Abstract. In contrast to mineral topsoils, in subsoils the origin and processes leading to the formation and stabilization of organic matter (OM) are still not well known. This study addresses the fate of litter-derived carbon (C) in whole soil profiles with regard to the conceptual cascade model, which proposes that OM formation in subsoils is linked to sorption–microbial processing–remobilization cycles during the downward migration of dissolved organic carbon (DOC). Our main objectives were to quantify the contribution of recent litter to subsoil C stocks via DOC translocation and to evaluate the stability of litter-derived OM in different functional OM fractions. A plot-scale stable isotope-labeling experiment was conducted in a temperate beech forest by replacing the natural litter layer with 13C enriched litter on an area of 20 m2 above a Dystric Cambisol. After 22 months of field exposure, the labeled litter was replaced again by natural litter and soil cores were drilled down to 180 cm soil depth. Water extraction and density fractionation were combined with stable isotope measurements in order to link the fluxes of recent litter-derived C to its allocation into different functional OM fractions. A second sampling was conducted 18 months later to further account for the stability of translocated young litter-derived C. Almost no litter-derived particulate OM (POM) entered the subsoil, suggesting root biomass as the major source of subsoil POM. The contribution of aboveground litter to the formation of mineral-associated OM (MAOM) in topsoils (0–10 cm) was 1.88±0.83 g C m−2 and decreased to 0.69±0.19 g C m−2 in the upper subsoil (10–50 cm) and 0.01±0.02 g C m−2 in the deep subsoil >100 cm soil depth during the 22 months. This finding suggests a subordinate importance of recent litter layer inputs via DOC translocation to subsoil C stocks, and implies that most of the OM in the subsoil is of older age. Smaller losses of litter-derived C within MAOM of about 66 % compared to POM (77 %–89 %) over 18 months indicate that recent carbon can be stabilized by interaction with mineral surfaces; although the overall stabilization in the sandy study soils is limited. Our isotope-labeling approach supports the concept of OM undergoing a sequence of cycles of sorption, microbial processing, and desorption while migrating down a soil profile, which needs to be considered in models of soil OM formation and subsoil C cycling.

Abstract. In contrast to mineral topsoils, the origin and processes leading to the formation and stabilization of organic matter (OM) in subsoils is still not well known. This study addresses the fate of litter-derived carbon (C) in whole soil profiles with regard to the conceptual cascade model, which proposes that OM formation in subsoils is linked to sorption-microbial processing-remobilization cycles during the downward migration of dissolved organic carbon (DOC). Our main objectives were to quantify the contribution of recent litter to subsoil C stocks via DOC movement and to evaluate the stability of litter-derived OM in different functional OM fractions. A plot-scale stable isotope labeling experiment was conducted in a temperate beech forest by replacing the natural litter layer with 13C enriched litter on an area of 20 m2 above a Dystric Cambisol. After 22 months of field exposure, the labeled litter was replaced again by natural litter and soil cores were drilled down to 180 cm soil depth. Water extraction and density fractionation were combined with stable isotope measurements in order to link the fluxes of recent litter-derived C to its allocation into different functional OM fractions. A second sampling was conducted 18 months later to further account for the stability of translocated young litter-derived C. Almost no litter-derived particulate OM (POM) entered the subsoil, suggesting root biomass as the major source of subsoil POM. The contribution of aboveground litter to the formation of mineral-associated OM (MAOM) in topsoils (0–10 cm) was 0.99 ± 0.45 g C m−2 yr−1, and decreased to 0.37 ± 0.10 g C m−2 yr−1 in the upper subsoil (10–50 cm) and 0.01 ± 0.01 g C m−2 yr−1 in the deep subsoil > 100 cm soil depth. This finding suggests a subordinate importance of recent litter layer inputs via DOC translocation to subsoil C stocks, and implies that most of the OM in the subsoil is of older age. Smaller losses of litter-derived C within MAOM of about 66 % compared to POM (77–89 %) indicate that recent carbon can be stabilized by interaction with mineral surfaces; although the overall stabilization in the sandy study soils was low. Our isotope labeling approach supports the concept of OM undergoing a sequence of cycles of sorption, microbial processing, and desorption while migrating down a soil profile, which needs to be considered in models on soil OM formation and subsoil C cycling.

Background Soils are important carbon (C) sinks or sources and thus of utmost importance for global carbon cycling. Particularly, subsoils are considered to have a high potential for additional C storage due to mineral surfaces still available for sorptive stabilization. Aims Little information exists about the extent to which additional litter‐derived C is transferred to and stabilized in subsoils. This study aimed at evaluating the role of litter‐derived dissolved organic matter (DOM) inputs for the formation of stable mineral‐associated C in subsoils. Methods We carried out a multiple‐method approach including field labeling with 13C‐enriched litter, exposure of 13C‐loaded reactive minerals to top‐ and subsoils, and laboratory sorption experiments. Results For temperate forest soils, we found that the laboratory‐based C sink capacity of subsoils is unlikely to be reached under field conditions. Surface C inputs via litter leachates are little conducive to the subsoil C pool. Only 0.5% of litter‐derived C entered the subsoil as DOM within nearly 2 years and most of the recently sorbed C is prone to fast microbial mineralization rather than long‐term mineral retention. Desorption to the soil solution and an adapted microbial community re‐mobilize organic matter in subsoils faster than considered so far. Conclusions We conclude that the factors controlling the current mineral retention and stabilization of C within temperate forest subsoils will likewise limit additional C uptake. Thus, in contrast to their widely debated potential to accrue more C, the role of forest subsoils as future C sink is likely overestimated and needs further reconsideration.

Litter-derived dissolved organic carbon (DOC) is considered to be a major source of stabilised C in soil. Here we investigated the microbial utilisation of litter-derived DOC within an entire soil profile using a stable isotope labelling experiment in a temperate beech forest. The natural litter layer of a Dystric Cambisol was replaced by 13C enriched litter within three areas of each 6.57 m−2 for 22 months and then replaced again by natural litter (switching-off the 13C input). Samples were taken continuously from 0 to 180 cm depths directly after the replacement of the labelled litter, and 6 and 18 months thereafter. We followed the pulse of 13C derived from aboveground litter into soil microorganisms through depth and over time by analysing 13C incorporation into microbial biomass and phospholipid fatty acids. Throughout the sampling period, most of the litter-derived microbial C was found in the top cm of the profile and only minor quantities were translocated to deeper soil. The microbial 13C stocks below 30 cm soil depth at the different samplings accounted constantly for only 6–12% of the respective microbial 13C stocks of the entire profile. The peak in proportional enrichment of 13C in subsoil microorganisms moved from upper (≤ 80 cm soil depth) to lower subsoil (80–160 cm soil depth) within a period of 6 months after switch-off, and nearly disappeared in microbial biomass after 18 months (< 1%), indicating little long-term utilisation of litter-derived C by subsoil microorganisms. Among the different microbial groups, a higher maximum proportion of litter-derived C was found in fungi (up to 6%) than in bacteria (2%), indicating greater fungal than bacterial dependency on litter-derived C in subsoil. However, in contrast to topsoil, fungi in subsoil had only a temporarily restricted increase in litter C incorporation, while in the Gram-positive bacteria, the C incorporation in subsoil raised moderately over time increasingly contributing to the group-specific C stock of the entire profile (up to 9%). Overall, this study demonstrated that microorganisms in topsoil of a Dystric Cambisol process most of the recently deposited aboveground litter C, while microbial litter-derived C assimilation in subsoil is low.

Dissolved organic carbon (DOC) from Oa horizons has been proposed to be an important contributor for subsoil organic carbon stocks. We investigated the fate of DOC by directly injecting a DOC solution from 13 C labelled litter into three soil depths at beech forest sites. Fate of injected DOC was quantified with deep drilling soil cores down to 2 m depth, 3 and 17 months after the injection. 27 ± 26% of the injected DOC was retained after 3 months and 17 ± 22% after 17 months. Retained DOC was to 70% found in the first 10 cm below the injection depth and on average higher in the topsoil than in the subsoil. After 17 months DOC in the topsoil was largely lost (-19%) while DOC in the subsoil did not change much (-4.4%). Data indicated a high stabilisation of injected DOC in the subsoils with no differences between the sites. Potential mineralisation as revealed by incubation experiments however, was not different between DOC injected in topsoil or subsoils underlining the importance of environmental factors in the subsoil for DOC stabilisation compared to topsoil. We conclude that stability of DOC in subsoil is primary driven by its spatial inaccessibility for microorganisms after matrix flow while site specific properties did not significantly affect stabilisation. Instead, a more fine-textured site promotes the vertical transport of DOC due to a higher abundance of preferential flow paths.