Old and stable soil organic matter is not necessarily chemically recalcitrant: implications for modeling concepts and temperature sensitivity

Abstract Soil carbon turnover models generally divide soil carbon into pools with varying intrinsic decomposition rates. Although these decomposition rates are modified by factors such as temperature, texture, and moisture, they are rationalized by assuming chemical structure is a primary controller of decomposition. In the current work, we use near edge X‐ray absorption fine structure (NEXAFS) spectroscopy in combination with differential scanning calorimetry (DSC) and alkaline cupric oxide (CuO) oxidation to explore this assumption. Specifically, we examined material from the 2.3–2.6 kg L −1 density fraction of three soils of different type (Oxisol, Alfisol, Inceptisol). The density fraction with the youngest 14 C age (Oxisol, 107 years) showed the highest relative abundance of aromatic groups and the lowest O ‐alkyl C/aromatic C ratio as determined by NEXAFS. Conversely, the fraction with the oldest C (Inceptisol, 680 years) had the lowest relative abundance of aromatic groups and highest O ‐alkyl C/aromatic C ratio. This sample also had the highest proportion of thermally labile materials as measured by DSC, and the highest ratio of substituted fatty acids to lignin phenols as indicated by CuO oxidation. Therefore, the organic matter of the Inceptisol sample, with a 14 C age associated with ‘passive’ pools of carbon (680 years), had the largest proportion of easily metabolizable organic molecules with low thermodynamic stability, whereas the organic matter of the much younger Oxisol sample (107 years) had the highest proportion of supposedly stable organic structures considered more difficult to metabolize. Our results demonstrate that C age is not necessarily related to molecular structure or thermodynamic stability, and we suggest that soil carbon models would benefit from viewing turnover rate as codetermined by the interaction between substrates, microbial actors, and abiotic driving variables. Furthermore, assuming that old carbon is composed of complex or ‘recalcitrant’ compounds will erroneously attribute a greater temperature sensitivity to those materials than they may actually possess.