The Relationship between the Chlorophyll Concentration and Spectral Signatures of Cinnamomum camphora Leaves

• Main Authors: Shu-Ching Huang, 黃淑清
• Other Authors: Chinsu Lin
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/83082447924466672790

碩士 === 國立嘉義大學 === 森林暨自然資源學系研究所 === 96 === The concentration of foliar chlorophyll is relative to photosynthesis. The use of remote sensing technique can not only evaluate the concentration of foliar chlorophyll but also provide the applications for wide range of ecosystems and driving ecosystem simulation model over large area. The remote sensing technique depends on spectrum information, but the resolution of traditional spectra limited the investigation of resource. However, the hyperspectral method exhibited excellent character of accuracy that overcame the problem of traditional measurement. This study used an ASD spectroradiometer to conduct a field measurement of Cinnamomum camphora leaves reflectance and measurement chlorophyll concentration. Furthermore, the multiple stepwise regression analysis with absorption-feature area, continuum removal, the first derivative reflectance, and six vegetation index (such as NDVI, SR, NDVI-G, TVI, OSAVI, MCARI) were used. The spectral behavior model of the tested samples was defined and predicted by the chlorophyll experimental results. The relation between reflectance of spectrum and concentration of total chlorophyll was established via the feature of absorption spectrum for foliar chlorophyll, and then proceeded the verification of hyperspectral model. The reflex intensity of wavelength at 690, 689, and 656nm of red spectra are the key feature for chlorophyll estimation. The experimental results from the multiple regression analysis showed that the coefficient of determination in chlorophyll prediction is 0.90. Particular signatures of the red bands derived from continuum removal reflectance (CR signatures), such as 705nm, the absorption-feature area of red bands ranged from 550nm-750 nm and first derivative reflectance also has 0.86-0.94 the coefficient of determination in chlorophyll prediction. And then NDVI vegetation index R2 is 0.91. Among those features the first derivative reflectance is the best variable for chlorophyll prediction.