Using transport diagnostics to understand chemistry climate model ozone simulations

S. E. Strahan (1) , A. R. Douglass (2) , R. S. Stolarski (2) , H. Akiyoshi (3) , Slimane Bekki (4) , P. Braesicke (5) , N. Butchart (6) , M. P. Chipperfield (7) , David Cugnet (4) , S. Dhomse (7) , S. M. Frith (8) , A. Gettelman (9) , S. C. Hardiman (6) , D. E. Kinnison (9) , J.-F. Lamarque (9) , E. Mancini (10) , Marion Marchand (4) , M. Michou (11) , Olaf Morgenstern (12) , T. Nakamura (3) , D. Olivié (13) , S. Pawson (2) , G. Pitari (10) , D. A. Plummer (14) , J. A. Pyle (5) , J. F. Scinocca (14) , T. G. Shepherd (15) , K. Shibata (16) , Dan Smale (12) , H. Teyssèdre (11) , W. Tian (7) , Y. Yamashita (3)

1 USRA - Universities Space Research Association [Washington]
2 GSFC - NASA Goddard Space Flight Center
3 NIES - National Institute for Environmental Studies
4 STRATO - LATMOS
5 Centre for Atmospheric Science [Cambridge, UK]
6 MOHC - Met Office Hadley Centre
7 SEE - School of Earth and Environment [Leeds]
8 SSAI - Science Systems and Applications, Inc. [Lanham]
9 NCAR - National Center for Atmospheric Research [Boulder]
10 UNIVAQ - University of L'Aquila [Italy]
11 CNRM - Centre national de recherches météorologiques
12 NIWA - National Institute of Water and Atmospheric Research [Lauder]
13 Department of Geosciences [Oslo]
14 CCCma - Canadian Centre for Climate Modelling and Analysis
15 Department of Physics [Toronto]
16 MRI - Meteorological Research Institute [Tsukuba]

We use observations of N2O and mean age to identify realistic transport in models in order to explain their ozone predictions. The results are applied to 15 chemistry climate models (CCMs) participating in the 2010 World Meteorological Organization ozone assessment. Comparison of the observed and simulated N2O, mean age and their compact correlation identifies models with fast or slow circulations and reveals details of model ascent and tropical isolation. This process-oriented diagnostic is more useful than mean age alone because it identifies models with compensating transport deficiencies that produce fortuitous agreement with mean age. The diagnosed model transport behavior is related to a model's ability to produce realistic lower stratosphere (LS) O3 profiles. Models with the greatest tropical transport problems compare poorly with O3 observations. Models with the most realistic LS transport agree more closely with LS observations and each other. We incorporate the results of the chemistry evaluations in the Stratospheric Processes and their Role in Climate (SPARC) CCMVal Report to explain the range of CCM predictions for the return-to-1980 dates for global (60°S-60°N) and Antarctic column ozone. Antarctic O3 return dates are generally correlated with vortex Cly levels, and vortex Cly is generally correlated with the model's circulation, although model Cl chemistry and conservation problems also have a significant effect on return date. In both regions, models with good LS transport and chemistry produce a smaller range of predictions for the return-to-1980 ozone values. This study suggests that the current range of predicted return dates is unnecessarily broad due to identifiable model deficiencies.

Mots clés: Stratospheric transport

Domaines: Physique Atmosphérique et Océanique [physics.ao-ph]; Climatologie

S. E. Strahan, A. R. Douglass, R. S. Stolarski, H. Akiyoshi, Slimane Bekki, et al.. Using transport diagnostics to understand chemistry climate model ozone simulations. Journal of Geophysical Research: Atmospheres, 2011, 116 (D17), pp.D17302. ⟨10.1029/2010JD015360⟩. ⟨hal-00621937⟩