Addition of iron to agricultural topsoil and subsoil is not an effective C sequestration strategy

The interaction of soil organic matter (SOM) with Fe-containing minerals represents a key mechanism that promotes carbon (C) stabilisation in soil. The addition of Fe-rich industrial by-products to soil may therefore help accelerate C storage. Our understanding of the effects of exogenous Fe addition (Fe (oxy)hydroxide vs. Fe chloride) on SOM dynamics and C dynamics in agricultural soils, especially in subsoils, however, remains poor. Here, we simulate the addition of Fe in an arable soil context and assess its effectiveness based on CO2 emissions and soil chemistry. We hypothesised that insoluble and soluble Fe would reduce the mineralization of newly added unprotected organic materials more than native SOM and that soluble Fe would cause mineralisation of native SOM. To investigate this, insoluble Fe(OH)3 or soluble FeCl2 (0–5 g kg−1) were added to arable top- (0–10 cm) or subsoils (50–60 cm) and CO2 emissions, pH and nutrient dynamics (e.g. P, N) measured in a laboratory incubation over a 45 d period. We also compared the effect of Fe on the turnover of native SOM and newly added C (i.e. 14C-labelled glucose, citrate and crop residues) which was pre-mixed with exogenous Fe. We found that: (1) despite a reduction in P and DOC, Fe(OH)3 did not suppress total CO2 efflux; (2) high FeCl2 rates induced a rapid and significant release of CO2, which we attribute almost entirely to FeCl2-induced soil acidification increasing DOC availability and carbonate dissolution; (3) 14C-substrate mineralisation was weakly suppressed by Fe(OH)3 but strongly by FeCl2 following the series: citrate < glucose < crop residues; and (4) Fe addition to subsoils induced a stronger C mineralisation response but weaker effect on soil solution chemistry compared to topsoil, possibly due to subsoils having a lower buffering ability and less microbial biomass. We conclude that addition of extra Fe was not effective in promoting greater C sequestration in the arable soil we tested.