Department Water Resources and Drinking Water

High nitrate levels in groundwater have been a concern in European politics for decades. Yet, nitrate is still one of the substances that most frequently exceeds the guideline limits in European and Swiss aquifers. In this project, we use machine learning approaches to model nitrate levels in groundwater across space and over time throughout Switzerland. This data-driven spatial modeling examines the individual contributions of the many factors that make an area more susceptible to nitrate pollution, including anthropogenic and ecological factors such as land use, nitrogen deposition from the atmosphere, climate, and soil properties. The temporal modelling aims to improve our understanding of the main drivers of groundwater nitrate contamination in Switzerland over the last 20 years, as well as what might be expected in the future.