Current models for seasonal climate prediction are limited due to poor initialization of the land surface soil moisture states. Passive microwave remote sensing provides quantitative information on soil moisture in a thin near-surface soil layer at large scale. This information can be integrated with a land surface process model through data assimilation to give better prediction of the near surface and deep soil moisture states than model predictions or remote sensing observations alone. A Kalman filter-based data assimilation strategy has been implemented in the catchment-based land surface model(CLSM) used by the NASA Seasonal-to-Interannual Prediction Project (NSIPP). In this study, we assimilated Scanning Multifrequency Microwave Radiometer (SMMR) data for the period of 1979-1987 and compared the resulting soil moisture with in-situ measurements collected in Russia, Mongolia and China. In this presentation, we will present the results with different assimilation approaches including standard assimilation, assimilating soil moisture index, assimilating soil moisture differences, assimilation with bias correction. We will present the advantages and disadvantage of each methods. Our comparison results show that the last three assimilation approach is better than the standard assimilation approach and also demonstrate that the data assimilation method used here improved our soil moisture estimation over either by model or remote sensing alone. It is demonstrating that Kalman filter-based assimilation is a feasible approach which can be used to combine remote sensing observations and land surface models for improved soil moisture initialization. Our algorithm can also be used to assimilate data collected from the Advanced Microwave Scanning Radiometer for the Earth (AMSR-E) observing system instrument on the current EOS Aqua satellite when it becomes available to provide better soil moisture states for real time forecasting.