Despite a number of hydrological studies on floods, the studies on droughts have been relatively inactive.
In 2015, droughts caused by water outage in Korea increased public attention to drought management across the country. In this study, the amount of rainfall is estimated based on duration and frequency of each probability distribution, applying the L-moments method to the long-term hydrologic data from Cheongju-Musim river catchment in Korea. The five probability distributions (Generalized Extreme Value (GEV), Generalized Pareto, Kappa, Pearson Type 3, and Wakeby) are used to obtain the ranges of frequency-based probabilistic estimates of low rainfall. This study could reduce the prediction errors for drought frequency using the suggested ranges of probabilistic estimates of low rainfall. It will be used as basic data for hydrological management at Cheongju-Musim river Catchment, Korea.