Comparison of the monthly streamflow forecasting in Maroon dam using HEC-HMS and SARIMA models

Streamflow forecasts are fundamental to the effective operation of flood control reservoirs and levee systems. Therefore, streamflow forecasting is of great importance. In this paper, the HEC-HMS conceptual model and SARIMA time-series model are compared to forecast streamflow in Maroon basin in the southwest of Iran to evaluate their ability and accuracy in monthly streamflow forecasting. First, the continuous rainfall–runoff was simulated monthly before the forecasting period by the HEC-HMS model. The monthly data from October 1991 to 2010 were used for verification. Also the data from 2011 to 2017 were used for calibrated HEC-HMS model. Streamflow forecast was conducted from 2018 to 2021 at the Idanak hydrometric station. To validate the SARIMA model based on the autocorrelation function, the partial autocorrelation of the residuals, Port-Manteau test, Akaike criterion and plotting the residual time series diagram on normal probability paper were used. The results showed that the accuracy of the HEC-HMS model in forecasting streamflow is higher than SARIMA model, the Root Mean Square Error (RMSE) of predicted and observed discharges for HEC-HMS and SARIMA models are 2.8 and 3.4 m3/s, respectively.