Frequency analysis of extreme precipitation in different regions of the Huaihe River Basin

Abstract Affected by climate change and human activities, the frequency and uncertainty of extreme precipitation have shown an increasing trend, which has brought great challenges to the prevention of flood disasters. The Huaihe River Basin (HRB) is located at the junction of northern and southern China and is more susceptible to climate change. In order to obtain the spatio‐temporal distribution characteristics of extreme precipitation in the HRB, in this study, the L‐moments method was used to analyse the regional frequency distribution of the annual precipitation (YP), the maximum one‐day rainfall (RM1day), the maximum five‐day rainfall (RM5day) and the maximum 15‐day rainfall (RM15day) in the HRB. The results show that the HRB can be divided into five sub‐zones on the basis of the division of natural sub‐basins to ensure regional consistency. The accuracy of quantile estimation will decrease as the return periods increase. When the return period is less than 100 years, the relative root mean square error (RMSE) can be guaranteed to be relatively small, and the cumulative RMSE is less than 0.3. However, after the return period is greater than 100 years, the 90% confidence interval of the quantile estimate increases rapidly, indicating that the estimated quantile error increased. The northwestern part of the HRB may experience extreme annual precipitation, which is different from the normal precipitation distribution. The extreme precipitation in Zone 1, Zone 2–2 and Zone 4 is not only strong but also has a large increase speed, making these areas the key focus zones for flood control and anti‐logging. This study can provide important reference for the prediction and prevention of extreme precipitation in the HRB.