Environmental pollution is often invisible: Nitrate in groundwater, pesticides in water bodies, air pollutants, or heavy metals in the soil cannot be directly measured across the entire area. Nevertheless, far-reaching decisions must be made based on limited measurement data—such as where protective measures are necessary, how natural resources should be managed, or which areas are considered contaminated.
At the Chair of Geoinformatics at the University of Jena, we are working on a project funded by the Federal Environment Agency to address the question of how such decisions can be made in a better, more transparent, and fairer way with the help of models. A key example is nitrate in groundwater. Nitrate is produced, among other things, by nitrogen fertilization in agriculture and, in excessively high concentrations, can endanger drinking water quality and aquatic ecosystems. The European Union therefore sets a threshold value of 50 mg/L. If this value is exceeded in a given area, stricter management rules may be required. This makes it all the more important to estimate the pollution level as reliably as possible—both to protect groundwater and to ensure a fair assessment of the affected areas.
However, environmental monitoring networks consist only of individual monitoring points. Between these points, it is necessary to model how high the pollution level is likely to be and where uncertainties exist. Modern statistical techniques and AI methods can help with this—but even they are only as good as the data on which they are based. New monitoring stations are therefore currently being installed nationwide, and this is expensive. This raises a key question: Where will additional monitoring stations yield the greatest insights?
Visitors can explore this very question for themselves at our interactive booth. In a simplified decision-making game, you’ll have a virtual budget of €30,000 to place three new groundwater monitoring stations. Various map layers help you assess the study area: Where are there indications of elevated nitrate levels? Where is uncertainty particularly high? And where could additional measurements contribute most to better assessing whether the 50 mg/L threshold is being exceeded? After all, we must use our limited resources as effectively as possible!
Once you’ve selected your monitoring sites, we’ll compare your choices with locations suggested by an optimization algorithm we developed. This reveals which strategies are particularly well-suited to reducing uncertainties and monitoring environmental impacts more effectively.
The nitrate example is representative of many similar environmental issues. Whether it’s groundwater, air, soil, or surface water: time and again, monitoring networks must be planned, data gaps assessed, and decisions made under conditions of uncertainty. Our booth demonstrates how models can help make sensible use of limited resources, assess environmental impacts more transparently, and place the protection of natural resources on an evidence-based foundation.
