The South Asian summer monsoon that occurs over 4 months from near summer solstice (June) to near autumnal equinox (September) evolves within the season due to interactions with other climate components. Since most climate patterns are seasonally phase-locked, their interaction and feedback with monsoon should also be season-dependent. Here, using rain gauge-based precipitation over India for 74 years, we show that the onset phase of the monsoon (June) is uncorrelated with the rest of the season (July-September). The propagation of the Madden-Julian oscillation (MJO) over the Indian Ocean triggers northward propagation of precipitable water that plays a key role during the onset phase. The MJO appears over the western equatorial Indian Ocean during both the early and late central Indian onset years. Normal and slow MJO phase propagation during early and late onset years, respectively, gives rise to interannual variations in June rainfall. After onset, till the end of the monsoon season, the interannual variability of the dominant seasonal mode of precipitation is regulated by the phase of El Ni & ntilde;o-Southern Oscillation (ENSO) through anomalous moisture convergence. We show that, in June, ENSO does not play a key role because of the opposing impact on wind and moisture. However, from July to September, ENSO perturbs wind and moisture in a way that gives rise to positive feedback between monsoon, ENSO, and large-scale circulation. Such feedback of monsoon with ENSO results in a coherent impact of ENSO on monsoon precipitation in July-September. Our results suggest that the dominant mechanisms for interannual variations in Indian monsoon rainfall in June and the rest of the season are different. SIGNIFICANCE STATEMENT: The variations in rainfall within the monsoon season in India are extremely important for agriculture and water resource management. For such purposes, prediction of the variations in rainfall within the season is more important than the prediction of the seasonal mean. However, very little study has been done to understand the generalized physical processes that govern the monthly variations in rainfall every year without assuming a particular intraseasonal feature. Here, we try to find the existence of the physical modes that are responsible for monthly variations in rainfall every year. We show that the dominant mechanisms behind the interannual variabilities of June and post-June rainfall are different. While June rainfall is governed by equatorial waves, post-June rainfall is a function of feedback between monsoon and ENSO. If the climate models capture these physical modes accurately, the prediction of the seasonal cycle of rainfall will improve substantially.
