Microbial Adaptation to Low DO Biological Nutrient Removal

Introduction Biological nutrient removal (BNR) in wastewater treatment (WWT) requires aeration for biological processes, such as nitrification and phosphorus (P) removal. Low dissolved oxygen (DO) BNR is a means of maintaining effective WWT and operational capacity while reducing the amount of aeration (and other resources) required as the operating DO is decreased. It is also a critical enabler for more advanced shortcut nitrogen (N) removal technologies and for improving the reliability of biological P removal. However, this field of study lacks a fundamental understanding of microbial metabolic kinetics under low DO conditions. To address this knowledge gap, pilot work was performed to investigate microbial kinetic adaptation to low DO while also monitoring key operational parameters including N removal efficiency, simultaneous nitrification-denitrification (SND), enhanced biological P removal (EBPR), settling, and filamentous bacteria growth. Methods At Hampton Roads Sanitation District's Virginia Initiative Plant (Norfolk, VA), primary effluent was fed to a continuous-flow BNR pilot consisting of an anaerobic selector followed by 5 aerobic tanks-in-series at a rate of 5.5 m3/d. Throughout the study, the DO concentration in the aerobic zone was incrementally decreased from 2 mg/L to 0.2 mg/L. Both aerobic solids retention time (SRT) and temperature were maintained consistently at 8.5 days and 20oC, respectively. Twenty-four-hour composite samples from the primary and secondary effluent were analyzed daily for COD, NHx, NO2, NO3, TKN, OP, TP, VFA, TSS, VSS, SVI, and total alkalinity. Nutrient profiles across the BNR reactors were completed twice a week and filamentous bacteria count was obtained via microscopic examination once a week. Sensors within the BNR configuration included: DO, pH, TSS, and ammonia ion selective electrode. Declining DO (DDO) and SND-P uptake (SND/P) batch tests were used to investigate changes in microbial kinetics. The DDO test provides a measure of the respiration rate of nitrifiers as DO in the batch declines to zero. The DDO test yields an oxygen half-saturation coefficient (Ko) and maximum growth rate (µmax) for ammonia oxidizing bacteria (AOB), nitrite oxidizing bacteria (NOB), and endogenous respiration. The SND/P test measures substrate (N and P) uptake rates at various DO set points to obtain nitrifier Ko and µmax, denitrifier oxygen half-saturation inhibition coefficient (Kio), polyphosphate accumulating organism (PAO) Ko, and maximum P uptake rate. Results The pilot achieved stable nitrification at each DO set point which resulted in an average ammonia removal of 99 +/- 2%, additionally 92 +/- 9% of P was removed on average (Figure 1). Nitrifier adaptation to low DO was observed in both in-situ measurements and batch tests. Nearly 100% ammonia removal was achieved at each DO set point and the rate of ammonia uptake observed in-situ did not decrease meaningfully. The DDO test results indicate that the AOB Ko decreased as the pilot DO set point was reduced (Figure 2). The SNDP test also showed a decrease in the average estimated AOB Ko but the decrease was not significant (Figure 3). This demonstrates that nitrifiers were capable of achieving a higher affinity for oxygen throughout the adaptation period while the maximum ammonia removal rates remained relatively constant. OHOs also adapted to low DO conditions, with a bit of an unexpected result. A decrease in the Kio for denitrification was observed as the pilot transitioned to low DO (Figure 4). SND was expected to improve under low DO conditions, however, due to the apparent adaptation of heterotrophs, minimal additional SND was observed. The SND/P test revealed that, at all pilot operating DO set points, PAOs have an extremely low Ko. Repeatedly, at the lowest batch test DO set point of 0.1 mg/L, the P-uptake rate observed was at a maximum; thus the Ko of PAOs is expected to be less than 0.1 mg/L (Figure 5). Filamentous bacteria counts remained low and SVI30 was consistently ≤ 100 g/L throughout the transition to low DO (Figure 6). The results of the SVI30 and filament count measurements demonstrate that settleability can be maintained under low DO conditions, even in an A/O process with some degradable COD making its way to the aerobic zone. Conclusion The low DO BNR pilot successfully transitioned to low DO conditions while maintaining settleability, complete ammonia removal, and excellent EBPR. Throughout the adaptation period, four impactful observations have been made: 1. AOB Ko decreased, and in-situ ammonia removal rates remained consistent as operating DO was decreased from 2 mg/L to 0.3 mg/L, indicating nitrifier adaptation to low DO conditions. 2. OHOs similarly adapted to low DO conditions, showing a reduced Kio for denitrification as operating DO was decreased. Because of the decrease in Kio, no increase in SND was observed as the operating DO was decreased. 3. EBPR performed well across the range of operating DOs tested. Kinetic experiments show PAOs have an extremely low Ko for aerobic P uptake. 4. Settleability did not deteriorate as operating DO was decreased; SVI remained stable within an acceptable range and filament counts were relatively unchanged