Simulation of Changes in Soil and Plant Nitrogen by CERES-Wheat Model

The DSSAT model is one of the most general and extensively used process-based crop growth simulation models. This model has been used worldwide to simulate crop biomass, yield, and soil
nitrogenleaching under different management practices and various climatic conditions. Among management agronomic factors, nitrogen fertilizer has a major effect on crops production. However, nitrogen fertilizer limiting causes to decrease crops production, but, high application rates of nitrogen would led to strong environmental consequences. Thus, management of nitrogen fertilizer consumption causes to decrease environmental pollution in the agroecosystems. Therefore, the objectives of the present study were: (1) determination of genetic coefficients and calibration of the CERES-Wheat modelof DSSAT-CSM, (2) evaluation the performances of model forsimulating wheat growth, development and grain yield and (3) simulationof changesof soil and plant nitrogen in different fertilizer nitrogen application rates under Kermanshah climate condition. Two experiments were established based on the randomized complete block design with three replications during 2015-2016. The treatments were included 4 levels of nitrogen fertilizer application
(90, 180, 300 and 360 kg ha-1 urea). The required model inputs describe field management, daily weather condition, soil profile characteristics, and cultivar characteristics. The cultivar coefficients calibrated under optimum conditions (i.e., minimum stress in weather and nutrients). The genetic coefficients (P1V, P1D, P5, G1, G2, G3 and PHINT) of the Pishtaz wheat cultivar were derived using the GenCal software of DSSAT v 4.6 for 300 kg Urea ha-1 treatment (optimum condition of nitrogen fertilizer based on the results of soil library). After model calibration process, the CERES-Wheat model validated by comparing simulated and measured values of wheat cultivars phenologicaldevelopment stages (DVS), leaf area index, total dry weight and grain yield for treatments of 90, 180, 300 and 360 kg Urea ha-1 fertilizer by root mean square error (RMSE), normalized RMSE (nRMSE) and index of agreement (d) by results ofan independent experiment from calibration experiment. The results indicated that the coefficient P1V was 54.45 °C day, the coefficient P1D was set 90.75 days hr-1, the value for P5 was 720 °C day, the value for G1 was 25, the values for G2 was 30
mg day-1, the value for G3 2 g, and the PHINT was 95°C day. The calibration results showed that the CERESWheat model was able to simulate growth, development stages and yield correctly, which indicate high accuracy in calculated genetic coefficients derived using the GenCal software of DSSAT v 4.6. In the simulated and measured conditions, leaf area index, total dry weight and grain yield improved by increasing of nitrogen fertilizer application. In the simulated and observed conditions, the highest grain yields were 7048 and 7874 kg ha-1 in the treatment of 360 kgnitrogen ha-1 and the lowest grain yields were 4006 and 4217 kg ha-1 in the treatment of 360 kgnitrogen ha-1, respectively. The validation results also indicated that the CERES-Wheat model had high ability to predictg growth, development stages and grain yield in the different fertilizer nitrogen application rates.
So that, the RMSE fordevelopment stages were about 3 to 4 days and the nRMSEwere about 7 to 8% of measured average, respectively. The index of agreement (d) for development stages was about 0.99. The RMSE for total dry weight were about 360 to 720 kg ha-1 and the nRMSE were about 5% to 9% of measured average, respectively. The index of agreement (d) for total dry weight were about 0.94 to 0.99. The amount RMSE for grain yield were 304 to 630 kg ha-1 and the nRMSE were 11% to 17% of measured average, respectively. The index of agreement (d) for grain yield ranged from 0.98 to 0.99. The simulation result also indicated that amount of soil NO3 and NH4 increased with nitrogen fertilizer application. The highestsoil NO3 were 41.3, 54.5, 72.1 and 80.9 kg ha-1 in the treatments of 90, 180, 300 and 360 kg Urea ha-1, respectively. The amount of nitrogen leaching increased with rising of nitrogen fertilizer. The nitrogen leaching were 259.3, 276.2, 310.4 and 335.5 kg ha-1 in the treatments of 90, 180, 300 and 360 kg Urea ha-1, respectively. The amount of nitrogen in the wheat biomass improved by increasing nitrogen fertilizer application. The results indicated that the CERES-Wheat calibrated correctly that confirm calculated genetic coefficient for Pishtaz cultivar under Kermanshah climate conditions. The results of validation also showed that the CERES-Wheat model was able to simulate all studied traits wheat cultivars except leaf area index accurately in different fertilizer nitrogen application rates. Excessive nitrogen consumption led to nitrogen leaching and groundwater pollution. Therefore, it is important to know the distribution of various forms of nitrogen and how they move in the soil.