TURUN YLIOPISTON JULKAISUJA ANNALES UNIVERSITATIS TURKUENSIS SARJA - SER. AII OSA - TOM. 239 BIOLOGICA - GEOGRAPHICA - GEOLOGICA ROLE OF NUTRIENTS IN REGULATION OF THE PHYTOPLANKTON COMMUNITY IN THE ARCHIPELAGO SEA, NORTHERN BALTIC SEA by Annika Lagus TURUN YLIOPISTO UNIVERSITY OF TURKU Turku 2009 From the Section of Ecology Department of Biology University of Turku FIN-20014 Turku Finland Supervised by Prof. Jouko Sarvala Department of Biology University of Turku Finland Dr. Pirjo Kuuppo Neste Oil Corporation Finland Reviewed by Prof. Jorma Kuparinen Department of Biological and Environmental Sciences University of Helsinki Finland Prof. Ulrich Sommer Leibniz-Institute of Marine Sciences Germany Opponent Prof. Anna-Stiina Heiskanen Research Manager Finnish Environment Institute Finland ISBN 978-951-29-3981-7 (PRINT) ISBN 978-951-29-3982-4 (PDF) ISSN 0082-6979 Painosalama Oy - Turku, Finland 2009 List of Original Articles 3 LIST OF ORIGINAL ARTICLES This thesis is based on the following articles, which are referred to by the Roman numerals in the text: I. Lagus, A., Suomela, J., Weithoff, G., Heikkilä, K., Helminen, H. & Sipura, J. 2004. Species-specific differences in phytoplankton responses to N and P enrichments and the N:P ratio in the Archipelago Sea, northern Baltic Sea. Journal of Plankton Research 26: 779-798. II. Lagus, A., Suomela, J., Helminen, H., Lehtimäki, J. M., Sipura, J., Sivonen, K. & Suominen, L. 2007. Interaction effects of N:P ratios and frequency of nutrient supply on the plankton community in the northern Baltic Sea. Marine Ecology Progress Series 332: 77-92. III. Lagus, A., Silander, M. & Suomela, J. 2002. Influence of nutrient enrichments on cyanobacteria in the Archipelago Sea, northern Baltic. Verhandlungen der Internationalen Vereinigung für Theoretische und Angewandte Limnologie 28: 607-612. IV. Vuorio, K., Lagus, A., Lehtimäki, J. M., Suomela, J. & Helminen, H. 2005. Phytoplankton community responses to nutrient and iron enrichment under different nitrogen to phosphorus ratios in the northern Baltic Sea. Journal of Experimental Marine Biology and Ecology 322: 39-52. V. Suomela, J., Gran, V., Helminen, H., Lagus, A., Lehtoranta, J. & Sipura, J. 2005. Effects of sediment and nutrient enrichment on water quality in the Archipelago Sea, northern Baltic: An enclosure experiment in shallow water. Estuarine, Coastal and Shelf Science 65: 337-350. VI. Lagus, A., Suomela, J., Helminen, H. & Sipura, J. 2007. Impacts of nutrient enrichment and sediment on phytoplankton community structure in the northern Baltic Sea. Hydrobiologia 579: 351-368. The original publications have been reprinted with the kind permission of Oxford University Press (Article I), Inter-Research Science Publisher (Article II), E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, Germany, http://www.schweizerbart.de (Article III), Elsevier Science (Articles IV & V) and Springer Science and Business Media (Article VI) 4 Contents CONTENTS LIST OF ORIGINAL ARTICLES ................................................................................3 CONTENTS....................................................................................................................4 1. INTRODUCTION ....................................................................................................5 1.1. Eutrophication and the Baltic Sea ......................................................................5 1.2. Nutrient regulation of phytoplankton growth ....................................................7 1.3. Coastal food webs ...............................................................................................9 1.4. Phytoplankton competition for limiting nutrients ............................................11 1.5. Nutrient-phytoplankton-zooplankton interactions ...........................................13 1.6. Aim of the study ...............................................................................................14 2. MATERIAL AND METHODS .............................................................................15 2.1. Study Area .......................................................................................................15 2.2. Mesocosm experiments ....................................................................................17 2.3. Plankton biomass .............................................................................................22 2.4. Primary production ...........................................................................................22 2.5. Hepatotoxins .....................................................................................................23 2.6. Nutrients and chlorophyll a ..............................................................................23 3. RESULTS AND DISCUSSION ..............................................................................24 3.1. Phytoplankton community ...............................................................................24 3.1.1. Nutrient limitation ..................................................................................24 3.1.2. Role of the N:P ratio ...............................................................................25 3.1.3. Frequency of nutrient supply .................................................................27 3.2. Diazotrophic cyanobacteria ..............................................................................28 3.2.1. Effects of nutrient enrichments and N:P ratio ........................................28 3.2.2. Effects of nutrient supply frequency .....................................................30 3.2.3. Effects of Iron ........................................................................................31 3.2.4. Cyanobacterial hepatotoxins ..................................................................32 3.3. The grazer community ......................................................................................32 3.3.1. Top-down effects of zooplankton ...........................................................32 3.3.2. Bottom-up effects of nutrient enrichments on higher trophic levels ......34 3.4. Role of the bottom sediment .............................................................................36 3.4.1. Sediment as a source of nutrients ..........................................................36 3.4.2. Effects of sediment on phytoplankton community structure .................38 4. CONCLUSIONS ....................................................................................................40 ACKNOWLEDGEMENTS ........................................................................................42 REFERENCES .............................................................................................................44 ORIGINAL ARTICLES ..............................................................................................57 Introduction 5 1. INTRODUCTION 1.1. Eutrophication and the Baltic Sea Increased nutrient loading due to human activity has caused eutrophication of coastal ecosystems throughout the world (Nixon 1995, Cloern 2001, Smith 2006). Eutrophication has been defined as “the enrichment of water by nutrients, especially nitrogen and/or phosphorus and organic matter, causing an increased growth of algae and higher forms of plant life to produce an unacceptable deviation in structure, function and stability of organisms present in the water and to the quality of water concerned, compared to reference conditions” (Andersen et al. 2006). The consequences of eutrophication include a possible shift in phytoplankton species composition (Wasmund & Uhlig 2003, Carstensen & Heiskanen 2007, Suikkanen et al. 2007), an increase in harmful algal blooms (Smayda 1990, Cloern 2001, Scavia & Bricker 2006), increased turbidity, oxygen deficiency in bottom waters (Rosenberg et al. 1990, Bishop et al. 2006, Kiirikki et al. 2006) and changes in fish (Baden et al. 1990, Rajasilta et al. 1999, Lappalainen et al. 2000) and benthos communities (Hansson & Rudstam 1990, Perus & Bonsdorff 2004, Perus et al. 2007). In the Baltic Sea the total load of nitrogen (N) and phosphorus (P) have been estimated to have increased fourfold and eightfold respectively from the 1940s to the 1980s (Larsson et al. 1985). Enhanced nutrient inputs have increased primary production by 30 to 70 % and sedimentation by 70 to 190 % (Elmgren 1989). This has resulted in increased oxygen consumption in the sediments, leading to an increase in oxygen-deficient bottom areas (Jansson & Dahlberg 1999). The increase in anoxic bottom sediments in turn enhances the benthic release of P. Oxic bottom sediments function as a sink for P, which is bound to iron (III) hydroxides, but when the sediments turn anoxic P is released via reduction of the metal oxides (Krom & Berner 1981, Lehtoranta et al. 1997, Pitkänen et al. 2003, Blomqvist et al. 2004). This may lead to a vicious circle, whereby eutrophication increases the sedimentation rate leading to anoxic sediments, triggering the release of P, causing more nutrients to enter the system and further exacerbating the eutrophication problem (Tamminen & Andersen 2007, Vahtera et al. 2007a). Although cyanobacterial blooms are a natural annual late summer phenomenon in the Baltic Sea (Bianchi et al. 2000), there are indications that they have increased during recent decades concomitantly with raised nutrient levels (Kahru et al. 1994, Finni et al. 2001, Kauppila & Lepistö 2001, Poutanen & Nikkilä 2001, Suikkanen et al. 2007). In open waters the blooms are dominated by the nitrogen-fixing species Aphanizomenon sp. and Nodularia spumigena Mertens ex Bornet and Flahault, while in coastal waters Anabaena spp. is also common (e.g. Sivonen et al. 2007). All three genera contain gas vesicles, making them buoyant; in calm weather they rise to the surface, where 6 Introduction they may form large aggregates (Walsby et al. 1995, Walsby et al. 1997, Ploug 2008). Cyanobacterial blooms are of particular concern because they are often toxic. In the Baltic Sea Nodularia spumigena produces the hepatotoxin nodularin (Sivonen et al. 1989), while Anabaena has recently been confirmed to produce microcystin, another hepatotoxin (Halinen et al. 2007). Baltic Aphanizomenon has not been found to be toxic (Sivonen et al. 1989, Repka et al. 2004), but freshwater species of both Anabaena and Aphanizomenon are capable of producing neurotoxins in addition to microcystins (Sivonen et al. 1990, Carmichael 1997, Willen & Mattson 1997). Toxin production by algae may be affected by nutrient availability (Granéli et al. 1998, Granéli & Flynn 2006, Lindehoff et al. 2009). Toxicity may increase under nutrient limitation due to cellular stress (Johansson & Granéli 1999). However, non-toxic algal blooms may also have negative effects due to their high biomass production. The blooms sometimes form an unpleasant scum on the water surface, reducing the recreational value of the water, and their decay may deplete water oxygen concentrations. Cyanobacterial blooms may moreover aggravate Baltic eutrophication by their nitrogen fixation (Savchuk & Wulff 1999, Rolff et al. 2007, Vahtera et al. 2007a). Recent studies suggest that diazotrophic cyanobacteria are responsible for one third to half of the total external nitrogen load in the Baltic proper (Larsson et al. 2001, Wasmund et al. 2001, Moisander et al. 2007, Rolff et al. 2007). The high occurrence of N -fixing cyanobacteria in the Baltic Sea has been suggested 2 to be due to the low N:P ratio in the area (Niemi 1979, Stal et al. 2003). A low N:P ratio is assumed to favour the growth of diazotrophic cyanobacteria because of the competitive ability provided by N fixation (Smith 1983, Granéli et al. 1990, Vrede et al. 2009). 2 Mesocosm experiments in the Baltic, however, have yielded contradictory results as to the effects of the N:P ratio and P-enrichment on cyanobacterial growth (e.g. Wallström et al. 1992, Rydin et al. 2002, Kuuppo et al. 2003). Thus the effect on cyanobacteria of an increased nutrient load and of the N:P ratio in the Baltic Sea is still a debated subject. It is well known that cyanobacterial growth is affected by many other factors in addition to nutrient availability, including water temperature, salinity and water column stability (Kononen et al. 1996, Laamanen 1997, Wasmund 1997, Hyenstrand et al. 1998, Kahru et al. 2000, Larsson et al. 2001, Rydin et al. 2002, Kanoshina et al. 2003, Stal et al. 2003, Lips & Lips 2008). It has also been proposed that the availability of trace elements, especially iron, may be an important factor affecting the growth of Baltic cyanobacteria (Howarth & Marino 1998, Stal et al. 1999, Schubert et al. 2008). Iron is a component of the nitrogenase enzyme complex, the enzyme responsible for nitrogen fixing, and is thus essential for N fixation (Fay 1992). It has been estimated that N -fixing cyanobacteria 2 2 require two orders of magnitude more iron than non-diazotrophic organisms (Raven 1988), and iron has been shown to limit the growth of N -fixing cyanobacteria in some 2 lakes (Elder & Horne 1977, Wurtsbaugh & Horne 1983, Hyenstrand et al. 2001). Introduction 7 Cyanobacteria have been the focus of interest in the Baltic Sea because of their increasingly frequent annual occurrence (Kahru et al. 2007); other algal blooms, which may also be harmful, have received less attention. Other potentially harmful species occur regularly in Baltic plankton (e.g. Leppänen et al. 1995, Lindholm & Öhman 1995, Hällfors 2004, Kuuppo et al. 2006), and mass occurrences of toxic dinoflagellates (Pertola et al. 2005, Hajdu et al. 2006, Kremp et al. 2009) and prymnesiophytes (Dahl et al. 1989, Lindholm & Virtanen 1992, Hajdu et al. 2008) are not unusual. Sometimes the blooms can have dramatic effects, as in 1988 in Skagerrak-Kattegat, where massive blooms of the toxic prymnesiophyte Chrysochromulina polylepis Manton & Parke caused mortality on all trophic levels, from phytoplankton to zooplankton, benthic macroalgae, fauna and fish (Dahl et al. 1989, Lindahl & Dahl 1990, Nielsen et al. 1990). Blooms of another prymnesiophyte, Prymnesium parvum Carter, have been associated with fish kills in Finland and Sweden (Lindholm & Virtanen 1992, Holmquist & Willén 1993, Lindholm et al. 1999). Nutrient availability and nutrient ratios may trigger the initiation of such blooms (Dahl et al. 2005, Hajdu et al. 2005). For example the abundance of Chryschromulina species has been associated with high N:P ratios (Dahl et al. 2005) while blooms of the invasive, potentially toxic dinoflagellate Prorocentrum minimum (Pavillard) Schiller have been correlated with high nutrient concentrations (Hajdu et al. 2005, Pertola et al. 2005). Knowledge of the responses of phytoplankton communities to changes in nutrient load and ratios is essential for making decisions in water management issues (Conley 2000, Olsen et al. 2001). It has been debated whether reductions in phosphorus, nitrogen or both are needed to deal with eutrophication in the Baltic Sea (e.g. Boesch et al. 2006). Some studies have indicated that in the short term a reduction in the N-load may increase the biomass of N -fixing cyanobacteria, due to their competitive advantage 2 during N-limitation (Elmgren & Larsson 2001). It has even been suggested that any reduction in the N-load may be offset by increased N -fixation, making N reduction 2 useless (Hellström 1996, Schindler et al. 2008). 1.2. Nutrient regulation of phytoplankton growth Phytoplankton biomass accumulation is a function of growth rates and losses. Growth rate and productivity are often regulated by the availability of resources (bottom-up regulation), such as light (e.g. Huisman et a. 2004), temperature (e.g. Hagström et al. 2001) or nutrients (e.g. Hecky & Kilham 1988), while losses are due to grazing (top-down regulation) (e.g. Carpenter et al. 1985, Sterner 1989, Kagami et al. 2002), sedimentation out of the photic zone (e.g. Heiskanen 1998), and viral and fungal parasitism (e.g. Suttle et al. 1990, Bratbak et al. 1993, Brussaard 2004). According to Liebig´s law of the minimum, the yield of any organism is limited by the factor present in the lowest amount in relation to its requirements (de Baar 1994). In most systems phytoplankton production is limited by the availability of light or the supply 8 Introduction of N and/or P (Schindler 1978, Hecky & Kilham 1988, Downing 1997). The general paradigm is that N is the nutrient most often limiting production in marine waters (Fisher et al. 1992, Oviatt et al. 1995, Howarth & Marino 2006), as well as in estuaries and coastal marine systems, while freshwater phytoplankton tends to be P-limited (Schindler 1974, Hecky & Kilham 1988, Smith 2003). However, this paradigm has been the subject of controversial debate (e.g. Hecky & Kilham 1988, Tyrrell 1999, Schindler et al. 2008). It is indeed obvious that nutrient limitation patterns vary both spatially and seasonally; moreover, co-limitation by both nitrogen and phosphorus is common in both freshwaters and marine waters (Elser et al. 1990, Kivi et al. 1993, Maberly et al. 2002, Howarth & Marino 2006, Smith 2006, Elser et al. 2007). Diatoms and some chrysophytes may additionally be limited by the availability of silica (Si), since they need Si in large amounts for their frustules (Egge & Aksnes 1992, Nelson & Dortch 1996). In eutrophicated waters the high supply of N and P often increases diatom growth and sedimentation, which in turn enhances Si accumulation in the sediments (Conley et al. 1993). This may lead to reduced Si concentrations and Si limitation in the water (Papush & Danielsson 2006), which will favour the growth of flagellates rather than diatoms (Smayda 1990, Wasmund & Uhlig 2003, Conley et al. 2008) and may also affect the diatom species composition (Olli et al. 2008). Ultimately, this shift in phytoplankton composition may cause major changes in the entire food web and may also lead to harmful algal blooms (Smayda 1990, Conley et al. 1993, Conley et al. 2008). In some situations phytoplankton growth may also be limited by the availability of certain trace elements; oceanic phytoplankton, for instance, has been shown to be limited by the availability of iron (Kolber et al. 1994, Coale et al. 1996, Hopkinson et al. 2007). Phytoplankton on average requires C, N and P in an approximate molar ratio of 106:16:1, the Redfield ratio (Redfield 1958). Deviations from this optimal ratio can be used to infer nutrient limitation of phytoplankton growth. However, species differ in their P and N requirements and the kinetics of nutrient uptake, and may thus have different optimal N:P ratios (Rhee & Gotham 1980, Hecky & Kilham 1988, Quigg et al. 2003, Klausmeier et al. 2004). In laboratory cultures, optimal molar N:P ratios measured for different phytoplankton species lie in the range between 7 and 84 (Rhee 1978, Healey & Hendzel 1979, Rhee & Gotham 1980, Smith 1982). The species-specific optimal ratios may vary depending on different factors, e.g. growth rate (Goldman et al. 1979, Terry et al. 1985, Elrifi & Turpin 1985, Turpin 1986), temperature (Tilman et al. 1986) light conditions (Healey 1985, Goldman 1986) or CO availability (Burkhardt & Riebesell 2 1997). In addition to ambient nutrient concentrations, the availability of nutrients for phytoplankton is affected by the regeneration rate of nutrients in the food web (e.g. Dugdale & Goering 1967, Andersen et al. 1991, Gaul et al. 1999). When primary Introduction 9 production is fuelled by recycled nutrients in the food web, it is referred to as regenerated production; production based on the external input of nutrients – i.e. from land and deep water, atmospheric fallout, allochthonous supply, and nitrogen fixation – is termed new production (Dugdale & Goering 1967, Eppley & Peterson 1979). Both the rate of nutrient cycling and the fate of new nutrient inputs in the system depend on the structure and function of the whole food web (Heiskanen et al. 1996, Verity & Smetacek 1996). Knowledge of the food web structure and of the mechanisms structuring community composition is thus of central importance in understanding the impact on these systems of environmental changes such as eutrophication. 1.3. Coastal food webs Food webs are divided into trophic levels, in which the first level, i.e. the base of the food web, is formed by primary producers (Fig. 1). In the pelagic food web, primary producers consist of picoplankton (0.2-2 µm), nanophytoplankton (2-20 µm) and large microphytoplankton (20-200 µm). Picoplankton consists of both prokaryotes (unicellular cyanobacteria) and eukaryotes (Stockner & Antia 1986). Nanophytoplankton is usually dominated by flagellates, while diatoms, dinoflagellates and filamentous cyanobacteria are the most common microphytoplankton. Large phytoplankton may be preyed upon Figure 1. Simplified schematic illustration of the major pathways of flow of energy and nutrients in the coastal food web. 10 Introduction by herbivorous zooplankton, which again serves as food for carnivorous zooplankton; this in turn forms an important food for small fish and mysids. The flow of energy from phytoplankton via zooplankton to fish is known as the classic herbivorous food chain (Hairston et al. 1960, Carpenter et al. 1985, 1987). Part of primary production is excreted or lost from the phytoplankton in the form of dissolved organic material (DOM) (Lignell 1990). DOM is also released due to incomplete ingestion and digestion by grazers, i.e. “sloppy feeding” (e.g. Lampert 1978, Strom et al. 1997, Titelman et al. 2008), and by excretion and leakage from their fecal pellets (e.g. Jumars et al. 1989, Urban-Rich 1999) as well as due to viral-induced cell lysis (Fuhrmann 1999, Riemann et al. 2009). This DOM is utilized by heterotrophic bacteria, which together with photosynthetic picoplankton form the basis of the microbial food web (Azam et al. 1983, Sherr & Sherr 1988, Titelman et al. 2008). The dominant grazers on bacteria and autotrophic picoplankton are heterotrophic flagellates (Kuuppo-Leinikki 1990, Kuuppo-Leinikki et al. 1994, Gasol et al. 2002), which in turn are grazed by ciliates and other protozoa and small zooplankton (Azam et al. 1983, Bernard & Rassoulzadegan 1990). Part of this ingested material is passed up the food chain to larger zooplankton, reconnecting the microbial food web with the classic food chain; part is regenerated into the water column and re-utilised by primary producers. Nutrient regeneration is of central importance in the microbial food web, but is difficult to measure since regenerated nutrients are so rapidly taken up by other cells. Omnivory, i.e. the ability of organisms to obtain food from more than one trophic level, and mixotrophy, their ability to gain nutrition through a combination of autotrophy and heterotrophy, are common in pelagic food webs, and further add to the complexity of the food web structure. Viruses are considered as part of the planktonic food web and may play an important role in regulating phytoplankton (Suttle et al. 1990, Bratbak et al. 1993, Brussaard 2004), bacteria (Fuhrmann 1999, Tuomi & Kuuppo 1999, Riemann et al. 2009) and protists (Garza & Suttle 1995). Energy transfer through the microbial food web is less efficient than through the classic food chain because of the increased average number of trophic links (Fenchel 1988, Pomeroy & Wiebe 1988, Berglund et al. 2007), but is considered highly significant for total energy throughput in the system (Pavés & González 2008). The importance of the microbial food web differs both seasonally and between different systems. The classic or herbivorous food chain has been considered more important in nutrient- rich waters, while the microbial food web is predominant in nutrient-constrained environments where productivity is based on nutrients regenerated within the system (Legendre & Rassoulzadegan 1995). The opposite, however, may also be true: high nutrient availability can stimulate the growth of either predation-resistant inedible algae or fast-growing small opportunistic primary producers which are not accessible to larger zooplankton, leading to a dominance of the microbial food web and reduced energy transfer to higher trophic levels (e.g. Andersson et al. 2006).