Arsenic speciation and turnover in intact organic soil mesocosms during experimental drought and rewetting

Wetlands are significant sources and sinks for arsenic (As), yet the geochemical conditions and processes causing a release of dissolved arsenic and its association with the solid phase of wetland soils are poorly known. Here we present experiments in which arsenic speciation was determined in peatland mesocosms in high spatiotemporal resolution over 10 months. The experiment included a drought/rewetting treatment, a permanently wet, and a defoliated treatment. Soil water content was determined by the TDR technique, and arsenic, iron and sulfate turnover from mass balancing stocks and fluxes in the peat, and solid phase contents by sequential extractions. Arsenic content ranged from 5 to 25 mg kg(-1) and dissolved concentrations from 10 to 300 pg L-1, mainly in form of As(III), and secondarily of As(V) and dimethylated arsenic (DMA). Total arsenic was mainly associated with amorphous iron hydroxides (R-2 > 0.95, alpha < 0.01) and deeper into the peat with an unidentified residual fraction. Arsenic release was linked to ferrous iron release and primarily occurred in the intensely rooted uppermost soil. Volumetric air contents of 2-13% during drought eliminated DMA from the porewater and suppressed its release after rewetting for >30 d. Dissolved As(III) was oxidized and immobilized as As(V) at rates of up to 0.015 mmol m(-3) d(-1). Rewetting mobilized As(III) at rates Of LIP to 0.018 mmol m(-3) d(-1) within days. Concurrently, Fe(I 1) was released at depth integrated rates of up 20 mmol m(-3) d(-1). The redox half systems of arsenic, iron, and sulfur were in persistent disequilibrium, with H,S being a thermodynamically viable reductant for As(V) to As(III). The study suggests that rewetting call lead to a rapid release of arsenic in iron-rich peatlands and that methylation is of lesser importance than co-release with iron reduction, which was largely driven by root activity. (C) 2008 Published by Elsevier Ltd.