Research On Hydrological Models Based On Multi Mixed Runoff Generation Mode

Hydrological model is the core tool to simulate rainfall-runoff process.At present,most hydrological models adopt single excess-saturation or excess-infiltration runoff generation mode with limited precision and great uncertainty.Mixed runoff generation mechanism as a new runoff generation mode is expected to solve the problem of single runoff generation structure.This paper is devoted to the study of hydrological model based on mixed runoff generation mode,focusing on the rationality and uncertainty of runoff generation structure in the model,in order to improve the simulation accuracy and identify the sources of uncertainty.The main research contents and results are as follows:(1)A Study on the theory and mode of runoff generation.Horton's runoff formation theory and Dunne's runoff formation theory,the excess-saturation runoff generation mode and the excess-infiltration runoff generation mode are introduced.On this basis,three typical mixed runoff generation modes(vertical mixed runoff generation mode,compatible runoff generation mode and VIC-3L mixed runoff generation mode)are selected,and their ideas and principles to realize mixed runoff mechanism are summarized.The discovery is that all the existing mixed runoff generation modes are coupling of excess-infiltration runoff generation mode and excesssaturation runoff generation mode in different ways,and the main difference between the three mixed runoff generation modes is only the difference of coupling ways and infiltration formulas.(2)Establishment of hydrological model based on mixed runoff generation mode and verification of simulation effect.Integral watershed hydrological model(Vertical mixed runoff model,compatible runoff model and VIC-3L mixed runoff model)based on the three mixed runoff generation mode are constructed to simulate 19 flood events in Luanchuan basin.The practical application effect of mixed runoff generation mode is tested and compared with that of excess-saturation runoff generation mode.The results show that the three mixed runoff generation modes are applicable to flood simulation in Luanchuan basin,and the simulation effect is better than that of excess-saturation runoff generation mode,which confirms that the mixed runoff generation mode is more perfect in runoff generation structure and better in simulation effect.(3)A comparative study of three mixed runoff models in different regions.13 typical river basins in 7 provinces were selected nationwide,and the simulation effects of vertical mixed runoff model,compatible runoff model and VIC-3L mixed runoff model in different regions were compared extensively.The results show that the three mixed runoff models are basically the same,and only little difference exists in the structure of these models.The three models can all simulate rainfall-runoff process better.and their simulation effect is basically the same.(4)Research on the structural uncertainty of mixed runoff generation mode.A quantitative framework based on ANOVA method is constructed to study the sources of structural uncertainty of mixed runoff generation mode.The uncertainties of coupling ways,infiltration formulas and their interaction in runoff generation simulation and runoff process simulation are quantified.The results show that coupling ways and infiltration formulas are important sources of uncertainty in runoff simulation,among which infiltration formula is the most important one that leads to the uncertainty of runoff simulation.The uncertain infiltration calculation method can produce greater uncertainty to the whole runoff process by influencing the runoff production process.Therefore,when studying the mixed runoff production mode and constructing the mixed runoff production model,great attention should be paid to the selection of reasonable infiltration calculation method.The current method is not enough to fully simulate the actual infiltration process.Research on a new infiltration calculation method with physical mechanism is an important way to reduce the uncertainty of runoff simulation.