Study on the factors affecting reflectance spectra of sweet potato leaves

• Main Authors: Pei-Shan Wang, 王珮珊
• Other Authors: Shin-Shinge Chang
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/43978633397600195748

碩士 === 國立臺灣大學 === 農藝學研究所 === 92 === The subject of this study was to utilize a spectrophotometer to measure the fresh, fresh-dry, depigment and depigment-dry sweet potato leaves. Leaf color of three sweet potato varieties included green, purple and yellow. The absorbance, transmittance and reflectance spectra of three sweet potato varieties leaves were measured. Explored the relationships between spectra and leave thickness, water content and pigment concentration. Calculated the normalized difference vegetation index（NDVI）with the reflectance at 705nm and 750nm. The thickness of three sweet potato varieties leaves was positively correlated with the absorbance, but negative to the transmittance and reflectance in the visible light. In NIR, leaf thickness was positive to the absorbance and reflectance, negative to the transmittance. The absorbance and transmittance spectra of fresh leaves were increased with the decreasing mesophyll water content. The absorbance spectra of depigmented leaves were increased with the increasing water content. The transmittance spectra of depigmented leaves were decreased with the increasing water content. The reflectance spectra were positively correlated with water content in NIR. In visible light, the reflectance of ‘Green’ and ‘Purple’ were negatively correlated with water content, but the ‘Yellow’ were positive. The reflectance of depigmented leaves was negatively correlated with water content. Pigments affected absorbance, transmittance and reflectance spectra of sweet potato leaves. The relative absorbance at 675nm in fresh leaves was positively correlated with chlorophyll content. The ‘Purple’ sweet potato leaves had abundant anthocyanin, its absorbance peak was around 550nm. The results also revealed that leaf transmittance was affected by depigmentation. Otherwise, the first order of reflectance indicated that reflectance spectra were affect by pigments. The peak around 700nm moved to the shorter wave when the chlorophyll content decreased. Computed the NDVI with the reflectance at R750 and R705. The NDVI was significantly correlated with leaf pigment such as chlorophyll and carotenoids, the R2 were 0.89-0.95. It can possibly be used for prediction of leaf pigment content.