ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on April 25, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(4):707-713; doi:10.1093/icesjms/fsm038
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A meta-analysis of the status of ICES fish stocks during the past half century
ICES, HC Andersens Boulevard 44–46, 15533 Copenhagen V, Denmark
Correspondence to H. Sparholt: tel: +45 33 38 67 00; fax: +45 33 93 42 15;e-mail: henriks{at}ices.dk
Sparholt, H., Bertelsen, M., and Lassen, H. 2007. A meta-analysis of the status of ICES fish stocks during the past half century. – ICES Journal of Marine Science, 64: 707–713.Based on a meta-analysis of time-series of stock size, recruitment, and fishing mortality, the general status of fish stocks within the ICES Area (i.e. the Northeast Atlantic) is evaluated. The analysis is based on data for 34 (7 pelagic, 27 demersal) commercial stocks. The stocks were selected based on the quality of the data and the length of the time-series. The analysis indicates that most pelagic stocks recovered to sustainable levels with high productivity after several had collapsed in the 1960s and 1970s. In contrast, most demersal stocks have continued to decline over the past half century and are now recruitment-overfished. By reducing fishing mortality on demersal stocks on average by half and building up the stocks by a factor of about two, management could be brought in line with international agreements. If recruitment-overfishing is avoided for all demersal stocks and discarding is minimized, their yield might be almost doubled over the current yield. Among the major management initiatives during the past half century, only the closure of the pelagic fisheries in the mid-1970s can be clearly identified in the time-series as having had a direct effect on stock status.
Keywords: fish, ICES, meta-analysis, status, stocks
Received 30 June 2006; accepted 6 February 2007; advance access publication 25 April 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Overfishing was an issue in the Northeast Atlantic even before ICES was created in 1902 (Went, 1972). At that time, North Sea plaice (Pleuronectes platessa) were of special concern. The massive expansion of the steam-trawler fleet from its introduction in the late 1800s is blamed for the depletion of demersal fish stocks and, in turn, explains the halt in new investment observed around 1900 (Blegvad, 1943). The increase in effort resulted in a reduction in catch per unit effort (cpue) of plaice (and possibly other stocks as well) by a factor of five (Rijnsdorp and Millner, 1996). In the first half of the 20th century, fishing pressure continued to increase, the only interruptions being the two World Wars (Pope and Macer, 1996; Rijnsdorp and Millner, 1996). In the second half of the century, fishing mortality generally increased further, as documented in various ICES Working Group reports. Although the numbers of vessels and crew decreased substantially, technological improvements and the increasing size and power of the vessels more than compensated for the declines.
Through ICES assessment Working Groups, time-series of total landings (C), spawning-stock biomass (SSB), recruitment (R), and fishing mortality (F) for most of the commercially important fish stocks in the ICES Area (Northeast Atlantic, FAO Statlant Area 27) have become available routinely. Many of the time-series go back < 20 y, some < 50 y. Most data are based on VPA-type analysis (XSA: Shepherd, 1999; ICA: Patterson and Melvin, 1996). We provide a meta-analysis of available time-series to investigate whether the overall status of the demersal and pelagic fish stocks has changed over time, and whether major events such as the introduction of the Common Fisheries Policy (CFP; Holden, 1994) and the precautionary approach (PA) to fisheries management (FAO, 1995; ICES, 1999) marked a clear change in overall status. Our aim is also to document recruitment-overfishing and to define a sustainable fishing mortality that will give a high sustainable yield on a meta-scale.
We use a meta-analysis approach, because general trends are more easily seen when all data are combined rather than on a stock-by-stock basis (Myers, 2001). This is especially true for identifying recruitment-overfishing because, for a single stock, this depends largely on whether a clear stock–recruitment relationship has been established. However, identification of a stock–recruitment relationship for a given stock is difficult because of the confounding effects of the environment. This has been shown in simulations (Sparholt, 1996; Megrey et al., 2005), and Gilbert (1997) showed that a simple model with just two levels of R, assumed to be only environmentally driven, explained variations in R better than fitted Ricker curves for 136 marine fish stocks, based on a data set compiled by Myers et al. (1995).
Garcia and De Leiva Moreno (2005) analysed 14 stocks in the ICES Area with time-series data from 1970 to 2003 in relation to PA reference points. Our analysis should be considered, in part, as an extension of their analysis with regard to time range (1946–2003), number of stocks (38), indicators considered (recruitment and catches), and a grouping into pelagic (P) and demersal (D) stocks.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Time-series data (up to and including 2005) on C, SSB, R, and F were taken from ICES (2006). For each series, we rejected the estimates for the most recent 2 y, based on their low precision owing to the convergence feature of VPA-type models (Pope, 1972). We included only stocks for which ICES (2006) regards the assessments as at least indicative of stock trends (Table 1). The table also contains the limit reference points, Blim for SSB (ICES, 2003, 2006), the stock type (D or P), unit stock area, and the time span of each series.
|
The total annual catch of these 38 stocks has been around 4–6 million tonnes per year during the most recent two decades, compared with a total in the Northeast Atlantic of about 10 million tonnes (ICES FishStat Plus Database, www.ices.dk; Figure 1). The difference is largely accounted for by capelin, sandeel, sprat, redfish, horse mackerel, and Norway pout.
|
To scale the time-series to the same units, SSB and R of individual stocks were standardized in two ways: (i) by dividing the annual values by their long-term means; and (ii) by dividing SSB by the associated Blim values and R by the mean R for SSB > Blim. The first method is sensitive to differences in time-series length among stocks: when a 50-y standardized index is combined with a 25-y standardized index, there will inevitably be a jump in the average at the halfway point. Standardization using Blim does not suffer from this type of weakness, but Blim is itself a construct and not always easy to estimate. ICES (2006) provides current estimates of Blim for most stocks considered here, and ICES (2003) tried to estimate Blim for all stocks based on a segmented regression analysis. This method fits a "hockey-stick" model (a straight line through the origin followed by a straight horizontal line) to stock–recruitment data, if this model is significantly better than a straight horizontal line (equivalent to assuming no correlation). We applied the set of estimates used by ICES (2006), supplemented with estimates given in ICES (2003), if available. Blim values from non-significant segmented regressions were included in the standardization, because the estimates were close to the minimum SSB observed in the series.
To make the analysis as simple and transparent as possible, we used means over stocks by year and simple linear regression analysis to explore temporal trends. Significance levels and confidence intervals were not calculated because the time-series generally are strongly autocorrelated; resolving the statistics of this complication is beyond the scope of his work. However, a repetitive leave-one-stock-out sensitivity analysis (Mosteller and Tukey, 1977) was carried out. The time-series selected for this analysis was the SSB/Blim index for demersal stocks (1980–2003), because this series was considered (i) more reliable than the index using mean SSB; (ii) more important for management than the indices of R; and (iii) more sensitive to autocorrelation than the series for pelagic stocks, because of the relatively long life of demersal fish.
No attempt was made to standardize F because F is already a standardized metric, reflecting the chance that a fish will be caught. There might be an issue of different exploitation patterns by stock over time (some herring, Clupea harengus, stocks have suffered from a large fishery on juveniles during some periods, which might make the documented mean F somewhat incomparable across years). However, we judged that standardizing F further would create more problems of interpretation than it might solve.
The meta-analysis may help to identify the F that provides a high and sustainable yield. We approach this issue by investigating the combined development of F and C over time. Theoretically, if fishing pressure during the period considered changes gradually from underexploitation, through a sustainable level, to overexploitation, one would expect C to increase simultaneously with F at the start, followed by a levelling off and a decline some years later. The F corresponding to the time when C stops increasing should be close to the F giving maximum sustainable yield. It is important that the change in F is slow (a few per cent per year), because the stock in each year may then be regarded as being approximately in a steady state. To support this argument, ICES (2005) showed by simulations that even large and sudden changes in F of 50–70% from one year to the next should stabilize SSB and C within 3–12 y for North Sea plaice and cod (Gadus morhua), and two stocks of Baltic cod.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Total landings in the Northeast Atlantic have been rather stable at about 10 million tonnes per year since the late 1960s (Figure 1). Landings of demersal fish increased until the mid-1970s, then decreased by 40% thereafter. Landings of pelagic fish increased until the mid-1970s and have remained stable since. However, the pelagic stocks not included in the analysis (i.e. mainly small industrial fish used only for fishmeal and oil) have declined markedly, whereas the pelagic stocks we include (mainly the larger species used partly for human consumption) have increased markedly. Landings of others (crustaceans, cephalopods, tuna-like species, etc.) were comparatively small during the entire period analysed.
The temporal trends in the two standardized SSB and R series for demersal stocks are shown in Figure 2. SSB divided by the long-term mean (Figure 2a) declined significantly at a rate of 1.3% per year over the entire period. Truncating the time-series at cumulative 10-y intervals (series starting in 1960, 1970, or 1980) gave similar rates of decline (1.2–1.4% per year). However, if only data until 1980 were used, the trend was not significant. For SSB standardized by dividing by Blim (Figure 2b), the long-term trend also decreased, but at a faster rate (2.3%). When split into two periods (1946–1979 and 1980–2003), the annual rates of decline were 3.3 and 0.9% per year, respectively. The relatively low value for the period 1980–2003 should be viewed against the background of a large drop in relative SSB of 1.65 at the end of the period 1946–1979 to a value of 1.48 at the start of the period 1980–2003. In other words, a drop of 0.17, spread over the period 1980–2003, gives 0.7% per year, and this, added to the 0.9%, gives 1.6% per year. Standardized recruitment declined more slowly, at just 0.8% per year over the entire period (0.8 and 0.7% per year for division by the long-term mean, and by the mean for SSB > Blim, respectively). The sensitivity to truncations is also small (0.8–0.9% and 0.7–1.3%, respectively). Therefore, the two types of standardization yielded similar results except for SSB for the period 1946–1979, for which the decline in SSB/mean (SSB) is not significant, and SSB/Blim decreased substantially at 3.3% per year. This might indicate the bias regarding the use of long-term means for calculating a standardized index, because several time-series start only shortly before 1980. Therefore, we conclude that standardization using Blim best reflects the development in SSB. Because overall R decreased by a factor of 1/3 (0.72%) less than SSB (2.26%), demersal stocks appear to have been able largely (2/3) to compensate for the decrease in SSB by producing more recruitment per unit of SSB. However, the compensation from 1980 to 2003 was much smaller (1/4). Therefore, R appears to have been almost proportional to SSB since 1980, suggesting that the stocks were generally on the steep left-hand side of the stock–recruitment curve and therefore recruitment-overfished.
|
Figure 3 shows the corresponding time-series for the standardized SSB and R indices for pelagic stocks. In this case, the two types of standardization yield mutually consistent trends, but for SSB they are more complex than for demersal stocks (Figure 3a and b). SSB declined steeply until around 1980, reflecting the collapse of several pelagic stocks in the 1970s. The temporary moratoria implemented on North Sea herring and Norwegian spring-spawning herring and subsequent harvest control rules (HCRs) led to the recovery of these stocks, but the North Sea mackerel (Scomber scombrus) stock remained depleted (ICES, 2006). R (Figures 3c and 3d) declined until 1980 and increased thereafter, in parallel with the developments in SSB, but both the rate of decline (3%) and the subsequent rate of increase (1%) were less than for SSB (15 and 5%, respectively; percentages based on regression analysis of the Blim-standardized time-series 1950–1979 and 1980–2003). This suggests that pelagic stocks are now at or close to the plateau of the stock–recruitment relationship and no longer recruitment-overfished.
|
The mean F for demersal stocks increased steadily during the 50 y, from
0.4 to 0.6 (Figure 4a). The most recent years show signs of stabilization or possibly even a slight decline, but it is still too early to determine its significance. F for pelagic stocks followed a more complex pattern (Figure 4b), increasing steeply until the late 1960s and the early 1970s, to record high levels. During the moratoria, F decreased steeply again and has remained stable at 0.3 since the late 1970s.
|
The leave-one-stock-out sensitivity analysis for demersal stocks during the period 1980–2003 gave decline rates in SSB varying between 0.58% (if cod-2532 are excluded) and 1.51% (if Sai-ARCT are excluded) per year, compared with a value of 0.85% for all stocks included. Of the 28 values obtained in this way, one for each stock left out, 19 values were between 0.75 and 0.95%. Therefore, the results are reasonably robust to this sensitivity test, and it may be concluded that it is not just one stock that is driving the observed trends in our analysis.
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
One might expect that the introduction of the CFP in 1983 (Holden, 1994), if effective, would have manifested itself initially through a reduction in F, followed later by an increase in SSB and R. Judging from Figure 4a, there has been no such effect for demersal stocks, and F for pelagic stocks had already been reduced to the current moderate level before the CFP was implemented (Figure 4b). Moreover, the introduction of the PA towards the end of the 1990s (ICES, 1999) coincides with an increase in F for demersal stocks rather than with a decrease, and had no apparent effect on F for pelagic stocks. Therefore, two of the major management initiatives of the second half of the 20th century generally had no clearly positive effect on stock development. In contrast, the prediction from the trend line for demersal stocks (Figure 2b) is that, on average, SSB would drop below Blim in 2009 (SSB/Blim<1). Consequently, there is a high risk that recruitment of most demersal stocks will be critically impaired. The only management initiative that coincided with a substantial change in F was the introduction of moratoria on some depleted pelagic stocks in the 1970s, which clearly initiated stock recovery. Although it is always possible that the situation would have been even worse without implementation of the CFP and the PA, the objective of sustainable exploitation has certainly not been met for the demersal stocks in general.
Comparing the development in landings of demersal stocks included in the meta-analysis (Figure 1) with the development in F (Figure 4a), both showed an increasing trend until 1970. Thereafter, F continued to increase as landings decreased. Assuming approximate equilibrium conditions, this suggests that those stocks were underutilized at the beginning of the 1950s (F 
0.40), fully utilized during the late 1960s and mid-1970s (F = 0.45–0.50), and overexploited thereafter. Annual landings were 70% higher in the late 1960s and mid-1970s than now.
For pelagic stocks, the steep increase in F during the 1960s and the collapse of the stocks during the 1970s cannot be expected to represent equilibrium conditions, so no inference about a sustainable F and corresponding yield can be drawn from that period. However, F was clearly far too high. Post-1980, F has remained at moderate levels, stocks have recovered, and landings have increased to an unprecedented high level. It would be unlikely that sustainable landings can be much higher than now because they currently constitute as much as half the total catch of all fish and shellfish in the Northeast Atlantic. Therefore, the present value of 0.3 appears a good candidate for a sustainable F for pelagic stocks.
The total catch in the ICES Area increased steadily from 6 million tonnes in 1950 to 13 million tonnes in 1976, and has been fairly stable around 10 million tonnes annually since then (Figure 1). However, landings of demersal stocks decreased as landings of pelagic stocks increased, suggesting some trade-offs among species groups. Based on official landing statistics, Grainger and Garcia (1996) concluded that fish stocks in the Northeast Atlantic have been fully or overexploited since the early 1980s and that no further increase is possible. This view seems consistent with our results. Therefore, the challenge for fishery management might be seen as getting the most value out of the 10 million tonnes per year, while meeting the requirements for an ecosystem approach and for rebuilding depleted stocks, as stipulated in several international agreements. Rebuilding demersal stocks to 1970s levels might be a sensible target, because these stocks generally are more valuable per unit weight than small industrial species of pelagic fish.
Considering 14 demersal stocks routinely assessed by ICES during the period 1970–2003, Garcia and De Leiva Moreno (2005) concluded that "...the status of the stocks in relation to the precautionary reference points has been degrading in terms of spawning biomass and excess fishing mortality...", but that "...the situation seems to have improved during the past decade ... ." This optimistic statement is based to a large extent on the data for 2001–2003. Although our analysis also reveals signs of a slight improvement in SSB and of a reduction in F in recent years, they are still within the range of variations observed around the general trend. In our opinion, therefore, it is premature to interpret this "slight improvement" as a break in the trend.
The advantage of a meta-analysis is that the general patterns become clearer than when looking at developments within individual stocks. Conclusions may be drawn about, for instance, a generalized, empirical stock–recruitment relationship that encompasses all the noisy and debatable relationships observed within single stocks. Moreover, the aggregated historical perspective provides the empirical outcome of what may be considered a series of "decadal fishing-intensity experiments". Of course, there will be some interdependence, but because annual changes are generally quite small (with the notable exception of the depletion of some pelagic stocks in the 1970s), the stock complex in each decade may be expected to be close to a steady state situation. Because these "experiments" have been carried out in the real world, the results by default reflect variations in all important ecosystem drivers such as carrying capacity, species interaction, climate and environmental effects, and disease. Therefore, the meta-analysis may help to synthesize the results of individual stock assessments into sound overall management strategies or to check whether the management strategies for individual stocks are mutually consistent in a broader context. On the other hand, a meta-analysis hides important stock-specific features, such as the potential for some stocks to tolerate greater levels of exploitation than others. Also, some demersal stocks are doing reasonably well at the moment (e.g. sole in the Skagerrak and Kattegat, and saithe in the northeast Arctic; ICES, 2006); others, of course, are severely depleted and urgently require conservation measures. Therefore, single-stock assessments remain a high priority.
Many stocks of small fish, fished mainly for meal and oil (sandeel, Ammodytes spp., and Norway pout, Trisopterus esmarkii, in the North Sea; capelin, Mallotus villosus, in the Barents Sea and around Iceland and Greenland waters), which historically comprised a major part of the total Northeast Atlantic fish catch, are currently depressed, allegedly because of ecosystem drivers rather than overexploitation (ICES, 2006). This meta-analysis offers no further insight on that issue. However, their current poor condition may be an expression of multispecies interactions, which by default are included implicitly in the analysis. The mechanism could be that large pelagic stocks have a negative impact on industrial species by predation on their offspring or by food competition. For instance, Norwegian spring-spawning herring play a major role as a predator on capelin larvae in the Barents Sea (ICES, 2006).
Perhaps too, the currently healthy pelagic stocks are exerting pressure on demersal stocks, preventing them from returning to their earlier productivity. Such a hypothesis would have to be rejected at the meta-scale level, because the average size of pelagic stocks is no larger now than in the late 1960s and mid-1970s. Also, the reverse hypothesis that the reduction in predation by demersal fish is the main reason for the healthy pelagic stocks has to be rejected because recruitment of pelagic fish did not improve during the period 1960–2003 (Figure 3d), and recruitment is the parameter mainly affected by prey-size selection of demersal predators (Sparholt, 1990). This is somewhat surprising because, on a finer scale, cod and sprat (Sprattus sprattus) stocks in the Baltic Sea exert these types of mutual influence on each other (Köster et al., 2003). In more complex ecosystems, such direct interactions might be counteracted by secondary or tertiary effects. In the North Sea, numbers of grey gurnard (Eutrigla gurnardus) increased, possibly in response to released predation from the reduced gadoid stocks, and they prey heavily on zero-group cod (Floeter et al., 2005).
In the North Sea, the late 1960s and early 1970s were characterized by a sudden and still largely unexplained increase in the abundance of cod, haddock (Melanogrammus aeglefinus), whiting (Merlangius merlangus), and saithe (Pollachius virens) through the production of a series of strong year classes, the so-called gadoid outburst (ICES, 2006). Against this background, it might be argued that the catch taken from demersal stocks in the entire Northeast Atlantic during the same period represents an anomalous situation. However, the mean annual catch of these four species in the North Sea in the period 1970–1975 was only some 300 000 t larger than in the period 1980–1985. Therefore, the gadoid outburst apparently had only a minor effect on the total annual catch of all demersal species, 5 million t in these years, and cannot be regarded as invalidating our results.
Misreporting and non-reporting of catches has been a continuous and major concern since the CFP was introduced in 1983. The total allowable catch and national quota system has been an incentive to stretch quotas by putting more fish in the boxes, by landing fish on black or grey markets, and by misreporting the position of capture. In some cases, situations have been documented where fishers collectively fished much more than their quota (ICES, 2006). Although the estimated unreported landings for demersal stocks entering the assessments have never exceeded 300 000 t (10% of the official catches of around 3 million tonnes), the actual quantities may have been substantially larger. Moreover, the meta-analysis may be invalidated by the inherent bias in SSB and F estimates derived from stock assessments.
In addition to the stocks of small industrial species, several commercially important stocks could not be included in the analysis because of various data-quality problems. Among pelagic stocks, an outstanding 1982 year class boosted the stock size of western horse mackerel (Trachurus trachurus), which gave rise to landings of several hundred thousand tonnes per year in the 1980s. However, the assessment is not regarded as being even indicative of stock trends because of bias in age reading (ICES, 2006). The demersal stocks not integrated include a large number of species that are generally taken as a bycatch in fisheries targeting the species listed in Table 1. Generally, the assessments of such stocks are uncertain owing to limited information on basic features such as stock boundaries and a shortage of samples for biological analysis (e.g. age). The landings of such fish have also been on a downward trajectory since the mid-1970s, although not as obviously strong as those of the demersal stocks included in the analysis, and their decline started later. Because such stocks are less targeted, fishing pressure is likely to have been lower, consistent with the delay in peak landings and the less severe decline.
The results of the sensitivity analysis for demersal species and the similarity between the results of the two alternative standardization procedures used lends support to our belief that our conclusions are robust to variations in the data set selected and the methods applied. The meta-analysis indicates that demersal stocks have declined continuously for the past half century and that they are currently recruitment-overfished. In contrast, pelagic stocks have largely recovered to sustainable levels after a period of some marked collapses, and they have not been recruitment-overfished in recent years. A general feature for both groups is that rates of change were higher for SSB than for R. This is consistent with the classic stock–recruitment models proposed by Ricker (1954) and Beverton and Holt (1957). For the demersal stocks, the rate of decrease in R was just a little less than the decrease in SSB since 1980, clearly supporting these models, because that result implies that the relationship approximates a proportional one close to the origin.
In recent years, harvest control rules (HCRs) have been evaluated for many demersal stocks (ICES, 2005, 2006). They generally indicate a lower target F (mostly between 0.3 and 0.4) than suggested by our analysis, which by its meta-nature does not specifically consider the risk of depleting individual stocks. In fact, the empirical F that should lead to sustainable yields for demersal stocks in general refers to mean values over many stocks, some of which may have been either under- or overexploited, even in the period of the analysis. Therefore, we conclude that proper management of individual stocks would have raised the overall yield even higher than that observed during this period. Moreover, if discarding of juvenile and marketable fish can be avoided, a further increase in the yield would be expected. However, if depletion of individual stocks is to be avoided with a high probability and at the cost of high yields, then a lower F should be sought. The range 0.3–0.4 suggested by HCR evaluations might then be more appropriate.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Beverton R. J. H., Holt S. J. On the Dynamics of Exploited Fish Populations. Fishery Investigations, Series II (1957) 19:533.
Blegvad H. Fiskeriet I Danmark. (1943) Danish: Selskabet for udgivelse af Kulturskrifter.
FAO. Code of Conduct for Responsible Fishing. (1995) http://www.fao.org/documents/show_cdr.asp?url_file=/DOCREP/005/v9878e/v9878e00.htm.
Floeter J., Kempf A., Vinther M., Schrum C., Temming A. Grey gurnard (Eutrigla gurnardus) in the North Sea: an emerging key predator? Canadian Journal of Fisheries and Aquatic Sciences (2005) 62:1853–1864.
Garcia S. M., De Leiva Moreno J. I. Evolution of the state of fish stocks in the Northeast Atlantic within a precautionary framework, 1970–2003: a synoptic evaluation. ICES Journal of Marine Science (2005) 62:1603–1608.
Gilbert D. J. Towards a new recruitment paradigm for fish stocks. Canadian Journal of Fisheries and Aquatic Sciences (1997) 54:969–977.
Grainger R. J. R., Garcia S. M. Chronicles of marine fishery landings (1950–1994). FAO Fisheries Technical Paper (1996) 359.
Holden M. J. The Common Fisheries Policy: Origin, Evaluation and Future. (1994) Oxford: Fishing News Books.
ICES. Report of the ICES Advisory Committee on Fisheries Management, 1998. (1999) 229. ICES Cooperative Research Report.
ICES. Report of the Study Group on Precautionary Reference Points for Advice on Fishery Management. In: ICES Document CM 2003/ACFM (2003) 15.
ICES. Report of the ICES Advisory Committee on Fishery Management, Advisory Committee on the Marine Environment and Advisory Committee on Ecosystems, 2005. In: ICES Advice (2005) 1–11.
ICES. Report of the ICES Advisory Committee on Fishery Management, Advisory Committee on the Marine Environment and Advisory Committee on Ecosystems, 2006. In: ICES Advice (2006) 1–10.
Köster F. W., Hinrichsen H. H., Schnack D., St John M. A., MacKenzie B. R., Tomkiewicz J., Möllmann C., et al. Recruitment of Baltic cod and sprat stocks: identification of critical life stages and incorporation of environmental variability into stock-recruitment relationships. Scientia Marina (2003) 67:129–154.
Megrey B. A., Lee Y-W., Macklin S. A. Comparative analysis of statistical tools to identify recruitment–environment relationships and forecast recruitment strength. ICES Journal of Marine Science (2005) 62:1256–1269.
Mosteller F., Tukey J. W. Data Analysis and Regression. (1977) Reading, MA: Addison-Wesley. 558.
Myers R. A. Stock and recruitment: generalizations about maximum reproductive rate, density dependence, and variability using meta-analytic approaches. ICES Journal of Marine Science (2001) 58:937–951.
Myers R. A., Bridson J., Borrowman N. J. Summary of worldwide spawner and recruitment data. Canadian Technical Reports of Fisheries and Aquatic Sciences (1995) 2024.
Patterson K. R., Melvin G. D. Integrated Catch-at-Age Analysis Version 1.2. In: Scottish Fisheries Research Reports (1996) 56.
Pope J. G. An investigation of the accuracy of virtual population analysis using cohort analysis. ICNAF Research Bulletin (1972) 9:65–74.
Pope J. G., Macer C. T. An evaluation of the stock structure of North Sea cod, haddock, and whiting since 1920, together with a consideration of the impacts of fisheries and predation effects on their biomass and recruitment. ICES Journal of Marine Science (1996) 53:1157–1169.
Ricker W. E. Stock and recruitment. Journal of the Fisheries Research Board of Canada (1954) 11:559–623.
Rijnsdorp A. D., Millner R. S. Trends in population dynamics and exploitation of North Sea plaice (Pleuronectes platessa L.) since the late 1800s. ICES Journal of Marine Science (1996) 53:1170–1184.
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) 56:584–591.
Sparholt H. Improved estimates of the natural mortality rates of nine commercially important fish species included in the North Sea multispecies VPA model. Journal du Conseil International pour l'Exploration de la Mer (1990) 46:211–223.
Sparholt H. Causal correlation between recruitment and spawning stock size of central Baltic cod? ICES Journal of Marine Science (1996) 53:771–779.
Went A. E. J. Seventy years agrowing. A history of the International Council for the Exploration of the Sea 1902–1972. Rapports et Procès-verbaux des Réunions du Conseil International pour l'Exploration de la Mer (1972) 165.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M.-J. Rochet and J. C. Rice Simulation-based management strategy evaluation: ignorance disguised as mathematics? ICES J. Mar. Sci., May 1, 2009; 66(4): 754 - 762. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||