基于数码相片Gamma校正的水稻叶面积指数估算

随着数码相机的日益普及,利用数码相机进行作物叶面积指数(LAI)测量不断得以应用。由于数码相机成像时会对入射光辐射强度进行Gamma编码变换,输出的相片DN(Digital Number)值与入射光辐射强度呈非线性关系,会造成在确定相片中植被叶片与背景的分割阈值时出现误差,并最终导致LAI估算存在较大不确定性。以水稻为研究对象,获取不同生长期水稻冠层相片并结合同步LAI 2000测量的LAI数据,基于相片Gamma校正原理,对水稻不同生长期冠层相片进行Gamma校正,在此基础上利用冠层孔隙率方法,估算不同生长期水稻LAI。结果表明,经过Gamma校正相片估算的水稻LAI总体精度有显著提高,相片估算的IMAGE LAI与LAI-2000测量值比较的决定系数达到0.71(P0.05)。在整个观测期内,两种方法观测的LAI值在时间变化趋势上表现一致,但在不同生长期内存在差别,在水稻分蘖期和拔节期相片估算的IMAGE LAI要高于LAI-2000测量值,孕穗期到抽穗期期间IMAGE LAI低于LAI-2000测量值,乳熟期到成熟期IMAGE LAI又高于LAI-2000的观测结果。     

基于光学仪器法测定谷地云冷杉林叶面积指数的季节变化

利用半球摄影法(DHP)和LAI-2000植物冠层分析仪两种光学仪器法(间接法)以及凋落物法(直接法),研究了小兴安岭谷地云冷杉林叶面积指数(LAI)的季节变化,并构建了不同季节直接法与间接法测定的LAI间的相关关系.结果表明: 在整个试验期间,DHP测定的LAI比直接法测定值低估40%～48%, LAI-2000植物冠层分析仪的低估范围为15%～26%;不同时期直接法与DHP和LAI-2000植物冠层分析仪测定的LAI均显著相关, 且均可合并为A、B、C 3类预测模型, 可以分别预测5和11月, 6、9和10月, 7和8月的LAI.本研究结果可为高效、准确地测定针叶林LAI的季节变化提供参考.     

利用数字图像估测棉花叶面积指数

叶面积指数是指示棉花长势、产量形成和高产群体调控等信息的重要结构特征参数。本研究的目的在于利用基于冠层图像光照叶片和光照土壤分量的图像透光率估测棉花叶面积指数。通过3年不同种植密度、品种、施氮量和灌水量的田间试验,在棉花不同的生育期用数码相机、LAI-2000冠层仪和线性光量子传感器采集数据并进行破坏性取样测定,分析图像透光率的有效性和建立LAI估测模型,进而对图像方法、LAI-2000和破坏性取样方法进行比较和分析。结果表明：(1)在太阳高度角最大且变化最小的正午时段,数码相机测量的图像透光率与线性光量子传感器测量的冠层透光率较一致且相对稳定。(2)图像透光率能反映除吐絮期以外各时期的冠层透光状况,但是当LAI大于5时图像透光率出现饱和。(3)综合分析2009和2010年数据,建立了图像透光率估测LAI的模型(R2=0.8438, SE=0.5605);利用2007年独立试验资料检验估测模型的性能,模型检验的拟合度较高(R2=0.8767)且预测误差较小(RMSE=0.4305),当LAI＞5时模型的预测能力降低。(4)数字图像、LAI-2000和破坏性取样三种方法测量的LAI值之间均呈现显著的线性相关(R2＞0.85),但是图像透光率的饱和性致使当LAI＞5时明显低估叶面积指数。     

黄土高原冬小麦田光谱特征变化及其与二氧化碳日收支的相关分析

根据光谱辐射仪对黄土高原冬小麦整个生育期光谱反射率的连续观测数据及CO₂通量观测数据,对冬小麦田光谱特征变化及其与CO₂日收支的相关性进行了分析.结果表明：冬小麦田不同波长光谱反射率和归一化植被指数（NDVI）呈现明显的日变化和季节变化.同一天内,反射率随太阳高度角的变化而变化,变化最大的波段（550 nm左右、700～1 050 nm）表现为峰.不同生育期同一时刻,可见光波段（350～670 nm）反射率变化不大,近红外波段（700～1 050 nm）出现较大差异,在出苗期、分蘖期和越冬期后红边位置向长波方向“红移”;越冬期前出现向短波方向“蓝移” 的现象;但成熟期“蓝移”现象不明显,表现为突变;其他生育时期没有观测到波谱位移.NDVI的日变化呈U型,13：00左右最低,16：00后出现较大波动,与抛物线有较好的拟合效果,小麦生长旺盛时期,对其地面遥感观测应选择在NDVI变化不大的13：00左右进行;整个冬小麦生长季11:00反射率及NDVI以播种后第140天为中心对称,NDVI的季节变化表现为M型,可用四次多项式拟合;在整个小麦生育期中NDVI与CO₂的日收支呈极显著负相关,但正午左右的相关性稍差.     

3类典型温带山地森林的叶面积指数的季节动态: 多种监测方法比较

 叶面积指数(leaf area index, LAI)是定量描述冠层结构的最有效指标之一。鉴于森林冠层三维结构的高度复杂性和异质性, 迄今仍没有形成统一标准的LAI测量方法。该文利用LAI-2000冠层分析仪、CI-110冠层分析仪和半球摄影法(digital hemispherical photograph, DHP), 对北京东灵山地区以蒙古栎(Quercus mongolica)为主的落叶阔叶林、华北落叶松(Larix gmelinii var. principis-rupprechtii)林和油松(Pinus tabuliformis)林的有效叶面积指数(effective leaf area index, LAI_e)进行了动态监测, 探寻其季节变化规律。为准确地估算温带山地主要森林类型的LAI, 对光学仪器测量值进行了去除木质成分、聚集效应等校正, 与基于凋落物收集法的相应实测值进行了比较分析。结果表明: 3种典型森林在生长季期间叶片生长均呈现单峰型; 3种光学仪器测量方法的同期LAI_e数值大小顺序为: LAI-2000冠层分析仪>DHP>CI-110冠层分析仪。光学仪器的直接测量值LAI_e包含了木质成分的贡献, 钝化了季节动态的变化幅度, 这对有明显季节交替的落叶林尤为突出。经校正, LAI-2000冠层分析仪和DHP的测量值与实测值都表现出显著的相关性, 其中LAI-2000冠层分析仪最适于采用基于空隙大小的校正方法, 而基于空隙度和空隙大小的综合算法则是校正DHP的最佳选择。结合经济成本和野外实际操作等因素考虑, DHP具有更大的推广优势, 特别适用于温带山地落叶林。     

利用鱼眼摄像机测量植被叶面积指数动态变化

叶面积指数(LAI)的快速准确测量可以为相关生态过程研究提供关键参数.本研究采用光学鱼眼相机拍摄了黄土高原北部典型植被柠条、沙柳、紫花苜蓿、荒草、大豆和玉米不同生长时期冠层照片,在玉米、大豆和柠条样地固定安装鱼眼摄像机定时拍摄冠层照片,并运用图像处理软件分析获取LAI.结果表明: 鱼眼相机法(DHP)获取的多种植被类型的LAI与LAI-2200测量结果存在显著线性相关关系(R²=0.85,P<0.05),均方根误差为0.256.太阳辐射值直接影响专业软件处理照片时的关键参数,镜头朝下情况下,计算LAI的关键参数——绿色指数随着拍摄照片时太阳辐射强度的增加而增加.但是镜头朝上情况下,亮度参数随着太阳辐射值的增加而减少.通过关键参数的调整,玉米、大豆和柠条样地LAI变化与LAI-2200测量结果一致,很好地反映了这些植物生长期的LAI动态变化,柠条样地镜头朝下测量结果更好.鱼眼摄像机可以用于定位监测LAI动态变化.     

采用不同光谱波段宽度的归一化植被指数估算水稻叶面积指数

通过大田试验,测量了水稻不同生育期的冠层光谱,并通过在TM红光波段和近红外波段范围内不断扩展波段宽度,计算了各个波段宽度对应的归一化植被指数(NDVI)及其与叶面积指数(LAI)的最大相关系数,再根据NDVI与LAI最佳拟合方程的最大R2值随波段宽度的变化特征确定了最佳波段宽度.结果表明:NDVI与LAI的相关系数和最佳拟合方程的最大R2值具有相同的变化趋势,在小于红光波段宽度60nm范围内,均随波段宽度的增大而减小,即窄波段NDVI比宽波段NDVI对LAI的估算效果好;在超过60nm后,由于受近红外波段宽度的影响而略有波动.对R2规律的分析表明,使用NDVI估算水稻LAI的最佳波段宽度为15nm.     

基于塔载辐射监测帽儿山落叶阔叶林冠层叶面积指数动态

基于通量塔常规辐射测量的宽带植被指数(BVI)具有高时间分辨率的优点,有利于获得更详细的森林冠层叶面积指数(LAI)动态信息.本文以帽儿山通量观测站的温带落叶阔叶林为例,研究宽带归一化差值植被指数(NDVI_B)、宽带增强型植被指数(EVIB)、近红外反射率与光合有效辐射反射率比值(SR_(NP))和太阳辐射反射率与光合有效辐射反射率比值(SR_(SP))4种BVI时间序列的控制因子及其滤波方法,并以凋落物收集法为参考,评估采用BVI估测冠层LAI的可行性.结果表明:Huemmrich、Wilson和Jenkins 3种方法计算的同一BVI值略有不同,但其季节变化趋势高度一致.BVI主要受太阳高度角和太阳高度角与坡度夹角的影响而呈现明显的日变化,太阳高度角与坡度夹角最大时刻(12:30)前后的BVI相对稳定.晴空指数可以作为BVI日值滤波的有效参数,不同时刻数据构成的日BVI时间序列的晴空指数阈值以及滤波后的有效数据率存在差异,应综合考虑平滑效果和有效数据率选择合适的时间点代表BVI日值.NDVI_B与凋落物收集法测定的LAI呈显著的线性关系,而EVIB、SR_(NP)和SRSP与LAI均呈显著的对数关系.因此,NDVI_B在表征冠层LAI季节动态和LAI外推中更为精确、方便.鉴于大部分森林碳水通量观测塔配备能量平衡观测系统,如果同时测定光合有效辐射反射率,即可实现冠层LAI长期连续联网监测.     

利用不同分辨率卫星影像的NDVI数据估算叶面积指数(LAI)——以岷江上游为例

短周期的低分辨率遥感数据为大面积估算LAI及季节动态和物候趋势提供了有利工具,但基于高分辨率LAI的遥感估算模型在低分辨率遥感数据上应用有很大的不确定性。研究利用LAI-2000冠层分析仪与跟踪辐射和冠层结构测量仪(TRAC),测定了岷江上游流域范围内490块野外调查样地(50m×50m样方)的LAI数据,结合同期较高精度卫星数据(TM)建立了不同植被类型的LAI-NDVI算法,在经过传感器的相对校正后,将这种算法应用到同期分辨率较低的MODIS数据和SPOT VEGETATION数据上。结果表明,30m分辨率的TM LAI的均值为4.53,250m MODIS LAI的均值为3.55,1000m VGT LAI的均值为4.20,随着栅格分辨率的降低,总体标准差有增加的趋势,并且LAI值也有不同程度的低估,其中MODIS LAI值被低估约22%。但利用TM LAI数据验证MODIS和VGT LAI数据后发现,250m的MODIS数据预测误差在30%左右,1000m的SPOT数据预测误差则高达50%,空间重采样分析表明,栅格分辨率的降低是导致预测误差扩大的主要原因,而这也是岷江流域植被分布破碎化的体现。     

辐射变化对中亚热带杉木人工林净CO2交换的影响

太阳总辐射是影响森林生态系统碳交换的重要因子.为认识辐射变化对杉木人工林碳交换的影响,本研究利用开路式涡度相关系统和气象梯度观测系统测得的CO2通量和气象因子长期定位监测数据,用晴空指数(kt)表示太阳辐射情况,分析了kt对中亚热带杉木人工林生长季(4-10月)净C02交换(NEE)的影响.结果 表明:晴天时的太阳总辐...     

Test of Accuracy of LAI Estimation by LAI-2000 under Artificially Changed Leaf to Wood Area Proportions

The accuracy of LAI-2000 Plant Canopy Analyzer for leaf (LAI) and plant (PAI) area indexes measurements was tested in 20-year-old Norway spruce stand using the reduction of canopy biomass. Needle and branch areas were reduced progressively upward every one meter. Values of effective leaf area index (LAI_e), as an uncorrected product of LAI-2000, were compared with directly estimated LAI and PAI values after each reduction step. LAI-2000 underestimates PAI and LAI values according to LAI-2000 rings readings, and varied proportions between leaf and wood areas. The values of LAI_c have been increased with decreasing of the view angle of the relevant LAI-2000 rings. Therefore, the underestimation of LAI becomes smaller when the readings near the horizon are masked. More accurate results, for projected LAI (LAI_p) calculation, are produced by LAI-2000 when some dense grids of measurement points and the most vertical ring readings (0 –13 °) are used. Correction factor 1.6 is possible to use for unreduced canopy hemi-surface LAI estimation, when the last rings (i.e. 5^th and 4^th rings, 47 –74 °) are excluded. Correction factor of 1.25 can be used to compute LAI_p if the angle readings under 43 °are also masked.     

The ratio of transmitted near-infrared radiation to photosynthetically active radiation (PAR) increases in proportion to the adsorbed PAR in the canopy

The daily total photosynthetically active radiation (400??00?nm, PAR) and near-infrared radiation (700??000?nm, NIR) were measured in the understory beneath the canopy (PARt and NIRt) and above the canopy (PARi and NIRi) of a Japanese cool-temperate deciduous broad-leaved forest during the snow-free period (May to November). The integration of spectral radiation for NIR and that for PAR, and the daily integrations of instantaneous NIR and PAR, reduced the noises from the optical difference in spectrum and from canopy structure heterogeneity, sky condition and solar elevation. PARi/PARt was linearly related to NIRt/PARt (R2?=?0.96). The effect of cloudiness was negligible, because the fluctuation of NIRi/PARi was quite small regardless of season and weather conditions compared with the range of NIRt/PARt in the forest. The ratio of NIRt/PARt beneath the canopy was log-linearly related to the in situ leaf area index (LAI) with a wide range from 0 to 5.25 (R2?=?0.97). We conclude that seasonal changes in fAPAR (=?1???PARt/PARi) and LAI of a canopy can be estimated with high accuracy by transmitted NIRt and PARt beneath the canopy.     

新疆喀纳斯国家自然保护区植被叶面积指数观测与遥感估算

叶面积指数(Leaf Area Index,LAI)是重要的植被结构参数,调控着植被与大气之间的物质与能量交换,在生态环境脆弱的我国西北部开展植被LAI的研究对阐明该地区植被对气候变化和人类活动的响应特征具有重要的科学意义.利用LAI-2200和TRAC仪器观测了新疆喀纳斯国家级自然保护区森林和草地的有效叶面积指数(LAIe)和真实LAI,构建了其遥感估算模型,生成了研究区LAIe和LAI的空间分布图.在此基础上,分析了LAI随地形因子(海拔、坡度、坡向)的变化特征,探讨了将其应用于估算研究区森林生物量密度的可行性,并评估了研究区MODIS LAI产品的精度.结果表明:研究区阔叶林、针阔混交林、针叶林、草地LAIe的平均值分别为4.40、3.18、2.57、1.76,LAI的平均值分别为4.76、3.93、3.27、2.30.LAIe和LAI的高值主要集中分布在湖泊和河流附近;植被LAI随海拔、坡度和坡向的变化表现出明显的垂直地带性的特点.LAI随海拔和坡度的增加呈现先增加后减小的变化趋势,坡向对针叶林和草地LAI的影响明显,但对阔叶林和针阔混交林LAI的影响较弱;森林生物量密度(BD)随LAI增加而线性增加(BD=44.396LAI-25.946,R2=0.83),研究区森林生物量密度平均值为120.3 t/hm2,估算的总生物量为5.0×l06 t;MODIS LAI产品与利用TM数据生成的LAI之间具有一定的相似性(森林R2=0.42,草地R2=0.53),但森林和草地的MODIS LAI产品分别比利用TM数据生成的LAI偏低16.5％和24.4％.     

三种回归分析方法在Hyperion影像LAI反演中的比较

借助GPS进行地面精确定位,利用LAI-2000冠层分析仅在攸县黄丰桥林场开展130个样地(60m×60m)的叶面积指数(Leaf Area Index,LAI)测量.采用FLAASH模块对Hyperion数据进行大气校正并与地面同步冠层观测数据进行拟合,通过研究地面实测LAI与Hyperion影像波段及其衍生的系列植被指数(NDVI、RVI等)的相关性,筛选出估算叶面积指数的植被指数因子.应用曲线估计、逐步回归及偏最小二乘三种回归分析技术分别建立叶面积指数的最优估算模型.结果表明:参与建模的因子中,比值植被指数(RVI)与LAI的相关性最大,敏感性最高,其次是SARVI0.1,NDVI705,NDVI,SARVI0.1,SARVI0.25;曲线估计、逐步回归分析和偏最小二乘回归三种分析方法所建的6个回归模型中,偏最小二乘回归的拟合效果最好,预测值与实测值的决定系数R2为0.84、曲线估计的拟合效果最低,预测值与实测值的决定系数R2为0.64;建模精度分析表明,选用5-6个自变量因子进行LAI建模是可靠的,以6个植被因子建立的偏最小二乘回归模型预测精度最高.     

An indirect method of estimating leaf area index in a tropical deciduous forest of India

Rapid, reliable and meaningful estimates of leaf area index (LAI) are essential to functional characterization of forest ecosystems including biomass and primary productivity studies. Accurate LAI estimates of tropical deciduous forest are required in studies of regional and global change modeling. Tropical deciduous forest due to higher species richness, multiple species association, varied phenophases, irregular stem densities and basal cover, multistoried canopy architecture and different micro-climatic conditions offers dynamism to the understanding of the LAI dynamics of different PFTs in an ecosystem. This investigation reports a new indirect method for measurement of leaf area index (LAI) in a topical moist deciduous forest in Himalayan foothills using LAI-2000 Plant Canopy Analyzer. We measured the LAI in two seasons (summer; leaf senescence stage and post-monsoon; full green stage) in three (dry miscellaneous, sal mixed and teak plantations) plant functional types (PFT) in Katerniaghat Wildlife Sanctuary, India. Ground LAI values ranged between 2.41 and 6.89, 1.17 and 7.71, and 1.92 and 5.19 during post-monsoon season and 1.36–4.49, 0.67–3.1 and 0.37–1.83 during summer season in dry miscellaneous, sal mixed and teak plantation, respectively. We observed strong correlation between LAI and community structural parameters (tree density, basal cover and species richness), with maximum with annual litter fall (R² > 0.8) and aboveground biomass (AGB) (R² > 0.75). We provided equations relating LAI with AGB, which can be utilized in future studies for this region and can be reasonably extrapolated to other regions with suitable statistical extrapolations. However, the relations between LAI and other parameters can be further improved with incorporation of data from optimized and seasonal sampling. Our indirect method of LAI estimation using litter fall as a proxy, offers repetitive potential for LAI estimate in other PFTs with relatively time and cost-effective way, thereby generating quicker and reliable data for model run for regional and global change studies.     

Improving accuracy of optical methods in estimating leaf area index through empirical regression models in multiple forest types

Key message

We developed the empirical regression models relating the direct LAI and optical LAI from initial leaf out to the leaf fall in different forest types in China.

Abstract

Optical methods have usually been used to estimate the leaf area index (LAI) in a forest stand because of rapidity and reduced labor requirements. However, few studies have reportedly improved the accuracy of the optical LAI estimates for seasonal dynamics using empirical models in different forest types. In the present study, we directly measured the seasonal dynamics of LAI from leaf out to leaf fall based on litter collection (defined as direct LAI) in a mixed evergreen–deciduous forest, an evergreen forest and a deciduous forest. Meanwhile, the effective LAI was estimated using digital hemispherical photography (DHP) and LAI-2000 instruments. Our main objective was to explore the seasonal changes in the relationship between direct LAI and effective LAI values and to find the best LAI empirical estimation model in different forest types. The season-dependent models relating direct LAI and effective LAI in each period were developed through a power function regression model in several forest types. Then, significance tests were applied to compare the different season-dependent models. The analysis showed that the season-dependent models can be merged into different aggregated models depending on forest types and optical methods. We confirm that the seasonal changes in LAI in different forest types can be fully estimated through aggregated models using both DHP and LAI-2000 methods with accuracies of more than 87 and 92 %, respectively. Meanwhile, our results suggest that the forest type (i.e., species composition of forest stand) and optical method should be seriously considered to correctly and quickly estimate the seasonal changes of LAI through the aggregated models.

     

Daylight spectra (400–740 nm) beneath sunny,blue skies in Tasmania,and the effect of a forest canopy

Abstract The photosynthetically active radiation (PAR) incident on a horizontal surface at an open mountain site is positively correlated with solar altitude for sunny, blue sky conditions. The proportion of red light in PAR decreases with increasing solar altitude, while that of blue increases. These results are consistent with the wavelength dependency of Rayleigh and Mie scattering. The ratio of near infrared radiation to PAR decreases with increasing solar altitude towards solar noon and with decreasing solar altitude towards sunset. Thus surface reflection seems to be an important part of the light climate. The relative transmission of daylight through a forest canopy to a horizontal surface is not correlated with solar altitude for sunny, blue sky conditions at a mountain site. The amount of diffuse daylight is negatively correlated with per cent canopy interception, and the amount of direct sunlight is negatively correlated with per cent solar track interception. Daylength is negatively correlated with both canopy and solar track interceptions. The proportion of red light in PAR increases with increasing solar altitude, while that of blue decreases. These results are opposite those for the open site and are due to the spatial patterns of canopy obstruction of the sky vault, and of the spectral quality of daylight across the sky. The ratio of near infrared radiation to PAR in shadelight increases with increasing canopy interception due to the selective scattering properties of the canopy. The ratio for shadelight is positively correlated with the ratio for sunflecks.