Characteristics of Soil Microbial Communities and Their Associated Environmental Variables Under Different Vegetation Types in Jianfengling, Hainan

Soil microorganisms play essential roles in maintaining the function and stability of forest ecosystems. In this study, conducted in the Jianfengling Nature Reserve in Hainan, we performed Illumina high-throughput sequencing, in combination with principal coordinate analysis (PCoA), redundancy analysis, and other statistical methods, to compare the composition, community structure, and diversity of the soil microbiota in four types of forest vegetation, namely, monsoon forest, lowland rainforest, montane rainforest, and artificial forest (AF), and to identify the environmental variables associated with differences among the forest types. Different vegetation types differed significantly with respect to pH, total phosphorus content (TP), elevation, and concentrations of manganese (Mn) and iron (Fe). In contrast, no significant differences were observed in soil water content, bulk density, total nitrogen (TN), and total organic carbon (TOC) content. Compared to AF, natural forests were characterized by higher concentrations of TN, TOC, TP, Mn, and Fe, although the differences in TN and TOC were not statistically significant. For all four of the assessed vegetation types, Pseudomonadota, Actinomycetota, Acidobacteriota, and Verrucomicrobiota were identified as the dominant bacterial phyla, whereas Ascomycota, Basidiomycota, and Mucoromycota were predominant among the fungal phyla, indicating that these phyla are capable of surviving in different soil ecosystems. Although no significant differences in bacterial diversity were detected among vegetation types, PCoA revealed pronounced differentiation in bacterial community composition (PERMANOVA, p = 0.0001). Notably, within-group multivariate dispersion also differed significantly (PERMDISP, p = 0.0129), indicating that the observed compositional differences may be associated with the combined effects of centroid location effect and within-group multivariate dispersion. In contrast, significant differences were observed in fungal diversity, whereas differences in fungal community composition were not significant in the PCoA (PERMANOVA, p = 0.1777). These results suggest that vegetation type primarily influenced fungal diversity rather than fungal community structure. Mn concentration exhibited the highest marginal explanatory power for the variation in the bacterial community, with a significant effect; it was positively correlated with the ACE, Chao1, and Simpson indices, and negatively correlated with the Shannon index. TOC exhibited the highest marginal explanatory power within the fungal community, showing a negative correlation with the ACE, Chao1, and Shannon indices and a positive correlation with the Simpson index. Collectively, these findings provide insights into microbial differentiation within forest soils associated with different vegetation types and highlight distinct patterns in bacterial and fungal diversity, which will potentially serve as a basis for the development of conservation strategies.