The fallacy in the use of the “best-fit” solution in hydrologic modeling

Using the parameters associated with best-fit simulation (i.e., highest objective function value) to represent a calibrated hydrological model is inadequate. The reason that models best value usually not significantly different from next or values after that. This non-uniqueness of causes problem because solution's are often set parameters. Therefore, only using as model's sole interpret watershed processes perform further modeling analyses could produce misleading results. Furthermore, lack pristine watersheds makes task watershed-scale calibration increasingly challenging. Subjective thresholds acceptable performance criteria suggested by some researchers, based on comparing measured and solution signals, achievable. Hence, obtain satisfactory fit, researchers practitioners forced compromise science behind their work. article discusses fallacy in hydrologic modeling. A two-factor statistic assess goodness calibration/validation discussed, considering output uncertainty.