Automatic calibration of semi-distributed conceptual rainfall–runoff model using MOSCEM algorithm

rouhani.hamed@yahoo.com dirk.gorissen@ugent.be ivo.couckuyt@ua.ac.be Jan.Feyen@ees.kuleuven.be Abstract We explore the effectiveness of multi-objective an evolutionary optimization algorithm to calibrate rainfallrunoff model. The Multi-objective Shuffled Complex Evolution Metropolis algorithm is used to find Pareto solutions. Log-Efficiency (LogE) of the observed vs. simulated flows which gives more weight to low flow and Nash and Sutcliffe (1970) defined the coefficient of efficiency (EF) which tends to give stronger emphasis on fitting the higher or peak output values are used as objective functions. The catchment model applied in this study is the Soil and Water Assessment Tool (SWAT). The case study area is Grote Nete basin located in the north-eastern Belgium.