Spatial variability of iron oxides in soils from Brazilian sandstone and basalt

• Published in: Catena (Giessen) 2020-02, Vol.185, p.104258, Article 104258
• Main Authors: Silva, Laércio Santos, Marques Júnior, José, Barrón, Vidal, Gomes, Romário Pimenta, Teixeira, Daniel De Bortoli, Siqueira, Diego Silva, Vasconcelos, Vinícius
• Format: Article
• Language: English
• Subjects: Diffuse reflectance spectroscopy; Goethite; Hematite; Mapping; Pedometrics
• ISSN: 0341-8162; 1872-6887
• DOI: 10.1016/j.catena.2019.104258

•Spatial variability is governed by the lithological and landscape contrast.•The iron oxides, hematite and goethite, are the main pedoenvironments indicators.•Diffuse reflectance spectroscopy reflects the different mineralogical composition.•Iron oxide mapping assists agronomic and environmental management. Iron oxides as goethite (Gt) and hematite (Hm) are key minerals to better understand the soil–landscape relationships. Soil samples were collected at three stages of landscape dissection from the geological formations of Vale do Rio do Peixe (sandstone) and Serra Geral (basalt) in the Western Paulista Plateau (WPP), Brazil. Both iron oxides were quantified by X-ray diffraction (XRD) and diffuse reflectance spectroscopy (DRS), and the results were subjected to geostatistical analysis in order to assess the usefulness of DRS for characterizing the spatial variability in Gt and Hm. The prevalence and spatial variability of Hm and Gt in the soils were governed by the sandstone/basalt lithological contrast and landscape dissection. Iron oxides in the clay fraction exhibited high spatial variability over a large area and can be robust indicators of geological diversity and landscape dissection in pedoenvironments with low or high contents of iron oxides. Goethite had the highest spatial variability. Based on the spatial pattern of the differences between DRS and XRD estimates, the saturated red color in soil made DRS less useful for quantifying Hm in environments with high iron oxide contents. The maps indicate the sensitivity of XRD and DRS techniques to represent Hm and Gt spatial variability patterns. Gt was more sensitive to landscape dissection while Hm sensitive to lithology. Thus, the DRS technique is efficient in characterizing the spatial variability of these soil oxides in large areas, even considering the complex relations between soil and landscape.