ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on May 21, 2008
ICES Journal of Marine Science: Journal du Conseil 2008 65(6):906-911; doi:10.1093/icesjms/fsn074
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Indications of a negative impact of herring on recruitment of Norway pout
Institute of Marine Research, PO Box 1870 Nordnes, N-5817 Bergen, Norway
Correspondence to G. Huse: tel: +47 55 236988; fax: +47 55 238687; e-mail: geir.huse{at}imr.no
Huse, G., Salthaug, A., and Skogen, M. D. 2008. Indications of a negative impact of herring on recruitment of Norway pout. – ICES Journal of Marine Science, 65: 906–911.The Norway pout (Trisopterus esmarkii) stock in the North Sea has experienced poor recruitment recently. Herring (Clupea harengus) has been suggested to be a major predator on fish larvae in the North Sea. We investigated possible interactions between herring and Norway pout using a simple statistical analysis and a modified stock–recruit relationship. There was a significant negative relationship (linear regression, r = –0.44, p < 0.05) between total herring biomass and recruitment of Norway pout. The spawning stock of Norway pout is typically dominated by 2-year-olds, and there was a strong negative relationship (linear regression, r = –0.79, p < 0.01) between herring biomass and Norway pout spawning-stock biomass (SSB) 2 years later. A Beverton–Holt model fitted to stock–recruit data of Norway pout produced a rather poor correlation (r2 = 0.04). However, when only the Norway pout SSB not overlapping with herring is considered, the fit between the model and the stock–recruit data improves (r2 = 0.31). The analyses indicate a negative impact by herring on recruitment of Norway pout, the most plausible cause for this being herring predation on Norway pout larvae, but field studies are needed to verify such predation.
Keywords: herring, North Sea, Norway pout, predation, stock–recruit relationship
Received 16 August 2007; accepted 27 March 2008; advance access publication 21 May 2008.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
The mortality of marine fish larvae is generally high and variable in space and time (Hjort, 1926; McGurk, 1986; Chambers and Trippel, 1997). Several hypotheses have been advanced to explain heavy larval mortality (reviewed by Chambers and Trippel, 1997), including predation by planktivorous fish owing to their potentially high densities and efficient foraging on fish larvae (Bailey and Houde, 1989). It has been hypothesized that predation by juvenile Norwegian spring-spawning (NSS) herring (Clupea harengus) on Barents Sea capelin (Mallotus villosus) larvae is the cause for the poor recruitment of the capelin stock in some years (Hamre, 1994). This hypothesis is supported by observations of negative correlations between herring abundance in the Barents Sea and capelin recruitment (Gjøsæter and Bogstad, 1998), and by observations of capelin larvae in herring stomachs (Huse and Toresen, 2000; Godiksen et al., 2006). Three recent collapses in Barents Sea capelin have been attributed to herring predation (Gjøsæter, 1998; Gjøsæter and Bogstad, 1998; ICES, 2006).
Recruitment of many fish stocks in the North Sea (Figure 1) has been exceptionally poor recently (ICES, 2007b), and this has led to a re-examination of the hypotheses about different factors affecting fish recruitment. A negative relationship between pelagic fish abundance and recruitment of demersal fish has been suggested for the North Sea, specifically because the so-called "gadoid outburst" during the 1960s coincided roughly with a collapse in North Sea herring and mackerel stocks (Cushing, 1980; Daan et al., 1985, 1994). However, the timing of various events in the 1950s and 1960s does not fully support a negative relationship between biomass levels of pelagic fish and recruitment of gadoids (Hislop, 1996). On the other hand, the recent poor recruitment to many North Sea stocks has coincided with a large herring stock, which again raises the question of predatory interactions. The picture is now even more complicated because the spawning-stock biomass (SSB) of many stocks is very low, which obviously should have a negative effect on recruitment (Myers and Barrowman, 1996). However, the stock of Norway pout (Trisopterus esmarkii) in the North Sea has experienced several years of poor recruitment, despite its initially large spawning stock (ICES, 2007b). This could be due to predation by herring, because there is potential for spatial overlap between the two stocks, although there is no information available on stomach content analysis to suggest such an interaction.
|
Verifying that predation takes place and has the magnitude needed to induce high rates of mortality is costly, because it entails extensive use of ship time and manpower. Moreover, it is difficult to detect predation of fish larvae by fish because of the rapid digestion of larval tissue (Folkvord, 1993). It is therefore valuable to utilize existing data to look for signs of predatory interactions between planktivorous fish and larvae. As spatial overlap is a prerequisite for a predatory interaction, maps of spatial distributions of species can provide information about the potential for predatory effects on recruitment. However, there is no standardized means of incorporating spatial overlap information into stock–recruit models. Gjøsæter and Bogstad (1998) showed that excluding years with the presence of herring in the Barents Sea from the analysis substantially improved the fit between the stock–recruit model and data for the Barents Sea capelin. Similarly, Fiksen and Slotte (2002) showed that the fit between the stock–recruit model and observations for NSS herring was improved by including a temperature term. Such examples illustrate the potential benefit that lies in modifying stock–recruit relationships to understand better and quantify the dynamics of fish stocks.
The distribution of Norway pout in the North Sea is indicated in Figure 1. The species is preyed upon by cod (Gadus morhua), whiting (Merlangius merlangus), saithe (Pollachius virens) and other fish in the North Sea (ICES, 1997). Herring are generally widespread, and there is considerable potential for overlap with Norway pout. The objectives of this study were to investigate whether there are patterns in spatial and population dynamics of Norway pout and herring in the North Sea that indicate interactions between the two stocks. We initially look for correlative evidence of predatory interactions, then estimate the extent of spatial overlap. Finally, we investigate the potential for using spatial information to improve the stock–recruit relationship of Norway pout. These analyses are complemented by correlations between prey abundance and abundance of juvenile Norway pout.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
For herring, estimated total-stock biomass was taken from the final assessment in the Report of the Herring Assessment Working Group for the Area South of 62°N (ICES, 2007a). For Norway pout, estimates of SSB, recruitment (numbers of fish aged 0 in quarter 3), proportion mature-at-age, mean weights-at-age, and stock numbers-at-age in the North Sea and Skagerrak were taken from ICES (2007c; the working group report that was the basis for the ICES advice in spring 2007).
The spatial distributions of herring and Norway pout were estimated based on abundance indices per ICES rectangle, derived from the quarter 1 International Bottom Trawl Survey (IBTS) for the period 1982–2006. The IBTS data are archived and updated by ICES (see ICES database http://www.ices.dk/datacentre/datras/datras.asp). The survey takes place in February just before the peak in hatching of Norway pout, in early April (Heath, 2007), and is therefore relevant to estimating overlap between Norway pout eggs and larvae and potential predators. To include the herring distribution in the Norway pout stock–recruit relationship, we estimated the SSB of Norway pout occupying ICES rectangles with no herring for each year. This was done by spatially resolving the estimates of Norway pout abundance based on virtual population analysis (VPA) to obtain a spatial SSB distribution, then summing the SSB from ICES rectangles without herring (age 1 and older):
|
| (1) |
|
| (2) |
If the entire Norway pout spawning stock overlaps with herring, SSBfree will be 0 and the overlap will be 1. Conversely, if there are no herring in the rectangles with Norway pout SSB, SSBfree will be equal to SSB and the overlap will be 0.
A Beverton–Holt (Beverton and Holt, 1957) model was fitted to the SSB and recruitment data for Norway pout. The Beverton–Holt spawner–recruit curve relates recruitment (R) to SSB (S) using the equation:
|
| (3) |
|
Although this work focuses on potential predation effects (i.e. top–down), limitation in prey availability (i.e. bottom–up effects) is often regarded as a key factor in determining the rate of survival of fish larvae (Cushing, 1990; Chambers and Trippel, 1997). In the North Sea, the copepod Calanus finmarchicus is the main prey of fish larvae that hatch during spring (Sundby, 2000). We therefore explored the relationship between the estimated relative abundance of C. finmarchicus collected by the Continuous Plankton Recorder (CPR) and recruitment and survival (recruitment per SSB) of Norway pout. Monthly CPR indices of total C. finmarchicus abundance (average number of individuals per sample per month) from the geographical area 57–0°N, 1°W–6°E from January to June of the period 1984–2004 were used. These data were kindly provided by the Sir Alistair Hardy Foundation for Ocean Science (SAHFOS). Monthly indices and indices averaged over longer periods (e.g. January–March) were investigated.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
There was a significant, but not very strong, negative relationship between Norway pout recruitment and total-stock biomass (TSB) of herring (linear regression = –0.033 x + 17 432, r = –0.44, p < 0.05; Figure 2a and b). The overlap between Norway pout and herring varied substantially over time (Figure 2c), and there was a significant negative relationship between the overlap and Norway pout recruitment (linear regression = –135.44x + 189.56, r = –0.52, p < 0.01). The variation in overlap is illustrated for a year with low degree of overlap in Figure 3a, and for one with a high degree of overlap in Figure 3b. The pattern between herring and Norway pout abundance is also seen with time-lags, and peaks between herring TSB and number of 2-year-old Norway pout 2 years later (linear regression = –144.2 x + 4 x 106, r = –0.68, p < 0.01). Moreover, the SSB of Norway pout is dominated by 2-year-olds, which typically constitute 60–80% of the biomass, and there is a strong negative relationship between herring TSB and SSB of Norway pout 2 years later (linear regression = –9 x 10–5 x + 409.4, r = –0.79, p < 0.01; Figure 4).
|
|
|
The Beverton–Holt model did not fit the stock–recruit data for Norway pout well (r2 = 0.04; Figure 5). However, when only the part of the Norway pout SSB not overlapping with herring (SSBfree) is included (see Figure 3), the fit of the model to the stock–recruit data improved substantially (r2 = 0.31; Figure 5). The SSBfree stock–recruit relationship in Figure 5 is influenced by 3 years with very high SSBfree. When these points are removed, the Beverton–Holt model maintains a good fit to the data (r2 = 0.25).
|
Calanus finmarchicus abundance estimated by the CPR was generally poorly correlated (p > 0.1) to Norway pout recruitment and survival, on both a natural and a log–log scale. The only exception was a significant positive relationship between C. finmarchicus abundance in January and subsequent Norway pout recruitment (Figure 6; linear regression = 16.323 x + 61.691, r = 0.47, p < 0.05). For log–log scale, the relationship was slightly weaker.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Our results demonstrate that taking into account the distribution of a suspected predator can improve the stock–recruit relationship for Norway pout substantially. Further, our analysis strengthens the suspicion that herring might influence recruitment of Norway pout. This was indicated both by a simple correlation analysis of herring biomass against Norway pout recruitment and through a modified stock–recruit relationship. Moreover, there was a strong negative correlation between the total biomass of herring and Norway pout SSB 2 years later. The analyses were based on partially independent data, the IBTS time-series and stock assessment estimates. The herring stock assessment utilizes catch per unit effort from the IBTS time-series to tune the VPA for the younger age groups, but the rest of the assessment is driven by landings data and other tuning series. This independence in the data used for our correlation and stock–recruit analyses strengthens the hypothesis that Norway pout recruitment can be impeded by herring. Estimates of recruitment and SSB of Norway pout rely on the VPA analysis and are therefore not completely independent.
In principle, the negative influence of herring on recruitment of Norway pout can be explained in terms of both food competition and predation (Hislop, 1996). Although herring and Norway pout larvae eat similar planktonic food, it seems rather unlikely that competition would result in the type of inverse pattern found here, because plankton availability will depend on a number of other factors, such as the annual productivity and the abundance of other food competitors. Predation by herring on Norway pout larvae, on the other hand, is a plausible explanation for the inverse pattern in population dynamics we have seen. Frederiksen et al. (2007) hypothesized that herring predation is important in regulating sandeel abundance around the Shetlands, based on negative correlations between local herring and sandeel abundance. Field studies have shown that herring feed on fish eggs (Daan et al., 1985; Ellis and Nash, 1997; Segers et al., 2007) and can be a major predator of fish larvae (Huse and Toresen, 2000; Godiksen et al., 2006; Hallfredsson et al., 2007), but further studies in the North Sea in the areas of overlap between herring and Norway pout are needed to verify that the proposed predation actually takes place to a marked degree. Episodes of massive predation by fish on fish have been witnessed in some instances (Godiksen et al., 2006; Temming et al., 2007), and an episodic view of predation might be a fruitful paradigm to improve our understanding of fish recruitment and natural mortality of fish in general.
Although our results indicate that recruitment of Norway pout could be influenced by herring, other environmental factors could impact recruitment. Hislop (1996) tried to disentangle the causes of the gadoid outburst in the 1960s in a retrospective analysis. He found that no single factor was responsible for the increase in gadoid abundance during that period, when examining various effects related to climate, predators, and prey. The North Sea is a warmer, more diverse ecosystem than the Barents Sea, which appears to be more strongly affected by interannual climate variation both directly and indirectly through cascading climatic effects. There, herring recruitment is strong typically in warm years (Toresen and Østvedt, 2000; Fiksen and Slotte, 2002; Sætre et al., 2002), and results in herring being widespread in the southern Barents Sea, where they overlap with and prey on capelin larvae (Huse and Toresen, 2000; Godiksen et al., 2006). This interaction has resulted in three recent collapses in the Barents Sea capelin stock, in 1986, 1994, and 2003 (ICES, 2006). In the Barents Sea, an inflow of warm water gives a positive direct effect on growth and recruitment, but it is also associated with better inflow of Calanus-rich water masses from the Norwegian Sea (Skjoldal et al., 1992). This results in a strong positive association between temperature and recruitment being seen for many stocks in the Barents Sea (Sætersdal and Loeng, 1987; Ottersen et al., 1994; Sundby, 2000). In the North Sea, inflow of water rich in C. finmarchicus from the cold Norwegian Sea is associated with low temperature (Reid et al., 2003). Bromley et al. (1997) found that C. finmarchicus made up the most of the diet of 0-group Norway pout. There has been a major change in the composition of copepods in the North Sea during the past few decades (Planque and Fromentin, 1996). The decrease in Calanus abundance and shift from C. finmarchicus to C. helgolandicus has therefore been used to explain the poor recruitment of North Sea fish stocks in recent years (ICES, 2007b, c). We investigated the relationship between Norway pout recruitment and abundance of C. finmarchicus by using the data from the CPR from the northern North Sea. The only significant relationship we found was between Norway pout recruitment and C. finmarchicus abundance in January (Figure 6). However, the relationship was not really strong and only provided modest indications of bottom–up control of Norway pout recruitment.
The stock–recruit relationship for Norway pout improved substantially when the SSB was modified by taking the herring distribution into account. Similar analyses can easily be made for other stocks, and might provide insight into stock dynamics. Additionally, they can be useful for predicting recruitment as part of the fishery management advisory process. The SSBfree data fit much better to the Beverton–Holt model, and give a much steeper curve and higher asymptotic level of Norway pout recruitment, indicating the increased productivity of the stock in a "herring-free" environment. The overlap measure used in our analysis was rather coarse and static, and did not take into account the abundance of herring in the squares, or migrations of herring during spring and the drift of Norway pout larvae. However, the first-quarter IBTS survey covers the spawning distribution of Norway pout fairly well in space and time, and is therefore suitable for studying the overlap between Norway pout larvae and potential predators in the absence of a direct survey targeting Norway pout larvae. A more dynamic approach would be to move the Norway pout larvae using a hydrodynamic model, and simulate the herring movements. However, this would require a substantial increase in effort and model complexity.
Our analyses indicate that herring have a negative effect on recruitment of Norway pout. The most plausible cause for this is herring predation on Norway pout larvae, but field studies are needed to verify this hypothesis. We note that the biomass of herring in the North Sea is currently declining, so we anticipate that this might lead to improved conditions for Norway pout soon.
|
Acknowledgements |
|---|
We thank Geir Ottersen and two anonymous referees for valuable comments on an earlier version of the manuscript. The Sir Alistair Hardy Foundation for Ocean Science kindly provided the Calanus data. The work was supported by the Research Council of Norway through the projects RECNOR and INFERNO.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Bailey K. M., Houde E. D. Predation on the eggs and larvae of marine fishes and the recruitment problem. Advances in Marine Biology (1989) 25:1–83.[Web of Science]
Beverton R. J. H., Holt S. J. On the dynamics of exploited fish populations. Ministry of Agriculture, Fisheries and Food. In: Fishery Investigations, London, Series II, 19 (1957) 533.
Bromley P. J., Watson T., Hislop J. R. G. Diel feeding patterns and the development of food webs in pelagic 0-group cod (Gadus morhua L.), haddock (Melanogrammus aeglefinus L.), whiting (Merlangius merlangus L.), saithe (Pollachius virens L.), and Norway pout (Trisopterus esmarkii Nilsson) in the northern North Sea. ICES Journal of Marine Science (1997) 54:846–853.
Chambers R. C., Trippel E. A. Early Life History and Recruitment in Fish Populations. (1997) London: Chapman and Hall. 632.
Cushing D. H. The decline of the herring stocks and the gadoid outburst. Journal du Conseil International pour l’Exploration de la Mer (1980) 39:70–81.
Cushing D. H. Plankton production and year-class strength in fish populations—an update of the match–mismatch hypothesis. Advances in Marine Biology (1990) 26:249–293.[Web of Science]
Daan N., Heessen H. J. L., Pope J. G. Changes in the North Sea cod stock during the twentieth century. ICES Marine Science Symposia (1994) 198:228–243.
Daan N., Rijnsdorp A. D., Vanoverbeeke G. R. Predation by North Sea herring Clupea harengus on eggs of plaice Pleuronectes platessa and cod Gadus morhua. Transactions of the American Fisheries Society (1985) 114:499–506.[CrossRef]
Ellis T., Nash R. D. M. Predation by sprat and herring on pelagic fish eggs in a plaice spawning area in the Irish Sea. Journal of Fish Biology (1997) 50:1195–1202.[CrossRef][Web of Science]
Fiksen O., Slotte A. Stock–environment recruitment models for Norwegian spring spawning herring (Clupea harengus). Canadian Journal of Fisheries and Aquatic Sciences (2002) 59:211–217.
Folkvord A. Prey recognition in stomachs of cannibalistic juvenile cod (Gadus morhua L). Sarsia (1993) 78:97–100.[Web of Science]
Frederiksen M., Furness R. W., Wanless S. Regional variation in the role of bottom–up and top–down processes in controlling sandeel abundance in the North Sea. Marine Ecology Progress Series (2007) 337:279–286.[CrossRef][Web of Science]
Gjøsæter H. The population biology and exploitation of capelin (Mallotus villosus) in the Barents Sea. Sarsia (1998) 83:453–496.[Web of Science]
Gjøsæter H., Bogstad B. Effects of the presence of herring (Clupea harengus) on the stock-recruitment relationship of Barents Sea capelin (Mallotus villosus). Fisheries Research (1998) 38:57–71.[CrossRef][Web of Science]
Godiksen J. A., Hallfredsson E. H., Pedersen T. Effects of alternative prey on predation intensity from herring Clupea harengus and sandeel Ammodytes marinus on capelin Mallotus villosus larvae in the Barents Sea. Journal of Fish Biology (2006) 69:1807–1823.[CrossRef][Web of Science]
Hallfredsson E. H., Pedersen T., Praebel K. Estimation of digestion rates for herring Clupea harengus L. feeding on fish larvae. Journal of Fish Biology (2007) 70:638–643.[CrossRef][Web of Science]
Hamre J. Biodiversity and the exploitation of the main fish stocks in the Norwegian–Barents Sea ecosystem. Biodiversity and Conservation (1994) 3:473–492.[CrossRef][Web of Science]
Heath M. R. The consumption of zooplankton by early life stages of fish in the North Sea. ICES Journal of Marine Science (2007) 64:1650–1663.
Hislop J. R. G. Changes in North Sea gadoid stocks. ICES Journal of Marine Science (1996) 53:1146–1156.
Hjort J. Fluctuations in the year classes of important food fishes. Journal du Conseil Permanent International pour l’Exploration de la Mer (1926) 1:5–38.
Huse G., Toresen R. Juvenile herring prey on Barents Sea capelin larvae. Sarsia (2000) 85:385–391.[Web of Science]
ICES. Report of the Multispecies Assessment Working Group. ICES Document CM 1997/Assess: 16 (1997).
ICES. Report of the Arctic Fisheries Working Group. ICES Document CM 2006/ACFM: 25 (2006) 594.
ICES. Report of the Herring Assessment Working Group South of 62N (HAWG). ICES Document CM 2007/ACFM: 11 (2007) a. 538.
ICES. Report of the Study Group on Recruitment Variability in North Sea Planktivorous Fish (SGRECVAP). ICES Document CM 2007/LRC: 07 (2007) b. 69.
ICES. Report of the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak (WGNSSK). ICES Document CM 2007/ACFM: 18 (2007) c. 700.
McGurk M. D. Natural mortality of marine pelagic fish eggs and larvae: role of spatial patchiness. Marine Ecology Progress Series (1986) 34:227–242.[CrossRef][Web of Science]
Myers R. A., Barrowman N. J. Is fish recruitment related to spawner abundance? Fishery Bulletin US (1996) 94:707–724.
Ottersen G., Loeng H., Raknes A. Influence of temperature variability on recruitment of cod in the Barents Sea. ICES Marine Science Symposia (1994) 198:471–481.
Planque B., Fromentin J. M. Calanus and environment in the eastern North Atlantic. 1. Spatial and temporal patterns of C. finmarchicus and C. helgolandicus. Marine Ecology Progress Series (1996) 134:101–109.[CrossRef][Web of Science]
Reid P. C., Edwards M., Beaugrand G., Skogen M., Stevens D. Periodic changes in the zooplankton of the North Sea during the twentieth century linked to oceanic inflow. Fisheries Oceanography (2003) 12:260–269.[CrossRef][Web of Science]
Segers F., Dickey-Collas M., Rijnsdorp A. D. Prey selection by North Sea herring (Clupea harengus), with special reference to fish eggs. ICES Journal of Marine Science (2007) 64:60–68.
Skjoldal H. R., Gjøsæter H., Loeng H. The Barents Sea ecosystem in the 1980s: ocean climate, plankton and capelin growth. ICES Marine Science Symposia (1992) 195:278–290.
Sundby S. Recruitment of Atlantic cod stocks in relation to temperature and advection of copepod populations. Sarsia (2000) 85:277–298.[Web of Science]
Sætersdal G., Loeng H. Ecological adaptation of reproduction in Northeast Arctic cod. Fisheries Research (1987) 5:253–270.[CrossRef][Web of Science]
Sætre R., Toresen R., Anker-Nilssen T. Factors affecting the recruitment variability of the Norwegian spring-spawning herring (Clupea harengus L.). ICES Journal of Marine Science (2002) 59:725–736.
Temming A., Floeter J., Ehrich S. Predation hot spots: large scale impact of local aggregations. Ecosystems (2007) 10:865–876.[CrossRef]
Toresen R., Østvedt O. J. Variation in abundance of Norwegian spring-spawning herring (Clupea harengus Clupeidae) throughout the 20th century and the influence of climatic fluctuations. Fish and Fisheries (2000) 1:231–256.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||