A Possible Mechanism for Simulating a Stronger Madden-Julian Oscillation in Atmosphere/Ocean Coupled Models

The Madden-Julian Oscillation (MJO) is a convective, intraseasonal atmospheric phenomenon that propagates eastward from the Indian Ocean and across the tropical Pacific and brings variations in global weather. Modeling experiments have suggested that the atmosphere/ocean coupled models can do a better job predicting this phenomenon than non-coupled models, but the mechanism behind the improved performance remains unclear. In this analysis, we use the linear mixed layer model (MLM) of Back and Bretherton (2009) to develop and test a hypothesis for why coupled models simulate enhanced MJO activity. The linear mixed layer model reproduces variations in surface convergence associated with the MJO, and is used to examine the relative influence of boundary layer and free-tropospheric processes on the distribution of convergence. We further divide the boundary layer components in the solution into contributions from SST gradients versus temperature gradients and wind contributions from the free troposphere. Results show that the MJO surface convergence is influenced by SST gradients. We conclude that the SST gradient-driven convergence may contribute to the better performance of the ocean-coupled model.