Are SST Anomaly Spatial Patterns or Extrema More Important for MJO Propagation in Coupled Simulations?
Previous work by the authors (DeMott et al. 2019) demonstrated that ocean feedbacks in coupled general circulation models (CGCMs) produce tropical mean moisture distributions that are more peaked on the Equator than in atmosphere-only GCMs (AGCMs) forced with CGCM monthly mean sea surface temperatures (SSTs). The sharper equatorward moisture gradient with ocean coupling is responsible for enhanced column moistening by meridional moisture advection east of MJO convection, and for improved MJO eastward propagation in CGCMs compared to their AGCM counterparts.
In this study, we take a closer look at how sub-monthly SST perturbations are communicated to mean state moisture patterns and the effects of each on MJO propagation. We perform two AGCM simulations where grid-point SST anomalies from the CGCM are randomized in time, added to the slowly varying background SST, then prescribed to the AGCM. In the first experiment, the spatial patterns of SST anomalies are randomized, which retains spatial coherence but degrades temporal coherence; in the second experiment, individual grid-point SST anomalies are randomized, which degrades spatial and temporal coherence. MJO propagation is somewhat degraded in the first experiment, and completely degraded in the second, indicating a non-trivial role for the SST anomaly pattern in MJO propagation. Mean state moisture changes and the moist static energy budget are analyzed to further investigate mechanisms of ocean coupled feedbacks to the MJO in these two simulations.