engleski

Seasonal skill and predictability of ECMWF PROVOST ensembles

Variations in seasonal-forecating skill and predictability during the 15 years (1979-93) of the ECMWF re-analysis have been studied using 120-day ensemble integrations of the ECMWF model. The model systematic error is found to be comparable with a typical amplitude of interannual variation. Ensemble-mean skill scores were found to be positive overall. The interannual variation of skill scores is much larger for the European region than for hemispheric domain. Estimates od intrinsic predictability were made for each year of the dataset. They indicate genarally high predictability in years when ENSO was strong. Overall SSTs have a dominant role though the impact of initial conditions is not negligible.

El Nino; ensemble prediction; seasonal predictability

nije evidentirano