Biochar and green manure have been widely applied in agricultural production and are important means to achieve sustainable agriculture. However, there is limited research systematically and comprehensively exploring the response of soil microbiota and the changes in soil metabolomics after the addition of two different carbon source amendments to the soil, and the differential mechanisms of soil metabolomics between them remain unclear. In this study, a long-term field experiment (initiated in 2019) was conducted to investigate the effects of biochar and green manure application on soil nutrients and soil functions driven by soil microbes. Compared to the pure fertilizer treatment, biochar increased soil total carbon by 14.54% to 27.04% and soil available potassium by 4.67% to 27.46%. Ryegrass significantly increased soil available phosphorus and organic matter. Under different fertilization regimes, the ecological niches of soil microbes changed significantly. Network analysis revealed that long-term ryegrass returning reduced the complexity of soil microbial networks. Ryegrass and biochar increased dispersal limitation in fungal assemblages (reaching 93.33% and 86.67%, respectively), with biochar particularly enhancing variable selection in bacterial assemblages (accounting for 53.33%). Variation partitioning analysis based on redundancy analysis indicated that humic substances had the highest explanatory power for microbial community variation, with humic substances explaining 38.49% of bacteria and 52.19% of fungi variation. The ryegrass treatment mainly changed the abundance of carbohydrates (CH), amines (AM), c (AH), and lipids (LP), while the BC treatment mainly altered the abundance of organic acids (AC), amines (AM), and carbohydrates (CH). Meanwhile, both treatments significantly reduced the bisphenol A, one of the soil pollutants. Ryegrass incorporation significantly increased the abundance of genes related to soil C, N, P, and S cycling, especially genes involved in carbon decomposition, while biochar significantly enhanced the abundance of nitrogen fixation genes nifH and Hao in soil. Random forest model results indicated that carbohydrates, alcohols, aromatics (AR), and ester (ES) were the main categories of metabolites in soil influenced by differential microbes, and Finegoldia served as a common important metabolic driving species. In summary, this study reveals the processes of soil function, microbial community succession, and metabolism driven by ryegrass and biochar, providing important insights for optimizing soil management and improving soil quality.Graphical
Read full abstract