Department Urban Water Management

Urban water systems in Switzerland are under growing pressure from climate change and increasing urbanization. More frequent extreme rainfall, reduced river flows, and rising pollution levels challenge the performance of existing wastewater infrastructure, threatening both environmental quality and public health. At the same time, costly upgrades to centralized treatment systems may no longer be sustainable or effective.

The EMPOWER-DD project aims to address this challenge by developing data-driven tools and strategies for smarter, more resilient wastewater management. By integrating real-time positioning data from mobile phones, historical data archives from wastewater treatment plants (WWTPs) across Switzerland, and other data sources, the project will improve our ability to predict and manage pollution dynamics across entire sewer networks. Combining expertise in environmental engineering, urban hydrology, and data science, EMPOWER-DD will develop and test novel approaches—such as pollution-based real-time control and decentralized waste design—that support adaptive infrastructure planning and help cities adapt to the effects of a changing climate.

• Improve prediction of wastewater generation and pollution using real-time mobile positioning data.
• Analyze stormwater pollution across hundreds of WWTPs using data mining and machine learning.
• Develop integrated models and monitoring frameworks for system-wide pollution assessment.
• Explore smart real-time control and waste design strategies to optimize wastewater system management.

Pollution dynamics in a sewer system depend on when and where people move through a city and generate wastewater—but current methods assume static populations. This PhD project explores the use of mobile positioning data (MPD) as a new type of sensor for urban wastewater systems. By analyzing anonymized, high-resolution mobility data from mobile phone networks, the project will investigate how population dynamics relate to wastewater flows and pollution loads. The goal is to develop data-driven (and/or hybrid) models that can predict when and where pollution peaks will occur—improving both monitoring and control. These tools will ultimately enable more responsive, efficient wastewater management, especially under rapidly changing conditions caused by urbanization and climate change.

Urban stormwater runoff is a major and growing source of water pollution—yet predicting its composition and dynamics remains a challenge. This PhD project focuses on using Switzerland’s extensive monitoring archives to understand and model stormwater pollution at a national scale. The work includes compiling and harmonizing over 20 years of influent data from hundreds of WWTPs, and combining this with high-resolution meteorological, land-use, and catchment data. Machine learning models will be used to identify key patterns in stormwater pollution and evaluate how rainfall intensity, land cover, and infrastructure shape pollutant loads. By benchmarking Swiss data against international datasets, this project will also assess how transferable these insights are across regions and climate zones—helping cities adapt to increasingly intense rainfall events.