The carbon-quality temperature hypothesis: Fact or artefact?

Abstract

Climate warming can reduce global soil carbon stocks by enhancing microbial decomposition. However, the magnitude of this loss remains uncertain because the temperature sensitivity of the decomposition of the major fraction of soil carbon, namely resistant carbon, is not fully known. It is now believed that the resistance of soil carbon mostly depends on microbial accessibility of soil carbon with physical protection being the primary control of the decomposition of protected carbon, which is insensitive to temperature changes. However, it is still unclear whether the temperature sensitivity of the decomposition of unprotected carbon, for example, carbon that is not protected by the soil mineral matrix, may depend on the chemical recalcitrance of carbon compounds. In particular, the carbon-quality temperature (CQT) hypothesis asserts that recalcitrant low-quality carbon is more temperature-sensitive to decomposition than labile high-quality carbon. If the hypothesis is correct, climate warming could amplify the loss of unprotected, but chemically recalcitrant, carbon and the resultant CO₂ release from soils to the atmosphere. Previous research has supported this hypothesis based on reported negative relationships between temperature sensitivity and carbon quality, defined as the decomposition rate at a reference temperature. Here we show that negative relationships can arise simply from the arbitrary choice of reference temperature, inherently invalidating those tests. To avoid this artefact, we defined the carbon quality of different compounds as their uncatalysed reaction rates in the absence of enzymes. Taking the uncatalysed rate as the carbon quality index, we found that the CQT hypothesis is not supported for enzyme-catalysed reactions, which showed no relationship between carbon quality and temperature sensitivity. The lack of correlation in enzyme-catalysed reactions implies similar temperature sensitivity for microbial decomposition of soil carbon, regardless of its quality, thereby allaying concerns of acceleration of warming-induced decomposition of recalcitrant carbon.

Keywords: carbon-quality temperature hypothesis; enzyme-catalysed reactions; global warming; microbial decomposition; soil organic carbon; temperature dependence.

FIGURE 1

Correlations between Q ₁₀ and the logarithm of respiration rate, ln(R _s). (a) Diagram of the carbon‐quality temperature hypothesis, showing a recalcitrant carbon compound with a higher Q ₁₀ compared with that for a labile carbon compound, resulting in a negative relationship between Q ₁₀ and carbon quality. (b) Temperature response curves of respiration rate for two hypothetical carbon compounds. Reanalyses of a global data set of soil incubation experiments, shown in (c), give a significant negative correlation between Q ₁₀ and ln(R ₀), the respiration rate at 0°C, but (d), no correlation between Q ₁₀ and ln(R ₄₃), the respiration rate at 43°C, and (e), the changes of correlation coefficient between Q ₁₀ and ln(R _s) at chosen reference temperatures from 0 to 60°C. The soil respiration data set came from Fierer et al. (2006) and Li et al. (2017).

FIGURE 2

Correlation between activation energy (E _a) and the logarithm of uncatalysed rates, ln(k _non), used here as the carbon quality index. (a) The correlation between ln(k _non) at 25°C and E _a of uncatalysed (open circles) and the corresponding enzyme‐catalysed (closed circles) reactions. The right y‐axis shows the corresponding Q ₁₀ values calculated between 20 and 30°C. The shaded band indicates the range of E _a or Q ₁₀ for enzyme‐catalysed reactions. (b) The correlation coefficient between E _a and ln(k _non) at different reference temperature from 0 to 200°C. The light‐shaded areas show the standard errors of the estimated correlation coefficients for both reactions using the jackknife resampling technique. The dark‐shaded areas indicate the biological temperature range of 0–60°C.

FIGURE 3

Energy diagram for difficult and simple reactions. In the absence of enzymes, a difficult reaction with recalcitrant substrates needs to overcome a higher energy barrier (∆Ea_non) than that of simple reactions. Enzymes have evolved structures that can lower the energy barriers (∆Ea_cat) of both difficult and simple reactions to similar levels through different rate enhancements (k _cat/k _non or ∆Ea).