Two adaptive radiative transfer schemes for numerical weather prediction models

[摘要] Radiative transfer calculations in atmospheric models are computationallyexpensive, even if based on simplifications such as the δ-two-streamapproximation. In most weather prediction models these parameterisationschemes are therefore called infrequently, accepting additional model errordue to the persistence assumption between calls. This paper presents twoso-called adaptive parameterisation schemes for radiative transfer in alimited area model: A perturbation scheme that exploits temporalcorrelations and a local-search scheme that mainly takes advantage ofspatial correlations. Utilising these correlations and with similarcomputational resources, the schemes are able to predict the surface netradiative fluxes more accurately than a scheme based on the persistenceassumption. An important property of these adaptive schemes is that theiraccuracy does not decrease much in case of strong reductions in the numberof calls to the δ-two-stream scheme. It is hypothesised that thecore idea can also be employed in parameterisation schemes for otherprocesses and in other dynamical models.

[学科分类] 大气科学