UC San Diego

Rice paddy soil-associated microbiota participate in biogeochemical processes that underpin rice yield and soil sustainability, yet continental-scale biogeographic patterns of paddy soil microbiota remain elusive. The soil bacteria of four typical Chinese rice-growing regions were characterized and compared to those of nonpaddy soils. The paddy soil bacteria were significantly less diverse, with unique taxonomic and functional composition, and harbored distinct cooccurrence network topology. Both stochastic and deterministic processes shaped soil bacteria assembly, but paddy samples exhibited a stronger deterministic signature than nonpaddy samples. Compared to other environmental factors, climatic factors such as mean monthly precipitation and mean annual temperature described most of the variance in soil bacterial community structure. Cooccurrence network analysis suggests that the continental biogeographic variance in bacterial community structure was described by the competition between two mutually exclusive bacterial modules in the community. Keystone taxa identified in network models (Anaerolineales, Ignavibacteriae, and Deltaproteobacteria) were more sensitive to changes in environmental factors, leading us to conclude that environmental factors may influence paddy soil bacterial communities via these keystone taxa. Characterizing the uniqueness of bacterial community patterns in paddy soil (compared to nonpaddy soils) at continental scales is central to improving crop productivity and resilience and to sustaining agricultural soils. IMPORTANCE Rice fields provide food for over half of the world's human population. The ecology of paddy soil microbiomes is shaped by human activities, which can have a profound impact on rice yield, greenhouse gas emissions, and soil health. Investigations of the soil bacteria in four typical Chinese rice-growing regions showed that (i) soil bacterial communities maintain highly modularized species-to-species network structures; (ii) community patterns were shaped by the balance of integrated stochastic and deterministic processes, in which homogenizing selection and dispersal limitation dominate; and (iii) deterministic processes and climatic and edaphic factors influence community patterns mainly by their impact on highly connected nodes (i.e., keystone taxa) in networks. Characterizing the unique ecology of bacterial community patterns in paddy soil at a continental scale may lead to improved crop productivity and resilience, as well as sustaining agricultural soils.