1Climatic Research Unit, School of Environmental Sciences, University of East Anglia, UK
2Canadian Centre for Climate Modelling and Analysis, Environment Canada, Canada
3National Institute of Water and Atmospheric Research, Lauder, New Zealand
4NOAA Aeronomy Laboratory, Boulder, USA
5Institute for Atmospheric and Climate Science, ETH Zürich, Switzerland
6Physical-Meteorological Observatory/World Radiation Center, Davos, Switzerland
Abstract. Stratospheric ozone recovery is expected to drive pronounced trends in atmospheric temperature and circulation from the stratosphere to the troposphere in the 21st century, but coupled chemistry-climate models (CCMs) vary widely in their predictions of future ozone evolution. In order to assess which models might be expected to better simulate future ozone evaluation, we assess the ability of twelve CCMs to simulate observed ozone climatology and trends and rank the models according to their errors averaged across the individual diagnostics chosen. According to our analysis no one model performs better than the others in all the diagnostics; however, combining errors in individual diagnostics into one metric of model performance allows us to objectively rank the models. The multi-model average shows better overall agreement with the observations than any individual model. Based on this analysis we conclude that the multi-model average ozone projection presents the best estimate of future ozone evolution. Our results also demonstrate a sensitivity of the analysis to the choice of reference data set for vertical ozone distribution over the Antarctic, highlighting the constraints that large observational uncertainty imposes on such model verification.