Abstract
Relevance
As part of the collective efforts to achieve 'net-zero' emission goals by 2050, leading water utilities around the globe have started including quantitative sustainability indices to be evaluated when major capital improvements and operational modifications are planned for water infrastructures [1]. The emission of greenhouse gases (GHG), mainly CH4 and N2O, is one important sustainability index that needs to be quantified. In addition, the annual GHG emissions from water resource recovery facilities (WRRFs) with emissions exceeding their relevant thresholds, are required to be reported to regulatory authorities.
Wastewater treatment contributes 2 — 9% of global GHG emissions [2, 3]. The noticeable uncertainty in emissions estimation is due in large part to the lack of direct measurement data. Several methods have recently been attempted for directly measuring the fugitive GHG emissions, such as drone flux method, mobile tracer gas dispersion method, and flux chamber method. However, due to the 'short duration' and 'point' nature of the selected approaches, only ‘snapshot’ and 'spot' emissions were captured. Therefore estimating the annual emissions of the entire site necessitates extrapolation of the limited data , which introduces substantial uncertainty to emissions quantification. Hence, the suitability of specific methods for quantifying the emissions from WRRFs need to be evaluated in terms of the methods' accuracy, detection limits, spatial and temporal coverage and representativeness.
Methodology
This study, beginning in June 2021, is one of the very few studies, and the first multi-year investigation, in North America that has performed a comprehensive characterization of GHG emissions from WRRFs. The overall purpose of this study was to develop a robust approach capable of quantifying the fugitive emissions at WRRFs. To achieve this purpose, a number of monitoring technologies have been evaluated on-site and the results have been cross validated. Emission data for CH4, N2O, and other gaseous species such as NH3 and CO2, have been obtained at different temporal and spatial resolutions.
This paper reports a subset of the data, focusing on the emission characteristics, in particular, the temporal and spatial variations of GHG emissions at WRRFs. Emission characterization is the first and most critical step in emissions monitoring, guiding the selection of suitable monitoring technologies, the design of the field investigation, and the analysis of the monitoring data. The reason most previous studies were unable to creating a sound baseline of the GHG emissions at the WRRFs was mainly because the selected methods were insufficient in capturing the temporal and spatial heterogeneities of the emissions. The primary study site is a biological nutrient removal WRRF in Canada, with an average daily flow of 348 million litre per day. A hybrid monitoring approach, utilizing various technologies, was assessed, improved, and utilized for directly monitoring GHG emissions at a wide range of temporal (seconds, diurnal, daily, monthly, annual) and spatial (facility, process, and site) scales. The technologies that were determined to be suitable for this application consisted of open-path and closed-path optical sensing, chemiresistive point sensors, meteorological station, sonic anemometer, satellite imagery, and reverse dispersion modeling (Backward Lagrange Stochastic model).
Results
Figure 1 reveals the spatial and temporal variations of CH4 and N2O within the plant. A clear seasonal pattern of spatial distribution was observed for both CH4 and N2O. Notably, CH4 plumes were not detected at the anaerobic digesters or cogeneration facility, but were instead observed in the aeras of primary sludge and secondary treatment. N2O plumes were observed near the secondary treatment bioreactors that are not designed for denitrification. The N2O plumes were most prominent during peak flow times.
Figure 2 illustrates the temporal variation of CH4 column concentrations from 2019 to 2023, showing a strong correlation with the wastewater temperature.
Figure 3 shows significant difference between the daytime and nighttime CH4 concentrations, implying that the emissions measured during the day cannot be used to represent the overall emissions.
Figure 4 shows the CH4 concentrations continuously measured with two different CH4 chemiresistive sensors. A highly dynamic emission pattern was observed, implying that snapshot measurement would cause bias when extrapolating the results for estimating the total emission over a longer period. Also, significant difference of the data quality between the two sensors was observed.
This study demonstrates that single-point or snapshot measurements are insufficient for accurate quantification, highlighting the need for developing hybrid monitoring approaches that combine multiple technologies for capturing the spatial and temporal variations of GHG emissions. A plausible approach is to combine several methods for performing multi-point and continuous measurements. In addition, it is critical to properly validate and rigorously calibrate the monitoring methods at ambient conditions for minimizing uncertainties. These findings provide a foundation for improving emissions monitoring practices in the water sector, ultimately supporting more accurate reporting and effective strategies for achieving the net-zero goals.
This paper was presented at WEFTEC 2025, held September 27-October 1, 2025 in Chicago, Illinois.
Author(s)Du, Ke, Mehrdad, Seyed Mostafa, Zhang, Bo, Regmi, Pusker
Author(s)K. Du1, S. Mehrdad1, B. Zhang2, P. Regmi2
Author affiliation(s)University of Calgary1, Stantec Inc.2
SourceProceedings of the Water Environment Federation
Document typeConference Paper
Print publication date Oct 2025
DOI10.2175/193864718825159964
Volume / Issue
Content sourceWEFTEC
Copyright2025
Word count16