Adapting CROPGRO for Simulating Canola Growth with Both RZWQM2 and DSSAT.

Currently, canola (Brassica napus L.) is gaining importance as a potential feedstock in biodiesel production industries, and high demand for canola requires more acreage for canola production. Agricultural system models may help to assess the feasibility of canola production under various agroclimatic conditions.  In this study, we adapted the CROPGRO model for simulation of canola in both RZWQM2 and DSSAT 4.0. Soil water, leaf area index, biomass, plant height, phenology  and grain yield data from irrigation experiments conducted in 2005 on a Weld silt loam soil in the semi-arid climate at Akron, Colorado were used for model parameterization and calibration. Similar data for 1993, 1994 and 2006 for validation and further improvement. Species and cultivar parameters for the model were developed using data from literature or by calibrating CROPGRO-faba bean parameters. Grain yields across various irrigation levels and seasons were simulated reasonably well by RZWQM2 with RMSE of 215 kg ha^-1 and index of agreement (d) of 0.98. Seasonal biomass development was simulated with RMSEs between 341 and 903 kg ha^-1, d between 0.55 and 0.99, and R² between 0.85 and 0.98.  The CROPGRO-canola parameters developed were also tested within the DSSAT 4.0 cropping systems model and found to produce results with similar accuracy. Results of the study indicate that the canola model developed is sufficient for predicting canola production under semi-arid conditions.