Alterations in nutrient limitations — Scenarios of a changing Baltic Sea

Available online at www.sciencedirect.com

s 73 (2008) 263–283www.elsevier.com/locate/jmarsys

Journal of Marine System

Alterations in nutrient limitations — Scenariosof a changing Baltic Sea

Å. Danielsson ⁎, L. Papush, L. Rahm

Department of Water and Environmental Studies, Linköping University, SE-581 83 Linköping, Sweden

Received 19 June 2006; accepted 12 October 2007Available online 2 January 2008

Abstract

Previous trend studies have shown increasing nitrogen and phosphorus as well as decreasing silica concentrations in the watermass of the Baltic Sea. This has had an impact on the amount of primary production, but also on the quality and succession ofplankton species. Present study examines the spatial and temporal patterns of potential nutrient limitations in the Baltic Sea for thetime period 1970–2000. Generally, low concentrations of DSi can limit the diatom blooms and such conditions are found in theGulf of Riga and Gulf of Finland during spring and summer. Nutrient ratios, DSi:DIN, DSi:DIP and DIN:DIP, are often used todetermine which nutrient may limit the primary production. Annual long-term temporal trends of silica to inorganic nitrogen andphosphorus respectively show consistent decreasing patterns. The largest slopes are detected during spring and summer for DSi:DIN and during spring for DSi:DIP ratios. For the DIN:DIP ratio significant slopes are only found in a few locations despiteincreasing levels for both nutrients, displaying a large variation in trends. In the open Baltic Proper the present trends are positiveduring winter and negative during spring and autumn. Gulf of Finland and Gulf of Riga are areas where both DSi:DIP and DSi:DINratios are found close to the Redfield ratios for diatoms. Together with the evaluated trends these suggest that the Gulfs maybecome silica limited in a relatively near future. These findings give some implications on the development and impact of changingnutrient concentrations.© 2008 Elsevier B.V. All rights reserved.

Keywords: DSi; DIN; DIP; Nutrient ratios; Redfield; Limitation; Trends; Baltic Sea

1. Introduction

Approximately one third of the global primary pro-duction is originated in marine waters and coastal watersrepresenting 30–50% of the global oceanic new produc-tion (Paerl et al., 1995). The size of this production largelydepends on the availability of inorganic nutrients, and the

⁎ Corresponding author.E-mail address: [email protected] (Å. Danielsson).

0924-7963/$ - see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.jmarsys.2007.10.015

relationship between them. The average composition ofcarbon and nutrients assimilated in algae generallyfollows the molar ratio C106:N16:P1, as first suggestedbyA.C. Redfield (see Redfield et al., 1963). However, thepicture is more complex involving e.g. Fe (Blomqvistet al., 2004) as well as light and turbulence (Huismanet al., 2004). Deviations in pelagic nutrient concentrationsfrom these proportions have been used as indicators ofprimary production limitation. While the nutrient ratios inoceanic water are close to the Redfield one (Tyrell, 1999),adjacent seas like the Baltic Sea deviate strongly from thisratio showing large regional and seasonal differences.

mailto:[email protected]

http://dx.doi.org/10.1016/j.jmarsys.2007.10.015

264 Å. Danielsson et al. / Journal of Marine Systems 73 (2008) 263–283

Marine environments are often assumed nitrogen (N)limited (Nixon et al., 1996), while freshwaters are mainlythought of as phosphorus (P) limited (Schindler, 1974).This is not entirely true, and often both nitrogen andphosphorus are limiting depending on time of year andlocation.

TheRedfield ratiowas later extended to incorporate alsosilica; C106:N16:P1: Si16 (Harrison et al., 1977). Silica (Si) ismainly used in forming diatom shells, but is also found inmany terrestrial plants etc. In contrast to the two above-mentioned nutrients N and P it is not directly influenced byhuman activities as it originates from weathering in theform of dissolved silica (DSi). It is often assumed to be inexcess compared to N and P. Damming usually decreasessilica load significantly and recent evidences show thatsilica has become a potential limiting factor in some coastalwaters (e.g. Conley et al., 2000). This limitation will affectthe plankton composition, which in a longer perspectivemay change the entire food web (Humborg et al., 1997).

There is also a more long-term aspect to take intoaccount. Globally the anthropogenic N input to coastalwaters has increased by a factor of two to three over recentdecades (Jickells, 1998). In the Baltic Sea, the corre-sponding N change is a factor four and for phosphorus theload had increased about eight times, mainly during thesecond half of the 20th century (Larsson et al., 1985).The Baltic Sea shows, due to this increased load, manysigns of eutrophication (Rönnberg and Bonsdorff, 2004).Previous studies have found increasing trends of N and Pand simultaneously decreasing trends of Si (Papush andDanielsson, 2006). Not surprising, decreasing coastalinorganic DSi:DIN ratios have also been observed (Rahmet al., 1996). Normally these changes have consequencesalso for the nutrient limitation patterns. Potential nutrientlimitation has been studied for a long time in the BalticSea (e.g. Howarth, 1988; Granéli et al., 1990; Lignellet al., 1993). Changes in nutrient ratios, linked to a re-duction of spring bloom diatoms, has already been ob-served (Wasmund andUhlig, 2003). It has been found thatthe diatom spring bloom is often terminated by the onsetof thermal stratification and depletion of, primarily,nitrogen (Smetacek, 1985; Heiskanen and Kononen,1994). A reduced DSi load at the same time as pro-gressing N and P loads, will result in changes in nutrientratios with the potential to modify the species compositionof the plankton community. Consequently, the presentstudy focuses on the spatial and temporal patterns of po-tential nutrient limitation in the Baltic Sea. This is dividedinto two objectives — silica limitation by low concentra-tion and alterations due to changing nutrient ratios. Thefindings are discussed in relation to the Redfield ratiosalong with possible implications on the ecosystem.

2. Data material and methods

2.1. Study area

The brackish Baltic Sea is located in Northern Europe(Fig. 1). It is a non-tidal, semi-enclosed water area,where shallow sills between the North Sea and BalticSea restrict the water exchange causing long hydraulicresidence times — about three decades on average(Stigebrandt and Wulff, 1987). The circulation andmixing are mainly forced by wind stress and heat flux,which results in an anti-clockwise residual circulation ineach major basin (Sjöberg, 1992). The freshwater dis-charge is limited and represents about half the brackishoutflow (Wulff et al., 2001). These conditions result inan area with a well-mixed surface layer and a saline deeplayer separated by a strong halocline. This halocline iswell below the entrance sills to the North Sea and limitsthe recycling of nutrients and the supply of oxygen to thedeep layers. Organic matter degradation often leads tohypoxia in these deep layers. Matthäus and Franck(1992) have shown how the infrequent major inflowsgovern the redox conditions in the deep basins in consortwith the organic matter load.

2.2. Nutrient load and trends

The Baltic Sea is a shallow, brackish sea shared by 14countries with a total drainage area about four times largerthan the sea area. The population is ~85 million in-habitants, with the majority located in its southern parts(Sweitzer et al., 1996). This gives a north–south gradientin nutrient loads with up to four times higher loads in thesouthern parts related to the distribution of population,agricultural activities, major rivers and point sources. TheGulf of Finland is strongly influenced by the largest riverin the Baltic Sea, Neva, and the largest coastal city, StPetersburg, while the Gulf of Riga receives its highnutrient load mainly from river Daugava and the city ofRiga. The annual atmospheric deposition of dissolvedinorganic nitrogen doubled during 1955–90, and nowrepresents about a quarter of the total N input to the Sea(Granat, 1990; HELCOM, 2005). These loads have led toa significant increase in nutrient concentration in thewatermass (Sandén et al., 1991; Sandén and Rahm, 1993;Kuparinen and Tuominen, 2001; Papush and Danielsson,2006). The degradation in environmental quality hastaken such proportions that an intergovernmental organi-sation, HELCOM, has been formed to counteract theseproblems. Today remedy actions have decreased the in-puts of nutrients, e.g. by improved wastewater treatment(HELCOM, 2002).

265Å. Danielsson et al. / Journal of Marine Systems 73 (2008) 263–283

Silica is another story. It is mainly a result of chemicalweathering controlled by temperature and vegetation(Berner andBerner, 1996).Alterations ofweathering rateshave not been affected by anthropogenic activities to suchdegree that it has caused a major impact of silica loads tothe water mass (Conley et al., 2000). River regulation, onthe other hand, has decreased the silica loads significantly

Fig. 1. The Baltic Sea and its major basins, i

as a result of increased water residence time, diatomgrowth and sedimentation (Humborg et al., 2000). Withinthe Baltic Sea eutrophication plays an essential part in thesilica cycle. An intensified algae bloom followed by anincreased deposition and preservation of diatomaceousbiogenic silica will consequently exhaust the silica pool inthe water column (Conley et al., 1993).

ncluding the 60 m water depth isoline.


2.3. Data material

Nutrient data are retrieved from the Baltic Envi-ronmental Database (http://data.ecology.su.se/Models/bed.htm). It is an environmental database includinghydrographical and chemical data based on nationalmonitoring data from the Baltic riparian countries (seeAcknowledgement for a list), but also from variousresearch projects (see Sokolov et al., 1997). Data is con-tinuously updated and quality checked.

The data used in present study include dissolvedsilicate (DSi), dissolved inorganic nitrogen (DIN; NO3+NO2+NH4) and dissolved inorganic phosphorus (DIP;PO4). The concentrations were analysed by methodsaccording to Grasshoff et al. (1983) in accordance withrecommendations by HELCOM (2000). The study spansover the last three decades, i.e. 1970–2000. There is nomajor difference in spatial coverage for different timeperiods, which should reduce possible bias due to varyingsampling frequency. There is a typical seasonality in thenutrient distribution (Fig. 2). Due to this the data weredivided into four seasons: January–March (Winter),April–June (Spring), July–September (Summer) andOctober–December (Autumn). For each sampling eventthe data is vertically averaged over the upper, well-mixed20 m to represent the photic zone. This water depth is anapproximation as it varies slightly over time (seasonally)and in space, but justifiable with regard to the threedecades of data to be screened.

2.4. Trend analysis

Temporal trend analysis was performed using theseasonal Mann–Kendall test elaborated by Hirsch et al.(1982) and modified by Hirsch and Slack (1984). It is anon-parametric algorithm resistant to the presence ofoutliers. The seasonal Kendall's estimator (Hirsch et al.,1982) was used to determine the slope of the trend. Thesemethods are frequently used methods for trend analysisin hydrological and hydrochemical time series analysis.They have also been applied to estimate nutrient trends inthe Baltic Sea (e.g. Sandén et al., 1991; Sandén andRahm, 1993; Papush and Danielsson, 2006).

3. Results

3.1. Nutrient concentrations and the occurrence of lowDSi levels

For all three nutrients (DSi, DIN and DIP) the meanvalues are slightly higher than the median indicating arightly skeweddistributionwith a fewhigh concentrations.

For DIN and DIP the highest mean concentrations arefound in the southern parts of the Baltic Sea, including thetwo Gulfs. The opposite is true for DSi, where the highestaverage concentrations are found in the northernmostbasins (Fig. 2). All basins show a typical seasonality forall three nutrients. During winter the nutrient pool is builtup but in spring a major drop is seen, which can bereferred to the spring bloom. In some basins, effects of latesummer blooms are discernible as a drop in the nutrientconcentration.

The high DSi concentrations in Bothnian Sea andBothnian Bay exclude the occurrence of low (potentiallylimiting) DSi concentrations. For the other three basins,Baltic Proper and the Gulfs of Riga and Finland, theoccurrences of low DSi concentrations are common(Fig. 3). The frequency varies over time, without a generalpattern (Fig. 3). In the Baltic Proper, the occurrence of lowDSi concentrations seems to be around 15–20% of theobservations over the entire time period, with a slightincrease in the occurrence of DSi observations ≤2 μMover time. Low DSi concentrations in the two Gulfs occurmore frequently, but the number of observations is fewerirrespective of N and P. In the Gulf of Riga, a majority ofthe measured concentrations are even below 2 μMindicating a possible silica limitation. In the Gulf ofFinland the number of very low DSi measurements wasmainly found in the mid 1990s. Low DSi concentrationsare more frequent during spring and summer than duringautumn andwinter (Fig. 4). These conditions are generallyfound in the Gulf of Riga, the Gulf of Finland and in thecoastal areas of the Baltic Proper. A majority of theobservations are limiting in both the western part of Gulfof Riga (spring) and in the inner parts of Gulf of Finland(summer). During autumn, these levels are more seldomseen, although frequencies up to 20% are not uncommon.

3.2. Seasonal patterns of nutrient ratios

Fig. 5 presents the seasonal pattern of nutrient ratios fordifferent basins. They are all compared to their respectiveRedfield ratio, which for DSi:DIN is equal to 1 and forboth DSi:DIP and DIN:DIP are 16. By subtracting theRedfield ratio from the observed ratios it is possible toview possible limitation patterns as deviations from theRedfield ratio. In all basins and for all three nutrient ratiosthere is a seasonal cycle similar to the one for singlenutrients (Fig. 5). Studying DSI:DIN and DSi:DIP it isseen that a possible silica limitation only occurs in theGulf of Riga (Fig. 5a). In the other basins, the DSi-ratiobars are well above the Redfield ratio, especially duringspring and summer when the nutrients are consumed inthe phytoplankton production.

http://data.ecology.su.se/Models/bed.htm

http://data.ecology.su.se/Models/bed.htm

Fig. 2. Interannual variation of mean DSi (μM), DIN (μM) and DIP (10⁎ μM) concentrations. It represents the averaged cycles for 1970–2001.



Focusing on the DIN:DIP ratio there is a larger var-iation in time and between basins (Fig. 5b). In the BalticProper, the DIN:DIP ratio is always below the Redfieldone suggesting a N-limitation all over the year. The op-posite is the Bothnian Bay with very high DIN:DIP ratiosand where there is a P-limitation throughout the year. Theother basins give more complex pictures, with patternschanging over time. The Gulf of Riga shows signs ofP-limitation during the productive season (spring andsummer). In the Gulf of Finland, there seems to be anN-limitation, with the only exception for some highpeaks in late summer/autumn. The Bothnian Sea, withits location between the Baltic Proper and the BothnianBay, gives a picture where the DIN:DIP varies over theyear following a seasonal cycle.

3.3. Annual variations of nutrient ratios

On a three-decade basis, the nutrient ratios show evenmore clear tendencies. Fig. 6 presents the annual winterratios (expressed as deviations from the Redfield ratio)over time. The DIN:DIP ratio in the Baltic Proper iscontinuously below the Redfield one with relatively smallvariations (Fig. 6b). The two silica ratios, on the otherhand, show rapidly decreasing ratios in the 1970s(Fig. 6a). From 1980 onwards these ratios are levellingof or increase slightly. In all the remaining basins, there isa shift in nutrient ratios in the 1990s related to the nitrogenconcentration, i.e. DIN:DIP is less than the Redfield ratiowhile DSi:DIN is above. DSi:DIP, on the other hand,seems to be more or less unaffected over time, except forthe Gulf of Riga where it goes from below to above thecritical Redfield ratio.

3.4. Temporal trends of nutrient ratios

Temporal trends for the ratios are presented in Fig. 7,with slopes given for respective station (Table 1). Mostof the trends are found for the Baltic Proper stations andmainly during winter and autumn. During summer, veryfew significant trends could be detected. For the DSi:DIN ratios most of the stations present decreasingtrends, with an overall average slope of −0.12 per year.Seasonally, the largest number of trends is found duringwinter, but the largest decreasing slopes are obtainedduring spring and summer. In the Gulf of Riga there is aslight positive trend. During summer, one positive trend

Fig. 3. The occurrence of low DSi concentrations in the surface layer0–20 m. The average number of observation per year for each bar is forthe Baltic Proper 346 (69–1205), Gulf of Riga 118 (5–422) and forGulf of Finland 85 (11–265).


is also detected at station BY32 in the western part of theBaltic Proper. The smallest, by absolute value, negativetrends are found in the southern and eastern parts of theBaltic Proper, while the highest ones are found in theBothnian Sea.

For the DSi:DIP ratios the average slope is −0.72 peryear. Seasonally, the largest number of trends is foundduring wintertime, but with moderate slopes. Duringautumn, the situation is very similar to the winter period,but with slightly larger, negative slopes. The situation isrelatively homogenous in the Baltic Proper, while noneof the two Gulfs show any significant change over time.In the Bothnian Sea the trends are decreasing drastically,with the largest slope during spring.

For the DIN:DIP ratio significant slopes are only foundat five stations, with a large variation from −2.6 in theGulf of Riga to +2.4 in the Bothnian Bay. The stations inthe Baltic Proper present positive trends duringwinter andnegative ones for both spring and autumn. The only trendduring summer is found in the Gulf of Riga, with arelatively large decrease over time.

4. Discussion

The changes in nutrient concentration have led toalterations in the balance of nutrient ratios — somethingthat may play a role in the primary production. TheRedfield ratio is often used as an indicator of whichnutrient puts a limit to the growth. Present study focuseson nutrient ratios in a spatiotemporal context, with pos-sible implications for diatom blooms.

4.1. Effects of a silica limitation

Indications of silica depletion during the springdiatom bloom have been observed for the Baltic Sea(e.g. Munk-Sörensen and Nielsen, 1992; Conley et al.,2000). This limitation might be a result of theavailability of Si. The levels of 2 μM and 4 μM areused in present study to distinguish sea areas where thediatom uptake of DSi can be limited. In the early 1970sPaasche (1973) reported half-saturation constants forDSi uptake of 0.8–3.37 μM. Azam and Chrisholm(1976) also studied how DSi is taken up by diatom cellsand found half-saturation constant in a similar range,1.6–2.5 μM, at rather low concentrations of DSi (3.3–3.6 μM). On the basis of their enrichment experiment

Fig. 4. The spatial occurrence of low DSi concentrations. Each gridcell show the frequency of DSi ≤2 μM, summed over years 1970–2001. Winter = Jan.–Mar., spring = Apr.–Jun., summer = Jul.–Sep andautumn = Oct.–Dec. No of observation per cell ≥5.

Fig. 5. Monthly deviations from the Redfield ratio. The data represent monthly mean concentrations for years 1970–2001 for each basin. These aresubtracted by the Redfield ratio, which equals 1 for DSi:DIN and 16 for DSi:DIP respective DIN:DIP. The bar charts represent a) DSi:DIP and DSi:DIN b) DIN:DIP.


Fig. 5 (continued ).



Fig. 6. Yearly deviations from the Redfield ratio. The data represent annual mean winter (January–March) concentrations for each basin. These aresubtracted by the Redfield ratio, which equals 1 for DSi:DIN and 16 for DSi:DIP respectiveDIN:DIP. The bar charts represent a) DSi:DIP andDSi:DINb) DIN:DIP.



they concluded that some of species might be Si limitedif their half-saturation constants are higher than the onedetermined for the entire multispecies sample. Almosttwo decades later, Egge and Aksnes (1992) suggested athreshold level of DSi of 2 μM for diatom uptake of DSi.In a recent study carried out in the waters of theOslofjord (Norway) the half-saturation constant for

Fig. 7. Seasonal temporal trends for different nutrient ratios at selected statiob) DSi:DIP and c) DIN:DIP.

silica uptake was found in the range 1.7–11.5 μM(Kristiansen et al., 2000). The upper range, however,corresponds to the maximal constant value measured inthe beginning of spring bloom, while the lower range isthe minimal value found at the end of the bloom. Thehalf-saturation constant in the fjord samples decreasedduring the spring bloom along with decreasing silicic

ns for the 1970–2001. Data represent seasonal trends for a) DSi:DIN,



acid concentration. Also specific uptake rates decreasedwhen the DSi dropped. This means that the DSi governsthe diatom growth under spring bloom even if limitedsilicon uptake does not necessarily mean that the diatomgrowth is significantly restricted. This suggests thatsome diatom species can experience limitation whileothers are in favourable conditions with respect to DSi.This can result in shifts within the diatom communitytoward species having lower requirements for DSi.

Smayda (1990) proposed that increases in theglobally observed toxic phytoplankton blooms were adirect consequence of increased N and P concentrationswith modifications on the Si:N and Si:P ratios. Thisfavouring of algae other than diatoms has been seen alsoin Scandinavian waters (Maestrini and Granéli, 1991).Observations of a decreasing diatom population infavour of dinoflagellates during spring bloom during thelast decades have also been reported (HELCOM, 1996).



However, in e.g. the Mecklenburg Bight and the ArkonaSea diatoms are still dominating (Wasmund et al., 1998).On a local scale, the community composition can besensitive to inorganic nitrogen and phosphorus concen-trations (Gasiunaitè et al., 2005). The diatom springbloom is often terminated by the depletion of nutrientsand the onset of thermal stratification restricting thesupply of nutrients from underlying water masses(Smetacek, 1985; Heiskanen and Kononen, 1994).Differences in community composition also vary withlight, temperature, salinity, nutrient concentration,exposure to wind actions etc. The small diatoms are

mainly affected by salinity distribution, while theabundance of the larger ones is more a function ofwind exposure (Ulanova and Snoeijs, 2006). Otherfactors affecting the abundance are the regeneration rate,competition between algae and bacteria, zooplanktongrazing, turbidity and light conditions (Pitkänen et al.,1986; Kivi et al., 1993; Cloern, 1999; Hänninen et al.,2000). In addition, there is a competition of speciesand variations between and within species dependingon environmental conditions and physiological states(Brzezinski, 1985). There is an observed decrease inabundance of diatom species with increasing salinity in

Table 1Long term annual trends for the entire year (total) and for separate seasons

Station DSi:DIN DSi:DIP DIN:DIP

Total Wi Sp Su Au Total Wi Sp Su Au Total Wi Sp Su Au

Baltic ProperBY5 −0.08 −0.16 n.s. n.s. n.s. −0.75 −0.83 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.BMP L1 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. −0.76BMP K1 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.BY15 −0.08 −0.1 n.s. n.s. n.s. −0.69 −0.58 −1.00 n.s. −0.73 n.s. 0.08 −0.21 n.s. n.s.BY31 −0.12 −0.12 n.s. n.s. n.s. −0.66 −0.62 −4.40 n.s. −0.41 0.05 0.10 n.s. n.s. −0.41BY32 −0.08 −0.93 n.s. 2.05 n.s. −0.48 −0.52 n.s. n.s n.s. 0.05 0.08 n.s. n.s. n.s.BY38 −0.12 −0.12 n.s. n.s. −0.14 −0.80 −0.74 n.s. n.s. −1.00 n.s. n.s. −0.130 n.s. −0.12BY27 −0.17 −0.17 −2.22 n.s. n.s. −1.03 −0.74 n.s. n.s. −1.60 0.11 0.15 n.s. n.s. n.s.

GoFBMP H1 −0.1 −0.06 n.s. n.s. −0.27 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. 0.74

GoR0.7 n.s. n.s. n.s. 0.12 n.s. n.s. n.s. n.s. n.s. −2.64 n.s. n.s. −3.06 n.s.

Bothnian SeaSR5 −0.27 n.s. −2.40 n.s. −0.24 −6.45 n.s. −20.80 n.s. −4.95 n.s. n.s. n.s. n.s. n.s.US5B −0.18 −0.11 n.s. n.s. −0.16 −4.30 n.s. n.s. n.s. −4.37 n.s. n.s. n.s. n.s. n.s.

Bothnian BayBO3 n.s. n.s. n.s. −0.36 n.s. n.s. n.s. n.s. n.s. n.s. 2.4 n.s. 7.77 n.s. n.s.

Winter (Wi) = January–March, Spring (Sp) = April–June, Summer (Su) = July–September and Autumn (Au) = October–December. The numbersrepresent the annual significant slopes (α=0.05) derived from 1970–2001. n.s. denotes that no significant trends could be detected.


the Baltic Sea. Inherent is also large variations betweenspecies during a life cycle, as some species are able tostore nutrients which may give them a competitiveadvantage (Sakshaug and Olsen, 1986; Andersson et al.,1996).

4.2. Nutrient limitation patterns of the Baltic Sea

The major concern regarding possible nutrient limita-tion has been focused on the human influenced nitrogenand phosphorus (e.g. Granéli et al., 1990). Major loads ofnitrogen and phosphorus are found in the southern Baltic(e.g. Sweitzer et al., 1996), where also eutrophicationeffects are more pronounced (Rönnberg and Bonsdorff,2004). The Gulfs of Finland and Riga are heavily affectedby river inputs. Large loads of DSi are only found in thenorthern parts. The main focus is, however, on nutrientlimitation. The results are discussed for each major basinseparately to get a more comprehensive overview of thecharacteristics governing the different basins of the BalticSea.

4.2.1. Bothnian BayPrimary production is light and temperature controlled

both in onset and in size (Hagström et al., 2001). There areno indications that DSi will become potentially limiting

during at least the next five decades, based on presentconcentrations, ratios and temporal trends. The meanwinter DSi concentration is about 31 μM and shows along-term positive trend over time (Papush and Daniels-son, 2006). The Bothnian Bay is since long assumed to bephosphorus limited (e.g. Zweifel et al., 1993). There are,however, indications of nitrogen depletion in its estuariesduring summer (Andersson et al., 1996; Humborg et al.,2003). In present study, the mean DIN:DIP is above theRedfield ratio of 16:1, suggesting a possible DIP lim-itation. This ratio varies greatly over season with lowvalues during spring, which is probably due to low loadsduring base flow and high DIN loads during the forestspring flood (Humborg et al., 2003). TheDIN:DIP trend isincreasing over time. This is in concordance with Papushand Danielsson (2006) who presented both increasingtrends of DIN and decreasing ones of DIP in these surfacewaters. To summarise, the basin seems phosphorus lim-ited and this condition will probably be more pronouncedover time.

4.2.2. Bothnian SeaThe Bothnian Sea act as a transition zone between

Bothnian Bay and the Baltic Proper. There is an excess ofDIP during the cold season while the warmer one is morebalanced, reflecting the transient character of the Bothnian


Sea between the phosphorus limited Bothnian Bay and thegenerally nitrogen limited Baltic Proper. Previous studieshave suggested a N-limitation. The DSi concentrations arestill very high, but there are a few observations of DSib4 μM. The mean winter concentration is around 16 μM,with a negative trend (Papush and Danielsson, 2006).Extrapolation of this trend suggests, at least, a time span offour decades for DSi to reach limiting levels on a generalbasis. Both DSi:DIN and DSi:DIP show a DSi surpluscompared to DIN respectively DIP, but with a long-termdecrease (Fig. 7a). This is a reflection of the decreasingDSi, increasing DIN and no DIP trends (Papush andDanielsson, 2006). DIN:DIP, on the other hand, shows alarge variation over time, although no significant long-term trend could be detected. This ratio has changedtowards a strongN-limitation from having ratios switchingbetween above and below the Redfield one. This is in linewith a previous suggested DIN limitation (e.g. Anderssonet al., 1996), but also suggests that DSi might becomelimiting in a not too far future.

4.2.3. Gulf of FinlandThere are signs of a progressing eutrophication in this

sub-basin (Pitkänen, 1994) as shown by increasingnutrient trends (Kahma and Voipio, 1989; Perttilä et al.,1995; Hänninen et al., 2000). Previous studies havesuggested that nitrogen is limiting during summer andearly autumn (Granéli et al., 1990; Pitkänen, 1991; Kiviet al., 1993; Pitkänen et al., 2001). However, DSi seems toplay an interesting role in this basin. The DSi concentra-tions are frequently below 4 μM and even below 2 μMsomething that seems to become even more frequent overtime. It indicates a possibility for DSi to potentially limitthe diatom production, as the diatoms have difficulties inassimilating low DSi concentrations. With a meanconcentration of ~8 μM (Fig. 2) and an annual negativetrend of −0.2 μM year−1 DSi (Papush and Danielsson,2006) the potential DSi limitation will probably continueto grow. The DSi stock is replenished during winter butrapidly depleted during spring bloom (Fig. 2). Thereafterthe recovery phase starts, only interrupted by summerdiatom blooms. The riverine loads of DSi are small de-spite the large river Neva. This is caused by the large lakeLadoga from which Neva originates.

The only monitoring station used for analysing trendsis located in the mouth of the basin. Since the waterexchange with the Baltic Proper is significant due to itsdeep entrance (Perttilä et al., 1995), it may not reflect thesituation in the inner basin. Sandén and Danielsson (1995)showed that the nutrient concentrations decreased sig-nificantly from the inner part to the mouth, a result of thenutrient load coming from riverNeva and St Petersburg. In

addition, an increased P mobilisation in the deeper layersdue to the hypoxic conditions is reported for recent years,which probably influence the conditions in the surfacelayer. Still, the DSi:DIN ratio is significantly decreasing atthe mouth, although it is still above the Redfield ratio(Fig. 5). DSi:DIP has a similar behaviour but does notshow any significant trend.

The long-term behaviour of DIN:DIP is a decreasingtrend during autumn when the ratio is close the Redfieldone or even slightly below it. Hänninen et al. (2000)have found indications of decreasing N:P ratios duringsummer, something that was not found in present work.With a few exceptions, the DIN:DIP ratio indicates aDIN limitation on a long-term basis as well as on aseasonal one. This could be a result of the location of thestation at the mouth of the Gulf of Finland. Spatially, thelow DSi concentrations suggest that there is a potentiallimitation by DSi within the Gulf.

4.2.4. Gulf of RigaThe Gulf is a semi-enclosed shallow bay with nu-

merous signs of eutrophication (Seppälä and Balode,1999). Previously, phosphorus has been assumed thelimiting nutrient (Yurkovskis et al., 1993). This view waschallenged by Wassmann and Tamminen (1999a,b), whopresented a more complete picture in which both nitrogenand phosphorus may limit the primary production. Thisseasonal switch in limitation has been found also byMaestrini et al. (1997). In addition, in the late 1990sYurkovskis et al. (1999) presented indications of a silica-limited system. Recently, Põder et al. (2003) have found amore complex structure with both seasonal and spatialvariations in nutrient limitation. They showed that DSilimits the spring bloom in the central part of the Gulf, DIPat the mouth of river Daugava and DIN at the entrance tothe open Baltic Proper. This pattern in time and space isseen also in the present study.

The Gulf of Riga is an area with frequently observedconcentrations below2μMindependent on time of the year(Figs. 3 and 4). DSi decreases from about 17 μM duringthe first half of the year down to 3 μM in summer. In thebeginning of the 1990s there was an exceptionally intensediatom spring bloom by heavily silicified species (Olliet al., this issue). Such blooms can potentially exhaust theDSi stocks in the water column. On a long-term basis, theDSi concentrations increase significantly over time(Papush and Danielsson, 2006). This is reflected also inDSi:DIN (Table 1), which shows an increasing long-termtrend going from being at the same order as the Redfieldratio to be well above it. The DSi:DIP shows no significanttrends. The summer months are characterised by high DSi:DIP ratios while the rest of the year shows low ones.

http://dx.doi.org/doi:10.1016/j.jmarsys.2007.009



DIN:DIP shows an interesting decreasing trend. Thedevelopment of present winter ratios takes a most peculiarpattern over the years. While the first half of the 1990susually had DIN:DIP ratios well above the Redfield one itchanged for the rest of the period in such a way that DINbecomes a limiting nutrient. Also seasonally it shows alarge variation although the summer months indicate asurplus of DIN to DIP. Present study, as well as someprevious ones, indicate that this basin is drasticallychanging over time. The previously often mentioned po-tential DIP limitation is probably changing to a morepronounced DSi but also DIN one.

4.2.5. Baltic ProperThe open Baltic Proper is generally assumed to be

nitrogen limited (Granéli et al., 1990; Wasmund et al.,1998), although most coastal areas are more or less phos-phorus limited due to their links to the terrestrial ecosystems(Elmgren and Larsson, 1997). This view of nitrogenlimitation has further changed with the occurrence ofextensive nitrogen-fixating cyanobacteria blooms. Thisnitrogen source is assumed to be of the same order as theatmospheric deposition of nitrogen (Savchuk and Wulff,1999; Rahm et al., 2000; Larsson et al., 2001). Thereby themagnitude of the summer bloom is mainly controlled bythe availability of phosphorus (Kivi et al., 1993; Savchukand Wulff, 1999). In fact, the role of the seasonalthermocline can be decisive in the regulation of the DIP(and DIN) supply from the intermediate layers and the latebloom. The intermediate layer between the seasonalthermocline and the halocline is impoverished on DIN insummer.However, amounts of DIP still are therewhich canbe supplied to the surface layer under given circumstances.There are also some indications of DSi limitation (Conleyet al., 2000).

DSi has an annual mean concentration of ~10 μM,with a negative trend of approximately −0.2 μM DSiyear−1 (Papush and Danielsson, 2006). This means that,on average, it will take with the present pace about threedecades to reach a potentially critical level of 4 μM. Ingeneral, about a fourth of all DSi observations are alreadybelow this level (Fig. 3), mainly occurring during springand autumn (Fig. 4) algae blooms. In the southern part afrequent occurrence of low DSi values is found thoughthey are also frequent along most stretches of the coast. Aspatial pattern with silica concentrations significantlyhigher than themedian on thewestern part of the basin hasalready been identified (Rahm and Danielsson, 2006).This spatial asymmetry is partly attributed to the largesilica inflow from the Bothnian Sea.

In present study, DSi:DIN shows decreasing trends inmost parts of the basin. This is a result of decreasing DSi

trends simultaneously with increasing DIN concentra-tions (Papush and Danielsson, 2006). This ratio is alwayswell above the Redfield one which makes DIN a prob-able limiting nutrient with regard to DSi but the frequentN-fixing cyanobacteria blooms make such reasoning use-less. The situation is similar, but slightly more pro-nounced, for DSi:DIP. It is always above the Redfield onesuggesting a surplus in DSi. Over time DSi:DIP and, butto a lesser extent, DSi:DIN decrease from very highratios in the early 1970s to aminimum in themiddle of the1990s after which they both increases again. Thesechanges in ratio over seasons are to a large part governedby DIN and DIP. A closer look at the DIN:DIP ratiosuggests that DIN is generally in short supply. It showsthree different trend patterns. In the southern part nosignificant trends are detected. During winter the trendsare generally increasing, while for other seasons the de-creasing ones dominate.

On a large-scale the Baltic Proper is relativelyhomogenous with regard to DSi, DIN and DIP. There areno significant differences between the sub basins (Sandénand Danielsson, 1995). On a finer scale there are coastalstretches significantly different from the overall conditions(Rahm and Danielsson, 2006). A high frequency ofobserved limitingDSi concentrations could be discerned inthe coastal areas. Presently DIN is characterized by lowconcentrations, but with present development a moregeneral DSi limitation is possible in future.

4.3. Primary production — diatoms

A changing nutrient composition in the Baltic Seashould have impacts on the primary production. DSiplays an important role in the ecosystem processes, bychanges in diatom species and food web dynamics(Officer and Ryther, 1980). The first study of diatomflora in the Baltic Sea was performed in the late 19thcentury by Juhlin-Dannfeldt (1882). Since then a numberof compilations have been made, where diatom com-munity and assemblages are used as indicators of eco-logical change. The spring bloom is dominated bydiatoms and dinoflagellates, where diatoms correspondon average to ~45% (Sommer, 1998). During the firstphase of the bloom the diatoms prevail. They havegrowth rates 5–50% larger than the one for dinoflagel-lates and often emerge as pioneer species in the planktonsuccession (Smayda, 1990; Snoeijs, 2001). In thenorthern Bothnian Bay diatoms dominate the primaryproduction over dinoflagellates (Alasaarela, 1979). Thisis also seen in Kattegat, where the observed sub-surfacemaxima during spring bloom may consist to ~90% ofdiatoms (Lund-Hansen et al., 2006).


Spatially, the spring bloom extent shows a regionalgradient with a three times increase from southwest tonortheast of the Baltic Proper (Schneider et al., 2006).Satellite images of the annual distribution of cyanobac-teria blooms show a pronounced asymmetry with minorblooms in the southern and south-eastern parts andfrequent ones along the western and northern parts (Kahruet al., 1994; RahmandDanielsson, 2006). The interannualvariations are, however, large as illustrated by the 2006blooms that showed the opposite pattern. During autumnthere is usually another bloom, now induced by frequentstorms increasing mixing and decreasing stratification.They cause increased nutrient levels in the photic zone.This gives rise to an autumn bloom with a communityconsisting of large phytoplankton, such as cold-waterdiatoms and dinoflagellates.

Diatoms rapidly consume inorganic nutrients andsink out of the water column (Kilham and Kilham,1980), the latter is facilitated by the shallow Baltic Sea.In contrary to dinoflagellates, they deposit on thebottoms supplying the benthic heterotrophic communitywith organic matter (Wasmund et al., 1998). A highproportion of the primary production is exported fromthe pelagic system, where the sinking loss rates may beas high as 30–60% per day of the suspended population(Heiskanen and Tallberg, 1999). However, silica re-cycles much slower than both nitrogen and phosphorus,which means that phosphorus can be used many timesduring its “life cycle” whereas silica is usually only usedonce (Conley et al., 1988).

Changing DSi levels will have impacts on the entirestructure of the phytoplankton community from diatoms tothose that do not require DSi, such as flagellates (Admiraalet al., 1990). Diatoms are the preferred food for manygrazers in the upper trophic layers. Turner et al. (1998) haveshown that due to lowered DSi concentrations theabundance of copepods have changed drastically, fromN75% to less than 30%of the total zooplankton population.This is seen globally, where dinoflagellates have increasedin relation to diatoms. Smaller non-siliceous phytoplanktonmay profit on the still available N and P, but those may notbe grazed due to lack of suitable predators. This can cause amajor change in the food web with impact on the fisheries(Turner et al., 1998; Ittekott et al., 2000).

Present study has shown a changing system in theBaltic Sea towards DSi depletion and changing nutrientratios. A more general DSi limitation is expected in theBaltic Proper and in the Gulf of Finland. In the Gulf ofRiga, the culmination of low DSi concentrations hasalready occurred. These changes have already been seenwith the stated ecological implications (Olli et al., thisissue).

5. Conclusions

Present study proposes evidence of an emergingsituation of low DSi concentrations and limiting condi-tions for diatoms. It also shows how the nutrient ratios arechanging over time implying alterations in the ecosystem.LowDSi concentrations are not a problemwithin the Gulfof Bothnia, where the concentrations are usually veryhigh. In the other basins, silica concentrations belowcritical limits are frequently found over time. The ratiotrends clearly indicate significant changes with generallydecreasing DSi:DIN and DSi:DIP trends. For DIN:DIPthe trends are also mainly decreasing but not so frequent.

Even though present large-scale analysis does notindicate a silica limitation in general, it clearly demon-strates that the there are a number of areas where lowconcentrations of DSi can limit the growth rates of diatomsand favour non-siliceous species. This persistent decreas-ing tendency even in the areaswith generally high levels ofDSi concentrations is a warning signal for unfavourablechanges in the ecosystem affected by anthropogenic ac-tivities, e.g. ongoing eutrophication in theBaltic Sea and inits catchments. Together these patterns show systems withchanging silica concentrations and ratios with probableeffects on the primary production and the ecosystem.

Acknowledgement

This research was a part of the EU-project Silicate andBaltic Sea EcosystemResponse (SIBER).We thankMiguelRoderiguez-Medina (SU) for supplying data from the BEDand Matti Perttilä (FIMR) and an anonymous referee forsuggesting significant improvements. The following mon-itoring organisations have contributed to BED: FIMR,Finland; EMI, Estonia;MarineMonitoringCentre, Instituteof Aquatic Ecology, Latvia; Lithuanian Maritime SafetyAdministration, Lithuania; Inst. of Meteorology and WaterManagement, Poland; Sea Fisheries Institute, Poland;Bundesamt für Seeschifffahrt undHydrographie, Germany;NERC; Denmark; SMF, UMF, SMHI and StockholmVatten, Sweden, and they are all greatly acknowledged.

References

Admiraal, W., Breugem, P., Jacobs, D., de Reuter van Steveninck, M.,1990. Fixation of dissolved silicate and sedimentation of biogenicsilicate in the lower river Rhine during diatoms blooms. Biogeochem-istry 9, 175–185.

Alasaarela, E., 1979. Ecology of phytoplankton in the north of theBothnian Bay. Acta Botanica Fennica 110, 63–70.

Andersson, L., Hajdu, S., Haecky, P., Kuparinen, J., Wikner, J., 1996.Succession and growth limitation of phytoplankton in the Gulf ofBothnia (Baltic Sea). Marine Biology 126, 791–801.




Azam, F., Chrisholm, S.W., 1976. Silicic acid uptake and incorporationby natural marine phytoplankton populations. Limnology andOceanography 21, 427–435.

Berner, E.K., Berner, R.A., 1996. Global Environment: Water, Air, andGeochemical Cycles. Prentice-Hall, Upper Saddle River, NewJersey. 376pp.

Blomqvist, S., Gunnars, A., Elmgren, R., 2004.Why the limiting nutrientdiffers between temperate coastal seas and freshwater lakes: a matterof salt. Limnology and Oceanography 49, 2236–2241.

Brzezinski, M.A., 1985. The Si–C–N ratio of marine diatoms—interspecific variability and the effect of some environmentalvariables. Journal of Phycology 21, 347–357.

Conley, D.J., Quigley,M.A., Schelske, C.L., 1988. Silica and phosphorusflux from sediments: importance of internal recycling in LakeMichigan. Canadian Journal of Fishery and Aquatic Sciences 45,1030–1035.

Conley, D.J., Schelske, C.L., Stoermer, E.F., 1993. Modification of thebiogeochemical cycle of silica with eutrophication. MarineEcology. Progress Series 101, 179–192.

Conley, D., Stålnacke, P., Pitkänen, H., Wilander, A., 2000. Thetransport and retention of dissolved silicate by rivers in Swedenand Finland. Limnology & Oceanography 45 (8), 1850–1853.

Cloern, J., 1999. The relative importance of light and nutrient limitationof phytoplankton growth: a simple index of coastal ecosystemsensitivity to nutrient enrichment. Aquatic Ecology 33, 3–15.

Egge, J.K., Aksnes, D.L., 1992. Silicate as a regulating nutrient inphytoplankton competition. Marine Ecology. Progress Series 83,281–289.

Elmgren, R., Larsson, U., 1997. Himmerfjärden: changes of a nutrient-enriched coastal ecosystem in the Baltic Sea. Swedish Environ-mental Protection agency. Report 4565. 197 pp.

Gasiunaitè, Z.R., Cardoso, A., Heiskanen, A.S., Henriksen, P.,Kauppila, P., Olenina, I., Pilkaityè, R., Purina, I., Razinkovas,A., Sagert, S., Schubert, H., Wasmund, N., 2005. Seasonality ofcoastal phytoplankton in the Baltic Sea: influence of salinityand eutrophication. Estuarine, Coastal and Shelf Science 65,239–252.

Granat, L., 1990. Estimating atmospheric nitrogen loading to the BalticSea. Ambio Special report 7, 20–21.

Granéli, E., Wallström, K., Larssson, U., Granéli, W., Elmgren, R.,1990. Nutrient limitation of primary production in the Baltic Seaarea. Ambio 19, 142–151.

Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of SeawaterAnalysis. Verlag Chemie, Weinheim. 419 pp.

Hagström, Å., Azam, F., Kuparinen, J., Zweifel, U.-L., 2001. Pelagicplankton growth and resource limitations in the Baltic Sea. In:Wulff, F., Larsson, P., Rahm, L. (Eds.), A System Analysis of theBaltic Sea. Ecological Studies, vol. 148, pp. 177–210.

Harrison, P.H., Conway, H., Holmes, W., Davis, D., 1977. Marinediatoms grown in chemostats under silicate or ammoniumlimitation. III. Cellular chemical composition and morphology ofChaetoceros debilis, Skeletonema costatum, and Thalassiosiragravida. Marine Biology 43 (1), 19–31.

Heiskanen, A.-S., Tallberg, P., 1999. Sedimentation and particulatenutrient dynamics along a coastal gradient from a fjord-like bay tothe open sea. Hydrobiologia 393, 127–140.

Heiskanen, A.-S., Kononen, K., 1994. Sedimentation of vernal and latesummer phytoplankton communities in the coastal Baltic Sea.Archiv fur Hydrobiologie 131 (2), 175–198.

HELCOM, 1996. Third periodic assessment of the Baltic Sea, 1989–1993; Background document. Baltic Sea Environment Proceedings64B, 252.

HELCOM, 2000. Manual for Marine Monitoring in the CombineProgramme of HELCOM Helsinki Commission RetrievedNovember 2004 from http://www.helcom.fi.

HELCOM, 2002. Environment of the Baltic Sea area 1994–1998.Baltic Sea Environment Proceedings 82B 215 pp.

HELCOM, 2005. Airborne nitrogen loads to the Baltic Sea. BalticMarine Environment Protection Comission. 23 pp. http://www.helcom.fi.

Hirsch, R.M., Slack, J.R., 1984. A nonparametric trend test for seasonaldatawith serial dependence.WaterResourcesResearch 20, 727–732.

Hirsch, R.M., Slack, J.R., Smith, R.A., 1982. Techniques of trendanalysis for monthly water quality data. Water Resources Research18, 107–121.

Howarth, R.W., 1988. Nutrient limitation of net primary production inmarine ecosystems. Annual Review of Ecology 19, 89–110.

Huisman, J., Sharples, J., Stroom, J., Visser, P.M., Kardinaal, W.E.A.,Vesphagen, J.M.H., Sommeijer, B., 2004. Changes in turbulentmixing shift competition for light between plankton species.Ecology 85, 2960–2970.

Humborg, C., Ittekkot, V., Cociasu, A.v., Bodungen, B., 1997. Effectof Danube River dam on Black Sea biogeochemistry andecosystem structure. Nature 386, 385–387.

Humborg, C., Schemewski, G., Bodo von Bodungen, Dannowski, R.,Steidl, J., Quast, J., Wallbaum, V., Rudolph, K.U., Mahlburg, S.,Muller, C.H., Erbguth, W., 2000. Marine science report OderBasin–Baltic Sea Interactions (OBBSI) Final Report No. 41,Institut fur Ostseeforchung Warnemunde.

Humborg, C., Danielsson, Å., Sjöberg, B., Gren, M., 2003. Nutrientland–sea fluxes in oligotrophic and pristine estuaries of the Gulfof Bothnia, Baltic Sea. Estuarine, Coastal and Shelf Science 56,1–13.

Hänninen, J., Vuorinen, I., Helminen, H., Kirkkala, T., Lehtilä, K.,2000. Trends and gradients in nutrient concentrations and loadingin the Archipelago Sea, Northern Baltic, in 1970–97. Estuarine,Coastal and Shelf Science 50, 153–171.

Ittekott, V., Humborg, C., Shäfer, P., 2000. Hydrological alterationsand marine biogeochemistry: a silicate issue? BioScience 50,776–782.

Jickells, T.D., 1998. Nutrient biogeochemistry of the coastal zone.Science 281, 217–222.

Juhlin-Dannfeldt, H., 1882. On the diatoms of the Baltic Sea. KungligaSvenska Vetenskapsakademiens Handlingar 6 52 pp.

Kahma, K., Voipio, A., 1989. Seasonal variation of some nutrients inthe Baltic Sea and the interpretation of monitoring results. FinnishInst. of Marine Research. Report C.M. 1989/C:31.

Kahru, M., Horstmann, U., Rud, O., 1994. Satellite detection ofincreased cyanobacteria blooms in the Baltic Sea: naturalfluctuation or ecosystem change? Ambio 23, 469–472.

Kilham, P., Kilham, S., 1980. The Evolutionary Ecology ofPhytoplankton. Univ. of California Press, Berkley, pp. 571–597.

Kivi, K., Kaitala, S., Kuosa, H., Kuparinen, J., Leskinen, E., Lignell,R., Marcussen, B., Tamminen, T., 1993. Nutrient limitation andgrazing control of the Baltic plankton community during annualsuccession. Limnology and Oceanography 38 (5), 893–905.

Kristiansen, S., Farbrot, T., Naustvoll, L.-J., 2000. Production ofbiogenic silica by spring diatoms. Limnology and Oceanography45 (2), 472–478.

Kuparinen, J., Tuominen, L., 2001. Eutrophication and self-puration:counteractions forced by large-scale cycles and hydrodynamicprocesses. Ambio 30 (4–5), 190–194.

Larsson, U., Elmgren, R., Wulff, F., 1985. Eutrophication and theBaltic Sea. Ambio 14, 9–14.

http://www.helcom.fi




Larsson, U., Hajdu, S., Walve, J., Elmgren, R., 2001. Baltic Sea nitrogenfixation estimated from the summer increase in upper mixed layertotal nitrogen. Limnology & Oceanography 46, 811–820.

Lignell, R., Heiskanen, A.-S., Kuosa, H., Gundersen, K., Kuuppo-Leinikki, P., Pajuniemi, R., Uitto, A., 1993. Fate of a phytoplank-ton spring bloom: sedimentation and carbon flow in the planktonicfood web in the northern Baltic. Marine Ecology. Progress Series94, 239–252.

Lund-Hansen, L.C., Casado De Amezua Ayala, P., Fragua Reglero, A.,2006. Bio-optical properties and development of a sub-surfacechlorophyll maxima (SCM) in southwest Kattegat, Baltic Sea.Estuarine, Coastal and Shelf Science 68 (1–22), 372–378.

Maestrini, S., Balode, M., Béchemin, C., Purina, I., Vérite, C., 1997.Nutrients limiting algal growth potential (AGP) in the Gulf ofRiga, eastern Baltic Sea, in spring and early summer 1996. La mer35, 49–68.

Maestrini, S.Y., Granéli, E., 1991. Environmental conditions andecophysiological mechanisms which led to the 1988 Chrysochromu-lina polylepis bloom: an hypothesis. Oceanologica Acta 14, 397–413.

Matthäus,W., Franck, H., 1992. Characteristics ofmajorBaltic inflows: astatistical analysis. Continental Shelf Research 12, 1375–1400.

Munk-Sörensen, H., Nielsen, K., 1992. Åhus bugt 1990–1991;Hydrografi, närsalter og plankton. Åhus Amt 1992.

Nixon, S.W., Ammerman, J.W., Atkinson, L.P., Berounsky, V.M.,Billen, G., Boicourt, W.C., Boynton, W.R., Church, T.M., Ditoro,D.M., Elmgren, R., Garber, J.H., Giblin, A.E., Jahnke, R.A.,Owens, N.J.P., Pilson, M.E.W.Q., Seitzinger, S.P., 1996. The fateof nitrogen and phosphorus at the land–sea margin of the NorthAtlantic Ocean. Biogeochemistry 35, 141–180.

Officer, C.B., Ryther, J.H., 1980. The possible importance of silicon inmarine eutrophication. Marine Ecology Progress Series 3, 83–91.

Olli, K., Clarke, A., Danielsson, Å., Aigars, J., Conley, D., Tamminen,T. Diatom stratigraphy and long-term dissolved silica concen-trations in the Baltic Sea. Journal of Marine Systems, this issue.doi:10.1016/j.jmarsys.2007.009.

Paasche, E., 1973. Silicon and the ecology of marine diatoms II.Silicate-uptake kinetics of five diatom species. Marine Biology 19,262–269.

Paerl, H.W., Pickney, J.L., Kucera, S.A., 1995. Clarification of thestructural and functional roles of heterocysts and anoxic micro-zones in the control of pelagic nitrogen fixation. Limnology andOceanography 40, 634–638.

Papush, L., Danielsson, Å., 2006. Silicon in the marine environment:dissolved silica trends in the Baltic Sea. Estuarine, Coastal andShelf Science 67, 53–66.

Perttilä, M., Niemistö, L., Mäkelä, K., 1995. Distribution, develop-ment and total amounts of nutrients in the Gulf of Finland.Estuarine, Coastal and Shelf Science 41, 345–360.

Pitkänen, H., Lehtoranta, J., Räike, A., 2001. Internal nutrient fluxescounteract decreases in external load: the case of the estuarialeastern Gulf of Finland, Baltic Sea. Ambio 30, 195–201.

Pitkänen, H., 1994. Eutrophication of the Finnish coastal waters:origin, fate and effects of riverine nutrient fluxes. Publication ofWater and Environment Research Institute of Finland 18, 1–44.

Pitkänen, H., 1991. Nutrient dynamics and trophic conditions in theeastern Gulf of Finland: the regulatory role of the Neva estuary.Aqua Fennica 21, 105–115.

Pitkänen, H., Niemi, J., Sarkula, J., Sandman, O., 1986. Phosphorus andNitrogenBalance of aCoastal Bay.Water Research Institute, Helsinki.

Põder, T., Maestrini, S.Y., Balode, M., Lips, U., Béchemin, C.,Andrushaitis, A., Purina, I., 2003. The role of inorganic andorganic nutrients on the development of phytoplankton along a

transect from Daugava River mouth to the open Baltic, in springand summer 1999. Journal of Marine Science 60, 827–835.

Rahm, L., Conley, D., Sandén, P., Wulff, F., Stålnacke, P., 1996. Timeseries analysis of nutrient input to the Baltic Sea and changing DSi:DIN ratios. Marine Ecology Progress Series 130, 221–228.

Rahm, L., Danielsson, Å., 2006. Spatial heterogeneity of nutrients inthe Baltic Proper, Baltic Sea. (revised).

Rahm, L., Jönsson, A., Wulff, F., 2000. Nitrogen fixation in the Balticproper: an empirical study. Journal of Marine Systems 25, 239–248.

Redfield, A., Ketchum, B., Richards, F., 1963. In: Hill, M. (Ed.), Theinfluence of organisms on the composition of seawater. The Sea,vol. 2. Wiley, New York, pp. 26–77.

Rönnberg, C., Bonsdorff, E., 2004. Baltic Sea eutrophication: area-specific ecological consequences. Hydrobiologia 514, 227–241.

Sandén, P., Danielsson, Å., 1995. Spatial properties of nutrientconcentrations in the Baltic Sea. Environmental Monitoring andAssessment 34, 289–307.

Sandén, P., Rahm, L., 1993. Nutrient trends in the Baltic Sea.Environmetrics 4, 75–103.

Sandén, P., Rahm, L., Wulff, F., 1991. Non-parametric trend test of theBaltic Sea data. Environmetrics 2, 263–278.

Sakshaug, E., Olsen, Y., 1986. Nutrient status of phytoplanktonblooms in Norwegian waters and algal strategies for nutrientcompetition. Canadian Journal of Fisheries and Aquatic Sciences43 (2), 389–396.

Savchuk, O., Wulff, F., 1999. Modelling regional and large-scaleresponse of Baltic Sea ecosystems to nutrient load reductions.Hydrobiologia 393, 35–43.

Schindler, D.W., 1974. Eutrophication and recovery in experimentallakes: implications for lake management. Science 184, 897–899.

Schneider, B., Kaitala, S., Maunula, P., 2006. Identification andquantification of plankton bloom events in the Baltic Sea bycontinuous pCO2 and chlorophyll a measurements on a cargoship. Journal of Marine Systems 59, 238–248.

Seppälä, J., Balode, M., 1999. Spatial distribution of phytoplankton inthe Gulf of Riga during spring and summer stages. Journal ofMarine Systems 23, 51–67.

Sjöberg, B. (Ed.), 1992. Sea and Coast. National Atlas of Sweden, SNA.Sokolov, A., Andrejev, O.A., Wulff, F., Rodriguez-Medina, M., 1997.

The data assimilation system for data analysis in the Baltic Sea.Contrib. Systems Ecol. StockholmUniversity, Stockholm, Sweden.

Smayda, T.J., 1990. Novel and nuisance phytoplankton blooms in thesea: evidence for a global epidemic. In: Granéli, E. (Ed.), Toxicmarine phytoplankton. Elsevier, London, pp. 29–40.

Smetacek, V., 1985. Role of sinking in diatom life-history cycles:ecological, evolutionary and geological significance.MarineBiology84, 239–251.

Snoeijs, P., 2001. Diatoms and environmental change in brackishwaters. In: Stermer, E.F., Smol, J.P. (Eds.), The Diatoms —Application for the Environmental and Earth Sciences. CambridgeUniversity Press, pp. 298–333.

Sommer, U., 1998. Biologische Meereskunde. Springer, Berlin. 475 pp.Stigebrandt, A., Wulff, F., 1987. A model for the dynamics of nutrients

and oxygen in the Baltic Proper. Journal of Marine Research 45,729–759.

Sweitzer, J., Langaas, S., Folke, C., 1996. Land cover and populationdensity in theBaltic Sea drainage basin: aGIS database.Ambio 25 (3),191–198.

Turner, R.E., Qureshi, N., Rabalais, N., Dortch, Q., Jusitc, D., Shaw,R., Cope, J., 1998. Fluctuating silicate:nitrate ratios and coastalplankton food webs. Proceedings of the National Academy ofSciences of the United States of America 95, 13048–13061.

http://dx.doi.org/10.1016/j.jmarsys.2007.009


Tyrell, T., 1999. The relative influences of nitrogen and phosphorus onoceanic primary production. Nature 400, 525–531.

Ulanova, A., Snoeijs, P., 2006. Gradient responses of ephilithic diatomcommunities in the Baltic Sea Proper. Estuarine, Coastal and ShelfScience 68 (3–4), 661–674.

Wassmann, P., Tamminen, T., 1999a. Eutrophication and sedimentationin the Gulf of Riga: an introduction. Journal of Marine Systems 23,1–10.

Wassmann, P., Tamminen, T., 1999b. Pelagic eutrophication andsedimentation in the Gulf of Riga: a synthesis. Journal of MarineSystems 23, 269–283.

Wasmund, N., Nausch, G., Matthäus, W., 1998. Phytoplankton springblooms in the southern Baltic Sea — spatio-temporal devel-opment and long-term trends. Journal of Plankton Research 20,1099–1117.

Wasmund, N., Uhlig, S., 2003. Phytoplankton trends in the Baltic Sea.ICES Journal of Marine Science 60, 177–186.

Wulff, F., Rahm, L., Hallin, A.-K., Sandberg, J., 2001. A nutrientbudget model of the Baltic Sea. In: Wulff, F., Rahm, L., Larsson, P.(Eds.), A Systems Analysis of the Baltic Sea. Ecological Studies.Springer Verlag, pp. 353–372.

Yurkovskis, A., Kostrichkina, E., Ikauniece, A., 1999. Seasonalsuccession and growth in the plankton communities of the Gulf ofRiga; Baltic Sea. Hydrobiologia 393, 83–94.

Yurkovskis, A., Wulff, F., Rahm, L., Andruzaitis, A.M., Rodriguez-Medina, M., 1993. A nutrient budget of the Gulf of Riga. BalticSea. Estuarine, Coastal and Shelf Science 37, 113–127.

Zweifel, U.L., Norrman, B., Hagström, Å., 1993. Consumption ofdissolved organic carbon by marine bacteria and demand forinorganic nutrients. Marine Ecology. Progress Series 101, 23–32.

Documents

Alterations in nutrient limitations — Scenarios of a changing Baltic Sea