Why do lower latitude jets shift further poleward in response to heating of the tropical stratosphere?

A meridional shift of the tropospheric mid-latitude jets is the dominant response of the mid-latitude circulation to many climate forcings such as increasing greenhouse gases, Antarctic ozone depletion and Solar variability. It has recently become apparent that the magnitude of such annular mode-like responses to forcings as well as the timescale of natural annular mode variability can be highly dependent on model specification. There is a tendency for models with lower latitude climatological jets to exhibit larger mid-latitude jet shifts and longer timescale annular mode variability. This is true of comprehensive GCMs, chemistry climate models and simplified GCMs.

Here we make use of simulations with a simplified GCM to understand why the annular mode-like response to a forcing depends on the structure of the climatological jet for one particular forcing case: heating of the tropical stratosphere. It is found that lower latitude jets exhibit a much larger response to this forcing than higher latitude jets. The reason for this dependence is a difference in the strength of the feedback between the eddies and the mean flow. It is proposed that the reason for this difference in feedback strength lies in the coherency of the behaviour of eddies of different phase speeds. When the mid-latitude jet exists at a higher latitude a much wider latitudinal range of critical layers are present. The result is a lack of coherence of the behaviour of the different phase speeds with low phase speeds actually acting to damp the effect of higher phase speed eddies resulting in an overall weaker feedback onto the zonal mean flow anomalies. In contrast, when the mid-latitude jet exists at lower latitudes, there is a much smaller latitudinal range of the critical layers and therefore a much more latitudinally coherent behaviour of the different phase speed eddies such that all phase speeds act together to provide a strong feedback onto the zonal mean flow anomalies.