Separating chemistry and transport effects in two-dimensional models

Type: Journal Article

Venue: Journal of Geophysical Research


Weisenstein, D. K., J. Eluszkiewicz, M. K. W. Ko, C. J. Scott, C. H. Jackman, E. L. Fleming, D. B. Considine, D. E. Kinnison, P. S. Connell, and D. A. Rotman (2004), Separating chemistry and transport effects in two-dimensional models, J. Geophys. Res., 109, D18310, doi:10.1029/2004JD004744.

Resource Link:

Representation of transport in numerical models is known to be a major uncertainty in modeling of the atmosphere. Models also differ in their treatment of gas phase and heterogeneous chemistry. This paper will describe a quantitative approach to diagnosing the source of intermodel differences in ozone assessment calculations. Our approach is applied to diagnosing the differences between two-dimensional (2-D) models from Atmospheric and Environmental Research, the NASA Goddard Space Flight Center, and the Lawrence Livermore National Laboratory. Surprisingly, we find that differences due to chemical formulation are often as large as those due to transport, despite the fact that all models use the same set of reaction rate coefficients. These differences are particularly large when polar stratospheric cloud (PSC) processes are included in the models, though differences due to photolysis rates and details of the sulfate chemistry are also apparent. Perturbation calculations for a scenario including supersonic commercial aircraft operating in the 2015 stratosphere reveal that differences in the accumulation of H2O and NO y emitted by aircraft are due almost entirely to transport, while differences in ozone due to chemical formulation are evident in the lower stratosphere even without differences in H2O and NO y and without PSCs. By demonstrating a capability of separating transport and chemical differences, it is hoped that the results described in this paper will stimulate analogous studies with other models and will thus lead to a deeper understanding of intermodel similarities and differences, along with a means to quantify uncertainties in model predictions of atmospheric response to perturbations.