A Study of Cumulus Parameterization in a Global Circulation Model.
Final rept. 8 Apr 82-8 Jun 85,
ILLINOIS UNIV AT URBANA DEPT OF ATMOSPHERIC SCIENCES
Pagination or Media Count:
Objectives of this research are 1 to critically evaluate the Kuo and Arakawa Schubert A-S cumulus parameterization schemes for numerical weather prediction and 2 to improve these parameterization schemes to improve precipitation forecasts on the global scale. Major improvements to the Kuo scheme include prediction of cloud top height and incorporation of the effect of entrainment on cloud temperature and mixing ratio profiles. This enables the Kuo scheme to parameterize shallow to medium clouds as well as deep clouds. Tests of the Kuo scheme using a semi-prognostic approach and a cloud cluster model indicate that the improved version verifies better with observation during weaker convective periods. Experimental predictions were made and the results clearly demonstrated the ability of the AFGL model to predict large-scale stratiform precipitation. With incorporation of the modified Kuo scheme, the area of convective precipitation can also be well predicted. However, the predicted convective precipitation area is generally broader and the amount of rainfall smaller than observed which may be attributed to the resolution of the AFGL model in that it cannot resolve the narrow band of the cold front, ther major mesoscale rain-producing system. Implementation of the A-S scheme in the AFGL model also produced a reasonable distribution of convective precipitation but the precipitation area is more concentrated and sometimes is produced in an observed clear area. This result may be caused by the current implementation of the A-S scheme in the AFGL model, in which the cloud base is assumed to be 500 m above the ground surface.
- Theoretical Mathematics