Abstract. The Brewer–Dobson Circulation (BDC) transports chemical tracers from the well-mixed tropical troposphere to the polar stratosphere, with many important implications for climate, chemistry, ozone distribution and recovery. Since the photochemical losses of nitrous oxide (N2O) are well-known, model differences in its rate of change are due to transport processes that can be separated in the mean residual advection and the isentropic mixing terms in the Transformed Eulerian Mean (TEM) framework. Here the climatological impact of the stratospheric BDC on the long-lived tracer N2O is evaluated through a comparison of its TEM budget in the Whole Atmosphere Community Climate Model (WACCM), a chemical reanalysis of the Aura Microwave Limb Sounder version 2 (BRAM2) and in a Chemistry-Transport Model (CTM) driven by four modern reanalyses (ERA-Interim, JRA-55, MERRA and MERRA2). The effects of stratospheric transport on the N2O rate of change, as depicted in this study, have not been compared across this variety of datasets and never investigated in a chemical reanalysis. We focus on the seasonal means and climatological annual cycles of the two main contributions to the N2O TEM budget: the vertical residual advection and the horizontal mixing terms. The N2O mixing ratio in the CTM experiments has a spread of approximately ~ 20 % in the middle stratosphere, reflecting the large diversity in the mean Age of Air obtained with the same experiments. In all datasets the TEM budget is well-closed and the agreement between the vertical advection terms is qualitatively very good in the Northern Hemisphere, and good in the Southern Hemisphere except above the Antarctic region. The datasets do not agree as well with respect to the horizontal mixing term, especially in the Northern Hemisphere where horizontal mixing has a smaller contribution in WACCM than in the reanalyses. WACCM is investigated through three model realizations and a sensitivity test where gravity waves are forced differently in the Southern Hemisphere. The internal variability of the horizontal mixing in WACCM is large in the polar regions, and comparable to the differences between the dynamical reanalyses. The sensitivity test has a relatively small impact on the horizontal mixing term, but significantly changes the vertical advection term and produces a less realistic N2O annual cycle above the Antarctic. In this region, all reanalyses show a large wintertime N2O decrease, which is mainly due to horizontal mixing. This is not seen with WACCM, where the horizontal mixing term barely contributes to the TEM budget. While we must use caution in the interpretation of the differences in this region, where the reanalyses show large residuals of the TEM budget, they could be due to the fact that the polar jet is stronger and not tilted equatorward in WACCM compared with the reanalyses. We also compare the inter-annual variability in the horizontal mixing and the vertical advection terms. As expected, the horizontal mixing term presents a large variability during austral fall and boreal winter in the polar regions. In the Tropics, the inter-annual variability of the vertical advection term is much smaller in WACCM and JRA-55 than in the other experiments. The large residual in the reanalyses and the disagreement between WACCM and the reanalyses in the Antarctic region highlight the need for further investigations on the modeling of transport in this region of the stratosphere.
On the Role of the Atmospheric Energy Transport in 2 × CO2–Induced Polar Amplification in CESM1
