Skip Navigation


ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on December 18, 2006
ICES Journal of Marine Science: Journal du Conseil 2007 64(2):369-376; doi:10.1093/icesjms/fsl033
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
64/2/369    most recent
fsl033v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Boulcott, P.
Right arrow Articles by Gibb, I. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Boulcott, P.
Right arrow Articles by Gibb, I. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2006 International Council for the Exploration of the Sea. Published by Oxford Journals. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Regional variation in maturation of sandeels in the North Sea

Philip Boulcott1,, Peter J. Wright1, Fiona M. Gibb1, Henrik Jensen2 and Iain M. Gibb1

1 Fisheries Research Services, Marine Laboratory, PO Box 101, Victoria Road, Aberdeen AB11 9DB, UK
2 Danish Institute for Fisheries Research, Department of Marine Fisheries, Charlottenlund Castle, DK-2920 Charlottenlund, Denmark

Correspondence to P. Boulcott: tel: +44 131 6513602; fax: +44 131 6506564; e-mail: p.boulcott{at}ed.ac.uk

Boulcott, P., Wright, P. J., Gibb, F. M., Jensen, H., and Gibb, I. M. 2007. Regional variation in maturation of sandeels in the North Sea. – ICES Journal of Marine Science, 64: 369–376.

The current assessment of lesser sandeel (Ammodytes marinus) in the North Sea assumes a single stock and a knife-edge maturity ogive. However, there is evidence that the North Sea stock consists of several reproductively isolated components, raising the possibility of demographic differences among regional aggregations. We examine regional variation in size- and age-at-maturity in four components of the North Sea stock. Surveys in 1999 indicated pronounced regional differences in length- and weight-at-age, implying a disparity in growth rate across the North Sea. Logistic regression revealed that the onset of maturity was significantly related to regional distribution, in addition to length and age, with a tendency for fish off the UK's northeast coast to mature later and smaller than elsewhere. No significant effect of year on either growth or length-at-maturity was revealed from a comparison with other data collected in 2004. The results show that important regional phenotypic variation not currently represented in stock assessments could have implications for the local sustainability of sandeel aggregations.

Keywords: Ammodytes marinus, fecundity, maturation, reproduction, sandeel, stock management

Received 6 June 2006; accepted 17 November 2006; advance access publication 18 December 2006.


   Introduction
Top
 Introduction
Methods
 Results
 Discussion
 References
 
Many stocks of marine fish consist of several spawning components with a variable degree of reproductive segregation (Sinclair, 1988; Stephenson, 1999). Differences in life history among these subpopulations may reflect local adaptation as well as a phenotypic response to spatial variation in environmental conditions (Olsen et al., 2004; Conover et al., 2005). Understanding such population variation is important in ensuring that fish harvests do not lead to localized depletion.

Sandeels are an abundant and important component of North Atlantic foodwebs (Sherman et al., 1981; Daan, 1990), and until 2003, they supported the largest fishery in the North Sea (primarily Ammodytes marinus), with annual landings exceeding a million tonnes in some years (ICES, 2005). Within the North Sea, there are several distinct aggregations of A. marinus, the distribution of which is largely defined by the availability of suitable sediment in which to reside following settlement (Wright et al., 1998, 2000), the limited scale of movement of settled sandeels (Kunzlik et al., 1986; Popp-Madsen, 1994), and the low rates of larval exchange among habitat patches (Proctor et al., 1998; Munk et al., 2002). In the light of these findings, the regional sustainability of these aggregations has become an important issue in managing North Sea sandeels (Monaghan, 1992; Wright, 1996). Moreover, concern over the local impact of fisheries has contributed to precautionary closure of the fisheries off the northeast coast of the UK (ICES, 1999a).

Sandeels are planktivorous throughout their life, calanoid copepods forming the main constituent of their diet (Macer, 1966). Given the relative spatial isolation of sandeel aggregations in the North Sea, it might be reasonable to expect that the great variability in abundance of zooplankton, on both temporal and spatial scales (Fransz et al., 1991), could lead to spatial differences in growth rate. Indeed, significant spatial differences in the growth of sandeels have been described, although the studies upon which such findings are based tend to have relied on data collected in different years or restricted to specific areas of the North Sea (Macer, 1966; Warburton, 1982; Wright, 1996; Bergstad et al., 2001, 2002). Differences in size-at-age north and south of 56°30'N reported in 1978 were, until 1995, used as a basis for dividing the North Sea into two regions for the purpose of sandeel assessment (ICES, 2005). However, temporal variation in size-at-age at some locations suggests that size differences may not persist (Wright and Bailey, 1996; Bergstad et al., 2002). Nevertheless, recent research suggests that although a notional north/south divide in the assessment of the stock appears to have little relevance (Wright et al., 1998; ICES, 1999a; Munk et al., 2002), there are differences in population dynamics between reproductively isolated aggregations in eastern and western areas of the central North Sea (Pedersen et al., 1999). A recent analysis of a 25-y time-series of sandeel size-at-age off the Firth of Forth, Scotland also indicates that sandeels there have always been slow-growing, despite a long-term decline in growth rates (Wanless et al., 2004).

Growth in sandeels is very seasonal, with the strongest period of growth between March and July (Bergstad et al., 2002) after emergence from the sandy substrata in which they over-winter (Macer, 1966; Wright et al., 2000). When they are active, sandeels tend to emerge only during daylight to feed (Winslade, 1974a), restricting foraging to nearby waters. Maturation of A. marinus is during July, with a transition to exogenous vitellogenesis (females) or spermatocyte production (males) in August or September (PB and PJW, pers. comm.). Experimental evidence provided by Boulcott and Wright (submitted) from a population of sandeels sampled off the east coast of Scotland suggests that maturation of fish aged 1 y is dependent on growth before a critical period of maturation in June/July of their second growing season, with few fish <10.7 cm becoming mature. Maturation in the second growing season appears typical for many parts of the North Sea, because nearly all sandeels are mature by January, when they are 2 y old (Macer, 1966; Gauld and Hutcheon, 1990).

Assessment of the North Sea sandeel stock currently uses a knife-edge maturity relationship, all fish >2 y old being assumed to be mature on 1 January. However, given that maturity is related to size (PB. and PJW, pers. comm.) and that growth differences have been reported among some of the reproductively isolated populations, there is potential for regional variability in the proportion of 2-y-old sandeels that mature, unless populations of slower-growing fish mature at a smaller size. As sandeel stocks are dominated by just a few age classes (Pedersen et al., 1999), even small changes in the proportion of fish maturing before or after the assumed maturation threshold could affect spawning-stock biomass. Slower growth rates for a given mortality should select for smaller size-at-maturity, as seen from an interspecies comparison of flatfish life history (Roff, 1983), but this may not completely compensate for potentially confounding variation in the energy accrual abilities of fish within populations. Hence, relationships between maturity and size of sandeels may vary regionally in relation to long-term differences in growth rate not accounted for in the knife-edge maturity relationship used for the assessment. Although all sandeels >13.5 cm collected from northern Scottish waters are mature, published accounts provide some support for the notion of a regional difference in maturity (Gauld and Hutcheon, 1990) with, for example, just 50% of fish 14 cm long maturing on the Klondyke Bank (Bergstad et al., 2001). Nevertheless, because earlier comparative studies were conducted across different years, it is possible that the reported differences reflect temporal rather than spatial variation in maturity–size relationships. If true, however, such regional variation will serve further to influence reproductive traits through the positive influence of growth on age-specific fecundity (Shine and Schwarzkopf, 1992) and egg size (Kjesbu et al., 1991, Marteinsdottir and Steinarsson, 1998).

Here, we characterize maturity-at-size and -age relationships of sandeels in four of the main fishing regions for the species in the North Sea. Sampling is undertaken in two time periods, and our aim is to test whether there are significant regional differences in maturity–size relationships and whether any spatial differences persisted over time. The implications of the results to the current maturity schedule used in the assessment are also examined.


   Methods
Top
 Introduction
 Methods
Results
 Discussion
 References
 
Dredge samples of sandeels were collected during surveys conducted by the Danish Institute for Fisheries Research and the FRS Marine Laboratory during 1999 from four key sandeel fishing grounds in the North Sea: NW Rough, Elbow Spit, Fisher Bank, and off the Firth of Forth (Figure 1). The surveys took place between late October and early December, after the summer growing season, at a time when sandeels return to bury themselves in the substratum, before re-emerging to spawn in December and January (Gauld and Hutcheon, 1990). Although little or no change in the length of sandeels has been noted during the sampling period (PB and PJW, pers. comm.), to ensure that any regional length differences were not an artefact of our sampling regime, all fish used in the analysis for 1999 were collected within a 30-d period. To provide data with which to examine temporal changes in maturity, a second, smaller-scale survey was undertaken in 2004. As regional variation in maturation in the 1999 survey was predominantly in age classes 0–2, the 2004 survey avoided collecting sandeels belonging to the upper limits of the species' length range, with a representative sample of 20 fish for each 0.5 cm length interval across the 7.5–17 cm length range targeted in each of the four regions. Regrettably, poor recruitment of sandeels across the North Sea in 2004 made it difficult to obtain the desired stratified sample. The collection in 2004 only gathered a sufficient sample of sandeels aged 1 y and was unable to obtain any data for the Fisher Bank. This situation was also mirrored in an earlier survey attempt in 2003, during which few sandeels aged 1 y were caught in any of the four target regions. Although sandeels aged 1 y dominated our sample collection, this age group constitutes some 78% of sandeel numbers when the 0-group is excluded from stock estimates (ICES, 2005). Once collected, all samples of A. marinus were frozen in the field for later analysis in the laboratory.


Figure 1
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Distribution of Danish (DIFRES) total landings of sandeels between 1996 and 2005. The four sample regions used in the study are shown: (1) NW Rough, (2) Elbow Spit, (3) Firth of Forth, and (4) Fisher Bank.

 
Age, total length (±1 mm), wet weight (±0.0001 g), and eviscerated wet weight (±0.0001 g) were recorded for each fish collected during the 2004 survey. Collection of weight data was not, however, possible during the 1999 survey. Where discernible, sex was recorded by direct observation of the gonads. For immature juvenile fish for which sex was indeterminate, a double entry was added to the data set used to analyse maturation, each fish being recorded once as female and then again as male. Maturity was classified macroscopically in females according to the standard ICES four-stage scale: virgin, maturing, spawning, and spent (ICES, 1999b). All mature fish of both sexes were in late maturation at the time of sampling.

The age of all fish was determined by examination of their sagittal otoliths. Counts of annual increments were made under x60 magnification with the sagittae immersed in 80% ethanol against a black background illuminated by reflected light. Interpretation followed the protocol of ICES (1995), 1 January being taken as the common birthdate. All estimates were made by an experienced reader, and a subset of annual increment counts was verified using counts of daily increments (Wright, 1993). As they represented just a small proportion of samples in each region, sandeels 4 + years old at capture were excluded from the analysis.

Statistical analyses
Analysis of variance (ANOVA) was carried out on length distributions to determine the overall variance between regions and, where applicable, year of survey. Fisher's Protected Least Squares Difference (PLSD) post hoc multiple comparisons were used to identify specific differences between regions. For fish for which sufficient information on sex was present, the size structures of sandeels sampled within each region were contrasted using a Kolmogorov–Smirnov two-sample test to test for sex-specific differences.

When assessing maturity, the data sets compiled from the two surveys were analysed using a logistic regression model (McCullagh and Nelder, 1983), which included length, age, sex, region, and year as explanatory variables. Changes in sandeel length are negligible during our selected sampling period (PB and PJW, pers. comm.), thus permitting reasonable comparison between samples collected on dates 30 d apart. Maturity in our model was expressed as binary (immature or mature), with region, age, and sex expressed as factorial variables. The proportions maturing in each instance were modelled as a response function using a logit link function, where


Formula 033UM1

The initial model used for the logistic regression included all explanatory variables and their interactions. Minimum adequate models could then be derived by stepwise deletion of all non-significant terms. We also calculated an additional measure of model fit, based on a pseudo-coefficient of determination (Swartzman et al., 1995), which was taken to be the fraction of the total variation explained by the model: r2=1–(residual variance/null variance).

The unavoidable omission of Fisher Bank from the 2004 data set prevented adoption of a fully balanced design that included all regions and ages across the two periods. Instead, the entire 1999 data set was first analysed in isolation, with length, age, sex, and region included as explanatory variables. Minimum adequate models were again derived by stepwise deletion of all non-significant terms. In all instances, models were re-built and tested with explanatory variables introduced in a different sequence, to ensure that order effects were not apparent. The only age class for which sufficient data existed across both 1999 and 2004 surveys belonged to the age 1 group from NW Rough, Elbow Spit, and Firth of Forth. This data set was subsequently analysed using a linear regression model, in which length, sex, region, and year of survey were used as explanatory variables.


   Results
Top
 Introduction
 Methods
 Results
Discussion
 References
 
In all, 4291 sandeels were collected over the two surveys for which length, age, and maturity data were available (Table 1). Examination of the contents of mature ovaries from sandeels caught in both 1999 and 2004 surveys showed them to contain a single batch of late vitellogenic oocytes. Mature male gonads were all in final stages of maturation. No fish sampled displayed any evidence of having spawned during the sample period.


View this table:
[in this window]
[in a new window]

  		Table 1. Number of sandeels sampled during two surveys (1999 and 2004) for each region for which age and maturity data were available.

 
Within the age range of 0–3 y sampled during the two surveys, total length ranged from 5.4 to 19.1 cm. Mean length- and weight-at-age of fish split by region are given in Table 2 for the 1999 and 2004 surveys. A two-tailed t-test revealed a significant difference between the mean lengths of males and females, 12.3 vs. 12.6 cm, respectively (d.f.=2588, t = –3.96, p < 0.001), with a slight but statistically significant difference in sex ratio (female/male proportion = 0.55) also discernible (goodness-of-fit test: d.f. = 1, {chi}2 = 32.2, p < 0.001). Comparison of the size distributions collected from the 1999 and 2004 data grouped according to region revealed a significant difference in male and female total length in the Firth of Forth (Table 3). No such difference was apparent in any of the other three regions tested. Consequently, male and female data were not pooled, with sex subsequently included as an initial explanatory variable during the process of fitting regression models to maturity ogives.


View this table:
[in this window]
[in a new window]

  		Table 2. Mean length (cm) and weight (g), the latter in parenthesis, of sandeels sampled by age class and region during the 1999 and 2004 surveys.

 


View this table:
[in this window]
[in a new window]

  		Table 3. A comparison (Kolmogorov–Smirnov two-sample test) of male and female sandeel size distributions by region.

 
Length distributions split according to age, sex, and region were unimodal, consistent with the relatively discrete spawning period associated with the species. A one-way ANOVA applied to each of the age groups tested in 1999 revealed significant differences in mean total length and mean weight for each of the four regions tested (Table 2). With regard to age 0 fish, this disparity in size was at its most pronounced when comparing fish from Fisher Bank and the Firth of Forth. The mean total length of samples collected from Fisher Bank was 2.6 cm greater and their mean weight 2.9 g heavier. The diminutive size of fish sampled from the Firth of Forth was also apparent in the three other age groups tested, although the relative ranking of the other three regions in terms of size altered in these age groups. A similar pattern of significant differences in mean total length and weight was recorded across the three regions sampled during 2004. Multiple post hoc analyses of these tests carried out on both sample years revealed that no single region was responsible for the significant effect of region on size. A two-way ANOVA applied to the combined 1999 and 2004 data set revealed a significant effect of region (two-way ANOVA: d.f. = 2, F = 555.1, p < 0.001) and its interaction with year (d.f. = 2, F = 39.2, p < 0.001), but was unable to detect a significant effect of year on mean length-at-age for the 2 years sampled (d.f. = 1, F = 0.003, p = 0.95).

Maturity-at-length and -age
Maturity ogives fitted by logistic regression for each of the four regions sampled during the 1999 survey are given in Figure 2, with overlying plots depicting the proportions maturing in each region across 0.5 cm intervals. Only one fish <9 cm matured in any region tested, and all fish >17 cm were mature. The proportion of fish that matured varied according to region and age during the 1999 survey (Table 4). With the notable exception of Fisher Bank, the percentage of age 0 fish to mature in each region amounted to <5%. Of the 579 age 0 fish sampled on Fisher Bank, 22% were mature. These fish would, therefore, be expected to spawn at age 1.


Figure 2
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Maturity ogives derived from the 1999 survey for each region. Plots derived from field data for the relationship between length (cm) and % maturity are given for each area: NW Rough (open triangles), Elbow Spit (filled squares), Firth of Forth (crosses), and Fisher Bank (filled circles). Curves depict the fitted maturity ogives for each region and were produced by logistical regression: NW Rough (solid line), Elbow Spit (dashed—dotted line), Firth of Forth (dashed line), and Fisher Bank (open dotted line).

 


View this table:
[in this window]
[in a new window]

  		Table 4. Proportion of sandeels mature according to age class and region.

 
As expected, total length (logistic regression: t = 29.82, p < 0.001) proved to be the most important explanatory variable, accounting for 49% of the variation in maturity when a logistic regression that contained all explanatory variables was fitted to the data collected during the 1999 survey. Both age (t = 24.89, p < 0.001) and region (t = –6.14, p < 0.001) influenced the maturity of A. marinus, accounting for an additional 6 and 2% of the variation, respectively. Sex, although statistically significant (t = 2.06, p = 0.04), explained just 0.001% of the variance in observed maturity values and was, therefore, omitted from the minimum adequate model. The maximum likelihood estimates of all parameters in the minimum adequate model are given in Table 5.


View this table:
[in this window]
[in a new window]

  		Table 5. Summary statistics from the logistic regression ln p/(1–p) = I+L · l + Ai + Rj, where I is the intercept, L the regression coefficient, l the length in cm, Ai the age effect for age class i, and Rj the regional effect for region j.

 
To assess the effect of year on maturation, data collected from sandeels aged 1 y from NW Rough, Elbow Spit, and the Firth of Forth were compared across the 1999 and 2004 surveys. The restriction of this analysis to age 1 fish was because of sampling limitations. Applying a logistic regression model to this combined data set revealed that length was the sole explanatory variable (t = 20.02, p < 0.001), accounting for 35% of the variation in maturity. Year (t = –1.09, p = 0.27), sex (t = 1.40, p = 0.16), and region (t = –0.65, p = 0.53) did not, however, influence maturation.

Given the omission of Fisher Bank from the combined data set, data from the 1999 survey were re-tested for all four regions using only sandeels aged 1 y. Again, length was the main explanatory variable in our logistical regression (t = 18.1, p < 0.001), explaining 18% of the variation in maturity; compare this result with region (t = –12.17, p < 0.001), which accounted for 8% of the variation. No effect of sex was apparent (t = 1.90, p = 0.06). When Fisher Bank was removed from the age 1 data set, total length became the sole explanatory variable in the model (t = 17.82, p < 0.001, r2 = 0.37), as it had been in the combined data set, with no discernible effect of region (t = –0.27, p = 0.79) or sex (t = 1.68, p = 0.09). However, analysis of the three regions in the 1999 data set across all three age classes revealed that length (t = 24.0, p < 0.001, r2 = 0.62), region (t = 3.54, p < 0.001, r2 = 0.02), and age (t = 11.6, p < 0.001, r2 = 0.02) remained significant explanatory variables in the minimal model.

Examining the age 1 sandeels from the 2004 data set, the sample for which a complete entry of weight measurements was available, length (t = 4.35, p < 0.001, r2 = 0.01), weight (t = –2.73, p < 0.01, r2 = 0.60), and condition (t = 3.75, p < 0.001, r2 = 0.03) were significant factors in the model. Nevertheless, length (t = 7.83, p < 0.001), when entered into the model alone, explained 60.5% of the variation in maturity.


   Discussion
Top
 Introduction
 Methods
 Results
 Discussion
References
 
The existence of regional differences in growth rate of A. marinus is evident from the range in sizes-at-age reported for different areas of the North Sea (Macer, 1966; Wright, 1996; Bergstad et al., 2002; Wanless et al., 2004). However, where growth patterns from several regions of the North Sea were compared, such comparison was invariably across different years, so it is possible that reported differences may reflect temporal rather than spatial variation in growth. Nevertheless, in a recent study comparing sandeel growth between an unexploited site and a commercially fished ground, Bergstad et al. (2002) detected differences in size-at-age of fish sampled within the same year. The results presented here provide clear evidence of marked regional differences in size-at-age across the commercially fished range of North Sea sandeels.

The North Sea habitat offers considerable spatial variation in the environmental conditions that can influence growth. At the most basic level, access to food resources will influence growth and, as such, environmental and seasonal cycles that influence zooplankton production throughout the North Sea might be expected to contribute to the observed differences in size-at-age between regions (Macer, 1966). That the seasonal distribution of copepods, the main component of sandeel diet in the North Sea (Ryland, 1964; Macer, 1966), varies over the North Sea supports this position. For example, copepod biomass peaks later and for a shorter period at the Firth of Forth aggregation than in the other areas examined here (Fransz et al., 1991). Temperature influences the emergence of sandeels from the substratum, and this will also affect feeding activity (Winslade, 1974a, b), as well as the scope for growth. However, although banks south and east of the Scottish east coast tend to have warmer sea surface temperatures in summer (FRS, 2004), little is known of the actual temperature that sandeels experience in the water column. Moreover, growth can rarely be attributed to one variable, and biotic factors, including density-dependent effects (Nagoshi and Sano, 1979) and reproductive investment (Roff, 1992), are liable also to be important in controlling growth of sandeels.

Our model revealed a distinct regional effect on maturity, with fish from faster-growing regions maturing earlier and larger, e.g. fish sampled from Fisher Bank matured earlier but larger. As is generally the case (Lambert et al., 2004), total length was the dominant factor in determining the maturity of North Sea sandeels, accounting for 49% of the variation in maturation rates in some instances. It is possible that the addition of body condition as a factor may improve the predictive ability of the model because, in fish species such as cod (Gadus morhua), indices of body condition have an important effect on spawning potential (Marteinsdottir and Begg, 2002; Yoneda and Wright, 2004). However, the lack of complete weight records from the 1999 survey precluded the inclusion of weight as an explanatory factor, whereas evidence from 2004 survey data suggests that weight recorded at the time of spawning does not increase model fit appreciably for sandeels. This could, in part, be because they spawn some 5–6 months after committing energetic resources to reproduction (PB and PJW, pers. comm.), and crucially, after experiencing a prolonged period of weight loss following their return to the substratum during September.

Within each region, age influenced maturation in addition to the effect of length, with a shift in the maturity ogive towards smaller length as age increased up to age 2, denoting an increase in the probability of maturation as fish get older. Although there was an apparent reversal of this trend for sandeels aged 3 y, this result could be influenced by year-class differences in maturity-at-age and the small sample size of the age class. Despite the existence of a significant difference in the sex ratio of sandeels in favour of females, a finding that supports that of Macer (1966), our model recorded a statistically significant effect of sex on maturation on just one occasion. Then, the variation in maturity accounted for by sex was negligible. Similarly, our analysis of the cumulative frequency distribution of male and female length data split according to region revealed sex to be a significant effect in only one region, the Firth of Forth. It is possible that this result is driven by the fact that fish there mature smaller, with the higher relative energy cost of producing ovaries in females acting to drive a gender-related difference in the size at which fish mature. Generally, however, our analysis suggests that differences in the timing of maturity between sexes appear mediated solely by differences in length recorded in both sampling periods. Such a mechanism may explain the observation that more male sandeels mature young than females (Bergstad et al., 2001).

The proportions of sandeels maturing in the age 0 class in Elbow Spit, NW Rough, and the Firth of Forth are similar to a previously published figure of 5% maturity recorded from Southernmost Rough (Macer, 1966), an area next to NW Rough. The prevalence of maturity in the age 2-group belonging to these three areas also compares favourably with Macer's (1966) published estimate of 97.9%. However, the maturity rates derived from Fisher Bank differ appreciably from published values for sandeels elsewhere in the North Sea (Macer, 1966; Bergstad et al., 2002). In particular, the faster-growing age 0 sandeels on Fisher Bank exhibited a considerably higher rate of maturity than estimates previously documented for this age group. Nonetheless, in the light of life history theory, such a finding might be expected, because rapid juvenile growth favours a younger age at first reproduction (Charlesworth, 1980), with maturation taking place across a trajectory of size and age ranges that are themselves determined by demographic conditions (Stearns and Crandall, 1984). Also, because sandeels aged 0 on the Fisher Bank grow faster, and because the size overlap between sandeel age classes is small, with maturation decided annually, the size range at maturity will necessarily become more extended. That the high prevalence of maturity in Fisher Bank fish relative to the other regions is not maintained in sandeels aged 1 or 2 is of note, and it is plausible that such low rates of maturation may reflect adverse environmental conditions or strong intraspecific competition experienced by these different year classes that formed the 1999 sample. Clearly, to improve our understanding of maturity–size relationships, future studies need to consider the probability of maturation within a year class, because the differences in age-at-maturity observed at Fisher Bank may well reflect differences in the growth rates of year classes.

It is also conceivable that the regional disparities we found in age- and size-at-maturity might reflect differences in commercial fishing activity between areas. As fish are removed from a local stock, the resultant reduction in competition for food resources may allow faster growth of juveniles and earlier maturity. Therefore, if high rates of fishing lower density dependence in terms of competition for food, then fishing mortality might be met in the short term by a decrease in age, but also an increase in length-at-maturity (Reznick, 1993; Rochet, 1998). Direct evidence for growth compensation attributable to enhanced mortality from commercial exploitation has not yet been found for A. marinus. Although Bergstad et al. (2002) noted a significant difference in length at 50% maturity between a commercially fished area and an unexploited area outside the main North Sea region, fish collected from the exploited area maturing larger, they were unable to conclude that their recorded differences were a response to differences in density. Nevertheless, Nagoshi and Sano (1979), looking at annual variations in growth of age 0 fish in the related A. personatus, found a negative correlation between population density and growth.

Data collected during 1999 and 2004 suggest that there were no changes in length at age 1 within the regions across the two periods. The significant interaction between sample year and region indicates that differences in length at age 1 between regions do, however, alter according to year sampled. That no year effect on length has been found in our study is surprising, given the variable nature of the conditions likely to influence growth of sandeels, and that none was found may be due to chance alone. Previous studies of sandeel growth have, in contrast, found significant year effects when examining growth over several years (Macer, 1966; Wright and Bailey, 1996; Bergstad et al., 2002; Wanless et al., 2004). Regrettably, owing to restrictions created by very poor recruitment to the stock in both 2003 and 2004, our study was only able to test across two periods in three of the regions initially tested, and it is conceivable that the 2 y for which we have data yielded similar growing conditions. With respect to maturation, our study did not indicate temporal variability in the maturity-at-length relationship within each region, suggesting that the maturity ogives depicted in Figure 2 may be stable throughout the time scales under observation. Given this, it is likely that any observed changes in the maturity rate of single age groups between years would be ascribed to changes in length rather than to a change in the shape of the maturity ogive.

The maturity-at-age key currently used by ICES assumes 100% maturity of North Sea sandeels at age 2. Although our estimates for two of the four regions analysed are consistent with this threshold, sandeels from the Firth of Forth and Fisher Banks, which recorded a prevalence of maturity of 79 and 58%, respectively, clearly do not conform to this broad assumption. Moreover, as fecundity also scales to size, lower length-at-age, such as in the Firth of Forth, will also drive reduced fecundity-at-age (Macer, 1966; Gauld and Hutcheon, 1990). The results presented here suggest that considerable regional differences exist within the North Sea with respect to growth and length-at-maturity, the sandeels spawning at age 1 on Fisher Bank contributing significantly more to egg production than sandeels of similar age elsewhere. In terms of sustainability, the regional difference in growth and maturity will influence the local reproductive potential of the different aggregations. Indeed, the aggregations appear largely self-recruiting (Proctor et al., 1998; Munk et al., 2002) and very site-attached following the larval phase (Kunzlik et al., 1986). The combined effects of a greater age-at-maturity and lower age-specific fecundity will tend to make the Firth of Forth sandeel aggregation more vulnerable to collapse through recruitment-overfishing, leading to reduced reproductive potential and hence a reduced capacity to produce recruits. As such, the Firth of Forth sandeel aggregation may be less able to support a level of fishing mortality as high as the other banks, and the time to recovery following a local population collapse may be longer than elsewhere. This sensitivity to fishing pressure is clearly an important issue and must be considered in any future re-opening of the region to fishing. Hence, in managing this important North Sea stock, we advocate a move towards a regionally based stock model that takes into account differences in biomass and potential fecundity per unit weight at a regional level. Attempts have already been made to conduct regional assessments for North Sea sandeels (Pedersen et al., 1999), although they used the stock maturity-at-age key for the total stock to estimate spawning-stock biomass. Work is currently underway to produce realistic spatial models of sandeel population dynamics that account for regional differences in life history traits and local mortality processes. The present study provides a valuable contribution to this work by providing input for spatially resolved estimates of egg production.


   Acknowledgements
 
Our work was supported by the Danish Institute for Fisheries Research and the CFP DG XIV 98/025 and PROTECT (FP6-2003-SSP-3) projects of the European Commission. We thank Cefas and the masters and crews of RV "Dana", FRV "Clupea", and FRV "Corystes" for their help in sample collection.


   References
Top
 Introduction
 Methods
 Results
 Discussion
 References
 

    Bergstad O. A., Høines Å. S., Krüger-Johnsen E. M. (2001) Spawning time, age and size at maturity, and fecundity of sandeel, Ammodytes marinus, in the north-eastern North Sea and in unfished coastal waters off Norway. Aquatic Living Resources 14:293–301.[CrossRef][Web of Science]

    Bergstad O. A., Høines Å. S., Jørgensen T. (2002) Growth of sandeel, Ammodytes marinus, in the northern North Sea and Norwegian coastal waters. Fisheries Research 56:9–23.[CrossRef][Web of Science]

    Charlesworth B. (1980) Evolution in Age-Structured Populations. (Cambridge University Press, Cambridge, UK)300.

    Conover D. O., Arnott S. A., Walsh M. R., Munch S. B. (2005) Darwinian fishery science: lessons from the Atlantic silverside (Menidia menidia). Canadian Journal of Fisheries and Aquatic Sciences 62:730–737.[CrossRef]

    Daan N. (1990) Data base report of the stomach sampling project 1981. ICES Cooperative Research Report, 164 144.

    Fisheries Research Services (FRS). (2004) The Scottish ocean climate status report 2002 and 2003. (Fisheries Research ServicesIn Hughes S. L. (Ed.). , Aberdeen)50 http://www.frs-scotland.gov.uk/Delivery/Information_resources/information_eresources_view_document.aspx?resourceId=37&documentId=1442.

    Fransz H. G., Colebrook J. M., Gamble J. C., Krause M. (1991) The zooplankton of the North Sea. Netherlands Journal of Sea Research 28:1–52.

    Gauld J. A. and Hutcheon J. R. (1990) Spawning and fecundity in the lesser sandeel (Ammodytes marinus Raitt) tagged in the north-western North Sea. Journal of Fish Biology 36:611–613.[CrossRef]

    ICES. (1995) Report of the ICESA Workshop on Sandeel Otolith Analysis. ICES Document CM 1995/G: 4 Ref. L 31.

    ICES. (1999a) Report of the Study Group on Effects of Sandeel Fishing. ICES Document 1999/ACFM: 19 14.

    ICES. (1999b) Manual For The International Bottom Trawl Surveys, Revision VI. ICES Document 1999/D: 2 Addendum 2 Ref. G The International Bottom Trawl Survey Working Group49.

    ICES. (2005) Report of the working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak. ICES CM 2005/ACFM: 09 981.

    Kjesbu O. S., Klungsoyr J., Kryvi H., Witthames P. R., Greer-Walker M. (1991) Fecundity, atresia, and egg size of captive Atlantic cod (Gadus morhua) in relation to proximate body composition. Canadian Journal of Fisheries and Aquatic Sciences 48:2333–2343.

    Kunzlik P. A., Gauld J. A., Hutcheon J. R. (1986) Preliminary results of the Scottish sandeel tagging project. 5 ICES Document CM 1986/G: 76.

    Lambert Y., Yaragina N. A., Kraus G., Marteinsdottir G., Wright P. J. (2004) Correlation between reproductive characteristics and environmental and biological indices as alternative methods of estimating egg and larval production. Journal of Northwest Atlantic Fisheries Science 33:115–159.

    Macer C. T. (1966) Sand eels (Ammodytidae) in the southern North Sea. Journal of the Marine Biological Association of the UK 45:187–207.

    Marteinsdottir G. and Begg G. (2002) Essential relationships incorporating the influence of age, size and condition on variables required for estimation of reproductive potential in Atlantic cod Gadus morhua stocks. Marine Ecology Progress Series 235:235–256.[Web of Science]

    Marteinsdottir G. and Steinarsson A. (1998) Maternal influence on the size and viability of Iceland cod (Gadus morhua L.) eggs and larvae. Journal of Fish Biology 52:1241–1258.[CrossRef]

    McCullagh P. and Nelder J. A. (1983) Generalized Linear Models. (Chapman and Hall, London)261.

    Monaghan P. (1992) Seabirds and sandeels: the conflict between exploitation and conservation in the northern North Sea. Biodiversity and Conservation 1:98–111.

    Munk P., Wright P. J., Pihl N. J. (2002) Distribution of the early life history stages of cod, plaice and sandeels across haline fronts in the North Sea. Estuarine, Coastal and Shelf Science 55:139–149.[CrossRef]

    Nagoshi M. and Sano M. (1979) Population studies of the sand eel, Ammodytes personatus, in Ise Bay. 1. Growth and its relation to population density. Japanese Journal of Ecology 29:1–10.

    Olsen E. M., Knutsen H., Gjøsæter J., Jorde P. E., Knutsen J. A., Stenseth N. C. (2004) Life-history variation among local populations of Atlantic cod from the Norwegian Skagerrak coast. Journal of Fish Biology 64:1725–1730.[CrossRef]

    Pedersen S. A., Lewy P., Wright P. J. (1999) Assessments of the lesser sandeel (Ammodytes marinus) in the North Sea based on revised stock divisions. Fisheries Research 41:221–241.[CrossRef][Web of Science]

    Popp-Madsen K. (1994) Tobis or not Tobis. Er vigtigt spørgsmål for dansk fiskeri. Fiskeri-Og Havundersogelsen 45:27–34.

    Proctor R., Wright P. J., Everitt A. (1998) Modelling the transport of larval sandeels on the Northwest European shelf. Fisheries Oceanography 7:347–354.[CrossRef][Web of Science]

    Reznick D. N. (1993) Norms of reaction in fishes. In Stokes T. K., McGlade J. M., Law R. (Eds.). The Exploitation of Evolving Resources(Springer, Berlin) pp. 72–90 Lecture Notes in Biomathematics.

    Rochet M-J. (1998) Short-term effects of fishing on life history traits of fishes. ICES Journal of Marine Science 55:371–391.[Abstract/Free Full Text]

    Roff D. A. (1983) An allocation model of growth and reproduction in fish. Canadian Journal of Fisheries and Aquatic Sciences 40:1395–1404.

    Roff D. A. (1992) The Evolution of Life Histories: Theory and Analysis. (Chapman and Hall, New York).

    Ryland J. S. (1964) The feeding of plaice and sandeel larvae in the southern North Sea. Journal of the Marine Biological Association of the UK 44:343–364.

    Sherman K., Jones C., Sullivan L., Smith W., Derrien P., Ejsymont L. (1981) Congruent shifts in sand eel abundance in western and eastern North Atlantic ecosystems. Nature 291:486–489.[CrossRef]

    Shine R. and Schwarzkopf L. (1992) The evolution of reproductive effort in lizards and snakes. Evolution 46:62–75.[CrossRef]

    Sinclair M. (1988) Marine Populations. An Essay on Population Regulation and Speciation. (Washington Sea Grant, Seattle, Washington)252.

    Stearns S. C. and Crandall R. E. (1984) Plasticity for age and size at sexual maturity: a life history response to unavoidable stress. In Potts G. and Wootton R. J. (Eds.). Fish Reproduction(Academic Press, London) pp. 13–33.

    Stephenson R. L. (1999) Stock complexity in fisheries management: a perspective of emerging issues related to population sub-units. Fisheries Research 43:247–249.[CrossRef][Web of Science]

    Swartzman G., Silverman E., Williamson N. (1995) Relating trends in walleye pollock (Theragra chalcogramma) abundance in the Bering Sea to environmental factors. Canadian Journal of Fisheries and Aquatic Sciences 52:369–380.

    Wanless S., Wright P. J., Harris M. P., Elston D. A. (2004) Evidence for decrease in size of lesser sandeels Ammodytes marinus in a North Sea aggregation over a 30-yr period. Marine Ecology Progress Series 279:237–246.[Web of Science]

    Warburton K. (1982) Sandeels—the elusive species. Scottish Fisheries Bulletin 4:22–27.

    Winslade P. (1974a) Behavioural studies on the lesser sandeel Ammodytes marinus (Raitt). 2. The effect of light intensity on activity. Journal of Fish Biology 6:577–586.[CrossRef]

    Winslade P. (1974b) Behavioural studies on the lesser sandeel Ammodytes marinus (Raitt). 3. The effect of temperature on activity and the environmental control of the annual cycle of activity. Journal of Fish Biology 6:587–599.[CrossRef]

    Wright P. J. (1993) Otolith microstructure of the lesser sandeel, Ammodytes marinus. Journal of the Marine Biological Association of the UK 73:245–248.

    Wright P. J. (1996) Is there a conflict between sandeel fisheries and seabirds? A case study at Shetland. In Greenstreet S. P. R. and Tasker M. L. (Eds.). Aquatic Predators and their Prey(Fishing News Books, Oxford) pp. 154–165.

    Wright P. J. and Bailey M. C. (1996) Timing of hatching in Ammodytes marinus from Shetland waters and its significance to early growth and survivorship. Marine Biology 126:143–152.[CrossRef]

    Wright P. J., Jensen H., Tuck I. (2000) The influence of sediment type on the distribution of the lesser sandeel, Ammodytes marinus. Journal of Sea Research 44:243–256.[CrossRef]

    Wright P. J., Pedersen S. A., Donald L., Anderson C., Lewy P., Proctor R. (1998) The influence of physical factors on the distribution of lesser sandeel, Ammodytes marinus and its relevance to fishing pressure in the North Sea. 17 ICES Document CM 1998/AA: 3.

    Yoneda M. and Wright P. J. (2004) Temporal and spatial variation in reproductive investment of Atlantic cod Gadus morhua in the northern North Sea and Scottish west coast. Marine Ecology Progress Series 276:237–248.[Web of Science]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 ICES J. Mar. Sci.Home page
A. Christensen, H. Mosegaard, and H. Jensen
Spatially resolved fish population analysis for designing MPAs: influence on inside and neighbouring habitats
ICES J. Mar. Sci., January 1, 2009; 66(1): 56 - 63.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
64/2/369    most recent
fsl033v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Boulcott, P.
Right arrow Articles by Gibb, I. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Boulcott, P.
Right arrow Articles by Gibb, I. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?