The ocean and climate change : relations with marine chemistry and biology (3)

The large uncertainties associated with ocean carbon balance are forcing geochemists to look for alternative methods of estimating ocean sequestration. Natural carbon contains around 1 % isotope 13. The ^13C/12Cratio (^δ13C) shows variations in the different reservoirs of the carbon cycle. These differences are due to thermodynamic equilibrium fractionations or kinetic fractionations linked to the speed of biological reactions. Compared with the huge reservoir of the deep ocean, atmospheric_CO2 has a ^δ13Cdeficit of - 7 ‰, and carbons derived from photosynthesis have ^δ13Cranging from - 20 to - 30 ‰. Through the balance effect of plankton photosynthesis, dissolved surface _TCO2 is ^{13C-surplus by}around 1 to 2 ‰. The oceanic distribution of the ^δ13Cratio of dissolved _TCO2 logically follows the effects of remineralization deduced from total carbon: positive correlation between ^δ13Cand _O2 content and negative correlations with nutrients. This natural stationary distribution has been disrupted by the injection of_CO2 generated by fossil fuel combustion and deforestation. Both sources are derived from photosynthesis, with coals, lignites and hydrocarbons showing varying ^δ13Cvalues depending on their origins and geological ages.

From measurements of the ^δ13Cof atmospheric_CO2, directly in the air or in air bubbles occluded in polar ice, or from the cellulose of tree rings, it is possible to quantify the progressive decrease in atmospheric ^δ13Clinked to anthropogenic emissions. This anomaly of around - 1.5 ‰ since the ^19th century has been propagating in the ocean, and it is therefore possible to track it by direct measurements on the _TCO2 of seawater since the 1970s (GEOSECS campaigns) as well as on massive corals or calcareous sponges with annual rings like trees. The ^δ13Cof the surface ocean follows that of the atmosphere with a time lag of a few years. Since the 1980s, seasonal measurements of oceanic _pCO2, _TCO2 and ^δ13Chave been carried out at several stations, notably in Bermuda and Hawaii. The parallel evolution of the different chemical species confirms the progressive decrease in ^δ13Cand its usefulness in distinguishing the propagation of anthropogenic carbon from natural seasonal effects linked to biological productivity and mixing of water masses.

Around 25,000 ^δ13Cmeasurements were taken at the surface and at depth during the WOCE oceanographic campaigns of the 1990s. The ^δ13Cprofiles are compared with those measured in the 70s to deduce an oceanic inventory of ^13Cpenetration ^thatdepends on each geographical area considered. Agreement with ^{13Cthermonuclear}inventories confirms the complementarity of the two isotopes for studying the sequestration of anthropogenic carbon by the ocean. Uptake for the period from the 70s to the 90s corresponds to 1.5 ± 0.6 GtC per year, compared with fossil carbon emissions of the order of 5 GtC over the same period (6 GtC/year with deforestation). This means that ocean pumping accounts for around 30 % of fossil carbon emissions, in line with methods based on total carbon measurements.