Data assimilation in an ice-ocean coupled model to improve sea ice predictions in the Arctic Ocean

Abstract

We have improved the accuracy of sea ice predictions in the Arctic Ocean by assimilating sea ice variables in an ice-ocean coupled model. The model domain is 25km in resolutions. The results of the model only predictions are bound by the uncertainties in initial conditions and forcing data. To minimize these uncertainties, sea ice concentration, sea ice thickness and sea ice velocity were assimilated in this study. Sea ice observations were obtained from AMSR2-data sets. The assimilation method that was used is an improved nudging method that minimizes the observation errors and model errors. As a result of assimilation, the ocean conditions have been greatly improved. This was evident in resulting ocean temperature. Compared with the other assimilated variables, sea ice concentration assimilation could yield better results.

Data assimilation in an ice-ocean coupled model to improve sea ice predictions in the Arctic Ocean (PDF Download Available). Available from: https://www.researchgate.net/publication/295857581_Data_assimilation_in_an_ice-ocean_coupled_model_to_improve_sea_ice_predictions_in_the_Arctic_Ocean [accessed Sep 28, 2017].