Estimates of Ocean Currents Using a Labrador Sea Ice Model

Abstract

Efforts to develop a dynamical model of Labrador pack ice have been hampered by a lack of input data, especially ocean current data. This work reports the results of approaching this problem in another way, where model simulations are used as a basis for adjusting the input currents to obtain agreement with ice drift data. The ice drift data comes from a period in February 1977 when a ship was fast in the northern Labrador pack. Unfortunately, the ship experienced navigation problems and the direct current observations were suspect. The model simulations bear this out indicating that it would have been very unlikely that the observed currents along with the other forcing terms could lead to the observed ice drift. A climatological current data set gave better agreement and was slightly modified to give the best agreement. This improved data set was used for a short sensitivity test of the rheological specifications of the model. It was found that the ice drift tracks were not very sensitive to the linear viscous stress law but were more sensitive to the parameterization used to reduce ice velocities in areas of high ice concentration and fast ice. Further plans are to use a model with a more realistic rheological component in order to assess its importance in estimating currents.