Unit hydrographs via nonlinear programing

A nonlinear programing model is developed for the optimal determination of unit hydrographs. The model determines both the rainfall losses for each rainfall event and the best unit hydrograph that minimizes deviations between observed and derived runoff hydrographs. Several multiperiod rainfall events can be considered simultaneously for the unit hydrograph determination. A nonlinear programing technique called the large‐scale generalized reduced gradient is used to solve the model. Several example applications are presented to illustrate the model.