Biological phosphorus removal under cold-climate using Sequencing Batch Moving Bed Biofilm Reactor

Sammanfattning

Biological phosphorus removal in cold-climate regions is historically challenged by suppressed kinetic rates and severe biomass washout during spring snowmelt events. This study evaluates the performance, resilience, and kinetic limitations of a pilot-scale Sequencing Batch Moving Bed Biofilm Reactor (SB-MBBR) for Enhanced Biological Phosphorus Removal (EBPR) un-der low temperatures (8.3–9.6°C) at the Fillan Wastewater Treatment Plant. The system's stabil-ity and treatment capacities were assessed across variable seasonal flows and two carbon sup-plementation strategies: a synthetic carbon source (SCS) and an on-site fermented internal car-bon source (ICS). Cycle tracking showed SBR2 achieved effective EBPR (anaerobic release of 2.45 mg/L, aerobic uptake to 0.155 mg/L), while identical reactor SBR1 failed due to differences in the initial start-up operational strategy. Ex-situ testing confirmed biological activity was localized on the carriers (6.62 mgP/gVSS/h) rather than the suspended sludge (1.85 mgP/gVSS/h). During the snow-melt the system transitioned to an ICS. Following Monod kinetics, the ICS increased volumetric phosphorus removal from 1.77 to 3.32 gP/m3day, but higher background nutrient loading caused effluent accumulation due to fixed phase timers. Under the dual stress of snowmelt and ICS, surface-specific uptake rates temporarily dropped from 0.05 to 0.033 gP/m2day. This ki-netic suppression was driven by a 152% increase in biofilm mass, where ordinary heterotrophs temporarily covered the PAOs, causing diffusion limitations. However, kinetics fully recovered to 0.05 gP/m2day by May, demonstrating that the biofilm is a dynamic system capable of adap-tation. Cross-validated PLS modelling confirmed the system was bottlenecked by aerobic dis-solved oxygen and influent P load rather than cold thermodynamics. Benchmarking showed the SB-MBBR maintained high specific kinetics (0.0146 gP/gVSSday) during snowmelt while the parallel IFAS line collapsed to 0.0005 gP/gVSSday due to sludge washout. However, to con-sistently meet the 0.3 mg/L effluent target requires extending operational phase times.