Modeling Impacts Of Nitrogen Fertilization And Irrigation On Winter Wheat Milling Quality,Baking Quality And Nutritional Quality

The research on the effects of different water and nitrogen management on wheat quality can not only provide theoretical basis for high yield and high quality,but also optimize the management and improve the competitiveness of Chinese wheat market.The purpose of this study is to study wheat quality and yield,a major grain crop in northwest China.The related literature was collated,and combined with field experiments under controlled conditions,the main physical,chemical,biological and plant physiological processes in the wheat farmland ecosystem under water and nitrogen management were analyzed and described quantitatively.Field experiments were conducted in the Guanzhong Plain of Shaanxi provice from2014-2016 for 2 wheat growing seasons.The CERES-Wheat model was modified to simulate wheat nutritional quality.The baking quality model,Sirius Quality2 could simulate impacts of nitrogen and water stress on winter wheat grain protein composition.Structural Equation Modeling was used to look for the most important wheat grain morphological features on milling quality.The main results are as follows:（1）The yield components all showed a trend of first increase and then decrease with the increasing of nitrogen fertilizer,and increased with the increasing of ET.With the increasing of nitrogen fertilizer,the milling quality decreased first and then increased,and decreased with the increasing of ET.Grain protein concentration（GPC）,essential amino acid（EAA）and amino acid increased with the increasing of nitrogen application.（2）The observed nitrogen-to-protein conversion factor（Fp）ranged from 5.14 to 5.86,and a certain significant negative linear relationship existed between the minimum of the two stresses(S_MIN)and Fp（P<0.001）.The relative root mean square error（RRMSE）values of Fp and GPC based on an Fp were 0.024 and 0.072,respectively,while the RRMSE of GPC estimated based on an Fp value of 5.70 was 0.098.GPC estimated using a variable Fp value that accounted for coupled water and nitrogen stresses was more precise than GPC estimates based on Fp value of 5.70.（3）The modified CERES-Wheat model was used to optimize water and nitrogen management.The selected final optimal management,with a basal dressing of 262.5 kg hm^-2and 120 mm irrigations at wintering stages,showed strong reliability under different climatic conditions in the Guanzhong Plain.（4）The CERES-Wheat model could accurately simulate winter wheat growth and GPC under different irrigation and nitrogen fertilization managements.While the aboveground and grain biomass were underestimated with severe nitrogen stress,and the relative root mean square error（RRMSE）values between simulated and measured were 10-30%which meant that the simulation results were good or medium.GPC were overestimated with slight nitrogen stress.The RRMSE of GPC was 3.77%,and the simulation results were still excellent.（5）Climate change and irrigation were beneficial to yield for all climatic regions in Shaanxi,and the yield for RCP8.5 was greater than that for RCP4.5.The GPC of the humid and semi-arid regions increased for RCP4.5 and RCP8.5,but that of the sub-humid region decreased.Irrigation responding to climate change had an active effect on GPC in the semi-arid region and a negative effect in the humid and sub-humid regions.（6）The leaf area index（LAI）of treatments without nitrogen fertilization（N0）was underestimated by Sirius Quality2.Additionally,the RRMSE values of LAI,aboveground dry matter,grain dry matter,and protein concentration were all larger than 0.2.However,the simulation results of phenology were excellent because of the set parameters for water and nitrogen stress.（7）Based on Structural Equation Modeling involving a total of 617 independent samples from Northern part of Huang-huai Region,results showed that:（a）test weight had the most significantly effects on flour yield,followed by the hardness index;（b）the effects of moisture content,falling number,and crude protein on flour yield were more the indirect effects by test weight and hardness index;（c）test weight and hardness index could excellently estimate flour yield by multiplicative effect of test weight and hardness index,and which could determine68%of the variation in flour yield.The RRMSE of regressive equation was 0.012 which was less than 0.1.The simulated result was excellent,and regressive equation could be used to predict flour yield.