Practical Control Strategies for Reducing N2O Emissions in Wastewater Treatment Plants | Praktiska kontrollstrategier för minskning av lustgasutsläpp från avloppsreningsverk

Wastewater treatment plants (WWTPs) contribute significantly, approximately 3%, of global nitrous oxide (N2O) emissions and must mitigate their environmental impact by mapping their carbon footprints, addressing emission sources, and exploring resource reuse, such as nutrients, water, and energy recovery. The purpose of this study was to contribute to the development of effective N2O mitigation practices in WWTPs by summarizing encountered mitigation strategies and implementation challenges. This thesis combines a literature review and data analysis with data-driven modeling to achieve this. A partial least squares, random forest, and supervised neural network model were tested for their suitability. Suitability was determined based on accuracy and interpretability for predicting N2O emissions. The research identified aeration control, feeding control, microbial community selection, and alterations to the process as mitigation strategies. The data analysis showed how multiple variables must be measured to explain the variations in emission but also some correlations between the variables. Regarding the models, the neural network model had the most accurate predictions but was not interpretable. The random forest model was deemed the best for analyzing and drawing conclusions about operational conditions and N2O emission, along with finding a potential mitigation strategy. The partial least squared model was deemed the worst due to its performance. However, could be suitable when a simple overview of the relationship between measured variables and N2O is needed. Overall, more rigorous on-site measurements, which also consider practical challenges, are needed to address the complexity of reducing the carbon footprint while maintaining a high treatment quality. Additionally, the carbon footprint needs to be accounted for when deciding on a mitigation strategy to ensure that the strategy does not conflict with minimizing the carbon footprint. There is still a need for better and simpler models to give insight into the underlying mechanisms and dominating pathways in the process, which can be used to calculate the carbon footprint of the whole WWTP, and additional research should be conducted. Specific long-term goals, financial aspects, and data availability should be used as basis when deciding on a mitigation strategy to ensure it is tailored to the WWTP.