Dataset of understory reflectance spectra and fractional cover in a boreal forest area in Finland

Published: 2 May 2023| Version 2 | DOI: 10.17632/2g9nkcdj53.2


The dataset includes forest floor spectral reflectance factors at wavelengths of 350–2500 nm, fractional cover data of forest floor, forest inventory data, and data on the light availability at forest floor (diffuse non-interceptance, DIFN) and tree canopy structure (effective plant area index, PAIe) in a boreal forest site in Europe. The data were collected near peak growing seasons 2018–2019 in situ in total of 36 forest stands in Hyytiälä, Finland (61°50’N, 24°17’E). Each study stand represents a forest site (fertility) type: OMT (herb-rich), MT (mesic), VT (sub-xeric), or CT (xeric), based on Cajander’s theory of forest types [1]. The first spread sheet contains descriptions of the variables, second spread sheet the stand characteristics, and third the spectral data. We thank Lauri Korhonen, Ville Ranta, and Daniel Schraik for collaboration. Data collection and preparation was supported by the Academy of Finland [BOREALITY, grant number 286390; and DIMEBO, grant number 3323004]; and by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme [grant agreement No 771049]. The article reflects only the authors’ view and the Agency is not responsible for any use that may be made of the information it contains. If you use this dataset, please cite Forsström et al. [2, 3]. Note that this is Version 2 of the dataset that includes new variables on tree canopy structure (DIFN, PAIe) and small changes to pre-processing of the spectra. If you use Version 1, please cite Forsström et all [4, 5]. A similar dataset for hemiboreal and temperate forest sites is available from Forsström et al. [6]. Together, these datasets can be used to reproduce the results of Forsström et al. [2]. [1] Cajander, A.K., 1926. The theory of forest types. Acta Forestalia Fennica. doi:10.14214/aff.7193. [2] Forsström, P.R., Hovi, A., Juola, J., Rautiainen, M., 2023. Links between light availability and spectral properties of forest floor in European forests. Agricultural and Forest Meteorology. doi:10.1016/j.agrformet.2023.109481 [3] Forsström, P.R., Juola, J., Hovi, A., Rautiainen, M., 2023. Dataset of understory reflectance spectra and fractional cover in a boreal forest area in Finland. Mendeley Data, V2, doi:10.17632/2g9nkcdj53.2 [4] Forsström, P.R., Juola, J., Rautiainen, M. 2021. Relationships between understory spectra and fractional cover in northern European boreal forests. Agricultural and Forest Meteorology. doi:10.1016/j.agrformet.2021.108604 [5] Forsström, P.R., Juola, J., Hovi, A., Rautiainen, M., 2021. Dataset of understory reflectance spectra and fractional cover in a boreal forest area in Finland. Mendeley Data, V1, doi:10.17632/2g9nkcdj53.1 [6] Forsström, P.R, Hovi, A., Juola, J., Rautiainen, M., 2023. Dataset of tree canopy structure, understory reflectance spectra and fractional cover in hemiboreal and temperate forest areas in Estonia and the Czech Republic. Mendeley Data, V1. doi:10.17632/9dx32rszp8.1


Steps to reproduce

Forest floor spectra were measured in near nadir viewing direction under diffuse illumination conditions from 15 measurements positions per stand along an 11-m-long spectral transect. ASD FieldSpec4 spectrometer and bare fiber sensor optics were used. The reflectance quantity is spectral hemispherical-conical reflectance factor (HCRF), calculated per wavelength (WL) from the ratio of measurement signal of the forest floor (DN_FF) and white reference panel (DN_WR), with dark current (DN_DC) readily subtracted and the correction term of the panel (i.e., panel’s reflectance, R_ref) applied. The formula for HCRF is: HCRF(WL) = (DN_FF(WL) - DN_DC(WL))/(DN_WR(WL) - DN_DC(WL)) * R_ref(WL). The fractional covers of vegetation and other material on the forest floor were estimated for each stand from nadir view photographs (four photographs per stand, each imaging a 1 x 1 m square on the forest floor). The estimated cover classes were vascular plants, non-vascular plants, intact plant litter, decomposed plant litter, and lichen. The tree canopy structural variables DIFN and PAIe were derived from canopy gap fraction data from hemispherical photographs (five photographs per stand). The processing chain for the photos incorporated an automatic thresholding method to separate sky and canopy pixels [7], followed by calculation of DIFN and PAIe according to LAI-2200 instructions manual [8]. Detailed descriptions of methodologies used in data collection and data processing are provided by Forsström et al. [2] and for forest inventory by Hovi et al. [9]. [7] Nobis, M., Hunziker, U., 2005. Automatic thresholding for hemispherical canopy-photographs based on edge detection. Agricultural and Forest Meteorology. doi:10.1016/j.agrformet.2004.10.002. [8] Li-COR Inc., 2012. LAI-2200 plant canopy analyzer instruction manual. Publication number 984–10633 rev 2 (accessed 13 May 2022). [9] Hovi, A., Schraik, D., Hanuš, J., Homolová, L., Juola, J., Lang, M., Lukeš, P., Pisek, J., Rautiainen, M. 2022. Assessment of a photon recollision probability based forest reflectance model in European boreal and temperate forests. Remote Sensing of Environment. doi:10.1016/j.rse.2021.112804.




Remote Sensing, Surface Reflectance, Spectral Characterization