Improvement of Estimation of Flood Hydrograph Using Modified Curve Number (non-linear Ia-S) Model

The Curve Number Model (SCS-CN) is in conventional mode is based on the linear relationship between initial absorption (Ia) and potential maximum retention (S) of the catchment but this model has been modified to consider non-linear Ia-S relation. The objective of this study is to compare the conventional curve number and modified curve number (non-linear Ia-S relation) models in flood hydrograph estimation in five Galikesh, Nodeh, Tamer, Vatana and Kechik catchments (37 rainfall-runoff events in calculation and selection of 14 events for results comparison in validation step). The root mean square error (RMSE), Nash-Sutcliff (NSE) and peak discharge estimation error (PEP) criteria were used for results comparison. Investigation of RMSE and NSE and PEP criteria shows that the application of modified curved number model (non-linear Ia-S) in all events of validation step improves the estimations of flood hydrograph and peak discharge in comparison with conventional curve number model (SCS-CN), therefore the results indicated that in studied catchments, the modified curve number model (non-linear Ia-S) has improved the conventional curve number SCS-CN model.