Journal of Watershed Management Research

Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

   The accurate determination of river flow in watersheds without sufficient data is one of the major challenges in hydrology. In this regard, given the diversity of existing hydrological models, selection of an appropriate model requires evaluation of the performance of the hydrological models in each region. The objective of this study was to compare the performance of artificial neural network (ANN) and IHACRES integrated model to predict the flow of sivand river in Tashak Bakhtegan watershed located in Fars province as a warm and arid area. Calibration and validation procedures were done by using data from 1982-1995 and 1996- 2012, respectively. Neural Network Toolbox of MATLAB software were used to evaluate the capabilities of neural networks. In both calibration and validation periods, simulated flows by the IHACRES model for flood flows, were less than the observed data. The determination coefficients of the model during calibration and validation were 0.62 and 0.54, respectively. The determination coefficients of dynamic neural networks and static neural networks during calibration and validation ranged from 0.88- 0.94 to 0.51- 0.69, respectively. The results demonstrated that artificial neural networks predicted monthly flow of sivand river more accurately than the IHACRES model.

Keywords: Watershed, Flood, Static networks, Dynamic networks

Full-Text [PDF 823 kb]   (1063 Downloads)    

Add your comments about this article : Your username or Email:

---

---