Study On The Estimation Method Of Areal Mean Free Water Storage Capacity In Xin’anjiang Model

In humid and semi-humid regions,Xin’anjiang model is widely used in rainfall runoff simulation and forecast because of its high prediction accuracy and strong applicability.The areal mean free water storage capacity(SM)is an important,sensitive and relevant parameter in Xin’anjiang model.Its value has a significant impact on the simulation results.At present,SM is mainly obtained by calibration,which inevitably leads to “ equifinality ” phenomenon.Based on the Xin’anjiang model theory,the estimation methods of the areal mean free water storage capacity of the basin and the areal mean free water storage capacity of the calculation unit are studied in this thesis.The main research contents and results are as follows:(1)A method for calculating the areal mean free water storage capacity(SM)of a basin based on Xin’anjiang model theory and the measured hydrograph is proposed.An ideal measured hydrograph and corresponding rainfall process are selected,according to the rainfall in the last period,the starting and ending time of the surface flow,the interflow and the groundwater flow formed by SM in the measured hydrograph,and travel time of three types of runoff in the slope and river network are determined.The ratio of RG and RI and SM and the recession constant of free water storage reservoir are given,the recession constants of the linear reservoir of the interflow and groundwater flow can be obtained through goodness-of-fit,and the inflow hydrograph of RI and RG to river network is obtained.Based on travel time in river network,dimensionless unit hydrograph of flow concentration in river network is generalized.The inflow hydrograph of RI and RG to the river network is convoluted with the unit hydrograph of flow concentration in river network,the hydrograph of RI and RG,their superposition are obtained respectively,so as to obtain the time to peak of the groundwater flow and the interflow and the groundwater flow formed by the interflow and groundwater flow can be obtained.According to the rise time,time to peak and measured recession process,the hydrograph formed by RG and RG and RI is obtained,and then SM,RG and RI are obtained.This method is applied to Yandu River basin and Shanxi Ying River basin respectively.Simulation results show: the mean value of coefficient of efficiency of 18 floods in the Yandu River basin is 0.860,and the qualified rate is 83.33%;the mean value of the coefficient of efficiency of 17 floods in Shanxi Ying basin is 0.921,and the qualified rate is 100%.It is shown that the ideal SM and KI/KG ratio can be obtained by the proposed method.(2)A method for calculating the areal mean free water storage capacity(SM)of calculation unit is proposed.SM is related to the thickness of the soil layer,and the thickness of the soil layer is related to the topographic slope.The trial and error method is used to analyze the functional relationship between SM of the calculation unit and the topographic slope,and SM with the topographic slope of 0 is introduced into the functional relationship as the parameter to be calibrated.Using the analyzed function relationship,SM of each calculation unit is calculated according to topographic slope.This method is applied to Yandu River Basin.The Xin’anjiang model considering the different SM of the calculation unit(referred to as the improved model in this thesis)is constructed.The SM of the basin calculated by the method proposed in(1)is directly brought into the model,the mean value of coefficient of efficiency of the simulation results is 0.872,the qualified rate is 83.33%.After calibrating the SM parameters,the mean value of coefficient of efficiency of the original model is 0.862,and the qualified rate is 83.33%.The mean value of coefficient of efficiency of the improved model is0.872,and the qualified rate is 88.89%.It can be seen that the simulation effect of the proposed method is improved.The sensitivity analysis of SM parameters of the model before and after improvement was carried out.The results show that the sensitivity of the SM parameter of the improved model is reduced.