Closing the Air Gap: Aeration Testing to Resolve Missing Demand and Unlock Savings at a 0.5 MGD Biological Nitrogen Removal Demonstration Plant

The Los Angeles County Sanitation Districts (LACSD) and the Metropolitan Water District of Southern California (MWD) have partnered to develop the Pure Water Southern California (Pure Water) program, which aims to produce up to 150 million gallons per day (MGD) of advanced treated potable reuse water from the A.K. Warren Water Resource Facility (Warren Facility). The Warren Facility is a 400-MGD capacity, non-nitrifying high-purity oxygen activated sludge (HPOAS) plant, producing secondary effluent with ammonia of approximately 40~50 mg N/L. As the product water needs to meet 2 to 4 mgN/L total inorganic nitrogen (TIN) targets for the intended applications, tertiary membrane bioreactor reactor (tMBR) is envisioned as a part of the nitrogen management strategy. To support process selection and full-scale design, MWD constructed a 0.6-MGD demonstration plant (Demo) at the Warren Facility. Between December 2023 and December 2025, the Demo was operated in nitrification-denitrification (NdN) tMBR mode with external carbon supplementation for denitrification (Fitzgerald et al., 2024). During this testing, the aeration air demand for NdN tMBR was consistently and significantly lower than the process design estimates, by approximately 20%. It was hypothesized that the tertiary biological nitrogen removal (BNR) operation resulted in improved oxygen transfer efficiency, thus lowering air demand. OBJECTIVE To evaluate whether improved aeration efficiency is behind the reduced air demand and to inform future full-scale aeration design, off-gas testing was conducted to assess aeration performance. METHODOLOGY Off-gas testing was conducted at the Demo in accordance with American Society of Civil Engineers 18-18 guidelines (ASCE, 2018) using an off-gas analyzer, a gas-capture hood connected to the analyzer, and a dissolved oxygen (DO) meter. The hood was floated on the aeration tank water surface to collect emitted gases, while off-gas and ambient air were alternately analyzed for oxygen concentration. DO was measured adjacent to the hood. These data were used to calculate oxygen transfer efficiency (OTE), alpha-fouling factor (αF), standard process-water OTE (αFSOTE), oxygen transfer rates (OTR), and oxygen uptake rates (OUR). Carollo's Inflatable Fast and Lightweight Off-gas Analysis Technology (I-FLOAT) platform was selected to accommodate the challenging Demo conditions, which included a 15-foot drop from the catwalk down to the water surface, limited available catwalk workspace, and instrumentation mounted along the handrails. The I-FLOAT's lightweight, inflatable gas-capture hood made it well suited to overcome these constraints. Testing was conducted in two campaigns: Phase 1 in November 2024 and Phase 2 in July 2025. Each phase included continuous off-gas testing for a minimum of three consecutive days in the aerobic zone (Zone #2), followed by three days in the swing zone (Zone #3) (Figures 1 and 2). At the start of Phase 2, a one-day sweep test was performed to assess spatial variabilities in aeration performance and support airflow balancing, with measurements collected across four locations spanning both the aerobic and swing zones. RESULTS Phase 1 (November 2024): Average OTE was 20.6% in Zone #2 and 9.0% in Zone #3, with corresponding αF values of 0.77 and 0.63. Average standard process-water OTE (αFSOTE) was 28.7% in Zone #2 and 16.2% in Zone #3. These results were likely influenced by imbalanced airflows between zones (0.29 vs. 1.53 scfm/diff) and, therefore, were considered non-representative of typical full-scale conditions. Phase 2 (July 2025): Airflows were rebalanced to approximate typical operating conditions (~1 scfm/diff), resulting in uniform distribution (0.83–0.87 scfm/diff). Under these conditions, OTE decreased to 9.8–10.6%, while αFSOTE (22.7-23.5) and αF (0.78–0.82) stabilized across the two zones, producing a representative spatial aeration profile. Across both phases, measured αF values ranged from 0.63 to 0.82 (Figure 3), exceeding typical mainstream conventional activated sludge values of 0.3–0.7 (Rosso et al., 2005). It is speculated the higher αF in tertiary NdN is due to HPOAS upstream of the Demo removing surfactants and inhibitory compounds. Additional details from the testing are summarized in Table 1. The improvement in oxygen transfer efficiency explains the low air demand observed during the Demo testing. Full-Scale Project Implications: Applying the validated αF of 0.8 in full-scale Pure Water aeration design, rather than the conservative 0.65 used in initial estimates, yields significant capital and O&M savings, including a 23% reduction in diffuser capital and a 24% reduction in required airflow (Table 2). For a 150-mgd tMBR NdN system, this translates to preliminary Class 5 aeration cost savings of $13-15 million. Industry Significance: To the authors' knowledge, there has been no to limited published reports of tertiary BNR aeration performance. This work addresses the data gap and shows optimization opportunities for such processes. CONCLUSION Off-gas testing confirmed that αF factors in an NdN tMBR system are substantially higher than in conventional activated sludge, demonstrating improved oxygen transfer efficiency as the cause of reduced air demand. Incorporating these validated αF values into full-scale design supports better optimized aeration strategies for tMBRs and similar tertiary BNR systems, yielding significant savings.