Application Of The Baseflow Segmentation In Flood Forecasting Of Upper Manghe Basin

The accurate prediction of the time,place,peak discharge and possible loss of the coming flood will play a guiding role in developing scientific and timely flood control work.Basal flow is an important replenishment source of river runoff.It is of great significance for water resources planning and maintaining river environmental health to select an appropriate base-flow segmentation method,and it is also an important research content for flood prediction and runoff yield and confluence calculation.Upstream as the research object,this article selects Manghe river basin in upper Manghe basin hydrological stations in the river basin control area as the study area,through the study collected regional meteorological hydrology,topography data,investigation and analysis of flood characteristics of the study area,and the space-time change source hydrological model is established to selected 11 floods(6 rate on a regular basis,5 field validation period)for flood forecasting and base flow simulation calculation,the change of time and space source in the applicability of the pythons river basin hydrological model.The main research results are as follows:(1)The flood characteristics in the study area are analyzed on the basis of hydrometeorological data in the study area.The annual precipitation in the study area was mostly concentrated in summer,which was consistent with the historical flood months.Because the topography of the study area is mostly mountainous and hilly,and the river system of the basin is fan-shaped,the confluence time is short and concentrated.Therefore,according to the observation of river system,topography and geomorphology features of the basin,and the flood process line of 6-field rate regular floods,it is found that floods tend to rise steeply and fall slowly,and the time difference between the rain peak and the flood peak is mostly within 4 hours.(2)The influence of filtering times and parameter values on the results when digital filtering method is used for base flow segmentation in small watershed is analyzed.The results show that the results of base flow segmentation are mainly affected by the filtering times,and the base flow decreases with the increase of filtering times.As the hydrological characteristics of small watershed,the author improved the UKIH method.Different base-flow segmentation methods are used to segment the measured runoff data in flood season in 2013 in the study area.The results show that the small watershed smoothing minimum method and the Boughton-Chapman filter method have good applicability in the study area.(3)The FFMS(fast flood model system)is used for the flood prediction of 11 selected floods(6 with periodic rate and 5 with verification period).The result of flood prediction shows that the qualified rate of regular flood field is 100% and the accuracy is grade A.The qualified rate of the flood field in the verification period is 75%,and the accuracy is B.The spatiotemporal variable source hydrological model has good applicability in the upper Manghe Basin.The base flow in the flow generation module of the FFMS is calculated by the linear reservoir method.Based on the characteristics of the base flow,the nonlinear reservoir method is proposed to replace the linear reservoir method.In order to verify the superiority of the nonlinear reservoir method,the Boughton-Chapman filtering method(3 times filtering,parameter 0.925)is used to evaluate the two methods.The results show that the base-flow segmentation results of the nonlinear reservoir method are closer to the evaluation value than that.