish, for example inside
fjords where important nursery grounds are located. Currently,
longlining is allowed inside all Icelandic fjords, whereas trawling
for fish is not permitted.
The main purpose of this study was to (i) locate areas in Icelandic
waters where the proportion of undersized cod and haddock is high
in the longline fishery and (ii) consider the feasibility of permanently closing these areas to increase the harvesting potential of the two
fish stocks.
B. Björnsson et al.
during the period 2005–2013. On average, 226 cod and 190
haddock were measured in each sample. The total number of length
measurements in longline samples was 301 741 cod and 202 083
haddock. Additionally, length measurements of cod caught with
bottom trawl (n ¼ 96 976), handline (n ¼ 15 584), and bottom
seine (n ¼ 48 569) were used in the study. For each sample, the following variables were recorded: date, geolocation, and depth at one or
both ends of the longline and total length of individual cod and
haddock. In the analysis, the average geolocation and average depth
of each longline was used when two values were recorded. These
data were also grouped according to statistical squares, 0.58 latitude ×
1.08 longitude in size. There were too few length measurements made
for wolffish (Anarhichas lupus), the third most important species
caught with longline, to include it in the analysis.
The distance between the geolocation of each longline set and the
main coast, ignoring islands, was calculated using an in-house geo
package in the computer program R version 3.0.2 (R Core Team,
2013), using a data array with 1323 locations along the coastline
of Iceland. The distance to every location was calculated, and the
lowest value was used to represent the distance to shore. To estimate
how far the temporary closure areas were from the coast, the distance of each corner point of the area’s perimeter from the main
coastline was calculated. In 90% of the cases, each closure area was
defined by four corner points, and in 8% of the cases, by 5 –6
corner points.
Fish samples from landed catches were used to assess the longterm (1951 –2013) trend in mean length of cod caught with trawl
vs. longline. The mean length in each fish sample was calculated
and the averages of these means for periods of 5 years, to obtain
sufficient sample size for all periods.
Catch density plots for cod, haddock, and wolffish were drawn
based on data from mandatory logbooks of all longliners in
Icelandic waters in 2005–2013. In this database, the catches are
amounts of fish in each set as estimated by captains during the
fishing trip and may differ slightly from the landed catch. The
average annual catch in tonnes per square mile during the whole
period was calculated with the geocontour.fill function in the geo
package, to smooth the data with pointkriging. These catch data
for individual species were also grouped into zones of various distances from the main coastline.
Additional information about catch, fishing location, and size of
individual longliners was obtained from the above-mentioned
database for the years 2010–2013 to assess the possible implications
of a fishery closure for the fleet. In the analysis, length of boats
was used as a measure of boat size. Vessels with a total catch of
cod ,10 t during the 4-year period were excluded from the analysis.
Results
Methods
Longline catches
For the study period 2005–2013, the criteria for a temporary closure
were unaltered for cod: .25% of fish ,55 cm in total length. For
haddock, the criteria were .30% of fish ,45 cm in length except
for the period November 2007–January 2009 when they were
.25% of fish ,41 cm in length. The reason for this alteration
was that the exceptionally large 2003 year class of haddock would
have made it difficult for the fleet to obtain the haddock quota
without catching too many sexually mature fish (T. Sigurdsson,
Marine Research Institute, pers. comm.). In this paper, we refer to
cod ,55 cm and haddock ,45 cm as undersized fish.
The present analysis is based on samples collected by government fishery inspectors on-board fishing vessels in Icelandic waters
Cod, haddock, and wolffish were the fish species of greatest importance in the longline fishery in Icelandic waters during 2005–2013,
especially in nearshore waters. Within a 10-km nearshore zone
(5.4 nautical miles), these three species constituted 94% of the
total longline catch (Table 1). During this period, the largest longline
catches of cod were taken in a relatively continuous and wide belt
reaching clockwise around Iceland from the southwest to the southeast coast (Figure 1a). The longline catches of haddock were taken in
more restricted areas closer to shore (Figure 1b), whereas the catches
of wolffish were mostly from a few small areas close to shore off the
northwest and the east coast of Iceland (Figure 1c). During the study
period, ca. 75% of the longline catches of cod, haddock, and wolffish
843
The proportions of undersized fish in the Icelandic longline fishery
Table 1. A list of fish species caught by longline in Icelandic waters
during 2005– 2013.
Species
Cod (G. morhua)
Haddock (M. aeglefinus)
Wolffish (A. lupus)
Ling (Molva molva)
Tusk (B. brosme)
Spotted wolffish (Anarhichas minor)
Redfish (Sebastes spp.)
Whiting (Merlangius merlangus)
Saithe (Pollachius virens)
Starry ray (Raja radiata)
Blue ling (Molva dypterygia)
Halibut (Hippoglossus hippoglossus)
Other species
All species
Catch <10 km
Catch >10 km
Tonnes
10 882
9040
2543
403
343
83
60
57
47
22
17
7
435
23 941
Tonnes
51 318
18 981
3613
4289
5600
1067
944
170
481
322
1750
165
1521
90 219
%
45.4
37.8
10.6
1.7
1.4
0.3
0.3
0.2
0.2
0.1
0.1
0.0
1.8
100.0
%
56.9
21.0
4.0
4.8
6.2
1.2
1.0
0.2
0.5
0.4
1.9
0.2
1.7
100.0
Mean annual catch in tonnes ,10 and .10 km from the main coast and the
catch of each species as a percentage of the catch of all species.
were taken 10– 80, 0 –30, and 0 –20 km from shore, respectively, and
18, 32, and 41% of their catches within 10 km of the coast (Table 2).
The sampling of fish by inspectors on-board longliners adequately reflects the area where most of the cod (Figure 2a) and
haddock (Figure 2b) were captured in the longline fishery
(Figure 1a and b) and represents various depths and distances
from the main coast. Cod samples were taken in 93 statistical
squares and haddock in 79 squares. The number of squares where
undersized fish represented .25% of the catch was 30 for cod
(Figure 2a), but only 7 for haddock (Figure 2b); only 4 squares for
haddock if the 30% rule of a temporary closure for this species is
used. Undersized cod were most common in a large area west,
north, and east of Iceland (Figure 2a), whereas high percentages of
undersized haddock were only found close to shore (Figure 2b).
For the whole dataset, the percentage of undersized cod in the
longline fishery decreased with depth and distance from shore
(Table 3, Figure 3). The criterion for a temporary closure for cod
(.25% undersized fish) was met for depths ,80 m and distances
from shore ,10 km. For haddock, the percentage also decreased
with depth and distance from shore (Table 3, Figure 4). The criterion
for a temporary closure for haddock (.30% undersized fish) was
not met for any depths or distances from shore.
There was a sharp increase in length frequencies from 54 to 55 cm
for cod and from 44 to 45 cm for haddock in samples taken in relatively shallow water and close to shore where the incidence of undersized fish was high. No such sharp increase was observed in samples
collected in deep water and far from shore where the incidence of
undersized fish was low (Figures 3 and 4).
Annual changes
The percentage of undersized cod within 10 km from shore ranged
from 31 to 44% during 2005–2013, except in 2011 when it was 20%
(Figure 5a). Farther than 10 km from shore, the percentage of
undersized cod was relatively low and stable, ranging between 14
and 23% (Figure 5a). The percentage of undersized haddock
within 10 km from shore decreased during 2007–2013 from 32 to
3% (Figure 5a). Farther than 10 km from shore, the percentage of
undersized haddock also decreased during 2007–2013 from 20 to
2% (Figure 5a). In each case, the number of length measurements
was .1000 and in more than half of the cases .10 000.
Figure 1. Location of annual fish catches (tonnes per nautical square
mile) in the Icelandic longline fishery during 2005 – 2012, (a) cod,
(b) haddock, and (c) wolffish. Depth contours at 100, 200, and 500 m.
Table 2. Average annual longline catch of cod, haddock, and
wolffish in tonnes and percentage of total catch in Icelandic waters
during 2005– 2013 by distance (km) from shore.
Cod
Distance (km)
,5
5 –10
10– 20
20– 30
30– 40
40– 80
.80
Total
% of cod at all depths
% of cod ,10 km
Haddock
Tonnes %
Tonnes %
4743
7.6 4136
14.8
6140
9.9 4903
17.5
12 707
20.4 7940
28.3
10 797
17.4 4613
16.5
7793
12.5 2325
8.3
15 326
24.6 3304
11.8
4694
7.5
799
2.9
62 200 100.0 28 021 100.0
100
45.0
100
83.1
Wolffish
Tonnes %
853
13.8
1690
27.4
2309
37.5
698
11.3
318
5.2
261
4.2
26
0.4
6156
100.0
9.9
23.4
Catch in each set as estimated by captains. The catch of each species at all
depths and ,10 km from shore expressed as percentage of the cod catch.
844
B. Björnsson et al.
Figure 2. Statistical squares where the percentage of undersized (a) cod and (b) haddock was ,15% (green), 15– 25% (yellow), 25– 30% (orange),
and .30% (red) in samples measured by fishery inspectors on-board longliners during 2005– 2013. Numbers of length measurements indicated for
each statistical square in thousands (i.e. 0 ¼ 100 – 499, 1 ¼ 500 –1499, etc.).
Seasonal changes
The percentage of undersized cod within 10 km from shore was
lowest during April and May (18%), but highest in July–
October (40– 52%) (Figure 5b). Farther than 10 km from shore,
the percentage of undersized cod was lowest in March (6%) and
highest in July (30%) (Figure 5b). The percentage of undersized
haddock within 10 km from shore was lowest in March (16%)
and highest in April (36%) (Figure 5b). Farther than 10 km from
shore, the percentage of undersized haddock was lowest in
February (8%) and highest in June (21%) (Figure 5b). In each
case, the number of length measurements was .1000 and in
one-third of the cases .10 000.
Other gear types
In Icelandic waters, most of the total catch of cod (Figure 6a),
haddock (Figure 6b), and wolffish (Figure 6c) is taken with
bottom trawl and longline. For the last three decades, the proportion
taken with longline has been increasing, while the proportion taken
with trawl and gillnet has been decreasing for both cod and haddock
(Figure 6a and b). During the last 3 years, 32, 39, and 56% of the total
catch of cod, haddock, and wolffish, respectively, were taken with
longline (Figure 6a– c). During the period 1977–2012, there were
two- to threefold changes in the total yearly catches of these three
species ranging from 151 000 to 460 000 t for cod, 35 000 to
109 000 t for haddock and 8000 to 18 000 t for wolffish (Figure 6d).
Comparison of the length distribution of cod caught with trawl,
longline, handline, and bottom seine, as measured by fishery inspectors at sea during 2005–2013, shows that the highest proportion
of undersized cod (,55 cm) was caught with handline and longline.
The highest proportion of large cod (≥70 cm) was caught with trawl
and bottom seine (Figure 7). The distance from shore at which most
of the cod are caught varies from 60 to 90 km for trawl, ,50 km for
845
The proportions of undersized fish in the Icelandic longline fishery
Table 3. Percentage of undersized cod and haddock caught with
longline during 2005– 2013 according to depth (m) and distance
from the main coast of Iceland (km).
Cod
Category
Depth (m)
,50
50–80
80–100
100–120
120–180
180–250
.250
Distance (km)
,5
5 –10
10–20
20–30
30–40
40–80
.80
Haddock
n
% <55 cm
n
% <45 cm
13 894
41 097
48 164
50 875
58 390
52 846
31 291
44.7
32.6
24.7
19.9
19.4
12.4
8.1
14 016
34 934
37 044
35 826
41 046
25 732
7310
23.6
19.9
20.4
15.0
11.1
9.5
7.0
20 156
28 411
57 925
47 703
43 775
79 853
23 588
40.6
33.6
25.0
23.7
14.5
15.4
6.4
25 940
29 604
49 524
30 533
18 785
38 812
8885
27.7
18.7
15.7
15.7
11.0
10.3
6.6
longline, and ,10 km for handline and bottom seine (Figure 8).
The distance from shore at which most of the haddock are caught
varies from 10 –50, ,25, and ,10 km for trawl, longline, and
bottom seine, respectively (Figure 8), whereas virtually no
haddock are caught with handline. The average length of cod in
samples from landed catch of trawlers and longliners fishing in
Icelandic waters decreased during 1951–1974, but increased
during 2000–2013 (Table 4). The mean length of cod was higher
in longline than in trawl catches during 1951–1974, but lower
during 1985–2013 (Table 4).
Temporary closures
In recent years, 50 –80% of all temporary closures in Icelandic
waters were due to high percentage of undersized cod and
haddock in the longline fishery (Table 5). During 2005–2013,
most of the temporary closures due to longlining were relatively
close to shore. The number of corner points used to define the
temporary closure areas declined rapidly with distance from shore
(Table 6). Considering all corner points during this period, 77, 19,
and 5% resulted from high percentage of undersized cod,
haddock, and tusk (Brosme brosme), respectively. Nearshore temporary closures were virtually only due to undersized cod and
haddock, but far offshore (.60 km) mainly due to undersized
cod and tusk. Almost half of all temporary closures due to cod
(43%) and more than half of all temporary closures due to
haddock (59%) were within 10 km from shore (Table 6). During
2005–2013, temporary closures due to undersized cod were
usually relatively small, with a mean size of 97.5 km2 (s.d. 133.4
and median 51.4 km2). Thus, an average temporary closure is less
than 0.1% of the main cod fishing area in Icelandic waters,
assumed to be 160 000 km2 (Figure 1a).
Location of the longline fleet
During 2010–2013, 82% of the longliners in the smallest vessel category (,9 m in length), but none of those in the largest vessel category (.20 m), took more than half of their cod catch within 10 km
from shore. However, boats in the smallest vessel category caught
Figure 3. Length distribution of cod captured in the longline fishery in
Icelandic waters during 2005 – 2013 according to distance from shore:
(a) ,5, (b) 5 – 10, (c) 10– 20, (d) 20 –30, (e) 30 –40, (f) 40 –80, and (g)
.80 km. Percentage of fish ,55 cm is indicated. All measurements
were made by observers on-board vessels. Total number of length
measurements was 301 741.
,1% of the total longline catch of cod, while boats in the largest
vessel category caught 53% (Table 7).
Discussion
Ontogenetic changes
Our results show that the proportion of undersized cod and haddock
caught with longline decreases with both depth and distance from the
coast of Iceland. For many demersal species, including cod and
haddock, the spawning and nursery areas are located close to shore,
whereas the feeding grounds for adult fish range over extensive
areas from shallow to deep waters (Jónsson, 1996; Thorsteinsson
et al., 2012). Thus, the average size of many species of fish in
various coastal and oceanic waters has a tendency to increase with
depth during the feeding season (Macpherson and Duarte, 1991;
Collins et al., 2005; Methratta and Link, 2007). Habitat use in
Atlantic cod has been found to shift ontogenetically, with juvenile
cod up to 3 years of age frequently associated with kelp and coarse
846
B. Björnsson et al.
Figure 5. Annual (a) and seasonal (b) changes in percentage of
undersized cod ,55 cm (orange) and haddock ,45 cm (black), as
measured by inspectors in the longline fishery in Icelandic waters
during 2005– 2013 according to distance from shore: ,10 km solid
lines and ≥10 km dotted lines.
Figure 4. Length distribution of haddock captured in the longline
fishery in Icelandic waters during 2005 – 2013 according to distance
from shore: (a) ,5, (b) 5– 10, (c) 10 –20, (d) 20 –30, (e) 30– 40, (f) 40–
80, and (g) .80 km. Percentage of fish ,45 cm is indicated. All
measurements were made by observers on-board vessels. Total number
of length measurements was 202 083.
substrata that offer shelter against predation in shallow water
(Gregory and Anderson, 1997; Gotceitas et al., 1997; Cote et al.,
2001), whereas older fish occur more frequently in deeper water
(Dalley and Anderson, 1997; Cote et al., 2001; Methratta and Link,
2007).
Predatory risk decreases (Lorenzen, 1996) and food demand
increases with fish size. Increased food demand may increase
inter- and intraspecific competition and drive older juveniles to
deeper waters (Macpherson and Duarte, 1991). Furthermore,
higher temperatures associated with shallow waters during the
main growing season are more important for juveniles than for
adult cod, since optimal temperatures for growth decrease with
body weight (Björnsson et al., 2007).
Migratory distances of cod tend to increase with fish size
(Svåsand and Kristiansen, 1990; Pihl and Ulmestrand, 1993;
Björnsson et al., 2011). Tagging studies within marine protected
areas in Iceland suggest that area closures on nursery grounds can
be useful in protecting immature cod, but are of little use in protecting highly migratory adults (Schopka et al., 2010). Therefore, such
measures to reduce the catch of undersized cod in shallow areas
close to shore will increase their chance of growing to a larger size
before they are eventually caught in areas open to fishing.
Longline catches
The percentage of undersized cod and haddock caught with longline,
decreasing both with depth and distance from shore, suggests that
fishing mortality of juveniles of these species can be reduced by permanent closures of nearshore areas. Technically, it would be easier
to implement closures of areas based on distance from shore rather
than depth. Within 10 km from shore, the mean proportion of undersized cod was above the 25% mark used as a reference point for temporary closures. This suggests that a large part of this area should be
considered when implementing long-term fishery closures.
For the whole dataset, the proportion of undersized haddock
exceeded 30% (the mark used as a reference point for temporary
closures for this species) only in restricted areas close to shore. The
age and size structure of the haddock stock was somewhat atypical
during the study period because the fishable haddock biomass was
847
The proportions of undersized fish in the Icelandic longline fishery
Figure 7. Length distributions of cod as measured by official inspectors
on-board (a) trawlers (n ¼ 96 976), (b) longliners (n ¼ 301 741),
(c) handliners (n ¼ 15 584), and (d) bottom seiners (n ¼ 48 569) in
Icelandic waters during 2005 – 2013. The percentage of undersized
(,55 cm) and large cod (≥70 cm) is indicated.
Sources of bias in the data
Figure 6. Relative changes in fishing gear used for (a) cod, (b) haddock,
and (c) wolffish, and (d) the total annual catch of cod (orange), haddock
(black), and wolffish (dotted line) in Icelandic waters during 1977–2012
(in thousands of tonnes). The legend in the stacked figures (Figure 6a,b,c)
is trawl white, longline light grey, gillnet grey, bottom seine dark grey,
handline dim black, and other gear white, respectively (data obtained
from Statistics Iceland, www.hagstofa.is, and Útvegur Fisheries Statistics).
dominated by the 2003 year class, the largest ever recorded in Icelandic
waters (Marine Research Institute, 2014). Initially, this resulted in unusually high and then unusually low proportions of juvenile haddock
during the 9-year study period. It was only in 2007 that the percentage
of undersized haddock caught with longline in nearshore waters
exceeded the 30% mark used for temporary closures, suggesting
that small haddock are not particularly vulnerable to capture with
the conventional type of longline used in Icelandic waters.
The data used in the present analysis are considered reliable because
they were collected at sea by official fishery inspectors. Samples from
landed catches were not used in the analysis because they may be
biased due to discarding of undersized fish (Pálsson et al., 2012).
However, the dents in the length frequencies below the critical
values used for temporary closure indicate ‘codependent behaviour’
by at least some of the inspectors. This effect was not observed in offshore areas where the incidence of undersized cod and haddock was
low and the probability of closure was minimal. When measuring
fish, there is some room for an error in the measurement, depending
on how well the fish is straightened out on the measuring board, how
tight the head is held against the end of the board, etc. In some cases,
fishery inspectors may want the fishers to benefit from the doubt in
their measurements. By smoothing the curve, this error can be
estimated as 2 percentage points in shallow waters. However, the
error may reduce the number of temporary closures more than that.
Another bias in the data, probably of greater importance, is the
effect that the presence of fishery inspectors may have on fishing behaviour (Benoı̂t and Allard, 2009). Due to cost and practical reasons,
848
B. Björnsson et al.
Table 5. Number of temporary closures due to high percentage
of small fish during 2001 – 2013 (according to Starfsskýrsla in
www.fiskistofa.is).
Year
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Trawl
80
61
27
11
36
43
50
12
16
35
12
15
17
Longline
32
34
44
40
44
99
108
75
132
69
44
112
85
Handline
0
2
5
9
2
9
4
0
0
3
13
48
58
Bottom seine
4
1
2
2
7
12
16
3
10
4
0
1
0
Other
8
41
35
11
6
8
2
3
7
3
2
12
5
Total
124
139
113
73
95
171
180
93
165
114
71
188
165
Table 6. Number of corner points of temporary closures in Icelandic
waters as a result of high catch of undersized cod, haddock, and tusk
in the longline fishery, grouped according to distance from shore
during 2005– 2013.
Km
,10
10– 20
20– 30
30– 40
40– 50
50– 60
.60
Total
Cod
1168
751
473
134
96
51
54
2727
Haddock
397
151
74
17
12
13
5
669
Tusk
3
11
24
14
32
18
60
162
All species
1568
913
571
165
140
82
119
3558
Actual number of closure areas was 649 and 156 due to cod and haddock,
respectively.
Figure 8. Average annual catches of cod (orange) and haddock (black)
according to distance from shore (km) for (a) trawlers, (b) longliners,
(c) handliners, and (d) bottom seiners in Icelandic waters during 2005–
2013 showing number of tonnes per year at each whole kilometers
(rounded).
Table 4. Mean length (cm) of cod in samples (n) of landed
catches from trawlers and longliners fishing in Icelandic waters
during 1951– 2013.
Trawl
Period
1951 –1954
1955 –1959
1960 –1964
1965 –1969
1970 –1974
1975 –1979
1980 –1984
1985 –1989
1990 –1994
1995 –1999
2000 –2004
2005 –2009
2010 –2013
Total mean
n
10
8
72
114
92
170
82
63
74
172
338
265
238
131
Longline
Mean
73.0
78.8
75.4
68.8
67.1
68.0
65.0
70.1
69.9
69.4
65.4
68.4
72.6
70.2
s.d.
13.1
10.3
8.6
9.6
7.8
8.0
6.0
11.0
11.2
7.3
7.8
5.6
6.9
8.7
n
208
253
169
155
301
163
109
97
122
137
186
187
165
173
Mean
81.1
82.4
84.1
75.8
70.5
66.1
66.8
62.6
63.5
64.9
65.1
65.8
69.8
70.6
s.d.
10.3
8.3
7.5
8.3
7.9
8.0
7.7
6.1
8.2
7.6
6.0
7.7
10.1
8.0
Difference in
mean length
8.1
3.5
8.7
6.9
3.4
– 1.9
1.8
– 7.5
– 6.4
– 4.6
– 0.4
– 2.6
– 2.9
0.5
Table 7. The relative importance of a 10-km nearshore zone for
longliners fishing in Icelandic waters according to vessel length (VL).
VL (m)
,9
9 –10
10– 12
12– 20
.20
Total
VN
34
60
82
67
32
275
VC (t)
1259
6539
34 834
53 497
108 901
205 029
%
0.6
3.2
17.0
26.1
53.1
100.0
N90
26
12
2
3
0
7
N75
50
23
8
4
0
15
N50
82
50
26
9
0
31
Data are for 2010– 2013 excluding vessels with total cod catch ,10 t. Total
number of vessels (VN), total catch (VC), and percentage of boats in each
vessel length category where .90% (N90), .75% (N75), and .50% (N50) of
the total catch came from the 10-km zone.
surprise boarding to inspect fishing vessels at sea is usually not
carried out in Icelandic waters. Instead, inspectors normally board
the ships before the fishing trips start. Thus, it is likely that when
under surveillance, experienced captains will not select fishing locations where they expect a high percentage of undersized fish.
Therefore, the percentages of undersized fish presented here may
be lower than for the longline fleet in general.
Other gear types
Although handlines, as well as longlines, may catch large proportions of undersized cod, there is a crucial difference between the
two gear types. With handlines, it is possible to stop fishing after
relatively few small fish have been caught, whereas once the longline,
849
The proportions of undersized fish in the Icelandic longline fishery
with thousands of hooks, has been set in an area with large proportions of small fish, the damage has been done. Furthermore, the
manual unhooking of handlined fish makes it possible to release juvenile fish with less injuries than is possible with the mechanical
unhooking of longlined fish, and more than half of the discarded
cod in the handline fishery have been observed to survive (Pálsson
et al., 2003). For the entire dataset, the length distribution was
similar for both handlines and longlines. However, a larger proportion of the handlined than longlined cod was taken within 10 km of
the mainland, and the percentage of undersized cod caught within
this zone was 26% for handlines vs. 36% for longlines. Therefore,
it seems less important to impose permanent closures of coastal
areas for the handline than for the longline fishery.
Compared with longlines, trawls caught a larger proportion of
large cod. Trawlers operate mainly on the outer parts of the continental shelf 60 –90 km off the northwest and southeast coasts of
Iceland, where large cod aggregate, except during the spawning
season. The length distribution of cod caught by trawl (Figure 7a)
was similar as for cod caught with longline .30 km from shore
(Figure 3e –g). For decades during the latter part of the 20th
century, old and large cod were uncommon in Icelandic waters
because of overfishing (Marine Research Institute, 2014), but in
recent years, the average weight of landed cod has been rising
(Table 4). Our finding, that more large cod have been caught with
trawls in recent years than with longlines, is contrary to earlier
studies in Norwegian waters (Løkkeborg and Bjordal, 1992; Huse
et al., 2000).
In 1958, large nearshore areas in Icelandic waters were permanently or seasonally closed to bottom trawling, extending as far as
12 nautical miles (22 km) offshore (Schopka, 2007). At that time,
44% of cod and 59% of haddock caught in Icelandic waters were
taken by foreign trawlers (ICES, 1976) and only a small fraction of
the total catch of these two species was taken with longlines.
During the past 35 years, the relative longline catch of cod and
haddock has tripled (Figure 6a and b), making it more important
now to consider the option of permanent closures of nearshore
areas to longlining.
Permanent closures
Our results indicate that the temporary closure system has been
unable to reduce catches of undersized cod sufficiently in the longline fishery. Normally, the closed areas are quite small and reopen
automatically after 2 weeks, usually without any follow-up inspection. The option of a long-term closure of a larger area is only considered if there are repeated temporary closures in a given area. A
study of temporary closures in the trawl fishery in Iceland showed
that in years when small cod were abundant, the fleet moved from
a temporary closure to a new area containing a similar size structure
of fish. Additional measurements usually resulted in a new temporary closure, etc. The conclusion of the study was that one temporary
closure was insufficient to prevent the capture of small cod by trawlers (Kristinsson et al., 2005).
By permanently closing a nearshore area to longlining, it is
likely that boats using handlines and bottom seines would benefit
to some extent because of an increase in catch per unit effort.
However, only a small part of the nearshore catch would most
likely be taken with handline and bottom seine because the former
gear is mainly restricted to summer and the latter one to limited
fishing areas. Therefore, in the absence of nearshore longlining,
undersized cod would have a greater chance to grow and migrate
to deeper waters in due time.
Permanent nearshore closures would have the greatest implications for small longliners, which take most of their catch close to
shore. On the other hand, large longliners get most of their catch
far from shore. Therefore, due to the ITQ system, permanent nearshore closures might result in a transfer of cod quota from small
longliners to larger fishing vessels, because small longliners are not
suitable for offshore fishing. This would add to the ongoing worldwide development in the longline fishery where large vessels
equipped with automatic baiting machines have gradually been replacing small vessels with manual baiting on land (Valdemarsen,
2001; Eigaard et al., 2011).
Although permanently closing nearshore areas is probably
the most effective way to reduce the catch of small fish in the longline
fishery, other measures have been considered. In a study of the
effects of hook size (O-type no. 10, 11, 12, 13, 14) and bait size
(10 and 30 g) on fish size caught with longlines in Icelandic waters,
there was not a significant effect of hook size on mean length, but a
significantly larger mean length was obtained with the larger bait
for both cod and haddock (Ingólfsson and Einarsson, 2009).
However, enforcing bait size regulation would be problematic especially for boats with automatic baiting machines.
Conclusion
Our results suggest that the permanent closure of large nearshore
areas for longlining is a feasible option to reduce the catch of undersized cod and haddock. It seems that the temporary closure system
aimed at protecting juvenile fish in Icelandic waters has not worked
adequately. The main problem in recent years has been high catches
of undersized cod in the longline fishery, especially in nearshore
waters where longlining is allowed with only limited restrictions,
whereas trawling is prohibited in these important nursery grounds.
The problem has been growing over time, with an increasing proportion of the total catch of cod taken by longline. It seems evident that
the most effective way to alleviate this problem is to permanently close
large nearshore areas to longlining.
It is beyond the scope of this paper to suggest definite closure
areas for longlining. One option would be a permanent closure
within a certain distance from shore, e.g. 5, 10, or 20 km from
shore (permanent meaning at least 1 year). Another option would
be a permanent closure independent of distance from shore in
areas where the capture of undersized cod has been particularly
high in recent years. The third option would be a seasonal closure,
where large areas would be closed during the time of year when
catches of undersized fish are high. The selection of closure areas
may be considered a compromise between (i) the national interest
in reducing catches of juvenile fish and (ii) the interests of the
stakeholders (fishers and vessel owners) honouring the principle
of proportionality.
Acknowledgements
We thank Graham Gaines for critically reading the manuscript,
co-workers at MRI for valuable discussions and the fishery inspectors for collecting data.
References
Benoı̂t, H. P., and Allard, J. 2009. Can the data from at-sea observer
surveys be used to make general inferences about catch composition
and discards? Canadian Journal of Fisheries and Aquatic Sciences,
66: 2025– 2039.
Beverton, R. J. H., and Holt, S. J. 1957. On the dynamics of exploited fish
populations. Her Majesty’s Stationary Office, London. 533 pp.
850
Bjordal, Å. 1989. Recent developments in longline fishing-catching performance and conservation aspects. In Proceedings of the World
Symposium on Fishing Gear and Fishing Vessel Design, St. Johns,
Nfld., Canada, Nov. 21– 24, 1988, pp.19 – 24.
Björnsson, B., Karlsson, H., Thorsteinsson, V., and Solmundsson, J.
2011. Should all fish in mark-recapture experiments be doubletagged? Lessons learned from tagging coastal cod (Gadus morhua).
ICES Journal of Marine Science, 68: 603 – 610.
Björnsson, B., Steinarsson, A., and Árnason, T. 2007. Growth model for
Atlantic cod (Gadus morhua): effects of temperature and body
weight on growth rate. Aquaculture, 271: 216– 226.
Collins, M. A., Bailey, D. M., Ruxton, G. D., and Priede, I. G. 2005.
Trends in body size across an environmental gradient: a differential
response in scavenging and non-scavenging demersal deep-sea fish.
Proceedings of the Royal Society B, 272: 2051– 2057.
Cote, D., Moulton, S., Scruton, D. A., and McKinley, R. S. 2001.
Microhabitat use of juvenile Atlantic cod in a coastal area of
Bonavista Bay, Newfoundland. Transactions of the American
Fisheries Society, 130: 1217 – 1223.
Dalley, E. L., and Anderson, J. T. 1997. Age-dependent distribution of
demersal juvenile Atlantic cod (Gadus morhua) in inshore/offshore
northeast Newfoundland. Canadian Journal of Fisheries and
Aquatic Sciences, 54 (Suppl. 1): 168 –178.
Diekert, F. K., Hjermann, D. Ø., Nævdal, E., and Stenseth, N. C. 2010a.
Non-cooperative exploitation of multi-cohort fisheries 2 the role of
gear selectivity in the North-East Arctic cod fishery. Resource and
Energy Economics, 32: 78– 92.
Diekert, F. K., Hjermann, D. Ø., Nævdal, E., and Stenseth, N. C. 2010b.
Spare the young fish: optimal harvesting policies for North-East
Arctic cod. Environmental and Resource Economics, 47: 455– 475.
Eigaard, O. R., Thomsen, B., Hovgaard, H., Nielsen, A., and Rijnsdorp,
A. D. 2011. Fishing power increases from technological development
in the Faroe Islands longline fishery. Canadian Journal of Fisheries
and Aquatic Sciences, 68: 1970– 1982.
Fenberg, P. B., and Roy, K. 2008. Ecological and evolutionary consequences of size-selective harvesting: how much do we know?
Molecular Ecology, 17: 209– 220.
Gotceitas, V., Fraser, S., and Brown, J. A. 1997. Use of eelgrass beds
(Zostera marina) by juvenile Atlantic cod (Gadus morhua).
Canadian Journal of Fisheries and Aquatic Sciences, 54: 1306– 1319.
Gregory, R. S., and Anderson, J. T. 1997. Substrate selection and use of
protective cover by juvenile Atlantic cod Gadus morhua in inshore
waters of Newfoundland. Marine Ecology Progress Series, 146: 9–20.
Gwinn, D. C., Allen, M. S., Johnston, F. D., Brown, P., Todd, C. R., and
Arlinghaus, R. 2013. Rethinking length-based fisheries regulations:
the value of protecting old and large fish with harvest slots. Fish
and Fisheries, doi:10.1111/faf.12053.
Halliday, R. G. 2002. A comparison of size selection of Atlantic cod
(Gadus morhua) and haddock (Melanogrammus aeglefinus) by
bottom longlines and otter trawls. Fisheries Research, 57: 63 – 73.
Hamley, J. M., and Skud, B. E. 1978. Factors affecting longline catch and
effort: II. Hook-spacing. International Pacific Halibut Commission
Scientific Report, 65: 15 – 24.
Huse, I., Løkkeborg, S., and Soldal, A. V. 2000. Relative selectivity in
trawl, longline and gillnet fisheries for cod and haddock. ICES
Journal of Marine Science, 57: 1271– 1282.
Huse, I., and Soldal, A. V. 2000. An attempt to improve size selection in
pelagic longline fisheries for haddock. Fisheries Research, 48: 43 –54.
ICES. 1976. Report of the North-Western Working Group. ICES
Document CM 1976/F: 6.
Ingólfsson, Ó. A., and Einarsson, H. 2009. Effects of hook size and bait
size on the size selection in longlining. Marine Research Institute,
Reykjavı́k. An unpublished report available on request (20 pp., in
Icelandic).
Jónsson, J. 1996. Tagging of cod (Gadus morhua) in Icelandic waters
1948-1986 and tagging of haddock (Gadus aeglefinus) in Icelandic
B. Björnsson et al.
waters 1953-1965. Journal of Marine Research Institute Reykjavı́k,
14: 1 – 108.
Kristinsson, K., Steinarsson, B. Æ., and Schopka, S. A. 2005. Temporary
closures in the bottom trawl cod fishery West and Northwest of
Iceland. Marine Research in Iceland, nr. 114. 29 pp. (in Icelandic).
Lorenzen, K. 1996. The relationship between body weight and natural
mortality in juvenile and adult fish: a comparison of natural ecosystems and aquaculture. Journal of Fish Biology, 49: 627– 647.
Løkkeborg, S., and Bjordal, Å. 1992. Species and size selectivity in longline fishing: a review. Fisheries Research, 13: 311– 322.
Løkkeborg, S., and Bjordal, Å. 1995. Size-selective effects of increasing
bait size by using an inedible body on longline hooks. Fisheries
Research, 24: 273– 279.
Macpherson, E., and Duarte, C. M. 1991. Bathymetric trends in demersal fish size: is there a general relationship? Marine Ecology Progress
Series, 71: 103 – 112.
Marine Research Institute. 2014. State of marine stocks in Icelandic
waters 2013/2014 – prospects for the quota year 2014/2015.
Marine Research in Iceland, 176. 188 pp. (in Icelandic).
Methratta, E. T., and Link, J. S. 2007. Ontogenetic variation in habitat
association for four groundfish species in the Gulf of Maine–
Georges bank region. Marine Ecology Progress Series, 338: 169–181.
Pálsson, Ó. K., Björnsson, H., Gı́sladóttir, H., Jóhannesson, G., and
Ottesen, Th. 2012. Discards of cod and haddock in demersal
Icelandic fisheries 2001-2010. Marine Research in Iceland, 160:
5 – 14 (in Icelandic).
Pálsson, Ó. K., Einarsson, H. E., and Björnsson, H. 2003. Survival
experiments of undersized cod in a hand-line fishery at Iceland.
Fisheries Research, 61: 73 – 86.
Pihl, L., and Ulmestrand, M. 1993. Migration pattern of juvenile cod
(Gadus morhua) on the Swedish west coast. ICES Journal of
Marine Science, 50: 63 – 70.
R Core Team. 2013. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
http://www.R-project.org/.
Ricker, W. E. 1945. A method of estimating minimum size limits for
obtaining maximum yield. Copeia, (2): 84– 94.
Ricker, W. E. 1975. Computation and interpretation of biological
statistics of fish populations. Bulletin of the Fisheries Research Board
of Canada, 191: 382 pp.
Rotabakk, B. T., Skipnes, D., Akse, L., and Birkeland, S. 2011. Quality
assessment of Atlantic cod (Gadus morhua) caught by longlining
and trawling at the same time and location. Fisheries Research,
112: 44– 51.
Schopka, S. A. 2007. Area closures in Icelandic waters and the real-time
closure system. A historical review. Marine Research in Iceland, 133.
86 pp. (in Icelandic).
Schopka, S. A., Solmundsson, J., Ragnarsson, S. A., and Thorsteinsson,
V. 2010. Using tagging experiments to evaluate the potential of
closed areas in protecting migratory Atlantic cod (Gadus morhua).
ICES Journal of Marine Science, 67: 1024– 1035.
Stephen, J. A., Harris, P. J., and Reichert, M. J. M. 2011. Comparison of
life history parameters for landed and discarded fish captured off
the southeastern United States. Fishery Bulletin US, 109: 292 – 304.
Svåsand, T., and Kristiansen, T. S. 1990. Enhancement studies of coastal
cod in western Norway. Part II. Migration of reared cod. Journal du
Conseil International pour l’Exploration de la Mer, 47: 13 – 22.
Thorsteinsson, G. 1980. Fishing and fishing gear. Almenna Bókafélagid,
Reykjavı́k. 186 pp. (in Icelandic).
Thorsteinsson, V., Pálsson, Ó. K., Tómasson, G. G., Jónsdóttir, I. G., and
Pampoulie, C. 2012. Consistency in the behaviour types of the
Atlantic cod: repeatability, timing of migration and geolocation.
Marine Ecology Progress Series, 462: 251– 260.
Valdemarsen, J. W. 2001. Technological trends in capture fisheries.
Ocean and Coastal Management, 44: 635– 651.
Handling editor: Emory Anderson