Prediction of Soil Organic Carbon Content by Spectroscopy at European Scale Using a Local Partial Least Square Regression Approach

Due to the large spatial variation of soil organic carbon (SOC) content, assessing the current state of SOC for large areas is costly and time consuming. Visible and Near Infrared Diffuse Reflectance Spectroscopy (Vis-NIR DRS) is a fast and cheap tool for measuring SOC based on empirical equations and spectral libraries. While the approach has been demonstrated to yield accurate predictions for databases containing samples belonging to soils with similar characteristics such as mineralogy, texture, iron and CaCO3 content, spectroscopic calibrations have been less successful when applied to large and diverse soil spectral libraries. About 20,000 samples collected all over the European Union were analyzed for physical and chemical properties, and scanned with a Vis-NIR spectrometer in a single laboratory. We implemented a modified local partial least square regression approach that, in addition to the spectra, uses other soil covariates (geographical coordinates, texture) for predicting the SOC content. The results showed good prediction ability for mineral soils under cropland (RMSE = 3.6 g C kg) and grassland (RMSE = 7.2 g C kg). Predictions of mineral soils under woodland (RMSE = 11.9 g C kg) and organic soils (RMSE= 51.1 g C kg ) were less accurate. The best results were obtained when sand content was used as covariate.