ARS | Publication request: Calibrating RZWQM2 model for maize responses to deficit irrigation

Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: November 6, 2011
Publication Date: December 9, 2011
Citation: Ma, L., Trout, T.J., Ahuja, L.R., Bausch, W.C., Saseendran, S.A., Malone, R.W., Nielsen, D.C. 2011. Calibrating RZWQM2 model for maize responses to deficit irrigation. Agricultural Water Management. 103 (2012):140-149.

Interpretive Summary: A well calibrated system model should help extend field research beyond the experimental location. In this study, the Root Zone Water Quality Model-DSSAT (RZWQM2) was used for simulating plant water stresses in corn in Eastern Colorado. The experiments were conducted in 2008 and 2009, in which the corn was irrigated to meet a certain percentage (100%, 85%, 70%, 55%, and 40%) of the potential crop evapotranspiration demand during a growing season. Simulated results showed that field estimated soil hydraulic properties provided better model responses to irrigation than laboratory measured ones, and manual calibrated plant parameters were very close to those obtained with automated optimization. The results demonstrated that there were multiple sets of plant parameters that achieved acceptable simulation results when only one treatment was used for model calibration. It is necessary to include multiple treatments and multiple years of data to decrease the degree of freedom in model parameterization, so that the calibrated model can be used in other locations.

Technical Abstract: Calibrating a system model for field research is a challenge and requires collaboration between modelers and experimentalists. In this study, the Root Zone Water Quality Model-DSSAT (RZWQM2) was used for simulating plant water stresses in corn in Eastern Colorado. The experiments were conducted in 2008 and 2009, in which the corn was irrigated to meet a certain percentage (100%, 85%, 70%, 55%, and 40%) of the potential crop evapotranspiration (PETc) demand during a growing season. The model was calibrated with both laboratory measured or field estimated soil water retention curves (SWRC) and evaluated for yield, biomass, leaf area index (LAI), and soil moisture under six irrigation treatments in 2008 and 2009. Simulated results showed that field estimated SWRC provided better model responses to irrigation than laboratory measured SWRC, and manual calibrated plant parameters were very close to those obtained with automated optimization. The results also demonstrated that there were multiple sets of plant parameters that achieved acceptable simulation when only one irrigation treatment was used for calibration. Therefore, it is necessary to include multiple treatments and multiple years of data to decrease the degree of freedom in parameterization. More detailed measurements of plant growth during a growing season should also help the calibration process. Standard methods of obtaining soil hydraulic properties, estimating PET, and initializing soil carbon pools should also be used if calibrating plant parameters are to be used under different soil and weather conditions.