Forecasting the peak anomalies of dominant intrinsic modes of Indian Ocean Dipole

It is important and highly desirable to identify and forecast the strength of dominant modes of Indian Ocean Dipole (IOD) events. A daily time series of dipole mode index (DMI) is derived from remotely sensed satellite SST observations. The empirical mode decomposition (EMD) technique is applied on this time series to identify intrinsic mode functions of observed IOD index. The spatial maps of the dominant IOD modes exhibit observed dipolar structure. For each dominant IOD mode, a forecasting rule is given to predict the peak SST anomalies (strength) of the positive and negative IOD regimes. The feasibility of using growth rate of SST anomalies in one regime for the prediction of peak SST anomalies in subsequent regime is also explored in an effort to develop an extended-range forecast model. It is shown that the peak SST anomalies in a regime corresponding to the most dominant mode of the IOD (with a time scale of 340–619 days) may be predicted with a lead-time of nearly 9–10 months.