The influence of ice-ocean coupling feedbacks on Arctic sea ice variability

Observations indicate that the Arctic ice cover is highly variable and has sizable trends in area and volume. These trends and variability are modified by feedback processes that occur at high latitudes. In this study, a developmental version of the NCAR community sea ice model is used to examine Arctic sea ice variability from 1958-1998. This model incorporates an elliptical yield curve based sea ice rheology, a sub-gridscale ice thickness distribution, and multi-level energy conserving thermodynamics. It is coupled to a slab ocean mixed layer model which allows for variable penetrating solar radiation and ice/ocean heat exchange to occur.

An amplifying feedback mechanism is associated with variations in ice/ocean coupling during the summer. When relatively large summertime lead fractions are present due to enhanced melting or dynamical effects, more solar radiation is absorbed in the ocean mixed layer, increasing the ice/ocean heat exchange and in turn further decreasing ice concentration and thickness. This positive feedback is an important component of the albedo feedback mechanism. In contrast, ice/ocean coupling acts as a negative feedback during the winter. When more open water or thin ice is present, the oceanic heat loss is relatively large, resulting in higher ice growth and reducing the anomalous ice conditions. Here we examine the influence of ice/ocean coupling feedbacks on modifying the Arctic sea ice trends and variability from 1958-1998. The strength of the feedback mechanism and its regional signature are discussed.

Additional sensitivity tests are performed to determine the influence of sea ice model physics on the feedback strength. These include the effects of the mean state of the ice cover, the resolution of the sub-gridscale ice thickness distribution, and the parameterization of the strength of the ice pack. These parameterizations modify the ice/ocean coupling feedbacks largely due to their influence on the open water formation that occurs in the model simulations.