Determining optimum nitrogen management as a function of planting date for spring wheat (Triticum aestivum L.) under semi-arid conditions using a modeling approach

A crop modeling approach can be used to assess various crop management options and water productivity to improve crop production for different environments. The goal of this study was to determine optimum planting dates and nitrogen management options using a systems analysis approach based on the Cropping System Model (CSM) CERES-Wheat of DSSAT v 4.6. Wheat experiments were conducted from 2007 to 2013 under semi-arid conditions at the experimental fields of the Nuclear Institute for Agriculture and Biology (NIAB), Faisalabad, Pakistan. The CSM-CERES-Wheat model was calibrated for wheat (Triticum aestivum L.) cultivar Sehar-2006 from a 3-year field experiment data on flowering day, maturity day, canopy cover, grain yield, biomass, grain nitrogen content, and nitrogen harvest index. After calibration, the CSM-CERES-Wheat model produced satisfactory simulations for wheat phenology and crop growth parameters. The model was then evaluated using six years of independent datasets for both irrigated and rainfed conditions. The evaluation showed that the model performed well as indicated by the accurate simulation of wheat phenology [the normalized root mean square error (NRMSE) = 3%] and crop growth parameters such as crop cover (NRMSE = 7%-13%), biomass (NRMSE = 17%), and grain yield (NRMSE = 9%) against measured data. The model's performance was less satisfactory for biomass production under high soil moisture or rainfed conditions. Focusing on differences in temperature and rainfall patterns during the potential growing season, the model was used to estimate the optimum planting dates and in-season nitrogen management options using 39 years (1974-2013) of historical weather data for short term adaptation against climate variability. The scenario simulations for planting dates showed that growing wheat under rainfed conditions at the study region is not a viable option. However, based on the simulation results, we concluded that early planting from November 1st to 10th for irrigated conditions with applications of 150 kg N ha(-1) in three equal splits with the first at sowing and the rest two splits during vegetative growth phase could result in higher crop yield and water use efficiency. This study showed that the model can be a promising tool for providing crop management recommendations under high-temperature conditions found in the semi-arid conditions.