Prescribing stratospheric chemistry overestimates southern hemisphere climate change during austral spring in response to quadrupled CO2

The interaction of stratospheric chemistry with a changing climate from an abrupt CO2 quadrupling is assessed using the coupled atmosphere-ocean Goddard Earth Observing System Chemistry-Climate Model (GEOSCCM). Two abrupt 4 x CO2 experiments were performed, one with interactive stratospheric chemistry and the other with a prescribed stratospheric chemistry that does not simulate stratospheric ozone response to 4 x CO2. The interactive and prescribed chemistry experiments simulate similar global mean surface temperature change. Nevertheless, interactive chemistry is critical to capture the Southern Hemisphere tropospheric midlatitude jet response to 4 x CO2. When stratospheric ozone response to 4 x CO2 is neglected, GEOSCCM overestimates Southern Hemisphere tropospheric circulation change. This stratospheric chemistry-induced climate impact has large seasonal variability. During the austral spring season September-October-November (SON), prescribed chemistry yields a stronger poleward shift and intensification of the Southern Hemisphere midlatitude tropospheric jet, surface wind stress, and the Southern Ocean meridional overturning circulation than occurs with interactive chemistry. In other seasons interactive and prescribed chemistry have similar effects on the Southern Hemisphere circulation. The seasonality of stratospheric chemistry-induced climate impact is related to the seasonality of Antarctic lower stratospheric ozone response to 4 x CO2. In contrast to this stratospheric ozone response to 4 x CO2, stratospheric ozone recovery from decline of the ozone depleting substances has its largest impact on the Southern Hemisphere tropospheric circulation in austral summer (December-January-February), but no effects in SON. It is found that the different seasonality for these two stratospheric ozone layer change scenarios is related to the different seasonality of tropopause meridional temperature gradient response.