1. Fontaine S., Barot S., Barré P., Bdioui N., Mary B., Rumpel C. Nature (2007), Vol. 450, 277-281

2.  Soil reservoir of organic C: more than biomass and atmospheric CO 2 combined  Mean residence time of SOC increases with depth  Which factors control the stability of carbon in deep soil layers? - Hypotheses: 1. Chemical structure 2. Fixation on minerals 3. Conditions for microbes

3. New theory of SOC dynamics: Slow SOC turnover at depth results from scarcity of fresh C  Soil humus → result of long-term accumulation of biochemically recalcitrant compounds with low energy content  Near soil surface: fresh C available → energy → decomposition  Deep soil layers: fresh C input extremely low → decomposition strongly reduced

4. Characterization of the soil profile:  3 independent soil samples in layers of 0.2 m down to depth of unweathered parent material  Organic C and bulk density measured to determine SOC content and storage  Studies of stability of SOC: focused on 2 layers: 0-0.2 m, 0.6-0.8 m Study site: grassland in France (for >50 yr, 2000 yr ago: chestnut/hornbearn forest)

5. Methods:  14 C content was measured to date SOC  MRT of SOC was determined with a flux model  Chemical composition was analyzed by 13 C CPMS, NMR and FTIR spectroscopy Results:  14 C content of SOC declined with depth  Surface layer: young fast-cycling C, subsoil: ancient slow-cycling C  Differences in MRT were not mirrored by changes in the chemical composition

6. Methods:  Amount of C bound to soil minerals was estimated by the demineralization technique  Clay mineralogy was determined by X-ray diffraction of oriented samples  Iron and aluminium oxides and oxyhydroxides were estimated by the dithionite-citrate-bicarbonate method Results:  Proportion of SOC bound to minerals increased slightly with depth (surface: 50 ± 0.5%, subsoil: 58 ± 1%)  Organo-mineral complexes at depth must be ten times more stable than in surface → not supported by results

7. Methods:  Soils were incubated at 20 ° C, at a water potential of -100kPa for 161 days  Dual-labelled cellulose was mixed with half of the incubated soils (1g cellulose C/kg soil). Control soils were also mixed  CO 2 evolved was trapped in NaOH and measured by continuous flow colorimetry  13 C and 14 C analysis of carbonates was carried out by IRMS and AMS Results:  Addition of cellulose stimulated microbial respiration and growth  Stimulation of decomposers induced significant increase in production of unlabelled soil-originated CO 2 (Priming)  Total microbial biomass and priming effect significantly decreased with exhaustion of cellulose

8. Conclusion: Although microbes decompose ancient C, acquisition of this energy is not sufficient to sustain long-term biological activity → energy required to break down recalcitrant SOC is higher than energy supplied by catabolism of such substrate → long-term activity of decomposer populations depends on permanent supply of fresh C

9.  Biological and physical processes that bury recalcitrant SOC below the deposits of fresh C protect it from decomposition and allow C storage over millennia  Even under favorable conditions for microbial activities, SOC from deep soil does not provide enough energy to sustain active microbial populations/production of enzymes  Deep SOC decomposition may be reactivated (changes in land use/agriculture practices)

10. Thank you for your attention!