A comparative method to evaluate and validate stochastic parametrizations

Hermanson, L., Hoskins, B. J. and Palmer, T. N. (2009) A comparative method to evaluate and validate stochastic parametrizations. Quarterly Journal of the Royal Meteorological Society, 135 (642). pp. 1095-1103. ISSN 1477-870X doi: 10.1002/qj.436

Abstract/Summary

There is a growing interest in using stochastic parametrizations in numerical weather and climate prediction models. Previously, Palmer (2001) outlined the issues that give rise to the need for a stochastic parametrization and the forms such a parametrization could take. In this article a method is presented that uses a comparison between a standard-resolution version and a high-resolution version of the same model to gain information relevant for a stochastic parametrization in that model. A correction term that could be used in a stochastic parametrization is derived from the thermodynamic equations of both models. The origin of the components of this term is discussed. It is found that the component related to unresolved wave-wave interactions is important and can act to compensate for large parametrized tendencies. The correction term is not proportional to the parametrized tendency. Finally, it is explained how the correction term could be used to give information about the shape of the random distribution to be used in a stochastic parametrization. Copyright © 2009 Royal Meteorological Society

Item Type	Article
URI	https://reading-clone.eprints-hosting.org/id/eprint/4637
Item Type	Article
Refereed	Yes
Divisions	Science > School of Mathematical, Physical and Computational Sciences > Department of Meteorology
Uncontrolled Keywords	numerical weather prediction • high resolution • subgrid scale
Publisher	Royal Meteorological Society
Download/View statistics	View download statistics for this item