Response of high resolution coupled sea ice/ocean model to the assimilation of ice motion fields derived from microwave satellite imagery

The interannual response of a coupled ocean/ice model to the assimilation of derived ice motion vectors is investigated. The prognostic model is an 18 km and 30 level coupled ocean/ice model with an elastic-viscous-plastic (EVP) rheology and a two-category ice thickness distribution.

The ice motion data, obtained from the JPL Polar Remote Group, is derived from 85 GHz SSM/I brightness temperature fields through use of a two-dimensional cross correlation feature tracking algorithm. For the purposes of computing the motion error, the International Arctic Buoy Program interpolated ice motions are used as truth.

A five year test period, from 1992 to 1997 is examined. The absolute error of ice motions for the prognostic model, the assimilated solution, and the data are compared. The resulting analysis shows that where there is data nearby, assimilation significantly reduces the error in the entire Arctic basin. The improvement in the MIZ typically ranges between 30 and 60 percent. Transport properties are also altered.