The co-utilization of green manure (GM) and rice straw (RS) in paddy fields has been widely applied as an effective practice in southern China. However, its effects on soil aggregate and soil organic carbon (SOC) stability remain unclear. In the present study, the effect of GM, RS, and co-utilization of GM and RS on particle size distribution of soil aggregates and SOC density fractions were measured in a field experiment. The experiment included six treatments, i.e., winter fallow (WF) without RS return (Ctrl), WF with 50% RS return (1/2RS), WF with 100% RS return (RS), GM without RS return (GM), GM with 50% RS return (GM1/2RS) and GM with 100% RS return (GMRS). The results showed that the proportion of small macro-aggregates (0.25–2 mm) and the mean weight diameter (MWD) of aggregates in the GMRS treatment was greater (by 18.9 and 3.41%, respectively) than in the RS treatment, while the proportion of silt+clay particles (<0.053 mm) was lower (by 14.4%). The concentration of SOC in microaggregates (0.053–0.25 mm) and silt+clay particles was higher in the GMRS treatment than in GM and RS treatments individually. The concentration and proportion of free light organic carbon (fLOC) in aggregates of various particle sizes and bulk soil was greater in the GMRS treatment than the RS treatment, whereas the concentration and proportion of mineral-associated organic carbon in small macroaggregates, microaggregates, and bulk was lower in the GMRS treatment than in the RS treatment. The proportion of intra-aggregate particulate organic carbon (iPOC) was greater in the GMRS treatment than in GM treatment. The GMRS treatment had strong positive effects on iPOC in small macroaggregates, suggesting that SOC was transferred from fLOC to iPOC. In conclusion, co-utilizing green manure and rice straw cultivated the SOC pool by increasing the concentration of fLOC and improved soil carbon stability by promoting the sequestration of organic carbon in iPOC as a form of physical protection.

Green manure–crop rotation is a sustainable approach to protect crops against diseases and improve yield.  However, the mechanism by which green manuring manipulates the crop-associated microbial community remains to be elucidated.  In this study, we explored the horizontal processes of bacterial communities in different compartments of the soil–root interface (bulk soil, rhizosphere soil, rhizoplane and endosphere) of tobacco by performing a field experiment including four rotation practices, namely, tobacco rotated with smooth vetch, ryegrass, radish, and winter fallow (without green manure).  Results showed that the co-occurrence networks constructed by adjacent compartments of the soil–root interface with green manuring had more edges than without green manuring, indicating  dramatic microbial interactions.  Green manuring increased the dispersal-niche continuum index between bulk soil and other compartments, indicating that it facilitated the horizontal dispersal of microbes.  For the different green manuring practices, the neutral community model explained 24.6–27.6% of detection frequency for bacteria, and at least one compartment under each practice had a normalized stochasticity ratio higher than the 50% boundary point, suggesting that the deterministic and stochastic processes jointly shaped the tobacco microbiome.  In conclusion, green manuring generally facilitates bacterial community dispersal across different compartments and enhances potential interactions among adjacent compartments.  This study provides empirical evidence for understanding the microbiome assembly under green manure–crop rotation.