Exploring the structural characteristics of arid soil microbial communities and their assembly mechanisms is important for understanding the ecological characteristics of arid zone soils and promoting ecological restoration. In this study, we used Illumina high-throughput sequencing technology to study soils in the arid zone of the Lake Ebinur basin, determined the differences among soil microbial community structures in the study area under different water–salt gradients, and investigated the effects of environmental factors on microbial community structure and assembly mechanisms. The results show the following: the microbial community alpha diversity exhibited a significantly higher low water–salt gradient (L) than high water–salt gradient (H) and medium water–salt gradient (M). The pH was most strongly correlated with soil microbial community structure, where the alpha diversity indices of the bacterial community and fungal community were significantly negatively correlated with pH, and the Bray–Curtis distance of bacterial community was significantly positively correlated with pH (p < 0.05). The complexity of bacterial community co-occurrence networks showed a significantly higher L than H and M, and the complexity of fungal community co-occurrence network showed a significantly lower L than H and M. The cooperative relationship of H and M in the co-occurrence networks was stronger than that of the L, and the key species of the microbial co-occurrence network were different under different water–salt gradients. Stochastic processes dominated the assembly mechanism of the microbial community structure of soil, and the explanation rates of deterministic and stochastic processes were different under different water–salt gradients, with the highest explanation rate of stochastic processes on the L accounting for more than 90%. In summary, the soil microbial community structure and assembly mechanisms significantly differed across water–salt gradients, and these findings can help provide a reference for further research on soil microbiology in arid zones.

Extreme rainfall and soil moisture play important roles in the survival, community composition, and ecosystem function of desert plants. This study focused on arid desert riparian forests ecosystems in the Ebinur Lake Basin of Xinjiang, China. We analyzed the effects of rainfall and soil moisture on species composition, indicator species, β diversity, species turnover, and nestedness using three consecutive years of community surveys. A zero-model combined with a Bayesian framework was used to explore the response of species turnover and nestedness to soil moisture and rainfall, and variance decomposition was used to quantify the relative importance of spatial distance, rainfall, and soil factors in determining species composition. The results indicated the following: (1) when rainfall was high, the richness and abundance of annual herbaceous plants increased. The proportion of the community based on richness (32%) and abundance (58.1%) of annual herbaceous plants in 2016 was higher than that in 2015 and 2017. The Jaccard, Bray–Curtis, and Chao indexes of the community in years with higher rainfall were significantly higher than in years with lower rainfall; however, a lag effect was also observed. (2) Soil factors explained 5% of the changes in community composition, rainfall explained 12% of the changes in community composition, and spatial distance, soil factors, and rainfall jointly explained 32% of the changes in community composition. (3) We also showed that high soil moisture leads to greater β diversity than low soil moisture. Rainfall had the greatest explanatory power on the measured values of β diversity (19.6%) and species turnover (38%), and the factor with the greatest explanatory power for species nestedness was the interaction between rainfall and soil moisture (26.2%). Our findings indicate that drought and rainfall drive differences in plant community composition, with rainfall playing a dominant role. These results provide a basis for understanding the impact of extreme rainfall events on arid ecosystem functions.