摘要：The lack of measured hydrological data is a problem in most of the developing countries resulted from inadequate gauging stations, short length of record period and relatively low accuracy of measurement. Due to the above mentioned problem, it is important to develop models which can carry out acceptable simulation of run-off behavior especially for the ungauged catchents. This research was designed to evaluate HEC-HMS model for prediction of flood and modeling of rainfall-run off process. After calibration of the model, the hyetograph and related hydrographs of 6 rainfall events were used and the observed and estimated hydrographs were compared from different points of view. It was observed that Curve Number (CN) and initial loss are the main parameters affecting the results. Another evaluation was the comparison of the lag time produced by Snyder and SCS methods, showing more accuracy of the lag time predicted by SCS approach. In this study the common hypothesis "initial loss is about 0.2 of S" was also evaluated and observed that in Toroq watershed, the value of 0.22S is more reliable for the initial loss. About the comparison of the estimated run off produced by SCS and Snyder approaches, it was seen that the results of SCS approach is more close to the measured values.
