A spectral analysis of common boreal ground lichen species

REMOTE SENSING OF ENVIRONMENT

卷:247

Article

Kuusinen, Nea¹  Juola, Jussi¹  Karki, Bijay¹  Stenroos, Soili²  Rautiainen, Miina^1,3 

[1] Aalto Univ, Sch Engn, Dept Built Environm, POB 14100, FI-00076 Aalto, Finland

[2] Univ Helsinki, Bot Unit, Finnish Museum Nat Hist, POB 7, FI-00014 Helsinki, Finland

[3] Aalto Univ, Sch Elect Engn, Dept Elect & Nanoengn, POB 15500, Aalto 00076, Finland

关键词: Cetraria;    Cladonia;    Hyperspectral;    Multiangular;    Reflectance spectra;    Stereocaulon;   

DOI  :  10.1016/j.rse.2020.111955

来源: Elsevier

【 摘 要 】

Lichens dominate a significant part of the Earth's land surface, and are valuable bioindicators of various environmental changes. In the northern hemisphere, the largest lichen biomass is in the woodlands and heathlands of the boreal zone and in tundra. Despite the global coverage of lichens, there has been only limited research on their spectral properties in the context of remote sensing of the environment. In this paper, we report spectral properties of 12 common boreal lichen species. Measurements of reflectance spectra were made in laboratory conditions with a standard spectrometer (350-2500 nm) and a novel mobile hyperspectral camera (400-1000 nm) which was used in a multiangular setting. Our results show that interspecific differences in reflectance spectra were the most pronounced in the ultraviolet and visible spectral range, and that dry samples always had higher reflectance than fresh (moist) samples in the shortwave infrared region. All study species had higher reflectance in the backward scattering direction compared to nadir or forward scattering directions. Our results also reveal, for the first time, that there is large intraspecific variation in reflectance of lichen species. This emphasizes the importance of measuring several replicates of each species when analyzing lichen spectra. In addition, we used the data in a spectral clustering analysis to study the spectral similarity between samples and species, and how these similarities could be linked to different physical traits or phylogenetic closeness of the species. Overall, our results suggest that spectra of some lichen species with large ground coverage can be used for species identification from high spatial resolution remote sensing imagery. On the other hand, for lichen species growing as small assemblages, mobile hyperspectral cameras may offer a solution for in-situ species identification. The spectral library collected in this study is available in the SPECCHIO Spectral Information System.

【 授权许可】

Free