Mitigating the MJO-Mean State Trade-Off Syndrome with a Parameterization of Convective Organization

Many contemporary Global Climate Models (GCMs) poorly represent the Madden-Julian Oscillation (MJO), the dominant mode of tropical intraseasonal variability. The MJO affects the global weather-climate system, providing the source of predictability on the intraseasonal time scale. The poor representation of the MJO in weather and climate models has been a barrier to a full exploitation of MJO’s predictability in predictions made using the models. Inhibiting too frequent deep, vigorous convection by tweaking parameters such as the fractional entrainment rate in the cumulus parameterizations has shown to enhance MJO variability in GCMs. The MJO amplification, however, typically accompanies a degradation of the mean state. This suggests that there exists a trade-off between the MJO variability and the mean state, which might have prevented modelers from implementing the known methods of enhancing MJO variability.

A recently-developed unified convection scheme (UNICON), when implemented in the Community Atmosphere Model version 5 (CAM5), resolves the MJO-mean state trade-off syndrome. The UNICON explicitly represents the interactions between convective updrafts and boundary layer cold pools through a parameterization of the convective organization. Results of sensitivity experiments, in which a select process in UNICON is disabled, show that the effects of convective organization on the MJO and the mean state are distinct from those of tweaking the fractional entrainment rate. While both methods enhance the MJO variability, increasing the fractional entrainment rate lowers moist stability of the tropical atmosphere while the addition of convective organization effect stabilizes the mean state. When the two effects are combined, CAM5 simulates a much enhanced MJO variability without a degradation in the mean state.