Groundwater assessment improves risk management
Water is both a precious and a vulnerable resource. In Europe, about 65 % of drinking water and 25 % of the water used in irrigation comes from groundwater. According to the European Environmental Agency (EEA), about a quarter of Europe’s groundwater contains chemical impurities. The FORENSHYD project sought to improve modelling tools used to identify the location, source and quantity of pollutants found in groundwater, as well as the timing of spills. This research was undertaken with the support of the Marie Skłodowska-Curie Actions programme.
Variables in forensic hydrology
Forensic hydrology is the application of science to the study of groundwater contamination to support the enforcement of environmental law. Many factors contribute to making this a difficult problem. The spatial variability of soil means that it reacts differently to contaminants, depending on various reasons. This impacts transport processes, such as dispersion and diffusion, which affect how contaminants move through soil and into groundwater. Chemical reactions caused by spills, such as acid-based reactions and the dissolution of minerals, add another layer of complexity when studying groundwater contamination. Collectively the combination of transport processes and chemical reactions is known as reactive transport processes, and deciphering their interactions requires sophisticated modelling techniques.
Advances in groundwater modelling
A complex array of algorithms known as an ensemble Kalman filter (ENKF) is used to assess groundwater contamination events. Traditional applications run into difficulties when it comes to determining multiple sources of pollution. Novel applications of machine learning made by Jaime Gómez-Hernández, FORENSHYD project coordinator, to the ENKF led to developments that improved results. The forecasting accuracy of the optimisation tool designed by Gómez-Hernández was tested in multiple settings. Tests were carried out in synthetic scenarios through computer simulation, as well as in controlled sandbox experiments. Additionally, tests were carried out in two demonstration sites. According to MSCA-IF beneficiary, Alicia Sanz-Prat: “The results demonstrated that it was feasible to simultaneously identify the locations of contaminant sources and estimate reactive parameters, even when dealing with complex and heterogeneous geological uncertainty.”
Open source technology
One of the strengths of the FORENSHYD project is that the novel technological advances that led to improved forensic hydrology applications are based on open-source packages. European governments, water agencies and other stakeholders all have access to the tools tested by the project. The long-term improvement of forensic hydrology depends on the open-source nature of the optimisation tool, and the success of the project is igniting discussion. According to Sanz-Prat: “Forthcoming peer-reviewed publications are expected to initiate an engaging dialogue on the applicability and correlation between uncertainty levels in science and judicial procedures. This discussion could lead to the identification of new commercial competitiveness and synergies in the European market among interdisciplinary and intersectoral stakeholders in soil and water.” Careful management of resources is crucial to society, and no resource is more essential than water. By improving the design of open-source hydrologic data assimilation and assessment tools and testing them in multiple controlled settings, FORENSHYD has contributed to the security and well-being of the European community.
Keywords
FORENSHYD, groundwater, contamination, optimisation tool, forensic hydrology, modelling, ensemble Kalman filter, open source technology
