| In their idealized model, Holton and Mass (1976) saw short-term variability in the stratospheric jet that was induced by a feedback between the mean flow and waves emanating from the troposphere. Several researchers have found longer term stratospheric variability with a tropospheric source in numerical simulations of the stratosphere-troposphere system, variability that could not be explained as a simple wave-mean flow interaction. What was causing a long term signal in the troposphere, where weather systems come and go quickly, was a mystery.; We have run a dry, nonlinear, primitive equation spectral model with no externally forced variability and with a realistic time mean state, and we have observed low frequency variability (LFV) in the stratosphere with timescales on the order of hundreds of days. Time lagged correlations have revealed that this variability is linked to LFV in the emission of longwaves from the troposphere. A set of linear model experiments is performed to determine the source of the stratospheric LFV. One set of runs reveal the lowest levels of the model troposphere as the source of most of the relevant forcing. A second set of runs forced with nonlinear terms has shown that the nonlinear interaction among shortwave, high-frequency eddy thermal anomalies in the troposphere has a "beating" effect which emits vertically propagating low-frequency longwaves. We also see that the eddies act in such a way as to offset the effects of linear temperature advection, allowing the thermal eddies to persist for long periods. |
