EGU26-20039, updated on 14 Mar 2026
https://doi.org/10.5194/egusphere-egu26-20039
EGU General Assembly 2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Monday, 04 May, 16:15–18:00 (CEST), Display time Monday, 04 May, 14:00–18:00
 
Hall A, A.92
Improving local groundwater management in Estonia and Latvia through real-time monitoring and machine learning
Liina Hints1,2, Magdaleena Männik1,3, Enn Karro3, Inga Retiķe4, Jānis Bikše5, Mārcis Tīrums6, and Andris Vīksna6
Liina Hints et al.
  • 1Department of Hydrogeology and Environmental Geology, The Geological Survey of Estonia, Tartu, Estonia (liina.hints@egt.ee)
  • 2Department of Geography, University of Tartu, Tartu, Estonia (liina.hints@ut.ee)
  • 3Department of Geology, University of Tartu, Tartu, Estonia
  • 4Department of Environmental Science, University of Latvia, Riga, Latvia
  • 5Department of Geology, University of Latvia, Riga, Latvia
  • 6Latvian Environment, Geology and Meteorology Centre, Riga, Latvia

As climate change accelerates, extreme events like droughts and floods are becoming more frequent and severe across Europe, and pollution adds further strain on groundwater resources. Groundwater sustains ecosystems and provides most of the drinking water in Estonia and Latvia, making its sustainable and proactive management more important than ever. In practice, however, that kind of management is difficult to achieve. Current monitoring systems in Estonia and Latvia still operate on slow cycles of data collection and manual analysis, leaving little room for early intervention. Because of this, municipalities often find out about issues with groundwater quality and quantity when they have already escalated – when wells have run dry or contaminants have reached drinking water supplies. Even when data is available, it often requires expert interpretation, making it difficult to act quickly and prevent problems in time.

The cross-border 'HydroScope' project addresses these challenges by developing a groundwater early warning system for two pilot municipalities: Saaremaa (Estonia) and Dienvidkurzeme (Latvia). In these municipalities, telemetry systems tailored to local groundwater conditions are installed in monitoring wells, introducing real-time groundwater monitoring in Estonia for the first time and expanding the network in Latvia. Real-time digital spring systems complement the well monitoring network. 

Machine learning models are developed to automatically detect patterns in real-time groundwater quantity and quality data and to generate short-term predictions. This information feeds into two municipality-specific early warning platforms. These platforms visualize insights from near real-time data in an easily interpretable way, along with recommendations for what actions to take under different scenarios – e.g. reducing water use if a groundwater drought is likely, or identifying potential contamination sources when thresholds are approached.

The ‘HydroScope’ project is a first step toward establishing a real-time groundwater monitoring paradigm in Estonia and Latvia, and also toward making that real-time data directly usable for local decision-making, thus supporting sustainable and proactive groundwater management practices.

The project HydroScope (EE-LV00250) is funded by the European Union through the European Regional Development Fund (ERDF) within the Interreg VI-A Estonia–Latvia Programme 2021–2027.

How to cite: Hints, L., Männik, M., Karro, E., Retiķe, I., Bikše, J., Tīrums, M., and Vīksna, A.: Improving local groundwater management in Estonia and Latvia through real-time monitoring and machine learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20039, https://doi.org/10.5194/egusphere-egu26-20039, 2026.