Feedback from the biological organic carbon pump

The marine biosphere is dominated by phytoplankton organisms, which synthesize organic matter from dissolved nutrients such as_CO2, nitrate and phosphate ions. The average stoichiometry of this photosynthesis reaction is expressed by the ratios introduced by Alfred Redfield in the late 50s. If the _pCO2 of surface water is in equilibrium with that of the overlying atmosphere, the_CO2 deficit linked to organic synthesis induces a diffusion of_CO2 from the atmosphere to the ocean. This is commonly referred to as the organic biological pump.

By consuming dissolved_CO2 and ^H3O+ ions, the organic synthesis reaction tends to raise the pH of seawater. The pH mapping of surface waters shows high values (8.2-8.4) for coastal zones and high latitudes, while the subtropical " deserts " are characterized by low pH values around 8.

After synthesis, organic matter is rapidly recycled in the photic zone. Nevertheless, a fraction of this organic matter (≈ 15 % in the open ocean) escapes at depth via sedimentation in the water column. It is then gradually degraded and remineralized by bacteria living at depth. Vertical pH profiles show a minimum at around 1 to 2 km, corresponding to the maximum of this remineralization. Deep-water masses are transported by the global thermohaline circulation, maintaining a _TCO2 and pH gradient between the North Atlantic and the North Pacific (around 0.2 pH units). The acidic, _TCO2-laden deep waters then rise to the surface through variousupwelling mechanisms.

Over the last few millennia, the ocean had reached a steady state with an atmospheric partial pressure of around 280 ppm. This equilibrium was maintained by compensation between primary organic production through photosynthesis and remineralization of _TCO2.

Photosynthesis depends on numerous parameters such as light intensity,_CO2 content and temperature. Recent changes in atmospheric _pCO2 and oceanic temperature are therefore likely to disrupt the marine biosphere, which could in turn have a feedback effect on atmospheric_CO2 levels. To assess the reality of this feedback, researchers are conducting controlled laboratory experiments by subjecting certain phytoplankton species to variable_CO2 and temperature conditions. For example, cultures of certain unicellular algae (coccolithophorids) show a " fertilization " effect when _pCO2 increases between 200 and 800 ppm.

Laboratory experiments on monospecific cultures are complemented by studies on entire ecosystems in enclosures in contact with the open ocean (mesocosms). Artificial fertilization experiments using_CO2 increase also show an increase in primary production. This in turn tends to lower dissolved_CO2 levels, leading to increased_CO2 flux at the air-sea interface(i.e. net diffusion from the atmosphere to the ocean). Mesocosm experiments also illustrate the complexity of an ecosystem's response, which is not simply an increase in primary productivity. Indeed, these experiments also show variations in ecosystem carbon losses linked to sedimentation.

Changes in mesocosm temperature indicate that a warming of water masses would tend to increase carbon remineralization and decrease sedimentation, thus limiting the effect of the organic pump on the_CO2 content of atmospheric air.

The complexity of the biospheric response to warming stems from the fact that changes in ocean dynamics have an indirect influence on primary productivity. Warmer surface waters increase water column stratification. At low and mid-latitudes, this stratification would tend to reduce nutrient recharge to surface waters, which should help to limit primary production. In contrast, biological productivity at high latitudes is generally not limited by nutrient supply, but by light intensity. Increased stratification tends to concentrate phytoplankton in the upper part of the water column, which would favour access to light energy, thus increasing productivity in these regions.

Predicting the response of the marine biosphere is therefore extremely difficult, as the combined direct and indirect effects of increased_CO2, changes in water temperature and salinity must be taken into account. A complementary approach to these mechanism studies is the direct measurement of variations in marine productivity over recent decades.

Plankton counts have been carried out since the early 1930s using the continuous plankton recorder. The series show that plankton abundance increased by a third in the 1980s at the surface of the North-East Atlantic.

To complement these in situ data, researchers use satellite-borne sensors to provide complete coverage of the world's oceans. Using empirical calibrations, it is possible to convert ocean color measurements - more precisely, the ratio of blue to green reflectances - into terms of surface chlorophyll concentration. The main measurement campaigns were carried out from 1979 to 1986 with the coastal zone color scanner (CZCS), followed by the sea-viewing wide field-of-view sensor (SeaWIFS) from 1997 to 2010. Geographical contrasts in chlorophyll content roughly follow the distribution of nutrients.

Satellite chlorophyll mapping also enables us to study temporal variations in primary productivity. SeaWIFS data show systematic variations at low and mid-latitudes. These high-frequency changes are due to the effect of ocean stratification caused by the ENSO oscillation on these time scales. For example, the 1998 El Nin͂o event is accompanied by surface warming and strong stratification, which generates a transient drop in biological productivity. A regional-scale study of satellite maps confirms the strong link between surface temperature and productivity, testifying to the effect of stratification in limiting nutrient recharge in the photic zone.

By normalizing the CSZC and SeaWIFS data, the researchers also attempted to detect longer-term trends. The comparison suggests a global increase in primary productivity of around 20 %, as well as significant regional variations. The latter are spatially correlated with temperature changes linked to several natural oscillations in the climate system.

Direct quantification of changes in primary productivity therefore highlights the importance of high-frequency natural cycles, which complicate detection of any long-term trends linked to warming and rising atmospheric _pCO2. The available series also highlight the weaknesses of current sampling: in particular, the absence of reliable estimates on a global scale before the early 1970s, and the problem of discontinuities between different satellite campaigns (e.g. the gap between CZCS and SeaWIFS).