An Application of ANFIS for Estimating River Flows: A Case Study of Marboreh River, Lorestan Province, Iran

River flow Simulation is the first and most important step for controlling flood in water resources management. In this study, daily flow data of Marboreh River was evaluated by using Nero-Fuzzy networks. To this end, different combinations of precipitation, temperature and runoff (as input data) were used for daily inflow estimation in the period 1998-2012. Correlation coefficient, root mean square error and Nash Sutcliff were used to evaluate the performance of this method. The results showed that the combination of structures using the intelligent method for estimating Marboreh River flows was quite acceptable. The correlation coefficient, root mean square error and Nash Sutcliff coefficient were 0.970, 0.001 m3/s, 0.928, respectively. To sum up, the results indicated that Neural-Fuzzy Inference has a better capability to estimate the minimum and intermediate values and has a good correlation with the observed values.