In this paper, a statistical downscaling methodology is developed to investigate its applicability in downscaling wind speeds over South Korea. The investigation includes calibration of the statistical downscaling model by using large-scale atmospheric variables encompassing NCEP/NCAR reanalysis data, validation of the model for the calibration period, and prediction of the future regional wind speed scenario based on the general circulation model (GCM) outputs of scenario A1B of the CGCM3. Selected as climate variables for downscaling were measured daily wind speed data (1971~2000) from 8 weather stations over South Korea. The results show that: (1) There exists a close relationship between the observed and the simulated daily wind speeds during the calibration period of 1971~2000. (2) Under the scenario A1B, during the prediction period of 2011~2100, effects of climate change influencing on the daily mean and 10 minute maximum mean wind speeds are expected, and these effects are different from region to region over South Korea. (3) However, based on the scenario A1B of the CGCM3, the potential impacts of climate change on the surface wind speed is relatively minor, and the current design wind velocity may not need to be changed for the predicted period.