大豆叶绿素含量高光谱反演模型研究

叶绿素是植物体进行光合作用、进行第一性生产的重要物质，能够间接反映植被的健康状况与光合能力，同时也能反映植被受环境胁迫后的生理状态。高光谱遥感为快速、大面积监测植被的叶绿素变化提供了可能。该研究实测了不同水肥耦合作用下，大豆冠层的高光谱反射率与叶绿素含量数据，对二者进行了相关分析；采用特定叶绿素敏感波段建立了植被指数叶绿素估算模型；最后采用相关系数较大的波段作为神经网络模型的输入变量进行了叶绿素含量的估算。经对比发现叶绿素A、B与光谱反射率在可见光与近红外波段的相关系数的变化趋势基本一致，在可见光谱波段呈负相关，近红外波段呈正相关，红边处相关系数由负变正。特定色素植被指数可以提高大豆叶绿素估算精度(R²>0.736)，但是人工神经网络模型可以大大提高大豆叶绿素含量的估算水平，当隐藏层节点数为4时，R²大于0.94，随着隐藏层节点数的增加，R²可高达0.99，表明神经网络模型可以大大提升高光谱反演大豆叶绿素含量的能力。     

基于光谱变换的滨海湿地土壤全氮含量建模预测

基于133个滨海湿地土样的全氮（TN）含量和光谱反射率（R）及其对数（lgR）、对数的一阶微分（（lgR）''）、倒数（1/R）、倒数的一阶微分（（1/R）''）、一阶微分（R''）、平方根（√R）、一阶微分的倒数（1/（R）''）变换，采用偏最小二乘回归（PLSR）、随机森林回归（RFR）和支持向量机回归（SVR）3种算法分别建立土壤TN含量估测模型。结果表明：①土壤TN含量与光谱变换形式相关性由高到低为：（1/R）''> R''> （lgR）''> 1/R > lgR > 1/（R）''> √R > > R，经光谱变换，土壤TN含量与变换光谱的相关性均高于R，其中与（1/R）''的Pearson相关系数最大为0.746。②PLSR和SVR基于R''、（1/R）''、（lgR）''和1/（R）''变换构建的模型、RFR方法构建的所有模型R²均大于0.732，均可用于滨海湿地土壤TN含量的估算。③基于1/（R）''建立的SVR模型预测精度最高，其R²为0.987，RMSE为0.057 g/kg，MAE为0.050 g/kg，是预测滨海湿地土壤TN含量的最优模型，可为准确获取滨海湿地土壤TN含量提供稳定方法。     

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。     

基于SPAD-502与GreenSeeker的冬小麦氮营养诊断研究

用GreenSeeker和SPAD-502测定了不同氮素处理的冬小麦冠层NDVI与叶片SPAD值, 分析了它们与叶片全氮、叶绿素含量及产量间的关系。结果表明: 冬小麦抽穗期SPAD值和NDVI值均与叶绿素含量呈极显著正相关; 除抽穗期和返青期外, SPAD值与叶氮含量、叶绿素含量的相关系数在其余各生育期均达到显著或极显著水平; NDVI值与叶氮含量、叶绿素含量在拔节期、乳熟期的相关性同SPAD值一致; SPAD值可以进行叶绿素的诊断, NDVI值可以进行氮的诊断。氮营养诊断时期应该选择拔节期。通过回归建立了基于SPAD值、NDVI值的产量估测模型, 可以通过SPAD值、NDVI值对冬小麦产量进行估算。     

基于半经验半机理建模的冬小麦LAI反演及长势评估

为了提高无人机遥感对冬小麦叶面积指数（leaf area index，LAI）反演模型的精度与泛化能力，该研究利用无人机搭载多光谱相机获取不同氮素处理和不同复种方式的冬小麦生长实测数据，结合PROSAIL辐射传输模型生成包含机理信息的模拟数据，基于不同组合方式建立了5种LAI反演混合数据集，结合多种机器学习方法，以期构建经验与机理相结合的LAI高精度反演模型。由于LAI反演受NIR波段反射率影响大，该研究筛选7种与NIR波段相关的植被指数提取冬小麦光谱特征，构建与混合数据集LAI的相关系数矩阵，进一步探究不同光谱特征对冬小麦LAI的影响程度。在此基础上，采用具有代表性和普适性的4种机器学习方法，即贝叶斯岭回归模型、线性回归模型、弹性网络模型和支持向量回归模型，构建不同冬小麦LAI反演模型，用以评估基于半经验半机理数据反演冬小麦LAI的可行性，进一步探索其对不同氮素水平和复种方式的冬小麦长势评估能力。结果表明：1）筛选的与NIR波段相关的植被指数与冬小麦LAI之间存在较强的相关性，其中归一化差异植被指数、增强植被指数、归一化差异红边指数、比值植被指数、红边叶绿素植被指数、土壤调节植被指数与LAI呈正相关，结构不敏感色素植被指数与LAI呈负相关；2）辐射传输模型中体现了冬小麦LAI影响太阳光线传播的机理，结果表明，与实测数据混合建立的模型，具有较强的鲁棒性和泛化能力。相比于其他3种模型，支持向量回归模型在各种数据组合下均取得了较好的LAI预测性能，在C1、C2、C3、C4这4种训练-测试组合的训练集中R²依次为0.86、0.87、0.88、0.91，RMSE依次为0.47、0.45、0.45、0.41；在测试集的R²依次为0.85、0.19、0.89、0.87，RMSE依次为0.45、1.31、0.49、0.50；3）使用支持向量机生成试验区LAI反演图，对4种氮素水平和2种复种方式的冬小麦长势评估，结果表明，适当的施加氮素处理能提高冬小麦LAI值，麦-豆复种方式下的冬小麦LAI值普遍高于麦-玉复种的LAI值。该研究为冬小麦LAI的反演提供了一种有效的方法，并为高效评估冬小麦长势研究提供了参考。     

基于傅里叶变换近红外光谱的籽棉含水率无损检测

为了实现对籽棉含水率的快速、无损检测，该研究采用傅里叶变换近红外光谱技术建立籽棉含水率定量检测模型。首先探究了籽棉样本密度对于光谱曲线的影响，该研究发现样本密度大小对光谱曲线影响显著，密度越小光谱信号越强，当样品密度不低于0.0886 g/cm³时，光谱曲线变化趋于平稳。通过采集籽棉样本在3900～11000 cm^-1波数范围的吸光度光谱数据，并应用了9种预处理方法对原始光谱数据进行处理。发现一阶导数结合消除趋势（first derivative- detrending，FD-DT）预处理方法在偏最小二乘回归（partial least squares regression，PLSR）模型建立时表现最佳。使用了竞争自适应重复加权法（competitive adaptive reweighted sampling，CARS）、信息增益法（information gain，IG）、连续投影法（successive projections algorithm，SPA）和相关系数（correlation coefficient，CC）等算法，来获取最佳的特征波长。构建PLSR和支持向量机（support vector machine，SVM）的籽棉水分含量预测模型，比较不同分析算法，确定了FD-DT-CARS-PLSR和FD-DT-CARS-SVM两种算法组合作为最佳预测模型，预测集决定系数（R²_P）分别为0.933和0.931，预测集均方根误差（root mean square error，R_MSE）分别为0.480和0.500，剩余预测偏差（residual prediction deviation，R_PD）分别为3.88和3.85。研究结果表明，利用近红外光谱技术可以无损和准确地检测籽棉样本的含水率。     

DNDC模型和TOPSIS法联合确定生物降解地膜覆盖农田合理施氮量

为明确干旱区最优覆盖地膜类型和施氮制度，于内蒙古河套灌区木垒滩节水试验站进行为期2年的不同类型地膜覆盖农田不同施氮量试验。在高氮水平（传统施氮336 kg/hm²）下设置3种覆膜处理，包括塑料地膜（PFM3）、生物降解地膜（BFM3）和无膜覆盖处理（NFM3）；同时在生物降解地膜覆盖下设立3个施氮水平，包括中氮（BFM2，276 kg/hm²）、低氮（BFM1，216 kg/hm²）和不施氮（BFM0，0 kg/hm²），共6个处理。利用2年观测的产量、吸氮量和氮淋失量对DNDC模型进行率定和验证，并基于改进的TOPSIS方法对地膜类型和施肥制度进行优化。结果表明：DNDC模型对地膜覆盖及氮肥调控下作物生长与氮素迁移较为敏感，产量、吸氮量与氮淋失量模拟的EF与R²均大于0.83，NRMSE均小于20%，能够为作物生产力与资源利用进行预测和评估。随施氮量增加，所有覆膜处理的氮淋失量呈线性增加，当施氮量增加至106 kg/hm²时，氮肥利用率达到峰值；当施氮量增加至256 kg/hm²时，覆膜处理的产量不再发生明显变化，同时生物降解地膜的净收益也达到最大值；但其成本高，导致净收益比塑料地膜降低6.84%，比无膜覆盖处理提高3.17%。塑料地膜和生物降解地膜覆盖下的氮淋失量、氮肥利用率和产量无明显差异，均大于无膜覆盖处理，平均提高8.22%~26.69%。利用改进的TOPSIS法对产量、氮淋失量、残膜量、氮肥利用率和净收益5个方面进行综合评价，选出在生物降解地膜覆盖下施用氮肥231~256 kg/hm²是干旱地区较合理的覆膜施氮制度。     

基于卫星光谱尺度反射率的冬小麦生物量估算

为探索基于光学卫星遥感数据的冬小麦地上生物量估算方法,本研究通过3年田间试验,获取冬小麦4个关键生育期（拔节期、抽穗期、开花期和灌浆期）和3种施氮水平下的地上生物量以及对应的近地冠层高光谱反射率数据。通过将高光谱数据重采样为具有红边波段的RapidEye、Sentinel-2和WorldView-2卫星波段反射率数据,构建任意两波段归一化植被指数。同时,将卫星波段反射率数据与6种机器学习和深度学习算法相结合,构建冬小麦生物量估算模型。研究结果表明：任意两波段构建的最佳植被指数在冬小麦开花期对生物量的敏感性最强（决定系数R2为0.50～0.56）。在不同施氮水平条件下,高施氮水平增强了植被指数对生物量的敏感性。Sentinel-2波段数据所构建的植被指数优于其他两颗卫星波段数据。对6种机器学习和深度学习算法,总的来说,基于深度神经网络（Deep Neural Networks,DNN）算法所构建的模型要优于其他算法。在单一生育期中,在拔节期（R2为0.69～0.78,归一化均方根误差为26%～31%）和开花期（R2为0.69～0.70,归一化均方根误差为24%～25%）的估算精度最高。Sentinel-2波段数据与DNN算法结合的估算精度最高,在全生育期中预测精度R2为0.70。施氮水平的提高同样增强了DNN模型的估算精度,3颗卫星波段数据在300 kg/hm2施氮条件下的预测精度R2都在0.71以上,均方根误差小于219 g/m2。研究结果揭示了光学卫星遥感数据在不同生育期和施氮条件下估算冬小麦生物量的潜力。     

斑潜蝇虫害叶片受害程度对其近红外反射光谱的影响

为探索实现作物虫害自动监测的方法，采用图像处理和光谱分析技术，测定了斑潜蝇虫害叶片的近红外反射光谱，计算了虫害叶片的破损率，对其破损率和干鲜比与近红外分光反射率的关系分别进行了回归分析。结果表明：在某些波段，叶片的破损率和干鲜比均与近红外分光反射率有较好的相关性。叶片的干鲜比与近红外分光反射率关系的决定系数：黄瓜为R²=0.79(在1452 nm)，番茄为R²=0.70(在1450 nm)。叶片的破损率与近红外分光反射率关系的决定系数，黄瓜为R²>0.81(在1436～1468 nm)，番茄为R²>0.69(在1436～1466 nm)。试验和分析结果证明斑潜蝇虫害叶片的虫害程度能很好地被近红外光谱信息反映。     

基于光谱指数与机器学习算法的土壤电导率估算研究

土壤盐分是干旱区土壤盐渍化评价的重要指标。以新疆维吾尔自治区渭干河-库车河三角洲绿洲为例，基于土壤电导率 (Electrical conductivity，EC) 及可见光-近红外 (Visible and near infrared, VIS-NIR) 光谱数据，通过蒙特卡洛交叉验证 (Monte Carlo cross validation, MCCV) 确定364个有效样本。采用原始光谱 (Raw reflectance, R) 及其经过微分、吸光度 (Absorbance, Abs)、连续统去除 (Continuum removal, CR) 等6种预处理后的数据构建光谱指数。基于遴选出的21个最优指数，采用BP神经网络 (Back propagation neural network, BPNN)、支持向量机 (Support vector machine, SVM)、极限学习机 (Extreme learning machine, ELM) 三种算法对EC进行估算，并引入偏最小二乘回归 (Partial least squares regression, PLSR) 进行比较。结果表明：在基于R与6种光谱预处理数据构建的21个最优光谱指数之中，R_FD_RSI (R₁₉₁₃，R₂₁₄₂) 表现最佳 (r = 0.649) ；与PLSR相比，机器学习算法能够显著提高模型的估算精度，R²提高了34.55%。三种机器学习算法模型中，ELM表现最优 (R² = 0.884, RMSE = 3.071 mS?cm^-1, RPIQ = 2.535) 。本研究中所构建的光谱指数在兼顾遥感机理的同时能深度挖掘更多的隐含信息，并且基于机器学习算法的土壤EC估算模型精度显著提高，为干旱区土壤盐分定量估算提供了科学参考。     

利用Landsat TM遥感数据监测冬小麦开花期主要长势参数

为精准农业技术体系中的小麦农艺处方管理决策提供详尽的全局性信息,该文以2007－2009年试验实测数据为基础,以Landsat TM影像为遥感数据源,分析了试验样点开花期冬小麦主要长势参数与品质和产量间以及与卫星遥感变量间的相关性,分别建立及评价了TM影像遥感变量监测冬小麦开花期SPAD值、生物量、叶面积指数和叶片氮含量的模型。结果表明：冬小麦开花期,选用作物氮反射指数、近红外波段反射率和归一化植被指数这些遥感变量分别反演冬小麦SPAD值、生物量、叶面积指数和叶片氮含量是可行的;SPAD值、生物量、叶面积指数和叶片氮含量遥感监测模型的精度较高,均方根误差分别为3.12、216.5 kg/hm2、0.269和0.162,以此为基础,制作出具有实际农学意义的冬小麦开花期不同等级SPAD值、生物量、叶面积指数和叶片氮含量遥感监测专题图,实现了主要长势参数空间分布量化表达。基于卫星影像的农田面状信息获取技术克服了点状信息的不足,为农业生产管理决策及时提供信息支持,使该研究技术更利于大面积应用和推广。     

Functional traits associated with nitrogen use efficiency in wheat

The association between functional traits and nitrogen use efficiency (NUE) was investigated to assist the breeding of nitrogen (N) use-efficient bread wheat (Triticum aestivum ssp. aestivum) varieties. This study combined results from a climate chamber experiment involving 41 spring wheat varieties and a field experiment involving six winter and six spring wheat varieties grown with and without the application of mineral N fertiliser. The climate chamber experiment was analysed by partial least squares (PLS) regression, with several predictors and NUE as response, to identify traits related to NUE. Specific hypotheses were then tested in the field experiment. The PLS indicated six traits of particular importance for overall NUE: leaf chlorophyll (SPAD value) of the top leaf at stem elongation, grains ear^?1, ears pot^?1, straw biomass pot^?1, days between emergence and anthesis, and days between emergence and completed senescence. In the field experiment, the SPAD value of flag leaves of winter wheat around anthesis was positively correlated with NUE and total grain N, at both N levels. Fast development was positively correlated with high NUE and N uptake efficiency in spring wheat. Early senescence of the flag leaf was positively correlated with grain N concentration and negatively correlated with grain-specific N efficiency in winter wheat at low N fertilisation levels. The results indicate that high SPAD value of the top leaf might be a candidate trait that could be used in wheat breeding for improved NUE, while genetic variation in senescence could possibly be used to tailor varieties for different end-use quality when grown at low N. More studies are needed to validate these findings in other environments and for other genotypes.     

Optimizing the Nitrogen Management Strategy for Winter Wheat in the North China Plain Using Rapid Soil and Plant Nitrogen Measurements

Excessive nitrogen (N) fertilizer with improper split-application in small-scale farming is widespread for reducing N use efficiency and polluting the environment. The objective of this study was to develop a strategy for providing winter wheat with twice-topdressing N by quickly measuring the soil and plant N status. During the period 2009–2011, a field experiment was conducted for winter wheat cultivar Zhongmai-175 in the North China Plain. The mineral N (Nmin) pool at a soil depth of 0–90 cm and topdressing N twice, as total N supply, was gradually increased from 0 to 420 kg N ha^–1 to mimic the farmers´ practices. Measurements with the Soil Plant Analysis Development (SPAD) meter were taken on the uppermost fully expanded leaf, and the SPAD index was expressed relative to SPAD readings of sufficiently fertilized plants. Grain yield exhibited linear-plus-plateau responses to total N supply with a significant difference between years, the r² ranged from 0.73 to 0.94. With a basal N application of 30 kg ha^–1, the soil Nmin at 0–90 cm supplemented by twice-topdressing N (1:1 ratio) at Zadoks growth stage (ZGS) 22–23 in early spring and ZGS 47–52 was required at 150–165 kg N ha^–1 to achieve a maximum grain yield of 3.9–5.3 t ha^–1. The SPAD index exhibited a strong exponential response to N supply irrespective of plant growth stage and year (r² = 0.95–0.97); the value of 0.94 was critical in denoting N deficiency from sufficiency status. The N topdressing at ZGS 47–52 could be precisely modified/estimated by the equation y = 161.7–218x^5.16, where x is the SPAD index. Since SPAD readings varied significantly from year to year, our study suggests that it might be difficult to precisely manage field N for winter wheat.     

Use of a hand‐held chlorophyll meter in winter wheat: Evaluation of different measuring positions on the leaves

A portable chlorophyll meter (Minolta SPAD‐502) was used to assess the nitrogen status of winter wheat (Triticum aestivum L.) in two fertilizer trials at Apelsvoll Research Centre, located in south‐east Norway. The midpoint of the last fully developed leaf was found to be the best position on the winter wheat plant on which to take chlorophyll meter readings. This conclusion was reached after examination of the relationships between soil‐plant analyses development (SPAD) readings taken at different positions on the plant and leaf nitrogen concentration, grain yield and grain protein content. Emphasis was also laid on finding a measuring position that was convenient from a practical point of view. The relationships between chlorophyll meter readings and the parameters investigated were better at Zadoks growth stage (GS) 49 than earlier in the season at GS 31.     

Leaf color diagnosis and nitrogen management for winter wheat “Kitanokaori” in Hokkaido

(Jpn. J. Soil Sci.Plant Nutr., 77, 293–298, 2006)

“Kitanokaori” is a new variety of wheat for bread use bred at the National Agricultural Research Center for Hokkaido Region. The grain protein content of wheat for bread use should be higher than 120 g kg?1. Much nitrogen application is necessary to obtain high grain protein content. Therefore, it is necessary to determine the optimum amount of nitrogen to obtain the required protein content and to prevent nitrogen from remaining in the soil. The nutrition diagnosis using leaf color was investigated to predict the need and the amount of top-dressing. Field experiments were conducted for four years with nitrogen treatments in Andosol, which has moderate nitrogen fertility, and in Histosol, which is a fertile soil. The leaf color was measured using a chlorophyll meter SPAD502 (CM value) at the middle part of the leaf, avoiding the center rib. The colors of the 10·15 uppermost second leaves were measured in one plot and averaged.

A close relation was found between leaf color at the full heading stage and grain protein content at harvest. Leaf color at the full heading stage is therefore a good index to control the protein content. Considering the effect of top-dressing at the full heading stage in each CM value, the diagnosis criterion was decided. When the CM value is over 52 at the full heading stage, more nitrogen application is not needed. When the CM value is 50·52, 30 kg N ha?1 of top-dressing at the full heading stage is needed, and when the CM value is 45·50, 60 kg N ha?1 of top-dressing is needed to obtain a grain protein content of more than 120 g kg?1.     

Spectral indices measured with proximal sensing using canopy reflectance sensor, chlorophyll meter and leaf color chart for in-season grain yield prediction of basmati rice

A number of optical sensing tools are now available and can potentially be used for refining need-based fertilizer nitrogen (N) topdressing decisions. Algorithms for estimating field-specific fertilizer N needs are based on predictions of yield made while the crops are still growing in the field. The present study was conducted to establish and validate yield prediction models using spectral indices measured with proximal sensing using GreenSeeker canopy reflectance sensor, soil and plant analyzer development (SPAD) chlorophyll meter, and two different types of leaf color charts (LCCs) for five basmati rice genotypes across different growth stages. Regression analysis was performed using normalized difference vegetation index (NDVI) recorded with GreenSeeker sensor and leaf greenness indices measured with SPAD meter and LCCs developed by Punjab Agricultural University, Ludhiana (India) (PAU-LCC) and the International Rice Research Institute, Philippines (IRRI-LCC). The exponential relationship between NDVI and grain yield exhibited the highest coefficient of determination (R²) and minimum normalized root mean square error (NRMSE) at panicle initiation stage and explained 38.3%-76.4% variation in yield using genotype-specific models. Spectral indices pooled for different genotypes were closely related to grain yield at all growth stages and explained 53.4%-57.2% variation in grain yield. Normalizing different spectral indices with cumulative growing degree days (CGDD) decreased the accuracy of yield prediction. Normalization with days after transplanting (DAT), however, did not reduce or improve the predictability of yield. The ability of each model to predict grain yield was validated with an independent dataset collected from two experiments conducted at different sites with a range of fertilizer N doses. The NDVI-based genotype-specific models exhibited a robust linear correlation (R²=0.77, NRMSE=7.37%, n=180) between observed and predicted grain yields only at 35 DAT (i.e., panicle initiation stage), while the SPAD, PAU-LCC, and IRRI-LCC consistently provided reliable predictions (with respective R² of 0.63, 0.60, and 0.53 and NRMSE of 10%, 10%, and 13.6%) even with genotype invariant models with 900 data points obtained at different growth stages. The study revealed that unnormalized values of spectral indices, namely NDVI, SPAD, PAU-LCC, and IRRI-LCC, can be satisfactorily used for in-season estimation of grain yield for basmati rice. As LCCs are very economical compared with chlorophyll meters and canopy reflectance sensors, they can be preferably used by small farmers in developing countries.     

Late seeding of soybean after rice transplanting period using variety "Yukihomare" in central area of Hokkaido

(Jpn. J. Soil Sci.Plant Nutr., 77, 293–298, 2006)
"Kitanokaori" is a new variety of wheat for bread use bred at the National Agricultural Research Center for Hokkaido Region. The grain protein content of wheat for bread use should be higher than 120 g kg−1. Much nitrogen application is necessary to obtain high grain protein content. Therefore, it is necessary to determine the optimum amount of nitrogen to obtain the required protein content and to prevent nitrogen from remaining in the soil. The nutrition diagnosis using leaf color was investigated to predict the need and the amount of top-dressing. Field experiments were conducted for four years with nitrogen treatments in Andosol, which has moderate nitrogen fertility, and in Histosol, which is a fertile soil. The leaf color was measured using a chlorophyll meter SPAD502 (CM value) at the middle part of the leaf, avoiding the center rib. The colors of the 10·15 uppermost second leaves were measured in one plot and averaged.
A close relation was found between leaf color at the full heading stage and grain protein content at harvest. Leaf color at the full heading stage is therefore a good index to control the protein content. Considering the effect of top-dressing at the full heading stage in each CM value, the diagnosis criterion was decided. When the CM value is over 52 at the full heading stage, more nitrogen application is not needed. When the CM value is 50·52, 30 kg N ha−1 of top-dressing at the full heading stage is needed, and when the CM value is 45·50, 60 kg N ha−1 of top-dressing is needed to obtain a grain protein content of more than 120 g kg−1.     

花后小麦叶面积指数与光合和产量关系的研究

以大穗型品种泰农18（T18）和中穗型品种山农15（S15）为材料,采用裂区设计,主区设置180 kg/hm2和240 kg/hm2 两个氮肥水平（纯氮）,裂区设置75104株/hm2,150104株/hm2和225104株/hm2三个种植密度,研究了叶面积指数与冬小麦光合和产量的关系。结果表明:冬小麦下部叶片叶面积指数（LLAI,倒4叶和倒5叶之和）与群体净光合速率（CAP）和产量呈极显著的正相关关系;从开花到花后14 d 之前,上部叶片叶面积指数（TLAI,旗叶、倒2叶和倒3叶之和）和全部叶面积指数（WLAI,下部叶片和上部叶片之和）与群体净光合速率（CAP）和产量并不呈必然的显著正相关关系（T18 显著正相关,S15相关不显著）,但花后14d 至成熟期,两个品种的 TLAI和 WLAI 均与群体净光合速率（CAP）和产量呈显著的正相关关系。氮肥水平和种植密度对产量存在显著的互作效应,无论是大穗型品种T18还是中穗型品种S15都可以通过适当增加密度（T18以225104//hm2为宜,S15以150104//hm2为宜）,降低氮肥用量（180 kg/hm2）实现高产。