A comparison of ice drift motion from modeled and buoy data

Abstract

U.S. Naval operations in the Arctic require an effective way to predict the movement and behavior of sea ice. This is currently provided by the Navy's PIPS model which is based on Hibler's (1979) sea ice model which combines the thermodynamic ice heat budget with a dynamic ice model sensitive to the effects of ice thickness and ice strength. The PIPS model simultaneously solves a system of four equations of ice momentum balance, ice rheology, ice thickness, and ice strength. In order to test the performance of Hibler's formulation, another version of it, developed by Lemke et al., (1990), was adapted to the Arctic Ocean. The model was initialized and run using 1986 forcing data and its performance evaluated using Arctic buoy drift data. Results indicate that the model ice drift is principally driven by wind forcing, that its response to changes in weather is rapid and essentially correct, and that it performs better at high wind speeds than at low wind speeds. Limitations to its accuracy were chiefly the result of limits to the precision and resolution of the input data provided to run the model, especially near the ice margins. Overall, the model performs well in depicting the ice flow pattern in all conditions in the Arctic.