Precipitation is a key variable to enhance the understanding of water cycle system and secure and manage the water resources efficiently. In this study, we evaluated the feasibility of GPM precipitation datasets through comparison with the 92 ASOS sites in South Korea during 2015. Additionally, three merging techniques (i.e., Geographical Differential Analysis, Geographical Ratio Analysis, Conditional Merging) were applied to improve accuracy of precipitation by fusing the advantages from point and satellite-based datasets. The results of this study are as follows. 1) GPM dataset indicated slightly overestimation with compared ASOS dataset, especially high uncertainties in summer season. 2) Validation of three merging techniques through jackniffe cross-validation showed that uncertainty were decreased as the spatial resolution increased. Especially, conditional merging showed the best performance among three methods.