Pathways to Better Prediction of the MJO: 1. Effects of Model Resolution and Moist Physics on Atmospheric Boundary Layer and Precipitation

Abstract

Abstract Despite recent advancements in numerical models, prediction of the Madden‐Julian Oscillation (MJO) remains a major challenge in numerical weather prediction and climate modeling. This study explores pathways for improving MJO prediction through systematic investigation of the effects of model resolution and moist physics on simulations of the MJO in Part 1, followed by effects of atmosphere‐ocean coupling in Part 2. The Unified Wave Interface—Coupled Model (UWIN‐CM) experiments with different cumulus parameterizations (CP) and explicit microphysics with convection‐permitting resolution are used to study the MJO convective initiation and eastward propagation. Observations of the atmosphere and ocean from the Dynamics of the MJO (DYNAMO) field campaign in 2011 are used to evaluate coupled model simulations. At lower resolution (12 km), the simulation with the Tiedtke CP produced an eastward propagating MJO event, while the simulation with the Kain‐Fritsch CP did not. The main difference between the two low‐resolution simulations is in the vertical structure of the atmospheric boundary layer which impacts the large‐scale MJO circulation and precipitation. The Kain‐Fritsch CP produced a persistent cloudy boundary layer that decoupled the tropospheric circulation from the surface and failed to produce the MJO. At higher resolution (4 km), convection‐permitting simulations are robust in capturing the observed eastward propagation of MJO precipitation and surface westerly winds. Increasing resolution improves the boundary layer structure and convective systems, which results in more realistic vertical structures of wind and moisture throughout the troposphere, as well as an improved eastward propagation of the MJO.

Keywords

• Madden‐Julian Oscillation
• boundary layer
• convective moist physics
• coupled modeling
• model resolution