Sensitivity of soil property predictions to instrumental spectral configurations from laboratory Vis-NIR spectroscopy

Laboratory Visible-Near Infrared (Vis-NIR) spectroscopy is a good alternative to costly physical and chemical soil analysis to estimate a wide range of soil properties. Various statistical methods relate soil Vis-NIR spectra to soil properties including partial least-squares regression (PLSR), the most common multivariate statistical technique in soil science. Most efforts are generally dedicated to the comparison of methodologies and their optimization for the estimation of soil properties. In this study, we focused on the impact of the spectral specifications of the spectroscopic sensor used to collect Vis-NIR spectra.
Specifically, this work examines the sensitivity of predictions of physico-chemical soil properties to different instrumental spectral configurations described by three parameters: the number of spectral bands, the spectral resolution and the spectral sampling interval. The initial database is composed of 1961 spectral bands, spectral resolutions of 3 and 10 nm in the 400-1000 nm and 1000-2500 nm ranges, respectively, and a spectral sampling interval of 1 nm. Seven instrumental spectral configurations were built from this database with a number of spectral bands decreasing from 328 to 10, a spectral resolution increasing from 3 nm to 200 nm, and a spectral sampling interval equaling the spectral resolution. All of these databases were composed of 148 soil samples collected at a Mediterranean site. Four soil properties were selected for their different spectroscopic behavior: clay, free iron oxides, calcium carbonate (CaCO3) and pH.
PLSR predicted these variables, and the results were as follows: (1) the prediction performance of the PLSR models were accurate and globally stable with a spectral resolution between 3 to 60 nm regardless of the soil properties, (2) the prediction performance decreased, but remained acceptable for clay, iron oxides and CaCO3 at spectral resolutions between 60 and 200 nm, (3) the sensitivity of a given soil property to instrumental spectral configurations depended on its spectral features and correlations with other soil properties.
This study highlights the ability of future lab Vis-NIR spectrometers designed with degraded spectral specifications to predict soil properties playing important roles in environmental and agronomic processes.