Abstract
Using outputs from the “twentieth-century climate in coupled models” (20c3m) control run of the Coupled Model Intercomparison Project, phase 3 (CMIP3), coupled GCMs, the authors have examined how seasonal variations of the Seychelles Dome (SD) are simulated in the southwestern Indian Ocean. The observed SD shows a dominant semiannual signal due to the semiannual variation in the local Ekman upwelling resulting from a combination of two terms related to the wind stress curl and the zonal wind stress. However, all models fail to reproduce this important mechanism. In particular, the latter contribution—that determined by the seasonal variation of the zonal wind stress associated with the Indian monsoon—is not well simulated. Successful models need to reproduce the asymmetric nature of the monsoon: a shorter and stronger summer monsoon and a longer and weaker winter monsoon. Possible remedies for the model bias are also discussed.
Corresponding author address: Takaaki Yokoi, Department of Earth and Planetary Science, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Email: yokoaki@eps.s.u-tokyo.ac.jp