Effects Of Leaf Hair On Estimation Of Chlorophyll Content With Hyperspectfal Vegetation Indices

Chlorophyll content is often estimated with hyperspectral vegetation indices todetermine the health status of plant. But the leaf surface structures, such as leaf hair,are seldom considered when most of the vegetation index are developed,.In this study, a variety of plants with different hair structure were selected toanalyze the effects of leaf hair on the leaf reflectance, hyperspectral vegetation indicesand the estimation of chlorophyll content with vegetation indices.The results are mainly as follow:Firstly, we found that the reflectance in400-1000nm wavelength had decreased.But the extents of decreasing are not consistent that the reflectance decreased invisible wavelength （400-700nm） more than in near infrared wavelength（700-1000nm）.Secondly, by comparing with38existing hyperspectral vegetation indices beforeand after hair treatment, we found that CTR1: R₆₉₅/R₄₂₀,REP, VOG1: R₇₄₀/R₇₂₀, whichare composed of only visible wavelength, or only near infrared wavelength are lessaffected by leaf hair. Two ratio indices were selected because the wavelength theyapplied had changed consistent before and after the hair treatment. These threevegetation indices are considered robust in estimating plant biochemistry.Thirdly, the greater density the plant surface hairs has, the greater effect on thehyperspectral vegetation index of the plant.Last, by comparing with the measured chlorophyll contents and the predictedones with the hyperspectral vegetation indices, we found that (R₈₅₀-R₇₁₀)/(R₈₅₀-R₆₈₀)is a relatively stable vegetation index when the regression coefficient and the slope ofthe regression line were checked.