Research On The Application Of Flood-forecasting Models In Semi-arid Regions Of North China

Xinanjiang model, TOPMODEL and HEC model are applied to Kuancheng watershed for flood simulation, Flood forecasting of semi-arid region is discussed in this thesis; the parameters' sensitivity and its uncertainty are also been discussed by choosing the case of TOPMODEL.Kuancheng watershed is located in the northern part of china. It has the characteristic of continental monsoon climate. The precipitation has great spatial and temporal difference. It usually occurs in summer. Because of the uneven distribution of precipitation, drought and flood often happen in this area. For this purpose, the study on flood forecasting method of this watershed is very important to the Flood-control, sustainable use of water resources, construction and management of water resources project, production of industry and agriculture and everyday life of people in this area. At the same time, the study of flood forecasting method also has the important instructive meaning to the flood forecasting of semi-arid region.According to the research, we can conclude that both of the repletion of storage and excess infiltration are probably apt to Kuancheng watershed, for flood simulation ,we should not only depend on the choosing of models to improve forecast accuracy, but also take into account of the influence of hydrologic data, water resources project and human activities; according to sensitivity analysis, we canconclude that parameters such as m ,SR_mm, T₀ and Rv are more sensitive, whileTd and CHv are less sensitive; for flood forecasting, the uncertainty has always been existed, Bayesian estimation, Generalized likelihood uncertainty estimation and modified probability forecast are introduced, we can get the confidence upper limit and confidence lower limit about different flood hydrograph by the method of modified probability forecast.