SWAIN: Sustainable Watershed Management Through IoT-Driven Artificial Intelligence

The contamination of water resources poses a significant threat to the environment. Rapid detection of pollutants and their sources in watersheds is vital for the sustainable management of these resources. Although there have been studies on tracking micropollutants in European water resources, the efficient integration of this data into decision-making processes to safeguard against chemical pollution remains elusive. Internet of Things (IoT) and Artificial Intelligence (AI) technologies offer promising solutions, enabling both immediate and long-term strategic responses to these environmental challenges.

Our proposed system aims to enhance understanding, provide near-real-time responses to pollution incidents, and improve predictions of pollution spread. The core of this approach is an integrated decision support system that combines micropollutant measurements with real-time hydrodynamic and meteorological data, facilitating sustainable water quality management. Micropollutants, owing to their persistence and source-specific nature, serve as reliable indicators of pollution and its origins.
This approach involves the integration of cutting-edge technologies to enhance water pollution management. First, it utilizes scalable IoT technology for optimal data collection for analysis. Furthermore, we propose a novel AI model that employs a graph-based representation to capture the relationships and dependencies among various data streams. This integrated approach not only addresses immediate pollution incidents but also contributes to long-term sustainable water resource management.

• University of Vienna University (Coordinator)
• TU Wien
• Università della Svizzera italiana
• Finnish Environment Institute
• Istanbul Technical University
• Bogazici University

01.03.2021 – 31.05.2024 (3 years)