Choosing meteorological input for the global modeling initiative assessment of high-speed aircraft

Citation
Ar. Douglass et al., Choosing meteorological input for the global modeling initiative assessment of high-speed aircraft, J GEO RES-A, 104(D22), 1999, pp. 27545-27564
Citations number
55
Categorie Soggetti
Earth Sciences
Volume
104
Issue
D22
Year of publication
1999
Pages
27545 - 27564
Database
ISI
SICI code
Abstract
The global modeling initiative (GMI) science team is developing a three-dim ensional chemistry and transport model (CTM) for use in assessment of the a tmospheric effects of aviation. This model must be documented, be validated against observations, use a realistic atmospheric circulation, and contain numerical transport and photochemical modules representing atmospheric pro cesses. The model must retain computational efficiency for multiple scenari os and sensitivity studies. To meet these requirements, a facility model co ncept was developed in which the different components of the CTM: are evalu ated separately. The assessment of the impact on the stratosphere of the ex haust of supersonic aircraft will depend strongly on the meteorological fie lds used by the CTM. Three data sets for the stratosphere were considered: the National Center for Atmospheric Research Community Climate Model (CCM2) , the Goddard Earth Observing System data assimilation system, and the Godd ard Institute for Space Studies general circulation model. Objective criter ia were developed to identify the data set that provides the best represent ation of the stratosphere. Simulations of gases with simple chemical contro l were chosen to test various aspects of model transport. The data sets wer e evaluated and graded on their performance on these tests. The CCM2 meteor ological data set has the highest score and was selected for GMI. This obje ctive model evaluation establishes a physical basis for interpretation of d ifferences between models and observations. Further, the method provides a quantitative basis for defining model errors, for discriminating between di fferent models, and for ready reevaluation of improved models. This will le ad to higher confidence in assessment calculations.