The first priority of this subgroup in the current fiscal year was removal of the remarkable low-temperature bias seen in the up troposhere-lower stratosphere in the general circulation model (CCSR/NIES/FRSGC AGCM) currently shared with a symbiosis first division title (Sumi group). In order to simulate actually the temperature of the high latitude lower stratosphere of spring, and tropical circulation from the winter which an ozone hole produces in addition to it, parameterization of gravity waves other than the geographical feature origin was introduced.

The low-temperature bias of AGCM existed until it resulted from the beginning now when this model was developed, and it had become a serious obstacle simulating correctly the substance exchange between the stratosphere and the troposhere, especially the water vapour content in the stratosphere. Into the radiation group of CCSR, improvement of the program used for calculation of radiation balance is performed, and it was able to test by including a new radiation code in AGCM.

Fig. 13 shows the value which pulled the observed value from the beltlike mean temperature of a Northern Hemisphere summer average by the result of the experiment which used the new radiation code (the both ends of a color -- +-10 -- K is expressed). The remarkable low-temperature bias (Fig. 14) looked at by the experiment using the conventional radiation code is improved sharply. If it will go improvement in piles further from now on, also in case the simulation which includes a chemical process in the future will be performed, it is expected that distribution of steam and a substance important for global warming called ozone will be reproduced better.

The thing of Hines (1997) which has spread most was chosen in introduction of non-orographic gravity wave resistance parameterization. It tried to derive the RMS amplitude of the gravity wave spread at least to all directions with the necessity of inputting into this parameterization from the simulation result by T106L250 AGCM, since observation of the gravity wave of the lower stratosphere is restricted. The perpendicular resolution of this model is 300m, and can express a small-scale gravity wave to an unresolvable vertical direction in the usual AGCM. Although it is the present testing phase, it has succeeded in reappearance of the realistic equator QBO using gravity wave sauce including geography change and a seasonal variation. It is necessary to aim at the improvement in reproducibility of high latitude circulation from now on.