Distinguishing AOB and Comammox Nitrifier Kinetics to Effectively Nitrify at Low Dissolved Oxygen Conditions

Background
Low dissolved oxygen (DO) operation (0.3 — 0.6 mg/L) enables biological nutrient removal (BNR) facilities to reduce energy and costs associated with aeration while achieving nitrogen and phosphorus removal. Although operator concerns about nitrification at low DO exist, our previous work shows that efficient nitrification is possible at low DO because low DO operation leads to biological selection of nitrification communities with high oxygen affinities1. Nitrification often occurs in a two-step process by autotrophic ammonia oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB). However, nitrifiers are a diverse community, with ammonia oxidizing archaea (AOA) and complete ammonia oxidizing bacteria (comammox, CMX) contributing to ammonia removal (Figure 1). Nitrifiers at low DO concentrations adapt with lower apparent half saturation constants (KDO) 1. It is unclear if these lower observed KDO values are due to selection of different nitrifiers or another mechanism. Initial qPCR results showed that the efficient nitrification at low DO coincides with an increase in the relative abundance of comammox. Yet current methods to determine the nitrification kinetics do not distinguish between active nitrifiers. The KDO of specific nitrifiers like comammox have yet to be reported 2. Additionally, investigating comammox kinetics is important for low DO operation, as comammox abundance may coincide with a decrease in nitrous oxide (N2O) emissions because comammox lacks nitric oxide reductases (NOR) that produce N2O through the nitrifier denitrification pathway3. In this study, the KDO for individual nitrifying communities were determined across low DO, mid DO, and high DO operating regimes, in addition to capturing the associated N2O generation. This will provide a better understanding of low DO nitrification, enabling more confidence in designing plants to operate at low DO. Objective: Determine KDO values for individual nitrifying organisms, including AOB and Comammox, to investigate if adaptations to low DO concentrations are driven by the selection of different nitrifiers.

Methods
A 200-L pilot-scale reactor was operated as an A2/O process, detailed in our previous work1. Three experimental phases were completed to facilitate varying nitrifier diversity based on changes to bulk DO: Phase 1 DO = 0.3 mg/L (Low DO), Phase 2 DO = 0.8 mg/L (Mid DO), Phase 3 DO = 2.0 mg/L (High DO). After three SRTs at the operating DO, steady state was established for the reactors. Reactor chemical performance was tracked weekly to evaluate nutrient removal, and DNA analysis was also performed weekly to evaluate microbial changes in the nitrifying communities.

The KDO and maximum ammonia removal rates were determined for ammonia oxidation, while monitoring N2O production with Unisense N2O probes. Mixed liquor samples from the pilot plant were controlled in batch tests at a specified DO setpoint (i.e., 0.2, 0.5, 0.7 and 2 mg O2/L). At the beginning of the test, 30 mg/L of NH3-N was added, and all nitrogen species were measured for 1-2 hours. Inhibitory chemicals were added to isolate nitrifier communities: Chlorate inhibits comammox, and 1-octyne inhibits AOB (Table 1) 6,7,8.9. Initial OUR testing determined the dosage of inhibitory chemicals to add.

Results
Nitrification was well maintained in all phases, with average effluent ammonia concentrations of 0.37 ± 0.71 mg/L at low DO, 0.05 ± 0.02 mg/L at mid DO, and 0.19 ± 0.11 mg/L at high DO. Low liquid phase N2O was usually observed in the reactor but peaks in N2O were observed when DO was outside of the desired operating band for an extended period. Monod curves were fitted for each experiment within each reactor operating phase to determine the KDO and the maximum activity rate. When considering the entire nitrifying community, or when no biomass was inhibited, lower KDO were observed at lower bulk DO conditions (Figure 2). Specifically, for the same test when all microbial community contributions to nitrification are considered, biomass acclimated to low DO had a KDO of 0.25 mg/L, compared to a high DO KDO of 1.21 mg/L, confirming the expectation that nitrifiers in lower DO systems adapt with lower observed KDO values (Table 2).

Initial qPCR results showed that the relative abundance of comammox was the highest at low DO (10.9%), but AOB and canonical NOB remained as the dominant nitrifiers in the system. All qPCR results will be included in the conference presentation. Inhibition testing confirmed lower KDO values for comammox at low and mid DO, demonstrating higher oxygen affinity and enhanced nitrification (Figure 3). Additionally, for all DO operation modes, the CMX KDO is lower than the overall KDO and the nitritation KDO, as well as the maximum growth rate. The CMX KDO for low DO and mid DO sludge ranged from 0.15 to 0.23 mg/L and since the comammox abundance appears to increase with the reduction in DO, the presence of comammox seems to help explain the overall KDO reduction with the low DO sludge1.

Conclusion and Significance Comammox KDO is lower than the AOB KDO at Low and Mid DO conditions, providing a selective advantage for comammox at low DO conditions and enabling nitrification. AOB may not adapt to low DO; however, due to the comammox selection at low DO, nitrification is possible.