Variations of extreme precipitation events with sub-daily data: a case study in the Ganjiang River basin

Abstract

Climate warming increases the intensity of extreme precipitation. Studying extreme precipitation patterns and changes is vital to reducing risk. This paper investigates thresholds, changes and timescales for extreme precipitation using sub-daily records from meteorological stations in the Ganjiang River basin. We use the gamma distribution and select the L-moment method to estimate the parameters α and β. Results show that (1) continuous precipitation events of 36 h contributed the most precipitation to the total but with lower frequency, which would be key events for flood monitoring; (2) the intensity and the occasional probability of extreme precipitation will increase in spring in the future in stations like Yifeng, Zhangshu and Ningdu, which will in turn increase the risk of storm floods; and (3) spatial distribution of extreme precipitation risk shows that the risk increases as elevation increases in the northern lowland and the Jitai Basin in the midstream region, while the risk in the southern mountainous region decreases as elevation increases. These findings will facilitate emergency preparedness, including risk management and disaster assistance, in the study areas.