Investigation of design rainfall distribution for Lower Da River Basin under climate change

Rainfall intensity expected great changes under changing climate. Therefore, using the intensity-duration-frequency (IDF) curves based on historic rainfall data likely leads to the underestimation of related-risks to the design of drainage systems. For future conditions, probability distributions used in the establishment of IDF curves might need to be re-evaluated. This study investigates the distributions of Gumbel, generalised extreme value, generalised Laplace and generalised exponential in rainfall frequency analysis for Lower Da River Basin, Northern Vietnam. The results indicated that generalised Laplace best-fitted to observation data, but generalised extreme value is the most appropriate distribution for generating IDF curves under changing climate. Importantly, at level of 50% percentile, the strongest changes in rainfall intensity range from 2.7% at 10 years to 41.4% at 200 years and from 5.5% at 25 years to 22.0% at 200 years of return period under RCP4.5 and RCP8.5, respectively for duration of 72 hours.