Faculty and Student Publications
Document Type
Article
Publication Date
8-11-2022
Abstract
Aim of study: The aim of this study was to verify the differences in VIS/NIR spectra and leaf color parameters of leaves of Eucalyptus badjensis, E. benthamii, E. dunnii, E. grandis, E. globulus and E. saligna, at four ages, and their influence on species discrimination. Area of study: São Mateus do Sul, Paraná, Brazil. Materials and methods: Seedlings of the six species, with four replicates for each, were acclimatized in the same environment, in Oc-tober 2015, in an entirely randomized design. Leaf samples were collected from plants that were 6, 8, 10 and 12 months old. Three leaves from each of four plants at each age were analyzed. Five parameters were recorded referring to the adaxial surface of each leaf, with a total of 15 records from repetitions and 60 per species at each age. The evaluation was performed in the spectral ranges from 360-740 nm (VIS) and 1000-2500 nm (NIR). Principal component analysis and linear discriminant analysis were performed. Main results: The influence of age differed within each species. In color data, the parameter with most variation among all samples was chromatic coordinate b*. In reflectance spectra (VIS), age of 12 months provided the best discrimination of species. Second derivative NIR spectra produced the best results of external prediction of Linear Discriminant Analysis models based on leaves of 12-month-old trees. Research highlights: Observation of color parameters and VIS/NIR spectroscopy have potential utility for discrimination of Eucalyptus species based on their green leaves.
Relational Format
journal article
Recommended Citation
Migacz, I. P., Manfron, J., Farago, P. V., Raman, V., De Muñiz, G. I. B., & Nisgoski, S. (2022). VIS/NIR spectra and color parameters according to leaf age of some Eucalyptus species: Influence on their classification and discrimination. Forest Systems, 31(2), e013. https://doi.org/10.5424/fs/2022312-19242
DOI
10.5424/fs/2022312-19242
Accessibility Status
Searchable text