© 2006 International Council for the Exploration of the Sea
A univariate and multivariate study of reef fisheries off northeastern Brazil
a Université de la Méditerranée, Centre d'Océanologie de Marseille UMR CNRS 6540, 13288 Marseille Cedex 9, France
b Departamento de Oceanografia, Universidade Federal de Pernambuco Recife, Pernambuco 50739-540, Brazil
*Correspondence to T. Frédou: Current address: Centro de Geociências, Universidade Federal do Pará, Campus Universitário do Guamá, Belém, Pará 66075-110, Brazil; tel: +55 91 3201 7983; fax: +55 91 3201 7609. e-mail: tfredou{at}ufpa.br.
This work was part of a programme to assess the potential living resources within the Exclusive Economic Zone (EEZ) of Brazil that collected information on the catch composition and biology of the main species within the zone (REVIZEE). It aimed to identify and, using statistical tools, to assess the factors that influence the dynamics of the demersal fishery, targeting the rocky, coral reef and associated sandy seabeds of the continental shelf. Within the reef fishery of northeastern Brazil, snappers (family Lutjanidae), in particular Lutjanus chrysurus, L. synagris, L. analis, L. jocu, and, to a lesser extent, L. vivanus, are the main artisanal catch and contribute most to the similarity between groups. Among the factors considered, the spatial effect (geopolitical state as a factor) appeared to be the strongest attribute in isolating groups. Of the technological factors, trip duration better discriminated the catch composition than fleet category. However, given some exceptions mainly linked to favourable strong wind, trip duration categories were usually related to fleet dynamics, because motorized boats generally undertake longer trips. Such characteristics are important for management decisions, because fleets are likely to exploit different stages of the life cycle of a fish as well as different species while operating in different geographical areas. The catch analysis was characterized by a "lutjanid community" typical of rock, coral, and coral-sand habitats, and it is clear that this community is dominant in Brazil as well as in the more usually quoted regions such as the Bahamas, Antilles and along the coast from Yucatan to Panama.
Keywords: artisanal fishery, coral reefs, Lutjanidae, multifleet, multigear
Received 15 February 2005; accepted 23 December 2005.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
One of the main purposes of a fishery management plan is to adapt catches to the potential of the exploitable resources (Hilborn and Walters, 1992). In a multispecies fishery, a situation common in tropical marine waters, regulations based on single-species assessments have been shown to have limitations (Polunin and Roberts, 1996); fisheries science has recognized that fish species do not exist in isolation from each other and that they are not harvested independently (Daan, 1987; Magnusson, 1995; Jennings et al., 2001). Technical interactions arise when gear comes into contact with stocks of different species, resulting in a mixed catch, and also when co-existing fleets exploit the same resource (Lucena et al., 2002).
Tropical ecosystems such as coral reefs contain many species living close together in small spatial areas (Longhurst and Pauly, 1987), and hard-bottom, demersal fishery dynamics and fishing impacts in tropical areas are better described multi-dimensionally (multispecies, multigear, multifleet, spatially and temporally). In the case of exploited fish, the identification and the quantification of factors influencing the dynamics of fishing activities and the structure of the community appear to be necessary prerequisites in developing a management framework. In turn, such a framework helps to predict the effects of, and hence to qualify and quantify, the variables that influence the fishery dynamics.
The study of fleet dynamics and fishing strategy through an analysis of the diversity of catch composition is important in fishery ecology (Hilborn, 1985). Such a line of research is particularly useful in the case of multispecies, multigear fisheries (Murawski et al., 1983; Biseau and Gondeaux, 1988). In fisheries that do not provide sufficient or accurate and reliable (or both) commercial data, analysis based on catch composition provides an objective and quantitative alternative means of classification (He et al., 1997). Many landings have missing biological or abiotic data, so such robust methods are valuable. Several studies have tried to characterize fleet dynamics using multivariate methods (Pauly, 1980; He et al., 1997; Pitcher et al., 1998; Preikshot and Pauly, 1998; Pelletier and Ferraris, 2000). Such techniques are useful in obtaining an integrated picture of the structure of the system, the factors that characterize the fishing activity, and an understanding of how fishers' techniques are adapted to the biological and ecological characteristics of target species. Moreover, multivariate approaches improve the definition of a typology of fishing activity in order to best define technological categories for fishing statistics.
Many community data sets show an a priori defined structure. In the fishery off northeastern Brazil, a vessel may carry more than one gear, and vessels with similar technological characteristics may operate on different grounds, so qualification and quantification of the factors that influence the catch are essential to developing a management plan. To date in the fishery, there have been no studies of fishing activity or its impact that can support development of a management plan to underpin sustainable exploitation. The programme REVIZEE represented a first attempt to constitute a comprehensive collection of information on the biota and the oceanographic conditions within the Brazilian Economic Exclusive Zone. This study aims to identify, describe, and quantify the factors that influence Holocenic coral reef formation (see Maida and Ferreira, 1997) and the hard-bottom, demersal fishery, referred to here as the reef fishery, off northeastern Brazil, using a univariate approach complemented by a multivariate one. In this context, the homogeneity of the Brazilian reef fishery in terms of spatial distribution (variations between states or in distance from shore, or both) and technological interactions was examined by looking at the catch composition by state, by gear when only one gear was used, by vessel, and by trip duration. The concept of "state" refers here not only to a geopolitical unit, but also to a geomorphological, environmental, and socio-cultural unit. The continental shelf varies by state as well as in terms of winds and currents (Frédou, 2004). Fishing communities have developed navigational and fishing skills that are adapted to their environment and that somehow match the geopolitical units. For example, when fishers from Ceará sail to remote fishing grounds (oceanic banks), they make use of trade winds and strong currents. In contrast, Pernambuco fleets mostly navigate by sight of coastal reference points, because the fishing grounds they use are close to the coast as a consequence of the narrowness of the continental shelf in their area.
Additionally, analysis of the reef fishery landed catch aims to characterize the target fish assemblage and the place of each species in the foodweb, leading to better understanding of fishing's effect on the ecosystem, which itself has implications for management.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Data were collected for the five years 1996–2000 within the REVIZEE framework, a local programme that aimed to gather information on the Brazilian marine ecosystem within the country's Exclusive Economic Zone (EEZ). Those data analysed here were collected at 10 landing sites of the artisanal, hard-bottom, demersal fishery, distributed along five states of the northeastern coast of Brazil: Ceará (CE), Rio Grande do Norte (RN), Pernambuco (PE), Alagoas, (AL), and Bahia (BA); (see Figure 1). Information on fishing operations (date, name of vessel, landing site, fishing ground, depth, time at sea, and moon phase) and vessel category (type, gear, horse power, and number of fishers) was obtained directly from fishers. Catches were identified where possible to species level. Species abundance, fish length, and sometimes fish weight were measured. Reef fishing takes place from shallow, fringing coral reefs to submerged reefs (Holocene reefs) over the continental slope. In all, 187 species were identified, but those considered as rare were excluded from the study. The criterion for inclusion in the analysis was for species to represent >1% of the total catch (by number) or >5% of the landings sampled. In addition, care was taken not to exclude species that would be common in one state and rare in others.
|
Abundance by state was analysed, but no extra species were added for the analysis. The data set used was 1667 landings and 60 400 individuals of 37 species (Table 1). These 37 species represented 86% of the total landings (Frédou, 2004).
|
Clarke and Warwick (1994) defined a framework for studying changes in assemblage structure. To them fish-assemblage analysis has three stages: representation of the assemblage, discrimination of site and conditions, and links between abiotic factors and the species assemblage. From a fishery point of view, analysing the catch determines those species that drive the fishing dynamics, discriminate fishing techniques or fishing conditions, and identify what factors influence catch composition.
Factor and data matrices
Four factors were considered relevant as descriptors of fishing tactics: state, vessel category, gear, and trip duration. The trip duration category is usually related to the vessel characteristics (means of propulsion and, if applicable, horse power, and storage and preservation capacity), and hence influences the fishing grounds that can be reached. Motorized vessels typically make longer trips and are able to reach fishing grounds farther from the coast. However, this statement cannot be generalized, because vessels propelled by sail can reach deeper water when winds are favourable. Therefore, in order to identify the best descriptor between vessel category and trip duration, two ordination analyses that consider two sets of factor were performed: (i) state, vessel, and gear; (ii) state, trip duration, and gear.
Considering trip duration as a category, each sampled vessel (see Table 2 for details) was discriminated by state and by four categories: A, B, C, and D (Table 3). The main fishing gears were lines (63–74% of the landings), gillnets (15–27%), and traps (2–6%). The line category includes bottom handline and longline gear. A handline is made up of a weighted line to which are fixed secondary lines with hooks, usually two secondary lines carrying one hook each. The mixed longline consists of lines with hooks attached at regular distances along a main line, moored to the seabed by weights or anchors. Gillnets are generally made of monofilament nylon and may be fixed to the bottom or drift. Their meshes range from 35 to 130 mm (knot to knot), and they are between 800 and 3600 m in overall length. Traps are constructed of wood, are hexagonal, and are covered with straw, wire, or plastic, and float above the bottom moored by weights or anchors.
|
|
Fishing types
Cluster analysis and multidimensional scaling (MDS) were complementary, so the entire analysis was based on non-metric properties that combined cluster and MDS. The similarity matrix was calculated using the Bray–Curtis distance coefficient. Owing to the large number of vessels sampled, abundance was averaged over groups constituted by state vs. trip duration vs. gear and state vs. vessel vs. gear. Hierarchical complete-linkage cluster and MDS analyses were used to discriminate landings (averaged by grouping factors). As the precise similarity value will not have any direct significance, the complete linkage was chosen as the most appropriate linkage option. Group characterization was carried out by considering the pre-defined factors (gears, state, and trip duration or vessel category) that were considered relevant in defining the artisanal northeastern Brazil reef fishery.
Each MDS representation plot here had stress values less than 0.15. Analyses were based on species-frequency data only, because information on weight was only available for the main species caught. Differences according to the crossed factors state, vessel or trip duration (defined a priori) were tested using a two-way analysis of similarity (ANOSIM) based on the original matrix (landings sampled) instead of averaged catch. The gear factor was not tested, because it was the least relevant, identified through cluster and MDS analysis, although the comments in the results section should be noted.
Fleet dynamics studied through catch composition
The Shannon index of diversity (H') was calculated considering the groups identified from MDS and ANOSIM that were thought to be relevant in defining typologies for the reef fisheries of northeastern Brazil.
A similarity of percentage analysis (SIMPER; Clarke and Warwick, 1994; Clarke and Gorley, 2001) was used to determine those species that were responsible for the dissimilarity between groups. An Indicator Species Analysis (ISA; Dufrêne and Legendre, 1997) specified the maximum indicator value (IV) of each species, with its significance assessed by a Monte Carlo procedure. Indicator values >20% were highlighted in order to visualize the core conservation area of such species (McGeoch and Chown, 1998).
Linking abiotic factors to catch composition
The relationship between abiotic parameters and catch was examined using a BVSTEP procedure, the stepwise alternative to the BIO-ENV routine that empirically links environmental variables to a biotic matrix (Clarke and Gorley, 2001). The abiotic set included the following attributes: vessel, gear, depth, number of fishers, moonlight state, period of a year (trimester), and effort in terms of number of fishers per day spent at sea. A Spearman's rank correlation (
) between abiotic and biological matrices was tested using a permutation procedure, under the null hypothesis that there is no relation whatsoever between the two matrices.
The importance of snappers to the fishery
From a management perspective, it is important to obtain the best understanding of those species that really matter in reef-fishery dynamics, such species being snappers in the studied area. A generalization of the BIO-ENV routine called BIO-BIO (in which the abiotic matrix is switched with the biotic matrix) allowed determination of the small subset of species whose similarity matrix best matched the entire set of species (Clarke and Ainsworth, 1993; Clarke and Warwick, 1994). Within the determined subset, the contribution of each species, particularly that of snappers, was analysed.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The catch of the reef fisheries of northeastern Brazil in the period 1996–2000 comprised 187 species distributed among 14 families. However, lutjanids dominated the catch by number (40%), followed by carangids (15%), scombrids (10%), and haemulids (6%). Although the fishery targeted hard-bottom species, a number of pelagic species were present in the catch in addition to scombrids, clupeids (5.4%), and coryphaenids (2.4%). Families at a high trophic level constituted almost the entire catch, with piscivores and large carnivores dominant. In contrast, herbivores and omnivores were insignificant.
The vessels were mainly motorized (BOM, >55% of those sampled), followed by jangadas (JAN, 30%), the other categories (PQT, BOT, BOV) representing the balance. The dominant gear category was line for all vessel categories (65–86%), then net (9–28%). Traps were used less, just 1.5% (PQT, BOV, BOT), 4% (BOM), and 7% (JAN).
Determination of groups
State and vessel category appeared to be the main factors driving the catch composition. Pernambuco, Alagoas, and Rio Grande do Norte landings had clearly identifiable catch compositions, although Bahia and Ceará catches could not be so clearly distinguished (Figure 2). Within each cluster of state, the factor vessel separated the average landings better than the gear factor, a result corroborated by two-way ANOSIM (Table 4). The pairwise tests performed between states revealed significant differences in the catch composition for all states, and also between Ceará and Bahia, and between Rio Grande do Norte and Pernambuco. Pairwise tests between fleets were not so conclusive (Table 4).
|
|
The MDS plot (Figure 3) was based on averaged landings by state, trip duration, and gear type. The factors state and trip duration appeared to be the main elements driving the catch composition. Catches from the states of Pernambuco and Ceará were split into three, Bahia catches stood alone, while Alagoas catches were similar to those from Rio Grande do Norte. Overall, subgroups were defined according to trip duration category. An ANOSIM test confirmed the differences according to the crossed factors state and trip duration (Table 5), so subsequent analyses were based on those two factors because they were judged to be the most relevant way of best defining the typologies of the reef fishery of northeastern Brazil.
|
|
Diversity of exploited fish species
The southern part of the studied area (Bahia and Alagoas) had the greatest diversity (H' > 2) and Rio Grande do Norte the least (H' < 1; Figure 4). Additionally, the category trip duration B showed the greatest diversity except for Pernambuco, and trip duration category A the least. The categories trip duration C and D had a similar range of diversity index by state.
|
Diversity differed significantly (two-way ANOVA, p < 0.001) between states and between trip duration categories. Interaction of the factors state and trip duration showed similar significant differences (p < 0.001) to the diversity indexes (Table 6).
|
Typology of groups
The reef fishery of northeastern Brazil exploits four lutjanid species: Lutjanus analis, L. chrysurus, L. synagris, and L. jocu (Table 7). Coryphaena hippurus and Scomberomorus cavalla are also important. A few species in certain groups were also important, e.g. the 21% contribution of Pseudopeneus maculatus to the return for PEA (state of Pernambuco, trip duration A), which also featured L. synagris (62%). L. synagris also contributed strongly in the return for CEA (state of Ceará, trip duration A) and RNA (state of Rio Grande do Norte, trip duration A). L. analis provided an important contribution in Pernambuco (trip duration B, C, and D) and in Alagoas (trip duration A; Table 7).
|
The largest indicator values (IV > 20%, emboldened in Table 8) were in Ceará (trip duration A, B, and D), Pernambuco (trip duration A), and Bahia (trip duration B). A few species characterized the fishing area, viz. Opisthonema oglinum for CEA, Holocentrotus adscensionis, Haemulon melanurum, and Malacanthus plumieri for CEB, Lutjanus chrysurus, Rachycentron canadum, and Acanthocybium solandri for CED, Pseudopeneus maculates, Lutjanus synagris, and Haemulon aurolineatum for PEA, and Caranx latus for BAB. Some species had low IVs and were categorized as rare, but restricted to a sector and therefore not considered to be caught by chance.
|
The relationship between catch composition and fishing typology
The best-factor combination from the BVSTEP procedure (with a higher Spearman's rank correlation) was identified for each state (Table 9). Each correlation was significantly different from zero, although there was no strong match of catch pattern with abiotic factors. Much of the catch variation could not be explained by the abiotic factors used in the analysis, although this statement depends on the state. For example, it appears that depth on the northern coast (Ceará) and trip duration on the southern coast (Bahia, Alagoas), respectively, maximized the correlation coefficient. The dry season (trimesters 1 and 4) and crew size were also important for the states in the south. Bahia had the lowest correlation coefficient and the largest set of abiotic variables.
|
Best subset of species to characterize catch composition
Only 13 species were required to maximize the correlation between biological and abiotic matrices, i.e. the species responsible for characterizing the typologies in the reef fishery of northeastern Brazil (Table 10). The family Lutjanidae was the most important family, with five species (Lutjanus chrysurus, L. synagris, L. analis, L. jocu, L. vivanus) that showed a hierarchical structure: the best combination at one level is always a subset of the best combination at the next level. Species of the families Carangidae (Carangoides crysos, C. latus, Seriola dumerili), Coryphaenidae (Coryphaena hippurus), Haemulidae (Haemulon plumieri), Scombridae (Scomberomorus cavalla, S. brasiliensis), and Serranidae (Cephalopholis fulvus) were also important. As mentioned above, these families are at a high trophic level, because they are piscivorous (Carangidae, Serranidae) or large carnivorous (Lutjanidae, Haemulidae, Scombridae), and belong to the shallow lutjanid complex, in contrast to L. bucanella, L. purpureus, Epinephelus niveatus, and E. morio, which live in deeper water. The fleets catching them operate mainly in shallow water.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Many multivariate methods have been applied to fishery data in an attempt to distinguish phenomena such as spatial pattern, fishery power, and gear selectivity related to fishing trip or survey in mixed fisheries (Clarke and Warwick, 1994; Pech and Laloë, 1997; Garcia et al., 1998; Legendre and Legendre, 1998; Pitcher et al., 1998; Preikshot and Pauly, 1998; Pelletier and Ferraris, 2000; Rice, 2000; Pitcher and Preikshot, 2001; Stergiou et al., 2002; Willis and Anderson, 2003). In this study, univariate and multivariate techniques were used to identify an integrated picture of the structure of the reef system of northeastern Brazil. Among the factors considered (gear, vessel category, state, and trip duration), spatial effect (state) appeared to have the greatest influence in separating the various groups. The dynamics of each state were influenced by a complex combination of technological and environmental factors that drive the catch composition of the landings in each state. Incorporation of non-biological factors provides an important tool for better understanding the fishing dynamics (Pitcher et al., 1998; Preikshot and Pauly, 1998), and such knowledge allows the researcher to focus on specific information to optimize the resources at his disposal. The combination of these techniques leads to knowledge of the structure and the dynamics of the fishing activity.
The factor "state" represented not only the geopolitical divisions of the waters off northeastern Brazil, but also reflected local fishing traditions. These were represented by the factors fishing gear, type of vessel, navigational skills, and preferred (traditional) fishing grounds, all of which have been developed to adapt to the local characteristics of continental shelf width, presence of hard substrata over the shelf, and the dominant weather conditions (wind and current). The factor "trip duration" best discriminated the catch composition. Although the trip duration categories clearly relate to the method of propulsion of the fleet (by rowing, by sail, and by motor), motorized boats generally making the longest trips, there are some exceptions. For example, wind-propelled boats may operate as far from home as the shelf break when winds are strong in the states of Rio Grande do Norte and Ceará. For this reason, trip duration was the factor that best determined the distance from the coast, the variety and the uniqueness of the fishing grounds reached, and hence whether more coastal or deep-water species dominated. Although it was expected that the gear types would separate groups of catches, the factor "gear" appears to be less influential in discriminating the catch. The reason for this may be that the northeastern Brazilian fishery is typically a multigear and multispecies one, and most fleet categories carry more than one type of gear during a trip.
A complex of significant factors seems to influence the catch composition and is related to distance from shore and depth of fishing. As Pelletier and Ferraris (2000) stated for the Senegalese artisanal fishery, this analysis could be improved if fishing locations were defined precisely. Trip duration can be used as a proxy for the position of fishing grounds, so should be considered as an alternative when information on depth is not available. The maximum diversity reached at the intermediate trip duration category may depend on habitat availability (Longhurst and Pauly, 1987; Levinton, 2001), and the ecological range of the species caught may explain the pattern of shallow- and deep-species distribution overlap in the intermediate range. The catch per unit effort (cpue) of snapper species varies with depth. Lutjanus synagris and L. vivanus were more abundant in shallow and deeper waters, respectively, but L. analis, L. chrysurus, and L. jocu were more evenly distributed though their abundance peaked in intermediate depths of the continental shelf (Frédou and Ferreira, 2005).
Although in some instances the diversity may have been partially influenced by the biogeography of the reef assemblage (Floeter et al., 2001; Joyeux et al., 2001; Araújo and Feitosa, 2003), it is of note that diversity was less in the state with a wide continental shelf (Rio Grande do Norte) than in the states with a narrow shelf (Alagoas and Bahia). Fleets that operate over a wide shelf do not reach deep water easily, so most of their catch comes from a homogeneous environment, whereas fleets operating on a narrow shelf can exploit a larger range of environments and quite probably a more diverse species assemblage.
Although the fishery was demersal, pelagic species were common in the catch (scombrids, coryphaenids, and clupeids). Scombrids and clupeids are commonly associated with reefs (Collette and Nauen, 1983; Whitehead, 1985) and are therefore potential targets. However, coryphaenids, which do not have links with hard substrata, can be captured by gillnets set throughout the water column. Also, it is a common practice that fishers operate drifting lines when moving to fishing grounds to obtain bait, and coryphaenids are susceptible to capture by such means.
Snappers dominated the artisanal catch and contributed most to the similarity between groups. Longhurst and Pauly (1987) stated that their so-called lutjanid community dominated around the Bahamas, the Antilles, and other Caribbean islands, and along the coast from Yucatan to Panama. Here too we found the community to be dominant along the Brazilian coast, presumably because coral reefs are distributed from shallow water to the shelf break (Maida and Ferreira, 1997). The subset that best defined catch variability was primarily represented by the snappers Lutjanus chrysurus, L. synagris, L. analis, L. jocu, and to a lesser extent L. vivanus. All but L. vivanus (Frédou and Ferreira, 2005) belong to the shallow lutjanid complex so were caught by fleets operating in shallower water. The deep lutjanid complex, by contrast, consists of L. purpureus, L. bucanella, Epinephelus niveatus, and E. morio.
Several criteria may characterize the reef fishery of northeastern Brazil, for instance life history and ecology of the species caught, technical and geographic characteristics (i.e. geomorphology of the shelf; Frédou, 2004), and the local culture (e.g. some fisher communities do not traditionally operate away from the coast). In an earlier study on the snapper fishery, Frédou and Ferreira (2005) identified a general trend towards greater size with increasing depth along with different abundance patterns, according to the species considered. As a result, the fleet comes into contact with stocks of different species and sizes. Fleets that operate in different ways therefore affect stocks differently, and fishing patterns are developed in response to simultaneous interaction of these issues. Each vessel has a catchability and efficiency specific to its design and to its operational range in relation to the species targeted, and similar fleets may have a distinct impact consistent with the fishing area in which they operate. Changes in the catch composition will incorporate technological effects through spatial or temporal changes or a mixture of both these factors.
The heterogeneity of the northeastern reef fishery described here will have consequences for the best choice of stock assessment model and the realization of management guidelines. Multispecies fishery models may be defined within two main categories, one of which accounts for biological interactions and models technological interactions. Incorporating biological interaction into multispecies models presents difficulties, as pointed out by Pikitch (1988). Technical interactions are more easily identified and can be applied in stock assessment models and management (Brugge and Holden, 1991; Gulland, 1991; Lucena et al., 2002). Technical interactions in the artisanal fishery rely on competing fleets exploiting the same resources and on mixed catches resulting from there being multiple target species (Hilborn and Walters, 1992). Fleets can exploit different stages in the life cycle of a fish stock, in simultaneous or sequential harvesting, and can also cover different geographical areas. Such cases are very common in tropical fisheries, which are typically multifleet and multispecies. Considering the stratified distribution of snappers, fleets with different operational capacities affect stocks on different ways. Fisheries of the same fish community are linked through exploitation practices. Because technical interaction models are straightforward (in contrast to biological interaction models) and are essentially derived from single-species models (Brugge and Holden, 1991), the impacts of fishing pressure on exploited species and the reef ecosystem in general should be investigated considering such factors. Other factors may also be of interest, because most catches involved species at high trophic levels, resulting in a significant impact on the overall functioning of reef systems (Jennings and Lock, 1996; Jennings and Kaiser, 1998). Also, reef systems have typically fragmented, patchy habitat that is sequentially harvested (when catches become sub-economic, fleets move to new fishing grounds).
Managers should be aware that the implementation of a common harvesting strategy may not be applicable for an entire fishing area. As discussed by Hilborn and Walters (1992), if the assumption is that the system is spatially structured, there is an opportunity to set up a range of alternative management policies for the various segments of the system. Alternative policies can be applied to a smaller system safely to provide a buffer against risks of overexploiting some areas. In the case of northeastern Brazil, modelling the artisanal, hard-bottom, demersal fishery and the development of harvesting strategies should take into consideration the fact that the region is not homogeneous (i.e. fleet composition, species interactions, and geomorphological characteristics are all different). In order to refine knowledge, information on the spatial characteristics should be determined by surveys in addition to collecting detailed information about commercial fishing activities.
This analysis has helped, for management purposes, to reduce the species complex significantly to just a few. Within this subset, lutjanids appear to be the main group because they dominate the catch and drive the fishery dynamics. They are much appreciated by the local market and are of high commercial value, a tradition set many decades ago during the boom of the southern red snapper fishery (Rezende et al., 2003). Therefore, further attention should be paid to the ecology of the group and to the impact of fishing them. Reef fishery science argues that, owing to the complexity of the system, single-species assessment may not be adequate. An alternative may be to develop biomass dynamic models for aggregate species groups covering jointly caught species (Ralston and Polovina, 1982). Models of this type can be appropriate for both interacting and non-interacting species (Hollowed et al., 2000). The advantage of such approaches is that there is no specific target species, but rather a group of species. In that case, our results suggest that the aggregated species should be the best subset that drives the fishing dynamics, namely L. analis, L. chrysurus, L. jocu, and L. synagris.
|
Acknowledgements |
|---|
This research forms part of the Programa Nacional de Avaliação do Potencial Sustentável dos Recursos vivos da Zona Econômica Exclusiva – REVIZEE, funded by the Ministério do Meio Ambiente (MMA) and the Secretaria da Comissão Interministerial para os Recursos do Mar (SECIRM). We thank André Vasconcelos, Kenia Cunha, Elton Nunes, Kátia Freire, Denis Hellebrandt, Marcelo Nóbrega, Moustapha Diedhiou, Roberto Kobayashi, Sergio Rezende (DTI/CNPq), Simone Teixeira, and several students for their assistance in data collection, and Flávia Lucena for helpful comments on an earlier draft.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Araújo M.E. and Feitosa V.C. (2003) Análise de agrupamento da ictiofauna recifal do Brasil com base em dados secundários: uma avaliação crítica. Tropical Oceanography 31:171–192.
Biseau A. and Gondeaux E. (1988) Methods of data-analysis in typological studies of fishery fleets. Journal du Conseil International pour l'Exploration de la Mer 44:286–296.
Brugge W.J. and Holden M.J. (1991) Multispecies management: a manager's point of view. ICES Marine Science Symposia 193:353–358.
Clarke K.R. and Ainsworth M. (1993) A method linking multivariate community structure to environmental variables. Marine Ecology Progress Series 92:205–219.[Web of Science]
Clarke K. R. and Gorley R.N. (2001) Primer v5: User Manual/Tutorial. PRIMER-E Ltd, 5. 91 pp.
Clarke K.R. and Warwick R.M. (1994) Change in Marine Communities: An Approach to Statistical Analysis and InterpretationPlymouth Marine Laboratory 144 pp.
Collette B.B. and Nauen C.E. (1983) FAO Species Catalogue. Scombrids of the World. An Annotated and Illustrated Catalogue of Tunas, Mackerels, Bonitos and Related Species Known to Date(FAO, Rome) vol. 2: 137 pp.
Daan N. (1987) Multispecies versus single-species assessment of North Sea fish stocks. Canadian Journal of Fisheries and Aquatic Sciences 44:360–370.
Dufrêne M. and Legendre P. (1997) Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs 67:345–366.[CrossRef][Web of Science]
Floeter S.R., Guimarães R.Z.P., Rocha L.A., Ferreira C.E.L., Rangel C.A., Gasparini J.L. (2001) Geographic variation in reef-fish assemblages along the Brazilian coast. Global Ecology and Biogeography 10:423–431.[CrossRef][Web of Science]
Frédou T. (2004) The fishing activity on coral reefs and adjacent ecosystems. A case study of the northeast of Brazil. PhD thesis, Universidade Federal de Pernambuco/Université de la Méditerranée. 200 pp.
Frédou T. and Ferreira B.P. (2005) Bathymetric trends of northeastern Brazilian snappers (Pisces, Lutjanidae): implications for the reef fishery dynamic. Brazilian Archives of Biology and Technology 48:787–800.[Web of Science]
Garcia C.B., Duarte L.O., von Schiller D. (1998) Demersal fish assemblages of the Gulf of Salamanca, Colombia (southern Caribbean Sea). Marine Ecology Progress Series 174:13–25.[Web of Science]
Gulland J.A. (1991) Under what conditions will multispecies models lead to better fisheries management? ICES Marine Science Symposia 193:348–352.
He X., Bigelow K.A., Boggs C.H. (1997) Cluster analysis of longline sets and fishing strategies within the Hawaii-based fishery. Fisheries Research 31:147–158.[CrossRef][Web of Science]
Hilborn R. (1985) Fleet dynamics and individual variation: why some people catch more fish than others. Canadian Journal of Fisheries and Aquatic Sciences 42:2–13.
Hilborn R. and Walters C.J. (1992) Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty(Chapman and Hall, New York) 570 pp.
Hollowed A.B., Bax N., Beamish R., Collie J., Fogarty M., Livingston P., Pope J., Rice J.C. (2000) Are multispecies models an improvement on single-species models for measuring fishing impacts on marine ecosystems? ICES Journal of Marine Science 57:707–719.
Jennings S. and Kaiser M.J. (1998) The effects of fishing on marine ecosystems. Advances in Marine Biology 34:201–352.[Web of Science]
Jennings S., Kaiser M.J., Reynolds J.D. (2001) Marine Fisheries Ecology(Blackwell Science, Oxford) 417 pp.
Jennings S. and Lock J.M. (1996) Population and ecosystem effects of reef fishing. In Polunin N.V.C. and Roberts C.M. (Eds.). Reef Fisheries(Chapman and Hall, London) pp. 193–218 477 pp.
Joyeux J-C., Floeter S.R., Ferreira C.E.L., Gasparini J.L. (2001) Biogeography of tropical reef fishes: the South Atlantic puzzle. Journal of Biogeography 28:831–841.[CrossRef][Web of Science]
Legendre P. and Legendre L. (1998) Numerical Ecology(Elsevier, Amsterdam) 853 pp.
Levinton J.S. (2001) Marine Biology. Function, Biodiversity, Ecology(Oxford University Press, Oxford, UK) 515 pp.
Longhurst A. and Pauly D. (1987) Ecology of Tropical Oceans(Academic Press, San Diego, CA) 407 pp.
Lucena F.M., O'Brien C.M., Reis E.G. (2002) Effect of exploitation of the bluefish Pomatamus saltatrix by two competing fleets in southern Brazil: an application of a catch-at-age model. Marine and Freshwater Research 53:1–13.[CrossRef][Web of Science]
Magnusson K.G. (1995) An overview of multispecies VPA – theory and applications. Reviews in Fish Biology and Fisheries 5:195–212.[Web of Science]
Maida M. and Ferreira B. P. (1997) Coral reefs of Brazil: overview and field guide. Eighth International Coral Reef Symposium pp. 263–274.
McGeoch M.A. and Chown S.L. (1998) Scaling-up the value of bioindicators. Trends in Ecology and Evolution 13:46–47.[CrossRef]
Murawski S.A., Lange A.M., Sissenwine M.P., Mayo R.K. (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.
Pauly D. (1980) A Selection of Simple Methods for the Assessment of Tropical Fish Stocks(FAO, Rome) 54 pp.
Pech N. and Laloë F. (1997) Use of principal component analysis with instrumental variables (PCAIV) to analyse fisheries catch data. ICES Journal of Marine Science 54:32–47.
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.
Pikitch E.K. (1988) Objectives for biologically and technically interrelated fisheries. In Wooster W.S. (Ed.). Fisheries Science and Management: Objectives and Limitations(Springer-Verlag, New York) 28: pp. 107–136 Lecture Notes on Coastal and Estuarine Studies.
Pitcher T., Mackinson S., Vasconcellos M., Nøttestad L., Preikshot D. (1998) Rapid appraisal of the status of fisheries for small pelagics using multivariate, multidisciplinarity ordination. Conference Proceedings: Fishery Stock Assessment Models. Alaska Sea Grant College Program pp. pp. 759–782.
Pitcher T.J. and Preikshot D. (2001) RAPFISH: a rapid appraisal technique to evaluate the sustainability status of fisheries. Fisheries Research 49:255–270.[CrossRef][Web of Science]
Polunin N.V.C. and Roberts C.M. (1996) Reef Fisheries(Chapman and Hall, London) 477 pp.
Preikshot D. and Pauly D. (1998) Multivariate interdisciplinary assessment of small-scale tropical fisheries. Conference Proceedings: Fishery Stock Assessment Models. Alaska Sea Grant College Program pp. pp. 803–814.
Ralston S. and Polovina J.J. (1982) A multispecies analysis of the commercial deep-sea hand-line fishery in Hawaii. Fishery Bulletin US 80:435–448.
Rezende S.M., Ferreira B.P., Frédou T. (2003) A pesca de lutjanídeos no nordeste do Brasil: histórico das pescarias, características das espécies e relevância para o manejo. Boletim Técnico Científico do CEPENE 11:1–17.
Rice J.C. (2000) Evaluating fishery impacts using metrics of community structure. ICES Journal of Marine Science 57:682–688.
Stergiou K.I., Moutopoulos D.K., Erzini K. (2002) Gill net and longline fisheries in Cyclades waters (Aegean Sea): species composition and gear competition. Fisheries Research 57:25–37.[CrossRef][Web of Science]
Whitehead P.J.P. (1985) FAO Species Catalogue. Clupeoid Fishes of the World (Suborder Clupeoidei). An Annotated and Illustrated Catalogue of the Herrings, Sardines, Pilchards, Sprats, Anchovies and Wolf-herrings. Part 1 – Chirocentridae, Clupeidae and Pristigasteridae(FAO, Rome) vol. 7: 303 pp.
Willis T.J. and Anderson M.J. (2003) Structure of cryptic reef fish assemblages: relationships with habitat characteristics and predator density. Marine Ecology Progress Series 257:209–221.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||