Hydrobiologia 514: 7378, 2004. H. Kautsky & P. Snoeijs (eds), Biology of the Baltic Sea.

2004 Kluwer Academic Publishers. Printed in the Netherlands.

73

Effects of grazing and excretion by pelagic mysids (Mysis spp.) on the size structure and biomass of the phytoplankton community
Eveliina Lind n1,2,3, & Harri Kuosa2 e
1 Department of

Ecology and Systematics, Division of Hydrobiology, P.O. Box 65, FIN-00014 Helsinki, University of Helsinki, Finland 2 Tvrminne Zoological Station, FIN-10900 Hanko, Finland 3 Finnish Institute of Marine Research, P.O. Box 33, FIN-00931 Helsinki, Finland E-mail: eveliina.linden@mr. ( Author for correspondence)

Key words: Mysis spp., grazing, excretion, phytoplankton community, bottom-up, top-down

Abstract The aim of this study was to determine the effects of pelagic mysids (Mysis mixta and M. relicta) on the biomass and size-structure of the phytoplankton community during the period following the spring bloom. Mysids excreted phosphate (4.5 0.7 nmol ind1 h1 ) and ammonium (123.6 31.6 and 45.0 3.2 nmol ind1 h1 ) and increased the total chlorophyll-a concentration of phytoplankton slightly. However, the presence of mysids affected different size-classes of phytoplankton differently. Mysids mainly grazed on large-sized (>10 m) phytoplankton cells. Small-sized (<10 m) algal cells avoided grazing, gained a competitive advantage and were able to utilize the nutrients excreted by mysids. According to this study, both top-down and bottom-up mechanisms simultaneously mould the structure of the phytoplankton community. A large zooplankton biomass might promote the increase of small agellates by a combination of repleting nutrient stores, selective grazing on large algal cells and heavy predation on protozoa which, consequently, might have a cascading effect on the most favoured protozoan food source, small agellates. iles, since the cell size is small (<100 m) and during the bloom, plenty of it is available. Of the main phytoplankton groups, mysids feed most on diatoms (Diatomophyceae), which in June constitute 58% of the total phytoplankton consumed (Viherluoto et al., 2000). They also feed on dinoagellates (Dinophyceae), green algae (Chlorophyceae) and blue-green algae (Cyanobacteria; Viherluoto et al., 2000). The use of zooplankton increases as summer proceeds, when mysids grow larger and are thus more able to utilize the increased biomass of zooplankters (Adare & Lasenby, 1994; Viherluoto et al., 2000). Nutrients, which have accumulated in the water during winter, are depleted from the epilimnion during the phytoplankton spring bloom in the Baltic Sea, and the concentrations of inorganic nutrients become very low (Tamminen, 1990). During summer, the production is mainly based on the regeneration of nutrients

Introduction Mysis mixta (Lilljeborg) and M. relicta (Lovn) are pelagic mysid shrimps in the Gulf of Finland. They start breeding in late autumn and release the juveniles after ice break-up, when the spring bloom of phytoplankton is most intense (Rudstam et al., 1986; Salemaa et al., 1986; Vinl, 1986; Rudstam & Hansson, 1990). The diet of mysids varies seasonally since the availability of different planktonic food item groups varies considerably during the growth season. The diet changes also as the juveniles mature, since small-sized mysids cannot capture actively moving zooplankton as well as larger adult individuals can (Cooper & Goldman, 1980; Viherluoto et al., 2000). In spring, mysids feed almost solely on phytoplankton (ca. 7080%; Viherluoto et al., 2000). Spring phytoplankton is a highly suitable nutrition source for the juven-

74 (Tamminen, 1990; Lignell et al., 1992). The availability of nutrients is linked to the mineralization rate, since the nutrient concentrations in the water would support primary production only for an hour or two (Tamminen, 1990). Some new nutrients do arrive in the ecosystem also during summertime, through e.g. physical mixing processes and nitrogen xation. In autumn, as the water column mixes, nutrients from the hypolimnion arrive, and the importance of regeneration decreases (Tamminen, 1990). With regard to nitrogen, the new production is based on the use of nitrate (NO ) and molecular nitrogen (N2 ); the regen3 erated production is based on ammonium (NH+ ) and 4 dissolved organic nitrogen (DON; Dugdale & Goering, 1967). The zooplankton plays an important role in nutrient regeneration, since it releases both inorganic and organic nutrients while handling and feeding (sloppy feeding), by excretion and as a result of the decay of faecal pellets (Lampert, 1978; Raymont, 1983; Banse, 1995). The aim of this paper is to study the effects of pelagic mysids on the biomass and size structure of the phytoplankton community during the period following the spring bloom, when phytoplankton is still abundant in the water column, as the bloom sinks towards the bottom. We wanted to study whether mysids feed equally on different phytoplankton size classes and if the nutrients excreted by mysids affect the growth of phytoplankton in the northern Baltic Sea. The hypothesis was that the mysids would graze more effectively on large-sized (diameter >10 m) cells or colonies, such as diatoms. Thus, small-sized phytoplankton could avoid grazing and gain a competitive advantage for nutrients excreted by the mysids. This process could lead to an increase of small-sized algae up to the point of a mass occurrence or bloom forming. We wanted to show the general impact of mysids on the planktonic community through both top-down (grazing) and bottom-up regulation (nutrient recycling). to the Gulf of Finland, or with a large zooplankton net from the 125 m deep monitoring station SR5 (6113 N; 19 97 E) in the Bothnian Bay. They were kept in ltered seawater and fed with commercial sh food (Aqua Win, Kerox Oy; Lngden) or with natural zooplankton (SR5). The length of the mysids was 2 to 3 cm. Seawater was ltered through a GF/F glassmicrober lter (Whatman) into experimental containers of 250 ml (Lngden) or 60 ml (SR5), and one mysid, after measuring its length, was added to each container. Containers without mysids served as controls. All the containers were incubated at 10 C (Lngden) or 3 C (SR5) for ca. 12 h, after which the concentrations of dissolved nutrients (NH+ and PO3 ) 4 4 were measured (Grasshoff, 1976; QuikChem Method, 1998). The amount of excreted nutrients in the containers with the mysids was calculated by subtracting the average concentration of the respective nutrient in the control containers from the nutrient concentration in the experimental container. Altogether 64 individual animals were tested. The data was tested for statistical signicance using a t-test, after log(x + 1) transformation of the data. Community experiments The water for the experiments was collected from Storgadden (5947 N; 23 19 E; depth 46 m) or Storfjrden (59 57 N; 2316 E; depth 33 m), which are situated southeast of the Hanko Peninsula. The natural phytoplankton community (ltered through a 100-m net) was collected with a Limnos sampler from ca. 1 m depth and the water temperature was recorded. The temperature of the climate room was set at the beginning of each experiment according to the water temperature at the time of collection (6 to 8 C). The experimental containers were illuminated from above by uorescent lamps with 60 E m2 s1 , corresponding to the light environment at about 5 m depth during the day. The light/dark period was 18/6 h. The mysids were collected from Lngden with an epibenthic sled, kept in an aquarium lled with seawater and fed with commercial sh food. The light/dark period was 18/6 h and the temperature of the room was ca. 10 C. The mysids were starved for 24 h before starting the experiments. The size of the mysids was 1 to 3 cm. Five experiments were conducted during May and June 2000, with three replicates in each treatment. 2 l transparent glass bottles were used as experimental containers. The treatments were mysid addition (ve

Materials and methods Excretion experiments The excretion experiments were conducted in May, June and August 2000. The mysids (M. mixta and M. relicta) were collected with an epibenthic sled from Lngden (5945 40 N; 23 13 90 E), an open deepsea area near the Hanko Peninsula at the entrance

75 mysids), nutrient addition (NH4 Cl-N 2.9 nmol l1 and Na2 HPO4 -P 0.32 nmol l1 ) and control (without nutrient additions). The containers were incubated in the climate room for three days. To facilitate gaseous exchange, the mysid containers were closed only with cotton plugs. On each day of the experiment, two 100 ml samples for chlorophyll-a analysis were taken from each container. One sample was used for total chlorophyll-a and the other sample was ltered through a 10-m polycarbonate lter (Poretics) to remove particles larger than 10 m. From each chlorophyll-a sample, two 50 ml samples were ltered onto GF/F lters, which were then placed into 10 ml 96% ethanol. The samples were ultrasonicated for 10 min and chlorophyll was extracted for at least 6 h. Before measuring the samples were ltered through GF/F lters to remove any impurities. Chlorophyll-a concentrations were measured with a Shimadzu spectrouorometer calibrated with pure chlorophyll-a (Sigma Chem. Co., U.S.A.), using 96% ethanol as the 0-sample. If the data fullled the assumptions of parametric tests (in some cases after log(x + 1) transformations of the data), one-way analysis of variance (ANOVA) and t-test were performed. Otherwise KruskalWallis oneway nonparametric analysis of variance and MannWhitney U -test were performed. The results presented in this paper are combined from all ve experiments.

Results The excretion of phosphorus by mysids was not signicantly different at the two locations (Lngden and SR5; t-test: t = 1.568, df = 47, p > 0.05). Thus, the results presented here are combined for the two locations. However, the excretion of ammonium varied signicantly (t-test: t = 3.955, df = 48, p < 0.001) between the two locations. The mysids excreted (mean SE) phosphate (PO3 ) 4.5 0.7 nmol ind1 h1 4 and ammonium (NH+ ) 123.6 31.6 nmol ind1 h1 4 (Lngden) and 45.0 3.2 nmol ind1 h1 (SR5). The mean N/P ratios of the excretion were 2.36 (Lngden) and 9.49 (SR5). The combined addition of the two main nutrients (N and P) had an immediate effect on total chlorophyll-a value (Fig. 1a). This effect levelled after the second day. The increased chlorophyll-a was mostly due to the increase in larger phytoplankton fraction, but also small-sized algal fraction showed a positive response (Figs 1b, c). Mysid addition in-

Figure 1. Chlorophyll-a concentration (mean SD) in three treatments. (a) Total, (b) in >10 m size fraction, (c) in <10 m size fraction. = p < 0.01, = p < 0.001, denotes the results of ANOVA or KruskalWallis.

creased the total chlorophyll-a concentration of the phytoplankton compared to control (day 3, t-test: t = 3.635, df = 22, p < 0.001; Fig. 1a), but its effect on different size classes of phytoplankton varied greatly. Mysids obviously grazed on large-sized (>10 m) phytoplankton cells, as chlorophyll-a concentration of that size-fraction decreased signicantly (day 3, ttest: t = 3.98, df = 23, p < 0.001; Fig. 1b). At the same time, the presence of mysids increased the

76 chlorophyll-a concentration of small phytoplankton (<10 m) signicantly, even compared to the nutrient addition treatments (day 3, MannWhitney U -test: p < 0.05; Fig. 1c). of 0.21 mol g dw1 h1 , but these values are not directly comparable to ours since the mysids in their experiments were starved several hours prior to the experiments. Pelagic mysids migrate diurnally between the benthos and the water column (Aneer, 1980), so excretion occurs not only in the euphotic zone. Thus, not all nutrients excreted are available to the phytoplankton. Mysids also transfer phytoplankton biomass away from the euphotic zone, when they migrate back to the bottom in the morning with their gut full after feeding (Madeira et al., 1982). Mysids are found in the euphotic zone during night time, when no light assimilation occurs, but the phytoplankton is able to take up nutrients also during darkness (Harvey, 1963). The patchy distribution of mysids (Haury et al., 1978) results in an uneven distribution of the excreted nutrients as well. Thus it is still premature to evaluate the real effect of mysid remineralization to bottom-up dynamics, but our results indicate a potential mechanism of nutrient replenishment during the summer minimum. Excretion experiments are technically difcult to conduct. The quality of the experimental water changes constantly; the oxygen concentration decreases, carbon dioxide and ammonium accumulate and the ionic concentrations and pH also change. These changes and the handling of mysids could cause stress and thus a change in excretion rates (Ikeda, 1977). High external ammonia concentration probably decreases the diffusion of ammonia from the tissues and can also stimulate excretion of urea (Mullin et al., 1975; Seale & Boraas, 1982). In addition, the organic or inorganic nutrients dissolved from the faecal pellets during the experiment may have affected the results. We did monitor the oxygen level during the experiments and found declining, but not critical, oxygen values. Taking into account the drawbacks of the experimental system, we give our results as preliminary and encourage further studies on the subject, whilst taking into account size variability and the nutritional history of mysid populations. According to our study, Baltic pelagic mysids do have a different effect on phytoplankton biomass in different size classes. The different succession of chlorophyll-a concentrations in different size fractions supports the hypothesis of size-selective feeding (Viherluoto et al., 2000). In containers with mysids, the chlorophyll-a concentration of large (>10 m) phytoplankton decreased signicantly compared to other treatments. Thus it can be concluded that mysids fed on this size class. In addition, Bowers & Gross-

Discussion Mysids actively excreted inorganic nutrients, ammonium and phosphate in the ratio of 2.369.49, which is in the same range as M. relicta in Lake Michigan (Madeira et al., 1982), where the measured N/P-ratio was 5.7, ammonia excreted at the rate of 0.17 g N mg dw1 h1 and phosphate 0.03 g N mg dw1 h1 , which would supply 110% of the nitrogen and phosphorus required by thermoclineassociated phytoplankton. Our N/P ratios also correspond with the N/P ratios in the two most common copepods in the northern Baltic Sea, Acartia bilosa (4.0) and Eurytemora afnis (4.57; M. Koski & M. Viitasalo, unpubl.), which formed a major part of the diet of the mysids in the excretion experiments conducted at SR5. Both nitrogen and phosphorus together are more efcient in enhancing primary production than either of the nutrients alone during early summer (Kivi et al., 1993). Thus the excretion of mysids was likely to have a bottom-up effect on the growth of phytoplankton. However, our excretion values are near maximum, due to the nutritional history of the animals, i.e. they were fed with food which was high in protein (zooplankton or commercial sh food). This, together with the lower temperature at the site, may also explain the lower ammonium excretion rates of mysids collected from SR5. Temperature, however, has been found not to have an effect on the amount of nitrogen excreted by zooplankton, but it depends more on the quantity and quality of the available food, or other environmental or physiological factors (Jacobsen & Comita, 1976; Seale & Boraas, 1982). When plenty of food is available, the zooplankton utilizes mostly carbohydrates for the production of energy and saves proteins for growth, and very little ammonium is excreted. When food is scarce, proteins are catabolized to energy, which leads to increased excretion of ammonium and decreased growth rate (Martin, 1968; Mayzaud, 1976). However, Weisse & Rudstam (1989) have shown that temperature does affect the excretion rates of ammonia and phosphate in Neomysis integer, though not the N/P ratio of the excreta. Their experiments gave N/P ratios of 726, ammonia excretion of 27 mol g dw1 h1 and phosphate excretion

77 nickle (1978) observed in their experiments in Lake Michigan that adult or juvenile Mysis relicta did not signicantly graze on algal cells smaller than 10 m and they grazed most effectively on the size class of over 53 m. In our experiments, the larger phytoplankton fraction consisted mainly of diatoms (Skeletonema costatum and Chaetoceros spp.), small dinoagellates and small colonial blue-green algae, all of which are considered to be suitable food sources for mysids, and which have previously been shown to be heavily grazed upon by mysids (Viherluoto et al., 2000). Large trichal and potentially toxic bluegreen algae were not present in the early summer experiments. As small algal cells avoided grazing, they gained a competitive advantage over larger ones and were able to utilize the nutrients excreted by mysids. In particular, the high surface-to-volume ratio of small cells aids in absorbing nutrients even from low concentrations. Another benecial factor for the success of small-sized phytoplankton is their release from the grazing pressure of pelagial ciliates by mysid predation on ciliates. This is a common cascading effect in Baltic summer communities, in which ciliates are tightly controlled by crustacean predators (Kivi et al., 1993). The smallest size-fraction consisted mainly of small agellated cells from several systematic groups (Chrysophyceae, Prasinophyceae, Cryptophyceae and Prymnesiophyceae). However, their similar functional characteristics, small cell size, naked cell surface and active swimming, make all of them optimal food source for protozoan grazers. Also, Kivi et al. (1996) found in their study on Baltic mesozooplankton, that the 1 to 10 m size-class phytoplankton increased in containers where both nutrients (nitrogen and phosphorus) and crustacean zooplankton were added. This size-class avoided mesozooplankton grazing due to their small size and gained from both the nutrient addition and the nutrient regeneration by the animals. anisms work simultaneously between these trophic levels.

Acknowledgements We wish to thank T. Sjlund for help in the eld, J. Leppnen for help with the chlorophyll-a analyses and A.-M. strm, P. Lemponen and I. Lastumki for laboratory analyses. Field and laboratory facilities were provided by the Tvrminne Zoological Station, University of Helsinki, and R/V Aranda, Finnish Institute of Marine Research. Stephen Venn revised the English of the manuscript. This study was nanced by the Walter and Andre de Nottbeck Foundation.

References
Adare, K. I. & D. C. Lasenby, 1994. Seasonal changes in the total lipid content of the opossum shrimp, Mysis relicta (Malacostraca: Mysidacea). Can. J. Fish. Aquat. Sci. 51: 1935 1941. Aneer, G., 1980. Estimates of feeding pressure on pelagic and benthic organisms by Baltic herring (Clupea harengus v. membras L.). Ophelia 1: 265275. Banse, K., 1995. Zooplankton: Pivotal role in the control of ocean production. ICES J. Mar. Sci. 52: 265277. Bowers, J. A. & N. E. Grossnickle, 1978. The herbivorous habits of Mysis relicta in Lake Michigan. Limnol. Oceanogr. 23: 767776. Cooper, S. D. & C. R. Goldman, 1980. Opossum shrimp (Mysis relicta) predation on zooplankton. Can. J. Fish Aquat. Sci. 37: 909919. Dugdale, R. C. & J. J. Goering, 1967. Uptake of new and regenerated forms of nitrogen in primary productivity. Limnol. Oceanogr. 12: 196206. Grasshoff, K., 1976. Methods of Seawater Analysis. Verlag Chemie, 2nd edn. Harvey, H. W., 1963. The Chemistry and Fertility of Sea Waters. Cambridge Univ. Press, London, 240 pp. Haury, L. R., J. A. McGowan & P. H. Wiebe, 1978. Patterns and processes in the time-space scales of plankton distributions. In Steele, J. H. (ed.), Spatial pattern in plankton communities. Plenum Press, New York: 227327. Ikeda, T., 1977. The effect of laboratory conditions on the extrapolation of experimental measurements to the ecology of marine zooplankton. IV. Changes in respiration and excretion rates of boreal zooplankton species maintained under fed and starved conditions. Mar. Biol. 41: 241252. Jacobsen, T. R. & G. W. Comita, 1976. Ammonia nitrogen excretion in Daphnia pulex. Hydrobiologia 51: 195200. Kivi, K., H. Kuosa & S. Tanskanen, 1996. An experimental study on the role of crustacean and microprotozoan grazers in the planktonic food web. Mar. Ecol. Prog. Ser. 136: 5968. Kivi, K., S. Kaitala, H. Kuosa, J. Kuparinen, E. Leskinen, R. Lignell, B. Marcussen & T. Tamminen, 1993. Nutrient limitation and grazing control of the Baltic plankton community during annual succession. Limnol. Oceanogr. 38: 893905. Lampert, W., 1978. Release of dissolved organic carbon by grazing zooplankton. Limnol. Oceanogr. 23: 831834.

Conclusions According to our results, mysids regenerate nutrients along with other zooplankton and thus maintain phytoplankton production. At the same time, the size selectivity of the feeding of the mysids moulds the size structure of the phytoplankton community in the Baltic Sea, favouring small-sized phytoplankton, mainly small agellates, and reducing larger phytoplankton. Thus both top-down and bottom-up mech-

78
Lignell, R., S. Kaitala & H. Kuosa, 1992. Factors controlling phytoand bacterioplankton in late spring on a salinity gradient in the northern Baltic. Mar. Ecol. Prog. Ser. 84: 121131. Madeira, P. T., A. S. Brooks & D. B. Seale, 1982. Excretion of total phosphorus, dissolved reactive phosphorus, ammonia and urea by lake Michigan Mysis relicta. Hydrobiologia 93: 145154. Martin, J. H., 1968. Phytoplanktonzooplankton relationships in Narragansett Bay III. Seasonal changes in zooplankton excretion in relation to phytoplankton abundance. Limnol. Oceanogr. 13: 6371. Mayzaud, P., 1976. Respiration and nitrogen excretion of zooplankton. IV. The inuence of starvation on the metabolism and the biochemical composition of some species. Mar. Biol. 37: 4758. Mullin, M. M., M. J. Perry, E. H. Renger & P. M. Evans, 1975. Nutrient regeneration by oceanic zooplankton: a comparison of methods. Mar. Sci. Commun. 1: 113. QuikChem Methods 31-115-01-3-A, 1998. Determination of Phosphorus by Flow Injection Analysis Colorimetry. Lachat Instruments, Milwaukee, U.S.A. Raymont, J. E. G., 1983. Plankton and Productivity in the Oceans. II. Zooplankton. Pergamon Press, Oxford, 824 pp. Rudstam, L. G. & S. Hansson, 1990. On the ecology of Mysis mixta (Crustacea, Mysidacea) in a coastal area of the northern Baltic proper. Ann. Zool. Fenn. 27: 259263. Rudstam, L. G., S. Hansson & U. Larsson, 1986. Abundance, species composition and production of mysid shrimps in a coastal area of the northern Baltic proper. Ophelia 4: 225238. Salemaa, H., K. Tyystjrvi-Muuronen & E. Aro, 1986. Life histories, distribution and abundance of Mysis mixta and Mysis relicta in the northern Baltic Sea. Ophelia 4: 239247. Seale, D. B. & M. E. Boraas, 1982. Inuence of experimental conditions on nitrogenous excretion by Lake Michigan Mysis relicta (Lovn): laboratory studies with animals acclimated in Fragilaria. Hydrobiologia 93: 163169. Tamminen, T., 1990. Eutrophication and the Baltic Sea: Studies on phytoplankton, bacterioplankton and pelagic nutrient cycles. PhD Thesis. University of Helsinki, Department of Environmental Protection, 22 pp. Viherluoto, M., H. Kuosa, J. Flinkman & M. Viitasalo, 2000. Food utilisation of pelagic mysids, Mysis mixta and M. relicta, during their growing season in the northern Baltic Sea. Mar. Biol. 136: 553559. Vinl, R., 1986. Sibling species and phylogenetic relationships of Mysis relicta (Crustacea: Mysidacea). Ann. Zool. Fenn. 23: 207221. Weisse, T. & Rudstam, L. G., 1989. Excretion and respiration rates of Neomysis integer (Mysidaceae): effects of temperature, sex and starvation. Hydrobiologia 178: 253258.