Rainfall-runoff Prediction Based on Artificial Neural Network (A Case Study: Jarahi Watershed)

The present study aims to utilize an Artificial Neural Network (ANN) to modeling the rainfallrunoff relationship in a catchment area located in a semiarid region of Iran. The paper illustrates the applications of the feed forward back propagation for the rainfall forecasting with various algorithms with performance of multi-layer perceptions. The monthly stream of Jarahi Watershed was analyzed in order to calibrate of the given models. The research explored the capabilities of ANNs and the performance of this tool would be compared to the conventional approaches used for stream flow forecast. Efficiencies of the gradient descent (GDX), conjugate gradient and Levenberg-Marquardt (L-M) training algorithms are compared to improving the computed performances. The monthly hydrometric and climatic data in ANN were ranged from 1969 to 2000. The results extracted from the comparative study indicated that the Artificial Neural Network method is more appropriate and efficient to predict the river runoff than classical regression model.