**4. Summary**

This article examined the impact of solar radiation data and its absorption schemes on ocean model simulation. Both are essential for modeling the upper ocean thermal structure.

Section 2 investigated the discrepancy between observed and OGCM-simulated anomalies in recent SSTs of the tropical Indian Ocean. Observed SSTs indicate warming beginning in the late 1990s, whereas simulated SSTs exhibit cooling over the same period. Examination of surface heat fluxes in the OGCM showed that the simulated SST cooling was caused primarily by a decreasing trend in the reanalyzed solar radiation used as the surface boundary condition. In the atmospheric reanalysis, the decrease in solar radiation was attributed to an increase in cloud cover, deduced from precipitation data, and in part, responsible for the observed local warming of the Indian Ocean SSTs prescribed as the lower boundary condition. Observation-based estimates of precipitation, however, showed no significant increasing trend; thus, no increase in cloud cover was indicated. Caution is necessary when atmospheric reanalysis data are used for surface boundary conditions for OGCMs.

Section 3 examined three absorption schemes for ocean model simulations: (1) a conventional scheme (Paulson & Simpson, 1977), (2) an introduction of varying solar angle (Ishizaki & Yamanaka, 2010), and (3) an introduction of the effect of local heating by chlorophyll-a concentration (Morel & Antoine, 1994) together with the second scheme. Introducing the new scheme resulted in a significant change especially in the equatorial Pacific, where the MLD decreased by about 10 m, and the surface current field showed a divergent flow. Associated with the surface current field, the equatorial upwelling was enhanced and the STC transport intensifies by more than 20 % in the Pacific. These changes in SLR run are explained by a dynamical response of the equatorial Pacific to the change in MLD (Sweeney et al., 2005).

These results indicate that both the solar radiation data and the employed absorption scheme of solar radiation were important, especially in the tropical ocean. Careful attention must be paid to the treatment of solar radiation data and the absorption scheme of radiation for ocean modeling.

Further observation-based studies are needed to clarify the long-term trend of precipitation in the Indian Ocean. In addition, from the standpoint of ocean modeling, further progress on reanalysis products is desired to improve sea surface fluxes, for example by including air-sea interaction processes (e.g., Fujii et al., 2009), which are lacking in the current atmospheric reanalyses.

Although this study set the ratio of the longer-wavelength (IR) to the total at the surface to a constant (*R* in (1)), atmospheric models generally treat direct rays and scattered light separately, and provide spectral intensities of radiation. Thus, the absorption of radiation in the sea can be accurately calculated by using coupled models. This is the next step of this study.
