A model-data study of the 1999 St. Lawrence Island polynya in the Bering Sea

Abstract

[1] A Coupled Ice Ocean Model (CIOM) and in situ measurements were used to investigate sea ice and the St. Lawrence Island polynya (SLIP) in the Bering Sea in 1999. The modeled 1999 seasonal cycle of ice cover compared well with satellite measurements. The simulated maximum sea ice coverage was ∼0.8 × 106 km2, and the simulated maximum sea ice volume was ∼344 km3. The polynya south of St. Lawrence Island was captured by the CIOM and investigated in depth against the measurements. It was found that an offshore wind was necessary, but not sufficient on its own, for the development of the SLIP. It was found that a strong offshore wind, offshore surface water velocity, and the angle (<60°) between wind and water current are the three major factors for the development of the SLIP. Multiple-variable, linear regression models were developed to confirm these three mechanisms. Yearly potential sea ice production in the SLIP area was estimated to be about 95.7 km3, which accounts for 2.8% of the total potential production of 3393 km3 in the whole Bering Sea. Sea ice contributes to approximately 63% of winter salinity changes in the Bering Shelf (<200 m), while the SLIP can contribute more than twice the local salinity changes. The relationships among wind, sea ice, and surface ocean current were examined. The classic Ekman drift theory (that surface water velocity drifts 45° to the right of the wind direction) is modified to be 50.4° on the ice-covered Bering Shelf due to the year-round existence of a background northward ocean transport.