Storage and stability of organic carbon in soils as related to depth, occlusion within aggregates, and attachment to minerals

Conceptual models suggest that stability of organic carbon (OC) in soil depends on the source of plant litter, occlusion within aggregates, incorporation in organomineral complexes, and location within the soil profile. Density fractionation is a useful tool to study the relevance of OC stabilization in aggregates and in association with minerals, but it has rarely been applied to full soil profiles. We aim to determine factors shaping the depth profiles of physically unprotected and mineral associated OC and test their relevance for OC stability across a range of European soils that vary in vegetation, soil types, parent material, and land use. At each of the 12 study sites, 10 soil cores were sampled to 60 cm depth and subjected to density separation. Bulk soil samples and density fractions (free light fractions - fLF, occluded light fractions - oLF, heavy fractions - HF) were analysed for OC, total nitrogen (TN), delta C-13, and Delta C-14. Bulk samples were also incubated to determine CO2 evolution per g OC in the samples (specific mineralization rates) as an indicator for OC stability. Depth profiles of OC in the light fraction (LF-OC) matched those of roots for undisturbed grassland and forest sites, suggesting that roots are shaping the depth distribution of LF-OC. Organic C in the HF declined less with soil depth than LF-OC and roots, especially at grassland sites. The decrease in Delta C-14 (increase in age) of HF-OC with soil depth was related to soil pH as well as to dissolved OC fluxes. This indicates that dissolved OC translocation contributes to the formation of subsoil HF-OC and shapes the Delta C-14 profiles. The LF at three sites were rather depleted in C-14, indicating the presence of fossil material such as coal and lignite, probably inherited from the parent material. At the other sites, modern Delta C-14 signatures and positive correlations between specific mineralization rates and fLF-OC indicate the fLF is a potentially available energy and nutrient source for subsurface microorganisms throughout the profile. Declining specific mineralization rates with soil depth confirm greater stability of OC in subsoils across sites. The overall importance of OC stabilization by binding to minerals was demonstrated by declining specific mineralization rates with increasing contributions of HF-OC to bulk soil OC, and the low Delta C-14 values of HF-OC. The stability of HF-OC was greater in subsoils than in topsoils; nevertheless, a portion of HF-OC was active throughout the profile. While quantitatively less important than OC in the HF, consistent older ages of oLF-OC than fLF-OC suggest that occlusion of LF-OC in aggregates also contributes to OC stability in subsoils. Overall, our results indicate that association of OC in soils, irrespective of vegetation, soil type, and land use.