As urbanization and transport demands rise, railway systems face higher risks from operational disruptions due to system failures and climate-related impacts, especially in urban areas. Understanding water flow and distribution beneath railway embankments is crucial not only for assessing potential contaminant transport in soil and groundwater but also for identifying possible structural weaknesses. This study implements numerical simulations to investigate these dynamics, focusing on a standard railway embankment designed according to German norms. The model setup includes ballast, a low-permeability protection subgrade surface layer (PSS), a subgrade (embankment base), and a subsurface to a depth of 10 m. Numerical simulations were conducted following a univariate approach. Variation parameters included key hydraulic characteristics of the embankment and subsurface. Further parameters were groundwater recharge rate, groundwater level, existence of a less permeable horizontal layer at multiple depths, and thickness of the embankment. The most sensitive parameters shaping water percolation fronts are identified as empirical coefficients (especially the Van Genuchten shape variable α), groundwater levels and hydraulic conductivity. The results indicate that the PSS layer retains the greatest amount of water from the rail system, particularly beneath the ballast. Low α values (0.8 m–1 to 7.7 m–1) led to total retention of water within the embankment, with no visible percolation fronts and the lowest values observed at the PSS layer. Shallow groundwater increased initial water content up to 0.30 at the PSS layer, rapidly approaching near fully water-saturated conditions. Embankments with low hydraulic conductivities (0.048 m d−1 to 1.207 m d−1) show the highest water contents, reaching up to 0.46, forming a clogging area prone to retain contaminants.

Mine Thermal Energy Storage (MTES) systems represent a promising solution for seasonal heat storage to balance the seasonal regenerative energy supply-demand mismatch in former mining regions. MTES may induce hydrochemical changes due to repeated thermal cycling. This study investigates the hydrochemical evolution of mine water during the operation of a pilot-scale MTES system in the Research and Teaching Mine Reiche Zeche of the TU Bergakademie Freiberg (Germany) and in laboratory experiments in the BMFTR-funded project 'MineATES'.The MTES pilot system consisted of a 20 m3 mine water filled basin located in the unsaturated zone of the mine. The Mine Water in the basin and the inflows showed typical acid mine drainage characteristics with original pH at 2.7, redox potential at 840 mV and sulphate values around 500 mg/l, up to 30 mg/l zinc and up to 26 mg/l dissolved iron. In total three heating and three cooling cycles were conducted at the test site from original 11.6 °C to water temperatures reaching 26 °C in the first two heating cycles and 39 °C in the last heating cycle. The basin was sampled weekly and the inlets into the storage basin were also monitored.Results indicate that the mine water chemistry was mostly controlled by temperature, mine water influx, and evaporation. We observed iron precipitation during heating after high inflow periods. After the inflow was significantly reduced iron precipitation did so, too. Iron concentration decreased from 23.7 mg/l to 2.5 mg/l during the first cycle with high inflow conditions. After the inflow was reduced a decrease from 1.5 mg/l to 0.2 mg/l iron was observed in the third heating cycle. The third heating cycle reached 39 °C and induced evaporation through the gaps of the basin cover leading to an enrichment in components by the factor 2 and to gypsum formation above the water line. For future optimisation of MTES systems, results suggest a reduction of new mine water inflow to prevent repeated iron precipitation and full contact with the surrounding rock to minimize evaporation effects.

Flooded and partially flooded mine galleries represent a largely untapped type of subsurface reservoir within underground thermal energy storage (UTES), known as mine thermal energy storage (MTES). This study presents a long-term field demonstration of a fully instrumented MTES test-bed at the Reiche Zeche underground mine in Germany. Three controlled heating-cooling cycles with a combined duration of 504 days were carried out and monitored through dense thermometry, tracer testing, and hydrochemical and materials analyses. A total of 38.0 MWh of heat was supplied. Approximately 90% of the stored energy resided in the surrounding gneiss, confirming that the rock mass functioned as the principal store while the basin water acted as a rapid carrier and exchanger interface. The rock warmed by 10.1 K at 1.8 m depth after the hottest cycle, consistent with a conduction-dominated regime. Tracer dilution determined a throughflow of 79 L h- 1 with a residence time of about 10.5 days, corresponding to an advective heat-loss coefficient of 0.092 kW K- 1. Warm phases triggered Fe (II) oxidation and precipitation of Fe-oxyhydroxides that dominated exchanger fouling, while hydrophobic coatings limited conductance losses to roughly 18%. Integrated field, laboratory, and numerical analyses showed that hydraulic isolation and oxygen control were the main levers for improving efficiency. The results demonstrate reproducible MTES operation under mine conditions and show that advective loss and exchanger fouling govern recoverability. The derived metrics provide a practical basis for MTES design in similar underground settings and highlight the potential of post-mining infrastructure to contribute to the underground thermal energy storage portfolio.

Predicting nitrate (NO3-) attenuation in groundwater requires solving the advection-dispersion-reaction dynamics governing donor limitation, nonlinear kinetics, and competition among electron acceptors. This study develops a physics-informed neural network (PINN) surrogate for reactive nitrate transport in saturated porous media, embedding the advection-dispersion-reaction equation, boundary conditions, non-negativity, and a redox-ordering constraint directly within the loss function. Four one-dimensional benchmarks of increasing geochemical complexity were constructed with PHREEQC, spanning linear denitrification, dual-linear redox competition, dual-substrate Monod kinetics, and fully coupled dual-Monod nitrate-Fe(III) reduction. The PINN was trained on full space-time fields using stochastic minibatches under a simulation-wise train-test split to avoid information leakage. Across all benchmarks, the surrogate reproduced the reference fields with high accuracy (test RMSE similar to 1 & sdot; 10-5 mol kgw-1; R2 >= 0.99 in nonlinear cases) while maintaining ADR residuals, boundary condition violations, and non-negativity penalties several orders of magnitude below characteristic concentration scales. The dual-Monod regime, representing the strongest kinetic nonlinearity, was captured with robust fidelity, including DOC-limited tailing and nitrate-Fe(III) competition. Uncertainty quantification using a 50-member ensemble with split-conformal calibration produced sharp, statistically valid prediction intervals whose width increased only near reactive fronts, consistent with the underlying nonlinear sensitivity. These results demonstrate that PINNs can serve as accurate, physically consistent surrogates for geochemical reactive transport and offer a computationally efficient pathway for scenario analysis, uncertainty propagation, and incorporation into groundwater management workflows.

The main objectives of the study are: (i) to evaluate the overall groundwater quality around the marble dumping sites in Kishangarh, Rajasthan, (ii) to assess the geochemical attributes of the water quality parameters with a focus on fluoride (F-) contamination, and (iii) to evaluate the potential anthropogenic impacts on groundwater quality. To achieve the objectives, groundwater samples are collected from the Kishangarh area. The water samples are analysed for different water quality parameters, and geochemical analysis is done to assess the major attributes. Spatial distribution of water quality parameters and land use patterns provide insight into the possible impact of dumping sites. The outcome of the study shows that the groundwater quality in the Kishangarh region is of poor quality, where up to 95

Nitrate ( ) contamination in groundwater often persists due to donor limitation, redox competition, and long residence times that constrain natural attenuation. This study presents a physics audited and uncertainty aware surrogate modeling framework for reactive nitrate transport in porous media. High-fidelity PHREEQC simulations were generated for four one-dimensional benchmarks of increasing geochemical complexity, ranging from linear heterotrophic denitrification to dual-substrate Monod kinetics with explicit - competition. The surrogate models were trained directly on space-time concentration fields using the CatBoost gradient boosting algorithm, with hyperparameters tuned via Bayesian optimization and simulation level data partitioning to prevent leakage. The surrogate takes hydrodynamic and geochemical inputs along with spatial and temporal coordinates, and predicts nitrate concentration fields across the domain. Uncertainty quantification was performed using a subsampling ensemble to characterize epistemic variability and conformalized quantile regression to provide calibrated prediction intervals. A physics audit was applied post hoc to verify consistency with the governing advection-dispersion-reaction balance, solute mass conservation, and non-negativity. Results show that the surrogates accurately reproduce nonlinear reaction fronts and donor-acceptor competition, achieving test R2 up to 0.997 with mass balance errors typically below 0.5%. Epistemic uncertainty remains small relative to aleatoric variability, which is concentrated along reactive transition zones. The presented framework is a reliable and computationally efficient tool for scenario analysis and risk-informed groundwater quality management in nitrate-impacted aquifers.

Underground mines, once vital industrial hubs, hold immense potential for innovative applications, including Mine Thermal Energy Storage (MTES). MTES repurposes partially and fully flooded mine cavities as reservoirs for storing surplus heat or cold, presenting a novel alternative to conventional Aquifer Thermal Energy Storage (ATES). While promising, MTES faces challenges such as scaling, corrosion, energy loss, and interactions between the geological matrix and technical infrastructure.To address these challenges, TU Bergakademie Freiberg has established a living MTES geo-lab at the historic Reiche Zeche silver mine. Key features of this facility include a 21-cubic-meter water reservoir and over 90 temperature sensors embedded in Freiberg Gneiss. The pilot-scale MTES simulator setup allows for continuously monitoring heat transfer during thermal energy injection and extraction cycles realized by a mobile heat pump system. Early findings are revealing an average background rock temperature of 12 °C, a fast conductive heat transport within the rock as well as a good storage potential with elevated rock temperatures of up to 25 °C in approximately 2 meters from the water body. However, significant heat losses across system boundaries have been observed, with advective heat transport via flowing water identified as the primary contributor.Parallel laboratory-scale experiments using column flow setups and batch reactors simulate MTES conditions, exposing rock and mine water to temperature cycles ranging from 10°C to 60°C. These experiments demonstrate significant chemical changes, including the precipitation of 90% of dissolved iron. These findings offer valuable insights into the chemical stability and thermal efficiency of MTES systems.Two complementary methods were employed to quantify effective inflow and energy dissipation caused by mine water movement. First, a dilution test with NaCl was conducted. Second, inflowing water volume was calculated based on reservoir water level reductions. Results indicate that the calculation based on inflowing water volume provided more reliable values, while the formula used in the dilution test requires further refinement.Additionally, numerical simulations using OpenGeoSys (OGS) software are being developed to assess the influence of fracture networks in the surrounding rock formation on heat storage and recovery performance. Preliminary results indicate that fractures enhance advective heat transport, leading to lower heat recovery ratios during cyclic operation.

Recently, Aquifer Thermal Energy Storage (ATES) systems gained increasing attention as a suitable storage method for local and temporary surplus thermal energy in aquifers. Among others, the success of ATES depends on the properties of the aquifer like hydraulic permeability, thermal conductivity, and porosity. During ATES operation, different pressure and temperature conditions above and below the surface can cause clogging and scaling processes, eventually leading to operational and maintenance issues or failures. In the research project „UnClog-ATES“ (funded by the Federal Ministry of Education and Research of Germany - BMBF), clogging and scaling processes are investigated on an interdisciplinary basis (microbiology, geology, hydrogeology, and geochemistry) and influencing factors for carbonate aquifers are determined. Based on the findings, countermeasures (e. g. scaling inhibitors or CO2 addition) are (further) developed. ATES conditions (pressure, temperature, hydraulics, and chemical composition) are systematically simulated: While 1-D column tests serve to model the transport processes taking place in the real system, 0-D batch reactor tests allow varying the hydrochemical environment and rock compositions. Two different ATES-relevant rocks are used as representatives of limestone in general: i) Jurassic limestone from Upper Malm, Germany ("Treuchtlinger Marmor”; mainly calcite) as a representative rock for the Malm, as well as ii) marble from Hammerunterwiesenthal, Germany (“Erzgebirgsmarmor”; calcite and dolomite). Water samples from the same Erzgebirge marble quarry are used as fluid. Batch and flow tests are conducted in cycles at ATES-typical temperatures between 5 °C and 60 °C. Quasi-continuous monitoring of fluid parameters in measuring cells and performing comprehensive hydrochemical and geochemical analyses before, during, and after the tests allows monitoring of alterations in hydrochemical and geochemical conditions. Hydrochemical analysis results of the first batch experiments (shaking “Treuchtlinger Marmor” with distilled water for approx. 30 days at 5, 40 and 55 °C) showed a decrease in the concentrations of Calcium and Magnesium with increasing temperature. Using a larger grain size showed an even greater decrease. Accompanying the experiments, hydrochemical modeling is used to quantify the processes and to estimate the experimental parameters a priori. The simulation results with PHREEQC first showed an equilibrium of the aforementioned rock and fluid materials at 15 °C and subsequently precipitating calcite while heating stepwise. Further simulations indicate that this precipitation can be prevented by adding a certain amount of dissolved CO2. UnClog-ATES intends to contribute to optimizing the prediction accuracy of hydrogeochemical reactions and to the creation of methods both for estimating the clogging potential and for developing and testing possible countermeasures.

In water-stressed regions, Managed Aquifer Recharge (MAR) is essential for water conservation, helping to sustain groundwater resources and increase resilience to drought. MAR typically involves using surface water, treated wastewater, stormwater, and runoff to address groundwater depletion. Since pharmaceuticals are commonly found in wastewater, stormwater, and treated effluent, it is crucial to understand their behavior in aquifers to prevent the unintended contamination of drinking water. This paper explores the transport dynamics of pharmaceutically active compounds (PhACs) under different environmental conditions commonly encountered in MAR systems using surface water and stormwater near agricultural fields, commonly known as Ag-MAR. The PhACs were chosen for their prevalence, toxicity, and ecological behavior, particularly in regions with varying agricultural parameters like pH, nitrate (NO3-) and zinc (Zn2+). Through laboratory experiments simulating typical Ag-MAR conditions, we examined how various parameters influence PhAC mobility, sorption, and degradation. Our results show that pH plays a significant role in PhAC transport, with acidic conditions (pH 4) generally enhancing retention, while neutral or alkaline soils (pH 6) increase mobility and the risk of groundwater contamination. Zn2+ and NO3- were found to further alter transport dynamics, with Zn2+ enhancing mobility through changes in soil adsorption and complexation, and NO3- influencing microbial degradation. Caffeine and ibuprofen exhibited higher mobility at neutral pH, while carbamazepine and diclofenac showed more complex behavior, depending on pH and the presence of Zn2+ and NO3-. The findings underscore the need for careful management of MAR systems in agricultural areas, taking into account the combined effects of pH fluctuations, heavy metal contamination, and nitrate presence to mitigate pharmaceutical pollution and ensure sustainable groundwater recharge.

Geochemical background values are commonly used for the authorization of remediation measures. However, in the case of the former Königstein uranium ore mine, located in Saxony (Germany), natural uranium concentrations in groundwater cannot be directly assessed, due to active mining activities. The Königstein uranium ore depositis located within the 4th aquifer and consists of the Oberhäslich Formation and 'Wurm'-Sandstone.This study aimed at deriving natural uranium content using batch shake test, assuming chemical equilibrium with regard to the speciation of uranium in solution and the binding to the rock matrix. For that purpose, representative subsamples from core material were taken for batch experiments and geochemical analyses. As uranium solubility strongly dependents on the redox state, the pH value, and the hydrochemistry of the target fluid, experiments were performed under various conditions.The rock samples were analyzed with respect to geochemical and mineralogical compositions, while uranium concentration in fluid samples was measured using inductively coupled plasma - mass spectrometry. Concentrations of major cations were analyzed using cation chromatography techniques. The PHREEQC software was used to analyses the species distribution of uranium under the hydrochemical conditions of the the unaffected inflowing water from the inflow area.Results show, that the sandstones in the middle of the 4th aquifer consist mainly of SiO2 with more than 98 wt.-%. Two-layer clay minerals and iron oxides were identified in another sample with fractions more than 50 wt.-% kaolinite. Uranium was found in the anaerobic zone in one rock sample at 1820 ppm and once at only 25.6 ppm.Equilibrium modeling revealed, that at a pH of 5.5 to 6 and under oxidizing conditions, uranium mainly occurs as UO22+, UO2OH+, and UO2CO3. It can therefore be expected that the uranium species present as cations will likely be adsorbed by the solid matrix. Initial shaking tests showed that combining reduced rock materials with oxidizing water led to excessive uranium fractionation into the fluid phase. Batch tests using deionised water showed uranium concentrations between 2 and 2.5 mg/l, contrasting expectation of natural uranium concentrations in the lower µg/l range.

Waterborne pathogens pose a particular threat to aquifers with low retention capacity, such as karst aquifers. To date, particle transport in groundwater is typically studied using tracer tests with microspheres or natural sediments, which might be unsuitable to derive transport characteristics of pathogenic bacteria or viruses. This study investigated the transport and attenuation behavior of two bacteriophages, MS2 and phiX174, as surrogates for pathogens in a karst aquifer in South-West Germany. Both bacteriophages and the solute reference tracer uranine were recovered within five days at a 9.1 km distant spring. Only a small proportion of infectious bacteriophages was recovered compared to uranine. Breakthrough curves indicated no observable retardation but major attenuation of bacteriophages during transport. The modeled transport parameters of bacteriophages were subject to high uncertainty relative to uranine due to insufficient analytical accuracy at the low concentrations observed. However, it was demonstrated that visual inspections of the breakthrough curves alone might lead to incorrect assumptions regarding dominating transport processes. The high attenuation of bacteriophages was attributed to attachment processes rather than inactivation, which affected the removal of bacteriophages only little in this study. Yet, the fast breakthrough of infectious bacteriophages at the spring indicates the high contamination potential of karst springs. Findings further suggest that only surrogates relevant to the specific requirements and injected at appropriate locations can accurately simulate the transport behavior of particles of interest. The results highlight the need for suitable experimental sites, tracer materials, and analytical methods for risk assessment studies concerning pathogens.

Decarbonizing the industrial and building heating and cooling sectors is a crucial step toward achieving carbon neutrality, necessitating innovative and sustainable solutions for the over-seasonal storage of excess heat energy. With Germany alone having more than 10,000 old mines, repurposing these sites to implement a controlled thermal energy storage strategy, known as mine-based thermal energy storage (TES), has emerged as a potential solution. To effectively utilize such partially flooded artificial cavities, it is crucial to fully understand the heat transport and storage behavior in these systems. In this work, a three-dimensional hydro-thermo-component (HTC) model was developed using the open-source simulation code OpenGeoSys (OGS). The model was initially verified against analytical solutions for the single fracture flow of heat and solute transport, respectively. Subsequently, stochastic discrete fracture matrix (DFM) geometries and meshes were generated using the computational suite Frackit, based on data from a pilot heat storage site in a water-filled mining cavity in Freiberg, Germany. This test site is geologically characterized as the Freiberg gneiss, a metamorphic fractured rock formation. The developed setup allows for investigating the thermal energy storage capacity and the energy recovery efficiency based on process simulations in OGS. The study evaluated the thermally affected zone in the fractured formation and quantified the amount of heat stored and recovered during cyclic operation. In addition, the solute transport distance within the surrounding rock can be evaluated under different hydraulic conditions. The general modeling workflow provides a basis for conducting techno-economic feasibility analysis of mine-based TES systems.

The (over-)seasonal storage of excess heat and cold in the subsurface is considered a promising solution to the manifold challenges of the energy transition. Many underground thermal energy storage (TES) systems are focusing on natural aquifers. In parallel, there has also been increasing attention on using artificial cavities in (partially) flooded underground mines. This special form is known as mine thermal energy storage (MTES). Like other underground TES systems, MTES faces several challenges. Many former mines are actively dewatered to keep a defined flooding level and, therefore, significant water flows can be present, especially in the main tunnels and shafts. The unintentional transport of stored heat energy out of the original storage area, whether through heat advection or conduction, ultimately leads to reduced recovery rates and suboptimal efficiency. Adoption of localized and hydraulically (more or less) isolated mine sections instead of entire levels may provide a solution to this technical challenge. An MTES Geo-Lab has been recently designed and established as part of the R&D project "MineATES", funded by the German Federal Ministry of Education and Research (BMBF). The in-situ Geo-Lab is located in the former silver mine "Reiche Zeche (Himmelfahrt Fundgrube)" at the TU Bergakademie Freiberg in Saxony, Germany, and focuses on the controlled simulation of TES cycles on a manageable scale. Specifically, a cuboid-shaped experimental reservoir (water capacity of approx. 21 cubic meters) in the northern field of the Reiche Zeche’s first level, slowly flown through by acidic precipitation water (pH values between 2 and 3), was chosen. The prevalent geological formation in this area is the Freiberg gneiss. Given the pilot-scale of the study site, heat losses across system boundaries are expected to be of an experimentally manageable magnitude – and are intentionally so. The immediate vicinity of the experimental reservoir has been equipped with an extensive thermal monitoring system. This includes more than 90 temperature sensors embedded at various distances from the reservoir walls, with some up to two meters deep and distributed across 18 boreholes. First measurements showed a background temperature in the rock of approx. 11.5°C on average. This monitoring system enables continuous tracking of transient temperature distributions in the surrounding rock, facilitating the quantification of heat losses and efficiency reductions during periodic heat/cold injection and extraction experiments, emulating real-world TES cycles. Furthermore, the Geo-Lab is equipped with multiple sampling points to monitor hydrochemical parameters over time.

This paper reviews the current research on aquifer thermal energy storage (ATES) and mine thermal energy storage (MTES) in Germany providing descriptions of 3 low-temperature ATES (LT-ATES), 8 high-temperature ATES (HT-ATES), and 2 MTES research sites. While the overview reveals a diverse field of investigations spanning various spatial scales, research objectives, and methodologies, the predominant focus is limited to early-stage research with low technology readiness levels (TRL). The high number of HT-ATES research sites suggests greater research interest compared to LT-ATES. The integration of ATES into district heating (DH) grids in particular is a prominent research focus, yet almost none of the projects are specifically intended for practical implementation. Future research should therefore prioritize real-world demonstration projects and identify key locations, which is crucial for showcasing the benefits of ATES. The need for a streamlined regulatory framework that addresses environmental risks and ensures installation quality and efficient permit procedures is also discussed.

Aquifer Thermal Energy Storage (ATES) systems are gaining attention as a method to store surplus thermal energy in aquifers. However, during ATES operation, changes in pressure and temperature conditions can initiate clogging and scaling processes, leading to operational and maintenance issues or failures. In the “UnClog-ATES” project (funded by the BMBF, Germany), we investigate clogging and scaling processes in carbonate aquifers and develop countermeasures such as scaling inhibitors or CO₂ addition through an interdisciplinary approach that combines microbiology, geology, hydrogeology, and geochemistry.Aiming at carbonate aquifers, we used two types of limestone: i) Jurassic limestone from Upper Malm, Germany ("Treuchtlinger Marmor”; primarily calcite) and ii) Marble from Hammerunterwiesenthal, Germany (“Erzgebirgsmarmor”; mainly composed of calcite and dolomite). Water samples from the Erzgebirge marble quarry served as fluid phase in all experiments, which were conducted at ATES-relevant temperatures (5–60 °C).Shaking experiments (0-D) assess the influence of hydrochemical environments and rock compositions on rock and fluid alteration. A series of time-dependent shaking experiments at 5, 40, and 60 °C revealed that, with Erzgebirgsmarmor, Ca concentrations in fluid decrease over time at all three temperatures, while Mg concentrations increase. Conversely, Treuchtlinger Marmor exhibits the opposite behavior. PHREEQC modeling of the 60 °C experiments predicts precipitation of dolomite, calcite, aragonite, and vaterite.1-D column experiments systematically simulate ATES conditions, including temperature and chemistry to model transport processes. Preliminary results at 12 °C with Treuchtlinger Marmor indicate precipitation of dolomite, calcite and aragonite. Early findings from a 40 °C test run comparing both carbonate rocks show differences over time and compared to the results at 12 °C in pH, electric conductivity and alkalinity.These results highlight the need of further site specific investigations to enhance our understanding of hydrochemical processes and reactions during ATES operations. Findings from this study will improve the prediction of dissolution and precipitation processes and the development of effective countermeasures for clogging and scaling during ATES in carbonate aquifers.

Managed aquifer recharge (MAR) is a vital water management strategy that infiltrates surface water or wastewater effluent into aquifers through soil and sediment. However, this process can introduce pharmaceutically active compounds (PhACs) and their metabolites, posing environmental risks. This study investigates the transport behavior of four selected PhACs—caffeine, carbamazepine, diclofenac, and ibuprofen—under neutral pH conditions using column experiments in both unsaturated and saturated porous media. PhACs and a tracer solution were introduced into the system, and experimental results were simulated using the CXTFIT model to determine retardation and degradation factors. Experimental findings indicate high mobility for carbamazepine and ibuprofen across both unsaturated and saturated conditions. Ibuprofen behaved similarly to the tracer with a retardation factor of ~1 and negligible degradation, while carbamazepine showed slight retardation and tailing effects showing higher persistence in the water. Diclofenac significantly degrades in saturated media (44% recovery) but increases release under unsaturated conditions (98% recovery). This indicates the release of diclofenac through the vadose zone but can undergo degradation and retardation in aquifers. Caffeine displayed high retardation and degradation under both conditions independent of the moisture content during transport. These findings highlight the differential transport of PhACs during MAR showing contaminant release to the groundwater, emphasizing the need for effective management practices to mitigate contamination risks and ensure groundwater quality.

Aquifer characterization is essential for optimizing Aquifer Thermal Energy Storage (ATES) systems. Single well tests, also known as push-pull tests, are a common method to identify effective solute and heat transport parameters of the aquifer, which are crucial for the design and assurance of long-term performance of ATES systems. Tracer breakthrough curves from push-pull tests are commonly used to calibrate analytical or numerical models of heat and solute transport in order to infer effective transport parameters like dispersivity of heat and solutes, retardation factors, and porosity. The main bottleneck of such multiparametric calibration is the non-uniqueness of the inverse problem solution which requires ensemble-based optimization to address the parametric uncertainty. In addition, the field measurements can only be performed up to a certain confidence as well, which introduces additional uncertainty to the calibration results. To account for parametric uncertainty while targeting computationally affordable simulation, we have developed a surrogate model-based optimization framework for stochastic parameter optimisation. The surrogate model uses Gaussian process regression (GPR) to train and predict the objective function (maximum RMSE) based on up to six aquifer and tracer properties. For training and fast model evaluation, we implemented a stable 1D radial finite difference representation of the advection-dispersion equation for sorbing compounds including measured input time-series as transient boundary condition and wellbore storage to accurately model push-pull tests. The surrogate model is used to calibrate this model and to propose plausible parameter combinations. The optimisation framework was applied to push-pull experiments using uranine, iodide, lithium, and heat as tracers in a sandy aquifer in Horonobe (Hokkaido, Japan). The samples drawn from the posterior distribution resulting from the GPR-based optimisation show an overall good fit to the field observations. Based on the posterior parameter distribution, it was possible to shrink the uncertainty intervals of the solute and heat dispersivity and porosity. The outcome suggests low sensitivity to the solute retardation factors. However, the study also reveals that slight sorption may be acting in the Horonobe aquifer for some of the solute tracers commonly assumed to be conservative. Moreover, the study shows that exact porosity measurements may reveal the presence of sorption and thus improve the understanding of the tracers' behaviour. We demonstrate the benefits of using multiple tracers and high-resolution measurements to improve calibration accuracy under measurement uncertainty. The demonstrated approach offers a computationally efficient framework for addressing parametric uncertainty in push-pull test analysis, improving the design and optimization of ATES systems.

The increasing demand for water in industry, agriculture, and private households as well as climate change are leading to more dynamic river and groundwater levels. This demands a new and improved monitoring approach to groundwater resources. The BMBF-funded research project iMolch (project number: 02WGW1667D) aims to develop sustainable water management concepts for Germany using innovative monitoring strategies. The general purpose of the investigations is to gain a complex understanding of hydrodynamic and hydrochemical processes in order to enable a more sustainable use of water resources on the basis of the indicator concept. This study investigates the spatial and temporal variation of different substances using as one example the urban bank filtration site in Düsseldorf, Germany. This site is used to draw conclusions on groundwater quality and dynamics as well as redox processes using the transport and retention of organic trace substances.Hydraulic and hydrochemical measurements were carried out fortnightly over a period of 1,5 years (31/01/2018–08/05/2019) across a study transect on the Rhine riverbanks with 15 measuring points. The analyses focus on changes in the concentrations of organic trace substances over time and the relationship to flow distance, flow duration and climatic conditions. Based on land use and occurrence, various organic compounds, such as fertilisers and pesticides, but also pharmaceuticals and detergents are monitored. The dependence between substance concentrations and Rhine river level decreases with increasing distance to the Rhine. Following the extreme drought in the summer of 2018, during which only low concentrations of trace substances were detected in the entire study area, there was an abrupt increase in substances entering the Rhine. This coincides with a significant Rhine high stand in the winter of 2018/2019. It is noticeable here that an increased concentration of substances discharged from the Rhine can also be detected at measuring points beyond the well gallery, on the land side of the measuring point transect.In dry periods, such as the summer of 2018, the proportion of bank filtrate in the raw water is significantly lower. In contrast, the Rhine floods the pumping well gallery during high water level periods, such as the winter of 2018/2019. During this time there is no landward flow to the pumping wells. The occurrence of organic trace substances even shows that the flood of the river Rhine pushes the water in the bank filtrate up to the well gallery and far beyond into the hinterland despite ongoing water pumping. These observations are crucial for the subsequent water management during different water levels of the river Rhine. Prediction models will be built up to help as water management tools to improve monitoring systems and for transfer of these results to other sites.