Evaluation and application of Finite Element Sea Ice-Ocean Model (FESOM) for the Arctic-North Atlantic region in variable resolution global simulations

Abstract

The changes and variability of the Arctic-North Atlantic Oceans and the inter-exchange of water mass and heat is of crucial relevance for the global ocean and climate. In this work the Finite Element Sea-ice Ocean Model (FESOM) is used to simulate the global ocean with focus on the Arctic-North Atlantic region. FESOM is formulated on unstructured meshes and offers variable-resolution functionality which is diffcult to achieve in traditional structured-mesh models. With this powerful tool we are able to resolve the key areas with locally refined resolutions in global simulations. The performance of FESOM in the Arctic Ocean and North Atlantic on large time scales is first evaluated in a 240-yr hindcast experiment. The model can reproduce realistic Atlantic Meridional Overturning Circulation (AMOC) and realistic Arctic freshwater content variability and sea ice extent. A water-hosing experiment is conducted to study the model sensitivity to increased freshwater input from Greenland Ice Sheet (GrIS) melting in a 0.1Sv discharge rate scenario. The released freshwater from Greenland can penetrate into the Arctic Ocean, especially in the Eurasian Basin. The anomalous freshwater also leads to a reduction in the AMOC strength and changes in freshwater exchange between the two oceans. Simulations with different local resolutions of 24 km and 9 km in the Arctic Ocean and surrounding regions are carried out to study the influence of resolution on the simulated Arctic Ocean and Arctic-Subarctic fluxes. Both simulations can reasonably simulate the mean state and variability of sea ice condition, freshwater content in the Arctic Ocean, and the fluxes through the Arctic gateways when compared to observations and previous model studies. Although the high resolution (9 km) run tends to improve the representation of fluxes through the Arctic gateways and the salinity structure in the Arctic basin, higher meso-scale eddy resolving resolution is required to further improve the simulation. The driving mechanism of the interannual variability of Barents Sea sea ice is studied. We found that sea ice import into the Barents Sea drives the sea ice interannual variability. The amount of sea ice flux determines the thermodynamic growth rate in the Barents Sea. The increasing trend of Atlantic Water heat flux through the Barents Sea Opening leads to the decline of Barents Sea ice volume.