ICES Journal of Marine Science, 63: 52e67 (2006)
doi:10.1016/j.icesjms.2005.08.011
Selectivity of diamond- and square-mesh codends in the
deepwater crustacean trawl fishery off the Balearic Islands
(western Mediterranean)
Beatriz Guijarro and Enric Massutı́
Guijarro, B., and Massutı́, E. 2006. Selectivity of diamond- and square-mesh codends in the
deepwater crustacean trawl fishery off the Balearic Islands (western Mediterranean). e
ICES Journal of Marine Science, 63: 52e67.
An analysis of 38 bottom trawl hauls at depths of 251e737 m off the Balearic Islands (western Mediterranean) during autumn 2002 and spring 2003 was used to compare the species
and size selectivity of 40-mm diamond- and square-mesh codends under commercial conditions. There was no difference in the catch composition or the yield that could be attributable to mesh shape, although the percentage of total and commercial species discarded with
a diamond mesh was higher than with a square mesh. At least in the short term, the escapement ratio and the economic loss with the diamond mesh were lower than with square mesh,
but economic efficiency was no different between them. For all the main species compared,
except one flatfish, size selectivity parameters were lower for the diamond- than for the
square-mesh codend. Selectivity values for the one flatfish species were similar. From
the results it is concluded that, within the context of precautionary management, introduction of a 40-mm square mesh in the codend could be an appropriate and plausible measure
to improve the state of the resources exploited by the deepwater crustacean trawl fishery of
the upper slope off the Balearic Islands, and to reduce the impact of the fishery on the
ecosystem.
Ó 2005 International Council for the Exploration of the Sea. Published by Elsevier Ltd. All rights reserved.
Keywords: Balearic Islands, bottom trawl, codend selectivity, deep water, discards,
diamond and square meshes.
Received 1 April 2005; accepted 25 August 2005.
B. Guijarro and E. Massutı́: IEO, Centre Oceanogràfic de les Balears, PO Box 291, 07080
Palma de Mallorca, Spain. Correspondence to B. Guijarro: tel: C34 971401561; fax: C34
971404945; e-mail: [email protected].
Introduction
Knowledge of gear selectivity is crucial to good fisheries
management. Its improvement contributes to minimizing
the capture of juveniles by regulating the size at first capture, increasing the yield per recruit of targeted species,
and reducing the discards and hence the impact of fishing
on ecosystems (Armstrong et al., 1990; MacLennan,
1992), some of the principles implicitly enshrined in the
Code of Conduct for Responsible Fisheries (Garcia,
2000). Recently, the General Fisheries Commission for
the Mediterranean (GFCM) has stressed the incongruence
between the minimum legal size for hake (Merluccius merluccius; 20 cm total length, TL) and Norway lobster (Nephrops norvegicus; 20 mm carapace length, CL), established
by European legislation, and the estimated length of first
capture (12 cm and 16 mm, respectively), given the legal
1054-3139/$32.00
minimum 40-mm diamond mesh in force for trawling in
the Mediterranean (GFCM, 2000), and has encouraged
studies aimed at improving the selectivity of trawls, in trying to eliminate these contradictions and reducing discards
(GFCM, 2001).
Size and shape of the mesh in the codend have been demonstrated as the main factors influencing the selectivity of
trawl catches (e.g. Robertson and Stewart, 1988; Reeves
et al., 1992). Diamond-shaped mesh in trawlnets stretches
under tension during the haul and has a tendency to close
when the codend fills, thus reducing its effective selectivity
compared with square mesh, which remains open during
a tow (Robertson and Stewart, 1988). For this reason,
many studies on square mesh have been carried out recently
(e.g. MacLennan, 1992; Campos et al., 2002).
Decapod crustaceans, specifically N. norvegicus and
Aristeus antennatus (red shrimp), are the prime target of
Ó 2005 International Council for the Exploration of the Sea. Published by Elsevier Ltd. All rights reserved.
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
the deepwater bottom trawl fishery carried out in the western Mediterranean (Sardà, 1998; GFCM, 2004). The fishery
is well developed on the upper slope off the Balearic Islands
(Merella et al., 1998; Carbonell et al., 1999; Garcı́a-Rodrı́guez and Esteban, 1999), and landings in Mallorca, the biggest island in the Archipelago, from which about half the
trawl fleet operates, are estimated at 10e20 and
100e200 t, respectively, 4e8% of the total landings and
30% of the total revenue. Stock assessments of these two
and other target species of the Mediterranean demersal
trawl fisheries (e.g. M. merluccius) suggest overfishing
(Sardà, 1998; Garcı́a-Rodrı́guez and Esteban, 1999;
GFCM, 2004).
Catch composition of the deepwater crustacean fisheries
off the Balearic Islands have been analysed by Moranta
et al. (2000), revealing that discards, mainly fish (73%)
and crustaceans (16%), represent 42% of the total catch.
However, some differences are evident by depth. In the
depths at which N. norvegicus is targeted, fish constitute
60e90% of the catch and 70e80% of the discards. The fish
discarded are mainly undersize commercial species such
as Helicolenus dactylopterus, Micromesistius poutassou,
Phycis blennoides, and M. merluccius. In the depths at which
A. antennatus is mainly caught, landings comprise both crustaceans (60%) and fish (40%), although fish (mostly unmarketable species) constitute 70% of the discards.
In the Mediterranean, studies on trawl selectivity using
square mesh in the codend have been undertaken mainly
in the eastern basin, and have focused on catch
composition and discarding (Stergiou et al., 1997b), as
well as selectivity parameters of target species (Petrakis
and Stergiou, 1997; Stergiou et al., 1997a). In the western
basin, a few studies have compared the effects of
diamond mesh of different sizes in the codend, for both
fish and crustaceans (Sardà et al., 1993; Ragonese et al.,
2001, 2002), but to our knowledge, only two studies have
assessed the influence of square mesh on target fish on
the shelf (Mallol et al., 2001; Sardà et al., 2004).
The current study was initiated to compare, under commercial fishing conditions in the trawl fishery for crustaceans off Balearics, catch composition, commercial yields,
retention efficiency, discards, and size selectivity parameters, using the 40-mm ‘‘traditional’’ diamond-mesh codend
mentioned in European legislation, and an ‘‘experimental’’
square-mesh codend of similar mesh size. The main objective is to analyse the effect of the introduction of such
a codend as a possible management measure to improve
the state of the resources and to reduce the impact of the
fishery on the ecosystem.
Material and methods
Sampling was conducted on the main fishing grounds on
the continental slope south of Mallorca during September
and October 2002 (autumn) and May and June 2003
53
(spring), on board the commercial bottom trawler FV ‘‘Moralti Nou’’ (length 22 m; 59 grt; nominal engine power
365 hp), which traditionally operates in the area. A conventional ‘‘huelvano’’-type trawl was used (Figure 1). Two
codends of 40 mm nominal mesh size but different mesh
shapes were used, employing the covered codend method.
The cover was a net of 20-mm diamond mesh, attached
directly to the funnel end of the net. In order to maintain
a good flow of water and to avoid masking the codend
meshes, the cover was 1.5 m wider and longer than the
two codends. This method has been considered appropriate
where catches are not very large (Wileman et al., 1996),
and it has been used in most of the trawl selectivity studies
in the Mediterranean.
In all, 38 hauls were carried out in daylight (18 in
autumn and 20 in spring) between 251 and 737 m,
following commercial fishing procedures, but trying to
make the same number of trawls with each shape of
mesh and at similar depths (Table 1). The average duration
and speed of trawls were 4.5 h and 2.5 knots, respectively.
Each codend was used on the same gear and changed
weekly. After each haul, catches in the codend and the
covernet were sorted by taxonomic and commercial
(landings and discards) categories, counted and weighed
separately. TL and CL of fish and crustaceans, respectively,
were measured.
The PRIMER package was used to analyse the standardized biomass matrix of species by trawl (kg h1 retained in
the codend). After square-root transformation, cluster analysis was applied to assess the different fishing strategies,
choosing BrayeCurtis as the similarity index and UPGMA
to link samples. Species recorded in fewer than 5% of the
samples were omitted from the analysis. A similarity percentage analysis (SIMPER) was also applied to estimate
the dissimilarity between groups and the contribution of
main species to each.
Two-way analysis of variance (ANOVA) was applied to
test differences between mesh shape and season in the commercial yields (kg h1 retained in the codend) of those species whose cumulative contribution, by SIMPER analysis,
was O90%. It was also used to analyse the total catch,
the catch of commercial species, total discards, and the discards of commercial species
retained in the codend. Data
pffiffiffi
were transformed to x or log(x C 1) as necessary and
checked for normality, and Cochran’s test was applied
to test the homogeneity of variance. In some cases, transformation did not produce homogeneous variance, but
ANOVA is considered a robust analysis when sample sizes
are equal (Zar, 1996).
To assess economic performance in the short term of
both mesh shapes, the escapement ratio (ER: proportion
of the catch escaping, as kg h1, in relation to the total
catch), the economic loss (EL: proportion of the value
of fish and shellfish escaping, as V h1, in relation to
the total value), and the economic efficiency (EE:
V kg1of the retained catch in relation to the total weight
54
B. Guijarro and E. Massutı́
Figure 1. Schematic diagram of the commercial ‘‘huelvano’’-type trawl gear used (PE, polyethylene; PA, polyamide; PP, polypropylene; H, hemp; >, diamond mesh; ,, square mesh;
ø, diameter).
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
55
Table 1. Main characteristics of the commercial hauls analysed by mesh shape. The mean depth and the speed are the average of that
registered every 15 min during trawling, and the duration is the effective fishing time of the net on the bottom, measured using
a SCANMAR system.
Haul
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Date
Mesh
shape
24/09/2002
25/09/2002
26/09/2002
27/09/2002
30/09/2002
01/10/2002
02/10/2002
03/10/2002
04/10/2002
07/10/2002
08/10/2002
09/10/2002
14/10/2002
15/10/2002
16/10/2002
17/10/2002
18/10/2002
21/10/2002
12/05/2003
13/05/2003
14/05/2003
15/05/2003
16/05/2003
19/05/2003
20/05/2003
21/05/2003
22/05/2003
23/05/2003
26/05/2003
27/05/2003
28/05/2003
29/05/2003
30/05/2003
02/06/2003
03/06/2003
04/06/2003
05/06/2003
06/06/2003
Diamond
Diamond
Diamond
Diamond
Square
Square
Square
Square
Square
Diamond
Diamond
Diamond
Diamond
Diamond
Square
Square
Square
Square
Diamond
Diamond
Diamond
Diamond
Diamond
Square
Square
Square
Square
Square
Diamond
Diamond
Diamond
Diamond
Diamond
Square
Square
Square
Square
Square
Initial
latitude (N)
Initial
longitude (E)
39(09.94
39(07.50
39(03.00
39(03.00
39(09.27
39(07.20
39(02.60
39(09.97
39(09.87
39(09.32
39(08.97
39(03.85
39(02.69
39(02.61
39(02.54
39(02.44
39(02.50
39(07.29
39(06.50
39(07.48
39(17.32
39(01.52
39(02.76
39(03.04
39(04.16
39(03.56
39(02.80
39(02.85
39(02.90
39(02.25
39(04.87
39(14.80
39(02.70
39(04.11
39(06.34
39(04.10
39(03.00
39(03.00
002(38.57
002(45.22
002(38.24
002(38.25
002(39.11
002(38.83
002(38.65
002(35.30
002(38.79
002(38.55
002(38.65
002(39.52
002(38.43
002(38.72
002(38.07
002(58.25
002(38.09
002(38.83
002(40.45
002(42.67
002(09.91
002(38.29
002(36.76
002(39.12
002(39.09
002(39.82
002(38.28
002(38.14
002(39.55
002(40.38
002(40.56
002(25.40
002(38.34
002(40.56
002(38.25
002(40.15
002(38.24
002(38.24
captured) were estimated by depth range, season, and
mesh shape. To test differences between mesh shape and
season, two-way ANOVA was applied. To transform the
proportional data (ER and EL) to a normal distribution,
the procedure was:
rffiffiffiffiffiffiffiffi
180
X
!arcsin
X# Z
p
100
ð1Þ
Homogeneity of the variance was tested with Cochran’s
test. Gear saturation for both mesh shapes was analysed
Mean
depth (m)
Course
(deg)
Speed
(knots)
Duration
(min)
417
251
694
644
410
430
696
427
431
431
450
463
710
707
726
717
716
433
423
283
399
737
702
475
473
465
717
726
467
471
463
687
721
477
483
469
715
719
307
208
165
168
303
305
176
306
304
291
305
314
177
185
180
182
178
312
312
185
98
180
179
314
340
318
177
178
298
331
315
181
176
308
314
314
177
178
2.7
2.7
2.5
2.5
2.5
2.7
2.4
2.7
2.7
2.7
2.6
2.6
2.5
2.5
2.4
2.5
2.5
2.7
2.6
2.5
2.6
2.5
2.5
2.5
2.6
2.5
2.5
2.5
2.7
2.7
2.6
2.4
2.5
2.5
2.5
2.5
2.5
2.4
262
262
285
280
180
285
196
275
310
293
293
277
285
279
260
260
185
285
289
255
265
309
303
267
255
267
301
303
270
265
284
296
313
180
270
267
311
310
by testing, through simple linear regression, the independence between the transformed ER and the retained catch.
For species in which ANOVA was applied and specimens
were present in the covernet, length frequency distributions
in the codend and the cover were calculated by haul, season,
and mesh shape. Size selectivity was modelled using the
generalized logistic curve
eðS1 CS2 !LÞ
SL Z
1CeðS1 CS2 !LÞ
ð2Þ
where SL is the retained proportion of length class L, and S1
and S2 are the parameters to estimate. Such a manner of
56
B. Guijarro and E. Massutı́
Table 2. Species and commercial categories captured during the two trawl selectivity surveys off Mallorca, during autumn 2002 and spring
2003, with their market code (high (H), OV6 kg1; medium (M), V3e6 kg1; low (L), !V3 kg1; discards (D), non-commercial species),
average price (V kg1), frequency of occurrence (F; %), proportion by weight (W; %), and economic value (V; %) by season and for both
seasons combined. The absolute numbers of analysed trawls and total catch, by weight (W; kg) and economic value (V; V), are also shown.
Autumn
Spring
Market
code
Price
(V kg1)
%F
Teleosts
Helicolenus dactylopterus
Merluccius merluccius
Micromesistius poutassou
Lepidorhombus spp.*
Lophius spp.y
Phycis blennoides
Mixed teleostsz
Mixed macruridsx
L
H
M
H
H
L
D,L,M,H
D
1.6
6.6
3.2
6.7
6.5
2.0
3.6
0.0
100
83
100
94
100
89
100
100
100
51.0
2.7
10.4
6.7
4.1
6.5
9.7
9.8
1.1
32.9
0.1
14.7
2.8
4.0
6.6
2.5
2.3
0.0
100
60
100
80
65
80
100
100
95
Elasmobranchs
Galeus melastomus
Mixed raysk
Mixed sharks{
L
L
L,M
0.6
1.7
1.7
100
100
100
100
16.4
9.8
3.6
3.1
0.8
0.0
0.7
0.1
Crustaceans
Aristaeomorpha foliacea
Aristeus antennatus
Geryon longipes
Nephrops norvegicus
Parapenaeus longirostris
Plesionika edwardsii
Plesionika martia
Mixed crustaceans**
H
H
L
H
H
H
M
D,L
27.7
25.8
2.5
25.5
11.9
7.7
3.2
0.6
100
0
44
50
72
56
33
44
100
29.1
0.0
7.8
2.1
4.2
6.9
3.3
2.5
2.3
H
D,L
D,L
11.2
0.7
1.6
94
56
78
100
Othersxx
Total
Taxon
Cephalopods
Loligo vulgaris
Mixed octopusesyy
Mixed squid
and cuttlefishzz
%W
%V
%F
%W
Both seasons
%V
%F
%W
%V
53.4
1.4
4.7
20.9
3.5
3.7
4.6
13.2
1.3
31.5
0.2
6.5
14.1
3.5
2.9
1.8
2.3
0.0
100
71
100
87
82
84
100
100
97
52.4
1.9
6.9
15.3
3.7
4.8
6.6
11.8
1.2
32.1
0.2
10.0
9.3
3.7
4.5
2.1
2.3
0.0
100
95
50
90
12.2
8.4
0.6
3.2
0.4
0.0
0.2
0.2
100
97
74
95
13.8
8.9
1.8
3.2
0.6
0.0
0.4
0.2
63.5
0.0
35.3
1.2
8.2
12.7
3.3
2.8
0.0
100
25
40
40
75
60
15
40
100
26.6
0.1
7.6
1.4
2.8
9.4
0.9
1.0
3.4
65.3
0.2
30.0
1.0
10.0
21.4
1.3
1.4
0.0
100
13
42
45
74
58
24
42
100
27.6
0.0
7.7
1.7
3.4
8.4
1.9
1.6
3.0
64.5
0.1
32.3
1.1
9.2
17.7
2.1
2.0
0.0
3.0
1.4
0.5
1.0
2.7
2.2
0.0
0.5
100
25
65
100
7.5
0.3
3.2
4.0
2.8
1.0
0.6
1.2
100
39
71
100
5.7
0.7
2.2
2.8
2.8
1.6
0.3
0.9
55
0.6
0.0
60
0.4
0.0
79
0.5
0.0
T
18
W
2 385
V
15 603
T
20
W
3 645
V
21 263
T
38
W
V
6 029 3 6865
*Lepidorhombus boscii and Lepidorhombus whiffiagonis.
yLophius budegassa and Lophius piscatorius.
zAnthias anthias, Argentina sphyraena, Argyropelecus hemigymnus, Arnoglossus rueppelli, Bathysolea profundicola, Boops boops, Capros aper, Cetrolophus niger, Chauliodus sloani, Chelidonichthys cuculus, Chlorophthalmus agassizi, Citharus linguatula, Conger conger,
Epigonus telescopus, Epigonus denticulatus, Gadiculus argenteus, Hoplostethus mediterraneus, Lampanyctus crocodiles, Lepidion lepidion, Lepidopus caudatus, Lepidotrigla cavillone, Molva dypterigia macrophtalma, Mora moro, Mullus surmuletus, Myctophidae, Nettastoma melanurum, Notacanthus bonapartei, Notolepis rissoi, Pagellus acarne, Pagellus bogaraveo, Peristedion cataphractum, Phycis
phycis, Pontinus kuhlii, Scorpaena elongata, Scorpaena notata, Scorpaena scrofa, Serranus cabrilla, Serranus hepatus, Stomias boa, Symphurus nigrescens, Symphurus ligulatus, Synchiropus phaeton, Trachinus draco, Trachurus mediterraneus, Trachurus trachurus, Trigla
lyra, Uranoscopus scaber, and Zeus faber.
xCaelorinchus caelorinchus, Hymenocephalus italicus, and Nezumia aequalis.
kDipturus oxyrinchus, Leucoraja naevus, Raja clavata, and Raja polystigma.
{Dalatias licha, Etmopterus spinax, Scyliorhinus canicula, and Squalus acanthias.
**Macropipus tuberculatus, Munida spp., Palinurus mauritanicus, Paromola cuvieri, Pasiphaea sivado, Pasiphaea multidentata, Plesionika giglioli, Plesionika heterocarpus, Plesionika antigay, Polycheles typhlops, and Sergestes arcticus.
yyBathypolypus sponsalis, Eledone cirrhosa, Eledone moschata, Octopus salutii, and Scaergus unicirrhus.
zzHistioteuthis bonnellii, Histioteuthis reversa, Illex coindetti, Sepia orbignyana, Sepietta oweniana, and Todarodes sagitattus.
xxEchinidae, Gryphus vitreus, Porifera, and Salpidae.
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
non-linear adjustment is one of the methods most recommended (Wileman et al., 1996). The selection parameters
L25, L50, and L75, or lengths at which 25, 50, and 75%, respectively, of the specimens are retained in the codend,
were calculated from:
S1 lnð3Þ
S1
S1 Clnð3Þ
; L50 Z
; L75 Z
L25 Z
S2
S2
S2
ð3Þ
Adjustment was with the SYSTAT package, following
a maximum likelihood procedure, and using the iteratively
re-weighted least-squares method and r2 as the index of
adjustment.
This logistic model was applied haul by haul. A mean
selectivity curve, taking into account between-haul variability, was estimated using ECWeb (ConStat), a software that
follows the methodology proposed by Fryer (1991). As in
some hauls the number of retained and escaped fish was
insufficient for estimating the selection curve, the logistic
model was also applied on the basis of pooled data from
all hauls.
Results
In total, the catch was 6029 kg with an economic value of
V3685 (Table 2). Teleosts were the most important group
in terms of weight (55%) and second in economic value
(32%); crustaceans were the most important group by value
(64%) and the second by weight (28%). Elasmobranchs and
57
cephalopods were less important, constituting 14% and 6%
by weight, respectively, and 1e3% by economic value.
Only two species appeared in all hauls (M. merluccius
and P. blennoides), both constituting O5% of the total
weight, but the most important species were M. poutassou,
Galeus melastomus, A. antennatus, and Parapenaeus longirostris, which constituted 15%, 9%, 7%, and 8% by
weight, respectively. The biggest contributions in terms
of economic value were A. antennatus (32%), followed
by P. longirostris (18%), and M. merluccius (10%). Although N. norvegicus comprised just 3% by weight, its importance in economic terms was great (9%).
In the similarity dendrogram (Figure 2), a first cluster
separated trawls in 399e483 m (SL1) from those in
644e737 m (SL2), with two hauls (!300 m) not taken
into account for further analysis. A second cluster
separated trawls made during spring (SL1-S and SL2-S)
from those made during autumn (SL1-A and SL2-A). There
was no association attributable to differences in mesh
shape. The results of SIMPER confirmed these clusterings
of trawls (Table 3). The greatest dissimilarity was when
depth intervals were compared, because they showed a different species composition: M. merluccius, P. longirostris,
M. poutassou, P. blennoides, N. norvegicus, and Lepidorhombus boscii were the dominant species in SL1, whereas
SL2 was dominated by A. antennatus, G. melastomus,
P. blennoides, Geryon longipes, and Plesionika martia.
Seasonal differences were less noticeable, a result attributable to distinct contributions by some of these species.
20
Similarity
40
60
SL2
SL2-S
SL1
SL2-A
SL1-S
SL1-A
100
SQ719S
DI737S
DI721S
SQ717S
SQ726S
DI702S
SQ715S
DI687S
SQ696A
DI694A
DI644A
DI710A
DI707A
SQ716A
SQ726A
SQ717A
DI251A
DI283S
SQ427A
SQ410A
SQ430A
DI417A
DI431A
DI450A
DI463A
SQ431A
SQ433A
DI471S
DI467S
DI463S
SQ469S
SQ465S
SQ483S
SQ477S
SQ475S
SQ473S
DI423S
DI399S
80
Figure 2. Dendrogram of trawls made during the two selectivity surveys. Mesh shape (DI, diamond; SQ, square), mean depth (in m), and
season (A, autumn; S, spring) are shown for each trawl.
58
B. Guijarro and E. Massutı́
Table 3. SIMPER results for each trawl group identified from the
dendrogram and for the species that contributed to at least 90%
of the differences between these groups: mean yields (Y as
kg h1 retained in the codend); average similarity ðSi Þ; s.d., standard deviation; percentage contribution to the similarity ð%Si Þ;
average dissimilarity ðdi Þ.
Species
Y
Si
Si =s:d: %Si
SL1-A; Si Z62:50
Merluccius merluccius
7.7 16.25 2.89 26.01
Parapenaeus longirostris 4.51 10.62 2.03 16.99
Phycis blennoides
3.24 8.49 3.59 13.58
Nephrops norvegicus
2.68 6.98 2.98 11.16
Lepidorhombus boscii
1.73 3.88 2.34 6.2
Micromesistius poutassou 1.83 3.52 1.73 5.63
Helicolenus dactylopterus 1.41 2.8 1.95 4.48
Lophius piscatorius
1.11 2.03 1.42 3.25
Loligo vulgaris
0.76 1.4 2.54 2.24
Lophius budegassa
0.56 1.13 1.41 1.82
SL1-S; Si Z64:41
Micromesistius poutassou 17.81 19.09 1.68 29.64
Parapenaeus longirostris 8.22 12.98 3.93 20.15
Merluccius merluccius
3.48 5.13 1.53 7.97
Lepidorhombus boscii
2.4
3.91 4.37 6.07
Phycis blennoides
2.49 3.81 3.19 5.91
Eledone cirrhosa
2.15 3.42 3.04 5.32
Nephrops norvegicus
2.35 3.36 1.75 5.22
Illex coindetii
1.41 2.37 2.88 3.68
Helicolenus dactylopterus 1.04 1.59 3.58 2.46
Galeus melastomus
0.84 1.07 1.89 1.66
Lophius piscatorius
1.12 1.05 1.37 1.63
Todarodes sagittatus
1.06 1.01 0.99 1.57
SL2-A; Si Z65:89
Aristeus antennatus
6.05 18.57 3.4 28.19
Galeus melastomus
7.67 15.1 2.17 22.91
Phycis blennoides
2.68 7.03 4.19 10.67
Micromesistius poutassou 2.53 5.77 1.76 8.76
Geryon longipes
1.74 5.32 2.92 8.08
Plesionika martia
2.44 4.93 1.32 7.48
Lophius piscatorius
2.6
4.09 0.83 6.21
SL2-S; Si Z65:17
Aristeus antennatus
7.58 28.59 2.89 43.88
Galeus melastomus
6.26 18.26 3.1 28.01
Geryon longipes
1.24 4.35 3.96 6.67
Phycis blennoides
1.51 3.14 4.75 4.81
Plesionika martia
0.82 2.42 5.73 3.71
Merluccius merluccius
0.51 1.67 2.21 2.56
Etmopterus spinax
0.61 1.37 1.46 2.1
Pairwise comparisons
SL1-A vs. SL2-A
SL1-A vs. SL1-S
SL1-S vs. SL2-S
SL2-A vs. SL2-S
X
Si (%)
26.01
42.99
56.57
67.74
73.94
79.57
84.05
87.29
89.53
91.35
29.64
49.79
57.76
63.83
69.74
75.06
80.28
83.96
86.42
88.09
89.72
91.29
28.19
51.1
61.77
70.53
78.61
86.09
92.3
43.88
71.89
78.56
83.37
87.09
89.65
91.75
di
78.53
53.21
89.89
41.77
Comparing yields of the main species and different catch
categories revealed that most differences were seasonal
(Table 4). Total catch, catch of commercial species, total
discards, and yields of A. antennatus, P. longirostris, and
M. poutassou obtained in SL1 during spring were higher
than in autumn, whereas yields of P. martia and G. longipes
in SL2 and of H. dactylopterus, M. merluccius, and P. blennoides in SL1 during autumn were higher than in spring.
The only differences related to mesh shape were observed
in SL2, where total discards and the discards of commercial
species were higher in the diamond mesh than in the square
mesh. The interaction of both factors (season and mesh
shape) was significant only for the catch of commercial
species and for H. dactylopterus and M. poutassou in SL1.
Catch composition by commercial and taxonomic categories (Figure 3) revealed that the percentage of catch discarded from the diamond-mesh net (18e45%) was greater
than from the square-mesh net (6e18%). Such a reduction
with the square mesh was also observed in the results for
commercial species discarded (7e17% with the diamond
mesh, 2e7% with the square mesh). Elasmobranchs and
teleosts were the dominant commercial species discarded
(31e91% and 6e33%, respectively), followed by crustaceans (7e43%) and cephalopods (%2%).
A clear increase in the escapement ratio from diamond to
square mesh was observed (Figure 4). This is reflected in
the economic loss, which is significantly higher with the
square than with the diamond mesh. By contrast, there
were no differences in economic efficiency between mesh
shapes, with season as the only significant factor. With
both factors combined (mesh and season), there was no significant difference. No saturation was detected, because the
relationships between the escapement ratio and the retained
catch did not fit a linear regression: p Z 0.322 for diamond
mesh, p Z 0.624 for square mesh.
The lengths of the retained and escaped animals by mesh
shape and season are summarized for the main species in
Table 5. Their selectivity parameters and curves, calculated
by mesh shape, season, and both seasons combined, taking
into account between-haul variability and pooled data, are
shown in Tables 6 and 7, and Figures 5 and 6. For those
species where it was possible to estimate size selectivity
with both methods, values were similar. In all species, there
was an increase in the length at first capture from diamond to
square mesh, the only exception being for L. boscii, for which
the length at first capture was similar for both mesh shapes.
Discussion
In this work we have compared the selectivity of the ‘‘traditional’’ diamond and an ‘‘experimental’’ square mesh in the
codend, under commercial conditions, in the deepwater
crustacean trawl fishery off the Balearic Islands. In contrast
to other studies comparing both meshes in the Mediterranean (Stergiou et al., 1997a, b; Petrakis and Stergiou,
1997), we took into account not only the size selectivity
parameters for the main species but also the selectivity in
relation to catch composition, commercial yield, and discards. These are important aspects in fisheries management.
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
59
Table 4. Yields (kg h1 retained in the codend, Gs.e.) for the main species, the total catch, the capture of commercial species, the total
discards, and the discards of commercial species within each trawl group identified from the dendrogram and two-way ANOVA, showing
the significance (ns, not significant; *, p ! 0.05; **, p ! 0.01; ***, p ! 0.001) of the factors season (S), mesh shape (M), and their interaction (S ! M).
SL1
SL2
Autumn
Species
A. antennatus
G. longipes
N. norvegicus
P. longirostris
P. martia
G. melastomus
H. dactylopterus
L. boscii
L. piscatorius
M. merluccius
M. poutassou
P. blennoides
Diamond
Square
e
e
2.4 G 0.5
5.3 G 1.2
e
e
2.2 G 0.3
1.9 G 0.4
0.9 G 0.2
8.6 G 1.3
2.4 G 0.6
4.1 G 0.3
e
e
2.9 G 0.4
3.2 G 0.8
e
e
0.7 G 0.1
1.6 G 0.4
1.2 G 0.5
7.0 G 2.3
1.2 G 0.4
2.4 G 0.3
Spring
Diamond
Autumn
Square
e
e
e
e
1.8 G 0.6 2.8 G 0.1
6.1 G 0.7 9.0 G 0.4
e
e
e
e
0.5 G 0.1 1.0 G 0.1
1.6 G 0.3 2.0 G 0.3
1.6 G 0.8 0.7 G 0.3
4.3 G 0.2 2.8 G 0.8
9.8 G 3.5 24.1 G 5.3
1.9 G 0.2 2.8 G 0.4
Spring
Diamond
Square
Diamond
Square
5.4 G 1.0
1.4 G 0.2
e
e
3.1 G 0.4
9.6 G 3.3
e
e
e
e
e
3.3 G 1.1
6.7 G 0.7
2.1 G 0.2
e
e
2.3 G 0.8
5.8 G 2.2
e
e
e
e
e
2.0 G 0.1
8.4 G 0.5
0.9 G 0.1
e
e
0.5 G 0.1
2.2 G 0.8
e
e
e
e
e
1.7 G 1.2
8.0 G 0.9
1.3 G 0.1
e
e
0.7 G 0.1
5.8 G 1.8
e
e
e
e
e
0.9 G 0.2
Total catch
34.2 G 3.5 29.0 G 4.5 43.4 G 4.5 59.3 G 6.5 30.4 G 5.3 26.2 G 5.0 19.1 G 0.5 24.8 G 4.0
Catch of commercial species
30.9 G 2.9 26.1 G 4.0 36.1 G 4.9 53.9 G 5.8 24.7 G 3.7 25.0 G 4.9 17.0 G 1.7 21.9 G 3.6
Total discards
2.3 G 0.8 1.9 G 0.7 4.7 G 1.5 3.4 G 0.6 6.1 G 1.9 1.8 G 0.8 5.7 G 0.7 4.7 G 0.8
Discards of commercial species 2.1 G 0.8 1.9 G 0.7 4.7 G 1.5 3.4 G 0.6 4.0 G 1.3 0.7 G 0.3 3.0 G 0.2 1.8 G 0.4
SL1
SL2
S
M
S!M
A. antennatus
G. longipes
N. norvegicus
P. longirostris
P. martia
G. melastomus
H. dactylopterus
L. boscii
L. piscatorius
M. merluccius
M. poutassou
P. blennoides
e
e
ns
**
e
e
**
ns
ns
**
***
*
e
e
ns
ns
e
e
ns
ns
ns
ns
ns
ns
Total catch
Catch of commercial species
Discards
Discards of commercial species
**
**
*
ns
ns
ns
ns
ns
No differences between mesh shape were observed in the
species composition of catches (Figure 2). By contrast,
bathymetric differences in catch composition confirmed
two fishing strategies, previously described for the trawlfishery of the study area targeting N. norvegicus between
350 and 600 m (Merella et al., 1998) and A. antennatus
S
M
S!M
e
e
ns
ns
e
e
***
ns
ns
ns
*
ns
*
**
e
e
***
Ns
e
e
e
e
e
Ns
ns
ns
e
e
ns
ns
e
e
e
e
e
ns
ns
ns
e
e
ns
ns
e
e
e
e
e
ns
ns
*
ns
ns
Ns
ns
ns
Ns
ns
ns
*
**
ns
ns
ns
ns
deeper than this (Carbonell et al., 1999). Besides these
two crustaceans, other bycatch species were also captured
(Table 2). In the first fishing strategy, M. merluccius and
P. longirostris, two species with high market value and
yields greater than N. norvegicus, can be also considered
target species. In the second strategy, G. melastomus is
60
B. Guijarro and E. Massutı́
SL1
AUTUMN
SPRING
L 55
DIAMOND
L 82
OT 4
CE 1
CR 4
EL 7
CE 2
OT 4
CE 1
CR 5
EL 7
CR 43
D 18
NC
C
EL 34
CE 2
C
NC
EL 31
TE 84
TE 83
7
11
28
TE 21
L 90
OT 4
CE 3
CR 13
EL 3
17
TE 24
L 91
OT 3
SQUARE
CR 43
D 45
CE 1
CR 7
D 10
NC
C
EL 60
CE 1
CE 3
CR 13
EL 2
CR 40
D9
NC
C
EL 26
TE 78
TE 78
3
7
3
6
TE 32
TE 33
SL2
L 81
L 74
DIAMOND
CE 2
CR <1
CR 1
C
NC
TE 39
EL 14
D 19
EL 59
CR 2
CE 4
CR 1
EL 89
D 26
C
NC
TE 81
11
12
7
15
TE 10
TE 10
L 82
L 94
CE 8
CE 1
CR 4
CR <1
SQUARE
EL 88
EL 15
EL 18
D6
NC
C
D 18
NC
EL 89
TE 77
4
CE 1
CR 2
CE 6
CR 1
EL 91
C
TE 75
2
11
TE 6
7
TE 6
Figure 3. Catch composition for trawl groups SL1 and SL2 identified from the dendrogram, by season and mesh shape (L, landings; D,
discards; C, commercial species; NC, non-commercial species; TE, teleosts; EL, elasmobranchs; CR, crustaceans; CE, cephalopods; OT,
others).
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
SL1
Escapement ratio
50
40
30
20
10
0
AUTUMN
SL2
Economic loss
SPRING
<
**
SQ
AUTUMN
DI
>
**
4
AUTUMN
=
Economic efficiency
SPRING
3
2
<
**
DI
50
40
30
20
10
0
>
**
<
***
1
SQ
SPRING
DI
4
SQ
AUTUMN
DI
=
DI
SQ
DI
1
SQ
DI
SQ
SPRING
DI
<
<
***
***
SQ
DI
SPRING
15
=
SQ
AUTUMN
10
0
<
***
=
0
2
<
***
AUTUMN
5
<
***
3
<
***
15
10
0
61
DI
<
*
SQ
SPRING
5
=
=
0
SQ
DI
SQ
DI
SQ
Figure 4. Average (Gs.e.) escapement ratio, economic loss, and economic efficiency for trawl groups SL1 and SL2 identified from the
dendrogram, by season and mesh shape. The results of two-way ANOVA, showing the significance (Z, not significant; *, p ! 0.05; **,
p ! 0.01; ***, p ! 0.001) of the factors season (autumn and spring) and mesh shape (DI, diamond; SQ, square), are shown.
the only bycatch species with yields bigger than A. antennatus, but this elasmobranch cannot be considered as a target species because of its low market value (Table 1).
No differences between mesh shape were observed by
comparing commercial yields for the main species (Table 3),
although the comparison showed some seasonal differences
for P. longirostris, H. dactylopterus, M. merluccius, M.
poutassou, P. blennoides, A. antennatus, G. longipes, and
P. martia, probably related to their time of recruitment,
which in the western Mediterranean is clearly seasonal
for some (Massutı́ et al., 1996, 2001; Company and Sardà,
1997; Recasens et al., 1998). The only differences in yield
between the mesh shapes were obtained for total discards
and discards of commercial species for hauls targeted on
A. antennatus, in both cases discarding being higher
with diamond than with square mesh. Such a reduction
was also observed in the percentage of the discarded
catch, both total and commercial species, using square
mesh in each fishing strategy (Figure 3). For total discards
it varied between 8% and 36%, depending on depth interval and season. These two factors affect the abundance and
size range of the exploited species in this fishery and they
are on the basis of temporal variations observed in
catches, landings, and discards (Moranta et al., 2000).
The reduction of 4e11% of commercial species discarded
mainly affected the elasmobranchs Scyliorhinus canicula
and G. melastomus, the teleosts P. blennoides, H. dactylopterus, and L. boscii, and crustaceans of the genus Plesionika. On the whole, such a reduction could benefit the
environment by decreasing the impact of the fishery upon
such a fragile ecosystem (Moranta et al., 2000), one that is
particularly important for elasmobranchs because fishing
affects this group more than it does most teleosts (Stevens
et al., 2000). The Balearic Islands has one of the most diverse and abundant elasmobranch populations in the western Mediterranean and, for this reason, harvesting
strategies should be linked to conservation of species in
the area (Massutı́ and Moranta, 2003).
According to Fryer (1991), the selectivity of a net can
vary between hauls, so the estimation of selection curves
from a model based on between-haul variation provides
more realistic parameters than consideration of pooled
data over hauls. In the Mediterranean, the small catches
mean that the number of fish and shellfish retained and
that escape per haul are small too, precluding estimation
of selectivity parameters that take into account betweenhaul variability (Petrakis and Stergiou, 1997). In fact,
such information is scarce (Ragonese et al., 2002) and, in
our study, estimation of selection curves in this manner
was not possible for M. merluccius and G. longipes, and
is limited to diamond mesh. In any case, previous and current results (Tables 5 and 6) show similar values for selectivity parameters estimated from both methods.
Selectivity parameters were clearly lower for the diamond
than for the square mesh, except for L. boscii, for which values were similar. However, the same parameters estimated
for this species in the eastern Mediterranean are higher for
diamond than for square mesh (Petrakis and Stergiou,
1997). In general, square mesh is more selective than
diamond mesh of similar size for roundfish (Robertson
and Stewart, 1988), but the opposite pertains to flatfish
(Millar and Walsh, 1992). Although nothing is known in
the Mediterranean about the survival of fish and shellfish
after escapement from codends, estimates for other areas
range between 48% and 89% for gadoids through
diamond-mesh codends (Sangster et al., 1996). In the case
of elasmobranchs and decapods, their tougher skin and rigid
integument, respectively, should allow better survival.
The estimated length at first capture (L50) with 40-mm
square mesh for M. merluccius is similar to that obtained
for the same species in the Aegean Sea (15 cm TL; Petrakis
and Stergiou, 1997). For N. norvegicus, the L50 with the
square mesh in the current study is equal to that estimated
by Stergiou et al. (1997a) for the eastern Mediterranean
with the same size mesh (24 mm CL), but larger than the
various values reported for diamond mesh: (i) 23 mm CL
62
B. Guijarro and E. Massutı́
Table 5. Descriptive statistics (n, number of individuals; x, average; s.d., standard deviation) from the length frequency distribution (fish,
total length in cm; crustaceans, carapace length in mm) for the main species captured with diamond and square meshes by season.
Codend
Species
M. merluccius
Season
Autumn
Spring
M. poutassou
Autumn
Spring
P. blennoides
Autumn
Spring
H. dactylopterus
Autumn
Spring
L. boscii
Autumn
Spring
G. melastomus
Autumn
Spring
A. antennatus
Autumn
Spring
P. longirostris
Autumn
Spring
P. martia
Autumn
Spring
N. norvegicus
Autumn
Spring
G. longipes
Autumn
Spring
Cover
Mesh
shape
n
xCs:d:
Mode
Range
Diamond
Square
Diamond
Square
1 183
1 169
775
303
27.6 G 7.3
26.3 G 6.8
26.4 G 9.1
32.0 G 9.4
24
22
15
36
13e61
16e60
10e68
13e60
Diamond
Square
Diamond
Square
635
610
1 300
3 379
23.7 G 4.2
25.9 G 3.7
27.7 G 2.6
28.1 G 2.4
21
27
27
28
Diamond
Square
Diamond
Square
2 069
752
1 075
869
18.9 G 5.7
23.2 G 5.7
19.5 G 8.0
21.5 G 7.0
Diamond
Square
Diamond
Square
1 652
589
1 022
839
Diamond
Square
Diamond
Square
n
xCs:d:
Mode
Range
2
18
23
21
9.0 G 0.0
18.3 G 1.4
12.0 G 2.9
15.6 G 3.9
9
17; 18
8; 9; 12
13; 14; 15
9e9
17e21
8e16
12e28
18e44
16e40
20e35
23e38
1
158
6
0
13.0 G 0.0
20.4 G 2.1
9.7 G 1.0
e
13
20
9
e
13e13
15e31
9e11
e
13
25
10
23
11e47
11e60
7e51
7e54
149
732
2 476
3 143
12.0 G 1.2
13.4 G 1.4
9.8 G 1.1
9.7 G 0.9
11
13
10
10
10e15
10e22
6e18
7e12
12.7 G 2.8
11.7 G 2.5
14.9 G 4.2
14.5 G 3.3
12
12
12
12
8e26
4e25
7e30
8e32
249
1 022
194
562
6.8 G 1.4
8.6 G 1.7
7.6 G 1.3
9.0 G 1.4
8
9
8
8
5e10
4e14
3e13
6e14
624
711
1 077
1 575
17.6 G 6.2
15.9 G 5.9
17.2 G 4.8
16.3 G 4.5
12
12
16
14
10e37
10e38
9e36
9e34
50
66
49
20
10.3 G 1.5
11.5 G 1.6
10.0 G 0.8
11.3 G 2.5
11
11
10
9
6e12
6e16
9e12
9e16
Diamond
Square
Diamond
Square
2 326
1 269
1 236
1 128
33.6 G 13.1
27.6 G 10.6
27.4 G 12.5
30.0 G 14.2
15
22
24
17
12e61
15e62
11e61
12e61
139
2 093
146
1 919
15.9 G 1.7
18.9 G 2.6
16.2 G 2.7
17.4 G 2.6
15
18
16
16
11e24
10e30
12e26
11e31
Diamond
Square
Diamond
Square
6 221
5 846
8 143
8 482
33.7 G 7.9
34.7 G 7.7
34.5 G 9.4
36.3 G 8.8
41
25; 39
26
28
18e58
19e60
16e60
17e62
131
482
169
1 392
20.9 G 2.9
24.6 G 4.1
20.5 G 2.4
23.6 G 3.6
19
22
19; 20
22
14e29
15e44
15e29
16e38
Diamond
Square
Diamond
Square
9 157
8 465
11 523
17 379
26.4 G 3.2
27.1 G 3.3
28.5 G 3.4
29.0 G 3.2
25
24
27
32
16e39
19e38
18e37
18e37
84
222
31
761
23.0 G 3.2
24.2 G 1.8
23.9 G 2.8
26.3 G 2.9
23; 24
24
23
27
12e27
21e30
17e33
21e37
Diamond
Square
Diamond
Square
5 686
5 089
2 447
1 920
21.6 G 1.7
22.1 G 1.6
22.0 G 1.8
22.9 G 1.7
23
22
23
23
18e25
17e30
16e26
19e29
65
301
13
237
18.2 G 2.2
20.5 G 1.9
21.1 G 2.7
21.7 G 2.0
18
21
20
22
13e23
14e26
18e28
16e27
Diamond
Square
Diamond
Square
1 524
2 062
1 081
1 710
34.2 G 4.5
35.2 G 5.0
38.6 G 5.1
39.4 G 5.4
33
35
37
37
25e60
24e60
27e68
25e66
0
48
0
38
e
28.2 G 3.6
e
28.3 G 2.2
e
28
e
29
e
21e42
e
24e32
Diamond
Square
Diamond
Square
261
389
325
278
53.0 G 8.4
51.0 G 9.5
50.3 G 7.8
49.1 G 9.2
61
49
52
51; 54
24e73
21e67
26e65
21e66
0
4
0
0
e
25.0 G 3.6
e
e
e
20; 25; 27; 28
e
e
e
20e28
e
e
Diamond
Species
Square
Season
S1
S2
L25
L50
L75
R11
R12
R22
S1
S2
L25
L50
L75
R11
R12
R22
M. poutassou
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
17.990
e
17.990
0.894
e
0.894
18.9
e
18.9
20.1
e
20.1
21.4
e
21.4
18.650
e
18.650
0.902
e
0.902
0.044
e
0.044
P. blennoides
A
S
T
12.050
7.425
8.987
1.005
0.589
0.735
10.9
10.7
10.7
12.0
12.6
12.2
13.1
14.5
13.7
27.170
4.741
4.623
1.887
0.341
0.341
0.131
0.025
0.026
10.430
7.639
7.881
0.704
0.559
0.548
13.3
11.7
12.4
14.8
13.7
14.4
16.4
15.6
16.4
5.994
2.015
1.188
0.463
0.142
0.082
0.037
0.010
0.006
H. dactylopterus
A
S
T
26.380
12.090
13.910
3.032
1.325
1.562
8.3
8.3
8.2
8.7
9.1
8.9
9.1
10.0
9.6
12.420
1.463
1.499
1.388
0.156
0.177
0.160
0.017
0.021
8.337
14.93
13.01
0.785
1.357
1.19
9.2
10.2
10.0
10.6
11.0
10.9
12.0
11.8
11.9
8.190
1.387
3.783
0.719
0.111
0.325
0.064
0.009
0.028
L. boscii
A
S
T
e
7.992
8.951
e
0.846
0.903
e
8.1
8.7
e
9.4
9.9
e
10.7
11.1
e
11.630
6.310
e
1.105
0.545
e
0.105
0.048
e
e
14.69
e
e
1.443
e
e
9.4
e
e
10.2
e
e
10.9
e
e
78.980
e
e
7.507
e
e
0.714
G. melastomus
A
S
T
e
3.779
4.025
e
0.283
0.305
e
9.5
9.6
e
13.4
13.2
e
17.3
16.8
e
1.285
0.954
e
0.062
0.045
e
0.003
0.002
10.63
7.792
8.840
0.470
0.358
0.395
20.3
18.7
19.6
22.6
21.8
22.4
24.9
24.8
25.1
0.847
0.493
0.443
0.020
0.020
0.015
0.001
0.001
0.001
A. antennatus
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
19.480
7.725
12.130
0.897
0.345
0.550
20.5
19.2
20.1
21.7
22.4
22.1
23.0
25.6
24.1
93.140
0.154
13.780
4.380
0.006
0.652
0.206
0.001
0.031
P. longirostris
A
S
T
e
e
9.754
e
e
0.587
e
e
14.7
e
e
16.6
e
e
18.5
e
e
7.318
e
e
0.310
e
e
0.013
28.800
13.980
19.070
1.437
0.658
0.943
19.3
19.6
19.1
20.0
21.2
20.2
20.8
22.9
21.4
51.680
7.958
21.910
2.715
0.407
1.116
0.142
0.021
0.057
P. martia
A
S
T
e
15.550
15.550
e
0.968
0.968
e
14.9
14.9
e
16.1
16.1
e
17.2
17.2
e
3.059
3.059
e
0.166
0.166
e
0.009
0.009
9.111
30.380
18.51
0.551
1.602
1.000
14.5
18.3
17.4
16.5
19.0
18.5
18.5
19.6
19.6
0.802
9.001
18.870
0.039
0.485
0.966
0.002
0.027
0.049
N. norvegicus
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
22.110
19.560
17.330
0.852
0.700
0.651
24.7
26.4
24.9
26.0
27.9
26.6
27.2
29.5
28.3
10.910
23.770
7.872
0.392
0.801
0.296
0.014
0.027
0.011
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
Table 6. Selectivity parameters (S1 and S2, selection curve parameters; L25, L50, and L75, lengths at which 25%, 50%, and 75%, respectively, of the specimens are retained in the codend; R11,
R12, and R22, variance matrix of parameter estimates) by mesh shape, season (A, autumn; S, spring), and both seasons combined (T) for the main species, estimated by the method of Fryer
(1991).
63
Diamond
Square
S1
S2
L25
L50
L75
r2
S1
S2
L25
L50
L75
r2
M. merluccius
A
S
T
e
32.435
33.344
e
2.801
2.878
e
11.2
11.2
e
11.6 (11.4e11.8)
11.6 (11.3e11.9)
e
12.0
12.0
e
0.998
0.998
9.437
14.262
15.026
0.618
0.928
0.983
13.5
14.2
14.2
15.3 (14.8e15.7)
15.4 (14.9e15.8)
15.3 (14.9e15.6)
17.0
16.5
16.4
0.942
1
0.987
M. poutassou
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
8.635
e
9.682
0.420
e
0.478
17.9
e
17.9
20.5 (18.6e22.5)
e
20.2 (19.1e21.4)
23.1
e
22.5
0.545
e
0.757
P. blennoides
A
S
T
23.688
6.099
6.409
2.039
0.483
0.533
11.1
10.4
10.0
11.6 (11.4e11.9)
12.6 (11.9e13.4)
12.0 (11.5e12.5)
12.2
14.9
14.1
0.967
1
0.978
12.610
11.209
8.818
0.777
0.816
0.542
14.8
12.4
14.2
16.2 (15.8e16.6)
13.7 (13.1e14.4)
16.3 (15.6e16.9)
17.6
15.5
18.3
0.995
1
0.996
H. dactylopterus
A
S
T
28.221
14.075
16.306
3.334
1.591
1.884
8.1
8.1
8.1
8.5 (8.4e8.5)
8.8 (8.6e9.1)
8.7 (8.4e8.9)
8.8
9.5
9.2
0.996
1
0.968
14.113
14.394
6.789
1.303
1.306
0.612
10.0
10.2
9.3
10.8 (10.6e11.0)
11.0 (10.9e11.1)
11.1 (10.3e11.8)
11.7
11.9
12.9
1
1
0.988
L. boscii
A
S
T
26.857
13.551
10.510
2.601
1.328
1.070
9.9
9.4
8.8
10.3 (8.9e11.8)
10.2 (10.1e10.3)
9.8 (9.7e10.0)
10.0
11.0
10.8
0.997
1
0.998
431.810
35.954
29.246
43.291
3.809
3.064
9.9
9.2
9.2
9.9 (9.8e10.2)
9.4 (9.4e9.5)
9.5 (9.5e9.6)
10.0
9.7
9.9
0.994
0.999
0.998
G. melastomus
A
S
T
59.472
2.499
2.854
4.675
0.229
0.248
12.5
6.1
7.1
12.7 (12.0e13.5)
10.9 (9.7e12.1)
11.5 (10.8e12.3)
12.9
15.7
15.9
0.976
0.997
0.931
10.381
6.810
7.334
0.465
0.312
0.331
20.0
18.3
18.8
22.3 (21.9e22.7)
21.8 (21.3e22.4)
22.2 (21.7e22.6)
24.7
25.4
25.5
1
1
1
A. antennatus
A
S
T
23.789
16.987
17.060
1.336
1.017
0.994
17.9
15.6
16.1
18.0 (17.9e18.0)
16.7 (16.3e17.1)
17.2 (16.7e17.6)
18.0
17.8
18.3
0.998
0.986
0.984
12.291
7.855
9.180
0.579
0.347
0.416
19.4
19.5
19.4
21.3 (20.7e21.9)
22.6 (22.1e23.1)
22.1 (21.6e22.5)
23.2
25.8
24.7
0.968
0.993
0.991
P. longirostris
A
S
T
15.460
e
22.821
0.883
e
1.324
16.3
e
16.4
17.5 (17.2e17.8)
e
17.2 (17.1e17.4)
18.7
e
18.1
0.993
e
0.998
26.106
14.511
21.566
1.284
0.696
1.046
19.5
19.3
19.6
20.3 (20.2e20.4)
20.8 (20.6e21.1)
20.6 (20.4e20.8)
21.2
22.4
21.7
0.999
0.981
0.998
P. martia
A
S
T
331.150
e
55.819
19.479
e
3.374
16.9
e
16.2
17.0 (16.9e17.0)
e
16.5 (16.2e17.0)
17.1
e
16.9
0.999
e
0.999
20.833
425.320
20.303
1.156
22.463
1.118
17.1
18.8
17.2
18.0 (17.1e18.9)
18.9 (18.9e19.0)
18.2 (17.3e19.0)
19.0
19.0
19.2
0.912
0.984
0.93
N. norvegicus
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
13.362
16.702
36.839
0.545
0.609
1.496
22.5
25.6
23.9
24.5 (24.2e24.8)
27.4 (26.9e28.0)
24.6 (24.3e25.3)
26.5
29.2
25.4
0.971
0.987
0.995
G. longipes
A
S
T
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
26.850
e
26.960
1.070
e
1.075
24.1
e
24.0
25.1 (24.9e25.3)
e
25.1 (24.6e25.5)
26.1
e
26.1
1
e
1
B. Guijarro and E. Massutı́
Season
Species
64
Table 7. Selectivity parameters (S1 and S2, selection curve parameters; L25, L50, and L75, lengths at which 25%, 50%, and 75%, respectively, of the specimens are retained in the codend; 95% CI, 95%
confidence interval of L50; r2, correlation coefficient) by mesh shape, season (A, autumn; S, spring), and both seasons combined (T) for the main species, estimated from pooled data.
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
M. merluccius
H. dactylopterus
100
100
100
75
75
75
50
50
50
25
0
25
b)
0
0 10 20 30 40 50 60 70
25
a)
0
10
20
30
40
M. poutassou
0
100
100
100
75
75
75
75
50
50
50
50
0
25
a)
0
10
20
30
40
50
0
25
b)
0
10
20
30
40
50
0
0
10
20
30
40
50
0
100
100
100
75
75
75
75
50
50
50
50
25
b)
0
0
10
20
30
40
b)
0
10
20
30
40
50
G. melastomus
100
a)
0
25
a)
P. blennoides
25
b)
L. boscii
100
25
65
25
a)
0
25
b)
0
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
Total length, TL (cm)
Total length, TL (cm)
Total length, TL (cm)
Total length, TL (cm)
Figure 5. Selectivity curves (seasons combined) of the main fish species by mesh shape (solid line, diamond mesh; dashed line, square
mesh) and method of analysis. Panels marked (a) are individual haul curves (thin lines) and the mean curve (thick line) calculated using the
method of Fryer (1991); those marked (b) display the data (>, diamond mesh; ,, square mesh) and selection curves from pooled hauls.
with 40-mm mesh in the eastern Mediterranean (Stergiou
et al., 1997a); (ii) 18 and 20 mm CL with 40- and 48-mm
mesh, respectively, in the same area (Mytilineou et al.,
1998); and (iii) 15 and 21 mm CL with 38- and 42-mm
mesh, respectively, in the northwestern Mediterranean (Sardà
et al., 1993). In any case, square mesh is clearly more efficient than diamond mesh when trawling for N. norvegicus.
Introduction of a 40-mm square mesh codend to the
deepwater trawl fishery off the Balearic Islands would improve the exploitation pattern of the main target species,
which currently show clear symptoms of overexploitation
(Sardà, 1998; Garcı́a-Rodrı́guez and Esteban, 1999;
GFCM, 2004), by reducing the fishing pressure on small
fish, generating improvements in the state of these resources and benefits in their yield per recruit. Although such
a change in codend mesh would produce a small, but significant, increase in the escapement ratio (5e15%) and an
economic loss (1.4e2.4%; Figure 4), the yields of main
species, in terms of biomass, and the economic efficiency
would be maintained. In part, this is because escaping individuals are small and of relatively low value. Further, even
given small economic loss in the short term, there may be
recovery in the medium and longer term, as demonstrated
for the P. longirostris fishery off the Gulf of Cádiz (Sobrino
et al., 2000). In addition, an increase in L50 with the use of
a square mesh would remove some of the contradictions in
management of the Mediterranean trawl fishery. For N. norvegicus, the estimated L50 with square mesh is greater than
the minimum legal landing size (20 mm CL), which contrasts with the pessimistic perspective of Sardà (1998),
who concluded that ‘‘fisheries regulations relating to
mesh trawl selectivity for Nephrops are not efficient and
should be abandoned’’. For P. longirostris and A. antennatus, species without a minimum legal landing size, the L50
with square mesh is at the size at first maturity, estimated at
20e22 mm CL (Mori et al., 2000; Ben Meriem et al., 2001)
and 16e29 mm CL (Carbonell et al., 1999; Garcı́a-Rodrı́guez and Esteban, 1999), respectively. Although effective,
a square mesh will not allow such a target objective to be
attained for M. merluccius, for which the legal landing
size is 20 cm TL and the length at first maturity estimated
to be 32 cm TL (Oliver, 1993). Reducing the capture of undersized marketable species, which only influences fishing
mortality without yielding economic benefit, could help
the fishery by minimizing the handling and sorting time
of catches and improving the quality of landings.
66
B. Guijarro and E. Massutı́
A. antennatus
100
100
75
75
50
50
25
In conclusion, within the context of precautionary management, introducing a 40-mm square mesh in trawl codends
could be an appropriate and plausible measure to improve
the state of the resources exploited by the deepwater crustacean trawl fishery off the Balearic Islands, and concomitantly
reduce the impact of the fishery on the ecosystem.
25
a)
b)
0
0
0 10 20 30 40 50 60 70
0 10 20 30 40 50 60 70
P. longirostris
100
100
75
75
50
50
25
25
a)
0
0
10
20
30
40
50
0
b)
0
10
20
30
40
50
P. martia
100
100
75
75
50
50
25
0
0
10
20
30
40
0
Caparace length, CL (mm)
b)
0
10
20
30
40
Caparace length, CL (mm)
N. norvegicus
100
100
75
75
50
50
25
25
a)
0
b)
0
0 10 20 30 40 50 60 70 80
0 10 20 30 40 50 60 70
We thank Joan Jesús Vaquero, Damià Gómez, Óscar Fernández, Manuel Salvà, Juan José Picó, and Vicente Sempere, the captain and crew of the FV ‘‘Moralti Nou’’, for
their help during the surveys, and Biel Pomar, Ramón
Mas, Maria Magdalena Guardiola, and José Luis Pellicer,
of the Centre Oceanogràfic de les Balears, for their assistance in both field and laboratory. We also thank Chris
Rodgers, John Gordon, Andrew Revill, and an anonymous
reviewer for their valued comments on an earlier draft. The
study was financed by the Spanish Ministry of Fisheries
(RAI-AP-22/2001 and RAI-AP-6/2002).
References
25
a)
Acknowledgements
Caparace length, CL (mm)
G. longipes
100
75
50
25
b)
0
0
20
40
60
80
Caparace length, CL (mm)
Figure 6. Selectivity curves (seasons combined) of the main crustacean
species by mesh shape (solid line, diamond mesh; dashed line, square
mesh), and method of analysis. Panels marked (a) are individual haul
curves (thin lines) and the mean curve (thick line) calculated by the
method of Fryer (1991); those marked (b) display the data (>, diamond
mesh; ,, square mesh) and selection curves from pooled hauls.
Armstrong, D. W., Ferro, R. S. T., MacLennan, D. N., and Reeves,
S. A. 1990. Gear selectivity and the conservation of fish. Journal
of Fish Biology, 37A: 261e262.
Ben Meriem, S., Fehri-Bedoui, R., and Gharbi, H. 2001. Taille à
maturation et période de ponte de la crevette rose, Parapenaeus
longirostris (Lucas, 1846) de Tunisie. Crustaceana, 74: 39e48.
Campos, A., Fonseca, P., and Erzini, K. 2002. Size selectivity
of diamond and square mesh cod end for rose shrimps
(Parapenaeus longirostris) and Norway lobster (Nephrops
norvegicus) off the Portuguese south coast. Fisheries Research,
58: 281e301.
Carbonell, A., Carbonell, M., Demestre, M., Grau, A., and Monserrat, S. 1999. The red shrimp Aristeus antennatus (Risso, 1816)
fishery and biology in the Balearic Islands, western Mediterranean. Fisheries Research, 44: 1e13.
Company, J. B., and Sardà, F. 1997. Reproductive patterns and
population characteristics in five deep-water pandalid shrimps
in the western Mediterranean along a depth gradient
(150e1100 m). Marine Ecology Progress Series, 148: 49e58.
Fryer, R. J. 1991. A model of between-haul variation in selectivity.
ICES Journal of Marine Science, 48: 281e290.
Garcia, S. M. 2000. The FAO definition of sustainable development and the Code of Conduct for Responsible Fisheries: an
analysis of the related principles, criteria and indicators. Marine
and Freshwater Research, 51: 535e541.
Garcı́a-Rodrı́guez, M., and Esteban, A. 1999. On the biology and
fishery of Aristeus antennatus (Risso, 1816), (Decapoda, Dendrobranchiata) in the Ibiza Channel (Balearic Islands, Spain).
Scientia Marina, 63: 27e37.
GFCM. 2000. Report of the Second Stock Assessment Sub-Committee Meeting. Scientific Advisory Committee of the General
Fisheries Commission for the Mediterranean. Madrid, Spain,
26e28 April 2000. 16 pp.
GFCM. 2001. Report of the Third Stock Assessment Sub-Committee Meeting. Scientific Advisory Committee of the General Fisheries Commission for the Mediterranean. Lacco Ameno, Italy,
10e13 September 2001. 27 pp.
GFCM. 2004. Report of the Sixth Stock Assessment Sub-Committee Meeting. Scientific Advisory Committee of the General Fisheries Commission for the Mediterranean. Málaga, Spain, 10e12
May 2004. 73 pp.
Diamond- and square-mesh codend selectivity in the Balearic Islands deepwater crustacean trawl fishery
MacLennan, D. N. (Ed.). 1992. Fishing gear selectivity. Fisheries
Research, 13: 201e352.
Mallol, S., Casadevall, M., and Garcı́a, E. 2001. Comparison of
discarded, escaped and landed fish using diamond and square
mesh codends. Rapport du Commission Internationale pour l’Exploration Scientifique de la mer Méditerranée, 36: 296.
Massutı́, E., and Moranta, J. 2003. Demersal assemblages and
depth distribution of elasmobranchs from the continental shelf
and slope trawling grounds off the Balearic Islands (western
Mediterranean). ICES Journal of Marine Science, 60: 753e766.
Massutı́, E., Morales-Nin, B., and Lloris, D. 1996. Bathymetric distribution and recruitment patterns of Phycis blennoides (Pisces:
Gadidae) from the slope of the northwestern Mediterranean. Scientia Marina, 60: 481e488.
Massutı́, E., Moranta, J., Gil de Sola, L., Morales-Nin, B., and
Prats, L. 2001. Distribution and population structure of the rockfish Helicolenus dactylopterus (Pisces: Scorpaenidae) in the
western Mediterranean. Journal of the Marine Biological Association of the UK, 81: 129e141.
Merella, P., Alemany, F., Carbonell, A., and Quetglas, A. 1998.
Fishery and biology of Norway lobster Nephrops norvegicus
(Decapoda: Nephropidae) in Mallorca (western Mediterranean).
Journal of Natural History, 32: 1631e1640.
Millar, R. B., and Walsh, S. J. 1992. Analysis of trawl selectivity
studies with an application to trouser trawls. Fisheries Research,
13: 205e220.
Moranta, J., Massutı́, E., and Morales-Nin, B. 2000. Fish catch
composition of the deep-sea decapod crustacean fisheries in
the Balearic Islands (western Mediterranean). Fisheries Research, 45: 253e264.
Mori, M., Sbrana, M., and De Ranieri, S. 2000. Reproductive biology of female Parapenaeus longirostris (Crustacea, Decapoda,
Penaeidae) in the northern Tyrrhenian Sea (western Mediterranean). Atti Società Toscana di Scienze Naturali, Serie B, 107:
1e6.
Mytilineou, C., Politou, C-Y., and Fortouni, A. 1998. Trawl selectivity studies on Nephrops norvegicus (L.) in the eastern Mediterranean Sea. Scientia Marina, 62(Suppl. 1): 107e116.
Oliver, P. 1993. Analysis of fluctuations observed in the trawl
fleet landings of the Balearic Islands. Scientia Marina, 57:
219e227.
Petrakis, G., and Stergiou, K. I. 1997. Size selectivity of diamond
and square mesh codends for four commercial Mediterranean
fish species. ICES Journal of Marine Science, 54: 13e23.
Ragonese, S., Bianchini, M. L., and Di Stefano, L. 2002. Trawl
cod-end selectivity for deepwater red shrimp (Aristaeomorpha
foliacea, Risso, 1827) in the Strait of Sicily (Mediterranean
Sea). Fisheries Research, 57: 131e144.
Ragonese, S., Zagra, M., Di Stefano, L., and Bianchini, M. L. 2001.
Effect of codend mesh size on the performance of the deep-water
67
bottom trawl used in the red shrimp fishery in the Strait of Sicily
(Mediterranean Sea). Hydrobiologia, 449: 279e291.
Recasens, L., Lombarte, A., Morales-Nin, B., and Torres, G. J.
1998. Spatiotemporal variation in the population structure of
the European hake in the NW Mediterranean. Journal of Fish Biology, 53: 387e401.
Reeves, S. A., Armstrong, D. W., Fryer, R. J., and Coull, K. A.
1992. The effects of mesh size, cod-end extension length and
cod-end diameter on the selectivity of Scottish trawls and seines.
ICES Journal of Marine Science, 49: 279e288.
Robertson, J. H. B., and Stewart, P. A. M. 1988. A comparison of
size selection of haddock and whiting by square and diamond
mesh codends. Journal du Conseil International pour l’Exploration de la Mer, 44: 148e161.
Sangster, G. I., Lehmann, K., and Breen, M. 1996. Commercial
fishing experiments to assess the survival of haddock and whiting after escape from four sizes of diamond mesh cod-ends.
Fisheries Research, 25: 323e345.
Sardà, F. 1998. Nephrops norvegicus (L.): Comparative biology
and fishery in the Mediterranean Sea. Introduction, conclusions
and recommendations. Scientia Marina, 62(Suppl. 1): 5e15.
Sardà, F., Conan, G. Y., and Fusté, X. 1993. Selectivity of Norway
lobster Nephrops norvegicus (L.) in the northwestern Mediterranean. Scientia Marina, 57: 167e174.
Sardà, F., Molı́, B., and Palomera, I. 2004. Preservation of juvenile
hake (Merluccius merluccius, L.) in the western Mediterranean
demersal trawl fishery by using sorting grids. Scientia Marina,
68: 435e444.
Sobrino, I., Garcı́a, T., and Baro, J. 2000. Trawl gear selectivity
and the effect of mesh size on the deep-water rose shrimp (Parapenaeus longirostris, Lucas, 1846). Fisheries Research, 44:
235e245.
Stergiou, K. I., Petrakis, G., and Politou, C-Y. 1997a. Size
selectivity of diamond and square mesh cod-ends for
Nephrops norvegicus in the Aegean Sea. Fisheries Research,
29: 203e209.
Stergiou, K. I., Politou, C-Y., Christou, E. D., and Petrakis, G.
1997b. Selectivity experiments in the NE Mediterranean: the effect of trawl codend mesh size on species diversity and discards.
ICES Journal of Marine Science, 54: 774e786.
Stevens, J. D., Bonfil, R., Dulvy, N. K., and Walker, P. A. 2000.
The effects of fishing on sharks, rays, and chimaeras (chondrichthyans), and the implications for marine ecosystems.
ICES Journal of Marine Science, 57: 476e494.
Wileman, D. A., Ferro, R. S. T., Fonteyne, R., and Millar, R. B.
(Eds). 1996. Manual of Methods of Measuring the Selectivity
of Towed Fishing Gears. ICES Cooperative Research Report,
215. 126 pp.
Zar, J. H. 1996. Biostatistical Analysis, 3rd edn. Prentice-Hall,
New Jersey. 662 pp.