Effects of seasonal closures in a multi-specific fishery · 2015. 9. 28. · Mohamed Samy-Kamal∗, Aitor Forcada, José Luis Sánchez Lizaso Departamento de Ciencias del Mar y Biología
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
E
MDA
a
ARRA
KCEEEFMT
1
rotmdmcr1tmae
(
h0
Fisheries Research 172 (2015) 303–317
Contents lists available at ScienceDirect
Fisheries Research
j ourna l ho me page: www.elsev ier .com/ locate / f i shres
ffects of seasonal closures in a multi-specific fishery
ohamed Samy-Kamal ∗, Aitor Forcada, José Luis Sánchez Lizasoepartamento de Ciencias del Mar y Biología Aplicada, Universidad de Alicante, PO Box 99, Edificio Ciencias V, Campus de San Vicente del Raspeig, E-03080licante, Spain
r t i c l e i n f o
rticle history:eceived 20 August 2014eceived in revised form 15 July 2015ccepted 26 July 2015
eywords:atch per unit effort (CPUE)ffort reductionvaluation of closuresx-vessel priceisheries managementanagement measures
rawl fishery
a b s t r a c t
In input-controlled multi-specific fisheries, seasonal closure has little biological rationale as a manage-ment measure, because it is difficult to adjust such closure for many target species and, in most cases, theyare adopted for economic purposes. We aimed to determine effects of closure in biologic and economicterms, using 10-year landing data from two representative trawling ports of the Western Mediterranean:Dénia and La Vila Joiosa. Analysis of Variance (ANOVA) was used to detect significant differences, beforeand after the closure, in standardized catch per unit effort (CPUE) at different seasons and sale prices athome/closed and neighbour/open ports. ANOVAs showed significantly higher CPUE after the closure fortotal landings and Mullus spp. of the Red mullet métier, Merluccius merluccius CPUE (in two years) andthe total landings of the Norway lobster métier. On the contrary, significant lower values were observedafter the closure for total CPUE (in early summer) and Aristeus antennatus of Red shrimp métier. SimilarCPUE was observed at all levels when the closure took place in late summer. In economic terms, mar-ket prices of target species have decreased or shown no changes after the closure at home/closed andneighbouring/open ports. The only exception was the significant increase of the price for A. antennatus
in Dénia during the closure in La Vila Joiosa. Depending on its timing, the closure would highlight somepositive biological effects on some target species. However, closure leads to an unavoidable reductionin most of target species prices. An alternative management measure that is based on effort reductionin input-controlled multi-specific fisheries could ban one day per week when market prices of targetspecies are lower.
A large number of fish stocks are overexploited in Mediter-anean multi-specific fisheries, and reductions of fishing mortalityn these stocks are often recommended (FAO, 2011). Fishing mor-ality is normally reduced through effort reductions, which can
ainly be done by decreasing the number of vessels or the fishingays. The adoption of closed fishing seasons is one of the simplesteasures used in the management of fisheries. Closure means a
omplete cessation of fishing activity for a certain period, whichesults in a reduction of annual effort (Lleonart and Franquesa,999). This management strategy is mainly based on effort con-rol which reduces fishing intensity and protects target stock from
ortality at a specific stage of the life history, i.e. when a speciesggregates in an area or in a specific season to spawn (Horwoodt al., 1998; Dinmore et al., 2003). This approach also can help repro-
ductive success and support recruitment (Arendse et al., 2007).However, it is well-known that in multi-species fisheries, such asthe Mediterranean Sea, there are many target species with differentrecruitment and reproduction periods. Consequently, a particularperiod may help the recruitment or the reproduction of certainspecies and not others (Lleonart and Franquesa, 1999). Therefore,in Mediterranean multi-specific fisheries, the adoption of closure,in some cases, is based on economic purposes in agreement withfishermen (Lleonart and Franquesa, 1999).
From an economic perspective, a temporary/seasonal closuremay have short-term benefits to fishermen: (i) the reduction ofoperating costs; (ii) financial compensation arising from the recov-ery of stocks where fishing has ceased; and (iii) compensationsubsidies (if the administration funds the closure) (Lleonart andFranquesa, 1999). However, ceasing the fleet for long periods (e.g.,monthly closure) results in serious logistical and economic prob-lems, namely: (i) fishermen unemployment during the closure
period; (ii) “border effect” the result of imbalances between thefleet activity belonging to adjacent ports without closure (Lleonartand Franquesa, 1999); (iii) the market for some luxury speciesbecoming devoid of highly appreciated local products (Guillen and
aynou, 2014); (iv) imbalances in market price due to the irregularupply of fish to the market (Guerra-Sierra and Sánchez-Lizaso,998); and (v) rise in administration cost in the form of state sub-idies.
Closure in Mediterranean multi-specific fisheries have a lit-le biological rationale because it is very difficult to adjust thelosure to reproductive periods of many target species (Table 1)Lleonart and Franquesa, 1999); also it generates some logisticroblems (Guerra-Sierra and Sánchez-Lizaso, 1998; Lleonart andranquesa, 1999; Guillen and Maynou, 2014). The closures are notntended to protect spawning stock at a vulnerable point in theirife cycle, thereby enhancing the probability of sustaining recruit-
ent; rather, they are adapted generally for economic purposes andeducing effort intensity. Closures can be justified in multi-specificsheries if it results in substantial biological or economic benefits,ther than effort reduction. These benefits can be seen by increases
n landings (e.g., in kg or in first sale price) that compensate somef the previously mentioned problems. Otherwise, effort reductionan be achieved by adopting other less-problematic managementeasures rather than closure, i.e., reduction of fishing days or hours.
Temporary/seasonal closures are widely studied in many fish-ries throughout the world (e.g., Ye, 1998; Pipitone et al., 2000;rendse et al., 2007; Shih et al., 2009). For instance, in the Gulff Castellammare (NW Sicily, Mediterranean Sea), Pipitone et al.2000) addressed that temporary closure based on year-roundrawling bans, may prove useful especially for multispecies and
ultigear artisanal fisheries. Studies in the Western Mediter-anean are limited to ecological effect on epibenthic communitiesDemestre et al., 2008) and on catch composition in the Catalan SeaSánchez et al., 2007). In the Adriatic and the Catalan Seas, Demestret al. (2008) reported a decrease of epibenthic faunal abundanceith the resumption of fishing activity after the closure at bothshing grounds. Further in both Seas, the species composition ofoth the retained and discarded fractions was analysed by Sánchezt al. (2007), where in both fishing grounds the retained fractionas slightly higher in the high fishing intensity periods than in the
ow intensity ones. Thereby the effectiveness of specific tempo-ary/seasonal closures as the most applied management measureor multi-specific fisheries should be rigorously evaluated in bothiological and economic terms using long-term landings data. Inddition, there are many target species with different recruitmentnd reproduction periods; thus the timing of the closure shoulde taken into account as suitable timing may or may not benefitarticular species.
The aim of this work was to determine the effect of seasonal clo-ure in biological (total landings and landings of target species) andconomic (ex-vessel prices “first sale price” of target species) terms,n a commercial Spanish trawling fishery. The data were derivedrom two representative fishing ports (Dénia and La Vila Joiosa) inhe Western Mediterranean.
. Material and methods
.1. Study area
This study was conducted in two ports, Dénia and La Vila Joiosa,ocated in the Southwestern Mediterranean Sea off the coast ofpain (Fig. 1). Along the gulf of Alicante, there are 12 fishing portshat have traditionally been important fishing activity locations.ccording to the number of trawlers, these two ports representbout 41% of the total trawlers operating on the Alicante coast (BOE,
013). They can be considered quite representative of this area,iven the similarity of the characteristics of the trawlers, and alsoave features similar to those operating in other areas of the West-rn Mediterranean (Samy-Kamal et al., 2014). The Mediterranean
search 172 (2015) 303–317
trawl fishery in Spain is an input-controlled fishery, where effortis controlled by limiting the time at sea: fishing is permitted for12 h/day from Monday to Friday, stopping the fishing activity com-pletely on weekends (Maynou et al., 2006). The fishing activity isceased normally for one month per year as seasonal closures, alter-nating the North ports (e.g., Dénia) with the south ports (e.g., La VilaJoiosa) to avoid the closure of the whole gulf at once (Table 2). Thespecies Mullus spp. (Linnaeus, 1758), Merluccius merluccius (Lin-naeus, 1758), Nephrops norvegicus (Linnaeus, 1758) and Aristeusantennatus (Risso, 1816) are the most targeted by fishermen andaccounted for almost 60% of the total income and 24% of the totalweight in the fishery (Samy-Kamal et al., 2014, 2015a). In regardsto stocks, in general, the Mediterranean and Black Sea had 33% ofassessed stocks fully exploited, while the great bulk (50%) overex-ploited (FAO, 2011). Almost all demersal fish and crustaceans stocksassessed were classified as overexploited including the four targetspecies studied herein (FAO, 2011).
2.2. Data collection
Two different data sets were used, one for each port. Datarecords of daily auctions were obtained from the fishing guild ofeach port for 10 years (2002–2011). For each fishing day, data onspecies landing weight (kg) and its first sale value (D ) were avail-able by vessel. Sale value (revenue) is the result of quantity landed(kg) and ex-vessel fish price (price obtained by fishers per kg oflanded fish). The sale value (D ) of each target species was dividedby its landings (kg) to calculate the first sale price per kg (ex-vesselfish price). Vessels with sporadic landings events (less than 3 years,and less than 3 months/year) within the ports were excluded fromthe analysis, considering only those vessels registered in the stud-ied ports (home port) to avoid possible biases in the data. Mostof the included vessels have had activity throughout the consid-ered period. The total number of collected samples (vessel/day) was102,187 fishing days. Technical characteristics of vessels within theanalysis were obtained from the Census of Fleet Operations of theGeneral Secretariat of Maritime Fisheries of Spain (BOE, 2013). Overthe 10 years studied, a total of 93 different fishing vessels werelisted in the official fleet register of Dénia and La Vila Joiosa (34and 59 vessels, respectively). The bulk of the fleet is composed ofvessels up to 23–25 m length, 40-80 GT, 40–60 GRT and 200–400registered HP (Samy-Kamal et al., 2014).
2.3. Data standardization
For multi-specific fisheries, a preliminary analysis of the fish-ing tactics in the fishery is essential to clearly determine the realeffort directed at the species under study (Maynou et al., 2003).Four principal métiers, Red mullet, European hake, Norway lob-ster and Red shrimp, were identified based on catch profiles andthe main target species, using the multivariate analysis: cluster,nMDS and SIMPER (Samy-Kamal et al., 2014, 2015a,b). Catch rateswere standardized to separate that large percentage of the vari-ability of data not directly attributable to variations in abundance.To standardize the catch per unit effort (CPUE), generalized linearmodels (GLM) were used (Maynou et al., 2003; Maunder and Punt,2004; Murawski et al., 2005). A minimum threshold of effort byvessel of 100 fishing days per year was considered; also, a selec-tion of vessels operating in the fishery for more than 4 years wascarried out with the intention of standardizing CPUE data fromvessels that would be representative of the fishery. Once the selec-tion of representative vessels was undertaken, a data matrix by
métier was constructed with the variables required for analysis.The initial set of explanatory variables considered was: temporalvariables (Year and Month) to capture temporal variations; techni-cal variables (vessel’s total length “TL” and gross tonnage “GT”) to
M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317 305
Table 1Spawning (gray cells) periods of the main target species: Mullus spp. Merluccius merluccius, Nephrops norvegicus and Aristeus antennatus by month.
Target species January February March April May June July August September October November December Reference
Mullus spp. (Relini et al., 1999;Voliani, 1999; Sieliet al., 2011 Sieli et al.,2011)
Merluccius merluccius (Martin et al., 1999;Martin et al., 1999)
Table 2Temporal/seasonal closures of trawling fisheries in Dénia and La Vila Joiosa ports during the studied 10 years (2002–2011). Shaded years were used in the analysis of variance(ANOVA).
apture differences between vessel characteristics; and the “indi-idual Vessel” was also used as an alternative in case if technicalactors were not significant. Regarding the “individual Vessel” fac-or in the analysis of Mediterranean fisheries CPUE, various authorsave used vessel factor, grouped into categories according to theirechnical characteristics (Goni et al., 1999), while others have usedhe “individual Vessel” factor (Maynou et al., 2003; Sbrana et al.,003). In the Mediterranean small and medium-scale fisheries,he experience and skills of the fishermen determine and influ-nce the result of fishing operations. This fact justifies that it isore appropriate to include the factor “individual Vessel” in theodels separately, rather than grouped into categories (Maunder
nd Punt, 2004). The “individual Vessel” factor includes other fac-ors that are not directly related to the technical characteristicsf the vessels, but that may influence catch rates (Maynou et al.,003). The initial model applied contains all factors, consideringear, Month and Vessel as factor, while TL and GT as variables:PUE ∼ Year + Month + TL + GT + Vessel.
The GLM was conducted on the total CPUE (kg · vessel−1 · day−1)s well as the CPUE of each target species (Mullus spp., M. mer-uccius, N. norvegicus and A. antennatus) in their respective métier.
hen the data was asymmetric, log transformation was made toorrect the extreme data and the constant K was added to the catchate to account for zero observation, where: K is 10% of the meanPUE. For each case, i.e., total CPUE and by each target species,he best model were fitted with a stepwise selection procedurey exact Akaike Information Criterion (AIC; Akaike, 1974), and fac-ors that were not significant were eliminated from the model. TheIC determines between adding or excluding each variable, creat-
ng a balance between the variability explained by each factor andhe degrees of freedom introduced in the model (Akaike, 1974).fter the models were fitted, the significance of each factor wasnalysed using F-values. Finally, we derived calibration coefficientsy back-transforming the parameter estimates (Quinn and Deriso,999) and transformed CPUE data by dividing the raw CPUE by theppropriate coefficient.
.4. Analysis of variance
To analyse the biological and economic effect of closure at the
ome/closed port, data of five years, where the closure occurred inarly and late summer, were selected for the analysis (Table 2), inhich two weeks before and two weeks after the closure were used.
or the economic effect at the neighbour/open port, two weeks
15 September to 15 October June September 15 January to 15February + October
June January June September
before, two weeks during and two weeks after the closure data werecompared. Analysis of Variance (ANOVA) was used to test for signif-icant differences in total standardized CPUE (kg · vessel−1 · day−1)and standardized CPUE of target species by métier (biologicaleffect), and first sale price of target species (euro · kg−1) at homeand neighbour/open port (economic effect) (Underwood, 1997).The experimental design for the biological analysis consisted ofthree factors: Closure (fixed); Season (fixed and orthogonal); andYear (random and orthogonal). The same experimental designwas used for the economic analysis, replacing the factor Seasonby the factor Port. An even numbers of samples were randomlyselected to maintain balanced data within each level of the fac-tors considered in the experimental design. However, métiers areknown to exhibit seasonality, in many occasions “disappearing” insome years (during the studied two weeks before and after theclosure). Therefore, levels number of factor Year and minimumsamples used to balance the model varied (Table 3). The temporalfactor Year was considered as orthogonal to separate the inter-annual variations from the effect of the closure. Factor Season wasused to separate the effect of season from closure, while factorPort accounted for the relation between both ports and first saleprice of target species. When the ANOVA F-test was significant,post hoc analyses were conducted using Student–Newman–Keuls(SNK) multiple comparisons (Underwood, 1981). Before ANOVAanalysis, Cochran’s test was used to test for homogeneity of vari-ance (Cochran, 1951). When significant heterogeneity was found,the data were transformed by
√x + 1 or ln(x + 1). When transfor-
mations did not remove heterogeneity, analyses were performedon the untransformed data, with the F-test ˛-value set at 0.01(Tables 5 and 6), since ANOVA is more restricted to departures fromthis assumption, especially when the design is balanced and con-tains a large number of samples/treatments (Underwood, 1997). Allanalysis (ANOVA and GLM) were conducted by R statistical comput-ing software (R Development Core Team, 2010) and the R’s packageGAD (Sandrini-Neto and Camargo, 2011).
3. Results
3.1. Data standardization
The GLMs were able to separate the percentage of data variabil-ity that do not account for abundance. The variability explainedby the model were between 27.51% and 55.20% for total CPUE andtarget species CPUE of Red mullet, European hake and Norway lob-
306 M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317
e locat
shfwCaRtaw
Fig. 1. Map of the study area (SW Mediterranean) showing th
ter métiers (Table 4). The factor Vessel contributed to separate theighest percentage of deviance in CPUE in most cases (e.g. 38.30%
or M. merluccius CPUE). In addition, factors Month and Year alsoere highly significant in most cases (e.g., 15% for Norway lobster
PUE), which clearly captured the temporal variability in the catch-bility of the target species. In contrast, the explained variance in
ed shrimp métier was about 23 to 33% (Table 4). This suggestshat factors other than the used variables cause most of the vari-bility within the CPUE data. In this métier, technical factors asell as Vessel account for the most (i.e., 31% for the total CPUE) of
ion of the two trawling ports La Vila Joiosa and Dénia (Spain).
the explained variability. The models within the last 5 AIC valuesof the best model, in each case, are reported in Appendix A. Alsothe resulting coeffecients used for standardization are reported inAppendices B and C.
3.2. Biological effect
In general, trends in CPUEs were higher after the closure, exceptfor the Red shrimp métier. For total landings and Mullus spp. ofRed mullet métier, significant higher CPUEs were observed after
M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317 307
Table 3Number of samples and levels per factor used in analysis of variance (ANOVA). Dash (–) indicates that the factor was not used in the analysis, because of the lack of data tobalance the model.
The analysis Levels per factor Number of samples
Métier Closure Season or Port Year Random samples per level Total samples
he closure in both seasons, early and late summer (Fig. 2a and b)Table 5).
Slightly increasing trends of CPUEs were observed after the clo-ure in total landings and M. merluccius of European hake métier,
ig. 2. Mean CPUE (kg · vessel −1 · day−1) and standard error of the total landings (left) oed shrimp, and target species (right): (b) Mullus spp., (d) Merluccius merluccius, (f) Nephrlosure.
2 25 3002 3 365 38 1140
in both seasons (Fig. 2c and d). For total landings, significant inter-annual variations were detected, while no effects were observedfor the closure (Table 6). M. merluccius CPUE showed significanttwo-way interactions between Closure and Year, as well as between
f the four métiers: (a) Red mullet, (c) European hake, (e) Norway lobster), and (g)ops norvegicus and (h) Aristeus antennatus, during two seasons before and after the
308 M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317
Table 4Analysis of deviance table for generalized linear models (GLMs) fitted to total CPUE and target species CPUE for the four métiers (from 2002 to 2011) in Dénia and La Vila Joiosa.Df.: degrees of freedom; Res. Df.: residual of degree of freedom; Resid. Dev.: residual deviance; Dev. ex (%): deviance explained; F:F value. Factors are arranged according tothe percentage of explained deviance.
Métier Model Df. Deviance Res. Df. Resid. Dev Dev. ex (%) F
Red mullet Total CPUE 27.51%NULL 15422 3868.448Vessel 61 599.9565 15339 2804.116 15.51% 53.80108***
eason and Year (Table 6). In SNK comparisons, significant higherPUE after the closure were detected in 2006 and 2007.
For Norway lobster métier, clear increasing trends werebserved after the closure at both total and target species levelsFig. 2e and f), but this difference was only significant for totalandings (Table 6).
On the contrary, decreasing trends of CPUEs in Red shrimpétier were observed after the closure mainly in early summer
Fig. 2g and h). At species level, A. antennatus CPUEs significantlyecreased after the closure in both seasons (Table 6). For the total
andings the two-way interaction between Closure and Season was
ignificant (Table 6). In SNK comparisons, significant lower CPUEsere obtained after the closure in early summer, while CPUEs in
ate summer were similar before and after the closure.
For the first sale price of Mullus spp. at the home/closed port, aslight decrease was observed in Dénia in contrast to a slight increasein La Vila Joiosa (Fig. 3a). In ANOVAs, the two-way interaction of Clo-sure and Port was significant (Table 5). In SNK comparisons, pricedecreased significantly after the closure in Dénia, while no differ-ences were detected in La Vila Joiosa. At neighbour/open port, adecreasing trend was observed by the closure at both ports (Fig. 3b).In ANOVAs, the three-way interaction was significant (Table 5),where the price in La Vila Joiosa was significantly higher before the
closure (in Dénia) than during and after the closure in the 3 yearsstudied (Fig. 3b). In Dénia, the same differences were detected butonly in 2010.
M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317 309
Table 5Results of analysis of variance (ANOVA) with 2 factors (C: closure; S: season) for biologic effect (the total CPUE of Red mullet métier and Mullus spp. CPUE). With 2 factors (C:closure; P: port) for economic effect (price at home/closed port) and with 3 factors (C: closure; P: port; Yr: year) for price at neighbour/open port. Df: degrees of freedom;MS: mean square; F: F value.
Sources of variation Df MS F F versus Sources of variation Df MS F F versus
Economic effect C 1 3.1 1.19 Residual C 2 67.67 11.67 C × YrP 1 192.93 74.40** Residual P 1 45 3.12 P × YrC × P 1 59.56 22.97** Residual Yr 2 58.86 12.98** ResidualResidual 36 2.59 C × P 2 48.06 2.84 C × P × YrTransform. -,a C × Yr 4 5.8 1.28 Residual
P × Yr 2 14.44 3.18 ResidualC × P × Yr 4 16.91 3.73* ResidualResidual 54 4.53Transform. -,a
a Indicates that there is no homogeneity of variance, the levels of significance being *p < 0.01; **p <0.001.- Dash (–) indicates that there is no transformation.* p < 0.05.
** p < 0.01.
rJstibdA
osihVbtLes
riwasadhVLHti
For M. merluccius, home/closed port prices showed a smalleduction after the closure in Dénia and similar prices in La Vilaoiosa (Fig. 3c). ANOVAs indicated that the interaction between Clo-ure and Year was significant (Table 6), showing higher price beforehe closure only in 2006. Inter-annual variation was detected as thenteraction between Port and Year was also significant. At neigh-our/open port, slight increase of prices was observed in Déniauring the closure in La Vila Joiosa and vice versa (Fig. 3d), althoughNOVA did not show any significant differences (Table 6).
For N. norvegicus, at home/closed port, similar mean prices werebserved in Dénia before and after the closure, in contrast to alight decrease in La Vila Joiosa (Fig. 3e). In ANOVAs, the three-waynteraction was significant (Table 6). Mean prices were significantlyigher in Dénia before the closure only in 2007 and 2010, while in Laila Joiosa, such differences were not significant (Fig. 3e). At neigh-our/open port, higher mean price in Dénia was observed duringhe closure in La Vila Joiosa (Fig. 3f). The opposite was evident ina Vila Joiosa, as prices decreased during the closure in Dénia. Noffect was detected in ANOVA for closure or port, while inter-annualignificant differences were present (Table 6).
Finally for A. antennatus, at home/closed port, a clear priceeduction was observed in Dénia after the closure, while a smallncrease was observed in La Vila Joiosa (Fig. 3g). In ANOVAs, there
ere significant two-way interactions between Closure and Year,s well as between Closure and Port (Table 6). Three years showedignificant lower mean price after the closure. Price also decreasedfter the closure in Dénia, while no significant differences wereetected in La Vila Joiosa (Fig. 3g). At neighbour/open port, a clearigher mean price in Dénia was observed during the closure in Laila Joiosa (Fig. 3h), while a small decreasing trend was detected in
a Vila Joiosa. The three-way interaction was significant (Table 6).igher mean price in Dénia was observed during, after and before
he closure in La Vila Joiosa, in all years (Fig. 3h). However, pricesn La Vila Joiosa did not show any effect by the closure in Dénia.
4. Discussion
The resumption of fishing activity, in both study ports, did notalways result in higher CPUE after the closure. Generally, increasingtrends were observed in Red mullet, European hake and Norwaylobster métiers at both total and target species CPUEs. However,the statistical analysis revealed significant differences only for totallandings and Mullus spp. CPUE of Red mullet métier, M. merlucciusCPUE (in two years) and the total landings of Norway lobster métier.On the contrary, Red shrimp métier showed a negative effect of sig-nificantly lower CPUE at both total landings (in early summer) andA. antennatus CPUE. In economic terms, market prices of the maintarget species have decreased or shown no changes after the clo-sure at home/closed and neighbour/open ports. The only exceptionwas the increased A. antennatus price in Dénia during the closurein La Vila Joiosa.
Standardized catch rates assumes that the total length, grosstonnage and individual vessel were able to separate a large per-centage of the variability of the data that not directly attributable tovariations in abundance. While the year, month and vessel mainlyexplained the total variance percentages ranging between 23% (inthe case of Red shrimp metier) and 55% (in European hake metier).Nevertheless, these percentages are very high despite consider-ing daily CPUE data instead of monthly average. The percentagesobtained by the models reflect the suitability of the selected fac-tors. One way to decrease the variability of the data, and thereforeincrease the variability explained by the model, is to aggregate thedata on a temporary basis; for example, monthly (Goni et al., 1999;Maynou et al., 2003). For our case of study, such aggregation wasnot useful because we wanted to see differences in CPUE to the
lower time scale, so we decided to keep the analysis on daily basis.More research is needed on individual species, fishing technology,and the environment to determine what factors are most influen-tial in determining CPUE. Mahévas et al. (2011) observed that the
310
M.
Samy-K
amal
et al.
/ Fisheries
Research
172 (2015)
303–317
Table 6Analysis of variance (ANOVA) results with 3 factors (C: closure; S: season; Yr: year) for biologic effect (total CPUE by métier and target species CPUE), and with 3 factors (C: closure; P: port; Yr: year) for economic effect (the firstsale price at home/closed and neighbour/open ports) of the target species Merluccius merluccius, Nephrops norvegicus and Aristeus antennatus. Df: degrees of freedom; MS: mean square; F: F value.
Sources of variation European hake métier Norway lobster métier Red shrimp métier F versus
Total landings Merluccius merluccius Total landings Nephrops norvegicus Total landings Aristeus antennatus
a Indicates that there is no homogeneity of variance, the levels of significance being *p < 0.01; **p < 0.001.- Dash (–) indicates that there is no transformation.* p < 0.05.
** p < 0.01.*** p < 0.001
M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317 311
F s: (a,ba ed shn
itootmtdt
fitiTriCcrh
ig. 3. Mean first sale price (euro · kg−1) and standard error of the main target specientennatus of the four métiers: Red mullet, European hake, Norway lobster, and Reighbour/open port before, during and after the closure (right).
mportance of the skipper/crew experience effect is weaker thanhe technical effect of the vessel and its gear. Also reported that,ther information (e.g. length of headline, weight of otter boards,r type of groundrope) should be taken into account to improvehe modelled relationships between CPUE and the variables that
easure relative fishing power (Mahévas et al., 2011). Other fac-ors such as the swept area, doors open, travelled distance, gearepth may greatly influence catch rates. However, we did not havehese data, so it could not be included.
Fishing closures during spawning season can most likely reduceshing mortality if the spawning stock is more aggregated during
he spawning season than at any other time of the year; however,n a multi-specific fishery, this not the case of all target species.he spawning seasons of the four main target species are summa-ized in Table 1. Adjusting the closure to benefit all target speciesn multi-specific fisheries is difficult. Changes were observed in thePUE of three main target species, as Mullus spp. and M. merluc-
ius increased after the closure, while A. antennatus decreased. Aise in total landings of Red mullet and Norway lobster métiersas been also observed. The European hake M. merluccius repre-
) Mullus spp., (c,d) Merluccius merluccius, (e,f) Nephrops norvegicus and (g,h) Aristeusrimp in the two ports before and after the closure (left) and their mean prices at
sents a spawning period extending almost throughout the year thatis interpreted as an adaptive strategy to maximize the survival ofearly life cycle stages (Martin et al., 1999; Domínguez-Petit et al.,2008). This large-scale spawning period has favoured the observedbenefits. The reproduction of Red mullet Mullus spp. (both Mul-lus barbatus and Mullus surmuletus) in the western Mediterraneanoccurs mainly between spring and summer, almost exclusivelyfrom May to July (Relini et al., 1999; Voliani, 1999; Sieli et al.,2011) which also has favoured the observed increase. In contrast,the spawning period of A. antennatus occurs between the months ofMay to October, but is more intense in July and August (Demestre,1995; García Rodríguez and Esteban, 1999). Although the spawningperiod concurs with the closure in early summer, decreased catcheshave been observed.
Moreover, a short closure period (one month) cannot sub-stantially raise biomass due to an increase of the abundance ofindividuals; while it could be solely due to the increase of fish
weight. An explanation of the increased CPUE after one monthof closure is linked to rapid-growing species, observed in Mullusspp., M. merluccius (Pineiro and Sainza, 2003) and total landings
3 ries Re
op2nmtilciirbcMwSeFihtd(eaipat
dFaiPbabaia
afit
12 M. Samy-Kamal et al. / Fishe
f Norway lobster métier, where species such as Micromesistiusoutassou and Phycis blennoides are abundant (Samy-Kamal et al.,014). But these closures are too short to affect benthic commu-ities, where these processes, recruitment and growth take placeuch more slowly (Demestre et al., 2008). From another perspec-
ive, Bas (2006) argued that the effect of closure, reflected in anncrease of catches following resumption of the activity, is moreikely due to species’ behavior. The absence of fishing activityhanges the species’ behavior to move around freely, thus occupy-ng more places, having previously been accustomed to escapingnto marginal places during the fishing activity (Bas, 2006). Aftereopening the fishery, it is likely these species are more suscepti-le to being caught. This is more evident in limited fisheries, such asontinental-shelf métiers, especially for fishes (e.g., Mullus spp. and. merluccius) as they are more mobile than benthic communitieshich could be another explanation of the results obtained here.
imilar changes in fish behaviour after closure have been reportedlsewhere (Jupiter et al., 2012; Januchowski-Hartley et al., 2014).or instance, Jupiter et al. (2012) observed that the main observedmpact of the closure was the decline of large-bodied species. Thisas reflected in differences in community composition as well ashe prevalence of small herbivores species, as a consequence of aecline in territorial aggression from the removal of large speciesJupiter et al., 2012). Also they suggested that the substantial ben-fits to fisheries from closures, when occurred, can be removed in
very short time period through focused fishing efforts. Similarly,n the Gulf of Alicante, Samy-Kamal et al. (2014) have observedeaks in the fishing effort intensity in both August and Octoberre mainly associated with the reopening of the fishery after theemporal closure.
In the short term, a closure may also involve losses, such as thoseerived from a reduction in sales or loss of markets (Lleonart andranquesa, 1999). Prices are a function of supply and demand, andre influenced by fish size, species, consumer preferences, fish qual-ty and the catch quantity-demand function (McClanahan, 2010).rices of most target species decreased by the closure, which maye related to loss of market due to shortage in the supply after
month of closure. The economic effect of closure at the neigh-our/open port was not so evident, except for the increase of A.ntennatus price in Dénia during the closure in La Vila Joiosa. Thiss explained as closure might produce more demand on the markett Dénia where A. antennatus is the main target species.
According to the results obtained here, the closure has one
pparent benefit, which is the overall reduction of fishing effortor that specified period. The seasonal closure reduces the fish-ng effort (fishing pressure) about 8.33% (one month per year) ofhe annual effort, which is the only apparent benefit. Despite this,
Métier
Red mullet Total CPUE
Mullus spp. CPUE
European hake Total CPUE
search 172 (2015) 303–317
choosing the suitable timing to schedule closure during the spawn-ing season of the main target species is difficult; it would bring upsome biological positive effects on some target species (e.g., Mul-lus spp. and M. merluccius). Notwithstanding, these increases incatches after the closure are so far to compensate the lost catchesby stopping the activity for a whole month. In addition, closuresmore likely lead to unavoidable reduction in market prices of manytarget species. An effective management measure should be easilyapplied, as in the case of seasonal closure, and be able to ensureenough net contribution to the income of fishers. At the same time,an economically consistent closure should be applied without sub-sidies and be accepted by the fishing community; otherwise, it willconvert into a structural compensation and will lose its economicsense (Lleonart and Franquesa, 1999). Despite these reductions inprices, the wide acceptance of seasonal closure as a managementmeasure by the fishing community is mainly because it is subsi-dized by the administration. An alternative management measure,based on effort reduction in input-controlled Western Mediter-ranean multi-specific fisheries, could target a day per week (otherthan weekend) when market prices of target species are lower(Guillen and Maynou, 2014; Samy-Kamal et al., 2015b). This wouldresult in the double annual amount of effort reduction, as well min-imize the short-term negative economic effect of seasonal closureon market prices and therefore on fishers’ income. Also, it is moreacceptable by the fishing community to stop fishing for one daythan a whole month, and can be easily applied without additionalcosts of subsidies.
Acknowledgements
The authors acknowledge the cooperation of the staff at LaVila Joiosa and Dénia fishermen’s guilds for their important rolein collecting the data. M. Samy-Kamal was supported by a grantof the Spanish Agency for International Development Cooperation(AECID). We acknowledge Julie Smith and Kelly Bucas for languagerevisions. We also extend our thanks to Dr. Jesus Jurado-Molina forhis help on the R script. We are also grateful to the two reviewersand the editor whose comments greatly improved the manuscript.We thank the FAO for species drawings for figures 2 and 3.
Appendix A.
Model selection procedure for the total CPUE and target speciesCPUE by métier shows the last 5 values based on AIC: Akaike Infor-mation Criterion.
Year + GT + Vessel 51705Year + Month + GT 55080Final model: Year + Month + GT + Vessel 51003
ppendix B.
Estimates and standard error from a generalized linear model fitted to total CPUE of Red mullet, European hake, Norway lobster anded shrimp métiers of the Dénia and La Vila Joiosa fleet from 2002 to 2011, incorporating the main effects of year, month, individual vesselV), total length (TL), and gross tonnage (GT). Estimates express the difference between each level of the factors and the first level.
Red mullet European hake Norway lobster Red shrimp
Estimate St. error Estimate St. error Estimate St. error Estimate St. error
M. Samy-Kamal et al. / Fisheries Research 172 (2015) 303–317 315
ppendix C.
Estimates and standard error from a generalized linear model fitted to CPUE of target species Mullus spp., M. merluccius, N. norvegicusnd A. antennatus of the Dénia and La Vila Joiosa fleet from 2002 to 2011, incorporating the main effects of year, month, individual vesselV), total length (TL), and gross tonnage (GT). Estimates express the difference between each level of the factors and the first level.
Mullus spp. M. merluccius N. norvegicus A. antennatus
Estimate St. error Estimate St. error Estimate St. error Estimate St. error
Relini, G., Bertrand, J., Zamboni, A., 1999. Synthesis of the knowledge on BottomFishery Resources in Central Mediterranean (Italy and Corsica). Biol. Mar.
kaike, H., 1974. A new look at the statistical model identification. EEE Trans.Autom. Control 19, 716–723.
rendse, C.J., Govender, A., Branch, G.M., 2007. Are closed fishing seasons aneffective means of increasing reproductive output? A per-recruit simulationusing the limpet Cymbula granatina as a case history. Fish. Res. 85 (1), 93–100.
as, C. 2006. The Mediterranean Sea: living resources and exploitation.CIHEAM-IAMZ.
OE, 2013. Boletin Oficial del Estado. Ministerio de Agricultura, Alimentación yMedio Ambiente. Resolución de 27 de marzo de 2013, de la Secretaría Generalde Pesca, por la que publican los censos actualizados de las modalidades dearrastre de fondo, artes menores, cerco y palangre de fondo del caladeroMediterráneo.BOE n◦ 88, Sec. III. Pág. 27442.
ochran, W.G., 1951. Testing a linear relation among variances. Biometrics 7,17–32.
emestre, M., 1995. Aristeus antennatus (Decapoda: Dendrobranchiata). Mar. Ecol.Prog. Ser. 127, 57–64.
emestre, M., de Juan, S., Sartor, P., Ligas, A., 2008. Seasonal closures as a measureof trawling effort control in two Mediterranean trawling grounds: effects onepibenthic communities. Mar. Pollut. Bull. 56 (10), 1765–1773.
inmore, T.A., Duplisea, D.E., Rackham, B.D., Maxwell, D.L., Jennings, S., 2003.Impact of a large-scale area closure on patterns of fishing disturbance and theconsequences for benthic communities. ICES J. Mar. Sci.: J. Cons. 60 (2),371–380.
omínguez-Petit, R., Korta, M., Saborido-Rey, F., Murua, H., Sainza, M., Pineiro, C.,2008. Changes in size at maturity of European hake Atlantic populations inrelation with stock structure and environmental regimes. J. Mar. Syst. 71 (3),260–278.
AO, 2011. Review of the state of world marine fishery resources. In: FAO Fisheriesand Aquaculture Technical Paper No. 569. FAO, Rome, 334 pp.
arcía Rodríguez, M., Esteban, A., 1999. On the biology and fishery of Aristeusantennatus (Risso, 1816), (Decapoda, Dendrobranchiata) in the Ibiza channel(Balearic Islands, Spain). Sci. Mar. 63 (1), 27–37.
oni, R., Alvarez, F., Adlerstein, S., 1999. Application of generalized linear modelingto catch rate analysis of Western Mediterranean fisheries: the Castellón trawlfleet as a case study. Fish. Res. 42 (3), 291–302.
uerra-Sierra, A., Sánchez-Lizaso, J. L. 1998. Fundamentos de Explotacion deRecursos Vivos Marinos. Ed. Acribia, Zaragoza, Spain. 249 pp.
uillen, J., Maynou, F., 2014. Importance of temporal and spatial factors in theex-vessel price formation for red shrimp and management implications. Mar.Policy 47, 66–70.
orwood, J.W., Nichols, J.H., Milligan, S., 1998. Evaluation of closed areas for fish
stock conservation. J. Appl. Ecol. 35 (6), 893–903.
anuchowski-Hartley, F.A., Cinner, J.E., Graham, N.A., 2014. Fishery benefits frombehavioural modification of fishes in periodically harvested fisheries closures.Aquat. Conserv.: Mar. Freshw. Ecosyst. 24 (6), 777–790.
Jupiter, S.D., Weeks, R., Jenkins, A.P., Egli, D.P., Cakacaka, A., 2012. Effects of a singleintensive harvest event on fish populations inside a customary marine closure.Coral Reefs 31 (2), 321–334.
Lleonart, J., Franquesa, R. 1999. La veda como medida de gestión. 4a Reunión delForo Científico sobre la pesca espanola en el Mediterráneo Málaga, junio.
Mahévas, S., Vermard, Y., Hutton, T., Iriondo, A., Jadaud, A., Maravelias, C.D.,Punzón, A., Sacchi, J., Tidd, A., Tsitsika, E., Marchal, P., Goascoz, N., Mortreux, S.,Roos, D., 2011. An investigation of human vs. technology-induced variation incatchability for a selection of European fishing fleets. ICES J. Mar. Sci. 68,2252–2263.
Martin, P., Sartor, P., García-Rodríguez, M., 1999. Exploitation patterns of theEuropean hake Merluccius merluccius, red mullet Mullus barbatus and stripedred mullet Mullus surmuletus in the western Mediterranean. J. Appl. Ichthyol.15 (1), 24–28.
Maynou, F., Sardà, F., Tudela, S., Demestre, M., 2006. Management strategies for redshrimp (Aristeus antennatus) fisheries in the Catalan Sea (NW Mediterranean)based on bioeconomic simulation analysis. Aquat. Living Resources 19 (2),161–172.
McClanahan, T.R., 2010. Effects of fisheries closures and gear restrictions on fishingincome in a Kenyan coral reef. Conserv. Biol. 24 (6), 1519–1528.
Maunder, M.N., Punt, A.E., 2004. Standardizing catch and effort data: a review ofrecent approaches. Fish. Res. 70, 141–159.
Maynou, F., Demestre, M., Sanchez, P., 2003. Analysis of catch per unit effort bymultivariate analysis and generalised linear models for deep-water crustaceanfisheries off Barcelona (NW Mediterranean). Fish. Res. 65, 257–269.
Murawski, S.A., Wigley, S.E., Fogarty, M.J., Rago, P.J., Mountain, D.G., 2005. Effortdistribution and catch patterns adjacent to temperate MPAs. ICES J. Mar. Sci.62, 1150–1167.
Pineiro, C., Sainza, M., 2003. Age estimation, growth and maturity of the Europeanhake (Merluccius merluccius (Linnaeus, 1758)) from Iberian Atlantic waters.ICES J. Mar. Sci.: J. Cons. 60 (5), 1086–1102.
Pipitone, C., Badalamenti, F., D’Anna, G., Patti, B., 2000. Fish biomass increase aftera four-year trawl ban in the Gulf of Castellammare (NW Sicily, MediterraneanSea). Fish. Res. 48, 23–30.
Quinn II, T.J., Deriso, R.B., 1999. Quantitative fish dynamics. In: Biological ResourceManagement Series. Oxford University Press, New York, 542 pp.
R Development Core Team, 2010: R: A language and environment for statisticalcomputing. R Foundation for Statistical Computing, Vienna, Austria. ISBN3-900051-07-0, URL http://www.R-project.org/
Mediterr. 6 (Supl. 1), 1–868.Samy-Kamal, M., Forcada, A., Sánchez-Lizaso, J.L., 2014. Trawling fishery of the
western Mediterranean Sea: métiers identification, effort characteristics,landings and income profiles. Ocean Coast. Manag. 102, 269–284.
Interpretation Using Analysis of Variance. Cambridge University Press,Cambridge.
Ye, Y., 1998. Assessing effects of closed seasons in tropical and subtropical penaeid
M. Samy-Kamal et al. / Fishe
amy-Kamal, M., Forcada, A., Sánchez-Lizaso, J.L., 2015a. Short-term effect ofselectivity change in a trawling fishery in the Western Mediterranean. J. Appl.Ichthyol. 31, 265–275.
amy-Kamal, M., Forcada, A., Sánchez-Lizaso, J.L., 2015b. Daily variation of fishingeffort and ex-vessel prices in a western Mediterranean multi-species fishery:implications for sustainable management. Mar. Policy, http://dx.doi.org/10.1016/j.marpol.2015.08.015, In Press.
ánchez, P., Sartor, P., Recasens, L., Ligas, A., Martin, J., De Ranieri, S., Demestre, M.,2007. Trawl catch composition during different fishing intensity periods in twoMediterranean demersal fishing grounds. Sci. Mar. 71 (4), 765–773.
andrini-Neto, L., and Camargo, M.G. 2011. GAD: an R package for ANOVA designsfrom general principles. Available on CRAN.
ardà, F., 1991. Reproduction and moult synchronism in Nephrops norvegicus (L.)
(Decapoda, Nephropidae) in the Western Mediterranean: is spawning annualor biennial? Crustaceana, 186–199.
brana, M., Sartor, P., Belcari, P., 2003. Analysis of the factors affecting crustaceantrawl fishery catch rates in the northern Tyrrhenian Sea (westernMediterranean). Fish. Res. 65 (1), 271–284.
search 172 (2015) 303–317 317
Shih, N.T., Cai, Y.H., Ni, I.H., 2009. A concept to protect fisheries recruits by seasonalclosure during spawning periods for commercial fishes off Taiwan and the EastChina Sea. J. Appl. Ichthyol. 25 (6), 676–685.
Sieli, G., Badalucco, C., Di Stefano, G., Rizzo, P., D’Anna, G., Fiorentino, F., 2011.Biology of red mullet, Mullus barbatus (L. 1758), in the Gulf of Castellammare(NW Sicily, Mediterranean Sea) subject to a trawling ban. J. Appl. Ichthyol. 27(5), 1218–1225.
Underwood, A.J., 1981. Techniques of analysis of variance in experimental marinebiology and ecology. Oceanogr. Mar. Biol.: Ann. Rev. 19, 513–605.
Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and
shrimp fisheries using a length-based yield-per-recruit model. ICES J.Mar. Sci.:J. Cons. 55 (6), 1112–1124.