Soil Feedbacks to the Climate System


A key question in Earth system science is: Will warming lead to increased soil organic matter decay and an accelerated release of soil carbon as CO2? If yes, a self-reinforcing feedback would result with warming begetting warming. In 1991, a replicated, in situ soil-warming experiment was established at the Harvard Forest in central Massachusetts to address this question. Rates of CO2 production have been measured monthly for microbial and root respiration from April through November. Initially, warmed plots had higher respiration than controls, but after about a decade, the warming-accelerated CO2 production decreased and returned to background levels. However, during the last seven years of the study (years 16–22), soil respiration again increased in the heated plots relative to the control plots – a long-term response to soil warming never before documented. Based on measurements made over the first 15 years that showed the depletion of the soil’s labile carbon pool, the investigators hypothesized that much of the carbon respired over the last seven years has come from the recalcitrant soil carbon pool. Using13C compound-specific soil incubation studies, they found that long-term soil warming increases the microbial carbon-use efficiency (CUE) associated with the degradation of complex (recalcitrant) carbon compounds such as phenol, but that the CUE of simple carbon compounds such as glucose was not temperature sensitive. Additional preliminary data shows a shift in microbial community structure in the heated plots that indicates an increase in taxa or pathways adapted to recalcitrant carbon decomposition. This long-term study suggests that the soil
microbial community will adapt to long-term warming in a way that will lead to a depletion of the recalcitrant soil carbon stocks and a self-reinforcing feedback to the climate system.


Frey, S. D., J. Lee, J. M. Melillo, J. Six. 2013. “Soil Carbon Cycling: The Temperature Response of Soil Microbial Efficiency and Its Feedback to Climate,” Nature Climate Change 3, 395–98. DOI: 10.1038/nclimate1796.