Vegetation type-associated variations in soil enzyme stoichiometry, bacterial composition and functional potential in an arid mining ecosystem of Xinjiang, China

Introduction Vegetation types are associated with different soil microbial and biochemical characteristics in naturally recovering arid mining areas; however, variations in soil enzyme stoichiometry, bacterial community composition, and functional potential among vegetation types remain inadequately understood. Clarifying these differences is important for understanding vegetation–soil relationships in mining subsidence areas. Methods We evaluated five vegetation types influenced soil microbial biomass, extracellular enzyme activities, and key stoichiometric indices. High-throughput sequencing was employed to characterize bacterial community structures and functional traits, thereby elucidating the relationships between soil physicochemical properties and microbial processes. Results Our results revealed clear vegetation-dependent differences in microbial biomass and enzyme activities. Microbial biomass carbon was greatest under Caragana microphylla, whereas microbial biomass nitrogen peaked under Anabasis salsa. Carbon-acquiring enzymes, nitrogen-acquiring enzymes, and the phosphorus-mineralizing enzyme generally showed their highest activities under Caroxylon passerinum. These patterns indicated that soil microorganisms across vegetation types were co-limited by carbon and phosphorus. The 10 dominant bacterial phyla collectively accounted for more than 90% of the bacterial community, with Proteobacteria, Gemmatimonadota, Actinobacteria, Acidobacteria, and Chloroflexi among the major groups. Proteobacteria showed relatively high abundance in A. alatavicum, whereas Acidobacteria were relatively more abundant in C. passerinum and A. salsa. Redundancy analysis (RDA) accounted for 55.63% (RDA1) and 25.05% (RDA2) of the variance in community composition, identifying soil organic carbon (SOC) and total nitrogen (TN) as primary environmental drivers. Microbial biomass, phosphorus availability, and soil texture were key determinants of carbon and phosphorus limitation, with biological factors (e.g., Actinobacteria) and physical attributes exerting dominant influences. Discussion Overall, five vegetation types were associated with distinct patterns of soil enzyme stoichiometry, bacterial diversity, and predicted metabolic functions, providing a comparative basis for vegetation selection and management in naturally recovering arid mining areas.