Satellite Cal/Val has traditionally relied on a wide range of datasets, including in-situ and aircraft campaigns, opportunistic inter-sensor comparisons, and modelled natural phenomena. While these approaches provide valuable information, it can be difficult to determine which datasets are of sufficient quality and optimised for satellite applications. ESA introduced the concept and label of FRMs to define a distinct class of observations specifically designed to meet the needs of satellite Cal/Val.

CEOS WGCV recognised the need for a formal framework to assess and endorse such measurements, leading to the development of the CEOS-FRM Assessment Framework. CEOS-FRM compliant data enable satellite operators to clearly understand the quality, traceability, and fitness for purpose of Cal/Val data for specific satellite products.

A number of in-situ reference networks and sensors have undergone preliminary evaluations against the CEOS-FRM Assessment Framework. These include:

• PGN (Pandonia Global Network)
• RadCalNet (Radiometric Calibration Network)
• ICOS (Integrated Carbon Observation System – Ecosystem Component)
• FRM4DOAS (Ground-Based DOAS Air-Quality Observations)
• BAQUNIN (Boundary-layer Air Quality-analysis Using Network of Instruments Super Site)
• CSIRO (DION)
• Brewer Spectrometer – NO2 Vertical Column Densities (VCDs) Measurements
• FRM4GHG 2.0 project – COCCON
• CREGARS (Centre for Reactive Trace Gases Remote Sensing)

The mandatory defining characteristics for CEOS-FRM are:

1. Traceability: FRM measurements are traceable to a community-agreed reference (ideally SI units) through metrology standards or best practices.
2. Independence of satellite-under-test: FRM measurements are independent from the satellite geophysical retrieval process.
3. Uncertainty budget: A comprehensive uncertainty budget for all FRM instruments and derived measurements is available and maintained.
4. Documented protocols: FRM measurement protocols, procedures and community-wide quality management practices are defined, published and adhered to.
5. Accessibility: FRM datasets, metadata, and processing information are openly available to allow independent verification of other processing systems.
6. Representativeness: FRM measurements are demonstrably comparable to what the satellite observes, with documented comparison methods and uncertainties.
7. Adequacy of Uncertainty: The uncertainty of the FRM and its comparison process meet the validation requirements of the satellite.
8. Utility: FRM datasets are designed to apply to a class of satellite missions and are not mission-specific.

In order to self-assess and verify a CEOS-FRM, a maturity matrix approach is used. Submissions are graded against specific criteria, and assigned colours indicating their level of compliance. Data providers pursuing CEOS-FRM compliance would self-assess their datasets against the matrix after collating the required documentation and evidence.

The matrix is a visual representation that provides a quick overview to assess the state of any FRM against each set of given criteria, making visible, for potential users, where it is mature and where evolution and effort are still needed. It contains categories to assess the FRM’s basic suitability for the intended measurand, the instrumentation, level of automation and processing, metadata format and availability, metrological quality, and the overall completeness, coverage and distribution of measurements.

WGCV is currently undertaking a series of pilot exercises to test the CEOS-FRM maturity matrix assessment, inviting providers of calibration and reference measurement networks to submit and evaluate FRMs against the framework. A second round of exercises is being conducted in 2026 and contributions from FRM providers are welcomed.

Interoperability refers to the ability of data, systems, and services that are often developed independently to work together seamlessly. For Earth observation (EO) data providers and distributors, this means enabling users to discover, access, integrate, and analyse data from multiple sources without unnecessary technical, semantic, or policy barriers.

Interoperability of data and services in the space-based EO domain is essential due to fundamental limitations of the data, including revisit times, potential launch or on-orbit failures that can result in data gaps, and spatial resolution constraints, as well as increasing cooperation between public and commercial space actors worldwide. 

Recognising the importance of EO interoperability, CEOS developed and published the CEOS Interoperability Handbook 1.1 in 2008. Fifteen years later, the rapid increase in the number of EO satellites and associated complex sensors providing ever-growing volumes of data, combined with more complex user requirements, motivated CEOS to develop the CEOS Interoperability Framework, which was endorsed at the 2023 CEOS Plenary. Building on this framework, CEOS subsequently reviewed the interoperability of data services in the current context and developed the CEOS Interoperability Handbook 2.0, which was endorsed by CEOS Plenary in 2025. The development was led by the CEOS Working Group on Information Systems and Services (WGISS), with contributions from various CEOS entities, including theWorking Group on Calibration and Validation (WGCV), CEOS Systems Engineering Office (SEO), the CEOS Analysis Ready Data Oversight Group, and the Land Surface Imaging Virtual Constellation (LSI-VC). The Handbook underwent broad community consultation throughout 2025, including in-depth review and discussion at WGISS, WGCV and Strategic Implementation Team (SIT) meetings. In line with CEOS recommendations, it is maintained openly on GitHub, enabling transparency, version control, and ongoing feedback from the wider EO community.

The handbook guides EO data providers, system architects, and policy stakeholders, across both the public and private sectors, in building interoperable data and service infrastructures. It emphasises modularity, openness, and machine-actionability, aligning with global initiatives under the Group on Earth Observations (GEO) including GEO Data Licensing Guidance and QA4EO, FAIR data principles, and open science mandates.

Factors of Interoperability

The Handbook provides a total of 76 recommendations across five complementary factors, each addressing a distinct but interrelated dimension:

• Vocabulary (Semantics) details how terms and expressions should be named and defined to ensure concepts are understood consistently across organisations.
  • 14 recommendations, across Semantic and Thesaurus sub-sections
  • Led by LSI-VC
• Architecture describes the organisational structure of concepts, processes, and data assets, including models and standards governing data collection, storage, archiving, documentation, and publication.
  • 25 recommendations, across Preservation, Data & Metadata, and Publishing sub-sections
  • Led by SEO and WGISS
• Interface (Accessibility) addresses how users and systems discover and access data and services, including data exchange protocols and application interfaces.
  • 16 recommendations, across Data Discovery, Data Access, and Authentication & Authorisation sub-sections
  • Led by WGISS
• Quality focuses on the indicators for informing users about the trustworthiness of the data, including accuracy, uncertainty, and consistency.
  • 9 recommendations
  • Led by WGCV
• Policy considers the legal, organisational, and governance context, including data licensing, access conditions, procurement, and alignment with open data and open science principles.
  • 12 recommendations
  • Led by WGISS and SEO

The Handbook recognises that modern EO systems must support distributed, cloud-native, and user-driven architectures. The emphasis is on enabling seamless integration across agencies, platforms, and applications, rather than simply connecting catalogues or services.

To support practical adoption of the Handbook, CEOS, led by WGISS, is developing two follow-on activities. The first is an Interoperability Maturity Matrix, which will allow data and service providers to assess their current level of interoperability across the five factors, track progress over time, and identify priority areas for improvement. The matrix will be aligned with existing CEOS Maturity Matrix frameworks developed by WGCV and WGISS.

The second activity is a set of Interoperability Demonstrators, which will be developed using the Interoperability Handbook and will help end users better understand the barriers to implementing interoperability.

CEOS encourages data providers, distributors, and service operators to review the CEOS Interoperability Handbook 2.0, consider its recommendations in their own operational contexts, and provide feedback through the GitHub repository.

CEOS-ARD team at LPS 2025

The Future of CEOS-ARD survey was opened in June 2025, and sought community feedback on matters that could be considered by an ambitious next-generation future CEOS-ARD Strategy. To date, the survey has collected over 110 responses, and remains open for those who haven’t yet shared their views. The survey was promoted at many events across the year, including European Space Agency’s Living Planet Symposium (LPS), the 2025 International Geoscience and Remote Sensing Symposium (IGARSS) and the 2025 International Astronautical Congress (IAC) – bringing in a wide range of perspectives. 

With this feedback in hand, the CEOS-ARD team identified fourteen categories that a future CEOS-ARD strategy should consider:

1. Consistent and Enhanced Metadata Specifications – There is an opportunity to increase alignment of CEOS-ARD Product Family Specification metadata, update requirements, and maximise compatibility with STAC and OGC/ISO standards.
2. Data and Metadata Quality –  While datasets might satisfy all of the reporting requirements of CEOS-ARD, there are no quantitative bounds placed on the ‘quality’ of these characteristics. Clearer expectations will build trust and aid data providers.
3. Ongoing Quality Assurance and Integrity Monitoring – Assessments are currently one-off exercises with no ongoing checks. Introducing continuous quality/integrity monitoring and provenance tracking would support ongoing tracking of performance and provide clarity for users.
4. Fitness for Purpose – Adding fitness-for-purpose guidance would help classify products, improve transparency, and support both users and providers.
5. Increased Support to Scientific Applications and Environmental Adaptation and Resilience – Including additional requirements related to long-term accuracy/stability  would facilitate long-term records, and support adaptation and resilience ambitions.
6. Solidifying the Business Case and Increasing Commercial Relevance – CEOS-ARD can support procurement as a clear and transparent common baseline and reduce duplication of effort and work on both the customer and data provider sides.
7. Tools to Aid Compliance Assessments and Peer Reviews – Manual compliance assessments are not scalable. Automated CEOS-ARD validation tools, web forms, and clearer guidance can streamline reviews, reduce burden, and increase uptake.
8. Supporting Opportunities and Reducing Risks Related to AI / ML – CEOS-ARD can be positioned as a “gold-standard” source for Geospatial Foundation Models (GFMs) and other AI/ML applications by requiring strong provenance and traceability information.
9. Measurand Consistency and Algorithm/Application Resilience – Improved measurand consistency would strengthen multi-sensor time-series use, operational resilience, and confidence in data supply.
10. Alignment with the Software Ecosystem – Uptake of CEOS-ARD will be limited if CEOS-ARD metadata and data are not readily interpreted by widely used software tools.
11. Analysis Ready Data Standards – CEOS-ARD can lead by developing a flexible, community-driven framework that matures into a formal standard at the appropriate time. CEOS should continue fostering broad engagement and develop a clear roadmap for ARD standards.
12. Thematic and Higher-level CEOS-ARD Products – Expanding to thematic and higher-level products can facilitate expanding user needs.
13. Training and Outreach – Expanded outreach, multilingual documentation, and provider/user support to boost user adoption.
14. Capturing Use and Impact – Compiling strong examples and compelling case studies will strengthen advocacy, demonstrate value, and encourage increased adoption.

CEOS Principals have now tasked the CEOS-ARD Oversight Group to develop a new 2026 CEOS-ARD Strategy, building on the concept note. This task has also been prioritised by the new CEOS Chair Team, formed of CSIRO, Geoscience Australia, and the Bureau of Meteorology. The 2026 CEOS-ARD Strategy will be delivered at the 40th CEOS Plenary, in November 2026. 

Community feedback is still welcome via the survey, and the team thanks everyone that has provided valuable feedback to date.

For more details on the survey responses and the topics described above, please see the concept note.

Due to their global coverage, satellites can provide critical information about active fires to disaster management agencies, to help identify new active fires and understand a fire’s size and movement. Sensors which image in the short (2.2μm) to mid-wave (3.7μm) infrared bands are widely used for active fire monitoring. 

Maximum revisit time (global aggregation) baseline scenario for fire monitoring/detection

Many CEOS Agencies have developed tools and services to operationally detect and monitor active fires. This includes: 

• Copernicus’ European Forest Fire Information System (EFFIS), which uses data from MODIS and VIIRS to detect active fires
• NASA’s Fire Information for Resource Management System (FIRMS), using data from MODIS, VIIRS, Landsat and various geostationary satellite sensors 
• NOAA’s Hazard Mapping System Fire and Smoke Product, using data from GOES ABI, MODIS, VIIRS and AVHRR
• NRCan’s Canadian Wildland Fire Information System, using MODIS and VIIRS
• Copernicus & GEO’s Global Wildfire Information System (GWIS), which brings together existing information sources at regional and national levels in order to provide a comprehensive view and evaluation of fire regimes and fire effects at global level

However, some of these operational tools are built on data from science missions with uncertain continuity (e.g. MODIS), and hence the continuity of these critical services are also uncertain. Furthermore, there are no current public missions carrying dedicated fire monitoring sensors, and there are often trade offs with other mission objectives when it comes to orbits and sensor characteristics. 

To respond to this challenge, WGDisasters established the Wildfire Pilot in 2021 with the aim to provide a comprehensive gap analysis for active-fire Earth observation, considering both the data supply and user requirements. The team have reached the conclusion of this activity, and are working with CEOS to develop recommendations to improve coordination of global active fire monitoring. 

Maximum revisit time in 2029 baseline scenario for fire monitoring

One of the critical characteristics for operational fire monitoring is short revisit times. To effectively detect active fires to support management, data needs to be collected regularly (ideally at least once per day) and downlinked and processed in near-real time. Resolution must also be high enough to pinpoint the fire – less than or equal to 1km.  This poses a challenge for many current systems, as those with the fastest revisit times have lower resolution. MODIS, which is carried on both NASA’s Terra and Aqua missions, has a revisit time of 1-2 days and a spatial resolution of 1km for thermal bands, and is hence frequently used for these applications. One way to overcome the challenges around ensuring adequate revisit frequency and spatial resolution is to combine observations from multiple missions in a global fire monitoring virtual constellation.

Following the analysis of the current and future data supply for active fire detection applications, the Wildfire Pilot team found that the global monitoring capability is unstable for at least the next ten years. Even with newer systems coming online, revisit times are not expected to improve, however spatial resolution is expected to improve by the early 2030s. However, many more systems exist that are capable of fire detection and/or monitoring than that have operational products available. This means that revisit times could be meaningfully improved just by leveraging systems that are already on orbit, or scheduled to be in the near future.   

The Wildfire Pilot team also surveyed the global fire management community to understand the use of space-based active fire products. The survey found that 87% (216) of the respondents stated that their organisations use space-based active fire products. The 247 respondents came from across the globe, with over 50 countries represented. The survey also indicated that fire managers have a high degree of trust in this data. The full survey report can be viewed here. 

With fire managers trusting this data, and relying on it to make critical decisions, it is up to CEOS to find a way forward to ensure these services can continue long into the future, without relying on a small handful of science missions. Better coordination, access and interoperability of data from existing and future missions could help alleviate these issues. 

Recognising the need for a holistic approach to space asset coordination and global wildfire monitoring, WGDisasters and the Wildfire Pilot leads are working with the CEOS community to propose a long term, active fire focussed group to bridge the gap between space agencies, private sector initiatives, product developers, and fire managers. 

Results of the Global Fire Management Community Survey, and their use of EO products

References:

Hope et al., (2024) https://doi.org/10.1080/17538947.2024.2420821 

Johnston et al. (2020) https://doi.org/10.3390/s20185081 

McFayden et al. (2024) https://ostrnrcan-dostrncan.canada.ca/handle/1845/275072