This session focuses on the intersection of mapping, modelling and visualization in geosciences, highlighting innovative tools, workflows, and applications.
We invite contributions addressing:
- Advanced techniques for mapping, analyzing, and visualizing data from all compartments and scales of the Earth;
- Geological field mapping;
- 2 D, 3 D and 4 D geological modelling;
- Co-designed tools that integrate interdisciplinary datasets for research and stakeholder engagement;
- Platforms and workflows for managing and visualizing high-frequency, high-dimensional, and metadata-enriched datasets;
- Harmonization of geoscience datasets using FAIR principles (e.g. cross-boundary data);
- Portrayal and presentation;
- Education and training initiatives for mapping, modelling, and visualization techniques.
By bringing together scientists, developers, and practitioners, this transdisciplinary session aims to foster a collaborative community, advancing mapping and visualization tools and ensuring their accessibility and impact across geosciences. The outcome will be applicable to and benefit various usages (e.g. mineral exploration geological risk assessment, protecting groundwater resources and coastal protection).
Orals: Tue, 29 Apr | Room -2.92
Transboundary geological baseline information, such as geological maps, datasets and 3D models following the FAIR data principles, is still scarce at European level. This is due to (1) lack of transboundary harmonisation, (2) lack of FAIRness of existing harmonised data, (3) non-existence of such data and (4) lack of a suitable framework to provide such data.
The Geological Mapping & Modelling Expert Group of EuroGeoSurveys gathers the European Geological Surveys Organisations (GSO) and establishes the data collection and model production framework as mentioned in (4). This is being done as part of the ongoing Coordination & Support Action "Geological Service for Europe"(GSEU) of the EU Horizon Europe Framework Programme.
Ongoing actions include (1) collecting metadata of geological maps, datasets and 3D models of Europe, (2) establishing a conceptual and physical data model capable of accommodating multiscale basic geological data as well as applied geoscientific data in 2.5D, (3) creating improved or new scientific vocabularies for lithology, anthropogenic deposits and lithotectonic units based on linked data and SKOS technology, (4) building a lithotectonic spatial map database using the vocabularies developed in (3), and (5) sharing experience and best practices in 3D geomodelling to federate European GSO around common approaches, as much as possible employing open source tools.
In addition to the technical framework, we are also working on appropriate organisational structures and workflows for the maintenance and future updating of all elements of the framework and its datasets.
Research and innovation needs for the next EU Framework Programme in the field of geological mapping and modelling have recently been addressed in the Strategic Research and Innovation Agenda (SRIA) of EuroGeoSurveys. These include additional scientific vocabularies (e.g. for lithogenetic units, structural and geomorphological features), tools for (semi)automatic generalisation of datasets, and a collaborative GIS platform for the user community for efficient data harmonisation. In addition, the design of innovative incentive schemes to stimulate the collection of still missing data and new technologies for subsurface exploration and geomodelling to produce geological maps and 3D models, are important parts of the SRIA.
Working well together as a large group of people with a wide range of skills and experience is essential to provide applications, such as those dealing with resources and risk, with the quantified and reliable geological information they need for their processes.
How to cite: Krenmayr, H. G., Asch, K., Calcagno, P., Čápová, D., Le Guern, C., Lopez, S., Németh, Z., Piessens, K., and Stępień, U.: A framework for making available Europe’s treasure of geological basic information: A collaborative effort, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12511, https://doi.org/10.5194/egusphere-egu25-12511, 2025.
Lithology is one geological key attribute on geological maps, serving as a fundamental framework for understanding the composition and structure of the Earth's crust. The development of a common lithological vocabulary has been essential for the harmonisation of geological maps at national and international level, enabling the creation of a semantically consistent map of Europe. Several international projects such as OneGeology-Europe, IQUAME, GSEU and numerous national initiatives like Polish GeoTezaurus, have provided valuable experience in this area.
These projects have highlighted the need for a unified approach to lithological mapping, bringing together geologists and petrologists to address a number of challenges. The production of standardised lithological maps requires cooperation and expertise, particularly in dealing with the different interpretations and terminologies used in different regions and languages.
One of the main challenges is the existence of multiple lithological classifications, each reflecting a different geological perspective. Classifications can vary based on factors such as grain size, composition and structure, complicating the process of creating a consistent mapping standard. In addition, many lithological terms are closely linked to various geological interpretations such as genetic process and depositional environment, further complicating the mapping process.
Another major challenge is the problem that the same lithological term can have different meanings depending on the context in which it is used. Therefore, a clear definition of a concept is much more important than which exact label (term) is chosen to represent that concept. A particular problem arises when translating terms between languages, as the meaning of a term can shift or become ambiguous in different linguistic and geological contexts. In addition, the translation process itself often reveals further problems, as terms that accurately describe geological features in one language may not have an exact equivalent in another. Comparing geological maps, we can find differences even in national datasets, such as using the singular or plural form of terms. Some languages have additional complications due to a rich vocabulary and a multitude of synonyms. This can lead to the creation of new descriptive terms, or adjustments to existing terminology.
Furthermore, national languages present additional challenges in ensuring consistency and clarity in lithological definitions. The process of translating scientific terms into national languages often reveals subtle differences in meaning and interpretation that may not be immediately apparent. As a result, new descriptive terms may be required to accurately convey the intended geological concepts, adding another layer of complexity to the standardisation effort.
Special attention must also be given to lithogenetic terms, particularly those associated with Quaternary surficial deposits. These deposits often present unique classification challenges due to their complex nature. Replacing lithogenetic terms with their very rich meanings, which include lithology, environment, process and form, by litological ones only, reduces map data content.
In conclusion, the development of a standardised lithological vocabulary, both nationally and internationally, is a complex but essential task for the advancement of geological research and communication. By addressing these challenges through collaboration and expertise, the global geological community can work towards a more integrated and comprehensive understanding of the Earth's lithology.
How to cite: Stępień, U., Asch, K., Bergman, S., Novak, M., Pantaloni, M., Bauer, H., Hackmann, P., and Krenmayr, H.-G.: The geological Gordian knot - lithological challenges in the world of geological mapping, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9972, https://doi.org/10.5194/egusphere-egu25-9972, 2025.
The Geological Service for Europe (GSEU) project is a five-year EU-funded Coordination and Support Action (CSA) that unites 35 geological surveys and 13 additional partners from 36 countries. Its primary objective is to establish a Geological Service for Europe (GSE), a geoscience-driven initiative delivering comprehensive data, information, and insights about Europe’s subsurface. The project addresses key challenges, including the sustainable management of critical raw materials, geothermal energy resources and storage, and groundwater systems. Work Package 6 (WP6), is dedicated to providing reference geological knowledge at European level: harmonized maps, concepts and 3D geological frameworks. The project is also anchored on the European Geological Data Infrastructure (EGDI) platform, which acts as a central repository for subsurface data collected before and during the project. This platform also serves as a dynamic resource, offering open access to a wide array of 3D geological models for diverse stakeholders.
Within WP6, Task 6.3 focuses on consolidating tools and best practices for designing and visualizing 3D geological models. This involves identifying and evaluating relevant tools, documenting their applications, and providing expert guidance to support the creation of shareable and reusable models. The task also tackles common challenges, emphasizes the strengths of individual tools, and explores opportunities for future advancements in modeling workflows.
To achieve these goals, we initiated efforts to inventory practices across European geological surveys and share in-house tools. Twice a year, we are also organizing open webinars on modeling workflows and related tools. These initiatives foster transparent and constructive discussions around 3D geological modeling methodologies. A core focus is ensuring accessibility and interoperability through the adoption of modular and open software components, enabling models and workflows to be easily adapted to diverse needs. The objective it to empower the geological surveys community to innovate and enhance existing frameworks.
The final deliverable of Task 6.3 is a comprehensive report that will consolidate best practices and documents available open-source tools. Beyond that, our action seeks to establish a sustainable network of 3D geological modeling experts across European geological surveys, fostering long-term collaboration and knowledge sharing.
How to cite: Lopez, S., Clausolles, N., Brogaard Pedersen, C., Calcagno, P., Colomer, M., d'Ambrogi, C., Kearsey, T., Heerema, C., Herms, I., Hölzel, M., Marin Lechado, C., Palotaï, M., Pizzella, L., Spoerlein, T., Stafleu, J., Stepien, U., Svagera, O., Szynkaruk, E., Terrington, R., and Wiese, M.: Towards a common geomodelling toolbox: sharing practices among European Geological Surveys, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13319, https://doi.org/10.5194/egusphere-egu25-13319, 2025.
Since the defining of plate tectonics in the 1960s the knowledge on geodynamics of European territory registered a remarkable progress in understanding of the multiple – polyorogenic overprints within individual lithotectonic zones and units of Europe, becoming by this way an etalon for orogenic interpretations to the rest of the world.
According to the classification by Neuendorf et al. (2011) and in a slightly modified form by the INSPIRE Geology data specifications (http://inspire.ec.europa.eu/codelist/GeologicUnitTypeValue/lithotectonicUnit), a Lithotectonic unit is a geologic unit defined on basis of structural or deformation features, mutual relations, origin or historical evolution. Contained material may be igneous, sedimentary, or metamorphic.
The Lithotectonic map of Europe will represent a combination of a lithological and a tectonic map, showing a collage of lithotectonic units and their boundaries, highlighting the geodynamic aspects of more than 2.5 billion years of crustal evolution.
This novel type of map provides a wealth of information through annotated data, contributing to the development of applied research, including raw materials exploration, environmental geology, geo-energy, etc. The evolution of lithotectonic units can be placed in that of orogenic cycles, which include the Svecokarelian, Sveconorwegian, Cadomian, Caledonian, Variscan, Alpine and Hellenic cycles. Different orogenic phases will be discriminated in these orogenic cycles. For the lithotectonic framework it is sensible to emphasize those orogenic phases which differ from the standard orogenic (Wilson) cycle of the 1960s: the post-orogenic phases of unroofing, intraplate stress elimination and regional extension.
The Lithotectonic map of Europe that is currently being compiled will be based on International Geological map of Europe and Adjacent Areas (IGME 5000; Asch, 2005, BGR, Hannover). Co-funding is provided by the EC – CINEA HORIZON-CL5-2021-D3-D2 project 101075609 Geological Service for Europe (GSEU), led by EuroGeoSurveys and its Geological Mapping and Modelling Expert Group, Work Package WP6 – Geological framework for the European geological data & information system.
Reference
Neuendorf, K.K.E., Mehl Jr., J.P. & Jackson, J.A., 2011: Glossary of Geology. Fifth Edition. American Geosciences Institute, Alexandria, Virginia, 1–779.
How to cite: Németh, Z., Asch, K., Krenmayr, H.-G., Pubellier, M., Piessens, K., Pascual, F. J. R., Padel, M., Luth, S., Pelech, O., and Heckmann, P.: Lithotectonic map of Europe – methodology, contribution to geosciences and further inspiration for territories outside Europe, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13479, https://doi.org/10.5194/egusphere-egu25-13479, 2025.
The objective of this study is to synthesize the geological and hydrogeological characteristics of the main aquifers in the Paris Basin within the Cenozoic sedimentary formations. This work is part of the multi-annual scientific program RGF (Référentiel Géologique de la France, https://rgf.brgm.fr/page/bassin-parisien), which aims to update the representation and mapping of French geology on a national scale. The program also funds several PhD projects in collaboration with the academic community, focusing on the geometry, distribution of petrophysical properties, modeling, and mapping of key Cenozoic geological formations in the basin. Particular emphasis is placed on the major aquifers as part of this modeling effort.
The geology of the Paris Basin has been extensively documented by various authors (Pomerol, 1967; Mégnien, 1980; Gély and Lorenz, 1991; Briais, 2015). Data from well logs and core descriptions collected from over 2,000 petroleum wells have been used to reconstruct the stratigraphic surfaces of the main formations, while also identifying the large-scale geometries of associated aquifers and aquitards. Recent studies have enhanced the dating of specific stratigraphic markers (Marlot, 2023; Moreau, 2023), described the geometries of alluvial formations in the Seine River (Chourio-Camacho, 2024), advanced knowledge in structural geology (Brown, 2024), and provided petrophysical characterizations of reservoir rocks (Moreau, 2023; Marie, 2024).
The hydrogeology of the Paris Basin has been studied and modeled for decades (Mégnien, 1980; Goncalves, 2003; Lamé, 2013). The lateral extent of aquifers varies significantly across regions. For instance, in Île-de-France, the hydrosystem comprises six primary aquifers: the Alluvial, Brie, Champigny, Lutetian, Ypresian, and Chalk aquifers. However, lateral facies variations can significantly alter hydrogeological properties, influencing groundwater resource potential. In the Oligocene formations, the main aquifers are primarily located in Île-de-France and the northern part of the Centre-Val de Loire region: i) on the Beauce plateau, commonly referred to as the "Beauce aquifer," ii) in the Yvelines area, primarily associated with the Fontainebleau Sands, and iii) on the Brie and Bière plateaus, where they are predominantly contained within the Brie Limestone.
These lateral facies variations, coupled with the presence of fractures or karstification, result in substantial differences in the petrophysical properties of the identified aquifers and aquitards. Pumping test data have been compiled and converted into permeability and transmissivity coefficients, which were subsequently mapped along lateral transects in the Brie region of the basin (Marie, 2024).
This study will also contribute to the harmonization of the lithostratigraphic framework across the 187 geological maps covering the territory. Furthermore, the 3D model will facilitate vertical and lateral interpolation of hydrological reference data from the BDLISA database (https://bdlisa.eaufrance.fr/), which currently provides detailed mapping of water bodies at the scale of metropolitan France.
How to cite: Audigane, P., Briais, J., Allier, D., Grataloup, S., Klinka, T., and Brugeron, A. and the RGF Bassin parisien Team: 3D Characterization of Aquifers in the Cenozoic Sedimentary Formations of the Paris Basin, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6265, https://doi.org/10.5194/egusphere-egu25-6265, 2025.
Exploration data (e.g. borehole records, geophysical sections) represent the essential input data for any geological model. Borehole records often play a decisive role in geological modeling. Usually, they contain descriptions and interpretations on petrography, lithology and stratigraphy. This information is crucial for modeling the spatial distribution of lithostratigraphic layers in three dimensions. However, these interpretations can be inconsistent or error-prone. The reasons include for instance, the date of recording (reflecting the prevailing state of knowledge at the time), the specific exploration target and techniques employed, the quality of digitalization, and the potential for human interpretative bias. Separating adequate from unsuitable borehole records is of great importance, yet rather difficult, especially evaluating large datasets. While visual inspection of the inferred geological model is a viable approach, it results in numerous iterations to identify inadequate drilling profiles, which is time-consuming and expensive.
In order to streamline the process of testing the quality of the data from borehole records, we developed the Python-based software package B-QualMT (borehole quality management tool) that can filter borehole records based on user-adjustable standards. The tool has a given set of deterministic tests depending on the user’s auxiliary information (e.g. previous 3D-models) and knowledge of the regional geological settings (e.g. sequence of geological layers), which can be used to select divergent drilling profiles for geologically comparable regions.
For our pilot study, we selected a former lignite mining area of Lusatia in the southeast of Germany bordering the Federal State of Brandenburg and the Free State of Saxony (Freistaat Sachsen). Here, the goal is to improve the previous geological model with 3000 additional borehole profiles from various exploration surveys spanning several decades. We show the evaluation process, how the deterministic tests work and will additionally give an outlook on the planned integration of machine learning algorithms identifying geological patterns in previously quality-tested borehole records.
How to cite: Schönfeldt, E., Hiller, T., Fahle, M., Giese, J., Hübschmann, M., and Grafe, F.: B-QualMT - A software quality management tool to select borehole records for 3D models, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20862, https://doi.org/10.5194/egusphere-egu25-20862, 2025.
Climate change in the Arctic is both a potential threat and a potential opener for new opportunities in sustainable development of the region. Better access to data, methods and tools, as well as better documentation of these are needed to advance science and support decision-making on public and private sector. While local, national and regional communities in different parts of the Arctic face individual challenges due to climate change, they experience similar problems when gathering and analysing data to address these issues.
The Arctic Ocean is one of the least explored oceans on the planet. There is lack of in situ observations in large parts of this region, especially on the seafloor and in the ocean under the sea ice. This means that the research communities working with climate, weather, ice-ocean processes, and geophysical hazards have limited knowledge about the processes below the sea ice. The lack of data limits the possibility to advance research in this region. It is therefore necessary to establish observing systems for data collection in the Arctic Ocean.
Fisheries plays a key role in the economy in many Arctic countries. With climate change affecting marine ecosystems and enabling accessibility of larger ocean areas in the region, there is a strong need for improved access to data and information tailored to the fisheries industry and the public sector monitoring and regulating Arctic Fisheries. Relevant data is available from many different sources, but efficient delivery chains for compiling, integrating, and analysing these in a common system is lacking.
These diverse stakeholder groups all need access to data from different sources, such as ice and underwater ocean observing systems, satellites, operational forecasting services and climate models, which provide data with different spatial and temporal resolution, in different formats and with varying levels of documentation. Furthermore, methods and tools are needed for data processing, analysis and visualisation to integrate heterogeneous multi-source data with reference and socio-economic data to support science driven as well as sector specific applications.
Blue Insight developed by Kongsberg Discovery AS offers a robust, modular platform designed for the processing, visualization, and sharing of ocean data. The core module integrates a cloud framework, data visualization tools, and comprehensive data storage and management capabilities. In addition, the system includes a data processing framework that facilitates the reuse of data processing methodologies. To enhance Blue Insight's functionality and cater to projects of varying scales, additional modules can be seamlessly integrated into this framework. This is done by means of container and workflow engine technologies, ensuring interoperability with existing digital twin for the ocean (DTO) and other research infrastructure initiatives through OGC standards.
The presentation will give an overview of the Blue Insight system and showcase how this digital platform is used and extended within the Horizon Europe HiAOOS project and the SBEP ARCFISH project.
How to cite: Hamre, T., Hestnes, A. J., Sagen, H., Bildøy, L. E., Hansen, K. H., Storheim, E., and Monsen, F.: Blue Insight – Using digital platforms for Arctic Ocean Science and sustainable fisheries, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17232, https://doi.org/10.5194/egusphere-egu25-17232, 2025.
The Mediterranean region faces critical challenges related to land degradation, requiring innovative and harmonized approaches for assessment and restoration. To address these challenges, the Land Degradation Decision-Support Toolbox (LanDS) has been developed as part of the REACT4MED project (https://react4med.eu/), funded by PRIMA (https://prima-med.org/). LanDS serves as a comprehensive and adaptable platform designed to evaluate land degradation and assess the impacts of restoration measures across diverse Mediterranean contexts.
LanDS exemplifies a novel approach to combating land degradation by bridging scientific research and practical application, promoting sustainable development, and supporting climate adaptation efforts in the Mediterranean region.
The toolbox integrates five core components:
- Geo-referenced Data Repository: a centralized knowledge base that aggregates site-specific data and resources from the project’s ecosystem restoration living labs, alongside broader datasets from global and regional repositories and satellite-based indices.
- Data Viewer: a suite of interactive visual analytics tools enabling effective data visualisation, sharing among project partners and stakeholders, and monitoring of restoration actions.
- Indicators Library: a modular and adaptable code library offering a wide array of indicators supporting analysis and comparisons across different spatial and temporal scales, drawn from an extensive dataset built from global repositories and project’s data.
- Machine-Learning-Based Procedure: a cutting-edge tool designed to identify and map potentially suitable areas for upscaling and outscaling restoration measures across the Mediterranean region.
- Interactive Web Dashboard: a user-friendly interface that delivers harmonized assessments of land degradation and evaluates the effectiveness and impact of the project’s restoration measures, while supporting dissemination of project findings.
By synthesizing knowledge from global and regional datasets with insights from living labs in pilot areas, LanDS facilitates informed decision-making for land restoration and sustainable resource management. The platform fosters the development of policy recommendations and investment opportunities aimed at addressing land degradation in the Mediterranean. Further, it enables policymakers, stakeholders, and private actors to identify investments opportunities based on maximum cost-effectiveness and impact criteria.
Built on an open-source technology stack, the LanDS toolbox ensures accessibility and transparency and is freely available at http://lands.soft-water.it.
How to cite: Matta, E., Micotti, M., Corti, S., and Weber, E.: Advancing Land Degradation Assessment and Restoration Planning in the Mediterranean through the LanDS Toolbox, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12904, https://doi.org/10.5194/egusphere-egu25-12904, 2025.
Earth system science increasingly relies on complex, high-dimensional datasets from advanced modelling and remote sensing. Transforming these data into actionable insights for stakeholders and decision-makers beyond academia remains a major challenge. In response, we developed a novel, open-source graphical user interface (GUI) for the CRAFTY (Competition for Resources between Agent Functional Types) agent-based land-use model. Built using JavaFX and available on GitHub, this GUI bridges the gap between model complexity and user accessibility. Key features of CRAFTY-UI include: (1) an intuitive workflow for configuring Agent Functional Types (AFTs) that represent diverse land managers, (2) real-time visualization of simulation progress, and (3) input/output data analysis. Users can customize model parameters such as AFT behavior, productivity settings, societal demand for ecosystem services (including pricing), and multiple protected-area and policy restrictions. Additionally, they can select from different mechanisms such as competition algorithms, mutation intervals, and neighbourhood effects according to their specific research or policy scenarios.
CRAFTY relies on a broad range of input data including land resource maps (capital maps) for each year under multiple SSP-RCP scenarios, along with AFT parameterizations, ecosystem service demands, and protected-area restrictions. The CRAFTY-UI enables users to analyse these datasets across spatial, temporal, and scenario dimensions. In addition to monitoring simulations in real time, users can also import previous runs and employ difference analysis tools to compare outcomes under varying parameters and mechanisms. These features foster a deeper understanding of how social, economic, and environmental capitals interact to shape land-use trajectories. Beyond its advanced analytical capabilities, the CRAFTY-UI simplifies large-scale simulation exploration, offering adjustments that instantly update interactive charts and spatial outputs. By clarifying feedback loops between natural capital, socio-economic drivers, and institutional influences, the interface facilitates evidence-based decision making. Ultimately, the CRAFTY-UI bridges complexity and usability, enabling diverse stakeholders to engage with modelling outputs, enrich policy discussions, and collaboratively shape sustainable land-use strategies.
How to cite: Byari, M., Zeng, Y., Brown, C., and Rounsevell, M.: CRAFTY-UI: Bridging Complexity and Usability in Large-scale Agent-Based Land-Use Modelling for Earth System Science, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20452, https://doi.org/10.5194/egusphere-egu25-20452, 2025.
In spite of the availability of satellite data, its complexity has prevented widespread adoption into the practices of small and independent farmers. We present in this paper a system (AgriVision AR) designed to aid user interaction with satellite data sources such as ESA’s Copernicus Data Space Ecosystem. AgriVision AR uses mobile augmented reality to simplify data visualisation through intuitive interaction. Augmented Reality (AR) is a technology that links together the physical and the virtual worlds. In AR, digital information seamlessly blend with the real world, creating a perception of immersion of the user. Geospatial information is overlaid over the real environment making more easily to understand the data and increasing user immersion. The application allows users to visualise agricultural indices and other information gathered through satellite imagery in the form of an animated color overlay on small scale landscapes. This approach will enable farmers to make informed decisions about crop management. It will also allow to optimize resource allocation, and mitigate localized issues such as soil fertility and pest infestation. AgriVision AR represents a step towards empowering small-scale farmers with advanced technology, fostering sustainable agricultural practices in the era of precision farming.
How to cite: Bacu, V. and Radu, M.: Augmented reality solution for visualization of agricultural data gathered from satellite imagery, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3976, https://doi.org/10.5194/egusphere-egu25-3976, 2025.
Posters on site: Wed, 30 Apr, 10:45–12:30 | Hall X4
The growing volume of high-resolution time-series data in Earth system science requires the implementation of standardised and reproducible quality control workflows to ensure compliance with the FAIR data standards. Automated tools such as SaQC1 address this need, but lack the capacity for manual data review and flagging. We are therefore planning to develop a Python-based tool with an intuitive graphical user interface (GUI) for local machines, thereby enhancing the functionality of SaQC. It is anticipated that the tool will be user-friendly, even for those with limited experience of Python. The GUI will be capable of interactively visualising the time-series data, highlighting the data that has already been automatically flagged. The selection of data points may be accomplished by clicking on them or via box-selection, and a flag may be assigned via a dropdown menu. An optional comment field can be used to record supplementary information, such as details of pollution events. Moreover, the option to unflag data that has failed the automated quality control process, but which is considered valid by the scientist, will be available.
The manual flagging tool will be based on SaQC, thereby facilitating a future integration into this software package. Consequently, integration into an existing SaQC workflow will be straightforward. It should be noted, however, that this is not exclusive to SaQC users; it can be easily applied to data created by another tool for automatic quality control. A simple conversion of the data via the pandas library will be sufficient for utilisation of the manual flagging tool. The flagging schemes can either be adopted from SaQC or user-specific schemes can be integrated. Once the flagging process is completed, the user is able to decide how to export the data set.
The manual flagging tool represents a valuable addition to existing toolkits for all scientists handling time-series datasets, effectively completing the data quality control process. From a scientific perspective, the benefits of this tool include increased efficiency and traceability in the data flow, as well as improved data quality through the fine-tuning of automatic controls based on experience and contextual knowledge.
1 Schäfer, David, Palm, Bert, Lünenschloß, Peter, Schmidt, Lennart, & Bumberger, Jan. (2023). System for automated Quality Control - SaQC (2.3.0). Zenodo. https://doi.org/10.5281/zenodo.5888547
How to cite: Büttner, N., Louisot, B., Lorenz, C., Schäfer, D., and Fösig, R.: Manual data review and quality control – An add-on to SaQC, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16190, https://doi.org/10.5194/egusphere-egu25-16190, 2025.
The open-source Data Analytics Software Framework (DASF) available at https://dasf.readthedocs.io is an advanced remote procedure call (RPC) framework designed to abstract Python code and make it securely callable over the internet. This framework ensures that computing resources remain protected from direct exposure to the internet. The latest innovation within DASF is the development of a flexible framework to automatically generate a web frontend for computing resources.
The basic concept of DASF lies in the transformation of original Python code into a JSON schema. This schema serves a dual purpose: it is used to generate the web frontend and to validate the input received via the web interface. By converting Python code into a JSON schema, DASF leverages the underlying code to create a dynamic and interactive web frontend. This approach simplifies the deployment of web interfaces and additionally ensures that the input data conforms to the expected format, thereby reducing the likelihood of errors and enhancing the overall user experience. The web interface includes forms, input fields, and other interactive elements necessary for user interaction. This automated generation process eliminates the need for manual coding of the web frontend, saving time and reducing the potential for human error, and makes it especially useful for scientists without background in web development.
One of the key benefits of this approach is the seamless integration of the web frontend with the underlying computing resources. Users can interact with the web interface to submit data and trigger computations, all while the JSON schema ensures that the input data is correctly formatted and validated. This validation step is crucial, as it prevents invalid data from being processed, which could otherwise lead to errors or security vulnerabilities. This security feature is particularly important in a web-based environment, where the potential for unauthorized access and data breaches is higher.
Our framework is especially useful for digital twins in the earth system sciences, where considerable amounts of data and computing resources are often required. Digital twins are virtual representations of physical systems, and they rely on such resources to function effectively. DASF is particularly well-suited for these applications because it allows the code to run on high-performance computing (HPC) resources without exposing them to the internet. This ensures that sensitive infrastructure remains protected while still providing the necessary computational power.
In our presentation we show how DASF works and provide live examples of the frontend implementation.
How to cite: Sommer, P. S., Saß, B. L., and Benninghoff, M.: Innovative Web Frontends for a secure access to High-Performance Computing resources via DASF, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3120, https://doi.org/10.5194/egusphere-egu25-3120, 2025.
Communicating marine geohazards to stakeholders can be challenging, and traditional media may not be sufficient to convey the full range of processes involved. In addition, news images of devastating events such as tsunamis and volcanic eruptions can be associated with prejudice and fear, hindering fact-based awareness-raising. With recent advances in computer graphics and virtual reality headset hardware, immersive visualization methods are becoming accessible to a wider scientific community. Virtual presentations of different local scenarios provide an opportunity to discuss with experts, policy makers and the general public, overcoming abstraction and prejudice and transforming scenarios into realistic and spatially explicit experiences.
The MULTI-MAREX collaborative project is establishing a living lab in the Aegean Sea to study extreme marine geological events and associated hazards, with the aim of developing the knowledge needed to manage geohazards at different scales. Digital reconstruction of real, physical study sites leads to and enhances situational awareness, resulting in a personalized, in-depth understanding of local scenarios. Accessibility of the communication format is important to reach the target user.
We are exploring a wide range of hardware options, from the portability and convenience of smartphones, to the highly immersive experiences offered by head-mounted displays, through immersive simulators such as video walls and dome theatres. These diverse platforms ensure that immersive visualizations are adaptable to different user needs and environments, facilitating greater accessibility and engagement across stakeholder groups. We focus on developing workflows for geoscientists to enable semi-automated, asset-enhanced, immersive visualization that synthesize collected remote sensing data, such as terrestrial and marine digital outcrop models, hydroacoustic and numerical simulations within popular game engines. The use of popular game engines to seamlessly integrate different data types enables dynamic and interactive environments where users can explore scenarios in real time, enhancing both scientific analysis and stakeholder engagement to bridge the gap between complex geohazard science and effective stakeholder understanding, enabling informed decision making and risk management.
How to cite: Eisermann, J. O., Gross, F., Vollert, J., Wolf, J., Petersen, L., Kopp, H., Berndt, C., Reicherter, K., Krastel, S., Hübscher, C., Wagner-Ahlfs, C., Kwasnitschka, T., and Bernstetter, A.: Immersive Visualizations for Marine Geohazards in the Aegean Sea: Bridging Science and Stakeholder Engagement., EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5968, https://doi.org/10.5194/egusphere-egu25-5968, 2025.
The Korea Meteorological Administration has developed a tool that can visualize various numerical model data in three dimensions by combining them with spatial information to support forecasters' weather analysis work.
The 3D visualization tool is a precise analysis tool that can compare and analyze weather data according to time, space, and altitude.
An open-source 3D visualization engine (CesiumJS) was applied to display numerical model data on 3D spatial information.
Spatial information displays major spatial information (national and administrative boundaries, major roads, rivers, etc.) and a three-dimensional numerical elevation model.
The numerical model data was developed to express the precursor model (ECMWF) and the local model (LDAPS), but it was flexibly implemented to express other numerical model data.
It is expected that more three-dimensional and precise analysis will be possible by combining numerical model data with spatial information and analyzing it in three dimensions.
How to cite: Kim, S.: Development of 3D visualization tool for KMA (Korea Meteorological Administration) numerical model, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2209, https://doi.org/10.5194/egusphere-egu25-2209, 2025.
Cloud MICAPS Engine is a next-generation cloud and component-based application development framework for Meteorological Information Comprehensive Analysis and Processing System(MICAPS) in China, which has the following features:
I.Microservice Architecture
It utilizes a hybrid microservices management framework composed of K8s (Kubernetes) and Spring Cloud. The primary computational services are based on the underlying K8s + container cloud to implement microservices, while the upper-layer business applications integrate business-oriented microservices functions through Spring Cloud.
II.Real-time Visualization and Analysis
Developed using B/S technology, all weather-related business data is encapsulated as meteorological data layers through methods such as OGC, resumable transmission, and streaming services for loading by front-end business applications. It includes professional analysis components for common meteorological business operations such as points, lines, and surfaces, as well as interactive analysis of time curves, vertical soundings, and arbitrary sections.
III.Real-time Visualization Rendering
Based on real-time drawing technologies such as WebGL and WebGPU, it supports real-time product map services for products with high access volumes through pre-processing and pre-service methods, forming a highly consistent map service data environment with an integrated approach.
IV.Collaborative Editing and Forecast Service
Based on a 2D/3D map engine, the high-resolution real-time data visualization rendering technology is supported by CogTiff data storage and service technology. Real-time synchronization of interactive editing operations is achieved through WebSocket to realize real-time collaboration among multi-terminals. The back-end data editing algorithm realizes the consistency of updated data in FIFO.
V.LLM-driven Interaction and Processing
Large language models technology is used, with aggregating algorithms in the whole process of intelligent digital forecast service business, including observation perception, analysis diagnosis, interactive judgment, processing and generation, inspection and evaluation. "AI Agent" is the core to drive human-computer intelligent interaction and information recommendation.
Cloud MICAPS Engine will be released later in 2025.
How to cite: Xue, F.: Progress in Cloud MICAPS Engine, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1941, https://doi.org/10.5194/egusphere-egu25-1941, 2025.
Standardisation of geological maps visualisation is crucial for improving data legibility and comparison across different scales and regions. In Poland, overview geological maps ranging from scales of 1:2,500,000 to 1:500,000 have been traditionally prepared using distinct graphical styles, each tailored to the particular characteristics of the mapped geological units. These maps used to employ individual patterns and colour palettes to enhance the visibility and readability of geological features, prioritising the requirements of printed editions. However, differences in the number and types of geological units across various maps led to inconsistent visual representations, limiting the ease of comparison between them.
To address the above, the Polish Geological Institute’s team embraced the idea of creating a unified, semantically harmonised graphical style for overview geological maps. The main objective was to develop a common style for all stratigraphic units, which could be applied across various maps of Poland, particularly those prepared for online publication in the frame of the Polish Geological Cartograhy Platform. This experiment aimed to standardise the colour and pattern schemes, building upon the stratigraphic classification system provided by the International Commission on Stratigraphy (ICS), but with necessary extensions to accommodate mixed stratigraphy.
While this approach slightly reduced the visibility of details in certain areas, it significantly enhanced the comparability of geological data across maps. By adopting a consistent visual language, the maps delivered a clearer cartographic message, particularly when zooming in and out in map viewers. The harmonisation of the graphical style also enhanced data visualisation across various scales, making it easier for geologists to interpret and compare geological units.
This initiative was inspired by earlier efforts, such as the OneGeology initiative and the INSPIRE Directive, both of which sought to standardise the visualisation of lithological and stratigraphic data. However, these frameworks primarily focused on older geological units and their principal lithologies, which could lead to potential misinterpretations of the data. For example, geological units spanning from the Cambrian to the Devonian period were often represented using a single colour, which could obscure their true geological diversity. To address this, PGI team proposed the use of distinct colours for each geological period, drawing inspiration from the Commission for the Geological Map of the World (CGMW) colour codes.
The results of this experiment demonstrate that a semantically harmonised approach to geological map visualisation not only enhances the clarity of individual maps but also makes data more comparable across different scales and regions. By providing a consistent and intuitive visual representation of geological units, this method helps to improve the overall understanding of geological data and facilitates its use in various scientific, educational, and practical contexts.
How to cite: Jóźwik, K., Urszula, S., and Joanna, P.: Zoom in - zoom out challenge: Semantically and visually coherent overview geological maps of Poland, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10206, https://doi.org/10.5194/egusphere-egu25-10206, 2025.
Although digitized, geological maps have poorly evolved since their inception and still retain the constraints imposed by printing: heterogeneity between map sheets, limited point data, and challenging update procedure that remains essentially nonexistent.
The heterogeneity of geological maps also complicates their integration to Information System. Maps, based on rules and concepts that evolved over time, are no longer suited to the precision required by database structures. As concepts and models evolve in geology together with the scientific knowledge, the choice of the representation on the map also depends on the authors’ interpretations and the period of production. As a result, the mapped geological units reflect arbitrarily chosen characteristics of the rocks concerning either their protolith nature, their metamorphic characteristics or alteration transformations, considered as the most representative at the date of publication.
In France, 1:50,000 scale geological maps provide a full territory coverage but do not escape these heterogeneity issues. To address this problem, the BRGM (French Geological Survey) has implemented the RGF program (Référentiel Géologique de la France), which aims to develop a methodology to overcome these limitations and propose an innovative approach to represent the geological knowledge. In this research program new data are acquired by PhD students allowing to improve geological knowledge and to produce updated and harmonized information on geological maps and boreholes.
The tools developed and implemented over the past decade, based on the establishment of an Information System structured as a knowledge database called the Geological Reference System, now make it possible to offer a standardized representation of geological knowledge derived from traditional geological maps. This knowledge can be represented through different reference systems: lithostratigraphic unit, event (and thus the geological history), or domain and zone (e.g. lithotectonic units).
Here, we present some of the results obtained from works conducted in the Alps, the Pyrenees, and the Montagne Noire, illustrating how this structuring of geological knowledge into reference system enables the creation of new maps tailored to the geological information one wishes to represent and, consequently, to the scientific and societal challenges at hand.
How to cite: Padel, M., Le Bayon, B., Bernachot, I., Cagnard, F., Plunder, A., Issautier, B., Baudin, T., Tissoux, H., Lacquement, F., Grataloup, S., and Ricodel, C.: The geological reference system: from the concept of an integrated geological knowledge management to cartographic representation., EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11034, https://doi.org/10.5194/egusphere-egu25-11034, 2025.
The RGF research program (French Geological Reference platform) aims to establish a continuous and coherent geological knowledge base covering the entire French territory. To achieve this, the program relies on the development and implementation of comprehensive geological reference systems designed to structure, harmonize, and manage geological data, including:
- The lithostratigraphic reference system, which provides a hierarchical classification of geological units (super-group, group, sub-group, formation, member),
- The geological events reference system, which organizes and ranks geological events to document and reconstruct the geological history of geological units,
- The lithotectonic reference system, encompassing structural zones, hierarchically organized tectono-stratigraphic units, and paleogeographic domains,
- The geological boundaries reference system, which catalogs structural and geologic contact information.
These reference systems are being conceived and structured in alignment with international standards such as GeoSciML and INSPIRE. Built as PostgreSQL databases, they are under development through close collaboration between geologists, computer scientists, and GIS experts to address scientific requirements and serve as a repository of geological knowledge.
At the same time, tools and applications are being developed to utilize these reference systems to constrain the attribution of geological elements across various datasets (e.g. boreholes, geological maps, 3D models), thus facilitating the harmonization, enrichment, and updating of legacy data. This includes, for example, the modernization of historical French geological maps at a 1:50,000 scale. The process involves linking map geometries to the reference systems through GIS tools and custom QGIS plugins developed by BRGM. This approach supports the transition from static maps to dynamic, multi-scale digital representations, and enables the creation of maps tailored to various scientific objectives and practical applications.
This presentation gives an overview of the lithotectonic and geological event reference systems, the underlying database model and the QGIS plugins developed for their application. Initially developed within the RGF and now within the Digital Earth project of the PEPR research program, these tools are also applicable to other projects, such as the ongoing development of the European Lithotectonic Map by the GSEU. The presentation will also give an overview of the work carried out in the Pyrenees region, demonstrating the dissemination of updated maps of lithostratigraphic units and the ability to query geological events associated with specific formations, all accessible through a dedicated ArcGIS mapping viewer.
How to cite: Bernachot, I., Le Bayon, B., Padel, M., Cagnard, F., Plunder, A., and Dechambenoit, G.: Development of geological reference systems: integrating databases and tools to improve geological knowledge and data harmonization, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11428, https://doi.org/10.5194/egusphere-egu25-11428, 2025.
Field observations and measurements are essential for reconstructing the geological history of a region and for producing accurate maps and models. For national geological surveys, these raw data are critical resources that need to be efficiently managed, stored and disseminated for effective reuse in research. At BRGM, GeoField1, an application developed within the framework of the Référentiel Géologique de la France (RGF)2 and subsequently extended to support other geological projects, allows field data to be managed and capitalised according to the geological reference system, ensuring compatibility with internal and international standards. This system facilitates the compilation, sharing and reuse of data in accordance with the FAIR principles (Findable, Accessible, Interoperable, Reusable).
Recently, a data acquisition process was implemented to streamline field data collection and enable direct integration into GeoField. This workflow relies on QField3, an open-source mobile application built on the QGIS engine, leveraging key features such as seamless integration with QGIS projects, customizable data forms, the ability to use proper vocabularies, and offline mapping capabilities. A dedicated master project was designed to meet the specific needs of field geologists, enabling them to capture essential information through tailored forms, complete with dropdown lists to ensure consistency in terminology. Adapted symbology allows for real-time visualization of structural measurements on the map. The master project is made available to users through a QGIS plugin, which loads the project template, including a predefined database structure and up-to-date BRGM lexicons. Users can then customize their project by adding relevant layers, such as map backgrounds and other vector or raster data (e.g., DEM, geochemical analysis, geophysical data), and load the project onto their mobile devices for field acquisition.
Upon returning to the office, the plugin facilitates the automatic transfer of field data into GeoField, ensuring seamless integration into the BRGM central database. This workflow provides a robust, efficient, and standardized approach to geological data collection, capitalizing on the synergies between QGIS/QField and GeoField to enhance data management, sharing, and reuse within the geoscientific community.
1GeoField page: https://rgf.brgm.fr/page/geofield
2RGF page: https://rgf.brgm.fr/
3QField - Efficient field work built for QGIS, url: https://qfield.org/
How to cite: Stephan-Perrey, J., Bernachot, I., Plunder, A., Padel, M., Le Bayon, B., and Bezard, M.: Streamlining geological field data collection, management, and integration with QField, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17590, https://doi.org/10.5194/egusphere-egu25-17590, 2025.
The island of Pantelleria is an active volcanic complex, rising 800 m from sea level, situated on a continental rift within the NW sector of the Sicily Channel (Italy).
The island is characterized by a bimodal magmatism, mafic and felsic, with this latter -by far more abundant- including metaluminous trachyte and pantellerite (peralkaline rhyolites) magmas, erupted either as lava flows, pumice falls or pyroclastic currents.
This scenario, in addition to discontinuous field exposures and closely spaced (in space and time) explosive events produced by a multitude of eruptive centers producing compositionally similar deposits, makes challenging the detailed stratigraphic reconstruction of the volcanological evolution of the island.
In the last 40 years many studies have focused on specific volcanological, geochronological geochemical and petrological aspects of the island, unravelling peculiar eruptive dynamics and petrographic aspects of peralkaline magmatism. These efforts produced great steps forward in the knowledge of the volcanological evolution of the island, strictly tied to the peculiarities of peralkaline magmas. Nevertheless, a detailed geological map comprehensive the entire volcanological evolution (pre- and post -Green Tuff ignimbrite eruption) is still missing.
In the mid 70’s and 80’s unofficial schematic geological maps have been realized, mostly focused on lithological aspects of the erupted products and their areal distribution, summarized in the seminal paper of Mahood & Hildreth (1986) the first to define a comprehensive of pre- and post-Green Tuff stratigraphy. These studies generated discordant stratigraphic subdivisions and a not univocal stratigraphic nomenclature, until the work of Jordan et al. (2018) who defined an accurate stratigraphy and nomenclature of the pre-Green Tuff ignimbrite eruptions, adjuvated by progresses made by slightly earlier geochronological and paleomagnetic studies. The stratigraphy of post-Green Tuff volcanism, mildly explosive to effusive, though much improved by recent Ar/Ar, field and petrographic studies, still has some doubtful points.
To summarize, harmonize and integrate all the existing data into a single detailed product, the Geological Survey of Italy (in collaboration with Palermo University) performed, between 2022 and 2024, a field survey at scale 1:10.000 in the framework of the Italian Geological Mapping Project 1:50.000 scale (CARG project).
Here, we present the preliminary geological field map of Pantelleria Island (according to CARG Project guidelines) corroborated by existent and new geochronological, geochemical and petrographic analyses on hundreds of samples collected from the different volcanic structures of the island.
Such detailed field mapping of the entire island, together with the geological survey of the offshore area (still in progress), updating the findings of previous studies, allows us to obtain the first official geological map of such an active volcanic island. This represents the first fundamental step for the development of future studies of volcanic hazard assessment.
References:
Mahood, G.A., Hildreth, W. Geology of the peralkaline volcano at Pantelleria, Strait of Sicily. Bull Volcanol 48, 143–172 (1986).
Jordan, N. J., Rotolo, S. G., Williams R., Speranza, F., McIntosh, W. C., Branney, M. J., Scaillet S. Explosive eruptive history of Pantelleria, Italy: Repeated caldera collapse and ignimbrite emplacement at a peralkaline volcano. JVGR, 349, 47-73, 2018
How to cite: Pensa, A., Bonomo, R., Cinquegrani, A., Ricci, V., Rotolo, S. G., Urbani, S., and Vita, L.: Unravelling the complex volcanological evolution of Pantelleria Island (Channel of Sicily, Italy): new insights from the Italian Geological Mapping Project (CARG Project), EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10643, https://doi.org/10.5194/egusphere-egu25-10643, 2025.
This study introduces a novel methodology to enhance the accuracy and efficiency of 87Sr/86Sr isoscape mapping by integrating deep learning (DL) techniques with geostatistical methods. Utilizing kriging-based data augmentation, the proposed framework addresses data scarcity by generating high-quality synthetic training data while incorporating spatial geological factors and geochemical elements as input variables to improve model performance. The study developed an isotopic basemap for South Korea using 409 soil samples and a feedforward deep neural network (FDNN) model. The FDNN model demonstrated superior accuracy (91.67%) compared to traditional kriging (76.8%) and convolutional neural network (CNN)-based models (86.14%). The robustness of the FDNN model was significantly enhanced by kriging-based data augmentation, which not only captured geological anisotropies but also incorporated uncertainty analysis to improve reliability. The resulting 87Sr/86Sr isoscape map revealed distinct isotopic distributions across South Korea, with higher ratios associated with metamorphic and granitic rocks, reflecting geological history and topographical influences. Notably, the predicted isotopic distributions closely aligned with the boundaries of tectonic provinces, underscoring the geospatial accuracy of the developed model. Validation using bone samples additionally confirmed the efficacy of the proposed method in accurately estimating isotopic levels. These findings highlight the potential of combining geostatistical and DL approaches to overcome traditional challenges in isotopic mapping, offering scalable solutions for applications in environmental monitoring, archaeology, and provenance studies.
How to cite: Lee, H. and Jeong, J.: Enhancing Deep Learning-based Strontium Isotopic Landscape Estimation Using Geostatistical Method: A Case Study in South Korea, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5423, https://doi.org/10.5194/egusphere-egu25-5423, 2025.
The application of complementary geochemical analysis alongside deep learning techniques serves as a powerful tool in identifying geographical origin of environmental samples on a national scale. This research presents methodologies aimed at enhancing the accuracy of origin determination for soil samples across South Korea, leveraging geochemical and geological data. It addresses challenges associated with integrating Sr isotopes and multivariate geochemical variables and preserving geological interpretability by incorporating Autoencoder deep learning algorithms,which facilitate efficient feature engineering for comprehensive data analysis. Through the analysis of 412 soil samples collected nationwide, a geographic origin distribution and classification model was developed, establishing a novel framework for environmental sample analysis. The analysis identified six origins within South Korea, each distinguished by its geological tectonic units, bedrock age, and bedrock type. Extensively wide areas with granite bedrock nationwide were mostly classified into the same origin, irrespective of their geological tectonic configurations. The findings highlight the efficacy of integrating isotopic with geochemical data through advanced analytical techniques, significantly improving origin tracing accuracy and efficiency. Such advancements have significant implications for disciplines including agriculture, forensics, and archaeology, showcasing the potential of these methodological innovations.
How to cite: Lee, S. and Jeong, J.: Utilizing deep learning-based feature engineering for effective geographical origin subdivision and classification of environmental soil samples in South Korea, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5488, https://doi.org/10.5194/egusphere-egu25-5488, 2025.
The segmentation of ice in X-ray Computed Tomography (CT) scans of permafrost samples has traditionally relied on the Hounsfield Unit (HU) thresholding approach while their accuracy is often limited by overlapping density ranges in complex and heterogeneous samples. Recent advances, including automated thresholding algorithms and machine learning techniques, offer improved precision by leveraging texture, contrast, and morphological features in CT images. This study investigates the evolution of ice segmentation methodologies by applying multiple approaches to a 164 cm long permafrost core drilled from a Yedoma upland in north-eastern Siberia. The core was analyzed using traditional HU thresholding, automated thresholding methods (e.g., Otsu and adaptive histogram-based segmentation), and machine learning models (e.g., random forests and convolutional neural networks). The results from CT scans and segmentation methods were validated and compared against laboratory measurements of ice content and density, ensuring a robust evaluation of each technique's accuracy and reliability.
The results provide critical insights into the strengths, weaknesses, and suitability of different segmentation methods for permafrost cores. These findings contribute to the development of standardized, high-precision methodologies for non-destructive characterization of ice-rich soils, supporting geotechnical and climate change studies in permafrost regions.
How to cite: Roustaei, M., Nitzbon, J., Harvey, J., Francis, E., Schlueter, S., Boike, J., and Froese, D.: Improving Ice Segmentation in Permafrost Cores using Computed Tomography, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20681, https://doi.org/10.5194/egusphere-egu25-20681, 2025.
