Logo image
Predictive modelling for carbon dioxide equivalent of greenhouse gas from wastewater treatment works : a case study of Northern wastewater plant in Durban
Thesis   Open access

Predictive modelling for carbon dioxide equivalent of greenhouse gas from wastewater treatment works : a case study of Northern wastewater plant in Durban

Kennedy Pahla
M.Eng., University of Johannesburg
2025
Handle:
https://hdl.handle.net/10210/517209

Abstract

Greenhouse gas mitigation|-South Africa-Mathematical models Environmental aspects-South Africa Carbon dioxide mitigation
The current estimation of greenhouse gas emissions in municipal wastewater treatment plants (WWTPs), based on the Intergovernmental Panel on Climate Change (IPCC) method, faces limitations in formulating targeted emission reduction strategies at the individual unit level. This research endeavours to enhance predictive accuracy for carbon dioxide equivalent (๐ถ๐‘‚!") emissions - a metric used to express the combined impact of multiple greenhouse gases, including methane (๐ถ๐ป#), nitrous oxide (๐‘!๐‘‚) and carbon dioxide (๐ถ๐‘‚!), in terms of their warming potential relative to carbon dioxide (๐ถ๐‘‚!) over a specific time horizon. The study focuses on Northern Wastewater Treatment Plant (WWTP) in Durban by leveraging advanced predictive tools, specifically the GPS-X and Bridle models. These models are employed in conjunction with measurements conducted following IPCC guidelines. Significant variations were observed in the magnitude of methane, carbon dioxide and nitrous oxide emitted in response to variations in component units' operating parameters and influent characteristics like influent flow rate, COD and temperature. The examined treatment system used a primary clarifier, anoxic tank, aerobic digester, secondary clarifier, thickener, belt presser and anaerobic digester along with chemical disinfection unit. The IPCC Northern wastewater treatment baseline estimate emission for the examined week was 611.29 kg๐ถ๐‘‚!". The Bridle model estimated an average daily emission of 8309.55 kg๐ถ๐‘‚!" with the sludge treatment process contributing between 68.70% and 70.8% of the total emissions. The average daily prediction differs from the actual IPCC emission with over 7698.26 kg๐ถ๐‘‚!". The GPS-X software simulation indicated an average daily emission of 2121.58 kg๐ถ๐‘‚!" with an anaerobic digester as the main emission component unit. The GPS-X daily average prediction was 247.06 % more than the actual IPCC daily average emission. The GPS-X model showed a stronger correlation with observed emissions (Rยฒ = 70.5%) compared to the Bridle model (Rยฒ = 47.95%). The 70.5% correlation indicates that the GPS-X model has a higher explanatory power, making it more effective in predicting the variability observed in the actual IPCC emission values than the Bridle model. This demonstrates that the GPS-X model offers a more accurate representation of emission patterns. The study examined the temporal dynamics of VI emissions, with a particular focus on daily variations over the seven-day data simulation period. Greenhouse gas production and emissions are driven by non-linear processes, which are influenced by daily fluctuations and varying operational conditions. GPS-X operates as a mechanistic model, simulating emissions on a component-specific unit basis by incorporating biochemical reaction, kinetics and mass balance principles. This assumes that emissions are directly influenced by microbial activity, reactor design, and operational conditions. The Bridle model adopts an empirical process-based approach, estimating emissions based on predefined treatment stages and historical emission factors, assuming a more generalized relationship between treatment processes and greenhouse gas outputs. This characteristic renders GPS-X a superior predictive model, particularly well-suited for formulating technical strategies aimed at emission control. Furthermore, the research explored the factors contributing to the variability emissions, with a specific emphasis on the influence of operational parameters, environmental conditions, and the unique characteristics of the influent. The sensitivity analysis revealed that the system's performance is highly responsive to changes in influent flow rate, with a sensitivity of 10%. A reduction in influent had a proportional impact on the emissions, and the system responded in a somewhat linear manner to changes in this parameter. The study quantifies direct emissions from wastewater treatment facilities and highlights the potential of biogas utilization to offset these emissions. By capturing and converting biogas from sludge treatment into renewable energy, the findings demonstrate how this process can substantially reduce greenhouse gas emissions. The analysis explores biogas systems as a sustainable alternative to traditional energy practices, with the capacity to meet internal energy needs and contribute to broader emission reduction goals. Additionally, the study identifies key operational factors influencing emission variability, such as biogas capture efficiency, energy recovery system scale, and sludge treatment management. Optimizing these factors enhances emission reductions. The findings provide data-driven insights for wastewater treatment facilities to implement targeted mitigation strategies, promoting sustainability within the sector. VI The study highlights the critical role of local environmental conditions in Durban, such as temperature, humidity, and precipitation patterns, which can significantly influence the efficiency of biogas production and emissions at wastewater treatment facilities. These unique climatic factors can lead to variations in the microbial processes involved in sludge treatment, potentially resulting in higher or lower emissions compared to regions with more temperate climates. Understanding these local environmental dynamics is essential for developing region-specific strategies to optimize biogas recovery and minimize emissions, ensuring more accurate and effective mitigation measures tailored to Durban's conditions. Future research could focus on refining predictive models using real-time data, exploring advanced treatment technologies for further emission reductions, and conducting comparative studies across diverse climatic regions to enhance the frameworkโ€™s applicability.
pdf
PAHLA_K_2211667922.56 MBDownloadView
Open Access

Metrics

15 File views/ downloads
37 Record Views

Details

Logo image