As a first step towards such a goal, we have developed a fully parallelized code for a coupled ocean, atmosphere and sea ice model designed to be run on the Earth Simulator. The three major components of our model: AfES, OfES and SIfES have been developed originally from CCSR/NIES AGCM, MOM3 and Hibler's sea ice model, respectively, and each of them has been tuned up to attain high computational performance on the machine. The high performance of our code enables us to execute several high-resolution coupled simulations covering the time scale of centennial order in a tolerable research period.
We have started to investigate the impact of the momentum, heat and water flux exchange on changing both atmospheric and oceanic circulations in response to the increase of a greenhouse gas forcing. Our experiments are basically composed of two runs, one is a control experiment and the other is a CO2 doubling one in which the concentration of CO2 is to be increased with the compound ratio of 1% per year until it reaches twice as much as its control value in 70 years and then it is fixed at this concentration level thereafter. The employed resolutions of our eddy-permitting oceanic and atmospheric models are as follows, AGCM : Horizontal resolution of T106 and there are 24 vertical sigma levels. OGCM : Horizontal resolution of about 1.0 deg and it has 37 vertical levels. A focal point of our study is the induced change of thermohaline circulations and of atmospheric water vapor transport associated with dominant flow patterns reflected, say, in atmospheric teleconnections in response to a given greenhouse gas forcing. We have not completed the analyses yet, however, promising features of the world oceans and atmosphere already began to emerge in simulated fields we have looked into. Sea ice variability was also reproduced with reasonable amplitude. This seems to suggest a possibility that further extensions of the present study would lead to an improvement in numerical studies on the global warming related climate variations. We will also discuss a possible spin-down mechanism of ocean circulations and heat redistribution in a simulated globally warmed condition with a longer extended integration after the CO2 doubling period.