For the model, three sets of ensembles are run, with MJO-like phase 1 and phase 5 heating, and without MJO heating. This approach allows for the examination of the sensitivity of the midlatitude flow evolution to (1) the initial midlatitude flow without MJO forcing and (2) the MJO heating and interaction with the midlatitude flow. For the reanalysis data, a systematic projection approach is used to obtain analog flows associated with MJO events. Analogs are obtained for those days when the MJO is both active (for all eight phases) and inactive. This approach allows for an estimation in the observations of the same questions addressed by the model runs.
The focus of the above analysis is on the Pacific-North-American region. It is found that large amplitude anomalies in this region are associated with a strong Rossby wave source and the excitation of poleward propagating Rossby waves, and are particularly sensitive to the initial presence of synoptic-scale disturbances located over eastern Asia and the North Pacific. The analog approach shows that the 1-2 week predictability of the midlatitude flow is enhanced when the MJO is active. Furthermore, when the MJO is active, it is found with the analog approach that the predictability of the midlatitude flow is dependent upon both the MJO tropical heating and the initial midlatitude flow, with the contribution by the tropical convection being much greater.