Evaluation of Different Global Optimization Methods for Calibration of a Daily Hydrological Model

Runoff simulation is a challenging issue in ungagged basins with missing data. Therefore hydrological modeling involved with new automatic calibration methods should be considered seriously. Since adjusting the models with high parameters manually is a time-consuming and labor-intensive task, automatic calibration techniques are generally required to obtain reasonable estimates of the model parameters. Thus in this research we aim to compare and assess calibrating of a conceptual daily rainfall-runoff model (Hymod) with five global optimization methods (GOMs). As well, two case studies (Karaj and Leaf Rivers) were selected to assess and present the results adopted from runoff modeling with two forcing data, rainfall and evapotranspiration series. A new performance criteria, KGE as a new criterion is decomposition of the widely used Nash–Sutcliffe efficiency (NSE) was applied in this study to analyse the different components that constitute NSE. Results revealed that particle swarm optimization (PSO) and shuffle complex evolution (SCE) may be more efficient and robust significantly and will be able to simulate daily runoff with better performance criteria.