This study describes an atmospheric-hydrological analysis to improve the water management system in the Korean peninsula. This study has two parts, the first part contains the proposal of a new high resolution (0.25°×0.25° latitude/longitude) gridded daily precipitation dataset (K-Hidra version 2020). This reanalysis data is based on 399-gauge observations in the Korea peninsula. Out of all the available gauges, only some stations have complete data. Therefore, to handle the missing data, 6 infilling method are evaluated by three supportive but separated evaluation frameworks. In our study area, the elastic net model has the best infilling capability. The elastic net model showed good results even in environments where data length and spatial expression are limited. After getting a complete dataset, spatial interpolation was performed. The produced K-Hidra was compared with various data including uninterpolated data, CPC, APHRODITE, TMAP and GPCP. K-Hidara showed the importance of infilling process in historical precipitation data reconstruction. It has been proven that the infilling process can improve spatial and temporal expressiveness of the product. Furthermore, through the use of K-Hidra, reasonable changes in precipitation trend such as annual daily maximum precipitation and total annual dry days have been discovered. In the study area, K-Hidra showed high performance in representing spatial variation than other products. It is then concluded that K-Hidra can be utilized for the second part of this paper. The second part explores the possibility of long-term prediction of early winter precipitation using the Ocean-atmospheric effect over South Korea. Lagged-correlation showed a high correlation value between the spring Atlantic Ocean and the winter precipitation in Korea. The variable selection and contraction method of the elastic net model show that the Atlantic Ocean Dipole (AOD) that we have confirmed in the Atlantic region in March and April, can be significant for prediction. Random sampling analysis using March and April AOD can get good predictions having high correlation values. Composite analysis and correlation tables show connection between AOD and ENSO. An increase in spring AOD, shows an increase in summer ENSO. While the increase in summer ENSO also affects the winter precipitation in South Korea. However, this effect train does not work in reverse. Finally, this research contributes to both the development and use of a precipitation data in South Korea. K-Hidra is expected to be useful for climate impact and rainfall variability analyses over the Korean Peninsula. Meanwhile, the Atlantic Ocean Dipole has been found to be vital in understanding the early winter precipitation over Korea.