Until recently the main motivation in sea ice modeling has been toward the development of large-scale models for climate studies. These models describe sea ice as a plastic material, with a smooth yield surface and ice strength dependent on a thickness distribution that is based on statistical representations of sea ice deformation through ridging. With tuning, they are found to reproduce ice extent and concentration in the Arctic and Antarctic, though velocity fields are overly smooth and many details, such as polynyas and leads, are not captured.

There is increasing interest in regional ice modeling. In the near-shore Beaufort and Chukchi seas, there is considerable interest from the oil industry in the formation and breakup of landfast ice, the propagation of oil spills, and prediction of sea ice conditions. The importance of resolving eddies in the ocean and modeling small-scale (sub-10-km) sea ice processes is becoming apparent, as we begin to understand the non-linear effect of small-scale processes on the large-scale motion. Recently, there have been advances in the direction of small-scale process research and regional ice-ocean model development. The most pertinent of these are outlined in this article.