This study aims to evaluate the impact of aerosol indirect effects (AIEs) on climate variations over tropical oceans through a three-dimensional cloud-resolving model, the System for Atmospheric Modeling, in idealized radiative-convective equilibrium (RCE). In RCE framework, the interactions among radiation, convection and surface fluxes are explicitly included while the effects of the large-scale circulation on convection are ignored. The AIEs on RCE are modeled by varying the number concentration of cloud condensation nuclei (CCN), served as a proxy for the aerosol amount in the environment, over a wide range starting from pristine maritime (50 cm-3) to polluted (1000 cm-3) conditions. Two sets of experiments are performed: (1) with an interactive sea surface temperature (SST) predicted by the simple slab ocean model and (2) with a prescribed SST fixed at 300 K. For simplicity, both experimental sets were run with constant insolation and removed diurnal cycle. In interactive SST runs, it took several hundred days until they reach a quasi-equilibrium state. Simulation results show that the presence of CCN causes reduced longwave cloud radiative forcing (0.6-2.5 W m-2) but enhanced shortwave cloud radiative forcing (0.3-1.5 W m-2) in both the interactive SST (ISST) and fixed SST (FSST) experiments. In the ISST runs with the highest CCN count, AIEs mitigate most, 1.5 K, of the greenhouse warming, 2 K, as simulated by the doubling-CO2 experiment. It is found in both ISST and FSST runs that the increase of CCN count tends to decrease the low-cloud and high-cloud covers, but increase the middle cloud cover, enhance cloud liquid water path, snow, and graupel water paths, but reduce cloud ice and rainwater paths. The qualitative differences in hydrological cycle between the ISST and FSST are also found. In ISST runs, cooler SSTs resulted from enhanced CCN counts tend to reduce precipitable water, latent heat flux and, consequently, decrease surface precipitation rate. In the FSST runs, on the other hand, the effects are opposite, that is slightly increased latent heat flux, constant PW and increased surface precipitation rate. These differences suggest that the estimates of AIEs over tropical oceans can be quite sensitive to the choice of the fixed or interactive SST framework.