Coupled Modes of the Warm Pool Climate System. Part I: The Role of Air–Sea Interaction in Maintaining Madden–Julian Oscillation

Abstract

Abstract Over the warm pool of the equatorial Indian and western Pacific Oceans, both the climatological mean state and the processes of atmosphere–ocean interaction differ fundamentally from their counterparts over the cold tongue of the equatorial eastern Pacific. A model suitable for studying the coupled instability in both the warm pool and cold tongue regimes is advanced. The model emphasizes ocean mixed layer physics and thermodynamical coupling that are essential for the warm pool regime. Different coupled unstable modes are found under each regime. In contrast to the cold tongue basic state, which favors coupled unstable low-frequency SST mode, the warm pool regime (moderate mean surface westerlies and deep thermocline) is conducive for high-frequency (intraseasonal timescale) coupled unstable modes. The wind–mixed layer interaction through entrainment/evaporation plays a central role in the warm pool instability. The cloud-radiation feedback enhances the instability, whereas the ocean wave dynamics have little impact. The thermodynamic coupling between the atmosphere and ocean mixed layer results in a positive SST anomaly leading convection, which provides eddy available potential energy for growing coupled mode. The relatively slow mixed layer response to atmospheric forcing favors the growth of planetary-scale coupled modes. The presence of mean westerlies suppresses the low-frequency SST mode. The characteristics of the eastward-propagating coupled mode of the warm pool system compares favorably with the large-scale features of the observed Madden–Julian Oscillation (MJO). This suggests that, in addition to atmospheric internal dynamic instability, the ocean mixed layer thermodynamic processes interacting with the atmosphere may play an active part in sustaining the MJO by (a) destabilizing atmospheric moist Kelvin waves, (b) providing a longwave selection mechanism, and (c) slowing down phase propagation and setting up the 40–50-day timescale.