Artificial acceleration of the Brewer-Dobson circulation due to stratospheric cooling

There is robust observational evidence that the troposphere is warming and the stratosphere is cooling in response to the radiative forcing of anthropogenic greenhouse gas (GHG) emissions. Temperature changes directly influence the vertical structure of the atmopshere. Numerous studies have analysed the thermal expansion of the troposphere, in particular the tropopause rise and its interaction with the Brewer-Dobson circulation (BDC). Stratospheric cooling, however, reduces the upward shift of pressure levels with increasing altitude so that it reverses sign at some height, leading to a downward shift of the middle to upper stratosphere. This "stratospheric shrinkage“ effect is a strong and robust feature of climate change and it is well documented through observations. Still, literature on this effect is relatively sparse and its impact on stratospheric dynamics is generally neglected.

In this study, we report and quantify the uncertainty in residual upward velocity (w*) trends that arises from the implicit neglection of stratospheric shrinkage in the data model request for the Chemistry-Climate Model Initiative part 1 (CCMI-1). Tropical w* is often taken as a proxy for diagnosing the BDC strength. In the data request, a constant scale height is assumed for conversion of w* from Pa/s to m/s . However, the scale height significantly decreases over time in the climate simulations as a result of stratospheric shrinkage.

We show that stratospheric cooling enhances the w* trends if the unit conversion is made with constant scale height, which can be misinterpreted as BDC acceleration. We quantify this effect to account for around 20% of the w* trend across the 21st century, consistently among the CCMI-1 climate projection simulations. Past studies that based w* trend analyses on these data therefore made a 20% error. Moreover, we call attention that other dynamical diagnostics are affected by the neglection of stratospheric shrinkage too and also the data requests of other multi-model assessments use the constant scale height assumtion for unit conversion in climate change simulations.