BackgroundThe relative importance of stochasticity versus determinism in soil bacterial communities is unclear, as are the possible influences that alter the balance between these. Here, we investigated the influence of spatial scale on the relative role of stochasticity and determinism in agricultural monocultures consisting only of wheat, thereby minimizing the influence of differences in plant species cover and in cultivation/disturbance regime, extending across a wide range of soils and climates of the North China Plain (NCP). We sampled 243 sites across 1092 km and sequenced the 16S rRNA bacterial gene using MiSeq. We hypothesized that determinism would play a relatively stronger role at the broadest scales, due to the strong influence of climate and soil differences in selecting many distinct OTUs of bacteria adapted to the different environments. In order to test the more general applicability of the hypothesis, we also compared with a natural ecosystem on the Tibetan Plateau.ResultsOur results revealed that the relative importance of stochasticity vs. determinism did vary with spatial scale, in the direction predicted. On the North China Plain, stochasticity played a dominant role from 150 to 900 km (separation between pairs of sites) and determinism dominated at more than 900 km (broad scale). On the Tibetan Plateau, determinism played a dominant role from 130 to 1200 km and stochasticity dominated at less than 130 km. Among the identifiable deterministic factors, soil pH showed the strongest influence on soil bacterial community structure and diversity across the North China Plain. Together, 23.9% of variation in soil microbial community composition could be explained, with environmental factors accounting for 19.7% and spatial parameters 4.1%.ConclusionsOur findings revealed that (1) stochastic processes are relatively more important on the North China Plain, while deterministic processes are more important on the Tibetan Plateau; (2) soil pH was the major factor in shaping soil bacterial community structure of the North China Plain; and (3) most variation in soil microbial community composition could not be explained with existing environmental and spatial factors. Further studies are needed to dissect the influence of stochastic factors (e.g., mutations or extinctions) on soil microbial community distribution, which might make it easier to predictably manipulate the microbial community to produce better yield and soil sustainability outcomes.Electronic supplementary materialThe online version of this article (10.1186/s40168-018-0409-4) contains supplementary material, which is available to authorized users.

The spatial distribution of bacteria in bulk soil has been well studied, but little is known about the bacterial biogeography in the rhizosphere of crops. Here, we investigated bacterial distribution in bulk soil, loosely-and tightly-bound soils, from wheat fields distributed across 800,000 km 2 of the North China Plain. Bacterial community composition differed dramatically among bulk and rhizospheric soils, and bacterial diversity decreased with the root proximity. Soil pH correlated with bacterial community composition and diversity in three compartments. Bacterial community in tightly bound soil formed a hub-based network topology with higher transitivity and greater number of central nodes compared with loosely bound and bulk soils, potentially as a result of more direct ecological interactions between the members of the tightly bound soil compartment. Bulk and rhizospheric soils maintained similar compositional distance decay patterns (with equal decay rates), but distinct phylogenetic distance decay patterns (with steeper slope of tightly bound soil). Geographical distance described a relatively greater proportion of bacterial spatial distribution in tightly bound soil, compared with loosely bound soil and bulk soil. Deterministic processes dominated the assemblage of bacterial communities in all soil compartments, while phylogenetic clustering was weaker in tightly bound soil. Taken together, our results suggest distinct bacterial network structure and distribution patterns among bulk soil, loosely bound soil and tightly bound rhizospheric soil, which could possibly result in potential functional differentiation.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.