The UIUC three‐dimensional stratospheric chemical transport model: Description and evaluation of the simulated source gases and ozone

Publisher Website
Google Scholar
Abstract
A new University of Illinois at Urbana‐Champaign (UIUC) three‐dimensional stratospheric chemical transport model is presented. The model consists of (1) a hybrid transport routine; (2) a chemical routine that includes the principal gas‐phase and heterogeneous reactions; and (3) the circulation, temperature, and tropospheric humidity fields acquired from the UIUC 24‐layer general circulation model. The model is applied to study chemistry‐transport processes in the stratosphere. The results of an 8‐year steady state simulation with 1995 boundary conditions are analyzed, and the distributions of three source gases, methane, water vapor, and nitrous oxide, and ozone are evaluated in comparison with appropriate UARS measurements. The comparison shows that the model is able to reproduce the main features of the distributions of long‐lived species obtained from satellite measurements, namely, the location of the tropical extremes in the summer hemisphere, the high horizontal gradient in the subtropics, the winter midlatitude mixing zone, and the high‐latitude minimum (or maximum) regions. The model also well captures the observed features of the ozone distribution in the stratosphere, including the intensity and location of the tropical maximum, the depletion in the lower stratosphere of the Southern Hemisphere, and the seasonal variations in the middle stratosphere. The magnitudes of the mixing ratios of the long‐lived species are found to be in reasonable agreement with the observed values, although the model overestimates the N2O mixing ratio over high latitudes and slightly underestimates the H2O mixing ratio in the stratosphere. Also, the simulated ozone mixing ratio is overestimated in the middle stratosphere and underestimated in the upper stratosphere by 5–15%. However, analysis and comparison of the simulated and observed species distributions and tracer‐to‐tracer correlations show a very good overall performance of the model.