Tropical Intraseasonal Variability Response to Zonally Asymmetric Forcing in an Idealized Moist GCM

The tropical intraseasonal variability in an idealized moist general circulation model (GCM) is investigated. The model has interactive water vapor and realistic radiative transfer, but no parameterization of clouds. Moist convection is parameterized using a Simplified Betts-Miller (SBM) convection scheme. The model is run in an aquaplanet configuration with an underlying slab ocean layer. In a zonally symmetric state, the tropical variability is dominated by westward-propagating convectively-coupled Rossby waves. A wavenumber-1 surface temperature pattern is introduced through the application of a prescribed ocean heat flux to create an equatorial warm pool. This change leads to the development of a slow, eastward-propagating mode of tropical variability in the warm pool sector which resembles the observed Madden Julian Oscillation (MJO). High frequency Kelvin waves appear alongside the MJO mode when the strength of the asymmetric heating is made sufficiently strong that the tropics of the cold hemisphere become climatologically dry. Due to the lack of clouds, the magnitude of intraseasonal surface temperature anomalies most resemble those of the observed MJO when a shallow slab depth (1 m) is used, but the model retains the non-monotonic dependence of MJO amplitude on slab depth seen in previous modeling studies. Sensitivity experiments which perturb the parameters of the model's convection and boundary layer schemes provide further insight into this intraseasonal mode. Analysis of the mode's column moist static energy (CMSE) budget shows that longwave radiative heating has only a small influence on the maintenance of CMSE anomalies relative to vertical advection. The CMSE budget also highlights the time-dependent nature of the gross moist stability (GMS), and shows that CMSE is exported from precipitating regions when convection is at a maximum, suggesting that the moisture mode instability paradigm may not be active in the model. Synergies between the simulation results and linear theories of the MJO are discussed as well, with a focus on the role of the GMS and the role of the weak temperature gradient approximation.