BioWin Modeling of a Three Reactor IFAS System

The City of Greensboro (City), North Carolina, owns and operates two water reclamation facilities, T.Z. Osborne (TZO) and North Buffalo Creek (NBC) that receive and treat wastewater from the City. TZO and NBC have average day maximum month rated capacities of 40 mgd and 16 mgd, respectively, and discharge to the Haw River Arm of Everett B. Jordan Reservoir (Jordan Lake). Based on the DENR TMDL strategy for Jordan Lake, the anticipated nitrogen and phosphorus loads for the TZO and NBC treatment plants equate to potential discharge permit limits of 5.29 and 0.66 mg/L as TN and TP, respectively, at average day maximum month flows. The City contracted with CDM/Hazen and Sawyer to evaluate nutrient removal alternatives for meeting future capacity requirements. The potential cost savings of IFAS relative to the 5-stage BNR based on conceptual level analysis prompted the one-year (April 2008 – April 2009) fullscale demonstration of the IFAS technology at TZO.With data collected over the demonstration period, calibration of the IFAS reactors with BioWin was performed for two different operation periods using the calibration options available. In the current released version of BioWin (File version 3.01.802), there are two primary methods of calibrating the media bioreactor element (i.e. IFAS) to field suspended solids and biofilm conditions: biofilm density factor or attachment/detachment rates. The biofilm density factor method calibrates to the average biomass for the three series of IFAS reactors and the attachment/detachment rates could be adjusted locally for match the individual biofilm biomass profiles found in the field. A beta version of BioWin was obtained from EnviroSim Associates Limited (www.envirosim.com), in which the biofilm density factor was made available as a local parameter allowing the modeler to mimic the field biomass profile.Using two different periods of operation (2.8 mgd 7/9 – 9/17/08 and 3.5 mgd 10/8 – 12/17/08), all three methods of calibration were performed to determine how well each method could be calibrated and how calibration to one operation period's conditions could be used to predict the results of a different operational condition. The paper discusses the calibrations and presents the results of the model simulations performed under each condition. Calibration was achieved for each period using each method of calibration, but those settings did not effectively predict the solids on the biomass and in the suspended phase under the other period loading conditions. Because each method calibrates specifically to the solids on the biomass and in the suspended phase and only involves adjustment of a single parameter, the complex influences on biofilm formation, maintenance, and process performance were not calibrated to field conditions. More study is needed to identify the optimum level of detail incorporated into the modeling of fixed film reactors specifically for engineering design.