Uncertainty Analysis in Prediction of Soil Water Balance Components in two Irrigated Field in Arid Region

Abstract

Uncertainty analysis is a useful tool to evaluate soil water simulations in order to get more information about the models output. These information provide more confidence for decision making processes. In this study, SWAP model is applied for soil water balance simulations in two fields which are planted by wheat and maize in an arid region. First the amount of uncertainty is estimated and compared for soil moisture simulation by using Generalized Likelihood Uncertainty Estimation (GLUE) in the two fields. Then based on the computed parameter uncertainty, the effect of uncertainty in soil moisture simulation is evaluated on soil water balance components. Results indicated that in arid regions with irrigated agricultural fields, prediction of actual evapotranspiration is relatively precise and the coefficient of variation for the two fields are less than 4%. Moreover, the prediction of deep percolation for the two fields are influenced by the uncertain hydraulic conductivity and showed lower precision according to the actual evapotranspiration.

Keywords

• Calibration
• SWAP
• Actual evapotranspiration
• Deep percolation