【作者】 李映雪；

【导师】 曹卫星；

【作者基本信息】 南京农业大学 ， 生态学， 2005， 博士

【摘要】 遥感技术在农业上的广泛运用为作物生长特征、氮素营养状况、籽粒产量和品质的实时监测与预报提供了方便快捷的方法，对于信息农业的发展具有重要的理论和实践意义。本研究综合运用作物冠层反射光谱探测技术及小麦农学参数和理化指标测试技术，分析了不同条件下小麦冠层多光谱和高光谱反射特征及其与小麦生长特征、氮素营养和籽粒品质性状的关系，明确了小麦叶面积指数、生物量、氮素状况及籽粒品质指标的敏感光谱特征参量及定量反演模型，并探明小麦植株氮素状况及氮素转运规律与籽粒蛋白质指标的定量关系，建立了基于植株氮素营养光谱监测的小麦籽粒蛋白质预测途径，从而为作物生长的无损监测和精确诊断提供了技术基础。 本研究首次综合运用多光谱和高光谱辐射仪，比较了不同氮素水平、不同生育时期和不同品种条件下两种仪器获得的小麦冠层光谱反射率的变化模式。结果表明，高光谱遥感比多光谱遥感更适合农作物水分的监测。随着施氮水平的提高，冠层反射光谱在近红外反射平台(750-1300nm)的反射率呈上升趋势，而可见光部分反射率则下降。小麦从拔节开始，冠层反射光谱在可见光波段先降低然后升高，以孕穗期反射率最低，而近红外区反射率则表现相反趋势，以开花期为分界先上升然后下降，直到成熟前降为最低。相同氮素水平下不同品种小麦冠层反射光谱的反射率值有所变化，且近红外部分差异较明显。这些研究结果为进一步利用冠层反射光谱监测小麦生长状况、氮素营养状况及籽粒品质提供了理论基础。 在分析不同氮素水平下小麦叶面积指数和干物重随生育期变化模式的基础上，讨论了LAI和叶茎干重与冠层多光谱和高光谱反射率及光谱参数的相关关系，提出了小麦LAI和叶茎干重的敏感光谱参数及预测方程。多光谱植被指数RVI(810，510)和DVI(810，560)可以较为准确地反演LAI；高光谱植被指数RVI[A(760-850)，A(350-400)]、RVI(810，510)、PVI以及红边参数(位置和斜率)和高光谱特征参数P_Area1100与LAI都存在极显著相关关系。多光谱植被指数RVI(1100，560)和RVI(870，560)以及高光谱植被指数A(760-850)／550和A(760-850)／700均可采用线性方程较为准确地反演小麦叶片干重，且不受氮素水平和生育时期的影响。多光谱差值植被指数DVI(560，460)和高光谱归一化植被指数NDVI(857，1210)可以反演茎干重，但其监测的精度低于叶片干重。此外，红边斜率和垂直植被指数PVI可以同时预测LAI和叶茎干重。 利用两种不同来源的冠层光谱反射率及演变的多种光谱参数，研究了小麦氮素状态与冠层反射光谱特征的定量关系，建立了小麦植株氮素状况的敏感光谱参数及预测方程。冠层多光谱指数NDVI(1220，710)和红边位置均可以预测不同类型小麦叶片氮含量，且具有较高的准确性。单位土地面积上叶片氮素积累量与宽波段的冠层反射光更多还原

【Abstract】 Remote sensing technology could be easily used to monitor and forecast growth characters, nitrogen status, yield and quality formation of crop plants, thus it was very important to the development of information agriculture. Based on the technology of canopy reflectance spectral and the test of agronomy parameters and physico-chemical index, the characteristics of canopy multispectral and hyperspectral reflectance under different conditions and their correlation to growth characters, nitrogen status and grain quality traits in wheat were analyzed in this paper. The sensitive spectrum parameters and quantitative regression models of leaf area index, biomass, nitrogen status and grain quality traits were confirmed. Based on the spectral monitoring of plant nitrogen status and the quantitative relationships between nitrogen status, nitrogen transportation and grain protein index, the indirect approach predicting mature grain protein status with reflectance spectra was forwarded, which provided technical basis for crop growth and precise diagnosis.In this study, the combination of multispectral and hyperspectral spectrum device was first used to compare the variation patter of canopy reflectance under different nitrogen supply, growing stage and cultivar in wheat. Results showed hyperspectal remote sensing was more feasible for monitoring water condition in crop plants than multispectral remote sensing. Reflectance at near infrared reflected flat (750-1300nm) increased with increasing nitrogen supply, whereas reflectance at visible band decreased. Reflectance at visible light initially decreased and then increased with growth progress after jointing, with the lowest value appeared at booting. However, reflectance in near infrared had opposite trend, which initially increased and then decreased to the lowest from anthesis to maturity. The canopy reflectance also was different for different cultivars under same nitrogen rate, with significant change at near infrared bands. These results provide background information for monitoring of growth characters, nitrogen status and grain qualities with canopy reflectance spectra in wheat.Based on the change patterns of leaf area index and dry weight under different nitrogen supply with growth stages, correlations of LAI and dry weight of leaf and stem to canopy multispectral and hyperspectral reflectance and spectral parameters were investigated, and put forward sensitive spectral parameters and quantitative equation forecasting LAI and dry weight of leaf and stem in wheat. Multispectral vegetation index, such as RVI (810, 510) and更多还原