【作者】 梁红霞；

【导师】 马友华； 王纪华； 章力干；

【作者基本信息】 安徽农业大学 ， 植物营养学， 2005， 硕士

【摘要】 本研究于 2003-2004 年在北京市小汤山国家精准农业研究示范基地进行,主要以ASD FR2500 便携式地物光谱仪及 SPAD 叶绿素测定仪为光谱数据获取手段,以冬小麦为研究对象,提取冬小麦冠层及叶片光谱信息为目标,进行了基于遥感数据的冬小麦变量施肥技术研究。对冬小麦冠层氮素时空分布特征、利用遥感数据进行冬小麦变量施肥技术的施肥机理进行了分析。建立了基于不同遥感数据的变量施肥遥感诊断模型,并就不同施肥模式对冬小麦籽粒产量、品质以及经济生态效益的影响进行了探讨。最后讨论了不同施肥模式的优劣,建立了适合我国冬小麦农业生产条件的基于高光谱数据的冬小麦变量施肥模型。主要研究结论如下: 1.基于冠层光谱特征的冬小麦变量施肥技术:初步建立了基于遥感数据的变量施肥模型,对利用优化土壤调节植被指数(OSAVI)进行冬小麦变量施肥的可行性、变量施肥经济、生态效益进行了分析,结果显示:冬小麦起身和拔节期的冠层 OSAVI 参数与最终产量之间以及与作物拔节期氮素吸收量之间有极显著的相关性,可以用来指导变量施肥实践。变量施肥处理显著增加了冬小麦籽粒产量和生物量,其空间变异也有了明显的降低,同时其经济效益以及生态效益也有明显的提高,与对照区相比,20-40cm 土层的硝态氮含量明显较低,其氮肥利用率也有显著的提高。但是,对籽粒品质指标的含量却没有明显增加,而且增加了其空间变异。 2. 基于不同叶位 SPAD 值的变量施肥技术:提出了利用倒一叶与倒二叶 SPAD 归一化值进行变量施肥研究,分析了利用 SPAD 进行变量施肥的机理,其对冬小麦籽粒产量,品质,施肥效益以及这种变量施肥处理对冠层光谱及叶色值的影响。结果表明:可以利用倒三叶与倒二叶的比值来分析氮素的丰缺状况,主要以 1 为标准,小于 1,缺氮,大于 1,过量。其 SPAD 值与 OSAVI 之间有很好的相关性,这种施肥模式对籽粒产量和生物量及其空间变异也有明显的增加,平均每公顷可以增收 500 元,土壤中的硝态氮的含量也有明显的降低。对降低地下水污染起到了抑制的作用。籽粒品质指标的含量也没有明显增加,且加大了蛋白质含量的空间变异,这与基于 OSAVI 进行变量施肥的施肥结果相似。 3. 基于土壤养分数据的冬小麦变量施肥技术:这种变量施肥处理的籽粒产量和生物量也是明显高于对照区和传统施肥处理,其产量的最大值与最小值分别出现在传统施肥区和对照区。变量施肥的产投比最高,为 9.77,明显高于传统施肥区,变量施肥可以大幅度提高冬小麦的产出、投入比例,增加农作物经济效益。 4. 基于冠层及叶片光谱数据的冬小麦变量施肥技术施肥模型:建立了适合我国农业生产条件的基于 OSAVI 和 SPAD 测定值的冬小麦变量施肥模型,为我国精准农业更多还原

【Abstract】 The experiment was carried out on the National Precision Agricultural ResearchBasement of Xiao Tanshang , Beijing from 2003 to 2005. This dissertation mainly studiedthe variable rate technology(VRT)through extracting the winter wheat(Triticum aestivumL.) spectrum information from the hyperspectral data and SPAD readings by ASD Fro2500spectroradiometers and SPAD meter(Chlorophyll Meter),puts great emphasis on the studyof canopy N dynamic distributing character and analyzed the fertilizer mechanism ofvariable rate technology based on hyperspectral data,and the timely obtained informationof wheat growth status, especially N status, will help us to apply fertilizer and irrigationmanagement. The study constituted the variable rate diagnostic model based on differentremote sensing data,and discussed the influence of winter wheat grain yield,quality andeconomical and environmental benefits,at last,compared with the advantage anddisadvantage of different fertilizer pattern,educed the optimum variable rate model fittingBeijing agricultural producing conditions based on hyperspectral data anddifferent-position leaf SPAD readings. The results are summarized as below: 1. Variable-Rate Nitrogen Fertilization Based on Remote Sensing Data:The Nfertilization optimization algorithm(NFOA) was developed to determine the needed N rateper 1 hm2 based on predicted yield and specific OSAVI per plot, The NFOA accounts forspatially variable potential yield, early stubble N uptake, and responsiveness of the crop toN input. This study demonstrated that crop reflectance measurements using optical sensorscan be used to set more efficient and profitable fertilization levels. The technologyincreased significantly grain yield and dry matter of the winter wheat,and decreased thevariable coefficients of grain yield and dry matter. The NO3-N concentration decreasedcompared with CK treatment in 20cm to 40cm soil,and improved the nitrogen useefficiency evidently. Variable-rate fertilization reduced soil nitrate concentration,andrestrict the potential of underground water pollution. But the grain quality retained thesame level as CK plots(balance N rates) and C plots(uniform N rates),and increasedthe spatial variety. 2. Variable-Rate Nitrogen technology Based on different site leaf SPAD readings:Thetest studied the VRT using normalized difference SPAD(NDSPAD) value for the firsttime,analyzed the influence of grain field ,grain quality,canopy spectrum anddifferent-position leaf SPAD readings. The research indicated that the ratio of inverse oneand inverse two could be used to estimate the N nutrient status,and A very high correlationhas been found between hand-held chlorophyll meter readings and N concentrations inplant tissue,it can be used to predict the total N absorbed. Compared with the CK plots andC plots,the grain yield and dry matter yield improved by VRT, and can increased 500yuan/hm2, the NO3-N concentration decreased obviously,and restrained the undergroundwater pollution. The protein content increased hardly,on the contrary,the VRT increasedthe spatial variety. 3. Variable-Rate Nitrogen technology Based on soil nutrient data:The technologyincreased significantly grain yield and dry matter of winter wheat,and decreased thevariable coefficients of grain yield and dry matter. The maximum yield and minimum yieldpresented to uniform rate plots and balance rate plots. The value/cost ratio, 6.56, was thehighest among three treatments; the variable rate treatment improved the grain yield output,cost proportion, and increased the economic benefits. 4. Variable rate technology models based on canopy spectrum and SPAD readings onwinter wheat: Set up the N fertilization optimization algorithm (NFOA) based on OSAVIvalue and SPAD readings in Beijing agricultural production conditions. It provided atechno-support for precision agriculture variable rate technology. 5. Selecting the N fertilization optimization algorithm: The results of three VRTshowed that the economic benefit of VRT based on SPAD readings is the best among threeVRT, the paper suggested tha更多还原