Threshold Effects of Moderate Wildfire Drive Depth-Dependent Responses in Subtropical Forest Soil Microbial Communities

Background and Aims Against the backdrop of global warming, the frequency and intensity of wildfires have significantly increased, exerting profound impacts on the structure and function of soil microbial communities. However, the mechanisms underlying microbial responses to varying levels of wildfire severity in subtropical montane forests remain poorly understood. Methods This study investigated subtropical forests in Huaxi District, Guizhou Province, employing a wildfire severity gradient design (unburned, light, moderate, severe) combined with depth-stratified soil sampling (topsoil: 0–20 cm, subsoil: 20–40 cm). Building on metagenomic sequencing and co-occurrence network analyses, we elucidate the coupled relationships among community diversity, interaction structure, and assembly processes. Results (1) The bacterial richness (ACE) increased continuously with wildfire severity, peaking at severe wildfire; evenness (Pielou_e) increased significantly only at moderate wildfire, exhibiting an intermediate-disturbance optimum. For fungi, richness in the topsoil layer increased with wildfire severity, whereas in the subsoil layer it peaked at moderate wildfire. (2) Co-occurrence networks showed a non-linear response: in bacteria, the proportion of positive edges rose sharply at moderate wildfire (> 90%); in fungi, modularity strengthened in the subsoil layer at moderate wildfire but decreased in the topsoil layer at severe wildfire, indicating “depth-differentiated” structural reorganization. (3) Neutral community model fitting indicated that bacterial assembly was dominated by stochastic processes (R²>0.78), whereas fungi deviated from the neutral model and were more strongly shaped by deterministic processes (environmental filtering/niche selection), with these effects being more pronounced in the subsoil layer. Conclusion Overall, moderate wildfire constitutes an ecological threshold that optimizes microbial community structure and functional potential, while soil depth reshapes post-fire successional trajectories by altering assembly processes and network topology. This study provides a theoretical basis for targeted post-fire microbial restoration in subtropical forests.