Unraveling the role of salinity in anammox-based nitrogen removal processes via data analysis from the literature and experimental validation

Abstract

As an energy-efficient nitrogen removal process, anaerobic ammonia oxidation (anammox) is now widely used to remove nitrogen in various wastewaters, some of which might contain high salinity. Yet, there are many uncertainties in the treatment of saline wastewater by anammox. In this study, the effect of salinity on the anammox nitrogen removal process was revealed by data analysis from the literature and experimental validation. First, the nitrogen removal performance of freshwater anammox bacteria (FAB) and marine anammox bacteria (MAB) in saline wastewater was different, and MAB can adapt to higher salinity (7.5%). Furthermore, the Edwards model can best describe the effect of salinity on the anammox process, due to its high fitted R² and reasonable modeled maximum nitrogen removal rate (NRR) ranging between 0.079 and 4.189 kg m⁻³ d⁻¹. In addition, the artificial neural network (ANN) model can simulate and accurately predict the NRR of anammox under different salinity conditions. The importance of environmental factors for the ANN model was lower than that of the effluent parameters. The performance of these models could be well verified by batch tests. Spearman correlation analysis showed that due to the coexistence of a stimulation period and inhibition period, the relationship between the NRR and salinity was not significant under both short-term and long-term salinity stress. Regarding functional bacteria for the nitrogen removal process, Kuenenia and Nitrosomonas had better salt tolerance than Brocadia and Nitrosospira, respectively. Overall, the effect of salinity on anammox-based nitrogen removal processes was elucidated by model fitting and experimental validation.