Modeling evapotranspiration for irrigation water management in a humid climate

摘要

Quantifying evapotranspiration (ET, consumptive crop water requirement) is critical to managing limited water resources for crop irrigations. Agricultural system simulation models that realistically simulate the ET processes are potential tools for integration, synthesis, and extrapolation of location-specific water management research data across soils and climates for limited-water management in agriculture. The objective of this investigation was to evaluate the accuracy of the Root Zone Water Quality Model v2.0 (RZWQM2) simulated ET against ET measured in corn, soybean, and cotton cropping systems in a predominantly clay soil under humid climate in the Lower Mississippi (MS) Delta, USA, in 2016, 2017, and 2018. Energy balance (EB) and eddy covariance (EC) methods were used for measuring ET. The RZWQM2 parameters calibrated in previous studies at the location were used in the simulations. Potential evapotranspiration (PET) in the model was simulated using an extended approach based on the Shuttleworth and Wallace (SW) model. Water infiltration into the soil was simulated using the Green and Ampt approach, and its further movement in soil layers and contributions to soil evaporation using Richard's equation. Across the three crops and their crop-seasons, simulated daily ET deviated from EC and EB estimates with RMSEs between 0.09 and 0.14 cm and RRMSEs between 21 and 37%. On a weekly basis, accuracies in simulated ET (ETS) improved significantly with RRMSEs between 9 and 17%, and on a seasonal basis RRMSEs were between -9 and 11%. The imbalance in incoming and outgoing energies accounted in the EC system varied between 2 to 12%; taking this uncertainty in estimated ET into account, the accuracies in weekly and seasonal ET simulations were reasonable for their use in irrigation management at these time-scales.