Soil carbon sequestration and turnover in a pine forest after six years of atmospheric CO₂ enrichment

During the first six years of atmospheric CO2 enrichment at the Duke Forest free‐air CO2 enrichment (FACE) experiment, an additional sink of 52 ± 16 g C·m−2·yr−1 accumulated in the forest floor (O‐horizon) of the elevated CO2 treatment relative to the ambient CO2 control in an aggrading loblolly pine (Pinus taeda L.) forest near Chapel Hill, North Carolina, USA. The experiment maintained an atmospheric CO2 concentration 200 μL/L above ambient levels in replicated (n = 3) FACE rings throughout the six‐year period. This CO2‐induced C sink was associated with greater inputs of organic matter in litterfall and fine‐root turnover. There was no evidence that microbial decomposition was altered by the elevated CO2 treatment. Consistent with ecosystem recovery following decades of intensive agriculture, the C and N content of the mineral soil increased under both the elevated CO2 treatment and the ambient CO2 control during the six‐year period. This increase is attributed to accumulation of plant residues derived from fine roots with relatively high turnover rates rather than accumulation of refractory or physically protected soil organic matter (SOM). The elevated CO2 treatment produced no detectable effect on the C and N content of the bulk mineral soils or of any particulate organic matter size fraction. Because the fumigation gas was strongly depleted in 13C, the incorporation of new C could be traced within the ecosystem. Significant decreases in δ13C of soil organic carbon (SOC) under the elevated CO2 treatment were used to estimate the mean residence times of intra‐aggregate particulate organic matter and mineral‐associated organic matter as well as the annual C inputs required to produce the observed changes in δ13C. Our results indicate that forest soils such as these will not significantly mitigate anthropogenic C inputs to the atmosphere. The organic matter pools receiving large annual C inputs have short mean residence times, while those with slow turnover rates receive small annual inputs.