THE INITIAL WIND SCHEME OF THE HEMISPHERIC SPECTRAL MODEL AND ITS EFFECT ON THE DATA ASSIMILATION

A series of data assimilation and forecast test have been carried out with a hemispheric spectralmodel(T42L9H).It is found that the numerical scheme for determining hemispheric initial wind isimportant to data assimilation and forecast.An inappropriate scheme may cause computationalsources of divergence near the equator,which are responsible for the spurious strong precipitationand corresponding latent heat release.Obviously,this problem differs from either thehemispheric/global domain effect or the tropical data effect pointed by Dalley et al.(1981).Basedon the previous studies,the new scheme of divergence and vorticity correction is presented,andthe difference with other schemes and its effects on the data assimilation are discussed against thecontrol test.Preliminary tests have shown that the new divergence correction scheme proposed in thispaper may be a preferable choice to overcome the initial computational errors in the hemisphericdata assimilation.

THE INITIAL WIND SCHEME OF THE HEMISPHERIC SPECTRAL MODEL AND ITS EFFECT ON THE DATA ASSIMILATION

A series of data assimilation and forecast test have been carried out with a hemispheric spectral model(T42L9H).It is found that the numerical scheme for determining hemispheric initial wind is important to data assimilation and forecast.An inappropriate scheme may cause computational sources of divergence near the equator,which are responsible for the spurious strong precipitation and corresponding latent heat release.Obviously,this problem differs from either the hemispheric/global domain effect or the tropical data effect pointed by Dalley et al.(1981).Based on the previous studies,the new scheme of divergence and vorticity correction is presented,and the difference with other schemes and its effects on the data assimilation are discussed against the control test. Preliminary tests have shown that the new divergence correction scheme proposed in this paper may be a preferable choice to overcome the initial computational errors in the hemispheric data assimilation.