Ocean data assimilation (ODA) is a procedure that combines satellite data and other more direct measurements from ships and buoys with information from predictive models to give the best possible estimate or analysis of the ocean state at a given time. This estimate can be used to initialize climate prediction models or to study ocean phenomena. The forecast skill of climate prediction models is sensitive to their initialization. Ideally, improvements in ODA are reflected in improved forecasts.

In ODA, observations are combined with information from predictive models in a manner that depends on statistical representations of the observational and model errors. By including more dynamical information in the model error representation, observations are used primarily to correct large-scale errors. In the NCEP ODA system, a component of IRI's two-tier forecast system, implementation of this approach has resulted in more dynamically balanced temperature correction fields and improved ocean currents.

Mean equatorial temperature correction (NCEP)	Mean equatorial temperature correction (IRI-NCEP)
Mean equatorial zonal velocity (NCEP)	Mean equatorial zonal velocity (IRI-NCEP)

IRI Participants

Collaborators