2.2 The importance of sea ice for biogeochemical cycles

Sea ice isn’t static. It offers a seasonal to multiyear reservoir for particles and dissolved substances and enables various ecosystem processes that circulate nutrients and organic substances and shape the polar marine biogeochemistry. As a natural barrier between the ocean and atmosphere, the sea ice has a major influence on biogeochemical processes. It consists of a dynamic matrix of pure ice, brine-filled interstices, and gas bubbles. Generally speaking, the initial concentrations of dissolved substances in the sea ice are determined by the composition of the seawater from which it was formed. During its formation, biogeochemical material (e.g. macronutrients, iron, organic substances, sediments) becomes trapped and is transformed in the sea ice, only to be released into the seawater when the ice melts. When melting occurs, dissolved and solid nutrients, along with organic material, are released into the water column, where they can influence pelagic processes or contribute to particle export. Sea-ice biogeochemistry is of critical importance for the polar sea-ice system  (Meiners and Michel, 2017). Material flows at the interfaces between the sea ice and ocean, and between the atmosphere, snow and ice, also influence the concentration of dissolved substances. In addition, organic substances accumulate in the sea ice due to autotrophic and heterotrophic production. Because of sea-ice drift, induced by winds and ocean currents, the biogeochemical material spreads laterally into other sea-ice regions (Vancoppenolle et al., 2013). Active biogeochemical processes in the sea ice involve macronutrients, trace elements, organic carbon, inorganic carbon, other climate-relevant gases (DMS, methane, laughing gas), the atmospheric halogen chemistry and its interactions with oceanic and atmospheric processes, and the precipitation of minerals  like plaster of Paris (calcium sulphate) or ikaite (a type of calcium carbonate that contains water) (Dieckmann et al., 2008 ; Wollenburg et al., 2018).


Some substances are chemically altered in the near-surface water itself or are absorbed by organisms; others make their way to the deeper water column or even the seafloor. When nutrients and other materials are released, the formation of algae blooms is initiated. This process is particularly pronounced in the Southern Ocean, because the water there for the most part contains very little iron – an important nutrient for plants. If, due to melting, large quantities of iron are transferred from the ice to the ocean, the level of algal growth, and therefore primary production, skyrockets. This can especially be seen at the sea-ice edge, where the light intensity is highest (Vancoppenolle et al., 2013).

Macronutrients

In the marine sciences, the term “macronutrients” generally refers to those inorganic chemical compounds that algae absorb in large quantities, and which they need in order to produce biomass. These nutrients include dissolved nitrate (NO3-), nitrite (NO2-), ammonium (NH4+), phosphate (PO43-) and silicic acid (Si(OH)4). Macronutrients, trace elements and other materials aren’t homogeneously distributed throughout the ice: there are areas with higher and lower nutrient concentrations. Analyses of vertical profiles for the most important biogeochemical tracers in sea ice reveal a variety of progressions. In the absence of biological activity or remineralisation, the inorganic macronutrients generally follow the salinity. As a result, in cold ice the concentrations in the brine fraction can be very high (Arrigo, 2014). All macronutrients are dissolved in the brine. Significant discrepancies between nutrient concentrations and salinity are connected with biological activity. Accordingly, nutrient uptake on the part of microalgae and the breaking down of organic substances on the part of bacteria can substantially change the relation between nutrients and salinity in the brine channels. Nutrients can also find their way into the sea ice after it is formed: the amount of brine that drains out of the ice via the brine channels during the desalination process is replaced by a roughly equivalent volume of seawater, full of new nutrients (Arrigo, 2014).

Classically, five processes are considered to contribute to the release of salt and dissolved substances from sea ice cover: the initial solute rejection at the ice-seawater freezing interface; diffusion of solutes; brine expulsion; gravity drainage; and flushing (Meiners and Michel, 2017). The last two of these processes have been identified as the driving forces behind salt loss and are therefore the most important physical processes for regulating macronutrients in the sea ice. In addition, macronutrients and various trace metals like iron are transported to the ice from the atmosphere. Tiny amounts of these substances cling to the surface of aerosols, which are reaching the polar regions by atmospheric transport. The particles are either deposited directly on the ice or through snow deposition. Particles contain relatively small amounts of macronutrients, but relatively large amounts of trace metals like iron (Arrigo, 2014). When the ice melts, these nutrients are made available to the sea-ice community or the pelagic ecosystem.

Iron

In regions in which iron (Fe) potentially limits phytoplankton growth (the Southern Ocean and North Pacific), evidence of Fe concentrations in the sea ice that are much higher than those in the water column has led to investigations of the role of seasonal sea-ice retreat as a potential trigger for phytoplankton blooms in Fe-limited surface waters. Iron is important for photosynthesis and nutrient assimilation processes. Low amounts of available Fe for phytoplankton reduce their growth rate and the frequency of large classes of phytoplankton, since smaller phytoplankton have a higher surface-to-volume ratio and can assimilate iron more efficiently than large phytoplankton (Vancoppenolle et al., 2013). However, in regions in which there is ample Fe (e.g. in the Arctic Ocean due to its proximity to the continents), Fe released by the sea ice is less relevant. The contributions of the mechanisms that potentially drive the Fe cycle in the sea ice are not yet fully understood (Vancoppenolle et al., 2013). Within the Antarctic sea ice, dissolved iron can accumulate up to a factor of three compared to the iron-deplete seawater, making the sea ice an important iron reservoir (Meiners and Michel, 2017) . In some cases, Fe released by melting sea ice can contribute to substantial blooms at the edges of ice floes. Yet it must be kept in mind that, after being released into the water column, Fe may not always be readily available to phytoplankton, since some forms can’t be used for assimilation in the plankton. The so-called bioavailability of Fe, which is still difficult to predict, is a function of the physical and chemical forms of Fe, and of the respective strategies used by various biota to gain access to Fe (Vancoppenolle et al., 2013). Compared to marine iron sources, atmospheric sources (I.e. snow deposition) appear to be less relevant  (Meiners and Michel, 2017). 

Trace gases in the ice

In addition to the nutrients, trace gases – which are produced by microorganisms in the ocean – find their way into the ice. The most frequently occurring trace gas is dimethyl sulphide (DMS). It is produced when bacteria break down the compound dimethylsulfoniopropionate (DMSP), which itself is produced by ice algae. Plants use DMSP e.g. as a form of natural antifreeze, which is why ice algae especially produce it at very low temperatures and high salinities. The sea-ice concentrations of DMS and DMSP are not only extremely high, but also extremely variable, as high-resolution vertical profiles and major regional and seasonal variations show (Vancoppenolle et al., 2013) . This variability is due to complex interactions between the physical, biological and chemical processes that drive the DMS(P) cycle on the one hand, and thermodynamic sea-ice processes on the other. Many of these processes are difficult to monitor in field studies and remain poorly quantified and understood (Vancoppenolle et al., 2013). The melting of sea ice significantly increases the concentration of DMS in the surface water of the polar ocean  – either directly, through the release of DMS(P), or through the formation of phytoplankton blooms. These DMS(P) impulses can dramatically increase regional oceanic DMS emissions.

If there were no sea ice, there would be a free exchange of gases between the ocean and atmos-phere in the polar regions. Worldwide, the ocean absorbs roughly one third of carbon dioxide (CO2) from the atmosphere. Yet even when the ocean is covered by ice, there are always certain open areas where gas exchange can (at least locally) take place. These can include polynyas, leads or wa-terways near coasts. For many years it was assumed that wide-scale CO2 exchange between the ocean and atmosphere only took place during the open-water season.
In seawater, carbon dioxide can react with various other carbonaceous substances , including bicarbonate (HCO3-) and carbonate ions (CO32-). The  CO2 in the ocean exists in three different kinds:

  1. dissolved inorganic carbon (DIC)
  2. dissolved organic carbon (DOC)
  3. particulate organic carbon (POC)

In marine systems DOC originates from either autochthonous or allochthonous sources. Autoch-thonous DOC is produced within the system, primarily by plankton organisms and in coastal waters additionally by benthic microalgae, benthic fluxes, and macrophytes, whereas allochthonous DOC is mainly of terrestrial origin, supplemented by groundwater and atmospheric inputs (Lønborg et al., 2020).

Like many other substances, DIC becomes trapped in the brine channels during the ice formation phase. A large part of the DIC is lost via brine rejection and possibly through outgassing to the at-mosphere. It’s also possible for CO2 to escape the brine and subsequently become part of the ice’s structure as CO2-rich gas bubbles (Vancoppenolle et al. 2013).  But various gas-release process-es can also allow CO2 to be directly released from the ice into the atmosphere.

In addition to dissolved inorganic carbon (DIC), seawater contains dissolved organic carbon (DOC), which can include carbohydrates, proteins, amino acids or complex substances like humic substanc-es and a host of other molecules. DOC is defined as organic substances smaller than 0.2 µm. DOC accumulates in the brine, where the concentration is up to three orders of magnitude higher than in the seawater below (Vancoppenolle et al. 2013). The DOC in the Arctic chiefly comes from rivers, whereas in the Southern Ocean, it is primarily autochthonous (Lønborg et al., 2020).

__________________________

Arrigo, K.R. (2014): Sea ice ecosystems, Annual Review of Marine Science, vol. 6, pp. 439-467.
Lønborg C., C. Carreira, T. Jickells & X. A. Álvarez-Salgado (2020): Impacts of Global Change on Ocean Dissolved Organic Carbon (DOC) Cycling, Frontiers in Marine Science, Volume 7, p. 466 - ;
https://www.frontiersin.org/article/10.3389/fmars.2020.00466; DOI=10.3389/fmars.2020.00466.
Dieckmann, G. S., G. Nehrke, S. Papadimitriou, J. Göttlicher, R. Steininger, H. Kennedy, D. Wolf-Gladrow & D. N. Thomas (2008): Calcium carbonate as ikaite crystals in Antarctic sea ice, Geophys. Res. Lett., 35, L08501, doi:10.1029/2008GL033540.
Meiners K. M. & C. Michel (2017): Dynamics of nutrients, dissolved organic matter and exopolymers in sea ice. In: D. N. Thomas (ed.), Sea Ice, 3rd edition, Wiley-Blackwell, Chichester (UK) Hoboken (NJ), pp. 415-432.
Vancoppenolle, M., K. M. Meiners, C. Michel, L. Bopp, F. Brabant, G. Carnat, B. Delille, D. Lannuzel, G. Madec & S. Moreau (2013): Role of sea ice in global biogeochemical cycles: emerging views and challenges. Quaternary science reviews 79, pp. 207-230.
J. E. Wollenburg, C. Katlein, G. Nehrke, E.-M. Nöthig, J. Matthiessen, D. A. Wolf-Gladrow, A. Nikolopoulos, F. Gázquez-Sanchez, L. Rossmann, P. Assmy, M. Babin, F. Bruyant, M. Beaulieu, C. Dybwad & I. Peeken (2018): Ballasting by cryogenic gypsum enhances carbon export in a Phaeocystis under-ice bloom. Scientific Reports (2018), DOI: 10.1038/s41598-018-26016-0