Statistical approaches for correcting bias in general circulation model’s monthly precipitation are compared. We evaluate the methods by focusing on the average and variance of the bias-corrected data. For average, the difference of results between methods is small. For variance, on the other hand, the difference is not small. The characteristics of bias corrected data are also compared between methods and between GCM outputs. The methods that correct the bias better in the retrospective period tend to reduce the differences of projection between GCM outputs. The methods are applied to the outputs of MIROC5, and their bias corrected data is analyzed in this study.
