This study focuses on attaining an improved understanding of the atmospheric phenomenon termed the Madden Julian Oscillation (MJO). The methodology used to explore the maintenance of the MJO is a scale interactions energetics formulation that follows Krishnamurti et al. (2000), Hayashi (1980) and Sheng (1986). Krishnamurti et al. (1998) derived an expression for the nonlinear energy transfer into a frequency from its interaction with other frequencies. Simply put, atmospheric phenomena that occur on the MJO time-scale hand energy to other time-scales (frequencies) and also receive energy from these time-scales. These interactions occur both in-scale (quasi-nonlinear quadratic) and between scales (nonlinear triad interactions).

The primary objective of this study is to perform multiple simulation experiments using a coupled model (containing an atmospheric and oceanic component) to examine the energy transfer between the MJO and other oscillations in the atmosphere that occur on varying time-scales.

An initial analysis of 44 years of global, full-atmosphere ‘re-analysis’ data available from the National Center for Atmospheric Research (NCAR) has been under-taken. Examination of the velocity potential (divergent) and stream function (rotational) components of the 200 and 850 hPa atmosphere has been investigated for the regions of the tropics as well as globally.

The coupled ocean-atmosphere model that will be employed for this study is similar to that used by Krishnamurti et al. (2000). The atmospheric component of this model is a global spectral model at the resolution of 42 waves using triangular truncation with 14 vertical layers between the earth’s surface and the top of the atmosphere. The oceanic component of this coupled model scheme is the Max Planck Institute global ocean model (Latif, 1987; LaRow and Krishnamurti, 1998; Krishnamurti et al. 2000).

The eventual maintenance of the MJO time scale may rely on its interaction with many different time-scales. The time-scales that will be investigated using the coupled model and re-analysis data are as follows:

· MJO (30-60 days)

· Decadal and beyond (10 years and beyond)

· El Nino Southern Oscillation (3-6 years)

· Annual

· Quasi Biennial Oscillation (~ 2 years)

· Semi-Annual

· Synoptic (3-7 days)

Preliminary results indicate and support the findings of previous studies where the rotational component of the wind dominates the divergent component for the triad interactions among all frequencies listed above.