Abstract
Excess nitrogen in water bodies can lead to eutrophication, posing a significant threat to aquatic ecosystems. Therefore, effective treatment of nitrogenous wastewater and the removal of nitrogen compounds from water bodies are essential for improving and maintaining water quality. In this study, the immobilized denitrifying bacterium Alcaligenes faecalis was employed to enhance the nitrogen removal efficiency of an ecological floating island (EFI). The research also aimed to determine the optimal operating conditions and assess the impact of different influent nitrogen concentrations on the treatment performance of the EFI. The results identified the optimal operating conditions for the integrated system as a C/N ratio of 16 and a DO concentration of 2 ~ 3 mg/L. Under this optimal condition, the system achieved a COD removal efficiency of 70.00 to 84.69% and a 100% total nitrogen (TN) removal efficiency across the entire tested range of influent TN concentrations (7.5 ~ 30 mg/L). Key microbial findings revealed that the highest microbial richness in the system was observed under the above optimal conditions; at the phylum level, the dominant microbial taxa were unclassified_k_norank_d_Bacteria and Proteobacteria, and at the order level, unclassified_p_Proteobacteria and unclassified_k_norank_d_Bacteria were the predominant groups, with unclassified_k_norank_d_Bacteria being the core functional taxa for nitrogen removal. Moreover, plant group 2 (Canna indica L. + Thalia dealbata Fraser + Vallisneria natans) and plant group 3 (Canna indica L. + Myriophyllum verticillatum L. + Vallisneria natans) exhibited significantly higher chlorophyll content and catalase activity than the other three plant groups, demonstrating that these two combinations are the optimal plant collocations with superior stress resistance and adaptability to the nitrogen-polluted wastewater environment. The study concluded that immobilizing denitrifying bacteria is an effective strategy for enhancing the water treatment performance of EFIs.