ICES Journal of Marine Science: Journal du Conseil Advance Access published online on January 25, 2008
ICES Journal of Marine Science: Journal du Conseil, doi:10.1093/icesjms/fsm192
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Evidence from survey data for regional variability in cod dynamics in the North Sea and West of Scotland
FRS Marine Laboratory, Aberdeen AB9 11DB, UK
Correspondence to S. J. Holmes: tel: +44 1224 295507, fax: +44 1224 295511; e-mail: s.holmes{at}marlab.ac.uk
Holmes, S. J., Wright, P. J., and Fryer, R. J. 2008. Evidence from survey data for regional variability in cod dynamics in the North Sea and West of Scotland.–ICES Journal of Marine Science, 65.Although cod (Gadus morhua) in the North Sea and ICES Division VIa are assessed as single units, recent research suggests that the stocks consist of reproductively isolated subpopulations within a metapopulation. We investigate whether temporal trends in stock indicators are asynchronous across subpopulations, which would support the metapopulation hypothesis. First quarter trawl survey data for the years 1983–2005 were aggregated into putative areas of high spawner fidelity (three in VIa, seven in the North Sea) to obtain indices of spawning–stock biomass (SSB) and recruitment (numbers-at-age 1). Asynchrony was investigated by fitting a smoother to the data for each of the ten spawning areas and testing whether the smoothers were parallel. Trends in SSB differed between spawning areas in both VIa and the North Sea. In VIa, SSB collapsed in the most southwesterly area, but remained more constant elsewhere. In the North Sea, there was a general decline in SSB, but areas thought to contain resident inshore populations showed more rapid declines than those in adjacent offshore areas. Recruitment results offered less support for a metapopulation, although recruitment in the southern North Sea declined rapidly before any trend was seen for the North Sea as a whole.
Keywords: cod, metapopulation, recruitment, smoothers, SSB, survey indices
Received 21 June 2007; accepted 3 December 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Although current fishery management is usually based on single-stock assumptions, many marine fish stocks consist of several spawning components with varying degrees of reproductive segregation (McQuinn, 1997; Smedbol and Stephenson, 2001; Thorrold et al., 2001). Such stocks are considered to be population rich (Sinclair and Iles, 1988). Failure to account for population richness in fishery management may lead to the depletion of stock components, with unknown ecological consequences (Stephenson, 1999; Frank and Brickman, 2000), and may critically affect the long-term stability and sustainability of an entire stock (Hilborn et al., 2003). The northern cod (Gadus morhua) stock off Newfoundland is perhaps the best known example of population richness, and its collapse involved several component populations (Hutchings, 1996; Ruzzante et al., 1999; Smedbol and Wroblewski, 2002). Since 1996 there has been evidence of recovery in the inshore populations, but no evidence of recovery offshore (DFO, 2007).
The cod stocks in the North Sea and off the Scottish west coast (ICES Division VIa) are in a poor state (ICES, 2007a, b), with spawning biomasses at historical lows. Although their management has primarily been through limitations on total stock landings, areas have also been closed (CEC, 2007). Advice on recovery plans and particularly the utility of closing areas needs to take into account any differences in the dynamics of spawning groups within the stocks, if such groups exist.
Cod in the North Sea and VIa tend to exhibit a high degree of spawning fidelity (Bedford, 1966; Neat et al., 2006; Wright et al., 2006a), and otolith microchemistry analyses indicate that adults from coastal regions originate from local nursery areas (Wright et al., 2006b). A study of microsatellite DNA variation has indicated at least four divergent cod populations in the North Sea (Hutchinson et al., 2001); a structure supported by the extent of movements estimated from tag recaptures (Bedford, 1966; Wright et al., 2006a; Righton et al., 2007). Despite this evidence for natal philopatry, there are examples from tagging studies too of quite distant straying (Bedford, 1966; Wright et al., 2006a). Such evidence of subpopulations along with evidence of straying between segregated spawning grounds suggests a metapopulation structure. Metapopulations are characterized by subpopulations with differing demographic rates and population fluctuations linked by some movement and gene flow, and by the potential for local populations to undergo extinction and recolonization (Hanski, 1999; Smedbol and Wroblewski, 2002; Kritzer and Sale, 2004). Evidence for recolonization of one North Sea spawning area has been found from a study of long-term genetic variation (Hutchinson et al., 2003). However, no study has tested whether the dynamics of different spawning groups in North Sea and VIa cod are asynchronous, as would be expected if the spawning groups behaved like local populations within a metapopulation (Hanski, 1998). Asynchronous population dynamics is important to metapopulation persistence, because it makes the simultaneous extinction of all local populations unlikely.
In this study, indices of the recruitment and spawning–stock biomass (SSB) of putative populations are used to test for asynchrony within the North Sea and VIa cod stocks. The indices are constructed from research vessel survey data derived from the first quarter of the years 1983–2005. Ten putative areas of high spawner fidelity are considered, three in VIa and seven in the North Sea. Temporal trends in indices of recruitment and SSB are modelled by fitting a separate smoother to the log-transformed indices for each area. The smoothers are then compared across areas, parallel smoothers indicating synchronous dynamics and divergent smoothers indicating asynchronous dynamics.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Spawning areas
Putative population boundaries were defined using published genetic, tag-recapture, otolith shape, and chemistry information, together with information on spawning areas from the distribution of mature cod in the first quarter of the year (Table 1, Figure 1). Otolith shape may reflect environmental segregation (Galley et al., 2006), and otolith microchemistry has been used to determine the nursery origin of spawning adults in Scottish coastal waters (Wright et al., 2006b). Significant fine-scale genetic differentiation between the northern North Sea off Bergen Bank, within the Moray Firth, off Flamborough Head, and the Southern Bight suggests that spawning grounds in areas 5, 6, 7–8–9, and 10 are reproductively isolated (Hutchinson et al., 2001). Where tag data were available, boundaries were based on the approximate extent of tag recaptures from spawning grounds. Defining boundaries that reflect the home range of cod from subpopulations is nevertheless difficult because cod may use several spawning grounds within a larger spawning area, such as off the north and west of Scotland (area 3; Wright et al., 2006a). The 100 and 50 m depth contours were used to delimit the northern boundaries of areas 8 and 10, because tag recaptures suggest that they reflect the range of cod released north and south of the Dogger Bank, respectively (Bedford, 1966; Neat and Righton, 2007). Flamborough (area 9) was separated from the German Bight (area 10) using tag-recapture data, although genetic evidence suggests that the Flamborough area was recolonized (after local depletion) by cod from adjacent areas in the 1970s (Hutchinson et al., 2003).
|
|
Area indices
Indices of numbers-at-age from 1983 to 2005 were obtained for areas 1–3 from the first quarter west coast Scottish groundfish survey (ScoGFS) conducted by Fisheries Research Services, Aberdeen, and for areas 4–10 from the first quarter international bottom-trawl survey (IBTS), with data downloaded from the ICES ‘DATRAS’ web-based database (ICES, 2001). Both surveys provide numbers-at-age by ICES statistical rectangle, 1° longitude by 0.5° latitude. The IBTS has two hauls per rectangle, and the ScoGFS usually has at least one haul per rectangle (ICES, 2001). Indices for each spawning area were obtained as arithmetic means over the rectangles within that area, the same procedure used by ICES to produce indices for its stock assessment areas (ICES, 2001).
Indices of SSB were obtained by the sum of products
|
| (1) |
Statistical analysis
Asynchrony was investigated by testing whether cod in different putative spawning areas showed similar or divergent SSB and recruitment dynamics, i.e. by testing whether the time-series for each area could be adequately modelled by parallel smoothers. The SSB and recruitment indices were log-transformed to homogenize variances within areas, with the occasional zero index first replaced by one-half of the lowest non-zero value from the same area, and an unusually low SSB index for the Shetland area replaced by one-half of the next lowest index from Shetland. An additive model with a separate smoother for each area was then fitted to the log-transformed indices and compared with a model in which the smoothers were constrained to be parallel. To homogenize variances across areas, indices were weighted by the size of the area (the number of ICES rectangles), with those derived from IBTS data given twice the weight of those derived from ScoGFS data, reflecting the different number of hauls per rectangle in each survey. The effective degrees of freedom of the smoothers were determined by generalized cross-validation (Wood, 2006) using the ‘mgcv' library of the R software package. An F-statistic was used to compare the two models. This was referred to an empirical F-distribution, constructed by bootstrapping, that accounted for any within-year correlation in indices across areas (Youngson et al., 2002). Within-year correlations were assessed using the correlation matrix of the residuals after fitting a separate smoother for each area.
The global tests described above show whether there are asynchronous dynamics somewhere in the North Sea or VIa, but do not indicate specifically where the differences are. The global tests were therefore supplemented by pairwise comparisons in which a smoother was fitted to the differences between the log indices from two areas, and compared with a constant fit by a standard F-test. Modelling the difference in log indices accounts for any correlation between areas without the need for bootstrapping (Youngson et al., 2002). Bonferroni corrections were used to account for the number of comparisons. As there appears to be little exchange of juveniles and adults between the west of Scotland and the North Sea (Wright et al., 2006a; Gibb et al., 2007), pairwise comparisons were considered separately for the two regions. The pairwise comparisons were also visualized using reference bands (Bowman and Azzalini, 1997): the trends (in SSB or recruitment) differ between two areas when the smoothers move outside their reference bands.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Spawning–stock biomass
All the putative spawning areas, with the possible exception of Shetland, show a decline in SSB (Figure 2). There are differences, however, in the extent of the decline across areas. In VIa, SSB appears to have collapsed in the Southwest area; in three of the last 4 years, no fish of mature age were recorded and the remaining year gives the lowest non-zero value. The VIa indices, using data from all hauls in areas 1–3, show a decline in SSB, but not one as severe as that in Southwest. In the North Sea, the decline in the large offshore areas appears greater passing from north to south (Viking to Fisher to Dogger). Also, the decline in inshore areas appears to be greater than in offshore areas of similar latitude (Flamborough compared with Dogger, East Coast compared with Fisher, and Moray compared with Fisher or Viking). The North Sea indices, using data from all hauls in areas 4–10, are an average for the individual areas, and as such gives no indication of the accelerating declines in Moray and Flamborough.
|
There was negligible evidence of any within-year correlation of SSB indices across areas (Table 2, lower left triangle). The global F-tests found significant differences in SSB trends between areas (all areas, p < 0.001; North Sea, p = 0.006; VIa, p = 0.011). The results of pairwise comparisons are shown in Table 3 and Figure 3. In VIa, the Southwest area has a much steeper decline in SSB than both the adjacent Minch area to the north (p < 0.01) and the Clyde (p < 0.05). In the North Sea, the relatively large number of areas made few pairwise comparisons significant once a Bonferroni correction was applied, but the trend in the Flamborough area still differed significantly from those in the offshore areas Viking, Fisher, and Dogger (p < 0.001, <0.01, <0.01, respectively) and that in Shetland (p < 0.05). The decline in SSB in Flamborough was more marked than in other areas and appeared to accelerate in recent years (Figure 3, second bottom row).
|
|
|
Recruitment
Time-series of recruitment indices (numbers-at-age 1) for the putative spawning areas are shown in Figure 4. The recruitment indices are more variable than the SSB indices, with strong within-year correlations across areas (Table 2, upper right triangle). Generally, the strongest correlations are between adjacent areas. There are visual differences in trends between areas (Figure 4). The Dogger area displays a long-term decline in recruitment. The large offshore areas farther north (Fisher, Viking) indicate improving recruitment, followed by a downturn relatively recently. Among the inshore areas, there is a suggestion of opposite trends for the Moray and East Coast areas.
|
The trends in recruitment differed between all areas (p = 0.004) and between areas within the North Sea (p = 0.009), but not between areas within VIa (p = 0.22). Pairwise comparisons are shown in Table 4 and Figure 5. The only pairwise comparison significant at the 5% level following Bonferroni correction was between Moray (increasing trend) and East Coast (decreasing trend). Without the correction, the trend in Moray differed from that in all other North Sea areas apart from Shetland, and the decreasing trend in Dogger differed from the trends in Viking, Fisher, and Flamborough, where recruitment was fairly constant or rising in the early part of the time-series.
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Most putative spawning areas showed a decline in SSB over the study period (1983–2005). Although this might indicate widespread environmental stress or overexploitation, it does not necessarily indicate a lack of population structure. It is legitimate to search for demographic patterns overlying a region-wide decline. For example, Rothschild (2007) identified five stock complexes of Northwest Atlantic cod, while finding a common peak in SSB among all stocks and coherent changes in weight-at-age. Here, the significant differences in SSB trends within both the North Sea and VIa are consistent with asynchronous population dynamics across spawning areas, which in turn is consistent with a metapopulation structure (Hanski, 1999).
There were few significant differences in recruitment trends across areas. In part, this could be due to the high interannual variability in the recruitment time-series (Figure 4), which will have limited the power of the analysis. The within-year correlations in the recruitment indices across areas suggest widespread environmental influences on recruitment, e.g. affecting survival of eggs and larvae. These correlations are unlikely to be due to year effects in the surveys, because the SSB indices showed negligible within-year correlations across areas. Rothschild (2007) concludes that SSB is a better stock differentiator than recruitment because it acts as a natural filter that integrates or averages recruitment.
Recruitment in Viking was low relative to the SSB found there, whereas in Fisher and Dogger, farther south, recruitment was high relative to SSB (Figures 2 and 4). This suggests that young fish move north as they become older, something that would undermine the subpopulations proposed. However, tag-recapture experiments on age 1 cod initially captured in the southeast North Sea indicate that their northward movements are limited to the central North Sea (Riley and Parnell, 1984). In contrast to older age classes, cod aged 1 typically concentrate in winter in shallow waters of many parts of the North Sea and coastal waters of the Scottish west coast, and this may influence their accessibility to first quarter surveys (Heessen, 1983; Riley and Parnell, 1984; Gibb et al., 2007). Several shallow sites are sampled at Dogger and Fisher, whereas the sampling sites at Viking are deeper. If juveniles at Viking are concentrated in coastal waters or on the top of banks, they may not be sampled adequately. A similar effect can be expected in VIa in the Minch area.
The spatial resolution of the surveys limited the minimum geographical size of the populations that could be differentiated by the approach used here. The variability of each index was approximately proportional to the number of hauls on which the index was based, and hence the large variability in the Shetland indices (Figures 2 and 4). A more powerful assessment of whether small areas hold populations with unique demographics would require time-series constructed from more-intensive sampling regimes. Care would be needed, however, when combining data from different surveys, which may have different selectivities for a given species. The use of two surveys in this study was not an issue, because both are ICES IBTS and have been standardized as far as possible. Also, the indices for a given area were compiled using only one survey, so were consistent over time.
From their analysis of decadal mean cod densities in first quarter IBTS surveys, Hedger et al. (2004) concluded that the spatial distribution of mature cod had shifted towards deeper water in the northern North Sea between the 1980s and 1990s. Using data from the third quarter English groundfish survey, Perry et al. (2005) also proposed a northward movement in cod distribution, and found significant correlations with three measures of North Sea warming. Such results could be interpreted as a distributional shift of a panmictic population, but our study here suggests that they may in fact represent changes in the relative abundance of local populations.
Even if separate local populations have been identified correctly, it is not possible to say to what extent differences in SSB trends are attributable to internally driven population dynamics or differing environmental or fishing pressures. Bjørnstad et al. (2004) showed how age-structured interactions and stochastic recruitment could induce low-frequency variability at the stock level when the model is based on cod life history (density-dependent survival through intra- and inter-cohort competition; lifespan and maturity-at-age based on Norwegian Skagerrak cod). The time-scale of the resulting low frequency variation was sufficiently long potentially to mimic variation in abundance linked to environmental change or overexploitation. A longer time-series of data would help to resolve the issue, as would spatially resolved data on effective fishing effort.
We conclude by considering the practical implications for stock assessments and management of accepting the metapopulation structure considered. Assessments of some stocks, such as cod in VIa and the North Sea, place greater emphasis on research vessel survey abundance indices than on commercial catches. Indices formed from arithmetic means across statistical rectangles are simple to construct and are proportional to abundance (ICES, 2001). Their use, however, will lead to any large and relatively homogenous area dominating the signal from distinct but smaller areas if they are combined. For example, a sharp decline in SSB in recent years has been evident in the stock assessments for cod in VIa, but because one index is produced for the whole of VIa, no account can be taken of the apparently contrasting fortunes of the Clyde and Minch areas compared with the offshore Southwest area.
Knowing that separate spawning areas belong to separate subpopulations rather than being alternative sites for a single stock might alter the perception of area closures designed to protect spawning aggregations. However, mixing of component subpopulations over a large area outside the spawning period would likely render effective permanent closed areas impracticable. Under such circumstances, focus on achieving a fishing mortality low enough to protect the most vulnerable stock component may be more appropriate. Notwithstanding, studies used in the definition of putative subpopulation areas (Figure 1, Table 1) have indicated generally that year-round cod movements remained within their boundaries (Wright et al., 2006b; Righton et al., 2007).
Of more immediate concern is the strong declines in recruitment and SSB in the Dogger and Flamborough areas. The Flamborough area is believed to have suffered a local extinction in the past (Hutchinson et al., 2003), and the trends in both recruitment and SSB point to future (commercial) extinction before anything similar is seen for the North Sea as a whole. Assuming metapopulation structure, and given genetic evidence, one might expect eventual recolonization from the adjacent Fisher area over a period of decades. Evidence from the Canadian northern cod stock, however, highlights the potential delay in recolonization, because there is no evidence for significant immigration of adults across inshore–offshore substock boundaries since the fishing moratorium in 1992 (DFO, 2007). We may be witnessing an even more advanced stage of subpopulation collapse in the Southwest area of VIa.
|
Acknowledgements |
|---|
We thank ICES and the member states that contribute to the IBTS survey for permission to use the data held on the DATRAS database. We are also grateful to E. Galley for assistance with Figure 1, and P. Kunzlik and C. Needle for useful discussions on the manuscript. Comments by George Rose, Georg Engelhard, and an anonymous reviewer greatly improved the manuscript. This work was conducted as part of a contract with the European Union METACOD Q5RS–2001–00953 and the Scottish Executive MF0756.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Bedford B. C. English cod tagging experiments in the North Sea. ICES Document CM 1966/G: 9 (1966) 29. pp.
Bjørnstad O. N., Nisbet R. M., Fromentin J. Trends and cohort resonant effects in age-structured populations. Journal of Animal Ecology (2004) 73:1157–1167.[CrossRef][Web of Science]
Bowman A. W., Azzalini A. Applied Smoothing Techniques for Data Analysis. The Kernel Approach with S-Plus Illustrations. (1997) Oxford: Clarendon Press. 193. pp.
CEC. Council Regulation (EC) No 41/2007 of 21 December 2006 fixing for 2007 the fishing opportunities and associated conditions for certain fish stocks and groups of fish stocks, applicable in Community waters and, for Community vessels, in waters where catch limitations are required. Official Journal of the European Union, L 15/1 (2007).
DFO. Stock Assessment of Northern (2J3KL) cod in 2007. DFO Canadian Science Advisory Secretariat Science Advisory Report, 2007/018 (2007) 25.
Frank K. T., Brickman D. Allee effects and compensatory population dynamics within a stock complex. Canadian Journal of Fisheries and Aquatic Sciences (2000) 57:513–517.
Galley E. A., Wright P. J., Gibb F. M. Combined methods of otolith shape analysis improve detection between spawning areas of Atlantic cod. ICES Journal of Marine Science (2006) 63:1710–1717.
Gibb F. M., Gibb I. M., Wright P. J. Isolation of Atlantic cod (Gadus morhua) nursery areas. Marine Biology (2007) 151:1185–1194.[CrossRef]
Graham M. The annual cycle in the life of the mature cod in the North Sea. Fishery Investigations London, Series 2 (1924) 6(6). 77 pp.
Hanski I. Connecting the parameters of local extinction and metapopulation dynamics. Oikos (1998) 83:390–396.[CrossRef][Web of Science]
Hanski I. Metapopulation Ecology. (1999) Oxford: Oxford University Press. 328. pp.
Hedger R., McKenzie E., Heath M., Wright B., Scott P., Gallego A., Andrews J. Analysis of the spatial distributions of mature cod (Gadus morhua) and haddock (Melanogrammus aeglefinus) abundance in the North Sea (1980–1999) using generalised additive models. Fisheries Research (2004) 70:17–25.[CrossRef][Web of Science]
Heessen H. J. L. Distribution and abundance of young cod and whiting in the south-eastern North Sea in the period 1980–1982. ICES Document CM 1983/G: 30 (1983) 4. pp.
Hilborn R., Quinn T. P., Schindler D. E., Rogers D. E. Biocomplexity and fisheries sustainability. Proceedings of the National Academy of Sciences USA (2003) 100:6564–6568.
Hutchings J. A. Spatial and temporal variation in the density of northern cod and a review of hypotheses for the stock's collapse. Canadian Journal of Fisheries and Aquatic Sciences (1996) 53:943–962.
Hutchinson W. F., Carvalho G. R., Rogers S. I. Marked genetic structuring in localised spawning populations of cod Gadus morhua in the North Sea and adjoining waters as revealed by microsatellites. Marine Ecology Progress Series (2001) 223:251–260.[CrossRef][Web of Science]
Hutchinson W. F., van Oosterhout C., Rogers S. I., Carvalho G. R. Temporal analysis of archived samples indicates marked genetic changes in declining North Sea cod (Gadus morhua). Proceedings of the Royal Society of London, Series B—Biological Sciences (2003) 270:2125–2132.
ICES. Development of a central database for European trawl survey data. DATRAS Database TRAwl Surveys. Final Report. In: Technical Report Project number QLRT-2001–00025 (2001) Copenhagen: ICES. 70.
ICES. Report of the Working Group on the Assessment of Northern Shelf Demersal Stocks (WGNSDS), 8–17 May 2007, Galway, Ireland. In: ICES Document CM 2007/ACFM: 22 (2007) A852.
ICES. Report of the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak (WGNSSK), 1–8 May 2007, ICES Headquarters, Copenhagen. In: ICES Document CM 2007/ACFM: 18 (2007) b701.
Kritzer J. P., Sale P. F. Metapopulation ecology in the sea: from Levins' model to marine ecology and fisheries science. Fish and Fisheries (2004) 5:131–140.[CrossRef][Web of Science]
McQuinn I. H. Metapopulations and the Atlantic herring. Reviews in Fish Biology and Fisheries (1997) 7:297–329.[CrossRef][Web of Science]
Neat F., Righton D. Warm water occupancy by North Sea cod. Proceedings of the Royal Society of London, Series B—Biological Sciences (2007) 274:789–798.
Neat F. C., Wright P. J., Zuur A. F., Gibb I. M., Gibb F. M., Tulett D., Righton D. A., et al. Residency and depth movements of a coastal group of Atlantic cod (Gadus morhua L.). Marine Biology (2006) 148:643–654.[CrossRef]
Perry A. L., Low P. J., Ellis J. R., Reynolds J. D. Climate change and distribution shifts in marine fishes. Science (2005) 308:1912–1915.
Riley J. D., Parnell W. G. The distribution of young cod. In: In The Propagation of Cod Gadus morhua L. pp. 563–580—Dahl E., Danielssen D. S., Moksness E., Solemdal P., eds. (1984) Flødevigen rapportser, 1. 895.
Righton D., Quayle V. A., Hetherington S., Burt G. Movements and distribution of cod (Gadus morhua) in the southern North Sea and English Channel: results from conventional and electronic tagging experiments. Journal of the Marine Biological Association of the UK (2007) 87:599–613.[CrossRef]
Robichaud D., Rose G. A. Migratory behaviour and range in the Atlantic cod: inference from a century of tagging. Fish and Fisheries (2004) 5:185–214.[CrossRef][Web of Science]
Rothschild B. J. Coherence of Atlantic cod stock dynamics in the northwest Atlantic Ocean. Transactions of the American Fisheries Society (2007) 136:858–874.[CrossRef]
Ruzzante D. E., Taggart C. T., Cook D. A review of the evidence for genetic structure of cod (Gadus morhua) populations in the Northwest Atlantic and population affinities of larval cod off Newfoundland and the Gulf of St Lawrence. Fisheries Research (1999) 43:79–97.[Medline]
Sinclair M., Iles T. D. Population richness of marine fish species. Aquatic Living Resources (1988) 1:71–83.[CrossRef]
Smedbol R. K., Stephenson R. L. The importance of managing within-species diversity in cod and herring fisheries of the north-western Atlantic. Journal of Fish Biology (2001) 59:109–128.[Web of Science]
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) 55:161–174.[CrossRef][Web of Science]
Stephenson R. L. Stock complexity in fisheries management: a perspective of emerging issues related to population sub-units. Fisheries Research (1999) 43:247–249.[CrossRef][Web of Science]
Thorrold S. R., Latkoczy C., Swart P. K., Jones C. M. Natal homing in a marine fish metapopulation. Science (2001) 291:297–299.
Wood S. N. Generalized Additive Models: An Introduction with R. (2006) FA: CRC/Chapman and Hall. 416. pp.
Wright P. J., Galley E., Gibb I. M., Neat F. C. Fidelity of adult cod to spawning grounds in Scottish waters. Fisheries Research (2006) a77:148–158.[CrossRef][Web of Science]
Wright P. J., Neat F. C., Gibb F. M., Gibb I. M. Evidence for metapopulation structuring in cod from the West of Scotland and North Sea. Journal of Fish Biology (2006) b69:181–199.[CrossRef][Web of Science]
Youngson A. F., MacLean J. C., Fryer R. J. Rod catch trends for early-running MSW salmon in Scottish rivers (1952–1997): divergence among stock components. ICES Journal of Marine Science (2002) 59:836–849.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Silva, D. W. Skagen, A. Uriarte, J. Masse, M. B. Santos, V. Marques, P. Carrera, P. Beillois, G. Pestana, C. Porteiro, et al. Geographic variability of sardine dynamics in the Iberian Biscay region ICES J. Mar. Sci., April 1, 2009; 66(3): 495 - 508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. F Hutchinson The dangers of ignoring stock complexity in fishery management: the case of the North Sea cod Biol Lett, December 23, 2008; 4(6): 693 - 695. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||