Interactions of Large-Scale Dynamics and Diabatic Heating in Multimodel MJO Simulations

The Madden-Julian Oscillation (MJO) is an intraseasonal disturbance that has direct impacts on climate extremes and tropical weather with connections to global weather. Studies have found the MJO is underrepresented in general circulation models (GCM). This study looks into the dynamics of the MJO in a group of climate models from the GEWEX Atmospheric System Study (GASS) and the MJO Task Force (MJOTF). The component of the MJOTF/GASS project focused on is the 20-year (1991-2010) climate simulations. Over 25 models were compared an observational data set containing Tropical Rainfall Measuring Mission (TRMM) precipitation, dynamics and thermodynamics from ECMWF’s Interim Re-Analysis (ERAI). Anomalies were projected onto a new precipitation-based index for the MJO and focuses on the months November-April.

Column-integrated moist static energy (MSE) is a proxy for tropical convection and thus, investigating its budget equation is useful for understanding MJO propagation. The column-integrated MSE budget in the MJOTF models has been investigated in a few studies and they identified low-level horizontal moisture advection as crucial to the simulation of coherent eastward MJO propagation. More specifically, the horizontal advection of background moisture by the intraseasonal large-scale circulation is most critical. While many studies have focused on the impact of background moisture gradients, this study focuses specifically on the anomalous circulation. One of the main ingredients separating good and poor MJOTF models is their representation of the anomalous low-level easterlies and convergence over the Maritime Continent. To better understand the mechanisms missing in many climate models, the anomalous circulation is broken into its contributions by different time and spatial scales including convectively coupled equatorial waves. Additionally, anomalies of the diabatic heating and apparent drying are examined to explicate the driving mechanisms of the poor and good model MJO simulations.