We invite scientists working on:
• Geological field mapping and cross-boundary harmonization
• Mapping of extreme environments such as marine areas, polar regions, deserts, volcanic terrains, high-mountain ranges, and planetary surfaces
• 3D geological modelling in any geological context
• Development of geomodelling methodologies and tools
• Application of AI methods to geological mapping and modelling
• Development of geological information systems
• Remote sensing, geophysical techniques, drilling, sampling, and specialized tools for inaccessible terrains for geological mapping and modelling
The session aims to be highly transdisciplinary, bridging geology, geophysics, geochemistry, mineralogy, hydrogeology, engineering geology, and planetary sciences. By integrating approaches across diverse contexts, from accessible outcrops to remote and hostile terrains, participants will explore common challenges in data acquisition, interpretation, 3D modelling, visualization, and knowledge synthesis.
Outcomes are expected to benefit a wide range of applications and research, including geothermal energy exploration, offshore wind energy, geological risk assessment, groundwater protection, coastal protection, habitat mapping, environmental impact assessment, marine protected area development, mineral and resource exploration, and planetary missions. Ultimately, this session seeks to foster dialogue between communities tackling mapping and modelling challenges in both familiar and extreme environments, to advance scientific understanding and practical applications across the geosciences.
Orals: Fri, 8 May, 08:30–12:25 | Room -2.33
Geological mapping stands at a methodological crossroads. While traditional chronostratigraphic and lithostratigraphic approaches are effective at documenting observable rock patterns and temporal sequences, modern geological applications increasingly demand maps that directly relate to geological processes and events. The Lithotectonic Framework (LTF), developed within the GSEU project (grant 101075609), revisits lithotectonic concepts from the 1970s with the first rigorous theoretical framework. Complementing parallel European initiatives (doi.org/10.1051/bsgf/2022017; doi.org/10.31223/X5RT28), it organizes geological knowledge based on our understanding of Earth's history, rather than from observed rock age or lithological composition only.
The LTF's boundary-first principle defines geological units based on the events that created them, producing maps that reflect uniform geological histories. Consider the Paris Basin and North Sea Basin that are chronostratigraphically continuous, but lithotectonically distinct: the former is linked to post-Variscan subsidence, and the latter to Atlantic rifting. This event-based approach complements traditional mapping methods: chronostratigraphy provides robust temporal correlation, lithostratigraphy captures compositional variation, while LTF reveals the tectonic and sedimentary processes that shaped Europe's geology. The framework is equally applicable to polydeformed basement and sedimentary sequences, offering a systematic treatment of overprinting relationships through a hierarchical structure.
Beyond cartographic advantages, LTF's conceptual foundation unlocks transformative digital capabilities. By describing geology conceptually rather than descriptively, its hierarchical structure translates directly into semantic knowledge systems. Unlike traditional geological databases that catalogue and describe map features, LTF knowledge bases formally encode the theoretical relationships between geological entities. This enables dynamic visualizations, such as temporal "undressing" to expose deeper or earlier geological levels, thematic extraction for applied research, and crucially, machine-assisted geological reasoning. Preliminary testing demonstrates that LTF's conceptual structure enables AI systems to reason correctly about novel geological questions, outperforming geologists unfamiliar with the framework.
The paradigm shift is profound: geological mapping evolves from producing static maps with implicit knowledge to dynamic knowledge bases, where maps become interactive visualizations of deeper insights. Traditional geological mapping discovered that rocks form traceable patterns across continents, leading to the realization that geology records Earth's history. The LTF builds upon that foundation, introducing a self-organizing framework – where structure emerges from conceptual principles – that transforms geological knowledge from implicit expertise into explicit, queryable infrastructure. For Europe's geological community, this is not a replacement but an evolution: a digital geological infrastructure that preserves the strengths of traditional mapping while unlocking computational capabilities essential for modern Earth science applications.
How to cite: Piessens, K., Asch, K., Bernachot, I., Heckmann, P., Hintersberger, E., Krenmayr, H.-G., Le Bayon, B., Luth, S., Mancebo Mancebo, M. J., Mink, S., Padel, M., Pelech, O., Rodriguez, J., Rubio Pascual, F. J., Tulstrup, J., and Walstra, J.: From Observation to Understanding: The Lithotectonic Framework as Foundation for Europe's Digital Geological Infrastructure, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-595, https://doi.org/10.5194/egusphere-egu26-595, 2026.
In 2007, the INSPIRE Directive became a catalyst for re-examining fundamental geological data represented on maps. A major milestone was the OneGeology-Europe project (2008–2010), which for the first time approached 1:1,000,000-scale geological maps as structured datasets. With the involvement of nearly all EuroGeoSurveys member surveys, GeoSciML, the OGC standard for geological data exchange, was adopted and tested, providing feedback that helped consolidate the standard. In parallel, datasets were documented using metadata compliant with ISO 19115 and the INSPIRE metadata profile.
These initiatives encouraged geological surveys to intensify efforts towards the development of geological vector maps at larger scales. However, such work is highly time-consuming and labour-intensive, and despite significant progress over the years, substantial challenges and data gaps remain. To address them effectively, gaps need to be identified and assessed to provide a clear basis for coordinated action. Advances in geoscientific knowledge frequently require renewed field investigations and the revision of existing maps/data sets to improve the accuracy and quality.
The GSEU project aims to identify gaps not only in terms of missing data, but also with respect to completeness and consistency, the nature of attributes describing geological units, as well as issues of semantic and geometric harmonisation across map series. Such harmonisation challenges often reflect the evolution of scientific knowledge, classification schemes and mapping best practice over time.
A robust foundation, provided by fundamental geological maps ranging from continental-scale overview products such as IGME5000 to highly detailed maps depicting local geological structures, is essential for guiding future research and development. Geological maps form the foundation for a wide range of applied and scientific activities, including mineral resource exploration, geo-energy assessments, groundwater modelling, geoengineering, vulnerability assessments, spatial planning, and subsurface management.
This contribution presents initial assessments of the current state of geological data coverage across Europe and highlights the importance of comprehensive, harmonised and well-structured geological map databases for emerging applications, including artificial intelligence (AI) and large language models (LLM).
The GSEU project will also provide an organisational, technical and semantic framework for the digitisation, harmonisation and presentation of datasets describing Europe’s fundamental geology at multiple scales.
How to cite: Stępień, U., Krenmayr, H.-G., Asch, K., Heckmann, P., Piessens, K., Capova, D., Kramolisova, P., and Mancebo, M.: Geological Maps and Data Gaps Assessment: The Key factors for a Solid Geological Background, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8882, https://doi.org/10.5194/egusphere-egu26-8882, 2026.
This work seeks to encourage reflection and discussion on the ability and suitability of traditional classified geological maps to represent the full complexities of geology in the wild, and to consider why this is important to think about in order to serve 21st century geological mapping purposes.
The key components of traditional classified geological maps are boundary lines (in 2D), and boundary surfaces (in 3D); both of which must be ‘closed’ to form polygons or volumes representing the various classes of the map. These lines, polygons, surfaces, and volumes carry geological meaning, but what exactly?
The boundary lines that we traditionally construct geological maps from represent changes in geology, such that the geological properties on one side of the line should be different from the geological properties on the other. But, at any point along a drawn boundary line, which geological properties are changing, and by how much, and how sharply is this change occurring?
The line-based construction of the traditional classified geological map gives a restrictive view of geology. A line gives an on/off binary indication of a change in geological properties. Are we to believe that the change in geological properties is equal at all points along the perimeter of any geological polygon? Logically, the magnitude of change in geological properties (perhaps assume the sum of magnitudes of change for all properties, but it could also be for an individual property) must have a maximum somewhere along the perimeter of the polygon – perhaps this is the point that is most deserving of being represented by the line, but does the entire perimeter deserve to be represented by that line?
The use of a line to indicate a boundary also implies infinite sharpness; that the change in geological properties is instantaneous on crossing the line. Whilst this may be appropriate for faults and unconformities, lines leave us unable to fairly represent the many gradational processes that are inherent to the geological system, examples of which include partial melting, fractional crystallisation, gradational sediment deposition and diagenesis.
So where do these limitations of traditional line-and-polygon based geological mapping leave us? Representing geology in its true complexity requires mapping the individual geological properties themselves through space rather than only delineating where they significantly and collectively change. If we map the geological properties as a collection of scalar fields (as in implicit geological modelling), then all changes – big and small – for all properties are revealed in the magnitude of their gradients. Correspondingly, it appears that traditional hand-drawn geological maps attempt to approximate the sum of the magnitude of the gradients of commonly considered geological properties (age, composition, texture), albeit with a thresholded presentation owing to the line-based approach (the line doesn’t have an intensity, it either is or is not) and some necessary inconsistencies to enable polygon closure. When we consider these points, going beyond the traditional classified geological map seems crucial for progressing the completeness of our geological knowledge in the 21st century.
How to cite: Kirkwood, C.: Geological boundary dispute: reflecting on the ability of the traditional classified geological map to fully represent geology, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21401, https://doi.org/10.5194/egusphere-egu26-21401, 2026.
When large areas of the UK were mapped over 100 years ago priority was given to identification of mineral resources. Many such ‘drift’ maps therefore are not consistent with modern scientific understanding, nor do they reflect current stakeholder interests. Surface and groundwater flooding represent a major hazard to homes, infrastructure, and land management across the Tweed catchment. Recent work by BGS Groundwater has indicated that slope deposits are far more widespread than previously identified and play a significant role in groundwater connectivity. Updating the superficial geology map across the ~5000 km² catchment is therefore critical for improving flood forecasting, and the design of a major baseline monitoring project, the Flood-Drought Research Infrastructure funded by NERC.
The Tweed Mapping Project applies spatial Random Forest models using DTM derivatives at 25 m resolution to predict twelve different deposit classes (e.g. till, alluvium, regolith, talus). Model training data are derived from detailed mapping surveys dated 2005, 2009 and 2012.
Initial results indicate that slope deposits have been under-mapped, with till being the dominant deposit predicted. Both over and under-sampling are a significant issue; sample adjustment methods are unable to compensate. Minor deposits are therefore under-represented in model outputs.
Model outputs have been checked in the field in Cheviot, Tweedsmuir and Galashiels areas during 2025. Geomorphological mapping, section logging, and bulk sampling of deposits are being used to provide up-to-date training data to enable more reliable and accurate model predictions. Outstanding issues include: (i) the absence of LiDAR data away from major river channels and settlements, (ii) over-representation of specific field observations, and (iii) limited geomorphological inputs to the model.
How to cite: Roberson, S.: The Tweed Mapping Project: machine learning methods for rapid Quaternary mapping, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14282, https://doi.org/10.5194/egusphere-egu26-14282, 2026.
The main objective of the study was to check how popular, free versions of AI chatbots cope with questions related to lithology. The assumption of the study was that a potential user is not a geologist, does not know how to formulate prompts correctly, and is sceptical enough about new technologies that they avoid logging in. Lithological issues may occur, for example, in descriptions of educational paths. The entire study was conducted in Polish. In order to shorten the study time, all prompts were formulated, and their order was imposed. The aim was, among other things, to see how the answer would differ depending on how precise the question was. In addition, the prompts were deliberately designed not to comply with the rules for asking questions, as we assumed that potential users would lack such knowledge. We asked people with geological knowledge to participate in the study so that they could assess its substantive value after receiving the results.
The rapid expansion of large language models (LLMs) into scientific workflows raises important questions concerning their reliability, transparency, and suitability for specialised disciplines such as the geosciences. This contribution presents the results of a survey-based assessment of selected AI-powered tools conducted in Polish between February and May 2025. The study involved 202 respondents, including professional geologists, academic staff, and students of geosciences, who evaluated AI-generated responses to seven tasks of varying complexity.
The study confirmed that the precise formulation of queries, especially those specifying source requirements and an expert-level perspective, substantially improves the quality of AI-generated content. This effect was particularly evident in questions involving linguistically ambiguous terms, where models often addressed only one interpretation while omitting alternative meanings relevant to geological sciences. Such omissions may result in incomplete or misleading answers when the user lacks sufficient domain knowledge to identify inaccuracies.
The opinions expressed in the Polish-language survey present an ambivalent picture. While the functional benefits and efficiency gains offered by AI tools are widely recognised, substantial methodological, substantive, and ethical limitations remain. The competence and awareness of the user have been identified as pivotal factors in determining whether the adoption of AI results in the creation of genuine value or the dissemination of errors and misinformation. The study emphasises the necessity for enhanced citation practices, the prioritisation of peer-reviewed literature, an augmentation in the number of high-quality non-English open geological publications, an enhancement in the semantic understanding of specialised terminology, and the development of regionally adapted language models. These measures are considered essential for ensuring transparent, reliable, and responsible use of AI in geoscientific research and communication.
How to cite: Stępień, U., Zaszewski, D., Fronczak, A., and Witkowski, W.: In search of lithological truth – sceptical non-geologists in the non-English speaking world, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6989, https://doi.org/10.5194/egusphere-egu26-6989, 2026.
Anthropogenic deposits are widespread in various environments. Some consist of displaced natural materials, and others of anthropogenic (human-made) materials, or they contain a mixture of both. Human-made materials include demolition materials (such as concrete), industrial waste and by-products (e.g., slags), mining residues, and domestic waste. Excavated soils and dredged sediments are examples of displaced natural materials. Anthropogenic deposits can be linked to hazards like geotechnical instability and contamination, potentially resulting in health and environmental risks (e.g., to soil, water, biodiversity, stable site foundation) with associated economic, legal, and social impacts. On the other hand, some deposits can represent valuable resources. Former mining or urban deposits, for example, may contain extractable amounts of critical raw materials (CRM). They may also be reused during land development or hold geoheritage value, such as in the case of prehistoric burial constructions. However, our knowledge of anthropogenic deposits is still poor. Improving their representation in geological maps and models is therefore crucial. Against this background, the European GSEU project is developing a set of coordinated vocabularies to standardise the describtion of anthropogenic deposits.
Existing national and international vocabularies and definitions were collected and compiled into a comprehensive list. In parallel, a conceptual data model was developed as a basis to systematically organise and classify the terms. This allowed establishing hierarchical lists of terms to structure the vocabularies and provide space for additional information on anthropogenic deposits, such as their purpose and geometry. A coherence and consistency check between the various vocabulary lists was conducted to ensure alignment across all terms. Real-world examples (use cases) of anthropogenic deposits were used to test the effectiveness and relevance of the vocabularies.
A “lithology-based” approach was chosen to describe anthropogenic deposits. The terms for displaced natural materials originate from the lithology vocabulary, which is being compiled in parallel within the GSEU project. For human-made materials an existing classification from materials science is used, with some adaptations and additions. The set of vocabularies includes additional attribute lists linked to the origin of the materials present in the deposit, the original purpose of the deposit, the shape of the deposit, as well as its environment (natural, anthropic). The selected use cases cover various situations (former landfill, redevelopment area, archaeological site, mine tailing, industrial residue, reclaimed land) in several environments (urban, rural, mining, industrial, coastal and fluvial environment). The associated environmental and social issues include sanitary aspects linked to soil pollution, surficial and groundwater quality, geotechnical stability (vulnerability to collapse, landslide, ground subsidence, erosion, etc.), and cultural heritage.
The developed scientific vocabularies dedicated to anthropogenic deposits are designed for use with multiscale spatial geological datasets in both 2D and 3D formats. These can be integrated within geological maps and 3D models to support various applications, such as spatial planning, area development, resource extraction, and risk management. The final hierarchical lists of terms will be delivered for implementation in EGDI, the European platform to share, integrate and access geological data.
How to cite: Le Guern, C., Schokker, J., Stępień, U., Walstra, J., Heckmann, P., Asch, K., and Krenmayr, H.-G.: An international vocabulary for anthropogenic deposits to improve geological mapping and modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16704, https://doi.org/10.5194/egusphere-egu26-16704, 2026.
Understanding the structural and stratigraphic connection between the exposed sectors of collisional belts and their external thrust front buried beneath foreland basin sediments remains a long-debated issue, largely due to the need for integration of heterogeneous surface and subsurface datasets. This research focuses on the Northern Apennines front, at the transition between the exposed Oltrepo Pavese hillslopes and the buried thrust front beneath the Po Plain, investigated within the Italian Geological Mapping Project (CARG, Pavia Sheet - 160). Detailed field mapping resulted in an original 1:10.000 scale geological map of the exposed belt, supported by petrographic and biostratigraphic analyses. These surface data were integrated with seismic profiles and deep well data, from a regional 3D model of the central Po Plain, which reconstruct the geometry of the buried thrust system displacing the Plio-Pleistocene sequence in the Po Plain, in order to link the exposed and buried stratigraphic units. The integration of surface and subsurface data allows the recognition of structures otherwise hided beneath vegetation and Quaternary colluvial covers in the hillslopes. In particular, seismic interpretation allows to localize buried structures and constrains their geometric relationships with respect to the attitude of the outcropping units.
The stratigraphic record in the exposed area includes Paleocene to Lower Eocene turbiditic succession of the Val Luretta Formation, which thrusts over a Tortonian to Piacentian sequence. This latter records a progressive shallowing upward trend in the environment, from a deep-sea setting to shallow marine, deltaic and then continental environments associated with the Messinian Salinity Crisis, followed by the Pliocene marine transgression.
Interpretation of subsurface dataset allows the recognition of a south-dipping, north-verging thrust system affecting both the exposed and the buried units, with multiple splays and blind thrusts active until the Lower Pleistocene.
These results provide new constraints on the geometry and the evolution of this sector of the Northern Apennines front, demonstrating the effectiveness of combining field-based geological data with subsurface data to link outcropping and buried portions of orogenic belts
How to cite: Bellotti, P., Stendardi, F., Barrera, D., Di Giulio, A., and Toscani, G.: Unravelling the tectono-stratigraphic link of buried and exposed structural fronts of the Northern Apennines through integrated geological mapping (CARG Sheet 160 Pavia, Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6514, https://doi.org/10.5194/egusphere-egu26-6514, 2026.
In 2024, Geosphere Austria initiated the project EAGLe (Establishing the Austrian General Geological Legend) to develop a harmonized nationwide geological dataset at a scale of 1:50,000 by the end of 2026. The primary objective is the creation of a hierarchically structured general legend by standardizing and harmonising the lithostratigraphic terms that are used in the different map sheets. This work is carried out by regional teams with varying starting conditions: The Quaternary and Neogene teams relied on already existing comprehensive lists, such as general legend only for Quaternary lithogenetic and geomorphological terms and the stratigraphic chart description for the Cenozoic eratherm. On the other hand, for the regions with basement rocks at the surface (such as the Tauern Window and the Bohemian Massif), regional teams faced the additional challenge of establishing coherent concepts for the lithostratigraphic and lithodemic terms in the respective regions. In some cases, legend descriptions —particularly from older maps—are either ambiguous or significantly outdated, yet they represent the only available information for certain geological units. Without field surveys, these entries can only be assigned to very general geological units. A comprehensive revision and mapping of all legend descriptions is therefore not feasible at this stage; consequently, the original legend descriptions will be included in the final dataset to ensure transparency.
The data base for this compilation consists of over 25000 legend descriptions from published geological map sheets at the scale of 1:50,000, added by GeoFAST maps at the same scale (maps compiled from selected archival material without additional fieldwork), as well as regional maps, partly at a scale of 1:25,000. However, the corresponding vector datasets exhibit considerable heterogeneity in both geological content and data structure. In some cases—particularly for older maps—vector data are entirely absent. Therefore, the second major objective is to consolidate these different datasets into a unified structure and to digitize older analogue maps to close existing digital gaps. It should be noted that this initial version will not include any geometric adjustments (e.g., correction of “sheet boundary faults”).
The first version of the integrated dataset, incorporating the preliminary general legend as far as possible, will be published on the Tethys Data Repository (www.tethys.at) by the end of 2026 and will be made publicly accessible via the Geosphere Austria web service (www.maps.geosphere.at). An additional metadata layer will provide information on the quality of the underlying published sources.
How to cite: Hintersberger, E. and Kettler, C. and the EAGLe-Team: Establishing the Austrian General Geological Legend (EAGLe), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20772, https://doi.org/10.5194/egusphere-egu26-20772, 2026.
Three-dimensional geologic modeling is a well-established technique developed in the last twenty years and currently applied in terrestrial mining, environmental management, and hydrogeology [1,2]. It also represents a critical frontier for planetary exploration, from fundamental scientific research and the search for subsurface life to operational applications such as mission planning or In-Situ Resource Utilization (ISRU). As missions increasingly target the shallow subsurface of the Moon and Mars, reconstructing subsurface architectures from available observations has become essential.
The primary challenge in planetary subsurface modeling lies in the extreme scarcity of direct subsurface data compared to the abundance of orbital remote sensing observations. Consequently, geologic mapping becomes the foundational prerequisite, providing the primary spatial and qualitative data needed to interpolate and propagate geologic contacts through three-dimensional volumes.
This work explores modeling approaches through experiments designed to test their applicability to planetary science. These include a volumetric model of Tempe Terra on Mars based solely on geological map information, and a benchmark study of a terrestrial impact crater using sparse drilling data to define the contact between bedrock and impact ejecta. Key findings relate to uncertainty evaluation and the importance of defining modeling objectives, which directly affect model complexity.
This research emphasizes avoiding "black box" solutions by adopting Free and Open Source Software workflows to ensure interoperability, traceability, and reproducibility—critical requirements in the demanding operational context of space exploration. Current results and modeling environments are promising for extraterrestrial applications. By integrating scientific reasoning with advanced interpolation algorithms, three-dimensional geologic modeling can generate robust predictive models essential for planning future robotic and human exploration missions.
References: [1] P. Calcagno et al. en. In: Physics of the Earth and Planetary Interiors 171.1-4 (Dec. 2008), pp. 147–157. [2] F. Wellmann and G. Caumon. In: Advances in Geophysics. Vol. 59. Elsevier, 2018, pp. 1–121.
Acknowledgements: This study is carried out within the Space It Up project funded by the Italian Space Agency, ASI, and the Ministry of University and Research, MUR, under contract n. 2024-5-E.0 - CUP n. I53D24000060005.
How to cite: Frigeri, A.: Three-Dimensional Geologic Modelling Beyond Earth: Challenges and Perspectives, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8462, https://doi.org/10.5194/egusphere-egu26-8462, 2026.
The precise global localization of the Mars rover serves as a fundamental prerequisite for long‑distance scientific traverses and in‑situ geological investigation. As Mars represents a typical GNSS‑denied environment, accurate positioning is typically accomplished through the registration of rover‑acquired imagery with orbital maps. Mainstream methodologies address the substantial perspective and scale differences between ground‑level and orbital images by first generating orthophotos from rover imagery, which are then aligned with satellite‑based imagery for localization.
The successful deployment of the Mars Helicopter (Ingenuity) enables the use of acquired UAV imagery as an intermediate bridge for the rapid and accurate global localization of the Perseverance rover. Accordingly, this study proposes an orbiter-UAV-rover collaborative matching framework, as illustrated in Fig.1. This framework sequentially performs three core steps: (1) cross-perspective matching between rover and UAV imagery, (2) cross-scale matching between UAV and orbiter imagery, and (3) a matching connection strategy that integrates the two matching sets to establish a continuous geometric transformation chain.
Figure 1. Schematic diagram of the proposed global localization framework.
Specifically, the rover-UAV image matching procedure is implemented through the following sequential steps, and the efficacy of this approach is demonstrated in Fig. 2.
(1) Horizon-based Pose Estimation: The visual horizon within the rover image is segmented using a Mask R-CNN model. This horizon line is then analytically processed to derive the pitch and roll angles of the rover camera.
(2) Cross-Perspective Image Rectification and Matching: Leveraging the estimated orientation angles, the rover image is orthographically rectified to approximate a nadir view, thereby aligning its perspective with that of the UAV imagery. A deep learning-based feature matching network is subsequently applied between the rectified rover image and the UAV image to establish dense, pixel-wise correspondences.
(3) Correspondence Projection: The matched feature points from the rectified image pair are back-projected onto their original coordinates in the raw rover image.
Figure 2. Comparison of cross-view feature matching results before and after orthographic rectification.
Following the establishment of correspondences between rover and UAV imagery, the matching results between the UAV and orbital data are subsequently derived using our previously proposed method [1]. This process culminates in the formation of a two-tier correspondence chain, effectively linking the rover, UAV, and orbiter, as visually summarized in Fig. 3.
Figure 3. Visualization of cross-platform feature matching results.
Figure 4. Results of collaborative matching and localization.
Table 1. Localization error of the Perseverance rover for different sites.
Localization experiments were conducted at multiple sites along the Perseverance rover's traverse. As shown in Fig. 4 and Table 1, multi-platform images were well-associated, achieving an average accuracy of 0.4 m (resolution of the orbital image is 0.25m). High-precision rover positioning information enables the precise fusion of multi-site local geological mapping products and ensures the accurate integration of rover and orbital-scale geological mapping products.
Reference:
[1] CAO Z, FU H, XU X, et al. A Novel Template Matching Method Incorporating a Multi-Candidate Region Optimization Strategy for the Initial Localization of Mars Helicopter. Transactions in GIS, 2025, 29(2): e70052.
How to cite: Cao, Z., Xu, X., Chen, Q., Xiao, C., Wang, C., Feng, Y., Xie, H., and Tong, X.: Global Localization of the Perseverance Rover via Orbiter-UAV-Rover Collaborative Matching, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2291, https://doi.org/10.5194/egusphere-egu26-2291, 2026.
Accurate characterisation of planetary surface topography and reflectance at metre and sub-metre scales is critical for geological interpretation, understanding regolith processes, and supporting surface exploration. We present LUMOS (LUminosity-constrained Multi-angular Observation Super-resolution), a physics-based framework for the joint reconstruction of super-resolution digital elevation models (DEMs), spatially varying surface reflectance, and uncertainty estimates from multi-angular orbital imagery. The method overcomes key limitations of classical shape-from-shading approaches, which typically assume Lambertian reflectance and provide no uncertainty quantification.
Figure 1 Area of the reconstructed terrain centred on the Apollo 15 landing site.
(a) LOLA elevation map at its native resolution. (b) LUMOS-derived DEM shown in nadir view.
(c,d) Oblique views of the LUMOS DEM.
LUMOS formulates surface reconstruction as a Bayesian inverse problem that explicitly couples topography and photometry. Observed radiance is modelled using a non-Lambertian, kernel-driven bidirectional reflectance distribution function (BRDF), adopting the Ross–Thick Li–Sparse (RTLS) formulation to represent isotropic, volumetric, and geometric scattering effects. This enables physically consistent treatment of anisotropic regolith scattering, shadowing, and viewing-geometry dependence. A low-resolution laser altimetry DEM is incorporated as a prior to constrain long-wavelength topography, while fine-scale surface structure is recovered from photometric variations across multiple illumination and viewing angles. The coupled system is solved efficiently using a Sylvester-equation-based formulation, avoiding empirical tuning parameters and allowing uncertainties in image radiance and prior information to propagate into the final products.
Figure 2 Slope uncertainty map. Uncertainty increases in shadowed regions and where viewing geometry is limited.
We demonstrate LUMOS using multi-angular LROC NAC observations of the Apollo 15 landing site. The reconstructed DEM achieves a spatial resolution of 0.53 m/pixel, corresponding to the native resolution of the NAC imagery and representing more than a two order of magnitude increase in sampling density relative to the Lunar Orbiter Laser Altimeter (LOLA) prior. Large-area comparisons show that the LUMOS DEM preserves consistency with LOLA-derived long wavelength trends while resolving fine scale morphological features, including small craters, subtle relief variations, and local undulations unresolved in altimetric data. Detailed views further illustrate surface continuity and the absence of illumination correlated artefacts.
Beyond elevation, LUMOS retrieves spatially resolved reflectance parameters and provides pixel-wise uncertainty estimates for both elevation and slope. Derived slope maps reveal metre-scale variations sensitive to reflectance modelling assumptions, with Lambertian-based reconstructions exhibiting systematic biases relative to the RTLS solution. These differences have implications not only for operational assessments, such as landing-site hazard evaluation, but also for scientific interpretation of small-scale morphology, regolith roughness, and slope-controlled geological processes.
The LUMOS framework is constrained primarily by observational resolution rather than algorithmic limitations. While the present results are bounded by the resolution of available NAC data, the methodology directly benefits from higher-resolution, multi-angular observations. As such, LUMOS constitutes a cornerstone of the ESA Máni mission (Phase A), which aims to acquire dense multi-angular imagery at spatial resolutions of approximately 0.17–0.2 m/pixel. Applied to Máni data, LUMOS is expected to further enhance topographic fidelity, reflectance characterisation, and uncertainty-aware surface mapping.
How to cite: Fernandes, I., Mosegaard, K., and Schmidt, F.: Physics-Informed Joint Super-Resolution Topography and Reflectance Inversion From Multi-Angular Planetary Imagery — The LUMOS Framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15157, https://doi.org/10.5194/egusphere-egu26-15157, 2026.
The shallow water nearshore area is often referred to as the white ribbon due to a low density or even a lack of data in this transition zone between land and sea. Historically, it was challenging to generate detailed 3D maps in this transition zone with the available technologies. However, emerging technologies during the last decade such as airborne lidar bathymetry (ALB) has enabled full-coverage, high-resolution seabed mapping in such environments (e.g. Andersen et al., 2017).
Seabed mapping in the shallow water coastal zone is paramount in relation to a wide spectrum of societal challenges, e.g. climate change adaptation with coastal protection in relation to storm surges and sea level rise, green energy transition with connection of offshore windfarms to land, nature restoration and protection for preserving or enhancing nature and biodiversity, and safety of critical infrastructure in nearshore areas.
We present examples of and experiences from national seabed mapping projects combining airborne lidar bathymetry and RGB imaging with ROV video imaging and seabed sampling for mapping seabed morphology, substrates and habitats in shallow water nearshore areas in Danish waters.
We demonstrate the potential of applying a combination of platforms (airborne, vessel borne and underwater) and instruments (optical and acoustical) in a multiscale remote sensing approach to acquire composite datasets tailored for seabed nature mapping in shallow water nearshore areas – filling in the white ribbon.
References
Andersen MS, Gergely A, Al-Hamdani Z, Steinbacher F, Larsen LR, Ernstsen VB (2017). Processing and performance of topobathymetric lidar data for geomorphometric and morphological classification in a high-energy tidal environment. Hydrology and Earth System Sciences, 21: 43-63, DOI: 10.5194/hess-21-43-2017.
How to cite: Ernstsen, V. B., Andersen, M. S., Hansen, L. Ø., Larsen, I. R., Nielsen, N. L. J., Blok, C. N., and Al-Hamdani, Z.: Filling in the white ribbon – Airborne lidar bathymetry and RGB imaging in combination with ROV video imaging and seabed sampling for seabed nature mapping in the coastal zone (Danish waters), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19869, https://doi.org/10.5194/egusphere-egu26-19869, 2026.
Within the framework of Portuguese policy for the energy transition and economy decarbonisation, the Portuguese Institute for the Sea and the Atmosphere (IPMA) is carrying out project RP-C21-i07.01 – Technical studies for offshore energy potential. This project, funded with 42 M€ by the European Recovery and Resilience Plan, through the component C21-REPOWEREU of the Climate Transition dimension, aims to support Portugal’s ambitions regarding energy independence and ecological transition, in the context of new geopolitical and energy market challenges.
Led by the Marine Geology and Geophysics Laboratory (SEISLAB) team at IPMA, the projects is developing studies to provide detailed data on the geological, geophysical and geotechnical properties of the seafloor, as well to define an environmental baseline. The main objective is to support offshore wind farm developers regarding engineering and financial planning, thereby providing the basis for launching auctions in offshore areas designated for windfarm development in the Portuguese Allocation Plan for Offshore Renewable Energy (PAER).
This project started in early 2024, has a duration of 2.5 years and focuses on surveys in the PAER areas of Leixões and Figueira da Foz, totalling approximately 2000 km2, located offshore the western Portuguese mainland coast, at water depths ranging from 120 m to 530 m.
Hydrographic and geophysical survey methodologies included multibeam echosounder (MBES), side scan sonar (SSS), magnetometer (MAG), two sub-bottom profilers (SBP) and multichannel ultra high-resolution seismic (UHRS) reflection data. Geotechnical methodologies included cone penetrating tests (CPT) and sedimentological and physical properties of sediments recollected with grabs and Vibrocoring (VC).
Preliminary works conducted in 2024 included desktop studies and exploratory surveys with the acquisition of approximately 2000 km of geophysical data (MBES, SBP, UHRS). Survey activities carried out in 2025 involved the acquisition of circa 15000 km of geophysical data (MBES, SSS, MAG, SBP, UHRS), 122 grabs samples, 71 VCs and 43 CPTs.
Seafloor surface characterisation relied on cartographic products derived from the MBES and SSS mosaic datasets, as well as on the identification of outcropping units from the seismo stratigraphic model calibrated with the geotechnical data. Seafloor features, including landforms and contacts were interpreted from the MBES and SSS data and validated against magnetic anomalies. These included anthropogenic features like shipwrecks, trawl marks and lost objects (e.g. fishing gear) and geological features like sorted bedforms, boulders, sinkholes and outcrops.
Sub-seafloor seismic data reveal a complex geological framework associated with the rifted margin and orogenic units. The upper units are dominated by unconsolidated sediments and polyphase channel complex events associated with sea level variations, while the lower units frequently display mass-transport deposits, extending for tens of kilometres, tectonic deformation and faulting.
Environmental analysis are based on water and sediment analytical work and on the characterisation of species communities, aiming to establish the biodiversity baseline and assess the environmental condition. Surveys were conducted in compliance with the Joint Nature Conservation Committee guidelines.
The thematic cartography resulting from these pioneering and unprecedented studies in Portugal constitutes a key asset for the development of the floating offshore wind industry, supporting the ongoing Portuguese energy transition
How to cite: Brito, P., Abrantes, F., Aires, C., Almeida, J., Batista, L., Borges, R., Costa, P., Drago, T., Neres, M., Magalhães, V., Noiva, J., Oliveira, D., Pereira, Â., Ribeiro, C., Rosa, M., Salgueiro, E., Silva, A., Trindade, L., Valadares, V., and Terrinha, P. and the PRR-RP-C21-i07.01 Team: Technical studies for offshore energy potential, geological and environmental mapping towards support of windfarm developers' decisions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21538, https://doi.org/10.5194/egusphere-egu26-21538, 2026.
From 2024 to 2030, the Swiss Geological Survey (swisstopo) leads the Swiss Alps 3D (SA3D) project as part of the swisstopo National Geological Model (NGM) program. The project brings together eight modelling and research teams from several universities with the objective to develop one coherent, large-scale 3D geological model of the Swiss Alps subsurface. The model targets the major structural and lithostratigraphic boundaries of the region and will serve as a regional geological reference framework for future high-resolution studies. It will support a wide range of applications, including infrastructure planning, groundwater management, georesource assessment, natural hazard analysis, as well as education and research.
This contribution presents results from the first two years of SA3D modelling in the Subalpine Molasse, Prealps, Helvetic, and Western Penninic tectonic domains, with emphasis on practical solutions developed to address key methodological challenges. The SA3D models are structured around four core components: (i) input datasets, (ii) 2D geological maps, (iii) reference cross-sections, and (iv) 3D meshes. Ensuring internal consistency among these elements, both at the surface and at depth, represents a primary challenge. This challenge is amplified by sparse subsurface data, limited seismic profiles and boreholes, the large extent of the study area, and the extreme structural complexity of the Alpine Orogen. These constraints limit the range of applicable modelling approaches (implicit versus explicit) and require rigorous integration of all components. Coordinating eight independent projects to produce a unified, technically and conceptually consistent model demands close collaboration and methodological harmonization across the different modelling teams.
By addressing these challenges, SA3D provides unprecedented insight into the largely unexplored Alpine subsurface. Reconstruction of the three-dimensional network of lithostratigraphic contacts and structures reveals large-scale structural and lithological patterns down to depths of 10 km, significantly improving our understanding of regional tectonic evolution. Beyond the resulting 3D model and its scientific outcomes, SA3D promotes a collaborative community of Alpine geologists and 3D geological modellers, setting the stage for continued for continued high-level research and exploration of the Alpine subsurface.
How to cite: Musso Piantelli, F., Kurmann, E., Nibourel, L., Wehrens, P., Baland, P., and Müller, H. R.: Building a large-scale 3D geological model of the Swiss Alps: First results, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4801, https://doi.org/10.5194/egusphere-egu26-4801, 2026.
How to cite: Wellmann, F. and de la Varga, M.: Differentiable Geomodelling: Towards Geomodel Insight and Task-Oriented Sensitivity Analysis, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7588, https://doi.org/10.5194/egusphere-egu26-7588, 2026.
At the Federal Institute for Geosciences and Natural Resources we develop a wide variety of 3D-Models of the subsurface. These models range from basin-wide structural models to small scale models of an artificial fracture.
In many cases it is important to present these 3D-Models to stakeholders or the general public. One big challenge lies in the fact that many of the spectators are not professionals in geology. Therefore, these complex 3D-Models have to be presented in a way non-professionals can easy access and understand.
Visualizing data and models in real 3D is not only very helpful in communicating our models to the general public. It can also be very helpful during the creation of 3D-model itself. Especially in very complex models, parts of the model may obstruct the view to other parts of the model. Seeing the 3D-model in real 3D provides the modeler with a better and easier impression of complex structures in the subsurface and allows
Experience has shown, that there is not one best way of visualizing 3D-data. In contrary, the 3D-visualisation has to be chosen and adapted not only for every model, but also for every target audience.
We present several methods of 3D-visualisation, ranging from 3D-Projectors and Virtual Reality over gamification (transferring 3D-models into computer games) to using 3D-printers. For each method we will present an application and evaluate the main advances and disadvantages.
How to cite: Steuer, S.: Look at it! – Visualizing 3D geological data in real 3D, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21613, https://doi.org/10.5194/egusphere-egu26-21613, 2026.
Geological mapping and 3D subsurface modelling require consistent geological interpretations across large datasets with heterogeneous spatial coverage and information density. In the Netherlands, several subsurface models rely heavily on borehole lithological descriptions to map lithostratigraphic units and geological structures. Automated interpretation approaches based on machine learning (ML) are being developed to transfer expert geological interpretations to previously unseen boreholes, thereby increasing the number of interpreted boreholes that can be incorporated into subsurface models. However, existing neural network-based approaches for borehole interpretation often struggle to consistently respect the stratigraphic and spatial relationships derived from expert geological knowledge. In practice, automated interpretations can produce stratigraphically inconsistent successions, with younger units incorrectly predicted to occur below older ones, or units appearing outside their known regional extent. This limitation stems from ML training objectives that prioritise local classification accuracy (e.g., categorical cross-entropy loss) over regional geological plausibility.
To improve the geological plausibility of ML-generated interpretations, we introduce geology-informed loss functions that account for stratigraphic consistency and the spatial extent of lithostratigraphic units. The proposed loss functions are combined with a standard classification loss during model training on expert-interpreted boreholes and evaluated on previously unseen boreholes drawn from the same national dataset, comprising 7,500 boreholes in total. By varying the relative weight of each loss function during model training, we found that ML models trained with a combination of geology-informed loss functions and standard categorical cross-entropy substantially reduce geologically implausible stratigraphic transitions, increasing the proportion of stratigraphically consistent transitions from approximately 90% to up to 95%, and making fewer predictions of lithostratigraphic units outside their known regional extent. These improvements in geological plausibility do not lead to a noticeable change in overall classification accuracy (≈ 75% across different loss-weight combinations). Incorporating geology-informed training objectives, therefore, provides a practical way to improve the plausibility and consistency of automated borehole interpretations used in large-scale subsurface modelling workflows.
How to cite: Garzón, S., Dabekaussen, W., De Boever, E., Busschers, F., Mehrkanoon, S., and Karssenberg, D.: Integrating geology-informed constraints into machine learning–based borehole interpretations for subsurface modelling: A case study from the Netherlands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14705, https://doi.org/10.5194/egusphere-egu26-14705, 2026.
Exploration datasets such as borehole logs and geophysical profiles form the fundamental basis of geological modeling. Among these, borehole records are particularly influential, as they typically include detailed descriptions and interpretations of petrography and stratigraphy. Such information is essential for constructing three-dimensional representations of lithostratigraphic units, which can be affected by inconsistencies or errors skewing borehole interpretations. Distinguishing reliable borehole data from problematic records is therefore critical, but becomes increasingly challenging when dealing with large datasets. Although visual assessment of the resulting geological models can help identify questionable boreholes, this approach typically requires many iterative modeling steps, making the process inefficient and costly.
To improve the efficiency of borehole data quality assessment, we developed B-QualMT, a Python-based borehole quality management tool with a GUI interface that enables automated filtering of borehole records using both a user-defined quality check as well as a purely data-driven approach. The software applies a suite of deterministic tests that incorporate auxiliary information such as existing 3D geological models and regional geological knowledge, including expected stratigraphic successions, to identify anomalous borehole logs within geologically similar areas. Furthermore, spatial outliers can be identified using a combination of borehole similarity analysis, various clustering techniques, and a Bayesian-based novelty detection system. To evaluate the functionalities and edge cases of these methods, synthetic borehole data besides real borehole data were used. Different test scenarios were utilized to systematically control and test the outlier detection approaches, enabling workflow optimization and a detailed assessment of their performance, limitations, and sensitivity under controlled synthetic conditions. The limitations identified during testing with synthetic data are subsequently used to inform and improve the interpretation of results derived from more complex real borehole logs.
How to cite: Schönfeldt, E., Hiller, T., and Giese, J.: How to find the baddies - a borehole quality management and outlier detection software for 3D-model data selection, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9273, https://doi.org/10.5194/egusphere-egu26-9273, 2026.
Building implicit 3D geological models requires the detailed integration of diverse data sources, including legacy drill logs, technical reports, and stratigraphic descriptions. While this process is fundamental to understanding the subsurface, the manual translation of unstructured text into quantitative model inputs is a time-intensive task. Large Language Models (LLMs) offer promising capabilities to assist in processing data presented as text, but their application requires rigorous control to ensure geological validity. We present an ongoing research project developing a Human-in-the-Loop (HITL) workflow that leverages uses a collaborative human-AI approach to structure raw descriptions into inputs that will be used for implicit modeling.
The proposed workflow grounds the LLM in a formal Axiom-Based reasoning framework designed to minimize hallucinations and ensure consistency. The process begins with entity extraction, where the LLM parses depths and lithological descriptions from raw logs, followed by an axiomatic reasoning phase where units are categorized based on standardized rules (e.g., the Lithotectonic Framework). Crucially, the workflow integrates a dedicated validation interfaces that empowers geologists to go beyond simple verification. Experts use this environment to contextualize interpretations, test different stratigraphic hypotheses, and inject external knowledge such as fault definitions or regional correlations, before the structured output is finalized. This effectively translates text into the specific geometric parameters and interface points required to initialize the GemPy modeling engine.
We are applying this workflow to legacy data from the Campine Basin. The objective is to demonstrate how AI can function as a reliable assistant for data structuring, potentially reducing the time required for model initialization. Our workflow shifts the priority from slow data processing to critical validation; we aim to allow geologists to focus more on conceptual definitions and uncertainty analysis rather than data management. Ultimately, this research seeks to facilitate the creation of self-updating geological models that can continuously ingest and interpret new textual data as it becomes available.
How to cite: Welkenhuysen, K., Rodriguez, J. D., and Piessens, K.: From Unstructured Geological Data to 3D Models: A Human-in-the-Loop LLM assisted Workflow for Automated Geological Model Building, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10918, https://doi.org/10.5194/egusphere-egu26-10918, 2026.
The Einstein Telescope (ET) research infrastructure is envisioned as Europe’s pioneering next-generation underground observatory for gravitational-wave detection.
Its engineering design requires a multi-criteria approach capable of identifying and addressing geological, geotechnical, environmental, and landscape challenges. To manage these complexities, the ET-3G Lab at Sapienza University of Rome (as part of the ETIC PNNR project), has produced an advanced digital multi-scale 3D model for the Sardinia site identified as a potential location.
The model integrates surface and subsurface data at both regional and local scales, consolidating all available geological, geophysical, and geotechnical datasets to support a coherent reconstruction of key subsurface features, including lithotypes, faults, and fracture networks. It incorporates data from surface observations and drilled calibration wells, encompassing geological and petrophysical information, laboratory tests on undisturbed samples, fracture analyses, and geophysical investigations conducted by the ET scientific community. This integrated representation strengthens the linkage between surface and subsurface information.
As a result, a comprehensive 3D geological model of the ET Sardinia site has been developed, enabling visualization of the subsurface down to a depth of approximately one kilometer.
This advanced modeling approach is intended to support the minimization of geotechnical risks, the optimization of construction strategies and associated costs, and the implementation of scenario-based design analyses.
How to cite: Lipparini, L., Cagnizi, M., Ferranti, F., D'Aranno, P. J. V., Sappa, G., Wahbeh, W., Napoleoni, Q., and Marsella, M.: 3D geological and geotechnical subsurface model for the Einstein Telescope study area in Sardinia (Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8376, https://doi.org/10.5194/egusphere-egu26-8376, 2026.
The efficiency of having a simple scheme for creating small scale international geological maps and to offer them in a simple, usable and standardised format has been showcased by the international collaboration of the Commission for the Geological Map of the World (CGMW), the Deep Time Digital Earth (DDE) and the CAGS (Chinese Academy of Geological Sciences) programme, and by some Geological Surveys. The success of the World 1:5M map pilot project and its follow-up toward multi-layers products has given us the confidence to achieve a unified World Geological Map at the scale of 1:1M., a dream initially envisaged by the OneGeology project.
A spectacular milestone of the global 1:5M map, the largest seamless digital geological map ever compiled, was the first phase. A following phase of this program is to create the first “basement map” of the world, by simply removing the youngest sediments from sedimentary basins and continental shelves.
While layering techniques such as basement mapping is accelerating, a new vivid vision is to compile a rigorous 1:1M global bedrock geology under protocols for sharing and regular updating of databases from willing Surveys. Compiling data into a harmonized Geological Map of the World at 1:1M scale is now the new ambitious objective of CGMW. The endeavour poses scientific, technical and geopolitical challenges, and will require the participation and efforts of partners from as many countries as possible, who must be willing to openly share information, as well as the active involvement of experts. Building on the robust methodology used for the 1:5M, we are exploring options to foster the harmonization, including using AI tools.
However, not all the national source maps are available in digital format and in English, use the same coordinate system, or comprehensive databases. Therefore, we anticipate the necessity to digitize or vectorize some geological data and to arrange a standardized database for all the maps. In some cases, boundary contrasts of resolution will require additional work. Another time-consuming task will be the cross-border correlation of geological structures and units by applying high-quality digital terrain models (DTMs), multi-spectral satellite data, or larger scale regional maps. Finally, the validation of the data by experts and Geological Surveys will be necessary. This initial digital mapping will be completed in 2D as a first step toward a future 3D geological map and a powerful Digital Twin. The multi-layer 3D version will be developed in the long term as data availability, priority, and partnerships allow. Our EGU2026 poster and associated discussions are an ideal opportunity to present the 1:1M project and to foster collaborations, for example with CGI and the OneGeology
How to cite: Pubellier, M., Thorleifson, H., Song, Y., Sautter, B., Ogg, J., Robida, F., Harisson, M., Nehlig, P., and Gomez Tapias, J.: Geological maps of the Future: Leveraging on the methodology of the 1:5M Map to construct a 1:1 Geologic Map of the World, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10293, https://doi.org/10.5194/egusphere-egu26-10293, 2026.
A powerful earthquake swarm related to accumulation of magma in a shallow reservoir beneath Svartsengi, on the Reykjanes Peninsula SW Iceland began in October 2023. On 10 November 2023 a large dike intrusion occurred beneath the town of Grindavík leading to the formation of a graben structure on the west side of town. Subsequently, 11 more dike intrusions have occurred along the Sundhnúkur crater row, with another graben forming on the east side of town. The maximum subsidence measured in the town is 1.5 m, and further fault movements were triggered throughout Grindavík. These events resulted in the opening of numerous fractures and caused damage to critical infrastructure. Following these events, the Icelandic Civil Protection authorities commissioned a detailed geological and geophysical investigation of the area.
A final report, alongside numerous technical memoranda, is now available, presenting the main results. One of the key outcomes of the project is a detailed fracture map of Grindavík. The map identifies seven distinct fracture zones that have been active during the ongoing unrest: Stamphólsgjá, Hópssprunga, Austurhópssprunga, Víðihlíðarsprunga, Bröttuhlíðarsprunga, Stakkavíkursprunga, and Strandhólssprunga (see:https://www.map.is/grindavik/). Stamphólsgjá is the deepest (>30 m) and widest fracture (3 m). In addition, depths greater than 20 m were measured within fractures of the Hópssprunga and Bröttuhlíðarsprunga zones. It is important to note that Stamphólsgjá and Hópssprunga are several thousand years old, and not all of the observed widening can be attributed to the current events. Historical aerial photographs show that Stamphólsgjá was already significantly open prior to the development of the town. No evidence of Austurhópssprunga, Víðihlíðarsprunga, Bröttuhlíðarsprunga, or Stakkavíkursprunga is visible on older aerial imagery, indicating that these fractures likely formed during the ongoing events. Most fractures are typically 20–60 cm wide and 1–5 m deep, while relatively few locations exhibit fractures wider than 80 cm and deeper than 8 m. It is important to consider that substantial material collapse has occurred into many fractures, and often only surface depressions and subsidence are visible, indicating the presence of open fractures beneath the surface. The investigation employed various methods, including aerial photo interpretation, LiDAR elevation measurements, ground-penetrating radar (GPR), magnetic surveys, electrical resistivity measurements, and visual inspection.
Excavations carried out in connection with road repairs provided valuable opportunities to examine several meters into the bedrock and assess its composition. These observations revealed that the upper 4–10 m of the bedrock consist of four postglacial lavas, separated by sedimentary layers and soil. No deeper hyloclastite formations from the last glacial period were observed. The youngest lava exposed at the surface is the Sundhnúkur (sh) lava (~2200 years old). Previously known fissures in Grindavík are prominent in older lava flows (>8000 years old) but are scarcely visible in Sh.
Importantly, the volcano-tectonic unrest in and around the town is ongoing, and further fracture movements may occur in the future, and existing surface fractures continue to evolve due to unconsolidated materials moving within the fractures underscoring the importance of continued monitoring.
How to cite: Erlendsson, Ö., Sigurgeirsson, M. Á., Einarsson, G. M., Friðsteinsson, J. Ö., Steingrímsson, J. H., Pascale, G. P. D., Johanna, E., Gallagher, C. R., Arngrímsson, H. Ö., Hauksdóttir, S., and Ben-Yehoshua, D.: Geological and Geophysical Investigation of Grindavík, Iceland, in Response to Volcanic Activity and Fissure Movements at the Sundhnúkar Eruption Fissure, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20682, https://doi.org/10.5194/egusphere-egu26-20682, 2026.
The Strathmore Basin is an extensive Silurian-Devonian basin which spans the entire width of Scotland. This basin has had a long and complex tectonic history, including periods of significant volcanic activity, faulting, basin folding, and several movements along the basin-bounding Highland Boundary Fault. Today, the basin is largely covered by substantial glacial deposits; bedrock exposure is limited.
Some areas of the basin were last mapped in the 1880’s (i.e., before aerial photography, and nearly a century before the theory of plate tectonics). Progressive mapping of adjacent map sheets up to the 1970’s has led to mismatches at sheet boundaries, significant inconsistencies in structural interpretation, and irregularities in stratigraphic relationships. Addressing these legacy issues in geological maps is critical for ensuring suitability for 21st century applications; these data are used to inform the management of the regional aquifer within the Devonian sandstones, and for evaluation of potential geothermal energy resources.
A novel basin-wide approach has been taken to revise the geological mapping to improve map quality and consistency across the Strathmore Basin. This has involved a range of techniques, including digital terrane analysis, targeted field visits, the integration of published geochronological data, and the compilation of basin-wide datasets of over 4,000 structural measurements and more than 20,000 observation points from multiple BGS data sources. This approach has allowed for a new large-scale structural interpretation of the fold and fault systems, particularly related to the Highland Boundary Fault, as well as a new understanding of key stratigraphic markers and a more coherent representation of the geology across the basin. This approach highlights the value of using both modern and historic datasets, and crucially, revisiting targeted outcrops in the field.
As traditional survey styles become less affordable, and the need for seamless maps more acute, regional approaches provide an important methodology, helping to maximise the value of existing data and targeting areas for new data collection. Understanding these strengths and limitations is essential for the future of resurvey, especially in countries such as the UK with a long surveying history and high demand for accurate and consistent geological information to manage energy, water, and mineral resources.
How to cite: Reeves, T., Whitbread, K., Kearsey, T., Stephens, T., Arkley, S., Unwin, H., Murphy, B., Callaghan, E., and Hughes, T.: Modern applications for basin-wide revision mapping in the Old Red Sandstone, Scotland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7883, https://doi.org/10.5194/egusphere-egu26-7883, 2026.
The sedimentary basin fill of the Cenozoic Roer Valley Graben System (the Netherlands) has gone through multiple phases of tectonic deformation during the Alpine orogeny, resulting in a variety of extensional and compressional structures, syn-tectonic sedimentary features and a complex and multidirectional fault pattern. The characteristics of these features, such as lithological properties, associated faults and their geometries, are crucial in geological investigations that focus on energy transition studies and/or water management. The present study seeks to enhance the geological understanding of a complex syn- and post-kinematic sedimentary feature, resembling a canyon-shaped collapse structure that formed on a relay ramp along the northern graben shoulder. Particular emphasis will be on methodology, mapping results and understanding the role of inherited faults on its development.
Since the late twentieth century, the Geological Survey of the Netherlands (GDN-TNO) has played an important role in advancing scientific understanding of the country’s subsurface geology. A major accomplishment of the GDN-TNO is the creation of comprehensive, country-wide subsurface models, using numerous 2D and 3D seismic surveys of various vintages as well as a substantial number of exploratory wells and more recently the results of the SCAN (Seismic Campaign for Accelerating Geothermal Energy) program.
Past and recent systematic inspection of this legacy data of the GDN enables us to examine both the geometry (i.e. the shape and spatial arrangement) and mechanisms of faults and associated specific sedimentary features, such as hanging-wall collapse and accretionary channel infill structures, as well as a plausible sedimentary wedge downslope of the hanging wall. Combining this with the results of our subsurface geological models, we present the potential relevance of inherited tectonics and fault reactivation on the development of these syn- and post-kinematic sedimentary features in the subsurface of the Netherlands. 2D seismic data may not always be sufficient to understand the fault orientation and length. To address these challenges and improve the accuracy of our geological modelling approach, we will incorporate and present our findings from adjacent 3D seismic datasets combined with conceptualized tectonic diagrams and real world analogues.
How to cite: Aksay, S., den Dulk, M., ten Veen, J., and Nelskamp, S.: Unravelling a fault-related footwall canyon feature along the Roer Valley Graben System, the Netherlands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13012, https://doi.org/10.5194/egusphere-egu26-13012, 2026.
The terrain classification through Terrain Mapping Units (TMU) consists of the definition of homogeneous relief units that integrate different aspects of the natural environment (geology, geomorphology, drainage, land use, vegetation, etc.), providing a solid basis for multidisciplinary studies focused on aspects such as mining, geotechnics, natural hazard analysis and environmental assessment, among others. This approach may be of particular interest in countries that lack a comprehensive geological and geomorphological mapping infrastructure, providing a basic characterization of their main geographical, geological and environmental characteristics. Currently, the wide variety of available remote sensing products constitutes an advantage when tackling this type of cartography.
The main goal of this study is to evaluate the usefulness of freely available remote sensing products, accessible online on a global scale, for producing TMUs. To achieve this goal, a combined analysis of several remote sensing products was addressed for the Campo de Cartagena (SE Spain), a semi-arid and heavily anthropized area including the Mar Menor lagoon, the Neogene and Quaternary detrital deposits from the Campo de Cartagena plain and the surrounding mountain ranges, formed by Palaeozoic, Permian and Triassic metamorphic rocks.
Remote sensing products used are: (1) a digital elevation model – DEM with spatial resolution of 30 m, derived from the Shuttle Radar Topography Mission (SRTM, NASA), and (2) a multispectral Sentinel-2 dataset, with spatial resolutions of 10 and 20 m. On this basis, two different spatial resolution TMU maps were developed and compared to test their different capabilities for mapping purposes: (1) based on the 30 m spatial scale DEM and Sentinel-2 bands at 20 m spatial resolution, and (2) based on the 30 m spatial scale DEM and the Sentinel-2 bands at 10 m spatial resolution. Processing the DEM using a Geographic Information System – GIS resulted in hillshade, slope and flow accumulation models, which were used to characterise the main topographic features. In addition, the combination of different spectral bands and the application of digital image processing techniques enabled the identification of differences in surface composition. Based on these observations, homogeneous TMUs were delineated according to three main criteria: (1) relief, (2) drainage network and (3) surface composition variability. Accuracy analysis and validation were implemented by field-work observations and by comparing the resulting terrain classification map with the already existing geological and geomorphological maps at 1:50000 scale from the Spanish Geological Survey (IGME). This study highlights the potential of freely available remote sensing products, accessible online on a global scale for mapping TMUs in an area affected by intense agricultural and mining activities.
Acknowledgements: Research Project PID2023-150229OB-100 (HYPERLANDFORM) financed by MICIU/AEI/10.13039/501100011033 and by FEDER, UE. The participation of Inés Pereira was supported by an FPU (FPU21/04495) contract from the Spanish Ministry of Universities.
How to cite: Rodríguez, I., Valenzuela, P., García-Meléndez, E., Pereira, I., and Ferrer-Julià, M.: Application of multispectral Sentinel-2 images for Geo-environmental terrain classification mapping based on landforms: an example of the Campo de Cartagena, SE Spain, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17659, https://doi.org/10.5194/egusphere-egu26-17659, 2026.
Geological mapping of the 25 × 25 km Torma 1:50,000 map sheet is challenged by:
- the crossing of the Ordovician–Silurian carbonates boundary,
- Devonian siliciclastic rocks overlapping parts of the area,
- alternating Quaternary cover of primarily glacial origin.
The bedrock geology is further complicated by a north–south oriented facies transition within the Ordovician succession, from relatively shallow carbonate facies towards more deep facies. Drilling-based constraints are limited: historical borehole information is sparse, descriptions too general, and locally conflicting, while available cores are of insufficient quality for reliable stratigraphic control. To improve geological understanding within restricted budgets, we selected towed time-domain electromagnetics (tTEM) as a rapid data acquisition method for regional-scale mapping.
We report results from over 100 km of tTEM profiling, acquired predominantly with a 3 × 3 m 1-turn transmitter configuration. Data were collected primarily along unpaved roads, smaller roads, and paths, complemented by targeted measurements on selected fields. This mixed acquisition strategy produces strongly variable lateral sampling density and enables an assessment of how survey geometry and data coverage influence interpretational confidence. Road-based acquisition enables rapid spatial coverage but with lower effective lateral resolution compared to field grids, and introduces additional noise and artefacts related to infrastructure. While mapped utilities can be considered during planning, abandoned cables and scattered ferrous objects (e.g., signs, posts, culverts) create intermittent interference that must be identified and mitigated during processing and interpretation.
Preliminary results do not support the presence of a large buried valley previously inferred from multiple older (now lost) drill cores; this is consistent with nearby seismic lines at the reported locations. Across most of the area, tTEM provides the most continuous constraint on Quaternary thickness, and field-based segments resolve internal variability sufficiently to discriminate between different Quaternary units with higher resistivity contrasts, providing a new tool for Quaternary mapping in Estonia as well. Bedrock-related contrasts are detectable in parts of the survey area, but not consistently across all geological situations. Thickness estimates of the uppermost bedrock units correlate well with drill-core control where available, yet indicate substantially higher spatial variability elsewhere than expected from existing conceptual models.
The dataset highlights the areas where drilling remains necessary to resolve key ambiguities, while providing a markedly improved basis for defining regional trends and constructing geological models and updated maps in a complex carbonate–siliciclastic setting.
How to cite: Ani, T. and Kuusk, C.: How much tTEM coverage is enough to trust a geological interpretation? Evidence from mixed road/field-based data acquisition across the Ordovician–Silurian boundary and Devonian cover in Estonia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17117, https://doi.org/10.5194/egusphere-egu26-17117, 2026.
Geologic maps are undergoing a paradigm shift from static illustrations to dynamic, intelligent knowledge platforms. Traditionally, geologic maps have served specialized fields in fixed image formats. However, their closed information systems, weak interactivity, and difficulties in cross-domain integration have limited the full release of their value. In recent years, advancements in Artificial Intelligence (AI) and Multimodal Large Language Models (MLLMs) have provided a new pathway for the digital reconstruction and intelligent application of geologic maps.
In collaboration with Microsoft Research Asia, our project team has proposed and constructed an open, extensible intelligent platform for geologic map comprehension and service. This platform is based on high-quality digitized geologic map datasets and utilizes MLLMs to achieve semantic parsing, knowledge association, and natural language interaction with geologic maps. The established platform not only supports the accurate identification and extraction of fundamental map elements (such as legends, lithology, and structures) but also enables the following multi-level application scenarios:
- Intelligent Interaction and Q&A: Users can directly query geologic information using natural language—for example, "What faults are distributed in this area?" or "What is the formation age of a certain rock layer?" The system generates accurate answers by integrating graphic-text information and domain knowledge.
- Scientific Research and Educational Tools: It provides an interactive, annotatable interface for geologic map learning, supporting classroom teaching, professional training, and interdisciplinary research.
The platform is supported by core technologies including the first-ever multimodal benchmark for geologic map understanding, GeoMap-Bench, and the intelligent agent framework, GeoMap-Agent, which significantly outperforms general-purpose vision-language models on multiple tasks. Geologic maps are no longer merely "base maps" or "reference maps"; they have become an intelligent knowledge base connecting geologic data, professional expertise, and multi-domain applications.
Looking ahead, the geologic map platform will further integrate real-time sensor data, remote sensing information, and socio-economic factors, driving the earth sciences towards a new era characterized by openness, collaboration, and intelligence. It is poised to play a central role in scientific discovery, engineering safety, sustainable resource utilization, and the building of societal resilience.
How to cite: Song, Y. and Huang, Y.: The Geologic Map Intelligent Platform: AI-Enabled Digital Transformation and Building a Multimodal Application Ecosystem, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16524, https://doi.org/10.5194/egusphere-egu26-16524, 2026.
The Orientale Basin, centered at ~19°S, ~93°W, is one of the most characteristic features on the surface of the Moon. Constituted by three concentric rings, the largest of which is between 930 and 950 km in diameter, this multi-ring basin is one of the youngest large impact basins on the Moon (Orientale is estimated to date back ~3.81 Ga) and one of the best-preserved large basins in the entire Solar System. Inside, its central depression hosts a relatively thin infilling of dark, smooth material interpreted as a mare basalt, whilst outside the outermost ring an ejecta blanket drapes the surrounding topography sometimes reaching over 1,400 km from the center of the basin. Throughout the years, the importance of the Orientale Basin has led to the creation of several geological maps at various scales, none of which, however, a scale greater than 1:200,000. Additionally, these maps never try to put together the two main methodological approaches adopted internationally up to this point at global scale for the Moon. Our work tries to bridge this gap by presenting a new medium-to-large-scale (1:118,000) geological map of both the inner and outer facies which makes use of a combination between a traditional planetary geological scheme and a more morphometric criterion.
The map was created with the latest long-time stable release of QGIS (vrs. 3.40) mainly using the 59 m/px resolution Lunar Orbiter Laser Altimeter (LOLA)-Kaguya Shaded Relief and the 59 m/px resolution LOLA-Kaguya DEM. These two datasets, only covering latitudes within ±60° were utilized to distinguish the different units and subunits based off their general morphology, textures, and locations, but also to identify the structures. The 100 m/px resolution, grayscale mosaic of the Lunar Reconnaissance Orbiter Wide Angle Camera (LROC-WAC) and the 118 m/px LOLA elevation model were additionally used to make up for the missing portion of the LOLA-Kaguya datasets. The Clementine UVVIS colored mosaic (200 m/px) and the mineral abundance (wt% of Ol, Cpx, Opx, Pl, FeO) Kaguya mosaics also allowed to add a layer of information regarding differences in composition of apparently visually uniform features and terrains.
We managed to identify over 20 between units and sub-units that we grouped based on the terrain or morphological feature they are related to (e.g. crater, mare, …), and over 10 classes of structures. Our final product represents the highest resolution map available for the Orientale Basin and when compared with already existing medium-scale maps, appears to depict with more detail and accuracy its complexity. Additionally, we made use of a color vision deficiency-friendly color scheme to make the map more accessible also to that part of the population having limited sensitivity to colors.
How to cite: Caddeo, Y., Nodjoumi, G., D'Incecco, P., and Di Achille, G.: A New High-Detail, Color Vision Deficiency-Friendly Geological Map of the Orientale Basin (Moon), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21060, https://doi.org/10.5194/egusphere-egu26-21060, 2026.
The Devonian-Carboniferous Admiralty igneous complex (i.e. Admiralty plutonites and Gallipoli volcanics) of northern Victoria Land, Antarctica, forms part of a widespread magmatic system comprising felsic volcanic, subvolcanic and plutonic lithologies. Due to extensive snow and ice coverage, aeromagnetic data has been used to interpret the extent of igneous bodies where surface exposure of igneous rocks is limited. However, some exposures generate strong positive magnetic anomalies, while others produce weak or negligible responses, raising questions about the factors controlling magnetic susceptibility and interpretation of aeromagnetic data where exposure is absent.
We focus on several key locations with exposed Admiralty igneous rocks showing strong positive anomalies (Everett, Salamander and southern Alamein ranges, Mariner Plateau), negligible anomalies (Tucker Glacier region), and a combination of weak and strong anomalies (Yule Bay) to explore how variations in rock properties and geochemical composition relate to observed magnetic anomalies.
Combining aeromagnetic surveys and in-situ susceptibility measurements with detailed petrology, modal mineralogy, whole-rock geochemistry (major, minor, and trace elements) and ongoing age dating allows a better understanding of the causes of low versus high magnetic anomalies in rocks previously ascribed to a single magmatic event. In particular we are testing whether (a) multiple, compositionally distinct magmatic pulses, (b) variable degrees of alteration, and/or (c) different levels of exposure can account for the observed discrepancies in magnetic anomalies.
Magnetic and susceptibility data, when combined with petrological and geochemical analyses, provide a powerful tool to investigate the origin of variations in magnetic susceptibility, particularly in regions with limited outcrops.
How to cite: Ruppel, A., Ratschbacher, B., Koglin, N., and Läufer, A.: Strong or weak? What controls magnetic anomalies in the Admiralty igneous complex, northern Victoria Land, Antarctica, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7419, https://doi.org/10.5194/egusphere-egu26-7419, 2026.
EMODnet Geology, one of the thematic pillars of the European Marine Observation and Data Network (EMODnet), harmonises and delivers pan European geoscientific data to support sustainable marine management. Since its launch in 2009, EMODnet Geology has successfully integrated diverse marine geological datasets covering seabed substrate, sedimentation rates, seafloor geology, coastal behaviour, geological events, marine minerals, and submerged landscapes into harmonised data products accessible via the EMODnet Portal: https://emodnet.ec.europa.eu/en. The thematic network spans the European regional seas and extends into the Caribbean Sea.
EMODnet Geology focuses on delivering harmonised data products (e.g., thematic maps) while providing metadata links to original data providers. By transforming fragmented datasets into standardized, interoperable products, it supports maritime spatial planning, environmental assessments, and sustainable resource management. The project also facilitates third-party data contributions via direct submission or through EMODnet Data Ingestion, engaging both public and private sector data holders.
A new project phase (September 2025–September 2027), coordinated by the Geological Survey of Finland GTK and executed by a consortium of 39 organisations from EuroGeoSurveys and other expert institutions, introduces significant enhancements in thematic coverage and data quality. These developments include compilation of novel datasets on organic carbon content of seabed sediments, carbon-14 measurements of strata, geotechnical properties of seabed as well as flora and fauna on the submerged landscapes. In addition, the network continues updating its existing data products with new and refined data. EMODnet Geology also contributes to the European Digital Twin Ocean (EDITO), by supporting the development of a shared, cloud-based data lake and enabling next-generation digital ocean applications.
EMODnet Geology, along with other EMODnet thematics: bathymetry, biology, chemistry, human activities, physics, and seabed habitats, provides open-access, FAIR in situ marine data and data products all accessible via the EMODnet Portal. These datasets support a wide range of scientific, policy, and industrial applications.
The current EMODnet Geology phase is funded by The European Climate, Environment and Infrastructure Executive Agency (CINEA) through contract CINEA/EMFAF/2024-25/3.6/4500124305 for European Marine Observation and Data Network (EMODnet) - Lot2/CINEA/2024/OP/0006 (Geology).
How to cite: Kaskela, A. M., Kihlman, S., Kotilainen, A. T., Wasiljeff, J., and network, E. G.: EMODnet Geology continues to advance marine geological data for Europe , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10753, https://doi.org/10.5194/egusphere-egu26-10753, 2026.
Human activities and increasing pressures on marine and coastal environments have highlighted the need for accessible, reliable, and harmonized marine information. Since 2009, the EMODnet (European Marine Observation and Data Network) Geology project has been collecting and harmonizing geological data from all European sea areas, and Caspian and Caribbean Seas. This work, carried out in collaboration currently with 39 partners and subcontractors, has focused on creating cross-boundary, multiscale datasets from scattered and heterogeneous sources for diverse applications.
Seabed substrate is one of the main parameters describing marine environment. Project addresses seabed substrates and related characteristics and over the years, EMODnet Geology has developed several data products such as harmonized seabed substrate maps based on sediment grain size, sedimentation rate datasets, and a seabed erosion index derived from literature. These products have evolved, incorporating additional attributes like seabed surface features (e.g., seagrass meadows, bioclastic bottoms, ferromanganese concretions) and confidence assessments to improve usability and usefulness.
Building on this foundation, the latest phase of the project introduces new data additions to the data catalogue. One of the additions to complement existing sedimentary information is organic carbon data, which is essential for understanding carbon cycling, climate regulation, and ecosystem health. At the same time, we have initiated work on identifying and classifying sedimentary environments within national datasets to better capture dynamic processes and environmental variability, to support modelling and interpretation of marine systems. Basic work on these new datasets is underway, and we are in the early stages of method development to integrate this new information.
After more than fifteen years, EMODnet Geology has established itself as one of the main providers of publicly available, harmonized in situ seabed data. Continued development, both updating existing products and introducing new datasets, will ensure the relevance of this information for addressing future challenges in marine and coastal management and research.
The EMODnet Geology project is funded by The European Climate, Environment and Infrastructure Executive Agency (CINEA) through contract CINEA/EMFAF/2024-25/3.6/4500124305 for European Marine Observation and Data Network (EMODnet) - Lot2/CINEA/2024/OP/0006 (Geology)
How to cite: Kihlman, S., Kaskela, A. M., Kotilainen, A. T., and Wasiljeff, J. and the EMODnet Geology network: Harmonized seabed substrate datasets and insights from EMODnet Geology, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16761, https://doi.org/10.5194/egusphere-egu26-16761, 2026.
The European EMODnet Geology project started in 2009. One of its aims is to provide geological map data of the European seas, harmonised as far as possible and made available according to FAIR data principles.
The EMODnet Geology Workpackage “Seafloor Geology” is not only compiling map layers of the geology of the seafloor (Quaternary and pre-Quaternary but is also mapping layers of the geomorphology of the European seas and beyond. Semantic and geometric harmonisation is essential to understand geological information across administrative (EEZ) boundaries. The main method to provide semantically harmonised data layers is common and agreed upon terms to describe a unit: a vocabulary.
To describe the characteristics of the seafloor geology, the vocabularies of the European INSPIRE Directive Data Specifications Geology (INSPIRE Thematic Working group Geology 2013) could be applied to describe the age, lithology and genesis (event environment, event process) of the marine geology.
While the INSPIRE vocabularies are comprehensive, they nevertheless lack terms to describe the marine geomorphological features. EMODnet Geology fills that gap and is developing hierarchical scientific vocabularies for marine geomorphology to describe the concepts to which geometrical descriptions (lines and polygons) can be linked. This controlled vocabulary consists of a hierarchical, machine-readable list of terms and definitions needed to describe the European seafloor geomorphological units.
The process to set up vocabularies for the marine domain faces considerable challenges, such as:
- Finding suitable terms and definitions
- Avoiding duplication
- Agreeing internationally on the terms and description
- Coping with obsolete and/or strictly regional terms
- Considering multiple hierarchies
The presentation demonstrates the project’s approach to build pan-European applicable vocabularies to describe marine geomorphological features and presents use cases for its application.
How to cite: Asch, K., Blischke, A., Ernsten, V. B., Hojgaard, B., Medialdea, T., Mortensen, L., Sakellariou, D., Heckmann, P., Schulz, M., Müller, A. M., and network, T. E. G.: EMODnet Seafloor Geology: The wavy cruise towards a hierarchical, machine-readable geomorphology vocabulary, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12405, https://doi.org/10.5194/egusphere-egu26-12405, 2026.
The Biodiversea LIFE IP project (2021–2029) is Finland’s largest coordinated initiative to safeguard biodiversity of the Baltic Sea and promote the sustainable use of its marine environment. The Geological Survey of Finland GTK conducted marine geological surveys around the Åland Islands to support informed marine management and conservation.
The work combined seismo-acoustic methods, including subbottom profiling, multibeam echosounder, and sidescan sonar, with extensive surface sediment sampling. These surveys produced detailed information on seabed geodiversity, sediment distribution, and substrate types, indicating a highly geologically diverse seafloor around the Åland Islands. The resulting datasets improve our understanding of the physical and geological properties of the seafloor, which form the foundation for biodiversity and habitat development in the area.
We describe the geological setting, the applied survey methods, and the contribution of geoscientific information to multidisciplinary marine conservation planning. The results highlight the importance of geological data for understanding marine ecosystems and for supporting science-based decision-making in marine management.
The Biodiversea LIFE IP project is coordinated by Metsähallitus. In addition to GTK, project partners include the Baltic Sea Action Group (BSAG), Finnish Environment Institute (SYKE), Ministry of the Environment, Natural Resources Institute Finland (Luke), Turku University of Applied Sciences, Åbo Akademi University, and the Åland Provincial Government. The project has received funding from the LIFE Programme of the European Union. The material reflects the views of the authors, and the European Commission or CINEA is not responsible for any use that may be made of the information it contains.
How to cite: Virtanen, S., Jokinen, S., Kaskela, A., Sahiluoto, M., Sainio, A., and Sanila, N.: Geodiversity and Seafloor Substrate Mapping to Support Marine Management in the Åland Islands, Baltic Sea – Results from the Biodiversea LIFE IP Project , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16620, https://doi.org/10.5194/egusphere-egu26-16620, 2026.
The inherent uncertainty of subsurface geological conditions remains a primary challenge in underground spatial planning and rock engineering. The rationality of engineering design is fundamentally dictated by the spatial distribution and continuity of geological structures. However, in complex environments—characterized by intense tectonic fracturing or rapid lithological transitions—traditional 2D projections often fail to capture the anisotropic nature and spatial evolutionary trends of the rock mass, leading to significant interpretative gaps. Discrepancies between predicted and encountered geology frequently stem from a 2D conceptual framework that oversimplifies the 3D connectivity of fault planes, shear zones, and joint sets. This study addresses these limitations by utilizing the Zhaishan Tunnel system in Kinmen, characterized by its granitic basement, as a research platform. By integrating UAV LiDAR, Terrestrial Laser Scanning (TLS), and SLAM technologies, we established a high-resolution 3D spatial database that bridges the gap between surface and subsurface geological data. The core research focus is the development of a workflow for continuous surface-subsurface 3D geological modeling. By incorporating surface topography, outcrop mapping, and in-situ structural measurements into a unified 3D coordinate system, the study employs multi-scale data constraints to enhance the reliability of geological interpretations. Macro-scale surface terrain data are utilized to constrain the meso-scale structural interpretations within the tunnel, ensuring that the model maintains structural consistency across different depths. The significance of this research lies in transforming geological outputs from static, post-survey records into dynamic, 3D interpretative engines. This approach allows for the visualization of discontinuity extensions in three dimensions, providing a data-driven framework for anticipating geological hazards. Ultimately, this shift ensures that geological interpretations are no longer fragmented, providing a high-integrity information base for modern underground space development and structural stability analysis.
How to cite: Chang, K.-J., Huang, M.-J., Wang, C.-C., and Haung, K.: Multi-Source Data Fusion and Multi-Scale Constraints for Continuous Surface-Subsurface 3D Geological Modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4181, https://doi.org/10.5194/egusphere-egu26-4181, 2026.
Geospatial data acquisition technology has been widely integrated into geological and engineering geology research, significantly enhancing the spatial precision and structural integrity of topographical interpretations. In the era of high-performance computing, 3D geological modeling has emerged as a pivotal trend for engineering applications. However, the practical depth of these models is often constrained by challenges in accuracy and reliability, arising from varying data resolutions and the complexities of integrating multi-source information. These issues complicate model validation, particularly in large-scale or high-complexity engineering environments. As data collection methods become increasingly diverse, Simultaneous Localization and Mapping (SLAM) technology has revolutionized traditional surveying by offering superior operational flexibility and mobility. Unlike static terrestrial laser scanning (TLS), handheld or mobile LiDAR systems (MMS) can efficiently traverse indoor spaces, narrow urban corridors, and densely vegetated areas, facilitating the construction of comprehensive, blind-spot-free 3D spatial datasets. Despite these advantages, achieving and maintaining engineering-grade precision in GNSS-denied or signal-unstable environments remains a critical technical bottleneck. This study aims to investigate a robust workflow for large-scale field model construction using a "batch processing and stitching fusion" strategy. Using the National Taipei University of Technology (NTUT) campus as an experimental field, high-density point cloud data were collected using the mobile mapping system. The research methodology focuses on optimizing geometric fidelity by rigorously analyzing two key variables: first, a comparative evaluation of trajectory adjustment modes, specifically contrasting loop-closure correction with Post-Processed Kinematic (PPK) technology; and second, an assessment of how the quantity and spatial distribution of Ground Control Points (GCPs) influence the model’s global stability and absolute correctness.
The experimental results demonstrate that through optimized GCP deployment and refined trajectory adjustment, the absolute accuracy of the point cloud model can be maintained within an RMSE of 5 cm, with the relative accuracy on ground surfaces controlled within 2 cm. Furthermore, in the measurement of high-rise structures, the ghosting effect (layering) is restricted to within 4 cm at a 30-meter operational radius, while an average point spacing of 4 cm is maintained to ensure the geometric integrity of model details. These findings confirm that mobile LiDAR systems, when supported by optimized workflows, can meet the stringent precision requirements of engineering-grade projects while retaining high flexibility.
In conclusion, this research establishes a high-precision 3D digital foundation for the campus. This methodology is highly extensible to geological fields, including outcrop geometric measurement, quantitative analysis of landslide volumes, and structural surveys in GNSS-denied environments such as tunnels and caves.
How to cite: Huang, K.-Y., Wang, C.-C., Chiang, J., and Chang, K.-J.: Optimization of Modeling Accuracy for Mobile Mapping Systems in Large-Scale Environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6560, https://doi.org/10.5194/egusphere-egu26-6560, 2026.
With the increasing demand for 3D spatial data in engineering and geological applications, constructing practical 3D models efficiently and effectively has become a critical challenge in geology, underground engineering, and architectural documentation. In recent years, Simultaneous Localization and Mapping (SLAM) technology has been widely adopted in complex environments to collect high-density point clouds with high efficiency. However, the reliability and applicability of SLAM-derived results in geological and engineering contexts still require verification through practical case studies. This research utilizes the Building of the Civil Engineering at National Taipei University of Technology as the primary experimental site. A mobile SLAM system was employed to collect 3D point cloud data, which was subsequently integrated into the Building Information Modeling (BIM) framework—a standard in Taiwan's engineering industry—to assist in model construction and application. Furthermore, the study extends to several representative engineering and geological sites, including the Zhaishan Tunnel in Kinmen, the Kinmen Ceramics Factory, and the coastal rock outcrops at Qixingtan in Hualien, to explore the feasibility of SLAM-based 3D modeling under diverse environmental conditions.Regarding engineering applications, this study compares different positioning modes, including pure SLAM, SLAM combined with PPK, and SLAM integrated with RTK. Both absolute and relative accuracy at the architectural scale were analyzed using control points. Additionally, the impact of control point distribution on the geometric consistency of the models was investigated. These findings serve as a technical reference for selecting SLAM positioning strategies and operational workflows in engineering practice.In terms of geological and underground engineering applications, the research focuses on using SLAM point clouds for the 3D reconstruction and visualization of tunnel morphology, rock wall geometric features, and coastal outcrops. The results demonstrate the potential of this technology in tunnel geological recording, engineering planning, and outcrop preservation, providing a foundation for geological modeling in analytical tasks. In conclusion, this study proposes a practice-oriented workflow that integrates SLAM point clouds with BIM. By balancing engineering precision analysis with geological modeling applications, this research provides a high-efficiency 3D modeling solution with significant practical value for the architectural, tunneling, and geological sectors.
How to cite: Wang, C.-C., Huang, K.-Y., Chiang, J., and Chang, K.-J.: Precision and Accuracy Evaluation of 3D Modeling in Indoor Confined Environments: Integrating Mobile Mapping System and BIM, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6674, https://doi.org/10.5194/egusphere-egu26-6674, 2026.
Hydrostratigraphic models are commonly used as structural frameworks for groundwater and subsurface studies. Traditionally, these models are treated as deterministic representations, providing a single “best estimate” of subsurface structure. While practical, this approach conceals the inherent uncertainty in geological interpretation, particularly in the spatial placement of layer boundaries, and limits the transparency and robustness of subsequent modelling workflows. Recognising and quantifying this uncertainty is a necessary step towards more probabilistic approaches to hydrostratigraphic modelling.
This contribution presents GDM (geology-driven modelling), a method for explicitly quantifying interpretation uncertainty in the placement of hydrostratigraphic layer boundaries through ensembles of 3D subsurface realisations. GDM operates on existing hydrostratigraphic models, assuming a fixed framework in terms of layer definition and conceptual interpretation, while focusing on the spatial variability of layer interfaces. The method is computationally efficient, enabling application at regional or national scales. Its national-scale implementation, allows interpretation uncertainties to be assessed across entire hydrostratigraphic frameworks, providing a consistent basis for revisiting legacy models.
As an illustration, we demonstrate how GDM was used to quantify interpretation uncertainties in the national-scale hydrostratigraphic model of Denmark and how the resulting ensemble of subsurface realisations was incorporated into the hydrological modelling workflow. The ensemble describes the range of equally plausible geometries supported by the available data and assumptions, providing a structured way to explore how interpretation uncertainty propagates through geological models.
This example serves as a starting point for reflecting on broader implications. In particular, it illustrates how approaches that explicitly quantify interpretation uncertainty can help bridge the gap between established deterministic models and future strategies that increasingly embrace probabilistic representations. At the same time, these approaches introduce new considerations for both modellers and users/end-users of geological models.
How to cite: Madsen, R. B., Møller, I., Falk, F., Troldborg, L., and Høyer, A.-S.: From Deterministic to Probabilistic: Quantifying Layer Boundary Uncertainty in Hydrostratigraphic Models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8372, https://doi.org/10.5194/egusphere-egu26-8372, 2026.
Within the framework of the National Geological Model (NGM), a long-termed federal program (2022–2030), the Swiss Geological Survey (swisstopo) is developing a series of three-dimensional geological models at national scale. The primary objective is to achieve full spatial coverage of Switzerland with harmonized 2D and 3D geological models representing the geometry of major tectonic structures, lithostratigraphic units, and the bedrock surface. These models form a consistent geological framework that supports sustainable subsurface use and long-term spatial planning.
Switzerland comprises three principal geological domains with contrasting structural styles and stratigraphic architectures: the Jura fold-and-thrust belt, the Foreland Plateau, and the Alpine orogenic domain. These domains differ significantly in terms of deformation mechanisms, lithological complexity, data availability, data type and depth of geological investigation. This requires domain-specific modelling strategies and tailored approaches to uncertainty management. In addition, subsurface utilization and associated societal demands, such as infrastructure development, groundwater management and hazard assessment, vary markedly between regions.
The 3D modelling group at swisstopo has implemented a domain-based modelling strategy by subdividing Switzerland into three regional modelling areas corresponding to the main geological domains. For each domain, regional-scale 3D geological models are constructed through the integrated interpretation of surface geological maps, borehole and geophysical data, cross-sections and geological concepts and constraints. These models provide a consistent structural and stratigraphic framework that translates traditional geological mapping into digital, reproducible subsurface representations suitable for national-scale applications.
This contribution presents an overview of the current status of four complementary modelling projects developed by the 3D Group at the Swiss Geological Survey: swissBEDROCK, Jura3D, GeoMol, and swissAlps 3D.
swissBEDROCK provides a nationwide 3D bedrock model of Switzerland based on an automated and reproducible workflow with explicit uncertainty representation and regular versioned updates. Jura3D focuses on high-resolution structural and stratigraphic modelling of the folded and thrust-faulted sedimentary sequences of the Jura fold-and-thrust belt. GeoMol addresses the Foreland Plateau at regional scale, emphasizing stratigraphic architecture and basin geometry. swissAlps 3D targets the structurally complex Alps, with a strong emphasis on the tectonic development of the main lithostratigraphic and structural units supported by scientific argumentation. This contribution further highlights the importance of collaborative workflows involving federal and cantonal authorities, academia, and private partners in the development of consistent national 3D geological models.
These projects together illustrate how diverse geological modelling approaches are integrated within a coherent national framework. Moreover, they bring together geological knowledge and 3D modelling workflows across contrasting geological domains.
How to cite: Kurmann-Matzenauer, E., Wehrens, P., Musso Piantelli, F., Signer, S., Ursprung, A., and Reynolds, L.: 3D geological modelling at the Swiss Geological Survey: Development of national-scale models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8046, https://doi.org/10.5194/egusphere-egu26-8046, 2026.
BRGM - the French Geological Survey – has launched an intriguing video series comprised of seven episodes that reveal the subsurface in 3D, offering a unique perspective on its applications across various fields. Topics include water resources, geothermal energy, natural risk, mineral resources, anthropic risk, and geological knowledge and training, along with significant insights into methodologies and tools that have been developed.
Each episode is designed to provide perspectives into how these different areas benefit from advanced geological modelling. Scenarios highlight the value of 3D approach both to describe geology and as a framework for simulating real-world processes. The stories are narrated from the perspective of practical applications, which makes them accessible and engaging for viewers. The collaboration with L’Esprit Sorcier TV enhances the production quality and ensures that complex information is presented in an engaging and accessible manner. Viewers can expect to see a blend of expert insights, practical applications, and captivating visuals, making the content both informative and enjoyable.
The episodes provide an essential resource for scientists, students, professionals and stakeholders in relation with the presented topics, and anyone interested in expanding their understanding of geology. By delving into real-world applications and contemporary issues, this series provides perspectives on how geological knowledge can inform better decision-making in various sectors.
Don’t miss the opportunity to explore these engaging episodes and renew your view of subsurface geology and its implications in our everyday lives.
This engaging series is freely available in French language with English subtitles on the BRGM’s YouTube channel:
https://www.youtube.com/playlist?list=PLfgMUGQz1vBPClcglLDF74GZrJQ0u6qrA.
Selection of references for the applications depicted in the series; more are available in the end credits of each episode:
- Audion, A.S. BRGM report BRGM/RP-62718 (2013)
- Burnol, A. et al. Remote Sensing 15, 2270 (2023) doi: 3390/rs15092270
- Calcagno, P. et al. Phys. Earth Planet. Inter.171, 147-157 (2008) doi: 1016/j.pepi.2008.06.013
- Courrioux, G. et al. 17th IAMG Conf proc. pp. 59-66 (2015)
- Janvier, T. BRGM report BRGM/RP-73278 (2023)
- Mas, P. et al. Sci Data 9, 781 (2022) doi: 1038/s41597-022-01876-4
- Saltel, M. et al. Hydrogeol J. 30, 79-95 (2021) doi: 1007/s10040-021-02410-3
How to cite: Calcagno, P., Burnol, A., Caritg, S., Janvier, T., Lopez, S., Saltel, M., Serrand, A.-S., Fauquet, J., Groc, B., Lievin, E., and Vassal, P.: Revealing the Subsurface in 3D: A Series of Short Films Focusing on Recent Applications, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12149, https://doi.org/10.5194/egusphere-egu26-12149, 2026.
Artificial intelligence is often expected to revolutionise geological modelling, but in practice its performance is strongly controlled by how geological information is collected, encoded, constrained, and by how well the AI workflow is tailored to the task. In this contribution we analyse what helps and what hurts AI-based geological modelling under data-scarce conditions, using shallow geothermal modelling in Cyprus as a testbed.
Within the WAGEs project on shallow geothermal energy, we compiled borehole profiles from across Cyprus, harmonising heterogeneous lithological descriptions into a simplified but consistent scheme and linking them to tectonic units and basic spatial information. Classical, off-the-shelf neural-network approaches performed poorly on this limited and noisy dataset, highlighting the vulnerability of generic architectures to inconsistent lithological classifications and incomplete metadata.
We therefore developed a tailored, sequence-based machine-learning workflow in which each borehole is encoded as a one-dimensional string combining depth-ordered lithologies, tectonic context, and location. A supervised learning algorithm was trained on existing boreholes and tested on independent control sites. In phase-one experiments, the model reached about 85% accuracy when the two top-ranked predicted lithological profiles were considered for the full borehole depth. This metrics was selected due to existing rock types that may be easily misclassified (marl-chalk) or interpreted (decayed rock at the surface – rock, soil or surface deposit). Algorithm’s skill was highest where lithological contrasts were strong, while more gradational successions remained difficult to distinguish. The model showed partial ability to infer the presence of faults from lithological patterns, while it was not designed to localise them nor supplied with relevant information.
From this case study we distil key factors that help tailored AI-based geological modelling (standardised, information-rich lithological logs; task-specific encoding that reflects geological settings; explicit tectonic context) and those that hurt it (lack of identification protocol; inconsistent rock descriptions; loss of detail during digitization). Our results indicate that robust AI-based geological modelling does not necessarily require massive datasets, as long as the available information is consistent and well structured. However, in data-scarce settings the main ceiling for AI performance is informational rather than algorithmic: more complex models add little once the underlying geological description is noisy or underspecified. In practice, tailored workflows are most powerful as tools for scenario ranking and for identifying where additional boreholes or geophysical surveys would most effectively reduce subsurface uncertainty, rather than as engines for fully automatic geological models. We conclude that the community should treat AI primarily as a tool for rapid, big-picture or illustrative geological modelling and for stress-testing geological knowledge. Its main value lies in exposing gaps in our subsurface descriptions (including quantitative uncertainty estimates), rather than providing a shortcut that can replace careful geological thinking.
How to cite: Ciapala, B., Papaefthymiou, E., Aresti, L., Pasias, D., Graikos, D., Florides, G. A., and Christodoulides, P.: What helps and what hurts tailored AI in geological modelling: beyond the hype, evidence from data-scarce shallow geothermal modelling in Cyprus, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17722, https://doi.org/10.5194/egusphere-egu26-17722, 2026.
Three-dimensional geological models can be used to simply return a visual representation of complex subsurface structures; however, when they are used to define the geometry and properties of bodies used in downstream numerical simulations (e.g., geothermal, geomechanical, and/or fluid flow simulations), their application is limited by the difficulty in generating computational meshes that preserve the geological topology. In particular, intersecting faults, unconformities, and stratigraphic contacts present challenges because numerical simulations require watertight models, with consistently defined surface intersections that do not pose any ambiguity whatsoever regarding the attribution of a certain 3D region to a given closed volume. As such, to generate watertight models and meshes is the critical step that quite often hinders practical downstream applications of geological models.
We present PyMeshIt (https://github.com/waqashussain117/PyMeshit), a pure-Python open-source modelling engine that addresses this bottleneck by automating the generation of conforming tetrahedral meshes from complex geological interpretations. PyMeshIt is available both as a standalone application and as an integrated meshing engine within the PZero geological modelling platform (https://github.com/gecos-lab/PZero), supporting a wide range of geomodelling workflows without imposing assumptions on downstream simulations.
PyMeshIt implements an interactive multistage workflow that supports point clouds, triangulated surfaces, well trajectories, and model boundaries. The central focus of the software is the explicit preservation of geological/topological relationships during meshing. Surface-surface and polyline-surface intersections are computed automatically, producing intersection polylines that trace fault cutoffs, unconformity truncations, and formation contacts. Locations where three or more geological features converge are identified as triple points and are retained as topological constraints. These intersections and junctions are used as constraints during surface reconstruction and volumetric meshing to ensure that element faces align with the geological boundaries in the final mesh.
Material regions are assigned through interactive seed-point placement, allowing tetrahedral volumes to be consistently attributed to geological units. The output formats include VTK/VTU for visualisation, STL for CAD applications, and EXODUS II for numerical modelling frameworks. When used within PZero, PyMeshIt directly accesses the geological model entities without intermediate file conversion, preserving pre-triangulated geometries and allowing the possibility of creating geological interpretations within a single framework, thereby ensuring a complete open-source workflow from geological interpretation and modelling to meshing.
How to cite: Hussain, W., Cacace, M., Bistacchi, A., and Monti, R.: PyMeshIt: An Open-Source Python modelling engine in PZERO and a standalone software for Conforming Tetrahedral Mesh Generation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16992, https://doi.org/10.5194/egusphere-egu26-16992, 2026.
