Projecting changes in the frequency and intensity of future precipitation and flooding is critical for the development of social infrastructure under climate change. The Mekong River is among the world's large‐scale rivers severely affected by clima...
Projecting changes in the frequency and intensity of future precipitation and flooding is critical for the development of social infrastructure under climate change. The Mekong River is among the world's large‐scale rivers severely affected by climate change. This study aims to define the duration of precipitation contributing to peak floods based on its correlation with peak discharge and inundation volume in the Lower Mekong Basin (LMB). We assessed the changes in precipitation and flood frequency using a large ensemble Database for Policy Decision‐Making for Future Climate Change (d4PDF). River discharge in the Mekong River Basin (MRB) and flood inundation in the LMB were simulated by a coupled rainfall‐runoff and inundation (RRI) model. Results indicated that 90‐day precipitation counting backward from the day of peak flooding had the highest correlation with peak discharge (R2 = .81) and inundation volume (R2 = .81). The ensemble mean of present simulation of d4PDF (1951–2010) showed good agreement with observed extreme flood events in the LMB. The probability density of 90‐day precipitation shifted from the present to future climate experiments with a large variation of mean (from 777 to 900 mm) and SD (from 57 to 96 mm). Different patterns of sea surface temperature significantly influence the variation of precipitation and flood inundation in the LMB in the future (2051–2110). Extreme flood events (50‐year, 100‐year, and 1,000‐year return periods) showed increases in discharge, inundation area, and inundation volume by 25%–40%, 19%–36%, and 23%–37%, respectively.
The ensemble mean of historical simulation of d4PDF dataset had good agreement with observed extreme flood events in the LMB.
The large ensemble members of d4PDF dataset could improve flood extreme prediction and reduced its uncertainty.
The relative changes of historical and future experiments for extreme flood events of 50‐, 100‐, and 1,000‐year return period showed an increase of discharge, inundation extent, and inundation volume by 25%–40%, 19%–36%, and 23%–37%, respectively.