Calibration models database of near infrared spectroscopy to predict agricultural soil fertility properties

Presented paper describes spectroscopic dataset and calibration models database of near infrared spectroscopy (NIRS) used to predict agricultural soil fertility properties. Near infrared spectra data in form of absorbance spectrum were acquired in wavelength range from 1000 to 2500 nm for a total of 40 bulk soil samples amounted of 10 g per each bulk. Soil fertility properties, presented as soil nitrogen (N), phosphorus (P). potassium (K), soil pH, magnesium (Mg) and calcium (Ca), were measured by means of wet chemical analysis. Calibration models, used to predict those soil fertility parameters were developed using two different regression algorithms namely principal component regression (PCR) and partial least square regression (PLSR) respectively. Prediction performance can be evaluated and justified by looking their statistical indicators: correlation of determination (R²), correlation coefficient (r), root mean square error (RMSE) and residual predictive deviation (RPD). Spectra data can also be corrected in order to improve and enhance prediction performance. Obtained NIRS dataset and models database can be used as a rapid and simultaneous method to determine agricultural soil fertility properties.