The boreal summer intraseasonal oscillation: Relationship between northward and eastward movement of convection

The summertime intraseasonal oscillation (ISO) is an important component of the south Asian monsoon. Lagged regressions of intraseasonally filtered (25-80 days) outgoing longwave radiation (OLR) reveal that centers of convection move both northward and eastward from the central equatorial Indian Ocean subsequent to the initiation of an ISO. Eastward movement of convection is also seen at Indian subcontinent latitudes (10°-20°N). Based on the regression results, the summertime ISO convection signal appears as a band tilting northwestward with latitude and stretching from the equator to about 20°N. Viewed along any meridian, convection appears to propagate northward while equatorial convection propagates to the east. To examine the robustness of the connection between eastward and northward movement, individual ISOs are categorized and composited relative to the strength of the large-scale eastward component of convection in the central equatorial Indian Ocean. It is found that the majority of ISOs that exhibit northward movement onto the Indian subcontinent (42 out of 54 ISOs, or 78%) also exhibit eastward movement into the western Pacific Ocean. It is also found that when convection in the central Indian Ocean is not followed within 10-20 days by convection in the western Pacific Ocean (12 out of 54 ISOs, or 22%), the independent northward movement of convection in the Indian Ocean region is somewhat stunted. The link between the eastward and northward movement of convection is consistent with an interpretation of the summertime ISO in terms of propagating equatorial modes. The northward moving portion of convection is forced by surface frictional convergence into the low pressure center of the Rossby cell that is excited by equatorial ISO convection. A similar convergence pattern is seen for the northern winter ISO, but it does not generate poleward movement due to relatively cool SSTs underlying the surface convergence.