The role of sea ice in the carbon cycle of Polar Seas: 1D to 3D modelling

• Main Authors: Moreau, S., Vancoppenolle, M., Delille, Bruno, Tison, J.-L., Zhou, Jiayun, Kotovitch, Marie, Thomas, D., Geilfus, N.-X., Goosse, H.
• Format: Conference Object
• Language: English
• Published: 2015
• Subjects: Physical; chemical; mathematical & earth Sciences; Earth sciences & physical geography; Physique; chimie; mathématiques & sciences de la terre; Sciences de la terre & géographie physique; Antarc*; Antarctica; Arctic Ocean; Sea ice; Southern Ocean
• Online Access: https://orbi.uliege.be/handle/2268/186747

Sea ice participates actively in the biogeochemical cycle of carbon of Polar Oceans, yet to which extent is not clear. We investigated the processes that govern sea ice carbon dy- namics in Polar Regions through 1D to 3D modelling developments. First, we constrained all major physical and biogeochemical processes with respect to CO2 dynamics (carbon- ate chemistry, biological activity, ikaite (CaCO3•6H2O) precipitation and dissolution and ocean-ice-air CO2 fluxes) in a one-dimensional sea ice model. According to our model, the CO2 budget is driven by the CO2 uptake during ice growth and release by brine drainage, whereas other processes such as brine-air CO2 fluxes, despite significant, are secondary. Subsequently, based on these preliminary conclusions, we evaluated the role of sea ice in the carbon dynamics of Polar Oceans by using an ocean-ice coupled Global Earth System Model. Carbon dynamics (e.g. ocean-atmosphere CO2 fluxes) are driven by the contribution of sea ice growth regions in the Arctic Ocean (mainly the Siberian coast) and sea ice melt regions in the Southern Ocean (off the coast of Antarctica). In addition, the production of deep waters is low in the Arctic Ocean but significant in the Southern Ocean. Therefore, sea ice only contributes to the deep water export of carbon in the Southern Ocean. The role of sea ice in the biogeochemical cycle of carbon is significant and its representation by Global Earth System Models should be improved.