Evaluation of DSSAT-CROPGRO-Canola Model to Simulate Growth and Yield of Two Canola (Brassica napus L.) Cultivars in Karaj

IntroductionCanola (Brassica napus L.) is known as the third most important oil crop in the world and is now cultivated over a large area of the world's farms in rotation with various crops, especially cereals (Reddy and Redi, 2003). Simulation models are a useful tool for predicting crop responses to different environments. The CSM-CROPGRO model (Jones et al., 2003) was integrated into the Decision Support System for Agrotechnology Transfer (DSSAT) for simulating spring rapeseed (Saseendran et al., 2010). Due to limited studies on simulating the growth and yield of rapeseed in Iran, especially using DSSAT models, the purpose of this study was to calibrate and evaluate the DSSAT-CROPGRO-Canola model for simulating the growth and yield of two canola cultivars with different treatments of planting date and nitrogen in Karaj, Iran.Materials and MethodsA field experiment was performed as a split-plot factorial based on a randomized complete block design with three replications in 2017 and 2018. Two spring canola cultivars (Dalgan and Hyola-420) were planted under three levels of nitrogen (0, 70, and 210 kg.ha-1) on two planting dates (28 Feb and 19 Mar). Planting date was considered as the main factor, and cultivars and nitrogen levels were considered as sub-factors. Measured data during the growing season were leaf area index (LAI), total dry matter (TDM), yield and yield components, and dates of flowering and physiological maturity. Daily weather data, management events, and soil characteristics are imported to DSSAT. The first-year experimental data were used for calibration, and second-year data were used for model evaluation of developmental stages, LAI, TDM, and grain yield. The performance of the DSSAT-CROPGRO-Canola model during the calibration and evaluation was assessed using different statistics, root mean square error (RMSE), normalized RMSE (nRMSE), Willmott’s index (d), and coefficient of determination (R2) of a 1:1 regression line.Results and DiscussionThe results of evaluating phenological stages (anthesis day, first pod day, first seed day, and physiological maturity day) showed that the RMSE for the Dalgan cultivar was less than four days, and for the Hyola-420 cultivar, it was less than five days. This indicates that the model performed excellently in accurately simulating developmental stages. The model was able to simulate LAI up to the pod formation stage in different treatments. The nRMSE and d were 24.88% and 0.92 for the Dalgan cultivar and 22.72% and 0.95 for the Hyola-420 cultivar, respectively.The model was also able to simulate the total dry matter at different planting dates as well as different levels of nitrogen fertilizer, and the values of nRMSE, d, and R2 for the Dalgan cultivar were 24.97%, 0.97 and 0.91**. For the Hyola-420 cultivar, the values were 22.73%, 0.98, and 0.94**. Additionally, the nRMSE, d, and R2 values for the number of grains per square meter were 14.97%, 0.98, and 0.91** for the Dalgan cultivar and 15.37%, 0.98, and 0.90** for the Hyola-420 cultivar, respectively.The evaluation results for grain yield of canola cultivars showed that the RMSE was 395 and 265 kg.ha-1, d was 0.97, and R2 was 0.89** and 0.91** for Dalgan and Hyola-420 cultivars, respectively, confirming the high accuracy of the calibration. Therefore, this model can be used to evaluate the different effects of crop management and make decisions in canola cultivation systems. One of these decisions is to determine the best planting date for spring canola cultivars in the region. Based on the long-term model simulation of cultivars in different planting dates, it is recommended to plant spring canola up to 11 March in this region.ConclusionThe results of this study showed that the DSSAT-CROPGRO-Canola model had reliably good performance under different management and environmental conditions. CSM-CROPGRO-Canola model predicts grain yield responses to management and environmental conditions well and can now be employed for assessing the impacts of various agronomic management strategies and decisions making in canola production systems in Iran.