Estimation of time-variant probable maximum precipitation for South Korea

The objective of this study is to suggest a methodology for estimating the time-variant probable maximum precipitation. First, the greatest amount of precipitation for a relatively long duration, e.g., a month or longer, defined as the Long-term Probable Maximum Precipitation (LPMP) in the present work was estimated. Then, the procedure for using the LPMP estimates for determining the temporally varied upper limit of precipitation during a certain period of the year, defined as the time-variant PMP (TPMP) in this study was implemented. Hershfield's statistical approach was used to estimate LPMP, i.e., monthly, calendar monthly, and cumulative monthly, for each of 61 weather stations located in South Korea, and nationwide distribution maps of three types of LPMP were made. In particular, using the cumulative monthly PMP estimates and total antecedent precipitation prior to a certain time, the TPMP for a certain duration was forecast for example problems. From the case study for precipitation data at Chungju weather station, TPMP forecasts at certain stormy periods were substantially smaller than the design PMP, which implies that reservoir operation for flood control based on the design PMP may be too restrictive for some period. It is expected that the TPMP can serve as a complementary index to increase the water supply of a reservoir.