Evaluation of SWAP Model for Simulating Soybean Yield, Yield Components and Water Productivity

In this research, SWAP model was evaluated to estimate soybean grain yield, biological yield, leaf area index, water productivity, and available soil water content in soil profile. The sensitivity of SWAP model was analyzed based on the field results obtained from soybean planting with four furrow irrigation treatments including full irrigation, conventional deficit irrigation at 75% and 50% soil moisture deficit compensation, and partial root zone drying at 50% soil moisture deficit compensation in 2008 agronomical year. Then it was calibrated, validated and evaluated based on the field results of the agronomical year 2009. The results of this study indicated that this model simulated grain yield better than biological yield. The best and the worst simulations of soybean water productivity were obtained at partial root zone drying and conventional deficit irrigation treatments at 50% soil moisture deficit compensation, respectively. Based on calculated statistical indices in this research, the best simulations of the model that had the best fitting with the measured data, were at grain yield, water productivity, leaf area index, biological yield, and volumetric soil water content (the least RMSE and ME and the highest R²), respectively. Also the results of the model sensitivity analysis showed that SWAP model was sensitive to soil input data, i.e. residual moisture and saturated hydraulic conductivity, and with any little change in these input data, output results (the amounts of simulated parameters) varied greatly.