1. Introduction
Let us begin our brief consideration of the ocean carbon cycle with a look at the atmosphere. The heat budget of the atmosphere and the Earth's surface is the result of incoming short wave solar radiation (and a bit of geothermal energy from the Earth's interior), reflection of short wave radiation back to space and outgoing long wave terrestrial radiation. The atmospheric composition is crucial for determining the surface temperature of the Earth. So-called greenhouse gases keep the lower part of the atmosphere – and hence also the Earth's surface, be it land or water – above the value for a hypothetical planet without an atmosphere. Without the pre-industrial greenhouse effect, the Earth's surface temperature would amount to 255 K instead of the observed 288 KReference Ramanthan, Callis, Cess, Hansen, Isaksen, Kuhn, Lacis, Luther, Mahlmann, Reck and Schlesinger1 and all water would be in the form of ice. The most important natural greenhouse gases are water vapour (H2O) and carbon dioxide (CO2). Both gases are minor constituents of the atmosphere, which mainly consists of oxygen, nitrogen, and some argon. Because of their molecular structure, the greenhouse gases get into resonance with the long-wave outgoing radiation from the Earth and create absorption bands for long-wave radiation at quite distinct bands and thus lead to a temperature increase. In the early stage of the climate system, the original pre-Cambrian atmosphere was mainly a CO2 atmosphere. The evolution of biota and photosynthesis gave rise later on to significant oxygen levels in the Earth's atmosphere, around 700 million years before present.Reference Schidlowski, Appel, Eichmann and Junge2 The current atmospheric composition consists of around 21% oxygen, and 78% nitrogen, while the key greenhouse gases H2O and CO2 provide only tiny amounts. The pre-industrial CO2 concentration amounted to 280 ppmv (parts per million of volume). Owing to human-derived CO2 emissions, the modern atmosphere is now close to 400 ppmv on the global average. For the regulation of the atmospheric CO2 concentration, physical, chemical, and biological processes act together, and over long periods of time (longer than several 1000 years) geological processes play a major role. Owing to its small atmospheric inventory, atmospheric CO2 can be altered by natural processes quite significantly. The largest atmospheric carbon perturbations over the past million years before the industrial revolution were the glacial-interglacial variations. These lowered the average interglacial (and pre-industrial) CO2 concentration of 280 ppmv repeatedly by 100 ppmv down to 180 ppmv. Since the beginning of the AnthropoceneReference Crutzen3 – which can be placed at year 1784, when the steam engine was produced – human activities (fossil fuel burning, land use change, and cement manufacturing) have increased the atmospheric CO2 concentrations by more than 100 ppmv within a very short time interval of somewhat more than 200 years. In May 2013, atmospheric CO2 concentrations in remote (‘clean’) tropospheric air (the troposphere is the lower part of the atmosphere, the lowermost 8–16 km depending on whether one is at polar or equatorial latitudes) approache 400 ppmv.
The oceans cover 71% of the Earth's surface and, seen from space, the Earth is predominantly a blue-coloured aqueous planet. What role does the ocean carbon cycle play in climate change? The ocean's role in the context of climate change is twofold: (1) through carbon dioxide gas exchange with the atmosphere and carbon cycle climate feedbacks, the ocean regulates the carbon dioxide concentration in the atmosphere and hence has a strong influence on the heat budget of the Earth; (2) the paleo-climatic marine sediment core record is largely based on biogenic matter fluxes from the surface ocean to the sea floor, and these are part of the marine carbon cycle. This sedimentary archive can be employed to understand the processes contributing to climate change and to reconstructing past climates, such as the climate during the ice ages. Let us now investigate why the ocean is important for carbon cycling in the Earth system.
2. Why is the Ocean Important for the Global Carbon Cycle?
The ocean is important for the global carbon cycle primarily due to three factors: (1) the ocean is a huge carbon reservoir with a relative short turnover time; (2) carbon dioxide in sea water is effectively dissociated inorganically into other substances; (3) marine plankton keeps the surface ocean carbon dioxide concentration at a lower level than would a lifeless ocean.
2.1. The Ocean as a Big Carbon Reservoir
For the pre-industrial situation, the inorganic oceanic carbon inventory was 38,000 GtC (the amounts of organically stored carbon in the ocean is significantly smaller), while those of the atmosphere and the terrestrial surface reservoir (living plants and dead soil organic matter; without rock and fossil fuel resources) amounted to only 550 GtC and 2000 GtC. 1 GtC is 109 tons of the atom carbon and is equal to 1 PgC (1015 grams carbon; please, note that the O2 in CO2 is not counted for the weight here, but only the C). From these numbers one can immediately deduce that a small change in the ocean carbon inventory could potentially have a strong effect on the atmospheric CO2 concentration. Carbon in the ocean is transported with the ocean currents. The global ocean circulation can be visualised conceptually (in the statistical mean sense) as big conveyor-belt-like water movement.Reference Broecker and Peng4 Dense water in the northern North Atlantic sinks to the bottom, flows along the west side of the continents, upwells to the surface in the Antarctic Circumpolar Current where it is re-cooled locally and then spreads as intermediate and deep waters into the Pacific and Indian Oceans. In the northern parts of the Pacific and Indian Oceans, ocean water upwells in a large-scale movement slowly to the surface, from where it returns through different pathways (around the southern tips of Africa or South America) back to the deep-water source regions of the North Atlantic. One average journey along this conveyor belt circulation takes ca. 1500 years. The turnover time of water in the upper few hundred metres is quicker, of course, but depends on the location in the ocean and the respective regional circulation conditions. Oceanic currents and mixing processes as well as water mass formation (production of water bodies of different characteristic temperature, salinity, and trace element composition) are driven by the wind field and processes causing density variations (heating/cooling, precipitation/evaporation, sea ice melting/freezing). From these considerations, one can already see that the ocean is a long-term factor in global carbon cycling.
2.2. The Inorganic Carbon Cycle – CO2 Reacts with Seawater
The reason for the huge oceanic carbon reservoir is the chemical reaction of CO2 with seawater. In the atmosphere, CO2 is close to inert, but not so in the ocean. The reason is the chemical composition of seawater due to its salinity and its trace element composition. CO2, entering the oceans from land (through river run-off) and the atmosphere (through gas exchange) becomes dissolved in seawater and is then quickly dissociated into bicarbonate ions HCO3– and carbonate ions CO32– (see, for example, details in Ref. 5). CO2 is also hydrated by CO2 to carbonic acid (H2CO3), but this is difficult to distinguish in practice from CO2 (the only difference being a water molecule H2O added). The sum of gaseous dissolved CO2 (plus carbonic acid), bicarbonate, and carbonate is called the total dissolved inorganic carbon (DIC). Average seawater contains a ratio of CO2 (plus carbonic acid): HCO3–: CO32– of roughly 1:100:10 and is slightly alkaline with a pH value of 8 (pH denotes a measure for the acidity of water, pH =−log10[H+], where pure fresh water has a neutral pH value of 7). The carbonate buffer system of the ocean allows large additions of CO2 to the ocean. Most of the CO2 reacts with water and carbonate ions to form bicarbonate. The net results of CO2 additions to seawater are decreasing concentrations of CO32− and increasing H+ ion concentrations (a process called ocean acidification). With increasing DIC content and increasing CO2 concentration or CO2 partial pressure in the surface ocean and the overlying atmosphere in which the ocean tries to achieve equilibrium, the ability of seawater to dissociate the CO2 decreases somewhat. Low pH and high CO2 water contains not as many CO32− ions anymore for neutralising the incoming CO2.
2.3. The Organic Carbon Cycle and the Role of Organisms
What role do organisms play within the ocean carbon cycle? The living biomass in the ocean is very small as compared to land. Most of the oceanic production of biogenic matter is due to plankton organisms growing in the ocean surface layer, provided a number of essential nutrients are available. Through the action of organisms, particulate organic carbon (POC) is produced, as well as dissolved inorganic carbon (DOC). After the death of the organisms originally produced, their remainders sink vertically through the water column as POC flux. DOC is also released by living organisms during their metabolism. Small organic particles (too small to sink any more) are also counted as DOC. DOC is mixed in the water like DIC. Most POC is degraded by microorganisms within the upper 1000 m and added to the local DIC. Only fairly small amounts reach deep sea sediments (whereas shelf sediments can be rich in organic material). A considerable amount of DIC is brought into the oceans through river run-off and a large percentage is degrading, but only slowly. The effect of biota on regulating the atmospheric CO2 concentration is primarily due to primary producers (plant plankton carrying out photosynthesis) and the amount of POC, which leaves the surface layer as so-called export production and thus is excluded from the seasonal cycling of the surface ocean. The export flux can be modified by zooplankton (animal plankton) which consumes the phytoplankton (plant plankton), e.g. through the size of particle aggregates formed. But ultimately the phytoplankton production determines how much DIC can be removed from the ocean surface layer through the action of biota. In an unperturbed steady state ocean, the export flux is balanced by vertical upwelling of water, which carries high amounts of carbon and nutrients again to the surface as so-called new production. These newly incoming carbon and nutrient compounds to the surface zone, which is full of light and enabling photosynthesis, are then again consumed by organisms. The net result is a vertical gradient of nutrients and DIC in the ocean, with low concentrations in the surface (except for strong upwelling areas) and increasing concentrations with depth. Marine phytoplankton thus keeps the surface DIC and CO2 partial pressure constantly about 200–300 ppmv lower than it would be in a lifeless ocean. As long as ocean circulation, nutrient availability, and other ecosystem drivers remain constant, marine biota would not contribute to oceanic uptake of human-produced CO2. If one of these controlling factors would change, biota can, however, have a significant impact on atmospheric CO2. The situation is further complicated by the fact that some plankton produced hard shell material is made of calcium carbonate (CaCO3). This particulate material – although, in the ocean, being almost exclusively produced by organisms – is called particulate inorganic carbon (PIC). For producing CaCO3, CO32– ions are removed from seawater. Therefore, although PIC contains carbon, its production increases the CO2 partial pressure in water. PIC contributes to quite effectively sinking particle aggregates. Its dissolution is determined first of all by pressure. Very deep areas of the ocean floor are, therefore, free of CaCO3 (PIC), while many mid-ocean ridges and other topographic highs are covered by CaCO3, like mountains by snow. The sedimentary CaCO3 provides an important reservoir of CO32− ions for neutralising human-produced carbon in the oceans on long time scales (more than 10,000 years). Many organisms produce siliceous hard parts instead of CaCO3 shells (in fact, if enough dissolved silica is available, organisms preferentially produce siliceous shells because of energetic reasons). As the availability of dissolved silica determines the specification of hard part production, also the silicon cycle has an influence on the carbon cycle in the ocean.
3. Natural Climate Variations and the Ocean Carbon Cycle – the Sedimentary Record
For natural carbon dioxide variations, such as during the glacial–interglacial cycles, biological carbon cycling presumably played a key role. Where do we have this knowledge from? As described in section 2.3, marine biota produce a flux of particles through the ocean water column. CaCO3, siliceous hard parts, and – to a lesser degree organic particulate material – which have accumulated in the sediment after deposition onto the sea floor through this particle flux (and hence luckily survived potential re-dissolution in the upper few centimetres of the sediment column) provide a time series of often varying composition and surface production. Marine geologists have developed technologies to drill into the ocean floor and recover sediment cores, which are typically several metres long. These cores can be analysed in the laboratory and through reconstructing the time history of accumulation, they provide important information about the functioning of the marine carbon cycle and other oceanic processes in the past. In addition, for correctly interpreting this sedimentary paleo-climate archive, a proper understanding of marine biological carbon cycling and related biogeochemical processes is vital. For this reason alone, ocean carbon cycle research is of key climatic importance. In addition to the varying composition and accumulation strength of marine sediment, stable carbon and oxygen isotopes stored in the sedimented CaCO3 provide the backbone for reconstructions of past oceans and their physical-biogeochemical functioning. In particular, varying contents of (13C/12C) and (18O/16O) in sedimented CaCO3 material indicate variations in nutrient concentration, oceanic overturning strength, vertical particle flux as well as carbon gradients (13C/12C) and, respectively, changes in temperature, sea level/ice volume, and salinity (18O/16O). Preferentially, CaCO3 material from zooplankton (foraminifera) is used for these analyses, as different species live at different depth ranges (even at the ocean floor) and thus enable a three-dimensional view on past oceans.
Marine biological processes in the ocean together with physical processes have resulted in a remarkable variation of atmospheric CO2 concentration through several glacial–interglacial cycles in the past million years.Reference Petit, Jouzel, Raynaud, Barkov, Barnola, Basile, Bender, Chappellaz, Davisk, Delaygue, Delmotte, Kotlyakov, Legrand, Lipenkov, Lorius, Pépin, Ritz, Saltzman and Stievenard6, Reference Siegenthaler, Stocker, Monnin, Lüthi, Schwander, Stauffer, Raynaud, Barnola, Fischer, Masson-Delmotte and Jouzel7 The CO2 variations as deduced from high accumulation Antarctic ice core analysis between 280 ppmv (at interglacials) and 180 ppmv (at glacial maxima) are highly correlated with surface air temperature variations. The atmospheric CO2 draw down at glacials is likely to be caused primarily by the ocean carbon cycle because the standing stock of living terrestrial biomass was probably smaller in a cold dry glacial climate.Reference Crowley8 The fact that bacterial decomposition of soil organic matter may also have slowed down in a cold climate provides a wild card in glacial carbon cycle research that has not yet been fully clarified.Reference Zeng9 Oceanic processes have probably changed, simultaneously involving the cycling of POC as well as PIC in order to achieve the drawdown,Reference Heinze and Hasselmann10 with the Southern Ocean being of key importance.Reference Sigman and Boyle11 The glacial–interglacial cycles are thought to be ultimately triggered by variations of the Earth's orbital parameters with respect to the sun and related variations in the solar incoming radiative forcing of the climate system. The ocean carbon cycle served then as an amplifier of the physically-induced climate change, but recent results have shown that the CO2 increase after the last glacial maximum at 20,000 years before present may have preceeded the warming after an initial southern hemisphere triggering.Reference Shakun, Clark, Feng, Marcott, Mix, Zhengyu, Otto-Bliesner, Schmittner and Bard12
4. Oceanic Uptake of Human-produced Carbon Dioxide
For the uptake of human-produced carbon dioxide, the inorganic uptake by the ocean (dissolution, dissociation, transport, and mixing) plays the most prominent role. It would practically play the only role if there were no changes in circulation, nutrient cycling, and ecosystem changes due to physical (change in temperature and salinity) and chemical (change in seawater pH and carbonate ion concentration due to high CO2) drivers. Circulation, nutrient cycling, and ecosystem changes can modulate the transfer of carbon between the atmosphere and the ocean – and are involved in feedback processes – but sensu strictu do not govern the transfer of human-produced excess carbon (that CO2 which has been emitted by human activities to the atmosphere and that leads to an atmospheric CO2 concentration well above the pre-industrial value).
The human-derived excess carbon uptake in the ocean works through the partial pressure difference between the atmosphere and the ocean: if the partial pressure in the atmosphere increases due to anthropogenic CO2 emissions, an imbalance between the oceanic and atmospheric CO2 partial pressures builds up and anthropogenic CO2 is pumped into the ocean through gas exchange attempting to remove this imbalance. In seawater, the bulk of the CO2 that has entered through gas exchange reacts with CO32− ions and water and is mainly converted into HCO3− ions. The efficiency of this mechanism depends on the carbon already dissolved inorganically in the seawater. The higher the dissolved inorganic carbon concentration and the seawater CO2 partial pressure, the lower the fractional uptake of further human-derived emissions to the atmosphere. It is thus expected that the surface ocean during the coming decades is expected to become less efficient in balancing further CO2 emissions to the atmosphere so that a larger fraction of emitted CO2 per year will remain in the atmosphere.
There are many factors that can modulate the air–sea transfer of anthropogenic excess carbon. Changes in ocean circulation as a consequence of climate change are expected to change the density structure of the ocean and hence ocean currents as well as mixing. These changes also induce changes in the cycling of nutrients, carbon, and oxygen, which themselves can lead to changes in oceanic organic export production. Ocean acidification (low pH, low carbonate saturation) may have a potentially damaging effect on ecosystems with changes in the export production of organic carbon as well as inorganic CaCO3 shell material.Reference Riebesell, Schulz, Bellerby, Botros, Fritsche, Meyerhöfer, Neill, Nondal, Oschlies, Wohlers and Zöllner13 The CaCO3 feedback is expected to be fairly small.Reference Heinze14 The alteration of the organic carbon cycling through low pH could potentially be larger.Reference Riebesell, Schulz, Bellerby, Botros, Fritsche, Meyerhöfer, Neill, Nondal, Oschlies, Wohlers and Zöllner13 On very long timescales, however, dissolution of CaCO3 from the sea floor (through repeated contact of waters with high CO2 load and biogenic CaCO3 in the surface sediment) plays a significant role in reducing the CO2 partial pressure in the oceans.Reference Archer15
On intermediate to long time scales, the ocean provides the most important sink for anthropogenic carbon dioxide. The marine uptake kinetics for carbon dioxide work on a longer time scale than current and projected emissions by humans. The reason is the long mixing time scale of the oceans for transporting human-produced excess carbon to deeper layers. Because of this bottleneck effect, saving fossil fuel emissions is of key importance for enabling efficient carbon dioxide buffering by the oceans, as demonstrated by global ocean general circulation models.Reference Maier-Reimer and Hasselmann16
5. Research Challenges and Concluding Remarks
For a better quantification of the air–sea carbon transfer, a number of key challenges exist. First of all, we still have only a comparatively small number of oceanic carbon observations, which in most regions of the ocean do not cover a full seasonal cycle. Satellite observations are useful for quantifying a number of surface variables, but do not allow a three-dimensional view into the water column. The anthropogenic carbon uptake by the oceans is, moreover, realised through a small net difference of large ingoing and large outgoing carbon fluxes. These are difficult to derive from (spatially and temporally sparse) observations. An extension of the measurement network also through the use of automated systems (commercial ships, buoys, and floats) is necessary. Ocean models need to be based on improved process descriptions which have to be derived from observational evidence. Oceanographers are starting to improve the models through systematically correcting them towards the observations by using a methodology called data assimilation. But due to the complexity of the oceanic inorganic as well as organic biogeochemical cycles this is not an easy task, as some processes are not yet well enough understood (and probably also to some degree not yet identified). International networks of researchers in big project consortia have considerably helped to improve our knowledge about ocean carbon cycling in a situation of imminent stronger climate change and ocean acidification. Currently it is not clarified how variable the oceanic sink is regionally over which time intervals.Reference Watson, Schuster, Bakker, Bates, Corbière, González-Dávila, Friedrich, Hauck, Heinze, Johannessen, Körtzinger, Metzl, Olafsson, Olsen, Oschlies, Padin, Pfeil, Santana-Casiano, Steinhoff, Telszewski, Rios, Wallace and Wanninkhof17, Reference McKinley, Fay, Takahashi and Metzl18 Certainly, next to some quite well established facts, ocean carbon cycling within climate research may still hold some surprises and makes such research an exciting undertaking.
Acknowledgments
The author would like to thank the organisers of the Aacademia Europaea's 24th Annual Conference (Northern seas: the European dimension, held from 11 to 13 September 2012, in Bergen, Norway) for the opportunity to present an overview on this topic during a plenary presentation. This work was supported through various sources, but in particular through the EU FP6 Integrated Project CARBOOCEAN (grant agreement no. 511176 GOCE) and EU FP7 large-scale integrating project CARBOCHANGE (grant agreement no. 264879). This article is a contribution to the Berknes Centre for Climate Research and its Centre for Climate Dynamics.
Christoph Heinze was educated as a physical oceanographer who switched to biogeochemical ocean modelling for his PhD studies. He is currently professor in chemical oceanography at the Geophysical Institute of the University of Bergen and leads the research group on biogeochemical cycles at the Bjerknes Centre for Climate Research. His work focuses on simulations of the ocean carbon cycle and other elemental cycles using predictive global biogeochemical ocean models and Earth system models. He coordinates the European FP7 large-scale integrated project CARBOCHANGE, and has coordinated the FP6 Integrated Project CARBOOCEAN, both dedicated to the oceanic uptake of carbon dioxide from the atmosphere and involving a large number of scientists from Europe, Africa, and North America.
