Wind energy has been recognized as one of the most important and fastest growing energy resources without emission of air pollutant. Thus, it is necessary to predict wind speed and direction accurately both in time and space toward the efficient usage of wind energy. Numerical simulation experiments using the Fifth-Generation Mesoscale Model (MM5) are carried out to clarify the impact of surface observation data assimilation on the estimation of wind energy resources. The EXP_Radius run was designed with respect to the radius of influence in the Four-Dimensional Data Assimilation (FDDA), and the EXP_Impact run was made by changing the nudging coefficient that determines the relative magnitude of the nudging term. The simulation period covers a clear-sky event on 3 - 5 June 2007 and another is on 2 - 4 December 2006. It is found that the simulated results are very sensitive to the radius of influence and nudging parameters in the FDDA. The further analysis of the results shows that the impact of the radius of influence tends to be stronger in weak synoptic flow episode than that in strong synoptic flows episode. The nudging factor is also sensitive to the intensity of the synoptic flows.