Sea Ice Variability in the Bering Sea

The Bering Sea consists of a shallow continental shelf in the northeast and a deep basin in the southwest. Sea ice in the Bering Sea concentrating in the shelf region exhibits large seasonal and interannual variations, which significantly impact the local marine ecosystem. Understanding of the physical mechanisms governing this sea-ice variability, however, remains incomplete. To better understand regional sea-ice variability we use a fine resolution (1/10-degree) global ocean and sea-ice model and available observations. The simulation consists of the Los Alamos National Laboratory Parallel Ocean Program (POP) and CICE models, and was run with Coordinated Ocean-ice Reference Experiment (CORE2) interannually varying atmospheric forcing for 1970-1989. Here we analyze 1980-1989; the first 10 years are treated as the spin-up period. We examine the partitioning of the ice volume tendencies into thermodynamic and dynamic components, as well as corresponding surface atmospheric and oceanic variables, in order to understand the relationship between sea ice variability and varying atmospheric and oceanic conditions. Focusing on the seasonal cycle first, we find that sea ice is mainly formed in the northern Bering Sea with the maximum ice growth rate occurring along the coast. Winds cause sea ice to drift southwestward from the north to the western ice edge. Along the ice edge, ice is melted by warm waters carried by the Bering Slope Current, especially in the west in winter; while in fall and spring, basal melting of sea ice spreads into the interior of ice pack. The ice growth rate is larger in winter than in fall and spring. Ice transport from the north to the southwest becomes weaker in fall and spring than in winter. The Bering Sea is ice free in summer. Surface melting is insignificant in all seasons. The interannual variability of sea ice in the Bering Sea can be largely explained by thermodynamics on the large scale. The dynamic ice transport, however, is often important locally, especially around the ice margins with ocean and land. Local dynamic and thermodynamic ice volume changes usually have opposite signs with similar magnitudes, implying a negative feedback between them. Through the surface heat flux budget, we find that sensible heat flux dominates the surface heat exchange with the atmosphere, which controls the ice growth in the north. Ocean-ice heat flux largely determines the basal melting along the ice edge in the south. Through the force balance analysis, we find that the ice motion is in steady state on the monthly timescale. Ice velocity correlates well with the wind stress, which is nearly balanced by the opposite ocean stress. Internal ice stress is not substantial except near the land boundaries in the north.