Stochastic processes and biotic interactions shape the structure of soil bacterial and fungal communities during secondary succession in high-altitude oldfields

Increasing evidence suggests that both deterministic (‘niche’) and stochastic (‘neutral’) processes shape soil microbial communities, ultimately influencing ecosystem functioning. Despite studies on microbial community assembly, the role of species interactions remains poorly understood due to difficulties of assessment in natural habitats. We investigated the balance between the two kinds of assembly processes during secondary succession, in which the arrival of propagules is stochastic, while interspecific interactions and their impact on ecosystem properties are deterministic. We categorized succession into five stages using a well-dated chronosequence of abandoned arable lands on the Tibetan Plateau: stage 1 (continued arable land), stage 2 (arable abandoned for 2 years), stage 3 (arable abandoned for 10 years), stage 4 (arable abandoned for 20 years), and natural grassland. We examined the relationship between community assembly and successional stage, employing co-occurrence networks analysis to assess assembly processes of the soil microbial communities. We found that stochastic processes, i.e. dispersal limitation and drift - as estimated with beta Net Relatedness Index (βNRI) - dominated in shaping the structure of bacterial and fungal communities during succession. The relative abundance of ecological modules within microbial communities changed strongly with succession, mainly concurrent with soil carbon content, nitrogen content and soil moisture. Moreover, species interactions, inferred from ecological modules, were related to homogeneous selection, drift and dispersal limitation processes. Our finding highlights the significant role of biotic interactions in microbial community assembly, suggesting that future studies should integrate species interactions to better predict community dynamics and ecosystem trajectories.