Stochastic assembly process dominates bacterial succession during a long-term microbial enhanced oil recovery

Microbial enhanced oil recovery (MEOR) has been successfully used in oil exploitation to increase oil production. However, the mechanisms of microbial interactions and community assembly related to oil production performance along MEOR process are poorly understood. Here, we investigated the microbiome of an oil reservoir for a period of 5 years under three phases of different treatments with the injection of a mixture of microbes, nutrients, and air at different intensity. During the MEOR process, amplification of functional genes revealed an increase of genes related to hydrocarbon degradation linked to methanogenesis, supported by stable isotope analysis for confirmation of the methanogenesis activity. Meanwhile, a lower contribution of the ubiquitous/common taxa, closer and more positive associations, and lower modularity were observed in bacterial co-occurrence networks, with the rare taxa being the keystone taxa. The null model analysis and structural equation modeling revealed that the contribution of stochastic processes affected by functional groups and co-occurrence patterns to bacterial community increased significantly with the increase of oil production. This provides new insight that stochastic assembly in bacterial community increased along with MEOR process, and it is worthwhile paying attention to the uncertain consequences caused by random evolution since the treatment effect of MEOR is closely related to the in-situ community in oil reservoir.