Overview
This study evaluates the performance of a low-cost, micro-electromechanical system-base for estimating soil organic carbon (OC) and total carbon (TC) in soil profiles from New South Wales, Australia. The research compares two commercial spectrometers, AgriSpec™ (ASD) and NeoSpectra™, which cover different spectral wavelength ranges. Three calibration models—Cubist tree model, partial least squares regression (PLSR), and support vector machine (SVM)—were assessed for predicting soil OC and TC using spectral data. The findings reveal that while the ASD spectrometer performed better, the NeoSpectrprovided comparable and cost-effective predictions. This demonstrates the potential of low-cost NIR spectrometer devices like NeoSpectra™ for efficient and affordable soil property assessment, particularly in agricultural applications.
