Abstract:The community composition and structural characteristics of soil bacteria are primarily shaped by deterministic processes based on niche theory and stochastic processes based on neutral theory. This study analyzed the contribution of stochastic processes for bacteria community assembly among four typical habitats in Yancheng coastal wetland using 16S rRNA gene high-throughput sequencing data and neutral community model. The results showed that: (1) Soil bacteria could be divided into neutral taxa, above-neutral taxa, and below-neutral taxa based on the NCM. The above taxa were primarily composed of Campylobacterota and Proteobacteria. The dominant phyla of the below taxa included Proteobacteria and Bacteroidota, while the neutral taxa was dominantly consisted of Proteobacteria, Desulfobacterota, and Chloroflexota. There existed some differences in soil bacterial composition among these three taxa. (2) Among the typical habitats studied, the composition and structural characteristics of soil bacterial communities were relatively similar in the surface and middle soil layers, but differed from those in the deep soil layer. The α diversity of soil bacteria was lowest in mudflat while it was highest in Spartina alterniflora community, both above taxa and neutral taxa were high in these two habitats. Instead, the soil bacterial diversity in the Suaeda salsa and Phragmites australis communities showed minor differences, with a higher prevalence of below taxa. (3) The neutral taxa showed consistent composition and community structure in different habitats, while the above- and below-neutral taxa varied significantly. Overall, stochastic processes predominantly influenced the assembly of soil bacterial communities in the mudflat and Spartina alterniflora communities, whereas the influence was relatively minor in the Suaeda salsa and Phragmites australis communities. The differences in soil bacterial communities were mainly reflected in the composition and structural characteristics of the above- and below-neutral taxa.