Plankton community structure and production along a ... - Springer Link

Marine Biology (2002) 141: 707–724 DOI 10.1007/s00227-002-0868-8

T.G. Nielsen Æ C.M. Andersen

Plankton community structure and production along a freshwater-influenced Norwegian fjord system

Received: 1 March 2002 / Accepted: 8 May 2002 / Published online: 26 June 2002 Springer-Verlag 2002

Abstract Plankton community structure and production was investigated along a freshwater-influenced Norwegian fjord system, from the innermost branch, across the sill, to more open waters. The water column shifted from being highly stratified, with a brackish surface layer (6 psu) separated from >30 psu water by a strong pycnocline, to a more open system outside the front. The zooplankton community in the brackish surface layer was dominated by ciliates, heterotrophic dinoflagellates, the cladoceran Evadne nordmanni and the copepod Eurytemora affinis. In and below the pycnocline and off the sill the diversity of the zooplankton increased, and 20 copepod and 36 protozoan taxa were observed. Mesozooplankton was sampled both with a WP-2 net (mesh size 200 lm) and with 10 l Niskin bottles. The biomass from the net samples was only 35–72% compared to the estimates from the Niskin bottles. Especially copepod nauplii, Oithona spp. and Microsetella norvegica were severely underestimated. Calanus finmarchicus dominated the copepod community. However, the bottle sampling demonstrated that also Oithona spp. and M. norvegica contributed significantly to copepod biomass and production. Protozooplankton contributed with 54–87% of the total zooplankton production and of potential grazing along the fjord, stressing the importance of the smaller grazers in the food web. A simple modeling approach shows that overall exchange processes between the fjord and coastal waters and in-fjord production of copepods are equally important forces determining in-fjord copepod biomass. In the surface layer, however, only E. affinis, E. nordmanni and the

Communicated by L. Hagerman, Helsingør T.G. Nielsen (&) Æ C.M. Andersen National Environmental Research Institute, Department of Marine Ecology, Frederiksborgvej 399, P.O. Box 358, 4000 Roskilde, Denmark E-mail: [email protected]

protozooplankton had growth rates that could compensate for advective loss.

Introduction Fjord systems are important sites in the carbon and nutrient dynamics at the land–sea interface, and here large quantities of nutrients are transported and incorporated into organic matter (Burrell 1988). Fjords are often highly influenced by freshwater (Farmer and Freeland 1983; Burrell 1988), and, although the direct impact of freshwater runoff is restricted to the upper few meters of the water column, it has major consequences for the physical and chemical oceanography by modulating stratification, nutrient input and current systems, as well as for the vertical distribution and production of plankton (Farmer and Freeland 1983; Skreslet 1986). Norwegian fjords have received much attention concerning their physical and biological oceanography, and large variations in plankton biomass and composition between the different fjord systems have been documented (Fosshagen 1980; Skjoldal et al. 1995; Falkenhaug 1997, and references therein). Numerous studies have dealt with the spatio-temporal distribution of the larger zooplankton, copepods (Magnesen 1988; Falkenhaug et al. 1997), larger crustaceans (Fossa˚ and Brattegard 1990; Onsrud and Kaartvedt 1998) and gelatinous plankters (Falkenhaug 1997), but fewer studies have focused on zooplankton production, and only rarely have small copepod species, copepod nauplii and protozooplankton been considered. Studies on the trophodynamic role of copepods in Norwegian fjords have traditionally focused mainly on the large Calanus species (Aksnes and Magnesen 1983; Tande 1991). The bias towards large copepods stems from previous undersampling of the smaller copepod species and nauplii by coarse plankton nets like the 180-lm-mesh Juday or 200-lm-mesh WP-2 net. Sampling with finer plankton nets gives a different impression of the zooplankton

708

community (Nichols and Thompson 1991; Hopcroft et al. 1998). Concerning the protozooplankton, few studies have previously attempted to assess their importance to the trophodynamics in Norwegian fjords (Verity and Vernet 1992; Archer et al. 2000), and, to our knowledge, only Nejstgaard et al. (1997) have concurrently evaluated the role of protozoo- and mesozooplankton. The present investigation was conducted during summer in a typical silled fjord system, Sandsfjord, in western Norway. This fjord system is well described due to a monitoring program initiated in connection with the construction of the largest hydroelectric power plant in Norway (Fosshagen 1980; Lie et al. 1992); the effect of power plant operation on the physical oceanography and distribution of the meso- and macrozooplankton is well documented (Kaartvedt and Svendsen 1990, 1995; Kaartvedt and Aksnes 1992). Prior to the present study, however, the power plant had been out of operation for 3 weeks, allowing the pelagic communities to establish in relation to the water column characteristics. The present paper therefore describes the ‘‘undisturbed’’ plankton community structure from the freshwater inlet, across the sill to the more marine influenced outer part of the fjord system. Special emphasis is given to estimates of zooplankton biomass and production in order to evaluate the role of small heterotrophs (protozooplankton and small copepod species) in food web dynamics. Based on a simple modeling approach we also attempt to assess the relative importance of in-fjord production in relation to physical exchange processes.

Fig. 1. Sandsfjorden system, showing the approximate positions of the stations sampled during the transect study 10 July 1995

Materials and methods Study area The survey was conducted onboard the R.V. ‘‘Ha˚kon Mosby’’ (University of Bergen) along the freshwater-influenced Sandsfjord system, situated on the southwestern coast of Norway (Fig. 1) on 10 July 1995. The weather was calm and sunny throughout the study period. The fjord system consists of three branches: the Hylsfjord (max. depth of 510 m), Sandsfjord (max. depth 420 m) and Saudafjord, of which only the two former were investigated. The Hylsfjord, which is the innermost branch, is usually strongly influenced by periodic outlets of freshwater from the power plant situated at the head of the fjord. However, the water column was not disturbed by the power plant during the present study. The Sandsfjord branch, on the other hand, continuously receives freshwater from the Suldalsla˚gen River, located at Sand (Fig. 1), providing a brackish surface layer in the entire fjord system. Towards the ocean, a 110 m deep sill separates the Sandsfjord from more open waters. The five sampling stations were located along a transect following the freshwater plume from the innermost part of the Hylsfjord to the more open waters, with the two outermost stations located off the sill. More detailed descriptions of the topography and hydrography of the fjord system are given by Lie et al. (1992) and Kaartvedt and Svendsen (1995). Sampling Sampling took place from 0600 to 1700 hours, and the stations were visited in the following order: stn 2, 1, 3, 4 and 5 (Fig. 1). Vertical profiles of temperature and salinity were recorded with a Sea Bird CTD system equipped with a rosette sampler with twelve 10-l Niskin water bottles. The in situ chlorophyll a fluorescence in the upper 12 m (including the pycnocline) was measured with a Hardt fluorometer. In situ fluorescence was calibrated against the spectrophotometrically determined chlorophyll a samples. From the vertical structure of the water column it was decided to

709 sample five depths: 1, 4, 7.5, 15 and 30 m for nutrient concentrations, chlorophyll a, proto- and mesozooplankton. The extension of the euphotic zone was estimated from Secchi depth measurements.

logðCmax Þ ¼ 1:491 0:23 logðPvol Þ

Nutrients

ð1Þ

and for heterotrophic dinoflagellates of

Samples for the determination of nitrate and nitrite (NO3– and NO2–), phosphate (PO43–) and silicate (SiO43–) concentrations were deep frozen immediately on board the ship. After arrival at the laboratory, the nutrient concentrations were measured on an automatic nutrient analyzer following Grasshoff (1976). Phytoplankton Samples of 1–2 l were filtered onto GF/F filters within 3 h after collection, extracted in 96% ethanol overnight (Jespersen and Christoffersen 1987), and the chlorophyll a concentration measured spectrophotometrically. The chlorophyll a fraction 10 lm, 300 ml samples were fixed in acid Lugol’s solution (2% final concentration). Samples were kept cold and dark until examination in a Leitz inverted microscope at 400· magnification. Depending on the concentration of organisms, 50 or 100 ml of water was settled for 24 h and counted using the Utermo¨hl technique (Utermo¨hl 1958). At least 400 cells were counted per sample. Identification of ciliates to species, group or morphotype was based on Montagnes and Lynn (1991). Dinoflagellates were identified according to Dodge (1985) and Thomsen (1992). Due to the fixative, the only mixotrophic species that were identified and enumerated separately were Myrionecta rubra and Laboea strobila. Biovolumes were estimated from linear dimensions using appropriate geometric shapes, and converted to biomass using conversion factors of 0.11 pg C lm–3 for ciliates and athecate dinoflagellates and 0.13 pg C lm–3 for thecate dinoflagellates (Edler 1979). No correction due to cell shrinkage was applied. The grazing and production of the protozooplankton was estimated by two different methods: (1) from the community growth

Table 1. Literature used to convert the length measurements of zooplankton to biomass. Carbon biomass was calculated by assumption of a specific carbon content of 40% of dry weight (Parsons et al. 1984) or 46% of ash-free dry weight (Paffenho¨fer and Harris 1976)

rates obtained from incubation of size-fractionated water samples at stn 1 (Andersen and Nielsen, 2002) and (2) from the general equations in Hansen et al. (1997; their Table 9), assuming maximum clearance rates for: ciliates of

logðCmax Þ ¼ 0:851 0:23 logðPvol Þ;

ð2Þ 5

–1

where Cmax is the maximum specific clearance rate (10 h ) at 20C and Pvol is the mean volume of the predator (ciliates and heterotrophic dinoflagellates, respectively). We further assumed that ciliates and heterotrophic dinoflagellates feed on the 11 lm fraction of the phytoplankton biomass, respectively. Ingestion for each sampling depth was temperature corrected by applying a Q10 of 2.8 (Hansen et al. 1997). Production was calculated assigning a gross growth yield of 33% (Hansen et al. 1997).

Mesozooplankton The vertical distribution of the mesozooplankton was investigated by the rosette sampler. Four to five Niskin bottles (40–50 l) were released at each of the selected depths, pooled, and concentrated on a 45 lm net and preserved in 2% buffered formalin (final concentration). Integrated samples were taken using a WP2 net (200 lm mesh size) equipped with a flowmeter (Hydrobios). The net was towed from 30 m to the surface at 0.25 m s–1, and the samples were further concentrated and preserved. In the laboratory, the samples were divided into appropriate subsamples using a modified Kott splitter (Kott 1953), and zooplankton specimens were identified and counted using a dissecting microscope. At least 400 individual copepods were identified to species or genus and classified as nauplii, copepodites, males or females. For each taxon the lengths of the first 40 nauplii, 40 copepodites, 10 males and 10 females in each sample were measured with 20 lm precision. Forty cladocerans meroplankters and appendicularians of each taxon were measured. The species-specific biomass was calculated from the abundance, the length measurements (copepods: prosome length; cladocerans: total length) and length/weight regressions from the literature according to Table 1. The biovolume of rotifers was estimated from linear dimensions (two dimensions measured) assuming the simple geometrical shapes from Ruttner-Kolisko (1977) and converted to biomass using a conversion factor of 0.12 pg C lm–3 (Hansen et al. 1997).

Taxon

Copepodites

Nauplii

Calanus finmarchicus Oithona spp. Microsetella norvegica Acartia longiremis Temora longicornis Pseudocalanus elongatus Microcalanus pusillus Oncaea sp. Eurytemora affinis Paracalanus parvus Centropages hamatus Metridia sp. Cladocerans Appendicularia Bivalvia Cirripedia Polychaeta Gastropoda

Hygum et al. (2000b) Sabatini and Kiørboe (1994) Satapoomin (1999) Klein Breteler et al. (1982) Klein Breteler et al. (1982) Klein Breteler et al. (1982) as Pseudocalanus Satapoomin (1999) Kankaala and Johanson (1986) as Pseudocalanus Klein Breteler et al. (1982) Hirche and Mumm (1992) Kankaala and Johanson (1986) Paffenho¨ffer (1976) Fotel et al. (1999) Rodhouse and Roden (1987) Hansen (1993) Hansen and Ockelman (1991)

Hygum et al. (2000a) Sabatini and Kiørboe (1994) As Oithona Klein Breteler et al. (1982) Klein Breteler et al. (1982) as Calanus as Calanus Not identified Kankaala and Johanson (1986) as Calanus Klein Breteler et al. (1982) Not identified

710 Copepod production The production of copepods was determined by three independent methods: (1) the egg production method, where it is assumed that the specific egg production rate of adult females corresponds to growth rates in the juvenile stages (Berggreen et al. 1988); (2) the general equation of Hansen et al. (1997); and (3) the equation of Huntley and Lopez (1992), in which copepod productivity is dependent on temperature only.

assume that copepod food items are heterotrophic dinoflagellates, ciliates and phytoplankton >11 lm, and that all food items were cleared at maximum rate. The factor converting copepod biomass to biovolume was 0.12 pg C lm–3 (Hansen et al. 1997). Production was calculated assigning a gross growth yield of 33% (Hansen et al. 1997).

Method 3 Finally, the copepod production was calculated following Huntley and Lopez (1992):

Method 1

P ¼ B 0:045 e0:111T Copepods for egg production experiments were collected by vertical tows with the WP-2 net equipped with a large, non-filtering codend. The depth of fluorescence maximum was included in the samples. On deck, the content of the cod-end was transferred to a large, insulated plastic container and diluted in 25 l of surface water (water from below the pycnocline at stns 1 and 2). Within the next 2 h, three to five females were transferred to 500-ml polycarbonate bottles, containing 45-lm-screened surface water. Three to six replicate bottles were incubated in running surface water for 24 h. Every 2 h the bottles were gently rotated manually. At the end of the experiments the spawned eggs were counted and the length of females and diameter of the eggs were measured (with 20 lm and 1 lm precision, respectively). Egg carbon was estimated from egg volume by assuming a conversion factor of 0.14 pg C lm–3 (Huntley and Lopez 1992), and the weight-specific egg production rate (SEP) was calculated. Unfortunately, at the two innermost stations the surface salinity was so low that all freespawning copepods had died when the net was on deck so no egg production experiments could be conducted. Instead the egg production rates from stn 3 or 4 (Calanus) were used to calculate production at stns 1 and 2. Temora longicornis did not produce eggs during any of the experiments, but since we do not believe that this result expressed the actual in situ egg production of Temora we applied the average SEP of Acartia and Centropages to calculate the production of this species. The same assumption was used to calculate the production of Microcalanus pusillus and Paracalanus parvus. SEP of the four dominant egg-carrying copepods, Eurytemora affinis, Oithona spp., Microsetella norvegica and Pseudocalanus elongatus, was calculated from the egg hatching rate (HR) at in situ temperature, the ratio of eggs to females (E/F) found in the fixed Niskin bottle samples, and the carbon content of eggs (eggC) and females (femC), respectively, according to: SEP ¼ HR ðE=FÞ ðeggC=femCÞ

ð3Þ

Additional E. affinis females for determination of the egg/female ratio were collected with the WP-2 net by taking horizontal hauls (10 min at 1 knot speed) in the brackish surface water at each station. The hatching rate of E. affinis was determined in temperature-controlled experiments (Andersen and Nielsen 1997). Other equations relating egg hatching rate to temperature were found in the literature, e.g. for Oithona spp. in Nielsen et al. (2002) and for P. elongatus in Corkett and McLaren (1970). In the case of M. norvegica, adult females could not readily be discriminated from older copepodite stages and no information about the egg hatching time was available. Therefore, a length criterion was applied (the smallest M. norvegica carrying eggs was 440 lm, so we assumed that all specimens of 440 lm were adult females) and the hatching rate of Oithona was used.

ð5Þ

where P is the production per day, B is the copepod biomass and T is the temperature in degrees Celsius. Cladoceran production As for the copepods the production of cladocerans was determined in three different ways. Method 1 Instead of SEP, the weight-specific neonate production (SNP) was determined. SNP was calculated from the number of embryos per female (E/F), the embryonic development rate (DR) and the carbon content of neonates (neoC) and females (femC), respectively: SNP ¼ DR ðE=FÞ ðneoC=femCÞ

ð6Þ

The number of embryos per female was calculated from the average brood size recorded from the fixed samples (discrete and depthintegrated samples pooled) and the percentage of pregnant females. The carbon content of neonates was calculated assuming a neonate length of about 300 lm equaling a carbon content of about 0.5 lg. Embryonic development times from Albjerg (1996) were converted to in situ temperatures assuming a Q10 of 2.8 (Hansen et al. 1997). For Podon, the SNP of Evadne was used. For calculation of ingestion a gross growth yield of 33% was assumed (Hansen et al. 1997).

Method 2 Hansen et al. (1997) presented the following equation for calculating cladoceran clearance: logðCmax Þ ¼ 0:656 0:23logðPvol Þ

ð7Þ

We assumed that the cladocerans cleared heterotrophic dinoflagellates, ciliates and both size fractions of phytoplankton at maximum rate. The factor converting cladoceran biomass to biovolume was 0.12 pg C lm–3 (Hansen et al. 1997). Method 3 Huntley and Lopez (1992) suggested that their general equation for calculating copepod production might be applicable to other zooplankton groups. We tested this hypothesis by using their equation to calculate cladoceran production.

Method 2

Production of appendicularians, meroplankton and rotifers

Hansen et al. (1997) presented the following equation (their Table 9):

The production of appendicularians was calculated according to Uye and Ichino (1995). Production of rotifers and meroplankton was calculated on the basis of general equations in Hansen et al. (1997) assuming maximum clearance rate. Potential food for rotifers was assumed to be phytoplankton. Potential meroplankton food was heterotrophic dinoflagellates, ciliates and phytoplankton.

logðCmax Þ ¼ 1:575 0:23logðPvol Þ

ð4Þ

where Cmax is the maximum specific clearance rate of the copepods (105 h–1) at 20C and Pvol is the mean volume of the copepods. We

711

Results Hydrography and nutrients At the inner station, the water column was strongly stratified with a lens of brackish water comprising the upper 2 m, but along the fjord the surface layer was gradually mixed up in the upper 10 m of the water column (Fig. 2A). Consequently, the surface salinity increased from 6 psu at stn 1 to 30 psu at stn 5 (Fig. 2A). The temperature followed the halocline with 14C at the surface, decreasing to 7–8C below the pycnocline (Fig. 2B). The vertical distribution of the nutrients followed the physical water column characteristics (Fig. 2C–E); the surface water had high concentrations of nitrate (max. 3.4 lM; Fig. 2D) and silicate (max. 4.6 lM; Fig. 2E). In the pycnocline, the concentration decreased, followed by increasing concentrations below the pycnocline. Phosphate concentration was low in the surface layer (0.2 lM; Fig. 2C), but increased below the pycnocline. Along the transect, the surface layer was stripped of all three major nutrients. Based on the Secchi depth measurements, the extension of the photic zone was estimated to be 30 m. Phytoplankton The phytoplankton distribution was characterized by a subsurface bloom, especially at the three innermost stations, where chlorophyll a concentrations reached 6 lg l–1. Outside the front, the phytoplankton was more evenly distributed in the euphotic zone (Fig. 2F). The phytoplankton community was generally dominated by cells of the 11 lm) to the total phytoplankton biomass increased along the transect from about 15% inside the sill to 40% at stn 5 (see Fig. 6A). This size class consisted almost entirely of dinoflagellates and the diatom genus Chaetoceros (Havskum and Hansen 1997). The integrated phytoplankton biomass was 2200–2600 mg C m–2 inside and just off the sill (stns 1–4; see Fig. 6A). At the more open station (stn 5), the biomass decreased to about 1700 mg C m–2. Protozooplankton distribution and community structure A total of 23 species or morphotypes of ciliates and 13 taxa of dinoflagellates were identified, and pronounced c

Fig. 2. Vertical distribution along the fjord of: A salinity (psu), B temperature (C), C phosphate (lM), D nitrate+nitrite (lM), E silicate (lM) and F chlorophyll a (lg l–1). Dots indicate the measuring/sampling depths

712

horizontal and vertical differences were observed in the distribution pattern of the two groups (Fig. 3A–E; see also Fig. 6B). In general, the diversity increased with salinity, i.e. both with depth and along the transect.

Ciliates and heterotrophic dinoflagellates contributed evenly to the protozooplankton biomass in and above the pycnocline. Below the pycnocline the relative importance of the dinoflagellates increased. The ciliate biomass was dominated by naked oligotrichs (Strobilidium spp. and Strombidium spp., 10–50 lm), Lacrymaria (15–25 lm) and Balanion sp. (10–30 lm). These groups had their highest biomass in the upper part of the water column, peaking at 10 lg C l–1 and 5–18 cells ml–1 in the frontal area at stn 3 (Fig. 3A). Tintinnid ciliates were dominated by Salpingella sp. (up to 5 cells ml–1). Other less abundant taxa were Eutintinnus tenuis, Parafavella denticulata, Helicostomella subulata and Tintinnopsis spp. The relative importance of tintinnid ciliates increased from the surface down to the tintinnid peak biomass of 8 lg C l–1 observed in association with the subsurface chlorophyll a layer at the three innermost stations (Fig. 3B). Off the sill the biomass of tintinnids decreased. The maximum abundance of tintinnid ciliates along the transect was 0.7–5.5 cells ml–1. The mixotrophic ciliate Myrionecta rubra was only observed off the sill and in relatively low biomass (Fig. 3C). Laboea strobila was rarely observed. The heterotrophic dinoflagellates were dominated by small naked Gymnodinium/Gyrodinium spp. (10–30 lm), the larger Gyrodinium spirale (50–80 lm) and thecates of the genus Protoperidinium, especially P. depressum (80– 160 lm). As for naked ciliates the highest biomass of athecate (7 lg C l–1; Fig. 3D) and thecate (11 lg C l–1; Fig. 3E) dinoflagellates was observed in the upper 5 m of the water column. However, compared with ciliates, the dinoflagellates had much higher biomasses below the pycnocline. Inside of the sill, thecate and athecate dinoflagellates contributed equally to the integrated biomass, but at stn 3 there was a pronounced peak of large thecates (Protoperidinium). Off the sill, athecate dinoflagellates became more important than the thecates. Maximum abundance along the transect of the two groups was 0.5–2 cells ml–1 for the thecates and 7–17 cells ml–1 for the athecates. The total integrated biomass of protozooplankton increased along the inner part of the fjord from 121 mg C m–2 at stn 1 to 268 mg C m–2 on the sill, after which it decreased to 169 and 222 mg C m–2 at stns 4 and 5, respectively (Table 2; see Fig. 6B). Mesozooplankton distribution and community structure

Fig. 3. Vertical distribution of protozooplankton biomass (lg C l–1) along the fjord for: A naked ciliates, B tintinnid ciliates, C Myrionecta rubra, D athecate dinoflagellates and E thecate dinoflagellates. Dots indicate the sampling depths

Copepods dominated the mesozooplankton, irrespective of the sampling method. Nevertheless, sampling with the 200 lm net and the Niskin bottles gave different impressions of the zooplankton community, with several small taxa being severely undersampled by the net (Table 3). Even Calanus finmarchicus was less abundant in the net samples, and the total copepod biomass in the net samples was only 35–72% of the corresponding water bottle samples. Thus, in order to obtain the most correct estimation of the abundance

713 Table 2. Depth (in parentheses) at the stations sampled, and meso- and microzooplankton biomass (mg C m–2) along the transect on 10 July 1995 Taxon

Calanus finmarchicus Oithona spp.b Microsetella norvegica Acartia longiremis Temora longicornis Pseudocalanus elongatus Microcalanus pusillus Oncaea sp. Eurytemora affinis Paracalanus parvus Centropages hamatus Metridia sp. Total copepods

Stage

CI–CVI Naupliia CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii CI–CVI Nauplii

Overall total Evadne nordmanni Podon sp. Total cladocerans Meroplankton Rotifera Appendicularia Tintinnid ciliates Naked ciliates Total ciliates Athecate dinoflagellates Thecate dinoflagellates Total dinoflagellates Total protozoans Myrionecta rubra a

Station and depth 1 (287 m)

2 (260 m)

3 (145 m)

4 (699 m)

5 (370 m)

147 25 82 55 50 9.1 24 1.0 14 11 8.0 – 19 – 4.6 – 2.1 0.6 0.6 – 0.7 0.2 0.0 – 351 102 453 12 0.7 13 29 0.8 0.1 12.7 21.9 34.5 42.3 44.7 86.9 121 3.6

457 42 77 38 57 15 39 1.0 16 4 18 – 14 – 4.9 – 2.6 0.8 1.5 – 0.0 0.2 0.0 – 687 101 788 11 0.1 11 45 4.3 0.0 36.9 28.6 65.5 83.0 79.9 163 228 1.6

270 42 86 37 44 7.6 30 8.8 19 15 6.6 – 1.3 – 0.8 – 0.0 0.8 2.4 – 0.0 1.0 1.0 – 460 111 571 5.4 0.9 6 15 0.2 2.5 27.3 106 133 42.7 91.5 134 268 2.2

916 80 55 27 16 1.2 7.4 6.3 12 19 19 – 2.2 – 0.3 – 0.2 0.2 2.4 – 0.4 1.3 3.6 – 1035 136 1170 4.4 0.6 5 41 0.0 3.2 10.5 71.6 82.1 55.6 31.1 86.7 169 6.3

1566 90 51 19 1.8 0.2 8 21 49 23 10 – 13 – 0.4 – 0.0 0.0 3.4 – 2.8 7.5 16.7 – 1727 161 1882 19 2.5 22 87 0.0 5.3 18.2 53.9 72.2 115 35.7 150 222 23

Calanus, Pseudo-, Micro- and Paracalanus O. similis, O. nana and O. plumifera

b

and biomass of small copepods, copepod nauplii and meroplankton the following results are based on the water bottle samples. A total of 20 copepod taxa were identified (Table 2, most uncommon species not presented) with Oithona spp. (dominated by O. similis) and Microsetella norvegica being the two most abundant copepods. The former constituted about 60% of both nauplii and older stages at all stations (abundance data not shown). The highest total abundance of nauplii (47–73 ind. l–1) and copepodites (20–44 ind. l–1) was generally found in the pycnocline below the brackish surface layer. Off the sill, where the surface salinity increased, the highest abundance of nauplii was found in the surface water. In terms of biomass, C. finmarchicus was the most important cope-

pod species, but Oithona spp. and M. norvegica also contributed significantly, and the three taxa together constituted 81–94% of the copepod biomass along the transect (Table 2). C. finmarchicus (mainly copepodite stages CIV and CV) gradually became more abundant and more deeply distributed outwards along the transect (Fig. 4A), and there was a tenfold increase in the integrated biomass of this species (147–1566 mg C m–2; Table 2). Calanus accounted for 88% of the copepod biomass at the outermost station. The increase in Calanus nauplii biomass was less pronounced, and in contrast to the older stages the distribution of nauplii moved towards the surface along the transect (Fig. 4A). The integrated biomass of Oithona spp. and M. norvegica was constant

714 Table 3. Abundance of copepodites and adults of some small-size copepod species and of Calanus (ind. m–2) in Niskin water bottles and in WP-2 net samples

Station

Sampling

Oithona

Microsetella

Microcalanus

Oncaea

Calanus

1

Niskin WP-2 Niskin WP-2 Niskin WP-2 Niskin WP-2 Niskin WP-2

230814 34300 207456 41218 257942 24632 143213 38380 130690 48332

119849 0 127170 0 96253 0 33641 0 3695 0

25712 5000 24169 3600 4084 133 2063 600 21563 7000

12249 400 9.362 400 2.156 267 925 0 1.068 0

3967 6300 16521 8800 6448 2133 8943 7600 13174 10333

2 3 4 5

from stns 1 to 3, decreasing off the sill (Table 2). Oithona adults and copepodites accumulated at the pycnocline, while the nauplii were more evenly distributed, with a larger proportion of the biomass found deeper in the photic zone (Fig. 4B). M. norvegica had its peak abundance deeper than Oithona, and the nauplii mirrored the distribution of the older stages (Fig. 4C). The integrated biomass of Acartia longiremis remained unchanged along the transect, but there was a pronounced shift in the relative importance of nauplii and older stages (Fig. 4D; Table 2). Temora longicornis biomass increased along the transect, and nauplii contribution to the biomass varied between 21% and 61%. Like Oithona, A. longiremis and T. longicornis accumulated in the pycnocline. Eurytemora affinis copepodites were present only in the brackish surface water at the two innermost stations and in very low numbers at stn 3 (Fig. 4F). A few nauplii were recorded in the surface water at stn 4. Nauplii contributed with 25% to the total Eurytemora biomass. Pseudocalanus elongatus and Paracalanus parvus accumulated in the upper part of the euphotic zone, Pseudocalanus throughout the transect (Fig. 5A), Paracalanus mainly off the sill (Fig. 5C), while Microcalanus pusillus and Metridia sp. peaked in the deeper part of the euphotic zone. Inside and off the sill, respectively (Fig. 5B, D). The integrated copepod biomass increased along the transect from 453 mg C m–2 at stn 1 to 1882 mg C m–2 at stn 5 (Table 2). Nauplii contributed between 23% and 9% to the total copepod biomass (decreasing along the transect). Cladocerans were dominated by Evadne nordmanni, with Podon sp. being much less abundant. Most of the cladocerans were found in the upper part of the photic zone, but at stn 2 there was a patch of E. nordmanni in the 15 m sample (Fig. 5E). The total biomass of cladocerans varied from 5 to 22 mg C m–2 (Table 2), with no consistent trend along the transect. Meroplankton (mainly bivalve and gastropod veligers, polychaete larvae and cirriped nauplii) was found mostly in the upper 10 m of the water column (Fig. 5F), and the biomass increased from 29 to 87 mg C m2 along the transect (Table 2). The biomass of appendicularians (Oikopleura and Frittilaria) was low, but increased along the transect, reaching 5.3 mg C m–2 at the outermost station (Fig. 5G). Rotifera (Synchaeta spp.) were present only below the pycnocline, inside of the sill,

with the highest biomass (4.3 mg C m–2) found at stn 2 (Fig. 5H). Production of the zooplankton community When growth rates from the size-fractionation experiments (Andersen and Nielsen, 2002) were applied, the integrated protozoan production largely reflected the biomass distribution along the transect (Fig. 6E; see Table 5). Ciliate production was lowest at stn 1 (10 mg C m–2 day–1) and highest at stn 3 (38 mg C m–2 day–1). Integrated P/B-ratios ranged between 0.26 and 0.31 day–1. Dinoflagellate production was of the same magnitude (20–40 mg C m–2 day–1), and P/B-ratios were from 0.23 to 0.29 day–1. The production of Myrionecta rubra was low at the three innermost stations, but reached 9.0 and 34 mg C m–2 day–1 at stns 4 and 5, respectively, corresponding to a P/B-ratio of about 1.5 day–1. Ciliate production calculated from the equations in Hansen et al. (1997) was about twice as high as the production calculated from experimental growth rates, except at stn 5, where it was the same (see Table 5). In contrast to the ciliates, the production of heterotrophic dinoflagellates was 5 to 11 times lower when calculated according to Hansen et al. (1997). Calanus finmarchicus produced 21 eggs female–1 day–1 at stn 4 (Table 4) and was the most important contributor to copepod production at all stations (Table 5). The rate of Acartia longiremis egg production increased along the transect, from 0.4 eggs female–1 day–1 at stn 3 to 17.1 eggs female–1 day–1 at stn 5, where Acartia contributed 4.2% to total copepod production. The egg production rate of Centropages hamatus increased from 10 eggs female–1 day–1 at stns 3 and 4 to 20 eggs female–1 day–1 at stn 5; however, Centropages did not contribute significantly to the total production due to its low biomass. Oithona spp. was the most important egg-carrying copepod, with egg production rates from 1.0 to 1.5 eggs female–1 day–1 (Table 4). The c –1

Fig. 4. Vertical distribution of copepod biomass (lg C l ) along the fjord for: A Calanus finmarchicus, B Oithona spp., C Microsetella norvegica, D Acartia longiremis, E Temora longicornis and F Eurytemora affinis. Left panels: stages CI–CVI; right panels: nauplii. Dots indicate the sampling depths

715

716

Fig. 5. Vertical distribution of mesozooplankton biomass (lg C l–1) along the fjord for: A Pseudocalanus elongatus, B Microcalanus pusillus, C Paracalanus parvus, D Metridia sp., E Evadne nordmanni, F meroplankton, G appendicularians and H rotifers. Dots indicate the sampling depths

Oithona contribution to copepod production was 30% at stn 1, but decreased to 5.4% at stn 5. Eurytemora affinis had the highest egg production rate among the egg-carrying copepods, with a maximum at stn 1 of 12.5 eggs female–1 day–1 (Table 4), equal to a SEP of 0.12 day–1. However, Eurytemora was restricted to the brackish surface layer and its biomass was low, so it contributed only 3% to the copepod production at stn 1. The egg production of Microsetella norvegica was low at all stations, and Pseudocalanus elongatus females with egg sacs were only caught at stns 2 and 5. The total copepod production was 10 mg C m–2 day–1 at stn 1, equivalent to a community P/B-ratio of 0.022 day–1 (Table 5). Along the transect production

increased relatively more than biomass, and was 55 mg C m–2 day–1 at stn 5, equivalent to a community P/B-ratio of 0.029 day–1. At stn 3, the production and P/B-ratio were comparable to values at stn 1. Calculating the copepod production on the basis of the equation in Hansen et al. (1997) yielded results similar to the egg production method (Table 5), but calculating the copepod production according to Huntley and Lopez (1992) gave results four to six times higher. Evadne nordmanni was the most productive cladoceran with a birth rate between 1.1 and 2.1 neonates female–1 day–1 (Table 4), equal to a SNP of 0.13–0.26 day–1. The birth rate of Podon sp. varied between 0.0 and 0.7 neonates female–1 day–1. Both cladoceran taxa were most productive at stn 5, and the production and P/B-ratio were 4.3 mg C m–2 day–1 and 0.20 day–1, respectively (Table 5). The relative importance of cladocerans though, was highest at stn 1, where they contributed 15% of the mesozooplankton production. When using the equation

717 Fig. 6. Integrated biomass (mg C m–2) along the fjord of: A phytoplankton, B protozooplankton, C cladocerans, meroplankton, rotifers and appendicularians, D copepods and E integrated production (mg C m–2 day–1) of copepods, ciliates, heterotrophic dinoflagellates, cladocerans and others. Integrated biomass (mg C m–2), ingestion and production (mg C m–2 day–1) were calculated by trapezoidal integration over the upper 30 m

Female length (lm)

136

568 679

94

478±2 482±1

918±8

Evadne nordmanni

Cladoceran

Acartia longiremis Centropages hamatus Calanus finmarchicus Temora longicornis

n

864±5

71

64

2443±20

728±10

1455

11

923±22

689±2

233

874±3

Free-spawning copepods

Eurytemora affinis Oithona spp. Microsetella norvegica Pseudocalanus sp.

Egg-carrying copepods

Taxon

–

–

150±1.0

70.5±1.3

70.0±1.0

130.0±2.7

56.2±0.4 44.0±0.4

79.8±0.3

Egg diam. (lm)

–

–

10

10

10

14

400 310

351

n

–1

(SNP, (neo. fem.–1 % day–1) day–1) 20.9 1.5 (71)

(neo. fem.–1 day–1) 1.4 (81)

–

–

–

–

–

–

–

–

–

(SEP, (egg fem.–1 % day–1) day–1) – –

(egg fem.–1 day–1) –

1.0±0.4 (2)

1.4±0.1 (45) 0.2±0.0 (98)

4.0±0.8 (67)

–1

0.0±0.0

3.2±0.2 0.5±0.0

11.5±1.3

(SEP, (egg fem. % day–1) day–1)

2

0.0±0.0 (3)

1.5±0.1 (83) 0.4±0.0 (94)

12.5±1.4 (32)

(egg fem. day–1)

1

Station

0.0±0.0 (1)

1.0±0.1 (75) 0.1±0.0 (75)

–

–1

–

–

10.4 (1)


–

–

–

(SEP, (egg fem.–1 % day–1) day–1) – 0.4±0.2 (7)

2.4±0.9

3.0±0.2 0.3±0.0

3.6±0.7


3

0.0±0.0 (2)

1.4±0.1 (78) 0.1±0.0 (58)

–

–1

0.0±0.0 (2)

21.0 (1)

9.7±2.5 (5)


–

–

6.3

(SEP, (egg fem.–1 % day–1) day–1) 0.3±0.1 2.2±0.4 (8)

0.0±0.0

2.1±0.1 0.1±0.0

–


4

–

–

19.6 (1)

(egg fem.–1 day–1) 17.1±1.8 (8)

0.54±0.0 (3)

1.5±0.1 (58) 0.2±0.1 (18)

–


0.0±0.0

4.6

4.2±1.0

(S EP, %day–1) 1.3±0.2

0.0±0.0

3.1±0.2 0.1±0.0

–

(SEP, (egg fem.–1 % day–1) day–1)

5

(SNP, % day–1) 26.3

–

–

8.1

(SEP, % day–1) 8.9±0.9

1.2±0.0

3.2±0.2 0.2±0.1

–

(SEP, % day–1)

Table 4. Transect study, 10 July 1995. Female length, egg diameter, egg production rate (eggs female–1 day–1) and specific egg production rate (SEP, % day–1) of copepods and female length, birth rate (neonates female–1 day–1) and specific neonate production rate (SNP, % day–1) of the cladoceran Evadne nordmanni (mean±SE) (–, no measurements). Parentheses indicate number of females (egg-carrying copepods and cladocerans), or number of experiments (free-spawning copepods). SEP values of free spawners and Evadne refer to surface temperature. All SEP values were transformed to in situ temperatures throughout the water column by applying a Q10 of 2.8 (Hansen et al. 1997), and the production of each copepod species was calculated from SEP multiplied by the biomass. Ingestion rate was calculated from production, assigning a gross growth efficiency of 33% (Hansen et al. 1997)

718

719 Table 5. Meso- and microzooplankton. Production (mg C m–2 day–1) and ratio of production to biomass (P/B, day–1) along the transect on 10 July 1995 Taxon

Copepods Calanus finmarchicus Oithona spp. Temora longicornis Microcalanus pusillus Eurytemora affinis Microsetella norvegica Oncaea sp. Acartia longiremis Pseudocalanus elongatus Centropages hamatus Metridia sp. Paracalanus parvus Total copepods P/B Productiona Productionb Cladocerans Evadne nordmanni Podon sp. Total cladocerans P/B Productiona Productionb Ciliates P/B Productiona Heterotrophic dinoflagellates P/B Productiona Myrionecta rubra P/B Meroplankton P/B Appendicularians P/B Rotifers P/B a

Station 1

2

3

4

5

5.3 3.0 0.6 0.4 0.3 0.3 0.2 0.0 0.0 0.0 0.0 0.0 10 0.022 11 56

16 3.8 0.5 0.3 0.1 0.3 0.2 0.1 0.4 0.0 0.0 0.0 21 0.027 22 103

7.6 2.9 0.8 0.0 0.0 0.1 0.0 0.1 0.0 0.1 0.0 0.1 12 0.020 20 66

26 2.9 0.8 0.0 0.0 0.0 0.0 0.1 0.0 0.1 0.1 0.1 30 0.025 20 131

43 3.0 4.7 0.5 0.0 0.0 0.0 2.3 0.1 0.7 0.4 0.1 55 0.029 42 211

1.9 0.0 1.9 0.145 0.2 1.9 10 0.284 18 21 0.246 2.0 4.0 1.12 0.5 0.018 0.02 0.233 0.05 0.061

1.6 0.0 1.6 0.142 0.2 1.4 18 0.276 51 40 0.245 5.1 2.0 1.27 1.3 0.028 0.00 – 0.38 0.087

0.6 0.1 0.7 0.109 0.2 1.0 38 0.283 77 39 0.291 3.4 2.2 1.00 0.4 0.025 0.89 0.357 0.03 0.128

0.6 0.0 0.6 0.127 0.1 0.7 21 0.259 41 20 0.233 2.8 9.0 1.43 1.4 0.034 1.16 0.366 0.00 –

3.9 0.4 4.3 0.196 0.4 3.5 22 0.311 23 40 0.264 8.8 34 1.48 2.1 0.024 2.08 0.395 0.00 –

Calculated according to Hansen et al. (1997) Calculated according to Huntley and Lopez (1992)

b

in Hansen et al. (1997), cladoceran production became 4 to 11 times lower compared with the neonate production method (Table 5). Calculating cladoceran production according to Huntley and Lopez (1992) yielded results similar to the neonate production method. The appendicularians had high growth rates (P/B up to 0.4 day–1), so despite the low biomass the production was about 1 mg C m–2 day–1 at stns 3 and 4, increasing to 2.1 mg C m–2 day–1 at stn 5, where it constituted 3.3% of the mesozooplankton production. The production of meroplankton was low due to low growth rates (P/B=0.02–0.03 day–1), and was 0.4–1.5 mg C m–2 day–1 at stns 1–4. At the outermost station, the meroplankton production was 2.1 mg C m–2 day–1, making up 3.3% of the mesozooplankton production. Due to the low biomass, the production of rotifers was low and insignificant.

Discussion Physical oceanography of the fjord High inputs of freshwater to the fjord system creates a pronounced stratification of the water column inside the sill (Fig. 2A, B), and the position of the salinity front observed during this study is relatively constant throughout the year, irrespective of the power plant scheme (Kaartvedt and Svendsen 1990; Lie et al. 1992). Currents in Hylsfjord are mainly driven by local freshwater discharge, and during summer, when the power plant is closed down, this branch acts as a backwater for the outlets of the Suldalsla˚gen River. Consequently, currents are weak and retention time is long in the sur-

720

face layer of Hylsfjord. Despite substantial freshwater output, the overall exchange between the fjord system and coastal water is mainly driven by coastal processes, upwelling that pushes dense water masses above the sill into the fjord or downwelling that pushes water masses into the fjord, in the upper layers underneath the brackish surface layer (Kaartvedt and Svendsen 1995). Nutrients and plankton community structure The distribution of nutrients and phytoplankton is strongly influenced by the water column structure and the freshwater discharge. Subsurface maxima of phytoplankton and low nutrient concentrations, in and below the pycnocline in the Sandsfjord system, during summer, have previously been observed (Erga 1989b; Lie et al. 1992). The supply of N and Si from different freshwater sources is of major importance in the overall nutrient budget in the fjord (Kaartvedt et al. 1991; Lie et al. 1992). Nitrate and silicate is supplied to the fjord system with the freshwater and is therefore constantly renewed, particularly when the power plant is operating, but other sources also contribute. Consequently, the concentration of these nutrients is relatively high in the brackish surface layer, and phytoplankton production in and above the pycnocline is generally phosphate limited (Thingstad et al. 1993). The constant supply of Si probably explains the relatively high contribution of diatoms. Moreover, dominant diatom species in the Sandsfjord system, Chaetoceros wighamii and C. simplex together with Katodinium rotundatum, are well-known brackish water species (Thomsen 1992). The observed increase in the proportion of large phytoplankton cells along the transect in the present study can be explained by the weakening of the pycnocline and more pronounced mixing of nutrients to the photic zone off the sill. These large cells could also be provided from coastal waters by advective transport in the intermediate layer (Erga 1989a). Distribution of zooplankton The water column structure had a strong impact on the zooplankton community structure, and pronounced vertical and horizontal differences in biomass and species composition were observed along the transect. Through the fjord the pelagic ecosystem shifted from a two-compartment system, separated by a the strong pycnocline, to a more open system outside the front, where all marine grazers potentially had access to the upper, more productive part of the photic zone. The fjord system thus offers an excellent test site for examination of the roles and strategies of different groups of zooplankton. The water column in the Sandsfjord system provides a diverse environment in which light, temperature, salinity, food and predation risk change with depth. These factors all influence growth and mortality and hence the magnitude and distribution of zooplankton biomass. In the brackish, upper part of the water column the zooplankton community was dominated by protozoo-

plankton, the copepod Eurytemora affinis and the cladoceran Evadne nordmanni, all known to tolerate low salinities in estuaries (Ackefors 1971; Roddie et al. 1984). Staying in the surface layer most likely minimizes both competition and the risk of predation, since the strong pycnocline probably acts as a physiological barrier, both for marine copepods and for all major copepod predators (planktivorous fish, euphausiids and ctenophores). Moreover, if one takes into account the relatively high biomass of naked ciliates and athecate dinoflagellates in the surface layer, the concentration of potential food for copepods and cladocerans is comparable to that in the pycnocline. Further, if we apply a Q10 value of 2.8, the difference in temperature across the pycnocline corresponds to a doubling of zooplankton clearance and hence of growth rates. The presence of the phytoplankton accumulation in the pycnocline apparently had major consequences for the distribution of zooplankton. Here patches of ciliates, heterotrophic dinoflagellates and mesozooplankton (Figs. 3, 4 and 5) were observed. As the phytoplankton accumulation became less distinct along the transect so did the biomass of zooplankton. Similar distribution patterns have been reported from other stratified ecosystems, after the main blooms have depleted the photic zone for nutrients (Nielsen et al. 1993; Levinsen et al. 1999), and it has been demonstrated in experiments that copepods (Tiselius 1992) and ciliates (Fenchel and Blackburn 1999) can locate and stay in thin layers of high phytoplankton biomass. Since the egg production of coastal copepods is often food limited during summer (Kiørboe and Nielsen 1994), this behavior will favor production. Additionally, higher phytoplankton biomass may provide shelter from visual predators. In and just below the pycnocline the diversity and biomass of mesozooplankton increased. Most prominent were Calanus finmarchicus, Oithona similis and Microsetella norvegica, and the two first species also dominated the surface water off the sill. We found much higher abundances of the small Oithona and Microsetella than previously reported from the Sandsfjord system (Kaartvedt and Svendsen 1995). As the stratification weakened outside the front the highest biomass of the marine copepod species shifted from the pycnocline towards the surface except for Calanus copepodites, for which the peak of the population was located deeper. We did not observe any ripe female C. finmarchicus on of the sill, so the fjord population is probably not maintained there. Another possible explanation could be that adult Calanus had already disappeared from the population and that CIV–CV had begun downward migration. C. finmarchicus and other marine species can be imported with the intrusions of coastal water from outside the fjord (Kaartvedt and Svendsen 1995). However, these intrusions do not propagate into the Hylsfjord, and the Calanus found here were imported with the compensatory current when the power plant was operating. Prior to the establishment of the power plant, the Hylsfjord contained few Calanus, so para-

721

doxically the establishment of the hydroelectric power plant and strong freshwater influence has made this branch more marine (Fosshagen 1980; Lie et al. 1992). In periods such as the present when the power plant is not operating, marine copepod species can survive and reproduce in the fjord system, but as soon as the plant starts to operate these species are entrained in the compensation current, brought to the head of the fjord and killed by osmotic shock in the mixing zone (Kaartvedt and Aksnes 1992). The fjord can therefore be a dead end for the more oceanic species. The distribution pattern of protozooplankton largely followed that of the copepods. An exception was the relatively high biomass of naked ciliates and athecate dinoflagellates in the surface water inside of the sill. Here the protozooplankton community was mainly composed of small (fast-growing) species like the ciliates Lacrymaria and Balanion. Off the sill protozooplankton were more homogeneously distributed, probably because of the higher mixing intensity and the more uniform predation pressure from copepods throughout the water column. The surface layer may act as a refuge for the protozooplankton, since the abundance of mesozooplankton is low. The protozoans that can tolerate the low salinities in the surface water will thus benefit from a lower predation impact caused by copepods and cladocerans and from the higher temperature in the surface layer. These obvious benefits should be balanced by the unstable nature of the surface layer. Paradoxically, the distribution pattern of protozooplankton and nauplii largely mirrored that of their potential predators (copepodites) (Fig. 7). The high predation risk in and just below the pycnocline was evidently overshadowed by the benefits of nutritional food and higher temperature. The estimated total copepod clearance potential (2–29% day–1, Fig. 7C) cannot harvest the daily protozoan production (22–50% day–1, Fig. 7B). In the brackish surface layer the copepods graze only about 5% of protozoan production. In the pycnocline the copepods grazed from 23% to 74% of the protozoan production. The most significant grazing impact was below the pycnocline, at the outermost station, where copepods cleared 90% of the daily protozoan production.

Fig. 7. Effects of zooplankton growth and grazing on: A total biomass of protozooplankton (lg C l–1), B specific growth of protozooplankton (% day–1), C copepod clearance potential (calculated from Hansen et al. 1997, % day–1) and D total copepod biomass (lg C l–1)

Production and potential grazing along the fjord Traditionally, investigations of zooplankton dynamics focus mainly on the large, biomass-dominant calanoid copepods. However, due to the general inverse relationship between size and growth rate (Fenchel 1974), productivity and grazing pathways comprising the protozooplankton and the small cyclopoid and harpacticoid copepods may be more important. The protozooplankton biomass contributed with 10–31% of the total zooplankton biomass in the Sandsfjord system, but the protozooplankton was responsible for 54–87% of the cycling of carbon. The two groups of protozooplankton considered here belong to the same size class of plank-

ton, but from a functional point of view they belong to different groups. Ciliates mainly predate on prey 1:10 of their cell volume, while dinoflagellates may have a 1:1 predator/prey ratio (Hansen et al. 1994). Consequently, heterotrophic dinoflagellates compete with their predators, copepods, for food, while ciliates mainly exploit the nanoplankton. In the present study heterotrophic dinoflagellates constituted from 50% to 72% of the protozooplankton biomass. Although numerous studies from temperate (Smetacek 1981; Hansen 1991) and high-latitude (Levinsen et al. 1999; Putland 2000) waters have documented the important role of heterotrophic dinoflagellates in pelagic food web dynamics, the liter-

722

ature is still biased towards the ciliated component of the microzooplankton. Only Paasche and Kristiansen (1982), Nejstgaard et al. (1997) and Archer et al. (2000) have previously reported on heterotrophic dinoflagellates in Norwegian fjords. Being responsible for 27–48% of the total zooplankton grazing impact, the heterotrophic dinoflagellates obviously need to be included in future studies of the plankton community structure and food web dynamics in fjords. The egg production rate of the free-spawning copepods increased towards the sea, which might be caused by a significant increase in the proportion of phytoplankton (>11 lm) biomass accessible to the copepods (Berggreen et al. 1988). The distinct phytoplankton patch in the inner part of the system consisted mainly of smaller forms, not readily grazed by the adult copepods. Moreover, the other potential food resources for the copepods, the protozoans, became increasingly available to the marine copepods along with the weakening of the pycnocline towards the sea. As documented for other marine systems the egg production rate of the egg-carrying cyclopoid copepods was lower and more constant than the co-occurring calanoid copepods (Kiørboe and Nielsen 1994; Nielsen and Sabatini 1996). In contrast, the egg-carrying calanoid copepod Eurytemora affinis had the highest SEP of all copepods, almost comparable to the SNP of the parthenogenetic Evadne nordmanni. For comparison, the production of the copepod community was computed from the temperaturedependent equation formulated by Huntley and Lopez (1992) and from the equations according to Hansen et al. (1997) assuming maximum clearance. The equations by Hansen et al. (1997) gave P/B-rates comparable to those calculated from the egg production experiments (Table 5), indicating that egg production was food limited in the present study. This has also been observed during summer in other stratified waters (Kiørboe and Nielsen 1994). The Huntley and Lopez equation, not considering food-limited production, led to production rates much higher than those obtained with the egg production method, illustrating the importance of considering potential food limitation outside the main bloom periods. The present investigation has demonstrated a range in P/B-ratios of an order of magnitude, from protozooplankton to copepods (Table 5). From a population point of view, one vital question arises when considering life in a perturbed fjord system like the present one.

(m3). The R-ratio indicates the degree of topographic ‘‘openness’’ and differs more than one order of magnitude among fjords (Aksnes et al. 1989; Lie et al 1992). The R-ratio for the Sandsfjord system is 5·10–6 m–1 (Aksnes et al. 1989; Lie et al. 1992). The advective rate (b) and plankton growth rates (r) have the same dimensions (s–1), and the ratio b/r decides which of the two processes dominates within the system. If b/r1 indicates advective dominance. According to Lie et al. (1992), the typical mean current velocity in the Sandsfjord system above sill depth is 10 cm s–1. In that case, the analysis of growth rates of the different zooplankton groups shows that appendicularians, parthenogenetic zooplankton (rotifers and cladocerans) and protist communities are clearly controlled by ‘‘in-fjord’’ production (Fig. 8A). During a similar modeling exercise in another Norwegian fjord, Aksnes et al. (1989) concluded that only phytoplankton could compensate for the advection through growth. However, they only considered zooplankton with growth rates of 5% day–1, i.e. mesozooplankton. In the present

Can local growth rates balance advection? Based on the equations of Aksnes et al. (1989) we estimated the balance between ‘‘in-fjord production’’ and the advective rate, b: b ¼ 0:5 vR ð8Þ R ¼ A=V where v is mean current velocity (m s–1), A is cross section area above the sill (m2) and V is the fjord volume

Fig. 8A, B. Scale analysis of the role of advection relative to zooplankton production. Y-values >1 indicate that advection dominates. The x-axis represents the mean current velocity above the sill. In this analysis, plankton is assumed to be homogeneously distributed above the sill. A Different zooplankton taxa and B copepods (1, Acartia; 2, Microsetella; 3, small free-spawners; 4, Oithona; 5, Calanus; 6, Centropages; 7, Eurytemora)

723

study, the copepod community seems to be equally controlled by ‘‘in-fjord’’ production and advective transport in and out of the fjord system. However, within the copepod community pronounced differences exist (Fig. 8B), and Acartia longiremis and Microsetella norvegica are clearly controlled by advection, while the growth rate of Centropages hamatus is high enough to overrule advection. It is important to note, however, that the model presumes an equal distribution of plankton above sill depth. If part of the copepod community resides below sill depth the advective influence will be less significant than predicted by the model. Organisms residing in the brackish surface layer need to have growth rates or behavior that allows them to compensate for being flushed out of the fjord system. During the period when the power plant is not operating (June–July) downfjord currents in the surface layer of the outer Sandsfjord generally vary between 20 and 30 cm s–1 (Kaartvedt and Svendsen 1995). The growth rate analysis illustrates that at current velocities of ca. 30 cm s–1 appendicularians, the parthenogenetic zooplankton (rotifers and cladocerans) and protists can easily balance the flushing of the surface layer (Fig. 8A). Among the copepods only E. affinis can balance the flushing. This assumption is supported by the results of the present study (Fig. 4F). The model thus indicates that strong flushing of the fjord system will reduce the zooplankton diversity in the surface layer and force the succession towards dominance of fast-growing opportunists such as protozooplankton and parthenogenetic metazoans. In conclusion, our investigation stresses that future studies, even in ‘‘Calanus-dominated’’ ecosystems such as Norwegian fjords, need to include the smaller, fastgrowing components of the zooplankton. The small heterotrophs have a growth potential to withstand flushing of the system, and need to be considered if the carbon and nutrient dynamics of these systems are to be thoroughly assessed and understood. Acknowledgements We thank K. Christoffersen, H. Levinsen, S. Kaartvedt and L. Riemann for commenting on an earlier version of the manuscript, the captain and crew on the R.V. ‘‘Ha˚kon Mosby’’, as well as L. Riemann, W. Martinsen and J.N. Larsen for assistance during the field work. We also thank E.F. Møller for help with sample analysis. The present study was supported by grants from the EU MAST II program, contract MAS2-92-0031‘‘MEICE’’, and The Danish National Research Council grant no. 9801391.

References Ackefors H (1971) Podon polyphemoides Leuckart and Bosmina coregoni maritima (P.E. Mu¨ller) in relation to temperature and salinity in field studies and laboratory experiments. J Exp Mar Biol Ecol 7:51–70 Aksnes DL, Magnesen T (1983) Distribution, development, and production of Calanus finmarchicus (Gunnerus) in Linda˚spollene, western Norway, 1979. Sarsia 68:195–208 Aksnes DL, Aure J, Kaartvedt S, Magnesen T, Richard J (1989) Significance of advection for the carrying capacities of fjord populations. Mar Ecol Prog Ser 50:263–274

Albjerg J (1996) Et studie af zooplankton dynamikken i danske farvande. M.Sc. thesis, University of Copenhagen, Copenhagen Andersen CM, Nielsen TG (1997) Hatching rate of the egg-carrying estuarine copepod Eurytemora affinis. Mar Ecol Prog Ser 160:283–289 Andersen CM, Nielsen TG (2002) The effect of a sharp pycnocline on plankton dynamics in a freshwater influenced Norwegian fjord. Ophelia 56 (3):1–26 Archer SD, Verity P, Stefels J (2000) Impact of microzooplankton on the progression and fate of the spring bloom in fjords of northern Norway. Aquat Microb Ecol 22:27–41 Berggreen U, Hansen B, Kiørboe T (1988) Food size spectra, ingestion and growth of the copepod Acartia tonsa during development: implications for determinations of copepod production. Mar Biol 99:341–352 Burrell DC (1988) Carbon flow in fjords. Oceanogr Mar Biol Annu Rev 26:143–226 Corkett CJ, McLaren IA (1970) Relationships between development rate of eggs and older stages of copepods. J Mar Biol Assoc UK 50:161–168 Dodge JD (1985) Marine dinoflagellates of the British Isles. Her Majesty’s Stationary Office, London Edler L (1979) Recommendations for marine biological studies in the Baltic Sea. Publ Baltic Mar Biol 5:1–38 Erga SR (1989a) Ecological studies on the phytoplankton of Boknafjord, western Norway. 1. The effect of water exchange processes and environmental factors on temporal and vertical variability of biomass. Sarsia 74:161–176 Erga SR (1989b) Ecological studies on the phytoplankton of Boknafjord, western Norway. 2. Environmental control of photosynthesis. J Plankton Res 11:785–812 Falkenhaug T (1997) Studies of spatio-temporal variations in a zooplankton community. Interactions between vertical behaviour and physical processes. Dr. scient. thesis, Institute of Marine and Freshwater Biology, The Norwegian College of Fishery Science, University of Tromsø, Tromsø Farmer DM, Freeland HJ (1983) The physical oceanography of fjords. Prog Oceanogr 12:147–219 Fenchel T (1974) Intrinsic rate of natural increase: the relationship with body size. Oecologia 14:317–326 Fenchel T, Blackburn N (1999) Motile chemosensory behaviour of phagotrophic protists: mechanisms for and efficiency in congregating at food patches. Protist 150:325–336 Fossa˚ JH, Brattegard T (1990) Bathymetric distribution of Mysidacea in fjords of western Norway. Mar Ecol Prog Ser 67:7–18 Fosshagen A (1980) How the zooplankton community may vary within a single fjord system. In: Freeland HJ, Farmer DM, Levings CD (eds) Fjord oceanography. NATO conference series IV, marine sciences, vol 4. Plenum, New York Fotel FL, Jensen NJ, Wittrup L, Hansen B (1999) In situ and laboratory growth by a population of blue mussel larvae (Mytilus edulis L) from a Danish embayment, Knebel Vig. J Exp Mar Biol Ecol 233:213–230 Grasshof K (1976) Methods for seawater analysis. Verlag Chemie, Weinheim, Germany Hansen B (1993) Aspects of feeding, growth and stage development by the trochophora larvae of the boreal polychaete Mediomastus fragile (Rasmussen) (Capitellidae). J Exp Mar Biol Ecol 166:273–288 Hansen B, Ockelmann KW (1991) Feeding behaviour in the opisthobranch Philine aperta. I. Growth and functional response at different developmental stages. Mar Biol 111:255–261 Hansen B, Bjørnsen PK, Hansen PJ (1994) The size ratio between planktonic predators and their prey. Limnol Oceanogr 39:395– 403 Hansen PJ (1991) Quantitative importance and trophic role of heterotrophic dinoflagellates in a coastal pelagical food web. Mar Ecol Prog Ser 73:253–261 Hansen PJ, Bjørnsen PK, Hansen BW (1997) Zooplankton grazing and growth: scaling within the 2–2,000 lm body size range. Limnol Oceanogr 42:687–704

724 Havskum H, Hansen AS (1997) Importance of pigmented and colourless nano-sized protists as grazers on nanoplankton in a phosphate-depleted Norwegian fjord and in enclosures. Aquat Microb Ecol 12:139–151 Hirche HJ, Mumm N (1992) Distribution of dominant copepods in the Nansen Basin, Arctic Ocean in summer. Deep-Sea Res 39(Suppl 2): 485–505 Hopcroft RR, Roff JC, Lombard D (1998) Production of tropical copepods in Kingston Harbour, Jamaica: the importance of small species. Mar Biol 130:593–604 Huntley M, Lopez MDG (1992) Temperature-dependent production of marine copepods: a global synthesis. Am Nat 140:201– 242 Hygum BH, Rey C, Hansen BW (2000a) Growth and development rates of Calanus finmarchicus nauplii during a diatom spring bloom. Mar Biol 136:1075–1085 Hygum BH, Rey C, Hansen BW, Tande K (2000b) Importance of food quantity to structural growth rate and neutral lipid reserves accumulated in Calanus finmarchicus. Mar Biol 136:1057–1073 Jespersen AM, Christoffersen K (1987) Measurements of chlorophyll a from phytoplankton using ethanol as extraction solvent. Arch Hydrobiol 109:445–454 Kaartvedt S, Aksnes DL (1992) Does freshwater discharge cause mortality of fjord-living zooplankton? Estuar Coast Shelf Sci 34:305–313 Kaartvedt S, Svendsen H (1990) Advection of euphausiids in a Norwegian fjord system subject to altered freshwater input by hydro-electric power production. J Plankton Res 12:1263–1277 Kaartvedt S, Svendsen H (1995) Effect of freshwater discharge, intrusions of coastal water, and bathymetry on zooplankton distribution in a Norwegian fjord system. J Plankton Res 17:493–511 Kaartvedt S, Johnsen TM, Aksnes DL, Lie U, Svendsen H (1991) Occurrence of the toxic phytoflagellate Prymnesium parvum and associated fish mortality in a Norwegian fjord system. Can J Fish Aquat Sci 48:2316–2323 Kankaala P, Johanson S (1986) The influence of individual variation on length–biomass regressions in three crustacean zooplankton species. J Plankton Res 8:1027–1038 Kiørboe T, Nielsen TG (1994) Regulation of zooplankton biomass and production in a temperate, coastal ecosystem. 1. Copepods. Limnol Oceanogr 39:493–507 Klein Breteler WCM, Fransz HG, Gonzales SR (1982) Growth and development of four calanoid copepod species under experimental and natural conditions. Neth J Sea Res 16:195–207 Kott P (1953) Modified whirling apparatus for the subsampling of plankton. Aust J Mar Freshw Res 4:387–393 Levinsen H, Nielsen TG, Hansen BW (1999) Plankton community structure and carbon cycling on the western coast of Greenland during the stratified summer situation. II. Ciliates and heterotrophic dinoflagellates. Aquat Microb Ecol 16:217–232 Lie U, Svendsen H, Kaartvedt S, Mikki S, Johnsen TM, Aksnes DL, Aswall RP, Golmen L (1992) Vannkraft og fjorder. Fysiske og biologiske konsenkvenser av Ulla-Førre utbyggingen. Rapp. 4/92, Center for Resources and Developmental Studies, University of Bergen, Bergen, Norway Magnesen T (1988) Horizontal distribution of zooplankton in Linda˚spollene, western Norway, 1979. Sarsia 73:193–204 Montagnes DJS, Lynn DH (1991) Taxonomy of choreotrichs, the major marine planktonic ciliates, with emphasis on the aloricate forms. Mar Microb Food Webs 5:59–74 Nejstgaard JC, Gismervik I, Solberg PT (1997) Feeding and reproduction by Calanus finmarchicus, and microzooplankton grazing during mesocosm blooms of diatoms and the coccolithophore Emiliania huxleyi. Mar Ecol Prog Ser 147:197–217 Nichols JH, Thompson AB (1991) Mesh selection of copepodite and nauplius stages of four calanoid copepod species. J Plankton Res 13:661–671 Nielsen TG, Sabatini M (1996) The role of cyclopoid copepods Oithona spp. in North Sea plankton communities. Mar Ecol Prog Ser 139:79–93

Nielsen TG, Løkkega˚rd B, Richardson K, Pedersen FB, Hansen L (1993) The structure of the plankton communities in the Dogger Bank area (North Sea) during a stratified situation. Mar Ecol Prog Ser 95:115–131 Nielsen TG, Møller EF, Satapoomin S, Ringuette M, Hopcroft RR (2002) Egg hatching rate of the cyclopoid copepod Oithona similis in arctic and temperate waters. Mar Ecol Prog Ser (in press) Onsrud MRS, Kaartvedt S (1998) Diel vertical migration of the krill Meganyctiphanes norvegica in relation to physical environmental, food and predators. Mar Ecol Prog Ser 171:209–219 Paasche E, Kristiansen S (1982) Ammonium regeneration by microzooplankton in the Oslofjord (Norway). Mar Biol 69:55–64 Paffenho¨fer G-A (1976) On the biology of Appendicularia of the southeastern North Sea. In: Persoone G, Jarpers E (eds) Proceedings of the 10th European symposium on marine biology, vol 2. Universa, Wetteren, Belgium, pp 437–455 Paffenho¨fer G-A, Harris RP (1976) Feeding, growth and reproduction of the marine planktonic copepod Pseudocalanus elongatus Boeck. J Mar Biol Assoc UK 56:327–344 Parsons TR, Takahasi M, Hargrave B (1984) Biological oceanographic processes. Pergamon, New York Putland JN (2000) Microzooplankton herbivory and bacterivory in Newfoundland coastal waters during spring, summer and autumn. J Plankton Res 22:253–278 Roddie BD, Leakey RJG, Berry AJ (1984) Salinity–temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth estuary. J Exp Mar Biol Ecol 79:191–211 Rodhouse PG, Roden CM (1987) Carbon budgets for a coastal inlet in relation to intensive cultivation of suspension-feeding bivalve molluscs. Mar Ecol Prog Ser 36:225–236 Ruttner-Kolisko A (1977) Suggestions for biomass calculations of planktonic rotifers. Arch Hydrobiol Beih Ergebn Limnol 8:71–76 Sabatini M, Kiørboe T (1994) Egg production, growth and development of the cyclopoid copepod Oithona similis. J Plankton Res 16:1329–1351 Satapoomin S (1999) Carbon content of some common tropical Andaman Sea copepods. J Plankton Res 21:2117–2123 Schlu¨ter L, Havskum H (1997) Phytoplankton pigments in relation to carbon content in phytoplankton communities. Mar Ecol Prog Ser 155:55–65 Skjoldal HR, Hopkins C, Erikstad KE, Leinaas HP (eds) (1995) Ecology of fjords and coastal waters. Elsevier, Amsterdam Skreslet S (1986) (ed) The role of freshwater outflow in coastal marine ecosystems. In: NATO ASI series, series G, ecological sciences, vol 7. Springer, Berlin Heidelberg New York Smetacek V (1981) The annual cycle of protozooplankton in the Kiel bight. Mar Biol 63:1–11 Tande KS (1991) Calanus in North Norwegian fjords and in the Barents Sea. In: Sakshaug E, Øritsland NA (eds) Proceedings of the Pro Mare symposium on polar marine ecology, 1990. Polar Res 10:389–407 Thingstad TF, Skjoldal EF, Bohne RA (1993) Phosphorus cycling and algal–bacterial competition in Sandsfjord, western Norway. Mar Ecol Prog Ser 99:239–259 Thomsen HA (ed) (1992) Plankton i de indre danske farvande. Miljøstyrelsen, Copenhagen Tiselius P (1992) Behavior of Acartia tonsa in patchy food environments. Limnol Oceanogr 37:1640–1651 Utermo¨hl H (1958) Zur Vervollkommung der quantitativen Phytoplankton Methodik. Mitt Int Ver Theor Angew Limnol 9: 1–38 Uye SI, Ichino S (1995) Seasonal variations in abundance, size composition, biomass and production rate of Oikopleura dioica (Fol) (Tunicata: Appendicularia) in a temperate eutrophic inlet. J Exp Mar Biol Ecol 189:1–11 Verity PG, Vernet M (1992) Microzooplankton grazing, pigments and composition of plankton communities during late spring in two Norwegian fjords. Sarsia 77:263–274