The global energy transition is driving unprecedented demand for critical and strategic raw materials such as copper, lithium, nickel, cobalt, graphite and rare earth elements. In Europe, the Critical Raw Materials Act (CRMA) sets ambitious targets for domestic extraction and processing, and requires Member States to establish national exploration programmes, including geoscientific surveys, mineral mapping and geochemical campaigns. These programmes are generating a new wave of regional and national-scale geophysical datasets and renewing interest in the reinterpretation of legacy data. With most near-surface deposits in well-explored terranes already discovered, future discoveries will increasingly depend on our ability to image deeper, covered and geologically complex targets.
This session invites contributions on all aspects of geophysics applied to mineral exploration, from national to deposit scale. We welcome studies using potential field (gravity, magnetics), electromagnetic, magnetotelluric, induced polarisation, seismic, radiometric and borehole methods, as well as airborne, drone-based and passive-source approaches.
Contributions presenting results, strategies and lessons learned from national exploration programmes, as well as work on low-impact exploration, social acceptance, and secondary resources such as mine waste and tailings, are also welcome. The session aims to bring together academia, geological surveys and industry to discuss how geophysics can support a secure and sustainable supply of raw materials in Europe and worldwide.

In recent years, technologies based on Artificial Intelligence (AI), such as image processing, smart sensors, and intelligent inversion, have garnered significant attention from researchers in the geosciences community. These technologies offer the promise of transitioning geosciences from qualitative to quantitative analysis, unlocking new insights and capabilities previously thought unattainable.
One of the key reasons for the growing popularity of AI in geosciences is its unparalleled ability to efficiently analyze vast datasets within remarkably short timeframes. This capability empowers scientists and researchers to tackle some of the most intricate and challenging issues in fields like Geophysics, Seismology, Hydrology, Planetary Science, Remote Sensing, and Disaster Risk Reduction.
As we stand on the cusp of a new era in geosciences, the integration of artificial intelligence promises to deliver more accurate estimations, efficient predictions, and innovative solutions. By leveraging algorithms and machine learning, AI empowers geoscientists to uncover intricate patterns and relationships within complex data sources, ultimately advancing our understanding of the Earth's dynamic systems. In essence, artificial intelligence has become an indispensable tool in the pursuit of quantitative precision and deeper insights in the fascinating world of geosciences.
For this reason, aim of this session is to explore new advances and approaches of AI in Geosciences.

Recent advances in geophysical observations, numerical simulations, remote sensing products, and high-performance computing are generating large and increasingly complex multidimensional datasets. Interpreting these datasets requires advanced processing and modelling together with effective methods for visualisation and integration.
This session focuses on the development and application of advanced visualisation methods for geophysical data. We particularly welcome approaches that improve the exploration and interpretation of multidimensional, multiscale, and time-dependent datasets, and that can improve geophysical interpretation and its practical application.
We welcome contributions addressing interactive visualisation, 3-D and 4-D representations, data fusion, and the integration of observational, experimental, and modelled data. Contributions combining geophysical data with artificial intelligence and machine learning, including AI-assisted visual analytics and interactive approaches for navigating large and heterogeneous datasets, are particularly encouraged.
The session will also consider practical applications of advanced visualisation across different operational and scientific scenarios. Examples include geophysical applications in engineering and infrastructure monitoring, environmental protection and assessment, natural hazard monitoring and early-warning systems, and the management and interpretation of large and heterogeneous datasets. We also welcome applications to planetary exploration, where observations acquired by spacecraft, landers, rovers, and orbital platforms need to be integrated with geological, geophysical, topographic, and numerical datasets to investigate planetary environments.
By bringing together developments across geophysics, engineering, environmental sciences, natural-hazard research, data science, and planetary exploration, the session will provide a forum to discuss how advanced visualisation can improve the interpretation of complex geophysical datasets and support more effective, transparent, and interactive approaches to understanding the Earth and planetary geophysics.

AI is rapidly transforming research in deep Earth geodynamics, shaping research strategy, driving new research pathways, and promoting new research directions. This session invites AI-driven contributions and Machine Learning applications from geodynamics and related disciplines focusing on the structure and evolution of the Earth's crust and upper mantle.

Machine learning is reshaping the representation of complex physical processes in Earth system models, offering new avenues for parameterisation, emulation, and hybrid modelling. This session focuses on the use of machine learning to emulate computationally expensive or unresolved processes, accelerate physical simulations, enable data-driven discoveries, and improve representation across domains such as convection, turbulence, radiation, hydrology, sea ice, and other Earth system components. Topics include (but are not limited to):

- Subgrid-scale parameterisations via machine learning
- Emulators of physical processes, model components, or whole weather and climate models (including end-to-end learning and foundation models)
- Data-driven discoveries
- Hybrid ML-physics modelling frameworks
- Physics-informed neural networks, neural operators, and differentiable programming
- Reinforcement learning and other approaches for ensuring physical consistency, stability, and optimising model behaviour
- Calibration and parameter optimisation using ML
- Physical behaviour, encoding and analysis of ML models
- Verification and explainability (XAI) of data-driven models (including AI forecasting)
- Representation of extremes and downscaling
- Coupling of ML models with physical models
- Cross-domain applications (atmosphere, ocean, cryosphere, land)
- Impacts of architecture choices and design on physical behaviour
- Links between ML methods and fundamental physics or applied mathematics
- Novel AI methods and applications

Time series are a common type of data generated by observational and modelling efforts across Earth, environmental and space sciences. Long-term observations are particularly important for understanding gradual changes and assessing risks, yet are often difficult to sustain and fund. Their characteristics can vary substantially, from short to long records, linear to nonlinear dynamics, univariate to multivariate data, and single- to multi-scale variability. These differences call for both tailored methodologies and general approaches.

A key challenge is distinguishing random fluctuations from long-term changes in order to better understand processes within and across Earth system components. This requires knowledge of temporal variability and, often, sufficiently long observations. For example, reliable sea-level trends may require several decades of continuous measurements because of decadal variability. Likewise, the stochastic variability of geophysical time series can exhibit power-law scaling, requiring long records for robust statistical assessment.

Time series analysis encompasses a broad range of tasks, including:
- characterizing nonlinear variability in the time and/or frequency domain;
- quantifying complexity, predictability and scaling properties;
- identifying statistical interdependencies within and between time series;
- distinguishing co-variability from causal relationships;
- reducing dimensionality and identifying meaningful modes of variability; and
- developing stochastic and deterministic statistical or dynamical models.

This session invites contributions on the development and application of modern methods for analysing observational and model time series across the EGU community, including geophysical, geodynamic, oceanographic, geodetic and climate observations from terrestrial observatories and remote sensing. Contributions addressing advances in sensors, instrumentation, monitoring, analysis and interpretation, as well as comparisons of different approaches, are welcome. Studies using novel methods, including AI, for the analysis of long time series are particularly encouraged. We aim to foster interdisciplinary exchange and cross-fertilization between different EGU divisions.

Urban sustainability is shaped by complex and dynamic interactions between environmental processes and human activities. These coupled human–environment interactions influence urban climate, land-use change, ecosystem services, carbon emissions, resource consumption, mobility, environmental exposure, inequality, and resilience. Recent advances in Earth observation, geospatial big data, in-situ and crowdsourced sensing, cloud computing, machine learning, and artificial intelligence provide unprecedented opportunities to observe, model and predict these interactions across spatial and temporal scales.
This session invites contributions that use new observations, integrated datasets, data-driven analytics, and AI-based approaches to advance the understanding of urban human–environment systems. We particularly welcome studies addressing multi-source urban data integration, machine learning and GeoAI, urban climate and heat, land-use change, urban ecosystems, carbon emissions, mobility, environmental exposure, inequality, vulnerability, resilience, open datasets, benchmarks, uncertainty assessment, explainable AI, and reproducible workflows.
The session aims to foster interdisciplinary exchange among remote sensing, geospatial science, urban studies, environmental science, climate science, and data science communities. We encourage contributions from local to global scales, especially those demonstrating how new data and analytics can reveal interactions between people and urban environments and support more sustainable and resilient urban futures.

Multi-modal machine learning models for atmospheric and Earth system dynamics that are trained on many different datasets and predict a range of outputs, such as the WeatherGenerator, ESFM, or WeatherNext, are emerging as the next frontier in machine learning-based Earth system modeling. These models pose unique challenges, such as how to combine many input datasets at different resolutions in space and time, how different datasets should be scheduled at different training stages, or what data should be used as target during pre-training and what can be decoded afterwards. They also open research questions such as how model parameter count should depend on the number and resolution of input datasets or how to quantify the expected benefit of adding an additional dataset. This session collects contributions both on the development of multi-modal machine learning models as well as their use in applications, from short- and medium-range forecasting to scientific investigations that probe the model itself. Submissions can address either methodological questions, evaluations (e.g. conservation of physics), technical aspects (training on HPC systems) or focus on applications.

This session concentrates on extreme rainfall events, surface water dynamics, and flood events, exploring innovative remote sensing, AI, and digital twin technologies for real-time monitoring, risk assessment, and mitigation. It invites submissions on advanced data integration, modeling approaches, early warning systems, and decision-support tools to improve understanding, forecasting, and management of flooding and related surface water hazards.

The integration of AI with digital twin improves the analytical and operational capabilities of geospatial systems, which through the analysis of historical data and the integration of real-time information (IoT) are able to highlight even “hidden patterns” in the data, identifying new models capable of improving forecasts with greater control over the quantification of uncertainty and the variability of the phenomenon analysed.

This session aims to focus on flood hazard and risk assessment, monitoring, and management. This Topic invites the submission of articles focused on, but not limited to, the following areas:
• Monitoring of extreme rainfall events and flood hazards for risk assessment and communication.
• Digital twins (DTs)/prototypes of DTs in flood hazard forecasting, early warning, monitoring, and supporting tools for urban governance.
• DSSs to extract meaningful information in the artificial intelligence era, eventually serving to reduce risk and provide support tools to mitigate flood hazards.
• The role of AI and digital twins to assess the economic impacts of flood hazards and the cost-effectiveness of various mitigation strategies.
• Novel techniques to analyse big data coming from Earth observation platforms, drones, and other geospatial data in order to provide timely information related to the extend, exposure, and impacts of flood hazards.

Scientific machine learning (SciML) and artificial intelligence (AI) are increasingly used to interpret heterogeneous subsurface data, accelerate multiphysics simulation, infer uncertain parameters, and support decisions in geotechnical, environmental, and geoenergy applications. Their practical value, however, depends on physical consistency, generalisation beyond training conditions, uncertainty characterisation, and validation against independent evidence.
This session invites contributions on recent methodological advances and innovative applications of data-driven, physics-informed, and hybrid physics-AI techniques for geotechnical, geothermal, and subsurface systems. Relevant applications include, but not limited to, site and subsurface characterisation; geothermal exploration, reservoir development, monitoring, and operation; groundwater flow, transport in porous media, multicomponent reactive transport, and geochemical processes; coupled thermo-hydro-mechanical-chemical processes; underground storage of heat, hydrogen, or carbon dioxide; and geotechnical infrastructure.
We welcome work on supervised and unsupervised learning, deep learning, physics-informed neural networks, neural operators, surrogate and reduced-order models, inverse modelling, data assimilation, Bayesian inference, digital twins, and other relevant studies. Contributions addressing heterogeneous or sparse data, multiscale modelling, uncertainty quantification, sensitivity analysis, model discrepancy, transferability, interpretability, and computational efficiency are particularly encouraged.
A central objective is to connect methodological development with physical and observational evidence. We therefore welcome studies integrating laboratory experiments, imaging and sensing methods, field monitoring, operational data, and numerical simulations for training, calibration, verification, and validation. Comparative studies that establish when machine learning or physics-informed methods improve upon conventional numerical, statistical, or inverse approaches and where they do not are especially valuable.

Machine learning (ML) is increasingly advancing cryospheric science, offering new ways to extract information from the rapidly growing volume of satellite, airborne, and in situ observations and to accelerate numerical modeling. Deep learning has proven particularly effective for computer vision tasks such as calving front detection, glacier outline delineation, mapping of permafrost landforms, sea ice classification, and snow cover mapping. Beyond image analysis, ML-based emulators can reproduce the behavior of computationally expensive physical models. This enables large ensembles and the inversion of poorly constrained parameters. Hybrid and physics-informed approaches further promise to combine the flexibility of data-driven methods with physical consistency. More recently, foundation models pretrained on large Earth observation archives have raised expectations of strong transferability across regions and sensors, particularly for applications where labeled training data are scarce. However, their benefit for cryospheric applications has yet to be demonstrated, as they have not consistently outperformed task-specific architectures trained from scratch. Systematic evaluation on shared benchmark datasets is therefore essential to measure real progress. We invite contributions on all aspects of machine learning for the cryosphere, including but not limited to computer vision and remote sensing, model emulation and hybrid modeling, benchmark datasets, as well as trust and interpretability. We also welcome critical perspectives, such as barriers to adoption, computational costs, and cases where ML approaches did not outperform physics-based methods.

Long spin-up times remain a major computational bottleneck in ocean and climate modelling, limiting our ability to investigate past climate states, understand model biases, and quantify parameter uncertainty. Developing more efficient spin-up methods can help overcome these limitations and support the use of past climate information to strengthen confidence in future climate projections.

This session will bring together results from the Past-to-Future Global Ocean Circulation Model Spin-Up Competition. The competition challenges participants to bring a global ocean model to equilibrium using as few computational resources as possible. Participants access the model as a time-forward black box, advancing the ocean state through a prescribed routine without modifying the underlying model or its physical configuration. This common framework enables a systematic comparison of alternative approaches.

The session will focus on three core aspects:

• presenting the methods developed by participating teams and comparing their performance against the common benchmark;
• discussing computational efficiency, convergence, and reproducibility, including the costs associated with training data where machine learning methods are used;
• exploring lessons learned and the potential for applying successful approaches more broadly in ocean and climate modelling.

We welcome competition participants and researchers interested in ocean and climate modelling, numerical analysis, scientific computing, and machine learning. The goal is to identify promising approaches, discuss remaining challenges, and encourage future collaborations on efficient model initialisation. Registration for the competition closes on 1 November 2026, with final results due on 1 March 2027.

The Atlantic Meridional Overturning Circulation (AMOC) plays a crucial role in shaping the dynamics of the Earth’s climate by distributing heat and nutrients across the Atlantic. It is important to understand the past, present and future changes in the dynamics of the AMOC, either gradual or abrupt, since such changes, and the possibility of its tipping, can have profound climatic and societal impacts. In this regard, big data and AI play an increasingly important role in studying AMOC dynamics based on diverse types of data. These range from geological proxies over contemporary in-situ and remote sensing observations to simulations of state-of-the-art ocean or coupled Earth system models and provide an ever increasing amount of more and more complex data on the AMOC. Advanced numerical methods and AI can help us to uncover critical aspects of the AMOC dynamics by extracting new patterns and highlighting the role of complex physical mechanisms and feedbacks, including early warnings of future regime shifts of the AMOC or some of its subcomponents like the Nordic Seas deep convection or the Northern hemisphere subpolar gyre.
In this session we welcome contributions exploring new ways of using big data and AI to elucidate AMOC dynamics. We aim to cover a broad variety of computational methods, making use of the wealth of AMOC-related observational and/or model data. These can range from statistical methods exploiting big datasets to machine learning and deep learning approaches, including neural-network emulators of the AMOC. The session is open to work on a wide range of timescales, from paleoclimate reconstruction, through current observations to future projections. Contributions may address the analysis of short-term AMOC dynamics, as well as longer-term behaviour, including tipping of the circulation and associated forecast and impact studies.

Geospatial foundation models (GeoFMs) are enabling increasingly large-scale applications across Earth observation and the geosciences. Leveraging advances in representation learning, GeoAI, and the growing volume of satellite observations available for pre-training, recent models transfer knowledge across regions, sensors, spatial and temporal scales, and downstream tasks. Since the previous EGU session on remote sensing geospatial foundation models (ESSI1.11), research in model development and applications has expanded, with models increasingly evaluated across diverse regions and Earth science disciplines.

This session focuses on large-scale, satellite-based foundation models built from image-based Earth observation modalities, including multispectral, optical, synthetic aperture radar (SAR), hyperspectral, and related remote sensing data. We distinguish this scope from foundation models primarily developed for climate and weather data, numerical weather prediction, or atmospheric modelling. Our focus is on models that learn generalizable representations from large-scale Earth observation imagery and their applications to mapping, monitoring, and understanding Earth’s surface.

We aim to investigate challenges shaping the scientific and practical impact of these models, including transferability across regions and datasets, real-world downstream performance, scaling with data and computational resources, and the uncertainty, interpretability, and reliability of foundation-model predictions.

We invite contributions covering, but not limited to:

* New geospatial foundation models
* Novel pre-training and representation-learning strategies
* Large-scale datasets and benchmarks
* Multimodal and multisensor learning
* Model evaluation and transferability
* Geospatial embeddings as data
* Applications of GeoFMs to Earth observation and Earth science

Relevant applications include:

* Natural and anthropogenic disaster mapping and monitoring
* Crop and agricultural monitoring
* Natural ecosystem monitoring
* Forest and biodiversity monitoring
* Built environment mapping and monitoring
* Environmental monitoring and Earth system science

The session brings together researchers developing GeoFMs and applying them to scientific and societal challenges, with the aim of identifying emerging opportunities, methodological gaps, and future directions for large-scale Earth observation foundation models.

Earth operates as a complex, self-regulating cybernetic system driven by both internal processes and external astronomical forcing. This session explores the intricate feedback loops between the Solar System’s gravitational dynamics and Earth’s interconnected geospheres—including the atmosphere, hydrosphere, lithosphere, biosphere, pedosphere, troposphere, magnetosphere, and ionosphere. We focus specifically on how these systemic interactions govern sub-system variations, with a core emphasis on water balance components and hydrological cycles.
To decipher these non-linear, multi-scale relationships, traditional deterministic models must be augmented with data-driven methodologies. We invite contributions that advance empirical research tasks, conceptual frameworks, and system-level models. We particularly welcome studies utilizing artificial intelligence (AI), statistical learning, machine learning, and advanced data-mining techniques to analyze observational and proxy datasets. By bridging planetary dynamics, cybernetic geoscience, and modern data science, this session aims to foster a holistic understanding of Earth’s systemic balance and improve our predictive capabilities in a changing cosmic and terrestrial environment.

Assessing the time-dependent probability of earthquakes and anticipating their cascading effects—including earthquake-generated tsunami waves and volcanic unrest—remain major challenges in Earth sciences. Recent advances in ground-based and spaceborne observations, laboratory experiments, numerical modelling, statistical analysis, computational methods and data-driven approaches offer unprecedented opportunities to investigate preparatory processes and evolving hazards by integrating heterogeneous observations across the different domains of the Earth system.
The session aims to foster interdisciplinary discussion on earthquake-related processes and to assess critically how multiparametric observations, physical knowledge and advanced data-analysis methods can improve our understanding, forecasting and time-dependent assessment of seismic and volcanic hazards and their impacts.
Particularly welcome are studies addressing:
- Ground-based and satellite observations of electric, magnetic, electromagnetic anomalies potentially associated with impending seismic and volcanic activity;
- Multi-parametric seismological, geodetic, geological, geochemical, hydrological, atmospheric and ionospheric observations;
- Lithosphere Atmosphere Ionosphere Coupling (LAIC) and related measurements;
- Transient deformation, fluid-related processes, seismicity patterns, earthquake-generated tsunami waves and other coupled phenomena;
- Possible relationships between seismic processes, Earth tides, lunar forcing, solar–Earth interactions and Space Weather;
- Theoretical models, numerical simulations and long-term data-analyses;
- Artificial intelligence, machine learning and deep-learning methods for anomaly detection, data integration, forecasting and uncertainty assessment.
Contributions presenting results from new instruments, innovative observational infrastructures and current or forthcoming satellite missions investigating atmospheric, ionospheric and electromagnetic disturbances potentially related to seismic and volcanic activity are especially encouraged.
Further contributions may focus on real-time monitoring, multimission and multiplatform data integration, open-data services and interoperable platforms supporting the development of integrated Earthquake Observing Systems.

Hydrologic data have traditionally been sparse in space and time, requiring process-based models to interpolate observations and predict fluxes under limited data availability. Rapid advances in observation technologies are now enabling near–real-time monitoring of complex hydrologic processes across a wide range of spatial and temporal scales, fundamentally transforming hydrology from data scarcity to data abundance. Fully exploiting these emerging observation systems requires parallel advances in digital tools for data processing, integration, modeling, and visualization. New digital solutions—including digital twins, real-time data platforms, and interactive visualization environments—play a critical role in translating novel observations and scientific insights into actionable information for water resources research, multi-sector management, and decision-making. Under accelerating environmental change, two-way data exchange between models, observations, and stakeholders is increasingly essential for responsive forecasting and intervention. This session invites contributions on next-generation hydrologic observation systems and multi-scale digital solutions that advance process understanding, modeling, and application. Topics of interest include, but are not limited to:
- Novel observation techniques enabling improved parametrization of hydrologic processes
- Integration and fusion of next-generation hydrologic observations into modeling frameworks
- Digital twins and other advanced digital representations of hydrologic systems
- Real-time forecasting, early-warning, and decision-support systems for water resources and hydrologic hazards
- Case studies demonstrating the operational use and societal impact of hydrologic digititalization
Hosted by the Digital Water Flagship (https://digitalwaters.fi)

Session Description

The Anthropocene is characterized by unprecedented human pressures on aquatic environments, with rapid urbanization, industrialization, intensive agriculture, mining, and climate change increasingly affecting the quality and availability of water resources. Emerging and persistent contaminants—including pharmaceuticals, pesticides, heavy metals, nutrients, microplastics, pathogens, and other chemical pollutants—pose complex challenges to freshwater and coastal ecosystems and may create significant risks to human and environmental health. Understanding the sources, transport, transformation, fate, and impacts of these contaminants is therefore essential for safeguarding water security and ecosystem integrity.

This session, “Safeguarding Water Quality in the Anthropocene: Contaminants, Health and Sustainable Management,” will bring together interdisciplinary research addressing contemporary and emerging challenges in water quality. Contributions are invited on contaminant monitoring and assessment, advanced analytical and sensing technologies, pollutant fate and transport, exposure pathways, ecological and human health impacts, risk assessment, and innovative approaches for water and wastewater treatment. The session will also explore nature-based solutions, circular approaches, remediation strategies, and data-driven and integrated frameworks for sustainable water-quality management.

By connecting environmental science, hydrology, engineering, public health, ecology, and policy, the session aims to foster knowledge exchange and identify science-based pathways for protecting water resources and supporting resilient, healthy, and sustainable aquatic systems in the Anthropocene.

Effective analysis of modern geospatial big data requires more than the application of existing ML tools; it requires the development of new strategies that account for the characteristics of spatial and temporal data and processes. This session highlights advances in the methodological foundations of geospatial AI, focusing on the proposal and development of new methods, tools, and strategies to improve prediction, interpretation, and uncertainty assessment.
This session discusses challenges and advances in spatial and spatio-temporal machine learning methods and the software and infrastructures to support them. Application-based studies are also welcome, particularly those that demonstrate the practical utility of new tools or methodological approaches. By focusing on the development of methods, this session aims to bridge the gap between theoretical AI advances and practical geoscientific challenges.

Digital Twins of the Earth system are emerging as an approach for combining Earth observations, modelling, high-performance computing, data-driven methods and interactive simulation to better represent environmental processes and their interactions across spatial and temporal scales. Initiatives such as Destination Earth (DestinE), together with national, European, and international Digital Twin developments, provide new opportunities to explore how these capabilities can contribute to understanding and reducing disaster risk.

This session invites contributions examining how Earth-system and environmental Digital Twins, and the data, models and technologies underpinning them, can contribute to disaster risk reduction (DRR). The focus extends from improving the representation and understanding of hazards and their interactions to assess exposure, vulnerability and impacts, exploring risk scenarios, and supporting preparedness, early warning and risk-reduction decisions.

We welcome contributions addressing single, compound, cascading and multi-hazard risks, including floods, droughts, heatwaves, wildfires, storms, landslides and other geophysical, climate- and weather-related hazards. Relevant topics include the integration of Earth observations and environmental data; coupling of Earth-system, hazard and impact models; high-resolution and multi-scale simulations; representation of exposure and vulnerability; scenario and what-if analysis; impact-based forecasting and early warning; and the translation of complex simulations into risk-relevant information.

The session also welcomes research on the role of artificial intelligence and machine learning (AI/ML) in Digital Twin workflows, including data integration, emulation and surrogate modelling, pattern and anomaly detection, hazard and impact modelling, and the extraction of risk-relevant information from large simulations and datasets. Contributions addressing uncertainty, validation, interpretability, interoperability and computational challenges are particularly encouraged.

We invite both methodological developments and case studies that critically assess the added value and limitations of Digital Twins for DRR, including their potential use by civil protection authorities, public administrations, infrastructure operators and other risk-management communities. Contributions from Destination Earth and other European and international Digital Twin initiatives are welcome.

Recent advances in large language models have enabled agentic systems that autonomously plan, reason, and act on problems in Earth Observation (EO) and the geosciences. By interacting with data catalogues, models, processing services, and computational environments, these systems move beyond generating answers to constructing and executing multi-step scientific workflows. This shift raises fundamental questions about how such systems can be designed to be not only capable, but also grounded, verifiable, reproducible, and trustworthy.

This session invites contributions on the design, integration, evaluation, and responsible deployment of agentic AI across EO and the geosciences. We welcome work on how agents reason over long-horizon tasks, interact reliably with heterogeneous data and computational infrastructure, and maintain coherent state, context, and provenance across multiple steps. Topics include agent architectures and orchestration, multi-agent systems, memory and self-evolving mechanisms, as well as protocols and shared conventions that enable agents to discover, communicate with, and use scientific tools and services.

A central focus is the evaluation and verification of agentic scientific reasoning: not only final answers, but also the trajectories through which agents reach them, including tool selection, planning, code execution, intermediate results, and provenance. Contributions addressing uncertainty, hallucination, and failure modes are particularly encouraged, as are approaches that promote physical consistency through domain knowledge, foundation models, retrieval, world models, and physics-aware constraints. Work on human oversight, intervention, and verification is also relevant, particularly for autonomous systems deployed in operational or high-consequence settings.

More broadly, the session aims to advance agentic AI as a paradigm for computational Earth Observation and the geosciences. By bringing together AI researchers, geoscientists, and providers of scientific data and computational infrastructure, we seek to characterize the capabilities and limitations of agentic systems, develop principled methods for evaluating and improving them, and establish what is required for autonomous scientific workflows to operate reliably and reproducibly.

The rapid growth of missions, observatories, and monitoring systems across the heliosphere and Solar System, together with surface-based and airborne facilities, has created an unprecedented volume and diversity of data. Making sense of these observations requires methods that can efficiently process large datasets while extracting meaningful physical insight. Machine learning has become an important tool in this effort, complementing established physics-based approaches by offering new ways to identify patterns, develop predictive models, and work with complex or incomplete measurements.

In recent years, increasing attention has been given to hybrid approaches that combine machine learning with physical models. These methods are now being applied across planetary science and heliophysics, from forecasting solar eruptions and solar wind conditions to automating the analysis of planetary surfaces and improving on-board data processing. They demonstrate how data-driven methods can benefit from physical knowledge, while physics-based models can in turn be enhanced through modern data analysis techniques.

This session aims to provide an inclusive and interdisciplinary forum for researchers applying machine learning in planetary science and heliophysics, as well as for those developing methods at the interface between data-driven and physics-based approaches. We particularly encourage contributions that demonstrate the wide range of possible applications, foster exchange across disciplines, and showcase the transition of new methods from research to operations.

The increasing complexity and resource intensity of state-of-the-art Earth System Science (ESS), from Earth System Model simulations to large-scale data processing and analysis, require advanced approaches to orchestrate, automate, document and reproduce scientific processes. Scientific workflows provide a fundamental means to manage this complexity and to enable scalable and reproducible research across increasingly complex computational and data infrastructures.

Researchers in ESS are faced with heterogeneous computing architectures, federated research infrastructures, distributed and increasingly large datasets and restrictive data-movement policies. Workflows thus have to operate across systems while making efficient use of available computational and data resources. Portability, interoperability and provenance therefore become essential characteristics of scientific workflows rather than optional additions.

Advances in artificial intelligence (AI) introduce an additional dimension. AI can support the composition, optimization and execution of scientific workflows. Emerging agentic approaches may increasingly enable workflows to adapt autonomously to data, computational environments and scientific tasks. Such capabilities also raise new questions: How can scientific intent, assumptions, dependencies and execution environments be represented in AI generated workflows? How can AI-assisted and agentic workflows remain transparent, verifiable and reproducible? How can their decisions and adaptations be captured through appropriate provenance and metadata?

This session explores advances in scientific workflow concepts, technologies and applications for ESS, ranging from established workflow management approaches to emerging AI-assisted and agentic systems. We particularly welcome contributions addressing workflow orchestration and portability, scalable HPC and data-intensive workflows, distributed, federated and hybrid infrastructures, data locality and efficient resource utilisation, workflow FAIRnes, and AI-assisted generation, adaptation, optimization and execution of scientific workflows.

We encourage contributions presenting methods, tools, lessons learned and concrete use cases from ESS, including European and international research infrastructures and initiatives. Contributions demonstrating both mature solutions and emerging approaches that explore emerging capabilities of scientific workflow technologies are particularly welcome.

Spatio-temporal datasets are constantly growing in size, due to increases in extent and resolution. Because of
this, existing software to read, write, and store data, and translate the data may not be able to perform the
work in a timely manner anymore. This limits the potential of numerical simulation models and machine learning
models, for example.

In this session we bring together researchers working on novel software for processing large spatio-temporal
datasets. By presenting their work to their colleagues we aim to further strengthen the field of
high-performance computation in the geosciences.

We invite everybody recognizing the problem and working on ways to solve it to submit an abstract to this
session. Topics of interest for this sessions are, for example, related to:

- High-performance computing, parallel computing, distributed computing, cloud computing, asynchronous
computing, accelerated computing, green computing
- Algorithms, libraries, frameworks
- Parallel I/O, data models, data formats, data compression, data cubes, HDF5, NetCDF, Zarr, COG
- Containerization, Docker, Kubernetes, Singularity, Apptainer
- Physically based modelling, physics informed machine learning, surrogate modelling
- Model coupling, model workflow management
- Large scale hydrology, remote sensing, climate modelling
- Lessons learned from case-studies

We recommend authors to highlight the generic aspects of their work that may be especially of interest to
their colleagues.

Cloud computing now underlies most industrial and academic work, making it well-suited for processing big data close to its source. Combined with HPC, cloud infrastructure, enhanced by GPUs, containerization, and microservices, they increasingly supports AI/ML workloads once limited to HPC environments. Quantum computing is also emerging as a complementary tool, with hybrid quantum-classical methods showing promise for optimisation, simulation, and machine learning in Earth system science.
Session Focus
This session focuses on use cases involving the entire computing spectrum, from Cloud to HPC and to emerging Quantum computing, to o assess current progress and identify steps for wider adoption of these resources in Earth Observation and Earth Modeling. Contributions are welcomed across the following areas:
• Big Data Infrastructures and Platforms: Case studies, techniques, models, and algorithms for data processing on the cloud and HPC.
• Cloud Federations, Interoperability, and Sustainability: Scalability and interoperability across domains, multi-provenance data management, security, privacy, and green/sustainable computing practices.
• Cloud-Native and HPC AI/ML Frameworks: Tools and frameworks for processing data using AI and ML on the cloud and/or HPC, including operational deployment and data lake/warehouse management.
• Convergence of Cloud Computing and HPC: Workload unification and hybrid architectures for Earth Observation (EO) data processing across cloud and HPC environments.
• Quantum Computing Algorithms for Earth Science: Quantum and quantum-inspired algorithms for optimisation, simulation, and data analysis in EO and Earth system modeling.
• Hybrid Quantum-Classical and Quantum-HPC Architectures: Integration of quantum processing units (QPUs) with classical HPC and cloud infrastructures, including workflow orchestration and resource scheduling.
• Quantum Machine Learning: Quantum-enhanced ML techniques for remote sensing, climate, and geospatial data processing.
• Benchmarking, Access, and Readiness for Quantum Resources: Evaluating quantum hardware/simulators, cloud-based quantum access models, and assessing technology readiness for operational Earth science applications.

We encourage researchers, practitioners, and industry experts to share their insights, case studies, and innovative solutions that promote the integration of cloud computing, HPC, and quantum computing in Earth Observation and Earth Modeling.

Remote sensing products have great potential to support the monitoring and modelling of water resources. Nevertheless, their uptake by water managers remains limited because of concerns related to data quality, spatial and temporal resolution, trust, accessibility, and technical expertise.

In this session, we look for contributions on new developments that support the use of remote sensing data for water-management applications, from local to global scales. We welcome research aimed at improving the quality and usability of remote sensing products, such as higher spatial and/or temporal resolution mapping of land use and agricultural practices; improved assessments of river discharge, lake and reservoir volumes, and groundwater resources; drought monitoring and modelling and its impacts on water-stressed vegetation; and the monitoring and modelling of irrigation volumes.

We are also interested in studies assessing the quality of remote sensing products through uncertainty analysis or validation against independent data sources. Contributions that combine different techniques —such as physically based models and artificial intelligence—or integrate multiple data sources from different sensors, including remote sensing and in situ observations, are particularly welcome. Studies using different types of imagery, including satellite, airborne, and drone data, are also encouraged.

Finally, we invite presentations on the development of user-friendly platforms that follow the FAIR principles and provide smooth access to remote sensing data for water-management applications. We are particularly interested in applications of remote sensing to characterise human–water interactions and assess the impacts of climate change across the entire water cycle, including the connections between inland and sea (coastal) systems.

Remote sensing is a critical tool for data collection and observation of floods and surface water dynamics, especially in regions where field surveys and gauging stations are limited, such as remote or conflict ridden areas and data-poor developing nations. The integration of dynamic variables derived from remote sensing (e.g. water extent, depth and width, river slope and cross sections, flow velocities, land cover and land use) into hydraulic models offers the potential to significantly enhance our understanding of hydrological processes and improve predictive capabilities. Case studies from the Global South are strongly encouraged, as are contributions from ungauged and transboundary basins where satellite observations are the primary record.

This session invites cutting-edge presentations on advancing surface water and flood monitoring and mapping through remotely-sensed data, focusing on:
-Flood and surface water mapping from satellite and airborne sensors, including commercial constellations and emerging observation concepts
-Machine learning, deep learning and foundation models for water body detection and flood segmentation
-Flood mapping in challenging environments such as urban, vegetated, arid, and cloud-persistent regions
-Multi-sensor and multi-frequency fusion for consistent flood extent retrieval
-Flood hazard, risk and impact mapping from Earth observation
-Assimilation of satellite-derived flood extent and inundation maps into hydrological and hydraulic models
-Calibration and validation of hydrological and hydraulic models using remotely sensed observations and field measurements, as well as the creation and use of benchmarks, uncertainty quantification methods, and standards for this purpose
-Monitoring surface water dynamics and change such as seasonal water body variation, wetland extent, river planform
-Operational rapid mapping and emergency response, and the path from research product to sustained service or tool
-Multi-satellite data products and services for improved monitoring of surface water dynamics including floods

Early career and underrepresented scientists are particularly encouraged to participate.

Increasing climate variability and water scarcity are placing unprecedented pressure on agricultural systems worldwide. To address these challenges, the agricultural sector is rapidly adopting precision technologies that combine remote sensing capabilities with artificial intelligence to optimize crop water management. This session focuses on cutting-edge applications of remote sensing technologies and AI-driven analytics in transforming agricultural water management practices. We welcome research that demonstrates innovative approaches to monitoring, predicting, and managing agricultural water resources by integrating datasets from Earth observation satellites, unmanned aerial systems, Internet of Things (IoT) sensors and using advanced computational methods.
Key topics include:
Multi-platform remote sensing applications (satellite, UAV, hyperspectral, SAR, thermal infrared) for estimation of soil moisture and crop water requirement.
Deep learning and AI-driven detection of crop water stress and yield prediction.
Precision irrigation systems and automated water management technologies.
Multi-sensor data fusion combining space-based, airborne, and ground observations.
Real-time monitoring systems and subseasonal-to-seasonal water demand forecasting models.
Digital agriculture platforms and decision support tools for sustainable water management.
LLM-augmented modeling and digital twin of the agri-climatic or agri-weather landscape.
This session aims to showcase practical solutions that bridge the gap between technological innovation and real-world agricultural applications, emphasizing scalable approaches that support both productivity and environmental sustainability.

Currently geological collections, including the institutions and professionals curating them, face growing operational and financial threats. Several University Earth Science Departments and scientific Institution face merging with other disciplines or closure and their historical collections are at risk of destruction. Museums and other institutions are struggling with financial resources and space. Yet, in an era dominated by modelling and big data, the strategic, scientific, and educational value of geo-collections is frequently unrecognized. While there is the need to preserve important and, in some cases, unique collections of specimens, historical photographs, field documents, geological maps and scientific instruments, there is an urgent need to discuss how to collect for the future and what it is the best way to use both existing and future collections.
This session provides a platform to showcase the scientific, educational, and societal value of physical and digital sample repositories. We invite contributions demonstrating how collections drive the advance of geological research. In this context, “collections” encompasses repositories managed by museums, universities, volcano observatories etc. We welcome submissions exploring case studies using physical or digital sample repositories for research, innovative methodologies for enhancing FAIR (Findable, Accessible, Interoperable, Reusable) data principles and Open Science, interdisciplinary applications of geo-collections, best practices for physical preservation, archiving and conservation, funding strategies, community engagement and lessons learned.
We aim to bring together curators and researchers to foster international collaboration, advocate for institutional support, and promote the active use of geo-scientific collections.

Environmental science, scientists and developers increasingly use or build Artificial Intelligence in tools, Digital Twins and models, but are concerned about trust, transparency and digital sovereignty when supporting Open Science practices. AI-ready research assets and VRE's are crucial enablers.
Fair Digital Objects (FDO)'s become the technical backbone that makes AI trustworthy enough to use. Agentic search is the clearest case for this: an agent querying across catalogues can only return trustworthy results if the underlying objects carry typed metadata, provenance and clear licensing.
Research Infrastructures, providing the standards for FDO's and trustworthy data are at the center of these developments and have a responsibility to do a risk and value assessment before any AI tool is adopted, model cards, transparency, and shared European compute rather than dependence on commercial cloud providers.
Virtual Research Environments (VRE) and workflow builders provided by Research Infrastructures need to take these requirements into account and provide solutions that are machine actionable while at the same time are trustworthy for any AI application.
We envision Virtual Research Environments (VREs) that are not only machine-actionable but also deeply trustworthy, designed to integrate AI assistance while upholding FAIR principles, digital sovereignty, and Open Science standards. These VREs should empower scientists to safely explore AI-driven opportunities -whether in data sharing, integration, or analysis- with full confidence in the tools, data, and workflows provided. By embedding provenance, interoperability, and ethical AI practices, we can ensure that VREs become catalysts for collaborative, reproducible, and impactful environmental research.

This session aims to bring together practitioners, developers and stakeholders to discuss the technical challenges and share practices with the aim to contribute to the emerging ENVRI & Data Terra/EOSC nodes.

abstracts can focus on (but are not limited to):
- Describing concepts, principles, visions and challenges underlying the use and implementation of AI-readyness in environmental research and data management
- Demonstrating tools and applications, research assets that take these requirements into account and at the same time revolutionise the possibilities of data sharing, integration and analysis
- Presenting challenges, lessons learned and achievements in AI enhanced environmental and earth science

Every day, research vessels collect continuous data underway, improving our understanding of the state of the ocean. Most importantly, they reveal how the marine environment is changing. These underway data streams together with stationary data are fundamental to global ocean observations. They support applications of societal significance that influence our everyday life, most prominently weather forecasts. However, workflows and digital infrastructures remain mostly standalone solutions scattered across institutions and countries. Unifying the work is essential, however there is often a lack of attention for addressing the issue in detail. Often the focus spreads across broader topics (sessions) where its unique challenges and solutions often remain unseen.

Standardization of data workflows can help to optimize data output of research cruises around the world. The German Marine Research Alliance (DAM) brought together a consortium of data experts from several German research institutions, universities and federal agencies. Since 2019, DAM has been developing an integrated framework to enhance the acquisition, transportation, processing and publication of data and metadata from German research vessels, particularly from permanently installed systems, to enable subsequent reuse. The extension and connection of research data infrastructures and existing modular workflows is used by a team of data stewards to combine sensor streams, automated quality‑control routines aiming for standardized data products and open data services. Early results show clear gains in consistency, reproducibility and accessibility leading to higher visibility and accelerated usage of acquired data from German research vessels. This session brings these stories together.

By gathering other national and international initiatives in one place, this session aims to build a shared space for a global community. Since similar efforts are emerging in Europe and worldwide this session shall support meaningful exchange and future global collaboration. We welcome submissions that share the overall vision of disseminating data globally. We invite contributions that advance shipboard data workflows, real‑time and delayed‑mode automation, interoperable observing infrastructures and FAIR data practices. We welcome innovative quality‑control methods, scalable data pipelines, and end‑to‑end sensor‑to‑archive solutions that strengthen transparency and cross‑platform compatibility.

Scientific software development dates back 80 years, however, software sharing is still a relatively new discipline, and many approaches and guidelines are still evolving. Research software is an essential component of modern geoscience, supporting data analysis, modelling, visualization, machine learning, and computational workflows. Like scientific data, software plays a critical role in enabling research outcomes, yet the effort involved in developing, adapting, maintaining, and applying software is often under-recognized.
In this session, we welcome contributions that demonstrate how software has enabled scientific discovery, streamlined research workflows, accelerated analysis, facilitated collaboration, or helped address geoscientific challenges. We particularly encourage real-world applications, case studies, lessons learned, and practical experiences from across the geosciences, including geophysics and seismology, environmental sciences, geochemistry, geomorphology, hydrogeology, and planetary sciences. Contributions may present software developed, adapted, or integrated to support scientific workflows, as well as best practices related to usability, reproducibility, FAIR and Open Science, interoperability, documentation, sustainability, and software reuse. Live demonstrations, videos, and interactive examples are particularly encouraged (supplementary materials can be hosted on the EGU27 platform for access after the conference).
Join us to share examples, experiences, and best practices that demonstrate the value of software as a research asset and help advance geoscience through more effective, accessible, and reusable scientific workflows.

Rapid advances in Artificial Intelligence (AI) are transforming scientific research. Although geoscience foundation models are still emerging, it is already clear that their development depends on access to large volumes of data. At the same time, the reliability and trustworthiness of AI systems depend on understanding the quality, provenance, timeliness, and governance of the data they are built on. As the saying goes: garbage in, garbage out.

Over the past two decades, the Open Science principle of “as open as possible, as closed as necessary” has encouraged greater transparency and accessibility across the scientific process. However, the rise of AI raises new questions about how to apply this principle in an era of large-scale data aggregation and automated knowledge extraction.

Who benefits from AI systems built on openly shared scientific data? Are our data, metadata, and repositories AI-ready? Can existing research infrastructures support the demands that AI agents may place on them? How can provenance, attribution, licensing, and data-governance principles (particularly Indigenous) be preserved when scientific outputs become inputs for AI models?

This session invites researchers, data stewards, repository operators, infrastructure providers, and AI practitioners to discuss the opportunities and challenges that AI presents for Open Science. Contributions are encouraged on AI readiness, data quality and provenance, repository sustainability, attribution and reward mechanisms, governance frameworks, and emerging tensions between the goals of scientific openness and the development of AI systems. Together, we will explore how the scientific community can uphold the values of Open Science while adapting to a rapidly evolving AI landscape.

In-situ observations are essential for understanding and monitoring the Earth system and supporting decision-making. At the same time, the in-situ data landscape is evolving, with new technologies and methods emerging, including low-cost sensors, Internet of Things (IoT), citizen science or drones. However, integrating these heterogeneous observations remains challenging due to differences in data sources, formats, quality and standards, making them difficult to discover, access and use, while relevant data gaps remain.

This session aims to focus on innovative in-situ observation approaches and their integration with existing observing networks and research infrastructures. Contributions on topics as data quality, interoperability, FAIR practices and metadata management, automated processing and aggregation, as well as approaches for defining user observational requirements and identifying the variables to be measured, are particularly encouraged. Examples of integrating in-situ observations with remote sensing satellite data, modelling and other geospatial information are also of particular interest.

By bringing together researchers and data users, observing networks, data managers, and open science enthusiasts, the session will highlight how emerging observing methods and open data sharing practices can address data gaps, strengthen existing data infrastructures and networks, and contribute to more sustainable and fit-for-purpose in-situ Earth observation system.

Geological mapping and modelling are fundamental pillars of the geosciences, providing the basis for understanding Earth and planetary systems. This session brings together contributions that span the full spectrum of geological mapping and modelling, including fieldwork, methodologies, tools, and case-studies. The session covers the range from traditional field-based methods to cutting-edge approaches, including AI, applied in the most extreme and inaccessible environments on Earth and beyond.
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.

Making data Findable, Accessible, Interoperable and Reusable (FAIR) is now widely recognised as essential to advance open and reproducible research. Increasingly, this requires not only shared principles but also concrete digital implementations that enable interoperability across systems, disciplines, and infrastructures. However, it is very difficult to translate these principles into practical data management guidelines or operational digital solutions across disciplines. The goal of the session is to explore how best data management practices are developed, implemented, and adopted across disciplines, including through interoperable digital objects, persistent identifiers, and emerging data space concepts.

As part of this session, we invite submissions that:
1) Share good or bad experiences developing, implementing, and adopting data practices that align with both FAIR principles and the evolving needs of specific research communities.
2) Propose strategies for engaging researchers in adopting and refining best practices, with a focus on community-driven approaches to technical standards and infrastructures.
3) Explore the role of cultural change in enabling adoption of sustainable data practices.
4) Highlight efforts that harmonise data formats and workflows across disciplines while respecting domain-specific requirements, for example through interoperable data architectures or research data spaces.
5) Present technical or conceptual approaches that support the transition from data silos to interoperable, FAIR-aligned data ecosystems.

This session is aligned with the objectives of the Research Data Alliance (RDA) Earth, Space, and Environmental Sciences (ESES) Data Community of Practice and aims to foster cross-disciplinary dialogue, particularly among researchers in hydrology, seismology, and ocean sciences. However, we welcome contributions from all disciplines, especially where they provide insights or novel approaches to community engagement.
By learning from diverse experiences, this session seeks to advance collective understanding of how to build and sustain data practices that are both FAIR and fit for purpose, from community processes to concrete, interoperable digital implementations.

The FAIR principles are key to reusing Earth system data for analytics and AI. A primary challenge for reuse is judging a dataset's fitness for purpose, which depends on its representational, relational, contextual, and error characteristics, aspects not necessarily captured by traditional curation and documentation. As machine learning and AI become central to Earth system modeling and prediction, unintentional misuse becomes a critical hazard: data unfit for training models or their components can lead to downstream societal harm.

The session covers observations, simulations, reanalyses, and synthetic data. Training, fine-tuning, evaluation, data assimilation, downscaling, bias correction, and attribution each impose distinct data requirements, and provenance of these processes helps trace how data characteristics and uncertainties propagate to results. A key question is how to tell users and automated processes whether a dataset supports the intended use. Topics include:

- Metadata standards and documentation of intended and unsupported uses and dataset limitations
- Assessment methods combining metadata-derived indicators, content checks, lineage, validation, and domain expertise, and the limits of each
- Purpose-specific assessment of resolution, coverage, representativeness (extremes, transferability across regions, periods, and observing systems), physical consistency, and uncertainty
- Time, financial, and energy costs of assessing and improving fitness, how they are distributed, and the consequences of unsuitable reuse

Some reuse challenges can be addressed through curation or supplementary information; others require alternative data or narrower scientific claims. Contributors are encouraged to state the intended use, present assessment evidence, and describe resulting decisions and their effects on conclusions. Comparative studies, negative results, and post-mortems with transferable lessons are particularly welcome, including assessments that find a dataset unsuitable. Contributions may address single assessment steps or complete data-to-discovery workflows.

The session convenes data producers, repositories, research infrastructures, AI practitioners, and domain scientists to identify shared assessment practices that avoid redundant effort and to clarify when scientific judgment is essential. Its focus is how data properties and assessment evidence support or constrain scientific claims.

Africa’s biodiversity and ecosystems are central to the continent's green transition, yet effective conservation strategies rely heavily on the availability of accurate, accessible, and interoperable environmental data. While emerging initiatives, such as the Africa Regional Centers of Excellence Programme (ArcX), are establishing critical reference points for biodiversity data, significant technical and systemic challenges remain in harmonizing fragmented datasets across borders.
This session addresses the intersection of open science, spatial data harmonization, and cloud-based environmental computing applied to African landscapes. We aim to showcase innovative technological solutions and collaborative frameworks that bridge the gap between data scarcity and actionable ecological insights.
We welcome contributions from researchers, data scientists, and practitioners focusing on, but not limited to:
• The development, deployment, and scaling of open-source architectures and Spatial Data Infrastructures (SDIs) tailored to African contexts.
• Novel Earth Observation (EO) applications and machine learning workflows for feature extraction and biodiversity monitoring in data-scarce regions.
• Strategies, standards, and case studies for cross-regional data harmonization (e.g., integrating data across Central, Western, and Eastern Africa).
• Successful capacity-building frameworks, FAIR (Findable, Accessible, Interoperable, and Reusable) data practices, and cross-border partnerships that enable open environmental data sharing.
By bringing together geospatial informatics and spatial ecology, this session aims to foster interdisciplinary dialogue on building robust data ecosystems that empower sustainable ecosystem management across Africa.

The Earth’s Critical Zone (CZ) is the “heterogeneous, near surface environment in which complex interactions involving rock, soil, water, air, and living organisms regulate the natural habitat and determine the availability of life-sustaining resources”. Understanding this heterogeneity requires integrating highly diverse datasets across space, time, formats, and domains. However, dealing with the extreme data fragmentation in environmental Earth sciences remains a major challenge. This session highlights the implementation of FAIR geospatial methodologies to accelerate data assimilation, predictive modelling, and the generation of actionable Earth Intelligence regarding Critical Zone behaviour.
We welcome contributions showing technical workflows, digital infrastructures, and methodologies designed to break down domain silos. These should address the seamless integration of in-situ monitoring networks with multi-sensor remote sensing (RS) data products, focusing on:
• Format harmonization strategies bridging disparate spatial-temporal models and interdisciplinary data sources.
• The application of standard APIs (e.g., OGC standards, openEO) for scalable data access.
• Modern, distributed metadata catalogues enhancing data discoverability.
• Semantic harmonization efforts, including domain ontologies and controlled vocabularies to eliminate ambiguity across disciplines (e.g., Essential Variables (EVs) across the hydrosphere, biosphere, atmosphere, and geosphere).
• The use of geospatial data pipelines for Analysis Ready Data (ARD) and AI-ready data to streamline machine learning workflows.
• Advanced AI modelling (e.g., physics-informed neural networks, deep learning) to extract scalable CZ Earth Intelligence.
• The alignment of CZ science with large-scale initiatives, such as Digital Twins (DestinE) and the Common European Data Spaces (e.g., Green Deal Data Space).
This session aims to foster an interoperable, open-science ecosystem that transforms fragmented observations into scalable Earth Intelligence and actionable insights for modelling and managing our living planet.

Artificial intelligence is rapidly transforming the Earth and environmental sciences, yet the effectiveness and reliability of AI depend fundamentally on the quality, interoperability, semantics, provenance, and machine-actionability of the underlying data. While FAIR principles have substantially advanced scientific data stewardship, emerging GeoAI, foundation models, multimodal AI, and agentic systems require research data infrastructures that move beyond data accessibility toward AI readiness.
This session brings together the geoscience, data science, research infrastructure, and semantic-web communities to explore how FAIR and interoperable data ecosystems can support trustworthy AI-enabled scientific discovery. We particularly welcome work addressing semantic interoperability, FAIR Digital Objects, knowledge graphs, machine-actionable metadata, controlled vocabularies and ontologies, provenance, AI-ready data standards, and emerging frameworks such as CDIF, Croissant, Semantic Croissant, and GeoCroissant.
We invite methodological developments, infrastructures, standards, demonstrators, and applications involving Earth observation, climate and environmental data, geospatial information, urban digital twins, sensor networks, social and human-generated data, and other heterogeneous scientific resources.
The session aims to connect data stewardship and interoperability communities with researchers developing the next generation of AI-enabled Earth and environmental science, and to identify practical pathways toward open, interoperable, trustworthy, and AI-ready scientific ecosystems.

The concept of Earth as the sole body in the Solar System with liquid water that can harbor microbial life has been overturned by the discovery of multiple 'ocean worlds'. The Solar System is home to several planetary bodies with subsurface oceans of liquid water, including icy satellites such as Europa, Ganymede, Callisto, Enceladus, Titan and Triton, as well as dwarf planets like Pluto, and chief among these ocean worlds, the Earth. Furthermore, new icy and ocean worlds are being continuously discovered in other planetary systems as well. Like Earth, the exploration of these oceans includes both aspects of planetary evolution and habitability. The geodynamic role of oceans in planetary evolution is thus a crucial aspect of understanding not only planet formation, but the onset of biological activity as well. In what ways can the oceans of Earth serve as analogs for other oceans of the Solar System? What instrumentation can be implemented on the Earth now to further our understanding of these ocean worlds, and what technological advances might we expect in future exploration of subsurface liquid water environments beyond Earth?

This session focuses on analog sites, laboratory simulation, modeling, instrumentation and mission proposals. Coordination between Earth, marine and planetary science communities is encouraged, as well as emphasis on upcoming (e.g. JUICE and Dragonfly) and proposed missions (e.g. Enceladus Orbilander). Analog sites might encompass either geological or biological themes in the broader frame of habitability. Interfaces of ice-water (e.g. underside of floating ice shelfs and subglacial lakes), clathrate-water (e.g. ocean floor sediments, veins/fractures/faults, layered horizons and atmosphere particulates), seafloor-ocean, and rock-ice (i.e. glaciers) are of particular curiosity. Instrumentation includes sensors, buoys, submersibles, drilling and coring, as well as satellite instrumentation (e.g. spectrometers, magnetometers and gravimeters).

This session invites contributions on the latest developments and results in lidar remote sensing of the atmosphere, covering • new lidar techniques as well as applications of lidar data for model verification and assimilation, • ground-based, airborne, and space-borne lidar systems, • unique research systems as well as networks of instruments, • lidar observations of aerosols and clouds, thermodynamic parameters and wind, and trace-gases. Atmospheric lidar technologies have shown significant progress in recent years. While, some years ago, there were only a few research systems, mostly quite complex and difficult to operate on a longer-term basis because a team of experts was continuously required for their operation, advancements in laser transmitter and receiver technologies have resulted in much more rugged systems nowadays, many of which are already operated routinely in networks and several even being fully automated and commercially available. Consequently, also more and more data sets with very high resolution in range and time are becoming available for atmospheric science, which makes it attractive to consider lidar data not only for case studies but also for extended model comparison statistics and data assimilation. Here, ceilometers provide not only information on the cloud bottom height but also profiles of aerosol and cloud backscatter signals. Scanning Doppler lidars extend the data to horizontal and vertical wind profiles. Raman lidars and high-spectral resolution lidars provide more details than ceilometers and measure particle extinction and backscatter coefficients at multiple wavelengths. Other Raman lidars measure water vapor mixing ratio and temperature profiles. Differential absorption lidars give profiles of absolute humidity or other trace gases (like ozone, NOx, SO2, CO2, methane etc.). Depolarization lidars provide information on the shapes of aerosol and cloud particles. In addition to instruments on the ground, lidars are operated from airborne platforms in different altitudes. Even the first space-borne missions are now in orbit while more are currently in preparation. All these aspects of lidar remote sensing in the atmosphere will be part of this session.

Two consecutive EGU sessions have demonstrated how the convergence of time-series aerial and satellite remote sensing, geospatial technologies, and artificial intelligence is transforming our ability to detect, monitor, and understand environmental change. Applications have ranged from local studies of agricultural systems and natural resources to large-scale analyses leveraging foundation-model embeddings derived from tens of millions of Earth observation scenes. As these capabilities mature, the next challenge is operational: translating scientific advances into decision-ready products and services that support action by hazard managers, conservation organizations, agricultural and water authorities, resource managers, and communities.
This third session in the Connected Earth series invites contributions that apply time-series Earth observation, geospatial technologies, and machine learning and AI to address environmental challenges across scales, from local sites to global systems. We welcome contributions using optical, SAR, and other remotely sensed data acquired from aerial and satellite platforms, as well as integrated geospatial workflows that connect Earth observation insights to real-world decisions.
Topics of interest include, but are not limited to:
• Operational and near-real-time monitoring systems based on cloud-native Earth observation platforms and GIS-integrated workflows.
• Multi-decadal environmental monitoring and approaches for maintaining consistency across evolving sensor generations.
• AI and machine learning methods, including foundation-model embeddings, for environmental analysis, classification, prediction, and change detection.
• Open, reproducible, and interoperable geospatial workflows that improve transparency and collaboration.
• Applications supporting hazard monitoring, natural resource management, agriculture, water resources, conservation, cryosphere research, climate adaptation, and environmental resilience.
The session welcomes both methodological and applied contributions from academia, government, industry, and community-based organizations. By bringing together advances in Earth observation, AI, and geospatial technologies, Connected Earth III aims to highlight pathways that move beyond proof-of-concept research toward scalable, operational solutions that support informed environmental decision-making.

Coastal zones are rapidly changing environments exposed to interacting processes and hazards across a wide range of spatial and temporal scales. Understanding and anticipating these changes requires continuous observations that capture both long-term trends and rapid responses to extreme events. Synthetic Aperture Radar (SAR) provides unique advantages for coastal monitoring through its all-weather, day-and-night imaging capability and sensitivity to surface roughness. Combined with multisource Earth observation, in situ measurements, numerical and data-driven modelling and artificial intelligence (AI), SAR offers new opportunities for integrated coastal monitoring, forecasting and dynamic virtual platforms for simulation and interactive scenario testing (digital twins).
This session welcomes contributions presenting new methods and innovative applications of SAR and advanced SAR techniques to support coastal monitoring, understanding, forecasting, adaptation and management. We invite studies addressing applications including multiscale coastal geomorphology, coastal hydro-morphodynamics and coastal–ocean interactions, coastal hazards and their drivers (e.g. flooding, erosion, subsidence and extreme events), coastal vegetation and ecosystems, and land-cover and land-use change.
Contributions combining SAR with complementary data sources, AI/ML and numerical modelling across different coastal typologies and scales are particularly encouraged, from the reconstruction of historical changes to forecasting, near-real-time applications, as well as the development and continuous updating of coastal digital twins.
The session aims to bring together the SAR and broader Earth observation, coastal and ocean science and management communities to exchange new methodologies and applications. It will explore how SAR-derived information can contribute to the continuum from coastal observation and monitoring to forecasting and digital twins, supporting coastal adaptation pathways, planning and management.
Attendees will gain an overview of emerging methods, applications and opportunities in this fast-developing frontier, while connecting with researchers and practitioners across the coastal, Earth observation, modelling and management communities. The session will provide an opportunity to hear updates on work progressing across Europe, understand how these efforts are being integrated, and identify potential synergies, knowledge gaps and avenues for future collaboration.

Agricultural systems are increasingly exposed to droughts, heatwaves, floods, and compound extremes, with impacts that can cascade across water resources, crops production, and rural livelihoods. Addressing such challenges requires better hazard monitoring or sensing technologies. A stronger link between observation, prediction, and the practical adaptation decisions is also needed that allow agricultural systems to anticipate, cope with, and recover from natural hazards.
This session brings together research that connects these different steps, from monitoring and prediction to adaptation and decision-making in agriculture under hydro-climatic extremes. Although hazard monitoring and precision agriculture technologies have advanced rapidly, turning these developments into tangible improvements in resilience remains challenging. This requires closer integration of hydrology, agriculture, remote sensing, climate science, and decision-support approaches.
We welcome contributions covering the full chain from observation to action, including Earth Observation and in situ monitoring of agricultural exposure and impacts; AI, machine learning, and hybrid modelling for impact prediction and early warning; agro-hydrological and land surface modelling under changing and compound climate conditions; and the use of monitoring and prediction to support adaptation, such as smart irrigation, precision water management, nature-based solutions, and climate-resilient cropping systems.
We particularly encourage studies that show how advances in observation and modelling can be translated into practical decision-support and adaptation. Contributions addressing compound and cascading hazards, especially those linking monitoring and modelling with agricultural impacts and management responses, are also welcome.

Data sovereignty has several scales. We can refer to the sovereignty of individual data producers. We can also refer to sovereignty in a community that works together in a data space. We can scale the issue to sovereignty of some territories that they are investing in a digital economy and they do not want benefits to leak to other territories. Among the different interpretations, there is also the sovereignty of scientists that want to secure their reputation by ensuring acknowledgment and recognition. Scientifics data sovereignty faces three main challenges: (1) the scientific community is globalised, and while scientists may suspect some issues surrounding lack of sovereignty, it is often perceived as a constraint on their autonomy and freedom to choose the most efficient tools and workflows; (2) there is no specific European directive for data sovereignty. Instead, a collection of policies was approved in different moments in time lacking a consistent framework; and (3) scientists are not familiar with the tools and technologies used to implement data sovereignty at different restriction levels.
While several efforts have been recently focusing on restricted data (e.g. EU Data Spaces), this session focuses on the open science spirit of “as open as possible, as closed as necessary” under Horizon Europe and other funding programs, but reinterprets it from the perspective of data sovereignty. This session will focus on why data sovereignty is important for Europeans, how scientists are currently addressing this issue and how we can ensure data sovereignty in Europe. Some possible topics to address in the session are acceptability of data to specific communities, governance of the data and interoperability of data and technologies. Papers participating in the session will collaborate to propose changes in practices in the scientific community; balancing open science and data sovereignty to maximize impact on society.
We are interested in understanding how you perceive the outcomes of restricting global open science with sovereignty safeguards. In particular we would like hear your thoughts and fears on why sovereignty is necessary and the barriers it can create for operating European Research Infrastructures.