Vis-NIR / NIR Spectral data for soil organic carbon prediction - Châteaudun/Melle (France)

On each core, one spectrum was acquired at 3 and 7 cm for 0-10 cm layer, 13 and 17 cm for the 10-20 cm layer and 23 and 27 cm for the 20-30 cm layer. The six spectra acquired per layer were then averaged. The NIR reflectance spectra were measured in the lab, using a Foss NIRSystems 5000 spectrophotometer (Laurel, MDS, USA).

The spectral resolution ranged from 1100 to 2498 nm, with a 2 nm interval. Each spectrum was the result of 32 co-added scans and for each sample, two replications were performed. The two spectra were then averaged.

All spectra were then converted from reflectance (R) to absorbance (A) : A = - log(R). Every spectrum in these datasets is an absorbance spectrum. SOC determination SOC content was determined by dry combustion, using an Elemantal Analyser (CHN Fisons/Carlo Erba NA 2000, Milan, Italy).

SOC content is here expressed in g/kg. SOC stock was computed as the product between the bulk density (calculated as the ratio of the mass of dry fine earth (<2mm) to the cylinder volume) and the SOC content. SOC stock is here expressed in kg/L. For further details on SOC determination, see Cambou et al. (2016).

Projects & Fundings This dataset is part of the INCA project (In-field spectroscopy for Carbon Accounting) , that belongs to the GESSOL program of the French Ministry of Ecology, sustainable development and energy and was funded by ADEME (Agence de l'environnement et de la maîtrise de l'énergie). This dataset is also part of the ALAMOD project (2024-2027), financed by the PEPR FairCarbon.