MADDEN-JULIAN OSCILLATION AND SEA SURFACE TEMPERATURE INTERACTIONS IN A MULTI-SCALE FRAMEWORK

The ocean-atmosphere coupling can play a role in initiating and sustaining the Madden-Julian Oscillations (MJOs), which are the major intraseasonal oscillations in the atmosphere. In this thesis, the oceanic influence on MJOs is studied with reanalysis products, numerical models, and idealized theoretical models. The energy sources for MJOs are calculated with NCEP reanalysis. The perturbed potential energy is found to be the most important energy source for most MJO events. In some MJO events, the sea surface is warmed due to the reduced latent heat flux during the suppressed phase of MJOs. As a result, warm sea surface temperature anomalies (SSTAs) occur, which appear to prolong the life time of these MJO events. In a minority of the MJO events, warm SSTAs can drive the atmosphere actively and trigger MJO events. In these events, the warm SSTAs are attributable to the internal oceanic processes influenced by the warm Indonesian Throughflow (ITF), which spreads from the southeastern Indian Ocean to the western Indian Ocean and modifies the subtle balance between stratification and mixing in the western Indian Ocean. In addition, during the transit period between monsoon seasons, a few MJO events are sustained by the energy obtained from the mean kinetic energy. Since the MJO events have different energy sources, their mechanisms should be considered in the context of these energy sources. While the spatial scale of the SSTAs in the Indian Ocean is only of order 100 km, the scale of MJOs is of order 1000 km, raising the potential for interactions between the oceanic and the atmospheric oscillations with different scales and this is demonstrated to be possible with analytical solutions to idealized linear governing equations. With a reasonable choice of parameters, the meso-scale oceanic and the large-scale atmospheric oscillations can interact with each other and lead to unstable waves in the intraseasonal band in this linear coupled model. The coupling and frequency shifts between oscillations with different scales and the atmospheric/oceanic responses to small variations in the external forcing are also tested with numerical models. Incorporating the oceanic influence on MJOs and the multi-scale interaction appropriately in a numerical model is supposed to help improve the simulation and forecast of MJOs. The hypothesis of multi-scale interaction is also expected to have wide applications in other studies, in addition to the MJO-SST interaction. The theoretical and numerical approach adopted here should also serve as a prototype for enhancing the process understanding of intraseasonal variability and lead to improved predictive understanding