Iron Compounds in Steppe Soils of the East-European Plain: Association with Soil Processes, Paleoclimatic Aspects

The total content, forms, and mineralogy of iron compounds in soils of forest-steppe, steppe, and semidesert zones of a large territory from the central chernozem regions to the Caspian lowland and from the Southern Urals to the Kerch Strait are studied. The study covers Chernozems (n = 40), Kastanozems (n = 15), Solonetzes (n = 7), and Calcisols (n = 7). Our results demonstrate the specific features in the distribution of total iron content, mass balance (τFe, Zr), forms of iron compounds, magnetic susceptibility (χ), and mineralogy in soil profiles. The distribution of τFe, Zr in the studied soils reflects the processes and conditions of soil formation, as well as specific lithological features. For a more comprehensive understanding of the transformation of iron compounds in steppe soils, particle-size fractions (<2, 2–5, 5–10, and 10–50 μm) have been analyzed by Mössbauer spectroscopy and magnetic susceptibility methods for Luvic Chernozem, Haplic Kastanozem, Haplic Kastanozem (Endosalic, Cambic), and Luvic Calcisol (Endosalic). In the clay fraction, a large share of Fe3+ is contained in highly dispersed oxides and hydroxides in a superparamagnetic state. In the humus-accumulative horizons of steppe soils, a decrease in the Fe2+ fraction in aluminosilicates caused by weathering is recorded. The observed correlation of the goethite/(hematite + goethite) ratio in humus-accumulative soil horizons with climatic parameters suggests the possibility to apply it for future paleoclimatic reconstructions. Mössbauer spectroscopy shows a significant increase in the content of nonsilicate iron in the humus-accumulative horizons of steppe soils as compared with the soil-forming material, which is an important confirmation for the formation of iron oxides during pedogenesis. A comparison of the methods for assaying nonsilicate iron in soils (Mössbauer spectroscopy and Mehra–Jackson extraction) demonstrates considerable differences in the results, suggesting that the chemical technique underestimates the amount of free iron forms.