Analysis of Influential Factors of Bankfull Discharge in the Lower Wei River

The main channel of the Lower Wei River,the largest tributary of the Yellow River,was deposited and shrinked seriously in recent years.It lead that the bankfull discharge was decreased significantly,therefore flood flows usually overflow the floodplains whether high or low flood flows.The flow and sediment load was very small in the nonflood season,so the effect of flow and sediment load in the nonflood season was ignored.This paper presents a study on the effects of flow and sediment load in the nonflood season and Tongguan elevation and peak flow on bankfull discharge at hydrometric stations of Huaxian in the Lower Wei River.BP neural network was used to develop models of bankfull discharge with the input conditions including the controlling factors,such as flow and sediment discharge.We compared the model with input data of water and sediment discharge in only flood season and that with water and sediment data during whole hydrological year.The results showed that after considering water discharge and suspended sediment load in nonflood season and Tongguan elevation and peak flow,the accuracy of the models increased at different degrees,with higher coefficient of efficiency and lower MRE.The calculation of bankfull discharge becomes more accurate with the flow and sediment load in nonflood season.