Extreme hydro-meteorological events have become the focus of more and morestudies in the last decade. Due to the complexity of the spatial pattern ofchanges in precipitation processes, it is still hard to establish a clearview of how precipitation has changed and how it will change in the future.In the present study, changes in extreme precipitation and streamflowprocesses in the Dongjiang River Basin in southern China are investigatedwith several nonparametric methods, including one method (Mann-Kendall test)for detecting trend, and three methods (Kolmogorov–Smirnov test, Levene'stest and quantile test) for detecting changes in probability distribution.It was shown that little change is observed in annual extreme precipitationin terms of various indices, but some significant changes are found in theprecipitation processes on a monthly basis, which indicates that whendetecting climate changes, besides annual indices, seasonal variations inextreme events should be considered as well. Despite of little change inannual extreme precipitation series, significant changes are detected inseveral annual extreme flood flow and low-flow series, mainly at thestations along the main channel of Dongjiang River, which are affectedsignificantly by the operation of several major reservoirs. To assess thereliability of the results, the power of three non-parametric methods areassessed by Monte Carlo simulation. The simulation results show that, whileall three methods work well for detecting changes in two groups of data withlarge sample size (e.g., over 200 points in each group) and largedifferences in distribution parameters (e.g., over 100% increase of scaleparameter in Gamma distribution), none of them are powerful enough for smalldata sets (e.g., less than 100 points) and small distribution parameterdifference (e.g., 50% increase of scale parameter in Gamma distribution).The result of the present study raises the concern of the robustness ofstatistical change-detection methods, shows the necessity of combined use ofdifferent methods including both exploratory and quantitative statisticalmethods, and emphasizes the need of physically sound explanation whenapplying statistical test methods for detecting changes.
展开▼
