Conclusion

We have tried to test the effectiveness of ensemble experiment in examining the initial development of the tropical atmospheric response to the introduction of the warm SST area.

It is verified that the ensemble experiment is effective in extracting the planetary scale responses against both the random noise produced by the small scale convective activities and the effects caused by intraseasonal variabilities embedded before the switch-on of the warm SST area.

In order to figure out the initial development of the response which has the characteristics consistent with that observed by the long term mean fields obtained by Hosaka et al. (1998), the number of runs to be averaged is at least 32. We recommend to use at least 64 runs to obtain ensemble mean to have a fairly good quality of data for the purpose of investigation of the development of the east-west asymmetry which is caused by the introduction of the warm SST area.