A new coupled chemistry-climate model for the stratosphere: The importance of coupling for future O-3-climate predictions

We have created a new interactive model for coupled chemistry-climate studies of the stratosphere. The model combines the detailed stratospheric chemistry modules developed and tested in the SLIMCAT/ TOMCAT off-line chemical transport models (CTM) with a version of the Met Office Unified Model (UM). The resulting chemistry-climate model (CCM), called UMCHEM, has a detailed description of stratospheric gas-phase and heterogeneous chemistry. The chemical fields Of O-3, N2O, CH4 and H2O are used interactively in the radiative heating calculation. We present results from a series of 10-year 'time-slice' experiments for 2000 and 2050 conditions. The UMCHEM model performs well in reproducing basic features of the stratosphere. The distribution of long-lived tracers and 'age of air' compare well with observations. For O-3, the model tends to underestimate the stratospheric column at high latitudes by similar to20 DU. This is due to an underestimate of poleward transport in the mid-low stratosphere. The UMCHEM reproduces well the seasonal cycle in monthly mean column O-3 at mid-high latitudes, though the variability is slightly smaller than observations and smaller than in the SLIMCAT CTM. Other comparisons with the CTM, which has an identical chemistry scheme, show differences resulting from the models' meteorologies. For example, while the CTM reproduces the observed NOy versus N2O correlation, the UMCHEM overestimates the slope by about a factor of 2. Including full chemistry in the UM causes important differences in the model's meteorology. As the zonal mean ozone climatology used in the UM is larger than that calculated in UMCHEM, with a maximum difference of 4 ppmv in the upper stratosphere, the UMCHEM temperature is about 4 K lower in the Antarctic lower stratosphere and 1-6 K higher in the upper stratosphere. The age of air is less in the basic UM by about 1-3 months in the lower stratosphere but slightly greater in the upper stratosphere. Coupling of the more realistic stratospheric chemistry water vapour warms the model stratosphere by similar to 1-2 K. This water vapour coupling also results in a decrease in ozone with a maximum difference of about 250 ppbv in the tropical and southern high-latitude upper stratosphere, while in the tropical lower stratosphere ozone concentrations are increased by up to 30 ppbv. For 2050 conditions, the model produces a column 03 5% higher than present-day values in the tropics, about 15% higher in the Arctic winter/spring and up to 90% higher in the much smaller Antarctic O-3 hole. This large O-3 increase more than offsets the effect of cooling in the Antarctic late spring induced by greenhouse gases.