composition of mobile matter in a floodplain topsoil: A comparative study with soil columns and field lysimeters

• Published in: Journal of plant nutrition and soil science 2016-02, Vol.179 (1), p.18-28
• Main Authors: Fritzsche, Andreas, Pagels, Britt, Totsche, Kai U
• Format: Article
• Language: English
• Subjects: alluvial soils; aluminum; anaerobic conditions; arsenic; Effluents; ferrolysis; floodplains; Floods; groundwater; imbibition; iron; lysimeters; manganese; nitrates; particulates; redox interface; riparian; river water; seepage; soil solution; sulfates; topsoil
• ISSN: 1436-8730; 1522-2624
• DOI: 10.1002/jpln.201500169

Floodplain soils are characterized by frequent and extreme redox changes caused by inundation with river water or imbibition of groundwater. Depending on the duration and extent of inundation, biogeochemical processes run at sub‐/anoxic conditions, which may result in the mobilization and relocation of dissolved and particulate matter within the soil. In this study, we investigated the effect of inundation events on the composition of mobilized matter in the topsoil horizon of a floodplain soil. We conducted experiments with soil columns in the laboratory and gravitational lysimeters in the field to identify redox‐mediated (im)mobilization processes and to estimate their relevance under field conditions. The lysimeters were filled with topsoil monoliths and run under in situ conditions during a ≈ 2.5‐y period. The soil columns were run with the same soil material either under strictly anoxic or mixed oxic–anoxic conditions. Effluents from mixed oxic–anoxic soil were composed fundamentally different [comparably high: Mn, Al, nitrate, sulfate; comparably low: pH, organic C (OC); not detected: Fe, As] compared to effluents from strictly anoxic soil (comparably high: pH, Fe, Mn, OC, As; comparably low: Al; not detected: nitrate, sulfate). Matter, which was mobile under anoxic conditions (e.g., Fe, As, OC), was effectively immobilized as soon as the mobile phase passed anoxic–oxic boundaries within soil (exception: Mn). We assume that the solution in the soil monoliths always passed such anoxic–oxic boundaries during downwards migration independent of lysimeter flooding with river water. This is indicated by the similar composition of the lysimeter seepage water and the effluents from mixed oxic–anoxic soil columns. Both solutions contained “fingerprints” from anoxic (Mn) and oxic conditions (nitrate). Inundations with river water and the duration of these floods (1–22 d) did not affect the composition of the lysimeter seepage water. In conclusion, immediate changes in the composition of the solution, which enters either the subsoil or nearby receiving waters, cannot be expected from regular topsoil flooding.