Mahlman, J. D., H. Levy II, and W. J. Moxim, 1986: Three-dimensional simulations of stratospheric N2O: predictions for other trace constituents. Journal of Geophysical Research, 91 (D2), 2687-2707.

Abstract: The Geophysical Fluid Dynamics Laboratory (GFDL) three-dimensional general circulation/tracer model has been used to investigate the stratospheric behavior of N2O under a range of photodestruction hypotheses. A comparison of observations with these simulations shows that the atmospheric N2O lifetime lies between 100 and 130 years. For the three experiments conducted, it was found that the model-derived global one-dimensional eddy diffusion coefficients Kz for one experiment are appropriate for the other two experiments as well. In addition, the meridional slopes of N2O mixing ratio isolines are virtually identical in the lower stratosphere for all three experiments. The generality of these two results was explored with a simple "two-slab" model. In this model the equilibrium meridional slopes of trace gas isolines and Kz values are solved directly. The model predicts that long-lived gases with weak photodestruction rates should have similar meridional slopes, but the effect of faster destruction is to flatten the meridional slopes. The simple model also predicts that Kz depends upon chemical processes through a direct dependence upon the meridional slope for a given gas as well as upon the intensity with which upward propagating tropospheric disturbances force the stratospheric zonal winds. The three N 2O experiments have been compared against detailed observational analyses. These analyses show that the model meridional N2O slopes are too flat by about 30%. The simple two-slab model indicates that this results from a somewhat weak forcing of the model stratospheric zonal winds. A comparision of the temporal variability of model N2O against the "Delta" statistics of Ehhalt et al. (1983) shows good agreement. Another simple theoretical model is proposed that shows why Delta statistics are so useful and predicts the circumstances under which different destruction chemistries should lead to different Delta statistics. These results have allowed a very general extrapolation of the three N2O numerical experiments to predicted structure for a wide class of long-lived trace gases. Specifically, the supporting theoretical developments allow predictions for the effect of chemistry on the global (one-dimensional) behavior, meridional-height (two-dimensional) structure and local temporal variability. Finally, some examples of transient behavior are presented through model time series at points corresponding to available measurements. These time series, with the support of horizontal N 2O charts, show complex behavior, including pronounced seasonal cycles, transport-produced N2O "inversions", and detailed meridional transport events associated with transient stratospheric disturbances.