Belowground processes are increasingly recognized as an important foundation for ecosystem responses to rising atmospheric [CO2]. Elevated [CO2] has been shown to increase the proportion of biomass in fine roots, and experimental evidence from a diverse set of forested ecosystems indicates that CO2-enrichment may lead to deeper rooting distributions. Increased production of fine roots at depth in the soil could drive changes in soil carbon (C) cycling because fine roots turn over quickly in forests. However, the consequences of increased fine-root proliferation and turnover at depth are still poorly understood; this is in part because belowground research is often truncated at relatively shallow soil depths. We examined soil C dynamics after 12 years of CO2-enrichment and at soil depths to 90 cm in soil pits harvested from the Oak Ridge National Laboratory (ORNL) Free-Air CO2 Enrichment (FACE) located in a sweetgum plantation in eastern Tennessee, USA. We hypothesized that: (1) soil C content would increase in response to elevated [CO2], especially at deeper soil depths where large increases in root production and mortality were observed, and (2) greater C inputs under elevated [CO2] would lead to increased potential C mineralization in long-term laboratory incubations.
As we hypothesized, total soil C content under elevated [CO2] was 20% greater throughout the soil profile to 90 cm depth. The CO2 effect was driven by an increase in the C content of the relatively labile particulate organic matter (POM) pool, which is likely derived primarily from fine roots. The POM was depleted in 13C throughout the soil profile, indicating substantial new C inputs over the course of the experiment. Contrary to what we hypothesized, we did not observe a significant increase in potential soil C mineralization under elevated [CO2]. While C mineralization rates were well-predicted by soil C content at shallower soil depths (i.e., 0 to 30 cm), this relationship did not hold at depths deeper than 30 cm. Therefore, root-derived C inputs to deeper soil may not decompose as quickly relative to inputs in shallower soil horizons, leading to increased ecosystem C storage under elevated [CO2]. However, C dynamics at depth in the soil are not currently incorporated in land surface models projecting forest response to elevated [CO2]. Progress in understanding and modeling the interface between deeper rooting distributions and soil C cycling will be critical in projecting the sustainability of forest responses to rising atmospheric [CO2].