Mineralogy of eroded sediments derived from highly weathered Ultisols of central Alabama

Abstract

Coarse-textured surface horizons are common in highly weathered southeastern US coastal plain soils. These soils have historically been managed under conventional tillage practices, but conservation tillage management practices are increasing. Although clay (<2 μm) contents are low in these surface horizons (typically <100 g kg −1), the reactive nature of this fraction tends to play a dominant role in colloidal facilitated pollutant transport. Studies have suggested that enrichment of certain minerals occurs in transported sediments, thus, we evaluated sediment size and the mineralogical partitioning of clay minerals of soil versus runoff sediment under simulated rainfall. In addition, because water dispersible clay (WDC) has been shown to be correlated with soil erodibility, we evaluated WDC differences as a function of tillage practices. Plots were established at a site in the upper coastal plain of central Alabama, where soils classified as coarse-loamy, siliceous, subactive, thermic Plinthic Paleudults and Typic Hapludults (FAO-Acrisols). Surface tillage treatments were established in 1988, and included conventional tillage (CT) versus no surface tillage (NT) treatments, with crop residue remaining or removed, and with or without paratilling (non-inversion subsoiling). Within these plots, simulated rainfall (target intensity=50 mm h −1 for 2 h) was applied to replicated 1 m×1 m areas, and runoff and sediment were collected. Mineralogical analyses of soils and sediment were conducted using thermogravimetric analysis (TGA) and X-ray diffraction (XRD) techniques. WDC quantities were highly correlated with percent soil organic carbon (SOC) ( r 2=0.76), which was a function of tillage treatment. Although no differences in the mineralogy of the <2 μm sediment were observed between tillage treatments, runoff sediments (<2 μm) were enriched in quartz (qtz) and relatively depleted with kaolinite compared to in situ soils. These findings will facilitate development of mechanistic models that predict sediment attached losses of nutrients and pesticides.