Variable emission factors of CH₄ and N₂O from WWTPs: a model-based analysis on available data

Climate change and carbon emissions is increasingly becoming a global concern, and thus wastewater treatment plants (WWTPs) are also receiving extensive attention due to direct greenhouse gas (GHG) emissions of methane (CH₄) and nitrous oxide (N₂O). Although there have been many emission factors (EFs) of CH₄ and N₂O in literature, they are changeful due to different processes and boundaries, which limits their values for reference and comparison. With this study, in situ monitored CH4 and N2O data reported in literature were retrieved for recalculating their EFs. The average EFs are found to be 0.0011 g CH₄/g BOD_5-influent, and 0.0017 g N₂O-N/g TN_influent, based on the secondary treatment. Subsequently, the data were analyzed using multivariate linear regression and neural network. The results indicate that BOD₅ is the first factor affecting the EF of CH₄, revealing a negative correlation and that TN is the second factor affecting the EF of CH₄, but having a positive correlation. On the other hand, the neural network is a powerfully predictive and generalizable tool for EF_N2O. BOD₅ is negatively correlated with EF_N2O, and EF_N2O reaches to its maximum value at TN=35 mg/L. Overall, the direct GHG emission intensity is the lowest in the AAO and AO processes, or with the BOD₅/TN ratio between 2.5-4.9. Medium-sized WWTPs and the Oceania region exhibit the highest GHG emission intensity. With this study, an approximate approach is established to estimate the EFs of CH₄ and N₂O, which can facilitate to account the carbon footprint of WWTPs and also to aid in optimizing their operational schemes.