The role of soil biodiversity in regulating multiple ecosystem functions is poorly understood, limiting our ability to predict how soil biodiversity loss might affect human wellbeing and ecosystem sustainability. Here, combining a global observational study with an experimental microcosm study, we provide evidence that soil biodiversity (bacteria, fungi, protists and invertebrates) is significantly and positively associated with multiple ecosystem functions. These functions include nutrient cycling, decomposition, plant production, and reduced potential for pathogenicity and belowground biological warfare. Our findings also reveal the context dependency of such relationships and the importance of the connectedness, biodiversity and nature of the globally distributed dominant phylotypes within the soil network in maintaining multiple functions. Moreover, our results suggest that the positive association between plant diversity and multifunctionality across biomes is indirectly driven by soil biodiversity. Together, our results provide insights into the importance of soil biodiversity for maintaining soil functionality locally and across biomes, as well as providing strong support for the inclusion of soil biodiversity in conservation and management programmes. Combining field data from 83 sites on five continents, together with microcosm experiments, the authors show that nutrient cycling, decomposition, plant production and other ecosystem functions are positively associated with a higher diversity of a wide range of soil organisms. Marie Sklodowska-Curie We thank N. Fierer, M. Gebert, J. Henley, V. Ochoa, F. T. Maestre and B. Gozalo for their help with laboratory analyses; O. Sala, C. Siebe, C. Currier, M. A. Bowker, V. Parry, H. Lambers, P. Vitousek, V. M. Pena-Ramirez, L. Riedel, J. Larson, K. Waechter, W. Williams, S. Williams, B. Sulman, D. Buckner and B. Anacker for their help with soil sampling in Colorado, Hawaii, Iceland, New Mexico, Arizona, Mexico and Australia; the City of Boulder Open Space and Mountain Parks for allowing us to conduct these samplings; C. Cano-Diaz for her advice about R analyses; S. K. Travers for her help with mapping. This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement no. 702057. M.D.-B. is supported by the Spanish Government under a Ramon y Cajal contract RYC2018-025483-I. This research is supported by the Australian Research Council projects (DP170104634; DP190103714). S.A. and F.D.A. are funded by FONDECYT 1170995, IAI-CRN 3005, PFB-23 (from CONICYT) and P05-002 (from Millennium Scientific Initiative). N.A.C. acknowledges support from Churchill College, University of Cambridge; and M.A.W. from the Wilderness State Park, Michigan for access to sample soil and conduct ecosystem survey. B.K.S. acknowledges a research award from the Humboldt Foundation. J.-Z.H. acknowledges support from the Australia Research Council (project DP170103628); and A.G. from the Spanish Ministry (project CGL2017-88124-R). F.B. thanks the Spanish Ministry and FEDER funds for the CICYT project AGL2017-85755-R, the CSIC project 201740I008 and funds from 'Fundacion Seneca' from Murcia Province (19896/GERM/15). P.T. thanks K. Little for her help with laboratory analyses. S.C.R. was supported by the US Geological Survey Ecosystems Mission Area. Any use of trade, firm or product names is for descriptive purposes only and does not imply endorsement by the US Government. S.N. was funded by the Austrian Science Fund (grant Y801-B16). Public domain – authored by a U.S. government employee
