ICES Journal of Marine Science: Journal du Conseil

Collie, Jeremy, S.;Gislason,, Henrik;Vinther,, Morten

doi: 10.1016/S1054-3139(03)00042-0pmid: N/A

Abstract In multispecies fish communities, predation levels change dynamically in response to changes in the abundance of predator and prey species, as influenced by the fisheries that exploit them. In addition to community-level metrics, it remains necessary to track the abundance of each species relative to its biological reference point. In situations with many interacting species, exploited by multiple fishing fleets, it can be complicated to illustrate how the effort of each fleet will affect the abundance of each species. We have adapted the AMOEBA approach to graph the reference levels of multiple interacting species exploited by multiple fleets. This method is illustrated with 10 species and eight fishing fleets in the North Sea. We fit a relatively simple response-surface model to the predictions of a fully age-structured multispecies model. The response-surface model links the AMOEBA for fishing effort to separate AMOEBAs for spawning stock biomass, fishing mortality, and yield. Ordination is used to give the shape of the AMOEBAs functional meaning by relating fish species to the fleets that catch them. The aim is to present the results of dynamic multispecies models in a format that can be readily understood by decision makers. Interactive versions of the AMOEBAs can be used to identify desirable combinations of effort levels and to test the compatibility of the set of single-species biological reference points. Introduction There is widespread acceptance that an ecosystem perspective is needed to manage marine fisheries but much less practical experience on how to do so. It is now recognized that ecosystems themselves cannot be managed; it is the human users of ecosystems that must be regulated (Bax et al., 1999). In marine fisheries, two approaches have emerged for incorporating ecosystem considerations into management decisions (Murawski, 2000). One is to use the metrics of community ecology, such as species diversity and size spectra, as indicators of ecosystem status, ecosystem health, and ecosystem services (Rice, 2000). The other approach is to incorporate additional ecosystem constraints into traditional management decisions. With respect to the first approach, criteria to aid in the selection of ecosystem objectives and their associated metrics have been put forward by the ICES Working Group on Ecosystem Effects of Fishing Activities (ICES, 2001, 2002a). Although rapid progress is being made, there is presently little theoretical understanding of how many of the proposed metrics respond to changes in harvesting, and what the desirable level of the metrics should be. For many important ecosystem properties, scientific understanding of the link between human impacts and ecosystem response is insufficient, and considerable effort is therefore needed before the approach can be fully implemented (ICES, 2002a). Although the concept of ecosystem health seems an intuitive analogy with the human body, it breaks down on closer examination because ecosystems can exist in multiple states, in all of which basic ecological functions are maintained (Hall, 1999). The second approach is to define reference levels for taxa other than the targets of the directed fishery. These ecosystem constraints have been considered as additional levers to nudge the management process toward meeting ecosystem, or at least community-level objectives (Bax et al., 1999). Another way to view these constraints is as additional dimensions or objectives to be satisfied in fishery management plans. Examples of ecosystem constraints are limits on the take of marine mammals in fisheries, catch limits on forage fish to preserve their predators, and area closures to protect structural epifauna. Progress can be expected with both approaches, but in the short-term it is more pragmatic to incorporate ecosystem considerations as additional constraints to existing fishery management plans (Murawski, 2000). Adding ecosystem constraints is likely to increase the complexity of the advice and may therefore increase the difficulties of reaching consensus if trade-offs cannot be presented to stakeholders and managers in an easily comprehensible way. Among the most important ecosystem processes in marine fish communities are trophic interactions among the fish species. There is empirical evidence that the mortality rate of prey species depends on predator abundance and, conversely, that predator growth rates depend on prey abundance (Collie, 2001). Competition is implied by many ecological models but is certainly more difficult to demonstrate than predation, and is also less likely to be structuring highly interconnected marine food webs. There is a strong parallel between the two approaches to incorporating ecosystem considerations and the types of multispecies models used for each approach. Models of the entire ecosystem (e.g. network models, dynamic ecosystem models) should lend themselves to the derivation of ecosystem metrics (Hollowed et al., 2000). On the other hand, community-level models of interacting species are more useful for adding ecosystem constraints to the single-species models that are widely used in fisheries management. A second dichotomy is whether the multispecies model is age structured or just tracks the total abundance or biomass of each species (Hollowed et al., 2000); age-structured models are most widely used in the management of temperate marine fish. When does multispecies advice matter in fisheries management? In the short term, the feeding requirements of predators must be considered when setting annual harvest quotas for forage fish species (e.g. capelin off Norway). In the medium term, biological reference points may need to be adjusted to account for variable predation rates on prey species and variable growth rates of predators (Collie and Gislason, 2001). Long-term management strategies need to account for the implicit trade-offs in prey and predator yields (May et al., 1979). In boreal ecosystems with a small number of interacting species, it may be straightforward to condition the reference levels of a target species on the abundances of interacting species (Livingston and Tjelmeland, 2000). In temperate ecosystems with a large number of fish species, the increased dimensionality necessitates different approaches. The North Sea is a good example of such an ecosystem. It harbours an intensive multispecies and multifleet fishery using a variety of gears. Demersal fisheries for human consumption catch a mixture of roundfish species, such as cod, haddock, whiting, and saithe, that are piscivorous to some extent, or target flatfish species such as sole and plaice, often with a bycatch of roundfish species. Pelagic fisheries for human consumption are directed at species such as herring and mackerel, while the industrial fishery targets forage species, such as sandeel, Norway pout, and sprat, and uses them for production of fishmeal and fish oil. Due to excessive levels of fishing mortality many of the demersal stocks are at present considered to be outside safe biological limits (ICES, 2002b). The European Common Fisheries Policy has multiple objectives (Halliday and Pinhorn, 1996). In the short term these are: (a) to ensure the continuity of each stock as a commercially viable resource; (b) to decrease the fishing effort on overexploited stocks in order to ensure yields that are stable from year to year; and (c) to ensure the highest possible catch from stocks, consistent with (a) and (b) and taking into account the relationships among stocks. ICES has established precautionary biomass (Bpa) and fishing mortality (Fpa) levels for each stock to meet objectives (a) and (b). These precautionary targets are intended to maintain each stock at a productive level and to provide a high probability of avoiding stock collapse. Objective (c) implies that fishery yields should be maximized, subject to the biological constraints and multispecies interactions. The North Sea fisheries have for many years been managed with a system of quotas or total allowable catches (TACs). Particularly for the species caught in the mixed demersal fisheries, this system has failed to produce the intended reductions in fishing mortality. There are several reasons for this failure, one of which is that the TACs often have been set independently for each species. This has contributed to extensive discarding and to little or no reduction in overall fishing mortality despite reductions in landings. As concluded by Holden (1994), the TAC system is a fundamentally flawed system for managing the mixed demersal fisheries in the North Sea. Due to technical interactions, fishing mortality cannot be regulated independently for each species. In addition there are important biological interactions taking place and these interactions need to be taken into account in the medium and long-term projections used for formulating rebuilding strategies (ICES, 1997). Age-structured multispecies and multifleet models (e.g. MSFOR) have thus been developed. Such models can be used to investigate the consequences of different fishing mortalities while accounting for interactions among fishing fleets and between the predators and prey (Pope, 1991). Although the models have now been available for more than a decade they have not been used routinely in fisheries management, partly because of their intensive data and computing requirements, but mainly because the increased complexity of multispecies models is thought to hinder decision making (Brugge and Holden, 1991). In an attempt to construct a simpler multispecies model, Pope (1989) proposed fitting a simpler response-surface model to the results of the more complicated multispecies model, and then using the response-surface model to investigate alternative levels of fishing effort. The ICES Multispecies Assessment Working Group (MSAWG) used a multispecies Schaefer or Fox model fit to the projections of the MSFOR model (ICES, 1992). This approach greatly simplifies the multispecies model, but leaves the problem of visualizing the results in as many dimensions as there are interacting species and fishing fleets. Joint levels of F0.1 and Fmsy for interacting fishing fleets can be calculated (Pope, 1989) but such community-wide indices do not ensure that reference levels for individual species will be met. Pope (1997) emphasized that “whatever model of complex, multispecies, multifleet, multiarea fisheries is adopted, it will be of little use in the real world unless its results can be presented to the managers in as clear and as unambiguous fashion as possible”. The results of multispecies models can be presented with decision tables or radar plots. AMOEBAs are extensions of radar plots that can be useful for visualizing multidimensional situations in which several constraints must be met simultaneously (Laane and Peters, 1993). Pioneered in the Netherlands in the context of water quality objectives, AMOEBA is the Dutch acronym for “a general method of ecosystem description and assessment” (Ten Brink et al., 1991). The AMOEBA approach has been applied to shellfish restoration in North Carolina (Wefering et al., 2000) and has also been proposed for displaying ecological quality objectives in the North Sea (Lanters et al., 1999). In this paper we show how AMOEBAs can be used to visualize the results of multispecies models applied to the North Sea fish community. We extend the AMOEBA concept by giving the shape of the AMOEBAs functional meaning and by making them change dynamically in response to changing effort levels. In creating the AMOEBA plots we followed Tufte's (1983) principles of graphic excellence. According to Tufte, graphic displays should: show the data in a way that makes large data sets coherent; induce the viewer to think about the substance rather than about the methodology; present many numbers in a small space; encourage the eye to compare the different pieces of data; and serve a reasonably clear purpose. The ultimate objective of this work is to present the results of multispecies fishery models in a format that can be readily understood and used by decision makers. Methods In this paper we analysed the multispecies, multifleet fishery of the North Sea. Multispecies VPA and projections were made using the 4M program (Vinther et al., 2001). The 4M (Multi-species, Multi-Fleet and Multi-Area Model) package is a newer and extended implementation of the MSVPA/MSFOR programs previously used by the ICES MSAWG. The forecasts were based on an MSVPA run similar to the so-called “key-run” made at the last MSAWG Meeting (ICES, 1997). This MSVPA included data for 10 VPA species (Table 1) for the period 1974–1995 such that 1996 became the first projection year. Recruitment in the projections was assumed to follow a Ricker stock-recruitment relation fitted to the MSVPA output for all VPA species except North Sea mackerel. An arithmetic mean of the estimated recruits in 1986–1995 was used for mackerel because the stock-recruitment relation was indeterminate. Abundance of species without analytical assessment (“other predators”) was kept constant at the 1995 level in the projections. Table 1 Species included in multispecies assessment. . Size or age groups . Predator/prey . Abbreviation for figures . VPA species  Cod 0–11+ Yes/Yes COD  Haddock 0–11+ Yes/Yes HAD  Whiting 0–10+ Yes/Yes WHG  Saithe 0–15+ Yes/(Yes) POK  Mackerel (North Sea stock) 0–15+ Yes/(Yes) MAC  Herring 0–9+ No/Yes HER  Norway pout 0–3+ No/Yes NOP  Sandeel 0–6+ No/Yes SAN  Plaice 0–15+ No/(Yes) PLE  Sole 0–15+ No/(Yes) SOL Other predators (abundance given as input)  Grey gurnards 0–3 Yes/No  Western stock mackerel 0–1 Yes/No  Raja radiata 0–3 Yes/No  Grey seals 1 Yes/No  Sea birds 1 Yes/No  Other species 1 Yes/No . Size or age groups . Predator/prey . Abbreviation for figures . VPA species  Cod 0–11+ Yes/Yes COD  Haddock 0–11+ Yes/Yes HAD  Whiting 0–10+ Yes/Yes WHG  Saithe 0–15+ Yes/(Yes) POK  Mackerel (North Sea stock) 0–15+ Yes/(Yes) MAC  Herring 0–9+ No/Yes HER  Norway pout 0–3+ No/Yes NOP  Sandeel 0–6+ No/Yes SAN  Plaice 0–15+ No/(Yes) PLE  Sole 0–15+ No/(Yes) SOL Other predators (abundance given as input)  Grey gurnards 0–3 Yes/No  Western stock mackerel 0–1 Yes/No  Raja radiata 0–3 Yes/No  Grey seals 1 Yes/No  Sea birds 1 Yes/No  Other species 1 Yes/No Prey (Yes) indicates very low predation mortality. Open in new tab Table 1 Species included in multispecies assessment. . Size or age groups . Predator/prey . Abbreviation for figures . VPA species  Cod 0–11+ Yes/Yes COD  Haddock 0–11+ Yes/Yes HAD  Whiting 0–10+ Yes/Yes WHG  Saithe 0–15+ Yes/(Yes) POK  Mackerel (North Sea stock) 0–15+ Yes/(Yes) MAC  Herring 0–9+ No/Yes HER  Norway pout 0–3+ No/Yes NOP  Sandeel 0–6+ No/Yes SAN  Plaice 0–15+ No/(Yes) PLE  Sole 0–15+ No/(Yes) SOL Other predators (abundance given as input)  Grey gurnards 0–3 Yes/No  Western stock mackerel 0–1 Yes/No  Raja radiata 0–3 Yes/No  Grey seals 1 Yes/No  Sea birds 1 Yes/No  Other species 1 Yes/No . Size or age groups . Predator/prey . Abbreviation for figures . VPA species  Cod 0–11+ Yes/Yes COD  Haddock 0–11+ Yes/Yes HAD  Whiting 0–10+ Yes/Yes WHG  Saithe 0–15+ Yes/(Yes) POK  Mackerel (North Sea stock) 0–15+ Yes/(Yes) MAC  Herring 0–9+ No/Yes HER  Norway pout 0–3+ No/Yes NOP  Sandeel 0–6+ No/Yes SAN  Plaice 0–15+ No/(Yes) PLE  Sole 0–15+ No/(Yes) SOL Other predators (abundance given as input)  Grey gurnards 0–3 Yes/No  Western stock mackerel 0–1 Yes/No  Raja radiata 0–3 Yes/No  Grey seals 1 Yes/No  Sea birds 1 Yes/No  Other species 1 Yes/No Prey (Yes) indicates very low predation mortality. Open in new tab Fishing mortalities (F) estimated for 1995 by the MSVPA were used as base line or status-quo levels in the projections. These F values were partitioned to partial F by fleet according to catch numbers given by the STCF database (Anon., 1991; Lewy et al., 1992), which includes detailed catch information for 56 national fleets fishing in the North Sea in 1991. These 56 fleets were aggregated into eight new fleets defined by the gear used or target species; “other gears” include national fleets that did not fit the grouping. Average partial Fs over the age range used by ICES in the calculation of reference F values are presented in Table 2. The exploitation patterns of the fishing fleets and the stock sizes of each species have changed considerably since 1991, such that each fleet probably accounts for a different percentage of the total catch of each species than in 1991. Therefore the status-quo exploitation patterns would need to be updated before the results of projecting different management options could be used for actual management decisions. Table 2 Average fishing mortality by fleet and species as used in the status-quo projection. Also listed are the precautionary reference levels for fishing mortality (Fpa) and SSB (Bpa) from ICES (2002b). For herring and mackerel only, the status-quo SSB levels were used as proxies for Bpa because these two stocks have components that are not resident in the North Sea. A plus sign (+) indicates fishing mortality <0.01 and a minus sign (−) indicates a species not caught by that gear. . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 0.03 + 0.01 + 0.35 0.35 + + − − Fixed gear 0.10 + + + 0.02 0.02 + + − − Industrial (small meshed trawl) 0.02 0.01 0.04 0.05 + + 0.05 0.02 0.36 0.36 Pelagic (purse seine and trawl) + + + − + + 0.47 0.10 + + Saithe (trawl) 0.01 0.01 + 0.09 + − + + − − Seine net 0.21 0.30 0.16 0.01 0.05 + + + − − Trawl 0.28 0.35 0.27 0.16 0.05 0.03 0.11 0.01 + + Other gears 0.16 0.05 0.01 0.11 0.07 0.11 0.20 + 0.01 + All fleets 0.81 0.73 0.49 0.42 0.55 0.51 0.83 0.12 0.36 0.36 Precautionary F level 0.65 0.70 0.65 0.40 0.30 0.40 0.25 0.17 0.59 0.84 Precautionary biomass level (kt) 150 140 315 200 300 35 311a 86a 600 150 . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 0.03 + 0.01 + 0.35 0.35 + + − − Fixed gear 0.10 + + + 0.02 0.02 + + − − Industrial (small meshed trawl) 0.02 0.01 0.04 0.05 + + 0.05 0.02 0.36 0.36 Pelagic (purse seine and trawl) + + + − + + 0.47 0.10 + + Saithe (trawl) 0.01 0.01 + 0.09 + − + + − − Seine net 0.21 0.30 0.16 0.01 0.05 + + + − − Trawl 0.28 0.35 0.27 0.16 0.05 0.03 0.11 0.01 + + Other gears 0.16 0.05 0.01 0.11 0.07 0.11 0.20 + 0.01 + All fleets 0.81 0.73 0.49 0.42 0.55 0.51 0.83 0.12 0.36 0.36 Precautionary F level 0.65 0.70 0.65 0.40 0.30 0.40 0.25 0.17 0.59 0.84 Precautionary biomass level (kt) 150 140 315 200 300 35 311a 86a 600 150 a Status-quo SSB used as a proxy for Bpa. Open in new tab Table 2 Average fishing mortality by fleet and species as used in the status-quo projection. Also listed are the precautionary reference levels for fishing mortality (Fpa) and SSB (Bpa) from ICES (2002b). For herring and mackerel only, the status-quo SSB levels were used as proxies for Bpa because these two stocks have components that are not resident in the North Sea. A plus sign (+) indicates fishing mortality <0.01 and a minus sign (−) indicates a species not caught by that gear. . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 0.03 + 0.01 + 0.35 0.35 + + − − Fixed gear 0.10 + + + 0.02 0.02 + + − − Industrial (small meshed trawl) 0.02 0.01 0.04 0.05 + + 0.05 0.02 0.36 0.36 Pelagic (purse seine and trawl) + + + − + + 0.47 0.10 + + Saithe (trawl) 0.01 0.01 + 0.09 + − + + − − Seine net 0.21 0.30 0.16 0.01 0.05 + + + − − Trawl 0.28 0.35 0.27 0.16 0.05 0.03 0.11 0.01 + + Other gears 0.16 0.05 0.01 0.11 0.07 0.11 0.20 + 0.01 + All fleets 0.81 0.73 0.49 0.42 0.55 0.51 0.83 0.12 0.36 0.36 Precautionary F level 0.65 0.70 0.65 0.40 0.30 0.40 0.25 0.17 0.59 0.84 Precautionary biomass level (kt) 150 140 315 200 300 35 311a 86a 600 150 . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 0.03 + 0.01 + 0.35 0.35 + + − − Fixed gear 0.10 + + + 0.02 0.02 + + − − Industrial (small meshed trawl) 0.02 0.01 0.04 0.05 + + 0.05 0.02 0.36 0.36 Pelagic (purse seine and trawl) + + + − + + 0.47 0.10 + + Saithe (trawl) 0.01 0.01 + 0.09 + − + + − − Seine net 0.21 0.30 0.16 0.01 0.05 + + + − − Trawl 0.28 0.35 0.27 0.16 0.05 0.03 0.11 0.01 + + Other gears 0.16 0.05 0.01 0.11 0.07 0.11 0.20 + 0.01 + All fleets 0.81 0.73 0.49 0.42 0.55 0.51 0.83 0.12 0.36 0.36 Precautionary F level 0.65 0.70 0.65 0.40 0.30 0.40 0.25 0.17 0.59 0.84 Precautionary biomass level (kt) 150 140 315 200 300 35 311a 86a 600 150 a Status-quo SSB used as a proxy for Bpa. Open in new tab Fishing mortality was assumed to be proportional to fishing effort. Different fishing efforts could then be modelled as multiples of the status-quo levels. Projections of yield and spawning stock biomass (SSB) were made for the status-quo F, and with changes in F of ±10, 25, and 50% both for all fleets combined, and by individual fleet. Each projection was run for 50 years to a (near) equilibrium state. The 4M model explicitly differentiates between retained and discarded catch. The multispecies production model was fit to the retained catch (yield) only. In this manner discarding is accounted for implicitly but is not tracked separately with the production model. Response-surface models were fit to the projections in which fishing effort for each fleet was increased by 10% in turn. The projections with other levels of fishing effort were used to compare the predictions of the simple response-surface model with those of the age-structured 4M model. The response-surface model is a generalization of the system of equations examined by Larkin (1966). Specifically, it is a multispecies production model of the Schaefer form: (1) where SSBs is the SSB of species s, q is a catchability coefficient, and ɛg is fishing effort in fleet g. The equilibrium SSB is: (2) where as is the SSB of species s in the absence of fishing and parameter bgs=qgas/rs measures the reduction of SSBs per unit of fishing effort in fleet g. For convenience, fishing effort (ɛg) was scaled to equal one in a reference year (1995). For each species, SSB was predicted with the 4M model for the status-quo effort level and a 10% increase in the effort of each fleet in turn. With m fishing fleets and n species, these calculations can be expressed in matrix notation as: (3) where S is an (m+1)×n matrix of SSB values, E is an (m+1)×(m+1) effort-change matrix, and A is the (m+1)×n matrix of parameter estimates, with one column for each species. Matrix E represents the changes in effort levels, with 1.1 on the diagonal, except for the first element which is 1.0, corresponding to the a parameters. This system of equations was solved by inverting the effort-change matrix, E. Separate equations were used to predict equilibrium SSB or yield in weight. Equilibrium yield is the product of equilibrium biomass and fishing mortality. Equations analogous to Equations (2) and (3) can be written to predict the equilibrium yield of a given fleet, f: (4) Dividing by ɛf, yield per unit effort, YPUE is: (5) Here αfs is the unfished YPUE of species s in fleet f and βfgs measures the reduction in YPUE per unit of fishing effort. In matrix notation, (6) where Uf is an (m+1)×n matrix of YPUE values and E is the same effort-change matrix. The (m+1)×n matrix of parameter estimates, Pf, can be estimated from the inverse of E, and once obtained it can be used to predict yield for different levels of fishing effort. A separate yield model was estimated for each fishing fleet. Fishing mortality rates corresponding to different combinations of fishing effort were calculated as: (7) where the partial fishing mortalities in 1995 were taken from Table 2. AMOEBA plots were used to display changes in SSB, yield, and fishing mortality resulting from changes in fishing effort and the resultant changes in species interactions. Angles for plotting the AMOEBAs were calculated with principal components analysis (PCA) of the table of yields by species and fleets, projected with status-quo effort levels. The PCA was calculated from the correlation matrix so that species and fleets with high yields would not dominate the principal components (PCs). We plotted the first two PCs using polar coordinates. The PCA loadings gave the angles for the fishing fleets and the PCA scores gave the corresponding angles for the fish species. The AMOEBAs were then used to investigate the consequences of different combinations of fishing effort. Effort levels corresponding to the multispecies maximum sustainable yield (MSY) can be found by maximizing the yield of each fleet as defined in Equation (4). Let (8) be the aggregate values for fleet f summed over species. Then aggregate yield can be expressed as (9) The partial derivative of yield with respect to effort in fleet f is: (10) The multispecies MSY is obtained when these partial derivatives are set to zero for all fleets simultaneously (Pope, 1989). Let B be the m×m matrix with −2βff on the main diagonal and −βfg in the remainder. In vector notation, MSY is obtained when (11) and the vector Emsy can be obtained from the inverse of B. Though easy to calculate, Emsy, is not a very useful reference point because it ignores the costs inherent in increasing fishing effort. Fishing costs are unknown, but their effect can be approximated by assuming that at status-quo effort levels, fishing costs equal the revenue or yield (Pope, 1997; Gislason, 1999). Most of the commercially important fish stocks in the North Sea suffer from overfishing and the majority of stocks are currently below the precautionary spawning biomass limits defined by ICES (ICES, 2002b). The North Sea fisheries are operating at a level exceeding that necessary to produce the maximum return both in tons caught and economic value. Although the situation differs from fleet to fleet and good economic data are lacking, the financial returns from fisheries are in many cases modest despite considerable subsidies. Financial returns fluctuate from year to year and in some cases costs and capital depreciation exceed revenues (European Commission, 2001). It is therefore not unreasonable to assume that the North Sea fisheries by and large are close to the level of effort corresponding to the bionomic equilibrium of the Gordon–Schaefer bio-economic fisheries model. This level is defined as the level of effort where opportunity costs would equal revenues in an equilibrium situation (Clark, 1985). Opportunity costs include costs due to fuel, gear and supplies, interest and depreciation on capital, as well as wages of skipper and crew. Assuming that each fleet is at the bionomic equilibrium, fishing costs equals revenue and the effort levels for maximum economic yield (Emey) can be calculated from: (12) where Ysq is the vector of status-quo yields for each fleet. Further constraints on the effort levels may be required to ensure that the SSBs of all species are above the precautionary levels (Bpa) and that the fishing mortality rates are below Fpa. Bounded nonlinear optimization was used to identify a set of effort levels to maximize yield while ensuring SSB≥Bpa and F≤Fpa for all species. Results The predictions of equilibrium SSB and yield made with the multispecies production model agree well with those of the 4M model for either a 50% increase or decrease in effort for all fleets (Figure 1). The differences between models are insubstantial (the points are largely superimposed) compared with the differences between the effort scenarios. With −50% effort the SSB of most species would be higher except for haddock, mackerel, Norway pout, and whiting (Figure 1a). For these species, an increase in predation mortality appears to compensate for the decrease in fishing mortality. With +50% fishing effort, the multispecies production model predicts the elimination of cod and herring, whereas the 4M model predicts small, but positive, SSB. With decreased fishing effort, yields of cod and herring were higher due to an increase in abundance (Figure 1b). Yields of haddock, Norway pout, sandeel, and whiting were lower, due to increased predation and lower fishing effort. With increased fishing effort, yield in the industrial fleet was lower, due to less sandeel while yield in the pelagic fleet was higher, reflecting the increase in herring (Figure 1c). In these validation runs, the multispecies production model was used to predict conditions other than the data that were used to estimate the model parameters (+10% effort). This close agreement indicates that the simpler response-surface model captures the main dynamics of the fishery and can be used to investigate different effort combinations, within a range of ±50% around the status-quo levels. Figure 1 Open in new tabDownload slide Predictions of the 4M model and the multispecies production model with 50% reduction or 50% increase in effort in all fishing fleets. The species abbreviations are defined in Table 1 and the fleet abbreviations are defined in Table 4. Figure 1 Open in new tabDownload slide Predictions of the 4M model and the multispecies production model with 50% reduction or 50% increase in effort in all fishing fleets. The species abbreviations are defined in Table 1 and the fleet abbreviations are defined in Table 4. The multispecies projections incorporate both technical and biological interactions (Table 3). Technical interactions occur because most fishing gears catch more than one species. Beam trawls catch sole and plaice and thus the SSB of both species would increase with a decrease in effort in the beam trawl fleet. Biological interactions occur because of predation among the modelled species. For example a decrease in seine effort would lead to an increase in the SSB of the predator cod but would decrease herring SSB because of increased predation. These species interactions can be plotted in three dimensions for a single species and pairs of fleets. Haddock SSB would increase with decreased effort in the trawl fishery and decrease with decreased effort in the industrial fishery due to increased abundance of the predators cod, whiting, and saithe (Figure 2). In this example, there was also close agreement between projections made with the multispecies production model and with the 4M model; the maximum difference between the two response surfaces was 6% (Figure 2). Figure 2 Open in new tabDownload slide Response-surface model for haddock SSB as a function of fishing effort in the trawl and industrial fleets. Fishing effort is expressed relative to 1 for the status quo and effort in the remaining six fishing fleets was fixed at 1. The flat plane corresponds to SSB predicted with the multispecies production model; the broken lines indicate the curved surface predicted by 4M model for effort levels at the edges of the plane. Figure 2 Open in new tabDownload slide Response-surface model for haddock SSB as a function of fishing effort in the trawl and industrial fleets. Fishing effort is expressed relative to 1 for the status quo and effort in the remaining six fishing fleets was fixed at 1. The flat plane corresponds to SSB predicted with the multispecies production model; the broken lines indicate the curved surface predicted by 4M model for effort levels at the edges of the plane. Table 3 Percent change in SSB of each species resulting from a 25% decrease in fishing effort of each fleet in turn, as estimated with the 4M model. Negative values result from increases in predator populations. Listed at the bottom are the scores of the first two PCs of the PCA of yields by species and fleet projected with status-quo effort levels. . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 2.89 −0.77 −0.14 0.05 19.80 25.41 −0.39 0.00 0.14 −0.03 Fixed gear 5.81 −0.62 −0.74 0.13 0.61 1.30 −0.10 0.00 0.22 0.40 Industrial (small meshed trawl) 4.47 −2.12 3.48 4.15 0.05 0.04 17.85 1.91 12.90 7.84 Pelagic (purse seine and trawl) 0.71 0.12 0.23 0.01 0.01 0.00 14.99 12.73 −0.37 −0.13 Saithe (trawl) 0.97 0.25 −0.10 6.45 0.08 0.01 −0.12 0.03 −0.02 0.21 Seine net 13.24 5.36 −0.20 1.34 2.46 0.06 −3.34 0.00 −0.21 −2.30 Trawl 20.93 3.57 0.23 12.03 3.02 1.86 −0.30 1.47 −0.02 −2.18 Other gears 13.82 0.01 −1.39 8.17 4.12 8.48 15.83 0.02 0.64 0.90 Principal component 1 scores 1.81 2.00 0.12 1.71 −1.00 −1.21 1.36 −1.17 −2.18 −1.42 Principal component 2 scores 1.34 1.45 0.73 −0.23 0.85 0.33 −3.87 0.11 −0.80 0.08 . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 2.89 −0.77 −0.14 0.05 19.80 25.41 −0.39 0.00 0.14 −0.03 Fixed gear 5.81 −0.62 −0.74 0.13 0.61 1.30 −0.10 0.00 0.22 0.40 Industrial (small meshed trawl) 4.47 −2.12 3.48 4.15 0.05 0.04 17.85 1.91 12.90 7.84 Pelagic (purse seine and trawl) 0.71 0.12 0.23 0.01 0.01 0.00 14.99 12.73 −0.37 −0.13 Saithe (trawl) 0.97 0.25 −0.10 6.45 0.08 0.01 −0.12 0.03 −0.02 0.21 Seine net 13.24 5.36 −0.20 1.34 2.46 0.06 −3.34 0.00 −0.21 −2.30 Trawl 20.93 3.57 0.23 12.03 3.02 1.86 −0.30 1.47 −0.02 −2.18 Other gears 13.82 0.01 −1.39 8.17 4.12 8.48 15.83 0.02 0.64 0.90 Principal component 1 scores 1.81 2.00 0.12 1.71 −1.00 −1.21 1.36 −1.17 −2.18 −1.42 Principal component 2 scores 1.34 1.45 0.73 −0.23 0.85 0.33 −3.87 0.11 −0.80 0.08 Open in new tab Table 3 Percent change in SSB of each species resulting from a 25% decrease in fishing effort of each fleet in turn, as estimated with the 4M model. Negative values result from increases in predator populations. Listed at the bottom are the scores of the first two PCs of the PCA of yields by species and fleet projected with status-quo effort levels. . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 2.89 −0.77 −0.14 0.05 19.80 25.41 −0.39 0.00 0.14 −0.03 Fixed gear 5.81 −0.62 −0.74 0.13 0.61 1.30 −0.10 0.00 0.22 0.40 Industrial (small meshed trawl) 4.47 −2.12 3.48 4.15 0.05 0.04 17.85 1.91 12.90 7.84 Pelagic (purse seine and trawl) 0.71 0.12 0.23 0.01 0.01 0.00 14.99 12.73 −0.37 −0.13 Saithe (trawl) 0.97 0.25 −0.10 6.45 0.08 0.01 −0.12 0.03 −0.02 0.21 Seine net 13.24 5.36 −0.20 1.34 2.46 0.06 −3.34 0.00 −0.21 −2.30 Trawl 20.93 3.57 0.23 12.03 3.02 1.86 −0.30 1.47 −0.02 −2.18 Other gears 13.82 0.01 −1.39 8.17 4.12 8.48 15.83 0.02 0.64 0.90 Principal component 1 scores 1.81 2.00 0.12 1.71 −1.00 −1.21 1.36 −1.17 −2.18 −1.42 Principal component 2 scores 1.34 1.45 0.73 −0.23 0.85 0.33 −3.87 0.11 −0.80 0.08 . Species . Fleet . Cod . Haddock . Whiting . Saithe . Plaice . Sole . Herring . Mackerel . Sandeel . Norway pout . Beam trawl 2.89 −0.77 −0.14 0.05 19.80 25.41 −0.39 0.00 0.14 −0.03 Fixed gear 5.81 −0.62 −0.74 0.13 0.61 1.30 −0.10 0.00 0.22 0.40 Industrial (small meshed trawl) 4.47 −2.12 3.48 4.15 0.05 0.04 17.85 1.91 12.90 7.84 Pelagic (purse seine and trawl) 0.71 0.12 0.23 0.01 0.01 0.00 14.99 12.73 −0.37 −0.13 Saithe (trawl) 0.97 0.25 −0.10 6.45 0.08 0.01 −0.12 0.03 −0.02 0.21 Seine net 13.24 5.36 −0.20 1.34 2.46 0.06 −3.34 0.00 −0.21 −2.30 Trawl 20.93 3.57 0.23 12.03 3.02 1.86 −0.30 1.47 −0.02 −2.18 Other gears 13.82 0.01 −1.39 8.17 4.12 8.48 15.83 0.02 0.64 0.90 Principal component 1 scores 1.81 2.00 0.12 1.71 −1.00 −1.21 1.36 −1.17 −2.18 −1.42 Principal component 2 scores 1.34 1.45 0.73 −0.23 0.85 0.33 −3.87 0.11 −0.80 0.08 Open in new tab Ordination was used to project the entire table of fleet-by-species interactions in two dimensions. The first two PCs accounted for 28 and 26% of the variance; the third and forth components explained only 15 and 13%. The species and fleets are well separated in two dimensions, expect whiting, which had low scores for PC1 and PC2 (Table 3), possibly because of its important role as both predator and prey species. Cod and haddock had similar scores for the first two PCs (Table 3), but different scores for PC3. Likewise sole, mackerel, and Norway pout clustered on the first two PCs and separated on the third, as they are caught by different fleets. With respect to fleets, some pairs had similar loading for the first two PCs (Table 4) and different loading for PC3 (fixed and seine) or PC4 (trawl and saithe; pelagic and other). Thus the PCA provides a useful ordination of species and fleets but there is additional variation not explained by the first two PCs. Table 4 Reference levels of fishing effort identified with the multispecies Schaefer model. Also listed are the loadings for the first two PCs of the PCA of yields by species and fleet projected with status-quo effort levels. . Effort relative to status quo . . PCA loadings . Fleet . Emsy . Emey . Epa . Abbreviations for plotting . PC1 . PC2 . Beam trawl 0.98 0.53 0.55 btr −0.176 0.159 Fixed gear 0.2a 2.08 0.73 fix 0.255 0.254 Industrial (small meshed trawl) 1.77 0.51 0.22 ind −0.320 −0.235 Pelagic (purse seine and trawl) 0.81 0.82 0.21 pel 0.187 −0.619 Saithe (trawl) 3.02 0.96 0.54 sth 0.287 −0.008 Seine net 2.39 0.71 0.56 sei 0.407 0.376 Trawl 1.42 0.86 0.24 trl 0.642 0.063 Other gears 1.11 0.86 0.54 oth 0.327 −0.572 . Effort relative to status quo . . PCA loadings . Fleet . Emsy . Emey . Epa . Abbreviations for plotting . PC1 . PC2 . Beam trawl 0.98 0.53 0.55 btr −0.176 0.159 Fixed gear 0.2a 2.08 0.73 fix 0.255 0.254 Industrial (small meshed trawl) 1.77 0.51 0.22 ind −0.320 −0.235 Pelagic (purse seine and trawl) 0.81 0.82 0.21 pel 0.187 −0.619 Saithe (trawl) 3.02 0.96 0.54 sth 0.287 −0.008 Seine net 2.39 0.71 0.56 sei 0.407 0.376 Trawl 1.42 0.86 0.24 trl 0.642 0.063 Other gears 1.11 0.86 0.54 oth 0.327 −0.572 a Constrained to prevent a negative estimate. Open in new tab Table 4 Reference levels of fishing effort identified with the multispecies Schaefer model. Also listed are the loadings for the first two PCs of the PCA of yields by species and fleet projected with status-quo effort levels. . Effort relative to status quo . . PCA loadings . Fleet . Emsy . Emey . Epa . Abbreviations for plotting . PC1 . PC2 . Beam trawl 0.98 0.53 0.55 btr −0.176 0.159 Fixed gear 0.2a 2.08 0.73 fix 0.255 0.254 Industrial (small meshed trawl) 1.77 0.51 0.22 ind −0.320 −0.235 Pelagic (purse seine and trawl) 0.81 0.82 0.21 pel 0.187 −0.619 Saithe (trawl) 3.02 0.96 0.54 sth 0.287 −0.008 Seine net 2.39 0.71 0.56 sei 0.407 0.376 Trawl 1.42 0.86 0.24 trl 0.642 0.063 Other gears 1.11 0.86 0.54 oth 0.327 −0.572 . Effort relative to status quo . . PCA loadings . Fleet . Emsy . Emey . Epa . Abbreviations for plotting . PC1 . PC2 . Beam trawl 0.98 0.53 0.55 btr −0.176 0.159 Fixed gear 0.2a 2.08 0.73 fix 0.255 0.254 Industrial (small meshed trawl) 1.77 0.51 0.22 ind −0.320 −0.235 Pelagic (purse seine and trawl) 0.81 0.82 0.21 pel 0.187 −0.619 Saithe (trawl) 3.02 0.96 0.54 sth 0.287 −0.008 Seine net 2.39 0.71 0.56 sei 0.407 0.376 Trawl 1.42 0.86 0.24 trl 0.642 0.063 Other gears 1.11 0.86 0.54 oth 0.327 −0.572 a Constrained to prevent a negative estimate. Open in new tab In the AMOEBA plots, the directions of the fleet vectors correspond to the directions of the species caught by that fishing gear (Figure 3). It can easily be seen that sandeel is caught by the industrial fishery and that sole and plaice are caught with beam trawls. The orientation of these arrows makes it easier to see which species will be affected by changes in fishing effort of particular fishing fleets. The flatfish vectors are in the upper left quadrant, the roundfish in the upper right, and the pelagic species are scattered in the other quadrants. It should be stressed that the PCA was used only to aid in displaying the results of the multispecies model and in no way influences the multispecies projections. The angles derived from the PCA provide a more informative grouping of fleets and species in two dimensions than could be achieved in a one-dimensional plot or table, or by regular spacing of the vectors around the circle. Beyond their utility as a plotting device, the angles derived from PCA are not important to the projection; in fact some of the angles were jittered slightly to avoid overlapping the vectors. Figure 3 Open in new tabDownload slide AMOEBA plots with status-quo effort levels. In each AMOEBA the circle represents the reference level and the arrows are the levels predicted with the multispecies model. Effort and landings are plotted relative to the status quo. Fishing mortality is plotted relative to Fpa and SSB is plotted relative to Bpa. Species abbreviations are given in Table 1 and fleet abbreviations in Table 4. Figure 3 Open in new tabDownload slide AMOEBA plots with status-quo effort levels. In each AMOEBA the circle represents the reference level and the arrows are the levels predicted with the multispecies model. Effort and landings are plotted relative to the status quo. Fishing mortality is plotted relative to Fpa and SSB is plotted relative to Bpa. Species abbreviations are given in Table 1 and fleet abbreviations in Table 4. The vector lengths indicate the magnitude of each quantity relative to its status-quo or reference level. These quantities were calculated with the multispecies production model, independently of the PCA. With status-quo effort levels, plaice, whiting, and cod SSB would all be below their precautionary biomass (Bpa) levels, and plaice, sole, cod, saithe, and herring fishing mortality would exceed the precautionary (Fpa) levels (Figure 3). The MSY effort levels were much higher than status quo for several of the fleets (Table 4), but yields in the industrial, seine, and saithe fleets would increase only slightly. At Emsy, SSB would be below Bpa for all species except sole, mackerel, Norway pout, and sandeel, and fishing mortality would exceed Fpa for all species except mackerel and Norway pout. Effort levels for maximum economic yield (Emey) were all less than the status quo except for the fixed gear (Table 4). Fixed gear has a very different exploitation pattern than trawl and seine nets. The mean age of cod, plaice, and sole in fixed gear is at least 1 year older than the other gear types. Effort reduction in the other fleets would increase the fixed gear yield because fixed gear targets older fish. For the remaining fleets Emey was less than one because fishing costs would exceed revenues at higher effort levels. At Emey cod SSB would be below Bpa and fishing mortality would exceed Fpa for plaice, cod, and herring (Figure 4), but all the other biological constraints would be met. With bounded nonlinear optimization, a combination of effort levels was identified that would satisfy all the biological constraints while maximizing yield (Epa, Table 4). This combination required substantial reduction in the industrial, pelagic, and trawl fleets in order to raise cod SSB above Bpa and to decrease fishing mortality on cod and herring (Figure 5). At these precautionary effort levels, the roundfish would be at or near their Bpa levels, while the other species would be well above Bpa. Figure 4 Open in new tabDownload slide AMOEBA plots with effort levels for MEY. The features of the AMOEBAs are explained in the caption to Figure 3. Figure 4 Open in new tabDownload slide AMOEBA plots with effort levels for MEY. The features of the AMOEBAs are explained in the caption to Figure 3. Figure 5 Open in new tabDownload slide AMOEBA plots with precautionary effort levels. The features of the AMOEBAs are explained in the caption to Figure 3. Figure 5 Open in new tabDownload slide AMOEBA plots with precautionary effort levels. The features of the AMOEBAs are explained in the caption to Figure 3. The shapes of the AMOEBAs represent the composition of the fishery and the fish community. Compared with the status quo (Figure 3), the precautionary effort levels would reduce the industrial and pelagic fleets and shift the fish community toward the prey species, especially sandeel and herring (Figure 5). The area of the AMOEBA, relative to its area at the reference levels, could be considered for use as a community-level index. Likewise, the volume of the AMOEBA could be calculated to capture more of the variance in species catch by fleet. However, such a summary index is much less informative than looking at the actual AMOEBA. Discussion We have shown how AMOEBAs can be constructed and used to display the main interactions in a multispecies, multifleet fishery on a single page. These plots sufficiently capture the trade-offs in multiple fishery objectives. The biological objectives require satisfying the Bpa and Fpa constraints for each species. Our results indicate that these constraints can be jointly met even when predator–prey interactions are included. For the prey species, the benefits of decreased fishing mortality appear to outweigh the increased predation mortality that occurs with increased predator abundance. At the precautionary effort levels (Table 4) the SSB of all species would be higher than the status-quo levels. This result differs from earlier MSFOR projections in which the result of increased mesh size in the roundfish fleet was to decrease the yields of the prey species because of increased predation (Pope, 1991). The earlier MSFOR projections did not include stock-recruitment relationships, but assumed constant recruitment, and thus decoupled recruitment from fishing mortality. In the 4M model, recruitment of the prey species increases with lower fishing mortality and higher SSB. However, the stock-recruitment relationships remain uncertain because they were fit to short time series of variable data. The incorporation of the stock-recruitment relationships also tends to cause oscillations in the projected abundances. Economic objectives operate at the fleet level. We used the yield in weight of each fleet as a surrogate for economic performance. It would be preferable to express yield in monetary units to account for price differences among species. However, we lacked price data that were appropriately averaged over time, subfleets, and size of fish. Therefore, in our estimate of MEY we implicitly assumed that the value per weight of each fleet's catch would remain constant with different effort levels. Pope (1997) also found that attaining MEY would require reducing effort in the roundfish and industrial fleets, with the other fleets kept near their status-quo levels. Social objectives are usually expressed at a finer level of geographic detail (e.g. fishing ports) than the main fleets in our model. One approach would be to include an AMOEBA for social objectives (e.g. employment) and to assume that ports with similar gears would be similarly affected by effort changes (Pope, 1997). However, if there are substantial national differences, even within the main gear groups, a two-tiered approach may be required. A coarser management model with aggregated fishing fleets would operate at the international level. The output from this model would then be made available to national groups to make second-stage models at further levels of disaggregation (Pope, 1997). In this study, we used a multispecies Schaefer model to describe the North Sea multispecies fishery. It was at first surprising that a simple production model could match the 4M model predictions so closely. However, the multispecies Schaefer model was fit to the 4M projections, and the simpler model appears to capture the main interactions. In this manner, simplified management advice can be given without further need for a more detailed biological model. The multispecies Schaefer model was very convenient for this application because the projections can be made almost instantly, which facilitates an interactive computer model. The model projections should be most reliable close to the status-quo effort levels; the MSAWG cautioned against extrapolating beyond a range of one half or twice the status-quo effort (ICES, 1992). We found very close agreement between the model predictions within a range of ±50% of status-quo effort levels for all fleets. At +50% effort the model projections began to diverge because the multispecies production model predicted extinction of cod and herring, whereas the 4M model predicted that both species would persist at low levels. The essential features of this display are that the AMOEBAs are linked with a multispecies model and that projections can be made simply by altering effort levels. Alternative model formulations could be used and/or extensions made to the present model. One approach would be to use the 4M model to make all the projections, without fitting the simpler production model. However, detailed accounting of age structure may be unnecessary unless changes in mesh size are investigated or there is a large price differential with size of fish. In this study the biological interactions appeared to be secondary to the direct fishery effects on each species. However, we may have down-played the biological interactions by ignoring variations in the predation mortality inflicted by the “other predators” in Table 1. Also, we did not consider the potential bottom-up effects of the prey species on the growth rates of their predators. Prey-dependent growth has been incorporated in other multispecies models (e.g. Gislason, 1999) but is thought to be less important in the North Sea because of a large variety of alternative prey species. An alternative approach is to fit the multispecies production model directly to catch and abundance data without accounting for age structure (e.g. Collie and DeLong, 1999). Preliminary attempts to fit a multispecies production model to North Sea data have not been successful even though the species were grouped into three larger functional units and MSVPA derived biomasses and catches were used as input. As concluded by Sullivan (1991) the number of parameters in multispecies production models is so large that very long data series often are needed to fit even the simplest multispecies system. Parameter estimation may be facilitated by introducing auxiliary data on fishing effort, recruitment indices, mean weight of fish in the stock, growth parameters, residual natural mortality, and food composition. Such an approach was followed by Horbowy (1996), who derived a multispecies production model for Baltic cod, herring, and sprat from the age-structured multispecies model of Andersen and Ursin (1977) and obtained biomass estimates consistent with results from age-structured models. The multispecies Schaefer model was fit to equilibrium conditions and therefore did not consider the time dynamics of moving from the status quo to the desired situation. These equilibrium solutions give useful targets relative to present conditions, but in practice it would be useful to have AMOEBAs for 1–5 year projections as well. It would also be useful to incorporate stochasticity, especially to account for uncertainty in the stock-recruitment relationships. If a stochastic multispecies model was used, the arrowheads in each AMOEBA could be replaced with error bars. Uncertainty in the ordination could be represented with wedges in place of the arrows. The bio-economic objectives could also be extended, for example by including effort–cost relationships and price elasticity; such extensions would give a higher value to reducing fishing effort. Social objectives could also be represented with AMOEBAs but the challenge is that, to be included, they must be quantified (Pope, 1997). While all these extensions are technically feasible, additions to the AMOEBA plots should only be made if they serve a reasonably clear purpose. The combinations of effort levels in Table 4 were meant more for illustration than for prognostication. The partial fishing mortalities were based on 1991 values (Table 2); fishing patterns have almost certainly changed since then. Before making actual projections, the status-quo effort levels would need to be updated from 1995 to present. Nevertheless, several general conclusions can be made regarding multispecies biological reference points. Joint levels of F0.1 and Fmsy can be calculated with the methods of linear algebra, but they are of limited usefulness because of the tendency for extremely high or low values for some fleets. Reference levels based on MEY appear to be more useful because they prevent extreme effort levels and because of the explicit link to bioeconomics. In the Baltic Sea, multispecies reference points based on MEY were also more sensible than the joint F0.1 and Fmsy levels (Gislason, 1999). A priority should be to incorporate more realistic cost functions. Our results suggest that it is possible to satisfy the Bpa and Fpa levels of all species but that substantial reductions in fishing effort of some fleets would be necessary. Relative to the status quo, there would also be foregone yield, although this loss would at least be partially compensated by increased catch per unit effort. In reality, we should not rely on projections with effort levels less than one half the status quo because they imply levels of stock abundance far different than those used to fit the models. Fishing effort is more likely to be reduced in a step-wise fashion, with multispecies models refit at each step. The AMOEBA plots are very useful for displaying the trade-offs among biological, economic, and social objectives. It is unlikely that any “optimal” effort combination will be chosen. More realistically, solutions will be sought that maximize the objectives while violating as few constraints as possible (Pope, 1997). The advantage of the AMOEBA approach is that these trade-offs can be viewed explicitly. We have also developed an interactive version of the program in which the effort levels can be changed in the graphical interface. In summary, we have demonstrated a method for the clear and concise presentation of advice for a multispecies, multifleet fishery. Incorporation of biological interactions does require a multispecies model, but the presentation of advice is no more complex than that required for the technological interactions among fishing fleets. We thank Allison DeLong, Terry Quinn, Marie-Joëlle Rochet, and Stephen Hall for helpful comments on earlier drafts of this paper. The research was funded by the European Commission (Project QLK5-CT1999-01609) and the Danish Ministry of Food, Agriculture and Fisheries. J.S.C. thanks the Danish Institute for Fisheries Research for hosting his sabbatical leave and Anders Nielsen for help with the AMOEBA plots. References Andersen K.P. , Ursin E. . A multispecies extension to the Beverton and Holt theory of fishing, with accounts of phosphorous circulation and primary production , Meddelelser fra Danmarks Fiskeri- og Havundersogelser , 1977 , vol. 7 (pg. 319 - 435 ) OpenURL Placeholder Text WorldCat Anon , 1991 Report of the meeting of the STCF working group on improvements of the exploitation pattern of the North Sea fish stocks June 1991 Charlottenlund Castle 166 pp Bax N. J. , Williams A. , Davenport S. , Bulman C. . Managing the ecosystem by leverage points: a model for a multispecies fishery , Ecosystem Approaches for Fisheries Management , 1999 (pg. 283 - 303 ) Alaska Sea Grant College Progam, AK-SG-99-01 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brugge W. J. , Holden M. J. . Daan N. , Sissenwine M. P. . Multispecies management: a manager's point of view , Multispecies Models Relevant to Management of Living Resources , 1991 (pg. 353 - 358 ) ICES Marine Science Symposium, 193 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Clark C.W. . , Bioeconomic Modelling and Fisheries Management , 1985 New York Wiley 291 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Collie J.S. . Steele J. , Thorpe S. , Turekian K. . Fisheries multispecies dynamics , Encyclopedia of Ocean Sciences , 2001 London Academic Press (pg. 1002 - 1009 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Collie J. S. , DeLong A. K. . Multispecies interactions in the Georges Bank fish community , Ecosystem Approaches for Fisheries Management , 1999 (pg. 187 - 210 ) Alaska Sea Grant College Progam, AK-SG-99-01 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Collie J.S. , Gislason H. . Biological reference points in a multispecies context , Canadian Journal of Fisheries and Aquatic Sciences , 2001 , vol. 58 (pg. 2167 - 2176 ) Google Scholar Crossref Search ADS WorldCat European Commission Green paper on the future of the common fisheries policy , 2001 Commission of the European Communities. COM (2001) 135 final, Brussels 20.03.2001. 40 pp Gislason H. . Single and multispecies reference points for Baltic fish stocks , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 571 - 583 ) Google Scholar Crossref Search ADS WorldCat Hall S.J. . , The Effects of Fishing on Marine Ecosystems and Communities , 1999 Oxford, UK Blackwell 274 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Halliday R.H. , Pinhorn A.T. . North Atlantic fisheries management systems: a comparison of management methods and resource trends , Journal of Northwest Atlantic Fisheries Science , 1996 , vol. 20 pg. 135 OpenURL Placeholder Text WorldCat Holden M. . , The Common Fisheries Policy. Origin, Evaluation and Future , 1994 Oxford Fishing News Books/Blackwell 274 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hollowed A.B. , Bax N. , Beamish R. , Collie J.S. , Fogarty M. , Livingston P.A. , Pope J. , Rice J.C. . Are multispecies models an improvement on single-species models for measuring fishing impacts on marine ecosystems? , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 707 - 719 ) Google Scholar Crossref Search ADS WorldCat Horbowy J. . The dynamics of Baltic fish stocks on the basis of a multispecies stock-production model , Canadian Journal of Fisheries and Aquatic Science , 1996 , vol. 53 (pg. 2115 - 2125 ) Google Scholar Crossref Search ADS WorldCat ICES , 1992 Report of the Multispecies Assessment Working Group. ICES Headquarters, 16–25 June 1992. ICES CM 1992/Assess: 16. 152 pp ICES , 1997 Report of the Multispecies Assessment Working Group. ICES Headquarters, 11–19 August 1997. ICES CM 1997/Assess: 16. 235 pp ICES , 2001 Report of the Working Group on Ecosystem Effects of Fishing Activities. ICES CM 2001/ACME: 09. 102 pp ICES , 2002 Report of the Working Group on Ecosystem Effects of Fishing Activities. ICES CM 2002/ACE: 03. 189 pp ICES , 2002 Report of the ICES Advisory Committee on Fisheries Management. ICES Cooperative Research Report, No 255. 948 pp Laane W.E.M. , Peters J.S. . Ecological objectives for management purposes: applying the amoeba approach , Journal of Aquatic Ecosystem Health , 1993 , vol. 2 (pg. 277 - 286 ) Google Scholar Crossref Search ADS WorldCat Lanters R. L. P. , Skjoldal H. R. , Noji T. T. . Ecological Quality Objectives for the North Sea , 1999 RIKZ/Institute of Marine Research, Report 99.015, The Hague, The Netherlands. 63 pp Larkin P.A. . Exploitation in a type of predator–prey relationship , Journal of the Fisheries Research Board of Canada , 1966 , vol. 23 (pg. 349 - 356 ) Google Scholar Crossref Search ADS WorldCat Lewy P. , Vinther M. , Thomsen L. . Description of the STCF North Sea database system and the prediction model ABC, assessments of bio-economic consequences of technical measures , 1992 ICES CM 1992/D: 17. 19 pp Livingston P.A. , Tjelmeland S. . Fisheries in boreal ecosystems , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 619 - 627 ) Google Scholar Crossref Search ADS WorldCat May R.M. , Beddington J.R. , Clark C.W. , Holt S.J. , Laws R.M. . Management of multispecies fisheries , Science , 1979 , vol. 405 (pg. 267 - 276 ) Google Scholar Crossref Search ADS WorldCat Murawski S. . Definitions of overfishing from an ecosystem perspective , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 649 - 658 ) Google Scholar Crossref Search ADS WorldCat Pope J. G. . Simple representations of the yield surface , 1989 Working paper for the 1989 meeting of the ICES Multispecies Assessment Working Group 4 pp Pope J. G. . Daan N. , Sissenwine M. P. . The ICES Multispecies Assessment Working Group: evolution, insights, and future problems , Multispecies Models Relevant to Management of Living Resources , 1991 (pg. 22 - 33 ) ICES Marine Science Symposium, 193 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pope J.G. . Montero C.C. . Integrating and presenting scientific fisheries management advice for complex fisheries subject to multiple objectives , Multiple Objectives and Fisheries Management: Strategies for the Future , 1997 Lisboa, Portugal Instituto de Investigaçao das Pescas e do Mar, Avenida de Brasília (pg. 29 - 53 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rice J. . Evaluating fishery impacts using metrics of community structure , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 682 - 688 ) Google Scholar Crossref Search ADS WorldCat Sullivan K. J. . Daan N. , Sissenwine M. P. . The estimation of parameters of the multispecies production model , Multispecies models relevant to management of living resources , 1991 (pg. 185 - 193 ) ICES Marine Science Symposium, 193 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ten Brink B.J.E. , Hosper S.H. , Colijn F. . A quantitative method for description and assessment of ecosystems: the AMOEBA approach , Marine Pollution Bulletin , 1991 , vol. 23 (pg. 265 - 270 ) Google Scholar Crossref Search ADS WorldCat Tufte E.R. . , The Visual Display of Quantitative Information , 1983 Cheshire, CT Graphics Press 197 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Vinther M. , Lewy P. , Thomsen L. . , Specification and Documentation of the 4M Package Containing Multi-species, Multi-Fleet and Multi-Area Models , 2001 Charlottenlund Danish Institute for Fisheries Research 70 pp. (contact [email protected]) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wefering F.M. , Danielson L.E. , White N.M. . Using the AMOEBA approach to measure progress toward ecosystem sustainability within a shellfish restoration project in North Carolina , Ecological Modelling , 2000 , vol. 130 (pg. 157 - 166 ) Google Scholar Crossref Search ADS WorldCat © 2003 International Council for the Exploration of the Sea

Rueda,, Mario;Defeo,, Omar

doi: 10.1016/S1054-3139(03)00096-1pmid: N/A

Abstract We estimated the spatial population structure by size and the economical potential (biovalue) of a multispecies fish resource in an estuarine lagoon in Colombia, based on fishery-dependent (catch and effort) and independent (seasonal fishing surveys) data. Model-based (geostatistics, kriging) estimations of such performance variables were used to quantify the uncertainty in abundance, individual price by size and variable costs per haul. Monte Carlo analysis was used to assess the status of the fishery. The spatial dimension of risk analysis was explored by indicator kriging, whereas effects of biovalue on the spatial allocation of fishing effort were evaluated using contingency tables. Fish abundance, individual sizes and biovalue were spatially structured, but the spatial patterns varied between seasons and species. Analysis of biovalue showed a moderate risk that fishers had economic losses derived from the fishing activity. Spatial risk analysis showed that no more than 30% of the total area from the lagoon registered profitable levels of fish abundance, which affected the spatial allocation of fishing effort. Management implications supported by our study suggest seasonal and spatial fishing closures to protect juveniles and spawning stock of fish species. Introduction The spatial structure of the environment and of the biological communities is not only one of the most important determinants of ecosystem functioning (Legendre, 1993), but it also defines the spatial allocation of fishing effort which affects fishery management (Castilla and Defeo, 2001). Indeed, spatially explicit analyses of exploited fish populations have demonstrated a strong spatial structure in abundance and in the fishing process (Pelletier and Parma, 1994; Orensanz and Jamieson, 1998; Caddy and Carocci, 1999; Rueda, 2001). However, the integrated analysis of the spatial dynamics of the composition by size and population abundance, jointly with concurrent spatio-temporal changes in the economic benefit, has been rarely documented (Anderson, 1989; Caddy, 1999; Pauly et al., 2001). This is relevant in small-scale fisheries located on tropical estuaries, where the unselective harvest of a multispecies stock of short-lived species (Blaber, 1997) together with economic pressures, cause fishers to allocate fishing effort in areas closer to port (Defeo and Castilla, 1998; Cabrera and Defeo, 2001). The application of spatial models, both analytical (Hilborn and Walters, 1987; Caddy and Seijo, 1998) and empirical (Caddy and Carocci, 1999; Taconet and Bensch, 2000) arise as a powerful tool for stock assessment and fishery management. In this setting, geostatistics has obtained increasing acceptance to address the spatial structure of variables observed across geographical space. Such applications in marine science have been mainly used to describe spatial patterns of benthic and pelagic species (Maravelias et al., 1996; Rueda, 2001; Defeo and Rueda, 2002) and to interpolate abundance at unsampled locations (Petitgas, 1993; Maynou et al., 1998; Rueda and Defeo, 2001). In this study we integrate biological and economic information for modeling and interpolate the spatial structure of fish abundance, individual size and the economic potential of the abundance (biovalue) in the Ciénaga Grande de Santa Marta (CGSM). This is a multispecies (Rueda and Urban, 1998; Sánchez and Rueda, 1999) and multifleet fishery (Rueda and Defeo, in press) based on the exploitation of the fishes Eugerres plumieri, Mulgil incilis and Cathorops spixii, whose harvestable biomass is mainly formed by juveniles (Rueda and Defeo, 2001). In addition, the effect of the abundance of different population components on the biovalue was assessed to quantify the uncertainty in fishery performance variables and to estimate the probability of exceeding limit reference points (LRPs) defined in terms of critical threshold values of biovalue. Potential consequences of biovalue spatial dimension on the spatial allocation of fishing effort are also explored. Methods Study area and sampling The CGSM is the most important estuarine lagoon of Colombia (Figure 1), because its fishery constitutes the main source of both food and income for ca. 20 000 persons (3500 fishers: Botero and Salzwedel, 1999). Four alternating seasons affect the life history traits of the fish fauna in the CGSM (Rueda and Santos-Martínez, 1999; Sánchez and Rueda, 1999); however, for the purpose of this study, we only considered data gathered in the contrasting rainy and dry seasons. One fishing survey was conducted in November 1993 (rainy season) and another in March 1994 (dry season), based on a systematic design of 115 stations spaced 2000 m apart and located using a GPS, covering the whole CGSM (Figure 1). At each station, a haul was carried out using a “boliche” or encircling gillnet, which enclosed an average circular area of 5000 m2. Eight “boliches” were used simultaneously to conduct each survey, which took approximately 8 h to be completed (see Rueda, 2001; Rueda and Defeo, 2001 for details). Individuals of E. plumieri, M. incilis and C. spixii collected per station were counted and measured for total length to the nearest 0.5 cm. Information on fishing effort of the “boliche” and bioeconomic variables was obtained from a fishery monitoring program conducted between 1993 and 1994 (Santos-Martínez and Viloria, 1998). Effort data (number of trips day−1) and location of fishing grounds were recorded during November 1993 and March 1994 using random sampling of commercial catches. Bioeconomic variables gathered per month focused on individual price by size (measurement of length–frequency samples and collection of size-at-price data). Variable costs of the “boliche” (VC; gas, gear repair, food and beverages) were obtained from logbooks and interviews with the fishers from the main port on the north of the CGSM (Tasajera, see Figure 1), where 80% of the “boliche” activity was carried out. Additional information on national minimum salary for 1993 and 1994 was gathered from Colombian government statistics (Banco de la República, 2002). Figure 1 Open in new tabDownload slide CGSM—Colombia, showing the fixed grid of 115 stations (+) sampled during rainy (1993) and dry (1994) seasons. (○) Denotes fishing ports, highlighting the port of Tasajera (●). Figure 1 Open in new tabDownload slide CGSM—Colombia, showing the fixed grid of 115 stations (+) sampled during rainy (1993) and dry (1994) seasons. (○) Denotes fishing ports, highlighting the port of Tasajera (●). Differences in fish abundance between seasons for each species were tested by the Kruskal–Wallis test by ranges, because the normality assumption was not fulfilled. Concerning the population structure by size, length–frequency distributions of E. plumieri, M. incilis and C. spixii were subjected to Kolmogorov–Smirnov (K–S) goodness of fit to test for differences between seasons. The length–price (L–P) relationship was modeled by the power function of the form P=aLb, where a and b are parameters, and were employed to calculate the price for each average size estimated per sampling station. An analysis of covariance (ANCOVA) was employed to compare the L–P relationship fitted for each species between seasons, using length as the covariate. Assessing spatial structure of fish abundance, size and biovalue Variographic analysis (Matheron, 1971) was used to characterize the spatial structure of fish abundance, size and biovalue. Each variable was considered as a spatial process observed in each season by means of 115 observations measured at a location x, defined by latitude and longitude in a two-dimensional space. Seasonal experimental semivariograms were estimated. Confirmation of a second-order stationarity assumption and assessing the possibility of isotropic and anisotropic processes were conducted by computing semivariogram surfaces (Isaaks and Srivastava, 1989). Structure functions for fish abundance, size and biovalue were estimated fixing the active lag distance to 20 400 m (65% of the maximum lag). This was done after looking for the large number of pairs available to estimate semivariograms which avoids the situation where they decompose at large lag intervals close to the maximum lag interval (Robertson, 2000). Calculated mean individual size per station was used to estimate experimental semivariograms of fish size, whereas the spatial autocorrelation of biovalue took into account species-specific information on individual price by size. The seasonal biovalue (BVijkl) for each species was calculated as BVijkl=(NP)ijkl, where N and P are the fish abundance and price, respectively, of the species i by size j at station k in season l. In all cases, theoretical models were fitted to the experimental semivariograms to relate the observed structure to hypothesized generating processes (Isaaks and Srivastava, 1989). The model that best explained the spatial structure was selected according to the coefficient of determination (r2) and the residual sum of squares. Semivariogram models provided the following parameter estimates: (1) the nugget effect (C0), which reflects microscale variation; (2) the sill (C0+C), which defines the asymptotic plateau in the semivariance; and (3) the range (A0), defined as the distance at which the variables ceases to be autocorrelated (Isaaks and Srivastava, 1989). An analysis of the residual sum of squares (ARSS; Chen et al., 1992) was performed to compare the semivariograms for fish abundance, size and biovalue fitted for each season. Spatial prediction of fish abundance, size and biovalue Maps of fish abundance, size and biovalue per species and seasons were obtained by block kriging (Matheron, 1971). Kriging predictions were evaluated using jackknife cross-validation, fitting observed (O) and estimated (E) values to a linear regression of the form O=α+βE and testing the significance of α and β (t-test) under the null hypotheses α=0 and β=1 (Power, 1993). Maps were performed over a regular interpolating grid of 424×414 m2 (internodal distances) covering the whole area of the CGSM (450 km2), whereas the neighborhood comprises at least the 16 nearest neighbors. Global biovalue and the standard deviation per species and seasons were computed as a linear combination of block estimates, with each estimate receiving a weight proportional to the sampled area (Journel and Huijbregts, 1978). Global estimates were corrected by the vulnerability of each species to the sampling gear experimentally estimated as 0.5, 0.43 and 0.4 for E. plumieri, M. incilis and C. spixii, respectively (Rueda and Defeo, 2001). Risk analysis and indicator probabilistic kriging We estimated multispecies biovalue (BV) by summing up the global biovalues obtained for each species, given by multiplying the mean fish abundance of block kriging analyses by its mean unit price per size. Thus, estimates of BV, VC and minimum threshold profit (Z) in each season were used to quantify the risk of falling below undesirable biovalue thresholds per haul. As a typical trip of the “boliche” involves around 20 hauls day−1 allocated to any site in the lagoon from any port (Rueda, 1995), VC corresponded to one global seasonal estimate weighed to a single haul. Z was fixed at 1.6 times the daily national minimum salary for 1993 (rainy season) and 1994 (dry season), representing revenue levels that fulfill daily economic expectations of fishers (Santos-Martínez and Viloria, 1998). This risk analysis was done with LRPs defined by two thresholds (1) prob(BV≤VC) and (2) prob(BV≤Z), representing two scenarios of fishery status. Scenario (1) could be considered as risk-prone, on the basis that BV≤VC implies economic losses or zero quasi rent, whereas scenario (2) is a risk-averse desirable profit margin. Monte Carlo analysis was used to explicitly account for the uncertainty associated with fish abundance, unit price per species and the variable costs per haul in order to quantify the BV risk of falling below the LRPs mentioned above. These variables were randomly generated by Monte Carlo resampling with lognormal (N and P; Chi-square test: p>0.05 in all cases) probability density functions, which allowed us to estimate the mean and standard deviation parameters needed for the simulation process. The VC was fitted to a uniform distribution, assuming that all values between the minimum and maximum, both fixed, occur with equal likelihood (Werckman et al., 2000). Two Monte Carlo runs of 1000 simulation trials were conducted for each year to obtain the probability distributions of BV by which the LRPs were assessed. To provide adequate signs of fishery status, the spatial dimension of the LRPs (Seijo and Caddy, 2000) were determined by indicator kriging (Burrough and McDonnell, 1998), by which we estimated the probability that the BV exceeds desirable thresholds [prob(BV>VC) and prob(BV>Z)] over the study area in each season. For this purpose, original BV data were transformed from a continuous to a binary scale to apply ordinary indicator kriging (Goovaerts, 1997) as follows: (1) where Xi is the BV datum location and Zk are the k thresholds (VC and Z). Indicator kriging is a non-linear form of ordinary kriging, where semivariograms are computed for the binary data in the usual way, and ordinary kriging proceeds with the transformed data (Burrough and McDonnell, 1998). The resulting maps displayed continuous data in the range 0–1, indicating the probability that BV has exceeded desirable thresholds. Effects of biovalue on temporal and spatial allocation of fishing effort Seasonal variations in fishing effort (daily number of fishing trips) were tested using a standard t-test. We do not have effort allocation data with the same spatial resolution as fishing surveys. To circumvent this, potential effects of the spatial structure of BV on spatial allocation of fishing effort were explored by dividing the CGSM area into three discrete sub-areas (fishing zones). These zones were defined according to the number of fishing trips per season allocated on each one, which significantly differed from a hypothesized 1:1:1 ratio for the rainy (χ test = 159.34; p⪡0.01) and the dry (χ test = 381.96; p⪡0.01) seasons. The distribution of the occurrence of the binary scale of BV (1 if BV>VC, and 0 otherwise) in each fishing zone was analyzed for each season by computing a 2×2 contingency table to test the null hypothesis that fishing effort between fishing zones is independent of the prob(BV>VC). Results Fish abundance and population structure by size Abundance of E. plumieri and C. spixii did not differ between seasons (Kruskal–Wallis test H1,228=0.29 and H1,228=0.31; p>0.05, respectively). However, fish abundance in the rainy season showed higher variability (Table 1). Although M. incilis presented higher abundance in the rainy season, seasonal comparison was not tested due to high number of hauls with zero catch in the dry season. The population structure by size of E. plumieri and C. spixii did not differ between seasons (K–S test: p>0.05), while M. incilis presented significantly lower mean lengths in the dry season (22.5±3.1 cm) with respect to the rainy season (26.3±2.6 cm) (K–S test: p<0.05, Table 1). Table 1 Seasonal mean (±s.d.) values of fish abundance and individual fish length in the CGSM. Different sample sizes (n) are due to a breakdown of the “boliche” (abundance data for rainy season) or stations with zero individuals (length data). . Abundance (ind 5000 m−2) . Total length (cm) . Species . X̄ . s.d. . n . X̄ . s.d. . n . 1993/Rainy  E. plumieri 6.2 11.1 113 17.6 3.1 70  M. incilis 4.6 14.7 113 26.3 2.6 59  C. spixii 2.0 4.4 113 22.3 2.6 48 1994/Dry  E. plumieri 3.3 4.0 115 17.5 2.3 85  M. incilis 0.1 0.5 115 22.5 3.1 12  C. spixii 2.3 3.5 115 23.3 2.0 62 . Abundance (ind 5000 m−2) . Total length (cm) . Species . X̄ . s.d. . n . X̄ . s.d. . n . 1993/Rainy  E. plumieri 6.2 11.1 113 17.6 3.1 70  M. incilis 4.6 14.7 113 26.3 2.6 59  C. spixii 2.0 4.4 113 22.3 2.6 48 1994/Dry  E. plumieri 3.3 4.0 115 17.5 2.3 85  M. incilis 0.1 0.5 115 22.5 3.1 12  C. spixii 2.3 3.5 115 23.3 2.0 62 Open in new tab Table 1 Seasonal mean (±s.d.) values of fish abundance and individual fish length in the CGSM. Different sample sizes (n) are due to a breakdown of the “boliche” (abundance data for rainy season) or stations with zero individuals (length data). . Abundance (ind 5000 m−2) . Total length (cm) . Species . X̄ . s.d. . n . X̄ . s.d. . n . 1993/Rainy  E. plumieri 6.2 11.1 113 17.6 3.1 70  M. incilis 4.6 14.7 113 26.3 2.6 59  C. spixii 2.0 4.4 113 22.3 2.6 48 1994/Dry  E. plumieri 3.3 4.0 115 17.5 2.3 85  M. incilis 0.1 0.5 115 22.5 3.1 12  C. spixii 2.3 3.5 115 23.3 2.0 62 . Abundance (ind 5000 m−2) . Total length (cm) . Species . X̄ . s.d. . n . X̄ . s.d. . n . 1993/Rainy  E. plumieri 6.2 11.1 113 17.6 3.1 70  M. incilis 4.6 14.7 113 26.3 2.6 59  C. spixii 2.0 4.4 113 22.3 2.6 48 1994/Dry  E. plumieri 3.3 4.0 115 17.5 2.3 85  M. incilis 0.1 0.5 115 22.5 3.1 12  C. spixii 2.3 3.5 115 23.3 2.0 62 Open in new tab Structural analysis and spatial prediction In four of the six cases analyzed, the spherical model described the spatial structure of E. plumieri, M. incilis and C. spixii abundance, which was unstable between seasons, as highlighted by significant differences in semivariograms (ARSS analyses: F-test >100; p⪡0.01) and the different values of variance explained by the spatial models [C/(C0+C)] (Table 2). The semivariance for each species was higher in the rainy season than in the dry season. This trend was also found for the range parameter A0, which describes the distance at which the variables ceases to be autocorrelated, indicating larger areas of spatial dependence in the rainy season. Consistent with this spatial heterogeneity, kriging maps of fish abundance showed a patchy structure, although M. incilis did not present spatial dependence in the dry season (Figure 2). C. spixii showed different distribution patterns between seasons, whereas E. plumieri consistently presented high-abundance patches in the north of CGSM. Mean sizes for each species tended to show patchy distribution with smooth autocorrelation between stations, as indicated by the best fit of exponential models in most cases (Table 2). Semivariograms for E. plumieri and C. spixii differed between seasons (ARSS analysis: F test >80; p<0.01). This was consistent with the variability in the spatially structured component of sizes, which ranged from 53 to 96%. Kriging maps of individual size evidenced the spatial segregation of different population components in each species, which varied seasonally (Figure 2). A simple visual inspection of abundance and size maps showed that high-abundance patches and larger sizes did not overlap in space (Figure 2), mainly in the rainy season. Negative correlations between mean size and abundance for E. plumieri (r=−0.40; p=0.009) and C. spixii (r=−0.26; p=0.047) corroborate our observations. Figure 2 Open in new tabDownload slide Ordinary kriging maps overlaying fish abundance (ind 0.176 km−2) and individual mean size (cm) for rainy (1993) and dry (1994) seasons in the CGSM. Darker surfaces correspond to fish abundance, whereas the numbers labeling isolines indicate mean length estimates for each species. For M. incilis, maps of fish abundance and sizes were not performed in the dry season, due to the lack of spatial autocorrelation. Figure 2 Open in new tabDownload slide Ordinary kriging maps overlaying fish abundance (ind 0.176 km−2) and individual mean size (cm) for rainy (1993) and dry (1994) seasons in the CGSM. Darker surfaces correspond to fish abundance, whereas the numbers labeling isolines indicate mean length estimates for each species. For M. incilis, maps of fish abundance and sizes were not performed in the dry season, due to the lack of spatial autocorrelation. Table 2 Parameters, goodness of fit criteria and cross-validation of the exponential (Exp), spherical (Sph) and Gaussian (Gau) models, fitted to fish abundance, size, biovalue and indicator experimental semivariograms during rainy and dry seasons in the CGSM. (C0) nugget effect, (C0+C) sill, (A0, in m) range, (%) spatially structured component, (r2) coefficient of determination, (RSS) reduced sum of squares, (α) intercept, (β) slope, (r) coefficient of correlation. For all surveys, α and β were not significantly different from 0 and 1, respectively, and r was significant (p<0.05), both for the geostatistical models fitted and the jackknife cross-validation. (BV) multispecies biovalue, (VC) variable costs of the “boliche” and (Z) minimum threshold profit. . . Parameters . Goodness of fit . Cross-validation . Species . Model . C0 . C0+C . A0 . % . r2 . RSS . α . β . r . Abundance 1993/Rainy  E. plumieri Sph 0.490 1.580 12 530 70 0.96 5×10−2 1.90 1.03 0.47  M. incilis Sph 0.432 1.295 13 750 66 0.91 6×10−2 0.91 1.60 0.40  C. spixii Gaua 0.001 1.106 4587 99 0.44 5.1 1.9 0.40 0.20 1994/Dry  E. plumieri Sph 0.371 0.743 7180 50 0.81 1×10−2 0.58 1.03 0.43  M. incilis None – – – – – – – – –  C. spixii Sph 0.270 0.766 6370 65 0.62 5×10−2 0.77 1.65 0.42 Fish length 1993/Rainy  E. plumieri Exp 0.014 0.056 51 100 74 0.82 3×10−4 4.54 0.75 0.32  M. incilis Exp 0.002 0.007 51 100 78 0.88 8×10−7 5.71 0.80 0.39  C. spixii Expb 3.250 6.980 3610 53 0.37 8×10−2 8.60 0.65 0.26 1994/Dry  E. plumieri Sph 0.0005 0.014 3370 96 0.47 9×10−5 5.72 0.66 0.28  M. incilis None – – – – – – – – –  C. spixii Expb 0.4000 3.990 840 90 0.10 2.4 11.30 0.52 0.20 Biovalue 1993/Rainy  E. plumieri Sph 0.083 0.173 21 780 51 0.89 8×10−3 0.04 1.10 0.35  M. incilis Sph 0.072 0.254 15 660 72 0.93 2×10−3 0.04 1.30 0.46  C. spixii Sph 0.0001 0.166 3830 99 0.56 2.4 0.35 0.26 0.10 1994/Dry  E. plumieri Sph 0.003 0.099 4090 96 0.56 4×10−4 0.14 0.62 0.40  M. incilis None – – – – – – – – –  C. spixii Sph 0.053 0.249 4940 79 0.51 7×10−3 0.29 0.60 0.34 Risk 1993/Rainy  prob(BV>VC) Sph 0.134 0.270 11 590 50 0.94 7×10−4 0.05 0.90 0.47  prob(BV>Z) Sph 0.110 0.221 15 240 50 0.92 8×10−4 0.05 0.87 0.41 1994/Dry  prob(BV>VC) Exp 0.019 0.146 1320 87 0.24 2×10−3 0.05 0.83 0.42  prob(BV>Z) Exp 0.017 0.170 680 90 0.22 1×10−4 0.06 0.67 0.33 . . Parameters . Goodness of fit . Cross-validation . Species . Model . C0 . C0+C . A0 . % . r2 . RSS . α . β . r . Abundance 1993/Rainy  E. plumieri Sph 0.490 1.580 12 530 70 0.96 5×10−2 1.90 1.03 0.47  M. incilis Sph 0.432 1.295 13 750 66 0.91 6×10−2 0.91 1.60 0.40  C. spixii Gaua 0.001 1.106 4587 99 0.44 5.1 1.9 0.40 0.20 1994/Dry  E. plumieri Sph 0.371 0.743 7180 50 0.81 1×10−2 0.58 1.03 0.43  M. incilis None – – – – – – – – –  C. spixii Sph 0.270 0.766 6370 65 0.62 5×10−2 0.77 1.65 0.42 Fish length 1993/Rainy  E. plumieri Exp 0.014 0.056 51 100 74 0.82 3×10−4 4.54 0.75 0.32  M. incilis Exp 0.002 0.007 51 100 78 0.88 8×10−7 5.71 0.80 0.39  C. spixii Expb 3.250 6.980 3610 53 0.37 8×10−2 8.60 0.65 0.26 1994/Dry  E. plumieri Sph 0.0005 0.014 3370 96 0.47 9×10−5 5.72 0.66 0.28  M. incilis None – – – – – – – – –  C. spixii Expb 0.4000 3.990 840 90 0.10 2.4 11.30 0.52 0.20 Biovalue 1993/Rainy  E. plumieri Sph 0.083 0.173 21 780 51 0.89 8×10−3 0.04 1.10 0.35  M. incilis Sph 0.072 0.254 15 660 72 0.93 2×10−3 0.04 1.30 0.46  C. spixii Sph 0.0001 0.166 3830 99 0.56 2.4 0.35 0.26 0.10 1994/Dry  E. plumieri Sph 0.003 0.099 4090 96 0.56 4×10−4 0.14 0.62 0.40  M. incilis None – – – – – – – – –  C. spixii Sph 0.053 0.249 4940 79 0.51 7×10−3 0.29 0.60 0.34 Risk 1993/Rainy  prob(BV>VC) Sph 0.134 0.270 11 590 50 0.94 7×10−4 0.05 0.90 0.47  prob(BV>Z) Sph 0.110 0.221 15 240 50 0.92 8×10−4 0.05 0.87 0.41 1994/Dry  prob(BV>VC) Exp 0.019 0.146 1320 87 0.24 2×10−3 0.05 0.83 0.42  prob(BV>Z) Exp 0.017 0.170 680 90 0.22 1×10−4 0.06 0.67 0.33 a Anisotropic model in northeast direction (64°). Only the value of A0 for the major axis is reported. b Data non-transformed. Open in new tab Table 2 Parameters, goodness of fit criteria and cross-validation of the exponential (Exp), spherical (Sph) and Gaussian (Gau) models, fitted to fish abundance, size, biovalue and indicator experimental semivariograms during rainy and dry seasons in the CGSM. (C0) nugget effect, (C0+C) sill, (A0, in m) range, (%) spatially structured component, (r2) coefficient of determination, (RSS) reduced sum of squares, (α) intercept, (β) slope, (r) coefficient of correlation. For all surveys, α and β were not significantly different from 0 and 1, respectively, and r was significant (p<0.05), both for the geostatistical models fitted and the jackknife cross-validation. (BV) multispecies biovalue, (VC) variable costs of the “boliche” and (Z) minimum threshold profit. . . Parameters . Goodness of fit . Cross-validation . Species . Model . C0 . C0+C . A0 . % . r2 . RSS . α . β . r . Abundance 1993/Rainy  E. plumieri Sph 0.490 1.580 12 530 70 0.96 5×10−2 1.90 1.03 0.47  M. incilis Sph 0.432 1.295 13 750 66 0.91 6×10−2 0.91 1.60 0.40  C. spixii Gaua 0.001 1.106 4587 99 0.44 5.1 1.9 0.40 0.20 1994/Dry  E. plumieri Sph 0.371 0.743 7180 50 0.81 1×10−2 0.58 1.03 0.43  M. incilis None – – – – – – – – –  C. spixii Sph 0.270 0.766 6370 65 0.62 5×10−2 0.77 1.65 0.42 Fish length 1993/Rainy  E. plumieri Exp 0.014 0.056 51 100 74 0.82 3×10−4 4.54 0.75 0.32  M. incilis Exp 0.002 0.007 51 100 78 0.88 8×10−7 5.71 0.80 0.39  C. spixii Expb 3.250 6.980 3610 53 0.37 8×10−2 8.60 0.65 0.26 1994/Dry  E. plumieri Sph 0.0005 0.014 3370 96 0.47 9×10−5 5.72 0.66 0.28  M. incilis None – – – – – – – – –  C. spixii Expb 0.4000 3.990 840 90 0.10 2.4 11.30 0.52 0.20 Biovalue 1993/Rainy  E. plumieri Sph 0.083 0.173 21 780 51 0.89 8×10−3 0.04 1.10 0.35  M. incilis Sph 0.072 0.254 15 660 72 0.93 2×10−3 0.04 1.30 0.46  C. spixii Sph 0.0001 0.166 3830 99 0.56 2.4 0.35 0.26 0.10 1994/Dry  E. plumieri Sph 0.003 0.099 4090 96 0.56 4×10−4 0.14 0.62 0.40  M. incilis None – – – – – – – – –  C. spixii Sph 0.053 0.249 4940 79 0.51 7×10−3 0.29 0.60 0.34 Risk 1993/Rainy  prob(BV>VC) Sph 0.134 0.270 11 590 50 0.94 7×10−4 0.05 0.90 0.47  prob(BV>Z) Sph 0.110 0.221 15 240 50 0.92 8×10−4 0.05 0.87 0.41 1994/Dry  prob(BV>VC) Exp 0.019 0.146 1320 87 0.24 2×10−3 0.05 0.83 0.42  prob(BV>Z) Exp 0.017 0.170 680 90 0.22 1×10−4 0.06 0.67 0.33 . . Parameters . Goodness of fit . Cross-validation . Species . Model . C0 . C0+C . A0 . % . r2 . RSS . α . β . r . Abundance 1993/Rainy  E. plumieri Sph 0.490 1.580 12 530 70 0.96 5×10−2 1.90 1.03 0.47  M. incilis Sph 0.432 1.295 13 750 66 0.91 6×10−2 0.91 1.60 0.40  C. spixii Gaua 0.001 1.106 4587 99 0.44 5.1 1.9 0.40 0.20 1994/Dry  E. plumieri Sph 0.371 0.743 7180 50 0.81 1×10−2 0.58 1.03 0.43  M. incilis None – – – – – – – – –  C. spixii Sph 0.270 0.766 6370 65 0.62 5×10−2 0.77 1.65 0.42 Fish length 1993/Rainy  E. plumieri Exp 0.014 0.056 51 100 74 0.82 3×10−4 4.54 0.75 0.32  M. incilis Exp 0.002 0.007 51 100 78 0.88 8×10−7 5.71 0.80 0.39  C. spixii Expb 3.250 6.980 3610 53 0.37 8×10−2 8.60 0.65 0.26 1994/Dry  E. plumieri Sph 0.0005 0.014 3370 96 0.47 9×10−5 5.72 0.66 0.28  M. incilis None – – – – – – – – –  C. spixii Expb 0.4000 3.990 840 90 0.10 2.4 11.30 0.52 0.20 Biovalue 1993/Rainy  E. plumieri Sph 0.083 0.173 21 780 51 0.89 8×10−3 0.04 1.10 0.35  M. incilis Sph 0.072 0.254 15 660 72 0.93 2×10−3 0.04 1.30 0.46  C. spixii Sph 0.0001 0.166 3830 99 0.56 2.4 0.35 0.26 0.10 1994/Dry  E. plumieri Sph 0.003 0.099 4090 96 0.56 4×10−4 0.14 0.62 0.40  M. incilis None – – – – – – – – –  C. spixii Sph 0.053 0.249 4940 79 0.51 7×10−3 0.29 0.60 0.34 Risk 1993/Rainy  prob(BV>VC) Sph 0.134 0.270 11 590 50 0.94 7×10−4 0.05 0.90 0.47  prob(BV>Z) Sph 0.110 0.221 15 240 50 0.92 8×10−4 0.05 0.87 0.41 1994/Dry  prob(BV>VC) Exp 0.019 0.146 1320 87 0.24 2×10−3 0.05 0.83 0.42  prob(BV>Z) Exp 0.017 0.170 680 90 0.22 1×10−4 0.06 0.67 0.33 a Anisotropic model in northeast direction (64°). Only the value of A0 for the major axis is reported. b Data non-transformed. Open in new tab The relationship between price and size was statistically significant for each species, accounting in all cases for ca. 50% of the total variance; moreover, parameters a and b were highly significant in all cases (p<0.001; Figure 3). For a same size, the price of E. plumieri was higher in the dry season (ANCOVA F1,147=6.32; p=0.012), whereas the price of M. incilis was higher in the rainy season (ANCOVA F1,51=4.71; p=0.034). Alternatively, the L–P relationship for C. spixii did not differ between seasons (ANCOVA F1,162=0.009; p=0.924). Figure 3 Open in new tabDownload slide L–P relationship fitted for fish species in rainy (R, —●—) and dry (D, —○—) seasons in the CGSM. Fitted models are also shown. P, price; L, total length. Note the different scale in the X-axis for M. incilis. Figure 3 Open in new tabDownload slide L–P relationship fitted for fish species in rainy (R, —●—) and dry (D, —○—) seasons in the CGSM. Fitted models are also shown. P, price; L, total length. Note the different scale in the X-axis for M. incilis. The spatial structure of BV for E. plumieri and C. spixii differed between seasons (ARSS analysis: F test >112; p<0.01), with areas of spatial dependence well defined by spherical models (Table 2). The spatially structured component of BV ranged between 51 and 99%, confirming different spatial features in the economic potential denoted by kriging maps (Figure 4). The distribution patterns of BV were very similar to those shown for abundance data (Figures 2 and 4). E. plumieri presented high-BV patches mainly in the north of CGSM in both seasons, with some important clusters in the dry season located toward southern, eastern and western ends of the lagoon. Consistently, C. spixii showed an aggregated distribution in each season, whereas the BV of M. incilis was clustered on the eastern coast during the rainy season. In all cases, cross-validation analysis showed that kriging predictions for fish abundance, size and biovalue were suitable, because the null hypotheses α=0 and β=1 were never rejected (p>0.05; Table 2). Global estimates of biovalue were higher in the dry season than in the rainy season, both for E. plumieri (13%) and C. spixii (37%) (Table 3). These estimates evidenced the combined effect of abundance by size and individual size price variations for each species on the economic potential of fish abundance. Figure 4 Open in new tabDownload slide Ordinary kriging maps of biovalue (US$ 0.176 km−2) for each species in rainy (1993) and dry (1994) seasons in the CGSM. The kriged map was not performed for M. incilis in the dry season, due to the lack of spatial autocorrelation. Figure 4 Open in new tabDownload slide Ordinary kriging maps of biovalue (US$ 0.176 km−2) for each species in rainy (1993) and dry (1994) seasons in the CGSM. The kriged map was not performed for M. incilis in the dry season, due to the lack of spatial autocorrelation. Table 3 Block kriging estimates of fish biovalue during rainy and dry seasons in the CGSM. B̄V̄ (US$ 0.176 km−2) is the mean biovalue and s.d.BV is its standard deviation; BVt and s.d.BVt are the mean total biovalue for the whole area (US$) and its standard deviation, respectively. . Biovalue . Species . B̄V̄ . s.d.BV . BVt . s.d.BVt . 1993/Rainy  E. plumieri 0.37 0.08 28 712 6208  M. incilis 0.37 0.09 28 786 7002  C. spixii 0.48 0.11 31 680 7260 1994/Dry  E. plumieri 0.42 0.08 32 424 6176  M. incilis – – – –  C. spixii 0.66 0.21 43 560 13 860 . Biovalue . Species . B̄V̄ . s.d.BV . BVt . s.d.BVt . 1993/Rainy  E. plumieri 0.37 0.08 28 712 6208  M. incilis 0.37 0.09 28 786 7002  C. spixii 0.48 0.11 31 680 7260 1994/Dry  E. plumieri 0.42 0.08 32 424 6176  M. incilis – – – –  C. spixii 0.66 0.21 43 560 13 860 Open in new tab Table 3 Block kriging estimates of fish biovalue during rainy and dry seasons in the CGSM. B̄V̄ (US$ 0.176 km−2) is the mean biovalue and s.d.BV is its standard deviation; BVt and s.d.BVt are the mean total biovalue for the whole area (US$) and its standard deviation, respectively. . Biovalue . Species . B̄V̄ . s.d.BV . BVt . s.d.BVt . 1993/Rainy  E. plumieri 0.37 0.08 28 712 6208  M. incilis 0.37 0.09 28 786 7002  C. spixii 0.48 0.11 31 680 7260 1994/Dry  E. plumieri 0.42 0.08 32 424 6176  M. incilis – – – –  C. spixii 0.66 0.21 43 560 13 860 . Biovalue . Species . B̄V̄ . s.d.BV . BVt . s.d.BVt . 1993/Rainy  E. plumieri 0.37 0.08 28 712 6208  M. incilis 0.37 0.09 28 786 7002  C. spixii 0.48 0.11 31 680 7260 1994/Dry  E. plumieri 0.42 0.08 32 424 6176  M. incilis – – – –  C. spixii 0.66 0.21 43 560 13 860 Open in new tab Risk analysis and indicator kriging Both LRPs [prob(BV≤VC) and prob(BV≤Z)] varied between seasons (Figure 5). For the first scenario, the probability that the BVs were equal or less than the variable costs (US$0.68 haul−1 in the rainy season and US$0.81 haul−1 in the dry season) was 0.47 and 0.58, respectively, suggesting a moderate risk that fishers obtain zero quasi rent or economic losses. The second scenario showed a high risk (0.79 and 0.92 for rainy and dry seasons, respectively) that fishers obtain a minimum profit threshold equal or less than that expected (Z=US$1.48 haul−1 in the rainy season and US$1.6 haul−1 in the dry season). Consequently, the probability of obtaining some fishing profit ranged between 0.53 in the rainy and 0.42 in the dry seasons. Such probabilities were substantially reduced under the second scenario (0.21 in the rainy and 0.08 in the dry season). Figure 5 Open in new tabDownload slide Risk analysis by season. Probability of falling below the LRPs (shaded bars) given by prob(BV≤VC) and prob(BV≤Z). The corresponding probability values are also shown. BV is the multispecies biovalue by haul. VC denotes variable costs by haul of the “boliche”, and Z is a minimum threshold profit (see text for details). Probabilities of falling below these undesirable thresholds are based on 1000 Monte Carlo simulation trials for rainy and dry seasons. Figure 5 Open in new tabDownload slide Risk analysis by season. Probability of falling below the LRPs (shaded bars) given by prob(BV≤VC) and prob(BV≤Z). The corresponding probability values are also shown. BV is the multispecies biovalue by haul. VC denotes variable costs by haul of the “boliche”, and Z is a minimum threshold profit (see text for details). Probabilities of falling below these undesirable thresholds are based on 1000 Monte Carlo simulation trials for rainy and dry seasons. Spatial risk analysis showed that probabilities of exceeding desirables thresholds of biovalue had spatial structure, which differed between seasons consistently with changes in the spatial variance explained by spherical (50%, rainy season) and exponential models (87–90%, dry season) (Table 2). Cross-validation analysis corroborated the appropriateness of the semivariogram models (0.33<r<0.47; p<0.05). Risk maps (Figure 6) showed that patches with probability of getting some profit [prob(BV>VC)] varied in location and size, being the smaller patches when the probability of achieving the expected rent by fishers was considered [(prob(BV>Z)]. Taking into account an arbitrary value of probability higher than 0.7, the potential areas for getting profits did not exceed 30% of the CGSM total area. Figure 6 Open in new tabDownload slide Probability maps of exceeding desirable thresholds given by prob(BV>VC) and prob(BV>Z) for rainy and dry seasons in the CGSM. Maps were produced by ordinary indicator kriging. BV, VC and Z are as defined in Figure 5. Figure 6 Open in new tabDownload slide Probability maps of exceeding desirable thresholds given by prob(BV>VC) and prob(BV>Z) for rainy and dry seasons in the CGSM. Maps were produced by ordinary indicator kriging. BV, VC and Z are as defined in Figure 5. Effects of biovalue on temporal and spatial allocation of fishing effort The number of fishing trips day−1 was significantly higher in the dry season than in the rainy season (t-test=−6.84; p⪡0.01) (Figure 7b). Moreover, the number of trips was significantly higher on fishing zone A, both in the rainy (55% of 571 trips) and dry (48% of 1882 trips) seasons. The lowest number of trips was allocated on zone B (12 and 13% of the total trips in rainy and dry seasons, respectively), whereas zone C had intermediate values (33 and 39% of the total trips in rainy and dry seasons, respectively) (Figure 7a, c, d). Consistent with this spatio-temporal pattern, the number of fishing trips between zones was dependent on the probability of obtaining some profit prob(BV>VC) in the rainy (χ2=10.07; p⪡0.01) and in the dry (χ test = 6.06; p<0.05) seasons. In this sense, zones A (in the rainy season) and C (in the dry season) were the most successful, considering that the higher frequency of indicators revealed the probability of finding fish abundance at profitable levels (Figure 7c, d). Figure 7 Open in new tabDownload slide Linkage between biovalue and spatial allocation of fishing effort in the CGSM. (a) Fishing zones (A, B and C). (b) Seasonal variations in fishing effort (fishing trips day−1: mean±SE). Frequency of indicators (—●—, BV>VC; —○—, BV≤VC) and total fishing effort (fishing trips) discriminated by fishing zones for rainy (c) and dry (d) seasons. Figure 7 Open in new tabDownload slide Linkage between biovalue and spatial allocation of fishing effort in the CGSM. (a) Fishing zones (A, B and C). (b) Seasonal variations in fishing effort (fishing trips day−1: mean±SE). Frequency of indicators (—●—, BV>VC; —○—, BV≤VC) and total fishing effort (fishing trips) discriminated by fishing zones for rainy (c) and dry (d) seasons. Discussion The spatial dimension has played an important role in building paradigms in ecological research, and currently constitutes one of the challenges to overcome in fisheries science for the present century (Caddy, 1999). In this study, model-based (e.g. geostatistics) approaches allowed us to describe successfully the spatial structure of fish abundance, individual size and biovalue in a tropical estuarine small-scale fishery. Such an approach was also useful to test hypotheses related to the spatial dynamics of those performance variables and to assess the status of fishery using bioeconomic reference points. Abundance of E. plumieri, M. incilis and C. spixii showed a strong spatial structure in the CGSM, and this was consistent with spatial variations in the population composition by size and the BV (Table 2). Fish distribution of these species has been related to gradients in salinity and to habitat features like substrate type (Rueda, 2001). The reproductive migration of M. incilis toward adjacent marine waters (Sánchez et al., 1998) precluded the determination of spatial structure for the abundance, size and biovalue in the dry season. Size distribution maps showed different spatial structures of population components (e.g. juveniles, spawning stock) within and between seasons, as a result of the co-occurrence of multiple annual cohorts (Sánchez et al., 1998; Tíjaro et al., 1998; Rueda and Santos-Martínez, 1999). High-abundance patches corresponded to juveniles, mostly for E. plumieri (L < 17 cm) and C. spixii (L < 23 cm) in the rainy season, whereas most aggregations of adults were present in the dry season (Figure 2). This picture was consistent with recruitment and sexual maturity peaks reported for these species during the rainy and dry seasons, respectively, in the CGSM (Tíjaro et al., 1998; Rueda and Santos-Martínez, 1999). Negative correlations between mean size and abundance for E. plumieri and C. spixii suggest spatial density dependence in the rainy season, thus giving strong support to our findings. Alternatively, M. incilis showed adult individuals (L > 24 cm) in the rainy season, mainly clustered in the eastern part of the CGSM before migrating to the sea to spawn. The L–P relationship fitted for each species was very useful to estimate spatial and temporal variations in the biovalue. In spite of this, abundance and structure by size did not differ between seasons for E. plumieri and C. spixii, and BV of these species was consistently higher in the dry season as a response to seasonal price-at-size variations. This demonstrates one of the main characteristics of artisanal fisheries, where market forces regulating prices affect the potential value of the catch (Defeo and Castilla, 1998; Castilla and Defeo, 2001). Higher BV values (Figure 4) coincided with scarce high-abundance patches for the three species (Figure 2). Such spatio-temporal heterogeneity of the economical potential determined that fishers had moderate probabilities of realizing economic losses from their fishing activity; whereas the risk of falling below an undesirable threshold of rent expected was always high. In agreement with this scenario, maps of risk emphasized that only a small portion of the stock could be harvested with high probabilities of obtaining profits and that, in general, when fish abundance was low variable costs were higher than the biovalue. The above situation determines high concentrations of fishers over the most productive grounds, generating crowding externalities (Seijo et al., 1998). Conflicts between fishers of different ports employing several gears support this assertion due to overlapping of fishing grounds (Santos-Martínez and Viloria, 1998). These findings were confirmed by the analysis of commercial fishing data, which showed: (1) higher amounts of fishing effort in the dry season in accordance with higher BV levels and (2) dependence between fishing effort allocated among zones and the probability of finding profitable levels of abundance [(prob(BV>VC)]. In conclusion, the analysis of spatial structure of fishery performance variables allowed us to map variations in abundance of different population components and in the economical potential of targeted species. These model-based estimates were useful to quantify spatially explicit probabilities of exceeding bioeconomic LRPs, which may be used to explain potential changes in the spatial and temporal allocation of fishing effort. In this context, a precautionary approach could include the reduction of fishing effort in areas and seasons with high concentration of juveniles (e.g. E. plumieri on the north of the lagoon during the rainy season), even though high abundance could generate high economic revenues. Moreover, effort on M. incilis should be restricted at the northeast of the lagoon during the rainy season, in order to protect the spawning stock. Such strategies might be combined with selectivity controls of the “boliche” (Rueda and Defeo, in press) and an appropriate institutional framework directed to strengthen the actually poor implementation and enforcement of management measures in this artisanal fishery. This paper is part of the PhD thesis of M.R. at CINVESTAV-IPN U. Mérida. The INVEMAR, COLCIENCIAS and GTZ provided logistical and financial support for fieldwork in Colombia. The first author would like to thank the “CGSM's Fisheries Group” of the INVEMAR for field and laboratory assistance, especially to J.C. Narvaez. Native fishers of the CGSM contributed with their fishing skills and their empirical knowledge of the area to the success of the fishing surveys. References Anderson L.G. . Optimal intra- and inter-seasonal harvesting strategies when prices varies with individual size , Marine Resource Economics , 1989 , vol. 6 (pg. 145 - 162 ) Google Scholar Crossref Search ADS WorldCat Banco de la República Estadísticas sobre información económica del Gobierno de Colombia , 2002 Available at: http://www.banrep.gov.co Blaber J.M. . , Fish and Fisheries of Tropical Estuaries , 1997 London Chapman and Hall (367 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Botero L. , Salzwedel H. . Rehabilitation of the Cienaga Grande de Santa Marta, a mangrove–estuarine system in the Caribbean coast of Colombia , Ocean and Coastal Management , 1999 , vol. 42 (pg. 243 - 256 ) Google Scholar Crossref Search ADS WorldCat Burrough P.A. , McDonnel R.A. . , Principles of Geographical Information Systems , 1998 New York Oxford University Press (333 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cabrera J.L. , Defeo O. . Daily bioeconomic analysis in a multispecific artisanal fishery in Yucatan, Mexico , Aquatic Living Resources , 2001 , vol. 14 (pg. 19 - 28 ) Google Scholar Crossref Search ADS WorldCat Caddy J.F. . Fisheries management in the twenty-first century: will new paradigms apply? , Reviews in Fish Biology and Fisheries , 1999 , vol. 9 (pg. 1 - 43 ) Google Scholar Crossref Search ADS WorldCat Caddy J.F. , Carocci F. . The spatial allocation of fishing intensity by port-based inshore fleets: a GIS application , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 388 - 403 ) Google Scholar Crossref Search ADS WorldCat Caddy J.F. , Seijo J.C. . Application of a spatial model to explore rotating harvest strategies for sedentary species , Canadian Special Publication of Fisheries and Aquatic Sciences , 1998 , vol. 125 (pg. 359 - 365 ) OpenURL Placeholder Text WorldCat Castilla J.C. , Defeo O. . Latin-American benthic shellfisheries: emphasis on co-management and experimental practices , Reviews in Fish Biology and Fisheries , 2001 , vol. 11 (pg. 1 - 30 ) Google Scholar Crossref Search ADS WorldCat Chen Y. , Jackson D.A. , Harvey H.H. . A comparison of von Bertalanffy and polynomial functions in modelling fish growth data , Canadian Journal of Fisheries and Aquatic Sciences , 1992 , vol. 49 (pg. 1228 - 1235 ) Google Scholar Crossref Search ADS WorldCat Defeo O. , Castilla J.C. . Harvesting and economic patterns in the artisanal Octopus mimus (Cephalopoda) fishery in a northern Chile cove , Fisheries Research , 1998 , vol. 38 (pg. 121 - 130 ) Google Scholar Crossref Search ADS WorldCat Defeo O. , Rueda M. . Spatial structure, sampling design and abundance estimates in sandy beach macroinfauna: some warnings and new perspectives , Marine Biology , 2002 , vol. 140 (pg. 1215 - 1225 ) Google Scholar Crossref Search ADS WorldCat Goovaerts P. . , Geostatistics for Natural Resources Evaluation , 1997 New York Oxford University Press (483 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hilborn R. , Walters C.J. . A general model for simulation of stock and fleet dynamics in spatially heterogeneous fisheries , Canadian Journal of Fisheries and Aquatic Sciences , 1987 , vol. 44 (pg. 1366 - 1369 ) Google Scholar Crossref Search ADS WorldCat Isaaks E.H. , Srivastava R.M. . , An Introduction to Applied Geostatistics , 1989 New York Oxford University Press (561 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Journel A.G. , Huijbregts C.J. . , Mining Geostatistics , 1978 London Academic Press (600 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Legendre P. . Spatial autocorrelation: trouble or new paradigm? , Ecology , 1993 , vol. 74 (pg. 1659 - 1673 ) Google Scholar Crossref Search ADS WorldCat Maravelias C.D. , Reid D.G. , Simmonds E.J. , Haralabous J. . Spatial analysis and mapping of acoustic survey data in the presence of high local variability: geostatistical application to North Sea herring (Clupea harengus) , Canadian Journal of Fisheries and Aquatic Sciences , 1996 , vol. 53 (pg. 1497 - 1505 ) Google Scholar Crossref Search ADS WorldCat Matheron G. . The theory of regionalized variables and its applications , 1971 Les Cahiers du CMM, facs. 5, Ecole Nat. Sup. Mines, Paris. 211 pp Maynou F.X. , Sardà F. , Conan G.Y. . Assessment of the spatial structure and biomass evaluation of Nephrops norvegicus (L.) populations in the northwestern Mediterranean by geostatistics , ICES Journal of Marine Science , 1998 , vol. 55 (pg. 102 - 120 ) Google Scholar Crossref Search ADS WorldCat Orensanz J. , Jamieson J. . The assessment and management of spatially structured stocks , Canadian Journal of Fisheries and Aquatic Sciences , 1998 , vol. 125 (pg. 441 - 459 ) OpenURL Placeholder Text WorldCat Pauly D. , Watson R. , Christensen V. . Ecological geography as framework for a transition toward responsible fishing , 2001 Proceedings of Reykjavik Conference on Responsible Fisheries in the Marine Ecosystem Reykjavik Iceland (pg. 1 - 18 ) Pelletier D. , Parma A. . Spatial distribution of Pacific halibut (Hippoglossus stenolepis): an application of geostatistics to longline survey data , Canadian Journal of Fisheries and Aquatic Sciences , 1994 , vol. 51 (pg. 1506 - 1518 ) Google Scholar Crossref Search ADS WorldCat Petitgas P. . Geostatistics for fish stock assessments: a review and an acoustic application , ICES Journal of Marine Science , 1993 , vol. 50 (pg. 285 - 298 ) Google Scholar Crossref Search ADS WorldCat Power M. . The predictive validation of ecological and environmental models , Ecological Modelling , 1993 , vol. 68 (pg. 33 - 50 ) Google Scholar Crossref Search ADS WorldCat Robertson G.P. . , GS+: Geostatistics for the Environmental Sciences , 2000 Plain Gamma Design Software (200 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rueda M. . Evaluación de la eficiencia y selectividad de redes de enmalle en la Ciénaga Grande de Santa Marta, Caribe colombiano , 1995 BSc thesis, Universidad del Magdalena, Santa Marta, Colombia. 107 pp Rueda M. . Spatial distribution of fish species in a tropical estuarine lagoon: a geostatistical appraisal , Marine Ecology Progress Series , 2001 , vol. 222 (pg. 217 - 226 ) Google Scholar Crossref Search ADS WorldCat Rueda M. , Defeo O. . Survey abundance indices in a tropical estuarine lagoon and their management implications: a spatially-explicit approach , ICES Journal of Marine Science , 2001 , vol. 58 (pg. 1219 - 1231 ) Google Scholar Crossref Search ADS WorldCat Rueda M. , Defeo O. . Linking fishery management and conservation in a tropical estuarine lagoon: biological and physical effects of an artisanal fishing gear , Estuarine, Coastal and Shelf Science in press OpenURL Placeholder Text WorldCat Rueda M. , Santos-Martínez A. . Population dynamics of the striped mojarra Eugerres plumieri from the Ciénaga Grande de Santa Marta, Colombia , Fisheries Research , 1999 , vol. 42 (pg. 155 - 166 ) Google Scholar Crossref Search ADS WorldCat Rueda M. , Urban H.J. . Population dynamics and fishery of the fresh-water clam Polymesoda solida (Corbiculidae) in Cienaga Poza Verde, Salamanca Island, Colombian Caribbean , Fisheries Research , 1998 , vol. 39 (pg. 75 - 86 ) Google Scholar Crossref Search ADS WorldCat Sánchez C. , Rueda M. . Space–time variation of the diversity and abundance of dominant fish species in the Magdalena River Delta, Colombia , Revista de Biología Tropical , 1999 , vol. 47 (pg. 1067 - 1079 ) OpenURL Placeholder Text WorldCat Sánchez C. , Rueda M. , Santos-Martínez A. . Dinámica poblacional y pesquería de la lisa Mugil incilis en la Ciénaga Grande de Santa Marta, Caribe colombiano , Revista de la Academia Colombiana de Ciencias Exactas Físicas y Naturales , 1998 , vol. 85 (pg. 507 - 517 ) OpenURL Placeholder Text WorldCat Santos-Martínez A. , Viloria E. . Evaluación de los recursos pesqueros de la Ciénaga Grande de Santa Marta y el complejo Pajarales, Caribe colombiano: estadística pesquera , 1998 Invemar Report, Colciencias, Bogotá. 62 pp Seijo J.C. , Caddy J.F. . Uncertainty in bio-economic reference points and indicators of marine fisheries , Marine and Freshwater Research , 2000 , vol. 51 (pg. 477 - 483 ) Google Scholar Crossref Search ADS WorldCat Seijo J.C. , Defeo O. , Salas S. . , Fisheries Bioeconomics. Theory, Modelling and Management , 1998 Rome FAO Fisheries Technical Paper 368 (108 pp.) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Taconet M. , Bensch A. . Towards the use of geographic information systems as a decision support tool for the management of Mediterranean fisheries , 2000 FAO COPEMED Project GCP/REM/057/SPA. 278 pp Tíjaro R. , Rueda M. , Santos-Martínez A. . Dinámica poblacional del chivo mapalé Cathorops spixii en la Ciénaga Grande de Santa Marta y Complejo de Pajarales, Caribe colombiano , Boletín de Investigaciones Marinas and Costeras , 1998 , vol. 27 (pg. 87 - 102 ) OpenURL Placeholder Text WorldCat Werckman C. , Hardy T. , Wainwright E. , Heywood B. . Crystal Ball User Manual , 2000 Denver Decisioneering Inc., CG Press (412 pp.) © 2003 International Council for the Exploration of the Sea

Gaspar,, M.B;Leitão,, F;Santos,, M.N;Chícharo,, L;Dias,, M.D;Chícharo,, A;Monteiro,, C.C

doi: 10.1016/S1054-3139(03)00023-7pmid: N/A

Abstract The white clam Spisula solida is harvested along the entire coast of Portugal using mechanical dredges. In this study, the total direct mortality of the macrobenthic community caused by three types of clam dredges (north dredge—ND, traditional dredge—TD, and the metallic grid dredge—GD) used in the S. solida fishery was determined and compared. The relationship between mortality and catching efficiency for each type of dredge was also assessed. Our results showed significant differences for total direct mortality between the ND and both the GD and TD dredges. This difference was largely attributed to the mortality of animals that died in the dredge track as a direct result of the physical damage inflicted by the dredge passing. It was also found that the damage to uncaught individuals is directly related to gear efficiency. The lower catching efficiency of the ND (64%) led to a higher proportion of damaged individuals being left in the dredge path, when compared with the more efficient GD (98%) and TD (90%) dredges. Short and long-term implications of the impact of dredging on the composition of benthic communities are discussed. From fisheries management and ecological points of view, there are obvious advantages to introduce into the bivalve dredge fisheries more efficient and selective dredges in order to reduce the number of damaged individuals and by-catch, and consequently decreasing the impact on the macrobenthic communities. Introduction The exploitation of subtidal bivalve beds along the Portuguese coast is relatively recent, having been initiated only in the late 1960s. Although several species of commercial importance are harvested, only the white clam Spisula solida is caught by the whole dredge fleet, as it is the only species that occurs along the entire Portuguese continental coast. For management purposes the Portuguese coast was divided into three main fishing areas, the northwest, the southwest and the southern areas. These were defined based on the distribution of clam beds and fishing ports, the coastal topography and environmental conditions. Although the majority of technical measures used to manage the fishery are similar in all three fishing areas, there are differences relating to the number of fishing licenses, boat engine power and daily quotas per boat and species. In this fishery, only mechanical dredges are allowed, made up of a rigid iron structure with a toothed lower bar, and a collecting system. The main differences between the dredges used in the S. solida fishery relate to the shape and length of the dredge mouth and the collecting system. Figure 1 shows photographs of the three types of dredges used in the fishery. Until 1999, the northwest dredge fleet only operated with the north dredge (ND) and the southwest and south dredge fleets with the traditional dredge (TD). Recently, a new dredge design (grid dredge—GD) was introduced into the fishery and since then the majority of the fleet operating along the southwest and south coasts of Portugal have started using this new gear. This dredge employs a metallic grid instead of using a net bag to retain the catch. Due to the extra weight of the GD only small boats still use the TD. Figure 1 Open in new tabDownload slide Photographs of the three dredge types used in the S. solida Portuguese fishery. (A) North dredge (ND); (B) traditional dredge (TD); (C) grid dredge (GD). Figure 1 Open in new tabDownload slide Photographs of the three dredge types used in the S. solida Portuguese fishery. (A) North dredge (ND); (B) traditional dredge (TD); (C) grid dredge (GD). These dredges were designed to dig clams out of the sediment, impacting on the benthic habitat, both in terms of its physical structure and its biological communities. Direct impacts include scraping and ploughing of the substrate, sediment re-suspension, destruction of the benthos and loss of biodiversity (e.g. Van Dolah et al., 1987; Eleftheriou and Robertson, 1992; Jones, 1992; Currie and Parry, 1996; Kaiser et al., 1996; Collie et al., 1997; Bergman and van Santbrink, 2000). Although the impact on the sediment caused by the three dredge types used in the S. solida fishery is expected to be similar (capture methods being identical), the impact on the macrofauna may be different. In order to introduce modifications to the dredges to reduce the mortality, or even to ban dredge types that cause greater impacts, it is important to estimate the direct mortality induced by each dredge type on the benthic macrofauna. During this study, the direct effects of three different dredge types on macrobenthic mortality were compared. The relationship between this mortality and the catching efficiency for each type of dredge was also assessed. Materials and methods Experimental design The study was undertaken in June 2001 in the Sines region on the southwestern coast of Portugal. The site is off Lagoa de Santo André (38°02′99″N, 08°49′78″W), and is one of the most important fishing grounds for S. solida in Portugal. The samples were collected from sandy bottoms between 8 and 10 m depth. The study was carried out using the research vessel “Donax”, which is of similar size and engine power to local commercial fishing boats. The dredges were identical to those used by the commercial dredge fleet. Throughout this study the dredge usually employed in the northwest coast of Portugal was referred to as a north dredge (ND), the dredge used by small boats was referred to as a traditional dredge (TD), while the dredge fitted with a metal grid collecting system was referred to as a grid dredge (GD). Table 1 summarises the gear specifications of these dredges. Table 1 Gear specifications of the dredges used in this study . North dredge . Grid dredge . Traditional dredge . Anterior part  Length of the mouth (cm) 150 64 64  Space between rods (cm) – 0.8 0.8  Number of teeth 49 10 10  Space between teeth (cm) 2 2.2 2.2  Tooth length (cm) 12 15 15  Tooth angle (degrees) 20° 20° 20° Net bag  Length (cm) 450 – 250  Mesh size (mm) 25 – 25 Grid  Space between rods (cm) – 1.2 – . North dredge . Grid dredge . Traditional dredge . Anterior part  Length of the mouth (cm) 150 64 64  Space between rods (cm) – 0.8 0.8  Number of teeth 49 10 10  Space between teeth (cm) 2 2.2 2.2  Tooth length (cm) 12 15 15  Tooth angle (degrees) 20° 20° 20° Net bag  Length (cm) 450 – 250  Mesh size (mm) 25 – 25 Grid  Space between rods (cm) – 1.2 – Open in new tab Table 1 Gear specifications of the dredges used in this study . North dredge . Grid dredge . Traditional dredge . Anterior part  Length of the mouth (cm) 150 64 64  Space between rods (cm) – 0.8 0.8  Number of teeth 49 10 10  Space between teeth (cm) 2 2.2 2.2  Tooth length (cm) 12 15 15  Tooth angle (degrees) 20° 20° 20° Net bag  Length (cm) 450 – 250  Mesh size (mm) 25 – 25 Grid  Space between rods (cm) – 1.2 – . North dredge . Grid dredge . Traditional dredge . Anterior part  Length of the mouth (cm) 150 64 64  Space between rods (cm) – 0.8 0.8  Number of teeth 49 10 10  Space between teeth (cm) 2 2.2 2.2  Tooth length (cm) 12 15 15  Tooth angle (degrees) 20° 20° 20° Net bag  Length (cm) 450 – 250  Mesh size (mm) 25 – 25 Grid  Space between rods (cm) – 1.2 – Open in new tab A total of 12 tows were undertaken, four for each dredge type. Dredges were towed for 5 min at a mean speed of 1.5 knots. Both the tow duration and fishing speed used in this experiment were similar to those used by commercial fishing vessels operating with these types of dredge. The duration of dredge hauls was measured from the time the winch stopped paying out the towing cable to the time when the winch was restarted. In order to determine the proportion of damaged individuals that entered the dredge but passed through the net bag during the fishing operation and during dredge retrieval, the cover method was adopted. This involves covering the net bag or grid with a second net bag with a smaller mesh size (20 mm mesh). The cover bag was stretched loosely over the entire back of the dredges, enough slack being left in the cover to reduce any masking effect on selection and to allow the natural flow of water through the net as suggested by Gaspar et al. (1999). The catches of each dredge, coming from the bag or the grid and from the cover, were always processed separately. In order to estimate macrofaunal mortality induced by each dredge, the extent of damage sustained was recorded for each organism caught using a 4 score scale (Table 2). This was visually assessed by the same person on all occasions immediately after sampling operations on board. Table 2 Criteria used in the attribution of a damage score for each taxon Score . 1 . 2 . 3 . 4 . Bivalvia In good condition Edge of shell chipped Hinge broken Crushed/dead Gastropoda In good condition Edge of shell chipped Shell cracked or punctured Crushed/dead Echinoidea In good condition <50% spine loss >50% spine loss/minor cracks Crushed/dead Ophiuroidea In good condition Arms missing Worn and arms missing/minor disc damage Major disc damaged/dead Cephalopoda In good condition Dead Crustacea  Anomura In good condition Out of shell and intact Out of shell and damaged Crushed/dead  Brachyura In good condition Legs missing/small carapace cracks Major carapace cracks Crushed/dead Osteichthyes In good condition Small amount of scales missing/small cuts or wound Large amount of scales missing/severe wounds Dead Score . 1 . 2 . 3 . 4 . Bivalvia In good condition Edge of shell chipped Hinge broken Crushed/dead Gastropoda In good condition Edge of shell chipped Shell cracked or punctured Crushed/dead Echinoidea In good condition <50% spine loss >50% spine loss/minor cracks Crushed/dead Ophiuroidea In good condition Arms missing Worn and arms missing/minor disc damage Major disc damaged/dead Cephalopoda In good condition Dead Crustacea  Anomura In good condition Out of shell and intact Out of shell and damaged Crushed/dead  Brachyura In good condition Legs missing/small carapace cracks Major carapace cracks Crushed/dead Osteichthyes In good condition Small amount of scales missing/small cuts or wound Large amount of scales missing/severe wounds Dead Open in new tab Table 2 Criteria used in the attribution of a damage score for each taxon Score . 1 . 2 . 3 . 4 . Bivalvia In good condition Edge of shell chipped Hinge broken Crushed/dead Gastropoda In good condition Edge of shell chipped Shell cracked or punctured Crushed/dead Echinoidea In good condition <50% spine loss >50% spine loss/minor cracks Crushed/dead Ophiuroidea In good condition Arms missing Worn and arms missing/minor disc damage Major disc damaged/dead Cephalopoda In good condition Dead Crustacea  Anomura In good condition Out of shell and intact Out of shell and damaged Crushed/dead  Brachyura In good condition Legs missing/small carapace cracks Major carapace cracks Crushed/dead Osteichthyes In good condition Small amount of scales missing/small cuts or wound Large amount of scales missing/severe wounds Dead Score . 1 . 2 . 3 . 4 . Bivalvia In good condition Edge of shell chipped Hinge broken Crushed/dead Gastropoda In good condition Edge of shell chipped Shell cracked or punctured Crushed/dead Echinoidea In good condition <50% spine loss >50% spine loss/minor cracks Crushed/dead Ophiuroidea In good condition Arms missing Worn and arms missing/minor disc damage Major disc damaged/dead Cephalopoda In good condition Dead Crustacea  Anomura In good condition Out of shell and intact Out of shell and damaged Crushed/dead  Brachyura In good condition Legs missing/small carapace cracks Major carapace cracks Crushed/dead Osteichthyes In good condition Small amount of scales missing/small cuts or wound Large amount of scales missing/severe wounds Dead Open in new tab Diving surveys were also conducted in order to estimate the percentage of damaged macrofaunal individuals left on dredge tracks, to determine the length of the dredge path and to estimate the dredges efficiency of capture. The efficiency of capture is defined as the proportion of the number of target clam species in the path of the dredge that enters through the dredge mouth (Caddy, 1971). For each haul, divers randomly collected 54 sediment samples using quadrats (area=0.0625 m2×0.15 m depth) within the dredge path: 27 quadrats in the furrow and 27 in the ridge. Samples were sieved in situ through a 5 mm mesh bag, and when back on board the boat, preserved in 70% ethanol. In the laboratory, the organisms were identified, counted, weighed and a damage score was attributed to each specimen caught using the damage scale (Table 2). The species identification was made according to Bucquoy et al. (1882–1898), Tebble (1966), FAO (1987) and Poppe and Goto (1993). The nomenclature adopted was that of FAO (1987). Data analysis The primer© software package (Clark and Warwick, 1994) was used to compare methods of capture (grid vs mesh), by investigating the number of individuals per species that escaped through the meshes of the bag (ND and TD) or through the bars of the grid (GD). Abundance data from the cover bag was square-root-transformed prior to cluster analysis using the Bray–Curtis method to produce a similarity matrix. The relationships between samples were examined by non-metric multidimensional ordination plots (MDS), while the analysis of similarities (ANOSIM) routine (Clark and Warwick, 1994) was used to detect any strong difference on dredge selectivity. Analyses of variance (ANOVA) or Kruskal–Wallis ANOVA were used to investigate differences between the fishing yields obtained from each dredge and to test the effect of dredge design on the proportion of damaged and dead individuals. The damage inflicted by dredges on macrofauna was analysed separately for the individuals that entered the dredge and for those organisms left on the dredge path. Multiple comparisons were performed using the Student–Newman–Keuls test. Prior to the application of ANOVA or Kruskal–Wallis ANOVA, data were standardised and transformed to arcsine square root values when expressed as a percentage. Statistical analyses were undertaken using SigmaStat© statistical software. Results A total of 29,119 individuals belonging to eight taxa were caught during the fishing experiments (Table 3). The catches from the GD, TD and ND comprised 52.9, 37.4 and 9.7% of the total number of individuals caught, respectively. Bivalvia was the taxon most represented with eight species, followed by Osteichthyes and Brachyura, with four and three species, respectively. Apart from the target species S. solida, the most abundant species were the bivalves Donax vittatus, Tellina tenuis and Ensis siliqua, the crabs Atelecyclus undecimdentatus and Liocarcinus depurator, and the heart urchin Echinocardium cordatum. Table 3 Total number of individuals that entered the dredges and retained in the cover bag . . North dredge . Grid dredge . Traditional dredge . . Species . Total Cover Total Cover Total Cover Polychaeta Polychaeta 12 12 5 0 Bivalvia Donax trunculus 1 1 Donax vittatus 309 89 1392 1385 1441 152 Dosinia exoleta 2 0 Ensis siliqua 2 1 117 110 45 0 Mactra corallina stultorum 1 0 33 9 19 0 Spisula solida 2347 31 12211 1343 8484 91 Tellina tenuis 11 10 640 638 87 47 Venus striatula 5 5 2 0 Cephalopoda Sepia officinalis 2 0 Anomura Pagurus spp. 4 4 5 2 Brachyura Atelecyclus undecimdentatus 61 10 705 47 478 2 Leucarcinus depurator 39 20 150 129 295 28 Polybius heslowi 4 1 21 2 14 0 Echinoidea Echinocardium cordatum 25 4 120 7 12 0 Ophiuroidea Ophiura texturata 1 1 Osteichthyes Citharus linguatula 1 0 2 0 Dicologoglossa cuneata 5 0 Trachinus draco 1 1 Trachinus vipera 7 1 Total 2815 174 15413 3689 10891 321 . . North dredge . Grid dredge . Traditional dredge . . Species . Total Cover Total Cover Total Cover Polychaeta Polychaeta 12 12 5 0 Bivalvia Donax trunculus 1 1 Donax vittatus 309 89 1392 1385 1441 152 Dosinia exoleta 2 0 Ensis siliqua 2 1 117 110 45 0 Mactra corallina stultorum 1 0 33 9 19 0 Spisula solida 2347 31 12211 1343 8484 91 Tellina tenuis 11 10 640 638 87 47 Venus striatula 5 5 2 0 Cephalopoda Sepia officinalis 2 0 Anomura Pagurus spp. 4 4 5 2 Brachyura Atelecyclus undecimdentatus 61 10 705 47 478 2 Leucarcinus depurator 39 20 150 129 295 28 Polybius heslowi 4 1 21 2 14 0 Echinoidea Echinocardium cordatum 25 4 120 7 12 0 Ophiuroidea Ophiura texturata 1 1 Osteichthyes Citharus linguatula 1 0 2 0 Dicologoglossa cuneata 5 0 Trachinus draco 1 1 Trachinus vipera 7 1 Total 2815 174 15413 3689 10891 321 Open in new tab Table 3 Total number of individuals that entered the dredges and retained in the cover bag . . North dredge . Grid dredge . Traditional dredge . . Species . Total Cover Total Cover Total Cover Polychaeta Polychaeta 12 12 5 0 Bivalvia Donax trunculus 1 1 Donax vittatus 309 89 1392 1385 1441 152 Dosinia exoleta 2 0 Ensis siliqua 2 1 117 110 45 0 Mactra corallina stultorum 1 0 33 9 19 0 Spisula solida 2347 31 12211 1343 8484 91 Tellina tenuis 11 10 640 638 87 47 Venus striatula 5 5 2 0 Cephalopoda Sepia officinalis 2 0 Anomura Pagurus spp. 4 4 5 2 Brachyura Atelecyclus undecimdentatus 61 10 705 47 478 2 Leucarcinus depurator 39 20 150 129 295 28 Polybius heslowi 4 1 21 2 14 0 Echinoidea Echinocardium cordatum 25 4 120 7 12 0 Ophiuroidea Ophiura texturata 1 1 Osteichthyes Citharus linguatula 1 0 2 0 Dicologoglossa cuneata 5 0 Trachinus draco 1 1 Trachinus vipera 7 1 Total 2815 174 15413 3689 10891 321 . . North dredge . Grid dredge . Traditional dredge . . Species . Total Cover Total Cover Total Cover Polychaeta Polychaeta 12 12 5 0 Bivalvia Donax trunculus 1 1 Donax vittatus 309 89 1392 1385 1441 152 Dosinia exoleta 2 0 Ensis siliqua 2 1 117 110 45 0 Mactra corallina stultorum 1 0 33 9 19 0 Spisula solida 2347 31 12211 1343 8484 91 Tellina tenuis 11 10 640 638 87 47 Venus striatula 5 5 2 0 Cephalopoda Sepia officinalis 2 0 Anomura Pagurus spp. 4 4 5 2 Brachyura Atelecyclus undecimdentatus 61 10 705 47 478 2 Leucarcinus depurator 39 20 150 129 295 28 Polybius heslowi 4 1 21 2 14 0 Echinoidea Echinocardium cordatum 25 4 120 7 12 0 Ophiuroidea Ophiura texturata 1 1 Osteichthyes Citharus linguatula 1 0 2 0 Dicologoglossa cuneata 5 0 Trachinus draco 1 1 Trachinus vipera 7 1 Total 2815 174 15413 3689 10891 321 Open in new tab From Table 3 it can be observed that the ND and the TD retained almost all individuals that entered the dredge (93.9 and 97.1%, respectively), while the GD retained a smaller proportion of individuals (76.1%). Cluster analysis and subsequent multidimensional scaling (MDS) applied to abundance data from all samples collected from the cover bag revealed two main groupings of points (Figure 2). One group corresponded to the GD and the other group contained the ND and TD. The ANOSIM test that accounted for retention type effects (grid vs mesh bag) showed significant differences between the GD and both the TD and ND (r=0.969, P<0.001), reflecting differences on the selectivity of these fishing gears. Figure 2 Open in new tabDownload slide Bray–Curtis cluster analysis and Multidimensional Scaling Ordination (MDS) plot from cover bag data. ND, north dredge; TD, traditional dredge; GD, grid dredge. Figure 2 Open in new tabDownload slide Bray–Curtis cluster analysis and Multidimensional Scaling Ordination (MDS) plot from cover bag data. ND, north dredge; TD, traditional dredge; GD, grid dredge. Table 4 summarises the data concerning the mean percentage of damaged (scores 2–4) and dead individuals (scores 3 and 4) that entered the dredges. Although the mean percentage of both damaged and dead individuals in the overall catch is very low, it was observed that the ND damages and kills a slightly lower proportion of individuals (mean damaged=3.3%; mean mortality=2.5%) than the GD (mean damaged=5.0%; mean mortality=4.8%) and the TD (mean damaged=7.4%; mean mortality=5.9%). However, the statistical analysis carried out showed that gear type had no effect on the percentage of damaged individuals (ANOVA, F=1.48, P=0.240) or dead individuals (K–W, H=5.538, df=2, P=0.057). Within the more abundant species, it was observed that those most affected by this kind of fishery were the thin-shelled bivalves E. siliqua and T. tenuis, the heart urchin E. cordatum and the crab A. undecimdentatus. Table 4 Mean number and mean proportion of individuals damaged and killed that entered the dredges, for each taxon and gear type . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Polychaeta  Polychaeta 2.93 0.59 20.00 0.59 20.00 1.17 0.59 50.00 0.59 50.00 Bivalvia  Donax trunculus 0.31 0.00 0.00 0.00 0.00  Donax vitattus 77.18 3.37 4.37 3.37 4.37 348.05 25.78 7.41 25.20 7.24 360.35 28.13 7.80 26.95 7.48  Dosinia exoleta 0.59 0.00 0.00 0.00 0.00  Ensis siliqua 0.61 0.61 100.00 0.61 100.00 29.30 22.27 76.00 22.27 76.00 11.13 9.38 84.21 9.38 84.21  Mactra corallina stultorum 0.31 0.00 0.00 0.00 0.00 8.20 4.69 57.14 4.10 50.00 4.69 2.34 50.00 2.34 50.00  Spisula solida 586.78 11.03 1.88 5.82 0.99 3052.73 41.60 1.36 35.74 1.17 2121.09 87.30 4.12 55.66 2.62  Tellina tenuis 2.76 0.31 11.11 0.31 11.11 159.96 10.55 6.59 10.55 6.59 21.68 4.69 21.62 2.93 13.51  Venus striatula 1.17 0.00 0.00 0.00 0.00 0.59 0.00 0.00 0.00 0.00 Cephalopoda  Sepia officinalis 0.59 0.59 100.00 0.59 100.00 Anomura  Pagurus spp. 0.92 0.00 0.00 0.00 0.00 1.17 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 15.31 3.68 24.00 3.68 24.00 176.37 63.87 36.21 63.28 35.88 119.53 37.50 31.37 35.16 29.41  Leucarcinus depurator 9.80 0.92 9.38 0.61 6.25 37.50 7.62 20.31 7.62 20.31 73.83 26.95 36.51 22.85 30.95  Polybius heslowi 0.92 0.61 66.67 0.61 66.67 5.27 2.93 55.56 2.93 55.56 3.52 2.34 66.67 2.34 66.67 Echinoidea  Echinocardium cordatum 6.13 2.14 35.00 2.14 35.00 29.88 12.89 43.14 12.89 43.14 2.93 1.76 60.00 1.76 60.00 Ophiuroidea  Ophiura texturata 0.31 0.31 100.00 0.31 100.00 Osteichthyes  Citharus linguatula 0.31 0.31 100.00 0.31 100.00 0.59 0.00 0.00 0.00 0.00  Dicologoglossa cuneata 1.17 0.00 0.00 0.00 0.00  Trachinus draco 0.31 0.00 0.00 0.00 0.00  Trachinus vipera 1.84 0.00 0.00 0.00 0.00 Total 703.76 23.83 3.31 17.76 2.52 3853.13 192.77 5.00 185.16 4.81 2722.85 201.56 7.40 160.55 5.90 . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Polychaeta  Polychaeta 2.93 0.59 20.00 0.59 20.00 1.17 0.59 50.00 0.59 50.00 Bivalvia  Donax trunculus 0.31 0.00 0.00 0.00 0.00  Donax vitattus 77.18 3.37 4.37 3.37 4.37 348.05 25.78 7.41 25.20 7.24 360.35 28.13 7.80 26.95 7.48  Dosinia exoleta 0.59 0.00 0.00 0.00 0.00  Ensis siliqua 0.61 0.61 100.00 0.61 100.00 29.30 22.27 76.00 22.27 76.00 11.13 9.38 84.21 9.38 84.21  Mactra corallina stultorum 0.31 0.00 0.00 0.00 0.00 8.20 4.69 57.14 4.10 50.00 4.69 2.34 50.00 2.34 50.00  Spisula solida 586.78 11.03 1.88 5.82 0.99 3052.73 41.60 1.36 35.74 1.17 2121.09 87.30 4.12 55.66 2.62  Tellina tenuis 2.76 0.31 11.11 0.31 11.11 159.96 10.55 6.59 10.55 6.59 21.68 4.69 21.62 2.93 13.51  Venus striatula 1.17 0.00 0.00 0.00 0.00 0.59 0.00 0.00 0.00 0.00 Cephalopoda  Sepia officinalis 0.59 0.59 100.00 0.59 100.00 Anomura  Pagurus spp. 0.92 0.00 0.00 0.00 0.00 1.17 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 15.31 3.68 24.00 3.68 24.00 176.37 63.87 36.21 63.28 35.88 119.53 37.50 31.37 35.16 29.41  Leucarcinus depurator 9.80 0.92 9.38 0.61 6.25 37.50 7.62 20.31 7.62 20.31 73.83 26.95 36.51 22.85 30.95  Polybius heslowi 0.92 0.61 66.67 0.61 66.67 5.27 2.93 55.56 2.93 55.56 3.52 2.34 66.67 2.34 66.67 Echinoidea  Echinocardium cordatum 6.13 2.14 35.00 2.14 35.00 29.88 12.89 43.14 12.89 43.14 2.93 1.76 60.00 1.76 60.00 Ophiuroidea  Ophiura texturata 0.31 0.31 100.00 0.31 100.00 Osteichthyes  Citharus linguatula 0.31 0.31 100.00 0.31 100.00 0.59 0.00 0.00 0.00 0.00  Dicologoglossa cuneata 1.17 0.00 0.00 0.00 0.00  Trachinus draco 0.31 0.00 0.00 0.00 0.00  Trachinus vipera 1.84 0.00 0.00 0.00 0.00 Total 703.76 23.83 3.31 17.76 2.52 3853.13 192.77 5.00 185.16 4.81 2722.85 201.56 7.40 160.55 5.90 Open in new tab Table 4 Mean number and mean proportion of individuals damaged and killed that entered the dredges, for each taxon and gear type . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Polychaeta  Polychaeta 2.93 0.59 20.00 0.59 20.00 1.17 0.59 50.00 0.59 50.00 Bivalvia  Donax trunculus 0.31 0.00 0.00 0.00 0.00  Donax vitattus 77.18 3.37 4.37 3.37 4.37 348.05 25.78 7.41 25.20 7.24 360.35 28.13 7.80 26.95 7.48  Dosinia exoleta 0.59 0.00 0.00 0.00 0.00  Ensis siliqua 0.61 0.61 100.00 0.61 100.00 29.30 22.27 76.00 22.27 76.00 11.13 9.38 84.21 9.38 84.21  Mactra corallina stultorum 0.31 0.00 0.00 0.00 0.00 8.20 4.69 57.14 4.10 50.00 4.69 2.34 50.00 2.34 50.00  Spisula solida 586.78 11.03 1.88 5.82 0.99 3052.73 41.60 1.36 35.74 1.17 2121.09 87.30 4.12 55.66 2.62  Tellina tenuis 2.76 0.31 11.11 0.31 11.11 159.96 10.55 6.59 10.55 6.59 21.68 4.69 21.62 2.93 13.51  Venus striatula 1.17 0.00 0.00 0.00 0.00 0.59 0.00 0.00 0.00 0.00 Cephalopoda  Sepia officinalis 0.59 0.59 100.00 0.59 100.00 Anomura  Pagurus spp. 0.92 0.00 0.00 0.00 0.00 1.17 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 15.31 3.68 24.00 3.68 24.00 176.37 63.87 36.21 63.28 35.88 119.53 37.50 31.37 35.16 29.41  Leucarcinus depurator 9.80 0.92 9.38 0.61 6.25 37.50 7.62 20.31 7.62 20.31 73.83 26.95 36.51 22.85 30.95  Polybius heslowi 0.92 0.61 66.67 0.61 66.67 5.27 2.93 55.56 2.93 55.56 3.52 2.34 66.67 2.34 66.67 Echinoidea  Echinocardium cordatum 6.13 2.14 35.00 2.14 35.00 29.88 12.89 43.14 12.89 43.14 2.93 1.76 60.00 1.76 60.00 Ophiuroidea  Ophiura texturata 0.31 0.31 100.00 0.31 100.00 Osteichthyes  Citharus linguatula 0.31 0.31 100.00 0.31 100.00 0.59 0.00 0.00 0.00 0.00  Dicologoglossa cuneata 1.17 0.00 0.00 0.00 0.00  Trachinus draco 0.31 0.00 0.00 0.00 0.00  Trachinus vipera 1.84 0.00 0.00 0.00 0.00 Total 703.76 23.83 3.31 17.76 2.52 3853.13 192.77 5.00 185.16 4.81 2722.85 201.56 7.40 160.55 5.90 . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Total . N . (%) . N . (%) . Polychaeta  Polychaeta 2.93 0.59 20.00 0.59 20.00 1.17 0.59 50.00 0.59 50.00 Bivalvia  Donax trunculus 0.31 0.00 0.00 0.00 0.00  Donax vitattus 77.18 3.37 4.37 3.37 4.37 348.05 25.78 7.41 25.20 7.24 360.35 28.13 7.80 26.95 7.48  Dosinia exoleta 0.59 0.00 0.00 0.00 0.00  Ensis siliqua 0.61 0.61 100.00 0.61 100.00 29.30 22.27 76.00 22.27 76.00 11.13 9.38 84.21 9.38 84.21  Mactra corallina stultorum 0.31 0.00 0.00 0.00 0.00 8.20 4.69 57.14 4.10 50.00 4.69 2.34 50.00 2.34 50.00  Spisula solida 586.78 11.03 1.88 5.82 0.99 3052.73 41.60 1.36 35.74 1.17 2121.09 87.30 4.12 55.66 2.62  Tellina tenuis 2.76 0.31 11.11 0.31 11.11 159.96 10.55 6.59 10.55 6.59 21.68 4.69 21.62 2.93 13.51  Venus striatula 1.17 0.00 0.00 0.00 0.00 0.59 0.00 0.00 0.00 0.00 Cephalopoda  Sepia officinalis 0.59 0.59 100.00 0.59 100.00 Anomura  Pagurus spp. 0.92 0.00 0.00 0.00 0.00 1.17 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 15.31 3.68 24.00 3.68 24.00 176.37 63.87 36.21 63.28 35.88 119.53 37.50 31.37 35.16 29.41  Leucarcinus depurator 9.80 0.92 9.38 0.61 6.25 37.50 7.62 20.31 7.62 20.31 73.83 26.95 36.51 22.85 30.95  Polybius heslowi 0.92 0.61 66.67 0.61 66.67 5.27 2.93 55.56 2.93 55.56 3.52 2.34 66.67 2.34 66.67 Echinoidea  Echinocardium cordatum 6.13 2.14 35.00 2.14 35.00 29.88 12.89 43.14 12.89 43.14 2.93 1.76 60.00 1.76 60.00 Ophiuroidea  Ophiura texturata 0.31 0.31 100.00 0.31 100.00 Osteichthyes  Citharus linguatula 0.31 0.31 100.00 0.31 100.00 0.59 0.00 0.00 0.00 0.00  Dicologoglossa cuneata 1.17 0.00 0.00 0.00 0.00  Trachinus draco 0.31 0.00 0.00 0.00 0.00  Trachinus vipera 1.84 0.00 0.00 0.00 0.00 Total 703.76 23.83 3.31 17.76 2.52 3853.13 192.77 5.00 185.16 4.81 2722.85 201.56 7.40 160.55 5.90 Open in new tab The scuba-diving surveys allowed an estimate of the efficiency of capture of the dredges. For the ND an efficiency of capture of 64% was estimated and the incidental mortality on uncaught white clams was in the range 5–20%. Higher efficiencies of capture were estimated for both the GD (98%) and TD (90%), while for both dredges no damage on the uncaught white clams was observed. It is interesting to note that higher catch efficiencies lead to a lower proportion of damaged individuals that are left in the path of the dredge. In Table 5, it can be seen that the ND damages and kills a higher proportion of the uncaught individuals than the GD and TD. Table 5 Mean number and mean proportion of damaged and dead individuals left in the dredge path, for each taxon and gear type. . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . % . N . % . Total . N . % . N . % . Total . N . % . N . % . Bivalvia  Donax vitattus 147.00 0.00 0.00 0.00 0.00 17.33 0.00 0.00 0.00 0.00 75.33 58.00 76.99 58.00 76.99  Spisula solida 326.00 146.67 44.99 33.33 10.22 58.00 0.00 0.00 0.00 0.00 233.33 0.00 0.00 0.00 0.00  Tellina tenuis 402.00 217.67 54.15 217.67 54.15 1170.67 208.67 17.82 208.67 17.82 1060.67 300.67 28.35 185.33 17.47 Anomura  Pagurus spp. 4.00 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 2.00 2.00 100.00 2.00 100.00 Echinoidea  Echinocardium cordatum 6.00 6.00 100.00 6.00 100.00 1.33 0.00 0.00 0.00 0.00 Total 881.00 370.33 42.04 257.00 29.17 1248.00 210.67 16.88 210.67 16.88 1374.67 358.67 26.09 243.33 17.70 . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . % . N . % . Total . N . % . N . % . Total . N . % . N . % . Bivalvia  Donax vitattus 147.00 0.00 0.00 0.00 0.00 17.33 0.00 0.00 0.00 0.00 75.33 58.00 76.99 58.00 76.99  Spisula solida 326.00 146.67 44.99 33.33 10.22 58.00 0.00 0.00 0.00 0.00 233.33 0.00 0.00 0.00 0.00  Tellina tenuis 402.00 217.67 54.15 217.67 54.15 1170.67 208.67 17.82 208.67 17.82 1060.67 300.67 28.35 185.33 17.47 Anomura  Pagurus spp. 4.00 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 2.00 2.00 100.00 2.00 100.00 Echinoidea  Echinocardium cordatum 6.00 6.00 100.00 6.00 100.00 1.33 0.00 0.00 0.00 0.00 Total 881.00 370.33 42.04 257.00 29.17 1248.00 210.67 16.88 210.67 16.88 1374.67 358.67 26.09 243.33 17.70 Open in new tab Table 5 Mean number and mean proportion of damaged and dead individuals left in the dredge path, for each taxon and gear type. . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . % . N . % . Total . N . % . N . % . Total . N . % . N . % . Bivalvia  Donax vitattus 147.00 0.00 0.00 0.00 0.00 17.33 0.00 0.00 0.00 0.00 75.33 58.00 76.99 58.00 76.99  Spisula solida 326.00 146.67 44.99 33.33 10.22 58.00 0.00 0.00 0.00 0.00 233.33 0.00 0.00 0.00 0.00  Tellina tenuis 402.00 217.67 54.15 217.67 54.15 1170.67 208.67 17.82 208.67 17.82 1060.67 300.67 28.35 185.33 17.47 Anomura  Pagurus spp. 4.00 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 2.00 2.00 100.00 2.00 100.00 Echinoidea  Echinocardium cordatum 6.00 6.00 100.00 6.00 100.00 1.33 0.00 0.00 0.00 0.00 Total 881.00 370.33 42.04 257.00 29.17 1248.00 210.67 16.88 210.67 16.88 1374.67 358.67 26.09 243.33 17.70 . North dredge . Grid dredge . Traditional dredge . . . Damaged . Mortality . . Damaged . Mortality . . Damaged . Mortality . . Total . N . % . N . % . Total . N . % . N . % . Total . N . % . N . % . Bivalvia  Donax vitattus 147.00 0.00 0.00 0.00 0.00 17.33 0.00 0.00 0.00 0.00 75.33 58.00 76.99 58.00 76.99  Spisula solida 326.00 146.67 44.99 33.33 10.22 58.00 0.00 0.00 0.00 0.00 233.33 0.00 0.00 0.00 0.00  Tellina tenuis 402.00 217.67 54.15 217.67 54.15 1170.67 208.67 17.82 208.67 17.82 1060.67 300.67 28.35 185.33 17.47 Anomura  Pagurus spp. 4.00 0.00 0.00 0.00 0.00 Brachyura  Atelecyclus undecimdentatus 2.00 2.00 100.00 2.00 100.00 Echinoidea  Echinocardium cordatum 6.00 6.00 100.00 6.00 100.00 1.33 0.00 0.00 0.00 0.00 Total 881.00 370.33 42.04 257.00 29.17 1248.00 210.67 16.88 210.67 16.88 1374.67 358.67 26.09 243.33 17.70 Open in new tab The mean percentage of both damaged and dead uncaught individuals from the TD was also found to be higher than those obtained from the GD. The results of one-way ANOVA showed that gear type has an effect on both the percentage of damaged (F=10.114, P=0.005) and dead (F=4.341, P=0.048) individuals left on the dredge track. A Student–Newman–Keuls multiple pairwise comparison showed significant differences between the ND and both the GD and TD, both in terms of damaged and dead individuals. Within the dredge tracks, bivalve species were the most abundant group, comprising nearly 100% of the total number of macrofaunal individuals collected. Among these, T. tenuis was the most affected species followed by D. vittatus. Fishing yield is known to be directly related to the efficiency of capture of dredge gears and therefore differences in the mean fishing yield (kg/5 min tow) obtained for each dredge were observed. From Figure 3 it can be seen that the mean fishing yield registered both for the GD and TD was substantially higher than that observed for the ND. The one-way ANOVA analysis performed revealed significant differences (F=16.486, P=0.004) in the mean fishing yield obtained for the dredges assayed. Application of the Student–Newman–Keuls test showed the existence of significant differences in the mean fishing yield (S–N–K, P<0.05), between the GD and ND, and between the TD and ND. Figure 3 Open in new tabDownload slide Standardised mean fishing yields (kg/5 min tow) obtained for the three dredges assayed. ND, north dredge; TD, traditional dredge; GD, grid dredge. Figure 3 Open in new tabDownload slide Standardised mean fishing yields (kg/5 min tow) obtained for the three dredges assayed. ND, north dredge; TD, traditional dredge; GD, grid dredge. Finally, for each dredge and tow, data from the bag, cover and dredge path were pooled in order to estimate and compare total mortality. Table 6 summarises the data obtained for each dredge and tow in terms of the percentage of damaged and dead individuals. Data analysis shows that for the overall community the ND damages and kills a higher proportion of macrofaunal individuals than the GD and TD. The Kruskal–Wallis one way ANOVA on Ranks revealed the existence of significant differences both in terms of damaged (K–W, H=8.769, df=2, P=0.001) and dead individuals percentage (K–W, H=6.615, df=2, P=0.024). The Student–Newman–Keuls Pairwise Multiple Comparison showed significant differences (P<0.05) between the ND and both the GD and TD for damaged and dead individuals. Table 6 Percentage of damaged and dead individuals obtained per tow and dredge type . Tow . Mean . . #1 . #2 . #3 . #4 . . . . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . North dredge 36.35 26.79 19.51 11.64 18.30 13.26 30.44 21.47 26.15 18.29 Grid dredge 11.92 11.40 9.82 9.57 6.03 6.00 5.64 5.50 8.35 8.12 Traditional dredge 11.27 8.44 16.27 12.75 10.78 7.12 17.26 11.73 13.89 10.01 . Tow . Mean . . #1 . #2 . #3 . #4 . . . . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . North dredge 36.35 26.79 19.51 11.64 18.30 13.26 30.44 21.47 26.15 18.29 Grid dredge 11.92 11.40 9.82 9.57 6.03 6.00 5.64 5.50 8.35 8.12 Traditional dredge 11.27 8.44 16.27 12.75 10.78 7.12 17.26 11.73 13.89 10.01 Open in new tab Table 6 Percentage of damaged and dead individuals obtained per tow and dredge type . Tow . Mean . . #1 . #2 . #3 . #4 . . . . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . North dredge 36.35 26.79 19.51 11.64 18.30 13.26 30.44 21.47 26.15 18.29 Grid dredge 11.92 11.40 9.82 9.57 6.03 6.00 5.64 5.50 8.35 8.12 Traditional dredge 11.27 8.44 16.27 12.75 10.78 7.12 17.26 11.73 13.89 10.01 . Tow . Mean . . #1 . #2 . #3 . #4 . . . . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . Damaged (%) . Mortality (%) . North dredge 36.35 26.79 19.51 11.64 18.30 13.26 30.44 21.47 26.15 18.29 Grid dredge 11.92 11.40 9.82 9.57 6.03 6.00 5.64 5.50 8.35 8.12 Traditional dredge 11.27 8.44 16.27 12.75 10.78 7.12 17.26 11.73 13.89 10.01 Open in new tab Discussion The direct mortality on the macrobenthic community inflicted by three types of clam dredges used in the S. solida fishery was both determined and compared in this study. Total direct mortality was assessed, taking into consideration the degree of damage sustained by individuals that entered the dredges, plus those individuals damaged and left in the dredges path. Our results showed significant differences in total direct mortality between the north dredge and both the grid and traditional dredges. These differences were largely attributed to the animals in the dredge track that died as a direct result of physical damage inflicted by the dredge passing. It was found during the study that damage on uncaught individuals was directly related to gear efficiency. The lower catching efficiency of the north dredge led to a higher proportion of damaged individuals left in the dredge track, when compared with the more efficient grid and traditional dredges. This relationship between catching efficiency and damage has also been observed by other authors. Meyer et al. (1981) reported that when the efficiency of dredges was low, larger clams, which burrowed deeper into the sediment, suffered mortalities as high as 92%, and when efficiency was high, mortalities decreased to 30%. Caddy (1973) noted that the low efficiency of the Alberton dredge was responsible for causing a high amount of lethal and sublethal damage to scallops left in the dredge's track. This amount of damage was also found to be higher on more rough seabeds. McLoughlin et al. (1991) concluded that in addition to its low catching efficiency, the Australian mud dredge damages many more scallops than it catches, producing a post-fishing mortality rate seven times the estimated natural mortality rate for Pecten fumatus. However, it should also be emphasised that the maximum dredging impact may not occur immediately after dredging, as exposed organisms may be predated. The attraction of epifaunal scavengers and predators to fished areas has been recorded in other studies (e.g. Meyer et al., 1981; Kaiser and Spencer, 1994; Lambert and Goudreau, 1996; Ramsay et al., 1996; Bergman and van Santbrink, 2000). Analysis of the diet composition of scavengers collected from trawled areas indicated that they feed primarily on animals that were damaged or disturbed by the trawl (Kaiser and Spencer, 1996). As well as direct mortality from being caught and indirect mortality due to predation on uncaught clams, there may be further mortality on discarded individuals (Veale et al., 2000), especially if sorting times are long and conditions on deck are unfavourable (Medcof and Bourne, 1964). Furthermore, re-location into unsuitable habitat and predation while returning to the seafloor after being discarded from the ship's deck may also contribute to increased mortality (Gaspar, 1996). This type of mortality also depends on many conditions, such as depth, type of species, individual's size, degree of damage and predator concentration. Gaspar and Monteiro (1999) reported that the length of exposure to air on deck was directly related to juvenile S. solida mortality. Robinson and Richardson (1998) found that undersized Ensis arcuatus individuals returned to the seabed were slow to re-bury, becoming highly vulnerable to attack from predatory crabs. These two examples illustrate the importance of designing more highly selective dredges. Our results showed that the GD retained a significantly smaller proportion of captured individuals than the TD and ND, reflecting differences in the collecting system used in the dredges (metallic grid vs net bag). When a net bag is used to retain the individuals, the mesh stretches while the dredge is being towed and prevents the escape of organisms through the mesh. The dredge therefore only becomes selective during the hauling process. When the metallic grid is used, selection of the captured individuals occurs throughout the tow. Gaspar et al. (2001) reported that the undamaged individuals that pass through the parallel rods of the GD grid burrow immediately (in the case of the infauna) or recover their activity (in the case of epifauna). This rapid reburying response decreases the probability of dislodged organisms being predated. From our results it can be concluded that there are significant direct effects of dredging on some benthic species, as certain groups of animals suffer heavy damage while others are less affected. Studies have demonstrated consistently that there is an immediate effect on the density of both target and non-targeted organisms after the impact of mobile fishing gears. The short-term environmental effects of dredging on the sea floor have received increased attention in the last decade and several studies have detected changes in benthic communities due to dredging (e.g. Hall et al., 1990; Eleftheriou and Robertson, 1992; Kaiser and Spencer, 1996; Lambert and Goudreau, 1996; Bergman and van Santbrink, 2000). Short-term effects are therefore also expected in the Portuguese bivalve dredge fishery, but the question is whether or not this type of fishing causes long-term effects in the benthic community structure. Biological communities that utilise a particular habitat have adapted to their environment through natural selection and the impact of mobile fishing gears on the habitat structure and biological community can be scaled against the magnitude and frequency of seabed disturbance due to natural causes (De Alteris et al., 1999). Although, for various species, mortality due to dredging appears to be fairly high, recolonisation can occur over a relatively short time period. Currie and Parry (1996), using a before-after-control-impact design experiment, reported the size and duration of scallop dredging impacts on soft sediment communities. The authors stated that reductions in density caused by dredging were usually small compared with annual changes in population density, where seasonal, and particularly inter-annual changes, were greater than those caused by dredging. Kaiser et al. (1998) found that immediately after fishing the composition of the community in stable sediments was significantly altered, while in mobile sediments the effects of fishing were not detectable. Nevertheless, after 6 months, seasonal changes had occurred in both communities and the effects of trawling disturbance were no longer evident. Similarly, Hall et al. (1990) found that despite the fact that suction dredging for Ensis sp. had profound immediate effects on benthic community structure, with consistent reduction in many macrofaunal species, after 40 days the abundance of species returned to pre-impact levels. By contrast, Pranovi and Giovanardi (1994) found that hydraulic dredging produced considerable long-term negative effects on the bottom environment of Venetian lagoon. These authors hypothesised that the slow recovery of the infaunal community was related to the medium/low energy conditions of the lagoon environment. Benthic communities inhabiting deeper waters may be less capable of sustaining and overcoming disturbance than benthic populations in shallow waters in more dynamic coarser sediments and accordingly have much longer recovery times (Jones, 1992). Besides sediment type and conditions at a site, the severity of accumulated fishing effects also depends on the scale and intensity of the activity. If a large proportion of a fishing area is affected, then it is quite conceivable that the scope for movement by the associated benthos would be reduced and recovery would take longer (Hall, 1994; Thrush et al., 1995). Furthermore, although the effects of a single passage of a dredge gear may be relatively limited, chronic fishing disturbance may produce long-term changes in benthic communities (Sainsbury, 1988; Collie et al., 1997; Jennings and Kaiser, 1998; Bradshaw et al., 2000). Evidence nevertheless suggests that long-term changes in mobile sediments are probably restricted to long-lived fragile species (Eleftheriou and Robertson, 1992). Therefore, population reductions may only persist if the sediments in which they live are immobile (e.g. Kaiser, 1998; Ball et al., 2000) or that the affected area is large relative to the remainder of the habitat and a dilution effect cannot occur (Kaiser, 1998). Thus, given the depth (<35 m) and the type of sediment (sandy bottoms) on which fishing is practised along the Portuguese coast and the relatively high natural disturbance found all year round, clam dredging is unlikely to have persistent effects on most infaunal communities. The effects on long-lived bivalve species could, however, be more serious. From this study, it was found that the ND damages and kills a higher proportion of macrofaunal individuals than the GD and TD. It was also found that for the same tow duration the ND mean fishing yield is significantly lower than those obtained with the GD and TD. Finally, our data showed that the GD is more selective than the other two dredges assayed. These results indicate that there are advantages in using the GD in the white clam fishery. Thus, in order to ban the use of the TD, fishermen of the small local dredge fleet should equip their boats with a small winch allowing for the use of the GD. The ND provokes a higher deleterious effect on the ecosystem than when the GD is used; therefore the ND should be banned from the fishery and replaced by the GD. From a fisheries management and ecological points of view, our results clearly showed that there are obvious advantages in developing more efficient and selective dredges in order to reduce the number of damaged individuals and by-catch, and consequently decreasing the impact of dredging on macrobenthic communities. We would like to thank Mr Mike Heasman for reviewing and for useful comments on the manuscript. We also thank Luz Abreu, António Laranjo and José Sofia for technical assistance during field work and the crew of R/V “DONAX” for their skilful handling of the boat and the gears. This study was in part funded by the European Commission (FAIR CT-98-4465; Project ECODREDGE). References Ball B , Murday B , Tuck I . Kaiser M.J , De Groot S.J . Effects of the otter trawling on the benthos and environment in muddy sediments , Effects of Fishing on Non-Target Species and Habitats. Biological, Conservation and Socio-Economic Issues , 2000 Oxford Blackwell Science (pg. 69 - 82 ) 416 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bergman M.J.N , van Santbrink J.W . Mortality in megafaunal benthic populations caused by trawl fisheries on the Dutch continental shelf in the North Sea in 1994 , ICES Journal of Marine Science , 2000 , vol. 57 5 (pg. 1321 - 1331 ) Google Scholar Crossref Search ADS WorldCat Bradshaw C , Veale L.O , Hill A.S , Brand A.R . Kaiser M.J , De Groot S.J . The effect of scallop dredging on gravelly seabed communities , Effects of Fishing on Non-Target Species and Habitats. Biological, Conservation and Socio-Economic Issues , 2000 Oxford Blackwell Science (pg. 83 - 104 ) 416 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bucquoy E. , Dautzenberg P. , Dollfus G. . Les mollusques marins du Rousillon , Pelecypodes, Tome 2 , 1882 Paris J. Baillière and Fils Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Caddy J. F. . Efficiency and selectivity of the Canadian offshore scallop dredge , 1971 ICES Shellfish and Benthos Committee Document. ICES CM 1971/K: 25 Caddy J.F . Underwater observations on tracks of dredges and trawls and some effects of dredging on scallop ground , Journal of the Fisheries Research Board of Canada , 1973 , vol. 30 (pg. 173 - 180 ) Google Scholar Crossref Search ADS WorldCat Clark K.R , Warwick R.M . , Change in Marine Communities: An Approach to Statistical Analysis and Interpretation , 1994 Plymouth Natural Environment Research Council, Plymouth Marine Laboratory 144 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Collie J.S , Escanero G.A , Valentine P . Effects of bottom fishing on the benthic megafauna of George Bank , Marine Ecology Progress Series , 1997 , vol. 155 (pg. 159 - 172 ) Google Scholar Crossref Search ADS WorldCat Currie D.R , Parry G.D . Effects of scallop dredging on a soft sediment community: a large-scale experimental study , Marine Ecology Progress Series , 1996 , vol. 134 1–3 (pg. 131 - 150 ) Google Scholar Crossref Search ADS WorldCat De Alteris J , Skrobe L , Lipsky C . Benaka L . The significance of seabed disturbance by mobile fishing gear relative to natural processes: a case study in Narragansett Bay , Fish Habitat: Essential Fish Habitat and Rehabilitation , 1999 Bethesda, MD American Fisheries Society (pg. 224 - 237 ) American Fisheries Society, Symposium 22 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Eleftheriou A , Robertson M.R . The effects of experimental scallop dredging on the fauna and physical environment of a shallow sandy community , Netherlands Journal of Sea Research , 1992 , vol. 30 (pg. 289 - 299 ) Google Scholar Crossref Search ADS WorldCat FAO Fischer W , Schneider M , Bauchot M.-L . Fiches FAO identification des especes pour les besoins de la peche. (Révision 1) , Mediterranee et Mer Noire , 1987 , vol. Vol. 1 Rome FAO (pg. 371 - 512 ) Zone de peche 37. Vegetaux et Invertebres 760 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gaspar M. B. . Bivalves do litoral oceânico algarvio. Aspectos da biologia, ecologia e das pescarias dos mananciais de interesse económico: aplicação à gestão dos recursos , 1996 PhD thesis, Universidade do Algarve, Faro. 282 pp Gaspar M.B , Monteiro C.C . Indirect mortality caused to juveniles of Spisula solida due to deck exposure , Journal of Marine Biological Association of UK , 1999 , vol. 79 (pg. 567 - 568 ) Google Scholar Crossref Search ADS WorldCat Gaspar M.B , Castro M , Monteiro C.C . Effect of tooth spacing and mesh size on the catch of the Portuguese clam and razor clam dredge , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 103 - 110 ) Google Scholar Crossref Search ADS WorldCat Gaspar M.B , Dias M.D , Campos A , Monteiro C.C , Santos M.N , Chícharo A , Chícharo L . The influence of dredge design on the catch of Callista chione (Linnaeus, 1758) , Hydrobiologia , 2001 , vol. 465 1–3 (pg. 153 - 167 ) Google Scholar Crossref Search ADS WorldCat Hall S.J . Physical disturbance and marine benthic communities: life in unconsolidated sediments , Oceanography and Marine Biology: An Annual Review , 1994 , vol. 32 (pg. 179 - 239 ) OpenURL Placeholder Text WorldCat Hall S.J , Basford D.J , Robertson M.R . The impact of hydraulic dredging for razor clams Ensis sp. on an infaunal community , Netherlands Journal of Sea Research , 1990 , vol. 27 1 (pg. 119 - 125 ) Google Scholar Crossref Search ADS WorldCat Jennings S , Kaiser M.J . The effect of fishing on marine ecosystems , Advances in Marine Biology , 1998 , vol. 34 (pg. 201 - 252 ) Google Scholar Crossref Search ADS WorldCat Jones J.B . Environmental impact of trawling on the seabed: a review , New Zealand Journal of Marine and Freshwater Research , 1992 , vol. 26 1 (pg. 59 - 67 ) Google Scholar Crossref Search ADS WorldCat Kaiser M.J . Significance of bottom-fishing disturbance , Conservation Biology , 1998 , vol. 12 6 (pg. 1230 - 1235 ) Google Scholar Crossref Search ADS WorldCat Kaiser M.J , Spencer B.E . Fish scavenging behaviour in recently trawled areas , Marine Ecology Progress Series , 1994 , vol. 126 (pg. 31 - 38 ) Google Scholar Crossref Search ADS WorldCat Kaiser M.J , Spencer B.E . The effects of beam-trawl disturbance on infaunal communities in different habitats , Journal of Animal Ecology , 1996 , vol. 65 3 (pg. 348 - 358 ) Google Scholar Crossref Search ADS WorldCat Kaiser M.J , Edwards D.B , Spencer B.E . Infaunal community changes as a result of commercial clam cultivation and harvesting , Aquatic Living Resources , 1996 , vol. 9 (pg. 57 - 63 ) Google Scholar Crossref Search ADS WorldCat Kaiser M.J , Edwards D.B , Armstrong P.J , Radford K , Lough N.E.L , Flat R.P , Jones H.D . Changes in megafauna benthic communities in different habitats after trawling disturbance , ICES Journal of Marine Science , 1998 , vol. 55 3 (pg. 353 - 361 ) Google Scholar Crossref Search ADS WorldCat Lambert J. , Goudreau P. . Performance of the New England Hydraulic dredge for the harvest of Stimpson's surf Clams (Mactromeris polynyma) , 1996 Canadian Industry Report of Fisheries and Aquatic Science, 235. vii+28 p McLoughlin R.J , Young P , Martin R , Parslow J . The Australian scallop dredge: estimates of catching efficiency and associated indirect fishing mortality , Fisheries Research , 1991 , vol. 11 (pg. 1 - 24 ) Google Scholar Crossref Search ADS WorldCat Medcof J , Bourne N . Causes of mortality of the sea scallop Placopecten magellanicus , Proceedings of National Shellfish Association , 1964 , vol. 53 (pg. 33 - 50 ) OpenURL Placeholder Text WorldCat Meyer T.L , Cooper R.A , Pecci K.J . The performance and environment effects of a hydraulic dredge , Marine Fisheries Review , 1981 , vol. 43 9 (pg. 14 - 22 ) OpenURL Placeholder Text WorldCat Poppe G.T , Goto Y . , European seashells Scaphopoda, Bivalvia and Cephalopoda , 1993 , vol. Volume II Germany Verlag Christa Hemmen 221 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pranovi F , Giovanardi O . The impact of hydraulic dredging for short-necked clams, Tapes spp., on an infauna community in the lagoon of Venice , Scientia Marina , 1994 , vol. 58 4 (pg. 345 - 353 ) OpenURL Placeholder Text WorldCat Ramsay K , Kaiser M , Hughes R . Changes in hermit crab feeding patterns in response to trawling disturbance , Marine Ecology Progress Series , 1996 , vol. 144 1–3 (pg. 63 - 72 ) Google Scholar Crossref Search ADS WorldCat Robinson R.F , Richardson C.A . The direct and indirect effects of suction dredging on a razor clam (Ensis arcuatus) population , ICES Journal of Marine Science , 1998 , vol. 55 5 (pg. 970 - 977 ) Google Scholar Crossref Search ADS WorldCat Sainsbury K.J . Gulland J.A . The ecological basis of multispecies fisheries management of a demersal fishery in tropic Australia , Fish Population Dynamics , 1988 Chichester Wiley (pg. 349 - 382 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tebble N . , British Bivalve Seashells. A Handbook for Identification , 1996 Edinburgh British Museum 212 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Thrush S.F , Hewitt J.E , Cummings V.J , Dayton P.K . The impact of habitat disturbance by scallop dredging on marine benthic communities: what can be predicted from the results of experiments? , Marine Ecology Progress Series , 1995 , vol. 129 1–3 (pg. 141 - 150 ) Google Scholar Crossref Search ADS WorldCat Van Dolah R.F , Wendt P.H , Nicholson N . Effects of a research trawl on a hard-bottom assemblage of sponges and corals , Fisheries Research , 1987 , vol. 5 1 (pg. 39 - 54 ) Google Scholar Crossref Search ADS WorldCat Veale L.O , Hill A.S , Brand A.R . An in situ study of predator aggregations on scallops (Pecten maximus (L.)) dredge discards using a static time-lapse camera system , Journal of Experimental Marine Biology and Ecology , 2000 , vol. 253 (pg. 111 - 129 ) Google Scholar Crossref Search ADS WorldCat © 2003 International Council for the Exploration of the Sea

Lewy,, P.;Nielsen,, A.

doi: 10.1016/S1054-3139(03)00080-8pmid: N/A

Abstract A new age-structured stock dynamics approach including stochastic survival and recruitment processes is developed and implemented. The model is able to analyse detailed sources of information used in standard age-based fish stock assessment such as catch-at-age and effort data from commercial fleets and research surveys. The stock numbers are treated as unobserved variables subject to process errors while the catches are observed variables subject to both sampling and process errors. Results obtained for North Sea plaice using Markov Chain Monte Carlo methods indicate that the process error by far accounts for most of the variation compared to sampling error. Comparison with results from a simpler separable model indicates that the new model provides more precise estimates with fewer parameters. Introduction The precautionary approach has become a basic concept in fish stock management (Anon., 1995). The concept implies that uncertainties have to be taken into account in the assessment of the fishery e.g. by estimating the risk that the stock biomass falls below a certain critical limit. The quantification of these uncertainties has emphasised the need for developing stochastic assessment approaches. Numerous stochastic assessment methods including frequentist, state space, time series and Bootstrap models have been suggested (e.g. Doubleday, 1976; Fournier and Archibald, 1982; Deriso et al., 1985; Gavaris, 1988; Lewy, 1988; Methot, 1990; Powers and Restrepo, 1993; Gudmundsson, 1994; Schnute, 1994; Patterson and Melvin, 1996). In the 1990s Bayesian methods have been used in connection with biomass dynamics models (Kinas, 1996; McAllister and Kirkwood, 1998; Millar and Meyer, 2000), with models that bridge the gap between biomass and age-structured models (McAllister et al., 1994; Meyer and Millar, 1999a) and with fully age-structured models (Ianelli and Fournier, 1998; Virtala et al., 1998; Patterson, 1999). However, only a few authors have considered the stock dynamics as a stochastic process or the stock size of a cohort as an unknown stochastic variable. Virtala et al. (1998) have formulated a consistent model, where the number of survivors in a cohort, the fish caught and the number dead from natural causes are assumed to be multinomially distributed. As noted by the authors the limitation of this model is that for stocks with millions of fish the model in practice becomes nearly deterministic. Sullivan (1992) and Gudmundsson (1994) applied Kalman filter approaches to length and age-structured state space models, respectively, and thus considered the survival in a cohort as a stochastic process. Schnute and Richards (1995) formulated an age-structured state space model including both process and measurement errors. The properties of estimates based on a simpler model that only include measurement error were evaluated using data generated from the general model. In this paper an age-structured assessment model with structural relations between variables and parameters is developed, where stock numbers are treated as unknown stochastic variables subject to process error and the catch variables subject to both sampling and other process errors. Estimates of parameters including process variances and predicted stock numbers have been obtained using likelihood-based Markov Chain Monte Carlo (MCMC). The assessment model enables the inclusion of detailed sources of information used in standard age-structured assessment such as catch-at-age and effort data from commercial fleets and research surveys. Catch data without effort information is combined into one fleet and all catch data by fleet are treated as stochastic variables subject to sampling and process errors. The usual problem of weighting the different sources of information (catch-at-age observations by fleet, the survival and recruitment processes) is solved by estimating the associated variances. Using data for North Sea plaice the importance of the process error is investigated and the properties of the estimated biomass and mortality rates are compared with the results of a simple separable model as well as with results from extended survivors analysis (XSA, Shepherd, 1999). Population dynamics models The stochastic age-based stock assessment models considered include the type of data used in many fish stock assessments in the North Atlantic, where the following information is available for a range of years: where subscripts refer to fleet f for age a in year y. The sum of catches for the fleets with effort data and residual catch equals total international catch. Catch-at-age in numbers and effort data for commercial fleets (Cf,a,y and ef,y) Catch-at-age in numbers without effort data for the remaining part of total international catches (residual catches, Cres,a,y) CPUE by age for research surveys (Is,a,y) All catch-at-age observations are assumed to be stochastic variables. The model applies residual catches, Cres,a,y, as well as commercial catches and effort data by fleet as observations instead of total international catches used in deterministic VPA approaches or partly stochastic approaches, which assume that total catches in numbers are known without error. The application of residual catches ensures that the catch observations for different fleets are independent variables. Effort data were incorrectly treated as covariates assumed to be known without errors. The numbers of survivors in a cohort are considered as unobserved variables, which are subject to stochastic variations caused by fishing and natural mortality processes. Similarly, the observed numbers of fish caught are subject to stochastic variation due to the fishing process as well. Further, the catches also are subject to sampling errors. The application of a survivor and catch dynamic model that includes process variation is natural, because even with perfect knowledge of the state of the system we would not be able to accurately predict tomorrow's survivors or catch. The following lognormal distributions have been used to describe the stochastic models used for describing the dynamics of stock and catch: (1) (2) (3) (4) where N denotes the stock number, F the fishing mortality, Z the total mortality, σsurvival, σres, σf and σs the standard deviations for the survival and fishing processes, q the catchability, e the effort, T the day of year when the survey takes place and the ɛs the standardised normal distribution. Theoretically, Equation (1) does not prevent stock numbers at a given time from exceeding stock numbers at an earlier time. To ensure this condition an alternative stock dynamics model was formulated: (1a) The output of the models (1) and (1a) will be compared. Equations (1)–(4) imply that the median of the four types of variables equals the standard deterministic stock and catch expressions and ⁠. It is assumed that the observed catches conditioned on stock numbers are mutually independent and that catchability remains unchanged over time for both commercial fleets and surveys. To simplify the model it has been assumed that effort is known without error and that the oldest age group is assumed to be a “true” age group and not a plus-group. To reduce the numbers of parameters, fishing mortality for the residual fleet defined in Equation (2) is assumed to be multiplicative: where Fres,y=1 is fixed to 1 to obtain a unique parameterisation. The number of parameters is further reduced by assuming that the age-dependent factors are identical for the three oldest age groups, i.e. Fres,a=A−2=Fres,a=A−1=Fres,a=A, where A is the maximum age considered. Recruitment to the stock is also considered a stochastic process and a Ricker stock-recruitment model has been used to model the relation between recruitment and spawning stock biomass, SSB. Recruitment is assumed to be lognormally distributed: (5) where minA is the recruitment age, σrecruit the standard deviation of the recruitment process and where the parameters, α, β and σrecruit will be estimated simultaneously with other parameters. The number of recruits in the first year is also assumed to be lognormally distributed: (6) where and μstart and σinitial are the standard deviation and variance of the stock number of the first age-class in the first year, respectively. Finally the remaining initial stock size-at-age in the first year is modelled assuming that the expected stock size is in equilibrium in the first year: (7) The variance parameters, σrecruit, and σsurvival, are allowed to differ because they are related to different processes of stock dynamics while σinitial2 is different from the two process variances because it relates to the assumption of equilibrium of the initial stock. The 55 parameters considered are: In Appendix A the models specified in Equations (1)–(4) are shown to be partly supported by stochastic extension of the standard deterministic stock and catch differential equations. The lognormal distribution of the survivors used, Equation (1), is shown to follow directly from the stochastic formulation in Appendix A. The variance of the survival process, σsurvial2, is shown to be the sum of the variances associated with the mortalities due to fishing and natural causes defined in Appendix A. The corresponding probability distributions of the catch observations cannot be analytically derived in support of the lognormal catch distributions assumed in Equations (2)–(4). Nevertheless it can be shown that the expected value of the catches equals the standard deterministic expressions, which approximately equals the assumption made in Equations (2)–(4). The stock numbers defined by Equation (1), are considered as unobserved random variables predicted from the estimates of the parameters involved. This is in contrast to most stock assessment models where the initial stock sizes are treated as parameters and estimated directly. The catch and survey observations, Cres,a,y, Cf,a,y and Is,a,y are implicitly sampling estimates used as estimates of the true values. Using the formula for conditional variances, and assuming that log sampling estimates are unbiased, the variances for each of the three types of log catch observations are shown to be the sum of two components: a fishing process contribution and a sampling contribution, i.e.: (8) (9) (10) where and etc. It should be noted that if the sampling error is known externally to the model then the process variances can be estimated by the residuals, and (“∧” indicates estimated values). For the case-study of North Sea plaice the sampling variance was externally known. Here the importance of the process variances, σprocess,res2, σprocess,f2 and σprocess,s2, is quantified by the proportions ⁠. To enable comparisons of the output from our model with other approaches, we considered a simple conventional method, which does not include the stochastic survival of the fish in the sea. The stochastic catch models in this method are exactly the same as described above, except that the initial stock numbers in the sea are treated as parameters and the process models, Equations (1) and (5)–(7), are replaced by the standard deterministic model: The initial stock sizes, that is the recruits, ⁠, and stock-size-at-age in the first year, ⁠, are selected as parameters. The remaining stock numbers by age and year are treated as functions of the initial stock size parameters and of the total mortality. The deterministic relationship between stock size, initial stock size and total mortality is: For the deterministic model, the 70 parameters included are: Natural mortality is assumed to be known and constant for all years and age groups. Estimation methods For the stochastic survival model it was only possible to formulate the likelihood for given values of the stock numbers. For the unconditional likelihood, however, it was not possible to analytically derive a closed form expression for the likelihood function and thereby obtain the ML estimates. Thus the parameters have been estimated using MCMC (Gilks et al., 1996) to simulate the (unconditional) likelihood function. For complex models with structural relationships between variables and parameters, such as the stochastic survival model considered, the so-called single component Metropolis–Hastings or Gibbs sampling (Metropolis et al., 1953; Hastings, 1970; Gilks, 1996) is an MCMC method especially suitable for simulating the likelihood function. For each of the parameters the method sequentially simulates a chain of values given the remaining parameters and variables, i.e. a Markov chain. This means that for each step the combined set of parameter values has the simultaneous distribution, corresponding to the normalised likelihood function or the posterior distribution of the parameters, as a stationary distribution. For each parameter the mean of these simulated values is used as an estimate. For the deterministic model the calculation of the likelihood function is possible and straightforward. However, for this model the MCMC was also used to estimate parameters since it is preferable to use the same type of estimator when comparing different models. The quantiles of the simulated distribution will be used as confidence intervals for the estimated parameters. For the deterministic survival model this is supported by simulations showing that these quantiles are in fact reasonable estimates for the confidence limits and are actually better estimates than the usual inverse Hessian matrix used for the ML estimates (Nielsen and Lewy, 2001). The method is a strong tool for simulating the likelihood especially in cases where it is not possible to analytically derive the likelihood function. The difference between the MLE and this estimator lies in the MLE being the maximum of the (normalised) likelihood function while the new estimator being the mean. When implementing the parameter estimation method it has been necessary to restrict parameter space to finite intervals. The limits have been chosen such that the intervals are sufficiently wide not to affect the sampled parameter distribution. For the stochastic survival model the restricted intervals are as follows: Fres,y, y > 1, Fres,a and qf,a should lie in the interval (0–10), qs,a, σres2, σf2, σs2, σinitial2, σrecruit2 and σsurvival2 in (0–2), ln(μstart) and α in (0–100) and β in (0–1). Due to numerical calculations, effort data have been normalised around the mean over time to avoid excessively small values for the catchability parameters (the parameters were found to lie between 0.01 and 1). For the deterministic model the chosen interval for initial stock size, NminA,y and Na,1, a>minA, is (0–1010). The MCMC estimates obtained may also be interpreted as Bayesian mean posterior estimates considering the uniform distributions over the restricted intervals as prior distributions of the parameters. The confidence intervals may correspondingly be treated as credibility intervals. A basic problem with MCMC is that one has to determine a simulation step for which the simulated chain is effectively at equilibrium. The values simulated before this step, the so-called “burn-in” period, then have to be discarded. The convergence of the chains was examined partly by visual inspection and partly by formal procedures. For the stochastic model the Gelman–Rubin convergence diagnostic (Gelman and Rubin, 1992) was computed in order to determine the “burn-in” steps. The idea of the diagnostic is to simulate a number of chains started on over dispersed values and see when they become indistinguishable. Formally, the convergence is evaluated by considering the ratio of the between and within sequence variances of the different chains. This ratio converges to one when the number of steps tends to infinity. Gelman and Rubin suggest that the burn-in period ends when the diagnostic is less than 1.1 or 1.2. Simulation experiments Simulation experiments were carried out to investigate the properties of the estimators. This was done by simulating the catch observations based on the models described and using fixed known parameters obtained from parameter estimation. The catch observations were simulated in the following way: The model used was the same as described by Equations (1)–(7). The parameters, Θ, used were the values estimated applying data described in the next section. Fres,a,y and Ff,a,y were calculated, the latter using effort data and catchabilities, qf,a. NminA,1 was predicted by randomly drawing from the lognormal distribution, (Equation (6)). Na,1, a=2,…,A were predicted by randomly drawing from the lognormal distribution, (Equation (7)). SSB1was calculated. For y=2 recruitment NminA,y was randomly drawn from Equation (5). For a=2,…,A Na,y was randomly drawn from Equation (1a) and SSBy calculated. Steps 7 and 8 were repeated as long as y<Y. The catch observations, Cres,a,y, Cf,a,y and Is,a,y, were generated from the lognormal distributions (Equations (2)–(4)). One hundred replications with 40 000 chains were generated and for each set of observations the parameters were estimated and stock numbers and SSB were predicted. The empirical mean and variance were calculated for the parameters and compared to the true values by calculating the relative bias, [(estimated−true)/true]×100. The relative bias was also calculated for predicted SSB and recruitment. As these quantities differ for each replication the relative bias was calculated separately for each. Finally, the mean and variance of the relative bias were calculated. Materials and software used The models have been applied to a set of data for North Sea plaice for the period 1988–1997. These data consist of: Catch-at-age and effort data for the Dutch (ages 2–9, years 1989–1997) and English (ages 4–10, years 1988–1997) commercial beam trawl fleets. Catch-at-age data for the combined fleet without effort data (the residual fleet) (age groups 1–12, years 1988–1997). Survey indices for the Dutch beam trawl (ages 1–7, years 1988–1997) and the Sole Net Survey (ages 1–3, years 1988–1997) These data and the mean weight-at-age used are the same as used by the ICES Assessment Working Group (ICES, 1999) except that the working group includes age groups 1–14 in their XSA analysis. The oldest age group considered here, age group 12, is treated as a real age group and not as a plus-group. However, as the catches of age groups 13 and 14 were very small the results should not be affected significantly. The XSA results presented in this paper are working group estimates (ICES, 1999). Estimates of the sampling error for plaice in the North Sea were provided by the EU project, EMAS (EMAS, 2001). Average CVs by age group for the Dutch, English and Danish fleets in the period 1991–1998 were calculated by bootstrapping the samples (Tables 3.16, 3.28 and 3.45 of EMAS, 2001) and were used as estimates of σsampling,Netherland2, σsampling,England2 and σsampling,res2, respectively. The Danish figures were used as estimates of σsampling,res2, because the Danish catch constitutes the largest national proportion of the residual catch (about 40% in 1997). The software package, WinBUGS 1.4 (Bayesian inference using Gibbs sampling, Spiegelhalter et al., 2000) was used to simulate the posterior distributions of the parameters. The package is designed to sample from complex models with structural relationships between variables and parameters. The package applies a single component Metropolis–Hastings algorithm to simulate the full conditional distributions (Gilks et al., 1996). An older version of the program has been described in detail for a biomass dynamics model (Meyer and Millar, 1999b). Results For each of the two stochastic models (1) and (1a) two independent chains with 50 000 steps were generated each with over-dispersed starting points. For each of the years 1988–1997, the Gelman–Rubin diagnostics for the spawning biomass chains for the first 5000 steps were found to lie between 1.03 and 1.2. Based on this and on visual inspection of both plots of parameters versus step number and of parameter distribution, each of these initial 5000 steps was discarded as “burn-in” period and only the remaining 90 000 steps used in the results. Comparisons of the two models showed that the estimated parameters were almost identical indicating that the truncation made in model (1a) had no practical influence on the results. For the deterministic model 11 000 steps were generated. Based on visual inspections of the same type of plots, the first 1000 steps were discarded as a “burn in” period. The results were also compared with other runs with up to 50 000 steps, which gave almost identical results. The fit of the model was examined by inspection of standardised residual plots (Figure 1). No systematic deviations from the model were found. However, the plots should be interpreted with caution, as log catches were only assumed normal for given stock numbers implying that residuals depending on the same stock number were dependent variables. The assumption of lognormal distribution was investigated by comparing the histogram of residuals with the density of the fitted normal distribution (Figure 2A) and by a Q–Q plot (Figure 2B). No serious indications of deviations from the normal distribution were found. Even though the Q–Q plot showed that residuals were a bit too “heavy tailed”, this may, however, be due to residual dependencies. Figure 1 Open in new tabDownload slide Standardised residuals of catch and survey observations from the stochastic model plotted against predicted value (A), fleet (B), age (C) and year (D). Figure 1 Open in new tabDownload slide Standardised residuals of catch and survey observations from the stochastic model plotted against predicted value (A), fleet (B), age (C) and year (D). Figure 2 Open in new tabDownload slide Histogram (A) and Q–Q plot (B) of standardised residuals from the stochastic model. Figure 2 Open in new tabDownload slide Histogram (A) and Q–Q plot (B) of standardised residuals from the stochastic model. For all but one year the stochastic survival method provides lower SSB and higher estimates than the deterministic method and XSA, which are quite similar (Figure 3). Although it is not possible to test whether these differences are significant, the SSB1997 from the stochastic approach seems to be a more conservative estimate, as it appears to be about 19% lower than that from the deterministic approach. Correspondingly, for 1997 the from the stochastic approach is about 15% higher than the deterministic one. The recruitment estimates indicate that there are substantial differences for the last two years, for which the XSA estimates are larger than the estimates for the two other methods. Figure 3 Open in new tabDownload slide Spawning biomass, recruitment and average fishing mortality estimated by the stochastic and deterministic survival models and XSA. Figure 3 Open in new tabDownload slide Spawning biomass, recruitment and average fishing mortality estimated by the stochastic and deterministic survival models and XSA. The importance of the process variances, σprocess,Netherland2, σprocess,England2 and σprocess,res2 illustrated by the proportions etc., indicates that in the case of North Sea plaice process error is by far the most dominant (Table 1). These results might also imply that the model is inconsistent for Danish catches of age groups 1, 11 and 12, for which the sampling variance is greater than total variance estimated. Table 1 Proportion of process variance to total variance for three national North Sea plaice fisheries (per cent). Age . Netherland . England . Denmark . 1 – – a 2 49 – 93 3 87 – 98 4 87 98 98 5 75 97 97 6 67 97 96 7 49 96 93 8 26 94 86 9 14 93 74 10 – 89 55 11 – – a 12 – – a Average 57 95 69b Age . Netherland . England . Denmark . 1 – – a 2 49 – 93 3 87 – 98 4 87 98 98 5 75 97 97 6 67 97 96 7 49 96 93 8 26 94 86 9 14 93 74 10 – 89 55 11 – – a 12 – – a Average 57 95 69b a Negative value. b Average 2–10. Open in new tab Table 1 Proportion of process variance to total variance for three national North Sea plaice fisheries (per cent). Age . Netherland . England . Denmark . 1 – – a 2 49 – 93 3 87 – 98 4 87 98 98 5 75 97 97 6 67 97 96 7 49 96 93 8 26 94 86 9 14 93 74 10 – 89 55 11 – – a 12 – – a Average 57 95 69b Age . Netherland . England . Denmark . 1 – – a 2 49 – 93 3 87 – 98 4 87 98 98 5 75 97 97 6 67 97 96 7 49 96 93 8 26 94 86 9 14 93 74 10 – 89 55 11 – – a 12 – – a Average 57 95 69b a Negative value. b Average 2–10. Open in new tab The CVs connected to the survival and recruitment processes, initial stock size and catch observations by fleet are given in Table 2. A comparison of the CVs of catch observations from the two models, for the two most important fleets (the Dutch and the residual), indicates that CVs for the stochastic model are lower than for the deterministic even though it is not possible to test the significance of the difference. Furthermore, the stochastic model apparently also provides more precise estimates of SSB and fishing mortality than the deterministic one, for all years (Figure 4). As the stochastic model has fewer parameters (55) than the deterministic model (70) one can conclude that the fit of the stochastic model is at least as good as that of the deterministic and apparently provides more precise estimates with fewer parameters. Figure 4 Open in new tabDownload slide Coefficient of variation of spawning biomass and average fishing mortality estimated by the stochastic and deterministic survival models. Figure 4 Open in new tabDownload slide Coefficient of variation of spawning biomass and average fishing mortality estimated by the stochastic and deterministic survival models. Table 2 Estimated CV of the survival and recruitment processes, initial stock and catch-at-age observations for the stochastic and deterministic survival models. . . Stochastic survival . Deterministic survival . Survival process CVsurvive 0.12 – Recruitment process CVrecruit 0.35 – Initial stock size CVinitial 0.68 – Dutch beam trawl CVf=1 0.14 0.19 Residual fleet CVres 0.30 0.34 English beam trawl CVf=2 0.38 0.37 Beam Trawl Survey CVs=1 0.51 0.54 Sole Net Survey CVs=2 0.67 0.65 . . Stochastic survival . Deterministic survival . Survival process CVsurvive 0.12 – Recruitment process CVrecruit 0.35 – Initial stock size CVinitial 0.68 – Dutch beam trawl CVf=1 0.14 0.19 Residual fleet CVres 0.30 0.34 English beam trawl CVf=2 0.38 0.37 Beam Trawl Survey CVs=1 0.51 0.54 Sole Net Survey CVs=2 0.67 0.65 ⁠. Open in new tab Table 2 Estimated CV of the survival and recruitment processes, initial stock and catch-at-age observations for the stochastic and deterministic survival models. . . Stochastic survival . Deterministic survival . Survival process CVsurvive 0.12 – Recruitment process CVrecruit 0.35 – Initial stock size CVinitial 0.68 – Dutch beam trawl CVf=1 0.14 0.19 Residual fleet CVres 0.30 0.34 English beam trawl CVf=2 0.38 0.37 Beam Trawl Survey CVs=1 0.51 0.54 Sole Net Survey CVs=2 0.67 0.65 . . Stochastic survival . Deterministic survival . Survival process CVsurvive 0.12 – Recruitment process CVrecruit 0.35 – Initial stock size CVinitial 0.68 – Dutch beam trawl CVf=1 0.14 0.19 Residual fleet CVres 0.30 0.34 English beam trawl CVf=2 0.38 0.37 Beam Trawl Survey CVs=1 0.51 0.54 Sole Net Survey CVs=2 0.67 0.65 ⁠. Open in new tab Figure 4 also shows that CVs are larger for the first and last years, especially for the average F, but also for SSB. For the first year this may mainly be due to catch and effort data for the Dutch beam trawl fleet not being available. For the last year(s) the large CVs may be due to few age groups being included in the cohorts in question. The survival process has a rather low CV of 0.12 compared to a recruitment CV of 0.35 (Table 2) indicating that the survival process model provides a better fit to generated stock numbers compared to the ability of the Ricker stock-recruitment process to model recruitment. This is not surprising, as no well-defined stock-recruitment relationship exist for North Sea plaice. Our results also indicate that the CVs associated with the Dutch beam trawlers and the residual fleet (accounting for the main parts of the catch) were lower than the CVs of the English beam trawlers and especially lower than the two surveys (Table 2). Hence, the exclusion of these surveys from the analysis would probably only result in insignificant changes. Results of simulation experiments The MCMC estimator of the SSB slightly underestimates the true SSB by about 8% while the estimators of overestimate the true values by about 15% (Table 3). Recruitment estimates and standard deviations of log catches for all fleets, σres, σf and σs, are almost unbiased while the survival process and initial standard deviations are slightly overestimated (16–18%). Stock-recruitment parameters, however, are significantly overestimated (α: 339%, β: 101% and σrecruit: 63%). The correlation between α and β is 0.91. For all parameters it can be shown that the true values lie within the 95% intervals of the empirical distributions. Table 3 Relative bias, [(estimate−true)/true]×100, of simulated stock parameters. . 1988 . 1989 . 1990 . 1991 . 1992 . 1993 . 1994 . 1995 . 1996 . 1997 . Average . SSB −5.8 −7.2 −8.2 −8.8 −8.4 −8.2 −7.2 −7.7 −8.3 −9.3 −7.9 F̄2–10 13.2 15.3 15.4 14.4 15.2 14.2 14.1 13.9 14.4 18.2 14.8 Recruitment −2.5 −1.0 −2.7 −3.1 −3.0 −2.0 −2.3 −1.5 −5.6 −4.3 −2.8 σres: 2.9 σf=1: 2.5 σf=2: 3.0 σs=1: 4.8 σs=2: 7.1 σsurvive: 16.5 σinitial: 18.6 . 1988 . 1989 . 1990 . 1991 . 1992 . 1993 . 1994 . 1995 . 1996 . 1997 . Average . SSB −5.8 −7.2 −8.2 −8.8 −8.4 −8.2 −7.2 −7.7 −8.3 −9.3 −7.9 F̄2–10 13.2 15.3 15.4 14.4 15.2 14.2 14.1 13.9 14.4 18.2 14.8 Recruitment −2.5 −1.0 −2.7 −3.1 −3.0 −2.0 −2.3 −1.5 −5.6 −4.3 −2.8 σres: 2.9 σf=1: 2.5 σf=2: 3.0 σs=1: 4.8 σs=2: 7.1 σsurvive: 16.5 σinitial: 18.6 Open in new tab Table 3 Relative bias, [(estimate−true)/true]×100, of simulated stock parameters. . 1988 . 1989 . 1990 . 1991 . 1992 . 1993 . 1994 . 1995 . 1996 . 1997 . Average . SSB −5.8 −7.2 −8.2 −8.8 −8.4 −8.2 −7.2 −7.7 −8.3 −9.3 −7.9 F̄2–10 13.2 15.3 15.4 14.4 15.2 14.2 14.1 13.9 14.4 18.2 14.8 Recruitment −2.5 −1.0 −2.7 −3.1 −3.0 −2.0 −2.3 −1.5 −5.6 −4.3 −2.8 σres: 2.9 σf=1: 2.5 σf=2: 3.0 σs=1: 4.8 σs=2: 7.1 σsurvive: 16.5 σinitial: 18.6 . 1988 . 1989 . 1990 . 1991 . 1992 . 1993 . 1994 . 1995 . 1996 . 1997 . Average . SSB −5.8 −7.2 −8.2 −8.8 −8.4 −8.2 −7.2 −7.7 −8.3 −9.3 −7.9 F̄2–10 13.2 15.3 15.4 14.4 15.2 14.2 14.1 13.9 14.4 18.2 14.8 Recruitment −2.5 −1.0 −2.7 −3.1 −3.0 −2.0 −2.3 −1.5 −5.6 −4.3 −2.8 σres: 2.9 σf=1: 2.5 σf=2: 3.0 σs=1: 4.8 σs=2: 7.1 σsurvive: 16.5 σinitial: 18.6 Open in new tab Discussion Our model of fish stock assessment which includes stochastic survival and recruitment, seemed to be more flexible than the deterministic separable model as the fit of the former is at least as good as that of the latter and appears to provide more precise estimates of biomass and mortality rates with fewer parameters. Furthermore, the biomass estimates of the stochastic model seem to be more conservative compared to those of the deterministic model, which is in agreement with retrospective analyses of the spawning stock biomass (ICES, 1999), indicating that the SSB in the last year was overestimated. The stochastic survival model estimates of SSB and recruitment were found to be lower than those of the deterministic model, and the opposite was the case for average F. Theoretically, this could be explained in that the stochastic model—contrary to deterministic model—includes a stock-recruitment model and that the average stock numbers in the first year are assumed to be in equilibrium. To investigate this, a third model was considered, which corresponds to the deterministic model but including the same stock-recruitment relationship and stock restrictions in the first year as the stochastic model. Again, the parameters for third model were estimated using WinBUGS and MCMC. Both recruitment estimates and stock size in the first year were found to be close to the estimates of the deterministic model. This implied that the differences between the results from the stochastic and deterministic survival models were caused by the survival model and not by initial stock constrains. Simulation experiments showed that apart from the stock-recruitment parameters, the MCMC estimates of the remaining parameters—including spawning biomass and average fishing mortality—were only slightly biased. The recruitment parameters, α, β and σrecruit, were all significantly biased and the estimates of α and β were highly correlated. However, despite being unable to obtain reliable recruitment parameter estimates the resulting recruitment estimates obtained using these parameters and the stock-recruitment function were almost unbiased (with a large variation). Errors associated with the catch-at-age by fleet used in stock assessment consist of sampling error and other errors denoted as process errors (including both fishing process error and model error). For the stochastic model, results for North Sea plaice indicate that process error is by far the most important factor while sampling error only plays a minor role. For assessment purposes this implies that even if information were available on the sampling errors of the catches for different fleet components, these should not be used directly to account for different variances associated with the various sources of information. The total variance of the catch by fleet still has to be estimated in the model. However, in cases where the sampling uncertainty is high relative to the process error, the total model variance may be estimated by the sampling variance. It should be noted that the concept of process error always refers to a specified model used. The survival process of fish in a cohort has been assumed to be lognormally distributed. It has been shown that this assumption can be derived from a stochastic reformulation of the standard deterministic differential equations that describe the stock dynamics. For the formulation chosen the log variance of the survival process has been shown simply to be the sum of the process errors due to fishing and natural mortality. The MCMC methodology, in particular the single component Metropolis–Hastings and graphical models, has proven to be a powerful tool for making inference in complex fish stock assessment models including structural relationships between variables and parameters. It is easy to implement such complex model in the WinBUGS program. In both the stochastic and deterministic models the catchability is assumed to be constant over years. A random walk extension of the deterministic model was formulated and implemented for which catchability was allowed to vary over years: qf,a,y|qf,a,y−1∼LN(ln(qf,a,y−1),σ2) where σ2 arbitrarily was fixed at 0.0252. This model was used for the catchability of the Dutch and English commercial beam trawl fleets. Markov chains with length of 60 000 were generated for the parameters and appeared unstable, which could indicate an over-parameterisation of the model. Therefore, if residuals for a fleet indicate a trend over time, as is the case for English beam trawlers, it is probably better to change the assumption of constant catchability to a parametric catchability model allowing for temporal trends. The present work is part of a project funded by the Danish Ministry of Food, Agriculture and Fisheries entitled “Development of improved models of fisheries impact on marine fish stocks and ecosystem”. We further thank Lasse Heje Pedersen, Stern School of Business, New York University, Søren Feodor Nielsen, the Department of Statistics and Operations Research, University of Copenhagen, for giving inspiration to develop the stochastic survival model, and Uffe Høgsbro Thygesen, DIFRES, for valuable discussions. Finally, we are grateful to two anonymous referees whose comments and suggestions considerably improved the manuscript. Appendix A Fish stock dynamics are usually based on the standard differential equations: ⁠, and ⁠, where t is the time, Nt is the number of fish in the sea, is the accumulated number of fish caught to time t, is the cumulative number of fish dead from natural causes at time t, Ft is fishing and Mt natural mortality. These deterministic models can be extended to be stochastic differential equations: (A1) (A2) (A3) where dWtfishing and dWtnatural are normally distributed stochastic process variables with mean 0 and variance dt, where Wtfishing and Wtnatural denote Wiener processes and where σfishing and σnatural are standard deviations associated with the two processes. For a given time, t0, Wt0fishing and Wt0natural are normal distributions, N(0,1). The two death processes are assumed to be mutually independent. The models chosen imply that ⁠, and dNt in a short time interval dt, is normally distributed with the mean equal to the deterministic values and the variances equal to σfishing2Nt2dt, σnatural2Nt2dt and σsurvival2Nt2dt, respectively, where σsurvival2=σfishing2+σnatural2. Using Ito's formula (see for example Gard, 1988) one can show that the following equation for given initial stock, N0, is the solution to the stochastic differential Equation (A3): where is a Wiener process including the processes due to both fishing and natural mortality. Assuming that Zt is constant over a time step, for instance a year, one can show that this corresponds to the number of survivors at the end of the year given the number at the beginning of the year following a lognormal distribution: (A4) For this distribution the mean is the usual deterministic value, E(Na+1,y+1 | Na,y) = Na,y exp(−Za,y) and variance σsurvival2. It is not possible correspondingly to derive the probability distributions for and (or the observed catches during a period, ⁠). However, it can be shown that the expected value of the catches during a period equals the standard value of ⁠. The coefficient of variation of the catches can also be derived analytically as a rather complicated function of F, M and σfishing and σnatural. It should be noted that only the variance associated with the survival, has been estimated in the model. In principle the variances, σfishing2and σnatural2, could be estimated separately as well as by using the complicated relationship mentioned and the equation, ⁠. When estimating parameters using Gibbs sampling it is assumed that catch observations conditioned the survivors and the mortality parameters are independent. It is not evident if this assumption is fulfilled for the model based on Equations (A1)–(A3). The WinBUGS code used in this study is available from the authors. References Anon Guidelines on the precautionary approach to capture fisheries and species introduction , 1995 FAO Fisheries Technical Paper No. 350 (1): 52 pp Deriso R.B. , Quinn P.R. Jr. , Neal T.J. . Catch-at-age analysis with auxiliary information , Canadian Journal of Fisheries and Aquatic Sciences , 1985 , vol. 42 (pg. 815 - 824 ) Google Scholar Crossref Search ADS WorldCat Doubleday W.G. . A least square approach to analysing catch at age data , International North Pacific Fisheries Commission Bulletin , 1976 , vol. 12 (pg. 69 - 81 ) OpenURL Placeholder Text WorldCat EMAS Evaluation of market sampling strategies for a number of commercially exploited stocks in the North Sea and development of procedures for consistent data storage and retrieval , 2001 Final Report to the EU Commission, CFP Study Project 98/075. European Commission, Directorate-General XIV Fisheries, Rue de la Loi 200, 1049 Brussels, Belgium Fournier D. , Archibald C.P. . A general theory for analyzing catch at age data , Canadian Journal of Fisheries and Aquatic Sciences , 1982 , vol. 39 (pg. 1195 - 1207 ) Google Scholar Crossref Search ADS WorldCat Gard T.C. . Introduction to Stochastic Differential Equations , 1988 New York, NY Marcel Dekker 234 pp Gavaris S. . An adaptive framework for estimation of population size , 1988 Canadian Atlantic Fisheries Scientific Advisory Commission, 88/29. 12 pp Gelman A. , Rubin D.B. . Inference from iterative simulation using multiple sequences (with discussion) , Statistical Science , 1992 , vol. 7 (pg. 457 - 511 ) Google Scholar Crossref Search ADS WorldCat Gilks W.R. . Gilks W.R. , Richardson S. , Spiegelhalter D.J. . Full conditional distributions , Markov Chain Monte Carlo in Practise , 1996 London, UK Chapman & Hall (pg. 75 - 86 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gilks W.R. , Richardson S. , Spiegelhalter D.J. . , Markov chain Monte Carlo in practise , 1996 London, UK Chapman & Hall 486 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gudmundsson G. . Time series analysis of catch-at-age observations , Applied Statistics , 1994 , vol. 43 (pg. 1117 - 1126 ) Google Scholar Crossref Search ADS WorldCat Hastings W.K. . Monte Carlo sampling methods using Markov chains and their applications , Biometrica , 1970 , vol. 57 (pg. 97 - 109 ) Google Scholar Crossref Search ADS WorldCat Ianelli J. N. , Fournier D. A. . Alternative age-structured analyses of the NRC simulated stock assessment data , Analyses of simulated data sets in support of the NRC study on stock assessment methods , 1998 (pg. 81 - 96 ) NOAA technical Memorandum, NMFS-F/SPO-30, U.S. Departments of Commerce Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ICES Report of the working group on the assessment of demersal stocks in the North Sea and Skagerrak , 1999 ICES CM 1999/ACFM: 8. 676 pp Kinas P.G. . Bayesian fishery stock assessment and decision making using adaptive importance sampling , Canadian Journal of Fisheries and Aquatic Sciences , 1996 , vol. 53 (pg. 414 - 423 ) Google Scholar Crossref Search ADS WorldCat Lewy P. . Integrated stochastic virtual population analysis: estimates and their precision of fishing mortalities and stock sizes for the North Sea whiting stock , Journal du Conseil International pour l'Exploration de la Mer , 1988 , vol. 44 (pg. 217 - 228 ) Google Scholar Crossref Search ADS WorldCat McAllister M.K. , Kirkwood G.P. . Using Bayesian decision analysis to help achieve a precautionary approach for managing developing fisheries , Canadian Journal of Fisheries and Aquatic Sciences , 1998 , vol. 55 (pg. 2642 - 2661 ) Google Scholar Crossref Search ADS WorldCat McAllister M.K. , Pikitch E.K. , Punt A.E. , Hilborn R. . A Bayesian approach to stock assessment and harvest decisions using the sampling/importance resampling algorithm , Canadian Journal of Fisheries and Aquatic Sciences , 1994 , vol. 51 (pg. 2673 - 2687 ) Google Scholar Crossref Search ADS WorldCat Methot R.D. . Synthesis model: an adaptable framework for analysis of diverse stock assessment data , International North Pacific Fisheries Commission Bulletin , 1990 , vol. 50 (pg. 259 - 277 ) OpenURL Placeholder Text WorldCat Metropolis N. , Rosenbluth A.V. , Rosenbluth M.N. , Teller A.H. , Teller E. . Equations of state calculations by fast computing machine , Journal of Chemical Physics , 1953 , vol. 21 (pg. 1087 - 1091 ) Google Scholar Crossref Search ADS WorldCat Meyer R. , Millar R.B. . Bayesian stock assessment using a state space implementation of the delay difference model , Canadian Journal of Fisheries and Aquatic Sciences , 1999 , vol. 56 (pg. 37 - 52 ) OpenURL Placeholder Text WorldCat Meyer R. , Millar R.B. . BUGS in Bayesian stock assessments , Canadian Journal of Fisheries and Aquatic Sciences , 1999 , vol. 56 (pg. 1078 - 1086 ) Google Scholar Crossref Search ADS WorldCat Millar R.B. , Meyer R. . Non-linear state space modelling of fisheries biomass dynamics by using Metropolis–Hastings within-Gibbs sampling , Applied Statistics , 2000 , vol. 49 Part 3 (pg. 327 - 342 ) OpenURL Placeholder Text WorldCat Nielsen A. , Lewy P. . Comparison of the frequentist properties of the Bayes and maximum likelihood estimators in a fish stock assessment model , Canadian Journal of Fisheries and Aquatic Sciences , 2001 , vol. 59 (pg. 136 - 143 ) Google Scholar Crossref Search ADS WorldCat Patterson K.R. . Evaluating uncertainty in harvest control law catches using Bayesian Markov chain Monte Carlo virtual population analysis with adaptive rejection sampling and including structural uncertainty , Canadian Journal of Fisheries and Aquatic Sciences , 1999 , vol. 56 (pg. 208 - 221 ) Google Scholar Crossref Search ADS WorldCat Patterson K. , Melvin G.D. . Integrated catch age analysis, version1.2 , Scottish Fisheries Research Report , 1996 , vol. 58 (pg. 1 - 60 ) OpenURL Placeholder Text WorldCat Powers J.E. , Restrepo V.R. . Evaluation of stock assessment research for Gulf of Mexico King Mackerel: benefits and costs to management , North American Journal of Fisheries Management , 1993 , vol. 13 (pg. 15 - 26 ) Google Scholar Crossref Search ADS WorldCat Schnute J.T. . A General framework for developing sequential fisheries models , Canadian Journal of Fisheries and Aquatic Sciences , 1994 , vol. 51 (pg. 1676 - 1688 ) Google Scholar Crossref Search ADS WorldCat Schnute J.T. , Richards L.J. . The influence of error on population estimates from catch-at-age models , Canadian Journal of Fisheries and Aquatic Sciences , 1995 , vol. 52 (pg. 2063 - 2077 ) Google Scholar Crossref Search ADS WorldCat Shepherd J.G. . Extended survivors analysis: an improved method for the analysis of catch-at-age data and abundance indices , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 584 - 591 ) Google Scholar Crossref Search ADS WorldCat Spiegelhalter D.J. , Thomas A. , Best N.G. . , WinBUGS Version 1.3 User Manual , 2000 Cambridge, UK Medical Research Council Biostatistics Unit, Institute of Public Health http://www.mrc-bsu.cam.ac.uk/bugs Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sullivan P.J. . A Kalman filter approach to catch-at length analysis , Biometrics , 1992 , vol. 48 (pg. 237 - 257 ) Google Scholar Crossref Search ADS WorldCat Virtala M. , Kuikka S. , Arjas E. . Stochastic virtual population analysis , ICES Journal of Marine Science , 1998 , vol. 55 (pg. 892 - 904 ) Google Scholar Crossref Search ADS WorldCat © 2003 International Council for the Exploration of the Sea

Massutí,, Enric;Moranta,, Joan

doi: 10.1016/S1054-3139(03)00089-4pmid: N/A

Abstract The analysis of 131 hauls from four bottom trawl fishing surveys carried out between depths of 46 and 1713 m in two different areas off the Balearic Islands yielded a total of 23 elasmobranch species belonging to eight families. Cluster analysis and multidimensional scaling (MDS) ordination were applied to detect zonation patterns and some ecological parameters (e.g. species richness, abundance and biomass, mean weight, diversity and evenness) were calculated for each assemblage. For each area, analysis of similitude (ANOSIM) and similarity percentage analysis (SIMPER) were also applied to detect differences between seasons and depths. For the most important species (Galeus melastomus, Scyliorhinus canicula, Centroscymnus coelolepis, Etmopterus spinax, Squalus blainvillei, Raja naevus, Raja asterias, Raja clavata, Raja miraletus and Raja oxyrhinchus), abundance and size distributions were analysed by depth. Introduction There have been numerous descriptions of the demersal fish assemblages in the Mediterranean Sea (e.g. Stefanescu et al., 1992; Papaconstantinou et al., 1994; Matarrese et al., 1996; among others). However, these studies include both selachians and teleosts. The only papers related exclusively to demersal elasmobranch assemblages are those by Capapé et al. (2000) in the Gulf of Lions, Relini et al. (2000) in Italian waters and Bertrand et al. (2000) in the northern Mediterranean. The latest study covered the whole area, from the Strait of Gibraltar to the Aegean Sea, but did not include the Balearic Islands. Papers on Mediterranean elasmobranchs are more numerous, but they are focused on the distribution and biology of certain species. In the Mediterranean, there has been an increasing international concern about changes in the abundance and diversity of elasmobranchs. There is increasing evidence that fishing exploitation affects their composition and biodiversity to a greater extent than most teleosts (Stevens et al., 2000). This applies to the Mediterranean Sea, in which there is a high level of exploitation over the continental shelf and upper slope down to a depth of 800 m. Evidence of changes in the number of elasmobranchs and decreases in the abundance and biomass of some species (e.g. Raja clavata) throughout the last decade have been reported for the highly exploited area Gulf of Lions (Aldebert, 1997; Bertrand et al., 1998). Elasmobranchs are widespread, although not too specious, resulting in an interesting group for biodiversity process studies. This paper characterises the assemblages of demersal elasmobranch on the bottom trawl fishing grounds along the continental shelf and upper slope, and in unexploited deeper areas of the middle and lower slope, off the Balearic Islands. Experimental trawl surveys are analysed for the main species in terms of species composition, community structure and distribution and population structure. Our aim was to provide information relating to the diversity and abundance of elasmobranches, which could serve as a reference for the monitoring of future trends in the same area and would allow comparison with other Mediterranean areas. Materials and methods Data were collected from 131 hauls made during four bottom trawl surveys off the Balearic Islands (Figure 1). Surveys were carried out in two different seasons (spring and autumn) and two different areas: around Mallorca and Menorca (northern area), and south of Eivissa and Formentera (southern area). Figure 1 Open in new tabDownload slide Location of the two areas studied off the Balearic Islands (south and north; showing the 50, 200, 600, 1000 and 2000 m isobaths) and the trawl stations surveyed during the different surveys: BALARs (♦) and QUIMERAs (●). Figure 1 Open in new tabDownload slide Location of the two areas studied off the Balearic Islands (south and north; showing the 50, 200, 600, 1000 and 2000 m isobaths) and the trawl stations surveyed during the different surveys: BALARs (♦) and QUIMERAs (●). Hauls in the northern area were made between 40 and 800 m depth during the BALAR cruises, on board the R/V “Francisco de Paula Navarro” (length: 30 m; engine power: 1100 hp) in April 2001 (41 hauls) and September–October 2001 (44 hauls). Hauls in the southern area were made between 200 and 1800 m depth during the QUIMERA cruises, on board the R/V “García del Cid” (length: 37 m; engine power: 1500 hp) in October 1996 (32 hauls) and May 1998 (14 hauls). In each haul, fish were sorted and abundance, biomass and length frequency (total length, in cm) of each species determined. Different sampling gears were used in each area. In the northern area, a GOC73 trawl towed by two warps at 2.8 knots was used. This gear has been used since 1994 by most surveys carried out in the Mediterranean Sea (Abelló et al., 2002). In the south area, an OTMS-27.5 benthic trawl was towed by a single warp at 2.5 knots (Sardà et al., 1998). In both cases, the arrival and departure of the net at the bottom, as well as its horizontal and vertical openings (on average, 16.4–2.8 m for the GOC73 and 14.0–1.9 m for the OTMS-27.5) were measured using a SCANMAR system. The position at the start and the end of each trawl was recorded using Global Position System (GPS). Using this information, catch data was standardised to a common sampled area of 10 000 m2. Trawl selectivity is mainly dependent on mouth area, mesh size, towing speed, power of the vessel and whether the net is towed on one warp or two (e.g. Merrett et al., 1991; Gordon et al., 1996). To avoid possible differences, no comparisons were made between areas. For each area, data on standardised abundance, biomass and mean fish weight were plotted over a depth axis to display trends with depth. The PRIMER package was used to analyse the abundance and biomass matrices of species composition (Clarke and Gorley, 2001). To identify assemblages, cluster analysis and multidimensional scaling (MDS) were applied after square root transformation. The Bray–Curtis index was chosen as the similarity coefficient and the UPGMA was applied to link samples into clusters. Samples in which only one species was caught (36), and species recorded in less than 5% of samples (13) were omitted from the analysis, since it was statistically more informative than when all samples and species were included. Analysis of similitude (ANOSIM) and similarity percentage analysis (SIMPER) were also applied to detect differences between seasons and depths. The ecological parameters, species richness and mean species richness, total abundance and biomass, mean fish weight and the Shannon–Wiener diversity index (H′) and evenness (J′) were also calculated in each group resulting from the cluster analysis. To show the bathymetric distribution of the main species along the whole depth range surveyed, the standardised mean abundance (fish per 10 000 m2) was calculated in each area at 10 established depth intervals. The overall length frequency distribution by sex was also calculated for these species. For the most abundant, the length frequency distribution was calculated by each of the depth intervals considered. Results A total of 6402 specimens (789 kg of biomass) belonging to 23 elasmobranch demersal species and eight families were collected from 131 bottom trawls carried out between 40 and 1800 m depth in two different areas off the Balearic Islands (Table 1). In the northern area (40–800 m depth-strata) 22 species (5379 specimens; 630 kg) were caught, while in the south (200–1800 m depth-strata) 10 species (1023 specimens; 159 kg) were caught. In both areas, the most abundant species were Galeus melastomus and Scyliorhinus canicula. Other important species in the overall assemblage were the sharks Etmopterus spinax, Squalus blainvillei and Centroscymnus coelolepis. Rays were captured almost exclusively in the northern area, with Raja miraletus, Raja clavata, Raja asterias, Raja naevus and Raja oxyrhinchus being the most important species. The remaining species were captured occasionally over the whole area. Table 1 Elasmobranch species caught between depths of 40 and 1800 m during BALAR and QUIMERA trawl surveys carried out in two different areas off the Balearic Islands. Total abundance (A; in number of specimens) and biomass (B; in kg), frequency of occurrence (F) and depth range (D; in metres) are shown by species for each surveyed area. . . Northern area . Southern area . Family . Species . A . B . F . D . A . B . F . D . Scyliorhinidae Galeus melastomus Rafinesque, 1810 2471 135.43 38 101–745 563 88.49 73 239–1713 Scyliorhinus canicula Linnaeus, 1758 2440 261.75 67 44–416 305 14.56 18 195–402 Triakidae Mustelus asterias Cloquet, 1821 1 0.17 1 103 – – – – Mustelus mustelus Linnaeus, 1758 1 0.76 1 68 – – – – Squalidae Centrophorus uyato Rafinesque, 1810 1 3.96 1 686 1 4 2 802 Centroscymnus coelolepis Bocage and Capello, 1864 – – – – 39 23.22 29 1012–1713 Dalatias licha Bonnaterre, 1788 2 3.73 2 624–698 5 9.01 9 595–892 Etmopterus spinax Linnaeus, 1758 65 5.90 19 616–745 76 13.31 53 311–1416 Squalus blanvillei Risso, 1826 53 39.95 7 103–649 24 2.4 2 241 Torpedinidae Torpedo nobiliana Bonaparte, 1835 1 0.22 1 371 – – – – Torpedo marmorata Risso, 1810 5 0.87 6 108–180 – – – – Rajidae Raja oxyrinchus Linnaeus, 1758 23 21.43 11 235–444 – – – – Raja naevus Müller and Henle, 1841 44 15.52 18 52–337 4 1.602 4 908 Raja asterias Delaroche, 1809 42 20.63 20 44–399 5 2.19 4 195–264 Raja brachyura Lafont, 1873 1 0.67 1 70 – – – – Raja clavata Linnaeus, 1758 92 83.26 28 85–400 – – – – Raja miraletus Linnaeus, 1758 112 20.56 32 69–399 – – – – Raja montagui Fowler, 1910 2 1.12 1 77 – – – – Raja polystigma Regan, 1923 7 2.14 4 63–127 1 0.22 2 398 Raja undulata Lacepède, 1802 1 1.40 1 53 – – – – Dasyatidae Dasyatis pastinaca Linnaeus, 1758 3 3.92 4 41–53 – – – – Myliobatidae Myliobatis aquila Linnaeus, 1758 10 10.69 4 41–46 – – – – Chimaeridae Chimaera monstrosa Linnaeus, 1758 2 0.15 2 494–538 – – – – . . Northern area . Southern area . Family . Species . A . B . F . D . A . B . F . D . Scyliorhinidae Galeus melastomus Rafinesque, 1810 2471 135.43 38 101–745 563 88.49 73 239–1713 Scyliorhinus canicula Linnaeus, 1758 2440 261.75 67 44–416 305 14.56 18 195–402 Triakidae Mustelus asterias Cloquet, 1821 1 0.17 1 103 – – – – Mustelus mustelus Linnaeus, 1758 1 0.76 1 68 – – – – Squalidae Centrophorus uyato Rafinesque, 1810 1 3.96 1 686 1 4 2 802 Centroscymnus coelolepis Bocage and Capello, 1864 – – – – 39 23.22 29 1012–1713 Dalatias licha Bonnaterre, 1788 2 3.73 2 624–698 5 9.01 9 595–892 Etmopterus spinax Linnaeus, 1758 65 5.90 19 616–745 76 13.31 53 311–1416 Squalus blanvillei Risso, 1826 53 39.95 7 103–649 24 2.4 2 241 Torpedinidae Torpedo nobiliana Bonaparte, 1835 1 0.22 1 371 – – – – Torpedo marmorata Risso, 1810 5 0.87 6 108–180 – – – – Rajidae Raja oxyrinchus Linnaeus, 1758 23 21.43 11 235–444 – – – – Raja naevus Müller and Henle, 1841 44 15.52 18 52–337 4 1.602 4 908 Raja asterias Delaroche, 1809 42 20.63 20 44–399 5 2.19 4 195–264 Raja brachyura Lafont, 1873 1 0.67 1 70 – – – – Raja clavata Linnaeus, 1758 92 83.26 28 85–400 – – – – Raja miraletus Linnaeus, 1758 112 20.56 32 69–399 – – – – Raja montagui Fowler, 1910 2 1.12 1 77 – – – – Raja polystigma Regan, 1923 7 2.14 4 63–127 1 0.22 2 398 Raja undulata Lacepède, 1802 1 1.40 1 53 – – – – Dasyatidae Dasyatis pastinaca Linnaeus, 1758 3 3.92 4 41–53 – – – – Myliobatidae Myliobatis aquila Linnaeus, 1758 10 10.69 4 41–46 – – – – Chimaeridae Chimaera monstrosa Linnaeus, 1758 2 0.15 2 494–538 – – – – Open in new tab Table 1 Elasmobranch species caught between depths of 40 and 1800 m during BALAR and QUIMERA trawl surveys carried out in two different areas off the Balearic Islands. Total abundance (A; in number of specimens) and biomass (B; in kg), frequency of occurrence (F) and depth range (D; in metres) are shown by species for each surveyed area. . . Northern area . Southern area . Family . Species . A . B . F . D . A . B . F . D . Scyliorhinidae Galeus melastomus Rafinesque, 1810 2471 135.43 38 101–745 563 88.49 73 239–1713 Scyliorhinus canicula Linnaeus, 1758 2440 261.75 67 44–416 305 14.56 18 195–402 Triakidae Mustelus asterias Cloquet, 1821 1 0.17 1 103 – – – – Mustelus mustelus Linnaeus, 1758 1 0.76 1 68 – – – – Squalidae Centrophorus uyato Rafinesque, 1810 1 3.96 1 686 1 4 2 802 Centroscymnus coelolepis Bocage and Capello, 1864 – – – – 39 23.22 29 1012–1713 Dalatias licha Bonnaterre, 1788 2 3.73 2 624–698 5 9.01 9 595–892 Etmopterus spinax Linnaeus, 1758 65 5.90 19 616–745 76 13.31 53 311–1416 Squalus blanvillei Risso, 1826 53 39.95 7 103–649 24 2.4 2 241 Torpedinidae Torpedo nobiliana Bonaparte, 1835 1 0.22 1 371 – – – – Torpedo marmorata Risso, 1810 5 0.87 6 108–180 – – – – Rajidae Raja oxyrinchus Linnaeus, 1758 23 21.43 11 235–444 – – – – Raja naevus Müller and Henle, 1841 44 15.52 18 52–337 4 1.602 4 908 Raja asterias Delaroche, 1809 42 20.63 20 44–399 5 2.19 4 195–264 Raja brachyura Lafont, 1873 1 0.67 1 70 – – – – Raja clavata Linnaeus, 1758 92 83.26 28 85–400 – – – – Raja miraletus Linnaeus, 1758 112 20.56 32 69–399 – – – – Raja montagui Fowler, 1910 2 1.12 1 77 – – – – Raja polystigma Regan, 1923 7 2.14 4 63–127 1 0.22 2 398 Raja undulata Lacepède, 1802 1 1.40 1 53 – – – – Dasyatidae Dasyatis pastinaca Linnaeus, 1758 3 3.92 4 41–53 – – – – Myliobatidae Myliobatis aquila Linnaeus, 1758 10 10.69 4 41–46 – – – – Chimaeridae Chimaera monstrosa Linnaeus, 1758 2 0.15 2 494–538 – – – – . . Northern area . Southern area . Family . Species . A . B . F . D . A . B . F . D . Scyliorhinidae Galeus melastomus Rafinesque, 1810 2471 135.43 38 101–745 563 88.49 73 239–1713 Scyliorhinus canicula Linnaeus, 1758 2440 261.75 67 44–416 305 14.56 18 195–402 Triakidae Mustelus asterias Cloquet, 1821 1 0.17 1 103 – – – – Mustelus mustelus Linnaeus, 1758 1 0.76 1 68 – – – – Squalidae Centrophorus uyato Rafinesque, 1810 1 3.96 1 686 1 4 2 802 Centroscymnus coelolepis Bocage and Capello, 1864 – – – – 39 23.22 29 1012–1713 Dalatias licha Bonnaterre, 1788 2 3.73 2 624–698 5 9.01 9 595–892 Etmopterus spinax Linnaeus, 1758 65 5.90 19 616–745 76 13.31 53 311–1416 Squalus blanvillei Risso, 1826 53 39.95 7 103–649 24 2.4 2 241 Torpedinidae Torpedo nobiliana Bonaparte, 1835 1 0.22 1 371 – – – – Torpedo marmorata Risso, 1810 5 0.87 6 108–180 – – – – Rajidae Raja oxyrinchus Linnaeus, 1758 23 21.43 11 235–444 – – – – Raja naevus Müller and Henle, 1841 44 15.52 18 52–337 4 1.602 4 908 Raja asterias Delaroche, 1809 42 20.63 20 44–399 5 2.19 4 195–264 Raja brachyura Lafont, 1873 1 0.67 1 70 – – – – Raja clavata Linnaeus, 1758 92 83.26 28 85–400 – – – – Raja miraletus Linnaeus, 1758 112 20.56 32 69–399 – – – – Raja montagui Fowler, 1910 2 1.12 1 77 – – – – Raja polystigma Regan, 1923 7 2.14 4 63–127 1 0.22 2 398 Raja undulata Lacepède, 1802 1 1.40 1 53 – – – – Dasyatidae Dasyatis pastinaca Linnaeus, 1758 3 3.92 4 41–53 – – – – Myliobatidae Myliobatis aquila Linnaeus, 1758 10 10.69 4 41–46 – – – – Chimaeridae Chimaera monstrosa Linnaeus, 1758 2 0.15 2 494–538 – – – – Open in new tab In both areas, the bathymetric distribution of standardised abundance and biomass of elasmobranches, as well as mean fish weight, showed similar trends above a depth of 800 m (Figure 2). Abundance reached its maximum between 300 and 400 m depth, whereas the biomass had minimum values around 500 m and mean fish weight reached its minimum between 400 and 500 m. In the southern area, abundance and biomass values showed a decreasing trend below depths of 500 and 800 m, respectively, while mean fish weight remained constant below 800 m. Figure 2 Open in new tabDownload slide Distribution by depth of the standardised abundance (A: fish per 10 000 m2) and biomass (B: g 10 000 m−2) and the mean weight (C: g per fish) of elasmobranchs captured during bottom trawl surveys carried out in two areas off the Balearic Islands. Figure 2 Open in new tabDownload slide Distribution by depth of the standardised abundance (A: fish per 10 000 m2) and biomass (B: g 10 000 m−2) and the mean weight (C: g per fish) of elasmobranchs captured during bottom trawl surveys carried out in two areas off the Balearic Islands. The similarity dendrograms for the trawls revealed the existence of four assemblages, which were confirmed by the MDS analysis (Figures 3 and 4), with the bathymetric gradient being the factor of association, without seasonal differences. In the northern area (Figure 3), the first cluster separated samples taken above a depth of 235 m (SH) from the rest, which were grouped in two depth intervals: 326–632 m (SL1) and 624–745 m (SL2). In the southern area (Figure 4), the first cluster separated samples taken above a depth of 264 m (SH) from the rest, which were grouped in three depth intervals: 335–415 m (SL1), 502–1322 m (SL2) and 1416–1713 m (SL3). Figure 3 Open in new tabDownload slide Dendrogram (A) and MDS ordination (B; indicating the groupings obtained from cluster analysis: ○ SH; □ SL1; ▵ SL2) of elasmobranch samples obtained during BALAR bottom trawl surveys carried out between depths of 40 and 800 m in the northern area off the Balearic Islands. The dendrogram shows the mean depth (in metres) and season (S, spring; A, autumn) for each sample. The groupings obtained from cluster analysis are indicated in MDS by different white (spring) and black (autumn) symbols. Figure 3 Open in new tabDownload slide Dendrogram (A) and MDS ordination (B; indicating the groupings obtained from cluster analysis: ○ SH; □ SL1; ▵ SL2) of elasmobranch samples obtained during BALAR bottom trawl surveys carried out between depths of 40 and 800 m in the northern area off the Balearic Islands. The dendrogram shows the mean depth (in metres) and season (S, spring; A, autumn) for each sample. The groupings obtained from cluster analysis are indicated in MDS by different white (spring) and black (autumn) symbols. Figure 4 Open in new tabDownload slide Dendrogram (A) and MDS ordination (B; indicating the groupings obtained from cluster analysis: ○ SH; ▵ SL1; □ SL2; ☆ SL3) of elasmobranch samples obtained during QUIMERA bottom trawl surveys carried out between depths of 200 and 1800 m in the southern area off the Balearic Islands. The dendrogram shows the mean depth (in metres) and season (S, spring; A, autumn) for each sample. The groupings obtained from cluster analysis are indicated in MDS by different white (spring) and black (autumn) symbols. Figure 4 Open in new tabDownload slide Dendrogram (A) and MDS ordination (B; indicating the groupings obtained from cluster analysis: ○ SH; ▵ SL1; □ SL2; ☆ SL3) of elasmobranch samples obtained during QUIMERA bottom trawl surveys carried out between depths of 200 and 1800 m in the southern area off the Balearic Islands. The dendrogram shows the mean depth (in metres) and season (S, spring; A, autumn) for each sample. The groupings obtained from cluster analysis are indicated in MDS by different white (spring) and black (autumn) symbols. The values of some ecological parameters in the different assemblages by area are given in Table 2. Large differences were obtained in species richness, with highest (17) and lowest (3) values in the SH and SL3 assemblages of the northern and southern areas, respectively. By contrast, mean species richness was similar in all assemblages and ranged between 1.5 in the SL3 of the south area and 2.8 in the SH and SL1 of the northern area. Although different sampling gear was used, mean abundance by assemblage was similar between areas, with maximum values (21–25 fish per 10 000 m2) for the SL1 assemblage. In the northern area, the highest mean biomass was for SH (2.5 kg 10 000 m−2 from GOC73), with a value very different from the rest. In the southern area, mean biomass were also similar among assemblages, except for SL3, which showed the lowest value (0.46 kg 10 000 m−2 from OTMS-27.5). In the northern area, the highest diversity and evenness were obtained for SH, while in SL1 and SL2 these parameters showed similar values. In the southern area, diversity was higher in the SH and SL2 assemblages, while maximum evenness was obtained for the SL2 and SL3 assemblages. Table 2 Number of hauls analysed and mean ecological parameters (standard error) for each group resulting from cluster and MDS analyses (see Figures 3 and 4) of elasmobranch samples obtained during bottom trawl surveys carried out in two areas off the Balearic Islands. . Northern area . Southern area . . SH . SL1 . SL2 . SH . SL1 . SL2 . SL3 . Hauls 52 15 18 3 6 26 10 Number of species 17 10 6 4 4 6 3 Mean species richness 2.8 (0.2) 2.8 (0.4) 2.1 (0.1) 2.3 (0.3) 2.1 (0.2) 2.1 (0.1) 1.5 (0.2) Fish per 10 000 m2 13.1 (1.8) 21.0 (5.6) 3.8 (1.0) 12.4 (1.3) 25.1 (7.2) 2.7 (0.2) 1.0 (0.2) g 10 000 m−2 2564.3 (365.2) 876.1 (217.1) 612.4 (141.6) 945.5 (205.9) 900.2 (242.9) 982.5 (157.4) 461.3 (132.6) Mean fish weight (g) 246.3 (25.7) 45.6 (5.5) 184.0 (38.3) 76.8 (18.4) 36.1 (7.4) 431.9 (92.6) 480.9 (139.1) Diversity (H′) 0.52 (0.06) 0.36 (0.09) 0.36 (0.05) 0.42 (0.23) 0.25 (0.08) 0.46 (0.16) 0.25 (0.10) Evenness (J′) 0.55 (0.04) 0.42 (0.09) 0.51 (0.05) 0.46 (0.18) 0.36 (0.12) 0.70 (0.09) 0.79 (0.02) . Northern area . Southern area . . SH . SL1 . SL2 . SH . SL1 . SL2 . SL3 . Hauls 52 15 18 3 6 26 10 Number of species 17 10 6 4 4 6 3 Mean species richness 2.8 (0.2) 2.8 (0.4) 2.1 (0.1) 2.3 (0.3) 2.1 (0.2) 2.1 (0.1) 1.5 (0.2) Fish per 10 000 m2 13.1 (1.8) 21.0 (5.6) 3.8 (1.0) 12.4 (1.3) 25.1 (7.2) 2.7 (0.2) 1.0 (0.2) g 10 000 m−2 2564.3 (365.2) 876.1 (217.1) 612.4 (141.6) 945.5 (205.9) 900.2 (242.9) 982.5 (157.4) 461.3 (132.6) Mean fish weight (g) 246.3 (25.7) 45.6 (5.5) 184.0 (38.3) 76.8 (18.4) 36.1 (7.4) 431.9 (92.6) 480.9 (139.1) Diversity (H′) 0.52 (0.06) 0.36 (0.09) 0.36 (0.05) 0.42 (0.23) 0.25 (0.08) 0.46 (0.16) 0.25 (0.10) Evenness (J′) 0.55 (0.04) 0.42 (0.09) 0.51 (0.05) 0.46 (0.18) 0.36 (0.12) 0.70 (0.09) 0.79 (0.02) Open in new tab Table 2 Number of hauls analysed and mean ecological parameters (standard error) for each group resulting from cluster and MDS analyses (see Figures 3 and 4) of elasmobranch samples obtained during bottom trawl surveys carried out in two areas off the Balearic Islands. . Northern area . Southern area . . SH . SL1 . SL2 . SH . SL1 . SL2 . SL3 . Hauls 52 15 18 3 6 26 10 Number of species 17 10 6 4 4 6 3 Mean species richness 2.8 (0.2) 2.8 (0.4) 2.1 (0.1) 2.3 (0.3) 2.1 (0.2) 2.1 (0.1) 1.5 (0.2) Fish per 10 000 m2 13.1 (1.8) 21.0 (5.6) 3.8 (1.0) 12.4 (1.3) 25.1 (7.2) 2.7 (0.2) 1.0 (0.2) g 10 000 m−2 2564.3 (365.2) 876.1 (217.1) 612.4 (141.6) 945.5 (205.9) 900.2 (242.9) 982.5 (157.4) 461.3 (132.6) Mean fish weight (g) 246.3 (25.7) 45.6 (5.5) 184.0 (38.3) 76.8 (18.4) 36.1 (7.4) 431.9 (92.6) 480.9 (139.1) Diversity (H′) 0.52 (0.06) 0.36 (0.09) 0.36 (0.05) 0.42 (0.23) 0.25 (0.08) 0.46 (0.16) 0.25 (0.10) Evenness (J′) 0.55 (0.04) 0.42 (0.09) 0.51 (0.05) 0.46 (0.18) 0.36 (0.12) 0.70 (0.09) 0.79 (0.02) . Northern area . Southern area . . SH . SL1 . SL2 . SH . SL1 . SL2 . SL3 . Hauls 52 15 18 3 6 26 10 Number of species 17 10 6 4 4 6 3 Mean species richness 2.8 (0.2) 2.8 (0.4) 2.1 (0.1) 2.3 (0.3) 2.1 (0.2) 2.1 (0.1) 1.5 (0.2) Fish per 10 000 m2 13.1 (1.8) 21.0 (5.6) 3.8 (1.0) 12.4 (1.3) 25.1 (7.2) 2.7 (0.2) 1.0 (0.2) g 10 000 m−2 2564.3 (365.2) 876.1 (217.1) 612.4 (141.6) 945.5 (205.9) 900.2 (242.9) 982.5 (157.4) 461.3 (132.6) Mean fish weight (g) 246.3 (25.7) 45.6 (5.5) 184.0 (38.3) 76.8 (18.4) 36.1 (7.4) 431.9 (92.6) 480.9 (139.1) Diversity (H′) 0.52 (0.06) 0.36 (0.09) 0.36 (0.05) 0.42 (0.23) 0.25 (0.08) 0.46 (0.16) 0.25 (0.10) Evenness (J′) 0.55 (0.04) 0.42 (0.09) 0.51 (0.05) 0.46 (0.18) 0.36 (0.12) 0.70 (0.09) 0.79 (0.02) Open in new tab In both areas, the ANOSIM analysis showed no seasonal differences, in terms of abundance and biomass, but a high dissimilarity between assemblages obtained from cluster and MDS analyses (Table 3). No differences were obtained only between the SH and SL1 assemblages from the south area. In all other instances, either abundance, or biomass or both were significantly different. The results of the SIMPER analysis showed the separate contributions, in terms of abundance, of the most important species to the average similarity within each assemblage and the average dissimilarity between them (Tables 4 and 5). These results indicated a high dissimilarity between assemblages and confirmed the existence of well-defined groups, with changes in the abundance of the main species. In the northern area, the species which characterised the different assemblages were S. canicula and R. miraletus for SH, G. melastomus, S. canicula and R. oxyrhinchus for SL1 and G. melastomus and E. spinax for SL2. In the southern area, the main species by assemblage were S. canicula for SH, S. canicula and G. melastomus for SL1, G. melastomus and E. spinax for SL2 and C. coelolepis and Centrophorus uyato for SL3. Table 3 Results of the ANOSIM routine to analyse differences between seasons and depths, using the groups resulting from cluster and MDS analyses for elasmobranch samples obtained during bottom trawl surveys carried out in two areas off the Balearic Islands. . R Global . . Northern area . Southern area . Comparison . Abundance . Biomass . Abundance . Biomass . Between seasons Autumn vs spring −0.001 −0.001 −0.066 −0.007 Between depth ranges 0.62* 0.72* 0.72* 0.53* SH vs SL1 0.52* 0.63* 0.23 0.13 SH vs SL2 0.81* 0.92* 0.86* 0.82* SH vs SL3 1.00* 1.00 SL1 vs SL2 0.52* 0.52* 0.64* 0.38* SL1 vs SL3 0.92* 0.81* SL2 vs SL3 0.69* 0.47* . R Global . . Northern area . Southern area . Comparison . Abundance . Biomass . Abundance . Biomass . Between seasons Autumn vs spring −0.001 −0.001 −0.066 −0.007 Between depth ranges 0.62* 0.72* 0.72* 0.53* SH vs SL1 0.52* 0.63* 0.23 0.13 SH vs SL2 0.81* 0.92* 0.86* 0.82* SH vs SL3 1.00* 1.00 SL1 vs SL2 0.52* 0.52* 0.64* 0.38* SL1 vs SL3 0.92* 0.81* SL2 vs SL3 0.69* 0.47* (*) Denotes a statistically significant difference at the 95% confidence interval. Open in new tab Table 3 Results of the ANOSIM routine to analyse differences between seasons and depths, using the groups resulting from cluster and MDS analyses for elasmobranch samples obtained during bottom trawl surveys carried out in two areas off the Balearic Islands. . R Global . . Northern area . Southern area . Comparison . Abundance . Biomass . Abundance . Biomass . Between seasons Autumn vs spring −0.001 −0.001 −0.066 −0.007 Between depth ranges 0.62* 0.72* 0.72* 0.53* SH vs SL1 0.52* 0.63* 0.23 0.13 SH vs SL2 0.81* 0.92* 0.86* 0.82* SH vs SL3 1.00* 1.00 SL1 vs SL2 0.52* 0.52* 0.64* 0.38* SL1 vs SL3 0.92* 0.81* SL2 vs SL3 0.69* 0.47* . R Global . . Northern area . Southern area . Comparison . Abundance . Biomass . Abundance . Biomass . Between seasons Autumn vs spring −0.001 −0.001 −0.066 −0.007 Between depth ranges 0.62* 0.72* 0.72* 0.53* SH vs SL1 0.52* 0.63* 0.23 0.13 SH vs SL2 0.81* 0.92* 0.86* 0.82* SH vs SL3 1.00* 1.00 SL1 vs SL2 0.52* 0.52* 0.64* 0.38* SL1 vs SL3 0.92* 0.81* SL2 vs SL3 0.69* 0.47* (*) Denotes a statistically significant difference at the 95% confidence interval. Open in new tab Table 4 Results of the SIMPER routine to analyse dissimilarity between groups resulting from cluster and MDS analyses for elasmobranch samples obtained during BALAR bottom trawl surveys, carried out between depths of 40 and 800 m in the northern area off the Balearic Islands and percentage contribution, in terms of abundance, of the main species to each group. Ā: abundance; S̄i: average similarity; δ̄i: average dissimilarity, SD: standard deviation. Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=30.44 S. canicula 10.72 27.16 1.02 89.22 89.22 R. miraletus 0.64 1.42 0.34 4.66 93.88 SL1 S̄i=30.56 G. melastomus 16.73 27.24 1.03 89.11 89.11 S. canicula 3.93 3.07 0.45 10.04 99.15 R. oxyrhinchus 0.19 0.24 0.44 0.79 99.94 SL2 S̄i=56.32 G. melastomus 3.59 50.57 2.36 89.79 89.79 E. spinax 0.45 5.74 1.1 10.19 99.98 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=89.36 G. melastomus 43.16 1.43 48.3 48.3 S. canicula 34.55 1.19 38.67 86.96 R. miraletus 3.06 0.43 3.42 90.39 SH vs SL2 δ̄i=98.40 S. canicula 48.66 1.54 49.45 49.45 G. melastomus 29.76 1.26 30.25 79.69 R. miraletus 4.65 0.56 4.72 84.42 E. spinax 4.09 0.86 4.16 88.58 SL1 vs SL2 δ̄i=74.66 G. melastomus 54.89 2.13 73.51 73.51 S. canicula 13.44 0.68 18 91.52 E. spinax 3.77 0.72 5.05 96.57 R. oxyrhinchus 1.2 0.44 1.6 98.17 Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=30.44 S. canicula 10.72 27.16 1.02 89.22 89.22 R. miraletus 0.64 1.42 0.34 4.66 93.88 SL1 S̄i=30.56 G. melastomus 16.73 27.24 1.03 89.11 89.11 S. canicula 3.93 3.07 0.45 10.04 99.15 R. oxyrhinchus 0.19 0.24 0.44 0.79 99.94 SL2 S̄i=56.32 G. melastomus 3.59 50.57 2.36 89.79 89.79 E. spinax 0.45 5.74 1.1 10.19 99.98 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=89.36 G. melastomus 43.16 1.43 48.3 48.3 S. canicula 34.55 1.19 38.67 86.96 R. miraletus 3.06 0.43 3.42 90.39 SH vs SL2 δ̄i=98.40 S. canicula 48.66 1.54 49.45 49.45 G. melastomus 29.76 1.26 30.25 79.69 R. miraletus 4.65 0.56 4.72 84.42 E. spinax 4.09 0.86 4.16 88.58 SL1 vs SL2 δ̄i=74.66 G. melastomus 54.89 2.13 73.51 73.51 S. canicula 13.44 0.68 18 91.52 E. spinax 3.77 0.72 5.05 96.57 R. oxyrhinchus 1.2 0.44 1.6 98.17 Open in new tab Table 4 Results of the SIMPER routine to analyse dissimilarity between groups resulting from cluster and MDS analyses for elasmobranch samples obtained during BALAR bottom trawl surveys, carried out between depths of 40 and 800 m in the northern area off the Balearic Islands and percentage contribution, in terms of abundance, of the main species to each group. Ā: abundance; S̄i: average similarity; δ̄i: average dissimilarity, SD: standard deviation. Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=30.44 S. canicula 10.72 27.16 1.02 89.22 89.22 R. miraletus 0.64 1.42 0.34 4.66 93.88 SL1 S̄i=30.56 G. melastomus 16.73 27.24 1.03 89.11 89.11 S. canicula 3.93 3.07 0.45 10.04 99.15 R. oxyrhinchus 0.19 0.24 0.44 0.79 99.94 SL2 S̄i=56.32 G. melastomus 3.59 50.57 2.36 89.79 89.79 E. spinax 0.45 5.74 1.1 10.19 99.98 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=89.36 G. melastomus 43.16 1.43 48.3 48.3 S. canicula 34.55 1.19 38.67 86.96 R. miraletus 3.06 0.43 3.42 90.39 SH vs SL2 δ̄i=98.40 S. canicula 48.66 1.54 49.45 49.45 G. melastomus 29.76 1.26 30.25 79.69 R. miraletus 4.65 0.56 4.72 84.42 E. spinax 4.09 0.86 4.16 88.58 SL1 vs SL2 δ̄i=74.66 G. melastomus 54.89 2.13 73.51 73.51 S. canicula 13.44 0.68 18 91.52 E. spinax 3.77 0.72 5.05 96.57 R. oxyrhinchus 1.2 0.44 1.6 98.17 Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=30.44 S. canicula 10.72 27.16 1.02 89.22 89.22 R. miraletus 0.64 1.42 0.34 4.66 93.88 SL1 S̄i=30.56 G. melastomus 16.73 27.24 1.03 89.11 89.11 S. canicula 3.93 3.07 0.45 10.04 99.15 R. oxyrhinchus 0.19 0.24 0.44 0.79 99.94 SL2 S̄i=56.32 G. melastomus 3.59 50.57 2.36 89.79 89.79 E. spinax 0.45 5.74 1.1 10.19 99.98 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=89.36 G. melastomus 43.16 1.43 48.3 48.3 S. canicula 34.55 1.19 38.67 86.96 R. miraletus 3.06 0.43 3.42 90.39 SH vs SL2 δ̄i=98.40 S. canicula 48.66 1.54 49.45 49.45 G. melastomus 29.76 1.26 30.25 79.69 R. miraletus 4.65 0.56 4.72 84.42 E. spinax 4.09 0.86 4.16 88.58 SL1 vs SL2 δ̄i=74.66 G. melastomus 54.89 2.13 73.51 73.51 S. canicula 13.44 0.68 18 91.52 E. spinax 3.77 0.72 5.05 96.57 R. oxyrhinchus 1.2 0.44 1.6 98.17 Open in new tab Table 5 Results of the SIMPER routine to analyse dissimilarity between groups resulting from cluster and MDS analyses of elasmobranch samples obtained during QUIMERA bottom trawl surveys, carried out between depths of 200 and 1800 m in the southern area off the Balearic Islands and percentage contribution, in terms of abundance, of the main species to each group. Ā: abundance; S̄i: average similarity; δ̄i: average dissimilarity, SD: standard deviation. Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=50.24 S. canicula 9.55 49.55 1.73 98.63 98.63 S. blainvillei 2.12 0.74 0.43 1.01 99.64 R. asterias 0.5 0.69 0.58 0.36 100 SL1 S̄i=20.4 S. canicula 9.29 10.97 0.5 53.79 53.79 G. melastomus 14.5 8.97 0.56 43.96 97.75 SL2 S̄i=54.98 G. melastomus 1.98 45.17 1.65 82.16 82.16 E. spinax 0.51 9.42 0.98 17.13 99.29 SL3 S̄i=51.20 C. coelolepis 0.87 47.51 2.08 92.80 92.80 C. uyato 0.20 3.69 0.41 8.20 100 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=72.88 S. canicula 28.78 1.6 39.49 39.49 G. melastomus 28.5 0.86 39.11 78.6 SH vs SL2 δ̄i=99.96 S. canicula 62.51 2.56 62.53 62.52 S. blainvillei 15.55 0.7 15.56 78.10 G. melastomus 12.86 1.81 12.86 90.96 SH vs SL3 δ̄i=100 S. canicula 69.26 2.58 69.26 69.26 S. blainvillei 17.42 0.69 17.42 86.68 C. coelolepis 6.35 1.51 6.35 93.03 SL1 vs SL2 δ̄i=86.97 S. canicula 37.64 0.99 43.27 43.27 G. melastomus 36.20 0.99 41.63 84.90 E. spinax 12.41 0.52 14.27 99.17 SL1 vs SL3 δ̄i=98.29 S. canicula 40.54 0.99 41.25 41.25 G. melastomus 38.31 0.93 38.97 80.22 E. spinax 14.47 0.47 14.72 94.94 SL2 vs SL3 δ̄i=82.62 G. melastomus 47.27 2.01 54.84 54.84 C. coelolepis 23.64 1.47 27.42 82.62 E. spinax 13.68 1.16 15.86 98.12 Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=50.24 S. canicula 9.55 49.55 1.73 98.63 98.63 S. blainvillei 2.12 0.74 0.43 1.01 99.64 R. asterias 0.5 0.69 0.58 0.36 100 SL1 S̄i=20.4 S. canicula 9.29 10.97 0.5 53.79 53.79 G. melastomus 14.5 8.97 0.56 43.96 97.75 SL2 S̄i=54.98 G. melastomus 1.98 45.17 1.65 82.16 82.16 E. spinax 0.51 9.42 0.98 17.13 99.29 SL3 S̄i=51.20 C. coelolepis 0.87 47.51 2.08 92.80 92.80 C. uyato 0.20 3.69 0.41 8.20 100 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=72.88 S. canicula 28.78 1.6 39.49 39.49 G. melastomus 28.5 0.86 39.11 78.6 SH vs SL2 δ̄i=99.96 S. canicula 62.51 2.56 62.53 62.52 S. blainvillei 15.55 0.7 15.56 78.10 G. melastomus 12.86 1.81 12.86 90.96 SH vs SL3 δ̄i=100 S. canicula 69.26 2.58 69.26 69.26 S. blainvillei 17.42 0.69 17.42 86.68 C. coelolepis 6.35 1.51 6.35 93.03 SL1 vs SL2 δ̄i=86.97 S. canicula 37.64 0.99 43.27 43.27 G. melastomus 36.20 0.99 41.63 84.90 E. spinax 12.41 0.52 14.27 99.17 SL1 vs SL3 δ̄i=98.29 S. canicula 40.54 0.99 41.25 41.25 G. melastomus 38.31 0.93 38.97 80.22 E. spinax 14.47 0.47 14.72 94.94 SL2 vs SL3 δ̄i=82.62 G. melastomus 47.27 2.01 54.84 54.84 C. coelolepis 23.64 1.47 27.42 82.62 E. spinax 13.68 1.16 15.86 98.12 Open in new tab Table 5 Results of the SIMPER routine to analyse dissimilarity between groups resulting from cluster and MDS analyses of elasmobranch samples obtained during QUIMERA bottom trawl surveys, carried out between depths of 200 and 1800 m in the southern area off the Balearic Islands and percentage contribution, in terms of abundance, of the main species to each group. Ā: abundance; S̄i: average similarity; δ̄i: average dissimilarity, SD: standard deviation. Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=50.24 S. canicula 9.55 49.55 1.73 98.63 98.63 S. blainvillei 2.12 0.74 0.43 1.01 99.64 R. asterias 0.5 0.69 0.58 0.36 100 SL1 S̄i=20.4 S. canicula 9.29 10.97 0.5 53.79 53.79 G. melastomus 14.5 8.97 0.56 43.96 97.75 SL2 S̄i=54.98 G. melastomus 1.98 45.17 1.65 82.16 82.16 E. spinax 0.51 9.42 0.98 17.13 99.29 SL3 S̄i=51.20 C. coelolepis 0.87 47.51 2.08 92.80 92.80 C. uyato 0.20 3.69 0.41 8.20 100 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=72.88 S. canicula 28.78 1.6 39.49 39.49 G. melastomus 28.5 0.86 39.11 78.6 SH vs SL2 δ̄i=99.96 S. canicula 62.51 2.56 62.53 62.52 S. blainvillei 15.55 0.7 15.56 78.10 G. melastomus 12.86 1.81 12.86 90.96 SH vs SL3 δ̄i=100 S. canicula 69.26 2.58 69.26 69.26 S. blainvillei 17.42 0.69 17.42 86.68 C. coelolepis 6.35 1.51 6.35 93.03 SL1 vs SL2 δ̄i=86.97 S. canicula 37.64 0.99 43.27 43.27 G. melastomus 36.20 0.99 41.63 84.90 E. spinax 12.41 0.52 14.27 99.17 SL1 vs SL3 δ̄i=98.29 S. canicula 40.54 0.99 41.25 41.25 G. melastomus 38.31 0.93 38.97 80.22 E. spinax 14.47 0.47 14.72 94.94 SL2 vs SL3 δ̄i=82.62 G. melastomus 47.27 2.01 54.84 54.84 C. coelolepis 23.64 1.47 27.42 82.62 E. spinax 13.68 1.16 15.86 98.12 Depth range . Species . Ā . S̄i . S̄i/SD . S̄i% . ∑Si% . SH S̄i=50.24 S. canicula 9.55 49.55 1.73 98.63 98.63 S. blainvillei 2.12 0.74 0.43 1.01 99.64 R. asterias 0.5 0.69 0.58 0.36 100 SL1 S̄i=20.4 S. canicula 9.29 10.97 0.5 53.79 53.79 G. melastomus 14.5 8.97 0.56 43.96 97.75 SL2 S̄i=54.98 G. melastomus 1.98 45.17 1.65 82.16 82.16 E. spinax 0.51 9.42 0.98 17.13 99.29 SL3 S̄i=51.20 C. coelolepis 0.87 47.51 2.08 92.80 92.80 C. uyato 0.20 3.69 0.41 8.20 100 Pair-wise comparisons δ̄i δ̄i/SD δ̄i% ∑δi% SH vs SL1 δ̄i=72.88 S. canicula 28.78 1.6 39.49 39.49 G. melastomus 28.5 0.86 39.11 78.6 SH vs SL2 δ̄i=99.96 S. canicula 62.51 2.56 62.53 62.52 S. blainvillei 15.55 0.7 15.56 78.10 G. melastomus 12.86 1.81 12.86 90.96 SH vs SL3 δ̄i=100 S. canicula 69.26 2.58 69.26 69.26 S. blainvillei 17.42 0.69 17.42 86.68 C. coelolepis 6.35 1.51 6.35 93.03 SL1 vs SL2 δ̄i=86.97 S. canicula 37.64 0.99 43.27 43.27 G. melastomus 36.20 0.99 41.63 84.90 E. spinax 12.41 0.52 14.27 99.17 SL1 vs SL3 δ̄i=98.29 S. canicula 40.54 0.99 41.25 41.25 G. melastomus 38.31 0.93 38.97 80.22 E. spinax 14.47 0.47 14.72 94.94 SL2 vs SL3 δ̄i=82.62 G. melastomus 47.27 2.01 54.84 54.84 C. coelolepis 23.64 1.47 27.42 82.62 E. spinax 13.68 1.16 15.86 98.12 Open in new tab The bathymetric distribution of mean abundance for the above mentioned main species in each area is shown in Figure 5. Clear differences were obtained among species, but for each species similar trends were obtained within them in the two surveyed areas. Within the sharks, S. canicula reached its maximum abundance above a depth of 100 m but was captured down to 500 m. S. blainvillei was captured almost exclusively between 101 and 300 m depth. G. melastomus appeared between depths of 301–1800 m, with clear maximum abundance between 301 and 500 m. E. spinax was captured between 301–1500 m depth, with similar values of abundance from 301 to 1300 m. C. coelolepis was only caught below a depth of 1301 m and reached its maximum abundance at the deepest interval surveyed. By contrast, most of the analysed rays were abundant above a depth of 300 m, reaching their maximum values at <100 m for R. miraletus, and between 101 and 300 m depth for R. asterias, R. clavata and R. naevus. The only exception was R. oxyrhinchus, which appeared from 101 to 500 m, reaching its maximum abundance between 301 and 500 m depth. Figure 5 Open in new tabDownload slide Mean standardised abundance (fish per 10 000 m2), calculated in the northern (●; from GOC73) and southern (○; from OTMS-27.5) areas surveyed and by depth interval, for the main species of elasmobranchs captured during bottom trawl surveys carried out off the Balearic Islands. Figure 5 Open in new tabDownload slide Mean standardised abundance (fish per 10 000 m2), calculated in the northern (●; from GOC73) and southern (○; from OTMS-27.5) areas surveyed and by depth interval, for the main species of elasmobranchs captured during bottom trawl surveys carried out off the Balearic Islands. Length frequency distribution by depth for S. canicula, G. melastomus and E. spinax showed clear differences. For S. canicula, the overall length frequency ranged from 5 to 50 cm (Figure 6), although specimens ≥25 cm were most frequent at depths of <100 m, while smaller fish were only distributed between depths of 101 and 500 m. By contrast, in G. melastomus, the length ranged between 10 and 70 cm (Figure 6), and specimens ≤30 cm were most common above a depth of 700 m, while females ≥40 cm predominated below this depth. Similar results were obtained for E. spinax (Figure 7); lengths ranged between 5 and 45 cm, with specimens ≤20 cm distributed almost exclusively from 301 to 900 m, while fish ≥30 cm predominated below a depth of 701 m. Figure 6 Open in new tabDownload slide Length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) and depth interval of S. canicula captured during bottom trawl surveys off the Balearic Islands. Figure 6 Open in new tabDownload slide Length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) and depth interval of S. canicula captured during bottom trawl surveys off the Balearic Islands. Figure 7 Open in new tabDownload slide Length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) and depth interval of G. melastomus and E. spinax captured during bottom trawl surveys off the Balearic Islands. Figure 7 Open in new tabDownload slide Length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) and depth interval of G. melastomus and E. spinax captured during bottom trawl surveys off the Balearic Islands. The other sharks S. blainvillei and C. coelolepis ranged between 20 and 70 and 20 and 90 cm, respectively, and showed a bimodal distribution (Figure 8). In S. blainvillei there was a dominance of large fish (between 40 and 70 cm, with a mode at 50 cm), while small specimens ranged from 20 to 30 cm. By contrast, small specimens (mode at 20–30 cm) dominated in C. coelolepis, which also showed a second mode at 50–65 cm. Figure 8 Open in new tabDownload slide Overall length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) of the rays R. miraletus, R. asterias, R. clavata, R. naevus and R. oxyrhinchus. Figure 8 Open in new tabDownload slide Overall length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) of the rays R. miraletus, R. asterias, R. clavata, R. naevus and R. oxyrhinchus. The overall length frequency distributions of rays R. miraletus (length range of 10–50 cm), R. asterias (15–90 cm) and R. naevus (10–55 cm) showed modes situated at 20, 20 and 35–40 cm, respectively (Figure 9). By contrast, no clear modes could be observed in R. clavata and R. oxyrhinchus, the two species with a major presence of large specimens (>40 cm), with length ranges ranging from 10 to 90 and 15 to 115 cm, respectively. Figure 9 Open in new tabDownload slide Overall length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) of the sharks S. blainvillei and C. coelolepis captured during bottom trawl surveys off the Balearic Islands. Figure 9 Open in new tabDownload slide Overall length frequency distribution (in percentage by 5 cm size classes) by sex (black bars for females and white bars for males) of the sharks S. blainvillei and C. coelolepis captured during bottom trawl surveys off the Balearic Islands. Discussion The analysis of demersal elasmobranch species distributed in two different areas off the Balearic Islands, along the continental shelf and slope between depths of 41 and 1713 m, has shown that some assemblages were related to depth. These results are similar to those obtained in Atlantic waters, when elasmobranch species were also analysed separately (Roel, 1987). The bathymetric boundaries obtained in this study are similar in both areas, and they are in accordance with those obtained in previous studies of fish communities (both selachians and teleosts) carried out in our study area (Massutı́ et al., 1996; Moranta et al., 1998) and in other areas of the western Mediterranean (Stefanescu et al., 1993; Demestre et al., 2000). The assemblages found in this study can be attributed to the different fish zonations proposed by Haedrich and Merret (1988) for Atlantic waters and corroborated in the Mediterranean by Stefanescu et al. (1993) and Demestre et al. (2000). Samples taken above a depth of 250 m correspond to the continental shelf (SH), over which the highest diversity of demersal elasmobranchs is reached. In this depth-strata, the most abundant species is S. canicula, although there are also other characteristic species such as the shark S. blainvillei and the rays R. miraletus, R. asterias, R. clavata and R. naevus. The low capture of ray species in the southern area could be attributed to the low number of samples taken on the continental shelf and the absence of samples above a depth of 195 m. In the northern area, where a large number of samples were taken on the shelf, higher numbers of rays (Raja brachyura, Raja montagui, Raja polystigma and Raja undulata), other sharks (Mustelus spp.) were captured, as well as other batoid species (Torpedo spp., Dasyatis pastinaca and Myliobatis aquila), which appeared at a very low frequency in bottom trawls (e.g. Massutı́ et al., 1996; Matarrese et al., 1996; Bertrand et al., 2000). This could be due to the scarcity of these species and their solitary habits, and to the low capture efficiency of the gear used. Along the slope, three different assemblages can be defined. In contrast to the shelf, these assemblages are characterised mainly by sharks, the only holocephalid species captured (Chimaera monstrosa), very few rays (R. oxyrhinchus is the only ray with an abundance peak on the slope) and the absence of other batoid species (e.g. the genera Torpedo, Dasyatis and Miliobatis). The shallowest slope assemblage corresponds to the upper slope (SL1; 300–500 m depth) and it is mainly characterised by G. melastomus, S. canicula and R. oxyrhinchus. The deepest slope assemblage, only surveyed in the southern area, corresponds to the lower slope (SL3; >1400 m depth) and it is mainly characterised by C. coelolepis, a species restricted to this depth and which, in the western Mediterranean, can occur down to a depth of 2250 m (Carrasón et al., 1992). Between these two assemblages, a third group (SL2; 500–1400 m depth) extends from the deep upper to the middle slope. It is characterised by E. spinax (a species restricted to this assemblage) and G. melastomus. The latter species is the most abundant elasmobranch captured, and has the widest bathymetric range (SL1, SL2 and SL3 assemblages). Some conclusions can be drawn concerning depth distribution patterns and the population structure of several abundant elasmobranch species collected in this study. In shark species, a clear segregation of sizes by depth was observed. For S. canicula, a species mainly distributed over the continental shelf but also occurring on the upper slope down to a depth of 500 m, the juveniles are found below 100 m while in shallower waters the population is composed exclusively of adults. Similar results have been obtained by D'Onghia et al. (1995) in the Northern Aegean Sea, who reported juveniles only at depths greater than 200 m. In addition, spawning in shallow waters on hard substrate off the Gulf of Lions has been suggested by Capapé et al. (1991). In G. melastomus and E. spinax, two species mainly distributed on the upper and middle slope, the different bathymetric distribution of juveniles and adults is more evident, with juveniles and adults in shallow and deep fishing grounds within the bathymetric range of the species, respectively. Similar results have been obtained by Tursi et al. (1993) in the Ionian Sea. In this area, G. melastomus found between 200 and 400 m were almost exclusively small (mainly concentrated at around 300 m), while between 400 and 650 m the population was found to comprise all length classes, including a considerable number of recruits. Considering the available information on length at first maturity for S. canicula (Capapé et al., 1991; Ungaro et al., 2002) and G. melastomus (Capapé and Zaouali, 1977) in the Mediterranean, the immature specimens of these two species off the Balearic Islands are mainly distributed between depths of 100 and 700 m. This depth range is widely exploited by the trawl fleet and for this reason, S. canicula and G. melastomus represent an important fraction of discards from this fishery (Moranta et al., 2000). The bathymetric distribution of R. miraletus in the study area is similar to that found in the central Mediterranean, where it is mainly concentrated between depths of 50 and 150 m (Relini et al., 1999) and off Tunisia, where it is distributed down to a depth of 200 m (Capapé and Quignard, 1974). The population found on the trawl fishing grounds off the Balearic Islands is mainly composed of immature specimens of 1 to 3 years of age (Abdel-Aziz, 1994). This species is part of the by-catch of the bottom trawl fishery, with a high proportion of individuals discarded. By contrast, the population structure of R. clavata shows a large proportion of mature specimens (>50 cm; Relini et al., 1999). Similar results are obtained for S. blainvillei, where a second mode of mature fish older than 3 years of age (Cannizaro et al., 1995) at around 50 cm in length can be observed. The analysis of available long-term data series has shown the impact of fishing activity on elasmobranchs, which is reflected in the reduction of species numbers and their declining abundance. Some biological factors may contribute to the vulnerability of this type of fish since they are long-lived and slow growing, mature at a late age and have a low fecundity. In the Atlantic Ocean, R. naevus and R. oxyrhinchus have been shown to be close to extinction in the north-west area (Casey and Myers, 1998) and in the Irish Sea (Dulvy et al., 2000), respectively. R. clavata has decreased both in abundance and in average length in the North Sea (Walker and Heessen, 1996). In the Mediterranean Sea, elasmobranch landings and number of species have decreased during recent decades in the Gulf of Lions, in direct relation to the development of the trawl fishery (Aldebert, 1997). In this area, the decline of abundance indices for R. clavata and reductions in its distribution have also been reported (Bertrand et al., 1998). Our results in the northern area can be compared with those obtained for the whole northern Mediterranean by Bertrand et al. (2000), where the same gear and sampling scheme as our study was used (Table 6). Diversity of demersal elasmobranchs in the Balearic Islands, even considering the low number of samples analysed, is higher than in adjacent waters off the Iberian Peninsula and similar to other insular Mediterranean areas (e.g. Sardinia, Corsica and Sicily islands), in which the highest values for the whole northern Mediterranean have been reported. Although biogeographic factors could form the basis of these differences, these results could also suggest the existence of some differences in fishing exploitation between areas, with lower intensity on the insular continental shelf and upper slope than along the peninsular bottoms. Table 6 Number of hauls analysed, elasmobranch species captured and standardised abundance (fish per km2) for the most abundant species reported from different areas of the western Mediterranean (Bertrand et al., 2000) and those obtained off the Balearic Islands from the BALAR surveys analysed in the present study, in which the same gear and sampling scheme were used. Abundance values from areas throughout the whole northern Mediterranean were obtained from an average of the 1994–1998 data series reported by Bertrand et al. (2000) at the different depth-strata in which the species were mainly distributed: (i) 10–200 m for Raja clavata and Raja miraletus; (ii) 200–800 m for Galeus melastomus and Etmopterus spinax; (iii) 10–800 m for Scyliorhinus canicula. Abundance values from the Balearic Islands were obtained from an average of spring and autumn data, in which no significant differences were detected (see Table 3). . . . Abundance: specimens per km2 . Area . Total hauls . Species number . R. clavata . R. miraletus . S. canicula . G. melastomus . E. spinax . Alboran Sea 170 16 0.0 0.4 50.2 1876.8 281.0 Central Iberian Peninsula 150 13 3.0 3.2 96.4 176.8 46.2 Northern Iberian Peninsula 215 10 2.0 0.0 231.4 107.4 8.4 Gulf of Lions 325 23 7.4 0.0 92.3 932.2 42.6 Corsica Island 120 26 40.2 101.2 590.4 641.4 54.2 Ligurian and Northern and Central Thyrrhenian 765 24 2.6 7.4 17.9 288.4 52.2 Sardinia Island 625 24 46.4 32.8 255.4 868.0 67.6 Sicily Island and South Thyrrhenian 705 29 7.6 115.6 34.0 253.6 67.8 Balearic Islands 85 22 54.0 88.0 804.0 1131.0 27.0 . . . Abundance: specimens per km2 . Area . Total hauls . Species number . R. clavata . R. miraletus . S. canicula . G. melastomus . E. spinax . Alboran Sea 170 16 0.0 0.4 50.2 1876.8 281.0 Central Iberian Peninsula 150 13 3.0 3.2 96.4 176.8 46.2 Northern Iberian Peninsula 215 10 2.0 0.0 231.4 107.4 8.4 Gulf of Lions 325 23 7.4 0.0 92.3 932.2 42.6 Corsica Island 120 26 40.2 101.2 590.4 641.4 54.2 Ligurian and Northern and Central Thyrrhenian 765 24 2.6 7.4 17.9 288.4 52.2 Sardinia Island 625 24 46.4 32.8 255.4 868.0 67.6 Sicily Island and South Thyrrhenian 705 29 7.6 115.6 34.0 253.6 67.8 Balearic Islands 85 22 54.0 88.0 804.0 1131.0 27.0 Open in new tab Table 6 Number of hauls analysed, elasmobranch species captured and standardised abundance (fish per km2) for the most abundant species reported from different areas of the western Mediterranean (Bertrand et al., 2000) and those obtained off the Balearic Islands from the BALAR surveys analysed in the present study, in which the same gear and sampling scheme were used. Abundance values from areas throughout the whole northern Mediterranean were obtained from an average of the 1994–1998 data series reported by Bertrand et al. (2000) at the different depth-strata in which the species were mainly distributed: (i) 10–200 m for Raja clavata and Raja miraletus; (ii) 200–800 m for Galeus melastomus and Etmopterus spinax; (iii) 10–800 m for Scyliorhinus canicula. Abundance values from the Balearic Islands were obtained from an average of spring and autumn data, in which no significant differences were detected (see Table 3). . . . Abundance: specimens per km2 . Area . Total hauls . Species number . R. clavata . R. miraletus . S. canicula . G. melastomus . E. spinax . Alboran Sea 170 16 0.0 0.4 50.2 1876.8 281.0 Central Iberian Peninsula 150 13 3.0 3.2 96.4 176.8 46.2 Northern Iberian Peninsula 215 10 2.0 0.0 231.4 107.4 8.4 Gulf of Lions 325 23 7.4 0.0 92.3 932.2 42.6 Corsica Island 120 26 40.2 101.2 590.4 641.4 54.2 Ligurian and Northern and Central Thyrrhenian 765 24 2.6 7.4 17.9 288.4 52.2 Sardinia Island 625 24 46.4 32.8 255.4 868.0 67.6 Sicily Island and South Thyrrhenian 705 29 7.6 115.6 34.0 253.6 67.8 Balearic Islands 85 22 54.0 88.0 804.0 1131.0 27.0 . . . Abundance: specimens per km2 . Area . Total hauls . Species number . R. clavata . R. miraletus . S. canicula . G. melastomus . E. spinax . Alboran Sea 170 16 0.0 0.4 50.2 1876.8 281.0 Central Iberian Peninsula 150 13 3.0 3.2 96.4 176.8 46.2 Northern Iberian Peninsula 215 10 2.0 0.0 231.4 107.4 8.4 Gulf of Lions 325 23 7.4 0.0 92.3 932.2 42.6 Corsica Island 120 26 40.2 101.2 590.4 641.4 54.2 Ligurian and Northern and Central Thyrrhenian 765 24 2.6 7.4 17.9 288.4 52.2 Sardinia Island 625 24 46.4 32.8 255.4 868.0 67.6 Sicily Island and South Thyrrhenian 705 29 7.6 115.6 34.0 253.6 67.8 Balearic Islands 85 22 54.0 88.0 804.0 1131.0 27.0 Open in new tab Differences in abundance indices for some of the most important species could be related to fishing pressure. In general, abundance off the Balearic Islands is higher than that reported from the Iberian Peninsula. It is also similar to the maximum abundance reported from other western Mediterranean areas off Corsica and Sicily for R. miraletus, off Corsica and Sardinia for R. clavata and off Corsica for S. canicula. In addition, the regular presence of R. oxyrhinchus on the slope bottoms of the Balearic Islands must also be pointed out. According to Bertrand et al. (2000), this species, which shows high vulnerability to fishing pressure, only occurs around Corsica and Sardinia, where trawling activity may be lower than in other Mediterranean adjacent areas. The only exceptions were G. melastomus and E. spinax, two species restricted to the slope which had maximum abundance off Alboran, with values much higher than those obtained from the other Mediterranean areas. The highest abundance indices of these two species in the Alboran Sea could be due to the low levels of fishing effort below a depth of 500 m in this area. This factor has also been used to explain differences in abundance and population structure obtained in a teleost species between this and other northern areas off the Iberian coast (Massutí et al., 2001). The present results form a reference point for the present status of demersal elasmobranchs in the Balearic Islands. This area, together with other insular areas, shows the most diverse and abundant elasmobranch assemblages in the western Mediterranean. For this reason, harvest strategies should be linked to the conservation of these species in these areas, and long-term monitoring programmes should be set up. This paper is a result of the Spanish and European Projects MEDER (IEO Project) and Deep-Sea Fisheries (DGXIV/FAIR/96/06-55), respectively. The authors are most grateful to all the participants in the cruises BALAR0401, BALAR0901, QUIMERA-I and QUIMERA-II as well as the crew of R/V “Francisco de Paula Navarro” and “García del Cid” for their help during the sampling, and to Dr C. Rodgers for help with improving the manuscript. References Abdel-Aziz S.H. . Observations on the biology of the common torpedo (Torpedo torpedo, Linnaeus, 1758) and marbled electric ray (Torpedo marmorata, Risso, 1810) from Egyptian Mediterranean waters , Australian Journal of Marine and Freshwater Research , 1994 , vol. 45 (pg. 693 - 704 ) Google Scholar Crossref Search ADS WorldCat Abelló P. , Bertrand J.A. , Gil de Sola L. , Papaconstantinou C. , Relini G. , Souplet A. . Mediterranean marine demersal resources: the MEDITS international trawl survey (1994–1999) , Scientia Marina , 2002 , vol. 66 Suppl. 2 (pg. 5 - 280 ) OpenURL Placeholder Text WorldCat Aldebert Y. . Demersal resources of the Gulf of Lions (Mediterranean). Impact on fish diversity , Vie et Milieu , 1997 , vol. 47 (pg. 275 - 284 ) OpenURL Placeholder Text WorldCat Bertrand J.A. , Aldebert Y. , Souplet A. . Temporal variability of demersal species in the Gulf of Lions from trawl surveys (1983–1997) , IFREMER Actes de Colloques , 1998 , vol. 26 (pg. 153 - 164 ) OpenURL Placeholder Text WorldCat Bertrand J.A. , Gil de Sola L. , Papaconstantinou C. , Relini G. , Souplet A. . Contribution on the distribution of elasmobranchs in the Mediterranean (from the MEDITS surveys) , Biologia Marina Mediterranea , 2000 , vol. 7 (pg. 1 - 15 ) OpenURL Placeholder Text WorldCat Cannizaro L. , Rizzo P. , Levi D. , Gancitano S. . Age determination and growth of Squalus blainvillei (Risso, 1826) , Fisheries Research , 1995 , vol. 23 (pg. 113 - 125 ) Google Scholar Crossref Search ADS WorldCat Capapé C. , Quignard J.P. . Contribution à la biologie des Rajidae des côtes tunisiennes. I. Raja miraletus Linné, 1758: répartition géographique et bathymétrique, sexualité, reproduction, fécondité , Archives Institut Pasteur de Tunis , 1974 , vol. 51 (pg. 39 - 60 ) OpenURL Placeholder Text WorldCat Capapé C. , Zaouali J. . Contribution à la biologie des Scyliorhinidae des côtes tunisiennes. VI: Galeus melastomus Rafinesque, 1810. Répartition géographique et bathymétrique, sexualité, reproduction, fécondité , Cahiers de Biologie Marine , 1977 , vol. 18 (pg. 449 - 463 ) OpenURL Placeholder Text WorldCat Capapé C. , Tomasini J.A. , Bouchereau J.L. . Observations sur la biologie de reproduction de la petite roussette, Scyliorhinus canicula (Linnaeus, 1758) (Pisces, Scyliorhinidae) du golfe du Lion (France méridionale) , Ichtyophysiologica Acta , 1991 , vol. 13 (pg. 87 - 109 ) OpenURL Placeholder Text WorldCat Capapé C. , Tomasini J.A. , Quignard J.P. . Les elasmobranches pleurotrêmes de la côte du Languedoc (France, Méridionale): observations biologiques et démographiques , Vie et Milieu , 2000 , vol. 50 (pg. 123 - 133 ) OpenURL Placeholder Text WorldCat Carrasón M. , Stefanescu C. , Cartes J.E. . Diets and bathymetric distributions of two bathyal sharks of the Catalan deep sea (western Mediterranean) , Marine Ecology Progress Series , 1992 , vol. 82 (pg. 21 - 30 ) Google Scholar Crossref Search ADS WorldCat Casey J.M. , Myers R.A. . Near extinction of a large, widely distributed fish , Science , 1998 , vol. 281 (pg. 690 - 691 ) Google Scholar Crossref Search ADS PubMed WorldCat Clarke K.R. , Gorley R.N. . , PRIMER v5: User Manual/Tutorial , 2001 Plymouth PRIMER-E 91 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Demestre M. , Sánchez P. , Abelló P. . Demersal fish assemblages and habitat characteristics on the continental shelf and upper slope of the north-western Mediterranean , Journal of the Marine Biological Association of the United Kingdom , 2000 , vol. 80 (pg. 981 - 988 ) Google Scholar Crossref Search ADS WorldCat D'Onghia G. , Matarrese A. , Tursi A. , Sion L. . Observations on the depth distribution pattern of the small-spotted catshark in the North Aegean Sea , Journal of Fish Biology , 1995 , vol. 47 (pg. 421 - 426 ) Google Scholar Crossref Search ADS WorldCat Dulvy N.K. , Metcalfe J.D. , Glanville J. , Pawson M.G. , Reynolds J.D. . Fishery stability, local extinctions and shifts in community structure in skates , Conservation Biology , 2000 , vol. 14 (pg. 283 - 293 ) Google Scholar Crossref Search ADS WorldCat Gordon J.D.M. , Merrett N.R. , Bergstad O.A. , Swan S.C. . A comparison of the deep-water demersal fish assemblages of the Rockall Trough and Porcupine Seabight, eastern North Atlantic: continental slope to rise , Journal of the Marine Biological Association of the United Kingdom , 1996 , vol. 49 Suppl. A (pg. 217 - 238 ) OpenURL Placeholder Text WorldCat Haedrich R.L. , Merret N.R. . Summary atlas of deep-living demersal fishes in the North Atlantic Basin , Journal of Natural History , 1988 , vol. 22 (pg. 1325 - 1362 ) Google Scholar Crossref Search ADS WorldCat Massutí E. , Reñones O. , Carbonell A. , Oliver P. . Demersal fish communities exploited on the continental shelf and slope off Majorca (Balearic Islands, NW Mediterranean) , Vie et Milieu , 1996 , vol. 46 (pg. 45 - 55 ) OpenURL Placeholder Text WorldCat Massutí E. , Moranta J. , Gil de Sola L. , Morales-Nin B. , Prats L. . Distribution and population structure of the rockfish Helicolenus dactylopterus (Pisces: Scorpaenidae) in the western Mediterranean , Journal of the Marine Biological Association of the United Kingdom , 2001 , vol. 81 (pg. 129 - 141 ) Google Scholar Crossref Search ADS WorldCat Matarrese A. , D'Onghia G. , Tursi A. , Basanisi M. . New information on the ichthyofauna of the south-eastern Italian coast (Ionian Sea) , Cybium , 1996 , vol. 20 (pg. 197 - 211 ) OpenURL Placeholder Text WorldCat Merrett N.R. , Gordon J.D.M. , Stehmann M. , Haedrich R.L. . Deep demersal fish assemblage structure in the Porcupine Seabight (eastern North Atlantic): slope sampling by three different trawls compared , Journal of the Marine Biological Association of the United Kingdom , 1991 , vol. 71 (pg. 329 - 358 ) Google Scholar Crossref Search ADS WorldCat Moranta J. , Stefanescu C. , Massutí E. , Morales-Nin B. , Lloris D. . Fish community structure and depth-related trends on the continental slope of the Balearic Islands (Algerian basin, western Mediterranean) , Marine Ecology Progress Series , 1998 , vol. 171 (pg. 247 - 259 ) Google Scholar Crossref Search ADS WorldCat Moranta J. , Massutí E. , Morales-Nin B. . Fish catch composition of the deep-sea decapod crustacean fisheries in the Balearic Islands (western Mediterranean) , Fisheries Research , 2000 , vol. 45 (pg. 253 - 264 ) Google Scholar Crossref Search ADS WorldCat Papaconstantinou C. , Vassilopoulou V. , Petrakis G. , Caragitsou E. , Mytilinaeou C. , Fourtouni A. , Politou C.-Y. . The demersal fish fauna of the North and West Aegean Sea , Bios , 1994 , vol. 2 (pg. 35 - 45 ) OpenURL Placeholder Text WorldCat Relini G. , Bertrand J.A. , Zamboni A. . Synthesis of the knowledge on bottom fishery resources in Central Mediterranean (Italy and Corsica) , Biologia Marina Mediterranea , 1999 , vol. 6 Suppl. 1 (pg. 94 - 98 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Relini G. , Biagi F. , Serena F. , Belluscio A. , Spedicato M.T. , Rinelli P. , Follesa M.C. , Piccinetti C. , Ungaro N. , Sion L. , Levi D. . I Selaci pescati con lo strascico nei mari italiani , Biologia Marina Mediterranea , 2000 , vol. 7 (pg. 347 - 384 ) OpenURL Placeholder Text WorldCat Roel B.A. . Demersal communities off the west coast of South Africa , South African Journal of Marine Science , 1987 , vol. 5 (pg. 575 - 584 ) Google Scholar Crossref Search ADS WorldCat Sardà F. , Cartes J.E. , Company J.B. , Albiol A. . A modified commercial trawl used to sample deep-sea megabenthos , Fisheries Science , 1998 , vol. 64 (pg. 492 - 493 ) Google Scholar Crossref Search ADS WorldCat Stefanescu C. , Lloris D. , Rucabado J. . Deep-living demersal fishes in the Catalan Sea (western Mediterranean) below a depth of 1000 m , Journal of Natural History , 1992 , vol. 26 (pg. 197 - 213 ) Google Scholar Crossref Search ADS WorldCat Stefanescu C. , Lloris D. , Rucabado J. . Deep-sea fish assemblages in the Catalan Sea (western Mediterranean) below a depth of 1000 m , Deep-Sea Research , 1993 , vol. 40 (pg. 695 - 707 ) Google Scholar Crossref Search ADS WorldCat Stevens J.D. , Bonfil R. , Dulvy N.K. , Walker P.A. . The effects of fishing on sharks, rays, and chimaeras (chondrichthyans), and the implications for marine ecosystems , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 476 - 494 ) Google Scholar Crossref Search ADS WorldCat Tursi A. , D'Onghia G. , Matarrese A. , Piscitelli G. . Observations on population biology of the blackmouth catshark Galeus melastomus (Chondrichthyes, Scyliorhinidae) in the Ionian Sea , Cybium , 1993 , vol. 17 (pg. 187 - 196 ) OpenURL Placeholder Text WorldCat Ungaro N. , Marano G. , Marzano M. C. . Vacchi M. , La Mesa G. , Serena F. , Séret B. . Size at sexual maturity of the smallspotted catshark, Scyliorhinus canicula, in the southern Adriatic Sea (Mediterranean Sea) , 2002 Proceedings of the Fourth European Elasmobranch Association Meeting 2000 Livorno (Italy) (pg. 171 - 175 ) ICRAM, ARPAT & SFI Walker P.A. , Heessen H.J.L. . Long-term changes in ray populations in the North Sea , ICES Journal of Marine Science , 1996 , vol. 53 (pg. 1085 - 1093 ) Google Scholar Crossref Search ADS WorldCat Author notes 1 Tel: +34 971611722; fax: +34 971611761. © 2003 International Council for the Exploration of the Sea

Junquera,, S;Román,, E;Morgan,, J;Sainza,, M;Ramilo,, G

doi: 10.1016/S1054-3139(03)00073-0pmid: N/A

Abstract Evidence for a prolonged ovarian development phase in Greenland halibut is presented. The reproductive cycle in this species was originally described based on the assumption that this phase should last about one year. The results of the present study, which involves data series covering a long time period and different geographic areas, show instead that there is more than one year between the mean age of the females that are at the onset of ovarian development and the mean age of the females that are actually spawning. There are two possible interpretations for this observation. One is that the ovarian development phase (vitellogenesis) could last more than one year and thus as a consequence, individual spawning would not necessarily occur on an annual basis. The other would be the existence of a high proportion of non-spawning females every year for other reasons not related with the natural rhythm of oocyte development. Introduction The cyclical change in the ovary during the development and maturation process is essentially similar in all oviparous teleosts. Eggs spawned in one year develop from a reservoir of resting or non-vitellogenic oocytes. During development they are supplied with yolk by the follicular cells during vitellogenesis. In most species, vitellogenesis ceases once oocytes reach their fully developed size and such oocytes eventually undergo maturation and ovulation after appropriate hormonal stimulation (Wallace and Selman, 1981). Oocyte maturation involves nuclear migration and hydration (Fulton, 1898). After maturation is complete, the oocyte is ovulated into the ovarian lumen and is ready to be spawned. The time of spawning and the duration of the reproductive cycle is genetically controlled, although it has been shown experimentally that it can be modified by environmental variation. The majority of fish species outside the tropics show seasonal reproductive cycles that are maintained for as long as the fish is reproductively competent (Breder and Rosen, 1966; Woodhead, 1979), and most usually complete the reproductive cycle within one year. Greenland halibut is considered to be a determinate, group synchronous spawning species, based on the existence of an hiatus separating the advance yolked oocytes, (Gundersen et al., 1999; Junquera et al., 1999). The fact that atresia rates show the highest values at the early vitellogenic stages (Junquera et al., 1999; Tuene et al., 2002) is also in support of this perception. This means that a single group of oocytes develops through vitellogenesis and matures to be spawned, without recruitment of any new group from the reservoir of non-vitellogenic oocytes. Up to now what was known about the Greenland halibut reproductive biology suggested the possibility that this species does not match the perception of a regular annual cycle of sexual maturity. Morgan and Bowering (1997) pointed out irregularities in the maturation process in the Northwest Atlantic Greenland halibut, possibly caused by variability in the distribution of adult fish, that hinders a precise estimation of maturity at age and size. Also, high variability in the main spawning seasons among areas and even within the same area is extensively documented. Besides the apparent lack of a clear seasonality in spawning, another outstanding characteristic in the Northwest Atlantic Greenland halibut population is that typically few mature (ripe) individuals are caught in either research surveys or commercial catches. To explain this several hypotheses have been proposed, including: spawning migrations of mature fish to areas not sampled; and misinterpretation of the maturity stage of the ovaries (Walsh and Bowering, 1981); a maturation cycle of more than one year (Jorgensen and Boje, 1994); an unsuitable temperature regime that leads to a high frequency of gonad resorption (Jorgensen and Boje, 1994) and population asynchrony in the maturation process (Fedorov, 1968; Junquera, 1994). In this study the hypothesis of a prolonged adolescent phase as a possible explanation of the peculiarities observed in the Greenland halibut sexual cycle is considered. The adolescent phase is the time span between the age a female changes from immature to adult and the age when a female actually spawns. Material and methods Spawning seasonality Data used in this part of the study have been recorded by scientific observers on board commercial deep-water trawlers in the Northwest Atlantic (NAFO Divisions 3LMN) during the period 1990–2000 (Table 1A). The sampling method in the case of maturity data is a length-stratified sampling from the catch. Maturity stages are assigned after macroscopic examination, based on a simple four-point scale (Junquera and Zamarro, 1994). In females those stages are: (1) immature (juvenile); (2) Adult resting or developing the ovaries at any level; (3) Fully hydrated (imminent spawning); and (4) Recent postspawning. This scale is intended to avoid as much as possible, the use of stages based on structures not visible to the naked eye, to which assignment is highly subjective. However the differentiation between immature (stage 1) from the adult stages (2 onwards) proved to be frequently uncertain, especially in the case of the first maturing females. But the aim of this part of the study is just to separate the stage of final growth of the oocytes, the duration of which is presently unknown, from the hydrated stage that indicates imminent spawning. This stage is perfectly identifiable to the naked eye due to the large size attained by the ooctytes (4–5 mm). The proportions of the fully hydrated females over the total adult females, first per sample and then accumulated to the corresponding monthly proportions have been obtained. Though in principle the total number of adults per sample would be here the total of females in stage 2 onwards, due to the uncertainty associated to the distinction between juveniles and adults in certain cases, inherent to this method of diagnosis, it was considered as a more accurate alternative to establish a separation threshold at the length of 50% maturity. Based on previous microscopic estimates on female Greenland halibut length at 50% maturity (Junquera et al., 1999), females were considered as “adult” from a total length of 67 cm. Table 1 Greenland halibut sampling description. (A) Northwest Atlantic (NAFO Divisions 3LMN) commercial sampling (scientific observers) from the Spanish deep-water bottom trawlers, analysed only macroscopically for maturity diagnosis. TF, total females sampled; AF, total adult females on the samples. (B) Number of females microscopically analysed in the NAFO area. A, number of females classified as adolescent; SP, number of females classified as spawning; C, commercial samples; R1, EU research survey in Flemish Cap (NAFO Division 3M); R2, EU selectivity research survey in NAFO Divisions 3LM; R3, Spanish long-line research survey in the same area; R4, Spanish research survey in the Grand Bank (NAFO Division 3N). (C) Number of females microscopically analysed in the Northeast Atlantic (ICES Subarea II). R5, autumn Spanish research survey. (D) Number of females macroscopically analysed from the autumn Canadian research survey (R6) in the NAFO area. . TF . AF . Sampling . . (A) Commercial sampling in NAFO Divisions 3LMN 1990 33581 7534 Monthly since May 1991 107227 13670 Monthly 1992 164818 15740 Monthly 1993 83340 4787 Monthly 1994 40136 6003 Monthly 1995 1459 133 Only January 1996 5497 853 Monthly 1997 3164 692 Monthly 1998 8242 1076 Monthly 1999 6889 973 Monthly 2000 11285 999 Monthly Total A SP Source (B) Sampling for microscopic analysis in NAFO Divisions 3LMN 1991 150 82 7 C 1992 250 58 93 C 1993 130 28 57 C 1994 433 126 66 C-R1 1995 90 27 18 R2 1996 526 95 111 C-R3-R1 1999 286 119 10 C-R4 2000 1054 409 23 C-R4 (C) Sampling for microscopic analysis in ICES Subarea II 1997 160 27 78 R5 1999 134 29 29 R5 2000 284 36 127 R5 (D) Research survey sampling in NAFO Subareas 0, 2 and Divisions 3KL 1978 191 70 26 R6 1979 131 74 1 R6 1980 201 162 6 R6 1981 547 320 97 R6 1982 208 187 21 R6 1984 113 16 8 R6 1985 101 3 13 R6 1986 259 205 26 R6 1988 192 75 1 R6 1990 24 22 0 R6 1991 160 14 0 R6 . TF . AF . Sampling . . (A) Commercial sampling in NAFO Divisions 3LMN 1990 33581 7534 Monthly since May 1991 107227 13670 Monthly 1992 164818 15740 Monthly 1993 83340 4787 Monthly 1994 40136 6003 Monthly 1995 1459 133 Only January 1996 5497 853 Monthly 1997 3164 692 Monthly 1998 8242 1076 Monthly 1999 6889 973 Monthly 2000 11285 999 Monthly Total A SP Source (B) Sampling for microscopic analysis in NAFO Divisions 3LMN 1991 150 82 7 C 1992 250 58 93 C 1993 130 28 57 C 1994 433 126 66 C-R1 1995 90 27 18 R2 1996 526 95 111 C-R3-R1 1999 286 119 10 C-R4 2000 1054 409 23 C-R4 (C) Sampling for microscopic analysis in ICES Subarea II 1997 160 27 78 R5 1999 134 29 29 R5 2000 284 36 127 R5 (D) Research survey sampling in NAFO Subareas 0, 2 and Divisions 3KL 1978 191 70 26 R6 1979 131 74 1 R6 1980 201 162 6 R6 1981 547 320 97 R6 1982 208 187 21 R6 1984 113 16 8 R6 1985 101 3 13 R6 1986 259 205 26 R6 1988 192 75 1 R6 1990 24 22 0 R6 1991 160 14 0 R6 Open in new tab Table 1 Greenland halibut sampling description. (A) Northwest Atlantic (NAFO Divisions 3LMN) commercial sampling (scientific observers) from the Spanish deep-water bottom trawlers, analysed only macroscopically for maturity diagnosis. TF, total females sampled; AF, total adult females on the samples. (B) Number of females microscopically analysed in the NAFO area. A, number of females classified as adolescent; SP, number of females classified as spawning; C, commercial samples; R1, EU research survey in Flemish Cap (NAFO Division 3M); R2, EU selectivity research survey in NAFO Divisions 3LM; R3, Spanish long-line research survey in the same area; R4, Spanish research survey in the Grand Bank (NAFO Division 3N). (C) Number of females microscopically analysed in the Northeast Atlantic (ICES Subarea II). R5, autumn Spanish research survey. (D) Number of females macroscopically analysed from the autumn Canadian research survey (R6) in the NAFO area. . TF . AF . Sampling . . (A) Commercial sampling in NAFO Divisions 3LMN 1990 33581 7534 Monthly since May 1991 107227 13670 Monthly 1992 164818 15740 Monthly 1993 83340 4787 Monthly 1994 40136 6003 Monthly 1995 1459 133 Only January 1996 5497 853 Monthly 1997 3164 692 Monthly 1998 8242 1076 Monthly 1999 6889 973 Monthly 2000 11285 999 Monthly Total A SP Source (B) Sampling for microscopic analysis in NAFO Divisions 3LMN 1991 150 82 7 C 1992 250 58 93 C 1993 130 28 57 C 1994 433 126 66 C-R1 1995 90 27 18 R2 1996 526 95 111 C-R3-R1 1999 286 119 10 C-R4 2000 1054 409 23 C-R4 (C) Sampling for microscopic analysis in ICES Subarea II 1997 160 27 78 R5 1999 134 29 29 R5 2000 284 36 127 R5 (D) Research survey sampling in NAFO Subareas 0, 2 and Divisions 3KL 1978 191 70 26 R6 1979 131 74 1 R6 1980 201 162 6 R6 1981 547 320 97 R6 1982 208 187 21 R6 1984 113 16 8 R6 1985 101 3 13 R6 1986 259 205 26 R6 1988 192 75 1 R6 1990 24 22 0 R6 1991 160 14 0 R6 . TF . AF . Sampling . . (A) Commercial sampling in NAFO Divisions 3LMN 1990 33581 7534 Monthly since May 1991 107227 13670 Monthly 1992 164818 15740 Monthly 1993 83340 4787 Monthly 1994 40136 6003 Monthly 1995 1459 133 Only January 1996 5497 853 Monthly 1997 3164 692 Monthly 1998 8242 1076 Monthly 1999 6889 973 Monthly 2000 11285 999 Monthly Total A SP Source (B) Sampling for microscopic analysis in NAFO Divisions 3LMN 1991 150 82 7 C 1992 250 58 93 C 1993 130 28 57 C 1994 433 126 66 C-R1 1995 90 27 18 R2 1996 526 95 111 C-R3-R1 1999 286 119 10 C-R4 2000 1054 409 23 C-R4 (C) Sampling for microscopic analysis in ICES Subarea II 1997 160 27 78 R5 1999 134 29 29 R5 2000 284 36 127 R5 (D) Research survey sampling in NAFO Subareas 0, 2 and Divisions 3KL 1978 191 70 26 R6 1979 131 74 1 R6 1980 201 162 6 R6 1981 547 320 97 R6 1982 208 187 21 R6 1984 113 16 8 R6 1985 101 3 13 R6 1986 259 205 26 R6 1988 192 75 1 R6 1990 24 22 0 R6 1991 160 14 0 R6 Open in new tab Duration of the ovary development phase The material used in this part of the study is listed in Table 1B, C and D. Data in Table 1B comes from the following sources: Spanish commercial deep-water trawlers in the Northwest Atlantic (NAFO Divisions 3LMN) (1991–1996, 1999–2000); Spanish annual research surveys on the Grand Bank of Newfoundland (NAFO Division 3N, 1999–2000); Spanish long-line research survey in the Northwest Atlantic (NAFO Divisions 3LM, 1996); European Union selectivity research survey in the same area (1995) and European Union annual research surveys on Flemish Cap (NAFO Division 3M, 1994 and 1996). The method used is an indirect approach that consists of measuring the distance in time between the age frequency distributions of the females at the adolescent stage and females ready to spawn. A similar method has been applied by Everson (1994) to Notothenia coriiceps. For this component of the study, maturity stages were determined by microscopic examination of ovary sections prepared using conventional histological processing and H&E staining (Junquera et al., 1999). The classification into maturity stages used here is an update of the one originally developed by Fedorov (1968), where Greenland halibut oogenesis is divided into oocyte growth (development), maturation and ovulation (Guraya, 1986). The presence/absence of oocytes in cortical alveoli, vitellogenesis, nuclear migration, hydration, previtellogenic and vitellogenic atresia, and postovulatory follicles has been recorded in each of the ovary sections, following the classification of Wallace and Selman (1981) and West (1990), and the photographic description of these stages in this species given by Fedorov (1968) and Walsh and Bowering (1981). Females were classified as “adolescent” when they were at either the cortical alveoli stage or any level of vitellogenesis, without signs of previous spawning, such as postovulatory follicles or vitellogenic atresia. Females were classified as “spawning” when they were either in the nuclear migration or hydration stages. All females analysed in this part of the study have been aged directly using their otoliths. To examine whether the observed maturity patterns are just local from this particular area of the Northwest Atlantic or could be generalized as a species characteristic, the geographic scope has been enlarged by analysing samples collected during the annual autumn Spanish research surveys in the Northeast Arctic (ICES Subarea II) in 1997, 1999 and 2000 (Table 1C). These samples were examined histologically as described earlier. In order to have a longer time perspective, data from the Canadian research surveys in NAFO Subareas 0, 2, and 3 during the 1978–1991 period have been included (Table 1D). These surveys were usually conducted annually in the NAFO Division 2J3KL area, and occasionally in the other Divisions. In 1987, 1989 and since 1991, no spawning females were found in these surveys. Only data from years in which females with hydrated eggs were observed are included. In this case, females were classified into maturity stages by the macroscopic approach. The maturity scale in this data set includes many stages, but among them only one (defined as “no evidence of previous spawning, new opaque eggs for spawning in the next year visible to the naked eye”) has been selected as a proxy of the adolescent stage as histologically described. It is assumed that this method of determining adolescent fish will not be as accurate as the histological method since it is more difficult to detect evidence of previous spawning macroscopically. As a spawning stage proxy, three macroscopic stages have been chosen, defined as follows: (i) “Opaque and clear eggs present with less than 50% of the volume being clear eggs; maturing to spawn in the present year” (ii) “50% or more of the volume are clear eggs; this stage also includes the ripe condition where the ovarian content is almost liquid with clear eggs to spawn or spawning in the present year” and (iii) “Partly spent, ovary not full as in the previous stage; some eggs extruded but many clear eggs remaining”. Females from this data set were also aged individually using otoliths. Results In the Northwest Atlantic the Greenland halibut spawning activity is irregularly distributed over the year (Figure 1). The low monthly proportions of spawning females, only occasionally exceeding 25% (in numbers) of the sampled adult females, is also striking. In the Northeast Atlantic (Table 1C) the proportion of spawning females in the single month sampled (October of 1997, 1999, 2000) is substantially higher (63% of the adult females in 1997, 34% in 1999 and 68% in 2000), however, no inference can be made on the seasonal pattern of occurrence due to the lack of year round sampling in this area. Figure 1 Open in new tabDownload slide Monthly proportions of Greenland halibut spawning females (hydrated stage) during the period 1990–2000, based on commercial sampling (Table 1A) in the Northwest Atlantic (NAFO Divisions 3LMN). In 1995 only data from January are available. Figure 1 Open in new tabDownload slide Monthly proportions of Greenland halibut spawning females (hydrated stage) during the period 1990–2000, based on commercial sampling (Table 1A) in the Northwest Atlantic (NAFO Divisions 3LMN). In 1995 only data from January are available. Frequency distributions at age of the maturity stages for the sample examined by histology show that adolescent (A) females appear for the first time at age 7 whereas spawning (SP) ones start to appear at age 10 in the Northwest Atlantic (Figure 2A). In the data set in which only the macroscopic approach has been used (Figure 2B), it is not possible to distinguish unambiguously the adolescent stage, but the spawning (hydrated) one can be determined easily by visual diagnosis. The stage used as a proxy of the adolescent one in this case appears first at age 5, while hydrated females first occur from age 8. However, in this data set the frequency distribution at age of both stages largely overlaps. In the Northeast Atlantic (Figure 2C) the adolescent stage appears first at age 6 and the spawning one at age 8. Figure 2 Open in new tabDownload slide Percentages at age of Greenland halibut females in adolescent (□) and spawning (■) stages. Northwest Atlantic (A) and Northeast Atlantic (C) panels have been obtained by microscopic examination of the ovaries. Northwest Atlantic (B) is based on a macroscopic classification. Figure 2 Open in new tabDownload slide Percentages at age of Greenland halibut females in adolescent (□) and spawning (■) stages. Northwest Atlantic (A) and Northeast Atlantic (C) panels have been obtained by microscopic examination of the ovaries. Northwest Atlantic (B) is based on a macroscopic classification. In order to compare the respective ‘A’ and ‘SP’ age distributions in the three data set analysed, a non-parametric two-sample Kolmogorov–Smirnov test has been applied. It showed that the differences between ‘A’ and ‘SP’ distributions are significant in all the three cases (KS=0.50; n=16; p<0.05 in (A); KS=0.56; n=16; p<0.05 in (B) and KS=0.31; n=16; p<0.001 in (C)). The linear trends of the cumulative frequency distribution of the maturity stages at age were produced to obtain an average difference in time between the adolescent and spawning stages (Figure 3). In both the Northwest and Northeast Atlantic Greenland halibut, a distance of four years is observed, though in the latter area both the age to become adult and to achieve spawning are younger. Figure 3 Open in new tabDownload slide Cumulative proportions at age of Greenland halibut females at the adolescent and spawning stages, and respective linear trends (only results from microscopic analysis are included). —adolescent; •—spawning. Figure 3 Open in new tabDownload slide Cumulative proportions at age of Greenland halibut females at the adolescent and spawning stages, and respective linear trends (only results from microscopic analysis are included). —adolescent; •—spawning. Discussion The results show an interval of about 4 years between when female Greenland halibut became adult and when they are actually ready for spawning. Since a similar pattern is observed on both sides of the Atlantic and across a fairly long time period, it might be considered as a characteristic of the species. The data set where the macroscopic approach for ovary staging was used gives only limited support to this conclusion, since the frequency distributions at age of the “proxy-adolescent” and spawning stages show substantial overlap. However it must be noted that while this later stage is clearly distinguishable by eye, the adolescent stage is impossible to define precisely without histological examination, since it requires the ability to distinguish adult females that never spawned before from adult repeat spawner resting females, and also from the immature. Nevertheless even in this case, a three year difference between the starting point of the two stages is observed, consistent with the results produced by the microscopic approach. Examples of species with lengthy sexual cycles are not uncommon. Two years to complete vitellogenesis are reported in south Atlantic grenadier (Alekseeva and Alekseev, 1984) and four years in Nototheniids and other Antarctic fishes (Shandikov and Faleeva, 1992; Everson 1994). Large proportions of non-reproductive adults in orange roughy have been documented (Bell et al., 1992). Also in Greenland halibut the idea of a lengthy reproductive cycle is not new. Fedorov (1971) concluded from visual and histological analysis of gonads that some sexually mature females in the Barents Sea Greenland halibut do not spawn annually and that the pause in reproduction could last for at least two years, with ovarian development halted in the initial stage. The situations of non-annual spawning can be classified as two types: (a) spawning is abnormally skipped due to some adverse condition which prevents the final development of the gametes. In this situation, massive resorption of sexual products occurs, and (b) species whose natural sexual cycle extends for several years. According to Fedorov (1971) this category usually includes long-lived species, with irregular spawning, inhabiting areas with extreme climatic conditions. Based on previous results (Junquera et al., 1999), atresia rates in the Northwest Atlantic Greenland halibut peak at initial vitellogenesis, as also pointed out Fedorov (1971), and become virtually absent in fully yolked oocytes prior to spawning. This should not be regarded always as a response to adverse conditions or a mechanism for spawning cancellation, but as a mechanism of fecundity regulation. Although in experimental conditions massive prespawning atresia is a common response to environmental stress, this is an unlikely event in physiologically normal females under natural conditions (Wallace and Selman, 1981). Oocytes resorbed at early development stages can be used later to produce interstitial gland cells (Guraya, 1972; Saidapur, 1978), thus constituting a feed back element in the normal cycle of the ovary development. But on the contrary, for determinate spawners it is a waste of energy to fully develop a significant number of oocytes that will not be spawned. If the process of vitellogenesis lasts up to 4 years, the question arises as to whether individual females spawn only once every 4 years or if they are able to spawn at shorter time intervals after their first spawning. If spawning normally occurs only once every 4 years, this would mean that only about 25% of the adult females would actually spawn per year, which is in agreement with the results from the Northwest Atlantic for most of the time series analysed. However, the proportion of spawning females found in the Northeast Atlantic is much larger than this. Sampling in this case was conducted in a known spawning area of the stock, and in a season close to the peak spawning, (Albert et al., 1998), where probably only the reproductively active adults would be present. The ratio of males/females in this survey area was 4:1, which is a very unusual one out of the breeding season, and supports the idea that the sampling took place in a spawning area, during the spawning season. In addition, Bowering and Nedreaas (2000) reported the existence of seasonal migrations of the mature fraction in this stock to the spawning areas. This could mean that the observed proportion of spawning females in this area was not reflective of the whole stock but rather that it was inflated by the absence of non-spawning females in the area. The prolonged vitellogenesis process also contributes to explaining the irregularity or lack of a clear-cut seasonality in the Northwest Atlantic Greenland halibut spawning. The timing of reproduction in female teleosts may be viewed as the product of numerous biotic and abiotic stimuli which exert both long-term effects on ovarian growth and short-term effects on final maturation and ovulation of the oocytes (Stacey, 1989). It is known experimentally that the oocyte maturation switch in Pleuronectiform species is not a simple consequence of oocyte development (vitellogenesis) being completed. Maturation and subsequent spawning would not be undertaken in the absence of specific endogenous stimuli. Individuals would be arrested at the vitellogenic stage for as long as appropriate conditions are met (Bone et al., 1995). In contrast to the Northwest Atlantic, spawning of Greenland halibut in the Northeast Atlantic is more synchronized in time, since the peak always occurs in the last quarter of the year (Albert et al., 1998). According to Bowering and Nedreaas (2000) in the Northwest Atlantic spawning occurs at considerably greater depths (beyond 1000 m) than in the Northeast Atlantic (between 500 and 800 m). It is possible therefore that in those shallower depths a variable photoperiod could act as a cue in switching the final maturation, and leading to a clear seasonality in spawning. Under a constant photoperiod, as is the case of spawning in deeper waters, the lack of seasonal cues may lead to less synchronous spawning within the population. Though there is the overwhelming perception that even in the constant environment of the deep seas fish species maintain circannual cycles in reproductive activity, the Northwest Greenland halibut case seems not to support this view. References Albert O. T. , Nilssen E. M. , Stene A. , Gundersen A. , Nedreaas K. H. . Spawning of the Barents Sea/Norwegian Sea Greenland halibut (Reinhardtius hippoglossoides) , 1998 ICES Council Meeting 1998/O: 22 Alekseeva Ye.I. , Alekseev F.Ye. . Polovye tskly ryb v izuchenii struktury vida i funktsional'noi struktury areala. The sexual cycles of fish in a study of the structure of a species and the functional structure of the range , Vnutrividovaya fifferentsiatsiya morskikh promyslovykh ryb i bespozvonochynkh , 1994 (pg. 23 - 38 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bell J.D. , Lyle J.M. , Bulman C.M. , Graham K.J. , Newton G.M. , Smith D.C. . Spatial variations in reproduction and occurrence of non-reproductive adults in orange roughy, Hoplostethus atlanticus Collet (Trachichthydae), from south-eastern Australia , Journal of Fish Biology , 1992 , vol. 40 (pg. 107 - 122 ) Google Scholar Crossref Search ADS WorldCat Bone Q. , Marshall N.B. , Blaxter J.H.S. . , Biology of Fishes , 1995 London Chapman and Hall 332 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bowering W.R. , Nedreaas K.H. . A comparison of Greenland halibut (Reinhardtius hippoglossoides (Walbaum)) fisheries and distribution in the Northwest and Northeast Atlantic , Sarsia , 2000 , vol. 85 (pg. 61 - 76 ) Google Scholar Crossref Search ADS WorldCat Breder C.M. , Rosen D.E. . , Modes of Reproduction in Fishes , 1996 New York Natural History Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Everson I. . Timescale of ovarian maturation in Notothenia coriiceps: evidence for a prolonged adolescent phase , Journal of Fish Biology , 1994 , vol. 44 (pg. 997 - 1004 ) Google Scholar Crossref Search ADS WorldCat Fedorov K.Ye. . Ovogenez i polovoi tsikli chernovo paltusa , Trudy PINRO , 1968 , vol. 23 (pg. 425 - 450 ) Oogenesis and the sexual cycle of the Greendland halibut. Canadian Fisheries and Aquatic Science Translation No. 4559, 1979 OpenURL Placeholder Text WorldCat Fedorov K.Ye. . The state of the gonads of the Barents Sea Greenland halibut (Reinhardtius hippoglossoides Walb.) in connection with failure to spawn , Journal of Ichthyology , 1971 , vol. 11 5 (pg. 673 - 683 ) OpenURL Placeholder Text WorldCat Fulton T.W. . On the growth and maturation on the ovarian eggs of teleostean fishes , Annual Report of the Fisheries Board of Scotland , 1898 , vol. 16 3 (pg. 88 - 124 ) OpenURL Placeholder Text WorldCat Gundersen A. , Kjesbu O.S. , Nedreaas K.H. . Fecundity of Northeast Arctic Greenland halibut (Reinhardtius hippoglossoides) , Journal of Northwest Atlantic Fisheries Science , 1999 , vol. 25 (pg. 29 - 36 ) Google Scholar Crossref Search ADS WorldCat Guraya S.S. . Histochemical observations on the interstitial gland cells of dogfish ovary , General and Comparative Endocrinology , 1972 , vol. 18 (pg. 409 - 412 ) Google Scholar Crossref Search ADS PubMed WorldCat Guraya S.S. . The cell and molecular biology of fish oogenesis , Monographs of Developmental Biology , 1986 , vol. 18 London Karger 223 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jorgensen O. , Boje J. . Sexual maturity of Greendland halibut in NAFO Subarea 1 , 1994 NAFO Scientific Council Research Document, 94/42. Serial Number N2412. 17 pp Junquera S. . Analysis of the variations in the spatial distribution and spawning of the Greendland halibut in divisions 3LMN (1990–93) , 1994 NAFO Scientific Council Research Documents, 94/25. Serial Number N2391. 12 pp Junquera S. , Román E. , Paz X. , Ramilo G. . Changes in Greenland halibut growth, condition and fecundity in the Northwest Atlantic (Flemish Pass, Flemish Cap and Southern Grand Bank) , Journal of Northwest Atlantic Fishery Science , 1999 , vol. 25 (pg. 17 - 28 ) Google Scholar Crossref Search ADS WorldCat Junquera S. , Zamarro J. . Sexual maturity and spawning of Greenland halibut (Reinhardtius hippoglossoides) from Flemish Pass Area , NAFO Scientific Council Studies , 1994 , vol. 20 (pg. 47 - 522 ) OpenURL Placeholder Text WorldCat Morgan M.J. , Bowering W.R. . Temporal and geographic variation in maturity at length and age of Greenland halibut (Reinhardtius hippoglossoides (Walbaum)) from the Canadian Northwest Atlantic with implications for fisheries management , ICES Journal of Marine Science , 1997 , vol. 54 (pg. 875 - 885 ) Google Scholar Crossref Search ADS WorldCat Saidapur S.K. . Follicular atresia in the ovaries of non mammalian vertebrates , International Review of Cytology , 1978 , vol. 54 (pg. 225 - 241 ) Google Scholar Crossref Search ADS PubMed WorldCat Shandikov G.A. , Faleeva T.I. . Features of gametogenesis and sexual cycles of six notothenioids fishes from East Antarctica , Polar Biology , 1992 , vol. 11 (pg. 615 - 621 ) Google Scholar Crossref Search ADS WorldCat Stacey N.E. . Potts G.W. , Wootton R.J. . Control of the timing of ovulation by exogenous and endogenous factors , Fish Reproduction—Strategies and Tactics , 1989 London Academic Press (pg. 207 - 222 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tuene S. , Gundersen A. , Emblem W. , Fossen I. , Boje J. , Steingrund P. , Ofstad L.H. . Maturation and occurrence of atresia of Greenland halibut (Reinhardtius hippoglossoides W.) in the waters of East Greenland, Faroe Island and Hatton Bank , Temanord , 2002 , vol. 519 (pg. 39 - 71 ) OpenURL Placeholder Text WorldCat Wallace R.A. , Selman K. . Cellular and dynamic aspects of oocyte growth in teleosts , American Zoologists , 1981 , vol. 21 (pg. 325 - 343 ) Google Scholar Crossref Search ADS WorldCat Walsh S.J. , Bowering W.R. . Histological and visual observations on Oogenesis and sexual maturity in Greenland halibut off Northern Labrador , NAFO Scientific Council Studies , 1981 , vol. 1 (pg. 71 - 75 ) OpenURL Placeholder Text WorldCat West G. . Methods of assessing ovarian development in fishes: a review , Australian Journal of Marine and Freshwater Research , 1990 , vol. 41 2 (pg. 199 - 222 ) Google Scholar Crossref Search ADS WorldCat Woodhead A.D. . Senescence in fishes , Symposium of the Zoological Society of London , 1979 , vol. 44 (pg. 179 - 205 ) OpenURL Placeholder Text WorldCat © 2003 International Council for the Exploration of the Sea

Figueiredo,, I;Bordalo-Machado,, P;Reis,, S;Sena-Carvalho,, D;Blasdale,, T;Newton,, A;Gordo,, L.S

doi: 10.1016/S1054-3139(03)00064-Xpmid: N/A

Abstract Black scabbardfish is a deep-water species, common in the NE Atlantic at depths between 450 and 1300 m, currently exploited by some European countries. Between May 1998 and April 2000, specimens collected at three different locations in the NE Atlantic—NW of Scotland, Sesimbra (mainland Portugal) and Funchal (Madeira)—were analysed. The evolution of maturity of both sexes throughout the year was studied based on the macroscopic and microscopic analysis of the gonads. Specimens with the largest total length were found in Funchal, whereas the smallest size was recorded in the NW of Scotland. Neither spawners nor post-spawners were ever observed in NW of Scotland and Sesimbra. In Sesimbra, only a few individuals attained pre-spawning stage and most of the early developing females exhibited atresia in their ovaries. In Funchal, all the maturity stages were found; spawners occurred from September to December (females) and from August to December (males). Length of first maturity for females was estimated to be around 1000 mm. Two groups of spawners with different sizes were observed during the spawning period off Madeira. Introduction Black scabbardfish (Aphanopus carbo Lowe, 1839) is a deep-water species of the family Trichiuridae. It has a world-wide distribution, with records in the NW and NE Atlantic, from Iceland to the south of Madeira Island (Anon., 1984; Gordon, 1986; Merrett et al., 1991), southern Indian Ocean (Piotrovskii, 1979) and north-western Pacific (Clarke and Wagner, 1976). This species occurs from 200 m depth around British Isles (Tucker, 1956) to 1800 m to the south of Madeira Island (Martins et al., 1987). The black scabbardfish is the target species of two Portuguese longline fisheries: the long established Madeira fishery (since the early 19th century) and the more recent fishery off the Portuguese mainland (since 1983). In northern European areas, the growing landings of this species result from multi-species deep-water trawl fisheries mainly from France and United Kingdom. Despite the increasing commercial interest in this species, little is known about its life cycle. The existing contributions on the reproduction usually allude to a short analysis of maturity and to the size range of captured specimens. In the waters to the north of the British Isles, the majority of caught specimens were immature or in an intermediate stage of maturity (Kelly et al., 1998); there is only reference to two individuals caught at the Porcupine Bank in January with ripe gonads (Ehrich, 1983). Specimens in a spent condition were found in Icelandic waters between January and March (Magnússon and Magnússon, 1995). No spawners were ever observed off the Portuguese continental coast (Machado et al., 1998; Anon., 2000). Specimens in pre-spawning and ripe condition were observed in Madeira waters between September and February (Carvalho, 1988; Anon., 2000) and in Azorean waters in August (Anon., 2000). The main objective of this article is to examine recent data on maturity from three distinct areas in the NE Atlantic—NW Scotland, Sesimbra (Portuguese continental coast) and Funchal (Madeiran waters) (Figure 1)—focusing the reproductive behaviour of this species and discussing its reproductive strategies. Figure 1 Open in new tabDownload slide NE Atlantic areas (black arrows) from where black scabbardfish samples were obtained. Figure 1 Open in new tabDownload slide NE Atlantic areas (black arrows) from where black scabbardfish samples were obtained. Methods Fish samples were obtained from bottom longline landings at Sesimbra (June 1998–April 2000) and Funchal (May 1998–December 1999) fishing ports and from bottom trawls held by the Marine Laboratory, Aberdeen (MARLAB) during research surveys (September 1998, October 1998 and March 1999). In the laboratory, all individuals were measured to the nearest millimetre and later grouped in 10 mm total length (TL) classes. Individual total weight was also recorded to the nearest 1 g and their gonads removed and weighted to the nearest 0.01 g. Macroscopic maturity stages were assigned to specimens using the maturity scale defined by Gordo et al. (2000). Histological sections of gonads were also made, especially in cases where the assignment of maturity stages was ambiguous. The preparation of sections included the preservation of gonads in Bodian's AFA (Lillie and Fulmer, 1976) during a period of time between 1 and 6 days, depending on their thickness. They were later embedded in paraffin wax, sectioned at 4–9 μm and stained using Masson trichrome (females) and Heidenhain Azan (males). Gonadosomatic index (GSI), which expresses gonad weight as a percentage of total weight, was estimated by month, for each sex (only for Funchal and Sesimbra, since no data on gonad weight of specimens from NW Scotland were available). Sex ratio was calculated by month as the fraction of the number of females in relation to the total number of individuals sampled. Based on the fraction of mature specimens (stages III, IV and V) per length class, the maturity ogive for the period where spawners occurred was estimated adjusting the simple logistic model (Zar, 1996) expressed by where M is the proportion of mature specimens at total length class L and β0 and β1 are model parameters. Incidence of atresia estimated as the proportion of atretic oocytes in relation to the total number of early-developed oocytes was also determined by month in specimens from Sesimbra. Results A total of 2443 fish were sampled in the three NE Atlantic areas: Funchal (1249), Sesimbra (826) and NW of Scotland (368). Females were predominant in the samples, with the exception of those collected in Funchal in July and December. Specimens sampled in the NW of Scotland (length range: 612–1175 mm) were smaller than those from Sesimbra (length range: 667–1365 mm) and Funchal (length range: 712–1510 mm). In the NW of Scotland, individuals were either in maturity stages I or II. The majority of males were in maturity stage I (87%), while females were more frequently found in stage II (52%) with the exception of October where stage I was more common (64%). These two maturity stages were almost exclusively observed in both males and females from Sesimbra, where no specimens in stage IV or V were found. In this area, some individuals began their gonadal development in July reaching maturity stage III in August—2% of the total number of specimens. However, between December and April, the majority of females in stage II showed a clear increase in the incidence of atresia (from 9.5 to 25%) in early-developed oocytes (Figure 2). In contrast to the other regions, all the maturity stages were recorded in Funchal for both sexes. Stage II (Figure 2) was found throughout the year, being more common between March and April. Stage III appeared in males mostly in May, while in females it appeared later in July. Stage IV occurred mainly from September to December (females) and from August to December (males). For the remaining maturity stages, the very low occurrences (usually below 4%) of stage I for both males and females and the long duration period of stage V (from November to June) must be pointed out. Figure 2 Open in new tabDownload slide Histological sections of stage II ovaries from a specimen sampled in Funchal and Sesimbra. In the section from Sesimbra some atretic oocytes are indicated by an arrow. Figure 2 Open in new tabDownload slide Histological sections of stage II ovaries from a specimen sampled in Funchal and Sesimbra. In the section from Sesimbra some atretic oocytes are indicated by an arrow. Spawning should occur preferentially during the last quarter of the year, where the highest GSI values were registered (Figure 3). Since insufficient data on maturity were available for males, the maturity ogive was only adjusted to female data for the period between September and February, which coincides with the spawning period. Based on the estimates obtained, the length of first maturity is about 1028 mm (Figure 4). Frequent macroscopic stage assignment errors (later corrected by microscopic analysis) were registered in stages I and V. However, the analysis of GSI indicated that specimens in maturity stage V presented higher values than those in maturity stage I (Figure 5). Actually, the GSI values of stages I and V were significantly different (t-test; p-value <0.01). As a consequence, this index can be considered a useful and expedite tool to help in the differentiation of these two stages. Figure 3 Open in new tabDownload slide Maximum (–) median (•) and minimum (–) values of GSI estimated by month for black scabbardfish sampled in Funchal. Figure 3 Open in new tabDownload slide Maximum (–) median (•) and minimum (–) values of GSI estimated by month for black scabbardfish sampled in Funchal. Figure 4 Open in new tabDownload slide Maturity ogive of black scabbardfish females sampled in Funchal for the period between September and February. Figure 4 Open in new tabDownload slide Maturity ogive of black scabbardfish females sampled in Funchal for the period between September and February. Figure 5 Open in new tabDownload slide Distribution of GSI values of maturity stages I and V by month in Madeira. Figure 5 Open in new tabDownload slide Distribution of GSI values of maturity stages I and V by month in Madeira. The analysis of histological sections from females in maturity stages III and IV suggested that oocyte development is essentially group-synchronous. Indeed, only one group of well-developed and bigger size oocytes was identified before spawning. To further understand the reproductive strategy of this species, a comparative analysis of GSI values from Sesimbra and Funchal was carried out. This analysis was restricted to ranges of length and maturity stages common to both areas (stage II females with TL between 1100 and 1210 mm). The highest GSI values were observed in specimens from Funchal, namely, in April and July (Figure 6). In addition, after the analysis of stage IV females sampled versus TL, two different groups of spawners were identified: (i) individuals with TL smaller than 1250 mm, which spawn early in the spawning season (between September and December) and (ii) individuals with TL larger than 1250 mm, which spawn preferentially in January and February. This finding is further corroborated through the monthly distribution of GSI values of maturity stage V females (Figure 5), which showed an increase from October to December followed by a decreasing trend that lasted until May. In June, the variability of GSI values was high. Furthermore, the analysis of GSI values by length from March to June (Figure 7) revealed the existence of two groups of post-spawners: one occurring in June with individuals above 1300 mm TL and other from March to May exhibiting lengths predominantly below 1300 mm TL. Figure 6 Open in new tabDownload slide Monthly distribution of stage II GSI values of females with TL between 1100 and 1210 mm sampled in Sesimbra and Funchal. Figure 6 Open in new tabDownload slide Monthly distribution of stage II GSI values of females with TL between 1100 and 1210 mm sampled in Sesimbra and Funchal. Figure 7 Open in new tabDownload slide Variation of GSI values versus total length of stage V specimens. Figure 7 Open in new tabDownload slide Variation of GSI values versus total length of stage V specimens. Discussion The range of black scabbardfish total length varied between areas. Largest specimens were sampled off Madeira (above 1400 mm TL) while smallest ones were collected in the NW of Scotland (below 650 mm TL). Such length differences between northern and southern areas (separated by parallel 40°N) were also verified in a recent study based on an enlarged length dataset (Carvalho and Figueiredo, 2001). Geographic and fishing gear (trawl in the north and longline in the south) factors may be responsible for these differences although their individual contributions are difficult to disentangle. In all the areas, juveniles (TL below 900 mm) were scarce, particularly during the pre-recruitment phase. So far, there are only two records of small specimens (100 and 150 mm TL) that were found in the stomach of an Alepisaurus ferox Lowe, 1833, captured off Madeira (Maul, 1954). The macroscopic assignment of maturity stages was sometimes difficult probably due to the slow rate of gonadal development. However, the analysis of GSI values by maturity stage constituted important auxiliary information for a correct assignment of maturity stages since the ranges of GSI from different stages were different and did not overlap. Based on our results, black scabbardfish exhibits temporal sexual maturation differences according to region in the NE Atlantic. Vitellogenesis begins in Funchal and Sesimbra at the same time of the year, however, only the specimens from Funchal continue their gonadal development towards maturation and egg release. No spawners were found in Sesimbra and NW Scotland, while off the Portuguese mainland a small percentage of specimens reached the pre-spawning stage, in NW Scotland only initial development stages (I and II) were found. The predominance of pre-adult specimens at NW Scotland, either immature or in an early development stage not evolving to more developed maturity stages, might reflect insufficient levels of energy to proceed with gametogenesis. In Sesimbra, early maturing specimens larger than the length of first maturity (ca. 1028 mm) enter into an intense process of atresia after October with a maximum in March. Atresia affects oocytes in different developing stages and the reabsortion process occurs throughout the ovary. This suggests that although those specimens are potentially capable of reproducing, they do not enter into a spawning process and remain in a resting phase that can extend far beyond the spawning season. Possible reasons for the observed cessation of the maturation process could also be related to insufficient levels of accumulated energy or unfeasible prospects of a successful reproduction. This arrest in the maturation development due to low levels of energy accumulated has also been observed for the north European eel (Anguilla anguilla) population and related with delays in reproductive migrations (Svedäng and Wickström, 1997). In Madeiran waters, spawners occurred mainly from September to December, which is in agreement with earlier results (Carvalho, 1988; Anon., 2000). Two distinct reproductive strategies seem to be followed by the species in that area as different size specimens spawn at different time periods. While smaller size spawners occur between September and December, larger individuals preferentially undertake spawning in January and February. These findings were also corroborated by the existence of two female post-spawner groups during the spawning period. At more northern latitudes (Icelandic waters), spent individuals were recorded between January and March by Magnússon and Magnússon (1995), which suggests that the species may also reproduce in northern waters in areas not surveyed by the present study and slightly after the spawning season off Madeira. Despite the fact that we only analysed data from three locations, it seems that black scabbardfish presents different reproductive strategies according to the sampled area in the NE Atlantic. The maturity data collected at these areas also favour the existence of reproductive migrations to spawning areas. This is in agreement with some authors who have already suggested that the species undertakes horizontal migrations to spawning and nursery grounds (Geistdoerfer, 1982; Kelly et al., 1998; Anon., 2000). This study was supported by DG XIV of the European Commission through Study project 97/0084 entitled “Environment and biology of deep-water species Aphanopus carbo in NE Atlantic: basis for its management (BASBLACK)”. References Anon INIP—Programa de apoio às pescas na Madeira–III (in Portuguese) , 1984 Cruzeiro de reconhecimento de Pesca e Oceanografia 020330981. Relatórios, INIP. 22, 132 pp Anon Final report of the EU study project CT 97/0084 Environment and biology of deep-water species Aphanopus carbo in NE Atlantic: basis for its management (BASBLACK) , 2000 DGXIV European commission. 94 pp Carvalho D. . Relatório final do estudo efectuado sobre o Peixe-Espada Preto (Aphanopus carbo, Lowe, 1839) capturado na ZEE da Madeira (in Portuguese) , 1988 EC Report. DG XIV/CE Doc. No. XIV/B/1-1987. 177 pp Carvalho M. L. , Figueiredo I. . Establishing a sampling programme for monitoring changes in the length distribution of the black scabbardfish (Aphanopus carbo Lowe, 1839) in the Northern Atlantic Ocean , 2001 Proceedings of the Advanced Workshop on Environmental Sampling and Monitoring 22–24 March Estoril, Portugal Clarke T.A. , Wagner P.J. . Vertical distribution and other aspects of the ecology of certain mesopelagic fishes taken near Hawaii , Fishery Bulletin, U.S. , 1976 , vol. 74 (pg. 635 - 645 ) OpenURL Placeholder Text WorldCat Ehrich S. . On the occurrence of some fish species at the slopes of the Rockall Trough , Archiv für Fischereiwissenschaft , 1983 , vol. 33 (pg. 105 - 150 ) OpenURL Placeholder Text WorldCat Geistdoerfer P. . L'exploitation commerciale des poisons de grande profounder dans l'Atlantique Nord , Oceanis (Doc. Oceanogr.) , 1982 , vol. 8 (pg. 29 - 55 ) OpenURL Placeholder Text WorldCat Gordo L. S. , Carvalho D. S. , Figueiredo I. , Reis S. , Machado P. B. , Newton A. , Gordon J. . The sexual maturity scale of black scabbardfish: a macro- and microscopic approach , 2000 Celta Editora, Oeiras. 35 pp. ISBN 972-774-060-X Gordon J.D.M. . The fish populations of the Rockall Trough , Proceedings of the Royal Society of Edinburgh , 1986 , vol. 88B (pg. 191 - 204 ) OpenURL Placeholder Text WorldCat Kelly C. J. , Connolly P. L. , Clarke M. W. . The deep-water fisheries of the Rockall Trough; some insights gleaned from Irish survey data , 1998 ICES CM 1998/O: 40, 22 pp Lillie R.D. , Fulmer H.M. . , Histopathologic Technique and Practical Histochemistry , 1976 New York McGraw-Hill 942 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Machado P. , Martins R. , Figueiredo I. , Gordo L. S. . Some notes on the biology of the black scabbardfish , 1998 ICES CM 1998/O: 69 Magnússon J.V. , Magnússon J. . Hopper A.G. . The distribution, relative abundance, and the biology of the deep-sea fishes of the Icelandic slope and Reykjanes Ridge , Deep-water Fisheries on the North Atlantic Oceanic Slope , 1995 Netherlands Kluwer Academic (pg. 161 - 199 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Martins M. R. , Leite A. M. , Nunes M. L. . Peixe-espada-preto. Algumas notas ácerca da pescaria do peixe-espada-preto (in Portuguese) , 1987 Instituto Nacional de Investigação das Pescas (publicações avulsas). 14 pp Maul G.E. . Notes and exhibitions. [A sketch of Madeiran Ichthyology with observations on the ecology of the most important fishes] , Proceedings of the Zoological Society of London , 1954 , vol. 123 (pg. 901 - 903 ) OpenURL Placeholder Text WorldCat Merrett N.R. , Gordon J.D.M. , Stehmann M. , Haedrich R.L. . Deep demersal fish assemblage structure in the Porcupine Seabight (eastern North Atlantic): slope sampling by three different trawls compared , Journal of the Marine Biological Association of the United Kingdom , 1991 , vol. 71 (pg. 329 - 358 ) Google Scholar Crossref Search ADS WorldCat Piotrovskii A.S. . On the distribution of the black scabbardfish Aphanopus carbo (family Trichiuridae) in the Indian Ocean , Journal of Ichthyology , 1979 , vol. 19 5 (pg. 145 - 146 ) OpenURL Placeholder Text WorldCat Svedäng H. , Wickström H. . Low fat contents in female silver eels: indications of insufficient energetic stores for migration and gonadal development , Journal of Fish Biology , 1997 , vol. 50 3 (pg. 463 - 690 ) Google Scholar Crossref Search ADS WorldCat Tucker D.W. . Studies on the Trichiuroid fishes—3. A preliminary revision of the family Trichiuridae , Bulletin of the British Museum (Natural History) Zoology , 1956 , vol. 4 (pg. 73 - 131 ) OpenURL Placeholder Text WorldCat Zar J.H. . , Biostatistical Analysis , 1996 3rd edition Upper Saddle River, NJ Prentice-Hall Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2003 International Council for the Exploration of the Sea

Dutil,, Jean-Denis;Lambert,, Yvan;Chabot,, Denis

doi: 10.1016/S1054-3139(03)00024-9pmid: N/A

Abstract Atlantic cod were food-deprived for a period of 84 days at three temperatures (2, 6, 10°C), and changes in the liver, gonads and somatic weights, and muscle and liver water contents were monitored and compared with changes observed in wild cod over winter in the northern Gulf of St. Lawrence. Total lack of food during the period January–April would have caused condition to decline to a level at which very high mortality takes place. Actual changes in condition in wild cod were less than predicted from the laboratory experiments except during the period April–May at the onset of spawning. Thus, wild cod were able to meet part of the metabolic costs during winter through occasional feeding, as confirmed by stomach content data. We conclude that previous estimates of natural mortality associated with poor condition in spring were not biased by the selective mortality of poor-condition fish in winter. Introduction Following the severe decline in the abundance of Atlantic cod (Gadus morhua) in eastern Canada, natural mortality (M) has been estimated to be higher than previously considered (Sinclair, 2001). Fishery biologists generally assume a fixed value (0.2, 18%) for M. While M can be assessed readily in stocks, which are not exploited, partitioning M into its major components is not a trivial task. This is even more complex in situations where stocks are exploited, particularly in large ecosystems. Thus, we have a limited understanding of natural mortality, its variability in space and time and its response to changes in the ecosystem. Two aspects of natural mortality in juvenile and adult cod have been examined in association with groundfish decline: predation by seals (Hammill and Stenson, 2000) and starvation (Dutil and Lambert, 2000). Climatic variations and changes in marine ecosystems that have a negative impact on fish condition may also increase the risk of mortality. Declining condition may, for instance, decrease metabolic and swimming capacities of cod (Martínez et al., 2003) and hence their ability to seize prey and avoid predation. Such a situation has been assessed recently in the Gulf of St. Lawrence, where cod have experienced a period of changing environmental conditions. The cooling of the cold intermediate water layer in the Gulf of St. Lawrence started in the 1980s (Gilbert and Pettigrew, 1997). In the northern Gulf, cod moved to deeper waters and shifted their latitudinal distribution, possibly to avoid being exposed to colder waters (Castonguay et al., 1999). Nevertheless, both size-at-age and individual energetic condition declined throughout the 1980s and early 1990s (Dutil et al., 1999). Individual fish experience marked seasonal changes in condition both in the northern (Lambert and Dutil, 1997b) and southern Gulf (Schwalme and Chouinard, 1999) with levels of energy reserves being minimal during the spring period when spawning takes place. Females in poor condition invested less energy in maturation, but in relation to available energy reserves, energy expenditures were greater than in females in good condition (Lambert and Dutil, 2000). As a result, energy invested in reproduction by poor-condition females, as well as low energy levels in reproductive and non-reproductive individuals, increased their risk of mortality (Dutil and Lambert, 2000). The range of potential values for several variables indicative of cod condition has been determined experimentally (Lambert and Dutil, 1997b; Dutil and Lambert, 2000). The proportion of wild cod with energy reserves as low as observed in starved cod in the laboratory experiments has been determined by comparing field and laboratory observations. The degree of overlap between frequency distributions was used as an estimate of natural mortality that is directly attributable to lack of food (Dutil and Lambert, 2000). However, there are two potential drawbacks to such a method. Firstly, fish in the wild may die before critical thresholds are reached, because life is more demanding in the wild than in the laboratory, i.e. critical thresholds may differ in the wild and in the laboratory. Secondly, mortality associated with poor condition may be underestimated when condition approaches critical thresholds, i.e. dead fish are not sampled representatively. To investigate whether our estimates of M associated with poor condition had been underestimated (case 2), we compared changes in physiological condition observed in the wild to changes predicted in unfed fish kept at three temperatures. Materials and methods Laboratory experiments Cod of two size categories (38–48 and 48–58 cm) were deprived of food for 84 days at three temperatures (mean±s.d.), 2.06±0.33°C (group 1), 6.14±0.15°C (group 2) and 9.97±0.12°C (group 3), from early-November 1995 to late January 1996. There were two tanks for each combination of size and temperature. One tank held 25 fish. These fish were measured (±1 mm) and weighed (±1 g) after 21, 42, 63 and 84 days: a random sample of 10 cod was used for dissection after 84 days. The other tank held 32 fish. A random sample of 10 fish was used for dissection after 21, 42 and 63 days. Twenty fish were sampled and dissected at the start of the experiment (group 4). Forty cod held in two other tanks, one tank for each size category, were maintained at 6.05±0.19°C, and were fed capelin to satiation once a week for a period of 63 days. They were measured and weighed and then used for dissection (group 5). All fish were measured, weighed and double-tagged (Visible implant tags, Northwest Marine Technology, Shaw Island, Washington, DC) at the start of the experiment. Capture sites, handling methods and tank setup were as described in Dutil et al. (1998). Condition factor (K), somatic condition factor (Ks), liver-somatic index (LSI), gonad-somatic index (GSI) and muscle (MW) and liver (LW) water contents were measured and calculated as described in Dutil and Lambert (2000). Temperature and size category effects on changes in the condition factor over 84 days (ΔK84d) were tested with a two-way factorial ANOVA without replication (Zar, 1996). Time course changes in body weight were described with linear regressions and tested with analysis of covariance (based on observations after 21, 42, 63 and 84 days). Changes in the condition factor and somatic condition factor over 84 days (ΔK84d and ΔKs84d, respectively) were described with a linear model pooling fish from the two size categories at each level of temperature and using fork length (Li) and initial condition or somatic condition factor (Ki and Ksi, respectively) as covariates. The number of fish with GSI below and above 3% was calculated for each tank, and differences in their proportions between tanks (i.e. size categories) were tested within each group (groups 1, 2, 3 and 5) using the χ2-test for independent samples. There was no tank (size category) effect (P>0.05), and thus we pooled the data for small and large fish to examine GSI and LSI at the start (group 4), after feeding (group 5) and after food-deprivation (groups 1, 2, 3), and for cod with GSI below and above 3%, separately. GSI was compared using the Kruskal–Wallis one-way ANOVA by ranks followed by a posteriori Tukey multiple comparisons. Group effects on the proportion of fish with GSI values below and above 3% were tested using the χ2 test for independent samples. The small number of females (10%) precluded a separate analysis for males and females. As no replicate tanks were available, group effects should be interpreted cautiously. Liver-weight/somatic-weight relationships were analyzed using linear regressions and the analysis of covariance. Relationships between LW, MW and Ks were fitted with a segmented quadratic model with a plateau representing the minimal liver or muscle water contents (Gauss–Newton method). Then a polynomial regression model adapted from Lambert and Dutil (1997a) fitted to the non-linear portion of the relationships was used to assess LW and MW from Ks. The analyses are based on observations made after 21, 42, 63 and 84 days. Field–laboratory comparisons Field samples were obtained from the estuary and northern Gulf of St. Lawrence in 1994 (Dutil and Lambert, 2000). Fish size varied among samples, but our analyses were restricted to a common size range (32–64 cm): January 20, mean length=41.0±5.2 cm (n=145); April 7, 37.1±4.0 cm (n=60); May 5, 47.4±6.9 cm (n=554); June 5, 39.7±4.3 cm (n=139). Ks and fork length were not correlated (P>0.05 and r2<0.02 in any of the four samples). The length of cod averaged 44.5±7.2 cm overall for field samples compared with 48.0±3.8 cm in the laboratory experiments. Changes in condition in the laboratory were modeled for a standard 48-cm fish and compared with changes in condition observed in field samples. For three-temperature (2, 6, 10°C) representative of cod geographic distribution (Brander, 1995), we tested whether changes in Ks observed in 1994 cod differed from those expected (ΔKs) for unfed cod in the laboratory experiments. The hypothesis for the period January–April was: where KsJ and KsA are the somatic condition factors observed in January and April, respectively, and ΔKs represents the expected change in the somatic condition factor for a 48-cm fish with a somatic condition factor at the start of the food-deprivation period equal to KsJ. The expected frequency distribution for the somatic condition factor in April was constructed by translation assuming a normal distribution with mean (KsAP): KsJ and KsAP were assumed to have equal coefficients of variability. Predicted (KsAP) and observed (KsA) somatic condition factors were compared by the t-test. This procedure was repeated for the three periods (January 20–April 7, April 7–May 5, and May 5–June 5) and three temperatures (see Equations (4)–(6) in Section “Results”). Variability in the predicted somatic condition factor in May (2 and 6°C) was further examined. The change in somatic condition factor from April to May (ΔKs28d) was calculated from Equations (4) and (5) (in Section “Results”) using the median length of the fish in the May sample (47.3 cm). The expected frequency distribution for the somatic condition factor in May was constructed by translation assuming a normal distribution. The variance (Var) was calculated as: where Cov (Ksi, ΔKs84d) is the covariance between Ksi and ΔKs84d. Differences in LSI, LW and MW among the four field samples were tested with Kruskal–Wallis one-way ANOVA by ranks followed by a posteriori Tukey multiple comparisons due to non-normal distributions, particularly in the May and June samples, and heterogeneous variances (P<0.001). The t-tests were used to compare LSI, LW and MW between cod sampled in January and cod in group 4. Normality was tested using Shapiro–Wilk's W and skewness (symmetry) following D'Agostino (in Zar, 1996). Feeding activity Stomachs were collected to assess feeding activity in the field. In January and August–September 1994, cod were caught using bottom trawls during depth-stratified random surveys covering most of the winter and summer distribution of northern Gulf of St. Lawrence cod. An additional survey was conducted in May 1994 that covered most of the spawning aggregation. For each set where cod were caught, three non-empty stomachs were taken for each 10-cm length-class. Empty stomachs were counted for each length-class until three non-empty stomachs were collected. In addition, the stomach of 56 cod collected in April 1994 near Sept-Iles, Quebec, and of 140 cod caught in June 1994 near Matane, Quebec, were weighed. For each stomach, a fullness index was calculated by dividing the mass of stomach content by the fork length raised to the third power, and the result was multiplied by 10 000 (Lilly, 1991). These analyses were restricted to cod measuring between 32 and 64 cm. Results Laboratory observations The decline in condition factor over 84 days (ΔK84d) was influenced by size and temperature (P<0.05; Figure 1). Weight loss over time was linear in all size and temperature combinations (P<0.01, r2>0.95), but daily weight loss was slightly greater from day 0 to day 21 than in the period from day 21 to day 84. Figure 1 Open in new tabDownload slide Decrease in the condition factor (mean and s.d.) of Atlantic cod in two size categories over an 84-day food-deprivation period at three temperatures. Figure 1 Open in new tabDownload slide Decrease in the condition factor (mean and s.d.) of Atlantic cod in two size categories over an 84-day food-deprivation period at three temperatures. When the two size categories were combined and fish length introduced as a covariate, ΔK84d was influenced by fish length and initial condition factor (Table 1), with fish of a small size and in better condition experiencing larger decreases in condition factor. Li had a significant effect at all temperatures (r>0.42), whereas Ki was significant at 2°C (r=−0.42) and 6°C (r=−0.20), but not at 10°C. Li and Ki showed a poor correlation. Table 1 Correlation coefficients between initial length (Li), condition factor (Ki) and decline in condition factor (ΔK84d) of Atlantic cod over an 84-day food-deprivation period at three temperatures. *indicates significant relationships with ΔK84d (P<0.05). Statistics for the multiple linear regression are shown. . Temperature . . 2°C . 6°C . 10°C . Correlation coefficient  Li−ΔK84d 0.53* 0.42* 0.58*  Ki−ΔK84d −0.42* −0.20* 0.10  Li−Ki 0.03 0.20 0.23 Model ΔK84d=f(Li, Ki)  P <0.001 <0.01 <0.001  r2 0.47 0.26 0.34 Test for factor=0  Li <0.001 <0.01 <0.001  Ki <0.001 0.04 0.78 . Temperature . . 2°C . 6°C . 10°C . Correlation coefficient  Li−ΔK84d 0.53* 0.42* 0.58*  Ki−ΔK84d −0.42* −0.20* 0.10  Li−Ki 0.03 0.20 0.23 Model ΔK84d=f(Li, Ki)  P <0.001 <0.01 <0.001  r2 0.47 0.26 0.34 Test for factor=0  Li <0.001 <0.01 <0.001  Ki <0.001 0.04 0.78 Open in new tab Table 1 Correlation coefficients between initial length (Li), condition factor (Ki) and decline in condition factor (ΔK84d) of Atlantic cod over an 84-day food-deprivation period at three temperatures. *indicates significant relationships with ΔK84d (P<0.05). Statistics for the multiple linear regression are shown. . Temperature . . 2°C . 6°C . 10°C . Correlation coefficient  Li−ΔK84d 0.53* 0.42* 0.58*  Ki−ΔK84d −0.42* −0.20* 0.10  Li−Ki 0.03 0.20 0.23 Model ΔK84d=f(Li, Ki)  P <0.001 <0.01 <0.001  r2 0.47 0.26 0.34 Test for factor=0  Li <0.001 <0.01 <0.001  Ki <0.001 0.04 0.78 . Temperature . . 2°C . 6°C . 10°C . Correlation coefficient  Li−ΔK84d 0.53* 0.42* 0.58*  Ki−ΔK84d −0.42* −0.20* 0.10  Li−Ki 0.03 0.20 0.23 Model ΔK84d=f(Li, Ki)  P <0.001 <0.01 <0.001  r2 0.47 0.26 0.34 Test for factor=0  Li <0.001 <0.01 <0.001  Ki <0.001 0.04 0.78 Open in new tab ΔK84d is described by the following equations: (1) (2) (3) where fork length is in millimeter. Removing Ki in the analyses at 10°C had no influence on the statistics in Table 1. GSI averaged 1.2% (n=20) at the start of the experiment. This value was used to obtain the initial somatic condition factor (Ksi): (4) (5) (6) GSI varied considerably among individuals particularly at the end of the experiment, with some fish having large gonads and others having smaller gonads lacking any sign of maturation (Table 2). Mean GSI was greater in food-deprived cod at 2°C as a larger proportion of the fish matured during the experiment. The proportion of maturing cod was similar in fed (63 days) and food-deprived (84 days) cod at 6°C, but mean GSI was much greater in fed cod. Slow maturation was associated with low energy reserves in the liver. LSI was greater in more mature fish (1.62±1.01% and 2.86±1.35% when GSI was smaller than 3% and larger than 3%, respectively). LSI was also greater in fed cod (6.24±1.48%) than in cod sampled at the start of the experiment (4.63±1.20) and much greater than in food-deprived cod (2.87±1.28, 2.29±1.32 and 1.62±1.18% at 2, 6 and 10°C, respectively). Table 2 Gonad-somatic index (GSI,%) of Atlantic cod before (S) and after 63 days of feeding at 6°C (F) or 84 days of food-deprivation at three temperatures (n=20). Numbers of fish, GSI and fork length (mean±s.d.) are shown for cod with GSI under (I) or over (M) 3%. Values with the same letter column-wise are not different (P>0.05). . GSI (range) . I:M . GSI (<3%) (length) . GSI (>3%) (length) . S 1.19c (0.42–3.26) 18:2 0.96±0.56bc (456±40) 3.21±0.07a (438±16) F 6.11ab (1.46–13.95) 8:12 1.98±0.44a (499±49) 8.86±3.23b (500±48) 2°C 6.48a (0.89–11.65) 2:18 1.63±1.04ac (482±36) 7.01±2.03ab (477±37) 6°C 3.75b (0.51–10.83) 11:9 1.29±0.54ab (485±40) 6.76±2.48ab (471±42) 10°C 1.66c (0.18–9.41) 16:4 0.66±0.43c (479±36) 5.68±2.73ab (524±50) . GSI (range) . I:M . GSI (<3%) (length) . GSI (>3%) (length) . S 1.19c (0.42–3.26) 18:2 0.96±0.56bc (456±40) 3.21±0.07a (438±16) F 6.11ab (1.46–13.95) 8:12 1.98±0.44a (499±49) 8.86±3.23b (500±48) 2°C 6.48a (0.89–11.65) 2:18 1.63±1.04ac (482±36) 7.01±2.03ab (477±37) 6°C 3.75b (0.51–10.83) 11:9 1.29±0.54ab (485±40) 6.76±2.48ab (471±42) 10°C 1.66c (0.18–9.41) 16:4 0.66±0.43c (479±36) 5.68±2.73ab (524±50) Open in new tab Table 2 Gonad-somatic index (GSI,%) of Atlantic cod before (S) and after 63 days of feeding at 6°C (F) or 84 days of food-deprivation at three temperatures (n=20). Numbers of fish, GSI and fork length (mean±s.d.) are shown for cod with GSI under (I) or over (M) 3%. Values with the same letter column-wise are not different (P>0.05). . GSI (range) . I:M . GSI (<3%) (length) . GSI (>3%) (length) . S 1.19c (0.42–3.26) 18:2 0.96±0.56bc (456±40) 3.21±0.07a (438±16) F 6.11ab (1.46–13.95) 8:12 1.98±0.44a (499±49) 8.86±3.23b (500±48) 2°C 6.48a (0.89–11.65) 2:18 1.63±1.04ac (482±36) 7.01±2.03ab (477±37) 6°C 3.75b (0.51–10.83) 11:9 1.29±0.54ab (485±40) 6.76±2.48ab (471±42) 10°C 1.66c (0.18–9.41) 16:4 0.66±0.43c (479±36) 5.68±2.73ab (524±50) . GSI (range) . I:M . GSI (<3%) (length) . GSI (>3%) (length) . S 1.19c (0.42–3.26) 18:2 0.96±0.56bc (456±40) 3.21±0.07a (438±16) F 6.11ab (1.46–13.95) 8:12 1.98±0.44a (499±49) 8.86±3.23b (500±48) 2°C 6.48a (0.89–11.65) 2:18 1.63±1.04ac (482±36) 7.01±2.03ab (477±37) 6°C 3.75b (0.51–10.83) 11:9 1.29±0.54ab (485±40) 6.76±2.48ab (471±42) 10°C 1.66c (0.18–9.41) 16:4 0.66±0.43c (479±36) 5.68±2.73ab (524±50) Open in new tab Liver weight correlated positively with somatic weight, but the relationships were different at 2, 6 and 10°C (analysis of covariance, P>0.05 for slopes and P<0.05 for elevations): (7) (8) (9) where LWt is liver weight and SWt is somatic weight. LW and MW correlated negatively with Ks. Water content reached minimum asymptotic values of 22.2% in the liver and 80.1% in the muscle, based on the quadratic model with a plateau, for Ks values above 0.93 and 1.04, respectively. Water content was higher at lower Ks values: (10) (11) Field observations The condition of cod declined from January to May, but improved from May to June. Ks differed significantly among the four samples (P<0.001). January and May differed from all other samples (Tukey HSD, P<0.05), but April and June were not significantly different (P>0.05). Ks was normally distributed for each monthly sample (Shapiro–Wilk, P>0.05) and the four variances were homoscedastic (Bartlett, P>0.05; Table 3). This change in Ks is shown in Figure 2 for a theoretical population of 1 million fish assuming a normal distribution and using means and s.d. in Table 3. The proportion of cod with Ks values below 0.75, for instance, would be 36.2% in January, 70.1% in April, 90.1% in May and 80.8% in June. Figure 2 Open in new tabDownload slide Frequency distribution of the somatic condition factor for Atlantic cod sampled in January, April, May and June 1994 assuming a normal distribution and a population of 1 million fish. The means and standard deviations (s.d.) of Table 3 were used. Figure 2 Open in new tabDownload slide Frequency distribution of the somatic condition factor for Atlantic cod sampled in January, April, May and June 1994 assuming a normal distribution and a population of 1 million fish. The means and standard deviations (s.d.) of Table 3 were used. Table 3 Somatic condition factor (Ks, mean±s.d.) of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. P is the Shapiro–Wilk probability statistic. Sample . Ks . Range . P . Skewness . Kurtosis . January 0.780±0.058 0.62–0.98 0.942 0.17 0.45 April 0.733±0.052 0.59–0.85 0.939 −0.20 0.10 May 0.690±0.054 0.55–0.91 0.411 0.37 0.57 June 0.715±0.052 0.59–0.84 0.412 0.06 −0.34 Sample . Ks . Range . P . Skewness . Kurtosis . January 0.780±0.058 0.62–0.98 0.942 0.17 0.45 April 0.733±0.052 0.59–0.85 0.939 −0.20 0.10 May 0.690±0.054 0.55–0.91 0.411 0.37 0.57 June 0.715±0.052 0.59–0.84 0.412 0.06 −0.34 Open in new tab Table 3 Somatic condition factor (Ks, mean±s.d.) of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. P is the Shapiro–Wilk probability statistic. Sample . Ks . Range . P . Skewness . Kurtosis . January 0.780±0.058 0.62–0.98 0.942 0.17 0.45 April 0.733±0.052 0.59–0.85 0.939 −0.20 0.10 May 0.690±0.054 0.55–0.91 0.411 0.37 0.57 June 0.715±0.052 0.59–0.84 0.412 0.06 −0.34 Sample . Ks . Range . P . Skewness . Kurtosis . January 0.780±0.058 0.62–0.98 0.942 0.17 0.45 April 0.733±0.052 0.59–0.85 0.939 −0.20 0.10 May 0.690±0.054 0.55–0.91 0.411 0.37 0.57 June 0.715±0.052 0.59–0.84 0.412 0.06 −0.34 Open in new tab LSI, LW and MW differed among samples (P<0.001; Table 4). LW increased from January to April and further increased into May and June. MW, in contrast, did not change from January to April, reached a high value in May and then declined slightly in June. Table 4 Mean (n) and 95% confidence interval for the liver-somatic index (LSI, %) and liver (LW, %) and muscle (MW, %) water contents of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. Sample . LSI . LW . MW . January 4.26 (144) 27.6 (46) 80.3 (46) 4.00–4.53 25.3–29.9 80.1–80.5 April 2.34 (56) 47.8 (57) 80.5 (57) 2.09–2.59 45.0–50.6 80.3–80.8 May 2.21 (379) 59.1 (117) 82.1 (119) 2.10–2.31 56.7–61.6 81.9–82.3 June 2.12 (129) 64.0 (88) 81.7 (88) 1.98–2.26 61.9–66.1 81.5–81.9 Sample . LSI . LW . MW . January 4.26 (144) 27.6 (46) 80.3 (46) 4.00–4.53 25.3–29.9 80.1–80.5 April 2.34 (56) 47.8 (57) 80.5 (57) 2.09–2.59 45.0–50.6 80.3–80.8 May 2.21 (379) 59.1 (117) 82.1 (119) 2.10–2.31 56.7–61.6 81.9–82.3 June 2.12 (129) 64.0 (88) 81.7 (88) 1.98–2.26 61.9–66.1 81.5–81.9 Open in new tab Table 4 Mean (n) and 95% confidence interval for the liver-somatic index (LSI, %) and liver (LW, %) and muscle (MW, %) water contents of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. Sample . LSI . LW . MW . January 4.26 (144) 27.6 (46) 80.3 (46) 4.00–4.53 25.3–29.9 80.1–80.5 April 2.34 (56) 47.8 (57) 80.5 (57) 2.09–2.59 45.0–50.6 80.3–80.8 May 2.21 (379) 59.1 (117) 82.1 (119) 2.10–2.31 56.7–61.6 81.9–82.3 June 2.12 (129) 64.0 (88) 81.7 (88) 1.98–2.26 61.9–66.1 81.5–81.9 Sample . LSI . LW . MW . January 4.26 (144) 27.6 (46) 80.3 (46) 4.00–4.53 25.3–29.9 80.1–80.5 April 2.34 (56) 47.8 (57) 80.5 (57) 2.09–2.59 45.0–50.6 80.3–80.8 May 2.21 (379) 59.1 (117) 82.1 (119) 2.10–2.31 56.7–61.6 81.9–82.3 June 2.12 (129) 64.0 (88) 81.7 (88) 1.98–2.26 61.9–66.1 81.5–81.9 Open in new tab Expected vs observed distributions From January to April, Ks in wild fish decreased much less than expected based on the declines observed in unfed fish in the laboratory for all three temperatures (P<0.001; Tables 3 and 5—April). Had the wild cod not fed between January and April (76 days between sampling events) or had feeding only made up for greater activity costs in the wild compared to the laboratory, a large proportion of the fish would have been threatened (Table 5—April). From April to May, observed and predicted declines did not differ at 2 or 6°C (P>0.05), but differed at 10°C (P<0.01). From May to June, condition actually improved and hence the observed and predicted values differed at all three temperatures (P<0.001). Table 5 Predicted somatic condition factor (Ks), liver-somatic index (LSI, %), liver water (LW, %) and muscle water (MW, %) contents for Atlantic cod in April and May 1994, and percentage of fish expected to reach two critical thresholds described in Dutil and Lambert (2000). Predicted values for April and May were calculated from the January and April observations, respectively, assuming no feeding took place. Temperature . Ks . Ks<0.54 . Ks<0.66 . LSI . LW . MW . April  2°C 0.649 1.2 58.7 2.78 59.7 83.5  6°C 0.618 4.4 81.8 2.35 69.2 84.1  10°C 0.578 18.7 97.1 1.60 84.3 85.2 May  2°C 0.687 <0.1 27.6 2.98 50.1 82.8  6°C 0.675 0.2 37.3 2.53 52.9 83.0 Temperature . Ks . Ks<0.54 . Ks<0.66 . LSI . LW . MW . April  2°C 0.649 1.2 58.7 2.78 59.7 83.5  6°C 0.618 4.4 81.8 2.35 69.2 84.1  10°C 0.578 18.7 97.1 1.60 84.3 85.2 May  2°C 0.687 <0.1 27.6 2.98 50.1 82.8  6°C 0.675 0.2 37.3 2.53 52.9 83.0 Open in new tab Table 5 Predicted somatic condition factor (Ks), liver-somatic index (LSI, %), liver water (LW, %) and muscle water (MW, %) contents for Atlantic cod in April and May 1994, and percentage of fish expected to reach two critical thresholds described in Dutil and Lambert (2000). Predicted values for April and May were calculated from the January and April observations, respectively, assuming no feeding took place. Temperature . Ks . Ks<0.54 . Ks<0.66 . LSI . LW . MW . April  2°C 0.649 1.2 58.7 2.78 59.7 83.5  6°C 0.618 4.4 81.8 2.35 69.2 84.1  10°C 0.578 18.7 97.1 1.60 84.3 85.2 May  2°C 0.687 <0.1 27.6 2.98 50.1 82.8  6°C 0.675 0.2 37.3 2.53 52.9 83.0 Temperature . Ks . Ks<0.54 . Ks<0.66 . LSI . LW . MW . April  2°C 0.649 1.2 58.7 2.78 59.7 83.5  6°C 0.618 4.4 81.8 2.35 69.2 84.1  10°C 0.578 18.7 97.1 1.60 84.3 85.2 May  2°C 0.687 <0.1 27.6 2.98 50.1 82.8  6°C 0.675 0.2 37.3 2.53 52.9 83.0 Open in new tab As the observed and predicted declines in Ks over the period April–May were similar (Tables 3 and 5—May), our May sample could have been biased as it may have missed fish, which died from energy exhaustion during that critical period. The laboratory experiments do not support this possibility. The percentage of fish below two critical thresholds described in Dutil and Lambert (2000) were not markedly different for the observed distribution in May (0.3 and 28.8% for Ks<0.54 and Ks<0.66, respectively) compared with the predicted distribution at 2 and 6°C (Table 5—May). Furthermore, Ks distribution did not depart from normality in May, although Ks and LSI observations were positively skewed (P<0.01) and LW observations negatively skewed (P<0.02). LSI, as estimated from predicted Ks was close to values actually observed in April, particularly at 6°C (Equations (7)–(9), somatic weights calculated for a 48-cm fish; Tables 4 and 5). LW and MW on the other hand were much greater (Equations (10) and (11); Tables 4 and 5). In contrast, LSI estimated from predicted Ks was slightly higher than actually observed in May (Tables 4 and 5). Similarly, LW was slightly lower than actually observed, although MW was slightly higher. There was no difference in LSI between wild cod in January and cod sampled at the start of the experiment (4.26±1.61% and 4.63±1.20%, respectively, mean and s.d., P>0.05). Differences in LW (27.58±7.67% and 32.06±7.67% in January and group 4 cod, respectively) and MW (80.33±0.64% and 80.81±0.58% in January and group 4 cod, respectively) were slight but significant (P<0.05 and P<0.01, respectively). Feeding activity The percentage of empty stomachs was greatest in January, decreased in May and further decreased in late summer (Table 6). Fullness index and weight of stomach contents were also lowest in January and increased to levels that were similar in May and late summer. These results may be taken as representative of the feeding activity of the stock. Cod were caught in 115 of 123 sets in January (stomachs sampled from 95 sets), 67 of 71 in May (55 sets sampled) and 83 of 198 in late summer (62 sets sampled). Two additional samples were collected in April and June. The cod sampled in April had all fed, whereas feeding activity was lower in June than in May. These two samples may not be representative of feeding activity for the whole stock as they were obtained from discrete areas and contained fewer fish than the other samples. Table 6 Feeding activity of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. Proportion of empty stomachs is expressed as a percentage of the total number of stomachs examined. Stomach fullness index is described in the Section “Materials and methods”. The average weight of stomach contents (g) is given with empty stomachs included (1) or excluded (2). Sampling period . Number of stomachs . Average fish length . Percentage of empty stomachs . Stomach fullness index . Weight of stomach contents (1) . Weight of stomach contents (2) . January 8–28 1188 449 84.8 0.11 0.8 5.3 April 7 56 373 0.0 2.57 14.1 14.1 May 4–11 1078 456 40.9 1.43 13.0 22.0 June 5 140 396 55.7 0.54 3.6 8.1 August 18–September 7 319 443 14.4 1.50 17.7 20.8 Sampling period . Number of stomachs . Average fish length . Percentage of empty stomachs . Stomach fullness index . Weight of stomach contents (1) . Weight of stomach contents (2) . January 8–28 1188 449 84.8 0.11 0.8 5.3 April 7 56 373 0.0 2.57 14.1 14.1 May 4–11 1078 456 40.9 1.43 13.0 22.0 June 5 140 396 55.7 0.54 3.6 8.1 August 18–September 7 319 443 14.4 1.50 17.7 20.8 Open in new tab Table 6 Feeding activity of Atlantic cod in 1994 in the estuary and northern Gulf of St. Lawrence. Proportion of empty stomachs is expressed as a percentage of the total number of stomachs examined. Stomach fullness index is described in the Section “Materials and methods”. The average weight of stomach contents (g) is given with empty stomachs included (1) or excluded (2). Sampling period . Number of stomachs . Average fish length . Percentage of empty stomachs . Stomach fullness index . Weight of stomach contents (1) . Weight of stomach contents (2) . January 8–28 1188 449 84.8 0.11 0.8 5.3 April 7 56 373 0.0 2.57 14.1 14.1 May 4–11 1078 456 40.9 1.43 13.0 22.0 June 5 140 396 55.7 0.54 3.6 8.1 August 18–September 7 319 443 14.4 1.50 17.7 20.8 Sampling period . Number of stomachs . Average fish length . Percentage of empty stomachs . Stomach fullness index . Weight of stomach contents (1) . Weight of stomach contents (2) . January 8–28 1188 449 84.8 0.11 0.8 5.3 April 7 56 373 0.0 2.57 14.1 14.1 May 4–11 1078 456 40.9 1.43 13.0 22.0 June 5 140 396 55.7 0.54 3.6 8.1 August 18–September 7 319 443 14.4 1.50 17.7 20.8 Open in new tab Discussion Earlier estimates of natural mortality associated with low condition factor and energy reserves in cod were reconsidered in the present study. The mortality of emaciated cod was hypothesized to occur during the winter leading to a biased estimate of poor condition and resulting mortality in the spring in the northern Gulf of St. Lawrence. The condition of cod steadily declined over the winter in 1994 (Lambert and Dutil, 1997b; Schwalme and Chouinard, 1999), suggesting that energy gains did not match energy costs. From January to April, however, wild cod exhibited a much smaller decline in condition factor than the food-deprived cod in the laboratory experiments. Other indicators of nutritional condition were consistent with this finding. Wild cod in May had smaller livers and higher LW contents than predicted from the laboratory experiments, but LSI (>2%) and LW content (<60%) suggested that liver was still a primary source of energy. Smaller livers and higher LW contents in the May sample may reflect a greater demand for energy in the pre-spawning period than could be estimated in our laboratory experiments. The laboratory experiments were conducted in the fall period at the onset of sexual maturation in cod (Lambert et al., 1994; Karlsen et al., 1995). Results may have been different had the laboratory experiments been conducted in the period January–May when the energy demands for maturation and mating increase. Food-deprived cod may also spend more energy on activity in the wild than in the laboratory. Greater activity costs in the laboratory, however, would have resulted in fish being in an even worst projected condition in spring. Temperature effects should be interpreted cautiously. The experimental design did not include replicate samples (tanks) for each level of temperature at the end of the experiment (day 84). Tank effects may not lead to an underestimate of minimal survival costs, but they may result in an overestimate at any given temperature. This would make our comparisons between actual and observed changes in the condition factor conservative. Thus, we reject our hypothesis and conclude that our earlier estimates of mortality were not biased (Lambert and Dutil, 1997b; Dutil and Lambert, 2000). Additional experiments would be required to better estimate weight loss–temperature relationships. While mid-winter in the northern Gulf of St. Lawrence appears to be a period of relative food shortage, as shown by the decline of condition in the field samples, part of the energy costs of cod are met through occasional feeding during mid-winter. Southern Gulf cod contained substantial food quantities only during the period from May to November, whereas a majority of empty stomachs and a reduction of stomach fullness occurred in the winter (Schwalme and Chouinard, 1999). This is consistent with stomach content data for northern Gulf cod in 1994. Few cod had fed in January, and a substantial proportion of empty stomachs were found in the cod sampled in a spawning aggregation in May (Ouellet et al., 1997) and in the St. Lawrence estuary in June, when spawning was not yet over. Nevertheless, stomach samples collected in April and May indicated that some fish had actually resumed feeding before spawning occurred. Temperature and food availability in the fall and winter may explain variations in reproduction in the spring period. Wild fish classified as being emaciated in Dutil and Lambert (2000) had a much larger gonad to liver dry weight ratio than non-emaciated fish. Dutil and Lambert (2000) suggested that two groups of fish, early and late spawners, might participate in reproduction, with early spawners having lower fat reserves earlier in the spring period. Alternately, poor- and good-condition fish may participate in reproduction with poor-condition fish being in a difficult position to meet the energy demand of maturation and spawning once committed to reproduction. Maturation is initiated early in the fall (Lambert et al., 1994; Karlsen et al., 1995). The marked decline in condition factor, which occurs in autumn and winter (Lambert and Dutil, 1997b; Schwalme and Chouinard, 1999) suggests that food shortage does occur during maturation right up to spawning in both northern and southern Gulf cod. Feeding level has been shown to affect the size of the liver (Karlsen et al., 1995) and the size of the gonads (Lambert et al., 1994) early in the process of maturation. Temperature has a marked incidence on metabolic rates and costs and thus can also potentially interfere with the process of maturation in food-deprived cod. In our study, lower GSI and smaller proportions of maturing to non-maturing fish were observed at higher temperatures suggesting that food-deprivation may be more detrimental to maturation at higher temperatures. Gulf cod inhabit warmer waters in winter than in summer, but individual cod are found in a range of temperatures in winter (2–6°C) (Swain et al., 1998; Castonguay et al., 1999). Temperature selection by individual cod during the winter may thus be critical in terms of both survival and reproduction in situations of food shortage and may have an incidence on individual variations observed in spawning patterns in the spring (Ouellet et al., 1997; Dutil and Lambert, 2000). We thank Y. Gagnon, S. Chouinard and M. Péloquin for their assistance in the field and in the laboratory. We also thank R. Miller, M.-F. Beaulieu, L. Chénard, L. Perreault and L. Girard for their work in analysing stomach samples. Thanks to R. Miller for her review of an earlier version and to H. Bourdages for his help with data analysis. The Department of Fisheries and Oceans funded this work under two projects: “Programme multidisciplinaire de recherche sur la morue du nord du Golfe Saint-Laurent (IML)” and “Partitioning the total mortality of Atlantic cod stocks”. References Brander K.M. . The effect of temperature on growth of Atlantic cod (Gadus morhua) , ICES Journal of Marine Science , 1995 , vol. 52 (pg. 1 - 10 ) Google Scholar Crossref Search ADS WorldCat Castonguay M. , Rollet C. , Fréchet A. , Gagnon P. , Gilbert D. , Brêthes J.-C. . Distribution changes of Atlantic cod (Gadus morhua) in the northern Gulf of St. Lawrence in relation to an oceanic cooling , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 333 - 344 ) Google Scholar Crossref Search ADS WorldCat Dutil J.-D. , Lambert Y. . Natural mortality from poor condition in Atlantic cod (Gadus morhua) , Canadian Journal of Fisheries and Aquatic Sciences , 2000 , vol. 57 (pg. 826 - 836 ) Google Scholar Crossref Search ADS WorldCat Dutil J.-D. , Lambert Y. , Guderley H. , Blier P.U. , Pelletier D. , Desroches M. . Nucleic acids and enzymes in Atlantic cod (Gadus morhua) differing in condition and growth rate trajectories , Canadian Journal of Fisheries and Aquatic Sciences , 1998 , vol. 55 (pg. 788 - 795 ) Google Scholar Crossref Search ADS WorldCat Dutil J.-D. , Castonguay M. , Gilbert D. , Gascon D. . Growth, condition, and environmental relationships in Atlantic cod (Gadus morhua) in the northern Gulf of St. Lawrence and implications for management strategies in the Northwest Atlantic , Canadian Journal of Fisheries and Aquatic Sciences , 1999 , vol. 56 (pg. 1818 - 1831 ) Google Scholar Crossref Search ADS WorldCat Gilbert D. , Pettigrew B. . Interannual variability (1948–1994) of the CIL core temperature in the Gulf of St. Lawrence , Canadian Journal of Fisheries and Aquatic Sciences , 1997 , vol. 54 Suppl. 1 (pg. 57 - 67 ) OpenURL Placeholder Text WorldCat Hammill M.O. , Stenson G.B. . Estimated prey consumption by harp seals (Phoca groenlandica), hooded seals (Cystophora cristata), grey seals (Halichoerus grypus) and harbour seals (Phoca vitulina) in Atlantic Canada , Journal of the Northwest Atlantic Fishery Science , 2000 , vol. 26 (pg. 1 - 23 ) Google Scholar Crossref Search ADS WorldCat Karlsen Ø. , Holm J.C. , Kjesbu O.S. . Effects of periodic starvation on reproductive investment in first-time spawning Atlantic cod (Gadus morhua) , Aquaculture , 1995 , vol. 133 (pg. 159 - 170 ) Google Scholar Crossref Search ADS WorldCat Lambert Y. , Dutil J.-D. . Can simple condition indices be used to monitor and quantify seasonal changes in the energy reserves of cod (Gadus morhua)? , Canadian Journal of Fisheries and Aquatic Sciences , 1997 , vol. 54 Suppl. 1 (pg. 104 - 112 ) Google Scholar Crossref Search ADS WorldCat Lambert Y. , Dutil J.-D. . Condition and energy reserves of Atlantic cod (Gadus morhua) during the collapse of the northern Gulf of St. Lawrence stock , Canadian Journal of Fisheries and Aquatic Sciences , 1997 , vol. 54 (pg. 2388 - 2400 ) Google Scholar Crossref Search ADS WorldCat Lambert Y. , Dutil J.-D. . Energetic consequences of reproduction in Atlantic cod (Gadus morhua) in relation to spawning level of somatic energy reserves , Canadian Journal of Fisheries and Aquatic Sciences , 2000 , vol. 57 (pg. 815 - 825 ) Google Scholar Crossref Search ADS WorldCat Lambert Y. , Dutil J.-D. , Munro J. . Effects of intermediate and low salinity conditions on growth rate and food conversion of Atlantic cod (Gadus morhua) , Canadian Journal of Fisheries and Aquatic Sciences , 1994 , vol. 51 (pg. 1569 - 1576 ) Google Scholar Crossref Search ADS WorldCat Lilly G.R. . Interannual variability in predation by cod (Gadus morhua) on capelin (Mallotus villosus) and other prey off southern Labrador and northeastern Newfoundland , ICES Marine Science Symposium , 1991 , vol. 193 (pg. 133 - 146 ) OpenURL Placeholder Text WorldCat Martínez M. , Guderley H. , Dutil J.-D. , Winger P.D. , He P. , Walsh S.J. . Condition, prolonged swimming performance and muscle metabolic capacities of cod (Gadus morhua) , Journal of Experimental Biology , 2003 , vol. 206 (pg. 503 - 511 ) Google Scholar Crossref Search ADS PubMed WorldCat Ouellet P. , Lambert Y. , Castonguay M. . Spawning of Atlantic cod (Gadus morhua) in the northern Gulf of St. Lawrence: a study of adult and egg distributions and characteristics , Canadian Journal of Fisheries and Aquatic Sciences , 1997 , vol. 54 (pg. 198 - 210 ) Google Scholar Crossref Search ADS WorldCat Schwalme K. , Chouinard G.A. . Seasonal dynamics in feeding, organ weights, and reproductive maturation of Atlantic cod (Gadus morhua) in the southern Gulf of St. Lawrence , ICES Journal of Marine Science , 1999 , vol. 56 (pg. 303 - 319 ) Google Scholar Crossref Search ADS WorldCat Sinclair A.F. . Natural mortality of cod (Gadus morhua) in the southern Gulf of St. Lawrence , ICES Journal of Marine Science , 2001 , vol. 58 (pg. 1 - 10 ) Google Scholar Crossref Search ADS WorldCat Swain D.P. , Chouinard G.A. , Morin R. , Drinkwater K.F. . Seasonal variation in the habitat associations of Atlantic cod (Gadus morhua) and American plaice (Hippoglossoides platessoides) from the southern Gulf of St. Lawrence , Canadian Journal of Fisheries and Aquatic Sciences , 1998 , vol. 55 (pg. 2548 - 2561 ) Google Scholar Crossref Search ADS WorldCat Zar J.H. . , Biostatistical Analysis , 1996 Upper Saddle River Prentice Hall 662 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2003 International Council for the Exploration of the Sea

Berg,, Erik;Albert, Ole, Thomas

doi: 10.1016/S1054-3139(03)00037-7pmid: N/A

Abstract The distribution of cod along the Norwegian coast and in fjords from 62°N north to the Russian border was examined using data from annual trawl surveys carried out between 1995 and 2001. Based on differences in growth zones of the otoliths, cod are traditionally classified into two types: Northeast Arctic cod and coastal cod. Both types were found throughout the area investigated. The catch rate of both increased northwards and from offshore to inshore. In a statistical model of length at age, abiotic factors such as area and year of capture explained more of the variance than biotic factors such as sex, stage of maturity, and type of cod. Length at age increased in a southward direction and was higher for cod captured offshore than for those captured inshore. In a statistical model of the proportion mature at age, area, type, and year of capture explained more of the variance than sex and depth of capture. On average, coastal cod attained 50% maturity (M50) more than a year younger than a year younger than Northeast Arctic cod attained maturity. Although there were relatively large differences in age at maturity between neighbouring areas, age at maturity was lowest in the south and inshore, and in general, lower inshore than offshore. As genetic analysis clearly indicates that cod in the study area consist of at least two genetically separated stocks, it is likely that the differences found here in age at M50 might have a genetic component. Introduction In the North Atlantic, cod (Gadus morhua, L.) are commonly found in open oceans, over coastal banks, in open fjords, and in semi-enclosed bays at depths from 0 to 600 m. The Norwegian coast includes all these marine habitats, and in addition the coast spans a large part of the latitudinal distribution of the species. Thus, off Norway, cod are found within a range of different environmental conditions, which in turn may influence population parameters and, hence, the sustainable level of the commercial catch. In most regions where cod are found, they are targeted by the commercial fishing fleet. The species has therefore been a major focus for stock assessment, and in the North Atlantic, some 20 cod stocks are assessed and managed as separate units. To manage cod in the Northeast Arctic, quotas are set for three different stocks: Northeast Arctic cod, Norwegian coastal cod, and Murman cod. Identification of the two types of cod found along the coast of North Norway, Northeast Arctic cod and Norwegian coastal cod, is based on differences in the structure of growth zones in the otoliths (Rollefsen, 1933). Otoliths have been successfully used to separate stocks of other species, such as king mackerel (DeVries et al., 2002). Other methods used to identify the stock structure of cod in Norwegian coastal areas include differences in the number of vertebrae (Løken et al., 1994; Noreide and Pettersen, 1998) and hemoglobin and genetic investigations (Møller, 1968, 1969; Mork et al., 1984; Jørstad and Nævdal, 1989; Dahle and Jørstad, 1993; Fevolden and Pogson, 1995, 1997; Árnason and Pálsson, 1996; Noreide and Pettersen, 1998; Mork and Giæver, 1999). Most of these investigations found differences between Northeast Arctic cod and coastal cod, although some did not, and the results from Fevolden and Pogson (1997) indicate that coastal cod probably comprises several more or less discrete stocks. This is also in accordance with the inferred stock structure of cod in the Northwest Atlantic (Smedbol and Stephenson, 2001; Smedbol and Wroblewski, 2002). However, it is difficult to draw firm conclusions on the basis of these investigations because neither methodology nor interpretation fully agree. Still, the two types of cod do seem to differ with respect to life history. Northeast Arctic cod migrate over a long distance, from their feeding area in the Barents Sea to spawning areas that are mainly around Lofoten, and also migrate along the coast, north of approximately 62°N (Bergstad et al., 1987). Tagging experiments on coastal cod, on the other hand, indicate only local migrations (Jakobsen, 1987; Godø, 1995; Nøstvik and Pedersen, 1999b; Skreslet et al., 1999). The spawning grounds used by coastal cod are at numerous locations inside fjords and in the same coastal areas used by Northeast Arctic cod (Jakobsen, 1987). Coastal cod utilize the same spawning grounds repeatedly from year to year (Jakobsen, 1987). Good management of cod within the heterogeneous environment of the Norwegian coastal zone relies on knowledge of spatial patterns in the biological parameters. Failure to account for possible isolation of stocks violates the precautionary principle of contemporary natural living resource management (Smedbol and Wroblewski, 2002). This study investigates the distribution, abundance, length at age, and the age and size at sexual maturity of cod along the Norwegian coast between 62°N and the Russian border. Differences among geographical area, year, and type of cod are discussed. The results are also discussed by way of comparison with corresponding data from Northeast Arctic cod in the Barents Sea. Material and methods Sampling and type separation The sampling area consists of numerous fjords and offshore banks between 62°N and the Russian border (Figure 1). The whole area was divided into three regions (northern, middle, and southern), and these regions were divided into 26 smaller areas. The 26 areas were stratified from distribution of the trawl stations, the density of cod, and the environmental heterogeneity (depth, fjord system). Fjords north of 68°N are mainly shallower than 300 m, while those farther south are generally deeper. The coastal banks outside the fjords range in depth from 50 to 400 m. Figure 1 Open in new tabDownload slide Map of the sampling area. The different areas correspond to areas used in the model for estimation of length at age and proportion mature at age. Figure 1 Open in new tabDownload slide Map of the sampling area. The different areas correspond to areas used in the model for estimation of length at age and proportion mature at age. In general, the water temperature increases from north to south. Although there are large temperature differences between the fjords, fjords in general are cooler than the coast in winter and warmer than the coast in summer (Hegseth et al., 1995). Sampling of fish was carried out on annual combined trawl and acoustic surveys conducted in autumn of the years 1995–2001. The bottom trawls were not randomized because the seabed in fjords and over the shelf zones is generally too rough to permit trawling. Therefore, trawling was carried out whenever the seabed conditions allowed, and catches are considered reasonably representative of seabeds suitable for trawling. Each survey lasted for approximately 30 days, and on each survey, approximately 250 hauls were made. On average, half the hauls were made with a pelagic trawl. The bottom-trawl hauls were conducted at more or less the same locations each year, whereas the pelagic-trawl hauls were conducted at different locations during each survey. The pelagic trawl was a 1600-mesh Harstad trawl with a 10 mm inner net in the codend. The bottom trawl was a 1800-mesh campelen shrimp trawl, also with a 10 mm inner net in the codend. For each haul, the round weight (g) and the total length (rounded down to the nearest centimetre) were recorded for all cod or for a random subsample. Sex and maturity were determined by visually inspecting the gonads, using a general maturity index (immature, maturing, running, spent). The spawning season for cod in the area peaks between mid-March and late April, but in some areas it may continue through to late June. All surveys were conducted in autumn between August and November. The periods of investigations were therefore midway between two spawning seasons, making it difficult to determine the stage of maturity. Although measures were taken to standardize the classification method, the precision of the maturity data is not known. The sampled cod were separated into coastal cod and Northeast Arctic cod on the basis of the structure of the growth zones on the otoliths, as described by Rollefsen (1933). The otoliths were broken along their mid-axis, and read under refracted light, as described by Williams and Bedford (1974). Approximately 19 300 cod were aged, measured, and separated into type, i.e. coastal cod and Northeast Arctic cod. Coastal cod have a smaller and more circular first winter zone than Northeast Arctic cod. The shape of the first winter zone in Northeast Arctic cod otoliths is similar to the shape of the otolith and to the other winter zones. The distance between the first and the second winter zone is also larger in coastal cod otoliths. This pattern is established at an age of 2 years, and error in differentiating between the two types does not increase with age. The accuracy of this classification technique is difficult to estimate, partly because, as far as we know, no relevant investigations have been conducted, and partly because the true answer is unknown, because other methods do not agree fully about either method or interpretation of the result. However, the otolith method of separating the two types of cod is to some extent supported by other investigations, such as haemoglobin variation (Dahle and Jørstad, 1993), DNA (Fevolden and Pogson, 1995), and number of vertebrae (Løken et al., 1994). In the stock assessments of Norwegian coastal cod and Northeast Arctic cod at ICES, the otolith method is used to separate both commercial catch and scientific survey data to type. In large-scale investigations, the method is convenient because it is possible to type numerous cod within a relatively short time. Statistical analyses Initially, generalized linear models (GLMs) of length at age were used to find appropriate expressions of the age effect. Linear, polynomial, and other relationships were tried, but analyses of residuals showed trends with respect to age, indicating inappropriate model specification. Therefore, a relative measure of length at age was used to facilitate comparisons across age groups. Relative length at age (RL) was expressed as the length of each fish divided by the mean length of the respective age group. A relative length of 1 was therefore the same as the mean length of the same age group. A GLM was applied to the RL data, and recommended procedures of model selection, model fitting, and checking of the available explanatory variables were followed (Aitken et al., 1989; McCullagh and Nelder, 1989). The model selected as full (see below) described the data adequately in the sense that residuals showed constant variance and no trend with either fitted values or any explanatory variables. For comparison of the effects of areas, two variables were established to account for the main geographical variability. A two-level factor was used to separate offshore (areas 101–302 in Figure 1) from inshore areas (areas 11–32). In addition, the longshore distance from Stadt, far south in the survey area, was calculated for each haul. A GLM with a logistic link function for the response probabilities was applied to data on maturity at age. For this model, the recommended procedures of model selection, model fitting, and checking of the available explanatory variables were those documented by Collett (1991). The model selected as the full model included those biologically relevant variables most likely to influence the probability of being mature. This model (see below) described the data adequately in the sense that the link function was valid, the form of the linear predictor was adequate, and the standardized deviance residuals revealed no unexpected features or patterns. For both models, the formal procedure of model simplification by way of F-tests was complicated because the large number of degrees of freedom (some 20 000 and 16 000, respectively) made even very small and probably biologically insignificant effects statistically significant. Therefore, model terms that contributed the least to the explained variance (<1%) were eliminated even if they were statistically significant. Reduced models were also checked for residual distribution and patterns. Further, individual factor levels were combined selectively to reduce the complexity of the models. The combination was based both on the estimates and their standard errors, and on the logical relationship between the levels. Thus, areas with similar estimates (within 95% confidence limits) were combined only if they also were geographical neighbours. Full model describing RL: Full model describing the probability of being mature: in both models, α denotes the intercept, Type is a two-level factor for otolith type (Northeast Arctic or coastal cod), Sex and Maturity are two-level factors for sex and stage of maturity (immature or mature), Area is a factorial variable representing the areas shown in Figure 1, Year is a factor for year of capture, Depth is a covariate given in metres, and p is the linear predictor, where probability factor = exp(P) / (1 + exp(p)). Results Distribution of cod Cod were found in all parts of the area surveyed, both offshore and well inside the fjords (Figure 2). However, >75% of the catches of cod by number were from the northern part of the area, 67°N and northwards (Figures 1 and 2). In the southern part surveyed, catch rates were generally low, except for a few locations far inside some fjords in shallow water (<150 m). Inshore areas had better catch rates than offshore ones, and likewise, shallow areas had better catch rates than deeper ones (Figure 3). The same pattern was found throughout the survey period. The catch rate decreased steadily from 1995 to 2001. Figure 2 Open in new tabDownload slide Mean catch rate of cod (number per nautical mile) stratified by region (data from all surveys combined, 1995–2001). Figure 2 Open in new tabDownload slide Mean catch rate of cod (number per nautical mile) stratified by region (data from all surveys combined, 1995–2001). Figure 3 Open in new tabDownload slide Mean cpue (kg per hour trawling) of cod by type, catch area, and catch depth interval (data from all surveys combined, 1995–2001). Figure 3 Open in new tabDownload slide Mean cpue (kg per hour trawling) of cod by type, catch area, and catch depth interval (data from all surveys combined, 1995–2001). The length composition of both types of cod combined was bimodal inshore (Figure 4), but only the larger mode was found offshore. Cod between 35 and 70 cm long (3–6 years) were most numerous in the catches. Northeast Arctic cod were larger than coastal cod both inshore and offshore (Table 1), and both types of cod were larger offshore than inshore. Throughout the areas and the survey period, cod younger than 2 years or older than 10 years were rarely caught. Figure 4 Open in new tabDownload slide Length distribution of the two types of cod by different areas (data from all surveys combined, 1995–2001). Figure 4 Open in new tabDownload slide Length distribution of the two types of cod by different areas (data from all surveys combined, 1995–2001). Table 1 Average length and number of cod aged and typed from inshore and offshore areas, 1995–2001 combined. . Inshore . Offshore . Type of cod . Average length (cm) . Number . Average length (cm) . Number . Coastal cod 44.9 10 187 50.0 4794 Northeast Arctic cod 55.3 2260 58.4 2059 . Inshore . Offshore . Type of cod . Average length (cm) . Number . Average length (cm) . Number . Coastal cod 44.9 10 187 50.0 4794 Northeast Arctic cod 55.3 2260 58.4 2059 Open in new tab Table 1 Average length and number of cod aged and typed from inshore and offshore areas, 1995–2001 combined. . Inshore . Offshore . Type of cod . Average length (cm) . Number . Average length (cm) . Number . Coastal cod 44.9 10 187 50.0 4794 Northeast Arctic cod 55.3 2260 58.4 2059 . Inshore . Offshore . Type of cod . Average length (cm) . Number . Average length (cm) . Number . Coastal cod 44.9 10 187 50.0 4794 Northeast Arctic cod 55.3 2260 58.4 2059 Open in new tab Approximately 15 000 (78%) of the cod were classified as coastal and 4300 (22%) as Northeast Arctic (Table 1). The catches from each area contained both types (Figure 5). The portion of coastal cod increased southwards and was higher inshore than offshore. Offshore in the northeast (areas 101–103), the proportion of coastal cod was almost 50%, and in the south (areas 32 and 302), almost all cod were classified as coastal. The proportion of Northeast Arctic cod increased with size, few cod <40 cm being classified as Northeast Arctic in any area or year. This pattern was stable throughout the study period. However, the proportion of Northeast Arctic cod increased from about 15–20% in 1995 to about 30% in 2001, this observation being widespread across all areas. Figure 5 Open in new tabDownload slide Proportion of coastal (black) and Northeast Arctic cod (white) caught in different areas (data from all surveys combined, 1995–2001). Figure 5 Open in new tabDownload slide Proportion of coastal (black) and Northeast Arctic cod (white) caught in different areas (data from all surveys combined, 1995–2001). Length at age Mean length at age from age 4 was near linear (Figure 6). However, the length of individual cod varied extensively within a single age group, the range being almost the same magnitude as the mean length for an age group. Coastal cod had a slightly larger mean length at age than Northeast Arctic cod. However, because the percentage of cod classified as coastal varied between areas, interpretations must be based on the statistical model. Figure 6 Open in new tabDownload slide Mean, minimum, and maximum length at age for coastal cod and Northeast Arctic cod, 1995–2001 combined. Only data points with at least 10 observations were used to draw the lines. Figure 6 Open in new tabDownload slide Mean, minimum, and maximum length at age for coastal cod and Northeast Arctic cod, 1995–2001 combined. Only data points with at least 10 observations were used to draw the lines. The full GLM model of RL included 68 parameter estimates and still explained only 19% of the variance in the data. By comparison, a similar model of length at age, including a linear age effect, explained 70% of the variance. However, age is a trivial explanation of length and the low R2 of the RL model only reflects the high variability inherent in length-at-age data. This variability is largely independent of the spatial, temporal, and biological variables normally included in a fish survey program. The reduced model of RL included 25 parameter estimates, retaining 17% of the total variance (89% of the variance explained in the full model). Most of the explained variance was due to the geographical component and year of sampling, whereas biological variables were of minor importance (Table 2). Mature females were on average larger than mature males and coastal cod were somewhat larger than Northeast Arctic cod of the same age. Table 2 Parameter estimates from the reduced GLM of RL (R2=0.17). Explanatory variable . Number of parameter estimates . Percentage of total variance explained . Level . Parameter estimate . Standard error . Intercept 1.1268 0.0070 Type of cod 1 1.3 Coastal 0.0246 0.0023 NE Arctic 0 Sex 1 2.4 Female 0.0350 0.0032 Male 0 Maturity 1 3.7 Immature −0.0119 0.0027 Mature 0 Female×immature 1 1.3 −0.0270 0.0038 Area 15 70.6 12 −0.1701 0.0074 13 −0.1205 0.0089 14 −0.1621 0.0071 15+16 −0.1913 0.0067 17 −0.1062 0.0070 18+21 −0.1592 0.0067 22 −0.1262 0.0070 23 −0.1981 0.0074 24 −0.1350 0.0288 25 −0.0995 0.0081 26+27 −0.1403 0.0069 31+32 −0.0317 0.0093 101+102 −0.1366 0.0070 103+104+201 −0.0915 0.0066 202+203+204 −0.0429 0.0067 301+302 0 Year 4 18.5 1995+1996 −0.0542 0.0029 1997 −0.0812 0.0040 1998 −0.0294 0.0033 1999+2000 −0.0125 0.0027 2001 0 Depth 1 2.2 7.6×10−5 9.7×10−6 Explanatory variable . Number of parameter estimates . Percentage of total variance explained . Level . Parameter estimate . Standard error . Intercept 1.1268 0.0070 Type of cod 1 1.3 Coastal 0.0246 0.0023 NE Arctic 0 Sex 1 2.4 Female 0.0350 0.0032 Male 0 Maturity 1 3.7 Immature −0.0119 0.0027 Mature 0 Female×immature 1 1.3 −0.0270 0.0038 Area 15 70.6 12 −0.1701 0.0074 13 −0.1205 0.0089 14 −0.1621 0.0071 15+16 −0.1913 0.0067 17 −0.1062 0.0070 18+21 −0.1592 0.0067 22 −0.1262 0.0070 23 −0.1981 0.0074 24 −0.1350 0.0288 25 −0.0995 0.0081 26+27 −0.1403 0.0069 31+32 −0.0317 0.0093 101+102 −0.1366 0.0070 103+104+201 −0.0915 0.0066 202+203+204 −0.0429 0.0067 301+302 0 Year 4 18.5 1995+1996 −0.0542 0.0029 1997 −0.0812 0.0040 1998 −0.0294 0.0033 1999+2000 −0.0125 0.0027 2001 0 Depth 1 2.2 7.6×10−5 9.7×10−6 Open in new tab Table 2 Parameter estimates from the reduced GLM of RL (R2=0.17). Explanatory variable . Number of parameter estimates . Percentage of total variance explained . Level . Parameter estimate . Standard error . Intercept 1.1268 0.0070 Type of cod 1 1.3 Coastal 0.0246 0.0023 NE Arctic 0 Sex 1 2.4 Female 0.0350 0.0032 Male 0 Maturity 1 3.7 Immature −0.0119 0.0027 Mature 0 Female×immature 1 1.3 −0.0270 0.0038 Area 15 70.6 12 −0.1701 0.0074 13 −0.1205 0.0089 14 −0.1621 0.0071 15+16 −0.1913 0.0067 17 −0.1062 0.0070 18+21 −0.1592 0.0067 22 −0.1262 0.0070 23 −0.1981 0.0074 24 −0.1350 0.0288 25 −0.0995 0.0081 26+27 −0.1403 0.0069 31+32 −0.0317 0.0093 101+102 −0.1366 0.0070 103+104+201 −0.0915 0.0066 202+203+204 −0.0429 0.0067 301+302 0 Year 4 18.5 1995+1996 −0.0542 0.0029 1997 −0.0812 0.0040 1998 −0.0294 0.0033 1999+2000 −0.0125 0.0027 2001 0 Depth 1 2.2 7.6×10−5 9.7×10−6 Explanatory variable . Number of parameter estimates . Percentage of total variance explained . Level . Parameter estimate . Standard error . Intercept 1.1268 0.0070 Type of cod 1 1.3 Coastal 0.0246 0.0023 NE Arctic 0 Sex 1 2.4 Female 0.0350 0.0032 Male 0 Maturity 1 3.7 Immature −0.0119 0.0027 Mature 0 Female×immature 1 1.3 −0.0270 0.0038 Area 15 70.6 12 −0.1701 0.0074 13 −0.1205 0.0089 14 −0.1621 0.0071 15+16 −0.1913 0.0067 17 −0.1062 0.0070 18+21 −0.1592 0.0067 22 −0.1262 0.0070 23 −0.1981 0.0074 24 −0.1350 0.0288 25 −0.0995 0.0081 26+27 −0.1403 0.0069 31+32 −0.0317 0.0093 101+102 −0.1366 0.0070 103+104+201 −0.0915 0.0066 202+203+204 −0.0429 0.0067 301+302 0 Year 4 18.5 1995+1996 −0.0542 0.0029 1997 −0.0812 0.0040 1998 −0.0294 0.0033 1999+2000 −0.0125 0.0027 2001 0 Depth 1 2.2 7.6×10−5 9.7×10−6 Open in new tab The parameter estimates for the four combined offshore areas (areas 101–302) clearly indicate a trend of decreasing RL from south to northeast in the survey area (Table 2). There appears to be a clear linear reduction (R2=0.80, p⪡0.01) in RL offshore along the coast from Stadt to East Finnmark (Figure 7). Inshore, RL was more variable, with only a weak linear trend along the coast (R2=0.39, p⪡0.01). Further, RL was significantly lower inshore than offshore, and the difference decreased in a northward direction. Average RL was 15% higher offshore in the south than offshore in the north, and 20% higher than inshore in the north. Most (76%) of the variance explained by the 25-parameter area-based model could be explained by a 12-parameter model replacing the area effect, with a two-level factor for inshore–offshore (representing areas 11–32 and 101–302, respectively), and a covariable representing longshore distance from Stadt. Figure 7 Open in new tabDownload slide RL for cod caught inshore and offshore. The horizontal axis expresses distance in kilometres from stadt (data from all surveys combined, 1995–2001). Figure 7 Open in new tabDownload slide RL for cod caught inshore and offshore. The horizontal axis expresses distance in kilometres from stadt (data from all surveys combined, 1995–2001). The year of catch was the variable that explained the secondmost variance in RL. The parameter estimate indicates a continuous increase in RL after 1997 (Table 2). Maturity at age Some male cod were already mature at an age of 2 years, and at an age of 10 years almost all cod were mature. A GLM model was applied to the data for examination of possible differences between geographic area, type of cod, sex, depth, and year of capture. The full GLM model of the probability of being mature included 70 parameter estimates and explained approximately 57% of the variance in the data. The reduced model included 27 parameter estimates, retaining 55% of the total variance in the data (Table 3). Most of the remaining 45% of variation in the data can probably be explained by natural individual variation. All parameters in the reduced model were statistically significant (p<0.05). Not surprisingly, age and length were the two parameters explaining most variance in the probability of being mature. Of the balance of explained variance, the parameters area of catch and year of catch contributed most, followed by type of cod. Sex and depth of catch were the two parameters that explained the least variance in the model. Table 3 Parameter estimates (linear predictor) from the reduced GLM of probability of being mature (maximum rescaled R2=0.55). Explanatory variable . Number of parameter estimates . Level . Parameter estimate . Standard error . Intercept 1 −7.8946 0.1281 Age 1 0.7379 0.0258 Length 1 0.0591 0.0030 Area 18 12 −0.3511 0.1121 13 0.4229 0.1392 14 0.0237 0.0934 15 0.3328 0.1059 16 0.0953 0.0887 17+18 0.4181 0.0736 21 0.0368 0.1307 22 0.4061 0.0962 23 0.6869 0.0988 24 −1.4504 0.9706 25 −0.1547 0.1196 26+27 −0.4961 0.1978 31 0.0725 0.0948 32 1.1424 0.1382 101+102+103 −0.3599 0.0788 104+201 −0.1151 0.0859 202 −0.6305 0.1044 203+204 −0.0742 0.1029 301+302 0 Depth 1 −0.0005 0.0002 Type of cod 1 Coastal 0.4652 0.0260 Sex 1 Female −0.1737 0.0192 Year 4 1995+1996 0.3971 0.0358 1997 0.7760 0.0580 1998 −0.0863 0.0473 1999+2000 −0.7679 0.0350 2001 0 Explanatory variable . Number of parameter estimates . Level . Parameter estimate . Standard error . Intercept 1 −7.8946 0.1281 Age 1 0.7379 0.0258 Length 1 0.0591 0.0030 Area 18 12 −0.3511 0.1121 13 0.4229 0.1392 14 0.0237 0.0934 15 0.3328 0.1059 16 0.0953 0.0887 17+18 0.4181 0.0736 21 0.0368 0.1307 22 0.4061 0.0962 23 0.6869 0.0988 24 −1.4504 0.9706 25 −0.1547 0.1196 26+27 −0.4961 0.1978 31 0.0725 0.0948 32 1.1424 0.1382 101+102+103 −0.3599 0.0788 104+201 −0.1151 0.0859 202 −0.6305 0.1044 203+204 −0.0742 0.1029 301+302 0 Depth 1 −0.0005 0.0002 Type of cod 1 Coastal 0.4652 0.0260 Sex 1 Female −0.1737 0.0192 Year 4 1995+1996 0.3971 0.0358 1997 0.7760 0.0580 1998 −0.0863 0.0473 1999+2000 −0.7679 0.0350 2001 0 Open in new tab Table 3 Parameter estimates (linear predictor) from the reduced GLM of probability of being mature (maximum rescaled R2=0.55). Explanatory variable . Number of parameter estimates . Level . Parameter estimate . Standard error . Intercept 1 −7.8946 0.1281 Age 1 0.7379 0.0258 Length 1 0.0591 0.0030 Area 18 12 −0.3511 0.1121 13 0.4229 0.1392 14 0.0237 0.0934 15 0.3328 0.1059 16 0.0953 0.0887 17+18 0.4181 0.0736 21 0.0368 0.1307 22 0.4061 0.0962 23 0.6869 0.0988 24 −1.4504 0.9706 25 −0.1547 0.1196 26+27 −0.4961 0.1978 31 0.0725 0.0948 32 1.1424 0.1382 101+102+103 −0.3599 0.0788 104+201 −0.1151 0.0859 202 −0.6305 0.1044 203+204 −0.0742 0.1029 301+302 0 Depth 1 −0.0005 0.0002 Type of cod 1 Coastal 0.4652 0.0260 Sex 1 Female −0.1737 0.0192 Year 4 1995+1996 0.3971 0.0358 1997 0.7760 0.0580 1998 −0.0863 0.0473 1999+2000 −0.7679 0.0350 2001 0 Explanatory variable . Number of parameter estimates . Level . Parameter estimate . Standard error . Intercept 1 −7.8946 0.1281 Age 1 0.7379 0.0258 Length 1 0.0591 0.0030 Area 18 12 −0.3511 0.1121 13 0.4229 0.1392 14 0.0237 0.0934 15 0.3328 0.1059 16 0.0953 0.0887 17+18 0.4181 0.0736 21 0.0368 0.1307 22 0.4061 0.0962 23 0.6869 0.0988 24 −1.4504 0.9706 25 −0.1547 0.1196 26+27 −0.4961 0.1978 31 0.0725 0.0948 32 1.1424 0.1382 101+102+103 −0.3599 0.0788 104+201 −0.1151 0.0859 202 −0.6305 0.1044 203+204 −0.0742 0.1029 301+302 0 Depth 1 −0.0005 0.0002 Type of cod 1 Coastal 0.4652 0.0260 Sex 1 Female −0.1737 0.0192 Year 4 1995+1996 0.3971 0.0358 1997 0.7760 0.0580 1998 −0.0863 0.0473 1999+2000 −0.7679 0.0350 2001 0 Open in new tab In general, the parameter estimates for the four outer areas combined (areas 101–302) indicated increasing age at M50 from south to northeast (Table 3). The parameter estimate for area 202 is, however, much lower than the estimates for all. For the 14 inshore areas (areas 12–32), the variation between neighbouring areas was rather high, and there was no obvious north–south trend. The very low parameter estimate in area 24 (−1.45) was due to low numbers (s.e.=0.97). However, as for the outer areas, M50 was lowest in the south (area 32; Table 3, Figure 8). Although the age at maturity differed extensively between neighbouring areas, cod caught inshore (areas 12–32) matured younger than cod caught offshore (areas 101–302; Figure 8). The average M50 for coastal and Northeast Arctic cod was 5.7 and 6.9 years, respectively. The geographical difference was much larger for coastal than for Northeast Arctic cod. The fitted values from the model for Northeast Arctic cod in the south (areas 32 and 302) are rather uncertain as a result of the low numbers (Table 3, Figure 8). The difference in age at maturity between the two types of cod was notable throughout the area surveyed, and it increased in a southward direction (almost 3 years in area 32). The year effect indicated a lower age at M50 in the period 1995–1997 than subsequently (Table 3). Figure 8 Open in new tabDownload slide Rate of maturation of cod (both sexes) by age, type, and area of catch (data from all surveys combined, 1995–2001). Solid lines represent inshore areas, dotted lines offshore areas, curves with symbols plotted coastal cod, and curves with no symbols plotted Northeast Arctic cod. Figure 8 Open in new tabDownload slide Rate of maturation of cod (both sexes) by age, type, and area of catch (data from all surveys combined, 1995–2001). Solid lines represent inshore areas, dotted lines offshore areas, curves with symbols plotted coastal cod, and curves with no symbols plotted Northeast Arctic cod. Discussion All surveys were conducted in autumn (September–November), at least 1–2 months before the spawning migration of Northeast Arctic cod towards the Norwegian coast starts (Bergstad et al., 1987). The observed distribution pattern of the two types of cod would therefore have been quite different if the surveys had been conducted during the first quarter of the year. However, the differences in length at age and age at maturity would most likely have been the same because the growth rate and the age at maturity observed for Northeast Arctic cod is the same as observed in the Barents Sea (ICES, 2002). Between 62 and 67°N, both inshore and offshore, the abundance of cod was much lower than farther north. This may partly be related to the bathymetry in the different regions, because cod density was greatest shallower than 300 m in all areas. The fjords in the south are generally deeper than 300 m, while those in the north tend to be shallower than 300 m, and therefore more suitable for cod. The bathymetry of the southern coastal banks (depth 100–500 m) is also different from that of banks farther north. In the south, some 75% of the banks are deeper than 300 m, and cod (and haddock Melanogrammus aeglifinus) abundance at all depths was much lower than in the north. There has never been large-scale fishing activity for cod and haddock on these southern banks, where catches are dominated by blue whiting (Micromesistius poutassou), greater silver smelt (Argentina silus), Norway pout (Trisopterus esmarkii), and Norway redfish (Sebastes viviparus). Other pelagic and demersal fish species preferred as prey by cod (Bergstad et al., 1987) are seldom found in the south. Very few cod smaller than 25 cm were caught in pelagic or bottom trawls. Engåas and God (1989) suggested that small cod may escape under the groundrope of bottom trawls (Engås and Godø, 1989), but because small cod are frequently caught in the Barents Sea with the same trawl, the absence of small cod in this study indicates that they are actually not very abundant on the trawling grounds. Small cod in the fjords and coastal areas are in shallow water close to shore, where trawling is impossible (Løken et al., 1994; Johansen et al., 1999; Nøstvik and Pedersen, 1999a; Berg and Pedersen, 2001). Løken et al. (1994) discussed settling strategies for coastal cod and suggested that in fjords and at the coast it may be advantageous for young cod to settle in the sublittoral. The macroalgae belt there may provide refuge for juvenile cod from the large cannibalistic cod that live in deeper waters. The lesser density of cod in fjords in the south may therefore also be associated with the absence of suitable areas for small cod to inhabit. Such a settling strategy is very different from that of juvenile Northeast Arctic cod in the Barents Sea, which settle in deeper water. Therefore, if eggs and larvae of Northeast Arctic cod drift into fjords, they are likely to settle in deep water, where they would probably be exposed to a higher rate of predation (Løken et al., 1994). The suggested difference in settling strategy between the two types of cod might be important in maintaining the stock structure between them. Other than food availability, temperature is the most important influence on growth rate of cod (Suthers and Sundby, 1996). The optimum temperature for large cod is 9–12°C and for small cod it is 11–15°C (Pedersen and Jobling, 1989). The bottom temperature in the investigation area seldom exceeds this. Cod living in areas with the highest temperature therefore have the fastest growth (Brander, 1995). During winter, the water temperature is higher offshore, whereas in summer it is higher in the fjords (Hegseth et al., 1995). The winter temperature is the most crucial because the temperature can approach 0°C in some fjords, and growth rate increases exponentially from this low level (Pedersen and Jobling, 1989; Brander, 1995). The temperature also decreases in a northward direction (Hegseth et al., 1995). The different temperature regimes are probably the main reason why the average length at age of cod increases from north to south and from inshore to offshore. There was only a small difference in length at age between coastal cod and Northeast Arctic cod when immature, confirming the results of laboratory experiments that revealed the same under identical conditions (Godø and Moksness, 1987; Svåsand et al., 1996). However, we found that, following maturity, length at age was slightly higher for coastal than for Northeast Arctic cod, and the difference increased with age. The average age at M50 for coastal cod (5.7 years) calculated here is similar to earlier estimates from a fjord system in the northern part of the area (Berg and Pedersen, 2001). The low age at M50 inshore in the south is the same as found in earlier studies in the same region (Godø and Moksness, 1987). The calculated M50 for Northeast Arctic cod (6.9 years) was more than a year higher than for coastal cod, and is the same as in the Barents Sea, the main feeding area of Northeast Arctic cod (ICES, 2002). This indicates that the cod determined as Northeast Arctic by the otoliths in this investigation are probably of the same origin as the same type of cod in the Barents Sea. This is also in line with life history theory, which predicts that migratory fish should mature later and at larger size than non-migrants (Roff, 1988). When reared under similar environmental conditions, the field-observed differences in growth rate and age at maturity between coastal cod and Northeast Arctic cod seemed to be eliminated (Godø and Moksness, 1987). Those authors also indicated that the differences found in the field were probably not of genetic origin. However, in laboratory experiments, fish do not have the chance to select for temperature and prey, so the results from such experiments should not automatically be applied to natural conditions. Fish from different cod stocks can, for instance, inhabit the same areas but prefer different prey and/or different ambient temperatures, resulting in different growth rates. An increased growth rate for a stock is usually associated with maturation at a younger age (Jørgensen, 1990). Our results showed only a small difference in length at age between coastal cod and Northeast Arctic cod, especially before maturation, whereas the difference in age at M50 was more than 1 year. Northeast Arctic cod were therefore 6 cm (10%) longer than coastal cod at M50. This finding counters the results of earlier investigations, which showed that the two types matured at the same length (Godø and Moksness, 1987). The difference in age at maturity cannot be explained by errors in age determination or in specifying maturity stage because any such error would be the same for both types of cod. Other possible sources of difference in these parameters between the two types of cod could be environmental, direct or indirect consequences of selection by fishing, or genetic. Environmental differences experienced early in the life of cod have been suggested as an explanation for differences in growth and maturity (Godø and Moksness, 1987). However, this is not likely because the difference in age at maturity between the two types in this investigation was demonstrated also offshore in the north, where the environment is similar to that in the open ocean. Differences in fishing mortality alone cannot explain the observed pattern because they would only change the abundance of old fish and not lead to a change in the percentage of immature cod in these age groups (Jørgensen, 1990). However, possible differences in exploitation pattern together with an inherited component in age at maturity might explain the differences. There was nothing in the data to indicate a difference in the condition of cod that could have caused earlier or delayed maturation of the two types. If Northeast Arctic cod had been feeding in other areas, such as the Barents Sea, for most of their life prior to capture, it would still be remarkable that the two types had approximately the same length at age but large differences in age at maturity. Besides, the pattern was found throughout the whole period of investigation. Most coastal cod spawning takes place inside fjords and close to shore (Jakobsen, 1987), so it is isolated from the spawning of Northeast Arctic cod. Investigations on the main spawning ground of Northeast Arctic cod revealed that the two types cluster in separated groups in the survey area (Dahle and Jørstad, 1993; Noreide and Pettersen, 1998). Samples taken for analysis of stock structure when cod were spawning revealed homogenous groups of the two types. Therefore, Northeast Arctic cod and coastal cod might be sufficiently isolated during spawning to maintain the stock structure revealed by all the investigations. As genetic analysis clearly indicates that cod in the study area comprise at least two genetically separated stocks (Fevolden and Pogson, 1995, 1997), it seems reasonable to assume that the differences we found in age at M50 might have a genetic component. We thank two anonymous referees for valuable suggestions on an earlier version of the article. References Aitken M. , Andersen D. , Francis B. , Hinde J. . , Statistical Modelling in GLIM , 1989 Oxford Clarendon Press 374 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Árnason E. , Pálsson S. . Mitochondrial cytochrome b DNA sequence variation of Atlantic cod Gadus morhua, from Norway , Molecular Ecology , 1996 , vol. 5 (pg. 715 - 724 ) Google Scholar Crossref Search ADS WorldCat Berg E. , Pedersen T. . Variability in recruitment, growth and sexual maturity of coastal cod (Gadus morhua L.) in a fjord system in northern Norway , Fisheries Research , 2001 , vol. 52 (pg. 179 - 189 ) Google Scholar Crossref Search ADS WorldCat Bergstad O.A. , Jørgensen T. , Dragesund O. . Life history and ecology of the gadoid resources of the Barents Sea , Fisheries Research , 1987 , vol. 5 (pg. 119 - 161 ) Google Scholar Crossref Search ADS WorldCat Brander K.M. . The effect of temperature on growth of Atlantic cod (Gadus morhua L.) , ICES Journal of Marine Science , 1995 , vol. 52 (pg. 1 - 10 ) Google Scholar Crossref Search ADS WorldCat Collett D. . , Modelling Binary Data , 1991 London Chapman and Hall 369 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dahle G. , Jørstad K.E. . Haemoglobin variation in cod—a reliable marker for Arctic cod Gadus morhua L , Fisheries Research , 1993 , vol. 16 (pg. 301 - 311 ) Google Scholar Crossref Search ADS WorldCat DeVries D.A. , Grimes C.B. , Prager M.H. . Using otolith shape analysis to distinguish eastern Gulf of Mexico and Atlantic Ocean stocks of king mackerel , Fisheries Research , 2002 , vol. 57 (pg. 51 - 62 ) Google Scholar Crossref Search ADS WorldCat Engås A. , Godø O.R. . Escape of fish under fishing line of a Norwegian sampling trawl and its influence on survey results , Journal du Conseil International pour l'Exploration de la Mer , 1989 , vol. 45 (pg. 269 - 276 ) Google Scholar Crossref Search ADS WorldCat Fevolden S.E. , Pogson G.H. . Skjoldal H.R. , Hopkins C. , Erikstad K.E. , Leinaas H.P. . Difference in nuclear DNA RFLPs between the Norwegian coastal and the Northeast Arctic population of Atlantic cod , Ecology of Fjords and Coastal Waters , 1995 Amsterdam Elsevier (pg. 403 - 415 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fevolden S.E. , Pogson G.H. . Genetic divergence at the synoptophysin (Syp I) locus among Norwegian coastal and north-east Arctic populations of Atlantic cod , Journal of Fish Biology , 1997 , vol. 51 (pg. 895 - 908 ) OpenURL Placeholder Text WorldCat Godø O.R. . Transplantation-tagging-experiments in preliminary studies of migration of cod off Norway , ICES Journal of Marine Science , 1995 , vol. 52 (pg. 955 - 962 ) Google Scholar Crossref Search ADS WorldCat Godø O.R. , Moksness E. . Growth and maturation of Norwegian coastal cod and Northeast Arctic cod under different conditions , Fisheries Research , 1987 , vol. 5 (pg. 235 - 242 ) Google Scholar Crossref Search ADS WorldCat Hegseth E.N. , Svendsen H. , Hellum von Quillfeldt C. . Skjoldal H.R. , Hopkins C. , Erikstad K.E. , Leinaas H.P. . Phytoplankton in fjords and coastal waters of northern Norway: environmental conditions and dynamics of the spring bloom , Ecology of Fjords and Coastal Waters , 1995 Amsterdam Elsevier (pg. 45 - 72 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ICES Report of the Arctic Fisheries Working Group , 2000 ICES CM 2002/ACFM: 18. 451 pp Jakobsen T. . Coastal cod in northern Norway , Fisheries Research , 1987 , vol. 5 (pg. 223 - 234 ) Google Scholar Crossref Search ADS WorldCat Johansen R. , Pedersen T. , Barret R.T. . Howell E. , Moksness E. , Svåsand T. . Cormorants Phalacrocorax carbo carbo as predators in a cod Gadus morhua enhancement area in North Norway , Stock Enhancement and Sea Ranching , 1999 Oxford Fishing News Books (pg. 334 - 349 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jørgensen T. . Long-term changes in age at sexual maturity of Northeast Arctic cod (Gadus morhua L.) , Journal du Conseil International pour l'Exploration de la Mer , 1990 , vol. 46 (pg. 235 - 248 ) Google Scholar Crossref Search ADS WorldCat Jørstad K.E. , Nævdal G. . Genetic variation and population structure of cod, Gadus morhua L., in some fjords in northern Norway , Journal of Fish Biology , 1989 , vol. 35 (pg. 245 - 252 ) Google Scholar Crossref Search ADS WorldCat Løken S. , Pedersen T. , Berg E. . Vertebrae numbers as an indicator for the recruitment mechanism of coastal cod of northern Norway , ICES Marine Science Symposia , 1994 , vol. 198 (pg. 510 - 519 ) OpenURL Placeholder Text WorldCat McCullagh P. , Nelder J.A. . , Generalized Linear Models , 1989 London Chapman and Hall 511 pp Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Møller D. . Genetic diversity in spawning cod along the Norwegian coast , Hereditas , 1968 , vol. 60 (pg. 1 - 32 ) Google Scholar Crossref Search ADS PubMed WorldCat Møller D. . The relationship between Arctic and coastal cod in their immature stages illustrated by frequencies of genetic charaters , Fiskeridirektoratets Skrifter Serie Havundersøkelser , 1969 , vol. 15 (pg. 220 - 233 ) OpenURL Placeholder Text WorldCat Mork J. , Giæver M. . Genetic structure of cod along the coast of Norway: results from isozyme studies , Sarsia , 1999 , vol. 84 (pg. 157 - 168 ) Google Scholar Crossref Search ADS WorldCat Mork J. , Giskeødegård R. , Sundnes G. . Population genetics studies in cod (Gadus morhua L.) by means of the haemoglobin polymorphism; observations in a Norwegian coastal population , Fiskeridirektoratets Skrifter Serie Havundersøkelser , 1984 , vol. 17 12 (pg. 449 - 471 ) OpenURL Placeholder Text WorldCat Noreide J.T. , Pettersen I.H. . Haemoglobin frequencies and vertebral numbers of cod (Gadus morhua L.) off northern Norway—test of a population structure hypothesis , ICES Journal of Marine Science , 1998 , vol. 55 (pg. 134 - 140 ) Google Scholar Crossref Search ADS WorldCat Nøstvik F. , Pedersen T. . Catching cod for tagging experiments , Fisheries Research , 1999 , vol. 42 (pg. 57 - 66 ) Google Scholar Crossref Search ADS WorldCat Nøstvik F. , Pedersen T. . Howell E. , Moksness E. , Svåsand T. . Movement pattern and growth of wild cod (Gadus morhua L.) and hatchery-reared cod released as 1-group , Stock Enhancement and Sea Ranching , 1999 Oxford Fishing News Books (pg. 315 - 333 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pedersen T. , Jobling M. . Growth rates of large, sexually mature cod, Gadus morhua, in relation to condition and temperature during an annual cycle , Aquaculture , 1989 , vol. 81 (pg. 161 - 168 ) Google Scholar Crossref Search ADS WorldCat Roff D.A. . The evolution of migration and some life history parameters in marine fishes , Environmental Biology of Fishes , 1988 , vol. 22 (pg. 133 - 146 ) Google Scholar Crossref Search ADS WorldCat Rollefsen G. . The otoliths of the cod , Report on Norwegian Fishery and Marine Investigations , 1933 , vol. 4 3 (pg. 4 - 18 ) OpenURL Placeholder Text WorldCat Skreslet S. , Albrigtsen I. , Andersen A.P. , Kolbeinshavn A. , Pedersen T. , Unstad K. . Howell E. , Moksness E. , Svåsand T. . Migration, growth and survival in stocked and wild cod (Gadus morhua L.) in the Vestfjord region, North Norway , Stock Enhancement and Sea Ranching , 1999 Oxford Fishing News Books (pg. 306 - 314 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smedbol R.K. , Stephenson R. . The importance of managing within-species diversity in cod and herring fisheries of the north-western Atlantic , Journal of Fish Biology , 2001 , vol. 59 Suppl. A (pg. 109 - 128 ) Google Scholar Crossref Search ADS WorldCat Smedbol R.K. , Wroblewski J.S. . Metapopulation theory and northern cod population structure: interdependency of subpopulations in recovery of a groundfish population , Fisheries Research , 2002 , vol. 55 (pg. 161 - 174 ) Google Scholar Crossref Search ADS WorldCat Suthers I.M. , Sundby S. . Role of the midnight sun: comparative growth of pelagic juvenile cod (Gadus morhua) from the Arcto-Norwegian and a Nova Scotian stock , ICES Journal of Marine Science , 1996 , vol. 53 (pg. 827 - 836 ) Google Scholar Crossref Search ADS WorldCat Svåsand T. , Jørstad K.E. , Otterå H. , Kjesbu O.S. . Differences in growth performance between Arcto-Norwegian and Norwegian coastal cod reared under identical conditions , Journal of Fish Biology , 1996 , vol. 49 (pg. 108 - 119 ) Google Scholar Crossref Search ADS WorldCat Williams T. , Bedford B.C. . Bagenal T.B. . The use of otoliths for age determination , Proceedings of an International Symposium on the Ageing of Fish , 1974 UK Old Woking (pg. 114 - 123 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2003 International Council for the Exploration of the Sea

Smith,, C.J;Papadopoulou,, K.-N

doi: 10.1016/S1054-3139(03)00065-1pmid: N/A

Abstract An underwater television technique was used to investigate burrow density of Nephrops norvegicus in a large semi-enclosed bay in the west central Aegean. Pagasitikos Bay has the shallowest occurrence of Nephrops in high densities in Greek waters with an estimated population area of 376 km2. As trawling is by law prohibited in the bay, Nephrops is fished only by bottom nets. Burrow densities were estimated seasonally (May, August and November 1998 and February 1999) by video sled transects at nine stations around the bay. Bottom trawls were undertaken to estimate animal abundance, mean individual carapace length and mean weight. Total stock estimate (number and weight) was undertaken for 100 and 75% burrow occupancy with the ground delimited by the 60 m contour. Annual removal by fishermen was estimated to be 1.5–2% of the stock. Densities were found to be higher than in other Greek Nephrops grounds. Variations in burrow density were found both between stations and within stations over time, with an overall decrease in density in 1999. The decrease in density during 1999, in conjunction with an increase in mean carapace length from the trawl catches was attributed to an eutrophication event evidenced by a flock layer settling onto the seabed and causing, in the worst cases, patches of anoxic surface sediments. The event was most probably the result of high nutrient run-off into the bay. Introduction The crustacean Nephrops norvegicus is one of the most important commercial species of the mixed demersal fishery for the shrimp Parapenaeus longirostris and hake, Merluccius merluccius in the Mediterranean (ICES, 1999). This is a widespread fishery across the northern Mediterranean region with highest landings in the Adriatic. Greek Nephrops landings in the period 1994–2000 represented an average of 9.7% of the total landings for the Mediterranean (FAO FISHSTAT data). Within Greece, the species represents 2.5% of the total fish (both demersal and pelagic) landing income at auction (ETANAL, 1998). The commercial fishery is primarily undertaken through trawling on the shelf of the northern Aegean, between 200 and 400 m depth with limited catch in other areas including slope grounds. The principal managerial measure for demersal fishing in Greece is a closed season in the summer (1st June to 30th September) but as a precautional measure there is a total ban of bottom trawling in Pagasitikos Bay. Pagasitikos Bay, is an enclosed bay in the western central Aegean, and is an area of high interest for Nephrops as this is the shallowest occurrence (<100 m depth) of the species in high quantities in Greek waters. The bay has an active all-year-round multispecies gillnet fishery including Nephrops as one of the target species. On occasion, net fishermen may further target Nephrops by using baited nets that are fished “heavy” so that they lie on the seabed. Maximum fishing pressure (April–August) coincides with the time of maximum availability of female Nephrops on the sediment surface. Many aspects of the biology and fishery of Nephrops have been studied in the Atlantic and the Western Mediterranean Sea (Tuck et al., 1997 a, b, 2000; Gonzalez-Gurrianan et al., 1998; Mytilineou et al., 1998; Sardà et al., 1998; Fariña et al., 1999; Chapman et al., 2000) and regular assessments of many European Atlantic Nephrops stocks are being carried out by the ICES Nephrops Working Group (ICES, 2001). Detailed studies of Aegean Nephrops have been undertaken through the activities of only a few recent research projects (e.g. Anon., 1994; Sardà, 1998; Smith et al., 2001). Traditional stock assessment methods rely on data from the fishery (landings, effort, discarding, etc.). Fishery dependent methods are known to have limitations for Nephrops since it has a burrowing lifestyle and availability to capture depends largely on emergence behaviour which varies with animal sex and size, time of day and season (Bailey et al., 1993). Fishery independent methods, on the other hand, have become an increasingly useful tool in assessment of Nephrops stocks (ICES, 2001) and could be a source of independent data with which to compare and fine-tune analytical methods (ICES, 1999). These include the underwater video technique (Bailey et al., 1993; Anon., 1994; ICES, 1995; Tuck et al., 1997b, 1999; Marrs et al., 1998; Smith et al., in press) and the larval production technique (Tuck et al., 1997b; Briggs et al., 2002). This is the first application of the underwater video technique for Nephrops stock assessment in the Eastern Mediterranean. The first application of the method to the Pagasitikos stock presented particular interest and advantages due to local topographical and managerial peculiarities of the area i.e. a semi-enclosed bay with a total ban on trawling and a sedimentary environment with fully developed burrowing communities and a sizable Nephrops stock. Environmental and socio-economic concerns over recurrent eutrophication events in the bay (NCMR, 2000; Smith et al., 2001) and the future of the fishery raise the issue of possible changes in management schemes. The work presented here is part of a larger research project concerned with periodic catch and release studies using tagging and traps and focussing on a number of aspects of Nephrops biology and fishery in Pagasitikos Bay including growth, mortality, maturity, fecundity and comparative evaluation of creeling, gillnetting and trawling. The stock assessment work involved seasonal sampling of a grid of stations in Pagasitikos Bay using towed underwater video and supplementary trawl sampling at one of the stations. Methods Sampling area Pagasitikos Bay is a semi-enclosed gulf approximately 15 nautical miles across at its widest, with a narrow mouth which through the Trikeri channel opens up into the northern Aegean (Figure 1). Average depth is approximately 90 m with one slightly deeper area of 100 m depth. The channel into the bay is 60 m depth and circulation is thought to be limited. Sediments (below 60 m depth) are mostly soft clay dominated muds (NCMR, 2000) allowing for full development of burrowing communities. Bottom water temperatures, as recorded by CTD in 1998 and 1999 from a grid of stations remain quite constant throughout the year (12.83–14.53°C) while surface water temperatures vary considerably (12.84–27.4°C). Surface salinity is also quite variable (35.74–38.00 psu) with more constant conditions on the bottom (37.98–38.75 psu). Figure 1 Open in new tabDownload slide Pagasitikos Bay in central west Aegean, showing the video sampling sites. Figure 1 Open in new tabDownload slide Pagasitikos Bay in central west Aegean, showing the video sampling sites. Video methods Assessment of Nephrops burrow density was undertaken by towed underwater video sledge at nine stations spread across the bay (Figure 1, Table 1) during May, August and November 1998 and February 1999. The video sledge was a modern Marine Laboratory (Aberdeen) design and the camera a colour CCD Osprey (OE1360 Osprey Electronics, Aberdeen) camera mounted on the sled looking obliquely forward with two wide-angle 500 W underwater lighting units. The camera had a fixed focal length, standard lens and wide field of view. The sled was towed astern of the RV Philia (IMBC) on a 12 mm trawl warp. Floatation was added to the warp at the sled end of the cable to prevent the towing cable from disturbing the sediment in front of the sled. Table 1 Sampling stations and depth (D) in metres. Station . Latitude . Longitude . D . A 39° 15.5′ 22° 55.0′ 66 B 39° 16.5′ 23° 02.0′ 92 C 39° 17.0′ 23° 06.5′ 88 D 39° 11.5′ 22° 57.0′ 72 E 39° 12.5′ 23° 03.0′ 91 F 39° 12.0′ 23° 07.5′ 90 G 39° 09.0′ 22° 59.5′ 85 H 39° 11.0′ 23° 08.5′ 91 I 39° 05.5′ 23° 01.0′ 82 Station . Latitude . Longitude . D . A 39° 15.5′ 22° 55.0′ 66 B 39° 16.5′ 23° 02.0′ 92 C 39° 17.0′ 23° 06.5′ 88 D 39° 11.5′ 22° 57.0′ 72 E 39° 12.5′ 23° 03.0′ 91 F 39° 12.0′ 23° 07.5′ 90 G 39° 09.0′ 22° 59.5′ 85 H 39° 11.0′ 23° 08.5′ 91 I 39° 05.5′ 23° 01.0′ 82 Open in new tab Table 1 Sampling stations and depth (D) in metres. Station . Latitude . Longitude . D . A 39° 15.5′ 22° 55.0′ 66 B 39° 16.5′ 23° 02.0′ 92 C 39° 17.0′ 23° 06.5′ 88 D 39° 11.5′ 22° 57.0′ 72 E 39° 12.5′ 23° 03.0′ 91 F 39° 12.0′ 23° 07.5′ 90 G 39° 09.0′ 22° 59.5′ 85 H 39° 11.0′ 23° 08.5′ 91 I 39° 05.5′ 23° 01.0′ 82 Station . Latitude . Longitude . D . A 39° 15.5′ 22° 55.0′ 66 B 39° 16.5′ 23° 02.0′ 92 C 39° 17.0′ 23° 06.5′ 88 D 39° 11.5′ 22° 57.0′ 72 E 39° 12.5′ 23° 03.0′ 91 F 39° 12.0′ 23° 07.5′ 90 G 39° 09.0′ 22° 59.5′ 85 H 39° 11.0′ 23° 08.5′ 91 I 39° 05.5′ 23° 01.0′ 82 Open in new tab Position of the towing vessel was recorded every 5 min (D-GPS position) along with depth, and the output from the TV camera was recorded on videotape (S-VHS), together with a time signal. Video recordings were carried out for at least 30 min of clear seabed viewing (Table 1). A total of over 19 000 m tow length (over 14 h video) was analysed for the purpose of this study with an average tow of 540 m per station per season. Although major features were continually noted by the operators in situ, counting the Nephrops burrows (complete burrow systems and not individual openings) and emergent Nephrops as they passed a line of 65 cm width one-third up from the bottom of the display (pre-calibrated view) was done back in the laboratory. Since the length of each 5 min segment of each tow was known, the area of sea bed viewed (m2) was computed. Burrow density was computed from the burrow count and area of sea bed viewed (length of 5-min tow×65 cm width) in each 5 min segment. Densities were averaged for each site in each period and standardised to 100 m2 areas for comparative purposes. Burrow identifications Through a combination of field studies, resin casting and laboratory observations Nephrops burrows, burrow construction, burrow longevity and burrow maintenance behaviour have been documented in detail (Chapman and Rice, 1971; Atkinson, 1974a; Chapman, 1980; Nash, 1980; Atkinson and Chapman, 1984; Tuck et al., 1994; Marrs et al., 1996). The burrows of the red band fish Cepola rubescens, co-occurring with Nephrops in some grounds, are very distinctive (Atkinson et al., 1977; Atkinson and Pullin, 1996) making their identification relatively straightforward. The burrow morphologies of Goneplax rhomboides, Lesueurigobius friesii and Calocaris macandreae, all of which are very common in some Nephrops grounds in the Aegean and elsewhere, are described by Atkinson (1974 a, b, 1986), Nash (1980), Nash et al. (1984) and Marrs et al. (1996). Particularly careful examination of the video material and a recent comprehensive compilation of existing and new information on the structural complexity of Nephrops burrows including a diagnostic burrow identification key provided by Marrs et al. (1996) aided the identifications. Confirmation of identification of Nephrops burrow associates was ground truthed through trawl collected material. To minimise interpretational variability all burrow identification work was made by an expert analyst. Trawl methods Estimates of animal densities were made from two replicate trawls in the vicinity of Station E during each of the sampling periods (Table 1). The bottom trawl was a traditional Greek commercial trawl with a codend mesh size of 26 mm (stretched). All trawls were conducted during daylight hours. Speed of tow was approximately 2 knots. D-GPS position and depth were noted when the trawl was on the bottom with wires fixed, thereafter position and depth were noted every 5 min until hauling. Trawl duration was approximately 60 min. All Nephrops were weighed and counted and densities were standardised to 100 m2 areas for comparative purposes. Regression analyses were used to investigate the relationship between the trawl estimates of animal density and weight and video estimates of burrow density. Stock assessment The area inhabited by Nephrops was defined as that below the 60 m depth contour and calculated to be equivalent to 376 km2 (pers. comm. A. Polani, IMBC). The mean burrow density from video (from burrow counts) and individual weight (from trawl data) were used to produce stock number (mean burrow count per unit area×total area) and biomass (mean individual weight per unit area×total area) estimates for the periods May to November and February. Although to date most research groups working on the subject, having carefully screened out any unoccupied burrows from their counts, assume 100% occupancy rates, stock estimates given here also include figures for 75% occupancy as a conservative measure. Commercial gillnet catches for the period 1 March 1998 to 28 February 1999 were recorded for all the boats employed in the Nephrops fishery and operating in the bay by G. Ecomomou (Fisheries Inspectorate, Volos). Fishing boats are 6–10 m long caiques with less than 200 HP engines. Sixteen fishing boats operate from Milina in the southeast of the bay representing the most active part of the fleet. Their area of operations is the central and eastern part of the bay at 80–90 m depth and on occasion deeper grounds on the eastern quarter. Six boats operate from Volos on a less regular basis (involved with other fishing gears and/or working outside the bay). Their main area of operations is the central part of the bay at 80–90 m depth with one fishing boat operating in shallower waters (40–80 m) and targeting mostly M. merluccius. Results Video appraisal of the grounds The video view of the seabed showed a high degree of microtopography caused by bioturbation at all nine stations in Pagasitikos Bay. This was conspicuously sculptured into a variety of burrows and mounds. Sediments appeared to be soft with a brownish colouration interspersed with lighter patches of freshly excavated material. Burrows seen were characteristic of the crustaceans N. norvegicus, G. rhomboides, Calocaris macandreae, and various thalassinideans shrimps as well as the burrowing fish C. rubescens and L. friesii. The largest burrows were those of C. rubescens, which were found in aggregations. Serranus hepatus, P. longirostris, an unidentified flatfish, Lepidotrigla cavillone and Trisopterus minutus capellanus were also occasionally seen. A small variety of sessile surface dwelling fauna was noted, mainly soft corals (Funiculina quadrangularis and Pennatula phosphorea) and cerianthid anemones. The echinoderms Stichopus regalis and Marthasterias glacialis were seen infrequently. In May 1998 the sediments at channel station (Station I) were covered by a “fluff”, assumed to be the fall-out and accumulation of planktonic detritus. During the February 1999 survey, all stations were affected to a much worse extent by thick patches of detritus on the surface and drifting particle aggregations. This was particularly noted in the northerly and westerly stations. Some areas of dark grey sediment were also observed where the seabed could have been anoxic, perhaps caused by the death of infaunal organisms. A number of fishing lines or thin anchor lines (for pelagic purse seine floats) were noted on the seabed as well as some old nets. Trawling marks were noted in the channel station (I) during the May sampling period. Estimates of Nephrops density Mean density ranged from less than one burrow per 100 m2 in February in Station I in the channel to almost 17 burrows per 100 m2 at Station B in May (Table 2). The highest densities were found in the northerly and westerly stations (A–E). Lowest densities were found in the channel area to the south of the bay, with slightly higher densities in the middle of the bay. No differences were found between stations in May, but significant differences were seen in August, November and February (Table 3). This is most easily visualised in Figure 2, showing contour plots (created with SURFER 7, gridding method: kriging) of burrow density in the bay for the different sampling periods. For most of the stations there was a significant difference in densities over time. The exceptions to this were Stations A, D and F. Stations A and D had high mean densities and large confidence intervals during some of the sampling periods. Station F was the one area to have relatively constant burrow density and variability. The notable overall trend for all the stations (except the shallowest station, Station A) was for a decrease in burrow density particularly in the February sampling period. Average decrease between May and subsequent sampling periods ranged between 4% (May–August) and 47% (May–February). Minimum and maximum average decrease between May and February was 22% in Station F and 83% in Stations I and G. Figure 2 Open in new tabDownload slide Contour plots (created with SURFER 7, gridding method: kriging) of Nephrops burrow density (mean density of Nephrops burrows 100 m−2, Table 3) across Pagasitikos Bay over time. Darker shading represents higher densities (range: 16–0 burrows 100 m−2, May 98: 14–4, August 98: 14–3, November 98: 11–3, February 99: 16–0). Figure 2 Open in new tabDownload slide Contour plots (created with SURFER 7, gridding method: kriging) of Nephrops burrow density (mean density of Nephrops burrows 100 m−2, Table 3) across Pagasitikos Bay over time. Darker shading represents higher densities (range: 16–0 burrows 100 m−2, May 98: 14–4, August 98: 14–3, November 98: 11–3, February 99: 16–0). Table 2 Mean density of Nephrops burrows (burrows 100 m−2) with 95% confidence intervals (95% CI) for each of the sampling stations and sampling periods. P values are from ANOVA within stations over the periods and between periods for the different stations (statistically significant values are shaded). . 05/98 . 08/98 . 11/98 . 02/99 . . Station . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . P value . A 9.60 10.92 10.25 3.99 7.51 1.60 16.60 6.80 0.0537 B 13.94 4.63 9.69 5.40 7.68 2.63 5.87 4.05 0.0491 C 11.46 2.77 8.35 4.82 5.88 3.24 3.69 2.92 0.0058 D 11.72 10.00 13.87 7.21 7.78 1.51 4.84 2.36 0.1010 E 9.39 3.71 5.94 4.05 9.79 1.48 3.60 1.99 0.0074 F 7.70 2.10 6.94 2.37 6.97 2.22 5.99 3.15 0.8361 G 8.46 4.28 11.34 8.20 3.52 4.63 1.55 1.03 0.0216 H 8.38 4.58 11.65 1.13 10.50 5.11 3.14 3.83 0.0093 I 4.46 1.68 3.23 2.18 4.23 3.16 0.75 0.76 0.0498 P value 0.1427 0.0112 0.0055 <0.001 . 05/98 . 08/98 . 11/98 . 02/99 . . Station . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . P value . A 9.60 10.92 10.25 3.99 7.51 1.60 16.60 6.80 0.0537 B 13.94 4.63 9.69 5.40 7.68 2.63 5.87 4.05 0.0491 C 11.46 2.77 8.35 4.82 5.88 3.24 3.69 2.92 0.0058 D 11.72 10.00 13.87 7.21 7.78 1.51 4.84 2.36 0.1010 E 9.39 3.71 5.94 4.05 9.79 1.48 3.60 1.99 0.0074 F 7.70 2.10 6.94 2.37 6.97 2.22 5.99 3.15 0.8361 G 8.46 4.28 11.34 8.20 3.52 4.63 1.55 1.03 0.0216 H 8.38 4.58 11.65 1.13 10.50 5.11 3.14 3.83 0.0093 I 4.46 1.68 3.23 2.18 4.23 3.16 0.75 0.76 0.0498 P value 0.1427 0.0112 0.0055 <0.001 Open in new tab Table 2 Mean density of Nephrops burrows (burrows 100 m−2) with 95% confidence intervals (95% CI) for each of the sampling stations and sampling periods. P values are from ANOVA within stations over the periods and between periods for the different stations (statistically significant values are shaded). . 05/98 . 08/98 . 11/98 . 02/99 . . Station . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . P value . A 9.60 10.92 10.25 3.99 7.51 1.60 16.60 6.80 0.0537 B 13.94 4.63 9.69 5.40 7.68 2.63 5.87 4.05 0.0491 C 11.46 2.77 8.35 4.82 5.88 3.24 3.69 2.92 0.0058 D 11.72 10.00 13.87 7.21 7.78 1.51 4.84 2.36 0.1010 E 9.39 3.71 5.94 4.05 9.79 1.48 3.60 1.99 0.0074 F 7.70 2.10 6.94 2.37 6.97 2.22 5.99 3.15 0.8361 G 8.46 4.28 11.34 8.20 3.52 4.63 1.55 1.03 0.0216 H 8.38 4.58 11.65 1.13 10.50 5.11 3.14 3.83 0.0093 I 4.46 1.68 3.23 2.18 4.23 3.16 0.75 0.76 0.0498 P value 0.1427 0.0112 0.0055 <0.001 . 05/98 . 08/98 . 11/98 . 02/99 . . Station . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . Mean . 95% CI . P value . A 9.60 10.92 10.25 3.99 7.51 1.60 16.60 6.80 0.0537 B 13.94 4.63 9.69 5.40 7.68 2.63 5.87 4.05 0.0491 C 11.46 2.77 8.35 4.82 5.88 3.24 3.69 2.92 0.0058 D 11.72 10.00 13.87 7.21 7.78 1.51 4.84 2.36 0.1010 E 9.39 3.71 5.94 4.05 9.79 1.48 3.60 1.99 0.0074 F 7.70 2.10 6.94 2.37 6.97 2.22 5.99 3.15 0.8361 G 8.46 4.28 11.34 8.20 3.52 4.63 1.55 1.03 0.0216 H 8.38 4.58 11.65 1.13 10.50 5.11 3.14 3.83 0.0093 I 4.46 1.68 3.23 2.18 4.23 3.16 0.75 0.76 0.0498 P value 0.1427 0.0112 0.0055 <0.001 Open in new tab Table 3 Density, weight and average individual weight of Nephrops from trawl catches for each of the sampling periods. P values are from ANOVA comparing between the sampling periods for each variable (significant values are shaded). Date . Individuals (inds 100 m−2) . Total weight (kg 100 m−2) . Average individual weight (kg) . 05/98 0.6385 0.0192 0.0300 08/98 0.3210 0.0114 0.0363 11/98 0.5749 0.0180 0.0317 02/99 0.1886 0.0068 0.0360 P value 0.0057 0.0014 0.6092 Date . Individuals (inds 100 m−2) . Total weight (kg 100 m−2) . Average individual weight (kg) . 05/98 0.6385 0.0192 0.0300 08/98 0.3210 0.0114 0.0363 11/98 0.5749 0.0180 0.0317 02/99 0.1886 0.0068 0.0360 P value 0.0057 0.0014 0.6092 Open in new tab Table 3 Density, weight and average individual weight of Nephrops from trawl catches for each of the sampling periods. P values are from ANOVA comparing between the sampling periods for each variable (significant values are shaded). Date . Individuals (inds 100 m−2) . Total weight (kg 100 m−2) . Average individual weight (kg) . 05/98 0.6385 0.0192 0.0300 08/98 0.3210 0.0114 0.0363 11/98 0.5749 0.0180 0.0317 02/99 0.1886 0.0068 0.0360 P value 0.0057 0.0014 0.6092 Date . Individuals (inds 100 m−2) . Total weight (kg 100 m−2) . Average individual weight (kg) . 05/98 0.6385 0.0192 0.0300 08/98 0.3210 0.0114 0.0363 11/98 0.5749 0.0180 0.0317 02/99 0.1886 0.0068 0.0360 P value 0.0057 0.0014 0.6092 Open in new tab The highest Nephrops density from trawl catches was found in May with 0.64 individuals 100 m−2 and the lowest in February with 0.18 individuals 100 m−2 (Table 3). Weight of Nephrops caught followed the same pattern as the animal density (highest in May, lowest in February). The average individual weight ranged between 0.030 and 0.036 kg. There were significant differences over time in both density and weight (Table 4), almost entirely due to the low values found in February. Both trawl and burrow density values follow the same trend of increase and decrease, with lowest values in February. As expected, animal density from trawling was far lower than burrow density, by a factor of 14.7–18.9 in May and February, respectively. Table 4 Regression coefficients between video and trawling estimates. Y-variable . X-variable . Intercept . Slope . R2 value . P value . Burrow density Trawl density 1.29 13.67 0.9454 0.0184 Burrow density Individual weight 0.19 503.99 0.9731 0.0090 Y-variable . X-variable . Intercept . Slope . R2 value . P value . Burrow density Trawl density 1.29 13.67 0.9454 0.0184 Burrow density Individual weight 0.19 503.99 0.9731 0.0090 Open in new tab Table 4 Regression coefficients between video and trawling estimates. Y-variable . X-variable . Intercept . Slope . R2 value . P value . Burrow density Trawl density 1.29 13.67 0.9454 0.0184 Burrow density Individual weight 0.19 503.99 0.9731 0.0090 Y-variable . X-variable . Intercept . Slope . R2 value . P value . Burrow density Trawl density 1.29 13.67 0.9454 0.0184 Burrow density Individual weight 0.19 503.99 0.9731 0.0090 Open in new tab Regression analyses were used to further investigate the relationship between the trawl estimates of animal density and individual weight and video estimates of burrow density (Table 4). Both regressions were significant with similar high level of fit, although with a marginally closer fit for burrow density against weight. Stock assessment for Pagasitikos Bay Using the overall means and confidence intervals for the May to November data and the February data (February was treated separately because of the eutrophication event) the total number and total weight of Nephrops were calculated assuming 100 and 75% burrow occupancy (Table 5). Overall values for May to November give some 31.6 and 23.7 million animals in the bay at 100 and 75% occupancy, respectively. This amounts to approximately 1034 and 775 t, respectively. Stock estimates for February indicate values 30 to almost 40% lower (33% in weight and 39% in numbers). Confidence intervals for the May to November period were based on a greater number of data values and were relatively low, but those of February were large. Table 5 Total stock (numbers and weight) of Nephrops in Pagasitikos Bay in 60+ m depth (area 376 km2) at 100 and 75% burrow occupancy (% Occ.) for the periods May–November and February. Mean values and 95% confidence intervals (95% CI) for N: number (.100 m−2) and W: weight (kg) are taken from all available data in the relevant periods. . . May–Nov . Feb . May–Nov . Feb . . % Occ. . Mean . 95% . Mean . 95% CI . Total stock . 95% CI . Total stock . 95% CI . N 100 8.411 0.79 5.12 1.65 31 631 768 2 972 334 19 271 424 6 220 761 75 6.308 0.593 3.84 1.24 23 723 826 2 229 250 14 453 568 4 665 570 W 100 0.033 0.036 1 034 359 97 195 693 771 223 947 75 775 769 72 896 520 328 167 961 . . May–Nov . Feb . May–Nov . Feb . . % Occ. . Mean . 95% . Mean . 95% CI . Total stock . 95% CI . Total stock . 95% CI . N 100 8.411 0.79 5.12 1.65 31 631 768 2 972 334 19 271 424 6 220 761 75 6.308 0.593 3.84 1.24 23 723 826 2 229 250 14 453 568 4 665 570 W 100 0.033 0.036 1 034 359 97 195 693 771 223 947 75 775 769 72 896 520 328 167 961 Open in new tab Table 5 Total stock (numbers and weight) of Nephrops in Pagasitikos Bay in 60+ m depth (area 376 km2) at 100 and 75% burrow occupancy (% Occ.) for the periods May–November and February. Mean values and 95% confidence intervals (95% CI) for N: number (.100 m−2) and W: weight (kg) are taken from all available data in the relevant periods. . . May–Nov . Feb . May–Nov . Feb . . % Occ. . Mean . 95% . Mean . 95% CI . Total stock . 95% CI . Total stock . 95% CI . N 100 8.411 0.79 5.12 1.65 31 631 768 2 972 334 19 271 424 6 220 761 75 6.308 0.593 3.84 1.24 23 723 826 2 229 250 14 453 568 4 665 570 W 100 0.033 0.036 1 034 359 97 195 693 771 223 947 75 775 769 72 896 520 328 167 961 . . May–Nov . Feb . May–Nov . Feb . . % Occ. . Mean . 95% . Mean . 95% CI . Total stock . 95% CI . Total stock . 95% CI . N 100 8.411 0.79 5.12 1.65 31 631 768 2 972 334 19 271 424 6 220 761 75 6.308 0.593 3.84 1.24 23 723 826 2 229 250 14 453 568 4 665 570 W 100 0.033 0.036 1 034 359 97 195 693 771 223 947 75 775 769 72 896 520 328 167 961 Open in new tab The commercial annual (3/98 to 2/99) Nephrops gillnet catch in the bay was 14.89 t. Sixteen fishing boats operated regularly from Milina across the central and eastern part of the bay with annual Nephrops catches of 11.35 t. Nephrops catches were higher in July 1998 (over 2000 kg) and lowest in Feb 1999 (129 kg). Nephrops represented 48% (by weight) of the total gillnet catch followed by Merluccius at 14%. The annual Nephrops catch of the remaining six boats was 3.49 t with higher catches in September 1998 (495 kg) and lowest in January 1999 (45 kg). Nephrops represented 38% (by weight) of the total gillnet catch followed by Merluccius at 23%. Discussion Nephrops burrow densities recorded here are much higher than other grounds in the Aegean (Anon., 1994; Smith et al., in press) and of comparable magnitude to those reported for the North Sea and the Adriatic (Bailey et al., 1993; Anon., 1994; Tuck et al., 1997b). The distribution of Nephrops within Pagasitikos was found to be heterogeneous, both within stations and between stations. In general the lowest number of burrows was found in the channel station and the highest in the north and west. Substantial variation in Nephrops burrow densities was also found in Scottish and Irish stocks (Chapman and Howard, 1988; Bailey et al., 1995; Tully and Hillis, 1995; ICES, 1996; Tuck et al., 1997a), mainly attributed to variability in environmental factors including hydrography and sediment particle size. While local hydrographic features (e.g. gyres) could lead to differential settlement of post larval juveniles, the relationship between Nephrops density and sediment grain size is “dome shaped” with an optimum particle size for peak Nephrops that is not the same for all the stocks (Afonso-Dias, 1997). The overall tonnage values estimated for Pagasitikos Bay (1034 t, 100% occupancy; 776 t, 75% occupancy) with the underwater video technique indicate a “large” albeit decreasing stock of Nephrops. The total annual commercial Nephrops catch from 22 fishing vessels operating mostly on the central and eastern part of the bay (at 80–100 m depth) was 14.89 t. This represents an annual removal of 1.5–2% (100–75% occupancy) of the stock. When compared to harvest rates of other stocks (ICES, 1998) this figure seems low (as this is a small scale small boat gillnet fishery), however, as catches are usually under-reported and removal due to other gears (including occasional illegal trawling evidenced by video) is not accounted for, it may be an underestimate. The most notable trend for almost all the stations (except the shallowest station A) was a decrease in burrow density over the study period and particularly so in February 1999 with an overall estimated stock decrease of 30%. A major eutrophication event in May 1997 resulted in fouling of bottom nets by smelly detritus that was difficult to remove and prevented local fishermen from fishing the central part of the bay for some months. All Nephrops catches that they did have, were noted to be very low. In May 1998 fall-out and accumulation of planktonic detritus was seen on video, but only in the channel station. By February 1999, however, this had spread across the larger part of the bay. Sediments covered by detritus and anoxic patches were noted in places. This was the most likely cause of the large reduction in both burrow density and Nephrops trawl catches during February. Eutrophication events in the Kattegat and Skaggerak in the mid-1980s have caused bottom oxygen depletion leading to eventual sediment anoxia. As Nephrops were forced out of their burrows, increased catches were recorded by the trawl fishery, then decreased as Nephrops were removed or died as a result of further decreases in oxygen content (Bagge and Munch-Petersen, 1979; Rosenberg, 1986). In a serious oxygen crisis in the Adriatic after the expected initial increase, Nephrops catches decreased substantially with a progressive increase of the mean CL in the surviving population due to heavy mortalities suffered by the juveniles (Froglia and Gramitto, 1987). In Pagasitikos, the mean CL increased significantly (p>0.0001) from May 1998 to February 1999 from 34.77 to 36.71 mm (Smith et al., 2001). In July 1996, 2 years prior to our May 1998 sampling and a year prior to the eutrophication event in May 1997, the mean CL from trawl sampling close to station E was 34.58 mm (Smith et al., in press). The mean average weight was the same as the May 1998 sampling, however the mean burrow and trawl density estimates were 30 and 70% higher, respectively (Smith et al., in press). The mean CL from trawl sampling in 1994 close to station F was even lower at 33.48 mm while trawl densities were twice that of May 1998 (Smith and Papadopoulou, 1999). It is therefore highly probable that eutrophication events have had a marked effect on Pagasitikos Nephrops populations. The support of the EU through the funding of the Study project 96/013 “Growth and natural mortality of N. norvegicus with an introduction and evaluation of creeling in Mediterranean waters” is gratefully acknowledged. Commercial gillnet catches from the local fishery were provided by G. Ecomomou, Fisheries Inspectorate, Volos. Dr A. Banks and A. Pollani (IMBC) assisted with data visualisation and area delimitation. Two anonymous referees provided helpful comments on the manuscript. References Afonso-Dias M. S. . Variability of Nephrops norvegicus (L.) populations in Scottish waters in relation to the sediment characteristics of the seabed , 1997 PhD thesis, University of Aberdeen. 282 pp Anon Nephrops norvegicus: stock variability and assessment in relation to fishing pressure and environmental factors , 1994 European Commission Final Report, DG XIV, Med. 91/003. 209 pp Atkinson R.J.A . Spatial distribution of Nephrops burrows , Estuarine and Coastal Marine Science , 1974 , vol. 2 (pg. 171 - 176 ) Google Scholar Crossref Search ADS WorldCat Atkinson R.J.A . Behavioural ecology of the mud-burrowing crab Goneplax rhomboides , Marine Biology , 1974 , vol. 25 (pg. 239 - 252 ) Google Scholar Crossref Search ADS WorldCat Atkinson R.J.A . Mud burrowing megafauna of the Clyde Sea Area , Proceedings of the Royal Society of Edinburgh , 1986 , vol. 90B (pg. 351 - 362 ) OpenURL Placeholder Text WorldCat Atkinson R.J.A , Chapman C.J . Resin casting: a technique for investigating burrows in sublittoral sediments , Progress in Underwater Science , 1984 , vol. 9 (pg. 15 - 25 ) OpenURL Placeholder Text WorldCat Atkinson R.J.A , Pullin R.S.V . Observations on the burrows and burrowing behaviour of Cepola rubsecens L. P.S.Z.N.I , Marine Ecology , 1996 , vol. 17 (pg. 23 - 40 ) Google Scholar Crossref Search ADS WorldCat Atkinson R.J.A , Pullin R.S.V , Dipper F.A . Studies on the red band-fish, Cepola rubsecens , Journal of Zoology (London) , 1977 , vol. 182 (pg. 369 - 384 ) Google Scholar Crossref Search ADS WorldCat Bagge O , Munch-Petersen S . Some possible factors governing the catchability of Norway lobster in the Kattegat , Rapports et Proces-Verbaux des Reunions Consell International pour l'Exploration de la Mer , 1979 , vol. 175 (pg. 143 - 146 ) OpenURL Placeholder Text WorldCat Bailey N. , Chapman C. J. , Kinnear J. , Bova D. , Weetman A. . Estimation of Nephrops stock biomass on the Fladen Ground by TV survey , 1993 ICES CM 1993/K: 34 Bailey N. , Chapman C. J. , Afonso-Dias M. , Turrell W. . The influence of hydrographic factors on Nephrops distribution and biology , 1995 ICES CM 1995/Q: 17 Briggs R.P , Armstrong M.J , Dickey-Collas M , Allen M , Mc Quaid N , Whitmore J . The application of fecundity estimates to determine the spawning stock biomass of Irish Sea Nephrops norvegicus (L.) using the annual larval production method , ICES Journal of Marine Science , 2002 , vol. 59 1 (pg. 109 - 119 ) Google Scholar Crossref Search ADS WorldCat Chapman C.J . Cobb S , Phillips B . Ecology of juvenile and adult Nephrops , The Biology and Management of Lobsters , 1980 , vol. Volume II New York Academic Press (pg. 143 - 178 ) Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Chapman C.J , Howard F.G . Environmental influences on Norway lobster (Nephrops norvegicus) populations and their implications for fishery management , Symposia of the Zoological Society of London , 1988 , vol. 59 (pg. 343 - 353 ) OpenURL Placeholder Text WorldCat Chapman C.J , Rice A.L . Some direct observations on the ecology and behaviour of the Norway lobster Nephrops norvegicus (L.) , Marine Biology , 1971 , vol. 10 (pg. 321 - 329 ) Google Scholar Crossref Search ADS WorldCat Chapman C.J , Shelton P.M.J , Shanks A.M , Gaten E . Survival and growth of the Norway lobster, Nephrops norvegicus in relation to light-induced eye damage , Marine Biology , 2000 , vol. 136 (pg. 233 - 241 ) Google Scholar Crossref Search ADS WorldCat ETANAL Statistical report of the fisheries markets , Fishing News , 1998 , vol. 208 (pg. 85 - 89 ) in Greek OpenURL Placeholder Text WorldCat Fariña A.C , Freire J , Gonzalez-Gurriaran E . Fecundity of the Norway lobster (Nephrops norvegicus (L.)) in Galicia (NW Spain) and a review of geographical patterns , Ophelia , 1999 , vol. 50 3 (pg. 177 - 189 ) Google Scholar Crossref Search ADS WorldCat Froglia C , Gramitto M.E . Caddy J.F , Savini M . , An estimate of growth and mortality parameters for Norway lobster (Nephrops norvegicus) in the central Adriatic Sea , 1987 Rome, Italy FAO GFCM (pg. 189 - 203 ) Report no. 394 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gonzalez-Gurrianan E , Freire J , Fariña A.C , Fernandez A . Growth at moult and intermoult period in the Norway lobster Nephrops norvegicus from Galician waters , ICES Journal of Marine Science , 1998 , vol. 55 (pg. 924 - 940 ) Google Scholar Crossref Search ADS WorldCat ICES Report of the Working Group on the assessment of Nephrops and Pandalus stocks , 1995 ICES CM 1995/Assess:11 ICES Report of the Study Group on life histories of Nephrops , 1996 ICES CM 1996/K:2 Ref. Assess. 107 pp ICES Report of the Study Group on life histories of Nephrops , 1998 ICES CM 1998/G:9 Ref. ACFM. 132 pp ICES Report of the Study Group on life histories of Nephrops , 1999 ICES CM 1999/G:13 Ref. ACFM. 41 pp ICES Overview of Nephrops stocks , 2001 (pg. 247 - 248 ) ICES CM 2001/Ref. ACFM Marrs S. J. , Atkinson R. J. A. , Smith C. J. , Hills J. M. . Calibration of the towed underwater TV technique for use in stock assessment of Nephrops norvegicus , 1996 European Commission Final Report, DG XIV, 94/069. 155 pp Marrs S. J. , Atkinson R. J. A. , Smith C. J. , Hills J. M. . The towed underwater TV technique for use in stock assessment of Nephrops norvegicus , 1998 (pg. 88 - 98 ) ICES CM 1998\G: 9 Mytilineou Ch , Castro M , Cancho P , Fourtouni A . Growth studies on Norway lobster Nephrops norvegicus (L.) in different areas of the Mediterranean Sea and the adjacent Atlantic , Scientia Marina , 1998 , vol. 62 Suppl. 1 (pg. 43 - 60 ) OpenURL Placeholder Text WorldCat Nash R. D. M. . The behavioural ecology of small demersal fish associated with soft sediments , 1980 PhD thesis, University of Glasgow. 156 pp Nash R.D.M , Chapman C.J , Atkninson R.J.A , Morgan P.J . Observations on the burrows and burrowing behaviour of Calocaris macandeae Bell (Crustacea: Decapoda: Thalassinoidea) , Journal of Zoology (London) , 1984 , vol. 202 (pg. 425 - 439 ) Google Scholar Crossref Search ADS WorldCat NCMR Development of policy for sustainable management of Pagasitikos bay: planktonic–benthic ecosystems and ecotoxicology (in Greek) , 2000 Final Report to Magnisia Local Authority. 139 pp Rosenberg R . Eutrophication—the future marine coastal nuisance , Marine Pollution Bulletin , 1986 , vol. 16 6 (pg. 227 - 231 ) Google Scholar Crossref Search ADS WorldCat Sardà F . Nephrops norvegicus comparative biology and fishery in the Mediterranean Sea , Scientia Marina , 1998 , vol. 62 Suppl. 1 pg. 143 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smith C. J. , Papadopoulou K. N. . The environmental impact of demersal fishing gears on the marine environment , 1999 European Commission Final Report, DG XIV, Study Project TR/MED93/012. 153 pp Smith C. J. , Papadopoulou K.-N. , Kallianiotis A. , Vidoris P. , Chapman C. J. , Vafidis D. . Growth and natural mortality of Nephrops norvegicus, with an introduction and evaluation of creeling in Mediterranean waters , 2001 European Commission Final Report, DG XIV, Study Project 96/013. 195 pp Smith C. J. , Marrs S. J. , Atkinson R. J. A. , Papadopoulou K.-N. , Hills J. M. . Evaluation of the underwater television technique for fisheries-independent stock assessment of Nephrops norvegicus from the Aegean (Eastern Mediterranean) , Marine Ecology Progress Series , 2002 (in press) OpenURL Placeholder Text WorldCat Tuck I.D , Atkinson R.J.A , Chapman C.J . The structure and seasonal variabillity in the spatial distribution of Nephrops norvegicus burrows , Ophelia , 1994 , vol. 40 1 (pg. 13 - 25 ) Google Scholar Crossref Search ADS WorldCat Tuck I.D , Chapman C.J , Atkinson R.J.A . Population biology of the Norway lobster, Nephrops norvegicus (L.) in the Firth of Clyde, Scotland—I: growth and density , ICES Journal of Marine Science , 1997 , vol. 54 (pg. 123 - 135 ) Google Scholar Crossref Search ADS WorldCat Tuck I.D , Chapman C.J , Atkinson R.J.A , Bailey N , Smith R.S.M . A comparison of methods for stock assessment of the Norway lobster, Nephrops norvegicus in the Firth of Clyde , Fisheries Research , 1997 , vol. 32 (pg. 89 - 100 ) Google Scholar Crossref Search ADS WorldCat Tuck I. D. , Bailey N. , Atkinson R. J. A. , Marrs S. . Changes in Nephrops densities in the Clyde Sea area, from underwater TV survey data , 1999 (pg. 24 - 31 ) ICES CM 1999/G: 13 Tuck I.D , Atkinson R.J.A , Chapman C.J . Population biology of the Norway lobster, Nephrops norvegicus (L.) in the Firth of Clyde, Scotland. II: fecundity and size at onset of sexual maturity , ICES Journal of Marine Science , 2000 , vol. 57 (pg. 1227 - 1239 ) Google Scholar Crossref Search ADS WorldCat Tully O , Hillis J.P . Causes and spatial scales of variability in population structure of Nephrops norvegicus (L.) in the Irish Sea , Fisheries Research , 1995 , vol. 21 (pg. 329 - 347 ) Google Scholar Crossref Search ADS WorldCat © 2003 International Council for the Exploration of the Sea