Stratospheric ozone-climate interactions in idealized DECK experiments from CMIP6

Rising greenhouse gases (GHG) and decreasing anthropogenic ozone-depleting substances (ODS) are the main drivers of stratospheric climate evolution in the 21st century. However, our understanding of the coupling between stratospheric composition, radiation and dynamics is still a subject to many uncertainties, partly because of the simplified representation of ozone in many current climate models. In our work, we study stratospheric ozone-climate interactions using idealized CMIP6 DECK experiments (pre-industrial control, abrupt quadrupling of CO2, and 1 % yr−1 CO2 increase). This set up provides longer time-series and stronger GHG forcing than in the historical period. The 6th phase of CMIP has a larger number of participating models with interactive chemistry (“CHEM”) to be contrasted against the models where it is prescribed (“NOCHEM”) than in previous generations of CMIP models. Our findings show that CMIP6 models broadly exhibit a similar ozone response to CO2 with increased ozone in the upper stratosphere (US), driven mostly by rapid adjustments (chemistry), and slow transport-driven decrease in the tropical lower stratosphere (LS), and increase in the extratropical LS. The total column ozone response is small in the tropics and positive at high latitudes, with large inter-model discrepancy, possibly arising from model biases in polar vortex dynamics. We also quantify, for the first time, the radiative and dynamical impacts of ozone and quantify their inter-model uncertainty, by means of radiative transfer calculations and careful comparison of chem vs nochem models. First, we find that CHEM models are colder than NOCHEM models in the UTLS region, consistent with the ozone changes in these regions. Second, we find that the large-scale circulation response is systematically different in CHEM and NOCHEM. Lastly, climate sensitivity tends to be lower in CHEM than NOCHEM models, although the uncertainty across models is large and processes that are not tied to ozone cannot be ruled out. Taken together, our work demonstrates that ozone changes can potentially modulate the modeled response to elevated CO2 levels, stressing the importance of interactive chemistry in the future generation of models, in order to correctly simulate the coupling between chemistry, radiative and dynamical processes under climate change.