Ensemble optimal interpolation, a simplified and computationally inexpensive version of ensemble Kalman filter, has been used to assimilate satellite-derived sea level anomaly and sea surface temperature in an Indian Ocean circulation model. In order to cut off the long-range spurious covariances, a localization technique has also been used where an assumption is made that only measurements located within a certain influence radius of a grid point can affect the analysis at that grid point. It has been found that the analyses fit the assimilated observations quite well. For a more stringent test of the power of the assimilation technique, selected regions have been demarcated in the analysis area. Simulated mixed layer depth, depth of the 20 ^°C isotherm, and temperature at 400-m depth have been compared with observations of these variables in these regions, and assimilation has been found to exhibit significant positive impact. Simulated surface and subsurface temperatures have been compared at isolated Research moored array for African-Asian-Australian monsoon analysis and prediction (RAMA) buoy locations, and again, the high positive impact of assimilation has been evident from these comparisons. Finally, the impact of assimilation on surface current simulation at selected RAMA buoy locations has been found to be somewhat marginal, and a significant improvement apparently needs additional assimilation of drifter data with a very high concentration of drifters.