USING VNIR-DRS TO ASSESS SOIL DEGRADATION DUE TO EROSION.

Large numbers of soil samples must be collected and analyzed when mapping soil degradation due to erosion. The reflective spectroscopy in the visible and near infrared part of electromagnetic spectrum (VNIR DRS spectroscopy) is a low-cost indirect method, which may be used to gain information about soil properties more effectively. The method for assessing soil degradation using soil spectra proposed and tested in this study was based on estimating oxidable organic carbon content, carbonate content, and contents of different forms of iron and manganese (Mehlich III extract, ammonium oxalate extract and dithionite-citrate extract). Study was performed in Brumovice. Tested soil types were Haplic Chernozem, Colluvic Chernozem, Colluvial soils and Regosol. Samples were taken from the topsoil within the regular grid covering the studied area. The soil spectra curves (of air dry soil and sieved using 0.2 mm sieve) were measured in the laboratory using the spectrometer FieldSpec®3 (350 - 2 500 nm). Partial least squares regression (PLSR) was used for modeling of the relationships between spectra and measured soil properties. The results showed the best prediction for carbonate content (R² = 0.87). The worse predictions were obtained for oxidable organic carbon content (R² = 0.78), and contents of Mn in ammonium oxalate extract (R² = 0.78), Mn in dithionite--citrate extract (R² = 0.72), Fe in ammonium oxalate extract (R² = 0.70), Fe in dithionite-itrate extract (R² = 0.57), Mn in Mehlich III extract (R² = 0.31) and content of Fe in Mehlich III extract (R² = 0.20). In general, the results confirmed that the measurement of soil spectral characteristics is a promising technology for digital soil mapping and predicting soil degradation due to erosion processes. [ABSTRACT FROM AUTHOR]