Multi-year Incubation Experiments Boost Confidence in Model Projections of Long-term Soil Carbon Dynamics

The Science

As the climate warms, soil carbon decomposition by microbes may be accelerated to release more carbon dioxide, but most predictions are based on short-term laboratory incubations that might not reflect rates in situ. Here the authors optimize model projections with the Microbial-ENzyme Decomposition (MEND) model using parameters derived from short- and long-term incubations, and find that only the projections from long-term incubations match long-term field-scale observational changes in soil organic carbon.

The Impact

Model simulations based on long-term experiments predicted small gains in soil organic carbon, similar to observations from many long-term field warming experiments.

Summary

Predictions of long-term changes in soil organic carbon are needed to understand future climate, but most projections are derived from model simulations based on lab incubations of short durations, e.g., hours to days. Here, model projections were compared from incubation datasets ranging from days to years, from four paired forest and grassland sites, and using substrates glucose and cellulose. Model projections derived from short-term experiments predicted greater losses of soil carbon than projections derived from long-term experiments. The projections from the long-term incubations (> 1.5 y) were more similar to the results of a meta-analysis of warming experiments in the field, which predicted small gains in soil carbon over 1- to 10-year time frames. Mechanistically, the findings represent feedbacks in the microbial community, where warming initially releases more organic carbon substrate for decomposition, but later limits reproduction and growth of the microbial community causing small positive increases in soil organic carbon. These findings suggest that long-term incubation experiments are required to accurately model long-term behavior of soil organic carbon.

Principal Investigator

Melanie Mayes
Oak Ridge National Laboratory
[email protected]

Program Manager

Daniel Stover
U.S. Department of Energy, Biological and Environmental Research (SC-33)
Environmental System Science
[email protected]

Funding

This work is financially supported by the U.S. Department of Energy (DOE) Office of Biological and Environmental Research through the Terrestrial Ecosystem Science Scientific Focus Area at Oak Ridge National Laboratory (ORNL), the U.S. National Science Foundation (NSF) HBCU-EiR, and the DOE Genomic Science Program. Financial support from ORNL to Tennessee State University (TSU) was provided through a subcontract. ORNL is managed by the University of Tennessee-Battelle, LLC with the U.S. DOE.

References