基于赤池信息量准则的冬小麦叶面积指数高光谱估测

冬小麦叶面积指数(leaf area index,LAI)是描述冠层结构的重要参数之一,对评价其长势和预测产量具有重要意义。该文利用灰色关联分析(grey relational analysis,GRA)对植被指数进行排序,用偏最小二乘法(partial least squares regression,PLS)选择不同的植被指数个数作为自变量进行回归建模,通过赤池信息量准则(Akaike’s information criterion,AIC)选择AIC值最小的模型作为冬小麦LAI最优估算模型,即GRA、PLS和AIC 3种方法整合建立冬小麦LAI最优估算模型。使用2008-2009年在中国北京通州区和顺义区获取的整个生育期冬小麦LAI和配套的光谱数据进行建模,利用2009-2010相关数据进行验证。研究表明:采用GRA评价标准与冬小麦LAI关联度最大的植被指数是VOG1,关联度最小的植被指数是SR;通过AIC建立的以8个植被指数作为自变量的冬小麦LAI模型效果最优,建模集的决定系数R2和标准误SE分别为0.76和0.009,验证集的R2和相对均方根误差RRMSE分别为0.63和0.004,预测模型和验证模型均具有较高的精度和可靠性。结果表明采用GRA-PLS-AIC方法进行冬小麦LAI反演是可行的,为提高冬小麦LAI遥感预测精度提供了一种有效的方法。     

利用高光谱指数进行冬小麦条锈病严重度的反演研究

通过选取不同条锈病抗性品种(高抗、高感、中间)进行田间不同梯度(对照、轻度、中度、重度)的接种试验,在接种后每隔7 d左右,同步测定了不同品种、不同处理的冠层光谱、单叶光谱和对应目标的病情指数以及叶面积指数、叶倾角等生物物理参数和叶绿素、SPAD数值等生物化学参数。通过对获取的光谱数据和生物物理参数和生物化学参数进行统计分析。研究结果表明,小麦被条锈病感染以后,叶片叶绿素含量急剧下降,通过研究叶片绿度值(SPAFD)值与叶绿素含量之间的关系,建立了叶片叶绿素含量和叶片SPAD数值之间的线性关系方程。通过在借鉴前人研究结果的基础上,通过筛选光谱指数,在冠层水平上构建作物冠层结构不敏感色素反演指数(CCII=TCARI/OSAVI)来反演全生育期不同处理的SPAD数值,此反演结果受品种类型、冠层结构和土壤背景的影响较小,线性方程的决定系数达到极显著的水平。在单叶水平选取归一化的光化学指数(NPRI)来反演单叶的病情指数(DI),线性方程的决定系数达到极显著的水平。所以该文通过选取适当的高光谱指数进行冬小麦条锈病严重度的反演的理论和方法是可行的。且反演结果受不同品种、不同叶面积指数和土壤背景等的影响均较小。     

Physiological interpretation of a hyperspectral time series in a citrus orchard

Hyperspectral remote sensing for monitoring horticultural production systems requires the understanding of how plant physiology, canopy structure, management and solar elevation affect the retrieved canopy reflectance during different stages of the phenological cycle. Hence, the objective of this study was to set up and to interpret a hyperspectral time series for a mature and healthy citrus orchard in the Western Cape province of South Africa considering these effects. Based on the remotely sensed data, biophysical parameters at the canopy level were derived and related to known observed physiological and phenological changes at the leaf level and to orchard management. Fractions of mature fruit, flowers, and sunburnt leaves were considered, and indices related to canopy structure chlorophyll content and canopy water status were calculated.Results revealed small cover fractions of mature fruit, flowers and sunburnt leaves of respectively 2.1%, 3.1% and 7.0%, but the high spectral contrast between flowers and leaves allowed a successful classification of flowering intensity. Furthermore, it was shown that canopy level time series of vegetation indices were sensitive to changes in solar elevation and soil reflectance which could be reduced by applying an empirical soil line correction for the most affected indices. Most trends in vegetation indices at the canopy level could be explained by a combination of changes at the leaf level (chlorophyll, carotenoids, dry matter), changes in canopy structure (leaf area index and leaf angle distribution) and changes in cover fractions of vegetative flushes, flowers and sunburnt leaves. The transformed chlorophyll absorption ratio index over the optimised soil adjusted vegetation index (MCARI/OSAVI) was best related to leaf level trends in chlorophyll content. Seasonal changes in the photochemical reflectance index (PRI) were linked to inverse changes in the carotenoid-to-chlorophyll ratio. Canopy structure indices (the modified triangular vegetation index or MTVI2 and the standardized leaf area index determining index or sLAIDI) were sensitive to changes in leaf area index, average leaf angle as well to management interactions (pruning and harvest). Canopy water status was highly impacted during the spring flush due to expanding leaves that concealed trends in the underlying mature leaves. Seasonal trends in soil and weeds reflectance were related to changes in volumetric soil water content and to the earlier and reduced growth period of non-irrigated weeds.     

基于无人机遥感植被指数优选的田块尺度冬小麦估产

田块尺度作物快捷精准估产对规模化农业经营管理具有重要意义。因此,急需选取最优植被指数和最佳无人机遥感作业时期,建立冬小麦无人机遥感估产模型,获取及时、快速、低成本的无人机遥感估产方法。该文以山东省滨州市典型规模化农田为研究对象,利用固定翼无人机遥感平台对冬小麦进行多期遥感观测与估产。基于2016年冬小麦返青拔节期、抽穗灌浆期和成熟期的无人机遥感影像数据集,采用最小二乘法,构建了基于不同植被指数与冬小麦实测产量的9种线性模型,并结合作物实测产量进行模型评价。多时相多种类植被指数的优选分析结果显示,抽穗灌浆期估产模型R~2最高,RMSE最低(n=34)。其中,模型R~2达到0.70的植被指数共6个,从高到低依次为EVI2、MSAVI2、SAVI、MTVI1、MSR和OSAVI;RMSE由低到高依次为EVI2、MSAVI2、SAVI、MTVI1、MSR和OSAVI。另外,该文进一步评价农田土壤像元对无人机遥感估产的影响,经过阈值滤波法处理后,返青拔节期估产模型的R~2(n=34)从约0.20提升至0.30以上,RMSE和MRE下降;抽穗灌浆期模型的RMSE降低,R~2(n=34)有所提升但不显著。综上所述,最佳无人机飞行作业时期为冬小麦抽穗灌浆期,最优植被指数为EVI2,土壤像元的滤除对抽穗灌浆期无人机遥感估产模型的影响不显著。因此,优化后的基于植被指数的无人机遥感估产模型,可以快速有效诊断和评估作物长势和产量,为规模化农业种植经营提供一种快捷高效的低空管理工具。     

用多角度光谱信息反演冬小麦叶绿素含量垂直分布

由于作物叶片具有一定的叶位空间垂直结构(倒一叶、倒二叶、倒三叶、倒四叶、倒五叶等)，且存在不同叶位叶绿素等生化组分垂直分布的特性，该研究提出利用遥感数据反演作物养分垂直分布，尤其是作物中、下层信息的方法。运用多角度光谱信息，通过不同角度条件下，反映的作物上层、中层、下层信息的差异等通过构建基于不同观测天顶角条件下的冠层叶绿素反演指数的组合值，形成上层叶绿素反演光谱指数、中层叶绿素反演光谱指数和下层叶绿素反演光谱指数来反演作物叶绿素的垂直分布，达到了极显著的水平。表明运用基于多角度光谱信息的光谱指数组合能够较好的反演作物叶绿素含量的垂直分布。对于生产上迫切需要对作物中、下层叶片氮素或叶绿素状况的监测来指导适时和适量施肥，保证获得既定的作物产量和品质目标，提高肥料利用率有重要意义。     

基于高光谱的夏玉米冠层SPAD值监测研究

开展夏玉米冠层SPAD值监测技术研究,建立叶绿素含量与敏感波段、光谱指数间的定量关系模型,以促进高光谱技术在玉米快速、无损长势监测及水肥精准管理的应用。以小型蒸渗仪夏玉米光谱反射率与植株冠层SPAD值的监测为基础,研究了夏玉米植株冠层光谱信息与SPAD值的响应关系,并优选出监测夏玉米冠层SPAD值的敏感波段与最优光谱指数。结果表明:夏玉米冠层光谱反射率在可见光波段随玉米冠层SPAD值增加而下降,在近红外波段却与之相反;采用原始光谱反射率、一阶微分光谱监测夏玉米冠层SPAD值的最敏感波段分别为700,690nm,与SPAD值的相关性分别为-0.498(p<0.05)和-0.538(p<0.01);而根据多元逐步回归分析获得的最优波段组合由405,408,700nm波段构成;从已报道的73个光谱指数中筛选出与夏玉米冠层SPAD值相关性较高的(SDr-SDb)/(SDr+SDb)、MCARI∥OSAVI、TCARI/OSAVI、SDr/SDb和MTCI等5个光谱指数,光谱指数(SDr-SDb)/(SDr+SDb)与SPAD值的相关性在各生育期均达极显著正相关,且在全生育期相关系数高达0.697(p<0.01),进一步优选出监测夏玉米冠层SPAD值最适宜的光谱指数为(SDr-SDb)/(SDr+SDb);在基于敏感波段、光谱指数和最优波段组合建立的夏玉米SPAD值的回归模型中,按照模拟效果由高到低排序依次为最优波段组合、光谱指数、原始光谱反射率、一阶微分光谱,其决定系数分别为0.777,0.539,0.351,0.282;推荐以(SDr-SDb)/(SDr+SDb)指数构建的二次多项式模型与基于405,408,700nm波段组合建立的线性回归监测模型作为夏玉米植株冠层SPAD值光谱监测适宜模型,且R2大于0.539,RMSE及MAE分别小于6.194和4.702。     

波段宽度对利用植被指数估算小麦LAI的影响

为了能够根据遥感数据类型实现指数的优化选择进而提高叶面积指数的反演精度,该研究分析了不同波段宽度(5~80 nm)对植被指数反演叶面积指数精度的影响。通过比较反演模型的决定系数均值,筛选出14个模型精度较高的植被指数,并探讨了不同波段宽度的选取对各指数叶面积指数反演精度的影响。结果表明,波段宽度对不同植被指数的影响可分为3类:1)OSAVI2等指数波宽越窄,反演精度越高,更适合应用于高光谱遥感数据;2)SR[800,680]等指数随着波段宽度的增加,反演精度先升后降,最适波宽为35 nm,适用于中等光谱分辨率的遥感数据;3)SR[675,700]等指数随着波段宽度的增大,反演精度不断提高,在多光谱数据中有更好的应用潜力。     

不同生育时期冬小麦籽粒蛋白质含量的高光谱遥感监测模型

为研究不同氮磷水平下冬小麦籽粒蛋白质含量高光谱遥感监测模型,提高模型精度,本文通过连续5年定位试验研究不同氮磷耦合水平下,不同生育时期冬小麦冠层光谱反射率、植株氮含量以及成熟期籽粒蛋白质含量,以相关、回归等统计分析方法,建立基于不同生育时期植株氮含量的籽粒蛋白质含量监测模型;然后通过灰色关联度分析,筛选植株氮含量的最佳植被指数,以偏最小二乘回归法,建立基于植被指数的植株氮含量监测模型;最后以植株氮含量为链接点,按照"植被指数—植株氮含量—籽粒蛋白质含量"之间的联系,建立融合植被指数与植株氮含量的冬小麦成熟期籽粒蛋白质含量监测模型。结果表明:在拔节期、孕穗期、抽穗期、灌浆期、成熟期基于植株氮含量建立的成熟期籽粒蛋白质含量监测模型,具有较好的监测精度;拔节期、孕穗期、抽穗期、灌浆期、成熟期分别基于修正叶绿素吸收反射率指数(MCARI_1)、归一化差值叶绿素指数(NDCI)、修正归一化差异指数(mNDVI)、MCARI_1、NDCI植被指数建立植株氮含量监测模型,监测精度(R~2)分别为0.826、0.854、0.867、0.859和0.819;以植株氮含量为链接点,通过"植被指数—植株氮含量—籽粒蛋白质含量"的间接联系,建立基于拔节期、孕穗期、抽穗期、灌浆期、成熟期植被指数且融合植株氮含量的籽粒蛋白质含量监测模型,R~2分别为0.935、0.972、0.990、0.979和0.936;以独立数据对模型进行验证,模型预测值与实测值间相对误差(RE)分别为11.26%、10.74%、8.41%、10.25%和11.36%,均方根误差(RMSE)分别为2.221 g×kg~(-1)、1.825 g×kg~(-1)、1.214 g×kg~(-1)、1.767 g×kg~(-1)和2.137 g×kg~(-1)。说明基于不同生育时期植被指数链接植株氮含量可以对成熟期籽粒蛋白质含量进行有效监测,且模型具有较好的年度间重演性和品种间适应性。     

基于高光谱植被指数的黑河流域作物叶绿素含量制图

Recent advances in optical remote sensing led to improved methodologies to monitor crop properties.The red-edge-based vegetation index considered to be one of the most powerful tools for estimating the chlorophyll content(Chl) was usually constructed from in-situ hyperspectral reflectance.In this paper, we present the work done to compare the Chl predictive quality by various red-edge-based vegetation indices based on the CASI data.The results indicated that among the selected vegetation indices, TCARI/OSAVI-based model estimated Chl(R2=0.46, RMSE =0.60 and P<0.01) with the best accuracy.To search the optimal vegetation index for Chl estimation, the normalized difference spectral index(NDSI) and ratio spectral index(RSI) were developed by using the waveband combination algorithm.A high linear correlation(R2=0.79, RMSE=0.38 and P<0.01) was acquired by combining the 869.20 and 754.90 nm wavebands, then NDSI(869.20, 754.90) was applied to the CASI image to generate the Chl distribution map.It suggests that more fertilizer should be provided for the southwest areas due to the lower Chl.     

基于GF-1卫星数据的冬小麦叶片氮含量遥感估算

以陕西关中地区大田和小区试验下的冬小麦为研究对象,探讨基于国产高分辨率卫星GF-1号多光谱数据的冬小麦叶片氮含量估算方法和空间分布格局。基于GF-1号光谱响应函数对地面实测冬小麦冠层高光谱进行重采样,获取GF-1号卫星可见光-近红外波段的模拟反射率,并构建光谱指数,利用与叶片氮含量在0.01水平下显著相关的8类光谱指数,分别建立叶片氮含量的一元线性、一元二次多项式和指数回归模型。通过光谱指数与叶片氮含量的敏感性分析,以及所建模型的综合对比分析,获取适合冬小麦叶片氮含量估算的最佳模型。结果表明:模拟卫星宽波段光谱反射率和卫星实测光谱反射率间的相关系数高于0.95,具有一致性;改进型的敏感性指数综合考虑了模型的稳定性、敏感性和变量的动态范围,敏感性分析表明比值植被指数对叶片氮含量的变化响应能力最强;综合模拟方程决定系数、模型敏感性分析、精度检验和遥感制图的结果,认为基于比值植被指数建立的叶片氮含量估算模型适用性最强,模拟结果与实际空间分布格局最为接近,为基于GF-1卫星数据的区域性小麦氮素营养监测提供了理论依据和技术支持。     

基于高光谱植被指数的黑河流域作物叶绿素含量制图（英文）

Recent advances in optical remote sensing led to improved methodologies to monitor crop properties.The red-edge-based vegetation index considered to be one of the most powerful tools for estimating the chlorophyll content(Chl)was usually constructed from in-situ hyperspectral reflectance.In this paper,we present the work done to compare the Chl predictive quality by various red-edge-based vegetation indices based on the CASI data.The results indicated that among the selected vegetation indices,TCARI/OSAVI-based model estimated Chl(R2=0.46,RMSE=0.60 and P0.01)with the best accuracy.To search the optimal vegetation index for Chl estimation,the normalized difference spectral index(NDSI)and ratio spectral index(RSI)were developed by using the waveband combination algorithm.A high linear correlation(R2=0.79,RMSE=0.38 and P0.01)was acquired by combining the 869.20 and 754.90 nm wavebands,then NDSI(869.20,754.90)was applied to the CASI image to generate the Chl distribution map.It suggests that more fertilizer should be provided for the southwest areas due to the lower Chl.     

Estimating the Nitrogen Concentration of Strawberry Plants from Its Spectral Response

Abstract

Estimating the nitrogen (N) status of plants as a function of their spectral response is a promising technique to diagnose and optimize N fertilization. An experiment was conducted in Jiquilpan (Michoacán, México) in which three N levels (0.3, 3, and 20 mM of NO₃ ^? in the irrigation water) were applied to strawberry (Fragaria vesca) in a randomized complete block design with three replicates. The spectral response of strawberry was measured at both the canopy and leaf level using individual wavebands as well as vegetation indices. Individual leaves were separated into three strata (young, mature, and old) and leaf dry matter, leaf area, and N content (% dry matter) were measured in each stratum. Leaf area, biomass, and N content differed significantly between strata. Leaf area, biomass, and N content in all strata were affected by N fertilization. At the canopy level, N content was highly correlated with green reflectance (R550) (r²=0.50) and red reflectance (R680) (r²=0.60) as well as the vegetation indices simple ratio (SR) (r²=0.56), normalized difference vegetation index (NDVI) (r²=0.56), and hyperspectral NDVI (HNDVI) (r²=0.56). For individual leaves, significant differences between strata were found with normalized total pigment to chlorophyll a ratio index (NPCI) and MERIS terrestrial chlorophyll index (MTCI) (p<0.001) as well as R550, photochemical reflectance index (PRI), red edge position (REP), and REP calculated using the MERIS satelite wavelengths (p<0.01). Relationships between spectral indices and N content at the leaf level were found with the youngest leaves only, with NPCI (p<0.01) and MTCI (p<0.05), whereas only R550 responded to N fertilization (p<0.05).     

关中地区夏玉米抽穗期叶绿素含量的高光谱估算

[目的]利用高光谱数据进行叶绿素估算,为快速获取作物的生长信息、生长诊断及精确管理提供依据。[方法]基于陕西省关中地区抽穗期夏玉米冠层光谱特征及叶绿素含量的测定,运用线性及非线性分析方法建立了基于原始光谱敏感波段和一阶微分光谱敏感波段叶绿素估算模型。[结果]夏玉米抽穗期反射光谱在可见光及中远红外区域,叶绿素含量越高,光谱曲线越向下偏移;在红边区域,叶绿素含量对光谱曲线影响不显著;在近红外波段,叶绿素含量越高,光谱曲线越向上偏移。基于一阶微分光谱敏感波段的夏玉米叶绿素含量估算模型拟合精度要优于基于原始光谱敏感波段估算模型,决定系数R2分别为0.81和0.60,均方根误差(RMSE)分别为2.39,4.41。[结论]基于一阶微分光谱敏感波段建模分析是估测抽穗期夏玉米冠层叶绿素含量的重要方法,对指导西北地区夏玉米种植与生产具有积极的借鉴意义。     

基于高光谱反射率的棉花冠层叶绿素密度估算

为了进一步提高棉花叶绿素密度高光谱估算精度,该研究以棉花冠层叶绿素密度以及冠层高光谱反射率为数据源,在分析叶绿素密度与原始高光谱反射率(R)、一阶导数光谱反射率(DR)、已有光谱指数及全波段组合指数相关性的基础上,采用线性及多元逐步回归技术构建了叶绿素密度高光谱诊断模型,系统对比分析了以上4种光谱形式用于棉花冠层叶绿素密度诊断的精度。结果表明:1)基于一阶导数光谱反射率的估算模型精度明显优于原始光谱反射率;2)基于比值指数或归一化指数形式的估算模型精度及稳定性要优于单波段或多波段的线性模型;3)单波段变量DR756、全波度组合比值指数DR635/DR643以及归一化指数(DR1055-DR684)/(DR1055+DR684)均可较好的实现叶绿素密度估算,其中由DR635/DR643为自变量的模型所得到棉花冠层叶绿素密度估算值与实测值拟合最好,相关系数达到0.821。该研究可为高光谱技术在棉花冠层叶绿素密度诊断中的更好应用提供参考。     

利用无人机多光谱数据监测玉米对不同灌溉模式的响应差异

[目的]作物水分状况的实时监测对于节水灌溉、缓解我国水资源紧缺具有重要意义.本研究旨在探寻利用无人机多光谱影像数据实时监测玉米干旱胁迫状况的可行性,比较无人机数据和田间实测农学指标对作物干旱胁迫的敏感程度.[方法]大田试验在河北吴桥进行,采用两个玉米品种'富民985'和'郑单958',设置畦灌、滴灌和雨养3种模式.分别...     

基于无人机数码影像和高光谱数据的冬小麦产量估算对比

作物产量准确估算在农业生产中具有重要意义。该文利用无人机获取冬小麦挑旗期、开花期和灌浆期数码影像和高光谱数据,并实测产量。首先利用无人机数码影像和高光谱数据分别提取数码影像指数和光谱参数,然后将数码影像指数和光谱参数与冬小麦产量作相关性分析,挑选出相关性较好的9个指数和参数,最后以选取的数码影像指数和光谱参数为建模因子,通过MLR(multiple linear regression,MLR)和RF(random forest,RF)对产量进行估算。结果表明:数码影像指数和光谱参数与实测产量均有很强的相关性。利用数码影像指数和光谱参数通过MLR和RF构建的产量估算模型均在灌浆期表现精度最高,在灌浆期,数码影像指数和光谱参数构建的MLR模型R~2和NRMSE分别为0.71、12.79%,0.77、10.32%。对模型对比分析可知,以光谱参数为因子的MLR模型精度较高,更适合用于估算冬小麦产量。利用无人机遥感数据,通过光谱参数建立的MLR模型能够快速、方便地对作物进行产量预测,并可以根据不同生育期的产量估算模型有效地对作物进行监测。     

Application of chlorophyll-related vegetation indices for remote estimation of maize productivity

Crop gross primary productivity (GPP) is an important characteristic for evaluating crop nitrogen content and yield, as well as the carbon exchange. Based on the close relationship observed between GPP and total chlorophyll content in crops, we applied a model that relies on a product of chlorophyll-related vegetation index and incoming photosynthetically active radiation for remote estimation of GPP in maize. In this study, we tested the performance of this model for maize GPP estimation based on spectral reflectance collected at a close range, 6 m above the top of the canopy, over a period of eight years from 2001 through 2008. Fifteen widely used chlorophyll-related vegetation indices were employed for GPP estimation in irrigated and rainfed maize, and accuracy and uncertainties of the models were compared. We also explored the possibility of using a unified algorithm in estimating maize GPP in fields that are different in irrigation, field history and climatic conditions. The results showed that vegetation indices that closely relate to total canopy chlorophyll content and/or green leaf area index were accurate in GPP estimation. Both green and red edge Chlorophyll Indices, MERIS Terrestrial Chlorophyll Index as well as Simple Ratio were the best approximations of the widely variable GPP in maize under different crop managements and climatic conditions. They were able to predict daily GPP reaching 30 gC/m²/d with RMSE below 2.75 gC/m²/d.     

利用无人机多光谱影像的多品种玉米成熟度监测

基于遥感监测多品种玉米成熟度进而掌握最佳收获时机，对提高其产量和品质至关重要。该研究在玉米成熟阶段获取无人机多光谱影像，同步采集叶片叶绿素含量（chlorophyll content，C）、籽粒含水率（moisture content，M）、乳线占比（proportion of milk line，P）等地面实测数据，以此构建玉米成熟度指数（maize maturity index，MMI），从而定量表征玉米成熟度。通过MMI与植被指数构建回归模型和随机森林模型，验证MMI适用性，并分析无人机遥感对不同品种玉米成熟度的监测精度。结果表明：1）不同品种玉米的叶片叶绿素含量、籽粒含水率、乳线占比的变化速率均存在差异。2）MMI与所选植被指数的相关性均可达到0.01显著水平，其中与归一化植被指数（normalized difference vegetation index，NDVI）、转换叶绿素吸收率（transformed chlorophyll absorbtion ratio index，TCARI）相关性最高，相关系数均为0.87。3）该研究基于不同组合的数据集进行了模型验证，其中随机森林模型对MMI的估测精度最高，测试集决定系数（coefficient of determination，R²）为0.84，均方根误差（root mean squared error，RMSE）为8.77%，标准均方根误差（normalized root mean squared error，nRMSE）为12.05%。此外，随机森林模型对不同品种MMI的估测精度较好，京九青贮16精度最优，其中R²、RMSE、nRMSE为0.76、10.67%、15.88%，模型精度证明了可以利用无人机平台对不同品种玉米成熟度进行监测。研究结果可为多光谱无人机实时监测农田多品种玉米成熟度的动态变化提供参考。     

用冠层高光谱反射估测玉米植株色素时的采样点和指标选择

初步分析了11个玉米品种在7个生育期的植株叶绿素含量、类胡萝卜素含量和叶绿素/类胡萝卜素比值在不同层位(上层、中层和下层)的变化特点及其与15个高光谱参量间的关系，结果发现1)玉米叶片叶绿素/类胡萝卜素比值较明显表现出随形态学高度的下降而升高；2)绝大多数的高光谱参量与叶绿素含量(mg·g^-1FW)间达到了显著或极显著相关，但相关性不强(最大r=0.56,n=77)；3)高光谱参量与叶绿素/类胡萝卜素比值间相关性，明显强于与叶绿素含量或类胡萝卜素含量间的相关性(最大r=0.75,n=77),并表现出明显的层间差异，下层明显强于中层和上层。研究表明，在用高光谱反射分析玉米的色素状况时，用叶绿素/类胡萝卜素比值代替叶绿素含量作为色素指标可提高分析精度。     

Remote estimation of gross primary production in wheat using chlorophyll-related vegetation indices

A number of recent studies have focused on estimating gross primary production (GPP) using vegetation indices (VIs). In this paper, GPP is retrieved as a product of incident light use efficiency (LUE), defined as GPP/PAR, and the photosynthetically active radiation (PAR). As a good correlation is found between canopy chlorophyll content and incident LUE for six types of wheat canopy (R² = 0.87, n = 24), indices aimed for chlorophyll assessment can be used as an indicator of incident LUE and the product of chlorophyll indices and PAR will be a proxy of GPP. In a field experiment, we investigated four canopy chlorophyll content related indices (Red edge Normalized Difference Vegetation Index [Red Edge NDVI], modified Chlorophyll Absorption Ratio Index [MCARI₇₁₀], Red Edge Chlorophyll Index [CI_{red edge}] and the MERIS Terrestrial Chlorophyll Index [MTCI]) for GPP estimation during the growth cycle of wheat. These indices are validated for leaf and canopy chlorophyll estimation with ground truth data of canopy chlorophyll content. With ground truth data, a strong correlation is observed for canopy chlorophyll estimation with correlation coefficients R² of 0.79, 0.84, 0.85 and 0.87 for Red Edge NDVI, MCARI₇₁₀, CI_{red edge} and MTCI, respectively (n = 24). As evidence of the existence of a relationship between canopy chlorophyll and GPP/PAR, these indices are shown to be a good proxy of GPP/PAR with R² ranging from 0.70 for Red Edge NDVI and 0.75 for MTCI (n = 240). Remote estimation of GPP from canopy chlorophyll content × PAR is proved to be relatively successful (R² of 0.47, 0.53, 0.65 and 0.66 for Red edge NDVI, MCARI₇₁₀, CI_{red edge} and MTCI respectively, n = 240). These results open up a new possibility to estimate GPP and should inspire new models for remote sensing of GPP.