Climate–soil gradients differentially shape bacterial and fungal assembly pathways across grasslands on the Mongolian Plateau

Soil bacterial and fungal responses to environmental gradients are increasingly well documented, but whether their diversity, turnover, network organization, and assembly vary in parallel across the same broad grassland climate–soil gradient remains unclear. In this study, we investigated 189 soil samples from steppified desert, desert steppe, typical steppe, and meadow steppe across the Mongolian Plateau using 16S rRNA gene and ITS amplicon sequencing, co-occurrence network analysis, iCAMP, and partial least squares path modeling (PLS-PM). The results indicated that observed bacterial richness remained relatively high in more arid grasslands, whereas bacterial evenness declined. In contrast, observed fungal richness and phylogenetic diversity increased toward wetter and more nutrient-rich grasslands. Bacterial turnover was most strongly associated with soil pH (R = 0.764), whereas fungal turnover was more closely associated with climatic variation, particularly aridity (R = 0.704), although soil pH also remained important. In co-occurrence networks, bacteria accounted for 62 – 77% of nodes and 90% of module hubs, while fungal contributions to module linkage increased toward less arid grasslands. Stochastic processes accounted for 55 – 61% of bacterial and 56 – 70% of fungal assembly, with dispersal limitation representing the largest iCAMP-assigned component. The PLS-PM explained 25% of the variation in bacterial assembly but only 3.8% in fungal assembly, indicating that the measured climate and soil variables captured substantially more bacterial than fungal assembly variation and explained little of the latter. Associations with field-based grazing categories varied among regional grassland groups and generally weaker than the broad climate–soil pattern. Overall, bacterial and fungal communities showed contrasting but partly overlapping ecological pathways across grasslands on the Mongolian Plateau. These findings highlight the need to evaluate bacteria and fungi separately when predicting soil microbiome responses to aridification and grazing disturbance.