Copepod grazing and its potential impact on the phytoplankton development in the Barents Sea ULF BAMSTEDT, HANS CHRISTIAN EILERTSEN, KURT S . TANDE, DAG SLAGSTAD and HEIN RUNE SKJOLDAL BQmstedt, U., Eilcrtsen, H. Chr.. Tande, K. S.. Slagstad. D. & Skjoldal. H. R. 1992: Pp. 339-353 in Sakshaug, E . , Hopkins. C. C. E. & Britsland. N. A. (eds.): Proceedings of the Pro Marc Symposium on Polar Marine Ecology. Trondheim, 12-16 May 1990. Polar Research I O ( 2 ) . Compiled data from published and unpublished sources on copepod grazing of the large-sized copepods in the Barents Sea give wide ranges in grazing rates. Approximate average values indicate daily rations of 7-18% for copepodite stages V and VI and considerably higher values for the earliest copepodite stages. It is demonstrated that individual variability in gut fullness of copepods from a given locality is typically very high and not closely related to variable food abundance o r depth of occurrence. There is no die1 feeding rhythm during the summer. and even when relating copepod grazing to a number of biotic and abiotic factors through stepwise linear regression analysis, much of the variability remains unexplained. It is suggested that feeding behaviour, food quality and feeding history of the copepods all play important roles as factors which regulate copepod grazing. Model simulations on the phytoplankton succession, using literature data on laboratory-determined growth characteristics for solitary cells and colonies of the prymnesiophyte Phaeocystis pouchetii and large diatoms, indicate that the extent of the mixed layer and selective grazing by zooplankton are important factors that may explain the occurrence of dense blooms of P. pouchetii colonies, frequently observed during the spring. U. Bdmstedt, Department of Fisheries and Marine Biology. High-Technology Centre N-5020 Bergen. Norway; H.Chr. Eilertsen and K . S . Tande. Norwegian College of Fisheries. Unioersity of Tromsm, P . O . Box 3083 Guleng. N-9001 Tromso, Norway: D . Slagstad, SINTEFAutomatic Control, N-7034 Trondheim- NTH, Norway; H . R . Skjoldal, Institute of Marine Research, P . O . Box 1870 Nordnes, N-5024 Bergen. Norway (revised August 1 9 9 1 ) . Introduction The Barents Sea is a very important area econ- omically for commercial fisheries; it is therefore of utmost interest t o study the biological pro- duction in this area. Since the main pelagic fish stocks, capelin (Mallotus uillosus) and herring (Clupea harengus), feed on zooplankton, there is a direct relation between production-related studies on zooplankton and the commercial fish- ery here. The limit for zooplankton production is primarily set by the amount of food consumed, and secondarily by the assimilation efficiency and the metabolic costs associated with life. Quantitative information on feeding by zooplank- ton may therefore be considered a tool which can be used to estimate zooplankton production, thereby providing important information for an estimation of the carrying capacity for plank- tivorous fish stocks. Zooplankton is also con- sidered a major regulating factor for lower trophic levels. Zooplankton grazing may influence the development and succession in the phytoplankton community directly by (1) depressing the total stock and (2) depressing part of the stock by selective grazing. Indirectly it may influence the phytoplankton by (3) reducing the grazing press- ure from microzooplankton by predation and (4) stimulating algal growth by regenerating plant nutrients. Estimates of the total impact of cope- pod grazing in the Barents Sea are restricted to point (1) above. These studies suggest that the impact is small in the spring ( 5 2 0 % of the pri- mary production, Eilertsen et al. 1989a) but con- siderably larger during summer (65-90% of the primary production, Eilertsen et al. 1989b; l& 400%, Hansen et al. 1990a). In view of the fairly extensive new literature on feeding behaviour and food selectivity in copepods (e.g. Huntley 1988; Jonsson & Tiselius 1990), points (2) and (3) are certainly also important. For example, one of the dominant algal species in the Barents Sea, the prymnesiophyte Phaeocystis pouchetii, may cause a reduction or even inhibition of the feeding in copepods (Dagg et al. 1982; Schnack et al. 1985, Verity & Smayda 1989). Till now, this effect has not been demonstrated in field studies from the Barents Sea (Eilertsen et al. 1989b) or in labora- 340 U. Bdmstedt et a1 Tabk 1 . Calculated individual carbon content (pg) of the copepodite stages of large-sized copepods from thc Barents Sca. Reference sources: (1) Protein content from the Barents Sea in May-June 1987 and conversion factors. Bimstedt 1986; (2) Slagstad & Tande 1990: (3) Eilertsen e l al. 1989b. Conover & Corner 1968: (3) Dry weight from the Barents Sea in August 1988 and conversion factor. Bimstedt 1986. Stage IV copepodites represent adult females. Species/Stage c-l C-11 c-Ill c-1v c - v c-VI Ref. Calanus finmarchicur 3 8 23 64 193 23 1 ( 1 ) Calanus glacialis 4 8 25 100 264 576 ( 2 ) Calanus hyperboreus I 16 17 230 743 1399 (3) Mefridia longa 193 (4) - - - - tory studies using the common copepod species in the Barents Sea (Tande & Bimstedt 1987; Huntley et al. 1987; Hansen et al. 1990b). However, Estep et al. (1990) used an incubation technique and image analysis in a field study and found evidences for a variable feeding behaviour of Calanus finmarchicus and C . hyperboreus, related to the physiological condition of P . p o u - chetii. In this paper we give a brief review of quantitative data on copepod grazing from the Barents Sea and formulate a simple mathematical model that simulates the effect of copepod grazing on the phytoplankton succession during the first part of the production season. Material and methods Copepod grazing rates The original papers should be consulted for a description of the material and methods used for the published data. The unpublished data from 1983 were based on in sitL incubations and sub- sequent analysis with an Elzonem particle coun- ter (Bimstedt et al. 1985). Unpublished results from 1984 and 1987 were produced as described by Tande & Bimstedt (1985), except that single copepods were analysed in the material from 1984. In 1987, zooplankton was sampled by a slow, vertical haul, using a 1-m diameter WP-2 net with 300 pm mesh and equipped with a non- filtering cod-end, 15 I volume. Healthy animals were sorted out and kept in 100ml of filtered seawater for up to 30min prior to incubation. -4mbient seawater samples from defined depths with the natural assemblage of particles were prepared in advance by mixing 3.7 X 103Bq (100pCi) of NaH14C03 to 41 water in poly- carbonate bottles and incubating them in natural light for 24 hours. The zooplankton samples (100 ml) were mixed with 400 ml radio labelled experimental water, using tissue-culture bottles of 500 ml capacity, and incubated for 0.7-1.0 h at ambient temperature and dim light exposure. Incubation was finished by sieving off the cope- pods, anesthetising them with MS 222 (ethyl m- aminobenzoate), rinsing them individually and putting them into a scintillation vial with 0.5 ml tissue solubiliser. Standard procedures were used in the subsequent preparation and scintillation counting of the, samples. Grazing rate, in terms of consumed food carbon, was calculated from the ambient concentration of particulate carbon and estimated clearance rate, as given below: pg C intake h-' = pg C ml-' (1) DPMcopepod X ( D P M s t a r t -k D P M s t o p ) m1-l h i n a b a t i o n Individual content of body carbon for each species/stage (Table 1) was used in the calculation of daily rations (food carbon intake as percentage of body carbon). Phytoplankton growth Laboratory-derived data on the growth charac- teristics of large diatoms and solitary cells and colonies of Phaeocystis pouchetii at 0°C (Verity et al. 1991 this volume) were used in the model formulation. These parameter values are summa- rised below and explained under Model formu- lation. ~~~ ~ Parameter Diatoms Phaeocysrir Phaeocysrrc Units solitary cells Colonies P 0.33 0.3 0.22 Chl a j C 0.04 0.0125 0.01 1 9 0.018 0.037 0.030 ac O.OOO7 0.0005 0.0003 Copepod grazing in the Barents Sea 341 unknown, but they usually begin to form simul- Model formulation A mathematical model has been designed to assess the effect of grazing on the development of the primary production during the spring. The model contains a two-compartment model for phytoplankton and a population of single species of copepods (Calanus finmarchicus). The bio- logical sub-models are driven by a 1-dimensional (1-D) model of the vertical water column. Phytoplankton growth and distribution is described by a 1-D model of a vertical water column where P(t, z) is the concentration of diatoms or Phaeocystis pouchetii, w is vertical velocity (usu- ally sinking velocity) and K, is the vertical eddy diffusion coefficient. fbiol represents the growth and mortality of the phytoplankton as described by equation 3: (3) where fp(I) is a function that describes the relationship between photosynthetic rate [mg C rng(Chla)-' h-I] and light intensity (I), r is the phytoplankton respiration rate and GZ is the zooplankton grazing rate. The phytoplankton res- piration constant was set at 0.05 d-' (Bimstedt & Tande 1985; Sakshaug & Slagstad 1991 this volume). Sedimentation loss was assumed significant only for diatoms and a coefficient of 0.02d-' has been applied. Assuming no pho- toinhibition, the P vs. I relationship is described by: P c = P c 1 - e x p -- ,I ( €21 ~ (4) where Ps is maximum carbon-normalised pho- tosynthetic rate and & is the carbon-normalised initial slope of the curve (see Table above, Webb et al. 1974 and Sakshaug & Slagstad 1991). The model simulates a spring situation, which means that there are no nutrient limitations. During the spring there is a transition from solitary cells to colonies of P. pouchetii, which has a somewhat lower growth rate (see above). However, the colonies seem to be controlled dif- ferently, and a mass bloom of colonies is a com- mon feature in the late spring. The trigger mechanism for the formation of colonies is taneously with the occurrence of diatoms. Results from P-I measurements performed in the field between 1984 and 1989 showed decreased growth rate (based on carbon) of solitary cells of P. pouchetii when the amount of diatoms increased (Eilertsen unpubl. data). In the model we have therefore assumed that the production of P. pou- chetii colonies is triggered by an increase in the diatom concentration and a switching from soli- tary cells t o colonies at a diatom concentration of 1 mg Chi a m-3 (25 mg C m-3). Surface irradiance was calculated from the theoretical high of the sun at 74"N latitude. An average 50% reduction of insolation due to clouds was assumed, as suggested by data from B j ~ r n ~ y a , provided by the Norwegian Meteoro- logical Institute. Of this light, 50% was assumed to be usable for photosynthesis. A reflection loss of >lo%, dependent on the solar elevation, was used in the simulations. The attenuatibn coefficient, k, of light in the water column is given below (Parsons et al. 1977): k = {k, + 0.0088 Chl* + 0.054 (Chl*)2/3}/p (5) where k, (m-I) represents the attenuation coef- ficient for pure seawater, p is the average cosine of the light field, which has been set at 0.6 (Kirk 1983), and Chl* is the concentration of light- absorbing pigments calculated by the formula of Sakshaug & Slagstad (1991): Chl* = Chl a + 1/2 Phaeophytin (6) Based on data from the Barents Sea, Sakshaug & Slagstad (1991) described the relationship between chlorophyll a and phaeophytin as: Phaeophytin = 0.45 Chl a + 0.02 (7) During the early spring, the copepod biomass is dominated by adult females of C. finmarchicus. In our model we have therefore simplified the copepod population to consist entirely of adult female C. finmarchicus. A detailed model of the grazing impact from this species, based on the energy requirements, has previously been pub- lished by Slagstad (1981) and Slagstad & Tande (1990). The ingestion rate is determined by the population size and the clearance rate. The maxi- mum clearance rate (FR,) for a copepod is (Slag- stad & Tande 1990): FR, = FRO eo (8) where FRO is the maximum clearance rate 342 U . Bdrnstedt et al. T a b l e 2 . Summary of grazing-rate data, expressed as daily ration (carbon intake per day as percentage of copepod carbon content), for the large-sized copepods in the Barents Sea. Analysis techniques: gut-fluor = based on gut content of algal pigments (e.g. Mackas & Bohrer 1976); “C = radiolabelling technique (e.g. Conover & Francis 1973); Elzone = in situ incubation and electronic particle counting (e.g. B h s t e d t et al. 1985). Species Stage Period Daily Analysis Reference ration technique Calanus finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus C . finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus C. finmarchicus Calanus glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C. glacialis C glacialis C. glacialis C. glacialis C. glacialis C. glacialiF C. glacialis Calanus hyperboreus C. hyperboreus C. hyperboreus C. hyperboreus C. hvperboreus C. hyperboreus C. hvperboreus C. hyperboreus C. hyperboreus C. hyperboreus C. hyperboreus C. hyperboreus Calanus spp. Calanus spp. Merridia longa M longa C - I c-I. I1 C-I1 c-I11 c-111 c-1v c - v c - v c - v ad. female ad. female ad. female C-I C-I1 c-Ill c-IV c-IV c-IV c-1v c-IV c-IV c-IV c-IV c - v c - v c - v c - v c - v ad. female ad. female ad. female c-I c-11 C-111 c-IV c-IV c-IV c-IV c - v c - v c - v ad. female ad. female c-I1 c-111 ad. female ad. female May May May May-June May-June May-June May-June May-June May-June May-June May-June May-June July-Aug. July-AUg. July-Aug. July-Aug. May-June May-June May-June May-June May-June July May-June July July May-June May-June May-June May-June May-June May-June July-Aug. July-Aug. July-Aug. July May-June May-June May-June May-June May-June May-June May-June July July May-June May-June July-Aug. 0-5.2 4.S148.5 0-1 7 CL1.9 O . M . 6 6.7-82.1 0.3-27.6 O . S l 6 . 8 0-14.9 0-15.4 0.3-55.3 2.8-32.0 75.0 27.9 15.1 58.4 1.9-20.2 1.8-10.6 0.2-10.6 0.1-27.4 1.9-4.5 2.0-21.2 0.1-1 1.5 2 . 6 2 9 . 9 1.7-79 Cb24.0 0.8-13.9 0.2-7.7 0-19.2 0.3-8.5 0.5-53.8 67.9 87.5 119.8 102.5 2.2-40 0-7.6 4.8-20.2 0.5-36.0 1.2-2.4 0.2-3.8 0.3-17.4 2 . 6 2 2 . 1 18.2-72.7 1 4 . M 3 . 7 0.2-6.3 CL3 8 gut-fluor I‘C gut-fluor gut-fluor “C 1°C “C gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor I ‘C 1‘C I‘C gut-fluor gut-fluor Elzone gut-fluor gut-fluor gut-fluor gut-fluor Elzone gut-fluor Elzone gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor gut-fluor IJC gut-fluor gut-fluor Elzone I ‘C “C gut-fluor gut-fluor gut-fluor “C gut-fluor “C Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Tande & BBmstedt 1985 Unpublished from 1987 Unpublished from 1987 Tande & Bimstedt 1985 Unpublished from 1987 Eilertsen et al. 1989b Eilertsen et al. 1989b Eilertsen et al. 1989b Eilertsen et al. 1989b Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Unpublished from 1987 Tande & Bimstedt 1985 Hansen et al. 1990a Unpublished from 1983 Hansen et al. 1990a Unpublished from 1984 Unpublished from 1987 Tande & BBmstedt 1985 Unpublished from 1983 Unpublished from 1987 Unpublished from 1983 Tande & Bimstedt 1985 Eilertsen et al. 1989b Eilertsen et al. 1989b Eilertsen et al. 1989b Eilertsen et al. 1989b Unpublished from 1984 Unpublished from 1984 Unpublished from 1987 Unpublished from 1987 Unpublished from 1983 Unpublished from 1984 Unpublished from 1984 Unpublished from 1987 Hansen et al. 1990a Hansen et al. 1990a Unpublished from 1987 Unpublished from 1987 (mi h-l) at a temperature (T) of 0°C. W e used a temperature of 3°C in the model simulations. The parameter value of FRO for adult female C. finmarchicus varies from 4 . 2 mi h-’ (Huntley 1981) to 12 ml h - ‘ (Tande & BAmstedt 1985). W e have used t h e value 10.0 ml h-I throughout our simulations. An index of electivity (E; Ivlev 1961, cited in Valiela 1984) was used in some of t h e simulations, in o r d e r t o define a preference towards diatoms, in accordance with current the- Copepod grazing in the Barents Sea 343 as given by the I4C-technique (Table 2 ) . may reflect a greater technical difficulty with the small- est organisms in the analytical procedures. The displayed values given for the early copepodite stages above, especially for C. hyperboreus, might therefore be somewhat biased. Nevertheless, con- sidering the total information given in Table 2 , the daily ration of copepodite stages I-IV appears significantly higher than that of for the older stages. This has subsequently been confirmed for C . finmarchicus from Balsfjorden, northern Nor- way, in laboratory experiments with Thalassiosira nordenskioldii and Phaeocystis pouchetii as food, two algal species that are commonly predominant in the Barents Sea (Hansen et al. 1990b). The low values on daily rations of Metridia longa (Table 2) indicate that this species does not graze solely on phytoplankton. Data from Balsfjorden, northern Norway (Blmstedt et al. 1985), the west coast of Norway (Blmstedt & Ervik 1984) and the west coast of Sweden (Biim- stedt & Tande 1988) all indicate an omnivorous trophic state of this copepod. ory (cf. Verity & Smayda 1989) E = (ri - pi)/(ri + Pi) (9) where q is the proportion of species i in the diet and pi, the corresponding proportion in the food environment. The energy requirements for a spawning C. finmarchicus female is taken to be 20pg C d-' (egg production from Hirche 1990; gross growth efficiency from Peterson 1988). This gives a criti- cal food concentration of 83 mg C m-3. Results and discussion Daily rations of the copepods The compiled data on grazing rate of the large- sized copepods (Table 2) cover the season from May to August, i.e. most of the production period of the year. The data are expressed as daily rations (carbon intake as percentage of the body-bound carbon) and are given either as ranges or as mean values, dependent on the data available from the different sources. Most of the data are for older developmental stages, and these may also be con- sidered most reliable. If the median values are used from those studies where ranges are given, and used together with the displayed mean values, a representative average value for each species and stage can be calculated. These values are given below: Individual Variability in grazing rate In a previous paper by Bimstedt (1988) the indi- vidual variability of copepod bioenergetics was f x u s e d upon and explained by a suggested vari- able nutritional history. The present study gives a comparable example from the Barents Sea, original data from July 1984 on gut fullness of Species/Stage c-1 c-11 c-Ill c-IV c - v c-v1 Calanus finmarchicus 40.3 39.5 25.2 44.4 10.0 17.6 Calanus glacialis 75.0 27.9 15.1 14.5 16.0 13.7 Calanus hyperboreus 67.9 87.5 119.8 35.0 7.4 10.6 2.6 Metridia longa - - - - - Excluding Metridia fongu, the displayed values indicate that the oldest developmental stages (c- V, c-VI) have a typical daily ration of 7-18%, whereas the younger copepodite stages may reach 100% or more. The highest original figures in Table 2 were given for Calanus hyperboreus, taken from Eilertsen et al. (1989b). They used a published single value for the gut evacuation rate of all copepods, and the gut fullness of c-I, c-I1 and c-111 C . hyperboreus was only estimated from similar-sized copepodite stages of C . glacialis. Furthermore, the very wide range in daily rations for the early copepodite stages of C. finmarchicus, copepodite stage IV Calanus hyperboreus (Fig. 1). In the extreme case, shown by copepods from 40 m depth, the individual variability covered a range from 12 to 340 ng pigment individual-', i.e. a factor of 28 in difference between lowest and highest value. By considering these results, together with the curve for in situ fluorescence, it is obvious that food availability alone explains neither the individual variability nor the variation with depth. However, high individual variability may be caused by non-synchronous individual rhythms in feeding, as suggested for C. glacialis by Bimstedt (1984), and this may in turn be 344 U . BBmsiedt el d. Gut contents (ng chl. equiv. copepod-1) 0 100 200 300 400 0 Colmiitr hvptrhoreir.c c- IV F i g . I . Calanur hyperboreus c-IV. July 1984. Gut contents of algal pigments (chlorophyll a + phaeophytin), measured on single individuals. as related to depth and relative phyto- plankton distribution (shape of in situ fluorescence profile given). governed by the nutritional history of the indi- vidual copepod. Diurnal variation in feeding Published information on possible die1 rhythms in feeding activity of the large-sized Arctic copepods suggests the absence of a rhythm during the light season with thq midnight sun (BQmstedt 1984; Hansen et al. 1990a; BQmstedt & Skjoldal unpubl. data) but a pronounced diurnal cycle in late sum- mer, with maximum ingestion during the dark period of the day (Head et al. 1985). Our results for Calanus hyperboreus c-IV and C . glacialis c-V from July 1984 are based on samples consisting of 10 individually analysed copepods (Fig. 2). For the first species, highest gut contents were recorded in the afternoon, but the temporal vari- ation was relatively small in relation t o the indi- vidual variability at each depth and between depths. C. glacialis showed pronounced vertical Fluorescence (V) 0 4 8 1 2 0 4 8 1 2 0 4 8 1 2 0 4 8 1 2 0 4 8 1 2 nl7-l-n - m ml7-n-l rl7-n-n GUT CONTENTS (ng chl.equiv. cop.’) 0 100 200 300 0- 16.30 40 0 100 200 300 23.50 0 100 200 300 06.40 0 100 200 300 12.40 10.30 F 0 100 200 300 v 18.50 C. hyperboreus m 16.50 C. glacialis Fig. 2 . Calanur hyperborem c-IV and C. glacialis c-V. Vertical profiles of grazing rate (mean and SE. + or -)of copepods sampled 5 times during a 24-hour period in July 1984. Each point represents I0 individually analysed copepods. The displayed curve represents relative in situ fluorescence (in volts). Copepod grazing in the Barents Sea 345 Fig. 3. Calanus finmarchicus and C. glacialis. Copepod grazing rate versus depth of occurrence. The data from May-June 1987 represent vertical profiles from up to 8 different localities, and the wide ranges have necessitated the use of a logarithmic scale. 0 a0 Ad. female u . . . . . . . . . . . . . . . . . . . . o m o m 0 0 > 0 0 0 W O 0 0 0 0 0 0 0 rn 00 03 0 0 0 0 4.0) in explaining the grazing rate of C. finmarchiclrs in mid-bloom and post-bloom situations. Depth contributed significantly in two cases and primary production in a single case for C . finmarchicus (Table 3 ) . For C . glacialis density was the only independent variable that showed a significant contribution, and in the post-bloom situation none of the four variables showed an F-value exceeding 4.0. The results show that the relationship between the environmental condition and copepod feeding intensity is not usually a simple one. The visual impression from the graphs indicates that maxi- mum grazing rate tends t o occur somewhat below the chlorophyll maximum, and data transform- ation might therefore have improved the cor- relations. However, our main purpose was to evaluate if a simple relationship existed in the natural habitat, with a direct coupling between environmental factors and the grazing rate. Our data indicate that the chlorophyll distribution and water density usually govern the grazing rate of copepods, although much of the variability in grazing rate is not explained by these two factors. Grazing selectivity A selective grazing behaviour of zooplankton may force the phytoplankton species succession and therefore may be considered a major structuring ecological factor. From the early considerations where cell size and concentration of microalgae were the main regulating factors (e.g. Frost 1972) feeding behaviour and importance of food quality have been major components in the explanation of grazing control (see e.g. Huntley 1988 and references therein). More recently, attention has also been put to toxicity and unpalatability of algae in Scandinavian and Arctic waters (Huntley et al. 1987; Tande & Bimstedt 1987; Nielsen et al. 1990; Estep et al. 1990). In the Barents Sea, where Phaeocystis pouchetii usually constitutes a major part of the phytoplankton community (Eilertsen et al. 1989a and unpubl. results), a Simulation results During the spring the mixed layer may vary from 20m near the ice border t o more than 100m in the Atlantic Water. Fig. 7A shows that the population succession of diatoms and Phaeocystis pouchetii when no grazing occurs, is influenced when using alternatively 25 m, 40 m and 75 m as mixed depth. Diatoms will gradually dominate in the first two cases, whereas P . pouchetii is better adapted than diatoms to the situation with 75 m mixed depth. Figure 7B shows that the relative initial concentration is of great significance for the balance between the two algal forms. With two times higher initial concentration of diatoms (PO = 0.5 x Do) P . pouchetii stabilises at a low level, whereas diatoms increase exponentially. With higher initial concentrations of P. pouchetii (PO = 2 x D o resp. 4 x Do), a dominance of P . ' pouchetii colonies will be established (Fig. 7B). The distribution of the grazing pressure from 2,000 adult females Calanus finmarchicus has a significant influence on the algal community suc- cession (Fig. 7C). With only a slight preference towards diatoms ( E = 0.2) total algal biomass will be low and slightly dominated b y diatoms. A stronger preference towards diatoms ( E = 0.4 and 0.9, respectively) generates a correspondingly higher dominance of P . pouchetii. In the case with very strong preference towards diatoms ( E = 0.9), the critical diatom concentration for trig- gering production of P. pouchetii colonies is not 350 U . Bdmstedt et a1 '? E 0 F p? E 0 E" 150 100 50 1 l o 20 '1 l o 20 '1 March April reached and production of P . pouchetii solitary cells is exponential (Fig. 7C). However, solitary cells of P . pouchetii a r e usually not found in concentrations higher than 0 . 5 m g Chl a m - 3 (40 mg C m - 3 , Eilertsen unpubl. results). Micro- zooplankton grazing may account for this dis- crepancy. Thus, T. Dale (pers. comm.) found that the ciliates rapidly increased their biomass during the spring bloom in the Barents Sea, and Eilertsen (unpubl. results) found a high phaeopigment/Chl a ratio (ranging from 0.3 t o 2 . 3 ) during times with dominance of P . pouchetii solitary cells. D u e to lack of quantitative data o n this problem, we have not included it in our model. None of the above simulations indicate any fast succession in spring towards an extensive 10 20 1 May Fig. 7. Simulated population succession (mg C m-') in a community with diatoms (broken lines) and Phaeocystis pouchetii (solid lines) with defined p and aC (see text). A . No grazing hut variahle extent of the mixed layer (75, 40 and 25 m). B . No grazing but variahle initial ratio between biomass of P . pouchetii and diatoms (P,,/D,) = 0.5; 2.0; 4 . 0 ) . C. Grazing by 2000 adult female Calanus finmarchicus with variable preference towards diatoms ( E = -0.2: 0.4: 0.9). dominance of P . pouchetii colonies in high con- centrations. Preliminary studies in the laboratory (Eilertsen unpubl. results) indicate that t h e growth characteristics of P . pouchetii may be more competitive than shown by Verity e t al. (1991 this volume). W e have therefore simulated the succession with three alternative values for p and & for the colonial form of P . pouchetii (Fig. 8A and 8 ) where the lowest values represent those given by Verity e t al. (1991). T h e simulation assumes a slight grazing selectivity towards dia- toms (E = 0.2) from 2000 copepods. T h e simu- lation results show that, under the conditions given, P . pouchetii can hardly maintain a high dominance over diatoms with the established growth parameters. An increase in p or & changes this situation considerably. and with a Copepod grazing in the Barents Sea 351 A 100 - 3 75 - E 0 E" 50 - Fig 8 Simulated population succession (mg C mg-') in a community with diatoms (broken lines) and Phaeocystrs pouchetii (solid lines), mixed depth = 40 m and grazing by 2000 adult females Calanus - 25 - -_.__ l o o - 8 0 75 finmarchicus with E = 0 2 (slight preference towards 0 diatoms) A Variable F W - E growth parameter p (0 22, 0 33, 0 46) for the colonial forms of P pouchetri. B Variable growth parameter d (0 0 0 0 3 . 0 ooo6, 0 Oaos) for the colonial forms of P pouchetii March April 25 - combined increase of both parameters the poten- tial for a rapid development of a heavy bloom of P . pouchetii colonies is given. The simulation results indicate that vertical stratification and grazing by zooplankton are vari- ables of great significance for the numerical bal- ance between diatoms and P. pouchetii. The occurrence of two different modes of P. pouchetii and lack of knowledge about the mechanisms involved in the production of these forms com- plicate the picture. The small solitary flagellates have a size of 3-8 pm (Tande & BQmstedt 1987) which indicate low retention efficiency for cope- pods (cf. Berggren et al. 1988). Microzooplankton can graze upon these with high efficiency (Johnsson 1986), whereas the macrozooplankton grazing on Phaeocysris colonies still is con- troversial (see above). Therefore, identification of the main micrograzers and quantitative measurements of microzooplankton grazing on microflagellates is an area which should render high priority in future ecological investigations in the Barents Sea. Furthermore, the importance of specific properties of P. pouchetii also needs much more attention. This includes a positive buoyancy of especially large colonies of P. pouchetii (Skres- let 1988), inhibitory effects on its own growth in dense populations (Kayser 1970) and release of dissolved organic matter (Batje & Michaelis 1986) that may affect growth of microalgae. None of these aspects have been included in our model formulation. Acknowledgements. - We thank all the colleagues that have contributed in the field work and laboratory work during the cruises, experiments and analyses from which the results in this paper were produced. We thank P. Verity for constructive critique. We also thank the organisers of the research program, the Norwegian Fisheries Research Council and the Ministry of the Environment for the very stimulating period with Pro Mare. References BAmstedt, U . 1984: Die1 variation in the nutritional physiology of Calanusglacialis from lat. 18"N in the summer. Mar. Biol. BAmstedt, U . 1986: Chemical composition and energy content. Pp. 1-58 in Corner E. D. S. & O'Hara S. C. M . (eds.): The biological chemistry of marine copepods. Oxford University Press, New York. BAmstedt. U. 1988: Ecological significance of individual varia- bility in copepod bioenergetics. Hydrobiologia 1671168. 43- 59. BAmstedt. U . & Ervik, A . 1984: Local variation in size and activity among Calanus finmarchicus and Metridia longa (Copepoda. Calanoida) overwintering on the west coast of Norway. J. Plankton Res. 6 , 843-857. BAmstedt, U . & Tande, K . S. 1985: Respiration and excretion rates of Calanus glacialis in arctic waters of the Barents Sea. Mar. Biol. 87, 259-266. 79, 251-261. 352 U . Bimstedt et al. Bimstedt, U & Tande. K . S 1Y88: Physiological responses of Calanus finmarchicus and Merridia longa (Copcpoda: Cal- anoida) during the wintcr-spring transition. Mar. Biol. Y9. Bimstedt, U.. Tande. K . S & Nicolajscn. H 1985: Ecological investigations on the zooplankton community of Balsfjorden. northern Norway: physiological adaptations in Merridta longa (Copepoda) to the overwintering period. Pp. 313-327 in Gray. J . S . & Christiansen. M. E. (eds.): Marine biology of polar regions and effects of stress on marine organums. J Wiley and Sons. Batje. M. & Mtchaelis. H. 1986: Phaeocvsris poucherii blooms in the east Frisian coastal waters (German Bight. North Sea). Mar. Biol. 93. 21-27. Berggren. U . . Hansen. B . . Ki0rboe.T. 1988. Food-size spectra. ingestion and growth of the copepod Acarria fonsa: impli- cations for the determination of copepod production Mar. Biol. 99. 341-352. Boyd. C. M.. Smith, S . L. & Cowles. T . J . 1980: Grazing patterns of copepods in the upwelling system of Peru. Limnol. Oceanogr. 24. 583-596. Claustre. H.. Poulet. S . A , . Williams. R . . Marty. J . C . Coombs. S.. Ben. Mlih. F.. Hapette, A . M. & Martin- Jezequel. M. 1990: A biochemical investigation of a Phaeo- cvsris sp. bloom in the Irish Sea. 1. Mar. Biol. Ass. UK 70. 197-207. Conover, R. J . & Corner. E . D. S . 1%8: Respiration and nitrogen excretion by some marine zooplankton in relation to their life cycles. 1. Mar. Biol. Ass. UK 48. 4!+75. Conover, R. J . & Francis, V. 1973: The use of radioactive isotopes to measure the transfer of materials in aquatic food chains. Mar. Biol. 18. 272-283. Dagg, M. J . & Grill, D. W. 1980: Natural feeding rates of Cenfropages rypicur females in the New York Bight. Limnol. Oceanogr. 25. 597-609. Dagg. M. J . , Vidal. J.. Whitledge, T. E . . lverson. R . L. & Goering, J . J . 1982: The feeding, respiration and excretion of zooplankton the Berings Sea during a spring bloom. Deep- Sea Res. 29. 45-63. Dagg, M. J. & Wyman. K. D. 1983: Natural ingestion rates of the copepods Neocalanur plumchrus and N . crisfarus cal- culated from gut contents. Mar. Ecol. Prog. Ser. 13. 37-46. Eilertsen. H. Chr., Tande. K. S. & Hegseth. E. N. 1989a: Potential of herbivorous copepods for regulating the spring phytoplankton bloom in the Barents Sea. Rapp. P . - o . Reun. Cons. lnt. Explor. Mer 188. 156163. Eilertsen. H . Chr., Tande, K. S . & Taasen. J. P. 1989b: Vertical distribution of primary production and grazing by Calanus glacialis Jasehnov and C. hyperboreur Krayer in Arctic waters (Barents Sea). Polar Biol. 9. 25%260. Estep, K. W.. Nejstgaard. J . Chr., Skjoldal, H . R. & Rey. F. 1990: Predation by copepods upon natural populations of Phaeocysris poucherii as a function of the physiological state of the prey. Mar. Ecol. Prog. Ser. 6 7 . 23S249. Frost. B. W . 1972: Effect of size and concentration of food particles on the feeding behaviour of the marine planktonic copepod Calanus pacilfcus. Limnol. Oceanogr. 17. 8 0 5 8 1 5 . Hansen. 8.. Berggreen. U . C . . Tande. K. S . & Eilertsen. H. Chr. 1990a: Post-bloom grazing by Calanus glacialis, C. finmarchicus and C. hyperboreus in the region of the polar front. Barents Sea. Mar. Biol. 104, 5-14. Hansen. B.. Tande. K. S . & Berggreen. U. C . 1990b: On the trophic fate of Phaeocystis poucherii (Hariot), I l l . Functional 31-38. anus finmarchicus (Copepoda) fed diatoms and Phaeocvsris. J. Plankton Res. 12. 1173-1187. Head. E . J . H . . Bedo. A . & Harris. L. R . 1988: Grazing, defecation and excretion rates of copepods from inter-island channels of the Canadian Arctic archipelago. Mar. B i d . 99, 333-340. Head. E . J . H.. Harris. L. R . & Abou Debs, C . 1985: Effect of daylength and food concentration on in situ diurnal feeding rhythms in Arctic copepods. Mar. Ecol. Prog. S e r . 24, 281- 288. Hirche. H.-J. 1990: Egg production of Calanusfinmarchicus at low temperature. Mar. B i d . 106. 53-58. Huntley. M . 1981: Nonselective. nonsaturated feeding by three calanid copepod species in the Labrador Sea. Limnol. Oce- anogr. 26, 831-842. Huntley, M. 1988: Feeding biology of Calanus: a new perspec- tive. Hydrobiologia 167/168. 83-99. Huntley. M.. Tande. K. & Eilertsen. H . Chr. 1987: On the trophic fate of Phaeocysris poucherii (Hariot). 11. Grazing rates of Calanus hyperboreus ( K r ~ y e r ) on diatoms and dif- ferent size categories of Phaeocysris poucherir. J . Exp. Mar. Biol. Ecol. 110. 197-212. Jonsson, P. R . 1986. Particle size selection, feeding rates and growth dynamics of marine planktonic oligotrichous ciliates (Ciliophora: Oligotrichina). Mar. Ecol. Prog. Ser. 33, 265- 277. Jonsson. P. & Tiselius, P. 1990: Feeding behaviour, prey detec- tion and capture efficiency of the copepod Acarria ronsa feeding on planktonic ciliates. Mar. Ecol. Prog. Ser. 60. 35- 44. Kayser. H. 1970: Experimental-ecological investigation on Phaeocysris poucherii (Haptophyceae): Cultivation and waste water test. Helgolander Wiss. Meeresunrers. 20, 195-212. K10rboe. T.. MBhlenberg. F. & Nicolajsen, H . 1982: Ingestion rate and gut clearance in the planktonic copepod Cenrropages hamarus (Lilljeborg) in relation to food concentration and temperature. Ophelia 21, 181-194. Kirk. J. T. 0. 1983: Lighf and phorosynrhesis in aquatic eco- s y s t e m . Cambridge University Press, Melbourne. Mackas. D. & Bohrer, R. 1976: Fluorescence analysis of zooplankton gut contents and an investigation of die1 feeding patterns. J. Exp. Mar. Biol. Ecol. 25. 77-85. Nielsen, T. G . . Kiarboe, T. & Bjarnsen. P. K. 1990: Effects of a Chrysochromulina polylepis subsurface bloom on the planktonic community. Mar. Ecol. Prog. Ser. 62, 21-35. Parsons. T . R., Takahasi, M. & Hargrave. B. 1977: Biological oceanographical processes. Pergamon Press, Oxford. 332 pp. Peterson. W. T. 1988: Rates of egg production by the copepod Calanus marshallae in the laboratory and in the sea off Oregon, USA. Mar. Ecol. Prog. Ser. 17. 229-237. Sakshaug. E. & Slagstad. D. 1991: Light and productivity in polar marine ecosystems: a physiological view. Pp. 6%85 in Sakshaug. E.. Hopkins. C. C . E. & Oritsland. N. A . (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology. Trondheim. 12-16 May 1990. Polar Research I O ( 1 ) Schnack, S . B.. Smetacec, V., von Bodungen. B. & Stegmann. P. 1985: Utilization of phytoplankton by copepods in Ant- arctic waters during spring. Pp. 65-81 in Gray, J. S. & Christiansen. M. E. (eds.): Marine Biology of polar regions and effecrs of stress on marine organisms. J . Wiley and Sons. Skreslet. S . 1988: Buoyancy in Phaeocysris pouchetit (Hariot) Lagerheim. J . Exp. Mar Biol. Ecol. 119, 157-166 responses in grazing demonstrated on juvenile stages of c a b Slagstad. D . 1981: Modeling and simulation of physiology and Copepod grazing in the Bnrents Sea 3 5 3 population dynamics of copepods. Modeling, fdentificafion and C o n f r o l 2 , 119-162. Slagstad, D. & Tande. K . S. 1990: Growth and production dynamics of Calanus glacialis in an arctic pelagic food web. Mar. Ecol. Prog. Ser. 63? 189-199. Tande. K. S . & Blmstedt, U . 1985: Grazing rates of the cope- pods Calanus glacialis and C . finmarchicus in Arctic waters of the Barents Sea. Mar. B i d . 87, 251-258. Tande. K . S . & BBmstedt. U . 1987: On the trophic fate of Phaeocysfis pouchefii. 1. Copepod feeding rates on solitary cells and colonies of P. poucherii. Sarsia 72. 312-320. Valiela. I . 1984. Marine ecological processes. Springer Verlag. New York. 546 pp. Verity. P. & Smayda. T. J . 1989: Nutritional value o f Phaeocystis pouchetii (Prymnesiophyceae) and other phy- toplankton for Acarfia spp. (Copepoda): ingestion, egg pro- duction, and growth of nauplii. Mar. Biol. 100. 161-171. Verity, P. G . . Smayda, T. J . , Sakshaug. E. 1991. Photo- synthesis. excretion, and growth rates of Phaeocysfis colonies and solitary cells. Pp. 117-128 in Sakshaug. E.. Hopkins. C. C. E. & Britsland. N . A. (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology. Trondheim. 12-16 May 1990. Polar Research l O ( 1 ) . Webb, W. L.. Newton, M., Starr, D. 1974. Carbon dioxide exchange of A l n m rubra: a mathematical model. Oecologia 17. 281-291.