Main drivers of Indian Ocean Dipole asymmetry revealed by a simple IOD model

Indian Ocean Dipole phenomenon (IOD) refers to a dominant zonal contrast pattern of sea surface temperature anomaly (SSTA) over tropical Indian Ocean (TIO) on interannual time scales. Its positive phase, characterized by anomalously warm western TIO and anomalously cold southeastern TIO, is usually stronger than its negative phase, namely a positively skewed IOD. Here, we investigate causes for the IOD asymmetry using a prototype IOD model, of which physical processes include both linear and nonlinear feedback processes, El Niño's asymmetric impact, and a state-dependent noise. Parameters for the model were empirically obtained using various reanalysis SST data sets. The results reveal that the leading cause of IOD asymmetry without accounting seasonality is a local nonlinear process, and secondly the state-dependent noise, the direct effect by the positively skewed ENSO and its nonlinear teleconnection; the latter two have almost equal contribution. However, the contributions by each process are season dependent. For boreal summer, both local nonlinear feedback process and the state-dependent noise are major drivers of IOD asymmetry with negligible contribution from ENSO. The ENSO impacts become important in boreal fall, along with the other two processes.