Abstract

In light of the historic Paris Agreement at the UN Climate Change Conference aimed at combating global warming, there has been increased momentum to quantify and mitigate greenhouse gas (GHG) emissions from Water Resources Recovery Facilities (WRRFs). However, the current methodologies for estimating GHG emissions from WRRFs are fraught with high degrees of uncertainty. To address this, a range of modelling approaches has been employed to estimate GHG emissions, specifically nitrous oxide (N2O) and methane (CH4), and to optimize and mitigate such emissions through linking operational processes. This article conducts a thorough and critical examination of GHG emissions modelling efforts in WRRFs, covering mechanistic, data-driven, and hybrid models for N2O and CH4, alongside empirical, steady-state, and dynamic plant-wide models. It emphasizes the applicability and limitations of these methods in full-scale applications, highlighting the calibration complexities of mechanistic models and the limited explainability of data-driven tools. The review also discusses innovative emerging approaches, such as hybrid modelling and knowledge-based AI, and stresses the necessity for novel, model-aided strategies to quantify and monitor fugitive methane emissions effectively. By elucidating knowledge gaps, addressing literature discrepancies, and reviewing diverse modelling methodologies, this article significantly enhances the current understanding of GHG modelling in WRRFs, paving the way for more sustainable and environmentally responsible wastewater management practices.