中国农业科学 ›› 2009, Vol. 42 ›› Issue (11): 3863-3870 .doi: 10.3864/j.issn.0578-1752.2009.11.013

• 耕作栽培·生理生化 • 上一篇    下一篇

大麦籽粒蛋白质含量预测模型

徐寿军,杨恒山,范富,王云,许如根,庄恒扬

  

  1. (内蒙古民族大学农学院)
  • 收稿日期:2009-03-02 修回日期:2009-06-09 出版日期:2009-11-10 发布日期:2009-11-10

A Model for Predicting Grain Protein Content in Barley

XU Shou-jun, YANG Heng-shan, FAN Fu, WANG Yun, XU Ru-gen, ZHUANG Heng-yang
  

  1. (内蒙古民族大学农学院)
  • Received:2009-03-02 Revised:2009-06-09 Online:2009-11-10 Published:2009-11-10

摘要:

【目的】构建大麦蛋白质含量预测模型,为建立大麦生产管理决策支持系统奠定基础。【方法】通过定量分析不同品种和氮肥处理大麦氮素吸收、积累、分配和转移的变化过程,建立了大麦花前氮素积累及分配和花后氮素吸收转移动态模型。模型利用抽穗期植株临界含氮量来表达氮素最大积累量,引入叶片潜在分配指数和茎鞘潜在分配指数2个品种遗传参数来区别不同品种在器官间的氮素分配差异,采用Richards方程来描述大麦花前氮素积累动态变化;采用指数函数方程来描述叶片氮的转移量随叶面积指数的动态变化以及籽粒从土壤中吸收的氮量随干物重的动态变化;采用非线性函数方程描述茎鞘和穗部的氮浓度随生理发育时间的动态变化。利用独立的观测资料对所构建的模型进行了检验。【结果】利用不同品种、氮肥、播期和种植地域试验数据检验模型,结果表明,大麦籽粒蛋白质含量模拟值与观测值的绝对预测误差为0.04%~1.27%,RMSE为0.20%~0.72%。精度良好。【结论】模型将经验性与机理性有机结合,具有较好的可靠性。

关键词: 大麦, 氮积累, 蛋白质, 模拟

Abstract:

【Objective】 Modelling grain protein composition in barley is of significant importance for establishing production management decision support system. 【Method】 Based on time-course observations on the nitrogen accumulation, assignment, absorption and remobilization under various nitrogen rate regimes with different cultivars, a dynamic model was developed to simulate the nitrogen accumulation and assignment before anthesis and the nitrogen absorption and remobilization model after anthesis in barley. The critical nitrogen content in heading period was used to calculate the biggest accumulation quantity of nitrogen. The latent distribution index of leaf and stem, which was the variety heredity parameter, were used to distinguish the difference in nitrogen assignment in different varieties. The dynamic content of the nitrogen accumulation before anthesis could be described with a Richards model. The dynamic content of the nitrogen remobilization in leaf with leaf area index (LAI) could be described with a power model, and the nitrogen absorption of grain from soil with dry weight could be described with a power model. The dynamic content of nitrogen density in stem and spike with physiological development time(PDT) could be described with a nonlinear model. The model was validated with independent experiment data. 【Result】 The results showed that the absolute prediction error ranges for grain protein contents were 0.04%-1.27%, and RMSEs were 0.20%-0.72%. 【Conclusion】 The present model with empirical and mechanism appears to provide a reliable prediction.

Key words: barley, nitrogen accumulation, protein, simulation