Application of Rainfall-runoff Models to Zard River Catchment's

Rainfall-runoff models are nonlinear processes according to the sequential and spatial distribution of the rainfall. So, it is difficult to explain the response of catchments systems with the simple models. In the present work simulation of the rainfall-runoff processes have been carried out by the Artificial Neural Networks (ANN) and the HEC-HMS models. The ANN models of Multi Layer Perceptron (MLP) with two hidden layers and Radial Basis Function (RBF), were used to simulate this process. It has been applied to the Zard river basin in Khuzestan province using daily rainfall and runoff data, during the period of 1991 to 2000. During this period, 14 flood events were selected to simulate rainfall-runoff processes by the HEC-HMS model. Results of two models were compared with the observed data of Zard river basin. It is shown that RBF model is much better than, MLP and HEC-HMS models for simulating of the rainfall-runoff process in Zard river basin.