Recent observational studies have discovered a dipole mode in the Indian Ocean: a zonal dipole structure of interannual variability that appears in various atmospheric and oceanic parameters, including SSTs, surface winds, OLR, sea surface height, etc. (Webster et al. 1999, Saji et al. 1999). These studies also suggest that the Indian Ocean dipole mode may primarily be an expression of internal dynamics, rather than a direct response to ENSO. Many regional climate anomalies, such as the heavy rainfall over the east Africa and severe droughts and forest fires over the Indonesian region, could be attributed to this anomaly mode in the Indian Ocean. With the use of general circulation models (GCMs), we will assess the impact of Indian Ocean dipole (IOD) events on the regional climate, under the condition that external forcing is absent.

Two GCMs are employed in the current study—the UCLA and ECHAM4 models. Both models are forced either by global climatological SSTs (“control runs”), or by an SST distribution of a dipole mode event in the Indian Ocean (IOD) and climatological values elsewhere (“IOD experiments”). The “control runs” of the two models reproduce a mean circulation of similar patterns. The simulated circulation in both models undergoes a seasonal cycle that is of comparable magnitude and phase to the observation. Both models also reproduce interannual variability with its patterns and intensity comparable to those of the observed, suggesting the consistency and capability of the two models on the application of interannual variability study.

For the “IOD experiments”, the multi-realization composites of different parameters show a robust atmospheric response to the Indian forcing in both models. The configurations observed during the IOD events are captured with general agreement in both models. However, some differences also exist. The common and disparate features of the response will be discussed, with particular interest in the Indian and the Southeast Asian summer monsoon.