Recently, the structures of the SCCs have been comprehensively described by using historical CTD data. Johnson and Moore (1997) derived various features of the SCCs; the jets' cores shift poleward from west to east, the shift is correlated with a shoaling of the pycnocline and the building pycnostad between the SCCs. They also present a inertial model, which accounts for many features of the SCCs, e.g., the poleward shift, the size and strength of the pycnostad from west to east. Their model is consistent with many observed features, however does not specify how the model jets are forced and maintained.
There are few numerical model research on the SCCs. Kitamura (1986) represented the subsurface countercurrents similar to the SCCs by using a GCM and suggested the eddy activities as generation mechanism. However, the model SCCs are not distinct to the EUC and their maximum speeds are weak to the observed ones. Ishida et al. (1997) described the ocean currents structures derived by a global high-resolution model with 1/4° horizontal resolution and 55 levels in the vertical. They showed successful simulation for the SCCs in their model and suggested the importance of high vertical resolution for equatorial currents simulation because high vertical mode motion dominates there.
In this study, we describe the mean structure and the variability of the SCCs simulated in the model of Ishida et al. (1997) comparing with observed ones and also discuss the importance of the eddy activity related to the Tropical Instability Waves (TIWs) on maintenance mechanism of the model SCCs.
2. Model and Experiment
The model in this study is based on the Modular Ocean Model version 2 (MOM2) (Pacanowski, 1995). The model covers the global ocean from 75｡? to 75｡? with 1/4 degrees horizontal resolution and 55 levels in the vertical with a spacing of 10m at the surface, smoothly stretching to 50m in the intermediate layer and to about 400m by 6000m. High vertical resolution distinguishes this model from the other global high-resolution models (e.g., Semtner and Chervin; 1988, 1992).
The model is driven from the initial state at rest with annual averaged temperature and salinity of Levitus (1982) climatology. The heat and freshwater flux are implemented as a linear restoring of temperature and salinity in the first model level toward Levitus data. The Hellerman and Rosenstein (1983) climatological wind stress is used to force the model ocean. Biharmonic operator with the coefficients of -1 x 1019cm4sec-1 for horizontal dissipation is used for both momentum and tracers. The vertical dissipation is handled through the Pacanowski and Philander (1981) formulation.
The model is forced by the annual averaged climatologies in the first 2 model years, and then forced by the monthly climatologies in 18 years. We use the results in the final model year for analysis. The details of the model parameters and conditions in the numerical experiment are described in Ishida et al. (1997).
3. Mean structures of the SCCs
In this section, we present the annual averaged model results and validate them comparing the observational results of Johnson and Moore (1997).
Meridional sections
The model meridional sections of potential temperature and salinity (Fig. 1), and potential density and eastward velocity (Fig. 2) along the latitudes for western (165｡?), central (155｡?), and eastern (110｡?) tropical Paciflc Ocean, corresponding to the observational sections shown by Johnson and Moore (1997).
