© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Recruitment indices of European hake, Merluccius merluccius (Linnaeus 1758), in the Northwest Mediterranean based on landings from bottom-trawl multispecies fisheries
a Instituto Español de Oceanografía, Centro Oceanográfico de Baleares Apartado 291, 07080 Palma de Mallorca, Spain
b University of Michigan School of Natural Resources and the Environment, 138 Museum Annex, 1109 N. University, Ann Arbor, MI, USA
c Instituto Español de Oceanografía Avenida de Brasil, 31, 28020 Madrid, Spain
d Instituto de Ciencias del Mar, CMIMA–CSIC Paseo Marítimo Barceloneta, 37–49, 08039 Barcelona, Spain
e Universitá di Pisa, Dipartimento di Scienze dell'Uomo e dell'Ambiente Via A, Volta 6, Pisa, Italy
*Correspondence to R. Goñi: tel: +34 971 401561; fax: +34 971 404945. e-mail: raquel.goni{at}ba.ieo.es.
Temporal and spatial variation in Merluccius merluccius recruitment in the Northwest Mediterranean is examined, and recruitment indices are derived from monthly M. merluccius catch rates of four bottom-trawl fleets operating in Spanish and Italian waters during the period 1991–1999. Where M. merluccius catches were not recorded by size category, multivariate techniques were applied to species proportions in order to identify catch records most representative of recruit abundance. Selected catch rates were analysed by generalized linear models (GLMs) to estimate recruitment indices. The GLMs explain large proportions of the variation in recruit abundance (50–84%) and indicate significant annual and seasonal variation in recruitment strength. Vessel was by far the most important factor affecting catch rates of recruits, stressing the need to account for vessel characteristics when analysing commercial catch data for stock assessment. Seasonal patterns of recruitment reveal similarities among the study areas and interannual variations within areas. One major recruitment peak was identified in each area during either spring/summer or late summer/winter. In most areas, recruitment between 1991 and 1999 seemed to decline, but this could not be confirmed owing to high interannual variability. The coherence of annual fluctuations in recruitment indices with those of MEDITS surveys supports the present results.
Keywords: catch rates, generalized linear models, Merluccius merluccius, multivariate analysis, Northwest Mediterranean, recruitment trends
Received 25 November 2003; accepted 21 April 2004.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
The European hake, Merluccius merluccius, is widely distributed in the Northeast Atlantic and Mediterranean and is found from 30 to 700 m deep, with main concentrations between 50 and 200 m (Oliver and Massutí, 1995). The bathymetric distribution is size dependent and varies seasonally (Oliver and Massutí, 1995; Maynou et al., 2003). In the western Mediterranean, hake eggs and larvae are found on the shelf edge and upper continental slope, where adults concentrate for spawning (Recasens et al., 1998; Sartini et al., 2002; Olivar et al., 2003), and juveniles and sub-adults are distributed over the continental shelf (Zupanovic, 1968; Oliver and Massutí, 1995). Young of the year (recruits) are distributed widely over sandy and muddy shelf substrata, but preferred nursery areas have been identified in depths of 100–200 m (Recasens et al., 1998; Belcari et al., 2001; Orsi-Relini et al., 2002; Maynou et al., 2003). Recruitment into the fisheries takes place 8–10 months after spawning (Recasens et al., 1998) at fish total length, TL, of 10–13 cm (Demestre and Sánchez, 1998).
Hake spawn throughout the year, but peaks have been reported during autumn/winter in the northwest (Recasens et al., 1998; Lleonart, 2001), and during winter/spring in the northern Tyrrhenian Sea (Biagi et al., 1995; Lleonart, 2001). Spatial variation in spawning and also in larval growth results in seasonal recruitment peaks that differ geographically (Table 1). M. merluccius recruitment studies in the Mediterranean are limited in their temporal and spatial coverage. Therefore, results from such studies tend to be specific for a particular year or location, and little is known about the driving forces. Recruitment windows linked to oceanographic processes have been associated with interannual variability in recruitment strength in the western Mediterranean (Oliver, 1993), but this hypothesis has not been tested. One of the limitations has been the lack of time-series of recruit abundance.
|
Traditionally, hake have been the target of demersal fisheries throughout their range for reasons of abundance and consumer appreciation. In the Mediterranean, M. merluccius is caught in multispecies fisheries by bottom trawlers that operate over the continental shelf and slope. Such fisheries rely heavily on hake recruits and juveniles (Recasens et al., 1998; Orsi-Relini et al., 2002), because local markets have developed for undersize fish. The vulnerability of M. merluccius to trawling in the Mediterranean appears to decline with fish size owing to the deeper distribution of adults or escapement (Abella et al., 1997; Martin et al., 1999; Sartor et al., 2001a). Therefore, adult fishing mortality in the Mediterranean has been typically low, and this "refuge effect" on adults has been put forward as the most plausible explanation for the apparent sustainability of the M. merluccius fisheries there, despite the high fishing pressure (Oliver, 1993; Abella et al., 1997). Longline fisheries targeting adults started in the northwest in the 1980s, increasing the rate of exploitation of adults (Recasens et al., 1998; Lleonart, 2004), and landings of small M. merluccius in the same area declined markedly throughout the 1990s (Lloret and Lleonart, 2002).
Traditionally, stock assessment of M. merluccius in the Mediterranean has been hampered by the lack of consistent time-series of catch (landings, length-at-age) and effort. Consequently, assessments have been occasional and short term (Lleonart, 2004). Nevertheless, basic catch and effort data have recently been collected from many important Mediterranean fleets and are now available for analysis. In addition, fishery-independent surveys have been conducted throughout the Mediterranean since 1994 to provide annual synoptic indices of abundance for groundfish species vulnerable to trawling (MEDITS surveys; Bertrand et al., 1997; Orsi-Relini et al., 2002).
In this study, we investigate temporal and spatial variation of M. merluccius recruitment in the Northwest Mediterranean and derive recruitment indices based on newly available catch and effort data from commercial bottom-trawl fisheries. As Mediterranean fisheries rely on juveniles, assessing recruitment strength is of special relevance. Indices are deduced from monthly M. merluccius catch rates of four fleets operating in Spanish and Italian waters. This is the first integrated study of recruitment for the western Mediterranean based on commercial catch data. The main merits of the analysis are that it covers a relatively long period (1991–1999) and an extensive area. To validate the results, we compare the annual recruitment indices for the later years in the analysis with indices drawn from the MEDITS surveys.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Study area and characteristics of the fleets
Data for the analysis are from the Mediterranean groundfish trawl fleets of the harbours of Santa Pola, Castellón, and Barcelona, located along the Spanish Mediterranean coast, and from Porto Santo Stefano on the northwest coast of Italy (Figure 1).
|
The Porto Santo Stefano trawl fleet operates over a stretch of the continental shelf 90 miles long and 60 miles wide along the Tuscany coast. During the study period, 58 vessels participated in the fishery. Part of the fleet used the Italian "traditional trawl", characterized by a vertical opening of 1–2 m. This gear is operated by both small and large trawlers at depths from 50 to 650 m in M. merluccius nursery areas (Lleonart, 2001). Other vessels used the "French trawl", which has a vertical opening of 4–10 m and is used in shallower waters (<150 m; Reale et al., 1995) to target red mullet (Mullus spp.; Alvarez et al., 2001).
The fleet from Barcelona fishes off the northeast Iberian Peninsula on a relatively narrow stretch of the continental shelf about 10 miles wide and 20 miles long, and over the adjacent slope and canyons. At the beginning of the study period, 32 vessels participated in the fishery, but only 24 were recorded in 1999. The trawl used by this fleet has a vertical opening of 1–3 m.
The fleet from Castellón operates on a stretch of the continental shelf of the eastern Iberian Peninsula 30 miles wide and 40 miles long. During the study period up to 44 vessels participated in the fishery, 29 of which fished consistently throughout. The trawlnet used by that fleet has a vertical opening of 1–2 m. The great width of the continental shelf in the area determines fleet behaviour. Small vessels generally operate inshore at depths <60 m, medium vessels over the midshelf between 50 and 100 m deep, and the largest vessels over the shelf and slope up to 350 m deep. A two-month fishery closure to protect recruitment of M. merluccius and of other species is in place every year during spring and summer (Llorca and Tegedor, 1997).
Santa Pola harbours one of the largest trawl fleets in the Spanish Mediterranean. Vessels operate on an 18-mile-wide, 40-mile-long stretch of the continental shelf off the southeastern Iberian Peninsula. Up to 112 vessels participated in the bottom-trawl fishery during the study period, although only 50 fished regularly. The trawl used by these vessels has a vertical opening of 1–2 m. The spatial dynamics of this fleet is similar to that of Castellón.
Data
Data available for the analysis for each fleet cover the period 1991–1999, and consist of landings by vessel, the number of days the vessel operated in the fishery, and the vessel characteristics (gross registered tonnage, horsepower, length). Fleets in all harbours regularly operate daily on a 12-h schedule. Landings were recorded by species or by commercial mixed-species category and were aggregated by month. For Porto Santo Stefano we used data from landings of vessels using the traditional trawl. The size structure of the landings there was sampled 3–5 days per month and, although the database from the fleet contained records from 23 trawlers, most vessels were excluded from the analysis because they were not sampled systematically for the whole study period. Characteristics of the vessels selected for the study and the number of records used in the analysis by fleet are presented in Table 2. The size composition of the vessels used for the study by fleet is shown in Figure 2.
|
|
Categorization, landings, PCA, and cluster analysis
M. merluccius landings in Santa Pola and in Porto Santo Stefano were recorded by commercial size category, whereas landings in Castellón and Barcelona were aggregated. In Santa Pola the size categories (TL) were: very small (<12 cm), small (13–20 cm), medium (21–30 cm), and large (>30 cm). The two smaller categories were classified as recruits. In Porto Santo Stefano, the size categories were: small (<15 cm), medium (15–22 cm), large (22–40), and very large (>40 cm). Only the smallest group there was categorized as recruits.
For the fleets of Castellón and Barcelona, where the size distribution of the hake catch was not available, it was necessary to identify M. merluccius catch records representative of recruit abundance. Although hake trawl catches in the western Mediterranean are primarily recruits, low and variable proportions of older juveniles and adults are also taken (Abella et al., 1997; Sartor et al., 2001a). We followed an approach based on ordination and classification similar to that used by Murawski et al. (1983), Lewy and Vinther (1994), and Pelletier and Ferraris (2000) to define fishing strategies. In these studies, multivariate methods were applied to identify métiers/fishing strategies within multispecies fisheries using the species composition of commercial catch data.
The landings at Castellón were dominated by M. merluccius (21%), followed by Mullus spp. (17%; M. barbatus and M. surmuletus), reflecting notable activity over the midshelf. Trachurus trachurus and Lophius spp. (L. piscatorius and L. budegasa), characteristic of deeper waters of the shelf, made up another 15% of the landings, and coastal species such as Murex brandaris, Squilla mantis, and Sepia officinalis together another 21%. In the Barcelona landings, M. merluccius was fourth in importance by weight (10%), after Aristeus antennatus (12.5%), Micromesistius poutassou (11.5%), and Octopus vulgaris (11%). This composition reflects fishing activity over the outer shelf and slope. Typical shelf species such as Mullus spp. were also present in significant quantities, but coastal species constituted a smaller proportion of the landings than at Castellón.
Principal component analysis (PCA) used a matrix of proportions of catches of selected species (or catch profiles) by month and vessel (Pelletier and Ferraris, 2000) as input data. Species were selected according to their consistency and relative importance in the catch, excluding occasional species and most species mixtures. In all, 16 species and 2 two-species groups (Lophius spp. and Mullus spp.) were used for Castellón and 18 species and 3 species groups (Mullus spp., Lophius spp., Liocarcinus spp.) for Barcelona.
Next, we selected the principal components that explained >85% of the variance and where additional components contributed little to the explained variance (Afifi and Clark, 1984; McGarigal et al., 2000). We used the score of the selected PCs as transformed variables to perform Agglomerative Hierarchical Clustering (AHC), and to classify the month–vessel observations into groupings reflecting similar fishing patterns (Pelletier and Ferraris, 2000). Clusters were built by successive pairwise agglomerations of elements, using the Euclidean distance as the metric of dissimilarity and Ward's minimum variance linkage as the clustering method (Afifi and Clark, 1984). Because there is no standard objective procedure to select an appropriate number of clusters (McGarigal et al., 2000), and selection depends on the purpose of the analysis and cluster interpretability (Lewy and Vinther, 1994; Pelletier and Ferraris, 2000), alternative numbers of clusters were identified using the distance criterion (Afifi and Clark, 1984). Next, each cluster classification was evaluated in view of the known fishing strategies of each fleet, and the classification that produced the most realistic and parsimonious result was chosen. Finally, from the chosen cluster classification, the clusters for which M. merluccius catch rates were identified as representative of recruit abundance were selected. The criteria used for cluster selection were the bathymetric distribution of the species contained in each cluster, the size of the vessels, and the seasonality of the fishing activities.
Estimation of recruitment indices
To estimate recruitment indices for each area, we used GLMs (McCullagh and Nelder, 1989; Hilborn and Walters, 1992; Goñi et al., 1999). Monthly catch rates of M. merluccius recruits in each fleet were modelled as a function of the vessel conducting the operation, month and year being explanatory variables. The vessel represents the fishing power as well as vessel-specific characteristics, such as fishing aids and skipper ability or knowledge. All variables were introduced as factors. In analysing Castellón data, a variable representing fishing closure was also introduced as a two-level factor (Goñi et al., 1999). Gamma probability distributions were used in the analyses, because the frequency distributions of the catch rates were skewed, and the variances were proportional to the square of the means (McCullagh and Nelder, 1989). The gamma density function is expressed within GLMs in terms of the mean µ and the parameter 
that determines the shape of the distribution. The parameter , assumed constant for all observations, is 
–2, where is the coefficient of variation. The gamma variance V(µ)=µ2/ and a logarithmic-link g(µ) function were used to relate the expected catch rates to the predictors according to the following GLM:
|
|
is the catch rate obtained by a reference vessel per month (January for all fleets) and year (1991 for Santo Stefano and Castellón, 1992 for Barcelona and Santa Pola). 
ve is the efficiency of vessel ve relative to the reference vessel, 
m the recruit abundance in month m relative to January, 
y the recruit abundance in year y relative to the reference year, and 
vemy represents the deviation between the observed catch rate and the expected value for vemy. Analysis of deviance was performed to evaluate the significance of explanatory variables by comparing models while excluding one term at a time. Interactions between year and month were also included to investigate departures from general seasonal recruitment patterns. Analyses were performed, applying routines contained in the S-Plus computing environment (Becker et al., 1988).
Comparison with MEDITS indices; spatio-temporal patterns of recruitment
Annual recruitment trends derived from GLMs for each fishery were compared with trends of selected indices from the MEDITS surveys taken from Orsi-Relini et al. (2002). MEDITS hake abundance indices are expressed as catch rates of M. merluccius in kg km–2 of swept area, and are derived by region and depth stratum. For comparison with the indices derived here, we selected MEDITS indices from the 100–200-m stratum, where recruits concentrate. On the basis of geographic proximity, MEDITS indices from the northern Tyrrhenian Sea were compared with those from the Porto Santo Stefano fleet, from the Catalan Sea with indices from the Barcelona fleet, and from the Alicante region with indices from the Castellón and Santa Pola fleets.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Catch rates representative of recruit abundance from the Castellón and Barcelona fleets
Results of the PCA of Castellón data expressed as a bi-plot of the first two PCs (not shown) indicated three main groups of fishing strategies, one dominated by M. merluccius, a second by coastal species (e.g. M. brandaris), and a third by Mullus spp., though the bulk of the observations fell in the middle of these three groups. The influence of the main species on the first five PCs (loadings) for Castellón is presented in Figure 3. More than 85% of the variability in the data was explained by the first six PCs, so their scores were used as input for the AHC analysis. Results of the stepwise AHC are represented as a dendrogram in Figure 4, which illustrates the distances where the clusters were combined. Observations were assigned to three, four, and six clusters depending on alternative selection of distances. The six-cluster classification was retained on the basis of the species composition and fishery characteristics of the three assignments. The average species composition of the catch in each of the six clusters, together with auxiliary information to interpret the clusters as fishing strategies, is summarized in Table 3. The species composition of cluster 1 is indicative of year-round fishing in shallow coastal waters, whereas that of cluster 2 represents fishing targeting Mullus spp. over the inner and midshelf during autumn. The species composition of cluster 3, with reduced representation of coastal species, is indicative of a fishery for midshelf species, including M. merluccius. The composition of cluster 4, dominated by M. merluccius and Lophius spp., and nearly devoid of coastal species, is indicative of fishing in deeper waters of the shelf. Clusters 5 and 6 denote fishing on the shelf edge and slope for M. merluccius and M. poutassou, respectively. Based on these cluster characteristics, we selected data from clusters 3 and 4 as most representative of M. merluccius recruitment, and for the GLM analysis, 1427 records from vessels that participated consistently in such fishing throughout the time-series.
|
|
|
Results of the PCA for the Barcelona fleet expressed as a bi-plot of the first two PCs (not shown) revealed three main groups, one characterized by A. antennatus, the second by M. poutassou, and the third by coastal species (e.g. Liocarcinus spp.). Again, though, the bulk of the observations fell in the middle of these three groups. For the AHC analysis of the Barcelona fleet, we used the scores of the first six PCs, which explained 87% of the variability in the data. From the resulting dendrogram (not shown), observations were assigned to three, four, and five clusters, but classification into four clusters best represented the fishing strategies of the fleet. The average species composition by cluster and auxiliary information to interpret the clusters as fishing strategies are summarized in Table 3. The species composition in cluster 1 is indicative of year-round fishing over the midshelf and outer shelf/slope, and that of cluster 2, with its high proportion of coastal species, fishing activity over the inner and midshelf. Cluster 3 has a variable species composition, reflecting mixed fishing on the continental shelf with a marked seasonal component. The strong presence of A. antennatus in cluster 4 indicates fishing over the slope and in canyons. Based on these cluster characteristics, data from clusters 1 and 2 are considered most representative of M. merluccius recruitment. After eliminating observations from vessels with sporadic representation in those clusters, the data set for the GLM recruitment analysis contained 1221 records.
Spatio-temporal variation in recruitment strength
Results from the GLM analyses of recruit catch rates from the four fleets indicate significant annual variation in recruitment strength, and that vessel and month affected catch rates significantly. The results also indicate that seasonal patterns of recruitment varied significantly between years in the four areas. Analysis of deviance (Table 4) indicates that differences in catch rates between vessels of the four fleets were far more important than annual and monthly variations. The variation in catch rates between vessels is highest in fishery operations within the Santa Pola fleet, the fleet with the largest range of vessel sizes. However, even for Porto Santo Stefano, where the fleet is comparatively homogeneous in terms of vessel size (Figure 2), the percentage of deviance explained by the factor vessel is double that explained by the annual variation. This result stresses the need to account for characteristics of the fleet when analysing commercial catch data for estimating abundance indices. In addition, results for the Castellón fleet show that catch rates during the month after a fishing closure are significantly greater than when not preceded by the closure (Table 4). Overall, models including interactions explain between 50 and 84% of the variation in catch rates (Table 4).
|
Annual recruitment indices fluctuated coherently between 1991 and 1995 in all areas except for Porto Santo Stefano (Figures 5a, 6a, 7a, 8a). During that period, recruitment in 1993 was among the strongest in the series, but was followed by a decline in 1994 and by a record low in 1995. During later years, the annual patterns tended to differ. On the grounds of the Porto Santo Stefano fleet, indices show intermittent and minor fluctuations around the mean, with an overall decline after 1995.
|
|
|
|
Seasonal patterns of recruitment varied among the study areas, but two broad patterns emerged. For the Barcelona and Castellón fleets, recruitment peaked generally during spring or summer, with lows in late autumn/winter (Figures 6b, 7b), whereas at Santa Pola and Porto Santo Stefano, maximum recruitment was in late summer/autumn or autumn/winter, and minimum in late spring/summer (Figures 5b, 8b). However, there were departures from these dominant patterns (not shown) in some years in the four areas (significant year:month interaction; Table 4).
Comparison of GLM and MEDITS indices
Patterns in GLM recruitment indices for the areas in this study show striking similarities to MEDITS survey indices available since 1994 for the 100–200-m depth stratum (Figure 9). This is particularly true for indices from the fleets of Barcelona and Castellón, whose patterns are closely matched (r=0.85–0.89). However, for the Santa Pola fleet, there is a slight difference between series for the 1994 and 1997 indices, decreasing to lower levels in the GLM than in the MEDITS (r=0.75) estimates. On the grounds fished by the Porto Santo Stefano fleet, there are also differences in the relative magnitudes of the 1994 and 1995 indices, GLM estimates being higher than MEDITS estimates (r=0.50).
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
By applying data for M. merluccius in the Northwest Mediterranean, it has been possible to demonstrate the value of commercial bottom-trawl catch and effort data in combination with multivariate and GLM analyses to investigate recruitment patterns. The strong correspondence between recruit abundance indices derived from this study and those from fishery-independent surveys supports this statement. The advantages of this approach for generating abundance indices over the use of data from scientific surveys such as MEDITS are the more extensive spatial and temporal coverage and the lower cost. The economic advantage makes the approach more likely to produce consistent time-series, whereas survey time-series may well be interrupted when financial priorities change and budgets shrink. So far, most assessments of M. merluccius stocks in the Mediterranean have been occasional and localized owing to budgetary problems and logistic constraints (e.g. the wide distribution of stocks shared by more than one country, and the large number of fleets and landing harbours involved). Meanwhile, bottom-trawl catch and effort information has become available from many harbours since the early 1990s, and its quality and coverage are improving as the implementation of the EU Directive on fishery data collection (Council Regulation 2597/1995) proceeds (Council Regulation 1543/2000). Therefore, we strongly recommend continuing fishery data collection programmes and making use of the data for estimating abundance indices of the main bottom-trawl target species.
Discarding practices could affect the present results and are potentially a major shortcoming of the use of catch and effort data to analyse M. merluccius recruitment. Mediterranean bottom-trawl fisheries are multispecific (Carbonell et al., 1997), a small number of high-value species, notably M. merluccius, dominating the landings (Alvarez et al., 1999). Whereas discards of M. merluccius in the three Spanish fisheries included in the study are negligible (Carbonell et al., 1997; Martin et al. 2001), at Porto Santo Stefano the discard rates of young of the year may be significant, though fortunately of little consequence to our analysis because landings and catches of recruits are correlated (Sartor** et al., 2001b; Martin et al., 2001). Another potential problem for this type of analysis is that the species composition of the landings used to identify fishing strategies may have been affected by discarding practices. In general, discard volumes vary depending on market conditions and fishery regulations. Here, we confronted the problem by selecting for the analysis species with a recurring presence in the landings that are usually the most valuable and suffer little or no discarding.
Our approach to identifying catch rates representative of recruit abundance can pose limitations. Although greatest concentrations of M. merluccius recruits are usually in the 100–200-m depth range (Orsi-Relini et al., 2002; Maynou et al., 2003), the distribution of recruits and juveniles over the shelf is wide and changes seasonally (Oliver and Massutí, 1995; Recasens et al., 1998; Maynou et al., 2003), and adults may at certain times of the year occupy the areas dominated by recruits (Zupanovic, 1968; Demestre and Sánchez, 1998; Recasens et al., 1998). As a result, selected records could contain variable proportions of non-recruit M. merluccius. The level of aggregation of our data into monthly activities by vessel also limits ability to resolve separation of fishing strategies, because vessels can operate in different areas during a single month. Nevertheless, M. merluccius recruits are targeted and dominate the catches of the Castellón and Barcelona fleets, so we are confident that the selection process increases the dominance further, and that the patterns observed are attributable to fluctuations in recruitment. The coherence of annual fluctuations of the recruitment indices derived here with those of the MEDITS surveys supports this claim.
The GLMs explained large parts of the variation in recruit catch rates of the four fisheries in the analysis (up to 84%). The advantage of this modelling over similar approaches to model catch rate variation is that the actual vessel was used as a factor representing fishing power (Large, 1992). In other studies, power had to be represented by specific characteristics of the vessels, such as gross registered tonnage (e.g. Kimura, 1981; Goñi et al., 1999). Including the vessel as a factor permitted accounting for other vessel-related characteristics that affect fishing power, such as fishing and navigation aids, and skipper ability (Cooke and Beddington, 1984; Hilborn and Walters, 1992).
Agreement of the results herein with previous studies on timing of recruitment peaks in the western Mediterranean depends on geographic location. M. merluccius recruitment in the region has been reported as year-round owing to the prolonged spawning period of the species (Orsi-Relini et al., 1989; Biagi et al., 1995; Recasens et al., 1998). Nevertheless, studies in most areas have identified one, and sometimes two, recruitment peaks in spring and/or autumn, the latter often of lower intensity (Table 1). It has been hypothesized that the presence of one major seasonal recruitment pulse, whose timing varies geographically, is due to the existence of the so-called "recruitment window" driven by local environmental factors, rather than by the presence of spawning peaks (Oliver, 1993). However, that hypothesis has not been verified owing to lack of data at the relevant temporal and spatial scales. During this study, we identified recruitment peaks either in spring/summer, as for the Barcelona and Castellón fleets, or in late summer/winter, as for the Porto Santo Stefano and Santa Pola fleets. For Castellón, the closure, which takes place in spring or summer, delays the arrival of the spring peak.
The results derived here show coherence of annual recruitment trends for the whole time-series only between Santa Pola and Castellón, the two southernmost ports of the study. In contrast with this result, Orsi-Relini et al. (2002) found coherence of M. merluccius recruitment patterns at a much larger spatial scale (i.e. Italy and Greece). Circulation in the western Mediterranean, with water flowing anticlockwise along the continental slope (Millot, 1999), should provide certain homogeneity of oceanographic conditions throughout the study area. We believe that the lack of large-scale coherence in the recruitment trends presented here results from mesoscale variability of the Northern Current in the western Mediterranean during winter, a main spawning period of M. merluccius (Lleonart, 2001). In winter, the Northern Current, which originates in the Ligurian Sea and flows along the continental slope, continuing south of the Balearic Sea, develops intense mesoscale meanders with amplitudes of 10–100 km, and phase speeds of 10–20 km day–1, which induce great variability (Millot, 1999). This has the potential to introduce a high degree of heterogeneity in the spatial pattern of survival and recruitment of M. merluccius in the region. Sánchez and Gil (2000) demonstrated for the Bay of Biscay a close association of M. merluccius larvae and these anticyclonic mesoscale structures, with recruitment areas close to locations characterized by upward motions of nutrient-rich deeper water stemming from variations in the vorticity field of the mesoscale eddies.
Results herein show a possible downward trend of M. merluccius recruitment during the study period, except for the Castellón fleet. This downward pattern is consistent with reports from other studies in the Northwest Mediterranean that have indicated declining trends in M. merluccius catches during the past decade (Lloret and Lleonart, 2002) and suggestive of growth-overfishing (Oliver, 1993; Lleonart and Maynou, 2003) and, more recently, recruitment-overfishing (Lleonart, 2004). The possible greater stability of recruitment indices in the Castellón fleet suggests that its two-month fishing closure established since 1991 could have a positive effect on the stock, although Martin (1995) argues that the immediate effect of the closure would only be to concentrate catches in the months following the closure.
A strategy including temporal fishing closures (delaying exploitation of incoming recruits, such as in Castellón), and spatial closures of nursery areas, which appear spatially stable (Sánchez and Gil, 2000), would seemingly have a beneficial impact on the status of western Mediterranean hake stocks. Further, the catch rates of bottom trawlers can be used by managers of the fleets to monitor hake recruitment in areas subject to different management and stock recovery regimes.
|
Acknowledgements |
|---|
Data from Porto Santo Stefano were provided by the Dipartimento di Scienze dell'Uomo e dell'Ambiente (University of Pisa), data from Barcelona by the Fishermen's Association, and data from Castellón and Santa Pola by the Instituto Español de Oceanografía. We thank M. Cruz Iglesias for editorial work, Pedro San Rafael for help with ancillary data from Castellón, and two anonymous referees for their valuable comments. The study was supported by funding provided by the European Commission, DG Fisheries, under project contract 98/053 "Factors affecting catch rates of NW Mediterranean trawl fleets and derivation of standardized abundance indices".
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Abella A., Caddy J.F., Serena F. (1997) Declining natural mortality with age and fisheries on juveniles: a Mediterranean demersal fishery yield paradigm illustrated for Merluccius merluccius. Aquatic Living Resources 10:257–269.[CrossRef][Web of Science]
Afifi A.A. and Clark V. (1984) Computer Aided Multivariate Analysis(Lifetime Learning Publications, Blemont, California) 458 pp.
Alvarez F., Alemany A., Ferrandis E. (1999) Modelling the relationship between fishing effort and fishing mortality in western Mediterranean trawl fleets. The case of hake Merluccius merluccius and striped red mullet Mullus surmuletus in Balearic Islands. Final Report, EU DGXIV, contract 96/025. 82 pp.
Alvarez F., Goñi R., García M., Adlerstein S., Sbrana M., Belcari P., Viva C., Reale B., Sartor P., Sánchez P., Demestre M., Recasens L., Lleonart J., Maynou F. (2001) Factors affecting catch rates of NW Mediterranean trawl fleets and derivation of standardized abundance indices. Final Report, EU DGXIV, contract 98/053. 102 pp.
Becker R., Chambers J., Wilks J. (1988) The New S Language. A Programming Environment for Data Analysis and Graphics(Wadsworth & Brooks/Cole Advanced Books & Software, Pacific Grove, California) 702 pp.
Belcari P., De Ranieri S., Reale B., Sartor P., Sbrana M., Viva C. (2001) Spatial distribution and seasonal concentration of European hake's juveniles, Merluccius merluccius, (L. 1758), in the North Tyrrhenian Sea. Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 36:238.
Bertrand J., Gil de Sola L., Papaconstantinou C., Relini G., Souplet A. (1997) An international bottom trawl survey in the Mediterranean: the MEDITS programme. ICES Document, CM 1997/Y: 03. 16 pp.
Biagi F., Cesarini A., Sbrana M., Viva C. (1995) Reproductive biology and fecundity of Merluccius merluccius (Linnaeus, 1758) in the northern Tyrrhenian Sea. Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 34:23.
Carbonell A., De Ranieri S., Martin P. (1997) Discards of the western Mediterranean trawl fleets. Final Report, EU DGXIV, contract MED 94/027. 150 pp.
Cooke J.G. and Beddington J.R. (1984) The relationship between catch rates and abundance in fisheries. IMA Journal of Mathematics Applied to Medicine and Biology 1:391–405.
Demestre M. and Sánchez P. (1998) Spatio-temporal distribution of the European hake Merluccius merluccius off Catalan coast (northwestern Mediterranean). Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 35:2420–421.
Goñi R., Alvarez F., Adlerstein S. (1999) Application of generalized linear modelling to catch rate analysis of the western Mediterranean: the Castellón trawl fleet as a case study. Fisheries Research 42:291–302.[CrossRef][Web of Science]
Hilborn R. and Walters C. (1992) Quantitative Fisheries Stock Assessment. Choice, Dynamics and Uncertainty(Chapman & Hall, London) 570 pp.
Kimura D.K. (1981) Standardized measures of relative abundance based on modelling log (CPUE), and their application to Pacific ocean perch (Sebastes alutus). Journal du Conseil International pour l'Exploration de la Mer 39:211–218.
Large P.A. (1992) Use of a multiplicative model to estimate relative abundance from commercial CPUE data. ICES Journal of Marine Science 49:253–261.
Lewy P. and Vinther M. (1994) Identification of Danish North Sea trawl fisheries. ICES Journal of Marine Science 51:263–272.
Lleonart J. (2001) Impact of the fishery and environment on Merluccius merluccius recruitment in the northwestern Mediterranean. (Co-ordinator) Final Report, EU DGXIV, contract 97-3522. 672 pp.
Lleonart J. (2004) A review of Mediterranean shared stocks, assessment and management. In Payne A.I.L., O'Brien C.M., Rogers S.I. (Eds.). Management of Shared Fish Stocks(Blackwell, Oxford) pp. 113–130 367 pp.
Lleonart J. and Maynou F. (2003) Fish stock assessment in the Mediterranean: state of the art. Scientia Marina 67:37–49.[Web of Science]
Llorca M. and Tegedor J. (1997) La pesca en la provincia de Castellón, 1985–1996. Un estudio científico, técnico y sociológico. Publicaciones de la Diputación de Castellón. 298 pp.
Lloret J. and Lleonart J. (2002) Recruitment dynamics of eight fishery species in the north-western Mediterranean Sea. Scientia Marina 66:77–82.[Web of Science]
Martin P. (1995) Impact of the two month closed season on the catches of hake (Merluccius merluccius) and red mullet (Mullus spp.) in a restricted area of the Catalan coast (NW Mediterranean). Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 34:249.
Martin P., Carbonell A., Belcari P. (2001) Estimation of trawl discards in the western Mediterranean. The European hake Merluccius merluccius as case study. Final Report, EU DGXIV, contract 00/009. 135 pp.
Martin P., Sartor P., Garcia Rodriguez M. (1999) Exploitation patterns of the European hake Merluccius merluccius, red mullet Mullus barbatus and striped red mullet Mullus surmuletus in the western Mediterranean. Journal of Applied Ichthyology 15:24–28.[CrossRef]
Maynou F., Lleonart J., Cartes J.E. (2003) Seasonal and spatial variability of hake (Merluccius merluccius L.) recruitment in the NW Mediterranean. Fisheries Research 60:65–78.[CrossRef][Web of Science]
McCullagh P. and Nelder J. (1989) Generalized Linear Models 2nd edn (Chapman & Hall, London) 509 pp.
McGarigal K., Cushman S., Stafford S. (2000) Multivariate Statistics for Wildlife and Ecology Research(Springer, New York) 283 pp.
Millot C. (1999) Circulation in the western Mediterranean Sea. Journal of Marine Systems 20:423–442.[CrossRef][Web of Science]
Murawski S., Lange A., Sissenwine M., Mayo R. (1983) Definition and analysis of multispecies otter-trawl fisheries off the northeast coast of the United States. Journal du Conseil International pour l'Exploration de la Mer 41:13–27.
Olivar M.-P., Quílez G., Emelianov M. (2003) Spatial and temporal distribution and abundance of European hake, Merluccius merluccius, eggs and larvae in the Catalan coast (NW Mediterranean). Fisheries Research 60:321–331.[CrossRef][Web of Science]
Oliver P. (1991) Dinámica de la población de merluza (Merluccius merluccius, L.) de Mallorca (reclutamiento, crecimiento y mortalidad). PhD thesis, University of the Balearic Islands. 392 pp.
Oliver P. (1993) Analysis of fluctuations observed in the trawl fleet landings in the Balearic Islands. Scientia Marina 57:219–227.
Oliver P. and Massuti E. (1995) Biology and fisheries of western Mediterranean hake (M. merluccius). In Alheit J. and Pitcher T.J. (Eds.). Hake: Biology, Fisheries and Markets(Chapman & Hall, London) 15: pp. 181–202 Fish and Fisheries Series 478 pp.
Orsi-Relini L., Capparena M., Fiorentini F. (1989) Spatial–temporal distribution and growth of Merluccius merluccius recruits in the Ligurian Sea. Observations on the 0 group. Cybium 13:263–270.
Orsi-Relini L., Papaconstantinou C., Jukic-Peladic S., Souplet A., Gil de Sola L., Piccinetti C., Kavadas S., Rossi M. (2002) Distribution of the Mediterranean hake populations (Merluccius merluccius smiridus Rafinesque, 1810) (Osteichthyes: Gadiformes) based on six years monitoring by trawl surveys: some implications for management. Scientia Marina 66:Suppl. 2, 21–38.
Pelletier D. and Ferraris J. (2000) A multivariate approach for defining fishing tactics from commercial catch and effort data. Canadian Journal of Fisheries and Aquatic Sciences 57:51–65.
Reale B., Sbrana M., De Ranieri S. (1995) Population dynamics of Merluccius merluccius exploited by different trawl-nets in the northern Tyrrhenian Sea. Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 34:254.
Recasens L., Lombarte A., Morales-Nin B., Torres D.J. (1998) Spatial–temporal variation in the population structure of the European hake in the NW Mediterranean. Journal of Fish Biology 53:387–401.[CrossRef][Web of Science]
Sánchez F. and Gil J. (2000) Hydrographic mesoscale structures and poleward current as a determinant of hake (Merluccius merluccius) recruitment in the southern Bay of Biscay. ICES Journal of Marine Science 57:152–170.
Sartini M., Belcari P., De Ranieri S. (2002) Occurrence of eggs and larvae of European hake Merluccius merluccius (L, 1758) in the northern Tyrrhenian Sea. Biología Marina Mediterránea 9:358–365.
Sartor P., Recasens L., Viva C., Lleonart J. (2001) Analysis of the impact of the fishery on the adult population of European hake Merluccius merluccius in the North-western Mediterranean. Rapport Commission Internationale pour l'Exploration de la Mer Méditerranée 36:321.
Sartor P., Sartini M., Reale B., Sbrana M. (2001) Analysis of the discard practices in the Merluccius merluccius (L, 1758) bottom trawl fishery of the northern Tyrrhenian Sea. Biología Marina Mediterránea 8:1771–774.
Zupanovic S. (1968) Study of hake (Merluccius merluccius L.) biology and population dynamics in the central Adriatic. Studies and Reviews, General Fisheries Council Mediterranean 32:1–24.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||