Pesticide wastewater emerges as a typical refractory wastewater, characterized by complex composition and high toxicity, posing significant treatment challenges. Bacterial communities are responsible for biological treatment of refractory wastewater in full-scale pesticide wastewater treatment plants (PWWTPs), providing important implications for optimizing system performance and improving management strategies. However, the knowledge of their composition, diversity, function, assembly patterns, and biological interactions remains limited. Therefore, this study applied high-throughput sequencing, machine learning models, and statistical analysis to investigate key features of bacterial communities in eight PWWTPs. We found that Proteobacteria and Bacteroidota were the most abundant phyla, with Pseudomonas, Hyphomicrobium, Comamonas, and Thauera being dominant genera. Bacterial community distribution and diversity varied significantly among influents, sludges, and effluents, with sludges and effluents exhibiting higher diversity, richness, and evenness compared to influents. Deterministic processes primarily shaped the bacterial communities, accounting for 77.12%, 61.44%, and 64.05% of variation in influents, sludges, and effluents, respectively. Homogeneous selection explained 47.71%, 31.37%, and 31.37% of variation across these communities. Key modules (Module 1 in influents, Modules 3 and 4 in sludges, and Module 1 in effluents) were significantly associated with various metabolic and degradative functions (p < 0.05). Core taxa identified by Random Forest analysis were strongly linked to key metabolic and degradation functions, such as the metabolism of cofactors and vitamins, carbohydrates, and amino acids as well as the degradation of benzoate, aminobenzoate, nitrotoluene, chloroalkane, and chloroalkene. This study deepens our understanding of bacterial community dynamics and key features in pesticide wastewater treatment systems, offering scientific guidance for process optimization, efficiency improvement, and system stability assessment.
Read full abstract