Effects of different rice straw returning methods in karst paddy fields on soil bacterial community structure and...

The effective utilization of crop straw can contribute to sustainable agricultural development. However, how different straw return methods regulate soil fertility and rice yield via bacterial communities in karst paddy fields remains elusive. This field study investigated five straw return treatments [deep plowing (PD); rotary tillage with incorporation (RTM); field rapid composting, (FRC); no-till mulching (NT); and bioreactor (BR)] and a blank control CK (no straw return, fertilizer only) on soil physicochemical properties, bacterial community structure, and rice yield, combined with 16S rRNA sequencing technology. Results indicate the following: (1) all straw incorporation treatments significantly increased soil organic matter (SOM) and nutrient content (p < 0.05), NT and BR treatments increased soil organic matter (SOM) by 38.2 and 36.4%, respectively, compared to CK, while total nitrogen increased by 42.1 and 48.4% with NT; (2) although RTM treatment did not achieve the highest SOM accumulation, it yielded the highest rice yield of 30.37 kg/plot (a significant increase of 13.2% compared to CK), revealing that yield is jointly regulated by soil physicochemical properties and bacterial communities; (3) straw return treatments did not significantly affect bacterial α-diversity (intergroup differences in Shannon index and Chao1 index, p > 0.05), but significantly influenced β-diversity, symbiotic network structure, and community assembly processes. BR treatment formed a complex and stable microbial network structure, while NT exhibited a highly modular community structure (modularity = 0.66); (4) bacterial community assembly under straw return was dominated by deterministic processes, with homogenous selection accounting for 45 and 42% in NT and BR treatments, respectively, significantly higher than CK (28%, p < 0.05); (5) pathwise linear structural equation modeling (PLS-SEM) confirmed that TN (path coefficient 0.97, p < 0.001) and bacterial β-diversity (path coefficient 0.83, p < 0.001) were the most critical factors influencing rice yield. This study elucidates the mechanisms by which different straw return methods drive soil functions by reshaping bacterial community assembly and interaction networks. It provides theoretical support for optimizing straw return technologies in karst paddy fields, such as applying RTM for "yield-priority" scenarios and NT for "Rapid fertilization" scenarios.