In this study, we investigate the significance of using an improved error modeling strategy to characterize the spatio-temporal characteristics of uncertainty in simulation of soil moisture fields from an off-line land surface model forced with satellite rainfall data. We coupled a Two-Dimensional Satellite Rainfall Error Model (SREM2D) with the Common Land Model to propagate ensembles of simulated satellite rain fields for the prediction of soil moisture at depths of 5 cm (near surface) and 50 cm (root zone). Our investigations revealed that multi-dimensional error modeling captures the spatio-temporal characteristics of soil moisture uncertainty with higher consistency than simpler bi-dimensional error modeling strategies. The proposed error modeling strategy appears to have the potential for delineating a more robust framework for the optimal integration of satellite rainfall data into models towards the study of global water and energy cycle.