A comparison of ANN and HSPF models for runoff simulation in Gharehsoo River watershed, Iran

Abstract

In this study, the capability of two different types of model including Hydrological Simulation Program-Fortran (HSPF) as a process-based model and Artificial Neural Networks (ANNs) as a data-driven model in simulating runoff were evaluated. The area considered is the Gharehsoo River watershed in northwest Iran. HSPF is a semi distributed Deterministic, continuous and physically-Based model that can simulate the hydrologic, associated water quality and quantity, processes on pervious and impervious land surfaces and streams. ANN is probably the most successful machine learning technique with flexible mathematical structure which is capable of identifying complex non-linear relationships between input and output data without attempting to reach understanding as to the nature of the phenomena. Statistical approach depending on cross-, auto- and partial-autocorrelation of the observed data is used as a good alternative to the trial and error method in identifying model inputs. The performance of ANN and HSPF models in calibration and testing stages are compared with the observed runoff values to identify the best fit forecasting model based upon a number of selected performance criteria. Results of runoff simulation indicate that simulated runoff by ANN were generally closer to observed values than those predicted by HSPF.