WP 5 Drift modelling and prediction
Dr. Kai Christensen (MET Norway) and Prof. Rune Graversen (UiT)
The objectives of this work package are to assimilate observations collected within CIRFA into an ocean-ice forecast model, and to produce probabilistic ocean, sea ice and drift forecasts.
Ocean and ice forecasting at high latitudes, including the forecasting of drift of icebergs, oil-spills and other pollutants, is challenging due to a severe lack of observations of oceanic ‘weather’. Improved operational forecast systems will require increased amounts of high-resolution observations and the assimilation of such data into ocean and ice models. Also, because of the large observational uncertainties and the chaotic nature of the flow, the forecasts have to be probabilistic, i.e. presented as a range of possible outcomes based on an ensemble of slightly different model runs.
So assimilation and ensemble forecasting are the central issues to be addressed within WP5. The work will utilize ocean-ice models and assimilation techniques already in use at MET Norway today, but substantial advances will need to be made. The most crucial steps ahead are 1) to form proper descriptions of the errors or uncertainties for the new types of observations to be assimilated, and 2) to optimize the coupling between the assimilation system and the ensemble prediction system (EPS). The coupled system will also allow for the optimization of exchange coefficients of fluxes (momentum, heat, fresh water) between ocean, sea ice and atmosphere.