Use of VNIR spectroscopy for assessment of Stagnosols properties based on linear and non- linear calibration methods

Soil organic and mineral compounds can be estimated non-destructively by visible and near infrared (VNIR) diffuse reflectance spectroscopy. However, results of calibration models differ in dependence of measurement precision, spectral range, variability of soil properties and calibration methods used for prediction. The objective of research was to estimate the ability of hyperspectral VNIR spectroscopy for field-scale prediction of soil total carbon (TC %) and total nitrogen (TN %) content, soil pH, plant-available potassium (K) and phosphorus (P), in arable Stagnosols. Total of 200 soil samples taken from field experiment (soil depth: 30 cm ; sampling grid: 15x15 m ; 2016) were scanned in laboratory using portable spectroradiometer (FieldSpec®3, 350-1050 nm). Partial least squares regression (PLSR) with full cross-validation and artificial neural networks (ANN) were used to build prediction models of selected soil properties based on VNIR spectra. Strong to full correlation and low root mean squared error were obtained between predicted and measured values for the calibration and validation dataset, and both calibration methods. ANN models were more efficient in capturing the complex link between selected soil properties and soil reflectance spectra. Key spectral features and algorithms defined in this study should help to support site-specific and real-time soil survey using hyperspectral remote sensing.