Alexis Awaitey, Aalto University, alexis.awaitey@aalto.fi
Wastewater treatment plants play a crucial role in protecting public health and the environment by removing pollutants to enable safe discharge of effluents to rivers, lakes, and coastal waters. In the context of the Baltic Sea, this is of great importance in reducing eutrophication, for example, which is currently an environmental challenge. However, these treatment processes involve complex biological systems that consume energy and produce greenhouse gases. As cities work toward climate targets, understanding and reducing these emissions has become an increasingly important challenge.
In my doctoral research, I am investigating how digital twin models could help wastewater treatment plants operate more sustainably. My focus is on reducing greenhouse gas emissions, particularly nitrous oxide (N₂O), while maintaining reliable treatment performance. Nitrous oxide is an especially important gas in this context due to its very high global warming potential compared with carbon dioxide. Although wastewater treatment plants are not usually thought of as major sources of greenhouse gases, emissions of nitrous oxide from biological nitrogen removal can significantly affect the overall climate footprint of a facility.
The goal of my research is to explore how digital twins could help treatment plants move from a reactive management approach to a more proactive one and strike an appropriate balance between operational cost and carbon footprint.
Digital Twins and Proactive Operation
A digital twin is essentially a virtual representation of a real system. In wastewater treatment, this means combining operational data from the plant with process models that describe the biological and physical processes occurring in the reactors. The main distinction from a conventional model is that digital twins have a near-real-time data connection with the physical plant.
The digital twin receives data from the real plant, such as flow rates, nutrient concentrations, and operational settings. Using this information, the model can simulate how the system is likely to behave under different conditions. This creates a virtual environment where operators and engineers can test ideas without affecting the real facility. For example, different operational strategies can be simulated to see how they might influence effluent quality, energy consumption, operating costs, and greenhouse gas emissions. This allows the plant to explore possible adjustments before implementing them in practice.
Wastewater treatment plants are traditionally managed in a reactive way. Operators monitor key parameters and respond when conditions start to deviate from their targets. This approach works well for maintaining regulatory compliance, but it does not always allow operators to anticipate changes or evaluate multiple objectives at the same time. In my research, the idea is to use the digital twin to perform short-term forecasting and scenario analysis. If incoming wastewater conditions are expected to change, the model can test different operational strategies in advance. Each scenario can be evaluated in terms of treatment performance, operational cost, and greenhouse gas emissions.
The goal is to help operators identify strategies that provide a good balance between these competing objectives. In practice, this could support decisions such as adjusting aeration intensity or changing process configurations in response to expected loading conditions.
Modelling Nitrous Oxide Emissions
One of the central challenges in this work is accurately modelling nitrous oxide emissions. Over the past decades, researchers have made significant progress in identifying the main pathways through which nitrous oxide is produced during biological nitrogen removal. These pathways are linked to the microbial processes of nitrification and denitrification. Under certain conditions, such as low oxygen concentrations or elevated nitrite levels, these processes can produce nitrous oxide as a byproduct.
Even though the main mechanisms are well documented in the scientific literature, predicting emissions at full scale remains difficult. Wastewater treatment relies on complex microbial communities that respond dynamically to environmental conditions and operational changes. Small variations in oxygen availability, substrate concentrations, or microbial activity can influence how much nitrous oxide is produced. Because of this complexity, translating theoretical knowledge into reliable predictive models is still an active area of research.
Looking ahead
The idea of using digital twins in wastewater treatment is still developing, but it offers an interesting opportunity to bridge scientific research and plant operations. By combining process models with real operational data, digital twins can provide a platform for exploring how treatment plants behave under different conditions. In the long term, tools like these could help wastewater utilities make more informed decisions that consider both environmental impact and operational efficiency. For cities that are working toward climate neutrality, better understanding and managing emissions from wastewater treatment will be an important step.
My research is one small part of that broader effort. By improving how we model and analyse treatment processes, I hope to contribute to more sustainable and climate-aware wastewater management in the future.
23.3.2026
