The Analysis Of Components And The Research Of Characteristic For Time Series Of Runoff In The Upstream Of The Fenhe River

Under the influences of climate changes and human activities, basin hydrological cycle and water resources systems have been constantly changing. The changes in hydrological cycle of velocity and route also alter the means of material transportation and energy exchange at the same time, and further alter the structure and function of water resources systems. These alterations accumulated day after day will have a profound impact in the local areas or even the whole watershed hydrological cycle and water resources system. So, what are the characteristics and laws in this process of alterations and affections, and how to identify and mine these characteristics and laws from the qualitative and quantitative aspects, has attracted widespread attention of home and abroad hydrologists.The Fenhe River is the second largest tributary of the Yellow River. It is just like a mother to raise the people of Shanxi province thousands of years. The Fenhe River basin has the densest population and distribution of productivity, so it occupies a very important economic and social position in the province. In recent years, Fenhe River basin water resources and water environment has undergone tremendous changes with the emergence of drying up of rivers, dropping of groundwater levels, drying up of springs, pollution of water and other phenomena under the influence of climate changes and human activities, which has seriously restricted the basin sustainable development of economy and construction of ecological environment.This paper mainly analyses the evolution of runoff in the upstream of the Fenhe River. It analyses the time series components of monthly runoff in the upstream of the Fenhe River which has contained the trend components, jump (mutation) components, cycle components and random components. These components are respectively identified using stochastic hydrology theoretical approaches of Kendall rank correlation test method, ordered clustering method, rank-sum test method, variance analysis method and autocorrelation analysis method. It applies the fractal theory approaches of the rescaled range analysis method to mine the long-range correlations properties to time series of monthly runoff at the upstream of the Fenhe River. It applies the chaos theory approaches of the saturated correlation dimension method to mine the chaotic properties to time series of monthly runoff at the upstream of the Fenhe River. At last, it makes preliminary judgments and analysis of the causes to evolution of time series of monthly runoff at the upstream of the Fenhe River from two influencing aspects of climate changes and human activities.There are three researching conclusions. First, the monthly runoff time series of upstream has contained declining trend components, apparent jump components, evident cycle components, and the surplus time series is stable random time series which has removed the trend, jump and cycle components. Second, the monthly runoff time series of upstream has long-range correlations properties and chaotic properties, all that reflect that the runoff time series will continuously maintain the declining trend in the period of future, and the hydrological system is consist of certain and random components which has chaotic character. Third, the evolution of time runoff series is controlled by climate changes in long period, but interfered by human activities in short period.