The global stratospheric circulation, termed the Brewer-Dobson circulation (BDC), controls the transport of trace gas species (e.g. ozone, water vapor – link to “water vapor”) which, in turn, can influence climate through radiative effects. Furthermore, variations in the stratospheric circulation may influence regional climate and weather patterns. Therefore, reliable climate predictions require an accurate representation of the stratospheric circulation in atmospheric models. In the tropics, the stratospheric circulation is characterized by upwelling of air masses which can transport polluted tropospheric air deep into the stratosphere. In the stratosphere, air masses are transported horizontally towards high, polar latitudes, where they sink back into the troposphere. The BDC is a mechanically forced circulation, with the driving force provided by atmospheric waves (e.g. planetary-scale Rossby waves, small-scale gravity waves) which propagate upwards from the troposphere and break at upper (so-called “critical”) levels in the stratosphere (analogous to ocean waves at the coast). When breaking, these waves dissipate, transfer their momentum to the mean flow and drive the meridional circulation of the BDC.

Overall, the BDC is a slow circulation and typical transit times for air masses during their circuit through the stratosphere are on the order of a few years. As related circulation velocities are very low, typically below 1mm/s, they cannot be directly observed and sophisticated diagnostics are needed to deduce observational constraints. The most commonly used diagnostic is stratospheric age of air, the transit time for an air mass through the stratosphere. Due to the presence of mixing processes on various scales, a macroscopic air parcel is characterized by a variety of transport pathways and related transit time scales through the stratosphere, and hence by a transit time distribution (the “age spectrum”). The mean transit time of an air parcel, the first moment of its age spectrum, is termed the mean age of air, and can be calculated from the mixing ratios of trace gas species which are linearly increasing in the troposphere and have no sources and sinks in the stratosphere. As these necessary assumptions do only hold approximately for real, observable trace species, the deduction of observational constraints for the stratospheric circulation remains challenging. Particularly long-term changes in the stratospheric circulation are not well understood, and climate models simulate an accelerating circulation which is not supported by observations. Additional variations in the circulation’s wave driving related to natural variability (e.g. QBO, ENSO, stratospheric aerosol – link) further complicate the deduction of long-term, anthropogenically-forced trends.