Analysis and simulation of extreme flow events at Yichang station under climate change

Abstract

The annual maximum, annual minimum and annual average flow data of Yichang station during 1882 -2001 were used to explore the statistical law of extreme flow events through the methods of descriptive analysis, frequency analysis, linear trend analysis, linear regression test and Mann-Kendall test. Then annual maximum flow series were respectively simulated by the generalized extreme value distribution model (GEV) and generalized Pareto distribution model (GPD), and the point and interval estimations of annual maximum flow with the return periods of 10 years, 50 years, 100 years, 200 years and 500 years were derived by the maximum likelihood estimates (MLE) and profile -likelihood method. The results showed that three flow series decreased slowly during 1882 -2001, in which linear trend of annual maximum flow series was not significant, yet the opposite for annual minimum and average flow series. Compared with GPD model, GEV model was more fitting for extreme flow of Yichang station, indicating that the annual maximum flow over that period followed GEV distribution. Estimated values of annual maximum runoff were 67802m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> /s and 69263m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> /s for 50 and 100 years' return period, which increased with return period.