Diagnostic Method On The Nonstationarity Of Precipitation And Runoff In River Basin And Case Study In North Haihe Reaches

Climate change and human activities have brought the dramatic change to hydrological cycle, including the increase of evapotranspiration, the decrease of the channel flow even the cutoff of the river and the extreme hydrological events. Thess changes challenge the stationarity hypothesis of the hydrological hydrologic series, as the foundamental theory of the frequency calculation, which provide the water resources management, plan and planning, flood control scheme, the design and implement of water projects the stationary design values. To examine changes of non-stationary evolution of the watershed rainfall and runoff, this disseratation analyze the content of non-stationarity and its definition by summarizing numerous related papers, classified the hydrological non-stationarity according to different forms, and baesd on the classification, proposed the corresponding non-stationaryity diagnostic methods for watershed rainfall and runoff. The diagnostic method has been successfully applied in the north part of Haihe River Basin and the Beijing-Tianjin area. The main contents and results are as follows:(1) By analyzing causes of the non-stationarity and the state of the water cycle system and itsoperation mechanism analysis, hydrological non-stationary are defined:with the background of a varying physical condition, the hydrological monitoring data no longer belong to the same totality, as the response to the changing environment, reflected in hydrological samples no longer meet the IID (Independent Identical Distribution) assumption. Reference to hydrologic series analysis and based on definition of the non-stationaryity, the non-stationarity of hydrology forms could be divided into two categories: the difference non-stationarityy and the trend non-stationarity, reflecting the changes in the mean and thevariance, respectively. The results complement the existing understanding on non-stationarity.(2) According to two catogories of hydrological non-stationarity, the non-stationary diagnostic method for single hydrologic series is proposed, identifying the change in the mean and variance, respectively, improved the current methodology emphasing on the mean while ignoring the variance. On this basis, the concept and the algorithm of the non-stationary degree for multiple hydrologic series was proposed which quantify the non-stationarity.(3) The methods were applied to 981 Regional hydrologic series of the rainfall and runoff of in the North part of Haihe River Basin and the Beijing-Tianjin area (NHBT). The results showed that the non-stationary degree of precipitation in NHBT is 80/380, non-stationary degree of extreme precipitation is 161/220, non-stationary degree of runoff is 280/380, the rainfall - runoff relationship shows non-stationarity. From views of the categories, trends and difference non-stationarity compose the precipitation (difference/ trends:39/46), indicating that the mean and variance of considerable changes in precipitation, extreme precipitation mainly shows a difference non-stationarity (difference/trends: 159/18), that the main change is in the variance, runoff mainly reflects the trend non-stationarityy (difference/trends:124/252), that the main change is in the mean. As for the spatial distribution of the non-stationary degree:Piedmont region has a high level non-stationary in precipitation, the mountain area has a high level non-stationary in the extreme precipitation, especially in variance; large irrigation districts and other areas in the upstream of the Sanggan River basin has a high degree of non-stationarity in runoff, especially in the mean.(4) Through the case study, the proposed non-stationarityy diagnostic method for basin rainfall and runoff is proved to be accurate in screening out non-stationary hydrologic series, and identifying the categories of non-stationarity of the non-stationary hydrologic series, and obtaining the time scale of most significant non-stationary mutation in watershed precipitation, runoff, and the characteristics of most significant non-stationary mutation of extreme precipitation. This study realize the quantification and comparation of non-stationarity of hydrological factors in multiple time scales, as well as the in multiple sites.