Rainfall-Runoff Modelling by Artificial Neural Network in Alkabeer Aljonobee Catchment


Abstract in English

The relation between rainfall and runoff forms one of the main hydrological cycle elements. It is one of the most complex hydrological phenomena because of the great numbers of parameters used in modeling the physical processes, the expansion of their parameter space, and the temporary change in watershed specifications. Thus, modeling the relation between rainfall and runoff is necessary for hydrological and hydraulic engineering design, integrated management of water resourses, and forecasting flood and preventing its dangers. This research aims at modeling the relation between rainfall and runoff in Alkabeer Aljononbee catchment. It depends on the technique of Artificial Neural Network (ANN). The mathematical model was built by the ntstool and nntool available in the Matlab program. This model depends on daily rainfall, evaporation, air temperature, and relative humidity data taken from meteorological stations that are distributed in the watershed. The daily runoff data have also been used for checking the performance accuracy of the network, using the Simulink technique. The results of this research confirm that artificial neural network technology offers good results in modeling the relation rainfall-runoff, depending on the set of data used. So it could be a better alternative than traditional approaches.

References used

Rientjes, T. H. M.؛ De Vos, N. J. Constraints of artificial neural networks for rainfallrunoff modelling: trade-offs in hydrological state representation and model evaluation. Hydrology and Earth System Sciences Discussions Belgium, 2005, 365– 415
Abbott, M. B.؛ Bathurst, J. C.؛ Cunge, J. A.؛ O’Connell, P. E.؛ Rasmussen, J. An introduction to the european hydrological system – Syste`me Hydrologique Europe´en, “SHE”, 2: Structure of a physically-based, distributed modelling system. 1986, 61–77
Nash, J. E.؛ Sutcliffe, J. V. River flow forecasting through conceptual models. Journal of Hydrology1970, 282–290
Modarres, R. Multi-criteria validation of artificial neural network rainfall-runoff modeling. Hydrologu & Earth System Science Iran, 2009, 411-421

Download