Research on the Effects of Leaf Biochemical Concentrations on Leaf Spectra:Case Study of Inversion of C:N Ratio Based on the Absorption Features Centered at 2100nm

SHI Run-he1; NIU Zheng2; ZHUANG Da-fang1

doi:10.11834/jrs.20050101

Views : 0 下载量: 800 CSCD: 21

PDF
Export
Share
Collection
Album

Research on the Effects of Leaf Biochemical Concentrations on Leaf Spectra:Case Study of Inversion of C:N Ratio Based on the Absorption Features Centered at 2100nm
Issue 1, Pages: 1-7(2005)
Published：2005
DOI： 10.11834/jrs.20050101
稿件说明：

移动端阅览

[1]施润和,牛铮,庄大方.叶片生化组分浓度对单叶光谱影响研究——以2100nm吸收特征的碳氮比反演为例[J].遥感学报,2005(01):1-7. DOI： 10.11834/jrs.20050101.

SHI Run-he1, NIU Zheng2, ZHUANG Da-fang1. Research on the Effects of Leaf Biochemical Concentrations on Leaf Spectra:Case Study of Inversion of C:N Ratio Based on the Absorption Features Centered at 2100nm[J]. Journal of Remote Sensing, 2005, (1): 1-7. DOI： 10.11834/jrs.20050101.

摘要

运用连续统去除法和波深归一化方法对构成植物叶片碳氮比的基本因素——总碳和总氮浓度进行深入研究

对其组成物质的浓度差异在2030—2220nm区间内光谱响应的物理机制进行深入分析。在对多组典型叶片的对比分析后发现

经连续统去除后的相对反射率光谱中可以明显观察到碳、氮浓度差异造成的影响

而其影响与纤维素、木质素和蛋白质在此范围的吸收特征密切相关。其中

以纤维素和木质素为代表的碳浓度的影响主要体现在2030—2050nm区间

而以蛋白质为代表的氮浓度的影响则体现在2054nm和2172nm附近存在的斜率突变上。另外利用波深归一化方法再次证明了氮浓度对2054nm和2172nm处光谱特征的影响

进一步证实蛋白质双峰特征在光谱上的响应

并通过相关分析和回归分析验证了相对反射率光谱与碳氮比之间的定量关系。

Abstract

This paper presents an in-depth research on the effects of leaf biochemical concentrations on leaf spectra. Absorption features centered at 2100nm are focused on in this study because the spectra of predominant carbon containing and nitrogen containing components in leaf are different here apparently. Spectral changes due to different carbon or nitrogen concentrations at this waveband are examined with several samples and two rules are concluded by using continuum removed spectra. First

the spectral changes due to different carbon concentrations can be seen from 2030nm to 2050nm

where the continuum-removed spectra are nearly straight and their slopes are indicators to distinguish different carbon concentrations. Second

the spectral changes at 2054nm and 2172nm are good indicators to distinguish different nitrogen concentrations. Correlation analysis and stepwise regressions are employed to verify the relations of C:N ratio and continuum-removed spectra and the results are satisfactory.