Are microbes the major determinant of global soil carbon storage?

Soils store a massive amount of carbon, much more than is stored in vegetation or in the atmosphere.  How soil organic carbon (SOC) forms and persists remains uncertain. This makes it challenging to predict how soil carbon will respond to climatic change. Recently, Tao et al. reported in Nature that microbes and their carbon use efficiency are the primary determinant of global SOC storage. If this holds true it would be a paradigm shift - as to date carbon inputs to the soil have been considered the primary determinant. In a Matter Arising we challenge this findings, and argue for the investigation of multiple mechanisms of SOC stabilization and loss, rather than solely focusing on CUE.

As direct observations of microbial carbon use efficiency are scarce, the study by Tao et al. relied on model-data assimilation to indirectly derive carbon use efficiency at global scale. Soil carbon models are infamously unreliable with little progress being made in capturing global patterns and underlying drivers in global models during the last decade.  With the help of one of the fastest computer and an impressively large dataset on 57,267 globally distributed vertical SOC profiles, Tao et al. optimized a selection of uncertain model parameters in a soil carbon model. The authors argue that this procedure enabled their soil model to produce reliable and robust predictions and base their analysis largely on model predictions.

Several shortcomings of the approach by Tao et al. that have lead to unreliable model predictions are now being reported in a Matters Arising in Nature lead by Xianjin He from the Le Laboratoire des Sciences du Climat et de l'Environnement (LSCE). We argue that the relationship between microbial carbon use efficiency and soil organic carbon critically depend on unconstrained model assumptions, and thus cannot be trusted. 

A major issue is that the parameter which controls the mortality of microbes was not included into the optimization based on the SOC profiles. More importantly, the parameter was set to a value which has been demonstrated earlier to minimizes the dependency of SOC to soil carbon inputs. As a consequence the model predictions are arguably predisposed to predict a major influence of microbes on global SOC storage - irrespectively of how much data was used to in the model optimisation. 

We highlight this and other shortcomings in the Matter Arising and warn that it is too early for a paradigm shift and that future research efforts should be allocated towards investigating multiple mechanisms of SOC stabilization and loss, rather than solely focusing on CUE. Tao et al. has published a reply to our critic. There is also further critic of Tao et al published in National Science Review.

References:

He, X., Abramoff, R.Z., Abs, E. et al. Model uncertainty obscures major driver of soil carbon. Nature 627, E1–E3 (2024). https://doi.org/10.1038/s41586-023-06999-1

Tao, F., Huang, Y., Hungate, B.A. et al. Microbial carbon use efficiency promotes global soil carbon storage. Nature 618, 981–985 (2023). https://doi.org/10.1038/s41586-023-06042-3

Georgiou, K., Abramoff, R.Z., Harte, J. et al. Microbial community-level regulation explains soil carbon responses to long-term litter manipulations. Nat Commun 8, 1223 (2017). https://doi.org/10.1038/s41467-017-01116-z

Tao, F., Houlton, B.Z., Frey, S.D. et al. Reply to: Model uncertainty obscures major driver of soil carbon. Nature 627, E4–E6 (2024). https://doi.org/10.1038/s41586-023-07000-9

Ke-Qing Xiao, Chao Liang, Zimeng Wang, Jingjing Peng, Yao Zhao, Ming Zhang, Mingyu Zhao, Shuiqing Chen, Yong-Guan Zhu, Caroline L Peacock, Beyond microbial carbon use efficiency, National Science Review, 2024;, nwae059, https://doi.org/10.1093/nsr/nwae059