Relationship of loss, mean age of air and the distribution of CFCs to stratospheric circulation and implications for atmospheric lifetimes

Projections of the recovery of the ozone layer are made with global atmospheric models using a specified time series of mixing ratios of ozone depleting substances (ODSs) at the lower boundary. This time series is calculated using atmospheric mixing ratio observations, emission rates, and an estimate for the atmospheric lifetime. ODS destruction and simulated atmospheric-lifetime vary among models because they depend on the simulated stratospheric transport and mixing. We investigate the balance between the annual change in ODS burden, its atmospheric loss, and the annual ODS input to the atmosphere using several models. Some models produce realistic distributions for the mean age of air and some do not. Back trajectory calculations relate the fractional release (one minus the amount of ODS at a location relative to its stratospheric entry value) to the mean age through the age spectrum, showing that, for the individual spectrum elements, the maximum altitude and loss increase with age. Models with faster circulations produce young'' distributions for the age of air and fail to reproduce the observed relationship between the mean age of air and the fractional release. Models with realistic mean age of air reproduce the observed relationship. These models yield a lifetime for CFCl3 of similar to 56 years, longer than the 45 year lifetime currently used to project future mixing ratios. Use of flux boundary conditions in assessment models would have several advantages, including consistency between the ODS evolution and simulated loss even if the simulated residual circulation changes due to climate change.