%0 Journal Article
%9 ACL : Articles dans des revues avec comité de lecture répertoriées par l'AERES
%A Barthès, Bernard
%A Chotte, Jean-Luc
%T Infrared spectroscopy approaches support soil organic carbon estimations to evaluate land degradation
%D 2020
%L fdi:010079541
%G ENG
%J Land Degradation 2020 Development
%@ 1085-3278
%K carbon concentration ; carbon stock ; mid-infrared reflectance spectroscopy ; multivariate regression ; near-infrared reflectance spectroscopy
%M ISI:000561867300001
%N 1
%P 310-322
%R 10.1002/ldr.3718
%U https://www.documentation.ird.fr/hor/fdi:010079541
%> https://www.documentation.ird.fr/intranet/publi/2020/09/010079541.pdf
%V 32
%W Horizon (IRD)
%X Soil organic carbon (SOC) is an acknowledged indicator for land degradation, but conventional determination of SOC remains tedious, especially regarding SOC stock (in kg C m(-2)for a given depth layer), which is the product of SOC concentration (g C kg(-1)) by volumetric mass (kg dm(-3)). Diffuse reflectance infrared spectroscopy (DRIS) is a time- and cost-effective approach, which uses calibrations for making predictions. The aim of this paper is to propose an overview of DRIS uses for estimating SOC, thus land degradation. Indeed, many papers have demonstrated the precision of DRIS for quantifying SOC concentration, at different scales. Current development of large soil calibration databases and improvements in spectral data analysis pave the way for ever-wider use of DRIS, which should help solving the soil data crisis, regarding SOC especially. The increasing availability of portable spectrometers allows SOC quantification in the field, which seems particularly promising; but large calibration databases made of soil spectra acquired in the field are difficult to build, while large collections of analyzed soil samples (air-dried, 2-mm sieved) already exist. Some recent studies indicate that DRIS can also be used for predicting SOC stock, even from sieved samples, which represents an efficient option because determining the volumetric mass is particularly tedious and an obstacle for exactly specifying the role of soils in the global carbon cycle. In short, DRIS has strong potential for supporting better evaluation of soil and land degradation, and the availability of spectrometers at increasingly affordable prices reinforces this potential.
%$ 068