Microbial carbon metabolic activity and bacterial cross-profile network in paddy soils of different fertility

Soil microbial carbon metabolic activity is critical to the carbon balance of terrestrial ecosystems. How the microbial interactions induce microbial carbon metabolic activity in rice-based ecosystems at different fertility levels are unclear. Here, we investigated the association of microbial carbon metabolic activity with soil phys-icochemical properties, microbial diversity and interactions at three fertilization practices and four soil profile levels (0-10 cm, 10-20 cm, 20-40 cm, 40-60 cm) in two paddy soils with different fertility levels. Our results found that microbial carbon metabolic activity and content of soil nutrient increased with fertilizer application practices but decreased with increasing soil depth. We provide an effective tool for studying microbial in-teractions across the soil profiles named cross-profile network model. The cross-profile showed that 10-20 cm had the highest average degree of edges and nodes in the co-occurrence network. The network indicated that the combination of organic and inorganic fertilizers increased the number of nodes and edges, while the single application of organic fertilizers decreased the number of nodes and edges in fertile soil. The taxonomic diversity of microorganisms and the carbon-related functional diversity were significantly correlated with microbial in-teractions by linear/non-linear fitted models. Partial least squares path modeling showed that microbial carbon metabolic activity in fertile soil was directly affected by dissolved organic carbon. However, soil physicochemical properties indirectly affected the carbon metabolic activity by mediating microbial diversity and interactions in infertile soil. For the first time, we incorporated potential microbial interactions between soil layers into a network model, emphasizing the importance of microbial diversity and interactions in regulating carbon metabolism activities. Our results pave the way for the integration of bacterial community diversity and in-teractions to the predict soil organic carbon dynamics.