A finite element sea ice model of the Canadian Arctic Archipelago.

The Canadian Arctic Archipelago (CAA) is a complex area formed by narrow straits and islands in the Arctic. It is an important pathway for freshwater and sea-ice transport from the Arctic Ocean to the Labrador Sea and ultimately to the Atlantic Ocean. The narrow straits are often crudely represented in coupled sea-ice-ocean models, leading to a misrepresentation of transports through these straits. Unstructured meshes are an alternative in modelling this complex region, since they are able to capture the complex geometry of the CAA. This provides higher resolution in the flow field and allows for more accurate transports (but not necessarily better modelling). In this paper, a finite element sea-ice model of the Arctic region is described and used to estimate the sea-ice fluxes through the CAA. The model is a dynamic-thermodynamic sea-ice model with elastic-viscous-plastic rheology and is coupled to a slab ocean, where the temperature and salinity are restored to climatology, with no velocities and surface elevation. The model is spun-up from 1973 to 1978 with NCEP/NARR reanalysis data. From 1979 to 2007, the model is forced by NCEP/DoE reanalysis data. The large scale sea-ice characteristics show good agreement with observations. The total sea-ice area agrees very well with observations and shows a sensitivity to the Arctic oscillation (AO). For 1998-2002, we find estimates for the sea-ice volume and area fluxes through Admunsen Gulf, McClure Strait and the Queen Elizabeth Islands that compare well with observation and are slightly better than estimates from other models. For Nares Strait, we find that the fluxes are much lower than observed, due to the missing effect of topographic steering on the atmospheric forcing fields. The 1979-2007 fluxes show large seasonal and interannual variability driven primarily by variability in the ice velocity field and a sensitivity to the AO and other large-scale atmospheric variability, which suggests that accurate atmospheric forcing might be crucial to modelling the CAA. [ABSTRACT FROM AUTHOR]