Evaluation of CERES-Wheat Under Variable Plant Population, Nitrogen, and Irrigation Treatments in Arizona.

Crop growth models can be used as a decision support tool in the management of agricultural systems.  In Arizona and other arid environments, there is potential to use crop growth simulations to monitor evapotranspiration and improve the scheduling of irrigations.  In this work, CERES-Wheat was evaluated for two plot-level wheat experiments, grown at Maricopa, Arizona during the winter of 2003-2004 and 2004-2005.  Experimental treatments included two methods for irrigation scheduling based on crop coefficients (FAO-56 and NDVI), two nitrogen application rates (111 and 241 kg N ha^-1 yr^-1), and three plant populations (75, 150, and 300 plant m^-2).  Neutron probes were used for bi-weekly measurements of soil water contents in each treatment plot, and detailed agronomic measurements were collected to characterize crop growth, development, and yield during each growing season.  Using this dataset, the CERES-Wheat model was calibrated and evaluated for simulating wheat growth under these treatment conditions.  Preliminary results suggest that the model underestimates potential evapotranspiration in this arid environment, which leads to an underestimate of soil water depletion.  It will be important to address this issue before applying the model for irrigation scheduling in Arizona.