Search for: All records

Abstract We present single‐column gravity wave parameterizations (GWPs) that use machine learning to emulate non‐orographic gravity wave (GW) drag and demonstrate their ability to generalize out‐of‐sample. A set of artificial neural networks (ANNs) are trained to emulate the momentum forcing from a conventional GWP in an idealized climate model, given only one view of the annual cycle and one phase of the Quasi‐Biennial Oscillation (QBO). We investigate the sensitivity of offline and online performance to the choice of input variables and complexity of the ANN. When coupled with the model, moderately complex ANNs accurately generate full cycles of the QBO. When the model is forced with enhanced CO₂, its climate response with the ANN matches that generated with the physics‐based GWP. That ANNs can accurately emulate an existing scheme and generalize to new regimes given limited data suggests the potential for developing GWPs from observational estimates of GW momentum transport. 

Abstract The annular modes of the extratropical atmosphere have received much attention for quantifying variability of the jet streams and storm tracks, despite the fact that the midlatitude circulation itself does not vary uniformly with longitude. While tropical fluctuations in geopotential height have lower amplitude than in the extratropics, they exhibit stronger zonal coherence, or dynamical annularity. A simple index is developed to characterize zonal‐mean anomalies of the tropical circulation. It reveals that anomalies in geopotential height and zonal wind migrate downward from the upper troposphere to the surface on a time scale of about 10 days. These features are distinguishable from known modes of tropical variability, the Madden‐Julian Oscillation in particular. Evidence from reanalysis and idealized model experiments confirms that this downward migration is quite generic and driven by mechanically forced variations in the strength of the Hadley circulation on subseasonal time scales.