Study On The Evolutionary Characteristics And Attribution Of Extreme Runoff Events In Watersheds Under Nonstationarity Conditions

In the context of global climate change,the frequent occurrence of extreme precipitation and runoff events poses a serious threat to human society.Under the influence of various factors such as climate change and human activities,the spatial and temporal evolution patterns of the global water cycle have changed,and how to identify the new characteristics of the evolution patterns of extreme runoff events affected by the changing environment is of some significance to regional water security issues.In order to explore the analysis method of non-coherent runoff changes in the current environment to achieve the purpose of scientifically and reasonably predicting the changes of extreme runoff events in the future,the Haihe River basin was selected as the research object,the change characteristics of extreme runoff in the Haihe River basin were analyzed,and the influence of climate change and human activities on extreme runoff events was studied by applying the non-coherent hydrological frequency analysis method.The results are as follows.(1)Non-consistent diagnosis of hydrological sequences.The non-consistency of extreme runoff was first identified before starting the analysis,and the Mann-Kendall test and Spearman test were adopted to test the trend of runoff;the Pettitt test,Mann-Kendall test and Buishand test were adopted to test the sudden change point of runoff.A comprehensive judgment of the authenticity of the variability of the runoff series shows that the runoff series in the Haihe River basin exhibit obvious non-consistent characteristics.The maximum one-day runoff and maximum fiveday runoff at two hydrological stations in the Haihe basin have a significant decreasing trend,the abrupt change in runoff at the Boundary River Pavement station is clearly influenced by the construction of its upstream reservoir.(2)Nonstationarity frequency analysis of runoff series in a changing environment.Climate indicators were filtered by calculating Kendall correlation coefficients while substituting reservoir indicators,and based on the GAMLSS model,one consistency model was constructed after using the AIC criterion rate determination,and two variables were different non-consistency models.The results of the non-coherent frequency analysis of runoff show that the non-coherent runoff series analysis model with the introduction of physical factors gives the best results,as well as good predictions and better adaptability to changing environments.This non-coherent model considering climate change and human activities describes accurately the fluctuations and abrupt changes of flood extremes.From the viewpoint of water engineering applications,the flood design values predicted by the non-consistent model with climatic and reservoir indicators as explanatory variables are more reliable and can provide theoretical support for the design of water engineering.(3)Attribution analysis of runoff changes.Analysis of the relationship between runoff variability and climate indicators using a model that has been constructed,The results show that the NINO1+2 index in March,April and May significantly influenced the runoff series at the Kuangmen Station and is negatively correlated with it.Increased frequency of extreme runoff events at lower NINO1+2 indices.The runoff series at Boundary River Pavement Station is negatively correlated with NINO1+2 in February,March and April and NAO in March,April and May at the same time.Low levels of both indicators indicate an increased frequency of extreme runoff events.Provides hydrologists with the possibility of anticipating extreme runoff events in advance.