Development of Kinetic Parameterization Methods for Nitrifying Bacteria using Respirometry
Understanding how nitrifiers react when exposed to low DO conditions could provide a greater understanding of low DO operations in full-scale biological wastewater treatment. Previous methods to observe nitrifier oxygen kinetics do exist in literature, however they are inefficient and labor intensive. Other more efficient methods require the use of selective inhibitors, which alter the characteristics of the biomass. This study developed a time and labor efficient respirometric method to distinctly measure oxygen half-saturation coefficients for both ammonia oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB) without the use of selective inhibitors. By eliminating the use of inhibitory substances, representative biomass characteristics were maintained throughout the tests. The developed method, called the declining DO method, consisted of using a high-speed dissolved oxygen (DO) probe to measure relative oxygen uptake rates (OUR) within a batch reactor when varying substrates (ammonia and nitrite) were present in excess within the system. A forward model was developed based on Monod kinetics to simultaneously fit Monod curves to the experimental OUR data. These curves were fit by solving for optimum oxygen kinetic parameters representing endogenous respiration, NOB, and AOB. An inverse model using Markov chain Monte Carlo analysis was applied to the results found in the forward model to provide statistical validation of the proposed respirometric method. A separate method, called the substrate utilization rate test, was conducted in parallel with the declining DO tests to compare and verify oxygen half-saturation coefficient results. Parallel tests were conducted using biomass samples from three different Hampton Roads Sanitation District (HRSD) full-scale facilities. Operating conditions between the three HRSD facilities were considered when performing parallel testing, including averages for DO, solids retention time (SRT), and floc size. Average floc size was found to have a significant effect on the observed oxygen half-saturation values. Observed trends for the KO values estimated using the two methods remained consistent throughout all tests, where KO,NOB was always lower than KO,AOB. The comparison of the two methods highlighted some faults associated with the substrate utilization rate test, which is commonly used in literature to observe nitrifier oxygen kinetics. The declining DO method appeared to be more resistant to potential experimental error and required less than half the time compared to the substrate utilization rate test. The development of the declining DO method without the use of selective inhibitors provided a more time and labor efficient technique for estimating apparent KO values for NOB and AOB without sacrificing biomass characteristics representative of the full-scale treatment process. Biomass samples collected from variable treatment process conditions yielded consistent parallel test results, providing further evidence that the proposed declining DO method can be a robust and reliable technique for distinctly measuring apparent oxygen half-saturation values for NOB and AOB.