© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Feeding preferences of herring (Clupea harengus) and sprat (Sprattus sprattus) in the southern Baltic Sea
National Board of Fisheries, Institute of Marine Research Box 4, 453 21 Lysekil, Sweden
*Correspondence to M. Casini: tel: +46(0)523 18728; fax: +46(0)523 13977. e-mail: michele.casini{at}fiskeriverket.se.
No field studies have been performed on the selectivity of herring and sprat in the southern Baltic Sea in relation to their entire range of prey. Accordingly, we tested in the field the following hypotheses: (i) sprat and herring are selective feeders and (ii) sprat and herring selectivity is size- and season-dependent. The results show that (i) smaller herring and all size classes of sprat are strictly zooplanktivorous, selecting principally Temora longicornis and Bosmina maritima during the autumn and Pseudocalanus elongatus in winter; (ii) larger herring are essentially nektobenthos feeders, predating on Mysis mixta during the autumn and amphipods and polychaetes during the winter; and (iii) herring and sprat seem to avoid Acartia spp. in both autumn and winter. During the autumn, herring are zooplanktivorous up to 18–20 cm, whereas in winter herring feed on nektobenthos starting from 14–15 cm. Selectivity was not an absolute process but it was related to prey relative abundance in the sea and, possibly, to prey profitability (e.g. size, conspicuousness, and reaction time).
Keywords: Bornholm Basin, clupeids, competition, size- and season-dependent food selectivity, zooplankton and nektobenthos
Received 26 June 2003; accepted 10 December 2003.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Herring (Clupea harengus) and sprat (Sprattus sprattus) are the two most important commercial pelagic fish species in the Baltic Sea (Cardinale and Arrhenius, 2000). They are also the dominant zooplanktivorous species and the principal prey for several piscivores, thereby playing a significant role in the ecosystem of the Baltic Sea (Hansson and Rudstam, 1990; Rudstam et al., 1992; Arrhenius and Hansson, 1993; Karlsson et al., 1999). Most Baltic Sea studies on the diet of sprat and herring have described their stomach contents, whereas issues on their prey selectivity often have been omitted (Flinkman, 1999). Sandström (1980) was the first to highlight the selective nature of herring predation. In two more recent papers, Rudstam et al. (1992) and Arrhenius (1996) studied the selective predation of herring and sprat in the Northern Baltic, although zooplankton taxa were grouped at the family or even subclass level. Flinkman et al. (1992, 1998) were the first to compare stomach contents of Baltic herring to contemporary zooplankton samples down to species or development stage level. Selectivity of herring in relation to zooplankton availability is well studied in the northern part of the Baltic (i.e. Flinkman et al., 1992, 1998; Rudstam et al., 1992; Arrhenius and Hansson, 1994; Viitasalo et al., 2001). On the other hand, while investigations on feeding of herring and sprat exist (i.e. Szypula, 1992; Köster and Schnack, 1994; Szypula et al., 1997a, b; Möllmann and Köster, 1999), herring selectivity is generally unknown in the southern Baltic Sea where the importance of prey other than zooplankton in the diet is greater due to the larger size range of herring in this area (Cardinale and Arrhenius, 2000). Moreover, there is no information on comparative selectivity of herring and sprat for the Baltic Sea in relation to the entire spectrum of available prey (i.e. zooplankton and nektobenthos). Sprat is a strict zooplanktivore while herring change its diet with fish size (Arrhenius and Hansson, 1993). Nevertheless, for the estimation of herring selection, nektobenthos resources have not been considered in earlier studies.
Sprat and herring growth rates have changed during the last two decades following similar patterns (Cardinale et al., 2002). These variations in growth rates have been explained by changes in clupeid total abundance and/or changes in zooplankton abundance and composition. Therefore, information on the composition and dynamics of the Baltic foodweb is of fundamental importance to increase our knowledge on the Baltic ecosystem. In addition, this information is needed for a more precise assessment and prediction of fisheries resources.
We studied the feeding interactions and ontogenesis of herring and sprat in relation to all potential available feeding resources sampled nearly simultaneously in time and space. We tested the following hypotheses: (i) both sprat and herring are selective feeders and (ii) sprat and herring selectivity is size and season dependent.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Fish sampling
Fish samples were taken at six sites in the Bornholm Basin (Figure 1) during the hydroacoustic trawl survey in October 1999 and the Baltic International Fish Survey in March 2000 onboard the Swedish RV "Argos" (Table 1). The Bornholm Basin is considered one of the main feeding areas for both herring and sprat in the Southern Baltic. Fish were sampled by means of trawling (30-min hauls) during the daylight hours and were immediately (within 15 min after trawl recovery) frozen. A subsample of at least 100 herring and 75 sprat individuals was randomly taken from each haul for stomach analysis. For all specimens, total fish length (to the nearest 0.5 cm) was recorded.
|
|
Zooplankton sampling
Zooplankton samples were collected at 21 stations in October and 14 stations in March (three replicates at each occasion) in the Bornholm Basin (Figure 1 and Table 1). Clupeids in the Southern Baltic Sea feed upon zooplankton mainly at twilight (herring) and the first hours of the day (sprat) mostly, even though not exclusively, in the uppermost 50 m of the water column (Cardinale et al., 2003) where food availability and light condition are possibly optimal (Orlowski, 2000; Cardinale et al., 2003). Moreover, it has been recognized that zooplankton, in response to predation, perform diel vertical migrations (DVM) in shallow areas of the Baltic Sea, inhabiting the deeper water during the daytime and the upper water column at night (Hansson et al., 1990). This pattern has been suggested also for deeper regions of the Baltic Sea (Kornilovs et al., 2001) and demonstrated in areas out of the Baltic basin (Pagano et al., 1993). Thus, although studies on zooplankton DVM are lacking in literature for the Bornholm Basin, in our study the zooplankton found at night in the uppermost 50 m of the water column was assumed to represent the food available for pelagic fish.
Zooplankton sampling took place during the night immediately after fish trawling using a vertical WP-2 net equipped with a 90-µm mesh size net and a flowmeter (bottom depth range 67–86 m). The plankton net was towed vertically from a depth of 50 m (i.e. from below the thermocline) to the surface at a speed of 0.5 m s–1. Zooplankton samples were preserved in 4% formalin solution buffered with disodiumtetraborate (Dybern et al., 1976). The samples were subsampled (six subsamples of the same volume) (Johansson et al., 1993) and analysed under an inverted microscope at a magnification of 60x. At least 100 specimens from each subsample were counted and identified at the species level. For copepods, adults and all the copepodite stages were considered while nauplii were excluded because a 90-µm mesh size net does not sample them quantitatively (Flinkman et al., 1992). Moreover, the incidence of nauplii in the stomachs was scarce during the study period.
Nektobenthos sampling
Nektobenthos samples were collected with a 1-m diameter 500-µm opening/closing Method Isaacs-Kidd (MIK) net at 21 stations, corresponding to the stations sampled for zooplankton. Due to limited ship time, sampling of nektobenthos was restricted to the autumn period. We assumed that nektobenthos found in the uppermost 70 m of the water column would represent the food available for pelagic fish. The MIK net was towed obliquely (45°) from a depth of 70 m (i.e. from below the thermocline) to the surface at a mean speed of 2.0 m s–1. Nektobenthos sampling took place during the night immediately after fish trawling (bottom depth range 67–86 m). All nektobenthos samples were preserved in 4% formalin buffered with disodiumtetraborate (Dybern et al., 1976). Nektobenthos individuals were identified at the species level using an inverted microscope at a magnification of 15x.
Fish stomach analysis
Stomach contents were collected for each individual of sprat and herring in order to determine prey abundance and composition. Zooplankton and nektobenthos, which were the two groups of food items found in the stomachs, were preserved separately in 70% alcohol. Zooplankton items were later sorted under an inverted microscope at a magnification of 60x into taxonomic groups and identified at the species level whenever possible. A subsample of at least 100 zooplankton individuals for each stomach of herring and sprat was identified counting the carapaces (Flinkman et al., 1998). Nektobenthos individuals were immediately identified at the species level using an inverted microscope at a magnification of 15x.
The importance of the different prey types was expressed using a numerical index (NI) defined as the average proportion of individuals of the ith prey type with respect to the total number of prey consumed for each fish size class (Hyslop, 1980; Linton et al., 1981; Orr and Bowering, 1997).
Statistical analysis
Data did not fulfil the assumptions of independence, homogeneous variance, normality, and linearity. Therefore, the data were log-transformed before the analysis (Sokal and Rohlf, 1995).
The mean NI of each prey item in the different seasons were tested against the different size classes (1 cm) of herring and sprat using one-way analysis of variance (ANOVA) and regression analysis (Sokal and Rohlf, 1995).
In order to determine trophic association (i.e. using the NI) among size classes of herring and sprat in the different seasons, a non-metric multi-dimensional scaling (NMDS) based on Bray–Curtis similarity was used for classification and ordination of size classes, respectively (Statistica, 1995). NMDS preserves the rank order of the inter-samples distance, as opposed to the linear relationship of classical metric scaling (i.e. Principal Component Analysis, Correspondence Analysis). NMDS has the advantage of robustness, e.g. not being sensitive to outliers and to normality and homoscedascity assumptions of classical metric scaling (Clark et al., 1996). As in factor analysis, the actual orientation of axes in the final solution is arbitrary. Thus, the final orientation of axes in the plane or space is mostly the result of a subjective decision taken by the researcher, who will choose an orientation that can be most easily explained (Statistica, 1995). NMDS uses a function minimization algorithm that evaluates different configurations with the goal of maximizing the goodness-of-fit. The most common measure of goodness-of-fit used to evaluate how well (or poorly) a particular configuration reproduces the observed distance matrix is the stress measure. Thus, the smaller the stress value, the better is the fit of the reproduced distance matrix to the observed distance matrix (Johnson and Wichern, 1998).
Selectivity was studied for each size class of herring and sprat in the different seasons using the V-index (Pearre, 1982; Flinkman et al., 1992, 1998), which is based on the chi-square:
|
|
The V-index value ranges between 1 (absolute selection) and –1 (absolute rejection) and corresponds to zero for no selection. We used Fisher's exact test for significant difference (Sokal and Rohlf, 1995) to test for significance of selection indices as recommended by Pearre (1982).
Statistical analysis was performed using Statistica (1995) and S-Plus (2000) computer software. The level of significance was set at 5% for all the statistical tests.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Autumn samples
The stomachs from 707 herring and 334 sprat were analysed (Table 2). The percentage of empty stomachs increased significantly with the length of fish for both herring (r2 = 0.94; p < 0.0001; n = 16) and sprat (r2 = 0.74; p < 0.003; n = 9).
|
Zooplankton biomass in the sea ranged between 4–6 g100 m–3 wet weight (Table 1). Temora longicornis, Acartia spp., Evadne nordmanni, and Bosmina maritima were the most abundant species in the zooplankton samples (Figure 2). Limnocalanus grimaldi, Pseudocalanus elongatus and nektobenthos species were nearly absent. Nektobenthos biomass varied from 0.1 to 1.9 g100 m–3 wet weight (Table 1).
|
There were marked differences in the abundance percentages of zooplankton and nektobenthos species between the sea and the stomachs of herring and sprat (Figure 2). Acartia spp. was frequent in the sea but not in the stomachs. Herring stomachs contained a larger abundance percentage of T. longicornis, while the cladocerans Podon intermedius, Podon polyphemoides, and B. maritima were more abundant in sprat stomachs. Mysis mixta and amphipods were only found in herring stomachs.
In herring, there was a general decrease in zooplankton with increasing fish size and a significant (r2 = 0.79; p < 0.001) increase of M. mixta (Figure 3). T. longicornis and B. maritima decreased significantly (r2 = –0.53; p < 0.003 and r2 = –0.76; p < 0.001, respectively) with increasing size of herring. On the other hand, all size classes of sprat contained a high abundance percentage of T. longicornis, whereas the percentage of B. maritima increased significantly (r2 = 0.71; p < 0.005) with the size of sprat.
|
Trophic interactions between size classes of sprat and herring were investigated by NMDS (Figure 4). Smaller herring (
20 cm) and larger sprat (
11 cm) constituted a separate group in the NMDS while large herring (21 cm) and small sprat (10 cm) were distinct in the ordination. A certain degree of separation was also shown for intermediate size herring (18–20 cm), which constituted a fairly distinct trophic aggregation in the NMDS compared to smaller herring.
|
All size classes of both herring and sprat showed a negative selection (Fisher's exact test, p < 0.05) for Acartia spp. (Figure 5). Smaller herring individuals (
20 cm) significantly selected both T. longicornis and B. maritima while larger herring (18 cm) selected M. mixta. Concerning sprat, most size classes selected B. maritima, P. polyphemoides, and T.longicornis whereas smaller individuals (7 cm) fed mostly on the latter species. Figure 5 shows only the species whose V-index was significant (either positive or negative) as calculated by Fisher's exact test (not shown).
|
Winter samples
A total of 734 herring and 691 sprat were used for stomach analysis (Table 2). As for the autumn samples, the percentage of empty stomachs increased significantly with the length of both herring (r2 = 0.51; p < 0.002; n = 16) and sprat (r2 = 0.51; p < 0.05; n = 8).
Zooplankton abundance in the sea varied from 0.5 to 1.3 g100 m–3 wet weight (Table 1). Acartia spp., Pseudocalanus elongatus, T. longicornis and Centrophages hamatus were the most abundant species in the zooplankton samples, whereas L. grimaldi and cladoceran species were nearly absent (Figure 2).
There were noticeable differences in the abundance percentage of zooplankton species in the sea and in herring and sprat stomachs (Figure 2). Acartia spp. was abundant in the environment but scarce in herring and sprat stomachs. P. elongatus was more frequent in sprat stomachs while nektobenthos (amphipods, M. mixta, and polychaetes) was found exclusively in herring stomachs.
The abundance percentage of P. elongatus and T. longicornis decreased significantly (r2 = –0.61; p < 0.001; n = 14) while polychaetes showed a significant increase (r2 = 0.64; p < 0.001; n = 14) with increasing size of herring (Figure 6). On the other hand, amphipods and M. mixta were present in almost all size classes of herring and without any specific trend. Concerning sprat, the abundance percentage of P. elongatus decreased significantly (r2 = 0.56; p < 0.05; n = 7) with fish size, whereas T. longicornis showed a significant (r2 = 0.68; p < 0.05; n = 7) increase with sprat length.
|
NMDS (Figure 7) showed the presence of three clearly distinct trophic groups: all sprat size classes and herring small individuals (11–13 cm), medium size herring (14–22 cm), and large herring (23–24 cm).
|
All size classes of herring and sprat significantly avoided (Fisher's exact test, p < 0.05) Acartia spp. and C. hamatus (Figure 8). Smaller herring (<14 cm) significantly selected P. elongatus, while larger individuals (
14 cm) avoided P. elongatus and T. longicornis. Concerning sprat, all the size classes showed a significant selection for P. elongatus. Figure 8 shows only the species whose V-index was significant (either positive or negative) as calculated by Fisher's exact test (not shown).
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This study is the first attempt to estimate in the field the selectivity of herring and sprat in relation to both zooplankton and nektobenthos availability. Here we show that (i) small herring (<13–15 cm) and sprat are strictly zooplanktivorous, principally selecting Temora longicornis and Bosmina maritima during the autumn and Pseudocalanus elongatus in winter; (ii) larger herring (>15–20 cm) are essentially nektobenthos feeders, predating Mysis mixta during the autumn and amphipods and polychaetes during the winter; and (iii) herring and sprat do not select Acartia spp. either in autumn or in winter.
Herring and sprat in the Baltic Sea form huge schools constituting around 97% of the pelagic biomass (Orlowski, 2001), performing extensive vertical migration (DVM) during the 24-hours (Orlowski, 1998, 2000, 2001; Cardinale et al., 2003) and feeding on zooplankton aggregations and, regarding herring, nektobenthos (Arrhenius and Hansson, 1993). This implies that herring and sprat are, at least partly, potential competitors for the same available resources. However, unequivocally demonstrating competition is not a simple task, requiring information about limitation of available food resources. Nevertheless, there is evidence that zooplankton resources are a limiting factor on the growth rates of pelagic fish in the Baltic Sea (Cardinale and Arrhenius, 2000; Cardinale et al., 2002), indicating a potential intra- and inter-specific competition among pelagic fish.
Results from this study support the hypothesis (i) of highly selective feeding of both herring and sprat and (ii) that herring and sprat prey spectra are size - dependent. The favoured zooplankton prey species for sprat and smaller herring were large and conspicuous copepods such as T. longicornis (autumn) and P. elongatus (winter) and low escape response (Flinkman et al., 1998; Viitasalo et al., 2001) cladocerans such as B. maritima for both species and P. polyphemoides for sprat (autumn). These results possibly furnish indirect evidence that prey selection of herring is based on prey size, conspicuousness, and reaction time (i.e. alertness to hydrodynamic signals), as predicted by Flinkman et al. (1998) and Viitasalo et al. (2001) for herring in the Northern Baltic. Nevertheless, the same is applicable to sprat, since selectivity was similar for the two species when comparing individuals of similar size. Interestingly, herring from the Northern Baltic negatively selected P. elongatus in the summer (Viitasalo et al., 2001) while P. elongatus was positively selected in the Southern Baltic (this study) during the winter. This could be explained by the fact that in our study P. elongatus was, together with Acartia spp., the most abundant species in the sea during the winter, while the abundance of the favourite species T. longicornis (Viitasalo et al., 2001; this study) was particularly low compared to the autumn. Moreover, in winter, cladocerans disappeared from the zooplankton. In these circumstances, we suggest that herring and sprat fed on P. elongatus plausibly because it was the most abundant prey with the relatively higher profitability. These results suggest that selectivity is not an absolute process but that it depends on both the profitability of the prey (i.e. size, conspicuousness, and reaction time) (Viitasalo et al., 2001) and the prey relative abundance (our study).
It must be held in mind that, whereas T. longicornis, Acartia spp., C. hamatus, and cladocera inhabit the 0–50-m depth range (Möllmann and Köster, 2002), in the Baltic deep basins the older copepodite stages (CVI) of P. elongatus may be distributed, especially in autumn, deeper in the water column (Möllmann and Köster, 2002). Thus, in our study, this part of the P. elongatus population could not have been included in the zooplankton sampling, that covered the uppermost 50 m only of the water column, and its total abundance might have been underestimated with a consequent overestimation of the V-index. Nevertheless, since in our study the V-index was negative in autumn for both herring and sprat, we consider this result not biased. Moreover, in the study by Möllmann and Köster (2002), zooplankton samples were collected during the daytime when there is strong evidence that zooplankton inhabits deeper strata of the water column (Hansson et al., 1990; Pagano et al., 1993; Kornilovs et al., 2001). Since in autumn in the Bornholm Basin, clupeids (especially sprat and smaller herring) feed mostly, even though not exclusively, between 0 and 50 m (Orlowski, 2000; Cardinale et al., 2003), the prey found in this depth interval could reliably represent their available prey. However, we are aware that limiting the zooplankton sampling to the uppermost 50-m depth could have introduced bias in our analysis on the feeding preferences of the larger, deeper dwelling, herring.
Acartia spp. seemed to be always deselected even in spite of its high abundance in the sea. This result is in agreement with Möllmann and Köster (2002) who found that Acartia spp., despite its large abundance, was not consumed by herring and sprat in the central Baltic Sea. This could be explained by the small size and high escape response of Acartia spp. (Viitasalo et al., 2001), but also by the fact that this genus is a surface dweller (Hansson et al., 1990; Möllmann and Köster, 2002) that performs less pronounced DVM. During the daytime (when the fish samples were collected), the clupeids are distributed and may continue feeding, deeper in the water column (Köster and Schnack, 1994; Cardinale et al., 2003) thus perhaps not always spatially overlapping with Acartia spp. during their DVM.
Another interesting result of our study is that the size of herring individuals predating on nektobenthos changed substantially with season. During the autumn, herring was zooplanktivorous until 18–20-cm, whereas in the winter it predated on nektobenthos starting from 14–15-cm mg. Therefore, in autumn there was a wider feeding overlap between herring and sprat, with small (11–14 cm) as well as medium size herring (15–20 cm) occupying the same trophic group as sprat. Generally, a change in diet occurs when fish approach a critical size (i.e. Ross, 1978; Cardinale, 2000). Such a scenario has been explained as means to optimize the energy intake and the time in obtaining and digesting prey as well as to avoid intra- and interspecific competition (i.e. Ross, 1978; Harmelin-Vivien et al., 1989; Cardinale et al., 1997). However, results from our study indicated that diet shifts are not only size (ontogenesis)-dependent but also dependent on the season and, therefore, on prey abundance and composition.
In our study we did not discriminate between zooplankton sexes and among the different copepodite stages of the copepods. Since, as showed by Viitasalo et al. (2001), herring can select single prey stages and sex, the results reported here must be taken cautiously and considered as representing averages on the entire stage spectrum of each prey. Additionally, we are aware that pooling 50-m depth is a wide range for fish to feed and that we were not able to resolve the feeding interactions existing between predator and prey at smaller vertical scale and to investigate the deeper strata where clupeids can continue feeding during their DVM. As a matter of fact, interpretations, of the selection process require good information on the vertical overlap of predator and prey. Therefore, further vertical small-scale investigations, perhaps by using stratified sampling, extended to the entire water column are needed to better understand the selective feeding behaviour of clupeids in the Baltic Sea.
In the late 1980s and early 1990s, the growth rates of herring and sprat significantly decreased in all the areas of the Baltic Sea (Cardinale and Arrhenius, 2000; Cardinale et al., 2002). This has been argued to occur, as a consequence of a decreased cod biomass, because of an abrupt increase in sprat abundance with a consequent density-dependent reduction in clupeid individual food intake (top–down mechanism) (Cardinale and Arrhenius, 2000; Cardinale et al., 2002). On the other hand, the decrease in clupeid growth rates could also have been related, owing to unfavourable environmental condition (e.g. low-salinity levels), to a reduction in their main food items, namely P. elongatus and T. longicornis (bottom–up mechanism) (Flinkman et al., 1998; Möllmann et al., 2000; Cardinale et al., 2002). A combination of the two mechanisms could possibly represent the most plausible explanation. Our study shows that in the southern Baltic Sea, herring and sprat utilize, at least partially, the same feeding resources by selecting/deselecting the same prey species. This could explain the positive correlation between herring and sprat growth rates observed in the Baltic Sea during the last 15 years of the past century (Cardinale et al., 2002).
The concept of selective predation by consumers on specific prey types and sizes is the major process involved in top–down processes (Kitchell and Carpenter, 1993). We showed that herring and sprat select certain prey items in different periods of the year. This shows the potential for clupeids to impact the zooplankton community (Flinkman et al., 1992; Kitchell and Carpenter, 1993; Möllmann and Köster, 1999, 2002). Möllmann and Köster (1999), and Rudstam et al. (1992, 1994) pointed out that predation by clupeids can contribute to the seasonal development of zooplankton species in the Baltic Sea. There are also indications that interannual variability of zooplankton abundance might be affected by clupeid predation pressure (Kornilovs et al., 2001; Möllmann and Köster, 2002). Therefore, further studies on selectivity of planktivorous fish should be performed in order to understand better the dynamics of zooplankton/nektobenthos populations and the mechanisms triggering the changes in clupeid growth rates.
|
Acknowledgements |
|---|
We are grateful to the crew of the RV "Argos" and the staff at the Institute of Marine Research of Lysekil for the assistance during the surveys in collecting the data. We are indebted to Italienska Kulturinstitutet, the University of Bologna, The Swedish Institute, and the European Commission Directorate (General DG XIV, Fisheries, FAIR GT96-1298, and study project 96/071) that provided financial support. Finally, we thank two anonymous referees for their valuable comments on an earlier version of the manuscript.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Arrhenius F. (1996) Diet composition and food selectivity of 0-group herring (Clupea harengus L.) and sprat (Sprattus sprattus L.) in the northern Baltic Sea. ICES Journal of Marine Science 53:701–712.
Arrhenius F. and Hansson S. (1993) Food consumption of larval, young and adult herring and sprat in the Baltic Sea. Marine Ecology Progress Series 96:125–137.[ISI]
Arrhenius F. and Hansson S. (1994) In situ food consumption by young-of-the-year Baltic Sea herring (Clupea harengus). A test of predictions from a bioenergetics model. Marine Ecology Progress Series 110:145–149.[ISI]
Cardinale M. (2000) Ontogenetic diet shifts of bull-rout, Myoxocephalus scorpius (L.) in the south-western Baltic Sea. Journal of Applied Ichthyology 16:231–239.[CrossRef][ISI]
Cardinale M. and Arrhenius F. (2000) Decreasing weight-at-age of Atlantic herring (Clupea harengus) from the Baltic Sea between 1986 and 1996: a statistical analysis. ICES Journal of Marine Science 57:1–12.
Cardinale M., Casini M., Arrhenius F. (2002) The influence of biotic and abiotic factors on the growth of sprat (Sprattus sprattus) in the Baltic Sea. Aquatic Living Resources 15:273–281.[CrossRef][ISI]
Cardinale M., Casini M., Arrhenius F., Håkansson N. (2003) Diel spatial distribution and feeding activity of herring (Clupea harengus) and sprat (Sprattus sprattus) in the Baltic Sea. Aquatic Living Resources 16:283–292.[CrossRef][ISI]
Cardinale M., Colloca F., Ardizzone G.D. (1997) Feeding ecology of Mediterranean razorfish Xyrichthys novacula in the Tyrrhenian Sea (Central Mediterranean Sea). Journal of Applied Ichthyology 13:105–111.
Clark B.M., Bennett B.A., Lamberth S.J. (1996) Temporal variations in surf zone fish assemblages from False Bay, South Africa. Marine Ecology Progress Series 131:35–47.[ISI]
Dybern B., Ackefors H., Elmgren R. (1976) Recommendations on methods for marine biological studies in the Baltic. Baltic Marine Biologists, Publication no. 1. 98 pp.
Flinkman J. (1999) Interactions between plankton and planktivores of the northern Baltic Sea: selective predation and predation avoidance. PhD thesis, University of Helsinki, Finland.
Flinkman J., Aro E., Vuorinen I., Viitasalo M. (1998) Changes in the northern Baltic zooplankton and herring nutrition from 1980's to 1990's: top–down and bottom–up processes at work. Marine Ecology Progress Series 163:127–136.
Flinkman J., Vourinen I., Aro E. (1992) Planktivorous Baltic herring (Clupea harengus) prey selectivity on reproducing copepods and cladocerans. Canadian Journal of Fisheries and Aquatic Sciences 49:73–77.
Hansson S., Larsson U., Johansson S. (1990) Selective predation by herring and mysids, and zooplankton community structure in a Baltic Sea coastal area. Journal of Plankton Research 12:1099–1116.
Hansson S. and Rudstam L.G. (1990) Eutrophication and Baltic fish communities. Ambio 19:123–125.
Harmelin-Vivien M.L., Kaim-Malka M., Ledoyer M., Jacob-Abraham S.S. (1989) Food partitioning among scorpaenid fishes in Mediterranean seagrass beds. Journal of Fish Biology 34:715–734.[CrossRef][ISI]
Hyslop E.J. (1980) Stomach contents analysis, a review of methods and their application. Journal of Fish Biology 17:411–429.[CrossRef][ISI]
Johansson S., Hansson S., Araya-Nunez O. (1993) Temporal and spatial variation of coastal zooplankton in the Baltic Sea. Ecography 16:167–173.
Johnson R.A. and Wichern D.W. (1998) Applied Multivariate Statistical Analysis 4th edn (Prentice-Hall Inc., Upper Saddle River, New Jersey) 809 pp.
Karlsson L., Ikonen E., Mitans A., Hansson S. (1999) The diet of salmon (Salmo salar) in the Baltic Sea and connection with the M74 syndrome. Ambio 28:37–42.
Kitchell J.F. and Carpenter S.R. (1993) Cascading trophic interactions. In Carpenter S.R. and Kitchell J.F. (Eds.). The Trophic Cascade in Lakes(Cambridge University Press, Cambridge) pp. 1–14 385 pp.
Kornilovs G., Sidrevics L., Dippner J.W. (2001) Fish and zooplankton interaction in the Central Baltic Sea. ICES Journal of Marine Science 58:579–588.
Köster F.W. and Schnack D. (1994) The role of predation on the early life stages of cod in the Baltic. Dana 10:179–201.
Linton L.R., Davies R.W., Wrona F.J. (1981) Resources utilisation index: an assessment. Journal of Fish Biology 18:411–429.[CrossRef]
Möllmann C., Kornilovs G., Sidrevics L. (2000) Long-term dynamics of main mesozooplankton species in the central Baltic Sea. Journal of Plankton Research 22:2015–2038.
Möllmann C. and Köster F.W. (1999) Food consumption by clupeids in the Central Baltic: evidence for top–down control? ICES Journal of Marine Science 56:Suppl., 100–113.[CrossRef][ISI]
Möllmann C. and Köster F.W. (2002) Population dynamics of calanoid copepods and the implications of their predation by clupeid fish in the Central Baltic Sea. Journal of Plankton Research 24:959–977.
Orlowski A. (1998) Acoustic methods applied to fish environmental studies in the Baltic Sea. Fisheries Research 34:227–237.[CrossRef][ISI]
Orlowski A. (2000) Diel dynamics of acoustic measurements of Baltic fish. ICES Journal of Marine Science 57:1196–1203.
Orlowski A. (2001) Behavioural and physical effect on acoustic measurements of Baltic fish within a diel cycle. ICES Journal of Marine Science 58:1174–1183.
Orr D.C. and Bowering W.R. (1997) A multivariate analysis of food and feeding trends among Greenland halibut (Reinhardtius hippoglossoides) sampled in Davis Strait, during 1986. ICES Journal of Marine Science 54:819–829.
Pagano M., Gaudy R., Thibault D., Lochet F. (1993) Vertical migration and feeding rhythms of mesozooplanktonic organisms in the Rhone River plume area (North-West Mediterranean Sea). Estuarine, Coastal and Shelf Science 37:251–269.[CrossRef]
Pearre S. (1982) Estimating prey preference by predators: uses of various indices, and a proposal of another based on 
2. Canadian Journal of Fisheries and Aquatic Sciences 39:914–923.
Ross S.T. (1978) Trophic ontogeny of the leopard sea robin, Prionotus scitulus, (Pisces: Triglidae). Fishery Bulletin 76:225–234.[ISI]
Rudstam L.G., Aneer G., Hildén M. (1994) Top–down control in the pelagic Baltic ecosystem. Dana 10:159–173.
Rudstam L.G., Hansson S., Johansson S., Larsson U. (1992) Dynamics of planktivory in a coastal area of the northern Baltic Sea. Marine Ecology Progress Series 80:159–173.[ISI]
Sandström O. (1980) Selective feeding by Baltic herring. Hydrobiologia 69:199–207.[CrossRef][ISI]
Sokal R.R. and Rohlf F.J. (1995) Biometry 3rd edn (W.H. Freeman and Company, New York) 859 pp.
S-Plus F.J. (2000) S-Plus 2000 Software(Statistical Sciences, Washington, Seattle).
Statistica F.J. (1995) Statistica for Windows. Statsoft, Inc. and its licensors, 1984–1996, Tulsa, Oklahoma.
Szypula J. (1992) Feeding of herring and sprat in the southern Baltic over 1971–1989. Acta Ichthyologica et Piscatoria 22:181–196.
Szypula J., Ostrowski J., Margo
ski P., Krajeska-Soltys A. (1997) Food of Baltic herring and sprat in the years 1995–1996 in light of the availability of components. Bulletin of the Sea Fisheries Institute Gdynia 141:19–31.
Szypula J., Wlodzimierz G., Wyszyski M. (1997) Feeding of Baltic herring and sprat in the period 1986–1996 in relation to their state and biomass. Bulletin of the Sea Fisheries Institute Gdynia 142:73–83.
Viitasalo M., Flinkman J., Viherluoto M. (2001) Zooplanktivory in the Baltic Sea: a comparison of prey selectivity by Clupea harengus and Mysis mixta, with reference to prey escape reactions. Marine Ecology Progress Series 216:191–200.[ISI]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Peltonen, M. Luoto, J.-P. Paakkonen, M. Karjalainen, A. Tuomaala, J. Ponni, and M. Viitasalo Pelagic fish abundance in relation to regional environmental variation in the Gulf of Finland, northern Baltic Sea ICES J. Mar. Sci., April 1, 2007; 64(3): 487 - 495. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||