Meet BeyondEarthSTAC and learn how the SpatioTemporal Asset Catalog (STAC) standard, widely used for Earth Observation data, can also be applied to planetary datasets. The proof-of-concept currently brings together six lunar collections from several space missions, providing a consistent way to discover and access imagery, topographic data and hyperspectral observations. CREODIAS users will find the mechanics familiar. BeyondEarthSTAC uses STAC in the same way it is used for Earth Observation data: one harmonised metadata layer is applied over products from multiple sources, with spatial and temporal search and direct access to assets over S3. This time, the subject is the Moon.
Lunar observation data is publicly available, but scattered across different archives, missions and formats. BeyondEarthSTAC explores how a common catalogue approach can make these datasets easier to discover and work with. As this is a proof of concept, a limited set of collections is currently available, with the architecture designed to accommodate further datasets.
What is in the catalogue
The demo release covers six collections, spanning panchromatic imagery, hyperspectral observations and topography:
- Chandrayaan-2 IIRS Calibrated- level-2 products from the Imaging Infrared Spectrometer (IIRS), covering 0.8 to 5.0 µm in 256 contiguous bands, with spectral radiance corrected for instrument effects. Used for lunar mineralogy, surface composition and reflectance studies.
- Chandrayaan-2 TMC-2 Calibrated- level-2 stereo imagery from the Terrain Mapping Camera-2 (TMC-2), a push-broom panchromatic camera acquiring data at approximately 5 m/pixel from a 100 km orbit, corrected for radiometric and geometric distortions. Used for lunar topographic mapping, geological studies and landing site characterisation.
- Chandrayaan-2 TMC-2 Derived Digital Terrain Model- elevation models at approximately 10 m/pixel, produced by photogrammetric processing of forward- and aft-looking stereo pairs, with heights referenced to the Moon's mean radius and provided in selenographic coordinates. Used for lunar topographic analysis, geomorphological mapping and mission planning.
- Chandrayaan-2 TMC-2 Derived Orthorectified Image- 5 m/pixel imagery projected onto a consistent map grid using the corresponding DTM, with terrain-induced distortions removed, giving planimetrically accurate lunar surface images in selenographic coordinates. Used as a precise base layer for lunar geological mapping, surface feature analysis and geographic registration of other datasets.
- SLDEM2015 - 512 Pixels/Degree Lunar Digital Elevation Model- the full-resolution global lunar DEM (approximately 59 m/pixel at the equator), derived from data acquired by the Lunar Orbiter Laser Altimeter (LOLA) aboard NASA's Lunar Reconnaissance Orbiter (LRO), co-registered with stereo imagery from the Terrain Camera (TC) on JAXA's SELENE/Kaguya mission between 60°S and 60°N, and LOLA-only data outside that range. The product is delivered as a set of geographic tiles, each accompanied by a derived slope map. This is one of the most precise lunar topographic products available. Used for high-resolution terrain analysis, mission planning and photometric correction workflows.
- SLDEM2015 - 256 Pixels/Degree Lunar Digital Elevation Model- the same LOLA and SELENE/Kaguya TC product at approximately 118 m/pixel at the equator, also tiled and accompanied by slope maps. The lighter data footprint makes it the more practical choice where full resolution is not required. Used for regional-scale analyses and global topographic modelling.
Together, these collections cover the Moon at scales from 5 m to approximately 120 m and characterise three complementary aspects of the terrain, its morphology, its topography and its surface composition. The source products come from different missions and instruments, but in BeyondEarthSTAC they are described using the same STAC standard, enabling more efficient search and analysis across all of them.
The catalogue is aimed at anyone who treats lunar data as an input to their own work: planetary scientists and remote sensing researchers mapping mineralogy or reflectance over a region of interest, mission planners and landing site analysts deriving slope and roughness for candidate sites, space industry teams working on surface operations and lunar infrastructure, and data engineers assembling training sets for crater and terrain detection models.
What's next
BeyondEarthSTAC is a proof of concept, and the architecture is deliberately open-ended. Because every collection follows the same STAC schema, datasets from other missions, historical and future, can be added without reworking the underlying infrastructure. Wider coverage of lunar missions and instruments is a possible next step, and the same model applies to other planetary bodies.
We encourage CREODIAS users to try out the catalogue and share feedback on what would make it more useful.
Exos, a next-generation access layer for EO data, has been deployed into production. The new solution replaces the legacy S3 Endpoint and introduces a more scalable, resilient and efficient architecture for accessing large satellite data repositories. Exos has been designed to improve the performance and reliability of EO platforms while preparing CloudFerro's infrastructure for the next generation of data-intensive applications.
User benefits include:
• faster access to large EO datasets,
• improved scalability for growing data repositories,
• resilient, highly available data access,
• more efficient use of computing resources,
• enhanced monitoring and operational transparency.
Early performance benchmarks indicate double-digit improvements in data access performance while reducing infrastructure resource requirements. These optimisations help users process and analyse large satellite datasets more efficiently without changing their existing workflows.
Future releases will further expand data discovery capabilities while continuing to improve the availability, resilience and scalability of the data access layer.
MODIS data collections are now available on CREODIAS and Copernicus Data Space Ecosystem, providing access to over 25 years of continuous, sensor-consistent Earth Observation data from Terra and Aqua platforms.
The repository includes a wide range of analysis-ready products including vegetation indices, land surface temperature, snow cover, land cover type, and many more. These datasets are gridded and ready for direct use in environmental monitoring and long-term time series studies. Data can be accessed via S3 object storage, OData, and STAC APIs, with full metadata support.
Read this article MODIS Data Collections now on CREODIAS where our experts present four use cases covering deforestation monitoring, desertification analysis, fire detection, and snow cover time series Two of them can be executed directly on CREODIAS virtual machines without any data transfer.
More about MODIS collections: creodias.eu/eodata/modis
We are pleased to invite you to our upcoming webinar designed to present you a practical introduction to Earth Observation in the cloud.
Join us for a practical introduction to Earth Observation in the cloud, where data, catalogues, and infrastructure form a single analytical ecosystem. The session will demonstrate how EO data can be discovered, explored, and processed directly in the cloud using real use cases and hands-on workflows.
The session is technical, but presented in an end-to-end manner, focusing on how all components work together.
WHEN: Wednesday, January 14th, 14:00 CET
DURATION: 60 min + Q&A
> This webinar is for:
- data analysts and data scientists who want to work with Earth Observation (EO) data without downloading it locally
- GIS and remote sensing specialists interested in modern, cloud-based approaches to data discovery and processing
- IT, cloud, and DevOps engineers who want to see practical use of virtual machines, Kubernetes, and object storage in an EO context
- students and PhD candidates in technical fields (GIS, geoinformatics, computer science, data science)
- people starting their journey with the Copernicus / CDSE / CREODIAS ecosystem, who want to understand the full stack: data, catalogue, and infrastructure
- participants of the Geospatial Innovation Competition for the best solutions and apps that rely on spatial data
> What you will learn:
- understand what data is available in the CREODIAS platform
- use data access points such as CREODIAS Data Explorer, the OData catalogue, STAC, effectively for data discovery
- perform data discovery and exploration in Jupyter notebooks without downloading the data
- understand how virtual machines, Kubernetes, and object storage form a coherent analytical environment
- launch and configure a virtual machine using the CloudFerro dashboard
- understand the basics of infrastructure as code, using Terraform and Kubernetes
- learn from real business and research use cases based on EO data
Technical presentation - agenda
- CloudFerro & the EO Ecosystem
An introduction to CloudFerro Cloud and the CREODIAS platform. - From Projects to Practice
How data, catalogues, and infrastructure come together in real-world EO workflows. Overview of CloudFerro’s projects and examples of real-world applications. - EO Data in the Cloud
Available datasets and how to access them efficiently without local downloads (CREODIAS) - Data Catalogue & Discovery
Using CREODIAS Data Explorer, the OData, STAC, and Jupyter for fast and effective data discovery. - Cloud Infrastructure for EO
Virtual machines, Kubernetes, and object storage as a unified analytical environment. - Real-World Use Cases
Practical examples including oil spill monitoring and AI-powered data discovery. - Key Takeaways & Q&A
CREODIAS as a complete, ready-to-use Earth Observation stack - combining data, catalogues, and cloud infrastructure.
Presenters
Marcin Niemyjski
Data Scientist at CloudFerro
Marcin is an Earth Observation Data Engineer with a strong focus on cloud computing and geospatial data processing. Works on turning raw satellite imagery into reliable, analysis-ready datasets by building scalable services, optimizing data pipelines, and enabling seamless access to Earth Observation data for real-world applications.
Tomasz Furtak
Junior Data Scientist at CloudFerro
Tomasz focuses on Earth Observation data cataloguing and cloud-based processing, with particular expertise in the STAC standard. Works on analyzing Copernicus data, developing practical EO use cases, and exploring very high-resolution datasets from Copernicus Contributing Missions. Involved in metadata standardization activities supporting the Copernicus Programme.
Bartosz Staroń
Junior Data Scientist at CloudFerro
Bartosz works with Earth Observation data cataloguing based on the STAC standard, GIS analyses, and cloud-based EO data processing. Uses Python to support scalable data workflows and geospatial analysis for a variety of EO-driven applications.
Bartosz Augustyn
Junior Data Scientist at CloudFerro
Bartosz is involved in R&D activities related to Earth Observation data embeddings. Works with Python and GIS tools, combining data science methods to support the development of innovative EO products.