KIMoDIs - AI-based monitoring, data management and information system for coupled forecasting and early warning of low groundwater levels and salinisation

Country / Region: Germany

Begin of project: March 1, 2023

End of project: February 28, 2026

Status of project: September 25, 2024

In recent years, parts of Germany have experienced groundwater shortages due to warmer and drier summers and lower annual rainfall. In addition to the death of trees, low water levels in rivers and the drying out of groundwater-dependent terrestrial ecosystems, there are also local bottlenecks in water supply. Climate projections up to the year 2100 show that the trend towards drier and warmer summer will continue. With presumably increasing water demand and further falling groundwater levels, there is also a risk of groundwater salinisation in some areas due to the upwelling of deep saline waters or seawater intrusion near the coast. So far, there is a lack of intelligent tools for forecasting extreme groundwater levels with corresponding possibilities for counteracting them, e.g. through adjusted resource management.

Visual abstract of KIMoDIs project concept

The aim of the KIMoDIs project is to develop an artificial intelligence (AI) based monitoring, data management and information system for short- (seasonal), medium- (1 to 10 years) and long-term (up to year 2100) prediction of groundwater levels and salinisation. In addition, the user-specific decision support system will provide early warning of groundwater low levels and salinisation as well as the related damages. Various groundwater extraction scenarios are intended to allow an intelligent planning of site-specific countermeasures, and thus ensure a sustainable groundwater management. This approach integrates all available data from water suppliers as well as conventional groundwater monitoring and further combines it with real-time remote sensing of irrigated agriculture. State-of-the-art AI methods (DeepLearning, Explainable AI) will be used for the coupled prediction of groundwater levels and salinisation, for optimising the monitoring network and for early warning of critical conditions for drinking water supply and groundwater-dependent ecosystems. AI methods offer significant benefits compared to established methods because, as data-based models, they are able to extract complex relationships from existing data and transfer them both spatially and temporally.

The methodological development and demonstration of the approach will be carried out on a supra-regional scale in the federal state of Brandenburg, considering aspects such as extraction rates for drinking water supply, industry and agriculture, monitoring of irrigated agriculture with high temporal resolution, and the risk of salinisation due to overexploitation. The developed approach will be transferred and tested on i) a regional scale, for a catchment area of the Harz waterworks in Lower Saxony, focusing on the problematic effects of low groundwater levels, and ii) a local scale, using the example of the island of Langeoog regarding the tourism-related strong variability of seasonal water demand in case of increasing drought, as well as the risk for drinking water supply due to seawater intrusion. In this way, the project is contributing to the fulfilment of the goals of the National Water Strategy.

In addition to coordinating the project, BGR`s work includes data management, scientific conceptual design of real-time monitoring systems, co-development of AI models for the coupled prediction of groundwater levels and salinisation, definition of future climate and groundwater extraction scenarios, development of a prototype for decision support in groundwater management, and evaluation of the methods in the pilot regions as well as their transferability.

The Federal Ministry of Education and Research (BMBF) is funding the joint project "KIMoDIs" within the "Sustainable groundwater management" funding measure as part of the federal research programme on water "Wasser: N", which is part of the BMBF strategy "Research for Sustainability (FONA)". The BMBF funding code is 02WGW1662B.

Project logos of the Federal Ministry of Education and Research (BMBF)

Literature:

Papers

• CHIOGNA, G., MARCOLINI, G., ENGEL, M. et al. (2024): Sensitivity analysis in the wavelet domain: a comparison study. - Stoch Environ Res Risk Assess 38, 1669-1684. DOI: 10.1007/s00477-023-02654-3
• ENGEL, M. & KÖRNER, M. (2024): Sentinel-2 Tiling Scheme Grid-Overlay for Efficient I/O-Operations Based on Spherical Voronoi Polygons and Local Optimization. - In IEEE International Geoscience and Remote Sensing Symposium. DOI: 10.1109/IGARSS53475.2024.10640984

Presentations

• DOLL, F., BRODA, S., HEUDORFER, B., KUNZ, S., THULLNER, M., WETZEL, M. & LIESCH, T. (2024): Die Eignung von Modellen des maschinellen Lernens für die gekoppelte Vorhersage von Grundwasserständen und Grundwasserversalzung. - 27. Tagung der Fachsektion Hydrogeologie e. V. in der DGGV e. V., RWTH Aachen University.
• ENGEL, M. & KÖRNER, M. (2024): Sentinel-2 Tiling Scheme Grid-Overlay for Efficient I/O-Operations based on Spherical Voronoi Polygons and Local Optimization. - IEEE International Symposium on Geoscience and Remote Sensing (IGARSS), Athens.
• ENGEL, M., KUNZ, S., WETZEL, M. & KÖRNER, M. (2024): Multiresolution Analysis based Assessment of Agricultural Effects on Groundwater Levels. - ESA Agriculture Under Pressure, Frascati.
• SCHULZ, A., KUNZ, S., NÖLSCHER, M., WETZEL, M., BRODA, S. & BIESSMANN, F. (2024): Entwicklung eines globalen Modells für kurzfristige Grundwasserstandsvorhersagen unter Anwendung des Temporal Fusion Transformers. - 27. Tagung der Fachsektion Hydrogeologie e. V. in der DGGV e. V., RWTH Aachen University.
• WETZEL, M. (2024): Künstliche Intelligenz in der Hydrogeologie. - Berlin-Brandenburger Brunnentage, Berlin.
• WETZEL, M., SCHMIDT, L.-K., HERMSDORF, A., KUNZ, S., LIESCH, T., HEUDORFER, B., DOLL, F. & BRODA, S. (2024): Advancing water resource management: Insights and implications from global machine learning models in groundwater prediction. - European Geosciences Union General Assembly (EGU24), Vienna.

Posters

• GONZÁLEZ, E., BRODA, S. & ELBRACHT, J. (2023): KIMoDIs – KI-basiertes Monitoring-, Datenmanagement- und Informationssystem zur gekoppelten Vorhersage und Frühwarnung vor Grundwasserniedrigständen und -versalzung. - Tagung Grundwasserströmungs-Modellierung, Hanover.
• GONZÁLEZ, E., ELBRACHT, J., WETZEL, M. & BRODA, S. (2024): KIMoDIs – Projektgebiet West-Hümmling. - 27. Tagung der Fachsektion Hydrogeologie e. V. in der DGGV e. V., RWTH Aachen University.
• WETZEL, M., SCHMIDT, L.-K., HERMSDORF, A., KUNZ, S., LIESCH, T., HEUDORFER, B., DOLL, F. & BRODA, S. (2024): KI-basiertes Grundwassermanagement in Brandenburg: Notwendigkeit und bisherige Entwicklungen. - 27. Tagung der Fachsektion Hydrogeologie e. V. in der DGGV e. V., RWTH Aachen University.
  

Partner:

• Karlsruhe Institute of Technology (KIT)
• Brandenburg State Office for the Environment (LfU)
• Mapular
• German Federal Institute of Hydrology (BfG)
• Technical University of Munich (TUM)
• Germany's National Meteorological Service (DWD)
• State Office for Mining, Energy and Geology of Lower Saxony (LBEG)
• Brandenburg State Office for Mining, Geology and Raw Materials (LBGR)