Drivers and Patterns of Iron Redox Cycling from Surface to Bedrock in a Deep Tropical Forest Soil: A New Conceptual Model

Iron (Fe) reduction and oxidation are important biogeochemical processes coupled to decomposition, nutrient cycling, and mineral weathering, but factors controlling their rates and spatial distribution with depth are poorly understood in terrestrial soils. In aquatic ecosystems, Fe reduction often occurs below a zone of oxic sediments. We tested an alternative conceptual model for Fe redox cycling in terrestrial soils using a deep humid tropical forest soil profile. We hypothesized that Fe reduction in anaerobic microsites scales with depth variation in labile C and Fe availability, as opposed to bulk oxygen (O2). We measured bulk O2 at multiple depths from 0.1 to 5 m quasi-continuously over 18 months and sampled soils from surface to bedrock (~7 m). Median O2 mixing ratios declined from 19.8 ± 1.2 % at 0.25 m to 16.1 ± 1.0 % at 1 m, but did not consistently decrease below 1 m, challenging a recent model of regolith development. Reduced Fe (Fe(II)) extractable in 0.5 M hydrochloric acid was greatest in 0–0.1 m soil and declined precipitously with depth, and did not correspond with visible gleying in B horizons. We observed similar depth trends in potential Fe reduction under anaerobic conditions. Depth trends in Fe(II) also closely mirrored short-term soil respiration and bulk soil C. Labile C stimulated Fe reduction at 0–0.1 m depth, whereas addition of short-range-ordered Fe oxides had no effect. Cultivable Fe-reducing bacterial abundance was four orders of magnitude greater in surface soil (0–0.1 m) than below 1 m. Although cultivable Fe oxidizing bacteria were typically also more abundant in surface soil, addition of labile C and nitrate stimulated Fe oxidizers in deep soil by two orders of magnitude under anaerobic conditions. This implies that infiltration of nitrate (and possibly C) from shallow soil water could potentially promote biotic Fe oxidation, a critical step in bedrock weathering, 7 m below. Together, these data suggest that C, Fe, and nutrient availability increase microbial Fe reduction and oxidation in surface (vs deeper) soil microsites despite high bulk O2, in contrast to the depth segregation of electron accepting processes often observed in aquatic ecosystems. Furthermore, the greatest capacity for Fe redox cycling can occur in A horizons that do not display gleying or mottling