The Role of Objective Functions in Assessment of Water Balance Components Using SUFI-2 Algorithm in Semi-arid Basin

Abstract

The Soil and Water Assessment Tool calibration and uncertainty program (SWAT-CUP) is equipped with many algorithms. The sequential uncertainty fitting 2 (SUFI-2) algorithm is one of the algorithms which is used in the Dharoi catchment of Sabarmati basin, India, for calibrating and validating the SWAT hydrologic model by considering seven objective functions. The different objective functions are examined to understand their influence on the results obtained during calibration, the best-fitted parameter values, and water balance components estimation. During calibration and validation processes, statistical performance indices of all the objective functions are found to give satisfactory results. The best-fitted parameter values using different objective function are different, except in case of Nash–Sutcliffe efficiency (NSE) and ratio of root mean square error and the standard deviation of observations (RSR). Due to same-fitted value generated by using NSE and RSR, the obtained water balance components, i.e., surface runoff (Qs), lateral flow (Lat), evapotranspiration (Ea), and percolation (Perc) are also found to be the same. The range of 45–59% (20–30%) (11–14%) evapotranspiration (surface runoff) (percolation) using different objective functions is estimated for the present area. It is concluded that the water balance estimated given by NSE and KGE is relatively reasonable than other objective functions. This analysis would help in selecting the objective functions used to produce predictive hydrological model during calibration in SWAT-CUP using SUFI-2 algorithm. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.