Elsevier

Journal of Marine Systems

Volume 178, February 2018, Pages 75-85
Journal of Marine Systems

The role of life cycle processes on phytoplankton spring bloom composition: a modelling study applied to the Gulf of Finland

https://doi.org/10.1016/j.jmarsys.2017.10.010Get rights and content

Highlights

  • Our ecosystem model includes the life cycles of three main phytoplankton groups.

  • It is able to reproduce the observed trends in the spring bloom composition in the Baltic Sea.

  • Bloom formation of dinoflagellates has increased, of diatoms decreased over the past 30 years.

  • The time lag between dinoflagellate germination and growth is crucial for bloom formation.

  • Warming has reduced the time lag between germination and growth.

Abstract

Diatoms are typical representatives of the spring bloom worldwide. In several parts of the Baltic Sea, however, cold-water dinoflagellates such as Biecheleria baltica have become dominant during the past decades. We have investigated the mechanisms behind this trend by using an ecosystem model which includes the life cycles of three main phytoplankton groups (diatoms, dinoflagellates and cyanobacteria). Coupled to a water column model we have applied the model system for the period 1981–2010 to the Gulf of Finland. In agreement with observations, the model results show an increasing trend in the proportion of dinoflagellates in the Gulf of Finland. Temperature and life cycle-related processes explain the relative increase of dinoflagellates and corresponding decrease of diatoms. Warming over the 30 years has enabled a head start of dinoflagellates by reducing the time lag between germination and growth of vegetative cells. Although diatoms have a much higher growth rate, they cannot compete with the high dinoflagellate concentrations that result from the inoculum. Diatoms will only dominate in years when the inoculum concentrations of dinoflagellates or the temperatures are low. Overall, the model results suggest that consideration of life cycle dynamics of competing phytoplankton groups may be crucial to understand trends and shifts in community composition.

Introduction

In most temperate waters spring blooms are dominated by diatoms (Smayda, 1980, Smayda and Reynolds, 2003). This is not the case in the Baltic Sea, where large (20–30 μm radius/diameter) cold-water dinoflagellates often compete successfully with diatoms (Niemi, 1975, Kononen and Niemi, 1984, Heiskanen, 1993, Heiskanen and Kononen, 1994, Heiskanen, 1998, Kremp and Heiskanen, 1999, Olli and Heiskanen, 1999, Jaanus et al., 2006, Spilling, 2007b). In recent decades, an increasing trend in the spring bloom proportion of cold-water dinoflagellates has been observed in several parts of the Baltic Sea, including the Gulf of Finland (Wasmund and Uhlig, 2003, Alheit et al., 2005, Jaanus et al., 2006, Klais et al., 2011, Wasmund et al., 2011, Klais et al., 2013). The causes behind this trend, however, are unclear.

Generally, dinoflagellates are considered to be less competitive compared to diatoms. Spring bloom dinoflagellates from the Baltic Sea for example have a lower maximum light utilization coefficient (Spilling, 2007b, Spilling and Markager, 2008), a narrower temperature window (Kremp et al., 2005) and a lower maximum growth rate (Spilling, 2007b, Spilling and Markager, 2008) than diatoms. Given these competitive disadvantages, the apparent paradox (low maximum specific growth rate but sudden bloom formation of dinoflagellates) may be related to life cycle aspects (see, e.g., Hense and Burchard, 2010).

Individual life cycle strategies seem to be important in regulating dinoflagellate blooms (Anderson and Rengefors, 2006, Bravo et al., 2010, Kremp, 2013, Wyatt and Zingone, 2014). For example, Biecheleria baltica (sensu Moestrup et al., 2009 = Woloszynskia halophile sensu Kremp et al. (2005)), one of the dominant spring bloom dinoflagellate species in the Gulf of Finland, forms numerous resting cysts at the end of the bloom period (Olli and Trunov, 2010); the accumulation of cysts in the sediment leads to the formation of seed banks. Enhanced encystment and the spreading of cyst beds are assumed to increase the abundance of this species in the water column (Olli and Trunov, 2010, Klais et al., 2011). Kremp et al. (2008) have emphasized that the size of the inoculum regulates bloom formation and the dominance of the cold-water dinoflagellates over diatoms. In their mesocosm experiments, nutrient concentrations and elemental ratios have played a minor role in influencing the competition between dinoflagellates and diatoms.

In general, however, changes in nutrient conditions as well as elemental ratios as a result of eutrophication are often made responsible for ecosystem changes in the Baltic Sea. In particular, high nitrogen and phosphorus loading over decades (see e.g. Elmgren, 2001) but a decrease in the land-sea flux of silica (Humborg et al., 2006) suggest that conditions for diatom growth have deteriorated in the Baltic Sea. However, the silica:nitrogen ratio, specifically in the Gulf of Finland, is quite high (2.0, Papush and Danielsson, 2006) and field studies in the entire Baltic Sea have shown that the spring bloom is limited by nitrogen (Tamminen and Anderson, 2007). Thus, although nutrient concentrations definitely control the total biomass concentrations, it seems unlikely that they can explain the observed long-term trends and neither do the corresponding nutrient ratios.

Other explanations regarding the increasing trend in the spring bloom proportion of dinoflagellates have been put forward, such as decadal climate oscillations (Wasmund et al., 1998, Alheit et al., 2005) or the decrease in ice cover (Klais et al., 2011, Klais et al., 2013). Both seem plausible, as they are consistent with a stronger stratification in winter and spring: since dinoflagellates prefer such stratified conditions, they may have a competitive advantage over diatoms, who do not. The same argument may be used to attribute the trend in species composition to global warming. However, rising temperatures have also more direct effects on growth and life cycle processes of phytoplankton. And some of these effects, e.g., the temperature-dependence of growth, or the life cycle transition regulation and germination control of dinoflagellates (Kremp and Anderson, 2000, Anderson and Rengefors, 2006), may or may not favor dinoflagellate species. In summary, existing observational evidence alone does not easily lead to an explanation of the positive trend in the proportion of dinoflagellates.

Ecosystem models may be a useful tool to address some of the open questions regarding the observed trend. As the inoculum and thus life cycle aspects seem to be important, such an ecosystem model needs to consider life cycle processes of both diatoms and dinoflagellates. So far, there are only few model studies that take these processes into account. Eilertsen and Wyatt (2000) have shown that the consideration of the seed stock better explains spring bloom dynamics in high latitudes. Warns (2013) has coupled a complex dinoflagellate life cycle model (Warns et al., 2013a, Warns et al., 2013b) with a simple diatom life cycle model and found that the alternating patterns between diatoms' and dinoflagellates' dominance can be explained by the abundance of the resting stages and the temperature gradient in spring. Neither of these studies, however, has addressed the observed increasing trend of the spring bloom proportion of dinoflagellates in the Baltic Sea.

By using an ecosystem model that resolves the life cycle of both dinoflagellates and diatoms, we specifically want to understand the underlying mechanisms leading to the increasing trend found in the spring bloom proportion of dinoflagellates in the Gulf of Finland for the period 1981–2010. Our ecosystem model is coupled to a water column model and considers the life cycle processes of three main Baltic Sea phytoplankton groups (diatoms, dinoflagellates and cyanobacteria). We are particularly interested in answering the following questions: (1) is the model able to reproduce the observed trend? If so, (2) what are the underlying mechanisms leading to it?

Section snippets

Observations

We will compare our model results with observations from the Gulf of Finland (longitude 24.8°E, latitude 54.8°N). Specifically, we use the Helsinki Commission (HELCOM) monitoring data for nutrients from the International Council for the Exploration of the Seas (ICES) database (ICES, 2016) and the phytoplankton monitoring datasets which are provided by national monitoring agencies (Klais et al., 2011). Although the phytoplankton data set covers all seasons and includes various taxonomic groups,

Model description

Our model is based on a Baltic Sea ecosystem model, the Ecological ReGional Ocean Model, (ERGOM, Neumann, 2000, Neumann et al., 2002), with an improved description of the phytoplankton groups: we now consider the life cycles of diatoms (Warns, 2013), dinoflagellates (Warns et al., 2013a, Warns et al., 2013b) and cyanobacteria (Hense and Beckmann, 2006, Hense and Burchard, 2010). All other ecosystem variables and processes of ERGOM (related to zooplankton, the three nutrients (nitrate, ammonium

Climatologies

Fig. 2 shows modeled and observed climatological seasonal patterns of sea surface temperature (SST) and surface DIN concentrations for the past three decades. Overall, the climatological seasonal cycles of environmental conditions in the Gulf of Finland are well reproduced. During winter (January–February) the simulated SST is slightly lower, during summer (June–August) somewhat higher than the observed SST (Fig. 2a). The simulated seasonal cycle of DIN concentrations is similar but with a time

Discussion and conclusions

We have investigated the trend in spring bloom composition in the Gulf of Finland, Baltic Sea for the period 1981–2010 by using a coupled water column-ecosystem model which includes the life cycle of three main Baltic Sea phytoplankton groups (diatoms, dinoflagellates and cyanobacteria). In agreement with observations, the model results show an increasing proportion of dinoflagellates in the spring bloom biomass. The altered dominance between diatoms and dinoflagellates is determined by changes

Acknowledgements

We thank two anonymous referees for their valuable comments, which helped to improve the manuscript. We are grateful to Riina Klais for making the observational data of diatoms and dinoflagellates available. We also thank ICES for the HELCOM monitoring data (temperature, salinity and nutrients) and ECMWF for the meteorological forcing data. This research was supported through the Cluster of Excellence “CliSAP” (EXC177), Universität Hamburg, funded through the German Science Foundation (DFG).

References (53)

  • K. Spilling et al.

    Ecophysiological growth characteristics and modeling of the onset of the spring bloom in the Baltic Sea

    J. Mar. Syst.

    (2008)
  • A. Warns et al.

    Encystment of a cold-water dinoflagellate — from in vitro to in sillico

    J. Mar. Syst.

    (2013)
  • N. Wasmund et al.

    Long-term trends in phytoplankton composition in the western and central Baltic Sea

    J. Sea Res.

    (2011)
  • T. Wyatt et al.

    Population dynamics of red tide dinoflagellates

    Deep-Sea Res. II

    (2014)
  • J. Alheit et al.

    Synchronous ecological regime shifts in the central Baltic and the North Sea in the late 1980s

    ICES J. Mar. Sci.

    (2005)
  • D.M. Anderson et al.

    Community assembly and seasonal succession of marine dinoflagellates in a temperate estuary: the importance of life cycle events

    Limnol. Oceanogr.

    (2006)
  • D.P. Dee et al.

    The ERA-Interim reanalysis: configuration and performance of the data assimilation system

    Q. J. R. Meteorol. Soc.

    (2011)
  • H. Eilertsen et al.

    Phytoplankton models and life history strategies

    Afr. J. Mar. Sci.

    (2000)
  • R. Elmgren

    Understanding human impact on the Baltic ecosystem: changing views in recent decades

    Ambio

    (2001)
  • A. Heiskanen

    Mass encystment and sinking of dinoflagellates during a spring bloom

    Mar. Biol.

    (1993)
  • A.S. Heiskanen

    Factors governing sedimentation and pelagic nutrient cycles in the northern Baltic Sea

    Monogr. Boreal Environ. Res.

    (1998)
  • A.S. Heiskanen et al.

    Sedimentation of vernal and late summer phytoplankton communities in the coastal Baltic Sea

    Arch. Hydrobiol.

    (1994)
  • I. Hense

    Regulative feedback mechanisms in cyanobacteria-driven systems: a model study

    Mar. Ecol. Prog. Ser.

    (2007)
  • J. Hinners et al.

    Evolution in temperature-dependent phytoplankton traits revealed from a sediment archive: do reaction norms tell the whole story?

    Proc. R. Soc. B

    (2017)
  • C. Humborg et al.

    Decreased silica land-sea fluxes through damming in the Baltic Sea catchment — significance of particle trapping and hydrological alterations

    Biogeochemistry

    (2006)
  • ICES

    Baltic Sea Monitoring data

    (2016)
  • Cited by (9)

    • Modeling the life cycle of four types of phytoplankton and their bloom mechanisms in a benthic-pelagic coupled ecosystem

      2022, Ecological Modelling
      Citation Excerpt :

      Warns (2013) coupled a complex dinoflagellate life cycle model (Warns et al., 2013a, 2013b) with a simple diatom life cycle model. Lee et al. (2018) developed an ecosystem model which includes the life cycles of three main phytoplankton groups (diatoms, dinoflagellates, and cyanobacteria) and showed the importance of phytoplankton life cycle mechanisms regulating their blooms. However, these models did not consider the early diagenetic vertical biogeochemical mechanisms in the benthic system which controls important nutrient supply to the pelagic systems in shallow coastal areas (Graf, 1992; Grenz et al., 2000; Liu et al., 2007; Griffiths et al., 2017).

    • Seed banks can help to maintain the diversity of interacting phytoplankton species

      2022, Journal of Theoretical Biology
      Citation Excerpt :

      A variety of models have endeavoured to explain and predict amplitude, timing and/or spatial distribution of blooms by explicitly modeling multiple stages in the life cycle of a particular species, but without interactions with other organisms (see for example McGillicuddy et al., 2005; Hense and Beckmann, 2006; Hellweger et al., 2008; Yñiguez et al., 2012). Two-to-four species (Yamamoto et al., 2002; Estrada et al., 2010; Lee et al., 2018) models also exist, but include a single species or compartement having a resting stage. This state of affairs means that we currently have no clear understanding of how the resting stage might help maintaining biodiversity in species-rich communities.

    • Recent environmental change in the Kosterhavet National Park marine protected area as reflected by hydrography and sediment proxy data

      2021, Marine Environmental Research
      Citation Excerpt :

      The acceleration in the B. baltica contribution to the spring bloom in the Baltic Sea proper during the past decades is the continuation of a process that started in the first half of the 20th century, when the species is assumed to have responded to the increasing anthropogenic eutrophication of the surface waters (Kremp et al., 2018; Sildever et al., 2019). However, ecosystem modelling suggests that rather than nutrient input, the superiority of B. baltica over diatoms is likely caused by climate change interfering with its life-cycle processes, i.e. an earlier temperature increase in spring over the past few decades has decreased the timespan between cyst germination and dinoflagellate blooms (Lee et al., 2018). The hydrographic data from northern Koster Island do not reveal a clear seasonal temperature increase mirroring the trend of cysts of B. baltica, but considering the species’ preference for lowered salinities, the increased abundance of B. baltica cysts in the Koster Trench sediments from ~2004 onwards may thus at least partly be caused by the changed seasonal runoff pattern in the area with higher summer precipitation and lower surface water salinities (Figs. 7 and 8) – the exact same changes that likely contributed to the decline of the P. dalei cyst bed.

    • Future HAB science: Directions and challenges in a changing climate

      2020, Harmful Algae
      Citation Excerpt :

      Herein lies a problem in HAB science; there are only scant data relating HAB species existence to other organisms they encounter in nature—data that are essential to formulate dynamic models. There also are important life history and behavioral traits (notably cyst formation and mixotrophy) that are typically absent from model descriptions (Azanza et al., 2018; Lee et al., 2018). A key priority in HAB modeling will be to identify the biggest model uncertainties, a non-trivial task not least because only correctly structured models offer a route to locating factors that exert greatest leverage on behavior.

    • Modelling phytoplankton adaptation to global warming based on resurrection experiments

      2019, Ecological Modelling
      Citation Excerpt :

      The first factor we consider for this purpose is cyst mortality. Cysts are known to play an important role in bloom formation (Klais et al., 2013; Lee et al., 2018) and the survival potential of resting stages can be affected under elevated temperatures (Ellegaard and Ribeiro, 2017). We therefore study the sensitivity of the adaptive response under global warming and a concurrent increase in cyst mortality.

    View all citing articles on Scopus
    View full text