Prediction of Monthly Streamflow at Ungauged Basin Using Flow Duration Curves

Abstract

In order to make a reliable analysis, the requirements of the data in terms of their qualification and quantity are crucial. In this study, monthly
streamflow values for target (ungauged) station were predicted by using a spatial interpolation approach based on flow duration curves. The
monthly streamflow data of 5 gauge stations in Middle Euphrates basin were examined. The data contains a period of 31 years without any
missing data in the dataset. In the first step of the procedure, flow duration curves (FDCs) of the 5 stations are constructed by gauged records.
Each of the 5 gauge stations is assumed to be ungauged in turn. The weight values were determined by correlation coefficients. A streamflow
prediction on any month at the target station was made by identifying the percentage point position on the flow duration curve of the streamflows for the same month at the source stations and reading off the streamflow value for the equivalent percentage point from the target station’s
duration curve. Then, each predicted streamflow value of target station was multiplied by the source station’s weight. This study showed that
promising results were obtained with this method, according to measured by the statistics of prediction performance, namely the Nash-Sutcliffe
efficiency (NSE). The method used in this study can also be applied to other meteorological and hydrological variables such as pressure, temperature, evaporation, precipitation, etc.