小麦氮素积累动态的高光谱监测
冯伟, 朱艳, 姚霞, 田永超, 庄森, 曹卫星
(南京农业大学/江苏省信息农业高技术研究重点实验室/农业部作物生长调控重点开放实验室, 南京 210095)
摘 要: 【目的】研究小麦地上部氮积累量与冠层高光谱参数的定量关系，分析多种高光谱参数估算地上部氮积累量的效果。【方法】连续3年采用不同蛋白质含量的小麦品种在不同施氮水平下进行大田试验，于小麦不同生育期采集田间冠层高光谱数据并测定植株不同器官生物量和氮含量。【结果】植株氮积累量随着施氮水平的提高而增加，不同地力水平间存在明显差异。植株氮积累量的光谱敏感波段主要存在于近红外平台和可见光区，而地上部氮积累量与冠层光谱的相关性明显降低。对植株氮积累量的光谱估算，在不同品种、氮素水平、生育时期和年度间可以使用统一的光谱模型。在籽粒灌浆期间植株氮积累量自开花期随时间进程的积分累积值与对应时期籽粒氮素积累状况存在显著的定量关系，根据特征光谱参数、植株氮素营养、籽粒氮积累量这一技术路径，以植株氮积累量为交接点将模型链接，建立高光谱参数与籽粒氮积累量间定量方程。将植株氮积累量与籽粒氮积累量相加，确立了基于高光谱参数的籽粒灌浆期间地上部氮积累量监测模型。经不同年际独立资料的检验表明，利用光谱参数SDr/SDb、VOG2、VOG3、RVI（810,560）、[（R750-800）/（R695-740）]-1和Dr/Db建立模型可以实时监测小麦地上部氮素积累动态变化，预测精度R2分别为0.774、0.791、0.803、0.803、0.802和0.778，相对误差RE分别为16.7%、15.5%、15.6%、18.5%、15.5%和17.3%。【结论】利用关键特征光谱参数可以有效地评价小麦地上部氮素积累状况，其中尤以植被指数VOG2、VOG3和[（R750-800）/（R695-740）]-1的效果更好。
关键词: 小麦; 高光谱遥感; 氮素积累; 监测模型
Monitoring Plant Nitrogen Accumulation Dynamics with Hyperspectral Remote Sensing in Wheat
FENG Wei; ZHU Yan; YAO Xia; TIAN Yong-chao; ZHUANG Sen; CAO Wei-xing
(Jiangsu Key Laboratory for Information Agriculture Nanjing Agricultural University/ Key Laboratory of Crop Growth Regulation Ministry of Agriculture, Nanjing 210095)
Abstract: 【Objective】This study aimed at identification of the quantitative relationship between plant nitrogen accumulation (PNA) and canopy reflectance spectra in winter wheat (Triticum aestivum L.)【.Method】Three field experiments with different wheat varieties and nitrogen levels were conducted. The time-course measurements were taken on canopy hyper spectral reflectance and weights and nitrogen contents in different plant parts during the experimental period. 【Result】Nitrogen accumulation in wheat plant increased with increasing nitrogen fertilization rates, with significant differences among growing seasons. The bands sensitive to PNA occurred during visible light and near-infrared region mostly, and correlation decreased between canopy reflectance and above-ground nitrogen accumulation (ANA). The regression analyses between vegetation indices and PNA indicated that several key spectral parameters could accurately estimate the changes in plant nitrogen status across different growth stages, nitrogen levels and growing seasons, with same spectral parameters for each wheat cultivar. The cumulative value of PNA from an thesis to specific day were highly correlated with grain nitrogen accumulation (GNA) at corresponding day, with the determination of coefficient (R2) as 0.883 and standard error (SE) of 1.736 from linear equation, respectively. Based on the technical route of key spectral parameters-plant nitrogen nutrition index-GNA, estimating models of GNA were constructed on the basis of canopy hyper-spectral parameters by linking the above two models with plant nitrogen nutrition as intersection in wheat. Total monitoring models on ANA during filling period were established using canopy hyper-spectral parameters by adding GNA to PNA. Tests with other independent dataset showed that several key spectral indices such as SDr/SDb, VOG2, VOG3, RVI (810, 560), [(R750~800)/(R695~740)]-1 and Dr/Db could be used to predict ANA. 【Conclusion】It can be concluded that ANA in wheat could be monitored directly by key vegetation indices, with more reliable estimation from VOG2, VOG3 and [(R750~800)/(R695~740)]-1.
Key words: Wheat; Hyper spectral remote sensing; Nitrogen accumulation; Monitoring model
基金: 国家自然科学基金(30671215);; 国家 863 计划项目(2006AA10Z202,2006AA10Z271);; 高校博士点基金项目(20060307031)
作者简介：冯伟(1976-)，男，河南人，博士，研究力向为作物生态与遥感监测。Tel: 025-84396242; E-mail: fengwei@126.com。通讯作者曹卫星 (1958-)，男，江苏南通人，教授，研究方向为作物生态与信息农学。Tel: 025-84396565; Fax: 025-84432420; E-mail: caowtinjan.edn.cn
文献出处：中国农业科学,第41卷(7) ,2008.7:1937-1946