Helgoland Roads meso- and macrozooplankton time-series 1974 to 2004: lessons from 30 years of single spot, high frequency sampling at the only off-shore island of the North Sea
- First Online:
- Cite this article as:
- Greve, W., Reiners, F., Nast, J. et al. Helgol Mar Res (2004) 58: 274. doi:10.1007/s10152-004-0191-5
The Helgoland Roads meso- and macrozooplankton time-series 1974 to 2004 is a high frequency (every Monday, Wednesday and Friday), fixed position monitoring and research programme. The distance to the coastline reduces terrestrial and anthropological disturbances and permits the use of Helgoland Roads data as indicators of the surrounding German Bight plankton populations. The sampling, determination and IT methodologies are given, as well as examples of annual succession, and inter-annual population dynamics of resident and immigrant populations. Special attention is given to the phenology and seasonality of zooplankton populations. The influence of winter sea surface temperature on the seasonality of spawning of the common sole Solea solea is given as an example for merozooplankton populations.
A single observation of nature permits the definition of the momentary state. Repeated observations in space or time permit the recognition of distributional rules and variances. Local repetitions of observations produce time-series as a documentation of the system dynamics. Time-series enable the development of analytical models defining state variables and rates of processes, and testing hypothetical rules of change. The definition of the rules, and variances determining abundance distributions of populations and their changes, are the main subjects in ecosystem research. The degree of the understanding of such processes is expressed in analytical and operative models predicting changes in abundance distributions. The rules of change concern physicochemical forcing, physiological response mechanisms and trophodynamic processes. Significant time-scales can cover several orders of magnitude. Therefore, causal time-series analysis has to rely on data sets from various sources. While experimental time-series may operate informatively on less complex systems (e.g. Greve 1995a), in situ time-series are real and their documentation and analysis are most important (Perry et al. 2004). The Helgoland Roads station enables the continuous monitoring of marine processes in an offshore location with little terrestrial and anthropogenic disturbance and, due to the Biologische Anstalt Helgoland, affordable sampling capacity is available.
The Helgoland Roads time-series on meso- and macrozooplankton was started in 1974 as a spin-off data collection stemming from the daily live-food catches from the small cutters of the Biologische Anstalt Helgoland (BAH) marine station. These zooplankton samples were provided to the laboratory for cultivation experiments and for eventual observation. Phytoplankton and nutrients had been measured every work day since 1962 (Hickel et al. 1993); the expected availability of this information, the then dominant bottom-up theory and the measurement of meso- and macrozooplankton led to the expectation that even in a hydrographically complex locality the trophic controls of zooplankton population abundance and distribution might become evident. The aim was the production of time-series which would permit the testing of simulation models on the trophodynamics of the pelagic ecosystem then believed to be the decisive mechanism for population control. The basis of the intended modelling in those days was a conceptual model of the IBP type (Greve 1972).
Sampling, preservation, sorting and counting
The plankton nets commonly used for the sampling of zooplankton on the small cutter “Ellenbogen” until April 1974 were not equipped with a counter to measure the volume filtered, nor with a conical design that avoided escape responses of zooplankton approaching the net. These nets were replaced by the HYDROBIOS quantitative collection device 438040 (150 μm hand-net, aperture 17 cm, net length 100 cm, FlowMeter, ballast weight) and the HYDROBIOS ring trawl net 438700 (Calcofi Net; mesh size 500 μm, 100 cm aperture, length 400 cm, depressor and FlowMeter). Both nets were used for oblique hauls under the prevailing conditions of Helgoland Roads, generally not permitting vertical hauls due to a depth of only 6 to 8 m.
Sampling was generally carried out at the position 54°11′18″N 7°54′ E) during morning hours, thereby shifting the sampling by 45 min daily through the tidal cycle. Samples were live-counted until 1981 on a daily basis. In 1982, sampling was reduced to Mondays, Wednesdays and Fridays. Samples are preserved in 4% formaldehyde. As some species (e.g. Bolinopsis infundibulum) cannot be preserved, occasional live counts were continued for limited time-periods on the basis of plane transports from Helgoland to Hamburg.
The taxonomic and numeric biodiversity was investigated and documented to the earliest possible date, handling meso- and macrozooplankton separately. While mesozooplankton was determined by F. Reiners and W. Greve only, macrozooplankton was determined by W. Greve, J. Nast, S. Hoffmann and various other assistants and students. The list of species and stages counted varied with taxonomic changes and increasing professional expertise. Fish larvae which were at first intentionally excluded from taxonomic analysis due to the institutional organisation, were determined and counted for 7 years from 1990.
Metazoa continuously change their characteristics with growth. Fixation does not preserve the pigments of fish larvae; they become bleached. Therefore, the taxonomic treatment of the complete size class, especially of macrozooplankton, requires expertise that can only be obtained after a long training period. Though this training could be supported by the experiences gained, individual inconsistencies cannot be excluded.
The training was supported by selected literature (e.g. Sars 1903; Zimmer 1909; Grimpe 1911–1942; Pesta 1928; Russell 1953, 1970, 1976; Williamson 1957; Kermack 1966ff; Jones 1976; Pierrot-Bults and Chidgey 1988; Ingle 1992), photographs and drawings, samples from the sorting exercises and a collection of reference samples.
The plankton was sub-sampled according to the number of organisms present. An aliquot of 60 individuals of each key species was regarded as an acceptable threshold level. Determination and counting were carried out in counting chambers using microscopes.
Coverage of the time-period sampled
The aim was to operate the investigation at the highest possible frequency. This was possible on a daily basis (weekdays) until 1981, then the sampling frequency had to be reduced to Mondays, Wednesdays and Fridays. Mesozooplankton sample analysis could be maintained at this frequency. For macrozooplankton, the taxonomic range was extended to ichthyoplankton in the 1990s. This taxonomic expansion of the analyses had to be countered by a reduction in the number of samples analysed to one sample per week. Due to organisational changes within the institutions involved, the macrozooplankton samples lag behind the mesozooplankton analysis.
The sorting and counting protocols are transferred into electronic storage via computer programs written in Visual Basic under Excel. Individuals m−3 are calculated as the standard value. Control programs check the plausibility of data with respect to value limits, parameter names and double entries.
Additional external information on weather, physical parameters, nutrients and phytoplankton is available as ASCII files in a comparable format from various sources [e.g. DWD (German Weather Service), BAH/AWI, BSH (Federal Marine and Hydrographic Agency of Germany)].
Alphabetic lists of taxonomic and ontogenetic entities (76 mesozooplankton and 367 macrozooplankton) determined and enumerated since 1 April 1994. Further variables have been separately documented. Meso- and macrozooplankton have been sampled every Monday, Wednesday and Friday since April 1 1974 and the macrozooplankton counted at least every Wednesday until 2001
Amphiura filiformis pluteus
Asterias rubens bipinnaria
Asterias rubens brachiolaria
Beroe spp. juvenile
Carcinus maenas megalopa
Carcinus maenas zoea
Centropages hamatus female
Centropages typicus female
Crangon crangon zoea
Euterpina acutifrons + Tisbe spp.
Galathea spp. zoea
Ophiura spp. pluteus
Pagurus bernhardus zoea
Paracalanus parvus + Pseudocalanus elongatus
Pleurobrachia pileus juvenile
Polychaeta larva undetermined
Psamechinus miliaris pluteus
Pseudocalanus elongatus female
Rathkea octopunctata + Lizzia blondina
Rathkea octopunctata + Lizzia blondina not differentiated
Temora longicornis nauplius
Aglantha digitale sum
Agonus cataphractus larva
Anguilla anguilla larva
Anomalocera patersoni not differentiated
Anomalocera patersoni sum
Anomura zoea undetermined
Anomura zoea sum
Anomura megalopa undetermined
Anomura megalopa sum
Anthomedusa undetermined not differentiated
Athanas nitescens zoea sum
Beroe cucumis <10 mm length
Beroe cucumis 10–20 mm length
Beroe cucumis 20–40 mm length
Beroe cucumis not differentiated
Beroe cucumis sum
Beroe gracilis <5 mm length
Beroe gracilis 5–10 mm length
Beroe gracilis 10–20 mm length
Beroe gracilis 20–30 mm length
Beroe gracilis not differentiated
Beroe gracilis sum
Beroe spp. <5 mm length
Beroe spp. 5–10 mm length
Beroe spp. 10–20 mm length
Beroe spp. not differentiated
Bougainvillia britannica not differentiated
Bougainvillia britannica sum
Bougainvillia macloviana not differentiated
Bougainvillia macloviana sum
Bougainvillia principis not differentiated
Bougainvillia principis sum
Bougainvillia ramosa not differentiated
Bougainvillia ramosa sum
Bougainvillia spp. not differentiated
Brachyura zoea undetermined
Brachyura zoea sum
Brachyura megalopa undetermined
Brachyura megalopa sum
Brachyura juvenile sum
Calanoid copepods sum
Calanus spp. copepodit
Calanus spp. female
Calanus spp. male
Calanus spp. not differentiated
Callianassa subterranea zoea not differentiated
Callianassa subterranea zoea sum
Callianassa subterranea megalopa
Callianassa zoea sum
Callianassa zoea and megalopa sum
Cancer pagurus zoea not differentiated
Cancer pagurus zoea sum
Cancer pagurus megalopa
Cancer pagurus zoea and megalopa sum
Candacia armata not differentiated
Candacia armata sum
Candacia spp. sum
Carcinus maenas zoea not differentiated
Carcinus maenas zoea sum
Carcinus maenas megalopa
Carcinus maenas juvenile
Carcinus maenas zoea, megalopa and juvenile sum
Caridion steveni zoea not differentiated
Caridion steveni zoea sum
Cladonema radiatum not differentiated
Cladonema radiatum sum
Corystes cassivelaunus zoea not differentiated
Corystes cassivelaunus zoea sum
Corystes cassivelaunus megalopa
Corystes cassivelaunus zoea and megalopa sum
Crangon allmanni zoea not differentiated
Crangon allmanni zoea sum
Crangon allmanni juvenile
Crangon allmanni zoea and juvenile sum
Crangon crangon zoea not differentiated
Crangon crangon zoea sum
Crangon crangon juvenile
Crangon crangon zoea and juvenile sum
Crangon spp. zoea not differentiated
Crangon spp. zoea sum
Crangon spp. juvenile
Crangon spp. zoea and juvenile sum
Crangon zoea sum
Crangon juvenile sum
Crangon zoea and juvenile sum
Crangonidae zoea undetermined
Crystallogobius sp. larva
Cyclopterus lumpus larva
Diastylis rathkei not differentiated
Diastylis spp. not differentiated
Ebalia spp. zoea not differentiated
Ebalia zoea sum
Ebalia zoea and megelopa sum
Eirene viridula not differentiated
Eirene viridula sum
Eualus occultus zoea not differentiated
Eualus occultus zoea sum
Eualus pusiolus zoea not differentiated
Eualus pusiolus zoea sum
Eualus spp. zoea not differentiated
Eualus spp. juvenile
Eualus zoea sum
Eualus zoea and juvenile sum
Eucheilota maculata not differentiated
Eucheilota maculata sum
Euphausiacea larva undetermined
Eutima gegenbauri not differentiated
Eutima gegenbauri sum
Eutima gracilis not differentiated
Eutima gracilis sum
Eutima insignis not differentiated
Eutima insignis sum
Eutonina indicans not differentiated
Eutonina indicans sum
Fish egg undetermined
Fish egg sum
Fish larva undetermined
Fish larva sum
Galathea spp. zoea not differentiated
Galathea spp. megalopa
Galathea zoea sum
Galathea megalopa sum
Galathea zoea and megalopa sum
Gastrosaccus spp. not differentiated
Helgicirrha schulzei not differentiated
Helgicirrha schulzei sum
Hippolyte varians zoea not differentiated
Hippolyte varians sum
Homarus gammarus zoea
Hyas spp. zoea not differentiated
Hyas zoea sum
Hyas zoea and megalopa sum
Hybocodon prolifer not differentiated
Idotea linearis not differentiated
Idotea linearis sum
Iphinoe trispinosa not differentiated
Iphinoe trispinosa sum
Labidocera wollastoni not differentiated
Labidocera wollastoni sum
Laodicea undulata not differentiated
Laodicea undulata sum
Leptomedusa undetermined not differentiated
Leptomedusa undetermined sum
Leuckartiara octona not differentiated
Leuckartiara octona sum
Liocarcinus spp. zoea not differentiated
Liocarcinus zoea sum
Liocarcinus spp. megalopa
Liocarcinus zoea and megalopa and juvenile sum
Lizzia blondina + Rathkea octopunctata not differentiated
Lizzia blondina + Rathkea octopunctata sum
Lovenella clausa not differentiated
Lovenella clausa sum
Macropodia rostrata zoea not differentiated
Macropodia rostrata zoea sum
Macropodia rostrata megalopa
Macropodia rostrata zoea and megalopa sum
Margelopsis haeckeli not differentiated
Mesopodopsis slabberi not differentiated
Mesopodopsis slabberi sum
Muggiaea atlantica gonozooid
Muggiaea atlantica nectophore
Muggiaea atlantica sum
Natantia-Caridea zoea undetermined
Natantia-Caridea zoea sum
Natantia-Caridea juvenile sum
Natantia-Caridea zoea and juvenile sum
Nyctiphanes couchii not differentiated
Nyctiphanes couchii sum
Obelia spp. not differentiated
Pagurus spp. zoea not differentiated
Pagurus spp. megalopa
Paguridae zoea sum
Paguridae zoea and megalopa sum
Palaemon spp. zoea not differentiated
Palaemon spp. zoea sum
Pandalina brevirostris zoea sum
Pandalus montagui zoea
Paramysis spiritus not differentiated
Paramysis spp. not differentiated
Paramysis spp. sum
Phialidium hemisphericum + Phialella quadrata not differentiated
Phialidium hemisphericum + Phialella quadrata sum
Philocheras bispinosus zoea sum
Philocheras trispinosus zoea not differentiated
Philocheras trispinosus zoea sum
Philocheras trispinosus juvenile
Philocheras trispinosus zoea and juvenile sum
Philocheras spp. zoea not differentiated
Philocheras zoea sum
Philocheras zoea and juvenile sum
Pilumnus hirtellus zoea not differentiated
Pilumnus hirtellus zoea sum
Pilumnus hirtellus megalopa
Pilumnus hirtellus zoea and megalopa sum
Pinnotheres pisum zoea not differentiated
Pinnotheres pisum zoea sum
Pinnotheres pisum zoea and megalopa sum
Pleurobrachia pileus <3 mm diameter
Pleurobrachia pileus 3–5 mm diameter
Pleurobrachia pileus 5–10 mm diameter
Pleurobrachia pileus 10–15 mm diameter
Pleurobrachia pileus 15–20 mm diameter
Pleurobrachia pileus not differentiated
Pleurobrachia pileus sum
Podocoryne borealis not differentiated
Podocoryne carnea not differentiated
Podocoryne carnea sum
Processa modica modica zoea not differentiated
Processa spp. zoea not differentiated
Processidae zoea sum
Processidae zoea and juvenile sum
Pseudocuma spp. not differentiated
Pseudocuma spp. sum
Rathkea octopunctata not differentiated
Rathkea octopunctata sum
Sarsia eximia not differentiated
Sarsia eximia sum
Sarsia gemmifera not differentiated
Sarsia gemmifera sum
Sarsia prolifera not differentiated
Sarsia prolifera sum
Sarsia tubulosa not differentiated
Sarsia spp. not differentiated
Sarsia eximia + Sarsia prolifera sum
Sarsia gemmifera + Sarsia tubulosa + Stauridiosarsia pr. + Dipurena ha. sum
Steenstrupia nutans not differentiated
Steenstrupia nutans sum
Thalassinidea zoea sum
Thia scutellata zoea sum
Thia scutellata megalopa
Thia scutellata zoea and megalopa sum
Tiaropsis multicirrata not differentiated
Tiaropsis multicirrata sum
Upogebia spp. zoea not differentiated
Upogebia zoea sum
Upogebia zoea and megalopa sum
The internet site http://www.senckenberg.de/dzmb/plankton gives further information on the research strategy, methodology and current results of the operative plankton prognoses. Since April 2004, a phenological prognosis is given for 50 populations. Based on the available time-series, the prognoses are recalculated automatically every day for each population. The automated phenological analysis calculates phenological predictions for every week according to the available abundance and temperature records, and chooses the week with the minimal deviation from that recorded during recent years for the current prognoses. The predictions span periods of up to 6 months. The error of the calculations is given as absolute times and as a percentage of the total observed phenophase period.
Auxiliary calculations: advection
Plankton advection occurs in three dimensions at variable orders of magnitude according to the tidal phase, and climate- and weather-driven currents. In order to be able to calculate this influence on the daily plankton catches, a simulation model was used. This model, of the “drawer type”, operates on the basis of former calculations of the operative advection model from BSH, from which the current conditions were extracted for the main wind directions and forces at Helgoland after a period of 3 days of continuous, almost identical, wind conditions. These values can then be used as approximations to the values to be calculated in the BSH operative advection model. While this model requires a broad input of climate data, computer capacity and time, the drawer model can easily be run for any period for which wind conditions at Helgoland are available from the German Weather Service (DWD). The drawer model DRIFT (Reiners, Müller-Navarra and Greve, unpublished data) is available as a GFA-Basic program from the authors.
Publications (papers, MURSYS, BLMP reports)
The results of the Helgoland Roads meso- and macrozooplankton analyses are partially published in the current “marine Umwelt reporting System MURSYS” from BSH as part of the national environmental monitoring programme. Reports of the Bund Länder Meßprogramm (BLMP) are available in regularly updated current versions on the internet, and as a printed version (e.g. BSH 2002). The Helgoland Roads zooplankton collection is also reported in the ICES WG Zooplankton Ecology (Valdez 2001).
A bibliography of the main results includes the following publications: Greve (1992, 1994, 1995a, 1995b); Greve and Reiners (1980, 1988, 1995, 1996); Greve et al. (1996, 2001); and Lange and Greve (1997).
In the 30 years samples have been taken at Helgoland Roads, 443 taxonomic entities have been determined (Table 1), counted and calculated for standardised electronic storage. Due to the long time period, the high taxonomic resolution and the high sampling frequency, a unique data pool exists. As the catching method permits the identification of gelatinous zooplankton as well as crustacean plankton and fish larvae, the description and analysis of interacting populations is possible with only the restrictions of single point sampling.
The information thus obtained includes measurements on the abundance, the change of abundance on timescales of days to decades and the variance of these changes for the major zooplankton species, representing the biodiversity of the meso- and macrozooplankton, including the ichthyoplankton, in the waters of the German Bight. These changes reveal distributions and response patterns which help the understanding of the marine ecosystem. Some of these aspects are exemplified here.
The composition of the mesozooplankton in the shelf sea near Helgoland is primarily determined by holoplanktonic populations, but for some periods meroplankton is more abundant in the mesozooplankton.
The macrozooplankton includes hydro- and laval scyphomedusae, decapod and fish larvae in quantities which can be larger than the abundance of ctenophores, chaetognaths, the large copepod Calanus helgolandicus, and mysids to mention the major groups. So the shelf sea macrozooplankton is influenced by meroplankton to a great extent.
Holo-zooplankton of the meso- and macrozooplankton size classes is composed of a rich variety of species, standing for diverse functional units. These represent omnivory with a preference for herbivory of selected size classes (e.g. Calanus helgolandicus feeding on larger particles and Oikopleura dioica feeding on small particles), omnivory with a preference for carnivory (e.g. Oithona helgolandica), omnivory with a preference for detritivory (e.g. Euterpina acutifrons), and primary carnivory (e.g. Pleurobrachia pileus) or secondary carnivory (e.g. Beroe gracilis).
Variance of zooplankton measurements
Zooplankton abundance and distribution varies under the influence of ethology (e.g. patchiness, diurnal vertical migration), physical forcing (advection, temperature dynamics), and physiological (e.g. temperature responses) and ecological conditions (e.g. trophodynamics, ontogeny) on any time and space scale with up to several orders of magnitude difference. The interpretation of the measurements obtained under these conditions must consider this variance, which can partially be countered by a high sampling frequency, long oblique plankton hauls sub-sampled successively and the calculation of the results (e.g. averaging).
Annual zooplankton dynamics are characterised by the seasonal succession of populations characterising trophic functional relationships between functional groups or populations (Greve and Reiners 1988). Key species may dominate for restricted periods. Besides autecological functional discrimination, biocoenotic separation of seasonal communities is possible (Greve and Reiners 1995) indicating successive functional ecosystem equilibria.
Not all sequential events are based on trophic interactions and succession. Advection and migration also leads to a sequence of events that influences the local time-series measurements.
The annual zooplankton succession varies considerably from year to year with respect to the seasonality and the abundance of single populations. In some cases, these changes represent a trend over a few or many years. Four examples are given for such behaviour:
In the case of the ctenophore Pleurobrachia pileus, measurements are available for 14 developmental stages, of which the newly hatched juveniles are measured in the merozooplankton samples which haven been analysed for the complete period of 27 years. Fig. 4b demonstrates, in agreement with the measurements of adult specimens, a basic change in annual distribution starting in 1988 and continuing in the years after. In the years to 1977 an abundance maximum was observed in June, this peak abundance then expanded into spring and autumn. The abundance maximum first gets reduced and distributed over almost the whole year, followed by a period of earlier population growth. This corresponds to the observation of the response of the seasonality of P. pileus to increased winter temperatures (Greve et al. 2001) and to the change in the distribution patterns observed in North Sea plankton (Reid and Edwards 2001).
Another drastic change in the abundance and distribution of zooplankton was monitored in the case of the turbellarian Alaurina composita which, according to our observations and to related limnetic populations, is a voracious predator on copepods and cadocerans. This species is hardly taken into account in ecosystem investigations, though one individual may eat four copepods per day. Until the early 1980s, 2 to 4 individuals m−3 were occasionally registered in the summer months: suddenly in the mid 1980s the abundance increased by four orders of magnitude. Abundances of up to 5,000 individuals m−3 were recorded. In the following years, the abundance decreased again by one to two orders of magnitude (Greve and Reiners 1996). The higher abundance level prevailed until 2002 (Fig. 4c).
The final example for faunal changes in the zooplankton is the immigration of the neozoon Penilia avirostris. This cladoceran is a Mediterranean and subtropical species. It was first observed as a few specimens in the winter of 1991. Since then it has reappeared irregularly with increasing abundance and length of season (Fig. 4d).
Appearance and disappearance of populations at Helgoland Roads. Mean annual abundances Ind·m–3. The generally abundant appendicularian Fritillaria borealis was missing in 1999 and almost missing in 2001. The populations of Penilia avirostris, Doliolum nationalis and Muggiaea atlantica started to appear in the zooplankton 1989 and 1990. Since then, they have been part of the local fauna
The ambient temperature permanently controls the physiology of marine organisms. This control concerns e.g. the digestive system, the ontogeny and the development of gonads (Lange and Greve 1997). One of the processes in which such a control becomes visible is the seasonal timing of populations (Greve et al. 2001).
Observation- and functionality-based prognoses represent two ways to understand the decisive role of phenology in marine ecology (Greve 2003).
The prediction of population dynamics is one aim of ecological research. The temporal prognosis based on the phenology of zooplankton was attempted on the basis of the Helgoland Roads time series (Greve et al. 2001). Regressions of temperatures with phenophases result in correlation coefficients of up to r2=0.8. This stands for a dominant population control, compared to the temporal influence of trophic relationships. The abundance, however, remains to be predicted. The match/mismatch analyses may contribute to this prognosis.
Unbiased observation of nature is almost impossible, as our scientific education includes learning and shaping. The shape we get is expressed in paradigms, according to which we design our investigations. The design of our studies can determine the outcome, the result, which is thus self-fulfilling and thus conserves our bias. As no reliable paradigm was available at the start of the measurements of the Helgoland Roads time-series, this project was started as a very general programme monitoring the possibly important zooplankton populations which might be required for future prognostic models. The resulting database enabled the description and spatial analysis of many specific processes, and the development of simulation and statistical models, which led to operative prognoses of the temporal distribution of the regional zooplankton populations. In the advancement of operative models, the data provide a permanent source of information.
The Helgoland Roads zooplankton time-series is biodiversity based. The species-specific changes in distribution and abundance—in some cases even the differences in developmental states—can be studied, specifically analysed or aggregated according to the investigated hypothesis. These aggregations can always be reduced again to single population dynamics for detailed analysis.
Shelf sea zooplankton is characterised by the influence of meroplankton, which is of increasing importance with increasing size of the organisms. Merozooplankton consists of the pelagic generation of coelenterates and the larval stages of zoobenthos and nekton. In this ontogenetic stage, the rate of annual recruitment is determined for many benthic and nektonic populations during the planktonic state, as the abundance and timing of the eggs and larvae is determined by the nutritional and physiological state of the parental generation. This bentho-pelagic coupling is especially great in shelf seas and contradicts any attempt to understand the benthic and the pelagic systems separately.
Zooplankton survival depends on the availability of primary production (bottom-up control) and on the presence or absence of the population-specific predators (top-down control). The annual development of these trophic controls was weighed by Greve and Reiners (1995).
An example of annual trophic control is given for 1987: it is not a causal analysis of the complete population dynamics in this year, as many more populations are part of the trophodynamic ecosystem regulation, and nor does it neglect the possible influences of e.g. advection. However, the comparison of Helgoland Roads dynamics with regional dynamics in the German Bight suggests that the local time-series can be regarded as an indicator of the regional population dynamics (Greve and Reiners 1988). This report described the local succession of trophic controls as the expression of population waves travelling from the coast to the central German Bight. Sequence and succession appears in the form of three-dimensional regional processes (Greve and Reiners 1980, 1988). The influence of regional temperature differences and advection remains to be analysed.
One year cannot be taken for a prototype of regional trophodynamics for many years. Averaging the annual dynamics tends to extinguish the trophic responses, as short time differences from year to year may lead to a contradictory temporal orientation.
In situ measurements provide important indicators and controls for ecosystem causal analysis, but only laboratory experiments and controlled ecosystem process studies linked by a theory in a common model will enable understanding of such complex process dynamics (Greve 1992).
A unifying concept of population control
Population prognosis is the final test of hypotheses, models and paradigms. The measurement of the skill of such prognoses requires the availability of independent time-series which were not used for the development of the model. The Helgoland Roads zooplankton time-series provide such data.
The numeric variance of zooplankton based on generation times, trophodynamics, advection, physiology and ambient temperatures must be understood for controlled ecosystem management.
The causal hypotheses derived from the ecosystem analysis can be tested in operational prognostic models which can be used continuously to disprove hypotheses and to improve them. The improvement of the prognostic skill must be of focal importance (Greve 1995b).
A first operative prognosis of Helgoland Roads zooplankton populations is available under www.senckenberg.de/dzmb/plankton. It operates on the basis of the phenology of zooplankton, which is mainly based on preceding temperatures (Greve et al. 2001). This statistical relationship is poorly understood as a functional causal relationship. The physiological requirements of zooplankton concern temperature and salinity. Each population has specific preference profiles for both parameters, as measured for Noctiluca scintillans (Uhlig and Sahling 1995). As these measurements are lacking for most zooplankton, the profiles can be determined in a first approximation by relating the annual abundance measurements and the distributional information to the ambient temperatures and salinities. In the appendicularians Fritillaria borealis and Oikopleura dioica, a temperature profile with an optimum value of 12°C for F. borealis and 14°C for O. dioica enabled the simulation of the seasonal variance of the two species with the extended niche model (Greve 1995b). These values were extracted from the Helgoland Roads time-series.
The potential of climatic influences on the distribution of zooplankton was shown by Southward et al. (1995) who registered a lateral displacement of populations, and by Greve et al. (1996) who detected a shift in the seasonal timing of zooplankton.
It is surprising that phenology, a well-established science in terrestrial ecology, has not made it into the aquatic sciences earlier (Greve 2003). Climate change requires extended phenological research in marine ecosystems. Recent global warming has led to phenological changes which it has been possible to observe in terrestrial systems (Menzel 2000). Phenological responses have also been observed in neritic copepods (Kioerboe et al. 1988). Recent observations of lateral shifts of zooplankton (Southward et al. 1995) and algae (Molenaar and Breeman 1997) prove that changes in the marine biota based on climatic alterations are already underway. Marine biometeorology should be further developed by observing the phenology of marine zooplankton over a wider range of latitudes than just Helgoland Roads.
The extensive multi-decade CPR plankton collection over the North Atlantic by the Alister Hardy Foundation for Ocean Sciences (SAHFOS), which enabled the recognition of faunal changes (Beaugrand et al. 2002), can be supplemented with such series, leading to the causal analysis of such changes. They can both contribute to anticipating changes in the pelagic plankton in the coming period of biocoenotic responses to global warming. The high frequency Helgoland Roads time-series permits the measurement of the regression analysis of temperature and phenology. Gelatinous plankton, which is only marginally represented in the CPR programme, is sampled and determined at Helgoland Roads and thus available for food web analysis.
Changes in the ecosystem due to new species immigrating and the loss of others which can no longer exist in the southeast North Sea in future warmer years can best be realised and valued in a time-series which can compare the present and the future with the documented past. Helgoland Roads is a permanent watch station for such changes which might otherwise alter the ecosystem without being recognised (e.g. Greve 1994).
The Helgoland Roads time-series on meso- and macrozooplankton is unique with respect to its length, sampling frequency, biocoenotic resolution (including the gelatinous populations) and the size spectrum of organisms investigated. The insight into the dynamics of the pelagic ecosystem obtained has revealed trophic control mechanisms, the relevance of these to the ecosystem, and the influence of temperatures from much earlier on the seasonality of zooplankton, including many benthic and nektonic populations. Thereby, high frequency zooplankton sampling has proven its importance for bentho-pelagic coupling and general marine ecosystem analysis.
Ecosystem management is based on our understanding of the ecosystems. The available knowledge will be converted into models (simulations and statistical models). Permanent operative prognoses, as started this year, will help to detect weaknesses in our understanding, leading to investigations and improvements to our operative models. Decisions will be optimised based on continuous observations of environmental indicators. This time-series could provide a further basis for such ecosystem management support.
This investigation was made possible by the Biologische Anstalt Helgoland, the Forschungsinstitut Senckenberg and the Bundesamt für Seeschiffahrt und Hydrography. It was supported by the captains and crews of the cutters “Ellenbogen”, “Aade”, “Diker”, “Uthörn I”, “Uthörn II”, “Friedrich Heincke” and “Heincke”, by the staff of the marine station preserving, packing and shipping the samples to Hamburg, by many assistants and students who dealt with the lab-work sorting and counting the samples, and putting the results onto electronic media. Uwe Lange, Christian Reick, Silke Johannsen, Peter Johanssen, Michael Jacobsen and the co-operation partners from Hamburg University, Prof. J.O. Backhaus and Prof. B. Page supported the analysis of the Helgoland Roads zooplankton in the BMBF project BEO 03F0223. K.H. Wiltshire initiated this collection of Helgoland Roads times-series reports. We thank all of those mentioned for their contribution to the success of the Helgoland Roads zooplankton time-series.