The bioavailability of arsenic in floodplain soils: a simulation of water saturation

The soils of the floodplains of the German rivers Elbe, Mulde and Saale are often contaminated with arsenic (As). The total As concentration in soil is, however, not related to As concentration in the plant biomass and thus is unsuitable for evaluating the risk of As transfer into the food chain. The aim of the present study was to establish a method that assesses the bioavailability of As in floodplain soils and takes possible changes in the mobility of As with changing redox potential into account. Samples from the floodplain of the river Mulde were saturated with water for 34 days. A soil solution was sampled every second day and the concentrations of As species and total iron (Fe) were determined. The redox potential was measured in situ. Lepidium sativum L. (garden cress) was cultured in cylinders to assess the availability of As to the plant. In addition, extraction methods were used to define operationally different As-binding forms. Arsenic in the soil samples was predominantly bound to poorly crystalline (hydr)oxides. These (hydr)oxides are likely to be affected by changing redox conditions. On saturation, 18 soil samples could be divided into two groups based on the temporary changes in redox potential and subsequent changes in the As concentration of the soil solution. Water-extractable organic carbon was identified as the reason for the differences between the two groups. As the field site is periodically flooded, it is likely that the samples that exhibit decreasing redox potential in the experimental system will also pass through redox cycles more frequently in the field. The frequency of redox cycles, in turn, strongly affects the Fe (hydr)oxide phase and the binding of As. In the present study these differences between the two groups were reflected in the different relationships between NH4 oxalate-extractable Fe (Fe-o) and As (As-o).