Cross-biome drivers of soil bacterial alpha diversity on a worldwide scale

We lack a defined suite of attributes that allow us to universally predict the distribution of bacterial diversity across and within globally distributed biomes. Using data from a global survey, including 237 locations and multiple environmental predictors, we found that only ultraviolet light, forest environments, soil carbon and pH can be considered as significant and globally consistent predictors of soil bacterial diversity, valid within and across biomes (arid, temperate and continental). Bacterial diversity always peaked in grasslands, with moderate-to-low carbon and ultraviolet light levels, and high soil pH. Using these environmental data, we generated the first global predictive map of the distribution of soil bacterial diversity. Our work helps to identify a unique set of environmental attributes for universally predicting the distribution of soil bacterial diversity. This knowledge is key to help predict changes in ecosystem functioning and the provision of essential services under changing environments.