Controlling factors for the stability of subsoil carbon in a Dystric Cambisol

- Higher specific CO2 production in subsoil raise question about subsoil SOC stability.
- Relatively more labile SOC and lower carbon use efficiency in deep subsoil
- Increasing SOC by 80 % had no positive priming effects in subsoil.
- Additional C inputs were more stable in soil horizons with low SOC concentration.
- SOC mineralization in subsoil is less temperature sensitive than in topsoils.

Subsoils store > 50 % of the total global soil organic carbon (SOC), and low SOC content and high mean residence times indicate that subsoils have the potential to sequester additional C on the long-term. Nevertheless, the mechanisms controlling the turnover of SOC in subsoils are poorly understood. The aim of this study was to assess the impact of temperature and substrate limitation on subsoil SOC turnover and evaluate the stability of additional C inputs in subsoils.In a 63-day microcosm incubation experiment, CO2 production of undisturbed soil samples from topsoil and two subsoil depth increments was measured at two different temperatures (10 °C and 20 °C). Additionally, 13C labeled root litter was added to the different samples and measurements of the isotopic signature of the respired CO2 allowed a differentiation between SOC mineralization and root mineralization. The CO2 production per unit soil mass was lower in deep subsoil than in the topsoil, but the CO2 production per unit SOC (specific mineralization) was three times higher in the deepest subsoil than in topsoil. This depth gradient of specific mineralization in undisturbed samples indicates that deep subsoil contained relatively more labile SOC than the topsoil. The temperature sensitivity of SOC mineralization expressed as Q10-q, decreased from around 3 to around 1 with increasing soil depth. In contrast, the mineralization of the added root material was solely determined by the recalcitrance of the added roots as indicated by a similar Q10-q through all three soil depths.Contrary to the SOC mineralization of undisturbed samples, significantly more added root litter was mineralized in the samples from the upper horizons than in the deepest subsoil samples, revealing a non-linear relationship between mineralization of added C and the SOC content. Thus, the distance between substrate units, as indicated by the SOC content, may be key factor for subsoil SOC dynamics. Moreover, root addition caused no positive priming effects in subsoil horizons indicating that enhanced C inputs to the subsoil can increase the SOC content and tap the unused C storage potential of subsoils.