Positive and Negative Eddy Feedbacks Acting on Midlatitude Jet Variability

The variability of midlatitude jets is investigated in a long-term integration of a dry three-level quasigeostrophic model on the sphere. As for most observed jets, the leading EOF of the zonal-mean wind corresponds to latitudinal shifts of the jet, and the second EOF to pulses of the jet speed. The first principal component (PC1) is also more persistent than the second one (PC2); this longer persistence arises from different eddy feedbacks both in the short-term, i.e within a few days following the peak of the PCs, and in the long-term.
The short-term eddy feedbacks come from two distinct mechanisms. First, a planetary waveguide effect acts as a negative feedback on both PCs. The positive phases of PC1 and PC2, which correspond to poleward-shifted and accelerated jets respectively, are first driven then canceled by planetary waves reflecting on the equatorward flank of the jet. A similar process occurs for the negative phases when planetary waves reflect on the poleward flank of the jet. Second, synoptic waves also exert a short-term negative feedback on PC2: when the jet accelerates, the enhanced meridional wind shear increases the barotropic sink of eddy energy and depletes it very rapidly, therefore preventing synoptic eddies from maintaining the accelerated jet.
Finally, at lags longer than their typical timescale, synoptic eddies drive a positive feedback on PC1 only. This feedback can be explained by a baroclinic mechanism in which the jet shift modifies the baroclinicity, causing first eddy heat flux then momentum convergence anomalies. This feedback is absent for PC2, despite some changes in the baroclinicity.