Correction of metabolic parameters and unit process performance data – Part II : Comparison of analytical approaches

Abstract

Oxygen consumption and other metabolic rates for commercial systems are typically measured under flow-through conditions and therefore are biased because of hydraulic lag. To evaluate potential correction approaches, three representative (SINE, STEP, MULTI-STEP) daily metabolic responses (Rta) were developed. Using the values of Rta, the resulting concentration of dissolved oxygen (Cti) on a minutely basis over the day was estimated from either a hydraulic mixing model or mass balance equations. Six approaches (STEADY, FRY, NORTHBY, NIIMI, SPLINE, and POLY) were used to estimate the metabolic oxygen consumption rate (Rt) for the 20-, 40-, and 60-minute periods. Three analytical approaches were tested (MEAN, POINT, and DETAILED).Based on all three test metabolic rates, the combination of the DETAILED analytical approach and the SPLINE correction was the most accurate. The DETAILED approach is based on estimation of the minutely oxygen consumption and computation of the average rate over a specific time period. The combination of DETAILED and NIIMI was more accurate for STEP and MULTI-STEP but was very inaccurate for the SINE response. It is not surprising that NIIMI is an excellent correction for STEP and MULTI-STEP as the derivation of this equation was based on a step change in metabolic rate. STEADY was the least accuracy for all metabolic response.For SINE, all of the equations can be used to estimate the daily average value (R¯daily). For STEP and MULTI-STEP, STEADY is less accurate. STEADY under-estimated the upper and lower peaking factors, but the other correction equations were very accurate.This same lag response can also bias other metabolic rates (carbon dioxide, ammonia, solids) and the computation of performance and efficiencies of unit processes such as biofilters and solids removal processes. The correction of unit process performance metrics may be further complicated by rapid changes in influent concentrations that were not considered in these correction approaches and the kinetic response of the process to changes in substrate concentrations.