Changes in species composition across communities, i.e., β-diversity, is a central focus of ecology. Compared to macroorganisms, the β-diversity of soil microbes and its drivers are less studied. Whether the determinants of soil microbial β-diversity are consistent between soil depths and between abundant and rare microorganisms remains controversial. Here, using the 16S-rRNA of soil bacteria and archaea sampled at different soil depths (0–10 and 30–50 cm) from 32 sites along an aridity gradient of 1500 km in the temperate grasslands in northern China, we compared the effects of deterministic and stochastic processes on the taxonomic and phylogenetic β-diversity of soil microbes. Using variation partitioning and null models, we found that the taxonomic β-diversity of the overall bacterial communities was more strongly determined by deterministic processes in both soil layers (the explanatory power of environmental distance in topsoil: 25.4%; subsoil: 47.4%), while their phylogenetic counterpart was more strongly determined by stochastic processes (the explanatory power of spatial distance in topsoil: 42.1; subsoil 24.7%). However, in terms of abundance, both the taxonomic and phylogenetic β-diversity of the abundant bacteria in both soil layers was more strongly determined by deterministic processes, while those of rare bacteria were more strongly determined by stochastic processes. In comparison with bacteria, both the taxonomic and phylogenetic β-diversity of the overall abundant and rare archaea were strongly determined by deterministic processes. Among the variables representing deterministic processes, contemporary and historical climate and aboveground vegetation dominated the microbial β-diversity of the overall and abundant microbes of both domains in topsoils, but soil geochemistry dominated in subsoils. This study presents a comprehensive understanding on the β-diversity of soil microbial communities in the temperate grasslands in northern China. Our findings highlight the importance of soil depth, phylogenetic turnover, and species abundance in the assembly processes of soil microbial communities.

Exploring the biogeographic patterns of soil microbial diversity is critical for understanding mechanisms underlying the response of soil processes to climate change. Using top- and subsoils from an ∼1,500-km temperate grassland transect, we find divergent patterns of microbial diversity and its determinants in the topsoil versus the subsoil. Furthermore, we find important and direct legacy effects of historical climate change on the microbial diversity of subsoil yet indirect effects on topsoil. Our findings challenge the conventional assumption of similar geographic patterns of soil microbial diversity along soil profiles and help to improve our understanding of how soil microbial communities may respond to future climate change in different regions with various climate histories.

Biogeographic patterns and drivers of soil microbial diversity have been extensively studied in the past few decades. However, most research has focused on the topsoil while the subsoil is assumed to have similar microbial diversity patterns as the topsoil. Here we compare patterns and drivers of microbial diversity in the top- (0-10 cm) versus subsoils (30-50 cm) of temperate grasslands in Inner Mongolia of China along an aridity gradient covering a ~1500-km transect from arid to mesic ecosystems. Counter to the conventional assumption, we find contrasting biogeographic patterns of diversity and influencing factors for different bacterial and archaeal groups and between depths. While bacterial diversity increases with increasing aridity, archaeal diversity decreases. Microbial diversity in the topsoil is most strongly influenced by aboveground vegetation, but is most strongly influenced by historical temperature anomaly since the Last Glacial Maximum (LGM) in the subsoil. Moreover, the biogeographic patterns of top-subsoil diversity difference varies for different microbial groups and is overall most strongly influenced by soil fertility difference between depths and historical temperature anomaly. These findings suggest that diversity patterns observed in the topsoil may not be readily applied to the subsoil horizons. For the subsoil in particular, historical climate plays a vital role in the distribution of various microbial groups. Overall, our study provides novel information for understanding and predicting soil microbial diversity patterns at depth. IMPORTANCE: Exploring the biogeographic patterns of soil microbial diversity is critical for understanding mechanisms underlying the response of soil processes to climate change. Using top- and subsoils from a ~1500-km temperature grassland transect, we find divergent patterns of microbial diversity and its determinants in the top-versus subsoils. Furthermore, we find important legacy effect of historical climate change on the microbial diversity of subsoil but not topsoil. Our findings challenge the conventional assumption of similar geographic patterns of soil microbial diversity along soil profiles and help to improve our understanding of how soil microbial communities may respond to future climate change in different regions with varied climate history.