How relevant is recalcitrance for the stabilization of organic matter in soils?

Abstract Traditionally, the selective preservation of certain recalcitrant organic compounds and the formation of recalcitrant humic substances have been regarded as an important mechanism for soil organic matter (SOM) stabilization. Based on a critical overview of available methods and on results from a cooperative research program, this paper evaluates how relevant recalcitrance is for the long‐term stabilization of SOM or its fractions. Methodologically, recalcitrance is difficult to assess, since the persistence of certain SOM fractions or specific compounds may also be caused by other stabilization mechanisms, such as physical protection or chemical interactions with mineral surfaces. If only free particulate SOM obtained from density fractionation is considered, it rarely reaches ages exceeding 50 y. Older light particles have often been identified as charred plant residues or as fossil C. The degradability of the readily bioavailable dissolved or water‐extractable OM fraction is often negatively correlated with its content in aromatic compounds, which therefore has been associated with recalcitrance. But in subsoils, dissolved organic matter aromaticity and biodegradability both are very low, indicating that other factors or compounds limit its degradation. Among the investigated specific compounds, lignin, lipids, and their derivatives have mean turnover times faster or similar as that of bulk SOM. Only a small fraction of the lignin inputs seems to persist in soils and is mainly found in the fine textural size fraction (<20 µm), indicating physico‐chemical stabilization. Compound‐specific analysis of 13 C : 12 C ratios of SOM pyrolysis products in soils with C3‐C4 crop changes revealed no compounds with mean residence times of > 40–50 y, unless fossil C was present in substantial amounts, as at a site exposed to lignite inputs in the past. Here, turnover of pyrolysis products seemed to be much longer, even for those attributed to carbohydrates or proteins. Apparently, fossil C from lignite coal is also utilized by soil organisms, which is further evidenced by low 14 C concentrations in microbial phospholipid fatty acids from this site. Also, black C from charred plant materials was susceptible to microbial degradation in a short‐term (60 d) and a long‐term (2 y) incubation experiment. This degradation was enhanced, when glucose was supplied as an easily available microbial substrate. Similarly, SOM mineralization in many soils generally increased after addition of carbohydrates, amino acids, or simple organic acids, thus indicating that stability may also be caused by substrate limitations. It is concluded that the presented results do not provide much evidence that the selective preservation of recalcitrant primary biogenic compounds is a major SOM‐stabilization mechanism. Old SOM fractions with slow turnover rates were generally only found in association with soil minerals. The only not mineral‐associated SOM components that may be persistent in soils appear to be black and fossil C.