Degradation dynamics of saflufenacil and tiafenacil in agricultural soils and their impacts on soil microbial networks and carbon-nitrogen cycling.

Herbicides are vital for sustaining global crop production; however, the limited understanding of their residual effects on soil microbial interactions and ecosystem processes leads to an underestimation of their long-term ecological risks to agricultural soil health. This study investigated the degradation dynamics of two novel PPO-inhibiting herbicides, saflufenacil and tiafenacil, and their impacts on microbial networks and carbon (C) and nitrogen (N) cycling across five contrasting agricultural soils. Both herbicides dissipated rapidly, with half-lives ranging from 0.8 to 4.1 days, and degradation rates were regulated by soil physicochemical properties, including cation exchange capacity, clay content, pH, and total nitrogen. Despite rapid dissipation, transient herbicide residues altered microbial network architecture. Bacterial networks exhibited increased modularity (9.9-21.2%), whereas fungal networks showed lower edge numbers (7.1-35.0%), average degree (5.7-34.8%), graph density (4.3-37.2%), and clustering coefficients (1.4-9.6%). Functionally, herbicide exposure increased the relative abundance of genes associated with labile C degradation (e.g., starch and pectin metabolism) and denitrification, while suppressing genes involved in recalcitrant C degradation (e.g., cellulose and chitin breakdown) and dissimilatory nitrate reduction. Potential herbicide-degrading taxa (e.g., Blastococcus) and a degradation-related gene (e.g., K11260) were identified, suggesting adaptive microbial responses. Collectively, these findings demonstrate that, for the two PPO-inhibiting herbicides tested here, even short-lived herbicide residues can restructure microbial interaction networks and redirect soil C and N transformation pathways. The results highlight the importance of soil-specific risk assessment and provide mechanistic insights to inform sustainable herbicide management within agroecosystems.