Soil contains more carbon than the atmosphere and vegetation combined1. Understanding the mechanisms controlling the accumulation and stability of soil carbon is critical to predicting the Earth’s future climate2,3. Recent studies suggest that decomposition of soil organic matter is often limited by nitrogen availability to microbes4,5,6 and that plants, via their fungal symbionts, compete directly with free-living decomposers for nitrogen6,7. Ectomycorrhizal and ericoid mycorrhizal (EEM) fungi produce nitrogen-degrading enzymes, allowing them greater access to organic nitrogen sources than arbuscular mycorrhizal (AM) fungi8,9,10. This leads to the theoretical prediction that soil carbon storage is greater in ecosystems dominated by EEM fungi than in those dominated by AM fungi11. Using global data sets, we show that soil in ecosystems dominated by EEM-associated plants contains 70% more carbon per unit nitrogen than soil in ecosystems dominated by AM-associated plants. The effect of mycorrhizal type on soil carbon is independent of, and of far larger consequence than, the effects of net primary production, temperature, precipitation and soil clay content. Hence the effect of mycorrhizal type on soil carbon content holds at the global scale. This finding links the functional traits of mycorrhizal fungi to carbon storage at ecosystem-to-global scales, suggesting that plant–decomposer competition for nutrients exerts a fundamental control over the terrestrial carbon cycle.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
We thank L. Nave and the International Soil Carbon Network for access to their database. C. Hawkes provided feedback during data collection and initial analyses of C storage. C. Iversen, J. Powers and M. Vadeboncouer provided unpublished data that contributed to this analysis. D. Jacquier provided the Australian soil database and E. Carlston helped to extract data from the Australian soil database. C. Shaw provided the Siltanen soil carbon database and the Forest Ecosystem Carbon Database of Canadian soils. T. Baisden provided scans of the California Soil-Vegetation Survey. E. Brzostek, N. Fowler, P. Groffman, E. Hobbie, B. Schlesinger and B. Waring provided feedback on earlier versions of this manuscript. The Center for Tropical Forest Science (CTFS) and Smithsonian Institution Geo-observatories (SIGEO) provided funding for the collection and analysis of soil profile data at large forest dynamics plots, and we thank the many collaborators, field assistants and laboratory technicians who assisted in the collection and analysis of soil profile data. This work benefited from extensive data contributions to the International Soil Carbon Network from both the USDA Natural Resources Conservation Service, National Cooperative Soil Survey, and the US Geological Survey. C.A. was supported by a fellowship from the University of Texas at Austin and by the National Science Foundation Graduate Research Fellowship Program (grant DGE-1110007). A.C.F. was supported by NSF grant number DEB 07-43564 and DOE grants 10-DOE-1053 and DE-SC0006916. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
Extended data figures
Extended data tables