Physical feedback on ocean pumping of excess atmosphericCO2

The complexity of the phenomenon stems from the fact that several climatic and biogeochemical system responses influence carbon sequestration in the ocean. A first climatic feedback is linked to the effect of temperature on_CO2 solubility and acid-base equilibria.

Following in the footsteps of the atmosphere, the ocean is warming from its surface, as shown by in situ oceanographic data since the end of the ^19th century, and more recently by satellite infrared observations and measurements relayed by the thousands of Argo floats that automatically sample the first few kilometers of the water column. Since the beginning of the ^20th century, the average temperature of surface waters has risen by around 1 °C. Over the century as a whole, virtually all regions of the ocean have warmed.

The solubility coefficient of_CO2 in water is a decreasing function of its temperature. In addition, the two acid-base dissociation constants are increasing functions of temperature. The combination of these effects leads to a decrease in oceanic pumping of excess atmospheric_CO2 in response to warming. For an atmospheric _pCO2 of 300 ppm, warming surface seawater by 1 °C to 20 °C leads to a new _pCO2 equilibrium at 313 ppm between the atmosphere and the ocean. Warming therefore generates an apparent outgassing of around 13 ppm. To a lesser extent, an increase in salinity also leads to a rise in dissolved_CO2 and thus to outgassing to restore equilibrium at the atmosphere-ocean interface.

In addition to the influence of physical parameters, we need to take into account the saturation effect of the acid-base buffer, which decreases with_CO2 diffusion. This is illustrated by the fact that Revelle's factor increases linearly with atmospheric _pCO2. Since the beginning of the industrial era, this factor has increased by about one unit in response to the 100 ppm increase in the atmosphere. Overall consideration of physico-chemical effects since 1950 shows that_CO2 sequestration in the ocean has been around 5 % lower than it would have been without these feedbacks.

Ocean feedbacks are not limited to the effects of physico-chemical equilibria. An additional feedback must be taken into account, involving changes in winds and ocean surface circulation. Indeed, even if excess atmospheric_CO2 imposes an average flux from the atmosphere to the ocean, net fluxes at the air-sea interface are highly variable according to geographical zones. Locally, the net flux is a linear function of the air-sea _pCO2 gradient, the temperature-dependent solubility of_CO2 in seawater, and a_CO2 transfer rate coefficient. This coefficient depends on sea state, which translates into a quadratic function as a function of wind speed at the ocean surface.

Over the past 40 years, the collection of more than 3 million measurements of the air-sea _pCO2 gradient has made it possible to map the_CO2 sink and source zones of the surface ocean, and to quantify the net mean annual flux of_CO2 from the atmosphere to the ocean. Recent studies have examined the temporal and spatial evolution of the pressure gradient, and hence of the flux of_CO2 to the ocean. In the absence of feedbacks, we expect a simple increase in flux linked to the acceleration of atmospheric content. Three decades of data show that the flux has indeed increased in many areas. Nevertheless, it appears that certain regions of the western tropical Pacific, Southern Ocean and North Atlantic show a decrease in the _pCO2 gradient, leading to a reduction in the flux of_CO2 to the ocean.

In parallel with these measurements, 3D numerical models of the ocean-atmosphere coupling are used to simulate and attempt to understand these complex feedbacks. When the model only takes into account the increase in atmospheric_CO2, the_CO2 flux increases almost everywhere on the ocean surface. On the other hand, if the numerical model is forced to simulate the climate variability of recent decades (temperatures and winds), the simulation of the pressure gradient shows major geographical contrasts, with zones where_CO2 flux decreases. Some of these areas correspond to the observations already mentioned (Western Tropical Pacific, Southern Ocean), but there is no agreement for all regions of the World Ocean.

Using this type of approach, several authors have quantified the increase in the oceanic sink over the last 50 years. The accelerating rise in atmospheric _pCO2 is thought to be responsible for a significant increase (around 2 GtC/year) in the oceanic sink over this period. On the other hand, warming waters and changing surface winds would have collectively reduced the ocean sink. The sum of these effects indicates that the oceanic_CO2 sink has continued to increase over the past 50 years, but less than without climate feedbacks (around 10-15 % on the 30-year_CO2 inventory).