Discrimination of land-use types in a catchment by energy dispersive X-ray fluorescence and principal component analysis

Differences in composition and chemical elemental concentration are important information for soil samples classification. The objective of this study is to present a direct methodology, that is non-destructive and without complex sample preparation, in order to discriminate different land-use types and soil degradation, employing energy dispersive X-ray fluorescence and multivariate analysis. Sample classification results from principal component analysis, utilizing spectral data and elemental concentration values demonstrate that the methodology is efficient to discriminate different land-use types.

► Soil degradation determination and land-use type classification was developed. ► EDXRF allied to PCA is an easy methodology for classification of samples. ► The novelty is the classification obtained directly from soil EDXRF spectral data.