MJO Prediction skill in ENSEMBLES

MJO Prediction Skill in ENSEMBLES Waqar Younas, Youmin Tang Natural Resources and Environmental Science, University of Northern British Columbia, Canada In this study, the variability and predictability of the Madden-Julian Oscillations (MJO) is evaluated with time average, using state-of-the-art dynamical multiple model ensemble of the ENSEMBLES data. Emphasis is placed on the evaluation of actual prediction skill of the ensembles mean and the information-based potential predictability, and comparison of information-based potential predictability measure with usual signal to noise ratio measures, examined at various time scales. It is well established that most of current dynamical models are still lacking to correctly simulate the MJO variability due to model deficiencies and imperfect initial conditions. Using combined EOF patterns of winds and precipitation, the MJO variability has been analyzed in different coupled models. The comparison of these patterns with observations revealed that models have better representation of the MJO variability. Using first two PC time series, actual and potential prediction skill is analyzed at daily, 3-days, 5-days and 10-days averaged data. It is found that both actual and potential predictability skill increased with time average and Multiple-

model ensemble (MME) skill was better than most of the individual models. In terms of potential skill, Relative Entropy (RE) is an effective measure in characterizing the potential predictability of individual prediction, whereas the Mutual Information (MI) is a reliable indicator of overall prediction skill. The comparison with conventional potential predictability measures of the signal-to-noise ratio, reveal that Mutual Information (MI) characterized more potential predictability when the ensemble spread varied over initial conditions. ace="Symbol"face="Symbol"face="Symbol"face="Symbol"face="Symbol"face="Symbol"face="Symbol"