Simulating hydrologic cycle and crop production in a subsurface drained and sub-irrigated field in Southern Quebec using RZWQM2

Abstract

Agricultural system models are promising tools in evaluating the agro-environmental effects of water management practices. However, very few models have been tested using a comprehensive hydrologic data set. The present study’s objective was to evaluate the hydrologic component of RZWQM2 (Root Zone Water Quality Model) using a comprehensive hydrological dataset including subsurface tile drainage, subirrigation, soil water content, sap flow and crop growth data such as leaf area index, crop yield and crop growth stages. Drawing on 2008 and 2009 data from a farm site in Southern Quebec, the RZWQM2 model showed accurate simulation in soil water content, sap flow, growth stage, leaf area index, and crop yield. While mean values for growing season tile flow under both free drainage (FD) and controlled drainage with subirrigation (CD-SI) were reasonably accurate, winter tile flow was significantly overestimated, indicating RZWQM2’s reliability to be compromised by its imperfect winter drainage process. Accordingly, a Kalman filter technique was applied to enhance model reliability and reduce predictive uncertainties. A novel RZWQM2 model equipped with a Kalman filter algorithm adequately simulated, in both calibration and validation phases, the hydrology and corn growth which occurred under both FD and CD-SI systems at the selected field site. Simulation results suggest that RZWQM2 model can be used for water management under subsurface drained and irrigated field and the Kalman filter technique significantly improved the accuracy of RZWQM2 model in simulating winter drainage in cold areas.