Ezra Haaf

The stability of a slope is governed by a combination of factors, where the hydromechanical properties of the soil are the most prominent ones. The groundwater conditions in such assessments are however, by Swedish practice, generally simplified to a two-dimensional (2D) linear interpolation between measured data, although the three-dimensional (3D) conditions may vary greatly at a site. This can lead to critical areas to be overlooked, especially for sites with variable topography, complex soil stratification or varying soil depth.This study thereby investigates the integration of a 3D groundwater model into 3D LEM slope stability analysis to account for spatial variations. The groundwater model is generated as a finite difference model via the open-source software MODFLOW and the LEM analysis is performed with PLAXIS 3D LE using the General Limit Equilibrium (GLE) with half-sine function. The PLAXIS 3D LE applies the two-directional 3D-method and the Cuckoo search method, which allow for asymmetrical failure mechanisms and does not require any predefined search area by the user, in contrast to e.g. SCOOPS3D.The study was applied to a geological site, Skälsbo, located along the Göta River valley. The site consists of thick deposits with soft sensitive clays with eroded slopes facing Göta River. Thorough geotechnical investigations have been performed at the site as a part of the Göta River Commission work to reduce landslide risks along the river.The results show that the advanced 3D groundwater model can be successfully imported into 3D LEM for a simple, yet computational efficient, uncoupled hydromechanical analysis of the slope stability at regional scale. Comparisons of results from dry 2D analysis shows comparative results between LEM and corresponding finite element analysis. The method has thereby great potential in incorporating future climate scenarios and their effect on regional stability, to detect both migration of critical stability areas and changes in its distribution over time. The method also shows that the user can seamlessly generate 2D models from the regional model for further assessment. The strength of using an advanced groundwater model, such as MODFLOW, is that both historical and future groundwater scenarios can be accounted for and thereby bring a robustness to the stability evaluation. This approach accounts for the complex groundwater situation, to ultimately better predict and optimize the need and extent of mitigation measures for cost- and environmental purposes.

The valley of the River Göta Älv in southwest Sweden is highly prone to landslides due to the presence of underlaying clay deposits. Landslides occur when driving forces and moments exceed resisting forces and moments, and this balance can be altered by elevated pore-water pressures thereby reducing effective stress. Pore pressure varies with climatic drivers (precipitation and evapotranspiration) and with changes in external loading, such as river stage fluctuations.Here we apply the impulse–response modelling framework Pastas to calibrate, validate, and simulate pore-pressure time series from piezometers installed along the Göta Älv valley. Since 2019, pore pressure has been monitored at multiple locations and depths. Model calibration used daily precipitation, temperature, potential evapotranspiration (PET), and river level time series. In total, 127 pore-pressure series were modelled using multiple combinations of impulse response functions and evapotranspiration formulations. Based on calibration and validation performance, 58 series were deemed suitable for climate-driven simulations. The most common causes of unsatisfactory model performance were (i) failure to reproduce rapid responses, (ii) threshold-like behaviour leading to underestimation of extreme high levels, (iii) short records limiting representation of inter-annual variability, (iv) shifted dynamics from the calibration to the validation period, and (v) potential outliers related to initialization of pressure sensors, measurement errors, or gas intrusion in instruments.The acceptable models were forced using the CMIP5 ensemble of EURO-CORDEX regional climate model (RCM) simulations for 1970–2100 (daily precipitation and temperature). Extreme high pore-pressure levels were quantified as 100-year return levels using a generalized extreme value (GEV) distribution under RCP8.5. For most series, the projected median change in the 100-year return level by 2100 is

The study of hydraulic head changes over time is a common task for groundwater hydrologists. Groundwater signatures are numerical metrics, or statistical aggregates, that quantify the behavior observed in hydraulic head hydrographs. Signatures can be helpful in a number of classical hydrological tasks, such as hydrograph classification, clustering, change detection, and model evaluation, selection, and calibration. Despite the potential benefits of using signatures in groundwater studies, their application has not yet been thoroughly explored. To support research into the application of signatures in groundwater studies, we introduce the new groundwater signatures module from the Pastas software. The signatures module is written in Python, fully tested and documented, and available as open‐source software under the MIT license. In this paper, it is shown how the signatures are tested and can be used in practical applications through two examples. In the first example, signatures are used to characterize and cluster monitoring wells in a nationwide monitoring network in Switzerland. In the second example, signatures are used to evaluate how well different groundwater model structures simulate the heads. Future research opportunities involving groundwater signatures are discussed.

Hell no, we haven’t yet fully explored the potential of spatial modeling of groundwater parameters—and also the Groundwater Spatial Modeling Challenge (https://groundwater-spatial-modeling-challenge.github.io/challenge2024/) hasn’t solved it. However, this challenge fosters learning and discussion within the community by providing standardized data and performance evaluation. At the same time, it helps uncover previously unknown members of the community and facilitates collaboration.This year, we present the interim results of all participating teams, comparing their methodological approaches and providing a preliminary analysis of model performances on the training and test data splits. Multiple metrics are used to account for their different focal points and limitations. Since quantitative metrics alone are insufficient to fully evaluate model predictions, we also analyze and discuss the generated maps of nitrate concentrations for the German state of Baden-Württemberg. Additionally, we provide a brief comparison of the prediction intervals submitted by each team.Beyond presenting these results, our main aim is to provide a platform for further exchange and collaboration on these topics.To encourage more teams to join, we are extending the participation deadline once again to July 15, 2025! Interested individuals or teams can find more information and participate via the link above. Submissions made by early April will have their results presented at the EGU.

This study aims to improve the regionalization of groundwater head dynamics using static environmental features. Recent machine- and deep learning studies have explored the use of these features for spatial and temporal imputation (Haaf et al., 2023) or improvement of global models (e.g., Chidepudi et al. (2024); Heudorfer et al. (2024); Nolte et al. (2024)). While physiographic features, including geology, land cover, anthropogenic factors, and topography, have been identified as important predictors of groundwater dynamics at regional and watershed scales (Haaf et al., 2020; Haaf et al., 2023; Rinderer et al., 2017; Zhao et al., 2023), there is still a lack of understanding on how to leverage static features to achieve significant model improvement for groundwater time series regionalization (e.g., Heudorfer et al., 2024; Nolte et al., 2024).In this study, we use a data-driven, static feature-based approach to regionalize groundwater head duration curves and reconstruct them based on similar donor sites (Haaf et al., 2023). We evaluate the similarity of static features compared to the geographical proximity of donor sites. The data set consists of more than 150 ten-year, daily groundwater head time series in the upper Danube catchment and more than 60 static features at each site.Our findings suggest that geographical proximity, related to both physiographic and climatic similarity, is the best default approach for selecting donor sites for regionalization. However, in specific cases where the nearest donor sites were located in different hydrogeological regimes, static features significantly improve regionalization. The study demonstrates the potential for improving the regionalization of groundwater dynamics using spatial features in diverse hydrogeological settings. Further research on larger and more diverse data sets is warranted to allow for robust feature selection strategies. ReferencesChidepudi, S. K. R., Massei, N., Jardani, A., Dieppois, B., Henriot, A., & Fournier, M. (2024). Training deep learning models with a multi-station approach and static aquifer attributes for groundwater level simulation: what’s the best way to leverage regionalised information? EGUsphere, 2024, 1-28. https://doi.org/10.5194/egusphere-2024-794Haaf, E., Giese, M., Heudorfer, B., Stahl, K., & Barthel, R. (2020). Physiographic and Climatic Controls on Regional Groundwater Dynamics. Water Resources Research, 56(10). https://doi.org/10.1029/2019wr026545Haaf, E., Giese, M., Reimann, T., & Barthel, R. (2023). Data‐Driven Estimation of Groundwater Level Time‐Series at Unmonitored Sites Using Comparative Regional Analysis. Water Resources Research, 59(7). https://doi.org/10.1029/2022wr033470Heudorfer, B., Liesch, T., & Broda, S. (2024). On the challenges of global entity-aware deep learning models for groundwater level prediction. Hydrol. Earth Syst. Sci., 28(3), 525-543. https://doi.org/10.5194/hess-28-525-2024Nolte, A., Haaf, E., Heudorfer, B., Bender, S., & Hartmann, J. (2024). Disentangling coastal groundwater level dynamics in a global dataset. Hydrol. Earth Syst. Sci., 28(5), 1215-1249. https://doi.org/10.5194/hess-28-1215-2024Rinderer, M., McGlynn, B. L., & van Meerveld, H. J. (2017). Groundwater similarity across a watershed derived from time-warped and flow-corrected time series. Water Resources Research, 53(5), 3921-3940. https://doi.org/10.1002/2016wr019856Zhao, F.-H., Huang, J., & Zhu, A. X. (2023). Spatial prediction of groundwater level change based on the Third Law of Geography. International Journal of Geographical Information Science, 37(10), 2129-2149. https://doi.org/10.1080/13658816.2023.2248215

Understanding groundwater dynamics is crucial for assessing groundwater resilience and supporting water management, particularly in transboundary areas where shared aquifers are often evaluated only at a national level, overlooking the broader aquifer system and data from neighboring countries. Groundwater resilience—the ability of groundwater systems to recover from disturbances such as droughts—is a spatially variable trait influenced by a range of spatial-temporal factors that do not obey borders. This study investigates the spatial and hydrodynamic controls on groundwater dynamics within the Baltic region, focusing on the challenges posed by data discrepancies and monitoring network inconsistencies across Latvia, Lithuania, and Estonia. We utilized groundwater timeseries indices and the groundwater memory effect to investigate dominant patterns and their correlations with physiographic and climatic controls. Machine learning algorithms were used to explore spatial patterns with similar hydrodynamic characteristics. The analysis of national groundwater level data revealed monitoring gaps, particularly in transboundary aquifers, along with different national approaches in groundwater monitoring networks. These challenges complicate the comprehensive assessment of groundwater dynamics at a regional scale.The results reveal that topographic, climatic and hydrological features are the most significant drivers of groundwater dynamics, followed by geological features. Groundwater indices and trends highlighted not only natural variability but also anthropogenic impacts on aquifer systems near large cities (e.g. Riga) and mining sites (e.g. Kohtla-Järve). We identify regions lacking monitoring wells and propose potential locations for new groundwater wells based on physiographic and hydrodynamic characteristics. This research is supported by the GRANDE-U “Groundwater Resilience Assessment through Integrated Data Exploration for Ukraine” (No. 2409395) project.

This study used groundwater memory timeseries as a proxy for groundwater drought vulnerability to unveil dominant patterns and their correlations with physiographic and climatic controls within hydrogeological systems of Baltic states. Clustering of groundwater memory timeseries was performed to identify regions with similar hydrodynamic systems, while random forest classification identified significant site-descriptive features, giving insights into system similarities and dissimilarities. Catchment characteristics showed the greatest importance followed by climate, topography and land use features. Moreover, spatial analysis of cluster distribution revealed feature combinations which are not covered by groundwater observations, presenting a challenge in understanding groundwater dynamics in these data-scarce locations. The study underscores the importance of considering not only locations with groundwater level data but also regions of typical feature patterns without monitoring infrastructure, thus identifying possible locations for new groundwater monitoring wells.

Ugh! - Haven’t we already fully explored the potential of spatial modeling of groundwater parameters? Haven’t we reached the limits there? In response to questions about whether we have fully explored the potential in these areas, we are launching this challenge to harness the swarm intelligence of the groundwater modeling community. Our aim is to dispel doubts and encourage creativity in feature development, model selection and training strategies. And why? Because regionalization, interpolation from point data into space, remains a crucial step in generating spatially continuous information and maps. And maps remain a crucial basis of information in sustainable water and resource management.For this purpose we provide a dataset of Nitrate-concentrations in approximately 1800 wells taken within a single month in southwestern Germany. The measured concentrations in the wells are representing concentrations in the most shallow aquifers. Besides Nitrate concentration as target variable, the dataset contains various features, describing the environmental and geological context of each sample site. Different types of models can be applied to model Nitrate concentrations, ranging from deterministic and geostatistical models to statistical/data-driven approaches such as machine learning models. We invite all interested researchers and data science enthusiasts to participate in this challenge as a team or single person. A ranking will be carried out while the challenge is open using a predefined set of model performance metrics on a secret test split. Further information on how to participate and the required data is available at https://groundwater-spatial-modeling-challenge.github.io/challenge2024/.The well defined rules of the challenge regarding the feature set, data splitting and metric choices, will allow the groundwater community to learn from different approaches and conduct a systematic comparison. The results of the challenge will be presented at the General Assembly of the EGU in 2025 and documented in a peer-reviewed paper with model contributors as co-authors on request. Through this challenge, we hope to increase the awareness in the groundwater community on the range of approaches available for (spatial) modeling of groundwater variables and their advantages and disadvantages.