"Recent field studies indicate an active biogeochemistry within sea ice. The latter is characterized by significant primary production, large loads of iron, and fluxes of climatically significant biogases, namely carbon dioxide (CO2) and dimethyl-sulfide (DMS). The role of sea ice in the carbon cycle and climate system via a large-scale action on ocean biogeochemistry is relatively poorly studied, potentially important and missing in Earth System Models.In this project, it is proposed to explore this role, in particular via an impact on carbon and iron cycles. The specific contribution of the relevant processes on carbon production and export at depth – be those processes physical (light, vertical mixing) or biogeochemical (nutrients, release of micro-algae, iron, heterotrophic feeding) – will be studied. In addition, an attempt to quantify the uncertainties associated with under-sampling and poor understanding of the processes involved will be made.The proposed analysis relies on a synthesis of existing observation data in both Polar Regions, on the development of numerical modeling tools, and on the analysis of numerical simulations both at local and global scales. In particular, the role of the specific processes at work will be studied via sensitivity experiments. The modeling tools include: (i) a one-dimensional sea ice model for the study of local-scale processes and (ii) NEMO, a numerical modeling system of the blue, white and green ocean (ocean, sea ice, and biogeochemistry) at large scales. As this ocean modeling system is part of a state-of-the-art climate model, it will be further possible to evaluate the role of sea ice on carbon cycle and climate.The project should take part into the reduction of uncertainties on climate and carbon cycle-climate projections in the Polar Regions and orient future observation campaigns. By directly implementing the developments directly in NEMO, this project should benefit to a wide community."