Analysis on the temporal scaling behavior of extreme rainfall in Korean Peninsula based on high-resolution radar-based precipitation data

South Korea To overcome the limitations of relying solely on ground precipitation gauges, this study utilizes radar precipitation data to investigate the temporal scaling behavior of extreme rainfall values. Maximum precipitation and extreme quantile values for each grid point were calculated using the rolling-window summation method. The investigation focuses on the relationship between maximum precipitation and duration. The study's findings are as follows: (1) Radar estimates higher maximum precipitation values than ground gauges, especially in mountainous areas with sparse gauge coverage; (2) The maximum precipitation-duration relationship deviates from a power-law relationship primarily due to unusual short-duration extreme rainfall events; (3) Lower-quantile high rainfall values show a stronger power-law relationship influenced by various rainfall mechanisms; (4) The East Asian rainy season induces greater extreme rainfall for durations up to 6 h, while longer durations are dominated by typhoons, indicating different flood risks; (5) Maximum precipitation values for most durations are observed on Jeju Island, primarily caused by typhoon events. Considering that climate change is expected to induce a northward shift in typhoon paths, appropriate flood defense measures should be implemented, especially in the southern part of the Korean Peninsula. These findings highlight how different observation methods can significantly impact flood risk assessment and the design of key hydraulic structures.