The effective monitoring of eleven crucial parameters that describe the condition of cultivated Haplic Luvisol soils using fourier-transformed near-infrared spectroscopy

The study aimed to develop a cost-effective and rapid approach for monitoring eleven crucial parameters that describe the health of cultivated Haplic Luvisol soils. To achieve this, the diffuse reflectance spectra in the NIR spectral region, between 1100 and 2500 nm, were registered, and partial least-squares (PLS) models to predict the concentrations of the critical parameters, including organic carbon, total nitrogen, the available soil nutrients (phosphorus, potassium and magnesium), the exchangeable cations (calcium, magnesium and potassium), the pH levels (determined in deionized water or KCl solution and hydrolytic acidity), were built. The validation results, which are expressed as RMSE and R2val that were obtained for validation the samples, demonstrate that the proposed approach is a reliable and efficient way to monitor the health of Haplic Luvisol soils. Specifically, excellent predictions were obtained for the total nitrogen, organic carbon and pH, which was determined in KCl solution with R2val values equal to 0.9696, 0.9573 and 0.9176, respectively. Additionally, very good PLS models were constructed to estimate the pH levels determined in deionized water, exchangeable magnesium, and po-tassium, which achieved R2val values of 0.8650, 0.8383, and 0.8329, respectively. For the remaining parameters, the corresponding R2val values were above 0.7294.