The aim of this study is to present an inverse modelling approach with a genetic algorithm for estimating soil hydraulic and solute transport parameters in two subsurface drainage systems in Iran. In this study, measured data were obtained from Amirkabir and Shoaybiyeh sugarcane fields equipped with subsurface drainage systems. The SWAP model was used for simulating the outputs of subsurface drainage systems. The accuracy of different objective functions which were based on the discrepancies between measured and simulated values of drainage discharge, groundwater depth and drainage water salinity was evaluated in the inverse modelling approach. Based on sensitivity analysis of the SWAP model in studied areas, n shape parameter, lateral hydraulic conductivity, depth to impermeable layer, saturated water content and α shape parameter were selected in the inverse modelling approach. By applying the objective function which is based on drainage discharge and salinity in the Amirkabir study area, the Nash–Sutcliffe Efficiency (NSE) values for predicting drainage discharge and salinity were 0.63 and 0.79, respectively. In the Shoaybiyeh study area, the objective function which includes drainage discharge, groundwater depth and drainage water salinity resulted in NSE values of 0.83, 0.95 and 0.89 for predicting drainage discharge, groundwater depth and drainage water salinity, respectively. Copyright © 2017 John Wiley & Sons, Ltd.