摘要
叶绿素是评估草地光合作用能力和生理状况的重要指标。高光谱遥感包含丰富的光谱信息,已经成为无损估算草地叶绿素含量的重要手段。然而,通常获取的草地冠层高光谱数据与实测叶片的叶绿素含量存在尺度不匹配问题,导致高光谱反演叶绿素的精度低。为此,提出基于绿被覆盖率上推的天然草地冠层叶绿素高光谱反演方法。以内蒙古呼伦贝尔典型天然草地为研究对象,利用ASD高光谱仪获取的地面冠层高光谱数据、SPAD叶绿素仪采集的叶片叶绿素相对含量实测值和样方手机数码相片,以绿被覆盖率为媒介,将叶片尺度的叶绿素相对含量实测值上推到冠层尺度。结果表明植被指数与SPAD相关性(-0.74~0.76)普遍高于场平均法上推SPAD(-0.63~0.50)。通过原始冠层光谱反射率、一阶导数光谱和42种常用叶绿素光谱指数构建基于绿被覆盖率上推的天然草地冠层叶绿素高光谱反演模型(SPAD_cover)。单变量的最优草地冠层叶绿素反演模型R^(2)=0.689,RMSE=2.714,RPD=1.752;多元线性逐步回归的最优草地冠层叶绿素反演模型R^(2)=0.833,RMSE=2.019,RPD=2.354。实验表明基于绿被覆盖率将草地叶片叶绿素含量实测值上推至冠层尺度,可以有效提高天然草地冠层叶绿素高光谱反演精度。
Chlorophyll is a crucial indicator for assessing grasslands'photosynthetic capacity and physiological condition.With its rich spectral information,hyperspectral remote sensing has become an important means for non-invasively estimating chlorophyll content in grasslands.However,there is a scale mismatch between the canopy hyperspectral data and the measured leaf chlorophyll values,leading to hyperspectral chlorophyll retrieval's low accuracy.Therefore,this paper proposes a hyperspectral retrieval method for natural grassland Canopy Chlorophyll based on the green cover rate.The typical natural grassland in Hulunbuir,Inner Mongolia,was selected as the research object.The measured leaf chlorophyll relative content values were obtained by ASD hyperspectral spectrometer,SPAD chlorophyll meter,and mobile phone digital photos.The results indicate that the correlation between vegetation indices and SPAD ranges from-0.74 to 0.76,which is generally higher than the average correlation of SPAD pushed up from-0.63 to 0.50.Green cover media pushed the measured values of leaf chlorophyll relative content to the sample canopy scale.First derivative spectra and 42 common chlorophyll spectral indices were used to construct a hyperspectral retrieval model(SPAD)of natural grassland Canopy Chlorophyll based on green cover rate_cover.The single variable optimal grassland Canopy Chlorophyll retrieval modelR^(2)=0.689,RMSE=2.714,RPD=1.752;The best regression model of grassland Canopy Chlorophyll wasR^(2)=0.833,RMSE=2.019,RPD=2.354.The results show that the hyperspectral retrieval accuracy of chlorophyll content in natural grassland canopy can be effectively improved by extrapolating the measured value of chlorophyll content in grassland leaves to the canopy scale based on the green cover rate.
作者
张爱武
李梦南
史剑聪
庞海洋
ZHANG Ai-wu;LI Meng-nan;SHI Jian-cong;PANG Hai-yang(Key Laboratory of 3D Information Acquisition and Application,Ministry of Education,Capital Normal University,Beijing 100048,China;Engineering Research Center of Space Information Technology,Ministry of Education,Capital Normal University,Beijing 100048,China;Center for Geographic Environment Research and Education,Capital Normal University,Beijing 100048,China)
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2024年第12期3513-3523,共11页
Spectroscopy and Spectral Analysis
基金
青海省科技成果转化项目(2022-NK-136)
北京市教委-市自然基金联合项目(KZ202110028044)
国家自然科学基金(42071303,41571369)资助。
关键词
天然草地
绿被覆盖率
高光谱
尺度上推
光谱指数
一阶导数光谱
Natural grassland
Green coverage rate
Hyperspectral
Scaling up
Spectral index
First derivative spectrum
