Journal of Environmental Hydrology
ISSN 1058-3912


Electronic Journal of the International Association for Environmental Hydrology

JEH Volume 19 (2011), Paper 5    Posted February 28, 2011
NEURAL NETWORK APPLICATION FOR MONTHLY PRECIPITATION DATA RECONSTRUCTION

Zohre Khorsandi1
Mohammad Mahdavi2
Ali Salajeghe2
Saeid Eslamian3

1Department of Natural Resources, Science and Research Branch, Islamic Azad University, Tehran, Iran
2Department of Natural Resources, Tehran University, Karaj, Iran
3Department of Water Engineering, College of Agriculture, Isfahan University of Technology, Iran

ABSTRACT
The need for precipitation data and the importance of its duration in hydrological and climatic phenomena motivate investigators to develop reconstruction methods. Four methods are artificial neural network, normal ratio, inverse distance weighting, and geographical coordinate. These methods are compared in this study using monthly precipitation data of three stations, Dolat-Abad, Kabootar-Abad and Refinery plant around the city of Esfahan, Iran. Mean absolute error and coefficient of correlation of the results are compared to select the most appropriate method. The neural network approach shows the best performance compared with the other methods.

Reference: Khorsandi, Z., M. Mahdavi, A. Salajeghe, and S. Eslamian. 2011. Neural network application for monthly precipitation data reconstruction. Journal of Environmental Hydrology, Vol. 19, Paper 5.
CONTACT:
Zohre Khorsandi
Department of Natural Resources
Science and Research Branch
Islamic Azad University
Tehran, Iran

E-mail: khorsandi_zohra@yahoo.com



Return to JEH 2011 Papers

Return to HydroWeb Homepage