Artificial vegetation restoration enhances soil organic carbon in abandoned mines through chemical property-driven modulation of bacterial communities

Abstract Restoration of artificial vegetation in mining areas can improve soil health and quality. A healthy soil ecosystem has suitable characteristics and diverse microbes, which can be assessed for balance and stability. This study investigated how cultivating alfalfa, Pennisetum giganteum , Chrysanthemum indicum , Chrysanthemum zawadskii , and Amorpha fruticosa affects mine soil rehabilitation. We analyzed soil chemical properties, bacterial diversity, and composition. Compared with the unplanted control, plant cultivation significantly improved multiple soil chemical properties; alfalfa notably increased soil pH, total carbon (TC), total nitrogen (TN), and total potassium (TK). Proteobacteria, Acidobacteriota , Bacteroidota , and Actinobacteriota dominated at the phylum level, with Haliangium and Sphingomonas as key genera. Alfalfa raised Bacteroidota levels, and all plants elevated Sphingomonas beyond CK. Soil bacterial communities differed significantly between control and cultivated soils. Proteobacteria correlated positively with soil potassium but negatively with phosphorus, nitrogen, and pH. Sphingomonas was positively associated with carbon and nitrogen, and negatively with phosphorus, pH, and potassium. Bacterial relationships varied: cooperative with MBB (bulk soil after cultivation of Chrysanthemum zawadskii Herbich ) and competitive with JB (bulk soil after cultivation of Pennisetum giganteum ). Soil pH and total nitrogen (TN) emerged as the primary chemical drivers that positively influenced soil organic carbon (SOC). Furthermore, bacterial community characteristics played crucial modulatory roles within this process: higher bacterial diversity significantly amplified the positive effects on SOC sequestration, whereas greater richness (indicative of shifts in community composition) was unexpectedly associated with a negative effect. These results collectively support our core hypothesis that vegetation restoration reshapes soil chemical properties, which in turn regulates bacterial community structure to influence carbon dynamics.