Comparison of petiole nitrate concentrations,SPAD chlorophyll readings,and QuickBird satellite imagery in detecting nitrogen status of potato canopies

Nitrogen (N) management is critical in optimizing potato yield and quality and reducing environmental pollution. Six N rates from 34 to 270 kg ha⁻¹, and different timing of N application were used in a 3-year field experiment to contrast SPAD-502 chlorophyll meter and QuickBird satellite imagery data against the conventional petiole sampling technique for assessing canopy N status. Overall treatment variations in SPAD readings were consistent with those in petiole nitrate-nitrogen (NO₃-N) concentrations. However, the ability of the SPAD meter to detect treatment differences varied with growth stage and growing season. Severe N deficiency was detected about 1 month after emergence with SPAD readings, but as early as 2 weeks after emergence with petiole NO₃-N concentrations. Petiole NO₃-N concentrations tended to differentiate more treatment variations than SPAD readings at all growth stages except at hilling. N deficiency was detected with QuickBird image-derived vegetation indices (VIs) at the hilling stage in 2002, but not in 2003. At the post-hilling stage, treatment differences in VI values were minimal and insignificant except very late in the growing season. SPAD meters could be used as an indirect method for detecting N deficiency at the hilling stage when making supplemental N applications, but they are not as sensitive as the petiole sampling method. The sensitivity of QuickBird imagery to canopy N variations needs to be further tested with more pixel data. However, cloud interference and high cost of images could limit the use of QuickBird data in making timely management decisions.     

冬小麦叶面积指数真实性检验尺度扩展方法研究

为在空间尺度上实现冬小麦LAI地面观测与遥感观测直接匹配,从1 m×1 m范围的实测LAI出发,通过优化采样方法扩展得到16 m×16 m范围的冬小麦LAI,然后利用空间分辨率为16 m的高分1号卫星的多光谱数据计算样本点的植被指数,建立其与冬小麦LAI的拟合模型,从四种植被指数的拟合模型中挑选表现最好的LAI估测模型,获得16 m×16 m尺度的LAI分布图,并经过重采样聚合为250 m×250 m尺度的LAI格点图,从而实现从地面点测量数据到卫星尺度数据的扩展。检验结果表明,16 m×16 m和250 m×250 m两个研究区域模拟点值和实测点值的相对误差分别为4.18%和3.64%,说明这种尺度扩展方法是科学可行的。     

利用RADARSAT-2雷达数据与改进的水云模型反演冬小麦叶面积指数

利用微波遥感反演植被参数往往受到植被分布不均、稀疏植被覆盖、地表裸土等因素影响,导致微波遥感用于农业参数估计的效果不佳。为解决微波遥感反演地表植被参数的问题,本研究在原有的水云模型基础上引入植被覆盖度以及裸土对于雷达后向散射系数的直接作用信息,提出一种改进的水云模型,并充分考虑地表植被的覆盖分布情况,结合地面实测数据及RADARSAT-2雷达数据对改进模型进行验证,然后根据改进模型通过查找表法反演出植被含水量,最后利用叶面积指数与植被含水量的经验关系间接得到叶面积指数的估测值。结果表明,改进的水云模型对后向散射系数的模拟精度比原有的水云模型精度高,模拟的决定系数在HH和VV极化时分别为0.850和0.739,均方根误差分别为0.918dB和1.475dB。由此可见,改进的模型对研究区植被条件更为敏感,能够较好地分离出植被与土壤信息对雷达后向散射系数的影响,同时利用其反演得到的叶面积指数精度较高,决定系数达到0.841,均方根误差为0.233。     

基于无人机RGB影像的小麦叶面积指数与产量估算研究

为探讨基于无人机RGB影像实现对小麦叶面积指数(leaf area index, LAI)和产量估算的可行性,设置不同生态点、品种和氮素处理的小麦田间试验,应用大疆精灵4 Pro无人机获取小麦拔节期、抽穗期、扬花期和灌浆期4个主要生育时期的RGB高时空分辨率影像,并同测定小麦LAI。采用相关性分析筛选出不同生育时期对LAI敏感的光谱与纹理特征集,并借助随机森林(random forest, RF)、偏最小二乘回归法(partial least squares regression, PLSR)、BP神经网络(back propagation neural network, BPNN)和支持向量机(support vector machine, SVM)分析方法,筛选出小麦不同生育时期最优的LAI估测模型。基于不同生育时期的光谱与纹理特征以及时期特征集,进一步建立产量预测模型,并在不同生态点验证叶面积估算模型与产量预测模型的普适性。结果表明,基于RF的LAI估测模型的验证精度最高,4个生育时期的均方根误差(root mean square error, RMSE)分别为2.26、1.44...     

New Vegetation Index and Its Application in Estimating Leaf Area Index of Rice

Leaf area index (LAI) is an important characteristic of land surface vegetation system, and is also a key parameter for the models of global water balancing and carbon circulation. By using the reflectance values of Landsat-5 blue, green and red channels simulated from rice reflectance spectrum, the sensitivities of the bands to LAI were analyzed, and the response and capability to estimate LAI of various NDVIs (normalized difference vegetation indices), which were established by substituting the red band of general NDVI with all possible combinations of red, green and blue bands, were assessed. Finally, the conclusion was tested by rice data at different conditions. The sensitivities of red, green and blue bands to LAI were different under various conditions. When LAI was less than 3, red and blue bands were more sensitive to LAI. Though green band in the circumstances was less sensitive to LAI than red and blue bands, it was sensitive to LAI in a wider range. When the vegetation indices were constituted by all kinds of combinations of red, green and blue bands, the premise for making the sensitivity of these vegetation indices to LAI be meaningful was that the value of one of the combinations was greater than 0.024, i.e. visible reflectance (VIS)>0.024. Otherwise, the vegetation indices would be saturated, resulting in lower estimation accuracy of LAI. Comparison on the capabilities of the vegetation indices derived from all kinds of combinations of red, green and blue bands to LAI estimation showed that GNDVI (Green NDVI) and GBNDVI (Green-Blue NDVI) had the best relations with LAI. The capabilities of GNDVI and GBNDVI to LAI estimation were tested under different circumstances, and the same result was acquired. It suggested that GNDVI and GBNDVI performed better to predict LAI than the conventional NDVI.     

新型植被指数及其在水稻叶面积指数估算上的应用

叶面积指数LAI不仅是陆表植被系统的一个重要属性,而且是全球水平衡、碳循环等模型中的重要输入参数。首先通过使用水稻小区试验冠层光谱数据模拟Landsat 5卫星蓝、绿、红光波段;其次分析了各个波段对LAI的敏感性;然后分析了由这个3个波段的所有组合分别代替常规NDVI中的红光波段构成的VNDVI对LAI变化的反应和对LAI的估算能力;最后使用不同条件下的水稻数据进行验证。结果表明,在不同的LAI范围,红绿蓝光3个波段对LAI有不同的敏感性。当LAI<3时,红蓝光波段敏感性较高。虽然这时绿光波段的敏感性不如红蓝光波段,然而绿光波段在更大的范围对LAI都有相当的敏感性。当采用红绿蓝光波段的各种组合构成植被指数时,如果要使这些植被指数不出现饱和现象,并使对LAI的敏感性有意义,其前提是要求这个波段或是波段组合的值要大于0.024,即VNDI(visible NDVI)公式中的VIS>0024,否则将可能产生饱和现象,而使LAI估算准确度降低。综合比较所有由红绿蓝光波段各种组合构成的植被指数对LAI的估算能力,认为GNDVI和GBNDVI与LAI有比较好的关系。使用其他条件下的水稻数据对各种NDVI的LAI估算能力进行了验证,仍然得到了同样的结论。可见,GNDVI和GBNDVI在估算LAI时确实比传统NDVI具有更好的效果。     

Impact of land‐use intensity on the relationships between vegetation indices,photosynthesis and biomass of intensively and extensively managed grassland fens

Vegetation indices are widely used as model inputs and for non‐destructive estimation of biomass and photosynthesis, but there have been few validation studies of the underlying relationships. To test their applicability on temperate fens and the impact of management intensity, we investigated the relationships between normalized difference vegetation index (NDVI), leaf area index (LAI), brown and green above‐ground biomass and photosynthesis potential (PP). Only the linear relationship between NDVI and PP was management independent (R² = 0·53). LAI to PP was described by a site‐specific and negative logarithmic function (R² = 0·07–0·68). The hyperbolic relationship of LAI versus NDVI showed a high residual standard error (s.e.) of 1·71–1·84 and differed between extensive and intensive meadows. Biomass and LAI correlated poorly (R² = 0·30), with high species‐specific variability. Intensive meadows had a higher ratio of LAI to biomass than extensive grasslands. The fraction of green to total biomass versus NDVI showed considerable noise (s.e. = 0·13). These relationships were relatively weak compared with results from other ecosystems. A likely explanation could be the high amount of standing litter, which was unevenly distributed within the vegetation canopy depending on the season and on the timing of cutting events. Our results show there is high uncertainty in the application of the relationships on temperate fen meadows. For reliable estimations, management intensity needs to be taken into account and several direct measurements throughout the year are required for site‐specific correction of the relationships, especially under extensive management. Using NDVI instead of LAI could reduce uncertainty in photosynthesis models.     

冬小麦叶面积指数的品种差异性与高光谱估算研究

为给小麦叶面积指数（LAI）的高光谱估算提供技术支持,基于2年大田试验,以4个河南主推品种为材料,对小麦LAI和冠层光谱变化特点、估算模型及其品种间的差异等进行了系统分析。结果表明,在生育期内不同冬小麦品种冠层光谱反射率的变化与LAI变化有差异;在相同LAI下,不同冬小麦品种的光谱曲线存在差异。利用400~900 nm范围内冠层光谱反射率的任意两波段组合的比值光谱指数（RSI）、归一化差值光谱指数（NDSI）和差值光谱指数（DSI）所建立的单品种模型以及不同品种综合模型的决定系数（r）均达到0.84以上,单品种模型的r和调整r分别较综合模型高出3.1%~4.8%和2.0%~4.2%。利用独立于建模样本以外的数据对上述模型进行检验,单品种模型预测的r较综合模型提高了0.6%~11.0%,而均方根误差降低了10.0%~37.0%。因此,在利用高光谱遥感技术估算冬小麦LAI时,可以通过建立单品种模型来提高估算精度。     

联合主成分分析与最小二乘支持向量机估测冬小麦叶面积指数

利用单一植被指数估测叶面积指数存在高光谱遥感丰富的波段信息易丢失和外界因素干扰大的缺点,但若将波段信息全部引入模型又会增加建模难度。为解决利用多波段信息估测叶面积指数的问题,利用主成分分析法(PCA)对光谱数据进行降维,之后将提取的主成分与最小二乘支持向量机(LS-SVM)模型相结合,构建冬小麦叶面积指数的高光谱估测模型,并与以4类植被指数作为LS-SVM输入参数建立的模型进行比较。结果表明,以主成分作为LS-SVM模型的输入参数建立的模型精度最高,模型检验集R2为0.71,检验集RMSE为0.56,估测结果较使用植被指数作为输入参数建立的模型精度高,稳定性好。该方法可为利用多波段信息进行大范围冬小麦叶面积指数的无损测定提供参考。     

监测花生叶面积指数和地上部生物量的最优植被指数及适宜波段带宽

为推动光谱遥感在快速无损监测花生生长中的应用，明确监测花生叶面积指数（leaf area index，LAI）和地上部生物量（aboveground biomass，AGB）的最优植被指数及适宜的核心波段带宽。设置2个花生品种、4个施氮水平的花生田间试验，在不同生育时期（苗期、开花下针期、结荚期、成熟期）用Analytical Spectral Devices（ASD）公司生产的FieldSpec HandHeld 2型野外高光谱辐射仪，采集325~1075 nm范围冠层反射光谱，筛选敏感植被指数，并研究核心波段带宽对其监测叶面积指数（LAI）和地上部生物量（AGB）时精度的影响。结果显示，对花生LAI和AGB敏感的植被指数均为归一化红边指数（normalized difference red edge），即NDRE（λ₇₉₀, λ₇₂₀）。进一步分析这一指数的监测精度随波段带宽的变化，发现监测LAI时，核心波段带宽（bandwidth，b）在（λ₇₉₀：1~33 nm，λ₇₂₀：41~59 nm）范围内时能使NDRE（λ₇₉₀, λ₇₂₀）保持较高监测精度，其中带宽组合（λ₇₉₀：33 nm，λ₇₂₀：53 nm）的带宽和值最大，对核心波段带宽的要求最低，利用其构建监测模型时决定系数（determination coefficient，R²）为0.7482，利用独立试验数据检验模型时相对均方根误差（relative root mean square difference，RRMSE）为13.88%。监测花生AGB时，核心波段带宽在（λ₇₉₀：1~101 nm，λ₇₂₀：19~101 nm）范围内时能使NDRE（λ₇₉₀, λ₇₂₀）保持较高的监测精度，其中带宽和值最大的核心波段带宽组合为（λ₇₉₀：89 nm，λ₇₂₀：89 nm），其建模R²为0.7103，检验RRMSE为20.42%。综上，在花生整个生长进程中，可用上述两个具有不同核心波段带宽的植被指数NDRE（λ₇₉₀-b₃₃, λ₇₂₀-b₅₃）和NDRE（λ₇₉₀-b₈₉, λ₇₂₀-b₈₉）分别对LAI和AGB进行监测，监测模型为LAI = 0.0296 × exp^{（14.365×NDRE）}和AGB = 0.6240 × exp^{（20.222×NDRE）}。在核心波段适宜带宽上的研究结果，可以为花生长势光谱监测设备研发及评估提供参考。     

基于Sentinel-2多光谱数据和机器学习算法的冬小麦LAI遥感估算

为了丰富大田尺度下冬小麦叶面积指数的遥感估算方法并提高估算精度,以关中地区冬小麦为对象,基于Sentinel-2多光谱卫星数据与地面同步观测的冬小麦叶面积指数样点数据,应用偏最小二乘回归(PLSR)、反向传播神经网络(BPNN)和随机森林(RF)法构建冬小麦叶面积指数估算模型,进行区域冬小麦叶面积指数遥感反演。结果表明,Sentinel-2多光谱卫星影像中心842nm近红外B8波段与冬小麦叶面积指数相关性最好,样本总体相关系数为0.778;植被指数中反向差值植被指数(IDVI)与冬小麦叶面积指数相关性最好,样本总体相关系数为0.776。各种估算模型中LAI-RF模型预测效果最佳,r~2为0.72,RMSE为0.53,RE为16.83%。基于LAI-RF估算模型,应用Sentinel-2多光谱卫星数据较好地反演了研究区冬小麦叶面积指数区域分布,其结果总体上与地面真实情况接近,说明以Sentinel-2卫星影像数据建立LAI-RF估算模型,可应用于区域冬小麦LAI反演制图。     

Deep neural network algorithm for estimating maize biomass based on simulated Sentinel 2A vegetation indices and leaf area index

Accurate estimation of biomass is necessary for evaluating crop growth and predicting crop yield.Biomass is also a key trait in increasing grain yield by crop breeding.The aims of this study were(i)to identify the best vegetation indices for estimating maize biomass,(ii)to investigate the relationship between biomass and leaf area index(LAI)at several growth stages,and(iii)to evaluate a biomass model using measured vegetation indices or simulated vegetation indices of Sentinel 2A and LAI using a deep neural network(DNN)algorithm.The results showed that biomass was associated with all vegetation indices.The three-band water index(TBWI)was the best vegetation index for estimating biomass and the corresponding R2,RMSE,and RRMSE were 0.76,2.84 t ha−1,and 38.22%respectively.LAI was highly correlated with biomass(R2=0.89,RMSE=2.27 t ha−1,and RRMSE=30.55%).Estimated biomass based on 15 hyperspectral vegetation indices was in a high agreement with measured biomass using the DNN algorithm(R2=0.83,RMSE=1.96 t ha−1,and RRMSE=26.43%).Biomass estimation accuracy was further increased when LAI was combined with the 15 vegetation indices(R2=0.91,RMSE=1.49 t ha−1,and RRMSE=20.05%).Relationships between the hyperspectral vegetation indices and biomass differed from relationships between simulated Sentinel 2A vegetation indices and biomass.Biomass estimation from the hyperspectral vegetation indices was more accurate than that from the simulated Sentinel 2A vegetation indices(R2=0.87,RMSE=1.84 t ha−1,and RRMSE=24.76%).The DNN algorithm was effective in improving the estimation accuracy of biomass.It provides a guideline for estimating biomass of maize using remote sensing technology and the DNN algorithm in this region.     

基于PROSAIL模型和无人机高光谱数据的冬小麦LAI反演

为及时准确高效监测小麦叶面积指数（leaf area index,LAI）,获取了冬小麦挑旗期和开花期地面实测光谱与无人机高光谱遥感影像数据,并基于查找表建立PROSAIL辐射传输模型得到冬小麦冠层模拟光谱数据,利用数学统计回归模型与偏最小二乘回归法分别构建冬小麦LAI单变量、多变量预测模型,以实测LAI数据对预测结果进行精度评价,将最佳预测模型应用于无人机高光谱影像以分析LAI空间分布情况。结果表明,冬小麦各生育时期的预测模型均具有较高的预测精度,单变量预测模型和多变量预测模型的决定系数分别为0.598~0.717和0.577~0.755,其中以基于植被指数的多变量预测模型表现最优,其在开花期的验证精度最高,RMSE和MAPE分别为0.405和12.90%。在LAI空间分布图中,开花期预测效果优于挑旗期,各试验小区的LAI分布较为均匀。     

Non-destructive estimation of potato leaf area index using a fish-eye radiometer

Summary Leaf area index (LAI) is widely used in many facets of potato (Solanum tuberosum L.) modelling but direct measurements have historically been difficult. This investigation tested the accuracy of a commercially available instrument (LI-COR LAI-2000) for measuring LAI non-destructively on a potato crop. Accurate estimates of LAI were difficult to obtain with small plots of≈1 m². Results from larger field plots were extremely favourable and indicate that non-destructive measurements of LAI in situ can routinely be estimated within 5 to 10% of the destructively measured LAI. Six thinning tests performed on four potato cultivars produced average root mean square error measurements of LAI that ranged from 0.09 to 0.27.     

玉米冠层内光合有效辐射和叶面积指数垂直分布模拟

基于辐射度模型(RGM),考虑冠层结构如垄宽、垄间距等建立玉米冠层内不同太阳高度角PAR垂直分布计算模型,结合指数递减光分布模型,考虑LAI与植被冠层内光分布的关系,运用Campbell椭球分布算法和BonhommeChartier算法两种算法分别计算LAI垂直分布,并就模型的参数如太阳高度角等对PAR垂直分布结果的影响进行分析。结果表明,RGM模型不同太阳高度角对封垄前的玉米冠层内PAR垂直分布的模拟精度均较高,60°太阳高度角精度比较高,顺垄和垂直于垄方向的RMSE值分别为0.037 307和0.064 702;两种算法对LAI垂直分布估算能力均较好,不同入射光条件下估算精度不同,Campbell椭球分布算法60°太阳高度角模拟各层LAI垂直分布精度更高。     

Discrimination of sterile oat (Avena sterilis) in winter barley (Hordeum vulgare) using QuickBird satellite images

Site-specific weed management implies detecting the location of weeds in order to generate maps of their spatial distribution. This information facilitates a more accurate application of herbicides, spraying them in the exact areas of weed growth and in the required doses. In order to explore the potential of commercial satellites to discriminate and map weeds, we used the information contained in high spatial resolution images acquired by the QuickBird satellite to assess the density of sterile oat (Avena sterilis) present in a winter barley field at two different dates (March and June). Our results confirmed the potential of using satellite images in the spectral discrimination of weed patches in infested fields. The results of binary logistic regressions showed that the best matches in the classification of three categories (low, medium, or high sterile oat densities) corresponded to the March image. QuickBird’s March image provided reliable estimates of sterile oat patches in barley crops when weed density was relatively high (between 86% and 94% of agreement between predicted and observed densities). However, when weed densities were lower than 10 plants/m² there were serious difficulties to distinguish them from weed-free zones (between 72 and 75% of global agreement in the classification) with large underestimation of medium density weed patches (10 plants/m²). This is a potential limitation considering than the thresholds used for herbicide application decisions are generally close to this density. However, the information obtained may still be useful for producing field maps to describe the spatial distribution of this weed. Moreover, these studies have provided valuable information on the best spectral regions and/or vegetation indices for approaching discrimination between sterile oat and cereal crops and the most suitable period for it.     

数据挖掘在玉米精准作业中的应用

随着农业信息系统的建立和农业数据的增长,对于海量数据的分析和处理变得越来越重要,这使得数据挖掘技术在农业领域特别是精准农业中得到了广泛应用。介绍了数据挖掘在玉米精准作业中的应用,包括玉米决策支持系统、玉米估产、玉米精准施肥、土壤性质空间分布分析中的应用等。     

不同类型夏玉米主要性状及产量的分析

以4种不同类型的夏玉米品种为材料,研究了不同时期其主要性状、干物质积累和分配及产量。结果表明：4种类型夏玉米的产量存在显著差异,在不同密度下产量也表现显著差异,其中大株特大穗型玉米品种CF2187的产量最高,具有相当大的生产潜力;4种类型夏玉米的LAI发展受密度和株型影响,高密度较低密度拥有更大的LAI及持续期,大株大穗型玉米品种的叶面积持续期最长;大株大穗型玉米品种的干物质积累最多,中株中大穗型玉米品种的经济系数最大。     

基于无人机影像的小麦株高与LAI预测研究

为快速、准确地估测不同生育时期小麦品种（系）株高与叶面积指数（LAI）表型性状，基于各生育时期小麦品种（系）数字正射影像（digital orthophoto map，DOM）和数字表面模型（digital surface model，DSM），分别构建不同生育时期株高估测模型和光谱指数LAI估测模型。借助一元线性回归、多元逐步回归（SMLR）和偏最小二乘回归（PLSR）分析方法，并采用决定系数（r）、均方根误差（RMSE）和归一化均方根误差（nRMSE）综合性评价指标，筛选出小麦不同生育时期最优的株高和LAI估测模型。结果表明，（1）全生育期株高估测效果最好，模型预测值与实测值高度拟合（r、RMSE、nRMSE分别为0.87、5.90 cm、9.29%）；在各生育时期中，灌浆期模型预测精度较好，成熟期预测精度最差，r分别为0.79和0.69。（2）所选的18种光谱指数与LAI相关性均较好，其中BGRI、RGBVI、NRI和NGRDI的相关系数达到极显著水平，且各时期三种回归估测模型均表现出较高的稳定性和拟合效果，其中SMLR回归模型对各生育时期LAI预测精度最好，其拔节期、孕穗期、扬花期、灌浆期和成熟期的预测集r分别为0.68、0.57、0.61、0.68和0.53。这说明，基于无人机获取的不同生育时期小麦DSM影像提取株高，并运用18种光谱指数构建LAI估测模型方法是可行的。     

多遥感光谱指标结合进行大田冬小麦叶片叶绿素含量估测研究

为解决大田冬小麦叶片叶绿素含量估测模型精度低、通用性弱的问题,在获取冬小麦拔节期和抽穗期冠层红光波段反射率(BR_red)和近红外波段反射率(BR_nir)的基础上,计算归一化差值植被指数(NDVI)、差值植被指数(DVI)、比值植被指数(RVI)、土壤调节植被指数(SAVI)、改进型比值植被指数(MSR)、重归一化植被指数(RDVI)、II型增强植被指数(EVI2)和非线性植被指数(NLI)等8个植被指数。经统计分析,选择与叶片叶绿素含量(SPAD值)相关性较好的5个遥感光谱指标(NDVI、MSR、NLI、BR_red和RVI)作为输入变量,建立了冬小麦叶片叶绿素含量的BP神经网络估测模型(WWLCC_BP),并对估测模型进行精度验证。结果表明,WWLCC_BP估测模型在拔节期估测的决定系数(r²)为0.84,均方根误差(RMSE)为5.39,平均相对误差(ARE)为9.87%。抽穗期的估测效果与拔节期较为一致。将WWLCC_BP和高分六号影像...