© 2005 International Council for the Exploration of the Sea
On the phenology of North Sea ichthyoplankton
a German Centre for Marine Biodiversity Research (FIS) and Federal Agency for Shipping and Hydrography of Germany (BSH) Notkestrasse 85, D-22607 Hamburg, Germany
b German Centre for Marine Biodiversity Research (FIS) Notkestr. 85, D-22607 Hamburg, Germany
c Federal Maritime and Hydrographic Agency Notkestrasse 85, D-22607 Hamburg, Germany
*Correspondence to W. Greve: tel: +49 40 89981870; fax: +49 40 89981871. e-mail: wgreve{at}meeresforschung.de.
The timing of fish larvae abundance was investigated at Helgoland Roads in the southern North Sea for the years 1990, 1993–1996, 1998, and 1999. From the 44 species identified, 27 were sufficiently abundant in more than 3 years and had well defined seasonal changes to calculate their phenology. The passage of the 15%, 50%, and 85% annual cumulative abundance thresholds was used to define the "start of season", "middle of season", and "end of season", respectively. The timing of a significant number of these events was correlated with the mean annual winter sea surface temperatures (SSTs) in weeks 1–10 with higher temperatures leading to earlier appearance. The timing of the end of season was also negatively correlated with the SSTs. Sole (Solea solea) is presented as an example of the general functional relationship.
Keywords: climatic forcing, fish larvae, North Sea, phenology, seasonality, sole, temperature, time-series
Received 13 August 2004; accepted 1 March 2005.
 |
Introduction |
|---|
 Top Introduction MethodsResultsDiscussionReferences |
|---|
It is believed that global warming will change the latitudinal, altitudinal, and temporal distribution of populations (Parmesan and Johe, 2003). Though generally sparsely investigated, a northward displacement of marine organisms has been observed (Southward et al., 1995), and a seasonal shift to earlier appearance of fish larvae has been described (Greve et al., 1996). Predictions of the seasonal timing of zooplankton populations on the species level can be made from temperature (Greve et al., 2001), and inter-decadal phenological trends for North Atlantic phytoplankton and zooplankton have been found (Edwards and Richardson, 2004).
The scientific discipline treating meteorological influences on biological processes is biometeorology (Weihe, 1997). At least two of its branches are concerned with marine ecology, i.e. climatic biogeography, which treats the impact of meteorology on the regional distribution of organisms; and phenology, which treats the impact of weather and climate on the temporal orientation of ontogenetic events (Schwartz, 2003). Thus, phenology is the study of the timing of recurrent biological processes. In temperate regions, the seasonal timing of population development varies from year to year within a limited time frame. Phenology was developed in terrestrial ecology, especially for agricultural applications. The timing of sprouting, blossoming, and harvesting permits the optimization of fertilization, pest control, and other management activities. Indeed, observation systems have been installed in many countries and by international bodies to monitor the phenology of plants and several animals (Schwartz, 2003).
Global warming has led to the expansion of the study of phenology (Sparks and Menzel, 2002; Walther et al., 2002). Though most examples are drawn from terrestrial systems, marine biota deserve equal attention (Rodwell and Rodwell, 1999; Marshall et al., 2001), if for no other reason than the commercial importance of fish. The year-class size of marine fish is greatly influenced by the timing of spawning and the resulting match–mismatch with their prey and predators (Cushing, 1990). This was confirmed recently on the basis of satellite remote sensing and a long-term data set of haddock recruitment (Platt et al., 2003).
A common obstacle limiting marine phenological studies is the requirement of long time-series of high frequency (at least weekly) observations, of which there are few in the ocean because of the considerable effort required. One such zooplankton time-series, including ichthyoplankton, has been collected near the only offshore island in the North Sea, Helgoland (Figure 1). This paper explores the changes in phenology of the ichthyoplankton using the Helgoland data set.
|
|
Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Since April 1974, the marine station Biologische Anstalt in Helgoland has been collecting macrozooplankton samples three times a week at Helgoland Roads (positioned about 50 km off the coast between the two small islands of Helgoland at 54°11'3''N 7°54'0''E) (Greve et al., 2004). Tidal currents, approaching from a depth of 50 m to the south or 30 m to the north, mix the water in the shallow, 6 m deep passage between the islands where sampling is possible under almost any wind condition. Mean residual currents pass along the German Bight counterclockwise. The river plumes of the Weser and Elbe reach as far offshore as Helgoland, so samples taken can be representative of coastal, estuarine, or stratified central North Sea water.
Macrozooplankton sampling was carried out with a 500-µm mesh Hydrobios CalCOFI net, using oblique hauls filtering approximately 100 m3 of water based upon flowmeter readings. Samples were collected every Monday, Wednesday, and Friday. They were preserved in 4% formaldehyde, stored, and processed later to determine the abundance of meso- and macrozooplankton (Greve, 2004). Since 1993, fish larvae have been sorted, identified and counted, and then combined into a single weekly sample. Prior to this, only the sum of the total ichthyoplankton was recorded. Following an initial analysis using the years 1993–1995, it appeared that thermal control might affect the timing of ichthyoplankton occurrences. Therefore, stored samples from the extremely cold winter of 1996 and the extremely warm winter of 1990 (Figure 2) were investigated along with 1998 and 1999 samples. Inclusion of more years has been prevented by limited resources for taxonomic discrimination of the fish larvae within the routine zooplankton work, although the stored samples are available for later processing.
|
In all 44 species were identified over the 7 years (Table 1). From the weekly abundance measurements, a cumulative sum of values was calculated. The annual sum was set as 100%, and the week of the passage of the 15%, 50%, and 85% cumulative abundance levels for each year was determined (Figure 3). These levels were defined as the "start of season", "middle of season", and "end of season", respectively. These dates were then correlated with the mean annual sea surface temperature (SST) for weeks 1–10, calculated from measurements at the Biologische Anstalt in Helgoland.
|
|
Standard linear regressions and correlation coefficients were calculated using Microsoft Excel with a significance level of p
0.05. |
Results |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Variability among the 44 species of ichthyoplankton was large. The number of samples in which there was at least one larva over the 7 years ranged from 191 (Limanda limanda) to 1 (e.g. Trisopterus luscus) (Table 1). Maximum abundance varied from 25.65 individuals m–3 (Ammodytes marinus) to 0.005 individuals m–3 (Trisopterus luscus). Mean weekly abundance varied from 2.096 individuals m–3 (Ammodytes marinus) to 0.005 individuals m–3 (Trisopterus luscus).
The common sole, Solea solea, because of its large interannual variability (Figure 4), demonstrated why relative cumulative abundance thresholds were used instead of absolute abundance values to define the phenophases (see Figure 3). There are also extreme interannual differences in the seasonal timing for Solea solea, e.g. the "end of season" occurred in week 22 during the warm year of 1990, while there was a late "start of season" in week 27 during the cold year of 1996. Note that SST accounts for 76% of the variance in the timing of the "middle of season" for this species (Figure 5, Table 1).
|
|
The duration of occurrence also varies from species to species. The maximum number of weeks of the year during which a species was observed was 48 for Atlantic herring, Clupea harengus, representing an extended reproductive phase. This is believed to be due to the location of several spawning sites within or near Helgoland with different spawning times (Greve et al., 2004). Another species with several annual spawning periods is Ammodytes tobianus. In such cases, the methodology for the determination of phenophases and phenological temperature correlations was assumed to be unreasonable and was not applied. However, in the case of species with more pronounced larval seasonal distribution, the calculation of the regression of the phenophases "start of season", "middle of season", and "end of season" with the winter SSTs was considered reasonable.
The statistical reliability of any regression relationship generally increases with the number of observations (i.e. years of data). Four years was chosen as the minimum number of data points in this study before undertaking a regression analysis between temperature and ontology, resulting in a total of 29 species with between 4 and 7 years of data. Two of these species (Clupea harengus, Ammodytes tobianus) were deleted because of their extended spawning times, as discussed above. This left 27 species for which regression analyses were calculated. In 11 of the 27 cases, the "middle of season" regressions against temperature were statistically significant. For both the "start of season" and "end of season", there were 10 significant cases (Table 1). Significance levels were based upon n – 1 degrees of freedom, where n varied from 4 to 7. The range of the variance in the seasonal distribution of the 24 most abundant species is given in Table 2.
|
The slope of the regression line varied depending upon the seasonal state (Figure 6) and from species to species (Figure 7). In some populations, as in the pelagic species Scomber scombrus and Sprattus sprattus, no influence of winter temperatures on the phenology was found, but in other pelagic species, e.g., Trachurus trachurus, there was a strong correlation (Table 1). The more demersal populations of Arnoglossus laterna, Ctenolabrus rupestris, Gadus morhua, Hyperoplus lanceolatus, Taurulus lilljeborgi, Platichthys flesus, Pleuronectes platessa, Solea solea, and Taurulus bubalis display a generally strong negative correlation with winter SSTs.
|
|
The total number of correlations undertaken between temperature and either the start, middle, or end of the season for the 27 species was 81. In 76 of these cases, the sign of the correlation was negative, providing further evidence of the significance of temperature.
|
Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
This study found that the timing of the occurrence of fish larvae is influenced by the SSTs the preceding winter, as had been determined for other zooplankton (Greve et al., 2001) and anticipated by Lange and Greve (1997). The phenophases, i.e. "start of season", "middle of season", and "end of season" defined by the 15%, 50%, and 85% cumulated abundance by numbers, respectively, are generally negatively correlated with the mean SST in weeks 1–10 in the North Sea. An extension of the spawning season indicated by a positive correlation with the "end of season", as observed in terrestrial systems (Parmesan and Johe, 2003), was not seen. Such an extension might occur in the case of marine organisms with short generation times, i.e. the possibility of two or more a year. We expect that the mechanism linking temperature to the timing of fish larval abundance is through faster gonadal development with increasing temperatures. If true, the monitoring of the gonadal state of the fish in fish markets could provide another and possibly more precise phenophase observation.
Sagarin (2001) warned against possible false conclusions concerning the advance of spring being attributed to climate. He calculated a precession of the vernal equinox of 0.78 days in 100 years on the basis of the conventions of our calendar. This trend towards earlier springs should be subtracted from observed phenological trends, but it is small relative to our observed changes and therefore was neglected.
The annual cumulative abundance-related determination of phenophases as used within the present study is suitable for populations with seasonal spawning but not for species with extended spawning periods or with several spawning sites near the sampling area that might have different spawning times.
Besides temperature, ichthyoplankton abundance is influenced by parental abundance and their nutrition, as well as by the mortality of the eggs and larvae themselves, but none of these factors were considered in this study. In spite of this, our results indicate that predictions of the timing of the abundance of fish larvae can be made on the basis of SSTs for a period of up to half a year ahead and with an accuracy of a single week for the "middle of season" in some species. As predictions on the seasonality of other zooplankton that are potential prey or predators of ichthyoplankton have reached a similar level of accuracy (Greve et al., 2001), match–mismatch projections and hence year-class recruitment prognoses may become possible.
Interestingly, the phenology of phytoplankton at Helgoland Roads does not follow the negative correlation with temperature described for fish larvae. The spring bloom has been delayed in spite of a warming by 1.1°C in the period from 1962 to 2004 (Wiltshire and Manly, 2004). The increased winter abundance of copepods and phytoplankton grazing, consistent with the phenological trend observed in zooplankton, was presented as possibly influencing the delay in the phytoplankton bloom.
Improved knowledge of the sensitive early life history periods of fish and the species-specific annual development could be used as protective measures for fish populations through possible short-term reduction of the effort of commercial fisheries during spawning.
Improved predictions by prognostic models are a major objective of research. In the case of the seasonality of fish larvae, our results narrow the band of possible appearance of many species in the southern North Sea by a substantial degree. Instead of a time frame of months as displayed by FishBase (Froese and Pauly, 2003), it is possible to focus almost on a single week as the period of most probable start of season, middle of season, and end of season. Of course, these results are based on only 7 years of data, and the investigations must extend over longer periods, more locations, and further populations to ensure reliable predictive models of the responses to climate change. Also, the methodology could be improved by including an optimization procedure for detecting the phenophases. Figure 3 shows the possible uncertainties of the temporal determinations, e.g. of the start of season, which could be shifted easily by 7 weeks by minor abundance changes. Such uncertainties become less important when based on longer sampling periods.
It is anticipated that global warming will change the latitudinal, altitudinal, and temporal distribution of marine populations just as it does with terrestrial populations. To better define the phenological functional relationships of marine species, two avenues are promising. First is the determination of the physiological thermal preference zone of each species (Uhlig and Sahling, 1995) representing this functional relationship within its ecological niche (Elton, 1927). This can be determined by laboratory experiments or, it is to be hoped in the future, by genetic analysis as a molecular trigger of functional biodiversity. Second is the traditional way as practised successfully in terrestrial ecology with the development of an observation system for all major regions. Weekly phenological observations for a decade would provide the basis for marine biometeorology. As volunteer observers carry the load for terrestrial observations, we expect to find sufficient potential volunteers dwelling near the seacoast to support biometeorological marine research and phenological studies, in particular by helping to observe changes in sea life that accompany globally changing temperatures.
The aim of the international GLOBEC programme is to advance our understanding of the structure and functioning of the global ocean ecosystem, its major subsystems, and its response to physical forcing so that a capability to forecast the responses of the marine ecosystem to global change can be developed. Biometeorology, with its disciplines phenology, thermal biology, and marine biogeography, treats such responses to global change. Generally, the field has been neglected in marine ecosystem studies until now, but offers an excellent way of monitoring the effects of climate change and as such needs to be encouraged.
|
Acknowledgements |
|---|
Time-series useful to biometeorology depend on the support of the participating institutions from the Biologische Anstalt Helgoland (now with the Foundation-Alfred-Wegener Institute), and the Senckenberg Research Institute and the Federal Agency for Shipping and Hydrography of Germany (BSH). The sailors and assistants of these research institutions merit our gratitude for their continuous, thorough efforts. Two unknown reviewers and Ken Drinkwater kindly helped us in the preparation of the manuscript.
|
References |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
-
Cushing D.H. (1990) Plankton production and year-class strength in fish populations: an update of the match/mismatch hypothesis. Advances in Marine Biology 26:1–122.[Web of Science]
Edwards M.A. and Richardson A.J. (2004) Impact of climate change on marine pelagic phenology and trophic mismatch. Nature 430:881–884.[CrossRef][Medline]
Elton C. (1927) Animal Ecology(Methuen, London) 207 pp.
Froese R. and Pauly D. (2003) FishBase, Kiel University World Wide Web electronic publication.
Greve W. (2004) Biometeorology of North Sea appendicularians. In Gorsky G., Youngbluth M., Deibel D. (Eds.). Response of Marine Ecosystems to Global Change: Ecological Impact of AppendiculariansGordon and Breach, Scientific Publishers pp. 283–296.
Greve W., Lange U., Reiners F., Nast J. (2001) Predicting the Seasonality of North Sea Zooplankton. In Kröncke I., Türkay M., Sündermann J. (Eds.). Burning Issues of North Sea Ecology, Proceedings of the 14th International Senckenberg Conference North Sea 2000 31:2 pp. 263–268 Senckenbergiana Maritima.
Greve W., Reiners F., Nast J. (1996) Biocoenotic changes of the zooplankton in German Bight: the possible effects of eutrophication and climate. ICES Journal of Marine Science 53:951–956.
Greve W., Reiners F., Nast J., Hoffmann S. (2004) Helgoland Roads time-series meso- and macrozooplankton 1975 to 2004: lessons from 30 years of single spot high frequency sampling at the only off-shore island of the North Sea. Helgoland Marine Research 58:274–288.[CrossRef][Web of Science]
Lange U. and Greve W. (1997) Does temperature influence the spawning time, recruitment and distribution of flatfish via its influence on the rate of gonadal maturation? Deutsche Hydrographische Zeitschrift 49:251–263.
Marshall J., Kushnir Y., Battisti D., Chang P., Czaja A., Dickson R., Hurrell J., McCartney M., Saravanan R., Visbeck M. (2001) North Atlantic climate variability: phenomena, impacts and mechanisms. International Journal of Climatology 21:1863–1898.[CrossRef][Web of Science]
Parmesan C.A. and Johe G. (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421:37–42.[CrossRef][Medline]
Platt T., Fuentes-Yaco C., Frank K.T. (2003) Marine ecology: spring algal bloom and larval fish survival. Nature 423:398–399.[Medline]
Rodwell M.J. and Rodwell D.P. (1999) Oceanic forcing of the wintertime North Atlantic Oscillation and European climate. Nature 398:320–323.[CrossRef]
Sagarin R. (2001) False estimates of the advance of spring. Nature 414:600–601.[Medline]
Schwartz M.D. (2003) Phenology: An Integrative Environmental Science(Kluwer, Netherlands) 564 pp.
Southward A.J., Hawkins S.J., Burrows M.T. (1995) 70 years observations of changes in distribution and abundance of zooplankton and intertidal organisms in the western English Channel in relation to rising sea temperature. Journal of Thermal Biology 1–2:127–155.
Sparks T.H.A. and Menzel A. (2002) Observed changes in seasons: an overview. International Journal of Climatology 22:1715–1725.[CrossRef][Web of Science]
Uhlig G.U. and Sahling G. (1995) Noctiluca scintillans: zeitliche Verteilung bei Helgoland und räumliche Verbreitng in der Deutschen Bucht (Langzeitreihen 1970–1993). Berichte der Biologischen Anstalt Helgoland 9:1–127 (In German).
Walther G-R., Post E., Convey P., Menzel A., Parmesan C., Beebee T.J.C., Fromentin J-M., Guldberg O.H., Bairlein F. (2002) Ecological responses to recent climate change. Nature 416:389–395.[CrossRef][Medline]
Weihe W.H. (1997) Review of the history of Biometeorology. International Journal of Biometeorology 40:19–15.[Web of Science]
Wiltshire K.H.A. and Manly B.F.J. (2004) The warming trend at Helgoland Roads, North Sea: phytoplankton response. Helgoland Marine Research 58:269–273.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Genner, N. C. Halliday, S. D. Simpson, A. J. Southward, S. J. Hawkins, and D. W. Sims Temperature-driven phenological changes within a marine larval fish assemblage J. Plankton Res., September 9, 2009; (2009) fbp082v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Pitois, M. Shaw, C. J. Fox, and C. L. J. Frid A new fine-mesh zooplankton time series from the Dove sampling station (North Sea) J. Plankton Res., March 1, 2009; 31(3): 337 - 343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Richardson In hot water: zooplankton and climate change ICES J. Mar. Sci., April 1, 2008; 65(3): 279 - 295. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||