An examination of several ice control mechanisms in a coupled ice‐ocean numerical model of the arctic

A fully prognostic, coupled ice‐ocean numerical model is used to examine the relative importance of several ice‐cover forcing mechanisms in large‐scale interannual simulations of Arctic ice cover. Validation of the simulations is conducted through statistical and qualitative comparisons between simulated and observed fields of ice concentration and thickness. A reduction of the ice strength permitted greater compaction and more realistic ice thicknesses. In contrast to less elaborate models, this model is insensitive to frozen‐surface albedo changes. The vertical ocean flux appears to be the dominant term controlling the position and extent of the simulated marginal ice zone. Within the pack, it is the dynamic forcing, particularly from the wind, which controls the thickness distribution. A strong feedback mechanism is proposed to explain the excessive summer ice‐melt in the Chuckchi, east Siberian and Laptev seas. The interaction between various ice control mechanisms is shown to be important, with considerable differences between regions.