The persistence and photochemical decay of springtime total ozone anomalies in CMAM integrated over 35o–80o latitude in both hemispheres is very realistic. The behaviour in the SH is similar for simulations with and without heterogeneous chemistry, i.e. with or without an ozone hole. The behaviour in the NH is realistic despite unrealistically low ozone variability, but is unrealistic for a simulation with very strong vertical diffusion. This shows that the diagnostic from F&S 2003 does not depend on how the springtime anomalies are created or on their magnitude, but reflects the transport and photochemical decay in the model – as one would expect. For the entire extratropical region the seasonality of the long-term ozone trends in CMAM for summer through early autumn can be explained by the persistence of the interannual springtime anomalies, as in the observations. This implies that summertime ozone trends are a result of the springtime trends, without the need to invoke changes in summertime chemistry. By considering the entire extratropical region, transport of ozone-depleted air from the polar into the midlatitude regions following the breakdown of the polar vortex does not affect the relation between springtime and summertime ozone anomalies and long-term trends. However this is not true for the midlatitudes alone if polar ozone variations are sufficiently large compared to midlatitude variations. This is true in the southern hemisphere for both observations and CMAM, where the ozone hole contributes to midlatitude summertime trends, elevating them above what would be expected based on midlatitude springtime trends and leading to a relatively weak seasonality of the midlatitude trends. For CMAM this is also true in the northern hemisphere – and in contrast to the observations – because of the relatively large impact, compared to observations, of the CMAM polar anomalies. This results from an unrealistically small midlatitude ozone variability rather than an unrealistically large polar variability.
Acknowledgements. This work was initiated during an earlier visit by S. Tegtmeier to the University of Toronto and has been supported by funding from the Natural Sciences and Engineering Research Council of Canada, the Canadian Foundation for Climate and Atmospheric Sciences, the Canadian Space Agency, and Environment Canada. The authors are indebted to S. Beagley, A. Jonsson, and J. de Grandpr´e for assistance with the CMAM data, to V. Fioletov for helpful discussion, and to R. Stolarski and S. Frith from NASA for making the merged satellite data set available. Work at AWI was supported by the EC under contract 505390-GOCE-CT-2004 (SCOUT-O3).